1

PLENARY TALKS 4

SUNDAY 28 AUGUST 2022: 17:00 – 17:30 4

PROTEOMES IN 3D 4
SUNDAY 28 AUGUST 2022: 17:30 – 18:00 5
HIGHLY MULTIPLEXED IMAGING OF TISSUES WITH SUBCELLULAR RESOLUTION BY IMAGING MASS
CYTOMETRY. 5
MONDAY 29 AUGUST 2022: 08:30 – 09:15 6
GETTING OUR FATS STRAIGHT: AN INTERNATIONAL ADVENTURE IN ISOMER-RESOLVED LIPIDOMICS 6
FRIDAY 2 SEPTEMBER 2022: 14:00 – 15:00 7
AIR QUALITY FROM SPACE: INDICATOR OF HUMAN ACTIVITY 7

AWARDS SESSIONS 8

TUESDAY 30 AUGUST 2022: 08:30 – 10:00 8

2020 THOMSON MEDAL AWARDS 8
WEDNESDAY 31 AUGUST 2022: 08:30 – 09:00 9
2020 CURT BRUNNÉE AWARD 9
WEDNESDAY 31 AUGUST 2022: 09:00 – 09:30 10
2022 CURT BRUNNÉE AWARD 10
WEDNESDAY 31 AUGUST 2022: 09:30 – 10:00 10
2022 JOCHEN FRANZEN AWARD 10
THURSDAY 1 SEPTEMBER 2022: 08:30 – 09:15 11
2022 THOMSON MEDAL AWARDS 11

ORAL PRESENTATIONS 14

MONDAY 29 AUGUST 2022: 09:30 – 11:30 14

SESSION: AD-1 ENVIRONMENTAL MS: GEO, WATER, AEROSOLS, VOC’S AND POC’S 14
SESSION: IM-1 NIBBERING SESSION ON ION CHEMISTRY 21
SESSION: IM-2 HIGH RESOLUTION MS - SESSION A 32
SESSION: LS-1 TRANSLATIONAL MS – CLINICAL AND LIQUID BIOPSIES 42
SESSION: LS-2 MS IN STRUCTURAL BIOLOGY - CROSSLINKING MS 52
MONDAY 29 AUGUST 2022: 15:30 – 17:30 61
SESSION: IM-3 ALTERNATIVE DISSOCIATION METHODS 61
SESSION: LS-3 LIPIDOMICS 69
SESSION: LS-4 SINGLE CELL MS AND IN CELL MS 76
SESSION: FP-1 TOXICOLOGY AND METABOLISM 85
SESSION: IM-4 DATA SCIENCES IN MS/AI/CHEMOMETRICS/IDENTIFICATION/MODELLING - SESSION A 95
TUESDAY 30 AUGUST 2022: 11:00 – 13:00 107
SESSION: HC-1 YOUNG MS SCIENTISTS - SESSION A 107
SESSION: IM-5 ION SPECTROSCOPY, PHYSICAL AND CHEMICAL PRINCIPLES UNDERLYING MS - SESSION B 119
SESSION: IM-6 DATA SCIENCES IN MS/AI/CHEMOMETRICS/IDENTIFICATION/MODELLING - SESSION B 129
SESSION: LS-5 GLYCOMICS & GLYCOPROTEOMICS - SESSION A 141
SESSION: FP-2 BIOPHARMACEUTICALS & VACCINES 151
TUESDAY 30 AUGUST 2022: 15:30 – 17:30 157
SESSION: AD-2 FORENSIC SCIENCES 157
SESSION: IM-7 INSTRUMENTATION DEVELOPMENT: MASS ANALYZERS 167
2
SESSION: LS-6 MS IN STRUCTURAL BIOLOGY - NATIVE MS, HDX-MS -SESSION A 179
SESSION: LS-7 CELLULAR SIGNALING PROCESSES AND MS IN SYSTEMS BIOLOGY 188
SESSION: IM-8 SEPARATION & HYPHENATION; CHROMATOGRAPHY, ELECTROPHORESIS – SESSION B, PROTEINS 199
WEDNESDAY 31 AUGUST 2022: 11:00 – 13:00 208
SESSION: AD-3 CULTURAL HERITAGE AND CONSERVATION SCIENCE 208
SESSION: IM-9 IONIZATION TECHNOLOGIES 218
SESSION: LS-8 TRANSLATIONAL MS – CANCER AND IMMUNOLOGY, AND MS 227
SESSION: LS-9 PROTEOMICS: TOP DOWN 238
SESSION: LS-10 IMAGING MS - APPLICATIONS IN LIFE SCIENCE & HEALTH - SESSION A 246
WEDNESDAY 31 AUGUST 2022: 15:30 – 17:30 258
SESSION: HC-2 MS IN THE NETHERLANDS (NVMS SESSION) 258
SESSION: IM-10 IMAGING MS - INSTRUMENTATION AND METHODS 267
SESSION: LS-11 PROTEOMICS: QUANTIFICATION 275
SESSION: BM-1 POLYMERS AND SYNTHETIC MOLECULES 283
SESSION: LS-12 GLYCOMICS & GLYCOPROTEOMICS - SESSION B 291
THURSDAY 1 SEPTEMBER 2022: 11:00 – 13:00 298
SESSION: IM-11 ION SPECTROSCOPY, PHYSICAL AND CHEMICAL PRINCIPLES UNDERLYING MS - SESSION A 299
SESSION: IM-12 SEPARATION & HYPHENATION; CHROMATOGRAPHY, ELECTROPHORESIS - SMALL MOLECULES 309
SESSION: LS-13 METABOLOMICS - SESSION A 318
SESSION: LS-14 CROSS-OMICS, DATA INTEGRATION AND BIOINFORMATICS FOR MS 329
SESSION: FP-3 BIOSIMILARS, BIOBETTERS & GLYCOENGINEERING 334
THURSDAY 1 SEPTEMBER 2022: 15:30 – 17:30 344
SESSION: AD-4 HOMELAND SECURITY, EXPLOSIVES AND ENVIRONMENTAL MONITORING 344
SESSION: IM-13 BEYOND MASS SPECTROMETRY: MAKING MS OBSOLETE…. 355
SESSION: HC-3 YOUNG MS SCIENTISTS - SESSION B 365
SESSION: LS-16 MS IN STRUCTURAL BIOLOGY - NATIVE MS, HDX-MS -SESSION B 374
SESSION: LS-15 PROTEOMICS: PROTEIN -PROTEIN INTERACTION 382
FRIDAY 2 SEPTEMBER 2022: 08:30 – 10:30 392
SESSION: AD-5 ENVIRONMENTAL AND EARTH MS 392
SESSION: IM-14 MINIATURIZATION, LAB-ON-A-CHIP, IN SITU APPLICATIONS 400
SESSION: LS-17 PROTEOMICS: POST-TRANSLATIONAL MODIFICATIONS AND THEIR CROSS-TALK 408
SESSION: IM-15 HIGH RESOLUTION MASS SPECTROMETRY - SESSION B 418
SESSION: LS-18 METABOLOMICS - SESSION B 427
FRIDAY 2 SEPTEMBER 2022: 11:00 – 13:00 434
SESSION: IM-16 AMBIENT TECHNOLOGIES (AND THEIR APPLICATIONS) 434
SESSION: LS-19 IMAGING MS - APPLICATIONS IN LIFE SCIENCE & HEALTH - SESSION B 443
SESSION: FP-4 FOOD, NUTRITION & AGRICULTURE 454
SESSION: HC-4 JMS AWARDEES SESSION 462
469

3
 Plenary Talks
Sunday 28 August 2022: 17:00 – 17:30
PROTEOMES IN 3D
Abstract ID: 848

Presenting author: Paola Picotti, ETH Zurich

Introduction

Proteomics has been broadly applied to detect changes in protein levels in response to perturbations and derive
information on altered pathways. Beyond protein expression changes, however, biological processes are also regulated
by molecular events such intermolecular interactions, chemical modification, aggregation and protein conformational
changes. These events do not affect protein levels and therefore escape detection in classical proteomic screens.
Protein structures integrate all these different types of regulatory molecular events. In my talk I will demonstrate that
global, in situ analyses of protein structures allow detecting various types of protein functional changes simultaneously,
yielding a more detailed and nuanced picture than measurement of abundance changes alone.

Methods

Our group developed the limited proteolysis-coupled mass spectrometry (LiP-MS) approach to monitor protein structural
changes directly within complex biological extracts and on a proteome-wide scale. It is based on the application of
unspecific proteases to native proteome extracts for a short time, followed by complete trypsin digestion under
denaturing conditions and mass spectrometric analysis. Comparison of structure-specific proteolytic fingerprints from
different conditions identifies structurally altered proteins and allows pinpointing structurally altered regions. Differential
proteolytic patterns are represented along the protein sequence in the format of structural barocdes.

Preliminary data (results)

In my talk, I will demonstrate that a global protein structural readout based on LiP-MS detects various types of functional
alterations in situ, such as enzyme activity changes, phosphorylation, protein aggregation, and protein complex formation
in microrganisms undergoing nutrient adaptation and responding to acute stress. The structural resolution of the
approach pinpoints individual regulated functional sites such as binding and active sites, thus supporting the generation
of mechanistic hypotheses and linking holistic and reductionist approaches. I will show how the structural readout allows
detecting novel regulatory interactions due to metabolite-protein interactions. Further, LiP-MS can be used to identify
drug targets o and binding sites in an unbiased manner and directly from cell and tissue extracts. Last, I will show how
the approach supports the identification of a novel type of protein biomarker based on altered protein structures
(structural biomarkers) and present an application to Parkinson’s disease. I will discuss the power and limitations of this
new concept.

Please explain why your abstract is innovative for mass spectrometry?

This structural MS readout substantially increases the coverage of classical proteomics screens and supports the direct
generation of mechanistic hypotheses. It paves the way for in situ structural systems biology.

4
Sunday 28 August 2022: 17:30 – 18:00
HIGHLY MULTIPLEXED IMAGING OF TISSUES WITH SUBCELLULAR
RESOLUTION BY IMAGING MASS CYTOMETRY.
Abstract ID: 807

Presenting author: Bernd Bodenmiller, Universit of Zurich, ETH Zurich

Introduction

Cancer is a tissue disease. Heterogeneous cancer cells and normal stromal and immune cells form a dynamic
ecosystem that evolves to support tumor expansion and ultimately tumor spread. The complexity of this dynamic system
is the main obstacle in our attempts to treat and heal the disease. The study of the tumor ecosystem and its cell-to-cell
communications is thus essential to enable an understanding of tumor biology, to define new biomarkers to improve
patient care, and ultimately to identify new therapeutic routs and targets.

Methods

To study and understand the workings of the tumor ecosystem, highly multiplexed image information of tumor tissues is
essential. Such multiplexed images will reveal which cell types are present in a tumor, their functional state, and which
cell-cell interactions are present. To enable multiplexed tissue imaging, we developed imaging mass cytometry (IMC).
IMC is a novel imaging modality that uses metal isotopes of defined mass as reporters and currently allows to visualize
over 50 antibodies and DNA probes simultaneously on tissues with subcellular resolution. In the near future, we expect
that over 100 markers can be visualized.

Preliminary data (results)

We applied IMC for the analysis of breast cancer samples in a quantitative manner. To extract biological meaningful data
and potential biomarkers from this dataset, we developed a novel computational pipeline called histoCAT geared for the
interactive and automated analysis of large scale, highly multiplexed tissues image datasets. Our analysis reveals a
surprising level of inter and intra-tumor heterogeneity and identify new diversity within known human breast cancer
subtypes as well as a variety of stromal cell types that interact with them.
In summary, our results show that IMC provides targeted, high-dimensional analysis of cell type, cell state and cell-to-cell
interactions within the TME at subcellular resolution. Spatial relationships of complex cell states of cellular assemblies
can be inferred and potentially used as biomarkers.

Please explain why your abstract is innovative for mass spectrometry?

We envision that IMC will enable a systems biology approach to understand and diagnose disease and to guide
treatment.

5
Monday 29 August 2022: 08:30 – 09:15
GETTING OUR FATS STRAIGHT: AN INTERNATIONAL ADVENTURE IN
ISOMER-RESOLVED LIPIDOMICS
Abstract ID: 804

Presenting author: Stephen J Blanksby, Queensland University of Technology

Introduction

Contemporary lipidomics has made significant strides over the last 20 years driven largely by rapid advances in mass
spectrometry. Notably, advances in mass-resolving power, the commercial availability of ion-mobility mass spectrometry
and the advent of novel ion activation modalities have helped reveal the vast complexity of the lipidome; including the
discovery of novel lipid structures and, consequently, new metabolic pathways. Despite these advances, the global effort
to describe the lipidome in a manner that can be reconciled with the proteome and the genome continues to face
significant challenges. Most significant among these is the identification and quantification of lipid isomers.

Methods

Our group has had the immense fortune to work alongside an international community of lipidomics researchers to
develop and implement novel workflows to address the specific challenges presented by lipid isomers. Here we present
the implementation of (i) ion-molecule and ion-ion reactions; (ii) photodissociation mass spectrometry and (iii) high
resolution ion mobility mass spectrometry for lipid structure elucidation. These technologies are demonstrated to resolve
different lipid isomer families both in classical direction infusion- and liquid chromatography-mass spectrometry
workflows as well as imaging modalities.

Preliminary data (results)

Application of these novel analytical workflows to lipid extracts from diverse sources including human plasma, cell lines
and vernix caseosa has led to the discovery of new lipids that point to hitherto undescribed metabolism within the source
cell or organism. In particular, explicit regiochemical assignment of position(s) of carbon-carbon double bonds and
methyl branch points in fatty acyl chains and the relative position of acyl chains within glycerolipids, point to a much
greater structural diversity within the lipidome than is typically considered. Moreover, the unambiguous assignment of
these structural features enables mapping changes in lipid isomers populations to changes in the underlying metabolism
and environment. While, application of isomer-resolved lipidomics in combination with imaging mass spectrometry has
proven to be an effective means to visualise changes in metabolism within tissue that can provide a level of contrast
unavailable in conventional imaging modalities. These results provide a strong motivation for achieving full lipid structure
elucidation of complex lipids by mass spectrometry and our preliminary efforts towards the resolution of lipid
stereochemistry will be discussed.

Please explain why your abstract is innovative for mass spectrometry?

New mass spectrometric approaches to lipid isomer resolution

6
Friday 2 September 2022: 14:00 – 15:00
AIR QUALITY FROM SPACE: INDICATOR OF HUMAN ACTIVITY
Abstract ID: 847

Presenting author: Pieternel F. Levelt, TU Delft

Introduction

In the 19th and 20th century the chemical composition of the atmosphere did change drastically as a result of human
activities. Therefore, the Dutch Nobel Prize winner Paul Crutzen called this time period the ‘anthropogenic’ epoch. The
rapid worldwide growth of megacities, and its associated strong increase in air pollution, are clear examples of this.
These are developments that will continue to be important in the coming decades, even with the agreements made
during the recent UN climate change conference in Glasgow (2026) and the Global Methane Pledge, signed by 100
countries around the world, to reduce methane emission with 30 % by 2030.

Methods

Nowadays we can measure the chemical composition of the atmosphere with satellites. With innovative satellite
instruments of Dutch origin, such as OMI and TROPOMI, daily global maps of air pollution and greenhouse gases are
measured on urban scale resolution.

Preliminary data (results)

During the lecture, an outline will be given of the major research questions in the atmospheric climate domain, and their
importance for air quality and climate policy. Further, the satellite measurement technique will be explained, and what
these measurements can bring for as well research as climate policy, now and in the future. Examples of COVID-19
lockdown reduction of air pollution, methane emission from the oil and gas industry, as well as the potential of satellite
measurements for the nitrogen deposition policy, will be shown. The use of satellite observations for countries and
continents from the global south, like Africa, will be addressed as well.

Please explain why your abstract is innovative for mass spectrometry?

TROPOMI Tropospheric NO2 column over Western Europe, average for April 2018 (courtesy Henk Eskes, KNMI).

7
 Awards Sessions
Tuesday 30 August 2022: 08:30 – 10:00
2020 Thomson Medal Awards

INVESTIGATING PROTEIN FOLDING, FUNCTION, AND INPUT TO DISEASE

USING MASS SPECTROMETRY
Abstract ID: 1037

Presenting author: Alison Ashcroft, University of Leeds

Introduction

The study of protein structure and dynamics is essential to our knowledge of protein folding and function, and
subsequently for the development of effective therapeutics targeted against protein-related diseases. Prevalent amongst
such illnesses are the amyloid-related diseases including Alzheimer’s, Parkinson’s, and type 2 diabetes mellitus, which
are associated with protein mis-folding followed by self-aggregation (Figure 1). For diseases associated with viruses
there is a paucity of anti-viral therapies; hence mapping virus capsid assembly from the individual protein monomers can
lead to the development of effective therapeutics. Native mass spectrometry is a powerful, versatile tool with which to
monitor biomolecular assembly pathways, identify key intermediates and test potential inhibitors of unwanted processes,
with the ability to deal with heterogeneous samples typically not amenable to high-resolution structural techniques such
as NMR, X-ray crystallography, or cryo-EM. ESI-IMS-MS can be used to study and record the mass, stability, cross-
sectional area and ligand binding capability of each transiently populated, non-covalently bound intermediate present in
the heterogeneous mixture of assembling species, in a single experiment in real-time. Here, examples from the Ashcroft
Group on the use and development of native mass spectrometry for protein identification will be presented, using IMS-
MS and labelling techniques including solution-phase HDX, hydroxyl radical footprinting, and chemical cross-linking
(Figure 2). The biomolecules under scrutiny include amyloid proteins, virus capsids, membrane proteins, and monoclonal
antibodies.

Please explain why your abstract is innovative for mass spectrometry?

8
TRANSLATIONAL IMAGING MASS SPECTROMETRY: INTERDISCIPLINARY
TEAM SCIENCE AT WORK
Abstract ID: 1042

Presenting author: Ron M.A. Heeren, The Maastricht MultiModal Molecular Imaging institute, Maastricht
University, The Netherlands

Introduction

Modern mass spectrometry in the “omics” arena plays a crucial role in many scientific disciplines ranging from material
sciences to clinical diagnostics. This requires researchers to think differently about the element of success. An essential
element in modern science is the concept of “team science”; it is virtually impossible to do anything by alone or within a
single discipline. Collaboration across scientific domains is key! Especially if hi-tech innovations are intended for
translation into a new application domain. This Thomson medal award lecture would not have been possible without very
enjoyable interdisciplinary team science. Mass spectrometry imaging technological advances are succeeding each other
with lighting speed. Increased methodological sensitivity allows researchers to acquire local metabolic, lipidomic and
proteomic information of smaller and smaller samples down to even single cells. The biggest challenge is to put that
concerted information in the context of the problem the samples originate from. This lecture describes how innovative
molecular imaging technologies, based on mass spectrometric innovations have impacted translational clinical research.
Or: how a mass spectrometer can be used as a sensitive and selective metabolic microscope. Innovative imaging
technologies now offer new insights in life’s complexity that can be employed for precision medicine, the understanding
of new (bio)materials and the processes that happen on the interface of living and ‘dead’ matter. Innovations in mass
spectrometry based chemical microscopes have now firmly established themselves in translational molecular research.
One key aspect of translational success is the ability to obtain this molecular information on thousands of molecules on a
process relevant timescale. Modern mass microscopes can now rapidly acquire images of metabolites, lipids, polymers,
peptides and proteins, depending on the spatial resolution chosen. Single cells can be analyzed in great molecular detail
and in the context of their native tissue. This is a true testimony and result of many interdisciplinary science teams at
work for translational Imaging Mass Spectrometry: from academic institutions to companies active in our field. This is
what changes our world, one image at a time.

Co-authors:

A Dynamic Team, Maastricht University

Wednesday 31 August 2022: 08:30 – 09:00

2020 Curt Brunnée Award

Guiding Medical Decisions with the MasSpec Pen Technology

Abstract ID: 1039

Presenting author: Livia S. Eberlin, Baylor College of Medicine

Introduction

Mass spectrometry techniques that allow direct and fast molecular analysis of tissues offer the exciting opportunity to
incorporate molecular data into medical practice to expedite treatment decisions and advance care for patients. In this
talk, I will discuss my lab’s efforts in the development and application of MasSpec Pen technology for direct tissue
analysis and disease diagnosis in the operating room, as well as in the microbiology and clinical pathology labs. I will
present results obtained in our ongoing clinical studies where the technology is being tested by medical professionals
and discuss the analytical and diagnostic performance metrics achieved to provide a critical assessment of potential
uses and limitations within the context of medical practice. Challenges and opportunities in implementing the technology
into clinical workflows will also be discussed.

9
 Wednesday 31 August 2022: 09:00 – 09:30
2022 Curt Brunnée Award

ION MOBILITY SPECTROMETRY: FROM INSTRUMENTAL ADVANCES TO

UNDERSTANDING THE INTERSECTION BETWEEN HUMAN HEALTH AND
THE ENVIRONMENT
Abstract ID: 1043

Presenting author: Erin S. Baker, University of North Carolina - Chapel Hill

Introduction

To date, the use of mass spectrometry (MS)-based analyses has provided extensive knowledge for many different omic
studies, especially due to recent advancements in complex sample extractions, front-end chromatography separation,
fragmentations strategies, database construction, and bioinformatics tools. However, numerous analytical challenges still
remain in MS studies including the analysis of isomeric molecules, suppression of low abundance species in complex
samples, disparities in ionization efficiency, and difficulties in data analysis and molecular annotation. Due to these
obstacles, both experimental and computational advancements are still greatly needed to enable better omic
measurements. In this presentation, I will showcase how ion mobility spectrometry (IMS) has advanced throughout the
last thirty years, resulting in IMS separations with higher sensitivity, throughput, and resolving power. These
improvements have enhanced the coupling of IMS with various chromatography and MS platforms and provided 3- and
4-dimensional analyses such as LC-IMS-MS or LC-IMS-CID-MS measurements. Currently, these multidimensional
analyses are facilitating the separation of highly similar molecules, increasing the number of observed features, and
providing improved identification confidence. Furthermore, applications ranging from clinical analyses to environmental
assessments are incorporating these simultaneous separations to enable the in-depth omic evaluations necessary for
the complex sample types.

Co-authors:

James N. Dodds, University of North Carolina - Chapel Hill

Kaylie I. Kirkwood, North Carolina State Univerity
Rebecca Beres, University of North Carolina - Chapel Hill
Jack Ryan, University of North Carolina - Chapel Hill
Anna Boatman, University of North Carolina - Chapel Hill
Jessie Chappel, North Carolina State Univerity
Melanie T. Odenkirk, University of North Carolina - Chapel Hill
Karen Butler, University of North Carolina - Chapel Hill

Wednesday 31 August 2022: 09:30 – 10:00

2022 Jochen Franzen Award

MASS SPECTROMETRY TECHNOLOGIES FOR THE IMAGING OF LIPIDS IN

SPACE AND TIME
Abstract ID: 1048

Presenting author: Shane R. Ellis, Molecular Horizons and School of Chemistry and Molecular Bioscience,
University of Wollongong, Wollongong, NSW, Australia 2522

Introduction

Despite lipids being arguably the most widely studied family of molecules using mass spectrometry imaging (MSI), it is
often only a tiny snapshot of lipids detected. This limitation primarily arises from the considerable variation in ionisation
efficiencies between different lipid classes, and many critical lipids species are often not detected with sufficient
sensitivity. The first part of this lecture will describe work over the past five developing and applying post-ionisation

10
technologies to improve lipid coverage, sensitivity, and spatial resolution of MALDI-MSI experiments. The first method we
have explored is laser post-ionisation (MALDI-2) implemented on Orbitrap platforms that allows an order of magnitude
improvement in lipid coverage whilst simultaneously enabling higher spatial resolution by compensating for the
degradation in ionisation efficiency typically countered using oversampling. For example, it allows the imaging of single
lipid droplets within human brain tissue containing active multiple sclerosis lesions tissue at a pixel size of 6 µm. In
addition, it will be shown how MALDI-2 using sub-threshold primary laser energies conditions enables an 11-fold
increase in ionisation efficiencies for aromatic analytes capable of undergoing direct 2-photon ionisation (REMPI) despite
the effective desorption area being reduced from 18 to 6 µm in diameter, a phenomenon attributed to the resulting
sparse plume conditions. Secondly, recent work describing the development of an AP-MALDI-MSI system coupled with
plasma post-ionisation and trapped-ion mobility mass spectrometer will be presented. The second part of this lecture will
describe recent efforts to image dynamic lipid synthesis in tissues using isotope labelling strategies. In particular, a dual
labelling strategy incorporating D9-choline and [U]-13C-PC16:0/16:0 to track the synthesis and catabolism of lung
surfactant. A key outcome of this work is the ability to localise acyl remodelling events within lung tissue, thus shifting
MSI from imaging sum-composition lipid distributions to localising distinct metabolic events.

Thursday 1 September 2022: 08:30 – 09:15

2022 Thomson Medal Awards

SURFACE COLLISIONS: FROM PEPTIDES TO NUCLEOPROTEIN

COMPLEXES
Abstract ID: 1038

Presenting author: Vicki Hopper Wysocki, Department of Chemistry and Biochemistry , The Ohio State
University

Introduction

Hyperthermal collisions with surfaces for characterization of projectiles ions in MS/MS were introduced by the Cooks lab
at Purdue University in the 1980s. It was clear in the early days that collisions with surfaces, more massive targets than
the projectiles used for collision-induced dissociation, CID, should be valuable for dissociation of massive
ions. Unfortunately, instruments of the day could not ionize and transmit high m/z ions. Over the years, The Wysocki
Research Lab has exploited surface collisions in development of the mobile proton model for peptide fragmentation and,
more recently for the characterization of protein and nucleoprotein complexes. As methods and instruments in the
community have morphed to accommodate more massive ions, surface collisions have been integrated into a variety of
instrument types (e.g., QqQ, QTOF, ICR, Orbitrap) and coupled with online separations, with ion mobility, and with other
activation methods, including ECD. The data are used throughout the stages of a biochemical/structural biology project
and in ways that are complementary to other structural biology tools. Examples will be provided to illustrate the value of
native mass spectrometry experiments that incorporate surface-induced dissociation, SID, for structural characterization
and the overlap/integration of the results with data from other approaches.

ISOTOPE-ELEMENT MASS SPECTROMETRY: NEW DEVELOPMENT IN

GEOCHRONOLOGY AND BIOCHEMISTRY.
Abstract ID: 1049

Presenting author: Lidiya N. Gall, Institute for Analytical Instrumentation of Russian Academy of Sciences IAI
RAS, ulitsa Ivana Chernykh, 31, lit. A, St. Petersburg, 198095, Russia

Introduction

Isotope and element analyses are analytical measurements performed exclusively by mass spectrometric methods and
tools. Isotope measurements, which in geochronology often require extremely high accuracy for determining isotope
ratios, are particularly challenging. The need to obtain such results dictates the choice of ionization methods with minimal
discriminations to achieve maximum consistency of the obtained ions composition to the composition of the sample, the
choice of a mass analyzer with high isotopic sensitivity and spectrographic registration capability, and the design of the
ion path without collimation of the ion beam by its elements. A mass spectrometer with a magnetic sector analyzer is the
optimal match for these requirements. The paper provides an overview of the newly developed ion-optical solutions for
obtaining the best sensitivity/resolution ratio when using electron ionization and surface thermoionization methods.
Additionally, it addresses the new biochemical application of isotope analysis related to the discovery of the magnetic

11
isotope effect in the living and advancement of biophysical and medical research in this area. In this case, it is usually
desirable to carry out isotope measurements simultaneously with molecular and elemental analysis, which requires the
use of new ionization methods. We also review the potential of the electrospray ionization method with in-source
atomization and fragmentation (ERIAD), allowing to consecutively obtain molecular, elemental, and isotopic spectra in
one analysis, which is extremely important for advancement of mass spectrometry in the “new isotopy”.

12
13
 Oral Presentations
Monday 29 August 2022: 09:30 – 11:30
Session: AD-1 Environmental MS: Geo, Water, Aerosols, VOC’s and
POC’s

A (BLUE) REVOLUTION: HIGH-RESOLUTION MASS SPECTROMETRY

APPLICATIONS IN THE (DRINKING) WATER SECTOR
Abstract ID: 851

Presenting author: Frederic Béen, KWR Water Research Institute

Introduction

Suspect and non-targeted screening (NTS) approaches using high-resolution mass spectrometers (HRMS) coupled with
liquid chromatography (LC) have increasingly been used for the qualitative and semi-quantitative assessment of
contaminants in water. In particular, these methodologies have been used to assess the removal of chemicals during
waste- and drinking water treatment, as well as to the monitor the quality of water bodies and drinking water sources. In
fact, the increasing amount of anthropogenic chemicals being used and eventually ending up in the aquatic
environments, combined with the shift towards more risk-based monitoring approaches, called for the implementation of
more comprehensive methods which allow to screen for a broader range of chemicals compared to conventional
targeted methods.

Methods

Combined with data-driven approaches combining statistical analysis, machine learning and toxicological information,
HRMS and NTS have been used to detect, prioritise and semi-quantify contaminants of concern, not part of routine
monitoring programs. Furthermore, NTS methods have been increasingly used to assess the removal of chemicals
during drinking- and wastewater treatment, including the formation of transformation products (TPs). More recently, NTS
and HRMS have been coupled with complementary and orthogonal separation techniques such as hydrophilic interaction
liquid chromatography (HILIC), mixed-mode chromatography (MMC) and supercritical fluid chromatography (SFC) for the
screening of (highly) polar compounds.

Preliminary data (results)

Suspect and non-targeted screening (NTS) methods have become an essential tool for (drinking) water laboratories
involved in the monitoring of water quality in The Netherlands. Despite the great improvements in recent years, some
bottlenecks still exist with respect to the translation of NTS-based methods and workflows from research to routine
practices. The goal of this presentation is to give an overview of the applications and paradigm shifts that HRMS and
NTS have brought to the (drinking) water sector, as well as the opportunities and challenges which lay ahead.

Please explain why your abstract is innovative for mass spectrometry?

HRMS and NTS had an important impact on the practices of laboratories monitoring emerging contaminants in water, yet
some bottlenecks still exist with respect to their routine implementation.

14
CHEMINFORMATICS AND HIGH RESOLUTION MASS SPECTROMETRY IN
HISTORICAL EXPOSOMICS OF THE MINETTE REGION
Abstract ID: 22

Presenting author: Dagny Aurich, Luxembourg Centre for Systems Biomedicine (LCSB), University of
Luxembourg, Avenue du Swing 6, L-4367 Belvaux, Luxembourg

Introduction

Investigating historical exposures can provide valuable insights into how the past contributes to current or even future
health conditions. While high resolution mass spectrometry (HR-MS) combined with non-target analysis (NTA) can help
detecting these contaminants in environmental samples, long term datasets do not yet exist, suitable historical samples
are rare and precious, and not all matrices or sample types available in the region of interest will help reconstruct the
past. The LuxTIME project (Luxembourg Time Machine) studies the long term impact of environmental changes on the
health of the local population of the Luxembourgish Minette region, characterized by its past of iron and steel industry in
a highly interdisciplinary manner.

Methods

This work explores how the boundaries of retrospective screening with NTA and HR-MS can be extended into a more
distant past, tackling the challenge of extracting chemical and event-based information from historical documents along
with the automated extraction of relevant chemicals from related patent information to apply in current open NTA
workflows such as patRoon and MS-Dial. This historical exposomics workflow was established initially focusing on
industrial contaminants, using surface waters of the Minette area. Retrospective HR-MS (QExactive HF) analysis
followed by NTA and suspect screening using a list of industrial chemicals (metallurgy industry) was performed on
samples from 2019-2021.

Preliminary data (results)

Relatively few industrial chemicals were found in the initial screening efforts, with pharmaceuticals dominating the
tentative identifications. More advanced suspect list creation based on patent and use information, as well as chemical
information extracted from historical archives is ongoing, with a focus on constraining patent matches (e.g. ~330K
chemicals in the metallurgy category) to more relevant subsets according to industrial disease or associated chemicals.
While patent information is a potentially powerful method of obtaining information, too many matches are found, such
that methods of refining the extracted chemicals are much needed.

The historical exposomics workflow will be applied on a hydrogeological sampling campaign interrogating specific layers
(‘dead zone’) of groundwater, to construct the chemical memory of the region, and establish whether groundwater dating
and chemical appearance can be connected with historical events and patent occurrence.

Please explain why your abstract is innovative for mass spectrometry?

Establishing an historical exposomics workflow using HRMS and patent information from metallurgic industry may help
connecting past chemical exposures from steel industry times to current health outcomes.

Co-authors:

Philippe Diderich, Administration de la gestion de l’eau (AGE), Avenue du Rock’n’Roll 1, L-4361 Esch-sur-Alzette,
Luxembourg
Emma Schymanski, Luxembourg Centre for Systems Biomedicine (LCSB), University of Luxembourg, Avenue du Swing
6, L-4367 Belvaux, Luxembourg

15
POLYETHYLENOXIDE IN THE AQUATIC ENVIRONMENT – DEVELOPMENT
AND OPTIMISATION OF A QUANTITATIVE TRACE-ANALYTICAL METHOD
AND FIRST OCCURRENCE DATA
Abstract ID: 156

Presenting author: Frances Pauelsen, Hochschule Fresenius gem. GmbH, Institute for Analytical Research,
65510 Idstein, Germany, Current address: Institute of Food Chemistry and Food Biotechnology, Justus Liebig
University of Giessen, 35392 Giessen, Germany

Introduction

The pollution of the environment by polymers has been discussed in scientific and political debates. Since these
discussions are mainly about particulate polymers, WSPs such as polyethylene oxide (PEO) are excluded or ignored
from these discussions. WSPs enter the environment directly and indirectly through their various applications (e.g. in
wastewater treatment). There, they either remain in the aqueous phase or sorb to particulate matter. The degradation of
WSPs leads to shorter polymer chains and transformation products with yet unknown impact on the environment. RP-LC
HRMS has shown possibilities for their characterisation. However, quantification was not possible due to the overall
intensity distribution over various homologues, charge states and ion species. Herein, we demonstrate a molecular-
weight independent approach for the quantification of PEO in aqueous matrices.

Methods

Extraction and purification was performed using C18 SPE cartridges. Measurements were performed on an SEC
(isocratic elution)-ESI-QTOF MS, which operated in full scan and SWATH acquisition mode. Quantification is performed
through the combined intensity of diagnostic fragments ([C 2nH4n+1On]+). Based on the SEC elution times, the signal could
be divided into molecular weight (MW)-dependent fractions. These fractions could be calculated and corrected
individually by MW-specific blank values, recoveries and response factors (influenced by ionisation and fragmentation
efficiency). The resulting concentrations of the respective MW-specific fractions are summed and subsequently corrected
by the calculated MW-independent matrix effects.

Preliminary data (results)

Environmental results were collected from a total of 18 surface waters, 2 wastewater treatment plants (WWTPs) (influent
and effluent respectively) and 5 drinking water samples. Since this method relies on the formation of diagnostic
fragments a differentiation between PEO, PEO-derivatives was not possible. The analysis of the WWTP samples
showed an overall removal of PEO and its derivatives of 80%. This was indicated by a decrease in concentration (405
µg/L and 254 µg/L (influent) and 13 µg/L and 20 µg/L (effluent)) and a shift in molar mass distribution. However, despite
the good degradability of PEO and the high removal efficiency of the WWTPs, a concentration of more than 10 µg/L was
detected in 6% and more than 1 µg/L in 61% of the surface waters sampled. The MW distribution of PEO in the surface
water samples was approximately 1300-4000 Da, which is similar to the MW distribution of PEO in the effluent samples,
indicating WWTPs to be the main discharge pathway for PEO and its derivatives into the aquatic environment. Among
the drinking water samples analysed, only one concentration of 0.6 µg/L was above the iLOD. The origin of this sample
was groundwater, which leads to the assumption that the PEO has already been largely degraded and mineralised.
Since concentrations in the double-digit µg/L range of a polymer described in the literature as readily degradable could
already be analysed, this increases the interest in the inclusion of further less readily degradable WSPs.

Please explain why your abstract is innovative for mass spectrometry?

Employment of an SEC-ESI-HRMS method utilizing an all-ion fragmentation approach for the formation of diagnostic
fragments independent of MW, charge state, and ion species for the quantification of PEO.

Co-authors:

Daniel Zahn, Hochschule Fresenius gem. GmbH, Institute for Analytical Research, 65510 Idstein, Germany
Sven Huppertsberg, Hochschule Fresenius gem. GmbH, Institute for Analytical Research, 65510 Idstein, Germany
Thomas P. Knepper, Hochschule Fresenius gem. GmbH, Institute for Analytical Research, 65510 Idstein, Germany

16
METAPROTEOMICS: A NEW TOOL IN WASTEWATER SURVEILLANCE
Abstract ID: 196

Presenting author: Claudia Tugui, Delft University of Technology

Introduction

Metaproteomics aims to measure the proteome of complex samples such as complete microbial communities to unravel
their microbial composition and metabolic activities. Since the coronavirus outbreak in 2019, wastewater-based
epidemiology has gained increasing momentum because it allows to monitor the spread of pathogens such as viruses on
a global level. Wastewater is also increasingly targeted to quantify the use of pharmaceuticals by the population and to
investigate distribution routes of antibiotic resistance genes. However, wastewater not only hosts bacteria and viruses,
but it also contains a high diversity of excreted (human) proteins and pharmaceuticals. Where the analysis of
pharmaceuticals and DNA became fairly routine over the past decade, there are currently no effective approaches that
measure the macromolecules – such as proteins – on a large-scale level.

Methods

The crude wastewater was subjected to lysis, protein precipitation and Trypsin digestion. The resulting peptide fraction
was isolated using a solid phase extraction well-plate. The enriched peptides were further analysed by shotgun
proteomics using an EASY nano LC coupled to a QE plus Orbitrap mass spectrometer, as described in Kleikamp et al.
(2021)1. The raw data were de novo sequenced to construct a taxonomic profile from the wastewater (NovoBridge 1). An
integrated API module was further used to construct from identified taxonomies a focused database from UniprotKB
entries for the final PEAKS database search and protein identification.

Preliminary data (results)

In the present study, we demonstrate the application of metaproteomics to provide large-scale proteomics data from
crude wastewater. The proteomic analysis of samples from such environments represents a challenge for mass
spectrometry due to its large chemical, as well as species diversity. Therefore, we developed a high-throughput sample
preparation procedure and a streamlined data processing pipeline employing a recently established de novo sequencing
approach (NovoBridge)1 that allows the highly efficient use of public sequence databases for species wide proteome
analysis.

In summary, we demonstrate the large microbial diversity found in wastewater streams ranging from common gut
microbes like Prevotella, Faecalibacterium or Acinetobacter to pathogenic Clostridium, Pseudomonas and
Cronobacter, or opportunistic pathogens such as Aeromonas or Campylobacter. Moreover, we observed a surprisingly
large array of human proteins, of which some are potential disease/health biomarkers like uromodulin, immunoglobulins,
alpha-1 antitrypsin and cancer related proteins. Ultimately, the approach aims to provide additional information to monitor
human health related parameters at a population scale, and possibly also to provide orthogonal data to DNA-based
approaches concerning (upcoming) epidemics.

References

[1] Kleikamp, H. B., Pronk, M., Tugui, C., da Silva, L. G., Abbas, B., Lin, Y. M., ... & Pabst, M. (2021). Database-
independent de novo metaproteomics of complex microbial communities. Cell Systems, 12(5), 375-383

Please explain why your abstract is innovative for mass spectrometry?

Our approach expands metaproteomics into the field of wastewater surveillance, realized by the rapid sample
preparation from crude wastewater and the de novo peptide sequencing based database construction approach.

Co-authors:

Willem van Holthe, Delft University of Technology

Hugo Kleikamp, Delft University of Technology
Mark van Loosdrecht, Delft University of Technology
Martin Pabst, Delft University of Technology

17
Figure 1: The metaproteomics-based wastewater survey approach.

SCREENING OF TRANSFORMATION PRODUCTS AND INTERMEDIATES

OF EMERGING CONTAMINANTS FROM SIMULATED AEROBIC
DEGRADATION TESTS COPULED WITH HPLC-MSN
Abstract ID: 704

Presenting author: Federico Ivanic, Institute of Environmental Research and Engineering (UNSAM-CONICET)

Introduction

Emerging contaminants (ECs) are chemical compounds and metabolites that have not yet been generally reached by the
existent water quality regulations, but still represent a harmful threat. ECs undergo degradation in natural environments
creating new metabolites, which could potentially be more harmful than their parent compound. HPLC-MS facilitates the
detection of multiple low abundance transformation products (TPs) in complex matrices, which may be generated
through lab experiments simulating diverse environmental conditions of stationary water bodies. Moreover, multiple
tandem mass spectrometry allows structural characterization of TPs.

Broad spectrum antibiotic oxytetracycline (OTC) was chosen as model contaminant due to its intensive use in areas of
animal husbandry, thus requiring a deeper study about its fate in the environment.

Methods

Aerobic biotic and abiotic OTC degradation experiments were carried out in aerated water/soils reactors that allowed
adjusting and measuring influential parameters such as pH, temperature, redox potential and dissolved O2. Aqueous
samples were withdrawn and analyzed at different times using HPLC - MSn. Mass spectra were acquired using a linear
ion trap mass spectrometer (Thermo LTQ XL) equipped with an electrospray source in positive-ion mode and multiple
stage fragmentation analysis was carried out. Search of transformation products was accomplished by a non-target
screening approach, using MatLab software to find potential candidates through the setting an intensity threshold.

Preliminary data (results)

OTC degradation followed a typical exponential decay corresponding to first order kinetics, reaching less than 5% of the
initial concentration in all of the experiments within the screened time. Half-life in biotic experiments was lower than in the
abiotic ones; similarly, lower pH decreased half-life of the pollutant.

Screening of transformation products and intermediates using HPLC-MS afforded several different m/z ions.
Furthermore, multiple tandem mass spectrometry proved to be a powerful tool in characterizing their chemical structures,
allowing to determine approximately 20 TPs. Generally, 5 or 6 fragmentation stages were needed to assign a possible

18
structure, depending on the ion abundance, as non-specific H2O and CO losses were commonly observed during the first
stages.

On the one hand, approximately 15 TPs were successfully identified at all tested conditions, although different rates and
prevalence were observed, proving that time evolution of TPs depends on the degradation condition. These metabolites
are generally products of oxidation reactions, although other degradation pathways were found, e.g. decarbonylation,
dehydration and condensation reactions. On the other hand, ions specific to certain degradation conditions accounted for
approx. 25% of the identified TPs, particularly, in biotic conditions, through reactions such as demethylation and
hydrolysis. Lastly, OTC degradation products adsorbed on the sediments by the end of the experiment were desorbed
and analyzed. This analysis, combined with time trends of intermediates in aqueous samples, allowed to classify
between persistent and short life products.

Please explain why your abstract is innovative for mass spectrometry?

Coupling of simulation tests with HPLC-MSn in extended experiments allowed the thorough characterization of the
chemical fate of an emerging contaminant under different environmentally significant conditions.

Co-authors:

Matias Butler, Institute of Environmental Research and Engineering (UNSAM-CONICET)

Roberto Candal, Institute of Environmental Research and Engineering (UNSAM-CONICET)

THE IDENTIFICATION OF THYROID HORMONE SYSTEM DISRUPTING

COMPOUNDS IN HUMAN CORD BLOOD SAMPLES USING EFFECT-
DIRECTED ANALYSIS
Abstract ID: 357

Presenting author: Jeroen Meijer, Department of Environment & Health, Faculty of Science, Amsterdam Institute
of Molecular and Life Sciences, Vrije Universiteit Amsterdam, De Boelelaan 1085, 1081 HV Amsterdam, The
Netherlands, Institute for Risk Assessment Sciences (IRAS), Utrecht University, Yalelaan 2, 3584 CM Utrecht, the
Netherlands

Introduction

Thyroid hormone (TH) system disruption is associated with adverse health effects in humans and might be caused by
exposure to xenobiotic substances that competitively bind to the T4 thyroid hormone transporter protein transthyretin
(TTR). There are many known compounds that show TTR-binding potential, however, it is likely that humans are also
exposed to yet unknown TTR-binding compounds. In previous work, an Effect-Directed Analysis (EDA) workflow was
developed that combines both toxicological testing and chemical analysis of fractionated serum extracts. This work
resulted in the identification of two unknown TTR-binding compounds in a fetal calf serum sample, namely, fipronil
sulfone and bisphenol S. The aim of the present study was to apply this EDA approach to identify unknown TTR-binding
compounds in human samples.

Methods

Six human cord blood sample extracts were tested in the T4-FITC TTR-binding assay, which shows a decreased
fluorescent signal when TTR-binding compounds are present. Furthermore, the extracts were chemically analyzed using
LC-QTOF MS(/MS) and fractionated into 80 fractions (each corresponding to 13.5 seconds of the chromatographic run).
The fractions were subsequently tested in the TTR-binding assay. The chemical features present in the chromatographic
window of the active fractions were prioritized for identification, reducing the complexity of the samples. All peak picking,
blank subtraction and identification steps were performed using Metaboscape.

Preliminary data (results)

All the unfractionated extracts showed TTR-binding activity. The TTR-binding activity was higher than what would be
expected if the activity resulted from the native T4 hormone alone. However, after fractionation, activity was only found in
two separate fractions from two samples. This discrepancy might be caused by the presence of a large number of TTR-
binding compounds in the extracts that are widely distributed across the chromatogram. In this case, the concentration of
each individual constituent of the mixture in the unfractionated extract might not be high enough to reach the limit of
detection of the assay. Chemical analysis (in both positive and negative mode) resulted in the detection of over 14,000

19
chemical features, of which 317 features were present in the retention time window of the most active fraction. Suspect
screening using the CECscreen database (>70,000 chemicals of emerging concern) resulted in the annotation of 142
compounds at identification confidence level 4 that might be responsible for the measured activity. MS/MS spectral
library searches resulted in seven matches, all of which were endogenous compounds. Future work will focus on further
prioritization and elucidation of those suspected candidates in order to find the chemical(s) responsible for the TTR-
binding activity in that active fraction. Furthermore, suspect screening will also be performed using the simulated
metabolite list of CECscreen (>300,000 structures).

Please explain why your abstract is innovative for mass spectrometry?

This work shows a multi-disciplinary screening technique (e.i. toxicity testing, high-resolution mass spectrometry and
novel data processing techniques) to identify unknown TH system disrupting compounds in real-life human samples.

Co-authors:

Jelle Vlaanderen, Institute for Risk Assessment Sciences (IRAS), Utrecht University, Yalelaan 2, 3584 CM Utrecht, the
Netherlands
Roel Vermeulen, Institute for Risk Assessment Sciences (IRAS), Utrecht University, Yalelaan 2, 3584 CM Utrecht, the
Netherlands
Timo Hamers, Department of Environment & Health, Faculty of Science, Amsterdam Institute of Molecular and Life
Sciences, Vrije Universiteit Amsterdam, De Boelelaan 1085, 1081 HV Amsterdam, The Netherlands
Marja Lamoree, Department of Environment & Health, Faculty of Science, Amsterdam Institute of Molecular and Life
Sciences, Vrije Universiteit Amsterdam, De Boelelaan 1085, 1081 HV Amsterdam, The Netherlands

20
Session: IM-1 Nibbering session on Ion Chemistry

REACTION ACCELERATION IN MICRODROPLETS: SCOPE AND

MECHANISMS
Abstract ID: 100

Presenting author: R. Graham Cooks, Purdue University

Introduction

The phenomenon of reaction acceleration in small volumes has been known for a decade from studies of reactions in
electrosprayed droplets. Reaction rates at the interface are much greater than those in the bulk and many organic
reactions are known to be accelerated in organic and aqueous solutions. Two major mechanisms have been suggested
for this phenomenon: (i) the partial solvation of reagents at the interface and (ii) the high electric fields at aqueous
interfaces. The microdroplet acceleration enables reactions to occur during millisecond droplet flight times so allowing
application to reaction screening in times of less than 1 second (screening rates approaching 10,000 reactions/hour). A
second major application is the collection of reaction products as a means to small-scale synthesis.

Methods

Sprayed and levitated droplets, as well as thin films, show reaction acceleration. Mechanistic studies use online MS
analysis as does high throughput reaction screening but the latter requires robotic array creation and rapid screening
under hardware control with automated data acquisition. Desorption electrospray ionization (DESI) is used in the high-
throughput platform while DESI and other electrospray methods are used in single-channel studies. Levitated droplets
are thermally or ultrasonically levitated and their surfaces are sampled and analyzed off-line. Small-scale synthesis
involves droplet collection on a suitable substrate after optimizing conditions to maximize yield in online analysis.

Preliminary data (results)

Some accelerated interfacial reactions proceed using the same reagents and parallel reaction mechanisms to those in
bulk, examples are the Katritzky transamination and the formation of imines. In these cases, which are observed in a
variety of solvents, the partial solvation mechanism rationalizes the increase in transition state stability and hence the
rate constant increase (Figure 1). In other cases, the extreme conditions at the interface cause unexpected reactivity,
e.g. formic acid or CO2 undergoing Lewis base attack by an amine to form a new N-C bond (Figure 2). Both electric field
and partial solvation effects likely operate in these cases. In still other cases, for example in the Dakin oxidation of an
aldehyde to a phenol, interfacial formation of a peroxide anion is suggested to be the result of the strong electric field at
the interface. Evidence is provided that the water radical cation, a species with a short lifetime but a relatively high
concentration in bulk water, may be involved in arylsulfone to sulfonic acid oxidation. This is also shown to be an
interfacial reaction favored by traces of water. Other accelerated reactions of wide interest include the interfacial
formation of amides in aqueous environments, a process relevant to the origin of life, as well as the generation of
catalysts by electrolytic spray (electrocorrosion) with deposition onto a suitable surface as nanoparticles. Control over
whether reaction acceleration occurs can be achieved by decreasing droplet size, increasing droplet transfer distance to
the mass spectrometry, and selecting an appropriate solvent.

Please explain why your abstract is innovative for mass spectrometry?

Reactions in confined volume solutions are accelerated at the interface due to partial solvation and strong electric fields.
This allows high-throughput reaction screening and small-scale organic and nanoparticle synthesis.

Co-authors:

Nicolas Morato, Purdue University

Lingqi Qiu, Purdue University
Kai-Hung Huang, Purdue University

21
Figure 1. Potential energy diagram illustrating the partial solvation mechanism.

Figure 2. Proposed mechanism of CO2 reaction at microdroplet interface.

22
THE FIRST MASS ANALYZER IN THE NETHERLANDS
Abstract ID: 5

Presenting author: Albert Heck, Utrecht University

Introduction

Pieter Zeeman was a Dutch physicist who at the University of Amsterdam discovered how in a static magnetic field a
beam of light splits into several components. This discoverey, now termed the Zeeman effect, was awarded a Nobel
Prize in 1902 for Zeeman and Lorentz. Zeeman was well known for his research and befriended with many of the giants
of science of the early 20th century. In 1920 for instance Albert Einstein and Paul Ehrenfest visited his laboratory in
Amsterdam.Zeeman did also get interested in, and inspired by, the work of Goldstein, Wien, Thomson and Aston on
cathode rays.

Methods

As we all know, Thomson had shown in 1909 that if a beam of charged particles is passed through a combined magnetic
and electric field, the mass of the nucleus can be determined from the deflection of the beam. At that time, it was found
that the so-called chemical elements are usually not singular but could also consist of a mixture or number of atoms,
whose masses are related almost as whole numbers. This investigation of the so-called isotopes had been brought to
great perfection especially by the work of Aston.

Preliminary data (results)

Zeeman’s PhD student Johannes de Gier constructed in between 1930-1934 an instrument based on Thomson's
channel beam method (parabolic method) and used it to determine the masses of a number of elements and discovered
some unknown isotopes, such as the argon isotope 38Ar and the nickel isotope 64Ni.Moreover, by obtaining deuterium
enriched hydrogen gas from Leiden, they described correctly the mechanisms leading to the formation of rays with
masses of 5 and 6 Da, interpreted earlier on by Thomson as He and HeH, but annotated by de Gier and Zeeman as
HD2+ and D3+. In that sense they postulated some of the first ion molecule reactions.Based on the thesis of Johannes
de Gier that I found a few years ago in the basement of a bookshop in Amsterdam I here describe and highlight some
applications of what likely was the first mass analyzer in the Netherlands.

Please explain why your abstract is innovative for mass spectrometry?

I here describe and highlight some applications of what likely was the first mass analyzer in the Netherlands.

23
24
EXCITED STATE N-ATOMS TRANSFORM AROMATIC HYDROCARBONS
INTO N HETEROCYCLES IN A LOW-TEMPERATURE PLASMA
Abstract ID: 94

Presenting author: Renato Zenobi, Department Chemistry and Applied Biosciences, ETH Zurich, 8093 Zurich,
Switzerland

Introduction

Recently, direct replacement of a carbon atom in aromatic compounds by nitrogen has been reported to occur in a low
temperature plasma. Low temperature plasmas generate reactive ions, neutrals, and excited species that react out of
thermodynamic equilibrium and can lead to chemically surprising products. Understanding what processes in nitrogen-
based low-temperature plasmas lead to nitrogen substitution reactions in aromatic compounds may help increase our
understanding of space chemistry and give clues to theories on the origin of life.

Methods

The formation of N-heterocycles directly from aromatic hydrocarbons was observed in a nitrogen dielectric barrier
discharge ionization source (DBDI) coupled online to an orbitrap mass spectrometer. The structure of the formed N-
heterocycle was confirmed by collision-induced dissociation (CID) and ionization of reference compounds. The
mechanism of the nitrogen replacement reaction was investigated by variable electron and neutral density attachment
mass spectrometry (VENDAMS) in a flowing afterglow Langmuir probe (FALP) at high discharge densities. The proposed
mechanism for the formation was supplemented by density functional theory (DFT) calculations.

Preliminary data (results)

The primary reactant studied in detail was toluene (C7H8), which after passing the nitrogen discharge plasma formed two
major products, one of a nitrogen replacement reaction resulting in picoline (C 6H8N+), and a nitrogenation (C7H8N+)
product. Several other minor products were detected, including pyridine (C 5H6N+) and dimerization products. The
nitrogen replacement reaction product increased in relative intensity with increasing plasma operating voltage (from 2.4
to 3.4 kVpp) of the discharge plasma and increasing concentration of the infused gas-phase toluene (5 to 20 ppm mol/mol
toluene in N2). The primary reactive nitrogen ions formed in the discharge plasma were identified as N +, N2+, N3+, and
N4+ with virtually no negative ions formed. The same trend in reactivity was observed for a series of substituted aromatic
compounds, each resulting in a nitrogen replacement product while retaining their original substituent. The products
resulting from the nitrogen replacement reaction of toluene could be reproduced in a microwave discharge plasma and
studied in the FALP. An increase in the electron density of the plasma lead to an increase of the nitrogen replacement
reaction products.

Please explain why your abstract is innovative for mass spectrometry?

This is the first mechanistic investigation detailing the direct formation of N-heterocycles from aromatic hydrocarbons in
low temperature plasmas.

Co-authors:

Nicholas S. Shuman, Air Force Research Laboratory, Space Vehicles Directorate, Kirtland AFB, New Mexico 87117,
USA
Bryan A. Long, Air Force Research Laboratory, Space Vehicles Directorate, Kirtland AFB, New Mexico 87117, USA
Robin Kämpf, Department Chemistry and Applied Biosciences, ETH Zurich, 8093 Zurich, Switzerland
Luzia Gyr, Leibniz Institute for Natural Product Research and Infection Biology – Hans Knöll Institute, Adolf-Reichwein-
Straße 23, 07745 Jena, Germany
Albert A. Viggiano, Air Force Research Laboratory, Space Vehicles Directorate, Kirtland AFB, New Mexico 87117, USA

25
SYNTHESIZING NEW MOLECULES IN THE CONDENSED PHASE USING
GASEOUS MOLECULAR FRAGMENT IONS
Abstract ID: 113

Presenting author: Jonas Warneke, Wilhelm-Ostwald-Institut für Physikalische und Theoretische Chemie,
Universität Leipzig, 04103, Leipzig, Germany, Leibniz-Institut für Oberflächenmodifizierung e.V. (IOM),
Permoserstr. 15, D-04318, Leipzig, Germany.

Introduction

Highly reactive fragment ions generated within mass spectrometers are commonly used for analytics and fundamental
gas phase ion chemistry studies. We intend to use these potent reagents as chemical building blocks for the synthesis of
new compounds in the condensed phase. Ion soft-landing (also called preparative mass spectrometry) enables the
gentle deposition of mass selected gaseous fragment ions onto surfaces covered with a reagent. The products are
accumulated on the surface and the molecular structures are determined using ex-situ analytical methods.

Methods

We use an electrospray-based dual polarity soft-landing instrument which contains a collision cell for efficient fragment
ion formation and a resolving quadrupole for mass selection, see Figure 1. [1] Ions are deposited on gold covered with
an alkane thiol self-assembled monolayer (SAM). The generated material layers are analyzed using ex-situ methods,
including Liquid Extraction Surface Analysis (LESA), Infrared (IR) and Nuclear Magnetic Resonance (NMR)
spectroscopy.

Preliminary data (results)

The generation of new molecules on surfaces using reactive fragments is demonstrated with ions of type
[B12X11]− (X=halogen, CN). Collision induced dissociation of highly stable [B12X12]2− dianions results in the formation of
[B12X11]− fragments which are able to bind noble gases at room temperature. [2] The reaction of [B 12X11]− with saturated
alkanes results in the formation of a C-B bond in the gas phase. The reaction mechanism is unraveled using
spectroscopic and computational methods. [3] Then, the reaction is applied to the condensed phase by soft-landing
[B12X11]− ions into layers containing different organic reagents, including peptides. It is shown that [B12X11]− binds
selectively to the alkyl moiety of the organic molecules. In addition, we show that the binding of [B 12X11]− to other anions
on the surface occurs resulting in multiply charged ions within the surface layers, see Figure 2. [4]

[1] H. Y. Samayoa-Oviedo et al.: “Design and Performance of a Soft-Landing Instrument for Fragment Ion
Deposition.” Anal. Chem. 2021, 93, 14489-14496.

[2] M. Mayer et al. “Rational Design of an Argon-Binding Superelectrophilic Anion. Proc. Natl. Acad. Sci. U.S.A. 2019,
116, 8167–8172.”

[3] J. Warneke et al.: “Direct Functionalization of C−H Bonds by Electrophilic Anions.” Proc. Natl. Acad. Sci. U.S.A. 2020,
117, 23374–23379.

[4] F. Yang et al. „Anion-Anion Chemistry with Mass-Selected Molecular Fragments on Surfaces.”Angew. Chem. Int.
Ed. 2021, 60, 24910-24914.

Please explain why your abstract is innovative for mass spectrometry?

Molecular fragment ions generated within mass spectrometers are used to synthesize new molecular structures in
condensed surface layers using ion soft-landing.

26
Figure 1: Scheme of the fragment ion deposition instrument.

Figure 2: Formation of the trianions [B24I23]3-.

27
CYCLIC PEPTIDE PROTOMERS DETECTION IN THE GAS PHASE: IMPACT
ON CCS MEASUREMENT AND FRAGMENTATION PATTERNS
Abstract ID: 228

Presenting author: Edwin De Pauw, Mass Spectrometry Laboratory, ULiege

Introduction

With the recent improvements in ion mobility resolution, it is now possible to separate small protomeric tautomers, called
protomers. In larger molecules above 1,000 Da such as peptides, protomers have rarely been observed, mainly because
the amino acid sequence often contains basic amino acids, that represent energetically favorable sites for protonation.
We explored the potential presence of protomers in the case of surfactin, a small lipopeptide with no basic site.

Methods

Density functional theoretical calculations were performed to evaluate the number and the position of energetically
favorable protonation sites in the gas phase, in positive ionization mode. Energy resolved breakdown curves were
performed after IMS separation to explore specific fragmentation pathways

Preliminary data (results)

Experimentally, at least three near-resolved IM peaks were observed in positive ionization mode while only one was
detected in negative ionization mode as only deprotonation site is available. These results were in good agreement with
the DFT predictions. CID breakdown curves analysis after IM separation showed different inflection points (CE 50)
suggesting that different intra-molecular interactions were implied in the stabilization of the structures of surfactin.The
fragments ratio observed after collision induced fragmentation were also different, suggesting different ring opening
localizations. All these observations support the presence of protomers on the cyclic peptide moieties of the surfactin.
These data strongly suggest that protomeric tautomerism can be observed on molecules above 1,000 Da if the IM
resolving power is sufficient. It also supports the idea that the proton localization involves a change in the 3D structure
that can affect the experimental CCS and the fragmentation channels of such peptides.

Please explain why your abstract is innovative for mass spectrometry?

Fisrt IMS analysis of protonation forms of cyclic lipopeptides and the CCS specific energy resolved MS/MS

Co-authors:

Christopher Kune, Mass Spectrometry Laboratory, ULiege

Philippe Massonnet, Mass Spectrometry Laboratory, ULiege
Johann Far, Mass Spectrometry Laboratory, ULiege
Marc Ongena, Gembloux Agro-Bio Tech, Uliege
Gauthier Eppe, Mass Spectrometry Laboratory, ULiege
Loic Quinton, Mass Spectrometry Laboratory, ULiege
Andrea McCann, Mass Spectrometry Laboratory, ULiege

28
CCS distribution of protonated surfactin

UNRAVELLING THE PEPTIDES’ AGGREGATION MECHANISM: THE

CHALLENGE OF IM-MS INSTRUMENTATION TO PROBE
HETEROGENEOUS AND FRAGILE PROCESSES.
Abstract ID: 164

Presenting author: Agathe Depraz Depland, Division of BioAnalytical Chemistry, AIMMS Amsterdam Institute of
Molecular and Life Sciences, Vrije Universiteit Amsterdam, De Boelelaan 1108, 1081 HV Amsterdam, The
Netherlands

Introduction

Neurodegenerative diseases, such as Alzheimer’s or Parkinson’s diseases, are directly correlated with the aging
process. On a molecular level, this is translated by the development of protein and peptide aggregates. However, the
mechanism of aggregation [Figure 1], and especially the transition from soluble oligomers to insoluble fibrils is still
unclear. This is particularly important as neurodegenerative toxicity originates from these conformational intermediates
that are formed in the early stages of the aggregation pathway. The challenge here is to obtain structural information of
the elusive intermediate conformers from their complex and heterogeneous environment. Therefore, we are developing
multi-dimensional technique based on ion mobility mass spectrometry (IM-MS).

Methods

In our MS-LaserLab team, we combine droplets-based microfluidics electrospray ionisation, mass and ion mobility
spectrometries and InfraRed spectroscopy to reveal the identity, the structure and the temporal evolution of these key
intermediates in a single experiment. Here, we are highlighting the heart of our experiment, by showing how IM-MS can
be used to follow oligomer formation [Figure 1]. Typically, we track the linear increase of both mass and charge of the
monomeric ion, over time points. To do so, complex instrumental parameters are optimised to allow the study of intact
aggregates and ensure their transmission towards detection [Figure 2].

Preliminary data (results)

In our studies, we initially focus on the well-characterized TDP-43 protein, found responsible for the aggregation process
in the case of amyotrophic lateral sclerosis (ALS) disease. Specific segments of this TDP-43 protein, such as the
cytotoxic peptide M307-N319 (wild type peptide (WT), two ALS-related peptide mutants (A315E and A315T), and one
non-toxic mutant (G314V)) have been studied in detail by Laos et al. using a drift tube ion mobility coupled to MS. Here,

29
we investigate the suitability and application of travelling wave ion mobility mass spectrometry, using our modified Photo-
Synapt, to probe the aggregation process of these TDP-43 segments.

Preliminary results were obtained on these TDP-43 segments systems using trapped ion mobility spectrometry,
displaying a unique distribution of oligomers testifying of the evolution of aggregation. With the workflow in place, we
observe the temporal evolution of aggregation, where the mobilograms recorded at longer time points show the
appearance of peaks at lower mobility values, translating the presence of larger aggregates.

Subsequently, we employ our developed methods [Figure 1 and 2] to study the aggregation mechanism of α-synuclein, a
protein implicated in Parkinson’s disease. Here, we focus on a segment, namely 68GAVVTGVTAVA78, because of its
critical role in α-synuclein amyloid formation and cytotoxicity. We will probe the structure and structural changes of the
aggregation of this segment and two other segments from the pre-NAC domain, both related to early-onset Parkinson’s
disease. This will help us to understand which key steps of that process are altered.

Please explain why your abstract is innovative for mass spectrometry?

The use of ion mobility mass spectrometry to study the toxic intermediate oligomers and follow their development over
time for diverse cases of neurodegenerative diseases.

Co-authors:

Sjors Bakels, Division of BioAnalytical Chemistry, AIMMS Amsterdam Institute of Molecular and Life Sciences, Vrije
Universiteit Amsterdam, De Boelelaan 1108, 1081 HV Amsterdam, The Netherlands
Iuliia Stroganova, Division of BioAnalytical Chemistry, AIMMS Amsterdam Institute of Molecular and Life Sciences, Vrije
Universiteit Amsterdam, De Boelelaan 1108, 1081 HV Amsterdam, The Netherlands
Anouk M Rijs, Division of BioAnalytical Chemistry, AIMMS Amsterdam Institute of Molecular and Life Sciences, Vrije
Universiteit Amsterdam, De Boelelaan 1108, 1081 HV Amsterdam, The Netherlands

Simplified scheme of aggregation process and the associated analytical method.

30
IM-MS instrumental parameters effect over mobilograms' shape.

31
Session: IM-2 High Resolution MS - Session A

TOWARDS SPATIAL AND CELL RESOLVED OMICS USING ADVANCED

FTMS APPROACHES
Abstract ID: 571

Presenting author: Ljiljana Paša-Tolić, Pacific Northwest National Laboratory

Introduction

Advances in Fourier transform mass spectrometry (FTMS) are revolutionizing the omics field. Common bulk omic
analyses report on large cell population averages, thus masking the presence of subpopulations, and individual cellular
variations. Cell and spatial measurements are needed to decipher mechanisms that give rise to phenotypic diversity and
link molecular-scale information to systems-level understanding of biology. Metabolomics and proteomics are lagging
behind genomics and transcriptomics due to considerable technical challenges. Low measurement throughput, as well
as limited size and volume of single cells, together with the low copy numbers of certain analytes of interest make their
characterization challenging. Herein, we highlight key advantages offered by advanced FTMS instrumentation for high-
throughput mapping of metabolites and complex proteoforms within tissue at near cellular resolution.

Methods

Fresh frozen tissue sections were analyzed using several MS imaging (MSI) approaches, including MALDI (Spectroglyph
LLC), and/or nanoDESI. MSI analyses were performed using a Thermo Scientific Q-Exactive HF Orbitrap MS upgraded
with UHMR boards and/or a custom FTICR mass spectrometer equipped with a 21 Tesla superconducting magnet.
Annotation of proteoforms was primarily based upon high-resolution accurate mass measurements, while metabolomics
and lipidomics annotations (METASPACE) were confirmed via LC-MS/MS analysis. Laser capture microdissection (LCM)
was used to cut tissue microstructures from serial sections for nanoPOTS top-down proteomics. Data were processed
with TopPIC and TDPortal, then manually examined.

Preliminary data (results)

Herein, we evaluate performance of advanced FTMS instrumentation for various MSI modalities and applications. The
21T FTICR MS demonstrated unparalleled ultra-high mass resolution in lower m/z ranges as illustrated by isotopic fine
structure measurements of metabolites and lipids across various tissue sections, further boosted by absorption mode
data processing (aFT). However, MALDI UHMR platform yielded enhanced performance for proteoform imaging
presumably due to slower reduction in resolution with increasing m/z with Orbitrap MS compared to FTICR MS.
Specifically, our novel UHMR MSI platform enabled routine high mass accuracy measurements allowing for confident
annotation of various proteoforms, and specifically core histones, within tissues. Enhanced proteoform mapping at near
cellular spatial resolution was facilitated by a significant boost in throughput, as high mass resolution was maintained
with 512 ms long transients (35k at m/z 11,306) within an extended mass range. This compares favorably to previously
reported (and simulated) FTICR MSI of proteoforms, providing a 6-fold improvement in duty cycle. Several key histone
proteoforms were observed with differential abundance across kidney tissue indicating differences in underlying
processes and serving as potential biomarkers, including proteoforms specifically associated with different forms of
cancer and trauma. Integration with alternative spatial proteomics approaches (i.e., nanoPOTS) facilitated confident
identification of post-translation modifications (PTMs) on proteoforms through experimental database matching. Similarly,
implementation of alterative tandem MS methods (e.g., ultraviolet photodissociation, UVPD) allowed for confident
characterization of proteoforms directly from tissue.

Please explain why your abstract is innovative for mass spectrometry?

Advanced FTMS instrumentation and methods enhance mass and spatial resolution for metabolite and proteoform
imaging.

Co-authors:

Kevin J. Zemaitis, Pacific Northwest National Laboratory

Dušan Veličković, Pacific Northwest National Laboratory
James M. Fulcher, Pacific Northwest National Laboratory
Yen-Chen Liao, Pacific Northwest National Laboratory
Sarah M. Williams, Pacific Northwest National Laboratory
David J. Degnan, Pacific Northwest National Laboratory

32
Lisa M. Bramer, Pacific Northwest National Laboratory
Gregory W. Vandergrift, Pacific Northwest National Laboratory
William R. Kew, Pacific Northwest National Laboratory
Mowei Zhou, Pacific Northwest National Laboratory
Ying Zhu, Pacific Northwest National Laboratory

33
HIGH-ACCURACY MS OF EXOTIC ATOMIC NUCLEI BY PHASE-IMAGING
ION-CYCLOTRON-RESONANCE AND MULTI-REFLECTION TIME-OF-
FLIGHT MS AT SHIPTAP AND ISOLTRAP
Abstract ID: 498

Presenting author: Lutz Schweikhard, Inst. of Physics, Univ. Greifswald

Introduction

Mass defects and excitation energies are important nuclear properties which - by comparison with calculations based on
modern nuclear theory - reveal the structure of atomic nuclei. Reaching the nuclear landscape's boundaries of existence
of is a major challenge of modern experimental physics. The high-accuracy mass-spectrometry setups ISOLTRAP and
SHIPTRAP, located at CERN/Geneva and GSI/Darmstadt, respectively, allow the determination of masses of exotic
nuclei with half-lives down to well below a second. The ions are delivered by the corresponding accelerator facilities in
only tiny amounts, such that the event rates are as low as just a few ion counts per hour. Nevertheless, methods have
been developed for mass resolving powers exceeding 10 million and relative uncertainties in the 10 -8 region and below.

Methods

The recent measurements have been enabled by the continuous development and implementation of highly sensitive
mass spectrometry (MS) methods and devices for "online" application, in particular Phase-Imaging Ion-Cyclotron-
Resonance MS (PI-ICR MS, see S. Eliseev et. al., Phys. Rev. Lett. 110 (2013) 082501 and Appl. Phys. B 114 (2014)
107) and Multi-Reflection Time-of-Flight MS (MR-ToF MS, see R.N. Wolf et al., Nucl. Instrum. Meth. A 686 (2012) 82 and
Int. J. Mass Spectrom. 349-350 (2013) 123). While PI-ICR MS is based on Penning traps highest mass resolving powers
and accuracies, MR-ToF MS uses electrostatic trapping to reach for shorter half-lives.

Preliminary data (results)

SHIPTRAP specializes in the study of very heavy atomic nuclei, i.e. elements beyond uranium. After the pioneering
measurements on nobelium (Z=102) isotopes (M. Block et al., Nature 463 (2010) 785 and Lawrencium (Z=103) (E.
Minaya Ramirez et al., Science 337 (2012) 1207) by use of the older method of Time-of-Flight Ion Cyclotron Resonance
MS, the studies have recently been extended wit PI-ICR MS and now reached the masses of the ground state and even
of low-lying metastable states of the rutherfordium (Z=104) isotope 257Rf. In addition, a precise measurement of the
dubnium (Z=105) isotope 258Db could be performed.

At ISOLTRAP, two Penning traps (M. Mukherjee et al., Eur. Phys. J. A 35 (2008) 1) have been complemented by an MR-
ToF MS section about a decade ago, for ion separation/purification (R.N. Wolf et al., Phys. Rev. Lett. 110 (2013) 041101)
as well as actual mass measurements (F. Wienholtz et al., Nature 498 (2013) 346). These experiments have continued
with recent highlights in the study of neutron-deficient indium isotopes in the vicinity of the tin isotope 100Sn – a doubly-
magic nuclide that is also the heaviest species with equal numbers of protons and neutrons. Published results include
the study down to the very light 99In (M. Mougeot et al., Nat. Phys. 12 (2021) 1099) and even more recent are mass
measurements of excited nuclear states including this isotope.

Please explain why your abstract is innovative for mass spectrometry?

Mass defects and nuclear excitation energies E = (m i-mg) c2 from mass doublets of isomer and ground-state masses
reveal the structure of atomic nuclei at the boundary of existence.

34
PI-ICR MS schematic [S. Eliseev et al., Appl.Phys.B-2014]

MR-ToF MS of ISOLTRAP [R.N. Wolf et al., IJMS-2013]

35
INSIGHTS INTO SURPRISINGLY BORING LIVES OF HIGH-MASS IONS BY
CHASING SINGLE PARTICLES USING SEGMENTED FOURIER
TRANSFORM.
Abstract ID: 690

Presenting author: Tobias P. Wörner, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for
Biomolecular Research and Utrecht Institute for Pharmaceutical Sciences, University of Utrecht, Padualaan 8,
3584 CH Utrecht, The Netherlands, Netherlands Proteomics Center, Padualaan 8, 3584 CH Utrecht, The
Netherlands, Thermo Fisher Scientific (Bremen), Bremen, Germany

Introduction

Native mass spectrometry (MS) analysis of novel biotherapeutics, like gene therapy vectors, vaccines and viruses, is
gaining steadily in importance. Single particle approaches and charge detection MS (CDMS) are ideally suited to tackle
the challenges associated with the characterization of these particle classes and have recently been adopted for
Orbitrap-based MS as well. These techniques offer a unique opportunity to deepen our understanding of high-mass ion
behavior in the Orbitrap. By closely monitoring individual ions in the trap, we were able to elucidate the distinctive stable
behavior of these analytes. The acquired data will ultimately lead to improved and robust experimental designs
facilitating the analysis of very large biomolecules and allow a broader adaption in academia and industry.

Methods

We performed here an in-depth analysis of individual ions in the Orbitrap™ mass analyzer at transient times of several
seconds. The analytes included therapeutic antibodies, gene therapy vectors, virus capsids, as well as intact viruses,
ranging in mass from 150 kDa to 9 MDa. We use a segmented Fourier Transform (FT) approach of the recorded
transient signal, which allows to monitor processes occurring during the transient acquisition over a wide pressure range,
such as ions decay, charge-stripping, and neutral losses. We linked these processes to specific peak shape artifacts,
which frequently occur and can impair single particle CDMS experiments.

Preliminary data (results)

We demonstrate that ions above one megadalton do not decay in the Orbitrap like smaller ions, although they do
experience multiple collisions with neutrals. Instead of fragmenting, these ions release the received energy through
neutral losses (and occasional charge stripping events) which can be temporally resolved through the use of a
segmented FT approach. These neutral losses and the accompanying frequency drifts correlate directly with the
background pressure in the instrument. Detailed analysis allows to establish that along with the actual collisions during
the transient acquisition, the injection from the C-trap is an equally strong contributor for activation.

The observed frequency drifts cause broadened or split FT peaks, which limit the sampling rate and achievable CDMS
resolution. We developed several strategies to reduce peak splitting and improve the sampling rate by more than 20-fold,
making Orbitrap based CDMS more robust.

The remarkable stability of these large ions and the possibility to avoid FT artifacts allowed us to explore the benefits of
longer transient times which, improves the mass resolution in CDMS experiments by almost two-fold (compared to
standard 1s transients). Furthermore, we could for the first time directly observe frequencies of the ions’ radial
oscillations via their modulation of the axial frequency. The determination of these radial frequencies offers the possibility
to identify unstable radial orbits and correct singe ion intensities for differing radial distances. This will be important for
reaching a charge resolution of less than one elementary charge in Orbitrap based CDMS.

Please explain why your abstract is innovative for mass spectrometry?

Chasing single particles with charge detection MS reveals longevity of high-mass ions, leading to improved strategies to
analyze large biotherapeutics.

Co-authors:

Konstantin Aizikov, Thermo Fisher Scientific (Bremen), Bremen, Germany

Joost Snijder, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research and Utrecht
Institute for Pharmaceutical Sciences, University of Utrecht, Padualaan 8, 3584 CH Utrecht, The Netherlands,
Netherlands Proteomics Center, Padualaan 8, 3584 CH Utrecht, The Netherlands
Kyle L. Fort, Thermo Fisher Scientific (Bremen), Bremen, Germany

36
Alexander A. Makarov, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research and
Utrecht Institute for Pharmaceutical Sciences, University of Utrecht, Padualaan 8, 3584 CH Utrecht, The Netherlands,
Thermo Fisher Scientific (Bremen), Bremen, Germany
Albert J.R. Heck, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research and
Utrecht Institute for Pharmaceutical Sciences, University of Utrecht, Padualaan 8, 3584 CH Utrecht, The Netherlands,
Netherlands Proteomics Center, Padualaan 8, 3584 CH Utrecht, The Netherlands

37
APPLICATION OF OCULAR APPROACH ON 15 T SOLARIX XR FOR
BITUMEN ANALYSIS
Abstract ID: 601

Presenting author: Benedict Gannon, University of Warwick

Introduction

Bitumen, a highly complex mixture, is a key component in 95% of the UK’s road surfacing. Bitumen contains
polyaromatic hydrocarbons (PAH), harmful materials released into the environment by road dust from road wearing and
in surrounding waters from runoff. Characterisation of bitumen by high resolution mass spectrometry is essential to
identifying any potential contamination of the environment. To achieve this in such a complex mixture by Fourier
transform ion cyclotron resonance mass spectrometry (FT-ICR MS), greater dynamic range and higher resolution must
be obtained. To achieve this, an improved stitching technique is applied to large m/z windows of width 100 obtained by
15 T FT-ICR using a dynamically harmonised cell.

Methods

A bitumen sample (0.05 mg/mL concentration) was analysed by direct infusion to a Bruker 15 T solariX XR FT-ICR MS.
Using a quadrupole, eleven 100 m/z wide windows with 10 m/z overlap between m/z 250 and 1250 were isolated. An
operation at constant ultrahigh resolution (OCULAR) approach was employed. Briefly, low mass cut-off for detection was
raised for each subsequent window resulting in a flat resolution of >1,000,000 across the mass range. Windows were
phased and calibrated, stitched together with in-house software Rhapso, assigned using Composer (Sierra Analytics),
then visualised and further analysed using in-house software.

Preliminary data (results)

Preliminary analyses of bitumen samples by electrospray ionisation (ESI) found them to be extremely complex (>38
peaks per nominal m/z)exhibiting high numbers of heteroatoms. To resolve mass splits of sub 1.6 mDa at m/z 650, much
higher resolution and dynamic range was required. The dynamic range obtained can be increased by utilising quadrupole
isolation due to reduction of space-charge effects. Traditional stitching experiments utilise this feature to enhance the
dynamic range across the whole mass range. Recently, an advancement of traditional stitching experiment (‘OCULAR’)
was demonstrated with a 12 T solariX, enabling an increase in dynamic range via isolation windows and an increase in
resolution by increasing the low mass cut-off. By sequentially increasing the low mass cut-off for each isolation window, a
constant resolution can be obtained across the mass range. Here we have utilised the OCULAR approach, showing
application to a 15 T instrument with a dynamically harmonised cell. We employed 100 m/z wide windows, as such only
eleven windows were required to cover the whole range from m/z 250 to 1250 which in turn reduced experiment time
significantly.

The full stitched spectrum resulted in an average resolution of 1,000,000 across the whole mass range, increasing 1.7x
with application of absorption mode, with >6,500 peaks per window. Preliminary analysis revealed nitrogen, oxygen, and
sulfur containing compounds. Further applications using atmospheric pressure photoionisation (APPI) will enable
ionisation of species such as PAHs, enhancing our ability to fully characterise this complex sample.

Please explain why your abstract is innovative for mass spectrometry?

Operating at constant ultrahigh resolution with broad isolation windows on a 15 T solariX XR with a dynamically
harmonised cell.

Co-authors:

Diana Palacio Lozano , University of Warwick

Hugh Jones, University of Warwick
Mark Barrow, University of Warwick

38
ANALYSIS OF SUPERCOMPLEX MIXTURES BY MULTI-STEP LIQUID
CHROMATOGRAPHY AND ONLINE HIGH-RESOLUTION MASS
SPECTROMETRY
Abstract ID: 660

Presenting author: Jens Dreschmann, Max-Planck-Institut für Kohlenforschung

Introduction

The understanding of structural entities in very complex samples like crude oil is not mundane. Although it serves
multiple purposes and forms the basis for many products the global energy consumption has nearly exhausted
conventional oil resources. This led to the exploitation of heavier feedstocks with higher amount of low soluble
compounds (called asphaltenes), which cause problems during their processing. Thus, a deeper understanding of these
materials is required in order to comply with quality requirements of refined products. The challenging task of analyzing
heavy crude oil makes the application of separation techniques inevitable. However, the detection still needs to be done
with high accuracy and resolution. Ultimately, high-resolution mass spectrometry offers both features and, additionally,
provides molecular information about individual compounds.

Methods

During this study, a two-dimensional chromatographic approach was developed. Chromatography was generally
performed using an UltiMate 3000 HPLC system (Thermo Fisher Scientific, Bremen, Germany). As first dimension, a
size-exclusion chromatography (SEC) was utilized for fraction collection and separation of an asphaltene fraction. In the
second dimension, the collected fractions were separated by a ligand-exchange chromatography (argentation
chromatography). The second dimension is coupled to a 7T LTQ FT Ultra ICR mass spectrometer (Thermo Fisher,
Bremen, Germany) to perform online measurements. This setup enables the opportunity to record full scans as well as
fragmentation experiments using APPI as ionization technique.

Preliminary data (results)

The obtained fractions of the SEC indicate a bulk separation of the previous asphaltene sample. Nevertheless, these
fractions feature a decreasing order in their averaged molecular weights, which can be applied as measure for the size.
However, the amount of ions per nominal mass in these fractions still exceeds the isolation capabilities of the technical
setup and impedes reasonable fragmentation studies. The argentation chromatography eliminates this problem and
provides rather clean isolation windows. As it turned out, the silver-(I)-mercaptopropyl silica based stationary phase is
also capable of separating NOS-compounds according to their heteroatom content besides separation via the strength of
the π-system. Furthermore, the direct coupling to HRMS analysis enables the possibility to perform online fragmentation
experiments (MS2). The fragmentation studies successfully revealed structural features of individual compounds.

Please explain why your abstract is innovative for mass spectrometry?

At this early stage, the investigation of asphaltene fractions already provides remarkable insights into single compound
structures and underlines the methods potential for the investigation of mixtures with unmatched complexity.

Co-authors:

Wolfgang Schrader, Max-Planck-Institut für Kohlenforschung

39
SUPER-RESOLUTIVE GENETIC ALGORITHM FOR IMPROVED FT-ICR MS
RESOLUTION
Abstract ID: 688

Presenting author: Marc Haegelin, Miniaturization for Synthesis, Analysis & Proteomics, USR 3290, CNRS,
University of Lille.

Introduction

Finding the amplitudes, frequencies and phases of all sines from a sum of exponentially damped sines waves which is
called harmonic inversion is a major step on FT-ICR MS, Orbitrap MS or NMR instruments. Classically this step is
performed by Discrete Fourier Transform (DFT), which has several limitations, the main one being that the resolution is
proportional to transient duration. To solve these issues, we developed a GPU-parallelized genetic algorithm called
Sinus_it, which performs the least square fitting of the transient. Our algorithm affords better parameters estimation than
DFT, in particular on frequencies and amplitudes, paving the way to optimal super-resolution, better chemical formula
identification and more accurate quantitation on mass spectra.

Methods

Experimental substance P and glutathione transients were recorded on a Bruker SolariX 9.4 Tesla FTICR mass
spectrometer. Simulated transients were generated with noise, damping and shift of frequency and phase. Sinus_it was
developed using CUDA Toolkit 11.1 in C++ language for maximal hardware control and speed and then executed on a 2
× 16 cores 2.1 Ghz Intel Xeon Dell Precision 7920 endowed with an NVIDIA RTX 2080 Ti card. The implementation has
two modes, the coarse mode, which finds the main isotopes, and the fine mode which works on a single isotope for
extracting fine isotopic structures.

Preliminary data (results)

First a band-pass Butterworth filter recursively selects the frequency range of interest determined by DFT. Then Sinus_it
is runs on one frequency of this range. In the first place, coarse mode identifies the main isotopes and gives the
amplitudes, frequencies and phases of the sines. In order to get accurate information on the spectrum, each frequency
obtained by a second band-pass filtering and submitted to Sinus_it in fine mode to get its fine isotopic structure. Sinus_it
automatically computes the number of peaks in the band of interest without a priori data. The 7 first isotopes of simulated
substance P were obtained on only 8k points by Sinus_it whereas DFT requires 32k. On experimental substance P
Sinus_it also resolved the 5 first isotopes on 8k points while DFT needed 32k points. When white noise was increased in
simulated substance P transients, peak identification was possible even when the white noise was equal to the amplitude
of the smallest peak. We also extracted the fine structures of the first, second and third isotope of substance P with twice
less points (1M points) than necessary for DFT (2M points). The low number of points required by Sinus_it allowed for
measuring the frequency shifts along the transient, which was found to be linear, the phase shifts, and the correlation
between the two. Finally, Sinus_it was optimized for narrowband acquisition. From a narrow 64k points signal
corresponding to a 16M broadband one the same peaks were obtained with the same precision on 128 points only.

Please explain why your abstract is innovative for mass spectrometry?

We provide a new super-resolutive signal processing technique useful for FT-ICR MS, Orbitrap MS or NMR, among
many other possible fields of application, even at low SNR.

Co-authors:

Ulviya Abdulkarimova, Azerbaijan State University of Oil and Gas Industry, ICube, Complex System and Translational
Bioinformatics, UMR CNRS 7357, University of Strasbourg
Pierre Collet, ICube, Complex System and Translational Bioinformatics, UMR CNRS 7357, University of Strasbourg
Christian Rolando, Miniaturization for Synthesis, Analysis & Proteomics, USR 3290, CNRS, University of Lille., Shrieking
Sixties, 1-3 Allée Lavoisier

40
Sinus_it workflow

Five super-resolved peaks from 8K experimental substance P transient

41
Session: LS-1 Translational MS – Clinical and Liquid Biopsies

TWO SIDES OF PRECISION MEDICINE: HEALTH SURVEILLANCE

Abstract ID: 795

Presenting author: Jennifer Van Eyk, Cedars-Sinai Medical Center

Introduction

Precision medicine is driven by the concept that an individual’s Omic (proteome) signature will provide a physician with a
clinically actionable diagnosis and subsequent mechanistic therapeutic route. Continuous monitoring of circulating
protein biomarkers is an unmet opportunity to assess an individual’s health in real-time as part of disease management
and preventive clinical care. Simultaneously, there is a need to increase the breadth of therapeutics available and to
identify the appropriate therapy(ies) for each individual based on analysis of their cellular avatar.

Methods

Enablement requires automation, scalability, and a move to normalization of proteomic data while balancing cost
effectiveness in the discovery and clinical domains.

Preliminary data (results)

To achieve this and with the goal for release into a CLIA clinical laboratory, we have developed and optimized a 60
protein MRM health surveillance assay, of which 29 proteins have FDA-approved or LDT assays. The assay quantifies
118 endogenous peptides and their matching heavy stable isotopic standards (480 transitions/assay) over a 24 min
gradient with performance criteria of >10 data point/peak, linearity of r2 > 0.9, matrix dependent recovery of > 80%
and process stability of >90%. In large cohort studies, the assay consistently has < 1% carryover and ~8% CV for
the N15 labelled intact C-Reactive Protein spiked as a digestion/process control in every sample. Concurrently, we are
determining whether equivalent accuracy can be achieved within a broader protein discovery hybrid assay which would
allow a subset of proteins to be released for clinical decision making while providing an opportunity to mine for additional
biomarkers across diverse conditions. Simultaneously, we are developing proteomic processes for cell-centric methods
(single or organoids) to rapidly screen response to therapeutics perturbations.

Please explain why your abstract is innovative for mass spectrometry?

Large scale automation for plasma and cell based assays with normalization of proteomic data.

Co-authors:

Qin Fu , Cedars-Sinai Medical Center

Esthelle Hoedt, Cedars-Sinai Medical Center
Weston Spivia, Cedars-Sinai Medical Center
Saeed Seyedmohammand, Cedars-Sinai Medical Center
Alekandra Binek, Cedars-Sinai Medical Center
Blandine Chazarin Orgel, Cedars-Sinai Medical Center
Simion Kreimer, Cedars-Sinai Medical Center
David Herrrington, Wake Forest University

42
IN SITU TISSUE PATHOLOGY FROM SPATIALLY ENCODED MASS
SPECTROMETRY CLASSIFIERS VISUALIZED IN REAL TIME THROUGH
AUGMENTED REALITY
Abstract ID: 13

Presenting author: Arash Zarrine-Afsar, University of Toronto

Introduction

A large number of publications describing developments of a variety of hand-held mass spectrometry analysis probes,
capable of determining tissue pathology with only a few seconds of sampling and analysis time are reported. To enhance
intrasurgical utility, tissue pathology results from hand-held sampling probes must be spatially encoded and displayed to
the surgeon in real time at the position of sampling to better streamline surgical decision making. We created a platform
for spatially encoded pathology classifications displayed at the site of sampling as color-coded pixels in an augmented
reality video feed of the surgical field of view. Integration between a hand-held mass spectrometry desorption probe
based on picosecond infrared laser technology (PIRL-MS) and an optical surgical tracking system was pursued to
demonstrate this concept.

Methods

A two-way communication between optical surgical navigation and a multivariate data analysis platform through a
custom-built interface on a Xevo qTOF equipped with PIRL-MS sampling system was completed. A variety of xenograft
models were used to determine registration accuracy and correct pathology prediction accuracy of the system.

Preliminary data (results)

Performance of the system was evaluated using murine models of human brain, breast and head & neck cancers
sampled in situ in the presence of body fluids with a technical pixel error of 1.0±0.2 mm, suggesting an 84% or 92%
cancer type classification rate across different molecular models that distinguish cell-lines of each class of breast, brain,
head & neck murine models. Further, through end-point immunohistochemical staining for DNA damage, cell death and
neuronal viability, spatially encoded PIRL-MS sampling is shown to produce classifiable mass spectral data from living
murine brain tissue, with levels of neuronal damage that are comparable to those induced by a surgical scalpel. This
highlights the potential of spatially encoded PIRL-MS analysis for in vivo use during neurosurgical applications of cancer
type determination or point-sampling in vivo tissue during tumor bed examination to assess cancer complete removal.
The interface developed herein for the analysis and the display of spatially encoded PIRL-MS data can be adapted to
other hand-held mass spectrometry analysis probes currently available.

Please explain why your abstract is innovative for mass spectrometry?

Spatially encoded ambient MS classifiers visualized in real time through augmented reality display will allow surgeons to
make immediate decisions regarding the course of surgery without switching field of view.

Co-authors:

Michael Woolman, University of Toronto

Jimmy Qiu, University Health Network
Claudia Kuzan-Fischer, Hospital for Sick Children
Isabelle Ferry, Hospital for Sick Children
Delaram Dara, University of Toronto
Lauren Katz, University of Toronto
Fowad Daud, University Health Network
Megan Wu, Hospital for Sick Children
Manuela Ventura, University Health Network
Nicholas Bernards, University Health Network
Harley Chan, University Health Network
Inga Fricke, University Health Network
Mark Zaidi, University of Toronto
Brad Wouters, University Health Network
James Rutka, Hospital for Sick Children
Sunit Das, University of Toronto
Jonathan Irish, University Health Network
Robert Weersink, University Health Network

43
Howard Ginsberg, University of Toronto
David Jaffray, University Health Network

Spatially encoded mass spectrometry classifiers for in situ pathology

Spatially encoded mass spectrometry classifiers for in situ pathology

44
AUTOMATION OF THE FILTER-AIDED SAMPLE PREPARATION (FASP)
PROTOCOL USING THE ROBOTIC PLATFORM BIOMEK I7 WITH FOCUS
ON PLASMA SAMPLES PREPARATION
Abstract ID: 166

Presenting author: Dana Hein, University Medical Center Mainz

Introduction

Within the DIASyM (Data-Independent Acquisition-based Systems Medicine) research core, at the University Medical
Center (UMC) Mainz, we focus on data-independent mass-spectrometry based proteome analysis of heart failure (HF)
patient plasma samples. HF is the leading cause for hospitalization of individuals over 65 years of age. Our aim is to
identify typical biomarkers for HF or those that are predicting a worsening of HF, as well as to better characterize
different subtypes of HF via deep phenotyping using mass spectrometry based proteomics. Samples need to be
prepared in a reliable, standardized, high-throughput manner using a robotic platform (Biomek i7, Beckman Coulter ©).
This enables sample preparation of up to 92 samples/day. Automated sample preparation was established with plasma
samples of a small study cohort (EmDIA).

Methods

A Biomek i7 (Beckman Coulter©) with an attached positive pressure ALP (Amplius) was used for sample preparation.
Samples were digested using the filter-aided sample preparation (FASP) protocol. The manual FASP-protocol was
automated and improved for plasma digestion using Biomek i7 with samples derived from the EmDIA study (patient
cohort with approximately 420 samples) and a plasma pool as control. Conditions of the FASP protocol were adjusted
and experiments comparing the manual and automated method were performed.

Preliminary data (results)

Washing-/incubation-conditions, pressure settings and liquid handling steps were optimized for the automated sample
preparation, to reach the same or higher levels of identified peptides/proteins as the manual digestion protocol.
Afterwards, the EmDIA samples were all prepared using the optimized automated sample preparation method.
Reproducibility within one 96 well plate and within the whole study cohort was tested.

Automated sample preparation is essential to speed up and standardize high-throughput analyses. After in depth
optimization, including pressure settings, buffer conditions and input amounts, we have successfully established a 96-
well FASP-protocol for efficient processing of patient plasma samples. In contrast to manual processing, the automated
protocol on a Biomek i7 minimizes hands-on time and enables routine high-throughput sample preparation. In
combination with data acquisition on a timsTOF Pro 2 platform using diaPASEF with DIA-NN data processing, the
workflow enables the quantification of typically around 350 proteins in 15 min instrument time from human plasma
samples.

Please explain why your abstract is innovative for mass spectrometry?

Automated sample preparation enables a high-throughput mass spectrometry analysis of clean samples. Our protocol
focuses on plasma, but can be used for many different cell types.

Co-authors:

Elena Kumm, University Medical Center Mainz

Ute Distler, University Medical Center Mainz
David Gomez-Zepeda, University Medical Center Mainz
Christiane Schönfeld, University Medical Center Mainz
Christina Jung, University Medical Center Mainz
Claudia Darmstadt, University Medical Center Mainz
Stefan Tenzer, University Medical Center Mainz

45
PROTEOFORM-REACTION-MONITORING (PFRM) AND THE DISCOVERY
OF BIOMARKER CANDIDATES IN LIVER TRANSPLANTED RECIPIENTS
Abstract ID: 400

Presenting author: Rafael D Melani, Department of Chemistry, Northwestern University, Proteomics Center of
Excellence, Northwestern University

Introduction

Acute rejection after liver transplantation is the leading cause of mortality among liver transplanted recipients (LTRs).
Unfortunately, patient prognosis is based on invasive procedures like biopsy. Therefore, we are applying mass
spectrometry-based top-down proteomics to detect differentially expressed proteoforms from LTR’s blood samples.
Ultimately, our goal is to develop a minimally invasive, fast, and sensitive assay for diagnostic purposes. Previously, we
performed a top-down quantitative analysis of peripheral blood mononuclear cells (PBMCs) from 75 LTRs and validated
a set of 24 proteoforms in a second cohort using selected ion monitoring (SIM). Here we describe the stepwise
development of a targeted quantitative assay named proteoform reaction monitoring (PfRM) and its optimization to
analyze a validation cohort of 100 LTR samples collected longitudinally over a year.

Methods

We analyzed 100 samples from 25 LTRs that were collected at four-time points. Patients were divided into three groups:
transplant excellent; acute dysfunction, no rejection; and acute rejection. Proteins from PBMCs were pre-fractionated by
size using gel electrophoresis. Standard proteins and the 0-30 kDa fraction of the PBMCs were analyzed by PfRM,
initially on an Orbitrap Eclipse and more recently in a modified Orbitrap Tribrid instrument (Thermo Scientific). The PfRM
method captures 24 immunoproteoforms, is consistently under development, and the obtained data were analyzed using
Proteoform Finder.

Preliminary data (results)

We are developing a PfRM approach that alternates SIM spectra with proteoform fragmentation via HCD, increasing
specificity and confidence in proteoform identification. The PfRM assay leverages improved ion transmission in the
newer Orbitrap tribrids, with faster data acquisition cycles.

Data collected on standards demonstrate that the PfRM method improves quantification metrics compared to the SIM-
only assay while providing the fragment ion specificity required for multiplexing the assay to include closely related, co-
eluting proteoforms. All proteins were quantified well in the low femtomole range using Proteoform Finder, developed for
proteoform-level targeted LC-MS/MS experiments.

Using the immune-PfRM assay in its current state of development, we applied it to longitudinal studies from a new cohort
of 25 LTRs to quantify rigorous proteoform targets in 100 patient samples. We could reliably detect the 24 proteoforms
from 23 proteins, including platelet factor 4 and profilin-1 proteoforms that were previously identified as upregulated in TX
recipients. Several proteoforms showed over 4-fold differential regulation over the time course samples, with a set of
proteoforms remaining high in patients with healthy liver transplants.

To increase the method sensitivity and enable the application of PfRM on isolated cell types, we are now in the process
of testing a modified Orbitrap Tribrid instrument that should enable several-fold improvements in LOD and LLOQ of the
PfRM targeted assay. Like SRM/MRM and peptide reaction monitoring (PRM), PfRM will improve the creation of high-
value, portable assays to monitor proteoform biology and biomarkers for future diagnostic purposes in clinical research.

Please explain why your abstract is innovative for mass spectrometry?

Targeted analysis of blood proteoforms is becoming as well-developed as peptide level assays and can enhance
diagnostic methods based on the discovery and validation of protein-based biomarkers.

Co-authors:

Che-Fan Huang, Proteomics Center of Excellence, Northwestern University

Fernanda Negrão, Proteomics Center of Excellence, Northwestern University
Graeme McAlister, Thermo Fisher Scientific
Christopher Mullen, Thermo Fisher Scientific
Micheal Goodwin, Thermo Fisher Scientific

46
Jesse Canterbury, Thermo Fisher Scientific
Xiao Wang, Thermo Fisher Scientific
Raman Mathur, Thermo Fisher Scientific
David Bergen, Thermo Fisher Scientific
Romain Huguet, Thermo Fisher Scientific
Joseph Greer, Proteomics Center of Excellence, Northwestern University
Matthew Robey, Proteomics Center of Excellence, Northwestern University
Bryan Early, Proteomics Center of Excellence, Northwestern University
Ryan Fellers, Proteomics Center of Excellence, Northwestern University
Paul Thomas, Proteomics Center of Excellence, Northwestern University
Luca Fornelli, Department of Biology, University of Oklahoma
Josh Levitsky, Comprehensive Transplant Center, Feinberg School of Medicine, Northwestern University
Mike Senko, Thermo Fisher Scientific
Neil Kelleher, Department of Molecular Biosciences, Northwestern University, Department of Chemistry, Northwestern
University, Department of Biochemistry and Molecular Genetics, Feinberg School of Medicine, Northwestern University

Proteoform reaction monitoring (PfRM) workflow.rafae

47
LC-MS ANALYSES OF PURIFIED CIRCULATING PLASMA EXOSOMES
FROM PATIENTS INFECTED WITH MYCOBACTERIUM TUBERCULOSIS
REVEALED DISEASE STATE-DEPENDENT PROTEIN AND LIPID PROFILES
Abstract ID: 716

Presenting author: Stefan Kalkhof, Fraunhofer Institute for Cell Therapy and Immunology - IZI, Institute for
Bioanalysis Coburg - IBICO

Introduction

Tuberculosis is one of the leading causes of death. We have developed custom lipidomics and proteomics assays to
purify exosomes from plasma samples of patients with either tuberculous lymphadenitis (TBL) or treated and untreated
pulmonary tuberculosis (PTB) and to determine lipids and proteins involved in the host-pathogen relationship.

Methods

Exosomes were extracted and purified from plasma samples of patients by size exclusion chromatography and
ultrafiltration. Quality of exosomes was confirmed by electron microscopy, nanoparticle tracking, and flow cytometry.
Lipids were quantified using HPTLC-ESI-MS on an Amazon SL. For quantitative proteomics after reduction, alkylation,
and proteolytic cleavage (LysC/Trypsin) proteolytic peptides were labeled with 10-plex TMT and measured by LC-MS/MS
on a Q Exactive HF coupled online to Easy-nLC 1200 system. Data processing and analysis were performed using
MaxQuant, various R packages, and DAVID bioinformatics tools. Additional targeted analyses of exosomal surface
proteins were performed by flow cytometry.

Preliminary data (results)

This is the first study of exosomes from plasma of TB patients using lipidomics and proteomics. Our results show that
treatment of PTB patients affects the fatty acid composition of sphingomyelins, phosphatidylcholins,
phosphatidylinositols, free fatty acids, triacylglycerols, and cholesteryl esters, and shifting them toward the composition of
healthy controls. Proteomic characterization of isolated exosomes identified three proteins of Mtb origin: RpoC, Rv2285,
and HycE, the latter of which was found to be expressed differently in patients with TBL. We also discovered that Mtb
infection alters host protein composition in circulating exosomes, with a total of 37 proteins significantly affected (fig. 1).
Apolipoproteins and the antibacterial proteins cathelicidin, SSC5D, and FCN3 were depleted in all TB patients. Protein
profiles of PTB and TBL patients were highly correlated, with 14 proteins coregulated compared with healthy controls.
However, disease-specific proteins could be extracted, including adhesion proteins (integrins, ICAM2, CD151, PRG4),
which were more abundant in PTB patients, and immunoglobulins, C1R, and GRIP1 in TBL patients. Anti-TB treatment
was found to have little effect on the protein composition of circulating exosomes, with only eight proteins altered in
abundance, including SAA4, FCN3, and APOB. Our results confirm previous studies on Mtb. We also characterize
previously unknown disease- and treatment-specific features of circulating exosomes, many of which are directly linked
to Mtb pathogenicity and immunity. This may open the possibility of using the profiles reported here to study host-
pathogen interactions in the development of new vaccines and drugs.

Please explain why your abstract is innovative for mass spectrometry?

Established protocol allowed identification of significantly regulated and disease-associated lipids (by HPLTC-ESI-MS)
and proteins (by TMT-based LC-MS/MS) from plasma exosomes as well as detection of pathogen proteins.

Co-authors:

Fantahun Biadglegne, College of Medicine and Health Sciences, Bahir Dar University, Institute of Medical Microbiology
and Epidemiology of Infectious Diseases, University of Leipzig, Institute of Clinical Immunology, Leipzig University
Johannes R. Schmidt, Fraunhofer Institute for Cell Therapy and Immunology - IZI
Kathrin M. Engel, Institute of Medical Physics and Biophysics, Leipzig University
Jörg Lehmann, Fraunhofer Institute for Cell Therapy and Immunology - IZI
Robert T. Lehmann, Fraunhofer Institute for Cell Therapy and Immunology - IZI
Anja Reinert, Faculty of Veterinary Medicine, Institute of Anatomy, Histology and Embryology, Leipzig University
Brigitte König, Institute of Medical Microbiology and Epidemiology of Infectious Diseases, University of Leipzig
Jürgen Schiller, Institute of Medical Physics and Biophysics, Leipzig University
Ulrich Sack, Institute of Clinical Immunology, Leipzig University

48
Disease and treatment specific clustering utilizing significantly regulated exosomal proteins

49
A METAPROTEOMICS BIOINFORMATICS WORKFLOW TO STUDY HOST-
MICROBE DYNAMICS IN CLINICAL SAMPLES
Abstract ID: 728

Presenting author: Pratik Jagtap, University of Minnesota

Introduction

The field of clinical metaproteomics has the potential to offer insights into host and microbiome interactions. However,
the analytical challenges associated with the detection of low-abundance microbial proteins make it difficult to arrive at
biological inferences. As a solution, we have developed a novel, integrated workflow coupling deep mass spectrometry-
based analysis, with customized bioinformatic processing of both host and microbial proteins. This involves the
generation of a verified host and microbe peptide panel, thus offering candidates suitable for targeted analysis within
individual patient samples. We have utilized this workflow in our ongoing work to identify a promising host and microbe
peptide panel for application to cystic fibrosis (CF) disease progression studies.

Methods

Protein digests from clinical samples were fractionated using high pH reverse-phase liquid chromatography and were
subjected to LC-FAIMS-MS/MS using a Thermo Orbitrap Eclipse mass spectrometer to generate RAW files. MetaNovo
software within the Galaxy platform was used to match the MS/MS to a large reference metaproteomics database to
generate a reduced database of potential microbial constituents. Software such as Fragpipe, MaxQuant,
SearchGUI/PeptideShaker, and FlashLFQ was used to detect peptides with associated peptide intensities. Customized
Galaxy tools such as Unipept were used to verify human and microbial peptides of interest for targeted validation studies.

Preliminary data (results)

Broncho-alveolar samples from pediatric CF and disease control (DC) patients were characterized for their microbial
diversity, pooled, digested with trypsin, and subjected to mass spectrometry (MS). A large combined microbial protein
sequence database of ~17M sequences was generated based on 16S rRNA taxonomic composition. The 17M
sequences database was processed using MetaNovo software to generate a reduced database containing 238,382
human and microbial sequences. The Fragpipe suite, MaxQuant software, and SearchGUI/PeptideShaker/FlashLFQ
were used for spectral matching against this reduced database, to generate reports with peptide intensities and protein
spectral counts. Microbial peptides were validated using the PepQuery tool within Galaxy, to generate a catalog of 666
high confidence microbial peptides. Analysis of the microbial data shows enrichment of Pseudomonas (a known CF
contributor) as well as Sphingomonas and Stenotrophomonas. Proteins from Moraxella and Streptococcus were
enriched in the DC samples. Currently, we have generated a peptide panel of 28 microbial peptides (12 from CF and 16
from DC) which will be subjected to further characterization. Analysis of the human proteome in CF, showed enrichment
of proteins involved in immune response and inflammation, as well as transport and cell motility proteins. Proteins
enriched in DC include membrane proteins involved in cytoskeletal protein regulation. Based on these results, we are
developing a novel microbial and human peptide panel, amenable to parallel reaction monitoring (PRM) to track host-
microbe protein dynamics in CF. Overall, this workflow should prove useful to others seeking to investigate host-microbe
interactions in challenging clinical samples.

Please explain why your abstract is innovative for mass spectrometry?

A bioinformatics workflow for host-microbiota proteomics in clinical samples, delivering well-characterized peptide panels
for targeted investigation of host-microbe dynamics in disease.

Co-authors:

Monica Kruk, University of Minnesota

Subina Mehta, University of Minnesota
James Johnson, University of Minnesota
Reid Wagner, University of Minnesota
Katherine Do, University of Minnesota
Chris Wendt, University of Minnesota
John O’Connor, Ann & Robert H. Lurie Children’s Hospital of Chicago
Theresa Laguna, Ann & Robert H. Lurie Children’s Hospital of Chicago, Northwestern University Feinberg School of
Medicine
Timothy Griffin, University of Minnesota

50
51
Session: LS-2 MS in Structural Biology - Crosslinking MS

CROSSLINKING MASS SPECTROMETRY: THE FRONTIER OF

PROTEOMICS
Abstract ID: 1047

Presenting author: Juri Rappsilber, TU Berlin

Introduction

Crosslinking mass spectrometry: the frontier of proteomics Crosslinking mass spectrometry is now a well-established
experimental technique for investigating protein interactions, structure and function. It is an indispensable tool in support
of single particle cryo-EM studies to illuminate areas with missing structural information due to flexibility. Crosslinking MS
also remains unparalleled in its ability to provide structural information in complex systems. However, many paths lead to
data in crosslinking MS, not all at equal outcome. The data and experimental strategies are ready to submit entrusted
believes that may govern current choices to scientific scrutiny. Will the current multitude of approaches persist, or a
single approach emerge as clearly superior?

COMBINING CROSS-LINKING MASS SPECTROMETRY AND

COMPLEXOME PROFILING FACILITATES THE SELECTIVE ANALYSIS OF
PROTEIN COMPLEXES
Abstract ID: 15

Presenting author: Johannes Hevler, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for
Biomolecular Research and Utrecht Institute for Pharmaceutical Sciences, University of Utrecht, Padualaan 8,
3584 CH Utrecht, The Netherlands, Netherlands Proteomics Center, Padualaan 8, 3584 CH Utrecht, The
Netherlands

Introduction

Cross-linking mass spectrometry (XL-MS) has emerged into a powerful method to structurally characterize proteins and
protein interactions. However, in-solution cross-linking does generally not provide sensible solutions for differentiating
cross-links obtained from co-occurring protein complexes and/or equilibria of protein assemblies (e.g. oligomers), which
significantly hampers an adequate structural characterization.

Methods

Here we combine in-gel XL-MS with complexome profiling into an approach we term IGX-MS, which enables the
complex specific assignment of detected cross-links.

Preliminary data (results)

We demonstrated this for the complement hetero-dimeric C5b6 complex in its equilibrium with monomeric C5 and C6.
Similarly, we employed this approach to study the interactome of heart mitochondria. We focused on complexes of
oxidative phosphorylation and found that the dimeric apoptosis-inducing factor 1 (AIFM1) forms a defined complex with
cytochrome c oxidase (COX or Complex IV) but hardly interacts with respiratory chain supercomplexes. Multiple AIFM1
intercross-links engaging six different COX subunits provided structural restraints to build a detailed and complete
structural model of the COX-AIFM12 complex. This model excludes direct electron transfer between AIFM1 and COX,
however, the binding site of cytochrome c remains accessible. The discovery of the previously overlooked COX-
AIFM12 complex and clues provided by the structural model hint at potential roles of AIFM1 in oxidative phosphorylation
biogenesis and in programmed cell death.

Please explain why your abstract is innovative for mass spectrometry?

52
In general, combining XL-MS with complexome profiling facilitates complex specific assignment of cross-links, and will
thus be useful for the structural biology community.

Co-authors:

Riccardo Zenezeni Chiozzi, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research
and Utrecht Institute for Pharmaceutical Sciences, University of Utrecht, Padualaan 8, 3584 CH Utrecht, The
Netherlands, Netherlands Proteomics Center, Padualaan 8, 3584 CH Utrecht, The Netherlands
Alfredo Cabrera-Orefice, Radboud Institute for Molecular Life Sciences, Radboud University Medical Center, 6525 GA
Nijmegen, The Netherlands
Matti Pronker, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research and Utrecht
Institute for Pharmaceutical Sciences, University of Utrecht, Padualaan 8, 3584 CH Utrecht, The Netherlands,
Netherlands Proteomics Center, Padualaan 8, 3584 CH Utrecht, The Netherlands
Vojtech Franc, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research and Utrecht
Institute for Pharmaceutical Sciences, University of Utrecht, Padualaan 8, 3584 CH Utrecht, The Netherlands,
Netherlands Proteomics Center, Padualaan 8, 3584 CH Utrecht, The Netherlands
Ulrich Brandt, Radboud Institute for Molecular Life Sciences, Radboud University Medical Center, 6525 GA Nijmegen,
The Netherlands, Cologne Excellence Cluster on Cellular Stress Responses in Aging-Associated Diseases (CECAD),
University of Cologne, 50931 Cologne, Germany.
Susanne Arnold, Radboud Institute for Molecular Life Sciences, Radboud University Medical Center, 6525 GA Nijmegen,
The Netherlands, Cologne Excellence Cluster on Cellular Stress Responses in Aging-Associated Diseases (CECAD),
University of Cologne, 50931 Cologne, Germany.

Combining Blue-native PAGE with IGX-MS

Two-tier experimental strategy for the analysis of proteome-wide protein-protein interactions

53
STRUCTURAL MASS SPECTROMETRY APPROACHES TO DECIPHER
INTERACTIONS WITHIN THE ~380 KDA RUVBL1/RUVBL2/DPCD COMPLEX
Abstract ID: 42

Presenting author: Sarah Cianferani, LSMBO-IPHC-CNRS-University of Strasbourg, French Proteomic

Infrastructure (FR2048 CNRS -CEA)

Introduction

DPCD is a protein that plays a role in cilia formation, its absence leading to primary dyskinesia. High-throughput
quantitative proteomic studies identified DPCD (23 kDa) as a new RUVBL1 (R1) and RUVBL2 (R2) interacting partner,
R1 and R2 (52 kDa each) being two ATPAses Associated with various cellular activities (AAA+) essential for many
cellular processes. R1R2 from yeast are reported to assemble into dodecamers (626 kDa), which are further disrupted
into hexamers (313 kDa) upon nucleotide binding to ensure R1R2 chaperone activity.
A combination of structural mass spectrometry (MS) methods, including native MS (nMS), ion mobility MS (nIM-MS) and
chemical cross-linking (XL-MS) was used to characterize the R1R2D complex and to generate 3D model of DPCD alone
or associated to R1R2.

Methods

HisDPCD (Q9BVM2) and R1R2 proteins (Q9Y265,Q9Y230) were overexpressed in E. coli, purified and incubated for
18h at 4°C to form the R1R2D complex.
nMS was performed on an Orbitrap Exactive Plus EMR (Thermo), nIM-MS on a Synapt G2 HDMS (Waters). XL-MS was
performed with a 100 molar excess of DSBU (CF+ chemical), a NHS ester MS-cleavable XL reagent, and analysed on
an Orbitrap Q-exactive HFX (Thermo). XL peptides were validated in 2/3 replicates using MeroX 2.0.1.4.4. DPCD ab
initio model was obtained using GASBOR ; R1R2D model was analyzed with CRYSOL under PRIMUS interface from
ATSAS (3.0.3).

Preliminary data (results)

For individual proteins, nMS evidenced that DPCD is a 23kDa monomer while R1R2 is a mixture between hexamers
(313 kDa) and dodecamers (626 kDa).
As no 3D structure of DPCD is available, we implemented nMS, nIM-MS and XL-MS in combination with ab
initio modeling (GASBOR) and provided the first 3D model for monomeric DPCD.
For R1R2D, nMS allowed detection of a ~383 kDa complex, corresponding to the binding of three DPCD to hexameric
R1R2, which confirms also the stabilization of the R1R2 hexameric ring upon DPCD binding. nIMS-MS revealed that
DPCD binding occurs with an increase in collision cross section (CCS) of ~16% upon DPCD binding, in agreement with
mass-based CCS estimations and in-solution SAXS data, suggesting that the global shape of the hexameric R1R2 is
maintained upon DPCD binding. To gain information about spatial proximities, XL-MS was performed from which 43
intra- and 45 inter- cross-links were identified and validated within the R1R2D complex. XL-MS results highlighted the
central role of DII domains on R1/R2 side, along with involvement of the N-terminal part and the CS putative domain of
DPCD.
Finally, rigid docking experiments were conducted under XL-MS constraints. Best models revealed that DPCD is located
at the periphery of the ring. XL-MS also allowed precise orientation of DPCD, evidencing its location not strictly below the
DII domains but likely on the side near DII, close to the ATPase domains. Altogether, the R1R2D model is in line with all
several other different biophysical approaches (EM, SAXS).

Please explain why your abstract is innovative for mass spectrometry?

Combination of complementary structural MS approaches (nMS, nIM-MS and XL-MS) to provide first 3D model for the
383 kDa RUVBL1/2-DPCD complex.

Co-authors:

Marie Ley, LSMBO-IPHC-CNRS-University of Strasbourg, French Proteomic Infrastructure (FR2048 CNRS -CEA)
Raphael Dos Santos Morais3 , IMoPA
Evolène DESLIGNIERE, LSMBO-IPHC-CNRS-University of Strasbourg, French Proteomic Infrastructure (FR2048 CNRS
-CEA)
Paulo E. Santo, iBET
Tiago M. Bandeiras, iBET
Bruno CHARPENTIER, IMoPA
Xavier Manival, IMoPA

54
Native MS and XL-MS analyses of R1R2D.

Rigid body model obtained under XL-MS constraints for R1R2D complex.

55
FAST FLUOROALKYLATION OF PROTEINS (FFAP): A NOVEL CROSS-
LINKING STRATEGY FOR AROMATIC AMINO ACIDS
Abstract ID: 192

Presenting author: Zdenek Kukacka, Institute of Microbiology, The Czech Academy of Sciences

Introduction

Chemical cross-linking in combination with mass spectrometry (CXMS) has been developed into a powerful method for
mapping protein structures, dynamics and interaction networks including molecular interfaces in protein-protein and
protein-nucleic acid complexes. Although many cross-linkers have been developed in last two decades, majority of
CXMS analyses still utilizes lysine-specific cross-linking reagents based on N-hydroxysuccinimide esters. Other cross-
linking reagents have only limited use for various reasons such as low reactivity, low occurrence of targeted residues,
unexpected side products. In this study we show new generation of cross-linkers based on recently published Fast
FluoroAlkylation of Proteins (FFAP) technology that enables targeting aromatic amino acid side chains.

Methods

Cross-linking reaction is based on two step mechanism which includes activation of hypervalent iodine bivalent reagent
using lewis acid and subsequent attack of aromatic residues within the protein sequence by the resulting radicals. Since
the induction of fluoroalkyl radicals is triggered by ascorbic acid and the labeling pulse is stopped by tryptophan, it
enables to perform protein labeling experiments in quench flow system in short time range (3s). The studied proteins
were analyzed by bottom up approach using high resolution mass spectrometry (solariX XR, 15T) where samples were
digested by trypsin, separated on reverse phase column online coupled to MS.

Preliminary data (results)

The results on several protein models such as horse heart myoglobin or binary haptoglobin-hemoglobin complex clearly
demonstrate the potential of our novel cross-linking strategy. The cross-links of aromatic residues generated by
fluoroalkyl radicals nicely correspond with protein crystal structures of studied proteins and provide information not
attainable by lysine-specific cross-links. Our data lead to an assumption that the fluoroalkyl radicals will be utilized for
fast cross-linking experiments studying structure and dynamics of proteins and protein assemblies in solution.

This work was supported by the Czech Science Foundation (grant numbers 22-27695S) and European Commission
H2020 (The European Proteomics Infrastructure Consortium providing access EPIC-XS - project agreement No.
823839).

Please explain why your abstract is innovative for mass spectrometry?

A novel cross-linking reagents targeting aromatic amino acids.

Co-authors:

Lukas Fojtik, Institute of Microbiology, The Czech Academy of Sciences, Faculty of Science, Charles University
Jan Fiala, Institute of Microbiology, The Czech Academy of Sciences, Faculty of Science, Charles University
Petr Novak, Institute of Microbiology, The Czech Academy of Sciences, Faculty of Science, Charles University

56
A NEW TRIFUNCTIONAL CROSS-LINKER FACILITATING THE MAPPING
OF MEMBRANE PROTEINS FOR IN VIVO PROTEOME-WIDE STUDIES
Abstract ID: 252

Presenting author: Lucienne Nouchikian, Institut Pasteur, Université de Paris, CNRS

Introduction

Within the past few years, in vivo cross-linking (XL) mass spectrometry (MS) has gained a lot of traction in studying
protein-protein interactions (PPIs) in a proteome-wide manner. Recently, we demonstrated highly efficient in vivo cross-
linking using our enrichable homo-trifunctional cross-linker, NNP9 (Rey et al., 2021), with primary amine specificity. Here,
we introduce a completely new enrichable cross-linker, NXL15, which contains a diazirine group that inserts into any C-X
bond. Our objective is to expand the type of amino acids targeted to improve the analysis of protein complexes lacking
lysine residues, such as membrane ones. Applied to the in vivo analysis of N. meningitidis, NXL15 shows an original
reactivity by targeting preferentially acidic amino acids, allowing for a very interesting expansion of the PPI network.

Methods

NXL15 is a trifunctional cross-linker containing an NHS-carbamate, a UV photo-reactive diazirine group, and an alkyne
moiety for further enrichment through click-chemistry. Our XL-MS workflow includes the following steps: cross-linking,
trypsin digestion, click-chemistry on photo-cleavable agarose beads, peptide elution and nanoLC-MS/MS analysis using
an Orbitrap Eclipse with HCD and EThCD. BSA was first used for optimization and for in vivo studies, 5x1010N.
meningitidis cells were labeled.Data was analyzed using MassSpec Studio with a false discovery rate of 1 % searching
against BSA or N. meningitidis Nm8013 protein databases (2125 protein entries).

Preliminary data (results)

We first optimized reaction conditions of NXL15 on purified BSA: protein/cross-linker ratio, NHS binding times, and UV
reaction time (for the diazirine reaction). We then determined NXL15 site specificity using EThCD and found it mainly
targets acidic residues (Glu/Asp). Like NNP9 (which contains two NHS carbamate groups and an alkyne moiety), NXL15
gave around 100 cross-linked peptides per run for BSA, with very complementarity data. Common highly reactive lysines
were labeled in both, however the second arm of each cross-linker targeted completely different areas of the protein,
increasing the quality of structural data achieved through improved XL coverage.

We then performed in vivo cross-linking on N. meningitidis. Compared to NNP9, NX15 allows to identify novel PPIs that
nicely complement the previously obtained bacterial interactome. Again, there were common sites to both cross-linkers
but completely different binding partners. For quality control, we mapped cross-links on a model ribosomal structure and
achieved an average distance of less than 30 Å, which agrees with the length of NXL15 (21 Å).

Very interestingly, and as expected, the ability of NXL15 to bind acidic residues gave us more interactions with
membrane proteins. Specifically, we were able to identify more interactions in the bacterium’s piliation machinery, a
major bacterial virulence factor that is structurally and mechanistically poorly understood. We currently pursue our XL-MS
experiments with NXL15 on mutants of N. meningitidis where the piliation machinery had been blocked at different
stages of assembly which will aid in understanding its role in the bacterial infection.

Please explain why your abstract is innovative for mass spectrometry?

Introduction of a new trifunctional bio-orthogonal cross-linker with a one-step purification approach enriching in
membrane protein cross-links, facilitating the mapping of their interactions in vivo.

Co-authors:

Martial Rey, Institut Pasteur, Université de Paris, CNRS

Agustin Zavala, Institut Pasteur, Université de Paris, INSERM
Guillaume Duménil, Institut Pasteur, Université de Paris, INSERM
Julia Chamot-Rooke, Institut Pasteur, Université de Paris, CNRS

57
General in vivo cross-linking workflow with our trifunctional crosslinker NXL15

58
SURVIVAL STRATEGIES IN THE DEEP - STRUCTURAL DYNAMICS OF A
HYPERTHERMOPHILIC PEP-SYNTHASE
Abstract ID: 395

Presenting author: Pascal Albanese, Biomolecular Mass Spectrometry and Proteomics Group, Utrecht Institute
for Pharmaceutical Sciences, Utrecht University, 3584CH Utrecht, The Netherlands, Netherlands Proteomics
Centre, Utrecht University, Utrecht University, 3584CH Utrecht, The Netherlands

Introduction

Despite life conquering nearly every habitat through adaptive evolution, the production of biomass relies on the
combination of only a few metabolites. Enzyme structural adaptations allow the basic metabolic reactions that are a
prerequisite for life, especially so in environments hostile to most life forms known to date. Key metabolic enzymes of
hyperthermophilic Archaea have adapted to conserve their catalytic cores to work at 90-100 °C, conditions similar to
those where earliest life forms may have originated. Here we characterize a key enzyme of the central carbon
metabolism, through an integrative approach combining structural mass spectrometry (MS), cryo-electron microscopy
(EM), and molecular dynamics simulations. From our investigation, we unveil the so far elusive structure of a 2.15 MDa
phosphoenolpyruvate synthase (PPSA) from P. furiosus.

