Methods in

Molecular Biology 2752

Miodrag Gužvić Editor

Single Cell
Analysis
Methods and Protocols
METHODS IN MOLECULAR BIOLOGY

Series Editor
John M. Walker
School of Life and Medical Sciences
University of Hertfordshire
Hatfield, Hertfordshire, UK

For further volumes:

http://www.springer.com/series/7651
For over 35 years, biological scientists have come to rely on the research protocols and
methodologies in the critically acclaimed Methods in Molecular Biology series. The series was
the first to introduce the step-by-step protocols approach that has become the standard in all
biomedical protocol publishing. Each protocol is provided in readily-reproducible step-by
step fashion, opening with an introductory overview, a list of the materials and reagents
needed to complete the experiment, and followed by a detailed procedure that is supported
with a helpful notes section offering tips and tricks of the trade as well as troubleshooting
advice. These hallmark features were introduced by series editor Dr. John Walker and
constitute the key ingredient in each and every volume of the Methods in Molecular Biology
series. Tested and trusted, comprehensive and reliable, all protocols from the series are
indexed in PubMed.
 Single Cell Analysis

Methods and Protocols

Edited by

Miodrag Gužvić
Department of Urology, University Hospital Regensburg, Regensburg, Germany
Editor
Miodrag Gužvić
Department of Urology
University Hospital Regensburg
Regensburg, Germany

ISSN 1064-3745 ISSN 1940-6029 (electronic)

Methods in Molecular Biology
ISBN 978-1-0716-3620-6 ISBN 978-1-0716-3621-3 (eBook)
https://doi.org/10.1007/978-1-0716-3621-3
© Springer Science+Business Media, LLC, part of Springer Nature 2024
This work is subject to copyright. All rights are reserved 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 Humana imprint is published by the registered company Springer Science+Business Media, LLC, part of Springer
Nature.
The registered company address is: 1 New York Plaza, New York, NY 10004, U.S.A.

Paper in this product is recyclable.

Preface

An exciting time lies ahead for cellular biology. We are finally able to move away from
assessing genome, transcriptome, proteome, or metabolome as a mean of a population and
instead focus on the features of single cells. Developments in physics, chemistry, and
molecular biology have pushed the boundaries of what is possible in cellular analysis to
give unprecedented improvements in sensitivity and scale that are allowing us the capability
to approach a full functional appraisal of single cells at a rapid rate.
The speed, sensitivity, accuracy, and scope of existing techniques for single-cell detec-
tion, isolation, and analysis (e.g., FACS, fluorescence microscopy, dielectrophoresis, whole
genome and transcriptome amplification, high-throughput sequencing, etc.) have improved
markedly, broadening research horizons, while complementary techniques in organic spec-
troscopy (e.g., MALDI imaging), inorganic spectroscopy (e.g., ICP-MS), and synchrotron
analysis support this with more detailed information on aspects previously overlooked.
This volume summarizes these techniques, their capabilities, and the type of information
that can be determined, and aims to give an overview of best practice for implementing them
in single-cell analysis in an important and necessary move away from the bulk analysis that is
constraining our boundaries.
Multi-omics analysis of single cells is gaining momentum in recent years. Such
approaches are somewhat underrepresented in this book, and future editions should more
comprehensively cover this important aspect of single-cell analysis. Furthermore, while an
attempt was made to make this volume thematically comprehensive, it is somewhat biased
toward the analysis of single cancer cells. The future editions of this book should broaden its
scope by including protocols on isolation and analysis of single prokaryotic or plant cells.
Still, many thematically overlapping or linked protocols presented in this volume enable
development of complex and complete workflows to isolate and analyze single cells, even
beyond cancer research.

Regensburg, Germany Miodrag Gužvić

v
Acknowledgments

I am grateful to Rob Hutchinson for support in early phases of preparation of this book, and
to contributors and the publisher for their limitless patience, which I tested many times
while wrestling many obstacles trying to bring this book to see the light of the day.
This book is dedicated to Christoph A. Klein, who introduced me to the multiverse of
single-cell analysis.

vii
Contents

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v
Acknowledgments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii
Contributors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
1 Single Cell Isolation from Surgically Resected Tissue Via Mechanical
Dissociation Using TissueGrinder. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Prama Pallavi and Stefan Scheuermann
2 Circulating Tumor Cell Enrichment and Single-Cell Isolation
Combining the CellSearch® and DEPArray™ Systems. . . . . . . . . . . . . . . . . . . . . . . 11
C€a cilia Köstler, Bernhard Polzer, and Barbara Alberter
3 Isolation of Viable Epithelial and Mesenchymal Circulating Tumor Cells
from Breast Cancer Patients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
Justyna Topa, Anna J. Żaczek, and Aleksandra Markiewicz
4 Single-Cell Recovery from Tumor Cell Xenotransplanted Zebrafish
Embryos for the Study of Metastasis-Initiating Cells . . . . . . . . . . . . . . . . . . . . . . . . 53
Pablo Hurtado, Inés Martı́nez-Pena, and Roberto Piñeiro
5 Isolation of Single Circulating Tumor Cells Using VyCAP Puncher System . . . . 65
Thais Pereira-Veiga, Bianca Behrens, Joska J. Broekmaat, Lisa Oomens,
Michiel Stevens, Arjan G. J. Tibbe, Nikolas Stoecklein,
Laura Muinelo-Romay, Roberto Piñeiro, and Clotilde Costa
6 Simultaneous Isolation and Amplification of mRNA and Genomic
DNA of a Single Cell . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
Miodrag Gužvić
7 Isolation and Genomic Analysis of Circulating Tumor Cell Clusters
in Cancer Patients. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
Carolina Reduzzi, Marta Vismara, Thomas Schamberger, Marco Silvestri,
Rosita Motta, Bernhard M. Polzer, and Vera Cappelletti
8 Establishing Single-Cell Clones from In Vitro-Cultured Circulating
Tumor Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
Teng Teng and Min Yu
9 Immunofluorescence Combined with Single-Molecule RNA
Fluorescence In Situ Hybridization for Concurrent Detection
of Proteins and Transcripts in Stress Granules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
Jakub Kochan and Mateusz Wawro
10 Highly Multiplexed and Simultaneous Characterization of Protein
and RNA in Single Cells by Flow or Mass Cytometry Platforms
Using Proximity Ligation Assay for RNA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
Andrew D. Duckworth, Joseph R. Slupsky, and Nagesh Kalakonda
11 Array-Based Comparative Genomic Hybridization for the Detection
of Copy Number Alterations in Single Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167
Giancarlo Feliciello, Zbigniew Tadeusz Czyz, and Bernhard M. Polzer

ix
x Contents

12 Single Cell Micro RNA Sequencing Library Preparation . . . . . . . . . . . . . . . . . . . . . 189

Sarah M. Hücker and Stefan Kirsch
13 Immunoblot Analysis from Single Cells Using Milo™ Single-Cell
Western Platform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201
Prashant V. Thakkar
14 Imaging of Subcellular Distribution of Platinum in Single Cells
Using Laser Ablation Inductively Coupled Plasma Mass Spectrometry . . . . . . . . 215
Amy J. Managh and Calum J. Greenhalgh
15 Patch-seq: Multimodal Profiling of Single-Cell Morphology,
Electrophysiology, and Gene Expression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227
Cathryn R. Cadwell and Andreas S. Tolias
16 Single-Nucleus ATAC-seq for Mapping Chromatin Accessibility
in Individual Cells of Murine Hearts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
Michail Yekelchyk, Xiang Li, Stefan Guenther, and Thomas Braun

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259
Contributors

BARBARA ALBERTER • Cellular and Molecular Diagnostics Group, Division of Personalized

Cancer Therapy, Fraunhofer Institute of Toxicology and Experimental Medicine ITEM-R,
Regensburg, Germany
BIANCA BEHRENS • Experimental Surgical Oncology, General, Visceral and Pediatric
Surgery, University Hospital and Medical Faculty, Heinrich-Heine-University Düsseldorf,
Düsseldorf, Germany
THOMAS BRAUN • Department of Cardiac Development and Remodelling, Max Planck
Institute for Heart and Lung Research, Bad Nauheim, Germany; German Centre for
Cardiovascular Research (DZHK), Partner site Rhein-Main, Frankfurt am Main,
Germany
JOSKA J. BROEKMAAT • VyCAP B.V, Deventer, The Netherlands
CATHRYN R. CADWELL • Departments of Neurological Surgery and Pathology, School of
Medicine, University of California San Francisco, San Francisco, CA, USA
VERA CAPPELLETTI • Biomarkers Unit, Fondazione IRCCS, Istituto Nazionale dei Tumori di
Milano, Milan, Italy
CLOTILDE COSTA • Roche-Chus Joint Unit, Translational Medical Oncology Group,
Oncomet, Health Research Institute of Santiago de Compostela (IDIS), Santiago de
Compostela, Spain; CIBERONC, Centro de Investigacion Biomédica en Red Cáncer,
Madrid, Spain
ZBIGNIEW TADEUSZ CZYZ • Experimental Medicine and Therapy Research, University
Regensburg, Regensburg, Germany
ANDREW D. DUCKWORTH • Department of Molecular and Clinical Cancer Medicine,
Institute of Systems, Molecular and Integrative Biology, University of Liverpool, Liverpool,
UK
GIANCARLO FELICIELLO • Cellular and Molecular Diagnostics Group, Division of
Personalized Cancer Therapy, Fraunhofer Institute of Toxicology and Experimental
Medicine ITEM-R, Regensburg, Germany
CALUM J. GREENHALGH • Department of Chemistry, Loughborough University,
Loughborough, UK
STEFAN GUENTHER • Department of Cardiac Development and Remodelling, Max Planck
Institute for Heart and Lung Research, Bad Nauheim, Germany
MIODRAG GUŽVIĆ • Department of Urology, University Hospital Regensburg, Regensburg,
Germany
SARAH M. HÜCKER • Fraunhofer Institut für Toxikologie und Experimentelle Medizin,
Abteilung Personalisierte Tumortherapie, Regensburg, Germany
PABLO HURTADO • Roche-Chus Joint Unit, Translational Medical Oncology Group, Gil
Casares Hospital, Oncomet, Health Research Institute of Santiago de Compostela, Santiago
de Compostela, Spain
NAGESH KALAKONDA • Department of Molecular and Clinical Cancer Medicine, Institute of
Systems, Molecular and Integrative Biology, University of Liverpool, Liverpool, UK
STEFAN KIRSCH • Fraunhofer Institut für Toxikologie und Experimentelle Medizin,
Abteilung Personalisierte Tumortherapie, Regensburg, Germany

xi
xii Contributors

JAKUB KOCHAN • Faculty of Biochemistry, Biophysics and Biotechnology, Department of Cell

Biochemistry, Jagiellonian University, Krakow, Poland
CA€ CILIA KÖSTLER • Cellular and Molecular Diagnostics Group, Division of Personalized
Cancer Therapy, Fraunhofer Institute of Toxicology and Experimental Medicine ITEM-R,
Regensburg, Germany
XIANG LI • Department of Cardiac Development and Remodelling, Max Planck Institute for
Heart and Lung Research, Bad Nauheim, Germany
AMY J. MANAGH • Department of Chemistry, Loughborough University, Loughborough, UK
ALEKSANDRA MARKIEWICZ • Laboratory of Translational Oncology, Intercollegiate Faculty of
Biotechnology, University of Gdansk and Medical University of Gdansk, Gdansk, Poland
INÉS MARTÍNEZ-PENA • Roche-Chus Joint Unit, Translational Medical Oncology Group, Gil
Casares Hospital, Oncomet, Health Research Institute of Santiago de Compostela, Santiago
de Compostela, Spain
ROSITA MOTTA • Biomarkers Unit, Fondazione IRCCS, Istituto Nazionale dei Tumori di
Milano, Milan, Italy
LAURA MUINELO-ROMAY • Liquid Biopsy Analysis Unit, Translational Medical Oncology
Group, Health Research Institute of Santiago de Santiago de Compostela (IDIS), Santiago
de Compostela, Spain; CIBERONC, Centro de Investigacion Biomédica en Red Cáncer,
Madrid, Spain
LISA OOMENS • VyCAP B.V, Deventer, The Netherlands
PRAMA PALLAVI • Department of Surgery, Universit€ atsmedizin Mannheim, Medical Faculty
Mannheim, Heidelberg University, Mannheim, Germany
THAIS PEREIRA-VEIGA • Roche-Chus Joint Unit, Translational Medical Oncology Group,
Oncomet, Health Research Institute of Santiago de Compostela (IDIS), Santiago de
Compostela, Spain; Department of Tumor Biology, Center of Experimental Medicine,
University Medical Center Hamburg-Eppendorf, Hamburg, Germany
ROBERTO PIÑEIRO • Roche-Chus Joint Unit, Translational Medical Oncology Group, Gil
Casares Hospital, Oncomet, Health Research Institute of Santiago de Compostela, Santiago
de Compostela, Spain; CIBERONC, Centro de Investigacion Biomédica en Red Cáncer,
Madrid, Spain
BERNHARD POLZER • Cellular and Molecular Diagnostics Group, Division of Personalized
Cancer Therapy, Fraunhofer Institute of Toxicology and Experimental Medicine ITEM-R,
Regensburg, Germany
BERNHARD M. POLZER • Cellular and Molecular Diagnostics Group, Division of Personalized
Cancer Therapy, Fraunhofer Institute of Toxicology and Experimental Medicine ITEM-R,
Regensburg, Germany
CAROLINA REDUZZI • Biomarkers Unit, Fondazione IRCCS, Istituto Nazionale dei Tumori
di Milano, Milan, Italy; Division of Hematology/Oncology, Weill Cornell Medicine, New
York, United States
THOMAS SCHAMBERGER • Experimental Medicine and Therapy Research, University
Regensburg, Regensburg, Germany
STEFAN SCHEUERMANN • Clinical Health Technologies, Fraunhofer Institute for
Manufacturing Engineering and Automation IPA, Mannheim, Germany
MARCO SILVESTRI • Biomarkers Unit, Fondazione IRCCS, Istituto Nazionale dei Tumori di
Milano, Milan, Italy
JOSEPH R. SLUPSKY • Department of Molecular and Clinical Cancer Medicine, Institute of
Systems, Molecular and Integrative Biology, University of Liverpool, Liverpool, UK
MICHIEL STEVENS • VyCAP B.V, Deventer, The Netherlands
 Contributors xiii

NIKOLAS STOECKLEIN • Experimental Surgical Oncology, General, Visceral and Pediatric

Surgery, University Hospital and Medical Faculty, Heinrich-Heine-University Düsseldorf,
Düsseldorf, Germany
TENG TENG • Department of Stem Cell Biology and Regenerative Medicine, University of
Southern California, Los Angeles, CA, USA; USC Norris Comprehensive Cancer Center,
Keck School of Medicine, University of Southern California, Los Angeles, CA, USA
PRASHANT V. THAKKAR • Department of Hematology and Medical Oncology, Weill Cornell
Medicine, New York, NY, USA; Department of Medicine, Weill Cornell Medicine, New
York, NY, USA
ARJAN G. J. TIBBE • VyCAP B.V, Deventer, The Netherlands
ANDREAS S. TOLIAS • Department of Neuroscience, Baylor College of Medicine, Houston, TX,
USA
JUSTYNA TOPA • Laboratory of Translational Oncology, Intercollegiate Faculty of
Biotechnology, University of Gdansk and Medical University of Gdansk, Gdansk, Poland
MARTA VISMARA • Biomarkers Unit, Fondazione IRCCS, Istituto Nazionale dei Tumori di
Milano, Milan, Italy
MATEUSZ WAWRO • Faculty of Biochemistry, Biophysics and Biotechnology, Department of Cell
Biochemistry, Jagiellonian University, Krakow, Poland
MICHAIL YEKELCHYK • Department of Cardiac Development and Remodelling, Max Planck
Institute for Heart and Lung Research, Bad Nauheim, Germany
MIN YU • Department of Stem Cell Biology and Regenerative Medicine, University of
Southern California, Los Angeles, CA, USA; USC Norris Comprehensive Cancer Center,
Keck School of Medicine, University of Southern California, Los Angeles, CA, USA;
Department of Pharmacology, University of Maryland School of Medicine, Baltimore, MD,
USA
ANNA J. ŻACZEK • Laboratory of Translational Oncology, Intercollegiate Faculty of
Biotechnology, University of Gdansk and Medical University of Gdansk, Gdansk, Poland
 Chapter 1

Single Cell Isolation from Surgically Resected Tissue Via

Mechanical Dissociation Using TissueGrinder
Prama Pallavi and Stefan Scheuermann

Abstract
Primary cells form the basis of modern-day in vitro research analysis tools. Many conventional procedures
for generating single-cell suspensions from solid tissue are neither robust nor reproducible. Here we
describe primary cells isolation from surgically resected tumor tissue via enzyme-free mechanical dissocia-
tion using TissueGrinder, a novel semi-automated benchtop device. The isolated cells can be used for any
downstream biochemical or cell-based analytic assay.

Key words Primary cells, Enzyme-free cell isolation, Single-cell isolation, Mechanical dissociation

1 Introduction

Primary cells isolated from tissue are very important in vitro tools to
answer several complex biological questions. In particular primary
cells from surgically resected tissue can provide insight into cellular
biological activity and hence new insights into disease [1]. Methods
to isolate primary cells from tissues can be broadly divided into two
main categories—explant and tissue dissociation techniques using
mechanical force and enzymatic digestion [2–5]. For more than
50 years, the explant method dominated the field of tissue culture
for obtaining primary cells [6, 7]. In fact, this is one of the simplest
techniques: the tissue is finely minced (1–2 mm3) and placed in a
culture flask [8], then the explant is covered with an appropriate cell
outgrowth medium. Within a week’s time, cells migrate out of the
explants and grow on the culture surface [9]. However, success is
limited by various factors such as long processing times, low yield,
and extensive manual workload, leading to a substantial increase in
the overall time from tissue resection to the generation of a cell line.
Furthermore, not all types of cells are able to migrate out of explant
tissue. The second method, dissociation with enzymes, is not
always desired because enzymatic digestion could affect the

Miodrag Gužvić (ed.), Single Cell Analysis: Methods and Protocols, Methods in Molecular Biology, vol. 2752,
https://doi.org/10.1007/978-1-0716-3621-3_1, © Springer Science+Business Media, LLC, part of Springer Nature 2024

1
2 Prama Pallavi and Stefan Scheuermann

proteins that might be later needed for labeling or staining, or are

required for downstream molecular biological analysis [2]. In the
mechanical approach, no additional enzymes or other agents are
used to isolate cells from tissue, which could potentially alter the
cells. During mechanical dissociation, however, the cells are
exposed to a certain level of mechanical stress.
TissueGrinder technology enables the isolation of single cells
from biological tissue of any origin by mechanical dissociation,
minimizing the mechanical stress to preserve native character of
cells to the maximum extent and aims for the utmost biological
comparability with the sample donor. With this technology, a com-
bination of shearing and cutting forces is applied to dissociate cells
from pre-cut tissue [10]. The duration and speed of shearing and
cutting forces can be optimized to develop tissue- and application-
specific dissociation protocols. Manually diced tissue sample is
transferred into grinding tubes, which contain an integrated grind-
ing gear [10]. After dissociation, centrifugation of the grinding
tubes allows easy collection of cells which can be either cultured
directly or used for further analytical tests. By combining the
process steps of dissociation with filtration and centrifugation in a
one-step process, the risk of contamination can be minimized.
Thus, TissueGrinder combines tissue-specific dissociation proto-
cols with gentle mechanical dissociation to provide high-quality
single-cell suspensions without the use of enzymes.
Due to the high demand for primary cells and the limited
availability of tissue samples, the isolation of primary cells should
be as efficient as possible. In this method protocol, we describe a
workflow for the generation of primary cell cultures using surgically
resected tissue obtained from an anal cancer patient via TissueGrin-
der. The method can be applied also to other cancer tissue
types [14].

2 Materials

Clinical tissue specimens are often stored under non-ideal condi-

tions and for long time periods and can therefore be of poor quality
at the time of processing. For this reason, it is important to ensure
cooling of the specimen after resection and to process it promptly
(see Notes).

2.1 Tissue Transport 1. Prepare centrifuge tube with 30 mL of DMEM medium sup-
plemented with 100 units/mL of penicillin and 100 μg/mL of
streptomycin.
2. Prepare a transport container with ice, to transport the tube
with medium. Tissue is collected as quickly as possible with a
cool box from surgery.
 Single-Cell Isolation Using TissueGrinder 3

2.2 Dicing of the 1. 10 cm diameter tissue culture grade petri dish.

Tissue 2. Scalpels.
3. Cell culture hood.
4. Sterile PBS without Ca+2 and Mg+2.
5. Sterile ammonium chloride solution: 0.8% NH4Cl, 0.1 mM
EDTA in water, buffered with KHCO3 to achieve a final pH of
7.2–7.6.
6. Sterile pair of tweezers.

2.3 Tissue Sample 1. Weighing scale with precision of 1 mg.

Processing with the 2. 10 cm diameter tissue culture grade petri dish.
TissueGrinder
3. TissueGrinder disposable sterile grinding tube set.
4. 1 mL of DMEM medium supplemented with 100 units/mL of
penicillin and 100 μg/mL of streptomycin.
5. TissueGrinder device (Fast Forward Discoveries GmbH,
Germany).
6. Benchtop centrifuge with swingout rotor available for 50 mL
centrifuge tube.
7. Dresden medium: mixture of two thirds of the common
DMEM (Sigma-Aldrich) supplemented with 10% FBS
(Gibco) and 1% of penicillin (10,000 units/mL)/streptomycin
(10,000 μg/mL) solution (Gibco) and one third of growth
medium for Keratinocytes (Gibco) supplemented with Bovine
Pituitary Extract (25 mg) and EGF Human Recombinant
(2.5 μg) (Gibco). This medium was established through previ-
ous successes in obtaining cell lines from primary tissue [8].
8. Sterile serological pipettes (10 mL and 5 mL).

2.4 Cell Counting 1. Luna counting slides.

and Viability 2. Luna instrument.
3. Sterile Trypan Blue solution 0.4%.
Cell counting can be also done using Neubauer chamber
and light microscope (see Notes).

2.5 Cell Seeding for 1. Six-well plate tissue culture grade.

Cell Culture 2. Culture medium.
Primary tumor cells isolated from primary tumors are
cultured in “Dresden Medium.”
3. Sterile serological pipettes (10 mL and 5 mL).
4 Prama Pallavi and Stefan Scheuermann

3 Methods

3.1 Tissue Transport 1. Aliquot 20 mL of the DMEM supplemented with 1% of peni-

cillin (10,000 units/mL)/streptomycin (10,000 μg/mL) solu-
tion in 50 mL centrifuge tube and place them in 4 °C fridge
near operation theater. The amount of medium should be
enough to submerge the tissue.
2. Place the surgically resected tumor tissue immediately in the
centrifuge tube containing precooled transport medium (4 °C)
and transport the tissue on ice from operation room to labora-
tory and process immediately (see Notes 1 and 2).

3.2 Pre-Cutting of Note: All these steps should be performed under sterile tissue
the Tissue Sample culture bench (see Notes 3 and 4).
1. Wash the tumor tissue with PBS supplemented with 1% Peni-
cillin/Streptomycin solution (Penicillin (10,000 units/mL)
and Streptomycin (10,000 μg/mL) solution)).
2. If the sample is still blood-soaked, then wash with an ammo-
nium chloride solution to lyse erythrocytes.
3. Pre-cut the tissue into small cubes with an edge length of about
1–2 mm with help of a single-use scalpel.
4. Weigh up to 10–400 mg of the pre-cut tissue pieces in a clean
and sterile petri dish.

3.3 Processing with 1. Take the disposable sterile grinding tube sets for the Tissue-
the TissueGrinder Grinder. Assemble by latching the stator into the cell sieve.
Next, wash stator and cell sieve with 1 mL of Dresden Medium
and transfer the cell sieve to the sterile 50 mL centrifuge tube
(see Note 5).
2. Place 10–400 mg of the prepared tissue pieces in the spare
space between the grinding teeth of the rotor (using a 1 mL
pipette) using a pair of sterile tweezers. Add 500 μl of the
Dresden medium and place the stator with cell sieve on the
rotor and close the tube. The detailed assembly of the grinding
unit is shown in Fig. 1.
3. Place the closed tube on a TissueGrinder benchtop port and
select a predefined tissue dissociation program. There are three
different generic programs—soft, medium, and hard—avail-
able for soft (easy to cut), medium (mildly fibroblastic), and
hard tissues (highly fibroblastic and somewhat calcified). In
addition, pre-installed programs for specific tissue types (such
as liver, spleen, heart muscle, etc.) can be selected or custo-
mized for specific user applications. The generic programs
should be selected and executed based on the texture of the
 Single-Cell Isolation Using TissueGrinder 5

Fig. 1 Assembly of single-use TissueGrinder sterile grinding set (1) and the tissue dissociation workflow
(2–12). The grinding set consists of a fitting lid to the 50 mL centrifuge tube with a centered hole, the rotor and
the stator both forming the grinding gear, a standard cell strainer with a pore size of 100 μm, and a 50 mL
centrifuge tube the assembled grinding set is locked with the rotor’s specific notch to its counterpart on the
TissueGrinder. The diced tumor tissues, each with edge lengths of 1–2 mm and a total weight of 10–400 mg,
are placed in the inner part of the rotor and the unit is assembled as shown under a cell culture bench. Diced
tissue fragments are processed into single cells by alternating processes of grinding and cutting which is
achieved in the TissueGrinder by clockwise or anticlockwise turning of the rotor. After the dissociation
protocol, the tube is transferred directly to a standard laboratory centrifuge and the generated single-cell
suspension is centrifuged through the 100 μm cell filter. The cell pellet is resuspended and transferred to a
new vessel

tissue. A fine-tuning of the program to suit specific require-

ments can also be done. Table 1 describes predefined tissue
dissociation programs.
4. Since anal cancer tissue is soft, run program soft. Following the
program run, centrifuge the tube for 5 min at 350 g (see
Notes/troubleshooting). After centrifugation, open the tube
under laminar flow cabinet. Flush the tissue remains with 1 mL
Dresden Medium through the cell sieve to ensure that all cells
are collected in the 50 mL centrifugal tube. Discard the tissue
fragments, which did not pass through the sieve.
5. Remove the cell sieve with the integrated stator from the tube.
Close the lid of the tube and centrifuge again for 5 min at 350 g
(see Note 6).
6. Resuspend cell pellet with 3 mL of Dresden Medium. Remove
50 μl of the resuspension in 1.5 mL microcentrifuge tube. This
generated single-cell suspension can be used to supply various
downstream applications (see Notes 7 and 8).
6 Prama Pallavi and Stefan Scheuermann

Table 1
Steps of three TissueGrinder programs used for processing of tumor
tissues

Soft Medium Hard

Program
Speed Duration Speed Duration Speed Duration
Steps Process (rpm) (s) (rpm) (s) (rpm) (s)
1 Cutting +30 20 70 30 70 30
2 Grinding -30 20 -70 30 -70 30
3 Cutting +30 20 50 25 80 30
4 Cutting +70 25 -50 25 -80 30
5 Grinding -70 25 50 15
6 Cutting +50 25 -50 15
A positive speed means clockwise turns which lead to a cutting of the tissue while a
negative speed represents anticlockwise turns which causes a grinding of the tissue. This
table is reproduced from [14]. The speed and duration for cutting and grinding can be
set for each of the tissue processing programs

Table 2
Representative cell counts after processing tissues from anal cancer sample

Weight of tissue processed per run

(mg) Living cells isolated per 1 mg of tissue
88 1.96 × 105
97 4.460 × 105

3.4 Cell Counting 1. Take 10 μl of the cell suspension and mix with 10 μl of the
and Viability (see Trypan blue solution.
Notes 9–12) 2. Load 10 μl of this cell solution on to Luna counting slide and
insert it in the Luna instrument to count the cells.
3. Note down the cell number (Table 2) and viability of the
isolated cells.

3.5 Cell Culture 1. Change the medium of the cells in the six well-plate twice a
week until a confluent cell monolayer is observed.
2. Once a monolayer is observed (Fig. 2), sub-culture the cells
into T25- and T75-flasks with passage ratio of 1:3 (see Notes
13 and 14).
 Single-Cell Isolation Using TissueGrinder 7

Fig. 2 A representative image showing anal cancer cells isolated via

TissueGrinder on the tenth day after processing. The images were taken with
4× Zeiss objective (LDA-Plan421231-991). The white scale bar represents
200 μm

4 Notes

1. A low temperature at around 4 °C helps to slow down meta-

bolic activities that can lead to biochemical changes that alter
gene expression, transcription, and the resulting protein pro-
files. This is especially important for the prevention of cell lysis.
2. Freeze/thaw cycles must also be avoided at this stage as cells
can be very sensitive to temperature fluctuations, which can be
very dramatic if the starting sample material is very limited.
3. Wear gloves when processing patient-derived tumor samples
and avoid closing of scalpel to avoid injuries.
4. Surgically resected tissue from patients suffering from an active
HIV-1, or other virologic infectious disease should be pro-
cessed based on the biological safety level of the laboratory.
5. Sterile grinding kits for TissueGrinder are available with differ-
ent sieve sizes. A sieve size of 70 μm should be used for cell
isolation, while a size of 100 μm can be used to isolate clumps
of cells which can be grown in Matrigel for microtissue culture.
6. Cells should always be washed in a buffer solution to remove
cell debris, contaminants, and/or medium containing
BSA/FCS. In addition, the centrifugation speed should be
adapted to the cell type. The generally accepted recommenda-
tion for mammalian cells is to centrifuge at 300–900 g for
5 min to avoid cell disruption. Nonetheless, it’s crucial to be
8 Prama Pallavi and Stefan Scheuermann

aware that excessively low centrifugation speeds can result in

the loss of cell pellets and, consequently, sample loss. Thus,
great care should be exercised during the washing steps.
Failed single cell analysis: Check cell viability and evaluate
cell suspension for total cell yield, cell debris, and aggregates.
7. A multitude of applications and potential exploitation in the
field of advanced purification of single-cell suspensions (Fluo-
rescence- or magnetic activated cell sorting (FACS, MACS),
laser capture microdissection (LCM), and microfluidic work-
flows) and single-cell analysis at genomic, transcriptomic, and
proteomic levels are emerging [11].
8. Evaluation of a solid tissue-derived single-cell suspension is an
important step prior to using cells for, e.g., flow cytometry. The
three critical parameters to evaluate after mechanical dissocia-
tion of solid tissue are: (i) cell viability, (ii) absence of cell
debris, and (iii) absence of aggregates [12]. The viability of
cells in a single-cell suspension can be assessed using the trypan
blue exclusion assay, in which dead cells incorporate trypan
blue into their cytoplasm, while live cells retain their selectively
permeable membrane and prevent trypan blue from entering
the cytoplasm [13].
9. Cell viability in a single-cell suspension can easily be screened
for using the trypan blue cell viability exclusion assay (Neu-
bauer chamber).
10. Cell debris and cell aggregates can also quickly be evaluated
using light microscopy.
11. In case of low cell yield, use polypropylene tubes and plates to
avoid cell adhesion. Add washing steps.
12. In case of low cell viability, reduce the applied rotational speed
of the TissueGrinder protocol by 50% and pipette gently while
resuspending.
13. The isolated cells should be monitored every day for the first
week to identify contaminated cultures.
14. Medium of the processed cultures should be changed every
3 days.

Acknowledgement

This research work was supported by Wilhelm Müller-Stiftung.

Ethical Statement This study was approved by the local ethics

committee of the University Medical Center Mannheim (2012-
293 N-MA), and all of the donors gave their written informed
consent.
 Single-Cell Isolation Using TissueGrinder 9

References
1. Lee J-K, Liu Z, Sa JK, Shin S, Wang J, 8. Ruckert F, Aust D, Bohme I, Werner K,
Bordyuh M, Cho HJ, Elliott O, Chu T, Choi Brandt A, Diamandis EP, Krautz C, Hering S,
SW, Rosenbloom DIS, Lee I-H, Shin YJ, Kang Saeger HD, Grutzmann R, Pilarsky C (2012)
HJ, Kim D, Kim SY, Sim M-H, Kim J, Lee T, Five primary human pancreatic adenocarci-
Seo YJ, Shin H, Lee M, Kim SH, Kwon Y-J, Oh noma cell lines established by the outgrowth
J-W, Song M, Kim M, Kong D-S, Choi JW, method. J Surg Res 172(1):29–39. https://
Seol HJ, Lee J-I, Kim ST, Park JO, Kim K-M, doi.org/10.1016/j.jss.2011.04.021
Song S-Y, Lee J-W, Kim H-C, Lee JE, Choi 9. Rückert F, Aust D, Böhme I, Werner K,
MG, Seo SW, Shim YM, Zo JI, Jeong BC, Brandt A, Diamandis EP, Krautz C, Hering S,
Yoon Y, Ryu GH, Kim NKD, Bae JS, Park Saeger HD, Grützmann R, Pilarsky C (2012)
W-Y, Lee J, Verhaak RGW, Iavarone A, Lee J, Five primary human pancreatic adenocarci-
Rabadan R, Nam D-H (2018) Pharmacoge- noma cell lines established by the outgrowth
nomic landscape of patient-derived tumor method. J Surg Res 172(1):29–39. https://
cells informs precision oncology therapy. Nat doi.org/10.1016/j.jss.2011.04.021
Genet 50(10):1399–1411. https://doi.org/ 10. Scheuermann S, Sch€afer A, Langejürgen J, Reis
10.1038/s41588-018-0209-6 C (2019) A step towards enzyme-free tissue
2. Mitra A, Mishra L, Li S (2013) Technologies dissociation. Curr Dir Biomed Eng 5(1):
for deriving primary tumor cells for use in per- 545–548. https://doi.org/10.1515/cdbme-
sonalized cancer therapy. Trends Biotechnol 2019-0137
31(6):347–354. https://doi.org/10.1016/j. 11. Hu P, Zhang W, Xin H, Deng G (2016) Single
tibtech.2013.03.006 cell isolation and analysis. Front Cell Dev Biol
3. Bols NC, Lee LEJ (1991) Technology and uses 4:116. https://doi.org/10.3389/fcell.2016.
of cell cultures from the tissues and organs of 00116
bony fish. Cytotechnology 6(3):163–187. 12. Reichard A, Asosingh K (2019) Best practices
https://doi.org/10.1007/BF00624756 for preparing a single cell suspension from solid
4. Gawad C, Koh W, Quake SR (2016) Single-cell tissues for flow cytometry. Cytometry A 95(2):
genome sequencing: current state of the sci- 219–226. https://doi.org/10.1002/cyto.a.
ence. Nat Rev Genet 17(3):175–188. https:// 23690
doi.org/10.1038/nrg.2015.16 13. Tennant JR (1964) Evaluation of the trypan
5. Esparza-López J, Martı́nez-Aguilar JF, Ibarra- blue technique for determination of cell viabil-
Sánchez MJ (2019) Deriving primary cancer ity. Transplantation 2:685–694. https://doi.
cell cultures for personalized therapy. Rev org/10.1097/00007890-196411000-00001
Investig Clin 71(6):369–380. https://doi. 14. Scheuermann S, Lehmann JM, Mohan RR,
org/10.24875/ric.19002832 Reißfelder C, Rückert F, Langejürgen J, Pallavi
6. Fischer A (1925) Tissue culture; studies in P (2021) TissueGrinder, a novel technology for
experimental morphology and general physiol- rapid generation of patient-derived single cell
ogy of tissue cells in vitro suspensions from solid tumors by mechanical
7. Parker RC, Morgan JF (1950) Methods of tissue dissociation. Front Med 9:721639.
tissue culture. Hoeber Medical Division- https://doi.org/10.3389/fmed2022.721639
Harper & Row, New York
 Chapter 2

Circulating Tumor Cell Enrichment and Single-Cell Isolation

Combining the CellSearch® and DEPArray™ Systems
C€acilia Köstler, Bernhard Polzer, and Barbara Alberter

Abstract
The analysis of circulating tumor cells (CTCs) has shown potential for detection of cancer spread,
prognosis, therapeutic target selection, and monitoring of treatment response. CTCs can be obtained
repeatedly by simple blood draws as so-called “liquid biopsy.” Thus, they can serve as a surrogate material
for primary or metastatic tissue biopsies. In addition, isolation of CTCs provides the possibility to investi-
gate those cells which may hold the (molecular) traits responsible for metastatic progression and ultimately
patient death. As such, CTCs represent a target of utmost importance in cancer research and therapy. In this
chapter, we describe a workflow for the enrichment of CTCs with the FDA-cleared CellSearch® system
followed by the isolation of single CTCs using the DEPArray™ technology enabling further molecular
single-cell analyses.

Key words CTC, CellSearch, DEPArray, Single-cell isolation, Dielectrophoresis, Liquid biopsy

1 Introduction

In most cancer patients, CTCs are low in abundance and heteroge-

neous in morphological and phenotypic profiles [1]. This compli-
cates their enrichment and isolation for subsequent
characterization [2]. However, a semi-automated and standardized
workflow composed of CTC enrichment with the CellSearch®
system and isolation of single CTCs with the DEPArray™ platform
produces reliable results and can be applied as a routine
workflow [3].
The CellSearch® system for the detection and enumeration of
CTCs has been approved for diagnostic use in breast [4], prostate
[5], and colorectal cancer [6]. The enrichment of CTCs with this
platform makes use of the epithelial origin of carcinoma cells sur-
rounded by mesenchymal cells of the blood compartment. Briefly,
Epithelial Cell Adhesion Molecule (EpCAM)-expressing cells are
enriched by antibody-coated ferromagnetic beads from a 7.5 mL
blood sample and stained for intracellular cytokeratin proteins

Miodrag Gužvić (ed.), Single Cell Analysis: Methods and Protocols, Methods in Molecular Biology, vol. 2752,
https://doi.org/10.1007/978-1-0716-3621-3_2, © Springer Science+Business Media, LLC, part of Springer Nature 2024

11
12 €cilia Köstler et al.
Ca

8, 18, and 19 and for DAPI to determine the presence of nuclear

DNA (positive identifiers), as well as for CD45 as a marker for
contaminating blood leukocytes (negative identifier). Subse-
quently, the number of putative CTCs is determined automatically
and verified by an experienced operator. Evaluation of CTC num-
bers at both early- and late-stage disease allows assessment of
patient prognosis and is predictive of progression-free survival and
overall survival [7, 8].
For research purposes, the precious patient sample analyzed by
CellSearch® can be further utilized to isolate CTCs which can then
be subjected to molecular analyses at the single-cell level. The
DEPArray™ technology has proven beneficial as downstream
single-cell isolation procedure. However, a mandatory buffer
change before the CellSearch® sample can be transferred to the
DEPArray™ cartridge is a critical step in terms of cell loss [2].
The DEPArray™ technology exploits the principles of dielec-
trophoresis (DEP) and the inducibility of electric dipoles in cells.
After polarization of cells, they are trapped one by one in so-called
electric “cages” on a chip. At this point, the cells can be detected by
microscopic scanning of the whole sample with different channels
simultaneously (brightfield and four fluorescence channels). Every
event is recorded as an overlay image of the detected object (i.e.,
putative cell) and can be selected for subsequent separation. This is
done by changing the electric field pattern of adjacent cages on the
chip making selected cells move forward to recover them either as
pools of cells or as single cells.
Once CTCs are isolated as single cells, their nucleic acids can be
amplified and a multitude of molecular analyses can be performed,
ranging from panel sequencing, copy number variation analysis to
genome/exome sequencing.

2 Materials

2.1 CellSearch® 1. CELLTRACKS® AUTOPREP® System User Manual.

System (see Fig. 1), 2. CELLTRACKS® AUTOPREP® System for enrichment, stain-
Kits, and Additional ing, and collection of the cells of interest.
Materials Required for
3. MAGNEST® cartridge mounts.
Cell Enrichment
4. Bottles with caps for waste, instrument buffer, de-ionized
2.1.1 CELLTRACKS® water, and cleaning solution.
AUTOPREP® System
5. Kit for the monthly maintenance.
6. Uninterrupted power supply.

2.1.2 CELLTRACKS 1. CELLTRACKS ANALYZER II® User Manual.

ANALYZER II® 2. CELLTRACKS ANALYZER II® for analysis of the enriched
cells.
 CTC Isolation Using DEPArray 13

Fig. 1 CellSearch® System consisting of CellSearch® Autoprep System and CellTracks® AnalyzerII. (© by
Menarini Silicon Biosystems SpA)

3. Two verification cartridges for the CELLTRACKS® System.

4. CELLTRACKS® justification cartridge.
5. Starter kit for lamp justification.
6. Mercury vapor lamp.
7. Uninterrupted power supply.
8. Local printer for reports.
9. Memorex® DVD + R to back-up results.
10. Fiber-free tissues.

