Advances in Experimental Medicine and Biology 1439

Proteomics, Metabolomics, Interactomics and Systems Biology

Taícia Pacheco Fill   Editor

Microbial
Natural
Products
Chemistry
A Metabolomics Approach
Advances in Experimental Medicine and Biology

Volume 1439

Series Editors
WIM E. CRUSIO, Institut de Neurosciences Cognitives et Intégratives
d’Aquitaine, CNRS and University of Bordeaux, Pessac Cedex, France
HAIDONG DONG, Departments of Urology and Immunology, Mayo Clinic,
Rochester, MN, USA
HEINFRIED H. RADEKE, Institute of Pharmacology and Toxicology,
Clinic of the Goethe University Frankfurt Main,
Frankfurt am Main, Hessen, Germany
NIMA REZAEI, Research Center for Immunodeficiencies, Children's
Medical Center, Tehran University of Medical Sciences, Tehran, Iran
ORTRUD STEINLEIN, Institute of Human Genetics, LMU University Hospital,
Munich, Germany
JUNJIE XIAO, Cardiac Regeneration and Ageing Lab, Institute of Cardiovascular
Sciences, School of Life Science, Shanghai University, Shanghai, China

Proteomics, Metabolomics, Interactomics

and Systems Biology
Subseries Editor
DANIEL MARTINS-DE-SOUZA, Institute of Biology,
Laboratory of Neuroproteomics, University of Campinas (UNICAMP),
Campinas, Brazil
Advances in Experimental Medicine and Biology provides a platform for scientific
contributions in the main disciplines of the biomedicine and the life sciences. This
series publishes thematic volumes on contemporary research in the areas of
microbiology, immunology, neurosciences, biochemistry, biomedical engineering,
genetics, physiology, and cancer research. Covering emerging topics and techniques
in basic and clinical science, it brings together clinicians and researchers from
various fields.
Advances in Experimental Medicine and Biology has been publishing exceptional
works in the field for over 40 years, and is indexed in SCOPUS, Medline (PubMed),
EMBASE, BIOSIS, Reaxys, EMBiology, the Chemical Abstracts Service (CAS),
and Pathway Studio.
2022 CiteScore: 6.2
This series of volumes focuses on concepts, techniques and recent advances in the
field of proteomics, interactomics, metabolomics and systems biology. Recent
advances in various ‘omics’ technologies enable quantitative monitoring of myriad
various biological molecules in a high-throughput manner, and allow determination
of their variation between different biological states on a genomic scale. Now that
the sequencing of various genomes, from prokaryotes to humans, has provided the
list and linear sequence of proteins and RNA that build living organisms, defining
the complete set of interactions that sustain life constitutes one of the key challenges
of the postgenomic era. This series is intended to cover experimental approaches for
defining protein-protein, protein-RNA, protein-DNA and protein-lipid interactions;
as well as theoretical approaches dealing with data analysis, integration and
modeling and ethical issues.
Taícia Pacheco Fill
Editor

Microbial Natural Products

Chemistry
A Metabolomics Approach
Editor
Taícia Pacheco Fill
Institute of Chemistry
Universidade Estadual de Campinas
Campinas, Brazil

ISSN 2662-3145     ISSN 2662-3153

Advances in Experimental Medicine and Biology
ISSN 0065-2598     ISSN 2214-8019 (electronic)
Proteomics, Metabolomics, Interactomics and Systems Biology
ISBN 978-3-031-41740-5    ISBN 978-3-031-41741-2 (eBook)
https://doi.org/10.1007/978-3-031-41741-2

© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature
Switzerland AG 2023
This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether
the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of
illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and
transmission or information storage and retrieval, electronic adaptation, computer software, or by similar
or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors, and the editors are safe to assume that the advice and information in this book
are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the
editors give a warranty, expressed or implied, with respect to the material contained herein or for any
errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Paper in this product is recyclable.

Preface

Since the discovery of penicillin, microbial natural products have been widely
explored concerning their bioactivities and biosynthetic pathways. Nowadays, natu-
ral products from microbial origin are applied in different sections of society,
including agriculture, pharmaceutical and food industries. From a chemical per-
spective, NPs are advantageous over synthetic molecules once they present great
scaffold diversity and structural complexity that have been optimized along millions
of years of evolution to interact with specific biological targets.
Although NPs have greatly contributed to society, the functionality of these
important molecules in their biological context remains unknown in many biologi-
cal systems. Researchers all over the globe are now focusing efforts on understand-
ing the biological function of secondary metabolites in their natural environment.
Understanding the way nature works is the first step to develop strategies to control
animal and plant diseases and design targeted medicines.
Classical approaches for bioactive natural products discovery usually involve the
isolation of the metabolites guided by biological activity screening of the fractions
and further characterization of the compounds using different spectroscopic and
spectrometric approaches. Although the classical methodologies are successful and
were greatly important to achieve the current knowledge concerning NPs chemistry,
the methodologies can be time-consuming and many times the efforts can lead to
the re-discovery of already known molecules. In this sense, metabolomics
approaches have been helping NPs chemists to unravel the metabolome of different
organisms, identify new biosynthetic pathways and screen for the metabolic profile
in accordance with different environmental variables or diseases.
This book focuses on natural products chemistry and the importance of omics
strategies to unveil new molecules from different natural sources with diverse chem-
ical structures and biological functions. Metabolomics strategies that could lead to
the understanding of the chemical interactions between microorganisms, plant-­
microorganisms and virus-microorganisms focusing in the secondary metabolites
involved in such interactions, are also highlighted here. We present all the newest
and ultimate source of information concerning state-of-the-art analytical tools used
for natural products discovery and microbial metabolomics, annotation protocols

v
vi Preface

for secondary metabolites using different databases and online platforms, and
dereplication strategies leading to new backbone molecules, including stepwise
pipelines.

Campinas, Brazil Taícia Pacheco Fill

Contents

1 Unveiling Chemical Interactions Between Plants

and Fungi Using Metabolomics Approaches ����������������������������������������    1
João Guilherme de Moraes Pontes, Mayra Suelen da Silva Pinheiro,
and Taícia Pacheco Fill
2 Metabolomics Applied to Cyanobacterial Toxins
and Natural Products������������������������������������������������������������������������������   21
Márcio Barczyszyn Weiss, Rhuana Valdetário Médice,
Fernanda Rios Jacinavicius, Ernani Pinto,
and Camila Manoel Crnkovic
3 Unveiling Microbial Chemical Interactions Based
on Metabolomics Approaches ����������������������������������������������������������������   51
Laís Castro de Carvalho, Arnaldo de Almeida Junior,
Fernanda Silva Ribeiro, and Célio Fernando Figueiredo Angolini
4 
Chemical-Biology and Metabolomics Studies in Phage-Host
Interactions����������������������������������������������������������������������������������������������   71
Rodolfo Dantas, Marcelo Brocchi, and Taícia Pacheco Fill
5 Advances in Mass Spectrometry-­Metabolomics Based
Approaches ���������������������������������������������������������������������������������������������� 101
Nerilson Marques Lima, Gabriel Franco dos Santos,
Gesiane da Silva Lima, and Boniek Gontijo Vaz
6 
Advances in Microbial NMR Metabolomics ���������������������������������������� 123
Ricardo Moreira Borges, Gonçalo Jorge Gouveia,
and Fernanda Oliveira das Chagas
7 Sample Preparation in Microbial Metabolomics:
Advances and Challenges������������������������������������������������������������������������ 149
Heiter V. M. Boness, Hanna C. de Sá, Emile K. P. dos Santos,
and Gisele A. B. Canuto

vii
viii Contents

8 
Discovering New Natural Products Using Metabolomics-Based
Approaches ���������������������������������������������������������������������������������������������� 185
Lívia Soman de Medeiros, Moysés B. de Araújo Júnior,
Eldrinei G. Peres, José Carlos Ipuchima da Silva,
Milena Costa Bassicheto, Giordanno Di Gioia,
Thiago André Moura Veiga, and Hector Henrique Ferreira Koolen
9 
Microbial Metabolites Annotation by Mass Spectrometry-Based
Metabolomics�������������������������������������������������������������������������������������������� 225
Paulo Wender P. Gomes, Talita Carla de Tralia Medeiros,
Naydja Moralles Maimone, Tiago F. Leão,
Luiz Alberto Beraldo de Moraes, and Anelize Bauermeister
Index������������������������������������������������������������������������������������������������������������������ 249
Chapter 1
Unveiling Chemical Interactions Between
Plants and Fungi Using Metabolomics
Approaches

João Guilherme de Moraes Pontes, Mayra Suelen da Silva Pinheiro,

and Taícia Pacheco Fill

1.1 Introduction

Metabolomics is the systematic identification and quantification of metabolites in a

certain organism, where the metabolic profile may vary in accordance to environmen-
tal variables, infections, nutrition, among others; and these are statistically compared
to a control group [1, 2]. Discussions about the physiological state of an organism
based in low-molecular-mass compounds and not only through observation of genes
and proteins has already occurred between the 70s and 90s; however, the term “metab-
olome” was first used in 1998 by Oliver et al. in a study related to the genome of the
yeast Saccharomyces cerevisiae [3]. In 2001, Oliver Fiehn introduced the concept of
“Metabolomics” and proposed different analytical approaches such as metabolite pro-
filing and metabolic fingerprinting [1, 5, 6]. In this sense, mass spectrometry (MS) and
nuclear magnetic resonance (NMR) spectroscopy are analytical platforms most used
in metabolomics studies, because of good sensitivity of MS and high spectral resolu-
tion of NMR with magnetic field higher than 500 MHz [7–10].
Although metabolomics is a well-established field in the scientific community
and used for various purposes, there is still much to be discovered concerning
fungal-­plant interactions [11]. Plants are considered an important reservoir of
microorganisms and many microorganisms species are still unknown and often
referred to as “microbial dark matter” in literature [11, 12]. The diversity of species
present in the plant microbiota makes it difficult to understand the contribution of
each microorganism in the different interactions [12, 13]. Furthermore, it is

J. G. de Moraes Pontes · M. S. da Silva Pinheiro · T. Pacheco Fill (*)

Universidade Estadual de Campinas (UNICAMP), Instituto de Química,
Laboratório de Biologia Química Microbiana (LaBioQuiMi),
Campinas, SP, Brazil
e-mail: [email protected]

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 1

T. Pacheco Fill (ed.), Microbial Natural Products Chemistry, Advances in Experimental
Medicine and Biology 1439, https://doi.org/10.1007/978-3-031-41741-2_1
2 J. G. de Moraes Pontes et al.

estimated a range between 200,000 and 1000,000 metabolites in the plant kingdom,
where 51,000 is recorded in databases [14]. In practice, not all metabolites from
metabolome can be identified, because there are limitations in the extraction meth-
ods, ionization of molecules, and analytical platforms used for detection of metabo-
lites [10].
In agriculture and biological sciences, the identification and quantification of
metabolites in plants infected by fungi and bacteria helps the understanding of the
physiological state of plants, since these metabolites undergo changes during the
phytopathogen-plant interactions. So, it is possible to evaluate the plant metabolic
pathways that were affected in the infection, the genes related to plant defense, and
virulence factors produced by phytopathogens as microbial infection mechanisms
[15, 16]. In this context, metabolomics have been applied in different applications
such as identification of biomarkers, disease diagnosis methods, chemotaxonomy,
signaling compounds production, plant defense mechanisms, and evaluation of
plant resistance. Figure 1.1 shows some examples of metabolomics applications
used in fungal-plant interactions research, which will be discussed throughout the
chapter.

1.2 MS-Based Metabolomics

Mass spectrometry technique has the versatility of being able to hyphenate with
chromatographic separation techniques such as high-performance liquid chroma-
tography (HPLC), gas chromatography (GC) and also other analytical platforms.
This feature further increases its applicability in metabolomics studies [17]. In this
context, LC-MS/MS and GC-MS spectra are commonly used to identify and anno-
tate primary and secondary metabolites, since the fragmentation pattern of these
spectra allow the construction of libraries for identification of compounds [18–20].
Data of m/z values, retention time, and chromatographic peak intensity are used
for chemometrics analysis such as principal component analysis (PCA), partial least
square discriminant analysis (PLS-DA), and soft independent modelling of class
analogies (SIMCA), which can indicate pattern recognition (PCA), discrimination
(PLS-DA), or classification of groups (SIMCA) [21–24]. PCA and PLS-DA can be
performed on MetaboAnalyst, an analytical web-based platform of free access [25],
while SIMCA can be performed on Pirouette® (InfoMetrix) or MATLAB® soft-
wares [26, 27].
The identification of metabolites can be done in four different levels in accor-
dance with the Metabolomics Standards Initiative (MSI) [28]. So, a compound clas-
sified as “identified” (level-1) is one that the MS analysis was done with the
co-injection of an authentic standard, while the “putative annotation” of a com-
pound (level-2) is done using the comparison between the similarity of experimen-
tal MS spectra with databases and literature. The classification of level-3 (putative
characterization) is done through comparison of physicochemical properties and
spectral similarity with a compound of a known chemical class. Finally, “unknown
1 Unveiling Chemical Interactions Between Plants and Fungi Using Metabolomics… 3

Fig. 1.1 Examples of metabolomics applications in fungal-plant interactions: (a) biomarkers

search: galactitol is a potential biomarker of white rot fungi (Trametes versicolor and Pleurotus
ostreatus) accompanied by a decrease in phenolic compounds and carboxylic acid levels; (b) diag-
nosis: 1-propanol and ethyl alcohol are used as biomarkers in early stage of Rhizopus soft disease
that affects sweet potatoes; (c) chemotaxonomy: phenylacetic acid is found in different levels in
the classification of anastomosis groups (AGs) of Rhizoctonia solani, while α,α-trehalose is found
in all AGs; (d) Signaling compounds and plant defense: jasmonic acid and salicylic acid signaling
are increased in peaches during the Monilinia fructicola infection after 12 h and 48 h, respectively;
(e) Plant resistance: methyl jasmonate is accumulated in resistant varieties of watermelon plants
upon Fusarium oxysporum f. sp. niveum infection

compounds” (level-4) are those that can be differentiated and quantified using data
from chromatograms and mass spectra, but are not identified/annotated through
spectral comparisons [10, 28].
Some databases and analytical platforms are available online for MS-spectra
comparison such as Global Natural Product Social Molecular Networking (GNPS)
[29], Feature-Based Molecular Networking (FBMN) [30], MassBank of North
America (MoNA) [31], MetFrag [32], and The Human Metabolome Database –
HMDB [33], among others [10]. Furthermore, there are databases of a diversity of
natural products as the Dictionary of Natural Products [34] and those specifically
produced by fungi as the FungalMet [35] or yeasts as the The Yeast Metabolome
Database – YMDB [36].
4 J. G. de Moraes Pontes et al.

1.3 NMR-Based Metabolomics

NMR spectra acquisition is done using different pulses sequences, which are chosen
in accordance to the finality of information to be obtained. Commonly, for the anal-
ysis of complex biological samples, it is acquired 1H-NMR spectra, which data from
chemical shift and peak intensity are converted and transferred to a chemometric
worksheet for the multivariate analysis [4].
Sometimes, the peak intensity of metabolites of interest is very low in relation to
the water peak, and then it is necessary to eliminate or reduce the intensity of the
water peak [37]. Therefore, NMR experiments of water suppression such as WATER
suppression by GrAdient-Tailored Excitation (WATERGATE) or the Water sup-
pression Enhanced Through T1 effects (WET) are used for this finality [38–40]. In
other cases, it is necessary to eliminate peaks of some metabolites with molecular
weight higher than 1000 Da (lipids, peptides, and proteins), which can be achieved
by adjusting echo time and relaxation delay parameters [41, 42]. Like this, it is used
a Carr-Purcell-Meiboom-Gill (CPMG) pulse sequence [43].
The assignments of metabolites can be done using data of chemical shift values,
chemical shift correlations of 2D NMR spectra, integration of peak area, multiplic-
ity of signs, and J-coupling constant in comparison to data available on databases
and/or reported in literature [44]. Some databases available online are: Biological
Magnetic Resonance Data Bank – BMRB [45] and The Human Metabolome
Database – HMDB [33]. Tables of assignment of metabolites of some compound
classes such as amino acids, carbohydrates, lipids, and organic acids can be con-
sulted on [46] or [47]. Chemical shift of impurities can be consulted on [48] or [49],
considering slight deviations in chemical shift values due to solvent and sample type.

1.4 Biomarkers Search in Plant-Microorganism Interactions

Biomarkers or biological markers is a term frequently used in clinical studies and

has been discussed in literature and defined in different manners. In general, bio-
markers are medical signs, which can be reproducibly measured and quantified indi-
cating biological, pathogenic processes or a response to therapeutic interventions
and medicine [50, 51]. There are countless searches aiming at the identification of
biomarkers to verify the health state of an organism and/or the progression of a
certain disease or disorder [7, 52]. However, the use of biomarkers is not restricted
only to clinical studies, these have also been used in different biological contexts
such as monitoring the quality of food that are affected by postharvest diseases [53],
metabolites associated to resistance of a plant species to a pathogen [54], biomark-
ers associated to plant stress [55], understanding the plant defense mechanisms
[56], among others.
An example of biomarkers identification related to the wood-degrading stage
affected by brown rot and white rot diseases is shown in Fig. 1.1a. Brown rot disease
1 Unveiling Chemical Interactions Between Plants and Fungi Using Metabolomics… 5

is caused by Rhodonia placenta and Gloeophyllum trabeum, while white rot is

caused by Trametes versicolor and Pleurotus ostreatus, which are microorganisms
classified as wood-decay fungi, because they can decompose lignified cell walls
[57, 58]. Brown rot fungi are considered the most destructive group of fungi decay
and they naturally occur in the northern coniferous forest from Germany [59], while
white rot fungi commonly colonize stems from fallen trees and are considered the
primary agents for cycling plant nutrients in forests [60].
In this study, Castaño et al. [57] performed for the first time an in planta extracel-
lular metabolomics study of the fungal biodegradation process of the wood for com-
parison between white and brown rot fungi. They reported about 100 metabolites
produced in the wood-degrading, which were detected though GC-MS analysis,
where galactitol was reported as a potential biomarker of white rot decay, since this
metabolite was detected only in white rot fungi. Furthermore, they reported general
metabolic patterns in decay such as a higher sugars release in early stage and a sig-
nificant decreasing of carboxylic acids (8-hydroxyoctanoic acid and heptanoic acid)
and phenolic compounds in white rot fungi, while brown rot fungi were character-
ized with a higher production of pyranones, furanones, and specific sugars (lactose
and glyceric acid) at late decay stages [57].
The higher sugars released in white rot decay in early stages may be related to a
capacity of T. versicolor and P. ostreatus degrade the hemicellulose earlier than
brown rot fungi. The higher levels of soluble carbohydrate available may be a bio-
logical advantage against competitors [57]. Furthermore, in general, white rot fungi
were more efficient in removing phenolic compounds of undecayed wood, which is
considered another biological advantage, since these compounds exhibit antimicro-
bial activities. However, in late stages, phenolic compounds were higher catabo-
lized by G. trabeum [57].
The metabolomics analysis of fungal biodegradation is important not only due to
environmental and economic issues, but also to study metabolites produced by fungi
during the biodegradation process, since the profile of metabolites produced will
change according to new interactions established between the fungus and the type
of degraded material. In addition, the metabolites studied may also have high aggre-
gate values [57, 61] and potential commercial use.

1.5 Disease Diagnosis in Plants Using

Metabolomics Approaches

In general, plants are often affected by different types of diseases, including those
caused by microorganisms [5]. Phytopathogenic fungi, which are among the main
causes of pathogenicity in plants represent a major hazard for global food security
and are often responsible for the reduction and/or extinction of certain agricultural
productions [62]. In this sense, it is essential that the method used to diagnose a
certain disease is reliable in the early stages of the disease (often an asymptomatic
6 J. G. de Moraes Pontes et al.

stage) and during the development of the infection [63], since it helps agricultural
producers to avoid inadequate control measures, as well as the excessive use of
harmful pesticides to human health and the environment [64].
Current advances in molecular approaches of Polymerase Chain Reaction (PCR)
[65], real-time PCR [66], nested PCR [67], among others [68], are being increas-
ingly applied in the diagnosis of plant diseases, replacing traditional methods of
microscopy and serology, which can lead to misdiagnosis, once many phytopatho-
gens have similar morphological characteristics, which promotes ambiguous results
and difficulty in the differentiation of fungal diseases [65, 69]. However, PCR is an
expensive technique, time-consuming for sample preparation, and it can lead to
false-negative results [70, 71].
In this context, metabolomics combined with sophisticated analytical techniques
as LC-MS, GC-MS and NMR, as well as chemometric analysis, emerged as a prom-
ising tool for the diagnosis of plant diseases [72]. Chemical and biological investi-
gations using metabolomics studies provide valuable information on metabolic
changes during the pathogen-plant interactions, especially in the identification and
quantification of metabolites that are strongly associated with the disease, also
called chemical biomarkers. After identified, this knowledge helps in the under-
standing of the pathogenesis process and in the development of potential treatments
of the diseases and, above all, in the early diagnosis [73, 74].
An example of metabolomics application in diagnosis of plant diseases is shown
in Fig. 1.1b, where Mohottige et al. [75] developed a method using GC-MS-based
metabolomics to diagnose Rhizopus soft disease during its asymptomatic stage in
sweet potatoes. In this metabolomics study, they used five replicates of infected
sweet potato samples and five control non-infected samples and analyzed volatile
organic compounds (VOCs) on different days during 4 weeks after the fungal inocu-
lation [75]. They observed a significant increase in alcohol production (1-propanol,
ethyl alcohol, and 3-methyl 3-buten-1-ol) and ethyl propionate during asymptom-
atic stage, where the combination of these four putative biomarkers could indicate
Rhizopus soft disease [75].
Besides the Rhizopus soft disease, several studies reported potential biomarkers
associated with plant diseases caused by fungi, including studies with maize tissues
infected with the necrotrophic pathogen Rhizoctonia solani and its phytotoxin phen-
ylacetic acid (PAA) [76], studies with strawberry plants infected with Botrytis cine-
rea [77] and with seeds of Solanum tuberosum L. (potato) infected by Fusarium
sambucinum and Fusarium oxysporum, where VOCs were identified as the main
markers of infection [78].
Tan Dai et al. (2019) developed a method for the early diagnosis of anthracnose
caused by Colletotrichum theobromicola in strawberry plants, via untargeted
metabolomics based on GC-MS. Through analyzes of the metabolic profile, 189
ions were annotated for control plants and 202 for infected plants. A PLS-DA analy-
sis model discriminated between healthy and infected plants. Based on the data,
about 20 metabolites were identified as potential biomarkers of the infection by
Colletotrichum theobromicola in the early stages, including citric acid, D-xylose,
erythrose, galactose, gallic acid, malic acid, methyl α-galactopyranoside, and
1 Unveiling Chemical Interactions Between Plants and Fungi Using Metabolomics… 7

shikimic acid. Therefore, this study has a potential applicability for the early diag-
nosis of anthracnose in strawberries [79].
In another study, Pancoro et al. [80] also identified about twenty biomarkers
associated with the basal stem rot (BSR) disease in the interaction between the phy-
topathogen Ganoderma boninense and tissues of oil palm stems, which were
observed at different stages of development of the disease using 1H-NMR-based
metabolomics analysis. The annotated biomarkers belong to different groups of
chemical compounds, including organic compounds, carbohydrates, heterocyclic
organics, nitrogenous organics, and phenolic compounds. Based on the results of
this study, the authors suggest that the model should be applied for purposes of
diagnosing BSR disease [80].
The correct diagnosis of sick plants provides the implementation of proper con-
trol strategies or prevention of fungal diseases [63]. In this sense, the studies previ-
ously presented [75, 79, 80] confirm the potential of metabolomics as an emerging
tool in the development of diagnostic methods, since the identification of biomark-
ers can indicate the presence, absence or state of a certain disease, which allows
establishing new strategies for the protection of plants against microbial diseases
[72, 81].

1.6 Chemotaxonomy

The term “chemotaxonomy” or “molecular taxonomy” shows different meanings in

different research fields. In a general sense, the chemotaxonomy is obtained using a
diversity of compound classes (fatty acids, proteins, carbohydrates, secondary
metabolites, and DNA sequences) for classification, identification, and taxonomic
annotation of an organism. For filamentous fungi this classification is mainly based
on secondary metabolite profiles, once fungi are good producers of secondary
metabolites and the production of each compound is limited to a number of fungal
species, genus or phylum [82].
There are a variety of studies that have successfully approximated the chemotax-
onomy of metabolomics approaches [83–85]. However, there are relatively few
studies aiming at the classification and identification of fungi [86]. Chemotaxonomic
studies associated to metabolomics have been highlighted once the metabolomics
approaches may be able to circumvent inherent problems of traditional chemotax-
onomy, which is mainly based on qualitatively and/or quantitatively analyzing a
specific group of secondary metabolites. So, the traditional chemotaxonomy may
not consider genetic and environmental variations arising from biochemical pro-
cesses and could lead to inconclusive results [85, 87].
An example of a chemotaxonomic study based on metabolomics analysis was
done by Aliferis et al. [86] using GC-MS metabolite profiling and footprinting to
differentiate anastomosis groups (AGs) of Rhizoctonia solani (Fig. 1.1c), since this
is a complex species of phytopathogenic filamentous fungus that have diverse
genetically isolated groups, which were named anastomosis groups and divided in
8 J. G. de Moraes Pontes et al.

13 AGs (AG-1 to AG-13) and other subdivisions [86, 88]. The understanding of
chemotaxonomy of these fungal species in each group is important not only for
taxonomy, but also for physiological studies, metabolic engineering, and monitor
mutations [86]. In this study, metabolomics was designed using a footprinting
approach of the R. solani infection that is based on the analysis of extracellular
metabolites of the fungus, which are secreted to the culture media [89]. Furthermore,
it was used GC-MS, since it is a highly reproducible technique and commonly
found in research institutes, which becomes suitable for the development of proto-
cols [86, 90].
Phenylacetic acid (PAA) was detected in most of AGs, with highest detection in
AG-3 PT, and in negligible amounts in AG-6. In addition, it was not detected in
AG-5 and AG-10. PAA and its derivatives have been reported as primary determi-
nant and/or virulence factor in the pathogenicity of R. solani, the causative agent of
Rhizoctonia disease on tomatoes, since PAA induces necrosis and can damage
tomato seedlings [86, 91]. PAA is a component of fungal sclerotia, which consists
of a hard structure present in many fungal bodies containing food reserves and it is
resistant to adverse environmental conditions [86, 92].
α,α-Trehalose was found in all AG. This metabolite is found in a variety of living
beings (plants, insects, and microorganisms) and in fungi have important physiolog-
ical functions such as to keep the cell integrity against biotic stresses and as cellular
reserve of carbohydrate [86].

1.7 Signaling Compounds and Plant Defense

Plants can produce a broad diversity of natural products, which exhibits different
biological functions such as physiological compounds (primary metabolism), com-
pounds acting against predatory insects, phytohormones, signaling compounds,
among others [93, 94]. In this context, secondary metabolism is directly involved in
the protection of plants against predators, pathogens, and abiotic stresses, being
important for the growth and development of plants [95]. The defense mechanisms
commonly used by plants against microbial phytopathogens are cell wall fortifica-
tion, antimicrobial metabolites production, and reactive oxygen species (ROS) gen-
eration [93].
Examples of secondary metabolites and phytohormones, which act as signaling
compounds produced in plant responses, are salicylic acid and jasmonic acid [96,
97]. Currently, Cheng et al. performed metabolomics and transcriptomics analysis
of peaches infected by Monilinia fructicola and identified 357 differential metabo-
lites using LC-MS [98]. M. fructicola is the etiologic agent of brown rot, a posthar-
vest disease that affects stone fruits and causes more than 50% of global losses of
peach fruits [98, 99]. In this study, Cheng et al. observed an increase in signaling of
jasmonic acid and salicylic acid after 12 h and 48 h of infection, respectively, as
shown in Fig. 1.1d. Based on previous studies and observation of up-regulation of
myc2 gene in peaches after M. fructicola infection, which is a gene that acts as a
1 Unveiling Chemical Interactions Between Plants and Fungi Using Metabolomics… 9

positive regulator of wound-responsive genes (VSP1, JR1, TAT, and LOX) in situa-
tions of insect herbivory attack and wounds, Cheng et al. suggested that the role of
jasmonic acid during the plant defense is associated with antioxidant protection of
fruit, while up-regulation of NPR1, TGA and PR1 genes are associated to plant
immunity, once NPR1 is a receptor of salicylic acid and participates of plant resis-
tance [98, 100].
Other secondary metabolite classes, which exhibit antimicrobial activities and
act in plant defenses, are phenolic compounds, terpenoids, and alkaloids [95, 101].
Figure 1.2 shows examples of these compounds against fungal pathogens.
Hu et al. [102] performed proteomics and metabolomics analysis of Verticillium
dahliae-tomato interactions in resistant and susceptible plants. V. dahliae is a soil-­
borne fungus and the causal agent of Verticillium wilt in numerous plant species, a
disease that causes vascular discoloration, stunting, wilting, and necrosis of the
leaves [102, 103]. Results of metabolomics analysis performed by HPLC-MS/MS
indicated that the phytopathogen induced the production of different phenylpro-
panoids such as caffeic acids derivatives (Fig. 1.2a), quinic acid derivatives, flavo-
noids, vanillic acid hexoside, syringic acid hexoside, and coumarins, which exhibit
antimicrobial activities and act as phytotoxins [102].
Phenylpropanoid pathway is triggered when pathogens interact with the plant
cell wall and then complex networks are regulated and start the production of phe-
nolic and antimicrobial compounds, signaling molecules, and the lignification pro-
cess [104]. Antimicrobial mechanisms proposed for phenolic compounds as the

Fig. 1.2 Examples in literature of secondary metabolites of different compound classes: (a) phen-
ylpropanoids; (b) terpenoids; (c) alkaloids; produced by plants as response to fungal infections
detected in metabolomics analysis
10 J. G. de Moraes Pontes et al.

caffeic acid (3, 4-dihydroxycinnamic acid) are based on protein synthesis inhibition
in the pathogen cells and pathogen’s membrane disintegration, due to the cell leak-
age [105].
Terpenoids are also associated to plant defense mechanisms [106]. Phytochemical
compounds such as Rhodexin A (Fig. 1.2b) and Bryophyllin A have been putatively
identified by Fourier-transform ion cyclotron resonance mass spectrometry (FT-ICR
MS) in positive and negative modes. These terpenoids were differentiated between
mycorrhizal roots from an oak tree (Quercus suber) colonized by Pisolithus tincto-
rius and the control group (non-colonized roots) [107]. Some other metabolites
(glucose, D-fructopyranose, quercitol, among others) were structurally elucidated
by NMR spectroscopy in fingerprinting analysis of root sample extracts from cork
oak (Quercus suber) using 1H NMR and bidimensional (COSY, TOCSY, HSQC,
and HMBC) spectra in comparison to databases and literature [107].
Pisolithus tinctorius is an ectomycorrhizae (ECM), which fungal mycelium can
access the soil nutrients and be transferred to roots of different plant species confer-
ring resistance and protection [107]. The mutualistic relationship established
between the ectomycorrhizal fungus and the plant host occurs because there are
fine-tuned interactions between the plant defense and the fungal attack, prevailing
the endophytism [108].
Ren et al. [109] performed metabolomics analysis by LC-MS of wheat infected
and non-infected with Tilletia controversa Kühn, the causal agent of the wheat
dwarf bunt, which has led to yield losses between 70% and 80% in wheat growing
areas worldwide [109, 110]. A total of 62 differential metabolites of different com-
pound classes were annotated. Furthermore, it was identified that eight metabolic
pathways were activated during the infection, including the isoquinoline alkaloid
biosynthesis, induced as a plant response to the phytopathogen invasion or also
activated in situations of biotic stresses (Fig. 1.2c) [109, 111]. Isoquinoline alka-
loids may participate in plant defense against microorganisms through virulence
factors inhibition of phytopathogens or due to its cytotoxic properties, being these
alkaloids accumulated in different plant parts [112, 113].

1.8 Plant Resistance

Plants are susceptible to several types of biological stress caused mainly by micro-
organisms, such as viruses, bacteria and fungi, which affect the growth and develop-
ment of agricultural crops [114]. On the other hand, plants have developed different
defense strategies in response to phytopathogenic stresses, including genetic modi-
fications and metabolic alterations [115]. The regulation of phytohormones and the
production of compounds that eliminate or inactivate the growth capacity of the
phytopathogen are directly involved in the defense mechanisms and increase of
plant resistance [116, 117]. Thus, understanding these plant mechanisms against
phytopathogen invasions is fundamental for the development of resistant plants and,
consequently, for the increase of agricultural productivity [118].
1 Unveiling Chemical Interactions Between Plants and Fungi Using Metabolomics… 11

Some alternatives reported in the literature for increase the Induced Resistance
(IR) in plants against pathogens can be done through Systemic Acquired Resistance
(SAR) and Induced Systemic Resistance (ISR), which are two biochemical path-
ways used in plant defense [119]. SAR occurs when the plant is attacked by the
pathogen and is dependent on hormones signaling, such as the salicylic acid [120],
followed by the production of compounds that play an active role in the plant’s
defense [121]. Therefore, SAR is characterized by increasing the salicylic acid con-
centration, increase the expression of pathogenesis-related protein genes such as the
NahG, a salicylate hydroxylase encoding gene [119, 120]. While ISR, happens
when the plant is colonized by microbial agents [119], including endophytic fungi
that improve plant resistance and are used as biocontrol agents against a wide spec-
trum of pathogens. ISR is also dependent on signaling compounds, mainly jasmonic
acid and ethylene [122–124].
An example of application using untargeted metabolomics in investigations of
resistant plants is shown in Fig. 1.1e. Kasote et al. [125] evaluated the hormonal
response and production of metabolites before and after infection caused by
Fusarium oxysporum f. sp. niveum (FON), monitoring the progression of wilting
disease in watermelon varieties that are resistant and susceptible to the disease. In
the study, hormones, melatonin, phenolic acids and amino acid profiles were identi-
fied in the leaves of the plant. The authors consider that jasmonic acid-isoleucine,
methyl jasmonate, melatonin and lysine may have important roles in the develop-
ment of defense responses against the pathogen FON, and indole-3-acetic acid may
be considered a potential biomarker of infection [125].
Li et al. [126] carried out a metabolomics study based on high-performance liq-
uid chromatography and tandem mass spectrometry (UHPLC-MS/MS) to evaluate
postharvest kiwi fruit resistance strategies during treatment with indole-3-acetic
acid (IAA) against Botrytis cinerea infection. The authors reported that the resis-
tance induced in kiwifruit by IAA treatment is directly related to the activation of
phenylpropanoids, including flavonoids and phenols, terpenoids, carbohydrate
metabolism, and hormonal signaling means. These results may reveal the mecha-
nism of IAA-induced resistance in kiwifruit and provide a new theoretical basis for
the safe and efficient control of postharvest diseases in kiwifruit [126].
Recently, Laupheimer et al. [127] in their research on GC-MS-based metabolo-
mics, analyzed the behavior of VOCs produced by Hordeum vulgare L. (barley)
after infection by Blumeria hordei. In this study, the authors suggest VOCs as
resistance-­inducing factors in barley. Green leaf VOCs in the barley-pathogen inter-
action revealed (Z)-3-hexenyl acetate (Z3HAC) as the major compound in the vola-
tile profile, which is a key compound in the defense against barley pathogens.
Furthermore, a mixture of VOCs and Z3HAC induced, for example, accumulation
of chlorophyll, linolenic acid and linolenate-conjugated lipids, as well as defense-­
related secondary metabolites such as hordatines in barley plants. Therefore, it
revealed the production of defense-active compounds as a barley resistance response
to the pathogen Blumeria hordei [127].
A metabolomics approach based on NMR and LC-MS was recently reported by
Fernandes et al. [128], where two citrus species and the role of metabolites of the
12 J. G. de Moraes Pontes et al.

coumarin class involved in the resistance of Citrus latifolia to black spot disease
(CBS) caused by Phyllosticta citricarpa were investigated. In the study, the authors
observed that the species Citrus limon (lemon) is susceptible to the disease,
while for Citrus latifolia (Tahiti lime) no damage was observed, assuming resis-
tance to the disease. Comparison of metabolic profiles showed that certain phenolic
compounds were strongly induced in plant defense against the pathogen, including
six coumarins: 5,7-dimethoxycoumarin (A), 8-methoxypsoralen (B),
7-­geranyloxycoumarin (C), 5-geranyloxypsoralen (D), psoralen (E), and 5-­geranyloxy-
7-methoxycoumarin (F) as shown in Fig. 1.3.
These isolated molecules and a mixed fraction were tested against P. citricarpa,
where no inhibition of isolated coumarins was observed, while for the fraction a
high antifungal activity was observed against P. citricarpa. A complementary in
silico study of the five coumarins was performed and each chemical structure was
evaluated individually. The data highlights the potential to use coumarin mixtures as
potential bio-insecticides for the control of P. citricarpa [128].
Metabolomics combined with analytical techniques is a powerful tool to be used
in investigations of the metabolome of resistant plants, since this technology allows
the simultaneous detection of a wide range of compounds through qualitative and
quantitative analysis of metabolites [93, 129]. Several metabolomic studies have
been published recently, where they report alterations of primary and/or secondary
metabolites in plant tissues in response to biotic stresses [130–135].

Fig. 1.3 Chemical structures of six coumarins, which the mixed fraction exhibited antifungal
activity against P. citricarpa
1 Unveiling Chemical Interactions Between Plants and Fungi Using Metabolomics… 13

1.9 Conclusions

Metabolomics approaches have been used in a variety of fungal-plant interactions

research, which makes possible the detection of biomarkers [57], secondary metab-
olites [107], virulence factors [136], signaling compounds [98], antimicrobial com-
pounds [102], and the identification of fungal species [86]. Despite its wide
applicability, metabolomics research in fungal-plant interactions still faces many
challenges such as: the identification (level-1) of unknown metabolites, since often
the isolation of a specific metabolite is an expensive and/or complicated process;
construction of databases and data standardization; analytical techniques covering
the detection of different metabolite classes; and the discrimination between the
metabolites produced by the pathogen or by the host [14, 93].
Although there are inherent difficulties in metabolomics research, it is consid-
ered a fundamental branch of systems biology [137]. Furthermore, metabolomics
has directly helped in the development of agricultural tools, which could be used in
the field [79, 138, 139].

Acknowledgements The authors acknowledge to the Coordenação de Aperfeiçoamento de

Pessoal de Nível Superior − Brasil (CAPES) − Finance Code 001, National Council for Scientific
and Technological Development (CNPq, Brazil) 142600/2021-0, Fundação de Amparo à Pesquisa
no Estado de São Paulo [FAPESP Grant number 2019/06359-7, 2021/00728-0; 2022/02992-0],
and Serrapilheira Institute for the Grant number Serra-1912-31626.

References

1. Fiehn O (2002) Metabolomics – the link between genotypes and phenotypes. Plant Mol Biol
48(1–2):155–171
2. Idle JR, Gonzalez FJ (2007) Metabolomics. Cell Metab 6(5):348–351
3. Oliver SG, Winson MK, Kell DB, Baganz F (1998) Systematic functional analysis of the
yeast genome. Trends Biotechnol 16(9):373–378
4. Pontes JGM, Brasil AJM, Cruz GCF, Souza RN, Tasic L (2017) 1H NMR metabolomic pro-
filing of human and animal blood serum samples. In: Guest PC (ed) Multiplex biomarker
techniques. Methods in molecular biology, vol 1546. Humana, New York, pp 275–282
5. Chen F, Ma R, Chen XL (2019) Advances of metabolomics in fungal pathogen–plant interac-
tions. Metabolites 9(8):169
6. Fiehn O (2001) Combining genomics, metabolome analysis, and biochemical modelling to
understand metabolic networks. Comp Funct Genomics 2(3):155–168
7. Pontes JGM, Brasil AJM, Cruz GCF, Souza RN, Tasic L (2017) NMR-based metabolomics
strategies: plants, animals and humans. Anal Methods 9(7):1078–1096
8. Marshall DD, Powers R (2017) Beyond the paradigm: Combining mass spectrometry and
nuclear magnetic resonance for metabolomics. Prog Nucl Magn Reson Spectrosc 100:1–16
9. Labine LM, Simpson MJ (2020) The use of nuclear magnetic resonance (NMR) and mass
spectrometry (MS)–based metabolomics in environmental exposure assessment. Curr Opin
Environ Sci Health 15:7–15
14 J. G. de Moraes Pontes et al.

10. Bauermeister A, Mannochio-Russo H, Costa-Lotufo LV, Jarmusch AK, Dorrestein PC (2022)

Mass spectrometry-based metabolomics in microbiome investigations. Nat Rev Microbiol
20:143–160
11. Zamkovaya T, Foster JS, Crécy-Lagard V, Conesa A (2021) A network approach to elucidate
and prioritize microbial dark matter in microbial communities. ISME J 15:228–244
12. Blacutt A, Ginnan N, Dang T, Bodaghi S, Vidalakis G, Ruegger P et al (2020) An in
vitro pipeline for screening and selection of Citrus- associated microbiota with poten-
tial anti-“Candidatus Liberibacter asiaticus” properties. Appl Environ Microbiol
86(8):e02883–e02819
13. Gupta S, Schillaci M, Roessner U (2022) Metabolomics as an emerging tool to study plant–
microbe interactions. Emerg Top Life Sci 6(2):175–183
14. Peters K, Worrich A, Weinhold A, Alka O, Balcke G, Birkemeyer C et al (2018) Current chal-
lenges in plant eco-metabolomics. Int J Mol Sci 19(5):1385
15. Liu R, Bao ZX, Zhao PJ, Li GH (2021) Advances in the study of metabolomics and metabo-
lites in some species interactions. Molecules 26(11):3311
16. Pontes JGM, Fernandes LS, Santos RV, Tasic L, Fill TP (2020) Virulence factors in the phy-
topathogen−host interactions: An overview. J Agric Food Chem 68(29):7555–7570
17. Ernst M, Silva DB, Silva RR, Vêncio RZN, Lopes NP (2014) Mass spectrometry in plant
metabolomics strategies: from analytical platforms to data acquisition and processing. Nat
Prod Rep 31(6):784–806
18. Schauer N, Steinhauser D, Strelkov S, Schomburg D, Allison G, Moritz T et al (2005)
GC-MS libraries for the rapid identification of metabolites in complex biological samples.
FEBS Lett 579(6):1332–1337
19. Halket JM, Waterman D, Przyborowska AM, Patel RKP, Fraser PD, Bramley PM (2005)
Chemical derivatization and mass spectral libraries in metabolic profiling by GC/MS and LC/
MS/MS. J Exp Bot 56(410):219–243
20. Rai AK, Khan S, Kumar A, Dubey BK, Lal RK, Tiwari A et al (2023) Comprehensive metab-
olomic fingerprinting combined with chemometrics identifies species- and variety-specific
variation of medicinal herbs: an Ocimum study. Metabolites 13(1):122
21. Wold S, Esbensen K, Geladi P (1987) Principal component analysis. Chemom Intell Lab Syst
2(1–3):37–52
22. Barker M, Rayens W (2003) Partial least squares for discrimination. J Chemom 17(3):166–173
23. Nocairi H, Qannari EM, Vigneau E, Bertrand D (2005) Discrimination on latent compo-
nents with respect to patterns. Application to multicollinear data. Comput Stat Data Anal
48(1):139–147
24. Wold S, Sjöström M (1977) SIMCA: a method for analyzing chemical data in terms of simi-
larity and analogy. In: Rowalski BR (ed) Chemometrics: theory and application, ACS sympo-
sium series. American Chemical Society, Washington, DC, pp 243–282
25. Pang Z, Chong J, Zhou G, Lima Morais DA, Chang L, Barrette M et al (2021) MetaboAnalyst
5.0: narrowing the gap between raw spectra and functional insights. Nucleic Acids Res
49(W1):W388–W396
26. Baldauf NA, Rodriguez-Romo RA, Männig A, Yousef AE, Rodriguez-Saona LE (2007)
Effect of selective growth media on the differentiation of Salmonella enterica serovars by
Fourier-Transform Mid-Infrared Spectroscopy. J Microbiol Methods 68:106–114
27. Khanmohammadi M, Garmarudi AB, Ghasemi K, Jaliseh HK, Kaviani A (2009) Diagnosis
of colon cancer by attenuated total reflectance-fourier transform infrared microspectroscopy
and soft independent modeling of class analogy. Med Oncol 26:292–297
28. Sumner LW, Amberg A, Barrett D, Beale MH, Beger R, Daykin CA et al (2007) Proposed
minimum reporting standards for chemical analysis: Chemical Analysis Working Group
(CAWG) Metabolomics Standards Initiative (MSI). Metabolomics 3(3):211–221
29. Wang M, Carver JJ, Phelan VV, Sanchez LM, Garg N, Peng Y et al (2016) Sharing and com-
munity curation of mass spectrometry data with Global Natural Products Social Molecular
Networking. Nat Biotechnol 34:828–837
1 Unveiling Chemical Interactions Between Plants and Fungi Using Metabolomics… 15

30. Nothias LF, Petras D, Schmid R, Dührkop K, Rainer J, Sarvepalli A et al (2020) Feature-­
based molecular networking in the GNPS analysis environment. Nat Methods 17:905–908
31. Horai H, Arita M, Kanaya S, NiheiY IT, Suwa K et al (2010) MassBank: a public repository
for sharing mass spectral data for life sciences. J Mass Spectrom 45(7):703–714
32. Ruttkies C, Schymanski EL, Wolf S, Hollender J, Neumann S (2016) MetFrag relaunched:
incorporating strategies beyond in silico fragmentation. J Cheminformatics 8:3
33. Wishart DS, Tzur D, Knox C, Eisner R, Guo AC, Young N et al (2007) HMDB: the Human
Metabolome Database. Nucleic Acids Res 35(Database issue):D521–D526
34. Dictionary of Natural Products CHEMnetBASE (2000) Available on: https://dnp.chemnet-
base.com. CRC Press/Taylor & Francis Group
35. FungalMet Database (2016) Available on: http://www.fungalmet.org/it/. Copyright by
Fungalmet
36. Jewison T, Knox C, Neveu V, Djoumbou Y, Guo AC, Lee J et al (2012) YMDB: the yeast
metabolome database. Nucleic Acids Res 40(Database issue):D815–D820
37. Saito Y, Tsuchida H, Matsumoto T, Makita Y, Kawashima M, Kikuchi J et al (2018)
Screening of fungi for decomposition of lignin-derived products from Japanese cedar. J
Biosci Bioeng 126(5):573–579
38. Piotto M, Saudek V, Sklenár V (1992) Gradient-tailored excitation for single-quantum NMR
spectroscopy of aqueous solutions. J Biomol NMR 2:661–665
39. Ogg RJ, Kingsley PB, Taylor JS (1994) WET, a T1- and B1-insensitive water-suppression
method for in vivo localized 1H NMR spectroscopy. J Magn Reson Ser B 104(1):1–10
40. Giraudeau P, Silvestre V, Akoka S (2015) Optimizing water suppression for quantitative
NMR-based metabolomics: a tutorial review. Metabolomics 11:1041–1055
41. Mazzei P, Vinale F, Woo SL, Pascale A, Lorito M, Piccolo A (2016) Metabolomics by pro-
ton high-resolution magic-angle-spinning nuclear magnetic resonance of tomato plants
treated with two secondary metabolites isolated from Trichoderma. J Agric Food Chem
64(18):3538–3545
42. Aiello F, Gerretzen J, Simons MG, Davies AN, Dani P (2021) A multivariate approach to
investigate the NMR CPMG pulse sequence for analysing low MW species in polymers.
Magn Reson Chem 59(2):172–186
43. Meiboom S, Gill D (1958) Modified spin-echo method for measuring nuclear relaxation
times. Rev Sci Instrum 29:688–691
44. Dona AC, Kyriakides M, Scott F, Shephard EA, Varshavi D, Veselkov K et al (2016) A guide
to the identification of metabolites in NMR-based metabonomics/metabolomics experiments.
Comput Struct Biotechnol J 14:135–153
45. Ulrich EL, Akutsu H, Doreleijers JF, Harano Y, Ioannidis YE, Lin J et al (2008)
BioMagResBank. Nucleic Acids Res 36(Database issue):D402–D408
46. Misra D, Bajpai U (2009) Metabolite characterization in serum samples from normal healthy
human subjects by 1H and 13C NMR spectroscopy. Bull Chem Soc Ethiop 23(2):211–221
47. Verpoorte R, Choi YH, Kim HK (2007) NMR-based metabolomics at work in phytochemis-
try. Phytochem Rev 6:3–14
48. Fulmer GR, Miller AJM, Sherden NH, Gottlieb HE, Nudelman A, Stoltz BM et al (2010)
NMR chemical shifts of trace impurities: Common laboratory solvents, organics, and gases in
deuterated solvents relevant to the organometallic chemist. Organometallics 29(9):2176–2179
49. Babij NR, McCusker EO, Whiteker GT, Canturk B, Choy N, Creemer LC et al (2016) NMR
chemical shifts of trace impurities: Industrially preferred solvents used in process and green
chemistry. Org Process Res Dev 20(3):661–667
50. Strimbu K, Tavel JA (2010) What are biomarkers? Curr Opin HIV AIDS 5(6):463–466
51. Tamhane M, Cabrera-Ghayouri S, Abelian G, Viswanath V (2019) Review of biomarkers in
ocular matrices: Challenges and opportunities. Pharm Res 36:40
52. Quintero M, Stanisic D, Cruz G, Pontes JGM, Costa TBBC, Tasic L (2019) Metabolomic bio-
markers in mental disorders: bipolar disorder and schizophrenia. In: Guest PC (ed) Reviews
16 J. G. de Moraes Pontes et al.

on biomarker studies in psychiatric and neurodegenerative disorders. Springer, Cham/New

York, pp 271–293
53. Belinato JR, Costa CP, Almeida A, Rocha SM, Augusto F (2021) Mapping Aspergillus niger
metabolite biomarkers for in situ and early evaluation of table grapes contamination. Foods
10(11):2870
54. Kumaraswamy KG, Kushalappa AC, Choo TM, Dion Y, Rioux S (2011) Mass spectrometry
based metabolomics to identify potential biomarkers for resistance in barley against fusarium
head blight (Fusarium graminearum). J Chem Ecol 37(8):846–856
55. Gavaghan CL, Li JV, Hadfield ST, Hole S, Nicholson JK, Wilson ID et al (2011) Application
of NMR-based metabolomics to the investigation of salt stress in maize (Zea mays).
Phytochem Anal 22(3):214–224
56. Camañes G, Scalschi L, Vicedo B, González-Bosch C, García-Agustín P (2015) An untar-
geted global metabolomic analysis reveals the biochemical changes underlying basal
resistance and priming in Solanum lycopersicum, and identifies 1-methyltryptophan as a
metabolite involved in plant responses to Botrytis cinerea and Pseudomonas syringae. Plant
J 84(1):125–139
57. Castaño JD, Muñoz-Muñoz N, Kim YM, Liu J, Yang L, Schilling JS (2022) Metabolomics
highlights different life history strategies of white and brown rot wood-degrading fungi.
mSphere 7(6):e00545-22
58. Blanchette RA (1991) Delignification by wood-decay fungi. Annu Rev Phytopathol
29:381–398
59. Kölle M, Horta MAC, Nowrouslan M, Ohm RA, Phillip Benz J, Pilgård A (2020)
Degradative capacity of two strains of Rhodonia placenta: From phenotype to genotype.
Front Microbiol 11:1338
60. Bari E, Daryaei MG, Karim M, Bahmani M, Schimidt O, Woodward S et al (2019) Decay
of Carpinus betulus wood by Trametes versicolor – an anatomical and chemical study. Int
Biodeterior Biodegradation 137:68–77
61. Almeida ACA, Pontes JGM, Alvarenga GR, Finocchio H, Fill TP (2021) The sustainable
cycle of a new cacao-based bioplastic: from manufacturing to exploitable biodegradation
products. RSC Adv 11:29976–29985
62. Fones HN, Bebber DP, Chaloner TM, Kay WT, Steinberg G, Gurr SJ (2020) Threats to global
food security from emerging fungal and oomycete crop pathogens. Nat Food 1(6):332–342
63. Oliveira MCF, Mattos MC, Filho ER (2021) Tópicos em química microbiana: oportunidades
acadêmicas e biotecnológicas. Imprensa Universitária UFC, 256p
64. He DC, He MH, Amalin DM, Liu W, Alvindia DG, Zhan J (2021) Biological control of plant
diseases: An evolutionary and eco-economic consideration. Pathogens 10(10):1311
65. Luchi N, Ioos R, Santini A (2020) Fast and reliable molecular methods to detect fungal patho-
gens in woody plants. Appl Microbiol Biotechnol 104:2453–2468
66. Ahmed Y, Hubert J, Fourrier-Jeandel C, Dewdney MM, Aguayo J, Ioos R (2019) A set of
conventional and multiplex real-time PCR assays for direct detection of Elsinoë fawcettii,
E. australis, and Pseudocercospora angolensis in citrus fruits. Plant Dis 103(2):345–356
67. Bhat RG, Browne GT (2010) Specific detection of Phytophthora cactorum in diseased straw-
berry plants using nested polymerase chain reaction. Plant Pathol 59(1):121–129
68. Driciru P, Mugasa MC, Acidri R, Adriko J (2021) Development of loop-mediated isothermal
amplification (LAMP) assay for detection of Pseudocercospora angolensis in sweet orange.
bioRxiv:1–39
69. Gaikwad PN, Sharma V, Singh J, Sidhu GS, Singh H, Omar AA (2023) Biotechnological
advancements in Phytophthora disease diagnosis, interaction and management in citrus. Sci
Hortic 310:111739
70. Pontes JGMP, Vendramini PH, Fernandes LS, Souza FH, Pilau EJ, Eberlin MN et al (2020)
Mass spectrometry imaging as a potential technique for diagnostic of Huanglongbing disease
using fast and simple sample preparation. Sci Rep 10:13457
1 Unveiling Chemical Interactions Between Plants and Fungi Using Metabolomics… 17

71. Barbieri HB, Fernandes LS, Pontes JGM, Pereira AK, Fill TP (2023) An overview of the most
threating diseases that affect worldwide citriculture: main features, diagnose, and current
control strategies. Front Nat Product 2:1045364
72. Razzaq A, Wishart DS, Wani SH, Hameed MK, Mubin M, Saleem F (2022) Advances in
metabolomics-driven diagnostic breeding and crop improvement. Metabolites 12(6):511
73. Mashabela MD, Piater LA, Dubery AI, Tugizimana F, Mhlongo MI (2022) Rhizosphere tri-
partite interactions and PGPR-mediated metabolic reprogramming towards ISR and plant
priming: A metabolomics review. Biology 11:346
74. Wilson AD (2018) Applications of electronic-nose technologies for noninvasive early detec-
tion of plant, animal and human diseases. Chemosensors 6(4):45
75. Mohottige CUG, Howard A, Baird R, Mlsna T (2022) Volatile biomarkers for the early
stage disease diagnosis of sweet potato Rhizopus soft tissue disease. ACS Agric Sci Technol
2(4):734–741
76. Hu W, Pan X, Li F, Dong W (2018) UPLC-QTOF-MS metabolomics analysis revealed the
contributions of metabolites to the pathogenesis of Rhizoctonia solani strain AG-1-IA. PLoS
One 13(2):e0192486
77. Hu Z, Chang X, Dai T, Li L, Liu P, Wang G et al (2019) Metabolic profiling to identify the
latent infection of strawberry by Botrytis cinerea. Evol Bioinforma 15:1–7
78. Steglińska A, Pielech-Przybylska K, Janas R, Grzesik M, Borowski S, Kręgiel D et al (2022)
Volatile organic compounds and physiological parameters as markers of potato (Solanum
tuberosum L.) infection with phytopathogens. Molecules 27(12):3708
79. Dai T, Chang X, Hu Z, Liang L, Sun M, Liu P et al (2019) Untargeted metabolomics based
on GC-MS and chemometrics: a new tool for the early diagnosis of strawberry anthracnose
caused by Colletotrichum theobromicola. Plant Dis 103(10):2541–2547
80. Pancoro A, Karima E, Apriyanto A, Effendi Y (2022) 1H NMR metabolomics analysis of oil
palm stem tissue infected by Ganoderma boninense based on field severity indices. Sci Rep
12:21087
81. Hsu MJ, Chang YCI, Hsueh HM (2014) Biomarker selection for medical diagnosis using the
partial area under the ROC curve. BMC Res Notes 7:25
82. Frisvad JC, Andersen B, Thrane U (2008) The use of secondary metabolite profiling in che-
motaxonomy of filamentous fungi. Mycol Res 112(2):231–240
83. Lee S, Oh DG, Lee S, Kim GR, Lee JS, Son YK et al (2015) Chemotaxonomic metabo-
lite profiling of 62 indigenous plant species and its correlation with bioactivities. Molecules
20(11):19719–19734
84. Santos VS, Macedo FA, Vale JS, Silva DB, Carollo CA (2017) Metabolomics as a tool for
understanding the evolution of Tabebuia sensu lato. Metabolomics 13:72
85. Sofrenić I, Anđelković B, Gođevac D, Ivanović S, Simić K, Ljujić J et al (2023) Metabolomics
as a potential chemotaxonomical tool: Application on the selected Euphorbia species grow-
ing wild in Serbia. Plan Theory 12(2):262
86. Aliferis KA, Cubeta MA, Jabaji S (2013) Chemotaxonomy of fungi in the Rhizoctonia
solani species complex performing GC/MS metabolite profiling. Metabolomics 9(Suppl
1):S159–S169
87. Ernst M (2013) Metabolomics in plant taxonomy: the arnica model. Masters Dissertation,
University of São Paulo, Faculty of Pharmaceutical Sciences of Ribeirão Preto, 209p
88. Misawa T, Kurose D (2019) Anastomosis group and subgroup identification of Rhizoctonia
solani strains deposited in NARO Genebank, Japan. J Gen Plant Pathol 85:282–294
89. Aszyk J, Biliński H, Namiesnik J, Kot-Wasik A (2018) Main strategies, analytical trends and
challenges in LC-MS and ambient mass spectrometry – based metabolomics. Trends Anal
Chem 108:278–295
90. Gullberg J, Jonsson P, Nordström A, Sjöström M, Moritz T (2004) Design of experiments: an
efficient strategy to identify factors influencing extraction and derivatization of Arabidopsis
thaliana samples in metabolomic studies with gas chromatography/mass spectrometry. Anal
Biochem 331(2):283–295
18 J. G. de Moraes Pontes et al.

91. Bartz FE, Glassbrook NJ, Danehower DA, Cubeta MA (2012) Elucidating the role of the
phenylacetic acid metabolic complex in the pathogenic activity of Rhizoctonia solani anasto-
mosis group 3. Mycologia 104(4):793–803
92. Willetts HJ (1971) The survival of fungal sclerotia under adverse environmental conditions.
Biol Rev 46:387–407
93. Castro-Moretti FR, Gentzel IN, Mackey D, Alonso AP (2020) Metabolomics as an emerging
tool for the study of plant–pathogen interactions. Metabolites 10(2):52
94. Rojo E, Solano R, Sánchez-Serrano JJ (2003) Interactions between signaling compounds
involved in plant defense. J Plant Growth Regul 22:82–98
95. Zaynab M, Fatima M, Abbas S, Sharif Y, Umair M, Zafar MH et al (2018) Role of secondary
metabolites in plant defense against pathogens. Microb Pathog 124:198–202
96. Mishra AK, Baek KH (2021) Salicylic acid biosynthesis and metabolism: A divergent path-
way for plants and bacteria. Biomol Ther 11(5):705
97. Ruan J, Zhou Y, Zhou M, Yan J, Khurshid M, Weng W et al (2019) Jasmonic acid signaling
pathway in plants. Int J Mol Sci 20(10):2479
98. Cheng C, Yan CY, Qi CT, Zhao XL, Liu LX, Guo YY et al (2023) Metabolome and transcrip-
tome analysis of postharvest peach fruit in response to fungal pathogen Monilinia fructicola
infection. LWT 173:114301
99. Xu Y, Wei J, Wei Y, Han P, Dai K, Zou X et al (2021) Tea tree oil controls brown rot in
peaches by damaging the cell membrane of Monilinia fructicola. Postharvest Biol Technol
175:111474
100. Dombrecht B, Xue GP, Sprague SJ, Kirkegaard JA, Ross JJ, Reid JB et al (2007) MYC2
differentially modulates diverse jasmonate-dependent functions in Arabidopsis. Plant Cell
19:2225–2245
101. Tiku AR (2018) Antimicrobial compounds and their role in plant defense. In: Singh A, Singh
IK (eds) Molecular aspects of plant-pathogen interaction. Springer, Singapore, pp 283–307
102. Hu X, Puri KD, Gurung S, Klosterman SJ, Wallis CM, Britton M et al (2019) Proteome
and metabolome analyses reveal differential responses in tomato -Verticillium dahliae-­
interactions. J Proteome 207:103449
103. Vermeulen JP, Hanika K, Thomma BPHJ, Bai Y, Schouten HJ (2022) Scarcity of major
resistance genes against Verticillium wilt caused by Verticillium dahliae. Plant Breed
141(5):681–694
104. Yadav V, Wang Z, Wei C, Amo A, Ahmed B, Yang X et al (2020) Phenylpropanoid pathway
engineering: An emerging approach towards plant defense. Pathogens 9(4):312
105. Riaz U, Kharal MA, Murtaza G, Zaman Q, Javaid S, Ahmed H et al (2019) Prospective
roles and mechanisms of caffeic acid in counter plant stress: a mini review. Pak J Agric Res
32(1):8–19
106. Toffolatti SL, Maddalena G, Passera A, Casati P, Bianco PA, Quaglino F (2021) Role of
terpenes in plant defense to biotic stress. In: Jogaiah S (ed) Biocontrol agents and second-
ary metabolites: Applications and immunization for plant growth and protection. Woodhead
Publishing, Cambridge, pp 401–417
107. Sebastiana M, Gargallo-Garriga A, Sardans J, Pérez-Trujillo M, Monteiro F, Figueiredo A
et al (2021) Metabolomics and transcriptomics to decipher molecular mechanisms underly-
ing ectomycorrhizal root colonization of an oak tree. Sci Rep 11:8576
108. Araújo EV, Pontes JGM, Silva SN, Amaral LS, Fill TP (2020) The chemical warfare
involved in endophytic microorganisms-plant associations. In: Kumar A, Radhakrishnan
EK (eds) Microbial endophytes: functional biology and applications. Woodhead Publishing,
Cambridge, pp 125–159
109. Ren Z, Fang M, Muhae-Ud-Din G, Gao H, Yang Y, Liu T et al (2021) Metabolomics analysis
of grains of wheat infected and noninfected with Tilletia controversa Kühn. Sci Rep 11:18876
110. Muhae-Ud-Din G, Chen D, Liu T, Chen W, Gao L (2020) Methyljasmonate and salicylic acid
contribute to the control of Tilletia controversa Kühn, causal agent of wheat dwarf bunt. Sci
Rep 10:19175
1 Unveiling Chemical Interactions Between Plants and Fungi Using Metabolomics… 19

111. Lu M, Yu M, Lian J, Wang Q, He Z, Feng Y et al (2021) Physiological and metabolomics

responses of two wheat (Triticum aestivum L.) genotypes differing in grain cadmium accu-
mulation. Sci Total Environ 769:145345
112. Othman L, Sleiman A, Abdel-Massih RM (2019) Antimicrobial activity of polyphenols and
alkaloids in middle eastern plants. Front Microbiol 10:911
113. Ziegler J, Facchini PJ (2008) Alkaloid biosynthesis: Metabolism and trafficking. Annu Rev
Plant Biol 59:735–769
114. Tang Y, He G, He Y, He T (2022) Plant resistance to fungal pathogens: bibliometric analysis
and visualization. Toxics 10:624
115. Sood M, Kapoor D, Kumar V, Kalia N, Bhardwaj R, Sidhu GPS et al (2021) Mechanisms of
plant defense under pathogen stress: a review. Curr Protein Pept Sci 22(5):376–395
116. Yang YXJ, Ahammed G, Wu C, Fan SY, Zhou YH (2015) Crosstalk among jasmonate, salicy-
late and ethylene signaling pathways in plant disease and immune responses. Curr Protein
Pept Sci 16(5):450–461
117. Arbona V, Gómez-Cadenas A (2016) Metabolomics of disease resistance in crops. Curr
Issues Mol Biol 19(1):13–30
118. Saberi Riseh R, Vazvani MG, Ebrahimi-Zarandi M, Skorik YA (2022) Alginate-induced dis-
ease resistance in plants. Polymers 14(4):661
119. Walling LL (2001) Induced resistance: from the basic to the applied. Trends Plant Sci
6(10):445–447
120. Ding P, Ding Y (2020) Stories of salicylic acid: a plant defense hormone. Trends Plant Sci
25(6):549–565
121. Moreno-Delafuente A, Garzo E, Fereres A, Viñuela E, Medina P (2020) Effects of a sali-
cylic acid analog on Aphis gossypii and its predator Chrysoperla carnea on melon plants.
Agronomy 10:1830
122. Bazioli JM, Belinato JR, Costa JH, Akiyama DY, Pontes JGM, Kupper KC, Augusto F,
Carvalho JE, Fill TP (2019) Biological control of citrus postharvest phytopathogens. Toxins
11(8):460
123. Wasternack C (2007) Jasmonates: an update on biosynthesis, signal transduction and action
in plant stress response, growth and development. Ann Bot 100(4):681–697
124. Hossain M, Sultana F, Hyakumachi M (2017) Role of ethylene signalling in growth and sys-
temic resistance induction by the plant growth-promoting fungus Penicillium viridicatum in
Arabidopsis. J Phytopathol 165:432–441
125. Kasote DM, Jayaprakasha GK, Ong K, Crosby KM, Patil BS (2020) Hormonal and metabo-
lites responses in Fusarium wilt-susceptible and -resistant watermelon plants during plant-­
pathogen interactions. BMC Plant Biol 20:481
126. Li ZX, Yang S, Wang X, Liao QH, Zhang WL, Liu J et al (2023) Widely targeted metabo-
lomics analysis reveals the effect of exogenous auxin on postharvest resistance to Botrytis
cinerea in kiwifruit (Actinidia chinensis L.). Postharvest Biol Technol 195:112129
127. Laupheimer S, Kurzweil L, Proels R, Unsicker SB, Stark TD, Dawid C et al (2023) Volatile-­
mediated signalling in barley induces metabolic reprogramming and resistance against the
biotrophic fungus Blumeria hordei. Plant Biol 25(1):72–84
128. Fernandes HP, Salomé-Abarca LF, Pereira RG, Seibert JB, Silva-Junior GJ, da Silva MFGF
et al (2022) Metabolomic investigation of Citrus latifolia and the putative role of coumarins
in resistance to black spot disease. Front Mol Biosci 9:934401
129. Mhlongo MI, Piater LA, Madala NE, Labuschagne N, Dubery IA (2018) The chemistry of
plant–microbe interactions in the rhizosphere and the potential for metabolomics to reveal
signaling related to defense priming and Induced Systemic Resistance. Front Plant Sci 9:112
130. Nam MH, Bang E, Kwon TY, Kim Y, Kim EH, Cho K et al (2015) Metabolite profiling of
diverse rice germplasm and identification of conserved metabolic markers of rice roots in
response to long-term mild salinity stress. Int J Mol Sci 16(9):21959–21974
131. Lacerda JWF, Siqueira KA, Vasconcelos LG, Belletti BS, Dall’Oglio E, Sousa Junior PT et al
(2021) Metabolomic analysis of Combretum lanceolatum plants pnteraction with Diaporthe
phaseolorum and Trichoderma spirale endophytic fungi through 1H-NMR. Chem Biodivers
18(10):e2100350
20 J. G. de Moraes Pontes et al.

132. Sidiq Y, Nakano M, Mori Y, Yaeno T, Kimura M, Nishiuchi T (2021) Nicotinamide effec-
tively suppresses fusarium head blight in wheat plants. Int J Mol Sci 22(6):2968
133. Lv Y, Wang F, Chen H, Zhang T, Yan J, Hu Y (2022) Unraveling arbuscular mycorrhizal
fungi-induced resistance of purple branch rose (Rosa rugosa ‘Zizhi’) to Lymantria dispar
based on metabolomics. Biol Control 172:104971
134. Kisiriko M, Anastasiadi M, Terry LA, Yasri A, Beale MH, Ward JL (2021) Phenolics from
medicinal and aromatic plants: Characterisation and potential as biostimulants and biopro-
tectants. Molecules 26:6343
135. Zhou S, Zhang YK, Kremling KA, Ding Y, Bennett JS, Bae JS et al (2019) Ethylene signaling
regulates natural variation in the abundance of antifungal acetylated diferuloylsucroses and
Fusarium graminearum resistance in maize seedling roots. New Phytol 221(4):2096–2111
136. Stuart AKC, Furuie JL, Cataldi TR, Stuart RM, Zawadneak MAC, Labate CA et al (2022)
Fungal consortium of two Beauveria bassiana strains increases their virulence, growth, and
resistance to stress: A metabolomic approach. PLoS One 17(7):e0271460
137. Carrera FP, Noceda C, Maridueña-Zavala MG, Cevallos-Cevallos JM (2021) Metabolomics,
a powerful tool for understanding plant abiotic stress. Agronomy 11(5):824
138. Santiago KAA, Edrada-Ebel R, dela Cruz TEE, Cheow YL, Ting ASY (2021) Biodiscovery
of potential antibacterial diagnostic metabolites from the endolichenic fungus Xylaria venus-
tula using LC–MS-based metabolomics. Biology 10(3):191
139. Villate A, Nicolas MS, Gallastegi M, Aulas PA, Olivares M, Usobiaga A et al (2021) Review:
metabolomics as a prediction tool for plants performance under environmental stress. Plant
Sci 303:110789
Chapter 2
Metabolomics Applied to Cyanobacterial
Toxins and Natural Products

Márcio Barczyszyn Weiss, Rhuana Valdetário Médice,

Fernanda Rios Jacinavicius, Ernani Pinto, and Camila Manoel Crnkovic

2.1 Introduction

Cyanobacteria are photosynthetic prokaryotes that occur as unicellular organisms,

in colonial forms, or composing multicellular filaments [1, 2]. In 2013, more than
250 genera and 2700 species of cyanobacteria had already been cataloged [3], and
this number continues to increase by an average of 15 new species per year. Their
first fossil records date back to 3.5 billion years and, during this long evolutionary
history, they were able to colonize almost all Earth’s ecosystems [4–6].
Cyanobacteria can be found forming mats on the sea bottom, as freshwater
blooms, colonizing rocks and sediments, in brackish water ponds and even surviv-
ing unfavorable conditions, such as draught, high salinity, and high temperature
[7]. Cyanobacteria may also live in symbiotic relationships with fungi, plants, and
animals, or as independent autotrophs [8]. They are producers of secondary
metabolites that shape ecosystems and have important biological activities

Authors M. B. Weiss and R. V. Médice have equally contributed to this chapter.

M. B. Weiss · C. M. Crnkovic (*)

School of Pharmaceutical Sciences, Department of Biochemical and Pharmaceutical
Technology, University of São Paulo, São Paulo, Brazil
e-mail: [email protected]
R. V. Médice · F. R. Jacinavicius
School of Pharmaceutical Sciences, Department of Clinical and Toxicological Analyses,
University of São Paulo, São Paulo, Brazil
E. Pinto
Centre for Nuclear Energy in Agriculture, Division of Tropical Ecosystem Functioning,
University of São Paulo, Piracicaba, Brazil

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 21

T. Pacheco Fill (ed.), Microbial Natural Products Chemistry, Advances in Experimental
Medicine and Biology 1439, https://doi.org/10.1007/978-3-031-41741-2_2
22 M. B. Weiss et al.

[9–11]. In this chapter, cyanobacterial secondary metabolites will be divided into

cyanotoxins and natural products according to their importance to toxicological
studies and drug discovery, respectively.
The proliferation of cyanobacterial harmful algal blooms (cyanoHABs) has
been reported in the scientific literature for more than 130 years [12] and, in
recent decades, their incidence and intensity have increased due to anthropogenic
activities and climate change [13]. The prevalence of cyanotoxins in cyanoHABs
causes several economic, environmental, and public health problems and need to
be considered for risk assessment [14]. Concerns regarding consequences to
human health and their degree of severity are associated with the route of expo-
sure, amount, and type of cyanotoxins. Oral is the main route of exposure. Low
concentrations of cyanotoxins can induce gastrointestinal warning signs, and
higher concentrations can cause liver, kidney, digestive, and neurological diseases
due to cytotoxicity [15]. Several occurrences of lethal poisoning have been associ-
ated with the ingestion of cyanotoxins by wild animals or domestic livestock [14].
Human intoxications caused by cyanotoxins have been reported in various coun-
tries, including the USA, Australia, England, Brazil, China, South Africa, Sweden,
and India [16–19]. It is important to note that not all cyanobacterial blooms are
toxic, and some are only toxic for a certain period of time due to the dominance
of specific strains [20]. Secondary metabolites produced by these microorgan-
isms, from various taxa and geographic origins, exhibit a great diversity of chemi-
cal structures [21–24]. According to Jones et al. (2021), more than 2000
cyanobacterial metabolites have been described. Such chemical diversity provides
opportunity for a variety of applications [9].
A promising future is reserved for natural products in drug discovery. Natural
products draw attention due to their special features in terms of scaffold diversity
and structural complexity [25]. Ensuring broad source diversity and enhancing
prospection approaches is imperative to successful natural product discovery [26].
Cyanobacteria represent an underexplored source of natural products with potential
for pharmaceutical applications. Cyanobacterial natural products have been shown
to display biological activities such as antibiotic activity, antifungal, antiviral, anti-
parasitic, anti-inflammatory, enzyme inhibition, and anticancer activity [9, 11, 22,
27, 28]. As a successful example, Dolastatin 10 is a peptidic natural product of
cyanobacterial origin [29] that inspired the development of the anticancer drug
brentuximab vedotin [30–32]. Modern chemical prospection approaches utilizing
metabolomics and genomics have expanded the frontiers of natural product discov-
ery from microbes, including cyanobacteria. In the following sections of this chap-
ter, we review selected articles highlighting the use of metabolomic approaches in
cyanobacterial toxicology and natural product discovery.
2 Metabolomics Applied to Cyanobacterial Toxins and Natural Products 23

2.2 Analysis of Cyanotoxins Using Metabolomics Approaches

Cyanobacteria are natural components of phytoplankton. As primary producers,

they play an important role in aquatic ecosystems [33]. However, there may be epi-
sodes in which an excessive growth of biomass occurs. These episodes are called
blooms or cyanoHABs, when toxic.
Although cyanoHABs can occur naturally in the environment, there is broad
consensus in the scientific community that the incidence and duration of cyano-
HABs increase worldwide due to anthropogenic activities and climate change [34,
35]. Some physical-chemical factors in reservoirs, such as abundant sunlight, high
temperatures, high nutrient load (mainly nitrogen and phosphorus), slow water flow,
and alkaline pH, can favor rapid reproduction and formation of cyanobacterial
blooms [36].
During blooms, cyanobacterial biomass can form a scum on the surface of reser-
voirs, increasing turbidity and shading submerged vegetation [37]. The degradation
of senescent biomass contributes to decrease dissolved oxygen concentrations in the
water, leading to the death of fish and other aquatic organisms [38]. CyanoHABs
cause generalized deterioration in water quality. They can produce compounds that
generate unpleasant taste and odor in water (i.e. methyl-isoborneal and geosmin),
impacting the recreational and scenic potential of a reservoir, and increasing treat-
ment costs for producing drinking water [33, 39].
In addition, cyanoHABs can produce a variety of cyanotoxins. Cyanotoxins are
toxic substances that can be classified as hepatotoxins (microcystins and nodular-
ins), cytotoxins (cylindrospermopsin), and neurotoxins (anatoxin-a, guanitoxin, and
saxitoxins) [40]. These bioactive compounds can cause liver, digestive, and neuro-
logical diseases [41] and pose a risk to livestock, wildlife, and human health [40].
There is also a hypothesis that some genera of cyanobacteria could produce the
neurotoxic nonproteinogenic amino acid β-N-methylamino-L-alanine, or BMAA,
which causes neurodegenerative disease. This topic is still controversial, and further
studies are necessary [42].
Microcystins are probably the most well studied class of cyanotoxins in the
world. Bishop et al. (1959) did the first chemical identification of a microcystin
from a Microcystis aeruginosa strain, hence the origin of the name [43]. They are
the most widely distributed cyanotoxins, and the majority of microcystin-producing
cyanobacteria are planktonic [14, 44].
Microcystins are chemically classified as cyclic heptapeptides. The general
structure of microcystins is cyclo-(d-Ala1-X2- d-MeAsp3-Z4-Adda5- d-γ-Glu6-­
Mdha7), in which X and Z are variable l-amino acids, d-MeAsp is d-erythro-β-­
methylaspartic acid and Mdha is N-methyldehydroalanine, respectively [9].
Microcystin-LR (1) for example, has the amino acids leucine (L) and arginine (R)
at positions X and Z respectively. The β-amino acid Adda, (2S,3S,4E,6E,8S,9S)-3-­
amino-­9-methoxy-2,6,8-trimethyl-10-phenyldeca-4,6-dienoic acid is a moiety typi-
cal of microcystins molecules. Structural variations in all seven amino acids have
been reported in microcystins, despite being more frequent substitutions in the two
24 M. B. Weiss et al.

variable l-amino (X and Z), and demethylation of amino acids in position 3 and/or
7 [44]. More than 320 structural variants of microcystins are known [45]. Their
molecular masses range from 800 to 2117 Da [9].

H2N
NH2
HO O N
O O

N N
H H O
O NH
HN O

NH
N O
O HN O
OH

O
1

The name of nodularins originates from the cyanobacterium Nodularia spumi-

gena [46]. Production of these molecules in planktonic species is limited to strains
of the genus Nodularia [45]. Nodularins have very similar structure to microcystins.
They are cyclic pentapeptides, and their chemical structure is identical in the amino
acids 1–4, cyclo-(d-MeAsp1-l-arginine2-Adda3-d-glutamate4-Mdhb5), and Mdhb is
2-(methylamino)-2-dehydrobutyric acid moieties (e.g. nodularin-R (2)) [46].
Nowadays, more than 16 structural variants of nodularins are known [45]. Their
molecular masses range from 700 to 1130 Da [9].

O
OH

O HN O

O NH O
O NH O
HO N NH2
N
H
O NH2
2

A hybrid pathway produces these two cyanotoxins through the action of high
molecular weight, multi-domain enzymes, called non-ribosomal peptide synthe-
tases (NRPS) and polyketide synthases (PKS). They are responsible for the synthe-
sis of unusual polyketide amino acids and, modification of amino acid residues, and
incorporation of seven amino acids in microcystins and five amino acids in nodula-
rins, followed by cyclization [47]. The biosynthetic gene cluster (BGC) comprises
9–10 genes for the microcystin (mcy) and 9 genes for the nodularins (nda) [48].
2 Metabolomics Applied to Cyanobacterial Toxins and Natural Products 25

All microcystins and nodularins are water-soluble. Their polarity is mainly due
to free carboxylic acids in the structure [49]. These cyanotoxins have high chemical
stability, remaining stable even after exposure to boiling temperature [50]. In natu-
ral environments, microcystins remain stable for up to 3 months [51]. They are
potent inhibitors of protein phosphatases PP1 and PP2A and phosphoprotein phos-
phatases PPP4, PPP5 [52]. In mammals, chronic exposure to low doses of microcys-
tins can promote renal toxicity and liver tumors [53, 54].
Cylindrospermopsins are cyclic guanidine alkaloids. Ohtani et al. (1992), char-
acterized the first cylindrospermopsin (3) from Raphidiopsis raciborskii strain (pre-
viously Cylindrospermopsis raciborskii) [55]. The biosynthesis of this cyanotoxin
involves genes encoding for amidinotransferase, NRPS, PKS, tailoring enzymes,
transport and regulation enzymes, and transposases [56]. Its BGC (aoa/cyr clusters)
comprises 11 genes and is highly conserved [56, 57]. Cylindrospermopsin (3) con-
sists of a tricyclic guanidine moiety combined with hydroxymethyl uracil ring. Five
structural variants are known [45, 58]. Cylindrospermopsins have a molecular mass
ranging from 319 to 415 Da [9]. Cylindrospermopsin is produced by various cyano-
bacteria in the Nostocales and Oscillatoriales orders isolated mainly from Europe
and Oceania [58–65]. Cylindrospermopsin is water-soluble and highly polar due to
its zwitterionic nature [66]. The stability of this molecule in natural environments
needs further clarification. In culture exposed to sunlight, cylindrospermospsin was
shown to have a half-life of 1.5 h [67]. The same study showed that it remained
stable after exposure to boiling temperature (100 °C, for 15 min), and degraded
slowly under temperatures ranging from 4 to 50 °C (4-week experiments) or in
aqueous solutions at pH 4, 7, and 10 (8-weeks experiment) [67]. Exposure to cylin-
drospermospsin causes hepatotoxicity, cytotoxicity and genotoxicity in mammalian
[68, 69].

O HN H
HN HN N+ O -
O
S
O N O
H H O
OH
3

Anatoxin-a (4) is a bicyclic alkaloid, originally named “very fast death factor
(VFDF)” [70]. Devlin et al. (1977), isolated and characterized the structure of 4
from a Dolichospermum flosaquae (previously Anabaena flosaquae) strain, hence
the origin of the name [71]. The biosynthesis of 4 has been deciphered recently. It is
produced from proline and acetate precursors by a pathway involving three
polyketide synthases (PKS) [72]. Its BGC comprises 10 genes (ana), and only part
is conserved when compared among different strains that produce these cyanotoxins
[73, 74]. Anatoxin-a (4) consists of a of 2-acetil-9-azabiciclo [4.2.1] non-2-eno
[71]. Anatoxin and analogs have a molecular mass ranging from 165 to 209 Da [9].
26 M. B. Weiss et al.

H O

HN

H
4

Twelve structural variants of 4 are known [9]. The coexistence of the different
anatoxin-a variants in the same strains has been reported [75]. Anatoxin-a has been
detected in freshwater bodies of North America, Europe, Australia, Oceania and
Asia, and is produced by several cyanobacteria genera, such as Dolichospermum
(previously Anabaena), Cuspidothrix (previously Aphanizomenon), Phormidium,
Cylindrospermum, and Rhapidiopsis (previously Cylindrospermopsis) [74–78]. The
stability of this molecule in natural environments needs further investigations.
Anatoxin-a (4) is unstable in water (half-life of 1–2 h); however, it has been effec-
tively extracted from cells using acidified solvents (water, methanol or a mixture of
both) [79]. In cultures exposed to abiotic factors, anatoxin-a degradation appears to
be related to strain and variant types [79, 80]. It is a potent postsynaptic depolarizing
agent, nicotinic agonist, and mimics the neurotransmitter acetylcholine [81].
Exposure to anatoxin-a (4) causes neurotoxicity [82].
Guanitoxin (5) is an alkaloid with guanidine and phosphate groups, originally
named anatoxin-a(S) “Salivary” [83]. Mahmood & Carmichael, characterized this
compound from Dolichospermum flosaquae strain (previously Anabaena flo-
saquae), proposing similar toxicology with anatoxin-a with the exception that 5
induces salivation. It has recently been renamed guanitoxin because it has no struc-
ture and actions mechanism resembling anatoxin-a [84]. Current knowledge on gua-
nitoxin biosynthesis is limited. The intermediates for the cyclic guanidine moiety
have been reported, and all carbons were shown to be derived from amino acids [85,
86]. The l-arginine precursor has been proposed for the guanidine group [87]. The
complete BGC of 5 has not yet been identified [64, 68]. Guanitoxin (5) has a molec-
ular mass of 252 Da [88]. This natural organophosphate consists of a cyclic
N-hydroxyguanine methyl phosphate ester moiety [85]. No structural variants of 5
are known. It has been detected sparsely in freshwater bodies of Canada, Denmark,
Portugal, USA, Scotland, and Brazil being produced by freshwater planktonic cya-
nobacteria genera, such as Dolichospermum, and Sphaerospermopsis [89–91]. The
stability of the molecule in natural environments needs further study. Guanitoxin (5)
decomposes rapidly in alkaline conditions; however, it is relatively stable under
neutral or acidic conditions (pH 1.5–5) [85, 92]. Guanitoxin has been effectively
extracted from cells using acidified water [92]. It is an irreversible anticholinester-
ase inhibitor [63, 65]. Exposure to guanitoxin (5) causes neurotoxicity [85].
2 Metabolomics Applied to Cyanobacterial Toxins and Natural Products 27

O
N
N P
O O
OH
H2N
5

Saxitoxins are natural alkaloids known as Paralytic Shellfish Poisons (PSP) [93].
Schantz et al. (1957) characterized the first saxitoxin (6) from a butter clam
(Saxidomus giganteus), hence the origin of the name [94]. Saxitoxins were later
discovered to be derived from a marine dinoflagellate symbiont Alexandrium
catenella (formerly Gonyaulax catenella) [95]. The first evidence of PSP in cyano-
bacterial strains was reported from an Aphanizomenon flosaquae bloom in 1969
[96]. Currently, these neurotoxins are known to be produced by freshwater cyano-
bacteria and marine dinoflagellates [97, 98]. Saxitoxin biosynthesis is initiated by a
type of polyketide synthase that uses arginines, acetyl-CoA, and methionine via
S-adenosylmethionine as precursors [99]. Additional enzymes include tailoring
enzymes, transporters, and saxitoxin-binding protein [99, 100]. The BGC (sxt) com-
prises genes encoding for 26 proteins [97, 100]. Saxitoxins have a molecular mass
of 224–491 Da [9]. Saxitoxin (6) consists of a tricyclic perhydropurine alkaloid that
can be substituted at various positions. Forty-two structural variants of saxitoxins
are known. They contain different functional groups at 4 different positions [9]. The
coexistence of different saxitoxin variants in the same strains has been associated
with biotransformation [101]. Saxitoxins are commonly divided in non-sulfated
(saxitoxins, dc saxitoxins and neosaxitoxin), mono-sulfated (gonyautoxins), di-­
sulfated (C-toxins), variants lacking a carbamoyl moiety (dc-toxins), or acetylated
(LWTXs) [102]. Saxitoxin and its analogs have been detected in all continents and
are produced by freshwater planktonic cyanobacterialike Dolichospermum (previ-
ously Anabaena), Raphidiopsis (previously Cylindrospermopsis), Aphanizomenon,
Planktothricoides (previously Planktothrix) and Oxynema [15, 89–91, 103].
Saxitoxins are highly polar molecules, and most are hydrophilic, mainly the ones
containing sulfate groups [101]. These compounds are highly soluble in water and
have been efficiently extracted from cells using acidified conditions [104]. They are
selective blockers of voltage-gated sodium, potassium, and calcium channels [105–
107]. Exposure to saxitoxins causes neurotoxicity [101].

H 2N
NH O
HO
HO N
O NH2
N N

NH2
6
28 M. B. Weiss et al.

Toxic secondary metabolites produced by cyanobacteria are abundant. They

belong to different chemical groups (cyclic peptides, alkaloids, and amino acid
derivatives) and have a large number of variants. Because of the high variability of
their chemical characteristics, the development of specialized analytical tools has
been a constant challenge.
In the past, cyanotoxins were often analyzed using bioassays with cell cultures,
animal models, or antibody-based systems [108, 109]. Later, liquid chromatography
and NMR systems were developed, allowing the targeted analysis of toxin classes
[110]. More recently, high-performance liquid chromatography (HPLC) coupled
with mass spectrometry (MS) was also introduced in toxin analysis, making it pos-
sible to quantify several toxins using single methods and targeted metabolomics
[111, 112]. Following the development of analytical techniques and bioinformatics,
untargeted metabolomics has become an important tool for studying cyanobacterial
metabolites.
In the following paragraphs, we will briefly discuss the concept of metabolo-
mics, its different approaches, analytical techniques that are used, and examples of
studies that have applied different metabolomics approaches to the research of
cyanotoxins.
Metabolomics can be defined as the study (identification and/or quantification)
of the set of metabolites (small molecules, products of metabolism) of a given bio-
logical system [113]. It is part of the set of “omics” sciences that include, among
others, genomics, transcriptomics, and proteomics [114].
In the last decades, metabolomics has been applied in various areas of knowl-
edge, such as in food and nutrition, sports, toxicology, environmental research,
pharmaceutical and medical research, clinical analysis, analysis of pathological
organisms, among others. This broad use has been due to the constant improvement
of technologies such as more accurate analytical tools, better sampling methods,
improvement of data quality, creation of shared analytical databases, and develop-
ment of bioinformatics pipelines [115].
Metabolomics approaches can be divided into two types: targeted and untargeted
metabolomics. Targeted metabolomics is defined as the analysis of few, pre-selected
metabolites of a certain chemical class, or certain metabolites that are associated
with specific metabolic pathways. It is often quantitative. Untargeted metabolomics
is the qualitative analysis of the largest possible number of metabolites, belonging
to different chemical classes, contained in the biological system under investigation
[116]. The choice of metabolomics approach directly depends on the research ques-
tion of the study.
Several analytical techniques can be used for metabolomic analysis. Each of
them has associated advantages and disadvantages, and obviously, no single analyti-
cal methodology is ideal for analyzing all metabolites in a system. Among the most
commonly used techniques for metabolomic analyses are nuclear magnetic reso-
nance (NMR) [117] and hyphenated techniques that couple high-resolution separa-
tion methods with mass spectrometry (MS), such as gas chromatography (GC-MS)
[118], liquid chromatography (LC-MS) [119] and capillary electrophoresis (CE-­
MS) [120].
2 Metabolomics Applied to Cyanobacterial Toxins and Natural Products 29

Regarding cyanotoxins, targeted metabolomic approaches have been used to: (a)
accelerate the identification and quantification of different classes of cyanotoxins in
environmental samples [121, 122], (b) monitor cyanotoxins in food and drinking
water [112, 123, 124], (c) determine the risk of exposure of certain toxins to living
organisms [125], (d) evaluate the effects of different growing conditions on the
metabolomic profiles of potentially toxic cyanobacteria [126].
Previous work tried to monitor PSP toxins by using different strategies. In 1995,
Oshima et al. developed a pioneering sensitive chromatographic method to analyze
PSPs [121]. This method used three different isocratic mobile phases and an oxidiz-
ing solution to produce fluorescent derivatives for detection. Despite the extended
analysis time due to three separate runs, this method enabled the identification of
sundry PSP variants with accurate results in different matrices such as dinoflagellate
cultures, natural blooms, and mussel and oyster extracts. Oshima’s method became
an official strategy to monitor and quantify saxitoxins based on HPLC-FD post
column derivatization [127, 128]. In 2007, Diener et al. improved Oshima’s method
by using Sequant HILIC column and mass spectrometry detection in a single run,
and being able to cover all PSP toxins [129].
The growing number of cyanoHABs in reservoirs requires quick decision-­
making from health agencies. Monitoring programs that rely solely on cyanobacte-
rial genus and species identification and cell counts may overestimate the risk and
lead to unnecessary health warnings [130]. Methods that use chromatographic sepa-
rations combined with mass spectrometry can accurately identify specific cyanotox-
ins [36]. Kaya and Sano (1999) developed a GC-MS method to monitor, in
freshwater, cyanotoxins containing Adda (2-amino-9-methoxy-2,6,8-trimethyl-­10-
phenyldeca-4,6-dienoic acid), i.e. microcystins [131]. The method was based on the
determination of the MMPB (erythro-2-methyl-3-methoxy-4-phenylbutyric acid), a
degradation product of the Adda moiety, using MMPB-d3 as internal standard.
Sample preparation involved a forced oxidation step followed by addition of the
internal standard. Selected ion monitoring (SIM) was used to detect ions derived
from MMPB and MMPB-d3. In addition to being specific, this method was shown
to be highly sensitive, with a detection limit of 0.43 pmol, corresponding to ca.
0.43 ng of microcystins [131]. Hiller et al. (2007) established an analytical method
based on liquid chromatography coupled with tandem mass spectrometry (LC-MS/
MS), using precursor ion mode for the simultaneous qualitative monitoring PSP
toxins, anatoxins (ANAs), cylindrospermopsins (CYNs), microcystins (MCs), and
nodularins (NODs), including rare and uncharacterized derivatives found in plank-
ton and water matrices [111].
The World Health Organization (WHO) recommends a provisional reference
value of 1.0 μg.L−1 MC-LR equivalents, 3.0 μg.L−1 for saxitoxins (saxitoxin.L−1
equivalent), and 1 μg.L−1 for cylindrospermopsin in drinking water [132]. Advances
in the determination of toxins in water using LC-MS/MS allow for additional cya-
nobacterial compounds to be monitored [133]. Pekar et al. (2016) developed a
“multitoxin method” based on LC-MS/MS that allows the simultaneous analysis of
22 cyanotoxins of different chemical classes in water, including anatoxins, cylin-
drospermopsins, nodularin, and microcystins [123]. Bogialli et al. (2017) presented
30 M. B. Weiss et al.

two analytical approaches, using an LC-QTOF system, for the targeted quantifica-
tion of specific cyanotoxins and screen of variants in freshwater [124].
Aquaculture species can be exposed to cyanotoxins through the food chain and
may accumulate them in their tissues, leading to biomagnification. Using UPLC-MS/
MS, Greer et al. (2017) developed and validated a targeted method capable of
detecting nine cyanotoxins (six microcystins, nodularin, cylindrospermopsin, and
anatoxin-a) in muscle, liver, and fish egg tissue samples [112].
The clinical approach of metabolomics, used to measure metabolic changes in
organisms exposed to certain conditions and/or substances, can also be used to
assess the risk of exposure to cyanotoxins. Lei et al. (2021) used 1H NMR-based
metabolomics combined with GC/LC-MS metabolic profiling to investigate the
toxic effects of exposure to MC-LR at environmentally relevant concentrations via
drinking water in rats [125].
The use of untargeted metabolomics to investigate the secondary metabolome of
cyanobacteria has revealed new cyanotoxin analogs [134–136] and has helped
determine variations in the production of these compounds in response to changes
in cultivation parameters [23, 137, 138]. Using LC-MS/MS molecular networking
on the Global Natural Products Social Molecular Networking (GNPS) platform fol-
lowed by isolation and structure elucidation, Via et al. (2018) discovered three new
members of the smenamide family, two of them with neurotoxic activity [134].
Using the same approach, Stewart et al. (2018) were able to describe 5 new
microginin variants [135]. Baliu-Rodriguez et al. (2021) identified two new micro-
cystin congeners using HPLC-MS/MS followed by data analysis in a newly devel-
oped software for spectral annotation and structure prediction of microcystins [136].
In an attempt to clarify the role and regulation of some secondary metabolites
produced by cyanobacteria, Briand et al. (2016) used an LC-MS/MS-based metabo-
lomics approach to investigate the production of bioactive compounds in response
to intraspecific strain interactions [26]. They showed changes in the relative concen-
tration of several small peptides, including some microcystins, under mono and
co-culture [137]. Moigne et al. (2021) used untargeted metabolomics to show sig-
nificant variations in metabolite dynamics when comparing exponential and station-
ary growing phases of a strain of cyanobacterium Aliinostoc sp. [138].
Using a “global metabolomics” approach based on LC-MS/MS fingerprinting,
Le Manach et al. (2019) found significantly different molecular profiles among 24
Microcystis strains [23]. Molecular networking on GNPS enabled the identification
of a broad set of chemically diverse metabolites, including microcystins, aerugino-
sins, microginins, cyanopeptolins, anabaenopeptins, and a set of unknown mole-
cules. Authors suggest that shotgun mass spectrometry chemotyping may be a
promising tool to rapidly characterize biomarkers for the toxicological evaluation of
cyanobacterial strains isolated from the field.
Identifying metabolites in metabolomic analyses is often a challenging process
that requires iterative searches in specialized databases. In 2021, Jones et al. pub-
lished a free database containing more than 2000 cyanobacterial secondary metabo-
lites and their associated metadata [9]. This database, named CyanoMetDB,
provides structural information (SMILES codes) for all metabolites and can be
2 Metabolomics Applied to Cyanobacterial Toxins and Natural Products 31

integrated to MS software. It is a powerful tool to complement MS screening and for

the dereplication of cyanotoxins and natural products.