Methods

Native PPSA complexes were purified directly from P.furiosus cells. PPSA fractions and cell lysates were cross-linked
with 2mM PhoX, digested, and analyzed by LC-MS/MS. Native MS measurements were performed on a Q-Exactive
UHMR (TFS). DDA-MS data were analyzed with MaxQuant, crosslinking (XL) MS data with XlinkX/PD2.4. Reproducible
cross-links, present in 3 out of 4 replicates, were used for further analysis. PPSA structure was determined by a
combination of single-particle cryo-EM for the stable core (residues 508-817) and molecular modeling (residues 1-507).
Molecular dynamics simulations were conducted in GROMACS using the Sirah force field to include acetylation and
phosphorylation.

Preliminary data (results)

The structure of the stable core, solved by Cryo-EM at ~3.4 Å, is assembled as a structure of six tetramers comprising
the C-terminal domain of the protein, one of the two catalytic cores. The structure is stabilized by the planar coordination
of iron by four methionine residues and inter-chain salt bridges; it however does not contain any disulfide bonds. We
additionally uncovered widespread non-enzymatic lysine acetylation of PPSA, as well as the major complexes of P.
furiosus, suggesting an overall stabilizing effect. Integration of Crosslinking MS and co-evolution analyses provided the
spatial distance restraints to model the N-terminal flexible domain of the protein, ultimately defining 2 major possible
arrangements within the full 24-mer. Crosslinking MS data also defined the position of an interacting partner,
Phosphoenolpyruvate carboxykinase (PCK), present at a stoichiometry of ~1 copy per complex, that competes for the
PPSA substrate, suggesting a key role for PCK in locally rerouting some intermediate metabolites. The structural models
were further subjected to extensive molecular dynamics simulations conducted at 90°C to assess the role of highly
acetylated lysine residues on the overall structure at extreme temperatures. Results indicate that the inter-domain
interfaces are disrupted in the absence of acetylation, indicating a very basic mechanism by which hyperthermophiles
may cope with harsh environmental conditions while maintaining an unusually abundant and intrinsically flexible
metabolic enzyme. The origin of extensive non-enzymatic acetylation in hyperthermophiles may lie with primordial
conditions and have therefore implications for protein molecular evolution and design.

Please explain why your abstract is innovative for mass spectrometry?

The integrative approach here used uniquely unveiled the role of extensive non-enzymatic lysine acetylation in the
structural stabilization of enzymes at elevated temperatures.

Co-authors:

Wenfei Song, Cryo-Electron Microscopy Group, Bijvoet Centre for Biomolecular Research, Faculty of Science -
Chemistry, Utrecht University, 3584CH Utrecht, The Netherlands
Siri Van Keulen, Computational Structural Biology, Bijvoet Centre for Biomolecular Research, Faculty of Science -
Chemistry, Utrecht University, 3584CH Utrecht, The Netherlands
Sem Tamara, Biomolecular Mass Spectrometry and Proteomics Group, Utrecht Institute for Pharmaceutical Sciences,
Utrecht University, 3584CH Utrecht, The Netherlands, Netherlands Proteomics Centre, Utrecht University, Utrecht
University, 3584CH Utrecht, The Netherlands
Jeroen Koendjbiharie, Laboratory of Microbiology, Wageningen University, 6708WE Wageningen, The Netherlands

59
Serve Kengen, Laboratory of Microbiology, Wageningen University, 6708WE Wageningen, The Netherlands
Alexandre Bonvin, Computational Structural Biology, Bijvoet Centre for Biomolecular Research, Faculty of Science -
Chemistry, Utrecht University, 3584CH Utrecht, The Netherlands
Friedrich Förster, Cryo-Electron Microscopy Group, Bijvoet Centre for Biomolecular Research, Faculty of Science -
Chemistry, Utrecht University, 3584CH Utrecht, The Netherlands
Richard Scheltema, Biomolecular Mass Spectrometry and Proteomics Group, Utrecht Institute for Pharmaceutical
Sciences, Utrecht University, 3584CH Utrecht, The Netherlands, Netherlands Proteomics Centre, Utrecht University,
Utrecht University, 3584CH Utrecht, The Netherlands

60
 Monday 29 August 2022: 15:30 – 17:30
Session: IM-3 Alternative Dissociation Methods

MASS SPECTROMETRY BASED PHOTODISSOCIATION STRATEGIES FOR

ILLUMINATING THE ‘HIDDEN’ DIVERSITY AND STRUCTURAL
COMPLEXITY OF THE LIPIDOME
Abstract ID: 20

Presenting author: Gavin Reid, University of Melbourne

Introduction

A major goal of analytical strategies for lipidomics is the systematic identification, characterization and quantitation of the
multitude of individual lipid species that may be present within a biological sample of interest. However, conventional
mass spectrometry (MS)- and tandem mass spectrometry (MS/MS)-based strategies for lipidome analysis are not suited
for the comprehensive identification and precise structural characterization of multiple categories of isobaric and isomeric
lipids that may be functionally involved in the regulation of cellular homeostasis, or in the onset and progression
of disease.

Methods

193 nm and 213 nm ultraviolet photodissociation (UVPD)- MS/MS, and ‘hybrid’ MSn methods involving the sequential
use of CID/HCD and UVPD, were implemented in either (i) a custom modified Orbitrap Q Exactive Plus [Ryan et al. J.
Am. Soc. Mass Spectrom. 2017, 28, 1406-1419; Fang et al. Anal. Bioanal. Chem. 2020, 412, 2339–2351., or (ii) an
Orbitrap Fusion Lumos mass spectrometers [White Buenger and Reid. Eur. J. Mass Spectrom. 2020, 26, 311-323; West
and Reid. Anal. Chim. Acta. 2021, 1141, 100-109.], as previously described.

Preliminary data (results)

This presentation will provide an overview of recent work undertaken to address these limitations, involving the
development and application of ultraviolet photo-dissociation (UVPD) tandem mass spectrometry (MS/MS) and ‘hybrid’-
MSn based instrumentation and methods for near-complete lipid structural characterization, including assigning the sites
of fatty-acid/-acyl chain unsaturation or cyclization, sn-linkage positions, and site specific characterization of lipid
modifications such as oxidation and hydroxylation, or complex head-group structures. Examples highlighting the
applications of these methods to obtain otherwise ‘hidden’ information regarding the complexity of the lipidome, as well
as complementary ‘integrated multi-omics’ analysis methods (e.g., transcriptomics and proteomics) will be used to
demonstrate the importance of these approaches for elucidating alterations in lipid metabolism that are functionally
associated with different cellular physiological or pathological conditions.

Please explain why your abstract is innovative for mass spectrometry?

UVPD-MS/MS and ‘hybrid’ MSn enable unambiguous identification and near complete lipid structural characterisation.

61
THE EXCLUSIVE ION ACTIVATION ARSENAL OF THE OMNITRAP
PLATFORM ILLUSTRATED – APPLICATIONS IN TOP DOWN AND BOTTOM
UP PROTEOMICS
Abstract ID: 44

Presenting author: Dimitrios Papanastasiou, Fasmatech Science & Technology

Introduction

Advanced methods for processing ions are demonstrated successfully in the Omnitrap platform, a novel linear ion trap
designed to support a unique functionality landscape. The enhanced performance is enabled by incorporating variations
of the entire range of ion activation methods into a single platform in a highly dynamic fashion. The ion activation arsenal
comprises reagent ions, radical neutral species, photons, electrons and collisions. All fragmentation methods involving
radical chemistry are shown to provide complete sequence coverage for partially unfolded ubiquitin. The first MS4
experiment with intact mAbs is performed successfully highlighting that complete sequence coverage of proteins with an
extended disulfide bond network is feasible. High efficiency electron induced dissociation is accomplished on
chromatographic time scales and performance is benchmarked using a BSA digest standard.

Methods

Experiments are performed on an Omnitrap connected to a Q Exactive MS. Ubiquitin was diluted to 5 μM solutions in a
1% acetic acid aqueous solution. Herceptin was diluted to 5μM in water:acetonitrile in 50:50 ratio (v/v) with 0.1% formic
acid. Static nanoelectrospray was performed using pulled borosilicate glass coated emitters. A 2 μL BSA digest sample
with C=200 fmol/μL was prepared in 3:5:92 ACN:FA:H2O (v/v/v) and injected in a C18 (25cm long, 75μm id) PepSep
column. Chromatographic separations were performed using the EASY-nLC 1000 system.

Preliminary data (results)

Complete sequence coverage with enhanced fragment pair complementarity is demonstrated in MS2 experiments of the
8+ charge state of ubiquitin using externally injected near-zero kinetic energy electrons (electron capture dissociation)
and also 35eV electrons (electron induced dissociation). Results with comparable quality are produced using a pulsed
beam of fast hydrogen ions and also a diffuse VUV light source, while the fragmentation efficiency and distinctive
features for each of the four fragmentation methods are highlighted. In addition, the extreme versatility of the Omnitrap
technology is demonstrated by entirely new multidimensional multiple-stage workflows developed for the analysis of
intact mAbs. The dissociation of intermolecular disulfide bonds is observed in resonance excitation CID under both native
and denatured conditions. Enhancements in the intensity of first generation fragments by one order of magnitude are
feasible by operating the Omnitrap platform in ion-accumulation mode to counterbalance the effect of signal partitioning
into an excessive number of dissociation pathways and to enable MSn (n≥3) experiments to be performed with high
efficiency using variable energy electrons and photons. Dissociation of the light chain intramolecular disulfide bonds is
observed at MS4 level raising sequence coverage to >85%. Finally, highly informative nLC-MS2 mass spectra based on
electron capture and electron induced dissociation of BSA peptides are generated and compared to HCD fragmentation.
Data are processed in Proteome Discoverer 2.5 and new software developed to trace side chain losses, while a
comparison of the results is used to evaluate the efficiency of electron-based reactions performed on chromatographic
time scales.

Please explain why your abstract is innovative for mass spectrometry?

Integration of a multitude of ion activation methods enables new levels in protein characterization to be accomplished.
High efficiency EID becomes available on LC time scales for the first time.

Co-authors:

Athanasios Smyrnakis, Fasmatech Science & Technology

Mariangela Kosmopoulou, Fasmatech Science & Technology
Roman Zubarev, Karolinska Institutet

62
Assembly of the Omnitrap platform designed for multidimensional multiple-stage tandem MS processing of gas phase
ions

MS4 workflow developed for the analysis of intact monoclonal antibodies

63
ELECTRON-INDUCED AND ELECTRON-CAPTURE DISSOCIATION IN
DATA-DEPENDENT ACQUISITION MODE PERFORMED ON THE
OMNITRAP PLATFORM COUPLED TO AN ORBITRAP MASS
SPECTROMETER
Abstract ID: 45

Presenting author: Mariangela Kosmopoulou, Fasmatech Science & Technology

Introduction

The functionality of the Omnitrap MS/MS platform is extended to bottom-up analytical workflows for the first time. The
fast electron-induced and electron-capture dissociation (EID, ECD) reactions enabled by external injection of electrons in
a linear ion trap driven by rectangular RF waveforms made these fragmentation techniques available on
chromatographic time scales. Highly informative MS2 mass spectra generated by the electron-based dissociation
methods available in the Omnitrap platform are compared to HCD mass spectra generated on the Q-Exactive Plus mass
spectrometer. Enhanced sequence coverage and complementarity are obtained by combining the different MS/MS
methods, increasing peptide identification confidence. The enhanced performance observed in EID generating
characteristic side chain losses and distinguishing between leucine and isoleucine residues is highlighted.

Methods

A 2μL BSA digest with C=200fmol/μL was prepared in 3:5:92 ACN:FA:H2O and injected in a C18 (25cm, 75μm id)
PepSep column. Chromatographic separations were performed using the EASY-nLC 1000 system. A 110min long
gradient was used and the eluting peptides are analyzed on a Q-Exactive Plus equipped with the Omnitrap platform. A
top-10 DDA method was employed. HCD is performed with 24eV kinetic energy, while in the ddMS2 ExD experiments
ions are transferred through the HCD cell with 4eV to eliminate collisional activation. EID is applied for 50ms and ECD for
150ms.

Preliminary data (results)

Data are analyzed in Proteome Discoverer and peptide lists are generated for each of the EID, ECD and HCD
experiments respectively. The BSA sequence coverage and the confidence in assignments are compared. The ddMS2-
EID experiment includes ~8500 MS2 scans and results in 63 peptide identifications corresponding to 83% coverage. The
longer reaction time required in ddMS2-ECD includes ~7000 MS2 scans with 42 identifications and limited sequence
coverage (58%). The faster reaction time in HCD results in a greater number of MS2 scans (~11400) and 72 peptides
corresponding to 91% coverage. Further data analysis is performed using new software developed in-house to process
ddMS2 experiments in greater depth by providing information on sequence coverage, complementarity and side chain
losses for each of the peptides identified. The sequence coverage for 2+ charge state peptides in HCD and EID
fragmentation is 95% and increases to 98% in ECD, while EID shows superior sequence coverage for the 3+ and 4+
charge states. Despite the greater number of peptides identified in HCD, ECD produced on average enhanced
complementarity for peptides in all charge states (63%), followed by EID (56%) and finally HCD (41%). Additional x, z, v
and diagnostic d, w fragments distinguishing between leucine and isoleucine are observed only in EID and ECD. In
almost 50% of the cases, EID generated high intensity w fragments diagnostic to leucine. The ddMS2 analysis is
extended to include singly charged peptides to highlight the unique features of EID further.

Please explain why your abstract is innovative for mass spectrometry?

High efficiency EID is made available on LC time scales for the first time.

Co-authors:

Athanasios Smyrnakis, Fasmatech Science & Technology

Giorgos Alevizos, Fasmatech Science & Technology
Dimitrios Papanastasiou, Fasmatech Science & Technology
Roman Zubarev, Karolinska Institutet

64
COUPLING ECD WITH IMS ON A WATERS QUADRUPOLE-IMS-TOF MASS
SPECTROMETER
Abstract ID: 66

Presenting author: Marcus Macht, MS Vision

Introduction

Electron capture dissociation (ECD) is well known to be effective for middle- and top-down fragmentation of proteins
while Ion Mobility-Mass Spectrometry (IM-MS) is a powerful tool for studying gas-phase protein conformations. Earlier we
had demonstrated the operation of the e-MSion ExD cell on Waters instrument in Q-ToF mode (i.e. without Ion Mobility
separation). Here we show how implementation of e-MSion ECD cell on the Waters Q-IMS-ToF provides new capabilities
to greatly improve the analysis of peptides and intact proteins.

Methods

The e-MSion ExD cell consists of six electrostatic lens elements, two high temperature ring magnets and an electron
emitting filament with all voltages under computer control from a separate power supply and allows ECD and EID to be
performed. Experiments with model peptides and proteins were carried out on a Waters Synapt G2-Si Q-IMS-ToF mass
spectrometer with an ExD cell placed in front of Ion Mobility Separation cell, after IMS cell and in both locations. Data
were analyzed using MassLynx v.4.2 and DriftScope v.2.9 software (both Waters Corporation).

Preliminary data (results)

The combined use of ExD cell(s) and Ion Mobility separation provides analytical capabilities previously inaccessible. In
the current work we demonstrate:

• Sequencing conformers by performing ECD after IM separation of ubiquitin

• Getting ECD spectra from all precursor charge states individually at once without precursor isolation by
performing ECD after IM separation of carbonic anhydrase
• Resolving isotope clusters in ECD spectra by performing ECD before IM separation of ubiquitin
• Doubling ECD efficiency by performing ECD twice, i.e. before IMS cell and after will be shown on several
examples.

These exemplary results show just some of capabilities from all possible combinations of two pairs of activation methods
(two ExD and two CID) and Ion Mobility separation which they can provide.

Please explain why your abstract is innovative for mass spectrometry?

ECD-IMS and ECD-IMS-ECD demonstrated for the first time

Co-authors:

Valery G. Voinov, e-MSion Inc., LPI, Oregon State University

Andrew K. Swansiger, Department of Chemistry and Biochemistry, University of Oregon
Samantha O. Shepherd, Department of Chemistry and Biochemistry, University of Oregon
Jesse W. Wilson, Department of Chemistry and Biochemistry, University of Oregon, PNNL, Richland (WA)
Amber D. Rolland Amber D. Rolland, Department of Chemistry and Biochemistry, University of Oregon
James S. Prell James S. Prell, Department of Chemistry and Biochemistry, University of Oregon
Joseph S. Beckman, LPI, Oregon State University, e-MSion Inc.

65
pre-IMS ECD fragmentation spectra of Ubiquitin7+ precursor

66
ELECTRON CAPTURE DISSOCIATION ON A CYCLIC ION MOBILITY
ENABLED Q-TOF MASS SPECTROMETER PROVIDES HIGHER
STRUCTURAL RESOLUTION IN HDX-MS EXPERIMENTS
Abstract ID: 240

Presenting author: Ross Chawner, Waters

Introduction

Hydrogen deuterium exchange mass spectrometry (HDX-MS) is a popular and valuable technique for the structural
analysis of proteins. It has been established that ‘soft’ fragmentation techniques such as electron transfer or electron
capture dissociation (ETD and ECD, respectively) can be used to fragment deuterated peptides while preserving the
deuterium uptake profile of the precursor, thus increasing the structural resolution of the experiment. Despite this, ETD is
rarely employed for this purpose due to the challenges associated with its use.

Here we present the use of ECD for fragmentation of deuterated peptides in HDX-MS experiments. Unlike ETD, ECD
requires no toxic reagents or rapid polarity switching of the instrument, and thus offers an advantageous alternative to
ETD.

Methods

A Q-ToF instrument fitted with a cyclic ion mobility device (Waters Corporation), fitted with an ECD cell (eMSion, Oregon,
USA) in the pre-transfer ion guide (after the cyclic IMS device), was tuned for minimal deuterium scrambling by adjusting
multiple ion source and transmission parameters, which will be described in detail. Deuterium scrambling was assessed
by observing deuterium uptake on the c and z fragment ions of the fully deuterated peptide P1 standard (Sequence:
HHHHHHIIKIIK). The ECD cell was tuned for maximum transmission and fragmentation efficiency by manually adjusting
voltages on focusing lenses within the cell.

Preliminary data (results)

With the instrument tuned for conditions of minimal deuterium scrambling, low deuterium content is observed on the N
terminal histidine residues of the P1 standard peptide, with good agreement observed between the measured deuterium
uptake of the peptide P1 fragment ions and the theoretical 0% scrambled deuterium uptake profile.

Following online pepsin digestion of the PhosB protein standard (accession no. P00489) and subsequent LC-MS
analysis of the resulting peptides, good sequence coverage was observed in the presence of the ECD cell. With ECD
fragmentation on - almost complete sequence coverage of the peptides studied was observed highlighting the utility of
ECD fragmentation for online, bottom-up residue resolution HDX experiments.

Additionally, low deuterium scrambling is also observed with IMS mode on with the precursor ions undergoing a single
pass of the cyclic device prior to ECD fragmentation. Good IMS drift time alignment is observed between ECD fragments
and their respective precursor ions for both the peptide P1 standard and the PhosB standard pepsin digest. Drift time
extraction of these data showed cleaner fragmentation spectra for the peptides studied with less interference from
coeluting ions. Given the frequent complexity of overlapping isotope distributions in peptide fragment ion spectra, owing
to the necessary use of short LC gradients in HDX-MS analyses, these data highlight the utility of IMS separation to aid
the deconvolution of fragmentation spectra, and assist in accurate quantification of deuterium uptake for c and z
fragment ions.

Please explain why your abstract is innovative for mass spectrometry?

ECD fragmentation used for single residue HDX-MS in a cyclic IMS equipped Q-ToF mass spectrometer

Co-authors:

Owen Cornwell, Waters Corporation

Dale Cooper-Shepherd, Waters
Nicola Johnston, Waters
Jonathan Fox, Waters
Malcolm Anderson, Waters

67
ELECTRON INDUCED FRAGMENTATION OF ADDUCT IONS AND
COLLISION INDUCED FRAGMENTATION OF RADICAL CATIONS FOR
STRUCTURAL ELUCIDATION OF METABOLITES BY LC-MS/MS
Abstract ID: 398

Presenting author: Gérard Hopfgartner, University of Geneva, LSMS

Introduction

LC-MS/MS-based global metabolomics or forensic investigations rely on exact mass determinations and collision
induced spectra for structural identification of compounds. The use of large MS/MS libraries, in silico fragmentation and
retention prediction tools plays more and more an important role to achieve that goal. Compared to peptides
fragmentation, metabolites do often not generate fragment rich CID spectra for protonated and deprotonated precursor
ions. In addition, adduct precursor ions (Na, K,…) also do not fragment well which leaves many potential hits
unexploited. This calls for additional fragmentation techniques such as photodissociation (PD) or electron induced
dissociation (EID).

Methods

Collision induced and electron induction fragmentation spectra were acquired on a hybrid QqTOF (6600 TripleTOF,
Sciex, Concord, ON, Canada) equipped with a prototype ExD fragmentation cell [1]. Furthermore, a differential mobility
cell (SelexIon, Sciex) was used to improve precursor ion isolation. Positive mode ionisation was performed using a dual
APCI/ESI source and an APCI source both from Sciex. Chromatographic separation of metabolites and drug of abuses
was achieved using microLC in reverse phase mode

[1] A.O. Ducati, et al. Anal Chim Acta, 1150 (2021) 338207.

Preliminary data (results)

The use of chimeric CID/EID collision cell mounted on a quadrupole-time-of-flight mass spectrometer, showed that
additional valuable information could be obtained from EID compared to CID for most molecules analyzed as protonated
precursors. . Compared to CID, EID fragmentation of sodium and potassium adducts and multimers provided additional
informative fragments with good sensitivity. Furthermore, in most cases the radical cation was present with some
characteristic electron impact (EI) ionization fragments [1]. Furthermore, EID and CID can be acquired in the same LC-
MS run. This opens the possibility to take advantage of EI libraries to support structural elucidation or analyte
confirmation in combination with CID libraries.

While electrospray ionization is largely applied for the detection of polar compounds the use of atmospheric pressure
photoionization (APPI) has mainly been described with dopants for the analysis of compounds lacking polar functional
groups. With APPI depending on the ionization conditions either radical cation or protonated ions can be formed
selectively for polar and apolar analytes. Radical cation precursors can further be fragmented by collision induced
dissociation and the spectra showed numerous fragments common to that observed in EI spectra. Even more, good
matches could be achieved with direct EI libraries search. In the present work we propose and describe an approach
combining LC-ESI-CID/EID and LC-APPI-CID on the same platform for the generation of EI fragment rich spectra from
various precursor ion type in addition to conventional CID spectra to support analyte identification in metabolomics and
forensic.

Please explain why your abstract is innovative for mass spectrometry?

Generation of EI like spectrum using Electron Induced Dissociation and Atmopheric pressure Photoionization for LC-
MS/MS analysis of metabolites and drug of abuse.

Co-authors:

Patrick Mueller, University of Geneva, LSMS

David Ruskic, University of Geneva, LSMS

68
Session: LS-3 Lipidomics

LIPID IMAGING USING NANOSPRAY DESORPTION ELECTROSPRAY

IONIZATION (NANO-DESI) MASS SPECTROMETRY
Abstract ID: 765

Presenting author: Julia Laskin, Purdue University

Introduction

Mass spectrometry imaging (MSI) is a powerful technique for studying the localization of lipids, metabolites, and proteins
in biological samples. We have developed nanospray desorption electrospray ionization (nano-DESI), an ambient
ionization technique that relies on localized liquid extraction of analyte molecules from the sample into a liquid bridge
formed between two glass capillaries. The extracted analytes are transferred to a mass spectrometer inlet and ionized by
electrospray ionization. Nano-DESI enables quantitative imaging of biomolecules in fully hydrated samples with minimal
or no sample pre-treatment. Recent developments in the nano-DESI MSI instrumentation have enabled quantitative
imaging of lipids in tissues with high sensitivity and spatial resolution down to 8-10 micron. We have also developed
methods for isomer-selective imaging of lipids in tissues.

Methods

Nano-DESI MSI experiments were performed on a QExactive HF-X and Agilent’s 6560 IM-QTOF mass spectrometers
equipped with custom-designed imaging platforms. Imaging experiments employed both the capillary-based nano-DESI
probe and a microfluidic probe fabricated using traditional photolithography, wet etching, and thermal bonding. We have
optimized the composition of the extraction solvent to enhance both the extraction and ionization efficiency of lipids.
Coupling of nano-DESI MSI with ion mobility separation and online photochemical derivatization enables imaging of lipid
isomers in biological tissues.

Preliminary data (results)

We have achieved a spatial resolution of better than 10 microns using both the capillary and microfluidic probe.
Furthermore, we have optimized the composition of the extraction solvent to ensure the efficient extraction of neutral
lipids. This was achieved by adding a nonpolar component to the solvent. Furthermore, we have demonstrated a
pronounced enhancement of lipid signals by adding ammonium fluoride to the solvent. This was attributed to the efficient
proton transfer from lipids to fluoride anion at interfaces, which enhances the ionization efficiency. A photosensitized
reaction between singlet oxygen with C=C double bonds in acyl tails of fatty acids and phospholipids has been used to
obtain ion images of lipid isomers differing by the position of the double bond. This reaction converts lipids into lipid
hydroperoxides, which generate unique diagnostic fragments in collision-induced dissociation experiments. MS/MS
imaging is subsequently used for imaging of lipid isomers in biological samples. Collectively, these results have opened
up new opportunities for biological research

Please explain why your abstract is innovative for mass spectrometry?

Advanced development of nano-DESI MSI for quantitative isomer-resolved imaging of lipids in tissues with high spatial
resolution and throughput.

69
DYSREGULATION OF BLOOD LIPIDOME IN VARIOUS TYPES OF CANCER:
THE WAY TOWARDS EARLY CANCER SCREENING BY UHPSFC/MS
Abstract ID: 221

Presenting author: Michal Holčapek, University of Pardubice

Introduction

Lipids are important biomolecules that have numerous functions in human metabolism with a great potential in biomarker
research of pathological states related to lipids, such as cancer. The lipid class separation approaches (UHPSFC or
HILIC) coupled with MS provide the most convenient way for high-throughput lipidomic quantitation, because the
exogenous lipid class standards coelute with lipid species from the same class, which guarantees the same matrix effect
and therefore quite reliable determination of molar concentrations of numerous lipids [Anal. Chem. 90 (2018) 4249].
Recent works have shown that lipidomic profiles of human blood enable to differentiate cancer patients from healthy
controls.

Methods

UHPSFC/MS measurements were carried out on UPC2 system coupled to Synapt G2-Si QTOF from Waters [Anal.
Bioanal. Chem. 412 (2020) 2375]. MarkerLynx 4.1 was used for data preprocessing, and then data tables
(m/z vs. intensity) for each lipid class were exported and further processed using LipidQuant 1.0 [Bioinformatics 37
(2021) 4591]. The identified species were isotopically corrected and quantified by calculating the molar concentrations of
the lipid species based on the comparison of the intensity of a particular lipid with the intensity of IS of the same lipid
class of known concentration.

Preliminary data (results)

We have developed and validated several MS based methods for high-throughput clinical lipidomic quantitation using
coupling with ultrahigh-performance supercritical fluid chromatography (UHPSFC) or ultrahigh-performance liquid
chromatography (UHPLC). Several hundred lipid species are typically quantified in biological samples (plasma, serum,
tumor tissues, cell lines, urine, etc.), but the major issue in lipidomic quantitation is the data reliability over a longer period
of time, the comparability of results among different groups, and the absolute molar quantitation based on the use of the
right lipid class exogenous internal standards. The lipidomic quantitation is achieved by using at least one internal
standard per each lipid class and evaluated by laboratory-made LipidQuant 1.0 software. The comprehensive MS
determination of a wide range of blood lipids reveals statistically significant differences between various types of cancer
patients and healthy controls visualized by multivariate data analysis. The most extensive results are obtained for
pancreatic cancer [Nat. Com. 13. (2022) 124], which showed the dysregulation of some very long chain sphingomyelins,
ceramides, and (lyso)phosphatidylcholines. The sensitivity and specificity to diagnose pancreatic cancer were over 90%,
which outperforms CA 19-9, especially at an early stage, and is comparable to established diagnostic imaging methods.
Similar dysregulation patterns were observed for kidney, breast, and prostate cancers [Sci. Rep. 11 (2021) 20322].
Multivariate data analysis including nonsupervised PCA, supervised OPLS-DA, S-plots, box plots, and correlation
analysis are applied. This work was supported by Czech Science Foundation (GAČR) project No. 21-20238S.

Please explain why your abstract is innovative for mass spectrometry?

High-throughput mass spectrometry-based workflows based on lipid class separation revealsignificant changes of lipid
profiles for multiple cancer types, which suggests the possibility of early cancer screening.

Co-authors:

Denise Wolrab, University of Pardubice

Michaela Chocholoušková, University of Pardubice
Robert Jirásko, University of Pardubice
Ondřej Peterka, University of Pardubice
Jakub Idkowiak, University of Pardubice
Zuzana Vaňková, University of Pardubice

70
CHARACTERISATION OF CARDIOLIPINS IN MITOCHONDRIA OF HELA
CELLS BY HPLC-MS/MS
Abstract ID: 246

Presenting author: Vera Schwantes, University of Münster, Department of Inorganic and Analytical Chemistry,
Corrensstrasse 30, 48149 Münster, Germany

Introduction

Cardiolipins (CL) are characterised as a phospholipid subclass located exclusively in the inner mitochondrial membrane
of eucaryotic cells. They are involved in membrane stability and several biochemical processes. Current research results
indicate a connection between widespread disease patterns, such as cardiovascular diseases, diabetes or
neurodegenerative diseases, and the functionality of mitochondria as power plants of the cells. A particular focus is
thereby set on alterations in the natural cardiolipin distribution induced by oxidative stress, which can possibly lead to
mitochondrial dysfunction. In order to evaluate these alterations it is essential to elucidate the natural cardiolipin
distribution pattern.

Methods

Using cell culture, complex biochemical processes become accessible for analytical investigations. For this project,
mitochondria were isolated from HeLa cells, before lipid extraction was performed. These lipid extracts were then
analysed by reversed-phase high-performance liquid chromatography (RP-HPLC) hyphenated to electrospray ionisation-
high resolution mass spectrometry (HRMS). Chromatographic separation was optimised with special focus on modified
CL species such as monolysocardiolipins (MLCL), dilysocardiolipins (DLCL), and oxidation products. The
chromatographic method achieves sufficient separation of homologous CL species based on the fatty acid residues,
resulting in the suitability for further analysis of mitochondrial lipid extracts.

Preliminary data (results)

A total of 38 CL species were identified in the lipid extracts of mitochondria from HeLa cells. Detection by high-resolution
mass spectrometry (HRMS) allowed identification of the species by their accurate masses, confirmed by additionally
performed data-dependent MS/MS experiments. Based on characteristic fragments, bound fatty acid residues were
identified for most CL species. The analysed lipid extracts show a broad distribution of even numbered CL species,
ranging from 64 to 74 carbon atoms in the fatty acid residues with varying degrees of unsaturation. Additionally the effect
of different incubation conditions of HeLa cells, such as nutrient deficiency and oxidant exposition, on the CL pattern was
analysed.

Please explain why your abstract is innovative for mass spectrometry?

The combination of HRMS detection and optimized data-dependent MS/MS experiments allows reliable identification of
CL species in complex sample matrices.

Co-authors:

Jannis Wittke, University of Münster, Institute of Integrative Cell Biology and Physiology; Schlossplatz 5, 48149 Münster,
Germany
Frank Schmelter, University of Münster, Institute of Integrative Cell Biology and Physiology; Schlossplatz 5, 48149
Münster, Germany
Karin Busch, University of Münster, Institute of Integrative Cell Biology and Physiology; Schlossplatz 5, 48149 Münster,
Germany
Heiko Hayen, University of Münster, Department of Inorganic and Analytical Chemistry; Corrensstrasse 30, 48149
Münster, Germany

71
MS-BASED TARGETED PROFILING OF OXYLIPINS TO UNDERSTAND THE
EVOLUTION AND SEVERITY OF COVID-19
Abstract ID: 309

Presenting author: Denise Biagini, University of Pisa

Introduction

Over 200 million reported cases and over 5 million deaths have been globally ascribed to the ongoing coronavirus
disease (COVID-19) pandemic. Despite the unprecedented effort of the scientific community, there are still many open
questions regarding the pathophysiology of the disease, the therapy, and the management of patients.The key role of
inflammation in COVID-19 induced many authors to study the cytokine storm, whereas the role of the oxylipin storm is
still poorly understood. Oxylipins are bioactive lipids generated from both ω-3 and ω-6 polyunsaturated fatty acids
through enzymatic and non-enzymatic oxidation reactions. During inflammation, oxylipins switch from pro-inflammatory
effectors to both anti-inflammatory and specialized pro-resolving lipid mediators, which could promote its resolution.

Methods

In the present work, a very powerful analytical platform, based on micro-extraction by packed sorbent (MEPS) coupled to
liquid chromatography-tandem mass spectrometry (UHPLC-ESI-MS/MS), was proposed for the determination of 60
plasmatic oxylipins in a single run at ppt levels. Our powerful in-house oxylipidomics platform was successfully employed
for a comprehensive characterization of the inflammatory cascade in COVID-19. A chemometric approach was employed
to compare inflammatory metabolites coming from both cytokine and oxylipin storms between Intensive Care Unit (ICU)
and non-ICU patients.

Preliminary data (results)

Unlike cytokines, oxylipins provided such a clear separation between ICU and non-ICU patients, when considered in a
multivariate way by principal component analysis. Moreover, the presence nearby the ICU-cluster of subjects who would
have ended up in ICU from 1 to 4 days after the oxylipin quantitation, suggested a potential predictive role of our panel of
lipid mediators. A multivariate ROC curve was obtained by application of the UNEQ class-modelling strategy, thus
showing an area under the curve equal to 0.92. As far as we know, our observational study is one of the first to suggest
the importance of targeting the lipid mediator class switching for a timely picture of the patient ability to respond to SARS-
CoV-2 infection.

Please explain why your abstract is innovative for mass spectrometry?

This is the first analytical method combining a fast and reproducible MEPS procedure and a highly sensitive and
selective UHPLC-ESI-MS/MS system for the analysis of oxylipins in plasma samples.

Co-authors:

Paolo Oliveri, University of Genova

Maria Franzini, University of Pisa
Silvia Ghimenti, University of Pisa
Tommaso Lomonaco, University of Pisa
Andrea Bonini, University of Pisa
Federico Maria Vivaldi, University of Pisa
Lisa Macera, University of Pisa
Laurence Balas, Institut des Biomolécules Max Mousseron, IBMM
Thierry Durand, Institut des Biomolécules Max Mousseron, IBMM
Camille Oger, Institut des Biomolécules Max Mousseron, IBMM
Jean-Marie Galano, Institut des Biomolécules Max Mousseron, IBMM
Fabrizio Maggi, University of Insubria
Alessandro Celi, University of Pisa
Aldo Paolicchi, University of Pisa
Fabio Di Francesco, University of Pisa

72
CHARACTERIZING CONTENT AND LOCALIZATION OF COMPLEX
GANGLIOSIDE PHENOTYPES IN A GBA MODEL OF PARKINSON'S
DISEASE BY ORTHOGONAL HIGH RESOLUTION ION MOBILITY MASS
SPECTROMETRY AND MASS SPECTROMETRY IMAGING
Abstract ID: 338

Presenting author: Kim Ekroos, Lipidomics Consulting Ltd, President of the International Lipidomics Society

Introduction

Mutations in GBA1, encoding lysosomal glucocerebrosidase (GCase), are a risk factor for Parkinson’s disease (PD).
Reduced GCase function results in accumulation of glycosphingolipid substrates and imbalances in downstream
ganglioside metabolites. Brain ganglioside spatial distribution in individuals with GBA deficits is largely
unknown. Resolving complex ganglioside profiles and their spatial distribution is challenging due to a lack of analytical
platforms capable of dissecting the complex structural heterogeneity and presence of isomeric species within
endogenous ganglioside species. Here, we combine High Resolution Ion Mobility (HRIM) with LC-MS to quantify
gangliosides at the level of molecular species and with Matrix-assisted Laser Desorption/Ionization Mass Spectrometry
Imaging (MALDI-MSI) to localize their distributions in brains from wildtype and D409V heterozygous and homozygous
GBA mutant mouse model of PD-related GCase deficiency.

Methods

Two identical sets of age-matched brains from wildtype, D409V heterozygous and homozygous GBA mutant mice were;
1) sectioned for MSI analyses, and 2) ganglioside extracted using a chloroform/methanol (1:2, v/v) and solid-phase
extraction for HRIM analyses. Samples for MALDI-MSI were prepared with 5 mg/ml 2,5-dihydroxyacetophenone matrix in
50 mM ammonium sulfate 70% ethanol aq. and analyzed on an Orbitrap Elite MS setup. Ion mobility (IM) experiments
were performed on a commercial-equivalent HRIM (Structures for Lossless Ion Manipulation technology) instrument
using an Agilent 1290 LC - Agilent 6545-XT QTOF setup applying a 30-minute HILIC gradient.

Preliminary data (results)

HILIC separation prior to HRIM acquisition resulted in reduced ion suppression and better ganglioside coverage.
Compared to our previous work using flow injection analysis (FIA), the LC dimension enabled detection of 457 new
features including gangliosides belonging to 17 different classes. Using FIA alone, GD1a and GD1b isomers are resolved
via IM separation as HRIM has a resolving power sufficient to distinguish these species. With the addition of LC
separation, additional isomers are observed by retention time with additional isomeric features observed within the IM
spectrum that have yet to be identified. Thus, we could identify three isomers of GT1, likely corresponding to GT1a,
GT1b and GT1c species respectively. When comparing wildtype to D409V heterozygous and homozygous mutant mice,
we observed ganglioside phenotypes correlating with mutation status. GBA1 mutation leads to statistically significant
differences in e.g. GD1a d36:1 and GT1a d36:1, suggesting a selectivity towards the a-series metabolism. MSI analyses
revealed distinct distributions of ganglioside species in brains of D409V mice corresponding to GM, GD, GT and GQ
species. Notably, we find selective ganglioside brain distributions on both class and molecular species level. GT1 d36:1
is particularly distributed in the outer regions of the cortex and hippocampus whereas GT1 d38:1 localizes to the inner
Dentate gyrus granule cell layer of the hippocampus and cerebellum. In conclusion, our orthogonal approach provides
new information within the complex brain ganglioside lipidome and identifies novel regio-specific localization in brain,
thus providing novel insights into dysfunctional glycolipid metabolism in neurodegenerative diseases.

Please explain why your abstract is innovative for mass spectrometry?

In-depth characterization and regio-specific localization of brain gangliosides in a GBA mutant mouse model using next-
generation ion mobility instrumentation and mass spectrometry imaging.

Co-authors:

Shadrack M Mutuku, Molecular Horizons and School of Chemistry and Molecular Bioscience, University of Wollongong
Rachel A Harris, MOBILion Systems
Kelly L Wormwood Moser, MOBILion Systems
Michiel Vandenbosch, Maastricht MultiModal Molecular Imaging (M4I) Institute, Division of Imaging Mass Spectrometry,
Maastricht University
Frances Carroll, MOBILion Systems
Leon Yao, Merck & Co., Inc

73
Ron M.A. Heeren, Maastricht MultiModal Molecular Imaging (M4I) Institute, Division of Imaging Mass Spectrometry,
Maastricht University
Nathan G Hatcher, Merck & Co., Inc
Shane R Ellis, Molecular Horizons and School of Chemistry and Molecular Bioscience, University of Wollongong,
Illawarra Health and Medical Research Institute

GT1 d36:1(red)/d38:1(green) in sagittal section.

Ganglioside content in brain with GT1 isomers.

74
CEREBRAL ORGANOIDS MODEL SYSTEM FOR NEUROLOGICAL
DISEASES
Abstract ID: 638

Presenting author: Zdenek Spacil, Masaryk University

Introduction

The advent of mass spectrometry-based lipidomics allowed for the exploration and profiling of neural tissue lipids in
biological samples with high selectivity, sensitivity, and multiplexing capacity. Neural membrane lipids and, in particular,
glycosphingolipids serve as primary traits of neuronal development, maturation, and degeneration. However, their
characterization remains challenging. Here, we present a workflow to profile significant classes of membrane lipids and
glycosphingolipids in cerebral organoidscultured from induced pluripotent stem cells. The cerebral complexity, missing in
2D cell cultures, was recently achieved with 3D cell culture. Cerebral organoids are organ-like structures recapitulating,
with a remarkable degree of detail, development, and degeneration of the human CNS, making them a unique tool for
disease modeling.

Methods

A single CO washed with PBS was used for lipidomics analysis; four biological replicates were analyzed in technical
duplicates. CO was homogenized, sonicated, and vortexed to determine the total protein content. Lipids were extracted
with IPA, and two-fold diluted, adding isotopically labeled standards. The sample was injected on the UHPLC system
equipped with C18 pre-column, and analytical column (CSHTM, 5 x 2.1 mm x 1.7 μm and 50 x 2.1 mm x 1.7 μm, Waters
Corp) thermostated to 40 °C. UHPLC system was coupled to a triple quadrupole mass spectrometer (Agilent 6495B) in
selected reaction monitoring (SRM) mode.

Preliminary data (results)

We applied mass spectrometry lipidomics to profile neuronal membrane lipids in a single cerebral organoid. Raw data
were processed in the Skyline (Version 20.1.0.76, MacCoss Lab., UW, USA). We determined the response factor for all
lipids relative to labeled internal standards to calculate concentrations reported as the average of two technical replicates
and relative to the level of a housekeeping protein. We characterized the composition of essential membrane lipids and
glycosphingolipids in the neural tissue and profiled lipid-based traits associated with developmental and
neurodegeneration stages of neural stem cells, radial glial cells, neurons, and astrocytes. Cerebral organoids revealed
temporal trends in membrane lipid composition relevant to study aging and neurodegeneration. For instance, levels of
neuron-specific glycosphingolipids increased with neuronal maturation and subsequently declined with the onset of
neurodegeneration. Other lipid levels were elevated at late time points of cultivation, indicating their functional role in the
growing astrocyte population. We are the first to report a workflow for a comprehensive lipid profiling in cerebral
organoids and preliminary data on changes to membrane lipid composition due to maturation/aging, varied
cytoarchitecture, and lipid perturbations relevant to neurodegenerative diseases.

Acknowledgment
This work was funded by The Grant Agency of the Masaryk University (project No. MUNI/G/1131/2017), the Ministry of
Health of the Czech Republic (NV19-08-00472), the RECETOX research infrastructure (No. LM2018121) financed by the
Ministry of Education, Youth and Sports, and Operational Programme Research, Development, and Innovation - project
CETOCOEN EXCELLENCE (No. CZ.02.1.01/0.0/0.0/17_043/ 0009632) and project CETOCOEN+ (No. CZ.02.1.01/0.0/
0.0/15_003/0000469).

Please explain why your abstract is innovative for mass spectrometry?

The first report on comprehensive lipidomics profiling and temporal trends and perturbation in membrane lipid
composition in cerebral organoids

Co-authors:

Durga Jha, Masaryk University

Marketa Nezvedova, Masaryk University
Darshak Gadara, Masaryk University

75
Session: LS-4 Single cell MS and in cell MS

EXPLORING FUNCTIONAL PROTEIN COVARIATION ACROSS SINGLE

CELLS
Abstract ID: 792

Presenting author: Nikolai Slavov, Northeastern University

Introduction

Biological functions are reflected in the natural variation of proteome configurations across individual cells. Emerging
single-cell proteomics methods may decode this variation and empower inference of biological mechanisms with minimal
assumptions.

Methods

I will describe the development and application of both established and emerging single-cell mass-spectrometry
methods, including prioritized single-cell proteomics, which can analyze thousands of proteins selected in order of their
priority for the biological question and project of interest. Furthermore, I will describe plexDIA, a method that enables
parallel analysis of both single cells and peptides. It quantifies about a thousand proteins per single cell while
parallelizing the analysis of both single cells and peptides and achiveing 98% data completeness within a set.

Preliminary data (results)

These methods allowed us to interpret protein covariation in different biological systems, including primary macrophages
and melanoma cells. The focus of my talk will be on conceptual innovations and strategies for data acquisition and
interpretation that make single-cell protein analysis accessible, robust and highly quantitative.

Please explain why your abstract is innovative for mass spectrometry?

Prioritized single-cell proteomics ensures duty-cycle time for analyzing prioritized peptides across all samples (thus
increasing data consistency) while analyzing identifiable peptides at full duty-cycle, thus increasing proteome coverage
2-fold.

76
SINGLE CELL QUANTITATIVE PROTEOMICS
Abstract ID: 563

Presenting author: David Goodlett, University of Victoria

Introduction

Single cell (sc) and near single cell (nsc) proteomics remain far from trivial for most laboratories and require further
refinement for more routine application. We build on previous sc and nsc proteomics protocols to assess the utility of
targeted proteomics for measuring heterogeneity in red blood cells (RBCs). We adapted the pioneering developments,
like nanodroplet processing in one-pot for trace samples (nanoPOTS) and automated preparation in one-pot for trace
samples (autoPOTS), to allow single cell proteomics on proteins with copy numbers <10 6.

Methods

autoPOTS was simplified replacing surfactant with 10% organic solvent and using heat and bath sonication to lyse and
emulsify proteins in a denatured state that also served to achieve high efficiency rapid proteolysis using heat stable
Trypsin/LysC, which eliminated the need for robotic liquid handling instrumentation. This one-pot sample preparation
method was combined with isotope labeled synthetic peptide standards for targeted quantitative isotope dilution parallel
reaction monitoring LC-MS (IDMS-PRM) to quantify Hemoglobin in RBCs in 2.5 h from sample preparation to data
analysis.

Preliminary data (results)

A one-pot sample preparation method was combined with isotope labeled synthetic peptide standards for targeted
quantitative isotope dilution parallel reaction monitoring LC-MS (IDMS-PRM) to quantify Hemoglobin in RBCs.
Specifically, an IDMS-PRM assay was developed to measure the endogenous N-terminal proteolytic peptide of
hemoglobin beta subunit and it’s N-terminal valine carboxymethyl (CMV) adduct, which were quantified in single digit red
blood cells (RBCs), isolated via limiting-dilution (LD) or a CellenONE single cell dispenser. Either isolation method
allowed the endogenous concentration of the N-terminal proteolytic peptide of hemoglobin beta subunit, and by inference
the hemoglobin tetramer, to be measured in single red blood cells (540-660 amol/RBC), which was comparable to the
calculated SI reference range for mean corpuscular hemoglobin (390-540 amol/RBC). Moreover, it was observed that
repeated measures using 5-25 RBCs could be used as an alternative and less difficult strategy than single RBC analysis
to measure protein heterogeneity, which ranged from 0.1-0.8% in 5-25 RBCs, with an interindividual coefficient of
variation ranging from 47—215%. In conclusion, a rapid, one-pot and relatively simplified autoPOTS protocol is
presented here for targeted quantification of proteins to measure red blood cell heterogeneity

Please explain why your abstract is innovative for mass spectrometry?

quantitation of proteins in a single cell by mass spectrometry

Co-authors:

Azad Eshghi, Department of Chemistry and Applied Biosciences, ETH Zurich, 8093 Zurich, Switzerland
Xiaofeng Xie, Brigham Young University
Darryl Hardie, University of Victoria Genome BC Proteomic Centre
Rachael Newman, University of Victoria Genome BC Proteomic Centre
Kelly Curtis, Discovery Clinical Services Ltd
Fabiana Izaguirre, Cellenion
Michael Chen, Vancouver Island Health Authority
Ying Zhu, Pacific Northwest National Laboratory
Ryan Kelly, Brigham Young University

77
MASS SPECTROMETRY IMAGING AND PROFILING OF SINGLE CELLS:
APPLICATION IN BREAST CANCER RESEARCH
Abstract ID: 172

Presenting author: Eva Cuypers, M4i, University Maastricht

Introduction

Single cell technologies in cancer research are recently strongly developing since these enhance the molecular
understanding. Biomolecules such as lipids are known to play a major role in cancer biology. Being able to visualize their
distribution on a single cell level will open up new molecular information on the development, growth and heterogeneous
composition of cancer cells. We present a single-cell imaging mass spectrometry method as powerful tool to map
biomolecule distribution.

Methods

Breast cancer cell lines were cultured, plated onto coated ITO slides and incubated overnight. Washing and drying steps
were applied. Prior to imaging, norhamane matrix was sublimed onto the cell containing slide. Single cell chemical
profiling was achieved using 5 µm spatial resolution MALDI imaging coupled to high mass resolution detectors (FTICR &
TimsTOF). SCiLS Lab software (Bruker) was used for visualization and statistical analysis.

Preliminary data (results)

The full chemical composition of breast cancer cell lines, including lipids identification, is generated on a single cell level.
Thanks to the applied imaging technique, lipid distributions within a single cell are visualized. Based on statistical
analysis of the single cell profiles, we generated a model that is able to distinguish different cell types 'on the fly' during
the imaging run.

Please explain why your abstract is innovative for mass spectrometry?

Cellular differences and compound distribution can be detected, possibly leading to the identification of specific
molecular pathways and biomarkers. Moreover, on-the-fly cell type recognition open-up digital pathology possibilities.

Co-authors:

Rianne Biemans, The M-Lab, Dep. Precision Medicine, University Maastricht

Natasja Lieuwes, The M-Lab, Dep. Precision Medicine, University Maastricht
Ludwig Dubois, The M-Lab, Dep. Precision Medicine, University Maastricht
Ron M.A. Heeren, M4i, University Maastricht

78
Single cell imaging mass spectrometry workflow

IMS images of single cell compound distributions

79
Analyzing cell type-specific dynamics of metabolism in kidney repair
Abstract ID: 200

Presenting author: Gangqi Wang, Department of Internal Medicine (Nephrology) & Einthoven Laboratory of
Vascular and Regenerative Medicine, Leiden University Medical Center

Introduction

Mass spectrometry-based single-cell metabolomics methods have been proposed previously, including those applying
matrix-assisted laser desorption/ionization mass spectrometry imaging (MALDI-MSI). To date, MALDI-MSI-based
metabolomics studies have focused on analyzing ‘static snapshots’ providing valuable information on in-situ metabolic
heterogeneity, however, crucially lack insight into the dynamic component of cellular metabolism. For a truly
comprehensive understanding of the interplay between biochemical alterations and cellular function, metabolic fluxes
and kinetic interpretations at single-cell level are mandatory. Here we describe a platform based on isotope tracing and
MALDI-MSI that allows in-situ dynamic measurementsof metabolism at single-cell resolution, and thus unravel cell
metabolism within its tissue architecture.

Methods

We applied different 13C-isotope-labeled nutrients on vibratome slices of fresh mouse kidney. MALDI-MSI at single-cell
resolution (i.e. pixel size of 5 × 5 µm2) was then applied to detect metabolites and lipids from the harvested tissues.
Following MALDI-MSI analysis, post-MSI-analyzed sections were stained and subsequently imaged using multiplexed
immunofluorescence (IF) microscopy for cell-type identification. An imputed dataset containing the complete 13C-labeling
information from different timepoints, and nutrients was established by integrating different datasets using k-nearest
neighbor analysis based on their lipid profile. Finally, flux analysis including metabolic rate and pathway convergence
was performed on single pixels and visualized in pseudo-images.

Preliminary data (results)

We show that our platform can map cell-type specific dynamic changes in the central carbon metabolism, as well as the
contribution of different nutrients to cellular energy metabolism on a complex heterogenous tissue like the kidney. In
combination with multiplexed immunofluorescence staining, we can detect the metabolic flux changes and nutrient
partitioning in targeted cell types. We observed differences in metabolic dynamics of kidney cells between renal cortex
and the medulla, revealing that proximal- and distal tubules from the cortex had a higher TCA cycle flux and lower
glycolysis flux compared to the medullar cells. In addition to different renal cell-types, metabolic heterogeneity of different
cell phenotypes within one cell-type can also be studied using this platform. For example, glomerular endothelial cells
could be singled out from other cortical renal endothelial cells by their higher glycolytic- and lower TCA cycle flux. This
platform allows to achieve single-cell resolution in situ and hence interpret the single cell metabolic kinetics in the context
of structure and metabolism of neighboring cells.

Please explain why your abstract is innovative for mass spectrometry?

This is a first-in-class platform to allow the study of spatial single cell metabolic flux within freshly derived organ sections.

Co-authors:

Bram Heijs, Center of Proteomics and Metabolomics, Leiden University Medical Center
Sarantos Kostidis, Center of Proteomics and Metabolomics, Leiden University Medical Center
Ahmed Mahfouz, Department of Human Genetics, Leiden University Medical Center, Leiden, The Netherlands, Leiden
Computational Biology Center, Leiden University Medical Center, Leiden, The Netherlands, Delft Bioinformatics Lab,
Delft University of Technology, Delft, The Netherlands
Rosalie G.J. Rietjens, Department of Internal Medicine (Nephrology) & Einthoven Laboratory of Vascular and
Regenerative Medicine, Leiden University Medical Center
Cathelijne W. van den Berg, Department of Internal Medicine (Nephrology) & Einthoven Laboratory of Vascular and
Regenerative Medicine, Leiden University Medical Center
Sébastien J. Dumas, Laboratory of Angiogenesis and Vascular Metabolism, Department of Oncology, KU Leuven and
Center for Cancer Biology, VIB, Leuven, Belgium.
Peter Carmeliet, Laboratory of Angiogenesis and Vascular Metabolism, Department of Oncology, KU Leuven and Center
for Cancer Biology, VIB, Leuven, Belgium., Laboratory of Angiogenesis and Vascular Heterogeneity, Department of
Biomedicine, Aarhus University, Aarhus 8000, Denmark.
Martin Giera, Center of Proteomics and Metabolomics, Leiden University Medical Center, Leiden, The Netherlands
Bernard M. van den Berg, Department of Internal Medicine (Nephrology) & Einthoven Laboratory of Vascular and
Regenerative Medicine, Leiden University Medical Center, Leiden, The Netherlands

80
Ton J. Rabelink, Department of Internal Medicine (Nephrology) & Einthoven Laboratory of Vascular and Regenerative
Medicine, Leiden University Medical Center, Leiden, The Netherlands

Workflow of single cell dynamic metabolic measurements and analysis.

81
SINGLE-CELL MALDI MASS SPECTROMETRY IMAGING ENABLES AN IN-
DEPTH VIEW OF MOLECULAR HETEROGENEITY IN CELL CULTURES
Abstract ID: 170

Presenting author: Jan Schwenzfeier, Institute of Hygiene, University of Münster, Robert-Koch-Str. 41, 48149
Münster, Germany

Introduction

Cell cultures express an inherent molecular heterogeneity between cells due to their different growing environment,
growth state, age or differing gene expressions. Despite this, most of the methods employed to investigate the molecular
composition of cell cultures are performed using bulk analysis, thereby ignoring the heterogeneous aspect within the
population. Recent developments in the field of MALDI mass spectrometry imaging enable a pixel size of 10 µm and
below that enable the spatial analysis of single cells. Combining this mass spectrometry dimensionality with other
imaging modalities like fluorescence microscopy allows to generate single-cell mass spectra which can be used to
investigate the molecular composition and inter- or intra-cellular heterogeneities within cell cultures using statistical
analysis and machine learning.

Methods

Cells were grown in culture flasks and then sewn in 8-well chamber slides for ease of use in a slide based workflow.
Vero-B4 and Caki-2 cells were grown as mono- as well as co-cultures. After cultivation cells were fixed with formalin and
stained with Hoechst and WGA fluorescence dyes. Afterwards cells were washed with PBS and NH 4OAc to remove salts
and were left for drying overnight. Microscopic images were obtained on an Olympus Slide Scanner before matrix
(DHAP) was applied via sublimation. MALDI-2-MSI was performed on a timsTOF fleX mass spectrometer at 8 µm pixel
resolution.

Preliminary data (results)

In our approach, generation of single-cell mass spectrometry data is based on co-registration of MALDI-MSI data with
corresponding fluorescence microscopy images. To ensure full integrity of the cells, microscopy images are recorded
prior to matrix application using staining protocols compatible with MALDI-MSI analysis of lipids. As a first step a
segmentation is performed on fluorescence microscopy images based on nuclei detection and a watershed algorithm to
determine the cell position and borders utilizing specific staining of cell nuclei and membrane, respectively. The second
step performs an overlay of the MALDI and fluorescence images using binarization and maximal cross-correlation.
Lastly, individual mass spectra are generated for each single cell by combining all pixels of the respective segmented
area. Pixels registered to more than one cell are discarded. Single-cell mass spectra were used for statistical analysis on
a single-cell level. In a proof-of-principle study, we successfully differentiated two different renal cell lines (Vero-B4 and
Caki-2) in a co-culture experiment, based on their molecular profiles using machine learning (support vector machine)
and multivariate statistical tools (principal components analysis). With the molecular information at hand, it was also
possible to investigate intercellular heterogeneity of certain lipid species within the co-culture by the use of histograms
that visualize the molecular distribution in the population. Co-registration of MSI and optical modality furthermore allowed
to compare and connect morphological features to the molecular phenotype of each individual cells within the
investigated culture.

Please explain why your abstract is innovative for mass spectrometry?

The approach combines the high spatial resolution of microscopy with the large depth of information in mass
spectrometry to investigate cellular heterogeneity in cell culture on a single cell level.

Co-authors:

Tanja Bien, Institute of Hygiene, University of Münster, Robert-Koch-Str. 41, 48149 Münster, Germany, Interdisciplinary
Center for Clinical Research (IZKF), University of Münster, Domagkstr. 3, 48149 Münster, Germany
Krischan Koerfer, Institute for Psychology and Otto Creutzfeldt Center for Cognitive and Behavioural Neuroscience,
University of Münster, Fliednerstr. 21, 48149 Münster, Germany
Klaus Dreisewerd, Institute of Hygiene, University of Münster, Robert-Koch-Str. 41, 48149 Münster, Germany,
Interdisciplinary Center for Clinical Research (IZKF), University of Münster, Domagkstr. 3, 48149 Münster, Germany
Jens Soltwisch, Institute of Hygiene, University of Münster, Robert-Koch-Str. 41, 48149 Münster, Germany,
Interdisciplinary Center for Clinical Research (IZKF), University of Münster, Domagkstr. 3, 48149 Münster, Germany

82
Workflow for single-cell analysis by combination of microscopy and MALDI-MSI.

83
SINGLE CELL MULTIMODAL IMAGING FOR EVALUATION OF CELLULAR
METABOLISM IN HUMAN FOCAL EPILEPSY
Abstract ID: 479

Presenting author: Isabeau Vermeulen, The Maastricht MultiModal Molecular Imaging Institute (M4i),

Introduction

Temporal lobe epilepsy is one of the most common types of refractory focal epilepsy. To identify and define the seizure
foci, fluorodeoxyglucose (FDG) – positron emission tomographic (PET) imaging is standard practice. FDG-PET is a non-
invasive neuroimaging technique that monitors brain’s glucose consumption. FDG-PET scans show in most but not all
cases focal hypometabolism in seizure foci (PET+). These scans are used to make presurgical decisions, making PET-
cases a significant problem in clinical practice. The goal of this study is to apply a multimodal imaging technique to
identify specific molecular profiles (lipidome, proteome) in hypometabolism (PET+) and normal metabolism (PET-)
epilepsy patients to understand the molecular mechanisms at a single cell level using mass spectrometry imaging (MSI)
and immunohistochemical staining (IHC).

Methods

Fresh-frozen human brain sections were sublimated (norharmane) and analyzed at 20µm and 5µm resolution on a
rapifleX. Data dependent acquisition was analyzed on an Orbitrap Elite hybrid ion trap for lipid identification. After MSI,
laser capture microdissection (LMD) was performed using a Leica LMD 7000 instrument. LMD samples were trypsin-
digested and further processed for LC-MS/MS peptide analysis. Data analysis was done in Scils, Lipostar and Proteome
Discover. A multiplex immunofluorescent staining was achieved using different antibodies. Staining images were
acquired with Nikon Ti-E microscope. Co-registration of MSI data and (immuno)fluorescent data is established using an
in-house MATLAB script.

Preliminary data (results)

Fresh frozen human hippocampus and neocortex were obtained upon resective treatment for temporal lobe epilepsy. In
these patients, the epileptic focus is hippocampal based on their EEG. The (hypo)metabolic FDG-PET region included
the ipsilateral temporal lobe. MSI lipidomics data of the hippocampus and neocortex clearly showed distinct features in
the lipid profile of PET+ and PET- epilepsy patients (4 samples per group). This information was based on the DDA data
where we could identify phosphocholines, phosphoethanolamines, phosphoinositols, phosphoserines, sphingomyelins,
triacylglycerols, prostaglandins and glycerophosphates. Further, proteomics analysis revealed distinctively different
profiles between PET+ and PET- epilepsy patients. The immunofluorescence and MSI data were co-registered using
fiducial markers on the slide. The combination of both imaging modalities made it possible to localize the different cell
types and extract the different lipid profiles: adult neurons, microglia, astrocytes. This multimodal setup can help to
understand the molecular profile of the individual cells and the overall molecular profile in the epileptic region of
hypometabolism and normal metabolism epilepsy patients. This work demonstrates that MSI and immunofluorescent
staining in a multimodal set-up can be applied in diagnostic settings. The distinct molecular profiles could potentially aid
in diagnostic decision making in the future.

Please explain why your abstract is innovative for mass spectrometry?

Single cell MSI combined with fluorescence staining that defines the molecular profile of individual cells can be applied to
differentiate hypo- from normo metabolism in focal epilepsy.

Co-authors:

Tobias Dancker, The Maastricht MultiModal Molecular Imaging Institute (M4i),

Govert Hoogland, Department of Neurosurgery, Maastricht University Medical Center, School of Mental Health and
Neuroscience
Kim Rijkers, Department of Neurosurgery, Maastricht University Medical Center, School of Mental Health and
Neuroscience
Olaf Schijns, Department of Neurosurgery, Maastricht University Medical Center , School of Mental Health and
Neuroscience, Academic Center for Epileptology (ACE), Maastricht University Medical Center
Cuypers Eva, The Maastricht MultiModal Molecular Imaging Institute (M4i),
Berta Cillero-Pastor, The Maastricht MultiModal Molecular Imaging Institute (M4i),

84
Session: FP-1 Toxicology and Metabolism

MASS SPECTROMETRY IN VENOM CHEMISTRY

Abstract ID: 421

Presenting author: Manjunatha Kini, Department of Biological Sciences, Faculty of Science, National University
of Singapore, Singapore 117543

Introduction

Venoms and toxins have involved in various animals to assist them in the acquiring food and predator/competitor
deterrence. We have been interested in structure-function relationships, mechanism of action and evolution of protein
toxins. Over the last several decades, toxin research has contributed significantly to both basic and applied sciences.

Methods

I will discuss the contributions of mass spectrometry in two main areas of toxin research: (a) Venom profiling and
identification of novel toxins and (b) antivenomics.

Preliminary data (results)

Mass spectrometry and proteomics play important roles in defining the toxin profiles of a venom, and the changes in
these profiles in relationship with ontogeny, geography and captivity. These studies help in the identification of novel
toxins. Structure-function relationships and mechanism of action of novel toxins contributes to the design and
development of therapeutic agents to treat various cardiovascular and neuromuscular diseases. Antivenomics
contributes in the development of better strategies for the production of antivenoms.

Please explain why your abstract is innovative for mass spectrometry?

In addition, I will discuss various strategies we use in venom profiling and identification of several novel toxins from
animal venoms and development of therapeutic agents.

85
MOLECULAR NETWORKING APPROACH WITH METWORK WEBSERVER
FOR DRUG ANNOTATION : BUILDING AN ARTIFICIAL INTELLIGENCE IN
SILICO DATABASE OF ALL COMMERCIALIZED DRUGS IN FRANCE AS AN
INNOVATIVE AND IDEAL ADDITION FOR CLINICAL TOXICOLOGY
Abstract ID: 377

Presenting author: Emmanuel Bourgogne, Hôpital Bichat, Laboratoire de Pharmacologie/toxicologie, Université

de Paris, faculté de Pharmacie, laboratoire de toxicologie

Introduction

In Toxicology, analysts are confronted with complex problems, resulting in important clinical consequences. Untargeted
screening remains a challenge, given the high number of molecules to be detected. LC-HRMS, the reference method for
screening, generates a large volume of high quality spectral data, without tools for visualizing and organizing MS data.
Another major issue lies in the absence of an exhaustive MS/MS database. Our objectives were to i) apply molecular
networking (MN) for screening interpretation, ii) build an in silico MS library of all commercialized drugs in France, iii)
apply this database (DB) for drug’s identification on a panel of 60 hospitalized patient’s plasma and iv) finally use
MetWork webserver for complementary metabolites identification.

Methods

Patient’s plasma was crushed with acetonitrile. After centrifugation, supernatant was injected onto an LC-Orbitrap
QExactive LC-HRMS system. For MS, positive MS and two data-dependent MS/MS scans were recorded using CID and
HCD activation types. For MN, data were generated using MZmine, analyzed and visualized using Metgem. MetWork
annotations were filtered and this file was used to annotate the previously obtained MN. In parallel, an in silico database
of all commercialized drugs in France was build using a python script and interrogated for drug annotation.

Preliminary data (results)

The rationale was to build a MS/MS in silico database of all commercialized drugs in France. Steps involved i) extraction
of smiles structures of all drugs from an open access dataset (data.gouv.fr website), ii) generation of in silico MS/MS
spectra using CFM-ID, iii) final compilation of all spectra in a MGF format to create this database, which will be
interrogated to annotate patient’s MN.

First, comparison was made between our experimental DB (containing 155 drugs/metabolites) and the in-silico MS/MS
spectra. CFM-ID gave comparable results in 50% of cases. In a second step, for patient, additional drugs annotations
were possible, enlarging the number of annotated compounds, even without standards available. Through a 60 patients
retrospective study, in addition to common drugs of abuse, benzodiazepines or neuropsychotics, drugs from antiepileptic,
antiretroviral, antifungal, immunosuppressant were found. In polymedicamented patients, those drugs are known to be
subject to drug-drug interactions (DDI). Having a better understanding of these DDI will improve patient management.
Finally, using MetWork, additional metabolite annotations were possible, such as mephedrone major metabolites; phase
2 metabolites like glucuronides; isobaric drugs like tramadol, amitriptyline, venlafaxine and their metabolites were
identified in patients without ambiguity.

For untargeted screening, our results show that MN opens perspectives in toxicology using LC-HRMS. Available
databases are now the major bottleneck for DDI. Combined with CFM-ID for generating DB and MetWork to extend the
annotation of new potential drugs or old drug’s metabolites, even without reference standard, it could help clinicians to
explain potential toxicity.

Please explain why your abstract is innovative for mass spectrometry?

In-silico database generation with CFM-ID and MetWork extend drug annotation and opens perspectives for molecular
networking approach in clinical analysis and toxicology

Co-authors:

romain magny, Université de Paris, faculté de Pharmacie, UMR 8038, Sorbonne Université UM80, INSERM UMR 968,
CNRS UMR 7210, Institut de la Vision, IHU ForeSight,
yann beauxis, Université de Paris, faculté de Pharmacie, UMR 8038
sacha guilhaumou, Université de Paris, faculté de Pharmacie, laboratoire de toxicologie
gregory genta-jouve, USR 3456 CNRS LEEISA
86
HYPHENATION OF ELECTROCHEMISTRY AND MASS SPECTROMETRY
FOR THE SIMULATION OF METABOLIC PROCESSES AND THE
GENERATION OF STABLE ISOTOPE LABELLED METABOLITE
STANDARDS
Abstract ID: 232

Presenting author: Valentin Göldner, University of Münster, Institute of Inorganic and Analytical Chemistry,
University of Münster, International Graduate School for Battery Chemistry, Characterization, Analysis,
Recycling and Application (BACCARA)

Introduction

During the safety evaluation of xenobiotics like drugs, pesticides or dietary supplements, the comprehensive elucidation
of their metabolic pathways is crucial. The online hyphenation of electrochemistry and mass spectrometry (EC/MS) has
proven to be a powerful tool for the simulation of oxidative and reductive metabolic processes. If a metabolic reaction is
mimicked successfully using EC/MS, the electrochemical conversion may be scaled up in order to generate milligram
amounts of the metabolite-like transformation products. These products can be used as reference substances in MS-
and HPLC/MS-based qualitative and quantitative analyses.

Methods

The oxidative or reductive transformation products of xenobiotics are generated and detected in an online hyphenation of
a commercially available electrochemical thin layer cell and a high resolution mass spectrometer. Transformation
products correlating to known metabolites of the xenobiotics are selected and the electrochemical transformation is
scaled up using an in-house built electrochemical cell comprising a working electrode made from porous graphite. The
optimized scaled up transformation method is applied to stable isotope labelled xenobiotics in order to generate labelled
metabolite standards, which find application in mass spectrometry based metabolism studies after purification using
semipreparative HPLC.

Preliminary data (results)

An electrochemical flow-through cell comprising an exchangeable porous graphite electrode is built and applied for
metabolism mimicry at the semipreparative scale. The cell scales up established, so-called “coulometric” electrochemical
cells from an analytical to a semipreparative scale. The method development process is supported by mass spectrometry
and online hyphenation of EC/MS and HPLC/MS giving insight in the underlying processes. Problems like electrode
fouling and cross contamination between syntheses are addressed by the modular cell design allowing an easy
exchange of the disposable working electrodes. The applicability of the developed cell for the synthesis of metabolite
standards required in mass spectrometric metabolism studies is demonstrated by the use of isotopically labelled
paracetamol as a model compound. Paracetamol is oxidized electrochemically in the developed cell and the resulting
transformation product is trapped by the addition of glutathione, thus forming an isotopically labelled standard of the
biologically relevant phase II metabolite 3-glutathionyl-paracetamol. Conversion rates of 39 mg/h and a yield of 71% are
achieved for the generated phase II metabolite. This highlights the potential of the method for aiding mass spectrometry
based metabolism studies by providing means for the synthesis of reference materials within only a few hours of
synthesis time. Due to the straightforward synthesis in a “top-down” approach, even the generation of stable isotope
labelled metabolite standards can be achieved in a cost efficient way by using the labelled xenobiotic itself as a
substrate.

Please explain why your abstract is innovative for mass spectrometry?

Mass spectrometry based metabolism studies are facilitated by this straightforward and cost efficient method for the
synthesis of stable isotope labelled standard materials.

Co-authors:

Jens Fangmeyer, University of Münster, Institute of Inorganic and Analytical Chemistry

Uwe Karst, University of Münster, Institute of Inorganic and Analytical Chemistry, University of Münster, International
Graduate School for Battery Chemistry, Characterization, Analysis, Recycling and Application (BACCARA)

87
Schematic view of the developed electrochemical cell.

88
A “CHEMICAL TOOLBOX” FOR THE GENERATION OF METABOLITE-LIKE
PRODUCTS OF NEW PSYCHOACTIVE SUBSTANCES
Abstract ID: 634

Presenting author: Peng Che, Vrije Universiteit Amsterdam, Amsterdam Institute of Molecular and Life Sciences
(AIMMS), Division of BioAnalytical Chemistry, de Boelelaan 1085, 1081 HV Amsterdam, The Netherlands, Center
for Analytical Sciences Amsterdam (CASA), Amsterdam, The Netherlands

Introduction

In the last decades, new psychoactive substances (NPS) have stepped in the recreational drug market. Due to the little
knowledge available on their effects and possible toxicity, NPS pose a serious threat to the public health. Notably, the
metabolism of NPS remains largely unknown. Moreover, for most NPS, no standards are commercially available (both
for the parent drugs and respective metabolites). However, NPS metabolites are highly relevant since metabolites
assessed in bio- or environmental samples bring additional forensic information on drug consumptions, compared with
the parent drugs. Most NPS Phase I metabolites are the result of oxidation reactions. In this study, we developed a
unique “chemical toolbox” able to chemically produce phase I metabolites that are typically also formed during metabolic
incubations.

Methods

NPS metabolites generated by rat liver microsomal incubations and metabolite-like products of the NPS obtained with
the chemical oxidant reagents (e.g., hydrogen peroxide, potassium permanganate and cumene hydroperoxide) were
analyzed using liquid chromatography - quadrupole time-of-flight mass spectrometry (LC-qTOF/MS). MS/MS data were
acquired, enabling the elucidation of biotransformation sites to identify NPS metabolites. The generated metabolite-like
products were compared with the real metabolites formed from the set of NPS under study (i.e., synthetic cathinones and
amphetamines). Finally, oxidation products corresponding to the metabolites obtained with microsomal incubations were
purified using solid-phase extraction or liquid-liquid extraction followed by liquid chromatography.

Preliminary data (results)

The tentative structures of the identified phase I metabolites from 3-methylmethcathinone (3-MMC) were elucidated
based on their MS and MS/MS spectra and are shown as an example in Fig 1 and 2. The general metabolism pathways
seen for several NPS mainly included de-alkylation of the amine group, hydroxylation and N-oxidation. Combinations of
multiple modifications were also observed for secondary metabolites. Results of the different oxidant reagents were
compared with the microsomal incubations, from which some metabolite-like products (de-alkylation products,
hydroxylation products and N-oxidation products) were selected for purification. With the proposed “Chemical Toolbox”,
we aim at offering a novel toolbox for straightforward chemical production of sufficient quantities of metabolites, which
are generated biologically during NPS biotransformation, to be used as analytical standards. This innovative “Chemical
Toolbox” will be highly beneficial in drug metabolism studies and other metabolomics-based approaches, providing a
protocol for the straightforward generation of metabolites of NPS and other drugs which are not commercially available.

Please explain why your abstract is innovative for mass spectrometry?

LC-high resolution-MS plays a crucial role in the metabolite-identification workflow and enables to study the correlation
between metabolites formed via microsomal incubations and metabolite-like products generated from chemical
oxidations.

Co-authors:

Kristina Still, Vrije Universiteit Amsterdam, Amsterdam Institute of Molecular and Life Sciences (AIMMS), Division of
BioAnalytical Chemistry, de Boelelaan 1085, 1081 HV Amsterdam, The Netherlands, Center for Analytical Sciences
Amsterdam (CASA), Amsterdam, The Netherlands
Wilfried Niessen, Vrije Universiteit Amsterdam, Amsterdam Institute of Molecular and Life Sciences (AIMMS), Division of
BioAnalytical Chemistry, de Boelelaan 1085, 1081 HV Amsterdam, The Netherlands, hyphen MassSpec, Margrietstraat
34, 2215 HJ Voorhout, The Netherlands
Jeroen Kool, Vrije Universiteit Amsterdam, Amsterdam Institute of Molecular and Life Sciences (AIMMS), Division of
BioAnalytical Chemistry, de Boelelaan 1085, 1081 HV Amsterdam, The Netherlands, Center for Analytical Sciences
Amsterdam (CASA), Amsterdam, The Netherlands
Isabelle Kohler, Vrije Universiteit Amsterdam, Amsterdam Institute of Molecular and Life Sciences (AIMMS), Division of
BioAnalytical Chemistry, de Boelelaan 1085, 1081 HV Amsterdam, The Netherlands, Center for Analytical Sciences
Amsterdam (CASA), Amsterdam, The Netherlands

89
Extracted ion chromatograms (EICs) of 3-MMC and its metabolites

MS/MS spectra of 3-MMC and its four metabolites

90
MULTIOMICS PIPELINE REVEALS ALPHA-KETOGLUTARATE AS A
COUNTERMEASURE FOR VX ORGANOPHOSPHATE POISONING
Abstract ID: 535

Presenting author: Phillip Mach, LANL

Introduction

Organophosphates (OPs) represent a wide range of structurally-related chemistries that have seen wide use as both
commercial insecticides and chemical warfare agents (CWAs). Their primary mechanism of action is widely accepted as
‘phosphorylation’ of acetylcholinesterase (AChE)’s catalytic site by the OP resulting in AChE inhibition; this inhibition
causes respiratory failure and death.

However, further research has suggested that OPs are more promiscuous than conventionally believed, and that some
secondary mechanism of action likely plays a significant role in OP toxicity. In the current study, shotgun proteomics and
untargeted metabolomics are performed on blood plasma samples obtained serially for 14 days from guinea pigs which
were intravenously exposed to 0.4 LD50 of VX.

Methods

Six guinea pigs were intravenously exposed to 0.4 LD50 of VX. Twenty-four hours prior to exposure, a control blood
sample was taken from each animal to serve as control samples. After exposure, samples were taken at 1 hour, 6 hours,
24 hours, 48 hours, 4 days, 7 days, 10 days and 14 days post-exposure. The blood was separated into plasma, and the
plasma was utilized for analysis.

Untargeted samples were analyzed via a Dionex 3000 liquid chromatography pump and a Thermo QE Plus Orbitrap &
Eclipse mass spectrometer.

Preliminary data (results)

More than 2200 proteins were identified across all animals in the study. By applying multivariate adaptive regression
splines (MARS), 72 unique proteins were identified with significant temporal profiles over the duration of the study.
Further filtering of these proteins was done by considering only proteins that changed +/- 2-fold in at least four of the six
animals. Twelve unique proteins were identified, and eight out of those 12 are directly involved in energy processes.
Several of those proteins are linked to lipid metabolic processes; for example, both apolipoprotein C-I and C-II were
identified and shown to decrease throughout the study starting at the six-hour time point, and their lowest concentration
occurring at 24-hours post exposure. Other significant proteins were identified to be directly involved in glycoloysis and
TCA cycle. IDH1 was significantly upregulated immediately following exposure to VX and remained elevated for two days
post exposure. IDH1 is a cytosolic form of isocitrate dehydrogenase which also exists within the mitrochondria (IDH2)
where it is critical for generated α-ketogluarate.

Verification of results were performed on an Agilent Excellegence platform with human cardiomyocytes. Dose response
and prophylatic effects were evaluated.

Please explain why your abstract is innovative for mass spectrometry?

Multiomics pipeline reveals that VX chemical warfare agent inhibits TCA cycle steps in addition to inhibition of
cholinesterase.

Co-authors:

Trevor Glaros, LANL

Elizabeth Dhummakupt, CBC
Ethan McBride, LANL

91
Metabolic and proteomic disruption of glycolysis/TCA
following VX exposure.

Primary cardiomyocytes as a model for OP exposure.

92
COMPOUND DEGRADATION STUDIES FOR A WIDE RANGE OF
MOLECULE SIZES AND HRMS DATA SOURCE
Abstract ID: 447

Presenting author: Ismael Zamora, Lead Molecular Design, S.L., Pompeu Fabra University

Introduction

The study of compound degradation in the context of Drug metabolism or/and Chemical degradation studies have an
important role in drug discovery with an impact in the development of a new chemical entities. These studies have 2
components: one related to the quantification and the calculation of compound clearance or degradation constants, and
another aspect that is related to the elucidation of the products (degradants or metabolites) formed during the
experiment. We are presenting in here a solution developed over the last 15 years that considers both aspects for
molecules of any size and for any vendor and acquisition mode data source.

Methods

A collection of algorithms was developed to perform the two major tasks that are needed for peak detection and structure
assignment. The peak detection algorithms consider the specific needs for each compound class, i.e., Monoisotopic
mass for molecules smaller than 4000 amu, the most abundant isotope for molecules until 20.000 amu or the average
mass for bigger ones. The structure assignment algorithm has been developed to be able to work with a set of atoms
and bonds like the ones described in small molecule or monomers like in oligonucleotides, peptides or proteins or a
combination of both molecule representations.

Preliminary data (results)

Using the developed approaches, we analyzed a collection of experimental data for a set of small molecules,
protacspeptides, oligonucleotides, or proteins where the data was collected on a Waters Synapt/Xevo/Vion/Bioaccord,
ThermoQ-Exactive, Agilent QTof, SciexTripleTof/Zeno, Bruker TimmTof instrument. The degradation studies presented
were done on in-vitro metabolite systems like human liver microsome, human hepatocyte, protease incubation or forced
degradation analysis.

Results will be presented for each of the compound classes:a) small molecular metabolism in Human liver microsome for
nefazodone measured in 4 different instrument/acquisition modes (Waters SynaptMSe, Sciex Tripe Tof SWATH, Bruker
QTof and Aglient All Ions); b) force degradation studies for lanzoprazol measured in Agilent QTof and Thermo Q-
Exactive; c) Protac Hepatocyte degradation study on a Thermo DDS d) Somatostatin metabolism study using a Waters
acquisition, e) Semaglutide degradation from a Waters Vion; f) Oligonucleotide (sense and antisense chain analysis)
from a ThermoDDS analysis.

In all cases the computation of the clearance and/or degradation contact will be presented together with a collection of
the metabolites found. The analysis of the fragmented data will be shown for one case of the Nefazone and the
oligonucleotide studies.

Please explain why your abstract is innovative for mass spectrometry?

Computation approach for data processing that is vendor/acquisition mode neutral able to analysis compounds of any
molecular size.

Co-authors:

Fabien Fontaine, Lead Molecular Design, S.L.

Tatiana Radchenko, Lead Molecular Design, S.L.
Luca Morettoni, Lead Molecular Design, S.L.
Nadia Zara, Molecular Horizon
Albert Garriga, Lead Molecular Design, S.L.
Pol Giménez, Lead Molecular Design, S.L.
Xavier Pascual, Lead Molecular Design, S.L.
Vera Lopéz, Lead Molecular Design, S.L.
Ramon Adalia, Lead Molecular Design, S.L.

93
A 3 sep procedure for peak finding and structure elucidation

94
Session: IM-4 Data sciences in
MS/AI/Chemometrics/identification/modelling - Session A

EMPOWERING LARGE CHEMICAL KNOWLEDGE BASES FOR

EXPOSOMICS: PUBCHEMLITE MEETS METFRAG
Abstract ID: 774

Presenting author: Emma Schymanski, LCSB, University of Luxembourg

Introduction

Exposomics researchers need to identify relevant chemicals covering the entirety of potential exposures over entire
lifetimes. With over 100 million chemicals in the largest chemical databases, coupled with broadly acknowledged
knowledge gaps, researchers are faced with too much—yet not enough—information at the same time. Improvements in
analytical technologies and computational mass spectrometry workflows coupled with the rapid growth in databases and
increasing demand for high throughput “big data” services from the research community present significant challenges
for both data hosts and workflow developers. The “PubChemLite for Exposomics” collection reduces candidate search
spaces in non-target small molecule identification workflows while increasing content usability. This allows users to profit
from both increasing size and information content of large compound databases, along with increased efficiency.

Methods

PubChemLite for Exposomics is a dynamic collection of ~380,000 chemicals that is built weekly from several categories
of annotation content in the PubChem database that are highly relevant for exposomics analysis. These ten categories
are: Agrochemical Information; Associated Disorders and Diseases; Biomolecular Interactions and Pathways; Drug and
Medication Information; Food Additives and Ingredients; Identification; Pharmacology and Biochemistry; Safety and
Hazards; Toxicity; Use and Manufacturing. Benchmarking datasets are used to show how experimental knowledge and
existing datasets can help detect and fill gaps in compound databases to progressively improve large resources such as
PubChem, and topic-specific subsets such as PubChemLite.

Preliminary data (results)

Interdisciplinary efforts and data sharing can facilitate research in exposomics and beyond. This effort demonstrates this
using the examples of PubChem (https://pubchem.ncbi.nlm.nih.gov/), the NORMAN Network Suspect List Exchange
(https://www.norman-network.com/nds/SLE/) and the in silico fragmentation approach MetFrag (https://msbi.ipb-
halle.de/MetFrag/). A subset of the PubChem database relevant for exposomics, PubChemLite for Exposomics, is
presented as a database resource that can be (and has been) integrated into current workflows for high resolution mass
spectrometry. The benchmarking analysis performed demonstrated dramatic performance improvements over MetFrag
coupled with PubChem, both in terms of candidate ranking (improving ranking to 81 % candidates correct in first place)
and runtime. PubChemLite for Exposomics is a living collection, updating as annotation content in PubChem is updated
(exemplified using the NORMAN Suspect List Exchange), and exported to allow direct integration into existing workflows
such as MetFrag. The dataset is currently built and checked weekly and updated monthly on Zenodo
(DOI: 10.5281/zenodo.5995885). It is integrated into the MetFrag Web interface (https://msbi.ipb-halle.de/MetFrag/) and
available for download for command line use. It is also integrated into the open source mass spectrometry workflow
patRoon (https://rickhelmus.github.io/patRoon/). The source code and files necessary to create and adjust this collection
are jointly hosted between the research parties. This effort shows that enhancing the FAIRness (Findability, Accessibility,
Interoperability and Reusability) of open resources can mutually enhance several resources for whole community benefit.

Please explain why your abstract is innovative for mass spectrometry?

PubChemLite for Exposomics is a dynamic collection of ~380,000 chemicals formed from several annotation categories
in PubChem designed to empower exposomics analysis using computational high resolution mass spectrometry

Co-authors:

Todor Kondic, LCSB, University of Luxembourg

Steffen Neumann, IPB Halle
Paul Thiessen, NCBI/NLM/NIH
Jian Zhang, NCBI/NLM/NIH
Evan Bolton, NCBI/NLM/NIH

95
PubChem Compound Table of Contents and Annotation content forming PubChemLite

Influence of annotation content (green) and MS2 data on performance

96
DEVELOPMENT OF DIAGNOSTIC TESTS FOR COVID-19 USING MALDI(+)
FT-ICR MS COMBINED WITH MACHINE LEARNING
Abstract ID: 687

Presenting author: Wanderson Romão, Federal Institute of Espírito Santo

Introduction

Rapid identification of the virus as SARS-CoV-2 is of utmost importance in combating outbreaks and epidemics. The
detection and diagnosis of the disease are essential to monitor the spread and to stop the transmission routes. 1 the
MALDI FT-ICR MS (matrix-assisted laser desorption ionization Fourier-Transform Ion Cyclotron Resonance Mass
Spectrometry) technique with the aid of chemometrics tools like machine learning may be promising in the classification
of positive samples for COVID-19 in complex matrices as biological samples with high accuracy.2

In this context, our work aims to develop rapid and efficient methods to perform screening of patients with suspected
COVID-19, analyzing saliva samples by MALDI FT-ICR MS technique combined with Support vector machines (SVM).

Methods

Initially, it was evaluation the volume of the trypsin used (5 or 10μL) and the time of protein digestion (2h and 16h).

Time spent on the analysis and the numbers of signals were evaluated in the MALDI FT-ICR analysis, using different
resolving power (512k, 1M, 2M, 4M, and 8M).

After optimization, 97 positive and 52 negative saliva samples were analyzed. The models PCA and SVM were
constructed by Matlab software and the variables were selected by Fisher's Discriminant. 3 And for SVM models, the
sample set was divided into calibration (70%) and prediction (30%) by the Kennard Stone method.

Preliminary data (results)

In the sample preparation optimization, the full scan spectra of samples that used 10uL of trypsin for digestion showed a
larger number of signals with a signal to noise ratio (S/N) > 4. the protein digestion in 16h also had a larger number of
signals. However, the shorter time (2h) was satisfactory and speeds up the diagnostic process in 14h.

In addition, using 10uL of trypsin and 2h of protein digestion in the sample preparation has a higher TIC for all peptide
sequences. For the MALDI FT-ICR MS analysis, using 1M resolving power got better results. It was showed a larger
number of signals (221 signals) in an individual analysis at a time of 60 seconds.

Two SVM models showed the best results, SVM1, and SVM2. The calibration group obtained 100% accuracy and the
test group 95.6% (SVM1) and 86.7% (SVM2). SVM1 selected 780 variables and has a False Negative Rate (TFN) of 0%,
while SVM2 selected just two variables (525.4 Da and 1410.8 Da) with a TFN of 3%.

Thus, the methodology created is non-invasive and avoids false-negative diagnoses. In addition, the better SVM model
had a sensitivity of 100%, specificity of 87.5%, the accuracy of 95.6%.

It was also observed that the signals with m/z lower than 1000 have higher intensity in positive samples and signals with
m/z greater than 1000 in negative samples.

Please explain why your abstract is innovative for mass spectrometry?

The development of a methodology using MALDI FT-ICR MS combined with machine learning can provide rapid results
with highly accurate to classify samples of diseases.

Co-authors:

Camila M. de Almeida, Petroleomic and Forensic Chemistry Laboratory, Department of Chemistry, Federal University of
Espírito Santo
Larissa C. Motta, Petroleomic and Forensic Chemistry Laboratory, Department of Chemistry, Federal University of
Espírito Santo
Gabriely S. Folli, Chemometrics Laboratory of the Center of Competence in Petroleum Chemistry – NCQP, Federal

97
University of Espírito Santo (UFES)
Wena D. Marcarini, Department of Physiological Sciences, Federal University of Espírito Santo (UFES)
Camila A. Costa, Federal University of Goiás – UFGO, School of Dentistry, Department of Stomatology
Carolina S. Vilela, Federal University of Goiás – UFGO, School of Dentistry, Department of Stomatology (Oral Pathology)
Valério G. Barauna, Department of Physiological Sciences, Federal University of Espírito Santo (UFES)
Luciene C. G. Campos, Department of Biological Science, Santa Cruz State University
Nádia L. Costa, Federal University of Goiás – UFGO, School of Dentistry, Department of Stomatology
Paula F. Vassallo, Clinical Hospital, Federal University of Minas Gerais, Belo Horizonte/MG
Denise C. Endringer, Department of Pharmacy, Universidade Vila Velha – UVV
José G. Mill, Department of Physiological Sciences, Federal University of Espírito Santo (UFES)
Paulo R Filgueiras, Chemometrics Laboratory of the Center of Competence in Petroleum Chemistry – NCQP, Federal
University of Espírito Santo (UFES)

MALDI (+) spectra among positive and negative samples.

PCA scores and loadings plots of MALDI(+) spectra

98
RELATIONAL GRAPH CONVOLUTIONAL NETWORK FOR ROBUST MASS
SPECTRUM CLASSIFICATION
Abstract ID: 144

Presenting author: Raphaël La Rocca, Mass Spectrometry Laboratory - ULiège

Introduction

Mass spectrometry and mass spectrometry imaging are methods of growing interest for non-targeted rapid detection of
different molecular classes from samples such as tissues, microorganisms, and more recently single-cells. Common
techniques rely on machine learning to classify mass spectra, which always depend on data specific preprocessing. The
recent use of deep learning with convolutional neural networks (CNN) shows promising results by avoiding the
preprocessing step. However, CNNs are not suited for high resolution mass spectrometry as the number of parameters
linearly increases with mass resolution. In this work, we propose a novel relational graph convolutional neural network
(R-GCN) to efficiently classify spectra according to their mass patterns.

Methods

Based on a mass spectrum, we build a graph where the peak centroids are represented as nodes. For each node, we
associate 3 features: the centroid intensity, its mass, and its estimated mass defect. We connect the nodes when their
mass difference matches a given set of edges whose type corresponds to commonly observed chemical transformations,
for example +H2, +CH2, or +O2. Then, we apply a 2-layer relational graph convolution to enhance the node embeddings.
Finally, we pool these embeddings to provide a compact representation of the graph for the classifying dense neural
network (Figure 1).

Preliminary data (results)

We evaluate our method on the classification task of several datasets, and compare it with CNN architectures and
machine learning approaches. The datasets from [1] are mainly low resolution. Hence, we introduce several high-
resolution image datasets, representing cell cultures on a glass or bacteria colonies on agar, acquired on a MALDI FT-
ICR-MS and a MALDI TOF-MS. The classification task consists of separating cells or bacteria from the background.

We use the same evaluation method as [1]. Our model approaches the performances of the CNN and sometimes
surpasses the classical machine learning approaches on most low-resolution datasets. However, on high resolution data,
our model performs better than the CNN as shown in Figure 2.

The advantages of our network are threefold: (1) it has only 3,000 parameters compared to the 7 millions of the CNN
architecture, making our method train and infer faster. (2) Our network is robust to mass shifts by design, whereas for
CNNs this advantage comes at the cost of a higher mass resolution. (3) CNNs such as those presented in [1] require a
mass binning of 0.1 Da to keep the network at a viable size, while our network can directly be applied without decreasing
the spectral resolution.

Further work will focus on automatically selecting the set of mass patterns from the dataset, and testing our method on
diverse high resolution datasets.

[1] Seddiki, K., et al. Cumulative learning enables convolutional neural network representations for small mass
spectrometry data classification. Nat. Commun. 11, 5595, 2020.

Please explain why your abstract is innovative for mass spectrometry?

We design a more efficient graph neural network fitted for mass patterns which improves on the previous state-of-the-art
CNN architecture for high-resolution mass spectrometry classification.

Co-authors:

Anthony Cioppa, TELIM - Montefiore Institute - ULiège

Marc Van Droogenbroeck, TELIM - Montefiore Institute - ULiège
Gauthier Eppe, Mass Spectrometry Laboratory - ULiège
Loïc Quinton, Mass Spectrometry Laboratory - ULiège

99
Figure 1. Overview of our pipeline for mass spectrum classification.

Figure 2. Classification accuracy for different datasets and methods.

100
UNIVERSAL FRAGMENTATION MODEL FOR TANDEM MASS
SPECTROMETRY BASED MOLECULAR STRUCTURE ELUCIDATION
Abstract ID: 259

Presenting author: Bela Paizs, The Rosalind Franklin Institute, deShape ltd.

Introduction

The current gold standard for tandem mass spectrometry (MS/MS) based molecular structure elucidation is based on
comparison of observed fragmentation patterns to those obtained for known compounds collected in spectral libraries.
Alternatively, computed fragmentation characteristics generated for entries in molecular structure databases might also
be matched against spectra of unknowns. The computational strategies applied range from simple bond-cutting
algorithms to generate plausible fragments to sophisticated machine learning approaches leading to complete spectra.
Here we introduce the Universal Fragmentation Model (UFM) for organic molecules that is a generalization of our
previous ‘Pathways in Competition’ (PIC) peptide fragmentation model (Paizs & Suhai, Mass Spectrom. Rev. 2005).
UFM is based on detailed consideration of the gas-phase ion chemistries governing fragmentation and modelling of the
associated phenomena.

Methods

UFM considers the entire complexity of the fragmentation process including excitation of molecular ions, structural
changes happening upon energy uptake and dissociation. UFM works with 3D molecular structures for both ionic and
neutral species appearing on fragmentation pathways. We use gas-phase ion chemistry to predict plausible
fragmentation patterns which are then assessed by modelling (using both classical and quantum chemical calculations)
to generate structures and energies. Based on these UFM also considers competition of the associated chemistries to
identify the most likely fragment ions of dissociating species and is capable of generating multiple-stage MSn patterns.

Preliminary data (results)

For [M+H]+ species UFM considers all plausible protonation sites and explores the energetics of protonation,
rearrangements by hydride transfer, ring formation and opening, dissociations from the original and rearranged
structures and re-association of fragments. To facilitate structural elucidation a software implementation of UFM termed
deFrag was developed that reads 2D molecular structures and automatically performs 3D computations associated with
UFM. The results of these complex calculations are concisely summarized as ‘computed fragmentation patterns’ (CFPs)
which are collected in libraries and searched by matching observed spectra against them.

Currently UFM and deFrag are capable to rapidly predict high quality fragmentation patterns for a range of molecules
which are usually available only in detailed labelling/spectroscopic/computational studies. For example, UFM accurately
predicts all the fragmentation chemistries for phenylalanine as described by El Aribi et al. (J. Phys. Chem. A 2004, 108,
3844). The fragmentation of the phenylalanine isomer 2-(hydroxymethyl)-N-methyl benzamide is dominated by species
at m/z 148, 119, and 91; these are all rearrangement products created by dissociations coupled to ring formation and
opening and hydride transfer, respectively. UFM and deFrag precisely predict this unique fragmentation pattern. When
searched against CFPs computed for a number of phenylalanine isomers QToF spectra of both phenylalanine and
benzocaine (another phenylalanine isomer) safely identify these species using quantum chemically computed
dissociation energies for scoring.

This presentation will discuss UFM’s performance on the CASMI 2016 dataset (Shymanski et al. J. Cheminform 2017)
and applications in metabolomics and natural product chemistry.

Please explain why your abstract is innovative for mass spectrometry?

Comprehensive fragmentation model capable of predicting accurate fragmentation patterns for general molecules for
MS/MS based molecular structure elucidation

Co-authors:

Daniel McGill, The Rosalind Franklin Institute

Daniel Simon, The Rosalind Franklin Institute
Josephine Bunch, The Rosalind Franklin Institute
Zoltan Takats, The Rosalind Franklin Institute, deShape ltd.

101
MSCLASSIFR: AN R PACKAGE FOR SUPERVISED CLASSIFICATION OF
MASS SPECTRA WITH MACHINE LEARNING METHODS
Abstract ID: 124

Presenting author: Alexandre Godmer, Département de Bactériologie, Hôpital Saint-Antoine, Sorbonne

université, Sorbonne Université, INSERM, Centre d’Immunologie et des Maladies Infectieuses, U1135

Introduction

In clinical microbiology, the identification of pathogens is essential to provide a reliable etiologic diagnosis and to offer
adequate treatments to patients. Allowing rapid identifications of pathogens with a relative low cost, Matrix Assisted
Laser Desorption Ionization - Time of Flight (MALDI-TOF) mass spectrometry is a technique that has spread in hospitals
all over the world. This technique consists in comparing a mass spectrum, acquired after the ionization of intact proteins
of a bacterial culture coming from a biological sample, to referenced mass spectra profiles. Although this technique
works well for many species, the performance of commercial systems remains disappointing to distinguish spectrally
close organisms, as it can often be the case between normal and antibiotic resistance strains, or between genetically
close sub-species.

Methods

Open source R packages already exist to preprocess MALDI-TOF mass spectra (MALDIquant, MALDIrppa), but
additional steps must be applied to classify pre-processed mass spectra, the main ones being the selection of
discriminating peaks, and the selection of an optimal mass spectra classification model. In this context, we developed
MSclassifR with the objective of providing a free and easy-to-use software allowing the construction of optimal data
analysis pipelines to routinely classify mass spectra in laboratory. It is freely downloadable from the CRAN (https://cran.r-
project.org/web/packages/MSclassifR/index.html) and will be regularly updated.

Preliminary data (results)

The MSclassifR package offers R functions that concerned three main topics (see Figure): (i) pre-processing of mass-
spectra and peack-picking ; (ii) selection of discriminative mass-over-charge values (m/z); (iii) estimation of a
classification model using several machine-learning methods (linear logistic regression, neural networks, random forests,
support vector machines or eXtrem Gradient boosting) to predict the category of a mass spectrum from shortlisted mass-
over-charge values. Each of these steps are associated with a function containing an optimized data analysis pipeline.

We applied the functions of our package to mass spectra of several bacterial species, notably the Mycobacterium
abscessus and Burkholderia cepacia complexes, and obtained classification models achieving accuracy close to 100%,
while algorithms provided by manufacturers have difficulty in providing reliable results. Thus, MSclassifR constitutes an
appealing alternative to commercial systems (Bruker Biotyper / Vitek MS) for more reliable clinical diagnosis and
ultimately better treatment of hospital patients.

Of note, although developed in the context of MALDI-TOF spectra, our R package can also be used to find discriminant
peaks and to build classification models from MS1 spectra in LC-MS/MS analyses.

Please explain why your abstract is innovative for mass spectrometry?

MSclassifR is easy-to-use and freely downloadable from the CRAN. It constitutes an alternative to commercial systems
(Bruker Biotyper / Vitek MS) for more reliable MALDI-TOF based diagnosis in routine laboratory.

Co-authors:

Mariette Matondo, Institut Pasteur, Université de Paris, Proteomics Platform, Mass Spectrometry for Biology Unit, USR
CNRS 2000
Alexandra Aubry, Sorbonne Université, INSERM, Centre d’Immunologie et des Maladies Infectieuses, U1135, AP-HP,
Hôpital Pitié-Salpêtrière, Centre National de Référence des Mycobactéries et de la Résistance des Mycobactéries aux
Antituberculeux
Quentin Giai Gianetto, Institut Pasteur, Université de Paris, Proteomics Platform, Mass Spectrometry for Biology Unit,
USR CNRS 2000, Institut Pasteur, Université de Paris, Bioinformatics and Biostatistics HUB, Computational Biology
Department

102
Workflow and typical output of the MSClassifR R package.

103
MODULAR ANTIBODY DE NOVO SEQUENCE ANALYSIS USING MULTI-
TIER PROTEOMICS DATA
Abstract ID: 53

Presenting author: Bastiaan De Graaf, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for
Biomolecular Research and Utrecht Institute for Pharmaceutical Sciences, University of Utrecht, Padualaan 8,
Utrecht 3584 CH, the Netherlands, Netherlands Proteomic Center, Padualaan 8, Utrecht 3584 CH, the
Netherlands

Introduction

While mass spectrometry (MS) is regarded the method of choice for sequencing proteins from cells and tissues,
sequencing polyclonal antibody mixtures still poses a significant unmet challenge. This is because sequencing is typically
performed by peptide-centric methods, which are impeded by the high sequence homology in between various antibody
sequences. Protein-centric methods, such as top-down and middle-down MS, represent an alternative that provides the
link between fragments and precursor molecules, although the field is still inmature, and the de novo sequence analysis
at the protein level remains challenging. Nevertheless, distinct MS tiers yield fruitful and complementary sequence
information. Here, we assess the power of combining peptide-centric and protein-centric MS/MS data for the fully
automated de novo sequencing of antibodies in polyclonal mixtures.

Methods

We process top-down/middle-down and bottom-up MS data to yield full length sequence predictions. We select five
template sequences for each analyzed precursor chain using ETD LC-MS/MS of reduced antibody chains. These are
used to untangle the de novo peptide predictions obtained by using multiple proteases and HCD/EThcD LC-MS/MS runs,
grouping them per antibody chain. Sequence tags in the top-down spectra are extended using the de novo peptide
predictions to yield long confident sequence predictions. We assess these in a modular fashion based on the germline
domains they cover, then recombine the results into full-length antibody sequences.

Preliminary data (results)

We analyzed several samples of increasing complexity, namely monoclonal IgG1s, mixtures of monoclonal IgG1, and
polyclonal antibody mixtures.

For a recombinant monoclonal sample (TZB), filtering the V-region database with bottom-up MS spectra yields 66 hits
out of 250 possibilities. We use fragments from the top-down MS and a custom sliding window fragment matching
algorithm to rank the remaining V-regions. The top-5 best scoring V-regions are chosen as templates for automatic
sequence reconstruction and have an average of 75% (heavy chain (HC)) and 85% (light chain (LC)) sequence identity .
A C-region is selected based on direct mass matching of C-terminal fragments, and a J-region is selected through
alignment to bottom-up de novo peptides. An accurate reconstruction of the target sequence is achieved for both the LC
and HC by reconstructing the framework region sequences, then filling the remaining gaps with mass matching de novo
reads.

Similarly accurate templates and full length sequence predictions, were obtained when TZB was analyzed in a mixture
with two other monoclonal antibodies.

When sequencing a clone from serum from a sepsis patient, the top-5 V-region templates had an average of 73.98%
(HC) and 82.42% (LC) sequence identity. The selected J-regions are 75% (HC) and 100% (LC) correct, and the selected
C-region is correct for both chains.

While long, accurate (>95%) predictions are obtained through automatic sequence reconstruction, contaminant clones
still hamper getting complete antibody sequences in a fully automated manner. However, we were able to obtain the
sequence through manual inspection of the data.

Please explain why your abstract is innovative for mass spectrometry?

Our pipeline aims at fully automated sequencing of antibodies from polyclonal mixtures with multi-tier MS, using data-
driven assignment of peptides to precursor chains, thereby reducing complexity of full sequence reconstruction.

Co-authors:

104
Sem Tamara, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research and Utrecht
Institute for Pharmaceutical Sciences, University of Utrecht, Padualaan 8, Utrecht 3584 CH, the Netherlands,
Netherlands Proteomic Center, Padualaan 8, Utrecht 3584 CH, the Netherlands
Albert Heck, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research and Utrecht
Institute for Pharmaceutical Sciences, University of Utrecht, Padualaan 8, Utrecht 3584 CH, the Netherlands,
Netherlands Proteomic Center, Padualaan 8, Utrecht 3584 CH, the Netherlands

Overview of the modular antibody de novo sequencing experimental workflow

Modular antibody sequencing pipeline: Top-down and Bottom-up data processing

105
106
Tuesday 30 August 2022: 11:00 – 13:00
Session: HC-1 Young MS Scientists - Session A

BUILDING A MULTIMODAL MOLECULAR ATLAS OF THE HUMAN KIDNEY

Abstract ID: 694

Presenting author: Elizabeth Kathleen Neumann, Department of Biochemistry, Vanderbilt University, Mass
Spectrometry Research Center, Vanderbilt University

Introduction

The human kidney is composed of over 26 cell types that actively coordinate to form higher order structures, such as
glomeruli and tubules. While the scientific community generally understand the roles of these cell types and structures, it
is not known how these cells vary molecularly and numerically throughout a single organ or between organs within the
human population, particularly as a function of demographic. Understanding both the underlying similarities and
differences between these demographics has major ramifications for functional efficiency, transition to disease, and
disease severity. We developed and employed a multimodal approach consisting of imaging mass spectrometry,
multiplexed immunofluorescence, autofluorescence microscopy, and histopathology to explore the molecular and cellular
differences between human kidneys as a function of sex and BMI.

Methods

Human kidney tissue was flash frozen, embedded in carboxymethylcellulose, and cryosectioned to a 10 µm thickness.
Autofluorescence microscopy was acquired. Samples for MALDI IMS were coated with DAN using an HTX TM Sprayer.
MALDI IMS was performed on a timsTOF fleX MS system in positive and negative ion modes at a 10 μm pixel size. For
multiplexed immunofluorescence, tissues underwent antigen retrieval, rehydration, fixation, primary antibody incubation,
and further fixation. The CODEX system automatically dispensed complementary, fluorescent barcodes. Microscopy was
performed using EGFP, DAPI, Cy7 and DsRed filters. Data were processed using a combination of commercial and in-
house software.

Preliminary data (results)

Here, we developed a workflow consisting of sample preparation, pathological assessment, multiplexed

immunofluorescence, MALDI imaging mass spectrometry, and histological staining to obtain rich molecular and cellular
profiles of 17 human kidneys. In total, we sampled from both men and women between the ages of 21 to 77 years old.
The size and depth of coverage enabled creation of the most extensive atlas of the human kidney to date, including >200
identified lipids and 23 antibody labels. We then used supervised and unsupervised approaches to create molecular and
cellular profiles of glomeruli and proximal tubules. By correlating our suite of analytical approaches with MALDI IMS, we
have hypothesized the physiological roles of cell type defining lipid features. For example, phosphatidylinositol lipids
were globally detected with lower abundance in female glomeruli with the largest differences observed in PI(34:2),
PI(36:4), and PI(32:0). Based on multimodal data, we hypothesize that these molecules are critical substrates for
prostaglandin synthesis in glomerular mesangial cells.The function of most lipids remains unknown because they are
difficult to assess, demonstrating the importance and power of the approach for understanding the role of discrete lipids.
Moreover, the established methodology is demonstrated for the human kidney but can be expanded and validated for
use in other organ systems, such as the heart or spleen, or animal models, such as the mouse. Because the workflow is
flexible, a variety of samples can be accommodated for high content atlas creation and exploration of biological systems.

Please explain why your abstract is innovative for mass spectrometry?

An atlas pertaining to human health and consisting of multiple analytical approaches, including mass spectrometry,
involving highly reproducibly data acquisition and integration.

Co-authors:

Nathan Heath Patterson, Department of Biochemistry, Vanderbilt University, Mass Spectrometry Research Center,
Vanderbilt University
Leonoor E.M. Tideman, Delft Center for Systems and Control, Delft University of Technology
Jamie L Allen, Mass Spectrometry Research Center, Vanderbilt University
Lukasz G. Migas, Delft Center for Systems and Control, Delft University of Technology
Madeline E Colley, Department of Biochemistry, Vanderbilt University, Mass Spectrometry Research Center, Vanderbilt

107
University
Emilio S. Rivera, Department of Biochemistry, Vanderbilt University, Mass Spectrometry Research Center, Vanderbilt
University
Melissa A. Farrow, Mass Spectrometry Research Center, Vanderbilt University
Carrie Romer, Mass Spectrometry Research Center, Vanderbilt University
Haichun Yang, Division of Nephrology and Hypertension, Department of Medicine, Vanderbilt University Medical Center
Maya Brewer, Division of Nephrology and Hypertension, Department of Medicine, Vanderbilt University Medical Center
Danielle B. Gutierrez, Mass Spectrometry Research Center, Vanderbilt University
Mark P. de Caestecker, Division of Nephrology and Hypertension, Department of Medicine, Vanderbilt University Medical
Center
Agnes B. Fogo, Division of Nephrology and Hypertension, Department of Medicine, Vanderbilt University Medical Center,
Department of Pathology, Microbiology and Immunology, Vanderbilt University Medical Center, Departments of Medicine
and Pediatrics, Vanderbilt University Medical Center
raf Van de Plas, Delft Center for Systems and Control, Delft University of Technology, Department of Biochemistry,
Vanderbilt University, Mass Spectrometry Research Center, Vanderbilt University
Richard M Caprioli, Department of Biochemistry, Vanderbilt University, Mass Spectrometry Research Center, Vanderbilt
University, Department of Chemistry, Vanderbilt University
Jeffrey M Spraggins, Mass Spectrometry Research Center, Vanderbilt University, Department of Chemistry, Vanderbilt
University, Department of Cell and Developmental Biology, Vanderbilt University

Intregrated dataset consisting of MALDI IMS, multiplexed immunofluorescence, and histology.

108
MALDI IMAGING MASS SPECTROMETRY EVALUATION OF GLYCANS AND
EXTRACELLULAR MATRIX PROTEINS AS BIOMARKERS OF RENAL CELL
CARCINOMA IMMUNOTHERAPY RESPONSE
Abstract ID: 546

Presenting author: Colin Mcdowell, Medical University of South Carolina, Department of Cell and Molecular
Pharmacology and Experimental Therapeutics

Introduction

The survival rates for kidney cancers, namely clear cell renal carcinoma (ccRCC), are still quite poor. Immunotherapeutic
checkpoint inhibitors have enhanced survival for a subset of ccRCC patients, yet unfortunately, many patients develop
therapeutic resistance or are non-responsive. Checkpoint inhibitors and their targets interact in the extracellular domain,
thus there is great clinical interest in understanding cell-surface or extracellular matrix (ECM) components which may
modulate this interaction or typify therapeutic response. Identified glycan and ECM signatures between different ccRCC
responders may facilitate understanding of resistance mechanisms, allowing prognostic stratification of patients for
precision cancer therapies. To this end, glycan and ECM-targeted imaging mass spectrometry (IMS) approaches were
applied to a pathology-annotated ccRCC tissue cohort from individuals with differential responses to PD-1/PD-L1-
targeted immunotherapy.

Methods

A cohort of ccRCC tissues were selected for treatment response, where 9 ccRCC tissues were from positive responders
(PR) to PD-L1 immunotherapy with > 18 months progression-free survival (PFS), compared to 12 tissues from
individuals who had no or very poor response (NR) with PFS < 6 months. Several normal kidney tissues were also
processed as reference samples. The formalin-fixed tissues were prepared for N-glycan and collagen/ECM IMS using
established protocols then imaged on a timsTOF fleX MALDI-QTOF MS (Bruker). Data was visualized using SCiLS Lab
software (Bruker), combined with existing databases for N-glycans and collagen/ECM peptides.

Preliminary data (results)

IMS was applied to 21 ccRCC tissues with a goal of linking glycan and extracellular matrix molecular tumor
microenvironment signatures with pathological annotations and associated clinical data. The tissues analyzed were
obtained from initial nephrectomy surgery and were treatment naïve, thus any N-glycans or ECM biomarkers identified
would be prognostic for treatment response. Tissue slides were first processed for N-glycan tissue imaging, resulting in
the detection of 120+ N-glycans per tissue. All spectra were collated into a single SCiLS file for analysis and co-
localization to specific histopathological locations. Reference tissues of normal kidneys were similarly processed, where
branched and sialylated N-glycans characterized glomeruli and branched, bisecting and multi-fucosylated species
distinguished tubules. In general, for most ccRCC regions, multiple branched, fucosylated and sialylated N-glycans were
detected which were not specific to NR or PR classifications. Of note, unique putative phosphorylated high mannose N-
glycans were detected at increased levels in the tumors of NR tissues. A phosphorylated high mannose glycan typifies
glycoprotein hydrolases normally targeted to the lysosome. If secreted, these enzymes will degrade ECM components
and facilitate metastatic spread. In PR tissues, 20 N-glycans associated with immune function were detected at
significantly elevated levels and were primarily localized to stroma ECM regions. Collagenase digestion for detection of
collagens and other ECM proteins is ongoing, with initial data demonstrating differential peptides in NR and PR tissues.
Multiplexed data analysis will be done for both glycan and ECM data, to be further linked with additional clinical data and
histopathology correlates.

Please explain why your abstract is innovative for mass spectrometry?

Imaging mass spectrometry analyses identified novel N-glycans and extracellular matrix proteins as potential predictors
of immune-oncological therapeutic response in ccRCC tissue, which could be applicable to other types of cancer.

Co-authors:

Harrison Taylor, Medical University of South Carolina, Department of Cell and Molecular Pharmacology and
Experimental Therapeutics
Grace Grimsley, Medical University of South Carolina, Department of Cell and Molecular Pharmacology and
Experimental Therapeutics
Philipp Ivanyi, Hannover Medical School, Department of Hematology, Hemostaseology, Oncology and Stem Cell
Transplantation
Viktor Grünwald, Essen University Hospital, Medical Oncology and Urology

109
Jessica Schmitz, Hannover Medical School, Nephropathology
Jan Hinrich Bräsen, Hannover Medical School, Nephropathology
Peggi M. Angel, Medical University of South Carolina, Department of Cell and Molecular Pharmacology and
Experimental Therapeutics
Richard R. Drake, Medical University of South Carolina, Department of Cell and Molecular Pharmacology and
Experimental Therapeutics

110
UNDERSTANDING THE ABILITY OF HOP TERPENE
BIOTRANSFORMATION TO ENHANCE BEER FLAVOUR USING PROTON
TRANSFER REACTION-TIME OF FLIGHT-MASS SPECTROMETRY (PTR-
TOF-MS)
Abstract ID: 415

Presenting author: Rebecca Roberts, University of Otago, Fondazione Edmund Mach (FEM)

Introduction

To meet consumer demand for hop-flavour driven beers, there is an increasing interest to control, optimise and predict
hop flavour generation during fermentation. Few aroma compounds present in hops directly contribute to beer flavour
due to changes during fermentation. Biotransformation can be mainly attributed to the contribution of enzymes in yeast.
However, current knowledge of the effect of yeast on hop volatile compounds during fermentation is limited. This makes
it impossible to accurately predict how hop additions will impact the aroma of finished beer. The overall aim of the
research was to gain a mechanistic understanding of the biotransformation reactions responsible for hop flavour
development in beer. PTR-ToF-MS allows for real-time monitoring of aroma generation, enabling changes of hop aroma
during fermentation to be detected.

Methods

To understand hop aroma generation and the role of yeast in biotransformation, six pure terpene compounds (linalool,
geraniol, nerol, β-pinene, citornellyl acetate and citronellol) and four Sacchromyces cervisiae strains (US-05, WB-06, W-
34/70 and S-23) were investigate. Standard brewing conditions were reproduced using a model beer system. Individual
terpene compounds (spiking concentration; 10 ppm), yeast (pitching rate: 1x106 cells/mL) and model wort were added to
20 mL glass head-space vials, sealed and incubated at 20oC for 5 days. Throughout fermentation, all samples were
measured automatically every 6 hours using an auto-sampler connected to a PTR-ToF-MS.

Preliminary data (results)

Preliminary experiments were conducted by measuring samples using SPME GC-MS after incubation at 20°C for 5 days.
When 10ppm of geraniol was added to model wort containing Sacchromyces cervisiae US-05, 1.6 ppm of citronellol was
produced. The formation of citronellol from geraniol aligns with literature, citronellol was the main product of geraniol
transformation in studies conducted by King and Dickinson (2003), Slaghenaufi et al (2020) and Takoi et al (2010). King
and Dickenson (2000) proposed that geraniol is primarily transformed to citronellol and can be further transformed to
linalool. This study tried to confirm these results however, no increase in linalool was detected, instead dihydrolinalool
was identified. Colomer et al. (2020) discussed the need to evaluate the bioflavouring of beer and whether compounds
such as nerol contribute to bioconversion. To evaluate nerol’s contribution to bioflavouring beer, 10 ppm of a pure nerol
standard was added to a model beer system. It was observed that ß-pinene (0.01 ppm), dihydrolinalool (0.012 ppm),
dihydrocitronellyl acetate (0.015 ppm), citronellyl acetate (0.07 ppm), α-terpineol (0.067 ppm), nerol acetate (0.135 ppm)
and citronellol (0.654 ppm) were produced. The production of citronellol and α-terpineol from nerol by yeast aligns with
literature however the formation of the other compounds identified in this study is novel. These compounds could impact
beer aroma, the generation of these compounds is complex and tracking the formation over the 5-day fermentation using
PTR-ToF-Ms is essential for improving the understanding of biotransformation’s contribution to beer aroma.

Please explain why your abstract is innovative for mass spectrometry?

Using PTR-ToF-MS to dynamically monitor compounds during fermentation will enable the discovery of transformation
pathways and allow identification of flavour compound generation, that influence overall beer aroma.

Co-authors:

Graham Eyres, University of Otago

Phil Bremer, University of Otago
Pat Silcock, University of Otago
Franco Basioli, Fondazione Edmund Mach (FEM)

111
Biotransformation of geraniol by ale yeast S. cerevisiae US-05.

112
THE CHARACTERISATION OF CHARGE LOCATION RESOLVED
PRECURSOR IONS USING ION MOBILITY TANDEM MASS
SPECTROMETRY
Abstract ID: 540

Presenting author: Anirudh Sharma, Teesside University

Introduction

Ion mobility has been proven to separate and characterise ions that form from molecules which have their charge located
at different positions. These structures have, in many cases, been shown to exhibit specific fragmentation patterns that
can help establish the site of the charge, and therefore enable the fragmentation mechanisms to be determined in much
greater detail. This has drawn particular interest for pharmaceutical compounds as it provides the means to extract
greater depth of information about the molecular structure of complex molecules containing more than one protomeric
structure. This can be related to the physiological action of the pharmaceutical compounds and can inform their mode of
action. It can provide conformationally isolated MS/MS spectra, allowing the characterisation of the separated ions.

Methods

Protomers of benzocaine (ethyl 4-aminobenzoate) were formed from electrospray ionisation;radical cations of
benzocaine were formed from Atmospheric pressure Solids Analysis Probe (ASAP) ionisation. Cyclic ion mobility-
enabled quadrupole time-of-flight (Q-cIM-oaToF) mass spectrometer (SELECT SERIES Cyclic IMS, Waters, UK) was
used for these studies. Using advanced features of this mass spectrometer, the two protomer structures, or the two
radimers structures were separated by IMS. The desired ion, in each case, was then shape-selectively isolated from all
other ions using the trapping features of the cyclic IMS device, and then dissociated prior to ToF MS measurement of the
product ion spectra.

Preliminary data (results)

It has been shown previously that two protomers of benzocaine can be resolved using IMS, and distinct collision-induced
dissociation spectra obtained for each protomer. In these studies, this research has been extended, by dissociating each
protomer with collision energies ranging from 2-26 eV. This has enabled fragmentation maps to be produced for all the
product ions, which clearly show the variety of fragmentation mechanisms in action. DFT calculations along with
experimental observations for benzocaine suggest that an ion with the proton at the amine nitrogen atom has a larger
CCS than the ion with the proton situated on the carbonyl oxygen atom. This has enabled two distinct fragmentation
mechanisms to be postulated. Comparative work with deuterated benzocaine (D5-ethyl 4-aminobenzoate) has been
undertaken to underpin the mechanistic postulations.

In a separate, but related study, the radical cations for benzocaine have been shown to be capable of being resolved
utilising ion mobility experiments. These species have been termed radimers. These separated radimers have been
individually isolated using selection and ejection experiments and fragmented utilising different collision energies.
Preliminary results suggest the radical cation resides on one atom for one or the resolved radimers, with the second
radimer existing as a distonic ion.

As a means to expand on these developments, larger molecules containing more functional groups are under study.

Please explain why your abstract is innovative for mass spectrometry?

Fragmentation of ion mobility resolved protomers, or radimers of benzocaine using novel functionalities of the cyclic IMS-
MSMS.

Co-authors:

Jim Scrivens,
Michael T Bowers, University of California Santa Barbara
Kalju Kahn, University of California Santa Barbara
Kevin Giles, Waters Corp
Martin Palmer, Waters Corp
Jakub Ujma, Waters Corp
Edward Calyton, Consultant
Jacqueline Mosely, Teesside University

113
Geometry optimised structure of benzocaine protonated at the amine group

Geometry optimised structure of benzocaine protonated at the carbonyl group

114
CAPILLARY ELECTROPHORESIS COUPLED TO TIMS-TOF MASS
SPECTROMETRY USING THE NANOCEASY INTERFACE
Abstract ID: 606

Presenting author: Jasmin Schairer, Aalen University

Introduction

Though ion mobility MS has been developing to a well-known technology, it is rarely coupled to capillary electrophoresis
(CE) so far. Both, CE and IM, separate ions according to their electrophoretic mobility, i.e. their charge-to-size ratio,
however CE separating solvated ions whereas IM separates ions in the gas phase. Thus, CE and IM have different
selectivity as shown for the analysis of glycans (https://doi.org/10.1007/s00216-018-1515-7). Furthermore, in CE,
separation efficiency and peak capacity is typically much higher than in IM making the coupling of both separation
techniques attractive.

In recent years innovative coupling technology of CE to MS has been developed. We recently introduced
the nanoCEasy interface, an easy-to-use, robust and flexible interface with nanoESI sensitivity
(https://doi.org/10.1021/acs.analchem.1c03213).

Methods

An Agilent CE was coupled via the nanoCEasy interface to a timsTOF Pro 2. An acidic background electrolyte (BGE)
was applied for the analysis of tryptic Hela-digest (commercial sample) and subunit analysis of proteins (IdeS digest of
mAb, fusion protein). A basic BGE was used to separate anionic metabolites from cell culture samples. Interface
parameter were set according to Schlecht et al. (https://doi.org/10.1021/acs.analchem.1c03213). The timsTOF Pro 2 was
operated in ESI positive (proteins and peptides) or negative mode (metabolites). Calibration of the tims was performed
by infusion of tuning mix directly through the nanoCEasy interface by a fused-silica capillary.

Preliminary data (results)

Setting up the CE-nanoCEasy-timsTOF was straightforward, leading to sensitive and robust measurements. In order to
have identical pressure settings the interface has also been used for mobility calibration of the tims. The valving option
was used for recoating of the CE capillary in between run. Set-up of the interface will be presented and discussed in
detail.

Different applications have been performed demonstrating the benefit of the CE-timsTOF MS set-up:

The analysis of subunits revealed the ability of both the CE and the tims to distinguish between various disulfide bridges.
Thus, the combination of CE and on-line tims TOF characterization allows structural information of proteins.

Peptide analysis of tryptic HeLa digest has been performed by the CE-timsTOF set-up using the PASEF approach. Short
CE capillaries can be used for this application with fast and efficient separation. Due to the different selectivity of CE and
LC, CE-tims MS/MS can be a complementary tool in proteomics.

Anionic metabolites from cell culture samples were analyzed using a basic BGE. For example, various sugar phosphates
could be separated in the CE. Based on migration times of standards and Collisional Cross Sections (CCS) the various
isobaric sugar derivatives could be unequivocally identified.

In summary, the CE-nanoCEasy-tims TOF is a powerful technique with various applications combining selectivity and
high separation efficiency of CE with sensitive ESI timsTOF characterization.

Please explain why your abstract is innovative for mass spectrometry?

· CE-IM-MS using the nanoCEasy interface

· First on-line CE-timsTOF MS data.