2.1.3 CellSearch® The CellSearch® Circulating Tumor Cell Kit (see Fig. 2) is used for
Circulating Tumor Cell Kit capturing CTCs of epithelial origin (CD45-, EpCAM+, and cyto-
keratins 8+, 18+, and/or 19+). It contains buffers and reagents for
16 enrichments. Up to its use it has to be stored at 2–8 °C (see
Notes 1 and 2). Once it has been opened, the kit can be used for up
to 30 days for diagnostic purposes (see Note 3).
The kit contains the following reagents:
1. 3.0 mL Anti-EpCAM Ferrofluid (brown cap):
Suspension of 0.022% magnetic particles conjugated to a
mouse monoclonal antibody specific for the cell surface marker
EpCAM present on epithelial cells in a buffer with 0.03%
bovine serum albumin (BSA) and 0.05% ProClin®
300 preservative.
2. 3.0 mL Staining Reagent (white cap):
0.0006% mouse monoclonal antibodies specific to cytoker-
atins conjugated to phycoerythrin (PE) and 0.0012% mouse
14 €cilia Köstler et al.
Ca

Fig. 2 Content of CellSearch® Circulating Tumor Cell (CTC) Kit. (© by Menarini Silicon Biosystems SpA)

anti-CD45 monoclonal antibody conjugated to allophycocya-

nin (APC) in buffer containing 0.5% BSA and 0.1% sodium
azide.
3. 3.0 mL Nucleic Acid Dye (blue cap):
0.005% 4′, 6-diamidino-2-phenylindole, dihydrochloride
(DAPI), and 0.05% ProClin®300.
4. 3.0 mL Capture Enhancement Reagent (clear cap):
0.02% proprietary reagent for controlled ferrofluid aggre-
gation, 0.5% BSA, and 0.1% sodium azide in buffer.
5. 3.0 mL Permeabilization Reagent (green cap):
0.011% proprietary permeabilization reagent and 0.1%
sodium azide in buffer.
6. 3.0 mL Cell Fixative (red cap):
25% proprietary fixative ingredients, 0.1% BSA, and 0.1%
sodium azide in buffer fixes cells for identification and
enumeration.
7. 2 × 110 mL Dilution Buffer:
A buffer with 0.1% sodium azide is used for dilution. It is
the only reagent of the kit, which should be stored at room
temperature after opening the first time (see Note 2).
8. 16 CellSearch®Conical Centrifuge Tubes (15 mL) and Conical
Tube Caps.
9. 16 CellSearch® Cartridges and Cartridge Plugs.

2.1.4 CellSearch® The CellSearch® Circulating Tumor Cell Control Kit is used for
Circulating Tumor Cell testing system performance and operator technique. It is sufficient
Control Kit for 24 applications and proves that the system detects both high as
well as low numbers of CTCs. It has to be stored at 2–8 °C and
contains:
 CTC Isolation Using DEPArray 15

1. 24 single-use vials of fixed cells from breast carcinoma cell line

(see Note 4).
2. 24 lot-specific barcodes (see Note 5).

2.1.5 Additional 1. CellSave Preservative Tubes for collecting 7.5–10 mL of sam-

Materials ple (see Note 6).
2. CELLTRACKS® AUTOPREP® Instrument Buffer for the
enrichment and staining procedure.
3. De-ionized water.
4. 0.26% sodium hypochlorite cleaning solution (see Note 7).
5. Horizontal swing out style rotor (i.e., swing bucket) centrifuge
capable of 800 × g and break of “0.”
6. Test tube racks.
7. Calibrated micro-pipettors and tips.
8. Vortex mixer.

2.2 Sample 1. CellSearch® Cartridge with enriched and stained cell

Preparation* for suspension.
(Single)-Cell Isolation 2. BSA lyophilized powder, essentially globulin and protease free,
with the DEPArray™ ≥ 98%.
Platform 3. 1 ×PBS.
4. Buffer for the DEPArray™ system (DABUF).
5. Protein gel loading tips round convenient for 200 μL-
micropipettes.
6. 1.5 mL Protein LoBind tubes.
7. 50 mL screw cap tubes.
8. Racks for 1.5 mL and 50 mL tubes.
9. Micropipettes (10, 20, 200, 1000 μL) and related LoRetention
filter tips (see Note 8).
10. Swinging bucket rotor centrifuge with adapters (see Note 9)
for 1.5 mL tubes.
11. Laminar flow hood.
*Alternatively, the VR volume reduction device (Menarini
Silicon Biosystems) can be used following the instructions of
the user manual to reduce the volume prior to loading the
DEPArray™ cartridge.

2.3 (Single)-Cell 1. DEPArray™ Plus for CTC Application Guide and DEPArray™
Isolation with the Instruction.
DEPArray™ Platform 2. DEPArray™ system (see Fig. 3).
16 €cilia Köstler et al.
Ca

Fig. 3 DEPArray™ Plus System. (© by Menarini Silicon Biosystems SpA)

3. DEPArray™ cartridges.
4. Buffer for the DEPArray™ system (DABUF).
5. 1.5 mL Protein LoBind tubes.
6. MicroAmp Reaction Tubes with cap 0.2 mL.
7. Micropipettes 5000 μL, 20 μL and related LoRetention filter
tips (see Note 8).
8. Ultrasonic bath with a pulse button.
9. Racks for 1.5 mL and 0.2 mL tubes.
10. Laminar flow hood.
11. Waterproof pen for the unique labeling of the tubes.

2.4 Volume 1. 0.2 mL MicroAmp Reaction tubes with the DEPArray™

Reduction of isolated cells inside.
DEPArray™ Isolated 2. 1 × PBS.
Cells*
3. 200 μL micropipette and related LoRetention filter tips (see
Note 8).
4. Fixed rotor centrifuge with adapters for 0.2 mL tubes.
5. Racks for 0.2 mL tubes.
6. Freezer for storage.
*Alternatively, the VR volume reduction device (Menarini
Silicon Biosystems) can be used following the instructions of
the user manual to reduce the volume after the isolation of the
cells.
 CTC Isolation Using DEPArray 17

3 Methods

3.1 CTC Enrichment Before starting a run, the system must be cleaned (see Note 10).
with the CELLTRACKS® Eight samples can be processed in parallel. Make sure that the
AUTOPREP® System samples fulfill all requirements for the run (see Note 11). The
necessity for a control sample arises whenever patient samples are
processed or when using a new lot number of kit reagents (see Note
12). Make sure to wear the required protective clothes and gloves
for all following steps.
1. Start the CellTracks® Autoprep® System.
2. Bring the CellSearch® Circulating Tumor Cell Kit and one
bottle of the CellSearch® Circulating Tumor Cell Control Kit
to room temperature (see Note 1).
3. Select “Run batch” and enter the password (see Note 13).
4. Make sure, that the system is clean. If necessary, perform the
“Daily Cleaning Procedure” (see Note 10), otherwise go on
with step 4.
5. Take as many CellSearch®Conical Centrifuge Tubes and Caps as
there are samples (including positive control if necessary). Put a
barcode label (if available, see Note 14) on the CellSearch®Co-
nical Centrifuge Tubes for the samples (see Note 15).
6. Mix the CellSave tube by inverting it five times (see Note 16).
7. Remove the cap of the CellSave tube and transfer 7.5 mL of the
blood sample into a CellSearch® Conical Centrifuge Tube. Add
6.5 mL dilution buffer, close it with a cap and mix gently by
inverting five times (see Note 16).
8. Repeat steps 6 and 7 for further samples.
9. Centrifuge the tubes with a horizontal swing-out style rotor at
800 g for 10 min at room temperature. Attention: Set break to
“0” (see Note 17).
10. Prepare the positive control—if necessary—during the centri-
fugation step, otherwise switch to step 13.
11. Put an orange ID label onto the conical tube with the positive
control (see Note 5).
12. Mix the 3 mL control tube gently by vortexing 5 s and invert-
ing the bottle five times (see Note 18). Open the cap and
transfer the complete volume—also from the cap—to a conical
centrifuge tube (see Notes 19 and 20).
13. Open all bottles of the CellSearch® Circulating Tumor Cell Kit
and inspect them (see Note 21).
14. Follow instructions on the screen for choosing the kit, select-
ing a marker and control.
15. After choosing the number of samples, including the control, the
system is “homing,” which means being prepared for the run.
18 €cilia Köstler et al.
Ca

16. Place the rack with the CellSearch® Circulating Tumor Cell Kit
into the CellTracks®AutoPrep® System (see Note 22) and
confirm the kit data with “Next.”
17. Confirm, if results shall be used for patient management (see
Note 23) or not. (If the kit is not valid anymore for in vitro
diagnostics, see Note 3).
18. Put an empty CellSearch® cartridge (without lid) into the
magnetic rack (MAGNEST®) and load it to the system (see
Notes 24 and 25).
19. Repeat step 18 for all remaining MAGNEST®s.
20. Go on with loading the samples and/or control (see Note 26).
If a control is included in the run, follow the loading order in
analogy to the order of the MAGNEST®s which were loaded in
step 18 (see Note 24).
21. Close the door and start the system with “Start.” During the
run, the estimated duration is indicated (see Note 27).
22. When the enrichment is finished, remove the MAGNEST®s
from the instrument. Carefully inspect the content of the
CellSearch® cartridge (see Note 28).
23. Store the MAGNEST®s with the filled and closed CellSearch®
cartridge at room temperature at a dark place in a horizontal
position and analyze them with the CELLTRACKS ANA-
LYZER II® system within the next 24 h (see Note 29). This
is described in Subheading 3.2.
24. Follow the instructions on the screen for removing the empty
sample/control tubes and inspect them (see Note 30). Remove
the kit and close the reagents (see Note 31).
25. Go on with the daily cleaning procedure if no further runs are
planned.

3.2 CTC Counting Keep in mind that enriched samples (also controls) can be scanned
with the CELLTRACKS at the earliest after 20 min of incubation (in a horizontal
ANALYZER II® System position in the dark at room temperature).
1. Switch on the device, computer, and the mercury vapor
lamp (see Note 32). The CELLTRACKS® software starts
automatically.
2. Sign in with the unique password.
3. If it is necessary to perform a system verification (see Note
33), click the “QC test” folder symbol, clean the system
verification cartridge with a fiber-free wipe, open the sam-
ple door, and load it to the sample rack. Push the “Start”
button in the field “System verification” (see Note 34).
After successful verification reload the verification
cartridge.
 CTC Isolation Using DEPArray 19

4. Check if there are ferrofluid aggregates in the first Cell-

Search® sample cartridge, but leave the cartridge in the
MAGNEST® (see Note 28).
5. Scan enriched patient samples by clicking to the “Patient
Test” folder. For scanning the CellSearch® control car-
tridge go to step 23.
6. Clean the cover of the CellSearch® sample cartridge with a
fiber-free wipe and load the MAGNEST® with the sample
to the sample rack.
7. Check if the sample and kit information are correct (see
Note 35).
8. Start scanning. When the edges of the CellSearch® car-
tridge are found, confirm by clicking “Accept” (see Note
36).
9. The system will start to focus automatically (see Note 37).
10. When focusing is finished the picture recording starts
automatically (see Note 38).
11. After the scan is finished, click “Ok” (see Note 39), open
the door, and remove the MAGNEST® with the Cell-
Search® cartridge (see Note 40).
12. Do not repeat scanning of a CellSearch® sample cartridge
more than once (see Note 29).
13. Scan all other CellSearch® sample cartridges by repeating
steps 4–12 or go on with their evaluation (step 14). For
analyzing CellSearch® control cartridges go to step 25.
14. For each sample scan, the system creates an auto-analysis of
the pictures and shows an image gallery of cells, which
have to be controlled by the user (see Note 41).
15. For evaluation, select the patient sample of interest in the
software (in “Patient data” ! “Sample table”) and review
the cells under the official guidelines of cell interpretation
(see Note 42 and refer to the user manual).
16. The auto-analysis generates an image of each event with
every filter and defined filter combinations (see Note 43).
17. Select each event in the DAPI/CK-PE channel (see Note
44), which fulfills the criteria for CTCs (see Note 42 and
user manual). The image is now framed orange and the
system counts this event as a CTC.
18. Make sure that all images of the gallery have been
reviewed, otherwise evaluation cannot be finished (see
Note 45).
19. The status of the sample changes to “Complete.” The
number of assigned and unassigned cells is shown.
20 €cilia Köstler et al.
Ca

20. If the quality of the scan is insufficient, the result is

denoted “Done—No Result.”
21. For printing or saving the data, click “Patient Data” !
“Sample Table,” select a sample, and click “Report”.
When the review of this sample is finished, a pdf document
can be printed or exported.
22. Save the data from time to time to a separate storage
location, e.g., by burning a DVD (see Notes 46 and 47).
23. Scan the enriched control by clicking “QC test.”
24. Proceed with steps 6–11 and treat the CellSearch® control
cartridge just like the CellSearch® sample cartridges.
25. The system creates an auto-analysis with a pre-count of
most of the “high” control cells. Nevertheless, the evalua-
tion of some of the “high” and all of the “low” control
cells (shown automatically at “QC Data”) has to be fin-
ished manually by the user to classify the remaining, not
automatically dedicated cells either to the high or the low
row (see Note 48).
26. If the number of the “high” and “low” cells are in the
correct range (see Note 4), the in vitro diagnostic (IVD)
control has “passed.”
27. The CellSearch® control cartridge can be discarded and
the system switched off.
28. If there is no sample or QC to report, the computer can be
shut down.

3.3 Sample After enrichment of CTCs with the CellSearch® system, the sample
Preparation for Single- needs to be subjected to a buffer change before applying it to
Cell Isolation with the the DEPArray™ cartridge (see Figs. 4 and 6) for automatic
DEPArray™ System isolation of single CTCs or blood leukocytes or pools thereof.
All following working steps should be performed under a lami-
nar flow hood wearing the required protective clothes and
gloves.
1. Thaw a vial of DABUF.
2. Prepare a 2% BSA in PBS solution (see Note 49).
3. Prepare.
– two aliquots of 325 μL DABUF.
– one labeled 1.5 mL Protein LoBind tube for the
sample.
– 1.5 mL 2% BSA/PBS solution (see step 2).
– one clean 1.5 mL tube for storing the tip.
 CTC Isolation Using DEPArray 21

Fig. 4 Backside of the DEPArray™ cartridge with buffer and sample inlet. Picture
adapted from the DepArray™ User Manual. (© by Menarini Silicon Biosystems
SpA)

4. Remove the plug of the CellSearch® cartridge (see Note

50) which contains the enriched and stained cell
suspension.
5. Set the micropipette to 200 μL and coat one protein gel
loading tip with the 2% BSA/PBS solution (see Note 51).
6. Use this tip to resuspend the sample by pipetting the
sample against all sides and the bottom of the CellSearch®
cartridge. Transfer first 200 μL and then the rest (~125 μL)
of the sample to the prepared and labeled 1.5 mL Protein
LoBind tube and keep this tip for later use (see Note 52).
7. Use a new protein gel-loading tip for transferring the first
aliquot of the 325 μL of DABUF into the empty Cell-
Search® cartridge in two steps. Throw away this tip.
8. Use the pre-coated tip from step 6 for washing the Cell-
Search® cartridge with the DABUF by pipetting the buffer
against all sides and the bottom. Transfer the suspension in
two steps into the 1.5 mL tube, which already contains
sample (step 6) without resuspending. Keep this tip for
later use (see Note 52).
9. Repeat steps 7 and 8 with the second 325 μL aliquot of
DABUF.
10. Centrifuge the labeled sample tube with a swinging bucket
rotor and the 1.5 mL tube adapters (see Note 9) with
1000 g for 5 min at room temperature.
11. Put the tube into the rack under the laminar flow hood,
open it carefully to avoid swirling up the cells, and remove
the supernatant (see Note 53) without touching the
22 €cilia Köstler et al.
Ca

bottom of the tube with the 1000 μL micropipette. Leave

a volume of approx. 30 μL in the tube.
12. Add 1 mL DABUF without resuspending. Close the tube
and flip with the finger against the wall to wash the pellet
gently.
13. Repeat step 10.
14. *Carefully open the tube under the laminar flow hood.
Remove the supernatant (see Note 53) without touching
the bottom of the tube with the 1000 μL micropipette,
then with the 200 μL micropipette, and finally with the
20 μL micropipette (see Note 54) and leave around 10 μL
in the tube.
15. Adjust the volume to 12 μL (see Note 55) without touch-
ing the cell pellet. Do not throw away the 20 μL tip
(compare its storage with Note 52), keep it for loading
the DEPArray™ cartridge.
16. Go on directly with loading the DEPArray™ cartridge (see
Subheading 3.4).
*As an alternative for manual volume reduction (steps
14 and 15), the VR volume reduction device (Menarini
Silicon Biosystems) can be used following the instructions
of the user manual to reduce the volume prior to loading
the DEPArray™ cartridge.

3.4 Cell Isolation All steps concerning the loading of the DEPArray™ cartridge have
with the DEPArray™ to be performed under a laminar flow hood wearing the required
System protective clothes and gloves.
1. Take the DABUF bottle and incubate them for 10 min in the
ultrasonic bath with the degas function (see Note 56).
2. Start the DEPArray™ system. Select the username in the drop-
down menu to sign in. Enter the password by touching the
“Code” button (see Note 57) and login.
3. Enable the VPN connection in order to have full service for
troubleshooting.
4. Touch the “Sorting,” “CTC,” “UDP fixed CTC” buttons on
the touch screen (see Note 58).
5. Go back to the laminar flow hood with this DEPArray™ car-
tridge. Open cover “1” (see Notes 59 and 60), remove the blue
protective tape, and leave the DEPArray™ cartridge inside the
laminar flow hood for loading.
6. Pipette 2.5 mL of the degassed and solved buffer (step 1) to
the buffer reservoir “B” (see Fig. 4) of the DEPArray™ car-
tridge by using the 5000 μL micropipette (see Note 61).
 CTC Isolation Using DEPArray 23

7. Check at the back of the DEPArray™ cartridge, if there are air

bubbles inside the buffer channel. Also make sure that the
meniscus of the buffer has not entered the valve of the main
chamber (see Note 62).
8. If there are no bubbles in the buffer channel, go on with
pipetting (see Note 63) the 12 μL sample (prepared in Sub-
heading 3.3) into the sample channel “S” (see Fig. 4) of the
DEPArray™ cartridge.
9. Check at the back of the DEPArray™ cartridge, if there are air
bubbles inside the sample channel and make sure that the
meniscus is in the correct area (see Note 64). Close the protec-
tive cover of the DEPArray™ cartridge and go back to the
DEPArray™ system for loading the DEPArray™ cartridge.
10. For opening the door of the system, touch “Open” (see Note
65).
11. Open cover “2” (see Note 59) of the DEPArray™ cartridge
protection cover and insert DEPArray™ cartridge into the
correct position of the tray (see Note 66).
12. Close the door of the system by pushing it gently. The car-
tridge tray is automatically pulled back, the door is locked, and
the system recognizes the cartridge ID.
13. Scan the barcode of the buffer. This step can also be skipped by
touching “Skip.”
14. Fill in the run code (name of experiment) by touching “Run
Code” and using the keyboard on the touchscreen of the
machine.
15. Touch “Start Process” to start the sorting execution.
16. The sample loading phase (see Note 67) is starting
automatically.
17. After the sample loading is finished, the sample scan (see Note
68) starts automatically.
18. For optimizing the scan settings, push the “Pause” field on the
right side of the monitor as early as possible to pause the scan.
19. Click several frames (see Note 69) in the DAPI channel for
finding cells and check the filter setting parameters for all
fluorescent channels (see Note 70). Once adapted correctly
push the “Play” button again. The system asks for confirmation
of the changes in chip scan settings. By clicking “Yes” the
system restarts scanning all frames under the new conditions.
If no re-scan shall be done under the new conditions, click
“No” and the system will go on with scanning from the posi-
tion where it had stopped before. The scan of the cartridge
takes around 30 min.
24 €cilia Köstler et al.
Ca

Fig. 5 Populations and Groups panels of the CellBrowser™ with functional buttons. Picture adapted from the
DepArray™ User Manual. (© by Menarini Silicon Biosystems SpA)

20. When the scan is finished, the cell selection workflow can be
started using the “Cell Selection Software” (see Note 71).
21. An automatic selection of cells (see Note 72) is shown in the
“Populations” panel (see Fig. 5).
22. For an overview of the cell list and the plot view in parallel, click
the “Grid and Data Visualization” icon.
23. Select the “Routable Cells” (see Note 73) by clicking on it
(they will be highlighted in blue) and click on the “Scatter
Plot” icon in the “Data Visualization” panel.
24. Select “Mean Intensity PE” for X-axis and “Mean intensity
APC” for Y-axis. Click the “Cross Selection” button to split
the “Routable Cells” selected for PE and APC into the four
subpopulations and move the lines for optimization (see Note
74).
25. Select the PE+/APC- subpopulation (“Q3”) and click the
“Create Table” icon (see Fig. 5) above the cell populations in
the “Populations” panel. A new table, called “[0]Table0 (0),”
is created (see Note 75) and will be filled in with the cells of
interest to be isolated.
26. Sort the cells in the “Grid View” list from highest to lowest
mean intensity of PE by clicking twice in the blue “Signal
Intensity PE” field (see Note 76) on the top of the list.
27. By clicking to the first cell in the “Grid View” list (highlighted
in blue), the fluorescent images of the selected cell are shown
below. Judge the cell by the official cell selection parameters (see
 CTC Isolation Using DEPArray 25

Note 77). Shift putative CTCs from the Q3 subpopulation to

the table “[0]Table0 (0)” (see Note 78) by typing the number
in squared brackets, i.e., 0 for “[0].” Proceed with all cells in
the “Grid View” and shift the putative CTCs to the respective
table.
28. Once all putative CTCs are selected, analogously repeat steps
25–27 for the selection of the putative leukocytes (APC+/PE-
subpopulation = Q1, create a new table for this subpopula-
tion) or other cell populations if needed.
29. In order to route the selected cells, the tables must be assigned
to groups in the “Groups” panel. For this, first highlight the
table containing the cells selected for routing, e.g., “[0]Table0
(XY)” in the “Populations” panel and then “Group1” in the
“Group” panel. For correlating both click the blue “Assign”
icon (see Fig. 5) between the “Populations” and the “Group”
panel. Repeat this for all tables containing cells to route (see
Note 79). The groups are automatically saved in the “Cell
Panel” (see Note 80).
30. Click the “Cell Routing” icon (top right) at the CellBrowser™
to start the “Recovery Manager™” (see Note 81).
31. Open the shutter in the “Live Configuration” section on the
lower left side. The DEPArray™ cartridge chamber becomes
visible. The squared frame in the main chamber view (top
right) can be moved for checking, if air bubbles are inside the
DEPArray™ cartridge chamber (see Note 82).
32. Highlight the first group to be routed to the “Park” section by
left-click. The group is marked to be parked in the parking
chamber by right-clicking and selection of “Move To Rout-
ing.” Repeat this for all groups to be parked and start “Park
Routing” by clicking the play button (see Note 83).
33. When the “Park Routing” is done a green check and the
comment “Park done” in the “Park” section notify the user
about successful parking (see Note 84).
34. For defining the recovery scheme, select the appropriate
“Recovery Support” in the software. For the “CTC-RUO
Fixed” application, the “200 μL-tube-rack” is preset and
recommended for maximum purity of the samples (see Note
85).
35. Select position “A1” in the “Recovery Support” layout by
right-click and set it as the priming recovery position (see
Note 86). To recover the putative CTCs as cell groups
(in order to obtain pools of cells), click the cell group name.
To recover single cells, click the cell ID. Drag selected group or
cell ID to the selected position of the “Recovery Support”
layout (see Note 87).
26 €cilia Köstler et al.
Ca

36. If more than one cell type is to be recovered (e.g., putative

CTCs and putative leukocytes), a blank recovery in between
each cell population should be defined (see Note 88).
37. To recover further cell types (e.g., putative leukocytes) as cell
groups or single cells, proceed in analogy to the CTC recovery
using the drag and drop function (see Note 87).
38. Prepare the necessary number of clean 200 μL tubes (see Note
89), label them with a unique name (corresponding to the
experiment), and put each tube to the respective position
selected in step 37/39 (see Note 90).
39. For insertion of the loaded “Recovery Support” into the
machine, click the “Play” button in the “Exit and Recovery”
section. Select the appropriate support “200 μL-tube-rack”
(or the support chosen in step 38, see also Note 85) in the
drop-down menu and click “Add.”
40. Switch to the DEPArray™ touch screen and touch “Trays” and
“Open Recovery” to eject the recovery tray.
41. Open the main door and insert the “Recovery Support” with
the opened 200 μL tubes (labeled) in the correct position
(consider the marks on the tray) and close the main door of
the DEPArray™ system.
42. Touch “Close Recovery” and switch to the recovery software
on the screen. A check box icon is shown, when the “Recovery
Support” is inserted correctly and has to be confirmed with the
“Ok” button.
43. After confirmation of the pop-up window, the initial washing
step starts (see Note 91). After the washing step, the system
automatically starts with the “Priming Recovery” and the
recovery of the cells (see Fig. 6).

Fig. 6 Front side of the DEPArray™ cartridge with the three-chamber system as the heart of the cartridge.
(© by Menarini Silicon Biosystems SpA)
 CTC Isolation Using DEPArray 27

44. When the recovery of a cell group or a single cell is completed,

a green check appears in the “Exit and Recover” field.
45. When recovery of all cells is finished, the samples can be
removed. Push the “Stop” button on the touch screen of the
DEPArray™ system and confirm with “Ok.”
46. Sign in with username and password (see Note 92).
47. Touch “Trays” at the touch screen of the DEPArray™ to open
the main door for removing the “Recovery Support” with the
samples and check if the tubes are filled.
48. Close the caps of the 200 μL PCR tubes.
49. Close the main door by touching “Close Recovery” on the
touch screen of the device.
50. Touch “Eject Cartridge” to open the DEPArray™ cartridge
tray and remove and throw away the used DEPArray™ car-
tridge (see Note 93). Close and lock the main door by touching
“Insert Cartridge.”
51. Touch “Back,” “Shutdown,” and “Ok” for shutting down the
device.
52. Before the system shuts completely down, a backup is per-
formed automatically (see Note 94). Go on with step 53 if
no further experiment is planned for this day or stop the
backup process, if another experiment follows (see Note 95).
53. Switch off the machine at the end of the backup process (when
the touchscreen is black and does not respond by touching it).

3.5 Preparation of Single cells or cell pools are recovered from the DEPArray™ system
Isolated Cells for in a cell number-specific volume (see Note 96). In order to proceed
Molecular Analyses with whole genome amplification (WGA) followed by molecular
analyses, the volume of the recovered single cells or cell pools needs
to be reduced to approx. 1 μL. Volume reduction of the isolated
samples has to be performed under a laminar flow hood wearing the
required protective clothes and gloves.
1. Spin down the 200 μL tubes at room temperature and
14,100 g at room temperature in a fixed rotor centrifuge for
200 μL tubes. The time depends on the number of cells in the
tubes (see Note 97).
2. Proceed with the following steps under a laminar flow hood.
3. If 21–85 cells are recovered in one tube (compare to Note 96),
carefully remove 50 μL of supernatant. Do not touch the
bottom of the tube. Instead, follow the meniscus of the
solution.
4. Add 100 μL of 1× PBS to all tubes (regardless of prior 50 μL
removal of the supernatant) without touching the solution.
28 €cilia Köstler et al.
Ca

5. *Spin down the samples at room temperature with 14,100 g

for 10 min at room temperature in a fixed rotor centrifuge and
consider the tube cap direction (compare to Note 97).
6. For reducing the volume to approx. 1 μL, go back under the
laminar flow hood, open the tube in the rack, and hold it
horizontally.
7. Be sure to stabilize arms by resting the elbows on the bottom of
the hood.
8. Aspirate the supernatant carefully and slowly by following the
meniscus and not touching the bottom of the tube until
approx. 1 μL is left. Use a 200 μL micropipette which is set to
150 μL. Use a fresh tip for each tube (see Note 98).
9. The sample is now ready for whole genome/transcriptome
amplification and subsequent molecular analyses.
10. Alternatively, the sample can be stored at -20 °C after volume
reduction.
*As an alternative for manual volume reduction (steps
5–8), the VR volume reduction device (Menarini Silicon Bio-
systems) can be used following the instructions of the user
manual to reduce the volume after the isolation of the cells.

4 Notes

1. The kit has to reach room temperature before it is used. Ideally,

the kit is transferred to room temperature approx. 30 min
before sample processing is started. To avoid evaporation,
remove the caps just directly before the run starts.
2. Once the dilution buffer has reached room temperature at first
opening, further storing has to be done at room temperature
and the buffer is stable for 4 weeks. Cooling of the buffer can
influence the sample and associated enrichment quality. The
dilution buffer is part of the CellSearch® Circulating Tumor
Cell Kit.
3. The CellTracks® Autoprep® System saves the first opening date
of the kit by scanning the kit barcode. To guarantee stability
under FDA guidelines, the kit can be used only 30 days for
patient management, thus diagnostic purposes. For research/
nondiagnostic experiments, it is possible to use the kit until the
date of expiry. The system has a safety warning message
installed, which precludes enrichment of important diagnostic
samples with a kit open for more than 30 days.
4. Each vial of the CellSearch® Circulating Tumor Cell Control
Kit contains two differently labelled cell types. A high number
of CK-PE+/DAPI+/FITC+/APC- cells (mean ~ 1000 cells;
 CTC Isolation Using DEPArray 29

“high” cells) is detected by the FITC-channel and a low num-

ber of CK-PE+/DAPI+/APC+/FITC- cells (mean ~ 50 cells;
“low” cells) is detected by the APC-channel. The exact range of
the cell numbers is indicated on the label of the control kit box
and can vary between different lot numbers. The number of
cells, detected in the control cartridge should fall into this
range, otherwise the control is regarded to have failed.
5. The orange barcode label included in the control kit contains
the lot-specific information (lot number, date of expiry, mean
cell number of “high,” and “low” cell type) and informs the
system, to skip the plasma aspiration, as it is done for the
patient samples.
6. Once the blood is collected in the CellSave Preservative Tube,
the proprietary preservative stabilizes the cells for up to 96 h.
Store the blood samples at room temperature until the run
starts and inspect the sample for clots before it is processed.
Wrong storage of the sample can have an influence on the
quality of the enrichment with the sensitive ferroparticles.
7. The bleacher Clorox® is suggested for the cleaning procedure.
Alternatively, it is also possible to prepare a 0.26% sodium
hypochlorite solution in a final volume of 4 L.
8. It is possible to use other pipettes with LoRetention filter tips
than the suggested ones from Eppendorf. However, according
to our experience, this combination fulfills the best require-
ments for an uncomplicated bubble-free run. Alternative tips
can be used as suggested from Menarini Silicon Biosystems.
9. The adapters for the 1.5 mL tubes are 50 mL screw cap tubes
with a hole, centered in the screw caps, fitting for 1.5 mL tubes.
They can also be self-made, if the balance of the centrifuge is
guaranteed.
10. The daily cleaning procedure should be performed:
(a) at the end of each day after sample processing.
(b) at least every 72 h.
(c) if a sample with more than 5000 CTCs was included in the
last batch.
Start the system, choose “Daily cleaning,” and enter the
password. Follow the instructions of the system for disconnect-
ing the bottle with the AutoPrep® Instrument buffer and for
connecting the bottles with de-ionized water, cleaning solution
and waste. Push “Enter” and the cleaning procedure starts. It
takes around 1 h. The system shows a reminder every 72 h to
perform the daily cleaning procedure. Follow the instructions
after finishing the cleaning procedure and push “Enter” to go
back to the main menu. For more information about cleaning,
30 €cilia Köstler et al.
Ca

weekly and monthly maintenance, please refer to the user

manual.
11. Only process samples which meet the following requirements:
(a) The sample was stored at room temperature in the Cell-
Save tube no longer than 96 h.
(b) The sample has volume of at least 7.5 mL.
(c) The sample is not coagulated.
12. Not every run has to include a control sample. But it is neces-
sary to run a control sample.
(a) once a day when patient samples are processed.
(b) when a reagent kit with a new lot number is used.
Keep in mind, that only seven more patient samples can be
processed at a time if a control is included.
If the system realizes a new lot number of the CellSearch®
Circulating Tumor Cell Control Kit, the barcode of the pack-
age has to be scanned to inform the system about the
lot-specific informations (see Note 5). It is possible to run
more than one control in one run.
13. Make sure that the de-ionized water bottle is disconnected, the
bottle with AutoPrep® Instrument buffer is filled and
connected, the waste bottle had been emptied, filled with
400 mL undiluted sodium hypochlorite solution and
connected.
14. There are no barcode labels included in the kit for the samples.
If it is necessary to code the sample tubes digitally, the Cell-
Tracks® AutoPrep® System can decode barcodes with the
CLSI-Standard Auto02-A2. The label should be no longer
than 50 mm. Fix the barcode sticker in the upper part of the
conical tube above the sample level. This ensures that the blood
sample or the control solution are not covered.
15. If there are no barcode stickers available, mark the sample tube
above the sample level with a unique sample name. At a later
stage of processing, it is possible to enter the sample name to
the system (see Note 26).
16. Invert the tube in order to obtain a homogeneous sample. This
is important for further processing.
17. The time to reach 800 g before and 0 g after centrifugation is
not included in the 10 min of centrifugation. Since the break is
set to “0,” it will take some minutes longer until the centrifuge
has finished.
18. For an improved and more homogeneous suspension of the
control cells, it is advisable to open the bottle and pipette five
times up and down with the 1000 μL micropipette and a clean
filter tip.
 CTC Isolation Using DEPArray 31

19. Tip over the empty and open control bottle and collect remain-
ing drops in the CellSearch®Conical Centrifuge tube for the
control.
20. There is no need for adding dilution buffer and centrifuging
the control tube.
21. Each bottle of the tray should be fixed properly by pushing
them down. To avoid evaporation, remove the caps directly
before the run. An inspection of the solutions in the bottles for
air bubbles is necessary before each run. A clean micropipette
tip can be used to remove them.
22. This is the last possibility to ensure that the reagents are at
room temperature. The system reports if the number of sam-
ples to be processed does not match the number of remaining
samples which can be processed with this kit.
23. Samples which are used for the patient management are always
written in black, on each and every
CELLTRACKS®AUTOPREP® System.
24. Every stained and enriched sample is collected in a single-use
CellSearch® cartridge, which is entered in a magnetic rack, the
so called MAGNEST®. There is no rule specifying the order of
loading samples and controls. However, in order to prevent
confusion, the gray MAGNEST® should be used for the con-
trol and the purple MAGNEST®s for the samples. Make sure
to load control/samples and gray/purple MAGNEST®s
accordingly using the same order. Every MAGNEST® has a
memory chip saving all the information created during the run.
It is deleted before every new run automatically, so the
MAGNEST® can be used continuously.
25. Each CellSearch® cartridge should be empty and clean inside.
The outside can be cleaned with a fiber-free tissue. The appro-
priate number of CellSearch® cartridges (corresponding to the
number of controls/samples) are inserted into the
MAGNEST®s and fixed properly by pushing them down.
MAGNEST®s must be undamaged since this could influence
the recovery of cells and the associated report. After entering a
MAGNEST®, the system shows the barcode of the CellSearch®
cartridge on the screen, closes the door, and moves the
MAGNEST® holder carousel to the next free position. All
filled MAGNEST®s have to be inserted.
26. The samples/controls have to be loaded in a way that the
barcode is directed toward the user and the sample tube is
freely rotatable. If there is no barcode label on the sample
tubes, the sample can be loaded anyway. By pushing “Enter,”
a warning message appears (see also Note 15) and the sample
ID and type can be edited. This information has to be saved.
32 €cilia Köstler et al.
Ca

27. The run takes 2 h 20 min plus 13 min per additional sample.
28. The CellSearch® cartridge must not be removed from the
MAGNEST® for inspection! It should be filled with an
orange-yellow colored solution (if not see Note 30). Be aware
that in rare cases, ferrofluid reagent can aggregate because of
disturbing factors in the blood of some patients. As a conse-
quence, the ferrofluids might mask CTCs during the scan
which could influence the results. Aggregated ferrofluids
close to the plug might have less influence to the results than
aggregates in the middle of the sample. If there are bubbles
inside the CellSearch® cartridge, place the plug onto the Cell-
Search® cartridge opening without pushing it down. While
keeping the plug in its place, knock the MAGNEST® gently
against a hard surface (e.g., bench) to destroy the bubbles.
Then re-seal the CellSearch® cartridge with the plug. If the
CellSearch® cartridge is empty, the associated conical tube
might still be filled with liquid. If this is the case, please seek
advice from the technical support.
29. The closed MAGNEST®s holding the CellSearch® cartridge
should be stored for a minimum of 20 min before scanning in a
horizontal position. This ensures that ferrofluid connected cells
sediment to the same layer for scanning. IVD samples and
controls have to be scanned with the CELLTRACKS ANA-
LYZER II® after 24 h at the latest and are allowed to be
scanned only twice. Note that if samples are scanned more
than twice within 24 h, they lose their IVD status.
30. Every CellSearch®Conical Centrifuge tube has to be inspected
before throwing it away. Ferrofluid accumulations might have
remained inside the tube. The results of such samples should be
considered with care since they might possibly be invalid.
31. Before removing the kit, it is possible to make a print-out with
the information of how many runs can still be performed with
the opened kit. For this, go back to the main menu (“View
Data” ! “Reagent Kit”) and print. For removing the kits,
close the bottles with their uniquely colored caps and store
the kit until next use at 2–8 °C. The dilution buffer has to be
stored at room temperature.
32. The mercury vapor lamp needs to warm up. This takes
approx.15 min. The lamp is switched on either after confirming
an automatically appearing request for warming up the lamp or
by starting it manually by clicking the button “lamp” in the
toolbar. The lamp signal blinks during warm-up phase and
remains green when the lamp is ready. The lamp has a total
run-time of 300 h. In order to save lifetime of the lamp, it is
advisable to switch it off when there are no samples to be
scanned and to avoid repeated switching on and off.
 CTC Isolation Using DEPArray 33

33. A system verification notice appears whenever the CELL-

TRACKS ANALYZER II® had been turned off, the applica-
tion was restarted, and 10 h have elapsed since the last system
verification. It verifies the optical performance and the cham-
ber skew. The system verification cartridge can be used for
180 verifications or until expiry date, depending on what
comes first. If it is not in use, protect the system verification
cartridge from light by storage in the corresponding box. Since
the chip of the MAGNEST® records (and counts) the use of
the system verification cartridge, the same MAGNEST® should
always be used for the system verification cartridge.
34. If the system verification fails, a second trial or the use of a new
system verification cartridge is possible. If it fails again, the user
manual helps for focusing the crosshairs, filling in and saving
the system verification information of the new system verifica-
tion cartridge.
35. Adding comments or editing sample information should be
done at this point. No changes can be done after the scan.
36. The system puts a rough focus to the ferrofluid and localizes
the CellSearch® cartridge edges. When the edges are detected,
four green dots appear on the image. If an edge is not detected,
a red dot appears. In this case, the scan has to be stopped and it
has to be confirmed that the CellSearch® cartridge has been
inserted correctly into the MAGNEST®. If it still does not
work, the edge has to be set manually by clicking to the
undetected edge within the picture.
37. The automatic focusing is done on nine spots of the Cell-
Search® cartridge. Six of nine spots have to be successful for
starting the scan.
38. The scan takes around 10 min. The CellSearch® cartridge is
scanned with each filter frame by frame. Since the frames are
overlapping, no cell is lost during the scan. After the scan is
finished, the system compares overlapping regions automati-
cally and removes duplicate cells.
39. Some frames appear gray after scanning. This means that the
camera has optimized the settings for making faint objects
visible.
40. The CellSearch® cartridge has to be removed from the
MAGNEST® and stored in a vertical position at 4 °C in the
dark until isolation of cells.
41. Results do not have to be evaluated immediately. All samples/
controls can be scanned first.
42. Tumor cell phenotype:
• CK-PE+, the cytokeratine signal is positive (green in the
composite image).
34 €cilia Köstler et al.
Ca

• DAPI+, the nucleus signal is positive.