2.3 Discovery of Cyanobacterial Natural Products Using

Metabolomics Approaches

Natural products continue to provide invaluable scaffolds for drug discovery.

According to a review by Newman and Cragg (2020), among the 1394 small-­
molecule drugs approved between 1981 and 2019, 67% were inspired in natural
compounds, including unmodified natural products and synthetic drugs based on
natural scaffolds [139]. Yet, there are concerns related to the increasing “rediscov-
ery” of known compounds and technical challenges associated with the complexity
of biological samples that hamper natural product discovery [140, 141]. The repeti-
tive investigation of similar natural sources (e.g., readily cultured microbes) with
commonly used methodologies has led to disappointing chemical redundancy [25].
Additionally, the chemical complexity of extracts complicates biological assay test-
ing and require extensive sample preparation, which tends to slow the pace of dis-
covery [141, 142]. Investing in understudied sources and incorporating new
methodologies are strategies that enhance the drug discovery process [143].
Cyanobacteria are ancient organisms with fossil records dating back to 3.5 bil-
lion years ago. They have evolved to colonize a wide variety of habitats, ranging
from aquatic to terrestrial, from deserts to hot springs, from marine to freshwater. To
thrive in these environments, often under strong competition, cyanobacteria have
developed survival strategies, such as the production of a broad range of biologi-
cally active metabolites [144]. Cyanobacterial natural products include a variety of
alkaloids, terpenoids, amino acid derivatives, fatty acid derivatives, polyketides, and
peptides [9, 11, 28]. It is generally accepted that, within the context of ecological,
evolutionary, and survival advantages, natural products are expected to bind targets
that can be explored for drug discovery [145]. Moreover, the biologically relevant
chemical space of natural products has been shown to improve hit rates in screening
campaigns [146]. As an underexplored source of natural products, cyanobacterial
metabolites can help meet the challenges of novel drug discovery.
Traditional natural product discovery workflows are often based on bioassay-­
guided fractionation, isolation, and structure elucidation. The incorporation of
metabolomics tools has provided new means to address challenges that are inherent
to natural products [119]. Advances include the possibility of an extensive view of
metabolite diversity within biological matrixes, new tools for sample prioritization,
correlations with biosynthetic pathways, associations between chemical and bioac-
tivity profiles, and the rapid identification of known molecules (dereplication),
expediting the discovery process. In the following paragraphs, we highlight selected
studies that applied metabolomics-based tools and strategies to facilitate natural
product discovery from cyanobacteria.
32 M. B. Weiss et al.

MS/MS molecular networking was used to expand the known chemical diversity of
an extensively studied cyanobacterium, Moorea producens JHB [147]. Briefly, MS/MS
molecular networking consists of a method for aligning fragmentation spectra based on
similarity. Results are displayed graphically to show consensus spectra as nodes and
their degree of similarity as edges, so that most similar spectra (according to an arbitrary
cut-off) are represented as clusters of connected nodes [148]. Considering that structur-
ally similar compounds tend to have similar fragmentation spectra, MS/MS molecular
networking enables the identification of molecular families and analogs of known com-
pounds in a sample. According to Boudreau et al. (2015), traditional workflows had led
to the discovery of hectochlorin (7) and jamaicamides A (8) and B (9) from Moorea
producens JHB. The use of MS/MS molecular networking guided the identification of
novel analogs of hectochlorin, namely hectochlorins B-D (10–12), and novel analogs of
jamaicamides, namely jamaicamides D (13) and E (14). The discovery of these com-
pounds shone light into the biosynthetic pathways of jamaicamides and hectochlorin
showing great plasticity for the production of a diversity of analogs. Further experiments
using halogen-supplemented media led to the discovery of an iodinated variant named
jamaicamide F (15) [147].

O
S
Cl O N OR2
R1
R3 O
O
O N
O
S
HO

7 R1 = Cl R2 = COCH3 R3 = H
10 R1 = Cl R2 = H R3 = H
11 R1 = H R2 = COCH3 R3 = H
12 R1 = Cl R2 = COCH3 R3 = CH3

R1
O O R2

N
H
N O
8 R1 = Cl R2 = Br
O 9 R1 = Cl R2 = H
13 R1 = H R2 = Br
14 R1 = H R2 = H
15 R1 = Cl R2 = I

Considering that as many as 99% of known microorganisms have not yet been
cultivated under laboratory conditions [149], environmental collections may expand
chemical diversity in the natural product discovery process. Luzzatto-Knaan et al.
(2017) have investigated over 300 field collections containing cyanobacteria and
macro-algae using untargeted mass spectrometry, multivariate analyses, and MS/
2 Metabolomics Applied to Cyanobacterial Toxins and Natural Products 33

MS molecular networking [150]. By associating LC-MS data with coordinates from

the field collections, Luzzatto-Knaan et al. (2017) identified biodiversity ‘hotspots’,
namely Panama-Portobelo, Palmyra Atoll, and Papua New Guinea, which are geo-
graphical areas of high chemical diversity. Detailed investigation of one cyanobac-
terial sample from Panama-Portobelo led to the discovery of yuvalamide A (16).
Molecular networking was used to propagate this annotation, guiding the identifica-
tion of yuvalamide B (17) [150]. This study illustrates the combination of metadata
attributes with analytical data to facilitate data visualization and assist massive
metabolomics analyses.

O
O
O
O NH HN

O
N O
H
O

R
16 R =

17 R =

Untargeted metabolomics along with bioassay-guided fractionation were used to

investigate an environmental sample of cf. Neolyngbya collected near Bangtang
Bay, South China [151]. Their approach led to the discovery of wenchangamide A
(18), a lipopeptide with seven sets of α-amino acids, shown to cause apoptosis to
human colorectal cancer cells. MS/MS molecular networking was used to detect
additional analogues of wenchangamide, among them wenchangamide B (19).
O
H2 N O

HO OH OH OH
O
HO
N N N N
H O N N N
O O H O H OH H
O

18

O
H2N O

HO OH OH OH OH
O
HO
N N N N
H O N N N
O O H O H OH H
O

19

Metabolomics has often been used in combination with genomics as orthogonal

methods to aid to natural product discovery. Advances in sequencing techniques and
reduction of costs have allowed genomics and genome mining to be incorporated in
natural product discovery workflows [152–154]. Genome mining refers to the bio-
informatics analysis of genomes aiming at detecting biosynthetic gene clusters
34 M. B. Weiss et al.

(BGCs), DNA regions predicted to code for the production of secondary metabo-
lites. Genome mining has greatly benefited from improved computational
approaches and web tools that make the process of BGC analysis automated and
accessible [155]. Examples of such tools include antiSMASH [156], NaPDoS [157],
PRISM [158], BAGEL [159], SMURF [160], BiG-SCAPE [161], ARTS [162],
among many others. While genome mining identifies the potential for secondary
metabolite production, the actual production of a certain molecule is assessed
through analytical chemistry techniques. The use of metabolomics and genomics
concurrently has been a key strategy to connect gene cluster to their respective
metabolites [163]. Kleigrewe et al. (2015) have employed the genome mining tools
AntiSMASH and NaPDos to analyze the genome sequences of three marine cyano-
bacterial strains, namely Moorea producens 3 L, Moorea producens JHB, and
Moorea bouillonii PNG [164]. These strains were known to produce previously
reported secondary metabolites and their respective BGCs were readily detected. A
specific regulatory gene was associated with hybrid nonribosomal peptide
synthetase-­polyketide synthase (NRPS-PKS) BGCs in two of the strains. A BLAST
search using this gene sequence revealed an undescribed NRPS-PKS BGC on the
third strain. Finally, MS/MS molecular networking was used to compare the metab-
olomic profiles of the three strains, leading to the discovery of columbamides A-C
(20–22), the putative product of the newly identified NRPS-PKS BGC [164].

R2 Cl OR1
N
Cl
O
20 R1 = Ac R2 = H O
21 R1 = Ac R2 = Cl
22 R1 = H R2 = H

Another strategy to link BGCs to their metabolites explores the use of stable iso-
tope-labeled precursors. It has been historically employed, in a targeted fashion, to
determine the biosynthetic origin of natural compounds [165]. Current bioinformatics
pipelines have enabled stable isotope labeling to be used in extensive metabolomic
analyses [165–167]. Cyanobacteria have the distinctive capacity to incorporate nitro-
gen directly from culture media into their metabolites. By utilizing 15N-labeled nitrate
instead of nitrate-14N in cyanobacteria culture media, it is possible to achieve full
isotope incorporation in secondary metabolites [168, 169]. May et al. have combined
genome mining analysis with stable isotope labeling to investigate the secondary
metabolism of the cyanobacterial strain Nostoc sp. UIC 10630 [168]. Analysis of the
genome sequence by antiSMASH detected six BGCs predicted to encode compounds
with >1 nitrogen atom. In parallel, comparative metabolomics using LC-MS analysis
was performed on cyanobacterial extracts obtained from cells cultured in 15N-labeled
and unlabeled culture media, revealing four nitrogen-­containing metabolites. To link
BGC and respective metabolites, the detected number of nitrogen atoms in each com-
pound was compared to the predicted number of nitrogens for each BGC. With this
approach, two previously known compounds, nostopeptolide A1 (23) and aeruginosin
865 (24), were identified and two novel natural products were discovered, namely
anabaenopeptin UIC827 (25) and nostopyrrolidonamide (26) [168].
2 Metabolomics Applied to Cyanobacterial Toxins and Natural Products 35

NH

O
N NH
O
HO
O O N
HN O
O O
O
H2N HN
NH
O H
N
O
N
O H O

23

OH O O
O O
O

HO OH
OH N
O
O NH
O
N NH2
H
OH HN
NH
HO 24

O
HN O
N
O H
OH
NH
O NH
O
NH
HN
N
O O
OH
25

N
HN
O O
O
O N

26
36 M. B. Weiss et al.

The so called “maldiisotopic” approach employs matrix-assisted laser desorp-

tion/ionization (MALDI) MS to rapidly detect nitrogen-containing compounds
from cultured cyanobacteria fed with unlabeled and 15N-labeled nitrate [169]. In the
study by Kinnel et al. (2017), the cyanobacterium Moorea producens JHB was cul-
tured in media containing nitrate-14N and -15N followed by MALDI MS analysis. In
addition to known metabolites, a feature with a relatively low m/z value ([M+H]+
m/z 400) and high nitrogen content (5 nitrogen atoms) was detected. Subsequent
isolation and structural elucidation revealed a novel secondary metabolite named
cryptomaldamide (27). Genome sequencing and mining were used to investigate the
biosynthetic origin of 27, which was determined to be produced by a hybrid NRPS-­
PKS [169].

O H
H -
H 2N N N COO
N
H
NH
+
O
HO
27

Sample preparation is known to be a laborious process in natural product discov-

ery and metabolomics. By performing analysis in situ (directly from the raw mate-
rial), one can save precious sample preparation time. Ionization methods such as
DESI (desorption electrospray ionization), LAESI (laser ablation electrospray ion-
ization), and MALDI have enabled mass spectrometry analysis to be performed
directly from the biological samples. These techniques also allow for mass spec-
trometry imaging, opening new possibilities in natural products discovery [170].
The droplet-liquid microjunction-surface sampling probe (or “droplet probe”) rep-
resents another important advance in the field, as it couples chromatography to in
situ analysis [171]. It consists of rapid sampling technique that requires little (if any)
sample preparation to generate a micro-extract that can be, in turn, analyzed in
LC-based systems. Crnkovic et al. (2018) used the droplet probe hyphenated to an
LC-UV-HRMS/MS system to analyze the chemical profile of freshwater cyanobac-
teria growing on agar plates [172]. It allowed not only the detection but also the
dereplication of compounds in situ. Using this method, the strain Calothrix sp. UIC
10520 was prioritized for large-scale growth based on the chemical novelty of its
metabolites. Detailed investigation led to the discovery of calothrixamides A (28)
and B (29). This work showed, as a proof of concept, an application of the droplet
probe for the chemical analysis of cyanobacteria in culture.

NH2

O O
R
O
HO
NH
28 R =
29 R =
2 Metabolomics Applied to Cyanobacterial Toxins and Natural Products 37

Culture conditions are known to influence the secondary metabolism of microor-

ganisms. The strategy named “one strain many compounds” (OSMAC) explores the
ability of microorganisms to produce different chemistry under different conditions
[173, 174]. Crnkovic et al. (2018) used a metabolomics approach to evaluate how
the levels of nitrate and phosphate in culture media impacted the chemical profile of
three freshwater cyanobacterial strains [175]. XCMS [176, 177] was used for com-
parative metabolomics analyses. While nitrate levels had a higher impact on pri-
mary metabolism, phosphate levels were shown to modulate the production of
secondary metabolites. The production of the cytotoxic merocyclophane C (30) by
the strain Nostoc sp. UIC 10110 increased under low phosphate levels, and the con-
tent of the actin inhibitor tolytoxin (31) in strain Scytonema sp. UIC 10036 increased
under high phosphate levels in comparison to the standard Z medium. Additionally,
a novel peptide was detected in the strain UIC 10036 under low nitrate and high
phosphate conditions. Isolation and structure elucidation of the novel compound led
to the discovery of scytodecamide (32), a linear peptide composed of 10 amino acid
residues showing a N-terminal N-methylation and a C-terminal amidation [178].
Further genome sequencing and genome mining identified a cyanobactin-like gene
cluster as the putative BGC of 33. Interestingly, the vast majority of cyanobactins
are cyclic peptides [179]. The first linear cyanobactins were short peptides discov-
ered by genome mining and reported in 2013 [180]. Genome mining and metabolo-
mics approaches have expanded the known chemical diversity of cyanobactins.

OH O
OMe
O
HO OH
O OMe OMe
OH N
HO OH O OH O O

OMe O
30 31

HO
O O O O O
H H H H
HN N N N N N
N N N N NH2
H H H H
O O O O O
OH

32

Leikoski et al. (2010) have combined genome mining, stable isotopic labeling,
and mass spectrometry to discover new cyclic cyanobactins from freshwater
Anabaena strains [181]. Genome sequencing of the cyanobacterial strain Anabaena
sp. 90 followed by bioinformatics annotation revealed a cyanobactin-like BGC. No
product could be predicted for this BGC due to the lack of conserved cleavage sites
in the precursor protein. Using PCR primers for a protease gene within the BGC, 26
other Anabaena strains were screened for the potential for cyanobactin production.
One of the strains showed a similar precursor protein encoding the sulfur-containing
amino acid methionine, which was selected for isotope-labeling studies. The strain
Anabaena sp. SYKE 844B was grown in medium containing either inorganic 34S or
32
S. The cell extracts were analyzed by LC-MS to identify compounds with 2-Da
38 M. B. Weiss et al.

mass shifts in the labeled and unlabeled extracts, indicating the presence of a sulfur
atom. Using this approach, anacyclamide B7 (33) was discovered and enabled the
prediction of the peptide product of the BGC initially identified in the strain
Anabaena sp. 90, anacyclamide A10 (34). Further anacyclamides were also detected
and their structures were confirmed by comparison with synthetic anacyclamides
and/or MS/MS fragmentation analyses. Finally, the cyanobactin-like BGC detected
in Anabaena sp. 90 BGC was expressed heterologously in Escherichia coli to con-
firm that it was responsible for the biosynthesis of anacyclamides [181].

S NH
N
H
N
HN
O
O O N
HN O O

HN
NH O
H
N
O

33

NH
OH
H O
N N
HN N
O H
O O
O NH
HN O HO
O O
H O
H2N N
N N N
H H
O O
HO

34

2.4 Conclusion and Perspectives

The analysis of cyanobacterial toxins and natural products has greatly benefited
from metabolomics. Targeted and untargeted approaches have been applied to
improve both detection and discovery of cyanobacterial compounds. Methods based
on mass spectrometry are especially useful due to its high sensitivity, selectivity,
2 Metabolomics Applied to Cyanobacterial Toxins and Natural Products 39

dynamic range, and resolving power. Additionally, novel sampling and sample
preparation techniques have been shown to expedite and/or enhance the chemical
analysis of cyanobacteria. From the perspective of the “dry laboratory”, advances in
bioinformatics have enabled the development of new pipelines that are continuously
added to the metabolomics toolbox. It should be noted, however, that the automated
identification of cyanobacterial metabolites remains challenging due to the rela-
tively low number of annotated spectra in databases. Yet, this bottleneck can be
mitigated as spectral databases become populated with more, high-quality spectra
of cyanobacterial compounds. Initiatives such as the GNPS and CyanoMetDB are
useful tools for the dereplication of cyanotoxins and natural products. In big data
analyses, metabolomics has often been combined with genomics, transcriptomics,
proteomics, as well as biological information to answer broader and more complex
scientific questions. New perspectives are being developed as machine learning and
artificial intelligence are incorporated into metabolomics pipelines. Most impor-
tantly, the combination of these tools and approaches continue to expand our knowl-
edge on the exceptional chemical diversity of cyanobacteria.

References

1. Vidal L, Ballot A, Azevedo SMFO, Padisák J, Welker M (2021) Introduction to cyanobacte-

ria. In: Chorus I, Welker M (eds) Toxic cyanobacteria in water, 2nd edn. Taylor & Francis,
Oxon, pp 163–211
2. Komárek J, Kaštovský J, Mareš J, Johansen JR (2014) Taxonomic classification of cyanopro-
karyotes (cyanobacterial genera) 2014, using a polyphasic approach. Preslia 86(4):295–335
3. Nabout JC, da Silva Rocha B, Carneiro FM, Sant’Anna CL (2013) How many species of
cyanobacteria are there? Using a discovery curve to predict the species number. Biodivers
Conserv 22(12):2907–2918. https://doi.org/10.1007/s10531-­013-­0561-­x
4. Schopf JW (2006) Fossil evidence of Archaean life. Philos Trans R Soc B Biol Sci
361(1470):869–885. https://doi.org/10.1098/rstb.2006.1834
5. Altermann W, Kazmierczak J (2003) Archean microfossils: a reappraisal of early life on
Earth. Res Microbiol 154(9):611–617. https://doi.org/10.1016/j.resmic.2003.08.006
6. Schirrmeister BE, de Vos JM, Antonelli A, Bagheri HC (2013) Evolution of multicellularity
coincided with increased diversification of cyanobacteria and the great oxidation event. Proc
Natl Acad Sci 110(5):1791–1796. https://doi.org/10.1073/pnas.1209927110
7. Sanchez-Baracaldo P, Hayes PK, Blank CE (2005) Morphological and habitat evolution in
the cyanobacteria using a compartmentalization approach. Geobiology 3(3):145–165. https://
doi.org/10.1111/j.1472-­4669.2005.00050.x
8. Adams DG, Duggan PS, Jackson O (2012) Cyanobacterial symbioses. In: Ecology of cyano-
bacteria II. Springer, pp 593–647
9. Jones MR, Pinto E, Torres MA, Dörr F, Mazur-Marzec H, Szubert K et al (2021) CyanoMetDB,
a comprehensive public database of secondary metabolites from cyanobacteria. Water Res
196:117017. https://doi.org/10.1016/j.watres.2021.117017
10. Leão PN, Engene N, Antunes A, Gerwick WH, Vasconcelos V (2012) The chemical ecology
of cyanobacteria. Nat Prod Rep 29(3):372–391. https://doi.org/10.1039/c2np00075j
11. Tidgewell K, Clark BR, Gerwick WH (2010) The natural products chemistry of cyanobacte-
ria. In: Liu H-W (Ben), Mander L (eds) Comprehensive natural products II. Elsevier, Oxford,
pp 141–188
40 M. B. Weiss et al.

12. Francis G (1878) Poisonous Australian lake. Nature 18(444):11–12. https://doi.

org/10.1038/018011d0
13. Paerl HW, Otten TG (2013) Blooms bite the hand that feeds them. Science (80-)
342(6157):433–434. https://doi.org/10.1126/science.1245276
14. Buratti FM, Manganelli M, Vichi S, Stefanelli M, Scardala S, Testai E et al (2017)
Cyanotoxins: producing organisms, occurrence, toxicity, mechanism of action and human
health toxicological risk evaluation. Arch Toxicol 91(3):1049–1130. https://doi.org/10.1007/
s00204-­016-­1913-­6
15. Chorus I, Welker M (2021) Toxic cyanobacteria in water: a guide to their public health con-
sequences, monitoring and management. CRC Press
16. Hilborn ED, Roberts VA, Backer L, DeConno E, Egan JS, Hyde JB et al (2014) Algal bloom-­
associated disease outbreaks among users of freshwater lakes: United States, 2009-2010.
Morb Mortal Wkly Rep 63(1):145–156
17. Chorus I, Ian R, Falconer H (2000) Health risks caused by freshwater cyanobacte-
ria in recreational waters. J Toxicol Environ Health Part B 3(4):323–347. https://doi.
org/10.1080/109374000436364
18. Kaebernick M (2001) Ecological and molecular investigations of cyanotoxin production.
FEMS Microbiol Ecol 35(1):1–9. https://doi.org/10.1016/S0168-­6496(00)00093-­3
19. Falconer IR (1994) Health implications of cyanobacterial (blue-green algae) toxins. In: Toxic
cyanobacteria current status of research and management-proceedings of an international
workshop. Adelaide
20. Meiβner K (1996) Toxic and non-toxic strains of the cyanobacterium Microcystis aerugi-
nosa contain sequences homologous to peptide synthetase genes. FEMS Microbiol Lett
135(2–3):295–303. https://doi.org/10.1016/0378-­1097(95)00469-­6
21. Janssen EML (2019) Cyanobacterial peptides beyond microcystins – a review on co-­
occurrence, toxicity, and challenges for risk assessment. Water Res 151:488–499. https://doi.
org/10.1016/j.watres.2018.12.048
22. Demay J, Bernard C, Reinhardt A, Marie B (2019) Natural products from cyanobacteria:
focus on beneficial activities. Mar Drugs 17(6):320. https://doi.org/10.3390/md17060320
23. Le Manach S, Duval C, Marie A, Djediat C, Catherine A, Edery M et al (2019) Global
metabolomic characterizations of Microcystis spp. highlights clonal diversity in natural
bloom-forming populations and expands metabolite structural diversity. Front Microbiol
10(Mar):1–16. https://doi.org/10.3389/fmicb.2019.00791
24. Huang I-S, Zimba PV (2019) Cyanobacterial bioactive metabolites – a review of their
chemistry and biology. Harmful Algae 83(June 2018):42–94. https://doi.org/10.1016/j.
hal.2018.11.008
25. Atanasov AG, Zotchev SB, Dirsch VM, Supuran CT (2021) Natural products in drug dis-
covery: advances and opportunities. Nat Rev Drug Discov 20(3):200–216. https://doi.
org/10.1038/s41573-­020-­00114-­z
26. Pye CR, Bertin MJ, Lokey RS, Gerwick WH, Linington RG (2017) Retrospective analysis
of natural products provides insights for future discovery trends. Proc Natl Acad Sci U S A
114(22):5601–5606. https://doi.org/10.1073/pnas.1614680114
27. Salvador-Reyes LA, Luesch H (2015) Biological targets and mechanisms of action of natural
products from marine cyanobacteria. Nat Prod Rep 32(3):478–503. https://doi.org/10.1039/
C4NP00104D
28. Chlipala GE, Mo S, Orjala J (2011) Chemodiversity in freshwater and terrestrial cyano-
bacteria – a source for drug discovery. Curr Drug Targets 12(11):1654–1673. https://doi.
org/10.2174/138945011798109455
29. Luesch H, Moore RE, Paul VJ, Mooberry SL, Corbett TH (2001) Isolation of dolastatin
10 from the marine cyanobacterium Symploca species VP642 and total stereochemistry and
biological evaluation of its analogue symplostatin 1. J Nat Prod 64(7):907–910. https://doi.
org/10.1021/np010049y
2 Metabolomics Applied to Cyanobacterial Toxins and Natural Products 41

30. Doronina SO, Toki BE, Torgov MY, Mendelsohn BA, Cerveny CG, Chace DF et al (2003)
Development of potent monoclonal antibody auristatin conjugates for cancer therapy. Nat
Biotechnol 21(7):778–784. https://doi.org/10.1038/nbt832
31. Senter PD, Sievers EL (2012) The discovery and development of brentuximab vedotin for use
in relapsed Hodgkin lymphoma and systemic anaplastic large cell lymphoma. Nat Biotechnol
30(7):631–637. https://doi.org/10.1038/nbt.2289
32. Richardson NC, Kasamon YL, Chen H, de Claro RA, Ye J, Blumenthal GM et al (2019)
FDA approval summary: brentuximab vedotin in first-line treatment of peripheral T-cell lym-
phoma. Oncologist 24(5):e180–e187. https://doi.org/10.1634/theoncologist.2019-­0098
33. Bortoli S, Pinto E (2015) Cianotoxinas: Características Gerais, Histórico, Legislação E
Métodos De Análises. In: Ecologia de reservatórios e interfaces. Instituto de Biociências da
Universidade de São Paulo, pp 163–164
34. Huisman J, Codd GA, Paerl HW, Ibelings BW, Verspagen JMH, Visser PM (2018)
Cyanobacterial blooms. Nat Rev Microbiol 16(8):471–483. https://doi.org/10.1038/
s41579-­018-­0040-­1
35. Paerl HW, Huisman J (2008) Climate: blooms like it hot. Science 320(5872):57–58. https://
doi.org/10.1126/science.1155398
36. Brunton C, Butler K, Collier M, Hamann J, Lilienfeld-jones J, Otness P et al (2021) EPA
needs an agencywide strategic action plan to address harmful algal blooms. Washington, DC
37. Scheffer M, Hosper SH, Meijer M-L, Moss B, Jeppesen E (1993) Alternative equilib-
ria in shallow lakes. Trends Ecol Evol 8(8):275–279. https://doi.org/10.1007/978-­3-
­642-­41714-­6_192672
38. Rabalais NN, Díaz RJ, Levin LA, Turner RE, Gilbert D, Zhang J (2010) Dynamics and dis-
tribution of natural and human-caused hypoxia. Biogeosciences 7(2):585–619. https://doi.
org/10.5194/bg-­7-­585-­2010
39. Jüttner F, Watson SB (2007) Biochemical and ecological control of geosmin and
2-­methylisoborneol in source waters. Appl Environ Microbiol 73(14):4395–4406. https://
doi.org/10.1128/AEM.02250-­06
40. Du X, Liu H, Yuan L, Wang Y, Ma Y, Wang R et al (2019) The diversity of cyanobacterial tox-
ins on structural characterization, distribution and identification: a systematic review. Toxins
(Basel) 11(9):1–34. https://doi.org/10.3390/toxins11090530
41. Merel S, Walker D, Chicana R, Snyder S, Baurès E, Thomas O (2013) State of knowledge
and concerns on cyanobacterial blooms and cyanotoxins. Environ Int 59:303–327. https://doi.
org/10.1016/j.envint.2013.06.013
42. Chernoff N, Faassen EJ, Hill DJ (2021) 2.7. 1 discrepancies introduced by incorrect BMAA
analysis. In: Chorus I, Welker M (eds) Toxic cyanobacteria in water: a guide to their public
health consequences, monitoring and management. CRC Press, Boca Raton, pp 123–125
43. Bishop CT, Anet EFLJ, Gorham PR (1959) Isolation and identification of the fast-death fac-
tor in Microcystis aeruginosa NRC-1. Can J Biochem Physiol 37(1):453–471. https://doi.
org/10.1139/y59-­047
44. Martins JC, Vasconcelos VM (2009) Microcystin dynamics in aquatic organisms. J Toxicol
Environ Health Part B 12(1):65–82. https://doi.org/10.1080/10937400802545151
45. Sivonen K, Jones GJ (1999) Cyanobacterial toxins. In: Chorus I, Bartram J (eds) Toxic cya-
nobacteria in water: a guide to their public health consequences, monitoring and manage-
ment. E & FN Spon, London, pp 41–91
46. Rinehart KL, Harada K, Namikoshi M, Chen C, Harvis CA, Munro MHG et al (1988)
Nodularin, microcystin, and the configuration of Adda. J Am Chem Soc 110(25):8557–8558.
https://doi.org/10.1021/ja00233a049
47. Tillett D, Dittmann E, Erhard M, von Döhren H, Börner T, Neilan BA (2000) Structural
organization of microcystin biosynthesis in Microcystis aeruginosa PCC7806: an integrated
peptide–polyketide synthetase system. Chem Biol 7(10):753–764. https://doi.org/10.1016/
S1074-­5521(00)00021-­1
42 M. B. Weiss et al.

48. Fastner J, Humpage AR (2021) Hepatotoxic cyclic peptides – microcystins and nodularins.
In: Toxic cyanobacteria in water, 2nd edn. CRC Press, Boca Raton, pp 21–52
49. Msagati TAM, Siame BA, Shushu DD (2006) Evaluation of methods for the isolation, detec-
tion and quantification of cyanobacterial hepatotoxins. Aquat Toxicol 78(4):382–397
50. Harada K, Tsuji K, Watanabe MF, Kondo F (1996) Stability of microcystins from cyanobac-
teria – III. Effect of pH and temperature. Phycologia 35(sup6):83–88
51. Funari E, Testai E (2008) Human health risk assessment related to cyanotoxins exposure. Crit
Rev Toxicol 38(2):97–125
52. Codd GA, Meriluoto J, Metcalf JS (2016) Introduction: cyanobacteria, cyanotoxins, their
human impact, and risk management. In: Handbook of cyanobacterial monitoring and cya-
notoxin analysis. Wiley, pp 1–8
53. Ji Y, Lu G, Chen G, Huang B, Zhang X, Shen K et al (2011) Microcystin-LR induces apopto-
sis via NF-κB/iNOS pathway in INS-1 cells. Int J Mol Sci 12(7):4722–4734
54. McLellan NL, Manderville RA (2017) Toxic mechanisms of microcystins in mammals.
Toxicol Res (Camb) 6(4):391–405
55. Ohtani I, Moore RE, Runnegar MTC (1992) Cylindrospermopsin: a potent hepatotoxin from
the blue-green alga Cylindrospermopsis raciborskii. J Am Chem Soc 114(20):7941–7942
56. Mihali TK, Kellmann R, Muenchhoff J, Barrow KD, Neilan BA (2008) Characterization of
the gene cluster responsible for cylindrospermopsin biosynthesis. Appl Environ Microbiol
74(3):716–722. https://doi.org/10.1128/AEM.01988-­07
57. Mazmouz R, Chapuis-Hugon F, Mann S, Pichon V, Méjean A, Ploux O (2010) Biosynthesis
of cylindrospermopsin and 7-epicylindrospermopsin in Oscillatoria sp. strain PCC
6506: identification of the cyr gene cluster and toxin analysis. Appl Environ Microbiol
76(15):4943–4949. https://doi.org/10.1128/AEM.00717-­10
58. Spoof L, Berg KA, Rapala J, Lahti K, Lepistö L, Metcalf JS et al (2006) First observation of
cylindrospermopsin in Anabaena lapponica isolated from the boreal environment (Finland).
Environ Toxicol 21(6):552–560. https://doi.org/10.1002/tox.20216
59. Stirling DJ, Quilliam MA (2001) First report of the cyanobacterial toxin cylindrospermopsin
in New Zealand. Toxicon 39(8):1219–1222. https://doi.org/10.1016/S0041-­0101(00)00266-­X
60. Harada K, Ohtani I, Iwamoto K, Suzuki M, Watanabe MF, Watanabe M et al (1994) Isolation
of cylindrospermopsin from a cyanobacterium Umezakia natans and its screening method.
Toxicon 32(1):73–84. https://doi.org/10.1016/0041-­0101(94)90023-­X
61. Kokociński M, Mankiewicz-Boczek J, Jurczak T, Spoof L, Meriluoto J, Rejmonczyk E et al
(2013) Aphanizomenon gracile (Nostocales), a cylindrospermopsin-producing cyanobacte-
rium in Polish lakes. Environ Sci Pollut Res 20(8):5243–5264
62. Seifert M, McGregor G, Eaglesham G, Wickramasinghe W, Shaw G (2007) First evidence
for the production of cylindrospermopsin and deoxy-cylindrospermopsin by the freshwater
benthic cyanobacterium, Lyngbya wollei (Farlow ex Gomont) Speziale and Dyck. Harmful
Algae 6(1):73–80
63. Preußel K, Stüken A, Wiedner C, Chorus I, Fastner J (2006) First report on cylindrospermop-
sin producing Aphanizomenon flos-aquae (Cyanobacteria) isolated from two German lakes.
Toxicon 47(2):156–162
64. Fastner J, Rücker J, Stüken A, Preußel K, Nixdorf B, Chorus I et al (2007) Occurrence of
the cyanobacterial toxin cylindrospermopsin in northeast Germany. Environ Toxicol An Int
J 22(1):26–32
65. Detoni AMS, Costa LDF, Pacheco LA, Yunes JS (2016) Toxic Trichodesmium bloom occur-
rence in the southwestern South Atlantic Ocean. Toxicon 110:51–55
66. Senogles P, Shaw G, Smith M, Norris R, Chiswell R, Mueller J et al (2000) Degradation of
the cyanobacterial toxin cylindrospermopsin, from Cylindrospermopsis raciborskii, by chlo-
rination. Toxicon 38(9):1203–1213
67. Chiswell RK, Shaw GR, Eaglesham G, Smith MJ, Norris RL, Seawright AA et al (1999)
Stability of cylindrospermopsin, the toxin from the cyanobacterium, Cylindrospermopsis
raciborskii: effect of pH, temperature, and sunlight on decomposition. Environ Toxicol An
Int J 14(1):155–161
2 Metabolomics Applied to Cyanobacterial Toxins and Natural Products 43

68. Froscio SM, Fanok S, Humpage AR (2009) Cytotoxicity screening for the cyanobacterial
toxin cylindrospermopsin. J Toxicol Environ Health Part A 72(5):345–349
69. Jacinavicius FR, Carneiro, Ronaldo Leal Sant’Anna CL, Rigonato J, Carvalho LR (2016)
Phenotypic plasticity and negative allelopathy in Microcystis strains. Ann Microbiol
66(3):1265–1276. https://doi.org/10.1007/s13213-­016-­1215-­5
70. Gorham PR (1964) Toxic algae as a public health hazard. J Am Water Works Assoc
56(11):1481–1488
71. Devlin JP, Edwards OE, Gorham PR, Hunter NR, Pike RK, Stavric B (1977) Anatoxin-a, a
toxic alkaloid from Anabaena flos-aquae NRC-44h. Can J Chem 55(8):1367–1371
72. Kust A, Méjean A, Ploux O (2020) Biosynthesis of anatoxins in cyanobacteria: identifica-
tion of the carboxy-anatoxins as the penultimate biosynthetic intermediates. J Nat Prod
83(1):142–151. https://doi.org/10.1021/acs.jnatprod.9b01121
73. Méjean A, Mann S, Maldiney T, Vassiliadis G, Lequin O, Ploux O (2009) Evidence that bio-
synthesis of the neurotoxic alkaloids anatoxin-a and homoanatoxin-a in the cyanobacterium
Oscillatoria PCC 6506 occurs on a modular polyketide synthase initiated by l-proline. J Am
Chem Soc 131(22):7512–7513. https://doi.org/10.1021/ja9024353
74. Méjean A, Paci G, Gautier V, Ploux O (2014) Biosynthesis of anatoxin-a and analogues (ana-
toxins) in cyanobacteria. Toxicon 91:15–22
75. Hodoki Y, Ohbayashi K, Kobayashi Y, Takasu H, Okuda N, Nakano S (2013) Anatoxin-
a-­producing Raphidiopsis mediterranea Skuja var. grandis Hill is one ecotype of non-­
heterocytous Cuspidothrix issatschenkoi (Usačev) Rajaniemi et al. in Japanese lakes. Harmful
Algae 21–22:44–53. https://doi.org/10.1016/j.hal.2012.11.007
76. Wood SA, Puddick J, Fleming R, Heussner AH (2017) Detection of anatoxin-producing
Phormidium in a New Zealand farm pond and an associated dog death. New Zeal J Bot
55(1):36–46. https://doi.org/10.1080/0028825X.2016.1231122
77. Sivonen K, Himberg K, Luukkainen R, Niemelä SI, Poon GK, Codd GA (1989) Preliminary
characterization of neurotoxic cyanobacteria blooms and strains from Finland. Toxic Assess
4(3):339–352. https://doi.org/10.1002/tox.2540040310
78. Osswald J, Rellán S, Gago A, Vasconcelos V (2007) Toxicology and detection methods of the
alkaloid neurotoxin produced by cyanobacteria, anatoxin-a. Environ Int 33(8):1070–1089.
https://doi.org/10.1016/j.envint.2007.06.003
79. Kaminski A, Bober B, Lechowski Z, Bialczyk J (2013) Determination of anatoxin-a sta-
bility under certain abiotic factors. Harmful Algae 28:83–87. https://doi.org/10.1016/j.
hal.2013.05.014
80. Rapala J, Sivonen K, Luukkainen R, Niemelä SI (1993) Anatoxin-a concentration in
Anabaena and Aphanizomenon under different environmental conditions and comparison
of growth by toxic and non-toxic Anabaena-strains – a laboratory study. J Appl Phycol
5(6):581–591. https://doi.org/10.1007/BF02184637
81. Thomas P, Stephens M, Wilkie G, Amar M, Lunt GG, Whiting P et al (1993) (+)-Anatoxin-a is
a potent agonist at neuronal nicotinic acetylcholine receptors. J Neurochem 60(6):2308–2311.
https://doi.org/10.1111/j.1471-­4159.1993.tb03519.x
82. Aráoz R, Molgó J, Tandeau de Marsac N (2010) Neurotoxic cyanobacterial toxins. Toxicon
56(5):813–828. https://doi.org/10.1016/j.toxicon.2009.07.036
83. Mahmood NA, Carmichael WW (1987) Anatoxin-a(s), an anticholinesterase from the cya-
nobacterium Anabaena flos-aquae NRC-525-17. Toxicon 25(11):1221–1227. https://doi.
org/10.1016/0041-­0101(87)90140-­1
84. Fiore MF, de Lima ST, Carmichael WW, McKinnie SMK, Chekan JR, Moore BS (2020)
Guanitoxin, re-naming a cyanobacterial organophosphate toxin. Harmful Algae 92:101737.
https://doi.org/10.1016/j.hal.2019.101737
85. Matsunaga S, Moore RE, Niemczura WP, Carmichael WW (1989) Anatoxin-a(s), a potent
anticholinesterase from Anabaena flos-aquae. J Am Chem Soc 111(20):8021–8023. https://
doi.org/10.1021/ja00202a057
44 M. B. Weiss et al.

86. Moura S, Pinto E (2010) Synthesis of cyclic guanidine intermediates of anatoxin-a(s) in both
racemic and enantiomerically pure forms. Synlett 06:967–969
87. Moore BS, Ohtani I, de Koning CB, Moore RE, Carmichael WW (1992) Biosynthesis of
anatoxin-a(s). Origin of the carbons. Tetrahedron Lett 33(44):6595–6598. https://doi.
org/10.1016/S0040-­4039(00)60994-­2
88. Dörr FA, Rodríguez V, Molica R, Henriksen P, Krock B, Pinto E (2010) Methods for detec-
tion of anatoxin-a(s) by liquid chromatography coupled to electrospray ionization-tandem
mass spectrometry. Toxicon 55(1):92–99. https://doi.org/10.1016/j.toxicon.2009.07.017
89. Testai E (2021) 2.5 anatoxin-a(s). In: Bartram J, Chorus I (eds) Toxic cyanobacteria in water,
2nd edn. Taylor & Francis, Boca Raton, p 109
90. Carmichael WW, Gorham PR (1978) Anatoxins from clones of Anabaena flos-aquae isolated
from lakes of western Canada. SIL Commun 1953–1996 21(1):285–295. https://doi.org/1
0.1080/05384680.1978.11903972
91. Molica RJR, Oliveira EJA, Carvalho PVVC, Costa ANSF, Cunha MCC, Melo GL et al (2005)
Occurrence of saxitoxins and an anatoxin-a(s)-like anticholinesterase in a Brazilian drinking
water supply. Harmful Algae 4(4):743–753. https://doi.org/10.1016/j.hal.2004.11.001
92. Fernandes K, Dörr FA, Pinto E (2021) Stability analyses by HPLC-MS of guanitoxin iso-
lated from Sphaerospermopsis torques-reginae. J Braz Chem Soc 32:1559–1567. https://doi.
org/10.21577/0103-­5053.20210053
93. Thottumkara AP, Parsons WH, Du Bois J (2014) Saxitoxin. Angew Chem Int Ed
53(23):5760–5784. https://doi.org/10.1002/anie.201308235
94. Schantz EJ, Mold JD, Stanger DW, Shavel J, Riel FJ, Bowden JP et al (1957) Paralytic shell-
fish poison. VI. A procedure for the isolation and purification of the poison from toxic clam
and mussel tissues. J Am Chem Soc 79(19):5230–5235. https://doi.org/10.1021/ja01576a044
95. Schantz EJ, Lynch JM, Vayvada G, Matsumoto K, Rapoport H (1966) The purification
and characterization of the poison produced by Gonyaulax catenella in axenic culture*.
Biochemistry 5(4):1191–1195. https://doi.org/10.1021/bi00868a011
96. Gentile JH, Maloney TE (1969) Toxicity and environmental requirements of a strain
of Aphanizomenon flos-aquae (L.) Ralfs. Can J Microbiol 15(2):165–173. https://doi.
org/10.1139/m69-­028
97. Kellmann R, Michali TK, Neilan BA (2008) Identification of a saxitoxin biosynthesis gene
with a history of frequent horizontal gene transfers. J Mol Evol 67(5):526–538. https://doi.
org/10.1007/s00239-­008-­9169-­2
98. Stüken A, Orr RJS, Kellmann R, Murray SA, Neilan BA, Jakobsen KS (2011) Discovery of
nuclear-encoded genes for the neurotoxin saxitoxin in dinoflagellates. PLoS One 6(5):e20096.
https://doi.org/10.1371/journal.pone.0020096
99. Kellmann R, Mihali TK, Jeon YJ, Pickford R, Pomati F, Neilan BA (2008) Biosynthetic inter-
mediate analysis and functional homology reveal a saxitoxin gene cluster in cyanobacteria.
Appl Environ Microbiol 74(13):4044–4053. https://doi.org/10.1128/AEM.00353-­08
100. Hackett JD, Wisecaver JH, Brosnahan ML, Kulis DM, Anderson DM, Bhattacharya D et al
(2013) Evolution of saxitoxin synthesis in cyanobacteria and dinoflagellates. Mol Biol Evol
30(1):70–78. https://doi.org/10.1093/molbev/mss142
101. Wiese M, D’Agostino PM, Mihali TK, Moffitt MC, Neilan BA (2010) Neurotoxic alkaloids:
saxitoxin and its analogs. Mar Drugs 8(7):2185–2211. https://doi.org/10.3390/md8072185
102. Pearson LA, Dittmann E, Mazmouz R, Ongley SE, D’Agostino PM, Neilan BA (2016)
The genetics, biosynthesis and regulation of toxic specialized metabolites of cyanobacteria.
Harmful Algae 54:98–111. https://doi.org/10.1016/j.hal.2015.11.002
103. Bartram J, Chorus I (1999) Toxic cyanobacteria in water: a guide to their public health con-
sequences, monitoring and management, 1st edn. E & F Spon, London
104. Haddad SP, Bobbitt JM, Taylor RB, Lovin LM, Conkle JL, Chambliss CK et al (2019)
Determination of microcystins, nodularin, anatoxin-a, cylindrospermopsin, and saxitoxin in
water and fish tissue using isotope dilution liquid chromatography tandem mass spectrom-
etry. J Chromatogr A 1599:66–74. https://doi.org/10.1016/j.chroma.2019.03.066
2 Metabolomics Applied to Cyanobacterial Toxins and Natural Products 45

105. Pearson L, Mihali T, Moffitt M, Kellmann R, Neilan B (2010) On the chemistry, toxicology
and genetics of the cyanobacterial toxins, microcystin, nodularin, saxitoxin and cylindrosper-
mopsin. Mar Drugs 8(5):1650–1680. https://doi.org/10.3390/md8051650
106. Su Z, Sheets M, Ishida H, Li F, Barry WH (2004) Saxitoxin blocks L-Type I Ca. J Pharmacol
Exp Ther 308(1):324–329. https://doi.org/10.1124/jpet.103.056564
107. Ritchie JM, Rogart RB (1977) The binding of saxitoxin and tetrodotoxin to excitable tissue.
In: Reviews of physiology, biochemistry and pharmacology, vol 79. Springer, pp 1–50
108. Deblois CP, Giani A, Bird DF (2011) Experimental model of microcystin accumulation in
the liver of Oreochromis niloticus exposed subchronically to a toxic bloom of Microcystis sp.
Aquat Toxicol 103(1–2):63–70. https://doi.org/10.1016/j.aquatox.2011.02.006
109. Mohamed ZA, Carmichael WW, Hussein AA (2003) Estimation of microcystins in the fresh-
water fish Oreochromis niloticus in an Egyptian fish farm containing a Microcystis bloom.
Environ Toxicol 18(2):137–141. https://doi.org/10.1002/tox.10111
110. Humpage AR, Magalhaes VF, Froscio SM (2010) Comparison of analytical tools and bio-
logical assays for detection of paralytic shellfish poisoning toxins. Anal Bioanal Chem
397(5):1655–1671. https://doi.org/10.1007/s00216-­010-­3459-­4
111. Hiller S, Krock B, Cembella A, Luckas B (2007) Rapid detection of cyanobacterial toxins in
precursor ion mode by liquid chromatography tandem mass spectrometry. J Mass Spectrom
42(July):1238–1250. https://doi.org/10.1002/jms.1257
112. Greer B, Maul R, Campbell K, Elliott CT (2017) Detection of freshwater cyanotoxins and
measurement of masked microcystins in tilapia from Southeast Asian aquaculture farms.
Anal Bioanal Chem 409(16):4057–4069. https://doi.org/10.1007/s00216-­017-­0352-­4
113. Demarque DP, Dusi RG, De Sousa FDM, Grossi SM, Silvério MRS, Lopes NP et al (2020)
Mass spectrometry-based metabolomics approach in the isolation of bioactive natural prod-
ucts. Sci Rep Nat Res 10(1051):1–9. https://doi.org/10.1038/s41598-­020-­58046-­y
114. Schwarz D, Orf I, Kopka J, Hagemann M (2013) Recent applications of metabolomics toward
cyanobacteria. Meta 3(1):72–100. https://doi.org/10.3390/metabo3010072
115. Shulaev V (2006) Metabolomics technology and bioinformatics. Brief Bioinform
7(2):128–139. https://doi.org/10.1093/bib/bbl012
116. Canuto GAB, Costa JL da, Cruz, Pedro LR da, Souza ARL, Faccio AT, Klassen A et al
(2018) Metabolômica: definições, estado-da-arte e aplicações representativas. Quim Nova
41(1):75–91. https://doi.org/10.21577/0100-­4042.20170134
117. Wishart DS (2019) NMR metabolomics: a look ahead. J Magn Reson 306:155–161. https://
doi.org/10.1016/j.jmr.2019.07.013
118. Fiehn O (2016) Metabolomics by gas chromatography–mass spectrometry: combined
targeted and untargeted profiling. Curr Protoc Mol Biol 114(1):232–235. https://doi.
org/10.1002/0471142727.mb3004s114
119. Krug D, Müller R (2014) Secondary metabolomics: the impact of mass spectrometry-based
approaches on the discovery and characterization of microbial natural products. Nat Prod Rep
31(6):768–783. https://doi.org/10.1039/c3np70127a
120. Boizard F, Brunchault V, Moulos P, Breuil B, Klein J, Lounis N et al (2016) A capillary
electrophoresis coupled to mass spectrometry pipeline for long term comparable assessment
of the urinary metabolome. Sci Rep 6(September):1–17. https://doi.org/10.1038/srep34453
121. Oshima Y (1995) Postcolumn derivatization liquid chromatographic method for paralytic
shellfish toxins. J AOAC Int 78(2):528–532. https://doi.org/10.1093/jaoac/78.2.528
122. Lawrence JF, Niedzwiadek B (2001) Quantitative determination of paralytic shellfish poi-
soning toxins in shellfish by using prechromatographic oxidation and liquid chromatog-
raphy with fluorescence detection. J AOAC Int 84(4):1099–1108. https://doi.org/10.1093/
jaoac/84.4.1099
123. Pekar H, Westerberg E, Bruno O, Lääne A, Persson KM, Sundström LF et al (2016) Fast,
rugged and sensitive ultra high pressure liquid chromatography tandem mass spectrometry
method for analysis of cyanotoxins in raw water and drinking water-first findings of anatox-
ins, cylindrospermopsins and microcystin variants in Swedish source waters and infiltration
ponds. J Chromatogr A 1429:265–276. https://doi.org/10.1016/j.chroma.2015.12.049
46 M. B. Weiss et al.

124. Bogialli S, Bortolini C, Di Gangi IM, Di Gregorio FN, Lucentini L, Favaro G et al (2017)
Liquid chromatography-high resolution mass spectrometric methods for the surveillance
monitoring of cyanotoxins in freshwaters. Talanta 170(December 2016):322–330. https://
doi.org/10.1016/j.talanta.2017.04.033
125. Lei H, Song Y, Dong M, Chen G, Cao Z, Wu F et al (2021) Metabolomics safety assessments
of microcystin exposure via drinking water in rats. Ecotoxicol Environ Saf 212:111989.
https://doi.org/10.1016/j.ecoenv.2021.111989
126. Jacinavicius FR, Geraldes V, Crnkovic CM, Delbaje E, Fiore MF, Pinto E (2021) Effect of
ultraviolet radiation on the metabolomic profiles of potentially toxic cyanobacteria. FEMS
Microbiol Ecol 97(1):fiaa243. https://doi.org/10.1093/femsec/fiaa243
127. Burrell S, Crum S, Foley B, Turner AD (2016) Proficiency testing of laboratories for paralytic
shellfish poisoning toxins in shellfish by QUASIMEME: a review. TrAC Trends Anal Chem
75:10–23. https://doi.org/10.1016/j.trac.2015.09.004
128. Etheridge SM (2010) Paralytic shellfish poisoning: seafood safety and human health perspec-
tives. Toxicon 56(2):108–122. https://doi.org/10.1016/j.toxicon.2009.12.013
129. Diener M, Erler K, Christian B, Luckas B (2007) Application of a new zwitterionic hydro-
philic interaction chromatography column for determination of paralytic shellfish poisoning
toxins. J Sep Sci 30(12):1821–1826. https://doi.org/10.1002/jssc.200700025
130. Farrer D, Counter M, Hillwig R, Cude C (2015) Health-based cyanotoxin guideline values
allow for cyanotoxin-based monitoring and efficient public health response to cyanobacterial
blooms. Toxins (Basel) 7(2):457–477. https://doi.org/10.3390/toxins7020457
131. Kaya K, Sano T (1999) Total microcystin determination using acid (MMPB-d 3) as the inter-
nal standard. Anal Chim Acta 386:107–112. https://doi.org/10.1016/S0003-­2670(99)00012-­4
132. Falconer I, Bartram J, Chorus I, Kuiper-Goodman T, Utkilen H, Burch M et al (1999) Safe
levels and safe practices. In: Toxic cyanobacteria in water. CRC Press, pp 175–198
133. Schymanski EL, Singer HP, Slobodnik J, Ipolyi IM, Oswald P, Krauss M et al (2015) Non-
target screening with high-resolution mass spectrometry: critical review using a collabora-
tive trial on water analysis. Anal Bioanal Chem 407(21):6237–6255. https://doi.org/10.1007/
s00216-­015-­8681-­7
134. Via CW, Glukhov E, Costa S, Zimba PV, Moeller PDR, Gerwick WH et al (2018) The
metabolome of a cyanobacterial bloom visualized by MS/MS-based molecular networking
reveals new neurotoxic smenamide analogs (C, D, and E). Front Chem 6(July):1–9. https://
doi.org/10.3389/fchem.2018.00316
135. Stewart AK, Ravindra R, Van Wagoner RM, Wright JLC (2018) Metabolomics-guided dis-
covery of microginin peptides from cultures of the cyanobacterium Microcystis aeruginosa.
J Nat Prod 81(2):349–355. https://doi.org/10.1021/acs.jnatprod.7b00829
136. Baliu-Rodriguez D, Peraino NJ, Premathilaka SH, Birbeck JA, Baliu-Rodriguez T, Westrick
JA et al (2022) Identification of novel microcystins using high-resolution MS and MS n with
python code. Environ Sci Technol:acs.est.1c04296. https://doi.org/10.1021/acs.est.1c04296
137. Briand E, Bormans M, Gugger M, Dorrestein PC, Gerwick WH (2016) Changes in secondary
metabolic profiles of Microcystis aeruginosa strains in response to intraspecific interactions.
Environ Microbiol 18(2):384–400. https://doi.org/10.1111/1462-­2920.12904
138. Le Moigne D, Demay J, Reinhardt A, Bernard C, Kim Tiam S, Marie B (2021) Dynamics
of the metabolome of Aliinostoc sp. PMC 882.14 in response to light and temperature varia-
tions. Meta 11(11):745. https://doi.org/10.3390/metabo11110745
139. Newman DJ, Cragg GM (2020) Natural products as sources of new drugs over the nearly
four decades from 01/1981 to 09/2019. J Nat Prod 83(3):770–803. https://doi.org/10.1021/
acs.jnatprod.9b01285
140. Pye CR, Bertin MJ, Lokey RS, Gerwick WH, Linington RG (2017) Retrospective analy-
sis of natural products provides insights for future discovery trends. Proc Natl Acad Sci
114(22):5601–5606. https://doi.org/10.1073/pnas.1614680114
141. Bisson J, McAlpine JB, Friesen JB, Chen S-N, Graham J, Pauli GF (2016) Can invalid bio-
actives undermine natural product-based drug discovery? J Med Chem 59(5):1671–1690.
https://doi.org/10.1021/acs.jmedchem.5b01009
2 Metabolomics Applied to Cyanobacterial Toxins and Natural Products 47

142. Beutler JA (2009) Natural products as a foundation for drug discovery. Curr Protoc Pharmacol
46(1):1–21. https://doi.org/10.1002/0471141755.ph0911s46
143. Harvey AL, Edrada-Ebel R, Quinn RJ (2015) The re-emergence of natural products for
drug discovery in the genomics era. Nat Rev Drug Discov 14(2):111–129. https://doi.
org/10.1038/nrd4510
144. Raja R, Hemaiswarya S, Ganesan V, Carvalho IS (2015) Recent developments in therapeu-
tic applications of cyanobacteria. Crit Rev Microbiol 42(3):1–12. https://doi.org/10.310
9/1040841X.2014.957640
145. Williams DH, Stone MJ, Hauck PR, Rahman SK (1989) Why are secondary metabolites
(natural products) biosynthesized? J Nat Prod 52(6):1189–1208. https://doi.org/10.1021/
np50066a001
146. Sukuru SCK, Jenkins JL, Beckwith REJ, Scheiber J, Bender A, Mikhailov D et al
(2009) Plate-based diversity selection based on empirical HTS data to enhance the num-
ber of hits and their chemical diversity. J Biomol Screen 14(6):690–699. https://doi.
org/10.1177/1087057109335678
147. Boudreau PD, Monroe EA, Mehrotra S, Desfor S, Korobeynikov A, Sherman DH et al (2015)
Expanding the described metabolome of the marine cyanobacterium Moorea producens JHB
through orthogonal natural products workflows. PLoS One 10(7):e0133297. https://doi.
org/10.1371/journal.pone.0133297
148. Yang JY, Sanchez LM, Rath CM, Liu X, Boudreau PD, Bruns N et al (2013) Molecular
networking as a dereplication strategy. J Nat Prod 76(9):1686–1699. https://doi.org/10.1021/
np400413s
149. Hofer U (2018) The majority is uncultured. Nat Rev Microbiol 16(12):716–717. https://doi.
org/10.1038/s41579-­018-­0097-­x
150. Luzzatto-Knaan T, Garg N, Wang M, Glukhov E, Peng Y, Ackermann G et al (2017) Digitizing
mass spectrometry data to explore the chemical diversity and distribution of marine cyano-
bacteria and algae. elife 6:1–20. https://doi.org/10.7554/eLife.24214
151. Ding L, Bar-Shalom R, Aharonovich D, Kurisawa N, Patial G, Li S et al (2021) Metabolomic
characterization of a cf. Neolyngbya cyanobacterium from the South China Sea reveals
Wenchangamide A, a Lipopeptide with in vitro apoptotic potential in colon cancer cells. Mar
Drugs 19(7):397. https://doi.org/10.3390/md19070397
152. Albarano L, Esposito R, Ruocco N, Costantini M (2020) Genome mining as new challenge in
natural products discovery. Mar Drugs 18(4):199. https://doi.org/10.3390/md18040199
153. Scherlach K, Hertweck C (2021) Mining and unearthing hidden biosynthetic potential. Nat
Commun 12(1):3864. https://doi.org/10.1038/s41467-­021-­24133-­5
154. Zerikly M, Challis GL (2009) Strategies for the discovery of new natural products by genome
mining. Chembiochem 10(4):625–633. https://doi.org/10.1002/cbic.200800389
155. Machado H, Tuttle RN, Jensen PR (2017) Omics-based natural product discovery and the
lexicon of genome mining. Curr Opin Microbiol 39:136–142. https://doi.org/10.1016/j.
mib.2017.10.025
156. Blin K, Shaw S, Kloosterman AM, Charlop-Powers Z, van Wezel GP, Medema MH et al
(2021) antiSMASH 6.0: improving cluster detection and comparison capabilities. Nucleic
Acids Res 49(W1):W29–W35. https://doi.org/10.1093/nar/gkab335
157. Ziemert N, Podell S, Penn K, Badger JH, Allen E, Jensen PR (2012) The natural product
domain seeker NaPDoS: a phylogeny based bioinformatic tool to classify secondary metabo-
lite gene diversity. PLoS One 7(3):e34064. https://doi.org/10.1371/journal.pone.0034064
158. Skinnider MA, Merwin NJ, Johnston CW, Magarvey NA (2017) PRISM 3: expanded pre-
diction of natural product chemical structures from microbial genomes. Nucleic Acids Res
45(W1):W49–W54. https://doi.org/10.1093/nar/gkx320
159. van Heel AJ, de Jong A, Montalbán-López M, Kok J, Kuipers OP (2013) BAGEL3: auto-
mated identification of genes encoding bacteriocins and (non-)bactericidal posttranslation-
ally modified peptides. Nucleic Acids Res 41(W1):W448–W453. https://doi.org/10.1093/
nar/gkt391
48 M. B. Weiss et al.

160. Khaldi N, Seifuddin FT, Turner G, Haft D, Nierman WC, Wolfe KH et al (2010) SMURF:
genomic mapping of fungal secondary metabolite clusters. Fungal Genet Biol 47(9):736–741.
https://doi.org/10.1016/j.fgb.2010.06.003
161. Navarro-Muñoz JC, Selem-Mojica N, Mullowney MW, Kautsar SA, Tryon JH, Parkinson
EI et al (2020) A computational framework to explore large-scale biosynthetic diversity. Nat
Chem Biol 16(1):60–68. https://doi.org/10.1038/s41589-­019-­0400-­9
162. Alanjary M, Kronmiller B, Adamek M, Blin K, Weber T, Huson D et al (2017) The Antibiotic
Resistant Target Seeker (ARTS), an exploration engine for antibiotic cluster prioritization and
novel drug target discovery. Nucleic Acids Res 45(W1):W42–W48. https://doi.org/10.1093/
nar/gkx360
163. van der Hooft JJJ, Mohimani H, Bauermeister A, Dorrestein PC, Duncan KR, Medema MH
(2020) Linking genomics and metabolomics to chart specialized metabolic diversity. Chem
Soc Rev. https://doi.org/10.1039/D0CS00162G
164. Kleigrewe K, Almaliti J, Tian IY, Kinnel RB, Korobeynikov A, Monroe EA et al (2015)
Combining mass spectrometric metabolic profiling with genomic analysis: a powerful
approach for discovering natural products from cyanobacteria. J Nat Prod 78(7):1671–1682.
https://doi.org/10.1021/acs.jnatprod.5b00301
165. McCaughey CS, van Santen JA, van der Hooft JJJ, Medema MH, Linington RG (2021) An
isotopic labeling approach linking natural products with biosynthetic gene clusters. Nat
Chem Biol. https://doi.org/10.1038/s41589-­021-­00949-­6
166. Huang X, Chen Y-J, Cho K, Nikolskiy I, Crawford PA, Patti GJ (2014) X13CMS: global
tracking of isotopic labels in untargeted metabolomics. Anal Chem 86(3):1632–1639. https://
doi.org/10.1021/ac403384n
167. Chokkathukalam A, Jankevics A, Creek DJ, Achcar F, Barrett MP, Breitling R (2013)
mzMatch–ISO: an R tool for the annotation and relative quantification of isotope-
labelled mass spectrometry data. Bioinformatics 29(2):281–283. https://doi.org/10.1093/
bioinformatics/bts674
168. May DS, Crnkovic CM, Krunic A, Wilson TA, Fuchs JR, Orjala JE (2020) 15 N stable isotope
labeling and comparative metabolomics facilitates genome mining in cultured cyanobacteria.
ACS Chem Biol 15(3):758–765. https://doi.org/10.1021/acschembio.9b00993
169. Kinnel RB, Esquenazi E, Leao T, Moss N, Mevers E, Pereira AR et al (2017) A maldiisotopic
approach to discover natural products: cryptomaldamide, a hybrid tripeptide from the marine
cyanobacterium Moorea producens. J Nat Prod 80(5):1514–1521. https://doi.org/10.1021/
acs.jnatprod.7b00019
170. Spraker JE, Luu GT, Sanchez LM (2020) Imaging mass spectrometry for natural products dis-
covery: a review of ionization methods. Nat Prod Rep. https://doi.org/10.1039/C9NP00038K
171. Oberlies NH, Knowles SL, Amrine CSM, Kao D, Kertesz V, Raja HA (2019) Droplet probe:
coupling chromatography to the in situ evaluation of the chemistry of nature. Nat Prod Rep
36(7):944–959. https://doi.org/10.1039/C9NP00019D
172. Crnkovic CM, Krunic A, May DS, Wilson TA, Kao D, Burdette JE et al (2018)
Calothrixamides A and B from the cultured cyanobacterium Calothrix sp. UIC 10520. J Nat
Prod 81(9):2083–2090. https://doi.org/10.1021/acs.jnatprod.8b00432
173. Pan R, Bai X, Chen J, Zhang H, Wang H (2019) Exploring structural diversity of microbe sec-
ondary metabolites using OSMAC strategy: a literature review. Front Microbiol 10(Feb):1–20.
https://doi.org/10.3389/fmicb.2019.00294
174. Romano S, Jackson SA, Patry S, Dobson ADW (2018) Extending the “one strain many com-
pounds” (OSMAC) principle to marine microorganisms. Mar Drugs 16(7):1–29. https://doi.
org/10.3390/md16070244
175. Crnkovic CM, May DS, Orjala J (2018) The impact of culture conditions on growth and
metabolomic profiles of freshwater cyanobacteria. J Appl Phycol 30(1):375–384. https://doi.
org/10.1007/s10811-­017-­1275-­3
2 Metabolomics Applied to Cyanobacterial Toxins and Natural Products 49

176. Tautenhahn R, Patti GJ, Rinehart D, Siuzdak G (2012) XCMS online: a web-based plat-
form to process untargeted metabolomic data. Anal Chem 84(11):5035–5039. https://doi.
org/10.1021/ac300698c
177. Gowda H, Ivanisevic J, Johnson CH, Kurczy ME, Benton HP, Rinehart D et al (2014)
Interactive XCMS online: simplifying advanced metabolomic data processing and subse-
quent statistical analyses. Anal Chem 86(14):6931–6939. https://doi.org/10.1021/ac500734c
178. Crnkovic CM, Braesel J, Krunic A, Eustμquio AS, Orjala J (2019) Scytodecamide from the
cultured Scytonema sp. UIC 10036 expands the chemical and genetic diversity of cyanobac-
tins. Chembiochem:1–9. https://doi.org/10.1002/cbic.201900511
179. Sivonen K, Leikoski N, Fewer DP, Jokela J (2010) Cyanobactins – ribosomal cyclic pep-
tides produced by cyanobacteria. Appl Microbiol Biotechnol 86(5):1213–1225. https://doi.
org/10.1007/s00253-­010-­2482-­x
180. Leikoski N, Liu L, Jokela J, Wahlsten M, Gugger M, Calteau A et al (2013) Genome mining
expands the chemical diversity of the cyanobactin family to include highly modified linear
peptides. Chem Biol 20(8):1033–1043. https://doi.org/10.1016/j.chembiol.2013.06.015
181. Leikoski N, Fewer DP, Jokela J, Wahlsten M, Rouhiainen L, Sivonen K (2010) Highly diverse
cyanobactins in strains of the genus Anabaena. Appl Environ Microbiol 76(3):701–709.
https://doi.org/10.1128/AEM.01061-­09
Chapter 3
Unveiling Microbial Chemical Interactions
Based on Metabolomics Approaches

Laís Castro de Carvalho, Arnaldo de Almeida Junior,

Fernanda Silva Ribeiro, and Célio Fernando Figueiredo Angolini

3.1 Introduction

Microorganisms are present on the most diverse habitats around the world, even on
spacecrafts that are supposed to be sterile and which are held in drastic environmen-
tal conditions, these organisms find ways to underpin their survival [1]. Unveiling
their interactions with the environment and with other organisms can be quite com-
plex and challenging and, therefore, different approaches need to be used [2–4]. For
that, it is fruitful not only to perform their chemical characterization but also com-
prehend their complex relations with the respective hosts (e.g., plants) and other
organisms present in the environment. Herein, we will present studies concerning
microorganisms with agricultural benefits. In addition, the tools presented here
could help other researchers to understand different microorganisms’ phenomena.
In nature, one common strategy that convergently evolved in different kingdoms
through co-evolutionary arm races, is the secretion of a wide range of molecules.
These biomolecules are called allelochemicals and may be beneficial (positive alle-
lopathy) or deleterious (negative allelopathy). Usually such molecules are second-
ary metabolites, that are not necessary for the basal metabolism of allelopathic
organisms, but play different roles in these organisms interactions [5]. Plants have
naturally developed, during its evolution, a range of chemically mediated interac-
tions with the surrounding heterotrophic community, especially microorganisms
[6]. Over the years, the catalog of molecules and extracts with biological properties
have continued to grow, however, it is still difficult to understand the individual role

L. C. de Carvalho · A. de Almeida Junior · F. S. Ribeiro · C. F. F. Angolini (*)

Mass Spectrometry and Chemical Ecology Laboratory (MC-CELL), Center for Natural and
Human Sciences, University of ABC (UFABC), São Paulo, Brazil
e-mail: [email protected]

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 51

T. Pacheco Fill (ed.), Microbial Natural Products Chemistry, Advances in Experimental
Medicine and Biology 1439, https://doi.org/10.1007/978-3-031-41741-2_3
52 L. C. de Carvalho et al.

and evolutionary origin of many of these molecules in the different interactions

studied in microbial ecology.
Therefore, we need to look beyond chemistry itself and understands the complex
interactions between hosts (plants) and the other components of its ecosystem
(microorganisms). We must unravel the whole chain of events that begins with the
release of these allelochemicals in the environment to the manifestation of their
impact on the system organization. To achieve this, we need tools and methods that
allow us not only to characterize those molecules, but also to define their origin,
once spatial distribution and its abundance are as important as their identification
for the understanding of the studied phenomena.
In this sense, traditional natural products chemistry classical approaches, in
which the discovery of new molecules involves several extraction and purification
steps for later identification of the isolated compounds, needs to be updated. An
obstacle in such approaches involves the small quantities of the biomolecules pro-
duced by these microorganisms, making the traditional methods very time consum-
ing and often unfeasible. Many other points are considered obsolete in the classical
methodologies, such as: artifact production, excessive solvent/reagent waste and
loss of spatial distribution information in the studied matrix.
Additionally, the access to quick and large genome information is allowing
unraveling a larger number of cryptic secondary metabolites. Today, the genome
sequencing of microorganisms can be performed on a large scale and relatively fast,
and the genome annotation and prediction tools for secondary metabolites are also
growing and becoming increasingly efficient [7]. In this scenario, the structural
characterization techniques of metabolites need to evolve and reach the speed and
resources required in molecular biology techniques. Recently, advances in analyti-
cal techniques, especially in mass spectrometry (MS) combined to bioinformatics
are allowing new approaches to emerge. For instance, nowadays it is possible to
access secondary metabolites directly from petri dish cultures using different mass
spectrometry techniques [8, 9].
Microbial metabolomics studies can have a variety of aims and approaches. The
studies can focus on how microorganisms live and adapt to a particular host from a
systemic biology point of view or even explore their biological potential for produc-
tion of bioactive molecules (such as antibiotics). Nonetheless, the most prolific
compounds regarding biological activities and ecological importance are secondary
metabolites, also known as natural products or specialized metabolites. They cor-
respond to a highly diverse group of molecules, with representatives that cover a
range of chemical classes, and which have received attention from researchers in
recent decades due to their high bioactive potential and application as anticancer
agents, [10] antibiotics, [11] chemotherapy, [12] pesticides, [13, 14] and others
[15]. Concerning microorganisms, the natural products produced can be even more
diverse and complex and classical approaches for metabolite characterization and
identification can be time consuming and sometimes unsuccessful.
An elegant way to overcome this situation, which is nowadays very common and
easily to perform in metabolomic studies, is the genome guided investigation (or
genome mining) [16]. A plethora of tools are available today to translate the genome
3 Unveiling Microbial Chemical Interactions Based on Metabolomics Approaches 53

information, predicting the ability of the target organism to produce secondary

metabolites. In general, three connected Omic approaches can be performed in
microbial metabolomics studies: genomic, metabolomic, and phenomic. In this
chapter, we intend to resume those approaches and their applications.

3.2 Genome Derived Approaches (Genomic)

Although specialized metabolites are classified into different molecule groups, the
biosynthetic machinery for their production remains conserved in many organisms;
and the genes responsible for encoding the enzymatic complexes responsible for
synthesizing them are found close within the genome, as gene clusters called BGC
(Biosynthetic Gene Clusters). In this sense, predicting and isolating bioproducts
based on genetic information has become a target study in natural products com-
munity and several research groups came together to develop methodologies that
make these goals feasible. Thus, tools such as antiSMASH, BiG-SCAPE, PRISM,
CLUSEAN and others emerged and are being applied in several microbial studies
[15, 17, 18].
The antiSMASH (antibiotics and Secondary Metabolite Analysis Shell), is a pio-
neering tool which its first version was released in 2011 and are continuously being
updated [19, 20]. It works as an autonomous server that enables automatic genomic
analysis of BGCs with prediction of polyketide synthases (PKS), [21] non-­ribosomal
peptide synthetases (NRPS), lanthipeptides and many other classes, which allows
us to perform identification, annotation and comparison of known and unknown
BGCs; in addition, it provides the potential chemical structure of several biomole-
cules based on sequence data already described in databases [20]. In short, the soft-
ware identifies different biosynthetic pathways involved in the production of
specialized metabolites and applies validated rules that define the essential biosyn-
thetic functions in a region of the genome for a BGC to be constituted.
The interface used by antiSMASH guarantees a certain level of simplicity to the
user when executing its functions, allowing in a few steps to perform the query of
the most varied types of genomes. As input data, the genomic sequence files in
FASTA, GBK or EMBL formats is requested or, alternatively, the software supports
the use of the GenBank/RefSeq accession number. Before starting the process, the
user can select the types of BGC they are interested in searching and which analysis
modules they want to include. As a result, antiSMASH provides: identification of
the respective BGCs; analysis and annotation of NRPS/PKS domains; prediction of
the core chemical structure of PKSs and NRPSs; comparative analysis of gene clus-
ters via ClusterBlast and analysis of protein families of the secondary metabolism
of smCOGs (secondary metabolite clusters of orthologous groups). The output is
visualized as an interactive page and all details are stored in an EMBL file for fur-
ther analysis, Fig. 3.1.
Currently, antiSMASH is the most widely applied resource for classifying BGCs
in fungi and bacteria and was recently updated to its seventh version (antiSMASH
54 L. C. de Carvalho et al.

Fig. 3.1 AntiSMASH genomic analysis for the genome of Streptomyces misionensis

7.0 beta is already available), which features, among other functions, the integration
of acquired results from other gene prediction tools, with greater effectiveness in the
detection of tailoring enzymes in ribosomally synthesized and post-translationally
modified peptides (RiPPs) [19].
With the advancement of BGC prediction techniques and the identification of
thousands of clusters, the databases used in these repositories are responsible for
storing a considerable amount of genetic information about the most diverse classes
of natural products. And it is a fact, that for the study of new BGCs is essential the
prior knowledge of the already characterized BGCs. From this perspective, in order
to standardize the maintenance of bioinformatics data in 2013 the ClusterMine360
was created, [22] the first database responsible for organizing information from
approximately 300 gene clusters.
Following the evolution of the genomic era, and still aiming to maintain the stan-
dard of file conservation, in 2015 the MIBiG (Minimum Information about a
Biosynthetic Gene Cluster) was established, [23] which immediately became a ref-
erence database for the search for clusters, presenting 2021 entries from known
function BGCs. With a user-friendly interface, MIBiG allows acquiring information
about secondary metabolites with structures similar to those found in the software
itself and detecting mass spectra linked to a specific molecule of interest; such attri-
butions complement existing connections with PubChem and other databases whose
main purpose is to match the structure of annotated BGCs and chemical data. For
instance, Bahram et al. used this tool to identify fungal BGCs associated with anti-
bacterial activity in 7560 metagenomic samples; in this work the authors were able
to demonstrate the correlation between the abundance of these BGCs and the pres-
ence of microbial resistance genes in certain soils [24]. An example of the interface
offered by MIBiG can be seen in Fig. 3.2, which is an example of the cluster
BGC0001792 identified in the species Streptomyces albidoflavus, this cluster still
3 Unveiling Microbial Chemical Interactions Based on Metabolomics Approaches 55

Fig. 3.2 Overview of the new version of MIBiG. The section found in the upper left corner allows
you to highlight specific genes from which information will be shown in the upper right corner
(gene detail section). At the bottom, the “Compounds” tab is selected, displaying all information
related to the BGC compound including chemical structure, molecular formula and related
databases

has incomplete annotation in MIBiG. Even so, it is possible to recognize Surugamide

A and Surugamide D as its main products, classify their biosynthetic classes and
visualize their defined structures, as shown in Fig. 3.2.
More recently, with the aim of expanding the work carried out by antiSMASH
and MIBiG, a new tool known as BiG-SCAPE (Biosynthetic Gene Similarity
Clustering And Prospecting Engine) has emerged. This platform allows the
researcher to explore the diversity of biosynthetic gene clusters (BGCs) across large
numbers of genomes by grouping orthologous BGCs (via comparison of structural
relationships) into families, “Gene Clusters Families” (GCFs) [25]. In other words,
this software uses BGCs obtained from antiSMASH analyses or annotated in
MIBiG as input and classifies them into families of clusters (multiple hierarchical
levels), which allows exploring gene cluster diversity linked to enzyme
phylogenies.
However, regarding the mapping and prediction of secondary metabolites, there
are some challenges to be overcome such as to connect these mapped BGCs to their
respective compounds. Although such clusters of genes can be directly identified
from the genome, the chemical characterization of the respective molecule requires
a lot of manual work and, therefore, aiming at its optimization, researchers have
been focusing on computational strategies and the use of omic approaches. In other
words, solving this question involves not only the characterization of BGCs but also
involves the identification of the produced biomolecules by these gene clusters.
According to Skinnider et al., one of the problems faced in the context of the
discovery of new bioproducts is linked to the development of recent methodologies
that have a greater focus on the detection of BGCs rather than the prediction of the
molecular structure of natural products from microbial genomes [26]. Thus, in this
study, the authors highlight the need for algorithms that take into account the
56 L. C. de Carvalho et al.

complete set of chemical reactions promoted by catalytic enzymes and present the
PRISM tool as a resource to overcome the aforementioned difficulties.
PRISM (PRediction Informatics for Secondary Metabolomes) is a server capable
of predicting the chemical structure of non-ribosomal peptides (NRPs) and
polyketides (type I and II), both modular enzymes for secondary metabolites pro-
duction, using microbial nucleotide sequences as input. (FASTA or Gen-Bank for-
mat) that are searched in HMMs (“hidden Markov models”) libraries associated
with secondary metabolism. Through the grouping of identified biosynthetic genes
and the analysis of protein sequences, the software organizes the data in a way that
it is possible to predict the structures [27].
Non-ribosomal peptides constitute a class of biomolecules synthesized indepen-
dently of the ribosome and often have branched structures that may contain struc-
tural modifications such as N-methylations, for example. However, it is not always
possible to predict exactly where these modification reactions will occur (“tailoring
reactions”) and, therefore, in addition to the information provided, PRISM gener-
ates combinatorial libraries of predicted structures indicating the variability of the
action of the given enzymes that promote these reactions. Additionally, the tool also
predicts the structure of ribosomally synthesized and post-translationally modified
peptides (RiPPs).
Also, according to the study published by Skinnider et al. (2017), the need to
expand the detection of “clusters” and the prediction of structures made by PRISM
to a broader collection of classes of natural products was identified; the algorithm,
which underwent a rewriting process to enable the prediction of more reactions and
to investigate the transformations that occur in the biosynthetic pathways of these
compounds, was then improved to the PRISM 3 version, which covers the predic-
tion of non-modular natural product assemblies and a broad set of adaptation
enzymes (enzymes that catalyse the so-called “tailoring reactions”). In this version,
the modelling of the biomolecule structure (scaffold) is presented as a chemical
graph and not as a linear permutation of modules. This approach is advantageous as
it enables the manipulation of structures at the atomic level, or individual bonds and
extends the prediction of structures to classes of molecules that cannot be effec-
tively modelled as linear sequences of residues.
Although the advancement of specialized tools in genomics and transcriptomics,
represents a leap in the bioinformatics era, it is still necessary that sophisticated
analytical techniques, like metabolomics, are applied together with what is offered
by in silico methods aiming at a better performance of the chemical characterization
of biological samples. These approaches allow us to deeply explore the chemical
diversity of biomolecule structures, and can help the understanding of new meta-
bolic pathways being corroborating in confirming the presence of compounds in
samples from target microorganisms.
3 Unveiling Microbial Chemical Interactions Based on Metabolomics Approaches 57

3.3 Chemical Diversity Investigation (Phenomics)

The dependence on synthetic pesticides and fertilizers in crops is mainly caused by

increasing abiotic stress caused by climate change, such as decreasing arable land
due to soil erosion, water scarcity, soil salinity, environmental degradation, and the
presence of heavy metals in the soil, which impedes the growth and productivity of
crops. The plant metabolism is negatively affected by unfavourable environmental
stimuli. Biofertilizers can be used as alternatives to chemical fertilizers and pesti-
cides and can provide new alternatives to improve plant growth and yield against
diseases caused by pathogens.
However, it is very important to note that not always the information obtained in
genomic approaches will be expressed in laboratory, or in vivo situations, since
these analyses show the potential ability of organisms to produce metabolites, how-
ever the secondary metabolite production is highly regulated and often produced
only under certain environmental conditions. In nature, these molecules have spe-
cific functions for their producers and their biosynthesis are energy and time con-
suming processes. Therefore along with their host evolutions, different strategies to
control the production of these metabolites have also evolved to avoid waste of
these resources. Understanding and being able to manipulate such strategies is one
of the most important steps in microbial ecology applications.
In nature, healthy plants host a large diversity of microorganisms (bacteria, fungi,
algae, and viruses), which are named the plant microbiome [28–30]. Distinct micro-
bial communities can be found on the same plant, on its surface or within its differ-
ent organs. These microbial sets, asymptomatic at first glance, present a continuum
of symbiotic interactions, ranging from commensalism to mutualism. So far, the
phylogenetic composition of bacterial communities in plants is relatively small con-
taining mainly bacteria of the phylum Actinobacteria, Bacteroidetes, Firmicutes and
Proteobacteria [31]. Fungi are more complex organisms and have a huge variety of
species. Fungi-plant interactions can be divided into three different niches: sapro-
phytic fungi, parasites and symbionts [32]. Each niche has a specific function within
the plant universe, and its comprehension is crucial to understand the ecological
interactions between host and the respective microorganisms.
For instance, in biological control, which is a process where there is suppression
of a pathogenic organism using another living organism, a huge variety of microor-
ganisms have been continuously isolated and have shown potential as biological
control agents (BCA) due to their ability to produce antimicrobial metabolites.
However, only a few strains are commercially available often from the genus
Bacillus sp. and Trichoderma sp.. That may occur once many of those isolated
microorganisms are not able to produce those antimicrobial molecules in the field or
even are unable to survive in the target environment. Therefore, it is important not
only to chemically characterize those strains, but also understands its adaptive dif-
ferences and search for the most suited strain for each particular application.
Furthermore different strains from a same microorganism but isolated from
58 L. C. de Carvalho et al.

different places (e.g. soil strains vs. phylosphere strain) are a crucial step to achieve
successful results in discovering a BCA.
Many of these bioactive molecules (allelochemicals) are derived from the sec-
ondary metabolism of these organisms. However, the major challenge in detecting
and identifying these molecules in the laboratory is related to the low production
that culminate in the need to perform large scale fermentations of these organisms
under conditions that are often not ideal for either growth or expression of those
metabolites. Thus, the development of improved identification techniques and,
therefore the understanding of the relationship between these organisms and their
hosts are of great importance.
For instance, an important source of these microorganisms is the soil around the
plant. Soil represents one of the richest microbial ecosystems on earth, however the
amount of nutrients is limited especially when compared to the rhizosphere region.
As a result, competition between organisms for these nutrient-rich regions shapes
the outcome of their interactions with the plant. Few bacterial phyla are found in the
most diverse soil biomes, including: Acidobacteria, Actinobacteria, Bacteroidetes,
Chloroflexi, Firmicutes and Proteobacteria; even smaller are the phyla found in the
rhizosphere and endosphere region (Proteobacteria, Actinobacteria and
Acidobacteria) [33, 34]. These results suggest that combinations between edaphic
and host factors form the composition of the microbiota in these regions. The litera-
ture points to a two-step process of selecting these microorganisms, which are found
in different proportions when compared to the surrounding soil biome. The first step
points to the host organism (plant) as the main driving force of attraction of micro-
organisms, in a process called rhizodeposition. This phenomenon is based on the
release of certain chemical biomarkers by the plant in the rhizosphere, responsible
for the attraction of these microorganisms. Some studies also show that the micro-
biota of these rhizodepposites is directly linked to the different parts of the root
(cap, hairy zone and mature root) reflecting local differences found in exudate
metabolites along the longitudinal axis of the roots [35]. The second step is based
on the host genotype, which is primarily responsible for the selection and differen-
tiation between rhizosphere and endosphere microorganisms. One hypothesis is that
the plant’s own innate immune system is primarily responsible for this selection of
endophytic microorganisms, which have evolved to camouflage their detection by
the plant’s immune system [36].
Therefore, given these nutritional limitations and adverse conditions in this envi-
ronment (soil), it is important not only to find a microorganism able to perform the
activity of interest (e.g., a plant growth promotion microorganism), but also strains
that could be able to survive in those environments. For instance, a strain that are
more efficient in recovering nutrients from the environment or that have a more
versatile metabolism have an advantage in mastering these ambient and are proba-
bly more adapted for promoting plant growth and/or pathogen inhibition.
Considering the vital and diverse functions of molecules secreted in interactions
with organisms, the importance of genomics and metabolomics is evident in order
to clarify the biology and ecology of plants and their associated microorganisms.
3 Unveiling Microbial Chemical Interactions Based on Metabolomics Approaches 59

Many secondary metabolites are not necessary for plant growth, but function as
signals and modifiers of the environment, so several of the secreted molecules act in
the interactions between organisms, which have elaborate systems for detecting
chemical clues from other organisms and from the environment [37]. Considering
the difficulty in understanding the role of secondary metabolites in biological inter-
actions, combinations of analytical characterization techniques and genomic tools
are commonly used as strategies to unveil these studies. The challenges encountered
in such metabolomics studies are related to metabolic dynamics, chemical diversity,
concentration gradient and cyclic, environmental or ontogenetic effects [38].
For the investigation on how environmental factors or experimental factors can
affect the metabolic profiles of plants and microorganisms, it is necessary to carry
out experiments in a systematic way in relation to the selection of variables and
optimization of these variables or factors, thus, metabolomic studies can be divided
into the following steps: experimental design, data acquisition, pre-processing, data
analysis, and data interpretation.
In general, the response to a phenomena in organisms comprise a series of
molecular, cellular cross-talk and signalling responses initiated through the detec-
tion of a specific or combined biotic or abiotic stresses that can result in the induc-
tion or suppression of secondary metabolites. Often understanding and simulating
these stimuli in a controlled environment within the laboratory is not an easy task,
but different approaches can be applied [39].
An alternative is elicitation, which, in addition to being an effective method to
enhance metabolite diversity production, it is also widely used to gain insights into
the understanding of the regulatory mechanisms of a particular biosynthetic path-
way. The elicitors, that belong to several chemical classes and present great struc-
tural diversity, act as signals and the beginning of the elicitation happens through
the perception of that signal, which in turn, happens through specific receptors pres-
ent in the host organism [40]. For instance, in plants, the method can consist in
inducing or increasing the production of a specific secondary metabolite through the
induction of defence or stress-induced responses in pilus root cutting and/or in plant
cell cultures of different species. Defence response mechanisms involve specific
modifications in the state of the metabolic gene expression network that affect pro-
tein synthesis that modulate primary and secondary biosynthetic pathways [41].
The elicitors are classified according to their nature, and may be biotic inducers,
with a defined or undefined composition; chemical stressors: salts of heavy metals,
osmotic stressors, gaseous substances, physical stressors and intracellular signalling
molecules, and also classified according to their origin, which can be exogenous
inducers or endogenous inducers [40].
The use of elicitors such as nutrients, trace elements, study of physical parame-
ters and chemical elicitors, as well as the co-cultivation of microbes and factors that
affect epigenetic control are methods that increase chemodiversity and, when com-
bined with genetic engineering, genomics, metabolomics and bioinformatics, exert
a great impact on the discovery of secondary metabolites in different microbial
species (e. g. Streptomyces).
60 L. C. de Carvalho et al.

The One Strain Many Componds (OSMAC) is another very interesting approach
to in promote the production of secondary metabolites in the laboratory. This
approach consists of systematized changes, among others, in the culture medium,
aiming to stimulate the production of different metabolites. Carbon and nitrogen are
major components in the culture medium, however different sources can be used.
The carbon is the basis for biomass production and is source of energy for all het-
erotrophs organisms and functions as source of carbon units for secondary metabo-
lites. The nitrogen is used in the synthesis of proteins and nucleic acids, and likewise
as source of N-containing units for secondary metabolites. The source type and the
C/N ratio have a significant influence on microbial secondary metabolism and fer-
mentation products [42].
Marine microbes, or microorganisms from different salinity environments,
inhabit a variety of environmental niches, which are ruled by different physical-­
chemical conditions. Salinity is an important factor and can determine some aspects
of the microbial metabolism along with temperature and pressure. Microorganisms
exposed to cultivation conditions with different levels of salinity can trigger differ-
ent synthetic pathways in order to restore osmotic imbalance, then activating hidden
biosynthetic gene clusters [43].

3.4 Metabolomics

Over the past few years, metabolomics has played a key role in basic research to
elucidate environmental effects, understand gene functions, and define cellular pro-
cesses seeking to improve the quality of human life. Microbial metabolomics is the
comprehensive analysis (qualitative and quantitative) of metabolites aiming objec-
tively to gather as much metabolic information as possible from an organism or
biological system to obtain a better understanding of the biochemical pathways or a
studied phenomena [3, 44]. In this analysis, the main interest is in molecules of
molar mass <1500 Da, hormones or signalling molecules, which act in vital func-
tions and participate in metabolic reactions.
Metabolites are categorized as end products of cellular processes and their roles
in biological systems can be considered as the final response to genetic, environ-
mental or changes resulting from some pathology and/or treatment. They can be
considered the closest representation of phenotype and, therefore, metabolomics
can provide insights into the cellular processes in response to some stimuli or inter-
actions [4]. The main challenges in the chemical analysis of these compounds are
related to the chemical complexity and heterogeneity, therefore, to characterizing
and quantifying these compounds is necessary to use specific methodologies and
equipment, according to the characteristics of each chemical class. Metabolomics
encompasses several analytical technologies, which need to be carefully selected,
according to the metabolites and metabolic pathway of interest, or the biological
question to be answered.
3 Unveiling Microbial Chemical Interactions Based on Metabolomics Approaches 61

Studies on microbial molecules have increased exponentially recently, and mass

spectrometry (MS) coupled with separation methods such as gas chromatography,
liquid chromatography to provide higher peak capacities and obtain qualitative and
quantitative analyses separation techniques, have become the commonly used tools
in this field. High-resolution mass spectrometry, such as time-of-flight mass spec-
trometry (ToF-MS) and Orbitrap are the most powerful mass analysers for the
global detection of metabolites. However, the tandem mass spectrometry method,
such as triple quadrupole mass spectrometry, is useful for target metabolomics stud-
ies and quantifications. Many studies can also use various MS-based platforms to
increase metabolite detection coverage. Other methods, such as nuclear magnetic
resonance (NMR) and capillary electrophoresis (CE-MS) have also been used, but
less frequently [3].
When it comes to microorganisms, researchers in the metabolomics field have
been continuously contributing to the discovery of unique metabolic pathways and
regulatory interactions, and these discoveries are essentially useful for understand-
ing the global metabolism of cells. Microbial metabolomics has been used in several
microbiological fields, such as identification of microorganisms, [45] cellular muta-
tions, research of functional genes [46] and identification of metabolic pathways
[47]. Another important point is also the applicability of metabolites due to their
bioactivities in various fields such as pharmacology, [48] agriculture, [49] cosmetol-
ogy. But when compared to other studies, the main disadvantage of the microbial
metabolomics studies is the high complexity of the samples and the presence of
classes of metabolites that are still little explored.
Metabolomics establishes that the understanding of organisms occurs through
the comparative analysis of metabolic profiles between individuals and/or popula-
tions subject to different genetic, environmental or pathological conditions, consid-
ering both the molecular size of the substances and the chemical class. Due to the
absence of a single technique capable of analysing the entire content of the metabo-
lome, a combination of tools can help to discover the nature, function and mode of
action of such metabolites.
As mentioned, the class of molecules to be investigated should be previously
selected before the selection of the analytical techniques used. For instance, studies
of microorganisms in environments such as the soil, that aims to seek microorgan-
isms’ communication or relationships among the microbial communities, probably
will obtain relevant information by studying volatile compounds, once they are fre-
quently associated to a greater reach in such environment. While for studies aiming
to characterize a direct interaction between microorganisms in a host (e.g., a plant)
or in a more compact environment, an interesting approach seeks to analyse water-­
soluble compounds. Nevertheless, given the complexity of these studies, the combi-
nation of different analytical techniques is important for a more comprehensive and
greater coverage of the detected metabolites in a biological interaction.
62 L. C. de Carvalho et al.