· Various applications towards peptides, proteins and metabolites

Co-authors:

115
Lukas Naumann, Aalen University
Stuart Pengelley, Bruker Daltonics GmbH & Co.KG
Eckhard Belau, Bruker Daltonics GmbH & Co.KG
Christian Albers, Bruker Daltonics GmbH & Co.KG
Christian Neusüß, Aalen University

Schematic view of CE-timsTOF coupling using the nanoCEasy interface

116
YIELD IMPROVEMENT IN SECONDARY ION MASS SPECTROMETRY
USING CHEMICALLY REACTIVE GAS CLUSTER ION BEAMS
Abstract ID: 469

Presenting author: Matija Lagator, The University of Manchester

Introduction

Amongst the greatest challenges in SIMS research are low ionization yields and the semi-quantitative nature of the
analysis caused by matrix effects. Through the development of polyatomic water beams for SIMS, it was previously
demonstrated that primary ion beam chemistry is an important parameter in increasing yields while also reducing the
matrix effects. Water gas cluster ion beams (GCIBs), typically use an inert gas (Ar or N 2) as a carrier. We report a recent
study (Lagator et al., 2022) investigating the potential of different carrier gases to alter the characteristics of water
clusters and hence increase yields. The study aimed to test the hypothesis that carbon dioxide would react with water
clusters thus making it more likely that the clusters would act as proton donors.

Methods

ToF-SIMS data was collected using a J105 3D Chemical Imager (Ionoptika Ltd, UK). A 70 keV GCIB source was used to
generate water clusters that contained approximately 26000 molecules (E/m ≈ 0.15 eV/amu). Depth profiles of drug
standards were collected over an area of 500×500 μm 2, with a total ion dose of 1×1013 ions cm-2. The spatial resolution
of the beam was ≈ 10 μm. Two different carrier gasses were applied to the water clusters: 1) pure argon; 2) mixture of
carbon dioxide, argon, and oxygen in ratio of 86:12:2.

Preliminary data (results)

By applying water clusters with a reactive carrier gas to drug standards (ranitidine) we were able to measure the effect
that the presence of carbon dioxide has on the ion yield. We show that compared to an inert carrier gas (pure argon), the
yield observed with reactive water clusters is between 20 and 50 times higher (Figure 1) for ranitidine ions (molecular ion
and fragments). This implies a change in cluster composition as a result of using a more reactive carrier gas, which is a
novel approach to altering the SIMS primary ions. With at least an order of magnitude of increase in yield, this approach
indicates a potential route to further improvements in SIMS methodologies and increases in resulting sensitivity.

Matrix affects are one of the biggest obstacles for further development of the SIMS field. They arise from the
unpredictable interaction of adjacent species which results in an ionisation suppression/enhancement effect. Analysis in
more complex environments present a greater obstacle due to a myriad of potential chemical interactions. We show that
by applying the newly developed chemically reactive water cluster primary ion beam, it is possible to reduce the matrix
effects. Additionally, by using these clusters we achieved a high level of quantification over a range of acetaminophen
concentrations doped onto biological tissue (Fig. 2). The high R 2 value and the overall good match of the linear line of
best fit indicates the potential for SIMS to be applied in a more quantitative manner, even for biological samples.

Please explain why your abstract is innovative for mass spectrometry?

Development of novel water-based and reactive primary ion beams has been shown to significantly increase SIMS
yields, while also having a potential of greatly improving analysis of biological materials.

Co-authors:

Irma Berrueta Razo, The University of Manchester

Thomas Royle, The University of Manchester
Nicholas Lockyer, The University of Manchester

117
Ratio of Ar/CO2 and pure Ar yields for ranitidine.

Calibration curve for acetaminophen molecular ion on mouse brain tissue.

118
Session: IM-5 Ion Spectroscopy, Physical and Chemical principles
underlying MS - Session B

LIGHTFOOTPRINTING PROTEINS
Abstract ID: 691

Presenting author: Perdita Barran, The University of Manchester

Introduction

Understanding the mechanism of photoreceptor proteins, ubiquitous in nature, is a growing field both to understand
fundamentals of the processes and to exploit this behaviour for biotechnology applications. Here we present a mass
spectrometry-based method that determines photoresponsive mechanisms in terms of the change in stoichiometry and
conformation on light activation. Previous work has described the use of native mass spectrometry to study
photoreceptor proteins widely located throughout the biological environment. Investigations have examined proteins from
plants involved with phototropism1 and opsin proteins which possess the retinal chromophore involved in vision. 2 The
prevalence of photoreceptors in the natural world, as well as the promiscuous biological behaviour that allows them to
perform multiple processes, means that they provide a rich and underexplored subject for investigation.

Methods

This method stems from the growing uptake of native mass spectrometry techniques to biological and biophysical
researchers. The ability to investigate large multimeric photoreceptor proteins with widely available mass spectrometry
apparatus and utilise LEDs rather than more expensive laser systems permits a facile implementation of this method with
minimal alignment. We use modified Ion Mobility Mass Spectrometers and high resolution Orbitraps that permit
phtoirradition in source and in instrument. To demonstrate the method, we focus on naturally occurring photoreceptor
proteins from plant and bacterial sources that absorb wavelengths across the UV, visible, and near-IR spectrum (200-
800nm).

Preliminary data (results)

We utilise standard, native ion mobility mass spectrometry techniques on three commercially available platforms, and we
show the ease of installation of LEDs in each source, permitting light activation experiments on photoreceptors across
the visible region. This method is demonstrated on a series of photoreceptor proteins activated by light from across the
near UV and visible regions of the spectrum. We initially implemented the method with investigations into the
photoreceptors TtCarH and full-length UVR8 alongside a truncated form of UVR8. CarH is an AdoCbl-dependent protein,
which uniquely utilises the coenzyme B12 cofactor, with an upper axial 5’-ado group, for light-sensing. Initial experiments
on the TtCarH photoreceptor complement UV-visible transient absorption data and aided the determination of the
photochemical mechanism. In the irreversible illumination of TtCarH tetramer the 5’-deoxyadenosyl group becomes
displaced by H132, which results in the TtCarH tetramer dissociating into monomers, each with a cob(III)alamin
covalently bound. These photoconverted monomers are thus of a different mass to both the apomonomer (i.e., with no
B12 cofactor bound) and the 5’-deoxyadenosylcoablamin bound monomer units that make up the tetramer. They are
therefore readily identifiable via mass spectrometry measurements. Further work focuses on the assembly of the
photoactive tetrameric species upon binding of AdoCbl in the dark. We show that, when AdoCbl is scarce, low
abundance tetrameric species form with sub-stoichiometric AdoCbl. The conformation of these species is similar to the
TtCarH tetramers with all four AdoCbl-binding sites occupied, suggesting each monomer does not need a bound-
chromophore.

Please explain why your abstract is innovative for mass spectrometry?

Photoactivation coupled with native mass spectrometry.

Co-authors:

Rachelle Black, The University of Manchester

Lennart Ramakers, The University of Manchester
Ines Camacho, National Physical Laboratory
Derren Heyes, The University of Manchester
Alex Jones, National Physical Laboratory

119
Experimental schematic of light footprinting to investigate photoreceptor proteins.

120
CHANGING THE TEMPERATURE DURING RESONANT EXCITATION IN
COMMERCIAL QUADRUPOLE ION TRAPS
Abstract ID: 733

Presenting author: Thomas Neugebauer, Université de Bretagne Occidentale

Introduction

Even decades after quadrupole ion traps have been made commercially available, they still have anuncontested role in
the structure elucidation due to their capability of performing Tandem MSnexperiments. The collision-induced resonant
excitation process itself in quadrupole ion traps is usuallythought of as a slow-heating process, as the internal energy of
ions increases only in small steps dueto the collisions with the light buffer gas helium. Therefore, the fragmentation
parameters likeexcitation voltage or excitation time normally play only a minor role in the appearance of fragment
ionspectra, if chosen sufficiently high. Consequently, they are usually not even mentioned in publicationsthat focus on
the chemical information represented by the mass spectra.

Methods

Ion trajectory calculations of the resonant excitation process have been performed by numerically solving the equation of
motion, explicitly including nonlinear field components of the quadrupole field. The problem of suitable starting conditions
has been taken care of by applying the phase-space theory of quadrupole fields. The evolution of the mean-square
velocity of the trajectories has been investigated compared to systematic collision-induced dissociation experiments in
which for different excitation frequencies the excitation voltage has been increased until 95% precursor dissociation was
achieved. The resulting harshness of the process was identified by the a4/b4 fragment ion ratio of Leucine-Enkephaline.

Preliminary data (results)

The effect that can be exploited to achieve these different excitation conditions is thedependence of the oscillation
frequency of ions on their oscillation amplitude which is caused by thenon-ideal electrode geometries that introduce
higher order potential impurities. Due to this, theexcitation frequency cannot be chosen to simply always match the ions’
oscillation frequencies.Hence, neither do the ions simply increase their amplitude until they are fast enough to endure
harshenough collisions to dissociate, nor do they reach a steady-state oscillation amplitude at whichexcitation and
damping cancel.

Results show a most efficient but also a harshest excitation frequency that are not the same. The most efficient excitation
frequency is the frequency that requires the lowest voltage to reach a certain degree of precursor dissociation and can
be found just slightly above the theoretical secular frequency in the pure field. The harshest excitation frequency occurs
at a slightly higher excitation frequency, because when the oscillation frequency of the ions starts to increase, due to
excitation, it can eventually become faster than the excitation frequency. At the moment this change in frequency order
happens, the phase shift between the two frequencies builds in the opposite direction which allows ions to stay longer in
on-resonance and to reach even higher oscillation amplitudes and thus higher velocities, resulting in the possibility of a
harsher fragmentation.

Please explain why your abstract is innovative for mass spectrometry?

We show with experimental evidence the direct connection between trajectory calculations and the resulting
fragmentation harshness during the collision-induced resonant excitation process in quadrupole ion traps.

Co-authors:

Antony Memboeuf, Université de Bretagne Occidentale

121
A slightly different excitation frequency can change everything.

Fragmentation harshness against excitation frequency.

122
NATIVE MS FOR INTERACTIONS AND GAS-PHASE CHEMISTRY OF
METALLOPEPTIDES AND METALLOPROTEINS
Abstract ID: 345

Presenting author: Sarah Brandner, TU Darmstadt

Introduction

It has been estimated that around 30% of proteins require metal cofactors to perform their biological function.
Conversely, disruption of metal homeostasis is associated with a range of diseases. Clearly then, there is a need to
study protein-metal interactions, and mass spectrometry has significant benefits this context, particularly when working
under native-like conditions. For example, stoichiometry of a protein-metal complex, oxidation state of a transition metal,
and covalent modifications to a protein (for example through metal redox chemistry) can all be rapidly and
unambiguously determined. Here, we report native MS results of a range of model peptide- and protein-metal
complexes, with particular focus on their gas-phase decomposition.

Methods

Measurements were carried out on an LTQ Orbitrap XL (Thermo) and a Synapt XS ion mobility – mass spectrometer
(Waters). Native nano-electrospray ionisation was performed with a home-built as well as a commercial source. Both
sources used glass emitters that were produced in-house, and solutions were prepared in aqueous solution. The various
peptide-metal samples were characterized by native top-down mass spectrometry (nTDMS). The precursor ions with the
metal complexation were isolated and excited by collision induced dissociation (CID).

Preliminary data (results)

We systematically investigated complex formation of a range of biologically relevant transition metals with a set of model
peptides and proteins across a broad mass range. We measured the relative stabilities of these complexes with native
MS. We used CID fragmentation to determine metal binding sites. Through analysis of the observed isotope distributions
in high-resolution MS, we determined the gas-phase metal oxidation states and compared these to those in solution. In
particular, we observed one-electron reduction of metal cations during decomposition of noncovalent complexes, which
can result in radical reactions such as side chain loss. We monitored this radical chemistry for samples containing
several physiologically relevant metals, including iron. Besides the common b and y ions for CID fragmentation, we also
observed fragments from radical-directed pathways in the MS2 spectra, including abundant a ions. Notably, these
fragments occur in the absence of exogenous electrons; however, their formation implies a radical interaction in the gas-
phase. We linked this radical behaviour to intrinsic properties of the metal centre and metal-interacting functional groups.

Please explain why your abstract is innovative for mass spectrometry?

In-depth description of the formation, interactions, and gas-phase redox chemistry of a range of peptide- and protein-
metal complexes.

Co-authors:

Tanja Habeck, TU Darmstadt

Frederik Lermyte, TU Darmstadt

123
CRYOGENIC INFRARED ION SPECTROSCOPY OF ISOMERIC LIPIDS
Abstract ID: 473

Presenting author: Carla Kirschbaum, Freie Universität Berlin, Fritz-Haber-Institut der Max-Planck-Gesellschaft

Introduction

Lipids constitute a heterogeneous class of biomolecules that are involved in energy storage, signaling and cell
compartmentalization. The multitude of biological functions is reflected in a tremendous structural diversity and
coexistence of multiple isomers, which fulfill distinct tasks in the metabolism. Mass spectrometry (MS) is the
uncontentested workhorse in lipidomics; however, several isomers cannot be resolved by MS and classical tandem MS
approaches (Fig. 1a). Therefore, the hyphenation of MS with infrared (IR) ion spectroscopy is investigated as a technique
to enhance the resolution of isomers and provide fundamental understanding of lipid fragmentation mechanisms.
Hyphenating MS and IR spectroscopy combines the high sensitivity and the possibility for mass-to-charge selection of
MS with the structure sensitivity of IR spectroscopy to identify isomers and fragment structures.

Methods

High-resolution IR spectra of lipids were obtained by cryogenic gas-phase IR spectroscopy in helium droplets employing
a free-electron laser. The lipids are ionized by nano-electrospray ionization, mass-to-charge selected in a quadrupole
and stored in a hexapole ion trap. Superfluid helium droplets traverse the trap and pick up ions. If an ion inside a droplet
absorbs an IR photon, the vibrational energy is dissipated by evaporation of the helium shell until the ion is released after
the absorption of multiple photons. The signal of released ions is detected on a time-of-flight detector and plotted against
the tunable wavenumber.

Preliminary data (results)

IR ion spectroscopy can distinguish subtle structural differences between isomers throughout various lipid classes. For
instance, stereoisomeric glycan headgroups in glycolipids, which are indistinguishable by (tandem) MS, yield diagnostic
IR fingerprints.[1] The detailedness of structural information is comparable to nuclear magnetic resonance spectroscopy.
However, biological applications, which are restricted by little sample material and complex matrices, are more readily
accessible by IR spectroscopy due to its high sensitivity and mass-to-charge selection of ions against the matrix
background (Fig. 1b).

Double bond isomers in lipid chains can be rendered visible by introducing IR-active double bond sensors, which are
designed to interact with double bonds and yield diagnostic frequency shifts depending on the double bond position and
geometry.[2] Furthermore, the sn-isomer specificity observed in tandem MS of glycerolipids was explained by
spectroscopic confirmation of the key fragment structure, which emerges by cyclization of the sn-2 fatty acyl into a
dioxolane ring.[3] Overall, IR ion spectroscopy is a powerful technique to complement MS-based workflows in lipidomics
by providing insight into biological isomer distributions and knowledge on lipid fragmentation.

[1] C. Kirschbaum, et al., Nat. Commun. 2021, 12, 1201.

[2] a) C. Kirschbaum, et al., Angew. Chem. Int. Ed. 2020, 59, 13638-13642; b) C. Kirschbaum, et al., Anal. Bioanal.
Chem. 2021, 413, 3643-3653.

[3] C. Kirschbaum, et al., J. Am. Chem. Soc. 2021, 143, 14827-14834.

Please explain why your abstract is innovative for mass spectrometry?

Coupling IR ion spectroscopy with MS revealed a high potential to tackle critical analytical challenges in lipidomics and
gain fundamental understanding of yet understudied lipid fragmentation mechanisms in tandem MS.

Co-authors:

Kim Greis, Freie Universität Berlin, Fritz-Haber-Institut der Max-Planck-Gesellschaft

Gerard Meijer, Fritz-Haber-Institut der Max-Planck-Gesellschaft
Gert von Helden, Fritz-Haber-Institut der Max-Planck-Gesellschaft
Kevin Pagel, Freie Universität Berlin, Fritz-Haber-Institut der Max-Planck-Gesellschaft

124
a) Challenging lipid isomers. b) Infrared spectra of stereoisomeric glycosphingolipids.

125
CONFORMATION-SELECTIVE INFRARED ION SPECTROSCOPY ON A
TIMS ENABLED FT-ICR MS PLATFORM
Abstract ID: 523

Presenting author: Kas Houthuijs, Radboud University, FELIX laboratory

Introduction

Infrared ion spectroscopy (IRIS) has become a powerful technique for structural characterization of gas-phase ions. It
has evolved from fundamental gas-phase ion chemistry to molecular identification in analytical chemistry.Yet, on its own,
IRIS is unable to measure isomer-selective IR spectra from a mixture. Trapped ion mobility spectrometry (TIMS) can be
employed to separate isomers based on their collisional cross-section, thereby extending the selectivity of MS and IRIS.
This combination was realized on a new instrument that includes ion mobility, ultra-high mass resolution and IRIS: a
prototype TIMS-enabled FT-ICR MS coupled to the beamline of the free-electron laser FELIX. First results from this
setup include characterization of polyalanine conformations and analysis of di- and tri-saccharide isomers.

Methods

A Bruker solariX XR 7T FT-ICR was coupled to the FELIX beamline, providing optical laser-access to the ions in the ICR
cell via an IR-transparent window. The instrument features an ESI/MALDI dual source. Additionally, the instrument was
equipped with a prototype version of trapped ion mobility spectrometry (TIMS) by exchanging the ion funnel in the source
vacuum housing with a modified funnel. IR spectra of mobility-selected ions can be recorded by operating this funnel as
a filter. Poly-alanine samples were generated via both ESI and MALDI. An LC-system is employed to introduce the
saccharides found in a body-fluid sample.

Preliminary data (results)

To perform spectroscopy with pulsed lasers such as FELIX, the MS sequence of the solariX was synchronized to an
external trigger through modifications to the standard pulse program. Additionally, wavelength scanning is implemented
via a handshake protocol thereby enabling automated recording of IR spectra. These software changes have been
implemented for all modes of operation, allowing for the measurement of IR spectra of ions produced by both ESI and
MALDI, as well as mobility selected ions by TIMS.

To demonstrate the performance of the setup, the gas-phase conformations of poly-alanines (Alan) were studied. Ion
mobility data indicates these polymers transition from a globular to helical conformation as they increase in size. We
elucidate the structures of these conformations by combining IRIS with DFT and CCS calculations and investigate the
transition-point dependence on adduct type. Additionally, we employ both ESI and MALDI to assess the influence of the
ionization mechanism on the gas-phase conformations.

To assess the analytical capabilities of the combined TIMS-IRIS approach we focus on the analysis of a set of isomeric
di- and tri-saccharides. Saccharides have a wide variety of isomers, which are often poorly differentiable with MS/MS or
LC. Ion mobility, however, can separate these isomers based on their shape that originates from varying stereochemistry
or branching of the monosaccharides. By hyphenating TIMS with IRIS, the separated isomers can be identified based on
their IR spectrum. We will utilize a set of poly-saccharide standards, as well as a body fluid sample introduced via LC.

Please explain why your abstract is innovative for mass spectrometry?

A unique prototype TIMS enabled FT-ICR MS with optical access enables IR ion spectroscopy on mobility-selected ions
for the characterization of poly-alanines and the identification of di- and tri-saccharides.

Co-authors:

Christopher Wootton, Bruker Daltonics GmbH & Co. KG

Pieter Kooijman, Radboud University, FELIX laboratory
Lara van Tetering, Radboud University, FELIX laboratory
Teun van Wieringen, Radboud University, FELIX laboratory
Christoph Gebhardt, Bruker Daltonics GmbH & Co. KG
Giel Berden, Radboud University, FELIX laboratory
Mark Ridgeway, Bruker Daltonics Inc.
Jonathan Martens, Radboud University, FELIX laboratory
Jos Oomens, Radboud University, FELIX laboratory

126
Schematic of the prototype TIMS-enabled FT-ICR MS coupled to FELIX.

127
THE GAS-PHASE HOST-GUEST CHEMISTRY OF
[N]CYCLOPARAPHENYLENES
Abstract ID: 86

Presenting author: Markus Freiberger, Physical Chemistry I, Friedrich-Alexander-University Erlangen-Nürnberg

Introduction

[n]Cycloparaphenylenes (CPPs) are strained macrocycles, comprising only sp2-hybridized carbon atoms. In recent years,
CPPs have become a promising candidate for fundamental studies in supramolecular chemistry. Their concave inside
and convex outside allow for the formation of host-guest complexes, in which CPPs serve as both guest and host,
thereby forming the shortest double-walled armchair carbon nanotubes.

Methods

In this work, we study host-guest complexes of [n]CPPs (n = 5-12) using electrospray and laser desorption ionization
mass spectrometry.

Preliminary data (results)

Complexes of two CPPs with a size difference of three to seven phenyl units are observed. Energy resolved tandem
mass spectrometry (MS2) experiments show that the charge is mainly located at the inner ring. Additionally, the
MS2 experiments reveal that complexes with a phenyl unit difference of five and six exhibit the same interaction strength
and that they are the most stable amongst the investigated host-guest complexes. However, size selectivity experiments
indicate that the formation of complexes with a size difference of five phenyl units is kinetically always preferred.

Please explain why your abstract is innovative for mass spectrometry?

Collectively, this study shows that mass spectrometry is an outstandingly precise tool for the study of non-covalently
bound host-guest complexes.

Co-authors:

Martin Minameyer, Physical Chemistry I, Friedrich-Alexander-University Erlangen-Nürnberg

Thomas Drewello, Physical Chemistry I, Friedrich-Alexander-University Erlangen-Nürnberg

128
Session: IM-6 Data sciences in
MS/AI/Chemometrics/Identification/Modelling - Session B

DETECTION STRATEGIES FOR CONJUGATE METABOLITES WITH

TANDEM MASS SPECTROMETRY DATA IN HUMAN BIOMONITORING AND
WASTEWATER-BASED EPIDEMOLOGY
Abstract ID: 445

Presenting author: Carolin Huber, Helmholtz Centre for Environmental Research - UFZ, Institute of Ecology,
Diversity and Evolution, Goethe University Frankfurt

Introduction

Library annotation of liquid chromatography-high resolution tandem mass spectrometry (LC-HRMS/MS) data is a reliable
approach for screening organic contaminants. For exposomics, also metabolites are analytes, but often spectra are
unavailable. Glucuronide conjugates are important phase II metabolites that can be used as biomarkers of exposure and
hold potential in wastewater-based epidemiology to identify compounds that passed the human compared to parent or
phase I metabolites, which might be released also without human uptake.

Methods

We present and evaluate a strategy including selective analysis of glucuronides by LC-HRMS/MS-based on all-ion
fragmentation and data-dependent MS/MS triggered by detection of a neutral loss. The spectra were truncated, which we
refer to as in-silico deconjugation, and finally searched against mass spectral libraries of the aglycones.

Preliminary data (results)

An optimisation of HRMS/MS detection method on model glucuronides showed that a combination of three different
collision energies allowed for a selective detection of neutral losses and MS/MS fragments from the aglycone part of the
molecules (i.e., diagnostic ions for annotation) for a wide variety of glucuronides in positive and negative ion mode. The
data processing workflow was tested on a dataset of in vitro–generated glucuronides of reference standard mixtures and
a dataset of 51 authentic urine samples collected from patients with known medication status, acquired on different
instrumentations. A total number of 75 different glucuronidated molecular structures were identified by in silico
deconjugation and spectral library annotation. We also identified specific molecular structures (sulfonamides, ether
bonds, di-glucuronides), which resulted in slightly different fragmentation patterns between the glucuronide and the
unconjugated compound. This led to a decreased spectral matching score and in some cases to a false-negative
identification.

Please explain why your abstract is innovative for mass spectrometry?

This method allows for selective detection of glucuronides, leading to no need for deconjugation steps or for reference
glucuronide spectra for screening approaches.

Co-authors:

Werner Brack, Helmholtz Centre for Environmental Research - UFZ, Institute of Ecology, Diversity and Evolution, Goethe
University Frankfurt
Tobias Schulze, Helmholtz Centre for Environmental Research - UFZ
Herbert Oberacher, Medical University of Innbruck
Martin Krauss, Helmholtz Centre for Environmental Research - UFZ

129
The LC-HRMS/MS approach for the selective detection of glucuronides.

In-silico deconjugation strategy for annotation of glucuronides

130
CAN WE PREDICT THE PREFERENCE FOR ADDUCT FORMATION IN
ELECTROSPRAY?
Abstract ID: 495

Presenting author: Stepan Stepanovic, Life Sciences Mass Spectrometry, Department of Inorganic and
Analytical Chemistry, University of Geneva, Quai Ernest Ansermet 24, 1211 Geneva, Switzerland

Introduction

When electrospray ionization method (ESI) is utilized, the obtained spectra are often complicated by adduct formation
with metal or small organic ions. Annotation methods, such as the multi-layered approach, are very powerful, but they do
not offer deeper insight into the microscopic mechanisms. In order to tackle these phenomena, we have developed a
simple theoretical approach that can be used to predict the preference of ion adducts formation, associated with several
protonation/deprotonation equilibria that also play an important role. The strategy consists of using explicit solvation of
reactive sites and density functional theory (DFT) as method of choice. It is simple and bypasses complicated ab-initio
molecular dynamics calculations, necessary to treat the system dynamics as well as bond making-bond breaking
properly.

Methods

All DFT calculations were performed with the Amsterdam Modelling Suite (AMS2021) program package. Initial structures
were refined with global minimization techniques available in AMS, namely Basin Hopping and systematic variation of
dihedral angles, using XTB semiempirical methodology. The obtained low energy structures were later reoptimized with
PBE DFT method, using full electron TZ2P Slater type orbitals basis, Grimme G4 dispersion correction and implicit (as
well as partial explicit) solvation, until the maximum gradient component was less than 5·10 -4 a.u (default is 10-3). The
nature of stationary points is confirmed by calculating analytical Hessians.

Preliminary data (results)

There are many complicated factors that influence the intensity of a specific ion signal in the mass spectrum. With ESI
ionization method, these factors include a delicate interplay between a condensed phase ionization process and
subsequent solvent expulsion.

We have chosen a succinic acid as a model system for our model development, since it gives solely the ion adducts, has
a complicated potential energy surface (with many close lying cyclic and acyclic conformers) and the most important
signals involve carboxylic group deprotonation. Experimental spectra of succinic acid with the ion mixture (conc. of all
salts is 2M) is shown on Figure 1. It can be seen that the most intensive peaks are from M 2+ ion adducts and that only
somewhat intense M+ adduct is with Li+ ion. Protonated succinic acid is not observed.

In order to obtain comparable results for both protonation, deprotonation and ion binding, we have treated all important
interaction centers with two explicit water molecules. The model is depicted on Figure 1. Simple energetics of the entire
process show that M2+ ions bind more favorably to succinic acid than M + ions and that acid protonation is the least
favorable option. Thus, it is experimentally demonstrated and theoretically confirmed that M 2+ ion adducts should no
longer be neglected.

We have also applied our model to methylated uric acid derivatives, and obtained the preference for protonation instead
of adduct formation, which is in excellent accordance with experimental results.

Please explain why your abstract is innovative for mass spectrometry?

The relatively simple methodology, to predict preference for ions adducts vs. protonation is presented.

Co-authors:

Gérard Hopfgartner, Life Sciences Mass Spectrometry, Department of Inorganic and Analytical Chemistry, University of
Geneva, Quai Ernest Ansermet 24, 1211 Geneva, Switzerland

131
Experimental mass spectra and theoretical model for succinic acid-ions mixture

132
NAIVE BAYES CLASSIFICATION MODEL FOR ISOTOPOLOGUE
DETECTION IN LC-HRMS DATA
Abstract ID: 130

Presenting author: Denice van Herwerden, University of Amsterdam

Introduction

Non-target analysis in combination with liquid chromatography high-resolution mass spectrometry is a comprehensive
approach for the characterization of unknown chemicals in complex sample matrices. One of the steps, to reduce the
number of individual features and group information for unique chemical constituents, is the identification of
isotopologues. For this, either predicted molecular formulas, which can be translated to predicted isotopic patterns, or a
theoretical mass difference of 1.0033 (i.e., CAMERA) is commonly used. However, the main shortcomings of the first
approach are the difficulties associated with accurate and reliable molecular formula prediction for unknown chemicals.
While for the mass difference method, an arbitrary mass tolerance is required as an input variable.

Methods

Therefore, a Naive Bayes isotopologue classification model was developed that does not require any thresholds or
molecular formula information. To construct the model, generated isotopic patterns were obtained for the chemicals in
the DDS-Tox database, using the fine and coarse isotopic pattern generators in pyOpenMS (Figure 1A). For the 2.3
million monoisotopic-isotopologue mass pairs, the elemental mass defects of six elemental ratios (i.e., CO, CN, CCl, CS,
CF, and CH) were calculated and used to construct the classification model (Figure 1B). The model performance was
benchmarked against an “in-house” mass difference method for both theoretical isotopologues and wastewater influent
samples.

Preliminary data (results)

The results showed that for the theoretical isotopologues, the classification model outperformed the "in-house" mass
difference with a true positive rate (TPr) of 99.0% and false positive rate (FPr) of 1.8% compared to a TPr of 16.2% and
an FPr of 0.02%, assuming no mass error for the mass difference method. As for the wastewater influent samples, the
classification model, with a TPr of 99.8% and a false detection rate (FDr) of 0.5%, again performed better than the "in-
house" mass difference method, with a TPr of 96.3% and FDr of 4.8%, for which a mass tolerance of 10 mDa was used
to better reflect the inherent mass error for real data. Therefore, it can be concluded that the classification model can be
used as an approach for isotopologue identification, requiring no thresholds or information on the molecular formula.

Please explain why your abstract is innovative for mass spectrometry?

Probabilistic isotopologue classification model requiring no molecular information or arbitrary thresholds (e.g., mass
tolerance).

Co-authors:

Jake O'Brien, The university of Queensland

Phil Choi, The university of Queensland, Queensland Health
Kevin Thomas, The university of Queensland
Peter Schoenmakers, University of Amsterdam
Saer Samanipour, University of Amsterdam

133
Overview of the isotopologue classification model construction and usage.

134
M2AIA EXTENSION FOR ACCESSIBLE ANNOTATION CREATION AND
ANNOTATION TRANSFER FOR MASS SPECTROMETRY IMAGING IN
MULTI-MODAL SETUPS
Abstract ID: 391

Presenting author: Jonas Cordes, Faculty of Computer Science, Mannheim University of Applied Sciences,
Medical Faculty Mannheim, University Heidelberg

Introduction

Applications of deep learning for mass spectrometry imaging (MSI) are still limited by the lack of high-quality ground truth
and benchmark datasets that include metabolite and spatial annotations [1]. Additional challenges arise from the size of
MSI data. Expert spatial annotations are usually expensive and difficult to obtain and mostly created using appropriate
non-MSI modalities. Transferring such annotations to MSI can be challenging and often requires image-based
registration methods [2]. M2aia [3] is an interactive application offering a wide range of MSI processing tools for
visualization, exploration, and image-based registration. We have extended M2aia with methods for creating annotations
and/or transfer of annotations from non-MSI images to corresponding MSI data [4] to meet the requirements of deep
learning and multi-modal setups in general.

Methods

The newly developed extension of M2aia allows to interactively create spatial and point annotations of anatomical
structures on non-MSI, like H&E or infrared spectroscopy imaging (IR) data, and ion images (Figure 1). Applying the
extended rigid and deformable registration methods of M2aia, annotations can be transferred within M2aia between
different modalities and the results can be interactively evaluated. Annotations are saved alongside the original data in a
commonly readable file format (e.g., .nrrd for spatial and .xml for point annotations).

Preliminary data (results)

M2aia (MSI application for interactive analysis in MITK, biotools:m2aia) [3] integrates MSI support into the platform-
independent, open-source Medical Imaging Interaction Toolkit (MITK) [4]. M2aia provides a single graphical user
interface allowing to perform all major MSI processing tasks (Figure 2) without requiring any programming skills.

To evolve M2aia into a platform for multi-modal annotation creation and annotation transfer, we have added support for
non-MSI image data and image-related file formats. M2aia now allows access to MSI (.imzML), IR (e.g. .fsm),
microscopy (e.g., .svs), radiological (e.g., .nrrd, DICOM), and annotation (e.g., .nrrd) image data. M2aia's registration
tools have been extended to allow the transfer of the computed transformations to annotations.

[1] Alexandrov T. Spatial Metabolomics and Imaging Mass Spectrometry in the Age of Artificial Intelligence. Annual
Review of Biomedical Data Science (2020).

[2] Balluff B, Heeren RM, Race AM. An overview of image registration for aligning mass spectrometry imaging with
clinically relevant imaging modalities. Journal of Mass Spectrometry and Advances in the Clinical Lab (2021).

[3] Cordes J, Enzlein T, Marsching C, Hinze M, Engelhardt S, Hopf C et al. M2aia — Interactive, fast, and memory-
efficient analysis of 2D and 3D multi-modal mass spectrometry imaging data. GigaScience (2021).

[4] Sammour DA, Cairns JL, Boskamp T, Kessler T, Guevara CR, Panitz V et al. Spatial Proba-bilistic Mapping of
Metabolite Ensembles in Mass Spectrometry Imaging. bioRxiv (2021).

[5] Wolf I, Vetter M, Wegner I, Böttger T, Nolden M, Schöbinger M et al. The Medical Imaging Interaction Toolkit. Medical
image analysis (2005).

Please explain why your abstract is innovative for mass spectrometry?

The open-source application M2aia provides a fast and interactive user interface for MSI processing, as well as features
for registration and annotation creation/transfer in multi-modal setups and machine learning.

Co-authors:

135
Thomas Enzlein, Center for Mass Spectrometry and Optical Spectroscopy (CeMOS), Mannheim University of Applied
Sciences
Denis Abu Sammour, Medical Faculty Mannheim, University Heidelberg, Center for Mass Spectrometry and Optical
Spectroscopy (CeMOS), Mannheim University of Applied Sciences
Carsten Hopf, Medical Faculty Mannheim, University Heidelberg, Center for Mass Spectrometry and Optical
Spectroscopy (CeMOS), Mannheim University of Applied Sciences
Ivo Wolf, Faculty of Computer Science, Mannheim University of Applied Sciences

Overview of extended multi-modal image and annotation processing in M2aia.

136
FT-IR and consecutive MSI images with example annotations in M2aia.

137
WORKFLOW BASED ON TIC ALIGNMENT FOR RETROSPECTIVE
ANALYSIS OF LOW-RESOLUTION ON-LINE SPE-GC-MS DATA
Abstract ID: 93

Presenting author: Nienke Meekel, KWR Water Research Institute

Introduction

The drinking water production in the Netherlands partially relies on surface water. Continuous online monitoring of this
source is essential for safeguarding the quality of the production. To map the chemical space of organic contaminants,
GC- and LC-MS are the methods of choice. High-resolution MS is gaining more and more attention within the field of
environmental analytical chemistry. Yet, this method is still mostly used in research laboratories, as the required
instruments are often expensive and data analysis can be more complex compared to low-resolution MS. Currently,
routine laboratories carrying out conventional monitoring activities still heavily rely on low-resolution/nominal mass
spectrometers for screening purposes. The lack of accurate mass combined with slight differences in instrumental
conditions complicate the (automatic) analysis of long time series.

Methods

Here, we present a workflow, which tackles these issues. The workflow has been developed and validated using a
particular data set. The data considered here covers multiple years of water samples collected from the river Rhine every
12 hours, analysis was performed using an automated (online) solid-phase extraction – gas chromatography mass
spectrometry (SPE-GC/MS) system. The workflow was developed to detect anomalies (e.g., sudden occurrence of
previously unknown chemicals) in this dataset and facilitate identification. Since low-resolution data lacks information
about the accurate mass of detected ions, the alignment of different chromatograms becomes more complicated than
when using HRMS data.

Preliminary data (results)

When using HRMS data, individual features can first be selected (using their accurate mass as criterium) in each
individual chromatogram and then these can be aligned across multiple chromatograms.In the case of low-
resolution/nominal data, alignment has to be applied to the whole signal before ions can be selected for further analysis.
Substantial shifts occur between these TICs since the data set consisted of chromatograms measured in individual
measurement series.

Therefore, an algorithm was developed for TIC alignment, using the retention time of mass-labelled internal standards
instead of n-alkanes like in the Kováts index, as given in Eq. 1. Where Rty corresponds to the corrected retention time,
which is calculated from the original retention time (Rtx), fixed retention times for the internal standards RtA and RtB and
the measured retention times from the internal standards Rt’A and Rt’B.

Rty=RtA+((RtB-RtA)/(Rt’B-Rt’A))*(Rtx-Rt’A) Eq. 1

The algorithm relies on the retention time range and specific mass spectrum of the internal standard. Existing R-
packages for the analysis of GC-MS data were evaluated and various combinations of preprocessing techniques (i.e.
standard normal variate, multiplicative scatter correction, Savitzky-Golay smoothing, polynomial smoothing, 1st and
2nd derivatives) were tested as well.

The resulting preprocessed data were compared using principal component analysis and hierarchical clustering
techniques in order to visualize and detect particularities within the data set that could correspond to discharges and
other calamities.

Please explain why your abstract is innovative for mass spectrometry?

Despite the low resolution, low-resolution/nominal MS data can contain valuable information and reveal spatial and
temporal trends, facilitated by the developed workflow using an algorithm developed for TIC alignment.

Co-authors:

Erik Emke, KWR Water Research Institute

Ton van Leerdam, KWR Water Research Institute
Chris Lukken, Rijkswaterstaat (CIV)

138
Henk Zemmelink, Rijkswaterstaat (CIV)
Frederic Béen, KWR Water Research Institute

DECONVOLUTION-FREE FEATURE EXTRACTION AND ANNOTATION VIA

TIME-DOMAIN TRANSIENT MODELLING IN ORBITRAP FTMS FOR
BIOPHARMA APPLICATIONS
Abstract ID: 379

Presenting author: Yury Tsybin, Spectroswiss

Introduction

Orbitrap FTMS analysis of protein biopharmaceuticals, including monoclonal antibodies (mAbs), often results in mass
spectral datasets of very high complexity that require advanced data processing and analysis software tools. Rigorous
feature extraction from the FTMS data determines the quality and confidence of the subsequent data analysis and
biologically-relevant conclusions. In addition, modern biopharma workflows demand data processing procedures to be
computationally efficient and suitable for automation. Deconvolution algorithms are available, but require additional data
processing steps and may introduce artifacts. Direct simulation of the mass spectra can be inaccurate due to the FTMS
specificity, such as isotopic beats. We will demonstrate how the rational use of the FTMS data specificity, namely the
time-domain transient processing, enhances the feature extraction performance and benefits sample analysis workflows.

Methods

Samples were obtained commercially or via collaborators. Experimental data (intact and subunit mass analysis, top-
middle down, and bottom-up) were acquired with diverse Hybrid and Tribrid Orbitrap instruments with or without time-
domain transients acquisition via an external data acquisition system (FTMS Booster, Spectroswiss). The original mass
spectra in the .RAW format (eFT data) were acquired in parallel to mass spectra in absorption mode FT (aFT) generated
from time-domain transients. In-silico FTMS mass spectra were generated using the FTMS Simulator (Spectroswiss)
from databases containing protein sequences and possible modifications. Data were processed and experimental
parameters were automatically extracted with Peak-by-Peak BioPharma (Spectroswiss).

Preliminary data (results)

Preliminary results demonstrate that accurate modelling of Orbitrap FTMS (profile) mass spectra for small (peptides, <
25 kDa), medium (25-50 kDa), and large (> 50 kDa) molecules via time-domain transients opens the doors toward
accurate, rapid, and deconvolution-free feature extraction. The Orbitrap data specificity manifests itself via resolution
dependence on m/z and isotopic beats, and peak interference.

We thus developed an Orbitrap data-specific software tool for targeted and semi-targeted feature extraction from the
direct infusion and LC-MS data for molecules of any size (including mAbs and viruses). The feature extraction process is
based on the similarity scoring between the resolved or unresolved isotopic envelopes of the experimental and simulated
data. The isotopic envelopes can be matched individually or grouped. The groups can contain charge state distributions
or proteoform clusters (e.g., glycosylation profiles of mAbs).

The initial step in feature extraction is the generation of the simulated library of isotopic envelopes based on the user-
specified library of elemental compositions or sequences, as well as any modifications. In addition to the targeted and
curated databases of compounds, large-scale public and in-silico generated databases can be employed.

The described approach and tool, supported with a graphical user interface, demonstrated excellent analytical specificity,
high sensitivity, and quantitative precision when applied to biopharma applications at different levels of structural
complexity: intact mass, middle-up, and bottom-up. It simplifies the processing and interpretation of FTMS data in the
experimental context for non-expert users and supports automation. Further approach extension to the MS/MS-based
workflows is on-going.

Please explain why your abstract is innovative for mass spectrometry?

Accurately-simulated Orbitrap FTMS data using time-domain transient modelling enables deconvolution-free targeted
and semi-targeted feature extraction and annotation for qualitative and quantitative analysis of protein
biopharmaceuticals at intact, subunit, peptide levels

139
Co-authors:

Anton Kozhinov, Spectroswiss

Natalia Gasilova, Ecole Polytechnique Fédérale de Lausanne
Laure Menin, Ecole Polytechnique Fédérale de Lausanne
Luca Fornelli, University of Oklahoma, Department of Biology
Konstantin Nagornov, Spectroswiss

140
Session: LS-5 Glycomics & Glycoproteomics - Session A

NEGOTIATING THE MAZE OF O-GLYCOSYLATION

Abstract ID: 786

Presenting author: Zsuzsanna Darula, Hungarian Centre of Excellence for Molecular Medicine, Szeged, Hungary,
Biological Research Centre, Szeged, Hungary

Introduction

Comprehensive characterization of O-glycosylation is a multilayer task with many aspects. The proteins glycosylated, the
glycan structures decorating the protein, and the sites of modifications all have to be assigned. Moreover, macro- and
microheterogeneity have to be addressed, as O-glycosylation frequently is non-stoichiometric and each glycosylation site
may feature a wide variety of glycans in different subpopulations of the glycoprotein. Our research team aims at the
characterization of mucin-type O-glycosylation in body fluids in a site-specific, high-throughput manner. Currently we are
focusing on human urine samples, and we have a complete story to tell, from the glycopeptide enrichment to the in-depth
data interpretation.

Methods

Glycopeptides were enriched from human urine tryptic digests by lectin affinity chromatography then analyzed by LC-
MS/MS using HCD fragment ion-dependent EThcD. Multiple search engines were used for iterative data interpretation
including Byonic, Protein Prospector and OPair. First, glycoproteins were identified using a full database search allowing
only the most common O-glycans followed by a second search on the subset of glycoproteins allowing a larger O-glycan
database. MS-Filter was also used to identify additional glycoforms of the peptides using HCD data. However, manual
data interpretation at the beginning of the project and evaluating the automatic assignments was still essential.

Preliminary data (results)

We analyzed a dataset of 60 LC-MS/MS experiments representing the urinary glycoproteome of ten donors. We found
that urinary O-glycan structures show a much higher diversity than expected based on available serum and plasma
reports. We identified more than 30 O-glycan structures that have never been reported in a site-specific manner including
blood-group antigens or O-acetylated sialic acids in glycans carrying disialic acid capping. The sugar-unit connectivity of
these structures was deciphered from specific glycan fragments formed by preferential single bond cleavages in EThcD
performed at minimal NCE. The novel structures were incorporated into our glycan database. Our iterative data approach
allowed more reliable interpretation of the data compared to single-run searches using large protein and glycan
databases simultaneously. At the same time, we tried to identify the strength and limitations of the different software tools
and come up with some ideas for their improvement.

This work was supported by the following grants: EU Horizon 2020 grant No. 739593, GINOP-2.3.2-15-2016-00001, and
GINOP-2.3.2-15-2016-00020.

Please explain why your abstract is innovative for mass spectrometry?

EThcD performed at minimal NCE favors single bond cleavages, that could be utilized in data interpretation. Deciphering
glycan unit connectivity and assignment of different glycoforms based on these cleavages.

141
DETERMINATION OF O-ACETYL POSITIONS ON SIALIC ACIDS WITH ION
MOBILITY-MASS SPECTROMETRY
Abstract ID: 117

Presenting author: Kevin Hooijschuur, Chemical Biology and Drug Discovery, Utrecht University

Introduction

Glycans are involved in many biological processes and diseases, and affect efficacy and pharmacokinetics of
biopharmaceuticals. Their analysis is complicated due to high heterogeneity, structural complexity, and high degree of
functionalisation. An important functionalisation is the O-acetylation of sialic acids on glycoproteins and glycolipids, which
is associated with cancer and pathogen infection. The analysis of O-acetylation with current techniques, however, is
challenging as O-acetyls are highly prone to migration and hydrolysis,and the acetylation, which occurs on the C-4, C-7,
C-8, and C-9 hydroxyls, leads to isomers that are difficult to distinguish. The lack of well-defined O-acetylated standards
hinders determination of exact acetylation position. Here we present a methodology using ion mobility-MS for structural
characterization of glycans with O-acetylated sialic acids.

Methods

A set of oligosaccharide standards, containing sialyl-N-acetyllactosamine (sialyl-LacNac) with N-acetylneuraminic

(NeuAc) and N-glycolylneuraminic (NeuGc) sialic acids attached via α2,3 or α2,6 linkage, was chemoenzymatically
synthesized. The sialic acids are O-acetylated on the C-4, C-7 and C-9 hydroxyls and most possible combinations
thereof and were applied to the development of a drift tube ion mobility spectrometry (DTIMS)-MS methodology for the
analysis of sialylated glycans. Multiplexing and associated high-resolution demultiplexing was used in the DTIMS
dimension to obtain high-resolution separation and collision cross section values of isomeric B-fragment ions.

Preliminary data (results)

It was found that DTIMS-MS can separate almost all the acetylated α2,3- and α2,6-linked sialyl-LacNac oxonium (B3)
ions. For standards where the B3 ions co-migrated, such as for the α2,6-linked NeuAc with C-7 and C-9 O-acetylation,
the acetylation could be determined by resolved sialic acid oxonium (B1) ions. This enables determination of both linkage
type as well as acetylation position on acetylated sialic acids isomers in glycans from biological samples.

The methodology was applied to the analysis of sialoglycans on mucins, biopharmaceuticals and glycolipids using a CCS
library obtained with the fragment ions of the oligosaccharide standards. O-glycans from isolated bovine submaxillary
mucin and porcine stomach mucin revealed a high abundancy of O-acetylated sialic acids and the positional expression
of O-acetylation was determined in both sample types. The analysis of sialoglycans, isolated from biopharmaceuticals
that were expressed in CHO cells, revealed the presence of O-acetylation on sialic acid residues of N-glycans. The
methodology was also applied to the determination of mouse brain gangliosides, showing exact O-acetylation position of
terminal sialic acids. These results demonstrate that DTIMS-MS in combination with a CCS library of fragment ions can
be applied for structural characterisation of glycans with O-acetylated sialic acids in biological samples without further
need for synthetic standards.

Please explain why your abstract is innovative for mass spectrometry?

High-resolution drift tube ion mobility-MS allows glycosidic bond type and exact O-acetylation position determination in
sialylated glycans without the need for intact synthetic standards.

Co-authors:

Gaël Vos, Chemical Biology and Drug Discovery, Utrecht University

Zeshi Li, Chemical Biology and Drug Discovery, Utrecht University
Chris Klein, Agilent Technologies
John Fjeldsted, Agilent Technologies
Javier Sastre Toraño, Chemical Biology and Drug Discovery, Utrecht University
Geert-Jan Boons, Chemical Biology and Drug Discovery, Utrecht University

142
MUCINS O-GLYCOMICS BY HIGH RESOLUTION ION MOBILITY-MASS
SPECTROMETRY
Abstract ID: 485

Presenting author: Leila Bechtella, Freie Universität Berlin

Introduction

The dense O-glycosylation of mucins plays an important role in the defensive barrier of the mucus hydrogel against
pathogens and aberrant glycosylation is often correlated with inflammation and diseases such as cancer and Crohn’s
disease. However, the inherent complexity of glycans combined with the diversity of the O-core structure constitute
fundamental challenges for mucins glycomics. The variability in composition, connectivity, and configuration of glycans
leads to the coexistence of multiple isomers, requiring efficient separative techniques. Therefore, multidimensional
workflows such as LC-MS are traditionally employed, using a Porous Graphitized Carbon (PGC) stationary phase to
separate the highly polar carbohydrates. However, this robust state-of-the-art method is time-consuming. As a high
throughput solution, Ion Mobility Spectrometry (IMS) coupled to MS constitutes a promising alternative for O-
glycomics.

Methods

O-glycans are released from mucins by β-elimination and reduced. The obtained alditols are ionized in negative ion
mode using an in-house 3D-printed offline nanoESI source connected to a timsTOF Pro (Bruker). The
deprotonated O-glycans are separated in the Trapped Ion Mobility Spectrometry (TIMS) cell and fragmented by
collision-induced dissociation (CID). The IM-MS/MS workflow is developed using a commercially available mucin
standard, and then applied to clinical sputum samples from healthy and cystic fibrosis patients as a proof-of-principle.

Preliminary data (results)

Compared to the LC-MS/MS workflow, the IM-MS/MS analysis time is drastically reduced, to acquisitions of few
minutes. The obtained isomers separation is compared to previously reported results using PGC and Traveling Wave
IMS (TWIMS).[1] The high IM-resolving power of the TIMS allows to separate the O-glycan isomers and even to
detect new structures (Figure 1). The workflow is successfully applied to clinical sputum samples and
characteristic glycosylation features are measured and compared between the healthy and Cystic Fibrosis patients.

[1] Jin, C., Harvey, D. J., Struwe, W. B., Karlsson, N. G. (2019). Separation of isomeric O-glycans by ion mobility and
liquid chromatography-mass spectrometry. Analytical Chemistry, 91(16), 10604-10613.

Please explain why your abstract is innovative for mass spectrometry?

A new method is presented for mucin glycomics using IM-MS/MS on a timsTOF instrument. Isomeric separation of O-
glycans is performed using Trapped Ion Mobility Spectrometry. New structures are identified.

Co-authors:

Güney Ertürk, Freie Universität Berlin

Weston B. Struwe, University of Oxford
Jin Chunsheng, University of Gothenburg
Niclas G. Karlsson, University of Gothenburg
Kevin Pagel, Freie Universität Berlin, Fritz-Haber Institute

143
Separation of O-glycan isomers using PGC, TWIMS or TIMS.

144
ANTIBODY ARRAY BASED N-GLYCAN IMAGING OF CAPTURED IMMUNE
CELLS
Abstract ID: 559

Presenting author: James W Dressman, Medical University of South Carolina

Introduction

Cell surface N-glycosylation plays an important role in both the innate and adaptive immune response through the
modulation of cell surface receptors. The study of immune cell N-glycosylation is increasingly becoming a field of
interest, but hindered by the complexity of cell type specific N-glycan analysis. Analytical techniques such as
chromatography, LC-MS/MS and lectin blotting all are currently used to analyze immune cell N-glycosylation. Pitfalls
from each analytical technique hinder either throughput or the acquisition of structural data, thereby reducing their
feasibility for N-glycan study. Here, we report development of a rapid antibody array-based approach for the capture of
specific nonadherent immune cells coupled with MALDI-IMS to analyze cell surface N-glycosylation, increasing the
feasibility of immune cell N-glycan analysis.

Methods

Cell N-glycosylation was analyzed using MALDI-IMS, relying on enzymatically released N-glycans from cells bound to
immobilized antibodies. Glycans were released by peptide N-glycosidase (PNGase F PRIME) digestion coupled with the
removal of terminal sialic acid residues via sialidase digestions or the stabilization of the sialic acid residues by an
amidation-amidation reaction. Cells analyzed included HepG2 cells and primary C57BL/6 T cells. Released N-glycans
were detected by a timsTOF-fleX mass spectrometer (Bruker) and data was analyzed using SCiLS lab software (Bruker
v2022a). Fluorescently labeled cells were imaged with an Odyssey Imaging System (LI-COR Biosciences).

Preliminary data (results)

The primary goal of the project was to determine the feasibility of cell capture with an immobilized antibody specific to a
surface receptor, and to analyze captured cell N-glycans via MALDI-IMS. HepG2 cells were initially tested with an anti-
transferrin receptor antibody coupled to a hydrogel coated slide. Following blocking and incubation, cells bound only to
the antibody spot, immobilized by the cell receptor antibody interaction. Utilizing well established N-glycan MALDI
imaging protocols, enzymatically released N-glycans were able to be detected from antibody captured cells. Following
confirmation of capture, the method was further optimized to improve reproducibility. The intended application of this
assay is to examine nonadherent immune cell N-glycosylation due to their inherent difficulty to analyze. Primary C57BL/6
mouse splenocytes were used to test the specificity and feasibility of immune cell capture. A mixture of splenocytes was
washed over spotted CD4 and CD8 immobilized antibodies, subsequently capturing CD4 and CD8 T cells. Abundant N-
glycans detected on both naive CD4 and CD8 T cells consisted of high mannose and biantennary structures. The most
predominate biantennary structure had one α(2,3) and one α(2,6) linked terminal sialic acid residue, both with and
without a fucose. Biantennary structures containing a fucose were the most abundant. Specificity post N-glycan imaging
was tested using biotinylated CD4 and CD8 antibodies followed by streptavidin 800 fluor incubation to confirm CD4 and
CD8 T cells were specifically captured from the heterogenous mixture of splenocytes.

Please explain why your abstract is innovative for mass spectrometry?

First method to image N-glycans from antibody array captured immune cells with MALDI-IMS.

Co-authors:

Colin T McDowell, Medical University of South Carolina

Xiaowei Lu, Medical University of South Carolina
Peggi M Angel, Medical University of South Carolina
Richard R Drake, Medical University of South Carolina
Anand Mehta, Medical University of South Carolina

145
Antibody captured mouse CD8 T cells

N-Glycan image of captured mouse CD8 T cells

146
ESTABLISHING STRUCTURAL MS TO UNDERSTAND PROTEIN
GLYCOSYLATION IN NEUROLOGICAL FUNCTION AND DISEASE
Abstract ID: 630

Presenting author: Melissa Bärenfänger, Division of Bioanalytical Chemistry, VU Amsterdam

Introduction

After numerous advances in genomics and proteomics, our understanding of the glycoproteome is vastly lacking behind.
This applies especially to the role of glycosylation in the central nervous system and its function in neurological diseases
such as Parkinson's or Alzheimer's. To understand the role of protein glycosylation in brain pathologies, we characterize
the glycoproteome of patients diagnosed with congenital disorder of glycosylation as a model system to link severe
changes in glycosylation to impaired neurological function. Beyond this correlation-based study, we aim to establish a
causal relationship between altered protein glycosylation and structural changes in protein conformation by employing
multidimensional MS technologies such as ion mobility-based collision induced unfolding. Understanding how distinct
glycoproteoforms impact protein structure helps to unravel how protein glycosylation affects disease mechanisms.

Methods

Here, we are using mass spectrometry-based glycoproteomics to analyze protein N-glycosylation in cerebrospinal fluid
(CSF) of patients diagnosed with different types of CDGs and healthy controls. After detecting glycopeptides showing
disease-related changes in glycosylation, we analyze corresponding glycoproteins as whole biomolecules. We use native
mass spectrometry in combination with ion mobility to reveal change in charge state distribution and drift time of properly
glycosylated proteins vs. proteins with altered glycosylation. By employing collision induced unfolding, we analyze the
impact of altered glycosylation on conformational stability of individual glycoproteoforms to understand the role of protein
glycosylation in disease mechanisms.

Preliminary data (results)

Our glycoproteomics approach in CSF allows us to monitor changes in protein glycosylation and the complete loss of
glycosylation simultaneously. With this, we were able to detect over 900 unique glycopeptides from almost 200 proteins
in non-enriched CSF samples to study N-glycosylation and site-occupancy of several glycoproteins. For most CDG
cases, we observe that both brain-derived proteins and non-brain-derived proteins exhibit a severe change in
glycosylation with highly truncated and absent glycans. To understand how these truncated or missing glycans affect
protein conformation, stability and ultimately function, we analyzed native glycoproteins and compared the protein
stability of proteins with intact glycosylation and proteins with shortened glycosylation. A first indication of structural
changes induced by altered glycosylation can be seen by changes in charge-state-distribution in native MS. Additionally,
ESI parameters were adjusted to detect different glycoproteoforms and to isolate the mass of multiply charged species.
The mass-isolated glycoproteoforms were analyzed by ion mobility to determine differences in protein shape. To
evaluate the effects of protein glycosylation on protein stability, we performed collision-induced unfolding experiments
with different mass-selected glycoproteoforms and demonstrated the importance of protein glycosylation for protein
stability.

Please explain why your abstract is innovative for mass spectrometry?

We demonstrate a glycoproteomic approach examining changes in glycosylation and site occupancy in neuropathology.
We use these results to establish a relationship between glycosylation and protein stability, using structural MS.

Co-authors:

Sjors Bakels, Division of Bioanalytical Chemistry, VU Amsterdam

Hans Wessels, Translational Metabolic Laboratory, Department of Laboratory Medicine, Radboud University Medical
Center
Merel Post, Department of Neurology, Donders Institute for Brain, Cognition, and Behaviour, Radboud University Medical
Center
Pieter Langenhorst, Translational Metabolic Laboratory, Department of Laboratory Medicine, Radboud University
Medical Center
Karin Huijben, Translational Metabolic Laboratory, Department of Laboratory Medicine, Radboud University Medical
Center
Fokje Zijlstra, Translational Metabolic Laboratory, Department of Laboratory Medicine, Radboud University Medical
Center
Marcel Verbeek, Department of Neurology, Donders Institute for Brain, Cognition, and Behaviour, Radboud University

147
Medical Center, Translational Metabolic Laboratory, Department of Laboratory Medicine, Radboud University Medical
Center
Dirk Lefeber, Department of Neurology, Donders Institute for Brain, Cognition, and Behaviour, Radboud University
Medical Center, Translational Metabolic Laboratory, Department of Laboratory Medicine, Radboud University Medical
Center
Anouk Rijs, Division of Bioanalytical Chemistry, VU Amsterdam

148
DECIPHERING SIALIC ACID PATHWAY REGULATION VIA NOVEL IN-
DEPTH MULTI-OMICS APPROACH IN TISSUE-SPECIFIC HUMAN MODELS
Abstract ID: 722

Presenting author: Merel Post, Department of Neurology, Donders Institute for Brain, Cognition and Behavior,
Radboud University Medical Center

Introduction

Sialic acid is the most abundant terminal monosaccharide on glycan structures of glycoproteins, glycolipids and
gangliosides. Due to their abundance and electronegative charge, they have many different functions including cell-cell
interactions, protein stabilization and ion transport. Sialic acid is also involved many disorders such as Salla disease,
atherosclerosis and congenital disorders of glycosylation (CDG). Mutations in sialic acid synthase gene (NANS) causes
NANS-CDG with diverse symptoms including skeletal dysplasia, short stature and facial dysmorphisms. They have a
clear neurological phenotype including neurodegeneration and intellectual disability. However, other genetic defects
related to sialic acid metabolism have different phenotypes, suggesting the existence of tissue-specific metabolic
mechanisms. Therefore, we studied sialic acid metabolism in patient plasma, fibroblasts, induced pluripotent stem cells,
its derived neurons and HAP1 knockout cells.

Methods

We applied a targeted MRM approach using UHPLC-QqQ mass spectrometry to analyze metabolites in the sialic acid
synthesis and hexosamine pathway in HAP 1 NANS knockout cells, patient-derived fibroblasts and patient iPSC derived
glutamatergic neurons (iNeurons). Separation of nucleotide sugars and HexNAc-phosphates was achieved by ion pair-
reverse phase chromatography using a HSS T3 column.In addition, we did glycomic and glycoproteomic analysis of
plasma of 15 NANS-CDG patients. Enriched glycopeptide mixtures were separated by nanoflow liquid chromatography
and analyzed on a timsTOF Pro 2 instrument optimized for glycopeptide analysis. All data was analyzed using Skyline
and MS Fragger.

Preliminary data (results)

First, we characterized all metabolites involved in the sialic acid and hexosamine biosynthesis pathway in
HAP1 NANS KO cells, patient fibroblasts and iNeurons to obtain an overview of aberrant metabolism caused by
defective NANS. HAP1 NANS KO cells, patient fibroblasts, patient derived iPSCs and its derived iNeurons showed
dramatically high levels of ManNAc-6P in line with enzymatic deficiency of NANS. Unexpectedly however, the levels of
CMP-sialic acid and sialic acid were only decreased in NANS KO cells. Surprisingly, UDP-HexNAc levels were clearly
decreased in NANS KO cells and moderately decreased in patient fibroblasts, iPSCs and iNeurons. Further detailed
analysis of the metabolic pathway revealed a block in hexosamine biosynthesis pathway due to accumulating ManNAc-
6P, which was confirmed by direct enzymatic assays. To study the effect of this mechanisms on glycans and
glycoproteins, total plasma glycomics and glycoproteomics was performed. Comparing healthy controls to NANS-CDG
patients no clear differences were observed using glycomics. However, using glycoproteomics, site-specific effects of
NANS deficiency on sialylation and N-glycan branching was discovered. In NANS patients a general overrepresentation
of hybrid and high-mannose glycans as well as hyposialylated glycans was perceived. The effect that we observed on N-
glycan branching is likely caused by the inhibition of hexosamine biosynthesis pathway, which we saw in
all NANS deficient cells, uncovering a novel regulatory pathway for the sialic acid pathway.

Please explain why your abstract is innovative for mass spectrometry?

We present a metabolomics workflow to study sugar metabolism in combination with a glycoproteomic approach for in-
depth analysis for a multi-omics workflow to analyze sugar metabolism from monosaccharide to glycoprotein.

Co-authors:

Rachel Mijdam, Department of Neurology, Donders Institute for Brain, Cognition and Behavior, Radboud University
Medical Center, Department of Human Genetics, Donders Institute for Brain, Cognition and Behavior, Radboud
University Medical Center
Melissa Bärenfänger, Division of BioAnalytical Chemistry, Amsterdam Institute for Molecular and Life Sciences, Vrije
Universiteit Amsterdam
Raisa Veizaj, Department of Neurology, Donders Institute for Brain, Cognition and Behavior, Radboud University Medical
Center
Fokje Zijlstra, Translational Metabolic Laboratory, Department of Laboratory Medicine, Radboud University Medical
Center

149
Johan Pijnenborg, Department of Synthetic Organic Chemistry, Institute for Molecules and Materials, Radboud University
Clara van Karnebeek, Department of Pediatric Metabolic Diseases, Emma Children's Hospital, Amsterdam University
Medical Center
Hans Wessels, Translational Metabolic Laboratory, Department of Laboratory Medicine, Radboud University Medical
Center
Thomas Boltje, Department of Synthetic Organic Chemistry, Institute for Molecules and Materials, Radboud University
Nael Nadif Kasri, Department of Human Genetics, Donders Institute for Brain, Cognition and Behavior, Radboud
University Medical Center, Department of Cognitive Neuroscience, Radboud University Medical Center, Donders Institute
for Brain, Cognition and Behavior
Dirk Lefeber, Department of Neurology, Donders Institute for Brain, Cognition and Behavior, Radboud University Medical
Center

150
Session: FP-2 Biopharmaceuticals & Vaccines

PROFILING HEAVILY GLYCOSYLATED BIO THERAPEUTICS WITH

CHARGE REDUCTION NATIVE MASS SPECTROMETRY
Abstract ID: 280

Presenting author: Wendy Sandoval, Genentech, Inc

Introduction

Characterization of bio therapeutic moieties typically involve the application of multiple orthogonal analytical methods to
evaluate product heterogeneity, quality and purity. Investigations into molecular traits include the evaluation of
aggregates, variants, degradation products, and post-translational modifications such as glycosylation. As modalities
evolve beyond monoclonal antibodies and increase in their complexity, strategies to resolve proteoforms within a single
molecule have become increasingly critical. Common approaches for resolving heterogeneous glycoforms include partial
or complete enzymatic removal of sugars or bottom up analyses, neither of which demonstrate the full representation of
the molecule. Here we showcase hyphenated native MS approaches to profile heavily glycosylated Fc fusion proteins,
other engineered bio therapeutics and biologically relevant proteins within a single experiment with no prior reduction of
complexity.

Methods

Analysis was performed on a modified Orbitrap Eclipse equipped with PTCR and extended mass range. Resolution was
obtained through a combination of gas phase proton transfer reduction reactions, ion isolation, and spectral stitching.
Samples included various heavily glycosylated Fc fusion proteins and other biomolecules from which charge state
overlap from intact mass analysis was extensive and did not provide sufficient differentiation for deconvolution under
standard native MS conditions.

Preliminary data (results)

Here we present the annotated results of heretofore ‘undeconvolvable’ spectra in which charge state resolution cannot
be determined. We detail the segmented acquisition methodology employed and demonstrate universal applicability on
various bioengineered molecules and also proteins containing complicated spectrally overlapped isoforms. We showcase
a bioengineered Fc fusion protein containing eight N-linked glycosylation sites, and identified over 160 validated
glycoforms from a single intact analysis. Using this method we also demonstrate the detection of differential sialic acid
modifications from a series of biomolecules with known sialic acid molar content.

Please explain why your abstract is innovative for mass spectrometry?

This methodology has universal applicability for profiling spatially unresolved proteoforms from a single moiety.
Regardless of heterogeneity, coincidental charge states are resolved to determine nominal masses of proteoforms
present.

Co-authors:

Luis Schachner, Genentech, Inc

Romain Huguet, Thermo Fisher Scientific
Kristina Srzentic, Thermo Fisher Scientific
Christopher Mullen, Thermo Fisher Scientific
Josh Hinkle, Thermo Fisher Scientific
David Bergen, Thermo Fisher Scientific
John Syka, Thermo Fisher Scientific

151
MASS SPECTROMETRY AS A KEY TECHNIQUE FOR THE
CHARACTERIZATION OF BISPECIFIC MONOCLONAL ANTIBODIES
Abstract ID: 84

Presenting author: Bastiaan Duivelshof, Institute of Pharmaceutical Sciences of Western Switzerland (ISPSO),
University of Geneva, School of Pharmaceutical Sciences, University of Geneva

Introduction

Bispecific monoclonal antibodies (bsAbs) combine the antigen recognition sites of two or more antibodies in a single
protein construct and therefore allow binding of two different target epitopes. High-Resolution Mass Spectrometry
(HRMS) plays an important role in the characterization of bsAbs since the expression of asymmetric IgG antibodies could
involve a potential mismatch between the heavy chains (HC) and light chains (LC) resulting in multiple undesirable by-
products and prevent the correct therapeutic functionalities. In addition, bsAbs are prone to similar post-translational
modifications (PTMs) as conventional mAbs, but due to the asymmetric nature of the bsAbs, these modifications can
occur on one or both of the heterodimer chains.

Methods

In this work, we combined HRMS with different chromatographic techniques to characterize the bsAb emicizumab
consisting of two different HCs, each with a specific antigenic recognition site, and a common LC fragment. Intact native
HRMS was coupled to cation exchange chromatography (CEX) and size exclusion chromatography (SEC) to verify the
correct chain heterodimerization and to perform charge- and size- variants characterization, respectively. To obtain the
site-specific information on the PTMs, a middle-up approach was additionally used in combination with reversed-phase
liquid chromatography (RPLC) and hydrophilic interaction chromatography (HILIC) coupled to HRMS.

Preliminary data (results)

Intact native SEC- and CEX-HRMS analysis showed that the emicizumab sample consisted of correct heterodimerization
and the absence of homo-dimer traces. The intact protein was further enzymatically digested and chemically reduced to
create protein subunits of ~25 kDa. Analysis of these subunits provided the important site- and chain-specific information
on the presence of PTMs, such as, N-terminal pyroglutamic acid formation and glycosylation on the bispecific antibody.
In this context, the proper characterization of the glycosylation profiles of each HC was enabled by using the extracted
ion chromatograms (EIC) of the subunits separated by HILIC-MS. This demonstrated a symmetrical profile between the
two HC of the bsAb.

In conclusion, our work demonstrates that HRMS is a key technique in the comprehensive characterization of new bsAbs
and that the combination of intact and middle-up level analysis provides the desired information on correct chain
heterodimerization as well as the important site-/chain-specific information on critical quality attributes.

Please explain why your abstract is innovative for mass spectrometry?

Direct hyphenation of native SEC and IEX to HRMS for determination of correct heterodimerization

Use of RPLC and HILIC-HRMS for the characterization of chain/site-specific PTMs on bsAbs

Co-authors:

Alain Beck, IRPF - Centre d’Immunologie Pierre-Fabre (CIPF)

Davy Guillarme, Institute of Pharmaceutical Sciences of Western Switzerland (ISPSO), University of Geneva, School of
Pharmaceutical Sciences, University of Geneva
Valentina D'Atri, Institute of Pharmaceutical Sciences of Western Switzerland (ISPSO), University of Geneva, School of
Pharmaceutical Sciences, University of Geneva

152
EXPANDING FUNCTIONAL ANTIBODY CHARACTERIZATION TO
PROTEOFORMS: AFFINITY CE-MS FOR ANTIBODY – FCRS BINDING
ASSESSMENT
Abstract ID: 185

Presenting author: Christoph Gstöttner, Leiden University Medical Center

Introduction

The Fc domain of monoclonal antibodies (mAbs) has several key functionalities, such as recruitment of immune cells via
different Fcγ receptors (FcRs), activation of the complement system and recycling of the antibody via binding to the
neonatal Fc Receptor (FcRn). Structural features of the Fc domain strongly influence these interactions and small
variations in the Fc region (e.g. glycosylation, oxidation) can severely impact their binding. Unfortunately, common
approaches, such as SPR provide an overall affinity response for all different mAb proteoforms and their individual
binding assessment requires tedious production or enrichment of the targeted proteoforms. We have exploited, for the
first time, the capabilities of Capillary Electrophoresis hyphenated with Mass Spectrometry (CE-MS) to study the binding
affinity of antibodies and FcRs in a proteoform-resolved fashion.

Methods

Affinity CE-MS experiments were performed using sheathless CE-MS. Sheathless integrated capillary electrophoresis
electrospray ionization was carried out on a CESI 8000 instrument (Sciex, Brea, CA) coupled to either an orbitrap-EMR
MS or a 15 tesla SolariX XR™ FTICR from Bruker Daltonics equipped with a nano-electrospray source. Protein
separations were performed using neutrally coated capillaries with a porous tip. Solutions of 50 mM ammonium acetate
at pH 6-7 with and without FcRs were used as background electrolytes.

Preliminary data (results)

We have developed different methods based on mobility-shift affinity CE-MS to study the binding of mAbs to various
FcRs, namely FcRn, FcγRIIa and FcγRIIb. To this end, the FcR receptors were added to the background electrolyte,
whereas the mixtures of antibody proteoforms were injected in the CE. As a first case, we studied the interaction towards
FcRn which determine antibody half-life. We will show that, by adding different amounts of FcRn to the background
electrolyte, we are able to determine the relative affinity of different proteoforms based on their mobility shifts. We
observed lower mobility shifts for singly and doubly oxidized mAbs compared to the unmodified antibodies indicating
lower binding affinity. For FcγRIIa (activating) and FcγRIIb (inhibitory) receptors, glycosylation of the antibody was key for
the binding. Hemiglycosylated antibodies showed strong decrease in the binding towards both FcγRIIs. Differences were
also observed within glycoforms with high mannose forms showing lower binding than complex type glycoforms.

The developed approach offers unique possibilities to study in solution binding of individual proteoforms and
simultaneously to address their structural heterogeneity. We believe that our approach will have a tremendous influence
and benefit in the study of the interactions of mAb proteoforms with different FcRs in biopharma. Understanding these
interactions is essential for developing new drugs as well as defining critical quality attributes of biopharmaceuticals.

Please explain why your abstract is innovative for mass spectrometry?

Hyphenation of liquid-phase affinity separation with native MS detection permitting the direct establishment of structure-
function relationships

Co-authors:

Elena Dominguez Vega, Leiden University Medical Center

153
TOP-DOWN MASS SPECTROMETRY: A VERY PROMISING TOOL TO
FOLLOW BIOPHARMACEUTICALS AND THEIR BIOTRANSFORMATION
PRODUCTS IN PLASMA
Abstract ID: 188

Presenting author: Jonathan Dhenin, DMPK, Sanofi, Chilly-Mazarin, France, Mass Spectrometry for Biology Unit,
Institut Pasteur, Université de Paris, CNRS USR2000, Paris, France

Introduction

Bottom-up proteomics is one of the methods of choice to evaluate pharmacokinetics of biologics. However, it is not the
most suitable for a complete structural characterization of all circulating forms and there is therefore a need for
alternative strategies. Top-down and middle-down mass spectrometry approaches are now commonly used for the
analysis of pure biologics and have recently been applied to in vivo plasma samples. However, a direct detection at the
molecular level of all circulating forms is often hampered by their large and diverse molecular weights, and their low
abundances in complex matrices. Complementary approaches combining various digestion enzymes and multiple mass
spectrometry strategies are proposed here to better characterize all potential proteoforms arising from the
biotransformation of biologics of interest in biological samples.

Methods

A commercial monoclonal antibody was spiked in mouse plasma across a wide concentration range (1-100 µg/mL),
enriched by immunocapture and deglycosylated. The mAbs were analysed intact or after a limited digestion (with IdeS or
GingisKHAN enzymes) into 25, 50 or 100 kDa subunits by both classical and low-flow LC coupled to a high field Orbitrap
mass spectrometer. MS files were processed with Genedata Expressionist® software for the detection and identification
of proteins. Calibration curves were generated by plotting XIC peak areas against concentrations. Targeted MS/MS was
also applied to fully characterize all proteoforms, including post-translational modifications (PTMs).

Preliminary data (results)

A top-down approach was developed for the analysis of an intact mAb and its potential biotransformation products in
plasma. To this aim, important instrument parameters (source energy, IRM pressure, maximum ion injection time …)
were optimized in the range of the molecular weights (from 25 to 150 kDa) expected after biotransformation. Using our
optimized workflows, truncated forms of the mAb could be identified and followed across the whole concentration range.
A limit of detection (LOD) and quantification (LOQ) of respectively 2.5 and 5 µg/mL could be achieved for the native mAb
with a classical UHPLC set-up using 10 µL of plasma. LOD and LOQ were found only slightly higher for the main
truncated proteoform with a molecular weight of 125 kDa. This proteoform nicely mimics the well-known in vivo clipping
process of mAbs. This highlights the unique added-value of a top-down mass spectrometry strategy, which allows not
only the identification but also the quantification of large biotransformation products present at low abundance. To further
improve our workflow, and in particular the sensitivity of the method at very low concentrations, a low-flow LC was
applied. After re-optimization of our experimental conditions, plasma samples obtained from mice treated with the
commercial mAb were analyzed to extend the applicability of top-down strategies to in vivo samples and showcase the
complementarity of this top-down approach to the more usual bottom-up one.

Please explain why your abstract is innovative for mass spectrometry?

Improved top-down Orbitrap mass spectrometry workflow for the analysis of biotherapeutics and their biotransformation
products in plasma

Co-authors:

Valérie Lafont, DMPK, Sanofi, Chilly-Mazarin, France

Norbert Zombori, DMPK, Sanofi, Chilly-Mazarin, France
Mathieu Dupré, DMPK, Sanofi, Chilly-Mazarin, France
Olivier Pasquier, DMPK, Sanofi, Chilly-Mazarin, France
Alain Krick, DMPK, Sanofi, Chilly-Mazarin, France
Christine Mauriac, DMPK, Sanofi, Chilly-Mazarin, France
Julia Chamot-Rooke, Mass Spectrometry for Biology Unit, Institut Pasteur, Université de Paris, CNRS USR2000, Paris,
France

154
Bioanalytical workflow combining intact mAb immuno-enrichment and top-down LC-MS analysis

155
CHARACTERISATION OF OLIGONUCLEOTIDES BY TANDEM MS AND IMS
IN NEGATIVE ION ESI
Abstract ID: 522

Presenting author: Fabien Hannauer, Chemistry, Faculty of Engineering and Physical Sciences, University of
Southampton, UK

Introduction

This project explores the use of tandem MS to identify therapeutic oligonucleotides and their impurities. The product ion
spectra are used to identify sequences and subsequently to determine mechanisms and rules of fragmentation. Where
different modifications are present in the structure, their effect on the fragmentation process and mechanisms will be
explored and de novo sequencing will be determined.

More and more therapeutic oligonucleotides are used in the pharmaceutical industry as a new generation of drugs. They
can help target diseases that are not accessible using current drugs on the market such as cancer and rare
diseases e.g. Duchenne muscular dystrophy or spinal muscular atrophy. More recently they have been used in vaccines
for COVID-19.

Methods

Different sequences of 21-mer oligonucleotides were analysed using negative ion electrospray ionisation using a Waters
Synapt G2-Si. Two sample introduction techniques were used, direct infusion and RP-IP-UHPLC with Waters Acquity
UHPLC and a C18 BEH column. Different charge states were isolated within the mass spectrum and fragmented by CID
with and without the use of IMS.

Preliminary data (results)

When the sample is analysed by direct infusion, different charge states are obtained compared to RP-IP-UHPLC-MS.
Each charge state was isolated, and optimised CID-MS conditions were applied to acquire product ion spectra that were
then interpreted to obtain sequence information. Furthermore, the use of IMS, after fragmentation, allows the separation
of the different charges into simplified data that are easier to interpret. IMS also permits the detection of different
conformations of oligonucleotides with the same m/z.

The fragmentation of different oligonucleotide sequences will give fragment ions based on McLuckey nomenclature and
internal fragment ions. Furthermore, new fragments produced by rearrangement can be observed and used for de
novo sequencing. Those fragments are depending on the sequence and particularly on the neighbour of each
nucleobase.

Please explain why your abstract is innovative for mass spectrometry?

New fragments produced by rearrangement from oligonucleotides are obtained which can be used for de
novo sequencing.

Co-authors:

G. John Langley, Chemistry, Faculty of Engineering and Physical Sciences, University of Southampton, UK
Eugen Stulz, Chemistry, Faculty of Engineering and Physical Sciences, University of Southampton, UK
Andrew D. Ray, New Modalities Product Development, Pharmaceutical Technology & Development, Operations,
AstraZeneca, Macclesfield, UK
Rachelle Black, New Modalities Product Development, Pharmaceutical Technology & Development, Operations,
AstraZeneca, Macclesfield, UK
Stephen W. Holman, Chemical Development, Pharmaceutical Technology & Development, Operations, AstraZeneca,
Macclesfield, UK

156
 Tuesday 30 August 2022: 15:30 – 17:30
Session: AD-2 Forensic Sciences

MASS SPECTROMETRY IN SPORTS DRUG TESTING - ADVANCES AND

CHALLENGES
Abstract ID: 7

Presenting author: Mario Thevis, German Sport University Cologne / Institute of Biochemistry

Introduction

Analytical approaches in sports drug testing are continuously updated and expanded, exploiting new information on drug
metabolism and disposition in humans as well as innovations in sample preparation and analysis, and also novel
strategies focusing on marker-based test methods have been assessed, developed, and implemented. The resulting
improved detection capability and retrospectivity of sports drug testing approaches has considerably limited the formerly
available options of substances and methods of doping. In addition, however, and similar to the general population, elite
athletes are exposed to a complex set of environmental factors including chemicals, biological and physical stressors,
which constitute an exposome that is, unlike for the general population, subjected to specific scrutiny for athletes due to
applicable anti-doping regulations and routine doping controls.

Methods

Test methods in sports drug testing, relying largely on chromatographic-mass spectrometric methods, were optimized
and applied to newly identified challenges, including e.g. the detection and characterization of superior metabolic
products of prohibited as well as non-prohibited substances, aiming at enhancing the analytical data available for
decision-making processes in test result management. Additional information, resuling from controlled (microdosed)
elimination studies and simulations of contamination scenarios, complements the dataset of routine doping controls.

Preliminary data (results)

Drug elimination profiles are an important aspect, contributing to the interpretation of analytical test results and
supporting the assessment of drug exposure scenarios concerning their plausibility. By means of examples including
ingredients of cosmetics, food potentially contaminated with doping agents such as anti-estrogens, and new anabolic
agents (SARMs) contributing to continuously increasing numbers of adverse analytical findings, the particularly important
role of chromatographic-mass spectrometric analyses in doping controls is illustrated. Optimized test methods allow for
utmost retrospectivity and, at the same time, can offer critical information as to the time point of drug exposure and/or the
source of the target analyte in athletes’ doping control samples.

Please explain why your abstract is innovative for mass spectrometry?

New target analytes and elimination profile data generated by mass spectrometry support decision-making processes in
anti-doping

157
STEROID PROFILING IN BLOOD AS AN EFFICIENT APPROACH IN DOPING
CONTROL ANALYSES
Abstract ID: 62

Presenting author: Tobias Langer, Swiss Laboratory for Doping Analyses

Introduction

For decades, anabolic steroids have been the most misused class of substances for performance enhancement.
Traditionally, they have been detected in urine after derivatisation and analysis by GC-MS. However, urine can be
manipulated during the sample collection, steroids can be subject to microbial degradation and steroid esters, which are
used as pro-drugs, are not detectable is this fluid. Blood on the other hand is more difficult to tamper with, and allows for
the analysis of steroid esters as well. Profiling of endogenous steroids in blood was proposed for the detection of oral
and transdermal testosterone administration in men. Later, it was shown that steroid profiling in blood was an efficient
tool to support the detection of testosterone gel application in women.

Methods

In this work, we have further explored the potential of these methods by including the detection of exogenous steroids
within the same analysis. Detection of steroid esters is more challenging due to suboptimal ionisation efficiency. To solve
this issue, steroid esters were derivatised and the sample was analysed a second time with different chromatographic
and mass spectrometric conditions.

Preliminary data (results)

With this combined method, it was possible to determine the endogenous steroid profile in blood, but also to detect
exogenous steroids and steroid esters in the same aliquot. The method was validated in compliance with ISO 17025 and
the International Standard for Laboratories by the World Anti-Doping Agency.
To overcome difficulties related to blood sample collection and transport, dried blood spots (DBS) have been proposed
as an alternative matrix. The lower sample volume results in higher detection limits, making the analysis of endogenous
steroids not trivial. However, we were able to show that the same methodology as for steroid profiling in serum could be
applied for the detection of endogenous and exogenous steroids as well as their esters in DBS.

Please explain why your abstract is innovative for mass spectrometry?

Detection of endogenous and exogenous steroids as well as steroid esters within the same blood sample using only
common and simple instrumentation.

Co-authors:

Olivier Salamin, Swiss Laboratory for Doping Analyses

Raul Nicoli, Swiss Laboratory for Doping Analyses
Tiia Kuuranne, Swiss Laboratory for Doping Analyses
Alessandro Musenga, Swiss Laboratory for Doping Analyses

158
General approach for steroid profiling in blood samples

159
ELUCIDATION OF CHLORINATED TYROSINE ADDUCTS IN BLOOD
PLASMA AS SELECTIVE BIOMARKERS OF CHLORINE EXPOSURE
Abstract ID: 212

Presenting author: Mirjam de Bruin-Hoegée, van ‘t Hoff Institute for Molecular Sciences, Faculty of Science,
University of Amsterdam, P.O. Box 94157, 1090GD Amsterdam, The Netherlands, TNO Defence, Safety and
Security, Dep. CBRN Protection, Lange Kleiweg 137, 2288GJ Rijswijk, The Netherlands

Introduction

Chlorine is a widely available toxic industrial chemical that has been used as chemical warfare agent in World War I and
which is allegedly still used in military conflicts. For this purpose, analysis of the highly persistent biomarkers 3-
chlorotyrosine and 3,5-dichlorotyrosine in biomedical samples might be used for forensic verification purposes. An
important shortfall of these biomarkers, however, is the relatively high incidence of elevated levels of chlorinated tyrosine
residues in individuals with inflammatory diseases who have not been exposed to chlorine. Therefore, more
unambiguous biomarkers of chlorine exposure are necessary to distinguish between endogenous formation and
exogeneous exposure. The present study aims to develop a diagnostic tool for identifying site-specific chlorinated
peptides as a more unambiguous indicator of exogeneous chlorine exposure.

Methods

Human blood plasma was exposed in vitro to various chlorine concentrations and the plasma proteins were subsequently
digested by pronase, trypsin and pepsin. After sample preparation, the digests were analyzed by liquid chromatography
tandem mass spectrometry (LC-MS/MS) and liquid chromatography high resolution tandem mass spectrometry (LC-
HRMS/MS).

Preliminary data (results)

It was found that the 3-chlorotyrosine and 3,5-dichlorotyrosine levels in blood plasma after chlorine exposure were higher
than levels reported in literature for diseased individuals, where 3,5-dichlorotyrosine is the most specific marker for
exogenous chlorine exposure. Additionally, in total 50 peptides were identified that could be used as more unambiguous
biomarkers for chlorine exposure. Chlorination of the peptides TYETTLEK, YKPGQTVK, YQQKPGQAPR,
HYEGSTVPEK and YLYEIAR could already be detected at moderate chlorine exposure levels. In addition, the latter two
peptides were found to have dichlorinated fragments. Especially, YLYEIAR, with a distinct chlorination pattern in the MS
spectra, could potentially be used to differentiate exogeneous exposure from endogenous causes as other studies
reported that this particular part of human serum albumin is nitrated rather than chlorinated under physiological
conditions. In conclusion, trypsin digestion combined with high resolution MS analysis of chlorinated peptides could
constitute a valuable technique for the forensic verification of exogeneous chlorine exposure.

Please explain why your abstract is innovative for mass spectrometry?

• An LC-HRMS/MS method was developed for the analysis of chlorinated peptides.

• Site-specific chlorine-tyrosine biomarkers of exogeneous chlorine exposure were identified.

Co-authors:

Irene van Damme, van ‘t Hoff Institute for Molecular Sciences, Faculty of Science, University of Amsterdam, P.O. Box
94157, 1090GD Amsterdam, The Netherlands
Daan Noort, TNO Defence, Safety and Security, Dep. CBRN Protection, Lange Kleiweg 137, 2288GJ Rijswijk, The
Netherlands
Arian. van Asten, van ‘t Hoff Institute for Molecular Sciences, Faculty of Science, University of Amsterdam, P.O. Box
94157, 1090GD Amsterdam, The Netherlands, CLHC, Amsterdam Center for Forensic Science and Medicine, University
of Amsterdam, P.O. Box 94157, 1090GD Amsterdam, The Netherlands

160
LC-HRMS/MS analysis of digested proteins after in vitro-chlorine exposure.

161
TRAPPED ION MOBILITY MASS SPECTROMETRY FOR THE RAPID
SEPARATION AND IDENTIFICATION OF POSITIONAL ISOMERS IN
DESIGNER-DRUG MIXTURES
Abstract ID: 285

Presenting author: Hany Majeed, Division of Bioanalytical Chemistry, Amsterdam Institute of Molecular and Life
Sciences, Vrije Universiteit Amsterdam, De Boelelaan 1085, 1081 HV Amsterdam, the Netherlands, Centre for
Analytical Sciences Amsterdam (CASA), 1098 XH Amsterdam, The Netherlands

Introduction

New psychoactive substances (NPS) are structural derivatives of conventional illicit drugs, designed to circumvent the
law. Some NPS are positional isomers of the original drug and pose a challenge to identify confidently in forensic
casework, especially when isomer mixtures are encountered. Current methods of identification require a combination of
gas chromatography with MS and infrared spectroscopic detection, which can be laborious and time consuming. Trapped
ion mobility mass spectrometry (TIMS-MS) is a relatively new gas-phase technique that separates ions based on their
charge, size and shape followed by mass-selective detection, potentially providing isomer resolution. We developed a
new TIMS-MS method and dedicated data-deconvolution tool for the fast and highly selective analysis of NPS isomers
found in real cases.

Methods

TIMS-MS experiments were carried out using a hybrid TIMS-time-of-flight (TOF) mass spectrometer (Bruker Daltonics).
Solutions (20µM) in water and 1% v/v formic acid of various NPS were directly infused (3 μL/min) into the electrospray
ionization source (positive mode). TIMS parameters, such as tunnel-in pressure, ramp time and mobility window, were
optimized for isomer resolution. 18-crown-6 was occasionally added (20µM) to the drug solutions to improve TIMS
performance. Recorded mass spectra and extracted-ion-mobilograms were processed with an in-house developed script.
Based on TIMS-MS measurements of pure standards, a model was created to deconvolute mobilograms of isomer
mixtures.

Preliminary data (results)

All positional isomers of the various tested drugs showed significantly different ion mobilities and could therefore be
distinguished with TIMS-MS. Optimal TIMS resolution of closely related NPS was achieved by selecting an increased
tunnel-in pressure of 3.0 mbar, a ramp time between 300 and 600 ms, and a narrow mobility window. For all individual
cathinones a bimodal mobility distribution was obtained and their mixture led to a distribution of unresolved peaks. We
hypothesize that the secondary bands are caused by gas-phase intermolecular interactions between two cathinone
derivatives to form a larger complex with reduced ion mobility. Addition of an interacting neutral crown ether to the
sample resulted in a single peak for each cathinone isomer, and greatly facilitated the analysis of mixture mobilograms.
The deconvolution algorithm appeared very useful and effective in resolving overlapping IM signals of positional isomers
obtained by TIMS-MS of mixtures of multiple isomers. This way the NPS composition of unknown samples could be
established also revealing the relative contribution of individual isomers to the total signal. The developed workflow was
successfully used for the unambiguous identification of NPS isomers in confiscated forensic case samples in less than 5
min per sample with minimal sample pretreatment.

This workflow represents a competitive new method for the screening and identification of positional isomers in NPS
mixtures in seized drug materials. Overall, this opens new perspectives for the study of isomer mixtures of small
molecules for forensic analysis.

Please explain why your abstract is innovative for mass spectrometry?

- Optimized TIMS method provides positional isomer resolution of NPS.

- Neutral crown-ethers prevent gas-phase intermolecular interactions, reducing spectral complexity

- Deconvolution algorithm aids automated isomer identification in NPS mixtures.

Co-authors:

Tijmen Bos, Division of Bioanalytical Chemistry, Amsterdam Institute of Molecular and Life Sciences, Vrije Universiteit
Amsterdam, De Boelelaan 1085, 1081 HV Amsterdam, the Netherlands, Centre for Analytical Sciences Amsterdam

162
(CASA), 1098 XH Amsterdam, The Netherlands
Ruben Kranenburg, Forensic Laboratory, Unit Amsterdam, Dutch National Police, Kabelweg 25, 1014 BA Amsterdam,
The Netherlands, Van’t Hoff Institute for Molecular Sciences, University of Amsterdam, P.O. Box 94157, 1090 GD
Amsterdam, The Netherlands
Arian van Asten , Centre for Analytical Sciences Amsterdam (CASA), 1098 XH Amsterdam, The Netherlands, Van’t Hoff
Institute for Molecular Sciences, University of Amsterdam, P.O. Box 94157, 1090 GD Amsterdam, The Netherlands , Co
van Ledden Hulsebosch Center (CLHC), Amsterdam Center for Forensic Science and Medicine, P.O. Box 94157, 1090
GD Amsterdam, The Netherlands
Govert Somsen, Division of Bioanalytical Chemistry, Amsterdam Institute of Molecular and Life Sciences, Vrije
Universiteit Amsterdam, De Boelelaan 1085, 1081 HV Amsterdam, the Netherlands, Centre for Analytical Sciences
Amsterdam (CASA), 1098 XH Amsterdam, The Netherlands
Isabelle Kohler, Division of Bioanalytical Chemistry, Amsterdam Institute of Molecular and Life Sciences, Vrije
Universiteit Amsterdam, De Boelelaan 1085, 1081 HV Amsterdam, the Netherlands, Centre for Analytical Sciences
Amsterdam (CASA), 1098 XH Amsterdam, The Netherlands

163
UNTARGETED MASS SPECTROMETRY METHODS FOR THE PROFILING
OF RICINUS COMMUNIS AND ABRUS PRECATORIUS SEED EXTRACTS –
A FORENSIC APPROACH
Abstract ID: 544

Presenting author: Lisa Scharrenbroch, Federal Criminal Police Office, Forensic Institute, 65193 Wiesbaden,
Germany, Technical University of Darmstadt, Department of Chemistry, 64287 Darmstadt, Germany

Introduction

Biological toxins are a focus of concern by public health and law enforcement on national and international levels due to
the increasing threat of their deliberate release in a bioterrorist attack. Among these biological toxins the plant toxins
ricin, which is present in seeds of the castor bean plant (Ricinus communis) and abrin, which is present in seeds of the
jequirity bean plant (Abrus precatorius) are of particular interest owing to their worldwide availability, ease of preparation,
high toxicity, and the lack of medical countermeasures. Over the last decade, they attracted much attention in the context
of threatening letters containing unknown `white powders´ and terrorist threat scenarios, culminating in the prevention of
a bioterrorist attack using ricin in Cologne, Germany, in 2018.

Methods

To raise the evidential value of forensic investigation in this field, analytical methods for the characterization of seed
extracts prepared from castor beans and jequirity beans have been developed. Known procedures were applied for
sample preparation and analytical methods were based on liquid chromatography coupled to high-resolution mass
spectrometry (LC-HRMS). Separation was achieved on a C8 analytical column using gradient elution to optimize the
sensitivity of the system. An LTQ Orbitrap XL mass spectrometer fitted with an electrospray (ESI) source and operated in
positive and negative ion mode at a scan range between 150-2000 m/z was used for detection.

Preliminary data (results)

The developed method was successfully applied to differentiate between seed extracts prepared from Ricinus
communis and Abrus precatorius. MSn-experiments were performed in both positive and negative ion mode to identify
specific compounds within the seed extracts and confirm distinct structural elements. Various characteristic lipid species,
including triglycerides, sterols and phosphatidylcholines as well as low molecular weight metabolites such as alkaloids
were identified with the help of obtained chromatograms and MS n-spectra. In addition, the influence of different sample
preparation techniques on the resulting chromatographic peak profile was studied. Obtained chromatograms
demonstrated that the developed method can not only provide plant specific information but also variations in the peak
profile between extracts of the same plant were observed which depended on the solvent used for extraction. This
suggests that molecular forensics can help linking a biotoxin found at a crime scene to a specific plant species and
preparation method by identifying unique, identifiable patterns characteristic for the agent. Obtained results were
converted into a database which allows the search of specific monoisotopic masses of identified substances. Such
databases can readily be used for forensic investigations to specify the composition and classification of an unknown
“white powder” suspected to be a preparation of a toxic plant extract.

Please explain why your abstract is innovative for mass spectrometry?

The developed method for the untargeted profiling of seed extracts provides additional information for the identification of
toxic plant extract compositions and promises forensic intelligence in terms of sample comparison.

Co-authors:

Dieter Kirsch, Federal Criminal Police Office, Forensic Institute, 65193 Wiesbaden, Germany
Thomas Schäfer, Federal Criminal Police Office, Forensic Institute, 65193 Wiesbaden, Germany
Björn Ahrens, Federal Criminal Police Office, Forensic Institute, 65193 Wiesbaden, Germany
Frederik Lermyte, Technical University of Darmstadt, Department of Chemistry, 64287 Darmstadt, Germany

164
VALIDATION OF A LC-ESI/MS/MS METHOD FOR SIMULTANEOUS
MEASUREMENT OF EIGHTEEN CANNABINOIDS IN PLANT MATERIALS OF
HEMP
Abstract ID: 14

Presenting author: Liguo Song, Western Illinois University

Introduction

Hemp is well known for its characteristic to produce a unique class of compounds, i.e., cannabinoids. Recently, the 2018
Farm Bill excluded hemp from the statutory definition of cannabis, if its total concentration of Δ 9-THC and Δ9-THCA is
0.3% or less.

For quantification of cannabinoids, recently published methods often used either LC-UV or LC-MS. Because LC-UV can
meet the required LOQ and corresponding instruments are widely available, it has been commonly accepted by
commercial suppliers. However, LC-UV is not an analytical method that provides definitive identification, whereas LC-MS
is. Therefore, more LC-MS methods have been published recently. However, most validated LC-MS methods only
focused on the identification and quantification of twelve or less abundant cannabinoids. Other cannabinoids have been
rarely identified and quantified.

Methods

Calibration solutions were prepared in methanol from 0.01 to 12.5µg/mL for the eighteen cannabinoids and contained
0.5µg/mL isotopically labelled internal standards. Plant materials of hemp were grinded, powdered, and combined with
methanol containing 75µg/mL ACBD in a 25mg/mL hemp/methanol mixture. After ultrasonication, centrifugation and
filtration, the extract was serially diluted with methanol as a solution at 25µg/mL sample containing 0.075µg/mL ACBD
and 0.5µg/mL isotopically labelled internal standards. The 25µg/mL extract was finally analyzed by LC-ESI/MS/MS using
an Agilent 1260 Infinity II LC coupled with an Agilent 6545 Q-TOF.

Preliminary data (results)

A thoroughly systematic optimization of LC separation of eighteen cannabinoids resulted in baseline resolution (≥1.5) of
all cannabinoids, including the seven structure isomers of Δ9-THC (Figure 1&2). In the literature, bassline resolution of
CBG/CBD and Δ9-THC/Δ8-THC has never been reported.

Method validation was performed according to the ISO 17025 guidelines. All cannabinoids achieved a linear calibration
range from 0.01 to 12.5µg/mL. Quality control samples of individual cannabinoid were prepared at 0.01, 0.25 and
6.25µg/mL and analyzed in triplicate in each day, consecutively in three separate days, with acceptable precision and
accuracy.

Published recovery/accuracy experiments were limited by the unavailability of cannabinoid-free matrix and the high cost
of cannabinoid standards, leading to extremely complicated experiments and inability to perform recovery/accuracy
experiments with any real samples. These problems have been solved once for all by spiking an appropriate internal
standard, i.e., ACBD, into all samples. Our assessment in triplicate showed that recovery/accuracy for all the analyzed
samples met the requirements by the ISO 17025 guidelines.

Five samples of hemp flower, one sample of hemp cigarette and two samples of Δ 8-THC fortified hemp flower were
analyzed in triplicates with acceptable precision. While the recent Δ 8-THC craze concerned chemists, our results clearly
showed that Delta8 Asteroids, a Δ8-THC fortified hemp flower,contained 5.2% Δ9-THC (including Δ9-THCA). While three
other samples contained Δ9-THC slightly lower than 0.3%, four other samples contained Δ9-THC slightly higher than
0.3%.

Please explain why your abstract is innovative for mass spectrometry?

To our best knowledge, this is the first report able to simultaneously measure eighteen cannabinoids and realistically
perform recovery/accuracy experiments with every sample.

Co-authors:

165
Gabrielle Valenzuela, Western Illinois University
Shelby Carlson, Western Illinois University

Figure 1: Chemical structure of eighteen cannabinoids plus ACBD

Figure 2. LC-ESI/TOFMS chromatogram of eighteen cannabinoids plus ACBD

166
Session: IM-7 Instrumentation development: Mass Analyzers

COMBINING ULTRAVIOLET PHOTODISSOCIATION AND 2-DIMENSIONAL

MASS SPECTROMETRY
Abstract ID: 481

Presenting author: Peter O'Connor, University of Warwick

Introduction

Two-dimensional mass spectrometry (2DMS) is an MS/MS technique, pioneered on FTICR mass spectrometers, which
allows fragmentation of a complex mixture of precursor ions while retaining the information of which fragment comes
from which precursor. 2DMS works by modulation of all ions spatially through a fragmentation zone, with modulation
frequencies programmed as a function of m/z. The Fourier transform can then extract those modulation frequencies,
linking each fragment to its precursor.

Key to 2DMS is setting up a defined fragmentation zone, which is readily done with ultraviolet photodissociation, and has
recently been done on 12T and 15T FTICR mass spectrometers with two different ICR cell geometries at the university of
Warwick. The instrumentation, results, and performance parameters will be presented.

Methods

A 193 nm Coherent Existar laser and a 213 nm Litron Nd:YAG frequency quintupled laser were used for ultraviolet
photodissociation on a 12T and a 15T FTICR mass spectrometer. In each case, the laser is far-field aligned with a red
photodiode laser beam, and then aligned through a BaF2 laser window and through the hollow cathode electron gun to
hit the trapped ions. The system uses a dichroic mirror to combine the IRMPD laser beam and the UVPD laser beam
which potentially allows combined IRMPD/UVPD experiments as well, although that approach has not been used to date.

Preliminary data (results)

A range of different samples has been analysed using ultraviolet photodissociation (UVPD) on the FTICR mass
spectrometer including vitamin D3 isomers, a mixture of agrichemical small molecules, tryptic peptides from a range of
different proteins (mABs, spike, scorpion venom proteins, and yeast whole cell lysate) and peptides including
phosphopeptides and glycopeptides, and we expect to have further data in the coming months to present. As UVPD
generates fragment-rich tandem mass spectra, the resulting 3D surfaces generated by a 2D mass spectrometry
experiment are exceptionally rich in peaks allowing for remarkable peak density, particularly in complex mixtures.

The instrumentation setup will be discussed, and the data shown for a range of sample types. Furthermore, the analytical
performance parameters of UVPD/2DMS will be studied. The radial profile of the fragmentation zone will be plotted out
as we’ve previously done with ECD and IRMPD fragmentation on both the Infinity cell and the dynamically harmonised
cell, which will allow us to profile the beam shape as it overlaps with the ion cloud. Since the dynamically harmonised cell
reportedly properly positions the ions in the center of the cell, we expect a sharper beam profile than previously observed
with IRMPD on the Infinity cell.

Overall, UVPD works very well with 2DMS on an FTICR mass spectrometer to the point that the experiment is close to
routine at the Warwick Ion Cyclotron Resonance Laboratory.

Please explain why your abstract is innovative for mass spectrometry?

Combining Ultraviolet Photodissociation and 2-Dimensional Mass Spectrometry

Co-authors:

Alina Theisen, Bruker Daltonics GmbH

Christopher Wootton, Bruker Daltonics GmbH
Bryan Marzullo, University of Warwick
Anisha Haris, University of Warwick
Mark Barrow, University of Warwick

167
A HIGH-PERFORMANCE QUADRUPOLE MASS FILTER WITH NOVEL
OPERATION PRINCIPLES
Abstract ID: 169

Presenting author: Emil Traykov, IPHC/CNRS

Introduction

Several innovative operation modes applied to Quadrupole Mass Filters (QMF) have been studied by extensive
simulations showing a significant improvement in performance compared to similarly sized QMF operating in standard
mode. The new operating modes are based on strongly mass-dependent grouped ion oscillations allowing controlled
node formations in both transverse planes of the QMF. Combining these node formations with physical apertures and
introducing regions of variable stability within the QMF allows increasing the mass resolution and/or the transmission
efficiency by multiple times (5 to 12times depending on various parameters). The exceptional results obtained in the
simulations led to the design and building of a prototype QMF with several unique properties allowing assessing and
studying the performance based on the introduced novel operation principles.

Methods

The novel operation modes and mass scanning have been studied by extensive simulations covering the dependence of
the performance on various parameters related to the ion beam, mechanical precision and alignment, and stability of
electronics. A dedicated experimental setup comprising the QMF prototype together with a surface-ionization ion source
and a complex detection system was built in 2020 and the first experiments were performed in 2021 successfully
confirming the formation of grouped oscillations and focusing nodes. An extended measurement campaign has started in
2022 aiming at reaching the highest mass resolution and studying various tuning parameters that influence the
performance.

Preliminary data (results)

The first measurements began in 2021 confirmed the applicability of the novel principles of operation. The latter include
formation of grouped oscillations of the ions in the QMF, the creation of focusing nodes in both transverse planes, and
the possibility to manipulate independently the positions of the latter as means to increase mass selectivity. Even without
applying yet the additional technique of using variable stability regions (VSR), a mass resolution of 7000 (FWHM) was
reached. With the VSR technique, the simulations show that at least twice higher mass resolution can be reached. The
upcoming measurements aiming at maximum performance will include also VSR. After reaching stable operation at
maximum resolution, the dependence of the QMF performance on various operation parameters will be studied. This
study should allow introducing simplifications to the QMF design and dedicated electronics, which should lead to
reduction of the overall cost and complexity of the QMF using the novel operation modes while preserving maximum
performance. The success of the study may allow introducing the novel modes of operation into existing and future QMF
devices with minor modifications to the mechanical design and the RF electronics.

Please explain why your abstract is innovative for mass spectrometry?

A single stage QMF based on novel operation modes is introduced. The new operation modes allow increasing
significantly the performance of QMF, both in mass resolution and in transmission efficiency.

168
Main principles of the novel operation modes - focusing and VSR.

Increase of resolution and efficiency compared to standard QMF operation.

169
IN SITU CHEMICAL EXPLORATION OF THE SKIN BARRIER EMPLOYING
RECENT ADVANCEMENTS IN CRYOGENIC GAS-CLUSTER SIMS WITH
HIGH MASS RESOLUTION
Abstract ID: 327

Presenting author: Alexander Pirkl, IONTOF GmbH

Introduction

The stratum corneum (SC) is the uppermost skin layer and provides the most substantial barrier against the permeation
of exogenous compounds. Understanding the native chemical composition of this barrier is crucial in order to study
fundamental skin biology and compound permeation into the skin, both desired (pharmaceutical, cosmetic) and
undesired (pollutants, pesticides). Numerous studies have identified and quantified the most significant lipid species in
the SC and have elucidated the presence of a lipid gradient. However, there are limitations regarding a detailed analysis
of distribution or chemical gradients present and these subtle changes are currently unable to be investigated in situ,
which is in part attributed to the limited mass resolving power even of state-of-the-art TOF-SIMS instruments.

Methods

Single layers were collected in vivo, using tape stripping, and high lateral and mass resolution image analysis was
carried out. Furthermore depth profiling experiments were performed on full thickness porcine skin samples, taken from
posterior ear tissue or human skin from donors undergoing cosmetic surgery.

Mass spectrometric depth profiling and imaging was performed on a Hybrid SIMS (IONTOF GmbH), a combined ToF-
SIMS/Orbitrap mass spectrometer in ultrahigh vacuum. Singly-charged Arn-clusters (20 keV, 1000<n<4000) served as
primary ions to generate secondary ions for the Orbitrap analyser (Q Exactive™ HF, Thermo-Fisher Scientific™) as well
as to sputter through the sample.

Preliminary data (results)

This work presents both, label-free and matrix-free depth profile analysis of ex vivo skin (porcine and human) and
surface imaging analysis of human SC single layers, collected in vivo, using cryogenic SIMS combined with high
resolution mass spectrometry.

Molecular depth profiling of ex vivo skin demonstrates the successful differentiation of the SC from the underlying
epidermal region and identified significant markers specific for each region, including fatty acid-related species localised
in the SC, and phospholipid-related species solely present in the underlying epidermis. In addition, intensity gradients are
identified for several chemical species within the SC region itself for the first time. The mass resolution of the data
enables the detection of species, and subtle changes in their intensity, impossible with traditional ToF-SIMS depth
profiling. Furthermore, our data shows that the above method can be used to follow the permeation of xenobiotics, here
an exogenous peptide highly used in anti-aging cosmetic products, into the skin.

Imaging experiments conducted on cells prepared by a tape stripping method show significant variation in lateral
distribution for both, endogenous and exogenous species. There is a clear difference observed between cholesterol
sulphate and several nitrogen-containing species. One species, attributed to the popular surfactant sodium lauryl
sulphate, shows an unusual localisation in only a few isolated cells within the layer.

Interestingly from a mechanistically point of view, comparison of mass spectra conducted from samples in frozen and
non-frozen state show substantial differences in relative ion intensities, with a number of new signals being detected
under cryogenic conditions.

Please explain why your abstract is innovative for mass spectrometry?

In situ molecular non-cryogenic and cryogenic mass spectrometric depth profiling for chemical exploration of skin and
skin permeation studies.

Co-authors:

Nichola Starr, University of Nottingham

Matthias Kleine-Boymann, IONTOF GmbH

170
Mike Bell, Walgreens Boots Alliance
Morgan R. Alexander, University of Nottingham
David J. Scurr, University of Nottingham

Schematic representation of the MS method used for skin analysis.

Typical mass spectrum obtained from a depth profile.

171
A METHOD FOR THE STRUCTURAL ANALYSIS AND TIME-RESOLVED
IMAGING OF BIOMACROMOLECULAR ASSEMBLIES IN MASS
SPECTROMETRY USING UV PHOTODISSOCIATION AND TIMEPIX
DETECTOR
Abstract ID: 28

Presenting author: Anjusha Mathew, Maastricht Multimodal Molecular Imaging (M4i) Institute, Division of
Imaging Mass Spectrometry, Maastricht University, 6229 ER Maastricht

Introduction

The retrieval of 3D structural features of biomacromolecular assemblies (MMAs) with mass spectrometry (MS) is of great
interest since the structure of these complexes strongly defines their functions and interactions with other molecular
species. Currently, MS based techniques (native MS, ion mobility MS, HDX-MS, XL-MS, tandem MS, etc.) only provide
limited 3D structural information. Complimentary methods such as Cryo-EM, X-ray diffraction and NMR have several
disadvantages. EM lacks molecular details, NMR requires relatively large amounts of purified sample, and X-ray
diffraction relies on extensive crystallization trials. We investigate an Orbitrap™-Time-of-flight (TOF) instrument that
utilizes MS combined with position-and time-sensitive Timepix (TPX) detectors and a UV laser to obtain both 3D
structural and molecular information at the same time on single molecules of MMAs.

Methods

The TPX detectors were characterized on two other MS instruments prior to their installation on the Orbitrap-TOF
instrument to evaluate their single-ion sensitivity and high mass detection capabilities. A dual microchannel plate (MCP)
stack-TPX quad detection assembly has been coupled to a nanoESI-orthogonal TOF MS (LCT) and a dual MCP stack-
scintillator-TPX3CAM assembly has been added to a MALDI-linear TOF MS (Ultraflex III). The TPX quad and TPX3
detection assemblies were coupled as axial and orthognal detectors and a 193 nm UV laser was integrated to the
custom-designed Q Exactive™ ultra-high mass range (UHMR) Orbitrap-linear/orthogonal-TOF MS platform (MMA-
imaging device).

Preliminary data (results)

HIGH MASS DETECTION: Initially, performance of different elements of the MMA imaging device was tested based on
the ability to analyze high molecular weight ions. The two TPX detection assemblies were coupled to the LCT and
Ultraflex MS, which allowed detection and ion imaging of multiply charged non-covalent protein complexes of molecular
weight in excess of 800 kDa and singly and doubly charged intact protein ions of mass to charge (m/z) ratio up to 970
kDa. The high mass transmission efficiency of the custom-designed TOF analyzer and associated ion optics was
investigated by coupling MMA-imaging device with ETP detectors and operating in three different modes. Cesium iodide
ions (m/z range:300-12,000 Da), stored in the quad trap (located between the Orbitrap MS and TOF analyzer), were sent
either to the Orbitrap analyzer, axial ETP detector or orthogonal TOF ETP detector.

SINGLE-ION ANALYSIS: This work targets the structural determination of MMAs by analyzing the fragment ions
produced from the single precursor MMA ion via UV photodissociation. Single frame TPX images collected on the LCT-
TPX quad assembly by spraying a protein mix that contains cytochrome c, BSA and β-amylase demonstrate the ability of
the TPX to unambiguously detect and image individual macromolecular ion events. We achieved single-ion sensitivity in
the MMA-imaging device through the isolation of the ion of interest with the quadrupole mass filter of the Q Exactive
UHMR instrument, and limiting the ion transmission by controlling RF voltage of the hexapole (placed between Orbitrap
MS and TOF analyzer).

Please explain why your abstract is innovative for mass spectrometry?

The innovative Orbitrap-TOF MS based instrument we developed here to obtain the molecular structure of MMAs is
expected to have a significant impact in structural/molecular biology.

Co-authors:

Gert B. Eijkel, Maastricht Multimodal Molecular Imaging (M4i) Institute, Division of Imaging Mass Spectrometry,
Maastricht University, 6229 ER Maastricht
Frans Giskes, Maastricht Multimodal Molecular Imaging (M4i) Institute, Division of Imaging Mass Spectrometry,
Maastricht University, 6229 ER Maastricht
Jean-François Greisch, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Centre for Biomolecular Research and

172
Utrecht Institute for Pharmaceutical Sciences, Utrecht University, Padualaan 8, 3584 CH Utrecht, Netherlands
Proteomics Center, Padualaan 8, 3584 CH Utrecht
Ian G. M. Anthony, Maastricht Multimodal Molecular Imaging (M4i) Institute, Division of Imaging Mass Spectrometry,
Maastricht University, 6229 ER Maastricht
Alexander Lekkas, Fasmatech Science and Technology, Demokritos NCSR, 15310 Agia Paraskevi
Jingming Long, Amsterdam Scientific Instruments (ASI), Science Park 106, 1098 XG Amsterdam
Kyle Fort, Thermo Fisher Scientific (Bremen), 28199 Bremen
Jord Prangsma, Amsterdam Scientific Instruments (ASI), Science Park 106, 1098 XG Amsterdam
Albert J. R. Heck, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Centre for Biomolecular Research and
Utrecht Institute for Pharmaceutical Sciences, Utrecht University, Padualaan 8, 3584 CH Utrecht, Netherlands
Proteomics Center, Padualaan 8, 3584 CH Utrecht
Dimitris Papanastasiou, Fasmatech Science and Technology, Demokritos NCSR, 15310 Agia Paraskevi
Alexander A. Makarov, Thermo Fisher Scientific (Bremen), 28199 Bremen
Shane R. Ellis, Molecular Horizons and School of Chemistry and Molecular Bioscience, University of Wollongong, NSW
2522
Ron M. A. Heeren, Maastricht Multimodal Molecular Imaging (M4i) Institute, Division of Imaging Mass Spectrometry,
Maastricht University, 6229 ER Maastricht

MALDI-MS spectra of IgM acquired on MCP and TPX3 channels.

173
Single TOF cycle-orthogonal TPX3 image acquired on MMA-imaging device (Concanavalin-A).

174
PROBING THE STABILITY OF THE Β-HAIRPIN STRUCTURE OF GB1P IN
THE GAS PHASE BY COUPLING MASS SPECTROMETRY AND
FLUORESCENCE SPECTROSCOPY
Abstract ID: 678

Presenting author: Lukas Raphael Benzenberg, ETH Zurich

Introduction

Whether GB1p, a small peptide forming a β-hairpin, unfolds after desolvation was the objective of previous research and
is still debated. The stability has never been thoroughly investigated by approaches that combine mass spectrometry to
fluorescence spectroscopy-based methodologies such as FRET, because the compact structure of β-sheets falls below
the working range of FRET. However, recently first evidence of transition-metal FRET (tmFRET) in the gas phase was
reported, providing a working range suitable for β-sheet structures. Thus, this work reports on the utilization of tmFRET
to probe the stability of the GB1p β-hairpin structure in the gas phase.

Methods

Wild-type GB1p was modified by attaching a cysteine residue at the C-terminus and a histidine-glycine-histidine motif at
the N-terminus. While the first provides a labeling site for a donor dye, carboxyrhodamine 6G maleimide, the latter non-
covalently binds copper ions acting as the quencher. Fluorescence lifetime measurements of the labeled peptide in the
gas phase were performed for every major charge state (z = 2+, …, 5+) in MS/MS mode. After adding copper chloride
(CuCl2) to the sample solution, similar lifetime measurements were carried out. Additional ion mobility spectra and
molecular dynamics simulations aided in obtaining further structural information.

Preliminary data (results)

Lifetimes of carboxyrhodamine 6G for every measured charge state without the addition of copper chloride did not differ
significantly from each other and matched the expected value of t = 6.5 ns. When copper ions (Cu 2+) were bound to the
di-histidine motif, a decrease of fluorescence lifetime was observed regardless of the charge state, proving the
occurrence of tmFRET. However, the reduction in fluorescence lifetime continuously decreased for higher charge states.
This is likely a result of Coulombic-driven unfolding of the β-hairpin, resulting in higher dye-copper distances. A stronger
quenching effect for lower charge states (2+, 3+) could inversely be interpreted as a retention of the solution-phase
native fold of the β-hairpin structure of GB1p after desolvation due to the close proximity of carboxyrhodamine 6G and
copper ions. Ion mobility data supported the trend observed for spectroscopic experiments by allocating lower collisional
cross sections (CCS) for lower charge states. Finally, molecular dynamics simulations assigned the folded β-hairpin
structure to the charge state z = 2+, consistent with retention of β-sheet structures in the gas phase for lower charge
states.

Please explain why your abstract is innovative for mass spectrometry?

This is the first application of tmFRET in a mass spectrometer to probe whether β-sheet structures retain their solution-
phase structure in the gas phase.

Co-authors:

Ri Wu, Department of Chemistry and Applied Biosciences, ETH Zurich, 8093 Zurich, Switzerland
Jonas Bastian Metternich, Department of Chemistry and Applied Biosciences, ETH Zurich, 8093 Zurich, Switzerland
Renato Zenobi, Department of Chemistry and Applied Biosciences, ETH Zurich, 8093 Zurich, Switzerland
Paul Katzberger, Department of Chemistry and Applied Biosciences, ETH Zurich, 8093 Zurich, Switzerland
Julian Harrison, Department of Chemistry and Applied Biosciences, ETH Zurich, 8093 Zurich, Switzerland
Sereina Riniker, Department of Chemistry and Applied Biosciences, ETH Zurich, 8093 Zurich, Switzerland

175
Structure of the labeled GB1p β-sheet structure.

176
SINGLE-PARTICLE MASS ANALYSIS APPLIED TO ANTIBODY−ANTIGEN
COMPLEXES, INTACT RIBOSOMES AND VIRUSES USING ORBITRAP-
BASED CHARGE DETECTION MASS SPECTROMETRY
Abstract ID: 464

Presenting author: Szu-Hsueh Lai, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for
Biomolecular Research and Utrecht Institute for Pharmaceutical Sciences, Utrecht University, 3584 CH Utrecht,
The Netherlands, Netherlands Proteomics Center, Padualaan 8, 3584 CH Utrecht, The Netherlands

Introduction

Standard methods for mass analysis by mass spectrometry measure ensembles of thousand to millions of ions. This
approach enables analysis of monodisperse recombinant proteins; however, co-occurring stoichiometries, sub-
complexes, and heterogeneous modifications pose challenges in studies of more heterogeneous protein assemblies. To
tackle the challenges posed by mass heterogeneity, single-particle mass analysis may come to the rescue. We recently
showed that with appropriate calibration, charge determination of single ion particles can be achieved on
OrbitrapTM mass analyzers. In contrast with the conventional native MS approach, this single-ion approach allows charge
detection mass spectrometry (CDMS) to circumvent the need for charge-state resolved mass spectra.

Methods

CDMS measurements were performed on an Orbitrap Q Exactive UHMRTM mass spectrometer (Thermo Fisher
Scientific, Bremen, Germany). Samples were buffer exchanged into the aqueous ammonium acetate solution and then
introduced into a gold-coated-borosilicate capillary for nanoelectrospray ionization. A resolution of 200,000 at 400 m/z
was set, recording 1 s transients. After the multiscan acquisition, an appropriate calibration factor was applied, which
correlates the measured intensities to the charges of individual single ions. According to the determined charge state, a
resulting formula mass = m/z × z − z was used to calculate the mass of each particle, separately.

Preliminary data (results)

The highlight of the CDMS data are listed below:

1.We examined samples of variable complexity, namely, IgG-RGY hexamers, heterogeneously glycosylated IgG-
RGY:sEGFR antibody−antigen complexes, and megadalton assemblies involved in complement activation to determine
(1) stoichiometries, (2) accurate masses, for extensively glycosylated species, and (3) assembly pathways of large
heterogeneous immune complexes.

2.We revealed the complex, multimeric interactions that occur between neutralizing IgG antibodies, the ACE2 receptor,
and the heavily glycosylated SARS-CoV-2 spike trimer. Our experiments show that antibodies targeting the S-trimer
typically prefer stoichiometries lower than the symmetry-predicted 3:1 binding. We determine that this behavior arises
from the interplay of steric hindrance and avidity effects. These substoichiometric complexes are fully effective at
blocking ACE2 binding despite containing free receptor binding sites.

3.We benchmarked the method for analyzing a relatively homogeneous human 40S ribosome (Hs40S). In addition, the
mass resolution limits of the method have been explored by focusing on the Hs40S particle when bound to the viral HCV
IRES RNA, in which the mass of the particles increases by 8% upon binding. Furthermore, we analyzed larger and more
complex ribosome particles from different origins, including the intact 80S ribosome from human cells and 70S
chloroplast ribosome from spinach. These samples are highly challenging for mass analysis using conventional native
MS.

4.Finally, we successfully measured the mass of an intact bacteriophage P22 (~20 MDa) by CDMS.

Please explain why your abstract is innovative for mass spectrometry?

We applied single-particle mass analysis to MDa macromolecular complexes, including complex glycoproteins,
ribosomes, and viruses using Orbitrap-based CDMS.

Co-authors:

177
Tobias P. Wörner, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research and
Utrecht Institute for Pharmaceutical Sciences, Utrecht University, 3584 CH Utrecht, The Netherlands, Netherlands
Proteomics Center, Padualaan 8, 3584 CH Utrecht, The Netherlands, Thermo Fisher Scientific (Bremen), Bremen,
Germany
Maurits A. den Boer, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research and
Utrecht Institute for Pharmaceutical Sciences, Utrecht University, 3584 CH Utrecht, The Netherlands, Netherlands
Proteomics Center, Padualaan 8, 3584 CH Utrecht, The Netherlands
Victor Yin, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research and Utrecht
Institute for Pharmaceutical Sciences, Utrecht University, 3584 CH Utrecht, The Netherlands, Netherlands Proteomics
Center, Padualaan 8, 3584 CH Utrecht, The Netherlands
Sem Tamara, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research and Utrecht
Institute for Pharmaceutical Sciences, Utrecht University, 3584 CH Utrecht, The Netherlands, Netherlands Proteomics
Center, Padualaan 8, 3584 CH Utrecht, The Netherlands
Xiaoguang Xue, Genmab, Uppsalalaan 15, 3584 CT Utrecht, The Netherlands
Muriel D. van Kampen, Genmab, Uppsalalaan 15, 3584 CT Utrecht, The Netherlands
Boris Bleijlevens, Genmab, Uppsalalaan 15, 3584 CT Utrecht, The Netherlands
Kyle Fort, Thermo Fisher Scientific (Bremen), Bremen, Germany
Alexander Makarov, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research and
Utrecht Institute for Pharmaceutical Sciences, Utrecht University, 3584 CH Utrecht, The Netherlands, Thermo Fisher
Scientific (Bremen), Bremen, Germany
Albert J.R. Heck, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research and
Utrecht Institute for Pharmaceutical Sciences, Utrecht University, 3584 CH Utrecht, The Netherlands, Netherlands
Proteomics Center, Padualaan 8, 3584 CH Utrecht, The Netherlands

178
Session: LS-6 MS in Structural Biology - Native MS, HDX-MS -Session A

NATIVE MASS SPECTROMETRY UNDER “CLOSE-TO-LIFE” CONDITIONS

Abstract ID: 785

Presenting author: Michal Sharon, Weizmann Institute

Introduction

A grand goal of structural biology is to understand the mechanics of the molecular machines that coordinate the functions
of the cell. This challenging task has traditionally been approached in a reductionist manner, in which cellular molecular
components are fractionated and purified before being studied individually. Recent years have witnessed a revolution in
the field of structural biology, stemming from the growing awareness to the importance of performing structural studies
under more physiologically relevant conditions, ones that retain the effects of the intracellular environment on protein
structure, stability, interactions, function, and dynamics. This notion has already given rise to cryo-electron tomography
techniques and in-cell nuclear magnetic resonance spectroscopy. Yet, conformational variability, heterogeneity, fast
dynamics, and asymmetric structures can be a challenge for these methods.

Methods

To propel the field of cellular structural biology forward complementary methods are required. One such approach, the
focus of my talk, is direct-MS that enables analysis under “close-to-life” conditions. Direct-MS is based on a conceptual
frameshift in the field by turning the inherent limited dynamic range of MS (wherein low-abundance proteins are masked
by highly abundant ones), typically considered a weakness, into an advantage. The method enables the biased detection
of a highly produced target protein, while disregarding the lower-abundant endogenous proteins. Thus, signal
suppression allows one to overcome the need for prior protein purification.