• CD45-APC-, the specific signal for leucocytes is negative.
• The unlabeled channel is negative for tumor cells.
Morphology:
The object in the CK-PE channel should be an intact
cell. This means:
• It has a size of at least 4 μM (compare to size of the cursor)
and a round or oval shape (but also a polygonal or an
elongated shape is possible).
• The nuclear area is smaller than the cytoplasmic area.
• More than 50% of the nucleus is visibly surrounded by
cytoplasm.
43. When the gallery is shown the first time, the pictures are not
assigned.
44. For a high-resolution picture of the event, the event number
has to be clicked.
45. The evaluation does not have to be finished immediately. If the
page is closed during evaluation, the sample state changes to
“Review.” This means that a review has been started. When
each page of the gallery down to the last picture has been
evaluated, the number of assigned and unassigned events is
shown and the “Done” button is active for saving the sample
evaluation. The sample status changes to “Complete.”
46. Eight to twelve samples can be saved on a DVD + R. Only
completely reviewed or “enabled” samples can be archived.
47. In the “Archive” folder, “Patient Samples” or “Control Sam-
ples” can be selected for archiving the results. By clicking
“Archive” and selecting “Archive Oldest Results,” the oldest
results are stored to the DVD and confirmed with “Ok.” If
specific tests have to be archived together, the tests of interest
have to be selected, confirmed by clicking “Ok” and selecting
“Archive selected results.” By clicking the “Start” button, the
system automatically displays the number of tests to be burnt
on one DVD and assigns a disc number. After labeling and
insertion of a blank DVD, the archiving can be started with
“Start.”
48. Cells have to be labeled (an orange frame appears after selec-
tion) in the respective row where a definite signal (high or low)
is seen. If one low and one high signal are in the picture, the
event has to be counted as one low cell, and if two high cells are
in the picture, it has to be counted as one high cell.
49. I.e., 0.5 g BSA in 25 mL 1 × PBS. The solution can be stored in
1.5 mL aliquots at 4 °C up to 1 week or at -20 °C up to
12 months. Avoid to freeze and thaw aliquots more than once.
 CTC Isolation Using DEPArray 35

50. By using a 1000 μL tip, the plug of the CellSearch® cartridge

can be removed by levering on the backside of the plug. Alter-
natively, clean forceps can be used.
51. The pipette tip is coated by slowly pipetting BSA solution five
times up and down. By immerging the tip down to the bottom
of the tube, also the outside of the tip is covered. The coating
of the tip prevents cells from sticking to it.
52. The used tip should be kept by putting it vertically in a clean
open 1.5 mL tube placed in a 1.5 mL rack, the narrow end
pointing up. To avoid contamination, the tip should not be
touched at the narrow end.
53. The supernatants can be collected after the centrifugation steps
in 1.5 mL tubes. This ensures that cells are not lost if the pellet
had been disturbed and the isolation experiment had failed. In
this case, the supernatants of the two washing steps should be
combined, centrifuged for 5 min at 1000 g at RT, and reduced
to 12 μL (see Note 55).
54. First approx. 800 μL with the 1000 μL micropipette, then
approx. 150 μL with the 200 μL micropipette, and finally the
rest of the solution with the 20 μL micropipette should be
removed until approx. 10 μL are left. It is important not to
touch the bottom of the tube, due to possible loss of cells. The
20 μL tip should be kept in the second clean 1.5 mL tube
already prepared (as explained in Note 52) for adjusting the
volume to 12 μL and loading the DEPArray™ cartridge.
55. The volume for loading the DEPArray™ cartridge has to be
exactly 12 μL. To reach this, measure the volume left in the
tube with the 20 μL micropipette and the stored 20 μL tip
(from Note 54). After measuring, store the tip again in
the 1.5 mL tube. If the volume is lower than 12 μL, add the
missing volume with a fresh 20 μL tip (without touching the
cell suspension on the bottom of the tube). If the measured
volume is higher than 12 μL, centrifuge the sample using the
adaptor (see Note 9) with 1000 g for 5 min at room tempera-
ture. Remove the extra volume with a 10 μL micropipette
without touching the cell pellet for reaching exactly 12 μL of
sample volume. Alternatively, the VRNxt device can be used for
preparing the sample prior to loading.
56. The sonication of the buffer in the ultrasonic bath is done at
room temperature and solves potential aggregations in the
buffer. The activation of the degas function improves bubble-
free buffer loading to the DEPArray™ cartridge.
57. It is possible to uncover the password by checking the box at
“Show Code.”
36 €cilia Köstler et al.
Ca

58. With this application, it is possible to park up to 105 single cells

in the park chamber of the DEPArray™ cartridge and to
recover single cells with high performance.
59. There are two possibilities to open the protective cover with
the DEPArray™ cartridge inside. By opening cover “1,” the
DEPArray™ cartridge has the right position for being loaded
(back side of the cartridge, see Fig. 4) with buffer and sample.
By opening cover “2” (turn the protective cover upside down),
the DEPArray™ cartridge is in the right position for being
loaded to the system. In this position, the buffer and sample
channels are visible as well as the main, the recovery, and the
exit chambers forming the core of the cartridge (front side of
the cartridge, see Fig. 6).
60. For loading the DEPArray™ cartridge, the instruction
enclosed in the DEPArray™ cartridge package has to be fol-
lowed. However, in order to correctly load the DEPArray™
cartridge and avoid both contamination of the system and
sample as well as loss of important patient samples, it is strongly
recommended to participate at a DEPArray™ user training.
Among other applications, the user is trained the specific
“CTC-RUO-Fixed Cells” application, which forms the basis
for isolating CTCs with the DEPArray™ from a CellSearch®
cartridge, enriched with the CellSearch® system.
61. The tip of the 5000 μL micropipette has to be filled without
creating air bubbles. Those can influence the success of the
experiment. Therefore, the loading tip has to be pressed firmly
into the buffer inlet “B” and the pipetting speed should be
very low.
62. In order to avoid air reaching the sensitive main chamber of the
DEPArray™ cartridge, bulges to trap air bubbles are embed-
ded in the buffer channel of the DEPArray™ cartridge. Bub-
bles collected in these “traps” or at the end of the channels
usually do not cause problems during the run. In case of
bubbles in the middle of the channels or at the meniscus
close to the main chamber, the DEPArray™ cartridge is not
usable anymore and needs to be exchanged for a new one.
63. For this step, it is recommended to use the 20 μL tip stored
during Subheading 3.3 step 15 in order to reduce the number
of lost cells sticking to the pipette tip. Resuspend the sample
gently without creating air bubbles before loading it on the
DEPArray™ cartridge.
64. Once the sample meniscus reaches the main chamber during
the loading step, the DEPArray™ cartridge cannot be used any
more. After removing the sample from the DEPArray™car-
tridge, it can be reloaded to a new DEPArray™ cartridge (refer
to user manual). However, this might cause a loss of cells. All
 CTC Isolation Using DEPArray 37

important steps for sample loading are described in the

enclosed product insert of the DEPArray™ cartridge.
65. To avoid contamination, it is recommended to touch the touch
screen of the device only with a pen.
66. For slotting in the DEPArray™ cartridge correctly, it is neces-
sary to rotate it by 90° to the left while cover “2” is turned
up. Open cover “2” and insert the DEPArray™ cartridge.
67. The “Sample Loading Phase” consists of a calibration step
(an internal reference system in the chip is controlled) and
the sample loading step (the recovery/exit chamber and the
parking chamber are filled with DABUF and the main chamber
is filled with the sample). The “Sample Loading Phase” takes
around 15 min. The adjustment of the chip scan settings needs
to be done at the end of the “Sample Loading Phase.” There-
fore, it is important to stay next to the machine in order not to
miss the time point to pause the automatically starting “Chip
Scan.”
68. During this step, the surface of the main chamber is scanned
with each filter selected during chip scan settings.
69. Due to the sample stream during loading, most of the cells
locate to the lower right side of the main chamber. For check-
ing the filter settings, it is therefore best to select frames in this
part of the main chamber. However, selecting other frames is
also possible.
70. All information for the correct adaption of the exposure time,
the camera gain, and the offset focus can be found in the user
manual. However, for a correct adaption of the filter settings, a
training for the system is strongly recommended.
71. The CellBrowser™ and the “Cell Panel” are the two compo-
nents of the “Cell Selection Software.” The latter contains the
“Grid View” (a list of detected cells), the “Data Visualization”
(population plots), and the “Populations” (generation of sub-
populations and selection groups) panel. Multiple options like
histogram, scatter plot, position on the chip, set operator, and
population cloning are available for informed cell selection (see
user manual).
72. In the “Populations” panel, the detected events can be
assigned to four hierarchical levels (see Fig. 5). The events
detected automatically via image analysis and gating tools rep-
resent the input data and are called population “All” (= hierar-
chy one). The subpopulation “Routable Cells” (= hierarchy
two) is also created automatically. (For events grouped to
hierarchies three and four, see Notes 77 and 78).
73. After an automatic calculation of routing ways between each
cell and a parking position in the parking chamber, not all cells
38 €cilia Köstler et al.
Ca

may be routable. This may be due to the quality of the sample

and the sample preparation procedure but also to particular
algorithms for calculating the routes of the cells. The latter
reflects a safety step of the system. In order to guarantee the
purity of the routed cells, only those routes are considered
which assure a free passage of the target cell without touching
other cells or particles.
74. After the correct selection, the four subpopulation groups
(APC+/PE- [leukocytes] = Q1, PE+/APC- [putative
CTCs] = Q3, APC+/PE+ = Q2, PE-/APC- = Q4) are
saved by clicking “Create” and appear as four groups in the
hierarchical level three under the “Routable Cells” in the
“Populations” panel (see Fig. 5). For a more detailed descrip-
tion of the “Scatter Plot Filtering” or other filtering options,
follow the user manual.
75. For selecting cells which are to be isolated, a table has to be
created. This is done by selecting the subpopulation (e.g., the
PE+/APC- subpopulation = Q3) and clicking the “Create
Table” icon. The new table called “[0]Table0 (0)” (= hierarchy
four) appears under the subpopulation (in this case under Q3).
This works in analogy for each subpopulation, e.g., for the
APC+/PE- subpopulation (= Q1), a second table called “[1]
Table1(0)” (= hierarchy four) is created, etc.
76. It is not obligatory to sort the cells in this way, but the CTCs
are detected by the CK-PE signal and the signal strength
correlates with the probability of the event representing a
CTC. This is done analogously with the leukocytes (subpopu-
lation Q1) by sorting the CD45-APC signal of this cell popu-
lation from the highest to the lowest “Signal Intensity APC.”
77. Several parameters are important for an optimal cell selection.
Trapping, morphology, intensity, score, and other parameters
(just available for some DEPArray™ applications) are
described in detail in the user manual. To select the correct
cells from the different subpopulations, it is possible to follow
the user manual, but in order to avoid cross contamination of
the populations of interest, an official user training is strongly
recommended.
78. For “shifting,” e.g., putative CTCs from the Q3 subpopulation
to table “[0]Table0 (0),” the cell of interest in the “Grid View”
list has to be marked and the table-associated number in
squared brackets, in this case [0], has to be entered. The
current number of cells appears in parentheses (e.g., “[0]
Table0 (1)”). This can be applied to all following subpopula-
tions and tables which have been created to select cells from.
The number of cells per table is limited depending on how
 CTC Isolation Using DEPArray 39

many tables have been created (see also Note 79 and the
“CTC-RUO Fixed” application in the user manual).
79. At maximum 10 routing groups can be created. The total
number of cells to be isolated is limited depending on the
number of groups and the total number of cells in the groups
(e.g., with one routing group a maximum of 105 cells can be
selected; with two routing groups 51 cells per group can be
selected, etc.; for more details see user manual). The name of
the groups can be changed by right-click if necessary.
80. The “Cell Panel” consists of the “Sidebar” (selection of groups
and parameters), the “Action Bar” (information about cell
numbers, icons to switch to “Cell Browser” or “Cell Rout-
ing”), and the “Panel Grid” (information and pictures of
selected cells). It is saved automatically and opens by clicking
“Cell Panel” on the top right of the screen in the CellBrow-
ser™ software. Refer to user manual for adaption of the “Cell
Panel.”
81. “Park Routing” (i.e., moving the cells from their place to the
parking chamber of the DEPArray™ cartridge) can be started
by using the “Recovery Manager™” interface. This software
allows to watch the cells moving (due to the incorporated live
camera), to control the exit and recovery of the samples and to
assign samples to the “Recovery Support” layout.
82. It is not necessary but recommended to check if air bubbles are
at the entrance of the parking/exit chamber, which might
block the routing ways. If air bubbles are visible, it is important
to contact the Customer Support before starting the park
routing, otherwise important cells might be lost.
83. It is possible to route all cells within a group (cell pools) or to
select single cells. The selected groups/cells for parking are
labeled with a blue dot.
84. The duration of park routing depends on the number of
selected cells. Both progress and time left are displayed in the
“Park” window.
85. Even if for the “CTC-RUO Fixed” application, the “200 μL-
tube-rack” is preset and recommended for maximum purity of
the samples, another “Recovery Support” (e.g., if more than
33 tubes are needed for one experiment) or other types can be
chosen by clicking the “+” field in the “Exit and Recovery”
section. Choose the appropriate “Recovery Support” from the
drop-down menu and “Add” it. By clicking “Ok” an additional
layout becomes visible in the “Recover Support” section. For
changing the trays during recovery, follow the instructions in
the user manual.
40 €cilia Köstler et al.
Ca

86. “A1” is the defined position for the priming recovery, which is
not to be confused with the cell recoveries. Once selected by
right-click and “Set Priming Recover Position” it is marked in
orange and also displayed in the “Exit and Recovery” section.
87. For collecting the putative CTCs as cell groups or single cells
within the “CTC-RUO Fixed” application, it is recommended
only to use positions in columns 3, 6, 9, and 12 of the “200 μL-
tube-rack” plate layout. This results in a maximum of 33 allow-
able positions (and tubes), “Priming Recovery” included. It is
suggested to use the positions for the tubes starting from
column 3 (A3, B3, C3, etc., then columns 6, 9, and 12, accord-
ingly). The positions selected for the recovery are marked in
blue and are also displayed in the “Exit and Recovery” section.
All cells which have not been selected for recovery yet, stay in
the “Unassigned” group. Refer to user manual for more
detailed information.
88. In order to ensure purity of the samples to be recovered, it is
recommended to set a “Blank Recovery” between the different
cell types by right-clicking the position in the plate layout and
“Set Blank Recover.” The position is marked in blue and is also
displayed in the “Exit and Recovery” section. The “Exit and
Recovery” section also indicates how many drops are used for
filling each tube.
89. The “Recovery Support” is optimized for the 200 μL tubes of
the company Applied Biosystems. If other tubes are to be used, it
is necessary to get confirmation by the customer support.
90. The lid of the tube at position “A1” should be opened and
directed toward “1” (printed on the recovery tray). The same
holds true for the tube in position “A3.” All other tubes should
have their lids directed toward the letters “A”–“H” (printed on
the recovery tray). For a detailed instruction, follow the user
manual.
91. Once washing is started, it is not possible to switch back to the
CellBrowser™ and no more cells can be parked in the
parking area.
92. The process can be stopped by any user regardless of who had
signed in for starting the machine.
93. DEPArray™ cartridges can be used only once.
94. For saving the data, the connection to the backup unit is
obligatory. The backup unit has to be switched on, otherwise
the backup will fail. For more details, follow the user manual.
95. The duration of the backup process depends on the amount of
data to be saved. If a new run is performed the same day, the
backup process can be cancelled by clicking the red button with
the cross at the software screen. It is obligatory to switch off
 CTC Isolation Using DEPArray 41

the device and wait for a minimum of 30 s before restarting it

for a new run. The created data which are not saved yet in the
backup unit are saved in an internal storage place. They will be
saved during the next backup run and will not be lost.
96. The volume of the tubes with 21–85 cells differs from the
volume of the tubes with 1–20 cells. The drop rules for the
cell recovery are defined depending in which recovery format
(e.g., 200 μL tube rack) the cells are collected (for more details
see user manual).
97. If 1–20 cells are recovered in one tube, centrifuge for 30 s at
14,100 × g. If cell pools with 21–85 cells are recovered in one
tube, centrifuge for 10 min at 14,100 × g. Make sure that all
tube caps are aligned equally with regard to the center of the
rotor. It is important to know where exactly the cells will be
located after centrifugation in order not to lose the cells when
reducing the supernatant to 1 μL in step 8.
98. This step is very critical in terms of losing the collected cells.
The handling is improved by experience.

References
1. Miller MC, Doyle GV, Terstappen LW (2010) 5. de Bono JS, Scher HI, Montgomery RB,
Significance of circulating tumor cells detected Parker C, Miller MC, Tissing H, Doyle GV,
by the cellsearch system in patients with meta- Terstappen LW, Pienta KJ, Raghavan D (2008)
static breast colorectal and prostate cancer. J Circulating tumor cells predict survival benefit
Oncol 2010:617421. https://doi.org/10. from treatment in metastatic castration-resistant
1155/2010/617421 prostate cancer. Clin Cancer Res 14(19):
2. Alberter B, Klein CA, Polzer B (2016) Single- 6302–6309. https://doi.org/10.1158/
cell analysis of CTCs with diagnostic precision: 1078-0432.CCR-08-0872
opportunities and challenges for personalized 6. Cohen SJ, Punt CJ, Iannotti N, Saidman BH,
medicine. Expert Rev Mol Diagn 16(1):25–38. Sabbath KD, Gabrail NY, Picus J, Morse M,
https://doi.org/10.1586/14737159.2016. Mitchell E, Miller MC, Doyle GV, Tissing H,
1121099 Terstappen LW, Meropol NJ (2008) Relation-
3. Polzer B, Medoro G, Pasch S, Fontana F, ship of circulating tumor cells to tumor
Zorzino L, Pestka A, Andergassen U, Meier- response, progression-free survival, and overall
Stiegen F, Czyz ZT, Alberter B, Treitschke S, survival in patients with metastatic colorectal
Schamberger T, Sergio M, Bregola G, Doffini A, cancer. J Clin Oncol 26(19):3213–3221.
Gianni S, Calanca A, Signorini G, Bolognesi C, https://doi.org/10.1200/JCO.2007.15.8923
Hartmann A, Fasching PA, Sandri MT, Rack B, 7. Bidard FC, Peeters DJ, Fehm T, Nole F,
Fehm T, Giorgini G, Manaresi N, Klein CA Gisbert-Criado R, Mavroudis D, Grisanti S,
(2014) Molecular profiling of single circulating Generali D, Garcia-Saenz JA, Stebbing J,
tumor cells with diagnostic intention. EMBO Caldas C, Gazzaniga P, Manso L, Zamarchi R,
Mol Med 6(11):1371–1386. https://doi.org/ de Lascoiti AF, De Mattos-Arruda L,
10.15252/emmm.201404033 Ignatiadis M, Lebofsky R, van Laere SJ, Meier-
4. Cristofanilli M, Budd GT, Ellis MJ, Stopeck A, Stiegen F, Sandri MT, Vidal-Martinez J,
Matera J, Miller MC, Reuben JM, Doyle GV, Politaki E, Consoli F, Bottini A, Diaz-Rubio E,
Allard WJ, Terstappen LW, Hayes DF (2004) Krell J, Dawson SJ, Raimondi C, Rutten A,
Circulating tumor cells, disease progression, Janni W, Munzone E, Caranana V, Agelaki S,
and survival in metastatic breast cancer. N Engl Almici C, Dirix L, Solomayer EF, Zorzino L,
J Med 351(8):781–791. https://doi.org/10. Johannes H, Reis-Filho JS, Pantel K, Pierga JY,
1056/NEJMoa040766 Michiels S (2014) Clinical validity of circulating
42 €cilia Köstler et al.
Ca

tumour cells in patients with metastatic breast Schneeweiss A, Lichtenegger W, Beckmann

cancer: a pooled analysis of individual patient MW, Friese K, Pantel K, Janni W, Group SS
data. Lancet Oncol 15(4):406–414. https:// (2014) Circulating tumor cells predict survival
doi.org/10.1016/S1470-2045(14)70069-5 in early average-to-high risk breast cancer
8. Rack B, Schindlbeck C, Juckstock J, patients. J Natl Cancer Inst 106(5). https://
Andergassen U, Hepp P, Zwingers T, Friedl doi.org/10.1093/jnci/dju066
TW, Lorenz R, Tesch H, Fasching PA, Fehm T,
 Chapter 3

Isolation of Viable Epithelial and Mesenchymal Circulating

Tumor Cells from Breast Cancer Patients
Justyna Topa, Anna J. Żaczek, and Aleksandra Markiewicz

Abstract
Circulating tumor cells (CTCs) undergoing epithelial-mesenchymal transition (EMT) may exhibit more
aggressive features than epithelial CTCs and are more frequently observed during disease progression.
Therefore, detection and characterization of both epithelial and mesenchymal CTCs in cancer patients are
urgently needed to allow for a better understanding of the metastatic process and more effective treatment.
Here we describe a method for detection and isolation of viable epithelial and mesenchymal CTCs from
peripheral blood of breast cancer patients. The method is based on density gradient centrifugation,
multiplex immunofluorescent staining, and negative anti-CD45 selection. Cells obtained after the proce-
dure are suitable for genomic or transcriptomic profiling, and they can also be isolated by micromanipula-
tion for single-cell analysis.

Key words Breast cancer, Circulating tumor cells, Epithelial-mesenchymal transition, Mesenchymal
phenotype

1 Introduction

Breast cancer (BC) is the most common cancer type in women

worldwide and there is an upward trend in the number of BC
cases in recent years [1]. A prognosis for BC patients strictly
depends on the disease advancement at the time of diagnosis [2]
and molecular subtype [3]. Distant metastases remain the main
cause of BC-related death, and metastasis mediators, circulating
tumor cells (CTCs) present in the bloodstream, are a factor of
poor prognosis, in both early and metastatic BC patients [2, 4–6].
CTCs are very heterogeneous, as they may change their phe-
notype epithelial-mesenchymal transition (EMT) [7]. In this pro-
cess, BC cells lose their epithelial characteristics (apico-basal
polarity, strong adhesion to other cells and basal membrane) and
gain aggressive features (motility, ability to degrade extracellular
matrix, features of stem cells) [8–10]. It has been shown that
mesenchymal CTCs are associated with shorter overall survival

Miodrag Gužvić (ed.), Single Cell Analysis: Methods and Protocols, Methods in Molecular Biology, vol. 2752,
https://doi.org/10.1007/978-1-0716-3621-3_3, © Springer Science+Business Media, LLC, part of Springer Nature 2024

43
44 Justyna Topa et al.

and unfavorable tumor characteristics [6, 7], and are more fre-
quently present in the advanced stages of the disease [11]. During
disease progression, cancer cells may change their phenotype into
the most favorable to particular conditions due to EMT and its
reverse process—mesenchymal-epithelial transition (MET)
[12, 13]; activation of EMT might promote the dissemination,
whereas reversing the mesenchymal phenotype might be crucial
for establishing new metastases [14].
Many methods allowing CTCs enrichment, detection, and
isolation were already described [15–17], with a vast majority
relying on epithelial markers (including golden standard in CTCs
detection—CellSearch®), limiting their ability to detect cells under-
going EMT, and further characterization of mesenchymal CTCs.
The method described here involves immunofluorescent (IF) stain-
ing of peripheral blood mononuclear cells (PBMCs) fraction
isolated from BC patient’s blood by density gradient centrifugation
and negative selection of CD45-positive blood cells, which is an
unbiased way of enrichment of CTCs with epithelial and mesen-
chymal features [6, 7]. For IF staining, we used antibodies directed
against common epithelial cell-surface markers, EpCAM and
E-cadherin; as well as MCAM (CD146), a protein present on the
surface of EpCAM-negative BC cell lines [18]. MCAM is associated
with mesenchymal phenotype and increased motility of BC cell
lines [19] and has been shown to improve CTCs detection in BC
patients [20]. As MCAM may be present on circulating endothelial
cells [21], the method includes CD31 (endothelial marker) in the
panel of exclusion markers, next to CD45 (blood cells marker) and
DAPI (dead cells marker). Our approach (see Subheading 2) allows
observation of epithelial markers in the “green” channel, whereas
mesenchymal markers in the “red” channel. All exclusion mar-
kers (CD31, CD45, DAPI) are visible in the “blue” channel. This
setup may be freely modified by the addition of other identified
CTCs markers. Performing IF staining before blood cells depletion
minimizes accidental loss of rare cancer cells during multiple wash-
ing steps. CD45-positive cells negative selection by Dynabeads™
CD45 is one of the most efficient methods for CTCs enrichment
regardless of their phenotype [17, 22] and in combination with IF
staining allows high recovery of living cells (Fig. 1a). The described
method also enables high enrichment of CTCs fraction, as after the
whole procedure about 98.4% of blood cells are depleted (Fig. 1b)
when 5 mL of blood was processed.

2 Materials

2.1 CTCs Enrichment 1. Low bind 1.5 mL tubes (e.g., Ultra High Recovery Tubes,
Starlab, cat. No. E1415-2600 or Protein LoBind Tubes,
Eppendorf, cat. No. 0030108116), 15 mL and 50 mL tubes.
 Viable CTCs Isolation Method 45

Fig. 1 (a) The recovery rate of 100 MCF-7 (epithelial) cells and 100 MDA-MB-231 (mesenchymal) cells of BC
cell lines spiked into 5 mL of blood. Recovery rate is shown as mean ± SD from three independent
experiments. (b) Number of nucleated cells after density gradient centrifugation and after the whole procedure
as determined by processing 5 mL of blood from healthy donors (with no diagnosed cancer) samples (n = 6)

2. K2EDTA-coated blood collection tubes (BD Vacutainer®).

3. Syringes and syringe filters with 0.22 μm pores.
4. Coating Buffer: 2 mM EDTA, 1% FBS in 1xPBS (pH 7.4).
Measure out 1 L of 1xPBS and transfer into 1 L glass bottle.
Weigh 0.744 g of disodium EDTA 2-hydrate and transfer into
a bottle. Mix on a roller or magnetic stirrer until EDTA dis-
solves. Sterilize in autoclave. After cooling down, add 10 mL of
sterile FBS, mix gently, and divide into 100 mL portions (see
Note 1). Store at 4 °C for up to 4 weeks.
5. Histopaque®-1077 (Sigma Aldrich).

2.2 Immuno- 1. Blocking Buffer: 50 mM glycine, 5% BSA in 1xPBS (pH 7.4).

fluorescent Staining Measure out 100 mL of 1xPBS and transfer into 100 mL glass
bottle. Weigh 0.375 g of glycine and 5 g of BSA. Transfer into a
bottle and mix on a roller or magnetic stirrer until glycine and
BSA dissolve. Perform filtration through filters with 0.22 μm
pores. Divide into 1 mL portions, leaving 10 mL for Staining
Buffer preparation. Store at -20 °C for up to 6 months.
2. Staining Buffer: 10 mM glycine, 1% BSA in 1xPBS (pH 7.4).
Measure out 10 mL of sterile Blocking Buffer and transfer into
50 mL bottle. Refill to the volume of 50 mL with 1xPBS.
Divide into 1 mL portions and store at -20 °C for up to
6 months.
3. Anti-EpCAM antibody: clone VU1D9, Alexa Fluor®
488-conjugated, Cell Signalling Technology, cat. No. 5198
(see Note 2).
46 Justyna Topa et al.

4. Anti-E-cadherin antibody: clone 67A4, Alexa Fluor®

488-conjugated, Santa Cruz Biotechnology, cat.
No. sc-21791.
5. Anti-MCAM antibody—clone P1H12, Alexa Fluor®
594-conjugated, Santa Cruz Biotechnology, cat. No. sc-
18837.
6. Anti-CD31 antibody—clone WM-59, Super Bright
436-conjugated, eBioscience, cat. No. 62-0319-42.
7. Anti-CD45 antibody—clone REA747, VioBlue®-conjugated,
Miltenyi Biotec, cat. No. 130-110-637.
8. DAPI—stock 1 mg/mL, intermediate stock 1:200 (stored at -
20 °C, aliquoted).
9. Sample Mixer (e.g., HulaMixer® Sample mixer,
Thermofisher).

2.3 CD45-Positive 1. Dynabeads™ CD45, Thermofisher, cat. No. 11153D.

Cells Negative 2. DynaMag™-15 Magnet, Thermofisher, cat. No. 12301D.
Selection

2.4 Additional 1. Centrifuge with temperature, acceleration, and brake control,

Equipment and with tilting rotor, for 15 and 50 mL tubes.
2. Centrifuge with temperature control for 1.5 mL tubes.

3 Methods

The CTCs isolation method is based on density gradient centrifu-

gation, multiplex IF staining, and anti-CD45 depletion (Fig. 2).

3.1 CTCs Enrichment 1. Prepare all buffers and reagents needed during the procedure:
Coating Buffer, 1xPBS, Histopaque® at room temperature,
and additional Coating Buffer and 1× PBS in 4 °C (see Note
3). Pre-coat 2× low-bind 1.5 mL tubes, 4 × 15 mL, and
1 × 50 mL tubes; add 1, 2, and 10 mL of Coating Buffer
(at room temperature) into tubes, respectively, and put on a
roller for 15 min to allow coating walls of the tubes with the
buffer and prevent cells sticking to the tube. After 15 min
remove the solution from the tubes.
2. By venipuncture collect blood sample into K2EDTA-coated
tubes. Discard the tube with the first 1 mL of blood that
might contain contaminating epithelial cells or fibroblasts due
to skin punctuation. Use a second K2EDTA-coated tube to
collect 5 mL of blood for the analysis (see Note 4).
3. To remove platelets, transfer blood into a pre-coated 15 mL
tube and centrifuge at 200 × g, 21 °C for 10 min. Discard the
 Viable CTCs Isolation Method 47

Fig. 2 The workflow of the CTCs isolation method

top layer of plasma and avoid taking any of the cellular fraction
beneath it (around 2=3 can easily be removed).
4. Refill the remaining sample with 1xPBS (room temperature) to
the volume of 9 mL, gently mix, and carefully layer onto 4 mL
of sterile Histopaque® in the new pre-coated 15 mL tube. Be
careful not to mix the layers of diluted blood and Histopaque®,
as it may affect the separation of cells. Centrifuge the samples at
400 × g, 21 °C for 30 min, with no brake and acceleration (see
Note 5).
5. From that moment carry out all the steps on ice. Collect PBMCs
fraction (see Note 6) into new pre-coated 15 mL tube and refill
to the volume of 10 mL with cold (4 °C) 1xPBS. Centrifuge at
450 × g, 4 °C, 10 min to obtain a cell pellet. Discard the
supernatant and immediately proceed to the next steps of the
procedure.

3.2 Immuno- 1. Suspend the pellet in 250 μl of Blocking Buffer (prepared

fluorescent Staining according to the description in the Subheading 2.2) and trans-
and Negative Selection fer it into a pre-coated 1.5 mL tube. To wash out any remaining
of CD45-Positive Cells cells (possibly also CTCs), rinse the 15 mL tube that contained
PBMCs fraction with another 250 μl of Blocking Buffer and
transfer it into the 1.5 mL tube containing suspended cell
pellet. Incubate at 4 °C for 15 min on the sample mixer.
Centrifuge at 400 × g, 4 °C, 5 min.
2. Carefully remove the supernatant with a pipette (see Note 7)
and add 200 μl of freshly prepared antibodies mix (Table 1)
into the pellet, with gentle pipetting. From that moment sample
must be protected from light. Incubate sample at 4 °C for 30 min
on sample mixer.
48 Justyna Topa et al.

Table 1
Antibodies mix used for immunofluorescent staininga

Component Final dilution Volume

anti-EpCAM Ab 1:200 1
anti-E-cadherin Ab 1:100 2.0
anti-MCAM Ab 1:200 1.0
anti-CD31 Ab 1:100 2.0
DAPI 1:10,000 2.0
Staining buffer – 192
a
Does not include CD45 antibody, which is added later during CD45-positive cells
depletion

3. During staining of the PBMCs fraction, prepare Dynabeads™

CD45 for further use. Transfer 125 μl of well-resuspended
Dynabeads™ CD45 into a pre-coated 15 mL tube (see Note
8). Add 1 mL of Coating Buffer and mix with Pasteur pipette.
Place the tube into the DynaMag™-15 Magnet for 1 min and
then discard the supernatant. Be careful not to touch magnetic
beads on the side of the tube. Take the tube off the magnet and
resuspend washed Dynabeads™ CD45 in 125 μl of Coating
Buffer.
4. When IF staining of PBMCs fraction is finished, centrifuge the
tube at 400 × g, 4 °C, 1 min, then gently suspend the cells, and
transfer into the pre-coated 15 mL tube containing washed
Dynabeads™ CD45. To collect residual cells from the tube,
rinse the 1.5 mL tube after PBMCs IF staining with 675 μl of
Coating Buffer and transfer it to a 15 mL tube with PBMCs
and Dynabeads™ CD45 (total volume should be 1 mL). Incu-
bate the sample for 5 min at 4 °C on the sample mixer. During
the incubation, CD45-positive cells are captured with anti-
CD45 antibody-coated magnetic nanoparticles (see Note 9).
5. Add anti-CD45 antibody directly to the tube (10 μl, to the final
dilution of 1:100) with PBMCs and Dynabeads™ CD45. After
25 min of incubation (at 4 °C), remove the tube from the mixer
and dilute the sample to the volume of 13 mL with cold (4 °C)
Coating Buffer (see Note 10). Mix sample with Pasteur pipette
and place tube in the DynaMag™-15 Magnet for 10 min
(cover the magnet with the tube to protect it from light)
CD45-positive cells attached to the magnetic beads remain
on the side of the tube while kept on the magnet, whereas
CD45-negative cells remain in the solution.
6. Carefully transfer CTCs-enriched cells suspension into a
pre-coated 50 mL tube and fill it to the volume of 25 mL
 Viable CTCs Isolation Method 49

Fig. 3 (a) Spiked-in MCF-7 cells positive for epithelial markers (EpCAM and E-cadherin; green), and MDA-MB-
231 positive for MCAM (red) after CTCs isolation procedure. Spiked cells show high viability (DAPI-negative)
and are negative for CD45 and CD31 (blue). (b) Picking of single MCF-7 cell positive for epithelial markers

with cold (4 °C) Coating Buffer. Centrifuge at 400 × g, 4 °C for

5 min and remove supernatant with a pipette controller or
Pasteur pipette (see Note 11), leaving about 100 μl of the
liquid above the pellet. Resuspend the pellet and transfer it to
a pre-coated 1.5 mL tube. To collect any remaining cells, rinse
the bottom of the 50 mL tube with 100–200 μl of Coating
Buffer and transfer it into 1.5 mL tube containing CD45-
depleted CTCs fraction.

3.3 Single CTCs Such prepared CTCs-enriched cell suspension may be further pro-
Isolation by cessed as bulk or used for single-cell analyses. For our research,
Micromanipulation single CTCs were isolated by micromanipulation (Fig. 3) described
elsewhere in this volume. Such captured single cells may be further
subjected to genomic and transcriptomic analyses, as demonstrated
in other protocols in this volume.

4 Notes

1. Add FBS under the laminar hood, to ensure the buffer remains
sterile. It will allow longer storage of the Coating Buffer.
100 mL of the buffer is enough to carry out the CTCs isolation
procedure from two 5 mL blood samples.
2. Fluorochromes should be matched to the lasers and filters
available on a particular microscope.
3. About 25 mL of 1xPBS (10 mL at RT and 15 mL at 4 °C) and
50 mL of Coating Buffer (20 mL at RT and 30 mL at 4 °C) are
enough to perform the procedure on one 5 mL blood sample.
Histopaque® may be aliquoted into 4 mL portions, to
reach room temperature faster.
50 Justyna Topa et al.

4. Collected blood should be processed as soon as possible, pref-

erably within 2 h after collection. After blood collection, the
sample should be stored at 4–8 °C, but it should be left to
equilibrate to room temperature for density gradient
centrifugation.
5. Ensure that acceleration and brake are turned off to avoid
mixing the layers after density gradient centrifugation.
6. Crucial step: ensure that the whole PBMCs fraction is col-
lected. It is better to aspirate a higher volume of Histopaque®
or diluted plasma than leave uncollected cells, among which
may be CTCs.
7. You can leave about 50 μl of the supernatant above the cell
pellet, to avoid accidental cells aspiration.
8. It is crucial to mix well Dynabeads™ CD45, to aspirate the
proper volume of Ab-coated magnetic particles. In this proce-
dure, we are using an excess of the reagent to ensure efficient
CD45-positive cells depletion from 5 mL of blood.
9. During immunomagnetic depletion, IF staining with anti-
CD45 antibody is performed. However, to avoid blocking
the CD45 molecule from binding immunomagnetic anti-
CD45 beads, first magnetic beads are added, followed by a
fluorescently labeled anti-CD45 antibody.
10. Make sure to aspirate also the solution that might be present in
the cap of the tube.
11. Supernatant may be aspirated with the use of glass Pasteur
pipette connected to vacuum pump, but attention should be
paid to not aspirate cell pellet.

Acknowledgments

This work was supported by the National Science Centre grant

number 2016/21/D/NZ3/02629.