3.4.1 Gas Chromatography – Mass Spectrometer (GC-MS)

The main advantages of GC-MS are their excellent separation efficiency, ease of
use, robustness, and high cost-effectiveness. Initially, the technique was used only
with low resolution mass analysers, but nowadays we easily find GC coupled with
high resolution analysers such as TOF and orbitraps [50]. In general, techniques
coupled to gas chromatography are limited to volatile, nonpolar and low molecular
weight molecules, which can still be expanded with some sample derivatization
methods [51]. However, despite this limitation, molecules detected by GC are often
not detected using other separation techniques, which makes GC essential in several
metabolomics studies of microorganisms, especially in the study of volatile com-
pounds. Microbes produce and consume organic and inorganic volatile compounds
that drive microbial metabolism, signalling, and ecosystem-to-global interactions.
However, volatile metabolism (volatilomic) remains frequently separate from gen-
eral studies on microbial metabolism, leaving the relevance of VOCs to metabolism
overlooked [52].
Today, combined with high-resolution techniques such as orbitrap and ToF, the
analysis can be done with accuracies of up to ppb or ~30 ppm, respectively. Still
using two-dimensional chromatography (GCxGC) usually coupled with ToF, the
best achieved chromatographic separation can increases approximately by 3 times
the number of resolved peaks, which shows that despite the limitations, the analysis
with GC-MS still has a lot of applications. Additionally, several derivatization
methods have been improved and when there is a greater procedure on sample prep-
aration, reproducibility becomes important. Miyagawa and co-workers utilized
sequential derivatization to evaluate and successfully obtain good repeatability for the
peak areas of 52 selected metabolites [53]. But again, there is no perfect derivatization
method and classes of compounds can be favoured in detriment of others, so the
choice of the method must be done with caution, especially with regard to MS librar-
ies in which derivatized compounds are available. Generally, such libraries are spe-
cific for each type of derivatization, as for example at Fiehn library used for compounds
derivatized with N-methylN-­trimethylsilyltrifluoroacetamide [54].
The more sophisticated the method or the more comprehensive, the greater the
amount of data that is generated by each sample and consequently there is a greater
need for data processing. The emergence of novel bioinformatic tools has remark-
ably reduced the time required to process GCxGC-MS data, making this technique
highly feasible for large scale metabolomics research [55, 56].
3 Unveiling Microbial Chemical Interactions Based on Metabolomics Approaches 63

3.4.2 Liquid Chromatography – Mass Spectrometer (LC-MS)

Electrospray ionization (ESI) is an ionization method widely used in mass spec-

trometry (MS) [57]. Its relative simplicity, ease of coupling with liquid chromato-
graph (LC) and usability for a large number of analytes has made it the primary ion
source for today’s LC-MS instruments. This “soft” ionization technique enables the
direct observation of natural compounds as intact ions, being the more predominant
approach in metabolomics studies when coupled with LC separation techniques.
This ionization method allows the detection and quantification of several compo-
nents simultaneously. However, the great ability to detect different kinds of mole-
cules such as proteins, lipids, nucleotides, inorganic salts, can become a problem
once those molecules can also be founded in the culture medium used, or being
from another organism present in sample or even from the environment. This can
generate misconceptions in the results interpretation, especially in guided studies.
Additionally, the identification of a large number of detected ions using untargeted
metabolomics remains the major bottleneck for the efficiency of this technique [58].
In general LC-MS analysis can be divided in three main groups, target, [59] untar-
geted (Data dependent analysis, DDA) [60] and all ion fragmentation (Data inde-
pendent analysis, DIA) [61].
In targeted studies, specific compounds are evaluated; they frequently are quanti-
fied and compared to reference ranges. In this situation the mass spectrometer is set
to monitor selected ion transitions (MRM), reflecting individual target compounds
(and their internal standards). However, targeted approaches are usually limited to
specific metabolites and are less effective at finding unknown compounds. Using
high resolution mass analysers, hybrid mass spectrometer can offer several modes
of fragmentation, with different tools for structural characterization and metabolite
annotation allowing performing untargeted metabolomics [62, 63]. We can think in
it as a “discovery mode” process, which relies on differential comparison between
groups of samples (for example, cases versus controls); but can also be applicable
to individual samples. Untargeted metabolomics studies seek to identify only the
relevant compounds for the phenomena. This approach makes the methods less time
consuming and robust. Generally, untargeted metabolic workflow follow these three
steps: Profiling, which involves the search for metabolites with statistically signifi-
cant variations within group samples sets; Compound identification, where the
annotation of relevant compounds are made; Interpretation, which seeks uncovering
biological connections between the metabolites and the studied phenomena, that
may result in insights for other experiments. Several approaches can be used in each
one of those steps, and will depend on the type of samples and instruments [64].
Recently, an MS analysis mode called all ion fragmentation (AIF), a data inde-
pendent analysis mode, has gained prominence in metabolomic studies, basically all
m/z range are analysed by fragmentation of small ranges of mass which are selected
in a stepped way. This method is also an untargeted approach, however it tries to
cover the losses in feature detections, which happens in untargeted analysis and
provides MS/MS experiments with no additional experimental time cost for
64 L. C. de Carvalho et al.

metabolite annotation. Nonetheless a more elaborated data processing is need after

sample analysis [65]. Once the MS/MS generated data were a sum of all features
within the selected range, we obtained a pseudo-MS/MS spectra, and such datasets
require deconvolutions and reconstruction of parent–fragment relationships before
data annotation. Recently, Ebbels and co-workers using AIF were able to demon-
strate the use of this workflow in three human serum datasets, achieving good preci-
sion and recall values for feature annotations (82–92% and 82–85%, respectively)
[61]. This approach using DIA data for metabolomics studies are quite new, and
despite the easiness of data acquisition, a lot of efforts are still needed to perform
data annotation for those kind of data (e.g. AIF, SWATH [66], etc.) As shown, dif-
ferent approaches can be used applying those techniques in microbial studies and
also different tools can be used for treatment of the generated data. For more in
formation concerning this topic, a recent review can be found in literature [4].

3.4.3 Direct Analysis of Biological Samples (DI, MALDI

and DESI)

Direct analysis, which does not require chromatographic separation and chemical
derivatization, is an effective alternative for high throughput metabolomics analysis.
Several possibilities are available, such as direct infusion (DI) for liquid samples or
MALDI or DESI for solid samples. Each one of those techniques having different
applications and sample preparation procedures. DI analysis is an easy way to
access a rapidly overview of the samples, however inherent shortcomings such as
the matrix effect, data alignment and isobars differentiation are a concern and lim-
ited the type of information retrieved about the phenomena. In general, this type of
analysis is used only for an initial assessment of the sample, for a better construction
of the pipeline of further MS analysis.
MALDI and DESI have similar disadvantages, but also share several advantages
including the ability to analyse complex mixtures with less ion suppression com-
pared to DI-ESI, high sensitivity, high tolerance to salts and contaminants, simple
and fast sample preparation, low time consumption to obtain the spectra, and the
production of mostly monocharged species. But, the most prominent advantage of
such techniques is the possibility to perform imaging experiments (IMS), which
opens a new era of information in microbial metabolomics, due to the possibility to
provide information concerning the spatial and temporal distribution of metabolites
in biological samples [67].
MALDI (Matrix Assisted Laser Desorption Ionization) is a method that allows
the ionization and transfer of a sample in the solid phase to the gas phase, which was
introduced by Karas and Hillenkamp in 1988. This technique has been used in the
metabolic studies of different strains of microorganisms, [68] especially in microor-
ganism identification, [69] in the analysis of protein and lipid fingerprints and
chemical screening. However, this technique has been expanded and used in the
3 Unveiling Microbial Chemical Interactions Based on Metabolomics Approaches 65

spatial study of metabolites and lipids in different biological matrices, analysis of

microbial natural products, microbial interactions between mixed cultures and met-
abolic characterization of microorganisms, and analysis of single plant cells [70, 71].
DESI (Desorption electrospray ionizations), introduced by Cooks in 2004, deliv-
ers mass spectra with very low fragmentation in either positive or negative ioniza-
tion modes. DESI is a technique for surface analysis where electrically charged
microdroplets of a solvent collide with the surface of a sample, and as a result,
secondary ions are formed upon impact. These secondary ions are then directed into
a mass spectrometer (MS), which allows for near-instantaneous analysis [72]. DESI
can be applied to investigate highly complex sample surfaces, from polymeric mate-
rials [73] to biological tissues and even fluids [74], allowing the detection and semi-­
quantification of a variety of polar or non-polar small molecules, which contribute
to the increasing popularity of DESI.
IMS has established itself as a highly efficient and reliable tool in capturing the
chemical image of both large and small molecules in different life science disci-
plines. IMS techniques combine the specificity, selectivity and sensitivity of MS
with spatially resolved chemical information and could be used for different pur-
poses [75]. At first, IMS methods applied to bacterial cultures were exclusively to
identify them through their chemical and lipid composition. A good example was
the cyclic lipopeptide Surfactin C15, a known metabolite commonly produced by
Bacillus sp., which can be observed on agar, making possible the taxonomic identi-
fication of the bacterium directly in a Petri dish, several other metabolites can be
used with those purpose even without identification [45, 76].
Another applicability of IMS is in the investigation of NP natural products (pri-
mary and secondary metabolites) from bacterial and/or fungal cultures. For exam-
ple, in a study by Angolini and co-workers, metal-sequestering siderophores were
imaged directly from the agar culture of Streptomyces wadayamensis allowing the
authors to understand the functions of siderophores and propose their roles in plant-­
microorganism interactions [9]. Other study performed with Penicillium strictum
found that polyhydroxy anthraquinones are inhibitors of quorum sensing, a type of
microbial communication. DESI-IMS demonstrated that polyhydroxyanthraqui-
nones were produced in the mycelia of the fungus and expressed differently over
time [77].
Therefore, IMS experiments are constantly proving to be a very effective method
to understand chemical interactions/exchanges between the same, similar or differ-
ent microorganisms, as well as the characterization of interaction of micro and
macro-organisms.

3.5 Perspectives

Understanding the evolutionary forces that shape chemically mediated ecological

interactions and use that knowledge to bring benefits to humanity and environment,
require a multidisciplinary approach, and in this scenario the modern tools and
66 L. C. de Carvalho et al.

technologies showed in this chapter is just a beginning of a promise future. The

microbial metabolomics in life science has just recently emerged and is evidently
increasing. Additionally, together with other Omic sciences such as genomics, phe-
nomics and proteomics, microbial metabolomics are becoming a robust platform
for fast and assertive chemical characterization of these biomolecules that will help
to understand the relationships between the organisms in all kind of studied phe-
nomena. We expected that with the developments of advanced analytical tools and
data analysis methods, new workflows on microbial metabolomics will be created
and utilized more often in the near future.

Acknowledgments This work was supported by São Paulo Research Foundation (FAPESP, grant
number 2019/08853-9); and by National Council for the Improvement of Higher Education
(CAPES) – Finance Code 001. And the National Council for Scientific and Technological
Development (CNPq).

References

1. Bijlani S, Stephens E, Singh NK et al (2021) Advances in space microbiology. iScience

24:102395. https://doi.org/10.1016/j.isci.2021.102395
2. Oyedeji AB, Green E, Adebiyi JA et al (2021) Metabolomic approaches for the determina-
tion of metabolites from pathogenic microorganisms: a review. Food Res Int 140. https://doi.
org/10.1016/J.FOODRES.2020.110042
3. Ye D, Li X, Shen J, Xia X (2022) Microbial metabolomics: from novel technologies to
diversified applications. TrAC Trends Anal Chem 148:116540. https://doi.org/10.1016/J.
TRAC.2022.116540
4. Bauermeister A, Mannochio-Russo H, Costa-Lotufo LV et al (2021) Mass spectrometry-based
metabolomics in microbiome investigations. Nat Rev Microbiol 203(20):143–160. https://doi.
org/10.1038/s41579-­021-­00621-­9
5. International Allelopathy Society (1998) Allelopathy: is this the definition we want? In: Willis
R (ed) lAS newsletter 2: 5. University of Melbourne, Parkville
6. Mitchell-Olds T, Gershenzon J, Baldwin I, Boland W (1998) Chemical ecology in the molecu-
lar era. Trends Plant Sci 3:362–365. https://doi.org/10.1016/S1360-­1385(98)01296-­5
7. Fraser CM, Eisen JA, Salzberg SL (2000) Microbial genome sequencing. Nature
4066797(406):799–803. https://doi.org/10.1038/35021244
8. Watrous J, Roach P, Alexandrov T et al (2012) Mass spectral molecular networking of living
microbial colonies. Proc Natl Acad Sci U S A 109. https://doi.org/10.1073/PNAS.1203689109/-­/
DCSUPPLEMENTAL/SAPP.PDF
9. Angolini CFF, Vendramini PH, Araújo FDS et al (2015) Direct protocol for ambient mass
spectrometry imaging on agar culture. Anal Chem 87:6925–6930. https://doi.org/10.1021/
ACS.ANALCHEM.5B01538
10. Newman DJ, Cragg GM (2014) Marine-sourced anti-cancer and cancer pain control agents
in clinical and late preclinical development †. Mar Drugs 12:255. https://doi.org/10.3390/
MD12010255
11. Rossiter SE, Fletcher MH, Wuest WM (2017) Natural products as platforms to over-
come antibiotic resistance. Chem Rev 117:12415–12474. https://doi.org/10.1021/ACS.
CHEMREV.7B00283
3 Unveiling Microbial Chemical Interactions Based on Metabolomics Approaches 67

12. Seca AML, Pinto DCGA (2018) Plant secondary metabolites as anticancer agents: suc-
cesses in clinical trials and therapeutic application. Int J Mol Sci 19. https://doi.org/10.3390/
IJMS19010263
13. Keswani C, Singh HB, García-Estrada C et al (2020) Antimicrobial secondary metabo-
lites from agriculturally important bacteria as next-generation pesticides. Appl Microbiol
Biotechnol 104:1013–1034. https://doi.org/10.1007/S00253-­019-­10300-­8
14. Balderas-Ruíz KA, Gómez-Guerrero CI, Trujillo-Roldán MA et al (2021) Bacillus velezensis
83 increases productivity and quality of tomato (Solanum lycopersicum L.): pre and post-
harvest assessment. Curr Res Microb Sci 2. https://doi.org/10.1016/J.CRMICR.2021.100076
15. Albarano L, Esposito R, Ruocco N, Costantini M (2020) Genome mining as new challenge in
natural products discovery. Mar Drugs 18. https://doi.org/10.3390/MD18040199
16. Ziemert N, Alanjary M, Weber T (2016) The evolution of genome mining in microbes – a
review. Nat Prod Rep 33:988–1005. https://doi.org/10.1039/C6NP00025H
17. Chandra Mohana N, Yashavantha Rao HC, Rakshith D et al (2018) Omics based approach for
biodiscovery of microbial natural products in antibiotic resistance era. J Genet Eng Biotechnol
16:1. https://doi.org/10.1016/J.JGEB.2018.01.006
18. Kenshole E, Herisse M, Michael M, Pidot SJ (2021) Natural product discovery through
microbial genome mining. Curr Opin Chem Biol 60:47–54. https://doi.org/10.1016/J.
CBPA.2020.07.010
19. Blin K, Shaw S, Kloosterman AM et al (2021) AntiSMASH 6.0: improving cluster detec-
tion and comparison capabilities. Nucleic Acids Res 49:W29–W35. https://doi.org/10.1093/
NAR/GKAB335
20. Medema MH, Blin K, Cimermancic P et al (2011) AntiSMASH: rapid identification, annota-
tion and analysis of secondary metabolite biosynthesis gene clusters in bacterial and fungal
genome sequences. Nucleic Acids Res 39. https://doi.org/10.1093/NAR/GKR466
21. Villebro R, Shaw S, Blin K, Weber T (2019) Sequence-based classification of type II polyketide
synthase biosynthetic gene clusters for antiSMASH. J Ind Microbiol Biotechnol 46:469–475.
https://doi.org/10.1007/S10295-­018-­02131-­9
22. Conway KR, Boddy CN (2013) ClusterMine360: a database of microbial PKS/NRPS biosyn-
thesis. Nucleic Acids Res 41. https://doi.org/10.1093/NAR/GKS993
23. Kautsar SA, Blin K, Shaw S et al (2020) MIBiG 2.0: a repository for biosynthetic gene clusters
of known function. Nucleic Acids Res 48:D454–D458. https://doi.org/10.1093/NAR/GKZ882
24. Bahram M, Hildebrand F, Forslund SK et al (2018) Structure and function of the global topsoil
microbiome. Nature 5607717(560):233–237. https://doi.org/10.1038/s41586-­018-­0386-­6
25. Navarro-Muñoz JC, Selem-Mojica N, Mullowney MW et al (2019) A computational frame-
work to explore large-scale biosynthetic diversity. Nat Chem Biol 161(16):60–68. https://doi.
org/10.1038/s41589-­019-­0400-­9
26. Skinnider MA, Merwin NJ, Johnston CW, Magarvey NA (2017) PRISM 3: expanded pre-
diction of natural product chemical structures from microbial genomes. Nucleic Acids Res
45:W49–W54. https://doi.org/10.1093/NAR/GKX320
27. Skinnider MA, Dejong CA, Rees PN et al (2015) Genomes to natural products PRediction
Informatics for Secondary Metabolomes (PRISM). Nucleic Acids Res 43:9645–9662. https://
doi.org/10.1093/nar/gkv1012
28. Bulgarelli D, Rott M, Schlaeppi K et al (2012) Revealing structure and assembly cues for
Arabidopsis root-inhabiting bacterial microbiota. Nature 488:91–95. https://doi.org/10.1038/
NATURE11336
29. Lindow SE, Brandl MT (2003) Microbiology of the phyllosphere. Appl Environ Microbiol
69:1875–1883. https://doi.org/10.1128/AEM.69.4.1875-­1883.2003/ASSET/3E78B043-­2567-­
4A8F-­953F-­5B19B0DA4010/ASSETS/GRAPHIC/AM0431728002.JPEG
30. Delmotte N, Knief C, Chaffron S et al (2009) Community proteogenomics reveals insights
into the physiology of phyllosphere bacteria. Proc Natl Acad Sci 106:16428–16433. https://
doi.org/10.1073/PNAS.0905240106
68 L. C. de Carvalho et al.

31. Bulgarelli D, Schlaeppi K, Spaepen S et al (2013) Structure and functions of the bac-
terial microbiota of plants. Annu Rev Plant Biol 64:807–838. https://doi.org/10.1146/
ANNUREV-­ARPLANT-­050312-­120106
32. Spiteller P (2015) Chemical ecology of fungi. Nat Prod Rep 32:971–993. https://doi.
org/10.1039/C4NP00166D
33. Fierer N, Strickland MS, Liptzin D et al (2009) Global patterns in belowground communities.
Ecol Lett 12:1238–1249. https://doi.org/10.1111/J.1461-­0248.2009.01360.X
34. Gans J, Wolinsky M, Dunbar J (2005) Computational improvements reveal great bacterial
diversity and high metal toxicity in soil. Science 309:1387–1390. https://doi.org/10.1126/
SCIENCE.1112665
35. DeAngelis KM, Brodie EL, DeSantis TZ et al (2008) Selective progressive response of soil
microbial community to wild oat roots. ISME J 32(3):168–178. https://doi.org/10.1038/
ismej.2008.103
36. Jones JDG, Dangl JL (2006) The plant immune system. Nature 4447117(444):323–329.
https://doi.org/10.1038/nature05286
37. Kellogg J, Kang S (2020) Metabolomics, an essential tool in exploring and harnessing microbial
chemical ecology. Phytobiomes J 4:195–210. https://doi.org/10.1094/PBIOMES-­04-­20-­0032-­
RVW/ASSET/IMAGES/LARGE/PBIOMES-­04-­20-­0032-­RVWF4.JPEG
38. Pilon AC, Selegato DM, Fernandes RP et al (2020) Metabolômica de plantas: métodos e desa-
fios. Quim Nova 43:329–354. 10.21577/0100-4042.20170499
39. Hautbergue T, Jamin EL, Debrauwer L et al (2018) From genomics to metabolomics, moving
toward an integrated strategy for the discovery of fungal secondary metabolites. Nat Prod Rep
35:147–173. https://doi.org/10.1039/C7NP00032D
40. Chamkhi I, Benali T, Aanniz T et al (2021) Plant-microbial interaction: the mechanism and
the application of microbial elicitor induced secondary metabolites biosynthesis in medicinal
plants. Plant Physiol Biochem 167:269–295. https://doi.org/10.1016/J.PLAPHY.2021.08.001
41. Halder M, Sarkar S, Jha S (2019) Elicitation: a biotechnological tool for enhanced production
of secondary metabolites in hairy root cultures. Eng Life Sci 19:880. https://doi.org/10.1002/
ELSC.201900058
42. Pan R, Bai X, Chen J et al (2019) Exploring structural diversity of microbe secondary
metabolites using OSMAC strategy: a literature review. Front Microbiol 10:294. https://doi.
org/10.3389/FMICB.2019.00294/BIBTEX
43. Romano S, Jackson SA, Patry S, Dobson ADW (2018) Extending the “One Strain Many
Compounds” (OSMAC) principle to marine microorganisms. Mar Drugs 16. https://doi.
org/10.3390/MD16070244
44. Müller R, Wink J (2014) Future potential for anti-infectives from bacteria – how to exploit
biodiversity and genomic potential. Int J Med Microbiol 304:3–13. https://doi.org/10.1016/J.
IJMM.2013.09.004
45. Singhal N, Kumar M, Kanaujia PK, Virdi JS (2015) MALDI-TOF mass spectrometry: an
emerging technology for microbial identification and diagnosis. Front Microbiol 6:791. https://
doi.org/10.3389/FMICB.2015.00791/BIBTEX
46. Andersen JS, Mann M (2000) Functional genomics by mass spectrometry. FEBS Lett
480:25–31. https://doi.org/10.1016/S0014-­5793(00)01773-­7
47. Heaney LM (2020) Applying mass spectrometry-based assays to explore gut microbial metab-
olism and associations with disease. Clin Chem Lab Med 58:719–732. https://doi.org/10.1515/
CCLM-­2019-­0974/ASSET/GRAPHIC/J_CCLM-­2019-­0974_FIG_001.JPG
48. Dehghan Esmatabadi MJ, Bozorgmehr A, Hajjari SN et al (2017) Review of new insights
into antimicrobial agents. Cell Mol Biol (Noisy-le-Grand) 63:40–48. https://doi.org/10.14715/
CMB/2017.63.2.6
49. Ray P, Lakshmanan V, Labbé JL, Craven KD (2020) Microbe to microbiome: a paradigm shift
in the application of microorganisms for sustainable agriculture. Front Microbiol 11:3323.
https://doi.org/10.3389/FMICB.2020.622926/BIBTEX
3 Unveiling Microbial Chemical Interactions Based on Metabolomics Approaches 69

50. Rey-Stolle F, Dudzik D, Gonzalez-Riano C et al (2022) Low and high resolution gas
chromatography-­mass spectrometry for untargeted metabolomics: a tutorial. Anal Chim Acta
1210. https://doi.org/10.1016/J.ACA.2021.339043
51. Fiehn O (2016) Metabolomics by gas chromatography-mass spectrometry: the combina-
tion of targeted and untargeted profiling. Curr Protoc Mol Biol 114:30.4.1. https://doi.
org/10.1002/0471142727.MB3004S114
52. Meredith LK, Tfaily MM (2022) Capturing the microbial volatilome: an oft overlooked “ome”.
Trends Microbiol. https://doi.org/10.1016/J.TIM.2021.12.004
53. Miyagawa H, Bamba T (2019) Comparison of sequential derivatization with concurrent meth-
ods for GC/MS-based metabolomics. J Biosci Bioeng 127:160–168. https://doi.org/10.1016/J.
JBIOSC.2018.07.015
54. Kind T, Wohlgemuth G, Lee DY et al (2009) FiehnLib: mass spectral and retention index
libraries for metabolomics based on quadrupole and time-of-flight gas chromatography/mass
spectrometry. Anal Chem 81:10038–10048. https://doi.org/10.1021/AC9019522/SUPPL_
FILE/AC9019522_SI_002.TIF
55. Tian TF, Wang SY, Kuo TC et al (2016) Web server for peak detection, baseline correction, and
alignment in two-dimensional gas chromatography mass spectrometry-based metabolomics
data. Anal Chem 88:10395–10403. https://doi.org/10.1021/acs.analchem.6b00755
56. Wei X, Shi X, Koo I et al (2013) MetPP: a computational platform for comprehensive two-­
dimensional gas chromatography time-of-flight mass spectrometry-based metabolomics.
Bioinformatics 29:1786–1792. https://doi.org/10.1093/BIOINFORMATICS/BTT275
57. Fenn JB, Mann M, Meng CK et al (1989) Electrospray ionization for mass spectrometry of
large biomolecules. Science 246:64–71. https://doi.org/10.1126/SCIENCE.2675315
58. Zhou B, Xiao JF, Tuli L, Ressom HW (2012) LC-MS-based metabolomics. Mol BioSyst
8:470. https://doi.org/10.1039/C1MB05350G
59. Roberts LD, Souza AL, Gerszten RE, Clish CB (2012) Targeted metabolomics. Curr Protoc
Mol Biol CHAPTER:Unit 30.2:1–24. https://doi.org/10.1002/0471142727.MB3002S98
60. Schrimpe-Rutledge AC, Codreanu SG, Sherrod SD, McLean JA (2016) Untargeted metabo-
lomics strategies – challenges and emerging directions. J Am Soc Mass Spectrom 27:1897.
https://doi.org/10.1007/S13361-­016-­1469-­Y
61. Graça G, Cai Y, Lau CHE et al (2022) Automated annotation of untargeted all-ion fragmenta-
tion LC-MS metabolomics data with MetaboAnnotatoR. Anal Chem 94:3446–3455. https://
doi.org/10.1021/ACS.ANALCHEM.1C03032/SUPPL_FILE/AC1C03032_SI_004.XLSX
62. Alonso A, Marsal S, Julià A (2015) Analytical methods in untargeted metabolo-
mics: state of the art in 2015. Front Bioeng Biotechnol 3:23. https://doi.org/10.3389/
FBIOE.2015.00023/BIBTEX
63. Gertsman I, Barshop BA (2018) Promises and pitfalls of untargeted metabolomics. J Inherit
Metab Dis 41:355. https://doi.org/10.1007/S10545-­017-­0130-­7
64. Tobolkina E, González-Rui V, Meister I et al (2022) Challenges in ESI-MS-based untargeted
metabolomics. Chimia (Aarau) 76:90. https://doi.org/10.2533/chimia.2022.90
65. Chaleckis R, Naz S, Meister I, Wheelock CE (2018) LC-MS-based metabolomics of biofluids
using All-Ion Fragmentation (AIF) acquisition. Methods Mol Biol 1730:45–58. https://doi.
org/10.1007/978-­1-­4939-­7592-­1_3
66. Shikshaky H, Ahmed EA, Anwar AM et al (2022) A novel approach of SWATH-based metabo-
lomics analysis using the human metabolome database spectral library. Int J Mol Sci 23:10908.
https://doi.org/10.3390/IJMS231810908/S1
67. Silva R, Lopes NP, Silva DB (2016) Application of MALDI mass spectrometry in natural
products analysis. Planta Med 82:671–689. https://doi.org/10.1055/S-­0042-­104800
68. Rahi P, Vaishampayan P (2020) Editorial: MALDI-TOF MS application in microbial ecology
studies. Front Microbiol 10:2954. https://doi.org/10.3389/FMICB.2019.02954/BIBTEX
69. Loucif L, Bendjama E, Gacemi-Kirane D, Rolain JM (2014) Rapid identification of
Streptomyces isolates by MALDI-TOF MS. Microbiol Res 169:940–947. https://doi.
org/10.1016/J.MICRES.2014.04.004
70 L. C. de Carvalho et al.

70. Li H, Li Z (2022) The exploration of microbial natural products and metabolic interac-
tion guided by mass spectrometry imaging. Bioengineering 9:707. https://doi.org/10.3390/
BIOENGINEERING9110707
71. Mehta A, Silva LP (2015) MALDI-TOF MS profiling approach: how much can we get from it?
Front Plant Sci 6:184. https://doi.org/10.3389/FPLS.2015.00184/BIBTEX
72. Takáts Z, Wiseman JM, Cooks RG (2005) Ambient mass spectrometry using desorption elec-
trospray ionization (DESI): instrumentation, mechanisms and applications in forensics, chem-
istry, and biology. J Mass Spectrom 40:1261–1275. https://doi.org/10.1002/JMS.922
73. Pierson EE, Midey AJ, Forrest WP et al (2020) Direct drug analysis in polymeric implants
using Desorption Electrospray Ionization – Mass Spectrometry Imaging (DESI-MSI). Pharm
Res 37. https://doi.org/10.1007/S11095-­020-­02823-­X
74. Garza KY, Feider CL, Klein DR et al (2018) Desorption electrospray ionization mass
spectrometry imaging of proteins directly from biological tissue sections. Anal Chem
90:7785–7789. https://doi.org/10.1021/ACS.ANALCHEM.8B00967/ASSET/IMAGES/
LARGE/AC-­2018-­00967J_0003.JPEG
75. Parrot D, Papazian S, Foil D, Tasdemir D (2018) Imaging the unimaginable: Desorption
Electrospray Ionization – Imaging Mass Spectrometry (DESI-IMS) in natural product
research. Planta Med 84:584. https://doi.org/10.1055/S-­0044-­100188
76. Strejcek M, Smrhova T, Junkova P, Uhlik O (2018) Whole-cell MALDI-TOF MS versus 16S
rRNA gene analysis for identification and dereplication of recurrent bacterial isolates. Front
Microbiol 9:1294. https://doi.org/10.3389/FMICB.2018.01294/BIBTEX
77. Figueroa M, Jarmusch AK, Raja HA et al (2014) Polyhydroxyanthraquinones as quorum
sensing inhibitors from the guttates of Penicillium restrictum and their analysis by desorption
electrospray ionization mass spectrometry. J Nat Prod 77:1351–1358. https://doi.org/10.1021/
NP5000704
Chapter 4
Chemical-Biology and Metabolomics
Studies in Phage-Host Interactions

Rodolfo Dantas, Marcelo Brocchi, and Taícia Pacheco Fill

4.1 Introduction

In the course of history involving the discovery of different molecules that changed
the course of civilization in the last century, one can highlight the importance of
antibiotics. This group of substances has made a milestone in the history of medi-
cine due to the advancement in treating different diseases caused by various bacteria
[1]. The introduction of these compounds as medicines has been saving millions of
lives worldwide, allowing targeted treatments and the performance of different sur-
gical procedures with a higher level of biosecurity, as an example of organ trans-
plantation [2].
However, with the inappropriate use of antibiotics [3], there has been a parallel
evolution of antimicrobial resistance (AMR) to these compounds, awakening a new
global public health problem: the need for alternative strategies to promote the con-
trol of different resistant pathogens [4, 5].
Bacteria are classified as significant pathogens causing morbidity and mortality
worldwide due to the possibility of presenting natural or acquired resistance to dif-
ferent antibiotics [6–8]. When resistance presents itself naturally, it can be intrinsic,
expressed by all species, or induced, where previous exposure to antibiotics is nec-
essary to activate the chromosomal or plasmid genes responsible for resistance.
Intrinsic resistance involves the mechanisms of permeability reduction of the outer
membrane (gram-negative bacteria), biosynthesis of enzymes that inactivate natural

R. Dantas · T. Pacheco Fill (*)

Institute of Chemistry, University of Campinas, Campinas, São Paulo, Brazil
e-mail: [email protected]; [email protected]
M. Brocchi
Institute of Biology, University of Campinas, Campinas, São Paulo, Brazil
e-mail: [email protected]

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 71

T. Pacheco Fill (ed.), Microbial Natural Products Chemistry, Advances in Experimental
Medicine and Biology 1439, https://doi.org/10.1007/978-3-031-41741-2_4
72 R. Dantas et al.

antibiotics, and the activity of efflux pumps [7, 9]. For induced resistance, mecha-
nisms involving multidrug efflux pumps are reported, among others [9].
These natural mechanisms allow resistance to antibiotics not as pronounced as
the mechanisms of acquired resistance since the acquired mechanisms are obtained
from horizontal gene transfer mediated by plasmids, other mobile DNA elements or
mutations in chromosomal DNA, which favors greater dissemination and expres-
sion of resistant characteristics [7]. The mechanisms involved in acquired resistance
include using efflux pumps for antibiotic excretion, modification of the drug target,
and drug inactivation [9].
All this specificity found in the metabolism of these pathogens in the face of
treatment processes with old and new molecules encourages global attention to new
government programs capable of outlining strategies to control the incidence of the
main strains with a high degree of resistance. In response, in 2017, the World Health
Organization (WHO) and several member states conducted a systematic survey of
the main strains of bacteria that cause the most significant risk to human and animal
health [10].
In a more current scenario and complementary way, in 2019, the Center for
Disease Control and Prevention (CDC) published a second list of pathogenic bacte-
ria and their respective levels of importance in the search for new strategies [11].
Examples of pathogens with a critical priority level include Acinetobacter bauman-
nii, Pseudomonas aeruginosa, and bacteria of the Enterobacteriaceae family, such
as Escherichia coli, Klebsiella spp., and Enterobacter. For pathogens with a high
priority level, there are Enterococcus faecium, Staphylococcus aureus, Helicobacter
pylori, Campylobacter spp., Salmonella spp., and Neisseria gonorrhoeae.
It is estimated that non-compliance with strategic plans aimed at the adequate
treatment of resistant strains will generate, by 2050, the death of 10 million people,
as well as the expenditure of 100 trillion dollars on security, economic, and health
services as a result of the antimicrobial resistance [12]. Therefore, investigating in
depth the mechanisms capable of discovering new ways for bacterial control puts us
in a position of social triumph in the face of so many possible consequences from a
new pandemic. Thus, using a more critical look at the interactions of bacteria with
the different actors present in the environment offers an opportunity to revisit the
strategies explored in the pre-antibiotic era.
One of the strategies pointed out in official WHO documents, which is closely
linked to the study of microbial ecology involving pathogenic bacteria, is the use of
bacteriophages for the treatment of infections. Bacteriophages, or simply phages,
are viruses capable of infecting bacteria by different mechanisms and using the
metabolic machinery of bacteria to replicate causing, in most of the cases, the host’s
death as the final event of the infection process [13].
The use of phages to control bacterial diseases appeared only two decades before
antibiotic-based therapy, at the end of the nineteenth century [13]. However, the
development of phage therapy was negatively impacted due to the favorable sce-
nario for the discovery and use of antibiotics, lack of reproducibility and research in
studies with phage therapy, and the existence of profit from the establishment of
large pharmaceutical industries in different countries [4, 14].
4 Chemical-Biology and Metabolomics Studies in Phage-Host Interactions 73

Currently, phage therapy is gaining prominence as a promising strategy in bacte-

rial control, as it is understood that phages are highly abundant infectious agents in
nature and possess an antimicrobial action mediated by high specificity against bac-
teria. In most cases of phages in bacterial control, there are no reports of harm to
human health, providing rare cases of induction of antibiotic resistance via horizon-
tal gene transfer to hosts [14, 15].
In addition to their use as biocontrol agents, phages can be used to detect and
diagnose bacteria and potential biological agents for the biotechnology industry, as
they encode enzymes that degrade biofilms from persistent infections [13].
Several studies have been published to use the ability and specificity of phages to
control different types of bacterial infections, whether in the agricultural [16, 17],
food [18–20], pre-clinical [21–23], or clinical sectors [23–25]. When studying the
ability to control infections caused by pathogens classified as high risk and critical
risk, researchers have pointed out the success of using phages in the control of
P. aeruginosa [26–29], A. baumannii [30–32], Klebsiella pneumoniae [33], E. coli
[34, 35], and Staphylococcus aureus [36, 37].
Several mechanisms are currently reported in the literature, both about how
phages interact with their hosts to replicate and the mechanisms used by bacteria to
dodge the invasion of these specialized viruses [14, 38, 39]. The high level of spe-
cialization of the virus-bacteria interaction leads us to question the biochemical
events that may be associated with the process of action of these promising agents
in the interaction with pathogenic bacteria. These interactions may involve forming
or regulating primary or secondary metabolites. Understanding these mechanisms
may be the key to the search for different classes of compounds involved in different
pathways of bacterial metabolism deactivation [40–42].
Within this bias of the production of compounds arising from the virus-bacteria
interaction, several studies have recently been developed considering the production
of peptides and enzymes as key compounds in controlling different pathogenic bac-
teria [43–57]. This relationship between phages and their hosts, derived from mil-
lions of years of evolution, reveals peptides and enzymes as leading compounds in
the experimental design of new drugs, capable of acting through different mecha-
nisms and sites compared to bacterial control by antibiotics.

4.2 Viruses, Bacteria, and Their Interactions

Unlike fungi, bacteria, and other microorganisms, viruses are not classified as cel-
lular organisms, having often been questioned regarding their origin, definition, and
classification as living entities [58–60]. According to Patrick Forterre (2010), when
infecting a cell, viruses can be understood as complex living entities that reprogram
the infected cell into a new organism. This characteristic makes them capable of
controlling the host’s metabolism to produce virions, refuting the idea that viruses
are not living organisms [58, 59].
74 R. Dantas et al.

With the advancement of science, new virus definitions have been presented to
consider the genome’s dissemination and propagation properties as essential param-
eters in its definition, starting to consider them as virion-producing organisms [60].
Virions, in turn, should not be used as synonyms for viruses and are defined as meta-
bolically inert infectious viral particles considered as the means responsible for dis-
seminating genetic information, as well as plant seeds [59, 60].
Soon, it became erroneous to understand viruses as tiny and microscopic biologi-
cal entities, as these physical characteristics define virions, not viruses. Another
factor that makes it impossible to define viruses within a spectrum of non-living
entities is the discovery of metabolic genes in the viral genomes of giant viruses,
which are considered responsible for the reconfiguration of the cellular metabolism
of the host organism, which goes against the requirements for classifying a biologi-
cal entity as living [61]. Therefore, we will understand, from this context and
according to Patrick Forterre (2016), that a virus, by free translation, can be concep-
tualized as being the process that integrates all aspects of viral reproduction.
Viral particles are the most abundant biological entities on the entire planet, and
in the marine environment scenario, they are considered ten times more abundant
than bacterial cells [59]. These, in turn, exist in their domain in the phylogenetic tree
of life, the Bacteria domain, emphasizing their diversity and uniqueness, being able
to be pathogenic or symbionts of several multicellular organisms [62–66]. As an
organism, bacteria are distinguished from viruses because we understand that they
do not need to become parasites to establish an active metabolism. After all, their
genetic material is dispersed in a cytoplasm, where the replication, transcription,
and translation steps provide their multiplication and protein synthesis by ribosomes
and messenger RNAs independently [67, 68].
On the other hand, bacterial viruses, bacteriophages, or simply phages constantly
interact with their hosts and depend exclusively on bacteria to develop their differ-
ent life cycles, classified as lytic, lysogenic, chronic and pseudolysogenic, the latter
being less common [69]. For the lytic cycle, the phage uses specific proteins located
at the ends of the tail fibers that interact with molecules present on the surface of
bacterial cells, such as lipopolysaccharides and peptidoglycans [51]. After molecu-
lar recognition, pores are formed in the host cell membrane by the action of lyso-
zymes, followed by the insertion of the viral genetic material, which is responsible
for recruiting all the bacterial metabolic machinery for the production of nucleic
acid and structural proteins that form virions [51]. The last step of this cycle occurs
by inducing transcription and translation of the synthesis of enzymes, such as
lysines, holins, or mureins, which are responsible for lysing the bacterial cell and
consecutively releasing the virions so that the whole process can start again [51].
For the lysogenic cycle, the steps of adsorption on the cell surface and insertion
of genetic material occur in the same way described above. However, the lysogenic
cycle differs from the lytic cycle by aggregating the virus’s genetic material to the
host bacterium’s genetic material or remaining as a plasmid [69, 70]. At this life
cycle stage, the phage is called a prophage, becomes latent, and can have its genetic
material replicated when the bacterial cell multiplies [69–71]. However, there may
be the occurrence of physical stimuli, such as temperature changes, or chemical
ones, such as communication between bacteria through the quorum sensing system.
4 Chemical-Biology and Metabolomics Studies in Phage-Host Interactions 75

These stimuli promote the transition from the lysogenic to the lytic cycle, returning
the prophage to the steps of protein synthesis, virion formation, and cell lysis to
release viral particles [72].
The lysogenic cycle is primarily responsible for the genetic variability of phages
and bacteria from the parallel occurrence of horizontal gene transfer during infec-
tion [15, 69]. Thus, this cycle ends up being responsible, for example, for affecting
bacterial growth [73, 74], favoring the excretion of virulence factors [75, 76], or
expressing genes that are not expressed in the usual course of infection, a phenom-
enon called a lysogenic conversion [70]. For this reason, when prospecting for
phages with therapeutic potential for the treatment of diseases, it is necessary to
previously investigate the life cycle of the phage under study, with priority given to
lytic phages. This condition is adhered to because the primary purpose of phage
therapy is to use the high specificity of these infectious agents regarding the recog-
nition of sites present in the bacterial cell wall without affecting human health
[16, 77].
For the last cycle, called pseudolysogenic, the phage remains unstable inside the
bacterial host as an episome (the viral genetic material that remains dispersed in the
cytoplasm without integrating with the bacterial DNA), which does not allow its
replication or establishment of functions such as those of the lytic cycle [78]. This
stage calls the bacteriophage a pre-prophage, occurring in conditions of nutrient
deficit for the host. When the host is in favorable nutritional conditions, it can trig-
ger integration into the genetic material, passing from pre-prophage to prophage,
thus enabling the beginning of a lytic or lysogenic cycle [79, 80]. Figure 4.1 sum-
marizes the presented lifecycles. The chronic infection, generally associated with
filamentous phages, is characterized by the replication and extrusion of viral parti-
cles without bacterial lysis to occur.
Despite the notable importance of bacteriophages, little is understood about their
diversity, their taxonomy being typically based on genome sequence and observa-
tion of phenotypic characteristics [81–83]. Phages classification is challenging
within the concept of taxonomic structuring, given that much of the diversity of
these biological entities is unknown [83, 84]. Even so, metagenomic sequencing has
enormously contributed to obtaining phage data, but taxon limitation and presenta-
tion of phage protein homologs clustered at different taxa cause ambiguity in the
classification process [83].
Most of the bacteriophages already reported with sequenced genomes belong to
the order Caudovirales, composed of three families: Myoviridae, Siphoviridae, and
Podoviridae. By the time this text was written, the order Caudovirales had com-
prised more than 57% (14,251) of the genomes already deposited for viruses in the
NCBI (National Center for Biotechnology Information) database, expressing the
magnitude of phage diversity [85]. The intrinsic characteristics of phages of this
order are the presence of capsids with the absence of lipids formed by subunits of
more than one type of protein containing double-stranded DNA. The capsids have
icosahedral symmetry and a ring at one of the vertices, which may contain different
tails. These tails are responsible for differentiating the families Myoviridae (long
contractile tails), Siphoviridae (non-contractile long tails), and Podoviridae (short
tails), as illustrated in Fig. 4.2 [86].
76 R. Dantas et al.

Fig. 4.1 Life cycles of a bacteriophage. Chronic infection usually caused by filamentous phages
is not considered in this description

Fig. 4.2 Morphological characteristics of viruses from the families Myoviridae (a), Shiphoviridae
(b), and Podoviridae (c). The long and contractile tails (a), long and non-contractile (b), and short
tails (c) are highlighted
4 Chemical-Biology and Metabolomics Studies in Phage-Host Interactions 77

Therefore, knowing the diversity and capacity of virions to infect bacterial cells,
reprogramming the host’s metabolism to benefit its replication for the production of
its progeny is configured as one of the evidence of induction of the production of
compounds in the virus-bacteria interaction [60]. However, this interaction has
many other facets. Next, some of the main mechanisms involved in the phage-­
bacteria interaction are highlighted, emphasizing the compounds produced in this
interaction.

4.2.1 Mechanisms of Attack and Defense

in the Virus-Bacteria System

The evolution of relationships involving parasitism and antagonism within the

phage-bacteria system is constantly transformed, with phages acting to infect the
hosts and the latter acting with defense mechanisms. However, phages co-evolve by
acquiring mechanisms capable of circumventing the host’s strategies and becoming
native parasites [87]. This process is present in the interaction of phages with their
respective pathogenic bacteria, with the cycle of defense and attack constantly
evolving. Some mechanisms and compounds that are present in these strategies
within the phage-bacteria system are discussed below, with the main objective of
clarifying the role of these compounds produced in the interaction dynamics. Some
mechanisms and compounds present in the attack and defense strategies within the
phage-bacteria system are discussed below. The primary objective is to clarify these
compounds’ role in the interaction dynamics.

4.2.1.1 Bacterial Defense Strategies Mediated by Macromolecules

Phage adsorption is primarily mediated by receptors on the bacterial cell surface,

usually peptides or polysaccharides. Within this process, some bacteria have molec-
ular defense strategies against invading phages [88]. Based on this possibility, there
are three molecular strategies capable of controlling the host’s infection: blocking
cell surface receptors, producing extracellular compounds, and producing com-
pounds that inhibit parasite action [39].
As for the strategy involving receptor blockade, researchers have sought to
understand the effect of proteins produced by the bacterial host in decreasing phage
adsorption in the infection process. A successful example of this strategy is the pro-
duction of protein A (OmpA) by Staphylococcus aureus, a human pathogen with a
high priority level, according to WHO, capable of triggering a generalized infection
(sepsis). When produced by S. aureus, OmpA is responsible for masking phage
adsorption to the specific binding site, thus preventing the insertion of viral DNA
[89]. The OmpA protein is a polypeptide formed by 172 amino acid residues con-
taining a barrel β domain in the bacterium’s outer membrane.
78 R. Dantas et al.

Another exciting example involving this defense strategy is the lipoprotein (Llp)
expressed by E. coli, which is responsible for blocking the membrane protein FhuA,
a specific receptor for the T5 phage. The bacterium can only produce Llp when the
T5 phage infects it. That is, the phage is responsible for inducing the production of
the lipoprotein that intercepts the same receptor that the phage uses to infect the
bacterium [90].
As for the mechanism involving the production of extracellular compounds, bac-
teria can use the ability to produce biopolymers as a physical barrier against phages.
Exopolysaccharides (EPS) are a class of compounds that act as a physical barrier to
protect the bacteria but can also be degraded by specific phages that bypass the host
defense system. As an example of this mechanism, researchers observed a phage
resistance mediated by alginate used to encapsulate the bacteria Azotobacter spp.,
which indicates that strains of Azotobacter spp., which are already known to pro-
duce alginate, may be resistant to phage infection [91, 92].
Regarding the external membrane modification, E.coli strains can be repro-
grammed by a prophage, inducing the activation of an O-acetyltransferase enzyme
in the modification of the O antigen of the lipopolysaccharide (LPS) present in the
external membrane of the bacterium. This mechanism means that other phages with
specificity for this antigen end up not adhering to the bacterial surface [93].
The phage-bacteria interaction can, for many times, present a mutualistic charac-
ter. An example of this process is the action of prophages that direct the use of
proteins present in the bacterial membrane to act in blocking the entry of DNA from
other phages by inhibiting the action of the channel through which the viral DNA
crosses the membrane [39, 94]. These proteins can also act by inhibiting the action
of other enzymes, such as lysozyme, present in the tail fibers of other phages capa-
ble of degrading the peptidoglycan layer [39]. This phenomenon is known as super-
infection exclusion.
On the other hand, E. coli strains may undergo mutations in genes responsible
for the biosynthesis of lipopolysaccharides present in the outer membrane, aiming
at resistance against U136B phage infection [95]. Within this same strand, strains of
Listeria monocytogenes show resistance to phages from mutations that cause loss or
deficiency of rhamnosylation of teichoic acid present in the cell wall [96].
Finally, the mechanism involving compounds that compete for phage receptors
may include substances that are present inside bacteria, such as the 21-amino acid
residue peptide with antimicrobial activity, Microcin J25 (Fig. 4.3), produced
mainly by E. coli [97]. Microcins are peptides known for forming a group of 15
substances with low molecular mass (<10 kDa), showing high stability to tempera-
ture and pH changes [98]. Microcin J25 is a type II bacteriocin that presents high
antimicrobial activity, being designated by the producing strains as agents that help
in microbial competition [97]. When produced by strains infected by phages, they
start competing for the FhuA receptor, considering that this is an iron transport
domain and used as a receptor for other phages, such as the T5 phage [99, 100].
In addition to the mechanisms mentioned earlier, there are other mechanisms
involving the action of specific proteins in protecting bacteria against phage DNA
insertion, such as restriction-modification mechanisms [102–104]. Other
4 Chemical-Biology and Metabolomics Studies in Phage-Host Interactions 79

Fig. 4.3 Molecular structure of microcin J25 (a). Secondary structure of the microcin peptide J25
(b) interacting with the E. coli membrane protein FhuA (c) (PDB ID: 4CU4) [101]

mechanisms in which the production of new compounds does not occur are the
activation of the immune system (CRISPR-Cas), which involves the neutralization
of exogenous DNA, and the abortive infection system, which aims to offer resis-
tance to the phage followed by cell death [105–109]. As they are outside the scope
of the text, these mechanisms will not be addressed.

4.2.1.2 Protein-Based Phage Parasitism Strategy

As an adaptive response to bacterial resistance mechanisms, some phages have

enzymes responsible for degrading exopolysaccharides present in the bacterial cell
wall as a strategy to overcome the host’s defense system. For example, there is a
study involving the degradation of alginates produced by strains of Pseudomonas
spp. from alginate lyase enzymes encoded by the F116 phage [110].
Enzymes that degrade polysaccharides can be classified as hydrolases and lyases.
For the lyases, these enzyme action sites are found in the bond between the mono-
saccharide and carbon 4 (C4) of uronic acid, introducing a double bond between C4
and C5 of uronic acid [111]. Hydrolases, however, break the glycosyl-oxygen bond
present in the glycosidic bond [112]. Both enzymes are located in the capsid struc-
ture, which interacts with polysaccharides of the bacterial membrane, or in a dis-
solved form present in the lytic cycle.
Similarly, the phages manage to circumvent the masking of the binding sites
present on the surface of the bacteria when there is the formation of a capsule (cohe-
sive layer of biopolymers of the class of polysaccharides covalently linked to the
cell wall) protecting the pathogen [113, 114]. This process was described when
observing that E. coli K1F phages could express the endosialidase enzyme, which
is capable of recognizing and degrading the E. coli capsule. This enzyme is present
in a gene that encodes the proteins present in the tail fibers of the T7 phage, a phage
that is similar to K1F [115].
80 R. Dantas et al.

Still considering the possibility of expressing enzymes to access the interior of

bacterial pathogens, Cornelissen et al. [116] identified the degradation action of
exopolysaccharides by an enzyme present in the tail fibers of the AF phage of
Pseudomonas putida. When investigated for efficiency against the reduction of bio-
films produced by different bacteria, phages also showed mechanisms similar to
those already reported involving the action of enzymes since biofilm is a matrix also
composed of exopolysaccharides, proteins, nucleic acids, and lipids [50, 117–121].
The importance of enzymes in the process of infection by bacteriophages is
remarkable. For this reason, the enzymatic richness present in the genome of differ-
ent phages is the focus of several studies within phage therapy since these enzymes
can be isolated and purified for subsequent use in in vitro or in vivo tests [51,
122, 123].

4.2.2 Phage Proteins in the Control of Bacterial Pathogens

Due to the functionality and specificity of phage proteins, efforts have been made
for the development of antimicrobial agents from phage enzymes. Of all the enzymes
encoded by phages, depolymerases stand out, which are responsible for helping the
viral particle to adsorb and invade the host, and holins, which are responsible for
lysing the plasma membrane and actively participating at the end of the lytic pro-
cess for the release of virions [124, 125]. In addition to these two groups, peptido-
glycan hydrolases (PGH) are considered essential and are divided into endolysins
and virion-associated peptidoglycan hydrolase (VAPGH).
Endolysins stand out because they are responsible for the internal lysis of the
peptidoglycan layer present in the cell wall of bacteria, allowing the release of viral
particles. VAPGH, on the other hand, differs from endolysins because they are
responsible for the peptidoglycan layer degradation at the time of adsorption of the
viral particle to the bacterial surface [122]. Endolysins receive pronounced promi-
nence because they do not present toxicity or induction of bacterial resistance and
present a broad spectrum of action with high efficiency. They are extremely capable
of providing cell death just seconds after their application [51, 126, 127].
Since the first report on the purification and application of endolysins in bacterial
control, several studies have focused on verifying their efficiency both in vitro and
in vivo. An example of the successful use of endolysins is the in vitro and in vivo
control (in mice) of 10 strains of Streptococcus spp. Nelson and co-workers [128]
evaluated the efficiency of action and specificity of a murein peptidoglycan hydro-
lase derived from a C1 streptococcal phage. They verified, in vitro, the total steril-
ization of a culture of approximately 107 Streptococcus in just 5 s. For in vivo
experiments, the absence of Streptococcus was verified in mice after 2 h of treat-
ment, concluding the verification of the potential use of this protein in the control of
infectious agents responsible for respiratory syndrome caused by Streptococcus spp.
The use of endolysins as a strategy to control multidrug-resistant strains of clini-
cal interest, such as Pseudomonas aeruginosa bacteria, has been reported with great
4 Chemical-Biology and Metabolomics Studies in Phage-Host Interactions 81

success, allowing us to envision great possibilities for use as therapy, either alone or
in combination with antibiotics [129, 130].
Despite the great importance of phages, as described here from several studies
involving different mechanisms of effective control of bacterial pathogens, their
approval for use as therapeutic agents is only allowed in countries such as Russia,
Georgia, and Poland [131, 132]. For enzymes, despite having great notoriety and
proven efficiency in several studies as specific compounds for bacterial control, so
far, endolysins are not approved for therapeutic use in humans [122, 133].

4.2.3 Toxins and Other Metabolites Associated

with Virus-Bacteria Interaction

The production of compounds in the virus-bacteria interaction is strongly linked to

phages with a lysogenic life cycle, once, the lysogenic cycle favors the phage’s abil-
ity to integrate its genetic material into the host’s genetic material and generate
genetic diversity from each cycle. This genetic diversity promotes the transfer of
genes from bacteriophages to bacteria, thus inducing the production of compounds
toxic to bacterial hosts [71].
The discovery that the phage-bacteria interaction could favor the production of
new substances capable of making the bacteria-host interaction even more critical
sparked an alert for understanding the chemical nature of these toxins. Compounds
produced by pathogens such as Corynebacterium diphtheriae, diphtheria-causing
bacterium [134]; Clostridium botulinum, the causative agent of botulism [135];
Vibrio cholerae, causal agent of cholera [75]; and Escherichia coli, which cause
profound watery diarrhea [136], are examples of chemical response encoded by
phages in the interaction with their hosts.
The discovery of the presence of phages as responsible for the transfer of genes
that encode toxins occurred shortly after the first outbreak of human infection with
E. coli, in 1983, as a result of the action of the Shiga toxin. Toxin production begins
with integrating the phage into the bacterial genome, followed by the induction of
the passage from the lysogenic to the lytic cycle, where the phage DNA is detached
from the bacterial genome. After these processes, the genes encoding the toxin are
replicated by several bacteria, where it is finally expressed. When cell lysis occurs,
the toxins are released into the extracellular environment. This toxin acts as a viru-
lence factor. It can be used to compete for space and nutrients within the bacterial
host, such as the bacteria present in the stomach [137].
All the toxins described here resulted from a gene transfer, which indicates that
although phages allow responses aimed at bacterial control, natural events also
occur where the host undergoes a genetic modification capable of transforming it
into a pathogenic strain, often lethal. This point can be seen as a plausible reason for
more significant investments in the knowledge of the genome of other parasitic
phages of important bacteria and in the studies involving phage-bacteria-human
host relationship.
82 R. Dantas et al.

H
N
O N
O O
H
O N N
N N N
H O
S S O NH
O

Colibactin

Fig. 4.4 Molecular structure of the natural product Colibactin produced by E. coli

Mechanisms involving bacterial natural products were recently reported, in

which it was possible to identify E. coli producing the toxin colibactin (Fig. 4.4) as
responsible for the induction of the lytic cycle of phages that presented lysogeny in
strains of E. coli not producing colibactin [138]. This secondary metabolite is a
reactive small-molecule classified as genotoxin with a reactive cyclopropane war-
head that interacts with DNA. The gene cluster responsible for its biosynthesis has
a hybrid nonribosomal peptide synthetase-polyketide synthase (NRPS-PKS)
domain, known as the pks island found in human-associated E. coli and
Enterobacteriaceae gut bacteria strains.
Understanding that prophages can be activated for the lytic cycle from the inter-
action with natural products present in the intestinal microbiota reveals the concern
in studies focused on phage therapy, emphasizing more diligently the need to under-
stand the life cycle of these biological agents. Another important factor presented in
this study [138] is the possibility of this natural product inducing the production of
Shiga toxin, previously discussed. Therefore, understanding the molecular mecha-
nisms involved in these interactions favors a better understanding of the relation-
ships within the human microbiome.
A second strategy involving the release of bacterial toxins into the medium, aim-
ing to favor competition for space and nutrients without the mediation of bacterial
genes, has been reported for strains of Salmonella enterica. The release of the toxin
colicin B was reported to be a product of the activation of the lytic pathway of pro-
phages present in bacterial strains. This mechanism answered a persistent question
about the regulation of colicin since its expression is not dependent on the expres-
sion of another gene in the bacterium [139]. Colicin B is a high molecular weight
(55 kDa) cytotoxic protein that recognizes E. coli membrane transporters as recep-
tors, causing the formation of pores that destabilize the cytoplasmic membrane,
leading to cell death [140].
Finally, considering the progress in understanding the mechanisms present in the
phage-bacteria interaction, Kronheim and co-workers [141] advanced in the search
for answers involving the production of secondary metabolites by bacteria against
their phage hosts. The authors reported host protective function mediated by DNA
intercalating compounds. Thus, four compounds (Fig. 4.5) previously reported in
Streptomyces species were identified as being responsible for inhibiting phages of
Streptomyces coelicolor, E. coli, and Pseudomonas aeruginosa. The authors report
4 Chemical-Biology and Metabolomics Studies in Phage-Host Interactions 83

NH2 NH2 NH2 NH2

OH OH OH O
H
O OH O O O O OH O O O O OH O O O O OH O O

OH OH OH O
O OH O O OH OH OH H
O O OH O O OH O
Idarubicin Daunorubicin Epirubicin Doxorubicin

Fig. 4.5 Molecular structures of the compounds idarubicin, daunorubicin, epirubicin and
doxorubicin

that the production of these compounds may be associated with an innate defense
system that aims to prevent the replication and transcription process from the inter-
calation of phage DNA [141].
A deeper study of the metabolic response arising from the phage-bacteria inter-
action can also be obtained from omics strategies, such as metabolomics. A vision
inside the metabolomics pipeline applied in phage-bacteria interaction study has
been used, and excellent results are emerging in this field. Many works have been
reported in the last years involving the study of the metabolomics of phage infec-
tion. Consequently, more key molecules have been associated with essential pro-
cesses in this interaction, helping to develop new strategies for treating bacterial
infections.

4.3 Metabolomics and Phage-Bacteria Interaction

4.3.1 Workflow Metabolomics in Phage-Host Interaction

Metabolomics consists of studying the metabolites alteration of different chemical

classes belonging to a biological system (cell, tissue, or organism) involved in a
given metabolic process, such as an alteration caused by a physical or biological
factor [142, 143]. The study can focus on a quantitative or semi-quantitative analy-
sis of a specific substance or a group of metabolites of interest, denominated as
target metabolomics [143, 144]. On the other hand, a qualitative or semi-­quantitative
analysis can also be carried out, considering the largest possible number of metabo-
lites obtained after establishing a specific condition of the biological system,
denominated as untargeted metabolomics [143, 145]. Evaluating the complete met-
abolic profile (metabolome) before and after setting a particular situation provides
information capable of relating genetic, environmental, and ecological factors to the
genotype and phenotype of the system under study [146, 147].
The use of metabolomic strategies permeates the phage-bacteria interaction to
understand how the phage, when it infects the bacteria, can modify the host’s
metabolism to induce the expression of genes responsible for the biosynthesis of the
metabolic arsenal involved in the assembly of new virions. The metabolomics work-
flow applied to biological systems also applies to this field of study. It consists of
84 R. Dantas et al.

establishing a biological question, experimental design with treated and control

groups, extraction of the metabolites of interest, sample preparation, use of high-­
performance analytical techniques to analyze metabolites, processing, and analysis
of data from univariate and multivariate strategies followed by biological interpreta-
tion of results [148, 149]. This pipeline favors verifying changes in the levels of
specialized metabolites involved in specific metabolic routes associated with the
infection process and the permanence of phages in the bacterial host.
Extraction of the metabolites of interest, followed by adequate sample prepara-
tion, are crucial steps before analysis and are directly related to the chosen experi-
mental design based on the objective established by the initial biological question.
The few existing articles with different approaches applied to study the phage-­
bacteria interaction direct their extraction protocols according to the biological sys-
tem of interest. Biological systems vary and may contain only bacteria infected with
phage at different incubation times [150–152], or, for example, a complex biologi-
cal system comprising an animal infected with a bacterium or set of bacteria and a
phage or group of phages [41, 153].
The variation of the experimental design also brings variations to the extraction
process, which can be intracellular, extracellular, or global. For protocols applied in
metabolomic studies involving intracellular metabolites from in vitro phage-­bacteria
systems, the extraction procedure comprises different steps. Initially, the system
under study (cells in culture medium) is subjected to cold centrifugation (4 °C) to
obtain the cellular content as a pellet. Subsequently, the cellular content is rinsed
with a PBS solution, ammonium carbonate, saline solution (0.9%), or similar strate-
gies to remove any remnants of the culture medium, maintaining a constant pH and
adjusted osmolarity. Then the washed cells are used to obtain the intracellular meta-
bolic content. Cell lysis can be induced by successive freezing and thawing, ultra-
sound after suspending the biomass in ultrapure water, or by incubating the biomass
in a 60% ethanol solution followed by brief heating at a controlled temperature. A
mixture of organic solvents (chloroform/methanol/water, 1:3:1, v/v/v) can be added
in the previous step for extracting the metabolites [154] or only ultrapure water can
be kept as the main solvent [150]. The metabolites of interest are then separated
from the higher molecular weight lysed cell contents by centrifugation or filtration.
The storage of samples containing metabolites free of any solvent must always be
cold (−80 °C) to reduce eventual degradation.
Studies involving an extracellular analysis target the phage-bacteria culture fil-
trate and the axenic culture of the bacteria, being obtained from filtration with a
porosity filter of 0.22 μm for cell retention [151]. On the other hand, when the inter-
est involves extracellular and intracellular metabolites of more complex systems,
such as an in vivo system using mice treated with a specific bacterium associated
with its respective bacteriophage, the extraction protocol focuses mainly on what is
excreted from the system as a whole. Thus, fecal samples can be used to study both
the phage-bacteria interaction and to verify the influence of this interaction on the
intestinal metabolome of the host [153]. In this scenario, the metabolites can be
directly extracted by the action of organic solvents such as methanol. The solid resi-
dues from the extraction can be separated by centrifugation, the supernatant
4 Chemical-Biology and Metabolomics Studies in Phage-Host Interactions 85

Fig. 4.6 Different sample extraction methods in vivo (a) and in vitro (b) systems for host-phage
metabolomics

containing the metabolites of interest can be dried, and the final mass can be stored
in an ultra-freezer or under nitrogen gas. Figure 4.6 outlines a workflow for sample
extraction in metabolomics of a phage-host system based on previously published
articles.
After obtaining the samples, several analytical techniques can be used to investi-
gate metabolic alterations in phage-bacteria systems. The most common techniques
used are Nuclear Magnetic Resonance (NMR) and Mass Spectrometry (MS) with-
out coupling with a separation system (flow-injection) or coupled to liquid chroma-
tography (LC-MS).
The nuclear magnetic resonance technique pioneered the techniques used in
studies of metabolome alteration in phage-host interactions. Within untargeted stud-
ies, the relationship between signals in the NMR spectrum and the structural nature
of the compounds present can be established, making it possible to associate chemi-
cal classes and unambiguous structures to predominant compounds in a complex
matrix such as the phage-bacteria system. In 2012, K. Sonkar, R. Purusottam, and
N. Sinha [155] used a 1H NMR-based metabolomics dataset of growth media in
planktonic cultures of Pseudomonas aeruginosa infected and uninfected with bac-
teriophage pf1 to determine the transient flux of associated metabolites to metabolic
disturbances resulting from phage infection. The authors verified a significant dif-
ference in the metabolic profile of the group of samples with infected bacteria com-
pared to non-infected bacteria by constructing a statistical model based on the
values of 1H signals. Regarding changes in metabolite levels, the authors
86 R. Dantas et al.

qualitatively reported a positive alteration in agmatine biosynthesis after phage

infection, which may trigger the inhibition of the biofilm formed by these bacteria
from the cytosolic metabolism of this metabolite. Furthermore, the authors also
observed pyruvate oxidation to lactate followed by further oxidation to acetate.
Other authors also report the efficient use of the NMR technique for differentiat-
ing the metabolic profile [150, 151], and the successful use of the liquid chromatog-
raphy technique coupled with mass spectrometry (LC-MS) [156]. Different
analytical configurations have been applied in studies of phage-host interaction by
LC-MS, ranging from not using a chromatographic system before analysis by mass
spectrometry, characterized as flow-injection (when using only the chromatograph
piping system, without performing chromatographic separation) [41, 152], such as
the use of high [154] and ultra-high performance chromatographic systems [153].
The configuration of the mass analyzers also varies within the study of the phage-­
bacteria system, going through analyses with time-of-flight (TOF) analyzers [41,
152], and analyses with Orbitrap, which has a high-resolution power, sensitivity and
selectivity [153, 154, 157, 158]. Still, within what involves the analysis system, dif-
ferent studies have reported the predominant use of electrospray ionization (ESI)
sources. The principal benefit of ESI sources consists in the ability to ionize small
molecules of polar and medium polarity, being a great strategy to ionize part of
molecules that participate in primary metabolism (known as building blocks) and
most molecules involved in secondary bacterial metabolism.
After the analysis, a large volume of data is generated from thousands of metabo-
lites extracted from the biological system under study. The designation of which
metabolites may be associated with an important alteration within the phage-host
system can be performed by processing and applying univariate and multivariate
data analysis methods [159]. Univariate methods (e.g., ANOVA, t-test, Volcano-­
plot) can reveal significant differences when comparing the levels of a given metab-
olite of interest in different studied groups [159, 160].
Unlike univariate analyses, multivariate analyses consider the interactions and
correlations of the metabolites studied in the different samples. Thus, when the
objective of the study involves a global approach, using unsupervised multivariate
methods is generally employed, as an example of Principal Components Analysis
(PCA). From the analysis by PCA, it is possible to verify a trend of the differential
grouping of the samples between treated and control groups from a resizing of the
data in a plane of components that explain the degree of variance of the data and that
differentially accommodate the variables that are involved in the differentiation of
the studied groups. In addition to separating groups based on variables, PCA analy-
sis also points out the existence of outliers and the quality and reproducibility of the
analysis method used. This last characteristic can be obtained through the central
positioning in the axes of the main components of a sample containing a pool of all
samples, called quality control (QC) [161, 162].
Supervised multivariate analysis, such as Partial Least Squares Discriminant
Analysis (PLS-DA), can associate additional information to the original data set,
making it a training system for obtaining a model capable of predicting a certain
related characteristic of the metabolic alteration of a system under known and
4 Chemical-Biology and Metabolomics Studies in Phage-Host Interactions 87

pre-­established conditions. This analysis provides a ranking of variables (analyzed

metabolites) as a function of their quantitative estimation and discriminatory power
within the data set. These variables are known as variable importance in projection
(VIP), where the variables with values of VIP > 1 are commonly significantly rele-
vant in separating the analyzed sample groups [162, 163].
After the statistical analysis and verification of the differential variables between
the treated and control groups, it is necessary to associate these variables with the
already-known metabolism of the system under study. Thus, some platforms can be
used both to integrate the data obtained with other omics (e.g., Kyoto Encyclopedia
of Genes and Genomes, KEGG, www.genome.jp/kegg/) and to carry out a putative
identification of the metabolites associated with the observed alteration from mass
spectrometry analyzes (e.g., Global Network Products Social Molecular
Networking – GNPS [164], SIRIUS 4 [165], MassBank [166] and MassBank of
North America (MoNa) [167]). For NMR analyses, the annotation of metabolites
can occur through databases such as Chenomx Compound Library [168]. Figure 4.7
outlines a general workflow in metabolomics of a phage-host system based on pre-
viously published articles.
Currently, the workflow in metabolomics studies has a set of recommendations
and guidelines previously established by the Metabolomics Standard Initiative
(MSI) to propose a standardization in the way of reporting the data obtained in the
different stages, such as sample preparation and analysis, pre-processing and data
processing, verification of data quality and annotation of metabolites at different

Fig. 4.7 Host-phage metabolomics workflow

88 R. Dantas et al.

levels based on the analytical technique used and information obtained from data-
bases [148, 169]. Below are reported studies involving metabolomics as a strategy
to investigate the role of phage in the modulation of host metabolism.

4.3.2 Metabolomics Studies in Phage-Host Interaction

The literature has few studies dedicated to understanding the modulation of the
metabolism of a bacterial host by phage infection [170]. The impact of phages on
essential pathways for maintaining gene transcription and translation up to host cel-
lular respiration can provide a clear view of the system’s phenotype in a specific
period of the infected host’s life cycle. In other words, verifying phenotype metabo-
lism level reveals central conditions related to the functions of genes encoded by
phages, called auxiliary metabolic genes (AMGs), leading us to the discover of new
paths capable of interrupting crucial steps in maintaining bacterial cell viability
[41, 150].
Few studies have deepened the questioning of different systems composed of
phages and hosts that are part of the class of human pathogens associated with the
incidence of antimicrobial resistance. As an example of research involving the
phage-Pseudomonas aeruginosa PAO1 interaction, De Smet et al. [41] questioned
the existence or not of a universal metabolic variation of this host against six lytic
phages (LUZ19, LUZ24, PEV2, 14-1, YuA and phiKZ). The authors evaluated the
impact of phage infection from an untargeted metabolomics approach based on ana-
lyzing intracellular metabolites by flow-injection mass spectrometry in negative
mode. In general, phage infection favored significant changes in the metabolites
analyzed during the course of infection compared to systems in which the bacteria
have not been infected. The main evidence of this study is that metabolomics can
provide insights concerning changes in phage-bacteria metabolism. The authors
verified changes in the metabolism of carbohydrates, nucleotides, nucleotide sug-
ars, amino acids, purines, glutathiones, pantothenate, and coenzyme A (CoA). As
discussed in the study, the pathways involving these metabolites are directly associ-
ated with the replication and biosynthesis process of macromolecules associated
with the assembly of new phages. The study also emphasized the alteration observed
in the metabolism of nucleotide sugars, associating the increased levels of these
metabolites with the presence of phage proteins responsible for inhibiting the pro-
cess of cell division and cell wall biosynthesis. Even given the important responses
obtained regarding the variation in the metabolomic profile after infection of PO1A
by different phages, the authors reported the visualization of only 2.5% of the
changes observed within the 518 intracellular metabolites identified, concluding
that there is not a universal response considering the infection by different phages.
Other authors have also focused on the power of metabolomics in providing
evidence at the molecular level of the ability of phages to hijack the host machinery
for its replication. In this regard, Chevallereau et al. [152] presented the application
of untargeted metabolomics based on mass spectrometry to analyze the
4 Chemical-Biology and Metabolomics Studies in Phage-Host Interactions 89

modification of intracellular metabolism associated with the infection of

Pseudomonas aeruginosa by the bacteriophage PAK_P3. The authors reported that
the infection favors the interference of specific host metabolic pathways, such as
pyrimidine metabolism, which was directly associated with increased viral RNAs
throughout the infection to maintain phage replication. Based on the metabolomic
strategy applied, the authors were able to understand the flow of metabolites at dif-
ferent times of infection, attributing, for example, the decrease in metabolites from
the amino acid class to the need for building blocks for the formation of new viral
particles. From this study, it was possible to understand how the host’s metabolism
is redirected to maintain the metabolism and replication of the phage.
On the other hand, in the search for specific antibacterial targets from the infec-
tion of bacteria by phages, Zhao et al. [150] investigated the metabolic alteration of
infected and uninfected P. aeruginosa by a lytic phage (PaP1) using a 1H NMR-­
based metabolomics approach. In this study, 48 metabolites of different classes
were identified, such as amino acids and derivatives (21), organic acids (10), nucleic
acid components (7), alcohols (3), sugar (1), and six other metabolites. VIP ana-
lyzes were performed, and metabolites such as thymidine and choline had their
levels increased intracellularly; an inverse trend was observed for betaine. By inte-
grating these data with transcriptomics data, the authors could faithfully correlate
the difference in metabolite levels in control and treated groups with the observed
differential gene expression. The integration of the results allowed the authors to
infer that the PaP1 phage can contribute to metabolic alterations in the choline-­
glycine betaine pathway. Studies like this one, involving an integrated approach of
metabolomics data with other omics, can lead us precisely to possible therapeutic
targets.
Metabolomics has also been used jointly and efficiently in studies to find alterna-
tives to control species of bacteria resistant to antimicrobial agents. The first study
involving the integration of metabolomics and pharmacology of biological systems
in the investigation of the antibacterial action of phages associated with antibiotics
was reported by Han et al. (2022) [154]. In this work, the authors evaluated the
modification of the metabolism of Klebsiella pneumoniae (Kp II-503) resistant to
polymyxin B under a new phage (pk8) action, under the action of just polymyxin B
and under the action of a phage-antibiotic combination. The authors were the first to
use an extraction method based on a mixture of organic solvents of different
polarities, chloroform/methanol/water (1:3:1, v/v), plus a mixture of internal: stan-
dards 3[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate (CHAPS),
N-cyclohexyl-3-aminopropanesulfonic acid (CAPS), piperazine-N,N′-bis(2-­
ethanesulfonic acid) (PIPES), and tris(hydroxymethyl)aminomethane (TRIS). An
untargeted metabolomic approach focused on intracellular metabolites was per-
formed, which were analyzed by a high-performance liquid chromatography system
in tandem with a high-resolution mass spectrometer. From the multivariate analysis
of the data through PCA analysis, the authors were able to visualize a differentiation
in the metabolic profile of the system containing the bacteria infected with the pK8
phage with and without polymyxin when compared with the profile of the bacteria
treated and not treated only with polymyxin. The study reported differentiation of
90 R. Dantas et al.

the metabolic profile also at different times of treatment (4 and 24 h), which, with
the aid of ANOVA analyzes and analysis of pathways by KEGG [171, 172] and
EcoCyc [173], it was possible to point out which metabolites were involved in the
significant inhibition of the pathways of pentoses phosphates and the citrate cycle,
as well as a disturbance in the metabolism of amino acids, nucleotides, and lipids.
The observed impact on the levels of substances involved in central carbon metabo-
lism was in line with the observation of the inhibition of bacterial regrowth after
24 h of the combination of the bacteriophage with the antibiotic. The study repre-
sents a significant initial impact on developing strategies to control pathogens that
are resistant to antibiotics, proving the existing potential and the possibility of inno-
vation in using phages as therapeutic agents.
Studies involving the verification of the metabolic alteration of the bacterial host
when in the presence of phage-derived antibacterial inhibitors were also developed
from a metabolomics approach based on 1H NMR [151]. In this study, the authors
analyzed the metabolic profile of Pseudomonas aeruginosa in the presence of the
phage protein gp70.1 encoded by the PaP3 phage, together with microarray-based
transcriptomic analysis and RT-qPCR. After verifying the partial inhibition of the
growth of P. aeruginosa containing the plasmid that encodes the gp70.1 protein, the
authors were able to visualize the decrease in the consumption of amino acids, cor-
roborating the observed inhibitory action. PCA analysis indicated a difference
between the metabolism of P. aeruginosa containing an overexpressed gp70.1 con-
tent and the metabolism of control P. aeruginosa. Thus, from the VIP analyses, the
authors observed 15 significantly different metabolites between the groups, namely
12 amino acids, two organic acids, and one sugar. From a quantitative estimate by
VIP analysis, the authors inferred that the highest concentration observed for 9 of
the 12 amino acids was directly associated with the effect of the gp70.1 inhibitor,
which reduced the consumption of amino acids by the bacteria. Studies like these
may provide evidence of which mechanisms are involved in host metabolism in the
presence of inhibitors of phage origin, expanding the horizon of studies within the
field of phage therapy.
More complex strategies involving phage interaction with intestinal bacteria in
an in vivo system have also been addressed through metabolomics. Hsu et al. (2019)
[153] demonstrated the effects on the intestinal metabolome caused by a bacterial
modulation induced by lytic phages (B40-8, F1, T4, VD13) and identified the pres-
ence of phage-modulated metabolites capable of affecting the mammalian host. The
study was guided by an untargeted metabolomics strategy based on analyzing the
fecal metabolome using high-resolution mass spectrometry coupled with ultra-­
high-­performance liquid chromatography (UPLC). Volcano plot analyses associ-
ated with Hierarchical clustering of significantly changing metabolites were used to
verify the differential metabolites at different stages of colonization in mice con-
taining phages and their respective hosts. A total of 860 metabolites were measured
and associated with KEGG pathways. Metabolites such as tyramine and tryptamine
were downregulated and were associated with the possibility of inducing ileum con-
tractions and increased gastric mobility of the host, respectively. This study sheds
light on the care that must be taken in initiatives to use phages for therapeutic
4 Chemical-Biology and Metabolomics Studies in Phage-Host Interactions 91

purposes, emphasizing the importance of knowing and understanding the product of

metabolic pathways arising from the interaction of phages and bacteria in a complex
microbiome such as that of a mammalian host.
Other studies on bacteria that are not considered a direct risk to human and ani-
mal health have also been reported. As an example of phytopathogenic bacteria that
cause disease in several economically important crops, Papaianni and co-workers
(2020) demonstrated the efficient use of the pipeline in metabolomics based on 1H
NMR (1D, 2D, and TOCSY) to demonstrate that the use of hydroxyapatite (HA),
eicosanoic acid (C20:0) and the phage Xccφ1 can destroy the biofilm produced by
Xanthomonas campestris pv. campestres, identifying specific biomarkers involved
in biofilm control [174].

4.4 Final Considerations and Perspectives

The success of the phage-host interaction through the action of particular com-
pounds and mechanisms opens doors to the molecular understanding of important
pathogenic processes in current and future society. Understanding the mechanisms
enables us to propose models and strategies to control possible natural processes
against human, animal, and plant life. The diversity of phages, linked to the tremen-
dous biochemical plasticity of these biological agents, provides us with an arsenal
that, when not exploited in the best way, can trigger adverse biological events.
Given all the possibilities studied so far, it is necessary to understand the mecha-
nisms that involve the phage-host interaction and to look for molecular strategies
involved in key processes for controlling bacterial pathogens. Then, from the suc-
cess of a parasite interaction, the newest discovery of crucial mechanisms capable
of describing and demystifying different paths within the strategies of circumvent-
ing the resistant behavior of pathogens of interest may emerge.
The ability to generate chemical diversity from a phage-bacteria interaction was
presented by George Smith in his 2018 Nobel Prize laureate work. The phage dis-
play strategy is a highly sophisticated technique and further substantiates what is
understood today as applied biotechnology. The phage-bacteria interaction within
the idea of phage display puts molecular ‘possibilities’ much closer to reality and
with a more immediate response factor, considering that it is a strategy that involves
high specificity, whether in the development of new drugs, as in the vaccine devel-
opment. Therefore, when discussing the production of compounds from virus-­
bacteria interaction within a current scenario, it is already possible to think about
the increase in chemical diversity within the possibilities encoded by the phage
display technique. Therefore, it is possible to provide new perspectives on the world
of new molecules.
92 R. Dantas et al.

References

1. Mohr KI (2016) History of antibiotics research. In: Current topics in microbiology and
immunology. Springer, pp 237–272
2. Collignon P, McEwen S (2019) One health – its importance in helping to better control anti-
microbial resistance. Trop Med Infect Dis 4:22. https://doi.org/10.3390/tropicalmed4010022
3. Klein EY, Van Boeckel TP, Martinez EM et al (2018) Global increase and geographic con-
vergence in antibiotic consumption between 2000 and 2015. Proc Natl Acad Sci U S A
115:E3463–E3470. https://doi.org/10.1073/pnas.1717295115
4. Davies J (2006) Where have all the antibiotics gone? In: Canadian Journal of Infectious
Diseases and Medical Microbiology. Hindawi Limited, pp 287–290
5. World Health Organization (1983) Antimicrobial resistance. Bull World Health Organ
61:383–394
6. Asadi A, Razavi S, Talebi M, Gholami M (2019) A review on anti-adhesion therapies of bac-
terial diseases. Infection 47:13–23. https://doi.org/10.1007/s15010-­018-­1222-­5
7. Peterson E, Kaur P (2018) Antibiotic resistance mechanisms in bacteria: relationships
between resistance determinants of antibiotic producers, environmental bacteria, and clinical
pathogens. Front Microbiol 9:2928. https://doi.org/10.3389/fmicb.2018.02928
8. Davies J, Davies D (2010) Origins and evolution of antibiotic resistance. Microbiol Mol Biol
Rev 74:417–433. https://doi.org/10.1128/MMBR.00016-­10
9. Reygaert WC (2018) An overview of the antimicrobial resistance mechanisms of bacteria.
AIMS Microbiol 4:482–501. https://doi.org/10.3934/microbiol.2018.3.482
10. Tacconelli E, Carrara E, Savoldi A et al (2018) Discovery, research, and development of new
antibiotics: the WHO priority list of antibiotic-resistant bacteria and tuberculosis. Lancet
Infect Dis 18:318–327. https://doi.org/10.1016/S1473-­3099(17)30753-­3
11. Antibiotic / Antimicrobial resistance | CDC. https://www.cdc.gov/DrugResistance/. Accessed
28 Apr 2023
12. Shallcross LJ, Howard SJ, Fowler T, Davies SC (2015) Tackling the threat of antimicrobial
resistance: from policy to sustainable action. Philos Trans R Soc B Biol Sci 370:20140082.
https://doi.org/10.1098/rstb.2014.0082
13. Iskandar K, Murugaiyan J, Hammoudi Halat D et al (2022) Antibiotic discovery and resis-
tance: the chase and the race. Antibiotics 11:182. https://doi.org/10.3390/antibiotics11020182
14. Wittebole X, De Roock S, Opal SM (2014) A historical overview of bacteriophage therapy
as an alternative to antibiotics for the treatment of bacterial pathogens. Virulence 5:226–235.
https://doi.org/10.4161/viru.25991
15. Rolain J-M, Hraiech S, Bregeon F (2015) Bacteriophage-based therapy in cystic fibrosis-­
associated Pseudomonas aeruginosa infections: rationale and current status. Drug Des Devel
Ther 9:3653. https://doi.org/10.2147/DDDT.S53123
16. Dy RL, Rigano LA, Fineran PC (2018) Phage-based biocontrol strategies and their appli-
cation in agriculture and aquaculture. Biochem Soc Trans 46:1605–1613. https://doi.
org/10.1042/BST20180178
17. Vu NT, Oh C-S (2020) Bacteriophage usage for bacterial disease management and diagnosis
in plants. Plant Pathol J 36:204–217. https://doi.org/10.5423/PPJ.RW.04.2020.0074
18. Dewanggana MN, Evangeline C, Ketty MD et al (2022) Isolation, characterization, molecu-
lar analysis and application of bacteriophage DW-EC to control Enterotoxigenic Escherichia
coli on various foods. Sci Rep 12:495. https://doi.org/10.1038/s41598-­021-­04534-­8
19. Kawacka I, Olejnik-Schmidt A, Schmidt M, Sip A (2020) Effectiveness of phage-based
inhibition of listeria monocytogenes in food products and food processing environments.
Microorganisms 8:1764. https://doi.org/10.3390/microorganisms8111764
20. Huang Z, Zhang Z, Tong J et al (2021) Phages and their lysins: toolkits in the battle against
foodborne pathogens in the postantibiotic era. Compr Rev Food Sci Food Saf 20:3319–3343.
https://doi.org/10.1111/1541-­4337.12757
4 Chemical-Biology and Metabolomics Studies in Phage-Host Interactions 93

21. Smith HW, Huggins MB (1982) Successful treatment of experimental Escherichia coli infec-
tions in mice using phage: its general superiority over antibiotics. Microbiology 128:307–318.
https://doi.org/10.1099/00221287-­128-­2-­307
22. Williams Smith H, Huggins MB (1983) Effectiveness of phages in treating experimental
Escherichia coli diarhoea in calves, piglets and lambs. J Gen Microbiol 129:2659–2675.
https://doi.org/10.1099/00221287-­129-­8-­2659
23. Melo LDR, Oliveira H, Pires DP et al (2020) Phage therapy efficacy: a review of the
last 10 years of preclinical studies. Crit Rev Microbiol 46:78–99. https://doi.org/10.108
0/1040841X.2020.1729695
24. Slopek S, Weber-Dabrowska B, Dabrowski M, Kucharewicz-Krukowska A (1987) Results
of bacteriophage treatment of suppurative bacterial infections in the years 1981-1986. Arch
Immunol Ther Exp 35:569–583
25. Cislo M, Dabrowski M, Weber-Dabrowska B, Woyton A (1987) Bacteriophage treatment of
suppurative skin infections. Arch Immunol Ther Exp 35:175–183
26. Wright A, Hawkins CH, Änggård EE, Harper DR (2009) A controlled clinical trial of a thera-
peutic bacteriophage preparation in chronic otitis due to antibiotic-resistant Pseudomonas
aeruginosa; a preliminary report of efficacy. Clin Otolaryngol 34:349–357. https://doi.
org/10.1111/j.1749-­4486.2009.01973.x
27. Fukuda K, Ishida W, Uchiyama J et al (2012) Pseudomonas aeruginosa keratitis in mice:
effects of topical bacteriophage KPP12 administration. PLoS One 7. https://doi.org/10.1371/
JOURNAL.PONE.0047742
28. Basu S, Agarwal M, Kumar Bhartiya S et al (2015) An in vivo wound model utilizing bac-
teriophage therapy of Pseudomonas aeruginosa biofilms. Ostomy Wound Manage 61:16–23
29. Maddocks S, Fabijan AP, Ho J et al (2019) Bacteriophage therapy of ventilator-associated
pneumonia and empyema caused by Pseudomonas aeruginosa. Am J Respir Crit Care Med
200:1179–1181. https://doi.org/10.1164/rccm.201904-­0839LE
30. Soothill JS (1992) Treatment of experimental infections of mice with bacteriophages. J Med
Microbiol 37:258–261. https://doi.org/10.1099/00222615-­37-­4-­258
31. Schooley RT, Biswas B, Gill JJ et al (2017) Development and use of personalized
bacteriophage-based therapeutic cocktails to treat a patient with a disseminated resis-
tant Acinetobacter baumannii infection. Antimicrob Agents Chemother 61. https://doi.
org/10.1128/AAC.00954-­17
32. Knirel YA, Shneider MM, Popova AV et al (2020) Mechanisms of Acinetobacter baumannii
capsular polysaccharide cleavage by phage depolymerases. Biochemist 85:567–574. https://
doi.org/10.1134/S0006297920050053
33. Volozhantsev NV, Shpirt AM, Borzilov AI et al (2020) Characterization and therapeutic
potential of bacteriophage-encoded polysaccharide depolymerases with β galactosidase
activity against Klebsiella pneumoniae K57 capsular type. Antibiotics 9:732. https://doi.
org/10.3390/antibiotics9110732
34. Tanji Y, Shimada T, Fukudomi H et al (2005) Therapeutic use of phage cocktail for control-
ling Escherichia coli O157:H7 in gastrointestinal tract of mice. J Biosci Bioeng 100:280–287.
https://doi.org/10.1263/jbb.100.280
35. Bryan D, El-Shibiny A, Hobbs Z et al (2016) Bacteriophage T4 infection of stationary phase
E. coli: life after log from a phage perspective. Front Microbiol 7. https://doi.org/10.3389/
fmicb.2016.01391
36. Lehman SM, Mearns G, Rankin D et al (2019) Design and preclinical development of a
phage product for the treatment of antibiotic-resistant staphylococcus aureus infections.
Viruses 88(11):88. https://doi.org/10.3390/V11010088
37. Petrovic Fabijan A, Lin RCY, Ho J et al (2020) Safety of bacteriophage therapy in severe
Staphylococcus aureus infection. Nat Microbiol 5:465–472. https://doi.org/10.1038/
s41564-­019-­0634-­z
38. Weinbauer MG (2004) Ecology of prokaryotic viruses. FEMS Microbiol Rev 28:127–181.
https://doi.org/10.1016/j.femsre.2003.08.001
94 R. Dantas et al.

39. Labrie SJ, Samson JE, Moineau S (2010) Bacteriophage resistance mechanisms. Nat Rev
Microbiol 8:317–327. https://doi.org/10.1038/nrmicro2315
40. Hendrix H, Zimmermann-Kogadeeva M, Zimmermann M et al (2022) Metabolic reprogram-
ming of Pseudomonas aeruginosa by phage-based quorum sensing modulation. Cell Rep
38:110372. https://doi.org/10.1016/j.celrep.2022.110372
41. De Smet J, Zimmermann M, Kogadeeva M et al (2016) High coverage metabolomics analysis
reveals phage-specific alterations to Pseudomonas aeruginosa physiology during infection.
ISME J 10:1823–1835. https://doi.org/10.1038/ismej.2016.3
42. Kutter E, Bryan D, Ray G et al (2018) From host to phage metabolism: hot tales of phage T4’s
takeover of E. coli. Viruses 10:387. https://doi.org/10.3390/v10070387
43. Pini A, Giuliani A, Falciani C et al (2005) Antimicrobial activity of novel dendrimeric pep-
tides obtained by phage display selection and rational modification. Antimicrob Agents
Chemother 49:2665–2672. https://doi.org/10.1128/AAC.49.7.2665-­2672.2005
44. Bishop-Hurley SL, Rea PJ, McSweeney CS (2010) Phage-displayed peptides selected for
binding to Campylobacter jejuni are antimicrobial. Protein Eng Des Sel 23:751–757. https://
doi.org/10.1093/protein/gzq050
45. Larpin Y, Oechslin F, Moreillon P et al (2018) In vitro characterization of PlyE146, a
novel phage lysin that targets Gram-negative bacteria. PLoS One 13:e0192507. https://doi.
org/10.1371/journal.pone.0192507
46. Bin Hafeez A, Jiang X, Bergen PJ, Zhu Y (2021) Antimicrobial peptides: an update on clas-
sifications and databases. Int J Mol Sci 22:11691. https://doi.org/10.3390/ijms222111691
47. Al-Wrafy F, Brzozowska E, Górska S et al (2019) Identification and characterization of phage
protein and its activity against two strains of multidrug-resistant Pseudomonas aeruginosa.
Sci Rep 9:13487. https://doi.org/10.1038/s41598-­019-­50030-­5
48. De Smet J, Wagemans J, Boon M et al (2021) The bacteriophage LUZ24 “Igy” peptide
inhibits the Pseudomonas DNA gyrase. Cell Rep 36:109567. https://doi.org/10.1016/j.
celrep.2021.109567
49. Kever L, Hünnefeld M, Brehm J et al (2021) Identification of Gip as a novel phage-encoded
gyrase inhibitor protein of Corynebacterium glutamicum. Mol Microbiol 116:1268–1280.
https://doi.org/10.1111/mmi.14813
50. Glonti T, Chanishvili N, Taylor PW (2010) Bacteriophage-derived enzyme that depolymer-
izes the alginic acid capsule associated with cystic fibrosis isolates of Pseudomonas aerugi-
nosa. J Appl Microbiol 108:695–702. https://doi.org/10.1111/j.1365-­2672.2009.04469.x
51. Drulis-Kawa Z, Majkowska-Skrobek G, Maciejewska B et al (2012) Learning from bacterio-
phages – advantages and limitations of phage and phage-encoded protein applications. Curr
Protein Pept Sci 13:699. https://doi.org/10.2174/138920312804871193
52. Sainath Rao S, Mohan KVK, Atreya CD (2013) A peptide derived from phage display
library exhibits antibacterial activity against E. coli and Pseudomonas aeruginosa. PLoS One
8:e56081. https://doi.org/10.1371/journal.pone.0056081
53. Briers Y, Walmagh M, Grymonprez B et al (2014) Art-175 is a highly efficient antibacterial
against multidrug-resistant strains and persisters of Pseudomonas aeruginosa. Antimicrob
Agents Chemother 58:3774–3784. https://doi.org/10.1128/AAC.02668-­14/SUPPL_FILE/
ZAC007143000SO1.PDF
54. Pires DP, Oliveira H, Melo LDR et al (2016) Bacteriophage-encoded depolymerases: their
diversity and biotechnological applications. Appl Microbiol Biotechnol 100:2141–2151.
https://doi.org/10.1007/s00253-­015-­7247-­0
55. Thandar M, Lood R, Winer BY et al (2016) Novel engineered peptides of a phage lysin
as effective antimicrobials against multidrug-resistant Acinetobacter baumannii. Antimicrob
Agents Chemother 60:2971–2979. https://doi.org/10.1128/AAC.02972-­15
56. Peng SY, You RI, Lai MJ et al (2017) Highly potent antimicrobial modified peptides derived
from the Acinetobacter baumannii phage endolysin LysAB2. Sci Rep 7:1–12. https://doi.
org/10.1038/s41598-­017-­11832-­7
4 Chemical-Biology and Metabolomics Studies in Phage-Host Interactions 95

57. Latka A, Maciejewska B, Majkowska-Skrobek G et al (2017) Bacteriophage-encoded virion-­

associated enzymes to overcome the carbohydrate barriers during the infection process. Appl
Microbiol Biotechnol 101:3103–3119. https://doi.org/10.1007/s00253-­017-­8224-­6
58. Forterre P (2010) Defining life: the virus viewpoint. Orig Life Evol Biosph 40:151–160.
https://doi.org/10.1007/s11084-­010-­9194-­1
59. Forterre P (2016) To be or not to be alive: how recent discoveries challenge the traditional
definitions of viruses and life. Stud Hist Philos Sci Part C Stud Hist Philos Biol Biomed Sci
59:100–108. https://doi.org/10.1016/j.shpsc.2016.02.013
60. Nasir A, Romero-Severson E, Claverie J-M (2020) Investigating the concept and origin of
viruses. Trends Microbiol 28:959–967. https://doi.org/10.1016/j.tim.2020.08.003
61. Moniruzzaman M, Martinez-Gutierrez CA, Weinheimer AR, Aylward FO (2020) Dynamic
genome evolution and complex virocell metabolism of globally-distributed giant viruses. Nat
Commun 11:1710. https://doi.org/10.1038/s41467-­020-­15507-­2
62. Woese CR, Fox GE (1977) Phylogenetic structure of the prokaryotic domain: the primary
kingdoms. Proc Natl Acad Sci 74:5088–5090. https://doi.org/10.1073/pnas.74.11.5088
63. Farner DS (1983) The growth of biological thought. Diversity, evolution, and inheritance.
Auk 100:507–549. https://doi.org/10.1093/auk/100.2.507
64. Woese CR (1987) Bacterial evolution. Microbiol Rev 51:221–271. https://doi.org/10.1128/
MR.51.2.221-­271.1987
65. Woese CR, Kandler O, Wheelis ML (1990) Towards a natural system of organisms: proposal
for the domains Archaea, Bacteria, and Eucarya. Proc Natl Acad Sci 87:4576–4579. https://
doi.org/10.1073/pnas.87.12.4576
66. Ochman H, Moran NA (2001) Genes lost and genes found: evolution of bacterial pathogen-
esis and symbiosis. Science (80- ) 292:1096–1099. https://doi.org/10.1126/science.1058543
67. Stanier RY, Niel CB (1962) The concept of a bacterium. Arch Mikrobiol 42:17–35. https://
doi.org/10.1007/BF00425185
68. Sapp J (2005) The prokaryote-eukaryote dichotomy: meanings and mythology. Microbiol
Mol Biol Rev 69:292–305. https://doi.org/10.1128/MMBR.69.2.292-­305.2005
69. Brüssow H, Canchaya C, Hardt W-D (2004) Phages and the evolution of bacterial pathogens:
from genomic rearrangements to lysogenic conversion. Microbiol Mol Biol Rev 68:560–602.
https://doi.org/10.1128/MMBR.68.3.560-­602.2004
70. Clokie MRJ, Millard AD, Letarov AV, Heaphy S (2011) Phages in nature. Bacteriophage
1:31–45. https://doi.org/10.4161/bact.1.1.14942
71. Little JW (2014) Lysogeny, prophage induction, and lysogenic conversion. Phages 37–54.
https://doi.org/10.1128/9781555816506.CH3
72. Correa AMS, Howard-Varona C, Coy SR et al (2021) Revisiting the rules of life for
viruses of microorganisms. Nat Rev Microbiol 19:501–513. https://doi.org/10.1038/
s41579-­021-­00530-­x
73. Edlin G, Lin L, Kudrna R (1975) λ Lysogens of E. coli reproduce more rapidly than non-­
lysogens. Nature 255:735–737. https://doi.org/10.1038/255735a0
74. Lin L, Bitner R, Edlin G (1977) Increased reproductive fitness of Escherichia coli lambda
lysogens. J Virol 21:554–559. https://doi.org/10.1128/jvi.21.2.554-­559.1977
75. Waldor MK, Mekalanos JJ (1996) Lysogenic conversion by a filamentous phage encoding
cholera toxin. Science (80- ) 272:1910–1914. https://doi.org/10.1126/science.272.5270.1910
76. Sekulovic O, Meessen-Pinard M, Fortier L-C (2011) Prophage-stimulated toxin produc-
tion in Clostridium difficile NAP1/027 lysogens. J Bacteriol 193:2726–2734. https://doi.
org/10.1128/JB.00787-­10
77. Chan BK, Abedon ST, Loc-Carrillo C (2013) Phage cocktails and the future of phage therapy.
Future Microbiol 8:769–783. https://doi.org/10.2217/fmb.13.47
78. Argov T, Azulay G, Pasechnek A et al (2017) Temperate bacteriophages as regulators of host
behavior. Curr Opin Microbiol 38:81–87. https://doi.org/10.1016/j.mib.2017.05.002
79. Abedon ST, Duffy S, Turner PE (2009) Bacteriophage ecology. In: Encyclopedia of micro-
biology. Elsevier, pp 42–57
96 R. Dantas et al.