Preliminary data (results)

I will demonstrate that direct-MS acquisitions enable the immediate, high-resolution assessment of a wide range of
structural features: sequence variations, assembly states, folding conditions, PTMs, overall structure, stability and the
association of ligands of the generated proteins. The applicability of the method to analyzing drug uptake and target
engagement in human cells will be discussed as well as its development towards full organ analysis.

Please explain why your abstract is innovative for mass spectrometry?

Direct-MS enables structural studies under “close-to-life” conditions, preserving as much as possible the natural
environment and biological diversity, features that are often lost during biochemical purifications.

Co-authors:

Gili Ben-Nissan, Weizmann Institute

Flowchart of the direct-MS approach

179
UNRAVELLING THE MECHANISM OF ROTAVIRUS VIRAL FACTORY
FORMATION USING STRUCTURAL MASS SPECTROMETRY
Abstract ID: 47

Presenting author: Alice Colyer, Astbury Centre for Structural Molecular Biology, University of Leeds

Introduction

Rotavirus infection accounts for >200,000 deaths every year primarily in children under the age of 5 in low-income
countries. While vaccines and treatments to alleviate symptoms are available, improved therapeutics are needed to ease
disease burden. Rotavirus replication occurs in membraneless organelles, termed viral factories. It has been proposed
two key non-structural proteins, NSP2 (an RNA chaperone) and NSP5 (an intrinsically disordered protein [IDP]), along
with viral RNA, drives the formation of viral factories by liquid-liquid phase separation. However, the mechanism of this
process remains obscure. Here we have harnessed the power of structural mass spectrometry (MS) to afford new
insights into the structure of NSP5 and interactions between NSP2, NSP5 and RNA that are essential to form viral
factories.

Methods

Native MS, conducted on Synapt G1 HDMS (Waters) / Q-Exactive UHMR (Thermo) instruments, has determined the
oligomeric state of NSP2 and NSP5. Ion mobility MS experiments on the Synapt G1 HDMS have investigated the
stability of NSP2 oligomers in the absence/presence of RNA. Hydrogen-deuterium exchange MS (HDX-MS) performed
on an automated HDX robot coupled to an Acquity M-class LC system with HDX manager and G2-Si mass spectrometer
(Waters) have probed regions of disorder in NSP5 and potential NSP2 / NSP5 interaction sites. Furthermore, native MS
and cross-linking MS (XL-MS) experiments using BS3 and tag-transfer XLs have validated these interaction sites.

Preliminary data (results)

Native MS methods have been utilised to characterise both the oligomeric state of NSP2 and NSP5. This has revealed
NSP2 exists in a tetramer-octamer equilibrium whereas NSP5 exists as a decamer. Collision induced unfolding
experiments have revealed that the NSP2 octamer is more stable than the NSP2 tetramer. Additionally, RNA binding to
the NSP2 octamer has been observed using native MS. HDX-MS has revealed the N-terminal of NSP5 is least protected
from exchange and therefore is likely to be intrinsically disordered while the C-terminal of NSP5 is most protected from
exchange and so is likely to be more structured. This is supported by circular dichroism spectroscopy (CD), showing that
whilst intact NSP5 was mostly disordered, the isolated C-terminal peptide is helical. Studying the C-terminal region of
NSP5 by native mass spectrometry has also revealed a potential role in oligomerisation. This provides the first glimpses
into the structure of this IDP and the interactions mediating assembly. Native MS experiments have further demonstrated
binding of an isolated NSP2 binding region peptide from NSP5 on full length NSP2, with XL experiments supporting this
interaction. These preliminary insights into the structure of NSP2 and NSP5 have provided a basis to characterise the
interaction of these proteins. This is with the hope to provide novel antiviral targets for future therapeutics to combat
rotavirus infection, potentially by interfering with these key interactions that are vital for viral factory assembly.

Please explain why your abstract is innovative for mass spectrometry?

Combining structural MS methods will uncover functionally important conformations, conformational changes, and
protein interactions, revealing the precise viral factory formation mechanism.

Co-authors:

Alexander Borodavka, Department of Biochemistry, University of Cambridge

Frank Sobott, Astbury Centre for Structural Molecular Biology, University of Leeds
Antonio Calabrese, Astbury Centre for Structural Molecular Biology, University of Leeds

180
NATIVE TOP-DOWN ELECTRON CAPTURE DISSOCIATION MASS
SPECTROMETRY WITH ISOTOPE DEPLETION FOR STUDYING THE EARLY
STAGES OF OLIGOMER FORMATION IN ALPHA-SYNUCLEIN
PROTEOFORMS.
Abstract ID: 219

Presenting author: Kiani Jeacock, University of Edinburgh

Introduction

Parkinson’s disease (PD) is the second most common neurodegenerative disorder, and the main proteinous component
of Lewy body aggregates found in the brains of PD patients is alpha-synuclein (αSyn), a 140-residue intrinsically
disordered protein. Despite aggregates of αSyn contributing to neuronal cell death in PD, little is known about how the
disordered monomer of αSyn can undergo the significant conformational change which results in β-sheet rich
amyloidogenic structures capable of bundling together into fibrils.

Here, multiple mass spectrometry techniques have been used to elucidate the structural properties associated with
various proteoforms of αSyn, including variants responsible for early-onset forms of PD, and variants with the
constitutively observed post-translational modification N-terminal acetylation.

Methods

WT αSyn and five PD-associated mutants were produced recombinantly in-house. Co-expression with the yeast NatB
acetylase was used to ensure total N-terminal acetylation, resulting in a complete panel of proteoforms and their
acetylated counterparts.

Using a Synapt G2S (Waters), native ion mobility mass spectrometry (IMMS) experiments show that higher order
oligomers such as dimers can be detected, and collision induced unfolding (CIU) demonstrates global conformational
differences between variants.

Additionally, native top-down electron capture dissociation (ECD) on a 12T SolariX 2XR (Bruker Daltonics), in
combination with our unique isotope depletion method, provides further information on the conformational behaviour of
this protein.

Preliminary data (results)

IMMS and CIU highlight differences in the gas-phase stability and conformational space occupied by the panel of αSyn
variants, suggesting that in the gas-phase, monomeric αSyn possesses some form of inherent structure. These
techniques, in combination with fluorescence assays and other biophysical methods, also demonstrate the influence of
N-terminal acetylation on the conformational profiles of these mutants. Interestingly, the most recently identified and least
studied genetic mutant, A53E, exhibits the most structural differences in comparison to the WT protein.

We postulate that differences in ECD fragmentation between WT and A53E αSyn may indicate why these proteoforms
exhibit different monomeric conformations, which in turn could explain a difference in their aggregation propensity and
therefore disease-initiating capability. In addition, we combine this high resolution mass spectrometry approach with an
isotope depletion method developed in our lab. Isotope depletion increases signal to noise ratio and/or reduces the
spectral complexity of fragmentation data, enabling the monoisotopic peak of low abundant fragment ions to be
observed. This enables the accurate and confident assignment of fragments unique to higher order oligomers of αSyn to
be assigned, and structural information about these species to be inferred.

Using this technique, fragments unique to dimeric species were identified. These fragments, derived from an early-
aggregation oligomer of αSyn hold information about the conformation and binding interface of the dimer, which has yet
to be identified. This study could further understanding of how αSyn transitions from a mostly disordered monomer to a
highly ordered amyloid fibril.

Please explain why your abstract is innovative for mass spectrometry?

Isotope depletion combined with native top-down ECD enables detection of fragments unique to the αSyn dimer,
providing the first insights into its structural properties.

181
Co-authors:

Kelly Gallagher, Utrecht University, Netherlands Proteomics Centre

Alex Chappard, University of Edinburgh
C. Logan Mackay, University of Edinburgh
Tilo Kunath, University of Edinburgh
Mathew Horrocks, University of Edinburgh
David Clarke, University of Edinburgh

Domain architecture of αSyn, showing the familial PD-associated missense mutations.

Standard and isotope depleted ECD workflow for the αSyn dimer.

182
HYDROGEN-DEUTERIUM EXCHANGE NATIVE MASS SPECTROMETRY OF
G-QUADRUPLEX DNA
Abstract ID: 241

Presenting author: Eric Largy, Univ. Bordeaux, CNRS, INSERM, ARNA, UMR 5320, U1212, IECB

Introduction

G-quadruplexes (G4s) constitute a family of nucleic acids secondary structures characterized by the stacking of tetrads
of guanines linked by H-bonds, and coordination of monovalent cations such as potassium at the stacking interface. G4s
are believed to be involved in essential cellular processes, but are challenging to characterize by traditional methods due
to their important polymorphism.

Recently, we demonstrated that HDX coupled to native MS and IMS is amenable to the study of structured nucleic acid
in native conditions. We now applied this approach to the biophysical study of a panel of G4s. To do so, we developed
an in-house script to model DNA isotopic distributions and deconvolute bimodal distributions, allowing the systematic
characterization of the exchange regimes (EX2/EX1).

Methods

Oligonucleotides were annealed in 100 mM TMAA, 1 mM KCl in 90% deuterium. HDX experiments were carried out at
the intact level, hence streamlining the analysis compared to typical bottom-up protein analysis. The exchange was
triggered by mixing deuterated samples with a buffer of the same composition albeit protonated, in either a continuous-
flow setup (1s—5min), or manually (2min – hours) resulting in a D-to-H exchange monitored by a Thermo Exactive
Orbitrap or an Agilent 6560 IMS-Q-TOF. In the continuous flow setup, the reaction time is controlled by changing the size
of the tubing carrying the sample to the MS source.

Preliminary data (results)

The protection from exchange by H-bonding is strongly linked to the secondary structure of the oligonucleotides.
Protected sites can be grouped in clusters of relatively similar protection (e.g. base pairs exchange much faster than
internal tetrads). As a result, similar structures exchange at similar (but distinct) rates. A correlation exists between the
exchange rates of G4s and their unfolding free energy, which can be leveraged to highlight species with non-canonical
features providing additional protection.

Native MS allows separating conformers binding different numbers of cations, giving access to the exchange kinetics of
single conformers in a single experiment. The stoichiometry of specific cation binding can be straightforwardly
determined since only specific binding events alter exchange rates.

Going further, an automated approach was implemented to detect and deconvolute bimodal distributions of deuterated
oligonucleotides, taking into account their sequence and the experimental conditions. The more stable G4s exchange
exclusively through EX2, wherein monomal isotopic distributions are observed. Several conformers can be detected and
characterized, even if they have identical masses, provided that they have different exchange rates, which is not possible
by native MS alone.

Less-stable species have large EX1 contributions, characterized by bimodal and correlated populations. This reveals
their cooperative opening, whose extent and rate can be determined by non-linear fitting of the deconvoluted data. This
gives unique insights into G4 dynamics in solution, highlighting that G4 can unfold/refold in solution even when no
unfolded species were observed at equilibrium by other techniques. EX1 can be promoted by destabilizing the G4s.

Please explain why your abstract is innovative for mass spectrometry?

HDX/native MS allows the biophysical and structural characterization of G4 DNA conformers in a way that goes beyond
a simple snapshot of the average solution content.

Co-authors:

Valérie Gabelica, Univ. Bordeaux, CNRS, INSERM, ARNA, UMR 5320, U1212, IECB

183
Guanine tetrad, G4 structures, their exchange kinetics and protection levels.

184
Characterization of the exchange regimes of G4s.

185
UNRAVELLING THE STRUCTURAL COMPONENTS BEHIND ALGAE’S
HIGHLY EFFICIENT PHOTOSYNTHETIC MACHINES.
Abstract ID: 356

Presenting author: Aneika Leney, University of Birmingham

Introduction

The light harvesting complex within microalgae, termed the phycobilisome, is the largest protein complexes known,
reaching up to 18 MDa in size. It functions to transmit broad wavelengths of light towards the photosystems for
photosynthesis and does so with remarkable efficiency. As such, phycobilisomes are highly sought for applications within
the solar power industry and could provide an innovative solution to convert sunlight to chemical energy in light-poor
geographical areas. But, how do they work? By combining native mass spectrometry, top down, bottom-up and cross-
linking mass spectrometry, we show how mass spectrometry is driving forward our ability to understand algae’s
fascinating photosynthetic machines.

Methods

Microalgae from a variety of algae species were grown under a variety of conditions and their phycobilisomes extracted
using sucrose gradient ultracentrifugation. The phycobilisomes were further dissociated into their phycobiliprotein sub-
complexes and their structures analysed through a combination of native mass spectrometry, top down, bottom-up mass
spectrometry. Mass spectrometry experiments were performed on either a QExactive HF mass spectrometer (Thermo
Fisher Scientific) or an Orbitrap Eclipse mass spectrometer (Thermo Fisher Scientific) coupled to a nanoESI source.

Preliminary data (results)

The mechanism by which phycobilisomes convert light energy to chemical energy is unknown. Moreover, photosynthetic
efficiency changes in different light, allowing the algae to rapidly adapt to their surrounding environment. Here, we
exposed algae to differing light intensities and colours and monitored the phycobilisomes structural properties. Using a
combination of native mass spectrometry, bottom-up, and top down mass spectrometry, we are able to decipher
differences in the protein components as well as their post-translational modification status that correlates with their
different photosynthetic efficiencies. Overall, the data highlights the critical components needed for efficient algae
photosynthesis and the power of using hybrid mass spectrometry approaches in structural biology.

Please explain why your abstract is innovative for mass spectrometry?

We present an innovative application of native mass spectrometry to study photosynthetic machines.

Co-authors:

Jeddidiah Bellamy-Carter, University of Birmingham

Jaspreet Sound, University of Birmingham

186
INVESTIGATING THE BINDING MODES OF A CONFORMATION-SELECTIVE
LIGAND TO PHARMACEUTICALLY RELEVANT IMMUNOPHILINS USING
NATIVE IM-MS AND CIU EXPERIMENTS
Abstract ID: 131

Presenting author: Silvana Smilla Zurmühl, Clemens-Schöpf-Institute, Department of Chemistry, Technical

University of Darmstadt

Introduction

Developing selective drug molecules is an important challenge, especially if structurally similar proteins are present in
vivo which have opposite biological effects to the target. One such example is provided by the immunophilins FKBP51
and FKBP52. In 2015 Gaali et al. developed a ligand that preferentially binds FKBP51 over other homologues through
conformational selection of a transient conformation unique to FKBP51.
In order to observe dynamic changes in proteins it is necessary to study them in an environment closely resembling their
state in solution. This is the idea behind native mass spectrometry, which has been used to investigate the solution-
relevant behaviour of proteins. When coupled with an ion mobility cell, information on their size as well as their mass can
be gathered.

Methods

We used a Waters SYNAPT XS HDMS instrument in positive ion mode for our measurements. Needles produced in-
house were used to introduce the samples in the nano-ESI source. The collision cross section measurements were
performed in nitrogen. The samples were kept in an aqueous solution of ammonium acetate to provide a native-like
environment for the proteins.

Preliminary data (results)

Our spectra reflected the different binding affinities of the FKBP51-selective ligand to other FK506-binding proteins. We
employed travelling wave ion mobility – mass spectrometry under native-like conditions to study the effect of ligand
binding on the collision cross section of the drug target FKBP51 as well as two additional non-target FK506-binding
proteins. Furthermore, we performed collision-induced dissociation and unfolding experiments which provided
information on the strength of the interactions between the ligand and proteins and the stability of the folded structure,
respectively.

As part of our data processing workflow, we have implemented a MATLAB script that speeds up and optimises the
calibration procedure to convert arrival time distributions into collision cross sections.

By varying the collision energy on the SYNAPT XS we were able to assess the stability of the ligand binding. At the same
time, we could monitor conformational changes caused by higher collision energies and ligand loss as shifts in the arrival
time distribution.

Please explain why your abstract is innovative for mass spectrometry?

The combination of native ion mobility – mass spectrometry and collision-induced unfolding provides insight into the
different binding modes of structurally homologous proteins to the same ligand.

Co-authors:

Thomas Nehls, Clemens-Schöpf-Institute, Department of Chemistry, Technical University of Darmstadt

Frederik Lermyte, Clemens-Schöpf-Institute, Department of Chemistry, Technical University of Darmstadt

187
Session: LS-7 Cellular Signaling Processes and MS in Systems Biology

DECRYPTING PROTEIN MODIFICATIONS AND DRUG ACTIONS BY

CHEMICAL PROTEOMICS
Abstract ID: 801

Presenting author: Bernhard Kuster, Technical University of Munich

Introduction

Most drugs act on proteins, are proteins themselves, lead to the production or degradation of proteins or otherwise use
the protein machinery of a cell to exert their therapeutic effects. Drugs frequently have more than one target leading to
desired or undesired polypharmacology. Given the molecular and organizational complexity of cellular systems, it is
important to characterize drug effects on a proteome-wide scale in order to understand their targets and mechanism(s) of
action (MoA). Therefore, chemical proteomics has become an important tool in drug discovery and chemical biology.
However, little information is available about drug action at the level of post-translational modifications (PTMs) of
proteins. This is surprising as many drugs work by modulating the activity of enzymes that regulate PTMs such as
kinases.

Methods

Here, we present a quantitative chemical proteomic approach termed decryptM able to assess target and pathway
engagement as well as the MoA of diverse cancer drugs in cells by measuring dose- (and time-) resolved modulation of
PTMs by drugs on a proteomic scale. Cells are treated with increasing doses of a drug and each proteome is encoded by
stable isotopes (TMT). Tryptic peptides bearing PTMs are enriched by immunoprecipitation (acetylation, ubiquitinylation)
or immobilized metal affinity chromatography (IMAC; phosphorylation). All PTM proteomes are combined and analysed
by LC-MS/MS. Dose-response characteristics are derived from the intensities of the TMT reporter ions.

Preliminary data (results)

Data collected for 34 drugs representing six drug classes in 15 human cell lines demonstrates that the approach is widely
applicable. The body of data represents millions of quantitative drug assays including thousands of drug- regulated p-
peptides and hundreds of acetylated and ubiquitinylated peptidess. Most PTMs are not regulated by most drugs and this
information is also highly valuable for understanding drug action in cells.

Because only few examples can be highlighted here, the decryptM data has been incorporated into ProteomicsDB where
data can be interactively explored. A drug-centric view enables users to view dose-response curves, annotations and
cross-references to UniProt and PhosphositePlus. The data can be filtered and searched for EC50 values (half maximal
effective concentration), curve quality, sequence motif and others.

DecryptM data is more informative than the conventional approach of measuring PTMs in replicates at one drug
concentration. This is because the former yields EC50 values, effect sizes and additional curve fit parameters instead of
only an effect size backed by a measure of statistical significance. The ability to assign potency to each regulated p-
peptide has important consequences for the interpretation of drug perturbation experiments.

Future integration of decryptM data and proteomic data collected from tumor patients in molecular tumor boards should
lead to a better understanding of tumor biology and improved patient stratification for therapy.

Please explain why your abstract is innovative for mass spectrometry?

Multiplexed stable isotope labeling enables proteome and PTM-wide dose- and time-resolved drug response
measurements in vitro and in vivo.

Co-authors:

Terrific TUM Team, Technical University of Munich

188
A CDK-MEDIATED PHOSPHORYLATION SWITCH OF DISORDERED
PROTEIN CONDENSATION
Abstract ID: 723

Presenting author: Maarten Altelaar, Utrecht University

Introduction

Cell cycle transitions arise from collective changes in protein phosphorylation states triggered by cyclin-dependent
kinases (CDKs), but conceptual and mechanistic explanations for the abrupt cellular reorganisation that occurs upon
mitotic entry are lacking. Specific interactions between distinct CDK-cyclin complexes and sequence motifs encoded in
substrates might result in highly ordered phosphorylation, while bistability in the mitotic CDK1 control network can trigger
switch-like phosphorylation. Yet the dynamics of mitotic phosphorylation has not been demonstrated in vivo, and the
roles of most cell cycle-regulated phosphorylations are unclear. Here, we show evidence that phosphorylation of
intrinsically disordered proteins (IDPs) by CDKs triggers switch-like mitotic cellular reorganisation by controlling liquid-
liquid phase separation (LLPS).

Methods

We collected single embryos at 15-minute intervals while recording cell divisions. Phosphopeptides from each embryo
were enriched using a Fe(III)-NTAand analyzed by nanoLC-MS. For the replicating or mitotic egg extracts, replication
was initiated by adding purified sperm chromatin to interphase egg extracts, while mitosis was triggered by adding
recombinant cyclin B. We also used egg extracts arrested at meiotic metaphase II. To determine high time-resolution
dynamics of individual phosphosites, we analysed 64 phosphosites in single embryos every 180-seconds using targeted
phosphoproteomics. We validated our results by analytical modeling and in vivo protein condensation experiments.

Preliminary data (results)

We performed quantitative phosphoproteomics in vivo, generating a dynamic map of protein phosphorylation from an
unfertilised egg to a 16-cell embryo. This resulted in the identification of 4583 high-confidence phosphosites on 1843
proteins. Hierarchical clustering of 1032 statistically significant changing sitesrevealed four distinct groups that reflect cell
cycle-regulated behaviour (Fig1). Interestingly, one of these clusters containedphosphosites with a clear oscillating
signature with upregulation preceding each cell division and coordinated phosphorylation of multiple members of protein
complexes, suggesting a common mechanism of regulation.We assigned in vivo embryo phosphosites to different cell
cycle stages by comparing with phosphorylation patterns of replicating or mitotic egg extracts, linking above mentioned
phosphosites to processes enriched in mitosis.

Next, we analyzed dynamics of 64 phosphosites sites from diverse protein complexes in single embryos every 180-
seconds using targeted phosphoproteomics. This revealed parallel and abrupt upregulation of all phosphosites preceding
each cell division, indicating switch-like phosphorylation of diverse protein complexes at mitotic onset (Fig2).

Bioinformatics analysis revealed that most cell cycle-regulated phosphosites occurred in CDK consensus motifs and
located to intrinsically disordered regions. Substrates of CDKs showed significantly more disorder than phosphoproteins
in general, a principle conserved from yeast to humans, while around half were components of membraneless organelles
(MLOs), whose assembly is thought to rely on LLPS. Analytical modelling of such disordered proteins predicts
modulation of LLPS by CDK-mediated phosphorylation, which was confirmed by biophysical and biochemical analysis of
a model IDP, Ki-67,suggesting a mechanism for CDK-mediated mitotic cellular reorganisation.

Please explain why your abstract is innovative for mass spectrometry?

Single embryo phosphoproteomics at high time resolution, reveals switch-like phosphorylation preceding each cell
divisionon diverse proteins with a high level of disorder and localization to MLOs.

Co-authors:

Juan Manuel Valverde, Utrecht University

Geronimo Dubra, IGMM, University of Montpellier
Henk van den Toorn, Utrecht University
Guido van Mierlo, Radboud University Nijmegen
Michiel Vermeulen, Radboud University Nijmegen
Albert Heck, Utrecht University
Carlos Elena-Real, CBS, University of Montpellier
189
Aurélie Fournet, CBS, University of Montpellier
Emile Al Ghoul, IGH, University of Montpellier
Dhanvantri Chahar, IGMM, University of Montpellier
Austin Haider, University of Denver
Matteo Paloni, CBS, University of Montpellier
Angelos Constantinou, IGH, University of Montpellier
Alessandro Barducci, CBS, University of Montpellier
Kingshuk Ghosh, University of Denver
Nathalie Sibille, CBS, University of Montpellier
Pau Bernado, CBS, University of Montpellier
Puck Knipscheer, Oncode Institute, Hubrecht Institute
Liliana Krasinska, IGMM, University of Montpellier
Daniel Fisher, IGMM, University of Montpellier

Hierarchical clustering of ANOVA significant phosphosites, zoomhighlights dynamic patterns

190
Phosphosite plots. Dots represent biological replicates, lines depict cell divisions.

191
THE ENDOTHELIAL INFLAMMATORY REPERTOIRE: A MULTI-OMIC
DELINATION OF DISTINCT AND SYNERGETIC ENDOTHELIAL
INFLAMMATORY STATES
Abstract ID: 336

Presenting author: Stijn Groten, Department of Molecular Hematology, Sanquin Research

Introduction

Vascular endothelial cells (ECs) form a dynamic interface between blood and tissue mediating critical steps in
maintaining anti- and pro- coagulation and inflammation states. Endothelial dysfunction through deregulation of cytokine
stimuli underlies several vascular inflammatory disorders. However, the endothelial response to many cytokines is not
known impeding the design of intervention strategies at the crossroads between vascular inflammation and hemostasis.

Methods

To generate an overview of the molecular response of ECs to cytokines, we evaluated changes in the proteome of
BOECs (Blood Outgrowth Endothelial Cells) stimulated to a panel of 92 cytokines separately and in combination. We
then performed a time-resolved multi-omics analysis on BOECs exposed to TNFα and IFNγ integrating transcriptome,
whole (phospho-) proteome, and secretome.

Preliminary data (results)

Inflammatory cytokines TNFα and IFNγ induced the highest number of significant proteomic events and exhibited distinct
proteomic profiles. Combined TNFα and IFNγ stimulation revealed a synergistic proteomic response and decrease in
endothelial barrier function.
To unravel the molecular basis of this synergistic response we performed a multi-omics approach which allowed for
complete mapping of molecular endothelial inflammatory responses, highlighting distinct signatures and a drastically
increased response through TNFα and IFNγ co-stimulation on all omics levels. Especially IFNγ induced (innate) immune
mediating processes in ECs among which a strong upregulation of MHCI and MHCII histocompatibility complexes.
Signaling was regulated via two main axes, the NFKB and JAK/STAT signaling pathways. However, key proteins in both
pathways such as RELA (NFKB) and JAK3 (JAK/STAT) were synergistically upregulated after combined exposure
revealing intricate transcriptional and translational control underlying the observed inflammatory states.
Moreover, ECs secreted a select subset of cytokines per inflammatory stimuli, which were (synergistically) increased
through combined TNFα and IFNγ stimulation, including CCL5, CCL8, CXCL9 and IL6.

In conclusion, our integrated analysis reveals 1) an in-depth molecular mapping of endothelial inflammation on multiple
omics levels 2) the presence of distinct endothelial inflammatory states and 3) a synergistic endothelial response through
combined TNFα and IFNγ stimulation. This study therefore supports the emerging role of endothelial cells as active
players in the progression of vascular inflammation.

Please explain why your abstract is innovative for mass spectrometry?

An extensive data integration of multiple omics levels, translated into a utilizable biological resource in the emerging field
of vascular inflammation

Co-authors:

Eva Smit, Department of Molecular Hematology, Sanquin Research

Esmeé FJ Janssen, Department of Molecular Hematology, Sanquin Research
Bart L van den Eshof, Department of Molecular Hematology, Sanquin Research
Floris PJ van Alphen, Department of Molecular Hematology, Sanquin Research
Carmen van der Zwaan, Department of Molecular Hematology, Sanquin Research
Alexander B Meijer, Department of Molecular Hematology, Sanquin Research, Department of Biomolecular Mass
Spectrometry and Proteomics, Utrecht Institute for Pharmaceutical Sciences (UIPS)
Arie J Hoogendijk, Department of Molecular Hematology, Sanquin Research
Maartje van den Biggelaar, Department of Molecular Hematology, Sanquin Research

192
Understanding molecular mechanisms of endothelial inflammation through multi-omic integration

Interaction network of the endothelial inflammatory response

193
INSIGHTS INTO ALZHEIMER PATHOLOGY USING MULTIMODAL
CHEMICAL IMAGING
Abstract ID: 794

Presenting author: Jörg Hanrieder, University of Gothenburg, University College London

Introduction

It is of critical importance to our understanding of Alzheimer’s disease (AD) pathology, to determine how key pathological
factors including beta-amyloid (Aβ) plaque formation are interconnected and implicated in nerve cell death, clinical
symptoms and disease progression. Exactly how Aβ plaque formation begins and how the ongoing plaque deposition
proceeds and initiates subsequent neurotoxic mechanisms is not well understood.

The primary aim of our research is to elucidate the biochemical processes underlying early Aβ plaque formation in brain
tissue.

Methods

For this, we will use an array of advanced chemical imaging modalities including hyperspectral microscopy and mass
spectrometry imaging that allows to delineate vivo Aβ build up and deposition at cellular length scales.

Specifically we advanced the integration of conformation sensitive hyperspectral mass spectrometry with MSI modalities
to elucidate plaque morphology associated changes in Aβ signatures.

We further pioneered means for amyloid chronology based on imaging stable isotope labelling kinetics (iSILK). Here,
transgenic AD mice are labelled metabolically with stable isotopes to follow the fate of aggregating Aβ species from
before and throughout the earliest events of precipitating plaque pathology.

Preliminary data (results)

Studies in AD mouse models and human post mortem brain revealed that distinct plaque phenotypes observed in the
brain (ie.cored and diffuse) are associated with distinct peptide patterns. Here, diffuse plaques consist predominantly of
Aβx-42 and Aβ4-42, while mature, cored plaques show significant Aβ1-40 deposition upon cre formation. Moreover, by
comparing plaque morphotypes in between AD and pathological ageing, we show that in AD plaques show characteristic
deposition of N-terminally truncated, pyroglumtamted full length Aβ1-42 (Aβ3pE-42). This suggests that early plaques
deposit as diffuse plaques consiting of Aβ1-42 and that maturation into cored plaques occurs upon hydrophobic
functionilization into Aβ3pE-42 that in turn seeds deposition of Aβ1-40.

In addition to the neuropathological studies we set out to follow Aβ dynamics in vivo. Here, our iSILK approach allowed
to visualize Aβ aggregation dynamics within single plaques across different brain regions. We show that formation of
structurally distinct plaques is associated with differential Aβ peptide deposition. Specifically, Aβ1-42 is forming an initial
core-structure followed by radial outgrowth and late secretion and deposition of Aβ1-38. These data, for the first time,
describe a detailed picture of the earliest events of precipitating amyloid pathology at scales not previously possible.

The results from these studies bring considerable novel information about the deposition mechanism of Aβ and its toxic
interactions with the surrounding. This will open up for developing tailored strategies to affect AD pathology prior to any
neurodegenerative mechanisms as well as to develop new biomarkers for AD.

Please explain why your abstract is innovative for mass spectrometry?

We advance the integration of MS with other chemical imaging and metabolic labelling techniques to follow endogenous
pathological processes underlying neurodegeneration.

194
Figure 1. Amyloid pathology and iSILK for dynamic peptide imaging

195
FOLLOWING FGF SIGNALING DYNAMICS IN BREAST CANCER USING A
TARGETED KINOME ASSAY
Abstract ID: 210

Presenting author: Tim Veth, Utrecht University, Netherlands Proteomics Centre

Introduction

Human cells contain 4 fibroblast growth factor receptors (FGFR) that are activated via 18 different fibroblast growth
factors (FGFs). FGFs regulate distinct and often paradoxical cellular processes, such as the epithelial-mesenchymal
transition or mesenchymal- epithelial transition. Phosphorylation signaling by kinases is pivotal for transmitting these
signals. Importantly, FGF-FGFR signaling is implicated in many types of cancer. In breast cancer FGF2, 3, 4, 10, and 19
are all associated with a bad prognosis and identified as possible drivers of tumor growth. FGFR activation is furthermore
identified as a key pathway that conforms estrogen inhibitor resistance in ER+ breast cancer. There is however limited
insights into the biological pathways that these FGFs activate, which is key to understanding what role these FGFs play
in breast cancer.

Methods

To investigate differential signaling induced by FGF2, 3, 4, 10, and 19 we measured the activation state of 46 kinases at
multiple time points using a dedicated selected reaction monitoring assay (SRM). The SRM assay quantifies activating
phosphorylation in the activation loop of kinases, as a proxy for kinase activity. We subsequently use this acquired
longitudinal kinase activity data to create a dynamic mechanistic model of the signaling pathway using logic-based
ordinary differential equations

Preliminary data (results)

We observed MAPK-ERK, PI3K, and PLCγ pathway activation by the different FGFs. Especially FGF2 and 4 show high
MAPK-ERK activation, which led to a substantial increase in cell proliferation. Contrarily to MAPK-ERK, PI3K activity is
downregulated by all FGFs, including novel associated kinases of the MARK family. The PI3K pathway is classically
associated with epithelial-to-mesenchymal transition but we observe proof of a more complicated regulation mechanism.
We subsequently verified our biological model using logic based dynamic modeling. This highlighted differences in ERK
activation dynamics and differential activation of cell cycle regulation kinases. Supported by (phospho)proteomics data
we show that FGF2 uniquely results in Rap1 driven activation. Besides, we confirm different phospho-signatures of cell
cycle regulation proteins for all FGFs. This highlights the differences in signaling across the FGFs. Concluding, we
provide a comprehensive systems map of FGF signaling in breast cancer.

Please explain why your abstract is innovative for mass spectrometry?

A targeted kinase activity SRM assay with high coverage of the FGFR pathway is introduced to follow FGF induced
signaling, and use the data for dynamic modelling of signaling events.

Co-authors:

Donna Debets, Utrecht University, Netherlands Proteomics Centre

Albert J.R. Heck, Utrecht University, Netherlands Proteomics Centre
Maarten Altelaar, Utrecht University, Netherlands Proteomics Centre

196
Kinome activation after stimulation with different FGFs.

Mapping of phosphorylations of proteins involved in ERK activations.

197
SPATIALLY RESOLVED PHOSPHOPROTEOMICS REVEALS FIBROBLAST
GROWTH FACTOR RECEPTOR RECYCLING-DRIVEN REGULATION OF
AUTOPHAGY AND SURVIVAL
Abstract ID: 70

Presenting author: Joanne Watson, University of Manchester

Introduction

The subcellular localisation of proteins, and their relocalisation upon perturbation, is critical to their functional roles and
contribution to fundamental cellular processes. This is well demonstrated by the internalisation and endocytic trafficking
of Receptor Tyrosine Kinases (RTKs), whose endocytosis-dependent signalling drives cell proliferation and motility
during development and adult homeostasis, but is dysregulated in diseases, including cancer. The recruitment of RTK
signalling partners during endocytosis, specifically during recycling to the plasma membrane, is still unknown. We
developed a spatially resolved phosphoproteomics approach to investigate signalling events regulated by the RTK,
Fibroblast Growth Factor (FGF) Receptor 2b (FGFR2b), when localised to the recycling endosome following stimulation
with its recycling ligand, FGF10.

Methods

We developed a bespoke mass Spectrometry (MS)-based, spatially resolved phosphoproteomics approach to resolve
subcellular phosphorylation events proximal to internalised RTKs. Our approach was based on the APEX2 proximity
dependent biotinylation system. APEX2-tagged FGFR2b and two subcellular compartment markers – RAB11 (recycling
endosome) and GFP (cytosol/cellular background) – were expressed in HeLa cells. Biotinylation was induced in
untreated cells or cells treated with FGF10 for 40 minutes. Protein lysates were enriched for biotinylated proteins,
digested with trypsin, and then enriched for phosphopeptides. Biological findings from bioinformatics analysis of MS data
were validating with functional, immunoblot and immunofluorescence assays.

Preliminary data (results)

By tagging FGFR2b and the recycling endosome marker RAB11 with APEX2, we could resolve intracellular FGFR
signalling partners proximal to recycling endosome. This FGFR2b-RE signature was distinct from the global
phosphoproteome, proving that our spatially resolved phosphoproteomics method could be used to distinguish
subcellular phosphorylation events. Functional and network-based bioinformatics analysis of the proximally
phosphorylated proteins revealed recycling endosome-localised regulation of mTOR signalling, including core signalling
players (e.g.RPTOR) and downstream effectors (e.g. ULK1). We validated these results with targeted assays and
uncovered that when FGFR2b is stimulated by FGF10, mTOR-dependent signalling leads to autophagy suppression via
ULK1. This had long-term effects on cell behaviour, including an increase in cell proliferation. These results were
reproduced in a separate cell line T47D, which has endogenous expression of FGFR2b. Our results add to the growing
importance of RTK recycling in orchestrating cell fate, by demonstrating that FGFR2b directly effects mTOR signalling
and autophagy from the recycling endosome.

Please explain why your abstract is innovative for mass spectrometry?

We successfully integrated proximity-dependent biotinylation approaches and mass spectrometry-based

phosphoproteomics. Spatially resolved phosphoproteomics, combined with other systems biology approaches, provides
a powerful tool to study cellular signalling.

Co-authors:

Harriet Ferguson, University of Manchester

Rosie M Brady, University of Manchester
Jennifer Ferguson, University of Manchester
Paul Fullwood, University of Manchester
Hanyi Mo, University of Manchester
Katherine H Bexley, University of Manchester
David Knight, University of Manchester
Jean-Marc Schwartz, University of Manchester
Michael Smith, University of Manchester
Chiara Francavilla, University of Manchester

198
Session: IM-8 Separation & Hyphenation; chromatography,
electrophoresis – Session B, proteins

HIGH THROUGHPUT VENOMICS

Abstract ID: 405

Presenting author: Jeroen Kool, VU

Introduction

Up to 5.4 million people suffer from snakebite annually, resulting in 2,7 million envenomings, 138,000 fatalities and more
than 400,000 cases of permanent disabilities. In order to assist in the process of developing next generation antivenoms,
a solid and ever-increasing knowledge basis on the in-depth composition of venom toxins in medically relevant snake
venoms is required. Venomics comprises of a combination of analytical and -omics approaches to elucidate venom
composition. Venomics is able to accurately detect, identify and nowadays also quantify the majority of venom
components. However, current day venomics approaches remain labour intensive and time consuming. In this study we
describe a High-Throughput (HT) venomics approach that is capable of delivering a full and automated proteomic
analysis of a snake venom within two days.

Methods

Snake venom is submitted to nanofractionation analytics, which involves liquid chromatographic separation followed by
mass spectrometry with parallel high-resolution fractionation on a 384 well-plate. After vacuum-centrifugation of the well
plates to evaporate eluents, automated tryptic digestion of fractionated toxins is performed robotically. Next, the well
plates are placed in an autosampler from a nanoLC-MS/MS system. Then, all digests are analysed using a fast-analytical
gradient, resulting in 100 measurements per day. The medically important snakes Calloselasma rhodostoma, Echis
ocellatus, Naja pallida, Bothrops asper, Bungarus multicinctus, Crotalus atrox, Daboia russelii, Naja naja, Naja nigricollis,
Naja mossambica and Ophiophagus hannah, were analysed.

Preliminary data (results)

Data obtained from all wells containing the tryptic digests of the fractionated toxins is automatically processed and
subjected to Mascot database searching. From there using an in-house written script, all Mascot results are compiled
into a single Excel sheet containing the proteomics data, sorted by fractionation time (i.e. retention time of elution for
each toxin; all toxins have eluted over a series of subsequent wells during the high-resolution fractionation). Then,
another script plots these data for each identified toxin in so called Protein Score Chromatograms (PSCs; resulting in one
chromatogram for each toxin). For this, for each toxin identified protein scores are plotted on the y-axis versus retention
times of adjacent series of wells in which a toxin was fractionated on the x-axis. Additionally, in a similar manner,
Sequence Coverage Chromatograms (SCCs) can also be plotted if desired. This same script integrates the peaks in
these chromatograms for semi-quantitation purposes. A final script combines all the X and Y information generated from
the individual toxins and plots all PSCs in a single Excel file to facilitate plotting all the data in one overview. Accurate
masses of the toxins could be correlated with PSCs via in-parallel acquired MS data enabled by post-column splitting to
nanofractionation and MS. Our data suggest that high throughput venomics will be a highly valuable tool for increasing
the throughput by which we can define venom variation and should aid the future development of new snakebite
treatments.

Please explain why your abstract is innovative for mass spectrometry?

HT venomics identified the toxins present in venoms under study and plotted this data as Protein Score Chromatograms.
Accurate mass information of intact toxins allows investigation of proteoforms and PTMs.

199
High Throughput Venomics

200
FROM DOWNSCALING TO SINGLE-CELL PROTEOMIC: UNDERSTANDING
AND MINIMIZING THE DOWNSCALING EFFECT
Abstract ID: 266

Presenting author: Christopher Kune, Mass Spectrometry Laboratory - ULiège

Introduction

Cellular systems consist of a variety of cells with distinct molecular and functional properties based on their location and
timing. The characterization of proteome heterogeneity in these systems is a key to enhancing medical research and
precision medicine, which requires quantification of proteins at the single-cell level. Liquid chromatography mass
spectrometry (LC-MS) is a well-suited technique for proteomics analysis. However, the drop in performance is an
inherent effect of decreasing the starting sample material amount. This downscaling effect is strongly related to the
sensibility of MS instrument and the peptide loss during sample preparation. A comprehensive study of all contributions
to the sample downscaling effect is essential to properly optimize single cell proteomic methods and thus for minimizing
the performance drop.

Methods

Here we present results from downscaled experiments together with a software-assisted strategy to evaluate the sample
downscaling contribution to the performance drop for bottom-up proteomic analysis. In this approach, a sample condition
(e.g., sample preparation protocol, LC-MS method, MS instruments) is evaluated by monitoring performance when the
injected quantity of peptides is reduced. Each peptide signal intensity is monitored in function of the injected quantity in
LC-MS. The signal drop for each peptide can then be study independently in function of their intrinsic properties (e.g.,
mass, charge, hydrophobicity factor, acidic/basis residue ratio) to highlight the causes of peptide loss.

Preliminary data (results)

This approach was first used to evaluate and compare the performances of LC-MS methods (e.g., sensitivity, feature
detection) in a high throughput context on QExactive (Thermo) and timsTOF (Bruker) instruments. This study led to a
comprehensive comparison of the performances of these two MS for proteomics of low-amount to single-cell level
samples. The same workflow was then applied to evaluate specific protein interactions (binding) with the vial surface.
Peptides from a HeLa tryptic digest standard were chosen as peptide mix model for this study to avoid the contribution of
sample preparation on the analysis performance. This study was first conducted with Total Recovery glass vials from
Waters. The loss of peptide signal was assessed by a downscaling experiment on a set of vials containing peptides at
different concentration levels (from 180ng/µL to 10ng/µL). As expected, the total peptide signal decreases as the total
peptide concentration is reduced. However, the drop in peptide signal was not homogenous in regards to peptide
hydrophobicity factors. This observation has been related to preferential and significant peptide binding on the vial
surface. These interactions becoming non-negligible when peptide concentration is downscaled. Based on these results,
vials molded in different polymeric materials were tested with our downscaling approach. The results helped determine
the best candidate polymeric material for vials or other laboratory consumables, minimizing peptide-surface interactions,
for single cell proteomic analysis or low starting material experiments LC-MS analyses.

Please explain why your abstract is innovative for mass spectrometry?

The monitoring of peptide signal in downscaling experiments allow to push further the comprehension of the downscaling
and to drive the optimization of single cell proteomic protocols.

Co-authors:

Sylvia Tielens, Stem Cells - Molecular Regulation of Neurogenesis - ULiège

Fléron Maximilien, Mass Spectrometry Laboratory - ULiège
Dominique Baiwir, Mass Spectrometry Laboratory - ULiège
Laurent Nguyen, Stem Cells - Molecular Regulation of Neurogenesis - ULiège
Gauthier Eppe, Mass Spectrometry Laboratory - ULiège
Gabriel Mazzucchelli, Mass Spectrometry Laboratory - ULiège

201
ONLINE NATIVE CEX-IM(CIU)-MS APPROACHES TO DECIPHER THE
CONFORMATIONAL LANDSCAPE OF THERAPEUTIC MONOCLONAL
ANTIBODIES CHARGE VARIANTS
Abstract ID: 370

Presenting author: Guusje Van Schaick, Center for Proteomics and Metabolomics, Leiden University Medical
Center, Leiden, the Netherlands

Introduction

The efficacy and safety of therapeutic monoclonal antibodies (mAbs) strongly rely on their chemical composition and
higher order structure (HOS), which can be impacted by post-translational modifications (PTMs). Usually, analytical
methods are focused on the identification of PTMs without exploring their effect on HOS. Cation exchange
chromatography (CEX) coupled to native MS (nMS) has emerged as valuable asset to characterize mAb charge variants.
Additionally, ion mobility (IM) and collision induced unfolding (CIU) have proven their utility to characterize the
conformational landscape of proteins. Here, we report on the multimodal characterization of therapeutic mAbs using an
innovative on-line coupling of CEX-CIU to record the CIU fingerprint of specific CVs, and decipher the influence of PTMs
on the HOS of mAbs.

Methods

Different mAbs (trastuzumab; IgG1, eculizumab; IgG2/4, and pembrolizumab; IgG4) were analyzed either at the intact or
middle-level (after IdeS digestion, Genovis). A BioResolve SCX column (Waters) was used with mobile phases
composed of 50 mM ammonium acetate at pH 5.0 (A) and 160 mM ammonium acetate at pH 8.6 (B). The (IM)-nMS and
CIU experiments were performed using a Synapt G2 HDMS (Waters) instrument. The CIU voltages were ramped with
steps of 10V during the elution time of mAb charge variants. Complete CIU fingerprints were acquired after four
independent CEX-CIU runs. CIU data was analyzed with the CIUSuite2 software.

Preliminary data (results)

We have developed an innovative online CEX-CIU methodology to provide an in-depth characterization of mAb charge
variants in a fast, robust and straightforward manner. In the first dimension, mAb charge variants are separated
according to their isoelectric points (pIs). Then, CIU unfolding is achieved by increasing the collision voltage in the trap
cell before IM separation during the elution of the selected charge variants. The experimental set-up was adapted to
record online the CIU fingerprint of the different CEX-separated mAb charge variants at intact- or middle- levels. Using
this novel approach, CEX-CIU measurements were allowed obtaining a complete fingerprint of at least three different
charge variant populations. This strategy was next applied to assess unfolding modifications of mAbs upon thermal
stress, and after the enzymatic modification of the glycan moiety on the Fc domain. Multiplexing capabilities of our CEX-
CIU set-up were demonstrated to the simultaneous analysis a mixture of three different mAb isotype subclasses with
different pIs. Altogether, these results clearly highlight the complementarity of nondenaturing LC dimensions such as
CEX combined with CIU to achieve an enhanced characterization of mAb charge variants. This approach provides not
only information on charge variant identification through nMS mass measurement but also reveals the influence of
primary sequence variations on the global conformation and unfolding porcess of mAbs without jeopardizing IM and CIU
quality data.

Please explain why your abstract is innovative for mass spectrometry?

All-in-one online CEX-CIU strategy for in-depth characterization of mAb charge variants by connecting proteoforms and
conformational landscapes.

Co-authors:

Manfred Wuhrer, Center for Proteomics and Metabolomics, Leiden University Medical Center, Leiden, the Netherlands
Elena Dominguez-Vega , Center for Proteomics and Metabolomics, Leiden University Medical Center, Leiden, the
Netherlands
Oscar Hernandez-Alba , Laboratoire de Spectrométrie de Masse BioOrganique, IPHC UMR 7178 Université de
Strasbourg, CNRS, France, Infrastructure Nationale de Protéomique ProFI, FR2048 CNRS CEA, Strasbourg France
Sarah Cianférani , Laboratoire de Spectrométrie de Masse BioOrganique, IPHC UMR 7178 Université de Strasbourg,
CNRS, France, Infrastructure Nationale de Protéomique ProFI, FR2048 CNRS CEA, Strasbourg France

202
INVESTIGATION OF METAL-PROTEIN-INTERACTIONS USING A
COMPLEMENTARY ANALYSIS SETUP COMPRISING HPLC-ESI-TIMS-MS
AND HPLC-ICP-MS
Abstract ID: 258

Presenting author: Catharina Erbacher, University of Münster, Institute of Inorganic and Analytical Chemistry

Introduction

Metal species are involved in many important mechanisms in the human organism, for example as enzymatic cofactors
or for signaling pathways. Additionally, metals can act as xenobiotic species, which enter the human body via
contaminations or drugs. For biological processes inside the organism, different metals are distributed via a protein
network. The physiological effects based on the metal-protein-interactions can be used pharmaceutically. Thus, different
metal compounds are applied as metallodrugs. Since binding of the metal species to proteins is especially relevant for
the mechanism, potential side effects and toxicity of the drug, research on this topic is highly important. Therefore,
reliable determination of active species and potential binding sites inside the human organism are crucial for
understanding the exact mode of action.

Methods

To determine the active metal species and potential binding to proteins, a complementary analysis setup comprising
HPLC-ESI-MS and HPLC-ICP-MS is suitable. Enzymatic digestions are frequently used for the specification of binding
sites. Usually, the digestions are analyzed via a time-consuming HPLC method. A more time efficient approach is the
implementation of TIMS-MS. To further accelerate the analysis of metal-protein-interactions, automation of the enzymatic
digestion can be helpful. Therefore, columns with immobilized proteolytic enzyme can be used for an online digestion
directly prior to mass spectrometric detection.

Preliminary data (results)

TIMS-MS analyses result in a three-dimensional dataset consisting of mobility, mass-to-charge-ratio (m/z) and intensity.
The fragments of digested protein samples form trajectories when their mobility is plotted against their m/z. Metal-binding
fragments drop out of these trajectories due to their increased mass combined with a similar mobility compared to metal-
free fragments and are thereby easily identified. Using this method, binding sites of the pharmaceutical Auranofin to
human serum albumin (HSA) and hemoglobin, which are both proteins of significant relevance for the human organism,
were determined verifying the involvement of free cysteine residues in the metal binding process. Furthermore, it was
demonstrated that the free cysteine groups of HSA and hemoglobin also act as binding sites for different organomercury
compounds, like methylmercury cations. Additionally, the implementation of online digestions resulted in a major
improvement of the overall analysis time, since overnight incubations are avoided and digestion of protein samples can
be carried out within several minutes.

In summary, the combination of online protein digestion and ESI-TIMS-MS enables a fast and easy determination of
metal binding sites in proteins. This results in a powerful method, which can be used for screenings of different metal-
protein-incubations to better understand the toxicity and the exact mechanism of action of metallodrugs.

Please explain why your abstract is innovative for mass spectrometry?

Compared to classical analysis techniques and data evaluation procedures the presented method simplifies and
accelerates the analysis of metal-protein-interactions significantly.

Co-authors:

Philipp Strohmidel, University of Münster, Institute of Inorganic and Analytical Chemistry

Michael Sperling, University of Münster, Institute of Inorganic and Analytical Chemistry
Uwe Karst, University of Münster, Institute of Inorganic and Analytical Chemistry

203
ESI-TIMS-MS heatmap of a digested methylmercury-HSA-adduct.

204
MULTI-NANOPARTICLE WORKFLOW ENABLES DEEP PLASMA
PROTEOMICS AT SCALE, WITH ENHANCED PRECISION, AND DEPTHS OF
COVERAGE.
Abstract ID: 365

Presenting author: Shadi Ferdosi, Seer Inc

Introduction

To overcome limitation of deep plasma proteomics, we have developed a fast and scalable technology that employs
intricate protein-nano interactions. Introducing a nanoparticle (NP) into blood plasma leads to the formation of a
selective, specific, and reproducible protein corona (PC) at the nano-bio interface driven by the relationship between
protein-NP affinity, protein abundance and protein-protein interactions. We previously demonstrated that this process,
incorporated within the Proteograph™, offers superior performance in terms of depth, breadth, precision, and throughput
compared to conventional deep workflows. The ratio of plasma-to-nanoparticles determines the competition between
proteins for binding surface, which affect the PC composition and can be optimized to enhance and differentiate protein
selectivity. Here we investigate effects of different conditions on PC composition enabling enhanced performance of
Proteograph.

Methods

We have investigated compositional changes of protein coronas from 5 NPs with blood plasma at different ratios.
Samples were analyzed with timsTOF Pro mass spectrometry and UltiMate3000 Dionex LC system using 30min
diaPASEF runs. The data was processed using the DIA-NN analytical software (version 1.8) in library-free mode. We
evaluated depth, dynamic range, coverage, and precision of quantification at a wide range of concentrations for each NP.

Preliminary data (results)

By limiting the available binding surface of NPs and increasing the binding competition, we are able to identify 20 – 60%
more proteins on the surface of each NP, specifically 2 – 5x better coverage of inflammation and hormone signaling
proteins. Moreover, by increasing the competition the proteins are more reproducibly identified and quantified across the
replicates of the same NP. In addition, protein selectivity was enhanced, leading to improved coverage of plasma
proteome when using multiple physicochemically distinct NPs. In summary, NP panels with optimized workflow, capture
a large and diverse set of proteins and biological pathways based on their specific physicochemical makeup.

Please explain why your abstract is innovative for mass spectrometry?

Upstream of the mass spectrometry, by leveraging the specific physicochemical properties of nanoparticles, we can
reproducibly sample plasma proteins across the dynamic range, enabling unbiased, deep proteomics at scale.

Co-authors:

Tianyu Wang, Seer Inc

Moaraj Hasan, Seer Inc
Eltaher M Elgierari, Seer Inc
Xiaoyan Zhao, Seer Inc
Martin Goldberg, Seer Inc
Asim Siddiqui, Seer Inc
Serafim Batzoglou, Seer Inc
Omid C. Farokhzad, Seer Inc
Daniel Hornburg, Seer Inc

205
COMPARISON OF HYDROPHILIC INTERACTION CHROMATOGRAPHY
AND NATIVE SIZE-EXCLUSION CHROMATOGRAPHY-MASS
SPECTROMETRY FOR THE CHARACTERIZATION OF HEAVILY GLYCATED
PROTEINS
Abstract ID: 635

Presenting author: Ziran Zhai, Van’t Hoff Institute for Molecular Sciences, University of Amsterdam, Amsterdam,
The Netherlands., Centre for Analytical Sciences Amsterdam, Science Park 904, 1098 XH Amsterdam, The
Netherlands.

Introduction

Glycation results from the non-enzymatic covalent attachment of sugar to the basic residues of a protein. Many
biotechnological products undergo glycation during storage conditions or in their application (e.g., enzymes used in food
production). This post-translational modification plays a significant role in modifying the protein structure and functions of
proteins. Therefore, during biotechnological product development, it is important to investigate the degree of glycation to
which a protein may undergo. However, the characterization of glycated protein is challenging due to their heterogeneity,
deriving from the type, number, and sites in which sugars may react. Both denaturing and native LC-MS methods can be
used to characterize glycated products. Here we compare the performance of hydrophilic interaction chromatography
(HILIC-MS; denaturing method) to native size-exclusion chromatography-mass spectrometry (SEC-MS).

Methods

We developed a simple protocol to accelerate the process of glycation of reference proteins. This was done using the
model protein BSA and glycating it with glucose. A novel acrylamide-based monolithic stationary phase for HILIC was
prepared, which guarantees high selectivity and allows limited amounts (0.005%, v/v) of trifluoroacetic acid (TFA) in the
mobile phase concerning 0.1% TFA needed in silica-based materials. The glycation products were analyzed by HILIC-
MS (with a low concentration of TFA) and native low flow SEC-MS (using ammonium acetate mobile phases) comparing
structural information that could be obtained in the analysis of heavily glycated proteins.

Preliminary data (results)

Using HILIC, the different glycated forms are separated on the basis of the number of sugar units attached. This was
demonstrated by the analysis of the reference protein BSA with different degrees of glycation. In contrast, SEC partially
differentiated the glycated forms on the basis of the mass and conformational changes introduced by the glycation
process. Between the two separations, the variation of retention time in HILIC was considerable, whereas in SEC only
limited variation of retention time and peak shape were observed. As a result of this, in our initial experiments, we
observed more complex co-elution of glycated forms in SEC-MS. Therefore although the native conditions result in a
lower charge state distribution, SEC-MS is characterized by convoluted MS spectra. On the contrary, in HILIC-MS a wide
charge state distribution and adducts are observed. However, the selectivity for the different numbers of glycans and
sites occupied of HILIC-MS results in more clear glycated MS spectra. Future experiments, optimizing LC-MS conditions
will be done to compare the outcome of the analysis with the two methods.

Please explain why your abstract is innovative for mass spectrometry?

The optimization of SEC-MS and HILIC as well as allows the analysis of proteins with a high degree of glycation.

Co-authors:

Marta Passamonti, Van’t Hoff Institute for Molecular Sciences, University of Amsterdam, Amsterdam, The Netherlands.,
Centre for Analytical Sciences Amsterdam, Science Park 904, 1098 XH Amsterdam, The Netherlands.
Boudewijn Hollebrands, Unilever R&D Vlaardingen, Olivier van Noortlaan 120, Vlaardingen 3133 AT, The Netherlands.
Hans-Gerd Janssen, Unilever Food Innovation Center, Wageningen, The Netherlands.
Garry Corthals, Van’t Hoff Institute for Molecular Sciences, University of Amsterdam, Amsterdam, The Netherlands.,
Centre for Analytical Sciences Amsterdam, Science Park 904, 1098 XH Amsterdam, The Netherlands.
Andrea Gargano, Van’t Hoff Institute for Molecular Sciences, University of Amsterdam, Amsterdam, The Netherlands.,
Centre for Analytical Sciences Amsterdam, Science Park 904, 1098 XH Amsterdam, The Netherlands.

206
207
 Wednesday 31 August 2022: 11:00 – 13:00
Session: AD-3 Cultural Heritage and Conservation Science

MASS SPECTROMETRY FOR THE RESEARCH OF OBJECTS OF

CULTURAL HERITAGE
Abstract ID: 746

Presenting author: Klaas Jan Van Den Berg, Cultural Heritage Agency of the Netherlands, University of
Amsterdam

Introduction

Diagnostic research is required to understand the making of cultural objects, such as archeological finds, and works of
art. Analysis of the degree of ageing or deterioration, influence of environmental factors as well as the detection of
materials added in past conservation treatments will support decisions on restoration, storage and exhibition of the
objects.

Although in recent years non-invasive analytical techniques have become prominent in cultural heritage materials
research, mass spectrometry remains a crucial element of the heritage scientist’s toolbox.

Methods

direct infusion, Atmospheric Solids Analysis Probe (ASAP), (Pyrolysis)GC, and LC in combination with quadrupole, time-
of-flight and orbitrap mass spectrometry

Preliminary data (results)

The Cultural Heritage Laboratory in Amsterdam employs a range of mass spectrometric techniques, which include direct
infusion, Atmospheric Solids Analysis Probe (ASAP), (Pyrolysis)GC, and LC in combination with quadrupole, time-of-
flight and orbitrap mass spectrometers.

The specificity and sensitivity of mass spectrometry, whether or not hyphenated with chromatography, are unrivalled
when it comes to the detection of materials used in objects of cultural heritage. These materials may have been used as
such, in paints, coatings and adhesives and include natural lipids and waxes, resins, colorants, proteins and
polysaccharides in different stages of ageing, as well as modern synthetic materials.

In this keynote, analytical procedures used at the Cultural Heritage Laboratory for the study of cultural heritage objects
and fundamental studies on ageing and deterioration will be discussed. Examples will include research on binding media
in Egyptian cartonnage, dyes in archeological and more recent textiles, and painting materials used by Rembrandt, Van
Gogh, Jackson Pollock and Karel Appel.

Please explain why your abstract is innovative for mass spectrometry?

direct infusion, high resolution mass spectrometry

Co-authors:

Sanne Berbers, Cultural Heritage Agency of the Netherlands

Wim Genuit, Cultural Heritage Agency of the Netherlands
Art Ness Proano Gaibor, Cultural Heritage Agency of the Netherlands
Saskia Smulders, Cultural Heritage Agency of the Netherlands
Inez D. van der Werf, Cultural Heritage Agency of the Netherlands

208
NON-PROXIMATE SAMPLING OF INTACT NATIVE AMERICAN BASKETS
WITH IN-LINE DOPANT PERMEATION ATMOSPHERIC PRESSURE
PHOTOIONIZATION
Abstract ID: 58

Presenting author: G. Asher Newsome, Smithsonian Museum Conservation Institute

Introduction

Ambient sampling and ionization mass spectrometry methods offer particular opportunities for culturally sensitive object
analysis because they can theoretically be performed without cutting material from a whole or preparing the surface.
However, to physically accommodate an intact object of even moderate size, the sample surface must be placed at some
distance from the MS and the analyte material transported between the two. To identify dyes on intact Native American
baskets while avoiding signal loss from moving ions at atmospheric pressure, analyte was thermally desorbed into the
gas phase as a neutral and transported to an MS-proximate photoionization source.

Methods

Samples were positioned 1.5 m from an LTQ Orbitrap Velos MS (Thermo Fisher). Neutral nitrogen was supplied at up to
250 °C and 1.0 L/min from a custom-built probe angled at 45° and mounted from overhead. An Arduino trigger regulated
sample exposure timing. Thermally desorbed material entered a 198 °C, SilcoNert-coated, 2 m long stainless steel
sample line. Toluene or anisole dopant was added to the transport line with a reservoir-jacketed, in-line permeation tube.
Analyte was conducted past a krypton UV lamp (Syagen) for atmospheric pressure photoionization (APPI) to the
differentially-pumped inlet to the mass spectrometer.

Preliminary data (results)

The APPI-MS system was first tested without the thermal desorption probe. The 13 cm-long concentric permeation
device was constructed from 6.4 mm o.d., 4.7 mm i.d. polytetrafluoroethylene (PTFE) tubing sealed within a larger
housing by ferrule-compression fittings at each end. Liquid reagent was added through external ports to the 2.3 mL
reservoir between the PTFE and outer housing. SilcoNert-coated stainless steel tubing liners were inserted into each end
of the device to shield all PTFE surface area except for the central, 16.3 mm reservoir-jacketed length. Clove oil and
mothball headspace showed a three-fold increase in [M+H]+ signal from safrole and M+• signal from naphthalene,
respectively, when permeated toluene dopant was added to a 5-6 L/min room air inflow.

Orientation of the heated, neutral nitrogen flow probe relative to a sample and transport tube inlet was optimized
empirically for efficient sampling from non-proximate sample surfaces, adjusting gas flows, temperatures, and exposure
time to minimize sample damage. A Swagelok Ultra-Torr tee fitting was modified to accommodate a 12.7 mm diameter
APPI lamp, positioned orthogonal to the mass spectrometer and millimeters from the straight flow pathway. The
complete sampling system was tested by identifying indigo on a dyed fabric swatch positioned on a 66 cm sphere (Figure
1). The system is used next to study Native American black ash baskets, positioned carefully but remaining intact, from
the National Museum of the American Indian. The blue-colored baskets are analyzed to identify the organic colorants.

Please explain why your abstract is innovative for mass spectrometry?

This is the first application of a non-proximate thermal desorption sampling-APPI system, which includes a recently-
designed in-line permeation device that adds dopant to analyte transport tubing.

209
Non-proximate desorption/MS-proximate ionization system test before basket analysis.

210
MAPPING HISTORICAL PIGMENTS BY MALDI-MS IMAGING
Abstract ID: 268

Presenting author: Alba Alvarez-Martin, AXIS, NANOlab Center of Excellence, Dept. of Physics. University of
Antwerp, Science department, Rijksmuseum

Introduction

From the moment an artist completes an oil painting, spontaneous and unwanted chemical transformations can lead to
the gradual loss of brightness and intensity of the intended color tones. Therefore, to understand the original palette used
by an artist, a full characterization of all materials used in his paintings is required. Mass spectroscopy can identify the
organic components in the artwork, but its coupling with separation techniques such as chromatography results in the
loss of any spatial information of the pigment within the sample. Here we exploit for the first time MALDI-TOF MS
Imaging (MSI) to visualize the spatial distribution of an historical organic pigment (eosin or geranium lake) and its
associated degradation products without prior extraction, purification and separation of the sample.

Methods

Oil paint samples containing eosin lake and linseed oil were artificially aged using light in a temperature and relativity
humidity controlled environment. Different embedding protocols were tested to check the solubility of the organic pigment
in the embedding material. For MSI experiments, samples were sectioned with 5, 10 and 20 µm thickness and mounted
on conductive tape on ITO slides. Two MALDI matrices (DHB and 9-AA) were tested and applied using self-built
sublimation apparatus. MS images were acquired on a Rapiflex MALDI Tissuetyper (Bruker Daltonics). Data was
processed with SCiLS (SCiL Lab, Bremen Germany) and R software (package Cardinal).

Preliminary data (results)

Herein, we describe the MALDI-MSI instrumental parameters and optimised experimental procedures for the analysis of
oil paint cross sections containing eosin. This pigment was frequently employed by Vincent Van Gogh and contemporary
artists to create various shades of purple and pink in their paintings. A preliminary investigation already proved
successful in visualizing the degradation products and the parent (historical) pigment. The data collected in this study
allows us to answer the following research questions: (a) Is the lateral distribution of the degradation products of eosin
the same as that of the parent molecule or do these distributions show evidence of significant migration of the secondary
products throughout the paint after their formation? (b) Is the pattern of secondary products everywhere the same vs.
depth below the surface or is it significantly different at the light-exposed paint surface than in the bulk?.

Please explain why your abstract is innovative for mass spectrometry?

The versatility of MALDI-MSI is opening a new frontier in the cultural heritage field by allowing mapping of traces of
organic pigments and their degradation products in oil paint samples.

Co-authors:

Jusal Quanico, Centre for Proteomics, University of Antwerp

Geert Baggerman, Centre for Proteomics, University of Antwerp, Unit Environmental Risk and Health, VITO, Mol
Koen Janseens, AXIS, NANOlab Center of Excellence, Dept. of Physics. University of Antwerp, Science department,
Rijksmuseum

211
STUDY OF COLLAGEN CROSSLINKING AND ASSOCIATED
MODIFICATIONS IN BONES USING PROTEOMICS
Abstract ID: 287

Presenting author: Catherine Gilbert, CBMN, UMR CNRS 5248, Proteome Platform, University of Bordeaux, F-
33000 Bordeaux, France

Introduction

Collagen is the most abundant extra-cellular matrix protein in the body, and significantly contributes to the high
mechanical strength of many tissues. This is caused by the enzymatically controlled cross-linking of the collagen
network, which can be a key biomarker in many biological processes, such as cancer, gestation, and osteogenesis
imperfecta. We are interested in analysing collagen cross-links in the context of the palaeoproteomic analysis of forensic
and archaeological bones. In such samples, collagen is the most abundant protein, and proteomics studies are broadly
covering protein sequence variations and species identification. 1 Collagen is significantly more well preserved than other
proteins present in bones, and as such, there is a strong interest in how collagen cross-linking contributes to this high
level of preservation.

Methods

Currently, collagen cross-linking is studied through a combination of the proteomic-based characterisation of

modifications indicative of collagen cross-linking; fluorescence- and radioactivity-based quantification of the cross-links
themselves; and immunoassay-based purification of cross-linked peptides.2 We propose novel cross-link enrichment
strategies including size-based fractionation and fluorescence detection. In addition to this, we propose the use of
targeted MS3 fragmentation to characterise the structure of cross-linked peptides. In the case of ancient samples,
minimising and simplifying the sample preparation protocol, and the use of sensitive, high resolution instrumentation,
including nanoLC and Orbitrap mass spectrometry, the required sample size is significantly reduced.

Preliminary data (results)

Despite collagen’s major role in the extracellular matrix structure, only a few structural analyses of collagen and its cross-
linking based on mass spectrometry are available.3 There are significant challenges in the characterisation of cross-
linked collagen peptides with mass spectrometry, such as (i) the lack of bioinformatic support for the proteomic analysis
of cross-links between three peptides; (ii) the high heterogeneity of collagen cross-linked structures; and (iii) the low
concentration of cross-links present in the sample. We propose a novel analysis approach from cross-link enrichment
procedures to targetted mass spectrometry analysis strategies.

This work focuses on the structural analysis of collagen cross-linking both in fresh and ancient bones using an optimized
proteomic workflow. We focus on the mass spectrometric structural analysis of trivalent, mature cross-links (such as
pyridinolines); as these are more likely to persist in archaeological samples, in comparison to the chemically unstable,
divalent, immature cross-links (such as ketonorleucines). Using conventional proteomic methodologies, several post-
translational modifications that are integral to the mechanism of the formation cross-links in collagen, such as
hydroxylysine, glycosylation, and galactoglycosylation, have been detected in collagen 1α(i) and α(ii) in both fresh
ancient bone samples. This has allowed us to identify specific regions in the collagen sequence to target for subsequent
cross-link investigations.
1 Dallongeville, S. et al. Chem. Rev. 116, 2–79 (2016)

2 Eyre, D. R. et al. Methods 45, 65–74 (2008)

3 Terajima, M. et al. J. Biol. Chem. 289, 22636–22647 (2014)

Please explain why your abstract is innovative for mass spectrometry?

A proteomics-based approach towards the analysis of cross-links between three peptides using targetting MS3 structural
analysis.

Co-authors:

Francesca Galluzzi, CBMN, UMR CNRS 5248, Proteome Platform, University of Bordeaux, F-33000 Bordeaux, France
Katell Bathany, CBMN, UMR CNRS 5248, Proteome Platform, University of Bordeaux, F-33000 Bordeaux, France

212
Stephane Claverol, CBMN, UMR CNRS 5248, Proteome Platform, University of Bordeaux, F-33000 Bordeaux, France
Quentin Scanvion, Univ. Lille, CHU Lille, ULR 7367 - UTML&A - Forensic Taphonomy & Anatomy, F-59000 Lille, France
Benoit Bertrand, Univ. Lille, CHU Lille, ULR 7367 - UTML&A - Forensic Taphonomy & Anatomy, F-59000 Lille, France
Valery Hedouin, Univ. Lille, CHU Lille, ULR 7367 - UTML&A - Forensic Taphonomy & Anatomy, F-59000 Lille, France
Caroline Tokarski, CBMN, UMR CNRS 5248, Proteome Platform, University of Bordeaux, F-33000 Bordeaux, France

213
UNDERSTANDING THE ROLE OF PIGMENTS IN THE SICCATIVATION OF
OIL-BASED PAINTS BY MEANS OF SOFT CHEMICAL
DEPOLYMERIZATION AND ULTRA-HIGH RESOLUTION MASS
SPECTROMETRY
Abstract ID: 612

Presenting author: Marie Yammine, Miniaturization for Synthesis, Analysis & Proteomics, USR 3290, CNRS,
University of Lille., Lesaffre International, 137 rue Gabriel Péri

Introduction

Pigments and in particular metal-containing pigments play a central though yet-not-fully explored role in oil-based paints
siccativation, which is based lipid autoxidation radical chain reactions. Recent studies have highlighted how different
pigments contribute in an early aging of the triglycerides (TAGs) composing 99% of vegetable oils employed in oil-based
paint formulations by means of the well known palmitic/stearic (P/S) ratio, which has definitively proven to be an
unreliable means for both oil discrimination and oil degradation index. This study aims at understanding pigment-related
aging of siccative oils together with the characterization of specific degradation products by means of soft chemical
depolymerization and ultra-high resolution mass spectrometry.

Methods

Two different sets of model oil paint samples were prepared by mixing reference methyl esters (linolenate, linoleate and
oleate) or traditional siccative oils with appropriate pigments amounts and spread on glass slides. Samples were
transamidated overnight using N,N′-dimethyl-1,3-propanediamine (DMAPA) under magnetic stirring. After quenching with
formic acid, samples were extracted with distilled water and petroleum ether after centrifugation. A second fraction was
obtained extracting the samples with ethyl acetate. MS analyses were performed by direct infusion using nanoESI or
MALDI ionization in absorption mode on a 9.4 Tesla Bruker™ SolariX™ FT-ICR instrument.

Preliminary data (results)

Both the reference samples prepared with methyl esters and the model samples prepared with traditional siccative oils
were analyzed at different aging stages (from fresh to multiples days and weeks of aging) to monitor the appearance of
common degradation products (aldehydes, ketones, (di)carboxylic acids, …) and possible pigment-related structures. In
general, aged samples show a dramatic degrease in the relative abundance of (poly)unsaturated fatty acids, playing a
central role in the polymerization process. The soft depolymerization of aged samples, performed through
transamidation, also afforded oxidized derivatized fatty acids homologues and in the low m/z area volatile degradation
products were also detected together with small carboxylic acids ( from C6:0to C8:0). Common dicarboxylic acids, as
azelaic-, succinic- and pimelic acid, described in literature as degradation markers, were also detected. In particular,
comparing two paint samples containing Prussian blue and Titanium white respectively, we can notice that Prussian blue
dramatically increases the polymerization rate, showing a large decrease in the quantity of polyunsaturated fatty acids. In
the same way, large degradation products had appeared, contrarily to the sample containing Titanium white, an inert
pigment. The quantity of pigment in the oil mixture was also found to play a role in the extent of degradation, giving rise
to pigment-specific or metal-specific features in MS spectra. Results from these model paints will be compared with
those of obtained from a reference historical canvases set dating from the early XIXth century up to present-day and
statistically correlated according to their pigment identification.

Please explain why your abstract is innovative for mass spectrometry?

This study proposes a comprehensive characterization of the pigment-dependent polymeric network formation upon oil
siccativation through the identification of cross-linked structures thanks to ultra-high resolution mass spectrometry.

Co-authors:

Caterina Bordin, Université de Lille, CNRS, USR 3290 - MSAP laboratory - Miniaturization for Synthesis, Analysis &
Proteomics
Fabrice Bray, Université de Lille, CNRS, USR 3290 - MSAP laboratory - Miniaturization for Synthesis, Analysis &
Proteomics
Christian Rolando, Université de Lille, CNRS, USR 3290 - MSAP laboratory - Miniaturization for Synthesis, Analysis &
Proteomics, Shrieking Sixties, 1-3 Allée Lavoisier, 59650 Villeneuve d'Ascq

214
Derivatization protocol for both fresh and polymerized paint samples

Extraction protocol for transamidated paint samples

215
PALEOPROTEOMICS BY ULTRAHIGH RESOLUTION MALDI FT-ICR FOR
IDENTIFICATION AND DATING BONES FROM UPPER PLEISTOCENE
Abstract ID: 659

Presenting author: FABRIZI Isabelle, MSAP UAR 3290

Introduction

Archaeological bones contains molecules which allow to obtain information about the genus, specie and their evolution.
Currently, bones dating is performed by Carbon 14, however, this method requires large quantities of bone sample
(starting from 10 mg), and the dating limit does not exceed 50000 years.Another mass spectrometry method (ZooMS)
allows to estimate deamidation by a comparison of experimental isotopic distribution with theorical isotopic distribution.
This method requires samples quantities in the range of hundreds of milligrams. MALDI FT-ICR and LC-MS/MS orbitrap
with ultra-high resolution allow us to visualize isotopic massif in order to quantify accurately the deamidation rate and
collagen crosslinking, starting with only 1 mg of bone powder. This method allow to identify the species and give an idea
regarding bones conservation.

Methods

One milligram of bone powder from 210 bones from Holocene (11,000 years ago - present, France) up to Pleistocene
(120,000 years ago, Europe) were analyzed on 96 wells plate with a 0.45 µm pore PVDF membrane.The bone powder
was demineralized with a TFA solution, washed and digested with trypsin. Then, peptides were purified on 96 well C18
separation and analyzed by MALDI on a SolariX XR 9.4 T FT-ICR MS and by nanoESI nanoLC-MS/MS on a Q-Exactive
plus Orbitrap MS. Raw data were analyzed with Data Analysis 5.1 for MALDI data and Proteome Discoverer and
MassSpecStudio for LC-MS/MS data.

Preliminary data (results)

Our high throughput MALDI FTICR MS proteomics workflow allows to identify taxa from 1 mg of bones with a correlation
of 96% with morphologic assignments from paleontologist. Ancient bones degradation can give some information thanks
to the presence of post-translational modification, such as glutamine deamidation representing by a 0.019 Da mass shift.
We hypothesized that deamidation and collagen crosslinking rates allows to estimate bones dating. The deamidation
rates of the highly conserved peptides between species at mass m/z 1105 (GVQGPPGPAGPR, COL1α1) are
significantly different between bones from different site. For Holocene bones, the deamidation rate ranges between 10 -
22% and from 35 - 55% for Pleistocene bones. For modern bones, it varies between 4 - 9%. LC-MS/MS and MALDI_FT-
ICR show a good correlation (R2=0.91). This results indicate a difference of conservation. Then, we studied
dehydroxylysinorleucine crosslinks. The quantification of the crosslink of COL1α1 peptide GPAGPSGPAGKGR with the
COL1α2 peptide GAAGEPGKGER corresponding to a triply charged peptide at mass m/z 767.3562, showed a significant
difference between modern, Holocene and Late Pleistocene samples with 0.1%, 28% and 72% in average respectively.
The data mining of other dehydroxylysinorleucine and other crosslinks is under review. These data indicate that the
increasing of collagen crosslinking is correlated with bone fossilization. This new study offers a new promising approach
for studying bone aging.

Please explain why your abstract is innovative for mass spectrometry?

High resolution MALDI_FT-ICR MS allows high-throughput identification and dating estimation of paleontological bones.
This method shows correlation of archaeological sites from different periods with different archaeological, conservation
and environmental context.

Co-authors:

Fabrice Bray, MSAP UAR 3290

AUGUSTE Patrick, EEP - UMR 8198
OUESLATI Tarek, HALMA - UMR 8164
ROLANDO Christian, MSAP UAR 3290, Shrieking Sixties

216
Samples deamidation correlated with time lapse (120,000/50,000 years).

Correlation between LC-MS/MS and MALDI FT-ICR deamidation rate.

217
Session: IM-9 Ionization technologies

ICP-MS - a powerful method for complementary elemental quantification

Abstract ID: 1044

Presenting author: Uwe Karst, University of Münster

Introduction

Elemental mass spectrometry, particularly inductively coupled plasma-mass spectrometry (ICP-MS), has evolved into
one of the most powerful analytical tools, which are currently available. While total elemental analysis by ICP-MS allows
the simultaneous multielemental analysis at concentations down to the ultratrace level already since decades, the
combination of ICP-MS with dedicated sample introduction and separation techniques is in the current focus of the
scientific community.

However, complex analytical challenges rarely can be addressed by only one individual analytical technique. The
complementary use of ICP-MS to electrospray ionization (ESI)-MS or matrix assisted laser desorption/ionization
(MALDI)-MS provides valuable additional information on the sample. The particular strength of ICP-MS is substance-
independent quantification due to atomization and ionization in a plasma of 6.000-10-000°C.

Methods

ICP-MS is based on the atomization and ionization of the analytes in an extremely hot plasma. This results in both
positive and negative consequences as, on the one hand side, any molecular information about the sample is lost, while,
on the other hand side, any atom of a particular element provides exactly the same signal intensity. Therefore, elemental
quantification can be carried our rapidly and easily. This is not only valid for the determination of total elemental
concentrations, but also for hyphenated techniques with ICP-MS, provided the plasma conditions remain the same.

Preliminary data (results)

Speciation analysis, based on the combination of analytical separation techniques with ICP-MS and complementary
electrospray ionization (ESI) MS, offers fascinating possibilities to better understand the role of metal species in biology,
medicine and the environment. While ESI-MS is used for identification of previously separated metal species, ICP-MS is
used for their quantification. Due to the substance-independent signals, calibration mostly can be performed with any
standard substance of the respective elements, so that the use of species-specific standards can be avoided.

Elemental bioimaging based on laser ablation (LA)-ICP-MS offers, in comparison to molecular mass spectrometric
imaging (MSI) techniques such as matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS), far
superior options for quantification, as well as similar spatial resolution. Again, this is due to the atomization and ionization
in the plasma, which leads to massively reduced, although not fully eliminated matrix interferences compared with ESI-
MS.

Single cell ICP-MS with the goal to address the natural elemental distribution at the individual cell level or to undergo
single cell immunohistochemical labelling experiments will (and partly already do) allow to better understand the
composition of individual cells. Again, elemental quantification in single cells can be performed exceptionally well using
specific sample introduction techniques.

The focus of this presentation will be directed towards strategical, technical and applicational aspects of hyphenated
techniques based on ICP-MS.

Please explain why your abstract is innovative for mass spectrometry?

Hyphenated techniques with ICP-MS are a powerful group of techniques, which deliver valuable complementary
information to ESI-MS and MALDI-MS in the analysis of biological, environmental and materials samples.

218
HYBRID IONIZATION SOURCE COMBINING NANO-ELECTROSPRAY AND
DIELECTRIC BARRIER DISCHARGE IONIZATION FOR SIMULTANEOUS
DETECTION OF POLAR AND NON-POLAR COMPOUNDS IN SINGLE
CELLS
Abstract ID: 80

Presenting author: Qinlei Liu, Department of Chemistry and Applied Biosciences, ETH Zurich, CH-8093 Zurich,
Switzerland

Introduction

Single-cell metabolomics is expected to deliver fast and dynamic information on cell function, therefore, it requires rapid
analysis of a wide variety of very small quantities of metabolites in living cells.

Methods

In this study, we introduce a hybrid nanoESI-DBDI source for single-cell analysis, which has two modes, an "ESI mode"
and a "hybrid mode" to detect intracellular metabolites with different polarities. As shown in Figure 1, a glass capillary
with a tip diameter of 1 µm is inserted into a cell for sucking in cell contents, and then the cellular metabolites are ionized
by the hybrid nanoESI-DBDI source, where the nanoESI primarily ionizes polar metabolites, followed by DBDI as a post-
ionization source to ionize less polar and non-polar metabolites that have low ESI ionization yield.

Preliminary data (results)

We were able to demonstrate detection of metabolites from single plant and animal cells. This hybrid source improved
the coverage of metabolites of different polarities from 50 metabolites (ESI mode) to 86 metabolites (hybrid mode) in
onion cells, and from 40 metabolites (ESI mode) to 111 metabolites (hybrid mode) in PANC-1 cells (a continuous tumor-
cell line from a human carcinoma of the exocrine pancreas).

Please explain why your abstract is innovative for mass spectrometry?

This hybrid ionization source improves the coverage, ionization efficiency, and limit of detection of metabolites with
different polarities, and could potentially contribute to the fast-growing field of single-cell metabolomics.

Co-authors:

Jiayi Lan, Department of Chemistry and Applied Biosciences, ETH Zurich, CH-8093 Zurich, Switzerland
Ri Wu, Department of Chemistry and Applied Biosciences, ETH Zurich, CH-8093 Zurich, Switzerland
Alina Begley, Department of Chemistry and Applied Biosciences, ETH Zurich, CH-8093 Zurich, Switzerland
Wenjie Ge, Department of Biology, ETH Zurich, CH-8093 Zurich, Switzerland
Renato Zenobi, Department of Chemistry and Applied Biosciences, ETH Zurich, CH-8093 Zurich, Switzerland

Single cell sampling process and photo of the nanoESI-DBDI source.

219
INCREASING MOLECULAR COVERAGE AND SENSITIVITY FOR MALDI-
MSI VIA DIRECT 2-PHOTON IONISATION OF ANALYTES ENABLED BY
MALDI-2 – APPLICATIONS TO AROMATIC ANTIOXIDANTS IN TISSUES
AND CELLS
Abstract ID: 102

Presenting author: Tassiani Sarretto, UOW

Introduction

Matrix-assisted laser/desorption ionisation-mass spectrometry imaging (MALDI-MSI) enables label-free imaging of

biomolecules in biological tissues. However, many molecules remain undetected due to their poor ionisation efficiencies,
leading to poor sensitivities and limiting spatial resolution. MALDI-2 is a recent post-ionisation method that can
dramatically improve the ionisation efficiencies of many molecules by inducing a second ionisation event mediated by
charge transfer from the photoexcited matrix. In addition to MALDI-like processes, MALDI-2 can also lead to direct
photoionisation of certain molecules via direct [1+1] resonant-enhanced multiphoton ionisation (REMPI). Herein, we
demonstrate that REMPI ion yields are improved using primary laser fluences at/or below the MALDI ionisation threshold
for aromatic antioxidants in mouse brain and human prostate cancer tissue as well as individual neuronal cells.

Methods

Data were acquired from mouse brain (7 and 50 µm pixel size), prostate cancer tissue (15 µm pixel size) and individual
neuronal cells (5 µm pixel size) using an Orbitrap Elite mass spectrometer (Thermo Fisher Scientific, Germany) coupled
to an intermediate pressure MALDI/ESI ion source (Spectroglyph, USA). MALDI-2 was performed using a 266 nm laser
NanoDPSS, Litron Lasers, UK). Both lasers were operated at 300 Hz with a time delay of 10 µs. Primary laser energy
was controlled using a motorised attenuator (PowerXP, Lithuania). Biological tissues were coated with 2,5–
dihydroxyacetophenone using an automatic sprayer (TM-Sprayer, USA).

Preliminary data (results)

Using a sub-threshold primary laser pulse energy of 0.6µJ, desorption craters ~6.6µm in diameter were produced on
mouse brain tissue, but no tissue-related signals were detected. When the MALDI-2 laser was enabled abundant signals
corresponding to the radical cations of vitaminE, ubiquinol Q9, and ubiquinol Q10 were detected. These are generated
via a process decoupled from the desorption event and independent of MALDI-like charge-transfer ionisation processes.
Under these conditions, very few MALDI/MALDI-2 ions (e.g., protonated lipids) were detected, an observation attributed
to the sparse analyte/matrix plume having insufficient density for MALDI/MALDI-2 ionisation processes. As the primary
laser energy was increased to 1.3µJ/pulse and 2.1µJ/pulse, spectra resemble those of conventional MALDI-2 with most
ions corresponding to alkali adducted or protonated species (Figure1). However, despite the favourable MALDI/MALD-2
conditions and 5-6-fold greater sampling volumes, signals for radical cations decreases. This counterintuitive observation
demonstrated a significantly increased direct REMPI ion signal per unit desorption area using primary laser energies
below the MALDI threshold while the use of reduced laser energies enables pixel sizes as small as 7µm without
oversampling (two-fold improvement compared to standard conditions). This was applied to the imaging of aromatic
antioxidants throughout mouse brain and human prostate cancer biopsies (Figure2). In addition, other low mass signals
were also observed arising from photoionisation of other endogenous metabolites. Extension of this work to mapping
metabolite distributions throughout individual neuronal cells at different stages of differentiation will also be presented.

Please explain why your abstract is innovative for mass spectrometry?

Simultaneous improvement in both spatial resolution and sensitivity for the mass spectrometry imaging of REMPI-
compatible analytes. Provides a means to overcome the sensitivity decrease encountered as pixel sizes are reduced.

220
MALDI-2 mass spectra at different desorption pulse energy.

Ion distribution of vitaminE and ubiquinolQ10 on prostate cancer tissue.

221
EFFECTS OF BIOCHEMICAL BUFFERS ON PROTEIN THERMAL STABILITY
MEASURED USING SUBMICRON EMITTERS AND FAST HEATING
Abstract ID: 316

Presenting author: Jacob S. Jordan, University of California, Berkeley

Introduction

Variable-temperature electrospray ionization (vT-ESI) experiments, wherein mass spectra are acquired as a function of
temperature, have been used to measure the melting process of proteins, macromolecular complexes, and antibodies.
These experiments are typically performed from ammonium acetate (AA) solutions by heating the electrospray capillary,
but the long thermal lag times of current vT-ESI sources (>45 s) make this method incompatible with high-throughput
analysis. Additionally, biomolecules are typically purified in buffers containing high ionic strengths of nonvolatile salts,
making comparisons between vT-ESI measurements and different techniques ambiguous. Here, protein ions are
produced from solutions containing between 10 mM – 175 mM nonvolatile salts and buffers and rapid laser-heating or
resistive heating of nESI emitters is compared to determine if protein thermochemistry measurements are achievable on
native-LC timescales.

Methods

A variable-temperature nESI source was constructed. A CO2 laser was aligned perpendicular to the electrospray emitter.
Measurements were performed on an Orbitrap Elite or Waters Synapt G2-Si mass spectrometer. Borosilicate capillaries
were pulled using a Flaming/Brown P-87 micropipette puller to produce emitters with inner diameters of 222 nm.
Solutions of proteins, protein complexes, and antibodies were prepared in 10 – 175 mM ammonium acetate, NaCl/Tris,
sodium phosphate, and citrate buffers titrated to pH 4.5 – 7.4. The capillary temperature and laser powers were varied
from 26 ℃ to 98 ℃ and between 0 W and 25 W, respectively.

Preliminary data (results)

vT-ESI was used to obtain data of proteins and protein complexes in solutions of AA or biochemically relevant buffers,
such as NaCl/Tris and sodium phosphate. At room temperature, the charge state distribution of cytochrome c (cyt c) is
narrow under both solution conditions (figure 1, left inset). At ~90 ℃, clear differences in the charge state distribution are
observed that indicate more compact species are present in the NaCl/Tris solution (figure 1, right inset). A fit of the
average charge to a two state model results in similar melting temperatures. However, the temperature-dependent
relative abundances of many individual charge states are dissimilar between the two solutions. For example, between 27
°C and 67 °C the relative abundance of the +8 charge state increases 43% and 6% in AA and NaCl/Tris solutions,
respectively. Similarly, the formation temperatures of the +11-+13 charge states are similar in AA solutions but are
unique in NaCl/Tris, suggesting that these may be unique conformers under biochemical conditions. For this system and
others we investigated, buffer choice has a prominent impact on the thermochemistry of proteins measured by vT-ESI,
emphasizing the value of using biochemically relevant buffers for native mass spectrometry. However, current vT-ESI
sources are limited by long thermal lag times, so resistive vT-ESI and rapid laser-heating of the nESI emitter will be
compared to determine if protein melting curves can be generated directly from biochemical buffers on a native-LC
timescale.

Please explain why your abstract is innovative for mass spectrometry?

Native mass spectrometry using laser-heated submicron emitters can accelerate the study of protein biophysical
properties over conventional variable temperature methods.

Co-authors:

Evan R. Williams, University of California, Berkeley

222
Average Charge of Cytochrome c as a Function of Temperature

223
HYPHENATION OF GAS CHROMATOGRAPHY TO A DUAL IONIZATION
SOURCE TOFMS FOR IMPROVED COMPOUND IDENTIFICATION
Abstract ID: 633

Presenting author: Steffen Bräkling, University of Wuppertal, TOFWERK

Introduction

Gas chromatography (GC) coupled to mass spectrometry (MS) is one of the most common and established hyphenation
techniques and has been used routinely for decades. However, most commonly coupled to electron ionization (EI) and
operated on a nominal mass resolution mass analyzer, for compound identification this technique strongly depends on
existing compound entries in mass spectral libraries. To overcome this drawback additional molecular ion information
gained via “soft ionization” e.g., chemical ionization (CI) in addition to a high resolution (HR) mass analyzer can increase
the identification probability for library unknown compounds drastically. To combine the advantages of EI and CI we
present a time of flight (TOF) mass spectrometer operating two ionization methods simultaneously, whereat being
hyphenated to a single GC separation column.

Methods

The presented mass spectrometer (EC-TOF; TOFWERK AG, Thun, Switzerland) operates a 70 eV EI-source parallel to
a medium pressure chemical ionization source both being coupled to one single GC (Agilent 7890A GC; Agilent
Technologies, Inc., Santa Clara, CA, USA). A specific GC-MS coupling is designed to deliver 50 % of the sample to each
ion source and subsequently allows a simultaneous detection of both types of ions of the analyte. To show the
advantage for compound identification using simultaneous EI & CI analysis the emissions of various materials were
sampled on desorption tubes and analyzed via thermal desorption-GC-MS.

Preliminary data (results)

The principle of operation of the dual ionization TOFMS will be shown via standard mix analysis based on various
parameters e.g., spectra quality and CI/EI chromatography alignment. Due to a careful designed coupling of the GC to
both ion sources the eluting compounds are ionized in parallel with both methods. Due to ion optical switching within the
ion transfer stage of the MS both types of ions are detected quasi-simultaneously. This provides ideal data alignment
conditions and therefore offers easy assignment of the EI to the corresponding CI mass spectra (Figure 1). Information
gained via each ion source are easily linked via GC retention time. Special attention was given to the GC-transfer
characterization following the effluent-splitter. Time shifts between CI and EI chromatogram are in a range below 100 ms.
The mass of each compound transferred to each ion source yield about 40-60 % of the total. No significant time
differences or split-ratio-shifts are discovered over a complete GC run. Increased identification probabilities using both
information are shown in suspect- and non-target-screening approaches.

It can be shown that compound identification probabilities compared to traditional GC-MS instruments using only 70 eV
EI are drastically improved with this setup. Further it is demonstrated that EI only approaches often do even end up in
false positive results. Therefore, VOCs emitted from materials used for e.g., car interiors were analyzed using the
described system. An example of a TD-GC-EC-TOF measurement of artificial leather emissions is shown in figure 2.

Please explain why your abstract is innovative for mass spectrometry?

Dual ionization, high resolution, accurate mass TOFMS; simultaneous structural and molecular information within one
single GC-MS run,

Co-authors:

Sonja Klee, TOFWERK

Christina Hinterleitner, Berner Fachhochschule
Luca Cappellin, University of Padua
Thorsten Benter, university of Wuppertal
Hendrik Kersten, university of Wuppertal

224
Set up of the dual ionization TOFMS.

Parallel CI and EI chromatograms of an artificial leather emission.

225
RAPID FORMATION OF HIGHLY OXYGENATED ORGANIC MOLECULES
(HOM) REVEALED WITH THE MULTI-SCHEME CHEMICAL IONIZATION
(MION) INLET
Abstract ID: 676

Presenting author: Matti Rissanen, Aerosol Physics, Tampere University, Tampere, Finland, Karsa Inc., Helsinki,
Finland

Introduction

The formation of highly oxygenated organic molecules (HOM) by rapid autoxidation of hydrocarbons and their relation to
the formation and growth of atmospheric secondary organic aerosol (SOA) has been known now for a decade. From the
pioneering studies it was already apparent how the general process can yield highly oxygenated material apparently
within sub-second time-scales, yet the instrumentation and the methodologies used were not able to record the details of
these fast events. Now, with the recently developed Multi-scheme chemical ionization inlet (MION) [1], we are able to
study the fast production with evermore shorter reaction times, allowing to inspect the details of the rapid HOM formation
reactions [2].

Methods

HOM formation from ozonolysis initiated autoxidation of the most common monoterpene a-pinene was investigated
under atmospheric conditions at variable short reaction times from 75 to 500 milliseconds. Several quartz flow tube
reactors equipped with movable injectors and coupled to a chemical ionization atmospheric pressure interface chemical
ionization mass spectrometer (CI-APi-ToF-MS), and the MION inlet, were applied for the work. The experimental work
was augmented by quantum chemical computations of the step-by-step reaction mechanism starting from the early
ozonolysis intermediates, and subsequent master equation modelling to obtain the all-important reaction rates.

Preliminary data (results)

The application of the MION inlet allowed to inspect the dynamics of the evolving radical product distribution at variable
short reaction times. With increasing reaction time we were able to record an increasing amount of reaction products,
together with the shift from radical dominated product distribution to closed-shell product domination, under these
relatively high reactant concentration experiments. Most importantly we were able to illustrate how the 8 oxygen atom
containing peroxy radical (i.e., C10H15O8) forms in less than 100 milliseconds timeframe, amply illustrating the high
efficiency of the atmospheric autoxidation process. In the IMSC meeting, I will concentrate on the experimental
application of the MION inlet to study the short reaction times under atmospheric conditions, and only shortly reveal the
computational details needed to shed light on this important series of reactions.

References

1. M. P. Rissanen, J. Mikkilä, S. Iyer, and J. Hakala, Multi-scheme chemical ionization inlet (MION) for fast switching of
reagent ion chemistry in atmospheric pressure chemical ionization mass spectrometry (CIMS) applications, Atmos.
Meas. Tech. (2019) 12, 6635-6646 (https://doi.org/10.5194/amt-12-6635-2019)

2. S. Iyer, M. P. Rissanen, R. Valiev, S. Barua, J. E. Krechmer, J. Thornton, M. Ehn, and T. Kurtén, Molecular
mechanism for rapid autoxidation in α-pinene ozonolysis, Nature comm. (2021) 12, 1-6 (https://doi.org/10.1038/s41467-
021-21172-w)

Please explain why your abstract is innovative for mass spectrometry?

Inlet design that allows for studying rapid gas-phase radical reactions under atmospheric conditions applying several
ionization schemes in fast repetition.

Co-authors:
Siddharth Iyer, Aerosol Physics, Tampere University, Tampere, Finland
Rashid Valiev, Department of Chemistry, University of Helsinki, Helsinki, Finland, Tomsk State University, Tomsk, Russia
Shawon Barua, Aerosol Physics, Tampere University, Tampere, Finland
Jordan E Krechmer, Aerodyne Research, Inc., Billerica, USA
Joel Thornton, Department of Atmospheric Science, University of Washington Seattle, Washington, USA
Mikael Ehn, Institute for Atmospheric and Earth System Research (INAR/Physics), University of Helsinki, Helsinki,
Finland
Theo Kurtén, Department of Chemistry, University of Helsinki, Helsinki, Finland

226
Session: LS-8 Translational MS – Cancer and Immunology, and MS

PROTEOMIC ANALYSIS OF CANCER INTERNAL HETEROGENEITY

Abstract ID: 789

Presenting author: Tamar Geiger, Weizmann Institute of Science

Introduction

Cancer heterogeneity is one of the major challenges that hampers the ability to cure the disease. Tumors differ in their
genetic profiles and the cellular interactions in the microenvironment, and each tumor may have multiple different clones
with distinct molecular characteristics. Therefore understanding cancer heterogeneity has major translational
implications. In breast cancer, different regions within single tumors may vary in the levels of key molecular markers, and
this heterogeneity affects metastasis and treatment response. Global analyses of internal heterogeneity have mainly
concentrated on the genomic layer, revealing evolutionary trajectories, and their association with immune selection.
However, focusing only on the genomic level ignores the functional proteomic layer of tumor subpopulations, and their
interactions with the tumor microenvironment.

Methods

Using mass spectrometry-based proteomics, we aim to understand the functional proteomic layer of cancer
heterogeneity in breast cancer. We combined analysis of clinical samples with histopathological analysis and functional
validations, to unravel novel regulators of cancer progression. To tackle the challenge of the very small sample amounts
from tumor regions, we implemented an automated high-throughput pipeline, which combines the Single Pot Solid Phase
Sample Preparation (SP3) technique with multiplexed TMT labeling. These methods enable highly sensitive sample
preparation from small formalin-fixed tissue (FFPE) samples and from single cells in culture.

Preliminary data (results)

Analysis of >300 breast cancer tumor regions unraveled the association between clinical parameters and the protein
networks, and showed their heterogeneity within single tumors. Our research showed the importance of each clinical
feature and the significance of the immune system in affecting tumor heterogeneity. For example, we found that immune
functions reduce tumor heterogeneity in high-grade luminal tumors, while cell proliferation is dominant in homogeneous
high-grade triple-negative tumors. In addition, we found that proteomic heterogeneity can be seen in tumors that seem to
be homogeneous based on histopathological analysis, and our analyses propose potential therapeutic targets that can
address heterogeneous tumors. Finally, a comparison of the proteomic heterogeneity to genomic heterogeneity showed
lower proteomic heterogeneity, suggesting that proteomic analyses are more likely to identify efficient drug targets.

Please explain why your abstract is innovative for mass spectrometry?

The FFPE-laser-capture-SP3-TMT workflow enabled highly sensitivity protein digestion from archived clinical samples to
allow large scale proteomic analysis of internal tumor heterogeneity.

Co-authors:

Mariya Mardamishina, Tel Aviv University

Anjana Shenoy, Weizmann Institute of Science
Einav Gal-Yam, Sheba Medical Center
Iris Barshak, Sheba Medical Center

227
SENSITIVE AND QUANTITATIVE DETECTION OF MHC-I DISPLAYED
NEOEPITOPES USING A SEMIAUTOMATED WORKFLOW AND TOMAHAQ
MASS SPECTROMETRY
Abstract ID: 18

Presenting author: Jennie Lill, Genentech

Introduction

Advances in several key technologies, including MHC peptidomics, have helped fuel our understanding of basic immune
regulatory mechanisms and the identification of T cell receptor targets for the development of immunotherapeutics.
Isolating and accurately quantifying MHC-bound peptides from cells and tissues enables characterization of dynamic
changes in the ligandome due to cellular perturbations. However, the current multistep analytical process is challenging,
and improvements in throughput and reproducibility would enable rapid characterization of multiple conditions in parallel.
Here, we describe a robust and quantitative method whereby peptides derived from MHC-I complexes from a variety of
cell lines, including challenging adherent lines such as MC38, can be enriched in a semiautomated fashion on reusable,
dry-storage, customized antibody cartridges.

Methods

This method involves 96 simultaneous enrichments at a similar level of quality as a manual workflow. TOMAHAQ
(Triggered by Offset, Multiplexed, Accurate-mass, High-resolution, and Absolute Quantification), a targeted mass
spectrometry technique that combines sample multiplexing and high sensitivity, was employed to characterize
neoepitopes displayed on MHC-I by tumor cells and to quantitatively assess the influence of neoantigen expression and
induced degradation on neoepitope presentation.

Preliminary data (results)

This unique combination of robust semiautomated MHC-I peptide isolation and high-throughput multiplexed targeted
quantitation allows for both the routine analysis of >4000 unique MHC-I peptides from 250 million cells using nontargeted
methods, as well as quantitative sensitivity down to the low amol/μl level using TOMAHAQ targeted MS.

Please explain why your abstract is innovative for mass spectrometry?

Development of a Triggered by Offset, Multiplexed, Accurate-mass, High-resolution, and Absolute Quantification

(TOMAHAQ) based method for the high throughput analysis of neo-antigen derived T cell epitopes.

MHC peptide identification and quantitation protocol using TOMAHAQ

228
TOMAHAQ is more sensitive than untargeted MS

229
PERSONALIZED RESPONSES TO VIRAL INFECTIONS REVEALED BY
MONITORING SERUM IGG1 REPERTOIRES
Abstract ID: 223

Presenting author: Danique van Rijswijck, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for
Biomolecular Research and Utrecht Institute for Pharmaceutical Sciences, University of Utrecht, Padualaan 8,
Utrecht 3584 CH, the Netherlands, Netherlands Proteomic Center, Padualaan 8, Utrecht 3584 CH, the
Netherlands

Introduction

The pandemic caused by SARS-CoV-2 has led to a huge increase in research aimed to discover possible treatment
strategies. The viral Spike (S) protein is thereby often used as a target as this protein is crucial for host infection.
However, the S protein of SARS-CoV-2 is prone to mutations, stirring up questions regarding the protection of antibodies
directed against one variant towards the other variants, especially for so-called Variants of Concern (VOCs). Existing
assays to measure antibody cross-reactivity against different S protein variants lack the discriminatory power to provide
insights at the level of individual antibody clones. Here, we developed a direct mass spectrometry (MS) based approach
enabling the analysis of the polyclonal response of individuals producing SARS-CoV-2 S -targeting IgG1 clones.

Methods

We investigated the S-directed IgG1 repertoires of eight donors, which had suffered an infection with different variants of
SARS-CoV-2. In brief, agarose beads coated with S trimer from either the Wuhan Hu-1, Alpha, Beta, or Gamma variant
were used to capture S -targeting antibodies from convalescent plasma of the eight donors. Subsequently, we generated
Fab fragments from the captured IgG1 clones which were directly measured by liquid chromatography (LC)-MS. Unique
IgG1 clones could be distinguished by their mass and their retention time (Figure 1).

Preliminary data (results)

We detected the highest amount of S-specific IgG1 antibodies in the more severely ill donors. Every donor revealed a
unique polyclonal repertoire of IgG1 Fabs, of which the majority were found to bind to all tested variants of the S protein,
albeit with varying affinity. Within donors, individual clones also displayed unique behavior. Intriguingly, the clonal profiles
and clonal behavior did not always relate to the observed effects in concomitantly performed neutralization assays.

In conclusion, by assessing IgG1 repertoires following infection we observe a polyclonal IgG1 response in SARS-CoV-2
infected people that are distinctive for each donor. Each clone exhibited a distinct pattern of cross-reactivity versus
SARS-CoV-2 S protein variants. Furthermore, the clonal repertoire analysis did not per se correlate with in
vitro neutralization assays, highlighting the need for a variety of assays to judge the full scale of antiviral fitness of a
patient. The knowledge that every clone shows a different SARS-CoV-2 binding pattern, is important to consider when
developing monoclonal antibody-therapies or novel vaccination strategies. Our approach can be extended beyond
SARS-CoV-2 to have a detailed look into the antigen-specific antibody response in a wide variety of infectious diseases.

Please explain why your abstract is innovative for mass spectrometry?

LC-MS based profiling reveals that polyclonal IgG1 responses following SARS-CoV-2 infection are highly individual, with
each clone exhibiting a distinct pattern of cross-reactivity against SARS-CoV-2 variants of concern.

Co-authors:

Albert Bondt, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research and Utrecht
Institute for Pharmaceutical Sciences, University of Utrecht, Padualaan 8, Utrecht 3584 CH, the Netherlands,
Netherlands Proteomic Center, Padualaan 8, Utrecht 3584 CH, the Netherlands
Max Hoek, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research and Utrecht
Institute for Pharmaceutical Sciences, University of Utrecht, Padualaan 8, Utrecht 3584 CH, the Netherlands,
Netherlands Proteomic Center, Padualaan 8, Utrecht 3584 CH, the Netherlands
Tom Caniels, Department of Medical Microbiology, Amsterdam UMC, University of Amsterdam, Amsterdam Institute for
Infection and Immunity, Meibergdreef 9, Amsterdam 1105 AZ, the Netherlands
Karlijn van der Straten, Department of Medical Microbiology, Amsterdam UMC, University of Amsterdam, Amsterdam
Institute for Infection and Immunity, Meibergdreef 9, Amsterdam 1105 AZ, the Netherlands, Department of Internal
Medicine, Amsterdam UMC, Vrije Universiteit Amsterdam, Amsterdam Institute for Infection and Immunity, Meibergdreef
9, Amsterdam 1105 AZ, the Netherlands.
Godelieve de Bree, Department of Internal Medicine, Amsterdam UMC, Vrije Universiteit Amsterdam, Amsterdam

230
Institute for Infection and Immunity, Meibergdreef 9, Amsterdam 1105 AZ, the Netherlands.
Rogier Sanders, Department of Medical Microbiology, Amsterdam UMC, University of Amsterdam, Amsterdam Institute
for Infection and Immunity, Meibergdreef 9, Amsterdam 1105 AZ, the Netherlands
Marit van Gils, Department of Medical Microbiology, Amsterdam UMC, University of Amsterdam, Amsterdam Institute for
Infection and Immunity, Meibergdreef 9, Amsterdam 1105 AZ, the Netherlands
Albert Heck, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research and Utrecht
Institute for Pharmaceutical Sciences, University of Utrecht, Padualaan 8, Utrecht 3584 CH, the Netherlands,
Netherlands Proteomic Center, Padualaan 8, Utrecht 3584 CH, the Netherlands

MS-based approach to monitor full IgG1 and S-directed IgG1 profiles

231
FIBROBLAST ACTIVATION PROTEIN TRIGGERS SELECTIVE RELEASE OF
DRUG PAYLOAD FROM SMALL MOLECULE-DRUG CONJUGATES IN
SOLID TUMORS
Abstract ID: 510

Presenting author: Ettore Gilardoni, Philochem

Introduction

To date, cancer is still one of the main causes of death in the world. The development of novel tools to selectively target
tumor cells (magic bullet concept) with small molecule-drug conjugates (SMDCs) is a topic which is gaining a lot of
consensus in the field. The possibility of targeting tumor cells for the delivery of cytotoxic drugs allows to increase the
antitumoral activity of the drug and to reduce its side effects. Furthermore, potent drugs that typically cannot be used
alone as chemotherapeutic agents due to their systemic toxicity, might find new therapeutic applications when used as
SMDCs, thanks to their reduced off-target effects. Philochem recently developed a small molecule directed against the
fibroblast activation protein (FAP) and conjugated with the antineoplastic agent MMAE.

Methods

A novel LC-MS method was developed for the ex-vivo quantification of the released cytotoxic payload (i.e., MMAE) in
biodistribution experimentsfrom mice treated with several hit compounds. Deuterated MMAE was used as internal
standard for quantification measurements. Sample preparation was optimized to reduce sample loss and increase
sensitivity. Tissues are mechanically homogenized, and proteins precipitated. Samples are then subjected to two solid
phase extraction steps prior to LC-MS injection. The precise amount of MMAE released from different SMDCs is
quantified by integrating its MS1 peak and the peak of the correspondent internal standard.

Preliminary data (results)

Biodistribution experiments play a crucial role in the drug discovery since they allow to correctly evaluate the
accumulation of the drug at the site of action. However, the interpretation of these experiments is very often challenging
for SMDCs. For this class of molecules, it is essential not only to evaluate the targeted delivery of the drug at the tumor
site, but also the selective release of the cytotoxic payload. MMAE is a very potent toxin, that can have critical adverse
effects if not carefully delivered at the action site. In this study, we developed a LC-MS method that allows us to quantify
MMAE released from several SMDC derivatives in preclinical studies. MS results contributed to the identification of the
best-in-class molecule and precisely correlate with therapy experiments (figure 1), providing a deep understanding of the
mechanism of action of these SMDCs in our preclinical model. To conclude, in this study we identified a compound with
high and selective release of MMAE in the tumor and not in other organs.

Please explain why your abstract is innovative for mass spectrometry?

We developed a LC-MS workflow which allows the precise quantification of cytotoxic payloads released from SMDCs in
preclinical biodistribution experiments.

Co-authors:

Riccardo Stucchi, Philochem

Aureliano Zana, Philochem
Galbiati Andrea, Philochem
Samuele Cazzamalli, Philochem
Dario Neri, ETH (Zurich), Philogen (Siena)

232
A) MMAE quantificationin the the tumor. B) Therapy experiments.

233
A NEW LC-MS/MS BASED METHOD TO QUANTIFY URINARY THYMINE
DIMERS AFTER ULTRAVIOLET RADIATION OF THE SKIN
Abstract ID: 669

Presenting author: Catharina Margrethe Lerche, Department of Dermatology, Copenhagen University Hospital—
Bispebjerg and Frederiksberg, Department of Pharmacy, University of Copenhagen

Introduction

Solar ultraviolet radiation (UVR) is a carcinogen and exposure of the skin results in DNA damage which may result in
mutagenesis, immunosuppression and skin cancer. Solar UVR contains approximately 95% UVA (315-400 nm) and
approximately 5% UVB (280-315 nm) and both categories of UVR can penetrate the skin and form DNA damage.
Cyclobutane pyrimidine dimers (CPDs) including thymine dimers are among the most frequent DNA damage and is a
reaction between the C=C bonds of adjacent pyrimidine in the DNA strand. Also 6-4 photoproduct and oxidative damage
occur. When CPDs are formed a nucleotide excision repair system is activated and CPDs are excreted in the urine

The aim of this study was to develop a mass spectrometry-based method to quantify urinary CPDs after UVR exposure.

Methods

Eight healthy volunteers were whole-body UVR exposed on 3 following days. Morning urine was collected at Day 1
(before irradiation) and the following 7-9 days.Prior to analysis, sample preparation consisted of a simple dilution. A
Waters Xevo TQ-XS tandem quadrupole mass spectrometer coupled to a Waters I-class UPLC was used for quantitative
analysis in MRM mode.The Danish Research Ethics Committee provided ethical approval (H-20076172). The study
adhered to the Declaration of Helsinki; the volunteers gave written, informed consent.

Preliminary data (results)

Le Curieux and Hemminiki published in 2001 a methodology to quantify CPDs in the urine using HPLC and 32P
postlabelling. This method has a very low limit of quantification but has also a very low throughput and it is not desirable
to work with isotopes. There is a need for a mass spectrometry-based method to quantify the thymine dimers. When
CPDs are formed a nucleotide excision repair system is activated and CPDs are excreted in the urine. Therefore, CPDs
in the urine can be a surrogate or proxy measure for a biomarker that reflects the amount of repaired UVR-induced DNA
damage on the total area of skin exposed

The LC-MS/MS results showed excretion of thymine dimers the morning after the first irradiation. With a maximum of
excretion on Day 5-6. The amount of thymine dimers was very decreased after 7-9 days but it was not as low as before
the irradiation.

A methodology able to quantify thymine dimers excreted in the urine can be useful in numerous areas of research
including DNA repair, DNA damage, evaluating possible candidates for systemic photoprotection and investigating
possible photocarcinogenic compounds.

Our search for photoproducts is based on the assumption that thymidine dimers released by DNA repair are not
degraded further and are excreted in urine as a dimer However, potential exists for dimers to be present in urine as other
oligomeric forms.

Please explain why your abstract is innovative for mass spectrometry?

A new LC-MS/MS based method to quantify thymine dimers excreted in the urine after UVR exposure of the skin was
developed

Co-authors:

Peter Alshede Philipsen, Department of Dermatology, Copenhagen University Hospital—Bispebjerg and Frederiksberg
Jakob Heydenreich, Department of Dermatology, Copenhagen University Hospital—Bispebjerg and Frederiksberg
Sigurd Hermansson, Waters Corporation
Hans Christian Wulf, Department of Dermatology, Copenhagen University Hospital—Bispebjerg and Frederiksberg

234
New LC-MS/MS-basedmethod to quantify thymine dimers in urine.

235
UBIQUITIN LIGASE STUB1 DESTABILIZES IFNΓ-RECEPTOR COMPLEX TO
SUPPRESS TUMOR SIGNALING
Abstract ID: 712

Presenting author: Onno B. Bleijerveld, Proteomics Facility, The Netherlands Cancer Institute

Introduction

Although immune checkpoint blockade (ICB) has been a major clinical success in the treatment of a variety of cancers,
the majority of patients fail to show durable clinical responses. In many cases, this is due to the intrinsic insensitivity that
tumors develop against cytokines secreted by cytotoxic T-cells, including IFNγ and TNF. IFNγ can promote anti-tumor
activity indirectly, for instance by inducing secretion of lymphocyte-attracting chemokines, but also directly, by improving
antigen presentation, inducing the expression of cell cycle inhibitors and pro-apoptotic proteins. Although the IFNγ
signaling pathway has been studied extensively, and different regulatory mechanisms have been uncovered, less is
known about cell-intrinsic regulation of IFNγ-R1, its essential ligand-binding receptor chain residing at the tumor cell
surface.

Methods

Genome-wide CRISPR/Cas9 knockout screens in D10 and SK-MEL-23 melanoma cells were used to uncover critical
factors regulating cell surface IFNγ-R1 abundance. The strongest negative regulator, STUB1, was further validated in
other cancer cell lines by flow cytometry. Immunoprecipitation of diGly peptides and LC-MS/MS were employed for
profiling the (ubiquitinated) proteome in Ctrl and STUB1 KO melanoma cells. Further mechanistic insights were obtained
by performing extensive co-precipitation experiments, Western blotting, colony forming and in vitro ubiquitination assays.
To explore the impact of STUB1 inactivation on IFNγ signaling and anti-PD1 treatment outcome, preclinical
immunotherapy mouse melanoma models were employed.

Preliminary data (results)

In two CRISPR/Cas9 knockout screens, we identified STUB1 as negative regulator of cell surface IFNγ-R1 expression in
melanoma cells, which was extended to other cancer cell lines by measuring the induction of cell surface IFNγ-R1 after
Cas9-mediated STUB1 knockout. LC-MS/MS-based proteome profiling confirmed increased IFNγ-R1 levels in sgSTUB1
melanoma cells and additionally revealed an increase in JAK1 protein. Inactivating JAK1 in a STUB1-deficient
background indicated that STUB1 deficiency promotes IFNγ-R1 stabilization in a largely JAK1-dependent fashion.
MG132 treatment, furthermore, showed that increased IFNγ-R1 and JAK1 protein levels were caused by reduced
proteasomal degradation. Immunoprecipitation of diGly peptides from sgCtrl and sgSTUB1 melanoma cells followed by
LC-MS/MS indicated that IFNγ-R1K285 and JAK1K249 are the most likely targets of ubiquitination by STUB1, which was
further corroborated by mutagenesis and in vitro ubiquitination assays revealing that STUB1 is able to directly
ubiquitinate IFNγ-R1K285 and JAK1K249. Our results support a model in which STUB1 through its E3 ubiquitin ligase
activity regulates proteasomal turnover of IFNγ-R1 and JAK1 proteins through ubiquitinating K285 and K249,
respectively. We further demonstrated that STUB1 inactivation sensitizes melanoma to cytotoxic T-cells through
amplified IFNγ signaling and this was clinically supported by a strong negative correlation between the expression of
STUB1 and IFNγ-response genes in patients undergoing anti-PD-1 treatment. Finally, by employing two preclinical
murine melanoma models, we demonstrated that STUB1 inactivation enhances IFNγ signaling and increases an anti-PD-
1 response in heterogeneous tumors with wildtype cells, but not in homogenous STUB1-deficient tumors.

Please explain why your abstract is innovative for mass spectrometry?

The power of mass spectrometry-based proteomics complemented genetics, molecular biology and animal studies to get
a mechanistic understanding of how a novel target might help improve future immunotherapy in cancer.

Co-authors:

Georgi Apriamashvili, Division of Molecular Oncology and Immunology, The Netherlands Cancer Institute, Oncode
Institute
David W. Vredevoogd, Division of Molecular Oncology and Immunology, The Netherlands Cancer Institute, Oncode
Institute
Oscar Krijgsman, Division of Molecular Oncology and Immunology, The Netherlands Cancer Institute, Oncode Institute
Maarten A. Ligtenberg, Division of Molecular Oncology and Immunology, The Netherlands Cancer Institute, Oncode
Institute
Beaunelle de Bruijn, Division of Molecular Oncology and Immunology, The Netherlands Cancer Institute
Joleen Traets, Division of Molecular Oncology and Immunology, The Netherlands Cancer Institute

236
Daniela D’Empaire Altimari, Division of Molecular Oncology and Immunology, The Netherlands Cancer Institute
Alex van Vliet, Division of Molecular Oncology and Immunology, The Netherlands Cancer Institute
Nils L. Visser, Division of Molecular Oncology and Immunology, The Netherlands Cancer Institute
James D. Londino, Division of Pulmonary, Critical Care and Sleep Medicine, The Ohio State University Wexner Medical
Center
Rebekah Sanchez-Hodge, Department of Pharmacology, McAllister Heart Institute, University of North Carolina at
Chapel Hill
Leah E. Oswalt, Department of Pharmacology, McAllister Heart Institute, University of North Carolina at Chapel Hill
Selin Altinok, Department of Pharmacology, McAllister Heart Institute, University of North Carolina at Chapel Hill
Jonathan C. Schisler, Department of Pharmacology, McAllister Heart Institute, University of North Carolina at Chapel Hill
Maarten Altelaar, Biomolecular Mass Spectrometry and Proteomics, Utrecht University, Utrecht, The Netherlands, Bijvoet
Center for Biomolecular Research, Utrecht University, Utrecht, The Netherlands, Utrecht Institute for Pharmaceutical
Sciences, University of Utrecht, Utrecht, The Netherlands, Netherlands Proteomics Center, Utrecht, The Netherlands,
Proteomics Facility, The Netherlands Cancer Institute
Daniel S. Peeper, Division of Molecular Oncology and Immunology, The Netherlands Cancer Institute, Oncode Institute

237
Session: LS-9 Proteomics: Top down

NEW FRONTIERS IN PROTEOMICS - PROTEOFORMS, PROTEOFORM

FAMILIES, AND THE HUMAN PROTEOFORM PROJECT
Abstract ID: 775

Presenting author: Lloyd M. Smith, University of Wisconsin - Madison

Introduction

Proteins are the primary effectors of function in biology, and thus complete knowledge of their structure and behavior is
needed to decipher function. However the richness of protein structure and function goes far beyond the linear amino
acid sequence dictated by the genetic code. Multigene families, alternative splicing, coding polymorphisms, and post-
translational modifications, work together to create a rich variety of proteoforms, whose chemical diversity is the
foundation of the biological complexes and networks that control biology. The term “proteoforms” refers to the specific
molecular forms in which proteins are present in biological systems; only direct analysis of the proteoforms themselves
can reveal their structures, dynamics, and localizations in biological systems.

Methods

Remarkably, the dominant paradigm of proteomics research, “bottom-up” proteomics, is unable to identify proteoforms –
rather, proteins are enzymatically digested into peptides, which are then identified, serving as surrogates for the likely
presence of their parent proteins in the sample. This strategy destroys the information as to what form of the protein the
peptide represents, and thus the critical information needed to identify proteoforms is lost. This limitation of todays
technology provides a “grand challenge” to the scientific community, to devise new strategies that are able to
comprehensively and quantitatively reveal the full breadth of the proteome at the proteoform level.

Preliminary data (results)

In this presentation I will provide an overview of this interesting problem, along with a variety of new tools and
approaches that we and others are developing to address it. A key integrating concept is the “proteoform family”, the set
of all proteoforms that derive from a given gene. The description of the proteome of a given sample of interest may thus
be considered as a set of proteoform families, one for each gene in the genome. Identifying and quantifying the members
of each proteoform family comprises a new way of conceptualizing proteome analysis in complex systems. Developing
the technology to accomplish this, building a comprehensive atlas of proteoforms present in human systems, and
eventually deciphering the functional roles they play in normal and disease biology, comprise central elements in the
quest to understand human biology.

Please explain why your abstract is innovative for mass spectrometry?

Integrating four data streams (top-down (MS1 and MS2), intact mass (MS1 only), bottom-up, and transcriptomic (RNA-
Seq data, both short-read and long-read) to generate tailored sample-specific databases for proteoform analysis.

238
CHARACTERIZATION OF BACTERIAL TOXIN ACTIVITY USING PROTON
TRANSFER REACTION TOP-DOWN MASS SPECTROMETRY
Abstract ID: 341

Presenting author: Martial Rey, Mass Spectrometry for Biology Unit, Université de Paris, Institut Pasteur,
USR2000, CNRS

Introduction

Bacteria have developed several secretion systems capable of delivering toxins into prokaryotic or eukaryotic cells which
are directly involved in virulence, and/or participate to clear and colonize niches by killing bacterial competitors. Toxin
activities include nucleases, proteases and ADP-ribosyltransferases.
Here, we functionally characterize a 44 KDa toxin, TreTu, which is delivered by the Type VI secretion system of
Salmonella enterica Typhimurium. We show that TreTu inhibits protein synthesis by transferring an ADP-ribose group
from NAD+ on the elongation factor EF-Tu. By applying high-end top-down MS3 experiments using ETD or HCD
activation in combination with Proton Transfer Reaction we were able to localize the modification on the sidechain of the
conserved Asp29 residue of the P-loop of the G domain, essential to EF-Tu activity.

Methods

EF-Tu protein was incubated with the TreTu toxin and NAD+ to promote ADP ribosylation.

Non-modified EF-Tu protein and TreTutreated EF-Tu were nano-electrosprayed on an Eclipse Orbitrap mass
spectrometer. Intact mass spectra were recorded at low (15K) and high (240K) resolution with 50 µscan per scan.
Fragmentation spectra were obtained at high resolution (240K) using a combination of fragmentation techniques (CID,
HCD, ETD, UVPD) at different energies. Proton Transfer Reaction methods were applied to decongest the spectra and
improve the deconvolution process. Data were extracted with Extract algorithm and identified fragments were matched
with ProSight Liite on EF-Tu sequence.

Preliminary data (results)

We first analyzed the intact mass of EF-Tu before or after action of TreTu to understand the toxin activity. Upon contact
with TreTu, EF-Tu exhibit a complete shift of +541.6 Da, corresponding to an ADP-ribosylation.

To validate the identity of the modification we performed a bottom-up analysis, but no modification could be identified
although a 100 % coverage was obtained, suggesting the loss of the modification during the processing of the sample.
We then performed the fragmentation of the modified intact protein. Four specific fragments of 136.062, 250.096,
348.074 and 428.041 (MH+), corresponding respectively to adenine, adenosine, adenosine-phosphate and adenosine-
diphosphate ions, were observed upon HCD activation, validating the identity of the modification.

To localize precisely the modification in the EF-Tu sequence, fragmentation spectra were recorded with a combination of
activation methods and energies. Top-down MS2 data generated 130 explainable fragments covering 30 % of the 401
amino acids of EF-Tu. This data allowed us to narrow the localization of the ADP-ribose to an 8 amino acid long stretch
(Asp29 to Thr36). To improve the localization, we took advantage of the Proton Transfer Reaction capability of our
instrument. We identified 108 more fragments pushing the sequence coverage to 45%. This high coverage for a 44 kDa
protein allowed us to precisely localize the modification on the side chain of the conserved Asp29 residue. This amino
acid is in the P-loop of the G domain essential to EF-Tu activity, explaining the loss of function of the elongation factor
EF-Tu upon TreTu action.

Please explain why your abstract is innovative for mass spectrometry?

Identification and localization at the amino acid level of a PTM introduced on 44 kDa protein by a bacterial toxin using
MS3 top-down approaches combining multiple fragmentation techniques and PTR.