References
1. Sung H, Ferlay J, Siegel RL, Laversanne M, 3. Fallahpour S, Navaneelan T, De P, Borgo A
Soerjomataram I, Jemal A, Bray F (2021) (2017) Breast cancer survival by molecular sub-
Global cancer statistics 2020: GLOBOCAN type: a population-based analysis of cancer reg-
estimates of incidence and mortality worldwide istry data. CMAJ Open 5(3):E734–E739.
for 36 cancers in 185 countries. CA Cancer J https://doi.org/10.9778/cmajo.20170030
Clin 71(3):209–249. https://doi.org/10. 4. Bidard FC, Peeters DJ, Fehm T, Nole F,
3322/caac.21660 Gisbert-Criado R, Mavroudis D, Grisanti S,
2. Dong G, Wang D, Liang X, Gao H, Wang L, Generali D, Garcia-Saenz JA, Stebbing J,
Yu X, Liu J (2014) Factors related to survival Caldas C, Gazzaniga P, Manso L,
rates for breast cancer patients. Int J Clin Exp Zamarchi R, de Lascoiti AF, De Mattos-
Med 7(10):3719–3724 Arruda L, Ignatiadis M, Lebofsky R, van
Laere SJ, Meier-Stiegen F, Sandri MT, Vidal-
 Viable CTCs Isolation Method 51

Martinez J, Politaki E, Consoli F, Bottini A, IL-6-induced breast cancer cell EMT and stem-
Diaz-Rubio E, Krell J, Dawson SJ, ness via c-myc activation. Breast Cancer 26(5):
Raimondi C, Rutten A, Janni W, Munzone E, 663–671. https://doi.org/10.1007/s12282-
Caranana V, Agelaki S, Almici C, Dirix L, Solo- 019-00966-3
mayer EF, Zorzino L, Johannes H, Reis-Filho 13. Ocana OH, Corcoles R, Fabra A, Moreno-
JS, Pantel K, Pierga JY, Michiels S (2014) Clin- Bueno G, Acloque H, Vega S, Barrallo-
ical validity of circulating tumour cells in Gimeno A, Cano A, Nieto MA (2012) Meta-
patients with metastatic breast cancer: a pooled static colonization requires the repression of
analysis of individual patient data. Lancet the epithelial-mesenchymal transition inducer
Oncol 15(4):406–414. https://doi.org/10. Prrx1. Cancer Cell 22(6):709–724. https://
1016/S1470-2045(14)70069-5 doi.org/10.1016/j.ccr.2012.10.012
5. Cristofanilli M, Budd GT, Ellis MJ, Stopeck A, 14. Markiewicz A, Topa J, Popeda M, Szade J,
Matera J, Miller MC, Reuben JM, Doyle GV, Skokowski J, Welnicka-Jaskiewicz M, Zaczek
Allard WJ, Terstappen LW, Hayes DF (2004) A (2021) Activation of epithelial-mesenchymal
Circulating tumor cells, disease progression, transition process during breast cancer
and survival in metastatic breast cancer. N progression—the impact of molecular subtype
Engl J Med 351(8):781–791. https://doi. and stromal composition. Acta Biochim Pol
org/10.1056/NEJMoa040766 68(3):385–392. https://doi.org/10.18388/
6. Markiewicz A, Nagel A, Szade J, Majewska H, abp.2020_5719
Skokowski J, Seroczynska B, Stokowy T, 15. Bailey PC, Martin SS (2019) Insights on CTC
Welnicka-Jaskiewicz M, Zaczek AJ (2018) biology and clinical impact emerging from
Aggressive phenotype of cells disseminated via advances in capture technology. Cell 8(6).
hematogenous and lymphatic route in breast https://doi.org/10.3390/cells8060553
cancer patients. Transl Oncol 11(3):722–731. 16. Rushton AJ, Nteliopoulos G, Shaw JA,
https://doi.org/10.1016/j.tranon.2018. Coombes RC (2021) A review of circulating
03.006 tumour cell enrichment technologies. Cancers
7. Markiewicz A, Topa J, Nagel A, Skokowski J, (Basel) 13(5). https://doi.org/10.3390/
Seroczynska B, Stokowy T, Welnicka- cancers13050970
Jaskiewicz M, Zaczek AJ (2019) Spectrum of 17. Topa J, Grešner P, Żaczek AJ, Markiewicz A
epithelial-mesenchymal transition phenotypes (2022) Breast cancer circulating tumor cells
in circulating tumour cells from early breast with mesenchymal features—an unreachable
cancer patients. Cancers (Basel) 11(1). target? Cell Mol Life Sci 79(2):81. https://
https://doi.org/10.3390/cancers11010059 doi.org/10.1007/s00018-021-04064-6
8. Kalluri R, Weinberg RA (2009) The basics of 18. Mostert B, Kraan J, Bolt-de Vries J, van der
epithelial-mesenchymal transition. J Clin Spoel P, Sieuwerts AM, Schutte M, Timmer-
Invest 119(6):1420–1428. https://doi.org/ mans AM, Foekens R, Martens JW, Gratama
10.1172/JCI39104 JW, Foekens JA, Sleijfer S (2011) Detection of
9. Mani SA, Guo W, Liao MJ, Eaton EN, circulating tumor cells in breast cancer may
Ayyanan A, Zhou AY, Brooks M, Reinhard F, improve through enrichment with anti-
Zhang CC, Shipitsin M, Campbell LL, CD146. Breast Cancer Res Treat 127(1):
Polyak K, Brisken C, Yang J, Weinberg RA 33–41. https://doi.org/10.1007/s10549-
(2008) The epithelial-mesenchymal transition 010-0879-y
generates cells with properties of stem cells. 19. Zabouo G, Imbert AM, Jacquemier J,
Cell 133(4):704–715. https://doi.org/10. Finetti P, Moreau T, Esterni B, Birnbaum D,
1016/j.cell.2008.03.027 Bertucci F, Chabannon C (2009) CD146
10. Moreno-Bueno G, Portillo F, Cano A (2008) expression is associated with a poor prognosis
Transcriptional regulation of cell polarity in in human breast tumors and with enhanced
EMT and cancer. Oncogene 27(55): motility in breast cancer cell lines. Breast Can-
6958–6969. https://doi.org/10.1038/onc. cer Res 11(1):R1. https://doi.org/10.1186/
2008.346 bcr2215
11. Zhang S, Wu T, Peng X, Liu J, Liu F, Wu S, 20. Onstenk W, Kraan J, Mostert B, Timmermans
Liu S, Dong Y, Xie S, Ma S (2017) Mesenchy- MM, Charehbili A, Smit VT, Kroep JR, Nortier
mal phenotype of circulating tumor cells is JW, van de Ven S, Heijns JB, Kessels LW, van
associated with distant metastasis in breast can- Laarhoven HW, Bos MM, van de Velde CJ,
cer patients. Cancer Manag Res 9:691–700. Gratama JW, Sieuwerts AM, Martens JW, Foe-
https://doi.org/10.2147/CMAR.S149801 kens JA, Sleijfer S (2015) Improved circulating
12. Gao X, Liu X, Lu Y, Wang Y, Cao W, Liu X, tumor cell detection by a combined EpCAM
Hu H, Wang H (2019) PIM1 is responsible for and MCAM cellsearch enrichment approach in
52 Justyna Topa et al.

patients with breast cancer undergoing neoad- 22. Guo M, Li X, Zhang S, Song H, Zhang W,
juvant chemotherapy. Mol Cancer Ther 14(3): Shang X, Zheng Y, Jiang H, Lv Q, Jiang Y,
8 2 1 – 8 2 7 . h t t p s : // d o i . o r g / 1 0 . 1 1 5 8 / Hao H (2015) Real-time quantitative
1535-7163.MCT-14-0653 RT-PCR detection of circulating tumor cells
21. Strijbos MH, Gratama JW, Kraan J, Lamers from breast cancer patients. Int J Oncol
CH, den Bakker MA, Sleijfer S (2008) Circu- 46(1):281–289. https://doi.org/10.3892/
lating endothelial cells in oncology: pitfalls and ijo.2014.2732
promises. Br J Cancer 98(11):1731–1735.
https://doi.org/10.1038/sj.bjc.6604383
 Chapter 4

Single-Cell Recovery from Tumor Cell Xenotransplanted

Zebrafish Embryos for the Study of Metastasis-Initiating
Cells
Pablo Hurtado, Inés Martı́nez-Pena, and Roberto Piñeiro

Abstract
The study of metastasis-competent cells at the single-cell level represents an opportunity to decipher the
molecular mechanisms associated with the metastatic cascade as well as to understand the functional and
molecular heterogeneity of these cells. In this context, preclinical in vivo models of cancer metastasis are
valuable tools to understand the behavior of cancer cells throughout the process. Here we describe a
detailed protocol for the isolation and recovery of individual viable human metastatic cells from zebrafish
embryos xenotransplanted with cancer cells for downstream molecular analysis. We cover the critical steps
for the dissociation of the xenografted zebrafish embryos to generate a single-cell suspension, and the
micromanipulation for their recovery as single cells.

Key words Metastasis, Zebrafish, Dissociation, Single-cell suspension, Micromanipulation

1 Introduction

The metastatic cascade is a complex multistep process by which

cancer cells abandon the primary tumor and reach distant organs
and tissues where they grow into new tumors. Cancer metastasis is a
highly inefficient process, and only a very small proportion of the
cells released by the primary tumor are able to survive this process
and successfully seed metastases [1]. The potential of cancer cells to
form metastases is mainly determined by their genetic composition
as well as by the interactions with the tumor microenvironment
(TME).
The use of animal models, and in particular the mouse, has
been an important tool to dissect the different stages of the meta-
static process and unravelling their interaction with the TME along
the metastatic process [2]. In recent years, the zebrafish embryo has
emerged as an alternative for murine xenograft models of metasta-
sis, allowing a more reliable dissection of the metastatic cascade and

Miodrag Guzvic (ed.), Single Cell Analysis: Methods and Protocols, Methods in Molecular Biology, vol. 2752,
https://doi.org/10.1007/978-1-0716-3621-3_4, © Springer Science+Business Media, LLC, part of Springer Nature 2024

53
54 Pablo Hurtado et al.

the study of the metastatic competency of cancer cells [3–5]. Their

fast extra-uterine development, their transparency, and the lack of a
mature adaptive immune system make them a very good organism
to model tumor growth and metastasis [3]. Therefore, zebrafish
xenografts allow to study in real time the behavior of just a few
hundred cancer cells and to follow them on time individually.
The recent development of methods and technologies to study
cancer cells at a single-cell level has opened new avenues to, among
others, characterize the intra-tumor cellular heterogeneity, and to
identify rare cancer cell subpopulations [6], e.g., metastasis-
initiating cells. Thus, nowadays it is possible to capture and isolate
individual cancer cells from different sources, such as tumor tissue
or blood biopsies, to study their mutation and gene expression
profile. The combination of these technologies with in vivo preclin-
ical models of cancer metastasis will make plausible to decipher the
molecular mechanisms underlying the metastatic competence of
tumor-disseminated cells under controlled conditions.
Here we describe a protocol for the isolation of individual
metastatic tumor cells from xenotransplanted zebrafish embryos
for the development of downstream molecular assays (e.g., DNA
or RNA analyses). This protocol explains in detail the different
steps for the efficient dissociation of the zebrafish embryo in
order to generate a single-cell suspension. It describes how to
manipulate and dissect the zebrafish embryos to obtain the tissue
sections where cells have disseminated, and the following dissocia-
tion process. The dissociation process combines an enzymatic and
mechanical digestion that allows to obtain a homogeneous
single-cell suspension. In addition, the protocol includes a detailed
step-by-step description of the micromanipulation process to suc-
cessfully obtain a pure and viable cell for molecular interrogation. It
describes the main considerations to take into account (e.g., cell
density and purity) for an easy and efficient recovery of the cells of
interest without carrying over contaminating material present in
the sample such as host cells. It also describes a few control tips to
ensure the adequate integrity of the collected cell as well as the
efficient transfer to the final recipient tube. Importantly, this proto-
col can be carried out using a manual micromanipulator and a
mechanical microinjector, without the need to use expensive auto-
mated equipment. Lastly, this protocol is suitable for the isolation
of intact viable cells, preserving nucleic acid integrity and allowing
to obtain genetic material compatible with downstream molecular
assays such as RNA gene expression analysis.
 Single-Cell Recovery of Metastatic Cells from Xenotransplanted Zebrafish 55

2 Materials

2.1 Reagents 1. Sterile PBS 1× (w/o Ca++ and Mg++).

2.1.1 Zebrafish Embryo 2. 0.4% Tricaine (ethyl 3-aminobenzoate methanesulfonate salt)
Euthanasia and solution in water.
Dissociation 3. 0.25% Trypsin-EDTA solution.
4. 100 mg/mL Collagenase Type I from Clostridium histolyticum
solution in PBS.
5. Fetal Bovine Serum (FBS).
6. Cell culture medium (e.g., DMEM, RPMI, etc.).

2.1.2 Single-Cell 1. Sterile PBS 1× (w/o Ca++ and Mg++).

Micromanipulation 2. 5% Bovine serum albumin (BSA) w/v solution in PBS.

2.2 Equipment 1. Surgical blade.

2.2.1 Zebrafish Embryo 2. Plastic Pasteur pipette (3 mL).
Euthanasia and 3. Micropipettes 2–20 μl, 20–200 μl, and 200–1000 μl.
Dissociation 4. Stereo microscope.
5. Heat block.

2.2.2 Single-Cell 1. Reaction tube (1.5 mL).

Micromanipulation 2. 0.2 mL thin-walled PCR tube, sterile.
3. Micropipette 0.2–2 μl.
4. Chambered cell culture slide, sterile.
5. Centrifuge capable of spinning down 0.2 mL PCR tubes.
6. Borosilicate thin wall capillaries, 1.0 mM OD, 0.78 mM ID,
75 mM.
7. Manual micromanipulator.
8. Mechanical microinjector.
9. Inverted fluorescence microscope.

3 Methods

The following protocol is designed to analyze individual cancer

cells (disseminated or metastatic cells) isolated from the tails of
five-day post-fertilization (dpf) embryos, although it could poten-
tially be applied for any cell found in the embryo’s body. Xeno-
transplanted tumor cells should be previously stained with
fluorescent lipid markers or transgenically engineered to express
fluorescent proteins (e.g., GFP and mCherry) in order to follow
56 Pablo Hurtado et al.

their behavior inside the fish body and allow an easy discrimination
between them and the host cells throughout the process.

3.1 Zebrafish 1. Prepare a 0.4% tricaine solution (anesthetic): dissolve 400 mg

Embryo Euthanasia of tricaine in 97.9 mL of ddH2O. Add 2.1 mL of Tris–HCl 1 M
and Dissection (see pH 9 solution. Adjust to pH 7.0. Keep the solution at 4 °C for
Note 1) short-term use and at -20 °C for long-term storage.
2. Dilute the 0.4% tricaine solution ten times in water to a final
concentration of 0.04% (tricaine overdose) and add 2 mL of the
solution to a small beaker.
3. Prepare a reaction tube (1.5 mL) adding 1 mL of PBS w/o Ca+
+
and Mg++.
4. Remove the xenotransplanted fish (≈15 zebrafish embryos at a
developmental stage of 5 days post fertilization) from the
incubation water using a 3 mL plastic Pasteur pipette (see
Note 2) and place them in the beaker containing the tricaine
solution to induce the dead by overdose of anesthetic.
5. Incubate the fish in this solution for 1–2 min, until they are
completely anesthetized (see Note 3).
6. Transfer one embryo at a time with the help of a plastic Pasteur
pipette to a clean dissection surface (e.g., microscopy glass
slide) and place it under a stereo microscope to proceed with
the dissection of the embryo’s body (see Note 4). Make sure to
add a few drops of the anesthetic solution to the fish to avoid
the sample from drying out (see Note 5).
7. With the help of fine-tip tweezers and a scalpel or razor blade,
orientate the fish as desired and physically separate the fish yolk
sac from the tail, where the cancer cells are found disseminated
(Fig. 1, step 1, and box) (see Note 6).
8. Cut the tail into smaller pieces in order to improve the diges-
tion efficiency. Collect all tail pieces from the dissection surface
and transfer it to the PBS-containing reaction tube for
washing.
9. Repeat this process with the rest of fish and sequentially transfer
the dissected tails to the reaction tube.
10. Invert the tube a few times to ensure that the tissues are washed
(Fig. 1, step 2).
11. With the help of a micropipette (200–1000 μl), carefully
remove the PBS from the tube avoiding the collection of the
embryo’s pieces (Fig. 1, step 3).
12. Repeat the washing step once more.
 Single-Cell Recovery of Metastatic Cells from Xenotransplanted Zebrafish 57

Fig. 1 Schematic representation of the protocol for zebrafish embryo dissociation. 1. Zebrafish embryo
dissection. 2. Washing of dissected tissue. 3–4. Enzymatic and mechanical digestion of the tissues.
5. Stopping the enzymatic reaction and pelleting of the cells. 6. Obtaining the single cell suspension

3.2 Zebrafish 1. Dissolve the collagenase in PBS at a final concentration of

Embryo Enzymatic and 100 mg/mL. This solution can be stored at -20 °C in small
Mechanical Digestion aliquots for later use.
2. Prepare 500 μl of the dissociation solution by mixing 20 μl of
collagenase (stock 100 mg/mL) and 480 μl of 0.25% Trypsi-
n-EDTA solution in a tube and keep it at 37 °C in a heat block.
3. Add the 500 μl of dissociation solution previously heated to the
reaction tube containing the tissues and transfer it to a heat
block at 37 °C (Fig. 1, step 3) (see Note 7).
4. With the help of a 200–1000 μl micropipette, every 2 min
pipette repeatedly up and down for a few times and vortex
58 Pablo Hurtado et al.

the sample to dislodge the cells. Transfer the tube to the heat
block in between mechanical processing (Fig. 1, step 4) (see
Note 8).
5. Incubate the sample for no longer than 10 min at 37 °C.
6. Prepare 5 mL of cell culture medium with 10% FBS by mixing
in a sterile tube 4.5 mL of medium and 500 μl of FBS.
7. Once the incubation time is finished, add 800 μl of cell culture
medium with 10% of FBS with a micropipette to neutralize the
action of the dissociation solution and avoid cell damage
(Fig. 1, step 5).
8. Transfer the tube to a bench-top centrifuge and spin down the
cells by centrifuging at 700 g for 5 min at room temperature
(Fig. 1, step 5).
9. Carefully remove the supernatant and add 50 μl of PBS + 2%
FBS to the cellular pellet and dislodge it with a micropipette
(Fig. 1, step 6). Keep the cell suspension at room temperature
and avoid putting them in ice.
10. Optional step. A DNA stain for live cells (e.g., Hoechst 33342)
can be added to the cell suspension to visualize the nuclear
morphology and integrity. For use, follow the manufacturer’s
instructions.
11. Visualize cells under the microscope to check for the presence
of fluorescent cells (tumor cells) and their morphology.

3.3 Preparation of 1. Prepare a 5% BSA w/v solution. Weigh 500 mg of BSA and
Collection Chamber for dissolve it in 10 mL of 1× PBS. Mix well until the BSA is fully
a Cellular Suspension dissolved. Filter the solution with a 0.22 μm pore. Add 100 μl
of the 5% BSA solution at every well of a sterile chambered cell
culture slide and remove it quickly (see Note 9).
2. Differentiate within the slide at least two chambers, one as the
picking field and the other one (or others) as the sample field
(Fig. 2, step 1).
3. Add 200 μl of PBS in each picking field (see Note 10).
4. Add 190 μl of PBS in the sample field plus 10 μl of the cell
suspension (Fig. 2, step 1). It is important that the maximum
concentration of cells in the chamber is not superior to 1 × 103
cells/chamber. It will ensure an efficient single-cell isolation
without contaminating cells being carried over (Fig. 3) (see
Note 11).
5. Place the chambered cell culture slide in the fluorescence
microscope specimen holder.
6. Allow cells to set at the bottom of the sample field for a few
minutes.
 Single-Cell Recovery of Metastatic Cells from Xenotransplanted Zebrafish 59

Fig. 2 Scheme depicting the main steps of the process of micromanipulation

Fig. 3 Requirement of cellular density at the sample field for single-cell micromanipulation. Representative
images of a too-high cellular density (a) and an adequate cellular density (b) at sample field

7. Check under the microscope that all cells are sitting at the
bottom of the slide (see Note 12).

3.4 Capillary 1. Add 100 μl of FBS to a sterile 1.5 mL reaction tube.

Preparation for 2. Immerse the capillary inside the reaction tube and let the FBS
Micromanipulation to enter the tip by capillarity.
3. Insert the capillary into the capillary holder of the microinjec-
tor, and place it onto the mechanical arm of the micromanipu-
lator (Fig. 4) (see Note 13).
60 Pablo Hurtado et al.

Fig. 4 Equipment setup for single-cell micromanipulation. This protocol is described for the use of a
mechanical microinjector and a manual micromanipulator, as shown in the images

4. With the micromanipulator control knobs, immerse the capil-

lary into the sample field and bring it to the bottom, softly
touching the surface of the slide (see Note 14). For easy visuali-
zation, use the 10× objective of the microscope; this will allow
an efficient focus of the capillary.
5. With the capillary inside the sample field, expulse a small vol-
ume of FBS from the capillary and absorb few microliters of
PBS, making use of the mechanical microinjector control knob.
Always make sure that the capillary is never void of any liquid
inside.

3.5 Cell Picking Before beginning with the micromanipulation step, make sure all
the necessary setup for the micromanipulation is previously done
(Fig. 4).
1. Add 1 μl of PBS to the lid of a 0.2 mL sterile PCR tube (see
Note 15). This tube will be used to collect the retrieved cell
from the cell suspension.
2. Use a 20× microscope objective to focus the capillary and for
the aspiration of the cells from the sample field.
3. Locate the cell of interest within the sample field (see Note 16)
and carefully approach the capillary to it making use of the
control knobs. Place the tip of the capillary opposite to the
cell of interest taking the precaution of not pushing or flushing
it away.
4. Clean the area surrounding the cell of interest by flushing away
the contaminating cells with the capillary using the microinjec-
tor control knob. This can be easily done by aspirating and
flushing PBS with the capillary, which will wash away the
possible contamination.
 Single-Cell Recovery of Metastatic Cells from Xenotransplanted Zebrafish 61

5. Proceed to aspirate the cell into the capillary. The disappear-

ance of the cell from the vision field should be verified before
retrieving the capillary.
6. Move the capillary away from the sample field, transfer it into
the picking field, and release the cell (Fig. 2, step 2) (see Note
17).
7. Check under the microscope the presence and purity of the
cell. In many occasions, the cell of interest is still accompanied
by contaminating material carried over from the sample field
(i.e., non-wanted zebrafish cells or debris that could interfere
with the downstream molecular analysis).
8. In the event that the cell is surrounded by contaminating
material, clean the surroundings as indicated in point 4 to
ensure that the cell remains alone (see Note 18).
9. Completely remove the capillary from the picking field.
10. Visually check for the integrity of the cell and the purity of the
area surrounding it. Take a second photograph of the cell.
11. Slowly rest the tip of a 0.2–2 μl micropipette on the surface of
the slide, next to the cell. Carefully retrieve the cell with the
micropipette previously set up to aspirate 1 μl (Fig. 2, step 3)
(see Note 19).
12. Transfer the 1 μl containing the cell to the lid of the 0.2 mL
PCR tube previously prepared by immersing the tip in the
liquid and releasing the cell inside (Fig. 2, step 3) (see Note
20).
13. Place the tube’s lid on the fluorescence microscope and check
for the presence of the cell (see Note 21).
14. Once the presence of the cell is confirmed, close the tube lid
and spin it down for a short time to bring the liquid to the
bottom of the tube.
15. Store the tube at -80 °C to preserve the integrity of the
genetic material or alternatively store it under the desired con-
ditions for downstream analysis.

4 Notes

1. It is convenient prior to the euthanasia and dissection of the

xenotransplanted embryos to verify under a fluorescence
microscope the presence of the desired tumor cell population
in the fish tails as well as to locate them within the tails. This
would help to minimize the size of the dissected tissue for a
more efficient digestion.
62 Pablo Hurtado et al.

2. Alternatively, a 200 μl micropipette tip can be cut off to collect

the embryos. Cut the tip wide enough to avoid damaging the
fish when manipulating them.
3. The addition of fish to the anesthetic solution should be done
sequentially. Avoid exposing the fish long times to the anesthe-
sia to preserve the viability of the cells.
4. The stereo microscope allows for a higher magnification and
better view and definition of the area of the fish to dissect.
5. Dry zebrafish tissue becomes very sticky and difficult to remove
from the glass slide surface.
6. This protocol is designed for the isolation of tumor cells
disseminated from the yolk sack or duct of Cuvier (injection
sites) to the tail of 5 days post-fertilization (dpf) embryos
(metastatic cells).
7. Alternatively, a water bath or incubator set at 37 °C can be used
for the incubation.
8. This procedure allows for a better enzymatic and mechanical
disassociation of the embryo’s tissues.
9. BSA helps to avoid cell attachment to the surface of the slides.
10. The volume in the chamber can be modified for comfort, but it
is convenient to have enough volume inside the chamber.
11. The total number of cells will depend on the size of the
digested tissue, the efficiency of the digestion, and the volume
in which the cell pellet is resuspended.
12. Make sure at this point that the dilution of the cells is adequate
for cell isolation. Dilute the sample if needed.
13. This protocol is described for the use of a manual micromanip-
ulator and a mechanical microinjector (Fig. 4), although auto-
mated versions could be used. Follow the manufacturer’s
instructions for the setup of the equipment.
14. As glass capillaries are flexible, they do not break when softly
touch the surface of the slide.
15. Instead of PBS, cell lysis buffer or another desired storage
buffer can be used at this point.
16. It is advisable to take a photograph of the cell before picking it
to contrast it with a photograph taken later in the process of
manipulation. It will allow to assess any possible damage
exerted on the cell or change on morphology.
17. Release only a small volume of PBS into the picking field to
avoid flushing the retrieved cell away from the field of view.
18. On occasions, the cell of interest can be attached to other cells
due to a lack of efficiency during tissue digestion. The capillary
 Single-Cell Recovery of Metastatic Cells from Xenotransplanted Zebrafish 63

can be used as a blade to cut off possible unions with other

cells.
19. A micropipette is used to retrieve the cell instead of the capil-
lary to minimize the risk of contamination, as undesired cells or
other material could be trapped inside the capillary. An alterna-
tive to the use of the micropipette would be the use of a new
clean capillary. However, removing the old capillary and attach-
ing a new one for every cell to be collected will substantially
increase the duration of the procedure, compromising the
quality of the biological material.
20. Immersing the tip in the liquid guarantees the efficient transfer
of the cell to the collection tube.
21. Placing the liquid in the lid of the tube, rather than at the
bottom, facilitates the visualization of the cell under the
microscope.

Acknowledgments

This work was supported by Roche-Chus Joint Unit (IN853B

2018/03), funded by Axencia Galega de Innovación (GAIN),
Consellerı́a de Economı́a, Emprego e Industria. I.M.-P. is funded
by the Training Program for Academic Staff fellowship (FPU16/
01018), from the Ministry of Education and Vocational Training,
Spanish Government.

References

1. Bystricky B, Mego M (2016) Circulating tumor Liverani C, Bongiovanni A, Spadazzi C, de

cells in breast cancer patients. Neoplasma 63(1): Vita A, van der Pluijm G, Giorgini A,
18–29. https://doi.org/10.4149/neo_2016_ Biagini R, Amadori D, Ibrahim T, Snaar-Jagalska
003 E (2016) Development of a patient-derived
2. Gomez-Cuadrado L, Tracey N, Ma R, Qian B, xenograft (PDX) of breast cancer bone metasta-
Brunton VG (2017) Mouse models of metasta- sis in a Zebrafish model. Int J Mol Sci 17(8).
sis: progress and prospects. Dis Model Mech https://doi.org/10.3390/ijms17081375
10(9):1061–1074. https://doi.org/10.1242/ 5. Teng Y, Xie X, Walker S, White DT, Mumm JS,
dmm.030403 Cowell JK (2013) Evaluating human cancer cell
3. Hill D, Chen L, Snaar-Jagalska E, Chaudhry B metastasis in zebrafish. BMC Cancer 13:453.
(2018) Embryonic zebrafish xenograft assay of https://doi.org/10.1186/1471-2407-13-453
human cancer metastasis. F1000Res 7:1682. 6. Saadatpour A, Lai S, Guo G, Yuan GC (2015)
https://doi.org/10.12688/f1000research. Single-cell analysis in cancer genomics. Trends
16659.2 Genet 31(10):576–586. https://doi.org/10.
4. Mercatali L, La Manna F, Groenewoud A, 1016/j.tig.2015.07.003
Casadei R, Recine F, Miserocchi G, Pieri F,
 Chapter 5

Isolation of Single Circulating Tumor Cells Using VyCAP

Puncher System
Thais Pereira-Veiga, Bianca Behrens, Joska J. Broekmaat, Lisa Oomens,
Michiel Stevens, Arjan G. J. Tibbe, Nikolas Stoecklein,
Laura Muinelo-Romay, Roberto Piñeiro, and Clotilde Costa

Abstract
Tumor heterogeneity has a major role in the development of tumor evasion and resistance to treatments. To
study and understand the intrinsic heterogeneity of cancer cells, the use of single-cell isolation technology
has had a major boost in recent years, gaining ground to bulk analysis in the study of solid tumors. In the
liquid biopsy field, the use of technologies for single-cell analysis has represented a major advance in the
study of the heterogeneity of circulating tumor cells (CTCs), providing relevant information about therapy-
resistant CTCs. However, single-cell analysis of CTCs is still challenging due to the weakness and scarcity of
these cells. In this chapter, we describe a protocol for CTCs isolation at a single-cell level using the VyCAP
Puncher system.

Key words CTC, Liquid biopsy, VyCAP, Puncher, Single cell, Sequencing, CellSearch, Tumor
heterogeneity

1 Introduction

Circulating tumor cells (CTCs) have been shown of high clinical

relevance, being directly involved in the metastatic process
[1]. Although they represent a unique opportunity to study cancer
metastasis and progression, they are at a low rate compared with
other blood cells. In recent years, a range of technologies for the
enrichment of CTCs has emerged to overcome this issue. Cell-
Search® is the only FDA-approved method for CTCs quantifica-
tion while Parsortix® has been cleared for use in metastatic breast
cancer patients for the capture and harvest of CTCs from whole
blood. This enumeration had proved to have prognostic value in
metastatic breast, prostate, and colorectal cancer patients [2]. Still,
CTC numbers offer limited information and do not allow for

Miodrag Gužvić (ed.), Single Cell Analysis: Methods and Protocols, Methods in Molecular Biology, vol. 2752,
https://doi.org/10.1007/978-1-0716-3621-3_5, © Springer Science+Business Media, LLC, part of Springer Nature 2024

65
66 Thais Pereira-Veiga et al.

physician intervention; therefore, the current trend is to analyze

genetic data from CTCs [3].
The strategies for CTCs isolation have limitations since specific
markers for CTCs have not yet been identified; there is no a clear
consensus about the definition of a CTC and the sensitivity of some
technologies is still low for addressing their molecular characteriza-
tion. It should be noted that currently, no system is directly capable
of isolating a population of pure CTCs. Thus, the available tech-
nologies isolate some normal blood cells from the hematopoietic
fraction in a non-specific way together with the CTCs. Therefore, a
combination of different isolation strategies is necessary, including
a first step of CTC enrichment and subsequent phenotypic charac-
terization of the isolated cells to determine their tumor origin.
Single-cell analysis is the best choice for an in-depth analysis of
heterogeneity studies as well as to avoid biased results due to
unspecific cells. Different technologies are currently in the market
for CTC isolation at a single-cell level. DEPArray [4, 5], CELLCE-
LECTOR™ [6], or VyCAP Puncher [7] are the most widely used.
Here, we describe the VyCAP Puncher protocol for CTCs isolation
and subsequent sequencing analysis. Here, we describe the VyCAP
Puncher protocol for CTCs isolation and subsequent sequencing
analysis. The starting specimens are CTC pre-enriched samples in
CellSearch® MAGNEST® cartridges, from the blood of metastatic
cancer patients. The CellSearch® system uses whole blood and
captures CTCs with magnetic beads, based on the positive expres-
sion of EpCAM. Subsequently, it performs immunofluorescence
staining of the enriched fraction of cells to identify those that are
positive for cytokeratins and negative for CD45, in combination
with round/oval morphology and nucleus/cytoplasm localization.
The current CellSearch® system defines a CTC as a positive event
when it has a well-defined nucleus (positive DAPI staining),
expresses cytokeratin (CK8, CK18, and CK19), does not express
CD45, and is larger than 4 × 4 μm in size.
VyCAP technology separates single cells into single microwells
after which the cells can be imaged and isolated on a Puncher
system. The system automatically creates images of all cells and
allows enumeration of CTCs. Besides, it gives the option of isolat-
ing individual cells from a CTC pre-enriched fraction that is com-
patible with other current technologies (CellSearch® system,
Parsortix, RosetteSep, etc.) (Fig. 1). The immunofluorescence
staining performed previously in the cells allows the recognition
of CTCs in the VyCAP Puncher software. For the isolation of
individual cells, the system is composed of two essential parts: a
disposable with microwell chip that is used to distribute individual
cells in a microwells and a Puncher system that automatically selects
and isolates cells. There is no limitation regarding the number of
cells that the user can isolate.
 Isolation of Single Circulating Tumor Cells Using VyCAP Puncher System 67

Cell
Label and
suspension
enrich CTC

CellSearch Puncher Single cells for

Pump-unit DNA/RNA analysis

Chip with 6400 microwells

cell suspension
Automated
fluorescent imaging
and selection of
CTCs
Captured single cells

Fig. 1 Workflow scheme for isolation of single CTCs using VyCAP Puncher from a pre-enriched sample by
Cellsearch technology

2 Materials

PBS 1× filtered (0.45 μm).

Plastic tubes (1.5 mL).
Long glass Pasteur pipettes or gel loading pipette tips.
Filtered ethanol (0.45 μm).
Mineral oil, embryo culture tested.
Sterile 0.2 mL thin-walled tube, classic semi-domed cap, regu-
lar profile.
Centrifuge capable of spinning down 0.2 mL PCR tubes.

3 Methods

3.1 Recovering 1. Remove the cartridge from the MAGNEST® cartridge holder
MAGNEST® Cartridge after scanning (see Note 1).
Content 2. Remove the cartridge plug by pushing it upwards from the
back of the cartridge (see Note 2).
3. Homogenize the sample inside the cartridge by pipetting it up
and down inside the cartridge at least 10 times, using a Pasteur
pipette or a gel-loading pipette tip. This procedure will detach
the cells from the cartridge.
4. Transfer the content of the cartridge to a 1.5 mL tube.
5. Using a pipette, add 300 μl of filtered PBS 1× inside the
cartridge by leaning the tip against the front side of the emptied
cartridge (glass side where the barcode is located). Pipette this
68 Thais Pereira-Veiga et al.

fluid up and down for ten times, rinsing the surface. The
removal of the cells can be checked by placing the cartridge
glass slide up under a standard upright fluorescence
microscope.
6. Add the 300 μl of PBS wash into the same 1.5 mL tube that the
sample was previously transferred to. The final sample volume
in the 1.5 mL tube is 600 μl.

3.2 Cell Seeding 1. Pre-wet the VyCAP microwell chip by adding 1 mL of filtered
100% ethanol on top of it. Let the chip sit with the ethanol for
45 min.
2. Connect a disposable filtration unit to the pump and insert the
microwell chip in the slot.
3. Add 1 mL of filtered PBS 1× to the remaining ethanol that is on
top of microwell chip.
4. Set the pressure of the pump station to 10 mbar (see Note 3).
5. Switch on the pump and pull most of the liquid through.
Switch off the pump action when there is only little amount
of fluid left. It is utterly important not to remove all fluid as this
will lead to air entering the microwells, which will limit the
number of available microwells.
6. Rinse the microwell chip by adding another 1 mL of filtered
PBS 1× to the chip and pulling most of the liquid through.
Repeating this step several times will completely remove the
ethanol.
7. Once the microwell chip is rinsed, add 1 mL of filtered PBS 1×
to the system and switch on the pump unit.
8. As soon as a drop of PBS has passed the microwells, add the
600 μl of the sample recovered from the MAGNEST® car-
tridge (see Note 4).
9. Switch off the pump after the whole volume of the sample has
passed.
10. Remove the microwell chip from the filtration station.
11. Wash the back side of the microwell chip with filtered PBS 1×.
12. Dry the back side of the microwell chip gently with a soft tissue
without directly touching the surface of the chip.

3.3 Isolation of CTCs 1. Transfer the microwell chip into the Puncher system to acquire
with the Punching fluorescence images.
System 2. The Puncher system contains the VyCAP Imaging system, an
automated image acquisition software. This system has auto
focus and an automatic filter cube changer that can hold a
maximum of six different filter cubes.
 Isolation of Single Circulating Tumor Cells Using VyCAP Puncher System 69

3. After image acquisition, the software presents a pre-selection of

the cells of interest, based on the fluorescent cell signature.
4. Pipette 35 μl of mineral oil in the cap of the 0.2 mL tube and
place it in the sample holder on the puncher device.
5. Bend the cap so that its top is at the same height as the opening
of tube.
6. Select the tube collection format and align the needle to punch
into the cap of the tube.
7. Select the desired cell to be punched into the cap of the 0.2 mL
tube (see Note 5).
8. Close the tube upside down, with the sample liquid still at the
cap, and vortex the tube upside down for 5 s.
9. Immediately spin down at 10,000 g for 1 min.
10. Add 1 μl of sterile 1× PBS to every sample. Spin down at
10,000 g for 1 min (see Note 6).
11. Sample can now be stored at 4 °C for up to 1 week prior to
quality control and genome wide amplification.

4 Notes

1. Ideally, proceed with this step immediately after scanning with

CellSearch®. Alternatively, the CellSearch® cartridges can be
stored in darkness, in horizontal position and at 4 °C to be used
up to a year after.
2. It is difficult to reach behind the plastic holder to be able to
push the plug out of the cartridge. Use a tool, such as a pair of
tweezers to push the plug out.
3. Depending on the sample, the pressure can be set higher, but a
higher pressure will induce more stress on the cells.
4. The flow rate for PBS without cells is 1–2 mL/min at 10 mbar.
As soon as more pores will become occupied by cells, the flow
rate will drop. In case all wells are filled, the flow rate will
approach zero.
5. After punching, an image of the punched microwell is acquired
to ensure that the microwell bottom and cell are removed.
6. Steps 8–10 must be performed within 10 min after step 7.
7. The number of microwells is 6400. Maximum number of cells
that you can add is depending on the size of the cells. The
diameter of the exit pore of the microwells is 5 microns. Cells
that are flexible or that have a small diameter will pass through
the pore. If the sample only contains large rigged cells that are
unable to pass through the 5 μm pore, the maximum number
of cells in the sample volume is 6400.
70 Thais Pereira-Veiga et al.

Acknowledgement

This work was supported by Roche-Chus Joint Unit (IN853B

2018/03), funded by Axencia Galega de Innovación (GAIN),
Consellerı́a de Economı́a, Emprego e Industria.

References
1. Massague J, Obenauf AC (2016) Metastatic col- 5. Reduzzi C, Motta R, Bertolini G, Miodini P,
onization by circulating tumour cells. Nature Martinetti A, Sottotetti E, Daidone MG, Cap-
529(7586):298–306. https://doi.org/10. pelletti V (2017) Development of a protocol for
1038/nature17038 single-cell analysis of circulating tumor cells in
2. Cristofanilli M, Budd GT, Ellis MJ, Stopeck A, patients with solid tumors. Adv Exp Med Biol
Matera J, Miller MC, Reuben JM, Doyle GV, 994:83–103. https://doi.org/10.1007/978-3-
Allard WJ, Terstappen LW, Hayes DF (2004) 319-55947-6_4
Circulating tumor cells, disease progression, 6. Neumann MH, Schneck H, Decker Y,
and survival in metastatic breast cancer. N Engl Schomer S, Franken A, Endris V, Pfarr N,
J Med 351(8):781–791. https://doi.org/10. Weichert W, Niederacher D, Fehm T, Neubauer
1056/NEJMoa040766 H (2017) Isolation and characterization of cir-
3. Keller L, Pantel K (2019) Unravelling tumour culating tumor cells using a novel workflow
heterogeneity by single-cell profiling of circulat- combining the CellSearch ((R)) system and the
ing tumour cells. Nat Rev Cancer 19(10): CellCelector(). Biotechnol Prog 33(1):
553–567. https://doi.org/10.1038/s41568- 125–132. https://doi.org/10.1002/btpr.2294
019-0180-2 7. Stevens M, Oomens L, Broekmaat J, Weersink J,
4. Di Trapani M, Manaresi N, Medoro G (2018) Abali F, Swennenhuis J, Tibbe A (2018)
DEPArray system: an automatic image-based VyCAP’s puncher technology for single cell
sorter for isolation of pure circulating tumor identification, isolation, and analysis. Cytometry
cells. Cytometry A 93(12):1260–1266. A 93(12):1255–1259. https://doi.org/10.
https://doi.org/10.1002/cyto.a.23687 1002/cyto.a.23631
 Chapter 6

Simultaneous Isolation and Amplification of mRNA

and Genomic DNA of a Single Cell
Miodrag Gužvić

Abstract
Many biological or pathological processes are driven by cells difficult to identify or isolate, i.e., rare cells.
Very often, these cells have elusive biology. Therefore, their detailed characterization is of utmost impor-
tance. There are many approaches that allow analysis of few or even many targets within one class of
biomacromolecules/analytes (e.g., DNA, RNA, proteins, etc.) in single cells. However, due to rarity of the
cells of interest, there is a great need to comprehensively analyze multiple analytes within these cells, in other
words to perform multi-omics analysis. In this chapter, I describe a method to isolate, separate, and amplify
total mRNA and genomic DNA of a single cells, using whole transcriptome (WTA) and whole genome
amplification (WGA). These WTA and WGA products enable simultaneous analysis of transcriptome and
genome of a single cell using various downstream high-throughput approaches.

Key words Single-cell analysis, Whole genome amplification, Whole transcriptome analysis, Multi-
omics analysis, Rare cells

1 Introduction

Analysis of single cells is becoming a routine approach in biomedi-

cal laboratory. In some cases, this type of analysis is enabling is a
deeper and more detailed insight into the biological phenomena
that we are already familiar with (e.g., comprehensive single-cell
transcriptome sequencing enabled unprecedented insight into the
heterogeneity of bone marrow populations [1]). In other cases,
single-cell analysis is a necessity to understand the mechanisms
driven by rare cells, whose biology is largely unknown (e.g., biology
of disseminated cancer cells that are founder cells of lethal metasta-
sis [2]). Analyzing individual aspects of the biology of these rare
cells (e.g., genome or transcriptome) is often not enough to paint a
complete molecular profile of these cells. Therefore, there is a need
to develop and use multiple-omics approaches to analyze single-cell
biology [3]. These approaches enable us to comprehensively and
simultaneously profile various biomacromolecules (e.g., RNA,

Miodrag Gužvić (ed.), Single Cell Analysis: Methods and Protocols, Methods in Molecular Biology, vol. 2752,
https://doi.org/10.1007/978-1-0716-3621-3_6, © Springer Science+Business Media, LLC, part of Springer Nature 2024

71
72 Miodrag Gužvić

DNA, proteins) or their various functional and structural states

(e.g., methylome, exome, chromatin states) [3]. In this chapter, I
present a method to isolate and amplify total mRNA and genomic
DNA of a single cell. The procedure has two major parts, whole
transcriptome amplification (WTA; nota bene: the name of the
method traditionally contains the term “transcriptome” although
it only captures mRNA, and not other expressed RNA species), and
whole genome amplification (WGA). Both methods have been
developed to profile disseminated cancer cells [4–6] in humans,
but have also been used for other types of human [7, 8] or murine
cells [9–11]. Generated WGA and WTA products are quite versatile
and can be further analyzed by various methods, like expression
microarrays [12], array-CGH [2, 13, 14], end-point PCR [2] and
qPCR [15, 16], genome and transcriptome sequencing [17], etc.
WTA is based on the capture of mRNAs on the solid phase [6, 12],
where reverse transcription (RT) takes place, followed by second
cDNA strand synthesis and global amplification (Fig. 1a). WGA is
based on fragmentation of the genome using a restriction enzyme
[4, 18], ligation of adapters, which enable generation of binding
sites for primers that drive the final step of global amplification
(Fig. 1b).