80. Feiner R, Argov T, Rabinovich L et al (2015) A new perspective on lysogeny: prophages

as active regulatory switches of bacteria. Nat Rev Microbiol 13:641–650. https://doi.
org/10.1038/nrmicro3527
81. Hatfull GF, Hendrix RW (2011) Bacteriophages and their genomes. Curr Opin Virol
1:298–303. https://doi.org/10.1016/j.coviro.2011.06.009
82. Ackermann HW (2011) Bacteriophage taxonomy. Microbiol Aust 32:90. https://doi.
org/10.1071/MA11090
83. Tolstoy I, Kropinski AM, Brister JR (2018) Bacteriophage taxonomy: an evolving discipline.
In: Methods in molecular biology. Humana Press, New York, pp 57–71
84. Shang J, Jiang J, Sun Y (2021) Bacteriophage classification for assembled contigs
using graph convolutional network. Bioinformatics 37:I25–I33. https://doi.org/10.1093/
bioinformatics/btab293
85. Taxonomy browser (Nematoda). https://www.ncbi.nlm.nih.gov/Taxonomy/Browser/
wwwtax.cgi?mode=Info&id=28883&lvl=3&lin=f&keep=1&srchmode=1&unlock.
Accessed 1 May 2022
86. Krupovic M, Prangishvili D, Hendrix RW, Bamford DH (2011) Genomics of bacterial and
archaeal viruses: dynamics within the prokaryotic virosphere. Microbiol Mol Biol Rev
75:610–635. https://doi.org/10.1128/mmbr.00011-­11
87. Buckling A, Rainey PB (2002) Antagonistic coevolution between a bacterium and a bacterio-
phage. Proc R Soc Lond Ser B Biol Sci 269:931–936. https://doi.org/10.1098/rspb.2001.1945
88. Bertozzi Silva J, Storms Z, Sauvageau D (2016) Host receptors for bacteriophage adsorption.
FEMS Microbiol Lett 363:fnw002. https://doi.org/10.1093/femsle/fnw002
89. Nordström K, Forsgren A (1974) Effect of protein A on adsorption of bacteriophages to
Staphylococcus aureus. J Virol 14:198–202. https://doi.org/10.1128/jvi.14.2.198-­202.1974
90. Pedruzzi I, Rosenbusch JP, Locher KP (1998) Inactivation in vitro of the Escherichia coli
outer membrane protein FhuA by a phage T5-encoded lipoprotein. FEMS Microbiol Lett
168:119–125. https://doi.org/10.1111/J.1574-­6968.1998.TB13264.X
91. Hammad AMM (1998) Evaluation of alginate-encapsulated Azotobacter chroococcum
as a phage-resistant and an effective inoculum. J Basic Microbiol 38:9–16. https://doi.
org/10.1002/(SICI)1521-­4028(199803)38:1
92. Cote GL, Krull LH (1988) Characterization of the exocellular polysaccha-
rides from Azotobacter chroococcum. Carbohydr Res 181:143–152. https://doi.
org/10.1016/0008-­6215(88)84030-­8
93. Perry LL, SanMiguel P, Minocha U et al (2009) Sequence analysis of Escherichia coli
O157:H7 bacteriophage ΦV10 and identification of a phage-encoded immunity pro-
tein that modifies the O157 antigen. FEMS Microbiol Lett 292:182–186. https://doi.
org/10.1111/j.1574-­6968.2009.01511.x
94. van Houte S, Buckling A, Westra ER (2016) Evolutionary ecology of prokaryotic immune
mechanisms. Microbiol Mol Biol Rev 80:745–763. https://doi.org/10.1128/MMBR.00011-­16
95. Burmeister AR, Fortier A, Roush C et al (2020) Pleiotropy complicates a trade-off between
phage resistance and antibiotic resistance. Proc Natl Acad Sci U S A 117:11207–11216.
https://doi.org/10.1073/pnas.1919888117
96. Trudelle DM, Bryan DW, Hudson LK, Denes TG (2019) Cross-resistance to phage infection
in Listeria monocytogenes serotype 1/2a mutants. Food Microbiol 84:103239. https://doi.
org/10.1016/j.fm.2019.06.003
97. Yu H, Shang L, Yang G et al (2022) Biosynthetic microcin J25 exerts strong antibacterial,
anti-inflammatory activities, low cytotoxicity without increasing drug-resistance to bacteria
target. Front Immunol 13:1. https://doi.org/10.3389/fimmu.2022.811378
98. Parker JK, Davies BW (2022) Microcins reveal natural mechanisms of bacterial manipula-
tion to inform therapeutic development. Microbiology 168:001175. https://doi.org/10.1099/
mic.0.001175
99. Vincent P, Morero R (2009) The structure and biological aspects of peptide antibiotic micro-
cin J25. Curr Med Chem 16:538–549. https://doi.org/10.2174/092986709787458461
4 Chemical-Biology and Metabolomics Studies in Phage-Host Interactions 97

100. Destoumieux-Garzón D, Duquesne S, Peduzzi J et al (2005) The iron-siderophore transporter

FhuA is the receptor for the antimicrobial peptide microcin J25: role of the microcin Val11-­
Pro16 β-hairpin region in the recognition mechanism. Biochem J 389:869–876. https://doi.
org/10.1042/BJ20042107
101. Mathavan I, Zirah S, Mehmood S et al (2014) Structural basis for hijacking siderophore recep-
tors by antimicrobial lasso peptides. Nat Chem Biol 10:340–342. https://doi.org/10.1038/
NCHEMBIO.1499
102. Luria SE, Human ML (1952) A nonhereditary, host-induced variation of bacterial viruses. J
Bacteriol 64:557–569. https://doi.org/10.1128/jb.64.4.557-­569.1952
103. Tock MR, Dryden DTF (2005) The biology of restriction and anti-restriction. Curr Opin
Microbiol 8:466–472. https://doi.org/10.1016/j.mib.2005.06.003
104. Luria SE (1953) Host-induced modifications of viruses. Cold Spring Harb Symp Quant Biol
18:237–244. https://doi.org/10.1101/SQB.1953.018.01.034
105. Garneau JE, Dupuis M-È, Villion M et al (2010) The CRISPR/Cas bacterial immune sys-
tem cleaves bacteriophage and plasmid DNA. Nature 468:67–71. https://doi.org/10.1038/
nature09523
106. Deveau H, Barrangou R, Garneau JE et al (2008) Phage response to CRISPR-encoded resis-
tance in Streptococcus thermophilus. J Bacteriol 190:1390–1400. https://doi.org/10.1128/
JB.01412-­07
107. Bondy-Denomy J, Pawluk A, Maxwell KL, Davidson AR (2013) Bacteriophage genes that
inactivate the CRISPR/Cas bacterial immune system. Nature 493:429–432. https://doi.
org/10.1038/nature11723
108. Wiedenheft B (2013) In defense of phage. RNA Biol 10:886–890. https://doi.org/10.4161/
rna.23591
109. Seed KD, Lazinski DW, Calderwood SB, Camilli A (2013) A bacteriophage encodes its own
CRISPR/Cas adaptive response to evade host innate immunity. Nature 494:489–491. https://
doi.org/10.1038/nature11927
110. Hanlon GW, Denyer SP, Olliff CJ, Ibrahim LJ (2001) Reduction in exopolysaccharide viscos-
ity as an aid to bacteriophage penetration through Pseudomonas aeruginosa biofilms. Appl
Environ Microbiol 67:2746–2753. https://doi.org/10.1128/AEM.67.6.2746-­2753.2001
111. Sutherland IW (1999) Polysaccharases for microbial exopolysaccharides. Carbohydr Polym
38:319–328. https://doi.org/10.1016/S0144-­8617(98)00114-­3
112. Linhardt RJ, Galliher PM, Cooney CL (1987) Polysaccharide lyases. Appl Biochem
Biotechnol 12:135–176. https://doi.org/10.1007/BF02798420
113. Leiman PG, Battisti AJ, Bowman VD et al (2007) The structures of bacteriophages K1E and
K1-5 explain processive degradation of polysaccharide capsules and evolution of new host
specificities. J Mol Biol 371:836–849. https://doi.org/10.1016/j.jmb.2007.05.083
114. Nwodo UU, Green E, Okoh AI (2012) Bacterial exopolysaccharides: functionality and pros-
pects. Int J Mol Sci 13:14002–14015
115. Scholl D, Adhya S, Merril C (2005) Escherichia coli K1’s capsule is a barrier to bac-
teriophage T7. Appl Environ Microbiol 71:4872–4874. https://doi.org/10.1128/
AEM.71.8.4872-­4874.2005
116. Cornelissen A, Ceyssens PJ, Krylov VN et al (2012) Identification of EPS-degrading activity
within the tail spikes of the novel Pseudomonas putida phage AF. Virology 434:251–256.
https://doi.org/10.1016/j.virol.2012.09.030
117. Coulon C, Vinogradov E, Filloux A, Sadovskaya I (2010) Chemical analysis of cellular and
extracellular carbohydrates of a biofilm-forming strain Pseudomonas aeruginosa PA14. PLoS
One 5. https://doi.org/10.1371/journal.pone.0014220
118. Colvin KM, Irie Y, Tart CS et al (2012) The Pel and Psl polysaccharides provide Pseudomonas
aeruginosa structural redundancy within the biofilm matrix. Environ Microbiol 14:1913–1928.
https://doi.org/10.1111/j.1462-­2920.2011.02657.x
98 R. Dantas et al.

119. Wu JA, Kusuma C, Mond JJ, Kokai-Kun JF (2003) Lysostaphin disrupts Staphylococcus
aureus and Staphylococcus epidermidis biofilms on artificial surfaces. Antimicrob Agents
Chemother 47:3407–3414. https://doi.org/10.1128/AAC.47.11.3407-­3414.2003
120. Aguinaga A, Francés ML, Del Pozo JL et al (2011) Lysostaphin and clarithromycin: a prom-
ising combination for the eradication of Staphylococcus aureus biofilms. Int J Antimicrob
Agents 37:585–587. https://doi.org/10.1016/j.ijantimicag.2011.02.009
121. Drilling AJ, Cooksley C, Chan C et al (2016) Fighting sinus-derived Staphylococcus aureus
biofilms in vitro with a bacteriophage-derived muralytic enzyme. Int Forum Allergy Rhinol
6:349–355. https://doi.org/10.1002/alr.21680
122. Roach DR, Donovan DM (2015) Antimicrobial bacteriophage-derived proteins and
therapeutic applications. Bacteriophage 5:e1062590. https://doi.org/10.1080/2159708
1.2015.1062590
123. Wan X, Hendrix H, Skurnik M, Lavigne R (2021) Phage-based target discovery and its
exploitation towards novel antibacterial molecules. Curr Opin Biotechnol 68:1–7. https://doi.
org/10.1016/J.COPBIO.2020.08.015
124. Yan J, Mao J, Xie J (2014) Bacteriophage polysaccharide depolymerases and biomedical
applications. BioDrugs 28:265–274. https://doi.org/10.1007/S40259-­013-­0081-­Y/METRICS
125. Rodríguez-Rubio L, Martínez B, Donovan DM et al (2013) Bacteriophage virion-­associated
peptidoglycan hydrolases: potential new enzybiotics. Crit Rev Microbiol 39:427–434. https://
doi.org/10.3109/1040841X.2012.723675
126. Sharma U, Paul VD (2017) Bacteriophage lysins as antibacterials. Crit Care 21:99. https://
doi.org/10.1186/s13054-­017-­1681-­6
127. Borysowski J, Weber-Da̧browska B, Górski A (2006) Bacteriophage endolysins as a novel
class of antibacterial agents. Exp Biol Med 231:366–377
128. Nelson D, Loomis L, Fischetti VA (2001) Prevention and elimination of upper respiratory
colonization of mice by group A streptococci by using a bacteriophage lytic enzyme. Proc
Natl Acad Sci U S A 98:4107–4112. https://doi.org/10.1073/pnas.061038398
129. Briers Y, Volckaert G, Cornelissen A et al (2007) Muralytic activity and modular structure
of the endolysins of Pseudomonas aeruginosa bacteriophages? KZ and EL. Mol Microbiol
65:1334–1344. https://doi.org/10.1111/j.1365-­2958.2007.05870.x
130. Walmagh M, Briers Y, Santos SB dos et al (2012) Characterization of modular bacteriophage
endolysins from myoviridae phages OBP, 201φ2-1 and PVP-SE1. PLoS One 7:e36991.
https://doi.org/10.1371/journal.pone.0036991
131. Chanishvili N (2012) Phage therapy-history from Twort and d’Herelle through soviet experi-
ence to current approaches. Adv Virus Res 83:3–40
132. Międzybrodzki R, Hoyle N, Zhvaniya F et al (2021) Current updates from the long-standing
phage research centers in Georgia, Poland, and Russia. Bacteriophages 921–951. https://doi.
org/10.1007/978-­3-­319-­41986-­2_31
133. Schmelcher M, Donovan DM, Loessner MJ (2012) Bacteriophage endolysins as novel anti-
microbials. Future Microbiol 7:1147–1171
134. Matsuda M, Barksdale L (1966) Phage-directed synthesis of diphtherial toxin in
non-­toxinogenic Corynebacterium diphtheriae. Nature 210:911–913. https://doi.
org/10.1038/210911a0
135. Fujii N, Oguma K, Yokosawa N et al (1988) Characterization of bacteriophage nucleic acids
obtained from Clostridium botulinum types C and D. Appl Environ Microbiol 54:69–73.
https://doi.org/10.1128/aem.54.1.69-­73.1988
136. O’Brien AD, Newland JW, Miller SF et al (1984) Shiga-like toxin-converting phages from
Escherichia coli strains that cause hemorrhagic colitis or infantile diarrhea. Science (80- )
226:694–696. https://doi.org/10.1126/science.6387911
137. Rodríguez-Rubio L, Haarmann N, Schwidder M et al (2021) Bacteriophages of shiga toxin-­
producing escherichia coli and their contribution to pathogenicity. Pathogens 10. https://doi.
org/10.3390/pathogens10040404
4 Chemical-Biology and Metabolomics Studies in Phage-Host Interactions 99

138. Silpe JE, Wong JWH, Owen SV et al (2022) The bacterial toxin colibactin triggers prophage
induction. Nature 6037900(603):315–320. https://doi.org/10.1038/s41586-­022-­04444-­3
139. Nedialkova LP, Sidstedt M, Koeppel MB et al (2016) Temperate phages promote colicin-­
dependent fitness of Salmonella enterica serovar Typhimurium. Environ Microbiol
18:1591–1603. https://doi.org/10.1111/1462-­2920.13077
140. Hilsenbeck JL, Park H, Chen G et al (2004) Crystal structure of the cytotoxic bac-
terial protein colicin B at 2.5 Å resolution. Mol Microbiol 51:711–720. https://doi.
org/10.1111/j.1365-­2958.2003.03884.x
141. Kronheim S, Daniel-Ivad M, Duan Z et al (2018) A chemical defence against phage infection.
Nature 564:283–286. https://doi.org/10.1038/s41586-­018-­0767-­x
142. Nicholson J, Lindon J, Xenobiotica EH-, 1999 undefined (2008) “Metabonomics”: under-
standing the metabolic responses of living systems to pathophysiological stimuli via mul-
tivariate statistical analysis of biological NMR. Taylor Fr 29:1181–1189. https://doi.
org/10.1080/004982599238047
143. Klassen A, Faccio AT, Canuto GAB et al (2017) Metabolomics: definitions and significance in
systems biology. Adv Exp Med Biol 965:3–17. https://doi.org/10.1007/978-­3-­319-­47656-­8_1/
FIGURES/2
144. Roberts LD, Souza AL, Gerszten RE, Clish CB (2012) Targeted metabolomics. Curr Protoc
Mol Biol 98:30.2.1–30.2.24. https://doi.org/10.1002/0471142727.MB3002S98
145. Vinayavekhin N, Saghatelian A (2010) Untargeted metabolomics. Curr Protoc Mol Biol.
https://doi.org/10.1002/0471142727.MB3001S90
146. Fiehn O (2002) Metabolomics – the link between genotypes and phenotypes. Funct Genomics
155–171. https://doi.org/10.1007/978-­94-­010-­0448-­0_11
147. Griffin JL (2006) The Cinderella story of metabolic profiling: does metabolomics get to go
to the functional genomics ball? Philos Trans R Soc B Biol Sci 361:147–161. https://doi.
org/10.1098/RSTB.2005.1734
148. Fiehn O, Robertson D, Griffin J et al (2007) The metabolomics standards initiative (MSI).
Metabolomics 3:175–178. https://doi.org/10.1007/S11306-­007-­0070-­6/FIGURES/2
149. Rosato A, Tenori L, Cascante M et al (2018) From correlation to causation: analysis of
metabolomics data using systems biology approaches. Metabolomics 144(14):1–20. https://
doi.org/10.1007/S11306-­018-­1335-­Y
150. Zhao X, Shen M, Jiang X et al (2017) Transcriptomic and metabolomics profiling of phage-­
host interactions between phage PaP1 and Pseudomonas aeruginosa. Front Microbiol 8:548.
https://doi.org/10.3389/FMICB.2017.00548/BIBTEX
151. Zhao X, Chen C, Jiang X et al (2016) Transcriptomic and metabolomic analysis revealed
multifaceted effects of phage protein Gp70.1 on Pseudomonas aeruginosa. Front Microbiol
7:1519. https://doi.org/10.3389/FMICB.2016.01519/BIBTEX
152. Chevallereau A, Blasdel BG, De Smet J et al (2016) Next-generation “-omics” approaches
reveal a massive alteration of host RNA metabolism during bacteriophage infection of
Pseudomonas aeruginosa. PLoS Genet 12:e1006134. https://doi.org/10.1371/JOURNAL.
PGEN.1006134
153. Hsu BB, Gibson TE, Yeliseyev V et al (2019) Dynamic modulation of the gut microbiota and
metabolome by bacteriophages in a mouse model. Cell Host Microbe 25:803–814.e5. https://
doi.org/10.1016/J.CHOM.2019.05.001
154. Han ML, Nang SC, Lin YW et al (2022) Comparative metabolomics revealed key pathways
associated with the synergistic killing of multidrug-resistant Klebsiella pneumoniae by a
bacteriophage-polymyxin combination. Comput Struct Biotechnol J 20:485–495. https://doi.
org/10.1016/J.CSBJ.2021.12.039
155. Sonkar K, Purusottam RN, Sinha N (2012) Metabonomic study of host-phage interaction
by nuclear magnetic resonance- and statistical total correlation spectroscopy-based analysis.
Anal Chem 84:4063–4070. https://doi.org/10.1021/AC300096J/SUPPL_FILE/AC300096J_
SI_001.PDF
100 R. Dantas et al.

156. Ankrah NYD, May AL, Middleton JL et al (2014) Phage infection of an environmen-
tally relevant marine bacterium alters host metabolism and lysate composition. ISME J
85(8):1089–1100. https://doi.org/10.1038/ismej.2013.216
157. Lorenzo-Rebenaque L, Casto-Rebollo C, Diretto G et al (2022) Examining the effects of
Salmonella phage on the caecal microbiota and metabolome features in Salmonella-free
broilers. Front Genet 13:3219. https://doi.org/10.3389/FGENE.2022.1060713/BIBTEX
158. Perez de Souza L, Alseekh S, Scossa F, Fernie AR (2021) Ultra-high-performance liquid
chromatography high-resolution mass spectrometry variants for metabolomics research. Nat
Methods 187(18):733–746. https://doi.org/10.1038/s41592-­021-­01116-­4
159. Saccenti E, Hoefsloot HCJ, Smilde AK et al (2014) Reflections on univariate and multi-
variate analysis of metabolomics data. Metabolomics 10:361–374. https://doi.org/10.1007/
S11306-­013-­0598-­6/FIGURES/6
160. Kumar N, Hoque MA, Sugimoto M (2018) Robust volcano plot: identification of differen-
tial metabolites in the presence of outliers. BMC Bioinform 19:128. https://doi.org/10.1186/
s12859-­018-­2117-­2
161. Pinto RC (2017) Chemometrics methods and strategies in metabolomics. Adv Exp Med Biol
965:163–190. https://doi.org/10.1007/978-­3-­319-­47656-­8_7/FIGURES/5
162. Considine EC, Thomas G, Boulesteix AL et al (2018) Critical review of reporting of the
data analysis step in metabolomics. Metabolomics 14:1–16. https://doi.org/10.1007/
S11306-­017-­1299-­3/TABLES/17
163. Cho HW, Kim SB, Jeong MK et al (2008) Discovery of metabolite features for the modelling
and analysis of high-resolution NMR spectra. Int J Data Min Bioinform 2:176. https://doi.
org/10.1504/IJDMB.2008.019097
164. Wang M, Carver JJ, Phelan VV et al (2016) Sharing and community curation of mass spec-
trometry data with global natural products social molecular networking. Nat Biotechnol
34:828–837. https://doi.org/10.1038/NBT.3597
165. Dührkop K, Fleischauer M, Ludwig M et al (2019) (2019) SIRIUS 4: a rapid tool for turning
tandem mass spectra into metabolite structure information. Nat Methods 164(16):299–302.
https://doi.org/10.1038/s41592-­019-­0344-­8
166. Horai H, Arita M, Kanaya S et al (2010) MassBank: a public repository for sharing mass spec-
tral data for life sciences. J Mass Spectrom 45:703–714. https://doi.org/10.1002/JMS.1777
167. Mehta SS (2020) MassBank of North America (MoNA): an open-access, auto-curating mass
spectral database for compound identification in metabolomics presentation
168. Chenomx Inc | Metabolite discovery and measurement. https://www.chenomx.com/.
Accessed 28 Apr 2023
169. Sumner LW, Amberg A, Barrett D et al (2007) Proposed minimum reporting standards for
chemical analysis. Metabolomics 3:211–221. https://doi.org/10.1007/s11306-­007-­0082-­2
170. Howard-Varona C, Lindback MM, Bastien GE et al (2020) Phage-specific metabolic repro-
gramming of virocells. ISME J 144(14):881–895. https://doi.org/10.1038/s41396-­019-­0580-­z
171. Kanehisa M, Subramaniam (2002) The KEGG database. Novartis Found Symp 247:91–103.
https://doi.org/10.1002/0470857897.CH8
172. Kanehisa M, Furumichi M, Sato Y et al (2021) KEGG: integrating viruses and cellular organ-
isms. Nucleic Acids Res 49:D545–D551. https://doi.org/10.1093/NAR/GKAA970
173. Keseler IM, Gama-Castro S, Mackie A et al (2021) The EcoCyc database in 2021. Front
Microbiol 12:2098. https://doi.org/10.3389/FMICB.2021.711077/BIBTEX
174. Papaianni M, Cuomo P, Fulgione A et al (2020) Bacteriophages promote meta-
bolic changes in bacteria biofilm. Microorganisms 8:480. https://doi.org/10.3390/
MICROORGANISMS8040480
Chapter 5
Advances in Mass
Spectrometry-­Metabolomics Based
Approaches

Nerilson Marques Lima, Gabriel Franco dos Santos,

Gesiane da Silva Lima, and Boniek Gontijo Vaz

5.1 Introduction

The search for pharmacologically active molecules and high value-added biotech-
nological products from natural sources has led to the development of more sensi-
tive and robust analytical tools due to the need to develop rapid and efficient analyses
to obtain metabolic profiles of complex sample matrices such as plants and micro-
organisms. Thus, the detection and identification of known molecules and the sepa-
ration of chemical constituents with unprecedented molecular characteristics in
complex extracts of natural products require efficient analytical methodologies
associated with state-of-the-art spectroscopic techniques.
In this context, state-of-the-art spectrometric techniques aiming at the character-
ization, and quantification of chemical species present in natural matrices have
emerged as promising tools in dynamic studies of complex metabolite mixtures,
enabling the analysis of the metabolome (set of metabolites produced by an organ-
ism), which includes a wide range of chemical constituents with very diverse
physico-chemical characteristics [39].
Mass spectrometry (MS) has established itself as an essential technique in
metabolomics approaches, an omics science characterized by qualitative and quan-
titative determination of the metabolome through various analytical platforms [11,
70]. Hence, it became a powerful analytical tool in assessing the molecular com-
plexity of natural products and rapidly obtaining metabolomic profiles through inte-
grated analyses of biological samples [40].

N. M. Lima · G. F. dos Santos · G. da Silva Lima · B. G. Vaz (*)

Federal University of Goias, Goiânia, GO, Brazil
e-mail: [email protected]

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 101
T. Pacheco Fill (ed.), Microbial Natural Products Chemistry, Advances in Experimental
Medicine and Biology 1439, https://doi.org/10.1007/978-3-031-41741-2_5
102 N. M. Lima et al.

MS has been successfully applied in Metabolomics due to its high precision,

versatility, sensitivity, and wide dynamic range [39]. Thus, it allows the detection of
small molecules even at low concentrations in different types of samples and pro-
vides important information for the precise determination of molecular mass, and
assists in structural identification studies. In addition, the coupling with chromato-
graphic and electrophoretic separation techniques has resulted in more efficient
technologies such as GC-MS, LC-MS, LC-NMR-MS, and CE-MS, whose advances
have allowed more comprehensive analyzes and expanded their applications in
metabolomics studies of natural products.
The development of more robust hyphenated techniques, the ability of MSn frag-
mentation, and obtaining thousands of spectra in short analysis times produce dense
metabolomic datasets. Therefore, data mining tools have to be used in order to get
the biological responses investigated. In this sense, MS-based hyphenated tech-
niques are the preferable choice for many natural product researchers, with a wide
range of potentialities, including resolution and structural quantification of isomers.
In addition, the development of ambient ionization techniques has expanded the
applications in microbial metabolomics research considerably [38].
New MS configurations and the application of automated tools for processing the
data have made classical natural product chemistry obsolete. They have allowed the
molecular fingerprints of samples without the need for compound isolation. The
ability to isolate and fragment the ion of interest to obtain structural information for
molecular characterization in a process known as tandem MS or MS/MS has been
essential in analyzing micro- and macromolecules [79]. However, the biochemical
interpretation of the obtained data essentially depends on the assignment of molecu-
lar structures from spectral data, the process of which has been a challenge in
metabolomic analyses [9]. Thus, the development of new approaches for in silico
annotation of natural products based on fragmentation pathways, reactivity, and
ionization mechanisms has been essential in the characterization of metabolites
[16, 50].
The metabolome is composed of hundreds of metabolites with very different
physico-chemical properties and thereby challenging to analyze, especially for nat-
ural sources chemically little explored such as microorganisms. Depending on the
origin of the analytes investigated, microbial metabolomics can be roughly classi-
fied as metabolic footprint analysis and metabolic fingerprint analysis [27]. Due to
the spectral complexity of the microbial samples, the investigation of metabolic
pathways and large-scale mining of the specialized metabolome is an essential step
in metabolomics studies of microorganisms. Hence, the use of hyphenated analyti-
cal techniques using gas chromatography (GC), liquid chromatography (LC), and
capillary electrophoresis (CE) coupled with high-resolution mass spectrometry
(HRMS) methods are indispensable in the qualitative and quantitative evaluation of
metabolites from microbial extracts [5, 91].
The evolution of MS techniques has enhanced the applications of microbial
metabolomics, aiming the detection and structural determination of metabolites, in
addition to identifying microorganisms in environmental or clinical samples. The
progress in mass resolution and precision combined with the development of
5 Advances in Mass Spectrometry-Metabolomics Based Approaches 103

easy-­to-­operate instruments, in addition to the expansion in the development of

MS-based analytical methods, enables the microbial fingerprint analyses to chemo-
taxonomy studies involving a wide range of species [53]. Furthermore, the tech-
nique allows single-cell metabolomics analysis from cultured microorganisms and
evaluates metabolite fluxes at the cellular level [36].
Although the technique provides accurate information regarding the molecular
composition of the metabolome of the microorganisms or microbiomes investi-
gated, the interpretation of data from untargeted experiments needs to be evaluated
in detail since these data can be influenced by protocols of sample collection, extrac-
tion, preparation, data acquisition and analysis [65].
In this work, we highlight current advances in the application of MS in the field
of microbial metabolomics and emphasize emerging mass spectrometry tools used
in such investigations. In addition, we provide a fundamental introduction to ambi-
ent ionization techniques applied to microbial metabolomics studies; imaging of
microbial samples in metabolomics applications; and the application of microbial
metabolomics in the discovery of new clinically useful bioactive agents. Figure 5.1
shows the increase in publications over the last 10 years concerning microbial
metabolomics and highlighted the impact of MS in microbial metabolomics field.

5.2 Mass Spectrometry-Based Metabolomics for Bioactive

Microbial Compounds Discovery

Metabolomics is an omics approach characterized by the large-scale quantification

or identification of small molecules produced by a biological system using analyti-
cal platforms. A comprehensive understanding of the spectroscopic/spectrometric
data obtained by the different analytical tools employed in metabolomics can

Publicaon number per year

600

500

400

300

200

100

0
2021 2020 2019 2018 2017 2016 2015 2014 2013 2012 2011

Fig. 5.1 The number of research papers published per year from microbial metabolomics
104 N. M. Lima et al.

contribute to the discovery or accelerate the search for bioactive substances from
natural sources. Thus, due to the advantages of mass spectrometry over other ana-
lytical techniques, including sensitivity, small sample volume, and the possibility of
coupling with a separation technique, MS is the technology of choice in many
metabolomics approaches. Besides the possibility of quantification in targeted stud-
ies, it has been widely used in untargeted studies since MS/MS fragmentation data
provide valuable information in structural elucidation and allow the dereplication of
extracts by database analysis [1, 42, 85].
In this context, mass spectrometry-based metabolomics constitutes a powerful
tool for the identification of metabolites of pharmacological interest. Thus, derepli-
cation strategies allow the selection of substances with unique molecular character-
istics and avoid the classical processes of laborious and costly isolation of known
molecules. For this purpose, MS-based metabolomics analysis techniques include
mainly GC-MS, LC-MS, and CE-MS, which are chosen based on the sample’s
chemical composition rather than on the species of microorganisms evaluated
[18, 27].
The specialized literature brings remarkable examples of applying mass
spectrometry-­based metabolomics in the search for new biologically active agents
from microbial sources. Moreover, metabolome determination provides valuable
information on the role of these living beings and their symbiotic, mutualistic, and
pathogenic relationships with other organisms, whose relationships may result in
the synthesis of metabolites with promising biological properties.
Metabolomics associated with genomic data and integration with molecular biol-
ogy information contributes to innovations in natural products studies aiming at
pharmaceutical and biotechnological applications. Hence, modern analytical tech-
nologies are fundamental in these approaches since they decrease the detection lim-
its and expand the diversity of detectable substances. Moreover, when the mass
spectrometer is incompatible with separation systems the microbial extracts can be
injected directly without the need for prior separation using chromatography [4, 63].
The ionization source most commonly employed in mass spectrometry-based
metabolomics studies is electrospray ionization (ESI) which produces charged ions
in a gas phase that occurs when samples are subjected to high voltage, nebulized in
a mist, and are desolvated into analyte ions [79]. However, it is not suitable for the
analysis of low oxygenated/nitrogenated metabolites such as terpenes or steroids,
which the most appropriate form of ionization is matrix-assisted laser desorption
ionization (MALDI) or atmospheric pressure chemical ionization (APCI) [76].
Aiming at metabolomic analyses, comprehensive information cannot be obtained
by only parent ion scan (MS1) analyses since it provides only m/z and molecular
formula data. Hence, structural data from fragmentation experiments commonly
using fragmentation methods such as collision-induced dissociation (CID), electron
capture dissociation, and electron transfer dissociation (ETD) is needed [23].
The application of microbiological extract samples by direct infusion to ultra-­
high resolution mass spectrometers has been an excellent strategy for obtaining the
metabolome of bacteria and fungi since it removes the polarity constraints imposed
by chromatographic columns [30, 63]. Spectrometers employed include Fourier
5 Advances in Mass Spectrometry-Metabolomics Based Approaches 105

transform ion cyclotron resonance (FTICR) and Orbitrap mass spectrometers.

However, FTICR is the least used due to its high costs and demanding analyst expe-
rience [30, 63].
Direct infusion methodology offers many advantages concerning the practicality
of experiments and analysis throughput. It is possible to discriminate microbiologi-
cal samples as yeast mutants from simple operations using high-resolution equip-
ment with accurate m/z detection such as FTICR, Orbitrap MS, and multipass TOF
MS. Whereas the direct injection can avoid the laborious sample pretreatment pro-
cedure, there are challenges associated with ion suppression, isomer differentiation,
and accurate quantification [35, 46].
Therefore, due to the complexity of microbiological samples, there is a need for
hyphenated analytical techniques. The most commonly employed methods in
metabolomic studies are GC-MS and LC-MS. Methodologies using GC-MS have
been widely applied in microbial metabolomics since they can profile a wide range
of substances with distinct physico-chemical properties in a single sample.
Considering some advantageous features of the technique such as chromatography
columns larger than 100 m, high resolution and reproducibility of separation, and
the stable fragmentation patterns, it is possible to obtain a higher peak capacity and
to obtain structural and quantitative information of a more significant number of
metabolites, including polar compounds such as carbohydrates and amino acids [27].
Furthermore, many advantages are achieved due to the versatility of column
types and the ease of configuration of two-dimensional GC-MS systems, which is
considered the most powerful tool for volatiles analysis [27]. The two-dimensional
system offers high selectivity, sensitivity, separation power, and speed. The different
configurations such as GC × GC-TOF MS and GC-triple quadrupole MS/MS enable
high acquisition and scanning and high quantification accuracy and detection sensi-
tivity [8, 27, 46]. Therefore, many applications are reported in the literature employ-
ing this type of microbial metabolomics system. Yao et al. [94] used the MS
GC × GC-time-of-flight (TOF) system to obtain metabolomic profiles of volatiles of
18 types of Luzhoulaojiao liquor, whose results showed the identification of 320
compounds and quantification of 13 bioactive compounds [94].
However, due to the limitations of GC-MS, liquid chromatography coupled to
mass spectrometry (LC-MS) is the most widely used technique in microbial metab-
olomics. There is a huge range of systems involving LC-MS such as triple quadru-
pole (QqQ), quadrupole ion trap (QIT), quadrupole time-of-flight (QTOF),
quadrupole Orbitrap, triple quadrupole linear ion trap (QTRAP), which are widely
employed in qualitative and quantitative evaluation of microbial metabolites with
interesting pharmacological properties [39].
For quantifying low concentration compounds often found in natural product
samples, the LC-QqQ-MS system is the most recommended due to the high sensi-
tivity achieved in MRM experiments (multiple reaction monitoring). At the same
time, the investigations involving metabolite composition require tandem MS
instruments that allow two (MS/MS) or more (MSn) sequential stages of analysis
since structural information obtained from the fragmentation profiles of the metabo-
lites is required [39]. Data can be obtained through data-dependent acquisition in
106 N. M. Lima et al.

these analyses, which works by acquiring a search scan. The process repeatedly
involves the isolation of the charged molecule based on signal intensity and subse-
quent fragmentation by MS/MS [7].
Furthermore, current advances in liquid chromatography involving the develop-
ment of ultra-high performance LC systems or ultra-fast LC systems as well as the
combination of different LC systems resulting in multidimensional LC-MS analy-
sis, have provided significant improvements in the discovery of new biologically
active agents from complex microbiological sample matrices [95]. Due to the prob-
lems related to the high volume ion pairs found in LC-MS systems, nanoelectro-
spray ionization MS systems have been developed [86]. These tools have been
successfully applied in the metabolomic analysis of natural products to discriminate
between active and inactive samples for certain bioactivity [18, 27].
Among all separation techniques hyphenated to mass spectrometry, capillary
electrophoresis coupled to mass spectrometry (CE-MS) is the least employed, since
it is recommended only for the analysis of highly polar and charged low molecular
weight metabolites and is considered a complementary technique to LC-MS and
GC-MS in metabolomics approaches [34, 60].
All the techniques mentioned above constitute valuable tools in mass
spectrometry-­based metabolomics for bioactive microbial compounds discovery,
whose main objective is to identify and quantify metabolites of molecular masses
below 2000 Da with biological properties from microbial samples ranging from
solids (e.g. directly from surfaces) to volatiles that the microbiome releases [2, 58,
74, 89].
Although it is impossible to obtain the entire microbial metabolome by mass
spectrometry due to the problems associated with extraction, ionization, and detec-
tion of metabolites, the possibility exists to obtain comprehensive information of
the microorganisms and their functional role metabolites [7]. Although the frag-
mentation profiles obtained by MS/MS and MSn cover only a fraction of the micro-
bial metabolome, untargeted metabolomics directs the isolation of bioactive
metabolites that are successful sources of drug leads [17].
Many advances have been made in mass spectrometry-based metabolomics and
these current advances have culminated in increased potential for targeted and
untargeted metabolomics approaches to the identification and quantification of bio-
logically active metabolites from microbial sources. Due to the challenges in sepa-
rating structural isomers, MS systems coupled with ion mobility spectrometry
(IMS) have been developed and successfully applied to natural products studies.
This system acts by separating the isomers because the different geometries of these
substances provide different SCC values, whose values have a very high reproduc-
ibility [10]. Current research has shown the application of this technique in the
analysis of microbial metabolites, such as in the detection of streptorubin B from
S. coelicolor species [54], and in the screening of mutant species of Nocardiopsis
bacteria that have antibiotic resistance effects on the production of secondary
metabolites [19].
5 Advances in Mass Spectrometry-Metabolomics Based Approaches 107

The development of more sensitive and versatile tools for the evaluation of the
metabolic coverage of microbiological samples is a great challenge since databases
register few structures of secondary metabolites from microbial origin and the iso-
lated microorganisms from different terrestrial and marine environments reveal an
arsenal of substances with novel structural chains that may contribute to the discov-
ery of new NPs with therapeutic potential. In this context, new MS-based analytical
technologies and methodologies have been developed to prioritize substances within
a biological context, perform rapid and high-throughput screening of the microbial
metabolome, and provide structural information to elucidate new bioactive agents
[17, 29]. Thus, advances have been made in increasing the sensitivity of mass
spectrometers.
The application of high-sensitivity spectrometric equipment such as TOF,
Orbitrap, and FT-ICR associated with ultra-efficient chromatographic systems has
revealed a large number of metabolites for antibiotic research based on natural
products. Some notable examples have been reported. An Agilent 1290 Infinity II
UHPLC system (Agilent Technologies) hyphenated to an ImpactII ultrahigh-­
resolution Qq-TOF mass spectrometer (Bruker Daltonics) equipped with electro-
spray ionization (ESI) source for mass spectrometry (MS/MS) analysis was used in
untargeted metabolomics approaches to analyze bacterial isolates from healthy and
diseased corals for the detection of antimicrobial components. Through metabolo-
mics and in silico analyses, candidates with biotechnological and pharmaceutical
potential were selected [20].
The association of metabolomics with tools guided by bioassays has been applied
to analyze microbial extracts for the selection of isolates in experiments to enhance
the production of antimicrobial compounds. Analysis of the extracts obtained on an
Accela HPLC (Thermo Scientific) coupled to an Exactive mass spectrometer
(Orbitrap) with an electrospray ionization source resulted in the isolation of 58 acti-
nomycetes from different soil samples and to discriminate the bioactive metabolites
(P < 0.01) present in a crude extract or fraction, even at nanogram levels [72].
Untargeted metabolomics evaluations of fungal extracts isolated from the tunica
of Ciona and the analysis of the molecular network data obtained by UPLC-­
QToF-­MS/MS showed a chemodiversity of compounds belonging to 22 natural
products families. The high bioactivity of the evaluated extracts towards human
pathogenic bacteria and fungi showed substances with potential for developing new
antibiotics [81].
Although analysis of microbial extracts is the majority of the studies in targeted
and untargeted metabolomics, advances in this field have been made through spatial
mapping of microbial natural products.
108 N. M. Lima et al.

5.3 Spatial Mapping of Natural Products by Mass

Spectrometry Imaging

Spatial mapping of biomolecules is one of the most innovative strategies in micro-

organism metabolomics, which detects and describes the spatial location of metabo-
lites involved in metabolic phenotypes, defense mechanisms, and interactions in the
microbiome [28]. Hence, mass spectrometry imaging (MSI) constitutes a powerful
tool in analyzing the microbial metabolome and elucidating the role of metabolites
in the interactions of these microorganisms with the host and the environment [73].
Different technologies employing MSI deliver more comprehensive analysis to
elucidate the metabolome involved in biological interactions as they enable visual-
ization of the spatial distribution of small molecules and macromolecules over a
broad mass range (~300–5000 m/z) directly from the microbial colony or histologi-
cal fraction of the biological sample [61, 87]. To obtain 2D/3D maps and visualize
the distribution of ions in the sample, usually, the total area of the sample is subdi-
vided into pixels (raster points x and y), which are individually analyzed by a laser,
and the mass spectra are obtained for each pixel [74]. Sample preparation involves
several steps and depends on the ionization sources and mass spectrometer analyz-
ers used, ending with data analysis and interpretation [92].
Despite the few reports described for MSI involving microbial metabolomics,
the technique has shown effective and reliable results for the distribution of biomol-
ecules in biological tissues [3, 12, 22]. To obtain the imaging data and the charac-
terization of the target metabolites, it is necessary to work with equipments whose
mass analyzers provide high mass accuracy and MS/MS or MSn spectra. For this
reason, TOF-type and orbitrap analyzers combined with MALDI (Matrix-Assisted
Laser Desorption Ionization), TOF-SIMS (Time-of-Flight Secondary Ion Mass
Spectrometry), LA-ICP (Laser Ablation Inductively Coupled Plasma), LAESI
(Laser Ablation Electrospray Ionization) and DESI (Desorption electrospray ioniza-
tion) ionization sources are the most common devices [3, 22, 49, 66, 75, 80].
Among these ionization techniques, ambient ionization sources such as DESI
and LA-ICP have gained prominence as they require no sample preparation and are
attractive alternatives for the analysis of bacterial colonies in the growth medium
itself [93]. Other promising atmospheric ionization methodologies used in the spa-
tial mapping of biological samples include Direct Analysis in Real Time (DART),
Desorption Atmospheric Pressure Chemical Ionization (DAPCI), Electrospary-­
assisted Laser Desorption Ionization (ELDI), Liquid Extraction Surface Analysis
(LESA), nanospray DESI (nanoDESI) and liquid microinjection surface sampling
probe (LMJ-SSP) [53].
On the other hand, vacuum conditions such as Fourier transform ion cyclotron
resonance (FT-ICR) mass spectrometer provide higher resolution for the ion species
of interest [93]. Although FTICR and Orbitrap mass spectrometers are excellent
alternatives as they achieve high resolving power and accuracy in structural deter-
mination, the acquisition time of HRMS-based MSI (1 pixel/s) is slower than that of
TOF (40 pixels/s) [66, 75, 80]. Thus, there is no ideal mass analyzer for microbial
5 Advances in Mass Spectrometry-Metabolomics Based Approaches 109

MSI analysis and the choice of each depends on the objective of the experiment,
either for simple mapping of the metabolite target in the sample or for direct evalu-
ation of microbial chemical processes at the atomic and molecular levels [93].
We will discuss the characteristics and applications of some of these ionization
sources in targeted metabolomics approaches to microbial samples with a focus on
metabolites present on the native surface of the microbial community.
Metabolomic analysis of endophytic fungi is the most reported work in the litera-
ture involving microbial natural products. Generally, bioactivity assays are per-
formed to select the most promising extracts to isolate bioactive constituents.
However, through MSI tools such as DESI, it is possible to obtain chemical profil-
ing information in situ culture and visualize temporal distribution patterns of chemi-
cal constituents in the tissue of intact biological samples with high speed, accuracy,
and versatility [64]. Besides, the technique offers several advantages over MALDI-­
MSI, as it does not require sample preparation [12].
MALDI-MSI is the most widely used spatial mapping technique that provides
the most metabolic coverage. MALDI-MSI has been applied for many applications
in the analysis of naturally occurring metabolites, including fungi and bacteria. The
method provides 2D spatial distribution data of the analytes present in the living
organism, such as in the mapping of antibiotics produced by Gram-positive bacteria
of the genus Streptomyces living in symbiosis with burrowing wasps [12]. Although
2D MALDI-MSI provides broad metabolic coverage of the sample surface, it still
has many limitations since the natural environment of the microorganism has 3D
architecture. Therefore, the development of technologies using 3D MALDI-MSI
has been enabling to overcome these challenges as it allows detection of the ana-
lytes within the environment and not just on the surface. 3D distribution data can be
achieved in this sample type by mounting MSI images of the tissue taken in con-
secutive sections at predefined intervals. An exciting example of this methodology
shows its application in obtaining depth profiles of microbial substances secreted in
the agar medium [88].
Numerous applications of MALDI-MSI in the microbiological field are reported
in the literature, and many findings obtained from this technique have helped
researchers to gain insights concerning interactions between different species of
microorganisms in the same environment, which information could not be obtained
by molecular fingerprints of microbial extracts. Some remarkable examples include
the application of MALDI-MSI for the investigation of allelopathic interactions
between bacterial species and the detection of competition molecules from micro-
bial culture of the freshwater cyanobacterial species Microcystis aeruginosa with its
antagonist Pseudomonas grimontii. This study allowed to obtain conclusive results
of the allelopathic interactions between these organisms and identify the metabo-
lites of the Gram-negative bacterium P. grimontii that inhibit the growth of the toxic
cyanobacterium M. aeruginosa [14].
Although MALDI-MSI is the most commonly used technique, it does not repre-
sent the most efficient technique for analyzing microbial culture metabolites. Thus,
the comparison of the ionization efficiency of the analytes present in microbial sys-
tems by MSI techniques and their influence on the metabolic coverage of microbial
110 N. M. Lima et al.

colonies has allowed targeting the most appropriate technique for metabolomics
investigations of selected microorganisms. Lukowski et al. [52] employed metal-­
assisted laser desorption/ionization (MetA-LDI) MSI and MALDI-MSI for detect-
ing biofilm metabolites from Bacillus subtilis colonies and compared the efficiency
of both techniques. The results showed that despite finding high overlap in the mol-
ecules in both methods, MetA-LDI MSI was more efficient in detecting neutral
lipids and small molecules, while MALDI MSI detected other lipids and surfactant
lipopeptides produced by Bacillus subtilis [52].
MALDI-MSI is a pioneering technique that allows the investigation of the distri-
bution of biomolecules through direct tissue analysis. In the last decades, several
other tools have been developed for the spatial mapping of substances of medicinal
interest and have been successfully applied to microbial metabolomics. A good
example is the Rapid Evaporative Ionization (REIMS), which allows real-time
microbial culture analysis and does not require solvent flow, thereby avoiding foul-
ing problems as the analytes evaporate with the water in the culture medium. This
technique has emerged as a powerful tool in microbial metabolomics assessments
because it allows metabolomic profiles of specific bacterial species and metabolites
produced by living microorganisms and can be performed without sample prepara-
tion [77].
More sophisticated hybrid methodologies such as LAESI-MS coupled with ion
mobility separation proved to be an efficient strategy for 2D and 3D mapping
Bacillus subtilis and Escherichia coli bacterial species directly from agar plates. In
this work, the authors obtained a metabolic profile with the detection of 400 metab-
olites and evaluated the interaction of an antibiotic with the bacterial culture [47].
Advances in microbial metabolomics have been achieved by combining different
MSI techniques. Current work shows the combination of fluorescence in situ hybrid-
ization microscopy (FISH) and high-resolution atmospheric pressure mass spec-
trometry imaging (AP-MALDI-MSI) to investigate host-microbe symbiosis
processes and their metabolic interactions. This technology brings advantages to the
individual technique because it allows the alignment and integration of metabolite
and fluorescent images in the same portion of the tissue for mapping host and sym-
biont metabolites, which has been successfully applied in metabolomics approaches
involving the elucidation of the mechanisms and the metabolites involved in the
symbiosis of molluscan species with bacteria [28]. The technique allows visualiza-
tion of the distribution of metabolites associated with this interaction using MS data
mapped on the surface or 3D volume [26, 28, 48, 67].
However, the arsenal of data obtained by these techniques requires complex pro-
cessing and treatment to the desired biological answers. Thus, very diverse software
has been developed for MSI evaluations that allow rapid and accurate analysis of the
images obtained, such as METASPACE. This software has been applied to the
investigation of the metabolic content and assessment of the chemical exchange
factors from the interactions existing between bacteria of the genera Actinomycetes,
Pseudomonas, and Bacillus, whose results discriminate the ions associated with the
5 Advances in Mass Spectrometry-Metabolomics Based Approaches 111

Fig. 5.2 Imaging mass spectrometry experiments were used to identify the spatial distributions of
metabolites from microbial colonies to detect biologically active compounds

metabolites responsible for these interactions [62]. In addition, orthogonal analysis

techniques can be used as complementary tools in imaging microbial samples [6].
The great potential of the techniques mentioned above for microbial metabolo-
mics lies in identifying and understanding the mechanisms involved in the meta-
bolic exchange of two or more microorganisms from microbial cultures or tissues.
The detection, identification, and quantification of metabolites involved in these
interactions allow the discovery of compounds with antimicrobial potential to
obtain an accurate Spatial-temporal assessment of complex microbial colonies
involved in metabolic and nutrient exchange processes [45]. Therefore, MSI-based
technologies enhance metabolomics assessments of microorganisms and open up
new perspectives for discovering new functional chemical constituents and antibiot-
ics from these natural sources. Figure 5.2 shows a spatial mapping methodology
applied to detect biologically active agents from microbial colonies.
Advances in MSI technologies offer an excellent opportunity for metabolomics
studies to investigate molecular interactions in biological samples and identify
metabolites involved in these interactions. Numerous versatile and sensitive tech-
niques are employed for this purpose, but they require high instrumental costs.
Recent advances have been achieved using ambient ionization mass spectrometry
(AIMS) techniques such as paper spray ionization (PSI), a simple technique that
provides satisfactory results for mass spectrometry-based metabolomics. In the next
section, we will address some of the modern techniques applied to obtain metabo-
lomic profiles of natural microbial products such as ion mobility spectrometry
(IMS), Paper Spray Ionization (PSI), Probe Electrospray Ionization (PESI) and
direct analysis in real time mass spectrometry (DART-MS).
112 N. M. Lima et al.

5.4 Ion Mobility Spectrometry (IMS) and Other Ambient

Ionization Techniques Applied to Metabolomic Profiles
of Microbial Natural Products

The molecular diversity of biological matrices such as fungal and bacterial samples
is a significant challenge for the natural products chemistry community since the
available analytical tools cannot obtain a global profile of the metabolites involved
in these biological systems. Considering that the microbial metabolome has been
little explored chemically and that the metabolic changes resulting from the biologi-
cal processes that microorganisms undergo produce molecules with distinct physi-
cochemical properties, there is a constant need for the development of more sensitive
and precise techniques with faster scanning rates to obtain more comprehensive
metabolomic profiles [12, 68].
In this context, many technologies involving mass spectrometry have been devel-
oped to address the deficiency in the evaluation of complex samples. Considering
the molecular complexity of microbial models, emerging tools such as ion mobility
spectrometry (IMS) have stood in this field, whose technique separates ions based
on hydrodynamic radius and charge [31]. IMS is a fast separation technique (milli-
second scale), compatible with chromatography, and is capable of separating bioac-
tive isobars and stereoisomers, thereby fundamental in the analysis of biological
samples [12, 41, 55, 90].
IMS is a key tool in microbial metabolomics investigations, once it is able to
discriminate between structurally similar substances and isomer pairs. In addition,
it can improve the quality of the mass spectra and uses collision cross-section (CCS)
values to provide structural information of metabolites. In addition, it improves the
signal-to-noise ratio (S/N) and the detection of minority substances, offering more
excellent metabolite coverage. In natural products, besides separating isomers and
isobars, the technique features the flexibility of operational modes and provides
greater confidence in the characterization of metabolites of natural origin [55].
Therefore, many are applications of this technique in microbial natural products
chemistry [41, 55, 90].
Many researchers report the application of IMS in natural product research. Both
ESI and MALDI ionization sources have been used in combination with traveling-­
wave IMS (TWIMS-MS) to the characterization of different natural products,
including halogenated metabolites from three species of Lyngbya, and to check the
distribution of these metabolites in microbial filaments [25, 54, 55].
Considering that microbial natural product samples constitute complex mixtures
containing hundreds of metabolites with very diverse molecular characteristics, the
coupling of IMS with chromatographic techniques has met many needs in microbial
metabolomics. Promising reports of the application of IMS coupled with RP-UHPLC
for the analysis of polyphenol mixtures are described. Thus, advances in connecting
5 Advances in Mass Spectrometry-Metabolomics Based Approaches 113

the IMS technique with separation techniques have resulted in more comprehensive
metabolomics evaluations of microbial cultures and directed studies toward detect-
ing bioactive substances [12, 37].
Many advances in metabolomics profiling of microorganisms using MS have
been achieved in recent years. Ambient Ionization Mass Spectrometry (AIMS) tech-
niques are widely used since they allow direct and rapid analysis of biological sam-
ples, resulting in high sensitivity and selectivity data, enabling detection and
quantification studies of natural microbial products [33, 51]. The direct detection of
metabolites from microbial extracts and colonies without the sample preparation
required by LC-MS makes AIMS a promising strategy in mass spectrometry-­
metabolomics based approaches. Hence, Paper Spray Ionization (PSI) and Probe
Electrospray Ionization (PESI) have been successfully applied to identify microor-
ganisms and obtain the microbial metabolome [33]. Some notable examples of the
application of PSI in microbial metabolomics include the investigation of the
metabolome of eight species of Candida fungi spread on filter paper and the dis-
crimination of the species using multivariate analysis of the spectral data obtained
by PSI-MS [33]. However, most studies reporting PSI’s application in microbial
sample analysis are directed to obtain the cellular unit fingerprint, whose experi-
ments are fast and do not require complex extraction protocols. However, the study
of the extracellular metabolome is still challenging [13].
An advantage of this technique is the possibility of analyzing low polarity sub-
stances by paper spray chemical ionization (PSCI) using the corona discharge phe-
nomenon. This versatility of the method allows the qualitative and quantitative
evaluation of low polarity aromatic compounds at concentrations up to femtomolar
[44]. In both methodologies, the colony of microorganisms is spread on a filter
paper and moistened with solvent for analysis in the mass spectrometer. Chemometric
analysis is employed to discriminate the ions associated with the target markers. An
example is the discrimination of Gram-positive and Gram-negative bacteria by
PSI-MS performed by Hamid et al. [32], whose results showed success rates close
to 90%. In addition, the technique allows for in vivo evaluations.
Direct analysis in real-time mass spectrometry (DART-MS) is another ambient
ionization mass spectrometry technique that has been used in the analysis of micro-
bial strains. Cody [15] used direct analysis in real-time time-of-flight mass spec-
trometry to evaluate the metabolic profiles of 17 strains of Saccharomyces. The
study has distinguished between yeast strains with 97% accuracy based on their
metabolic differences. Furthermore, the authors discuss the application of these
results to biotechnological products from fermentative processes [15].
The application of the aforementioned techniques has contributed significantly
to obtaining microbial metabolic profiles and directed the search for new substances
with pharmaceutical and biotechnological potential. However, to avoid the re-­
isolation of known substances and rationalize the discovery of new biologically
active agents, the natural products community uses a strategy named dereplication.
114 N. M. Lima et al.

5.5 Dereplication of Microbial Natural Products

Through MS

Derreplication is a strategy used to detect novel substances and identify known

compounds through chemical screening of microbial extracts and cultures, which
was introduced in the early 1990s. This crucial process in metabolomics studies
enables the discovery of potential drugs since it discriminates from complex sam-
ples those substances with chemical structures already described from those with
novel molecular characteristics and, therefore, avoids the re-isolation of substances
and reduces the cost and time of analysis [57].
To perform this strategy, it is necessary to obtain fingerprints of the natural prod-
uct samples using spectroscopic/spectrometric techniques such as MS and NMR
combined with chromatographic (LC, GC) or electrophoretic (CE) separation tech-
niques, and the spectral data obtained are compared with online databases and in
silico identification tools [59]. The combination of different analytical instrumenta-
tion techniques used to analyze the metabolome of microorganisms, either from
metabolites of the endometabolome (metabolic fingerprint) analysis or from metab-
olites extracted from the exometabolome (metabolic footprint). However, the meta-
bolic footprint is the least employed analysis because it does not provide information
on the changes that occur intracellularly and evaluates a limited amount of chemical
constituents [27].
For these metabolite profiles, liquid chromatography coupled to mass spectrom-
etry (LC-MS) is the technique of choice for organic chemists because it requires
shorter analysis times, a smaller amount of sample, and offers broad metabolite
coverage. Spectral and chromatographic data from samples are compared with
structural and analytical information from commercial standards or isolated com-
pounds such as exact mass, isotopic patterns, fragmentation profiles, UV spectra,
retention times, and evaluation of spectral similarity with data from compounds
deposited in MS/MS libraries. However, due to the molecular diversity and the scar-
city of spectral data on microbial metabolites, dereplication of natural product
extracts becomes a challenge and offers the opportunity for isolation of many novel
constituents [7, 69, 84].
Considering that the dereplication of microbial extracts is mainly based on the
fragmentation profile of MS/MS spectra of the substances, techniques that provide
only MS1 data are rarely used in this methodology. In untargeted metabolomics, the
individual analysis of spectra (MS1 and MS/MS) of each substance becomes infea-
sible, so computational metabolome mining tools are often employed to compare
the spectral data of substances present in the samples with chemical structures of
compounds deposited in integrated or online databases, which evaluate spectral
similarity and building blocks of molecules [12, 69, 84].
With the evolution in the development of more sophisticated analytical tools based
on orthogonal techniques, dereplication workflows are combined with natural product
databases such as SciFinder, AntiMarin, MarinLit, ChemSpider, Antibase. However,
there are still many challenges associated with the mining and interpretation of
5 Advances in Mass Spectrometry-Metabolomics Based Approaches 115

spectral and chromatographic data from microbial small molecules due to the scarcity
of information in the available bioinformatics platforms [7].
Current advances in the dereplication of small molecules of microbial origin are
a consequence of the development of new chemoinformatics tools used to extract
and evaluate large datasets. Despite the challenges associated with identifying the
molecular structures of compounds present in complex microbial samples, the
development of tools that allow extraction, annotation, biological contextualization,
and integration with omics data has resulted in significant advances in microbial
metabolomics [21]. These dereplication tools use algorithms that match the spectra
of the substances present in the sample with chemical structures present in spectral
libraries and publicly available molecular structure databases such as KEGG and
PubChem, which evaluate spectral similarity and provide candidates. Although the
entire structure of metabolites is still challenging, knowledge of the substructure,
presence of particular molecular scaffolds, and a chemical class of compounds from
complex metabolite mixtures is an efficient alternative in dereplication strategies, as
it allows to perform chemical annotation propagation [7].
Thus, although not directly obtaining the chemical structure of the compound
present in the sample, it is possible to get important structural information such as
functional groups, building blocks, biosynthetic origin, structural similarity with
known metabolites or shared substructures between substances, and thereby employ
this information in the dereplication of samples [9]. Due to this limitation of spec-
tral libraries, in silico annotation tools emerge as a promising alternative in the
dereplication of natural products. However, the level of annotation and precision of
compounds is not the same as that of spectral libraries because they correspond to
computational analyses that do not consider important fragmentation factors such as
rearrangements [56].
The in silico annotation tools use combinatorial fragmentation methodologies
and retrieve candidate structures and the competitive fragmentation modeling to
explain the fragments generated in the MS/MS spectrum of the compound evalu-
ated. Classic examples of machine learning-based methods that employ this meth-
odology include MetFrag and CFM-ID, which disconnect bonds from known
metabolites and use the obtained substructures to present molecules for that MS/MS
spectrum. Other sophisticated in silico analytical tools include SIRIUS and
ZODIAC, which employ MS/MS fragmentation profiling based on fragmentation
trees for spectra obtained using structure data of known compounds and retrieve the
structure of ranked candidates [10, 82].
Natural product dereplication has significantly grown in recent years due to the
increased volume of data deposited in spectral repositories such as PubChem,
METLIN, NIST, GOLM, BMRB, Super Natural II, and the NuBBEDB [10, 82].
Undoubtedly, the most famous worldwide and freely accessible platform that uses
information obtained from metabolite or tissue analysis utilizing the mass spec-
trometry technique is GNPS (Global Natural Products Social Molecular
Networking).
GNPS is a versatile, robust, freely accessible, and widely used online platform
for natural product chemists to investigate metabolomics and proteomics approaches.
116 N. M. Lima et al.

Although it was developed for proteomic analysis, the platform has been applied
mainly to store and process mass spectrometry datasets of metabolites from natural
origin, especially microorganisms, marine organisms, and plants [24, 65]. The
dereplication of natural product extracts for the discovery of new pharmaceutical
ingredients through molecular network analysis and structural annotation of metab-
olites and their analogs are the main objectives of the GNPS community. The meth-
odologies employed for chemical annotation are based on spectral similarity of
experimental MS/MS spectra to reference libraries. In this process, ions with simi-
lar fragmentation profiles are grouped into clusters, and the cluster analysis pro-
vides structural information of chemically related compounds and their analogs
[18, 65].
Large-scale mining of the specialized metabolome using dereplication platforms
such as GNPS provides results of possible chemical structures of compounds depos-
ited by other users and similar substances that contribute to the propagation of the
chemical annotation. Thus, aiming to make the confidence of these annotations
accurate, the Metabolomics Standards Initiative (MSI) and the Schymanski system
have presented levels of annotation confidence [71, 78]. The highest level of confi-
dence, known as annotation validation, indicates structure confirmation using data
from an analytical reference standard that associates mass spectrometry data (MS1,
MSn) and chromatographic data such as retention time and UV spectra obtained by
hyphenated analytical techniques. Whereas the other levels of annotation involve
only hits of possible molecular structures based on exact mass data, spectral infor-
mation (adduct patterns, isotope, and fragmentation pattern), do not enable discrim-
ination of positional isomers, but lack evidence information of the proposed
structure [71, 78]. Evaluating parameters that associate spectral similarity and bio-
synthetic pathways of chemically related metabolites is important because metabo-
lites from the same natural source may share common substructures, the same
building blocks, or belong to the same biosynthetic pathway [9].
Currently, many online dereplication platforms also allow gas chromatography-­
mass spectrometry (GC-MS) data analysis. GC-MS/MS data using electron ioniza-
tion (EI) sources are deconvoluted and checked against public and commercial
spectral libraries [2, 43, 85]. However, in this technique, it is not possible to obtain
molecular ion information, but other parameters such as Kovats index and retention
time contribute to the dereplication of the substances [2, 43, 83, 85].

Fig. 5.3 Replicating secondary metabolites from microbial extracts using hyphenated mass spec-
trometry techniques and database search
5 Advances in Mass Spectrometry-Metabolomics Based Approaches 117

An important step in dereplication studies is data processing, which includes

data filtering, normalization, and alignment [27]. All of these steps are crucial to
analyze the specialized metabolome comprehensively.
Figure 5.3 shows the main experimental steps of microbial samples’ dereplica-
tion process using mass spectrometry and hyphenated techniques.
Overall, the studies presented here show the advances and new spectrometric
approaches for microbial metabolomics.

5.6 Concluding Remarks

This chapter summarized recent systematic literature involving advances in mass

spectrometry-metabolomics-based approaches. Different MS techniques have been
contemplated, and the main analytical methodologies, benefits, limitations, and
advances were also demonstrated in some studies. Current advances in the detection
and quantification of bioactive metabolites of microbial origin by mass spectrome-
try attest that MS is an indispensable tool in microbial metabolomics. The main
benefits of the MS progress are the absence of sample preparation protocols and
faster and more direct screening of metabolites from microbial colonies using ambi-
ent ionization methods. In addition, the development of specialized databases and in
silico dereplication tools associated with machine learning has accelerated the pro-
cess of discovering new biologically active molecules clinically useful for human
health. Metabolomics has emerged as a novel approach in microbiome studies, and
the development and coupling of MS with new technologies dramatically expand its
applications.

Acknowledgments The authors aknowledge Coordenação de Aperfeiçoamento de Nível

Superior - CAPES, Conselho Nacional de Desenvolvimento Científico e Tecnológico - CNPq for
fellowships and financial support.
Conflict of Interest None to declare.

References

1. Alonso A, Marsal S, Julià A (2015) Analytical methods in untargeted metabolomics: state of

the art in 2015. Front Bioeng Biotech 3:23
2. Aksenov AA, da Silva R, Knight R, Lopes NP, Dorrestein PC (2017) Global chemical analysis
of biology by mass spectrometry. Nat Rev Chem 1(7):1–20
3. Alexandrov T (2020) Spatial metabolomics and imaging mass spectrometry in the age of arti-
ficial intelligence. Ann Rev Biomed Data Sci 3:61–87
4. Bader CD, Haack PA, Panter F, Krug D, Müller R (2021) Expanding the scope of detectable
microbial natural products by complementary analytical methods and cultivation systems. J
Nat Prod 84(2):268–277
5. Baidoo EE, Benke PI, Keasling JD (2012) Mass spectrometry-based microbial metabolomics.
In: Microbial systems biology. Humana Press, Totowa, pp 215–278
118 N. M. Lima et al.

6. Balog J, Kumar S, Alexander J, Golf O, Huang J, Wiggins T, Abbassi-Ghadi N, Enyedi A,

Kacska S, Kinross J, Hanna GB, Nicholson JK, Takats Z (2015) In vivo endoscopic tissue
identification by rapid evaporative ionization mass spectrometry (REIMS). Angew Chem
127(38):11211–11214
7. Bauermeister A, Mannochio-Russo H, Costa-Lotufo LV, Jarmusch AK, Dorrestein PC (2022)
Mass spectrometry-based metabolomics in microbiome investigations. Nat Rev Microbiol
20:143–160
8. Bean HD, Dimandja JMD, Hill JE (2012) Bacterial volatile discovery using solid phase micro-
extraction and comprehensive two-dimensional gas chromatography–time-of-flight mass
spectrometry. J Chromatogr B 901:41–46
9. Beniddir MA, Kang KB, Genta-Jouve G, Huber F, Rogers S, Van Der Hooft JJ (2021) Advances
in decomposing complex metabolite mixtures using substructure-and network-based computa-
tional metabolomics approaches. Nat Prod Rep 38(11):1967–1993
10. Blaženović I, Kind T, Ji J, Fiehn O (2018) Software tools and approaches for compound iden-
tification of LC-MS/MS data in metabolomics. Metabolites 8(2):31
11. Borges RM, Resende JVM (2021) (DES) Construindo a metabolômica em produtos naturais:
Um convite a discussão. Química Nova 44:1392–1394
12. Bouslimani A, Sanchez LM, Garg N, Dorrestein PC (2014) Mass spectrometry of natural prod-
ucts: current, emerging and future technologies. Nat Prod Rep 31(6):718–729
13. Chamberlain CA, Hatch M, Garrett TJ (2021) Extracellular vesicle analysis by paper spray
ionization mass spectrometry. Metabolites 11(5):308
14. Chen Q, Wang L, Qi Y, Ma C (2020) Imaging mass spectrometry of interspecies metabolic
exchange revealed the allelopathic interaction between Microcystis aeruginosa and its antago-
nist. Chemosphere 259:127430
15. Cody RB (2020) Saccharomyces cerevisiae and S. pastorianus species and strain differentia-
tion by direct analysis in real time time-of-flight mass spectrometry. Rapid Commun Mass
Spectrom 34(17):e8835
16. Da Silva RR, Wang M, Nothias LF, van der Hooft JJJ, Caraballo-Rodríguez AM, Fox E,
Balunas MJ, Klassen JL, Lopes NP, Dorrestein PC (2018) Propagating annotations of molecu-
lar networks using in silico fragmentation. PLoS Comput Biol 14(4):1–26
17. Demain AL (2014) Importance of microbial natural products and the need to revitalize their
discovery. J Ind Microbiol Biotechnol 41(2):185–201
18. Demarque DP, Dusi RG, de Sousa FD, Grossi SM, Silvério MR, Lopes NP, Espindola LS
(2020) Mass spectrometry-based metabolomics approach in the isolation of bioactive natural
products. Sci Rep 10(1):1–9
19. Derewacz DK, Goodwin CR, McNees CR, McLean JA, Bachmann BO (2013) Antimicrobial
drug resistance affects broad changes in metabolomic phenotype in addition to secondary
metabolism. Proc Natl Acad Sci 110(6):2336–2341
20. Deutsch JM, Mandelare-Ruiz P, Yang Y, Foster G, Routhu A, Houk J et al (2022) Metabolomics
approaches to Dereplicate natural products from coral-derived bioactive bacteria. J Nat Prod
85(3):462–478
21. Dhariwal A, Chong J, Habib S, King IL, Agellon LB, Xia J (2017) MicrobiomeAnalyst: a web-­
based tool for comprehensive statistical, visual and meta-analysis of microbiome data. Nucleic
Acids Res 45(W1):W180–W188
22. Dunham SJ, Ellis JF, Li B, Sweedler JV (2016) Mass spectrometry imaging of complex micro-
bial communities. Acc Chem Res 50(1):96–104
23. Ernst M, Silva DB, Silva RR, Vêncio RZ, Lopes NP (2014) Mass spectrometry in plant metab-
olomics strategies: from analytical platforms to data acquisition and processing. Nat Prod Rep
31(6):784–806
24. Ernst M, Kang KB, Caraballo-Rodríguez AM, Nothias LF, Wandy J, Chen C, Wang M, Rogers
S, Medema MH, Dorrestein PC, Van Der Hooft JJ (2019) MolNetEnhancer: enhanced molecu-
lar networks by integrating metabolome mining and annotation tools. Metabolites 9(7):144
25. Esquenazi E, Daly M, Bahrainwala T, Gerwick WH, Dorrestein PC (2011) Ion mobility mass
spectrometry enables the efficient detection and identification of halogenated natural products
from cyanobacteria with minimal sample preparation. Bioorg Med Chem 19(22):6639–6644
5 Advances in Mass Spectrometry-Metabolomics Based Approaches 119

26. Floros DJ, Petras D, Kapono CA, Melnik AV, Ling TJ, Knight R, Dorrestein PC (2017) Mass
spectrometry based molecular 3D-cartography of plant metabolites. Front Plant Sci 8:429
27. Gao P, Xu G (2015) Mass-spectrometry-based microbial metabolomics: recent developments
and applications. Anal Bioanal Chem 407(3):669–680
28. Geier B, Sogin EM, Michellod D, Janda M, Kompauer M, Spengler B, Dubilier N, Liebeke M
(2020) Spatial metabolomics of in situ host–microbe interactions at the micrometre scale. Nat
Microbiol 5(3):498–510
29. Genilloud O (2014) The re-emerging role of microbial natural products in antibiotic discovery.
Antonie Van Leeuwenhoek 106(1):173–188
30. Ghaste M, Mistrik R, Shulaev V (2016) Applications of fourier transform ion cyclotron reso-
nance (FT-ICR) and orbitrap based high resolution mass spectrometry in metabolomics and
lipidomics. Int J Mol Sci 17(6):816
31. Goodwin CR, Fenn LS, Derewacz DK, Bachmann BO, McLean JA (2012) Structural mass
spectrometry: rapid methods for separation and analysis of peptide natural products. J Nat
Prod 75(1):48–53
32. Hamid AM, Jarmusch AK, Pirro V, Pincus DH, Clay BG, Gervasi G, Cooks RG (2014) Rapid
discrimination of bacteria by paper spray mass spectrometry. Anal Chem 86(15):7500–7507
33. Hamid AM, Wei P, Jarmusch AK, Pirro V, Cooks RG (2015) Discrimination of Candida spe-
cies by paper spray mass spectrometry. Int J Mass Spect 378:288–293
34. Hirayama A, Wakayama M, Soga T (2014) Metabolome analysis based on capillary
electrophoresis-­mass spectrometry. TrAC Trends Anal Chem 61:215–222
35. Hsu CC, ElNaggar MS, Peng Y, Fang J, Sanchez LM, Mascuch SJ, Møller KA, Alazzeh EK,
Pikula J, Quinn RA, Zeng Y, Wolfe BE, Duttonγ RJ, Gerwick L, Zhang L, Liu X, Månsson
M, Dorrestein PC (2013) Real-time metabolomics on living microorganisms using ambient
electrospray ionization flow-probe. Anal Chem 85(15):7014–7018
36. Ibáñez AJ, Fagerer SR, Schmidt AM, Urban PL, Jefimovs K, Geiger P, Dechant R, Heinemann
M, Zenobi R (2013) Mass spectrometry-based metabolomics of single yeast cells. Proc Natl
Acad Sci 110(22):8790–8794
37. Jarmusch AK, Cooks RG (2014) Emerging capabilities of mass spectrometry for natural prod-
ucts. Nat Prod Rep 31(6):730–738
38. Jarmusch SA, van der Hooft JJ, Dorrestein PC, Jarmusch AK (2021) Advancements in captur-
ing and mining mass spectrometry data are transforming natural products research. Nat Prod
Rep 38(11):2066–2082
39. Jorge TF, Rodrigues JA, Caldana C, Schmidt R, van Dongen JT, Thomas-Oates J, António C
(2016) Mass spectrometry-based plant metabolomics: metabolite responses to abiotic stress.
Mass Spectrom Rev 35(5):620–649
40. Kang KB, Ernst M, van der Hooft JJ, da Silva RR, Park J, Medema MH, Sung SH, Dorrestein
PC (2019) Comprehensive mass spectrometry-guided phenotyping of plant specialized
metabolites reveals metabolic diversity in the cosmopolitan plant family Rhamnaceae. Plant J
98(6):1134–1144
41. Kaufmann A, Butcher P, Maden K, Walker S, Widmer M (2020) Does the ion mobility resolv-
ing power as provided by commercially available ion mobility quadrupole time-of-flight mass
spectrometry instruments permit the unambiguous identification of small molecules in com-
plex matrices? Anal Chim Acta 1107:113–126
42. Kellogg JJ, Todd DA, Egan JM, Raja HA, Oberlies NH, Kvalheim OM, Cech NB (2016)
Biochemometrics for natural products research: comparison of data analysis approaches and
application to identification of bioactive compounds. J Nat Prod 79(2):376–386
43. Keppler EAH, Jenkins CL, Davis TJ, Bean HD (2018) Advances in the application of com-
prehensive two-dimensional gas chromatography in metabolomics. TrAC Trends Anal Chem
109:275–286
44. Kim D, Yim UH, Kim B, Cha S, Kim S (2017) Paper spray chemical ionization: highly sen-
sitive ambient ionization method for low-and nonpolar aromatic compounds. Anal Chem
89(17):9056–9061
120 N. M. Lima et al.

45. Lanni EJ, Masyuko RN, Driscoll CM, Aerts JT, Shrout JD, Bohn PW, Sweedler JV (2014)
MALDI-guided SIMS: multiscale imaging of metabolites in bacterial biofilms. Anal Chem
86(18):9139–9145
46. Lei Z, Huhman DV, Sumner LW (2011) Mass spectrometry strategies in metabolomics. J Biol
Chem 286(29):25435–25442
47. Li H, Balan P, Vertes A (2016) Molecular imaging of growth, metabolism, and antibiotic inhi-
bition in bacterial colonies by laser ablation electrospray ionization mass spectrometry. Angew
Chem 128(48):15259–15263
48. Liebeke M, Strittmatter N, Fearn S, Morgan AJ, Kille P, Fuchser J, Wallis D, Palchykov V,
Robertson J, Lahive E, Spurgeon DJ, McPhail D, Takáts Z, Bundy JG (2015) Unique metabo-
lites protect earthworms against plant polyphenols. Nat Commun 6(1):1–7
49. Lima GS, Dos Santos GF, Ramalho RRF, de Aguiar DVA, Roque JV, Maciel LIL, Simas RC,
Pereira I, Vaz BG (2022) Laser ablation electrospray ionization mass spectrometry imaging
as a new tool for accessing patulin diffusion in mold-infected fruits. Food Chem 373:131490
50. Lima NM, Lima GS, dos Santos GF, Preet G, Maciel LIL, Andrade TJAS, Jaspars M,
Chaves AR, Vaz BG (2023) Assessing the effectiveness of chemical marker extraction from
Amazonian plant Cupuassu (Theobroma grandiflorum) by PSI-HRMS/MS and LC-HRMS/
MS. Metabolites 13:367
51. Ling C, Shi Q, Wei Z, Zhang J, Hu J, Pei J (2022) Rapid analysis of quinones in complex
matrices by derivatization-based wooden-tip electrospray ionization mass spectrometry.
Talanta 237:122912
52. Lukowski JK, Bhattacharjee A, Yannarell SM, Schwarz K, Shor LM, Shank EA, Anderton CR
(2021) Expanding molecular coverage in mass spectrometry imaging of microbial systems
using metal-assisted laser desorption/ionization. Microbiol Spectr 9(1):e00520-21
53. Luzzatto-Knaan T, Melnik AV, Dorrestein PC (2015) Mass spectrometry tools and workflows
for revealing microbial chemistry. Analyst 140(15):4949–4966
54. Marshall AP, Johnson AR, Vega MM, Thomson RJ, Carlson EE (2020) Ion mobility mass
spectrometry as an efficient tool for identification of Streptorubin B in Streptomyces coelicolor
M145. J Nat Prod 83(1):159–163
55. Masike K, Stander MA, de Villiers A (2021) Recent applications of ion mobility spectrometry
in natural product research. J Pharm Biomed Anal 195:113846
56. Medema MH, Fischbach MA (2015) Computational approaches to natural product discovery.
Nat Chem Biol 11(9):639–648
57. Middlesworth FV, Cannell RJ (1998) Dereplication and partial identification of natural prod-
ucts. In: Natural products isolation. Humana Press, pp 279–327
58. Misra BB (2020) The connection and disconnection between microbiome and metabolome: a
critical appraisal in clinical research. Biol Res Nurs 22(4):561–576
59. Mohimani H, Gurevich A, Shlemov A, Mikheenko A, Korobeynikov A, Cao L, Shcherbin
E, Nothias L-F, Dorrestein PC, Pevzner PA (2018) Dereplication of microbial metabolites
through database search of mass spectra. Nat Commun 9(1):1–12
60. Monton MRN, Soga T (2007) Metabolome analysis by capillary electrophoresis–mass spec-
trometry. J Chromatogr A 1168(1–2):237–246
61. Moore JL, Caprioli RM, Skaar EP (2014) Advanced mass spectrometry technologies for the
study of microbial pathogenesis. Curr Opin Microbiol 19:45–51
62. Nguyen DD, Saharuka V, Kovalev V, Stuart L, Del Prete M, Lubowiecka K, De Mot R, Venturi
V, Alexandrov T (2021) Facilitating imaging mass spectrometry of microbial specialized
metabolites with METASPACE. Metabolites 11(8):477
63. Panter F, Bader CD, Müller R (2021) Synergizing the potential of bacterial genomics and
metabolomics to find novel antibiotics. Chem Sci 12(17):5994–6010
64. Patil SG, Patil MP, Maheshwari VL, Patil RH (2021) In situ probing of endophyte natural
products with DESI-imaging mass spectrometry. In: Endophytes, pp 177–193. https://doi.
org/10.1007/978-­981-­15-­9371-­0_9
65. Pilon AC, Vieira NC, Amaral JG, Monteiro AF, Silva RRD, Spíndola LS, Gamboa IC, Lopes
NP (2021) Redes moleculares: uma análise sobre anotações e descoberta de novos ativos.
Química Nova 44(9):1168–1179
5 Advances in Mass Spectrometry-Metabolomics Based Approaches 121

66. Prentice BM, Caprioli RM (2016) The need for speed in matrix-assisted laser desorption/
ionization imaging mass spectrometry. Postdoc J 4(3):3
67. Quinn RA, Melnik AV, Vrbanac A, Fu T, Patras KA, Christy MP, Bodai Z, Belda-Ferre P,
Tripathi A, Chung LK, Downes M, Welch RD, Quinn M, Humphrey G, Panitchpakdi M,
Weldon KC, Aksenov A, da Silva R, Avila-Pacheco J, Clish C, Bae S, Mallick H, Franzosa EA,
Lloyd-Price J, Bussell R, Thron T, Nelson AT, Wang M, Leszczynski E, Vargas F, Gauglitz JM,
Meehan MJ, Gentry E, Arthur TD, Komor AC, Poulsen O, Boland BS, Chang JT, Sandborn
WJ, Lim M, Garg N, Lumeng JC, Xavier RJ, Kazmierczak BI, Jain R, Egan M, Rhee KE,
Ferguson D, Raffatellu M, Vlamakis H, Haddad GG, Siegel D, Huttenhower C, Mazmanian
SK, Evans RM, Nizet V, Knight R, Dorrestein PC (2020) Global chemical effects of the micro-
biome include new bile-acid conjugations. Nature 579(7797):123–129
68. Rath CM, Yang JY, Alexandrov T, Dorrestein PC (2013) Data-independent microbial metabo-
lomics with ambient ionization mass spectrometry. J Am Soc Mass Spectrom 24(8):1167–1176
69. Ren JL, Zhang AH, Kong L, Wang XJ (2018) Advances in mass spectrometry-based metabo-
lomics for investigation of metabolites. RSC Adv 8(40):22335–22350
70. Rinschen MM, Ivanisevic J, Giera M, Siuzdak G (2019) Identification of bioactive metabolites
using activity metabolomics. Nat Rev Mol Cell Biol 20(6):353–367
71. Schymanski EL, Jeon J, Gulde R, Fenner K, Ruff M, Singer HP, Hollender J (2014) Identifying
small molecules via high resolution mass spectrometry: communicating confidence. Environ
Sci Tech 48(4):2097–2098
72. Sebak M, Saafan AE, AbdelGhani S, Bakeer W, El-Gendy AO, Espriu LC, Duncan K, Edrada-­
Ebel R (2019) Bioassay-and metabolomics-guided screening of bioactive soil actinomycetes
from the ancient city of Ihnasia, Egypt. PLoS One 14(12):e0226959
73. Shank EA (2018) Considering the lives of microbes in microbial communities. MSystems
3(2):e00155-17
74. Shih CJ, Chen PY, Liaw CC, Lai YM, Yang YL (2014) Bringing microbial interactions to light
using imaging mass spectrometry. Nat Prod Rep 31(6):739–755
75. Škrášková K, Claude E, Jones EA, Towers M, Ellis SR, Heeren RM (2016) Enhanced capabili-
ties for imaging gangliosides in murine brain with matrix-assisted laser desorption/ionization
and desorption electrospray ionization mass spectrometry coupled to ion mobility separation.
Methods 104:69–78
76. Spraker JE, Luu GT, Sanchez LM (2020) Imaging mass spectrometry for natural products
discovery: a review of ionization methods. Nat Prod Rep 37(2):150–162
77. Strittmatter N, Jones EA, Veselkov KA, Rebec M, Bundy JG, Takats Z (2013) Analysis
of intact bacteria using rapid evaporative ionisation mass spectrometry. Chem Commun
49(55):6188–6190
78. Sumner, L. W., Amberg, A., Barrett, D., Beale, M., Beger, R., Daykin, C., -M.Fan, T., Fiehn,
O., Goodacre, R., Griffin, J., Hankemeir, T., Hardy, N., Harnly, J.M., Higashi, R., Kopka, J.,
Lane, A., Lindon, J., Marriott, P., Nicholls, A., Reily, M., Thaden, J., Viant, M., 2007. Proposed
minimum reporting standards for chemical analysis. Metabolomics 3, 211–221. https://doi.
org/10.1007/s11306-007-0082-2
79. Tabang DN, Ford M, Li L (2021) Recent advances in mass spectrometry-based Glycomic and
Glycoproteomic studies of pancreatic diseases. Front Chem 9:707387
80. Tian H, Six DA, Krucker T, Leeds JA, Winograd N (2017) Subcellular chemical imaging
of antibiotics in single bacteria using C60-secondary ion mass spectrometry. Anal Chem
89(9):5050–5057
81. Utermann C, Echelmeyer VA, Blümel M, Tasdemir D (2020) Culture-dependent microbiome
of the Ciona intestinalis tunic: isolation, bioactivity profiling and untargeted metabolomics.
Microorganisms 8(11):1732
82. Valli M, Russo HM, Pilon AC, Pinto MEF, Dias NB, Freire RT, Castro-Gamboa I, Bolzani
VS (2019) Computational methods for NMR and MS for structure elucidation II: database
resources and advanced methods. Phys Sci Rev 4(11):20180108
83. Viegas MC, Bassoli DG (2007) Utilização do índice de retenção linear para caracterização
de compostos voláteis em café solúvel utilizando GC-MS e coluna HP-Innowax. Quím Nova
30(8):2031–2034
122 N. M. Lima et al.

84. Wandy J, Davies V, van der Hooft JJJ, Weidt S, Daly R, Rogers S (2019) In silico optimization
of mass spectrometry fragmentation strategies in metabolomics. Metabolites 9(10):219
85. Wang M, Carver JJ, Phelan VV, Sanchez LM, Garg N, Peng Y, Nguyen DD, Watrous J, Kapono
CA, Luzzatto-Knaan T, Porto C, Bouslimani A, Melnik AV, Meehan MJ, Liu WT, Crüsemann
M, Boudreau PD, Esquenazi E, Sandoval-Calderón M, Kersten RD, Pace LA, Quinn RA,
Duncan KR, Hsu CC, Floros DJ, Gavilan RG, Kleigrewe K, Northern T, Dutton RJ, Parrot D,
Carlson EE, Aigle B, Michelsen CF, Jelsbak L, Sohlenkamp C, Pevzner P, Edlund A, McLean
J, Piel J, Murphy BT, Gerwick L, Liaw CC, Yang YL, Humpf HU, Maansson M, Keyzers
RA, Sims AC, Johnson AR, Sidebottom AM, Sedio BE, Klitgaard A, Larson CB, Boya PCA,
Torres-Mendoza D, Gonzalez DJ, Silva DB, Marques LM, Demarque DP, Pociute E, O’Neill
EC, Briand E, Helfrich EJN, Granatosky EA, Glukhov E, Ryffel F, Houson H, Mohimani H,
Kharbush JJ, Zeng Y, Vorholt JA, Kurita KL, Charusanti P, McPhail KL, Nielsen KF, Vuong
L, Elfeki M, Traxler MF, Engene N, Koyama N, Vining OB, Baric R, Silva RR, Mascuch SJ,
Tomasi S, Jenkins S, Macherla V, Hoffman T, Agarwal V, Williams PG, Dai J, Neupane R,
Gurr J, Rodríguez AMC, Lamsa A, Zhang C, Dorrestein K, Duggan BM, Almaliti J, Allard
PM, Phapale P, Nothias LF, Alexandrov T, Litaudon M, Wolfender JL, Kyle JE, Metz TO,
Peryea T, Nguyen DT, VanLeer D, Shinn P, Jadhav A, Müller R, Waters KM, Shi W, Liu X,
Zhang L, Knight R, Jensen PR, Palsson BO, Pogliano K, Linington RG, Gutiérrez M, Lopes
NP, Gerwick WH, Moore BS, Dorrestein PC, Bandeira N (2016) Sharing and community cura-
tion of mass spectrometry data with global natural products social molecular networking. Nat
Biotechnol 34(8):828–837
86. Wang L, Lv W, Sun X, Zheng F, Xu T, Liu X, Li H, Lu X, Peng X, Hu C, Xu G (2021) Strategy
for nontargeted Metabolomic annotation and quantitation using a high-resolution spectral-­
stitching Nanoelectrospray direct-infusion mass spectrometry with data-independent acquisi-
tion. Anal Chem 93(30):10528–10537
87. Watrous JD, Dorrestein PC (2011) Imaging mass spectrometry in microbiology. Nat Rev
Microbiol 9(9):683–694
88. Watrous JD, Phelan VV, Hsu CC, Moree WJ, Duggan BM, Alexandrov T, Dorrestein PC
(2013) Microbial metabolic exchange in 3D. ISME J 7(4):770–780
89. Weisskopf L, Schulz S, Garbeva P (2021) Microbial volatile organic compounds in intra-­
kingdom and inter-kingdom interactions. Nat Rev Microbiol 19(6):391–404
90. Werres T, Leonhardt J, Jäger M, Teutenberg T (2018) Critical comparison of liquid chroma-
tography coupled to mass spectrometry and three different ion mobility spectrometry systems
on their separation capability for small isomeric compounds. Chromatographia 82(1):251–260
91. Wolfender JL, Nuzillard JM, Van Der Hooft JJ, Renault JH, Bertrand S (2018) Accelerating
metabolite identification in natural product research: toward an ideal combination of liquid
chromatography–high-resolution tandem mass spectrometry and NMR profiling, in silico
databases, and chemometrics. Anal Chem 91(1):704–742
92. Yang JY, Phelan VV, Simkovsky R, Watrous JD, Trial RM, Fleming TC, Wenter R, Moore BS,
Golden SS, Pogliano K, Dorrestein PC (2012) Primer on agar-based microbial imaging mass
spectrometry. J Bacteriol 194(22):6023–6028
93. Yang H, Goodlett DR, Ernst RK, Scott AJ (2020) Mass spectrometry imaging of microbes.
Mass Spectrom Lett 11(3):41–51
94. Yao F, Yi B, Shen C, Tao F, Liu Y, Lin Z, Xu P (2015) Chemical analysis of the Chinese liquor
Luzhoulaojiao by comprehensive two-dimensional gas chromatography/time-of-flight mass
spectrometry. Sci Rep 5(1):1–6
95. Zhou B, Xiao JF, Tuli L, Ressom HW (2012) LC-MS-based metabolomics. Mol BioSyst
8(2):470–481
Chapter 6
Advances in Microbial NMR Metabolomics

Ricardo Moreira Borges, Gonçalo Jorge Gouveia,

and Fernanda Oliveira das Chagas

6.1 NMR Fundamentals

Among the repertoire of analytical techniques for structure identification of organic

chemical compounds, the Nuclear Magnetic Resonance (NMR) is deemed the most
informative, providing atomic-level details about the chemical structure and envi-
ronment of nuclei and molecules. Since NMR detects magnetically active nuclei,
which is a consequence of a processing property called ‘spin’, and each signal
shows information about the respective chemical environment to the spin that pro-
duces it, an NMR spectrum is characteristic of a single compound as a fingerprint.
The simplest NMR experiment yields the chemical shifts (according to the local
electronic distribution in a molecule) of resonances and their respective multiplici-
ties (according to the quantity and relative spatial location of neighboring spins). In
metabolomics, most of the analysis is done focusing on the detection of 1H and 13C
(less commonly: 15N and 31P) due to their abundance in nature. Virtually all organic
compounds contains hydrogen atoms and the natural abundance of its magnetic-­
active isotope is 99.98%, building up an advantageous case for 1H NMR sensitivity
in comparison to 13C NMR.
Briefly, NMR is a spectroscopy technique that relies on radiofrequency wave-
lengths radiation pulses and nuclear magnetization, and it depends on an external
magnet to align the spins of a sample to an equilibrium before excitation and detec-
tion. The importance of this alignment of spins is such that the power of the external

R. M. Borges (*) · F. O. das Chagas

Instituto de Pesquisas de Produtos Naturais Walter Mors,
Universidade Federal do Rio de Janeiro, Rio de Janeiro, Brazil
G. J. Gouveia
Department of Biochemistry and Molecular Biology, University of Georgia,
Athens, GA, USA

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 123
T. Pacheco Fill (ed.), Microbial Natural Products Chemistry, Advances in Experimental
Medicine and Biology 1439, https://doi.org/10.1007/978-3-031-41741-2_6
124 R. M. Borges et al.