Co-authors:

Dukas Jurenas, Laboratoire d'Ingénierie des Systèmes Macromoléculaires, Institut de Microbiologie, Bioénergies et
Biotechnologie (IM2B), Aix-Marseille , UMR 7255, CNRS
Laurent Terrado, Laboratory of Molecular Microbiology and Structural Biochemistry, Institut de Biologie et Chimie des
Protéines, UMR 5086, CNRS, Université de Lyon
Eric Cascales, Laboratoire d'Ingénierie des Systèmes Macromoléculaires, Institut de Microbiologie, Bioénergies et

239
Biotechnologie (IM2B), Aix-Marseille , UMR 7255, CNRS
Julia Chamot-Rooke, Mass Spectrometry for Biology Unit, Université de Paris, Institut Pasteur, USR2000, CNRS

240
TOP-DOWN IDENTIFICATION OF PROTEIN-PROTEIN AND PROTEIN-
LIGAND COMPLEXES USING NATIVE AMBIENT MASS SPECTROMETRY
DIRECTLY FROM TISSUE
Abstract ID: 272

Presenting author: James Hughes, University of Birmingham

Introduction

Native ambient mass spectrometry (NAMS) permits the analysis of folded intact proteins and protein complexes
providing information on the stoichiometry of quaternary structure and allowing endogenous ligands to be identified. So
far, the extraction and identification of protein complexes by NAMS has been limited to lower molecular weight species.
Through the optimisation of instrument parameters including source dissociation voltage, source compensation voltage,
and instrument pressures we demonstrate the detection and identification of several intact protein complexes up to >145
kDa including dimeric, trimeric, and tetrameric species. High resolution mass spectrometry and/or deconvolution
following proton transfer charge reduction provide accurate (< 20 ppm) measurements of protein complex and protein
sub-unit masses. Collision induced dissociation facilitated determination of both complex stoichiometry and primary
structure.

Methods

Liquid extraction surface analysis (LESA) using a solvent system comprising 200 mM ammonium acetate and 0.5-2 x
critical micellar concentration of the detergent C8E4 was performed on thin (10 µm) sections of fresh frozen rat liver and
kidney tissue. The solvent extract was deposited into a well plate and transferred into gold coated borosilicate nano-ESI
emitters. Nano-ESI was performed on an Orbitrap Eclipse. Optimisation for the transmission of high m/z ions while
maintaining non-covalent interactions in the gas phase was required. Collision induced dissociation and source
fragmentation were utilised for top-down characterisation.

Preliminary data (results)

Intact protein-protein and protein-ligand complexes up to >145 kDa have been successfully characterised directly from
both rat kidney and rat liver tissue. Ion optics for the Orbitrap Eclipse were optimised for the transmission of large and
intact protein complexes, namely the source dissociation voltage and the source compensation voltage. The ion routing
multipole was operated at a high pressure to improve the trapping of high mass protein complexes. The stoichiometry of
protein complexes was determined by using collision induced dissociation to dissociate complexes into monomer sub-
units. Further activation provides sequence fragments which combined with subunit information allows the identity of the
protein and its native stoichiometry to be determined. The accurate mass of the protein complexes was determined by
high resolution mass spectrometry and/or using deconvolution following proton transfer charge reduction (PTCR) of the
precursor ion peak. α-enolase (94.175 kDa complex) was also detected in the kidney as a homodimeric complex with
four endogenous magnesium ions bound to the complex (two per subunit). Ornithine transcarbamylase (108.529 kDa
complex) was detected in the liver as a homotrimeric complex and the L-lactate dehydrogenase A chain (145.449 kDa
complex) was also detected in the liver as a homotetrameric complex.

Please explain why your abstract is innovative for mass spectrometry?

The extraction of protein complexes in excess of 145 kDa directly from tissue has been achieved and stoichiometry,
endogenous ligands, and primary structure have been determined using top-down mass spectrometry.

Co-authors:

Oliver Hale, University of Birmingham

Emma Sisley, University of Birmingham
Helen Cooper, University of Birmingham

241
CONFORMATION-SPECIFIC TOP-DOWN MASS SPECTROMETRY
Abstract ID: 202

Presenting author: Hannah Britt, Institute of Structural & Molecular Biology, University College London

Introduction

Native top-down mass spectrometry (MS) is a powerful tool for analysis of proteins and their complexes, characterising
backbone organisation and post-translational modifications (PTMs). Use of structure-dependent fragmentation methods,
notably electron capture dissociation (ECD), further enhances top-down MS by probing features of higher order protein
structure such as solvent accessibility and binding interfaces. To date, however, such experiments have sampled across
the entire conformational landscape, therefore the results reflect an average across multiple protein structures which
cannot be delineated. Here, we combine ECD fragmentation with cyclic ion mobility-mass spectrometry (cIM-MS) to
perform native top-down MS in a conformation-specific manner. To demonstrate the utility of this approach, the method
was applied to study the structures, dynamics, and ligand interaction interfaces of calmodulin.

Methods

Calmodulin samples were prepared in 10 mM ammonium acetate buffer at a concentration of 3 μM. Where peptide
ligands melittin or CaMKII peptide were added, these were present at a final concentration of 4.5 μM. Analyses were
performed on a SELECT SERIES Cyclic IMS QToF (Waters Corp.) fitted with a post-mobility ECD modification (e-MSion
Inc.). Samples were infused into the instrument using nano electrospray (nESI) capillaries prepared in house using a
Flaming-Brown P97 micropipette puller, and gold-coated with a Quorum Q150RS sputter coater. Data were processed
using UNIFI Scientific Information System (Waters Corp.).

Preliminary data (results)

Native MS using the high mass resolution of the cIM-MS instrument identified two charge state distributions of
calmodulin, attributed to native and partially denatured states. Each of these charge states exhibits a heterogeneous
array of calcium-binding from 0 to 4 Ca2+ additions. Within this context two distinct conformers of calmodulin were
identified though ion mobility separation, attributed by collision cross section calculations to be the reported compact
globular and extended dumbbell structures. To further confirm the identity of these structures, each conformation was
isolated and subjected to ECD fragmentation. This conformation-specific top-down analysis revealed distinct differences
in fragmentation patterns between the two structures, an observation which would not be possible without mobility
isolation. Furthermore, conformers were subjected to collision activation with simultaneous ECD in order to probe their
structural dynamics in detail. Fragmentation of these transitional intermediates identified their solvent exposed regions,
providing insight into the unfolding pathway of calmodulin. Comparable experiments were performed in the presence of
two target peptides known to bind calmodulin, with conformation-specific ECD confirming the interaction interfaces for
ligand binding.

Using the multi-functional and structurally flexible protein calmodulin along with its known binding partners, conformation-
specific top-down MS has been shown to be an excellent workflow to perform in-depth study of protein structure,
dynamics and ligand interactions. Having successfully developed this method, the future prospects now include its
application to the study of exciting biological applications, for example in the case of protein misfolding diseases.

Please explain why your abstract is innovative for mass spectrometry?

Application of native top-down mass spectrometry in a conformation-specific manner to characterise isolated protein or
complex structures and interaction interfaces.

Co-authors:

Aisha Ben-Younis, Institute of Structural & Molecular Biology, University College London
Nathanael Page, Institute of Structural & Molecular Biology, University College London, LGC Group
Konstantinos Thalassinos, Institute of Structural & Molecular Biology, University College London, Institute of Structural &
Molecular Biology, Birkbeck College

242
BENEFITS OF NATIVE TOP-DOWN ECD FRAGMENTATION FOR THE
SEQUENCING OF DIVERSE IMMUNOGLOBULIN FORMATS
Abstract ID: 92

Presenting author: Kelly Gallagher, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for
Biomolecular Research and Utrecht Institute of Pharmaceutical Sciences, Utrecht University, Padualaan 8, 3584
CH Utrecht, The Netherlands, Netherlands Proteomics Center, Padualaan 8, 3584 CH Utrecht, The Netherlands

Introduction

Native top-down mass spectrometry is gaining momentum for the characterization of immunoglobulins, providing an
effective method for the determination of the sequence and identity, as well as investigating diverse structural
modifications. Besides IgG, the lesser studied immunoglobulin classes, like IgA and IgM, are of particular interest as they
are an integral part of the adaptive immune response. However, these immunoglobulins pose significant challenges due
to their large mass, large number of isoforms and extensive glycosylation. Here we extend native top-down ECD
fragmentation, progressing from IgG, via IgA, to even sub-million Da IgM and IgG oligomeric immunoglobulins.

Methods

Immunoglobulins were analyzed with a Q-Exactive Ultra-High Mass Range (UHMR) Orbitrap instrument equipped with
an ECD cell developed by eMSion. Samples included different IgGs, glycosylated Fab-fragments, IgAs, a J-chain
coupled IgM pentamer (~935 kDa) and an engineered IgG hexamer (~895 kDa). All samples were subjected to
fragmentation by ECD. For the oligomeric immunoglobulins, mass spectrometry conditions were optimized for
transmission and recording of these sub-million Da species while suppressing the third harmonic peak, thereby,
improving the detection of ECD fragments within the lower m/z range. Dedicated software was developed to analyze and
annotate the resulting ECD spectra.

Preliminary data (results)

Here we describe the optimization of ECD for the native top-down analysis of IgM and (IgG)6. Compared to experiments
on monomeric immunoglobulins, here we also observe many highly abundant ‘ECnoD’ species. However, for these large
oligomers, a variety of fragment ions could be generated, informative for sequencing. Interestingly, backbone cleavages
of both the HC and LC occur only outside of the disulfide loops that maintain the Fab structure, creating sequence tags of
nearly exclusively c-ions spanning from the crucial hypervariable complementary determining region 3 (CDR3) into the
constant region. Observed fragment ions cover exactly the same region as observed for monomeric IgG and IgA. To
further test this, IgM and IgG monomer antibodies with sequences matching those of the oligomers used above were
likewise subjected to ECD. Comparison of the ECD fragmentation spectra revealed a correlation of at least 70% with the
observed fragmentation spectra of the of IgM and (IgG)6 oligomers. For a glycosylated recombinant IgG Fab, the ECD
data reveal that the glycans remain intact, and primarily backbone fragmentation are observed. The presence of the
glycan allows can be utilized to aid analysis and permit disentangling of which fragments originate from the heavy or the
light chain.

The culmination of these analysis focus on the various benefits of applying ECD fragmentation for the sequencing and
study of diverse immunoglobulin formats.

Please explain why your abstract is innovative for mass spectrometry?

We extend the feasible range of native top-down proteomics using ECD to about 1 MDa immunoglobulin oligomers
producing informative sequence tags within the variable regions facilitating de novo sequencing.

Co-authors:

Jean-Francois Greisch , Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research
and Utrecht Institute of Pharmaceutical Sciences, Utrecht University, Padualaan 8, 3584 CH Utrecht, The Netherlands,
Netherlands Proteomics Center, Padualaan 8, 3584 CH Utrecht, The Netherlands
Maurits den Boer, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research and
Utrecht Institute of Pharmaceutical Sciences, Utrecht University, Padualaan 8, 3584 CH Utrecht, The Netherlands,
Netherlands Proteomics Center, Padualaan 8, 3584 CH Utrecht, The Netherlands
Szu-Hsueh Lai, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research and Utrecht
Institute of Pharmaceutical Sciences, Utrecht University, Padualaan 8, 3584 CH Utrecht, The Netherlands, Netherlands
Proteomics Center, Padualaan 8, 3584 CH Utrecht, The Netherlands
Shengya Cao, Genentech, San Francisco, USA

243
Albert Bondt, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research and Utrecht
Institute of Pharmaceutical Sciences, Utrecht University, Padualaan 8, 3584 CH Utrecht, The Netherlands, Netherlands
Proteomics Center, Padualaan 8, 3584 CH Utrecht, The Netherlands
Jan Commandeur, MSVision, Televisieweg 40, 1322 AM Almere, The Netherlands
Wendy Sandoval, Genentech, San Francisco, USA
Albert Heck, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research and Utrecht
Institute of Pharmaceutical Sciences, Utrecht University, Padualaan 8, 3584 CH Utrecht, The Netherlands, Netherlands
Proteomics Center, Padualaan 8, 3584 CH Utrecht, The Netherlands

Comparison of ECD fragments from (IgM)5J and monomeric IgM.

244
TWO-DIMENSIONAL MASS SPECTROMETRY AND TOP-DOWN
PROTEOMICS: POST-TRANSLATIONAL MODIFICATIONS AND PROTEIN
CONFORMATIONS
Abstract ID: 120

Presenting author: Maria Van Agthoven, BIOCEV

Introduction

Two-dimensional mass spectrometry (2DMS) is a method for tandem mass spectrometry that relies on ion radius
modulation instead of ion isolation to correlate between precursor and fragment ion peaks. 2D mass spectra show all the
fragmentation patterns of the analytes in a sample. Signal multiplexing yields higher signal-to-noise ratios and therefore
more complete sequence coverage (e.g. for biomolecules) than standard tandem mass spectrometry. Modifications can
easily be assigned and located visually with precursor ion scans and dissociation lines. The development of narrowband
mode 2DMS and phase correction for absorption mode 2DMS has enabled ultra-accurate precursor-fragment correlation.

Methods

The potential of 2DMS has been tested for both top-down and bottom up approaches. 2DMS has also successfully been
used for label-free relative quantification of modified synthetic histone peptides. Histone proteins were selected in order
to determine the extent of posttranslational modifications and thus interpret the histone code. Since Fast photochemical
oxidation of proteins (FPOP) is a very successful covalent labelling method to probe tertiary and quaternary structures of
proteins, e.g. to study protein-ligand binding, it has been selected as a model case for top-down with 2DMS.

Preliminary data (results)

We use the accuracy of the precursor-fragment correlation to identify and locate the modifications in the sequence and
we use fragment ion abundances for label-free relative quantification.

Please explain why your abstract is innovative for mass spectrometry?

We show that 2DMS is a useful tool for the analysis of histone modifications, and that FPOP combined with 2DMS yields
accurate information on protein tertiary and quaternary structure.

Co-authors:

Marek Polak, BIOCEV

Michael Palasser, University of Innsbruck
Marc-André Delsuc, IGBMC, CASC4DE
Kathrin Breuker, University of Innsbruck
Petr Novak, BIOCEV

245
Session: LS-10 Imaging MS - applications in Life Science & Health -
Session A

TOWARD IN VIVO INTRAOPERATIVE MASS SPECTROMETRY IMAGING

Abstract ID: 797

Presenting author: Isabelle Fournier, Univ. Lille, Inserm, CHU Lille, U1192 - Protéomique Réponse Inflammatoire
Spectrométrie de Masse - PRISM, F-59000 Lille, France

Introduction

Ambient Ionization Mass Spectrometry (AIMS) has paved the way for novel applications and applied for MS imaging
(MSI) they have demonstrated a great potential for health applications. Dedicated AIMS such as Water-Assisted Laser
Desorption/Ionization (WALDI) can even perform in vivo MS, advancing clinical applications to in-man use. So far, there
is an unmet need in the field of surgery; and despite advanced robotic solutions are helping to perform a more precise
surgery, the decision making remains solely based on the surgeon's judgment. Retrieving molecular data in real-time by
MS is the best match to meet this need. This requires the development of in vivo MSI to screen areas of interest with a
specific acquisition so that the topography of the tissue can be considered.

Methods

The WALDI-MS is based on the resonant excitation of endogenous water to perform micro-invasive sampling of tissues
(Figure 1). The WALD-MS was coupled to a stiff 6D-axis robotic arm of 5µm accuracy by attaching the handpiece of the
laser fiber to the robotic arm through an adaptor equipped with a distance sensor. This setup enables to measure in a
single acquisition the sample topography and acquire the MSI data adapting the distance between the laser and sample.
The topography is reconstructed in real-time and by the end of the acquisition, the molecular images are plotted back
onto the topology maps.

Preliminary data (results)

We initially tested the ability to perform in vivo image acquisition by mounting the laser on a bracket to be placed above
volunteers. The system enabled automatic adjustment of the laser probe above the surface and this first setup was
tested during a clinical trial realized on volunteers with skin pathologies (Figure 1). Then the WALDI-MS laser probe was
mounted onto a 6-axis stiff robotic of high accuracy and a distance sensor was added to the setup to measure the
distance to the sample. The measurement from the distance sensor is used both to get the sample topography and
adjust the laser to keep it in focus. An acquisition software was developed under MATLAB to control the image
acquisition. First sample topography and MSI data were recorded separately and fused post-acquisition to plot the
molecular distribution onto the topographic maps. In a second step, the hardware was modified to enable the acquisition
of both the topography and the MSI data in a single run integrating a real-time reconstruction of the topography. The
system was first evaluated from tissue sections. The laser spot size is 500 µm but using oversampling, we imaged down
to 200 µm spatial resolution were produced with a speed up to 1.6 pixels/s. First in vivo experiments were performed
from a volunteer finger. Then, mice models developing spontaneously mammary tumours were imaged enabling
classification models to be built and then reinterrogated in real-time to get automated image annotation indicating cancer
and healthy tissue areas.

Please explain why your abstract is innovative for mass spectrometry?

Development of in vivo Mass Spectrometry Imaging for intraoperative analysis

246
Figure 1: WALDI-MS system and its use in vivo

247
N-GLYCOSYLATION AND ITS ROLE IN THE MALIGNANT
TRANSFORMATION OF ADENOMA TO EARLY-STAGE COLORECTAL
CANCER
Abstract ID: 231

Presenting author: Bram Heijs, Center for Proteomics and Metabolomics, Leiden University Medical Center

Introduction

Population-based screening for colorectal cancer (CRC) is performed in the Netherlands since 2014 and led to a
substantial increase in the detection and diagnosis of early (T1) CRC cases. The development of CRC is a multistep
process, starting from normal epithelium, to benign adenoma, and ultimately malignant carcinoma when invading into the
submucosal regions. N-linked glycosylation is known to play a key role in the molecular processes underlying cancer
progression, including cell invasion, angiogenesis and immune modulation. To better understand the disease
progression from normal epithelium to adenoma, and ultimately carcinoma we applied mass spectrometry imaging (MSI)
to study the N-glycome of the malignant transformation in the colon.

Methods

A cohort of 21 formalin-fixed, paraffin-embedded human CRC biopsies was obtained from the Department of Surgery at
the Leiden University Medical Center. Tissues were sectioned (6 µm thickness) and mounted on indium-tin-oxide-coated
glass slides. After paraffin removal, rehydration, and linkage-specific sialic acid derivatization, N-glycans were
released in-situ by PNGase-F and analyzed on by matrix-assisted laser desorption/ionization mass spectrometry imaging
(MALDI-MSI). Post-MSI histopathological staining was performed and normal epithelium (NE), low-grade dysplasia
(LGD), high-grade dysplasia (HGD) and carcinoma (CA) were annotated. Identically treated consecutive sections were
used for N-glycan identification by capillary electrophoresis electrospray ionization tandem mass spectrometry (CE-ESI-
MS/MS).

Preliminary data (results)

Following data curation, a total of 58 N-glycans were detected consistently in all tissues. Based on these glycans, relative
intensities and glycosylation traits were calculated for the morphology-based data interrogation. We observed that high-
mannose glycans (HM) were higher in NE than in CA (FCCA/NE = 1,53), which is in line with literature where different
cancer types including more advanced CRC stages show an upregulation of HM. Surprisingly, we observed an even
higher abundance of HM glycans in dysplastic lesions compared to NE (FC HGD/NE = 2,33), and HM was significantly lower
in CA compared to HGD (FCCA/HGD = 0,75). This specific pattern was found persistent in all individual HM glycans. It is
often hypothesized that the increase in HM glycans is related to the increased proliferation of cells in neoplastic tissue, in
combination with a reduced or incomplete N-glycan maturation. To test this hypothesis we performed
immunohistochemistry of Ki-67, a known proliferation marker, on consecutive sections. It was found that proliferation and
HM glycans followed similar patterns. Additionally, an increase in the relative abundance of α2,3-linked sialylation in the
microenvironment surrounding the invasion site was observed. This observations corroborates our previous findings in
T2 CRC. Finally and surprisingly, a relatively low abundance of tri- and tetraantennary complex-type glycans was
observed in CA. This could be explained by the incomplete synthesis principle as postulated by Hakomori and Kannagi
commonly occurring in early cancer lesions.

Please explain why your abstract is innovative for mass spectrometry?

The integrated on-tissue and off-tissue MS analysis of CRC glycans proved particularly valuable to unravel the spatial
distribution of structurally defined glycans.

Co-authors:

Fanny Boyaval, Department of Pathology, Leiden University Medical Center, Center for Proteomics and Metabolomics,
Leiden University Medical Center
Hans Dalebout, Center for Proteomics and Metabolomics, Leiden University Medical Center
René van Zeijl, Center for Proteomics and Metabolomics, Leiden University Medical Center
Wenjun Wang, Center for Proteomics and Metabolomics, Leiden University Medical Center
Guinevere Lageveen-Kammeijer, Center for Proteomics and Metabolomics, Leiden University Medical Center
Manfred Wuhrer, Center for Proteomics and Metabolomics, Leiden University Medical Center
Hans Morreau, Department of Pathology, Leiden University Medical Center

248
UNRAVELLING AMYLOID BETA PLAQUE PATHOLOGY ASSOCIATED
LIPID DYNAMICS IN VARIOUS ALZHEIMER’S DISEASE MOUSE MODELS
Abstract ID: 371

Presenting author: Junyue Ge, University of Gothenburg

Introduction

Amyloid-β (Aβ) plaque formation is one of the main hallmarks of Alzheimer’s disease (AD). Over the years, mounting
evidence that changes in the neuronal lipid biochemistry are associated with protein misfolding and aggregation into
mature, neurotoxic plaques. However, the overall molecular events underlying neurodegenerative Aβ pathology in AD
are largely unknown. Therefore, more efforts are needed to delineate the changes of neuronal lipids correlated to
different amyloid aggregates in the brains of various AD mouse models. The primary goal of this project is to employ
advanced chemical imaging to unravel the exact lipid-Aβ plaque interactions in various AD model systems.

Methods

Here, we employed a multimodal imaging paradigm combining MALDI-IMS and fluorescent amyloid staining. Tri-modal
MALDI-IMS under negative- and positive ion mode lipid analysis and subsequent peptide/protein ion imaging were
performed at 10 μm spatial resolution on the same tissue section from tgAPP Swe, tgAPPArcSwe and tgAPPUppSwe mouse
brains. Furthermore, employed LCO-based, hyperspectral, fluorescent amyloid microscopy paradigm provided the
insights into degree of Aβ aggregation associated with plaque polymorphism. Multivariate statistical analysis was used to
interrogate the IMS datasets and region-specific differences in Aβ peptide pattern were correlated with changes in lipid
distributions revealed by MALDI-IMS lipid analysis.

Preliminary data (results)

MALDI-IMS and fluorescent amyloid staining revealed chemical features associated with heterogeneous distributions of
different individual Aβ deposits. For the tgAPPSwe mouse model, it was primarily the wild type peptides that aggregated in
Aβ plaques including both diffuse and core formation. For the tgAPP ArcSwe mouse model, there were massive, cored
plaques containing primarily the C-terminally truncated peptides. Meanwhile, there were smaller and diffused plaques
containing primarily AβUpp peptides in the tgAPPUppSwe mouse model. Furthermore, heterogeneous distributions of lipids
were revealed by high-spatial resolution MALDI-IMS. GM1 aggregated at the core region of Aβ plaques, while
arachidonic acid (AA) conjugated phosphatidylinositols (PI) and their corresponding degradation product,
lysophosphatidylinositols (LPI) were located to the periphery of the plaques in the tgAPPSwe and tgAPPArcSwe mouse
models. On the contrary, GM2 and GM3 were localized to the periphery of the plaques in the tgAPP Swe and
tgAPPArcSwe mouse models. For the tgAPPUppSwe mouse model, the depletion of sulfatides and the increase of the
ceramides showed plaques-associated pattern. Also, GM2 and GM3 were localized to the plaques, while GM1 and GD1
showed no increase at the plaques.

Please explain why your abstract is innovative for mass spectrometry?

MALDI-IMS and fluorescent amyloid staining were monitored to delineate lipid-protein interactions associated with Aβ
plaque pathology in various mouse models of AD.

Co-authors:

Srinivas Kouterapu, University of Gothenburg

Jörg Hanrieder, University of Gothenburg

249
MALDI images of ganglioside and Aβ peptide.

Hyperspectral data of plaques in different AD mouse models.

250
ORBITRAP-SIMS IMAGING REVEALS CELL-TYPE SPECIFIC
LOCALIZATION OF TOMATO SECONDARY METABOLITES AND ALLOWS
PUTATIVE METABOLITE ANNOTATION
Abstract ID: 609

Presenting author: Uwe Heinig, Weizmann Institute of Science

Introduction

Metabolite mass spectrometry imaging has become a valuable tool for localization of plant primary and secondary
metabolites within a tissue section. Using Matrix-assisted laser desorption ionization mass spectrometry (MALDI-MS)
imaging, the lateral distribution of various compounds in tomato tissues was determined during development and the
change in metabolites due to genetic modifications examined. The lateral resolution of commercially available MALDI-MS
imaging instruments of at best 10 µm, allows metabolites to be assigned to specific tissues, such as seed, embryo or fruit
skin, but not on a (sub)cellular level. In order to overcome this limitation and better understand processes like production,
storage, and transport of secondary metabolites in specific cell types an approach employing secondary ion mass
spectrometry (SIMS) using an OrbitrapTM analyzer was evaluated.

Methods

SIMS imaging was performed using a recently developed Hybrid SIMS instrument (IONTOF GmbH), which allows high
mass resolution imaging of metabolites and other biological molecules at µm lateral resolution by employing an Orbitrap
detector (Q Exactive HF, Thermo-Fisher Scientific) and 20 keV Ar gas clusers. The TOF-SIMS part of the instrument can
be operated using Bi-clusters, allowing even sub-µm lateral resolution, but usually accompanied by intense molecule
fragmentation.

Preliminary data (results)

Here we describe the Orbitrap-SIMS analysis of tomato secondary metabolites, like steroidal glycoalkaloids or
flavonoids, inorganic ions, and lipids/fatty acids. Using the Argon gas cluster ion beam in combination with the Orbitrap
mass analyzer high mass resolution images were acquired, while TOF-SIMS mode with the Bi- ion beam was used to
achieve sub-µm lateral resolution at the moderate mass resolution of the TOF analyzer (Figure 1).

We detected and annotated 10s of compounds/ions in both positive and negative ionization mode, including inorganic
ions, lipids, fatty acids, primary metabolites and secondary metabolites at single cell resolution.

Furthermore, we explored the possibility to identify characteristic fragment ions of secondary metabolites directly from
the Orbitrap-SIMS imaging run. In contrast to MALDI that does not cause metabolite fragmentation, SIMS using the Bi
ion beam leads to strong fragmentation, while using the Argon gas cluster ion beam allows the parallel detection of
molecular ions, adducts and characteristic fragments. We analyzed the obtained images using correlation analysis of ion
distributions and manual identification of co-localized masses as fragments of steroidal glycoalkaloid molecular ions.
Thereby we were able to distinguish between differentially localized molecular ions and assign their characteristic
fragments accordingly. This allowed the putative metabolite identification directly from imaging data without performing
MS/MS analysis or complementary analysis methods, like LC/MS for compound assignment.

Please explain why your abstract is innovative for mass spectrometry?

We present here mass spectrometry imaging of plant metabolites using Orbitrap-SIMS at (sub)cellular resolution with
high mass resolution & metabolite identification partially using inherent fragmentation.

Co-authors:

Julia Zakel, IONTOF GmbH

Alexander Pirkl, IONTOF GmbH
Yonghui Dong, Weizmann Institute of Science
Matthias Kleine-Boymann, IONTOF GmbH
Yana Kazachkova, Weizmann Institute of Science
Asaph Aharoni, Weizmann Institute of Science

251
Hybrid-SIMS (A), TOF-SIMS (B) and Orbitrap-SIMS (C) example scan modes.

252
ISOMER-RESOLVED LIPID IMAGING OF BREAST CANCER USING HIGH-
PRESSURE OZONE-INDUCED DISSOCIATION MASS SPECTROMETRY
IMAGING
Abstract ID: 460

Presenting author: Britt Claes, Maastricht MultiModal Molecular Imaging (M4i) Institute, Maastricht University

Introduction

Breast cancer presents with intra-tumoral spatial heterogeneity and has unique molecular characteristics that contribute
to the aggressiveness of the cancer. Mass spectrometry imaging (MSI) detects, localizes, and identifies molecular
distributions of compounds in tissue. In previous work, we showed a correlation between specific phosphatidylcholine
(PC) species in necrotic and hypoxic regions1, but only gross lipid compositions were determined. Biological actuators
are structure-dependent and current MSI approaches, even at high mass resolving power, remain ambiguous to
structural alterations introduced by isomeric lipids. Here, we identified distinct isomer-specific molecular signatures
through the introduction of lipid isomer separation with ozone-induced dissociation (OzID)-MSI. This helped to improve
the understanding of the molecular networks that define the tumor microenvironment.

Methods

Fresh frozen MDA-MB-231 and SUM159 breast tumor xenografts were cryo-sectioned (Leica, 10 µm, -16°C) and thaw-
mounted onto indium tin oxide (ITO) slides. Slides were spray-coated with sodium acetate with a TM-sprayer, followed by
2,5-DHA matrix sublimation (HTX technologies). OzID data were acquired on a SYNAPT G2-Si mass spectrometer
(Waters) combined with a prototype µMALDI source. Collision-induced dissociation (CID)-OzID data were acquired on an
Orbitrap Elite mass spectrometer (Thermo Fisher Scientific) coupled to a MALDI/ESI source (Spectroglyph). OzID was
conducted by mixing high concentration ozone (~18% O3 in O2) with the buffer gas.

Preliminary data (results)

The ozonolysis step in our setup enabled imaging of isomeric lipids in breast tumor tissue sections. Four isomers were
identified for PC 34:1, as fragments of both sn-isomers 16:0/18:1 and 18:1/16:0 were detected, as well as the double
bond isomers n-7 and n-9. Distributions of individual isomers were compared against the total abundance of product ions
from all isomers to generate fractional distribution images (FDI) to visualize relative changes between isomers.
Interestingly, the intensity of PC 34:1n-7 showed higher abundance in the necrotic and possibly hypoxic regions
compared to viable regions. Four sn-isomers were detected for PC 32:1, which were 16:0/16:1, 16:1/16:0, 14:0/18:1, and
18:1/14:0, respectively. Here, the FDI of 14:0/18:1 over 14:0_18:1 showed 14:0/18:1 to be more abundant in necrotic
regions than viable regions. Similar to PC 32:1, four sn-positional isomers were detected for PC 36:1, namely 18:0/18:1,
18:1/18:0, 16:0/20:1, and 20:1/16:0. Surprisingly, the FDI of 18:0/18:1 over 18:0_18:1 was homogenously distributed
across the whole tumor section and consisted of ~90% of the 18:0/18:1 isomer. Further validation is currently being
performed by histological staining and immunohistochemistry. Additionally, enzymes involved in phospholipid
remodeling, such as phospholipase A2 and lysophosphatidylcholine acyltransferase 1, will be targeted to correlate with
the OzID-MSI data.

Please explain why your abstract is innovative for mass spectrometry?

The introduction of lipid isomer separation with OzID-MSI enables the identification of distinct isomer-specific molecular
signatures. This generates more knowledge on the molecular networks that drive breast cancer progression.

Co-authors:

Annet Duivenvoorden, Department of Surgery, NUTRIM School of Nutrition and Translational Research in Metabolism,
Maastricht University
Caitlin Tressler, The Russell H. Morgan Department of Radiology and Radiological Science, Division of Cancer Imaging
Research, The Johns Hopkins School of Medicine
Ethan Yang, The Russell H. Morgan Department of Radiology and Radiological Science, Division of Cancer Imaging
Research, The Johns Hopkins School of Medicine
Kanchan Sonkar, The Russell H. Morgan Department of Radiology and Radiological Science, Division of Cancer Imaging
Research, The Johns Hopkins School of Medicine
Shane Ellis, Maastricht MultiModal Molecular Imaging (M4i) Institute, Maastricht University, Molecular Horizons and
School of Chemistry and Molecular Bioscience, University of Wollongong, Illawarra Health and Medical Research
Institute
Kristine Glunde, The Russell H. Morgan Department of Radiology and Radiological Science, Division of Cancer Imaging

253
Research, The Johns Hopkins School of Medicine, The Sidney Kimmel Comprehensive Cancer Center, The Johns
Hopkins School of Medicine, Department of Biological Chemistry, The Johns Hopkins School of Medicine
Ron Heeren, Maastricht MultiModal Molecular Imaging (M4i) Institute, Maastricht University

H&E and OzID-MSI of breast xenograft show similar distributions.

254
AP-SMALDI IMAGING OF COMPOUNDS AND METABOLITES IN
PARASITES AND INFECTED HOSTS
Abstract ID: 735

Presenting author: Carolin Morawietz, Institute of Inorganic and Analytical Chemistry, Justus Liebig University,
Giessen, Germany

Introduction

Host-parasite systems are characterized by complex molecular interaction profiles. Also, the involved, interacting tissues
of host, guest and infection/inflammation-based reactions such as encapsulation or granuloma formation exhibit distinct
tissue-specific molecular profiles. Atmospheric-pressure scanning microprobe MALDI mass spectrometry imaging (AP-
SMALDI MSI) on Orbitrap mass spectrometers allows to characterize such profiles with high spatial resolution and high
mass accuracy on the level of lipids, peptides, drugs and metabolites. [1] Tissue profiles and interaction profiles
of Schistosoma mansoni worms, causing the neglected tropical disease (NTD) schistosomiasis (bilharziosis) and
of Fasciola hepatica (causing fasciolosis in cattle and humans) were determined by AP-SMALDI MSI at 5 µm lateral
resolution, providing detailed information about granuloma formation and nutrition supply of schistosoma eggs in infected
liver tissue.

Methods

AP-SMALDI systems for high-performance imaging of planar and non-planar (three-dimensional) surfaces were coupled
to Orbitrap Exploris mass spectrometers. Besides investigating tissue sections, intact species were mass
spectrometrically imaged by means of 3D-surface imaging [3] and coaxial laser and ion beams normal to the sample
surface [2].

Preliminary data (results)

Unicellular and multicellular parasites were studied, regarding both, their anatomical molecular structure and their
chemical interactions with the environment and with their hosts. 5-µm pixel size at 5-µm laser spot size was employed in
2D flat-tissue scanning mode and, for investigating intact worms and worm couples, in 3D-surface topography scanning
mode.

Intact bisexual Schistosoma mansoni worm couples were mass spectrometrically imaged showing chemical
communication between animals on their surface (the so-called tegument) as well as organ-specific and sex-specific
markers in tissue sections. Uptake of anthelminthic drugs was characterized in treated worms in a time profile, giving
hints about efficient and less-efficient oral or transdermal uptake routes. Also Fasciola flatworms were investigated [4]
regarding the uptake of anthelminthic drugs including triclabendazole and imatinib (as an example for drug repurposing
studies).

Infection of host liver with Schistosoma eggs was investigated to characterize encapsulation, granuloma formation and
inflammation reactions [5]. Markers of mechanistically relevant structures in the tissue and in eggs were determined, also
giving hints about nutrition supply of the eggs with essential lipid species inside the granuloma.

[1] B. Spengler B, Anal Chem 2015, 87, 64−82. [2] M. Kompauer et al., Nature Methods 2017, 14, 90-96. [3] M.
Kompauer et al., Nature Methods 2017, 14, 1156-1158. [4] C. Morawietz et al., Parasitology Research2022, DOI
10.1007/s00436-021-07388-1 [5] K. Wiedemann et al., Anal and Bioanal. Chem. 2022, under revision.

Please explain why your abstract is innovative for mass spectrometry?

Infections with parasitic worms and their eggs were investigated using high-resolution MS imaging of tissue sections and
intact worms.

Co-authors:

Stefanie Gerbig, Institute of Inorganic and Analytical Chemistry, Justus Liebig University, Giessen, Germany
Katja Wiedemann, Institute of Inorganic and Analytical Chemistry, Justus Liebig University, Giessen, Germany
Annika Mokosch, Institute of Inorganic and Analytical Chemistry, Justus Liebig University, Giessen, Germany
Domenic Dreisbach, Institute of Inorganic and Analytical Chemistry, Justus Liebig University, Giessen, Germany
Kerstin Strupat, Thermo Fisher Scientific (Bremen) GmbH, Bremen, Germany
Martin Roderfeld, Institute of Parasitology, Justus Liebig University, Giessen, Germany

255
Elke Roeb, Institute of Parasitology, Justus Liebig University, Giessen, Germany
Christoph G. Grevelding, Institute of Parasitology, Justus Liebig University, Giessen, Germany
Simone Haeberlein, Institute of Parasitology, Justus Liebig University, Giessen, Germany
Bernhard Spengler, Institute of Inorganic and Analytical Chemistry, Justus Liebig University, Giessen, Germany
Thomas Quack, Institute of Parasitology, Justus Liebig University, Giessen, Germany

MS imaging of anthelmintic drug in Fasciola tissue section

256
High-resolution imaging of Schistosoma mansoni worm couple

257
 Wednesday 31 August 2022: 15:30 – 17:30
Session: HC-2 MS in the Netherlands (NVMS session)

MASS SPECTROMETRY IN BIOMARKER AND BIOPHARMACEUTICAL

RESEARCH
Abstract ID: 815

Presenting author: Rainer Bischoff, University of Groningen, Department of Analytical Biochemistry

Introduction

The analysis of proteins by mass spectrometry has greatly contributed to research on disease biomarkers and
biopharmaceuticals. I will address these topics based on recent results concerning the soluble Receptor for Advanced
Glycation Endproducts (sRAGE), a biomarker for Chronic Obstructive Pulmonary Disease (COPD), and the
biotransformation of the monoclonal antibodies Trastuzumab and Pertuzumab in breast cancer patients.

COPD is currently the 3rd cause of disease-related death worldwide. Effective treatment for COPD is lacking partially due
to a lack of suitable biomarkers to assess disease heterogeneity.

Biopharmaceuticals are widely used to treat diseases that are not effectively treatable. While biopharmaceuticals and
notably monoclonal antibodies are extensively characterized during and after production, little is known about their fate in
the human body after drug administration.

Methods

sRAGE was enriched from plasma or serum using various approaches comprising antibodies, affimers or charge-based
principles. sRAGE was quantified by Selected Reaction Monitoring (SRM) LC-MS/MS and the results compared to
commercial ligand binding assays (LBAs). All procedures were validated according to international guidelines.

Trastuzumab and Pertuzumab were enriched from plasma using selective affimers, their proteoforms separated by
cation-exchange chromatography with pH-gradient elution and the proteoforms characterized by LC-MS/MS-based
peptide mapping. Affinity to Her2 and Fc-gamma receptors was assessed by surface plasma resonance (SPR).

Preliminary data (results)

We developed sensitive methods to quantify sRAGE in plasma and serum at the low pM level using LC-MS/MS in the
SRM mode. We notably focused on sample preparation by comparing enrichment with antibodies to enrichment with
specific affimers that were selected by phage display and to charge-based, ion-exchange solid-phase extraction (SPE).
All three methods were validated according to international guidelines and compared to a commercial immunoassay as
well as to other assay principles (e.g. an aptamer-based LBA). The levels of sRAGE were interpreted in the context of
emphysema development as assessed by High-Resolution CT (HRCT) in clinical cohorts showing that increasing
sRAGE levels correlate with the development of centrilobular emphysema. This new finding may open the way to patient
stratification and targeted therapeutic interventions.

The analysis of Trastuzumab and Pertuzumab after stress testing at 37 oC in Phosphate-Buffered Saline (PBS) showed
the appearance of multiple proteoforms due to changes in surface charge (charge variants). Detailed analysis related
these changes to the deamidation of Asn residues in the Complementarity Determining Regions (CDRs) of Trastuzumab
as well as in the Fc-part of Pertuzumab. Analysis of plasma samples from Her2-positive breast cancer patients after
enrichment produced similar charge-variant-profiles. These results indicate that in vitro stress testing is an appropriate
way of predicting charge-based biotransformations occurring in patients.

Please explain why your abstract is innovative for mass spectrometry?

Mass spectrometry-based analysis of proteins in complex biological samples requires highly selective enrichment
strategies as well as powerful separation of proteoforms.

258
sRAGE after enrichment with affimers (A) or antibodies (B)

Charge state profile of trastuzumab

259
MS: THE INDISPENSIBLE DRIVER IN INDUSTRIAL FOOD AND BIOTECH
RESEARCH
Abstract ID: 944

Presenting author: Maurien Olsthoorn, DSM

Introduction

DSM is a company active in Health, Nutrition and Bioscience. For the development and biotechnological production of
ingredients in food such as enzymes, cultures, hydrocolloids, plant-based proteins, yeast extracts, mass spectrometry is
an indispensable driver. For supporting the protein transition towards more sustainable food production, development of
novel ingredients requires deep molecular and physical-chemical insights into the food matrix to create plant-based foods
while maintaining good taste, texture and health.

Methods

* Untargeted flavor metabolomics workflow using SPME-GC-MS and LC-MS

* Flow-injection Electron Activated Dissociation MS/MS for lipids

* LC-MS-based intact enzyme molecular and activity analysis

Preliminary data (results)

Therefore, we have set-up and applied an untargeted metabolomics workflow using SPME-GC-MS to map flavors from
meat and plant-based burgers to find leads that can close the flavor gap. Next to that, we are developing a targeted and
untargeted metabolomics workflow using LC-MS to broaden the coverage to non-volatile flavor compounds.

For the structural elucidation of lipids we are developing Electron Activated Dissociation (EAD) MS/MS combined with
flow-injection, which allows rapid lipid class determination including double bond- and Sn-positioning in a single
experiment.

In addition, MS-based proteomics and metabolomics are drivers to steer microbial strain development for ingredient
bioproduction by monitoring the proteome expression and metabolites from the pathways of interest. To provide these
insights at the right pace of the strain optimization, requires not only rapid MS methods but tuning of the whole workflow
including sample preparation, data handling and visualization.

Finally, for studying structure-function relationship of our food enzymes under representative food conditions, we are
developing MS methods to rapidly screen for enzyme activity while in parallel performing molecular analysis of the
different enzyme variants. In summary, mass spectrometry is at the heart of our industrial biotech and food innovation.

Please explain why your abstract is innovative for mass spectrometry?

Flow-injection Electron Activated Dissociation MS/MS for rapid lipid class determination including double bond- and Sn-
positioning

Co-authors:

Rob van der Hoeven, DSM

Nicolas Abello, DSM
Brenda Ammerlaan, DSM
Andre Vente, DSM
Erwin Kaal, DSM
Leon Coulier, DSM

260
MASS SPECTROMETRIC CHARACTERIZATION STRATEGIES OF
GLYCOENGINEERED BACTERIAL VACCINES CARRYING O-ANTIGEN
POLYSACCHARIDES
Abstract ID: 490

Presenting author: Agnes Hipgrave Ederveen, Center for Proteomics and Metabolomics, Leiden University
Medical Center

Introduction

As a common bacterial pathogen in humans, E. coli can cause a broad range of infections with possible detrimental
effects on health. Vaccine development is essential as multi-drug resistance in bacterial infections is a rising concern.
Bacterial vaccines are often carbohydrate-based, and recombinantly produced proteins carrying versatile O-antigen
glycosylation are promising glycoconjugate vaccine candidates.The immunogenic O-antigen component of
lipopolysaccharides (LPS) is exposed on the bacterial cell surface and can be enzymatically transferred to specifically
engineered protein carriers. Since O-antigen polysaccharides are typically polydisperse and the recombinantly
expressed carrier proteins may have multiple glycosylation sites, the analysis of these highly heterogeneous complex
molecules is not trivial.

Methods

Here, we present the comprehensive characterization of glycoconjugate vaccine glycosylation by combining two
approaches: 1. Dopant-enriched nitrogen (DEN) gas-assisted bottom-up approach with reversed-phase liquid
chromatography-mass spectrometry (LC-MS) provided site-specific glycosylation information. The effect of acetonitrile
DEN gas, as well as plain nitrogen nebulization and ambient air, were evaluated. 2. Using a complementary intact mass
approach based on capillary electrophoresis (CE) coupled to MS further information on the overall number of repeat units
(RU) per protein was obtained.

Preliminary data (results)

Monitoring and comparing glycoconjugate vaccine glycosylation between batches is pivotal for assessing the integrity of
glycoconjugate bacterial vaccines, which is generally understood to be related to their safety and efficacy. In this study,
the bottom-up analysis revealed insights into chain length distribution and characterization of the RU structural
heterogeneity on a site-specific level. The effect of ionization adjustments, including enriched DEN gas were evaluated
with regard to charge state distribution and ionization efficiency. Overall, the presence of acetonitrile-enriched nebulizing
gas led to a significant increase in the glycan-associated signals derived from the tryptic digestion. Enriched nitrogen gas
showed up to a 6-fold increase in signal intensity and a 4-fold increase in S/N ratios. Next to the expected sensitivity
improvement for mainly the glycopeptide species, the non-glycosylated peptides were also positively affected in the
presence of DEN gas. Analyses revealed information on the target molecule carrying extensive glycan moieties per N-
glycosylation site, as well as partial site occupancy. Of note, the masses of the detected glycopeptides were between 2
and 16 kDa. The heterogeneity on intact glycoconjugate was addressed, providing additional information on the total
number of RU. We found a heterogeneous mixture of RU attached per protein. Finally, the described workflows and
analysis methods can be integrated in current O-polysaccharides protein conjugate mapping workflows, making it an
attractive and precise method for the antibacterial vaccine development sector.

Please explain why your abstract is innovative for mass spectrometry?

We developed complementary mass spectrometric approaches for the detection and characterization of large,
heterogeneous, highly glycosylated bioconjugates.

Co-authors:

Elena Domínguez-Vega, Center for Proteomics and Metabolomics, Leiden University Medical Center
Simone Nicolardi, Center for Proteomics and Metabolomics, Leiden University Medical Center
Guinevere Lageveen-Kammeijer, Center for Proteomics and Metabolomics, Leiden University Medical Center
Renzo Danuser, Janssen Vaccines AG (Branch of Cilag GmbH International)
Ali Al Kaabi, Janssen Vaccines AG (Branch of Cilag GmbH International)
Viktoria Dotz, Bacterial Vaccines Discovery and Early Development, Janssen Vaccines and Prevention B.V.
Michel Beurret, Bacterial Vaccines Discovery and Early Development, Janssen Vaccines and Prevention B.V.
Chakkumkal Anish, Bacterial Vaccines Discovery and Early Development, Janssen Vaccines and Prevention B.V.
Manfred Wuhrer, Center for Proteomics and Metabolomics, Leiden University Medical Center

261
ACRYLAMIDE MONOLITHS FOR THE HYDROPHILIC INTERACTION LIQUID
CHROMATOGRAPHY-MASS SPECTROMETRY OF INTACT
(GLYCO)PROTEINS
Abstract ID: 850

Presenting author: Annika van der Zon, University of Amsterdam

Introduction

One-third of the approved biopharmaceuticals are proteins that are glycosylated. Due to the influence of glycosylation on
bioactivity and pharmacokinetics, it is essential for us to understand their molecular structures. Hydrophilic interaction
liquid chromatography coupled to mass spectrometry (HILIC-MS) is emerging as alternative selectivity to reversed-phase
chromatography for the characterization of intact glycoproteins, capable of chromatographically resolving glycoforms of
glycoproteins. A drawback of reversed-phase is the presence of the residual sites of ionic interactions from the silica-
based materials that negatively influence the chromatographic performance or ionization when an ion-pairing agent is
used. Next to that, with HILIC, low abundance modifications can be identified

Methods

Therefore, we synthesized polymer monoliths capillary for HILIC of intact proteins using polymerization mixtures based
on acrylamide monomers, from which the high peak resolution and capacity were obtained that enormously improves
characterization ability for high degree complex glycoproteins. Capillaries with a small diameter (100 – 800 µm) were
used to achieve high chromatographic performance. A method was developed to separate and identify glycoforms of
intact proteins.

Preliminary data (results)

Two proteins were studied with different characteristics: the receptor-binding domain (RBD) of spike proteins which was
used in COVID-19 studies (MW: ~24 kDa) and the NIST mAb (MW: ~145 kDa). The RBD protein is heavily glycosylated
with N-glycans and O-glycans whereas the NIST mAb has two N-glycosylation sites located on the heavy chain. In this
approach, we were able to achieve highly efficient separations of intact proteins and glycoforms of glycoproteins up to
150 kDa by using HILIC. It allows the separation and detection of small protein quantities (down to 5 ng injected on
column) and reveals more excellent resolution ability in minor differences compared with other separation modes, e.g.
reversed-phase or capillary electrophoresis.

Please explain why your abstract is innovative for mass spectrometry?

The HILIC-MS exerts a unique function in studying biopharmaceuticals. Small protein quantities can be detected with
high resolution. With this favorable morphology, high chromatographic performance is accomplished for large molecules.

Co-authors:

Marta Passamonti, University of Amsterdam

Jesse Wilson, Pacific Northwest National Laboratory
Mowei Zhou, Pacific Northwest National Laboratory
Andrea Gargano, University of Amsterdam

262
HYPHENATING ION MOBILITY AND ACTION SPECTROSCOPY IN A
SYNAPT G2 TO PROBE THE STRUCTURE AND KINETICS OF
AGGREGATING PEPTIDES.
Abstract ID: 830

Presenting author: Steven Daly, MS Vision

Introduction

Nature provides us with an impressive array of examples of both function and malfunction arising from extreme
complexity. An important example of a highly complex molecular network is that of aggregating peptides and proteins.
Aggregation, the transition from soluble functioning proteins into insoluble amyloid aggregates, is directly related to
neurodegenerative diseases including Alzheimer’s and Parkinson’s disease. To control and prevent protein aggregation,
a full understanding of the initial, neurotoxic steps of the aggregation process is vital. However, the complex, heterogenic
and dynamic nature of this process, makes this a huge experimental challenge. To tackle this, we have developed a
novel, multidimensional spectroscopy- and mass spectrometry-based method, which allows us to probe the structure and
kinetics of the initial steps of aggregation in a single measurement.

Methods

MS Vision and Anouk Rijs’s MS-LaserLab group (VU Amsterdam) have collaborated to modify a Synapt G2 such that we
can hyphenate droplets-based microfluidics, electrospray ionization, mass spectrometry, ion mobility spectrometry and
ion spectroscopy in a single experiment (the Photo-Synapt). To be able to measure mass-and mobility-selected UVPD
and IRMPD spectra, the Synapt G2 was modified with two additional hexapoles between mobility and TOF stages. The
use of pin traps allows the storing, manipulation and irradiation with UV or IR photons of mass- and mobility selected
ions.

Preliminary data (results)

In this presentation, we will focus on the design and implementation of the Photo-Synapt, and characterization of its
different operational modes. Ion optical simulations were used to guide the design the hexapole length and geometry of
the pin traps, and the operation of the different trapping modes.

The pin traps are three sets of six pins positioned between the hexapole rods. Two sets of the pins are form potential
well within the hexapole to trap the ions. The first set of pins is used to eject ions from the trap and into the second
hexapole for trapping and irradiation or to the TOF. We show that the addition of the additional hexapoles does not affect
the standard IM-MS and MSn performance, that we can trap in both pin traps, and transfer the ions between the two
traps and to the TOF. The influence of the trapping time, gas pressure and pin trap voltages are explored to find optimal
trapping conditions. Successful UVPD experiments of iodotyrosine demonstrate that we can trap and induce
photodissociation. We are currently being extended to optimize conditions to perform IRMPD measurements.

Finally, the ability to perform ion mobility slicing is demonstrated. This is achieved by pulsing the IMS exit lens. The
trapping and irradiation of a mass and mobility selected population of ions probed by IR or UV photons is the final goal of
the characterization experiments. Initial experiments have shown this is possible, although optimisation of the protocol is
still required.

Please explain why your abstract is innovative for mass spectrometry?

The development of a novel, multidimensional MS-based approach to probe structure and kinetics of peptide
aggregation, or other heterogeneous, dynamic samples, in a single experiment.

Co-authors:
Sjors Bakels, Division of BioAnalytical Chemistry, AIMMS Amsterdam Institute of Molecular and Life Sciences, Vrije
Universiteit Amsterdam
Agathe Depraz, Division of BioAnalytical Chemistry, AIMMS Amsterdam Institute of Molecular and Life Sciences, Vrije
Universiteit Amsterdam
Iuliia Stroganova, Division of BioAnalytical Chemistry, AIMMS Amsterdam Institute of Molecular and Life Sciences, Vrije
Universiteit Amsterdam
Jan Commandeur, MS Visison
Anouk M Rijs, Division of BioAnalytical Chemistry, AIMMS Amsterdam Institute of Molecular and Life Sciences, Vrije
Universiteit Amsterdam

263
HIGH-THROUGHPUT MASS-RESOLVED MICROSCOPE MODE SIMS AND
MALDI IMAGING OF BIOLOGICAL SURFACES
Abstract ID: 135

Presenting author: Aljoscha Körber, M4i/University of Maastricht

Introduction

Mass Spectrometry Imaging (MSI) allows the untargeted detection, identification and localization of hundreds to
thousands of molecules from a surface. One challenge is the low throughput of MSI compared to other imaging
techniques. This is because most MSI images are acquired pixel-by-pixel in so-called microprobe mode. Measurements
can take impractically long in microprobe mode, especially at higher spatial resolutions. A different image acquisition
approach is microscope mode, which acquires tens of thousands of mass spectra in parallel and therefore allows for
orders of magnitudes higher throughput. Here, we use a fast CMOS detector to conduct mass-resolved microscope
mode imaging of murine tissues with Secondary Ion Mass Spectrometry (SIMS) and Matrix-Assisted Laser
Desorption/Ionization (MALDI).

Methods

A TRIFT II mass spectrometer (Physical Electronics Inc., Chanhassen, MN, USA) was equipped with a C60 ion gun
(Ionoptika Ltd., Chandler’s Ford, UK) and an optical setup linked via a multimode fiber to an Explorer One (Spectra-
Physics, Stahnsdorf, Germany) Nd:YLF laser operated at its third harmonic. The initial CCD detector was replaced by a
TPX3CAM (Amsterdam Scientific Instruments B.V., Amsterdam, The Netherlands), a fast hybrid array CMOS detector
with a time resolution of 1.56 ns. Slides of murine tissue were imaged with Metal-Assisted SIMS and MALDI. The data
were processed using self-written scripts on a desktop computer.

Preliminary data (results)

Doing MSI in microscope mode increases throughput by at least two orders of magnitude, and allows us to achieve a
spatial resolution near 1 µm for SIMS. The spatial resolving power of microscope mode only depends on the quality of
the ion optics and not on the type and focus size of the employed surface probe beam. This makes microscope mode a
promising approach for high-resolution MSI, especially for MALDI imaging. To demonstrate the capabilities of our setup,
we measured an entire rat brain section with 2 µm pixel size within only 24 hours (Figure 1). We expect to achieve
additional enhancements in both acquisition time and lateral resolution by increasing the sensitivity of our setup.

Please explain why your abstract is innovative for mass spectrometry?

Decoupling acquisition time from pixel size while achieving high lateral resolution enables much higher throughput. This
will enable MSI to generate more high-quality data and to expand into new applications.

Co-authors:

Ian G.M. Anthony, M4i/University of Maastricht

Ron M.A. Heeren, M4i/University of Maastricht

264
A 16.5 by 18 mm image of a rat brain.

265
RESEARCH PROPOSAL: IMMUNOPRECIPITATION COUPLED WITH
µFLOW-TOP-DOWN HIGH-RESOLUTION MASS SPECTROMETRY FOR THE
QUANTIFICATION OF THE PROTEIN TUMOR BIOMARKER NEURON-
SPECIFIC ENOLASE
Abstract ID: 40

Presenting author: Sebastian van den Wildenberg, Eindhoven University of Technology

Introduction

Introduction: LC-MS methods using bottom-up based (e.g.: protein digestion using Trypsin) and/or middle-down (e.g.:
protein/antibody reduction using DTT) have been popular methods for the quantification of proteins. However, the
development of these methods often come with extensive sample preparation that requires thorough assay optimization.
In addition to this using bottom-up proteomics sequence coverage is often limited and information about Post
Translational Modification’s (PTMs) is lost. Using top-down proteomics intact protein are analyzed, without digestion,
retaining the maximum amount of information. Although, top-down proteomic approaches have their own specific
challenges, such as sensitivity and the availability of internal standards and reference material.

Methods

Objectives: The primary objective of this study is to develop an immunoprecipitation assay combined with protein
elution, followed by intact top-down protein quantification using µflow LC-QToF-MS for the quantification of the small cell
lung cancer (SCLC) tumor biomarker Neuron-Specific Enolase (NSE).

Methods: Protein precipitation will be performed by protein-biomarker-specific-antibodies that are coupled to Protein G-
linked magnetic beads. The protein will be eluted from the antibody. The eluted protein will be separated using a µflow-
liquid chromatography method on a Waters ionKey device coupled to a Waters Xevo G2-XS QToF, resulting in a µflow
LC-QToF-HRMS method.

Preliminary data (results)

Outlook and challenges: As mentioned top-down proteomics have their own specific challenges. Briefly some
challenges to overcome:

1. Sensitivity, since proteins can have a widely spread charge state distribution, spreading the peak intensity over
multiple charge stages. Sensitivity needs to be increased or charged states need to be reduced to maximize intensity of
a limited number of charged states.

2. After isolation using the magnetic beads and coupled antibodies the antibody-antigen-complex needs to be disrupted
to isolate the protein solely. Where using bottom-up proteomics the antibody-protein complex can be digested and
analyzed even without elution.

3. There are multiple ways of data collection modes and MS that can be used, such as mass deconvolution,
single/multiple reaction monitoring (precursor > product), pseudo-single reaction monitoring (precursor > precursor),
extracted ion chromatograms (XIC) and combinations. Since intact protein analysis and High Resolution-MS are
combined single or multiple charge states can be used, single or multiple isotopes can be selected, and the XIC-window
can be varied. A review of the (limited) available literature did not result in an agreement on one single data collection
and analysis method.

Please explain why your abstract is innovative for mass spectrometry?

• Top-Down MS
• Intact Protein Quantification
• HRMS Quantification
• µFlow-LC
• IonKey-MS

266
Session: IM-10 Imaging MS - Instrumentation and Methods

CONTEXTUALIZING MALDI IMAGING: TECHNOLOGIES FOR IMPROVING

MOLECULAR AND BIOLOGICAL SPECIFICITY
Abstract ID: 989

Presenting author: Jeffrey Spraggins, Department of Cell and Developmental Biology, Vanderbilt University,
Nashville, TN USA, Department of Biochemistry, Vanderbilt University, Nashville, TN USA, Department of
Chemistry, Vanderbilt University, Nashville, TN USA, Mass Spectrometry Research Center, Vanderbilt University,
Nashville, TN USA

Introduction

Imaging mass spectrometry (IMS) is a powerful technology that enables untargeted, highly multiplexed in situ molecular
imaging for a wide range of biomolecular classes. However, without improved specificity, interpretability and biological
insight can be limited for IMS experiments. Molecularly, the ability to exhaustively identify the ions observed in imaging
MS experiments is critical for determining the exact molecular and mechanistic changes associated with biological
processes. It is also necessary to be able to link observed molecular distributions directly to specific cell types and
anatomical features unambiguously to provide proper biological context. Here we describe our recent advancements in
MALDI IMS instrumentation and multimodal data integration for addressing these key challenges.

Methods

Tissues were flash frozen, embedded in carboxymethylcellulose, and cryosectioned. Autofluorescence microscopy was
acquired using EGFP, DAPI and DsRed filters on an Axio Scan Z1 Slide Scanner. MALDI IMS was performed on a
MALDI timsTOF fleX MS system in positive and negative ion modes with a pixel size of 5-10µm using 150 laser
shots/pixel. A prototype version of timsControl software was used allowing for precursor ion selection and fragmentation
during MALDI acquisition. Multiplexed immunofluorescence images were collected using EGFP, DAPI, Cy7 and DsRed
filters on an Axio Observer. Data were processed using a combination of commercial and in-house software.

Preliminary data (results)

Molecular identification is one of the most significant challenges in IMS experiments, which generally rely on MS1 mass
accuracy for annotating ions. Parallel-accumulation-serial fragmentation (PASEF) is a powerful technology that leverages
trapped ion mobility spectrometry for sequential isolation and fragmentation based on a species’ mobility. Recently, our
research group has demonstrated the ability to perform highly multiplexed MS/MS on a pixel-by-pixel basis at high spatial
resolution using PASEF IMS. For example, we performed 25 MS/MS experiments from a single 10 µm pixel collected
and more than 100 MS/MS experiments from a 30 µm pixel by embedding multiple TIMS frames into a subdivided pixel.
Variable, m/z-dependent collision energies from 30 to 45 ev were used to ensure efficient fragmentation across the entire
mass range.

In addition to improving structural characterization for MALDI IMS, we have also developed multimodal imaging
strategies for determining molecular profiles of specific cell types and multicellular functional tissue units in situ. Our
workflow brings together IMS, various forms of microscopy, and machine learning to mine MALDI signals based on
segmented cell types to determine the molecular markers of these key tissue features. In one example from our efforts
as part of the NIH Human Biomolecular Atlas Program (HuBMAP), we have used this approach to develop an extensive
lipid and cellular atlas of the human kidney consisting of over 3 million cells comprising 75,000 functional tissue units
(i.e., glomeruli, proximal tubules, distal tubules, and collecting ducts) from 13 human subjects.

Please explain why your abstract is innovative for mass spectrometry?

Here, we describe MALDI imaging technologies for performing pixel-wise highly multiplexed MS/MS using PASEF-IMS
and linking molecular profiles to specific cell types and anatomical features through multimodal imaging approaches.

267
MALDI MSI AND M²AIA ENABLE MOLECULAR 3D RECONSTRUCTIONS OF
SPHEROIDS
Abstract ID: 176

Presenting author: Stefania Alexandra Iakab, a. Center for Mass Spectrometry and Optical Spectroscopy
(CeMOS), Mannheim University of Applied Sciences, 68163 Mannheim, Germany

Introduction

MALDI-MSI is the go-to technique for describing in situ the molecular composition of biological samples. Organoids and
spheroids are very attractive 3D cell cultures because they mimic specific in vivo physiological conditions, can be
developed from human cells, and contribute towards animal-free research. Their analysis brings light on cell-cell and cell-
microenvironment molecular composition and interactions, which creates opportunities in cancer research and drug
screening. However, 3D molecular characterization is required for a complete understanding of the molecular
composition and chemical changes of the system. Therefore, molecular image-to-image registration has become a
necessity but it is still not straightforward. M2aia is an open-source application that enables a fast, user-friendly, and
interactive exploration of large MS imaging datasets. Here, it enables molecular 3D reconstructions of spheroid models.

Methods

Biculture spheroids were fresh frozen and sectioned at 20 µm with a CM 1950 cryostat (Leica Biosystems), sprayed with
matrix using a TM-Sprayer M3 (HTX Technologies), and imaged at 20 µm lateral resolution and 40k spectral resolution
using timsTOFflex MS (Bruker Daltonics) in tims OFF mode, negative ionization mode. The acquired MS images were
individually exported into the imzML format using SCiLS Lab (Bruker Daltonics) before importing into M 2aia. MS image
processing (e.g. normalization) was done using M2aia.

Preliminary data (results)

We used M2aia to reconstruct spheroids in 3D by automatically registering ~30 adjacent tissue slices. Lipid ion images
were used to align all sections by rigid and deformable image-based registration. The interactive capabilities of M2aia
allowed visualizing the simultaneous spatial distribution of multiple m/z features in 3D. We identified morphologically
representative features which illustrated the layered distribution of the two cell lines used to construct the spheroid,
different distributions of cell-specific molecules, and intensity gradients of molecules present in both cell types. Cell-cell
interactions, invasive cell migration, tumor progression, and other spatially relevant interactions could be monitored in 3D
using MALDI-MSI in combination with M2aia.

Please explain why your abstract is innovative for mass spectrometry?

Automatic molecular 3D reconstruction of bicellular spheroid models for MALDI-MSI.

Co-authors:

Jonas Cordes, b. Faculty of Computer Science, Mannheim University of Applied Sciences, Paul-Wittsack-Straße 10,
68163 Mannheim, Germany, c. Medical Faculty Mannheim, University Heidelberg, Theodor Kutzer-Ufer 1-3, 68167
Mannheim, Germany
Florian Keller, d. Institute of Molecular and Cell Biology, Mannheim University of Applied Sciences, 68163 Mannheim,
Germany, e. Institute of Medical Technology of Heidelberg University and Mannheim University of Applied Sciences,
68167 Mannheim, Germany
Ivo Wolf, b. Faculty of Computer Science, Mannheim University of Applied Sciences, Paul-Wittsack-Straße 10, 68163
Mannheim, Germany
Rüdiger Rudolf, d. Institute of Molecular and Cell Biology, Mannheim University of Applied Sciences, 68163 Mannheim,
Germany, e. Institute of Medical Technology of Heidelberg University and Mannheim University of Applied Sciences,
68167 Mannheim, Germany
Carsten Hopf, a. Center for Mass Spectrometry and Optical Spectroscopy (CeMOS), Mannheim University of Applied
Sciences, 68163 Mannheim, Germany

268
TRANSMISSION-MODE MALDI-2 ON A TRAPPED ION MOBILITY
QUADRUPOLE TIME-OF-FLIGHT INSTRUMENT FOR SUB-CELLULAR
RESOLUTION MS IMAGING AT HIGH DATA ACQUISITION SPEEDS
Abstract ID: 695

Presenting author: Alexander Potthoff, Institute of Hygiene, University of Münster, Germany

Introduction

Transmission-mode MALDI combined with laser-induced postionization (t-MALDI-2) enables sensitive MS imaging with
pixel sizes down to about 1 µm. For most mammalian cells, this enables the analysis at cellular and sometimes sub-
cellular resolution. Decreasing pixel size, however, substantially increases measuring time, calling for fast data
acquisition and precise definition of the region of interest on the sample. Here we present the integration of a
transmission-mode setup into a state-of-the-art qTOF-MALDI-2-MSI platform (timsTOF fleX MALDI-2, Bruker). The
combination allowed for high data acquisition speeds, high dynamic range, and coupling with existing dedicated software
solutions streamlined for MSI applications. We demonstrate the capabilities of the method with selected tissues
generated from murine organs (e.g., brain, testes, kidney, retina) and by imaging cell cultures.

Methods

A timsTOF-fleX MALDI-2 instrument (Bruker) was heavily modified to integrate a piezo-actuated XYZ-stage (SmarAct)
and an UV-transmitting objective (Mitutoyo) in transmission-mode geometry. A DPSS-laser (wavelength: 355 nm,
repetition rate: 1 kHz) was used for material ejection producing ablation marks of ~1 µm width on the sample. Stage and
laser controls were connected to existing software solutions (TIMS Control, flexImaging; both Bruker) to enable
unattended automated operation. SCiLS Lab software (SCiLS/Bruker) was used for data analysis. Tissue sections of
different murine organs were prepared to 7-10 µm thickness and cell cultures grown directly on glass slides.

Preliminary data (results)

MALDI-MSI analyses at micrometer spatial resolution require dedicated sample preparation protocols that produce a
homogenous microcrystalline or amorphous matrix layer. To optimize sample coating on the different tested tissue types
and cell cultures we screened a set of different MALDI matrices and deposition methods. Well-prepared samples allowed
for the analysis at cellular to sub-cellular resolution. This is demonstrated in Fig. 1 at the example of murine cerebellum
covered with HABA matrix using a sublimation protocol. Depicted are overlays of t-MALDI-2-MSI data of different lipid ion
species and a bright-field microscopy image recorded pre-MALDI-MSIon a state-of-the-art slide-scanning microscope
(VS200, Olympus). MSI data was acquired with an acquisition speed of 5 pixels/s using a pixel size of 1 x 1 µm².
Notably, small-scale features of the granular layer, the Purkinje cell layer and the white matter regions are all well
resolved in the t-MALDI-2 images. Laser-postionization (MALDI-2) critically increases sensitivity and chemical depth of
the MSI analysis, allowing for the visualization of numerous glyco- and phospholipid classes as well as that of small
metabolites. For the cell cultures we demonstrate a sub-cellular resolution and present first applications of single cell
analysis to decipher chemical heterogeneity.

Please explain why your abstract is innovative for mass spectrometry?

First report of the integration of t-MALDI-2 MSI on a state-of-the-art QTOF-type mass analyzer and dedicated data
acquisition framework.

Co-authors:

Marcel Niehaus, Bruker Daltonics GmbH & Co. KG, Bremen, Germany
Andreas Höhne, Bruker Daltonics GmbH & Co. KG, Bremen, Germany
Arne Fütterer, Bruker Daltonics GmbH & Co. KG, Bremen, Germany
Jens Höhndorf, Bruker Daltonics GmbH & Co. KG, Bremen, Germany
Klaus Dreisewerd, Institute of Hygiene, University of Münster, Germany
Jens Soltwisch, Institute of Hygiene, University of Münster, Germany

269
t-MALDI-2-MSI of mouse cerebellum at 1 µm pixel-size

270
OVERCOMING THE RESOLUTION GAP: INCORPORATING MALDI-IMS
DATA INTO SINGLE CELL PHENOTYPING BY IMAGING MASS
CYTOMETRY
Abstract ID: 567

Presenting author: Jake Griner, Medical University of South Carolina

Introduction

Multimodal imaging mass spectrometry (IMS) aims to leverage the strengths of different instrumental approaches.
Matrix-assisted laser desorption/ionization (MALDI) IMS has been used extensively to study N-linked glycosylation of
proteins. Glycosylation can be attributed to cellular neighborhoods within a tissue, but for low abundance analytes, a
supra-cellular laser size is often required. Historically, therefore, MALDI is unable to resolve specific cell types. Imaging
mass cytometry (IMC), however, provides subcellular resolution and high dimensional single cell phenotyping by using
lanthanide tagged antibodies. We merge these technologies on serial sections of tissue using accurate automated, non-
rigid image co-registration, attributing MALDI analytes to cells defined by IMC.

Methods

Human liver biopsies were obtained under IRB Protocol 111306. Serial sections of FFPE tissue were cut at 5um
thickness and brightfield scans were made using a Hamamatsu Nanozoomer. Glycans were released using PNGaseF
Prime applied with an HTX M5 sprayer, followed by CHCA matrix application. Acquisition was performed on a Bruker
timsTOF Flex at 30μM lateral resolution in positive ion mode. IMC was performed on a serial section by overnight
incubation at 4℃ with lanthanide conjugated antibodies and acquisition on a Fluidigm Hyperion at 1μM lateral resolution.
Image registration was performed using open-source IMS MicroLink and wsireg.

Preliminary data (results)

Greater than 100 liver biopsies were included in this study, representing the spectrum of non-alcoholic fatty liver disease
(NAFLD). Data was analyzed by comparing samples of varying histologic severity as defined by their NAFLD Activity
Score (0-8) and fibrosis score (0-4). IMC analysis uses nuclear and cell membrane markers to create ‘masks’ of cell
boundaries. Uniform Manifold Approximation and Projection (UMAP) and cell clustering analysis then group cells based
on phenotypic similarity. More than fifteen clusters of phenotypically distinct cells are identified, notably a milieu of
immune cells around the portal triad. Through alignment of series of whole slide brightfield scans taken before and after
imaging, laser ablation marks from MALDI IMS are aligned with the IMC ablated regions (a link of each’s explicit spatial
origin in microscopy) on a serial section (using IMS MicroLink and wsireg) as well as with whole slide H&E images. We
show here that it is possible to attribute N-glycans identified on MALDI to distinct cell clusters identified on IMC. For
example, glycosylation of hepatocytes remains relatively unchanged across the disease spectrum and is predominated
by glycans containing mannose, N-acetylglucosamine, and galactose. Development of portal inflammation is associated
with an increased abundance of glycans containing fucose and sialic acids, as these glycans associate with cell clusters
identified as lymphocytes. UMAP and clustering can also be used to incorporate IMS data, where glycans such as
Hex5dHex1HexNAc4NeuAc1 (m/z 2137.7663) and Hex9HexNAc2 (m/z 1905.6338) are used along with IMC markers
such as CD38 and Ki67

Please explain why your abstract is innovative for mass spectrometry?

Demonstration of a novel computational approach to overcome resolution gaps between two IMS modalities linking
highly multiplex, high resolution, cell-type specific IMC images with multiplex glycan images from MALDI IMS.

Co-authors:

Reto Gerber, University of Zurich

Nathan Patterson, Mass Spectrometry Research Center, Vanderbilt University, Department of Biochemistry, Vanderbilt
University
Peggi Angel, Medical University of South Carolina
Carsten Krieg, Medical University of South Carolina
Mark Robinson, University of Zurich
Anand Mehta, Medical University of South Carolina

271
KINETICMSI, AN R-BASED FRAMEWORK FOR RELATIVE
QUANTIFICATION OF SPATIAL ISOTOPIC INCORPORATION
Abstract ID: 406

Presenting author: Berin Boughton, Australian National Phenome Centre, Murdoch University

Introduction

Kinetic mass spectrometry imaging (kMSI) has emerged as an innovative technique integrating the imaging capability of
mass spectrometry imaging (MSI) with stable isotope labelling to determine metabolic flux in a spatiotemporal manner
within tissues. Recent efforts in studying in vivo metabolic flux have been hampered by the complexity and high volumes
of data generated by kMSI and the difficulties in accessing freely available computational resources incorporating
workflows for automated data analysis. Addressing these challenges, we have developed KineticMSI, an open-source
bioinformatics workflow written using base R programming language providing an automated pipeline for processing and
analysingkMSI data. We validate our method by discerning metabolic changes in samples obtained from the
hippocampus of a Huntington's disease (HD) neurodegenerative disease mouse model compared to wild-type (WT).

Methods

Animal model - R6/1 mouse model of Huntington’s disease (HD) and age-matched WT controls (n = 6/group) at 16
weeks were injected with deuterated water (99% 2H2O and 0.9% NaCl) via intraperitoneal injection. Additional, 8% 2H2O
was maintained in drinking water for 8 days post-injection. Tissues were harvested, immediately flash frozen, then
sectioned, dried and norharmane matrix applied using a TM Sprayer.

MALDI-MSI - A Bruker SolariX XR FT-ICR-MS was used to acquire spectra in negative ionisation mode, over m/z 100-
2500, using a 30x30 µm raster.

Data Procesing - Date pre-processing was performed using SCiLS Lab prior to analysis with KineticMSI.

Preliminary data (results)

By segregating pixels into clusters based on tracer incorporation and comparing the cluster means between two
conditions, we uncovered distinct metabolic states in diseased mice, where conventional approaches comparing overall
means across entire brain regions failed.

Wepresent a comprehensive method, incorporating a range of statistical tools that can be applied to any system to
conduct relative quantification of isotopic tracer incorporation across different treatment groups displaying intra-tissue
spatial heterogeneity. KineticMSI allows users to take data-driven decisions by providing a tool for the elucidation of
affected pathways that are associated with detected metabolic turnover changes, thus providing mechanistic insights in a
wide range of biological systems.

Please explain why your abstract is innovative for mass spectrometry?

An automated R-based pipeline for processing and analysing kMSI data, allowing for interrogation of spatially
incorporated stable isotope labels in tissue sections and elucidation of affected metabolic pathways.

Co-authors:

Farheen Farzana, Florey Institute of Neuroscience & Mental Health, University of Melbourne, Bio21 Molecular Science
and Biotechnology Institute, University of Melbourne
Federico Martinez-Seidel, Max Planck Institute of Molecular Plant Physiology, School of Biosciences, University of
Melbourne
Anthony Hannan, Florey Institute of Neuroscience & Mental Health, University of Melbourne
Danny Hatters, Bio21 Molecular Science and Biotechnology Institute, University of Melbourne

272
MASS SPECTROMETRY IMAGING AT 500,000 PIXELS PER SECOND
Abstract ID: 145

Presenting author: Ian G. M. Anthony, Maastricht University

Introduction

Mass spectrometry imaging (MSI) is usually performed below one thousand pixels per second (pps). Secondary
Ionization Mass Spectrometry and Matrix Assisted Laser Desorption/Ionization MSI instruments are usually operated
below 600 pps and 50 pps, respectively. For comparison, modern cell phones acquire video at over 62 million pps. A
conundrum with microprobe-mode MSI is that the number of pixels that cover an area grows quadratically as pixel size
shrinks, leading to smaller imaging areas or longer imaging times. A solution is microscope mode MSI that decouples
spatial resolution from spot size. Herein, we discuss our method of fast microscope mode MSI. We show images
acquired at more than 500,000 pps with pixels smaller than 1 μm and discuss the advantages and limitations of our
approach.

Methods

A 20 kV C60 ion gun (Ionoptika Ltd., Chandler’s Ford, UK) was mounted to a TRIFT II-based mass spectrometer (Figure
1, Physical Electronics, Inc, Chanhassen, MN) equipped with a TPX3CAM, a hybrid pixel detector (Amsterdam Scientific
Instruments, Amsterdam, NL). Slides of murine tissue and fingerprints were prepared by first placing a 300 mesh TEM
grid (Agar Scientific LTD) in an unobtrusive area of the slide and then sputter coating a layer of 1 nm thick gold. Imaging
was performed by linking the output of the TPX3CAM with calculated stage positions. Data were processed using custom
scripts on a desktop computer.

Preliminary data (results)

Results from the fast microscope mode MSI approach applied to a slide with three fingerprints (Figure 2) demonstrate
that large-field areas (i.e., a 42 by 23 mm area that is the unfrosted area of a standard glass slide) can be acquired at
very high throughputs (under 35 minutes of imaging time resulting in an image of over 1 billion pixels and a data file of
more than 60 GB). The acquired MSI data requires new file formats and processing strategies, as such high pixel counts
are inefficiently stored in traditional MSI file formats, such as .IMZML, even when intelligent choices are made about pixel
storage. Furthermore, this technique shows promise for having a relatively large focal range (of at least 135 micrometers
of z-height) and for applications for imaging biological tissue. Because of the lack of need for focusing of the ion beam,
instrumental tuning needs are reduced. The instrumental modification is easily accomplished and the data acquisition
methods and processing strategies and can be performed on a desktop computer. Advancements for this technique will
focus on futher improving throughput, mass resolution, mass range, and image construction techniques.

Please explain why your abstract is innovative for mass spectrometry?

An increase in mass spectrometry imaging speed of over 500 times currently achievable. Whole slide imaging is now
possible at under 1 micrometer pixel size.

Co-authors:

Aljoscha Körber, Maastricht University

Ron M. A. Heeren, Maastricht University

273
Scheme of the fast microscope-mode MSI instrumental setup and pipeline

The acquired image of fingerprints with different pixel sized zoom-ins

274
Session: LS-11 Proteomics: Quantification

MS REDOX-PROTEOMICS FOR GLOBAL ANALYSIS OF OXIDATIVE

RESPONSE
Abstract ID: 788

Presenting author: Sara Zanivan, Cancer Research UK Beatson Institute

Introduction

Increased levels of reactive oxygen species (ROS) have been implicated in the development of many diseases including
cancer and fibrosis. ROS are well recognised signalling molecules because they modify, reversibly or irreversibly, protein
cysteine residues.

Methods

To be able to study ROS-induced signalling, we have developed a stable isotope cysteine labelling with iodoacetamide
method (SICyLIA), which enables to quantify cysteine oxidation levels at global scale with no enrichment steps required.

Preliminary data (results)

We have applied SICyLIA to models of acute and chronic oxidative stress and discovered that cysteine oxidation plays a
key role in cell metabolism regulation. Currently, we are further developing SICyLIA for basic and clinical research.

Please explain why your abstract is innovative for mass spectrometry?

Development of a new MS-based approach for global analysis of cysteine oxidation.

275
DETECTING AND QUANTIFYING TRANSLATIONAL ERRORS BY DATA-
INDEPENDENT ACQUISITION
Abstract ID: 141

Presenting author: Jonas Pöhls, Max-Planck-Institute of Molecular Cell Biology and Genetics, Center for
Systems Biology Dresden

Introduction

Ribosomal frameshifting is an error in mRNA translation where the ribosome changes the reading frame. At a given
position, frameshifting happens stochastically, with estimated rates from 10-5 to more than 50% at some specific sites.
Due to the usually low abundance and the fact that frameshifting results in a completely different amino acid sequence,
the resulting, non-canonical proteins are often missed by conventional analysis. We identify frameshifted protein forms
on a proteome-wide scale by combining high-coverage, high specificity data-independent acquisition (DIA) and a
customized sequence database of predicted frameshifted sequences.

Methods

We optimized our DIA scheme for high specificity by using very narrow, overlapping isolation windows. We avoided
losses in coverage with repeat injections of each sample, acquiring a small m/z range in each injection.
To identify frameshift errors, we predicted the protein sequences resulting from a frameshift at any codon position in all
proteins. We digested and filtered the resulting sequences, restricting ourselves to fully tryptic, proteotypic peptides.
Using DIA-NN, we searched the acquired spectra against the custom database, together with the reference proteome
and contaminants. Identified precursors were further filtered by replicate identification and a strict q-value cutoff.

Preliminary data (results)

We applied our method to S. cerevisiae and identified approx. 4 500 canonical proteins or ca. 75% of the proteome with
more than 40 000 peptides. In addition, we identified over 100 potential frameshifted peptides. The identifications were
confirmed by manual inspection of peptide-spectra matches. Since these peptides did not match to any sequence in the
complete Uniprot database or the six-frame translation of the genome, even when considering single amino acid
substitutions, they are indicative of true ribosomal frameshifting events.
We developed a method for proteome-wide characterization of frameshifting errors, applicable to any organism with
sufficient sequence information. By adapting the strategy for assembling the custom database, this can be extended to
detect other non-canonical peptides. We are now using absolute quantification methods (AQUA, FUGIS) to determine
abundances of frameshifted peptides and their respective 'canonical' proteins. Quantification of the non-canonical protein
forms will provide a proteome-wide overview of the frequencies of translational errors.

Please explain why your abstract is innovative for mass spectrometry?

We show the use of a custom database including non-canonical proteins together with high-coverage DIA. The method
can identify peptides present in cells but previously not annotated in proteomics experiments.

Co-authors:

Tobias Jumel, Max-Planck-Institute of Molecular Cell Biology and Genetics

Andrej Shevchenko, Max-Planck-Institute of Molecular Cell Biology and Genetics
Agnes Toth-Petroczy, Max-Planck-Institute of Molecular Cell Biology and Genetics, Center for Systems Biology Dresden

276
IDENTIFYING NOVEL TRANSCRIPTIONAL REGULATORS USING AFFINITY
PURIFICATIONS COUPLED TO QUANTITATIVE MASS SPECTROMETRY
Abstract ID: 374

Presenting author: Cathrin Graewe, Department of Molecular Biology, Faculty of Science, Radboud Institute for
Molecular Life Sciences, Oncode Institute, Radboud University Nijmegen, Nijmegen, The Netherlands

Introduction

Gene expression is driven by the binding of transcription factors to regulatory elements in the genome, such as
enhancers and promoters. In recent years, DNA-protein interactions have been extensively studied and many essential
transcription factors have been identified. Nevertheless, the mammalian genome contains numerous ‘orphan’ motifs for
which the interacting transcription factors are not yet known.

Methods

A powerful technique to identify transcription factors for orphan motifs is affinity purification followed by mass
spectrometry. Classic affinity purifications coupled to quantitative mass spectrometry provide information about binding
specificity. Binding of transcription factors to regulatory elements in vivo, however, also depends on binding affinity, so
the strength of an interaction. To obtain this information, we recently developed a technique called PAQMAN that uses a
series of DNA affinity purifications to quantify apparent binding affinities proteome-wide.

Preliminary data (results)

We applied classic DNA affinity purifications and PAQMAN to the orphan CGCG motif and identified the essential protein
Banp as a strong transcriptional activator binding this motif with high affinity when unmethylated (Fig. 1). Further
functional experiments showed that Banp opens up chromatin of genes containing the CGCG motif in their promoter.

This example illustrates that DNA affinity purifications coupled to quantitative mass spectrometry represents are a
powerful technology to investigate DNA-protein interactions in a proteome-wide, unbiased manner. We recently
expanded the PAQMAN workflow to provide information about binding specificity and affinity in a single experiment. To
this end, we combine quantitation at the MS1 level with quantitation at the MS2 level, a strategy that is known as higher
order multiplexing (Fig. 2). In the future, we anticipate that this new workflow will be a useful tool to investigate
transcription factor biology.

Please explain why your abstract is innovative for mass spectrometry?

We show, to our knowledge, for the first time that higher order multiplexing can be applied to affinity purification - mass
spectrometry experiments.

The CGCG element is bound by BANP with high affinity.

277
Determination of binding specificity and affinity in a single experiment.

278
PROTEOME WIDE, REAL-TIME SPECTRAL LIBRARY MATCHING TO
IMPROVE SENSITIVITY AND EFFICIENCY OF QUANTITATIVE
PROTEOMICS WORKFLOWS.
Abstract ID: 583

Presenting author: Chris Mcgann, University of Washington

Introduction

Intelligent data acquisition strategies improve instrument efficiency and quantitative accuracy in sample multiplexed
proteomics experiments. Here, we combine advances in data acquisition strategies on a new, modified
Orbitrapinstrument. Previous work has shown proof-of-principle utility of real-time database searching (RTS, Schweppe
et al. 2020), now we demonstrate the utility of an integrated real-time library searching method (RTLS, Figure 1A) to
selectively trigger MS3 scans downstream of highconfidence peptide spectra matching. Extending these methods, we
paired RTS and RTLS modules to enable synergistic filtering, improve triggering specificity, and construct highly flexible
multiplexed methods. When coupled with the sensitivity improvements of the modified Orbitrap instrument, RTS and
RTLS demonstrate consistent improvements of at least 2-fold for sample throughput.

Methods

A two proteome HyPro standard (90% human, 10% yeast, Figure 1B) labelled with TMTpro was used to assess
sensitivity and quantitative accuracy. Unfractionated peptides were injected into a new modified Orbitrap instrument
running new instrument control software and data was compared to those from an Orbitrap Eclipse. We employed both
empirical spectral libraries based on fractionated samples (SpectraST; human, yeast) and predicted TMTpro libraries
using Prosit (Gessulat, et al. 2019) or Prosit-TMT. RTS comparisons were run with either iAPI instrument control or
native instrumental control software. Custom library building software was written in R.

Preliminary data (results)

Real-time data acquisition methods increase protein identifications, quantitative accuracy, and instrument efficiency. We
tested the potential and benefits of coupling hardware improvements of a new modified Orbitrap instrument with
improved real-time informatics (RTLS) for sample multiplexed quantitative proteomics. Adaptive informatics match
peptides to databases or spectral libraries in real-time (Figure 2A), allowing the MS to selectively perform time-intensive
secondary/tertiary scans only on high quality spectra. When quantifying only yeast peptides from the HyPro standard
(Figure 1B), RTLS raised the selectivity of yeast peptides compared to canonical methods from 10% to ~50% (Figure
2B).

In proof-of-principle demonstrations, we show that RTLS can efficiently match spectra both from predicted and empirical
libraries in real time, with instrument acquisition median search times for all methods under 15ms. These speeds enable
RTLS to match against libraries ranging from small, targeted libraries with a few hundred library spectra to a proteome
wide library consisting of >2 million spectra. To facilitate library building and rapid prototyping, we demonstrate the use of
a new custom library-processing tool to convert most common spectral library formats to be compatible with the modified
instrument control software.

We go on to compare the utility of RTS and RTLS as well as their integration with ion mobility, RTS-FAIMS and RTLS-
FAIMS, against canonical SPS-MS3 and HRMS2 methods. RTLS consistently improved selective triggering of HyPro
yeast peptides by 35% (Figure 2C). We observed improved instrument efficiency and sensitivity of the modified Orbitrap
instrument in terms of quantified peptides per hour (Figure 2D).

Please explain why your abstract is innovative for mass spectrometry?

We establish the utility and extensibility of RTLS for intelligent data acquisition on a modified Orbitrap instrument with
application to improved sensitivity for sample multiplexed quantitative proteomics.

Co-authors:

William Barshop, Thermo Fisher Scientific

Jesse Canterbury, Thermo Fisher Scientific
Wassim Gabriel, Technical University Munich
Vlad Zabrouskov, Thermo Fisher Scientific
Romain Huguet, Thermo Fisher Scientific

279
Shannon Eliuk, Thermo Fisher Scientific
Mathias Wilhelm, Technical University Munich
Graeme McAlister, Thermo Fisher Scientific
Devin Schweppe, University of Washington

Overview of real-time library searching (RTLS) and experimental workflow.

Search time, quantitative accuracy and quantified peptides from preliminary experiments.

280
LC-MS PLATFORM FOR HIGH-THROUGHPUT QUANTITATIVE
PROTEOMICS OF WHEAT GRAIN IN LARGE BREEDING PROGRAMS
Abstract ID: 643

Presenting author: Malte Sielaff, Institute for Immunology, University Medical Center of the Johannes
Gutenberg-University Mainz

Introduction

Wheat contributes to up to 20% of the calorie and protein intake of the global population. Its popularity is partly due to its
unique dough rheology and baking properties, which are mainly determined by the protein content and composition of
the grain. On the other hand, flour proteins are implicated in disorders and intolerances related to wheat consumption.
Quantitative proteome screens of wheat flour could therefore provide important insights for breeding and selecting better
and healthier wheat varieties. By combining semiautomated sample preparation, the Evosep One system, ion mobility-
enhanced data-independent acquisition (diaPASEF) and state-of-the-art data processing software, we developed a liquid
chromatography-mass spectrometry (LC-MS)-based bottom-up label-free quantitative (LFQ) proteomics platform
enabling in-depth characterization of wheat flour proteomes in the context of large breeding programs.

Methods

Proteins were extracted from flour under chaotropic and reductive conditions and extracts were transferred to 96-well
microtiter plates. Plate-wise semiautomated sample preparation was performed using a Biomek i7 liquid handling robot
(Beckman Coulter) and an adapted SP3 protocol including reduction, alkylation, magnetic bead-based protein clean-up,
tryptic digestion and peptide recovery. Peptides were loaded on Evotips and analyzed by LC-MS using an Evosep One
(60 samples/day method) coupled to a Bruker timsTOF Pro mass spectrometer (diaPASEF, 0.9 s cycle time). Library-
free search of raw data using a UniProtKB FASTA database of wheat proteins and LFQ was performed in DIA-NN.

Preliminary data (results)

Using the established high-throughput workflow, a quantitative proteome map of 600 wheat flours from a diverse set of
genotypes grown at two different field locations could be generated. While the semiautomated sample preparation
significantly increased throughput and minimized the required manual handling, the integrated desalting step of the
Evosep One system made it possible to eliminate the time-consuming vacuum-drying of peptides that would have been
required after offline peptide desalting. Taking advantage of the high-speed acquisition of the timsTOF Pro and adapting
a diaPASEF isolation scheme to the sample type and LC gradient an average number of three data points per peak at
FWHM (full width at half maximum) could be recorded, providing good quantitative precision. The time-consuming
library-free search in DIA-NN could be decoupled from the final cross-run analyses utilizing the “--use-quant” option
leading to an efficient way of processing large datasets. On average, more than 40.000 peptide precursors were
identified per sample corresponding to over 41.000 stripped peptide sequences and over 7.100 protein groups across the
whole dataset as reported by DIA-NN using the “--relaxed-prot-inf” parameter. Notably, for more than 3.000 protein
groups, a data completeness of 100% was achieved. The intra-batch quantitative reproducibility was very high (r=0.99)
as measured by the correlation of log-transformed protein group quantities of a standard that was prepared in triplicates
on every sample preparation plate and subsequently analyzed at the beginning, in the middle and at the end of each
batch. Cross-batch evaluation of the quantitative reproducibility is currently ongoing.

Please explain why your abstract is innovative for mass spectrometry?

High-throughput LFQ proteomics platform for the analysis of wheat flours in the context of large breeding programs
utilizing automated sample preparation, high-throughput LC and fast ion mobility-enhanced data-independent acquisition
MS.

Co-authors:

Khaoula El Hassouni, State Plant Breeding Institute, University of Hohenheim

C. Friedrich H. Longin, State Plant Breeding Institute, University of Hohenheim
Stefan Tenzer, Institute for Immunology, University Medical Center of the Johannes Gutenberg-University Mainz

281
PROTEOMICS PROFILING OF SALIVA FOR IDENTIFICATION OF NOVEL
BIOMARKER IN ADENOMATOUS POLYP AND COLORECTAL CANCER
PATIENTS VS. HEALTHY CONTROLS
Abstract ID: 717

Presenting author: Sama Rezasoltani, 1- Department of Clinical Chemistry and Laboratory Medicine, University
Medical Center Hamburg-Eppendorf, Hamburg, Germany

Introduction

Colon Cancer (CRC) is a major burden to healthcare systems, accounting for approximately one million new cancer
cases, worldwide, mostly due to the lack of validated clinically beneficial biomarkers with appropriate sensitivity and
specificity to detect such disease at early stages. However, it is well recognized that the CRC pathogenesis is a
progressive accumulation of mutations in several genes, much less is identified at the proteome level. The aim of this
study is to explore salivary proteome biomarkers that could discriminate adenoma polyp (AP) and CRC from healthy
controls (HC) based on proteomics profiling of saliva samples.

Methods

For proteome, mass spectrometry was performed using saliva samples from 50 controls, 53 CRC, and 44 AP.
Specimens were treated with urea, followed by protein reduction with dithiothreitol and alkylation with iodoacetamide. For
protein digestion, urea was diluted to a final concentration of 1.6 M with 50 mM ammonium bicarbonate, and 1 mM of
calcium chloride was added to the samples for trypsin digestion for 16 h at 37 °C. The reaction was quenched with 0.4%
formic acid, and peptides were desalted, dried in a vacuum concentrator, reconstituted in 0.1% formic acid and stored at
−80 °C.

Preliminary data (results)

Current study provides evidence for the clinical utility of a newly proteomic method for CRC early detection. Also, the
novel described CRC-related proteins might become tools for cancer risk assessment in CRC cases.

Please explain why your abstract is innovative for mass spectrometry?

The nwely CRC-related proteins can become tools for cancer risk assessment in CRC patients.

Co-authors:

Hartmut Schlüter, 1- Department of Clinical Chemistry and Laboratory Medicine, University Medical Center Hamburg-
Eppendorf, Hamburg, Germany
Mohammad Mehdi Feizabadi, 2- Department of Microbiology, School of Medicine, Tehran University of Medical
Sciences, Tehran, Iran
Hamid Asadzadeh Aghdaei, 3- Basic and Molecular Epidemiology of Gastrointestinal Disorders Research Center,
Research Institute for Gastroenterology and Liver Diseases, Shahid Beheshti University of Medical Sciences, Tehran,
Iran

282
Session: BM-1 Polymers and Synthetic Molecules

ELUCIDATION OF TADPOLE AND COMB POLYMER ARCHITECTURES BY

TANDEM MASS SPECTROMETRY AND ION MOBILITY TECHNIQUES
Abstract ID: 784

Presenting author: Chrys Wesdemiotis, The University of Akron

Introduction

Polymer composition, size, primary structure, and architecture can greatly impact the properties of a polymeric material.
While techniques for the determination of polymer size and composition abound, deciphering the exact polymer
connectivity (i.e., primary structure) and architecture remains difficult. To address this issue, our study utilized both
tandem mass spectrometry (MS/MS) to elucidate the primary structure and architecture of thiol-yne copolymers based on
their fragmentation patterns, as well as ion mobility mass spectrometry (IM-MS) to separate and distinguish isomeric
comb oligomers based on their unique ion mobilities. Combined, these methods enabled the conclusive identification of
tadpole and comb architectures in the thiol-yne copolymers.

Methods

The polymeric materials were synthesized via thiol‐yne chemistry, resulting in copolymers containing a poly(thioether)
main chain and poly(ethylene oxide) pendants. The polymers were dissolved in methanol and diluted to appropriate
concentrations. MALDI-MS and MS/MS analysis was performed on a Bruker UltraFlex-III ToF/ToF instrument using
DCTB matrix and sodium trifluoroacetate (NaTFA) cationizing salt; ESI-MS/MS and IM-MS experiments were performed
on a Waters Synapt G1 Q/ToF mass spectrometer using NaTFA cationization.

Preliminary data (results)

ESI- and MALDI-MS analysis of the thiol‐yne copolymers showed two main product distributions. ESI‐MS/MS on the low‐
mass oligomers provided conclusive evidence of a tadpole structure based on three distinct fragments from dissociations
in the tadpole tail. In sharp contrast, the fragmentation products observed from the higher‐mass oligomers were
indicative of either unsaturated linear, or cyclized comb species. MALDI‐MS/MS analysis of saturated comb chains,
which can only be linear, showed similar fragmentation patterns as the unsaturated comb polymer, indicating the need
for additional analysis techniques in order to distinguish between an unsaturated linear and a macrocyclic comb
architecture. Using IM-MS, drift time data were collected for both the saturated combs, which were forced to be linear as
a result of having an additional dithiol end group, and the unsaturated combs, which could be linear or cyclized. The
unsaturated comb polymers consistently showed lower drift times than the saturated linear species (at the same mass),
indicating that the unsaturated comb had undergone cyclization. Overall, this work shows the benefits of combining
complementary techniques in order to gain a complete understanding of polymeric architectures within a mixture.

Please explain why your abstract is innovative for mass spectrometry?

Complete primary structure and architecture verification of complex thiol-yne copolymer mixtures.

Co-authors:

Kayla N Williams-Pavlantos, The University of Akron

Brennan Curole, Tulane University
Scott M. Grayson, Tulane University

283
The tadpole undergoes unique fragmentations in the tail.

The unsaturated comb copolymer forms a macrocyclic architecture.

284
WATER OXIDATION ON FREE CALCIUM-MANGANESE-OXIDE CLUSTERS:
GAS PHASE MODEL SYSTEMS FOR THE CATALYTICALLY ACTIVE
CENTER OF PHOTOSYSTEM II
Abstract ID: 802

Presenting author: Sandra Lang, Ulm University

Introduction

The catalytic oxidation of water in plants takes place at an inorganic Mn 4CaO5 cluster located in photosystem II. To aid
the design of new artificial water oxidation catalysts we embark on a novel hierarchical modeling strategy, starting with
small clusters and increasing the model system’s complexity in a staged, controlled manner.

Methods

We uniquely combine gas-phase ion trap reactivity studies with infrared multiple-photon dissociation (IR-MPD)
spectroscopy and first-principles calculations.

Preliminary data (results)

In a first steps of the hierarchical modeling strategy, we studied the reactivity of isolated manganese oxide cluster ions,
MnxOy+, of different size and composition with D216O and H218O. Combined experimental and theoretical work revealed
the facile water deprotonation and the exchange of the oxygen atoms of the cluster with water oxygen atoms as well as
the potential of di-manganese oxide clusters as water oxidation catalysts (to H 2O2). In a further step we investigated
binary calcium manganese oxide clusters (Ca4-xMnxO4+ and Ca5-xMnxO5+) and identified cluster compositions which are
able to mediate the water oxidation reaction. Several parameters were found to be crucial for the reaction to take place at
such small non-ligated clusters. Finally, we started modeling the ligand environment of the manganese oxide clusters by
small acids.

Please explain why your abstract is innovative for mass spectrometry?

Ion trap mass spectrometry used to study properties of bioinspired molecular systems.

285
CELL INSTRUCTIVE MATERIALS FOR NEXT GENERATION MEDICAL
DEVICES: WHAT’S MASS SPECTROMETRY GOT TO DO WITH IT?
Abstract ID: 762

Presenting author: Morgan Alexander, The University of Nottingham

Introduction

Millions of medical devices are surgically implanted per year, with annual sales approaching US$500 billion; we all
probably know someone with a coronary stent, catheter, hip/knee joint, pacemaker or surgical mesh. The range of
biomaterials found in the clinic today are dominated by materials that have been chosen largely on the basis of their
availability and mechanical properties. Failure of implanted devices due to infection or inflammation can be as high as
20%, impacting patients’ quality of life and burdening health services. It is desirable to design our way forward from this
situation to new and better biomaterials chosen for positive interactions with surrounding cells and tissues. Unfortunately,
our understanding of the interface between most materials and biology is poor. Only in isolated cases is there a good
understanding of cell-material interactions and fewer still where material-tissue interactions are well characterised and
understood. This paucity of information on the mechanism of biomaterial interactions within the body acts as a roadblock
to rational design. Consequently, we have taken a high throughput screening approach to discover new bio-instructive
polymers from large chemical libraries of synthetic monomers presented as micro arrays. [1,2] This approach is akin to
engineering the process of serendipitous discovery. It will be illustrated using examples where our polymers are coated
onto devices to control mammalian and microbial cells, including a catheter that has been taken from the lab all the way
to the clinic. This talk will cover the role of mass spectrometry in this endeavour. This started with determining the role of
the polymer surface chemistry [1], to identification of the proteins adsorbed to the material surface from cell culture media
using solution extraction from surfaces. [3] Most recently, it has progressed to in situ analysis using the 3D OrbiSIMS
approach to achieve high spectral resolution imaging mass spectra through direct analysis of solid samples by
incorporating an OrbiTrap with a time-of-flight SIMS instrument. [4] Protein analysis using gas cluster ion beams has
recently been illustrated [5] and complex samples like the plethora of small molecules in bacterial biofilms have become
accessible [6,7] along with cellular markers of immune cell polarisation for next generation implant materials [8].

[1] Hook et al. Nature Biotechnology (2012).

[2] Celiz et al. Nature Materials (2014).

[3] Rostam et al. Matter (Cell Press) (2020).

[4] Passarelli et al. Nature Methods (2017).

[5] Kotowska et al. Nature Communications (2020).

[6] Zhang et al. Analytical Chemistry (2020).

[7] Towards comprehensive analysis of the 3D chemical makeup of Pseudomonas aeruginosa biofilms Kotowska et
al. unpublished.

[8] Single cell metabolomics of macrophages using 3D OrbiSIMS: correlations with phenotype Suvannapruk et
al. unpublished.

Methods

Preliminary data (results)

Please explain why your abstract is innovative for mass spectrometry?

The 3D OrbiSIMS approach allows high spectral resolution imaging mass spectra through direct analysis by
incorporating an OrbiTrap with a time-of-flight SIMS instrument. [Passarelli et al Nature Methods 2017]

286
CORRELATIVE APPROACHES BASED ON MASS SPECTROMETRY FOR
SEMICONDUCTOR APPLICATIONS
Abstract ID: 798

Presenting author: Jean-Paul Barnes, Univ. Grenoble Alpes, CEA, Leti, F-38000

Introduction

Advanced Semiconductor technology such as nano- optoelectronic and photovoltaic applications involve complex
structures and a large variety of materials. The characterization of such devices can be very challenging and require the
development of new approaches, for example, the correlation of several techniques on the same specimen to obtain
information that is more reliable. In many cases, there is also a need to give fast feedback to remain competitive for the
development of new technology and the new processes and materials that are involved. This presentation will address
developments in mass spectrometry based compositional analysis for a range of semiconductor technologies. The
importance of sample preparation will be addressed also.

Methods

Time-of-flight secondary mass spectrometry (TOF-SIMS) was performed on either an ION TOF TOF-SIMS 5 or a PHI
nanoTOF II. Plasma profiling time-of-flight mass spectrometry (PP-TOFMS) was performed on a Horiba instrument.
Transfer between AFM, XPS and TOF-SIMS instruments was achieved using a nitrogen filled transfer capsule. Access to
buried layers is either obtained by sputter depth profiling, or by the preparation of cross-sections using Ga or argon
cluster ion beams. Part of this work, carried out on the Platform for Nanocharacterisation (PFNC), was supported by the
“Recherches Technologiques de Base” program of the French National Research Agency (ANR).

Preliminary data (results)

The first example that will be presented is the characterization of degradation in OLED layers due to ageing. We have
developed a correlative protocol where argon cluster sputtering is used to create a bevel crater in the OLED film stack.
The energy per atom in the argon clusters was optimized to avoid any measurable damage to the molecules present.
The bevel crater means the same zone can be analysed by both TOF-SIMS and XPS, or other techniques, which
facilitates data correlation. The sample is transferred in a nitrogen-filled capsule to avoid surface modification.

A second example is 3D imaging using TOF-SIMS. This is required to analyse the complex composition of modern
semiconductor technology and is necessary at several length scales from a few nanometers up to several tens of
microns. One problem with 3D analysis of devices with many different materials is the different sputter rates. To correct
for this we have developed protocols that use AFM to measure the topography at several points in the TOF-SIMS
analysis, to estimate and correct for the variations in sputter rate. Another problem can be the size of the object to be
analysed or the presence of cavities for example in copper interconnections. Here the use of in-situ Ga ion beam milling
can be used to perform tomography of relatively large sample volumes.

Lastly the use of the PP-TOFMS in combination with TOF-SIMS analysis will be presented. This approach has been
shown to enable fast-feedback for thin-film deposition process development.

Please explain why your abstract is innovative for mass spectrometry?

Specific sample preparation and protocols to enable a correlative approach involving mass spectrometry techniques.

Co-authors:

Claire Guyot, Univ. Grenoble Alpes, CEA, Leti, F-38000

Olivier Renault, Univ. Grenoble Alpes, CEA, Leti, F-38000
Maiglid Moreno, Univ. Grenoble Alpes, CEA, Leti, F-38000
Eric Langer, Univ. Grenoble Alpes, CEA, Leti, F-38000
Nicolas Gauthier, Univ. Grenoble Alpes, CEA, Leti, F-38000
Pierre Hirchenhahn, Univ. Grenoble Alpes, CEA, Leti, F-38000
Tony Maindron, Univ. Grenoble Alpes, CEA, Leti, F-38000
Yann Mazel, Univ. Grenoble Alpes, CEA, Leti, F-38000
Emmanuel Nolot, Univ. Grenoble Alpes, CEA, Leti, F-38000
Nicolas Chevalier, Univ. Grenoble Alpes, CEA, Leti, F-38000
Brice Gautier, Université de Lyon, INSA Lyon, Institut des Nanotechnologies de Lyon, UMR CNRS 5270, F- 69621
Laurent Houssiau, Lise, Namur Institute for Structured Materials (NISM), Université de Namur, rue de Bruxelles, 61,
287
Agnès Tempez, HORIBA France SAS
Sébastien Legendre, HORIBA France SAS
Gregory L Fisher, Physical Electronics, Chanhassen, 55317

288
BIOMEDICAL ACCELERATOR MASS SPECTROMETRY
Abstract ID: 782

Presenting author: Esther Van Duijn, TNO

Introduction

The availability of human data on the absorption, distribution, metabolism and excretion (ADME) early during drug
development can significantly impact the overall drug discovery program. Microdosing/microtracer studies can provide
this information and be executed even before or during Phase I. A very small amount of radiolabelled (often 14C) drug is
administered to a human volunteer solely (microdose) or on top of the intended therapeutic dose (microtrace). This
labeled drug can be analyzed by extremely sensitive accelerator mass spectrometers (AMS).

Here we provide an overview on the use of microdosing/microtracing studies for the safer and faster development of
drugs. This is not only applicable for human adults, this approach has also successfully been applied in pediatrics,
including neonates.

Methods

Various clinical study designs have been applied. To determine the absolute bioavailability of a drug a microtrace is
administered intraveneously while the therapeutic unlabeled drug is administered orally. A microtrace is included in the
oral dose to monitor the excretion/metabolism. Plasma, urine or fecal samples are collected at several timepoints after
dosing. Samples can be analyzed for total radioactivity by AMS or further processed and analyzed by UPLC-AMS-high
resolution mass spectrometer (hrMS). An automated CO2-combustion AMS method is used for the quantitative analysis
of 14C activity. A high resolution MS is used for metabolite identification.

Preliminary data (results)

AMS evolved into a full-fledged biomedical analytical technology and is now capable to analyze a large number of
samples in a short time. Any Phase I study can easily be extended into a human mass balance and metabolite profiling
study, simply by adding a microtracer to the unlabeled dose. The current speed of the analysis even supports subject
discharge from clinical units in the mass balance part. Numerous microtracer studies were performed, generating full
human metabolite profiles by using a combination of UPLC-AMS-hrMS. Relevant metabolites could be selected for
further toxicity testing in the appropriate animal species.

However, not in all cases these studies were performed directly in Phase I. Although even at later stages this approach is
still beneficial (a separate high radioactive mass balance study can be saved), the advantage of early human data is lost.
In some of these studies a human unique metabolite was found, or a metabolite that was present at much higher levels
in human compared to what was seen previously in animals. These late stage surprises severely delay the overall
development program.

Besides the advances made in drug development for human adults by the application of AMS microdosing/microtracer
studies, we also show the added value of this approach in pediatric drug development. Especially the ontogenic
differences raise an enormous challenge between the different age classes of the pediatric population. 14C-paracetamol
and 14C-midazolam have been administered to children, including neonates. Data on the absolute bioavailability,
pharmacokinetics, mass balance and metabolite profiles were generated.

Please explain why your abstract is innovative for mass spectrometry?

AMS allows analysis of samples in the attomole range. This extreme sensitivity of the technology enables the
investigation of new drugs in human at very early stages of development.

Co-authors:

René Braakman, TNO

Elwin Verheij, TNO
Arjan de Vries, TNO
Marta Pelay, TNO
Wouter Vaes, TNO

289
SEQUENCE DETERMINATION OF COPOLYMERS BY MASS
SPECTROMETRY AFTER PYROLYSIS-GAS CHROMATOGRAPHY
Abstract ID: 114

Presenting author: Wouter Knol, Analytical Chemistry Group, van ’t Hoff Institute for Molecular Sciences (HIMS),
Faculty of Science, University of Amsterdam, Science Park 904, Amsterdam, The Netherlands, Centre for
Analytical Sciences Amsterdam, Science Park 904, Amsterdam, The Netherlands

Introduction

A polymer is not a single defined molecule but a collection of molecules featuring distributions in molecular weight,
chemical composition, end-groups etc. One of these distributions, is the sequence distribution, which describes the
average order of monomers in a copolymer. The sequence distribution affects material properties and to develop new
sustainable and high-performance polymers, it is important to determine the sequence length in relation to the final
properties. Although NMR serves as the gold standard, the measurement of the sequence distribution is notoriously
challenging, as it requires the measurement of subunits i.e. diads and triads. Ideally, NMR quantifies the average
copolymer sequence length directly. However, the resolution between various subunits such as trimers is often lacking.

Methods

An alternative method which can be applied to copolymer sequence determination is based on the use of MS, after
degradation of the polymer by pyrolysis. The polymer is fragmented in smaller subunits such as diads and triads. In-
depth MS spectra elucidation make the identification of diads and triads possible, including the monomeric order of these
diads and triads. Overall, the coupling of pyr-GC with MS unlocks quantification and identification of chemically similar
trimers. This allows for the calculation of sequence lengths.

Preliminary data (results)

To investigate the possibilities and thus the application range, this MS based approach was investigated, developed, and
optimised for the sequence analysis of various copolymers. These sequence lengths were linked to NMR results to verify
the obtained values. This presentation presents our findings and details the advantages of py-GC combined with MS in
copolymer sequence analysis and the application range on complex copolymers. To determine the sequence length over
the MWD range, py-GC-MS was coupled to SEC, which also will be addressed.

Please explain why your abstract is innovative for mass spectrometry?

In-depth MS spectra elucidation identifiying triads gives new insights in the sequence of colpolymers, ultimately leading
to a deeper understanding of material properties aiding in the development of future materials.

Co-authors:

Till Gruendling, BASF, Carl-Bosch-Strasse 38, Ludwigshafen am Rhein, Germany

Bob Pirok, Analytical Chemistry Group, van ’t Hoff Institute for Molecular Sciences (HIMS), Faculty of Science, University
of Amsterdam, Science Park 904, Amsterdam, The Netherlands, Centre for Analytical Sciences Amsterdam, Science
Park 904, Amsterdam, The Netherlands
Ron Peters, Analytical Chemistry Group, van ’t Hoff Institute for Molecular Sciences (HIMS), Faculty of Science,
University of Amsterdam, Science Park 904, Amsterdam, The Netherlands, Centre for Analytical Sciences Amsterdam,
Science Park 904, Amsterdam, The Netherlands, Covestro Group Innovation, Sluisweg 12, Waalwijk, The Netherlands

290
Session: LS-12 Glycomics & Glycoproteomics - Session B

EXACT STRUCTURE DETERMINATION OF ISOMERIC GLYCANS BY ION

MOBILITY-MASS SPECTROMETRY
Abstract ID: 108

Presenting author: Javier Sastre Toraño, Chemical Biology and Drug Discovery, Utrecht University

Introduction

Glycans play essential roles in biological processes. To understand their biology at a molecular level and exploit
glycoscience for the development of biopharmaceuticals, diagnostics and nutraceuticals, it is essential to quickly and
easily determine exact glycan structures in complex biological samples. Currently, accurate mass measurements are
used for compositional assignment and additional MS/MS fragmentation experiments can provide structural information.
Due to their isobaric nature, however, the assignment of exact glycan structures from MS/MS spectra remains very
challenging. Furthermore, the lack of well-defined standards is another major hurdle for exact structure identification. In
this work we present a high-throughput and de novo sequencing ion mobility-MS methodology for glycans, that can
unambiguously determine isomeric glycan structures in the absence of synthetic standards.

Methods

A limited set of chemoenzymatically synthesized isomeric glycans is used to develop a drift tube ion mobility
spectrometry (DTIMS)-MS methodology for de novo sequencing and high-throughput identification of isomeric glycan
structures, based on their arrival time distribution (ATD) fingerprints and collision cross section (CCS) values. The DTIMS
methodology uses multiplexing and associated high-resolution demultiplexing to obtain high resolution ATD fingerprints
(up to 267 Ω/ΔΩ) of intact isomeric glycans and their fragments.

Preliminary data (results)

Complex glycans possess multiple flexible glycosidic linkages and as a result can adopt a number of distinct
conformations in the gas phase, determined by glycan branching and positions of carbohydrate residues. We
demonstrate that ATD fingerprints of glycans, which resemble the conformational populations in the gas phase, are
unique and that they can be applied for high-throughput isomeric glycan identification, even for closely related structures.
An ATD reference library with conformer distributions of synthetic glycan standards allows, for example, for high-
throughput identification of all sialic acid linkage‐isomers of biantennary N‐glycans.

For unknown glycans in biological samples without an entry in the reference library and synthetic standard available, an
IMS-MS de novo sequencing method is presented. Fragments of human milk oligosaccharides and N-glycans derived
from proteins, such as from monoclonal antibodies and cell surfaces, are identified by their ATD and CCS values, using
fragment ion entries from the ATD library, and used for de novo sequence assembly to elucidate the glycan structure.
The ATD of the elucidated intact structure is then added to the ATD reference library and used for future high-throughput
identification in other biological samples, creating a self-expanding ATD reference library .

This new methodology will allow the glyco-workfield to rapidly and unambiguously identify exact glycan structures in
biological samples, independently of the peptide backbone or reducing end label, without the need for synthetic
standards.

Please explain why your abstract is innovative for mass spectrometry?

IM-MS reveals unique glycan conformers for high-throughput isomeric structure identification and facilitates de
novo sequencing to elucidate new structures and create a self-expanding ATD library in the absence of standards.

Co-authors:

Gaël Vos, Chemical Biology and Drug Discovery, Utrecht University

Kevin Hooijschuur, Chemical Biology and Drug Discovery, Utrecht University
Chris Klein, Agilent Technologies
John Fjeldsted, Agilent Technologies
Geert-Jan Boons, Chemical Biology and Drug Discovery, Utrecht University

291
STRUCTURAL STUDIES OF SARS-COV-2 SPIKE WT AND OMICRON B.1
PROTEIN TRIMERS AND THEIR HUMAN CELL SURFACE RECEPTORS
Abstract ID: 591

Presenting author: Catherine Costello, Boston University School of Medicine

Introduction

The efficiency and consequences of infection by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)
respond dramatically to sequence variations and posttranslational modifications of the viral spike protein and its
receptors. These need to be considered during design of vaccines and therapies. The extensive N- and O-linked
glycosylation of these proteins is host- and organ-specific. The level of the best-known receptor for the spike protein,
ACE2, is low in some target cells/tissues. We have shown that CD209/CD209L are alternative receptors [1] and that
vimentin can serve as an accessory protein [2]. Results reported here contrast the N-,O-linked glycosylation of the initial
SARS-CoV-2 trimeric spike protein with the recent, highly infective omicron variant and explore the glycosylation of
human CD209 and CD209L.

Methods

Recombinant trimeric SARS-CoV-2 spike protein, hCD209 and hCLEC4M [Creative BioMart] and Omicron spike protein
trimer (B.1.1.529) [ACROBiosystems], all expressed in HEK293 cells, were digested with multiple proteases, then
analyzed directly or fractionated (high pH) before C18-nUPLC-MS analysis on a Fusion Lumos Orbitrap Tribrid MS
[ThermoFisher] with both stepped collision energy HCD and EThcD modes. Signals recorded in parallel by the Thermo
system and an FTMS Booster X1 [Spectroswiss] were processed by Thermo XCalibur or Spectroswiss PeakByPeak, and
analyzed using Peaks Studio X+ [Bioinformatics Solutions] and Byonic 3.8-11 [Protein Metrics]. MS/MS assignments
were manually verified.

Preliminary data (results)

The efficiencies of cell entry and infection by SARS-CoV-2 are strongly affected by sequence changes and PTMs on the
trimeric spike protein of the virus, cell surface receptors, and elements present on host cell surfaces and extracellular
environment. Omicron BA.1 variant spike protein exhibits 35 aa-sequence changes vs. wt, but preserves all 22 potential
N-glycosylation sites, whereas seven changes occur in potential O-linked sites (S/T). HCD and EThcD analyses were
performed on the Fusion Lumos Orbitrap Tribrid MS equipped with Waters Mclass nUPLC, Advion TriVersa NanoMate,
and FTMS-Booster-X1 to investigate site-specific N-,O-glycosylation of wt and omicron spike protein trimers and human
CD-209(DC-SIGN)and CD209L(L-SIGN) receptor constructs. Results are compared with one another and our previous
data obtained for spike monomer, its receptor-binding domain (RBD), and ACE2 and CD209/209L constructs expressed
at BUSM. Omicron showed increased N- and O-linked site occupancy and altered glycoform distributions in the RBD and
elsewhere. CD209 exhibits a wider range of N-glycoforms than CD209L. We have shown that both these receptors can
serve as alternative receptors or facilitate ACE2 binding in tissues where ACE2 has low concentration [1] and that
additional cell surface or extrcellular factors impact SARSCoV-2 binding to ACE2 [2]. Site-specific structural information
obtained using this methodology should advance understanding of viral entry and may lead to improvements in diagnosis
or therapy.

[1] R. Amraei, et al. ACS Cent. Sci., 2021, 7, 1156-1165.

[2]. R. Amraei, et al. Proc. Natl Acad. Sci. USA, 2022, in press.

Please explain why your abstract is innovative for mass spectrometry?

Multiple proteolytic digestions, multistage separations, HCD and ExD, and advanced data acquisition/interpretation
methods elucidate virus and receptor proteins; particular emphasis on full definition of site-specific N- and O-glycoform
distributions.

Co-authors:

Chaoshuang Xia, Boston University School of Medicine

292
TOWARDS REAL-TIME GLYCOPEPTIDE IDENTIFICATION ON THE
TIMSTOF PRO - PASER PLATFORM: VIRTUAL PRECURSOR ENABLED
PEPTIDE-MOIETY IDENTIFICATION
Abstract ID: 520

Presenting author: Gad Armony, Translational Metabolic Laboratory, Department of Laboratory Medicine,
Radboud Institute for Molecular Life Sciences, Radboud University Medical Center

Introduction

Protein glycosylation greatly affects protein function and may serve as useful biomarkers in biomedical applications.
Holistic glycoproteomics in blood can provide site-specific data for glycosylation of up to hundreds of proteins in a single
measurement and therefore may be applied to functional diagnoses of human diseases. We developed software
modules for the PaSER (Parallel Search Engine in Real-time) computational platform to efficiently handle data generated
by PASEF-DDA on timsTOF Pro instruments. These modules enable glycoproteomics in clinical environments by
performing (semi) real-time data processing, data analysis, on-the-fly acquisition parameter adjustment, reporting, and
data management.

Methods

To enable real-time glycopeptide identification capabilities on PaSER, we split the glycopeptide identification into two
distinct processes: peptide moiety identification and glycan moiety identification. The raw fragmentation spectra are
streamed (Kafka) to a classifier module which uses oxonium ions to determine if a spectrum was derived from a
glycopeptide. If so, the classifier uses the fragmentation pattern of the constant N-glycan core structure to determine the
mass of the peptide moiety part. The peptide moiety is identified by modifying the spectrum and submitting it to the
database search engine in PaSER (ProLuCID).

Preliminary data (results)

We developed a glycopeptide fragmentation spectra classifier module for PaSER. We optimized glycopeptide spectrum
classification and peptide moiety mass determination on pre-recorded raw data from plasma glycopeptide samples.

First, we enhanced the sensitivity and specificity of the glycopeptide fragmentation spectrum classification by comparing
glycan oxonium ions to results of a glycopeptide identification search. We optimized the number of characteristic
oxonium ions, their intensity, and the mass tolerance to distinguish between fragmentation spectrum of a glycopeptide
and a non-glycosylated peptide. An oxonium ion filter with the optimized parameters was implemented as the first
classification step.

Next, we optimized the peptide-moiety mass determination by evaluating parameters of the N-glycan core fragmentation
patterns in the raw spectrum. We inspected the number of (consecutive) pattern matches, pattern intensity, charge, and
relative m/z values of identified patterns. Results showed that a combination of the number of pattern matches and
pattern intensity performs best in finding the correct peptide-moiety mass. However, the classifier selectivity can still be
improved, therefore, we are developing a machine learning model to help us reduce the number of incorrect peptide-
moiety mass determinations.

With the classifier implemented in the PaSER platform, glycopeptide characterization from fragmentation spectra is
performed in real-time during data acquisition. Therefore, right after the end of the LC-MS/MS measurement the
glycopeptide identification results are available for subsequent downstream analysis. Currently, these glycopeptide
search results include the identification of the peptide moiety and potential glycan composition(s) based on the glycan
moiety mass and oxonium ion information.

Please explain why your abstract is innovative for mass spectrometry?

Real-time glycopeptide identification from unprocessed instrument data streams for ready-when-run diagnostics
reporting.

Co-authors:

Sven Brehmer, Bruker Daltonics GmbH & Co. KG

Lennard Pfennig, Bruker Daltonics GmbH & Co. KG
Sebastian Wehner, Bruker Daltonics GmbH & Co. KG

293
Fokje Zijlstra, Translational Metabolic Laboratory, Department of Laboratory Medicine, Radboud Institute for Molecular
Life Sciences, Radboud University Medical Center
Tharan Srikumar, Bruker Ltd
Gary Kruppa, Bruker S.R.O.
Dirk Lefeber, Translational Metabolic Laboratory, Department of Laboratory Medicine, Radboud Institute for Molecular
Life Sciences, Radboud University Medical Center, Department of Neurology, Donders Institute for Brain, Cognition and
Behavior, Radboud University Medical Center
Alain van Gool, Translational Metabolic Laboratory, Department of Laboratory Medicine, Radboud University Medical
Center
Hans Wessels, Translational Metabolic Laboratory, Department of Laboratory Medicine, Radboud Institute for Molecular
Life Sciences, Radboud University Medical Center

294
STRUCTURAL CHARACTERIZATION OF ANTIGEN-LIKE
OLIGOSACCHARIDE STRUCTURES BY GAS PHASE INFRARED
SPECTROSCOPY
Abstract ID: 251

Presenting author: Baptiste Moge, Institut Lumière Matière. Université Lyon 1/CNRS. Villeurbanne. France

Introduction

Structural characterization of carbohydrates is of great importance, since they are one of the most important class of
biomolecules constituting living organisms. They play a vital role in many processes, e.g. energetic metabolism, cell-cell
recognition. Despite their crucial importance, these biomolecules are still not well characterized by current analytical
methods this is due to their inherent structural complexity.