2 Materials

2.1 Isolation of Ideally, this procedure should be performed under laminar flow
mRNA and Genomic bench in a room where no DNA amplification takes place and no
DNA, and Whole amplified DNA is stored. During execution of the protocol, gloves
Transcriptome should be exchanged frequently, especially upon touching anything
Amplification that is outside of the bench (face, chair, drawer handles, etc.). It is
advised not to wear wristwatches, rings, bracelets, etc.

2.1.1 Reagents and Kits All reagents should be molecular biology grade, and DNase and
RNAse free. All dilutions and mixtures should be made with
nuclease-free water. Reagent dilutions made from powder should
be filtered using 0.22 μm filter. All dilutions should be aliquoted to
avoid frequent freeze-thaw cycles.
. mTRAP kit (Active Motif), containing lysis buffer, protease,
streptavidin-coated beads, and biotinylated poly-T gripNA
Probe (peptide nucleic acid, PNA) oligonucleotides. Beads are
stored at 4 °C, the rest is stored at -20 °C. Individual reagents
should be prepared according to manufacturer’s instructions
(lysis buffer and beads are ready to use, protease and PNAs are
lyophilized and need to be dissolved).
. Nuclease-free water.
. Linear polyacrylamide carrier (e.g., AM9520 from Ambion, see
Note 1). Stored at 4 °C.
 Single-Cell Genome and Transcriptome Amplification 73

Fig. 1 Schematic representation of WTA procedure (a) and WGA procedure (b). Please note that the
oligonucleotide sequences shown do not have to match those presented in the protocol; the figure serves
to display the principles of individual steps
74 Miodrag Gužvić

Fig. 1 (continued)
 Single-Cell Genome and Transcriptome Amplification 75

. SuperScript II reverse transcriptase (SSIIRT) kit (Thermo) con-

taining 5× RT buffer, 0.1 M DTT, and SS II reverse transcriptase
(RT). Stored at -20 °C (see Note 2).
. 10% IGEPAL CA-630 in nuclease-free water. Stored at -20 °C.
. dNTP mix in nuclease-free water, 10 mM each. Store at -20 °C.
. Oligonucleotide N8: 5′-CCC CCC CCC CCC CCC GTC TAG
ANN NNN NNN-3′ (N = A/C/T/G)—mixture of oligonu-
cleotides containing random octamer sequences. 200 mM dilu-
tion. Store at -20 °C (see Note 3).
. Oligonucleotide T24: 5′- CCC CCC CCC CCC CCC GTC
TAG ACT TGA GTT TTT TTT TTT TTT TTT TTT TTT
TVN-3′ (V = A/C/G; N = A/C/T/G)—mixture of oligonu-
cleotides containing random 3′ termini. 100 mM dilution. Store
at -20 °C (see Note 4).
. IGEPAL-wash buffer: 50 mM Tris–HCl, pH 8.0, 75 mM KCl,
10 mM DTT, 0.25% IGEPAL CA-630. Store at -20 °C.
. Tween-wash buffer: 50 mM Tris–HCl, pH 8.0, 75 mM KCl,
10 mM DTT, 0.50% Tween 20. Store at -20 °C.
. Ethanol, absolute.
. MgCl2, 40 mM.
. KH2PO4, 200 mM. Store at -20 °C.
. DTT, 1 mM (dilute 0.1 M DTT from SSIIRT kit).
. dGTP, 2 mM. Store at -20 °C.
. Terminal deoxynucleotidyl Transferase (TdT; e.g., 72,033 from
Affymetrix, see Note 5).
. Mineral oil (e.g., M1028 from Sigma).
. Tailing-wash buffer: 50 mM KH2PO4 pH 7, 1 mM DTT, 0.25
IGEPAL CA-630. Store at -20 °C.
. Expand Long Template PCR System (ELTPS) kit (Roche) con-
taining Buffer 1 and PolMix. Stored at -20 °C (see Note 6).
. Formamide, 20%. Stored at -20 °C.
. Oligonucleotide CP2: 5′- TCA GAA TTC ATG CCC CCC
CCC CCC CCC-3′. 24 mM dilution Store at -20 °C (see
Note 7).

2.1.2 Plasticware Plasticware in contact with sample should have low nucleic acid
binding properties.
. 0.2 mL domed cap microtubes.
. Microtubes (e.g., 0.5, 1.5, or 2.0 mL) for preparing reagent
mixes.
. 10 μl, 20 μl, 200 μl, and 1000 μl filter-tips.
. 50 mL screw-cap tube.
76 Miodrag Gužvić

2.1.3 Equipment . Laminar flow bench.

. Microtube racks.
. Microtube ice racks (optional).
. 0.5–2 μl, 10 μl, 2–20 μl, 20–200 μl, 200–1000 μl micropipettes
(recommended).
. Vortexer.
. Benchtop microcentrifuge for microtubes listed above.
. Thermal cycler.
. Tube roller.
. Hybridization oven (see Note 8).
. Magnetic racks (e.g., DynaMag-96 Side Magnet from Thermo).

2.1.4 Cycling Programs Ideally, dedicated thermal cycler used only for WTA and WGA
should be used, and it should be in the same room. Cycling pro-
grams are shown in Table 1.

2.1.5 Reagent Mixes For buffers, final concentrations of individual components are
given. For reagent mixes, volumes of individual reagents sufficient
for one sample are given; it is recommended to prepare 5–10%
more of the total reagent mix volume than needed for desired
number of samples, to account for pipette errors and “dead”
volumes. Reagent mixes composition are given in Table 2.

2.2 DNA Ideally, this procedure should be performed in the same room and
Precipitation and using the same equipment as for the WTA (see Subheading 2.1).
Whole Genome During execution of the protocol, gloves should be exchanged
Amplification frequently, especially upon touching anything that is outside of
the bench (face, chair, drawer handles, etc.). It is advised not to
wear wrist watches, rings, bracelets, etc.

2.2.1 Reagents and Kits All reagents should be molecular biology grade, and DNase and
RNAse free. All dilutions and mixtures should be made with
nuclease-free water. Reagent dilutions made from powder should
be filtered using 0.22 μm filter. All dilutions should be aliquoted to
avoid frequent freeze-thaw cycles.
. Nuclease-free water.
. 70% ethanol.
. WGA buffer pH 7.5: 100 mM Tris-acetate pH 7.8, 100 mM Mg
(CH3COO)2, 500 mM CH3COOK (see Note 9). Store at -
20 °C.
. 10% Tween 20. Store at -20 °C.
. 10% IGEPAL CA-630. Store at -20 °C.
 Single-Cell Genome and Transcriptome Amplification 77

Table 1
Cycling programs used in Subheading 3.1

Steps Time/temperature Repetition

LYSIS
Step 1 10 min/45 °C
1×
Step 2 1 min/74 °C 1×
Step 3 10 min/22 °C 1×
Step 4 1/22 °C
DENATURATION
Step 1 4 min/94 °C 1×
Step 2 1/94 °C 1×
TAILING
Step 1 60 min/37 °C 1×
Step 2 1/22 °C 1×
TINACTIVATE
Step 1 5 min/70 °C 1×
Step 2 1/4 °C
AMPLIFICATION
Step 1 30 s/78 °C 1×
Step 2 15 s/94 °C
Step 3 30 s 65 °C 20×
Step 4 2 min/68 °C
Step 5 15 s/94 °C
Step 6 30 s/65 °C 20×
Step 7 2 min 30 s/68 °C
+10 s/cycle
Step 8 15 s/94 °C
Step 9 30 s/65 °C 1×
Step 10 7 min/68 °C
Step 11 1/4 °C
1—forever

. 10 mg/mL Proteinase K (e.g., 3115828001 from Roche). Store

at -20 °C.
. MseI, 50 U/μl (e.g., R0525M from NEB; see Note 10) Store at
-20 °C.
. ddMse11 oligonucleotide: 5′-TAA CTG ACAG ddC-3′.
100 μM. Store at -20 °C.
. Lib1 oligonucleotide: 5′-AGT GGG ATT CCT GCT GTC
AGT-3′. 100 μM. Store at -20 °C.
78 Miodrag Gužvić

Table 2
Composition of reagent mixes used in Subheading 3.1

Reagent Volume [μl]

RT-mix 1
5× SSIIRT buffer 2.0
0.1 M DTT 1.0
10% IGEPAL CA-630 0.5
Nuclease-free water 0.5
100 μM T24 oligonucleotide 3.0
200 μM N8 oligonucleotide 3.0
RT-mix 2
5× SSIIRT buffer 2.0
0.1 M DTT 1.0
Nuclease-free water 5.0
dNTPs (10 mM each) 1.0
Add later (see step 21 of Subheading 3.1)
SSII RT enzyme 1.0
Tailing-mix
40 mM MgCl2 1.0
1 mM DTT 1.0
2 mM dGTP 1.0
0.2 KH2PO4 0.5
Nuclease-free water 6.5
PCR-mix 1
ELTPS buffer 1 4.0
20% formamide 7.5
Nuclease-free water 24.0
PCR-mix 2
24 μM CP2 oligonucleotide 2.5
dNTPs (10 mM each) 1.75
Add later (see step 39 of subheading 3.1)
ELTPS PolMix enzyme 1.5
 Single-Cell Genome and Transcriptome Amplification 79

. 10 mM ATP (e.g., 11140965001 from Roche). Store at -

20 °C.
. T4 DNA ligase, 5 U/μl (e.g., 10799009001 from Roche). Store
at -20 °C.
. dNTP mix, 10 mM each. Store at -20 °C.
. Expand Long Template PCR System (ELTPS) kit (Roche) con-
taining Buffer 1 and PolMix. Store at -20 °C.

2.2.2 Plasticware Plasticware in contact with sample should have low nucleic acid
binding properties.
. 10 μl, 20 μl, 200 μl, and 1000 μl filter-tips.
. 0.2 mL domed cap microtubes.
. Microtubes (e.g., 0.5, 1.5, or 2.0 mL) for preparing reagent
mixes.

2.2.3 Equipment . Laminar flow bench.

. Microtube racks.
. 0.5–2 μl, 10 μl, 2–20 μl, 20–200 μl, 200–1000 μl micropipettes
(recommended).
. High speed microcentrifuge with controlled temperature.
. Thermomixer.
. Thermal cycler.

2.2.4 Reagent Mixes For reagent mixes, volumes of individual reagents sufficient for one
sample are given; it is recommended to prepare 5–10% more of the
total reagent mix volume than needed for desired number of sam-
ples, to account for pipette errors and “dead” volumes. Reagent
mixes composition are given in Table 3.

2.2.5 Cycling Programs Ideally, dedicated thermal cycler used only for WTA and WGA
should be used, and it should be in the same room. Cycling pro-
grams are shown in Table 4.

2.3 Control of the This procedure should not be performed in the room where WTA
WTA Quality or WGA takes place, nor the same equipment, plasticware, or
reagents should be used. This procedure can be performed under
the same conditions as any other conventional PCR in your lab.

2.3.1 Reagents and Kits All reagents should be molecular biology grade, and DNase and
RNAse free. All dilutions and mixtures should be made with
nuclease-free water. Reagent dilutions made from powder should
be filtered using 0.22 μm filter. All dilutions should be aliquoted to
avoid frequent freeze-thaw cycles.
80 Miodrag Gužvić

Table 3
Composition of reagent mixes used in Subheading 3.2

Reagent Volume [μl]

Proteinase K digestion mix
WGA buffer 0.5
10% tween 20 0.13
10% IGEPAL CA-630 0.13
10 mg/mL Proteinase K 0.26
MseI digestion mix
MseI (50 U/μl) 0.25
Nuclease-free water 0.25
Adapter annealing mix
WGA buffer 0.5
100 μM Lib1 primer 0.5
100 μM ddMse11 primer 0.5
Nuclease-free water 1.5
Ligation mix
Adapter annealing mix 3.0
10 mM ATP 1.0
T4 DNA ligase (5 U/μl) 1.0
PCR-mix
ELTPS buffer 1 3.0
dNTPs (10 mM each) 2.0
ELTPS PolMix enzyme 1.0
Nuclease-free water 34.0

. Taq polymerase with appropriate buffer (e.g., FastStart kit

04738381001 from Roche, containing FastStart Taq polymer-
ase, PCR buffer with 20 mM MgCl2, and dNTP mix). Store at -
20 °C.
. BSA, 20 mg/mL. Store at -20 °C.
. dNTP mix, 10 mM each (optional, in case it is not contained in
Taq polymerase kit). Store at -20 °C.
. Nuclease-free water. Store at -20 °C.
. GAPDH forward primer: 5′- CCA TCT TCC AGG AGC GAG
AT-3′, 100 μM. Store at -20 °C.
 Single-Cell Genome and Transcriptome Amplification 81

Table 4
Cycling programs used in Subheading 3.2

Steps Time/temperature Repetition

PROTEINASE
Step 1 15 h/42 °C 1×
Step 2 10 min/80 °C 1×
Step 3 1/4 °C 1×
MSE1
Step 1 3 h/37 °C 1×
Step 2 5 min/65 °C 1×
Step 3 1/4 °C 1×
ANNEAL
Step 1 1 min/65 °C 50×
-1 °C/cycle
Step 2 1/15 °C 1×
LIGATION
Step 1 Overnight/15 °C 1×
(see Note 11)
Step 2 1/15 °C
AMPLIFICATION
Step 1 3 min/68 °C 1×
Step 2 40 s/94 °C
Step 3 30 s/57 °C 15×
Step 4 1 min 30 s/68 °C
+1 s/cycle
Step 5 40 s/94 °C
Step 6 30 s/57 °C
+1 °C/cycle 9×
Step 7 1 min 45 s/68 °C
+1 s/cycle
Step 8 40 s/94 °C
Step 9 30 s/65 °C 23×
Step 10 1 min 53 s/68 °C
+1 s/cycle
Step 11 3 min 40 s/68 °C
1×
Step 12 1/4 °C
1—forever
82 Miodrag Gužvić

. GAPDH reverse primer: 5′- CAG TGG GGA CAC GGA

AGG-3′, 100 μM. Store at -20 °C.
. ACTB forward primer: 5′- GCG TGA CAT TAA GGA GAA
GCT G-3′, 100 μM. Store at -20 °C.
. ACTB reverse primer: 5′- CGC TCA GGA GGA GCA ATG
AT-3′, 100 μM. Store at -20 °C.
. EEF1A1 forward primer: 5′- CTG TGT CGG GGT TGT AGC
CA-3′, 100 μM. Store at -20 °C.
. EEF1A1 reverse primer: 5′- TGC CCC AGG ACA CAG AGA
CT-3′, 100 μM. Store at -20 °C.

2.3.2 Plasticware Plasticware in contact with sample should have low nucleic acid
binding properties.
. 0.2 mL microtubes.
. Microtubes (e.g., 0.5, 1.5, or 2.0 mL) for preparing reagent
mixes.
. 10 μl, 20 μl, 200 μl, and 1000 μl filter-tips.

2.3.3 Equipment . PCR bench.

. Microtube racks.
. Microtube ice racks (optional).
. 0.5–2 μl, 10 μl, 2–20 μl, 20–200 μl, 200–1000 μl micropipettes
(recommended).
. Benchtop microcentrifuge for microtubes listed above.
. Thermal cycler.

2.3.4 Reagent Mixes For reagent mixes, volumes of individual reagents sufficient for one
sample are given; it is recommended to prepare 5–10% more of the
total reagent mix volume than needed for desired number of sam-
ples, to account for pipette errors and “dead” volumes. Reagent
mixes composition are given in Table 5.

2.3.5 Cycling Programs Cycling programs are shown in Table 6.

2.4 Control of the This procedure should not be performed in the room where WTA
WGA Quality or WGA takes place, nor the same equipment, plasticware, or
reagents should be used. This procedure can be performed under
the same conditions as any other conventional PCR in your lab.

2.4.1 Reagents and Kits All reagents should be molecular biology grade, and DNase and
RNAse free. All dilutions and mixtures should be made with
nuclease-free water. Reagent dilutions made from powder should
be filtered using 0.22 μm filter. All dilutions should be aliquoted to
avoid frequent freeze-thaw cycles.
 Single-Cell Genome and Transcriptome Amplification 83

Table 5
Composition of reagent mixes used in Subheading 3.3

Reagent Volume [μl]

Primer-mix
Each primer 8 (6 × 8 μl = 48 μl)
Nuclease-free water 152
PCR-mix
10× FastStart PCR buffer (with 20 mM MgCl2) 1.0
Primer-mix 1.0
20 mg/mL BSA 0.2
dNTPs (10 mM each) 0.2
FastStart Taq polymerase 0.1
Nuclease-free water 6.5

Table 6
Cycling programs used in Subheading 3.3

Steps Time/temperature Repetition

WTAQC
Step 1 4 min/95 °C 1×
Step 2 30 s/95 °C
Step 3 30 s/58 °C 32×
Step 4 1 min 30 s/72 °C
Step 5 7 min/72 °C 1×
Step 6 1/4 °C
1—forever

. Taq polymerase with appropriate buffer (e.g., FastStart kit

04738381001 from Roche, containing FastStart Taq polymer-
ase, PCR buffer with 20 mM MgCl2, and dNTP mix). Store at -
20 °C.
. 20 mg/mL BSA. Store at -20 °C.
. dNTP mix, 10 mM each (optional, in case it is not contained in
Taq polymerase kit). Store at -20 °C.
. Nuclease-free water. Store at -20 °C.
84 Miodrag Gužvić

. KRAS Forward primer: 5′-ATA AGG CCT GCT GAA AAT

GAC-3′, 100 μM. Store at -20 °C.
. KRAS Reverse primer: 5′-CTG AAT TAG CTG TAT CGT CAA
GG-3′, 100 μM. Store at -20 °C.
. D5S2117 Forward primer: 5′-CCA GGT GAG AAC CTA GTC
AG-3′, 100 μM. Store at -20 °C.
. D5S2117 Reverse primer: 5′-ACT GTG TCC TCC AAC CAT
GG-3′, 100 μM. Store at -20 °C.
. KRT19P1 Forward primer: 5′-GAA GAT CCG CGA CTG GTA
C-3′, 100 μM. Store at -20 °C.
. KRT19P1 Reverse primer: 5′-TTC ATG CTC AGC TGT GAC
TG-3′, 100 μM. Store at -20 °C.
. TP53 Forward primer: 5′-GAA GCG TCT CAT GCT GGA
TC-3′, 100 μM. Store at -20 °C.
. TP53 Reverse primer: 5′-CAG CCC AAC CCT TGT CCT
TA-3′, 100 μM. Store at -20 °C.

2.4.2 Plasticware Plasticware in contact with sample should have low nucleic acid
binding properties.
. 0.2 mL microtubes.
. Microtubes (e.g., 0.5, 1.5, or 2.0 mL) for preparing reagent
mixes.
. 10 μl, 20 μl, 200 μl, and 1000 μl filter-tips.

2.4.3 Equipment . PCR bench.

. Microtube racks.
. Microtube ice racks (optional).
. 0.5–2 μl, 10 μl, 2–20 μl, 20–200 μl, 200–1000 μl micropipettes
(recommended).
. Benchtop microcentrifuge for microtubes listed above.
. Thermal cycler.

2.4.4 Reagent Mixes For reagent mixes, volumes of individual reagents sufficient for one
sample are given; it is recommended to prepare 5–10% more of the
total reagent mix volume than needed for desired number of sam-
ples, to account for pipette errors and “dead” volumes. Reagent
mixes composition are given in Table 7.

2.4.5 Cycling Programs Cycling programs are shown in Table 8.

 Single-Cell Genome and Transcriptome Amplification 85

Table 7
Composition of reagent mixes used in Subheading 3.4

Reagent Volume [μl]

Primer-mix
Each primer 8 (8 × 8 μl = 64 μl)
Nuclease-free water 136
PCR-mix
10× FastStart PCR buffer (with 20 mM MgCl2) 1.0
Primer-mix 1.0
20 mg/mL BSA 0.2
dNTPs (10 mM each) 0.2
FastStart Taq polymerase 0.1
Nuclease-free water 6.5

Table 8
Cycling programs used in Subheading 3.4

Steps Time/temperature Repetition

WGAQC
Step 1 4 min/95 °C 1×
Step 2 30 s/95 °C
Step 3 30 s/58 °C 32×
Step 4 1 min 30 s/72 °C
Step 5 7 min/72 °C 1×
Step 6 1/4 °C
1—forever

3 Methods

3.1 Isolation of The starting point is the sample in 0.2 mL domed cap microtubes
mRNA and Genomic containing one or more cells in 4.4 μl of mTRAP lysis buffer
DNA, and Whole supplemented with 10 ng of E. coli tRNA (see Note 12). The cells
Transcriptome can be freshly isolated or kept at -80 °C.
Amplification 1. Before bringing samples, prepare mixture containing mTRAP
lysis buffer and mTRAP protease mixed in 20:1 ratio. Take 1 μl
of this mixture and mix with 1 μl of mTRAP biotinylated poly-
T PNA oligonucleotides (see Notes 13, 14, and 15).
86 Miodrag Gužvić

A B C D
tip

beads

rack

magnet duct tape

Fig. 2 Important steps during WTA. (a) Microtubes in 50 mL screw cap tube stuffed with paper wipes (grey). (b)
Position of the pipette tip during bead washing and supernatant removal. (c) Fastening the microtubes on the
stripe of the masking duct tape. (d) Fastening the microtubes using masking duct tape to the vial/bottle of
hybridization oven

2. Bring samples into the laminar flow bench and keep on ice (see
Note 16).
3. Add 2 μl of lysis buffer:protease/PNA mix into the sample (see
Notes 17 and 18). Shortly spin in microcentrifuge.
4. Lyse the sample in thermal cycler, using cycling program LYSIS
(Table 1). See Note 19.
5. Prepare and label two sets of fresh 0.2 mL domed cap micro-
tubes. One set will be used for downstream WTA procedure
and the other for WGA procedure (see Subheading 3.2). See
Note 20.
6. Add 0.8 μl of polyacrylamide carrier into the WGA-set of
0.2 mL tubes from previous step (see Note 21). Put these
microtubes aside.
7. Thoroughly vortex mTRAP streptavidin-conjugated microbe-
ads and shortly spin in microcentrifuge (see Note 22).
8. Once the lysis is done, bring the samples under the laminar flow
bench, shortly spin them, place them on ice, and add 4 μl of
beads into each sample.
9. Stuff some paper towels into 50 mL tube, and put 0.2 micro-
tubes with samples in. Make sure that the microtubes are not
moving (Fig. 2a). Place the 50 mL tube onto the roller and
incubate 45 min at room temperature (see Notes 23 and 24).
10. Shortly before the incubation is done, turn on the hybridiza-
tion oven and set it to 44 °C.
11. Prepare RT-Mix 1 and RT-Mix 2 (without RT enzyme) and
keep them on ice (Table 2). See Note 25.
12. Once the incubation is over, spin the microtubes and place
them on ice.
 Single-Cell Genome and Transcriptome Amplification 87

13. Add 10 μl of IGEPAL-wash buffer into each tube, gently mix

by pipetting up and down, and place on magnetic microtube
rack kept on ice (see Notes 17 and 26). Place the microtubes in
a row where the magnet is on the same side of the well where
the hand is that you are using to pipet (Fig. 2b; see Note 27).
14. Carefully, placing the pipette tip on away from the beads, take
out 20 μl of supernatant and transfer it to WGA-microtubes
with polyacrylamide carrier (see Notes 28 and 29).
15. Add 20 μl of Tween-wash buffer into each tube and gently mix
by pipetting up and down. Collect the compete sample into the
pipette tip and transfer to a fresh WTA microtube (from step 5)
and place it on magnetic microtube rack kept on ice (see Note
30).
16. Repeat step 14.
17. Add 20 μl of IGEPAL-wash buffer into each tube, gently mix
by pipetting up and down, and place on magnet microtube rack
kept on ice.
18. Carefully, placing the pipette tip on away from the beads, take
out all of the supernatant (slightly more than 20 μl) and trans-
fer it to WGA-microtubes with polyacrylamide carrier (see
Notes 31 and 32).
19. Add 10 μl of the RT-Mix 1 to the beads, mix by pipetting up
and down, and incubate 10 min at room temperature (in a
non-magnetic microtube rack; see Note 33).
20. Add 120 μl of 100% ice-cold ethanol to collected supernatants
in WGA-microtubes. Mix by inverting microtubes (see Note
34). Store the microtubes at -20 °C until processing (see
Subheading 3.2).
21. Add 1 μl of SSII RT enzyme to RT-mix 2. Mix well.
22. Add 10 μl of RT Mix 2 to each sample.
23. Fasten the microtubes to the outer side of the hybridization
vial and place inside the pre-heated hybridization oven (Fig. 2c,
d; see Note 35). Start the rotation and incubate for 45 min at
44 °C (see Notes 8 and 36).
24. After incubation, shortly spin the microtubes and place them
on magnetic rack kept on ice.
25. Prepare Tailing-mix (Table 2).
26. Carefully, placing the pipette tip away from the beads, take out
20 μl of supernatant and discard (Fig. 2b).
27. Add 20 μl of Tailing-wash buffer, gently mix by pipetting up
and down, and place on magnetic microtube rack kept on ice.
28. Carefully, placing the pipette tip away from the beads, take out
20 μl of supernatant and discard (see Note 37).
88 Miodrag Gužvić

29. Add 10 μl of Tailing-mix.

30. Add 40 μl of oil (see Note 38).
31. Start DENATURATION (Table 1) program on the thermal
cycler and wait until the block temperature starts to rise. Pause
the program, and wait until the block temperature reaches 94 °
C. Place the microtubes inside, close the lid, and restart the
program (see Note 39).
32. After 4 min open the lid, take the microtubes out, and place
them on ice.
33. Carefully, add 0.8 μl of TdT enzyme underneath the oil.
34. Put the microtubes in the thermal cycler and start the program
TAILING (Table 1; see Notes 40 and 41).
35. After tailing, inactivate TdT by starting TINACTIVATE pro-
gram on thermal cycler (Table 1).
36. Prepare PCR-mix 1 and PCR-Mix 2 (without PolMix enzyme;
Table 2).
37. Spin the microtubes and place them on ice.
38. Add 35.5 μl of PCR-mix 1 into each sample (see Note 42).
39. Add 1.5 μl of PolMix enzyme into PCR-Mix 2.
40. Place the microtubes in thermal cycler and start AMPLIFICA-
TION program (Table 1).
41. Wait until the block temperature reaches 78 °C and pause the
program.
42. Quickly add 5.5 μl of PCR-mix 2 into each sample (see Note
43).
43. Close the lid of the thermal cycler and restart the program (see
Notes 44 and 45).

3.2 DNA Supernatants collected during WTA procedure (step 20 of Sub-

Precipitation and heading 3.1) are stored at -20 °C before processing. The proce-
Whole Genome dure described in this section takes 4 consecutive days. Breaks in
Amplification terms of skipping days are possible, but are strongly discouraged
and should be used only in case of emergency (see Note 46).
1. Cool the centrifuge to 4 °C.
2. Centrifuge the DNA-containing supernatants in ethanol
45 min at 18400 rcf at 4 °C (see Note 47).
3. After centrifuging small white pellets should be visible at the
bottom of the tube (see Note 48). Total volume in the micro-
tube is around 180 μl.
4. Remove and discard supernatant, and leave approximately 20 μl
inside (see Notes 49 and 50).
5. Gently add 180 μl of 70% ethanol (see Note 51).
 Single-Cell Genome and Transcriptome Amplification 89

6. Place the tubes in Thermomixer and incubate 10 min on 19 °C

at 350 rpm (see Note 52).
7. Set the centrifuge to room temperature.
8. After incubation, centrifuge the samples 10 min at 18400 rcf at
room temperature.
9. Repeat steps 4–8 two more times, for a total of three ethanol
washing/thermomixer/centrifuging steps.
10. After the final centrifugation, remove and discard complete
supernatant (see Note 53).
11. Leave microtubes open for the pellets to dry. This may take
between few minutes, e.g., 30 min (see Notes 54 and 55).
12. Once the pellets are dry, add 3.48 μl of water (see Note 56).
13. Incubate overnight in Thermomixer on 19 °C at 350 rpm.
14. On the next day, take the samples out of Thermomixer, shortly
spin, and place on microtube rack.
15. Prepare Proteinase-K digestion mix (Table 3). Add 1.02 μl to
each sample. Incubate overnight in thermal cycler using pro-
gram PROTEINASE (Table 4). See Notes 57 and 58.
16. On the following day, prepare MseI digestion mix (Table 3).
Add 3 μl to each sample, and incubate in thermal cycler using
program MSE1 (Table 4). See Note 59.
17. Prepare Adapter annealing mix (Table 3). Incubate the com-
plete reagent mix in thermal cycler using program ANNEAL
(Table 4). See Note 60.
18. After MseI digestion and adapter annealing are finished, pre-
pare Ligation mix (Table 3). Add 5 μl of the Ligation mix to
each sample, and incubate overnight in thermal cycler using
program LIGATION (Table 4). See Note 61.
19. On the final day, prepare PCR-mix (Table 3). Add 40 μl to each
sample and run the program AMPLIFICATION (Table 4) in
thermal cycler. See Notes 62, 63, and 64.

3.3 Checking the This procedure is a multiplex end-point PCR that amplifies three
Quality of WTA fragments of different lengths, mapping to three “housekeeping”
genes (see Note 65).
1. Label enough 0.2 microtubes for WTA samples, positive, and
negative control (see Note 66).
2. Prepare the PCR-mix (Table 5) and distribute 9.0 μl of the mix
to the tubes.
3. Add 1 μl of the WTA product. Shortly spin the tubes, place
them in thermal cycler, and run program WTAQC (Table 6).
90 Miodrag Gužvić

4. After amplification, perform the fragment analysis using your

preferred way (see Note 67).
5. Note down the number of expected amplified fragments. The
number of bands reflects the quality of the WTA sample (see
Notes 68 and 69).

3.4 Checking the This procedure is a multiplex end-point PCR that amplifies four
Quality of WGA fragments of different lengths, located on MseI digested fragments
of different lengths and mapping to different regions of the
genome (see Note 70).
1. Label enough 0.2 microtubes for WGA samples, positive, and
negative control (see Note 71).
2. Prepare the PCR-mix (Table 7) and distribute 9.0 μl of the mix
to the tubes.
3. Add 1 μl of the WGA product. Shortly spin the tubes, place
them in thermal cycler, and run program WGAQC (Table 8).
4. After amplification, perform the fragment analysis using your
preferred way (see Note 67).
5. Note down the number of expected amplified fragments. The
number of bands reflects the quality of the WGA sample (see
Note 72).

4 Notes

1. Alternatively, linear polyacrylamide can be prepared using pub-

lished protocol [19].
2. This is an established protocol that uses SuperScript II reverse
transcriptase. It is possible to use other SuperScript enzymes
(e.g., III or IV) or enzymes from other manufacturers, as long
as they have optimal processivity at 44 °C and are RNaseH-
deficient. Replacement of an RT system can be done with
sufficient comparison and testing using SS II system.
3. N8 oligonucleotide contains XbaI restriction site that facilitates
removal of oligonucleotide adapters, cloning of individual frag-
ments, or preparation of sequencing libraries. The sequence of
this oligonucleotide can be modified by introducing or adding
other restriction sites.
4. T24 oligonucleotide contains XbaI and BpuEI restriction sites
that facilitate removal of oligonucleotide adapters, cloning of
individual fragments, or preparation of sequencing libraries.
The sequence of this oligonucleotide can be modified by intro-
ducing or adding other restriction sites; e.g., previous version
of this oligonucleotide contained only XbaI site (5′-CCC CCC
CCC CCC CCC GTC TAG ATT TTT TTT TTT TTT TTT
TTT TTT TVN-3′).
 Single-Cell Genome and Transcriptome Amplification 91

5. Other TdT enzymes can be used, but they have to be compati-

ble with the Tailing reaction mix used here (Table 2); i.e., own
TdT reaction buffer cannot be used.
6. Other polymerase blends can be used. Prerequisites are that the
blend has to generate long DNA fragments and that it has
proofreading activity. For some downstream applications, it
may be necessary that the composition of the reaction buffer
is known.
7. CP2 oligonucleotide contains EcoRI restriction sites that facil-
itate removal of oligonucleotide adapters, cloning of individual
fragments, or preparation of sequencing libraries. The
sequence of this oligonucleotide can be modified by introdu-
cing or adding other restriction sites.
8. Any alternative approach that enables fixing the tubes to a
rotating element at 44 °C is also possible.
9. pH of the WGA buffer is set the following way. After mixing of
all components in a large volume, an aliquot is taken and pH
measured. Then, using NaOH or HCl, pH is further set.
Again, an aliquot is measured on pH meter. And so on. This
is done to avoid possible contamination of the buffer by general
pH meter. After the pH is finally set, the buffer should be
filtered, aliquoted, and frozen.
10. Alternative MseI enzyme can be used, as long as it has the same
enzyme concentration and it works with WGA buffer.
11. The “overnight” ligation step does not have a fixed amount of
time. 12 h should be enough. This offers some flexibility in the
execution of the different steps of WGA protocol.
12. E. coli tRNA serves two purposes. First, it coats the walls of the
microtube, thereby preventing passive binding of precious
sample’s RNA. Second, since it is added in excess, it serves as
a substrate for any RNases that may be active before the sample
is processed. However, if addition of tRNA is not possible or is
not practical, it can be left out. This protocol really starts after
the cell isolation step.
13. All reagents should be thawed, thoroughly mixed, and spun in
microcentrifuge before preparing mixes. Upon adding
reagents or mixes to the samples, unless otherwise noted, mix-
ing by gently pipetting up and down and a short spin in a
microcentrifuge should always be done. Additional spin should
also be done after any incubation steps, and before adding new
reagents/mixes.
14. It is recommended not to process more than 12 samples at
once (including controls). Each run should have at least one
“cell-free/reagents only” negative control. Optionally, a posi-
tive control containing one or more cells of, e.g., cell line or
peripheral blood lymphocytes (to name the few that can be
easily obtained) can be included.
92 Miodrag Gužvić

15. Depending on the number of samples, the simplest approach is

to mix 9.5 μl of lysis buffer with 0.5 μl of protease, or 19 μl of
lysis buffer with 1 μl of protease. Then, another mix is made by
mixing equal volumes of lysis buffer/protease mix and PNA
oligonucleotides (e.g., 13 μl + 13 μl for 12 samples).
16. Microtube racks with samples should always be kept on ice,
unless otherwise noted. Instead of ice, one can use microtube
ice racks.
17. Pipetting steps should be done by most appropriate pipette and
tip for the volume being handled. That being said, it is better
to pipet, for example, 2 μl with a 0.5–2 μl pipet than 2–20 μl,
and it is better to pipet 20 μl with a 2–20 μl pipet than
20–200 μl pipet.
18. The reagent mix should be added by touching the microtube
wall near the liquid surface. Touching of the sample should be
avoided due to capillary forces that may pull in minuscule
amounts of liquid that may result in loss of nucleic acid
molecules.
19. During this step, cell is lysed, enzymes thermally inactivated
and digested, and biotinylated poly-T PNAs capture poly-A
tails of mRNAs (Fig. 1a).
20. This protocol can be used for WTA only. If genomic DNA for
WGA is not needed, the second set of 0.2 mL microtubes with
polyacrylamide carrier is not needed.
21. Polyacrylamide carrier serves to facilitate and visualize DNA
precipitation.
22. mTRAP beads should be vortexed or mixed until no bead
precipitate is visible and the liquid becomes homogeneous.
Short spin in microcentrifuge is needed to remove the liquid
from the cap, but it should lead to repeated precipitation of
beads. If pipetting takes too long, it should be checked whether
the beads started to precipitate, which may require another
mixing/vortexing.
23. Any alternative approach that enables fixing the tubes to a
rotating element on room temperature is also possible.
24. After this step, biotinylated poly-T PNAs hybridized to poly-A
tails of mRNAs are captured by streptavidin-coated beads
(Fig. 1a). From this step onward, all reactions take place on
beads, i.e., on solid phase.
25. IGEPAL prevents aggregation of beads. T24 oligonucleotide
hybridizes to the beginning of the poly-A tail of mRNA (not
hybridized by PNA oligonucleotide). N8 is similar, but instead
of poly-T and following dinucleotide sequence, it contains
random octamer, to bind randomly along the mRNA, thus
creating short cDNA fragments. Octamer is used instead of
 Single-Cell Genome and Transcriptome Amplification 93

commonly used hexamers to reduce the number of binding

sites and thereby increase the length of cDNA fragments.
26. Detergents reduce the surface tension and facilitate superna-
tant removal.
27. If you are right-handed, the beads will precipitate on the right
wall of the microtube, and pipette tip can be placed on the
opposite side to collect the supernatant.
28. If genomic DNA is not needed, the supernatant can be
discarded.
29. The washing of the beads is the critical step and should be done
very carefully and slowly. When adding wash buffers, bubbles
should be avoided. If they appear, be very careful in removing
the supernatant, since the bubbles pull the beads into the
pipette tip. In that case, it is better to remove less than 20 μl.
One can remove supernatant from a microtube and immedi-
ately proceed with the next washing step and put back the
microtube to the magnetic rack. Alternatively, supernatant
can be sequentially removed from up to 4 microtubes (close
the lid of the microtubes where the supernatant is removed)
and then new wash can be performed. In any case, it is critically
important not to let the beads to dry.
30. This is the earliest safest point to get rid of the original sample
microtube, thereby reducing the risk to carry-over any
unwanted residues.
31. To ensure that complete supernatant is removed, one approach
can be to press the pipette plunger slightly into the “second”
step before inserting the tip into the sample, thereby being able
to collect slightly more than 20 μl.
32. If mRNAs and WTA are not needed, the sample with the beads
can be discarded, and supernatant in WGA microtubes can be
processed further, as described in Subheading 3.2. However, if
only genomic DNA and WGA of single cell are needed, there
are faster, simpler, and cheaper protocols [20, 21].
33. During this step, T24 and N8 oligonucleotides are binding to
mRNAs (Fig. 1a).
34. This step will enable precipitation of DNA.
35. For this we usually stick the microtubes to a strip of paper/
masking duct tape, and fasten it to the vial/bottle of the
hybridization oven (Fig. 2c, d).
36. During this step, reverse transcription takes place (Fig. 1a).
37. These steps of beads washing and supernatant removal are
critical. The purpose is to remove any remaining dNTPs from
94 Miodrag Gužvić

the RT step, since they may interfere with subsequent poly-G

tailing (by introducing dNTPs other than dGTP).
38. Oil serves to prevent evaporation of the sample during
subsequent denaturation step.
39. During the denaturation step, mRNAs and cDNAs are released
from beads. From this step onward, the reactions are not
happening on the solid phase.
40. In this step, TdT will incorporate dGTPs and will generate 3′
poly-G tails on single-stranded cDNAs (Fig. 1a).
41. Although not recommended, this is a step where it is possible
to introduce a break, e.g., overnight. One can simply leave the
samples in the cycler or at 22 °C.
42. Formamide is used to decrease the annealing temperature of
G-C pairs formed by CP2 oligonucleotides. CP2 oligonucleo-
tides will bind to the poly-G tails generated in step 34 of
Subheading 3.1. The same nucleotide will bind to the poly-G
sequences complementary to the 5′ end of the T24 and N8
oligonucleotides introduced during reverse transcription step.
Therefore, CP2 oligonucleotide serves both as forward and
reverse primer (Fig. 1a).
43. This is basically a “hot-start” approach. The reason for the hot
start is to avoid unspecific priming and extension of the CP2
primers (that bound to the single-stranded cDNA at low tem-
peratures) until 94 °C is reached.
44. This is a final step of a global mRNA amplification (i.e., WTA).
45. Theoretically, original WTA product can perpetually be used to
reamplify itself. PCR mix can be prepared according to Table 9
and, with 1 μl of WTA product added, can be amplified using
cycling program in Table 10. Quality can be controlled by
using the procedure shown in Subheading 3.3.
46. There is a kit based on WGA protocol presented here. This kit
(Ampli1 WGA PLUS Kit) is marketed by Menarini Silicon
Biosystems, and its use is previously published 20 and also
described elsewhere in this book.
47. Centrifuging time can also be a bit longer, e.g., 60 min.
48. Pellets may also have the shades of brown, coming from beads
that were collected during supernatant removal (steps 13–18
of Subheading 3.1). These beads do not interfere with the
procedure.
49. Pipet very slowly. Avoid detaching of the pellet from the bot-
tom of the microtube, touching the pellet by the tip, or pipet-
ting near the pellet.
50. If the pellet is detached, it is not a big problem, just be very
cautious not to suck it in in the tip. If it does happen, expel it
 Single-Cell Genome and Transcriptome Amplification 95