Fig. 6.1 Key NMR fundamental representations. (a) the splitting pattern of the degenerate nuclear
energy levels resulting from the application of an external magnetic field and the fundamental
equation expressing energy as a function of the magnetic field. (b) examples of NMR active nuclei.
(c) a representative 1H NMR of urine with the identification of phenylalanine

magnet (measured in Tesla but generally referred to as the counterpart 1H frequency;

e.g.: 18.8 T or 800 MHz), which is responsible for the distribution of nuclei between
energy states according to the Boltzmann distribution, reflects the high cost of NMR
instrumentation (Fig. 6.1) [1]. Recent technical developments include, among oth-
ers, higher field strengths to increase spectrometers sensitivity and resolution.
Indeed, the intrinsic low sensitivity of NMR (mainly compared to Mass Spectrometry;
MS) is due to the small gap between the Zeeman energy sublevels (α and β states)
resulting in a small excess populating the lower energy state (conventionally called
α), where only a few spins among hundreds of thousands are responsible for the
resulting magnetization (net magnetization). This equilibrium state must be dis-
rupted by the application of a radiofrequency field in a short period of time (RF
pulse) for signal detection. Such a signal is detected as an exponential free induced
decay (FID) once the spins complete a sequence of excitation (coherence state) and
relaxation (back to the equilibrium state). The full theoretical intricacies of the
NMR spectroscopy can be reviewed elsewhere [2], since it is beyond the scope of
this chapter. It suits to mention, though, that the net magnetization can be handled
through different quantum states to provide different information in the form of
many different experiments and these can be accessed to collect data concerning
compound identification, molecular dynamics, molecular interactions, etc.
Especially important for the discussion intended here, the indirect spin-spin interac-
tion can be accessed through analysis of scalar coupling, which occurs via electrons
in chemical bond connecting spins (J-coupling or scalar coupling). This assessment
is done by the use of a set of experiments with the goal of tackling such a puzzling
construction to enable an unequivocal chemical structure elucidation [3].
6 Advances in Microbial NMR Metabolomics 125

Concerning NMR applications within metabolomics, it is important to point out

certain aspects of mixture analysis. When dealing with a mixture of organic com-
pounds, the resulting NMR spectra will present every signal of each spin of each
chemical structure, and, more important, in intensities respective to their molar con-
centrations [4, 5]. This last feature is of utmost importance because, under confident
acquisition parameter choices, NMR is uniquely quantitative for many metabolites
using a single internal reference. Thus, the intensity of NMR signal can be inferred
as the direct concentration of that spin in the analyzed sample. In other words, the
only requirement for quantification is the presence of signal of known (and certifi-
able) intensity; ideally, this is achievable by using of a certified reference material
(CRM) [5].

6.2 Complementary Aspects of NMR to MS

An important disadvantage of NMR is its low sensitivity, which is accepted to be at

least of 3 orders of magnitude lower than MS [6]. Sensitivity is an important aspect
to be considered when dealing with microorganism due to the limited access to
biomass in laboratory scale and low extraction yield, especially of secondary
metabolites. It is often said that MS enables the detection of picomolar (10−12) con-
centration as an example of its high sensitivity, even though in routine studies, such
low concentrations of a compound is virtually impossible due to background and
matrix noise. The limit of detection (LOD) is defined as the lowest quantity of com-
pound with a certain analytical method, which includes sample preparation and
most methods yield micromolar (10−6) to nanomolar (10−9) concentrations at its
minimum [7]. The main outcome of this high sensitivity in MS, together with its
selectivity is seen in the capability of detecting thousands of features within a single
analysis. However, the required ionization for MS analysis imposes a limitation of
its quantification capabilities because different compounds have different ionization
potential, and so respond differently. Thus, one needs to use pre-determined authen-
tic samples with specific requirements such as structural similarity (e.g. 13C-labeled
derivative of the target compound for co-injection). In NMR, the single requirement
for peak intensity is the occurrence of an active spin, such as 1H that occurs in very
high abundance naturally. In principle, a common and simple standard compound
(e.g. maleic acid CRM) might be used in quantification routines when its resonance
does not overlap with other sample resonances.
MS-driven databases have much broader coverage than NMR-driven data-
bases; for instance, the GNPS [8] efforts to design a community-driven database
is enabling a progressive expansion of MS focused metabolites. It is not the goal
here to point out to detailed number, but the technical aspects of MS analysis
(relying on standards for quality control) together with new approaches for
library construction enable users to easily contribute to those databases [8].
126 R. M. Borges et al.

NMR users, mainly natural products and synthetic chemists, are generally less
inclined to provide their data to organized open-access databases (e.g.:
nmrshiftdb [9]). Fortunately, some efforts have been made to establish the
acquisition of pure compounds in a standardized manner to populate NMR-
based databases [10–12], even though they focus on primary metabolites. An
interesting alternative to tackle this limited database issue is the development of
methods for spectral simulations [13–15]. Ideally, users can simulate spectral
data (e.g: MS fragments or NMR chemical shifts) and compare to their experi-
mental data for compound annotation. The simulation can be achieved through
different approaches, from the most computer/time-consuming quantum
mechanics and machine learning to faster methods such as hierarchically
ordered spherical environment (HOSE). This last method depends highly on
structure variations in an existent chemical-shift database, but once similar
structures are found, the results are highly accurate [16].
Another comparison commonly made between MS and NMR for metabolo-
mics studies is the high cost of NMR instrumentations. Indeed, a basic NMR
setup is a lot more expensive and it occupies more space than a MS basic instru-
mentation; needless to say, even more expensive when the maintenance of cryo-
probes is considered. Once the main price limitation is overcome, either because
the users already have the instrumentation on site or because they outsource the
data acquisition, NMR data might be acquired within a remarkable lower cost-
service per sample and, certainly, each dataset will weight computationally less
(hundreds of megabits) than the equivalent MS dataset (dozens of gigabits).
Finally, the decision on which platform will be used in a metabolomics study will
be made by the users and they must balance a sensitive trade-off between advan-
tages and disadvantages of each technology and the chemical space in question.
Novel approaches, so called data-­fusion, are rising in the literature where users
acquire data from both techniques to explore the complementary aspects of each
platform [17–24]. Table 6.1 displays a brief list of NMR advantages and disadvan-
tages compared to MS.

Table 6.1 Quick list of NMR advantages and disadvantages compared to MS

Advantages Disadvantages (compared to MS)
Robust instrumentation
Very reproducible Lower sensitivity
Absolute quantification Lower metabolite coverage
Allows for accurate structure identification Limited spectral databases
Non-destructive for samplesa Requires cryogen fillings
Large concentration range Expensive (and large) instrumentation required
Automation
Throughput
Specifically important for in vivo analysis
a
6 Advances in Microbial NMR Metabolomics 127

6.3 Practical Aspects

No details about microorganism growth will be discussed here, but it is worth men-
tioning some words concerning samples size [25], technical variations and quality
control, as those are important details to be aware of in metabolomics. The number
of replicates, or sample size, to be used in a statistic-driven study such as in metabo-
lomics is important to guarantee representativeness among possible biological vari-
ations. Sample size can be estimated mathematically from a pilot-study [26].
Biological variations are expected but they should not be overtaken by technical
variations. Sample replicates are expected to be obtained by a reproducible cultiva-
tion procedure that will detail the specific strain, growth media and cultivation con-
ditions (light, temperature, aeration, etc.).
Regarding quality control samples, blank samples from the growth media, blank
samples from the extraction procedure and blank samples from analysis sequence
are considered good practices [27]. The pool-samples obtained from the combina-
tion of a small percent of each sample is also indicated for system suitability and
sample preparation quality evaluation and assurance. More on quality control can
be retrieved elsewhere [28–31].

6.4 Sample Preparation

In metabolomics, each datum represents a snapshot of a sample collection instant

and for this to be true, an effective method of quenching is needed. Liquid N2 and
cold methanol are generally used to achieve this break in enzyme metabolism [32].
This step might impose an important limitation to the procedure since it can slow
down the sample preparation workflow. In liquid media, users need to choose from
filtering the cells before quenching or to do it with the whole medium-cells solution.
Centrifugation might be used when cells tend to decant down without compromis-
ing their integrity, and so the media volume can be directly removed. This process
can be followed by lyophilization to enable mass balance before the extraction tak-
ing place. Nevertheless, sample preparation for NMR analysis is generally straight-
forward and deemed to be minimals which facilitates automation and higher sample
throughput [1, 4, 33–39, 41, 42]. However, the sample preparation planning needs
to be a part of the study design and, in many cases, it is important to include clean
up steps to ensure compounds of interest (targeted or non-targeted) to be detected
more selectively.
Zabek et al. [43] reported a method for distinguishing filamentous fungal species
that also includes centrifugation of cell pellets followed by liquid N2 quenching
before extraction with 0.05 M D2O phosphate buffer for the investigation of endo-­
metabolites [43]. On the other hand, Palama et al. [44], proposed a direct addition
of 60 μL of 1.25 M D2O phosphate buffer to 540 μL of centrifuged media at 4000 × g
for 10 min without quenching for discriminate between different bacterial species
128 R. M. Borges et al.

using their exo-metabolome [44]. They successfully evaluated 576 samples, identi-
fied 43 compounds and reported representative 1H NMR spectra for the test species.
Newton et al. [36] sought after an optimized extraction method for NMR analysis of
the unicellular protozoan Blastocystis. They established a method where they first
centrifuged (at 1000 × g for 5 min at 4 °C) each sample to exclude the excess of
liquid media and subsequently wash the cell pellets in Locke’s solution (aqueous
sodium chloride, calcium chloride, potassium chloride, sodium bicarbonate, and
glucose adjusted to pH 7.4), for then to quench the washed pellets in liquid N2.
Noteworthy, they provided strong indications that methanol extraction and bead
beating at room temperature for breaking the cell wall composes a consistent, repro-
ducible, and efficient method for metabolomics studies using Blastocystis [36].
These results can be extrapolated, albeit carefully, to other studies.
This minimal sample preparation step allows the entire extract to be sampled
together and, consequently, major compounds will be detected in their full concen-
trations. For this reason, NMR metabolomics is less frequently used in natural prod-
ucts studies as it is presumably assumed to yield results only regarding primary
metabolites. While this is the perceived consensus, because of NMR high dynamic
range these low intensity signals of interest are indeed detected, but challenging to
be analyzed due to overlap with high intensity signals. However, as demonstrated by
Beltran et al. [45], using appropriate extraction methods, sequential analysis of
extracts can provide complementary data otherwise hard to capture, where, the
chromatographic separation step generally associated with MS platforms aptly
addresses this NMR limitation. Compounds with certain chemical properties are
retained favorably for resolution in the “sweet-spot region” while others will be
discarded with the eluting solvents. For example, highly polar compounds such as
amino acids and simple sugars will elute quickly in the solvent front of a reverse-­
phase chromatography separation, while non-polar compounds will be retained to
elute together with the wash at the end of the gradient elution.
The use of deuterated solvents for NMR detection is mandatory for frequency
locking in routine use. Thus, in the majority of cases, samples will be solubilized in
deuterated high purity solvents, but first the compounds of interest will need to be
extracted from complex matrices. The cell lysis method can have a significant
impact on the number, type and levels of metabolites which in conjunction with the
extraction solvent can determine the metabolite coverage [46]. For solution NMR,
samples are generally obtained from the extraction with high purity solvents, evapo-
rated to dryness, and then, solubilized in the deuterated solvent of choice. The
choice of the extraction solvent and the NMR analysis solvent needs to take into
consideration the polarity range of the constituents. Generally, polar constituents
are extracted using 80% methanol-water and less-polar constituents, including lip-
ids, are extracted with mixtures with chloroform or isopropanol in different propor-
tions [39, 40, 47–49]. In any case, the extraction procedure must be designed to be
robust and to minimize possible metabolite degradation. Specific considerations
must be taken for the analysis of the endo- vs exo-metabolome of microorganisms
[44, 47]. Some general approaches are sketched in Fig. 6.2.
6 Advances in Microbial NMR Metabolomics 129

Fig. 6.2 Scheme for extraction of microorganism-derived samples. Three possible procedures are
shown: (1) extraction directly from the media plate; (2) extraction from the wet pool of cells after
media centrifugation; (3) extraction from the filtered cells in which both the cells and the media are
physically separated for analysis of the endo- and exo-metabolome

6.5 Data Acquisition

This topic has been extensively reviewed by other authors and will not be detailed
here [1, 6, 33, 34, 50–54]. Instead, a quick list of the most used experiments is pre-
sented (Table 6.2).

6.6 Data Processing

Data processing consists in a sequence of operations on the raw data: apodization,

Fourier transformation (FT), phasing, baseline correction, chemical shift calibra-
tion, normalization, and scaling. All these procedures, except for normalization and
scaling, are commonly applied in routine NMR analysis and they are extensively
reviewed elsewhere [33, 66, 67]. These operations have a significant impact on data
quality and should be thoroughly reported as they define in-phase sharp lines (pure
Lorentzian curves whenever possible) and a stable baseline over the spectral window.
Normalization adjustment across samples is needed to minimize unwanted
experimental or biological variances. These might be due to differential mass bal-
ance across different extracts, different number of cells collected per sample in a
study or simply due to manual sample preparation procedures [68, 69]. Different
methods were designed for normalization: area (or integral) normalization, probabi-
listic quotient normalization (PQN) [70], range normalization and reference nor-
malization are among the most commonly used. The algorithm behind area
normalization promotes a division between each data point by an equal total area of
the spectral window. The PQN method was developed using urine samples, and the
130 R. M. Borges et al.

Table 6.2 Most used NMR experiments in metabolomics and natural products for complex
mixtures analysis
NMR
Experiments Application Comments References
1
H Metabolomics Basic 1D, fast and reliable one-pulse
1
H NMR experiment
Noesypr Metabolomics Successful in water suppression Giraudeau et al.
unaffected by gradient imperfections [55]
PURGE Metabolomics Successful in water suppression, but it Simpson and
is limited by possible gradient Brown [56], Le
imperfections Guennec et al.
[57]
CPMG Metabolomics Focus on T2 filters to remove Meiboom and
unwanted broad signals from Gill [58], Le
macromolecules Guennec et al.
[57]
PROJECT Metabolomics An alternative to CPMG with lower Aguilar et al.
intensity losses [59], Le Guennec
et al. [57]
J-resolved Metabolomics + Experiment that decomposes a 1D Ferretti et al.
Compound spectra into pure resonance signals as [60], Huang et al.
Identification projections and scalar coupling [61]
splitting as crosspeaks of a 2D
spectra.
TOCSY Compound Mostly used to reduce peak Bingol et al. [11]
Identification overlapping on an orthogonal axis and
HSQC Metabolomics + use these cross-peaks as features for Bingol et al. [11],
Compound compound identification Wang et al. [39]
Identification
HSQC-TOCSY Compound Bingol et al. [11],
Identification Wang et al. [39]
HMBC Compound Kuhn et al. [62]
Identification
13
C Metabolomics + Bakiri et al. [63],
Compound Bruguiere et al.
Identification [64]
INADEQUATE Compound Clendinen et al.
Identification [65]

concept consists in the calculation of a most probable dilution factor by looking at

the distribution of the quotients of the amplitudes of a test spectrum by those of a
sample or calculated reference spectrum [70]. Range normalization rescales each
data point to a new scale, usually 0–1. Moreover, reference normalization assumes
that there is a stable reference signal (i.e. TSP, DSS, etc.) in the spectra, which can
be used to normalize each of the remaining features for every spectrum. Caution
must be taken when applying these methods. These can be particularly detrimental
in cases where samples have very different metabolic profiles or situations where a
few compounds are present in extreme concentrations. In either scenario,
6 Advances in Microbial NMR Metabolomics 131

normalization strategies will transform the variance across all samples and therefore
the resulting transformations should be carefully inspected.
Scaling adjustment across features is needed to equate the contributions of peaks
at the extremes of the dynamic range. If normalization is a procedure applied across
samples, scaling is a procedure applied across variable/factor within samples [68,
69]. It is the accepted strategy to limit the detrimental effect of heteroscedasticity of
intense peaks on variance-dependent multivariate techniques (e.g. PCA). The most
commonly used methods are: autoscaling [71], Pareto scaling, and Vast scaling [69,
72, 73]. Autoscaling (or unit variance scaling) is used to convert all variables/factors
to fit an interval of unit variation by operating on each value with the standard devia-
tion obtained after mean centering. Since this method is applied for each variable/
factor, it might produce unwanted value for noise regions. The Pareto scaling
method uses the square root of the standard deviation. This leads to misbalanced
significance for small variations in detriment of large variations. In Vast scaling,
data is operated not only by the standard deviation but also by the coefficient of
variation to emphasize stable peaks after a step of mean-centering [73]. The draw-
back from scaling pretreatment is a possible inflation of small intensity values since
these small peaks can contain great relative measurement error across samples [69].
Better understanding of these critical pretreatment methods can be retrieved in the
following references [26, 42, 66, 68–77].

6.7 Statistical Assessment

Following the usual metabolomics workflow, once all data is processed and stored,
the statistical assessment is applied according to thespecific study design. To dive
deep into this topic is beyond the scope of this chapter, instead, suggestions are
made for further reading [33, 78–85]. Different from application with MS-based
studies, using NMR data in its full-resolution (alternatively to binned data), users
can assess reconstructed spectra according to the Loadings results from a principal
component analysis (PCA), for example, and highlight peaks (factors or variables)
that contribute strongly (positively or negatively) to each principal component. This
view is interesting because it enables users to pinpoint several peaks responsible for
grouping or classification, and often some of these peaks may compose an NMR
spectrum of a compound of interest [86]. PCA and partial least square-­discriminatory
analysis (PLS-DA) are the main statistical techniques used in untargeted metabolo-
mics [81, 87]. These are generally used to identify trends, similarities, or differences
among samples and to achieve reduced-dimensionality for a final goal of an ideal
interpretation of the data collected.
As an example of unsupervised technique, PCA requires no underlying groups
information, and so it is deemed as a purely data-driven exploration. PCA is indeed
generally used as a first look into sample distributing within a metabolomics study.
Briefly, PCA reorganizes data into a new set of orthogonal axes (principal compo-
nents; PCs) by maximizing variance among sample data, thus the multidimensional
132 R. M. Borges et al.

data is reduced by abating redundancy. However, for a number of reasons, the bio-
logical response might not fully emulate the data variance and grouping might be
due to faulty technical procedures (e.g. batch effects and sample order bias). On the
other hand, PLS-DA and its variant OPLS-DA are supervised techniques that
require sample classification beforehand. PLS-DA is usually used after sample visu-
alization with PCA. PLS techniques use the sample classification to explore the
maximum relationship between those and samples variation expressed as covaria-
tion of scores. Thus, as a tool for predictive modeling, PLS includes the calculation
of quality parameters (R2 and Q2) which, together with other cross-validation tech-
niques (e.g. permutation and receiver operating characteristic curve, AUROC), can
provide information about the predictive ability of the model. The evaluation of
these cross-validation parameters is mandatory for any rationalization drawn from
supervised techniques (e.g. ideally, Q2 should be close to R2) to verify the occur-
rence of overfitting [81, 85, 88, 89]. Other methods such as hierarchical cluster
analysis (HCA), random forest (RF) and support vector machine (SVM) are less
used in metabolomics studies [90].

6.8 Tools for NMR-Based Metabolomics

In this section, a list of tools available for NMR-based metabolomics is presented

with their respective references (Table 6.3). It is not the goal of this document to
discuss and compare each tool nor to comprehensively list all the available resources,
but highlight the most common and familiar to the authors.

6.9 Applications

Microorganisms’ short life cycles, easy manipulation and fast responses to a chang-
ing environment have helped unravel biochemical processes and mechanisms
throughout the history of metabolism. In recent years, NMR metabolomics has
emerged as a complement to these experiments, providing a global view of their
metabolic output. This output has been shown capable of describing microorgan-
ism’s physiological states even when there is a lack of observable phenotypes. This
has been aptly demonstrated by Mielko et al. [99] discerning P. aeruginosa col-
lected from Cystic Fibrosis patients and from environmental products. Multivariate
analysis of 1H NMR spectra allowed not only discerning the origin of each strain,
grown under the same conditions, but also indicating pathways that could be respon-
sible for a metabolic shift and therefore pathogenicity. Similarly, Borchert et al.
[100], used 1H NMR profiles and multivariate analysis (PLS-DA) of both Salmonella
enterica spent growth media and bacterial pellets, in conjunction with a metabolite
supplementation strategy, to isolate the cascading effects of the accumulating reac-
tive metabolite 2-aminoacrylate caused by the ridA gene knockout. Furthermore,
 6

Table 6.3 List of available tool commonly used in NMR metabolomics

Tools Where to find them References
NMR data processing
TOPSPIN Bruker
MNOVA Mestrelab
iNMR http://www.inmr.net/
NMRPipe https://www.ibbr.umd.edu/nmrpipe/
ccpNMR https://www.ccpn.ac.uk/ Skinner et al. [91]
NMRbox https://nmrbox.org/ Maciejewski et al. [92]
Statistical assessments
MVAPACK toolbox http://bionmr.unl.edu/mvapack.php Worley and Powers [93]
Edison Lab Metabolomics toolbox (Matlab) https://github.com/artedison/Edison_Lab_Shared_Metabolomics_UGA Robinette et al. [37]
MetaboAnalyst https://www.metaboanalyst.ca/ Chong et al. [26]
Advances in Microbial NMR Metabolomics

NMR database
Nmrshiftdb2 https://nmrshiftdb.nmr.uni-­koeln.de/ Steinbeck and Kuhn [9]
BMRB https://bmrb.io/ Bingol et al. [10]
HMDB https://hmdb.ca/ Wishart et al. [94]
COLMAR http://spin.ccic.ohio-­state.edu/index.php/colmar Bingol et al. [11], Wang et al. [39]
E. coli Metabolome Database (ECMDB) http://www.ecmdb.ca Sajed et al. [95]
Spectral Data Base System (SDBS) https://sdbs.db.aist.go.jp/sdbs/cgi-­bin/cre_index.cgi
Compound identification
NMRfilter https://github.com/stefhk3/nmrfilter Kuhn [19], Kuhnet al. [62]
SMART 2.0 https://smart.ucsd.edu/classic Reher et al. [96]
MADByTE https://www.madbyte.org/ Egan et al. [97]
COLMAR http://spin.ccic.ohio-­state.edu/index.php/colmar Bingol et al. [11], Wang et al. [39]
NP-MRD https://npmrd-­project.org/ Wishart et al. [98]
133
134 R. M. Borges et al.

the integration of NMR metabolomics with biochemical experiments can expand

previously unknown or alternate mechanistic understanding of specific biochemical
reactions. Marshall et al. [101], illustrated this point resolving longstanding contra-
dictory literature by uncovering an alternative biosynthetic pathway of D-alanine in
Mycobacterium smegmatis. The authors used an untargeted approach using 1H
NMR profiles but also targeted specific metabolites by collecting HSQC0 of 13C
labelled metabolites, which allowed building a calibration curve for quantification
and concluding that inhibitors of the Alr protein would not be bactericidal.
For similar reasons, microorganisms have gained a lot of attention in recent years
from natural products research specifically in respect to drug discovery. NMR
metabolomics was used to correlate different metabolite profiles of Streptomyces
spp. and the production of antibiotic against the phytopathogenic bacteria
Burkholderia spp.. Streptomyces spp. were cultivated in four different liquid media
to induce diverse metabolic profile and samples were collected time-dependently.
After centrifuging, the supernatant was extracted with ethyl acetate to obtain the
samples for NMR data acquisition and antimicrobial assay. OPLS-DA of the NMR
data over the different fractions using results from the antimicrobial assay as sample
classifier enabled the authors to find chemical shifts held responsible for the distinc-
tion between active and non-active samples. Later, they followed a traditional bio-­
guided procedure to isolate two compounds and to validate the metabolomics
approach [102] (Fig. 6.3). This approach combining chemical profiling, supervised

Fig. 6.3 Brief scheme representing the study [102]. (Adapted from “Betancur et al. [102].”
Reprinted under license number 5253730166382 (Feb 21, 2022; Microbiological Research,
Elsevier))
6 Advances in Microbial NMR Metabolomics 135

exploratory multivariate analysis (e.g. PLS), and classification between active vs

non-active samples (NMR/MS-PLS-Bioactivity) to find promising biologically
active compounds (e.g. VIP compounds obtained from PLS) is a straightforward
application of metabolomics tools in natural products [103–106]. However, the
main limitations for this application are related to the access to the analytical instru-
mentation and knowledge on chemometrics. In a phytochemical study, Carvalho
et al. [103] designed a combination of metabolic profiling and a series of antioxi-
dant, antiglioma, and antibacterial bioassays to identify bioactive compounds. The
analysis of the different extracts obtained led to differential chemical profiles and
differential bioassay results. PLS-DA was applied using the bioassay result as clas-
sifier to determine VIP resonances characterizing compounds as promising bioac-
tive candidates [103].
Using a One Strain - Many Compounds (OSMAC) approach, 4160 fractions of
13 actinomycetes cultivated under 32 growth conditions to diverse their secondary
metabolism were analyzed using a metabolomics-like approach on their NMR data.
Additionally, the author included an assay to evaluate the inhibition effects of frac-
tions against Mycobacteria bovis Bacillus Calmettee-Guerin (BCG). Aided by a
PCA score plot, they were able to prioritize 37 fractions for searching for bioactive
compounds. This study has shown the value of an OSMAC-NMR-PCA approach to
accelerate drug discovery [107]. Fernand et al. [108] also applied an OSMAC
approach on an Aspergillus niger strain isolated from the shrimp Penaeus monodon
to study the global differential metabolite production. Seven different agar culture
media were used and the mycelium with the agar layer were extracted to obtain both
the endo- and exo-metabolome [108]. The differentiation among the samples was
done using a PCA analysis on data collected from both LC-MS and 1H NMR and
the overall outcome was shown to be highly comparable between analytical meth-
ods; however, only the LC-MS data were used to annotate secondary metabolites.
A common use of NMR metabolomics is found within pipelines for taxonomical
differentiation of microbial species. Zabek et al. [43] have shown the distinction
among three models of fungal strains (Aspergillus pallidofulvus, Fusarium oxyspo-
rum, and Geotrichum candidum) by analyzing their endo-metabolome. For this,
they cultivated the strains in 100 mL of potato dextrose broth (PDB) inoculated to a
final concentration of 106 conidia per mL of medium. Aliquots of 100 mg of the
filtered and washed fungal biomass were frozen, homogenized, and extracted with
phosphate buffer containing 10% of D2O. After centrifugation, the supernatant was
transferred to NMR tubes. Representative 1H NMR spectra were obtained and
assigned using the BMRB and HMDB database and a PCA was used to assess spe-
cies dissimilarity. Interestingly, a similar approach was used to distinguish among
576 samples of 6 different bacterial species using only their exo-metabolome [44].
Remarkably, eight different strains of Escherichia coli were discriminated based on
their metabolite profiles using 1H NMR data processed with PCA [109]. The authors
went further to investigate their data under an enrichment pathway analysis using 38
identified compounds in order to rationalize over the strain differentiation and fur-
thercharacterize metabolic clues for each of them.
136 R. M. Borges et al.

Aries and Cloninger [110] applied NMR metabolomics to differentiate wild type
and mutant E. coli strainsas part of a study to elucidate mechanisms for drug resis-
tance in bacteria. The researchers designed an experiment to assess the metabolic
pathways that are altered when E. coli is threatened with antibacterial agents. They
studied the effect of dendrimer functionalized with mannose and a quaternary
ammonium compound as endgroup, called DABCOMD, to elicit bacterial resis-
tance under minimum inhibitory concentrations after 33 growth cycles. Once the
mutant strains were produced, a comparison study was established between those
mutated strains and the wild type E. coli. The bacteria were grown in Mueller
Hinton II broth and centrifuged cell pellets extracted with methanol-water followed
by a partition with chloroform to remove lipids and a protein precipitation step. The
final aqueous extract was dissolved in D2O phosphate buffer and used for 1H NMR
data acquisition. The authors used the MetaboAnalyst Pathway Analysis tool
together with KEGG and PLS-DA to assess the results and elucidate the effects of
treatment with DABCOMD in E. coli. It appears that mutant strains presented
energy related pathways altered and an increase of metabolites involved in the
membrane peptidoglycan synthesis [110] (Fig. 6.4).
With the same goal of understanding microbial cellular mechanisms for develop-
ing new antimicrobials, the metabolic changes on E. coli were evaluated after treat-
ment with electrolyzed water as it is used by the food industry to yield oxidative
damage to pathogenic microorganisms [111]. Together with morphology evaluation
by atomic force microscopy, NMR metabolomics was used to assess the affected
metabolic pathways after treatment. Four time-points were obtained to enable
chronological evaluation aided by the PCA score plot. The NMR data was anno-
tated using the E. coli Metabolome Database [95] and the binned data together with
the annotated compounds were submitted to pathway analysis using MetaboAnalyst
3.0. Five pathways related to energy metabolism were deemed affected by the treat-
ment with electrolyzed water in agreement to the results seen with the OPLS-DA
and VIP compounds [111].
Rona et al. [112] designed a protocol to evaluate a functional link between human
oncoprotein NSD3s and cancer metabolic reprogramming using Saccharomyces
cerevisiae overexpressing NSD3s or Pbp3 and 1H NMR experiments. Briefly, con-
centrated cell pellets were quenched under −80 °C and extracted with aqueous etha-
nol at 75%. The extracts were resuspended in 0.05 M D2O phosphate buffer at

Fig. 6.4 Brief scheme representing the study [110]. (Adapted from Aries and Cloninger [110])
6 Advances in Microbial NMR Metabolomics 137

pH 7.4 for NMR data acquisition. PCA score plots showed clear grouping distin-
guishing recombinant strains from the wild type strain samples. Database compari-
son using HMDB, BMRB, and COLMAR enabled authors to drawn their conclusions
concerning the increasing of glutaminolysis rate and the adaptation of metabolic
pathways.
Microorganism biomass obtained in laboratory scale is usually limited, and so
isthe extract yield,what might end up restricting the detection of secondary metabo-
lites above the limit of detection in NMR. In plants, this is different; the amount of
extract obtained is frequently more than enough and enable users to safely proceed
with a more selective and multi-step sample preparation method (e.g. liquid-liquid
extraction, solid phase extraction, or fractionation). Thus, plant NMR metabolomics
for detection of secondary metabolite are more common [4, 33, 34, 38, 83, 113–
118]. An interesting approach to grow large scale amounts of the nematode
Caenorhabditis elegans to collect analytical and phenotypic data was taken by
using cake pans [119]. Their procedure was proven to be consistent and reproduc-
ible for the study of stage related metabolome of C. elegans.
NMR finds remarkable advantages when isotopic labels are used, for example, in
in vivo metabolomics and metabolic pathways. Within natural products, the most
common application of isotopic labels is in biosynthesis studies and 13C is usually
the isotope of choice. Some aspects of 13C NMR are important to call attention such
as the low sensitivity (compared to 1H NMR) due to its low natural abundance
(1.1%; where 1H has a natural abundance of 99.98%) and the wide spectral window
(~0–200 ppm for 13C; where 1H has a spectral window ~0–12 ppm) and conse-
quently broader chemical shift dispersion. In contrast to what occurs in MS, with
the use of isotopic labels in NMR it is feasible to filter out non-labeled peaks mak-
ing the spectra simpler to be analyzed.
Polozsányi et al. [120] investigated whether conidia from filamentous fungi are
metabolic active or not by analyzing selected enzyme activities, 45Ca2+ inward trans-
port, and 13C-label bicarbonate (HCO3−). Then, the authors used 1H and 13C NMR
and LC-MS metabolomics to access variations of metabolites, but unfortunately the
use of H13C-CO3− did not led to satisfactory labelling; when only the 1H NMR data
were used. More specifically, they used 1H NMR to inspect metabolic changes dur-
ing the conidial maturation. They were able to track concentration variations in
more than 30 identified metabolites during 43 days. Interestingly, conidia in their
yearly days presented characteristic variations for different metabolites clearly indi-
cating enzymatic/metabolic activity (Fig. 6.5).
By applying a novel method called continuous in vivo monitoring of metabolism
by NMR (CIVM-NMR), Judge et al. [121] successfully characterized metabolic
variations in time using a High Resolution-Magic Angle Spinning (HR-MAS) probe
for solid state NMR. A key feature of this study was the minimum sample prepara-
tion required: the mycelium pellet imbibed with D2O Vogel’s media was pushed into
the rotor for direct analysis in the HR-MAS probe. NMR data were acquired 16
times over 65 min to create high temporal resolution and annotated peak intensities
were tracked for visualization and interpretation, which is out the scope here. Other
example exploring similar aspects of NMR for in vivo cell suspensions can be find
elsewhere [122] (Fig. 6.6).
138 R. M. Borges et al.

Fig. 6.5 Brief scheme representing the study [120]. (Adapted from [120]. Reprinted under license
number 5253740496369 (Feb 21, 2022; Fungal Biology, Elsevier))

Fig. 6.6 Brief scheme representing the study [121]. (Adapted from Judge et al. [121]. Copyright
© 2019 Judge, Wu, Tayyari, Hattori, Glushka, Ito, Arnold and Edison)

6.10 Conclusion and Future Perspectives

Instrumental developments on NMR on cryoprobes, sample volume capabilities,

ultra-high field magnets, pulse sequence experiments, detection methods, etc. fore-
shadow a bright future for NMR metabolomics. The advent and spread of cryo-
probes and microprobes (1 to 30 μL; Fig. 6.7) are among the main hardware
improvement in the way to tackle the intrinsic low sensitivity of NMR. The sensitiv-
ity gain, reflected on increased signal-to-noise ratio, resulted from the cryogenically
cooling the radiofrequency coils in the probe to near absolute zero temperature,
which is up to four times greater comparing to room temperature probes. On the
other hand, sample volume reduction leads to more concentrated sample and more
molecules in the active region of the receiver and decoupling coils. Samples that are
more concentrated generally yields higher intensity signals [123], save situations
when oversaturated solutions may lead to signal distortions. Higher-field magnets
also increase sensitivity and resolution, but to achieve ultra-high fields is still too
expensive for general use. Dynamic nuclear polarization (DNP) is an exciting new
venue to be explored for signal intensity improvements. It relies on phenomenon
called hyperpolarization, which produces higher nuclear spin polarization, and so
favoring the equilibrium distributions between the α-β states (Fig. 6.1). DNP NMR
might become the tool responsible for the expansion of 13C NMR in routine metabo-
lomics in the future [124].
6 Advances in Microbial NMR Metabolomics 139

Fig. 6.7 Scheme to represent common NMR tube diameters and their sample volumes

Fig. 6.8 Representation of the measurement time in 1H/13C 2D Correlation Spectra with and with-
out NUS. (Adapter from Delaglio et al. [125])

Regarding pulse sequence designs, faster methods and experiments are under
constant developments. Even though most of the NMR metabolomics is made by
using a 1D 1H-based experiment (e.g.: noesypr from Bruker), bidimensional experi-
ments can be used mainly with the goal of reducing signal overlapping. The main
limitation for the use of 2D NMR on metabolomics is the amount of time needed for
each data to be collected. Some novel pulse sequences were developed to shorten
the experimental time of 2D data collection. Band-selective optimized flip angle
short transient (SOFAST), band selective excitation short transient (BEST) and
acceleration by sharing adjacent polarization (ASAP) based methods are some
examples.
The use of non-uniform sampling (NUS) has been shown to significantly reduce
experimental time for higher dimensional experiments [125] (Fig. 6.8). In NUS,
points in the data collection are sparse instead of regularly recorded, consequently
140 R. M. Borges et al.

the overall time of a single experiment can be reduced by up to 75% depending on

the sample.
Important advances have been made in NMR to improve its applications for
users to explore its unique features as an outstanding analytical technique [1, 50,
51]. Clearly, a lot can be done with NMR in Metabolomics, especially for microbial
research and here some cases were presented aiming to show possible ways forward.

References

1. Wishart DS (2019) NMR metabolomics: a look ahead. J Magn Reson 306:155–161

2. Shoulders B (2000) High-resolution NMR techniques in organic chemistry. Tetrahedron
organic chemistry series volume 19 by Timothy D. W. Claridge (Dyson Perrins Laborotory,
Oxford). Pergamon: Oxford. 1999. ix + 384 pp. Hardbound $134.50, ISBN 0-08-042799-5.
Paperback $49.50, ISBN 0-08-042798-7. J Am Chem Soc 122(33):8104–8104
3. Soares V, Bezerra T, Lafetá R, Borges R, da Silva AJ (2016) Three new steroidal Glycoalkaloids
from Solanum pseudoquina A. St.-Hil. (Solanaceae). J Braz Chem Soc 28:782–789
4. Simmler C, Napolitano JG, McAlpine JB, Chen SN, Pauli GF (2014) Universal quantitative
NMR analysis of complex natural samples. Curr Opin Biotechnol 25:51–59
5. Westwood S, Yamazaki T, Huang T, Garrido B, Ün I, Zhang W, Martos G, Stoppacher N,
Saito T, Wielgosz R (2019) Development and validation of a suite of standards for the purity
assignment of organic compounds by quantitative NMR spectroscopy. Metrologia 56(6)
6. Emwas A-HM (2015) The strengths and weaknesses of NMR spectroscopy and mass spec-
trometry with particular focus on metabolomics research. In: Bjerrum JT (ed) Metabonomics:
methods and protocols. Springer New York, New York, pp 161–193
7. Evard H, Kruve A, Leito I (2016) Tutorial on estimating the limit of detection using LC-MS
analysis, part I: theoretical review. Anal Chim Acta 942:23–39
8. Wang M, Carver JJ, Phelan VV, Sanchez LM, Garg N, Peng Y, Nguyen DD, Watrous J, Kapono
CA, Luzzatto-Knaan T, Porto C, Bouslimani A, Melnik AV, Meehan MJ, Liu WT, Crusemann
M, Boudreau PD, Esquenazi E, Sandoval-Calderon M, Kersten RD, Pace LA, Quinn RA,
Duncan KR, Hsu CC, Floros DJ, Gavilan RG, Kleigrewe K, Northen T, Dutton RJ, Parrot D,
Carlson EE, Aigle B, Michelsen CF, Jelsbak L, Sohlenkamp C, Pevzner P, Edlund A, McLean
J, Piel J, Murphy BT, Gerwick L, Liaw CC, Yang YL, Humpf HU, Maansson M, Keyzers RA,
Sims AC, Johnson AR, Sidebottom AM, Sedio BE, Klitgaard A, Larson CB, Torres-Mendoza
CABPD, Gonzalez DJ, Silva DB, Marques LM, Demarque DP, Pociute E, O'Neill EC, Briand
E, Helfrich EJN, Granatosky EA, Glukhov E, Ryffel F, Houson H, Mohimani H, Kharbush
JJ, Zeng Y, Vorholt JA, Kurita KL, Charusanti P, McPhail KL, Nielsen KF, Vuong L, Elfeki
M, Traxler MF, Engene N, Koyama N, Vining OB, Baric R, Silva RR, Mascuch SJ, Tomasi
S, Jenkins S, Macherla V, Hoffman T, Agarwal V, Williams PG, Dai J, Neupane R, Gurr J,
Rodriguez AMC, Lamsa A, Zhang C, Dorrestein K, Duggan BM, Almaliti J, Allard PM,
Phapale P, Nothias LF, Alexandrov T, Litaudon M, Wolfender JL, Kyle JE, Metz TO, Peryea
T, Nguyen DT, VanLeer D, Shinn P, Jadhav A, Muller R, Waters KM, Shi W, Liu X, Zhang
L, Knight R, Jensen PR, Palsson BO, Pogliano K, Linington RG, Gutierrez M, Lopes NP,
Gerwick WH, Moore BS, Dorrestein PC, Bandeira N (2016) Sharing and community cura-
tion of mass spectrometry data with global natural products social molecular networking. Nat
Biotechnol 34(8):828–837
9. Steinbeck C, Kuhn S (2004) NMRShiftDB -- compound identification and structure elucida-
tion support through a free community-built web database. Phytochemistry 65(19):2711–2717
6 Advances in Microbial NMR Metabolomics 141

10. Bingol K, Li DW, Bruschweiler-Li L, Cabrera OA, Megraw T, Zhang F, Bruschweiler R

(2015) Unified and isomer-specific NMR metabolomics database for the accurate analysis of
(13)C-(1)H HSQC spectra. ACS Chem Biol 10(2):452–459
11. Bingol K, Li DW, Zhang B, Bruschweiler R (2016) Comprehensive metabolite identification
strategy using multiple two-dimensional NMR spectra of a complex mixture implemented in
the COLMARm web server. Anal Chem 88(24):12411–12418
12. Sorokina M, Steinbeck C (2020) Review on natural products databases: where to find data in
2020. J Cheminform 12(1):20
13. Borges RM, Colby SM, Das S, Edison AS, Fiehn O, Kind T, Lee J, Merrill AT, Merz KM Jr,
Metz TO, Nunez JR, Tantillo DJ, Wang LP, Wang S, Renslow RS (2021) Quantum chemistry
calculations for metabolomics. Chem Rev 121(10):5633–5670
14. Jonas E, Kuhn S (2019) Rapid prediction of NMR spectral properties with quantified uncer-
tainty. J Cheminform 11(1):50
15. Luiz Paranhos Costa F, de Albuquerque ACF, Borges RM, dos Santos Junior FM, de
Amorim MB (2014) High cost-effectiveness ratio: GIAO-MPW1PW91/6-31G(d)//
MPW1PW91/6-31G(d) scaling factor for <SUP>13</SUP>C nuclear magnetic resonance
chemical shifts calculation. J Comput Theor Nanosci 11(1):219–225
16. Kuhn S, Johnson SR (2019) Stereo-aware extension of HOSE codes. ACS Omega
4(4):7323–7329
17. Hayden HL, Rochfort SJ, Ezernieks V, Savin KW, Mele PM (2019) Metabolomics approaches
for the discrimination of disease suppressive soils for Rhizoctonia solani AG8 in cereal crops
using (1)H NMR and LC-MS. Sci Total Environ 651(Pt 1):1627–1638
18. Jousse C, Dalle C, Abila A, Traikia M, Diogon M, Lyan B, El Alaoui H, Vidau C, Delbac
F (2020) A combined LC-MS and NMR approach to reveal metabolic changes in the
hemolymph of honeybees infected by the gut parasite Nosema ceranae. J Invertebr Pathol
176:107478
19. Kuhn S, Colreavy-Donnelly S, Santana de Souza J, Borges RM (2019) An integrated
approach for mixture analysis using MS and NMR techniques. Faraday Discuss 218:339–353
20. Leggett A, Wang C, Li DW, Somogyi A, Bruschweiler-Li L, Bruschweiler R (2019)
Identification of unknown metabolomics mixture compounds by combining NMR, MS, and
cheminformatics. Methods Enzymol 615:407–422
21. Morgan MA, Griffith CM, Dinges MM, Lyon YA, Julian RR, Larive CK (2019) Evaluating
sub-lethal stress from roundup((R)) exposure in Artemia franciscana using (1)H NMR and
GC-MS. Aquat Toxicol 212:77–87
22. Silva EL, Almeida-Lafeta RC, Borges RM, Staerk D (2018) Athenolide a, a new steroidal
lactone from the leaves of Athenaea martiana (Solanaceae) determined by means of HPLC-­
HR-­MS-SPE-NMR analysis. Chem Biodivers 15(1)
23. Tomita S, Nakamura T, Okada S (2018) NMR- and GC/MS-based metabolomic characteriza-
tion of sunki, an unsalted fermented pickle of turnip leaves. Food Chem 258:25–34
24. Walker LR, Hoyt DW, Walker SM 2nd, Ward JK, Nicora CD, Bingol K (2016) Unambiguous
metabolite identification in high-throughput metabolomics by hybrid 1D (1) H NMR/ESI
MS(1) approach. Magn Reson Chem 54(12):998–1003
25. Blaise BJ, Correia G, Tin A, Young JH, Vergnaud AC, Lewis M, Pearce JT, Elliott P, Nicholson
JK, Holmes E, Ebbels TM (2016) Power analysis and sample size determination in metabolic
phenotyping. Anal Chem 88(10):5179–5188
26. Chong J, Soufan O, Li C, Caraus I, Li S, Bourque G, Wishart DS, Xia J (2018) MetaboAnalyst
4.0: towards more transparent and integrative metabolomics analysis. Nucleic Acids Res
46(W1):W486–W494
27. Halouska S, Zhang B, Gaupp R, Lei S, Snell E, Fenton RJ, Barletta RG, Somerville GA,
Powers R (2013) Revisiting protocols for the NMR analysis of bacterial metabolomes. J
Integr OMICS 3(2):120–137
28. Broadhurst D, Goodacre R, Reinke SN, Kuligowski J, Wilson ID, Lewis MR, Dunn WB
(2018) Guidelines and considerations for the use of system suitability and quality control
142 R. M. Borges et al.

samples in mass spectrometry assays applied in untargeted clinical metabolomic studies.

Metabolomics 14(6):72
29. Dunn WB, Broadhurst DI, Edison A, Guillou C, Viant MR, Bearden DW, Beger RD (2017)
Quality assurance and quality control processes: summary of a metabolomics community
questionnaire. Metabolomics 13(5):1–6
30. Garcia CJ, Yang X, Huang D, Tomas-Barberan FA (2020) Can we trust biomarkers identified
using different non-targeted metabolomics platforms? Multi-platform, inter-laboratory com-
parative metabolomics profiling of lettuce cultivars via UPLC-QTOF-MS. Metabolomics
16(8):85
31. Gouveia GJ, Shaver AO, Garcia BM, Morse AM, Andersen EC, Edison AS, McIntyre LM
(2021) Long-term metabolomics reference material. Anal Chem 93(26):9193–9199
32. Maier TS, Kuhn J, Muller C (2010) Proposal for field sampling of plants and processing in
the lab for environmental metabolic fingerprinting. Plant Methods 6:6
33. Deborde C, Moing A, Roch L, Jacob D, Rolin D, Giraudeau P (2017) Plant metabolism as
studied by NMR spectroscopy. Prog Nucl Magn Reson Spectrosc 102-103:61–97
34. Kim HK, Choi YH, Verpoorte R (2010) NMR-based metabolomic analysis of plants. Nat
Protoc 5(3):536–549
35. Nagana Gowda GA, Raftery D (2017) Recent advances in NMR-based metabolomics. Anal
Chem 89(1):490–510
36. Newton JM, Betts EL, Yiangou L, Ortega Roldan J, Tsaousis AD, Thompson GS (2021)
Establishing a metabolite extraction method to study the metabolome of blastocystis using
NMR. Molecules 26(11):3285
37. Robinette SL, Brüschweiler R, Schroeder FC, Edison AS (2012) NMR in metabolomics and
natural products research: two sides of the same coin. Acc Chem Res 45(2):288–297
38. Schripsema J (2010) Application of NMR in plant metabolomics: techniques, problems and
prospects. Phytochem Anal 21(1):14–21
39. Wang C, Timari I, Zhang B, Li DW, Leggett A, Amer AO, Bruschweiler-Li L, Kopec RE,
Bruschweiler R (2020a) COLMAR lipids web server and ultrahigh-resolution methods for
two-dimensional nuclear magnetic resonance- and mass spectrometry-based Lipidomics. J
Proteome Res 19(4):1674–1683
40. Wang R, Li B, Lam SM, Shui G (2020b) Integration of lipidomics and metabolomics for
in-depth understanding of cellular mechanism and disease progression. J Genet Genomics
47(2):69–83
41. Wist J (2017) Complex mixtures by NMR and complex NMR for mixtures: experimental and
publication challenges. Magn Reson Chem 55(1):22–28
42. Zacharias HU, Altenbuchinger M, Gronwald W (2018) Statistical analysis of NMR metabolic
fingerprints: established methods and recent advances. Meta 8(3)
43. Zabek A, Klimek-Ochab M, Jawien E, Mlynarz P (2017) Biodiversity in targeted metabo-
lomics analysis of filamentous fungal pathogens by (1)H NMR-based studies. World J
Microbiol Biotechnol 33(7):132
44. Palama TL, Canard I, Rautureau GJ, Mirande C, Chatellier S, Elena-Herrmann B (2016)
Identification of bacterial species by untargeted NMR spectroscopy of the exo-metabolome.
Analyst 141(15):4558–4561
45. Beltran A, Suarez M, Rodriguez MA, Vinaixa M, Samino S, Arola L, Correig X, Yanes O
(2012) Assessment of compatibility between extraction methods for NMR- and LC/MS-based
metabolomics. Anal Chem 84(14):5838–5844
46. Mielko KA, Jablonski SJ, Lukaszewicz M, Mlynarz P (2021) Comparison of bacteria disin-
tegration methods and their influence on data analysis in metabolomics. Sci Rep 11(1):20859
47. Dong C, Reilly DK, Bergame C, Dolke F, Srinivasan J, von Reuss SH (2018) Comparative
ascaroside profiling of Caenorhabditis exometabolomes reveals species-specific (omega)
and (omega - 2)-hydroxylation downstream of peroxisomal beta-oxidation. J Org Chem
83(13):7109–7120
6 Advances in Microbial NMR Metabolomics 143

48. Lugo Charriez K, Soledade Lemos L, Carrazana Y, Rodriguez-Casariego JA, Eirin-Lopez JM,
Hauser-Davis RA, Gardinali P, Quinete N (2021) Application of an improved chloroform-­
free lipid extraction method to staghorn coral (Acropora cervicornis) lipidomics assessments.
Bull Environ Contam Toxicol 107:92–99
49. Shiva S, Enninful R, Roth MR, Tamura P, Jagadish K, Welti R (2018) An efficient modified
method for plant leaf lipid extraction results in improved recovery of phosphatidic acid. Plant
Methods 14:14
50. Edison AS, Colonna M, Gouveia GJ, Holderman NR, Judge MT, Shen X, Zhang S
(2021) NMR: unique strengths that enhance modern metabolomics research. Anal Chem
93(1):478–499
51. Emwas AH, Roy R, McKay RT, Tenori L, Saccenti E, Gowda GAN, Raftery D, Alahmari F,
Jaremko L, Jaremko M, Wishart DS (2019) NMR spectroscopy for metabolomics research.
Meta 9(7):123
52. Forseth RR, Schroeder FC (2011) NMR-spectroscopic analysis of mixtures: from structure to
function. Curr Opin Chem Biol 15(1):38–47
53. Hansen AL, Kupce ER, Li DW, Bruschweiler-Li L, Wang C, Bruschweiler R (2021) 2D
NMR-based metabolomics with HSQC/TOCSY NOAH supersequences. Anal Chem
93(15):6112–6119
54. Malet-Martino M, Holzgrabe U (2011) NMR techniques in biomedical and pharmaceutical
analysis. J Pharm Biomed Anal 55(1):1–15
55. Giraudeau P, Silvestre V, Akoka S (2015) Optimizing water suppression for quantitative
NMR-based metabolomics: a tutorial review. Metabolomics 11(5):1041–1055
56. Simpson AJ, Brown SA (2005) Purge NMR: effective and easy solvent suppression. J Magn
Reson 175(2):340–346
57. Le Guennec A, Tayyari F, Edison AS (2017) Alternatives to nuclear overhauser enhancement
spectroscopy presat and Carr-Purcell-Meiboom-Gill presat for NMR-based metabolomics.
Anal Chem 89(17):8582–8588
58. Meiboom S, Gill D (1958) Modified spin-Echo method for measuring nuclear relaxation
times. Rev Sci Instrum 29(8):688–691
59. Aguilar JA, Nilsson M, Bodenhausen G, Morris GA (2012) Spin echo NMR spectra without
J modulation. Chem Commun 48(6):811–813
60. Ferretti JA, Highet RJ, Pohl LR, Monks TJ, Hinson JA (1985) Two-dimensional J-resolved
nuclear magnetic resonance spectral study of two bromobenzene glutathione conjugates.
Environ Health Perspect 61:139–145
61. Huang Y, Zhang Z, Chen H, Feng J, Cai S, Chen Z (2015) A high-resolution 2D J-resolved
NMR detection technique for metabolite analyses of biological samples. Sci Rep 5(1):8390
62. Kuhn S, Colreavy-Donnelly S, de Andrade Silva Quaresma LE, de Andrade Silva Quaresma
E, Borges RM (2020) Applying NMR compound identification using NMRfilter to match
predicted to experimental data. Metabolomics 16(12):123
63. Bakiri A, Hubert J, Reynaud R, Lanthony S, Harakat D, Renault JH, Nuzillard JM (2017)
Computer-aided (13)C NMR chemical profiling of crude natural extracts without fraction-
ation. J Nat Prod 80(5):1387–1396
64. Bruguiere A, Derbre S, Coste C, Le Bot M, Siegler B, Leong ST, Sulaiman SN, Awang
K, Richomme P (2018) (13)C-NMR dereplication of Garcinia extracts: predicted chemical
shifts as reliable databases. Fitoterapia 131:59–64
65. Clendinen CS, Pasquel C, Ajredini R, Edison AS (2015) (13)C NMR metabolomics:
INADEQUATE network analysis. Anal Chem 87(11):5698–5706
66. Emwas AH, Saccenti E, Gao X, McKay RT, Dos Santos V, Roy R, Wishart DS (2018)
Recommended strategies for spectral processing and post-processing of 1D (1)H-NMR data
of biofluids with a particular focus on urine. Metabolomics 14(3):31
67. Palaric C, Pilard S, Fontaine JX, Boccard J, Mathiron D, Rigaud S, Cailleu D, Mesnard F, Gut
Y, Renaud T, Petit A, Beaumal JY, Molinie R (2019) Processing of NMR and MS metabolo-
144 R. M. Borges et al.

mics data using chemometrics methods: a global tool for fungi biotransformation reactions
monitoring. Metabolomics 15(8):107
68. Hendriks MMWB, van Eeuwijk FA, Jellema RH, Westerhuis JA, Reijmers TH, Hoefsloot
HCJ, Smilde AK (2011) Data-processing strategies for metabolomics studies. TrAC Trends
Anal Chem 30(10):1685–1698
69. van den Berg RA, Hoefsloot HC, Westerhuis JA, Smilde AK, van der Werf MJ (2006)
Centering, scaling, and transformations: improving the biological information content of
metabolomics data. BMC Genomics 7:142
70. Dieterie F, Ross A, Schlotterbeck G, Senn H (2006) Probabilistic quotient normalization as
robust method to account for dilution of complex biological mixtures. Application in 1H
NMR metabonomics. Anal Chem 78:4281–4290
71. Jackson JE (1991) Scaling of data. John Wiley & Sons, Inc
72. Blaise BJ, Correia GDS, Haggart GA, Surowiec I, Sands C, Lewis MR, Pearce JTM, Trygg J,
Nicholson JK, Holmes E, Ebbels TMD (2021) Statistical analysis in metabolic phenotyping.
Nat Protoc 16(9):4299–4326
73. Keun HC, Ebbels TMD, Antti H, Bollard ME, Beckonert O, Holmes E, Lindon JC, Nicholson
JK (2003) Improved analysis of multivariate data by variable stability scaling: application to
NMR-based metabolic profiling. Anal Chim Acta 490(1–2):265–276
74. Bararpour N, Gilardi F, Carmeli C, Sidibe J, Ivanisevic J, Caputo T, Augsburger M, Grabherr
S, Desvergne B, Guex N, Bochud M, Thomas A (2020) Visualization and normalization of
drift effect across batches in metabolome-wide association studies. BioRxiv
75. Kohl SM, Klein MS, Hochrein J, Oefner PJ, Spang R, Gronwald W (2012) State-of-the art
data normalization methods improve NMR-based metabolomic analysis. Metabolomics
8(Suppl 1):146–160
76. Reinhold D, Pielke-Lombardo H, Jacobson S, Ghosh D, Kechris K (2019) Pre-analytic con-
siderations for mass spectrometry-based untargeted metabolomics data. Methods Mol Biol
1978:323–340
77. Silva LP, Lorenzi PL, Purwaha P, Yong V, Hawke DH, Weinstein JN (2013) Measurement
of DNA concentration as a normalization strategy for metabolomic data from adherent cell
lines. Anal Chem 85(20):9536–9542
78. Barnes S, Benton HP, Casazza K, Cooper SJ, Cui X, Du X, Engler J, Kabarowski JH, Li
S, Pathmasiri W, Prasain JK, Renfrow MB, Tiwari HK (2016a) Training in metabolomics
research. I. Designing the experiment, collecting and extracting samples and generating
metabolomics data. J Mass Spectrom 51(7):461–475
79. Barnes S, Benton HP, Casazza K, Cooper SJ, Cui X, Du X, Engler J, Kabarowski JH, Li
S, Pathmasiri W, Prasain JK, Renfrow MB, Tiwari HK (2016b) Training in metabolomics
research. II. Processing and statistical analysis of metabolomics data, metabolite identi-
fication, pathway analysis, applications of metabolomics and its future. J Mass Spectrom
51(8):535–548
80. Fay DS, Gerow K (2013) A biologist’s guide to statistical thinking and analysis. In:
WormBook, pp 1–54
81. Gromski PS, Muhamadali H, Ellis DI, Xu Y, Correa E, Turner ML, Goodacre R (2015) A
tutorial review: metabolomics and partial least squares-discriminant analysis--a marriage of
convenience or a shotgun wedding. Anal Chim Acta 879:10–23
82. Mutihac L, Mutihac R (2008) Mining in chemometrics. Anal Chim Acta 612(1):1–18
83. Sumner LW, Mendes P, Dixon RA (2003) Plant metabolomics: large-scale phytochemistry in
the functional genomics era. Phytochemistry 62(6):817–836
84. Wishart DS (2008) Quantitative metabolomics using NMR. TrAC Trends Anal Chem
27(3):228–237
85. Worley B, Powers R (2016) PCA as a practical indicator of OPLS-DA model reliability. Curr
Metabolomics 4(2):97–103
6 Advances in Microbial NMR Metabolomics 145

86. Selegato DM, Freire RT, Tannús A, Castro-Gamboa I (2016) New Dereplication method
applied to NMR-based metabolomics on different Fusarium species isolated from rhizo-
sphere of Senna spectabilis. J Braz Chem Soc 27:1421–1431
87. Kumar N, Bansal A, Sarma GS, Rawal RK (2014) Chemometrics tools used in analytical
chemistry: an overview. Talanta 123:186–199
88. Bevilacqua M, Bro R (2020) Can we trust score plots? Meta 10(7)
89. Westerhuis JA, Hoefsloot HCJ, Smit S, Vis DJ, Smilde AK, van Velzen EJJ, van Duijnhoven
JPM, van Dorsten FA (2008) Assessment of PLSDA cross validation. Metabolomics
4(1):81–89
90. Ghosh T, Zhang W, Ghosh D, Kechris K (2020) Predictive modeling for metabolomics data.
Methods Mol Biol 2104:313–336
91. Skinner SP, Fogh RH, Boucher W, Ragan TJ, Mureddu LG, Vuister GW (2016) CcpNmr
AnalysisAssign: a flexible platform for integrated NMR analysis. J Biomol NMR
66(2):111–124
92. Maciejewski MW, Schuyler AD, Gryk MR, Moraru PR II, Romero EL, Ulrich HR, Eghbalnia
M, Livny FD, Hoch JC (2017) NMRbox: a resource for biomolecular NMR computation.
Biophys J 112(8):1529–1534
93. Worley B, Powers R (2014) MVAPACK: a complete data handling package for NMR metab-
olomics. ACS Chem Biol 9(5):1138–1144
94. Wishart DS, Tzur D, Knox C, Eisner R, Guo AC, Young N, Cheng D, Jewell K, Arndt D,
Sawhney S, Fung C, Nikolai L, Lewis M, Coutouly MA, Forsythe I, Tang P, Shrivastava S,
Jeroncic K, Stothard P, Amegbey G, Block D, Hau DD, Wagner J, Miniaci J, Clements M,
Gebremedhin M, Guo N, Zhang Y, Duggan GE, Macinnis GD, Weljie AM, Dowlatabadi R,
Bamforth F, Clive D, Greiner R, Li L, Marrie T, Sykes BD, Vogel HJ, Querengesser L (2007)
HMDB: the human metabolome database. Nucleic Acids Res 35(Database issue):D521–D526
95. Sajed T, Marcu A, Ramirez M, Pon A, Guo AC, Knox C, Wilson M, Grant JR, Djoumbou
Y, Wishart DS (2016) ECMDB 2.0: a richer resource for understanding the biochemistry of
E. coli. Nucleic Acids Res 44(D1):D495–D501
96. Reher R, Kim HW, Zhang C, Mao HH, Wang M, Nothias LF, Caraballo-Rodriguez AM,
Glukhov E, Teke B, Leao T, Alexander KL, Duggan BM, Van Everbroeck EL, Dorrestein PC,
Cottrell GW, Gerwick WH (2020) A convolutional neural network-based approach for the
rapid annotation of molecularly diverse natural products. J Am Chem Soc 142(9):4114–4120
97. Egan JM, van Santen JA, Liu DY, Linington RG (2021) Development of an NMR-based
platform for the direct structural annotation of complex natural products mixtures. J Nat Prod
84(4):1044–1055
98. Wishart, D. S., Z. Sayeeda, Z. Budinski, A. Guo, B. L. Lee, M. Berjanskii, M. Rout, H. Peters,
R. Dizon, R. Mah, C. Torres-Calzada, M. Hiebert-Giesbrecht, D. Varshavi, D. Varshavi,
E. Oler, D. Allen, X. Cao, V. Gautam, A. Maras, Ella F. Poynton, P. Tavangar, V. Yang, Jeffrey
A. van Santen, R. Ghosh, S. Sarma, E. Knutson, V. Sullivan, Amy M. Jystad, R. Renslow,
Lloyd W. Sumner, Roger G. Linington and John R. Cort (2021). "NP-MRD: the natural prod-
ucts magnetic resonance database." Nucleic Acids Res
99. Mielko KA, Jabłoński SJ, Wojtowicz W, Milczewska J, Sands D, Łukaszewicz M, Młynarz
P (2020) Possible metabolic switch between environmental and pathogenic Pseudomonas
aeruginosa strains: 1H NMR based metabolomics study. J Pharm Biomed Anal 188:113369
100. Borchert AJ, Gouveia GJ, Edison AS, Downs DM (2020) Proton nuclear magnetic reso-
nance metabolomics corroborates serine Hydroxymethyltransferase as the primary target of
2-Aminoacrylate in a ridA mutant of salmonella enterica. mSystems 5(2):10–1128
101. Marshall DD, Halouska S, Zinniel DK, Fenton RJ, Kenealy K, Chahal HK, Rathnaiah G,
Barletta RG, Powers R (2017) Assessment of metabolic changes in mycobacterium smegma-
tis wild-type and alr mutant strains: evidence of a new pathway of d-alanine biosynthesis. J
Proteome Res 16(3):1270–1279
102. Betancur LA, Forero AM, Vinchira-Villarraga DM, Cardenas JD, Romero-Otero A, Chagas
FO, Pupo MT, Castellanos L, Ramos FA (2020) NMR-based metabolic profiling to fol-
146 R. M. Borges et al.

low the production of anti-phytopathogenic compounds in the culture of the marine strain
Streptomyces sp. PNM-9. Microbiol Res 239:126507
103. Carvalho FV, Fonseca Santana L, Diogenes A d SV, Costa SL, Zambotti-Villelae L,
Colepicolo P, Ferraz CG, Ribeiro PR (2021) Combination of a multiplatform metabolite pro-
filing approach and chemometrics as a powerful strategy to identify bioactive metabolites in
Lepidium meyenii (Peruvian maca). Food Chem 364:130453
104. Kellogg JJ, Cech NB (2020) Uncovering bioactive natural products via biochemometric
methodologies. In: Comprehensive natural products III, pp 271–279
105. Ory L, Nazih EH, Daoud S, Mocquard J, Bourjot M, Margueritte L, Delsuc MA, Bard
JM, Pouchus YF, Bertrand S, Roullier C (2019) Targeting bioactive compounds in natural
extracts - development of a comprehensive workflow combining chemical and biological
data. Anal Chim Acta 1070:29–42
106. Wen C, Wang D, Li X, Huang T, Huang C, Hu K (2018) Targeted isolation and identifica-
tion of bioactive compounds lowering cholesterol in the crude extracts of crabapples using
UPLC-DAD-MS-SPE/NMR based on pharmacology-guided PLS-DA. J Pharm Biomed Anal
150:144–151
107. Liu M, Grkovic T, Liu X, Han J, Zhang L, Quinn RJ (2017a) A systems approach using
OSMAC, Log P and NMR fingerprinting: an approach to novelty. Synth Syst Biotechnol
2(4):276–286
108. Fernand MG, Roullier C, Guitton Y, Lalande J, Lacoste S, Dupont J, Ruiz N, Pouchus YF,
Raheriniaina C, Ranaivoson E (2017) Fungi isolated from Madagascar shrimps - investiga-
tion of the aspergillus Niger metabolism by combined LC-MS and NMR metabolomics stud-
ies. Aquaculture 479:750–758
109. Chen L, Zhao X, Wu J, Liu Q, Pang X, Yang H (2020) Metabolic characterisation of eight
Escherichia coli strains including “Big Six” and acidic responses of selected strains revealed
by NMR spectroscopy. Food Microbiol 88:103399
110. Aries ML, Cloninger MJ (2020) NMR metabolomic analysis of bacterial resistance pathways
using multivalent quaternary ammonium functionalized macromolecules. Metabolomics
16(8):82
111. Liu Q, Wu JE, Lim ZY, Aggarwal A, Yang H, Wang S (2017b) Evaluation of the metabolic
response of Escherichia coli to electrolysed water by 1 H NMR spectroscopy. LWT Food Sci
Technol 79:428–436
112. Rona GB, Almeida NP, Santos GC Jr, Fidalgo TK, Almeida FC, Eleutherio EC, Pinheiro AS
(2019) (1) H NMR metabolomics reveals increased glutaminolysis upon overexpression of
NSD3s or Pdp3 in Saccharomyces cerevisiae. J Cell Biochem 120(4):5377–5385
113. Imai A, Lankin DC, Godecke T, Chen SN, Pauli GF (2020) Differentiation of Actaea species
by NMR metabolomics analysis. Fitoterapia 146:104686
114. Kroymann J (2011) Natural diversity and adaptation in plant secondary metabolism. Curr
Opin Plant Biol 14(3):246–251
115. Leiss KA, Choi YH, Verpoorte R, Klinkhamer PG (2011) An overview of NMR-based metab-
olomics to identify secondary plant compounds involved in host plant resistance. Phytochem
Rev 10(2):205–216
116. Nephali L, Piater LA, Dubery IA, Patterson V, Huyser J, Burgess K, Tugizimana F (2020)
Biostimulants for plant growth and mitigation of abiotic stresses: a metabolomics perspec-
tive. Meta 10(12)
117. Schauer N, Fernie AR (2006) Plant metabolomics: towards biological function and mecha-
nism. Trends Plant Sci 11(10):508–516
118. Zulak KG, Weljie AM, Vogel HJ, Facchini PJ (2008) Quantitative 1H NMR metabolomics
reveals extensive metabolic reprogramming of primary and secondary metabolism in elicitor-­
treated opium poppy cell cultures. BMC Plant Biol 8:5
119. Shaver AO, Gouveia GJ, Kirby PS, Andersen EC, Edison AS (2021) Culture and assay of
large-scale mixed-stage Caenorhabditis elegans populations. JoVE 171:e61453
6 Advances in Microbial NMR Metabolomics 147

120. Polozsányi Z, Kaliňák M, Babjak M, Šimkovič M, Varečka Ľ (2021) How to enter the state
of dormancy? A suggestion by Trichoderma atroviride conidia. Fungal Biol 125(11):934–949
121. Judge MT, Wu Y, Tayyari F, Hattori A, Glushka J, Ito T, Arnold J, Edison AS (2019)
Continuous in vivo metabolism by NMR. Front Mol Biosci 6:26
122. Koczula KM, Ludwig C, Hayden R, Cronin L, Pratt G, Parry H, Tennant D, Drayson M,
Bunce CM, Khanim FL, Gunther UL (2016) Metabolic plasticity in CLL: adaptation to the
hypoxic niche. Leukemia 30(1):65–73
123. Dalisay DS, Rogers EW, Edison AS, Molinski TF (2009) Structure elucidation at the nano-
mole scale. 1. Trisoxazole macrolides and thiazole-containing cyclic peptides from the nudi-
branch Hexabranchus sanguineus. J Nat Prod 72(4):732–738
124. Pintér G, Schwalbe H (2017) Unprecedented carbon signal enhancement in liquid-state NMR
spectroscopy. Angew Chem Int Ed 56(29):8332–8334
125. Delaglio F, Walker GS, Farley KA, Sharma R, Hoch JC, Arbogast LW, Brinson RG, Marino
JP (2017) Non-uniform sampling for all: more NMR spectral quality, less measurement time
<non-uniform sampling for all more NMR spectral quality, less measurement time.pdf>. Am
Pharm Rev 20(4):1–12
Chapter 7
Sample Preparation in Microbial
Metabolomics: Advances and Challenges

Heiter V. M. Boness, Hanna C. de Sá, Emile K. P. dos Santos,

and Gisele A. B. Canuto

Abbreviations

ADP Adenosine Diphosphate

AMP Adenosine Monophosphate
ATP Adenosine Triphosphate
BCA Bicinchoninic Acid
BSTFA N,O-bis(trimethylsilyl)trifluoroacetamide
C. Candida
CE Capillary Electrophoresis
CIMR Core Information for Metabolomics Reporting
E. Escherichia
ECF Ethyl Chloroformate
EI Electron Impact
EP Electroporation
FIE-HRMS Flow Infusion Electrospray High-Resolution Mass Spectrometry
GC Gas Chromatography
GC-MS Gas Chromatography-Mass Spectrometry
1
H NMR Proton Nuclear Magnetic Resonance
HEPES (4-(2-Hydroxyethyl)-1-Piperazineethanesulfonic Acid)
HILIC-MS Hydrophilic Interaction Chromatography-Mass Spectrometry
HEPES 4-(2-Hydroxyethyl)-1-Piperazineethanesulfonic Acid
HS Headspace
IS Internal Standards
LC Liquid Chromatography
LPS Lipopolysaccharides
MCF Methyl Chloroformate

H. V. M. Boness · H. C. de Sá · E. K. P. dos Santos · G. A. B. Canuto (*)

Department of Analytical Chemistry, Institute of Chemistry, Federal University of Bahia,
Salvador, BA, Brazil
e-mail: [email protected]

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 149
T. Pacheco Fill (ed.), Microbial Natural Products Chemistry, Advances in Experimental
Medicine and Biology 1439, https://doi.org/10.1007/978-3-031-41741-2_7
150 H. V. M. Boness et al.

MO Microorganisms
MS Mass Spectrometry
MSI Metabolomics Standards Initiative
MSTFA N-Methyl-N-(Trimethylsilyl)Trifluoroacetamide
NMR Nuclear Magnetic Resonance
O.D. Optical Density
PBS Phosphate-Buffered Saline
PGN Peptidoglycan
RI Retention Index
RPLC-MS Reversed Phase Liquid Chromatography-Mass Spectrometry
S. Saccharomyces
SFE Supercritical Fluid Extraction
SFE-CO2 Supercritical Carbon Dioxide Extraction
SPE Solid Phase Extraction
SPME Solid Phase Microextraction
TMCS Trimethylchlorosilane
TMS Trimethylsilylate
UV Ultraviolet
VOCs Volatile Organic Compounds

7.1 Introduction

The products of metabolism reflect the physiological state of an organism, influ-

enced by genetic characteristics in addition to environmental factors, lifestyle, and
diet. Metabolomics, one of the key strategies for studying systems biology, emerged
to help the understanding of the phenotype of complex organisms. This omics strat-
egy aims to determine metabolites (<1.5 kDa) produced and altered in living organ-
isms. Metabolomics analyzes are performed through comparative studies (test
versus control groups) and can be conducted using two approaches: untargeted and
targeted. Untargeted metabolomics, a global approach, aims to identify and semi-­
quantify the largest number of altered metabolites in the biological system. In turn,
targeted metabolomics, a hypothesis-driven approach, focuses on the quantitative
determination of a small set of selected metabolites or metabolic pathways [1, 2].
The two primary tools for metabolomics analysis are based on mass spectrometry
(MS) and nuclear magnetic resonance (NMR) detection. MS has been gaining
prominence in the last years due to good coupling with separation techniques, such
as liquid chromatography (LC-), gas chromatography (GC-), and capillary electro-
phoresis (CE-). These couplings provide broad coverage of the metabolome [3], the
set of all metabolites present in the biological organisms. The complex data gener-
ated are processed in bioinformatics tools, and statistical analysis is applied to
extract relevant biological information [1]. The accurate determination of metabolic
changes in living cells allows its performance in discovering biomarkers, develop-
ing or monitoring therapies, diagnosis, and prognosis, among other applications.
7 Sample Preparation in Microbial Metabolomics: Advances and Challenges 151

Among the various areas of the metabolomics field, the study of microorganisms
(MO) has grown in importance since they are used as model organisms in microbi-
ology [4, 5], and has been recently revised [6–9]. Microbial metabolomics has been
applied in different fields, such as disease control and mechanism, microorganism
identification, food safety, microbiome comprehension, development of new drugs,
microbial infection, and toxicity, among others [5]. Like other metabolomics stud-
ies, microbial metabolomics has the great challenge of accurately determining the
alterations of different classes of metabolites present in a variable concentration
range. The high complexity of the sample matrices requires careful handling and
preparation. In this chapter, we will discuss general aspects of sample preparation
workflow applied to microbial studies, highlighting the challenges for current meth-
ods and advances. The discussion will focus on untargeted and targeted methods.
Correlations with analysis and data normalization will also be presented according
to the sample type.

7.2 Metabolomics Sample Preparation Workflow

The sample type applied to metabolomics studies varies and depends on the objec-
tive of the work, as the analytical technique and the studied organism. Regarding
in vivo clinical investigations by using human and animal models, biofluids such as
urine, plasma/serum, saliva, and tissues are frequently used. For in vitro studies,
more specifically for MO, the complete understanding of the metabolism requires
knowledge of both the extracellular metabolome (exometabolome), also called met-
abolic footprinting, and the intracellular metabolome (endometabolome), defined as
metabolic fingerprinting [10–13].
The sample preparation is one of the most critical steps in the metabolomics
workflow to provide reliable results and good correlations with the physiological
changes or states of the studied organism [14]. The great variability of the metabo-
lome, which comprises a variable range of concentrations of metabolites with dis-
tinct physicochemical characteristics, combined with intracellular leakage and
metabolite degradation, makes the extraction process even more challenging for
microbial metabolomics [15, 16]. Sample preparation involves a series of complex
steps. Figure 7.1 presents an overview of the workflow focused on sample harvest-
ing and extraction of microbial samples. The procedures generally involve previous
considerations of cell growth and collection prior to the extraction itself.
Preconcentration, filtration, and derivatization steps are also usually associated to
the process, the latter being applied for metabolomics analysis using gas
chromatography-­ mass spectrometry (GC-MS). For a good performance of the
method, some prerequisites such as quenching efficiency, with instantly stopping
the metabolism without membrane damage, followed by efficient extraction, with
no degradation of metabolites, are necessary [12]. Several review articles are found
in the literature with excellent descriptions of the most applied sample preparation
procedures for metabolomics [16–19]. Still, others compilations focus on sample
152 H. V. M. Boness et al.

Fig. 7.1 Overall sample preparation workflow for microbial metabolomics. (“Created with
BioRender.com”)

preparation for microbial metabolomics [10, 14, 20–22]. In the following sections
of this chapter, each step of the sample preparation workflow applied to microbial
metabolomics will be discussed in detail.

7.2.1 Experimental Design and Pre-Analytical Factors

Samples collected non-clinically, without planning for experiments in metabolo-

mics, are susceptible to variations, such as temperature and time, during collection,
transport, and storage [23]. Variability in these pre-analytical factors, and sample
extraction and analysis, are considered pivotal and will directly affect the quality of
data and interpretation of results. Ideally, experiments should be conducted after
planning all the steps involved in the workflow; this includes the pre-analytical fac-
tors [23, 24]. The reliability of the metabolomics studies requires workflow stan-
dardization. In order to accomplish that and to homogenize metabolomics protocols,
the Metabolomics Standards Initiative (MSI) [25] and the Core Information for
Metabolomics Reporting (CIMR) [26] proposed a minimum information report for
metabolomics and more specific microbial metabolomics experiments. These docu-
ments provide important recommendations for protocol development and reporting,
7 Sample Preparation in Microbial Metabolomics: Advances and Challenges 153

helping to reduce errors and facilitate the repetition and comparison of experiments
in different laboratories. Thus, for cellular metabolomics, experimental steps such
as metabolic quenching, culture medium composition, cell number, and washing
must be carefully performed and standardized [24]. Next, important aspects of these
pre-analytical factors applied to microbial metabolomics will be detailed.

7.2.1.1 Cultivation Conditions

For cell cultures, an important step to be considered in the experimental design is

the preparation of the culture medium. Controlled conditions of temperature, pH,
substrate composition, light, and dissolved gases (CO2 and O2) are important factors
for cell growth [12] and, consequently, can affect the metabolome. For microorgan-
isms, cell cultures usually take place in bioreactors. Two methods are often used:
fed-batch and continuous. The fed-batch, more applied in industrial processes,
operates under fixed nutrient conditions. Since there is no replacement of nutrients
and the medium, the cellular environment changes continuously, and different
phases of cell growth and metabolic production will change over time. In continu-
ous mode (also designated as chemostat), there is a continuous renewal of the cul-
ture medium and constant removal of cell biomass. So, a steady state is always
reached, improving experimental repeatability [12, 27–29]. Since the reproducibil-
ity of the results is directly related to the composition and preparation conditions of
the cell culture, cell growth under continuous mode seems to be preferred in metab-
olomics studies [12, 14].
Little is known about the effect of the type of culture medium on the production
of secondary metabolites. However, it is known that different nutrients will influ-
ence the growth of microorganisms and consequently affect the metabolic content
[27]. Some authors have verified the need for supplementation to improve biomass
production. Mielko et al. [30] observed the need to add 0.5% glucose to the medium
on Miller’s LB broth agar for bacterial growth to obtain an adequate amount of
biomass for untargeted studies. Crnkovic and co-workers [31] verify the influence
of the content of nitrate and phosphate in biomass production in different cyanobac-
teria strains. A rich nitrate medium produces better secondary metabolites extrac-
tion, while the presence of phosphate produces different profiles depending on the
microorganism. The volatile organic compounds (VOCs) profile was evaluated in
Pseudomonas strains, in which the authors observed metabolic variation by supple-
menting Vogel’s broth with glucose or protein (egg white powder) [32]. Kim and
Kim (2017) compared the metabolic profile of Escherichia coli and Saccharomyces
cerevisiae grown in complex and minimal media. The authors found different intra-
cellular profiles in which MO accumulated energy storage metabolites on minimal
media, and in complex media, they grew faster. According to the available cell nutri-
ents, these results suggested a cellular adaptation mechanism [33]. The knowledge
of the effect on intracellular metabolism of different cultivation conditions and car-
bon sources can serve as an important starting point for developing bioprocess con-
trol [34] and metabolomics cell cultivation protocols.
154 H. V. M. Boness et al.

Another factor that affects the variability of results and proves to be fundamental
in metabolomics studies is the number of biological replicates. Biological replicates
are composed of a repetitive analysis of pooled individuals from a population or
individual samples from different cultivation flasks or bioreactors using the same
growth conditions. According to MSI, a minimum triplicate is required for cell sam-
ples, and quintuplicate is preferred [25]. Biological replicas are an important point
in experimental design once they include biological variance in the experiment,
improving statistical confidence, validation, and data interpretation.

7.2.1.2 Storage and Stability

Storage time and temperature of the samples, both for intracellular and extracellular
analysis, will depend on the type and stability of the metabolites. Metabolites are
generally better preserved at low temperatures (−80 °C). Thus, it is necessary to
guarantee that no chemical reaction or degradation will occur until the sample prep-
aration [22]. The storage time and stability of each sample type against analytical
procedures are challenging to predict. There is consensus that biological samples
should be stored at −40 or −80 °C for a few years (about 5 or 10, respectively). The
defrosting must be gradual and performed at low temperatures (on an ice bath, for
example). It also seems appropriate that the sample preparation should be performed
at least at 4 °C and that freeze-thaw cycles should be avoided [11, 24]. Particularly
for cells, the pellets integrity must be preserved at low temperatures, avoiding enzy-
matic action and, consequently, alterations in the organism’s metabolism. Although
it is known that extracellular medium samples are stable and that a simple lyophili-
zation would be sufficient to leave them at room temperature during storage, it is
also recommended to keep these samples in the dark and at −20 to −80 °C until
sample extraction and analysis [22].

7.2.1.3 Sample Harvesting

Metabolomics must produce a real response of the metabolic state of the microor-
ganisms compatible with the conditions of growth and cell culture at the time of
sample collection. Sampling has to be representative, reflecting a reliable metabolic
profile [19]. However, due to the action of enzymes, significant alterations can be
observed. To circumvent such issues, cellular metabolism must be stopped quickly
[21]. This step, essential in the pre-preparation and preparation of biological sam-
ples, is called metabolic quenching. Cell sampling is routinely carried out manually.
In order to perform a reproducible sampling, the analyst must be well trained and
work under controlled temperature conditions [21]. Thus, metabolic quenching is
achieved by placing the sample in liquid nitrogen or adding a quenching solution.
Some rapid sampling techniques combined with quenching have been developed in
recent years, mainly using the continuous mode of bioreactor operation [35–37].
7 Sample Preparation in Microbial Metabolomics: Advances and Challenges 155

These sampling techniques aim to reduce the turnover rates of energy metabolism
products, such as glucose-6-phosphate [12].
The cell incubation period directly influences the metabolites produced or altered
by MO. Since they are associated with the physiological state of the cell, different
metabolites are produced in each phase of cell growth (lag, exponential, stationary,
and decline phase) [27]. The lag phase is responsible for cell adaptation. In the
exponential phase, cells divide and grow. The stationary phase is a feature of deple-
tion of nutrients and carbon sources causing growth inhibition, followed by the
decline phase with cellular death or lyse [38]. The exponential growth phase seems
to be the most suitable for microbial metabolomics studies, in which there is greater
uniformity of microorganisms in terms of chemical composition, the number of
cells, and metabolic activity [30]. However, the best harvesting period must be
investigated, so growth curves must be prepared.
As metabolites are present in biological organisms in a variable concentration
range, in order to obtain a detectable amount of the chemical compounds, especially
those present in low concentrations, the cell density must be optimized [24, 39].
Indeed, the growth phase of the MO influences cell density. Halouska and co-­
workers [39] discuss their experiences for bacteria cultivation, in which for staphy-
lococcal and mycobacteria cultures, the optical density at 600 nm (O.D.600nm) of 1–2
and 3–7 is achieved from 15 to 50 mL (exponential phase) and 3–5 mL (stationary
phase) of media volumes, respectively. Since the content of biomass is directly pro-
portional to the concentration of intracellular metabolites, some researchers suggest
working with a minimum of 10 mg of biomass for microbial metabolomics [40].
Separating the pellet and culture medium is also an important step in harvesting.
The method must be efficient enough to separate intracellular e extracellular materi-
als. Generally, fast vacuum filtration and cold centrifugation are applied for this
purpose (Fig. 7.1). Pezzatti and co-workers [41] evaluated both fast filtration and
cold centrifugation methods in gram-negative Caulobacter crescentus. The authors
verify that fast filtration results in better extraction regarding the number and abun-
dance of extracted metabolites due to method speed and minimum metabolite leak-
age. Nevertheless, cold centrifugation is most recommended since its results in
better reproducibility of the data. After the separation of pellet and culture medium,
a washing step is required in the pellet to prevent extracellular metabolome carry-
over and contamination, which could result in low recoveries of some metabolites.
Cell culture media are a complex mixture of constituents containing sugars and
anions (such as Cl−, SO4−, and PO43−) in high concentration, which cause signal
interference in the proton nuclear magnetic resonance (1H NMR) spectra and elec-
trospray ionization suppression in MS [24]. In addition, some additives in the cul-
ture media, such as fetal bovine serum, also can cause ion suppression of analytical
signals in MS [17]. As well as the use of buffer solutions, like HEPES
(4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid), is a problem particularly for
NMR since it adds unwanted signals to the spectra [24].
Saline (NaCl 0.9% w/v) [42–46] and phosphate-buffered saline (PBS) solutions
[47–51] are commonly used for cellular wash. The methods involve three to six pel-
let washes to complete removal of the culture medium and interferents. Less usual
156 H. V. M. Boness et al.

methods, such as washing Azospirillum brasilense pellets with NFbHPN-­lactate

medium and water (1:3, v/v) [52] or the use of pure water at room temperature for
E. coli and S. cerevisiae [33] were also found. The use of saline solutions with
physiological ionic strength seems suitable since they preserve the integrity of cell
membranes, preventing intracellular losses. Bolten and collaborators [53] evaluated
the effect of different wash solutions and quenching in six gram-positive and nega-
tive bacteria. The authors found that washing with pure deionized water resulted in
90% of metabolite loss. On the other hand, they observed that a low ionic strength
solution results in severe metabolite leakage, especially for gram-negative bacteria.
Thus, the authors recommended using solutions with high ionic strength, such as
2.6% NaCl [53]. The wash step must also be evaluated and optimized when prepar-
ing microbial samples. This step must be carried out quickly before quenching,
preferably at low temperatures, improving the reliability of the measurements of the
intracellular metabolome.

7.2.1.4 Quenching

Metabolic quenching is a continuous process that ideally should be performed dur-

ing the sampling, transport, and extraction of metabolites in order to eliminate enzy-
matic activity, prevent metabolite turnover, and reduce degradation [19]. The
quenching process is nothing more than “exposing” the biological sample to severe
conditions that inactivate endogenous enzymes. Metabolism deactivation is chal-
lenging in cell samples and should be faster than metabolic changes. Usually,
quenching is performed by adding pure solvents or mixtures and applying extreme
conditions of pH and temperature [14, 19]. One of the major problems of these
procedures is the high susceptibility of the cell wall of some microorganisms and
the consequent leaching of intracellular metabolites to the extracellular medium,
causing significant losses of metabolic information [21].
Determining the efficiency of metabolic quenching is not a trivial task. An estab-
lished strategy uses energy metabolism phosphate species as a parameter since they
have fast turnover rates (half-lives <0.1 s) in cells [54]. Equation 7.1 is used to cal-
culate the efficiency:

Energy charge 
 ATP   0.5  ADP  (Eq. 7.1)
 ATP    ADP    AMP 
In which [ATP], [ADP], and [AMP] are the molar concentrations of adenosine
triphosphate, adenosine diphosphate, and adenosine monophosphate, respectively.
The ratio of 0.80 to 0.95 is expected to indicate an adequate quenching [17, 53, 54].
As mentioned before, one quenching strategy is adding quenching solutions to
the samples. Alkali (NaOH or KOH) or acidic (HCl or HClO4) solutions have been
the oldest quenching solutions applied in microbial metabolomics [19, 21].
Currently, this approach has fallen into disuse since its inefficiency has been proven
7 Sample Preparation in Microbial Metabolomics: Advances and Challenges 157

due to the easy rupture of the cell membrane and the occurrence of ionization sup-
pression by the use of non-volatile species. Cold organic solvents, pure or mixtures
[42–51], such as acetonitrile [49] and methanol [41, 55, 56] have shown greater
efficiency. Pre-cooled methanol is a suitable quenching solution as it is water-­
miscible, non-viscous, and has a low freezing point compared to other quenching
solutions (like glycerol and ethanol). Furthermore, it is a solvent compatible with
the analytical techniques used in metabolomics [19]. However, it is known that pro-
karyotic organisms (such as some bacteria) compared to eukaryotic organisms (fila-
mentous fungi and yeast) have different behaviors when exposed to methanol [12].
Different works have demonstrated that some bacteria and yeast species have plas-
matic membrane composition susceptible to pure or 60% methanol [57, 58]. Thus,
some buffer solutions, such as tricine, ammonium carbonate, and HEPES, are fre-
quently used as quenching in order to control ionic strength in the medium, preserv-
ing bacterial cell membrane [19]. Villas-Boas and Bruheim [59], developed an
alternative quenching method, using the glycerol-saline solution at −23 °C. The
authors state that although they could not determine whether there was intracellular
metabolite leakage, this method provides an excellent recovery of key metabolites,
such as glutamate, phosphoenolpyruvate, NAD+, and NADP+, in S. cerevisiae. The
use of the glycerol-saline solution for metabolic quenching is controversial. Glycerol
can be an important interferent because it is a highly hygroscopic compound, mak-
ing it difficult to carry out silylation reactions in derivatization procedures [59, 60].
Instead, alkylation methods could be used in metabolite derivatization and unaf-
fected by the remaining glycerol in the quenching solution [57]. However, it is
known that silylation reactions are more comprehensive in metabolic coverage than
alkylations.
Other quenching methods include fast filtration [33, 61–63] and liquid nitrogen
[47, 52, 64, 65]. Nevertheless, it is important to consider that liquid N2 (−196 °C)
can also favor the formation of crystals in the pellets, provoking rupture of the cell
membrane [19, 21]. The manual realization of fast filtration is subject to filter clog-
ging, resulting in long sampling periods and, consequently, alteration of cellular
metabolism [57]. Another problem with this practice is cellular exposure to mechan-
ical stress, which could also change the metabolic profile pattern [24]. Recently
published quenching protocols applied in microbial metabolomics are summarized
in Table 7.1.
As can be seen in Table 7.1, the quenching protocols are variable, especially
regarding the applied solutions. There is no consensus in the literature on the most
appropriate method, and many authors recommend a careful assessment of quench-
ing conditions and the development of protocols. In the last years, some researchers
have spent efforts on investigating and optimizing metabolic quenching in microbial
studies [61, 63, 68, 69, 74, 75]. Fast filtration and pure methanol (at −40 °C) were
recently evaluated for rapid quenching in yeast S. cerevisiae. The authors verified
that using organic solvent resulted in significant losses of intracellular metabolites
but good recovery of the glycolysis metabolic pathway intermediates. On the other
hand, fast filtration, besides reducing metabolite leakage and avoiding extracellular
contamination, proved more robust and more straightforward since it was applied in
158 H. V. M. Boness et al.

Table 7.1 Recent quenching protocols applied in microbial metabolomics

Quenching solution/protocol Microorganism Ref.
Bacterial metabolomics
Liquid nitrogen Saccharopolyspora erythraea [45]
filamentous Alphanizomenon flos-aquae [47]
Citrobacter brasiliense [48]
Azospirillum brasiliense [52]
Escherichia coli [66]
Fast filtration, liquid nitrogen, acetonitrile/ Escherichia coli [33]
water (1:1, v/v) (−20 °C)a
Methanol 40% (−20 °C) Corynebacterium glutamicum [55]
Methanol/water (8:2, v/v)a Caulobacter crescentus [41]
Methanol/acetonitrile/water (40:40:20, v/v/v) Escherichia coli [49]
with 0.1% formic acida
20% methanol/0.9% NaCl Lactobacillus plantarum [67]
60% methanol/0.9% NaCl and Bacillus licheniformis [68]
60% methanol/0.9% NH4CO3
Methanol/ethylene glycol Escherichia coli [69]
(45:55, v/v) (−60 °C)
40% ethanol/0.8% NaCl (−20 °C) Escherichia coli [70]
Fungi metabolomics
Liquid nitrogen Mortierella alpina [62]
Candida albicans [64]
Mortierella elongata [71]
Fast filtration, liquid nitrogen, acetonitrile/ Saccharomyces cerevisiae [33]
water (1:1, v/v) (−20 °C)a
Methanol 40% Saccharomyces cerevisiae and [70]
Kluyveromyces marxianus
60% methanol with Pichia pastoris [56]
0.11 mol/L NH4HCO3
Cold glycerol and cold methanol Yarrowia lipolytica [72]
Rapid cooling at −80 °C Candida tropicalis [73]
Quenching and extraction performed simultaneously
a

a vacuum manifold system [61]. Similar findings were found by Zheng and collabo-
rators [63] in Aspergillus niger. The authors observed a significant loss of cell integ-
rity with methanol (pure and mixtures) compared to fast filtration with liquid N2.
Wang and co-workers [68], on the other hand, evaluated five quenching agents,
varying the composition of methanol and salts (NaCl, NH4HCO3, and NaO2CCH3)
with other sample preparation conditions for Bacillus licheniformis. The authors did
not find a unique optimal condition. The results pointed to 60% methanol/0.9%
NH4HCO3 and 60% methanol/0.9% NaCl as the bests quenching solutions. The
authors claim that the presence of salts in the medium probably reduced the osmotic
pressure of the medium, leading to lower losses of intracellular metabolites [68].
7 Sample Preparation in Microbial Metabolomics: Advances and Challenges 159

Finally, the combination of hot and cold quenching and extraction solutions was
evaluated in Pseudomonas taiwanensis [75]. Combined with fast filtering, four
solutions, including mixtures of methanol, acetonitrile, ethanol, and water, were
tested. Quenching efficiency by applying Eq. 7.1, associated with metabolite quan-
tification, and data reproducibility showed boiling ethanol as the best quenching/
extractant solvent. As the authors point out, the metabolic profile is entirely affected
by the applied method [75], corroborating that there is no standard protocol for
quenching. Furthermore, studies prove that the best condition is organism-­
dependent. Thus, a continuous evaluation of this workflow step in each microbial
study is recommended.