Methods

Gas phase infrared spectroscopy is performed on a Thermo LTQ XL ion trap modified to allow injection of an infrared
laser to the trap. Ions are isolated in the trap and irradiated by an infrared laser in the 3µm range. This setup acquires
simultaneously MS spectrum and gas phase IR spectrum of carbohydrates allowing for structural study of the compound.

Preliminary data (results)

The performance of a new MS/IR setup equipped with a kHz IR laser for rapid and automated acquisition of IR data will
be presented.1 A library of MS/IR fingerprints of antigen-like oligosaccharide and their MS/MS fragments will be
presented, as well as the interrogation of a homemade database for the automated identification of unknown
compounds. The case of unknown compounds that are not referenced in the database will be discussed.

1. Yeni, O., Schindler, B., Moge, B. & Compagnon, I. Rapid IRMPD (InfraRed multiple photon dissociation) analysis for
glycomics. Analyst (2021) doi:10.1039/D1AN01870A.

Please explain why your abstract is innovative for mass spectrometry?

A mid Infrared laser irradiates compounds in the ion trap. This setup allows the study of structure of the carbohydrates.

Co-authors:
Oznur Yeni, Institut Lumière Matière. Université Lyon 1/CNRS. Villeurbanne. France
Baptiste Schindler, Institut Lumière Matière. Université Lyon 1/CNRS. Villeurbanne. France
Isabelle Compagnon, Institut Lumière Matière. Université Lyon 1/CNRS. Villeurbanne. France

Higher repetition rate laser allow faster MS/IR spectrum acquisition

295
DEVELOPMENT AND APPLICATION OF ION MOBILITY TANDEM MASS
SPECTROMETRY FOR THE INVESTIGATION OF HUMAN CEREBROSPINAL
FLUID GANGLIOSIDOME
Abstract ID: 417

Presenting author: Mirela Sarbu, Department of Condensed Matter, National Institute for Research and
Development in Electrochemistry and Condensed Matter, Timisoara, Romania, Faculty of Physics, West
University of Timisoara, Timisoara, Romania

Introduction

The proximity of cerebrospinal fluid (CSF) with the brain, its permanent renewal, and better availability for research than
tissue biopsies, as well as ganglioside (GG) shedding from brain to CSF, impelled lately the development of protocols for
the characterization of GGs and discovery of central nervous system (CNS) biomarkers expressed in CSF. Although
CSF sampling is more difficult than urine or blood, as CSF is in direct contact with the brain and spine, CSF testing is
more effective in diagnosing a variety of CNS conditions. Recently, we have implemented ion mobility separation mass
spectrometry (IMS MS) and tandem mass spectrometry (MS/MS) for the exploration of human CSF gangliosidome and
the characterization of rare human CSF glycoforms, with potential biomarker role.

Methods

The normal lumbar CSFs investigated here were obtained from adult individuals exhibiting no signs of tumors,
intracranial haemorrhage or acute inflammatory process of the central nervous system (CNS). The extracted GG sample
of 5 pmol/µL in methanol was infused into a Synapt G2S and the signal was acquired in the negative ion mode at 1.6 kV
ESI voltage and 45V for cone, respectively. To enhance the separation, IMS wave velocity was set at 650 m/s and IMS
wave height at 40 V. Fragmentation experiments were performed after mobility separation in the transfer cell, using
energies between 40-65 eV.

Preliminary data (results)

IMS MS separation and screening revealed 113 distinct GG species in CSF; of these, more than 76% were found
polysialylated, which is similar with the percentage of the polysialylated GG species (over 78%) detected in human brain
using the same approach. As evidenced by IMS MS, the GG classes found in CSF are GT1>GQ1>GD1>
GA3>GD3>GD2=GT2>GM3>GM2=GT3>GM4>GM1=GT4, given here in descending order, based on their abundances.
IMS MS evidenced the presence of four O-Fuc GGs and five O-Ac GGs. Further, the structural confirmation of
GD3(d18:1/18:0) and GD2(d18:1/18:0), exhibiting shorter carbohydrate chain, a feature of CSF, was achieved by
collision-induced dissociation (CID) MS/MS. The detailed evaluation of the GG profile in CSF revealed also the incidence
of some new and biologically relevant species. The doubly charged ions at m/z 1018.004 and m/z 1019.008 assigned,
according to mass calculation, to GalNAc-GD1(d18:1/18:1) and GalNAc-GD1(d18:1/18:0) were further submitted to CID
MS/MS for detailed investigation of the oligosaccharide and ceramide structures. The incidence of only one mobility
feature for each of the investigated species, together with the diagnostic fragment ions allowed the unequivocal
identification, among the six possible structures for each of the fragmented structures, of the exact isomers in the CSF.
Hence, by IMS MS/MS, GalNAc-GD1c(d18:1/18:1) and GalNAc-GD1c(d18:1/18:0) having both Neu5Ac residues and
GalNAc attached to the external galactose were for the first time discovered in CSF and structurally characterized.

Please explain why your abstract is innovative for mass spectrometry?

Human CSF and brain display a similar ganglioside pattern, which might serve in clinical of CNS diseases. IMS MS and
CID MS/MS revealed and structurally confirmed rare GG isomers.

Co-authors:

Dragana Fabris , Department of Chemistry and Biochemistry, University of Zagreb Medical School, Zagreb, Croatia
Željka Vukelić , Department of Chemistry and Biochemistry, University of Zagreb Medical School, Zagreb, Croatia
David Clemmer, Department of Chemistry, Indiana University, Bloomington, Indiana, United States of America
Alina Zamfir, Department of Condensed Matter, National Institute for Research and Development in Electrochemistry and
Condensed Matter, Timisoara, Romania, Department of Technical and Natural Sciences, “Aurel Vlaicu” University of
Arad, Arad, Romania

296
THE USE OF MASS SPECTROMETRY AND GLYCOGENOMICS FOR THE
DISSECTION OF THE HUMAN O-GLYCOME
Abstract ID: 160

Presenting author: Noortje De Haan, University of Copenhagen

Introduction

Protein O-glycosylation affects many cellular functions, including protein cleavage and cellular signaling, and yet we only
have a limited understanding of how glycans impose such functions at the structural level. This lack of knowledge is
partly due to the lack of analytical methodologies to investigate O-glycosylation in biological material.

Protein O-GalNAc glycosylation is particular complex as its site localization and occupancy are controlled by 20 different
enzymes and mature glycans are hugely diverse. Importantly, most steps in glycan biosynthesis follow strict pathways.
Knowledge of these pathways allows us to employ glycoengineering to create standards for glycan structural
identification, and to decomplexify samples before analysis. Here, we present two mass spectrometry-based approaches
aided by glycogenomics to elucidate glycan structures, and site localization and occupancy.

Methods

For the in-depth analysis of glycan isomers, we developed a minimally destructive, non-reductive release of O-glycans
from proteins in tissues and cells. Labeling of the glycans enabled efficient C18 LC-MS analysis, using a standard
proteomics set-up and featuring glycan isomer separation. Glycoengineered human cell lines provided the standards for
structural annotation.

To further investigate O-GalNAc glycan localization and site occupancy, we applied a glycoproteomics strategy on
glycoengineered material expressing monomeric GalNAc glycans. While most O-glycoproteomics relies on glycopeptide
enrichment, inherently losing information on site occupancy, the current method allowed the simultaneous analysis of
peptides and glycopeptides in a complex background.

Preliminary data (results)

The developed glycomics and glycoproteomics strategies were applied on human keratinocyte cell lines as model
system.

The new approach for released glycan analysis allowed multiplexed sample preparation in a 96-well format as well as the
sequential release of N- and O-glycans from the same sample. Application of the method on an array of glycoengineered
human cell lines, e.g., POFUT1, POGLUT1, EOGT, C1GALT1, GCNT1 and B4GALNTs knock outs, resulted in the
structural annotation of O-glycans, including the annotation of initiating monosaccharides and O-GalNAc core elongation.
In the human keratinocytes, we found a wide variety of O-glycan structures, including O-fucose, O-glucose, O-GlcNAc
and O-GalNAc glycosylation, with the latter carrying both elongated core1 and core2 structures and varying numbers of
fucoses and sialic acids. Application of the method on human tissue and disease models will allow the characterization of
the potential change of specific glycan structures during tissue differentiation and disease.

In the secretome of the human keratinocytes we were able to quantify the occupancy of about 70 O-GalNAc glycan sites
or regions, derived from 33 different glycoproteins. While about 60% of the quantified sites exhibited high occupancy
(above 70%), a substantial number of sites showed to be rather low occupied, with 20% of the sites having an occupancy
between 0.1 and 10%. This finding questions the relevance of the high number of O-glycan sites previously reported, and
provides help for the selection of relevant sites which can be pursued in future functional studies.

Please explain why your abstract is innovative for mass spectrometry?

The analysis of protein O-glycosylation by mass spectrometry is challenging. Here we developed sample preparation
strategies that allow O-glycomics and O-glycoproteomics in complex samples by C18 LC-MS/MS.

Co-authors:
Mathias Nielsen, University of Copenhagen
Yoshiki Narimatsu, University of Copenhagen
Sally Dabelsteen, University of Copenhagen
Ieva Bagdonaite, University of Copenhagen
Sergey Vakhrushev, University of Copenhagen
Hans Wandall, University of Copenhagen

297
298
 Thursday 1 September 2022: 11:00 – 13:00
Session: IM-11 Ion Spectroscopy, Physical and Chemical principles
underlying MS - Session A

ELECTRONIC CIRCULAR DICHROISM ION SPECTROSCOPY

Abstract ID: 224

Presenting author: Valérie Gabelica, Université de Bordeaux, CNRS & Inserm, Institut Européen de Chimie et
Biologie (IECB, UAR3033, US01), Université de Bordeaux, Inserm & CNRS, Laboratoire Acides Nucléiques:
Régulations Naturelles et Artificielles (ARNA, U1212, UMR5320)

Introduction

DNA and proteins are chiral: their three-dimensional structure cannot be superimposed with its mirror image. However,
mass spectrometry-based measurements are typically blind to chirality. Characterizing chiral compounds by mass
spectrometry currently requires a physical interaction with other chiral molecules, either by separation on chiral phases in
front of the mass spectrometer or by forming complexes with chiral auxiliaries in the gas phase. Here we report how to
record electronic circular dichroism spectra on DNA helices separated in a mass spectrometer, based on the interaction
between the ions and chiral light.

Methods

We analyzed several rigid DNA structures (G-quadruplexes, silver-bound G-duplexes) having distinct solution CD
spectra, and validated their fold preservation in the gas phase by ion mobility spectrometry. The ions are electrosprayed
in the negative mode and analyzed in a quadrupole ion trap mass spectrometer, with holes in the trap to shine
nanosecond wavelength-tunable OPO laser, in the range 220-300 nm. Circularly polarized laser pulses were generated
using achromatic broadband quarter wave plates. The laser enters in the ion trap through a fused silica window in the
vacuum manifold. The percentage of circular polarization was greater than 95%.

Preliminary data (results)

To unambiguously prove that the gas-phase CD effect comes from the sample and not from an instrumental artifact, we
performed the experiment on d-[(TGGGGT)4·(NH4+)3] formed from the natural DNA backbone (all d-sugars), and from its
enantiomer (all l-sugars). The gas-phase CD signals have opposite signs (Fig. 1A). Then we compared the solution and
gas-phase CD spectra for other typical DNA helices (Fig. 1B—D; symbols for gas-phase CD, lines for solution CD). The
solution and gas phase CD spectral shapes are similar in terms of sign and position of the maxima and minima. Given
that solution-phase spectra result from the CD in absorption, we conclude that the gas-phase CD effect in
photodetachment yields is also due to the resonant absorption of circularly polarized light by a chiral molecule. However,
the gas and solution phase asymmetry factors differ by their magnitude, which is consistently larger in the gas phase
than in solution. We hypothesize that the electronic states responsible for the CD effect being most likely to be
delocalized on the entire DNA helices, this may result in more efficient auto-detachment after resonant excitation. The
rest of the presentation will show the most recent applications of this new measurement technique.

Daly S, Rosu F, Gabelica V, “Mass-resolved electronic circular dichroism ion spectroscopy”, Science (2020) 368, 1465.

Please explain why your abstract is innovative for mass spectrometry?

First realization of electronic circular dichroism ion spectroscopy on electrosprayed multiply charged ions. This abstract is
submitted for the invited keynote lecture by Anouk Rijs.

Co-authors:

Steven Daly, Université de Bordeaux, Inserm & CNRS, Laboratoire Acides Nucléiques: Régulations Naturelles et
Artificielles (ARNA, U1212, UMR5320)
Aram Hong, Université de Bordeaux, Inserm & CNRS, Laboratoire Acides Nucléiques: Régulations Naturelles et
Artificielles (ARNA, U1212, UMR5320)
Frédéric Rosu, Université de Bordeaux, CNRS & Inserm, Institut Européen de Chimie et Biologie (IECB, UAR3033,
US01)

299
Gas-phase circular dichroism (symbols) compared to solution phase spectra (lines).

300
OPTIMIZATION OF IONIC LIQUID CLUSTERS IONIZATION BY
EXPERIMENTAL DESIGN AND INTERACTION STRENGTH COMPARISON
USING ESI-MS/MS
Abstract ID: 450

Presenting author: Alexis Dubuis, L’Oréal Research & Innovation

Introduction

Ionic liquids (ILs) can be used as green alternatives to solvents in various application fields. However, interactions
energies between partners and their impact on physico-chemical properties remains uncertain. Previous studies carried
out in electrospray mass spectrometry (ESI-MS) showed (i) the desorption of series of (Cq+1Aq)+ and (CqAq+1)- clusters (A
and C for anion and cation respectively) (ii) relative stability of non-covalent interaction using survivor ion method under
CID conditions. In the present study, cluster ionization/desorption conditions were optimized using a design of
experiment (DOE) approach based on various solvents and electrospray source parameter optimization. Then, a model
was proposed to describe cluster's structure and non-covalent strengths were compared using energy resolved mass
spectra (ERMS: varying collision energy, CE).

Methods

Seven ILs constituted by different anions (acetate, octanoate, chloride, tetrachloroaluminate, saccharinate, tosylate,
methylsulfate) and 1-ethyl-3-methylimidazolium (Emim) were analyzed at 0.1 mmol.L-1 directly introduced at 5 µL.min-1.
An Orbitrap Exploris 240 (ThermoScientific) with an ESI source was used in negative mode with a 45.000 resolution
on m/z 50 –500. DOE comprising 25 experiments was built with four quantitative factors: spray tension, sheath gas,
auxiliary gas, RF lens and one qualitative factor: solvent. Transfer tube temperature was 320 °C. ERMS with higher-
energy collision dissociation (HCD) were performed with 1 eV < CE < 20 eV.

Preliminary data (results)

The general trend observed on the mass spectra of ILs series is that cluster’s abundance decreases with its size
increase. During the solvent screening, isopropanol appeared as the best candidate to promote formation/stability of
clusters, followed by methanol. Conversely, adding water or additives led to a drop of cluster abundances. Regarding the
source parameters, three parameters seem to be relevant for the DOE: sheath gas, auxiliary gas and the spray tension.
These observations were made on the m/z 229 intensity corresponding to the (A, C, A)- cluster of Emim acetate, but for
further parameters optimization in Emim series study, the DOE approach was applied. It allowed to determine optimal
parameters in negative ESI mode specific to each IL, with a trend observed at a maximum spray tension associated with
a minimum sheath gaz. The mass spectra recorded with optimized parameters showed an increase of the cluster
abundance up to 10. Subsequently their fragmentation under HCD conditions could be carried out, showing the loss of
the neutral anion-cation salt. A model involving salt bridges and hydrogen bonds was proposed to explain gas phase
formation of series of Emim clusters. In a second step, HCD mode gave access to ERMS. ERMS of the positive and
negative cluster ions did not show similar trend, specifically for Emim chloride (in negative) and Emim acetate (in
positive) which show higher interaction strength than the rest of the series.

Please explain why your abstract is innovative for mass spectrometry?

This methodology provides experimental data on ILs interactions and new insights for a rational choice according to their
use.

Co-authors:

Anais Berthollet, L’Oréal Research & Innovation

Priscilla Riva, L’Oréal Research & Innovation
Olivia Dufour, L’Oréal Research & Innovation
Natali Stojiljkovic, L’Oréal Research & Innovation
Stéphane Sabelle, L’Oréal Research & Innovation
Jean-Claude Tabet, Sorbonne Université, Faculté des sciences et de l’ingénierie, Institut Parisien de Chimie Moléculaire
(IPCM), Université Paris-Saclay, CEA, INRAE, Département Médicaments et Technologies pour la Santé (DMTS),
MetaboHUB

301
COMBINING NATIVE ION MOBILITY-MASS SPECTROMETRY AND
FLUORESCENCE SPECTROSCOPY FOR STRUCTURAL
CHARACTERIZATION OF BIOMOLECULES IN THE GAS PHASE
Abstract ID: 65

Presenting author: Ri Wu, ETH Zurich

Introduction

Conformational studies on biomolecules in the gas phase based on Ion mobility mass spectrometry (IM-MS) and Förster
resonance energy transfer (FRET) have been of increasing interest in recent years.

Methods

In this study, we propose a powerful new structural characterization method by combining of a home-built differential ion
mobility spectrometry (DMS) device, to IM-MS (i.e., TWIM and cIM), and gas-phase fluorescence spectroscopy based on
a modified quadrupole ion trap (QIT) MS. The DMS can be easily moved between instruments, allowing FRET
measurements, as well as determination of the differential mobility(ΔK), and collision cross sections (CCS). Molecular
dynamics (MD) simulations were then conducted to provide conformational insights.

Preliminary data (results)

The [M+3H]3+ ion of a singly-labelled polyanaline peptide (PAV5-Atto 532) showed two separated peaks in DMS with 0.3
mol % isopropanol as gas modifier in the buffer gas. Downstream IM-MS measurements confirmed 3 and 4
conformations with different ratios for each DMS separated peaks. Further, isomeric doubly-labelled polyanaline peptides
(5+, 6+, and 7+ charge states) could also be separated in DMS with 0.4 mol % ratio acetonitrile as gas modifier. FRET
measurements showed that the calculated donor-acceptor distances (rDA) of the [M+6H]6+ ions of PAV5-crh6g-QSY7 and
PAV6-crh6g-QSY7 are 77.4 ± 1.0 Å, and 71.5 ± 0.9 Å, respectively. The structural information stored in the differential
mobility (ΔK) due to ion-modifier clusters binding (Gibbs free energy, ΔG), shape or CCS, and r DA, could define structural
constraints for MD simulations. The combination of these three complementary techniques promises more accurate
information and thus a better picture on gas-phase biomolecular conformations together with computational studies.

Please explain why your abstract is innovative for mass spectrometry?

DMS, native IM-MS, and fluorescence spectroscopy (i.e., FRET) are coupled with MD simulations for highly accurate
structural analysis of biomolucules in the gas phase.

Co-authors:

Jonas B. Metternich, ETH Zurich

Table of content

302
INVESTIGATION OF PEPTOID IONS SECONDARY STRUCTURES BY ION
MOBILITY MASS SPECTROMETRY - MASS SPECTROMETRY
Abstract ID: 194

Presenting author: Perrine Weber, University of Mons

Introduction

Peptoids represent an emergent class of synthetic polymers differing from peptides by the side chain position on the
amide nitrogen atom instead on the α-carbon atom. They can adopt specific secondary structures such as helices in
solution. Their characterization is usually achieved by Nuclear Magnetic Resonance (NMR) and Circular Dichroism (CD),
methods requiring an extensive purification. Nowadays, Mass Spectrometry (MS) represents an efficient method to
decipher primary and secondary structures of macromolecules, even in complex mixtures. Currently, specific studies are
combining Ion Mobility Spectrometry MS (IMS-MS) and computational chemistry to provide data on 3D ion structures in
gas phase. Due to ionization/desolvation processes, huge structural modifications can be induced and the structure
retention from the solution to the gas phase represents a great challenge.

Methods

Short peptoids presenting defined sequences are usually prepared by a step-by-step solid-phase approach. For longer
chains, a ring-opening polymerization is preconized to prepare, in one step, a large number of N-substituted glycines
oligomers with various degrees of polymerization (DP) called “poly(N-substituted glycines)” or “polypeptoids”. By
associating these methods, we may prepare a large diversity of peptoids with different substituents and/or chain length to
determine the essential factors responsible for the structuring. We will take benefit of this communication to present
some of our results concerning the (poly)peptoids synthesis and their investigation by IMS-MS combined to
computational chemistry.

Preliminary data (results)

N-(S)-phenylethyl (Nspe) peptoids are known to form helical structures in solution. To evaluate whether this conformation
is conserved upon ESI, we investigated the conformations adopted by the protonated poly(N-(S)-phenylethylglycine)
(PNsPeGlyH+) in the gas phase by IMS-MS for a wide range of DPs. The helicity loss in favour of a loop due to the «
charge solvation effect » is invariably detected whatever the instrumental conditions (Figure (a)).

To maintain peptoid helices in the gas phase, 2 strategies have been explored.

The first strategy relied on the macrodipole stabilization typical of helical peptide ions. We considered negatively charged
peptoids and, by positioning the negative charge at the positive side of the helical peptoid macrodipole, structural
changes described in the positive ionization mode should be circumvented in the negative mode. A N-(S)-(1-carboxy-2-
phenylethyl) (Nscp) residue is grafted to a Nspe backbone at specific positions, including the extremities, as the negative
charge carrier (Figure (b)). However, no evidence of the helix presence was obtained, even if, for selected sequences
and lengths, different conformations are detected.

The second strategy was to compensate the energy gained during the charge solvation by the loop formation introducing
strong interactions all along the side chains. We selected a substituent able to provide hydrogen bonds such as the Nscp
side chain and designed full Nscp peptoids (Figure (c)). We demonstrate that the peptoid helix is preserved in the gas
phase by the hydrogen bond network creation compensating the charge solvation process.

Please explain why your abstract is innovative for mass spectrometry?

Ion mobility investigation of synthetic foldamers and secondary structure conservation upon Electrospray ionization.

303
Conformations observed in gas phase with the different sequences mentioned.

304
GAS-PHASE INTRAMOLECULAR PROTON TRANSFER CATALYSIS OF
PARA-AMINOBENZOIC ACID
Abstract ID: 581

Presenting author: Boris Ucur, University of Wollongong, School of Chemistry and Molecular Bioscience

Introduction

Intramolecular proton transfer catalysis and the implication of solvent molecules underpins many cases of protonation
mechanisms present within electrospray ionization mass spectrometry, where these processes are responsible for
producing distributions of protonation site isomers.

Para-aminobenzoic acid (pABA) has become the archetype molecule for studying intramolecular proton transfer and the
role of ESI source conditions. From our previous work, we showed that for pABA, proton transfer between protonation
sites is catalysed by one methanol molecule under ion trap conditions. A generalized scheme for [pABA—
H]+ intramolecular proton transfer catalysis is shown in Figure 1. This general scheme provides a framework for
predicting the change in rate of intramolecular isomerism caused by modifying the molecular properties of the interacting
neutral catalyst.

Methods

Experiments were performed on a Thermo Fisher Scientific LTQ-XL mass spectrometer, modified with a gas-handling
manifold. Electrospray ionization was used to produce [p ABA—H]+ ions that were isolated and stored in the ion trap. The
[pABA—H]+ ions were reacted with the following neutrals: water, formic acid, methanol, ethanol, propanol, ammonia and
acetonitrile. Mass spectra were acquired after storage times spanning 1–10 000 ms and kinetic data was extracted for
relevant m/z channels. Isomerism reactions were repeated six times at varying number densities. Energy schemes were
constructed for intramolecular proton transfers and calculated using DSD-PBEP86/aug-ccpVDZ.

Preliminary data (results)

The first-order-rate constants were plotted against six different number densities for each catalyst exhibiting isomerism.
Each catalyst series was fitted with a straight line with the following trend in protonation site isomerism efficiency: water <
formic acid < methanol < ethanol < propanol < ammonia. Intramolecular proton transfer schemes were constructed with
each of the catalysts, which were calculated with DSDPBEP86/aug-cc-pVDZ. These surfaces revealed that stabilization
of the intramolecular proton transfer transition state followed the same order as the increase in proton transfer efficiency.
These transition state energies were subsequently plotted against the corresponding log k2nd values (Figure 2)and this
yielded an inversely proportional relationship, indicating that the transition state controls the reactivity. The transition
state energies were then plotted against the catalyst proton affinities that also showed an inversely proportional
relationship. By correlation, this demonstrates that increasing the proton affinity of the catalyst increases the reactivity.

The results from this study can be used to help control the placement of protons in polyfunctional molecules in mass
spectrometry experiments and predict the mechanisms of proton transfer that can occur during electrospray ionization.

Please explain why your abstract is innovative for mass spectrometry?

Ion-molecule reactions and quantum chemical calculations provide a useful framework for predicting the mechanism and
reactivity for intramolecular proton transfer.

Co-authors:

Oisin Shiels, University of Wollongong, School of Chemistry and Molecular Bioscience

Shane Ellis, University of Wollongong, School of Chemistry and Molecular Bioscience, Illawarra Health and Medical
Research Institute, Wollongong, New South Wales 2522, Australia
Stephen D. Blanksby, Central Analytical Research Facility, Institute for Future Environments, Queensland University of
Technology, Brisbane 4001, Australia

305
Schematic for intramolecular proton transfer catalysed by a neutral R.

Predictive framework for the pABAH+ intramolecular proton transfer catalysis.

306
MASS SPECTROMETRY BASED FOOTPRINTING FOR MEMBRANE
PROTEINS
Abstract ID: 770

Presenting author: Michael Gross, Washington University in St Louis

Introduction

Integral membrane proteins (IMPs) perform key roles in cellular signaling, transport, adhesion, and catalysis, motivating
their choice for 60% of drug targets. Membrane proteins are sensitive to the stabilizing environment, and footprinting
IMPs in their native cell membrane environment is significant but challenging owing to the limitations of current
footprinting approaches. Based on our recent NanoPOMP development (Sun, J. et al. Nanoparticles and photochemistry
for native-like transmembrane protein footprinting. Nat Commun 12, 7270 (2021)), we are extending TiO2 nanoparticles
for laser-initiated footprinting of the membrane transporter, humanGLUT1 to elucidate the protein motions in transport
cells containing this prototypal MP. The presentation will explain NanoPOMP and describe its extension to in-cell
footprinting, demonstrating the potential to footprint membrane proteins in cells.

Methods

NanoPOMP was developed using the standar flow system used for OH radical footprinting. To extend this platform,
proteins were footprint with a new platform coupling a 96-well plate and Nd-YAG laser at 355 nm. A 20 μL sample in
each well was irradiated by the laser. After labeling, proteins were extracted and digested by trypsin or chymotrypsin.
After digestion, 5 μL of sample was loaded onto a custom-built silica capillary column and a Thermo QExactive Plus
hybrid mass spectrometer coupled with a nanospray ion source and submitted to proteomic analysis. The data analysis
was done with Protein Metrics software.

Preliminary data (results)

In preliminary experiments, we used a fixed concentration of TiO2 nanoparticles for incubation with cells. Under 355 nm
laser irradiation, the NPs adsorbed on the cell surface produce a high local concentration of free OH radicals to footprint
the IMPs. After the footprinting, we extracted, digested the target protein with chymotrypsin, analyzed each proteolytic
peptide by using nanoHPLC MS/MS, and showed > 90% coverage. Peptide fragmentation in the LC/MS/MS mode
showed that the generated free radicals react with residues mainly located on the extramembrane domain. After
modification, the HPLC chromatography peaks representing peptides showed a shift to shorter retention time, owing to
the decreased hydrophobicity of the OH-modified peptides. All the identified peptides were manually checked.
Summarizing these preliminary results, we found 11 oxidized peptides, and 8 of them were from the extramembrane
domain, demonstrating potential for in cell footprinting of IMPs.

In subsequent experiments, we will i) screen several incubation conditions for NPs with cells to achieve a higher
adsorption efficiency and, thus, a higher footprinting level; ii) optimize the approach to increase the cell membrane
permeability to increase the footprinting coverage of the transmembrane domain; and iii) optimize the laser conditions
(e.g., pulses per well, energy per pulse) to achieve more extensive footprinting performance.

The poster will show the adaption to the FPOP platform for in-cell footprinting and the outcome of transmembrane
footprinting.

Please explain why your abstract is innovative for mass spectrometry?

Laser-initiated photocatalytic nanoparticles were employed for the first time to footprint integral membrane proteins in
liposomes and cells

Co-authors:

Jie Sun, Washington University in St Louis

Weikai Li, Washington University in St Louis

307
NanoPOMP protocol for OH radical footprinting of membrane proteins.

308
Session: IM-12 Separation & Hyphenation; Chromatography,
Electrophoresis - Small Molecules

MULTI-DIMENSIONAL LIQUID CHROMATOGRAPHY TECHNIQUES FOR

THE ANALYSIS OF ORGANIC MICROPOLLUTANTS IN ENVIRONMENTAL
SAMPLES
Abstract ID: 793

Presenting author: Deirdre Cabooter, University of Leuven (KU Leuven)

Introduction

In recent years, the discharge of organic micropollutants (pharmaceuticals, hormones, personal care products,
pesticides, …) in environmental waters has become a problem of increasing concern. Organic micropollutants (OMPs)
are released into the environment via human and animal excretions, agricultural processes and industrial activities. Even
though most water discharges are subjected to treatment in wastewater treatment plants (WWTPs), many OMPs can
pass these treatment plants unaffected. This is due to the fact that the current generation of wastewater treatment plants
cannot degrade these micropollutants adequately, resulting in environmental concentrations that can reach ng/L and
even µg/L levels for a variety of OMPs.

Methods

Due to the impact these micropollutants can have on aquatic organisms, their presence needs to be monitored carefully.
For this purpose, adequate and sensitive analytical techniques are required, such as liquid chromatography in
combination with mass spectrometry (LC-MS). Since OMPs can display a wide variety in characteristics (in terms of
polarity, ionization state, molecular weight…) and the environmental water matrices wherein they are present are often
moreover very complex (containing inorganic salts, natural organic matter…) a single LC-MS method is often not
sufficient for the full characterization of all compounds of interest.

Preliminary data (results)

Multi-dimensional LC techniques, combining multiple LC methods by transferring fractions from a first-dimension column
to a second-dimension column with a different selectivity, are interesting in this perspective as they can provide an
enlarged separation performance for complex samples. This presentation will give an overview of the different multi-
dimensional LC-MS solutions our team has recently developed to deal with the problem of emerging OMPs in the
environment.

A commercially available 2D-LC instrument will be evaluated for the combination of a hydrophilic interaction liquid
chromatography (HILIC) column and a reversed-phase (RP) LC column for the analysis of OMPs with a wide variety in
polarity. The possibilities of different 2D-LC techniques, such as (multiple) heart-cutting and selective comprehensive LC
will be demonstrated. To deal with the incompatibility of the mobile phases used in HILIC and RPLC, active solvent
modulation will be evaluated, wherein a specifically designed valve is used for the on-line dilution of the first-dimension
fractions before transfer into the second-dimension column by means of restriction capillaries. This commercial solution
will then be compared with an in-house developed 2D-LC solution wherein a single pump is used to couple a HILIC and
RPLC column in series. Similarly, an in-house developed solution based on restriction capillaries is used to dilute the
effluent of the first dimension column before introducing it into the second dimension column. Both commercial and in-
house developed 2D-LC solutions will be compared in terms of robustness, separation capacity, flexibility and cost for
the analysis of OMPs in environmental waters.

Please explain why your abstract is innovative for mass spectrometry?

The improved separations of complex samples that can be obtained via multi-dimensional LC techniques facilitate
identification using high-resolution mass spectrometry

Co-authors:
Marie Pardon, University of Leuven (KU Leuven)
Rafael Reis, University of Leuven (KU Leuven)
Peter de Witte, University of Leuven (KU Leuven)
Soraya Chapel, University of Leuven (KU Leuven)

309
SEPARATION OF COELUTING ISOMER AND ISOBAR COMPOUNDS FROM
COMPLEX HALOGENATED POLLUTANT MIXTURES BY GC-APCI-TIMS-
TOFMS
Abstract ID: 283

Presenting author: Hugo Muller, Mass Spectrometry Laboratory - ULiège

Introduction

Nowadays, the ubiquitous presence and continuous emission of a wide range of pollutants in the environment is of
particular concern due to the adverse effects they represent for human kind and ecosystems. In particular, the food and
environmental monitoring of regulated Persistent Organic Pollutants (POPs) such as dioxins, furans, PCBs is performed
by well-established targeted reference methods such as GC-EI-magnetic sectors high resolution mass spectrometry in
selected ion monitoring (SIM) or GC triple quadrupole in tandem mass spectrometry mode. However, the number of
compounds of concern to be monitored is steadily increasing, including poorly studied, overlooked and emerging
compounds (e.g., brominated dioxins/biphenyls & PFAS). Thus, it is essential to build solutions that are based on a
different vision of analytical chemistry and develop non-targeted, rather than targeted, methods for the analysis of POPs.

Methods

Our research aims to develop a non-targeted method for halogenated pollutants analysis by the integration of ion mobility
(IM) between GC and TOFMS detector to improve selectivity, sensitivity and identification capabilities. A TIMS TOFpro II
mass spectrometer (Bruker, Bremen) equipped with a GC-APCI source for sample ionization (GC-APCI II, Bruker,
Bremen) was used for this purpose.

Preliminary data (results)

A common problem in hyphenated chromatography-MS applications is the case of coeluting isomeric/isobaric

compounds that can neither be separated in the chromatographic nor the m/z dimension. IM is often presented as an
additional dimension of separation added to MS (and chromatographic methods), able to separate isomeric and isobaric
compounds. Here we present through a few examples how the additional separation provided by IM can be useful in the
above-mentioned scenarios, with particular emphasis on the analysis of persistent halogenated pollutants in complex
food samples. We also introduce a novel concept of “ion mobility windows”, an approach developed on our TIMS-TOF
instrument that uses the correlation between CCS and GC retention times to enhance IM resolving power while avoiding
loss of ion transmission or scan rate.

Please explain why your abstract is innovative for mass spectrometry?

Novel non-targeted analytical strategy that relies on the direct coupling of gas chromatography (GC) with trapped ion
mobility (TIMS)-mass spectrometry to probe halogenated pollutants in complex food/environmental samples with
additional separation, sensitivity and identification capabilities compared to classical methods.

Co-authors:

Georges Scholl, Mass Spectrometry Laboratory - ULiège

Johann Far, Mass Spectrometry Laboratory - ULiège
Alexandre Collgros, Bruker France
Edwin De Pauw, Mass Spectrometry Laboratory - ULiège
Gauthier Eppe, Mass Spectrometry Laboratory - ULiège

310
ACETYLATION IN COMBINATION WITH GAS CHROMATOGRAPHY
COUPLED TO ULTRAHIGH RESOLUTION MASS SPECTROMETRY FOR
THE DETERMINATION OF FUNCTIONAL GROUPS IN COMPLEX MIXTURES
Abstract ID: 658

Presenting author: Diana Catalina Palacio Lozano, University of Warwick

Introduction

Bio-oils are complex mixtures not simply because of the large number of poly-oxygenated compositions, but because
each molecular composition can present many structural arrangements (isomers) with multiple functional groups.
Functional groups are specific rearrangements of atoms within the molecule that confer specific properties and reaction
chemistry. For instance, carbonyl groups and carboxylic acids can promote instability and increased corrosiveness. For
highly complex samples, only chromatographic techniques coupled to ultrahigh resolution mass spectrometry (UHR/MS)
enables isomeric separation while detecting thousands of co-eluting species. These experiments, however, do not reveal
structural information. Herein, the acetylation of a bio-oil with acetic anhydride in combination with hyphenated data sets
is proposed for the determination of alcohols functional groups in complex mixtures such as bio-oils.

Methods

A GC was coupled to an APCI source and a 15T Fourier transform ion cyclotron resonance mass spectrometry (FTICR
MS) resulting in the hyphenated set-up. Direct infusion experiments were performed by using a negative-ion nano ESI
source.

A bio-oil was diluted in three different solvents (methanol, dimethyl sulfoxide, and acetone) to a final concentration of 71
mg/mL. Each dilution was then spiked with 0 mL, 0.36 mL and 0.89 mL of acetic anhydride. The sample were analysed
by GC APCI-FTICR MS at a concentration of 1 mg/mL.

Composer and in-house software, KairosMS, were used for the data processing.

Preliminary data (results)

The reactivity of the bio-oils with acetic anhydride was monitored by direct infusion FTICR-MS. Our preliminary results
shown a shift towards highly oxygenated molecular compositions and higher number of double bond equivalents (DBE)
with the increased volume of acetic anhydride (Figure 1). This indicates, that by increasing the concentration of acetic
anhydride, the acetylation was greatly accelerated. Additionally, a greater reaction of the samples was observed when
diluted in dimethyl sulfoxide.

Here, we proposed the combination of a derivatization method with a hyphenated approach to allow the separation of
isomeric compositions and the identification of the individual isomers that effectively react under the reaction conditions.
More specifically, the method consists in the comparison of GC-FTICR MS data of the bio-oils and the samples spiked
with acetic anhydride to identify isomeric compositions that are only detected before or after the reaction.

Our preliminary data shows that semi-volatile isomers of bio-oils can be clearly separated by GC-FTICR MS. For
instance, 11 isomers were detected for a composition assigned to C 7H8O2 (one of which may represent guaiacol, Figure
2). Additionally, by comparing the extracted ion chromatogram corresponding to this molecular composition, it was
possible to identify three new isomers that are clearly concentrated by increasing the amount of acetic anhydride. A
database/library consisting of retention times, elemental molecular composition and the possible presence of an alcohol
group of each elemental isomer detected will be created to further advance knowledge of structural information of this
bio-oil.

Please explain why your abstract is innovative for mass spectrometry?

A new method for the Identification of functional groups of individual isomers in complex mixtures using GC-FTICR MS

Co-authors:

Hugh E. Jones, University of Warwick

Bryan P. Marzullo, University of Warwick

311
Martin Wills, University of Warwick
Mark P. Barrow, University of Warwick

Class distribution and DBE plots of the bio-oils in DMSO.

Extracted ion Chromatogram of the composition C7H8O2[H]

312
ADVANCES IN THE ION MOBILITY SPECTROMETRY STRATEGIES TO GO
BEYOND SEPARATION
Abstract ID: 686

Presenting author: Darya Hadavi, Maastricht Multimodal Molecular imaging (M4i) Institute, Universiteitssingel
50, 6229ER Maastricht, The Netherlands

Introduction

Ion mobility spectrometry (IMS) is an analytical technique that opens new windows on separation and identifications of
molecules that are not primarily differentiated by mass spectrometry. Even though the application of IMS has been
mainly focused on its separation capability, this techniques enables characterization of molecules in an extra level when
it is combined with tandem mass spectrometry (MS/MS) and different ionization techniques such as electrospray
ionization (ESI) and laser-induced postionization (MALDI-2). Here we are introducing strategies to a) separate isomeric
molecules in a complex sample, b) investigate MS/MS pathways based on adduct ion formation and ions' mobility c)
annotate IMS without requiring standards molecules and d) combine experimental and theoretical data to explain gas
phase behaviour of isomers.

Methods

The IMS investigations have been performed on trapped ion mobility spectrometry (TIMS), travelling wave ion mobility
spectrometry (TWIMS), drift tube ion mobility spectrometry (DTIMS). The ESI and MALDI-2 have been used for
ionization and time of flight was used as mass analyser. Collision-induced dissociation (CID) experiments performed
either before or after the ion mobility cell. The density functional theory (DFT) and collision cross section calculations
were performed on SPARTAN and IMOS (respectively).

Preliminary data (results)

Our results demonstrated the impact of adduct ion formation in the separation of structural and stereoisomers. The
presence of multiple adducts changes the 3D conformation of isomers, enabling separation of them in a complex sample.
When the experimental data were combined with the theoretical DFT predictions and CCS calculation, the gas-phase
behaviour of molecules could be explained and predicted. We also demonstated the importance of performing MS/MS
before IMS to enable the differentiation of structral isomers and even enantiomers. In addition, by doing MS/MS after
IMS, we performed stability test on ions and annotated ion mobility peaks. When these strategies used in flow-chemistry,
we could monitor a model reaction in on-line fashion and identify the elements of the reaction, including the isomeric
reaction products, impurities and side-products. Our findings demonstrate the added value of IMS and MS/MS in different
application areas including metabolites, flow-chemistry, imaging and ionisation research.

Please explain why your abstract is innovative for mass spectrometry?

Separate isomers in a complex sample. Explain MS/MS pathways based on adduct-ions formation. Annotate IMS without
standards. Combine experimental and theoretical-data to explain and predict gas-phase behaviour of isomers

Co-authors:

Marina Borzova, Maastricht Multimodal Molecular imaging (M4i) Institute, Universiteitssingel 50, 6229ER Maastricht, The
Netherlands
Tiffany Porta Siegel, Maastricht Multimodal Molecular imaging (M4i) Institute, Universiteitssingel 50, 6229ER Maastricht,
The Netherlands
Maarten Honing, Maastricht Multimodal Molecular imaging (M4i) Institute, Universiteitssingel 50, 6229ER Maastricht, The
Netherlands

313
CHARACTERIZATION OF THE NANOCEASY CE-MS INTERFACE:
ANALYTICAL PROPERTIES AND FLOW RATES OF THE NANOFLOW
SHEATH LIQUID COUPLING
Abstract ID: 618

Presenting author: Jasmin Schairer, Aalen University

Introduction

CE-MS is a powerful tool in various fields, such as protein and metabolite analysis. For many years the triple tube sheath
liquid interface has been the standard technology due to its ease-of-use, flexibility and robustness. However, in recent
years nanoESI interfaces have been developed enabling strong improvements in sensitivity. We have recently
introduced the nanoCEasy interface, an easy-to-use, robust and flexible interface with nanoESI sensitivity and an elegant
valving functionality (https://doi.org/10.1021/acs.analchem.1c03213). Here we present key properties of the interface,
such as robustness and spray stability. Furthermore, the flow rate is determined for the first time for such a glass emitter-
based nanoSL-interface using isotopically labeled analytes. The dependency on emitter size and applied voltage is
studied in detail.

Methods

A CE was coupled to a QTOF MS via the nanoCEasy interface, consisting of a glass emitter with an opening of 10-40µm
which is inserted in a plug-and-play polymer block. Both, sheath liquid (SL) and separation capillary are inserted into the
emitter.

For the determination of the flow rate atrazine was added to the SL. 5D-labeled atrazine was pumped through the CE
capillary. Different emitter sizes and voltages for ESI pos. were investigated.

Robustness and spray stability were evaluated using several amino acids and a BSA digest. An acidic system was used
for the SL and BGE.

Preliminary data (results)

The robustness of the interface was investigated using the critical parameters (i) ESI voltage, (ii) distance between
emitter tip and MS orifice and (iii) distance between separation capillary and emitter tip. The interface is shown to be
robust regarding all tested parameters and reproducible regarding different emitters. Overall, this leads to long term
stability of the spray.

The nanoCEasy interface has a higher sensitivity than the commercially available triple tube sprayer, regardless of the
used emitter. This is the consequence of the lower analyte dilution correlating with the flow rate of the ESI. The flow rate
is expected to be dependent on ES voltage and emitters size in the open nanoCEasy interface. In our experiments the
flow rate was determined based on the isotopic ratio of atrazine in the SL and D5-atrazine flushed through the CE
capillary with a defined pressure.The measured ES flow strongly depends on the ES voltage. In contrast, the size of the
emitter opening has only a limited influence. At low voltages, the ES flow is nearly independent of the emitter orifice.
Nevertheless, smaller emitters, can be operated at lower ES voltage, which results in a lower flow rate. Here, we present
and discuss in detail the relationship between flow rate and sensitivity and the consequences regarding different CE
separation conditions.

Please explain why your abstract is innovative for mass spectrometry?

· Function and benefits the nanoCEasy CE-MS interface

· Demonstration of the robustness and spray stability of the nanoCEasy

· First accurate flow rate determination of CE-nanoSL-MS interface using isotopically labeled analytes

Co-authors:

Ralf Höneise, Aalen University

Alexander Stolz, Aalen University

314
Johannes Schlecht, Aalen University
Christian Neusüß, Aalen University

Schematic view of the nanoCEasy interface

315
3D-PRINTED OPEN PORT PROBE-ELECTROSPRAY INTERFACE FOR
HIGH-THROUGHPUT FLOW INJECTION OR LIQUID CHROMATOGRAPHY
ANALYSIS
Abstract ID: 529

Presenting author: Xiaobo Tian, Life Sciences Mass Spectrometry, Department of Inorganic and Analytical
Chemistry, University of Geneva, Switzerland

Introduction

The widely used flow-injection mass spectrometry directly delivers the analytes into the ionization source without
chromatographic separation prior to the mass analysis yet it suffers from carry-over. As a variation of flow-injection, open
port probe (OPP), usually applied with self-aspirating liquid introduction ionization source such as electrospray ionization
(ESI), possesses the merit of self-cleaning that efficiently minimizes the carry-over issue (Figure 1A). Three-dimensional
printing (3DP) is an accessible approach for analytical chemists to bring their virtual conception into physical form. OPP
can also be used as an interface for micro-liquid chromatography (µLC) where it enables to decouple the LC mobile
phase conditions from conditions necessary for optimal ESI signal.

Methods

Stereolithography (SLA) 3DP was applied to construct OPP (Figure 1B) and other accessories like the liquid handler.
The merit of self-cleaning was demonstrated by comparison with flow-injection analysis (FIA). The application of 3DP-
OPP device for high-throughput measurements with and without differential ion mobility as well as the coupling with µLC
was exemplified with the detection of illegal drugs and pesticides in various matrices.

Preliminary data (results)

In the comparison between FIA-MS and 3DP-OPP-MS, the sample of benzoylecgonine showed a broad peak with tailing
lasting more than one minute for the former, while the latter generated a sharp Gaussian-like shape of signal that lasts
around 15s (Figure 2). The OPP system was utilized for analysis of cocaine and benzoylecgonine in urine, as well as
amphetamines in plasma. Carry-over was completely removed by using disposable pipette tips. 3D printing technology
enables optimization of interface and accessories design for high-throughput analysis as well as for µLC in various
modes, where it enables to decouple the mobile phase for LC separation from that optimal for ESI ionization.

Please explain why your abstract is innovative for mass spectrometry?

3D-printed-open port probe interface for high-throughput analysis or liquid chromatography-mass spectrometry of
complex samples

Co-authors:

Piotr Sosnowski , Life Sciences Mass Spectrometry, Department of Inorganic and Analytical Chemistry, University of
Geneva, Switzerland
Gérard Hopfgartner , Life Sciences Mass Spectrometry, Department of Inorganic and Analytical Chemistry, University of
Geneva, Switzerland

316
Figure 1. A) scheme of OPP-ESI-MS; B) 3DP-OPP

Figure 2. 3DP-OPP-MS VS flow-injection MS

317
Session: LS-13 Metabolomics - Session A

HIGH RESOLUTION TAILORED METABOLOMICS IN THE FOOD-

NUTRITION-HEALTH CHEMICAL CONTINUUM
Abstract ID: 796

Presenting author: Philippe Schmitt-Kopplin, Helmholtz Munich, Technical University Munich

Introduction

Metabolomics, as the comprehensive study of metabolic reactions in complex dynamic/living systems and thus in
holobionts is growing very rapidly and integrate analytical approaches (LC-MS, NMR and ICR-FT/MS) covering the
description of only 10% of the experimental signals in databases. Important approaches thus are related to the
description of the dark metabolome with adapted strategies. Especially direct injection FTICR/MS enables a high
throughput description of highly complex mixtures and Holometabolomes at the level of the elementary composition
space. FTICR/MS will be presented is a strong tool to understand the chemical diversity in various study fields from food
chemistry, biology, microbiomes towards the discovery of new bioactives.

Methods

FTICR-MS, LC/MS

Preliminary data (results)

Schmitt-Kopplin Ph. , D. Hemmler, F. Moritz, R. D. Gougeon, M. Lucio, M. Meringer, C. Müller, M. Harir, N. Hertkorn,
“Systems Chemical Analytics: Introduction to the challenges of chemical complexity analysis”, Faraday discussions,
2019, DOI: 10.1039/c9fd00078j

Laber, S., S. Forcisi, L. Bentley, J. Petzold, F. Moritz, K. S. Smirnov, L. Al Sadat, I. Williamson, S. Strobel, T. Agnew, S.
Sengupta, T. Nicol, H. Grallert, M. Heier, J. Honecker, J. Mianne, L. Teboul, R. Dumbell, H. Long, M. Simon, C. Lindgren,
W. A. Bickmore, H. Hauner, P. Schmitt-Kopplin, M. Claussnitzer and R. D. Cox (2021). "Linking the FTO obesity
rs1421085 variant circuitry to cellular, metabolic, and organismal phenotypes in vivo." Science Advances 7(30).

Nina Sillner, Alesia Walker, Marianna Lucio, Tanja Verena Maier, Monika Bazanella, Michael Rychlik, Dirk Haller,
Philippe Schmitt-Kopplin, “Longitudinal profiles of dietary and microbial metabolites in formula-and breastfed
infants”, Frontiers in molecular biosciences, 202, 490, 8.

Berger, M. T., D. Hemmler, A. Walker, M. Rychlik, J. W. Marshall and P. Schmitt-Kopplin (2021). "Molecular
characterization of sequence-driven peptide glycation." Scientific Reports 11(1).

Pieczonka S., S. Paravicini, M. Rychlik, Ph. Schmitt-Kopplin “On the trail of the German purity law: distinguishing the
metabolic signatures of wheat, corn and rice in beer”, 2021, Frontiers in Chemistry, 9, 715372

Please explain why your abstract is innovative for mass spectrometry?

FTICR-MS in metabolomics within the Food/Nutrition-Health continuum

318
INFECTION METALLOMICS - BASED DIFFERENTIATION OF
ASPERGILLUS FUMIGATUS COLONIZATION AND INVASION
Abstract ID: 362

Presenting author: Rutuja H. Patil, Institute of Microbiology of the Czech Academy of Sciences, Prague, Czech
Republic, Department of Analytical Chemistry, Faculty of Science, Palacký University, Olomouc, Czech Republic

Introduction

Aspergillus fumigatus is an opportunistic fungal pathogen whose colonization in immunocompromised patients can
develop into invasive pulmonary aspergillosis causing 200,000 deaths in 2019. However, the distinction between
colonization and infection caused by A. fumigatus, in a host remains challenging. Iron, an essential trace element, is
required for A. fumigatus conidial germination, which is a crucial process for initiating fungal proliferation. As a result, A.
fumigatus secretes high-affinity ferric ion-specific chelators, called siderophores, for iron acquisition and storage.
Infection metallomics (IM) was recently introduced as a mass spectrometry platform based on the central concept that
microbial metallophores are specific, sensitive, noninvasive and promising biomarkers of invasive infectious diseases.
Using IM, fungal growth dependent siderophores production can provide a new borderline between asymptomatic
colonization and invasive infection.

Methods

A human isolate of A. fumigatus was grown in an iron limited mineral medium at 37 °C for 72 h. Germination of conidia
was observed using bright field microscopy. A. fumigatus specific intracellular ferricrocin (FC), hydroxy-ferricrocin (HFC)
and extracellular triacetylfusarinine C (TafC) and fusarinine C (FusC) siderophores were extracted using liquid-liquid
extraction and quantified by liquid chromatography and Fourier-transform ion cyclotron resonance mass spectrometry.
Iron-containing species were determined by isotope data filtering. Qualitative and quantitative data analysis was
performed using in-house Cyclobranch version 2.0.19 and DataAnalysis v.5.0 (Bruker Daltonics, Germany) software,
respectively.

Preliminary data (results)

Different morphotypes can be distinguished during conidial germination: dormancy, isotropic growth (conidia doubled in
size) and polarized growth (formation of germ tubes) (Figure 1a). HFC essential for conidial iron storage was
biosynthesized during the non-pathogenic dormancy. From isotropic to polarized phases (6-10 h), conidia notably
increased the expression of both intracellular and extracellular siderophores (Figure 1b). FC was gradually increased
from isotropic growth (6 h), from 12-72 h approximately 1.6-3 mg/g concentration of FC was accumulated in mycelia and
considerable amounts were also detected in culture supernatants (Figure 2). Extracellular TafC secretion was triggered
from 8 h during polarized growth of conidia. Most of the TafC was quantified in its main cellular location supernatant, with
a rapid increase over the time period of 24 h up to the concentration of 0.5 mg/mL (Figure 2b). We also detected FusC
from 6 h of incubation at isotropic phase while TafB, a derivative of TafC, was observed in the late polarized stage (9 h).
Therefore, each A. fumigatus morphotype demonstrated both qualitatively and quantitatively different profiles of
siderophores. Our results showed that HFC and FC indicated the beginning of germination and isotropic growth
of A.fumigatus, respectively, whereas TafC and TafB produced during later stages of germination marked the phase of
proliferation. Further, this transition from a non-pathogenic to a pathogenic state can be monitored in animals and
humans using mass spectrometry.

Acknowledgment

The authors gratefully acknowledge the support from the Czech Science Foundation (21-17044S).

Please explain why your abstract is innovative for mass spectrometry?

1. Use of mass spectrometry based infection metallomics platform for identification and quantification of
siderophores.
2. Distinction of asymptomatic colonization and symptomatic infection of A. fumigatus using metallomics.

Co-authors:

Dominika Luptáková, Institute of Microbiology of the Czech Academy of Sciences, Prague, Czech Republic
Radim Dobiáš, Department of Bacteriology and Mycology, Public Health Institute in Ostrava, Ostrava, Czech Republic,
Department of Biomedical Sciences, Faculty of Medicine, University of Ostrava, Ostrava, Czech Republic
David A. Stevens, California Institute for Medical Research, San Jose, CA, USA, Division of Infectious Diseases and
319
Geographic Medicine, Stanford University School of Medicine, Stanford, CA, USA
Tomáš Pluháček, Institute of Microbiology of the Czech Academy of Sciences, Prague, Czech Republic, Department of
Analytical Chemistry, Faculty of Science, Palacký University, Olomouc, Czech Republic
Andrea Palyzová, Institute of Microbiology of the Czech Academy of Sciences, Prague, Czech Republic
Vladimír Havlíček, Institute of Microbiology of the Czech Academy of Sciences, Prague, Czech Republic, Department of
Analytical Chemistry, Faculty of Science, Palacký University, Olomouc, Czech Republic

(a) Transition of germinating conidia (b) Siderophore production during germination.

Quantitative siderophore analysis in A. fumigatus (a) mycelia (b) supernatant.

320
THE UNEXPLORED WORLD OF NON-CANONICAL FATTY ACIDS
Abstract ID: 472

Presenting author: Jan Philipp Menzel, School of Chemistry and Physics and the Central Analytical Research
Facility, Queensland University of Technology, Centre for Materials Science, Queensland University of
Technology, Centre for Data Science, Queensland University of Technology

Introduction

The recent discovery of fatty acids with unusual double bond positions in human cells and tissues suggests an
incomplete picture of lipid metabolism in human, plant and animal biology and highlights the need for new analytical tools
for de novo lipid discovery. Ozone-induced dissociation (OzID) has shown significant potential for structural elucidation of
fatty acids where a key advantage is the ability to predict fragmentation patterns based on the well-described reaction of
ozone with carbon-carbon double bonds. Here we present an end-to-end workflow for fatty acid analysis based on lipid
extraction, fixed-charge derivatization, liquid chromatography-OzID-mass spectrometry, and automated data analysis
without requiring analytical standards or libraries (de novo discovery). Data analysis is performed with custom-written
python code as well as the Skyline Mass Spectrometry environment.

Methods

Lipids are extracted by MTBE-based methods, subjected to hydrolysis and derivatization with 1-(4-(amino-
methyl)phenyl)pyridinium. LC-MS was performed using a Waters Acquity (UPLC Class 1; CSH, C18 reverse phase, 2.1 x
100 mm, 1.7 mm, linear gradient from 20% B (ACN + 0.1% formic acid) and 80% A (H2O + 0.1% formic acid) to 100% B)
liquid chromatography system coupled with a Waters SYNAPT G2Si (T-Wave Ion Mobility; TOF) mass
spectrometer. The IMS cell of the instrument is filled with ozone enriched oxygen. Dataset analysis is a stepwise process
using Windows Batch Files, Skyline Runner and python scripts.

Preliminary data (results)

Quantification of charge-tagged fatty acids is possible by direct infusion ESI-MS. The fixed positive charge minimises
ionization bias and direct infusion (without separation) ensures equivalent spray conditions for each fatty acid, enabling
quantitation. Double bond positions are assigned, and isomers quantified, based on an LC-OzID-MS (& MS/MS) analysis
of the same sample.The data analysis workflow assigns double bond positions from the detected aldehyde and Criegee
product ions (i.e., ozonolysis products). We assessed pooled human plasma, identifying fatty acids previously reported
from GC- and LC-based analyses, as well as an unexpected variety of hitherto unreported species with non-canonical
positions of unsaturation. While only a few dozen fatty acids are commonly identified in a standard reference material
(NIST 1950 SRM, frozen metabolites in pooled human plasma), we gained evidence of more than 170 double bond
isomers therein. Most of these have been described in different biological samples, whereas a small number has never
been reported to date. Most novel fatty acid identifications arise from low abundant double bond isomers that coelute
with highly abundant known isomers. Preliminary investigations of cancer cell lines show that unusual fatty acid isomers,
previously reported based on lengthy manual searches, were found automatically through our new method. Furthermore,
several additional fatty acid isomers were detected that have not previously been reported in mammalian tissue or cell
lines and are not yet included in the LIPID MAPS database.

Please explain why your abstract is innovative for mass spectrometry?

The automated workflow creates comprehensive target lists (>1000 targets) for LC-OzID-MS/MS acquisition, leading to
spectra for each fatty acid isomer containing characteristic fragments detected at S/N >> 3.

Co-authors:

Reuben S.E. Young, School of Chemistry and Physics and the Central Analytical Research Facility, Queensland
University of Technology, Centre for Materials Science, Queensland University of Technology
Aurelie H. Benfield, School of Biomedical Sciences, Faculty of Health, Queensland University of Technology, Princess
Alexandra Hospital, Translational Research Institute
Sonia T. Henriques, School of Biomedical Sciences, Faculty of Health, Queensland University of Technology, Princess
Alexandra Hospital, Translational Research Institute
Berwyck L.J. Poad, School of Chemistry and Physics and the Central Analytical Research Facility, Queensland
University of Technology, Centre for Materials Science, Queensland University of Technology
Stephen J. Blanksby, School of Chemistry and Physics and the Central Analytical Research Facility, Queensland
University of Technology, Centre for Materials Science, Queensland University of Technology

321
Discovery of fatty acid isomers by UPLC-OzID-MS/MS.

322
METHOD DEVELOPMENT AND TROUBLESHOOTING OF METABOLOMIC
MALDI IMAGING REVEALS SAMPLE PREPARATION DEPENDENT TISSUE
SPECIFICITIES
Abstract ID: 575

Presenting author: Marlene Leibetseder, Technische Universität Wien

Introduction

Metabolites play a crucial role in disease development. Concerning healthcare approaches, a paradigm change from
reactive medicine to predictive approaches, targeted prevention, and fundamental molecular understanding of pathology
is highly desirable. In parallel, mass spectrometry-based imaging (MSI) has seen a rapid development as it provides a
platform to investigate molecular compositions and alterations within a biological sample for a variety of diseases.

Comprehensive metabolic profiling is a powerful tool for identification of disease affiliated metabolites or
targets. However, metabolites exist in a very broad range of concentrations and exhibit remarkable chemical diversity.
We evaluated and present sample preparation strategies for different tissues taking a critical look at metabolite detection
and distribution.

Methods

Fresh frozen kidney, liver and prostate tissues (kindly provided by L.Kenner and I. Yotova, Med. Univ. Vienna) were cryo-
sectioned, cut, mounted on glass slides, sprayed with different matrices and Girard T (GirT) for derivatization, using
different devices (TM/HTX, Aero/Shimadzu). Samples were analyzed using a Synapt G2 HDMS (Waters) or MALDI 7090
(Shimadzu). Raw files generated were loaded into MSI reader (NC State Univ.) or Imagereveal (Shimadzu). Compounds
were identified with Metaspace annotation platform (https://metaspace2020.eu) and LipidMaps
(https://www.lipidmaps.org), statistics performed using Perseus (MPI) and Origin Pro v. 9.8.5.204. A schematic overview
of the workflow is shown in figure 1.

Preliminary data (results)

In this study, we compare different MSI based metabolomic approaches (targeted vs. untargeted) in tissues, coming from
various cancerous (lung cancer, prostate cancer) as well as chronic inflammatory diseases (endometriosis).

We show how different sample preparation methods such as varying tissue thickness, cutting methods as well as varying
matrix application methods influence the metabolomic read-out, depending on the origin of the tissue. We observed
changes in signal-to-noise ratio for high and low abundant primary metabolites analyzing the same tissue; also the type
of matrix, the solvent and spraying conditions highly influence ionization and extraction efficiencies generating “different”,
yet to some extent “similar” images. Here for example organic solvent dependent extraction efficiency is a crucial
influence for apolar secondary metabolites.

We also apply derivatization (GirT) to enable the detection of apolar, secondary metabolites like corticosteroids (e.g.
cortisone) and hormones at physiologically relevant concentrations. We test limits of detections for different tissue types
to get assess limitations for spatially resolved metabolite analysis using MALDI MSI.

Please explain why your abstract is innovative for mass spectrometry?

Insights into tissue-specific steroid/sterol biology and disease affiliated cortisol production in a systematic approach with
method development using different matrices for MALDI.

Co-authors:

Peter Sandbichler, Technische Universität Wien

Martina Marchetti-Deschmann, Technische Universität Wien

323
Fig. 1 Graphical abstract: Schematic of workflow (created with BioRender)

324
AN INTERPLATFORM IM-MS STUDY OF STEROID COLLISION CROSS
SECTIONS (CCSS) ACROSS THREE DIFFERENT IM-MS TECHNOLOGIES
Abstract ID: 646

Presenting author: Tim Causon, University of Natural Resources and Life Sciences, Vienna, Department of
Chemistry, Institute of Analytical Chemistry, Muthgasse 18, 1190 Vienna, Austria

Introduction

One of the most recent analytical technologies applied to steroid analysis is ion mobility-mass spectrometry (IM-MS). In
particular, ion collision cross section (CCS) obtained from different types of commercial IM-MS instrumentation has been
proposed as an ideal additional identification point for steroid analysis as the long-term repeatability and interlaboratory
reproducibility are excellent, while matrix effects are generally absent or low. However, the routine use of CCS across the
major IM-MS technology platforms (i.e., DTIM-MS, TWIM-MS and TIM-MS) requires assessment of the analytical
performance and ion conformation dissimilarities that may be influenced by different instrument characteristics,
calibration methods and IM transport mechanisms specific to each instrument type.

Methods

A comprehensive interlaboratory and cross-platform comparison of 142 different CCS values derived from drift tube
(DTIM-MS), travelling wave (TWIM-MS) and trapped ion mobility (TIM-MS, supported by Bruker Daltonik) platforms with
87 steroids and steroid phase II metabolites was undertaken. Experimental datasets from analytical standards and
spiked urine samples were generated using a standardized LC-IM-MS methodology adapted to each IM-MS platform.
Statistical analysis was used to assess analytical performance of each platform including the identification of
conformational outliers, ruggedness of external CCS calibration, and respective influence upon identification via
database matching. Finally, computational approaches (DFT) were used to investigate conformational outliers.

Preliminary data (results)

The new experimental dataset provides a set of three instrument-specific CCS databases covering 142 different ion
species of steroids and phase II metabolites. All instrument platforms studied provided excellent stability; 95% of ions
had CCS bias within ±1% for TIM-MS and within ±2% for TWIM-MS with respect to DTIM-MS data and minimal or no
matrix effects were observed in spiked urine samples. However, differences in CCS observed on each platform for
individual ions demand for detailed evaluations with respect to the external calibration, ion species, and IM transport
mechanism.

For example, to assess trends with respect to the transport mechanism of IM separation, bias data was also assessed
using a modified CCS (i.e., CCS’ = CCS ∙√µ). This reflects the IM separation order and is used to study correlation of
bias observed on different platforms with respect to CCS’ according to ion species. This provides key hints in our
attempts to delineate the influences of external calibration from true conformational differences realized on different IM-
MS platforms, which require detailed computational studies for confirmation.

Finally, the current potential of using CCS databases applied to single or interplatform assessments is critically
assessed. The identity confirmation of different isomeric ions within the dataset can be considered in terms of typically
applied arbitrary thresholds (i.e., 1-2% difference) or with reference to uncertainty estimates applied to DTCCSN2 values
revealing both the challenges of using CCS for standards-free isomer differentiation, and the as-yet unresolved
discrepancies appearing in different IM-MS platforms (Figure 1).

Please explain why your abstract is innovative for mass spectrometry?

A first comparison of extensive experimental CCS datasets from all three major IM-MS technologies applied to small
molecule identity confirmation workflows (case study steroids).

Co-authors:

Max L. Feuerstein, University of Natural Resources and Life Sciences, Vienna, Department of Chemistry, Institute of
Analytical Chemistry, Muthgasse 18, 1190 Vienna, Austria
Maykel Hernández-Mesa, LABERCA, Oniris, INRAE, 44307 Nantes
Younes Valadbeigi, University of Natural Resources and Life Sciences, Vienna, Department of Chemistry, Institute of
Analytical Chemistry, Muthgasse 18, 1190 Vienna, Austria
Bruno Le Bizec, LABERCA, Oniris, INRAE, 44307 Nantes
Stephan Hann, University of Natural Resources and Life Sciences, Vienna, Department of Chemistry, Institute of
325
Analytical Chemistry, Muthgasse 18, 1190 Vienna, Austria
Gaud Dervilly, LABERCA, Oniris, INRAE, 44307 Nantes

Comparison of DTCCSN2, TIMCCSN2 and TWIMCCSN2 data for isomer examples.

326
NON-INVASIVE MONITORING OF SHORT-CHAIN FATTY ACIDS IN
EXHALED BREATH USING PROTON TRANSFER REACTION – TIME-OF-
FLIGHT – MASS SPECTROMETRY
Abstract ID: 599

Presenting author: Joris Meurs, Department of Analytical Chemistry and Chemometrics, Institute for Molecules
and Materials, Radboud University

Introduction

Volatile organic compounds (VOCs) in exhaled breath contain rich information about the metabolic processes in the
body, allowing to monitor health status and the response to various stimuli. We used proton transfer reaction – time-of-
flight – mass spectrometry (PTR-ToF-MS) to monitor changes in the breath profile of participants during a multi-day
walking event (Four Days Marches 2018, 2019) in real-time. The participants were included in two studies: one
comparing statin and non-statin users and one on inflammatory bowel disease (IBD). No restrictions were placed on the
participants regarding consumption of food and drinks. The aim of this work was to 1) identify signatures of how a
pathological state affects the response to exercise and 2) validate the sampling and analysis of breath biomarkers.

Methods

Participants were asked to exhale directly into the PTR-Tof-MS via a breath sampler or provide a single exhalation into a
Tedlar® bag following a standard protocol. The detected ions were subjected to multivariate analysis (ANOVA
simultaneous component analysis (ASCA)) and multilevel partial least squares – discriminant analysis (M-PLS-DA) to
identify significant group differences and the VOCs responsible for them. Validation of compounds of interest was done
using the respective chemical standards.

Preliminary data (results)

We designed and validated an online breath sampling and analysis protocol using PTR-ToF-MS for real-time
measurement in field conditions. This approach allowed a high throughput of samples with participants being able to give
duplicate breath samples in approximately 2 minutes.

The effect of prolonged exercise was reflected in the exhaled breath of participants (Figure 1). Changes in VOCs, such
as acetone, isoprene and short-chain fatty acids (SCFAs), were found in relation to exercise and use of the cholesterol-
lowering drug statin. We showed that the SCFAs (acetic, propanoic and butanoic acid) can be used as indicators for non-
invasive monitoring of changes in human metabolism and especially for the gut microbiome activity in relation to exercise
and drug therapy (Henderson et al., 2021).

Furthermore, butanoic acid was found to significantly correlate with plasma inflammation marker IL-6 (Spearman r =
0.551, p < 0.001). No difference in butanoic acid concentration was found between IBD and non-IBD participants. This
indicates that butanoic acid may be used as a potential non-invasive marker to monitor exercise-induced inflammation
(Henderson et al., submitted)

To validate the protocol for analysis of SCFAs in exhaled breath using PTR-ToF-MS, chemical standards were used to
optimize the instrumental parameters and determine the products of the reaction of SCFAs with reagent ions H 3O+,
O2+ and NO+. Using optimized parameters, exhaled breath samples were used to determine the robustness of the
analysis. The repeatability of the response was within analytically acceptable limits (<15% relative standard deviation) for
each SCFA.

Please explain why your abstract is innovative for mass spectrometry?

An online protocol for PTR-ToF-MS analysis of SCFAs in exhaled breath was developed and validated for field campaign
measurements allowing a step further towards standardization of breath research.

Co-authors:

Ben Henderson, Department of Analytical Chemistry and Chemometrics, Institute for Molecules and Materials, Radboud
University
Evangelia Sakkoula, Department of Analytical Chemistry and Chemometrics, Institute for Molecules and Materials,
Radboud University
Guilherme Lopes-Batista, Department of Analytical Chemistry and Chemometrics, Institute for Molecules and Materials,

327
Radboud University
Carlo G. Bertinetto, Department of Analytical Chemistry and Chemometrics, Institute for Molecules and Materials,
Radboud University
Dušan Materić, Institute for Marine and Atmospheric Research, Utrecht University
Coen C.W.G. Bongers, Department of Physiology, Radboud Institute for Health Sciences, Radboud University Medical
Center
Neeltje A.E. Allard, Department of Physiology, Radboud Institute for Health Sciences, Radboud University Medical
Center
Thijs M.H. Eijsvogels, Department of Physiology, Radboud Institute for Health Sciences, Radboud University Medical
Center
Rupert Holzinger, Institute for Marine and Atmospheric Research, Utrecht University
Frans J.M. Harren, Department of Analytical Chemistry and Chemometrics, Institute for Molecules and Materials,
Radboud University
Jeroen J. Jansen, Department of Analytical Chemistry and Chemometrics, Institute for Molecules and Materials,
Radboud University
Carlijn R. Lamers, Division of Human Nutrition and Health, Wageningen University & Research
Maria T.E. Hopman, Department of Physiology, Radboud Institute for Health Sciences, Radboud University Medical
Center
Simona M. Cristescu, Department of Analytical Chemistry and Chemometrics, Institute for Molecules and Materials,
Radboud University

M-PLS-DA revealing the exercise effect on the breath VOC profile

328
Session: LS-14 Cross-omics, Data Integration and Bioinformatics for MS

A CLOUD-SCALABLE SOFTWARE SUITE FOR LARGE-SCALE

PROTEOGENOMICS DATA ANALYSIS AND VISUALIZATION.
Abstract ID: 107

Presenting author: Margaret Donovan, Seer

Introduction

Assessment of the flow of genetic information through multi-’omic data integration can reveal the molecular
consequences of genetic variation underlying human disease. Next-generation sequencing (NGS) can be used to identify
genetic variants, while mass spectrometry can be used to assess the proteome. The Proteograph™ Product Suite,
leverages multiple nanoparticles with distinct physiochemical properties to enable large-scale, deep plasma proteome
analyses. Integration of proteomics and genomics data requires many tools of which require complex workflows that can
act as a barrier for researchers to adapt new analysis tools. Here, we present a cloud-based, data analysis software
platform called Proteograph Analysis Suite (PAS) for proteogenomic data analyses through the integration of
Proteograph proteomics data with NGS variant information.

Methods

PAS includes an experiment data management system, analysis protocols, analysis setup wizard, and result
visualizations. PAS can support both Data Independent Analysis (DIA) and Data Dependent Analysis (DDA) workflows
and can integrate variant call format (.vcf) files, enabling personalized database searches. PAS includes metrics for
identified peptides and protein groups like intensity, protein sequence coverage, abundance distributions, and counts.
Visualizations, including PCA, hierarchical clustering, and heatmaps highlight experimental trends. To enable biological
insights, differential abundance results are displayed as volcano plots, protein interaction maps, and protein-set
enrichment. PAS provides an easy-to-use and efficient suite of tools to enable data analysis.

Preliminary data (results)

Here, we apply PAS by analyzing 141 Proteograph NSCLC plasma dataset1 and performing a database search). This
search was launched through the user interface requiring only 3 clicks, and in the background this search provisioned
142 servers and completed in approximately five and half hours. Together, these results show the utility of PAS for
seamless and fast proteomic data analysis.

Reference: 1Blume, J. E. et al. Nature Communications, 2020

Please explain why your abstract is innovative for mass spectrometry?

We present a cloud-based, proteomic and proteogenomic data (MS and NGS) analysis and visualization software
platform called Proteograph Analysis Suite (PAS).

Co-authors:

Harsharn Auluck, Seer

Arjun Vadapalli, Seer
Yan Berk, Seer
Aaron Gajadhar, Seer
Khatereh Motamedchaboki, Seer
Yuandan Lou, Seer
Theo Platt, Seer
Asim Siddiqui, Seer

329
EXPERIMENTAL REPRODUCIBILITY LIMITS THE CORRELATION
BETWEEN MRNA AND PROTEIN ABUNDANCES IN TUMOUR PROFILES
Abstract ID: 369

Presenting author: Swathi Ramachandra Upadhya, Systems Biology Ireland, University College Dublin, School
of Computer Science, University College Dublin

Introduction

mRNAs, the precursors of proteins, have long been used as proxies for protein measurements. With recent
advancement in protein quantifying technologies, large-scale studies of human proteomes have revealed only a
moderate correlation between mRNA and protein abundances. It is unclear to what extent this moderate correlation
reflects post-transcriptional regulation and to what extent it reflects measurement error.

Methods

We applied a standardized pipeline to assess the mRNA-protein correlation across 13 different studies. We primarily
focussed on analysing studies with replicate proteomic profiles and obtained an aggregate reproducibility score’. We
compared the variation in mRNA-protein correlation that was explained by biological (protein complex membership) and
technical errors or noise (protein-reproducibility score). We also looked at studies with replicate transcriptomic profiles
and its impact on mRNA-protein correlation briefly.

Preliminary data (results)

We show that there is considerable variation in the reproducibility of measurements of individual proteins. Proteins with
more reproducible measurements tend to have higher mRNA-protein correlation, suggesting that a substantial fraction of
the unexplained variation between mRNA and protein abundances may be attributed to limitations in the experimental
reproducibility of proteomic and transcriptomic quantification. We also find that proteins that have high reproducibility in
one study tend to have high reproducibility in others and exploit this to develop an 'aggregate protein reproducibility'
score. This score can explain a substantial amount of the variation in mRNA-protein correlation across multiple studies of
both healthy and tumour samples. Finally, we show that pathways previously reported to have higher-than-average
mRNA-protein correlation may simply contain members that can be more reproducibly quantified.

A preprint of the paper is uploaded on biorxiv and can be viewed here

(https://www.biorxiv.org/content/10.1101/2021.09.22.461108v1) for more details.

Please explain why your abstract is innovative for mass spectrometry?

We highlight the importance of protein reproducibility in MS and it's impact on mRNA-protein correlation.

Co-authors:

Colm J Ryan, Systems Biology Ireland, University College Dublin, School of Computer Science, University College
Dublin

330
REVEALING THE MOLECULAR UNIVERSE OF THE HUMAN KIDNEY WITH
MALDI-MSI: FROM SPATIAL METABOLOMICS TO SPATIAL GLYCOMICS
Abstract ID: 443

Presenting author: Christopher Anderton, Pacific Northwest National Laboratory

Introduction

The Kidney Precision Medicine Project (KPMP) consortium, in part, aims to create a human kidney tissue atlas from
evaluating healthy and diseased biopsies. We have developed and optimized matrix-assisted laser desorption/ionization
mass spectrometry imaging (MALDI-MSI)-based spatial metabolomics, lipidomics, and N-glycomics assays for analysis
of human kidney biopsy tissues obtained from tissue recruitment sites (TRS) across the USA. The KPMP TRS acquire
tissue from patients with varying disease states (e.g., Acute kidney injury, Chronic kidney disease, and Diabetic kidney
disease). We have also linked our data to other omics-based analyses being performed within the consortium. By
combining data obtained within our tissue integration site (TIS) with data from other TISs, we have begun to identify key
molecular markers of cell types and disease states.

Methods

Fresh frozen and FFPE biopsy and nephrectomy samples were obtained internally, the University of Michigan, and from
KPMP TRS. Sectioning, matrix application (HTX TM-Sprayer), and high mass resolution MALDI analysis was performed
using a Bruker 15T-FTICR-MS and a Thermo Orbitrap QE HF-X MS equipped with a MALDI source (Spectroglyph LLC).
For spatial N-glycomics, peptide-N-glycosidase F was applied with the TM-Sprayer. All imaging data was converted to
imzML and uploaded to METASPACE for putative annotations. Metabolomics and lipidomics annotations were confirmed
via LC-MS/MS analysis. Autofluorescence- and histochemical-based microscopy was performed to correlate different
kidney tissue features to the MSI data.

Preliminary data (results)

To date, we have analyzed over 600 human kidney sections from nearly 40 different patients. As part of this effort, we
have created best practices for tissue handling, developed QC and QA metrics, and established metabolite, lipid, and N-
glycan markers for different compartments within the kidney. These data are open and readily accessible, and they can
be found in METASPACE (https://metaspace2020.eu/) under the NIH KPMP project. Using the custom CoreMetabolome
and the SwissLipids databases, we regularly annotate over 400 and 250 metabolite and lipids in METASPACE,
respectively, using our optimized protocols. To build robustness into our analysis, we developed data-based QCs to
establish confidence in longitudinal studies. We are now expanding our efforts to lipids that more readily ionize in
negative ion mode. We recently developed a N-glycan database (NGlycDB) that can be utilized by researchers
worldwide for annotating and visualizing MSI-based N-glycan results across all sample types (Veličkovićet al., Anal.
Chem. 2021). Using, NGlycDB we determined that 49 common N-glycans can be seen across human kidney tissue. By
correlating MALDI-MSI data generated from our TIS with regional proteomics and single-cell RNA sequencing data of
tissue generated at other KPMP TISs from the same patient, we were able to validate metabolite, lipid, and N-glycan
markers for podocytes. For example, we have identified N-palmitoylsphingomyelin as a near perfect correlate with
healthy glomerular kidney regions and it is correlated with the Cers6 enzyme activity. We are now beginning to explore
spatial O-glycomics and proteomics with MALDI-MSI of human kidney tissue.

Please explain why your abstract is innovative for mass spectrometry?

Our robust MALDI-MSI methods for obtaining spatial omics data from human kidney tissue has revealed cell and
potential disease biomarkers, providing prospective molecular links to normal and disease states.

Co-authors:

Jessica Lukowski, Pacific Northwest National Laboratory

Dušan Veličković, Pacific Northwest National Laboratory
Annapurna Pamreddy, The University of Texas Health San Antonio
Guanshi Zhang, The University of Texas Health San Antonio
Jeffrey Hodgin, University of Michigan
Ljiljana Paša-Tolić, Pacific Northwest National Laboratory
Theodore Alexandrov, European Molecular Biology Laboratory
Kumar Sharma, The University of Texas Health San Antonio

331
MULTI-OMICS INVESTIGATION OF AMINO ACID DYNAMICS IN
AUTOPHAGY
Abstract ID: 573

Presenting author: Kathrin Thedieck, University of Innsbruck, University of Groningen, Carl von Ossietzky
University Oldenburg

Introduction

Amino acids (AA) have a central role in metabolic homeostasis and signaling. mTORC1 (mechanistic target of rapamycin
complex 1) is a protein kinase complex that integrates metabolic stimuli including AA abundance and controls
metabolism, growth and survival. AA restriction inhibits mTORC1 and enhances autophagy, a catabolic process that
degrades macromolecules in autophagolysosomes and replenishes pools of AA and intermediary metabolites.
Autophagy is tightly controlled with complex cross-regulation at the level of metabolism, signaling, protein synthesis and
degradation. We developed a method for absolute quantification of underivatized amino acids by RP-AEX-PRM-MS,
offering enhanced speed and accuracy. In a simultaneous proteo-metabolomic approach, we investigated and correlated
dynamics of amino acids, proteins, and phosphorylation during autophagy.

Methods

Cell cultures were subjected to repeated shifts from amino acid rich to poor media.

A CHCl3-MeOH based simultaneous proteo-metabolome liquid-liquid extraction (SPM-LLE) was used for simultaneous
AA and protein extraction. 13C,15N-labelled AA standards were spiked in during SPM-LLE. Absolute quantification of
intra- and extracellular AA was performed using a novel RP-AEX-OT-PRM-MS approach. Phosphopeptides were
enriched by IMAC and separated by high pH reversed chromatography prior to LC-MS/MS analysis. Quantitative
proteomics and phosphoproteomics was performed by nanoLC-OT-MS/MS using lable-free and TMT-based
quantification, respectively.

Preliminary data (results)

We present a novel method for the quantitative analysis of intra- and extracellular free amino acids. Mixed mode RP-AEX
resulted in narrow, symmetric peaks for free amino acids including baseline separation of isobaric amino acids. The
method has a limit of quantification in the low micromolar range enabling the quantification of amino acids in biological
samples. Reproducibility of the extraction- and measurement methods were determined to be sufficient for biological
experiments, with CV values lower than 10% obtained for all amino acids in the 13C,15N-labelled standard spiked into cell
culture samples.

Applying this method to cells restricted for amino acids, we investigated changes in intra- and extracellular amino acid
levels over time during autophagy. We identified three clusters of amino acids with unique behavior. Time-dependent
differential proteomics identified 2682 proteins of which 53 were altered. Integrated analysis of amino acid, protein and
signaling dynamics reveal differential dynamic patterns of distinct amino acid pools, indicative of complex regulation of
different branches of amino acid metabolism during autophagy.

Please explain why your abstract is innovative for mass spectrometry?

A new method for absolute quantification of underivatized amino acids by RP-AEX-PRM-MS in combination with
simultaneous proteo-metabolomics opens new avenues to systems studies of signaling and metabolism.

Co-authors:

Anna-Sophia Egger, University of Innsbruck

Madlen Hotze, University of Innsbruck
Anja Reintjes, University of Innsbruck
Alienke van Pijkeren, University of Innsbruck, University of Groningen
Marcel Kwiatkowski, University of Innsbruck

332
SIMULTANEOUS PROTEO-METABOLOMICS REVEALS METABOLIC
SHIFTS IN AN IN VITRO MODEL OF TUBEROUS SCLEROSIS COMPLEX
(TSC)
Abstract ID: 631

Presenting author: Alienke Van Pijkeren, University of Innsbruck, University of Groningen

Introduction

Tuberous Sclerosis Complex (TSC) is a genetic conditioncaused by mutations in the genes coding for TSC1 andTSC2,
which form the TSC protein complex. The TSC complex suppresses the metabolic master regulator mechanistic target of
rapamycin complex 1 (mTORC1).TSC is the most common genetic cause for epilepsy and leads to neuropsychiatric
conditions and benign tumors in almost all organ systems. mTORC1 inhibitors are approved for the treatment of TSC-
related tumors and epilepsy but are not always effective and can cause adverse effects. A detailed understanding of the
TSC-related metabolic shift may reveal new points of intervention for better treatment and quality of life. Towards this
goal, we investigated the polar metabolome and proteome in a TSC in vitro model by simultaneous proteo-metabolomics.

Methods

CRISPR-engineered TSC2-/- (KO) and TSC2+/+ (control) cells, treated with pharmacological inhibitors, were
investigated by simultaneous proteo-metabolomics. The polar metabolome and proteome were extracted from the same
samples by a MeOH-CHCl3 based simultaneous proteo-metabolome liquid-liquid extraction (SPM-LLE). Polar
metabolites were quantified by IC-SIM-MS and RP-AEX-PRM-MS using an Orbitrap HF-X. Label-free differential
proteome analysis was performed by nanoLC-MS/MS using an Orbitrap Fusion Lumos instrument.

Preliminary data (results)

SPM-LLE in combination with IC-SIM-MS and RP-AEX-PRM-MS resulted in quantification of 59 polar metabolites and
simultaneous differential proteome analysis identified 4,557 proteins of which 365 exhibited quantitative changes.
Integrated analysis of metabolites and proteins revealed multiple changes in glycolysis, TCA cycle and amino acids
metabolism.

Please explain why your abstract is innovative for mass spectrometry?

High-resolution, simultaneous quantitative proteo-metabolomics enables deep quantitative analysis of the interactions
between energy metabolism and the proteome, providing new insights into the metabolic shifts associated with TSC.

Co-authors:

Madlen Hotze, University of Innsbruck

Anna-Sophia Egger, University of Innsbruck
Karen van Eunen, University of Groningen
Anja Reintjes, University of Innsbruck
Elisabeth Zimmermann, University of Innsbruck
Marcel Kwiatkowski, University of Innsbruck
Kathrin Thedieck, University of Innsbruck, University of Groningen, University of Oldenburg

333
Session: FP-3 Biosimilars, Biobetters & Glycoengineering

CHARACTERIZATION AND MONITORING OF BISPECIFIC ANTIBODY

VARIANTS BY NATIVE MASS SPECTROMETRY
Abstract ID: 811

Presenting author: Markus Haberger, Roche Diagnostics GmbH

Introduction

Bispecific antibody formats providing novel functionalities compared to classical antibodies. However, these new
antibody formats come along with new analytical challenges and product variants. In this study, an approach employing
size-exclusion and charge variant separation combined with native mass spectrometry for the characterization of
bispecific antibodies variants was applied. Size variants derived from incorrect light or heavy chain association typically
represent critical product-related impurities for bispecific antibody formats. Furthermore, post translational modifications
of bispecific antibodies were investigated. The herein presented native size exclusion and ion exchange separation
online mass spectrometry methods represent a complementary or even alternative approach for multiattribute monitoring
of biologics at the intact level.

Methods

size exclusion chromatography mass spectrometry

charge variant exchange chromatography mass spectrometry

Please explain why your abstract is innovative for mass spectrometry?

methods adopted to bispecific antibodies

334
DISCLOSING THE QUANTITATIVE POTENTIAL OF MIDDLE-UP HILIC-MS
FOR THE N-GLYCAN PROFILING OF THERAPEUTIC MONOCLONAL
ANTIBODIES
Abstract ID: 111

Presenting author: Valentina D'Atri, Institute of Pharmaceutical Sciences of Western Switzerland (ISPSO),
University of Geneva, School of Pharmaceutical Sciences, University of Geneva

Introduction

The glycosylation profile of therapeutic monoclonal antibodies (mAbs) is of crucial importance as it might have a major
role in the stability, immunogenicity and the clinical efficacy of the mAbs. For this reason, it is considered a key
post/translational modification (PTM) and an important critical quality attribute (CQA) that requires comprehensive
characterization. Current reference techniques for glycan analysis are based on laborious and time-consuming workflows
including the release and labelling of the glycans prior to their analytical characterization performed by hydrophilic
interaction chromatography (HILIC) coupled to fluorescence detection (FLD) or mass spectrometry (MS).

In this context, the use of HILIC-MS for the analysis of protein subunits (middle-up approach) has emerged as a powerful
technique for the qualitative glycan analysis of mAbs and mAb-related products.

Methods

Different therapeutic mAb products with distinct glycosylation profiles (e.g., low fucose, bisecting and biosimilar products)
were selected and analyzed using HILIC-MS performed at the subunit level and by using the released N-glycan
approach as a reference method. mAb subunits of approximately 25 kDa were generated through a rapid enzymatic
digestion with IdeS and chemical reduction of the disulfide bonds using DL-dithiothreitol (DTT). Released N-glycans were
obtained by using the PNGase F enzyme and then labelled with two different approaches (2-AB and RFMS).

Preliminary data (results)

The aim of this project was to evaluate the quantitative performance of the middle-up HILIC-MS approach in comparison
to the reference released 2-AB and RFMS N-glycan approaches. First, adalimumab, a well characterized mAb, was
analyzed to compare the relative abundance levels obtained from the subunit mass spectra with the glycan abundance
levels obtained from 2-AB and RFMS released glycan profiling. At subunit level, a relative quantification based on the
summation of the extracted ion chromatograms (XIC) of three different charge states of the ions corresponding to the
subunits carrying different glycoforms was performed. Overall, a good correlation with the reference 2-AB method was
found. In addition, the subunit analysis enabled the detection of other PTMs, such as lysine clipping and glycation. Then,
to demonstrate the potential of the quantitative middle-up HILIC-MS approach for more complex samples, mAbs with
different N-glycan characteristics were selected, such as the glyco-engineered products atezolizumab (aglycosylated),
benralizumab (low fucose), and obinituzumab (bisecting and low fucose). In addition, the innovator product of infliximab
was compared with two biosimilar products that have known differences in the N-glycosylation profile.

It was observed that HILIC-MS analysis performed at the middle-up level not only provided a means for a fast qualitative
evaluation of mAb glycosylation profiles but also the basis of a robust relative quantification technique by also holding a
great potential as multi-attribute monitoring method.

Reference: Duivelshof et al., Pharmaceutics. 13, 1744 (2021).

Please explain why your abstract is innovative for mass spectrometry?

Independent XIC profiles of co-eluting glycoforms were extracted and used to calculate the relative abundance levels of
each glycan species separately, allowing FLD abundance levels to be corrected.

Co-authors:

Bastiaan Duivelshof, Institute of Pharmaceutical Sciences of Western Switzerland (ISPSO), University of Geneva, School
of Pharmaceutical Sciences, University of Geneva
Steffy Denorme, Research Institute for Chromatography (RIC)
Koen Sandra, Research Institute for Chromatography (RIC)
Xiaoxiao Liu, Waters Corporation
Alain Beck, IRPF - Centre d’Immunologie Pierre-Fabre (CIPF)

335
Matthew Lauber, Waters Corporation
Davy Guillarme, Institute of Pharmaceutical Sciences of Western Switzerland (ISPSO), University of Geneva, School of
Pharmaceutical Sciences, University of Geneva

336
MASS SPECTROMETRY-BASED DE NOVO SEQUENCING OF ANTIBODIES
USING MULTIPLE PROTEASES AND A DUAL FRAGMENTATION SCHEME
Abstract ID: 253

Presenting author: Weiwei Peng, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for
Biomolecular Research and Utrecht Institute of Pharmaceutical Sciences

Introduction

Antibodies have become invaluable research tools in the life sciences and ever more widely developed as therapeutic
agents. Sequence information is crucial for the use, production, and validation of these important research tools and
biopharmaceutical agents. Antibody sequences are typically obtained from DNA sequencing. This technology always
requires isolation of the antibody-producing cells from peripheral blood monocytes, or spleen and bone marrow tissues,
which is often not available. In this context, mass spectrometry (MS)-based sequencing of the secreted antibody
products tackles the challenge.

Methods

Here, we demonstrate a method for direct de novo sequencing of recombinant monoclonal IgG (mab) and hybridoma
mab with multiple light chains. The method uses a panel of multiple complementary proteases to generate suitable
peptides for de novo sequencing by liquid chromatography-tandem mass spectrometry (LC-MS/MS) in a bottom-up
fashion. Furthermore, we apply a dual fragmentation scheme, using both stepped high-energy collision dissociation
(stepped HCD) and electron-transfer high-energy collision dissociation (EThcD), on all peptide precursors.

Preliminary data (results)

The method achieves full sequence coverage of the monoclonal antibody herceptin, with an accuracy of 99% in the
variable regions. We applied the method to sequence the widely used anti-FLAG-M2 mouse mab, which we successfully
validated by remodeling a high-resolution crystal structure of the Fab and demonstrating binding to a FLAG-tagged target
protein in Western blot analysis (The results were recently published in Journal of Proteome Research 2021 20 (7),
3559-3566). Moreover, the sequences of an anti-muc1 antibody with multiple light chains have been obtained based on
this method and confirmed by western blot.

Please explain why your abstract is innovative for mass spectrometry?

This method not only offers robust and reliable sequencing of monoclonal antibodies but also more complicated samples,
which shows prospective applications for sequencing secreted antibodies from bodily fluids.

Co-authors:

Matti Pronker, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research and Utrecht
Institute of Pharmaceutical Sciences
Koen Giesbers, Department Biomolecular Health Sciences, Division Infectious diseases and Immunology, Faculty of
Veterinary Medicine, Utrecht University
Karin Strijbis, Department Biomolecular Health Sciences, Division Infectious diseases and Immunology, Faculty of
Veterinary Medicine, Utrecht University
Douwe Schulte, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research and Utrecht
Institute of Pharmaceutical Sciences
Joost Snijder, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research and Utrecht
Institute of Pharmaceutical Sciences

337
Mass spectrometry-based de novo sequence of the antibody herceptin

Mass spectrometry-based de novo sequence of anti-FLAG™-M2 antibody

338
AT-LINE MONITORING AND QUANTIFICATION OF MONOCLONAL
ANTIBODY PRODUCTS DURING BIOPROCESSES USING HPLC-MS
Abstract ID: 677

Presenting author: Katharina Böttinger, Paris Lodron University of Salzburg

Introduction

Despite making up only 3-5% of the total mass of a therapeutic monoclonal antibody (mAb), N-glycosylation strongly
affects the drug’s safety and efficacy. Even minor variations in manufacturing conditions during bioprocesses
considerably manipulate a mAb’s N-glycosylation profile. Thus, monitoring changes in the glycan profile in addition to
process parameter screening is critical during the bioprocess. (Regl, et al., 2019)

Mass spectrometry allows for quick analysis of mAbs at the intact protein level directly from fermentation broth. Besides
determination of the intact mass and detection of misassembled protein species, an overview of the mAb’s glycan
distribution is obtained. Bioinformatics tools facilitate automated N-glycan quantification at the deconvoluted and raw-
data level enabling a fast and reliable screening in dependence of process parameters.

Methods

Fermentation broth obtained from different bioprocesses was centrifuged and the supernatant was diluted to 50 ng
µL-1 final mAb concentration. mAb variants were separated on a polymeric reversed-phase column employing a 15 min
acetonitrile gradient. Mass spectrometry data was acquired in full scan mode within a mass range of m/z 1800-5,500 at a
resolution setting of 17,500. Quality control samples were injected at regular intervals in-between measurements. (Paul,
et al., 2020) In addition to quantification based on deconvoluted data (BioPharma Finder 4.0, Thermo Scientific™),
fractional abundances of N-glycans were determined using an in-house written R-package named fragquaxi.

Preliminary data (results)

First, HPLC-MS allowed for detection of three mAb species present in the fermentation broth eluting in the following
order: light chain (LC), light chain dimer (LC2) and an intact mAb, which was identified to be trastuzumab. Next, the intact
mAb was relatively and absolutely quantified on the basis of mass spectrometric and UV spectroscopic detection. mAb
concentrations increased from 29.9 to 710 ng µL-1 during the course of fermentation. Likewise, LC and LC2 slightly
increased with time.

Second, minor and major glycosylation variants were assigned to deconvoluted raw data. Five major glycovariants with
increasing galactosylation and one afucosylation species were identified. Minor variants included doubly afucosylated
and fully galactosylated glycoproteins. Subsequently, changes in glycosylation patterns during the time course of
fermentation were relatively quantified based on deconvoluted and raw data. An increase in galactosylation over time
was observed and both quantification approaches delivered similar results. However, quantification based on raw data
gave more precise results.

Third, the workflow was evaluated using quality control samples. Assessed parameters were retention time, peak area,
height and width at 50% intensity at the level of HPLC-UV as well as extraction ion current chromatograms (EICCs).
Additionally, quality and long-term stability of mass spectrometric detection was evaluated.

In conclusion, this workflow provides a useful tool for quick and reliable N-glycan profiling of mAbs directly from
fermentation broth employing a fast data evaluation based on an in-house written R-package.

Please explain why your abstract is innovative for mass spectrometry?

Usage of an in-house written R-package to relatively quantify N-glycans at the intact protein level from raw data without
the need for prior deconvolution.

Co-authors:

Wolfgang Esser-Skala, Paris Lodron University of Salzburg

Marius Segl, Paris Lodron University of Salzburg
Christoph Herwig, Technische Universität Wien
Christian G. Huber, Paris Lodron University of Salzburg

339
STANDARDIZED GRAPHITIC CARBON CHROMATOGRAPHY
HYPHENATED TO TANDEM MS/MS USED FOR ISOMER SPECIFIC N-
GLYCOMICS
Abstract ID: 109

Presenting author: Johannes Helm, 1 Department of Chemistry, Institute of Biochemistry, Universität für
Bodenkultur Wien, Muthgasse 11, A-1190 Vienna, Austria

Introduction

The importance of protein glycosylation in the biomedical field requires methods that not only quantitate structures by
their monosaccharide composition, but also resolve and identify the many isomers expressed by mammalian cells.

The art of unambiguous identification of isomeric structures in complex mixtures, however, did not yet catch up with the
fast pace of advance of high-throughput glycomics. Tandem MS, operated in positive or negative mode, is often used,
although, due to inherent limitations, this method alone is only partially suitable to differentiate between the omnipresent
glycan isomers (Figure 1). The full potential of tandem MS for the discrimination of glycan isomers can be exploited by
coupling it with an LC-system equipped with a shape-selective stationary phase like porous graphitic carbon (PGC).

Methods

Retention on PGC combined with diagnostic fragment ions derived from negative mode MS/MS leads to unambiguous
identification of even closely related glycan isomers. As the retention time on PGC is very unstable between runs, we
normalized the acquired retention times with a tightly spaced grid of isotope-labeled N-glycans used as internal standard.
Furthermore, we generated a library of normalized retention times based on isotope labelled N-glycans biosynthesized
with recombinantly expressed glycosyltransferases and glycosidases, which is already equipped with around 130, partly
isomeric, structures. Furthermore, MS/MS data generated from these structures will be

Preliminary data (results)

Application of this method to the human brain N-glycome led to the definitive and isomer-specific identification of most
occuring structures. Furthermore, we discovered glycan structures, which have not been described in the human brain
before. We identified hybrid-type glycans with galactosylated and even Lewis X containing bisected N-
acetylglucosamine, which actually have not yet been discovered in a natural source (Figure 2). We also identified
structures with fucose and N-acetylgalactosamine on the same arm, the so-called LDNF epitope often associated with
parasitic worms.

Negative and positive mode MS/MS data generated from the biosynthetically generated N-glycan library will be used to
gain further insights into the fragmentation of these molecules and to identify possible new diagnostic fragment ions for
the identification of N-glycans.

Please explain why your abstract is innovative for mass spectrometry?

• Combination of MS/MS data with normalized retention on PGC to identify closely related isomeric N-glycans
• Biosynthesized N-glycan library for the generation of reference MS/MS data

Co-authors:

Friedrich Altmann, 1 Department of Chemistry, Institute of Biochemistry, Universität für Bodenkultur Wien, Muthgasse 11,
A-1190 Vienna, Austria

340
PGC-ESI-MS chromatograms of isomeric brain N-glycans (m/z 895.34)

Negative mode MS/MS spectra of selected brain N-glycans

341
STRUCTURAL CHARACTERIZATION OF ANTIBODY-DRUG CONJUGATES
USING HYDROGEN/DEUTERIUM EXCHANGE AND LIMITED
PROTEOLYSIS-MASS SPECTROMETRY
Abstract ID: 492

Presenting author: Liana Tsiatsiani, Byondis B.V.

Introduction

The efficacy of biopharmaceuticals, such as monoclonal antibody-drug conjugates (ADC), is strongly dependent on
maintaining structural integrity throughout industrial process development, and this is traditionally monitored using
spectroscopic techniques. Such methods monitor important aspects of protein structure, but offer limited resolution
regarding the exact sites of structural change. At the same time, mass spectrometry (MS)-based methods are gaining
ground in the pharmaceutical industry because they allow peptides indicative of structural integrity to be detected with
shotgun or bottom-up proteomics workflows in a single chromatographic run, and thus localize structural changes with
increased accuracy. In this work, we set out to evaluate the complementarity of limited proteolysis-MS (LiP) and
hydrogen-deuterium exchange-MS (HDX) for the analysis of protein higher-order structure (HOS) in a biopharmaceutical
setup.

Methods

First, we compared different randomly-conjugated ADC species with a drug-to-antibody ratio (DAR) of 0, 2 and 4 using
Intrinsic Tryptophan Fluorescence spectroscopy (ITF). Then, the different DAR species were also examined with MS-
based assays, such as LiP-MS and HDX-MS. First, LiP-MS was performed using two different proteolytic enzymes and
label-free peptide quantification. To confirm these observations and further increase the resolution of conformational
changes in the different DAR species, we monitored the hydrogen-deuterium exchange rates under native conditions
using HDX-MS. Finally, both the LiP and HDX-MS data sets were compared and shared regions of conformational
change were identified.

Preliminary data (results)

ITF analysis of DAR0, DAR2 and DAR4 species showed that the F350/F330 and λmax parameters gradually increased
upon linker drug (LD) conjugation due to higher solvent exposure of tryptophan upon conformational changes. LiP-MS
was used to infer structural changes from peptide abundances as a result of higher protease accessibility to the protein
structure. The analysis revealed protein regions that are more amenable to enzymatic digestion than others, due to the
increasing drug load. Even though, the overall extent of protein unfolding in DAR4 was low, the most affected domains
included predominantly the CL and CH1, which are located near the LD conjugation sites, and to a lesser extent the CH2
domain in Fc region. No significant changes were observed in the complementarity-determining regions that are involved
in target binding. To confirm the LiP-MS observations, each DAR species was analyzed with MS after several time
intervals of HDX. Again, the differences in solvent exposure between the different DAR species were small, but the
acquired spatiotemporal deuterium uptake allowed us to locate precisely the affected sites in CL, CH1 and CH2 domains
with near amino acid resolution. Both LiP and HDX-MS confirmed that the same protein regions became more accessible
to the proteases, or the solvent, depending on the drug load. In this way, we can conclude that both LiP-MS and HDX-
MS can detect small conformational changes due to random LD conjugation onto the protein sequence, but HDX offers
higher resolution, while LiP-MS is faster and easier to implement.

Please explain why your abstract is innovative for mass spectrometry?

LiP-MS and HDX-MS were used orthogonaly for the structural characterization of biopharmaceuticals. MS offers higher
resolution than conventional spectroscopic methods, and allows for monitoring of structural changes and their impact.

Co-authors:

Yunus Saricay, Byondis B.V.

Lang Xin Zhang, Byondis B.V.
Eef H.C. Dirksen, Byondis B.V.

342
Analysis of antibody-drug conjugates with mass spectrometry and biophysics

343
 Thursday 1 September 2022: 15:30 – 17:30
Session: AD-4 Homeland Security, Explosives and Environmental
Monitoring

DETERMINATION OF PER- AND POLYFLUOROALKYL SUBSTANCES

(PFAS) AND POLYFLUOROALKYL PHOSPHATE ESTERS (PAPS) IN FOOD
PACKAGING MATERIAL BY LC-MS
Abstract ID: 701

Presenting author: Jacob de Boer, Amsterdam Institute for Life and Environment, Vrije Universiteit

Introduction

Per- and polyfluroalkyl substances (PFAS) are highly fluorinated compounds. They are fat, water and dirt-repellent.
Therefore, they have many useful applications. However, due to their extreme persistency, they have been detected in
the environment and humans, and have been associated with different diseases. Therefore, PFAS are considered
environmental pollutants of high concern. The polyfluoroalkyl phosphate esters (PAPs) have been identified as potential
precursors of PFAS. Some studies have evidenced that they are more toxic than PFAS themselves. PAPs are mainly
used in food packaging materials. They have also been detected in sewage sludge and human serum.

We have developed an LC/MS method for the determination of several PFAS, including PAPs, in food-contact materials.

Methods

Instrument: Bruker, ELUTE LC with a triple-quad Elite EVOQ MS, neg. ESI. Source parameters: spray voltage 4500V;
cone temperature 350˚C; cone gas flow 25 L/h; heated probe 375˚C; probe gas flow 45 L/h; nebulizer 55 L/h. Columns:
XBridge BEH C18 (150x2.1 mm, 2.5 mm), pre-column: XBridge BEH C18 (50x2.1 mm, 2.5 mm). Mobile phase: Milli-Q
water (A) and methanol (B) both containing 0.1% NH4OH, flow rate 0.3 ml/min. Gradient: 0.5 min 5% B, increasing to
50% (1.5 min) and to 99% (8 min), maintained 8 min, decreasing to initial conditions, 4 min.

Preliminary data (results)

NH4OH was added to the mobile phase to increase the pH until 9, to improve the mono-PAPs peak shape. PAPs are
very polar and have strong interactions with the polar part of the stationary phase. In this way, the phosphate groups are
ionized and polar interactions are avoided. Figure 1A shows the differences in the TIC chromatograms with NH 4OH in
the mobile phase between 0.05% and 0.1%. The best resolution was obtained with 0.1% NH 4OH, due to the higher
dissociation of the phosphate group. Various flow rates were tested (0.2, 0.3 and 0.35 ml/min). Those TIC PAP
chromatograms are shown in figure 1B. Due to the higher signal, a flow of 0.3 ml/min was selected above 0.35 ml/min,
while a better peak resolution was obtained compared to a flow of 0.2 ml/min. Due to the low sensitivity of PAPs, pre-
concentration will always be necessary. The SPE step that is already included in the method allows that. Figure 2 shows
PAP chromatograms (A: TIC; B: 8:2 diPAP; C: 10:2 diPAP) with the PAPs in two different concentrations (1 and 0.01
mg/ml). In the case of 8:2 diPAP (2B) and 10:2 diPAP (2C) the signal is completely absent at low concentrations, which
complicates their detection in real samples where concentrations might be very low, down to 0.1 ng/g for di-PAPs.
However, we plan to use a Sciex LC/MS for the final analyses (Exion 6500+ QTRAP LC/MS, ion spray 4000V).

Please explain why your abstract is innovative for mass spectrometry?

Application of Sciex QTRAP LC/MS

Co-authors:

Maria Jesús Dueñas Mas, Department of Analytical Chemistry, Institute of Fine Chemistry and Nanochemistry, Campus
of Rabanales, University of Córdoba

344
TIC chromatograms at different NH4OH concentrations (A) and flows (B)

TIC Chromatograms (A), 8:2diPAP (B) and 10:2diPAP (C)

345
TOXICITY PREDICTIONS OF UNIDENTIFIED CHEMICALS IN WATER BY
NONTARGET LC/HRMS
Abstract ID: 129

Presenting author: Pilleriin Peets, Stockholm University

Introduction

For evaluating the safety of different water samples toxicity and quantity of different chemicals presents must be
analysed. While liquid chromatography with high resolution mass spectrometry (LC/HRMS) is excellent tool for detecting
chemicals in nontarget analysis, previous experiences show that only small fraction of chemicals are fully identified. Here
we present a novel methodology to predict lethal concentration for 50% of the population (LC50) of individual chemicals
directly from HRMS data without identification step using structural fingerprints calculated with SIRIUS+CSI:FingerID
software. For best model 75% of the predicted LC50 differed less than one order magnitude from experimental values
and in 98% of the cases difference was less than two orders of magnitude.

Methods

LC50 for three fishes (rainbow trout, bluegill, fathead minnow) were retrieved from CompTox for ~800 compounds. The
molecular fingerprintswere obtained in two different ways. For training and testing the fingerprints were calculated
structures’ SMILES with R package rcdk, while for validating and final application SIRIUS+CSI:FingerID software was
used for calculating fingerprints from MS1 and MS2 data. MS2 data for validation was gathered from MassBank (EAWAG,
Athens University and University of Luxembourg). Models were evaluated by analysing variance importance in model
and root mean square error and squared correlation coefficientbetween predicted and experimentally measured toxicity
values.

Preliminary data (results)

LC/HRMS spectra carry information about the polarity, size, and functional groups present in the detected compounds,
which also affect the toxicity of the compounds. Unfortunately, the number of compounds for which toxicity values and
LC/HRMS spectra are simultaneously available is insufficient for training a toxicity predictive machine learning model
directly from LC/HMRS spectra. Therefore, the model was trained from LC 50 values from CompTox and from theoretical
fingerprints calculated from SMILES representation of the compound. Compounds with available toxicity value and
MS2 spectra in MassBank were kept for validation set.

LC50 values for rainbow trout, bluegill, fathead minnow were combined into common fathead minnow scale using general
additive model (GAM) to increase the data amount. All calculations were done in logarithmic scale of mmol/L and the
range for LC50 varied from -6.38 to 2.94. Different machine learning algorithms like gradient boosting, random forest and
support vector machine were used. The best results were gained with extreme gradient boosting Dropouts Additive
Regression Trees (xgbDART). Root mean square error for validatation was 0.88 log(mmol/L), while standard deviation
for experimental toxicity values was 0.45, meaning that the predicted toxicity error is comparable to eperimentral
one. This indicates that empirical structural information relevant for toxicity predictions can be successfully extracted from
the MS2 spectra.

As an outcome of this work, fish LC50 prediction tool was created that was also succesfully tested for analysing in-house
measured spiked solutions measured with LC-Orbitrap system.

Please explain why your abstract is innovative for mass spectrometry?

An accurate approach to estimate LC50 values for compounds detected with nontarget LC/HRMS analysis.

Co-authors:

Wei-Chieh Wang, Stockholm University

Matthew MacLeod, Stockholm University
Magnus Breitholtz, Stockholm University
Jonathan Martin, Stockholm University
Anneli Kruve, Stockholm University

346
HRMS data is converted into fingerprints to predict LC50 values

347
TWO-DIMENSIONAL ULTRAVIOLET PHOTODISSOCIATION MASS
SPECTROMETRY FOR THE DATA-INDEPENDENT ANALYSIS OF SINGLY-
CHARGED AGROCHEMICALS AND THEIR METABOLITES IN
ENVIRONMENTALLY RELATED MATRICES
Abstract ID: 604

Presenting author: Bryan Marzullo, University of Warwick

Introduction

Structural characterization of complex biological mixtures may prove difficult, requiring high-resolution analytical
techniques. Ultraviolet photodissociation (UVPD) has been shown to produce extensive structurally informative data for a
variety of chemically diverse compounds. Coupling two-dimensional mass spectrometry (2DMS), a data-independent
analysis (DIA) technique, and UVPD provides many advantages for the structural characterisation of mixtures. Recent
developments in data extraction methods can reduce spectral complexity and increase the applicability of this DIA
technique.

Methods

The samples analysed were composed of an agrochemical (parent compound) and some of their related metabolites
within environmentally relevant matrices (soil and tomato extract). The compounds present in these samples were
simultaneously analysed via two-dimensional mass spectrometry (2DMS) on a 12 T FT-ICR MS using ultraviolet
photodissociation (UVPD) as the fragmentation technique. 2DMS allows all fragments to be correlated to their
corresponding precursors, without the need for prior chromatography or precursor isolation. A 3D peak picking was
utilized to extract spectral information providing sub-ppm assignment errors.

Preliminary data (results)

The results obtained from this study demonstrate that this data-independent analytical technique can be applied to a
complex sample composed of an agrochemical parent species and its related metabolites within environmentally relevant
matrices, resulting in characterisation of many related species simultaneously. 2D-UVPD-MS analysis was performed in
both positive and negative polarity to obtain structural informative data on all compounds of interest present in the
sample. Most species of interest displayed extensive fragmentation, readily resulting in the characterisation of their
structure. The precursor (vertical) scan lines, indicating common m/z values allow for the identification of the chemical
classes – grouping a range of related species by their intact functional groups. The 3D peak picking method has many
advantages for analysing 2DMS data, overcoming the limitations of simply extracting lines and successfully
distinguishing fragments from two species by >5 sigma. This specificity is applicable throughout the broadband 2DMS.
Coupling with internal calibration results in sub-ppm assignment error values. UVPD combined with 2DMS is shown to be
a viable and powerful technique for small molecule and metabolite analysis; allowing this technique to be applied to the
analysis of a wide range of small molecule types e.g. pharmaceutical industries and forensics, especially since
agrochemicals have similar characteristics to chemical warfare agents.

Please explain why your abstract is innovative for mass spectrometry?

High resolution data-independent 2D-UVPD-MS analysis allowing swift correlation of compounds in dual polarity.

Co-authors:

Alina Theisen, Bruker Daltonics GmbH

Anisha Haris, University of Warwick
Christopher Wootton, Bruker Daltonics GmbH
Simon Perry, Product Metabolism & Analytical Sciences, Jealott’s Hill International Research Centre, Syngenta
Mansoor Saeed, Product Metabolism & Analytical Sciences, Jealott’s Hill International Research Centre, Syngenta
Mark Barrow, University of Warwick
Peter O'Connor, University of Warwick

348
Negative mode 2D-UVPD-MS and auto-correlation line displaying the different compounds.

349
Extracted precursor scan lines displaying common moieties between different compounds.

350
DOES SEA-DUMPED CHEMICAL WEAPONS POSE A RISK TO MARINE
ECOSYSTEM?
Abstract ID: 621

Presenting author: Paula Vanninen, VERIFIN, University of Helsinki

Introduction

During and after the World War II, totally 40,000 tons and 170,000 tons of chemical weapons (CWs) as aircraft bombs,
artillery shells, high-explosive bombs, containers and drums have been dumped in the Baltic Sea and Skagerrak strait,
respectively. The estimated amount of dumped-CWs into seas and oceans all around the world is as high as 1 million
tons. This presentation focusses on risk assessment based on marine sediment and biota studies in the Baltic Sea and
Skagerrak strait performed during the EU-funded projects (CHEMSEA, DAIMON). Mustard gas and phenylarsenic
chemicals (e.g. Clark I and II, Adamsite) have been dumped in those studied areas. This presentation gives an overview
on the status of chemical studies on phenylarsenic chemical warfare agents (CWAs).

Methods

The sample preparation of sediment samples included determination of dry weight, fractionation for gas
chromatographic-tandem mass spectrometric (GC-MS/MS) and liquid chromatographic-tandem mass spectrometric (LC-
MS/MS) analysis based on volatility of studied chemicals. For both techniques derivatization and multiple reaction
monitoring were applied. Biota samples e.g. fish muscle and liver tissues were homogenized, extracted, purified using
solid phase extraction and oxidized prior the LC-MS/MS analysis for the detection of CWAs. Liquid chromatography-high
resolution mass spectrometry (LC-HRMS) was used to confirm presence of newly detected degradation products of
phenylarsenic chemicals in the sediment and their metabolites in marine biota samples.

Preliminary data (results)

More than thousand sediment samples have been analysed using validated methods for selected CWAs and their
degradation products. In the marine environment, Clark I and II form same degradation products, first being hydrolyzed
and then oxidized further to diphenylarsinic acid (DPA). In sediment samples the detected concentrations of DPA have
been up to 1000 mg/kg dw. Some previously unknown chemicals were also found in marine sediments using non-
targeted screening and identified using LC-HRMS. These chemicals most likely have a significant role in the total burden
caused by CWAs in marine sediments.

A total of ca. 300 different marine biota samples were analysed from three known dumping areas, showing that
contaminated marine biota species with CWAs could be found in all those studied dumping areas. The results contribute
to the risk assessment of dumped-CWs by increasing knowledge on sediment concentrations, bioaccumulation in marine
biota, biotransformation and toxicity of sea-dumped phenylarsenic chemicals. Based on the in vitro studies, reactive
metabolites are produced as a result of metabolism of CWA-related phenylarsenic chemicals which might lead to
enhanced toxicity of these chemicals.

Please explain why your abstract is innovative for mass spectrometry?

Non-targeted screening and identification using LC-HRMS of new target phenylarsenic chemicals from sediment and
marine biota sample analysis could be successfully performed.

Co-authors:

Hanna Niemikoski, SYKE

351
AMBIENT IONIZATION TECHNIQUES FOR THE HIGH-THROUGHPUT AND
LOW-COST SCREENING AND CHARACTERIZATION OF PFAS.
Abstract ID: 390

Presenting author: Patrick Fedick, Naval Air Warfare Center Weapons Division

Introduction

Per-and-polyfluoroalkyl substances (PFAS) is an emerging contaminant class that is a multi-billion-dollar problem. These
environmentally hazardous PFAS were widely used in fire retardants, aqueous film forming foams (AFFF), and water-
resistant textiles, making it necessary to monitor PFAS levels in water sources and soil. AFFF has been utilized heavily
by the military since the 1970s for fire suppression, to extinguish liquid fuel fires on ships, shore fixed systems, and in
aircraft hangars. The widespread use of these materials, coupled with their recalcitrant nature, has resulted in
contamination of water sources and soil at a number of DoD sites. Even though AFFF is a known contaminant, there are
currently no approved alternatives, and is still being used due to the necessity of protecting the Warfighter and assets.

Methods

A new screening methodology, 3D-printed cone spray ionization mass spectrometry (3D-PCSI-MS), has been developed
at NAWCWD for the rapid and high-throughput screening of PFAS in bulk matrices (soil, sediment, concrete). This
methodology takes low-cost 3D-printed conductive cones to act as the sampling and ionization device. Concurrently, a
methodology for screening thermal breakdown products of PFAS at various temperatures has been developed through
pyrolysis-DART. By understanding the thermal breakdown products of PFAS, optimal remediation strategies can be
developed and improved screening lists of PFAS can be established that are needed to monitor after an AFFF release
event.

Preliminary data (results)

Rugged, conductive 3D-PCSI sources constructed using a 3D-printer and ESD-Safe PETG filament doped with carbon
nanotubes, was used to develop the 3D-printed cone ionization sources. The added rigidity of the conductive plastic aids
in the reproducibility of manufacturing, making it possible to develop an autosampler.
Once the cone was printed, solid material (contaminated soil, sediment, and concrete) was placed within the cavity of the
cone and 1 mL of solvent and high voltage was applied. PFAS was rapidly identified through database screening
based on their characteristic mass spectra and mass transitions. Our group has recently reported limits of detection
(LOD) for various PFAS analyzed by 3D-PCSI-MS, in a variety of soil types, ranging from 100 parts per trillion (ppt) to 1
part per billion (ppb) which are the relevant concentrations for soil analysis.

Concurrently, to examine the broad variety of PFAS needed to be screened, we employed thermal desorption/pyrolysis
direct analysis in real time- mass spectrometry (TD/pyro-DART/MS) to characterize products of incomplete combustion
(PIC) and determine extent of thermal degradation of PFAS in standards and AFFF concentrates ranging from 25 – 600
°C. Several PFAS standards, ranging from 4 – 14 carbon atoms were monitored in situ followed by representative legacy
and current AFFF mixtures. Head group scission were observed, followed by sequential carbon-carbon cleavage in the
structures resulting in fragments differing by -CF2 and -C2F4 units. In addition, large molecular weight PFAS compounds
resulted in more detectable pyrolytic fragments than low molecular weight counterparts.

Please explain why your abstract is innovative for mass spectrometry?

Typically, LC-MS is the standard for environmental screening for PFAS analysis. Presented here are two ambient
ionization techniques for high-throughput and low-cost analysis, with relevant performance metrics.

Co-authors:

Hilary Brown, Naval Air Warfare Center Weapons Division

Brian Molnar, Naval Air Warfare Center Weapons Division
Christopher West, Naval Air Warfare Center Weapons Division, Purdue University

352
3D-PCSI manufactured on a benchtop 3D-printer.

The ionRocket pyrolysis-DART mass spectrometry platform used for this study.

353
DEVELOPMENT OF A LC-ESI-TIMS-QTOFMS DATABASE TO ENHANCE
THE PERFORMANCE OF WIDE-SCOPE TARGET SCREENING IN
ENVIRONMENTAL (BIO)MONITORING STUDIES
Abstract ID: 737

Presenting author: Konstantina S. Diamanti, Laboratory of Analytical Chemistry, Department of Chemistry,

National and Kapodistrian University of Athens, Panepistimiopolis Zografou, 15771, Athens, Greece

Introduction

Thousands of chemicals are estimated to be dispersed throughout the environment and may pose a threat to
ecosystems and human health. Exposure to chemicals can be assessed by the monitoring of environmental
contaminants in the ambient environment (air, water, soil), in organisms from various trophic levels, and in human
biological samples (blood, urine). The increased availability of high-resolution mass spectrometric (HRMS) techniques in
the chemical analysis allows the wide-scope screening of complex samples. The hyphenation of ion mobility
spectrometry (IMS) with HRMS has come to add a new dimension for compound separation and identification;
namely collision cross section (CCS). This additional dimension enhances the identification confidence in target
screening. To-date, very few studies are available in the literature providing integrated LC-IMS-HRMS databases of
organic contaminants.

Methods

Here, a database containing more than 1,000 environmental contaminants (pharmaceuticals, illicit drugs, pesticides,
industrial chemicals, etc.) was built by utilizing liquid chromatography coupled to electrospray ionization—quadrupole-
time-of-flight mass spectrometry equipped with trapped ion mobility spectrometry (LC-ESI-TIMS-TOFMS).

Preliminary data (results)

The overall aim of this study was the establishment of a CCS-aware database for wide-score target screening. For the
database development, a quality control protocol was followed to ensure high-quality CCS measurements based on the
guidelines for the Unified CCS Compendium. Accuracy was assessed by analyzing a standard solution with reference
CCS values before, during and after the analyses, and the individual CCS error was found to be below 1%. To assess
CCS values reproducibility, all solutions were analyzed in triplicate resulting in ΔCCS <0.7% being achieved in most
cases. Therefore, beyond molecular formula (accurate mass, isotopic pattern), retention time, and qualifier ions (adducts,
fragments etc.) information for each compound, the developed database was enriched with ion mobility derived CCS
values.To the best of our knowledge, this is the first study reporting such an extensive dataset incorporating CCS values
for a broad range of environmentally relevant compounds. Wide-scope target screening with the developed database
was applied in different samples. Selected examples unveiling the performance of adding another dimension for analyte
separation and sample characterization are presented.

Environmental chemistry and exposure assessment: analysis, monitoring, fate and modeling

Please explain why your abstract is innovative for mass spectrometry?

To the best of our knowledge, this is the first study reporting such an extensive dataset incorporating CCS values for a
broad range of environmentally relevant compounds.

Co-authors:

Bob Galvin, Bruker Daltonik GmbH, Fahrenheitstrasse 4, 28359 Bremen, Germany

Dimitrios E. Damalas, Laboratory of Analytical Chemistry, Department of Chemistry, National and Kapodistrian University
of Athens, Panepistimiopolis Zografou, 15771, Athens, Greece
Georgios O. Gkotsis, Laboratory of Analytical Chemistry, Department of Chemistry, National and Kapodistrian University
of Athens, Panepistimiopolis Zografou, 15771, Athens, Greece
Eleni I. Panagopoulou, Laboratory of Analytical Chemistry, Department of Chemistry, National and Kapodistrian
University of Athens, Panepistimiopolis Zografou, 15771, Athens, Greece
Maria-Christina Nika, Laboratory of Analytical Chemistry, Department of Chemistry, National and Kapodistrian University
of Athens, Panepistimiopolis Zografou, 15771, Athens, Greece
Nikolaos S. Thomaidis, Laboratory of Analytical Chemistry, Department of Chemistry, National and Kapodistrian
University of Athens, Panepistimiopolis Zografou, 15771, Athens, Greece
Carsten Baessmann, Bruker Daltonik GmbH, Fahrenheitstrasse 4, 28359 Bremen, Germany

354
Session: IM-13 Beyond Mass Spectrometry: Making MS obsolete….

THE EMERGING LANDSCAPE OF SINGLE-MOLECULE PROTEIN

SEQUENCING TECHNOLOGIES
Abstract ID: 773

Presenting author: Chirlmin Joo, Delft University of Technology

Introduction

Single-cell profiling methods have had a profound impact on our understanding of cellular heterogeneity. While genomes
and transcriptomes can be explored at the single-cell level, single-cell profiling of proteomes is not yet established.

Methods

I will give an overview on new single-molecule protein identification technologies that will eventually enable broad
sequence coverage in single-cell profiling. These technologies will in turn facilitate biological discovery and open new
avenues for ultra-sensitive diseases diagnostics.

355
NOVEL NANO-ELECTRO-MECHANICAL RESONATOR MS SYSTEM
DESIGN
Abstract ID: 354

Presenting author: Christophe Masselon, CEA IRIG

Introduction

Nano-electro-mechanical sensors (NEMS) are silicon nanostructures vibrating at their resonance frequencies through
electrical actuation. These devices are sensitive to the mass deposited onto their active surface. A particle landing onto
the NEMS results in a sharp drop in resonance frequencies in proportion to the added particle’s mass.