Table 9
PCR-mix for WTA reamplification

Reagent Volume [μl]

PCR-mix
ELTPS buffer 1 5.0
24 μM CP2 oligonucleotide 6.0
dNTPs (10 mM each) 1.75
20% Formamide 7.5
ELTPS PolMix enzyme 1.5
Nuclease-free water 27.25

Table 10
Cycling profile for WTA reamplification

Steps Time/temperature Repetition

WTAQC
Step 1 1 min/95 °C 1×
Step 2 15 s/94 °C
Step 3 1 min/55 °C 5×
Step 4 3 min 30 s/65 °C
Step 5 15 s/94 °C
Step 6 1 min/55 °C 3×
Step 7 3 min 30 s/65 °C
+10 s/cycle
Step 8 7 min/65 °C 1×
Step 9 1/4 °C
1—forever

out, and briefly spin the microtube again. If the pellet sticks to
the surface of the tips, just pipet vigorously. At this point,
damaged pellet is better than the lost pellet. Repeat superna-
tant removal, as described in step 4 of Subheading 3.2.
51. Just add ethanol along the microtube wall. Do not mix or
resuspend.
52. These steps serve to wash away buffer residues or other con-
taminating molecules from the DNA pellet.
53. In this step the complete supernatant is removed. Therefore,
pipetting near the pellet cannot be avoided. Just be extra slow
96 Miodrag Gužvić

and careful. If it is not possible to remove all of the liquid

without the danger of sucking in the pellet, rather leave some
liquid inside. This will lead to longer drying in step 11 of
Subheading 3.2.
54. White pellets become transparent when dry. Dark brown pel-
lets become light brown when dry. The pellets should not have
any visible liquid left (ethanol interferes with subsequent
steps), but should also not over-dry, as this may interfere with
dissolving in water.
55. Extra caution needed—dry pellets may react to static electricity
coming from the hand gloves, and may change location inside
of the microtube, or even fly out of the tube, helped by the air
flow in the bench. Before taking the tubes to check if the pellet
is dry, close the cap, and open it again once the microtube is
back in the rack if additional drying is needed.
56. Add water drop near the pellet, close the microtube, and gently
flick the tube to enable contact of water and pellet. Do not mix
or resuspend.
57. This step may seem superfluous since the protease step was
already done during WTA procedure (Steps 3–4 of Subhead-
ing 3.1). However, this step is part of the original WGA proto-
col and it makes sure that the DNA is completely available for
subsequent restriction digestion.
58. From this point on, reagent mixes are just added to the sample
by pipetting on the microtube wall and shortly spinning the
microtube to bring the mix to the sample. Avoid touching the
sample with pipette tip, as this may result in a loss of single-
cell DNA.
59. During this step, MseI digest complete genomic DNA of a
single cell, thereby generating so called “MseI fragments”
(Fig. 1b).
60. During this step, oligonucleotides Lib1 and ddMse11 generate
short double-stranded adapters perfectly matching the MseI
restriction sites (Fig. 1b).
61. During this step, ligation of adapters takes place. 3′-dideoxy
nucleotide of ddMse11 primer prevents extension of this
primer and therefore random priming events. Due to the lack
of 5′ phospahate, this primer will not be ligated to the 3′ of the
MseI fragment overhang and dissociates during the first dena-
turation step of the global amplification (Fig. 1b). Therefore,
only Lib1 primer gets ligated to MseI fragments and forms an
identical adapter sequence on both termini of each MseI
fragment.
62. Lib1 primer drives the amplification of MseI fragments by
acting as both forward and reverse primer (Fig. 1b).
 Single-Cell Genome and Transcriptome Amplification 97

Table 11
PCR-mix for WGA reamplification

Reagent Volume [μl]

PCR-mix
ELTPS buffer 1 5.0
10 μM Lib1 5.0
dNTPs (10 mM each) 1.75
20 mg/mL BSA 1.25
ELTPS PolMix enzyme 0.5
Nuclease-free water 35.5

Table 12
Cycling profile for WGA reamplification

Steps Time/temperature Repetition

WTAQC
Step 1 1 min/94 °C 1×
Step 2 30 s/60 °C 1×
Step 3 2 min/68 °C 1×
Step 4 30 s/94 °C
Step 5 30 s/60 °C
11×
Step 6 3 min/68 °C
+20 s/cycle
Step 7 1/4 °C
1—forever

63. This is a final step of a global genomic DNA amplification

(i.e., WGA).
64. Theoretically, original WGA product can perpetually be used to
reamplify itself. PCR mix can be prepared according to
Table 11 and, with 1 μl of WGA product added, can be ampli-
fied using cycling program in Table 12. Quality can be con-
trolled by using the procedure shown in Subheading 3.4.
Please note that elongation can also be performed at 65 °C.
This approach generates slightly different fragment distribu-
tion. Alternatively, two reamplifications can be done with dif-
ferent elongation temperatures, and resulting WGA
reamplification products can be mixed. The utility of each
approach should be tested by individual lab according to their
downstream applications.
98 Miodrag Gužvić

65. GAPDH primers generate fragment 489 bp long, ACTB pri-

mers generate fragment 378 bp long, and EEF1A1 primers
generate fragment 290 bp long.
66. Positive control can be WTA product of a cell pool (cell lines or
blood lymphocytes).
67. Standard agarose gel electrophoresis can be used, or devices
that use capillary electrophoresis can be used (e.g., Qiaxcel).
68. We have observed that the samples with all three fragments
coming in QC-PCR have the best WTA quality. However,
these are not absolutes, and certainly samples with 2 or
1 band can be further analyzed by, e.g., sequencing. However,
a caution should be exercised in interpretation, since
sub-optimal WTA quality may affect both the presence and
the amount of transcripts, which may interfere with biological
interpretation. Therefore, each application would need a series
of validation experiments.
69. Further quality control steps can be undertaken. It is possible
to measure the concentration of DNA by using, e.g., Nano-
drop or Qubit, or fragment length distribution can be assessed
by, e.g., Bioanalyzer. Due to the presence of unbound oligo-
nucleotides, dNTPs, etc., it is advised to purify WTA product
before the measurement. This can be done by standard PCR
purification kits. In this case, only an aliquot of original WTA
product should be used for this, in case anything goes wrong
during purification.
70. KRAS primers generate fragment 91 bp long, amplifying por-
tion of MseI fragment 192 bp long, originating from chromo-
some 12. D5S2117 primers generate fragment approx. 140 bp
long (this is microsatellite locus and the actual fragment length
may vary by few bp or appear as a double band), amplifying
portion of MseI fragment 1376 bp long, originating from
chromosome 5. KRT19P1 primers generate fragment 621 bp
long, amplifying portion of MseI fragment 1146 bp long,
originating from chromosome 12. TP53 primers generate frag-
ment 301 bp long, amplifying portion of MseI fragment
1374 bp long, originating from chromosome 17.
71. Positive control can be WGA product of a cell pool (cell lines or
blood lymphocytes).
72. QC PCR for WGA has 4 PCR fragments. Practice shows that
the more fragments are appearing in QC PCR, the WGA
product is of better quality. However, these are not absolutes,
and certainly samples with less than 4 fragments can be further
analyzed by, e.g., sequencing. We advise using GII (genomic
integrity index, [21]) to judge the WGA quality. It is consid-
ered that WGA products with GII 3 and 4 have good quality,
while the WGA product with GII 2 can be used with caution.
 Single-Cell Genome and Transcriptome Amplification 99

References
1. Hay SB, Ferchen K, Chetal K, Grimes HL, Sca-1+/CD34+,- cells--the putative murine
Salomonis N (2018) The human cell atlas lung stem cells. PLoS One 8(12):e83917.
bone marrow single-cell interactive web portal. https://doi.org/10.1371/journal.pone.
Exp Hematol 68:51–61. https://doi.org/10. 0083917
1016/j.exphem.2018.09.004 10. Suzuki T, Asami M, Hoffmann M, Lu X,
2. Gužvić M, Braun B, Ganzer R, Burger M, Gužvić M, Klein CA, Perry ACF (2016) Mice
Nerlich M, Winkler S, Werner-Klein M, Czyz produced by mitotic reprogramming of sperm
ZT, Polzer B, Klein CA (2014) Combined injected into haploid parthenogenotes. Nat
genome and transcriptome analysis of single Commun 7:12676. https://doi.org/10.
disseminated cancer cells from bone marrow 1038/ncomms12676
of prostate cancer patients reveals unexpected 11. Asami M, Lam BYH, Hoffmann M, Suzuki T,
transcriptomes. Cancer Res 74(24): Lu X, Yoshida N, Ma MK, Rainbow K,
7383–7394. https://doi.org/10.1158/ Gužvić M, VerMilyea MD, Yeo GSH, Klein
0008-5472.CAN-14-0934 CA, Perry ACF (2023) A program of successive
3. Lee J, Hyeon DY, Hwang D (2020) Single-cell gene expression in mouse one-cell embryos.
multiomics: technologies and data analysis Cell Rep 42(2):112023. https://doi.org/10.
methods. Exp Mol Med 52(9):1428–1442. 1016/j.celrep.2023.112023
https://doi.org/10.1038/s12276-020- 12. Hartmann CH, Klein CA (2006) Gene expres-
0420-2 sion profiling of single cells on large-scale oli-
4. Klein CA, Schmidt-Kittler O, Schardt JA, gonucleotide arrays. Nucleic Acids Res 34(21):
Pantel K, Speicher MR, Riethmuller G (1999) e143. https://doi.org/10.1093/nar/gkl740
Comparative genomic hybridization, loss of 13. Fuhrmann C, Schmidt-Kittler O, Stoecklein
heterozygosity, and DNA sequence analysis of NH, Petat-Dutter K, Vay C, Bockler K,
single cells. Proc Natl Acad Sci U S A 96(8): Reinhardt R, Ragg T, Klein CA (2008) High-
4494–4499. https://doi.org/10.1073/pnas. resolution array comparative genomic hybridi-
96.8.4494 zation of single micrometastatic tumor cells.
5. Klein CA, Seidl S, Petat-Dutter K, Offner S, Nucleic Acids Res 36(7):e39. https://doi.
Geigl JB, Schmidt-Kittler O, Wendler N, org/10.1093/nar/gkn101
Passlick B, Huber RM, Schlimok G, Baeuerle 14. Czyz ZT, Hoffmann M, Schlimok G, Polzer B,
PA, Riethmuller G (2002) Combined tran- Klein CA (2014) Reliable single cell array CGH
scriptome and genome analysis of single micro- for clinical samples. PLoS One 9(1):e85907.
metastatic cells. Nat Biotechnol 20(4): https://doi.org/10.1371/journal.pone.
3 8 7 – 3 9 2 . h t t p s : // d o i . o r g / 1 0 . 1 0 3 8 / 0085907
nbt0402-387 15. Hoffmann M, Pasch S, Schamberger T,
6. Klein CA, Zohlnhofer D, Petat-Dutter K, Maneck M, Mohlendick B, Schumacher S,
Wendler N (2003) Gene expression analysis of Brockhoff G, Knoefel WT, Izbicki J, Polzer B,
a single or few cells. Curr Protoc Mol Biol Stoecklein NH, Klein CA (2018) Diagnostic
Chapter 25:Unit 25B 28. https://doi.org/10. pathology of early systemic cancer: ERBB2
1002/0471142727.mb25b08s61 gene amplification in single disseminated can-
7. Hahtola S, Burghart E, Jeskanen L, Karenko L, cer cells determines patient survival in operable
Abdel-Rahman WM, Polzer B, Kajanti M, esophageal cancer. Int J Cancer 142(4):
Peltomaki P, Pettersson T, Klein CA, Ranki A 833–843. https://doi.org/10.1002/ijc.
(2008) Clinicopathological characterization 31108
and genomic aberrations in subcutaneous 16. Durst FC, Grujovic A, Ganser I, Hoffmann M,
panniculitis-like T-cell lymphoma. J Invest Ugocsai P, Klein CA, Czyz ZT (2019) Tar-
Dermatol 128(9):2304–2309. https://doi. geted transcript quantification in single
org/10.1038/jid.2008.6 disseminated cancer cells after whole transcrip-
8. Chambers KF, Pearson JF, Pellacani D, Aziz N, tome amplification. PLoS One 14(8):
Gužvić M, Klein CA, Lang SH (2011) Stromal e0216442. https://doi.org/10.1371/journal.
upregulation of lateral epithelial adhesions: pone.0216442
gene expression analysis of signalling pathways 17. Werner-Klein M, Grujovic A, Irlbeck C,
in prostate epithelium. J Biomed Sci 18(1):45. Obradovic M, Hoffmann M, Koerkel-Qu H,
https://doi.org/10.1186/1423-0127-18-45 Lu X, Treitschke S, Kostler C, Botteron C,
9. Hittinger M, Czyz ZT, Huesemann Y, Weidele K, Werno C, Polzer B, Kirsch S,
Maneck M, Botteron C, Kaeufl S, Klein CA, Gužvić M, Warfsmann J, Honarnejad K,
Polzer B (2013) Molecular profiling of single Czyz Z, Feliciello G, Blochberger I,
100 Miodrag Gužvić

Grunewald S, Schneider E, Haunschild G, carrier. Nucleic Acids Res 18(2):378. https://

Patwary N, Guetter S, Huber S, Rack B, doi.org/10.1093/nar/18.2.378
Harbeck N, Buchholz S, Rummele P, 20. Czyz ZT, Klein CA (2015) Deterministic
Heine N, Rose-John S, Klein CA (2020) whole-genome amplification of single cells.
Interleukin-6 trans-signaling is a candidate Methods Mol Biol 1347:69–86. https://doi.
mechanism to drive progression of human org/10.1007/978-1-4939-2990-0_5
DCCs during clinical latency. Nat Commun 21. Polzer B, Medoro G, Pasch S, Fontana F,
11(1):4977. https://doi.org/10.1038/ Zorzino L, Pestka A, Andergassen U, Meier-
s41467-020-18701-4 Stiegen F, Czyz ZT, Alberter B, Treitschke S,
18. Stoecklein NH, Erbersdobler A, Schmidt- Schamberger T, Sergio M, Bregola G,
Kittler O, Diebold J, Schardt JA, Izbicki JR, Doffini A, Gianni S, Calanca A, Signorini G,
Klein CA (2002) SCOMP is superior to degen- Bolognesi C, Hartmann A, Fasching PA, San-
erated oligonucleotide primed-polymerase dri MT, Rack B, Fehm T, Giorgini G,
chain reaction for global amplification of min- Manaresi N, Klein CA (2014) Molecular
ute amounts of DNA from microdissected profiling of single circulating tumor cells with
archival tissue samples. Am J Pathol 161(1): diagnostic intention. EMBO Mol Med 6(11):
43–51. https://doi.org/10.1016/S0002- 1371–1386. https://doi.org/10.15252/
9440(10)64155-7 emmm.201404033
19. Gaillard C, Strauss F (1990) Ethanol precipita-
tion of DNA with linear polyacrylamide as
 Chapter 7

Isolation and Genomic Analysis of Circulating Tumor Cell

Clusters in Cancer Patients
Carolina Reduzzi, Marta Vismara, Thomas Schamberger, Marco Silvestri,
Rosita Motta, Bernhard M. Polzer, and Vera Cappelletti

Abstract
The role of circulating tumor cell (CTC) clusters in the metastatic dissemination process is gaining
increased attention. Besides homotypic clusters, heterotypic clusters that contain tumor cells admixed
with normal cells are frequently observed in patients with solid tumors. Current methods used for cluster
detection and enumeration do not allow an accurate estimation of the relative fractions of tumor cells. Here
we describe a method for estimating tumor fraction of clusters including isolation and collection of single
clusters, assessment of copy number alterations of single clusters by low-pass whole genome sequencing,
and bioinformatic analysis of sequencing data.

Key words CTC clusters, Micromanipulation, Genomic analysis, Copy number alteration, Tumor
fraction prediction, Low-pass whole genome sequencing

1 Introduction

The role of circulating tumor cell (CTC) clusters in the metastatic

dissemination process is gaining increased attention [1]. CTC clus-
ters, either representing a grouping of CTCs, i.e., homotypic clus-
ters, or associations between CTCs and accessory normal cells, i.e.,
heterotypic clusters, do have distinct features with respect to single
CTCs and possibly a different biological role and clinical relevance
in the metastatic dissemination [2–4]. As such, CTC clusters repre-
sent a unique biomarker which deserves to be separately studied.
No standardized assay is currently available for CTC-cluster
enumeration and literature data addressing the clinical role of
CTC clusters have been produced with an array of different
approaches including CellSearch® [5], the only FDA-approved
method for enumeration of single CTCs.
Moreover, due to their heterotypic nature, CTC clusters pose
additional challenges beyond simple isolation and enumeration,
dealing with specifying their composition. Here we describe a

Miodrag Gužvić (ed.), Single Cell Analysis: Methods and Protocols, Methods in Molecular Biology, vol. 2752,
https://doi.org/10.1007/978-1-0716-3621-3_7, © Springer Science+Business Media, LLC, part of Springer Nature 2024

101
102 Carolina Reduzzi et al.

comprehensive protocol spanning from CTC-cluster collection to

prediction of their tumor fraction, which defines the percent con-
tent of tumor cells in each single cluster based on the content of
genomically aberrant cells estimated by copy number alteration
(CNA) profiles generated by shallow whole genome sequencing.
The protocol has been developed for breast cancer patients [6]
following the observation of a higher number of CTC cluster in
women with early breast cancer than in those with metastasis [7], as
a first step toward clarifying this biologically and clinically
intriguing observation. Although not yet validated in other tumor
types, it can be simply extended by running a new calibration curve
with ad hoc cell lines.

2 Materials

2.1 CTC-Cluster • Blood collection tubes: K2-EDTA BD Vacutainer tubes (BD).

Enrichment and • Blood filtration devices: ScreenCell CYTO kit (ScreenCell).
Collection of Single
• 15 mL and 50 mL sterile conical tubes.
CTC Clusters (See
Subheading 3.1) • Phosphate-buffered saline (PBS) 1×.
• Parafilm.
• May Grunwald (Merck Millipore).
• Giemsa (Merck Millipore).
• Tap water.
• Paper towels.
• Tweezer.
• Serological pipettes (5 and 10 mL) with pipette controller.
• 200 μL micropipette tips with micropipette.
• 4 °C refrigerator.
• Inverted microscope (e.g., Olympus IX81).
• Patchman NP2 micromanipulator (Eppendorf).
• DMZ Capillary puller (Zeitz-Instruments Vertriebs GmbH).
• CellTram Air pump (Eppendorf).
• Glass capillaries: micro-hematocrit capillary, non-heparinized
length 75 mm × 1.1/1.2 mm (Nunc).
• BSA (Sigma-Aldrich).
• Fetal bovine serum (FBS), sera plus, (PAN Biotech).
• Nunc LabTek Chamber Slides; 8 fields (Thermo Fisher
Scientific).
• Ice.
• 200 μL MAXYMum Recovery® PCR Tubes (Axygen).
 Isolation and Genomic Analysis of CTCs 103

2.2 Genomic • Ampli1™ WGA Kit (Menarini Silicon Biosystems; see Note 1).
Analysis of CTC • DNA-OFF DNA removal agent (MP Biomedicals).
Clusters: WGA, QC, and
• Thermal Cycler (Applied Biosystems GeneAmp 9700 or supe-
Low-Pass CNA (See
rior; see Note 2).
Subheading 3.2)
• Set of dedicated micropipettes (Gilson).
2.2.1 Whole Genome
• PCR microcentrifuge tube 0.2 mL (Amplitube™ thin wall reac-
Amplification (WGA)
tion tube 200 μL flat snap cap, Starlab).
• Dual-filter/barrier tips (Eppendorf).
• Mini Centrifuge suitable for 0.2 mL PCR tubes.
• Laminar flow hood.
• -20 °C storage freezer.
• Vortex mixer.
• See Notes 3 and 4.

2.2.2 Quality Control (QC) • Ampli1™ QC Kit (Menarini Silicon Biosystems).

• Dedicated micropipette set for post-PCR reactions.
• 0.2 mL PCR tubes.
• 1.5 mL PCR-clean tubes.
• Microcentrifuge for 1.5 and 0.2 mL tubes.
• Barrier tips.
• -20 °C Storage Freezer
• Thermal Cycler.
• Agilent 2100 Bioanalyzer (Agilent Technologies).
• Agilent DNA 1000 Kit (Agilent Technologies).

2.2.3 LowPass Copy • Ampli1™ LowPass Kit for Ion Torrent (Menarini Silicon Bio-
Number Alteration (LpCNA) systems; see Note 6).
Sequencing • DNA OFF DNA removal agent (MP Biomedicals).
• Thermal Cycler (Applied Biosystems GeneAmp 9700 or supe-
rior; see Note 2).
• Ethanol Bio Ultra, for molecular biology, ≥99.8%.
• Nuclease-free water (Molecular Biology Grade).
• Low TE: 10 mM Tris–HCl, pH 8.0, 0.1 mM EDTA).
• 0.2 mL PCR tubes (Axygen Maxymum Recovery thin wall PCR
tubes with flat cap)
• Aerosol-resistant tips and pipettes range from 1 to 1000 μL.
• SPRI select reagent (Beckman Coulter).
104 Carolina Reduzzi et al.

• Magnetic rack for 0.2 mL tubes (DynaMag-96 Side magnet,

Thermo Fisher Scientific).
• Programmable thermal cycler with adjustable ramp rate,
operating within manufacturer’s specifications.
• Dedicated micropipette sets for pre-PCR and post-PCR
reactions.
• Filter tips (Gilson Diamond sterilized).
• Mini centrifuge for 0.2 mL tubes.
• Vortex.
• Agilent 2100 Bioanalyzer (Agilent Technologies).
• Agilent High Sensitivity DNA Kit (Agilent Technologies).
• E-Gel Precast Agarose Electrophoresis Systems (Thermo Fisher
Scientific).
• E-Gel Size Select Agarose Gels 2%, (Thermo Fisher Scientific).
• 50 pb DNA Ladder.
• Laminar flow hood.
• Tape station (Agilent Technologies).

2.3 Bioinformatic • Torrent Suite™.

Analysis and
Interpretation of
Sequencing Data
2.3.1 Pre-processing and
Alignment

2.3.2 Alignment Quality • Qualimap [8].

Control • FastQC (http://www.bioinformatics.babraham.ac.uk/pro
jects/fastqc).

2.3.3 Copy Number • R.

Alteration Analysis and • HMMcopy package.
Tumor Fraction Estimation
• GenomeInfoDb package.
• GenomicRanges package.
• ichorCNA [9].

2.3.4 Data Visualization • Bash.

and Operative System • R.
• Linux/Unix.
 Isolation and Genomic Analysis of CTCs 105

3 Methods

3.1 CTC-Cluster 1. Collect 9 mL of peripheral venous whole blood in K2-EDTA

Enrichment and tubes (see Notes 7 and 8). Invert the tubes 10 times and store
Picking of Single CTC the blood at 4 °C until blood filtration, for a maximum of 4 h
Clusters (see Note 9).
2. Open one ScreenCell filtration unit (included in the ScreenCell
CYTO kit) and place it on the bench top with the adhesive
protective membrane facing down (see Note 10).
3. Prepare the dilution buffer included in the ScreenCell CYTO
kit (FC2 buffer) by fully screwing the cap on the vial until the
clap of the yellow plunger in the cap is opened (the molded
tamper-evident ring will break). Shake the vial by hand at least
10 times and check the buffer pH using the pH indicator strip
provided in the kit. If needed, add up to 3 μL of 30% NaOH
(included in the kit) to reach a pH between 6.7 and 7.3. Use
the dilution buffer within 24 h.
4. Transfer 3 mL of blood in a 15 mL sterile conical tube, add
4 mL of the prepared FC2 dilution buffer (see step 3), gently
invert the tube 10 times, and incubate at room temperature for
8 min.
5. At the end of the incubation, transfer the 7 mL of diluted
blood in the module A of the filtration unit prepared in step
2. Remove the adhesive protecting membrane from the lower
part of the filtration unit (module B), being careful not to spill
the blood contained in module A. Insert module B onto an
empty blood collection tube provided in the ScreenCell CYTO
kit (module C) and push down completely to start the filtra-
tion. When the blood reaches the yellow line of module A, add
1.6 mL of PBS 1× into module A. Wait until all the liquid has
gone through the membrane, it should take approximately
3 min (see Note 11).
6. Open the device by twisting module A and eject the isolation
support (IS, formed by the filtration membrane and an inox
O-ring) onto a piece of absorbent paper. Wash the IS by
putting 70 μL of PBS on the IS and let the PBS be absorbed
by the paper, through the IS. Using the tweezer, put the filter
on a piece of well-extended parafilm and let the IS 1 h at 37 °C
or dry overnight at room temperature. Then proceed with
staining.
7. Incubate the filter in May Grunwald for 2.5 min at room
temperature, transfer the filter into diluted May Grunwald
(diluted 1:2 with distilled water pH 7.0), and incubate for
2.5 min. Rinse the IS in tap water 2–3 times. Incubate for
106 Carolina Reduzzi et al.

Fig. 1 The pictures show a cluster (dotted white circle) before and after cell detachment with the microma-
nipulator. The red circles indicate two single cells detached from the cluster

10 min the IS in Giemsa diluted 1:10 with distilled water

pH 7.0, then rinse in tap water (see Note 12).
8. Let the IS air-dry for 15 min. Store the IS at 4 °C for 1–2 weeks
maximum and proceed with the procedure for CTC clusters
isolation as follows.
9. Prepare 0.2 mL PCR tubes containing 2 μL of lysis buffer from
the WGA kit (see Subheading 3.2) and store them on ice.
10. Put a glass capillary to the Zeitz capillary puller. Two pick
capillaries are formed from one glass capillary, each with an
opening of 30–50 μm (see Note 13).
11. Put the IS onto the microscope and add approximately 0.5 mL
of 1×PBS to the slide.
12. Coat one field of an 8-field chamber slide with BSA, add
200 μL 1×PBS (= picking field), and place the slide next to
the IS on the microscope.
13. Add by hand FBS into the picking capillary and connect the
capillary to the CellTram Air pump (see Note 14).
 Isolation and Genomic Analysis of CTCs 107

14. Focus the picking field and move the picking capillary by using
the Patch Man joystick into the middle of the visible field.
Lower the capillary to the surface of the picking field and
flush out the FBS (not all) using the air pump. Lift the capillary
out of the picking field (see Note 15).
15. Take 1 μL of PBS from the picking field and transfer into a
0.2 mL PCR tube containing 2 μL lysis buffer from the WGA
kit (see Subheading 3.2) and store it on ice until the end of the
isolation procedure. This refers to picking (negative) control.
16. Focus a target CTC cluster, lower the picking capillary in front
of the cluster. Try to remove the cluster by pushing or scratch-
ing it with the glass capillary. Be careful and do not destroy the
cells/cluster (Fig. 1) (see Note 16).
17. Aspirate the target cluster into the glass capillary, lift the capil-
lary out of the sample slide, focus the picking field, and lower
the glass capillary to the surface (see Note 17).
18. Take a 10 μL pipette including 10 μL pipette tip. Flush the
picked cluster out of the capillary and collect it with the pipette
within 1 μL (see Note 18).
19. Transfer the isolated cluster into a 0.2 mL PCR tube contain-
ing 2 μL of the lysis buffer from the WGA kit prepared in step 9
and store on ice until the end of the isolation procedure, and
then continue with the WGA procedure (see Note 19).

3.2 Genomic The Ampli1™ WGA method was originally designed to amplify
Analysis of CTC DNA from a single cell, but is also suitable for amplification of CTC
Clusters clusters whose cell number ranges from 2 cells to 1000. Due to the
extreme sensitivity of the WGA, it is important to avoid contami-
3.2.1 Whole Genome nation by exogenous DNA during the CTC-picking step.
Amplification (WGA) The Ampli1™ WGA method is based on a ligation-mediated
adaptor linker PCR technique, which assures a balanced amplifica-
tion due to restriction enzyme-promoted fragmentation of the
genome. The workflow for Ampli1™ WGA includes five steps
that may be executed in 1 day.
All reagents necessary to successfully execute the WGA proce-
dure are provided in the Ampli1™ WGA Kit.
1. Prepare control samples.
It is strictly recommended to process the following con-
trols along with samples in each run of Ampli1™ WGA
procedure:
• Two no-cell controls: by adding 1 μL of Ampli1™ Water.
• One positive control: by adding 1 μL of purified DNA
(1 ng/μL) extracted from a cell line. For this step use a
dedicated and decontaminated 10 μL micropipette. It is
108 Carolina Reduzzi et al.

strongly recommended to process the positive control as the

last sample of each step.
2. Open the Ampl1™WGA kit in the dedicated laboratory space
and proceed with the four steps according to kit instructions
(see Notes 3–6).
3. Proceed with STEP 1 (cell lysis) using the PCR tubes deriving
from the picking procedure, containing CTC clusters and lysis
buffer; prepare new PCR tubes for the negative and positive
WGA controls. Continue with STEP 2A (DNA digestion) and
STEP 2B always referring to the Ampli1™WGA kit user man-
ual v6 (see Note 20 for STEP 2B).
4. Continue with STEP 3 (ligation) and STEP 4 (primary PCR) as
described in the Ampli1™WGA kit user manual v6 (see Note
21 for the final step).

3.2.2 Quality Control (QC) The Ampli1™QC kit contains reagents for a PCR-based assay
of WGA Products useful for the qualitative evaluation of the obtained Ampli1™WGA
products (for human-origin samples). It predicts the suitability of
each sample for downstream analyses and is based on a multiplex
PCR for four markers. Positivity of at least two markers predicts
successful genome-wide analyses to detect copy number alterations
(CNA).
Store the Ampli1™ QC Kit at -15/-20 °C. Transfer the tube
containing the enzyme to ice just prior to use. Other kit compo-
nents should be thawed, stored on ice, and briefly vortexed before
use (see Note 22).
1. It is recommended to prepare the following controls for
Ampli1™ QC reaction:
• Blank control: by adding 1 μL of PCR water.
• Positive control: by adding 1 μL of 100 ng/μL human
genomic DNA.
2. Prepare the reaction mixture according to the instructions of
the Ampli1™QC kit. Add 1 μL of Ampli1™WGA product,
prepare a blank and a positive control, and incubate in the
thermal cycler as indicated.
3. For the qualitative evaluation of PCR products, we load 1 μL of
the final product on the Agilent DNA1000 chip on the Bioa-
nalyzer Agilent 2100.

3.2.3 Low Pass CNA Ampli1™ LowPass Kit is designed to generate multiplexed,
Sequencing sequencing-ready libraries in a streamlined single-day protocol, to
detect chromosomal aneuploidies and CNA, by low-pass whole
genome sequencing using the Ion PGM or the Ion S5 systems
(see Note 6). Libraries are created through a single-reaction step
 Isolation and Genomic Analysis of CTCs 109

and offer the possibility of generating up to 48 barcoded libraries

suitable for both Ion Torrent NGS platforms.
The kit offers a robust and reproducible method for shallow
whole genome to assess genome-wide copy number alterations.
Upon receipt, store the Ampli1™ LowPass Kit at -20 °C.
Transfer the tube containing the enzyme to ice just prior to use,
add the enzyme to the reaction mix as the last component, and
immediately put the tube containing the enzyme back at -20 °C.
The other kit components should be thawed, stored on ice, and
briefly vortexed before use.
Open the Ampl1™ Low pass Kit in the dedicated laboratory
space (see Notes 23 and 24) and proceed through four steps
according to kit instructions.
1. Proceed with STEP 1 (Clean up of Ampli1™ WGA product)
referring to the user manual of Ampli1™ LowPass Kit v5 (see
Note 25).
2. Continue with STEP 2 (Barcoding Reaction) according to the
user manual of Ampli1™ LowPass Kit v5 (see Note 26).
3. Thereafter, proceed with STEP 3 (Clean up of Final Library)
according to the user manual of Ampli1™ LowPass Kit v5 (see
Note 27).
4. Next, perform STEP 4 (Pool Barcode Libraries by Tape Sta-
tion) as specified in the user manual v5 (see Note 28).
5. Finally go to STEP 5 (Size Selection) referring to the user
manual of Ampli1™ LowPass Kit v5 and to Note 29 and
conclude with STEP 6 (Clean up of Final Library) according
to the user manual of Ampli1™ LowPass Kit v5 and to Note
30.
Refer to your sequencing service provider for the final steps:
• STEP 7: Final Library Quantification.
• STEP 8: Sequencing setup.

3.3 Bioinformatic 1. Open Torrent Suite software and perform the pre-processing
Analysis and steps on raw sequences as indicated by the manufacturer:
Interpretation of • Quality check.
Sequencing Data
• Alignment to hg19 (TMAP aligner).
3.3.1 Pre-processing and 2. Obtain alignment file (.bam) from Torrent Suite. Exclude
Alignment (See Note 31) samples not passed quality control and/or samples with aligned
reads counts lower than 400,000.

3.3.2 Alignment Quality 1. Perform alignment quality control using FastqQC with stan-
Control dard settings. Do not consider “Per base sequence content,”
“Per sequence GC content,” “Sequence length distribution,”
110 Carolina Reduzzi et al.

and “Overrepresented sequences” parameters during the anal-

ysis. Exclude samples with >1 “Fail” returned by software.
2. Perform alignment quality control using Qualimap selecting
“BAM QC” as analysis type. Exclude samples with “mapping
quality mean” < 50.

3.3.3 Copy Number 1. Run ichorCNA on alignment file (.bam) with the following
Alteration Analysis and settings (see Note 32):
Tumor Fraction Estimation • Window = 1 Mb.
• Ploidy = 2.
• estimatePloidy = TRUE.
• estimateNormal = TRUE.
• normalPanel = TRUE.
• normal = c(0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9).
2. Retrieve the Copy number alteration calls for each segmented
genomic region (1 Mb) and estimated tumor fraction
(TF) from <sampleID>.seg.txt and < sampleID>.params.txt
files, respectively (see Note 33).
3. Do not consider chr19 during the evaluation of CNA profile
(see Note 34). Classify as unclear the CNA profiles showing
one of the following features:
• Normal profile but only 1 genomic region with amplifica-
tion/deletion lower than 125 Mb.
• Normal profile but sum of amplification/deletion of differ-
ent genomic regions lower than 375 Mb.
Classify all CNA profiles with alterations above these
thresholds as aberrant.
4. Classify samples as Normal, Tumor, or Unclear based on cri-
teria shown in the following table (see Notes 35 and 36).
Please see Fig. 2 for an example of two clusters containing
cells with aberrant DANA and one with normal diploid cells
(Table 1).

4 Notes

1. Store all reagents and enzymes according to the guidelines

specified in the corresponding data sheets provided by the
manufacturer, in a separate box from the DNA samples.
Avoid using reagents that have expired or reagents for which
you suspect an improper storage. Completely thaw and
pre-chill to 4 °C all reagents before preparation of reaction
master mix, as use of non-properly chilled reagents may cause
errors in pipette volumes. Since multiple freezing/thawing
 Isolation and Genomic Analysis of CTCs 111

Fig. 2 CNA profiles obtained with low-pass WGS from 3 CTC clusters: two with aberrant CNA profiles (upper)
and one with normal CNA profile (bottom). Color codes refer to 1 copy, 2 copies, 3 copies, and more than
4 copies for each single genomic region

Table 1
CNA classification of isolated clusters

TF + CNA profile = Final output

0 ≤ TF ≤ 0.05 + Normal/aberrant/unclear = Normal CTC cluster
0.05 < TF ≤ 1 + Aberrant = Aberrant CTC cluster
0.05 < TF ≤ 1 + Unclear = Unclear CTC cluster

cycles of temperature-sensitive reagents (i.e. enzymes, ATP,

and dNTPs) may impair their performance, we recommend
generating smaller aliquots to prevent their decomposition.
Put special attention to the water used for dilutions and con-
trols by using DNAse-free sterile water (provided in the kit)
stored in small aliquots and use a new aliquot for each run.
112 Carolina Reduzzi et al.