7.2.2 Sample Extraction

Ideally, any extraction procedure should be as simple and effective as possible with-
out changing the physicochemical characteristics or inducing metabolite degrada-
tion. The extraction methods are planned around the metabolomics approach and
the analytical technique. For untargeted metabolomics, the metabolites extraction
must be fast, efficient, and obtain maximum metabolome coverage. For the targeted
approach, the extraction must be highly selective and efficient, with good recovery
of the metabolites of interest. Additional steps, such as solubilization in deuterated
solvents for NMR and derivatization for GC-MS, may be necessary [14, 19].
The extraction of metabolites from microbial samples will differ not only by the
approach and analytical technique used but, above all, whether the analysis is aimed
at the intracellular metabolome or those secreted by cells. In general, intracellular
analysis requires a more elaborate and time-consuming preparation, with mechani-
cal apparatus for cell lysis associated with the solvent extractant [10]. Extractants
are composed of organic solvents at high or low temperatures, pure or in mixtures.
The solubility of some metabolites is a limiting factor for extractions. More selec-
tive compositions are applied to targeted methods, while methanol, ethanol, and
chloroform are preferred for untargeted approaches. The use of organic solvents for
extraction, more specifically alcohols, is preferable in metabolomics methods. In
addition to protein precipitation and cell permeability, they are compatible with the
mobile phases used in LC methods, do not introduce undesirable salts into MS sys-
tems, and are easy to evaporate for preconcentration and GC derivatization [19].
The analysis of the extracellular metabolome involves centrifugation or filtration
and, for some studies, additional steps of protein precipitation and supernatant evap-
oration [10]. Other methodologies, more selective and applied to targeted methods,
such as solid phase extraction (SPE) and solid phase microextraction (SPME), can
also be applied. Details of the different extraction methods will be discussed in the
following subsections.
160 H. V. M. Boness et al.

7.2.2.1 Microbial Intracellular Extraction

Intracellular metabolites are contained within the cell, protected by the cell mem-
brane. Determining the concentration of intracellular substances is essential for
understanding the metabolic dynamics reflected in the phenotype of the biological
organisms [29]. Extracting intracellular metabolites involves exposing the cells to
extreme conditions to permeate the cell wall extracting as many metabolites as pos-
sible [10, 14, 19].
There is no standard or universal method for intracellular extraction in microbial
metabolomics. The choice will be highly influenced by the type of microorganism
studied, the metabolomics approach, and the analytical technique. Each type of MO
has complexity and variation in the composition of the plasmatic membrane [10, 14,
21]. For instance, bacteria have a cell membrane composed of multiple polymers,
and their type defines the MO as gram-positive or gram-negative bacteria [21]. The
cell wall of gram-negative is composed essentially of lipopolysaccharides (LPS),
while gram-positive bacteria have higher concentrations of peptidoglycan (PGN).
Such differences are reflected in cell wall resistance and chemical permeability.
Regarding mechanical resistance, the greatest thickness of the PGN layer in gram-­
positive bacteria provides higher resistance when compared to gram-negative bacte-
ria [14, 30]. On the other hand, the higher concentration of lipids in the cell wall in
gram-negatives provides greater resistance to solvents [20]. Yeasts have an intracel-
lular envelope composed of mannoproteins, which give protection against external
osmotic changes and provide mechanical resistance [20]. Thus, the strategies
applied for cell membrane disruption must be carefully evaluated according to the
MO studied.
As previously discussed in the subsection “Sample harvesting”, before to intra-
cellular extraction, the cells are separated from the culture medium by centrifuga-
tion or filtration, followed by pellet washing, to removal of extracellular constituents.
Intracellular extractions can be conducted through chemical agents or mechanical
processes, or most commonly applied, a combination of both.
Mechanical methods are high-efficient and low-cost physical processes. Although
widely applied in analytical chemistry, mechanical methods are difficult to use in
microbial samples due to low selectivity and simultaneous release of cytoplasmic
components to the medium, making the purification of samples difficult. The greater
extraction efficiency is observed when chemical methods are combined with
mechanical procedures [10, 20, 21]. Among the most applied mechanical methods
to extract MO are:
(a) High-pressure homogenization: cells are forced through narrow channels at
high speed. The samples are subjected to pressurization and, consequently, cell
lysis. The pressure exerted will depend on the cellular membrane resistance
[20]. The extraction conditions (biomass concentration, pressure, and the num-
ber of passes) were evaluated in a French press for E. coli membrane disruption.
The predictive models generated by machine learning analysis demonstrated
the system’s efficiency. The results corroborate with previous works, which
verify the minimal damage to the biochemical composition of this MO [76].
7 Sample Preparation in Microbial Metabolomics: Advances and Challenges 161

(b) Sonication: cells are disrupted by shock waves caused by high-frequency sonic
waves, resulting in cavitation. Due to the temperature increase caused by the
shocks, the procedures must be carried out in an ice bath to avoid metabolite
degradation. Sonication-based methods are inexpensive and selective for the
extraction of intracellular metabolites; for this reason, they are widely applied
in metabolomics [20]. Higher cell membrane permeabilization efficiency is
observed when this physical method is combined with organic solvents [77].
(c) Freezing-thawing: the cell membrane disruption occurs by volume expansion
due to the formation of crystals inside the cell [20]. Liquid N2 is often applied
in this process [43, 74]. Like sonication, freeze-thaw methods are better com-
bined with extractant solvents. As recently applied by Chen and co-workers,
which extract intracellular metabolites from E. coli using pure methanol (at
−20 °C) with three cycles of freezing with nitrogen and thawing in an ice
bath [78].
(d) Milling: the cells are squeezed and damaged by collision with the walls of the
container and beads particles added to the extractor medium. The beads of dif-
ferent sizes are made of glass or ceramic and are chosen according to the cell
size [20, 30]. For instance, a vibration mill apparatus associated with tungsten
carbide beads and the mixed solvent methanol/chloroform/water (2:1:1, v/v/v)
significantly increased the extraction efficiency in Mycobacterium tuberculo-
sis [79].
Chemical lysis is also frequently applied for extraction and is considered less
aggressive than mechanical methods but quite effective. Chemical lysis consists of
adding a chemical agent to the sample, promoting the extraction by permeabilizing
the extractant through the cellular membrane and, consequently, the concentration
of the metabolites in the liquid phase [10, 20, 21]. It is a low-cost method whose
efficiency depends on the correct choice of solvents and temperature in order to
maintain the chemical integrity of the extracted metabolites [67].
Different extractor chemical compositions can be used in the intracellular extrac-
tion of MO for applications in metabolomics. Acidic (HClO4) [80] or alkaline
(KOH) solutions were applied [81]. However, these extractors are being less used,
mainly due to low solubility and efficiency, in addition to the known problems of
analytical signal interference for NMR and ionization suppression in MS [19, 20,
29]. Thus, due to the lipid composition of the cell membrane of microorganisms,
organic solvents (such as methanol, acetonitrile, and ethanol) are more suitable for
metabolite extraction. These solvents, pure or in mixtures, are compatible with the
analytical techniques used in metabolomics and present adequate cell permeability.
Methanol at low temperatures (−20 °C) is considered the gold standard for intra-
cellular extraction in metabolomics and can be applied simultaneously as extracting
and quenching solutions. It is a solvent easily removed by evaporation, facilitating
the extraction of thermolabile metabolites, and has been widely applied in microbial
metabolomics as a pure extractant [50, 51, 66, 72, 82, 83] or in aqueous solution
[41, 47, 52, 56, 64, 68, 73]. Alternatively, aqueous acetonitrile [33, 61, 62], metha-
nol/acetonitrile [42, 48], and methanol/acetonitrile/water mixture [69] have also
162 H. V. M. Boness et al.

been used. Overall, the methods combine the use of those extractants with mechani-
cal devices, such as beads milling [61, 68, 84], vortexing [50], freeze-thaw cycles
[43, 48, 72, 85], and sonication [42, 52], increasing extraction efficiency. Methanol
also can be combined with other solvents such as chloroform or dichloromethane
[31]. Methods involving methanol/chloroform/water (3:1:1, v/v/v) [43, 46, 86] and
(2:1:1, v/v/v) [48, 79] mixtures applied to bacterial extractions have been shown to
be efficient for the extraction of polar and non-polar metabolites. Although highly
toxic, chloroform has the advantage of promoting the denaturation of enzymes,
interrupting metabolic activity, and favoring quenching while extracting metabo-
lites. Hot extraction methods, using boiling ethanol, are also found for gram-­
negative bacteria such as E. coli [69, 70] and Pseudomonas taiwanensis [75].
Heating the system increases the solvent capacity of cell lysis. However, this method
is not indicated for the extraction of thermolabile metabolites [21]. On the other
hand, intracellular metabolites of gram-positive bacteria such as Staphylococcus
aureus [87] and Bacillus licheniformis [88] were efficiently extracted using cold
ethanol solutions. Table 7.2 summarizes some recent works of microbial metabolo-
mics, highlighting intracellular extraction methods and applied analytical techniques.
Several studies have shown that each extraction method will provide a different
intracellular metabolic profile, regardless of the organism under study [30, 89].
Thus, some works focus on optimizing extraction parameters, like solvents and lyse
[41, 63, 69, 77, 90]. Overall, optimizations still involve systematic studies for
quenching and extraction, which may or may not occur simultaneously. A complete
optimization of the sample preparation workflow, comprising the evaluation of
quenching, washing steps, intracellular extraction, and metabolite derivatization for
the smut fungus Ustilago maydis was conducted by Phan and co-workers [77].
Regarding the extraction, ethanol/water (7:2, v/v), methanol/water (7:2, v/v), and
methanol/chloroform/water (5:2:2, v/v/v) solutions were studied. In addition, hot,
cold, and sonication extractions were combined with the solvents to improve cell
lyse. All methods were suitable with good reproducibility. The best choice was the
cold extraction with the ethanol solution, as it extracts the most polar portion of the
metabolome, being a simple, non-carcinogenic, and environmentally friendly
method [77]. Three quenching solutions were compared in combination with three
solvent mixtures for intracellular extraction in cultures of Acidithiobacillus ferro-
oxidans by Doran et al. [90]. The authors found different responses among the nine
combinations evaluated, in which acidic methanol/water as quenching solution
combined with isopropanol/methanol/water for extraction produced better results.
The authors also pointed out the need for studies of sample preparation conditions
according to the type of metabolites of interest [90].
The association of different methods or sequential extraction procedures also
influences the extraction efficiency [89]. Two successive extractions with pure
methanol and saline solution were sufficient to extract Klebsiella pneumoniae intra-
cellular metabolites efficiently [85]. For Candida albicans, the combination of two
solvents, 50% methanol followed by methanol/chloroform (3:1, v/v), was preferred
[64]. A recent study by Mielko and co-workers [30] evaluated the effect of sand mill
extraction, sonication, and tissuelyzer using methanol/water (1:1, v/v) as the
7 Sample Preparation in Microbial Metabolomics: Advances and Challenges 163

Table 7.2 Intracellular extraction methods for microbial metabolomics

Analytical
Organism Extraction Method Derivatization Technique Ref.
Oscillatoria sp., Methanol/dichloromethane – RPLC-MS [31]
Scytonema sp., (1:1 v/v) maceration at
and Nostoc sp. 4 °C. Supernatants were dried
in a vacuum and resuspended
in methanol containing IS,
followed by a clean-up step
using a C18 cartridge
Escherichia coli Acetonitrile/water (1:1 v/v) at 5 μL of 40 mg/mL of GC-MS [33]
and −20 °C and freeze in liquid N2 methoxyamine
Saccharomyces and thawed on ice. hydrochloride in
cerevisiae Supernatants were dried in a pyridine (90 min at
vacuum concentrator 30 °C). 45 μL of
MSTFA (30 min at
37 °C)
Penicillium Methanol/acetonitrile (1:1 v/v) – RPLC-MS [42]
oxalicum at −80 °C by 60 s vortexed
shocking and 30 min
cryogenic ultrasonic extraction
(2x). Supernatants were
lyophilized and reconstituted
in methanol/acetonitrile
(1:1 v/v)
Pseudomonas Cold methanol/chloroform/ – HILIC-MS [43]
aeruginosa water (3:1:1 v/v/v). Cell lyses
with freeze-thawing operation
in liquid N2 and vortexed (3x)
Acinetobacter Methanol/chloroform/water – HILIC-MS [46]
baumannii (3:1:1 v/v/v) at −80 °C
containing IS were freeze in
liquid N2 and thawed on ice
(3x)
Aphanizomenon Methanol/water (3:1 v/v) – RPLC-MS/ [47]
flos-aquae and homogenized at 35 Hz for MS
Microcystis 4 min and sonicated in an
aeruginosa ice-water bath for 5 min (2x).
Incubation at −40 °C for 1 h
and the supernatant dried. The
residue was resuspended in the
extraction solution containing
labelled internal standards
Citrobacter Methanol/acetonitrile/water – RPLC-MS [48]
sedlakii (2:2:1 v/v/v) and freeze in and
liquid N2 and thawed in a HILIC-MS
water bath at 37 °C (3x).
Supernatants were dried under
pure N2 and reconstituted in
aqueous acetonitrile (50%)
(continued)
164 H. V. M. Boness et al.

Table 7.2 (continued)

Analytical
Organism Extraction Method Derivatization Technique Ref.
Staphylococcus Cold methanol vortexed for – HILIC-MS/ [50]
aureus 1 min. After mixing with IS, MS
supernatants were dried on a
vacuum concentrator, followed
by resuspension in acetonitrile/
water (50:50 v/v)
Lactobacillus Methanol at −20 °C for – HILIC-MS/ [51]
acidophilus, L. 20 min. Supernatants were MS
fermentum, and dried and resuspended in
L. delbrueckii acetonitrile/water (50:50 v/v)
Azospirillum Aqueous methanol (80%) and 50 μL of 15 mg/mL of GC-MS [52]
brasilense sonication for 1 h. Supernatantmethoxyamine
was dried in N2 hydrochloride in
pyridine, (maintained at
50 °C until resuspension
of the residue). 50 μL
MTSFA +1% TMCS
(1 h at 50 °C)
Pichia pastoris Aqueous methanol (60%) with – 1
H NMR [56]
0.11 mol/L of ammonium
bicarbonate quickly vortexed
and freeze-thawing operation
in liquid N2 and ultrasonic
bath for 15 min. Supernatants
were dried under reduced
pressure and reconstituted in
NMR buffer
Saccharomyces Acetonitrile/water (1:1 v/v) 10 μL of 40 mg/mL of GC-MS [61]
cerevisiae and glass beads vortexed for methoxyamine
3 min hydrochloride in pyridine
(90 min at 30 °C and
200 rpm). 45 μL of
MSTFA (30 min at 37 °C
and 200 rpm)
Actinosynnema Cold aqueous acetonitrile 200 μL of NaOH 1% GC-MS [62]
pretiosum (50%) and freeze in liquid N2, for residue dissolution
thawed on ice, and vortexed. and mixture with
Supernatants were mixed with 167 μL of methanol and
IS and dried in a vacuum 34 μL of pyridine.
concentrator Addition of 20 μL of
MCF and vortexed for
30 s (2x). Resulting
solutions were mixed
with 400 μL of
chloroform and 400 μL
of NaHCO3
(50 mmol/L). Organic
supernatant phases were
dried by adding
anhydrous Na2SO4
(continued)
7 Sample Preparation in Microbial Metabolomics: Advances and Challenges 165

Table 7.2 (continued)

Analytical
Organism Extraction Method Derivatization Technique Ref.
Candida albicans Aqueous cold methanol (50%) 50 μL of 20 mg/mL of GC-MS [64]
(30 min at −20 °C). methoxyamine
Re-extraction with methanol/ hydrochloride in
chloroform (3:1 v/v). pyridine (2 h at 37 °C).
Combined supernatants were 50 μL of MSTFA +1%
lyophilized TMCS (2 h at 37 °C)
Escherichia coli Methanol at −20 °C vortexed – RPLC-MS [66]
for 1 min. Cell lyse with and
freeze-thawing operation in HILIC-MS
liquid N2 (3x). Re-extraction
with identical conditions.
Combined supernatants were
dried and resuspended in
acetonitrile/water (50:50 v/v,
with 0.1% formic acid)
Escherichia coli, Boiling aqueous ethanol (75%) – RPLC-MS/ [70]
Saccharomyces and IS vortexed and incubated MS
cerevisiae, and at 95 °C for 3 min. Supernatants
Kluyveromyces were dried under a vacuum and
marxianus resuspended in water
Candida Cold methanol/water 25 μL of BSTFA +1% GC-MS [73]
tropicalis (60:40 v/v) incubated for TMCS, immersion in a
30 min at −20 °C, followed by flux of air heated for 1 h
re-extraction with methanol/ at about 100 °C
chloroform (3:1 v/v). Organic
supernatant phases and second
extraction supernatant were
dried under N2 stream
Escherichia coli Boiling in perchloric acid. Water – 1
H NMR [80]
content supernatants were
mixed with deuterated water
Staphylococcus Cold methanol vortexed for – HILIC-MS [83]
aureus 1 min and addition of labelled
IS (incubation for 20 min at
−20 °C). Supernatants were
dried under a vacuum and
reconstituted in aqueous
acetonitrile (50%)
Escherichia coli Cold methanol/acetic acid/ – RPLC-MS [84]
water (4:4:2 v/v/v) containing
IS. Homogenization with a
ball mill (40 Hz for 5 min) and
sonicated (5 min). Protein
precipitation by incubation for
1 h at −20 °C. Supernatants
were dried under a vacuum
and reconstituted with aqueous
acetonitrile (50%), vortexed
(30 s), and sonicated (30 s at
4 °C for 10 min)
(continued)
166 H. V. M. Boness et al.

Table 7.2 (continued)

Analytical
Organism Extraction Method Derivatization Technique Ref.
Klebsiella Methanol at −20 °C and sterile – 1
H NMR [85]
pneumoniae saline solution vortexed,
followed by freeze at −80 °C
for 30 min and thawed on ice
for 30 min (3x). Re-extraction
with identical conditions.
Combined supernatants were
evaporated under reduced
pressure and reconstituted in
NMR buffer
Mycobacterium Methanol/chloroform/water 5 μL of 20 mg/mL of FIE-HRMS [86]
smegmatis (3:1:1 v/v/v). Four freeze-thaw methoxyamine and GC-MS
cycles and re-extraction using hydrochloride in
identical mixture solvents. pyridine (90 min at
Supernatants were combined 30 °C). 10 μL of
and divided for FIE-HRMS MSTFA (30 min at
and GC-MS analysis. GC-MS 37 °C)
extracts were dried under a
vacuum
Staphylococcus Aqueous ethanol (60%) at – RPLC-MS [87]
aureus −20 °C vortexed 10x (10 s at and
4 °C). Cell lysis in a Precellys HILIC-MS
homogenizer 3x (30 s at
3800 rpm at 4 °C).
Supernatants dried under N2
stream and solubilized in
10 mmol/L ammonium
carbonate pH 10.5 containing
internal standards
Bacillus Cold aqueous ethanol (75%) 100 μL of 20 mg/mL of GC-MS [88]
licheniformis containing IS, freeze in liquid methoxyamine
N2 for 3 min, thawed on ice, hydrochloride in
and vortexed for 5 min (3x). pyridine (1 h at 37 °C).
Supernatants were dried 100 μL of MSTFA (3 h
at 60 °C)
BSTFA N,O-bis(trimethylsilyl)trifluoroacetamide, FIE-HRMS flow infusion electrospray high-­
resolution mass spectrometry, HILIC-MS hydrophilic interaction chromatography-mass spectrom-
etry, IS internal standard, MSTFA N-methyl-N-(trimethylsilyl)trifluoroacetamide, RPLC-MS
reversed phase liquid chromatography-mass spectrometry, TMCS trimethylchlorosilane

extractant in six bacterial strains (gram-positive and gram-negative). The authors

found no significant differences in the qualitative metabolic profile of the microor-
ganisms. Although, some metabolites were extracted in higher concentrations
depending on the organism and mechanical apparatus used. For instance, sand mills
were more effective for Enterococcus faecalis, while tissuelyzer extracted higher
concentrations for Bacillus cereus. This study proves that there is no method more
suitable for global metabolomics, and the findings serve as good guides for future
7 Sample Preparation in Microbial Metabolomics: Advances and Challenges 167

targeted studies. Finally, a less usual solvent, such as methyl t-butyl ether, was also
investigated to extract Mortierella alpine [65] and Pseudomonas aeruginosa [44].
However, it demonstrated low extraction efficiency against solvents based on mix-
tures of methanol/water [65] and methanol/chloroform/water [44].
Conventional intracellular extraction methods are not considered “green meth-
ods” since they use large volumes of toxic solvents, generating hazardous waste
[91]. An attractive and environmentally friendly alternative is supercritical fluid
extraction (SFE), which uses a supercritical fluid, such as nitrous oxide, carbon
dioxide, or xenon, as the extracting solvent. Supercritical carbon dioxide extraction
(SFE-CO2) is the most applied since CO2 is a recurrent and abundant source of vari-
ous technological processes, such as fermentation and combustion [92]. In addition
to being considered a green technique, SFE works at relatively low temperatures
due to the high pressure applied to the system, increasing extraction efficiency. On
the other hand, it requires more expensive instrumentation than conventional meth-
ods. SFE has consolidated applications in the area of ​​natural products from plants.
In microbial natural products, it has been used essentially in the investigation of
microbial activity of cyanobacteria [92–94]. SFE was applied in the extraction of
myxobacterial cells by Bader and co-workers [95]. In this work, different co-­
solvents and SFE-CO2 were evaluated by targeted and untargeted metabolomics
with the microbial activity of the extracts. The SFE proved to be a good technique
for the extraction and elucidation of new secondary metabolites, and although little
explored in metabolomics, it presents a promising tool.

7.2.2.2 Microbial Extracellular Extraction

Extracellular metabolites produce a response that reflects the internal metabolic

state of a microorganism in response to environmental conditions and the culture
medium [6, 22, 96]. Footprinting analysis has been applied in several areas such as
bioprocess involving the monitoring of fermentation [34, 97], mineralization of
xenobiotic compounds (bioremediation) [98], biomarker discovery [56, 99], food
control [100], and degradation processes by microorganisms [101, 102]. Finally,
this approach became popular when associated with genomics, allowing the differ-
entiation of microbial strains [103, 104].
Extracellular metabolites originate from cellular secretion processes, cell leak-
age, or media composition. The changes result from factors associated with the
cellular growth environment, such as temperature, pH, and nutrients, among others
[6, 22]. Little is known about the mechanism of metabolite secretion, and one of the
explanations is based on the metabolic overflow theory [22, 27]. This theory is
based on the premise that an extracellular metabolite will increase when its levels
are also increased intracellularly. However, some researchers disagree. Granucci
et al. [105] performed a time-series metabolomics study in Saccharomyces cerevi-
siae growing on a glucose-limited culture medium under aerobic and anaerobic
conditions. Aspartate, 2-oxoglutarate, benzoate, and decanoate showed opposite
regulation levels in intracellular and extracellular analyzes, suggesting an
168 H. V. M. Boness et al.

anti-­metabolic overflow behavior. This work is one of the few systematic studies
that evaluate the dynamic production of intra and extracellular metabolites. Thus,
drawing conclusions about this mechanism is not an easy task, requiring more com-
prehensive investigations.
Extracellular metabolomics analysis is conducted with rapid preparation meth-
ods in a few steps (Fig. 7.1) [22]. In addition, the quenching step is unnecessary
since there are no metabolic processes in the culture medium [10]. The exometabo-
lome has low concentrations of metabolites compared to the intracellular metabo-
lome and lacks enzymatic activity [27]. Removal of cellular debris is important for
extracellular evaluation since suspended particles can cause clogs in the chromato-
graphic systems and contaminate the ionization sources, impairing the analysis [10,
22]. Centrifugation and filtration can be performed concurrently. A simple centrifu-
gation and supernatant collection procedure were performed for direct analysis of
the exometabolome of Acinetobacter baumannii by hydrophilic interaction
chromatography-­mass spectrometry (HILIC-MS) [46]. However, depending on the
objective of the metabolomics analysis and the analytical technique used, it is nec-
essary to apply additional steps, such as desalination, deproteinization, pre-­
concentration, or evaporation [6]. For instance, sample preparation for extracellular
analysis of Staphylococcus aureus cultures involved precipitation of proteins using
methanol [82]. Jiang et al. [42], in their evaluation of the influence on the production
of extracellular acids in Penicillium oxalicum with different phosphorus sources,
used a 0.22 μm polytetrafluoroethylene (PTFE) membrane filter after culture cen-
trifugation. Zhong et al. [49] in footprinting and fingerprinting study of E. coli,
evaluated the extraction step using five organic solutions with variable pH and com-
position. A mixture of methanol/acetonitrile/water (40:40:20, v/v/v) with 0.1% for-
mic acid demonstrated better results. This composition ensured good precipitation
of proteins, extracted more metabolites, and showed less variability between repli-
cates [49]. Salt removal and protein precipitation methods use large volumes of
solvents and are time-consuming. In addition, they produce low analyte recovery,
making them unpopular for metabolomics analysis [22]. To circumvent these issues,
SPE and SPME have been applied to microbial metabolomics and will be discussed
in the next subsection.
Additional concentration steps may be required for metabolomics footprinting
analysis to improve detection due to metabolite dilution in the culture medium. The
pre-concentration will allow the direct analysis or derivatization of the metabolites,
depending on the analytical technique used. Vacuum evaporation or drying systems
are the most widespread, as observed by Tredwell et al. [56], in which the cell
medium of Pichia pastoris was dried under reduced pressure before 1H NMR analy-
sis. The nitrogen stream has also been applied efficiently [106]. Finally, an alterna-
tive is a freeze-drying method, in which samples are quickly frozen and then dried
by sublimation. This method is quite efficient for concentration; however, extra care
must be taken to avoid the degradation of labile metabolites. Some specificities,
such as susceptibility to oxidation and high content of sugars or organic solvent,
must be considered before this sample concentration procedure [22].
7 Sample Preparation in Microbial Metabolomics: Advances and Challenges 169

7.2.2.3 SPE and SPME

Solid phase extraction (SPE) is an extraction and preconcentration procedure with

excellent precision and sensitivity. The method is similar to chromatography. The
separation is based on the difference between the retention of chemical species on a
solid sorbent and elution through a suitable mobile phase. Different solid phases are
commercially available, such as silica, ion exchange species, polymers, and carbon-­
based. The sorbent of choice will be carried out according to the physicochemical
characteristics of the metabolites of interest. Modulations on mobile phase compo-
sition in terms of polarity, pH, and ionic strength will aid in separating interferents
from trapped metabolites in the SPE cartridge [22, 107].
In metabolomics, SPE is applied mainly in targeted analyses, in which selected
metabolites or chemical classes are isolated from a liquid sample. Also, it is used to
extract non-volatile and semi-volatile analytes [22]. A great advantage of SPE, in
addition to the selective extraction and removal of interferents, is its ability to con-
centrate the metabolites in low concentrations in the biological samples. It also
dramatically decreases ionization suppression effects by reducing matrix compo-
nents. On the other hand, its application for untargeted analyzes is not advisable due
to the high selectivity [22]. A few microbial metabolomics works use SPE in the
preparation of cellular samples. The optimization and validation of targeted extrac-
tion methods for phosphorylated sugars from Saccharopolyspora erythraea were
performed by Hong and co-workers [45]. Different intracellular extraction methods,
including grinding and freeze-thawing in liquid N2, and boiling ethanol, followed
by the selection of the phosphorylated sugars using a quaternary amine silane cova-
lently bonded SPE column, were evaluated. In addition, quantification of Acyl-­
CoAs was performed after extraction in polymeric reversed-phase SPE column,
showing good recovery (above 82%) for six analytes. Crnkovic and collaborators
[31] applied C18 cartridges and methanol to cleanup extracts of methanol/dichloro-
methane (1:1, v/v) from three filamentous freshwater cyanobacteria prior to LC-MS
analysis.
Solid phase microextraction (SPME), invented in 1990 by Dr. Pawliszyn and co-­
workers, is a portable and environmentally friendly technique for extracting small
amounts of samples (below 100 μL) [108, 109]. In SPME, coated fibers in a micro-­
syringe are exposed to a sample until an equilibrium is established. The extraction
occurs through the partition or adsorbing of the analytes in the fiber when it is
coated with a high molecular weight polymeric liquid or a solid sorbent, respec-
tively. Finally, the analytes (metabolites) are taken for desorption and analyzed. One
of the great advantages of SPME is the simultaneous sampling and extraction of
volatile analytes present in low concentrations (up to ppt), both in solid and liquid
samples [22]. Like SPE, SPME is highly selective and therefore better suited to
targeted metabolomics analysis.
SPME can be combined with headspace (HS-SPME), which collects analytes
after volatilization by heating in the headspace, and is an excellent alternative for
GC-MS metabolomics analysis [22, 107] The HS-SPME is a solvent-free method
that analyzes volatile compounds without derivatization. The methods require less
170 H. V. M. Boness et al.

extraction time than direct SPME and when performed in automated systems,
present highly reproducible results [22, 109]. HS-SPME has been extensively
applied in the analysis of VOCs in microbial metabolomics [110–115], with the
advantage of not requiring elaborate sample preparation procedures [116]. VOC
profiles were evaluated in seven mycobacteria species for diagnostic purposes.
HS-SPME using a Polydimethylsiloxane/Carboxen/Divinylbenzene fiber was
used to extract the metabolites, proving to be a fast methodology (20 min of fiber
exposure) and adequate in the selection of discriminatory metabolites among bac-
teria strains [110]. An interesting work on the differentiation of metabolic profil-
ing of biothreat agents was carried out by Dailey et al. [117]. Liquid cultures of
Francisella tularensis, Burkholderia pseudomallei, Brucella melitensis, and
Yersinia pestis were analyzed by gas chromatography with a flame ionization
detector (GC-FID). VOCs were extracted with an HS-SPME multifiber using
fibers based on Carboxen/Polydimethylsiloxane, Polydimethylsiloxane,
Polyethylene Glycol, Polydimethylsiloxane/Divinylbenzene, Divinylbenzene/
Carboxen/Polydimethylsiloxane, and Polyacrylate. Using the multifiber method,
the authors drastically reduced the extraction time (approximately 2 h in tripli-
cate), improving the analytical frequency for sample preparation and increasing
the coverage of VOC analyses by using the different fibers.

7.2.2.4 Derivatization

GC-MS is a powerful analytical technique successfully used to analyze microorgan-

ism metabolomes [9]. This technique combines the high-resolution power of gas
chromatography with the sensitivity and versatility of mass spectrometry detection.
In addition, GC-MS coupling facilitates the accurate identification of hundreds of
metabolites in a single run. Due to the hard ion source, the electron ionization tech-
nique (EI), the system generates reproducible mass spectra, enabling libraries for
metabolite identification [118]. However, one of the utmost limitations of this tech-
nique is the need for volatility and thermal stability of the analytes. Not all metabo-
lites, such as sugars, are volatiles [7]. Thus, in order to increase the volatility, an
additional derivatization step must be included in the sample preparation workflow
(Fig. 7.1). Two types of derivatization have been employed in metabolomics stud-
ies: silylation and alkylation procedures.
Silylation, the classical procedure, has been mostly applied in metabolomics
because it encompasses various derivatized species, such as some amino acids, sug-
ars, and sugar alcohols [60]. This method can be performed in two steps, oximation
(or methoximation) followed by the silylation itself. The oximation is a reaction in
an anhydrous medium whose aim is to protect ketone and carbonyl groups, inhibit
the cyclization of reducing sugars, and the formation of keto-enol tautomers (of
aldehydes and ketones). Silylation is responsible for increasing metabolite volatility
by the replacement of acidic hydrogen (-OH, -COOH, R’NH2, -SH) by a trimethyl-
silylate (TMS) group (-Si(CH3)3) [118, 119]. In metabolomics, the N,O-­
bis(trimethylsilyl)trifluoroacetamide (BSTFA) and the N-methyl-N-(trimethylsilyl)
7 Sample Preparation in Microbial Metabolomics: Advances and Challenges 171

trifluoroacetamide (MSTFA) are the most silylation reagents used [60]. When
silylation does not occur rapidly, and hydroxyl groups do not react directly, 1%
(v/v) of trimethylchlorosilane (TMCS) is used as a catalyst [119].
Some variations in the derivatization reagents and protocols are found in recently
published works, as presented in Table 7.2. In general, the methods include two
steps, ranging from 2 to 4 h of procedure at 30 or 37 °C [33, 61, 86], as observed in
the analyzes of filamentous fungus Mortierella alpine [65], E. coli and S. cerevisiae
[33], and Bacillus licheniformis [88]. Higher temperatures for oximation (50 °C)
and silylation (60 °C) are also found in protocols for Mycobacterium tuberculosis
analysis [120] and Azospirillum brasilense [52]. Longer derivatization procedures
are also found, in which the oximation step was performed in 16 h at room tempera-
ture for derivatization of Beauveria bassiana [121]. Works involving only the
silylation step are less usual, mainly due to the advantages mentioned above of the
oximation step. Even so, standardized protocols are using BSTFA with TMCS
reacting for 2.5 h at 95 °C in the derivatization of metabolites present in polymicro-
bial biofilms of Candida albicans-Klebsiella pneumonia [106, 122], and 30 min at
50 °C of MSTFA with 1% TMCS for extracellular metabolomics of Candida
auris [123].
Different derivatization methods are notable, and their effectiveness is also a
crucial factor in metabolomics sample preparation workflow, impacting directly on
the reproducibility of the results [60]. Some metabolites, such as amines, amides,
and phosphates, exhibit poor accuracy and recovery after silylation. This is due to
incomplete reaction and the formation of artifacts, which impair the quantification
of these metabolites [124, 125]. Thus, some preventive measures such as the addi-
tion of excess reagent and raise temperature during silylation increase the reaction
kinetics, promoting better derivatization [118]. Therefore, it is necessary to opti-
mize the reaction parameters in order to obtain the best results for different micro-
organisms. Wang and collaborators [68] evaluated different times, from 0.5 to 2.5 h,
for MSTFA derivatization of intracellular extracts of Bacillus licheniformis. The
results showed that there was no significant changes in the quantification of metabo-
lites after 2 h of derivatization and short times, such as 0.5 h, and are not enough to
detect several compounds such as nucleotides, sugars, and some organic and
fatty acids.
Alkylation methods have been used as derivatization alternatives. Villas-Bôas
and coworkers developed a rapid derivatization reaction with methyl chloroformate
(MCF) [126]. This reaction converts primary and secondary amines and carboxyl
groups into volatile carbamates and esters. The great advantage over conventional
silylation is its speed. The reactions occur at room temperature in approximately
1 min, favoring automated methods. Furthermore, unlike silylation, which requires
an anhydrous medium, MCF derivatization works in an aqueous phase with rela-
tively inexpensive reagents [40]. Alkylation was compared to silylation (using
MSTFA) in the derivatization of extracts of Clostridium proteoclaticum and
Acidovorax temperans. The silylation proved effective for detecting alcohols, pre-
nols, and organic acids (fatty, mono and dicarboxylic). On the other hand, it showed
low recovery for nucleotides, amino and non-amino organic acids. Alkylation
172 H. V. M. Boness et al.

proved to be a suitable method to be used as a complement to silylation, enabling

the improvement of metabolic coverage [127]. In microbial metabolomics, MCF
derivatization has been applied to intracellular sample preparation of S. cerevisiae
[128] and Actinosynnema pretiosum [62], intra and extracellular analysis of
Yarrowia lipolytica [72], among others.
One of the great disadvantages of the derivatization methods is the manual exe-
cution, being, therefore, a laborious and time-consuming methodology, subject to
many errors, especially from inexperienced analysts. To circumvent such issues,
automated systems for online derivatization have been developed [129–131].
Robotic systems have modular stations containing shakers, temperature controllers,
and syringes to inject previously derivatized samples. Derivatization steps are pro-
grammed, including the addition and volume of reagents, as well as reaction times
and temperatures, improving the integrity and reproducibility of GC-MS data [132].
An automated system was applied to derivate intracellular metabolites for untar-
geted analysis of Candida albicans. GERSTEL MPS autosampler, a commercially
available robotic workstation, was used, and oximation and silylation steps were
performed in 4 h [64]. This commercial autosampler was also used to develop a new
high-throughput MCF and ECF (ethyl chloroformate) derivatization method applied
to human biofluids (such as urine, serum, and feces) and cell samples of E. coli,
enabling the preparation and analysis of the samples in a few minutes. The combi-
nation of the two derivatization solvents facilitates the identification of the metabo-
lites in the spectra library since the comparison was performed by the fragments of
the MCF/ECF adducts and the retention index (RI). This method proved to be accu-
rate in metabolite assignment and is interesting in uncovering unknown metabo-
lites [72].

7.3 Metabolomics Analysis and Sample Normalization

Size, weight, or volume of samples can significantly affect the content of metabo-
lites extracted in a metabolomics experiment. Thus, in addition to the variations
introduced by analytical factors, from sample preparation to analysis and data pro-
cessing, the biological variations of each sample are sources of error and variability
between measurements. Thus, a normalization step should be included in metabolo-
mics experiments. Normalization aims to minimize variations in the amounts of
metabolites between sample groups and highlight inter-group variations [133],
making the data comparable within a studied context.
As mentioned before, there is a great challenge in standardization in metabolo-
mics, given a large number of metabolites present in complex organisms, in a vari-
able concentration range. As recommended for any metabolomics protocol, adding
internal standards (IS) to samples has also been a practice. The IS can correct vari-
ability resulting from errors during sample preparation, incomplete derivatization of
analytes, and ionization inefficiency or instrumental instability. The addition of
isotope-labeled IS has gained strength in the last years, especially for untargeted
7 Sample Preparation in Microbial Metabolomics: Advances and Challenges 173

analysis, due to the chemical complexity of the samples and the difficulty of choos-
ing an adequate IS.
Other commonly applied normalization methods include post-analysis strate-
gies, such as median, sum of total abundance, among others. Although interesting,
they are recommended for biological samples whose volume or weight does not
guarantee reproducibility of sample collection. Regarding to cellular metabolomics,
such as in microbial metabolomics, some normalization strategies are more appro-
priate, such as cell number or cell density, biomass, and total protein content. Cell
counting is a complex and inconvenient practice in the case of MO due to the small
size of these organisms. Thus, determining optical density at 600 nm (OD600) is a
viable alternative for data normalization [43, 46, 50]. The OD is measured in a spec-
trophotometer in which radiation at 600 nm passes through the culture medium. The
absorption measurement estimates the number of cells [133].
The determination of dry or fresh biomass has been a strategy frequently applied
in normalization in MO studies [31, 34, 52, 72]. While effective, these strategies are
time-consuming. Intercurrences such as long drying periods and errors in weighing
small masses are observed in the dried samples. In the case of fresh biomass,
depending on the variation of the water content present in the samples, the normal-
ization becomes less accurate. Lu et al. [65] evaluated normalization procedures
using fresh and dried biomass of filamentous fungi Mortierella alpina and found
better reproducibility and recovery using freeze-dried biomass values. Normalization
based on the total protein content is performed using well-established methods such
as the Bradford assays and the bicinchoninic acid (BCA assay) [48]. However, such
procedure has been questioned in metabolomics since given the complexity of bio-
logical systems, protein levels and expression do not directly correlate with concen-
trations of metabolites in microorganisms [134].
Alternative normalization methods involve the use of labeled compounds. Wu
and Li [133] present the dansylation metabolite assay for normalization applied to
E. coli. In this method, the metabolites are labeled with 12C-dansyl chloride and
measured at 338 nm using an LC with ultraviolet (UV) detection. The total concen-
tration of labeled metabolites allows an adjustment in the sample volume to perform
the metabolomics analysis, performing normalization before data acquisition. The
authors claim that the method is simple, fast, and inexpensive and may be a good
alternative to the analog BCA assay. Considering the particularities of each method
and type of sample worked, it is important to emphasize that the normalization step
must be considered in the metabolomics workflow, ensuring quality in interpreting
the results.

7.4 Advances in Microbial Sample Preparation

Conventional sample preparation methods for microbial metabolomics have been

described in this chapter. One of the major challenges in the area is the lack of stan-
dard protocols, especially for intracellular analysis, due to the differences in
174 H. V. M. Boness et al.

composition and resistance of cell membranes of different MOs. In recent years,

high-throughput extraction techniques have been developed and applied to micro-
bial metabolomics to increase the analytical frequency and produce reliable and
reproducible results.
Microfluidics technologies are a prominent field in cell biology and have been
used as alternative extraction methods. Microfluidic devices are particularly useful
as they require low volumes of solvents, producing reproducible cell handling and
low dilution of the samples [135, 136]. Methods based on electroporation (EP) have
recently been applied to extract metabolites from bacteria and yeast [137–139]. In
these methods, the cells were exposed to an electric field with sufficient strength and
duration to cause ruptures by the formation of pores in the membrane and conse-
quently promote the extraction of intracellular metabolites. Rockenbach and col-
laborators [138] proposed a chip-based setup using irreversible EP to extract
intracellular metabolites from Escherichia coli and Saccharomyces cerevisiae. Low
values of electric field strengths (below 10 kV/cm) were efficient for complete cell
lysis. Li and co-workers [140] developed the highly sensitive single-cell MS-analysis
based on electromigration and EP to manipulate ultrasmall size cells. The metabolic
profile was characterized for S. cerevisiae and other unicellular organisms. When
evaluating aspects such as heterogeneity between single cells and metabolic altera-
tions induced by external stimuli, the authors verified that the developed strategy is
highly effective.
Automated microscale procedures have also been developed for microbial
metabolomics. Footprinting analysis of Pseudomonas putida and Pseudomonas
aeruginosa were conducted by a high-throughput dilution-resolved cultivation
method. The results were consistent with the time-resolved conventional method,
indicating the robustness of the new cultivation tool [141]. Microscale culture and
extraction platforms were developed by Barkal and collaborators [139]. The avail-
able technology allows studying metabolic alterations in fungal and bacterial
metabolomes after external stimuli, such as variations in culture.
Other automated methods involve the high-throughput SPME sample prepara-
tion protocol, developed for untargeted analysis [142]. An autosampler, capable of
preparing 96 samples simultaneously was used to optimize the extraction of intra
and extracellular metabolites from E. coli. The authors claim the efficiency of this
new system in extracting metabolites, ranging from polar to lipophilic, in a short
time. Finally, an automated platform was recently developed by Donati and co-­
workers [143] for microbial multi-omics workflow. The customized platform is
capable of operating from cultivation, sampling, sample extraction, analysis, and
data processing. In this paper, the authors describe the design of the platform for
targeted analysis of exometabolome, untargeted endometabolome, and proteomics,
pointing out advantages and limitations. Better performance was observed when it
operates with a capacity of 264 batch cultivars per day [143]. Automated and min-
iaturized methods have been the focus of several works in recent years. These mod-
ern strategies enable the high throughput of sample preparation using minimal
amounts of solvents, being fast and environmentally friendly.
7 Sample Preparation in Microbial Metabolomics: Advances and Challenges 175

7.5 Conclusions

Microbial metabolomics studies are relevant in understanding system biology and

biotechnology since this omics approach helps to understand fundamental cellular
processes, with flow and redistribution of metabolites in cells [12]. There is no stan-
dard sample preparation protocol for microbial metabolomics analysis. All the dis-
cussions presented in this chapter consolidate the need for a systematic investigation
of the sample preparation steps and the development of individualized protocols.
Such protocols must consider the aim of the study based on the type of microorgan-
ism, metabolomics approach, and analytical technique employed. The workflow
involves important steps of cell culture planning, control of the culture medium and
available nutrients for cell growth, sample harvesting, and separation of cell and
culture medium. One of the greatest challenges in this area is the interruption of
enzymatic activity, called metabolic quenching. Most procedures described cannot
prevent metabolic leakage during quenching, requiring a careful evaluation of the
methods, with particular attention to the type of microorganism under study.
Extracellular analyses have gained prominence in biotechnology and to differen-
tiate types of microorganisms, providing important insights into the metabolic state
of organisms. The preparation of exometabolome samples is simple and fast, con-
sisting of dilutions and filtration or centrifugation. Sample preparation is more elab-
orate and time-consuming for intracellular analyses, which provide an instant
picture of cellular metabolism. Such procedures can combine extracting solvents
and mechanical devices to increase the efficiency of cell lysis and can also be per-
formed simultaneously with metabolic quenching. The set of information provided
by footprinting and fingerprinting analysis allows a better understanding of the
dynamics of cell metabolism.
The use of automated techniques and microscale platforms has been highlighted
in recent years. The development of modern techniques for extracting cellular
metabolites has shown to be efficient, with good reproducibility, reduction in the
volume of toxic reagents, and satisfactory analytical frequency.
Due to the increasing interest in microbial metabolism, a meticulous examina-
tion of the workflow of cell sample preparation should be considered for metabolo-
mics. A good experimental design and systematic study of the parameters to assess
the metabolites will provide reliable data for accurate metabolic interpretation.

References

1. Canuto GAB, da Costa JL, da Cruz PLR, de Souza ANL, Faccio AT, Klassen A, Rodrigues
KT, Tavares MFM (2018) Metabolômica: definições, estado-da-arte e aplicações representa-
tivas. Quím Nova 41:75–91
2. Klassen A, Faccio AT, Canuto GAB, da Cruz PLR, Ribeiro HC, Tavares MFM, Sussulini A
(2017) Metabolomics: definitions and significance in systems biology. In: Sussulini A (ed)
Metabolomics: from fundamentals to clinical applications. Springer, Switzerland, pp 3–17
176 H. V. M. Boness et al.

3. Kuehnbaum NL, Britz-McKibbin P (2013) New advances in separation science for metabolo-
mics: resolving chemical diversity in a post-genomic era. Chem Rev 113:2437–2468
4. Covington BC, McLean JA, Bachmann BO (2017) Comparative mass spectrometry-based
metabolomics strategies for the investigation of microbial secondary metabolites. Nat Prod
Rep 34:6–24
5. Xu YJ, Wang C, Ho WE, Ong CN (2014) Recent developments and applications of metabo-
lomics in microbiological investigations. Trends Anal Chem 56:37–48
6. Ye D, Li X, Shen J, Xia X (2022) Microbial metabolomics: from novel technologies to diver-
sified applications. Trends Anal Chem 148:116540
7. Bauermeister A, Mannochio- Russo H, Costa-Lotufo LV, Jarmusch AK, Dorrestein PC (2022)
Mass spectrometry-based metabolomics in microbiome investigations. Nat Rev 20:143–160
8. Oyedeji AB, Green E, Adebiyi JA, Ogundele OM, Gbashi S, Adefisoye MA, Oyeyinka SA,
Adebo OA (2021) Metabolomic approaches for the determination of metabolites from patho-
genic microorganisms: a review. Food Res Int 140:110042
9. Belinato JR, Bazioli JM, Sussulini A, Augusto F, Fill TP (2019) Metabolômica microbiana:
inovações e aplicações. Quím. Nova 42:546–559
10. Patejko M, Jacyna J, Markuszewski MJ (2017) Sample preparation procedures utilized in
microbial metabolomics: an overview. J Chromatogr B 1043:150–157
11. Álvarez-Sánchez B, Priego-Capote F, de Castro MDL (2010) Metabolomics analysis I. selec-
tion of biological samples and practical aspects preceding sample preparation. Trends Anal
Chem 29:111–119
12. Mashego MR, Rumbold K, De Mey M, Vandamme E, Soetaert W, Heijnen JJ (2007)
Microbial metabolomics: past, present and future methodologies. Biotechnol Lett 29:1–16
13. Kell DB, Brown M, Davey HM, Dunn WB, Spasic I, Oliver SG (2005) Metabolic footprint-
ing and systems biology: the medium is the message. Nat Rev Microbiol 3:557–565
14. Kamal KM, Maifiah MHM, Rahim NA, Hashim YZHY, Sani MSA, Azizan KA (2022) Bacterial
metabolomics: sample preparation. Methods 2022. https://doi.org/10.1155/2022/9186536
15. Lu W, Su X, Klein MS, Lewis IA, Fiehn O, Rabinowitz JD (2017) Metabolite measurement:
pitfalls to avoid and practices to follow. Annu Rev Biochem 86:277–304
16. Vuckovic D (2012) Current trends and challenges in sample preparation for global metabolo-
mics using liquid chromatography-mass. Anal Bioanal Chem 403:1523–1548
17. Kanu AB (2021) Recent developments in sample preparation techniques combined with
high-performance liquid chromatography: a critical review. J Chromatogr A 1654:462444
18. Barnes S, Benton HP, Casazza K, Cooper SJ, Cui X, Du X, Engler J, Kabarowski JH, Li S,
Pathmasiri W, Prasain JK, Renfrow MB, Tiwari HK, Training in metabolomics research.
I. (2016) Designing the experiment, collecting and extracting samples and generating metab-
olomics data. J Mass Spec 51:461–475
19. Álvarez-Sánchez B, Priego-Capote F, de Castro MDL (2010) Metabolomics analysis II: prep-
aration of biological samples prior to detection. Trends Anal Chem 29:120–127
20. Gomes TA, Zanette CM, Spier MR (2020) An overview of cell disruption methods for intra-
cellular biomolecules recovery. Prep Biochem Biotechnol 50:635–654
21. Pinu FR, Villas-Boas SG, Aggio R (2017) Analysis of intracellular metabolites from micro-
organisms: quenching and extraction protocols. Meta 7:53
22. Pinu FR, Villas-Boas SG (2017) Extracellular microbial metabolomics: the state of the art.
Meta 7:43
23. Stevens VL, Hoover E, Wang Y, Zanetti KA (2019) Pre-analytical factors that affect metabo-
lite stability in human urine, plasma, and serum: a review. Meta 9:156
24. Smith L, Villaret-Cazadamont J, Claus SP, Canlet C, Guillou H, Cabaton NJ, Ellero-Simatos
S (2020) Important considerations for sample collection in metabolomics studies with a spe-
cial focus on applications to liver functions. Metabolites 10:104
25. Sumner LW, Amberg A, Barrett D, Beale MH, Beger R, Daykin CA, Fan TWM, Fiehn O,
Goodacre R, Griffin JL, Hankemeier T, Hardy N, Harnly J, Higashi R, Kopka J, Lane AN,
Lindon JC, Marriott P, Nicholls AW, Reily MD, Thaden JJ, Viant MR (2007) Proposed mini-
7 Sample Preparation in Microbial Metabolomics: Advances and Challenges 177

mum reporting standards for chemical analysis chemical analysis working group (CAWG)
metabolomics standards initiative (MSI). Metabolomics 3:211–221
26. CIMR: In vitro Biology/Microbiology Context. <https://github.com/MSI-­Metabolomics-­
Standards-­Initiative/CIMR/blob/master/cimr-­invitro.pdf> Accessed August 2022
27. Horak I, Engelbrecht G, Jansen van Resburg PJ, Claassens S (2019) Microbial metabolomics:
essential definitions and the importance of cultivation conditions for utilizing bacillus species
as bionematicides. J Appl Microbiol 127:326–343
28. Pilizota T, Yang YT (2018) “Do it yourself” microbial cultivation techniques for synthetic and
systems biology: cheap, fun, and flexible. Fron Microbiol 9:1666
29. Shen Y, Fatemeh T, Tang L, Cai Z (2016) Quantitative metabolic network profiling of
Escherichia coli: an overview of analytical methods for measurement of intracellular metabo-
lites. Trends Anal Chem 75:141–150
30. Mielko KA, Jabłoński SJ, Łukaszewicz M, Młynarz P (2021) Comparison of bacteria disin-
tegration methods and their influence on data analysis in metabolomics. Sci Rep 11:20859
31. Crnkovic CM, May DS, Orjala J (2018) The impact of culture conditions on growth and
metabolomic profiles of freshwater cyanobacteria. J Appl Phycol 30:375–384
32. Zareian M, Silcock P, Bremer P (2018) Effect of medium compositions on microbially medi-
ated volatile organic compounds release profile. J Appl Microbiol 125:813–827
33. Kim J, Kim KH (2017) Effects of minimal media vs. complex media on the metabolite pro-
files of Escherichia coli and Saccharomyces cerevisiae. Process Biochem 57:64–71
34. Panagiotou G, Christakopoulos P, Olsson L (2005) The influence of different cultivation con-
ditions on the metabolome of fusarium oxysporum. J Biotechnol 118:304–315
35. van Gulik WM (2010) Fast sampling for quantitative microbial metabolomics. Curr Opin
Biotechnol 21:27–34
36. Mashego MR, van Gulik WM, Vinke JL, Heijnen JJ (2003) Critical evaluation of sam-
pling techniques for residual glucose determination in carbon-limited chemostat culture of
Saccharomyces cerevisiae. Biotechnol Bioeng 83:395–399
37. Visser D, van Zuylen GA, van Dam JC, Oudshoorn A, Eman MR, Ras C, van Gulik WM,
Frank J, van Dedem GWK, Heijnen JJ (2002) Rapid sampling for analysis of in vivo kinet-
ics using the BioScope: a system for continuous pulse experiments. Biotechnol Bioeng
79:674–681
38. Maier RM, Pepper IL (2009) Bacterial growth. In: Maier RM, Pepper IL, Gerba CP (eds)
Environmental microbiology. Elsevier/Academic Press, Burlington, pp 37–54
39. Halouska S, Zhang B, Gaupp R, Lei S, Snell E, Fenton RJ, Barletta RG, Somerville GA,
Powers R (2013) Revisiting protocols for the NMR analysis of bacterial metabolomes. J
Integr OMICS 3:120–137
40. Smart KF, Aggio RBM, Houtte JRV, Villas-Bôas SG (2010) Analytical platform for metabo-
lome analysis of microbial cells using methyl chloroformate derivatization followed by gas
chromatography–mass spectrometry. Nat Prot 5:1709–1729
41. Pezzatti J, Bergé M, Boccard J, Codesido S, Gagnebin Y, Viollier PH, González-Ruiz V,
Rudaz S (2019) Choosing an optimal sample preparation in Caulobacter crescentus for untar-
geted metabolomics approaches. Meta 9:193
42. Jiang Y, Ge F, Li F, Zhang D, Deng S, Tian J (2020) Intracellular metabolomics switch-
ing alters extracellular acid production and insoluble phosphate Solubilization behaviour in
Penicillium oxalicum. Meta 10:441
43. Hussein M, Han ML, Zhu Y, Zhou Q, Lin YW, Hancock REW, Hoyer D, Creek DJ, Li J,
Velkov T (2020) Metabolomics study of the synergistic killing of Polymyxin bin combina-
tion with amikacin against Polymyxin-susceptible and -resistant Pseudomonas aeruginosa.
Antimicrob Agents Chemother 64:e01587–e01519
44. Han ML, Zhu Y, Creek DJ, Lin YW, Anderson D, Shen HH, Tsuji B, Gutu AD, Moskowitz
SM, Velkov T, Li J (2018) Alterations of metabolic and lipid profiles in Polymyxin-resistant
Pseudomonas aeruginosa. Antimicrob Agents Chemother 62:e02656–e02617
178 H. V. M. Boness et al.

45. Hong M, Mou H, Liu X, Huang M, Chu J (2017) 13C-assisted metabolomics analysis
reveals the positive correlation between specific erythromycin production rate and intracel-
lular propionyl-­CoA pool size in Saccharopolyspora erythraea. Bioprocess Biosyst Eng
40:1337–1348
46. Maifiah MHM, Creek DJ, Nation RL, Forrest A, Tsuji BT, Velkov T, Li J (2017) Untargeted
metabolomics analysis reveals key pathways responsible for the synergistic killing of colistin
and doripenem combination against Acinetobacter baumannii. Sci Rep 7:45527
47. Jin H, Ma H, Gan N, Wang H, Li Y, Wang L, Song L (2022) Non-targeted metabolomic
profiling of filamentous cyanobacteria Aphanizomenon flos-aquae exposed to a concentrated
culture filtrate of Microcystis aeruginosa. Harmful Algae 111:102170
48. Kuang E, Marney M, Cuevas D, Edwards RA, Forsberg EM (2020) Towards predicting gut
microbial metabolism: integration of flux balance analysis and untargeted metabolomics.
Meta 10:156
49. Zhong F, Xu M, Metz P, Ghosh-Dastidar P, Zhu J (2018) A quantitative metabolomics study
of bacterial metabolites indifferent domains. Anal Chim Acta 1037:237–244
50. Schelli K, Rutowski J, Roubidoux J, Zhu J (2017) Staphylococcus aureus methicillin resis-
tance detected by HPLC-MS/MS targeted metabolic profiling. J Chromatogr B 1047:124–130
51. Xu M, Zhong F, Zhu J (2017) Evaluating metabolic response to light exposure in lactobacil-
lus species via targeted metabolic profiling. J Microbio Methods 133:14–19
52. Kukolj C, Pedrosa FO, de Souza GA, Sumner LW, Lei Z, Sumner B, do Amaral FP, Juexin W,
Trupti J, Huergo LF, Monteiro RA, Valdameri G, Stacey G, de Souza EM. (2020) Proteomic
and Metabolomic analysis of Azospirillum brasilense ntrCMutant under high and low nitro-
gen conditions. J Proteome Res 19:92–105
53. Bolten CJ, Kiefer P, Letisse F, Portais JC, Wittmann C (2007) Sampling for metabolome
analysis of microorganisms. Anal Chem 79:3843–3849
54. Coulier L, Bas R, Jespersen S, Verheij E, van der Werf MJ, Hankemeier T (2006) Simultaneous
Quantitative Analysis of Metabolites Using Ion-Pair Liquid Chromatography−Electrospray
Ionization Mass Spectrometry. Anal Chem 78:6573–6582
55. Zhang Q, Zheng X, Wang Y, Yu J, Zhang Z, Dele-Osibanjo T, Zheng P, Sun J, Jia S, Ma Y
(2018) Comprehensive optimization of the metabolomic methodology for metabolite profil-
ing of Corynebacterium glutamicum. Appl Microbiol Biotechnol 102:7113–7121
56. Tredwell GD, Aw R, Edwards-Jones B, Leak DJ, Bundy JG (2017) Rapid screening of cel-
lular stress responses in recombinant Pichia pastoris strains using metabolite profiling. J Ind
Microbiol Biotechnol 44:413–417
57. Jäpelt KB, Christensen JH, Villas-Boas SG (2015) Metabolic fingerprinting of lactobacillus
paracasei: the optimal quenching strategy. Microb Cell Factories 14:132
58. Spura J, Reimer LC, Wieloch P, Schreiber K, Buchinger S, Schomburg D (2009) A method
for enzyme quenching in microbial metabolome analysis successfully applied to gram-­
positive and gram-negative bacteria and yeast. Anal Biochem 394:192–201
59. Villas-Boas SG, Bruheim P (2007) Cold glycerol-saline: the promising quenching solution
for accurate intracellular metabolite analysis of microbial cells. Anal Biochem 370:87–97
60. Koek MM, Jellema RH, van der Greef J, Tas AC, Hankemeier T (2011) Quantitative
metabolomics based on gas chromatography mass spectrometry: status and perspectives.
Metabolomics 7:307–328
61. Yun EJ, Seo SO, Kwak S, Oh EJ, Lee SH, Jin YS, Kim KH (2021) Fast filtration with a vac-
uum manifold system as a rapid and robust metabolome sampling method for Saccharomyces
cerevisiae. Process Biochem 110:195–200
62. Liu T, Jin Z, Wang Z, Chen J, Wei LJ, Hua Q (2020) Metabolomics analysis of Actinosynnema
pretiosum with improved AP-3production by enhancing UDP-glucose biosynthesis. J Biosci
Bioeng 130:36–47
63. Zheng X, Yu J, Cairns TC, Zhang L, Zhang Z, Zhang Q, Zheng P, Sun J, Ma Y (2019)
Comprehensive improvement of sample preparation methodologies facilitates dynamic
metabolomics of aspergillus Niger. Biotechnol J 14:e1800315
7 Sample Preparation in Microbial Metabolomics: Advances and Challenges 179

64. Miao Q, Zhao L, Wang Y, Hao F, Sun P, He P, Liu Y, Huang J, Liu X, Liu X, Deng G,
Li H, Li L, Tang Y, Wang L, Feng M, Jia W (2020) Microbial metabolomics and network
analysis reveal fungistatic effect of basil (Ocimum basilicum) oil on Candida albicans. J
Ethnopharmacol 260:113002
65. Lu H, Chen H, Tang X, Yang Q, Zhang H, Chen YQ, Chen W (2019) Evaluation of metab-
olome sample preparation and extraction methodologies for oleaginous filamentous fungi
Mortierella alpina. Metabolomics 15:50
66. Li H, Wang Y, Meng Q, Wang Y, Xia G, Xia X, Shen J (2019) Comprehensive proteomic
and metabolomic profiling of mcr-1-mediated colistin resistance in Escherichia coli. Int J
Antimicrob Agents 53:795–804
67. Yang Q, Tao R, Yang B, Zhang H, Chen YQ, Chen H, Chen W (2018) Optimization of the
quenching and extraction procedures for a metabolomic analysis of lactobacillus plantarum.
Anal Biochem 557:62–68
68. Wang HB, Feng YR, Gui SQ, Zhang Y, Lu FP (2017) A sample pretreatment method to
suit the metabolomic analysis of bacillus licheniformis based on GC-MS. Anal Methods
9:2299–2304
69. Ye D, Li X, Wang C, Liu S, Zhao L, Du J, Xu J, Li J, Tian L, Xia X (2021) Improved sam-
ple preparation for untargeted metabolomics profiling of Escherichia coli. Microbiol Spectr
9:e00625–e00621
70. Takenaka M, Yoshida T, Hori Y, Bamba T, Mochizuki M, Vavricka CJ, Hattori T, Hayakawa
Y, Hasunuma T, Kondo A (2021) An ion-pair free LC-MS/MS method for quantitative metab-
olite profiling of microbial bioproduction systems. Talanta 222:121625
71. Li Z, Yao Q, Dearth SP, Entler MR, Gonzalez HFC, Uehling JK, Vilgalys RJ, Hurst GB,
Campagna SR, Labbé JL, Pan C (2017) Integrated proteomics and metabolomics suggests
symbiotic metabolism and multimodal regulation in a fungal-endobacterial system. Environm
Microbiol 19:1041–1053
72. Zhao L, Ni Y, Su M, Li H, Dong F, Chen W, Wei R, Zhang L, Guiraud SP, Martin FP, Rajani
C, Xie G, Jia W (2017) High throughput and quantitative measurement of microbial metab-
olome by gas chromatography/mass spectrometry using automated alkyl Chloroformate
derivatization. Anal Chem 89:5565–5577
73. Salvatore MM, Maione A, Albarano L, de Alteriis E, Carraturo F, Andolfi A, Salvatore F,
Galdiero E, Guida M (2021) An integrated analysis of intracellular metabolites and virulence
gene expression during biofilm development of a clinical isolate of Candida tropicalis on
distinct surfaces. Int J Mol Sci 22:9038
74. Liu X, Wang T, Sun X, Wang Z, Tian X, Zhuang Y, Chu J (2019) Optimized sampling proto-
col for mass spectrometry-based metabolomics in Streptomyces. Bioresour Bioprocess 6:32
75. Wordofa GG, Kristensen M, Schrübbers L, McCloskey D, Forster J, Schneider K (2017)
Quantifying the metabolome of Pseudomonas taiwanensis VLB120: evaluation of hot and
cold combined quenching/extraction approaches. Anal Chem 89:8738–8747
76. Bhilare KD, Patil MD, Tangadpalliwar S, Dev MJ, Garg P, Benerjee UC (2018) Machine
learning modelling for the high-pressure homogenization-mediated disruption of recombi-
nant E. coli. Process Biochem 71:182–190
77. Phan ANT, Blank LM (2020) GC-MS-based metabolomics for the smut fungus Ustilago
maydis: a comprehensive method optimization to quantify intracellular metabolites. Front
Mol Biosci 7:211
78. Chen L, Zhao X, Wu J, Liu Q, Pang X, Yang H (2020) Metabolic characterization of eight
Escherichia coli strains including “Big Six” and acidic responses of selected strains revealed
by NMR spectroscopy. Food Microbiol 88:103399
79. du Preez I, Beukes D, Loots DT (2019) Fatty acid metabolome extraction from mycobacte-
rial cells for GC-MS metabolomics analysis. In: Baidoo EEK (ed) Microbial metabolomics:
methods and protocols, methods in molecular biology. Humana Press, New York, pp 111–120
180 H. V. M. Boness et al.

80. Planchon M, Léger T, Spalla O, Huber G, Ferrari R (2017) Metabolomic and proteomic
investigations of impacts of titanium dioxide nanoparticles on Escherichia coli. PLoS One
12:e0178437
81. Winder CL, Dunn WB, Schuler S, Broadhurst D, Jarvis R, Stephens GM, Goodacre R (2008)
Global metabolic profiling of Escherichia coli cultures: an evaluation of methods for quench-
ing and extraction of intracellular metabolites. Anal Chem 80:2939–2948
82. Groves RA, Mapar M, Aburashed R, Ponce LF, Bishop SL, Rydzak T, Drikic M, Bihan DG,
Benediktsson H, Clement F, Gregson DB, Lewis IA (2022) Methods for quantifying the
metabolic boundary fluxes of cell cultures in large cohorts by high-resolution hydrophilic
liquid chromatography mass spectrometry. Anal Chem 94:8874–8882
83. Rutowski J, Zhong F, Xu M, Zhu J (2019) Metabolic shift of Staphylococcus aureus under
sublethal dose of methicillin in the presence of glucose. J Pharm Biomed Anal 167:140–148
84. Ma H, Lai B, Jin Y, Tian C, Liu J, Wang K (2020) Proteomics and metabolomics analysis
reveal potential mechanism of extended-spectrum blactamase production in Escherichia coli.
RSC Adv 10:26862–26873
85. Foschi C, Salvo M, Laghi L, Zhu C, Ambretti S, Marangoni A, Re MC (2018) Impact of
meropenem on Klebsiella pneumoniae metabolism. PLoS One 13:e0207478
86. Baptista R, Fazakerley DM, Beckmann M, Baillie L, Mur LAJ (2018) Untargeted metabolo-
mics reveals a new mode of action of pretomanid (PA-824). Sci Rep 8:5084
87. Aros-Calt S, Castelli FA, Lamourette P, Gervasi G, Junot C, Muller BH, Fenaille F (2019)
Metabolomic investigation of Staphylococcus aureus antibiotic susceptibility by liquid
chromatography coupled to high-resolution mass spectrometry. In: Wang X, Kuruc M (eds)
Functional proteomics: methods and protocols, methods in molecular biology. Springer
Nature, New York, pp 279–293
88. Yuan H, Xu Y, Chen Y, Zhan Y, Wei X, Li L, Wang D, He P, Li S, Chen S (2019) Metabolomics
analysis reveals global acetoin stress response of bacillus licheniformis. Metabolomics 15:25
89. Duportet X, Aggio RBM, Carneiro S, Villas-Boas SG (2012) The biological interpretation
of metabolomic data can be misled by the extraction method used. Metabolomics 8:410–421
90. Doran ML, Mykytczuk N, Bieniek A, Methé A, Merritt TJS (2017) Evaluation of quenching
and extraction procedures for performing metabolomics in Acidithiobacillus ferrooxidans.
Metabolomics 13:156
91. Anastas P, Eghbali N (2010) Green chemistry: principles and practice. Chem Soc Rev
39:301–312
92. Esquivel-Hernández DA, Rodríguez-Rodríguez J, Alemán-Nava GS, Cuéllar-Bermúdez SP,
García-Pérez JS, Mancera-Andrade EI, Núñez-Echevarría JE, Ontiveros-Valencia A, Rostro-­
Alanis M, García-García RM, Torres JA, Chen WN, Parra-Saldívar R (2017) Effect of super-
critical carbon dioxide extraction parameters on the biological activities and metabolites
present in extracts from Arthrospira platensis. Mar Drugs 15:174
93. Gouda M, Sankar U, Sarada R, Ravishankar GA (2015) Supercritical CO2 extraction
of functional compounds from spirulina and their biological activity. J Food Sci Technol
52:3627–3633
94. Esquivel-Hernández DA, López VH, Rodríguez-Rodríguez J, Cuéllar-Bermúdez SP, Rostro-­
Alanis M, Parra-Saldívar R (2016) Supercritical carbon dioxide and microwave-assisted
extraction of functional lipophilic compounds from Arthrospira platensis. Int J Mol Sci 17:658
95. Bader CD, Neuber M, Panter F, Krug D, Müller R (2020) Supercritical fluid extraction enhances
discovery of secondary metabolites from Myxobacteria. Anal Chem 92:15403–15411
96. Aurich MK, Paglia G, Rolfsson Ó, Hrafnsdóttir S, Magnúsdóttir M, Stefaniak MM, Palsson
BØ, Fleming RMT, Thiele I (2015) Prediction of intracellular metabolic states from extracel-
lular metabolomic data. Metabolomics 11:603–619
97. Carneiro S, Villas-Bôas SG, Ferreira EC, Rocha I (2011) Metabolic footprint analysis of
recombinant Escherichia coli strains during fed-batch fermentations. Mol BioSyst 7:899–910
98. Zhuang X, Han Z, Bai Z, Zhuang G, Shim H (2010) Progress in decontamination by halo-
philic microorganisms in saline wastewater and soil. Environ Pollut 158:1119–1126
7 Sample Preparation in Microbial Metabolomics: Advances and Challenges 181

99. Kang HJ, Yang HJ, Kim MJ, Han ES, Kim HJ, Kwon DY (2011) Metabolomic analysis of
meju during fermentation by ultra-performance liquid chromatography-quadrupole-time of
flight mass spectrometry (UPLC-Q-TOF MS). Food Chem 127:1056–1064
100. Lee JE, Hong YS, Lee CH (2009) Characterization of fermentative behaviours of lactic acid
bacteria in grape wines through 1H NMR- and GC-based metabolic profiling. J Agric Food
Chem 57:4810–4817
101. Panagiotou G, Pachidou F, Petroutsos D, Olsson L, Christakopoulos P (2008) Fermentation
characteristics of fusarium oxysporum grown on acetate. Bioresour Technol 99:7397–7401
102. Meijer S, Panagiotou G, Olsson L, Nielsen J (2007) Physiological characterization of xylose
metabolism in aspergillus Niger under oxygen-limited conditions. Biotechnol Bioeng
98:462–475
103. Villas-Bôas SG, Moon CD, Noel S, Hussein H, Kelly WJ, Cao M, Lane GA, Cookson AL,
Attwood GT (2008) Phenotypic characterization of transposon-inserted mutants of clostrid-
ium proteoclasticum B316T using extracellular metabolomics. J Biotechnol 134:55–63
104. Allen J, Davey HM, Broadhurst D, Heald JK, Rowland JJ, Oliver SG, Kell DB (2003) High-­
throughput classification of yeast mutants for functional genomics using metabolic footprint-
ing. Nat Biotechnol 21:692–696
105. Granucci N, Pinu FR, Han TL, Villas-Boas SG (2015) Can we predict the intracellular meta-
bolic state of a cell based on extracellular metabolite data? Mol BioSyst 11:3297–3304
106. Galdiero E, Salvatore MM, Maione A, de Alteriis E, Andolfi A, Salvatore F, Guida M (2021)
GC-MS-based metabolomics study of single- and dual-species biofilms of Candida albicans
and Klebsiella pneumoniae. Int J Mol Sci 22:3496
107. Dettmer K, Aronov PA, Hammock BD (2007) Mass spectrometry-based metabolomics. Mass
Spectrom Rev 26:51–78
108. Pawliszyn J (1997) Solid phase microextraction: theory and practice. Wiley-VCH Verlag
GmbH & Co., New York, p 264p
109. Pawliszyn J (2001) Solid phase microextraction. In: Rouseff RL, Cadwallader KR (eds)
Headspace analysis of foods and flavors, Advances in experimental medicine and biology.
Springer, Boston
110. Beccaria M, Franchina FA, Nasir M, Mellors T, Hill JE, Purcaro G (2021) Investigating
bacterial Volatilome for the classification and identification of mycobacterial species by
HS-SPME-GC-MS and machine learning. Molecules 26:4600
111. Monedeiro F, Railean-Plugaru V, Monedeiro-Milanowski M, Pomastowski P, Buszewski B
(2021) Metabolic profiling of VOCs emitted by bacteria isolated from pressure ulcers and
treated with different concentrations of bio-AgNPs. Int J Mol Sci 22:4696
112. Don SMY, Schmidtke LM, Gambetta JM, Steel CC (2020) Aureobasidium pullulans vola-
tilome identified by a novel, quantitative approach employing SPME-GC-MS, suppressed
Botrytis cinerea and Alternaria alternata in vitro. Sci Rep 10:4498
113. Costa CP, Bezerra AR, Almeida A, Roch SM (2020) Candida species (volatile)
Metabotyping through advanced comprehensive two-dimensional gas chromatography.
Microorganisms 8:1911
114. Azollini A, Boggia L, Boccard J, Sgorbini B, Lecoultre N, Allard PM, Rubiolo P, Rudaz S,
Gindro K, Bicchi C, Wolfender JL (2018) Dynamics of metabolite induction in fungal co-­
cultures by metabolomics at both volatile and non-volatile levels. Front Microbiol 9:72
115. Farbo MG, Urgeghe PP, Fiori S, Marcello A, Oggiano S, Balmas V, Hassan ZU, Jaoua S,
Migheli Q (2018) Effect of yeast volatile organic compounds onochratoxin A-producing
aspergillus carbonarius and A. ochraceus. Int J Food Microbiol 284:1–10
116. Reyes-Garcés N, Gionfriddo E (2019) Recent developments and applications of solid phase
microextraction as a sample preparation approach for mass-spectrometry-based metabolo-
mics and lipidomics. Trends Anal Chem 113:172–181
117. Dailey A, Saha J, Zaidi F, Abdirahman H, Haymond A, Alem F, Hakami R, Couch R (2020)
VOC fingerprints: metabolomic signatures of biothreat agents with and without antibiotic
resistance. Sci Rep 10:11746
182 H. V. M. Boness et al.

118. Mastrangelo A, Ferrarini A, Rey-Stolle F, García A, Barbas C (2015) From sample treat-
ment to biomarker discovery: a tutorial for untargeted metabolomics based on GC-(EI)-
Q-MS. Anal Chim Acta 900:21–35
119. Khakimov B, Motawia MS, Bak S, Engelsen SB (2013) The use of trimethylsilyl cyanide
derivatization for robust and broad-spectrum high-throughput gas chromatography–mass
spectrometry based metabolomics. Anal Bioanal Chem 405:9193–9205
120. Beukes D, du Preez I, Loots DT (2019) Total metabolome extraction from mycobacterial
cells for GC-MS metabolomics analysis. In: Baidoo EEK (ed) Microbial metabolomics.
Humana Press, New York, pp 121–131
121. Stuart AKC, Furuie JL, Cataldi TR, Stuart RM, Zawadneak MAC, Labate CA, Pimentel
IC (2022) Fungal consortium of two Beauveria bassiana strains increases their virulence,
growth, and resistance to stress: a metabolomic approach. PLoS One 17:e0271460
122. Galdiero E, Salvatore MM, Maione A, Carraturo F, Galdiero S, Falanga A, Andolfi A, Salvatore
F, Guida M (2021) Impact of the peptide WMR-K on dual-species biofilm Candida albicans/
Klebsiella pneumoniae and on the untargeted Metabolomic profile. Pathogens 10:214
123. Semreen MH, Soliman SSM, Saeed BQ, Alqarihi A, Uppuluri P, Ibrahim AS (2019)
Metabolic profiling of Candida auris, a newly-emerging multi-drug resistant Candida spe-
cies, by GC-MS. Molecules 24:399
124. Kanani HH, Klapa MI (2007) Data correlation strategy for metabolomics analysis using gas
chromatography-mass spectrometry. Metab Eng 9:39–51
125. Koek MM, Muilwijk B, van der Werf MJ, Hankemeier T (2006) Microbial metabolomics
with gas chromatography/mass spectrometry. Anal Chem 78:1272–1281
126. Villas-Bôas SG, Delicado DG, Åkesson M, Nielsen J (2003) Simultaneous analysis of amino
and nonamino organic acids as methyl chloroformate derivatives using gas chromatography-­
mass spectrometry. Anal Biochem 322:134–138
127. Villas-Bôas SG, Smart KF, Sivakumaran S, Lane GA (2011) Alkylation or Silylation for
analysis of amino and non-amino organic acids by GC-MS? Meta 1:3–20
128. Casu F, Pinu FR, Stefanello E, Greenwood DR, Villas-Bôas SG (2018) The fate of linoleic
acid on Saccharomyces cerevisiae metabolism under aerobic and anaerobic conditions.
Metabolomics 14:103
129. Mairinger T, Weiner M, Hann S, Troyer C (2020) Selective and accurate quantifica-
tion of N-Acetylglucosamine in biotechnological cell samples via GC-MS/MS and
GC-TOFMS. Anal Chem 92:4875–4883
130. Fritsche-Guenther R, Gloaguen Y, Bauer A, Opialla T, Kempa S, Fleming CA, Redmond HP,
Kirwan JA (2021) Optimized workflow for on-line derivatization for targeted metabolomics
approach by gas chromatography-mass spectrometry. Meta 11:888
131. Abbiss H, Rawlinson C, Maker GL, Trengove R (2015) Assessment of automated trimeth-
ylsilyl derivatization protocols for GC–MS-based untargeted metabolomic analysis of urine.
Metabolomics 11:1908–1921
132. Macedo AN, Faccio AT, Fukuji TS, Canuto GAB, Tavares MFM (2021) Analytical platforms
for mass spectrometry-based metabolomics of polar and Ionizable metabolites. In: Simionato
AVC (ed) Separation techniques applied to omics sciences, advances in experimental medi-
cine and biology. Springer, Switzerland, pp 215–242
133. Wu Y, Li L (2016) Sample normalization methods in quantitative metabolomics. J Chromatogr
A 1430:80–95
134. Misra BB (2020) Data normalization strategies in metabolomics: current challenges,
approaches, and tools. Eur J Mass Spectrom 26:165–174
135. Wang S, Lee LJ (2013) Micro−/nanofluidics based cell electroporation.
Biomicrofluidics 7:11301
136. Young EWK, Beebe DJ (2010) Fundamentals of microfluidic cell culture in controlled micro-
environments. Chem Soc Rev 39:1036–1048
137. Pudasaini S, Perera ATK, Ng SH, Yang C (2021) Bacterial inactivation via microfluidic elec-
troporation device with insulating micropillars. Electrophoresis 42:1093–1101
7 Sample Preparation in Microbial Metabolomics: Advances and Challenges 183

138. Rockenbach A, Sudarsan S, Berens J, Kosubek M, Lazar J, Demling P, Hanke R, Mennicken

P, Ebert BE, Blank LM, Schnakenberg U (2019) Microfluidic irreversible electroporation - a
versatile tool to extract intracellular contents of bacteria and yeast. Meta 9:211
139. Barkal LJ, Theberge AB, Guo CJ, Spraker J, Rappert L, Berthier J, Brakke KA, Wang CCC,
Beebe DJ, Keller NP, Berthier E (2016) Microbial metabolomics in open microscale plat-
forms. Nat Commun 7:10610
140. Li Z, Wang Z, Pan J, Ma X, Zhang W, Ouyang Z (2020) Single-cell mass spectrometry
analysis of metabolites facilitated by cell electro-migration and electroporation. Anal Chem
92:10138–10144
141. Perdesen BH, Gurdo N, Johansen HK, Molin S, Nikel PI, La Rosa R (2021) High-throughput
dilution-based growth method enables time-resolved exo-metabolomics of Pseudomonas
putida and Pseudomonas aeruginosa. Microb Biotechnol 14:2214–2226
142. Mousavi F, Bojko B, Pawliszyn J (2019) High-throughput solid-phase microextraction–liq-
uid chromatography–mass spectrometry for microbial untargeted metabolomics. In: Baidoo
EEK (ed) Microbial metabolomics: methods and protocols, methods in molecular biology.
Humana Press, New York, pp 133–152
143. Donati S, Mattanovich M, Hjort P, Jacobsen SAB, Blomquist SD, Mangaard D, Gurdo N,
Pastor FP, Maury J, Hanke R, Herrgård MJ, Wulff T, Jakočiūnas T, Nielsen LK, McCloskey
D (2022) An automated workflow for multi-omics screening of microbial model organisms.
2022; Preprint. https://doi.org/10.1101/2022.07.18.500181
Chapter 8
Discovering New Natural Products Using
Metabolomics-Based Approaches

Lívia Soman de Medeiros, Moysés B. de Araújo Júnior, Eldrinei G. Peres,

José Carlos Ipuchima da Silva, Milena Costa Bassicheto, Giordanno Di Gioia,
Thiago André Moura Veiga, and Hector Henrique Ferreira Koolen

8.1 Introduction

The comprehensive knowledge of the chemical diversity of microbial natural prod-

ucts represents a prominent access route to bioactive molecules that has diverse
implications for humanity. In this sense, the recognition of metabolites, especially
those produced by microorganisms, presents itself as a great attraction to the scien-
tific community, in the most diverse frontiers of study. This is because, historically,
many of these compounds have been and are still used for the benefit of society in
biotechnological and biomedical applications. On the other hand, many also repre-
sent toxic agents, making them targets of constant monitoring, due to their harmful
effects [1–3].
For decades, the screening and acquisition of new entities of natural origin
adopted (and often still today, depending on the available laboratory infrastructure,
adopt), classic preliminary processes of chromatographic fractionations of crude
extracts, with concomitant monitoring of biological activities of fractions and sub-­
fractions of these extracts. In these cases, the use of spectroscopic techniques
(NMR, MS, IV, UV) for molecular identifications is traditionally performed only at
the end of the process, in other words, after purification of the targets determined by
biological responses [4–6]. The strategy of biomonitored fractionations for the dis-
covery of molecules was established in the past once it has been successful in

L. S. de Medeiros (*) · M. C. Bassicheto · G. Di Gioia · T. A. M. Veiga

Grupo de Pesquisas LaBiORG – Laboratório de Química Bio-orgânica Otto Richard Gottlieb,
Universidade Federal de São Paulo, Diadema, Brazil
e-mail: [email protected]
M. B. de Araújo Júnior · E. G. Peres · J. C. I. da Silva · H. H. F. Koolen
Grupo de Pesquisa em Metabolômica e Espectrometria de Massas,
Universidade do Estado do Amazonas, Manaus, Brazil

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 185
T. Pacheco Fill (ed.), Microbial Natural Products Chemistry, Advances in Experimental
Medicine and Biology 1439, https://doi.org/10.1007/978-3-031-41741-2_8
186 L. S. de Medeiros et al.

encountering thousands of new compounds of natural origin. However, over the

years, the problem of frequent re-isolation of previously known compounds came to
the fore [7]; thus, for years causing some obscurantism in studies involving the
prospection of new natural compounds.
As a consequence, over the years, there became a need for new shortcuts to be
put into practice so that the workflows would lead to the discovery of new molecules
in a more assertive and effective way. With the modernization and the implementa-
tion of faster processes in large-scale screening (high-throughput screening), hun-
dreds of thousands of samples and compounds began to be inspected concomitantly,
through spectroscopic and spectrometric techniques, and also in various bioassays,
in an attempt to define targets without the prior isolation of samples and compounds
[8, 9]. However, the demand for the curation and treatment of an extensive set of
analytical and biological data became one of the bottlenecks for revealing new bio-
active metabolites, due to the large consumption of time required for data analysis
[10], quality of the acquired data [11], the reliability of the inspection method, and
even due to the fact that the detection of biological activities is susceptible to the
effect of synergism between the concomitantly biosynthesized compounds [12].
Particularly, in the case of microorganisms, with the advent of more accessible
genetic sequencing technologies and the modernization of bioinformatics, there has
been increasing use of genome mining tools for microbial DNA, which has led to
the unequivocal realization that, from a chemical and biological point of view, there
is in fact still an immense portion of metabolites to be unraveled. According to the
atlas of biosynthetic clusters, more than 400,000 microbial biosynthetic gene clus-
ters (BCGs) have been predicted to date, with less than 1% of them confirmed by
compound detection/identification [13], which ratifies their metabolic potential. In
the case of fungal metabolites for example, it is estimated that currently 80% of this
metabolic potential has not yet been accessed as a result of cryptic BCGs [14].
The discovery of BCGs with reduced or inactive gene expression (silenced
genes) has brought an even more promising horizon in the search for new natural
compounds [15]. Strategies for the induction and activation of gene expressions in
the exploration of compounds not yet detected have been increasingly implemented
over the last two decades. One can cite, for example, the use of practices that test the
biotic and abiotic fermentative modifications in the development of the microorgan-
ism, which are traditionally known as the OSMAC (one strain many compounds)
approach [16]. The execution of nutritional variations, and variations in tempera-
ture, pH, luminosity, and use of chemical elicitors, epigenetic modifiers and, finally,
the use of crosstalking techniques, with the experimentation of combined microbial
cultures (co-cultures) are the main possibilities encompassed in the approach [14,
17]. Although such techniques are historically one of the most direct and successful
with respect to activation of BCGs, genomics-based approaches also represent an
alternative and prominent path to accelerate access to microbial metabolic diversity
that is possibly not revealed through methodologies that involve direct nutritional,
mechanical, chemical or biological disturbances in the fermentative process [18,
19]. Gene manipulation, whether with the application of genetic engineering, regu-
lation of transcription factors or epigenetic regulation, currently represents the most
8 Discovering New Natural Products Using Metabolomics-Based Approaches 187

modern face of studies aimed at the biosynthetic understanding of microbial natural

products and can lead to new discoveries regarding chemical entities [20–23].
Additionally, a key point that must be considered for broader access to the meta-
bolic potential of microorganisms is the biodiversity that is intrinsic to unexplored
organisms or unknown species. Studies estimate that there are about 1 trillion spe-
cies of microorganisms, with only a tiny fraction of these known to man [24]. In the
case of fungi, for example, it is estimated that between 3% and 8% of the total
number of possible species have been identified [25]. This immense, though still
latent biodiversity, combined with all the genetic code inherent in thousands of
unexplored species, reflects the great potential that nature still hides regarding the
uniqueness of its microbial chemodiversity.
Therefore, there is a high demand for analytical processes that are aimed at the
investigation of a significant number of biological systems and their respective bio-
synthetic machinery. In order to achieve new natural products and the understanding
of their chemical and biochemical functions inside and outside nature, the prelimi-
nary recognition of the molecules present in these biological systems is crucial. In
this sense, currently, the use of metabolomics-based approaches is practically man-
datory to achieve a faster and more efficient screening process.
Due to the existence of several concepts and tools inherent to modern metabolo-
mics, the chapter covers a compendium of relevant points and techniques, which
have come to the forefront in recent years, since they function as powerful shortcuts
for unraveling an immense hidden microbial chemodiversity. Molecular examples,
whose detection and identification were guided by the use of metabolomics-based
approaches will also be explained throughout this chapter (Fig. 8.1), and serve as
inspiration for the application of such approaches in workflows focused on the
search for new natural products.

8.2 Metabolomics and Integrated Analytical Techniques

Metabolomics is one of the most applied “omics” fields today. By providing the
visualization of molecules that regulate biological interactions, it is sometimes con-
sidered the connecting link between the omics sciences, and offers the possibility of
mechanistic insights into biological processes and the correlation of the biochemi-
cal activities involved in the observed phenotypes [26, 27].
By definition, metabolomics deals with the broad investigation of small mole-
cules, smaller than 2000 Da that are contained in a biological system, whether in
cells, tissues or body fluids [28, 29]. In microbial natural product research, its
implementation makes it possible to qualitatively and/or quantitatively detect sec-
ondary (also called specialized) metabolites that are produced by the organisms
under certain conditions to which they are subjected [30].
Metabolomics makes it possible to monitor chemical changes in the metabolome
of biological systems [31] and today, it is a tool that is widely used in an integrated
way with other strategies, such as in combination with genome mining and
188 L. S. de Medeiros et al.

Fig. 8.1 Examples of microbial metabolites detected and identified with the aid of modern
approaches based on metabolomics. In green, the molecules isolated from fungi and, in red, the
molecules isolated from bacteria

biological screening of specialized metabolites [32]. By allowing the chemical map-

ping of a large number of samples more efficiently, metabolomic studies facilitate
the process of discovering molecules, and also, as already mentioned, of under-
standing the biological role of secondary metabolites in nature. In this context, in
8 Discovering New Natural Products Using Metabolomics-Based Approaches 189

2014, Krug and Müller advocated the use of the term “secondary metabolomics”,
thus highlighting its specificity and relevance within the area of natural products as
a whole [33].
The workflows within metabolomics can include the use of different methodolo-
gies in sample preparation (item not covered in this chapter), data acquisition, anal-
ysis and visualization and, in particular, the preliminary identification of metabolites.
The establishment of these flows within metabolomics are mainly to enable the
interpretation of their biological functions in the system to which they are involved
[34–36]. Therefore, one of the main bases in metabolomic studies is founded on the
design of metabolic profillings of explored targets. Because the use of some con-
cepts sometimes present ambiguities, depending on the context in which they are
being used, it is important to highlight that the definition of metabolic profiling
involves the set of analytical data that lead to the determination of metabolites that
can be detected in a sample. In other words, it is the inspection and comparison of
spectroscopic (and sometimes also chromatographic) data of complex samples
(usually extracts and/or fractions) that leads to the identification and/or quantifica-
tion of specific groups of metabolites that share chemical and structural characteris-
tics [31, 37].
Although metabolomic analyses cannot be dissociated from the evaluation of
metabolic profillings, and also the fact that some authors consider the terms “metab-
olomics” and “metabolic profiling” interchangeable, in this chapter, we consider
that metabolomics is much broader, according to the discussion made by Hubert
et al. (2017): “Metabolomics is an interdisciplinary field that combines high resolu-
tion analytical systems, multivariate statistics and data mining tools, chemical and
biological knowledge, and sometimes metabolic network modeling in an attempt to
understand metabolic pathways, gene-function relationships, or states of an organ-
ism in response to environmental changes, drug perturbations, phylogeny, geno-
typic or phenotypic variabilities” sic [38].
For the broad detection of several biosynthetic classes, due to the demand for
simultaneous analysis (often from a large volume of specimens), it is necessary to
use analytical techniques that are at the same time reliable, robust, selective, accu-
rate and that have high resolution with respect to the capacity of molecular analyses
[14, 30]. In this sense, the analytical techniques of high resolution mass spectrom-
etry (HRMS) and nuclear magnetic resonance (NMR), represent the core of the
techniques used in metabolomics studies [35, 39], not only for fulfilling the afore-
mentioned characteristics and for the high degree of technological development
intrinsic to them, but also for the possibility of integration with bioinformatics tools
[5, 36].
As detection techniques, NMR, and especially HRMS, can be implemented indi-
vidually or orthogonally, complementing data in the acquisition of metabolic pro-
files [32]. Both methods can be coupled to liquid chromatography with the
hyphenation LC-NMR or LC-HRMS, or in the fully integrated form of LC-HRMS-­
SPE-NMR techniques [40], thus ensuring the most effective separation and evalua-
tion of complex samples by combining the advantages of all the techniques used
[36]. However, profiles obtained directly from NMR data, without chromatography,
190 L. S. de Medeiros et al.

is one of the most traditionally used methods in complement to data obtained sepa-
rately, from analyses via LC-HRMS and also via gas chromatography (GC-MS) [41].
The use of NMR offers advantages to studies of natural products as a whole.
Firstly, because it is a non-destructive technique, which offers a high degree of
detail regarding the chemical structures of metabolites and often allows the unequiv-
ocal molecular identification of them. In the second instance, when implemented in
metabolomic studies, it offers impartiality in the quantitative and qualitative detec-
tion of compounds, due to the fact that the technique is not susceptible to signal
variations that are intrinsic to the structural chemical particularities of the com-
pounds. It therefore provides good reproducibility in data acquisition and is appli-
cable to all classes of compounds [5, 36, 42].
Despite the advantages mentioned, NMR has a lower incidence in screening
studies that seek new microbial metabolites when compared to HRMS [43]. This is
because the application of HRMS can provide important advantages to data analy-
sis. The sensitivity in the detection of compounds in complex mixtures, for exam-
ple, can be much higher than in NMR; up to about 1000 times more sensitive [44].
Additionally, the molecular data set can be very broad and informative due to the
direct acquisition of accurate masses, with high resolving power between ions, and
mainly due to the possibility of comparisons and confirmations of structures via the
possibility of acquiring fragmentation profiles in MSn experiments [45].
Another very significant point in the context of HRMS is the ease of integration
of modern computational tools with the processing systems of data obtained via
mass spectrometry (MS), which accelerates the dynamics of screening and results
acquisition. This set of characteristics makes HRMS an essential technique for the
global interpretation of metabolites produced in biological systems and drives its
increasing implementation in recent metabolomic studies [46, 47].
The imaging of results for visual enrichment during the comparison of metabolic
profiles is an additional possibility also brought about by MS. As an example, it is
possible to list the use of different ionization sources used in MS, such as MALDI
and DESI [27]. The traditional hyphenation with liquid chromatography coupled to
a UV-vis detector or PDA also brings more information to be mined in the process
of assigning chemical profiles [41]. The direct correspondence between the respec-
tive mass and UV spectra of each compound (when chromophore groups are pres-
ent) is an additional parameter that can optimize target identification. However, it is
valid to point out that even HRMS is subjected to obstacles in the detection of
metabolites, mainly due to the problem of their selective ionization [48, 49].
Therefore, if it is feasible to implement the two techniques in an integrated way, this
can enhance the possibility of finding new natural products.
In this scenario, due to the need for molecular recognition and assignment of tens
of thousands of compounds in a sample, the practice of untargeted metabolomics is
the preferred approach over targeted metabolomics [50]. While the targeted
approach focuses on a specific set of predefined metabolites, the untargeted
approach, also called the global approach, does not prioritize any molecular group
per se, but rather aims to classify qualitatively and/or quantitatively as many entities
as possible that are present in the biological matrix according to their respective
8 Discovering New Natural Products Using Metabolomics-Based Approaches 191

spectral characteristics [43, 51]. Due to the possibility of detecting thousands of

signals corresponding to a diversity of compounds in a single sample, the use of the
global methodology represents the most comprehensive strategy and, therefore, has
greater applicability in the establishment of new discoveries [29].
Untargeted metabolomics encompasses a workflow with often very complex
data analyses, often resulting in one of the main challenges in the field of natural
products, which is the constant rediscovery of compounds [52]. Therefore, linking
processes that direct the molecular recognition contained in microbial metabolomes
effectively and quickly is an essential step for the discovery of either new deriva-
tives/analogs of known compounds, or rare/novel scaffolds, thus avoiding time-and
resource-consuming isolation of unwanted compounds [36]. In this sense, it is pos-
sible to highlight the consolidation of processes and methodologies of dereplication
(chemical and biological) and molecular annotation, which help us to carry out
modern metabolomics studies.