NEMS-based MS offers unique capabilities for the analysis of ultra-high mass analytes in the MDa to GDa mass range
such as whole viruses or nanoparticles, independently of their charge state. Acting as both an analyzer and a detector,
since they both sense the particle and determine the mass directly, NEMS devices open unprecedented capabilities for
MS instrumentation.

Methods

A nano-resonator array was mounted as detectors into a novel home-made NEMS-based MS prototype. Briefly, the
system consists in an atmospheric pressure ESI nebulization source coupled to three successive differentially pumped
sections, without moving parts or electromagnetic fields for ion guiding. Nebulized analyte particles were aspirated into
the vacuum system through an 11-cm-long, 1/100-in. i.d. heated stainless steel capillary tubing (Sigma-Aldrich, St. Louis,
USA); the incoming particle beam was then collimated by an aerodynamic lens onto an array of 20 frequency addressed
NEMS sensors operating in a high vacuum environment.

Preliminary data (results)

Taking in the lessons learned from a prior system design, we set out to develop a new generation NEMS-MS system.
Whereas the original design was cumbersome, incorporating a cryostat and substantial pumping requirements, we
strived for a more compact, transportable design, relieved of unwarranted ancillary features. The new system features a
compact modular design with three interchangeable chambers to facilitate further developments, and a single turbo
pump. The aerodynamic lens is now aligned with the skimmer by design, and the array of NEMS detectors mounted on
x/y micromanipulators for ease of alignment. Perhaps the most striking difference between the two systems is the
significantly reduced pumping requirements (1e-2 Torr vs. 1e-5 Torr). Using data from the first prototype, we found
through numerical simulations and validated experimentally that the aerodynamic lens performance and the resonator’s
reponse were only moderately affected by this change of pressure. Relaxing pumping requirements opens new
opportunities for further system miniaturization.

Please explain why your abstract is innovative for mass spectrometry?

A novel NEMS-MS system operating at moderate vacuum.

Co-authors:

Adrien Reynaud, CEA LETI

Louis Dartiguelongue, CEA IRIG
Vaitson Çumaku, University Grenoble Alpes, CEA IRIG
Thomas Fortin, CEA IRIG
Marc Sansa-Perna, CEA LETI
Sébastien Hentz, CEA LETI

356
Novel NEMS-MS system design

357
SINGLE PARTICLE MASS MEASUREMENTS FOR WEIGHING VIRAL GENE
DELIVERY PARTICLES AND MONITORING SARS-COV-2 / ANTIBODY
INTERACTIONS
Abstract ID: 142

Presenting author: Victor Yin, Universiteit Utrecht

Introduction

Native mass spectrometry provides information on protein assemblies that can be difficult to ascertain using other
experimental approaches. However, native MS typically struggles in cases where high levels of micro-heterogeneity (e.g.
extensive glycosylation, multiple isoforms) hampers charge state assignment. Mass photometry (MP) and Orbitrap-based
charge-detection MS (CD-MS) are two recently developed single particle mass analysis techniques that both solve the
charge assignment problem, and enable the successful mass measurement of challenging, heterogeneous assemblies.
Here, we demonstrate and compare the unique capabilities of these two methods in studying two challenging, high-
interest, case studies: (1) genome loading in adeno-associated viruses (AAVs), and (2) binding of different antibody
formats to the SARS-CoV-2 spike (S) protein trimer [1].

[1] Yin et al., ACS Central Sci 2021.

Methods

Conventional native MS relies on ESI charge state inference to determine mass. In contrast, CD-MS directly detects the
charge of individual ions in the gas phase, while MP utilizes photon scattering to determine the mass of individual
particles in solution, thereby circumventing ESI entirely. MP measurements were performed on a Refeyn OneMP
(Refeyn Ltd., Oxford) following sample dilution into Tris or PBS buffer without further purification. CD-MS measurements
were performed as previously described (Worner et al., Nat. Meth. 2020) on an Orbitrap Q-Exactive UHMR mass
spectrometer (Thermo Scientific Ltd., Bremen), using samples first buffer exchanged into aqueous AmAc.

Preliminary data (results)

We focused on two systems in which mass determination is untenable using conventional native MS approaches.

Case1: AAVs are promising vectors for gene therapy applications, but are challenging to characterize due to their large
size (~3.8 MDa) and micro-heterogeneity arising from the three contributing capsid proteins (VP1, VP2, VP3). This
heterogeneity is exacerbated upon genome loading. Despite this complexity, we demonstrate that single particle
measurements allow facile determination of the AAV masses, and can accurately measure and quantify the
efficiency/integrity of genome loading via direct detection of the +-genome-loaded AAV mass abundance.

Case2: The extensive glycosylation of the SARS-CoV-2 S trimer (~66 glycosites) and other viral spike proteins have
hampered conventional native MS analysis. By contrast, both MP and CD-MS successfully determine the mass of the
intact S trimer, with excellent agreement between the two. Using these methods, we next accessed the binding
stoichiometries of a panel of anti-SARS-CoV-2 antibodies (as full IgGs) against the S trimer. Surprisingly, we observed a
diversity of stoichiometries, with no IgG displaying the “complete” 3:1 stoichiometry predicted from the 3-fold S trimer
symmetry. Equivalent experiments using truncated antibodies (Fabs) exhibited binding patterns distinct from the IgGs
and suggest that this lower-than-expected stoichiometry in the full antibodies arises from the interplay of steric clashes
and avidity effects that are absent in Fabs. These divergent results suggest that truncated antibody constructs commonly
employed in structural biology may paint an incomplete picture of the multimeric antibody/S trimer interactions.

Please explain why your abstract is innovative for mass spectrometry?

Single particle mass measurement methods allow the study of heterogeneous, high-interest assemblies beyond the
range of conventional native MS approaches.

Co-authors:

Paul Devine, AstraZeneca

Nicholas Bond, AstraZeneca
Szu-Hsueh Lai, Universiteit Utrecht
Tom Caniels, Amsterdam University Medical Centers
Philip Brouwer, Amsterdam University Medical Centers

358
Mitch Brinkkemper, Amsterdam University Medical Centers
Yoann Aldon, Amsterdam University Medical Centers
Hejun Liu, Scripps Research Institute
Meng Yuan, Scripps Research Institute
Ian Wilson, Scripps Research Institute
Rogier Sanders, Amsterdam University Medical Centers
Marit van Gils, Amsterdam University Medical Centers
Albert Heck, Universiteit Utrecht

359
EXTERNAL INJECTION OF TRAPPED IONS INTO A HEMISPHERICAL
ELECTROSTATIC SECTOR ANALYZER FOR CHARGE DETECTION MASS
SPECTROMETRY – A SIMULATION STUDY
Abstract ID: 227

Presenting author: John Hoyes, TrueMass

Introduction

Electrospray of very large molecules yields a continuum of possible charge states meaning that conventional mass
analysers that only detect m/z ratios fail to give distinct spectral peaks providing little useful analytical data. CDMS
overcomes this limitation by injecting individual ions and measuring both m/z and z allowing true mass measurements.
Our novel analyzer consists of two sets of 199.2o spherical electric sectors placed opposite to each other such that ions
describe a figure of eight trajectory. A weak focussing lens is placed at the plane of symmetry to prevent excessive
arcuate excursions of the ions, along with a pair of optimally positioned charge tubes for inductive ion detection.We
examine the phase space acceptance of this analyzer from our bespoke Electrospray interface.

Methods

Two SIMON models were constructed. Model 1 is a medium resolution (200μm/gu) array of the complete instrument
comprising a segmented RF hexapole ion trap, two RF transfer hexapole units, a focusing lens and the mass analyzer.
Model 2 is a high resolution (50μm/gu) stand-alone array of the analyzer. These two potential arrays are imported into
our inhouse software for tracing ions, and to produce phase space plots at chosen test planes. We evaluated ion
transmission, collimation and acceptance using Model 1, while Model 2 provided a more detailed evaluation of ion
stability and tuning of the analyzer.

Preliminary data (results)

Using Model 1 a group of Myoglobin ions of 16,941 Da each carrying 20 charges are thermalized in the RF hexapole trap
at a nitrogen pressure of 1.2x10-2 mbar and ejected into the downstream optics by pulsing the potential applied to a
differential aperture below the DC level of the RF hexapole trap. Results indicate that the energy spread of the ions at the
entrance of the analyzer is established during ejection from the high-pressure trap and can be kept below 2 eV
thereafter. We also examined the characteristics of a three-stage accelerating immersion lens at an energy of 300eV/Th.
The voltages applied to this lens along with those applied to the sectors were varied to probe the sensitivity and stability
of the system during the transient. Model 2 is used to map the phase space acceptance of the mass analyzer. Initial
positions and velocities are registered for ions surviving for 13 turns of the analyzer with a total flight time of 1 ms, the
calculated phase space acceptance is 15 mm*mrad. Simulation results indicate that the energy acceptance of the mass
analyzer is ±1% at 300eV which is more than the simulated energy spread of 2eV. The phase space emittance of the
transfer interface at the exit of the immersion lens is also calculated and compared to the phase space acceptance of the
CDMS analyzer.

Please explain why your abstract is innovative for mass spectrometry?

The first use of an electric sector based analyzer for Charge Detection Mass Spectrometry. Advantages include high
resolution, increased space charge and greater experimental throughput over existing designs.

Co-authors:

Dimitris Papanastasiou, Fasmatech

Alexandros Lekkas, Fasmatech

360
DIRECT SINGLE MOLECULE IMAGING ON A MODIFIED Q EXACTIVE
UHMR WITH ELECTRON HOLOGRAPHY CAPABILITY
Abstract ID: 230

Presenting author: Albert Konijnenberg, Thermo Fisher Scientific

Introduction

To understand protein function it is critical to have access to protein structure. Today structures predominantly come
from either X-ray crystallography(XRD) or cryo-electron microscopy (Cryo-EM). However, both techniques have
limitations that require them to either force the protein into an unnatural state possibly altering the conformation (XRD) or
depend on substantial averaging of thousands of particles for the reconstruction protein structure (Cryo-EM).
Furthermore, neither technique allows the user to directly correlate proteoform information such as post-translational
modifications to the protein conformation, which is critical to understand protein function. The ability of mass
spectrometry to provide unaveraged data allows for detection of coexisting proteoforms or conformations. However,
structural mass spectrometry only provides inferred structure and requires a detailed 3D structure to interpret the data.

Methods

Recently, low energy electron holography (LEEH) showed the capacity to image individual nanoscale particles. Critical to
this breakthrough was to deliver particles to an ultraclean substrate in vacuo by ion beam deposition using a mass
spectrometer, after which the substrate was transferred using a vacuum suitcase to a separate instrument for imaging.
The use of a mass spectrometer allows to characterize and filter the species of interest from a complex matrix for
subsequent imaging. Furthermore, making use of low energy electrons, LEEH is ideal to image proteins, as there is little
to no damage associated with these energies during imaging.

Preliminary data (results)

Here we present an integrated mass spectrometry-based system that can image individual single particles with high
contrast and over extended periods of time. Our system combines ion beam deposition of native proteins on a modified
Q Exactive UHMR mass spectrometer with direct protein imaging capabilities using single particle LEEH within a single
instrument. The instrument workflow consists of 3 consecutive steps: acquisition of a native mass spectrum to select the
protein or proteoform of interest, energy controlled deposition of the mass selected protein on an ultraclean graphene
monolayer at ultrahigh vacuum and imaging of the deposited proteins on the substrate using LEEH. In order to validate
the instrument and assess its performance we will show how our setup imaged various types of proteins over a broad
size range. We will demonstrate how the resulting holograms can be reconstructed to generate 2D protein images and
how due to the non-damaging nature of the low energy electrons, even videos can be acquired following the behavior
and dynamics of the particles on the freestanding graphene. Finally, we show how single particle LEEH can image the
structure and dynamics for a FAB-antigen complex (~80 kDa) that was too small and dynamic for Cryo-EM studies, and
where the conformational dynamics were absent in the crystal structure.

Please explain why your abstract is innovative for mass spectrometry?

We developed a single instrument using ion beam deposition of proteins on an ultraclean substrates for subsequent low
energy electron holography imaging, allowing protein structure and dynamics to be studied.

Co-authors:

Josh Gilbert, Thermo FIsher Scientific

Kyle Fort, Thermo FIsher Scientific
Alexander Makarov, Thermo FIsher Scientific
Alan Bahm, Thermo FIsher Scientific

361
Single molecule holographic reconstruction of a FAB-antigen complex(~80kDa)

362
NANO-ELECTRO-MECHANICAL SENSOR MASS SPECTROMETRY FOR
VIRAL PARTICLES CHARACTERIZATION.
Abstract ID: 271

Presenting author: Vaitson Çumaku, UGA (Université Grenoble Alpes), CEA/IRIG (Commissariat à l'Energie
Atomique et aux Energies Alternatives)

Introduction

Nano-electro-mechanical sensor mass spectrometry (NEMS-MS) is an emerging technology to analyze species in the
MDa to GDa mass range, beyond the reach of conventional MS instruments. Allowing single particle measurement
independently on charge state, NEMS-MS represents a paradigm shift in MS. Moreover, NEMS-MS features mass-
independent resolution, which makes it increasingly valuable as species become more massive. This is the case for the
vast majority of viruses, having mass in excess of tens of MDa. Therefore, we investigated the conditions for NEMS-MS
for virus and virus-like particles (VLPs) characterization.

Methods

The NEMS devices considered in this work are doubly pinned beams oscillating in plane (see Fig 1). The mass of
individual particle landing onto the device’s active surface are determined through changes in the device’s resonance
frequencies. An array of 20 NEMS is mounted as detectors in a home-made NEMS-based mass spectrometer prototype
consisting of an atmospheric pressure electrospray nebulization source coupled to three successive differentially
pumped sections, without moving parts or electromagnetic fields for ion guiding. An aerodynamic lens is used to guide
and focus incoming particles onto the detector array.

Preliminary data (results)

Besides sample preparation issues, several nano-ESI source parameters were identified as influencing the mass
measurement of viral particles using NEMS-MS. These include the solvent composition, the solution feeding flow rate,
the ESI voltage, the inlet capillary temperature, and ambient humidity levels. We set out to investigate these parameters
to determine optimal conditions for viral particles analysis. Clearly, solvent composition has a strong influence both on
the viral particle itself and on the ESI process, and optimal buffer conditions may depend on the virus studied. Much
effort went into verification of viral particle integrity and aggregation status as a function of the buffer used. For this
purpose, we used nano-characterization methods to determine the hydrodynamic radius of viral particles prior to NEMS-
MS analysis. In addition, we designed a controlled experiment in which the nano-ESI flow rate was varied over the
course of a NEMS-MS acquisition and the ESI voltage adjusted on the fly to compensate for spray instabilities. This
allowed us to isolate the effect of desolvation on the resulting mass measurement. In these experiments, we observed
signs of several water monolayers on bacteriophage T5 capsid-like particles electrosprayed from pure water solutions
(see Fig 2). The observed mass differences were evaluated in light of known measurement uncertainties of NEMS, and
our results may provide insights on the nebulization and desorption mechanism of large particles with sizes as large as
the droplets formed by nano-ESI.

Please explain why your abstract is innovative for mass spectrometry?

Refinement of conditions for the measurement of single viral particle mass using nano-ESI-NEMS-MS

Co-authors:

Kavya Clement, UGA (Université Grenoble Alpes), CEA/IRIG (Commissariat à l'Energie Atomique et aux Energies
Alternatives)
Adrien Reynaud, CEA/LETI (Commissariat à l'Energie Atomique et aux Energies Alternatives)
Szu Hsueh Lai, CEA/IRIG (Commissariat à l'Energie Atomique et aux Energies Alternatives)
Thomas Fortin, CEA/IRIG (Commissariat à l'Energie Atomique et aux Energies Alternatives)
Sébastien Hentz, UGA (Université Grenoble Alpes), CEA/LETI (Commissariat à l'Energie Atomique et aux Energies
Alternatives)
Christophe Masselon, UGA (Université Grenoble Alpes), CEA/IRIG (Commissariat à l'Energie Atomique et aux Energies
Alternatives)

363
Doubly clamped beam nanoresonator and its vibration modes.

Measured mass of VLPs as a function of flow rate.

364
Session: HC-3 Young MS Scientists - Session B

DECIPHERING O-GLYCOPROTEASE SUBSTRATE PREFERENCES WITH

O-PAIR SEARCH
Abstract ID: 764

Presenting author: Nicholas M. Riley, Stanford University

Introduction

O-glycoproteases are an emerging class of enzymes that selectively digest glycoproteins at positions decorated with
specific O-linked glycans. O-glycoprotease substrates range from any O-glycoprotein (albeit with specific O-glycan
modifications) to only glycoproteins harboring specific O-glycosylated sequence motifs, such as those found in mucin
domains. Their utility for multiple glycoproteomic applications is driving the search to both discover new O-
glycoproteases and to understand how structural features of characterized O-glycoproteases influence their substrate
specificities. Here, we show how O-Pair Search, a recently developed O-glycopeptide-centric identification platform that
enables rapid searches and confident O-glycosite localization, can be used to determine substrate specificities of various
O-glycoproteases de novo from LC-MS/MS data of O-glycopeptides.

Methods

All experiments were performed on an ETD-enabled Orbitrap Fusion MS equipped with a Dionex Ultimate 3000
chromatography system. Data were collected in a product-dependent data-dependent fashion, where HCD MS/MS scans
were collected for every precursor and EThcD MS/MS scans were triggered upon the observation of common oxonium
ions. All raw data were searched using O-Pair Search implemented in MetaMorpheus (0.0.317), which is available
at https://github.com/smith-chem-wisc/MetaMorpheus. Recombinant O-glycoproteins podocalyxin, MUC16, PSGL-1, and
CD43 (3 µg) were digested at ~1:10 protease:protein ratio with several O-glycoproteases, including mutant versions, to
explore how glycan-binding residues can alter substrate preferences.

Preliminary data (results)

O-glycoproteases are dramatically expanding the available toolkit for characterizing O-glycoproteins. Active forms of O-
glycoproteases have been used as tools to improve O-glycopeptide characterization, while catalytically inactive point
mutants have proven useful as O-glycoprotein enrichment and imaging reagents. Key to the utility of each O-
glycoprotease is a detailed understanding of its substrate specificity, which is reliant on the combination of amino acid
sequence and O-glycosylation status. Previous efforts to characterize O-glycoprotease specificities largely relied on slow
and/or unreliable search strategies coupled with manual interpretation of O-glycopeptide spectra to piece together
cleavage motifs in de novo fashion. Our recently developed search strategy, O-Pair Search, provides several features
that can directly benefit these efforts, including an ultrafast identification algorithm that enables semi-specific and non-
specific database searching of O-glycopeptide MS/MS spectra and the ability to confidently localize O-glycan
modifications.

Here we demonstrate how to use O-Pair Search to determine O-glycoprotease specificity de novo, focusing on how
search parameters affect O-glycopeptide identifications that are used for downstream analysis while also showing how
O-Pair Search results can be processed to yield O-glycoprotease cleavage motifs complete with glycosylation
preference. We demonstrate this workflow using multiple recombinant O-glycoprotein standards and several O-
glycoproteases. In particular, we show how O-Pair Search can be used to refine previously described cleavage motifs for
two mucinases, StcE and AM0627, where the degree of glycosylation at each position can be measured and an
unbiased sequence motif can be generated using hundreds of peptides.

Please explain why your abstract is innovative for mass spectrometry?

This work highlights an improved approach for determining O-glycoprotease cleavage specificity mapping through
analysis of O-glycopeptide MS/MS spectra using the newly developed O-Pair Search tool.

Co-authors:

D. Judy Shon, Stanford University

Carolyn R. Bertozzi, Stanford University

365
IN VIVO MONITORING OF FLAVOR RELEASE USING PTR-MS: EFFECT OF
ORAL PROCESSING BEHAVIOR AND FOOD COMPOSITION
Abstract ID: 61

Presenting author: Karina Gonzalez Estanol, Department of Food Quality and Nutrition, Edmund Mach
Foundation, San Michele all’Adige (TN), Italy, Food Quality and Design, Wageningen University, Wageningen,
The Netherlands, Department of Agri-food and Environmental Sciences, Trento University, Trento, Italy

Introduction

Flavour analysis has shown to be an important application area for PTR-MS. For instance, it has been used not only to
understand and manipulate flavour release at different stages of food process conditions, but also for setting quality
control parameters and product specifications. Furthermore, the evolution of aroma resulting from the release of Volatile
Organic Compounds (VOCs) from the food into the mouth and nasal cavity during consumption and the corresponding
organoleptic sensation has proven to be critical for the sensory perception of food and its preference, thus is of major
interest to understand its underlying mechanisms.

The aim of the study was to investigate the effect of oral processing behaviour, food structure and composition on aroma
release and sensory perception of foods.

Methods

Two carriers (bread and sponge cake) were combined with three formulations of strawberry jams varying in pectin and
sugar content. Jams were spiked with citral 0.4% (w/w) and limonene 0.4% (w/w). In vivo release of citral and limonene
and its corresponding perception (citrus flavour intensity) were characterized using a standardized fast and slow chewing
protocol (chewing 25s with a chewing rate of 1.33 or 0.66 chews/s) (n=8women).

Target volatiles released during eating were measured using PTR-ToF-MS (Proton. Transfer Reaction Time-of-Flight
Mass Spectrometer, Ionicon Analytik, Innsbruck, Austria), while the sensory stimulus was rated using continuous Time-
Intensity (TI) profiling.

Preliminary data (results)

We hypothesized that (a) oral processing with a higher chewing rate leads to stronger structural breakdown of foods
leading to flavour release and higher aroma perception and (b) flavor release and perception are strongly affected by
composition of strawberry jams and addition of carrier foods.

Strong effects of chewing rate, food texture and composition on in vivo flavour release and aroma perception were
observed. Data analysis will be completed soon. The coupling of both methods allowed to reveal the presence of cross
modal associations between food texture and aroma perception in complex, real food matrices and to demonstrate the
impact of mastication behaviour on flavour release and perception. Better understanding of inter individual differences
will help in the design of successful food products tailored to specific populations, with specific characteristics

Please explain why your abstract is innovative for mass spectrometry?

By coupling dynamic sensory methods with simultaneous nose-space analysis by high sensitivity direct injection mass
spectrometry, information about in vivo flavour release and aroma perception, during food consumption is obtained.

Co-authors:

Iuliia Khomenko, Department of Food Quality and Nutrition, Edmund Mach Foundation, San Michele all’Adige (TN), Italy
Monica Fontova-Cerdà , Food Quality and Design, Wageningen University, Wageningen, The Netherlands
Markus Stieger , Food Quality and Design, Wageningen University, Wageningen, The Netherlands
Franco Biasioli, Department of Food Quality and Nutrition, Edmund Mach Foundation, San Michele all’Adige (TN), Italy

366
HIGH-MASS MALDI-MS QUANTITATIVELY ANALYSIS OF NONCOVALENT
INTERACTIONS OF MEMBRANE
Abstract ID: 279

Presenting author: Na Wu, ETHz

Introduction

Exploring protein-protein interactions (PPIs) that are dominated by noncovalent binding helps to develop a better
understanding of various disease mechanisms and thus contributes to the development of protein-related drugs and to
treatment optimization. G-protein coupled receptors (GPCRs) constitute the largest family of membrane proteins in
eukaryotes. They are important pharmaceutical targets for the treatment of a broad spectrum of diseases. Upon ligand
binding, GPCRs initiate intracellular signaling pathways by interacting with partner proteins. Assays that quantify the
interplay between ligand binding and initiation of downstream signaling cascades are critical in the early stages of drug
development.

Methods

Quantification of PPIs using MALDI-MS remains challenging. This is because, first, the peak intensity of an analyte in
MALDI-MS is not directly related to its concentrations; Second, noncovalent protein complexes often dissociate during
the MALDI process. Here, we developed an absolute quantitative MALDI-MS strategy that combines chemical
crosslinking and internal standard calibration techniques to study PPIs.

Preliminary data (results)

Using the quantitative MALDI-MS strategy, we determined the selectivity profile and binding affinities of three GPCRs
(rhodopsin, beta-1 adrenergic receptor [β1AR], and angiotensin II type 1 receptor) to engineered Gα-mimicking partner
proteins (mGs, mGo, mGi, mGq), nanobody 80 (Nb80), as well as wild type Gαiβγ and β-arrestin. We found that GPCRs
in the absence of ligand can bind mGo, and that the role of the G protein C-terminus in GPCR recognition is receptor-
specific. We exemplified our quantification method using β1AR and demonstrated the allosteric effect of Nb80 binding in
assisting displacement of nadolol by isoprenaline. We also quantified complex formation with wild-type heterotrimeric
Gαiβγ and β-arrestin 1 and showed that carvedilol induces an increase in the coupling of β-arrestin 1 and Gαiβγ to β1AR.

Please explain why your abstract is innovative for mass spectrometry?

We developed an absolute quantitative MALDI-MS strategy that combines chemical crosslinking and internal standard
calibration techniques to quantify the PPIs.

367
Schematic of the internal standard quantification of the protein-protein interactions.

368
METABOLIC INVESTIGATION OF INFLAMMATION AND OXIDATIVE
STRESS TO FACILITATE COVID-19 DISEASE PREDICTION
Abstract ID: 375

Presenting author: Lieke Lamont, Analytical Biosciences and Metabolomics, Systems Biomedicine and
Pharmacology, Leiden Academic Centre for Drug Research, Leiden University, the Netherlands

Introduction

Most COVID-19 patients suffer from classical respiratory symptoms such as coughing, fever, and nasal congestion.
However, a small group of patients progresses into a life-threatening COVID-19 disease with a possible fatal outcome.
In-depth understanding of the immune response against COVID-19 is still limited. Endocannabinoids (ECs) hydrolyze
into free fatty acids (FFAs) and are important in the immune response. In addition, FFAs and their derived oxylipins
(including prostaglandins, leukotrienes, and thromboxanes) are known as important coagulation and inflammation
markers. We not only hypothesize that dysregulation of these inflammatory markers is triggered by a SARS-CoV-2
infection but also that immune specific differences in metabolite abundances will provide more insight in the
heterogeneity in COVID-19 disease severity.

Methods

We have developed an LC-MS/MS method which enables profiling of inflammatory markers in plasma of COVID-19
patients. The patient cohort consisted of 44 adults (severe and critical) admitted to the Amphia hospital in Breda after a
positive SARS-CoV-2 PCR test; EDTA plasma samples were collected every 3-4 days. Metabolites were measured by
high pH and low pH chromatography after a liquid-liquid extraction. Isotope-labelled internal standards were used to
acquire relative metabolite concentrations. Quality control samples were replicated throughout the batches to assess
data quality. Correlation between several immune markers and inflammatory metabolites were calculated.

Preliminary data (results)

Cytokine IL-6 is often used to monitor the inflammatory process. Paired analysis with IL-6 revealed a strong correlation
with 2-arachidonoylglycerol (2-AG) and its congeners during disease progression. ICU patients showed a 4-11 fold
increase in three acyl glycerol ECs (AG, linoleoylglycerol, and oleoylglycerol) which was not observed in patients towards
recovery. Strong positive correlations were observed for 2-AG and its congeners with pro-inflammatory immune markers
such as neutrophils, CRP, ferritin, TNF-a, and IL-6. FFAs and oxylipins correlated negatively with several pro-
inflammatory immune markers, such as CRP, CXCL10, GMCSF, and IL-6. This study provides biochemical insight of
inflammatory metabolites to facilitate further COVID-19 research in disease prediction and to support potential
therapeutic treatments.

Please explain why your abstract is innovative for mass spectrometry?

This sensitive MS-based metabolomics platform allows the detection of a unique combination of important inflammatory
and oxidative stress metabolites in one single analysis.

Co-authors:

Naama Karu, Analytical Biosciences and Metabolomics, Systems Biomedicine and Pharmacology, Leiden Academic
Centre for Drug Research, Leiden University, the Netherlands
Wei Yang, Analytical Biosciences and Metabolomics, Systems Biomedicine and Pharmacology, Leiden Academic Centre
for Drug Research, Leiden University, the Netherlands
Alida S. D. Kindt, Analytical Biosciences and Metabolomics, Systems Biomedicine and Pharmacology, Leiden Academic
Centre for Drug Research, Leiden University, the Netherlands
Adriaan J. van Gammeren, Department of Clinical Chemistry and Hematology, Amphia Hospital, Breda, The Netherlands
Anton M. M. Ermens, Department of Clinical Chemistry and Hematology, Amphia Hospital, Breda, The Netherlands
Amy C. Harms, Analytical Biosciences and Metabolomics, Systems Biomedicine and Pharmacology, Leiden Academic
Centre for Drug Research, Leiden University, the Netherlands
Lutzen Portengen, Department of Population Health Sciences, Institute for Risk Assessment Sciences, University
Utrecht, Utrecht, the Netherlands
Roel C. H. Vermeulen, Department of Population Health Sciences, Institute for Risk Assessment Sciences, University
Utrecht, Utrecht, the Netherlands
Wim A. Dik, Department of Immunology, Laboratory Medical Immunology, Erasmus MC University Medical Center,
Rotterdam, the Netherlands
Anton W. Langerak, Department of Immunology, Laboratory Medical Immunology, Erasmus MC University Medical

369
Center, Rotterdam, the Netherlands
Vincent H. J. van der Velden, Department of Immunology, Laboratory Medical Immunology, Erasmus MC University
Medical Center, Rotterdam, the Netherlands
Thomas Hankemeier, Analytical Biosciences and Metabolomics, Systems Biomedicine and Pharmacology, Leiden
Academic Centre for Drug Research, Leiden University, the Netherlands

370
TOWARDS SINGLE CELL GLYCOMICS
Abstract ID: 499

Presenting author: Guinevere Lageveen-Kammeijer, Center for Proteomics and Metabolomics, Leiden University
Medical Center

Introduction

While more and more single-cell proteomics studies are performed, the analysis of their post-translational modifications
(PTMs) in a single-cell manner remains limited, as high-resolution and high sensitive platforms are required. Moreover,
high-throughput approaches to profile one of the most abundant PTMs (glycosylation) of minute amount of cells or low
abundant proteins are scarce. This is unfortunate as there is a need for more in-depth studies to investigate the
associations between glycosylation and various diseases, including cancer. In this study, to enhance the release
efficiency and sensitivity, we further developed the release of N-glycans which would require only minute amounts with a
96-well polyvinylidene difluoride (PVDF) membrane-based approach. The optimized sample preparation was combined
with a highly sensitive, capillary electrophoresis-mass spectrometry (CE-MS) platform.

Methods

Several steps in the PVDF membrane approach were evaluated, such as membrane blocking and the N-glycan release
conditions. By implementing the release on PVDF membrane as well as using a hydrazide labeling procedure, no
additional purification steps were required. Analysis was performed on a CE-MS platform and the complete workflow
(Figure 1) was further evaluated by its application for the characterization of the N-glycome from total plasma and a
pancreatic cancer cell line (PaTu-S) using various amount of lysed cells (5 – 5000 cells).

Preliminary data (results)

As the PaTu-S cell line does not express a high abundance of sialylation, the ability of the CE-MS platform for isomeric
separation was investigated using N-glycans released from total plasma. This analysis revealed the coverage of a highly
diverse N-glycome as well as the ability to separate and identify differently linked sialic acids (Figure 2). It is important to
note that this workflow does not require a specific sialic acid derivatization step as well as no further purification was
required after eluting the N-glycans from the PVDF membrane. A total of 58 N-glycans could be relatively quantified from
5000 PaTu-S cells using the optimized PVDF membrane-aided protocol. The four most abundant N-glycans could still be
detected using ~50 cells after cell lysis (~0.2 cells were hydrodynamically injected using 90 nL). While, we already
achieved an exceptional sensitivity, our future endeavors will focus on optimizing the CE-MS conditions (e.g. injection
volume and leading electrolyte composition) to gain even more insights in the N-glycome using minute amount of
samples and maintaining isomeric separation. Finally, we envision to use laser-capture microdissection to define the
minimal amount of material that is required to obtain a representative N-glycome profile of a specific area of interest in
tissues. This would allow us to investigate whether specific glycomic features are solely present in the invasive front of
the tumor, cancerous region or neighboring healthy tissue. Eventually this research will contribute to gaining more in-
depth knowledge about the role of glycosylation in tumor progression.

Please explain why your abstract is innovative for mass spectrometry?

Hitherto, this is the first workflow that enables to detect the N-glycome of approximately 0.2 cells (~50 cells at start of
sample preparation) using a PaTu-S cell line.

Co-authors:

Wenjun Wang, Center for Proteomics and Metabolomics, Leiden University Medical Center
Katarina Madunić, Center for Proteomics and Metabolomics, Leiden University Medical Center
Manfred Wuhrer, Center for Proteomics and Metabolomics, Leiden University Medical Center

371
Figure 1. Workflow for high-sensitivity N-glycomics .

Figure 2. Analysis of released N-glycans from total plasma.

372
OPENING NEW HORIZONS IN LIPIDOMICS – ULTRA-HIGH MASS
RESOLUTION MASS SPECTROMETRY IMAGING WITH AN ORBITRAP
COUPLED TO AN EXTERNAL DATA ACQUISITION SYSTEM
Abstract ID: 339

Presenting author: Andrej Grgic, The Maastricht MultiModal Molecular Imaging (M4I) institute, Division of
Imaging Mass Spectrometry (IMS), Maastricht University, 6229 ER Maastricht, The Netherlands

Introduction

Ultra-high-resolution (UHR) is becoming a prerequisite for resolving the mass spectral complexity encountered during the
mass spectrometry imaging (MSI) of biological samples. However, in the standard configuration, Orbitraps are limited in
mass resolution to 240,000-480,000 at 200 m/z. This poses a problem for MSI of lipids as thousands of lipid-related
signals are concentrated in an m/z range only several hundred Da wide. Until now the UHR MSI measurement with the
highest reported mass resolution was performed by using a 21 T ion cyclotron resonance (FT-ICR) mass spectrometer.
The reported mass resolution was up to 875,000 at m/z 800. On the other hand, it is known that some Orbitraps may
achieve comparable UHR performance. However, this capability is not accessible to general users.

Methods

Here, we present the development of a MALDI-MSI-enabled Q Exactive HF Orbitrap (QE HF) (Thermo Fisher Scientific)
instrumentation set-up coupled to the updated FTMS Booster X2. Such configuration uniquely enables the acquisition
and processing of the phased time-domain transients, providing access to the unreduced data (full profile aFT mass
spectra) from a benchtop Orbitrap platform. Analyzed samples include brain, kidney, and spleen tissue sections.
Norharmane was sprayed on all slides with HTX M3+ sprayer as images were recorded in both positive and negative
mode. The resulting pixel size was 50 µm.

Preliminary data (results)

First, we show that the unreduced data (full profile aFT mass spectra) provide an increase in dynamic range and mass
accuracy compared to the conventional MALDI QE HF instrument with a regular mass resolution performance (reduced
profile eFT mass spectra). Second, with the acquisition of longer, up to 10 seconds, time-domain transients and use of
aFT processing we demonstrate UHR performance exceeding 1,000,000 in the lipid-relevant mass region of 700–900
m/z. We thus report better mass resolution performance than reported for the FTMS MSI platforms up to date. This
research demonstrates that due to the unprecedented performance, including UHR, this advanced MSI set-up is ready to
support further biological discoveries and understanding of lipid metabolism in heterogeneous tissue.

Please explain why your abstract is innovative for mass spectrometry?

Novel mass spectrometry imaging (MSI) instrumentation set-up for ultra-high mass resolution measurements.

Co-authors:

Konstantin O. Nagornov, Spectroswiss, 1015 Lausanne, Switzerland

Anton N. Kozhinov, Spectroswiss, 1015 Lausanne, Switzerland
Ian G. M. Anthony, The Maastricht MultiModal Molecular Imaging (M4I) institute, Division of Imaging Mass Spectrometry
(IMS), Maastricht University, 6229 ER Maastricht, The Netherlands
Yury O. Tsybin, Spectroswiss, 1015 Lausanne, Switzerland
Shane R. Ellis, The Maastricht MultiModal Molecular Imaging (M4I) institute, Division of Imaging Mass Spectrometry
(IMS), Maastricht University, 6229 ER Maastricht, The Netherlands, Molecular Horizons and School of Chemistry and
Molecular Bioscience, University of Wollongong, Wollongong, New South Wales 2522, Australia, Illawarra Health and
Medical Research Institute, Wollongong, NSW 2522, Australia
Ron M. A. Heeren, The Maastricht MultiModal Molecular Imaging (M4I) institute, Division of Imaging Mass Spectrometry
(IMS), Maastricht University, 6229 ER Maastricht, The Netherlands

Scheme of the developed instrumentation set-up for UHMR measurements.

373
Session: LS-16 MS in Structural Biology - Native MS, HDX-MS -Session
B

CYCLIC ION MOBILITY – MASS SPECTROMETRY AND ELECTRON

CAPTURE DISSOCIATION PROBE DIMERISATION OF AGGREGATION-
PRONE IAPP
Abstract ID: 186

Presenting author: Aisha Ben-Younis, Institute of Structural & Molecular Biology, University College London

Introduction

Oligomers formed by human islet amyloid polypeptide (hIAPP) contribute to pancreatic β-cell dysfunction and death in
type II diabetes. Low order oligomers are believed to be the most toxic. Rat IAPP (rIAPP) differs from the human
sequence at just six residues but is non-amyloidogenic and non-toxic to β-cells, although it forms oligomers. The
question arises whether these sequence differences lead to structural differences in the dimeric forms of these peptides.

Here, high resolution cyclic ion mobility – mass spectrometry (cIM – MS) with new electron capture dissociation (ECD)
capability has been applied to investigate detailed gas phase dynamic behaviour and observe conformational transitions
of early oligomers of hIAPP and rIAPP and describe the self-assembly region for their dimers.

Methods

Lyophilised peptides were dissolved in 100% DMSO at a concentration of 1 mM and incubated for 24 h at 37°C. Samples
were prepared by dilution into 100 mM ammonium acetate pH 7.4 at a concentration of 10 μM, with a final DMSO
concentration of 1 % (v/v) before direct infusion into a SELECT SERIES Cyclic IMS QToF (Waters Corp.) fitted with a
post-mobility ECD cell (e-MSion Inc.). Data were analysed using UNIFI Scientific Information System (Waters Corp.) and
our in-house software, AmphitriteX.

Preliminary data (results)

Individual oligomeric species were mobility separated and subjected to collision activation on reinjection into the cyclic
device. Conformer-specific collision activation experiments for the +5 dimer species showed different gas phase
behaviour and stabilities between human and rat IAPP samples. Human IAPP dimers were seen to extend on activation
before forming a super-extended state at high energies immediately before dissociation into monomer. Rat IAPP dimers
however tended to recompact with increasing activation, did not form the super-extended state, and were generally less
stable.

ECD fragmentation experiments localised the dimerisation interface of hIAPP to between residues Arg11 – Ser21,
correlating extremely well with previously published solution phase peptide mapping array studies. The rIAPP
dimerisation interface however was shown not to be in this region, indicating a different structural arrangement.

cIM – MS combined with ECD has been used to show that the highly amyloidogenic hIAPP and non-amyloidogenic
rIAPP dimers display different conformational behaviour, correlating to their solution-phase characteristics. The contrast
between hIAPP and rIAPP dimerisation interfaces provides an insight into why hIAPP proceeds to form β-sheet fibrils
and rIAPP does not and suggests that targeting the human dimer interface could be a strategy to prevent formation of
toxic species

Please explain why your abstract is innovative for mass spectrometry?

Using amyloidogenic IAPP as a model system, cIM – MS combined with ECD has proved a powerful technique for the in-
depth study of protein conformation within a single experiment.

Co-authors:
Alexander Zhyvoloup, Institute of Structural & Molecular Biology, University College London
Hannah Britt, Institute of Structural & Molecular Biology, University College London
Daniel Raleigh, Institute of Structural & Molecular Biology, University College London, Department of Chemistry, Stony
Brook University

374
Konstantinos Thalassinos, Institute of Structural & Molecular Biology, University College London, Institute of Structural &
Molecular Biology, Birkbeck College

THE DYNAMICS OF SARS-COV-2 NSP7-11 POLYPROTEIN PROCESSING

AND COMPLEX FORMATION
Abstract ID: 346

Presenting author: Kira Schamoni-Kast, Leibniz Institute for Experimental Virology, Centre for Structural
Systems Biology (CSSB), Deutsches Elektronen Synchroton (DESY)

Introduction

Since its emergence in 2019 the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) causes COVID-19, an
ongoing pandemic of respiratory disease with a wide spectrum of clinical presentations and outcomes.

On a molecular basis, the virus requires the replication of a very large genome. The major part encodes for two
overlapping polyproteins containing the non-structural proteins (nsps). These are proteolytically processed by two
internal proteases and assemble into a large replication transcription complex. The C-terminal part of the polyprotein
includes amongst other nsp7 and nsp8 as co-factors of the RNA-dependent RNA polymerase (nsp12) and nsp11 that is
disordered in solution and is likely not to be only a side-product of frameshifting. Thus, correct processing of these
regions is essential for viral growth.

Methods

To decipher the molecular mechanisms in SARS-CoV-2 polyprotein processing, biochemical analysis of cleavage order
and complex formation is required.

Native MS was used to monitor the proteolytic processing of the C-terminal end of polyprotein nsp7-11 in a time-resolved
manner. In order to follow the initial dynamics of the processing, the proteolytic activity was recorded as in-capillary
processing. To observe the entire order of cleavage and the coordinated release, an in vitro processing was conducted
simultaneously.

Preliminary data (results)

In our studies, the proteolytic processing of the C-terminal end of SARS-CoV-2 polyprotein (pp1a) nsp7-11 was analyzed
by using native MS. Hereby, the proteolytic processing, cleavage, and complex formation were monitored in vitro in a
time-resolved manner. One central point was to investigate the structural influence of cleavage efficiency of nsp11
domain on the processing order. Furthermore, the processing was conducted in presence of nsp core enzymes and
binders of the cleaved domains. Our latest results on the processing and complexation of different variants with other
CoV proteins will be presented.

Please explain why your abstract is innovative for mass spectrometry?

In this work, dynamics of proteolytic processing of SARS-CoV-2’s polyprotein were investigated. Especially, initial
dynamics of the processing were monitored by recording the reaction as an "in-capillary" process.

Co-authors:

Boris Krichel, Leibniz Institute for Experimental Virology, Centre for Structural Systems Biology (CSSB), Deutsches
Elektronen Synchroton (DESY), University of Siegen
Charlotte Uetrecht, Leibniz Institute for Experimental Virology, Centre for Structural Systems Biology (CSSB), Deutsches
Elektronen Synchroton (DESY), University of Siegen

375
KNOW YOUR TARGET: CHARACTERIZING SNAKE VENOM PROTEIN
COMPONENTS BY NATIVE MASS SPECTROMETRY
Abstract ID: 167

Presenting author: Irina Oganesyan, ETH Zurich

Introduction

Snakebite envenoming is considered a neglected tropical disease (NTD) that affects impoverished rural communities
worldwide with up to 2.7 million cases a year, ~100K of which result in death and three times as many in amputations
and permanent disabilities. Snakebites have been categorized as one of the main NTDs by the World Health
Organization (WHO) since 2017.

Currently, the best treatment of snakebite envenoming is an antivenom consisting of monoclonal immunoglobulins from
large animals (e.g., horses) that were hyperimmunized against a specific venom. A significant challenge in this treatment
lies in the complexity and variation of the protein/peptide mixtures that constitute each venom. Better treatments are
desirable since immunoglobulin production is expensive, time-consuming, and limited to certain venoms.

Methods

Characterization of venoms is important for the development of new types of universal treatments. In-depth knowledge of
the protein profile could promote the development of cross-reactive antibodies/molecules. In this study, five different
snake venoms (O. hannah, N. naja, N. kaouthia, N. melanoleuca, and B. polylepis) were first fractionated by size
exclusion chromatography, and the resulting fractions were analyzed by native mass spectrometry (MS) and ion mobility
(IM) spectroscopy.

Preliminary data (results)

In this study, five different snake venoms (O. hannah, N. naja, N. kaouthia, N. melanoleuca, and B. polylepis) were first
fractionated by size exclusion chromatography, and the resulting fractions were analyzed by native mass spectrometry
(MS) and ion mobility (IM) spectroscopy. Results revealed multiple families of non-enzymatic and enzymatic toxins, with
masses ranging from 5 to 150 kDa, with large variation from one venom to another. The lower mass fractions mostly
consisted of various cytotoxins, for example, long and short neurotoxins, cytotoxins, various phospholipases, nerve
growth factors, and their homologs. Higher mass species were identified as L-amino oxidases, metalloproteases, and
cysteine-rich secretory proteins. Currently, the relative fraction of each protein family is assessed, for further comparison
between the five venoms. The next steps of this research will be focused on tertiary conformations of higher mass
species utilizing ion mobility and collisional- and surface-induced dissociation MS.

Please explain why your abstract is innovative for mass spectrometry?

Assessing the protein profile of each venom will help to recognize the similarities and differences in targets between
snake species and help to create a more universal treatment for snakebites.

Co-authors:

Julian Alexander Harrison, ETH Zurich

Line Jensen Ledsgaard, Technical University of Denmark
Timothy Patrick Jenkins, Technical University of Denmark
Andreas Hougaard Lautsen, Technical University of Denmark
Renato Zenobi, ETH Zurich

376
Snake venom proteins are the targets of this research.

377
BINDING OF A POTENT SARS-COV-2 PEPTIDE INHIBITOR REVEALED BY
INTEGRATED HYDROGEN-DEUTERIUM EXCHANGE MASS
SPECTROMETRY AND CRYO ELECTRON MICROSCOPY
Abstract ID: 328

Presenting author: Joost Snijder, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for
Biomolecular Research and Utrecht Institute for Pharmaceutical Sciences, Utrecht University, Utrecht, The
Netherlands, Netherlands Proteomics Center, Utrecht, The Netherlands

Introduction

SARS-CoV-2 has caused an ongoing pandemic with a tremendous burden on global public health and the economy.
Whereas effective vaccines are available, they are sensitive to immune escape observed in novel variants of concern.
There is still an unmet need for antiviral drugs to treat infection of a broad array of SARS-CoV-2 variants. We identified a
novel peptide inhibitor of SARS-CoV-2 that binds the spike, but its binding site could initially not be determined by cryo-
EM, due to the small size of the inhibitor and flexibility in the spike structure. Here we mapped the inhibitor binding site by
HDX-MS, enabling a focused reconstruction of the EM data that recovered the peptide inhibitor density and revealed its
mode of binding to atomic detail.

Methods

Two deuterated states were acquired, unbound state consisted of the SARS-CoV-2 spike subdomain and the binding
state consisted of SARS-CoV-2 spike subdomain and peptide. These solutions were exposed to buffered 95% D 2O
solution for a duration of 10 seconds, 1 min, 60 min and 240 min. The deuterium exchange reaction was quenched with
6M urea, 300mM TCEP (final pH 2.5). Analysis was performed using the HDX/nanoACQUITY UPLC system coupled to a
Xevo Qtof G2 instrument. By comparing the exchange between bounded and non-bounded state, it was examined where
the peptide binds to the SARS-CoV-2 spike subdomain.

Preliminary data (results)

The deuterium uptake of free SARS-CoV-2 spike subdomain is compared to inhibitor-bound SARS-CoV-2. For the
experiments a sequence coverage of 97.3% is achieved. Upon binding from the peptide to SARS-CoV-2 three protected
areas are observed with an independent t-test (p<0.001). With the above described HDX workflow, the protected areas
are identified. We can conclude from our HDX data where the peptide binds. With these observations it was possible to
reconstruct the binding of SARS-CoV-2 spike subdomain and peptide inhibitor with single particle cryo-EM.

Please explain why your abstract is innovative for mass spectrometry?

Integrating HDX-MS in the structural analysis workflow enabled reconstruction by single particle cryoEM of a challenging
target that was otherwise invisible.

Co-authors:

Nadia Mokiem, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research and Utrecht
Institute for Pharmaceutical Sciences, Utrecht University, Utrecht, The Netherlands, Netherlands Proteomics Center,
Utrecht, The Netherlands
Daniel Hurdiss, Cryo-Electron Microscopy, Bijvoet Center for Biomolecular Research, Department of Chemistry, Faculty
of Science, Utrecht University, Utrecht, The Netherlands, Virology Section, Infectious Diseases and Immunology
Division, De-partment of Biomolecular Health Sciences, Faculty of Veterinary Medicine, Utrecht University, Utrecht, The
Netherlands
Vito Thijssen, Chemical Biology and Drug Discovery, Utrecht Institute for Pharmaceutical Sciences, Utrecht University,
Utrecht, The Netherlands
Berend Jan Bosch, Virology Section, Infectious Diseases and Immunology Division, De-partment of Biomolecular Health
Sciences, Faculty of Veterinary Medicine, Utrecht University, Utrecht, The Netherlands
Frank van Kuppeveld, Virology Section, Infectious Diseases and Immunology Division, De-partment of Biomolecular
Health Sciences, Faculty of Veterinary Medicine, Utrecht University, Utrecht, The Netherlands
Seino Jongkees, Chemical Biology and Drug Discovery, Utrecht Institute for Pharmaceutical Sciences, Utrecht
University, Utrecht, The Netherlands, Chemistry and Pharmaceutical Sciences, Amsterdam Institute of Molecular and
Life Sciences, Vrije Universiteit Amsterdam, Amsterdam, The Netherlands

378
STRUCTURAL MASS SPECTROMETRY REVEALS INSIGHTS INTO DNA
TRIPLEX ASSEMBLIES FOR ANTIGENE TECHNOLOGIES
Abstract ID: 582

Presenting author: Tara Pukala, University of Adelaide

Introduction

Growing evidence suggests that many variations from the well-known Watson–Crick duplex DNA structure play key roles
in a range of cellular processes. For example, nucleic acid triplexes, which form when a third oligonucleotide strand binds
within the major groove of a double helix, have been proposed as under-explored regulators of gene expression and
candidates for therapeutic development.

Compared to other non-covalent biomolecular assemblies, nucleic acid triplexes are poorly characterised, and robust
techniques to investigate them in detail are lacking. Here we have performed fundamental structural mass spectrometry
(MS)-based studies on model oligonucleotide systems to probe the formation and stability of biologically relevant DNA
triplex assemblies, underpinning the rational design of chemically modified DNA oligonucleotides for applications in
antigene technologies.

Methods

A suite of structural mass spectrometry-based experiments, including native mass spectrometry, ion mobility and
collision induced dissociation, in conjunction with molecular dynamics and solution phase measurements have been
employed to probe fundamental features controlling formation, structure and stability of oligonucleotide triplex
assemblies. We have explored a range of oligonucleotide structures of varying length, sequence composition and
backbone chemistry (DNA, RNA, PNA etc) as well as various chemically modified nucleobase structures.

Preliminary data (results)

Our systematic studies of triplex formation for a variety of oligonucleotide structures reveal a range of behaviours
dependant on composition. For example, formation of Ytype triplexes (where the triplex forming third oligonucleotide
strand contains a pyrimidine stretch) is significantly favoured over the R-type (purine stretch), particularly in the absence
of consecutive cytosines. Flanking sequences around the triplex forming site notably alter the propensity to form higher
order assemblies, and different backbone chemistries influence the formation kinetics and stabilities of the triplex
structures formed. This insight has been used to design novel structures for targeted triplex formation in the context of
antigene antibacterial agents.

Of particular interest, we have identified a range of unexpected higher order assemblies such as purine-purine dimer
strands that confound analysis by traditional methods such as UV-vis spectroscopy-based measurements of DNA
melting transitions. This highlights the importance of MS as a means to probe the composition of DNA complexes.

Although our experiments emphasise the importance of MS in the structural study of nucleic acid triplexes, importantly,
this work also highlights key considerations for analysis of these structures in the gas phase. For example, experimental
factors such as buffer composition and ionisation mode can significantly influence detection of such structures, and
collision cross-section measurements along with collision induced dissociation analysis reveal differences between
solution and gas-phase behaviour for some systems.

Please explain why your abstract is innovative for mass spectrometry?

Native ion mobility mass spectrometry provides new insight into non-covalent DNA assemblies important in gene
regulation, and underpins rational design of non-natural triplex forming oligonucleotides.

Co-authors:

Alexander Begbie, University of Adelaide

Jack Klose, University of Adelaide

379
CRYO-EM SAMPLES OF GAS-PHASE PURIFIED PROTEIN ASSEMBLIES
FROM NATIVE ELECTROSPRAY ION-BEAM DEPOSITION
Abstract ID: 484

Presenting author: Tim Esser, Department of Chemistry, University of Oxford, Oxford, UK

Introduction

An increasing number of studies on biomolecular function indirectly combine mass spectrometry (MS) with imaging
techniques such as cryo-electron microscopy (cryo-EM), to obtain information on homogeneity, stoichiometry, shape,
and interactions of native protein complexes, complementary to high-resolution protein structures. This approach proved
effective for optimization of buffer conditions for cryo-EM sample preparation by plunge freezing, as well as for
identification or validation of small ligands and flexible protein regions that can have lower resolution in cryo-EM density
maps. Here we present a soft-landing instrument that provides a direct link between native MS and cryo-EM, which forms
a potential new route to reliable preparation of homogeneous cryo-EM samples and a better understanding of the relation
between native solution-phase and native-like gas-phase structures.

Methods

Mass-selected cryo-EM samples are prepared using native electrospray ion-beam deposition (native ES-IBD, see figure
1). Protein assemblies are transferred into the gas phase via native electrospray ionization. The complex of interest is
separated from fragments, aggregates, and contaminants using mass-selection. The mass-selected ion-beam is
deposited with defined landing energy on TEM grids. Grids are removed from deposition chamber and plunged into liquid
nitrogen, followed by cryogenic transfer to the microscope. Samples are then imaged and micrographs are processed
according to established single-particle analysis (SPA) procedures, resulting in 2D classes and 3D EM density maps.

Preliminary data (results)

We present our implementation of a native ES-IBD instrument (see figure 2), based on a commercial, high-resolution,
tandem MS, designed for analysis of ions with a mass-to-charge ratio (m/z) up to 80,000 m/z (Thermo Scientific™ Q
Exactive™ UHMR). We have increased the ion-source transmission, modified the instrument operation mode, and added
home-made deposition ion-optics and control software. Key parameters for preparation of cryo-EM samples of protein
assemblies, include ion-beam intensity, landing-energy control, and deposition spot size. Typical ion-beam intensities of
tens of pA allow to cover TEM grids within 30 minutes. Careful thermalization and guiding of the mass-selected ion-beam
allows to keep landing energies below 2 eV per charge.

First results demonstrate fabrication of mass-selected and ice-free cryo-EM samples for a diverse selection of protein
assemblies (apo/holo-ferritin, GroEL, ADH, and β-galactosidase). Particles stand out clearly against the background and
different orientations can typically be distinguished by eye in unfiltered micrographs. Protein conformation is largely
preserved during transfer onto TEM grids by carefully controlled native MS and soft landing. Further, we show that low-
resolution 3D reconstruction from ice-free samples is possible using SPA. While protein shapes match expectations from
the literature, a markedly lower spatial resolution can be understood by small but random changes in the secondary and
tertiary structure of each protein, due to dehydration, landing, substrate interaction, transfer under ambient conditions,
and radiation damage. We will address current limitations and ongoing instrumental improvements, required to obtain
performance comparable to state-of-the-art cryo-EM.

Please explain why your abstract is innovative for mass spectrometry?

A new native electrospray ion-beam deposition instrument allows for direct extraction of mass-selected, folded and,
assembled protein complexes from the gas-phase for direct imaging via cryo-EM.

Co-authors:

Jan Böhning, Sir William Dunn School of Pathology, University of Oxford, Oxford, UK
Tanmay Bharat, Sir William Dunn School of Pathology, University of Oxford, Oxford, UK, Structural Studies Division,
MRC Laboratory of Molecular Biology, Cambridge, UK
Joseph Gault, Department of Chemistry, University of Oxford, Oxford, UK
Stephan Rauschenbach, Department of Chemistry, University of Oxford, Oxford, UK, Max Planck Institute for Solid State
Research, Stuttgart, Germany

380
Native electrospray ion-beam deposition (native ES-IBD) workflow.

Q-Exactive UHMR with custom deposition stage.

381
Session: LS-15 Proteomics: Protein -Protein Interaction

BIOPLEX: CURRENT STATUS AND FUTURE PROSPECTS FOR AP-MS

MAPPING OF THE HUMAN INTERACTOME
Abstract ID: 783

Presenting author: Edward L. Huttlin, Harvard Medical School

Introduction

The proteome comprises constellations of proteins whose interactions govern spatial and functional organization and
define cellular state. Yet our knowledge of this network has been fragmentary and largely focused on a small number of
well-studied proteins and complexes, leaving much of this interaction landscape beyond our view. Thus, we established
a pipeline for large-scale AP-MS to profile protein interactions across the entire human proteome. Ten years later, we
have completed over 26,000 AP-MS experiments targeting more than half of all human proteins and identified nearly
280,000 distinct protein-protein interactions across five contrasting cell lines. Here we summarize the biological
insights these networks encode and describe our plans to complete AP-MS profiling of nearly all remaining
proteins in the human proteome.

Methods

Our affinity-purification mass spectrometry platform features lentiviral expression and immunoprecipitation of HA-FLAG-
tagged bait proteins followed by LC-MS analysis on ThermoFisher Q-Exactive or Exploris mass spectrometers with
interaction identification via CompPASS-Plus (Huttlin Cell 2015). Using clones from the Human ORFeome collection, we
have completed AP-MS analysis of over 10,000 genes in 293T and HCT116 cells and have further profiled a subset of
~2000 in U2OS, RPE1, and HeLa cells as well. More recently, we have obtained clones targeting an additional ~7,500
human genes and are now targeting these additional clones as AP-MS baits in both 293T and HCT116 cells.

Preliminary data (results)

Since establishing our AP-MS platform, we have produced multiple interaction networks. The first were obtained in 293T
cells (Huttlin Cell 2015; Huttlin Nature 2017), and we have since incorporated HCT116 cells (Huttlin Cell 2021) as well as
U2OS, RPE1, and HeLa cells (unpublished). Altogether, 10,724 proteins have been baits in at least one cell line,
producing a composite network spanning 279,899 interactions among 15,630 proteins. Thus, 54% of human proteins
have been baits, while 78% reside in the combined network.

Individually and together, these networks enable biological discovery. Specific interactions reveal myriad communities
and complexes, while network context associates uncharacterized proteins with known and novel complexes and
enables prediction of subcellular localization, function, and disease associations. Moreover, comparison of cell-specific
networks reveals extensive remodeling.

Nevertheless, despite the progress we have made, for every bait subjected to AP-MS thus far, nearly as many remain to
be targeted. Targeting each protein is important to identify interactions with maximal sensitivity. Moreover, pulling down
complexes repeatedly via different baits enables “intra-network” confirmation. Thus, targeting additional baits will
increase network depth, coverage, and quality while reducing the sampling bias that results when just a subset of
proteins are targeted as baits. As we target ~7,500 additional baits for AP-MS in 293T and HCT116 cells, we will obtain
independent experimental confirmation for many interactions while revealing interactome remodeling in distinct cellular
contexts. We expect to target up to 91% of human proteins as baits and reveal interactions among up to 95% of
the human proteome.

Please explain why your abstract is innovative for mass spectrometry?

We are mapping the human interactome with unprecedented depth and cell-specific detail via AP-MS of nearly every
human protein.

Co-authors:

Laura Pontano Vaites, Harvard Medical School

J. Wade Harper, Harvard Medical School
Steven Gygi, Harvard Medical School

382
Combined BioPlex Network Summary

Example Complexes by Cell Line: Baits dark, preys light.

383
HIGH-THROUGHPUT MASS SPECTROMETRY METHODOLOGY TO FISH
PEPTIDE TOXINS FROM CRUDE VENOMS BY AFFINITY CAPTURE ON
CELL MEMBRANE RECEPTORS
Abstract ID: 256

Presenting author: Lou Freuville, Mass Spectrometry Laboratory-ULiège

Introduction

Animal venoms are complex chemical cocktails, containing a wide range of biologically active peptides whose selectivity
and efficiently act against membranes targets, such as ion channels. Venom peptides are largely used in pharmaceutical
domains. They serve for the study of their corresponding receptors and in pharmacophore modeling to identify leads for
the development of novel therapeutic molecules. The scope of venoms for drug discovery is rapidly emerging but is still
mostly undone, mainly due to the complexity of venoms, but also due to the low throughput observed when venoms have
to be screened towards various molecular receptors of interest.
In this context, our work aims to propose a faster methodology of toxin selection based on affinity capture on cellular
membranes and monitored by mass spectrometry.

Methods

The venoms of ten arthropods species were analyzed by shotgun proteomics, following the classical bottom-up
approach. The sequences of digested peptides were compared to toxins already known to be active on potassium
channels.
Cellular membranes overexpressing potassium channels were produced and purified in-house, from transfected HEK293
cells.
These membranes were incubated in multi-well plates together with crude venoms or chromatographic fractions. Toxins
displaying affinities for the channels bind to the membranes, whereas toxins which do not display any affinity are
maintained in solution. The mixture containing the candidates is analyzed by MALDI mass spectrometry to identify the
potential ligands.

Preliminary data (results)

To allow a fast and robust screening of the crude venoms to select the most promising species, we exploit the
methodology developed in the laboratory by Echterbille et al. (2017). It was set up to discover new ligands of nicotinic
acetylcholine receptors (nAChRs) from complex mixtures of peptides. They used this one to demonstrate the specificity
of the binding of a-conotoxins to nAChRS. Also, the method had to be able to discover new nAChRs ligands from this
crude venom. They were demonstrated that their workflow is able to detect multiple ligands in only one experiment and a
new toxin targeting their receptors has been discovered. We will apply this preliminary methodology to perform a
discovery tool to highlight potential new toxins targeting K + channels.
In this context, we have already analyzed the ten venoms by shotgun proteomics (LC-Q-Exactive). To do that, we have
realized the reduction, alkylation and digestion with trypsin protocol (bottom-up approach). Moreover, the membrane
overexpressing the receptors of interest were prepared in-house. The receptor concentrations were measured with
saturation binding elements, where membranes were incubated with increasing concentrations of radioligand. After
sample filtration, the radioactivity was quantified to estimate the maximal number of binding sites available.

Please explain why your abstract is innovative for mass spectrometry?

This study enables to highlight potentially new toxins targeting potassium channels. Analysis by MALDI permits to isolate
the best ligand candidates, although they were initially in minority in the venom.

Co-authors:

Fernanda Gobbi Amorim, Mass Spectrometry Laboratory-ULiège

Romain Vitello, Laboratory of Medicinal Chemistry and C.I.R.M-ULiège
Rudy Fourmy, Alphabiotoxine Laboratory sprl
Aude Violette, Alphabiotoxine Laboratory sprl
Jean-François Liégeois, Laboratory of Medicinal Chemistry and C.I.R.M-ULiège
Alain Brans, Centre for Protein Engineering-ULiège
Loïc Quinton, Mass Spectrometry Laboratory-ULiège

384
WEIGHTING THE SWEET GLUE OF ANTIBODY-RECEPTOR
INTERACTIONS: STRUCTURAL AND FUNCTIONAL GLYCOPROTEIN
CHARACTERIZATION BY HYPHENATED MS TECHNIQUES
Abstract ID: 380

Presenting author: David Falck, Center for Proteomics and Metabolomics, Leiden University Medical Center

Introduction

The interaction between Fc gamma receptors (FcγRs) and immunoglobulin G (IgG) is a key element in the human
immune system.(Falk Nimmerjahn, Jeffrey V. Ravetch, Nat.Rev.Immunol.,2008, 8, 34) Glycans on both proteins are
essential for their interaction, forming critical carbohydrate-carbohydrate and carbohydrate-peptide interactions. The
influence of IgG glycosylation is well established and has led to therapeutic advances. Despite an equally essential role,
FcγR glycosylation in humans is largely uncharacterized and its function unknown, mainly due to a lack of suitable
techniques. In addition, in many (patho-)physiological conditions the impact of aberrant glycosylation of IgG, and even
more so other antibody species, remains to be fully elucidated.

Methods

Constantly, we have pushed the boundaries of classical bottom-up approaches in biopharmaceutical, biomedical and
clinical glycomics studies. With our unique and flexible balance between molecular resolution, sensitivity and throughput,
we have been able to study thousands of patient samples and address such challenging questions as differences
between the glycosylation of all antibodies and pathogenic autoantibodies in one patient. For example, we can now
perform relative quantitation of IgG and IgA - consisting of five subclasses with over 100 glycoforms in total - in a site-
and protein-specific manner in a 21 min measurement.(Ana Momčilović et al., Anal.Chem., 2020, 92(6), 4518)

Preliminary data (results)

Furthermore, we have developed a targeted glycoproteomics platform for mapping the site-specific glycosylation of
FcγRIII in primary human cells (Figure1).(Iwona Wojcik et al., Anal. Chem., 2020, 92(19), 13172) For neutrophil FcγRIIIb,
we observed striking differences between neutrophil-bound and free receptor, associated with resting cells and immune
activation, respectively.

The measurement of affinity of the numerous potential combinations of FcγR and IgG glycoforms is an unsolved
technical challenge. By combining FcγR affinity chromatography (AC) with native protein mass spectrometry (MS), we
can simultaneously assess the FcγR affinity of a mixture of IgG glycoforms. We applied the method to the neutrophil
FcγRIIIb, whose activation is important in the mechanism of action of anti-cancer therapeutic monoclonal antibodies
(Figure2).(Steffen Lippold et al., mAbs,2021, 13(1), 1982847) The relatively low affinity towards FcγRIIIb is a challenge,
even for classical binding assays. However, utilizing the excellent flexibility of the method, we could combine superior
affinity and molecular resolution with a broad coverage of affinities and sizes. Interestingly, we observed a significant
influence of the antigen-binding domain on receptor binding, normally assumed to be fully controlled by the Fc fragment.

Please explain why your abstract is innovative for mass spectrometry?

Combining molecular resolution, sensitivity and throughput for glycoprotein analysis. FcγR AC-MS was first
demonstrated by us. We combine multiple diverse MS approaches to study a complex and fundamental immunological
phenomenon.

Co-authors:

Steffen Lippold, Center for Proteomics and Metabolomics, Leiden University Medical Center
Iwona Wojcik, Center for Proteomics and Metabolomics, Leiden University Medical Center
Thomas Sénard, Center for Proteomics and Metabolomics, Leiden University Medical Center
Ana Momčilović, Center for Proteomics and Metabolomics, Leiden University Medical Center
Simone Nicolardi, Center for Proteomics and Metabolomics, Leiden University Medical Center
Elena Domínguez-Vega, Center for Proteomics and Metabolomics, Leiden University Medical Center
Alexander Knaupp, Pharma Research and Early Development, Roche Innovation Center
Dietmar Reusch, Pharma Technical Development, Roche Innovation Center
Peter A. van Veelen, Center for Proteomics and Metabolomics, Leiden University Medical Center
Gestur Vidarsson, Department of Experimental Immunohematology, Sanquin Research, Landsteiner Laboratory,
Academic Medical Center

385
Tilman Schlothauer, Pharma Research and Early Development, Roche Innovation Center, Biological Technologies,
Genentech Inc, South San Francisco, USA
Manfred Wuhrer, Center for Proteomics and Metabolomics, Leiden University Medical Center

Glycopeptide mass spectrum of the human neutrophil Fc gamma receptor.

Fc gamma receptor affinity chromatography - mass spectrometry.

386
LASER-FREE FLASH OXIDATION (FOX) HYDROXYL RADICAL PROTEIN
FOOTPRINTING SYSTEM ACCURATELY MAPS THE PARATOPE AND
EPITOPE OF TNFΑ BOUND TO ADALIMUMAB
Abstract ID: 553

Presenting author: Emily Chea, GenNext Technologies

Introduction

The Fox Protein Footprinting System is a novel hydroxyl radical protein footprinting (HRPF) method that uses a
proprietary flash oxidation lamp which obviates the use of synchrotron or dangerous excimer laser sources to promote
covalent modification. HRPF methods generate hydroxyl radicals (•OH) that irreversibly modify solvent exposed side
chains of amino acids. As solvent accessibility changes, the •OH modification concordantly changes. With the addition of
real time radical dosimetry, the effective •OH concentration can be determined and easily adjusted, allowing conditions
with varying scavenging effects to be easily compared. With the Fox System, the protein surface involved in the paratope
and epitope of TNFα, a pro-inflammatory cytokine, with Adalimumab, a monoclonal antibody prescribed to treat
inflammatory diseases, is determined.

Methods

Using the Fox System, three conditions (3 µM TNFα, 1.5 µM adalimumab, and 3 µM TNFα + 1.5 µM adalimumab) were
labeled. Samples were mixed with 1 mM adenine and 100 mM H 2O2. Samples were introduced to the Fox System using
a 12 µL injection loop. Running buffer pushed the sample passed the flash oxidation lamp, photolyzing H 2O2 to form •OH.
Downstream of the photolysis module, a dosimeter module accurately determined the effective •OH concentration.
Finally, samples were collected in 20 µL of quench, 0.3 mg/mL catalase and 35 mM methionine, and underwent a tryptic
digestion for bottom-up proteomics.

Preliminary data (results)

Each condition, 3 µM TNFα, 1.5 µM adalimumab, and 3 µM TNFα + 1.5 µM adalimumab, was labeled using the Fox
System. Using real time radical dosimetry, the differing •OH scavenging effects between the conditions were
compensated by adjusting the flash lamp voltage, so each condition experienced the same effective •OH concentration.
Following tryptic digestion and bottom-up proteomics, the chromatographic peak areas for all unmodified and modified
peptides were determined and the average peptide and residue oxidation was calculated using FoxWare™ Protein
Footprinting Data Processing Software. Regions were TNFα and Adalimumab interact are protected from •OH
modification, resulting in a significant decrease in average oxidation. All TNFα and Adalimumab regions with a significant
change in oxidation are a part of the paratope or epitope identified in the available crystal structure. Taken collectively,
our results illustrate that the Fox Protein Footprinting System can confidently and accurately map an antibody-antigen
interface.

Please explain why your abstract is innovative for mass spectrometry?

Epitope and paratope mapping using a laser-free or non-synchrotron oxidation footprinting system coupled to mass
spectrometry has not been previously performed.

Co-authors:

Sandeep Misra, Department of BioMolecular Sciences, School of Pharmacy, University of Mississippi

Joshua Sharp, Department of BioMolecular Sciences, School of Pharmacy, University of Mississippi, Department of
Chemistry and Biochemistry, University of Mississippi, GenNext Technologies
Scot Weinberger, GenNext Technologies

387
HOS of Adalimumab upon TNFa binding.

HOS of TNFa upon Adalimumab binding.

388
NATIVE TOP-DOWN MASS SPECTROMETRY WITH COLLISION- AND
ELECTRON-BASED DISSOCIATION YIELDS HIGHER ORDER STRUCTURE
INFORMATION FOR PROTEIN COMPLEXES
Abstract ID: 60

Presenting author: Joseph Loo, University of California Los Angeles

Introduction

Native mass spectrometry (MS) of proteins and protein assemblies reveals size and binding stoichiometry. But
elucidating their structures to understand their function is more challenging. We show that native MS and native top-
down MS (nTDMS), i.e., fragmentation of the gas-phase “native” protein, can be effective for deriving structural
information for soluble and membrane protein complexes (including ligand and lipid binding), and much of this
information can be correlated to their solution-phase structures. However, the activation/dissociation methods used to
fragment the gas-phase proteins and derive sequence-specific information can greatly affect the resulting information on
protein higher order structure. We show that even slower heating methods that can be implemented on many
instruments can yield information on protein 3D structures.

Methods

nTDMS was performed using a Bruker solariX 15T Fourier transform ion cyclotron resonance instrument (with ECD and
CAD) and a Thermo Q-UHMR orbitrap system with CAD (HCD) and ECD. Deconvoluted peaks were assigned by our
program, ClipsMS (terminal and internal fragments), with an error tolerance of 1-5ppm.

Preliminary data (results)

nTDMS endeavors to fragment covalent bonds in an intact biomolecule or complex in a conformation-sensitive manner,
such that information about higher-order structure can be inferred from the fragmentation pattern. nTDMS generates
information on the surface topology, ligand binding sites, post-translational modifications (PTMs), and proteoforms of
proteins and protein complexes. Methods such as electron capture/transfer dissociation (ECD and ETD, or ExD) and
ultraviolet photodissociation (UVPD) typically fulfills this requirement. Surface accessible residues are primarily released
as fragments in a nTDMS experiment. A long-held belief is that slower heating collisional activation induces protein
unfolding and subunit release, and thus collisionally activated dissociation (CAD) and higher energy C-trap dissociation
implemented on Orbitrap platforms are not useful for nTDMS, i.e., for probing 3D structures. For some protein complexes
up to 400 kDa, however, compared to complex-down experiments in which fragmentation of released protein subunits in
an initial activation step yields results similar to dissociation of denatured proteins, we show that direct nTDMS of the
complex using CAD directly dissociates protein complexes without releasing subunits and bound ligands to yield
sequence-bearing product ions consistent with their native 3D structures, potentially providing higher order structural
information about native-like complexes. In addition, assignment of internal fragments (i.e., fragments containing neither
terminus) resulting from fragmentation of intact proteins using CAD, ExD, and UVPD increase sequence coverage and
gives further information on higher order structure.

Please explain why your abstract is innovative for mass spectrometry?

Similar to ExD, CAD fragmentation of large protein complexes can yield terminal and internal product ions that give
information on their 3D structures.

Co-authors:

Carter Lantz, University of California Los Angeles

Benqian Wei, University of California Los Angeles
Wonhyeuk Jung, University of California Los Angeles
Jessie Le, University of California Los Angeles
Andrew Goring, University of California Los Angeles
Boyu Zhao, University of California Los Angeles
Rachel Ogorzalek Loo, University of California Los Angeles

389
DEEP TUMOUR PROFILING OF HER2+ BREAST CANCER BIOPSIES
IDENTIFIES SIGNATURES PREDICTIVE OF TREATMENT RESPONSE
Abstract ID: 187

Presenting author: Donna Olivia Debets, Utrecht University

Introduction

Breast cancer is a highly heterogeneous disease, which is classified in subtypes with distinct molecular signatures, each
with their own treatment strategy. Trastuzumab and Pertuzumab form the foundation of the therapeutic regimen for
HER2-overexpressing tumours. These therapies have proven very effective in some patients, yet treatment resistance is
prevalent. Full understanding of the molecular mechanisms underlying treatment resistance in vivo is lacking and protein
biomarkers to predict treatment success of HER2-targeted therapies are still absent. Therefore, we performed
(phospho)proteomics analysis of 45 treatment-naïve HER2+ breast cancer biopsies to identify molecular signatures
predictive of treatment response. Additionally, we measured the kinase activity profiles of a subset of these biopsies
(n=32) using a system-wide kinase activity analysis by targeted mass spectrometry.

Methods

Patients enrolled in the TRAIN2 study were included in this analysis. After needle biopsies were taken, patients
completed multiple drug treatment cycles and treatment response was determined at surgery. Biopsies were lysed and
digested. For (phospho)proteomics analysis, peptides were labelled using TMT and deep (phospho)proteomics profiling
was obtained using off-line high pH fractionation. Phosphorylated peptides were enriched using Fe-IMAC and analysed
by LC-MS/MS DDA. For the kinase activity measurements, after protein digestion, heavy labelled synthetic peptides of
kinase activation loop sequences were spiked in prior to enrichment of endogenous phosphorylated peptides. These
were analysed by SRM.

Preliminary data (results)

In this study, we achieve deep (phospho)proteomics profiling and high coverage kinase activation loop phosphorylation
with high data quality using limited sample input, demonstrating the feasibility of microscale clinical proteomics. Our data
reveals that accurate quantification of ESR and HER2 biomarkers, combined with the assessment of associated
biological features, can improve treatment outcome prediction. HER2 protein levels were significantly downregulated
amongst the treatment resistant tumours compared to responsive tumours, and low kinase activity within the HER2
pathway was associated with poor treatment outcome. Furthermore, we identified a subgroup of patients with an ESR-
dependent signature, which showed enrichment of treatment resistant samples. Within this subgroup, a significant
upregulation of PRKD2 kinase activity was found amongst the treatment resistant samples. Additionally, we identified
three cellular mechanisms that precondition tumours to resist therapy: unfolded-protein-response (UPR) induced cellular
dormancy (accompanied by low kinase activity of CDKs), a metabolic switch towards oxidative phosphorylation and
reduced numbers of tumour immune cell infiltration. Lastly, MARK kinase activity was significantly upregulated amongst
drug-responsive tumours, which represents a potential new player in therapy resistance.

Multiple resistance mechanisms coincide in treatment resistance tumours, indicating that the combination of drug-
evading mechanisms determines the tumour’s fate. We identified five proteins representative of these resistance
mechanisms in the proteomics dataset and four activated kinases that both constitute to a strong signature associated
with treatment success. Our results highlight the multifactorial nature of drug resistance in vivo and demonstrate the
necessity of deep tumour profiling.

Please explain why your abstract is innovative for mass spectrometry?

The application of targeted MS approach to perform system-wide kinase activity profiling to predict treatment outcome in
vivo.

Co-authors:

Kelly Stecker, Utrecht University

Maarten Altelaar, Utrecht University
Erik de Graaf, Pepscope

390
391
 Friday 2 September 2022: 08:30 – 10:30
Session: AD-5 Environmental and Earth MS

ENVIRONMENTAL DART: HITTING COMPOUNDS QUALITATIVE AND

SEMI-QUALITATIVELY
Abstract ID: 220

Presenting author: Jan Jordens, VITO

Introduction

PFAS are getting more attention on the worldwide agenda as a compound of concern in the environment. PFAS is a
generic term for a broad family (>4700 identified compounds) of polyfluorinated compounds that are highly persistent due
to their chemical properties. In most analyses only a small subset of PFAS (1-50) are quantitively analysed. Drawback of
target analysis, is the time consuming sample preparation and analysis of the compounds.

We develop a DART-MS screening of different sample types (for example; dust, water, foam) to determine if PFAS are
present above a certain threshold. This allows a preselection of samples and compounds in subsequent targeted or
untargeted analyses. The advantage is a high-throughput of samples for the target analysis with a limited well-defined
compound set.

Methods

The DART ionization source, set in a temperature range of 250-500 °C is coupled to a Q-Exactive mass spectrometer,
measuring both in positive and negative ion mode. The obtained mass spectra are filtered for PFAS-specific mass defect
range and are exported to an external database used for the identification of PFAS marker peaks. The results can be
linked to parent PFAS compounds and concentration ranges of these PFAS

Preliminary data (results)

First PFAS reference compounds were analysed using DART-MS at different temperatures and ionization settings.
DART-MS uses plasma ionization at high temperatures, so it can be expected that certain compounds degrade during
analysis. The measurements show that different PFAS classes behave differently in DART-MS. For example N-MeFOSA
remains intact up to 500 °C, while larger perfluorinated carboxylic acids seem to degrade to shorter chain lengths. The
behaviour of known PFAS with reference compounds is studied. The spectral information obtained from these references
is exported to an external home-made database, where the peaks are linked to their actual composition as well as to
intact PFASes or PFAS classes they originate from.

Besides for PFAS identification, DART-MS is also used for an indication of the quantity of PFAS present in the sample.
For example in water, PFAS can be detected in the ppb range, in dust detection ranges of ng/filter to capture the dust,
could be detected.

The combination of qualitative and semi-quantitative information obtained from fast DART-screening allows improved
sample selection for further targeted and untargeted research, increasing sample throughput and total PFAS coverage.
Moreover, the database is extended towards non-PFAS compounds, allowing detection of untargeted compounds of
concern outside the requested window of compounds in provided samples.

Please explain why your abstract is innovative for mass spectrometry?

Coupling of DART-MS with an external database and automation of data interpretation, leading to qualitative and semi-
quantitative information on PFAS in different matrices.

Co-authors:

Griet Jacobs, VITO

Stefan Voorspoels, VITO

392
DART-MS spectrum of dust after mass defect filtering.

393
THE APPLICATION OF MALDI MASS SPECTROMETRY IMAGING IN THE
EVALUATION OF PATHOGENESIS OF AGROBACTERIUM TUMEFACIENS
IN CULTIVATED DICOTYLEDONS
Abstract ID: 668

Presenting author: Katarzyna Suśniak, Department of Genetics and Microbiology, Maria Curie-Sklodowska
University

Introduction

Agrobacterium tumefaciens are soil, Gram-negative bacteria, causing crown gall deformations on roots and stems of
many varieties of dicotyledons and some gymnosperms. A. tumefaciens attacks mainly wounded plants, recognising
them by compounds released at the places of injury, such as sugars and phenols. After infection the pathogen
transforms hosts’ cells, forcing their excessive division, which leads to the formation of tumour tissue.

One of the main factors determining pathogenic interactions between Agrobacterium and plant is bacterial outer
membrane lipid composition. The main components of these bacteria’s membrane are phosphatidylethanolamine,
phospphatidylcholine and phosphatidylglicerol. However, any changes in membrane lipid composition may cause
alterations in Agrobacterium virulence.

Methods

In this study, the ability of mutant in the phosphatidylserine synthase gene (Atu1062), defective in
phosphatidylethanolamine synthesis was analyzed, to conduct the transformation of plant cells and their virulent
capabilities, and the wild (virulent) strain A. tumefaciens C58 was used as the positive control. We monitored the
development of the infection and tumor growth using MALDI mass spectrometry and MALDI mass spectrometry imaging.
The MS and MSI was used to analyse irreversible metabolic changes under the influence of bacterial infection as well as
its spatial distribution changes.

Preliminary data (results)

In our study the MALDI MSI technique was used to analyse the plant tissues under the influence of wild type and by the
mutant and its complementats. The tested mutant of A. tumefaciens C58 in the phosphatidylserine synthase gene
despite the change in the outer membrane composition, is capable of infecting plants, which is manifested by the
formation of callus tissue. The MS analyses allowed to examine signalling compound participating in plants antimicrobial
reaction in the stem and callus tissue – glycoalcaloid α-tomatin. The mutant infection cause smaller changes in the
spatial distribution of alfa-tomatin in the tumor tissue and in the stem compared to the wild strain and complementants.

This work was supported by National Science Centre within the OPUS project (2018/31/B/NZ9/01755).

Please explain why your abstract is innovative for mass spectrometry?

MALDI MS imaging gave deeper insight into the process of pathogen infection and plants antimicrobial reaction and its
changes as a result of changes in bacterial lipid composition.

Co-authors:

Anna Sroka-Bartnicka, Independent Unit of Spectroscopy and Chemical Imaging, Medical University of Lublin
Adam Choma, Department of Genetics and Microbiology, Maria Curie-Sklodowska University
Iwona Komaniecka, Department of Genetics and Microbiology, Maria Curie-Sklodowska University

394
STRUCTURE-AIDED PROFILING OF NATURAL ORGANIC MATTER BY
FTICR MS
Abstract ID: 637

Presenting author: Alexander Zherebker, Skolkovo institute of science and technology

Introduction

Natural organic matter (NOM) is ubiquitous in the environment and important subjects of biogeochemical
research. Continuous attempts are made to go beyond molecular formulae to the structural information about NOM
individual components in non-target profiling in order to predict their chemical properties, biolability and fate. However,
excluding sporadic reports on fragmentation experiments, limited formula-based estimation are widely applied by the
community: e.g. aromaticity index (AI) proposed as a proxy for carbon skeleton accounted unsaturation. However, such
approximations are based on integral parameters about NOM samples from orthogonal analyses, and ignore the
structural features of NOM constituents. At the same time, chemical tagging can be used to enumerate structural
moieties for NOM components enabling structural profiling and development more precise formula-based indices.

Methods

NOM samples were isolated from various sources including soil, peat, coal, river and permafrost. Fractionation was
performed by step-wise extraction on nonionic sorbent at steadily lowered pH values. Using a set of selective chemical
reactions functional groups and aromatic fragments were determined in the parent samples and their fractions. Parent
and tagged samples were analysed using 7T FT MS Bruker Apex Ultra with harmonized cell (Bruker Daltonics) equipped
with ESI source operating in negative ion mode. FTICR MS data were processed in order to extract corresponding
reaction series using custom python-scripts.

Preliminary data (results)

Applying the proposed pipeline, we obtained distributions of functional groups through molecular ensembles of NOM of
drastically different origins. FTICR MS enabled to resolve reaction series and reliably enumerate carboxylic, carbonyl and
phenolic groups. Data treatment included the measuring the length of reaction series and the distribution of peaks within
series. Repeating the reaction on the tagged samples, we observed differences in the samples with high transformation
degree and permafrost. In highly processed materials, the peak distribution shift was observed reaching its maximum
after second iteration. This indicated the high isomeric complexity of the samples. In the permafrost NOM reaction
repetition resulted in the distribution kink indicating the presence of several isomers. For some species the number of
COOH determined by deuteromethylation occupied all oxygen atoms implying ionization through C-H bonds.
Respectively, another tagging with 15N-labeled glycine was conducted. This method proved the possibility of C-H
dissociation in NOM. Considering functional groups in variety of NOM, constrained AI con was proposed. This index
provided the smalles deviation toward experimental data as compared to conventional AI mod index. The applicability of
aromaticity index was tested using mild bromination reaction. The distribution of aromatic compounds for NOM samples
was obtained. Ultimately, attribution of individual compounds to aromatic structures by AI con was well-supported by
bromination.

Please explain why your abstract is innovative for mass spectrometry?

Selective chemical tagging enabled to enhance the application of FTICR MS and obtain structural and constrained
formula-based profiling of geochemical samples

Co-authors:

Oliver Lechtenfeld, Helmholtz Centre for Environmental Research

Tatyana Mikhnevich, Skolkovo institute of science and technology
Gleb Rukhovich, Skolkovo institute of science and technology
Alexey Orlov, Skolkovo institute of science and technology
Evgeny Nikolaev, Skolkovo institute of science and technology

395
DEVELOPMENT AND VALIDATION OF A PYROLYSIS - GAS
CHROMATOGRAPHY – HIGH RESOLUTION MASS SPECTROMETRY
METHOD FOR THE DETERMINATION OF NANO- AND MICROPLASTICS IN
RIVER WATER AND SEDIMENT SAMPLES
Abstract ID: 673

Presenting author: Eva De Rijke, University of Amsterdam

Introduction

Micro- and nanoplastics (MNP) are emerging contaminants in the environment. Measuring concentrations and sizes of
micro- and nanoplastics in the environment is essential to assess their risks . MNP have been detected globally in a
variety of ecosystems. Their determination, however, is still challenging (1). Recently, pyrolysis gas chromatography to
mass spectrometry (Py-GCMS) was introduced for mass related quantification of microplastics (2). Under high
temperature and oxygen-free conditions thermal breakdown of polymers takes place and characteristic volatile pyrolytic
products are generated. Under reproducible pyrolysis conditions following gas chromatographic separation and mass
spectrometric detection, individual polymer types can be quantified (3). Pyrolysis GC-MS has shown to be promising for
the detection of MNP in environmental samples and can give information on the different polymer types and quantities

Methods

In this study, different commonly occurring MNP polyethylene (PE), polypropylene (PP), polystyrene (PS),
polymethylmetacrylate (PMMA), polybutylmetacrylate (PBMA), polyvinylchloride (PVC), polyethylene terephthalate
(PET), polytetrafluoroethylene (PTFE) were investigated. The preparation of standards was optimized by comparing
cryo-milling with quartz sand and dissolution in various solvents. A low resolution pyrolysis GC-MS method was
developed and transferred to a py-GC-HRMS Orbitrap system. Both methods were validated. Furthermore, a double-shot
method was developed to qualitatively determine volatiles additives present in the MNP. As an application, pre-
concentrated nanoplastics from aqueous environmental samples (surface water and river sediment) were qualitatively
and quantitatively investigated with the optimized methods.

Preliminary data (results)

For all MNP except PP three indicator pyrolysis products were found and limits of detection were below 0.02 ng absolute
in all cases, with R2 values rangeing between 0.990 and 0.999. With the optimized py-GC-HRMS method PS an PE were
successfully identified in sediment samples and PET in the water samples.

1 1. Mintenig et al. Environ. Sci.: Nano, 2018, 5, 1640

2 2. M. Fischer and B. M. Scholz-B¨ottcher, Environ. Sci. Technol., 2017, 51, 5052–5060.

3 3. T. P. Wampler and E. J. Levy, J. Anal. Appl. Pyrolysis, 1987, 12, 75–82.

Please explain why your abstract is innovative for mass spectrometry?

The development, validation and application of py-GC-HRMS with an orbitrap system for MNP is novel and has great
potential in environmental research.

Co-authors:

Samira Absalah, University of Amsterdam

Tristan Jansen, University of Amsterdam, Hogeschool Utrecht
Rachel A Addoquaye, University of Amsterdam, Fresenius Hochschule Idstein Germany
Annemarie van Wezel, University of Amsterdam

396
APPLICATION OF CLUSTERIZATION ALGORITHMS FOR ANALYSIS OF
SEMI VOLATILE POLLUTANTS IN ARKHANGELSK SNOW
Abstract ID: 168

Presenting author: Dmitrii Mazur, Lomonosov Moscow State University, Chemistry Department

Introduction

Environmental exposure assessment is an important step in establishing a list of local priority pollutants and finding the
sources of the threats. Arkhangelsk is the largest city of North-European part of Russia, where a number of industrial
facilities are located. Arkhangelsk experiences subarctic climate with heavy snowfalls during 7 months. Therefore, snow
analysis represents an efficient approach for the estimation of long-term air pollution during winter period due to
accumulation and preservation of environmental contaminants by snow.

Methods

Ten snow samples were collected in January 2021 in Arkhangelsk. Classical liquid-liquid extraction of molten water with
dichloromethane was used for sample preparation. The extracts were analyzed using Pegasus GC-HRT+ 4D high-
resolution mass spectrometer combined with Agilent 7890A gas chromatograph in the GC×GC mode. Identification and
quantification of priority pollutants was done using standards, while NIST20 mass spectral library together with manual
structural elucidation were used for identification of unexpected or unknown compounds. Quantification of some other
environmentally relevant compounds, for which the standards were missing, was carried out considering the response
factors being equal to 1.

Preliminary data (results)

In addition to the classic priority pollutants listed by the US EPA (polycyclic aromatic hydrocarbons, phenols, phthalates,
etc.), non-target analysis allowed identifying a wide range of organic compounds, including alkyl pyridines,
organophosphates, furans, organic acids, etc. 3 different data processing approaches were used for the obtained data
set. Chemometric data processing methods were used to identify the differences in the nature of pollutants between the
snow sampling sites. The novel ChromaTOF Tile program provided by LECO Company, free access Metaboanalyst
software and custom designed software were used for sample clusterization. ChromaTOF Tile program automatically
determines the differences in two-dimensional spectra between samples and uses the found features to plot PCA graph
and hence perform clusterization.Metaboanalyst is well known free of charge internet resource for data processing which
allows using different chemometric approaches, while the custom software was used to have the ability of varying the
clusterization parameters in a more detailed manner and determine the number of clusters. Application of chemometric
methods of data processing made it possible to identify the most significant pollutants at each sampling point and to draw
up a pollution map of the Arkhangelsk city.

Please explain why your abstract is innovative for mass spectrometry?

GC-HRMS revealed identification of unexpected classes of organic pollutants, while mass spectrometry based data
processing approaches found the most relevant compounds.

Co-authors:

Anna Sosnova, Lomonosov Moscow State University, Chemistry Department

Thomas Latkin, Lomonosov Northern (Arctic) Federal University, Core Facility Center "Arktika"
Dmitriy Koluntaev, Scietegra
Albert Lebedev, Lomonosov Moscow State University, Chemistry Department
Viatcheslav Artaev, LECO Corporation, St Joseph, MI, USA
Kevin Siek, LECO Corporation, St Joseph, MI, USA

397
RAPID CYANOBACTERIA SPECIES IDENTIFICATION WITH HIGH
SENSITIVITY USING NATIVE MASS SPECTROMETRY
Abstract ID: 257

Presenting author: Jaspreet Sound, School of Biosciences, University of Birmingham

Introduction

Algae blooms are ubiquitous, being found in rivers, lakes and oceans worldwide, with their numbers expanding rapidly
due to climate change. Due to their potential to produce toxins, algae blooms can have devastating consequences on
water quality. Algae blooms can kill fish and other aquatic organisms, harm animals and humans, and result in great
economic losses to the recreational and tourism industries. Determining which species of algae are present within the
bloom and whether they produce toxins is critical to mitigate the harmful effects of toxic algae blooms. Here, we show
how a novel method, based on native mass spectrometry, can be used to create photosynthetic ‘fingerprints’ of
cyanobacteria (commonly known as blue-green algae) that enable different species to be rapidly identified.

Methods

To ‘fingerprint’ cyanobacteria, nine different species of cyanobacteria were cultured, and the cells lysed through the
addition of pure H2O. The lysates were then buffer exchanged into 100 mM ammonium acetate (pH 6.8) using a 30kDa
molecular weight cut-off filter and subsequently analysed on a Q-Exactive HF mass spectrometer using a nanoESI
source.

Preliminary data (results)

The photosynthetic complexes within different species of cyanobacteria are structurally similar yet contain unique protein
primary amino acid sequences. We show, using native mass spectrometry, that these photosynthetic complexes can be
easily identified without prior purification due to their high molecular weights and high abundances within cyanobacteria.
Through high-resolution mass spectrometry analysis of these complexes, we show different species of cyanobacteria
can readily be distinguished from one another, even when multiple species are present. Finally, by comparing with UV-
visible spectroscopy methods, we show our native mass spectrometry method is highly sensitive and can be used to
identify cyanobacterial species at levels prior to bloom formation – a factor that is essential in the early detection and
mitigation of toxic algae blooms.

Please explain why your abstract is innovative for mass spectrometry?

We present the first use of native mass spectrometry to provide identification ‘fingerprints’ of cyanobacterial species.

Co-authors:

Anna Peters, School of Biosciences, University of Birmingham

Jeddidiah Bellamy-Carter, School of Biosciences, University of Birmingham
Cecilia Rad-Menendez, Scottish Association for Marine Science, Culture Collection of Algae and Protozoa, Scottish
Marine Institute
Karen MacKechnie, Scottish Association for Marine Science, Culture Collection of Algae and Protozoa, Scottish Marine
Institute
David Green, Scottish Association for Marine Science
Aneika Leney, School of Biosciences, University of Birmingham

398
Native mass spectrometry of cyanobacterial photosynthetic complexes produce identification 'fingerprints'.

399
Session: IM-14 Miniaturization, Lab-on-a-chip, In Situ Applications

Portable mass spectrometry: importance, developments, and future

perspectives
Abstract ID: 1046

Presenting author: Sarfaraz U.A.H. Syed, Next Generation Sensors B.V.

Introduction

Mass spectrometry (MS) is widely regarded as one of the most definitive techniques used to confirm the composition of
sample of interest. Because of its high sensitivity and low identification errors, in the last couple of decades, an
increasing trend has emerged to take the mass spectrometer beyond its laboratory setting i.e., to bring the lab to the
sample, in the form of portable MS and this continues to be a growing area of research and development. This talk will
introduce why miniaturization of mass spectrometry is very important to solve some very pressing real-life problems in
various markets such as agri-food, security, and health. Key developments in the field of portable mass spectrometry will
be presented. We will discuss what challenges lie ahead to bring portable mass spectrometry for a wider market/range of
applications. We will give an overview on how Next Generation Sensors BV is addressing those challenges for certain
applications.

400
FOOD IMPACT ASSESSMENT ON EXHALED BREATH VOLATILE ORGANIC
COMPOUNDS USING A PORTABLE MEMBRANE INLET MASS
SPECTROMETER
Abstract ID: 82

Presenting author: Milena Jakšić, BioSense Institute, University of Novi Sad

Introduction

It is well known that exhaled breath composition is instant reflection of the organism since volatile metabolites tend to
move from the blood into the air in the lungs. Therefore, breatholomics is becoming very popular as а new non-invasive
screening diagnostic tool. This research brings a portable, low-cost screening sensor prototype based on membrane-inlet
mass spectrometry (MIMS) for food impact assessment, by tracking changes of volatile organic compounds (VOCs) in
exhaled breath. Analytes of interest are breath VOCs that are related to the metabolism of the main food constituents.

Methods

150 exhaled breath samples were collected from 50 adult healthy volunteers with informed consents and ethical approval
completed (Approval No. 2021-01-3/70-1). Each participant provided a sample before meal (after 12h fasting period), 60
min and 120 min after the meal. All samples were analyzed using MIMS system for selected VOC mass fragments – m/z
58 for acetone, m/z 42 for n-pentane, m/z 45 for ethanol and m/z 67 for isoprene. Sample introduction was done using
PDMS sheet membrane probe heated at 70◦C. VOC gas standards were prepared by static dilution technique and used
for quantification.

Preliminary data (results)

Significant change (more than 10% increment or decrement) upon meal consumption was observed in ~85% of
participants for acetone and ethanol, ~70% for n-pentane and ~50% for isoprene. Changes in selected analytes’ signals
were compared for 60 min and 120 min after the meal, and greater change was observed 120 min after the meal for all
analytes, except for ethanol. Obtained mean values for acetone, ethanol, n-pentane, and isoprene respectively were: 809
ppb, 488 ppb, 30 ppb and 70 ppb in samples before meal, 875 ppb, 532 ppb, 29 ppb and 68 ppb in samples collected 60
min after the meal and 559 ppb, 474 ppb, 27 ppb and 66 ppb in samples collected 120 min after the meal. According to
the results obtained, it can be said that MIMS system used in this study could be employed in food impact assessment
by analyzing breath VOCs. It has been established that major food impact can be monitored 120 min after the meal for
analyzed mass fragments. Additionally, obtained mean values were ranked in expected order of magnitude and not
surprisingly – the most significant food impact was observed for acetone. Further research will include more participants
with different nutritional and health backgrounds, which will expand the knowledge in this area.

Please explain why your abstract is innovative for mass spectrometry?

Employment of portable MIMS system in food impact assessment offers a great advantage for possible preventive
screening analyses that would provide an alert for nutritional imbalances.

Co-authors:

Andrea Mihajlović, BioSense Institute, University of Novi Sad

Djordje Vujić, BioSense Institute, University of Novi Sad
Boris Brkić, BioSense Institute, University of Novi Sad

401
APTAPAPER – AN APTAMER-FUNCTIONALIZED GLASS FIBER PAPER
PLATFORM FOR RAPID UPCONCENTRATION AND DETECTION OF SMALL
MOLECULES
Abstract ID: 189

Presenting author: Sandra Martínez-Jarquín, Department of Chemistry and Applied Biosciences, ETH Zürich

Introduction

Analysis of trace compounds from complex samples continues to pose challenges for many analytical chemistry
techniques. Routine trace analysis of small molecules from complex environmental samples such as wastewater or
sludge require extraction of low-concentration analytes from their matrix. Paper Spray ionization is a rapid and affordable
method for the analysis of compounds from samples directly from complex matrixes. Chemical modification of the paper
surface is a strategy to overcome interference of the matrix, which leads to poor detection limits. Aptamers are single-
stranded DNA or RNA molecules selected for their ability to bind to a target molecule with high affinity and specificity. In
this work we characterized and tested a paper-based platform ("Aptapaper") for the upconcentration and analysis of
small molecules from complex matrixes.

Methods

As proof of concept, we used two well-characterized aptamers, quinine and serotonin binding aptamers (QBA and SBA).
We used specific aptamer friendly conditions to ensure the correct folding and binding of both aptamers.
We then characterized the QBA aptapaper system using fluorescence microscopy. We used paper spray ionization
coupled with high-resolution mass spectrometry (PS-MS) to detect the target molecules. A washing step after the
molecule's binding was incorporated to lower the limits of detection (LOD) and reduce matrix effects.

Preliminary data (results)

LODs were determined to be 810 pg for quinine and 44.5 ng for serotonin, respectively, from different matrices. Random
absorption was analyte-specific and observed for quinine but for not serotonin. Based on the results obtained we are now
adapting the system to other aptamers of relevance in different areas that require the analysis of compounds present in
low concentrations, as environmental or food safety. The use of glass microfiber paper makes it low-cost, and the
sampling conditions allow non-experienced persons to collect samples from remote areas. In principle, selective binding
and paper spray ionization can later be applied to more portable techniques, or the shelf life of bound molecules can be
investigated to use aptapaper to store sampled molecules, which can then be sent to a facility for MS readout. However,
each aptamer should be tested separately in order to optimize the specific conditions necessary for the aptamer to fold
properly.

Please explain why your abstract is innovative for mass spectrometry?

Platform for upconcentration of trace molecules that can be adapted to several aptamers for specific and sensible
detection of target compounds. The system can be extrapolated for enviromental measurements.

Co-authors:

Alina Begley, Department of Chemistry and Applied Biosciences, ETH Zürich

Yin-Hung Lai, Department of Chemistry and Applied Biosciences, ETH Zürich, Department of Chemical Engineering,
National United University, Institute of Food Safety and Health Risk Assessment, National Yang Ming Chiao Tung
University
Giovanni-Luca Bartolomeo, Department of Chemistry and Applied Biosciences, ETH Zürich
Adam Pruška, Department of Chemistry and Applied Biosciences, ETH Zürich
Christian Rotach, Department of Chemistry and Applied Biosciences, ETH Zürich
Renato Zenobi, Department of Chemistry and Applied Biosciences, ETH Zürich

402
Methodology. A. Aptapaper synthesis. B Target binding. C. PS-MS Analysis.

403
3-DIMENSIONAL TISSUE SAMPLING IN THE NANOLITER-VOXEL SCALE
BY NANOSECOND-IR-LASER ABLATION FOR ANALYSIS OF
PROTEOFORMS WITH TOP-DOWN MASS SPECTROMETRY
Abstract ID: 675

Presenting author: Hartmut Schlüter, University of Hamburg

Introduction

For investigating molecular mechanisms in the organism analysis of the composition of molecules in the tissue should be
preferred compared to cultured cells. Extracting molecules out of tissues however is associated with the risk of artificial
conversion of the original molecules present in intact tissues, because during conventional mechanical extraction
compartments are destroyed thereby releasing enzymes responsible for the conversion of the original molecules. This
problem is especially critical for the original proteoforms which may be proteolyzed and post-translational modifications
cleaved of. In this study we describe a nanosecond-infrared-laser for soft and ultrafast conversion of voxels of tissue in
the nano-liter range into an aerosol as sample preparation step for analysis of proteoforms with top-down mass
spectrometry.

Methods

Frozen tissue is inspected with optical coherence tomography (OCT) for areas of interest and selected parts irradiated
with a nanosecond-infrared-laser (2.91 µm wavelength) and the aerosol condensed by freezing or drying. The
condensate is mixed with solvent systems allowing the enrichment of proteoforms prior to top-down mass spectrometry,
performed with Q-TOF-, orbitrap and MALDI mass spectrometers. Protein compositions obtained by laser ablation
versus mechanical homogenization are analyzed with respect to their yield. In addition, for demonstrating the precision of
the ablation method condensates tissue layers of the colon are analyzed with bottom-up proteomics for looking for tissue
cell specific markers.

Preliminary data (results)

Guided removal of voxels of tissues in the 3-dimensional fashion in the nano-volume scale by IR-laser ablation is
documented with OCT. Quantities and numbers of proteoforms obtained from the tissues with laser ablation are
significantly higher compared to mechanical homogenization. Analysis of the proteoform composition with 2-dimensional
electrophoresis showed a significantly lower number of intact proteoforms and a high degree of degradation products in
the case of mechanical homogenization compared to laser ablation. The condensates of tissue aerosols can be either
directly mixed with MALDI matrices for top-down analysis with MALDI mass spectrometry or purified and fractionated
with subsequent liquid chromatography prior to ESI mass spectrometry of intact proteoforms. By the analysis of different
cellular layers of colon tissue obtained by laser ablation sampling and subsequent bottom-up proteomics is yielding the
cell-specific protein markers, verified by using the knowledgebase “Human Protein Atlas”.

Please explain why your abstract is innovative for mass spectrometry?

OCT-guided 3D-tissue sampling by nanosecond-IR-laser-ablation is significantly improving top-down mass spectrometric

analysis of proteoforms and is giving access to proteoforms closer to their original composition in intact tissue.

Co-authors:

Jan Hahn, University Medical Center Hamburg Eppendorf

Manuela Moritz, University Medical Center Hamburg Eppendorf
Hannah Voß, University Medical Center Hamburg Eppendorf

404
DETECTION OF METABOLIC CHANGES IN HFD-APOE-/- MODEL BY SP6
PEPTIDE USING MRMS
Abstract ID: 24

Presenting author: Matthias Witt, Bruker Daltonics GmbH & Co. KG

Introduction

Metabolic disorders such as hypertension and dyslipidemia are comorbid pathologic conditions often found in
combination and characterized by a deep connection of altered molecular pathways. The employment of natural
compounds in combination to pharmacological treatment can be an attractive option as preventive therapy. To
understand the molecular pathways that are influenced by bioactive compounds, metabolomics has emerged as a
leading approach. The objective of this work was to investigate the effect of the novel spirulina peptide SP6 in a mice
model of atherosclerosis and evaluate the altered molecular pathways by a fast and accurate FIA-MRMS approach.
Results showed evidence of a distinct modulation of key molecular analytes involved in the disease development and
opened the way to further large-scale studies.

Methods

HFD ApoE-/- mice were divided in control (saline) and SP6 treated group. At the end of treatment and after sacrifice (4th
week), blood was rapidly centrifuged at 2200 rpm for 15 min to obtain plasma samples which was extracted with Matyash
method. Analyses were performed in direct infusion nano-electrospray by an automated multi sample chip-based nESI
sample ionization platform TriVersa NanoMate (Advion BioSciences Ltd, Ithaca, NY, USA). Data were acquired on a
Bruker MRMS 7T solariX XR mass spectrometer. Data processing was performed by MetaboScape 2021.

Preliminary data (results)

The FIA-MRMS method used an automated nanoelectrospray direct infusion system. The sample analysis time was less
than 2 min including sample draw, analysis time and tip change. Mass accuracy values were on average 0.10 and 0.22
ppm for the detected polar metabolites and lipids respectively, which reflects the extreme good mass accuracy of the MS
platforms used. MRMS also provided extreme mass resolution and isotopic fine structure (ISF) leading to unambiguous
molecular formula assignment. The coefficient of variation (CV%) relative to peak intensity was between 0.11 and
10.95% for polar extract samples (relative to the m/z range 100-750), and between 0.59 and 11.19% for lipid extract
samples (relative to the m/z range 150-885) indicating a good repeatability for both metabolite classes. Data
preprocessing was based on sample filtering to remove peaks that were not present in at least 80% of a single group.
Statistical PLS-DA score plots for plasma polar and lipids extracts clearly showed that the two groups were clearly
separated. Metabolites that contributed to the clustering and discrimination were extracted based on the variable
importance in projection (VIP), which were generated after PLS-DA processing. Different metabolite classes were found
dysregulated, and in particular: hydroxyl and tricarboxylic-organic acids, amino acids, lysophosphatidylcholines,
sphingomyelins, and other glycerophospholipids. The treatment with SP6 was able to modulate the levels of these key
metabolites which are involved in atherosclerotic plaque progression and development.

Please explain why your abstract is innovative for mass spectrometry?

Ultra fast metabolomic and lipidomic profiling by FIA-MRMS allows rapid detection of metabolic changes useful for
mechanism elucidation.

Co-authors:

Eduardo Sommella, University of Salerno

Fabrizio Merciai, University of Salerno
Albino Carrizzo, University of Salerno
Carmine Vecchione, University of Salerno
Pietro Campiglia, University of Salerno

405
Workflow for the extraction and analysis of metabolites and lipids

406
DEVELOPMENT AND APPLICATION OF A ROBUST, AUTOMATED HDX-MS
SYSTEM THAT CAN CONTINUOUSLY MEASURE EXCHANGE TIMES FROM
MILLISECONDS TO HOURS
Abstract ID: 59

Presenting author: Cristina Lento, York University

Introduction

Traditional HDX-MS techniques allow for the characterization of protein structural transitions in relatively ordered regions
of proteins. Millisecond Time Resolved HDX (TR-HDX) allows for the characterization of conformational transitions in
weakly ordered regions and intrinsically disordered proteins. Traditionally, TR-HDX has been performed at fixed time
points requiring manual pullback and a high degree of user involvement. We have created a continuous pullback system
to perform millisecond-second HDX-MS automatically, thus reducing the amount of user input required and increasing
the amount of reproducible data collected, all in a fraction of the amount of time required for conventional analyses. Two
model proteins were used for proof-of-concept experiments; cytochrome C and human Tau protein.

Methods

All solvents were delivered using HPLC pumps and autosamplers, with a concentric capillary system serving as the
reaction chamber for performing millisecond timescale studies. Flow rates utilized were 8μL/min for the protein and D2O,
and 38uL/min of 10% acetic acid for the quench. Protease XIII crosslinked to agarose beads packed in a microfluidic chip
was used to localize the deuterium uptake on the proteins. A syringe pump anchored to the protein line was programmed
to automatically pullback the inner capillary delivering protein to obtain reaction times ranging from 150 milliseconds to
12 seconds.

Preliminary data (results)

The resulting uptakes were compared to the analogous experiment using fixed timepoints. As expected, short time scale
measurements yield largely ‘flat’ uptake vs. time profiles in structured proteins such as cytochrome C, whereas uptake
kinetics are observed for the intrinsically disordered Tau protein. In both systems, measured uptake was comparable, but
the continuous workflow provided a much more accurate picture of uptake kinetics due to the vastly increased number of
‘timepoints’ in the profile. In addition, removal of user error allows for more consistent and reproducible results for
systems that are extremely sensitive to deuterium uptake as is the case for intrinsically disordered domains and proteins.
Continuous pullback HDX-MS can also potentially become a routine method in high-throughput settings where large data
sets must be collected in a short period of time.

Please explain why your abstract is innovative for mass spectrometry?

Automation of continuous pullback HDX-MS at millisecond-second time scales reduces the amount of user involvement
and increases the amount of data collected.

Co-authors:

Joe Anacleto, York University

Banafsheh Mehrazma, York University
Dineesha Subasingha, York University
Derek J. Wilson, York University

407
Session: LS-17 Proteomics: Post-Translational Modifications and their
Cross-talk

KINASE SIGNALLING CIRCUITS IN HEALTH AND DISEASE

Abstract ID: 799

Presenting author: Pedro Beltrao, ETH Zurich

Introduction

Cells need to constantly adapt to changes in conditions and use post-translational regulation as a fast way to transfer
information from sensors to effectors of cellular responses. Advances in mass-spectrometry allow us to identify post-
translational modification (PTMs) sites on a large scale and to quantify their changes across different conditions.
However, the interpretation of these measured changes remains challenging.

Methods

We have worked on approaches that try to predict the kinase-kinase regulatory network and how to use large scale
phosphoproteomics to infer the activation state of kinases.

Preliminary data (results)

As an example we have applied these approaches to study the changes in kinase signalling across tumour samples or
occurring during SARS-CoV-2 infection. Our work on the tumour samples revealed a disconnect between the mutational
status of the tumour and the predicted activation state of kinases, indicating substantial compensatory mechanisms.
Based on the viral phosphorylation studies we could show how SARS-CoV-2 infection promoted casein kinase II (CK2)
and p38 MAPK activation and the inhibition of cell-cycle kinases. These were linked to production of diverse cytokines,
cell cycle arrest and stimulation of filopodial protrusions.

Please explain why your abstract is innovative for mass spectrometry?

These approaches are starting to give us a less biased understanding of the kinase signalling network and uncovering
the importance of phospho-regulation across multiple aspects of cell biology and disease.

408
UNRAVELING T CELL - TUMOR CELL COMMUNICATION USING HYBRID
QUANTITATIVE MS
Abstract ID: 457

Presenting author: Kelly Stecker, Utrecht University

Introduction

Intercellular communication involves the complex integration of multiple cell-surface signaling nodes undergoing
simultaneous activation. Due to technical limitations, MS-based proteomic analysis of cell communication often employs
a simplified approach where receptors are activated independently, for example by exogenous ligands. This
recapitulation misses dynamic and cumulative responses that uniquely arise from simultaneous activation of cell surface
receptors. Herein we implement a new hybrid quantitative MS approach to barcode different cell types and enable
quantification of signaling dynamics between and within Tcells and tumor cells simultaneously.

When it comes to immune:tumor-cell engagement, capturing inter/intracellular signaling dynamics in a comprehensive

manner is essential. To find new therapeutic opportunities, we must identify mechanisms tumors employ to evade
immune cell clearance. Our method enables such studies using co-culture systems.

Methods

Cytotoxic CD8+ T cells are co-cultured with resistant or sensitive tumor cells using a ‘matched’ system where cells are
genetically engineered to express cognate receptor-ligand pairs that facilitate engagement. Incubation with untransduced
(UT) Tcells are used as a control. SILAC labels are used to barcode the different cell types: tumor cells are labeled
heavy and T cells light. Medium SILAC labels are added to the media during the co-culture experiment to indicate new
protein translation upon cell engagement. Proteome and phosphoproteome analysis is conducted over a 6hr timecourse
and quantification is performed on each SILAC channel separately using LFQ.

Preliminary data (results)

Preliminary data demonstrates our method works well to quantify cell-type specific signal transduction events occurring
within Tcell and tumor cell upon engagement. In our mixed proteome samples, we detect 3,755 proteins and 5,860
phosphorylation sites specific to Tcells, including positive controls such as phospho-activation of Tcell receptors (TCR).
We detect 5,180 proteins 17,752 phosphosites specific to tumors including expected activation of EGFR and STAT
signaling. Finally, detection of newly translated proteins upon Tcell-tumor engagement show production of known
regulators of Tcell cytotoxicity, including granzyme B and IFNγ. Together this data shows we can detangle mixed cell
types to study simultaneous signaling events occurring in interacting cells.

We observed that Tcells respond differently when engaged with resistant vs. sensitive tumors. We found specific
activation of repressive TFs upon interaction with resistant tumors and conversely upregulation of proliferative signaling
upon engagement with sensitive tumors. In contrast, tumor cells respond similarly to both matched and UT Tcells that
lack the TCR for specific recognition. This finding suggests that tumors do not require specific engagement to prime
themselves for defense against Tcell challenge, shedding new light onto mechanisms of tumor immunosurveillance.

Finally, we find that tumor cells resistant to Tcell killing activate distinct signaling pathways upon Tcell challenge,
including regulation of endocytosis and autophagy. By identifying shared mechanisms of resistance across tumor cell
lines, we can expose key regulators mediating tumor resistance. Our co-culture platform and hybrid quantitative MS
approach enables analysis of inter/intracellular crosstalk and is a powerful tool for future studies.

Please explain why your abstract is innovative for mass spectrometry?

We demonstrate a successful quantitative MS workflow that enables deconvolution of mixed cell types to allow for MS
analysis of inter and intracellular signaling dynamics between and within cells.

409
Cell-specific (phospho)proteome analysis of tumor:Tcell co-culture using hybrid-quantitative MS.

Tumors respond similarly to Tcells, regardless of specific TCR engagement.

410
COMBINED METABOLIC AND CHEMICAL (COMETCHEM) LABELING
USING STABLE ISOTOPES TO REVEAL SITE-SPECIFIC HISTONE
ACETYLATION/DEACETYLATION RATES BY LC-MS/MS
Abstract ID: 554

Presenting author: Marcel Kwiatkowski, University of Innsbruck, University of Groningen

Introduction

Histone acetylation is an important, reversible posttranslational modification (PTM) and hallmark of epigenetic regulation.
However, little is known about the dynamics of histone acetylation, due to the lack of analytical methods that can capture
site-specific reaction rates of acetylation and deacetylation. To tackle this issue, we developed a combined metabolic
and chemical labeling (CoMetChem) approach. CoMetChem combines metabolic labeling using [U-13C]-glucose and
chemical acetylation of non-acetylated lysines at the protein level using stable isotope labeled acetic anhydride. Thereby,
chemically equivalent, fully acetylated histone species are generated that only differ in their isotopic composition. These
isotopologue species have same chemical properties, permitting accurate relative quantification by LC-MS and
calculation of site-specific acetylation and deacetylation rates, a critical parameter to monitor reversible PTM dynamics.

Methods

RAW 264.7 cells were cultured in [U-12C]-glucose containing media for 22h, followed by pre-incubation (t=16h)
with/without histone deacetylase inhibitors (HDACi, MS-275, SAHA). Medium was replaced by [U-13C]-glucose
containing medium with/without HDACi. Histones were extracted at different timepoints, chemically acetylated with
13C4,D6-acetic anhydride (AA), followed by tryptic digestion. H3(18-26) isotopologues were analyzed by LC-MS/MS. For
quantification, extracted ion chromatograms (EICs) were generated, followed by natural isotope abundance correction
(PICor). Half-lives, turnover, acetylation rates and deacetylation rates were calculated using R and Python.

Preliminary data (results)

Chemical acetylation with 13C4,D6-AA resulted in a high acetylation yield for histones, allowing quantification of site-
specific acetylation abundances. H3(18-23) species occurred in the following abundance: non-acetylated (52.4%), K23ac
(28.4%), K18acK23ac (14.2%) and K18ac (5.0%).

Metabolic labeling ([U-13C]-glucose) allowed to quantify label incorporation and calculate acetylation half-lives. However,
half-lives of different acetylated species are not directly comparable. An acetylated species whose substrate is 20x more
abundant compared to another one with the same half-life, is turned over 20x times faster. Half-lives also represent a
simplified description of acetylation dynamics, since no distinction can be made between acetylation and deacetylation
rates. However, CoMetChem allows the abundance of all isotopologues to be compared, thus quantifying the acetylation
and deacetylation rates of the different H3(18-26) species.

We demonstrated with CoMetChem that K23 (0.019/h) had a fivefold higher acetylation rate than K18 (0.0036/h) and that
the 2x-acetylated species had faster acetylation rates 1x-acetylated species (K23: 0.054/h, K18: 0.008/h). The
differences in deacetylation rates were less pronounced (K23: 0.059/h, K18: 0.032/h) than in acetylation rates. These
results suggest that acetylation and deacetylation rates differ between lysines within the same peptide and that a pre-
existing acetyllysine promotes acetylation of a neighboring lysine.

Finally, using CoMetChem, we were able to show that MS-275 and SAHA decreased deacetylation rates as expected.
Interestingly, however, we were also able to show that MS-275 increased the acetylation rate of the doubly acetylated
species at both K18 and K23.

Please explain why your abstract is innovative for mass spectrometry?

CoMetChem generates chemically equivalent acetylated histone isotopologues that allow, for the first time, quantification
of site-specific acetylation and deacetylation rates. Thereby, CoMetChem allows a comprehensive assessment of
reversible acetylation dynamics.

Co-authors:

Alienke van Pijkeren, University of Innsbruck, University of Groningen

Jörn Dietze, University of Tromsø
Alejandro Sánchez Brotons, University of Groningen

411
Anna-Sophia Egger, University of Innsbruck
Tim Lijster, University of Groningen
Andrei Barcaru, University of Groningen
Madlen Hotze, University of Innsbruck
Philipp Kobler, University of Innsbruck
Frank J. Dekker, University of Groningen
Peter Horvatovich, University of Groningen
Barbro N. Melgert, University of Groningen
Mathias Ziegler, University of Bergen
Kathrin Thedieck, University of Innsbruck, University of Groningen, University of Oldenburg
Ines Heiland, University of Tromsø
Rainer Bischoff, University of Groningen

412
OPTIMIZATION OF SUSPENSION TRAPPING (S-TRAP) METHODS FOR
PHOSPHOPROTEOMICS
Abstract ID: 386

Presenting author: Fujia Wang, Biomolecular Mass Spectrometry and Proteomics, Center for Biomolecular
Research and Utrecht Institute for Pharmaceutical Sciences, Utrecht University

Introduction

Suspension trapping (S-Trap), as a novel method of sample preparation for proteomics, is increasingly applied in bottom-
up proteomics studies. It is a rapid, efficient, and simple method, combining the advantages of filter aided sample
preparation method and SDS for protein solubilization. In the existing S-Trap protocol, the addition of phosphoric acid
and methanol buffer creates the fine protein suspension that is captured, washed, and digested on filter. Our results
indicate that this addition of large amounts of phosphoric acid has negative consequences for downstream
phosphopeptide enrichment. Currently, the use of S-Trap digestion for phosphopeptide enrichment is suboptimal, this
obstructs the expansion of its application in phosphoproteomics. Here, we evaluate the performance of commercial S-
Trap filter for phosphopeptide enrichment and explore a solution to the problem.

Methods

Hela cells are used as a standard sample. Trifluoroacetic acid (TFA), formic acid, glycolic acid and phosphoric acid (PA)
are chosen for acidifying proteins in the S-Trap method. S-Trap digests are performed following the manufacturer’s
instructions. As reference samples, in-solution digest samples with sodium deoxycholate (SDC) and standard desalting is
used. The phosphopeptides are enriched using Fe(III)-NTA cartridge on AssayMAP Bravo Platform (Agilent
Technologies). All samples are analyzed using an Orbitrap Exploris480 mass spectrometer coupled with an
UltiMate3000 UHPLC system (Thermo Scientific). All raw MS files are searched by MaxQuant software and analyzed
data in Perseus and Excel.

Preliminary data (results)

Our results indicate that exchanging the PA for another acid has no negative influence on the S-Trap proteome. In
comparisons of the proteomic coverage between different groups, the number of MS/MS matches, peptides, proteins,
and unique peptides are similar in all sample groups prepared by S-Trap. The digestion efficiency is also similar, and all
samples showed good quantitative reproducibility. There are no quantitative or qualitative differences in the full proteomic
analysis between samples prepared with different acids.

We demonstrate that using PA in the S-Trap protocol is problematic for downstream phosphorenrichment. In the
phosphopeptides results, PA negatively affects the number of phosphopeptides identified by LC-MS and samples display
poor phosphopeptide enrichment efficiency. Replacing the acidifier with TFA gave better phosphopeptide enrichment and
detection. After optimization, samples prepared by S-trap (TFA as acidifier) have a similar number of phosphopeptides
as SDC in-solution digestion samples. When comparing the coverage of replicates in phosphopeptides and
phosphosites(STY) identification between samples using PA as acidifier with other samples, the samples handled with
PA also have worse performance. FurthermoreIn conclusion, exchanging the PA in the protocol for another acid,
especially TFA, will increases the depth of coverage for phosphoproteomics data with no perceived consequence on the
proteome.

Please explain why your abstract is innovative for mass spectrometry?

The sample prepared by suspension trapping (S-Trap) is problematic for downstream phosphoenrichment and an
optimized method is explored.

Co-authors:

Kelly E. Stecker, Biomolecular Mass Spectrometry and Proteomics, Center for Biomolecular Research and Utrecht
Institute for Pharmaceutical Sciences, Utrecht University
Maarten Altelaar, Biomolecular Mass Spectrometry and Proteomics, Center for Biomolecular Research and Utrecht
Institute for Pharmaceutical Sciences, Utrecht University

413
ISOBARIC LABELING MASS SPECTROMETRY TO MONITOR
UBIQUITINATION DYNAMICS UPON PROTEASOME MODULATION BY
SMALL MOLECULE INHIBITORS
Abstract ID: 476

Presenting author: Jeroen AA Demmers, Erasmus MC

Introduction

Proteins are tagged with the small protein ubiquitin to target them for degradation by the proteasome. Malfunctioning of
the ubiquitin–proteasome system (UPS) leads to proteome imbalance and – thus – to cancer and neurodegenerative
disorders. Since malignant cells have a higher dependency on the UPS compared to normal cells, proteasome inhibitors
such as Bortezomib are used in the clinic for the treatment of multiple myeloma. Using quantitative isobaric labeling
mass spectrometry in combination with ubiquitination enrichment technologies, we can monitor the dynamics of the
ubiquitinome upon proteasome inhibition in great detail. We use this technology to study the modes of action of different
(second generation) proteasome inhibitors.