2. Use a programmable thermocycler with adjustable ramping

rates.
3. Use only dual-filter/barrier pipette tips to prevent the cross-
contamination of the samples. We recommend using flat-lid
PCR reaction tubes with shield cap which facilitates labeling of
samples.
4. It is recommended to designate a separate laboratory or work-
ing space dedicated to the WGA only. Always pay great atten-
tion to maintain high standards of purity and sterility in the
working space, which must absolutely be physically separated
from areas of the laboratory used for other molecular biology
activities, in particular post-PCR processing of samples, as this
could result in contaminations. Ideally, all steps of the proce-
dure should be performed under a dedicated laminar flow hood
equipped with the full set of laboratory equipment (i.e., dual
filter tips, dedicated set of pipettes, which must be used exclu-
sively used for the Ampi1™ WGA) and consumables. Pipettes
are particularly critical for a successful reaction and must be
calibrated and properly handled. Please refer to manufacturer
instructions for pipette-calibration and for pipette decontami-
nation. We recommend decontamination steps to be per-
formed each time experiments are run after a long
interruption time. Before starting each experiment always
clean the top of the laminar flow hood with the DNA-off
solution followed by sterile water and 70% ethanol. Always
use disposable lab coats and nitrile/powder-free gloves. To
save time all incubation steps of the protocol should be
pre-programmed on the thermal cycler.
5. It is critical that the operators are experienced in pipetting very
small volumes. Several tips may be useful: always pipette the
small volumes of reaction mixes gently, spin the tubes using a
microcentrifuge, pipette the desired volume of the reaction mix
onto the wall of the tube, without disturbing the fluid inside,
and collect it on the bottom of the tube with a short centrifu-
gation step. Samples should be briefly centrifuged after each
subsequent step of the procedure. To compensate for volume
losses during pipetting, it is suggested to include a 5% volume
excess in each reaction mix.
6. You can also use the Ampli1™ LowPass Kit for Illumina
(Menarini Silicon Biosystems).
7. Each filtration unit can process up to 3 mL of blood. To process
a higher volume, collect the blood in multiples of 3 and process
it using multiple filtration kits.
8. To avoid the risk of epithelial skin cell contamination, discard
the first ml of blood or collect the tubes for CTC analysis after
collection of blood samples for other analyses.
 Isolation and Genomic Analysis of CTCs 113

9. Filter the blood as soon as possible, better within 1 h. The

longer the blood is kept before filtration, the higher is the
possibility that clots will form, causing the stop of the filtration
before the end of the sample.
10. 50 mL tube racks can be used to hold the filtration units in
place during processing.
11. In case blood filtration stops, it is possible to try to restart it by
inserting the filtration unit into a new empty blood collection
tube (module C). No additional tubes are provided with the
kit, but it is possible to use conventional 10 mL EDTA tubes.
12. In alternative to May Grunwald Giemsa stain, it is possible to
stain with hematoxylin and eosin using the following protocol.
Put 70 μL of filtered hematoxylin on the IS positioned on the
parafilm. Incubate for 1 min and then, using the tweezers,
move the IS on the absorbent paper to absorb the hematoxylin.
Immediately rinse the IS by dipping it in a 50 mL conical tube
containing tap water, for 2/3 times, using the tweezers to hold
the IS. Put the filter on absorbent paper to absorb the excess of
water and put it back on the parafilm. Put 70 μL of eosin on the
IS and incubate for 30 s. Put the IS on the absorbent paper and
absorb the eosin. Immediately rinse the IS by dipping it in a
50 mL conical tube containing tap water, for 2/3 times, using
the tweezers to hold the IS. Put the filter on absorbent paper to
absorb the excess of water.
13. You can use one capillary per patient slide. Be careful not to
destroy the tip; use gloves.
14. Fill 50 μL of FBS in a 200 μL tube. Take the capillary and put it
into the FBS. FBS absorbs itself. Do not touch the tube surface
with the tip.
15. Do not flush out too much of the FBS. A little amount of FBS
should stay to prevent the cell attachment to the inner surface
of the picking capillary. If you have too little FBS in the
capillary, you will not be able to aspirate a cell. If you have
too much FBS in, the capillary pressure is too high and you
might lose the isolated cell.
16. If there are some other unwanted cells near the target cell,
remove the cells using the glass capillary by scratching, but be
sure that no one of the unwanted cell is attached in or outside
the glass capillary.
17. Do not aspirate the cell too less or too much into the picking
capillary. Too less: cell lost by moving out; too much: high
pressure. Should you aspirate too much, put out the liquid very
carefully until you can see the cluster.
18. To ensure a secure and stable hold, it is helpful to lean your
forearm against the microscope. If necessary, a box can also be
114 Carolina Reduzzi et al.

placed on the table. Then place your elbow on the box and lean
your forearm against the microscope. Caution: do not touch
any moving parts such as the microscope table with your
forearm.
19. This is a possible stopping point. Cells can be stored at -20 °C
until WGA.
20. This step can be done in parallel with STEP 1/STEP 2A by
using a different thermal cycler. Otherwise, the two steps must
be done subsequently in the same thermal cycler storing the
samples at +2–8 °C. This step is a pre-preparation of a compo-
nent for the reaction mix of STEP 3.
21. Once the program is ended, remove the tunes and store them
at -20 °C in a freezer located in a separate lab space dedicated
to downstream analysis of the amplified products.
22. Ampli1™ QC Kit is a POST WGA amplification kit. The
reaction must be prepared in the PCR working area, and
NOT in WGA working area. Pipettes for WGA should not be
used for this step. To save time all incubation steps of the
protocol should be pre-programmed on the thermal cycler.
23. In order to prevent any contamination, it is strongly recom-
mended to wear powder-free nitrile gloves for all protocol
steps, use only freshly opened plastic ware, dedicate a separate
working area, and use separate set pipettes for pre-PCR and for
library preparations. We recommend using a laminar-flow
hood and cleaning the top with DNA-off, sterile water, and
ethanol 70° before starting. To save time, all incubation steps
of the protocol should be pre-programmed on the thermal
cycler. Before starting, clean the laminar flow hood, prepare
the magnet, prepare the requested volume of ethanol 80%, the
dedicated pipettes, the LowTE buffer, the labeled 0.2 mL
tubes, and do not forget to vortex the beads making sure to
obtain a homogeneous solution. We suggest processing
12 samples in each single experiment.
24. To compensate for volume losses during pipetting, it is sug-
gested to include a 5% volume excess in each reaction mix.
When handling the barcodes, open one tube at time to prevent
any cross-contamination. Touch the side of tube with the
pipette tip and slowly dispense reagent on the side to form a
droplet.
25. Be very careful during removal of the supernatant to avoid
touching the beads loaded with the Ampli1™ WGA fragments.
Before discarding the supernatant make sure that there are no
beads in the tip. Should the pipette tip contain beads, carefully
discharge the supernatant back into its tube and wait a few
minutes to allow the optimal capture of the beads by the
magnet before removing the supernatant. For the first washing
 Isolation and Genomic Analysis of CTCs 115

step, 200 μL of ethanol 80% is added to beads; before proceed-

ing with the second washing step, remove 150 μL and replace
them with 150 μL of fresh ethanol, thereafter remove the
entire supernatant, by removing twice 110 μL of ethanol, and
remove all residual ethanol using a 10 μL micropipette. Do not
leave the beads get too much dry, to avoid reducing the effi-
ciency of the TE-elution step.
26. Always check that the PCR program is still stored in the Ther-
mal Cycler memory. Prepare the mix for 12.5 samples. Thaw all
the reagents except the Taq enzyme. Remove the Taq enzyme
from the freezer only last minute. Thaw, vortex, and briefly
spin the tubes containing the barcodes.
27. Use the same precautions as for step 1! If you decide not to
proceed immediately with step 4b, store the samples at 4 °C
(for 24 h storage) or at -20 °C (for long-time storage).
28. After the 1:5 dilution of the samples with LowTE, briefly
vortex and spin. Annotate the concentration (ng/μl) by select-
ing the area between 300 and 450 bp for each sample profile on
the TapeStation. Prepare a final equimolar pool of barcoded
libraries based on the concentrations measured by the TapeSta-
tion. For the pool preparation step, order the samples based on
their concentrations starting from the lowest one and divide
them into two sets. Use each half of the samples to prepare
pools paying attention to avoid having too large volume differ-
ences between the samples. This can be achieved by slightly
adjusting the volumes to maintain the equimolarity, but avoid
pipetting too small volumes. Do not pipette less than 1 μL per
sample to avoid inaccurate volumes. The final pool should be at
least 15 ng/μL and no more than 38 ng/μL in a final volume
of 44 μL. Each pool contains a maximum of 24 samples.
29. Where specified in the manufacturer’s instructions, use LowTE
instead of water (nuclease-free water) and before loading the
samples add to the wells 20 μL instead of 25 μL. The run
should be stopped when the 350 bp band starts to exit from
the central well (loaded with the marker) and the 450 bp band
starts entering the well. For this step, refer to the fluorescent
bands 350 and 450 from the reference lane, and monitor very
well the run-in order to avoid missing the best time for stop-
ping the run. Now you can collect the pool fraction containing
fragments ranging from 350 to 450 bp. Add a small volume of
LowTE to the wells and recover the pool fraction touching the
bottom of the well to be sure to collect the entire volume; do
not worry if you end up breaking the gel. The recovered
volume should be around 50–60 μL. Spin the pool at
14,000 × g for 10 min to remove gel leftovers and collect the
116 Carolina Reduzzi et al.

supernatant leaving a small volume of liquid on the bottom of

the tube.
30. Use the same precaution as in step 1! Store the samples at 4 °C
or at -20 °C (for long-time storage) if you decide not to
proceed with the next step.
31. In the case of samples processed using Illumina NGS platform,
consider the following Pre-processing & alignment step.
• Perform raw sequence quality control using FastqQC with
standard settings. Do not consider “Per base sequence con-
tent,” “Per sequence GC content,” “Sequence length dis-
tribution,” and “Overrepresented sequences” parameters
during the analysis. Exclude samples with >1 “Fail”
returned by software.
• Perform raw sequencing trimming and adapter clipping
using trimmomatic.
• Download reference genome hg19 from https://
hgdownload.soe.ucsc.edu/downloads.html.
• Perform alignment to human reference genome hg19 using
BWA-MEM with standard settings.
32. ichorCNA gives the possibility to create reproducible and scal-
able pipeline using snakemake (see https://github.com/
broadinstitute/ichorCNA/wiki/SnakeMake-pipeline-for-
ichorCNA).
33. The <sampleID>.seg.txt file reports copy number state in
terms of median log2 ratio and absolute copy number. In
order to classify a CNA profile as normal, aberrant, or unclear
(see step 3), only median log2 ratio could be considered using
the following threshold:
• Median log2 ratio ≥ 0.1: Gain.
• Median log2 ratio ≤ -0.1: Loss.
• -0.1 < Median log2 ratio < 0.1: Normal.
Complete CNA profile image is reported in <sam-
pleID>_genomeWide.pdf file.
34. Chr19 should not be considered during the evaluation of CNA
profile due to its biased deletion calls associated with the high
CG base percentage [10].
35. Use R software (or other programming language for data
analysis) to handle and process all data from steps 2–4.
36. It is important to note that the TF value returned by ichorCNA
is estimated only depending on the “amount” of genomic
aberration not corrected by the different degrees of CNAs
characterizing cancer cells within the CTC clusters. To cope
with this problem, we have previously published a method to
 Isolation and Genomic Analysis of CTCs 117

obtain a TF value corrected for the different degrees of DNA

alteration and for sequencing coverage [6].

References
1. Castro-Giner F, Aceto N (2020) Tracking can- (2021) Circulating tumor cell clusters are fre-
cer progression: from circulating tumor cells to quently detected in women with early-stage
metastasis. Genome Med 12(1):31. https:// breast cancer. Cancers (Basel) 13(10).
doi.org/10.1186/s13073-020-00728-3 https://doi.org/10.3390/cancers13102356
2. Aceto N (2020) Bring along your friends: 8. Okonechnikov K, Conesa A, Garcia-Alcalde F
homotypic and heterotypic circulating tumor (2016) Qualimap 2: advanced multi-sample
cell clustering to accelerate metastasis. Biom J quality control for high-throughput sequenc-
43(1):18–23. https://doi.org/10.1016/j.bj. ing data. Bioinformatics 32(2):292–294.
2019.11.002 https://doi.org/10.1093/bioinformatics/
3. Heeke S, Mograbi B, Alix-Panabieres C, Hof- btv566
man P (2019) Never travel alone: the crosstalk 9. Adalsteinsson VA, Ha G, Freeman SS,
of circulating tumor cells and the blood micro- Choudhury AD, Stover DG, Parsons HA,
environment. Cell 8(7). https://doi.org/10. Gydush G, Reed SC, Rotem D, Rhoades J,
3390/cells8070714 Loginov D, Livitz D, Rosebrock D,
4. Schuster E, Taftaf R, Reduzzi C, Albert MK, Leshchiner I, Kim J, Stewart C,
Romero-Calvo I, Liu H (2021) Better Rosenberg M, Francis JM, Zhang CZ,
together: circulating tumor cell clustering in Cohen O, Oh C, Ding H, Polak P, Lloyd M,
metastatic cancer. Trends Cancer 7(11): Mahmud S, Helvie K, Merrill MS, Santiago
1020–1032. https://doi.org/10.1016/j. RA, O’Connor EP, Jeong SH, Leeson R,
trecan.2021.07.001 Barry RM, Kramkowski JF, Zhang Z,
5. Pineiro R, Martinez-Pena I, Lopez-Lopez R Polacek L, Lohr JG, Schleicher M,
(2020) Relevance of CTC clusters in breast Lipscomb E, Saltzman A, Oliver NM,
cancer metastasis. Adv Exp Med Biol 1220: Marini L, Waks AG, Harshman LC, Tolaney
93–115. https://doi.org/10.1007/978-3- SM, Van Allen EM, Winer EP, Lin NU,
030-35805-1_7 Nakabayashi M, Taplin ME, Johannessen CM,
Garraway LA, Golub TR, Boehm JS, Wagle N,
6. Silvestri M, Reduzzi C, Feliciello G, Getz G, Love JC, Meyerson M (2017) Scalable
Vismara M, Schamberger T, Kostler C, whole-exome sequencing of cell-free DNA
Motta R, Calza S, Ferraris C, Vingiani A, reveals high concordance with metastatic
Pruneri G, Daidone MG, Klein CA, Polzer B, tumors. Nat Commun 8(1):1324. https://
Cappelletti V (2021) Detection of Genomically doi.org/10.1038/s41467-017-00965-y
aberrant cells within circulating tumor micro-
emboli (CTMs) isolated from early-stage breast 10. Zhao M, Wang Q, Wang Q, Jia P, Zhao Z
cancer patients. Cancers (Basel) 13(6). https:// (2013) Computational tools for copy number
doi.org/10.3390/cancers13061409 variation (CNV) detection using next-
generation sequencing data: features and per-
7. Reduzzi C, Di Cosimo S, Gerratana L, spectives. BMC Bioinform 14(Suppl 11).
Motta R, Martinetti A, Vingiani A, https://doi.org/10.1186/1471-2105-14-
D’Amico P, Zhang Y, Vismara M, S11-S1
Depretto C, Scaperrotta G, Folli S, Pruneri G,
Cristofanilli M, Daidone MG, Cappelletti V
 Chapter 8

Establishing Single-Cell Clones from In Vitro-Cultured

Circulating Tumor Cells
Teng Teng and Min Yu

Abstract
Cancer is a common health problem with more than 90% of deaths due to metastases. Circulating tumor
cells (CTCs) contain precursors that can initiate metastases. However, CTCs are rare, heterogeneous, and
difficult to expand in culture. We have previously created CTC-derived cell lines from stage IV breast cancer
patients. These CTC lines were used to establish single-cell CTC clones using flow cytometry cell sorting.

Key words Circulating tumor cells, Single-cell clones, CTC line culture, Breast cancer, Heterogeneity

1 Introduction

Cancer is a common disease worldwide, with survival rates varying

based on the type of cancer and geographical region [1]. For those
who don’t survive, metastatic tumors at sites distant from the
primary tumors are the main cause of cancer-related death
[2]. The “seeds” initiating the multi-step cascade of hematogenous
metastasis are a population of tumor cells in the circulatory system,
called circulating tumor cells (CTCs). CTCs originate from primary
or metastatic lesions, encounter a new set of stresses in the circula-
tion during their transit to distant sites, and extravasate into the
secondary organs. Here, some of these CTCs eventually create a
new metastatic lesion, or remain dormant [3, 4]. To demonstrate
this tumor-initiating capacity of CTCs in vivo, scientists recently
established CDX (CTC-derived xenograft) mouse models
[5]. Many studies have demonstrated the potential value of CTCs
in diagnosing disease, monitoring progression, predicting patient
outcomes, and designing therapeutic targets [6]. Recently, clinical
trials implemented CTC enumeration in order to monitor the
chemotherapeutic responses of prostate cancer patients [7]. There-
fore, CTCs hold promise as a “liquid biopsy” for cancer patients.

Miodrag Gužvić (ed.), Single Cell Analysis: Methods and Protocols, Methods in Molecular Biology, vol. 2752,
https://doi.org/10.1007/978-1-0716-3621-3_8, © Springer Science+Business Media, LLC, part of Springer Nature 2024

119
120 Teng Teng and Min Yu

Traditionally, biopsies require surgery, which is invasive. In

contrast, clinicians can isolate CTCs from the peripheral blood via
venipuncture, which is much less invasive. However, CTCs are
extremely rare and admixed in a huge number of normal blood
cells, presenting a significant challenge for isolation [8–10]. Tech-
nological advancement has led to the development of various
approaches to detect [11] and isolate CTCs [12–15]. Recent stud-
ies have found CTCs as single cells or clusters, which vary in
number depending on cancer type and stage. Many CTCs are
expected to be nonviable or fragile in vivo due to anoikis—cell
death due to loss of attachment—or stressful environmental con-
ditions in the blood, such as shear force. Additional CTCs may die
during the process of isolation.
Despite these challenges, we have successfully established
patient-derived CTC lines from luminal metastatic breast cancer
patients via ex vivo culture [16]. Using these CTC lines, we have
further succeeded in establishing single-cell CTC clones. Studies
have shown that CTCs demonstrate inter- and intra-patient hetero-
geneity [17, 18], and that clonal events can drive metastasis
[19]. Therefore, every viable CTC counts. Cultured single-cell
CTC clones will help us understand these cells’ genetic and epige-
netic heterogeneity, and correlate this heterogeneity with pheno-
typic analysis both in vitro and in vivo. In addition, established
single-cell clones will enable us to develop the right culture condi-
tions to establish more patient-derived CTC lines. Ultimately, we
may discover which of these single cell clones can create metastatic
lesions—providing crucial information for guiding the clinical
monitoring of cancer patients. In this protocol, we present a
method to generate single cell clones from heterogeneous in vitro
cultures of CTCs (Fig. 1).

2 Materials

2.1 Flow Cytometry 1% BSA in PBS buffer, keep on ice.

0.22 μ Syringe Filters, Sterile.
7-AAD 40×.
Trypan blue.
DRAQ5 (optional).
Hemocytometer.
Foil.

2.2 CTC Culture RPMI 1640 (with phenol red) supplemented with: 20 ng/mL
Recombinant Human EGF, 20 ng/mL Recombinant Human
FGF-basic, 0.02% v/v B27 Supplement, 10 mg/mL Gentamicin,
 Establishing Single-Cell Clones from In Vitro-Cultured Circulating Tumor Cells 121

a b c
Sorting Single CTC culture Clone expansion
Single CTC

Single CTC
suspended Cells reach
80% confluence
collection
Single CTC Separate to
Settle down several 96-wells

PBS+1%BSA

Take out media

from platform
dye Collect to
in the middle
24-wells

Single CTC
stay at bottom
space

Collect to
6-wells

Sorting single CTC into Adding fresh

GravityTRAP™ ULA Plate 96 well media

Fig. 1 Illustration of the protocol. (a)-(c) show the method of establishing single cell clones from in vitro
cultured circulating tumor cells: (a) flow cytometry (Subheading 3.1); (b) single CTC culture (Subheading 3.2);
and (c) establishment of single-cell clones (Subheading 3.2)

0.5% v/v amphotericin B/sodium deoxycholate (Gibco). Mix,

shake to achieve uniformity. Keep in 4 °C (see Note 1).
Alternatively, the following media can be used:
RPMI 1640 (with phenol red) supplemented with: 20 ng/mL
Recombinant Human EGF, 20 ng/mL Recombinant Human
FGF-basic, 0.02% v/v B27 Supplement, 0.01% Antibiotic-
Antimycotic 100×.
Mix, shake to achieve uniformity. Keep in 4 °C (see Note 1).
GravityTRAP™ ULA Plate 96 well (PerkinElmer) (see Note 6).
Ultra-low attachment 6-well plates.
Ultra-low attachment 6 cm plates.
Ultra-low attachment 24-well plates (optional).
Cell culture incubator with regulated levels of CO2 and O2.

2.3 Incubator Cells were cultured in a humid 37 C incubator with 5% CO2 and
Conditions for Single- 4% O2. Culture media was changed every 3–4 days.
Clone Cell Culture
122 Teng Teng and Min Yu

3 Methods

Keep all aforementioned materials at room temperature, unless

otherwise specified.
Prepare ice.

3.1 Flow Cytometry 1. Take the CTCs cultured in suspension in ultra-low attachment
plates and check the cell conditions under the microscope.
Generally, cells cluster in small clumps and stay in the center
of the well, and media looks clean and clear. Under the micro-
scope, CTCs should look healthy, round, clean, and clear with a
distinguishable cytoplasmic membrane, allowing light to pass
through and with minimal debris in the media (Fig. 2h).
2. Transfer CTCs to a 1.7 mL sterile Eppendorf tube and centri-
fuge for 10–12 s (speed around 1200r cf., recommended bench
mini-centrifuge; see Note 2). The cell pellet should be visible
(see Note 3). Remove the supernatant slowly using a pipet.
3. Wash the cell pellet using cold sterile 1% BSA in PBS buffer,
pipet up and down to mix, and centrifuge for 10–12 s (speed
around 1200r cf., recommended bench mini-centrifuge). The
cell pellet should be visible. Remove the supernatant slowly
using a pipet (optional; see Note 4).
4. Add 1 mL of cold sterile 1% BSA in PBS buffer to Eppendorf
tube, pipet up and down carefully about 100 times to fully mix,
and separate cells from each other (This step will affect the
sorting result.).
5. Remove the bubbles at the very top.
6. Transfer 5 μL to another Eppendorf tube, add 5 μL of Trypan
blue, mix, check under the microscope using a hemocytometer,
and ensure that the majority of cells are alive, healthy, and
single cells.
7. Add cold 1% BSA in PBS buffer to Eppendorf tube and make
the final volume 1 mL.
8. Add 7-AAD, which stains dead cells. Keep on ice, ready for
sorting (see Note 5).
9. Add DRAQ5, which stains viable cells (optional; see Note 5).
10. Sort single cells into GravityTRAP™ ULA Plate 96 well (Per-
kinElmer) that was pre-filled with CTC media 90–100 uL/
well, 1 cell per well (see Notes 6–9).

3.2 CTC Culture 1. After flow cytometry sorting, culture cells in a humid 37 °C
incubator with 5% CO2 and 4% O2 for 2–3 h. This will allow
the single cells to settle to the bottom of the well.
 Establishing Single-Cell Clones from In Vitro-Cultured Circulating Tumor Cells 123

Fig. 2 Example images of an established single CTC clone at different time points. (a)–(h) show a typical cell
expansion from single BRx-68 CTC under the microscope. Magnification times is 5× for (a)–(g), and 10× for
(h). A single BRx-68 CTC in the well (a) immediately after sorting, and after being in culture for (b) 6 days, (c)
18 days, (d) 39 days, (e) 51 days, (f) 60 days, (g) 69 days. (h) A BRx-68 single clone line being moved to a
24-well plate

2. Centrifuging the plates at 200 g for 1 m, with a recommended

acceleration and deceleration speed of 6, will remove bubbles
and provide a clear view (see Note 10).
3. Check plates under the microscope and record cell numbers in
each well, and discard wells without single cells.
4. Change media every 3–4 days using 65–75 μL/well (see Notes
6 and 11; Fig.1b), and count every 2–4 weeks. Try not to
disturb the cells too much.
5. When you see the cells cover more than 80% of the well’s space,
move them to several wells of a 96-well plate using procedure
described in Subhedading 3.1.
6. Once cells reach 80% confluence (Fig. 2g), collect them and
move them into a 24-well plate, using procedure described in
Subheading 3.1.
7. Repeat the steps 5 or 6 until you get enough cells (Fig. 1c; see
Note 12).

4 Notes

1. We have previously published this media for CTC culture and

routine maintenance [16].
2. In this step, we find that it is better to centrifuge CTCs at a
high speed for a short time (200 g, 5 min.). When using a low
124 Teng Teng and Min Yu

speed and longer time, it is difficult to appropriately adjust the

acceleration and deceleration, and the small number of cells can
stick to the walls, leading to the loss of many CTCs. In addi-
tion, shorter experimental times will lead to higher survival
efficiency.
3. Minimum cell number depends on your lab’s flow cytometry
efficiency and CTC cell culture survival rate.
4. This step will result in the loss of a lot of cells (optional).
5. Concentration of dye depends on the datasheet from respective
manufacturer.
6. In this kind of plate, the well has a super tiny flat bottom to
hold cells in place (diameter of bottom is 1 mm). In compari-
son, the diameter of a regular 96-well plate is 6.4 mm. This
design facilitates finding the cells within the well under 4× or
10× magnification, monitoring them, and enhancing their sur-
vival (Fig. 2a). In the middle of each well, there is an additional
platform for resting the pipet tip while changing media, in
order to prevent accidentally aspirating suspended CTCs at
the bottom (Fig. 1b).
7. Do this step as fast as possible, since cells can die if they spend a
long time outside the incubator. If there is a long distance
between the lab and sorting facility, you can wrap the plates
with aluminum foil to protect them from the light and ambient
oxygen.
8. Flow cytometry sorting equipment has differing efficiencies,
which can significantly influence results.
9. Avoid sorting the cells into the edges of this 96-well plate, since
evaporation at the edges can kill cells (optional).
10. After sorting, give the cells time (3 h or less) to naturally settle
to the bottom. If you centrifuge immediately after sorting, you
can potentially miss CTCs that stick to the walls of the well. If
you let the cells settle for more than 3 h, single cells may
proliferate, causing you to discard wells as having more than
one cell.
11. Change the tip for each well in order to avoid cell
contamination.
12. It is best to keep the cells at a high confluence, in order to
promote survival and proliferation.

Acknowledgments

Our research is funded by the NIH K22 award (K22 CA175228-

01A1), NIH DP2 award (DP2 CA206653), Donald E. and Delia
B. Baxter Foundation, Stop Cancer Foundation, Wright
 Establishing Single-Cell Clones from In Vitro-Cultured Circulating Tumor Cells 125

Foundation, the Pew Charitable Trusts, and the Alexander & Mar-
garet Stewart Trust (Pew-Stewart Scholar for Cancer Research) to
M.Y. We are grateful to Ms. C. Lytal for assisting with manuscript
editing.

References
1. Allemani C, Weir HK, Carreira H, Cancer Res 12(14 Pt 1):4218–4224. https://
Harewood R, Spika D, Wang XS, Bannon F, doi.org/10.1158/1078-0432.CCR-05-2821
Ahn JV, Johnson CJ, Bonaventure A, Marcos- 7. Heller G, McCormack R, Kheoh T, Molina A,
Gragera R, Stiller C, Azevedo e Silva G, Chen Smith MR, Dreicer R, Saad F, de Wit R, Aftab
WQ, Ogunbiyi OJ, Rachet B, Soeberg MJ, DT, Hirmand M, Limon A, Fizazi K,
You H, Matsuda T, Bielska-Lasota M, Fleisher M, de Bono JS, Scher HI (2018) Cir-
Storm H, Tucker TC, Coleman MP, Group culating tumor cell number as a response mea-
CW (2015) Global surveillance of cancer sur- sure of prolonged survival for metastatic
vival 1995–2009: analysis of individual data for castration-resistant prostate cancer: a compari-
25,676,887 patients from 279 population- son with prostate-specific antigen across five
based registries in 67 countries (CONCORD- randomized phase III clinical trials. J Clin
2). Lancet 385(9972):977–1010. https://doi. Oncol 36(6):572–580. https://doi.org/10.
org/10.1016/S0140-6736(14)62038-9 1200/JCO.2017.75.2998
2. Weigelt B, Peterse JL, van’t Veer LJ (2005) 8. Khoo BL, Grenci G, Lim YB, Lee SC, Han J,
Breast cancer metastasis: markers and models. Lim CT (2018) Expansion of patient-derived
Nat Rev Cancer 5(8):591–602. https://doi. circulating tumor cells from liquid biopsies
org/10.1038/nrc1670 using a CTC microfluidic culture device. Nat
3. Yadavalli S, Jayaram S, Manda SS, Madugundu Protoc 13(1):34–58. https://doi.org/10.
AK, Nayakanti DS, Tan TZ, Bhat R, 1038/nprot.2017.125
Rangarajan A, Chatterjee A, Gowda H, Thiery 9. Stott SL, Lee RJ, Nagrath S, Yu M, Miyamoto
JP, Kumar P (2017) Data-driven discovery of DT, Ulkus L, Inserra EJ, Ulman M, Springer S,
extravasation pathway in circulating tumor Nakamura Z, Moore AL, Tsukrov DI, Kemp-
cells. Sci Rep 7:43710. https://doi.org/10. ner ME, Dahl DM, Wu CL, Iafrate AJ, Smith
1038/srep43710 MR, Tompkins RG, Sequist LV, Toner M,
4. Aceto N, Bardia A, Miyamoto DT, Donaldson Haber DA, Maheswaran S (2010) Isolation
MC, Wittner BS, Spencer JA, Yu M, Pely A, and characterization of circulating tumor cells
Engstrom A, Zhu H, Brannigan BW, Kapur R, from patients with localized and metastatic
Stott SL, Shioda T, Ramaswamy S, Ting DT, prostate cancer. Sci Transl Med 2(25):25ra23.
Lin CP, Toner M, Haber DA, Maheswaran S https://doi.org/10.1126/scitranslmed.
(2014) Circulating tumor cell clusters are oli- 3000403
goclonal precursors of breast cancer metastasis. 10. Hodgkinson CL, Morrow CJ, Li Y, Metcalf
Cell 158(5):1110–1122. https://doi.org/10. RL, Rothwell DG, Trapani F, Polanski R,
1016/j.cell.2014.07.013 Burt DJ, Simpson KL, Morris K, Pepper SD,
5. Baccelli I, Schneeweiss A, Riethdorf S, Nonaka D, Greystoke A, Kelly P, Bola B, Krebs
Stenzinger A, Schillert A, Vogel V, Klein C, MG, Antonello J, Ayub M, Faulkner S,
Saini M, Bauerle T, Wallwiener M, Holland- Priest L, Carter L, Tate C, Miller CJ,
Letz T, Hofner T, Sprick M, Scharpff M, Blackhall F, Brady G, Dive C (2014) Tumori-
Marme F, Sinn HP, Pantel K, Weichert W, genicity and genetic profiling of circulating
Trumpp A (2013) Identification of a popula- tumor cells in small-cell lung cancer. Nat Med
tion of blood circulating tumor cells from 20(8):897–903. https://doi.org/10.1038/
breast cancer patients that initiates metastasis nm.3600
in a xenograft assay. Nat Biotechnol 31(6): 11. USFDA (2008) CallSearch-circulating tumor
539–544. https://doi.org/10.1038/nbt. cell kit: K073338
2576 12. Karabacak NM, Spuhler PS, Fachin F, Lim EJ,
6. Hayes DF, Cristofanilli M, Budd GT, Ellis MJ, Pai V, Ozkumur E, Martel JM, Kojic N,
Stopeck A, Miller MC, Matera J, Allard WJ, Smith K, Chen PI, Yang J, Hwang H,
Doyle GV, Terstappen LW (2006) Circulating Morgan B, Trautwein J, Barber TA, Stott SL,
tumor cells at each follow-up time point during Maheswaran S, Kapur R, Haber DA, Toner M
therapy of metastatic breast cancer patients pre- (2014) Microfluidic, marker-free isolation of
dict progression-free and overall survival. Clin
126 Teng Teng and Min Yu

circulating tumor cells from blood samples. 16. Yu M, Bardia A, Aceto N, Bersani F, Madden
Nat Protoc 9(3):694–710. https://doi.org/ MW, Donaldson MC, Desai R, Zhu H,
10.1038/nprot.2014.044 Comaills V, Zheng Z, Wittner BS, Stojanov P,
13. Sarioglu AF, Aceto N, Kojic N, Donaldson Brachtel E, Sgroi D, Kapur R, Shioda T, Ting
MC, Zeinali M, Hamza B, Engstrom A, DT, Ramaswamy S, Getz G, Iafrate AJ,
Zhu H, Sundaresan TK, Miyamoto DT, Benes C, Toner M, Maheswaran S, Haber DA
Luo X, Bardia A, Wittner BS, Ramaswamy S, (2014) Cancer therapy. Ex vivo culture of cir-
Shioda T, Ting DT, Stott SL, Kapur R, culating breast tumor cells for individualized
Maheswaran S, Haber DA, Toner M (2015) A testing of drug susceptibility. Science
microfluidic device for label-free, physical cap- 345(6193):216–220. https://doi.org/10.
ture of circulating tumor cell clusters. Nat 1126/science.1253533
Methods 12(7):685–691. https://doi.org/ 17. De Mattos-Arruda L, Cortes J, Santarpia L,
10.1038/nmeth.3404 Vivancos A, Tabernero J, Reis-Filho JS, Seoane
14. Fachin F, Spuhler P, Martel-Foley JM, Edd JF, J (2013) Circulating tumour cells and cell-free
Barber TA, Walsh J, Karabacak M, Pai V, Yu M, DNA as tools for managing breast cancer. Nat
Smith K, Hwang H, Yang J, Shah S, Rev Clin Oncol 10(7):377–389. https://doi.
Yarmush R, Sequist LV, Stott SL, org/10.1038/nrclinonc.2013.80
Maheswaran S, Haber DA, Kapur R, Toner M 18. Messaritakis I, Stoltidis D, Kotsakis A, Dermit-
(2017) Monolithic chip for high-throughput zaki EK, Koinis F, Lagoudaki E,
blood cell depletion to sort rare circulating Koutsopoulos A, Politaki E, Apostolaki S,
tumor cells. Sci Rep 7(1):10936. https://doi. Souglakos J, Georgoulias V (2017) TTF-1-
org/10.1038/s41598-017-11119-x and/or CD56-positive circulating tumor cells
15. Chen L, Peng M, Li N, Song Q, Yao Y, Xu B, in patients with small cell lung cancer (SCLC).
Liu H, Ruan P (2018) Combined use of Sci Rep 7:45351. https://doi.org/10.1038/
EpCAM and FRalpha enables the high- srep45351
efficiency capture of circulating tumor cells in 19. Hunter KW, Amin R, Deasy S, Ha NH, Wake-
non-small cell lung cancer. Sci Rep 8(1):1188. field L (2018) Genetic insights into the morass
https://doi.org/10.1038/s41598-018- of metastatic heterogeneity. Nat Rev Cancer
19391-1 18(4):211–223. https://doi.org/10.1038/
nrc.2017.126
 Chapter 9

Immunofluorescence Combined with Single-Molecule RNA

Fluorescence In Situ Hybridization for Concurrent Detection
of Proteins and Transcripts in Stress Granules
Jakub Kochan and Mateusz Wawro

Abstract
Immunofluorescence (IF) microscopy is arguably one of the most commonly used methods for studying
structure and composition of stress granules (SGs). While in most cases standard IF protocols are sufficient
to visualize protein components of SGs, concurrent detection of proteins and transcripts in stress granules
requires more sophisticated and problematic approaches. Here we present a well-established, simple,
robust, and fluorescent protein-compatible method for simultaneous detection of proteins and transcripts
in individual stress granules using combination of IF and single-molecule RNA fluorescence in situ
hybridization (smRNA FISH).

Key words Microscopy, Immunofluorescence, smRNA FISH, Stress granules, RNA, lncRNA, Single-
cell analysis

1 Introduction

Stress granules (SGs) are large ribonucleoprotein complexes pre-

dominantly comprised of stalled translation initiation complexes,
RNA-binding proteins, and signaling factors engaged in numerous
aspects of cellular metabolism. These non-membrane-bound
assemblies of RNAs and proteins form when cells are subjected to
adverse conditions such as heat shock (e.g., hyperthermia), viral
infections, oxidative stress, UV light, amino acid deprivation, or
osmotic shock. SGs are dynamic structures involved in RNA triage
and modulation of various signaling cascades determining whether
stressed cells should eventually survive or undergo apoptosis [1, 2].
The key feature of stress granules is that virtually all known
protein components found in SGs pre-exist dispersed in the cyto-
plasm of resting cells and rapidly and reversibly accumulate by
liquid-liquid phase separation into a condensed state upon expo-
sure to stress conditions. In this condensed phase, the RNAs and

Miodrag Gužvić (ed.), Single Cell Analysis: Methods and Protocols, Methods in Molecular Biology, vol. 2752,
https://doi.org/10.1007/978-1-0716-3621-3_9, © Springer Science+Business Media, LLC, part of Springer Nature 2024

127
128 Jakub Kochan and Mateusz Wawro

RNA-binding proteins behave as a single organelle with liquid-like

properties forming cytologically discernible foci [2–4].
The protein composition, structure, and dynamics of SG for-
mation have been extensively studied over the last two decades.
Due to the fact that formation of SGs is a result of subcellular
re-localization of specific components rather than change in their
cellular level, the method of choice in studies on SGs was and still is
fluorescence microscopy (Fig. 1). This approach has provided valu-
able insights into the spatio-temporal organization of protein com-
ponents of stress granules and shed light on the molecular
mechanisms governing formation of SGs [5].
Whereas the dynamics of SGs formation and their main protein
components were recently characterized [1, 6–8], the RNA com-
position of SGs still remains poorly explored. Only a few specific
mRNAs have been shown to localize to SGs. The full population of
mRNAs in SGs, and whether SGs contain any specific noncoding
RNAs (ncRNAs), is not known. Only recently, in an attempt to
describe the mammalian stress granule transcriptome using
RNA-seq analysis, Khong et al. revealed that while essentially
every mRNA can be targeted to stress granules, there are also
certain lncRNAs that are overrepresented in these structures
[9]. These observations along with spectacular progress in SG
research led to increased interest in procedures allowing for con-
current observations of specific SG markers (proteins) and tran-
scripts in stress granules at the level of single cells.
Though the concept of linking immunofluorescence with sin-
gle-molecule RNA fluorescence in situ hybridization (smRNA
FISH) seems apparent, the specific materials used during IF and
smRNA FISH make this approach difficult to perform on the same
specimen. Here we present a well-established, straightforward,
robust, and fluorescent protein-compatible method for simulta-
neous detection of proteins and transcripts in individual SGs
using combination of IF and smRNA FISH. The presented proto-
col is based on our previous seminal technical report published in
2015 [10]. To date, our procedure has been successfully used by
several groups. Lately we managed to adopt the original protocol to
visualize components of stress granules. Here we present protocol
allowing for simultaneous observation of a key protein marker of
SGs, G3BP1, and lncRNA NORAD recently found to be one of the
top five most abundant lncRNAs in human SGs [9].

2 Materials

Read this section carefully and ensure to have all reagents prepared
before beginning the procedure. Also make sure to read the
detailed protocol and respective notes in advance. Observe caution
not to introduce RNase contamination into your reagents and
 Protein and mRNA Detection Using Simultaneous IF and RNA-FISH 129

Fig. 1 Formation of stress granules is a result of subcellular re-localization of specific components rather than
change in their cellular level. (a) Western blot analysis of extracts from HeLa cells, control or sodium arsenite-
treated (500 μM, 30 and 60 min), using anti-G3BP1 (upper panel), anti-phospho-eIF2α (Ser51, middle panel),
and anti-lamin (loading control, lower panel) antibodies. While no change in total G3BP level is detected upon
treatment with sodium arsenite, observed phosphorylation of eIF2α clearly reveals exposure of cells to stress
conditions. (b) Immunofluorescence-based approach allows observation of stress granules formed upon
treatment of the cells with sodium arsenite. Images of HeLa cells with visualized G3BP1, a key marker of
stress granules (red) in control and sodium arsenite-treated cells. Blue: DAPI (nuclei). Scale bar: 10 μm

processed specimens. Wear gloves throughout the whole experi-

ment and change them often, especially when you suspect contact
with skin, hair, or other surfaces potentially contaminated with
RNases (e.g., keyboards, doorknobs etc.). Use of sterile, RNase-
free containers and consumables at every step is recommended (see
Note 1). Prepare all solutions using ultrapure (i.e., 18.2 MΩ·cm at
25 °C) RNase-free water (see Note 2). We do not add sodium azide
or other preservatives to the prepared reagents.

2.1 Cell Culture 1. HeLa cells (human cervix adenocarcinoma).

2. Dulbecco’s Modified Eagle Medium (DMEM) with 1.0 g/L
D-glucose (Lonza) supplemented with GlutaMAX (Gibco) and
10% fetal bovine serum (FBS, heat inactivated, South America
origin, Biowest).
3. Phosphate buffered saline (PBS, without calcium and
magnesium).
4. TrypLE Express dissociating reagent (Gibco). No antibiotics
(i.e., penicillin or streptomycin) were used. Cells were main-
tained in cell culture vessels from BD Falcon.
5. Sterile glass coverslips (see Note 3).
6. Sterile tweezers (see Note 4).

2.2 Fixation/Post- 1. RNase-free Phosphate-Buffered Saline (PBS), 1× sterile:

fixation 137 mM NaCl, 2.7 mM KCl, 8 mM Na2HPO4, and 2 mM
KH2PO4. Prepare by diluting Ambion 10× RNase-free PBS
pH 7.4 (Invitrogen) to 1× solution in UltraPure DNase/
130 Jakub Kochan and Mateusz Wawro

RNase-Free Distilled Water (Invitrogen) or equivalent (see

Note 2). Store at room temperature (RT).
2. Fixative: 4% (w/v) methanol-free formaldehyde in 1× RNase-
free PBS. In a 50 mL sterile, RNase-free conical tube (or ap-
propriate sterile, RNase-free glass bottle), mix 10 mL of 16%
(w/v) methanol-free formaldehyde (Invitrogen; use whole
ampule at once), 4 mL of 10× RNase-free PBS pH 7.4 (Invi-
trogen), and 26 mL of UltraPure DNase/RNase-Free Distilled
Water (Invitrogen) or equivalent. Store at 4 °C in a container
wrapped with aluminum foil.

2.3 Immuno- 1. RNase-free PBS, 1× sterile: see previous section.

fluorescence 2. 10% (v/v) Triton X-100 in RNase-free water: mix 5 mL of
molecular biology-grade Triton X-100 and 45 mL of Ultra-
Pure™ DNase/RNase-Free Distilled Water (Invitrogen) or
equivalent. Sterile filter through a 0.2 μm syringe filter. Store
at 4 °C (see Note 5).
3. 200 mM vanadyl ribonucleoside complexes (VRC): we use
molecular biology-grade VRC (Merck/Sigma-Aldrich, prod-
uct number 94740). Reconstitute the powder in sterile,
RNase-free water to a green-black clear solution by incubating
securely sealed vial at 65 °C (about 10 min with occasional
mixing). Once prepared, aliquot the entire volume into smaller
samples and keep frozen at -20 °C. After thawing ensure that
the solution is clear. If a visible precipitate formed, re-heat the
aliquot at 65 °C.
4. 10% (w/v) sterile-filtered, RNase-free (acetylated) bovine
serum albumin (Ac-BSA). We prepare Ac-BSA according to
the protocol “Removal of RNase from Bovine Serum Albumin
by Acetylation Reaction” available online from web resources
of Malaysian Cocoa Board (http://cocoabiotech.koko.gov.
my/RNases%20Elimination1.html). The protocol can be
scaled-up to at least 200 mL. After dialysis against sterile,
RNase-free water, we concentrate the obtained Ac-BSA using
Amicon Ultra centrifugal filters (Ultracel 50 K) in refrigerated
centrifuge (4 °C) equipped with swing-bucket rotor capable of
accommodating 50 mL tubes to obtain 10% w/v Ac-BSA.
Determine the final protein concentration using BCA assay.
Filter sterilize through a 0.2 μm filter and store 2 mL aliquots
at -20 °C (see Notes 6 and 7).
5. Blocking/permeabilizing buffer: 2 mM VRC, 1% (w/v)
Ac-BSA, 0.3% (v/v) Triton X-100, 1× sterile, RNase-free
PBS. Prepare in a 50 mL sterile, RNase-free conical tube (or ap-
propriate sterile, RNase-free glass bottle). For 20 mL, mix
200 μl of 200 mM VRC, 2 mL of 10% Ac-BSA, 608 μl of
10% (v/v) Triton X-100, 2 mL of sterile 10× RNase-free
 Protein and mRNA Detection Using Simultaneous IF and RNA-FISH 131

Fig. 2 The humidifying chamber assembly. Components used to build the humidifying chamber are presented
in the left picture: (1) paper towels, (2) 50 mL conical tube screws, (3) parafilm, (4) distilled water, (5) glass
plate, and (6) glass Petri dish. Right image presents a typical view of an assembled chamber (a – a glass plate
with piece of parafilm placed on top (stage for coverslips), (b) – an orientation mark on the parafilm used for
proper alignment of coverslips)

PBS, and add sterile, RNase-free water to 20 mL. Sterile filter

through a 0.2 μm syringe filter. Store at 4 °C in a container
wrapped with aluminum foil.
6. Primary and secondary antibodies. We use mouse monoclonal
antibodies anti-human G3BP1 (250 μg/mL, BD Transduction
Laboratories, 611126), diluted 1:50 and goat anti-mouse IgG
(H + L) highly cross-adsorbed secondary antibodies conju-
gated with Alexa Fluor 488 (2 mg/mL, Invitrogen,
A-11029), diluted 1:500).
7. Humidifying chamber (Fig. 2).