8.3 Processes and Concepts: Dereplication and Annotation

of Compounds

Dereplication and molecular annotation are closely connected processes for exploit-
ing known data in order to reveal data that are still hidden from the point of view of
identity. The term “dereplication” seems to have appeared in the literature prior to
1978, since the use of the word was reported in 1980 in the first CRC Handbook of
Antibiotic Compounds [53, 54]. Only in 1990 did its resurgence occur through the
study of plant extracts to identify the bioactivity of compounds, without their previ-
ous isolation [55].
Many works deal with the definition of the concept of “dereplication”. In his
work, Phelan (2020) simply uses the word “re-identification” [56], while Aron et al.
(2020), in a succinct way uses the definition “Rapid identification of previously
characterized (known) molecules” [57]. More comprehensive descriptions used in
studies of natural products can also be found. In Mohimani et al. (2018), we find the
description: “The process of using the information about the chemical structure of a
known natural product to identify this compound in an experimental sample (with-
out having to repeat the entire isolation and structure-determination process)”,
sic [52].
Regardless of the interpretation that is made regarding the concept itself, the fact
is that, today, dereplication can be considered a multidisciplinary process, and can
add improvements to new drug discovery programs. Inevitably, for this reason, it
has become a growing practice and of great interest in the context of natural prod-
ucts [58]. For example, in the last surveys carried out respectively by Gaudêncio
[58] and Hubert [38], who provide the number of publications and citations on the
subject, more than 800 citations were found and an average of 50 articles were pub-
lished in 2014 on the Web of Science™Core Collection platform (Thomson
192 L. S. de Medeiros et al.

Reuters). The trend of an evolution in the dissemination of the approach in the fol-
lowing years is ratified when the graph in Fig. 8.2 is observed. When comparing the
occurrence of the terms “dereplication” and “de-replication” specifically for the
years 2014 (the last year cited in the aforementioned reviews) and 2022, it is noted
that the number of citations was approximately five times higher in 2020.
As identified by these two reviews by Gaudêncio & Pereira (2015) [58] and
Hubert et al. (2017) [38], dereplication methodologies can be found in the literature
in different ways, and in variable systematic steps. In general, methodologies can be
based on analyses of a single sample or even a variety of crude extracts, sometimes
in combination with high-throughput screening methodologies. In most cases, the
process is assisted by the use of the analytical separation and detection techniques
already mentioned here (GC, LC, HRMS, UV, NMR), and the incidence of the
untargeted approach is more frequent in relation to the targeted approach in these
cases. Although not a mandatory practice, processes can also be aided by bioactivi-
ties, after or during the analysis of spectral data. Finally, with the evolution of
genomics allied to the advances of bioinformatics, today, it is common to find meth-
odologies directed mainly to studies of microorganisms, which are based on the
selection of more talented species from the biotechnological point of view, or on the
screening of organisms that are still unknown, via the prospecting of the respective
genomes. In these cases, taxonomic dereplication is established, which can help the
access to an unprecedented biosynthetic potential of previously selected lineages,
prior to chemical or biological analyses. Therefore, based on the screening of works
found in the last three decades, Hubert et al. (2017) categorized dereplication into
five different strands, named DEREP1 through DEREP5. All are linked to the search
for biomolecules of natural origin, and specify a particular workflow, which is sum-
marized in Fig. 8.3. For the details of each category, see Hubert et al. (2017) [38].

Fig. 8.2 Citation report and publications per year during the period of 1993–2022 for dereplica-
tion/de-replication. Boolean operator applied: “OR”. (Data sourced from Web of Science™ Core
Collection (all rights reserved))
8 Discovering New Natural Products Using Metabolomics-Based Approaches 193

Fig. 8.3 Workflow for identification of biomolecules of natural origin. DEREP1 undirected derep-
lication aimed at the rapid identification of the main known compounds, regardless of their chemi-
cal class, in a single natural sample; DEREP2 dereplication for the acceleration of bioactivity-guided
fractionation procedures; DEREP3 dereplication fully incorporated in undirected metabolomic
analyses of collections of natural extracts; DEREP4 targeted dereplication focused on identifying
a predefined group of known molecules and DEREP5 taxonomic dereplication based on gene
sequence analyses of collections of microbial strains. (Adapted from Ref. [38])

It is essential to warn that although the identification of metabolites without prior

isolation, namely the annotation process, is inherent to the different dereplication
methodologies, it is most often performed putatively [38]. In other words, although
the annotation of compounds is a primary guide for recognizing the biosynthetic
potential of a sample of natural origin, unambiguous molecular identification still
depends on the use of standards and/or the use of 3D structural elucidation tech-
niques after isolation and purification of the target compound, including informa-
tion on the molecular configuration (relative or absolute), such as through the use of
techniques such as NMR and X-ray crystallography [9, 59]. Annotation is actually
a process of assigning the putative chemical structure of a detected molecule [57]
194 L. S. de Medeiros et al.

and, therefore, great caution should be taken to avoid possible erroneous statements
about the identification of metabolomes. In this way, the annotation of detected
metabolites must be reliable in order to ensure the correct interpretation and infer-
ence of results [60].
It can be said that there is a consensus on the importance of establishing and
executing annotations based on a reliability ranking for metabolomic profiling
reports [29]. In this scenario, through proposals and refinements established with
the Metabolomics Standards Initiative (MSI), the Metabolomics Society proposed 5
different levels of metabolic annotation, on a scale of 0–4, considering the wide-
spread use of analytical techniques [61]. Figure 8.4 illustrates the summary of the
different levels proposed. In it, the levels are associated with a given identification
situation for the metabolite to be recognized, according to the analytical require-
ments fulfilled for obtaining the results. In this way, the identification is given from
the highest to the lowest degree of reliability, going through situations of “unequivo-
cal molecular identification”, “reliable molecular identification”, “putatively anno-
tated compounds”, “putatively characterized compound classes” and, finally,
“unidentified compounds”. For more details, see Sumner et al. (2007) [61].
Other ranking systems can also be found, especially those with the use of experi-
ments conducted via MS. This is probably due to the fact that this is the analytical
technique that is most used for annotation of compounds nowadays [41]. For

Fig. 8.4 Levels of identification of molecules in metabolomics found in the literature. (Adapted
from [29, 61, 62])
8 Discovering New Natural Products Using Metabolomics-Based Approaches 195

example, the system established by Schymanski et al. (2014) makes additional and
more specific considerations to the reality of MS when compared to the system
proposed by the MSI [62]. In this case, level 1 is represented by the structural con-
firmation of the target metabolite, when there is the matching of MS, MS/MS data
and retention times to those of the reference standards, and there may be the combi-
nation of orthogonal detection methods. The intermediate levels of identification
apply the use of comparison and correspondence of the detected MS and MS/MS
data with data from the literature and data from spectral libraries, as well as the
matching of diagnostic characteristics obtained experimentally for the evaluated
compounds, without the manipulation of in situ standards. When the investigated
targets are compatible with the hits found in the libraries and there is experimental
evidence of diagnostic profiles, the attribution of a possible structure occurs, thus
characterizing level 2 reliability. With the absence of possible hits in the databases
used, level 3 makes the proposal feasible for candidates within a specific class;
however, it makes information about exact structures impossible to achieve. For
level 4, the confirmation of characteristic isotopic profiles, the detection of different
types of adducts or clusters formed in the ionization source, and details of fragmen-
tation of the structure can collectively corroborate the assertive acquisition of the
molecular formula of the species; while at level 5, which implies the use of only the
accurate mass (obtained via high resolution and high accuracy equipment), there is
the possibility of the screening of probable predetermined targets, but without con-
firmatory validity.
Even more complex and thorough systems are found in the literature, as is the
case of the work described by Rochat (2017) [63]. In this example, the reliability of
annotation is assigned by a tiered system, named by a combination of letters and
numbers which in turn is associated with a detailed scoring system. For each score,
very specific criteria are stipulated with respect to the data obtained via
LC-HRMS. For more details, see Rochat (2017) [63]. Regardless of the system
considered, it is important that, within a metabolomic study, the dereplication pro-
cesses are based on and implement criteria regarding credibility during molecular
annotation, with the aim of conducting work that leads to reliable results.

8.4 Strategic Tools for Mining Microbial Natural Products

8.4.1 Databases and Data Analysis Platforms

Analytical data mining strategies are currently taking on the role of revealing true
“snapshots” of the microbial metabolome through the increasingly broad knowl-
edge of the chemodiversity produced by organisms. These strategies have as their
main goal the highlighting of known compounds, but also the identification of
unknown entities, thus making the search for novel metabolites a faster process, as
well as enabling the prioritization of lineages, extracts and, finally, targeting
196 L. S. de Medeiros et al.

molecules to be purified and characterized unequivocally [26, 32, 64]. One of the
key points within the dereplication and, consequently, for the annotation of com-
pounds is to have the most comprehensive knowledge possible about the informa-
tion reported in the scientific literature regarding candidate metabolites. Therefore,
the consultation of databases of compounds that house repositories of chemical and
spectral information, such as biosynthetic origin, molecular formulas, exact masses,
chemical structures and respective spectra acquired mainly via NMR and MS, is
essential [14, 43].
The availability of commercial or free databases, which are increasingly becom-
ing more complete and better organized, has revolutionized all the fields of “omics”
due to the possibility of access to a massive amount of information, and today these
are a valuable resource especially for metabolomic analysis. Interestingly, in not-so-­
distant times, a simple query regarding a single compound could represent the most
complex challenge of “finding a needle in a haystack” via laborious searches in
printed literature papers available at local libraries. Prior to the 90s, the number of
data collections was very small, especially with regard to natural products. Since it
was the beginning of the digital age, obviously online access to information was in
its infancy. In the same way, computational search tools have been following the
technological development of computing over the decades and, today, it is possible
to have access to online databases that are completely dedicated to natural prod-
ucts [65].
Some publications list a robust compendium of these databases, including bases
with free access and with data obtained in silico [5, 66]. In a recent and fairly com-
plete review published by Sorokina & Steinbeck (2020) [67], more than 100 data-
bases and collections devoted to the subject have been catalogued. However, even
with the large number of these reservoirs of information, one of the main obstacles
to the use of these platforms in metabolomic studies of microorganisms is due to the
fact that most of the deposited structures come from other natural sources. Few
databases are dedicated exclusively to microorganisms, for example, Antibase [68],
Atlas of Natural Products [69] and StreptomeDB [70], the latter having only depos-
its for bacteria of the genus Streptomyces.
Other problems involve partial accessibility to the total set of deposits (in the
case of in-house collections for example), absence of spectral data, lack of updating,
and servers being out of operation. The inspection of data can also be difficult in
some cases due to the limitation in the types of filters and also the type of formatting
of entries assigned by each platform, bringing the underutilization of important fea-
tures that may be used during curation. For example, the Collection of Open Natural
Products (COCONUT) [71], which currently represents the most robust open access
platform for natural products, does not enable the search for compounds through
exact or isotopic masses (only molecular weight), which makes it difficult to directly
screen candidates based on accurate mass data acquired via HRMS. In this case, for
a more selective query, the use of previously deduced molecular formulas should be
applied, restricting the direct flow of search via accurate masses and deviations
selected in ppm.
8 Discovering New Natural Products Using Metabolomics-Based Approaches 197

Even with the highlighted problems, the manual or automated inspection of these
libraries through the use of software represents an important tool for the annotation
of molecules. As such, Table 8.1 summarizes data sources that have some record of
information deposited regarding natural products of microbial origin and that can be
eventually consulted and used by the reader.
Due to querying versatility (intrinsic structural organizations and, thus, various
ways to query), and especially the differential compatibility regarding the number
and variety of compounds present in each of these repositories [5], the concomitant
use of one or more collections can be quite advantageous during microbial informa-
tion mining. Even if each molecule database presents certain limitations, the use of
these platforms is essential during the prospection of microbial secondary metabo-
lites, and facilitates the discovery of novel compounds. For example, dereplication
based on MS data and Antibase consultation aided in the confirmation of seven
depsides (thielavin S (1), T (2), U (3) and V (4); leocanorin D (5), E (6) and F (7))
produced by the fungus Setophoma sp. [117]. The use of the same repository showed
the biosynthetic precursor rugulovasine in extracts of Talaromyces wortmannii, and
aided in the isolation of new dichlorinated analogs, the indole alkaloids
2,8-­dichlororugulovasine A (8) and B (9) [118]. Recently, after studies involving in
silica fragmentation of compounds also found in Antibase, 33 new cyclic lipopep-
tides (CLPs) produced by the proteobacteria Pseudomonas sp. stechlisins B2 (10),
C3 (11) and D3 (12) were reported [119]. Another repository traditionally used is
the Dictionary of Natural Products. Workflows involving chemical dereplication
with the application of queries to this database are cited in several works involving
the discovery of microbial metabolites, such as the dipeptide xestostreptin (13),
which is produced by a species of Streptomyces associated with the marine sponge
Xestospongia muta [120], and the polyketide 5,9-dihydroxy-2,4,6,8,10-­
pentamethyldodeca-­2,6,10-trienal (14), which was isolated from Aspergillus floc-
culus [121].
Currently, the most robust molecular database that is fully dedicated to com-
pounds of microorganisms is the Natural Products Atlas [122]. More than 32,000
compounds are currently deposited on this platform, with its latest update integrat-
ing additional repositories, such as CyanoMetDB, a database that covers metabo-
lites exclusively from cyanobacteria [123]. The use of the Natural Products Atlas in
the investigative process of the secondary metabolism of cyanobacteria Nostoc sp.
UIC 10771 and Oscillatoria sp. aided the discovery of two unusual alkaloids aulo-
sirazoles B (15) and C (16) [124] and the heptapeptide lingaoamide (17) [125],
respectively.
In untargeted metabolomics, in complement to molecular databases, there is the
use of spectral libraries. Access to reference spectra plays a vital role in compound
identification processes. The depositing of spectral data intrinsic to molecules of
microbial origin occurs in several repositories, including MassBank, Metabolights,
NIST, METLIN, MONA and mzCloud, among others (see Table 8.1). Despite the
advantages brought about by libraries, in the challenge of deciphering the identity
of molecular targets through MS or NMR data, the Achilles heel of the system is
certainly in the number of entries in these spectral databases. The molecular
Table 8.1 Avaiable databases for the search of MS and NMR data

Search Search entry Search Search Search Search

entry for for accurate entry for entry for entry for entry for
molecular mass or exact molecular MS MS/MS NMR NMR 13C 2D NMR
1
Database formula mass weight spectra spectra H shifts shifts data Access type Link References
Ambinter-Green-­pharma No No No No No No No No Commercial https://www.ambinter.com/ [72]
natural compound library
(GPNCL)
AntiBase Yes Yes No Yes No No Yes No Commercial https://mswil.com/wp-­content/ [68]
uploads/2020/02/
AntiBase2014.pdf
BiGG No No No No No No No No Free http://bigg.ucsd.edu/ [73]
Binding DB No No Yes No No No No No Free http://www.bindingdb.org/ [74]
bind/index.jsp
BIOFAQUIM No No No No No No No No Free https://biofacquim.herokuapp. [75]
com/
ChemSpider Yes Yes Yes No No Yes Yes No Free http://www.chemspider.com/ [76]
CNPD (Chinese Natural No No Yes No No No No No Free https://www.cmnpd.org./ [77]
Products Database) browse/organism
COCONUT Yes No Yes No No No No No Free https://coconut.naturalproducts. [71]
net/search/advanced
DMNP (Dictionary of Yes No No No No No No No Commercial https://dmnp.chemnetbase. [78]
Marine Natural Products) com/faces/chemical/
ChemicalSearch.xhtml;jsession
id=FB8467F1F5C803DA4937
10C8EFFFA789
DNP (Dicionary os Natural Yes No No No No No No No Commercial https://dnp.chemnetbase.com/ [78]
Products) faces/chemical/
ChemicalSearch.xhtml;jsession
id=4969FDEA9FA328D8229D
42C30ED684F7
Drugbank NPs No Yes Yes Yes Yes Yes Yes Yes Free https://go.drugbank.com/ [79]
GNPS (Global Natural No Yes No Yes Yes No No No Free https://gnps.ucsd.edu/ [80]
Products Social Molecular ProteoSAFe/static/gnps-­splash.
Networking) jsp
HMDB (Human No No Yes Yes Yes Yes Yes No Free https://hmdb.ca/ [81]
Metabolome Database)
Indofine Chemical No No No No No No No No Free https://indofinechemical.com/ [82]
Company Inc. natural
products
InflamNat No No Yes No No No No No Free http://www.inflamnat.com/#/ [83]
main/home
InterBioScreen Ltd (IBS) Yes No No No No No No No Free https://www.ibscreen.com/ [84]
KEGG Yes Yes No No No No No No Free https://www.genome.jp/kegg/ [85]
compound/
Marine Natural Products No No Yes No No No No No Free https://www.cmnpd.org/A [86]
Database (MNPD)
MarinLit Yes Yes Not Yes Not Yes Yes No Commercial https://marinlit.rsc.org/ [87]
specified specified
Massbank Yes Yes No Yes Yes Yes Yes Yes Free https://massbank.eu/ [88]
MassBank/
Metabolights No No No Yes No Yes No No Free https://www.ebi.ac.uk/ [89]
metabolights/compounds
Metabolomics WorkBench Yes Yes No Yes Not No No No Free https://www. [90]
specified metabolomicsworkbench.org/
data/mb_structure_ajax.php
MetaCyc Yes Yes Yes No No No No No Free https://metacyc.org/ [91]
METLIN Yes Yes No Yes Yes No No No Free https://metlin.scripps.edu/ [92]
landing_page.
php?pgcontent=mainPage
MONA Yes Yes No Yes Yes No No No Free https://mona.fiehnlab.ucdavis. [93]
edu/
mzCloud No No No Yes Yes No No No Free https://www.mzcloud.org/ [94]
NCI DTP data No No No No No No No No Free https://dtp.cancer.gov/ [95]
databases_tools/data_search_
instructions.htm
(continued)
Table 8.1 (continued)

Search Search entry Search Search Search Search

entry for for accurate entry for entry for entry for entry for
molecular mass or exact molecular MS MS/MS NMR NMR 13C 2D NMR
1
Database formula mass weight spectra spectra H shifts shifts data Access type Link References
NIST Yes Yes Yes No Yes No No No Commercial https://webbook.nist.gov/ [96]
chemistry/
NMRDATA – – – – – No Yes No In-house http://en.nmrdata.com/ –
collection
NMRShiftDB Yes No No Yes No Yes Yes No Free https://nmrshiftdb.nmr. [97]
uni-­koeln.de/
NPASS No No Yes No No No No No Free http://bidd.group/NPASS/ [98]
index.php
NPAtlas Yes Yes Yes No No No No No Free https://www.npatlas.org/ [99]
NPEdia Yes No Yes No No No No No Free http://www.cbrg.riken.jp/ [100]
npedia/keywords.php
Open Source Malaria Yes Yes No No No Not Not Not Free https://www.cheminfo.org/ [101]
specified specified flavor/malaria/Display_data/
index.html
PAMDB Yes No Yes Yes Yes No No No Free http://pseudomonas.umaryland. [102]
edu/
PubChem Yes No No No No No No No Free https://pubchem.ncbi.nlm.nih. [103]
gov/
StreptomeDB No No Yes Yes Yes Yes Yes No Free http://132.230.56.4/ [70]
streptomedb/
Super Natural II No No Yes No No No No No Free https://bioinf-­applied.charite. [104]
de/supernatural_new/
TargetMol Natural No No No No No No No No Commercial https://www.molport.com/ [105]
Compound Library shop/supplier/
TargetMol-­Chemicals/6883
UNPD (Universal Natural No No Yes No No Not Not Not Free https://bioinf-­applied.charite. [106]
Products Database) specified specified specified de/supernatural_new/index.
php?site=compound_input
Yeast Metabolome No Yes Yes Yes Yes Yes Yes Yes Free http://www.ymdb.ca/ [107]
Database
ZINC natural producst No No Yes No No No No No Free https://zinc12.docking.org/ [108]
catalogue catalogs/specsnp
BMRB (Biological Yes Yes No No No Yes Yes Yes Free https://bmrb.io/metabolomics/ [109]
Magnetic Resonance Data
Bank)
NAPROC-13 No No No No No No Yes No Free https://c13.materia-­medica.net/ [110]
NMR ShiftDB Yes No No No No Yes Yes No Free https://nmrshiftdb.nmr. [111, 112]
uni-­koeln.de/
SDBS Yes No Yes Yes No Yes Yes No Free https://sdbs.db.aist.go.jp [113]
DEREP-NP No No No Yes Yes Yes No Yes Commercial https://github.com/clzani/ [114]
DEREP-­NP
MP-MRD (Natural Yes Yes No No No Yes Yes Yes Free https://np-­mrd.org [115]
Products Magnetic
Resonance Database)
CH-NMR-NP (JEOL) Yes No Yes No No Yes Yes No Free https://www.j-­resonance.com/ [116]
en/nmrdb/
202 L. S. de Medeiros et al.

representativeness for compounds of microbial origin is expressively lower when

compared to molecular databases, and currently represents one of the greatest bar-
riers to metabolic annotation processes. Additionally, METLIN which today repre-
sents the broadest database of MS/MS, with deposited spectra referring to more
than 800,000 standards [126], is only commercially available (https://metlin.
scripps.edu).
In contrast to spectral databases with limited accessibility is the Global Natural
Products Social Molecular Networking (https://gnps.ucsd.edu). GNPS represents
an online bioinformatics platform that was developed for curation, analysis and
sharing of data acquired via mass spectrometry, with emphasis on natural products.
The platform features an important public library of MS/MS spectra and incorpo-
rates data from several other spectral databases available in its own system and also
from a repository powered by the community itself (GNPS-MassIVE), thus extend-
ing the effectiveness during dereplication and annotation processes for its users.
In addition to dereplication itself, which occurs through the automated compari-
son of spectra in this virtual community infrastructure, the acquisition of molecular
networks is another important application provided by the platform [127, 128]. By
grouping molecules with possibly correlated structures, the information obtained
through these networks operate as facilitators of the screening process of new com-
pounds and, perhaps for this reason, it is the most used tool within this modern
ecosystem of data analysis [57, 129].
In summary, in the case of the GNPS platform, molecular networks are obtained
from computational operations that use alignment algorithms and spectral similarity
calculations of MS2 data via the conversion of pairs of spectra into vectors and com-
parisons of products of these vectors correlated to the cosine angle. From the cosine
angle between two vectors, a cosine similarity score of distinct spectra can be estab-
lished, which varies between 0 and 1. In other words, there is the correspondence of
the smallest to the greatest degree of similarity between two spectra of MS2, where
1 represents identical spectra [130]. This type of connection stems from the concept
that molecules that have correlated structural portions generate a similar fragmenta-
tion profile and can be therefore grouped according to the degree of similarity
between these profiles. However, molecular networks are the visual record of puta-
tively related metabolites and, interestingly, they had the beginnings of their appli-
cation focused precisely on the screening of secondary metabolites present in vivo
in colonies of microorganisms [131].
The consolidation of the use of this tool in the microbial metabolomics context
and also on other fronts is notorious. This is because, over the last few years, its
implementation has made it possible to a more quickly identification of already
known metabolites, as well as comparative analyses between samples through the
detection and automated correlation of analogous molecular portions and com-
pounds of the same biosynthetic origin [132, 133].
Another software that can be applied for the generation of molecular networks
and comparison of MS2 spectra is MetGem [134]. In this application, molecular
networks can be obtained through the t-SNE (t-distributed stochastic neighbor
embedding) algorithm, without information losses that are eventually faced in
8 Discovering New Natural Products Using Metabolomics-Based Approaches 203

networks generated via the GNPS system due to the application of processing fil-
ters. The two formats for obtaining networks are complementary and the integrated
use of the two computational applications is recommended for better robustness of
the chemical information recovered from the data set. The advantages of the com-
bined use of the approaches have been reported for the process of identifying new
communesins of natural (Communesin M (18) and Communesin O (19)) and semi-­
synthetic origin, which are rare heptacyclic alkaloids associated with the chemodi-
versity of a marine strain of Penicillium expansum [135].
Even if software with functionalities similar to those found in GNPS are avail-
able, a differential to be highlighted here is undoubtedly the democratization of all
the resources that this particular platform makes available. As long as an analyst has
access to a computer, or even a cellular phone, that is connected to the internet, it is
possible to take advantage of this data processing network for metabolomic analy-
sis. Various educational and informative materials on new trends in data processing
using the GNPS platform are widely disseminated and made available by the GNPS
founding group itself through online video sharing platforms and social networks
(e.g., Youtube, Twitter). Additionally, well-structured protocols detailed in technical
parameters can be found in the literature, thus making it easier for users to perform
the step-by-step creation of molecular networks in GNPS in a reproducible way.
This is the case of the protocol reported by Aron et al. (2020) [57], for example, for
the classical workflow and also for feature-based molecular networking, in which
features such as retention times and ionic abundance are incorporated for data
acquisition with distinction between isobar species.
Although the processes of metabolic investigation for the discovery of new natu-
ral products have been expanded by the use of the platform, the diminutive avail-
ability of spectra derived mainly from microbial compounds cannot be
underestimated. To solve this problem, GNPS has been adding other in sillico data
processing tools (some of them mentioned in the following items) in the expectation
of boosting and consolidating metabolic annotation processes, especially in the case
of molecules produced by microorganisms [136].
Although the NMR technique is typically applied to pure compounds for the
characterization and structural elucidation of molecules [43], it is also a detection
technique that is widely used in the dereplication of natural products [115]. In both
cases, commercial or in-house NMR spectra collections can be consulted, while
metadata sharing platforms with open access have a greater emphasis on NMR-­
based dereplication processes [66].
With regard to the studies of microbial natural products, specific spectral reposi-
tories are limited. It is possible to mention StreptomeDB [70] win which unique
NMR spectra of bacteria of the genus Streptomyces are found. However, open
access databases that also include spectra of compounds of microbial origin can be
accessed: NMR database of natural products 13C (NAPROC-13) [110], NMR
shiftDB [112], spectral database for organic compounds (SDBS) [113] and Natural
Products Magnetic Resonance Imaging (NP-MRD) database [115]. Of these, the
largest database is NMR shiftDB, which contains 449,966 NMR spectra of organic
compounds including 1D (13C, 1H, 15N, 31p and etc.) and 2D. NAPROC-13 is
204 L. S. de Medeiros et al.

restricted to 13C spectra, presenting more than 6000 spectra of various natural prod-
ucts. Different to other platforms, it allows one to search by type of carbon (C, CH,
CH2, CH3, C-O, and CH-O, etc.). SDBS contains 15,900 1H NMR spectra and
14,200 13C NMR spectra. NP-MRD is the latest database (released in 2020) and has
spectra of nearly 41,000 natural products from over 7400 species. For the most part,
these platforms make it possible to search by name, molecular formula, structure
and lists of chemical displacements. Other open-access and commercial databases
are listed in Table 8.1.

8.4.2 Automated Data Processing and In Silico

Computational Tools

8.4.2.1 Examples in Mass Spectrometry

Traditionally, metabolomic studies comprise an extensive sample size to be investi-

gated and, consequently, a massive number of spectra to be inspected. The manual
and complete evaluation for this volume of information is a challenge during the
interpretation of results. Therefore, pre-treatment methods and automated data anal-
ysis become essential in conducting experiments in metabolomics. Fortunately,
with the evolution and increasing supply of computational tools that offer such a
possibility of automation, nowadays, it is possible to more effectively organize a
volume of spectral information, which enables faster and more assertive interpreta-
tions of the data sets to be explored.
For metabolomic studies via both MS and NMR, data pretreatment from raw
data files should be performed prior to statistical analysis and annotation. Usually,
pretreatments are performed using software and algorithms, which perform peak
picking procedures (detection of peaks and integration of their areas), extraction,
classification, normalization, standardization and grouping of signals, as well as
alignment (correspondence of peaks between samples) and peak deconvolution
(operation of distinguishing signals that coalesce or coexist in a spectrum) [137–
140]. Most of the software packages available, which are marketed by the different
equipment manufacturers (i.e., Bruker, Agilent, and Waters, etc.), usually integrate
the tools that enable such procedures in an automated way [141]. Other software
packages available online have greater employability and reach, since they can per-
form pretreatment of data in raw data obtained in different instruments and brands,
such as NMRProcFlow [142], OpenMS [143], XCMS [144] and mzMine [145].
The last two stand out as the most popular, and are widely used in untargeted metab-
olomics for processing of data obtained via LC-MS or GC-MS [137, 138].
Particularly in the case of analyses involving samples of microbial origin, there
is the frequent manipulation of organic extracts (or possibly extracts of aqueous
origin) that contain complex mixtures of compounds in varying concentrations. As
previously explained, LC-MS hyphenation represents one of the primary analytical
techniques for finding new molecular targets without the need for prior isolation of
8 Discovering New Natural Products Using Metabolomics-Based Approaches 205

compounds, which is mainly due to its combined working characteristics, such as

high selectivity, sensitivity and analyte separation capacity [138, 146]. In these
cases, the processing of untargeted data (manual or automated) is usually initiated
with the detection of all the ions in their respective retention times, which are called
features [60, 63]. During the interpretation of the spectra, the inspection of features,
through the integration of information on exact masses, isotopic profile, type of ions
(formation of adducts, clusters, and ion-source fragments) and retention times, is
crucial for the selection of targets.
Although the verification of features and their intrinsic characteristics is not
something extremely complicated to be carried out manually, it requires time, some
practice and interpretative experience for the definition of molecular formulas from
the types of ions detected. This occurs especially when there is the use of electro-
spray (ESI) as a source of ionization due to the high probability of the formation of
various types of pseudomolecular ions. ESI in positive ion mode can generate
mainly species such as [M+H]+, [M + Na]+, [M+H+CH3CN]+, [M+H+CH3OH]+ and
[M + H−H2O]+ or [M−H]−, [M+HCO2]−, and [M−H+CO2]− in negative ion mode.
However, other forms can often exist or co-exist, thus implying more complex cases
[45, 46]. Additionally, as already stated, one of the main bottlenecks for the inter-
pretation of metabolomic results is the large volume of data to be filtered. In this
sense, the establishment of molecular formulas associated with more than one type
of ionic species in a given retention time can be confirmed more quickly and reli-
ably through the use of tools and/or algorithms found in software packages, such as
the traditional XCMS-CAMERA. The CAMERA tool is coupled together with the
XCMS platform and assists the identification of metabolites with the reduction of
the redundancy of features in approximately 50% of cases, as well as the reduction
of false positives [147]. Another tool, mzAdan, was recently launched for annota-
tion in single or multiple accurate high-resolution ESI-MS spectra, via direct inser-
tion, which enables the assignment of different ionic species without the application
of chromatographic hyphenation [46].
In untargeted analyses, it is possible for a single type of ionic species to be
detected for a single feature, which makes it difficult to confirm the pseudomo-
lecular ion via clusters, adducts, or losses originating from the ionization source.
In addition, several compounds can share the same exact mass and molecular
formula, causing a stalemate in the differentiation between isomers with informa-
tion that is only from MS1 data. Thus, to assist in the identification of compounds
and increase the reliability of dereplication and annotation processes, the applica-
tion of mass spectrometry in tandem mode (MS/MS) is seen as a valuable strat-
egy, especially when performing the Data Dependent Acquisition (DDA) mode
[35]. In this mode, a given number of precursor ions are selected for fragmenta-
tion according to criteria of threshold, charge type and exclusion of unwanted
ions. The DDA mode is characterized as the main method of data acquisition in
MS/MS. Another possible mode is Data-Independent Acquisition (DIA), in which
all ions are fragmented (or in a range of m/z) during a scan. Although it provides
a lot of information about the possible fragments, this mode is not as widely used
206 L. S. de Medeiros et al.

as the DDA mode due to the need for informatics retreatment and the lack of
spectral libraries of data acquired via DIA [5, 148].
The great trend of application of experiments in the DDA mode has enabled
dereplication processes and metabolic annotation through the correlation of frag-
ments of compounds sought, with MS/MS spectra of reference compounds depos-
ited in platforms and/or spectrum databases (such as METLIN, GNPS, etc.), in
addition to diagnostic information from the acquisition of molecular networks of
reference compounds. However, one of the main obstacles found in metabolomic
studies aimed at the preliminary identification of compounds is the fact that the
construction of fully complete platforms is impractical and/or unpredictable. With
regard to microbial metabolomics, the scenario is even less favorable since the set
of data related to compounds known to be produced by microorganisms and depos-
ited in libraries is considered much lower than the amount estimated to exist
[29, 149].
Fortunately, in the last decade, new perspectives have emerged as a result of the
advancement of computer technologies. These advances have provided a rapid evo-
lution in the improvement of databases and platforms, allowing the availability of
increasingly robust libraries and mainly leading to the intense sharing of experimen-
tal data among the scientific community at a global level [150]. Complementarily,
the development of programs and methodologies based fundamentally on in silico
fragmentation have also represented one of the most promising tools to circumvent
the limitations brought by a still insufficient number of microbial compound data,
especially with regard to the lack of spectral libraries.
The great advantage of these programs is that they allow the putative identifica-
tion of metabolites, even if spectral data deposited in databases are unavailable. Due
to the in silico fragmentations being developed through recent artificial intelligence
methods and based on hypothetical molecular fragmentation patterns (from the
molecular models present in libraries and which are in continuous supply), the
annotation process of compounds has been conducted at a high level of reliability.
Ironically, only very recently it was noticed that in silico tools for fragmentation
data analysis (MSn) of natural products were scarce [60]. However, after less than a
decade, the appearance of a range of options to be explored for annotation of struc-
tures and substructures of natural products as a whole has been observed.
In silico fragmentation methodologies can be based on three main approaches:
rule-based prediction, combinatorial fragmentation and fragmentation trees [151].
In the first case, there is the simulation and comparison of spectra according to theo-
retical breaks of known chemical structures through the compendium of rules
described in the literature on mass spectrometry. MS-Finder is a software that
implements this type of approach and applies hydrogen rearrangement rules for the
prediction of fragmentation [152]. Its application in the investigation of the co-­
culture of Aspergillus sydowii with Bacillus subtilis allowed, for example, the iden-
tification of 25 newly biosynthesized metabolites, including 2 new compounds with
antibacterial activity, serine sydonate (20) and macrolactin U′ (21) [6]. In another
microbial interaction, between the endophytic fungi Copinforma mamane and
Fusarium solani, the use of the tool confirmed the biosynthetic induction of de novo
8 Discovering New Natural Products Using Metabolomics-Based Approaches 207

compounds that previously had not been produced in monoculture; altenuene (22),
pestalotin (23) and N-palmitoyl proline (24) [153].
Although this type of methodology has been helping in the structural elucidation
of unknown targets, there are phenomena and mechanistic behaviors that are intrin-
sic to it and that are not yet understood for given fragmentations [153]. In contrast,
methods that employ combinatorial fragmentation are not based on theoretical rules
of fragmentation, but rather on the best combination of molecular breaks [154]. In
these cases, the indication of more likely candidate metabolites results from combi-
natorial data processing, in which there is the ranking of the best options for match-
ing between a given molecular substructure (estimated by the software) and
fragment ions detected in the MS/MS spectra, regardless of whether there is a mech-
anistic justification that explains such correspondence and the observed fragmenta-
tion [136, 154, 155]. The MetFrag algorithm present in the METLIN platform is one
of the examples that uses such a method. Its application allows the identification of
compounds through a very rapid screening among thousands of compounds con-
sulted in molecular libraries that are compatible with the information of the data
acquired from MS/MS [43, 126, 151]. Its use during identification studies of the
composition of organic matter in marine environments promoted the annotation for
the first time of lyso-sulfolipid with a C14:0 fatty acid chain, 2-Hydroxy-3 – (tetra-
decanoyloxy)propyl 6-deoxy-6-sulfo-α-D-mannopyranoside (25), in the cyanobac-
teria Synechococcus elongatus [156]. Despite the recent refinement in its algorithm
system and scoring methodology [157], its use for annotations of microbial com-
pounds seems to be less frequent when compared to the number of studies reporting
its use in plant metabolomics. Therefore, the application of MetFrag in combination
with other platforms, such as Lipidmaps [158], or its recent incorporation into other
tools, such as the NAP tool (mentioned below), has significant potential for applica-
tion in workflows for the discovery of compounds produced by microorganisms.
Finally, nowadays, methods that include fragmentation trees represent intense
employability for the prediction of fingerprinting fragments in silico, and present
increasing prominence in annotation processes. In this approach, the peaks of pre-
cursor ions detected in the MSn spectra are annotated according to the predicted
molecular formulas and in the sequence; the relationship between these peaks and
their respective fragment ions is established through the observed mass losses. The
most plausible molecular formula for each loss is considered in each case.
The results are visualized via a graph (Fig. 8.5), which is similar to a fragmenta-
tion diagram, named as “tree”, in which each molecular formula represents a node
corresponding to a given structure or substructures. These nodes are connected by a
bridge (edge). If a node is a subformula of another node, the bridge indicates a prob-
able loss. All possible fragmentation trees are computed, but combinatorial calcula-
tions optimize for the best tree suggestion, in other words, the one that explains the
greatest number of detected peaks. Obtaining a fragmentation tree does not require
the use of spectral libraries or molecule databases, and therefore can be used to
identify the molecular formula of any unknown metabolite [33, 149, 159].
SIRIUS software represents one of the modern computational tools that best
employs fragmentation tree methods for fingerprint prediction. In association with
208 L. S. de Medeiros et al.

Fig. 8.5 (a) Sclerotiorin MS2 spectrum as a representative model for fragmentation tree data
processing. (b) Optimal fragmentation tree graph of sclerotiorin computed from MS2 data, indi-
cated by solid lines (c) Fragmentation mechanism of sclerotiorin and the corresponding molecular
formulas from fragment ions [160]. (Adapted from Ref. [161])

screening in libraries, it offers higher rates of molecular identification with high

levels of assertiveness of results [161]. In its most current version, SIRIUS 4, the
CSI:FingerID tool was incorporated, which is used as a web service precisely to
confront fragmentation information in silico in databases of molecular structures.
Additionally, the tool performs operations via algorithms trained from machine-­
learning techniques to provide more plausible molecular predictions, including of
unknown compounds, through a score system for classification of molecular candi-
dates, thereby increasing the rates of correct identifications.
The current workflow proposed by the use of the SIRIUS platform promises to
offer high performance annotations, with a high degree of accuracy, reaching about
70% assertiveness in identifications [159]. The recent application of these tools has
aided, for example, the discovery of several N-acyl amides (1 N-acyl glycine (26),
N-acyl alanine (27), N-acyloxyacyl lysine (28), N-acyloxyacyl glutamine (29),
N-acyl serinol (30), N-acyl lysine (31), N-acyl ornithine (32)) that are produced by
commensal bacteria and which appear to be associated with the regulation of gas-
trointestinal mechanisms by the human gut microbiota. Such compounds had not
previously been detected via searches in traditional libraries [161, 162].
The COSMIC (Confidence of Small Molecule IdentifiCations) [163] and
CANOPUS [164] tools are other apparatuses of automated use that have also
recently been added to the SIRIUS platform. COSMIC is a workflow that provides
greater reliability to the metabolic annotation process and eliminates the need for
queries in spectral libraries [163]. Even if it is not possible to confirm whether a
suggested annotation is completely right or wrong, by using the tool, it is possible
to assign degrees of confidence to annotations previously predetermined solely as
correct or incorrect in CSI:FingerID. In other words, in this case, one annotation
8 Discovering New Natural Products Using Metabolomics-Based Approaches 209

may have a higher level of accuracy than another, although both were classified as
probably correct. In studies also directed to the intestinal microbiota, the application
of COSMIC promoted the identification of 12 new natural bile acid conjugates,
such as phenylalanine-conjugated chenodeoxycholic acid (33) found in Escherichia
coli and in bacteria of the genus Stenotrophomonas [163]. In comparison, CANOPUS
is an alternative that can be used for putative identification of new scaffolds via the
prediction of classes of unknown compounds from its high-resolution MS/MS data.
The software uses deep neutral networks in the inference of the biosynthetic clas-
sification of metabolites. Through the use of this resource, coumarin ferulenol (34),
indicated as a chemotherapeutic agent, was detected in microbial cell-engineering
experiments with recombinant strains of E. coli [165].
Global Natural Products Social Molecular Networking is a web-based mass
spectrometry ecosystem that certainly represents one of the most used platforms in
research of natural products today, and has become prevalent even in microbial
metabolomics. More than a mere analytical option for data processing, the use of
the platform has become an essential strategy not only for molecular dereplication,
for the identification of structural analogues of known metabolites, but also for the
discovery of specimens with unknown scaffolds. This is because, nowadays, the
platform offers approximately 50 in silico tools and worflows for molecular mining,
with free public access. Among the most recently launched is the DERREPLICATOR+
algorithm, which offers specialized dereplication in ribosomal and non-ribosomal
peptides, as well as polyketides, flavonoids and terpenes, from the detection of simi-
lar characteristics between the experimental fragmentation profiles and the MS/MS
spectra generated in silico. In this aspect, correlations are also stipulated based on
theoretical predictions from structures deposited in databases. The use of the
DERREPLICATOR+ in investigations of microbial interactions with
Pseudoalteromonas strains aided in the isolation of two new dibrominated chromo-
peptides, dibromoalterochromides D (35) and D′ (36) [166]. The unexpected pro-
duction of several cryptic cyclicdepsipeptides in Neurospora crassa, such as
myxochromide S2 (37) and syringostatin A (38), were also detected with the imple-
mentation of the tool, through the analysis of extracts of the fungus grown in culture
media supplemented with ionic liquids based on choline [167].
NAP (Network Annotation Propagation) [150] and MS2LDA [168] configure
other possibilities for the mining of data acquired in tandem mode through auto-
mated processing, and can be manipulated directly through the GNPS platform. In
order to improve the process of molecular identifications, the NAP workflow pro-
poses to perform the re-ranking of annotations of candidate structures, which are
incorporated into the same molecular family, to suggest metabolites not previously
annotated in preliminary processes. The operation of the tool is inspired respec-
tively by the combinatorial fragmentation algorithms MetFrag and MetFusion. Its
methodology stems from the consistency of the structural similarity between the
molecules grouped in the same network. Assuming that molecules grouped in a
given family are structurally similar to each other, the structural information shared
between entities can be propagated among any member of a family. For this
210 L. S. de Medeiros et al.

reclassification of annotations, NAP employs precisely the propagation of structural

information intrinsic to the given molecular family, and explores the topology of
molecular networks of grouped metabolites [136]. Additionally, according to the
compatibility of candidate structures found in the MetFrag database by in silico
predictions, these are also annotated and combined with the re-ranking of detected
and predicted metabolites. The execution of this workflow led, for example, to the
detection of potentially novel bioactive metabolites observed in metabolomic stud-
ies of cyanobacteria and in co-culture between Streptomyces lunalinharesii and
Rhizoctonia solani [169], and also aided in the isolation and identification of cryp-
toporic acid T (39), a novel trimeric cryptoporic acid produced by the basideomy-
cete Cryptoporus volvatus [170].
MS2LDA is a tool that enables the annotation of compounds through informa-
tion on the sharing of chemically relevant portions between molecules, in other
words, when there is synchronicity between molecular substructures [168]. These
molecular moieties, called Mass2Motifs, represent groups of unsupervised product
and neutral loss ions, which are detected in common among mass features [171].
Even if there is low similarity between molecules in the total fragmentation profile,
the processing of Mass2Motifs by algorithms present in MS2LDA enables grouping
of molecules that have coincident structural portions. Multiple Mass2Motifs can
co-exist for a single MS/MS spectrum, but the prevalence of a motif can be ascer-
tained in different samples in large datasets by processing via MS2LDA. A database
of Mass2Motifs from analyses via MS2LDA, with open data access and storage
system, called MotifDB, was recently launched [171], which allows the approach to
be used even for the annotation of molecules that are not necessarily located in
known spectral databases. The integration of the software in investigations of the
secondary metabolism of the fungus of marine origin Aspergillus sp. LS116 led, for
example, to the discovery of two rare skeletons of tetracyclic alkaloids, perinadine
B (40) and C (41) [172].
Many halogenated compounds of natural origin are described with diverse bio-
logical properties. It is estimated that up to 20% of still undiscovered natural prod-
ucts contain halogen atoms in their constitution [173]. In this scenario, automated
approaches to the targeted search for halogenated structures also reveal an impor-
tant strategy for maximizing the discovery of novel bioactive microbial metabolites.
The MeHaloCoA (Marine Halogenated Compound Analysis) algorithm developed
in R provided, for example, the discovery of antiproliferative polyketides, griseo-
phenone I (42) and chlorogriseofulvin (43) from Penicillium canescens [174].
Although the analysis of complex extracts can be performed via this device to filter
compounds containing for example Cl and Br, metabolites with masses greater than
800 Da, multicharged or containing S atoms, may not be detected or differentiated.
In these cases, the automated dynamic cluster analysis (DCA) technique offers the
possibility of identifying Cl, Br, S and other A+2 element containing metabolites in
untargeted MS data analysis, including species above 800 Da [173]. Its application
has allowed, for example, the detection of several undescribed prymnesin-like
molecular features in cultures of the microalgae Prymnesium parvum.
8 Discovering New Natural Products Using Metabolomics-Based Approaches 211

8.4.2.2 Examples in Nuclear Magnetic Resonance

Examples of in silico tools based on NMR data are much less frequent when com-
pared to tools based on MS data, and this is possibly due to the greater speed of
development and frequency of implementation of the latter [133]. Among the soft-
ware dedicated to natural products, SMART 2.0 from 2020 and MADbyTE from
2021 can be cited. The first is based on the convolution of neural networks and
system learning that is trained with HSQC spectra for annotation of natural products
[9]. The second, allows the dereplication of a large data set through the construction
of spin system feature networks based on spectral data from HSQC and TOCSY
[175]. As MADbyTE requires a spectral database, as a proof of concept, in Flores-­
Bocanegra et al. (2022) [175], an in-house database of HSQC and TOCSY spectra
was built from nineteen resorcyclic acid lactones and ten spirobisnaphthalenes for
targeted screening of compounds of these classes in extracts of species of the genera
Setophoma, Neosetophoma, Pochnia and Pleosporales. In this case, MADbyTE
was used as an automated dereplication tool, which, via information from NMR, led
to the guided fractionation of microbial extracts, and gave rise to the isolation of
three new palmarumycins, CP20 (44), CP21 (45) and CP22 (46).
NMR-based in silico tools have been shown to be effective in the discovery of
bioactive compounds of microorganisms, especially when combined with derepli-
cation protocols based on biological data and also MS data. One example is the
discovery of a new cytotoxic macrolide called symplocolide A (47), which was
obtained from the marine cyanobacteria Symploca sp. The study implemented the
biomonitoring of molecules against cytotoxicity in cancer cells, and integrated the
prioritization of molecular targets through the use of the SMART 2.0 platform and
molecular networks using data from MS2 [9, 176].
Integrated applications with other omics sciences are also possible, such as the
work of Hu and collaborators (2021) [177] who discovered a new allenic macrolide
called archangiumide (48), which is produced by the myxobacterium Archangium
violaceum SDU8. Genome mining revealed a biosynthetic gene cluster (BGC) asso-
ciated with macrolide biosynthesis, whose prediction indicated an allene portion
previously unseen in polyketide macrolides. To verify the expression of BGC, the
authors obtained the chemical profile of the extracts based on NMR methods and
observed the presence of this type of compound also through the SMART 2.0 tool.
Subsequently, manual curation of HMBC and COSY spectra indicated partial mac-
rolide structures corresponding to the prediction based on genome mining. Finally,
the extract obtained in large-scale cultivation was fractionated and, aided by NMR,
led to the isolation of archangiumide.
The application of NMR for metabolomic analysis results in large datasets, usu-
ally generating overlapping signals and making them difficult to differentiate
metabolites in complex samples. To overcome this problem, deconvolution tools
have been developed, in order to make easier the identification of compounds
through NMR-based metabolomics studies [178]. In silico tools firstly developed
were aimed at the analysis of biofluid metabolites, but during the last years, tools for
processing natural products have emerged.
212 L. S. de Medeiros et al.

Some examples are: Bayesian Automated Netabolite Analyser (BATMAN) (Hao

et al., 2012) [179]; Bayesil [180] (Ravanbakhsh et al., 2015); Automated
Quantification Algorithm (AQuA) [181] (Röhnisch et al., 2017); Automatic
Statistical Identification in Complex Spectra [182] (ASICS) (Tardivel et al., 2017);
rDolphin [182] (Cañueto et al., 2018); COLMARm [183] (Bingol et al., 2016);
Metabolomics and Dereplication by Two-Dimensional Experiments (MADbyTE)
[184] (Egan et al., 2021); MetaboMiner [185] (Xia et al., 2008); Mixture of Natural
Products (MixONat) [186] (Bruguière et al., 2020); Small Molecule Accurate
Recognition Technology (SMART) [187] (Zhang et al., 2017).
These tools are used for the identification and/or quantification of metabolites by
means of NMR 1D (1H, 13C e DEPT) spectra and/or 2D (TOCSY, COSY, HSQC e
HMBC). Particularly regarding the last ones, MixONat enables the dereplication of
natural products by the use of 1D spectra and it is able to distinguish similar struc-
tural compounds [186] (Bruguière et al., 2020), while SMART provides the derep-
lication and other analysis based on artificial intelligence technologies. The method
uses NMR 2D (HSQC) data to match isolated natural products to the corresponding
analogues through neural networks [187] (ZHANG et al., 2017).

8.4.3 Multivariate Statistical Analyses

Within the context of the biology of systems, connecting an abundant universe of

spectral information to the chemical and biological functions of compounds is one
of the challenges in microbial metabolomic analysis. In this sense, multivariate sta-
tistical analysis models assume an essential role for the organization and interpreta-
tion of massive and highly complex data from NMR and MS [51, 141, 146].
Therefore, it is also possible and advantageous to integrate these types of tools to
workflows aimed at the discovery of bioactive compounds.
For the discrimination of metabolites in large sets, multivariate analyses are
based on the evaluation of the extracted data matrix, from the projection of values
and series of variations, and the relationship between all the variables involved.
Multivariate methods can apply supervised and unsupervised approaches to analy-
sis [188, 189]. In the first case, models that have already been labeled (labeled data)
are used in the training of patterns to lead to the prediction of results, as is the case
of partial least squares (PLS-DA), and orthogonal projections for latent structures
(OPLS-DA) [190]. On the other hand, for unsupervised methods, there is no basis
in labels and, therefore, they impact on characteristics of greater impartiality (unbi-
ased) in the analyses. With this, there is the contribution to the delineation of trends
to the data set, thereby favoring the prediction of models and the detection of clus-
ters. Among them is one of the most-used methods in metabolomics, principal com-
ponent analysis (PCA), as well as hierarchical clustering analysis (HCA) [146,
191, 192].
Abdelmohsen et al. (2014) identified actinosporins A (49) and B (50), glycosyl-
ated polyketides with antitrypanosome activity, in an extract obtained from the
8 Discovering New Natural Products Using Metabolomics-Based Approaches 213

bacterium Actinokineospora SP EG49. The authors used NMR and MS spectra in a

complementary way in the dereplication of the chemical constituents of the extract
and noted several anthraquinone derivatives. Through the combined methods of
PCA, HCA and OPLS-DA, they evaluated extracts from different forms of cultiva-
tion for the selection of variants of this biosynthetic subclass, which revealed the
two new O-glycosylkated angucyclines [193].
It should be noted that multivariate statistical analyses can be performed through
platforms, such as MetaboNexus [194] and MetaboAnalyst [195], which integrate
data preprocessing. The use of the combination of supervised and unsupervised
multivariate analyses via the MetaboAnalyst platform in metabolomic studies of
Colletotrichum species aided in the discovery of the 26-member macrolactone, a
colletotrichumolide (51) from a Brazilian fungal lineage [196]. However, through
these and other examples described in the literature [169, 197–200], it is clear that
the implementation of statistical techniques in a single or combined way can also
function as a prioritization mechanism for prospective microbial species, thus facili-
tating the selection of molecular targets of interest [28].

8.5 Conclusion

The use of metabolomic approaches in research of natural products is in full prog-

ress. This has been driven by constant developments in analytical instrumentation
and computational methods and has resulted in more and more molecules being
discovered via approaches involving this methodology. In addition, through the
review of the literature, it is observed that, with the involvement of genomics
approaches, the results have been even more expressive, with the discovery of many
molecules of the most varied classes. In this sense, the branch of chemistry of
microbial natural products increasingly shows promise in providing the molecules
of the future. In addition, this chapter indicates the need for the expansion of the use
of such analytical and computational tools that make it possible to direct the discov-
ery of new molecules that are useful to humanity.

References

1. Kjærbølling I, Mortensen UH, Vesth T, Andersen MR (2019) Strategies to establish the

link between biosynthetic gene clusters and secondary metabolites. Fungal Genet Biol
130:107–121. https://doi.org/10.1016/j.fgb.2019.06.001
2. Ramírez-Rendon D, Passari AK, Ruiz-Villafán B et al (2022) Impact of novel microbial
secondary metabolites on the pharma industry. Appl Microbiol Biotechnol 106:1855–1878.
https://doi.org/10.1007/s00253-­022-­11821-­5
3. Gruber-Dorninger C, Novak B, Nagl V, Berthiller F (2017) Emerging mycotoxins: beyond
traditionally determined food contaminants. J Agric Food Chem 65:7052–7070. https://doi.
org/10.1021/acs.jafc.6b03413
214 L. S. de Medeiros et al.

4. Caesar LK, Kellogg JJ, Kvalheim OM, Cech NB (2019) Opportunities and limitations for
untargeted mass spectrometry metabolomics to identify biologically active constituents
in complex natural product mixtures. J Nat Prod 82:469–484. https://doi.org/10.1021/acs.
jnatprod.9b00176
5. Wolfender J-L, Nuzillard J-M, van der Hooft JJJ et al (2019) Accelerating metabolite iden-
tification in natural product research: toward an ideal combination of liquid chromatogra-
phy–high-resolution tandem mass spectrometry and NMR profiling, in silico databases, and
chemometrics. Anal Chem 91:704–742. https://doi.org/10.1021/acs.analchem.8b05112
6. Sun Y, Liu W-C, Shi X et al (2021) Inducing secondary metabolite production of Aspergillus
sydowii through microbial co-culture with Bacillus subtilis. Microb Cell Factories 20:42.
https://doi.org/10.1186/s12934-­021-­01527-­0
7. Wakefield J, Hassan HM, Jaspars M et al (2017) Dual induction of new microbial second-
ary metabolites by fungal bacterial co-cultivation. Front Microbiol 8:1284. https://doi.
org/10.3389/fmicb.2017.01284
8. Thornburg CC, Britt JR, Evans JR et al (2018) NCI program for natural product discovery: a
publicly-accessible library of natural product fractions for high-throughput screening. ACS
Chem Biol 13:2484–2497. https://doi.org/10.1021/acschembio.8b00389
9. Reher R, Kim HW, Zhang C et al (2020) A convolutional neural network-based approach for
the rapid annotation of molecularly diverse natural products. J Am Chem Soc 142:4114–4120.
https://doi.org/10.1021/jacs.9b13786
10. Misra BB, van der Hooft JJJ (2016) Updates in metabolomics tools and resources: 2014–2015.
Electrophoresis 37:86–110. https://doi.org/10.1002/elps.201500417
11. Allard P-M, Genta-Jouve G, Wolfender J-L (2017) Deep metabolome annotation in natural
products research: towards a virtuous cycle in metabolite identification. Curr Opin Chem Biol
36:40–49. https://doi.org/10.1016/j.cbpa.2016.12.022
12. Caesar LK, Cech NB (2019) Synergy and antagonism in natural product extracts: when 1 + 1
does not equal 2. Nat Prod Rep 36:869–888. https://doi.org/10.1039/C9NP00011A
13. Palaniappan K, Chen I-MA, Chu K et al (2020) IMG-ABC v.5.0: an update to the IMG/Atlas
of Biosynthetic Gene Clusters Knowledgebase. Nucleic Acids Res 48:D422–D430. https://
doi.org/10.1093/nar/gkz932
14. Hautbergue T, Jamin EL, Debrauwer L et al (2018) From genomics to metabolomics, moving
toward an integrated strategy for the discovery of fungal secondary metabolites. Nat Prod
Rep 35:147–173. https://doi.org/10.1039/C7NP00032D
15. Gross H (2007) Strategies to unravel the function of orphan biosynthesis pathways:
recent examples and future prospects. Appl Microbiol Biotechnol 75:267–277. https://doi.
org/10.1007/s00253-­007-­0900-­5
16. Bode HB, Bethe B, Höfs R, Zeeck A (2002) Big effects from small changes: possible ways
to explore nature’s chemical diversity. ChemBioChem 3:619–627. https://doi.org/10.1002/
1439-­7633(20020703)3:7<619::AID-­CBIC619>3.0.CO;2-­9
17. Zhang X, Hindra EMA (2019) Unlocking the trove of metabolic treasures: activating silent
biosynthetic gene clusters in bacteria and fungi. Curr Opin Microbiol 51:9–15. https://doi.
org/10.1016/j.mib.2019.03.003
18. Luo Y, Cobb RE, Zhao H (2014) Recent advances in natural product discovery. Curr Opin
Biotechnol 30:230–237. https://doi.org/10.1016/j.copbio.2014.09.002
19. Pfannenstiel BT, Keller NP (2019) On top of biosynthetic gene clusters: how epigenetic
machinery influences secondary metabolism in fungi. Biotechnol Adv 37:107345. https://
doi.org/10.1016/j.biotechadv.2019.02.001
20. Baral B, Akhgari A, Metsä-Ketelä M (2018) Activation of microbial secondary meta-
bolic pathways: avenues and challenges. Synth Syst Biotechnol 3:163–178. https://doi.
org/10.1016/j.synbio.2018.09.001
21. Mattern DJ, Valiante V, Unkles SE, Brakhage AA (2015) Synthetic biology of fungal natural
products. Front Microbiol 6:775. https://doi.org/10.3389/fmicb.2015.00775
22. Ziemert N, Alanjary M, Weber T (2016) The evolution of genome mining in microbes – a
review. Nat Prod Rep 33:988–1005. https://doi.org/10.1039/C6NP00025H
8 Discovering New Natural Products Using Metabolomics-Based Approaches 215

23. Scherlach K, Hertweck C (2021) Mining and unearthing hidden biosynthetic potential. Nat
Commun 12:3864. https://doi.org/10.1038/s41467-­021-­24133-­5
24. Locey KJ, Lennon JT (2016) Scaling laws predict global microbial diversity. Proc Natl Acad
Sci 113:5970–5975. https://doi.org/10.1073/pnas.1521291113
25. Hawksworth DL, Lücking R (2017) Fungal diversity revisited: 2.2 to 3.8 million species. In:
Heitman J, Howlett BJ, Crous PW et al (eds) The fungal kingdom. ASM Press, Washington,
DC, pp 79–95
26. Patti GJ, Yanes O, Siuzdak G (2012) Metabolomics: the apogee of the omics trilogy. Nat Rev
Mol Cell Biol 13:263–269. https://doi.org/10.1038/nrm3314
27. Brunetti AE, Carnevale Neto F, Vera MC et al (2018) An integrative omics perspective for the
analysis of chemical signals in ecological interactions. Chem Soc Rev 47:1574–1591. https://
doi.org/10.1039/C7CS00368D
28. Covington BC, McLean JA, Bachmann BO (2017) Comparative mass spectrometry-based
metabolomics strategies for the investigation of microbial secondary metabolites. Nat Prod
Rep 34:6–24. https://doi.org/10.1039/C6NP00048G
29. Blaženović I, Kind T, Ji J, Fiehn O (2018) Software tools and approaches for compound iden-
tification of LC-MS/MS data in metabolomics. Metabolites 8:31. https://doi.org/10.3390/
metabo8020031
30. Tawfike AF, Viegelmann C, Edrada-Ebel R (2013) Metabolomics and dereplication strategies
in natural products. In: Roessner U, Dias DA (eds) Metabolomics tools for natural product
discovery, Methods in molecular biology, vol 1055. Humana, Totowa, pp 227–244
31. Wolfender J-L, Rudaz S, Hae Choi Y, Kyong Kim H (2013) Plant metabolomics:
from holistic data to relevant biomarkers. Curr Med Chem 20:1056–1090. https://doi.
org/10.2174/092986713805288932
32. Genilloud O (2019) Natural products discovery and potential for new antibiotics. Curr Opin
Microbiol 51:81–87. https://doi.org/10.1016/j.mib.2019.10.012
33. Krug D, Müller R (2014) Secondary metabolomics: the impact of mass spectrometry-based
approaches on the discovery and characterization of microbial natural products. Nat Prod Rep
31:768–783. https://doi.org/10.1039/c3np70127a
34. Barkal LJ, Theberge AB, Guo C-J et al (2016) Microbial metabolomics in open microscale
platforms. Nat Commun 7:10610. https://doi.org/10.1038/ncomms10610
35. Fenaille F, Barbier Saint-Hilaire P, Rousseau K, Junot C (2017) Data acquisition work-
flows in liquid chromatography coupled to high resolution mass spectrometry-based
metabolomics: where do we stand? J Chromatogr A 1526:1–12. https://doi.org/10.1016/j.
chroma.2017.10.043
36. Atanasov AG, Zotchev SB, Dirsch VM, Supuran CT (2021) Natural products in drug discov-
ery: advances and opportunities. Nat Rev Drug Discov 20:200–216. https://doi.org/10.1038/
s41573-­020-­00114-­z
37. Mapelli V, Olsson L, Nielsen J (2008) Metabolic footprinting in microbiology: methods
and applications in functional genomics and biotechnology. Trends Biotechnol 26:490–497.
https://doi.org/10.1016/j.tibtech.2008.05.008
38. Hubert J, Nuzillard J-M, Renault J-H (2017) Dereplication strategies in natural product
research: how many tools and methodologies behind the same concept? Phytochem Rev
16:55–95. https://doi.org/10.1007/s11101-­015-­9448-­7
39. Alves S, Rathahao-Paris E, Tabet JC (2013) Potential of Fourier transform mass spectrom-
etry for high-throughput metabolomics analysis. In: Rolin D (ed) Metabolomics coming of
age with its technological diversity, Advances in botanical research, vol vol 67. Academic,
Amsterdam, pp 219–302
40. Ghosh R, Bu G, Nannenga BL, Sumner LW (2021) Recent developments toward integrated
metabolomics technologies (UHPLC-MS-SPE-NMR and MicroED) for higher-throughput
confident metabolite identifications. Front Mol Biosci 8:720955. https://doi.org/10.3389/
fmolb.2021.720955
216 L. S. de Medeiros et al.

41. Henke MT, Kelleher NL (2016) Modern mass spectrometry for synthetic biology and
structure-­
based discovery of natural products. Nat Prod Rep 33:942–950. https://doi.
org/10.1039/C6NP00024J
42. Wu C, Choi YH, van Wezel GP (2016) Metabolic profiling as a tool for prioritizing anti-
microbial compounds. J Ind Microbiol Biotechnol 43:299–312. https://doi.org/10.1007/
s10295-­015-­1666-­x
43. Caesar LK, Montaser R, Keller NP, Kelleher NL (2021) Metabolomics and genomics in natu-
ral products research: complementary tools for targeting new chemical entities. Nat Prod Rep
38:2041–2065. https://doi.org/10.1039/D1NP00036E
44. Bertrand S, Schumpp O, Bohni N et al (2013) De novo production of metabolites by fungal
co-culture of Trichophyton rubrum and Bionectria ochroleuca. J Nat Prod 76:1157–1165.
https://doi.org/10.1021/np400258f
45. De Vijlder T, Valkenborg D, Lemière F et al (2017) A tutorial in small molecule identifica-
tion via electrospray ionization-mass spectrometry: the practical art of structural elucidation.
Mass Spectrom Rev 37:607–629. https://doi.org/10.1002/mas.21551
46. Stricker T, Bonner R, Lisacek F, Hopfgartner G (2021) Adduct annotation in liquid chro-
matography/high-resolution mass spectrometry to enhance compound identification. Anal
Bioanal Chem 413:503–517. https://doi.org/10.1007/s00216-­020-­03019-­3
47. Machushynets NV, Elsayed SS, Du C et al (2022) Discovery of actinomycin L, a new mem-
ber of the actinomycin family of antibiotics. Sci Rep 12:2813. https://doi.org/10.1038/
s41598-­022-­06736-­0
48. Nielsen KF, Månsson M, Rank C et al (2011) Dereplication of microbial natural products by
LC-DAD-TOFMS. J Nat Prod 74:2338–2348. https://doi.org/10.1021/np200254t
49. Schmid R, Petras D, Nothias L-F et al (2021) Ion identity molecular networking for mass
spectrometry-based metabolomics in the GNPS environment. Nat Commun 12:3832. https://
doi.org/10.1038/s41467-­021-­23953-­9
50. Cajka T, Fiehn O (2016) Toward merging untargeted and targeted methods in mass
spectrometry-­ based metabolomics and lipidomics. Anal Chem 88:524–545. https://doi.
org/10.1021/acs.analchem.5b04491
51. Percival B, Gibson M, Leenders J et al (2020) Univariate and multivariate statistical
approaches to the analysis and interpretation of NMR-based metabolomics datasets of
increasing complexity. In: Wilson PB, Grootveld M (eds) Computational techniques for ana-
lytical chemistry and bioanalysis. The Royal Society of Chemistry, London, pp 1–40
52. Mohimani H, Gurevich A, Shlemov A et al (2018) Dereplication of microbial metabolites
through database search of mass spectra. Nat Commun 9:4035. https://doi.org/10.1038/
s41467-­018-­06082-­8
53. Langlykke A (1980) Foreword to CRC handbook of antibiotic compounds. CRC, Boca Raton
54. Van Middlesworth F, Cannell RJP (1998) Dereplication and partial identification of natural
products. In: Cannell RJP (ed) Natural products isolation, Methods in biotechnology, vol 4.
Humana, Totowa, pp 279–327
55. Beutler JA, Alvarado AB, Schaufelberger DE et al (1990) Dereplication of phorbol bioactives:
Lyngbya majuscula and Croton cuneatus. J Nat Prod 53:867–874. https://doi.org/10.1021/
np50070a014
56. Phelan VV (2020) Feature-based molecular networking for metabolite annotation. In: Li S
(ed) Computational methods and data analysis for metabolomics, Methods in molecular biol-
ogy, vol 2104. Humana, New York, pp 227–243
57. Aron AT, Gentry EC, McPhail KL et al (2020) Reproducible molecular networking of
untargeted mass spectrometry data using GNPS. Nat Protoc 15:1954–1991. https://doi.
org/10.1038/s41596-­020-­0317-­5
58. Gaudêncio SP, Pereira F (2015) Dereplication: racing to speed up the natural products discov-
ery process. Nat Prod Rep 32:779–810. https://doi.org/10.1039/C4NP00134F
59. Přichystal J, Schug KA, Lemr K et al (2016) Structural analysis of natural products. Anal
Chem 88:10338–10346. https://doi.org/10.1021/acs.analchem.6b02386
8 Discovering New Natural Products Using Metabolomics-Based Approaches 217

60. Kluger B, Lehner S, Schuhmacher R (2015) Metabolomics and secondary metabolite profil-
ing of filamentous fungi. In: Zeilinger S, Martín J-F, García-Estrada C (eds) Fungal biology:
biosynthesis and molecular genetics of fungal secondary metabolites, vol Vol 2. Springer,
New York, pp 81–101
61. Sumner LW, Amberg A, Barrett D et al (2007) Proposed minimum reporting standards for
chemical analysis. Metabolomics 3:211–221. https://doi.org/10.1007/s11306-­007-­0082-­2
62. Schymanski EL, Jeon J, Gulde R et al (2014) Identifying small molecules via high resolution
mass spectrometry: communicating confidence. Environ Sci Technol 48:2097–2098. https://
doi.org/10.1021/es5002105
63. Rochat B (2017) Proposed confidence scale and ID score in the identification of known-­
unknown compounds using high resolution MS data. J Am Soc Mass Spectrom 28:709–723.
https://doi.org/10.1007/s13361-­016-­1556-­0
64. Wolfender J-L, Litaudon M, Touboul D, Queiroz EF (2019) Innovative omics-based
approaches for prioritisation and targeted isolation of natural products – new strategies for
drug discovery. Nat Prod Rep 36:855–868. https://doi.org/10.1039/C9NP00004F
65. van Santen JA, Kautsar SA, Medema MH, Linington RG (2021) Microbial natural product
databases: moving forward in the multi-omics era. Nat Prod Rep 38:264–278. https://doi.
org/10.1039/D0NP00053A
66. Johnson SR, Lange BM (2015) Open-Access metabolomics databases for natural product
research: present capabilities and future potential. Front Bioeng Biotechnol 3:22. https://doi.
org/10.3389/fbioe.2015.00022
67. Sorokina M, Steinbeck C (2020) Review on natural products databases: where to find data in
2020. J Cheminform 12:20. https://doi.org/10.1186/s13321-­020-­00424-­9
68. Laatsch H (2014) AntiBase 2014: the natural compound identifier. Wiley-VHC, Weinheim
69. van Santen JA, Jacob G, Singh AL et al (2019) The natural products Atlas: an open access
knowledge base for microbial natural products discovery. ACS Cent Sci 5:1824–1833. https://
doi.org/10.1021/acscentsci.9b00806
70. Klementz D, Döring K, Lucas X et al (2016) StreptomeDB 2.0—an extended resource of
natural products produced by streptomycetes. Nucleic Acids Res 44:D509–D514. https://doi.
org/10.1093/nar/gkv1319
71. Sorokina M, Merseburger P, Rajan K et al (2021) COCONUT online: Collection of Open
Natural Products database. J Cheminform 13:2. https://doi.org/10.1186/s13321-­020-­00478-­9
72. Petrova ON, Lamarre I, Fasani F et al (2020) Soluble guanylate cyclase inhibitors discov-
ered among natural compounds. J Nat Prod 83:3642–3651. https://doi.org/10.1021/acs.
jnatprod.0c00854
73. King ZA, Lu J, Dräger A et al (2016) BiGG models: a platform for integrating, standard-
izing and sharing genome-scale models. Nucleic Acids Res 44:D515–D522. https://doi.
org/10.1093/nar/gkv1049
74. Gilson MK, Liu T, Baitaluk M et al (2016) BindingDB in 2015: a public database for medici-
nal chemistry, computational chemistry and systems pharmacology. Nucleic Acids Res
44:D1045–D1053. https://doi.org/10.1093/nar/gkv1072
75. Pilón-Jiménez B, Saldívar-González F, Díaz-Eufracio B, Medina-Franco J (2019)
BIOFACQUIM: a Mexican compound database of natural products. Biomolecules 9:31.
https://doi.org/10.3390/biom9010031
76. Pence HE, Williams A (2010) ChemSpider: an online chemical information resource. J Chem
Educ 87:1123–1124. https://doi.org/10.1021/ed100697w
77. Shen J, Xu X, Cheng F et al (2003) Virtual screening on natural products for discover-
ing active compounds and target information. Curr Med Chem 10:2327–2342. https://doi.
org/10.2174/0929867033456729
78. Kong D-X, Jiang Y-Y, Zhang H-Y (2010) Marine natural products as sources of novel scaf-
folds: achievement and concern. Drug Discov Today 15:884–886. https://doi.org/10.1016/j.
drudis.2010.09.002
218 L. S. de Medeiros et al.

79. Wishart DS, Feunang YD, Guo AC et al (2018) DrugBank 5.0: a major update to the
DrugBank database for 2018. Nucleic Acids Res 46:D1074–D1082. https://doi.org/10.1093/
nar/gkx1037
80. Wang M, Carver JJ, Phelan VV et al (2016) Sharing and community curation of mass spec-
trometry data with Global Natural Products Social Molecular Networking. Nat Biotechnol
34:828–837. https://doi.org/10.1038/nbt.3597
81. Wishart DS, Feunang YD, Marcu A et al (2018) HMDB 4.0: the human metabolome database
for 2018. Nucleic Acids Res 46:D608–D617. https://doi.org/10.1093/nar/gkx1089
82. Chen Y, de Bruyn KC, Kirchmair J (2017) Data resources for the computer-guided discovery
of bioactive natural products. J Chem Inf Model 57:2099–2111. https://doi.org/10.1021/acs.
jcim.7b00341
83. Zhang R, Lin J, Zou Y et al (2019) Chemical space and biological target network of anti-­
inflammatory natural products. J Chem Inf Model 59:66–73. https://doi.org/10.1021/acs.
jcim.8b00560
84. Singh P, Bast F (2015) Screening of multi-targeted natural compounds for receptor tyrosine
kinases inhibitors and biological evaluation on cancer cell lines, in silico and in vitro. Med
Oncol 32:233. https://doi.org/10.1007/s12032-­015-­0678-­8
85. Kanehisa M, Furumichi M, Tanabe M et al (2017) KEGG: new perspectives on genomes,
pathways, diseases and drugs. Nucleic Acids Res 45:D353–D361. https://doi.org/10.1093/
nar/gkw1092
86. Lei J, Zhou J (2002) A marine natural product database. J Chem Inf Comput Sci 42:742–748.
https://doi.org/10.1021/ci010111x
87. Blunt JW, Carroll AR, Copp BR et al (2018) Marine natural products. Nat Prod Rep 35:8–53.
https://doi.org/10.1039/C7NP00052A
88. Horai H, Arita M, Kanaya S et al (2010) MassBank: a public repository for sharing mass spec-
tral data for life sciences. J Mass Spectrom 45:703–714. https://doi.org/10.1002/jms.1777
89. Haug K, Cochrane K, Nainala VC et al (2019) MetaboLights: a resource evolving in response
to the needs of its scientific community. Nucleic Acids Res 48:D440–D444. https://doi.
org/10.1093/nar/gkz1019
90. Wajiki M, Yamamoto T, Maruki-Uchida H et al (2022) Effect of sucrose on amino acid absorp-
tion of whey: a randomized crossover trial. Metabolites 12:282. https://doi.org/10.3390/
metabo12040282
91. Caspi R, Billington R, Fulcher CA et al (2018) The MetaCyc database of metabolic pathways
and enzymes. Nucleic Acids Res 46:D633–D639. https://doi.org/10.1093/nar/gkx935
92. Guijas C, Montenegro-Burke JR, Domingo-Almenara X et al (2018) METLIN: a technol-
ogy platform for identifying knowns and unknowns. Anal Chem 90:3156–3164. https://doi.
org/10.1021/acs.analchem.7b04424
93. Li Y, Kind T, Folz J et al (2021) Spectral entropy outperforms MS/MS dot product simi-
larity for small-molecule compound identification. Nat Methods 18:1524–1531. https://doi.
org/10.1038/s41592-­021-­01331-­z
94. Verstraete A, Hyde E (2020) Une méthode de dépistage des NPS utilisant les plateformes
collaboratives HighresNPS et Mzcloud. Toxicol Anal Clin 32:S47–S48. https://doi.
org/10.1016/j.toxac.2020.09.017
95. Monga M, Sausville E (2002) Developmental Therapeutics Program at the NCI: molecu-
lar target and drug discovery process. Leukemia 16:520–526. https://doi.org/10.1038/
sj.leu.2402464
96. Linstrom P (2017) NIST chemistry WebBook – SRD 69. National Institute of Standards and
Technology. https://doi.org/10.18434/T4D303. Accessed 26 May 2023
97. Kuhn S, Schlörer NE (2015) Facilitating quality control for spectra assignments of small
organic molecules: nmrshiftdb2 – a free in-house NMR database with integrated LIMS for
academic service laboratories. Magn Reson Chem 53:582–589. https://doi.org/10.1002/
mrc.4263
8 Discovering New Natural Products Using Metabolomics-Based Approaches 219

98. Zeng X, Zhang P, He W et al (2018) NPASS: natural product activity and species source
database for natural product research, discovery and tool development. Nucleic Acids Res
46:D1217–D1222. https://doi.org/10.1093/nar/gkx1026
99. Capecchi A, Reymond J-L (2020) Assigning the origin of microbial natural products by
chemical space map and machine learning. Biomolecules 10:1385. https://doi.org/10.3390/
biom10101385
100. Tomiki T, Saito T, Ueki M et al (2006) RIKEN Natural Products Encyclopedia (RIKEN
NPEdia), a Chemical Database of RIKEN Natural Products Depository (RIKEN NPDepo). J
Comput Aided Chem 7:157–162. https://doi.org/10.2751/jcac.7.157
101. Williamson AE, Ylioja PM, Robertson MN et al (2016) Open source drug discovery: highly
potent antimalarial compounds derived from the tres cantos arylpyrroles. ACS Cent Sci
2:687–701. https://doi.org/10.1021/acscentsci.6b00086
102. Huang W, Brewer LK, Jones JW et al (2018) PAMDB: a comprehensive Pseudomonas aeru-
ginosa metabolome database. Nucleic Acids Res 46:D575–D580. https://doi.org/10.1093/
nar/gkx1061
103. Kim S, Thiessen PA, Bolton EE et al (2016) PubChem substance and compound databases.
Nucleic Acids Res 44:D1202–D1213. https://doi.org/10.1093/nar/gkv951
104. Banerjee P, Erehman J, Gohlke B-O et al (2015) Super Natural II—a database of natural
products. Nucleic Acids Res 43:D935–D939. https://doi.org/10.1093/nar/gku886
105. Ibrahim MAA, Abdeljawaad KAA, Abdelrahman AHM, Hegazy M-EF (2021) Natural-like
products as potential SARS-CoV-2 M pro inhibitors: in-silico drug discovery. J Biomol Struct
Dyn 39:5722–5734. https://doi.org/10.1080/07391102.2020.1790037
106. Gu J, Gui Y, Chen L et al (2013) Use of natural products as chemical library for drug dis-
covery and network pharmacology. PLoS ONE 8:e62839. https://doi.org/10.1371/journal.
pone.0062839
107. Ramirez-Gaona M, Marcu A, Pon A et al (2017) YMDB 2.0: a significantly expanded ver-
sion of the yeast metabolome database. Nucleic Acids Res 45:D440–D445. https://doi.
org/10.1093/nar/gkw1058
108. Sterling T, Irwin JJ (2015) ZINC 15 – ligand discovery for everyone. J Chem Inf Model
55:2324–2337. https://doi.org/10.1021/acs.jcim.5b00559
109. Ulrich EL, Akutsu H, Doreleijers JF et al (2007) BioMagResBank. Nucleic Acids Res
36:D402–D408. https://doi.org/10.1093/nar/gkm957
110. López-Pérez JL, Therón R, del Olmo E, Díaz D (2007) NAPROC-13: a database for the
dereplication of natural product mixtures in bioassay-guided protocols. Bioinformatics
23:3256–3257. https://doi.org/10.1093/bioinformatics/btm516
111. Steinbeck C, Krause S, Kuhn S (2003) NMRShiftDBConstructing a free chemical informa-
tion system with open-source components. J Chem Inf Comput Sci 43:1733–1739. https://
doi.org/10.1021/ci0341363
112. Steinbeck C, Kuhn S (2004) NMRShiftDB – compound identification and structure elucida-
tion support through a free community-built web database. Phytochemistry 65:2711–2717.
https://doi.org/10.1016/j.phytochem.2004.08.027
113. Yamamoto O, Someno K, Wasada N et al (1988) An integrated spectral data base system
including IR, MS, 1H-NMR, 13C-NMR, ESR and Raman spectra. Anal Sci 4:233–239.
https://doi.org/10.2116/analsci.4.233
114. Zani CL, Carroll AR (2017) Database for rapid dereplication of known natural products using
data from MS and fast NMR experiments. J Nat Prod 80:1758–1766. https://doi.org/10.1021/
acs.jnatprod.6b01093
115. Wishart DS, Sayeeda Z, Budinski Z et al (2022) NP-MRD: the natural products magnetic res-
onance database. Nucleic Acids Res 50:D665–D677. https://doi.org/10.1093/nar/gkab1052
116. Asakura K (2015) A NMR spectral database of natural products “CH-NMR-NP”. J Synth Org
Chem Japan 73:1247–1252. https://doi.org/10.5059/yukigoseikyokaishi.73.1247
220 L. S. de Medeiros et al.

117. de Medeiros LS, Abreu LM, Nielsen A et al (2015) Dereplication-guided isolation of

depsides thielavins S–T and lecanorins D–F from the endophytic fungus Setophoma sp.
Phytochemistry 111:154–162. https://doi.org/10.1016/j.phytochem.2014.12.020
118. de Medeiros L, da Silva J, Abreu L et al (2015) Dichlorinated and Brominated Rugulovasines,
Ergot Alkaloids produced by Talaromyces wortmannii. Molecules 20:17627–17644. https://
doi.org/10.3390/molecules200917627
119. Marner M, Patras MA, Kurz M et al (2020) Molecular networking-guided discovery and
characterization of Stechlisins, a group of cyclic lipopeptides from a Pseudomonas sp. J Nat
Prod 83:2607–2617. https://doi.org/10.1021/acs.jnatprod.0c00263
120. Harinantenaina Rakotondraibe L, Rasolomampianina R, Park H-Y et al (2015)
Antiproliferative and antiplasmodial compounds from selected Streptomyces species. Bioorg
Med Chem Lett 25:5646–5649. https://doi.org/10.1016/j.bmcl.2015.07.103
121. Tawfike AF, Romli M, Clements C et al (2019) Isolation of anticancer and anti-trypanosome
secondary metabolites from the endophytic fungus Aspergillus flocculus via bioactivity
guided isolation and MS based metabolomics. J Chromatogr B 1106–1107:71–83. https://
doi.org/10.1016/j.jchromb.2018.12.032
122. van Santen JA, Poynton EF, Iskakova D et al (2022) The Natural Products Atlas 2.0: a data-
base of microbially-derived natural products. Nucleic Acids Res 50:D1317–D1323. https://
doi.org/10.1093/nar/gkab941
123. Jones MR, Pinto E, Torres MA et al (2021) CyanoMetDB, a comprehensive public data-
base of secondary metabolites from cyanobacteria. Water Res 196:117017. https://doi.
org/10.1016/j.watres.2021.117017
124. Davis LJ, Maldonado AC, Khin M et al (2022) Aulosirazoles B and C from the Cyanobacterium
Nostoc sp. UIC 10771: analogues of an isothiazolonaphthoquinone scaffold that activate
nuclear transcription Factor FOXO3a in ovarian cancer cells. J Nat Prod 85:540–546. https://
doi.org/10.1021/acs.jnatprod.1c01030
125. Iwasaki A, Kurisawa N, Wang T et al (2021) Lingaoamide, a cyclic heptapeptide from a
Chinese freshwater cyanobacterium Oscillatoria sp. Tetrahedron Lett 75:153214. https://doi.
org/10.1016/j.tetlet.2021.153214
126. Montenegro-Burke JR, Guijas C, Siuzdak G (2020) METLIN: a tandem mass spectral library
of standards. In: Li S (ed) Computational methods and data analysis for metabolomics,
Methods in molecular biology, vol 2104. Humana, New York, pp 149–163
127. Fox Ramos AE, Evanno L, Poupon E et al (2019) Natural products targeting strategies involv-
ing molecular networking: different manners, one goal. Nat Prod Rep 36:960–980. https://
doi.org/10.1039/C9NP00006B
128. Qin G-F, Zhang X, Zhu F et al (2022) MS/MS-based molecular networking: an efficient
approach for natural products dereplication. Molecules 28:157. https://doi.org/10.3390/
molecules28010157
129. Leao TF, Clark CM, Bauermeister A et al (2021) Quick-start infrastructure for untar-
geted metabolomics analysis in GNPS. Nat Metab 3:880–882. https://doi.org/10.1038/
s42255-­021-­00429-­0
130. Yang JY, Sanchez LM, Rath CM et al (2013) Molecular networking as a dereplication strat-
egy. J Nat Prod 76:1686–1699. https://doi.org/10.1021/np400413s
131. Watrous J, Roach P, Alexandrov T et al (2012) Mass spectral molecular networking of liv-
ing microbial colonies. Proc Natl Acad Sci 109:E1743–E1752. https://doi.org/10.1073/
pnas.1203689109
132. Perez De Souza L, Alseekh S, Brotman Y, Fernie AR (2020) Network-based strategies in
metabolomics data analysis and interpretation: from molecular networking to biologi-
cal interpretation. Expert Rev Proteomics 17:243–255. https://doi.org/10.1080/1478945
0.2020.1766975
133. Beniddir MA, Bin KK, Genta-Jouve G et al (2021) Advances in decomposing complex
metabolite mixtures using substructure- and network-based computational metabolomics
approaches. Nat Prod Rep 38:1967–1993. https://doi.org/10.1039/D1NP00023C
8 Discovering New Natural Products Using Metabolomics-Based Approaches 221

134. Olivon F, Elie N, Grelier G et al (2018) MetGem software for the generation of molecular net-
works based on the t-SNE algorithm. Anal Chem 90:13900–13908. https://doi.org/10.1021/
acs.analchem.8b03099
135. Hoang TPT, Roullier C, Evanno L et al (2023) Nature-inspired chemistry of complex alkaloids:
combining targeted molecular networking approach and semisynthetic strategy to access rare
communesins in a marine-derived Penicillium expansum. Chem Eur J:e202300103. https://
doi.org/10.1002/chem.202300103
136. Bauermeister A, Mannochio-Russo H, Costa-Lotufo LV et al (2022) Mass spectrometry-­
based metabolomics in microbiome investigations. Nat Rev Microbiol 20:143–160. https://
doi.org/10.1038/s41579-­021-­00621-­9
137. Domingo-Almenara X, Siuzdak G (2020) Metabolomics data processing using XCMS. In:
Li S (ed) Computational methods and data analysis for metabolomics, Methods in molecular
biology, vol 2104. Humana, New York, pp 11–24
138. Du X, Smirnov A, Pluskal T et al (2020) Metabolomics data preprocessing using ADAP
and MZmine 2. In: Li S (ed) Computational methods and data analysis for metabolomics,
Methods in molecular biology, vol 2104. Humana, New York, pp 25–48
139. Pathmasiri W, Kay K, McRitchie S, Sumner S (2020) Analysis of NMR metabolomics data.
In: Li S (ed) Computational methods and data analysis for metabolomics, Methods in molec-
ular biology, vol 2104. Humana, New York, pp 61–97
140. Kumar K, Schweiggert R, Patz C-D (2020) Introducing a novel procedure for peak alignment
in one-dimensional 1 H-NMR spectroscopy: a prerequisite for chemometric analyses of wine
samples. Anal Methods 12:3626–3636. https://doi.org/10.1039/D0AY01011A
141. Perez de Souza L, Naake T, Tohge T, Fernie AR (2017) From chromatogram to analyte to
metabolite. How to pick horses for courses from the massive web resources for mass spectral
plant metabolomics. Gigascience 6:gix037. https://doi.org/10.1093/gigascience/gix037
142. Jacob D, Deborde C, Lefebvre M et al (2017) NMRProcFlow: a graphical and interactive
tool dedicated to 1D spectra processing for NMR-based metabolomics. Metabolomics 13:36.
https://doi.org/10.1007/s11306-­017-­1178-­y
143. Röst HL, Sachsenberg T, Aiche S et al (2016) OpenMS: a flexible open-source software plat-
form for mass spectrometry data analysis. Nat Methods 13:741–748. https://doi.org/10.1038/
nmeth.3959
144. Huan T, Forsberg EM, Rinehart D et al (2017) Systems biology guided by XCMS Online
metabolomics. Nat Methods 14:461–462. https://doi.org/10.1038/nmeth.4260
145. Katajamaa M, Miettinen J, Orešič M (2006) MZmine: toolbox for processing and visualiza-
tion of mass spectrometry based molecular profile data. Bioinformatics 22:634–636. https://
doi.org/10.1093/bioinformatics/btk039
146. Ye D, Li X, Shen J, Xia X (2022) Microbial metabolomics: from novel technologies to
diversified applications. TrAC Trends Anal Chem 148:116540. https://doi.org/10.1016/j.
trac.2022.116540
147. Johnson CH, Ivanisevic J, Benton HP, Siuzdak G (2015) Bioinformatics: the next frontier of
metabolomics. Anal Chem 87:147–156. https://doi.org/10.1021/ac5040693
148. Naake T, Gaquerel E, Fernie AR (2020) Annotation of specialized metabolites from high-­
throughput and high-resolution mass spectrometry metabolomics. In: Li S (ed) Computational
methods and data analysis for metabolomics, Methods in molecular biology, vol 2104.
Humana, New York, pp 209–225
149. Rasche F, Svatoš A, Maddula RK et al (2011) Computing fragmentation trees from tandem
mass spectrometry data. Anal Chem 83:1243–1251. https://doi.org/10.1021/ac101825k
150. da Silva RR, Wang M, Nothias L-F et al (2018) Propagating annotations of molecu-
lar networks using in silico fragmentation. PLoS Comput Biol 14:e1006089. https://doi.
org/10.1371/journal.pcbi.1006089
151. Scheubert K, Hufsky F, Böcker S (2013) Computational mass spectrometry for small mol-
ecules. J Cheminform 5:12. https://doi.org/10.1186/1758-­2946-­5-­12
222 L. S. de Medeiros et al.