Methods

A quantitative mass spectrometry approach was based on SILAC and TMT labeling combined with diGly peptide
enrichment. First, specific proteasome modules were selectively inactivated by RNAi knockdown or CRISPR/Cas
knockout or by small molecule inhibitors. Next, the effects of several different proteasome inhibitors on the ubiquitinome
were explored by combining peptide enrichment with isobaric (TMT) labeling over time. Finally, we have explored the
added value of ion mobility mass spectrometry to the analysis of TMT labeled enriched modified peptides.

Preliminary data (results)

First, we developed an improved workflow for the enrichment and detection of diGly peptides that originate from
ubiquitinated proteins upon tryptic digestion. Using a combination of crude peptide fractionation, optimized use of diGly
antibody beads and an efficient peptide fragmentation regime, we were able to routinely identify >23,000 diGly peptides
from a single sample. We then perturbed proteasome activity by selective depletion of subunits or by small molecule
inhibitors. Malfunctioning of the proteasome resulted in a largely affected proteome, characteristic for changes in stress
response, cell cycle regulation, apoptosis and the UPS. The effects were even more pronounced for the ubiquitinome,
which was dramatically remodeled upon proteasome modulation. Although the far majority of proteins became
increasingly ubiquitinated, many proteins showed heterogeneous ubiquitination patterns on different lysine residues. In
addition, we selectively depleted the three proteasome associated deubiquitinases (DUBs) and observed remarkably
differences in proteome and ubiquitinome remodeling, suggesting unique targeting specificities. Next, several small
molecule DUB inhibitors such as Bortezomib and b-AP15 were used to target the proteasome in a specific manner and
we used SILAC quantitation to monitor the deep ubiquitinome. To characterize ubiquitination profiles over time, isobaric
labeling was used in a multiplexed fashion. This strategy allowed us to accurately measure and detect subtle changes of
the ubiquitinome.

Please explain why your abstract is innovative for mass spectrometry?

Multiplexed quantitative proteomics combined with enrichment strategies allows for the monitoring and comparison of
ubiquitination dynamics over time.

Co-authors:

Karel Bezstarosti, Erasmus MC

Lennart Van der Wal, Erasmus MC
Karen A Sap, Erasmus MC, Amsterdam UMC

414
Workflow for isobaric labeling of diGly peptides.

415
UNLOCKING THE ROLE OF TUBULIN POLYGLUTAMYLATION ENZYMES
INVOLVED IN NEURODEGENERATION WITH TOP-DOWN PROTEOMICS
Abstract ID: 267

Presenting author: Megan Gant, Mass Spectrometry for Biology Unit, Université de Paris, Institut Pasteur, CNRS
USR2000 Paris, 75015, France

Introduction

Microtubules (MTs) are composed of α- and β-tubulin dimers and are essential for neuronal structure. Tubulin post-
translational modifications (PTMs) are regulators of MTs, for example polyglutamylation contributes to neuronal
homeostasis and is associated with neurodegeneration. In neurons, enzymes TTLL1 and TTLL7 catalyze the
glutamylation of α- and β-tubulin, but the number and location of added glutamate residues remains unknown. TTLL1 -
/- mice suffer from neurodegeneration and impaired fertility, whilst the phenotype for TTLL7 -/- mice is not yet described.

Understanding the role of TTLL1 and TTLL7 requires the comprehensive characterization of tubulin polyglutamylation
patterns. To achieve this goal, we developed a top-down proteomics (TDP) workflow, which allowed us to characterize,
for the first time, the TTLL1/7 enzymatic activity on tubulins extracted from WT and TTLL-/- mouse brains.

Methods

Tubulin was purified from WT, TTLL1-/-, TTLL7-/- and TTLL1/7-/- mouse brains by polymerization/depolymerization cycles 1.
Tubulins were diluted with 0.1% formic acid in water before top-down analysis by Vanquish Horizon LC with a C4 column
coupled to an Orbitrap Tribrid Eclipse mass spectrometer. Data were collected at Orbitrap resolution 15 or 120K. HCD,
EThcD, CID, ETD and UVPD fragmentation were used. TDP data were analyzed using BioPharma Finder and ProSight
Lite. Tubulins were also digested with trypsin or thermolysin, desalted using C18 Empore stage-tips before analysis by
LC-MS/MS. Data were analyzed using PEAKS.

Preliminary data (results)

To optimize the TDP pipeline, we tested different LC methods to produce the optimal tubulin (50 kDa) chromatographic
separation. We then analyzed WT tubulin and found that we can detect 34 distinct proteoforms from 6 different tubulin
sequences. Identifications were made by matching the observed and theoretical molecular masses from the MS1 data.
Different fragmentation methods were then tested to achieve the best sequence coverage in targeted MS2 experiments
(HCD, ETD, EThcD, UVPD). A large diversity of polyglutamylation could be observed, with up to 6 glutamate residues for
the TBB2A tubulin proteoform family. To further assign the modification sites, we selected the most abundant
polyglutamylated proteoforms and fragmented them in targeted TDP experiments. These experiments confirmed the
tubulin identification and assigned some modification sites. We then analyzed the TTLL1 -/- mutant for which we observed
similar complexity and diversity as for the WT, although some tubulin species are lost. Lack of α-tubulin
polyglutamylation could explain some of the phenotypes observed for TTLL1-/- mice. In contrast, the TTLL7-/- and the
TTLL1/7-/- mutants both showed a drastic decrease in the number of proteoforms observed, with almost no
polyglutamylation. A BUP analysis was also performed to confirm site assignments. Thermolysin, which led to large C-
terminal peptides, where most PTMs are located, gave the most interesting results. Our findings are important for the
characterization of any emergent TTLL7-/- phenotype and allow for the first time the role of TTLL1 and TTLL7 to be
characterized from a molecular standpoint.

Please explain why your abstract is innovative for mass spectrometry?

For the first time, tubulin polyglutamylation by enzymes TTLL1 and TTLL7 has been characterized by top-down
proteomics, revealing insights into the role of polyglutamylation on proteins linked to neurodegeneration.

Co-authors:

Thibault Chaze, Mass Spectrometry for Biology Unit, Université de Paris, Institut Pasteur, CNRS USR2000 Paris, 75015,
France
Maria M. Magiera, Institut Curie, Université PSL, CNRS UMR3348, Orsay, France, Université Paris-Saclay, CNRS
UMR3348, Orsay, France
Mariya Genova, Institut Curie, Université PSL, CNRS UMR3348, Orsay, France, Université Paris-Saclay, CNRS
UMR3348, Orsay, France
Sinda Khanfir, Institut Curie, Université PSL, CNRS UMR3348, Orsay, France, Université Paris-Saclay, CNRS
UMR3348, Orsay, France
Philippe Bastin, Unité de Biologie Cellulaire des Trypanosomes, Institut Pasteur, INSERM, Paris France
Martial Rey, Mass Spectrometry for Biology Unit, Université de Paris, Institut Pasteur, CNRS USR2000 Paris, 75015,

416
France
Mariette Matondo, Mass Spectrometry for Biology Unit, Université de Paris, Institut Pasteur, CNRS USR2000 Paris,
75015, France
Carsten Janke, Institut Curie, Université PSL, CNRS UMR3348, Orsay, France, Université Paris-Saclay, CNRS
UMR3348, Orsay, France
Julia Chamot-Rooke, Mass Spectrometry for Biology Unit, Université de Paris, Institut Pasteur, CNRS USR2000 Paris,
75015, France

Workflow for the characterization of tubulin polyglutamylation froum mouse brain

417
Session: IM-15 High Resolution Mass Spectrometry - Session B

CHARACTERIZATION AND QUANTIFICATION OF LIPID NANOPARTICLE

COMPONENTS AND THEIR DEGRADANTS IN VIVO USING AN LC-HRAM
MS PLATFORM
Abstract ID: 566

Presenting author: Siegrun Mohring, Thermo Fisher Scientific

Introduction

Lipid nanoparticles (LNPs) have emerged across the pharmaceutical industry as promising vehicles to deliver a variety of
therapeutic agents including the mRNAs. A key aspect to design and optimize a LNP formulation is the development of
biodegradable ionizable lipids which improve LNPs clearance and reduce toxicity in vivo while maintaining the structural
features required for lipid potency. In order to rapidly measure the clearance rate of administrated lipid components of
LNP and monitor the degradants/metabolites of these lipid components in vivo with minimum consumption of biological
samples, we developed a HPLC MS-MS/MS method which enables simultaneous targeted quantification of ionizable
lipid/PEG-lipid and unknown lipid metabolite characterization within a single HPLC MS run. The results are reported
here.

Methods

DOTMA, 14:0 PEG 2000, Cholesterol, DSPC & DSPE were used. To mimic the biological matrix samples at different
time points after the LNP administration, the five lipids were spiked in the bovine liver total lipid extract at nine
concentration levels. A Thermo ScientificTM Orbitrap ExplorisTM 120 mass spectrometer coupled with a Thermo
Scientific™ Vanquish™ Horizon HPLC pump was used for all experiments. A full MS experiment was followed by a data
dependent MS/MS experiment first, then followed by a targeted MS/MS experiment for data collection. Data were
processed with small molecular specific software.

Preliminary data (results)

Preliminary Data (206/250 words)

It is challenging to rapidly monitor the clearance rate and biodegradation pathway of the LNP lipid components in vivo
with limited volume of biological matrix (such as tissues, serum, plasma) samples. The high resolving power, accurate
mass measurement and instrument acquisition versatility offered by Orbitrap Exploris 120 MS allowed us to develop a
single LC-MS method for both metabolite profiling and targeted lipid quantification. The developed method was applied to
the analysis of the dilution sample series. The targeted MS/MS quantification approach uses integrated peak areas of
extracted unique fragment ion(s) for quantification and significantly improves LOD/LOQ of targeted components
compared with the full scan MS quantification approach. The LOD was 0.1ng/mL for the targeted DOTMA and 0.5 ng/ml
(0.001 ng on column) for the targeted DMG PEG 2000. Excellent linearity was observed with R2=0.9995 for DOTMA
over the nine concentration levels of the dilution series. The data collected from data dependent MS/MS was used for
unknown metabolites characterization using either Thermo Scientific™ Compound Discoverer™ 3.3 software or Thermo
Scientific™ LipidSearch™ 4.1 software. Very low abundant DOTMA degradants with oxidization in multiple sites were
confidently identified using the predefined metabolite workflow template “MetID w Stats Expected w Fish Scoring”
included in the Compound Discoverer 3.3 software.

Please explain why your abstract is innovative for mass spectrometry?

Simultaneous untargeted metabolite profiling and targeted LNP lipid component quantification in biological matrix
samples with a single LC MS-MS/MS method.

Co-authors:

Reiko Kiyonami, Thermo Fisher Scientific

Ralf Tautenhahn, Thermo Fisher Scientific
Min Du, Thermo Fisher Scientific
Sinae Lee, Thermo Fisher Scientific
Evgenia Verovskaya, Thermo Fisher Scientific
Jason Potter, Thermo Fisher Scientific

418
FINE STRUCTURAL ELUCIDATION OF PHOSPHOLIPIDS WITH PRACTICAL
ELECTRON-BASED FRAGMENTATION ON Q-TOF INSTRUMENTS
Abstract ID: 718

Presenting author: Christian Klein, Agilent Technologies

Introduction

Chemically diverse lipid species are recognized to play a widening role in biological process and disease pathology.
However, it remains challenging to distinguish lipid isomers and unambiguously reveal fine structural differences.
Contemporary mass spectrometry approaches have largely been based on collisional induced dissociation (CID) which
provide only partial information on lipid structure. Alternative techniques such as ozone-induced dissociation MS have
been developed to identify double bond location, among other structural features. In this work, we demonstrate that a
practical electron-based dissociation (ExD) cell, easily retrofitted into an LC/Q-TOF instrument, provided unique
fragmentation spectra that discriminated phosphatidylcholine (PC) lipid class, distinguished PC sn-1/sn-2 regioisomers,
and determined PC double bond location, with sufficient sensitivity to accommodate a chromatographic timescale.

Methods

Phosphatidylcholine (PC) standards were reconstituted and diluted in 1:1 methanol/isopropanol containing 5mM
ammonium formate, 0.2mM ammonium fluoride. Total lipids were extracted from a 10 µL aliquot of NIST SRM 1950
plasma with a modified butanol-methanol single phase extraction procedure. Standards and plasma lipid extracts were
separated with a 16-minute C18 RP-LC method. Syringe-based infusions and LC eluents were analyzed with an Agilent
6546 LC/Q-TOF with an electromagnetostatic ExD cell (e-MSion). Targeted MS/MS and Auto MS/MS spectra (ExD and
CID) were analyzed with Agilent Qualitative Analysis and NIST MS Interpreter software.

Preliminary data (results)

The ExD cell was tuned for Electron Induced Dissociation (EID), and spectra were observed from infusions of the
regioisomer pair PC 18:0/18:1 and PC 18:1/18:0. Each isomer displayed a single unique EID fragment ion from sn-1
cleavage that defined the sn-2 acyl chain, demonstrating the ability of ExD to distinguish PC regioisomers. The typical
headgroup PC/SM fragment ion (m/z 184) typically observed with CID was also observed with EID, but EID spectra
additionally contained C-Type (m/z 224) and O-Type (m/z 226) fragment ions specific to PC but not SM, demonstrating
the ability of ExD to distinguish lipid classes.

EID spectra contained a series of fragment ions largely corresponding to dissociation of the acyl chain constituents.
Comparison of PC 18:1(9)/18:1(9) and PC 18:1(6)/18:1(6) showed similarities but also 2-Da shifts and diminished
fragment intensities that corresponded to the positions of the acyl chain double bonds, demonstrating the ability of ExD
to distinguish PC double bond location.

Standard-flow LC/Q-TOF experiments were conducted to assess whether EID lipid fragmentation patterns could be
observed on a LC timescale. Sufficient sensitivity was obtained for the standards to detect the unique fragmentation
patterns described above. Additionally, an injection equivalent of 0.5 µL plasma was acquired with typical AutoMS/MS
parameters. Preliminary analysis found the endogenous lipid PC 16:0_18:2 (based on CID) could be further annotated
with EID to PC 16:0/18:2(9,12).

Please explain why your abstract is innovative for mass spectrometry?

ExD technology produced fragmentation spectra that could discriminate lipid class, determine sn-1/sn-2 acyl chain
location, and localize double-bond position.

Co-authors:

Hania Khouri, Agilent Technologies

Mark Sartain, Agilent Technologies
Valery Voinov, e-MSion Inc, Department of Biochemistry and Biophysics, Oregon State University
Joseph Meeuwsen, e-MSion Inc
Joseph Beckman, e-MSion Inc, Department of Biochemistry and Biophysics, Oregon State University
Stefan Schaebler, Agilent Technologies

419
EID spectral comparison demonstrates differing double bond location in standards

420
ESI AND MALDI FTICR MS ANALYSIS OF SKIN-RELEVANT LIPIDS AFTER
EXPOSURE TO LONG WAVELENGTH UV RADIATION
Abstract ID: 640

Presenting author: Samuele Zoratto, Institute of Chemical Technologies and Analytics, TU Wien, Vienna,
Austria, Christian Doppler Laboratory (CDL) for Skin Multimodal Imaging of Aging and Senescence –
(SKINMAGINE), Vienna, Austria

Introduction

Our skin is constantly exposed to solar radiation, high oxygen levels, and environmental pollutants. These are accelerant
stress factors for premature skin aging, tissue inflammation, and photocarcinogenesis. Such oxidative stress activates
cutaneous lipoxygenases and nonenzymatic lipid peroxidation. Oxidized lipids can act as danger-associated molecular
patterns (DAMPs) and as members of the senescence-associated secretory phenotype (SASP), the signaling cocktail of
senescent cells. Our study aims to target bioactive oxidized phospholipids of the SASP and expand the list of possible
candidates. We present a systematic investigation of lipids and their chemically-driven oxidation products (oxLipS)
generated in a controlled environment by reactive oxygen species (ROS) after UV exposure. OxLipS are analyzed by
ESI and MALDI FTICR MS. Results are correlated to results from skin-equivalents investigated by MALDI imaging.

Methods

Lipid standard solution (LipS) is comprised of PAPC and the internal, not UV affected reference DPPC and DNPC. Solar-
like conditions were simulated by exposure to 0/80 J/cm2 of UVA radiation. Skin equivalents (3D cell cultures) were
doped with a) LipS; b) LipS pre-exposed to UV (oxLipS); c) LipS and then exposed to UV; d) LipS and Fibroblasts again
followed by UV exposure. A 7T scimaX FTICR MRMS (Bruker) equipped with a dual ESI/MALDI source was used for
analyses. Direct infusion for ESI and skin equivalent sections mounted on ITO slides for MALDI Imaging using 1,5-
Diaminonaphthalene as matrix.

Preliminary data (results)

We aim to investigate and characterize oxLipS generated via chemical pathways to deepen the knowledge for dermal
tissue investigations. Oxidized products of PAPC result from reactions with peroxides, from hydroxylation, consecutive
fatty acid chain cleavage or loss of arachidonic moiety, to name a few. DNPC and DPPC instead, cannot generate
oxLipS, thus serving as internal standard for semi-quantitative information.

ESI FTICR MS analysis provided crucial information on oxLipS generated by ROS formed in solution after UV irradiation.
Accurate, high-resolution MS data allowed identification of newly formed species through database search and tentative
structural assignment for m/z values not found in the database. LipS and oxLipS were investigated by MALDI FTICR MS
to identify ionization effects. m/z values were compared between ESI and MALDI, and dilution series allowed for LoDs
determination.

These findings were translated to skin-equivalent samples by doping 3D cell cultures with LipS and oxLipS of known
concentration and composition to study the effect of collagen on LoDs. This also mimics, to some extent, real-life
conditions. Qualitative, semi-quantitative, and spatial information about oxLipS in tissue was generated by MALDI FTICR
MS imaging. Finally, fibroblasts embedded in collagen matrix together with LipS were also exposed to UV light to assess
the additional effect of cells on lipid oxidation. Chemistry-driven PAPC oxidation will be superimposed by UV-induced
biological effects and results were carefully considered for future research on human skin samples.

The authors thank the Federal Ministry of the Republic of Austria and CHANEL Parfums et Beauté financial support.

Please explain why your abstract is innovative for mass spectrometry?

Systematic investigation of chemically driven lipid oxidation studied in solution and in organotypic skin equivalents.

Co-authors:

Selma Avdic, Institute of Chemical Technologies and Analytics, TU Wien, Vienna, Austria, Christian Doppler Laboratory
(CDL) for Skin Multimodal Imaging of Aging and Senescence – (SKINMAGINE), Vienna, Austria
Christopher Kremslehner, Department of Dermatology, Medical University of Vienna, Vienna, Austria, Christian Doppler
Laboratory (CDL) for Skin Multimodal Imaging of Aging and Senescence – (SKINMAGINE), Vienna, Austria
Michaela Sochorova, Department of Dermatology, Medical University of Vienna, Vienna, Austria, Christian Doppler

421
Laboratory (CDL) for Skin Multimodal Imaging of Aging and Senescence – (SKINMAGINE), Vienna, Austria
Florian Gruber, Department of Dermatology, Medical University of Vienna, Vienna, Austria, Christian Doppler Laboratory
(CDL) for Skin Multimodal Imaging of Aging and Senescence – (SKINMAGINE), Vienna, Austria
Martina Marchetti-Deschmann, Institute of Chemical Technologies and Analytics, TU Wien, Vienna, Austria, Christian
Doppler Laboratory (CDL) for Skin Multimodal Imaging of Aging and Senescence – (SKINMAGINE), Vienna, Austria

422
HIGH-DEPTH MULTIPLEXED DRUG PROFILING WITH A MODIFIED
TRIBRID MASS SPECTROMETER
Abstract ID: 579

Presenting author: Steven R. Shuken, Department of Cell Biology, Harvard Medical School

Introduction

Integrating tandem mass tags (TMT)-based sample multiplexing and chemical proteomics is emerging as a powerful
approach for drug discovery. However, sensitivity and throughput remain limiting factors toward routine full-proteome
interrogation of drug effects. We evaluated a modified Tribrid mass spectrometer system and used it in a multiplexed
assessment of whole-proteome drug effects. This modified system offers several improved hardware and software
features, including an upstream quadrupolar ion manipulation device for better ion duty cycles and faster
FTMSn acquisition rates, and improvements to the real time search method filter. Multidimensional comparisons of these
features showed significant sensitivity improvement for TMT-based proteomics. With this system we assessed effects of
novel inhibitors of peptidyl-prolyl isomerase PIN1 on proteins across the whole proteome, revealing changes at superior
depth.

Methods

TMT11plex yeast standard was obtained from ThermoFisher Scientific. Biological quadruplicate of four human cell lines
(RPE1, U2OS, HEK293T and HCT116) were digested with LysC and trypsin, labeled using TMTpro16 reagents, pooled,
and fractionated on reverse-phase HPLC at high pH. HCT116 cells were treated with PIN1 inhibitors in triplicate at 10 μM
for 24 h, then processed in the same manner. Samples were analyzed on the modified Tribrid mass spectrometer and
the Orbitrap Eclipse spectrometer using both high-resolution MS2 and SPS-MS3 with real-time search. Parameters
including sample amount, injection time, sensitivity, and quantitative accuracy were investigated and compared.

Preliminary data (results)

In both the four-cell-line study and the PIN1 inhibitor study, the modified Tribrid system showed performance
improvement over the Orbitrap Eclipse system across several metrics, including total numbers of proteins identified and
quantified with smaller sample input while maintaining good quantitative accuracy. The modified ion path allowed for
more sensitive data acquisition for both MS2 and SPS-MS3 workflows (faster spectral acquisition rates with more
parallelizable injection time). As a result of the sensitivity gain, we were able to achieve deeper proteome profiling with
smaller sample input and much improved throughput compared to the current Orbitrap Eclipse system.

Please explain why your abstract is innovative for mass spectrometry?

The new hardware and software contained within a modified Tribrid mass spectrometer show multidimensional
improvement with multiplexed quantitative proteomics and thereby enable more efficient characterization of whole-
proteome changes.

Co-authors:

Graeme C. McAlister, ThermoFisher Scientific

William D. Barshop, ThermoFisher Scientific
Jesse D. Canterbury, ThermoFisher Scientific
David Bergen, ThermoFisher Scientific
Jinjing Huang, ThermoFisher Scientific
Romain Huguet, ThermoFisher Scientific
Joao Paulo, Department of Cell Biology, Harvard Medical School
Vlad Zabrouskov, ThermoFisher Scientific
Steven P. Gygi, Department of Cell Biology, Harvard Medical School
Qing Yu, Department of Cell Biology, Harvard Medical School

423
CHARGE STATE SEPARATION MASS SPECTROMETRY ON TOF
PLATFORM FOR TOP-DOWN ANALYSIS
Abstract ID: 561

Presenting author: Pavel Ryumin, Sciex

Introduction

Top-down analysis is becoming an emerging tool for proteomics analysis. One of the main challenges of top-down
analysis is spectral complexity, which often leads to signal convolution. Recently, charge detection mass spectrometry
(CDMS) implemented on FTMS instruments has shown a potential to tackle this problem. It has been previously reported
that in electron multiplier detection systems employed by TOF mass analyzers, the number of generated primary
electrons depends on the charge state of the oncoming ion[ref1]. Therefore, an amplified detection signal generated by
the cascade multiplication of these electrons encode information about the charge state of the registered ion. In this
work, we utilize this property for separation of overlapping ion products in top-down experiment.

Methods

A prototype ZenoTOF 7600 system equipped with a 4-channel detector was used for this work and all data was acquired
with “Zeno” pulsing enabled. We modified the acquisition system to record raw data, which contained pulse amplitude
and arrival time information for each detection event. For concept validation a direct infusion ECD top-down experiment
using carbonic anhydrase ([M+33H]33+) was performed. Acquired data was either summed into a single spectrum
(conventional data) or banded, wherein a range of the detection event intensities appropriate for the charge state of
interest was selected and summed for further analysis.

Preliminary data (results)

In a top-down experiment, the signal is distributed across many reaction channels with relatively low ion flux per each
channel. In this work, we operated the mass analyzer at 1.5 kHz frequency necessary for “Zeno” pulsing with signal
recorded at four independent detector channels. We confirmed that the instrument speed matched our typical top-down
workflow ion flux maintaining individual ion strike regime without the need for acquisition slow down or sample dilution.
Under those conditions, we observed strong separation of overlapping species with a significant difference in charge
state, thus improving peak capacity over conventional acquisition strategies (Figure 1).

In the described experiment, the achieved cleavage coverage was 90% for summed and 93% for banded data. However,
the number of fragments at different charge state supporting the assignment increased 35% for c ions (351 vs 259) and
22% (371 vs 303) for z ions. Larger gains were observed for higher charge fragments (>15+) – 79 vs 32 for c ions and 54
vs 37 for z ions confirming that the approach helps uncovering longer complementary fragments often buried under lower
charged background or product ions.

Further, we tested the existing fragment matching algorithms optimized for non-overlapping isotopic clusters typically
present in peptide experiments. In this experiment we observed a similar ~30% increase in the total number of matched
fragments in banded data compared to the summed data.

[ref1] Journal of Mass Spec. Proc. Volume 131, 24 February 1994, Pages 181-192

Please explain why your abstract is innovative for mass spectrometry?

Charge state separation on a TOF instrument applied to ECD top-down analysis of a protein

Co-authors:

Gordana Ivosev, Sciex

Takashi Baba, Sciex
Nic Bloomfield, Sciex
Wen Jin, Sciex
Anjali Chelur, Sciex

424
16+ C100 fragment conventional acquisition (top), banded acquisition(bottom)

425
ASSESSING KEY ATTRIBUTES OF ADENO-ASSOCIATED VIRAL
PROTEINS USING HPLC-FLD-INTACT ACCURATE MASS ANALYSIS
Abstract ID: 569

Presenting author: Julia Kraegenbring, Thermo Fisher Scientific

Introduction

Recombinant Adeno-associated viral (rAAV) vectors have emerged as the leading gene delivery vehicles for gene
therapy due to their high-efficiency transduction and safety. AAV capsid proteins (VPs) are critical for viral infectivity and
vector potency and their key attributes, such as identity and relative ratio of VPs and their PTMs need to be fully
characterized and monitored during the viral vector development and manufacturing to ensure the safety, quality, and
efficacy of AAV products. To address this analytical need, we developed a HPLC-FLD-HRAM MS method for
simultaneous measurement of relative expression ratios of VP1, VP2 and VP3 using FLD and direct accurate intact mass
measurement of the VPs and their truncated protein forms using HRAM mass spectrometer.

Methods

Multiple AAV serotype samples were used. The AAV samples were buffer exchanged and concentrated into 80%
H₂O/20% acetonitrile containing 5 mM TCEP and 0.1% formic acid. The collected sample was incubated at room
temperature and used for HPLC-FLD-MS analysis. All data were collected on a Thermo Scientific™ Orbitrap Exploris™
MX mass detector coupled with a Thermo Scientific™ Vanquish™ Horizon UHPLC system and a Thermo Scientific™
Vanquish™ Fluorescence Detector F. The MS data were processed with intact mass analysis workflow in the Thermo
Scientific™ BioPharma Finder™ 4.1 software.

Preliminary data (results)

The goal of this work is to develop a rapid and robust HPLC-FLD-HRAM MS method to monitor several key quality
attributes of VPs for helping to optimize the AAV product development and control the quality of the AAV products during
the manufacturing process. The Orbitrap Exploris MX mass detector, a simple to operate system which offers high
resolution accurate mass (HRAM) and compliance-ready Chromeleon chromatography data system (CDS) software was
used in the developed method. As initial work, we first analyzed an AAV9 sample in triplicate for evaluating the intact
mass measurement accuracy and reproducibility of the observed relative expression ratios of VP1, VP2 and VP3. The
relative expression ratios were observed as 1/0.9/10 (VP1/VP2/VP3) consistently over the three replicate runs. The
observed intact average mass accuracies for the VP1, VP2 and VP3 were below 10 ppm. Additionally, three low
abundant truncated VP proteoforms (50470 Da, 47192 Da and 39475 Da) were detected from the AAV9 sample. To
mimic AAV products with different production methods, we further analyzed two AAV6 samples with the developed
method. The observed intact average masses and the PTMs of VP1, VP2 and VP3 were comparable between the two
AAV6 samples. The truncated VP proteoform (43221 Da) was also detected in both AAV6 samples. However, lower
relative expression ratios of VP1/VP2/VP3 (0.7/0.6/10) was observed from one AAV6 sample compared to another one
which had relative expression ratios of VP1/VP2/VP3 at 1.3/0.9/10.

Please explain why your abstract is innovative for mass spectrometry?

Rapid monitoring of key attributes of adeno-associated viral proteins for AAV product quality control using a HPLC-FLD-
HRAM MS method.

Co-authors:

Reiko Kiyonami, Thermo Fisher Scientific

Roberto Gamez, Thermo Fisher Scientific
Min Du, Thermo Fisher Scientific
Kenneth Thompson, Thermo Fisher Scientific
Chao Yan Liu, Thermo Fisher Scientific
Catharina Crone, Thermo Fisher Scientific

426
Session: LS-18 Metabolomics - Session B

MONITORING GUT MICROBIOTA ACTIVITY BY SECONDARY

ELECTROSPRAY IONIZATION-HIGH RESOLUTION MASS
SPECTROMETRY
Abstract ID: 128

Presenting author: Jiayi Lan, Department of Chemistry and Applied Biosciences, ETH Zurich

Introduction

Hundreds of bacteria species are living in our gut, representing a huge diversity, in terms of genetics and function. An
Easily Accessible Microbiota (EAM1) mouse model, containing 3 bacterial strains (E.coli HS, Bacteroides
thetaiotaomicron, Eubacterium rectale) from the most abundant phyla in the human gut, was developed as a model
system to study host-microbiota interaction. The EAM strains reproducibly colonize germ-free mice, with little variance in
their abundance over time. Although the abundance of the strains was stable, hydrogen generated by the microbiota
varied following a day-night shift, indicating a difference in bacterial activity. To better understand this, we introduce
secondary electrospray ionization (SESI)-mass spectrometry to monitor volatile metabolites produced by the host and
their microbiota in real time.

Methods

In this study, a SESI-Orbitrap mass spectrometer system was employed to profile the real-time volatile metabolome of
either the headspace over a bacterial culture or differently colonized mice. By using sparseSVM classification models, we
selected 133 features and were able to differentiate the bacterial cultures based on the volatiles they released. We also
measured the volatile metabolome of differently colonized mice, including germ-free (GF, zero microbiota), EAM1 mice
(ex-GF mice colonized for 4 weeks with the three EAM1 species) and SPF mice (complete microbiota). Isotopic labeled
D-arabinose was used to targeted feed the gut microbiota to see bacterial activities.

Preliminary data (results)

Compared to GF mice, several pathways were found in mice exhalome to be upregulated in SPF/EAM1 mice, including
fatty acid biosynthesis, metabolism of xenobiotics and amino sugar and nucleotide sugar metabolism. Volatile profiles of
the EAM mice closely clustered with the B. theta profile, this alines with the fact that B. theta has the highest in
vivo abundance. By targeted activating B. theta metabolism using labeled D-arbainose feeding, altered labelling patterns
were observed in various B. theta and E. rectale fermentation products. In vivo cross-feeding between EAM1 species
can thus be inferred.

Please explain why your abstract is innovative for mass spectrometry?

This study provides a volatile metabolites-based approach to non-invasively study microbiota and host metabolism
simultaneously.

Co-authors:

Giorgia Greter, Department of Health Sciences and Technology, ETH Zurich

Markus Arnoldini, Department of Health Sciences and Technology, ETH Zurich
Bettina Streckenbach, Department of Chemistry and Applied Biosciences, ETH Zurich
Renato Zenobi, Department of Chemistry and Applied Biosciences, ETH Zurich
Emma Slack, Department of Health Sciences and Technology, ETH Zurich

427
DEVELOPING AND APPLYING A SEPARATED POOLED QUALITY
CONTROL STRATEGY TO UNTARGETED LC-MS/MS EXPOSOMICS
Abstract ID: 270

Presenting author: Gianfranco Frigerio, Luxembourg Centre for Systems Biomedicine (LCSB), University of
Luxembourg, 6 avenue du Swing, L-4367 Belvaux, Luxembourg, Department of Clinical Sciences and
Community Health, University of Milan, Italy, Occupational Health Unit, Fondazione IRCCS Ca' Granda Ospedale
Maggiore Policlinico, Milan, Italy

Introduction

In metabolomics, untargeted LC-MS/MS methods are conducted to maximise the coverage of annotated metabolites.
Pooled quality controls (QCs), usually prepared by mixing an aliquot from each sample, are useful to improve the quality
of the dataset: the data matrix can be filtered by removing features with a low detection rate in pooled QCs, or those with
a high relative standard deviation across QCs. However, when applied to exposomics studies conducted on groups of
exposed and non-exposed subjects, low-concentrated compounds present only in exposed subjects can be diluted, such
that they are not detected in the total pooled QCs and thus filtered from the dataset. The aim of this work was to propose
a different strategy to prepare pooled QCs in exposomics.

Methods

Urine samples were collected from 38 non-smoking subjects and 22 smoking subjects. Three different pooled QCs were
prepared: mixing an aliquot from every sample (QC-T), from non-smoker samples (QC-NS) and from smoker samples
(QC-S). Samples and QCs were diluted with an acetonitrile:methanol solution, centrifuged, injected, and analysed in four
different chromatographic runs (RPLC and HILIC, both with negative and positive polarity). For each run, the feature
table was obtained using XCMS, then filtered using QC-T (T-feature list), QC-S, and QC-NS separately. The last two
feature lists were merged (SNS-feature list). 152 analytical standards were analysed to confidently identify compounds.

Preliminary data (results)

For all the four chromatographic runs (RPLC-NEG, RPLC-POS, HILIC-NEG, and HILIC-POS) a higher number of
features were kept in the SNS-feature list compared with the T-feature list. A t-test identified significantly different
features between smokers and non-smokers. For the RPLC-NEG run, a total of 6 significantly different compounds
between smokers and non-smokers were confidentially identified, including one (N-acetyl-S-(hydroxypropyl)cysteine)
detected only when considering the SNS-feature list. For RPLC-POS, 6 significantly different compounds were identified,
among which 5 (nicotine, N-acetyl-S-(3,4-dihydroxybutyl)-L-cysteine, N-acetyl-S-(3-hydroxy-1-methylpropyl)-L-cysteine,
phenylglyoxylic acid, and methyluric acid) were detected only with the SNS-feature list. Likewise, in HILIC-POS, 3
significant compounds were identified, among which 2 detected only with the SNS-feature list (3-benzoylpropionic acid
and N-acetyl-S-(3-hydroxy-1-methylpropyl)-L-cysteine). The results obtained showed that filtering the list of features with
dedicated pooled QCs for each group of subjects can prevent the loss of compounds of interest that are clearly relevant
to an understanding of the study context. Besides the compounds identified confidently via the analytical standards
mentioned above, several other compounds were tentatively annotated with varying levels of confidence.

Please explain why your abstract is innovative for mass spectrometry?

Preparing separated pooled QC samples for an untargeted smoking-related exposomics study increases the number of
significant case-study related features retained for further interpretation.

Co-authors:

Camilla Moruzzi, Department of Clinical Sciences and Community Health, University of Milan, Italy
Rosa Mercadante, Department of Clinical Sciences and Community Health, University of Milan, Italy
Emma L. Schymanski, Luxembourg Centre for Systems Biomedicine (LCSB), University of Luxembourg, 6 avenue du
Swing, L-4367 Belvaux, Luxembourg
Silvia Fustinoni, Department of Clinical Sciences and Community Health, University of Milan, Italy, Occupational Health
Unit, Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, Milan, Italy

428
Visual representation of the separated pooled QCs strategy.

Procedure implemented for data elaboration, for each chromatographic run.

429
CHEMICAL PROFILING OF THE HUMAN SKIN SURFACE FOR MALARIA
VECTOR CONTROL USING A NON-INVASIVE SORPTIVE SAMPLER WITH
GC×GC-TOFMS AND UPLC-IMS-HRMS
Abstract ID: 433

Presenting author: Madelien Wooding, University of Pretoria

Introduction

Odour mosquito lures are currently being used as part of integrated vector control strategies in the fight against malaria.
Variation in inter-human attractiveness to mosquitoes, and the preference of mosquitoes to bite certain regions on the
human host, are possible avenues for identifying potential mosquito attractants and repellants. The chemical complexity
of the human skin surface and ethical considerations when sampling humans call for non-intrusive sampling solutions
that enable the detection of a wide range of chemical compounds, whilst not impeding on the dignity of the individual
sampled, and as well as for sophisticated analytical techniques that allow the detection of low concentration chemicals in
a complex matrix.

Methods

A practical non-invasive sampling approach was employed to sample the wrist and ankle skin surface area of human
volunteers. In-house developed polydimethylsiloxane (PDMS) sorptive samplers were fashioned into anklets and
bracelets and placed in direct contact with the skin surface for ease of sampling. Sampling was followed by analyses with
comprehensive gas chromatography – time of flight mass spectrometry (GC×GC-TOFMS) and ultra-performance liquid
chromatography – ion-mobility spectrometry – high-resolution mass spectrometry (UPLC-IMS-HRMS). The sampler was
thermally desorbed directly in the inlet liner of a GC system, while for LC analyses the PDMS sampler was solvent back-
extracted prior to analyses.

Preliminary data (results)

Compounds from a broad range of chemical classes were detected and tentatively or unequivocally identified as
contributing to the differences in human skin surface chemical profiles using GC×GC-TOFMS and UPLC-IMS-HRMS.
Known mosquito semiochemicals were also detected on the human skin surface. In addition to known human skin
compounds, three compounds, (-)-carvone, (E)-2-decenal and caffeine, not previously reported were unequivocally
identified. Limits of detection, using GC×GC-TOFMS, ranged from 1 pg (carvone) to 362 pg (indole). Method limits of
detection ranged from 8.7 ng (sulfadimethoxine) to 95 ng (taurine) for the UPLC-IMS-HRMS method. Furthermore, 77
compounds, of which 64 to the authors’ knowledge have not previously been reported, were detected on the human skin
surface using accurate mass, collision cross section (CCS) values and fragmentation patterns. These complementary
approaches enabled the detection and identification of as yet unknown human skin surface chemicals consisting of
highly-volatile to non-volatile compounds from endogenous and exogenous sources. Furthermore, the study employed a
simplified non-invasive sampling method that can potentially be used in mass screening of the human skin surface
metabolome for not only vector control applications, but also the application of human health screening, such as dietary
studies and detection of disease indicators.

Please explain why your abstract is innovative for mass spectrometry?

This study showed the importance of combining sophisticated analytical equipment, such as UPLC and HRMS with
newer techniques such as IMS, to detect and identify biomarkers in complex biological matrices.

Co-authors:

Yvette Naudè, University of Pretoria

Egmont Rohwer, University of Pretoria

430
STRUCTURAL ANNOTATION OF NOVEL BIOMARKERS FOR INBORN
ERRORS OF METABOLISM WITH INFRARED ION SPECTROSCOPY
Abstract ID: 647

Presenting author: Pieter Kooijman, Radboud University, Institute for Molecules and Materials, FELIX
Laboratory, Nijmegen, the Netherlands

Introduction

Modern metabolic approaches, such as untargeted liquid chromatography mass spectrometry (LC-MS) screening, make
it possible to confidently link low abundant molecular features found in patients to metabolic disease states. Assigning
the molecular structure of these features can often lead to the discovery or refinement of metabolic pathways and
provide novel insights to the underlying biochemistry and pathophysiology.

Standard MS/MS or NMR approaches are often not able to distinguish the subtle structural differences (e.g. isomerism)
between low-abundant isobaric features in a metabolic pathway. Infrared ion spectroscopy (IRIS) fills this gap, as it can
be performed directly on patient samples by coupling to an LC-MS method. The vibrational bands of an infrared
spectrum often contain the information needed to confidently assign the structure of biomarkers.

Methods

Features were detected using untargeted LC-MS in patient materials such as cerebrospinal fluid, blood plasma and
urine. These compounds, often the nM range, were fractioned by heart-cut reversed phase (RP) or hydrophilic interaction
liquid chromatography (HILIC) and analysed by infrared ion spectroscopy (IRIS) in an ion trap mass spectrometer
coupled to the free-electron laser at the FELIX laboratory. Quantum-chemical calculations were used to determine the
optimal wavelength range for IRIS on each target feature and to direct further investigation. Final confirmation was
obtained by comparison to IRIS analysis of synthesized standards.

Preliminary data (results)

Using the method described above multiple previously unknown features have been assigned in patients with inborn
errors of metabolism, such as AICA-ribosiduria, SSADH deficiency and GLUT1 deficiency syndrome. These
identifications have led to new hypotheses on the pathways of the disease states and further discovery is ongoing.

Please explain why your abstract is innovative for mass spectrometry?

We have combined online heart-cutting HILIC and RPLC methods with IRIS methodology to generate IR spectra from
nM range components of patient samples that enable confident assignment of molecular structures.

Co-authors:

Tessa Peters, Radboud University Medical Center, Translational Metabolic Laboratory, Nijmegen, the Netherlands
Karlien Coene, Radboud University Medical Center, Translational Metabolic Laboratory, Nijmegen, the Netherlands
Udo Engelke, Radboud University Medical Center, Translational Metabolic Laboratory, Nijmegen, the Netherlands
Jona Merx, Radboud University, Institute for Molecules and Materials, Synthetic Organic Chemistry, Nijmegen, the
Netherlands
Thomas Boltje, Radboud University, Institute for Molecules and Materials, Synthetic Organic Chemistry, Nijmegen, the
Netherlands
Jos Oomens, Radboud University, Institute for Molecules and Materials, FELIX Laboratory, Nijmegen, the Netherlands,
van’t Hoff Institute for Molecular Sciences, University of Amsterdam, Amsterdam, the Netherlands
Jonathan Martens, Radboud University, Institute for Molecules and Materials, FELIX Laboratory, Nijmegen, the
Netherlands

431
COLLISION INDUCED DISSOCIATION AND ULTRAVIOLET
PHOTODISSOCIATION FOR QUALITATIVE AND QUANTITATIVE LC-MS/MS
ANALYSIS OF LOW MOLECULAR WEIGHT COMPOUNDS
Abstract ID: 558

Presenting author: Romain Giraud, Life Sciences Mass Spectrometry, Department of Inorganic and Analytical
Chemistry, University of Geneva, Geneva, Switzerland

Introduction

Ultraviolet Photodissociation (UVPD) appeared a few decades ago and is now becoming an alternative and/or
complementary fragmentation technique to traditional fragmentation such as collision induced dissociation (CID),
electron transfer dissociation (ETD) or electron capture dissociation (ECD). Selective absorption of the UV radiation for
LC/MS analysis depends on the chromophores of molecules, the wavelength and repetition rate of the laser. UVPD
fragmentation has been applied mostly for the analysis of proteins, peptides, lipids, or sugars. In the present work, UVPD
was compared to CID for structural elucidation of a large set of low molecular weight compounds including
pharmaceuticals and pesticides. Furthermore, the benefit of UVPD for selective LC-MS quantification in biological fluids
is investigated.

Methods

A 266 nm ultra-violet (Nd:YAG, 20 kHz, TeemPhotonics) has been combined with a modified QTRAP 6500+ System
(SCIEX). Software modifications and a controller allow to trigger laser shots. Fragmentation is carried out at the end of
the high pressure collision cell (q2) followed by transfer into the linear ion trap (LIT). Qualitative analysis LC-MS/MS
acquisitions (CID and UVPD) using positive mode electrospray can be performed in a single LC analysis using Data
Dependent Acquisition and for qualitative analysis (Enhanced Product Ion (EPI), MS 3). For quantitative LC-MS/MS
analysis of bosentan and its metabolite in plasma UVPD-EPI was used.

Preliminary data (results)

The performance of UVPD fragmentation was investigated by analysing a set of 200 molecules which contains different
molecule classes including aromatics, alkaloids and derivatives, lipids and lipids like molecules, organic acids and
derivatives and organic oxygen compounds. UVPD fragmentation showed sometimes better fragmentation efficiency
compared to CID with similar sensitivities. Compared to CID the MS/MS spectra recorded in UVPD can be ranked into 3
categories: identical spectra (common fragments), hybrid (unique and common fragments) and different spectra (unique
fragments). In some cases, UVPD fragmentation spectra are more informative than CID which can lead to a better
identification of the molecule considering library search. According to this set of molecules, two stereoisomers,
testosterone and epi-testosterone show variations of fragments intensity sufficiently different to be able to distinguish
them using fragment intensities ratio.

The use of UVPD fragmentation was evaluated for the quantitative LC-MS/MS analysis in plasma of a pharmaceutical
compound (bosentan) and one of its metabolites (desmethyl bosentan) using enhanced product ion mode (EPI). EPI/CID
and MRM/CID has been compared with regards of assay selectivity, sensitivity, precision, accuracy and dynamic range
to EPI/UVPD. Similar performance was obtained for both methods supporting that UPVD is a practical approach for
bioanalysis but offering different selectivity tuning possibilities by selecting a different fragment or versus endogenous
interference from the matrix.

Please explain why your abstract is innovative for mass spectrometry?

LC-MS/MS UVPD-CID analysis of 200 low molecules weight compounds and application for quantitative analysis in
plasma on a QqQLIT.

Co-authors:

Mircea Guna, Sciex, Toronto, Ontario, Canada

Yves Le Blanc, Sciex, Toronto, Ontario, Canada
Gérard Hopfgartner, Life Sciences Mass Spectrometry, Department of Inorganic and Analytical Chemistry, University of
Geneva, Geneva, Switzerland

432
SYSTEMATIC COMPARISON OF DIFFERENT DERIVATISATION
REAGENTS FOR DETERMINATION OF MULTIPLE VITAMIN D3
METABOLITES USING LC-MS/MS
Abstract ID: 636

Presenting author: Anastasia Alexandridou, Department of Chemistry, Bioanalytical Chemistry, Humboldt-

Universität zu Berlin, Brook-Taylor-Str. 2, 12489 Berlin, Germany

Introduction

Liquid chromatography/tandem mass spectrometry is firmly established today as the gold standard technique for analysis
of vitamin D. The major vitamin D metabolites are 25-hydroxyvitamin D3 (25(OH)D3), 3α-25(OH)D3 and 3β-25(OH)D3,
1,25-dihydroxyvitamin D3 (1,25(OH)2D3) and 24,25-dihydroxyvitamin D3 (24,25(OH)2D3). Their quantification can be very
challenging, in particular for the low abundant species. Chemical derivatisation can enhance the detection sensitivity by
increasing the ionization efficiency, shifting the mass to higher m/z values with less isobaric noise and providing specific
fragmentation patterns for MS/MS. Moreover, derivatisation can also have a positive effect on the selectivity of LC
separation.

Methods

In this study, we compared the detection sensitivity and LC separation of vitamin D 3, 3β-25(OH)D3, 3α-25(OH)D3,
1,25(OH)2D3 and 24,25(OH)2D3 using four dienophile reagents (PTAD, DMEQ-TAD, Amplifex, 2-nitrosopyridine), three
reagents for hydroxyl functions (dansyl chloride, 2-fluoro-1-methylpyridinium p-toluenesulfonate) and a combination of
both, including PTAD derivatisation and acetylation of the hydroxyl groups using acetic anhydride. We also present
separation of isomers and epimers utilizing C-18 and PFP HPLC columns as well as different compositions of the mobile
phase. The measurements were conducted using Sciex QTRAP 6500+.

Preliminary data (results)

Derivatisations using Cookson-type reagents are well established but often lead to problems during chromatographic
separation. More than a single peak for each vitamin D analyte are present at the chromatogram as a result of
diastereomer (R and S) formation during reaction. This can be problematic when different diastereomer’s peaks coelute.
Moreover, the existence of more than one peaks has a negative impact on the sensitivity of the method. R and S
diastereomers can be present in different ratios for each metabolite and their elution order can be different depending on
the applied stationary and mobile phases.

Derivatisation reagents targeting hydroxyl group can be an alternative choice offering advantages and disadvantages.
Depending on the stereochemical hindrance of each hydroxyl group and the amount of derivatisation reagent used, one
or more hydroxyl groups can be derivatised resulting in 1-3 different precursor ions depending on the structure of each
analyte.

The comparison of the derivatisation reagents showed that isomer separation was improved when hydroxyl derivatisation
reagents were used in combination with dienophiles. Detection sensitivity depended on the analyte’s structure, with
vitamin D metabolites giving better sensitivity for the dienophile reagents, because of the more selective derivatisation.

Please explain why your abstract is innovative for mass spectrometry?

No systematic comparison of multiple derivatising agents for MS analysis of vitamin D metabolites has ever been
presented. This information is important, as limited literature data show contradictory information.

Co-authors:

Pascal Schorr, Department of Chemistry, Bioanalytical Chemistry, Humboldt-Universität zu Berlin, Brook-Taylor-Str. 2,

12489 Berlin, Germany
Dietrich A. Volmer,

433
 Friday 2 September 2022: 11:00 – 13:00
Session: IM-16 Ambient Technologies (and their applications)

LATEST DEVELOPMENTS IN NATIVE AMBIENT MASS SPECTROMETRY

Abstract ID: 1034

Presenting author: Helen J. Cooper, University of Birmingham

Introduction

Work in our laboratory focuses on combining native mass spectrometry with ambient mass spectrometry imaging to
enable simultaneous acquisition of spatial and structural information on intact proteins directly from their physiological
environment. Sampling approaches include liquid extraction surface analysis (LESA) and nanospray desorption
electrospray ionization (nano-DESI). Latest developments with LESA and nano-DESI, and their integration with ion
mobility spectrometry, for the identification, structural characterisation, and imaging of monomeric proteins, protein
assemblies, and protein-ligand complexes, directly from a range of tissue types and pathologies, and living microbial
colonies, will be presented.

Methods

Preliminary data (results)

Please explain why your abstract is innovative for mass spectrometry?

Co-authors:

Oliver J. Hale, University of Birmingham

Emma K. Sisley, University of Birmingham
James W. Hughes, University of Birmingham
Yuying Du, University of Birmingham

434
IMMUNO-ENRICHED MICROSPHERE - MAGNETIC BLADE SPRAY TANDEM
MASS SPECTROMETRY FOR DOMOIC ACID IN MUSSELS
Abstract ID: 36

Presenting author: Ariadni Geballa-Koukoula, Wageningen Food Safety Research, Wageningen University and
Research

Introduction

Paramagnetic microspheres are used in (multiplex) planar array fluorescent immunoassays for easy and fast screening
of residues in food quality and safety testing. However, no structural confirmation of the screened contaminant(s) can be
obtained. Coated blade spray (CBS) is a recently developed ambient ionization technique that permits direct ionization of
the analyte. In combination with triple quadrupole mass spectrometry (MS/MS) this allows confirmation of food
contaminants, but lack of chromatography compromises the specificity of the measurement and the number of
identification points. A simple and rapid immuno-magnetic blade spray (iMBS) method was developed to enhance the
specificity. The applicability of iMBS was demonstrated for domoic acid (DA) detection at the maximum permitted levels
in contaminated shellfish samples.

*Briefly alternative configurations will be discussed.

Methods

Following a simple extraction process for the shellfish screening assay, the same sample extract is used with
paramagnetic microspheres coupled with monoclonal antibodies (mAb) against domoic acid (DA). The microspheres
immunocapture DA from the extract, and following washing, the cleaned microspheres are pipetted onto the metal blade.
Using a magnet, the microspheres are held on the blade, and following pipetting of the optimized dissociation/spray
solution and application of high voltage, DA is dissociated and detected in the MS. Monitoring of two fragments in MRM
mode, ion ratios plus immunocapturing, provide the selectivity needed for the unequivocal confirmation of DA.

Preliminary data (results)

For the blade spray method, the spray voltage, cone voltage, capillary temperature and fragmentation conditions of DA
were optimized. Moreover, several spray solvents were tested to ensure sufficient dissociation of DA from the mAb and
maximum ionization. A solution of 2.5% v/v formic acid in methanol was the optimum one among those tested. The
iMBS-MS/MS method was also validated in-house following the EU 2021/808 regulation, and it was benchmarked
against a commercial lateral flow immunoassay (LFIA) for on-site screening of DA. The applicability of iMBS-MS/MS was
further demonstrated by analyzing incurred mussel samples. Despite the robust ion ratios monitored, the calculated
probability of the selectivity of the transitions was not sufficient. However, the immunocapture preceding the MS analysis
adds to the overall selectivity of the method, which is demonstrated by the rapid differentiation between DA and its
structural analog kainic acid, which cannot be achieved with the LFIA alone. The final experimental protocol includes
incubation of the immuno-enriched paramagnetic microspheres with the mussel extract, and after several washing steps,
deposition on the tip of the blade. After application of the optimum dissociation/ spray solution, ionization and
confirmation is achieved. It should be stressed that the newly developed rapid iMBS-MS/MS method is generic and can
be easily adapted to include other immuno-captured food contaminants provided monoclonal antibodies are available.
Moreover, thanks to its speed of confirmation, the method can bridge the logistics gap between massive future on-site
testing and classical laborious confirmatory testing in official laboratories.

Please explain why your abstract is innovative for mass spectrometry?

The workflow with immuno-enriched paramagnetic microspheres combined with blade spray ionization and MS/MS
identification is the first iMBS example.

Co-authors:

Arjen Gerssen, Wageningen Food Safety Research, Wageningen University and Research
Marco H. Blokland , Wageningen Food Safety Research, Wageningen University and Research
Christopher T. Elliott , School of Biological Sciences, Institute for Global Food Security, Queen's University Belfast
Janusz Pawliszyn , Department of Chemistry, University of Waterloo
Michel W. F. Nielen, Wageningen Food Safety Research, Wageningen University and Research, Laboratory of Organic
Chemistry, Wageningen University

435
INLINE CARTRIDGE EXTRACTION MASS SPECTROMETRY: A SIMPLE
AND RELIABLE TOOL FOR MOLECULAR PROFILING
Abstract ID: 570

Presenting author: Igor Popov, Moscow Institute of Physics and Technology, Dolgoprudny, Russian Federation,
National Medical Research Center for Obstetrics, Gynecology and Perinatology Named After Academician V.I.
Kulakov, Healthcare of Russian Federation, Moscow, Russian Federation

Introduction

The inline cartridge extraction mass spectrometry (ICE-MS) technique couples electrospray ionization with analyte
molecule extraction in a solvent flow. This method can be used for rapid molecular profiling of tissues, which can find its
uses both in surgical practice where a number of biopsy samples needs to be analyzed intra-operatively to determine
tumor margins, and in the research of systemic metabolic changes. The method doesn’t require additional manipulations
or equipment within the operating room, and eliminates cross-contamination risks, meaning that one mass spectrometer
can be utilized by several operating rooms simultaneously, providing a high-throughput, cost-efficient solution.

Methods

The solid sample (i.e. biopsy tissue sample) is placed into a disposable cartridge that consists of a stainless steel tube,
integrated electrospray emitter, and connectors (see Fig.1). The solvent flow is routed through the cartridge to enable
both analyte extraction and electrospray ionization. Electrospray voltage source is connected via the stainless steel tube.
The emitter position relative to the mass spectrometer inlet is controlled using two cameras and a 3D positioning system.
Signal stability and method reproducibility are assessed using principal component analysis (PCA) and a cosine similarity
metric between individual scans.

Preliminary data (results)

ICE-MS has been shown to be capable of determining glioblastoma TCP within minutes of obtaining the biopsy sample,
with a precision similar to that of express histological analysis. The disposable cartridge with an integrated ESI emitter
eliminates sample cross-contamination during measurements. The electrospray emitter is incorporated into the cartridge
by melting polyether ether ketone (PEEK) connector, because such a connection is consistently watertight and it was
demonstrated to not introduce significant contaminant peaks into the resulting mass spectra. It was shown, using PCA
and cosine similarity metric calculations, that the cartridge provides signal stability and reproducibility similar to, or better
than, those achieved using a commercially available, non-disposable emitter.

Please explain why your abstract is innovative for mass spectrometry?

An improved variant of the ambient ICE-MS technique is introduced, eliminating sample cross-contamination and
expediting analysis. It can be easily implemented in any MS laboratory without additional specialized equipment.

Co-authors:

Denis Bormotov, Moscow Institute of Physics and Technology, Dolgoprudny, Russian Federation
Vasily Eliferov, Moscow Institute of Physics and Technology, Dolgoprudny, Russian Federation
Olga Peregudova, Moscow Institute of Physics and Technology, Dolgoprudny, Russian Federation
Denis Zavorotnyuk, Moscow Institute of Physics and Technology, Dolgoprudny, Russian Federation
Konstantin Bocharov, V. L. Talrose Institute for Energy Problems of Chemical Physics, N. N. Semenov Federal Research
Center for Chemical Physics, Russian Academy of Sciences, Moscow, Russian Federation
Stanislav Pekov, Skolkovo Institute of Science and Technology, Skolkovo, Moscow region, Russian Federation
Anatoly Sorokin, Moscow Institute of Physics and Technology, Dolgoprudny, Russian Federation, Department of
Biochemistry and Systems Biology, Institute of Systems, Molecular and Integrative Biology, Faculty of Health and Life
Sciences, University of Liverpool, Liverpool, UK
Eugene Nikolaev, Skolkovo Institute of Science and Technology, Skolkovo, Moscow region, Russian Federation

436
Figure 1. The disposable cartridge with integrated electrospray emitter.

437
NATIVE IN SITU TOP-DOWN IDENTIFICATION AND NATIVE AMBIENT
MASS SPECTROMETRY IMAGING OF PROTEINS AND PROTEIN
COMPLEXES FROM RAT BRAIN
Abstract ID: 143

Presenting author: Emma Sisley, University of Birmingham

Introduction

Nanospray-desorption electrospray ionisation (nano-DESI) is an ambient MS imaging technique which can analyse
native (folded) proteins from biological substrates such as tissue sections. Nano-DESI provides improved spatial
resolution over liquid extraction surface analysis (LESA) which until recently was the only sampling technique capable of
imaging native proteins directly from tissue samples. Here, we present nano-DESI images of native proteins from rat
brain along with in situ top-down identification of proteins of interest including protein-ligand, protein-metal and protein-
protein complexes as well as a membrane protein.

Methods

Nano-DESI MSI was performed using a home-built ion source coupled to an Orbitrap Eclipse mass spectrometer. Thin
tissue sections (10 µm) from fresh frozen rat brain were thaw mounted onto glass slides. Nano-DESI images were
conducted using 200 mM ammonium acetate + 0.125% C8E4 at a spatial resolution of 200 µm. Tandem mass
spectrometry (MSn) experiments to identify proteins were performed using higher-energy collisional dissociation (HCD)
and a range of extraction methods (LESA, nano-DESI, and LESA extractions transferred into borosilicate pulled tips)
depending on the protein. MS3 was performed to identify ligands bound to proteins.

Preliminary data (results)

Nano-DESI images of proteins in the brain ranging from 11.7 to 66.3 kDa were produced with some proteins displaying
distinct spatial distributions. Examples of proteins identified and imaged include the monomer α-synuclein (14.5 kDa), the
homodimer malate dehydrogenase 2 (66.3 kDa) and the homotrimer cytokine macrophage migration inhibitory factor
(37.0 kDa). The higher molecular weight proteins (>30 kDa) were imaged in a separate experiment where the ion optics
and gas pressures had been optimised for higher m/z transmission.

Ligand-bound proteins were also observed in the nano-DESI images. For example, the protein ADP ribosylation factor
(ARF) 3 with GDP bound was identified. MS2 sequence fragments confirmed the protein as ARF3, while MS3 performed
in negative mode was used to confirm the identity of the GDP ligand. ARF1 and GTPase Ran were also observed to
have GDP bound. GTPase Ran was also observed with an Mg 2+ ion bound. In addition, parvalbumin was observed with
two Ca2+ ions bound and carbonic anhydrase was observed with Zn 2+ bound.

Other proteins that have been observed and identified directly from brain tissue include those with higher molecular
weight such as the homodimer creatine kinase B type (88.5 kDa). The membrane protein VAMP2 was also identified
from rat brain. To achieve this a solvent system comprising 200 mM ammonium acetate + 0.5% C8E4 (2x its critical
micelle concentration) was used to perform a LESA extraction which was then transferred into a borosilicate pulled tip.

Please explain why your abstract is innovative for mass spectrometry?

In situ top-down identification and native mass spectrometry imaging of ligand-bound and metal-bound proteins from
tissue using nano-DESI

Co-authors:

Oliver Hale, University of Birmingham

Iain Styles, University of Birmingham
Helen Cooper, University of Birmingham

438
ADVANCES IN SAMPLE ANALYSIS SPEED AND CLINICAL DIAGNOSTICS
USING LIQUID AP-MALDI MS
Abstract ID: 511

Presenting author: Rainer Cramer, University of Reading

Introduction

Liquid atmospheric pressure (AP)-MALDI MS was recently introduced, adding several functionalities to the toolbox of
MALDI MS. For instance, due to the liquid nature of the MALDI sample a highly stable ion beam can be achieved. Apart
from improvements in quantitative data analysis, this feature greatly facilitates further advances in sample analysis speed
and the accuracy of MALDI MS profiling analysis, as shown by clinical biotyping methods. Liquid AP-MALDI MS
employed on typical high-performing ESI instruments can provide low-noise mass spectra with high mass accuracy and
the detection of a vast array of analytes, from metabolites and lipids to multiply charged proteins, all in the same mass
spectrum of a single acquisition. Our latest data on speed and disease classification will be presented.

Methods

A Q-TOF instrument was fitted with a heated stainless-steel inlet tube (https://doi.org/10.1021/acs.analchem.9b05202).
Samples were irradiated with a 337nm/343nm (up to 2000Hz) laser using an extraction potential of ~3kV between
sample and inlet. Liquid support matrices (LSMs) consisting of α-CHCA (5 mg/mL in 50:50 H2O/MeCN, v/v) and
propylene glycol or ethylene glycol (+ 60% v/v) were mixed 1:1 with the sample. To ensure sufficient temporal resolution
of TOF scans, SONAR software (Waters) was modified to store 200 distinct mass spectra per scan while allowing the
quadrupole to operate without scanning and voltage ramping.

Preliminary data (results)

Analysis of individual samples were achieved at speeds of up to 60 samples/second with 10 or more desorption events
per sample. Using Angiotensin Converting Enzyme (ACE) in an exemplar assay both Angiotensin I and N-Hippuryl-His-
Leu and their products Angiotensin II and His-Leu were observed and monitored in the non-enzyme treated samples and
the treated samples, respectively. Ion suppression was observed for different buffers.

Disease classification using liquid AP-MALDI MS is exemplified by the detection of preclinical bovine mastitis from small
amounts of bovine milk samples using a biobank of 12,000 samples from a longitudinal study over six months of sample
collection. From this data, bovine mastitis was detected in preclinical samples (up to 2 days before symptoms) with a
sensitivity of up to 70% and 100% specificity.

For this and the analysis of clinically relevant bacterial cultures longer acquisition times were used (around 1
min/sample). Bacterial analysis showed that both lipid and protein profiles can accurately identify bacteria with the added
benefit of MS/MS analysis of the multiply charged proteins for further improvements in specificity.

Please explain why your abstract is innovative for mass spectrometry?

Highest individual sample analysis speed in MS and new improvements in MALDI MS profiling (biotyping) of disease and
bacterial cultures.

Co-authors:

Cristian Piras, University of Reading, "Magna Græcia University" of Catanzaro

Henriette Krenkel, University of Reading
Sophie Lellman, University of Reading
Jeff Brown, Waters Corporation
Michael Morris, Waters Corporation
Nick Taylor, University of Reading
Barney Jones, University of Reading
Chris Reynolds, University of Reading

439
440
DIRECT REAL-TIME ANALYSIS OF LIVING CELLS USING LASER
DESORPTION-RAPID EVAPORATIVE IONIZATION MASS SPECTROMETRY
(LD-REIMS)
Abstract ID: 560

Presenting author: Stefania Maneta-Stavrakaki, Imperial College London

Introduction

Laser Desorption-Rapid Evaporative Ionization Mass Spectrometry (LD-REIMS) is an ambient ionization technique
utilizing lasers for the generation of a molecule-rich aerosol, which is then analyzed by mass spectrometry. Here, we aim
to expand the application of the method, using a prototype, automated LD-REIMS platform for the high-throughput,
sample preparation-free analysis of living or frozen cells. The cells were analyzed directly from the cell culture well plate,
allowing fast (2 seconds per well) real-time measurements. Human breast and colorectal cancer cell lines were used for
the method development and validation, while the method was also tested in a previously established metabolic
pathway, linking oncogenic PIK3CA to enhanced arachidonic acid metabolism and increased cPLA2 activity.

Methods

Cells were cultured in 96-well microplates until 70-80% of confluency was reached. The cell monolayer attached to the
well was washed with ammonium acetate.Duplicate plates were generated. One of them was flash-frozen and
cryopreserved until analyzed, while the other was analyzed straight after washing the cells.

An optical parametric oscillator laser, incorporated in a prototype LD-REIMS autosampler was used. The 96-well plate
with the cells was placed on a moving two-dimensional stage. Each well was aligned with the laser beam and an aerosol
transfer capillary. The generated aerosol was aspirated and analyzed with a Xevo G2-XS QToF.

Preliminary data (results)

Classification models were generated based on the MS profiles of 5 breast and 5 colorectal cancer cell lines. The models
demonstrated excellent performance with 100% prediction accuracy for living and 99.7% for flash-frozen cells. The
repeatability of the method was assessed using Pearson correlation coefficient and standard deviation between the
measurements. For both living and frozen cells the correlation coefficient was higher than 0.9 and the standard deviation
was below 0.05, indicating high repeatability between the measurements. Average spectra were compared, with living
cells exhibiting richer small metabolite region, including glucose and amino acids, while frozen cells exhibited richer
glycerophospholipid region.

The method was tested in a previously established metabolic pathway, linking oncogenic mutation in PIK3CA to
enhanced arachidonic acid metabolism through increased cPLA2 activity.Two PIK3CA mutant breast cancer cell lines
(MCF10A) were analyzed (living and frozen) with LD-REIMS and compared to wild type. Arachidonic acid was increased
in mutant compared to wild type, while glycerophospholipids containing arachidonate, such as PI(18:0/20:4) were
decreased in the mutant cells, indicating enhanced cPLA2 activity. In cells treated with the cPLA2 inhibitor ASB14780 the
arachidonic acid levels were found to be significantly lower in the mutant cells, as opposed to the wild type, where it
remained unchanged, confirming that the increase in arachidonic acid in the context of PIK3CA mutation is driven by
increased cPLA2 activity. The next steps of this study will focus on time-course experiments to monitor the dynamic
changes of the metabolome and the lipidome in the cells.

Please explain why your abstract is innovative for mass spectrometry?

LD-REIMS is capable of the high-throughput, direct analysis of cells, in their native state, with no sample preparation,
allowing the real-time measurements of cell cultures.

Co-authors:

Aurelien Tripp, The Institute of Cancer Research

Daniel Simon, Imperial College London
Yuchen Xiang, Imperial College London
Efstathios Elia, National Physical Laboratory
Josephine Bunch, National Physical Laboratory
George Poulogiannis, The Institute of Cancer Research
Zoltan Takats, Imperial College London

441
LD-REIMS spectra of frozen and living cancer cells

442
Session: LS-19 Imaging MS - applications in Life Science & Health -
Session B

APPLICATIONS OF HIGH RESOLUTION MASS SPECTROMETRY IMAGING

IN PHARMACEUTICAL RESEARCH AND FOOD SCIENCES
Abstract ID: 800

Presenting author: Andreas Römpp, University of Bayreuth

Introduction

The detection of drug compounds is one of the most popular applications of MS imaging, while the analysis of food
samples is much less established. However, both application areas share a number of requirements and challenges.
They both deal with very complex samples that require high mass resolution / mass accuracy for reliable identification of
target compounds. The requirements for spatial resolution strongly depend on the sample/question at hand and range
from the analysis of cellular features to entire sections of rodent model animals or food plant organs. They also have in
common that preparation of suitable sections can be challenging, to say the least. These aspects, along with data
analysis approaches, will be discussed on a range of examples from ‘Food & Pharma’.

Methods

MS imaging experiments were carried out on different combinations of Orbitrap mass spectrometers (Thermo Fisher
Scientific, Bremen, Germany) and AP-SMALDI ion sources (TransMIT GmbH, Giessen, Germany). All spectra were
acquired with high mass accuracy (< 1 ppm) and high mass resolution (up to R=240,000 @m/z 200). A range of different
matrix compounds, solvent compositions and spray parameters was used depending on analyte and sample properties.
Measurements were acquired with a step size between 5 and 200 µm. Data analysis was largely based on the open
format imzML. MSiReader, SpectralAnalysis and Mirion were used to generate MS images.

Preliminary data (results)

A number of MS imaging studies cover raw food, mainly plants. For example, in Bhandari et al. 2015
(https://doi.org/10.1039/C5AN01065A) we analyzed wheat seeds at 5 µm and detected a wide range of endogenous
compounds. However, little data is available on processed food. In our study “MALDI mass spectrometry imaging: from
constituents in fresh food to ingredients, contaminants and additives in processed food” (Kokesch-Himmelreich, Wittek
et. al., 2022 https://doi.org/10.1016/j.foodchem.2022.132529) we analyzed a range of plant-based and meat-based food.
The analysis of natamycin in cheese and acrylamide in gingerbread constitute the first mass spectrometry imaging
measurements of a food additive and a food contaminant, respectively.

For drug compounds we reported the first combination of high mass accuracy (1 ppm) and high spatial resolution (10
µm) in 2011 (Römpp et al. https://doi.org/10.1007/s00216-011-4990-7). Subsequent work focused on the analysis of anti-
tuberculosis drugs in the framework of the German Center for Infection Research (DZIF). We were able to detect all first
line anti-TB drugs at therapeutical concentrations in mouse lung. In our most recent study “Do anti-TB drugs reach their
target? High resolution MALDI MS imaging provides information on drug penetration into necrotic granulomas.”
(Kokesch-Hilmmelreich, Treu et al. 2022, https://doi.org/10.1021/acs.analchem.1c03462) we analyzed clofazimine and
lipids at 5 µm pixel size allowing a detailed investigation of drug distribution in TB-induced granuloma. Our latest
experiments cover a new anti-TB drug which is currently undergoing phase II clinical trials.

In both application areas we employed a new approach for the analysis of diffusion / transport processes by MS imaging.

Please explain why your abstract is innovative for mass spectrometry?

First detection of contaminant and additive in (processed) food by MSI. Detailed analysis of anti-TB drug in necrotic
granuloma. New data analysis approach to investigate diffusion/transport processes by MSI.

443
Anti-tuberculosis drug clofazimine in necrotic TB-granuloma (10.1021/acs.analchem.1c03462).

Food additive natamycin in cheese (https://doi.org/10.1016/j.foodchem.2022.132529).

444
TOWARDS A NEW ERA IN SPATIALOMICS: M3-IMAGING OF INTACT
PROTEINS AND OTHER BIOMOLECULES IN TISSUES USING MASS
SPECTROMETRY
Abstract ID: 313

Presenting author: Mark Lim, AmberGen Incorporated

Introduction

While the advent of mass spectrometric imaging (MSI) has extended the “omic” capabilities of mass spectrometry to the
spatial dimension, it is generally limited to untargeted analysis of small molecules and proteolytic protein fragments. We
have developed a novel spatialomic approach based on photocleavable mass-tag (PC-MT) labeled probes, such as
antibodies, for MSI of targeted macromolecules in tissues and workflows which enable highly multiplex, multiomic and
multimodal imaging (referred to as M3-Imaging) [Yagnik, Liu et al. (2021) J Am Soc Mass Spectrom 32: 977-988].
Overall, M3-Imaging holds significant promise for use in the fields of tissue pathology, tissue diagnostics, drug discovery,
and precision medicine as well as in research aimed at understanding the mechanisms of human disease.

Methods

Figure 1 depicts the overall M3-Imaging workflow. Untargeted MALDI MSI of endogenous small molecules is first
performed on a fresh frozen tissue section. The matrix compound used for MSI is then washed away and the tissue
simultaneously fixed, followed by pathology H&E staining and brightfield microscopy, staining with PC-MT probes such
as antibodies (including in some cases dual-labeled fluorescent-PC-MT-probes), and finally fluorescence imaging as well
as highly multiplex MALDI MSI of targeted macromolecules such as intact proteins.

Preliminary data (results)

Highly multiplex, multiomic and multimodal MSI has been achieved on brain, tonsil and cancer tissues, reflecting the
known molecular composition, anatomy and pathology of the targeted biomarkers. Multiplex: The PC-MT probes
coupled with MSI significantly exceed the multiplex capabilities of conventional immunohistochemistry and previous
cleavable mass-tag based methods for tissue imaging. Multiomic: In addition to compatibility with different PC-MT probe
classes including antibodies (proteomics), lectins (glycomics) and nucleic acids (transcriptomics), when these probes are
combined with label-free MSI of untargeted small biomolecules (e.g., lipids, metabolites and drugs), it provides an
integrated workflow and systems biology- based approach to study on the same tissue section the spatial distribution and
interaction of a wide range of biomolecules. Multimodal: Compatibility with conventional H&E pathology staining and
dual-labeled fluorescent-PC-MT-probes extend this approach to additional imaging modes such as brightfield and
fluorescence microscopy on the same tissue section.

Figure 2 shows example results: 19-plex M3-Imaging has been applied to a pathologist certified breast cancer FFPE
tissue specimen using 19 targeted PC-MT antibody probes. The following image panels are all from the same tissue
section: (a.) MSI on a Bruker rapifleX of the PC-MTs with six representative PC-MTs colorized in the image according to
the key provided. (b.) Two of the PC-MT probes (targeting vimentin and αSMA) were dual-labeled with different
fluorophores and a 2-color fluorescence image of the whole tissue section is shown (5 µm resolution). (c.) MSI (20 µm
resolution) of the two dual-labeled PC-MT antibodies is shown for comparison.

Please explain why your abstract is innovative for mass spectrometry?

Highly multiplex, multiomic and multimodal mass spectrometry tissue imaging using probes such as antibodies, nucleic
acids and lectins directly labeled with novel, NHS-activated, photocleavable peptide mass-tags.

Co-authors:

Gargey Yagnik, AmberGen Incorporated

Ziying Liu, AmberGen Incorporated
Kenneth Rothschild, AmberGen Incorporated, Molecular Biophysics Laboratory, Department of Physics and Photonics
Center, Boston University

445
M3-Imaging Workflow.

Example 19-Plex M3-Imaging on FFPE Breast Cancer Tissue.

446
DESI MASS SPECTROMETRY IMAGING OF LIPIDS IN ADVANCED HUMAN
ATHEROSCLEROTIC PLAQUE
Abstract ID: 393

Presenting author: Nuria Slijkhuis, Erasmus Medical Center

Introduction

Carotid atherosclerosis is an arterial disease driven by inflammation and lipid accumulation in the vessel wall, causing
plaque formation. Plaque rupture and subsequent thrombus formation can result in ischemic stroke. The risk of rupture
depends on plaque composition, which critically impacts biomechanical stability. There is a large local variation in plaque
phenotype, and evolution as well as clinical sequelae are unpredictable, leading to suboptimal patient risk-stratification.
Lipids play a major role in atherosclerosis and are therefore potential biomarker targets. Investigating the spatial
distribution of lipids in plaques can aid in understanding their role in local pathobiology, plaque progression and
subsequent biomarker discovery. We used DESI-MSI to image lipids in 14 plaque samples to elucidate plaque lipid
content and distribution with respect to histologically-determined plaque composition.

Methods

Plaques from 14 patients that underwent a carotid endarterectomy were collected. Cross-sectional plaque segments
were embedded in gelatin, cryosectioned into 10µm thick sections and thaw mounted onto glass slides. We performed
imaging experiments on a Waters Synapt XS mass spectrometer with DESI-XS source in positive- and negative
ionization mode with a 100µm pixel size. Additionally, a subset of plaque sections was measured with a 20µm pixel size
in positive ionization mode. Histology was prepared on consecutive sections. An NMF clustering algorithm was
performed to extract the main spectral features from the data to correlate with histologically-determined areas of interest.

Preliminary data (results)

Using DESI-MSI we found a wide range of lipids from different lipid classes. We selected a subset of lipid m/z values that
were present in at least half of the samples. This resulted in 225 m/z in positive ionization mode. This subset of lipids
was used for NMF clustering analysis.

NMF clustering analysis described the data in 8 spectral components, each component consisted of one or two lipid
classes. Some components co-localized with histological features of the plaque specimen: component 2 and 4 consisting
of sphingomyelins and sphingomyelins and cholesterylesters, respectively, co-localized to the calcified areas as shown
by Alizarin-Red staining, see Figure 1. This co-localization could be indicative of the process of sphingomyelin-mediated
arterial calcification, as is described in the literature.

Furthermore, component 3, consisting of phosphatidylcholines, showed co-localization with macrophage-infiltration of

plaques, as shown by CD68 immunohistological staining. Macrophages play an important role in the inflammation
cascade of atherosclerosis and have an increased phospholipid biosynthesis, which might explain the co-localization of
these cells with phosphatidylcholines.

Data analysis is ongoing. Outcomes of this study will be correlated to patient characteristics and to plaque and plasma
lipidomics of the same patients performed on a Sciex Lipidyzer platform to investigate possible lipid markers for high-risk
atherosclerotic plaque.

Please explain why your abstract is innovative for mass spectrometry?

We demonstrate the suitability of the new DESI-XS source for imaging lipids in complex tissues with high sensitivity and
high resolution.

Co-authors:

Mark Towers, Waters Corporation

Mina Mirzaian, Erasmus Medical Center
Ingeborg Nieuwenhuizen, Erasmus Medical Center
Kim van Gaalen, Erasmus Medical Center
Eric Sijbrands, Erasmus Medical Center
Yolanda de Rijke, Erasmus Medical Center
Heleen van Beusekom, Erasmus Medical Center

447
Kim van der Heiden, Erasmus Medical Center
Emmanuelle Claude, Waters Corporation
Gijs van Soest, Erasmus Medical Center

448
QUANTITATIVE MASS SPECTROMETRY IMAGING TO STUDY LIPID
METABOLISM IN PARKINSON’S DISEASE MODEL
Abstract ID: 290

Presenting author: Michiel BR Vandenbosch, Maastricht MultiModal Molecular Imaging (M4I) Institute, Division
of Imaging Mass Spectrometry, Maastricht University, Maastricht, Netherlands

Introduction

Mass spectrometry imaging (MSI) provides insight into the molecular distribution of a broad range of compounds. This
work investigates the suitability of MSI for quantitatively visualizing disease-related alterations in brain lipid concentration
and distribution related to Parkinson’s disease (PD). For this purpose, a new MALDI internal standard (IS) mix was
developed for the simultaneous quantitative MSI of multiple lipid classes. As proof-of-principle, the IS mix was applied for
quantitative detection of glycosphingolipids (GSL) in a transgenic mouse model carrying D409V mutations in the
glucocerebrosidase gene (GBA1) that encodes the enzyme glucocerebrosidase (GCase). Alterations in GSL
concentration and distribution were compared across brain sections from wild-type and heterozygous or homozygous
D409V GBA mutant mice to determine spatial lipid phenotypes across conditions of increasing GBA mutation severity.

Methods

The IS mix consists of ten isotopically labelled or odd-chain analogues of phospholipids and sphingolipids (Avanti Polar
Lipids) with relative concentrations optimised to mimic endogenous signals obtained in MALDI-2 experiments. The IS
was applied using a TM-sprayer (HTX imaging) prior to matrix deposition. MALDI-2 data was acquired on a TIMSTOFflex
(Bruker) using a 266 nm laser (NL204-1K-FH, Ekspla) at a spatial resolution of 30µm. Similar experiments were
performed on biological replicates using a MALDI-2 Orbitrap Elite mass spectrometer (Thermo Scientific) coupled to a
reduced pressure MALDI interface (Spectroglyph, LLC) equipped with a 266 nm laser (Nano-L, Litron).

Preliminary data (results)

Initial investigations have revealed fatty acid chain length-dependent distribution of hexosylceramide (HexCer) species
such as HexCer (d18:1/20:0) and HexCer (d18:1/24:1) across the brain that represent the combined signal of glucose-
and galactose containing. Both HexCer (d18:1/20:0) and HexCer (d18:1/24:1) appear concentrated within white matter
fiber tracts in the midbrain, cerebellum and brain stem but with different relative peak intensities. In addition, the data
showed clear elevation of the HexCer brain levels correlating with GBA mutation status. In heterozygous D409V mice
HexCer (d18:1/24:1) accumulation was highest in the brainstem and reached a maximum of 0.6 µmoles/mm². While
HexCer (d18:1/20:0) accumulation was highest within hindbrain regions and reached a maximum of 1.2 µmoles/mm².

Furthermore, other lipids such as ceramides, Hex2Cer, sphingomyelins (SM), phospatidylcholines (PC) and
phosphatidylethanolamines (PE) could be quantified simultaneously using this approach. For example, SM (d18:1/18:0)
was distributed more homogenously across the brain with concentration ranging from 0–1.4 nmole/mm². PC (16:0/18:1)
concentration ranged from 0-1.3 nmole/mm² with the highest concentrations in the corpus callosum.

These results were used to identify region-specific lipid phenotypes related to deficits in GCase activity. In conclusion,
our presented IS-based Q-MSI approach provides a selective and high-throughput analytical platform to study
quantitative alterations in lipid composition throughout biological tissues.

Please explain why your abstract is innovative for mass spectrometry?

In-depth characterization and simultaneous quantitation of multiple lipid classes in a GBA mutant mouse model using
next-generation mass spectrometry imaging.

Co-authors:

Shadrack M Mutuku, Molecular Horizons and School of Chemistry and Molecular Bioscience, University of Wollongong,
Wollongong, NSW Australia
Nathan G Hatcher, Merck & Co., Inc, Kenilworth, NJ, USA
Kim Ekroos, Lipidomics Consulting, Ltd, Esbo, Finland
Shane R Ellis, Molecular Horizons and School of Chemistry and Molecular Bioscience, University of Wollongong,
Wollongong, NSW Australia, Illawarra Health and Medical Research Institute, Wollongong, NSW, Australia
Ron M.A. Heeren, Maastricht MultiModal Molecular Imaging (M4I) Institute, Division of Imaging Mass Spectrometry,
Maastricht University, Maastricht, Netherlands

449
Quantification of regio-Specific GSL alterations in GCase deficient mice.

Simultaneous quantification of different lipids classes.

450
COMBINED O-GLYCANASE AND O-GLYCOPROTEASE APPROACH FOR
ANALYSIS OF OGLYCOSYLATION IN MUCINOUS TUMOR TISSUES BY
MALDI IMAGING MASS SPECTROMETRY
Abstract ID: 576

Presenting author: Richard Drake, Medical University of South Carolina

Introduction

Applications of imaging mass spectrometry (IMS) to characterize N-glycan distributions in tissues have been effective,
and are possible due to the efficient activity of peptide Nglycosidase, the enzyme that releases the glycans attached to
asparagine on the carrier glycoproteins. Protein glycosylation also occurs via O-glycans attached to serine or threonine.
Compared to N-glycans, O-glycans frequently contain fewer glycans in normal conditions, but can become larger and
more complex in diseases with increased mucin content. Mucinous tumors are more aggressive clinically and difficult to
treat. There is no universal glycosidase specific to all O-glycan modifications, only certain structures. However, new
mucin proteases specific to O-glycosylated proteins have been recently identified. Combinations of these enzymes were
evaluated for IMS of mucinous tumor tissues.

Methods

Formalin-fixed paraffin embedded tissues from human colon and pancreatic mucinous tumors were selected for method
development. Multiple enzyme combinations were evaluated for spraying onto tissues. O-glycanase was from
commercial sources (Genovis and New England Biolabs) and the O-glycoproteinase/mucinase StcE was from the
Malaker lab. PNGase F PRIME and sialidase PRIME (N-Zyme Scientific) were also used for pre-treatment of tissue prior
to O-glycosylation analysis. After spraying CHCA matrix, MALDI IMS was done on a timsTOF fleX MALDI-QTOF mass
spectrometer (Bruker). Data was visualized using SCiLS Lab software (Bruker) and initial structure compositions were
determined by accurate mass.

Preliminary data (results)

Mucinous tumor tissues were selected because of their known high density of Oglycosylated mucin glycoproteins. Using
a colon tissue with extensive mucinous regions, O-glycanase digestion alone resulted in detection by MALDI IMS of
multiple O-glycan species localized to the mucinous tumor regions. Structure compositions of 4-10 sugars consisted of
different combinations of hexose, N-acetylhexosamine (HexNAc) and fucose residues, detected as single sodium
adducts in the 500-2000 m/z. These are consistent with the known substrate specificity of the O-glycanase. Compared to
detection of Nglycans in the same tissue, the relative intensities of the detected O-glycans were significantly lower.
Analysis of the same tissue using the mucinase StcE alone or in combination with PNGase F, yielded multiple peptide
peaks that co-localized to three specific mucinous tumor regions. Over 100 peptide peaks were evident in the mass
range of 1000-4000 m/z. Many of the peptides differed by the mass of a known sugars, suggesting these are
glycopeptides being detected. Both O-glycans and O-glycopeptides co-localized to the same mucinous tumor regions.
PNGase F and sialidase digestions prior to O-glycanase or mucinase are being done to assess whether this improves
access of the enzymes to their O-glycan targets. Complementary analysis of O-glycopeptides extracted from the same
tissues for glycan and peptide identification is ongoing using methods developed in the Malaker lab. Other ionization
methods and ion mobility separations are other options. The combined approaches of differential enzyme digests
targeting O-glycopeptides and O-glycan show promise for evaluating cancers and other diseases with mucinous
features.

Please explain why your abstract is innovative for mass spectrometry?

Combinations of O-glycoprotein mucinases and O-glycanase enzymes can be used to characterize complex mucinous
tumor tissues, providing proof of principle workflows for IMS and tandem MS applications.

Co-authors:

Colin McDowell, Medical University of South Carolina

Grace Grimsley, Medical University of South Carolina
Stacy Malaker, Yale University

451
Mucinase and O-glycanase IMS of colon mucinous tumor (black oval)

452
VISUALIZING THE INTERACTION OF THE TWO FUNGI TRICHODERMA
ATROVIRIDE AND BOTRYTIS CINEREA WITH MALDI IMAGING
Abstract ID: 697

Presenting author: Stefan Kirnbauer, Institute for Chemical Technologies and Analytics, TU Wien (Vienna
University of Technologies)

Introduction

Trichoderma atroviride, a filamentous fungus used as a biocontrol agent, is able to parasite a wide range of aerial and
soilborne plant pathogenic fungi, like Botrytis cinera. Our study aims for a better understanding of underlying
mechanisms allowing T.atroviride to actually sense a pathogen. For this we developed an agar-based micro assay to
study especially the formation of secondary metabolites (SMs) that play a crucial role in fungal sensing and interaction.
Applying MALDI MSI on an 7T FT-ICR instrument we monitor the spatial distribution of known and unknown substances
in an interaction zone (almost) not harboring fungal hyphae. By this we aim to identify highly mobile SMs migrating
towards the interaction partner resulting in a response in the interaction partners T.atroviride and B.cinerea.

Methods

Pre-cultures grown at 25°C in petri dishes were transferred on opposite ends of an agar-medium deposited on glass or
ITO slides. Incubation was performed until growth fronts were 1cm apart (confrontation zone). After quenching (liquid
nitrogen) and vacuum-drying, matrix deposition was done with a home-built sublimation unit or a TM sprayer (HTX).
Extractions of the confrontation zone were done with acidified water/acetonitrile. MALDI MSI of the confrontation zone
and direct-infusion ESI was performed on a 7T scimaX FT-ICR-MRMS equipped with an ESI/MALDI dual source
(Bruker). Data analysis was done with SCiLS and DataAnalysis 5.0 (Bruker).

Preliminary data (results)

A confrontation assay of standardized measures was developed for regular glass or ITO slides. This standardization in
preparation makes parameters like fungal growth highly reproducible. The miniaturization on commercially available
glass slides also allows for samples measurement by MALDI directly after quenching, drying and matrix application using
commercial sample holders.

The distribution of peptaibols and other metabolites formed by T. atroviride as well as potentially novel metabolites
from B.cinerea and T. atroviride were visualized by MALDI MSI in the confrontation zone.

The agar is producing high chemical backgrounds for every MSI experiment. To overcome this limitation of our approach
we tested different matrices (e.g. 2,5-DHB, CHCA, 1,5-DAN, 3-Hydroxycoumarin, 6-Aza-2-Thiothymine and mixtures
thereof) for efficiency and analysis of the analyte signals we can detect. Furthermore, sublimation and spraying for matrix
deposition were studied in more detail for agar samples and experiments for derivatization of functional groups were
conducted to improve the signal to noise ratio of interesting target compounds.

To confirm our findings, the agar from the confrontation zone was extracted with a mixture of 50:50 (v:v) ACN:H2O
acidified with 0.1% Formic acid or 0.1% Acetic acid. Direct infusion ESI FT-ICR MS analyses allowed further evaluation
of ESI versus MALDI ionization efficiencies of metabolites to assess the potential for detecting metabolites in the
presented MALDI MSI setup. Our putative results were then compared with metabolomics outcomes from experiments
using isotopically-labeled standards in the same assay to verify the fungal origin of the metabolites.

Please explain why your abstract is innovative for mass spectrometry?

Combination of Imaging mass spectrometry and supporting techniques to track fungal metabolite production in a
biologically relevant scenario.

Co-authors:
Kristina Missbach, Institute of Bioanalytics and Agro-Metabolomics, Department of Agrobiotechnology (IFA-Tulln),
University of Natural Resources and Life Sciences
Rainer Schuhmacher, Institute of Bioanalytics and Agro-Metabolomics, Department of Agrobiotechnology (IFA-Tulln),
University of Natural Resources and Life Sciences
Susanne Zeilinger-Migsich, Institute of Microbiology, University of Innsbruck
Daniel Flatschacher, Institute of Microbiology, University of Innsbruck
Martina Marchetti-Deschmann, Institute for Chemical Technologies and Analytics, TU Wien (Vienna University of
Technologies)
453
Session: FP-4 Food, Nutrition & Agriculture

USING MASS SPECTROMETRY BASED TECHNIQUES TO DISCOVER

MOLECULAR TARGET FOR FLAVOR INNOVATIONS IN FOOD
Abstract ID: 755

Presenting author: Corinna Dawid, Technical University of Munich, Chair of Food Chemistry and Molecular
Sensory Science

Introduction

The expected continuous population growth, climatic changes as well as changes in dietary patterns faces our
agricultural-food sector a great balancing act in the near future. At the same time, the consumer behavior is changing,
particularly in the industrialized countries. Food is increasingly central to how people identify themselves; they are
looking for healthy, sustainably produced, personalized and performance-increasing nutritional concepts.The
development of healthier food products, for example, reduced in fat or salt, are well-known to induce nonacceptable
flavor and has, thus, created unexpected flavor challenges for the food industry. In response to the consumers’ demand
for healthy but tasty foods, novel ingredient discovery is essential to overcome such flavor challenges.

Methods

The presentation will highlight mass spectrometry based SENSOMETABOLOMICS strategies to identify key taste- and
odor-active compounds and taste modulators in foodstuffs. The knowledge of such chemosensory key molecules then
opens new avenues towards a better understanding of their sensory activity by correlating functional data from human
psychophysics studies and taste receptor activation patterns. Moreover, the population of chemosensory active
molecules (“sensometabolome”) is used as a molecular blueprint for visualizing the changes in sensometabolite profiles
throughout breeding programs, post-harvest manufacturing processes, and/or storage of food.

Preliminary data (results)

Within a current project, the Sensometabolome of a typical milk dessert containing 15 % fat was decoded to serve as
blueprint for processing fat-reduced and flavor-optimized analogs based on functionalized plant-based proteins. With the
help of the Sensomics concept, dairy tastants and odorants were quantitated via UHPLC-MS/MS, based on stable
isotope dilution analyses. Sensory reconstitution experiments of the quantitated compounds also highlighted important
flavor-active analytes. Based on these findings, flavor changes within the functional food systems could be identified
during manufacturing, explained, and consequently minimized by further pre-treatments. In summary, a new way of
reducing fat by implementing functionalized plant-based proteins could successfully be established and led to functional
food products with pleasant flavor profiles.

Please explain why your abstract is innovative for mass spectrometry?

In sum, the presentation will highlighted how a combination of metabolomics and human sensory science helps to
discover flavor innovations in foods.

Co-authors:

Florian Utz, Technical University of Munich, Chair of Food Chemistry and Molecular Sensory Science
Johanna Kreißl, Leibniz-Institute for Food Systems Biology
Andrea Spacasassi, Technical University of Munich, Chair of Food Chemistry and Molecular Sensory Science
Timo D. Stark, Technical University of Munich, Chair of Food Chemistry and Molecular Sensory Science
Christian Schmid, Technical University of Munich, Chair of Food Chemistry and Molecular Sensory Science
Caren Tanger, 3Chair for Food and Bioprocess Engineering, Technical University of Munich
Ulrich Kulozik, 3Chair for Food and Bioprocess Engineering, Technical University of Munich
Thomas Hofmann, Technical University of Munich, Chair of Food Chemistry and Molecular Sensory Science

454
STRUCTURAL ELUCIDATION OF PHOTO-MOLECULAR HEATER
BYPRODUCTS USING INFRARED ION SPECTROSCOPY
Abstract ID: 549

Presenting author: Matthias Vink, Felix Laboratory, Radboud University

Introduction

Agricultural activities in regions with lower temperatures limit plant growth and reduce crop yields. Recently, research
projects have been initiated to investigate novel photo-molecular heating agrochemical compounds that can induce
localized heating to increase crop yields by functionalizing the part of the solar radiation that is not used for
photosynthesis. However, an issue that needs to be addressed for these chemical-based solutions to agricultural
problems is that the degradation of compounds increases chemical complexity. The profile of degradation products must
be characterized to assess their impact on environmental and human safety.

In this study, we have focused on the degradation arising from solar radiation exposure of agrochemicals and aimed to
structurally elucidate the byproducts originating from exposure by employing infrared ion spectroscopy (IRIS).

Methods

We first used LC-MS to identify molecular features only found after simulated solar radiation exposure. LC-MS revealed
the presence of various isomeric degradation products, which challenge traditional LC-MS(/MS) approaches offering
limited molecular structure information. These features were mass-isolated in an ion trap mass spectrometer, and their
IRIS spectra in the 500 - 2000 cm-1 range were recorded using the FELIX free-electron laser. Molecular structures were
assigned by comparing experimental infrared spectra with computationally predicted IR spectra of candidate structures.
Calculations employed the B3LYP density functional with a 6-311++G(d,p) basis set.

Preliminary data (results)

In panel A of Figure 1, the chromatographic profile of sinapoyl malate (SM) before (black trace) and after simulated solar
irradiation (orange trace) is shown. After irradiation, byproducts are generated as indicated by the additional peaks in the
chromatogram. The peak at a retention time of 6.2 minutes has the same high-resolution accurate mass and MS/MS
fragmentation spectrum as the original SM peak at 6.1 minutes, suggesting that the extra feature is a close isomer of
SM. This feature is fractioned and characterized using IRIS, resulting in the spectrum in orange in panel B. A density-
functional theory (DFT) computed IR spectrum for the cis isomer of SM (blue) agrees favorably with the measured
spectrum regarding frequency positions of the vibrational bands, confirming that this product is likely due to trans-to-cis
isomerization.

We applied this workflow to assign various other byproducts generated after simulated solar irradiation, enabling us to
structurally elucidate all major byproducts of SM. Experiments on other photo-molecular heating compounds are in
progress. Additionally, we currently investigate byproducts from photo-molecular heaters sprayed on actual crops as the
next step from the solution experiments.

Please explain why your abstract is innovative for mass spectrometry?

Byproducts of agrochemicals (photo-molecular heaters) were elucidated using a reference-free identification approach
utilizing computationally predicted IR spectra of candidate structures. This workflow could allow for improved
identification of agrochemical byproducts.

Co-authors:

Giel Berden, Felix Laboratory

Jonathan. K. Martens, Felix Laboratory
Jos Oomens, Felix Laboratory, Radboud University

455
Fractionation-IRIS experiment of isomerized SM.

456
FOOD AUTHENTICATION: MASS SPECTROMETRIC STRATEGIES FOR
DETECTING FOOD FRAUD IN ROUTINE LABORATORIES
Abstract ID: 494

Presenting author: Marina Creydt, University of Hamburg, Hamburg School of Food Science

Introduction

Food fraud is a multi-million-dollar business, that has increased significantly in recent years and requires objective
analytical methods. In this context, non-targeted metabolomics strategies using mass spectrometric techniques offer
great potential. Nevertheless, this approach is not yet very widespread in routine analyses.

The example of determining the geographical origin of white asparagus is used to show how metabolomics methods can
also be established in routine laboratories. Asparagus is a particularly good example because the labeled geographical
origin of this food is very often falsified. For this reason, asparagus is usually also analyzed by state food inspection
agencies using isotope ratio mass spectrometry (IRMS). However, this method often proved to be ambiguous, which is
why alternatives are in demand.

Methods

Based on method development with a UHPLC-ESI-QTOF platform, the workflow of a non-targeted metabolomics study
with approximately 400 asparagus samples is described. Since screening approaches with mass spectrometers often do
not provide reproducible results over long time periods, a housekeeping approach with selected metabolites is presented
to ensure normalization. In addition, the transfer of the method to a commercial laboratory is shown, where the non-
targeted approach was converted to a simple targeted method using a robust UHPLC-ESI-QQQ instrument.

Preliminary data (results)

In the development of the UHPLC-ESI-QTOF method, the detection of non-polar compounds proved to be particularly
useful in determining the geographical origin of the asparagus samples. Some samples were also measured using IRMS
to evaluate the performance of the mass spectrometric method compared to the established approach. The newly
developed metabolomics method proved to be at least equivalent and in some cases even better. By using four
metabolites that showed no concentration differences depending on geographical origin (housekeeping metabolites), a
reproducible evaluation could be ensured that can be performed independently of the typical technically induced
fluctuations of a mass spectrometer. In this way, it is no longer necessary to measure all samples in a single batch, but to
split them up. This drawback has often meant that non-targeted MS methods are not yet widely used for routine
applications and that NMR approaches, for example, are preferred due to their robustness.

Despite the subsequent reduction of the non-targeted method from several thousand metabolites to 20 metabolites, a
reliable result was still obtained in terms of geographical origin assignment. By quantifying these metabolites, the
procedure could be transferred to a robust targeted method using an UHPLC-ESI-QQQ instrument. These devices,
unlike high-resolution mass spectrometers, are already available in most routine laboratories, ensuring a widespread
transfer.

Please explain why your abstract is innovative for mass spectrometry?

- Development of a non-targeted metabolomics study to determine the geographical origin of asparagus

- Downstream process of a non-targeted into a targeted method for routine analysis

Co-authors:

Markus Fischer, University of Hamburg, Hamburg School of Food Science

457
Figure 1. Detection of the geographical origin of asparagus.

458
HUMAN EXPOSURE TO EXTRACTABLES AND LEACHABLES FROM
SINGLE-USE PLASTIC FOOD CONTACT MATERIALS: A DISPOSABLE
SORPTIVE SAMPLER, UHPLC-IMS-HRMS AND GC×GC-TOFMS
Abstract ID: 432

Presenting author: Madelien Wooding, University of Pretoria

Introduction

280 million tonnes of plastic are produced annually around the globe, of which half is single-use items such as straws,
polyethylene terephthalate (PET) bottles, polypropylene (PP) food packaging and polystyrene (PS) take-away
containers. Plasticisers are added to plastic food contact materials (FCMs) to promote flexibility but have been found to
leach into food and drink media. The World Health Organisation confirms that trace levels of plasticisers are sufficient to
compromise human health, especially in the case of repetitive exposure through various exposure routes. Five million
tonnes of plasticisers are incorporated into plastic annually, which is alarming considering certain plasticisers display
endocrine-disrupting behaviour. This study provides a simple detection and quantification method for extractable and
leachable (EL) compounds that emerge from FCMs.

Methods

Food contact materials were exposed to ambient conditions, elevated temperatures, UV/Vis radiation and microwaving to
simulate various scenarios, such as leaving PET-bottled beverages in a car or in the sun, microwaving take-away food in
PS or PP packaging, drinking hot beverages out of PS cups or drinking cooldrink out of PP cups. Sampling was followed
by analyses with comprehensive gas chromatography – time of flight mass spectrometry (GC×GC-TOFMS) and ultra-
performance liquid chromatography – ion-mobility spectrometry – high-resolution mass spectrometry (UPLC-IMS-
HRMS). These complementary techniques enabled the detection of a wide volatility range of ELs at trace levels.

Preliminary data (results)

Detection and quantification of ELs migrating out of plastic FCMs using an in-house developed PDMS loop for sorptive
extraction are reported for the first time. Diethyl phthalate, a known endocrine-disrupting chemical (EDC), was detected
in levels up to 167 µg per single-use food contact material. Other EDCs detected included several phthalates and
plasticisers, as well as BPA. Toxic compounds released upon microwaving plastic FCMs included styrene, styrene oxide,
benzonitrile and hydrogen azide. Toxic hydrocarbons sorbed onto plastic FCMs during vehicle transportation included
nonadecane and pentadecane. The high trapping power of an in-house developed PDMS sorptive sampler enabled the
extraction of a diversity of compound classes from the FCMs analysed. PDMS loops were desorbed directly in the GC
inlet to optimise analysis time, prevent sensitivity loss and avoid the use of expensive cryogenics. LODs and LOQs for
LC analysis ranged from 0.014 μg/L (Octabenzone) to 7.4 μg/L (Irganox 1076) and 0.042 μg/L (Octabenzone) to 22 μg/L
(Irganox 1076), respectively. Extractable and leachable target compounds were detected in water stored in single-use
FCMs in concentrations ranging from 0.019 to 25 μg/L.

Please explain why your abstract is innovative for mass spectrometry?

These complementary approaches enabled the detection of harmful extractable and leachables at trace levels. Accurate
mass and CCS values enabled the tentative identification of non-targets present in food contact materials.

Co-authors:

Gizelle van Niekerk, University of Pretoria

Yvette Naudè, University of Pretoria

459
VALIDATION OF IMMUNOAFFINITY-BASED MASS SPECTROMETRY
APPROACHES FOR THE DETECTION OF RUMINANT-SPECIFIC PEPTIDES
IN ANIMAL FEED
Abstract ID: 381

Presenting author: Tessa Höper, Department of Food Safety, National Reference Laboratory for Animal Protein
in Food, German Federal Institute for Risk Assessment (BfR)

Introduction

To prevent transmission of bovine spongiform encephalopathy strict feeding restrictions for animal by-products in
livestock breeding intended for human consumption were enacted by the European Commission (regulation (EG)
999/2001). Increasing reauthorizations have arisen analytical challenges for governmental regulatory agencies, as
previously established PCR- and microscopy-based systems are insufficient to discriminate authorized (e.g. milk) and
non-authorized materials (e.g. meat and bone meals) in a straightforward manner. We thus previously developed two
analytical methods for the detection of illegal protein additives using antibody-based immunoaffinity enrichment coupled
with mass spectrometry for the detection of ruminant-specific proteins by surrogate peptides in animal feed. These
methods now underwent validation in accordance with the FDA recommendation - Bioanalytical Method Validation
Guidance for Industry.

Methods

One method was designed for the detection of ruminant hemoglobin using commercially available antibodies and
materials. Hemoglobin was enriched on the protein level using magnetic beads and a manual magnetic separator.
Tryptic peptides were analyzed using an LC-MS/MS method. The method is easily transferrable to other labs.

The second method represents a sophisticated two-tier multiplex assay for the detection of tissue-specific ruminant
proteins. Target analytes were enriched on the peptide level after tryptic digestion using customized antibodies and an
automated magnetic separator. Analysis was performed using LC-MS/MS.

Preliminary data (results)

For method validation, five blood products and nine meat and bone meals from ruminant origin were purchased or kindly
provided by the manufacturers. The animal protein powders were mixed into vegetal feed and ground to a fine powder.
Both methods then comprise tryptic digestion, magnetic bead-based immunoaffinity enrichment of the protein surrogate
peptides and isotope-labelled peptide analogues, as well as analysis by nanoHPLC-MS/MS (parallel reaction
monitoring). Ruminant- and tissue-specific peptides were quantified using internal standards and an external calibration
curve. The following proteins allow discrimination of tissues: myosin-7 and matrilin-1 (meat and cartilage), complement
component 9 and α-2-macroglobulin (plasma proteins), hemoglobin subunit α (blood and meat meal) and osteopontin
(bone meals and milk). Accuracy, linearity, parallelism, precision, recovery and sensitivity of the method were determined
and analyte stability was determined for multiple storage conditions. Twelve animal feeds ranging from vegetal cattle
feed to high protein aquaculture feed served as matrices to demonstrate that a wide variety of feeds neither produce
false positive signals nor interfere with the detection of the surrogate peptides. To be able to exclude suppression of the
analytes by other proteins present in the feed matrix, interference of the analytical method with milk powder additives up
to 60 % (w/w) was assessed. Validation results for the analytical methods shown here demonstrate that adulterants with
ruminant protein as low as 0.1% (w/w) in the feed matrix can be qualitatively and 1% (w/w) can be quantitatively detected
in laboratory routine operation.

Please explain why your abstract is innovative for mass spectrometry?

Validation of two analytical methods using antibody-based immunoaffinity enrichment coupled with mass spectrometry
for the detection of ruminant-specific proteins by surrogate peptides in animal feed to reveal illegal protein additives.

Co-authors:

Andreas Steinhilber, Signatope GmbH

Hannes Planatscher, Signatope GmbH
Cornelia Sommersdorf, Signatope GmbH
Oliver Pötz, Signatope GmbH, NMI Natural and Medical Sciences Institute at the University of Tuebingen
Uta Herfurth, Department of Food Safety, National Reference Laboratory for Animal Protein in Food, German Federal
Institute for Risk Assessment (BfR)

460
CONSIDERATIONS FOR DEVELOPING AN ANALYTICAL STRATEGY FOR
SMALL MOLECULE MS-BASED SCREENING IN INDUSTRIAL
BIOTECHNOLOGY
Abstract ID: 436

Presenting author: Leon Coulier, Center of Analytical Innovation, Biodata & Translation, DSM Science and
Innovation

Introduction

DSM is a company active in Health, Nutrition and Bioscience. In our bioprocess and bioproducts R&D, mass
spectrometry is one of the core technologies used in our labs for small molecule analysis to generate insight for microbial
strain development and screening.

In this presentation we will show an overview of different technologies investigated for fast MS analysis for various
classes of small molecules. An extra challenge in this investigation is the fact that the targeted metabolites can be of very
different chemical composition for different innovations and the same is true for the microbial matrix of interest. As a
result, the analytical technologies should be widely applicable for various classes of compounds

Methods

We will show examples for amino acids, organic acids and peptides of fast LC-MS analysis, application of dual
LC/multiplexing LC-MS and flow injection analysis (FIA)-MS. In addition, we also investigated external technologies like
fast solid-phase extraction (SPE)-MS and acoustic droplet injection (ADE)-MS as high-throughput alternatives.

Preliminary data (results)

Normal LC-MS runs last between 5-10 min. Pushing the limits of UHPLC-MS typical can reduce runtimes to 2-3 minutes
with still sufficient separation. The next step is tackling the inefficient use of the MS due to washing and equilibration
steps by performing LC-MS in multiplexing mode with which the efficient runtime can be reduced by roughly a factor of 2.
Separation however takes time, so leaving the separation out will result in a further significant decrease in runtime. Flow
injection analysis by MS (FIA-MS) is an example of this with typical runtimes of 30-60 sec, but with no separation and
therefore more matrix interference. If even more speed is necessary, more dedicated equipment can be used, like fast
SPE-MS with typical runtimes of 10-20 seconds or ADE-MS with runtime of only 1-3 seconds.

Each of the technologies presented have their specific advantages and the specific bioproduct development
requirements will determine which approach for small molecule analysis is the most appropriate, there is no one-size-fits-
all solution. The subtle balance between speed, separation, quantification, robustness and costs and the expected
(future) needs will determine which of these technologies are the right analytical strategy for each small molecule MS-
based screening.

Please explain why your abstract is innovative for mass spectrometry?

investigating the newest developments for fast MS in an industrial environment.

461
Session: HC-4 JMS Awardees session

HARNESSING LOW-TEMPERATURE PLASMA CHEMISTRY TO

DISTINGUISH ALKYLATED AROMATIC ISOMERS WITH MASS
SPECTROMETRY
Abstract ID: 1041

Presenting author: Alina Begley, Department Chemistry and Applied Biosciences ETH Zürich, 8093 Zurich,
Switzerland

Introduction

Low-temperature plasma sources, for example, atmospheric pressure chemical ionization (APCI), dielectric barrier
discharge (DBD), or direct analysis in real time (DART), are used in mass spectrometry to ionize gaseous samples.
These sources generate reactive neutrals, excited species, and ions that generally result in the intact molecular analyte
ion. However, variation in the operating conditions can lead to chemically surprising products, for example, we recently
reported that excited state N atoms react with aromatic hydrocarbons to form N-heterocycles at high operating voltage.
Here, we use the voltage-dependent changes in plasma chemistry to differentiate alkylated aromatic hydrocarbon
isomers with varying alkyl chains (at low voltage) and the number of alkyl substituents (at high voltage).

Methods

We infused pure alkylated aromatic hydrocarbon isomers with the formulae C8H10 and C9H12 through a nitrogen dielectric
barrier discharge ionization source (DBDI) coupled online to an orbitrap mass spectrometer. We determined the products
of the compounds at low (2.44 kVpp) or high (3.44 kVpp) operating voltage from their exact masses and fragments
produced by collision-induced dissociation (CID). The products and mechanisms were compared to electron impact
ionization database spectra.

Preliminary data (results)

At low voltage, the products vary depending on the branching of the alkyl substituent. At the same concentration and
plasma conditions, dimethylbenzene C8H10 isomers form the radical cation [M]•+ and lose a methyl group, whereas
ethylbenzene is further fragmented. Likewise, all isomers of C 9H12 form the radical cation [M]•+, but propylbenzene also
forms [C7H7]•+, isopropylbenzene [C8H11]•+, 1-ethyl-4-methylbenzene [C9H11O]•+, and trimethylbenzenes lose a methyl
group. These reaction products are attributed to reactions with ions (N 2+, N3+, N4+, NO+) and atomic oxygen O(3P) from
impurities in the plasma gas. At high voltage, all alkylated aromatic hydrocarbons form two ionization products: an N-
addition product [M+N]+ and an N-replacement product [M-C+N] •+, which are attributed to elevated concentrations of
nitrogen ions and excited state N atoms N(2P), respectively. The branching ratio of the addition and replacement
products depend on both the concentration and degree of substitution of the aromatic hydrocarbon. The greater the gas-
phase concentration of the analyte, the higher the ratio of N-replacement to -addition product. In a preliminary
experiment, when adjusted for concentration, the higher the number of alkyl substituents (toluene<xylene<mesitylene),
the lower the ratio of the substitution to replacement product.

Co-authors:

Renato Zenobi, Department Chemistry and Applied Biosciences ETH Zürich, 8093 Zurich, Switzerland

462
DETERMINING STRUCTURAL MOTIFS AND CONFORMATIONS OF
GLYCOSYL CATIONS BY CRYOGENIC GAS-PHASE ION INFRARED
SPECTROSCOPY
Abstract ID: 1040

Presenting author: Kim Greis, Fritz Haber Institute of the Max Planck Society, Faradayweg 4-6, 14195 Berlin

Introduction

The structural characterization of reactive intermediates, such as the glycosyl cation, is particularly challenging. The high
reactivity leads to short lifetimes and impedes their isolation. We developed a workflow to isolate and structurally
characterize glycosyl cations in the gas phase. They are generated in the “clean-room” environment of a mass
spectrometer via electrospray ionization of glycosyl donors.

Methods

The investigated ions are then accumulated in an ion trap and picked up by superfluid helium droplets (0.4 K). The doped
droplets are excited by the infrared beam of a free-electron laser. Resonant excitation of vibrational modes leads to the
release of the ions from the droplets, which are detected by a time-of-flight detector. The ion count as a function of the
photon wavenumber provides an infrared spectrum.

Preliminary data (results)

The resulting infrared pattern yields the ions’ structural motif (e.g. stabilization of the charge by neighboring protecting
groups) and conformation (e.g. orientation of protecting groups and pyranose ring pucker) by comparison with infrared
spectra of candidate structures computed employing density functional theory.The results indicate that stabilization of the
positive charge at the anomeric carbon is stronger in benzoyl than for the previously probed acetyl protecting groups due
to resonance effects. Furthermore, site-selective fluorination has a strong effect on the conformation of glycosyl cations.
The data show that the conformation is dependent on steric effects of the protecting groups. Finally, for a subset of
glycosyl cations, we compare the gas-phase infrared spectra obtained using cryogenic infrared spectroscopy in helium
droplets to those obtained via infrared multiple photon dissociation (IRMPD) spectroscopy.Here, the results show that the
spectra obtained with the former technique are significantly narrower, allowing for the assignment based on subtle
structural differences.

Please explain why your abstract is innovative for mass spectrometry?

In conclusion, we get unprecedented structural insight into glycosyl cations that allows for predicting the reactivity of the
precursors in condensed phase glycosylation reactions.

Co-authors:

Carla Kirschbaum, Freie Universität Berlin, Institut für Chemie und Biochemie, Arnimallee 22, 14195 Berlin
Gerard Meijer, Fritz Haber Institute of the Max Planck Society, Faradayweg 4-6, 14195 Berlin
P.H. Seeberger, Max Planck Institute of Colloids and Interfaces, Am Mühlenberg 1, 14476 Potsdam
Gert von Helden, Fritz Haber Institute of the Max Planck Society, Faradayweg 4-6, 14195 Berlin
Kevin Pagel, Freie Universität Berlin, Institut für Chemie und Biochemie, Arnimallee 22, 14195 Berlin

463
HIGH-THROUGHPUT BIOANALYSIS USING DESORPTION
ELECTROSPRAY IONIZATION MASS SPECTROMETRY (DESI-MS)
Abstract ID: 101

Presenting author: Nicolas Morato, Purdue University

Introduction

The analysis of complex biological samples using MS typically relies on sample purification prior to analysis. Despite
useful at simplifying complex matrices, the sample work-up limits the throughput that can be achieved. This is a critical
drawback in cases where high-throughput (≥1 sample/s) bioanalysis is required, for instance in the bioactivity
assessment of large compound libraries, large-scale population studies, or the generation of extensive spectral libraries.
We demonstrate the use of desorption electrospray ionization (DESI) as an efficient ambient ionization strategy for the
versatile, direct (work-up free) and rapid (>6,000 samples/h) analysis of complex biosamples including label-free
bioassay mixtures, microorganisms, and tissue, where the high-quality data obtained on the native samples provide
unique opportunities in enzymology, directed evolution and personalized medicine.

Methods

Commercial and custom software and hardware have been developed and integrated into a high-throughput DESI-MS
platform capable of direct analysis of complex biological samples at rates of ≥1Hz. Briefly, a Biomek-i7 workstation
equipped with a pin-tool is used to automatically prepare assays and spot 50-nL aliquots of each sample from microtiter
plates onto PTFE-coated glass slides, thus generating high-density arrays (up to 6,144 samples/plate). The slides are
automatically transferred to a DESI stage and analyzed in a spot-to-spot fashion, producing large data volumes that are
processed in real-time. Both quantitative and qualitative analyses are performed using this workflow.

Preliminary data (results)

The utility and versatility of high throughput DESI-MS (Fig1) has been demonstrated for several types of complex
biological samples and diverse applications. In particular, the excellent quantitative performance of this methodology (low
nM LOQs, RSDs ≤10%) has been repeatedly shown in the case of label-free enzymatic assays. In these cases, crude
bioassay mixtures, including those with high salt/buffer/detergent concentrations, are directly analyzed
(substrates/products are quantified). Following this approach, we have studied complex enzymatic phenomena, such as
the inhibition-reactivation of acetylcholinesterase by chemical agents (Fig2A), and the characterization of enzymes with
potential as novel pharmaceutical targets, such as the cytosolic sulfotransferases 2B1b and 1A1 (Fig2B, Fig2C). In both
cases we have demonstrated better performance than established biochemical approaches relying on labeling,
particularly in terms of increased simplicity and absence of interferences. The same workflow has been extended to
competitive binding assays using opioid receptors, where binding is monitored by determination of a constant
endogenous ligand competitor.

Tissue samples have also been analyzed using the platform. Crude extracts are used to determine activity of specific
bio-pathways upon spiking of an enzymatic substrate, such as the renin-angiotensin system in different mouse organs
upon addition of angiotensin II. Brain biopsies are spotted and rapidly profiled to determine IDH mutation status for
comparison with intrasurgical glioma analyses. Similarly, bacteria species can be rapidly profiled and identified using a
simple extraction step compatible with clinical laboratory workflows, or, alternatively, measured directly from agar plates
to automatically assess product formation upon directed evolution.

Please explain why your abstract is innovative for mass spectrometry?

Automated high-throughput (≥1 Hz) platform for the direct (no work-up) and quantitative mass spectrometry analysis of
various complex biological samples including bioassay mixtures, tissue sections and bacterial colonies.

Co-authors:

Samadhi Kulathunga, Purdue University

My Phuong Le, Purdue University
Rui Huang, Indiana University
Veronica Feng, Purdue University
Hannah Brown, Purdue University
Edwin Gonzalez, Purdue University
Dylan Holden, Purdue University
Indigo Williams, Purdue University

464
Christina Ferreira, Purdue University
Qing Zhou, Purdue University
Jared Lewis, Indiana University
Andrew Mesecar, Purdue University
R. Graham Cooks, Purdue University

Figure 1. General workflow using the high-throughput DESI-MS platform.

Figure 2. Representative results of quantitative high-throughput DESI-MS bioassays.

465
THUNDERSTORMS: CLOUD ELECTRIFICATION OR SPONTANEUS WATER
IONIZATION?
Abstract ID: 828

Presenting author: Luan Felipe C. Oliveira, University of Campinas - UNICAMP

Introduction

Thunderstorms has been studied using instruments to measure electromagnetic properties. This methodology have been
very useful to describe the electromagnetic behavior of atmospheric electricity, for example the mechanisms of cloud
electrification, distribution of charges, initiation and propagation of discharges [1-5]. However the nature of the charges and
their electrification is not yet totally understand. Mass spectrometry (MS) is able to study molecular ions. Herein were
simulated experiments of physical changes in water, that could happen in a cloud during a thunderstorm and their
respective ionic current was measure to study the respective spontaneous water ionization process. This work is a prove
of concept of the use of MS to understand the molecular aspects involved in thunderstorms.

Methods

A quadrupole-mass spectrometer (LCMS shimadzu 2010 EV) with Ventury Easy Ambient Sonic-Spray ionization (V-
EASI) was used to measure ionic current (IC) in positive and negative mode with a range of m/z 10-1000 of Milli Q water
during: melting process (to simulated the transition solid-liquid) and warming in the range of 25 °C – 80 °C (to simulated
the thermal instability), while the boiling process (to simulating the transition liquid-vapor) was measured without any
source of ionization.

Preliminary data (results)

Transition Solid-Liquid: it was detected a range of 25 % in the IC along melting process, comparatively the range of IC in
ambient temperature had a behavior with a range between 15%. The mass spectrum shows that the molecular species
responsibile for range in IC could are cluster of water because of gaussian distributions of clusters . Transition Liquid-
Steam: it was possible to detect IC during the vaporization and it was directly proportional at that the volume of steam
and the temperature. The mass spectrum acquired present few signals in the range of m/z 500-700 whats suggest a
limited range of cluster water distribution other alternative was the formation of a cluster with a compound which promote
the nucleation of the cluster. Thermal instabilitity: it was possible to detect IC during the warming process a rise of two
times in the IC only between 25 °C – 60 °C was present in the positive mode, between 60°C – 80°C the IC was constant.
On the other hand, in the negative, a rise with exponential behavior in the whole interval studied was observed and an
increase of four times, in the spectrum was observed a profile similar during melting process that could be a distribution
of cluster of water. The rise of charge during the melting is in agreement with other works [6-8], while the charge increase
during the range of temperature and liquid-vapor transition. These results presents spontaneous water ionization that
could be associated to cloud electrification.

Please explain why your abstract is innovative for mass spectrometry?

This work presents a new application of MS, the study of thunderstorms, showing to be efficient to study processes of
electrification already studied and new ones.

466
INTERMITTENT FASTING INDUCES SEXUALLY DIMORPHIC HEPATIC
INTERFERON ALPHA SIGNALING
Abstract ID: 471

Presenting author: Dylan Harney, Charles Perkins Centre and School of Life and Environmental Sciences,
University of Sydney, 2006, Sydney, Australia

Introduction

Intermittent fasting (IF) is a beneficial dietary treatment for obesity which improves liver metabolic health independent of
weight loss. The liver is a key fasting responsive organ that stores glucose for immediate use and co-ordinates switching
of metabolic substrates. Previous IF studies have demonstrated the liver upregulates fatty acid catabolism and
scavenging pathways in response to IF and increases lipid storage within the liver. However, these studies have only
examined the IF response in males. There are significant differences in metabolism between males and female. In this
study, we examined the differences in the liver’s response to IF using proteomics. Additionally, we compared two
different approaches for bottom-up proteomic analysis: standard DIA on a long gradient versus scanningSWATH on a 5-
minute gradient.

Methods

Male and female C57B/6J mice were subject to a 2-week every-other-day fasting (EODF) model with chow and
compared against an ad libitum control group. Metabolic health was assessed throughout the model, tissues were frozen
on LN2 and liver tissue was lysed and submitted to a standard trypsin digest and stage-tip cleanup. Samples were
analysed using either conventional DIA proteomics on a Thermo Eclipse instrument coupled to a Dionex U3000 nanoLC
or using Scanning-SWATH on Sciex 6600/7600 instruments coupled to a high-flowrate ExionLC. Data was analysed
using the DIA-NN search engine and subject to statistical analysis by two-way ANOVA.

Preliminary data (results)

Results (250 words)

To compare the liver proteome we performed bottom-up proteomics using trypsin digestion and peptide cleanup using
SDB-RPS Stagetips. These analyses consistently identified >4,500 proteins across the cohort where EODF had a
stronger influence on protein abundance than sex. Animals of both sexes had increased fatty acid catabolism, but this
change was much greater in females whereas males had greater increases in fatty acid synthesis. Interestingly, females
had increased interferon-alpha signalling after IF and increased abundance of downstream IFNα targets, whereas males
had minimal change. To validate these findings, we applied IF to castrated mice and repeated the proteomic analysis,
which showed testosterone signalling represses IFNα pathway induction in the liver after IF. Conversely, analysis of liver
tissue from ovariectomised mice did not diminish this IFNα response, suggesting the sexual dimorphism arises from
testosterone induced suppression via the androgen receptor. Next, mice that had one of the IFNα receptors knocked out
(IFNAR1) were subject to IF, in which these animals produced no IFNα signalling response demonstrating that receptor
engagement is necessary for this response. Future work will test the hypothesis that IFNα signalling is tied to cholesterol
biosynthetic flux. This work highlights an interesting intervention that improves metabolic health even in the absence of
weight loss and is a novel opportunity to develop new dietary treatment strategies.

Please explain why your abstract is innovative for mass spectrometry?

In this research, we conducted the first analysis of complex tissue lysate and plasma using both conventional DIA
proteomics and Scanning-SWATH.

Co-authors:

Michelle Cielesh, Charles Perkins Centre and School of Medical Sciences, University of Sydney, 2006, Sydney, Australia
Matthew Yang, Charles Perkins Centre and School of Medical Sciences, University of Sydney, 2006, Sydney, Australia
Barney Viengkhou, Charles Perkins Centre and School of Life and Environmental Sciences, University of Sydney, 2006,
Sydney, Australia
Markus Hofer, Charles Perkins Centre and School of Life and Environmental Sciences, University of Sydney, 2006,
Sydney, Australia
Mark Larance, Charles Perkins Centre and School of Life and Environmental Sciences, University of Sydney, 2006,
Sydney, Australia, Charles Perkins Centre and School of Medical Sciences, University of Sydney, 2006, Sydney,
Australia

467
Mouse intermittent fasting model and proteomics workflow.

Principle component analysis of intermittent fasting model between sexes.

468
469
470