152. Tsugawa H, Kind T, Nakabayashi R et al (2016) Hydrogen rearrangement rules: computa-

tional MS/MS fragmentation and structure elucidation using MS-FINDER software. Anal
Chem 88:7946–7958. https://doi.org/10.1021/acs.analchem.6b00770
153. Triastuti A, Haddad M, Barakat F et al (2021) Dynamics of chemical diversity during co-­
cultures: an integrative time-scale metabolomics study of fungal Endophytes Cophinforma
mamane and Fusarium solani. Chem Biodivers 18:e2000672. https://doi.org/10.1002/
cbdv.202000672
154. Verdegem D, Lambrechts D, Carmeliet P, Ghesquière B (2016) Improved metabolite identifi-
cation with MIDAS and MAGMa through MS/MS spectral dataset-driven parameter optimi-
zation. Metabolomics 12:98. https://doi.org/10.1007/s11306-­016-­1036-­3
155. Wolf S, Schmidt S, Müller-Hannemann M, Neumann S (2010) In silico fragmentation for
computer assisted identification of metabolite mass spectra. BMC Bioinformatics 11:148.
https://doi.org/10.1186/1471-­2105-­11-­148
156. Longnecker K, Kujawinski EB (2017) Mining mass spectrometry data: using new compu-
tational tools to find novel organic compounds in complex environmental mixtures. Org
Geochem 110:92–99. https://doi.org/10.1016/j.orggeochem.2017.05.008
157. Ruttkies C, Schymanski EL, Wolf S et al (2016) MetFrag relaunched: incorporating
strategies beyond in silico fragmentation. J Cheminform 8:3. https://doi.org/10.1186/
s13321-­016-­0115-­9
158. Fahy E, Sud M, Cotter D, Subramaniam S (2007) LIPID MAPS online tools for lipid research.
Nucleic Acids Res 35:W606–W612. https://doi.org/10.1093/nar/gkm324
159. Ludwig M, Fleischauer M, Dührkop K et al (2020) De novo molecular formula annotation
and structure elucidation using SIRIUS 4. In: Li S (ed) Computational methods and data
analysis for metabolomics, Methods in molecular biology, vol 2104. Humana, New York,
pp 185–207
160. Peres E, Souza M, Sousa T et al (2023) Dereplication of sclerotiorin-like azaphilones pro-
duced by Penicillium meliponae using LC-MS/MS analysis and molecular networking. J
Braz Chem Soc 00:1–15. https://doi.org/10.21577/0103-­5053.20230027
161. Dührkop K, Fleischauer M, Ludwig M et al (2019) SIRIUS 4: a rapid tool for turning tandem
mass spectra into metabolite structure information. Nat Methods 16:299–302. https://doi.
org/10.1038/s41592-­019-­0344-­8
162. Cohen LJ, Esterhazy D, Kim S-H et al (2017) Commensal bacteria make GPCR ligands that
mimic human signalling molecules. Nature 549:48–53. https://doi.org/10.1038/nature23874
163. Hoffmann MA, Nothias L-F, Ludwig M et al (2022) High-confidence structural annota-
tion of metabolites absent from spectral libraries. Nat Biotechnol 40:411–421. https://doi.
org/10.1038/s41587-­021-­01045-­9
164. Dührkop K, Nothias L-F, Fleischauer M et al (2021) Systematic classification of unknown
metabolites using high-resolution fragmentation mass spectra. Nat Biotechnol 39:462–471.
https://doi.org/10.1038/s41587-­020-­0740-­8
165. Klamrak A, Nabnueangsap J, Puthongking P, Nualkaew N (2021) Synthesis of ferulenol by
engineered Escherichia coli: structural elucidation by using the in silico tools. Molecules
26:6264. https://doi.org/10.3390/molecules26206264
166. Atencio LA, Boya P CA, Martin HC et al (2020) Genome mining, microbial interac-
tions, and molecular networking reveals new dibromoalterochromides from strains of
Pseudoalteromonas of Coiba National Park-Panama. Mar Drugs 18:456. https://doi.
org/10.3390/md18090456
167. Sequeira P, Rothkegel M, Domingos P et al (2022) Untargeted metabolomics sheds light on
the secondary metabolism of fungi triggered by choline-based ionic liquids. Front Microbiol
13:946286. https://doi.org/10.3389/fmicb.2022.946286
168. van der Hooft JJJ, Wandy J, Barrett MP et al (2016) Topic modeling for untargeted sub-
structure exploration in metabolomics. Proc Natl Acad Sci 113:13738–13743. https://doi.
org/10.1073/pnas.1608041113
8 Discovering New Natural Products Using Metabolomics-Based Approaches 223

169. Maimone NM, de Oliveira LFP, Santos SN, de Lira SP (2021) Elicitation of Streptomyces lun-
alinharesii secondary metabolism through co-cultivation with Rhizoctonia solani. Microbiol
Res 251:126836. https://doi.org/10.1016/j.micres.2021.126836
170. Pham HT, Lee KH, Jeong E et al (2021) Species prioritization based on spectral dissimilarity:
a case study of polyporoid fungal species. J Nat Prod 84:298–309. https://doi.org/10.1021/
acs.jnatprod.0c00977
171. Rogers S, Ong CW, Wandy J et al (2019) Deciphering complex metabolite mixtures by unsu-
pervised and supervised substructure discovery and semi-automated annotation from MS/MS
spectra. Faraday Discuss 218:284–302. https://doi.org/10.1039/C8FD00235E
172. Liu Y, Ding L, Shi Y et al (2022) Molecular networking-driven discovery of antibacterial
perinadines, new tetracyclic alkaloids from the marine sponge-derived Fungus Aspergillus
sp. ACS Omega 7:9909–9916. https://doi.org/10.1021/acsomega.2c00402
173. Andersen AJC, Hansen PJ, Jørgensen K, Nielsen KF (2016) Dynamic cluster analysis: an
unbiased method for identifying A + 2 element containing compounds in liquid chromato-
graphic high-resolution time-of-flight mass spectrometric data. Anal Chem 88:12461–12469.
https://doi.org/10.1021/acs.analchem.6b03902
174. Roullier C, Guitton Y, Valery M et al (2016) Automated detection of natural halogenated
compounds from LC-MS profiles–application to the isolation of bioactive chlorinated com-
pounds from marine-derived fungi. Anal Chem 88:9143–9150. https://doi.org/10.1021/acs.
analchem.6b02128
175. Flores-Bocanegra L, Al Subeh ZY, Egan JM et al (2022) Dereplication of fungal metabolites
by NMR-based compound networking using MADByTE. J Nat Prod 85:614–624. https://doi.
org/10.1021/acs.jnatprod.1c00841
176. Zhang C, Idelbayev Y, Roberts N et al (2017) Small Molecule Accurate Recognition
Technology (SMART) to enhance natural products research. Sci Rep 7:14243. https://doi.
org/10.1038/s41598-­017-­13923-­x
177. Hu J-Q, Wang J-J, Li Y-L et al (2021) Combining NMR-based metabolic profiling and
genome mining for the accelerated discovery of Archangiumide, an Allenic Macrolide from
the Myxobacterium Archangium violaceum SDU8. Org Lett 23:2114–2119. https://doi.
org/10.1021/acs.orglett.1c00265
178. Emwas A-H et al (2019) NMR spectroscopy for metabolomics research. Metabolites
9(7):123. https://doi.org/10.3390/metabo9070123
179. Hao J et al (2012) BATMAN – an R package for the automated quantification of metabo-
lites from nuclear magnetic resonance spectra using a Bayesian model. Bioinformatics
15:2088–20901. https://doi.org/10.1093/bioinformatics/bts308
180. Ravanbakhsh S et al (2015) Accurate, fully-automated NMR spectral profiling for metabolo-
mics. PLoS ONE 5:e0124219. https://doi.org/10.1371/journal.pone.0124219
181. Röhnisch HE et al (2018) AQuA: an automated quantification algorithm for high-throughput
NMR-based metabolomics and its application in human plasma. Anal Chem 3:2095–2102.
https://doi.org/10.1021/acs.analchem.7b04324
182. Tardivel PJC et al (2017) ASICS: an automatic method for identification and quantification of
metabolites in complex 1D 1H NMR spectra. Metabolomics 10:109. https://doi.org/10.1007/
s11306-­017-­1244-­5
183. Bingol K et al (2016) Comprehensive metabolite identification strategy using multiple two-­
dimensional NMR spectra of a complex mixture implemented in the COLMARm web server.
Anal Chem 24:12411–12418. https://doi.org/10.1021/acs.analchem.6b03724
184. Egan JM et al (2021) Development of an NMR-based platform for the direct structural
annotation of complex natural products mixtures. J Nat Prod 4:1044–1055. 10.1021/acs.
jnatprod.0c01076
185. Xia J et al (2008) MetaboMiner – semi-automated identification of metabolites from 2D
NMR spectra of complex biofluids. BMC Bioinformatics 1:507008. https://doi.org/10.118
6/1471-­2105-­9-­507
186. Bruguière A et al (2020) MixONat, a software for the dereplication of mixtures based
on 13 C NMR spectroscopy. Anal Chem 13:8793–8801. https://doi.org/10.1021/acs.
analchem.0c00193
224 L. S. de Medeiros et al.

187. Zhang C et al (2017) Small Molecule Accurate Recognition Technology (SMART) to enhance
natural products research. Sci Rep 1:14243. https://doi.org/10.1038/s41598-­017-­13923-­x
188. Bujak R, Struck-Lewicka W, Markuszewski MJ, Kaliszan R (2015) Metabolomics for
laboratory diagnostics. J Pharm Biomed Anal 113:108–120. https://doi.org/10.1016/j.
jpba.2014.12.017
189. Chong J, Xia J (2020) Using MetaboAnalyst 4.0 for metabolomics data analysis, interpreta-
tion, and integration with other omics data. In: Li S (ed) Computational methods and data
analysis for metabolomics, Methods in molecular biology, vol vol 2104. Humana, New York,
pp 337–360
190. Ghosh T, Zhang W, Ghosh D, Kechris K (2020) Predictive modeling for metabolomics data.
In: Li S (ed) Computational methods and data analysis for metabolomics, Methods in molec-
ular biology, vol 2104. Humana, New York, pp 313–336
191. Sinha R, Sharma B, Dangi AK, Shukla P (2019) Recent metabolomics and gene editing
approaches for synthesis of microbial secondary metabolites for drug discovery and devel-
opment. World J Microbiol Biotechnol 35:166. https://doi.org/10.1007/s11274-­019-­2746-­2
192. Debik J, Sangermani M, Wang F et al (2022) Multivariate analysis of NMR-based metabolo-
mic data. NMR Biomed 35:e4638. https://doi.org/10.1002/nbm.4638
193. Abdelmohsen U, Cheng C, Viegelmann C et al (2014) Dereplication strategies for targeted
isolation of new antitrypanosomal actinosporins A and B from a marine sponge associated-­
Actinokineospora sp. EG49. Mar Drugs 12:1220–1244. https://doi.org/10.3390/md12031220
194. Huang S-M, Toh W, Benke PI et al (2014) MetaboNexus: an interactive platform for inte-
grated metabolomics analysis. Metabolomics 10:1084–1093. https://doi.org/10.1007/
s11306-­014-­0648-­8
195. Pang Z, Chong J, Zhou G et al (2021) MetaboAnalyst 5.0: narrowing the gap between raw
spectra and functional insights. Nucleic Acids Res 49:W388–W396. https://doi.org/10.1093/
nar/gkab382
196. Takaki M, Williams DE, Freire VF et al (2022) Metabolomics reveals a 26-membered macro-
lactone produced by Endophytic Colletotrichum spp. from Alcatrazes Island, Brazil. Org Lett
24:9381–9385. https://doi.org/10.1021/acs.orglett.2c03531
197. Deutsch JM, Mandelare-Ruiz P, Yang Y et al (2022) Metabolomics approaches to dereplicate
natural products from coral-derived bioactive bacteria. J Nat Prod 85:462–478. https://doi.
org/10.1021/acs.jnatprod.1c01110
198. Kešnerová L, Mars RAT, Ellegaard KM et al (2017) Disentangling metabolic functions
of bacteria in the honey bee gut. PLoS Biol 15:e2003467. https://doi.org/10.1371/journal.
pbio.2003467
199. Hou Y, Braun DR, Michel CR et al (2012) Microbial strain prioritization using metabo-
lomics tools for the discovery of natural products. Anal Chem 84:4277–4283. https://doi.
org/10.1021/ac202623g
200. Wu C, Zacchetti B, Ram AFJ et al (2015) Expanding the chemical space for natural products
by Aspergillus-Streptomyces co-cultivation and biotransformation. Sci Rep 5:10868. https://
doi.org/10.1038/srep10868
Chapter 9
Microbial Metabolites Annotation by Mass
Spectrometry-Based Metabolomics

Paulo Wender P. Gomes, Talita Carla de Tralia Medeiros,

Naydja Moralles Maimone, Tiago F. Leão,
Luiz Alberto Beraldo de Moraes, and Anelize Bauermeister

9.1 Introduction

Specialized metabolites (previously called secondary metabolites) from natural

sources, such as plants and microorganisms, are of great interest to be investigated
because of their pharmacological and biotechnological properties. These organisms
generally present complex biosynthetic pathways, which usually lead to the produc-
tion of a wide variety of chemical structures [1, 2]. Microorganisms, in particular

Authors Paulo Wender P. Gomes, Talita Carla de Tralia Medeiros and Naydja Moralles Maimone
have equally contributed to this chapter.

P. W. P. Gomes
Collaborative Mass Spectrometry Innovation Center, Skaggs School of Pharmacy and
Pharmaceutical Sciences, University of California San Diego, La Jolla, CA, USA
T. C. de Tralia Medeiros · L. A. B. de Moraes
Departamento de Química, Faculdade de Filosofia, Ciências e Letras de Ribeirão Preto,
Universidade de São Paulo, Ribeirão Preto, São Paulo, Brazil
N. M. Maimone
Departamento de Ciências Exatas, Escola Superior de Agricultura ‘Luiz de Queiroz’,
Universidade de São Paulo, Piracicaba, São Paulo, Brazil
T. F. Leão
Centro de Energia Nuclear na Agricultura, Universidade de São Paulo,
Piracicaba, São Paulo, Brazil
A. Bauermeister (*)
Collaborative Mass Spectrometry Innovation Center, Skaggs School of Pharmacy and
Pharmaceutical Sciences, University of California San Diego, La Jolla, CA, USA
Instituto de Ciências Biomédicas, Universidade de São Paulo, São Paulo, Brazil
e-mail: [email protected]

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 225
T. Pacheco Fill (ed.), Microbial Natural Products Chemistry, Advances in Experimental
Medicine and Biology 1439, https://doi.org/10.1007/978-3-031-41741-2_9
226 P. W. P. Gomes et al.

those with large genomes, are known to encode several Biosynthetic Gene Clusters
(BGCs) on their DNA, reflecting the potential to produce specialized metabolites.
Here, we can mention key enzymes for natural products production such as NRPS
(nonribosomal peptide synthetase), PKS-I (polyketide synthase type I), PKS-II,
NRPS-PKS hybrids, and aminoglycoside, all are known to generate drug-like mol-
ecules [1, 3].
The first reports of using natural products to improve human health are from
2900 to 2600 BC, in ancient Mesopotamia [4]. In the early 1900s, approximately
80% of all medicines were obtained from plants, and only with the discovery of
penicillin by Alexander Fleming in 1928, microorganisms started to be considered
as a potential source of compounds with pharmaceutical and biotechnological prop-
erties. Since then, compounds derived from microorganisms have been largely
applied in medicine, agriculture, food industry, cosmetics, among others.
Aproximately 60% of the FDA-approved small molecule drugs are related to natu-
ral products and, more particularly, 69% of all antimicrobial agents are originated
from natural sources [5]. Indeed, several bioactive metabolites, including com-
pounds already used for pharmaceutical or biotechnological purposes, have been
isolated from microbial cultures [6, 7].
Traditional compound discovery methods, which are based on the isolation of
active compounds usually guided by bioassays [8], have been used for at least 8
decades and still dominate the scenario of research on natural products, with the
potential to lead us to the discovery of new structures [9–11]. However, despite their
undeniable relevance, these methodologies demand huge amounts of solvents, and
are time-consuming. Not to mention that the classical methodologies present a high
rediscovery rate (isolation of already known compounds), which is not desired in
most investigations of natural products [12]. Furthermore, the biological activity
observed in a crude extract may be due to synergism of metabolites, and therefore,
fractionation and isolation processes do not favor the identification of the active
compounds [13]. Besides, fractionation of extracts may cause loss or degradation of
the bioactive metabolites [14]. In the beginning of the 2000s, there was an increas-
ing development of new methods to optimize culture conditions to instigate the
production of new structures (e.g. with the development of high-throughput micro-
culture approaches), nonetheless, the identification of the metabolites still remains
a problem to be overcome [15].
With few exceptions, most microbial samples comprehend complex mixtures of
metabolites, and a considerable part of these are produced in low amounts, making
their isolation even more difficult [16]. Mass spectrometry (MS) stands out as an
analytical tool for the detection and identification of metabolites due to its high
sensitivity and selectivity, particularly when coupled with liquid chromatography
(LC-MS). Its ability to separate, detect and identify many metabolites within intri-
cate samples largely decrease the necessity of isolation steps [17, 18]. In the context
of complex samples, the mass spectrometer detects multiple ions, resulting in a
substantial amount of data which is challenging to process manually. Over the past
decade, several automated approaches have been developed to enhance data analy-
sis in response to this issue.
9 Microbial Metabolites Annotation by Mass Spectrometry-Based Metabolomics 227

The development of MS open-access databases, such as MoNA [19], HMDB

[20, 21], MassBank [19], ReSpect [22], mzCloud [23], GNPS (Global Natural
Products Social Molecular Networking) spectral library [24], have also been very
helpful for metabolites annotation using library comparison. However, MS data-
bases are still very limited in number of known structures when compared to struc-
tural databases, such as PubChem (110 M), ChemSpider (114 M) or Dictionary of
Natural Products (328 K). Unfortunately, most structures deposited in structural
databases do not present spectral data, especially MS spectra, and therefore, other
approaches are necessary. To fill this gap, in silico tools have been developed to con-
nect somehow known structures to experimental MS spectra. The use of these tools
has increased the number of annotations from 4–5% to even more than 50% in a
metabolomics dataset [25], which is of great importance in investigations of micro-
bial samples.
In this chapter, we describe metabolomics approaches to assist in metabolites
annotation process, including the most commonly employed method (spectral
library search) as well as recently developed approaches for annotation of substruc-
tures and chemical classes. The approaches presented here are applied in the discov-
ery of the specialized metabolites from microorganisms, as well as to improve the
annotation by employing in silico tools and structure databases. We also discuss
how to check and validate an annotation, as well as how to classify them by the lev-
els defined by the Metabolomics Society [26, 27]. In addition, we present some
examples and discussion of investigations of microbial metabolites in which anno-
tation tools based on metabolomics have made great contributions.

9.2 Mass Spectrometry to Detect Microbial Metabolites

Mass Spectrometry is an analytical technique that detects metabolites as ions (mass

(m) over charge (z), m/z) at gas phase [17]. A mass spectrometer is composed of an
ion source to generate ions at gas phase, a mass analyzer to analyze the ionized
metabolites according to their m/z (select, fragment, etc) and a detector. The ions are
registered in a mass spectrum that shows the m/z value (x axis) and the intensity (y
axis). Nowadays, there are several types and models of ionization sources and mass
analyzers available. This variety of equipaments, including different methods of
ionization and fragmentation, allows the analysis of metabolites with different
physical-chemical properties. That way, different equipments can generate different
mass spectra for the same metabolite, which hinders the analysis and the compari-
son with literature.
For a long time, mass spectrometry was used coupled to gas chromatography,
once electron impact (EI) was the main ionization source, however the technique is
limited to volatile and thermally stable samples. In EI, a beam of accelerated elec-
trons bombards the volatile molecules, removing an electron and creating typically
high energy radical molecular ions with positive charge. These ions simultaneously
fragment to dissipate the energy generating fragment ions [28]. However,
228 P. W. P. Gomes et al.

considering that a great part of metabolites in nature are not volatile, not even ther-
mally stable, which includes several microbial metabolites, John Bennett Fenn
introduced electrospray ionization source (ESI) [29] to allow the analysis of a
broader range of metabolites. ESI can be coupled to liquid chromatography tech-
nique (LC-MS), and it has been largely applied in the metabolomics field to analyze
small metabolites of biological systems including microorganisms, plants, humans,
and several other kinds of samples [30].
In electrospray, the analysis can be performed in positive ionization mode, where
the ions are usually generated with proton [M + H]+, sodium [M + Na]+, potassium
[M + K]+, and others; or negative ionization mode, where the metabolites lose a
proton [M − H]− or clusters with chloride [M + Cl]− for instance. These ions can be
fragmented in the mass analyzer by collision-induced dissociation (CID) experi-
ment with an inert gas (usually argon or helium), when the ions gain energy due to
the collision and this energy is dissociated by breaking chemical bonds. The frag-
ment ions are detected and a fragmentation spectrum (also known as MS2 or MS/
MS) is generated. The MS2 spectrum contains important structural information that
a mass spectrometrist specialist can use for metabolite identification and to contrib-
ute to chemical structure characterization [31, 32]. Furthermore, some equipment,
such as traps, are able to develop MSn experiments, a method with multiple steps of
fragmentation (MS1 > MS2 > MS3 > MS4 > MSn) [33].
In an experiment called untargeted metabolomics, all the ions in each sample can
be fragmented generating thousands of MS2 for a single sample. Therefore, in untar-
geted metabolomics using large datasets, automated tools are desired and necessary
to assist to unveil the chemical complexity the data encompasses [17, 34]. However,
the annotation of the metabolites in untargeted metabolomics should be done care-
fully, since many of the ions can be contaminants from sample extraction and prepa-
ration process. The presence of more than one ion for a single metabolite should
also be considered, once a metabolite can form complexes with different metals
([M + H]+, [M + Na]+, [M + K]+, etc). As a result, one single metabolite can ionize
with different adducts, which several times also present different fragmentation pat-
terns. For some cases, there are specific tools to assist in the data analysis and avoid
misinterpretation of the results as well as contaminant libraries for mass spectrom-
etry analysis that will be discussed later in this chapter.

9.3 Spectral Library Search for Microbial

Metabolites Annotation

Microbial extracts encompass a wide array of metabolites. Usually mass spectrom-

etry data, such as MS1 and/or MS2, of these extracts can be used to search for some
correspondence in spectral libraries (process called annotation). Annotation is the
assignment of a putative identity (molecular formula, name, structure, etc) to each
feature (or spectrum) in a dataset [35]. When there is a spectral match with the
9 Microbial Metabolites Annotation by Mass Spectrometry-Based Metabolomics 229

Fig. 9.1 General overview employed for the annotation of microbial metabolites. Extracts pro-
duced by microorganisms are usually analyzed by liquid chromatography coupled to mass spec-
trometry (LC-MS/MS). In untargeted LC-MS/MS, MS1 data are generated for ions detected in the
ionization source, followed by the fragmentation of the most intense ions for each scan automati-
cally generating several of MS2 data. The MS2 data, which can be considered as a fingerprint of a
metabolite, can be search against a spectrum library (e.g., GNPS spectral libraries) to find a spec-
tral match. The results are visualized by a MS2 spectrum mirror plot comparing the experimental
and the reference spectrum from the library. A good spectral match indicates the possible presence
of the annotated metabolite in the sample analyzed

library, the results are usually presented as a mirror plot, in which the experimental
spectrum is presented in the top, and the reference spectrum from the library in the
bottom (Fig. 9.1).
One of the main bottlenecks in the metabolite annotation process is the number
of structures with MS2 spectra available. In fact, this number represents a small frac-
tion of the real number of metabolites produced by microorganisms. Among the
largest sources to search for metabolites we can mention PubChem, which contains
more than 110 millions of unique chemical structures, COCONUT, with 406 K
unique structures from natural sources [36], and LOTUS, with 276 K metabolites
from natural products [37]. There are other sources, smaller but not less important,
that should be also mentioned here once they are widely known in the field, such as
KEGG [38], ChEBI [39], and ChemSpider [40]. When these structural databases
are compared with spectral databases there is an evident discrepancy of numbers.
For example, MoNA contains 10 K molecular standards with MS2 data, and others
may have even smaller numbers such as mzCloud and ReSpect (9 K), GNPS (6 K),
and HMDB (2 K) [41, 42]. Therefore, the limited number of available MS2 spectra
causes a low percentage of metabolite annotations, and this rate often represents
230 P. W. P. Gomes et al.

less than 10% of the total number of compounds in complex mixtures [43]. In addi-
tion, some spectral databases such as METLIN and NIST are not freely available and
therefore not readily accessible for most of the scientific community. It is important
to highlight that the lack of spectral annotations in the dataset does not necessarily
mean unknown compounds, these spectra may belong to compounds that does not
have mass spectra available in the spectral library employed. There are many metabo-
lites available only in structural libraries and/or published in manuscripts.
Currently, the GNPS offers one of the main infrastructures for MS data deposit,
sharing, and data analysis. The library encompasses gold-level spectra (those origi-
nated from purified compounds and submitted by authorized users), silver-level
spectra (when the spectra were submitted in association with a publication) or
bronze-level spectra (which means that the spectra were submitted by common
users of the platform) [44], and this library includes spectra from compound specific
collections (as phytochemicals, pesticides, and common contaminants from sol-
vents, culture media and chromatographic equipment). In addition to library search,
this platform also allows users to perform molecular networking (MN) [24] analysis
based on their MS2 data. MN are visual representations of compound relatedness
constructed from the alignment of all MS2 spectra within a dataset, and for this rea-
son it constitutes an approach that enables the propagation of metabolite annota-
tions. When spectral similarity between two features reaches or exceeds the minimal
similarity level settled by the user, edges connecting nodes (which represents MS2
spectra) are defined. Those edges are determined by a modified cosine scoring
scheme that sets the similarity of two spectra with scores ranging from 0 to 1 (com-
pletely different and totally identical, respectively). Then, a group of features related
to each other will constitute a molecular family, formed by chemically similar com-
pounds. For a detailed review about this methodology and platform, please check
[44]. Recently, some algorithms have been developed to improve the accuracy of
MNs. Spec2Vec [45] uses an algorithm based on natural language processing
(Word2Vec) to produce a similarity score that constitutes a better proxy for struc-
tural similarity than cosine correlation. Still regarding the traditional MN, it is suit-
able to this chapter to mention that this approach was created for microbial
metabolomics studies when metabolomic profiling of live microbial colonies was
accessed in situ [46]. Since then, MN has been applied to assist in the discovery of
new microbial compounds [47, 48], to support strain prioritization, culture condi-
tions and extraction methods for drug discovery purposes [49], microbial interac-
tion investigations [50], chemical diversity profiling [51], among others.
MN can also be misinterpreted once a single metabolite can generate different
adducts (different ions) in mass spectrometry data. Therefore, these ions can appear
as different nodes in the molecular network due to different m/z and MS2 profile.
Sometimes, these adducts are even split in different molecular families or remain
unconnected. To overcome this issue Ion Identity Molecular Networking (IIMN)
[52] was launched to recognize these adducts and improve the connectivity of them
in the MN. This tool combines chromatographic data, such as retention time of the
precursor ions to create the molecular networking topology, and therefore, it is able
to connect different ion species of the same metabolite, which reduces feature
9 Microbial Metabolites Annotation by Mass Spectrometry-Based Metabolomics 231

redundancy (a common problem in MS metabolomics studies). IIMN also has a

specific spectral library within GNPS to annotate the possible adducts detected. In
the microbial metabolomics context, IIMN is a promising tool to be applied in the
study of ionophores for example [53], since it is currently a challenge to annotate
these compounds [54].
Regarding those who work with gas chromatography, recently a workflow to
GC-MS data processing was launched, and it can be a new alternative to improve
microbial metabolite annotation [55]. MetaboliteDetector [56] was developed for
GC/MS-based metabolomics data, and it uses an in-house mass spectral library,
retention time and mass spectrum to process automatic metabolite annotations. For
example, that tool was applied to understand how microbial ecosystems respond to
disturbances in mixed cultures in a microbial oilseed population in a biological
wastewater treatment, and four distinct fundamental niche types were found in this
ecosystem [57]. In addition to approaches and tools to annotation of metabolites
using MS data, in recent years, based on many efforts of researchers of the metabo-
lomic field, the concept of annotation used in mass spectrometry was also applied to
1D/2D NMR data. For instance, Complex Mixture Analysis by NMR (COLMAR)
is one of the first NMR database, and in summary this database search for reference
NMR spectra to annotate candidates, including metabolites from complex mixtures,
such as reported to Drosophila melanogaster and Escherichia coli [58]. Also, an
extract of a marine cyanobacterium was analyzed using another tool named Small
Molecule Accurate Recognition Technology” (SMART 2.0) which also uses 1D/2D
NMR data, and it can be associated with LC-MS/MS data by neural network analy-
sis [59].

9.4 Using In Silico Tools to Improve Annotations

Although largely employed for annotation, spectral libraries are far from complete
when compared to structural libraries. These libraries contain a large amount of
structural information and, especially in the last decade, researchers have explored
computational approaches to develop methods to improve annotations [60].
Generally, in silico tools calculate a score between the experimental and the pre-
dicted spectra and create a list of candidate matches for the studied features. Besides
the need of manual inspection of the results in order to avoid false-positive annota-
tions, it is also important to apply database type and taxonomy restrictions and
combine multiple tools to analyze a given dataset [61]. For example, for a microbial-­
derived dataset, it is interesting to consider databases which are exclusive for micro-
bial metabolites (e.g. Natural Products Atlas [62]), and some tools (e.g. NAP that
will be discussed later) that allow the user to employ in-house databases that can be
restrictive for the studied taxa. Although in silico fragmentation algorithms still
need to be improved, it is expected that the average accuracy of spectral predictions
will enhance with the increasing of the public available data. In this section, we
232 P. W. P. Gomes et al.

Fig. 9.2 Methods used for metabolite annotation (a) Combinatorial fragmentation that purpose to
explain the peaks in the experimental MS2 spectrum considering the breakdown of the bonds
between parts of the molecule (e.g., MetFrag, and CFM-ID); (b) Fragmentation tree use machine
learning to predict a fingerprint about classes, subclasses and structure (e.g., Sirius); (c) Propagation
of annotation using spectral library matching (NAP); and d) MS2LDA which can provide a finger-
print concerning specific substructures in one or a group of metabolites

discuss some of the currently available in silico tools and bring examples on their
use applied to microbial metabolite annotations.
Combinatorial fragmentation (Fig. 9.2a), employed in MetFrag [63, 64] and
CFM-ID 4.0 [65], is a well-known method that uses a heuristic strategy to break
chemical bonds of known structures from structural libraries in order to predict the
fragment ions in a given fragment spectrum. MetFrag filters candidates from large
databases through Molecular Formula (e.g. PubChem) and then uses theoretical dis-
sociation energy to calculate substructures for fragments, making a top-rank for
those structures that best explain most of the fragments in an experimental spec-
trum. This approach has two drawbacks that should be considered with caution.
First, the Normalized Dissociation Energy (NCE) used by MetFrag is theoretical,
and in a laboratory experiment there are significant differences involving factors
such as vacuum, collision gas pressure, and other electronic settings. Second, the
suggested substructures for fragments do not show relationships with a precursor
fragment. In this sense, we recommend the use of chemical knowledge to evaluating
the possibility of candidates and fragments proposed by MetFrag, as well as neutral
9 Microbial Metabolites Annotation by Mass Spectrometry-Based Metabolomics 233

losses or cleavage homolytic/heterolytic, to avoid false-positive results based on

combinatorial analysis without rationalizing the chemical mechanism. For more
details about the application, here are some examples [64, 66–70].
Unlike MetFrag, the CFM-ID 4.0 [65] uses unsupervised machine learning to
predict MS2 spectra [71–74]. Overall, the in silico-predictor CFM provides a method
to assist the annotation of metabolites using MS2 spectra. This tool was trained
using competitive fragmentation modeling to produce probabilistic results closer to
what is expected for the fragmentation process of a known and/or unknown metabo-
lite. The results need to be severely delineated using chemotaxonomic information
about metabolites already reported for the family, genus, and/or species to have an
annotation. In addition, this method is recommended to small metabolites because
they tend to have more limited fragmentation possibilities [75]. There are some
examples in the literature [76–79] using microbial diversity and fragment MS2 con-
sistent with spectra predicted by CFM.
Fragmentation tree is another well-known in silico method widely used in anno-
tation tools (Fig. 9.2b). The method uses machine learning (support vector machine)
and a spectral library of known structures to learn the patterns of the fragmentation
related to structures and then, it creates a fingerprint that better explains each spec-
trum [80]. The trees are graphically characterized by nodes (representing precursor
and fragment ions observed in a spectrum, designated by their putative molecular
formulas), which are interconnected when there is a suggested fragmentation pat-
tern between nodes (such as neutral losses) [81]. Spectrum peaks without corre-
sponding nodes in the computed tree are considered noise, and the molecular
formula of the tree’s root node corresponds to the putative molecular formula for the
precursor ion of the compound [80]. In a prediction phase, an experimental MS2
spectrum is given to the method, and it generates a fingerprint based on the pattern
learned from the library. In the scoring and last phase, the fingerprint from the
experimental spectrum is compared to the other fingerprints generated before and
scores the structures that better match the fingerprints. SIRIUS 4, for instance, is an
open-source software [82] that uses fragmentation trees along with isotope pattern
analysis in a framework that supports the initial fingerprint of large metabolomics
data by enabling the determination of accurate molecular formulas for compounds
from a HRMS dataset. The SIRIUS ecosystem also harbors the Gibbs-sampling-­
based ZODIAC algorithm as an alternative to predict molecular formulas [83],
allowing more accuracy in the annotation of features of higher molecular masses (>
500 Da); the CSI: FingerID, an in silico workflow [84] that combines fragmentation
tree and machine learning approaches for the prediction of molecular properties for
an experimental spectrum, allowing searches for candidate structures to unknown
features from the analyzed datasets in several molecular structure databases; and the
CANOPUS [85] which uses machine learning methods such as support vector
machines, to predict the fingerprints of the analyzed metabolites from their mass
spectra and deep neural networks, to perform automated classification of com-
pounds based on the 2497 compound classes from the Classyfire ontology [86],
covering several biologically relevant chemical classes from up to different levels
such as Kingdom, SuperClass, Class, etc. [85]. CANOPUS assigns chemical
234 P. W. P. Gomes et al.

classifications to the majority of MS2 spectra from a dataset, even in cases where
there are no annotated structures from databases or reference MS2 spectra available
for training the algorithm. More information can be found in recent examples that
investigated the metabolic profile produced by Burkholderia cenocepacia [87] and
Escherichia coli [88]. COSMIC is another tool to improve the annotation of micro-
bial metabolites [89], which allows ranking annotations using scores that range
from 0 to 1. This score does not mean that an annotation is completely correct, for
example, a structure can show a high score, but it does not correspond to the real
structure, or show a low score and be the correct structure. Recently the COSMIC
was incorporated into SIRIUS, and all annotations should pass by double-checking
for final validation. Likewise, a single-cell method named RespectM [90] was
recently reported for metabolomics analysis using mass spectrometry data, and as
prospects, that tool can be applied to identify cellular populations with different
chemical profiles in microorganisms datasets [91].
MS-FINDER [92–94] is another software option for structure elucidation based
on MS2 data (also applicable for GC-EI-MS) and fragments in silico annotation. It
also provides molecular formula predictions (98% of assignment accuracy when
tested with 5063 MassBank MS2 records) based on accurate mass, isotope ratio and
ion fragmentation information. In addition to considering mass accuracies, frag-
ment linkages and bond dissociation energies to rank putative structures,
MS-FINDER considers the Hydrogen Rearrangement (HR) rules, used to predict
patterns of fragment formation in low-energy collision-induced dissociation. This
tool was used on a multi-software approach to analyze and identify the metabolites
induced in the co-culture of Aspergillus sydowii and Bacillus subtilis [95]. It also
assisted the annotation of compounds representing microbiome-related metabolo-
mic signatures in mice, providing clues on how the microbiota may have influence
in the host metabolism and inter-organ communication [96]. Furthermore,
MS-FINDER enabled the characterization of Vitis vinifera upregulated compounds
due to heavy microbial endophyte colonization [97].
MAGMa was created to support LC-MSn data, n being equal or higher than 2
[97–99]. It is based on annotation of substructure candidates by multistage spectral
trees (MSn data) [99], without the need of reference spectral data. MAGMa’s algo-
rithm computes hierarchical trees for in silico generated substructures, considering
for each MS level assignments the substructures assigned in predecessor and suc-
cessor levels as well. Besides that, to determine the most likely substructures for
each fragment peak, penalty scores are designated to candidates based on the type
of bonds that are broken (e.g. single, double, triple, aromatic bonds with heteroatom
and carbon-carbon bonds) and the presence/absence of non-matched peaks. Taking
these annotations into account, a match score ranks the different candidate molecu-
lar structures that can be retrieved from PubChem and HMDB. It is important to
note that related molecules (e.g. positional isomers) may result in the same score. In
these cases, MAGMa’s developers suggest that the candidate with higher biological
relevance be considered in the manual validation of results.
Another approach that can be applied to improve the annotation and discovery of
analogs from known metabolites or new metabolites is the Network Annotation
9 Microbial Metabolites Annotation by Mass Spectrometry-Based Metabolomics 235

Propagation (NAP) [100]. This is a tool available through the GNPS [100], and
overall, NAP can use the spectral similarity between annotations from the GNPS
library and unknown compounds within the same molecular families to propagate
in silico metabolite proposals using top ranked candidates retrieved from several
public databases or in house libraries, but the tool does not necessarily require spec-
tral matches to annotate the structures that compose a molecular family. In cases
where no matches are obtained, it still manages to correctly annotate most of the
substructures using in silico predictions, which can be indicative of chemical
classes, from compounds ranked with highest scores. Due to this fact, using NAP in
association with hierarchical chemical classification algorithms such as Classyfire
[86] and NPClassifier [101] can also favor the annotation of molecular families in
higher levels of chemical taxonomy, such as chemical classes. NAP has been applied
to discover bioactive compounds from cyanobacteria [102]; to characterize plants
[103, 104] and microbial derived extracts [105, 106].
For some specific chemical classes such as peptides, there are also some specific
tools. DEREPLICATOR [107] was launched as a new algorithm to support the
dereplication process of peptide natural products (PNPs). Due to particularities
observed for PNPs, such as the fact that they can be constituted from hundreds of
different amino acids (and not only by essential amino acids) as well as being able
to form complex spatial configurations (which is not commonly observed for pro-
tein generating peptides), among other factors, tools traditionally employed in pro-
teomic analysis cannot be fully effectively applied in PNPs analysis.
DEREPLICATOR relies on calculating false discovery rate scores for putative
matches based on the tool’s internal PNP databases and a decoy database, as well as
the statistical significance for the matches using the algorithm proposed by
Mohimani [108]. The tool also integrates VARQUEST [109], an algorithm that
allows the annotation of modified variants of known PNPs, and takes advantage of
the molecular networks topology to improve the confidence on the suggested candi-
dates. The utility of DEREPLICATOR was demonstrated when the tool was able to
annotate surugamide in five different and independent datasets from the GNPS plat-
form, four from Streptomyces albus J1074 (a widely studied strain) data and one
from Streptomyces sp. CNY228 data, when original studies encompassing those
datasets did not identify this compound previously [110]. This tool was also used to
annotate bromoalterochromide analogues from pigment-producing
Pseudoalteromonas strains [111], an approach that resulted in the isolation of two
new compounds (D/D′ dibromoalterochromides). In addition, a new algorithm
named DEREPLICATOR+ [112], based on DEREPLICATOR approaches, is avail-
able to assist in the annotation of several classes of natural products such as
polyketides, terpenes, flavonoids, and alkaloids. An example of this approach can be
consulted in the characterization process to discover a polyketide antibiotic from
Streptomyces Mg1 [112, 113]. DEREPLICATOR+ also allowed the annotation of
derivatives from the macrolide novonestimycin, defined as the cytotoxic compounds
in the extracts produced by Streptomyces BRB-298 and BRB-302 from Rocas Atoll
in Brazil [114].
236 P. W. P. Gomes et al.

Regarding those who work with NMR, NMRfilter is a potential tool to combine
in silico to experimental data [115]. Furthermore, NMRfilter can be used to validate
compound annotation after dereplication in GNPS. 1D/2D NMR data can be
uploaded into the tool and based on machine learning it is possible to predict a
potential structure. However, this approach is currently under development and the
results should be evaluated with caution.

9.5 Annotation of Substructures

Sometimes it is not possible to putatively identify the structure or even the chemical
class of a metabolite. However, the identification/annotation of its chemical sub-
structures can be very helpful to guide the next steps in an investigation. For
instance, the isolation procedure can be settled in favor of a specific chemical group.
For these cases, there are computational tools to allow the annotation of substruc-
tures in metabolomics datasets. These tools can also be used for the classification of
the moiety or functional group from specific chemical classes and subclasses.
MS2LDA [116–118] is an unsupervised text mining algorithm to search substruc-
tures in a dataset using fragmentation spectral patterns (MS2) and characteristic
neutral losses, also known as Motifs (Fig. 9.2d). A single experimental mass spec-
trum can harbor several Motifs that can be annotated through the MotifDB database,
which is integrated into the tool. Using this approach, the diversity of the deferox-
amine siderophores was investigated for an isolate of Streptomyces sp. [119], and
several new potential derivatives were detected. The authors discussed that the total
diversity of these compounds in the analyzed extracts could be 2.5 times greater
than indicated by spectral annotations using the GNPS library only. With the anno-
tation of substructures added to the analysis of molecular networks, MS2LDA was
used to annotate thiopeptides produced by poorly studied lineages of actinobacteria
of the genus Planomonospora [120]. Analyzing the molecular networks, genetic
data of these strains and information available in the literature to annotate
Mass2Motifs, these authors identified several possible analogues of known
compounds.
Likewise, MESSAR [121] was inspired by association rule mining [122], in
other words, compounds that contain the same substructure often show some spec-
tral characteristics in common. These tools correspond to the same manual method
used in the recent past to characterize substructures in metabolites from natural
products, evaluating spectrum by spectrum to create building blocks of metabolites.
Furthermore, the GNPS platform recently launched an innovative approach named
Mass Spec Query Language (MassQL), and that tool searches for specific MS2
information in big metabolomic datasets acquired by LC-MS/MS. For example, two
fragments or a neutral loss can describe a specific glycoside from known anthracy-
cline, and new compounds can be found using that information and predefined code
with m/z, mass defect, and peak intensity.
9 Microbial Metabolites Annotation by Mass Spectrometry-Based Metabolomics 237

Although these tools represent advances in the substructure annotation of spe-

cialized metabolites, their application is generally limited to specific groups of com-
pounds. In this sense, a universal approach based on chemical taxonomy and
chemical ontology can help researchers to verify and interpret large diversity of
classes from annotations obtained from spectroscopic data. For example, ClassyFire
[123] is a tool that uses both approaches and allows easy automated classification
for a given set of compounds relying on structural properties (as chemical classes
correspond to certain structural patterns), using as input data information such as
SMILES or InChI from compound annotations [86]. ClassyFire was used to classify
natural products from a Cyanobacteria in previous reports [102], as well as lipids
and lipid-like molecules superclass from microbial datasets [124]. Due to the suc-
cess in many fields and applications, other advanced bioinformatics tools have
incorporated ClassyFire to improve metabolite annotation process, e.g. Qemistree
[125], a new tree-based approach (based on hierarchical clustering) to compute
(using SIRIUS and GNPS library hits) and represent chemical characteristics from
samples in metabolomics studies as phylogenetic trees (which resembles DNA
sequence-based trees).
Considering all the approaches described before, integration of these tools
becomes very important to improve chemical insight from a dataset. Currently,
MolNetEnhencer [126] combines the outputs generated by the other workflows
available on the GNPS platform (such as NAP, DEREPLICATOR and
DEREPLICATOR+, MS2LDA and VARQUEST, but there is also support for exter-
nal tools such as CSI:FingerID) in a single molecular network, and in addition, it
performs automated hierarchical chemical classification through the Classyfire
ontology for the representative structures on molecular families, which allows and
facilitates the visualization of compound classes inferred for molecular families
within a molecular network. For instance, MolNetEnhancer was applied to discover
the chemical diversity of antimicrobial metabolites from fungi of the Gulf of Mexico
[127], as well to reveal compounds from Aspergillus sp. [128] and microbial com-
munity in mangrove sediments [129], and to integrate the annotations obtained from
the GNPS spectral library and the NAP workflow to evaluate the metabolic diversity
observed in extracts of actinobacteria isolated from different polar arctic environ-
ments [130].

9.6 How to Validate a Metabolite Annotation

The annotation process using mass spectrometry is characterized by assigning a

possible chemical identity such as a metabolite name, chemical structure (without
stereochemistry), or molecular formula to a molecule or a group of specific com-
pounds detected in a sample. In MS-based metabolomics, using MS tandem data,
generally considers the precursor ion data (MS1) and fragmentation data (MS2) to
annotate a structure [89, 131]. Some key points need to be considered here, the m/z
ratio of the precursor ion can be used to calculate and suggest molecular formulae,
238 P. W. P. Gomes et al.

and in this stage, it must evaluate the precursor ion region to understand the isotopic
pattern (e.g. if Cl, Br or other atom with characteristic isotopic pattern is present).
In addition to the points described previously, it needs to be severely evaluated:
(1) the mass shift or error in ppm for the precursor ion (for low resolution data, it is
accepted up to 2 Da of difference, while for high-resolution data it is recommended
0.02 Da); (2) there is not a minimum or specific number of fragment ions to con-
sider for a good spectral match, all fragment ions should be checked; (3) in some
cases, different structures may results to the same MS2 spectrum, for instance, three
isomers of glutathione conjugates of estrone and estradiol [132], therefore, the
researcher should have in mind that the metabolite annotated could have an isomer
with different structure, especially if there are stereocenters, once MS is blind for
the spatial distribution of the atoms; (4) the acquisition of fragmentation spectra in
different equipment can results in different spectra for the same structure; and (5)
the computational tools for annotation are automated, therefore, they search for
spectral similarity and there is always a “range” of false-positive results.
The Metabolomics Society Initiative (MSI) [26] created four levels to classify an
annotation. The level 1 refers for unambiguous identification of metabolites by iso-
lation and complete structure characterization (including NMR, MS, MS2, and RT
data), or by comparison with standards. Level 2 for probable structure when at least
2 data are compatible between experimental data and library spectra data, e.g. spec-
tral match with spectral libraries and precursor ion. Care is needed and it is impor-
tant to check when there are different acquisition settings and the evaluation of the
mirror plot should consider e.g., resolution, normalized collision energy, and ioniza-
tion. Level 3 is applied as tentative of candidate annotation or chemical class level,
in other words, when there are pieces of evidence of those compounds (e.g., molec-
ular formula, MS1, and MS2). Level 4 is used for unknown metabolites that still can
be detected and quantified. Schymanski and coworkers [27], suggested to split level
4 into levels 4 (molecular formula assignment) and 5 (no molecular formula assign-
ment). In other words, level 5 uses exact mass to measure the mass defect, generally
below 10 ppm (e.g. for a mass weight of 500 Da) [27].
After all the discussions presented in this chapter, we highlight the importance of
detailing the level of annotation used by researchers in the works using the annota-
tion tools, since this will bring greater reliability and robustness to the studies devel-
oped as well as reduce the number of false- positives described in the literature.

9.7 Perspectives

Metabolomics has provided deeper knowledge about the cellular and biological
transformations at different stages of microorganism growth and interactions [133].
Many challenges are still recurrent in microbial metabolomics studies, including the
identification of metabolites; how to attribute which microorganism produces a spe-
cific metabolite in a microbiome, or what are the biological function of specialized
metabolites [134]. Through the correct use of the bioinformatics approaches and
9 Microbial Metabolites Annotation by Mass Spectrometry-Based Metabolomics 239

tools described in this chapter, we show that it is possible to perform reliable metab-
olomics investigations with microbial datasets, besides being one of the main
emerging trends in metabolomics. Furthermore, the integration of data obtained by
NMR and MS has shown encouraging results, and many case studies presented in
this chapter confirmed the efficiencies of these approaches, and in silico tools. The
creation and development of public platforms, tools and libraries can improve the
development of computational approaches and the annotation of metabolites, once
the higher the number of public available spectra, the higher the possibilities of
annotations. We also believe that the integration of data, such as metabolomics and
genomics can largely contribute to reveal the chemical content present in microbial
samples, and the development of platforms such as the Paired Omics Data Platform
[135] and tools that connect such data, such as NPOmix [136], NPLinker [137],
MetaMiner [138] and NRPMiner [139] will have a great impact in the field.
The main challenge so far remains the necessity for reliable annotation of metab-
olites from complex mass spectral data; even with the emergence of many microbial
natural product databases, it still means a very small number of annotations.
However, this scenario is changing and has shown promise with platforms for shar-
ing natural products data. The main one to be highlighted here is the Global Natural
Products Social Molecular Networking, which has been helpful for researchers in
the field of metabolomics and currently brings together a significant number of tools
for annotation of metabolites, as well as mass spectrometry data. Therefore, it is
hoped that these advances can encourage MS data sharing to automate global anno-
tation of specialized metabolites. Lastly, the potential of metabolomics is beginning
to be explored and associated with other fields, e.g. genomics and proteomics. Thus,
for the next few years, we look forward to the development of this field and a fast
advance in annotation of metabolites, specifically natural products from
microorganisms.

References

1. Katz L, Baltz RH (2016) Natural product discovery: past, present, and future. J Ind Microbiol
Biotechnol 43:155–176
2. Land M, Hauser L, Jun S-R, Nookaew I, Leuze MR, Ahn T-H, Karpinets T, Lund O, Kora
G, Wassenaar T, Poudel S, Ussery DW (2015) Insights from 20 years of bacterial genome
sequencing. Funct Integr Genomics 15:141–161
3. Newman DJ, Cragg GM, Grothaus PG (2017) In: Newman DJ, Crag GM (eds) Chemical
biology of natural products |, 1st edn. Taylor & Francis
4. Dias DA, Urban S, Roessner U (2012) A historical overview of natural products in drug
discovery. Metabolites 2:303–336
5. Patridge E, Gareiss P, Kinch MS, Hoyer D (2016) An analysis of FDA-approved drugs: natu-
ral products and their derivatives. Drug Discov Today 21:204–207
6. Núñez-Montero K, Barrientos L (2018) Advances in Antarctic research for antimicrobial dis-
covery: a comprehensive narrative review of bacteria from Antarctic environments as poten-
tial sources of novel antibiotic compounds against human pathogens and microorganisms of
industrial importance. Antibiotics (Basel) 7. https://doi.org/10.3390/antibiotics7040090
240 P. W. P. Gomes et al.

7. Newman DJ, Cragg GM (2016) Natural products as sources of new drugs from 1981 to 2014.
J Nat Prod 79:629–661
8. Schneider HG, Tener GM, Strong FM (1952) Separation and determination of antimycins.
Arch Biochem Biophys 37:147–157
9. Gu C-Z, Yuan S-H, Lü J, Qiao Y-J, Song Y-Y, Abdalla Elzaki ME, Yang C-R, Zhang Y-J, Zeng
R-S (2019) Albocycline-type macrolides with antibacterial activities from Streptomyces sp.
4205. Chem Biodivers 16:e1800344
10. Harunari E, Imada C, Igarashi Y (2019) Konamycins A and B and Rubromycins CA1
and CA2, aromatic polyketides from the tunicate-derived Streptomyces hyaluromycini
MB-PO13T. J Nat Prod 82:1609–1615
11. Hong S-H, Ban YH, Byun WS, Kim D, Jang Y-J, An JS, Shin B, Lee SK, Shin J, Yoon YJ, Oh
D-C (2019) Camporidines A and B: Antimetastatic and anti-inflammatory polyketide alka-
loids from a gut bacterium of Camponotus kiusiuensis. J Nat Prod 82:903–910
12. Cooper MA, Shlaes D (2011) Fix the antibiotics pipeline. Nature 472:32
13. Verpoorte R, Choi YH, Kim HK (2005) Ethnopharmacology and systems biology: a perfect
holistic match. J Ethnopharmacol 100:53–56
14. Nothias L-F, Nothias-Esposito M, da Silva R, Wang M, Protsyuk I, Zhang Z, Sarvepalli A,
Leyssen P, Touboul D, Costa J, Paolini J, Alexandrov T, Litaudon M, Dorrestein PC (2018)
Bioactivity-based molecular networking for the discovery of drug leads in natural product
bioassay-guided fractionation. J Nat Prod 81:758–767
15. Baltz RH (2019) Natural product drug discovery in the genomic era: realities, conjectures,
misconceptions, and opportunities. J Ind Microbiol Biotechnol 46:281–299
16. Onaka H (2017) Novel antibiotic screening methods to awaken silent or cryptic secondary
metabolic pathways in actinomycetes. J Antibiot 70:865–870
17. Gowda GAN, Djukovic D (2014) Overview of mass spectrometry-based metabolomics:
opportunities and challenges. Methods Mol Biol 1198:3–12
18. Yamashita M, Fenn JB (1984) Electrospray ion source. Another variation on the free-jet
theme. J Phys Chem 88:4451–4459
19. Horai H, Arita M, Kanaya S, Nihei Y, Ikeda T, Suwa K, Ojima Y, Tanaka K, Tanaka
S, Aoshima K, Oda Y, Kakazu Y, Kusano M, Tohge T, Matsuda F, Sawada Y, Hirai MY,
Nakanishi H, Ikeda K, Akimoto N, Maoka T, Takahashi H, Ara T, Sakurai N, Suzuki H,
Shibata D, Neumann S, Iida T, Tanaka K, Funatsu K, Matsuura F, Soga T, Taguchi R, Saito
K, Nishioka T (2010) MassBank: a public repository for sharing mass spectral data for life
sciences. J Mass Spectrom 45:703–714
20. Wishart DS, Tzur D, Knox C, Eisner R, Guo AC, Young N, Cheng D, Jewell K, Arndt D,
Sawhney S, Fung C, Nikolai L, Lewis M, Coutouly M-A, Forsythe I, Tang P, Shrivastava S,
Jeroncic K, Stothard P, Amegbey G, Block D, Hau DD, Wagner J, Miniaci J, Clements M,
Gebremedhin M, Guo N, Zhang Y, Duggan GE, Macinnis GD, Weljie AM, Dowlatabadi R,
Bamforth F, Clive D, Greiner R, Li L, Marrie T, Sykes BD, Vogel HJ, Querengesser L (2007)
HMDB: the human metabolome database. Nucleic Acids Res 35:D521–D526
21. Wishart DS, Feunang YD, Marcu A, Guo AC, Liang K, Vázquez-Fresno R, Sajed T, Johnson
D, Li C, Karu N, Sayeeda Z, Lo E, Assempour N, Berjanskii M, Singhal S, Arndt D, Liang Y,
Badran H, Grant J, Serra-Cayuela A, Liu Y, Mandal R, Neveu V, Pon A, Knox C, Wilson M,
Manach C, Scalbert A (2018) HMDB 4.0: the human metabolome database for 2018. Nucleic
Acids Res 46:D608–D617
22. Shteynberg D, Mendoza L, Hoopmann MR, Sun Z, Schmidt F, Deutsch EW, Moritz RL
(2015) reSpect: software for identification of high and low abundance ion species in chimeric
tandem mass spectra. J Am Soc Mass Spectrom 26:1837–1847
23. MzCloud – advanced mass spectral database. https://www.mzcloud.org/. Accessed 14
Feb 2022
24. Wang M, Carver JJ, Phelan VV, Sanchez LM, Garg N, Peng Y, Nguyen DD, Watrous J,
Kapono CA, Luzzatto-Knaan T, Porto C, Bouslimani A, Melnik AV, Meehan MJ, Liu W-T,
Crüsemann M, Boudreau PD, Esquenazi E, Sandoval-Calderón M, Kersten RD, Pace LA,
9 Microbial Metabolites Annotation by Mass Spectrometry-Based Metabolomics 241

Quinn RA, Duncan KR, Hsu C-C, Floros DJ, Gavilan RG, Kleigrewe K, Northen T, Dutton
RJ, Parrot D, Carlson EE, Aigle B, Michelsen CF, Jelsbak L, Sohlenkamp C, Pevzner P,
Edlund A, McLean J, Piel J, Murphy BT, Gerwick L, Liaw C-C, Yang Y-L, Humpf H-U,
Maansson M, Keyzers RA, Sims AC, Johnson AR, Sidebottom AM, Sedio BE, Klitgaard A,
Larson CB, P CAB, Torres-Mendoza D, Gonzalez DJ, Silva DB, Marques LM, Demarque
DP, Pociute E, O’Neill EC, Briand E, EJN H, Granatosky EA, Glukhov E, Ryffel F, Houson
H, Mohimani H, Kharbush JJ, Zeng Y, Vorholt JA, Kurita KL, Charusanti P, McPhail KL,
Nielsen KF, Vuong L, Elfeki M, Traxler MF, Engene N, Koyama N, Vining OB, Baric R,
Silva RR, Mascuch SJ, Tomasi S, Jenkins S, Macherla V, Hoffman T, Agarwal V, Williams
PG, Dai J, Neupane R, Gurr J, Rodríguez AMC, Lamsa A, Zhang C, Dorrestein K, Duggan
BM, Almaliti J, Allard P-M, Phapale P, Nothias L-F, Alexandrov T, Litaudon M, Wolfender
J-L, Kyle JE, Metz TO, Peryea T, Nguyen D-T, Van Leer D, Shinn P, Jadhav A, Müller
R, Waters KM, Shi W, Liu X, Zhang L, Knight R, Jensen PR, Palsson BO, Pogliano K,
Linington RG, Gutiérrez M, Lopes NP, Gerwick WH, Moore BS, Dorrestein PC, Bandeira
N (2016) Sharing and community curation of mass spectrometry data with global natural
products social molecular networking. Nat Biotechnol 34:828–837
25. Bowen BP, Northen TR (2010) Dealing with the unknown: metabolomics and metabolite
atlases. J Am Soc Mass Spectrom 21:1471–1476
26. Sumner LW, Amberg A, Barrett D, Beale MH, Beger R, Daykin CA, Fan TW-M, Fiehn O,
Goodacre R, Griffin JL, Hankemeier T, Hardy N, Harnly J, Higashi R, Kopka J, Lane AN,
Lindon JC, Marriott P, Nicholls AW, Reily MD, Thaden JJ, Viant MR (2007) Proposed mini-
mum reporting standards for chemical analysis Chemical Analysis Working Group (CAWG)
Metabolomics Standards Initiative (MSI). Metabolomics 3:211–221
27. Schymanski EL, Jeon J, Gulde R, Fenner K, Ruff M, Singer HP, Hollender J (2014) Identifying
small molecules via high resolution mass spectrometry: communicating confidence. Environ
Sci Technol 48:2097–2098
28. Eljarrat E, Barceló D (2005) MASS SPECTROMETRY | electron impact and chemical ion-
ization. In: Worsfold P, Townshend A, Poole C (eds) Encyclopedia of analytical science, 2nd
edn. Elsevier, Oxford, pp 359–366
29. Whitehouse CM, Dreyer RN, Yamashita M, Fenn JB (1985) Electrospray interface for liquid
chromatographs and mass spectrometers. Anal Chem 57:675–679
30. Dettmer K, Aronov PA, Hammock BD (2007) Mass spectrometry-based metabolomics. Mass
Spectrom Rev 26:51–78
31. Ionization Techniques in Mass Spectrometry: A Review. https://www.omicsonline.org/open-­
access/ionization-­techniques-­in-­mass-­spectrometry-­a-­review-­2469-­9861-­1000126-­102758.
html. Accessed 14 Feb 2022
32. Banerjee S, Mazumdar S (2012) Electrospray ionization mass spectrometry: a technique
to access the information beyond the molecular weight of the analyte. Int J Anal Chem
2012:282574
33. Nolting D, Malek R, Makarov A (2019) Ion traps in modern mass spectrometry. Mass
Spectrom Rev 38:150–168
34. Theodoridis G, Gika HG, Wilson ID (2008) LC-MS-based methodology for global metabo-
lite profiling in metabonomics/metabolomics. Trends Anal Chem 27:251–260
35. Domingo-Almenara X, Montenegro-Burke JR, Benton HP, Siuzdak G (2018) Annotation:
a computational solution for streamlining metabolomics analysis. Anal Chem 90:480–489
36. Sorokina M, Merseburger P, Rajan K, Yirik MA, Steinbeck C (2021) COCONUT online:
collection of open natural products database. J Cheminform 13:2
37. Rutz A, Sorokina M, Galgonek J, Mietchen D, Willighagen E, Gaudry A, Graham JG,
Stephan R, Page R, Vondrášek J, Steinbeck C, Pauli GF, Wolfender J-L, Bisson J, Allard P-M
(2021) The LOTUS initiative for open natural products research: knowledge management
through Wikidata. bioRxiv:2021.02.28.433265
38. Kanehisa M, Furumichi M, Tanabe M, Sato Y, Morishima K (2017) KEGG: new perspectives
on genomes, pathways, diseases and drugs. Nucleic Acids Res 45:D353–D361
242 P. W. P. Gomes et al.

39. Hastings J, Owen G, Dekker A, Ennis M, Kale N, Muthukrishnan V, Turner S, Swainston N,

Mendes P, Steinbeck C (2016) ChEBI in 2016: improved services and an expanding collec-
tion of metabolites. Nucleic Acids Res 44:D1214–D1219
40. Pence HE, Williams A (2010) ChemSpider: an online chemical information resource. J Chem
Educ 87:1123–1124
41. Wishart DS, Guo A, Oler E, Wang F, Anjum A, Peters H, Dizon R, Sayeeda Z, Tian S, Lee
BL, Berjanskii M, Mah R, Yamamoto M, Jovel J, Torres-Calzada C, Hiebert-Giesbrecht M,
Lui VW, Varshavi D, Varshavi D, Allen D, Arndt D, Khetarpal N, Sivakumaran A, Harford
K, Sanford S, Yee K, Cao X, Budinski Z, Liigand J, Zhang L, Zheng J, Mandal R, Karu N,
Dambrova M, Schiöth HB, Greiner R, Gautam V (2022) HMDB 5.0: the human metabolome
database for 2022. Nucleic Acids Res 50:D622–D631
42. Xue J, Guijas C, Benton HP, Warth B, Siuzdak G (2020) METLIN MS2 molecular standards
database: a broad chemical and biological resource. Nat Methods 17:953–954
43. da Silva RR, Dorrestein PC, Quinn RA (2015) Illuminating the dark matter in metabolomics.
Proc Natl Acad Sci U S A 112:12549–12550
44. Aron AT, Gentry EC, McPhail KL, Nothias L-F, Nothias-Esposito M, Bouslimani A, Petras
D, Gauglitz JM, Sikora N, Vargas F, van der Hooft JJJ, Ernst M, Kang KB, Aceves CM,
Caraballo-Rodríguez AM, Koester I, Weldon KC, Bertrand S, Roullier C, Sun K, Tehan RM,
Boya PCA, Christian MH, Gutiérrez M, Ulloa AM, Tejeda Mora JA, Mojica-Flores R, Lakey-­
Beitia J, Vásquez-Chaves V, Zhang Y, Calderón AI, Tayler N, Keyzers RA, Tugizimana F,
Ndlovu N, Aksenov AA, Jarmusch AK, Schmid R, Truman AW, Bandeira N, Wang M,
Dorrestein PC (2020) Reproducible molecular networking of untargeted mass spectrometry
data using GNPS. Nat Protoc 15:1954–1991
45. Huber F, Ridder L, Verhoeven S, Spaaks JH, Diblen F, Rogers S, van der Hooft JJJ (2021)
Spec2Vec: improved mass spectral similarity scoring through learning of structural relation-
ships. PLoS Comput Biol 17:e1008724
46. Watrous J, Roach P, Alexandrov T, Heath BS, Yang JY, Kersten RD, van der Voort M,
Pogliano K, Gross H, Raaijmakers JM, Moore BS, Laskin J, Bandeira N, Dorrestein PC
(2012) Mass spectral molecular networking of living microbial colonies. Proc Natl Acad Sci
U S A 109:E1743–E1752
47. Machushynets NV, Wu C, Elsayed SS, Hankemeier T, van Wezel GP (2019) Discovery
of novel glycerolated quinazolinones from Streptomyces sp. MBT27. J Ind Microbiol
Biotechnol 46:483–492
48. Marner M, Patras MA, Kurz M, Zubeil F, Förster F, Schuler S, Bauer A, Hammann P,
Vilcinskas A, Schäberle TF, Glaeser J (2020) Molecular networking-guided discovery
and characterization of stechlisins, a group of cyclic lipopeptides from a sp. J Nat Prod
83:2607–2617
49. Crüsemann M, O’Neill EC, Larson CB, Melnik AV, Floros DJ, da Silva RR, Jensen PR,
Dorrestein PC, Moore BS (2017) Prioritizing natural product diversity in a collection of 146
bacterial strains based on growth and extraction protocols. J Nat Prod 80:588–597
50. Oppong-Danquah E, Parrot D, Blümel M, Labes A, Tasdemir D (2018) Molecular networking-­
based metabolome and bioactivity analyses of marine-adapted fungi co-cultivated with
Phytopathogens. Front Microbiol 9:2072
51. Bauermeister A, Pereira F, Grilo IR, Godinho CC, Paulino M, Almeida V, Gobbo-Neto L,
Prieto-Davó A, Sobral RG, Lopes NP, Gaudêncio SP (2019) Intra-clade metabolomic profil-
ing of MAR4 Streptomyces from the Macaronesia Atlantic region reveals a source of anti-­
biofilm metabolites. Environ Microbiol 21:1099–1112
52. Schmid R, Petras D, Nothias L-F, Wang M, Aron AT, Jagels A, Tsugawa H, Rainer J, Garcia-­
Aloy M, Dührkop K, Korf A, Pluskal T, Kameník Z, Jarmusch AK, Caraballo-Rodríguez
AM, Weldon KC, Nothias-Esposito M, Aksenov AA, Bauermeister A, Albarracin Orio A,
Grundmann CO, Vargas F, Koester I, Gauglitz JM, Gentry EC, Hövelmann Y, Kalinina SA,
Pendergraft MA, Panitchpakdi M, Tehan R, Le Gouellec A, Aleti G, Mannochio Russo H,
Arndt B, Hübner F, Hayen H, Zhi H, Raffatellu M, Prather KA, Aluwihare LI, Böcker S,
9 Microbial Metabolites Annotation by Mass Spectrometry-Based Metabolomics 243

McPhail KL, Humpf H-U, Karst U, Dorrestein PC (2021) Ion identity molecular networking
for mass spectrometry-based metabolomics in the GNPS environment. Nat Commun 12:3832
53. Aron AT, Petras D, Schmid R, Gauglitz JM, Büttel I, Antelo L, Zhi H, Nuccio S-P, Saak
CC, Malarney KP, Thines E, Dutton RJ, Aluwihare LI, Raffatellu M, Dorrestein PC (2021)
Native mass spectrometry-based metabolomics identifies metal-binding compounds. Nat
Chem 14:100–109
54. von Eckardstein L, Petras D, Dang T, Cociancich S, Sabri S, Grätz S, Kerwat D, Seidel
M, Pesic A, Dorrestein PC, Royer M, Weston JB, Süssmuth RD (2017) Total synthesis and
biological assessment of novel Albicidins discovered by mass spectrometric networking.
Chemistry 23:15316–15321
55. Aksenov AA, Laponogov I, Zhang Z, Doran SLF, Belluomo I, Veselkov D, Bittremieux W,
Nothias LF, Nothias-Esposito M, Maloney KN, Misra BB, Melnik AV, Smirnov A, Du X,
Jones KL 2nd, Dorrestein K, Panitchpakdi M, Ernst M, van der Hooft JJJ, Gonzalez M,
Carazzone C, Amézquita A, Callewaert C, Morton JT, Quinn RA, Bouslimani A, Orio AA,
Petras D, Smania AM, Couvillion SP, Burnet MC, Nicora CD, Zink E, Metz TO, Artaev V,
Humston-Fulmer E, Gregor R, Meijler MM, Mizrahi I, Eyal S, Anderson B, Dutton R, Lugan
R, Boulch PL, Guitton Y, Prevost S, Poirier A, Dervilly G, Le Bizec B, Fait A, Persi NS, Song
C, Gashu K, Coras R, Guma M, Manasson J, Scher JU, Barupal DK, Alseekh S, Fernie AR,
Mirnezami R, Vasiliou V, Schmid R, Borisov RS, Kulikova LN, Knight R, Wang M, Hanna
GB, Dorrestein PC, Veselkov K (2021) Auto-deconvolution and molecular networking of gas
chromatography-mass spectrometry data. Nat Biotechnol 39:169–173
56. Hiller K, Hangebrauk J, Jäger C, Spura J, Schreiber K, Schomburg D (2009)
MetaboliteDetector: comprehensive analysis tool for targeted and nontargeted GC/MS based
metabolome analysis. Anal Chem 81:3429–3439
57. Herold M, Martínez Arbas S, Narayanasamy S, Sheik AR, Kleine-Borgmann LAK, Lebrun
LA, Kunath BJ, Roume H, Bessarab I, Williams RBH, Gillece JD, Schupp JM, Keim PS,
Jäger C, Hoopmann MR, Moritz RL, Ye Y, Li S, Tang H, Heintz-Buschart A, May P, Muller
EEL, Laczny CC, Wilmes P (2020) Integration of time-series meta-omics data reveals how
microbial ecosystems respond to disturbance. Nat Commun 11:5281
58. Bingol K, Li D-W, Bruschweiler-Li L, Cabrera OA, Megraw T, Zhang F, Brüschweiler R
(2015) Unified and isomer-specific NMR metabolomics database for the accurate analysis of
13C–1H HSQC spectra. ACS Chem Biol 10:452–459
59. Reher R, Kim HW, Zhang C, Mao HH, Wang M, Nothias L-F, Caraballo-Rodriguez AM,
Glukhov E, Teke B, Leao T, Alexander KL, Duggan BM, Van Everbroeck EL, Dorrestein PC,
Cottrell GW, Gerwick WH (2020) A convolutional neural network-based approach for the
rapid annotation of molecularly diverse natural products. J Am Chem Soc 142:4114–4120
60. Blaženović I, Kind T, Ji J, Fiehn O (2018) Software tools and approaches for compound
identification of LC-MS/MS data in metabolomics. Metabolites 8. https://doi.org/10.3390/
metabo8020031
61. Schymanski EL, Ruttkies C, Krauss M, Brouard C, Kind T, Dührkop K, Allen F, Vaniya
A, Verdegem D, Böcker S, Rousu J, Shen H, Tsugawa H, Sajed T, Fiehn O, Ghesquière B,
Neumann S (2017) Critical assessment of small molecule identification 2016: automated
methods. J Cheminform 9:22
62. van Santen JA, Jacob G, Singh AL, Aniebok V, Balunas MJ, Bunsko D, Neto FC, Castaño-­
Espriu L, Chang C, Clark TN, Cleary Little JL, Delgadillo DA, Dorrestein PC, Duncan KR,
Egan JM, Galey MM, Haeckl FPJ, Hua A, Hughes AH, Iskakova D, Khadilkar A, Lee J-H, Lee
S, LeGrow N, Liu DY, Macho JM, McCaughey CS, Medema MH, Neupane RP, O’Donnell
TJ, Paula JS, Sanchez LM, Shaikh AF, Soldatou S, Terlouw BR, Tran TA, Valentine M, van
der Hooft JJJ, Vo DA, Wang M, Wilson D, Zink KE, Linington RG (2019) The natural prod-
ucts atlas: an open access knowledge base for microbial natural products discovery. ACS
Cent Sci 5:1824–1833
63. Ruttkies C, Schymanski EL, Wolf S, Hollender J, Neumann S (2016) MetFrag relaunched:
incorporating strategies beyond in silico fragmentation. J Cheminform 8:3
244 P. W. P. Gomes et al.

64. Ruttkies C, Neumann S, Posch S (2019) Improving MetFrag with statistical learning of frag-
ment annotations. BMC Bioinformatics 20:376
65. Wang F, Liigand J, Tian S, Arndt D, Greiner R, Wishart DS (2021) CFM-ID 4.0: more accurate
ESI-MS/MS spectral prediction and compound identification. Anal Chem 93:11692–11700
66. Veličković D, Chu RK, Myers GL, Ahkami AH, Anderton CR (2020) An approach for visu-
alizing the spatial metabolome of an entire plant root system inspired by the Swiss-rolling
technique. J Mass Spectrom 55:e4363
67. Delgado de la Torre MP, Priego-Capote F, Luque de Castro MD (2015) Tentative identifica-
tion of polar and mid-polar compounds in extracts from wine lees by liquid chromatography-­
tandem mass spectrometry in high-resolution mode. J Mass Spectrom 50:826–837
68. Woo S, Kang KB, Kim J, Sung SH (2019) Molecular networking reveals the chemical diver-
sity of Selaginellin derivatives, natural Phosphodiesterase-4 inhibitors from Selaginella tam-
ariscina. J Nat Prod 82:1820–1830
69. Nothias-Esposito M, Nothias LF, Da Silva RR, Retailleau P, Zhang Z, Leyssen P, Roussi F,
Touboul D, Paolini J, Dorrestein PC, Litaudon M (2019) Investigation of Premyrsinane and
Myrsinane esters in euphorbia cupanii and Euphobia pithyusa with MS2LDA and combinato-
rial molecular network annotation propagation. J Nat Prod 82:1459–1470
70. Kešnerová L, Mars RAT, Ellegaard KM, Troilo M, Sauer U, Engel P (2017) Disentangling
metabolic functions of bacteria in the honey bee gut. PLoS Biol 15:e2003467
71. Allen F, Pon A, Wilson M, Greiner R, Wishart D (2014) CFM-ID: a web server for annota-
tion, spectrum prediction and metabolite identification from tandem mass spectra. Nucleic
Acids Res 42:W94–W99
72. Allen F, Greiner R, Wishart D (2015) Competitive fragmentation modeling of ESI-MS/MS
spectra for putative metabolite identification. Metabolomics 11:98–110
73. Allen F, Pon A, Greiner R, Wishart D (2016) Computational prediction of electron ionization
mass spectra to assist in GC/MS compound identification. Anal Chem 88:7689–7697
74. Djoumbou-Feunang Y, Pon A, Karu N, Zheng J, Li C, Arndt D, Gautam M, Allen F, Wishart
DS (2019) CFM-ID 3.0: significantly improved ESI-MS/MS prediction and compound iden-
tification. Metabolites 9. https://doi.org/10.3390/metabo9040072
75. Nguyen DH, Nguyen CH, Mamitsuka H (2019) Recent advances and prospects of compu-
tational methods for metabolite identification: a review with emphasis on machine learning
approaches. Brief Bioinform 20:2028–2043
76. Utermann C, Parrot D, Breusing C, Stuckas H, Staufenberger T, Blümel M, Labes A,
Tasdemir D (2018) Combined genotyping, microbial diversity and metabolite profiling stud-
ies on farmed Mytilus spp. from Kiel Fjord. Sci Rep 8:7983
77. Alden N, Krishnan S, Porokhin V, Raju R, McElearney K, Gilbert A, Lee K (2017)
Biologically consistent annotation of metabolomics data. Anal Chem 89:13097–13104
78. Liu J, Nothias L-F, Dorrestein PC, Tahlan K, Bignell DRD (2021) Genomic and Metabolomic
analysis of the potato common scab pathogen Streptomyces scabiei. ACS Omega
6:11474–11487
79. Yang D, Jang WD, Lee SY (2021) Production of Carminic acid by metabolically engineered
Escherichia coli. J Am Chem Soc 143:5364–5377
80. Böcker S, Dührkop K (2016) Fragmentation trees reloaded. J Cheminform 8:5
81. Vaniya A, Fiehn O (2015) Using fragmentation trees and mass spectral trees for identifying
unknown compounds in metabolomics. Trends Anal Chem 69:52–61
82. Dührkop K, Fleischauer M, Ludwig M, Aksenov AA, Melnik AV, Meusel M, Dorrestein
PC, Rousu J, Böcker S (2019) SIRIUS 4: a rapid tool for turning tandem mass spectra into
metabolite structure information. Nat Methods 16:299–302
83. Ludwig M, Nothias L-F, Dührkop K, Koester I, Fleischauer M, Hoffmann MA, Petras D,
Vargas F, Morsy M, Aluwihare L, Dorrestein PC, Böcker S (2020) Database-independent
molecular formula annotation using Gibbs sampling through ZODIAC. Nat Mach Intell
2:629–641
9 Microbial Metabolites Annotation by Mass Spectrometry-Based Metabolomics 245

84. Dührkop K, Shen H, Meusel M, Rousu J, Böcker S (2015) Searching molecular struc-
ture databases with tandem mass spectra using CSI:FingerID. Proc Natl Acad Sci U S A
112:12580–12585
85. Dührkop K, Nothias L-F, Fleischauer M, Reher R, Ludwig M, Hoffmann MA, Petras D,
Gerwick WH, Rousu J, Dorrestein PC, Böcker S (2021) Systematic classification of unknown
metabolites using high-resolution fragmentation mass spectra. Nat Biotechnol 39:462–471
86. Djoumbou Feunang Y, Eisner R, Knox C, Chepelev L, Hastings J, Owen G, Fahy E, Steinbeck
C, Subramanian S, Bolton E, Greiner R, Wishart DS (2016) ClassyFire: automated chemical
classification with a comprehensive, computable taxonomy. J Cheminform 8:61
87. Jaiyesimi OA, McAvoy AC, Fogg DN, Garg N (2021) Metabolomic profiling of Burkholderia
cenocepacia in synthetic cystic fibrosis sputum medium reveals nutrient environment-specific
production of virulence factors. Sci Rep 11:21419
88. Klamrak A, Nabnueangsap J, Puthongking P, Nualkaew N (2021) Synthesis of Ferulenol by
engineered Escherichia coli: structural elucidation by using the in silico tools. Molecules 26.
https://doi.org/10.3390/molecules26206264
89. Hoffmann MA, Nothias L-F, Ludwig M, Fleischauer M, Gentry EC, Witting M, Dorrestein
PC, Dührkop K, Böcker S (2021) High-confidence structural annotation of metabolites
absent from spectral libraries. Nat Biotechnol. https://doi.org/10.1038/s41587-­021-­01045-­9
90. Meng X, Tao F, Xu P (2021) Single-cell metabolomics reveals the metabolic heterogeneity
among microbial cells. bioRxiv:2021.11.08.467686
91. Gonçalves E, Frezza C (2021) Genome and metabolome: chance and necessity. Genome
Biol 22:276
92. Tsugawa H, Kind T, Nakabayashi R, Yukihira D, Tanaka W, Cajka T, Saito K, Fiehn O, Arita
M (2016) Hydrogen rearrangement rules: computational MS/MS fragmentation and structure
elucidation using MS-FINDER software. Anal Chem 88:7946–7958
93. Lai Z, Tsugawa H, Wohlgemuth G, Mehta S, Mueller M, Zheng Y, Ogiwara A, Meissen J,
Showalter M, Takeuchi K, Kind T, Beal P, Arita M, Fiehn O (2018) Identifying metabolites
by integrating metabolome databases with mass spectrometry cheminformatics. Nat Methods
15:53–56
94. Sun Y, Liu W-C, Shi X, Zheng H-Z, Zheng Z-H, Lu X-H, Xing Y, Ji K, Liu M, Dong Y-S
(2021) Inducing secondary metabolite production of aspergillus sydowii through microbial
co-culture with Bacillus subtilis. Microb Cell Factories 20:42
95. Lai Y, Liu C-W, Yang Y, Hsiao Y-C, Ru H, Lu K (2021) High-coverage metabolomics uncov-
ers microbiota-driven biochemical landscape of interorgan transport and gut-brain communi-
cation in mice. Nat Commun 12:6000
96. Stranska M, Lovecka P, Vrchotova B, Uttl L, Bechynska K, Behner A, Hajslova J (2021)
Bacterial endophytes from vitis vinifera L. – metabolomics characterization of plant-­
endophyte crosstalk. Chem Biodivers 18:e2100516
97. Ridder L, van der Hooft JJJ, Verhoeven S, de Vos RCH, Vervoort J, Bino RJ (2014) In silico
prediction and automatic LC-MS(n) annotation of green tea metabolites in urine. Anal Chem
86:4767–4774
98. Ridder L, van der Hooft JJJ, Verhoeven S, de Vos RCH, Bino RJ, Vervoort J (2013) Automatic
chemical structure annotation of an LC-MS(n) based metabolic profile from green tea. Anal
Chem 85:6033–6040
99. Ridder L, van der Hooft JJJ, Verhoeven S, de Vos RCH, van Schaik R, Vervoort J (2012)
Substructure-based annotation of high-resolution multistage MS(n) spectral trees. Rapid
Commun Mass Spectrom 26:2461–2471
100. da Silva RR, Wang M, Nothias L-F, van der Hooft JJJ, Caraballo-Rodríguez AM, Fox E,
Balunas MJ, Klassen JL, Lopes NP, Dorrestein PC (2018) Propagating annotations of molec-
ular networks using in silico fragmentation. PLoS Comput Biol 14:e1006089
101. Kim H, Wang M, Leber C, Nothias L-F, Reher R, Kang KB, van der Hooft JJJ, Dorrestein P,
Gerwick W, Cottrell G (2020) NPClassifier: a deep neural network-based structural classifi-
cation tool for natural products. ChemRxiv 10.26434/chemrxiv.12885494.v1
246 P. W. P. Gomes et al.

102. Ferreira L, Morais J, Preto M, Silva R, Urbatzka R, Vasconcelos V, Reis M (2021) Uncovering
the bioactive potential of a cyanobacterial natural products library aided by untargeted metab-
olomics. Mar Drugs 19. https://doi.org/10.3390/md19110633
103. Ernst M, Nothias L-F, van der Hooft JJJ, Silva RR, Saslis-Lagoudakis CH, Grace OM,
Martinez-Swatson K, Hassemer G, Funez LA, Simonsen HT, Medema MH, Staerk D, Nilsson
N, Lovato P, Dorrestein PC, Rønsted N (2019) Assessing specialized metabolite diversity in
the cosmopolitan plant genus euphorbia L. Front Plant Sci 10:846
104. Di Ottavio F, Gauglitz JM, Ernst M, Panitchpakdi MW, Fanti F, Compagnone D, Dorrestein
PC, Sergi M (2020) A UHPLC-HRMS based metabolomics and chemoinformatics approach
to chemically distinguish “super foods” from a variety of plant-based foods. Food Chem
313:126071
105. Pham HT, Lee KH, Jeong E, Woo S, Yu J, Kim W-Y, Lim YW, Kim KH, Kang KB (2021)
Species prioritization based on spectral dissimilarity: a case study of Polyporoid fungal spe-
cies. J Nat Prod 84:298–309
106. Maimone NM, de Oliveira LFP, Santos SN, de Lira SP (2021) Elicitation of Streptomyces
lunalinharesii secondary metabolism through co-cultivation with Rhizoctonia solani.
Microbiol Res 251:126836
107. Mohimani H, Gurevich A, Mikheenko A, Garg N, Nothias L-F, Ninomiya A, Takada K,
Dorrestein PC, Pevzner PA (2017) Dereplication of peptidic natural products through data-
base search of mass spectra. Nat Chem Biol 13:30–37
108. Mohimani H, Kim S, Pevzner PA (2013) A new approach to evaluating statistical significance
of spectral identifications. J Proteome Res 12:1560–1568
109. Gurevich A, Mikheenko A, Shlemov A, Korobeynikov A, Mohimani H, Pevzner PA (2018)
Increased diversity of peptidic natural products revealed by modification-tolerant database
search of mass spectra. Nat Microbiol 3:319–327
110. Horsman ME, Boddy CN (2016) Natural products: mapping an amazing thicket. Nat Chem
Biol 13:6–7
111. Atencio LA, Boya PCA, Martin HC, Mejía LC, Dorrestein PC, Gutiérrez M (2020)
Genome mining, microbial interactions, and molecular networking reveals new
Dibromoalterochromides from strains of Pseudoalteromonas of Coiba National Park-­
Panama. Mar Drugs 18. https://doi.org/10.3390/md18090456
112. Mohimani H, Gurevich A, Shlemov A, Mikheenko A, Korobeynikov A, Cao L, Shcherbin
E, Nothias L-F, Dorrestein PC, Pevzner PA (2018) Dereplication of microbial metabolites
through database search of mass spectra. Nat Commun 9:4035
113. Kersten RD, Ziemert N, Gonzalez DJ, Duggan BM, Nizet V, Dorrestein PC, Moore BS
(2013) Glycogenomics as a mass spectrometry-guided genome-mining method for microbial
glycosylated molecules. Proc Natl Acad Sci U S A 110:E4407–E4416
114. Velasco-Alzate KY, Bauermeister A, Tangerina MMP, Lotufo TMC, Ferreira MJP, Jimenez
PC, Padilla G, Lopes NP, Costa-Lotufo LV (2019) Marine bacteria from Rocas atoll as a
rich source of pharmacologically active compounds. Mar Drugs 17. https://doi.org/10.3390/
md17120671
115. Kuhn S, Colreavy-Donnelly S, de Andrade Silva Quaresma LE, de Andrade Silva Quaresma
E, Borges RM (2020) Applying NMR compound identification using NMRfilter to match
predicted to experimental data. Metabolomics 16:123
116. van der Hooft JJJ, Wandy J, Barrett MP, Burgess KEV, Rogers S (2016) Topic model-
ing for untargeted substructure exploration in metabolomics. Proc Natl Acad Sci U S A
113:13738–13743
117. Wandy J, Zhu Y, van der Hooft JJJ, Daly R, Barrett MP, Rogers S (2018) Ms2lda.org: web-­
based topic modelling for substructure discovery in mass spectrometry. Bioinformatics
34:317–318
118. Rogers S, Ong CW, Wandy J, Ernst M, Ridder L, van der Hooft JJJ (2019) Deciphering
complex metabolite mixtures by unsupervised and supervised substructure discovery and
semi-automated annotation from MS/MS spectra. Faraday Discuss 218:284–302
9 Microbial Metabolites Annotation by Mass Spectrometry-Based Metabolomics 247

119. Jarmusch SA, Lagos-Susaeta D, Diab E, Salazar O, Asenjo JA, Ebel R, Jaspars M (2021)
Iron-meditated fungal starvation by lupine rhizosphere-associated and extremotolerant
Streptomyces sp. S29 desferrioxamine production. Mol Omics 17:95–107
120. Zdouc MM, Iorio M, Maffioli SI, Crüsemann M, Donadio S, Sosio M (2021) Planomonospora:
a metabolomics perspective on an underexplored Actinobacteria genus. J Nat Prod 84:204–219
121. Liu Y, Mrzic A, Meysman P, De Vijlder T, Romijn EP, Valkenborg D, Bittremieux W, Laukens
K (2020) MESSAR: automated recommendation of metabolite substructures from tandem
mass spectra. PLoS One 15:e0226770
122. Vu TN, Mrzic A, Valkenborg D, Maes E, Lemière F, Goethals B, Laukens K (2014)
Unravelling associations between unassigned mass spectrometry peaks with frequent itemset
mining techniques. Proteome Sci 12:54
123. Run Classification – ClassyFire. http://classyfire.wishartlab.com/. Accessed 14 Feb 2022
124. Moreno-Ulloa A, Sicairos Diaz V, Tejeda-Mora JA, Macias Contreras MI, Castillo FD,
Guerrero A, Gonzalez Sanchez R, Mendoza-Porras O, Vazquez Duhalt R, Licea-Navarro A
(2020) Chemical profiling provides insights into the metabolic machinery of hydrocarbon-­
degrading Deep-Sea microbes. mSystems 5. https://doi.org/10.1128/mSystems.00824-­20
125. Tripathi A, Vázquez-Baeza Y, Gauglitz JM, Wang M, Dührkop K, Nothias-Esposito M,
Acharya DD, Ernst M, van der Hooft JJJ, Zhu Q, McDonald D, Brejnrod AD, Gonzalez
A, Handelsman J, Fleischauer M, Ludwig M, Böcker S, Nothias L-F, Knight R, Dorrestein
PC (2021) Chemically informed analyses of metabolomics mass spectrometry data with
Qemistree. Nat Chem Biol 17:146–151
126. Ernst M, Kang KB, Caraballo-Rodríguez AM, Nothias L-F, Wandy J, Chen C, Wang M,
Rogers S, Medema MH, Dorrestein PC, van der Hooft JJJ (2019) MolNetEnhancer: enhanced
molecular networks by integrating metabolome mining and annotation tools. Meta 9. https://
doi.org/10.3390/metabo9070144
127. Villanueva-Silva R, Velez P, Riquelme M, Fajardo-Hernández CA, Martínez-Cárdenas A,
Arista-Romero A, Wan B, Ma R, Qader M, Franzblau SG, Figueroa M (2021) Chemical
diversity and antimicrobial potential of cultivable fungi from Deep-Sea sediments of the Gulf
of Mexico. Molecules 26. https://doi.org/10.3390/molecules26237328
128. Martínez-Cárdenas A, Cruz-Zamora Y, Fajardo-Hernández CA, Villanueva-Silva R, Cruz-­
García F, Raja HA, Figueroa M (2021) Genome mining and molecular networking-based
metabolomics of the marine facultative aspergillus sp. MEXU 27854. Molecules 26. https://
doi.org/10.3390/molecules26175362
129. Marfil-Santana MD, Martínez-Cárdenas A, Ruíz-Hernández A, Vidal-Torres M, Márquez-­
Velázquez NA, Figueroa M, Prieto-Davó A (2021) A meta-omics analysis unveils the shift
in microbial community structures and metabolomics profiles in mangrove sediments treated
with a selective Actinobacterial isolation procedure. Molecules 26. https://doi.org/10.3390/
molecules26237332
130. Soldatou S, Eldjárn GH, Ramsay A, van der Hooft JJJ, Hughes AH, Rogers S, Duncan KR
(2021) Comparative Metabologenomics analysis of polar Actinomycetes. Mar Drugs 19.
https://doi.org/10.3390/md19020103
131. Broeckling CD, Ganna A, Layer M, Brown K, Sutton B, Ingelsson E, Peers G, Prenni JE
(2016) Enabling efficient and confident annotation of LC-MS metabolomics data through
MS1 Spectrum and time prediction. Anal Chem 88:9226–9234
132. Ramanathan R, Cao K, Cavalieri E, Gross ML (1998) Mass spectrometric methods for dis-
tinguishing structural isomers of glutathione conjugates of estrone and estradiol. J Am Soc
Mass Spectrom 9:612–619
133. Vogt N (2020) Putting on the squeeze. Nat Methods 17:24
134. Bauermeister A, Mannochio-Russo H, Costa-Lotufo LV, Jarmusch AK, Dorrestein PC (2021)
Mass spectrometry-based metabolomics in microbiome investigations. Nat Rev Microbiol.
https://doi.org/10.1038/s41579-­021-­00621-­9
135. Schorn MA, Verhoeven S, Ridder L, Huber F, Acharya DD, Aksenov AA, Aleti G,
Moghaddam JA, Aron AT, Aziz S, Bauermeister A, Bauman KD, Baunach M, Beemelmanns
248 P. W. P. Gomes et al.

C, Beman JM, Berlanga-Clavero MV, Blacutt AA, Bode HB, Boullie A, Brejnrod A, Bugni
TS, Calteau A, Cao L, Carrión VJ, Castelo-Branco R, Chanana S, Chase AB, Chevrette MG,
Costa-Lotufo LV, Crawford JM, Currie CR, Cuypers B, Dang T, de Rond T, Demko AM,
Dittmann E, Du C, Drozd C, Dujardin J-C, Dutton RJ, Edlund A, Fewer DP, Garg N, Gauglitz
JM, Gentry EC, Gerwick L, Glukhov E, Gross H, Gugger M, Guillén Matus DG, Helfrich
EJN, Hempel B-F, Hur J-S, Iorio M, Jensen PR, Kang KB, Kaysser L, Kelleher NL, Kim
CS, Kim KH, Koester I, König GM, Leao T, Lee SR, Lee Y-Y, Li X, Little JC, Maloney KN,
Männle D, Martin HC, McAvoy AC, Metcalf WW, Mohimani H, Molina-Santiago C, Moore
BS, Mullowney MW, Muskat M, Nothias L-F, O’Neill EC, Parkinson EI, Petras D, Piel J,
Pierce EC, Pires K, Reher R, Romero D, Roper MC, Rust M, Saad H, Saenz C, Sanchez
LM, Sørensen SJ, Sosio M, Süssmuth RD, Sweeney D, Tahlan K, Thomson RJ, Tobias NJ,
Trindade-Silva AE, van Wezel GP, Wang M, Weldon KC, Zhang F, Ziemert N, Duncan KR,
Crüsemann M, Rogers S, Dorrestein PC, Medema MH, van der Hooft JJJ (2021) A commu-
nity resource for paired genomic and metabolomic data mining. Nat Chem Biol 17:363–368
136. Leao TF, Wang M, da Silva R, van der Hooft JJJ, Bauermeister A, Brejnrod A, Glukhov E,
Gerwick L, Gerwick WH, Bandeira N, Dorrestein PC (2021) A supervised fingerprint-based
strategy to connect natural product mass spectrometry fragmentation data to their biosyn-
thetic gene clusters. BioRxiv
137. Hjörleifsson Eldjárn G, Ramsay A, van der Hooft JJJ, Duncan KR, Soldatou S, Rousu J,
Daly R, Wandy J, Rogers S (2021) Ranking microbial metabolomic and genomic links in
the NPLinker framework using complementary scoring functions. PLoS Comput Biol
17:e1008920
138. Cao L, Gurevich A, Alexander KL, Naman CB, Leão T, Glukhov E, Luzzatto-Knaan T, Vargas
F, Quinn R, Bouslimani A, Nothias LF, Singh NK, Sanders JG, Benitez RAS, Thompson LR,
Hamid M-N, Morton JT, Mikheenko A, Shlemov A, Korobeynikov A, Friedberg I, Knight R,
Venkateswaran K, Gerwick WH, Gerwick L, Dorrestein PC, Pevzner PA, Mohimani H (2019)
MetaMiner: a scalable Peptidogenomics approach for discovery of ribosomal peptide natural
products with blind modifications from microbial communities. Cell Syst 9:600–608.e4
139. Behsaz B, Bode E, Gurevich A, Shi Y-N, Grundmann F, Acharya D, Caraballo-Rodríguez
AM, Bouslimani A, Panitchpakdi M, Linck A, Guan C, Oh J, Dorrestein PC, Bode HB,
Pevzner PA, Mohimani H (2021) Integrating genomics and metabolomics for scalable non-­
ribosomal peptide discovery. Nat Commun 12:3225
Index

A E
Ambient ionization techniques, 102, Extracellular metabolome, 113, 151, 155, 159
103, 112–113
Annotation, 2, 7, 30, 33, 37, 52, 53, 55, 63, 64,
87, 102, 115, 116, 126, 191–197, F
202–211, 227–239 Fungal-plant interactions, 1–3, 13

B G
Bacteriophage, 72, 74–76, 80, 81, 84, Genomics, 22, 28, 33, 34, 39, 53–59, 66, 104,
85, 89, 90 167, 186, 192, 213, 239
Bioinformatics, 28, 33, 34, 37, 39, 52, 54, 56,
59, 62, 115, 150, 186, 189, 192, 202,
237, 238 H
Biotechnological applications, 104 Host-phage, 85, 87

C I
Cellular cultivation, 153 Intracellular metabolome, 151, 156, 159, 168
Chemical elicitation, 59
Chemical interactions, 1–13, 51–66
Cyanobacteria, 21–23, 25–32, 34, 36, 39, 153, L
167, 169, 197, 207, 210, 211, 237 Liquid chromatography coupled to mass
Cyanotoxins, 22–31, 39 spectrometry (LC-MS), 2, 6, 8, 10, 11,
28–30, 33, 34, 37, 63–64, 85, 86, 102,
104–106, 113, 114, 135, 137, 169, 204,
D 226, 228, 229, 236
Data analysis, 30, 39, 59, 66, 86, 108, 116,
186, 190, 191, 195–204, 206, 210, 226,
228, 230 M
Dereplication, 31, 36, 39, 104, 113–117, Mass spectrometry (MS), 1, 28, 52, 85, 101,
191–197, 202, 203, 205, 206, 209, 124, 150, 190, 226
211–213, 235, 236 Mass spectrometry imaging, 36, 108–111

© The Editor(s) (if applicable) and The Author(s), under exclusive license to 249
Springer Nature Switzerland AG 2023
T. Pacheco Fill (ed.), Microbial Natural Products Chemistry, Advances in
Experimental Medicine and Biology 1439,
https://doi.org/10.1007/978-3-031-41741-2
250 Index

Metabolic quenching, 153, 154, 156, 157, 175 NMR spectroscopy, 1, 10, 124
Metabolite annotation, 63, 64, 229, 230, Nuclear magnetic resonance (NMR), 1, 4, 6,
232, 237–238 10, 11, 28, 61, 85–87, 90, 91, 114,
Metabolite extraction, 161 123–140, 150, 155, 159, 161, 164–166,
Metabolomics, 1–3, 5–13, 22–31, 39, 52, 56, 168, 185, 189, 190, 192, 193, 196–201,
58–65, 83–91, 101–107, 109–111, 203, 204, 211–213, 236,
113–115, 117, 123, 125–128, 238, 239
130–132, 134–140, 150–172,
185–213, 225–239
Microbial metabolites, 105, 106, 114, 188, S
190, 197, 210, 227–236 Sample preparation, 6, 29, 31, 36, 39, 62, 64,
Microbial metabolomics, 52, 53, 60, 61, 64, 84, 87, 108–110, 113, 117, 125,
66, 102, 103, 105, 108, 110–113, 115, 127–129, 137, 151–175, 189
117, 150–175, 202, 206, 209, 212, 230, Secondary metabolites, 2, 7–9, 11–13, 21, 22,
231, 238 28, 30, 34, 36, 37, 51–57, 59, 60, 65,
73, 82, 106, 107, 116, 125, 135, 137,
153, 167, 188, 197, 202
N Specialized metabolites, 52, 53, 84, 188,
Natural products discovery, 36 225–227, 237, 239