2.4 smRNA FISH 1. 20× SSC buffer: 3 M sodium chloride, 300 mM sodium citrate.
We use Ambion 20× SSC solution (Invitrogen). Store at
RT. 20× SSC can be also easily prepared in house (see Note 8).
2. 2× SSC buffer. Prepare in a 50 mL sterile, RNase-free conical
tube. For 50 mL mix 5 mL 20× SSC and 45 sterile, RNase-free
water. Store at RT.
3. Wash buffer: 2× SSC, 10% (v/v) formamide. Prepare in a
50 mL sterile, RNase-free conical tube. For 50 mL, mix 5 mL
20× SSC, 5 mL formamide, and add sterile, RNase-free water
to 50 mL (see Note 9). Sterile filter through a 0.2 μm syringe
filter. Store at RT.
132 Jakub Kochan and Mateusz Wawro

Fig. 3 Specificity verification of used probe blend targeting lncRNA NORAD. (a) Detection of NORAD (red) using
smRNA FISH in wild-type HeLa cells (left panel), mock-transfected (middle panel), and HeLaiKD cells with
induced NORAD knock-down (right panel). Blue: DAPI (nuclei). Scale bar: 10 μm (b) Quantification of the
smRNA FISH experiments shown in (a). Twenty-five cells were analyzed for each condition. (C) qPCR analysis
of NORAD expression in mock-transfected and HeLaiKD cells with induced NORAD knock-down. Bars: average
from two independent experiments with two technical replicates. Error bars: 95% CI

4. Hybridization buffer: 2× SSC, 10% (v/v) formamide, 10%

(w/v) dextran sulfate. Prepare in a 50 mL sterile, RNase-free
conical tube. For 10 mL, mix 1 mL of 20× SSC, 1 mL form-
amide, 1 g of dextran sulfate, and add sterile, RNase-free water
to 10 mL. Store 1 mL aliquots at -20 °C (see Note 10).
5. smRNA FISH probe blend. We used custom-designed probe
blend labelled with Quasar 570 dye (LGC Biosearch Technol-
ogies) targeting lncRNA NORAD at the concentration of
125 nM (Fig. 3). Probes were designed using the Stellaris
Probe Designer (version 4.2) available online free of charge
(www.biosearchtech.com).

2.5 Imaging In principle every modern fluorescence microscope can be used to

image prepared specimens; therefore, we suggest to consult your
local microscopy facility operators for possible options. All images
presented in this chapter were obtained using a Leica DM6B
upright widefield fluorescence microscope (Leica Microsystems).
We used a 63× oil immersion objective and a 12-bit Leica
DMC5400 CMOS camera (Leica Microsystems) with Leica LAS
X image acquisition software. The following filter sets (Leica
Microsystems) were used: A4 for detection of DAPI (nuclei),
GFP ET for detection of Alexa Fluor 488 (IF signal), and RHOD
ET for Quasar 570 dye (smRNA FISH signal). After acquisition of
about 30 z-sections with 0.3 μm spacing, images were deconvolved
using Huygens Professional Software (Scientific Volume Imaging).
Final image adjustments were performed using ImageJ 1.53c
(National Institutes of Health) and Adobe Illustrator 2020
(Adobe Systems).
 Protein and mRNA Detection Using Simultaneous IF and RNA-FISH 133

3 Methods

The presented protocol, including all reagent volumes, is optimized

for 12-well plate format but of course it can be scaled up or down
according to specific needs. The procedure usually requires 4–5
consecutive days of work. Remember to book a proper fluorescence
microscope in advance as imaging freshly prepared specimens gives
the best quality of results. It is not recommended to store the speci-
mens longer then overnight. If longer storage of prepared specimens
is needed, diligently follow the storage instructions provided by the
manufacturer of the used mounting medium.

3.1 Plating and 1. Using tweezers place sterile glass coverslips in a multi-well cell
Treatment of the Cells culture plate and seed the cells at the desired density suitable for
single-cell imaging and analysis. Usually 50,000–60,000 cells
per well of 12-well plate in 1 mL of medium should be suffi-
cient. In general, it is not recommended to let the cells reach
100% confluence (i.e., monolayer) at the time of fixation (see
Note 11).
2. Place the cells in a dedicated incubator for 24–48 h to allow
proper attachment and accommodation. If needed, change the
medium at desired time point or treat the cells according to
your experimental requirements.
3. Induce stress granule formation (see Note 12) and proceed
immediately to subsequent steps.

3.2 Fixation 1. Using a sterile, RNase-free Pasteur pipette connected to a

vacuum pump, remove and dispose properly the culture
medium.
2. Wash the cells twice with 1× PBS (1 mL/well) (see Note 13).
Use a serological RNase-free pipette. Work gently and fluently.
Do not let the cells to dry out.
3. Remove PBS, add 1 mL of fixative, and incubate the cells for
10–15 min at room temperature (see Note 14).
4. Remove fixative and wash the cells 3 times with 1× PBS (1 mL/
well, 5 min at room temperature). Gentle agitation using an
orbital shaker will facilitate proper washing (see Note 15).

3.3 Immuno- 1. Following fixation, remove PBS and block/permeabilize the

fluorescence cells for 60 min at RT with gentle shaking using 1 mL/well of
blocking buffer (see Note 16).
2. While the cells are being blocked/permeabilized, prepare pri-
mary antibody by diluting them in blocking buffer. When
calculating the volume of the needed antibody solution, take
50 μl of antibody solution per specimen (coverslip) and add
134 Jakub Kochan and Mateusz Wawro

Fig. 4 Picture of the bent needle attached to a syringe used to manipulate coverslips. Zooms present the tip of
the bent needle in more details

50 μl of excess to compensate for possible liquid losses and

pipetting errors.
3. Assemble the humidifying chamber omitting the parafilm-
coated glass plate and equilibrate it at 4 °C (see Fig. 2).
4. Carefully place 50 μl droplets of primary antibody solution on
the parafilm. Avoid any air bubbles in the droplets as the cells
will not be stained in these areas. While placing the droplets pay
attention to arrange them in a manner allowing to place the
coverslips separately and not touching each other.
5. Place the coverslips (cells toward the liquid) using tweezers and
a bent needle (fresh, sterile, RNase-free) attached to a syringe
on the primary antibody droplets (Figs. 4 and 5). Do not squeeze
the cells.
6. Place the glass plate with specimens in humidifying chamber
and incubate the cells with primary antibodies overnight at 4 °
C.
7. Following the overnight incubation with primary antibodies,
gently transfer the specimens into a fresh, sterile cell culture
plate wells filled with 1× PBS (rotate the coverslips so the cells are
facing upwards). Again, be careful not to disturb or dry the
cells.
8. Wash the cells three times with 1× PBS (5 min at RT with
gentle agitation each).
9. While washing the specimens, equilibrate the humidifying
chamber to RT and set up a glass plate with a new piece of
parafilm placed on top.
Protein and mRNA Detection Using Simultaneous IF and RNA-FISH 135

Fig. 5 Photographs presenting the essential steps during coverslip manipulation.

Upper panel shows technique used to lift the coverslip from cell-culture plate
well. Below a series of images presents how to place the coverslip (cells toward
the droplet) using tweezers on the antibody/probe droplet
136 Jakub Kochan and Mateusz Wawro

10. Prepare secondary antibody dilution in blocking buffer (vol-

ume of secondary antibodies should be calculated similarly to
the volume of primary antibodies as in step 2).
11. Place 50 μl droplets of secondary antibodies on the parafilm.
12. Use tweezers and a bent needle (a new one) attached to a
syringe to place coverslips (remember to keep the cells towards
the droplet) onto the secondary antibody droplets.
13. Gently transfer the glass plate with coverslips into the humidi-
fying chamber and incubate the specimens with secondary
antibodies for 90 min at RT, in the dark (see Note 17).
14. Following the incubation with secondary antibodies, transfer
the specimens into a fresh, sterile cell culture plate wells filled
with 1× PBS (remember to rotate the coverslips and keep the cells
facing upwards). Be careful not to disturb or dry the cells.
15. Wash as in step 8.
16. Remove PBS after last wash, add 1 mL of fixative (4% formal-
dehyde in PBS), and incubate for 10 min at RT (in the dark).
The post-fixation step prevents removal of antibodies during
smRNA FISH steps.
17. Remove the fixative and wash as in step 8. Keep the cells in PBS
until starting smRNA FISH procedure.

3.4 Single-Molecule 1. Prepare fresh smRNA FISH wash buffer and thaw a fresh
RNA Fluorescence In aliquot of hybridization buffer.
Situ Hybridization 2. Remove PBS and equilibrate the specimens by incubation in
smRNA FISH wash buffer (1 mL/well) for 5–10 min at RT.
3. During specimen equilibration, prepare probe blend working
solution. Dilute the probe blend stock in hybridization buffer.
When calculating the volume of the probe, you need to follow a
similar routine as for the primary and secondary antibodies.
Take 50 μl of probe droplet per coverslip and add 50 μl excess
to compensate for liquid losses.
4. Equilibrate the humidifying chamber to 37 °C and prepare a
glass plate with a piece of parafilm placed on it (use the same
chamber construction as in previous steps).
5. Place droplets of probe blend (50 μl/coverslip) on the parafilm.
6. Use tweezers and a bent needle (yet another new one) attached
to a syringe to place coverslips (cells toward the droplet) onto the
probe blend droplets (see Note 18).
7. Hybridize the cells with probes at 37 °C overnight in the dark.
8. Following the overnight hybridization, gently transfer the spe-
cimens into a fresh, sterile cell culture plate wells filled with
smRNA FISH wash buffer prewarmed to 37 °C (rotate the
 Protein and mRNA Detection Using Simultaneous IF and RNA-FISH 137

coverslips so the cells are facing upwards). Again, be careful not

to damage the cells (see Note 19).
9. Wash the specimens twice with 1 mL of smRNA FISH wash
buffer per coverslip (30 min at 37 °C, agitation is not required).
Add nuclear counterstain (e.g., DAPI or equivalent) to the
smRNA FISH wash buffer during the second wash (see Note
20).
10. Remove smRNA FISH wash buffer and immerse the specimens
in 2× SSC while preparing to mount the coverslips onto micro-
scope slides.
11. Mount the specimens in an appropriate fluorescence micros-
copy mounting medium and proceed to imaging.

3.5 Imaging We recommend acquiring three-dimensional stacks of images using

widefield fluorescence microscope. A good starting point is a 1 s or
longer exposure time to detect the smRNA FISH signal. Often we
use 1500 ms exposure time when working with probe blends
labeled with Quasar 570 Dye. Acquire stacks of about 30 Z-slices
at intervals of 300 nm in the Z direction that can be then recon-
structed (deconvolved) using appropriate software (we use Huy-
gens Professional Software). We strongly advice to test different
deconvolution settings at initial stages of your experiments as it may
differ between used microscopy platforms. It is not uncommon to
be unable to see separated RNA spots before deconvolution
(Fig. 6). Figure 7 presents anticipated results of the described
procedure.

4 Notes

1. In most cases disposable cell culture plasticware is certified to

be RNase-free. To remove RNases from glassware or metalware
bake them at 300 °C for 4 h or 180 °C or higher for several
hours. Alternatively, small metalware (e.g., spoons or spatulas)
can be quickly decontaminated by holding them in a Bunsen
burner flame for several seconds.
2. In most cases we use UltraPure DNase/Rnase-Free Distilled
Water purchased from Invitrogen. However, when larger
volumes are needed, we prefer to prepare sterile, RNase-free
water in-house by diethyl pyrocarbonate (DEPC) treatment.
To prepare DEPC-treated water, add DEPC to ultrapure water
to a final concentration of 0.1% (v/v), i.e., 500 μl DEPC per
500 mL of water, mix the solutions thoroughly to disperse the
DEPC droplets in whole volume, place the bottles in an incu-
bator shaker providing further continuous mixing overnight,
138 Jakub Kochan and Mateusz Wawro

Fig. 6 Comparison of smRNA FISH images (maximum intensity projections) before (left panel) and after
deconvolution (right panel). Scale bar: 10 μm

Fig. 7 Hyperthermia and sodium arsenite induce stress granule formation and lead to sequestration of lncRNA
NORAD in stress granules. IF-smRNA FISH images of HeLa cells stained to visualize NORAD (red) and G3BP1, a
key marker of SGs (green), during two different types of stresses (hyperthermia and sodium arsenite). Blue:
DAPI (nuclei). Scale bars: 10 μm (upper panels), 1 μm (lower panels)

and then autoclave to remove the DEPC. Store at room tem-

perature. Use sterilely.
3. Choice of shape and dimensions of coverslips depends on the
cell culture plate format to be used and specific experimental
requirements. As standard, for 12-well plate, we use
15 mM × 15 mM glass coverslips (No. 1.5H (170 ± 5 μm),
Protein and mRNA Detection Using Simultaneous IF and RNA-FISH 139

D 263 M Schott borosilicate glass, Menzel-Gl€aser). Sterilize

the coverslips in a glass Petri dish by baking for at least 2 h at
180 °C.
4. Clean the tweezers using a regular detergent followed with
extensive wash with tap water, distilled (demi) water, and even-
tually with 96% ethanol. Next, place the precleaned tweezers in
the laminar flow hood and either burn them using Bunsen
burner (remember to let it cool down before use) or
UV-sterilize. Sterile tweezers may be kept on a freshly opened
Petri dish or cell culture dish until needed.
5. Preparation of 10% (v/v) Triton X-100 may take a long while.
While preparing mix the contents gently to avoid foaming.
Prepare in advance and store at 4 °C.
6. For determination of Ac-BSA concentration using BCA assay,
an Ac-BSA standard must be used. Due to severe modification
after acetic anhydride treatment, normal BSA standards cannot
be used. Aliquots of Ac-BSA with determined concentration
may be found as components of restriction enzyme kits or
purchased separately. We can also provide small aliquots of
our own Ac-BSA.
7. Preparation of Ac-BSA requires use of acetic anhydride which is
a restricted reagent. Because of its use in the synthesis of
heroin, acetic anhydride is listed as a drug precursor and its
purchase is strictly controlled. Special permissions may be
required.
8. To prepare 50 mL of 20× SSC buffer, dissolve 8.765 g of
sodium chloride and 1.410 g of trisodium citrate (dihydrate)
in sterile, RNase-free water. Bring total volume to 50 mL.
9. Formamide is a known teratogen and should be used in a
chemical fume hood only. Once opened, the bottle should be
purged with inert gas (we use argon or nitrogen) and stored
frozen to prevent oxidation. The breakdown products of form-
amide may degrade nucleic acids. Prepare fresh wash buffer for
each experiment and only the volume that is needed. Use right
after preparation.
10. We use dextran sulfate sodium salt from Leuconostoc spp.
(Merck, D8906). When preparing hybridization buffer, add
dextran sulfate as the last component and not at once but
part after part to prevent clog formation. It may take up to
45 min. Mix gently. The hybridization buffer is very viscous.
We recommend preparation of the hybridization buffer in bulk
(10 mL) immediately after opening a fresh bottle of formamide
and store 1 mL aliquots of the prepared buffer at -20 °C until
use. This avoids any freeze-thaw effects and offers fresh hybri-
dization buffer for each experiment. We have not done
140 Jakub Kochan and Mateusz Wawro

extensive stability studies but, from our experience, the ali-

quots will perform to expectation for 2–3 months when stored
frozen.
11. Be aware that the exact culture conditions depend on the type
of used cells and desired experiment conditions. In this proce-
dure, we used HeLa cells that usually do not require any special
treatment, but some cell lines do not attach to bare (i.e.,
uncoated) glass coverslips and a cell line-specific treatment of
the surface may be required (e.g., coating of the coverslips with
poly-L-lysine, fibronectin, laminin, or other agents).
12. In the presented protocol, we induced formation of stress
granules using the most commonly applied stressors: hyper-
thermia (42 °C, 1 h) or sodium arsenite (500 μM, 1 h). Wear
protective clothing, eyewear, and gloves when handling
sodium arsenite. Discard all excess sodium arsenite and media
containing sodium arsenite reagents according to your institu-
tion’s guidelines and all applicable local requirements.
13. From now on observe extreme caution not to introduce any
RNase contamination into the specimens. All used reagents
must be RNase-free.
14. Do not discard the remaining fixative. It will be used in post-
fixation step. Store the fixative at RT, in the dark (we wrap it in
aluminum foil).
15. Possible rest point. From our experience, at this step the speci-
mens may be kept in RNase-free PBS for some time (several
hours) but we strongly recommended to process the specimens
immediately after fixation to obtain best results. Do not store
them overnight. In our hands the specimens are kept in RNase-
free PBS at room temperature only for the time needed for
preparation of blocking buffer.
16. Remember not to discard the remaining blocking buffer. It will
be used to dilute primary and secondary antibodies. Store the
buffer at 4 °C, in the dark (we wrap it in aluminum foil).
17. From this step onward, the specimens must be protected from
light. Wrap the plate in which the specimens are washed in
aluminum foil.
18. Droplets of probe in hybridization buffer exhibit different
behavior compared to droplets of antibodies. Due to the pres-
ence of formamide and dextran sulfate, the hybridization
buffer is more viscous and the coverslips seem not to touch
fully the parafilm. It may give an impression of a stack of
parafilm, buffer and coverslip with clearly distinguishable
layers.
19. After overnight incubation with probe blend in hybridization
buffer, the coverslips may stick to the parafilm. Observe
 Protein and mRNA Detection Using Simultaneous IF and RNA-FISH 141

extreme caution while levering them with needle and forceps.

Use gentle and smooth move. When placed in wash buffer, the
coverslips may float on the surface of the liquid. Ensure they are
covered with the wash buffer and sunk to the bottom of
the well.
20. In our hands addition of nuclear counterstain to the wash
buffer during second wash instead of using mounting medium
with counterstain results in more even staining of nuclei and
prevents the halo-like effect when imaging nuclei.

Acknowledgement

This work was supported by grant from National Science Center,

Poland (project number 2020/39/D/NZ3/02328 to J.K).

References
1. Panas MD, Ivanov P, Anderson P (2016) Franzmann TM (2020) RNA-induced confor-
Mechanistic insights into mammalian stress mational switching and clustering of G3BP
granule dynamics. J Cell Biol 215(3): drive stress granule assembly by condensation.
313–323. https://doi.org/10.1083/jcb. Cell 181(2):346–361 e317. https://doi.org/
201609081 10.1016/j.cell.2020.03.049
2. Protter DSW, Parker R (2016) Principles and 7. Sanders DW, Kedersha N, Lee DSW, Strom
properties of stress granules. Trends Cell Biol AR, Drake V, Riback JA, Bracha D, Eeftens
26(9):668–679. https://doi.org/10.1016/j. JM, Iwanicki A, Wang A, Wei MT,
tcb.2016.05.004 Whitney G, Lyons SM, Anderson P, Jacobs
3. Molliex A, Temirov J, Lee J, Coughlin M, WM, Ivanov P, Brangwynne CP (2020) Com-
Kanagaraj AP, Kim HJ, Mittag T, Taylor JP peting protein-RNA interaction networks con-
(2015) Phase separation by low complexity trol multiphase intracellular organization. Cell
domains promotes stress granule assembly and 181(2):306–324 e328. https://doi.org/10.
drives pathological fibrillization. Cell 163(1): 1016/j.cell.2020.03.050
123–133. https://doi.org/10.1016/j.cell. 8. Yang P, Mathieu C, Kolaitis RM, Zhang P,
2015.09.015 Messing J, Yurtsever U, Yang Z, Wu J, Li Y,
4. Perez-Pepe M, Fernandez-Alvarez AJ, Boccac- Pan Q, Yu J, Martin EW, Mittag T, Kim HJ,
cio GL (2018) Life and work of stress granules Taylor JP (2020) G3BP1 is a tunable switch
and processing bodies: new insights into their that triggers phase separation to assemble stress
formation and function. Biochemistry 57(17): granules. Cell 181(2):325–345 e328. https://
2488–2498. https://doi.org/10.1021/acs. doi.org/10.1016/j.cell.2020.03.046
biochem.8b00025 9. Khong A, Matheny T, Jain S, Mitchell SF,
5. Van Treeck B, Parker R (2019) Principles of Wheeler JR, Parker R (2017) The stress gran-
stress granules revealed by imaging approaches. ule transcriptome reveals principles of mRNA
Cold Spring Harb Perspect Biol 11(2). accumulation in stress granules. Mol Cell
https://doi.org/10.1101/cshperspect. 68(4):808–820 e805. https://doi.org/10.
a033068 1016/j.molcel.2017.10.015
6. Guillen-Boixet J, Kopach A, Holehouse AS, 10. Kochan J, Wawro M, Kasza A (2015) Simulta-
Wittmann S, Jahnel M, Schlussler R, Kim K, neous detection of mRNA and protein in single
Trussina I, Wang J, Mateju D, Poser I, cells using immunofluorescence-combined sin-
Maharana S, Ruer-Gruss M, Richter D, gle-molecule RNA FISH. Biotechniques
Zhang X, Chang YT, Guck J, Honigmann A, 59(4):209–212, 214, 216 passim. https://
Mahamid J, Hyman AA, Pappu RV, Alberti S, doi.org/10.2144/000114340
 Chapter 10

Highly Multiplexed and Simultaneous Characterization

of Protein and RNA in Single Cells by Flow or Mass
Cytometry Platforms Using Proximity Ligation Assay
for RNA
Andrew D. Duckworth, Joseph R. Slupsky, and Nagesh Kalakonda

Abstract
In situ hybridization of oligonucleotide probes to intracellular RNA allows quantification of predefined
gene transcripts within millions of single cells using cytometry platforms. Previous methods have been
hindered by the number of RNA that can be analyzed simultaneously. Here we describe a method called
proximity ligation assay for RNA (PLAYR) that permits highly multiplexed RNA analysis that can be
combined with antibody staining. Potentially any number of RNA combined with antigen can be analyzed
together, being limited only by the number of analytes that can be measured simultaneously.

Key words Transcriptome analysis, Flow cytometry, Single-cell analysis, Single-cell multi-omics, In
situ hybridization, Mass spectrometry, Proteome analysis, Proximity ligation assay for RNA (PLAYR)

1 Introduction

Single-cell RNA analysis can currently be performed using two

main complimentary methods of sequencing or in situ hybridiza-
tion. Reverse transcription of RNA to DNA and its sequencing
using high-throughput platforms can provide transcriptome-level
information in single cells, yet a trade-off is ultimately observed
between the sensitivity of the procedure and the number of cells
analyzed due to sequencing depth and the resulting high financial
cost. In situ hybridization on the other hand provides sensitive
analysis of predefined RNA targets in hundreds of cells per second
using cytometry-based platforms. In situ RNA hybridization meth-
ods usually contain three main steps: firstly, transcripts are specifi-
cally targeted using oligonucleotide probes; secondly, a transcript-
associated sequence is replicated to amplify the signal; thirdly, this
replicated transcript-specific sequence is then labeled using a com-
plimentary DNA detection probe (DDP) that is labeled with

Miodrag Gužvić (ed.), Single Cell Analysis: Methods and Protocols, Methods in Molecular Biology, vol. 2752,
https://doi.org/10.1007/978-1-0716-3621-3_10, © Springer Science+Business Media, LLC, part of Springer Nature 2024

143
144 Andrew D. Duckworth et al.

analyte. Previous methods [1–8] have been limited in the specificity

as well as the number of RNA that can be simultaneously analyzed.
Here we describe a proximity ligation assay for RNA (PLAYR)
[9, 10], an in situ single-cell RNA quantification technique that
removes multiplexing constraints and greatly minimizes off-target
signals. In addition, as PLAYR is performed at physiological con-
ditions, the assay can be combined with the study of antigen
expression using conventional antibody staining. Specific RNA
detection with minimal background staining by PLAYR is facili-
tated through a method of proximal ligation [11, 12]. Paired oli-
gonucleotide probes (probe pairs) are designed to anneal adjacently
along their target transcript within fixed and permeabilized cells via
their 5′ DNA sequence. We recommend using four to six probe
pairs along the length of each target RNA to provide sufficient assay
sensitivity. The number used will depend upon the copy number of
the specific transcript within each cell and the efficiency of binding
for each probe pair. The requirement of multiple probe pairs per
transcript to maintain sensitivity of the assay usually limits PLAYR
to measuring RNA which have sufficient length (e.g., mRNA,
lncRNA) and viable sequence composition (i.e., sufficient GC con-
tent) to accommodate binding of multiple probe pairs. When the
probe pairs are annealed and positioned alongside one another on
their target RNA, their 3′ sequences combine to produce a plat-
form for the circularization of two further DNA oligonucleotides
termed the ‘insert’ and ‘backbone’. The insert sequence is specific
for each transcript while the backbone is universal. Once annealed
to both probe pairs the insert and backbone form a single-stranded
DNA circle that can be ligated. This completed circular DNA
molecule is then copied using rolling circle amplification (RCA)
which is catalyzed by the Phi29 polymerase enzyme [13]. An illus-
tration of how the PLAYR oligonucleotides bind together is shown
in Fig. 1. We recommend 2 h and 4 h RCA times for flow and mass
cytometry, respectively. Increasing the RCA time beyond this can
increase the signal but not generally the sensitivity of the assay (i.e.,
negative cells remain negative). DNA amplification during RCA
initiates from the 3′ end of one of the probe pairs, producing a
linear DNA molecule that is composed of potentially hundreds of
concatenated complimentary copies of the original rolling circle.
Thus, transcript-specific complimentary insert sequences, which
have been replicated during the RCA, can then be detected using
DDPs which are conjugated to a specific analyte of your choice.
PLAYR is particularly powerful technique when combined with
mass cytometry [14, 15] (PLAYR-CyTOF), where up to 50 analytes
can be concurrently measured. To this end, we present an addi-
tional sub-method for labeling DDPs with metal analytes for use in
mass cytometry in Methods Subheading 3.1 of this protocol. In
addition, cells analyzed by PLAYR-CyTOF may be pooled using
live-cell barcoding (LCB) techniques [16–18], in which
 Highly Multiplexed and Simultaneous Characterization of Protein and RNA in. . . 145

A
Probe1
5’-NNNNNNNNNNNNNNNNNNNAAAAAAAAAANNNNNNNNNGACACTCTT-3’

Probe2
5’-NNNNNNNNNNNNNNNNNNNAAAAAAAAAAGACGCTAATNNNNNNNNN-3’
RNA-binding sequence Linker backbone- insert-
binding binding
sequence sequence

B
Backbone

Insert
…CTGCGATTA NNNNNNNNNTTTNNNNNNNNN CTGTGAGAA…
GACGCTAATNNNNNNNNN NNNNNNNNNGACACTCTT RCA begins
AAAAAAAAAA

AAAAAAAAAA

here
Probe2 Probe1

NNNNNNNNNNNNNNNNNNN NNNNNNNNNNNNNNNNNNN
5’ –NNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNN-3’ RNA

Fig. 1 Overview of PLAYR probe design and cognate interactions between oligonucleotides within the PLAYR
systems. (a) Overview of the PLAYR probe design, with each sequence component shaded differently. (b)
Diagram showing sequence interactions between different oligonucleotides during PLAYR. “N” represents
nucleotides that vary depending upon the target (RNA-binding sequence) and the PLAYR system (insert-
binding sequence) used

ubiquitously expressed abundant antigen(s) are targeted to label

each sample with different combinations of heavy-metal isotopes.
See Table 1 for an example of an LCB scheme using Cadmium-
conjugated antibodies for barcoding of up to 20 samples.
There are currently 27 sets of PLAYR systems (i.e.,
corresponding probe pair 3′ ends, inserts, backbones, and DDPs)
each allowing simultaneous quantification of a transcript. All
27 PLAYR system sequences are provided (Table 2), with room
for expansion if required. In addition, we provide validated prede-
signed 5′ probe pair sequences for targeting common
“housekeeping” transcripts (Table 3). Instructions on installing
and using PLAYRDesign software, an R software which aids in
picking 5′ probe pair sequences and designing PLAYR probes for
your RNA target of interest, is available at https://github.com/
nolanlab/PLAYRDesign. Specificity and accurate quantification of
transcripts using designed PLAYR probes should be validated by
146 Andrew D. Duckworth et al.

Table 1
Example of a 6-choose-3 LCB scheme using Cadmium isotopes

Cadmium Isotopes
106 110 111 113 114 116
Barcode 1 x x x
2 x x x
3 x x x
4 x x x
5 x x x
6 x x x
7 x x x
8 x x x
9 x x x
10 x x x
11 x x x
12 x x x
13 x x x
14 x x x
15 x x x
16 x x x
17 x x x
18 x x x
19 x x x
20 x x x
An antibody (or antibodies) that target constituently and highly expressed antigen(s) are
labeled with different cadmium isotopes and combined in sets of three (x’s) to produce
LCBs. Cadmium isotopes can be substituted for other metal isotopes and can be tailored
around your panel design. If antibodies target the same epitope and their staining signal
is weak due to competition for antigen binding, barcoding may perform better with just
two isotopes per barcode (e.g., a 7-choose-2 scheme provides 21 barcodes)

comparison of mean PLAYR signal with measurements within bulk

samples using quantitative PCR. We have found that perturbation
of the targeted transcripts using inhibition or stimulation of an
appropriate cell line produces time point samples in which
measured RNA quantity closely correlates between both techni-
ques (Fig. 2).
Figure 3 shows a schematic of the antigen and RNA staining
procedure by PLAYR. Methods Subheading 3.2 provides the steps
required to stain surface antigen, label dead cells, and prepare
samples prior to PLAYR. If using mass cytometry, LCB can be
Table 2
Oligonucleotide sequences for 27 PLAYR systems

System 3′ Probe1 3′ Probe2 Insert Detection Probe

1 AAAAAAAAAACTCAGTCG AAAAAAAAAAGACGCTAATA P-ACGACTGAGTTTGGTCAC Z-ACGACTGAGTTTGGTCAC
TGACACTCTT TCGTGACC GAT GAT
2 AAAAAAAAAATATCG AAAAAAAAAAGACGCTAATC P-CGGACGATATTTGTCTC Z-CGGACGATATTTGTCTC
TCCGGACACTCTT TTCGAGAC GAAG GAAG
3 AAAAAAAAAAGATTCC AAAAAAAAAAGACGCTAATC P-CGAGGAATCTTTCATTGG Z-CGAGGAATCTTTCATTGG
TCGGACACTCTT TGCCAATG CAG CAG
4 AAAAAAAAAACTACC AAAAAAAAAAGACGCTAA P-CCAAGGTAGTTTAG Z-CCAAGGTAGTTTAG
TTGGGACACTCTT TCAGGCTACT TAGCCTG TAGCCTG
5 AAAAAAAAAACTC AAAAAAAAAAGACGCTAA P-CTCGAAGAGTTT Z-CTCGAAGAGTTT
TTCGAGGACACTCTT TCACCAGTTG CAACTGGTG CAACTGGTG
6 AAAAAAAAAACTTAGCC AAAAAAAAAAGACGCTAA P-CAGGCTAAGTTTA Z-CAGGCTAAGTTTA
TGGACACTCTT TCCAGACTGT CAGTCTGG CAGTCTGG
7 AAAAAAAAAACCGCTTA AAAAAAAAAAGACGCTAATC P-CATAAGCGGTTTGCCATG Z-CATAAGCGGTTTGCCATG
TGGACACTCTT TACATGGC TAG TAG
8 AAAAAAAAAACTCGATC AAAAAAAAAAGACGCTAA P-CAGATCGAGTTTAC Z-CAGATCGAGTTTAC
TGGACACTCTT TCAACCTGGT CAGGTTG CAGGTTG
9 AAAAAAAAAATACG AAAAAAAAAAGACGCTAA P-CGGAACGTATTTCTCAG Z-CGGAACGTATTTCTCAG
TTCCGGACACTCTT TCATCCTGAG GATG GATG
10 AAAAAAAAAACTGCTCA AAAAAAAAAAGACGCTAA P-CATGAGCAGTTTA Z-CATGAGCAGTTTA
TGGACACTCTT TCGCAAGTCT GACTTGCG GACTTGCG
11 AAAAAAAAAATGACTC AAAAAAAAAAGACGCTAATC P-CGAGAGTCATTTGATTCC Z-CGAGAGTCATTTGATTCC
TCGGACACTCTT TCGGAATC GAG GAG
Highly Multiplexed and Simultaneous Characterization of Protein and RNA in. . .

12 AAAAAAAAAACTGTC AAAAAAAAAAGACGCTAA P-CGTAGACAGTTTGA Z-CGTAGACAGTTTGA

TACGGACACTCTT TCACAGTGTC CACTGTG CACTGTG
147

(continued)
Table 2
148

(continued)

System 3′ Probe1 3′ Probe2 Insert Detection Probe

13 AAAAAAAAAATTC AAAAAAAAAAGACGCTAATC P-CCTGGAGAATTTCCGATT Z-CCTGGAGAATTTCCGATT
TCCAGGGACACTCTT TCAATCGG GAG GAG
14 AAAAAAAAAACACTTG AAAAAAAAAAGACGCTAA P-CGACAAGTGTTTAGG Z-CGACAAGTGTTTAGG
TCGGACACTCTT TCAGATGCCT CATCTG CATCTG
15 AAAAAAAAAACTTC AAAAAAAAAAGACGCTAA P-CTGCAGAAGTTTA Z-CTGCAGAAGTTTA
TGCAGGACACTCTT TCCAGGATCT GATCCTGG GATCCTGG
Andrew D. Duckworth et al.

16 AAAAAAAAAATCTA AAAAAAAAAAGACGCTAATC P-CCGGATAGATTTGGTCTA Z-CCGGATAGATTTGGTCTA

TCCGGGACACTCTT TGTAGACC CAG CAG
17 AAAAAAAAAACGCATC AAAAAAAAAAGACGCTAATC P-CAAGATGCGTTTATGTGC Z-CAAGATGCGTTTATGTGC
TTGGACACTCTT TGGCACAT CAG CAG
18 AAAAAAAAAATCTCACG AAAAAAAAAAGACGCTAA P-CACGTGAGATTTCATTC Z-CACGTGAGATTTCATTC
TGGACACTCTT TCCTCGAATG GAGG GAGG
19 AAAAAAAAAATCGCTAC AAAAAAAAAAGACGCTAA P-CAGTAGCGATTTAT Z-CAGTAGCGATTTAT
TGGACACTCTT TCGCCATGAT CATGGCG CATGGCG
20 AAAAAAAAAATACGCTC AAAAAAAAAAGACGCTAA P-CAGAGCGTATTTC Z-CAGAGCGTATTTC
TGGACACTCTT TCACACTTGG CAAGTGTG CAAGTGTG
21 AAAAAAAAAACGTC AAAAAAAAAAGACGCTAA P-CGTAAGACGTTTAGTCG Z-CGTAAGACGTTTAGTCG
TTACGGACACTCTT TCATGCGACT CATG CATG
22 AAAAAAAAAACCATTCG AAAAAAAAAAGACGCTAA P-CACGAATGGTTTACGCT Z-CACGAATGGTTTACGCT
TGGACACTCTT TCATCAGCGT GATG GATG
23 AAAAAAAAAACCTAG AAAAAAAAAAGACGCTAATC P-CGAACTAGGTTTCCTAGG Z-CGAACTAGGTTTCCTAGG
TTCGGACACTCTT TACCTAGG TAG TAG
24 AAAAAAAAAACTCCGA AAAAAAAAAAGACGCTAA P-CAATCGGAGTT Z-CAATCGGAGTT
TTGGACACTCTT TCAACGCGTT TAACGCGTTG TAACGCGTTG
 25 AAAAAAAAAATTCGCAC AAAAAAAAAAGACGCTAA P-CAGTGCGAATTTCCG Z-CAGTGCGAATTTCCG
TGGACACTCTT TCAATTCCGG GAATTG GAATTG
26 AAAAAAAAAATCC AAAAAAAAAAGACGCTAA P-CCTGAAGGATTTACT Z-CCTGAAGGATTTACT
TTCAGGGACACTCTT TCCGCTAAGT TAGCGG TAGCGG
27 AAAAAAAAAACGTTAC AAAAAAAAAAGACGCTAATC P-CGAGTAACGTTTGCGCT Z-CGAGTAACGTTTGCGCT
TCGGACACTCTT TTAAGCGC TAAG TAAG
Universal backbone: P-ATTAGCGTCCAGTGAATGCGAGTCCGTCTAGGAGAGTAGTACAGCAGCCGTCAAGAGTGTC
P, 5′-phosphorylated; Z, 5′-C6-thiol modification for mass cytometry or fluorophore modification for flow cytometry
Highly Multiplexed and Simultaneous Characterization of Protein and RNA in. . .
149
150 Andrew D. Duckworth et al.

Table 3
List of 5′ PLAYR probe sequences that target constituently expressed RNA in human cells

Probe
Gene pair 5′ Probe1 5′ Probe2
GAPDH 1 TGATGGCAACAATATCCACTTT GTTAAAAGCAGCCCTGGTGA
2 TGGAAGATGGTGATGGGATT ATTGATGACAAGCTTCCCGT
3 TGGACTCCACGACGTACTCA CATCGCCCCACTTGATTTT
4 CCTGCTTCACCACCTTCTTG TCATCATATTTGGCAGGTTTTT
ACTB 1 TGCCAATGGTGATGACCTG GTCAGGCAGCTCGTAGCTCT
2 TGTTGGCGTACAGGTCTTTG ATGTCCACGTCACACTTCATG
3 CGTCATACTCCTGCTTGCTG ATCTGCTGGAAGGTGGACAG
4 TGTTTTCTGCGCAAGTTAGGT TCAAGAAAGGGTGTAACGCA
RPL27 1 GTTTCATGAACTTGCCCA GGCCCTACCAGCAAAAAGGAAAG
TTTCGG
2 GCGATCTGAGGTGCCATCATCAA TCTTCACGATGACAGCTTTGCG
T TC
3 TCTCTTGGCGATCTTCTTC GCTGTCACTTTGCGGGGGTAG
TTGCC
4 TCTTCAAACTTGACCTTGGCC GCCTTGCGTTTAAGAGCAGGATC
TCC T
5 AGCATCTAAAACCGCAGTTTC ACCACTTGTTCTTGCCTGTCTTG
TGG T
Table shows predesigned 5′ sequences for PLAYR probe pairs targeting the indicated “housekeeping” transcripts. To
complete the PLAYR probe, the 3′ Probe1 or 2 sequence from the PLAYR systems table (Table 2; any system may be
chosen) needs to be added to their 3′ end

performed as the first procedure, allowing samples to be pooled

into one tube prior to further downstream processing. All incuba-
tions for antigen staining are performed with live cells which are
placed on ice to avoid any potential changes in transcript expression
caused by antibody binding. To reduce loss in surface antibody
staining during PLAYR we perform an additional amine-to-amine
crosslinking step also on ice. LCB and phenotyping antibody stain-
ing, as well as live/dead cell discrimination, are optional and can be
individually excluded without affecting one another or the RNA
detection by PLAYR. Once stained, cells are treated with parafor-
maldehyde and permeabilized in methanol, after which there is an
opportunity to pause the protocol by storing the sample(s) at
-80 °C. Staining the RNA using PLAYR (Methods Subheading
3.3) takes approximately 10 h to complete for mass cytometry and
8 h for flow cytometry, with frequent incubation steps. If required,
internal antigen staining can be performed in conjunction with the
 Highly Multiplexed and Simultaneous Characterization of Protein and RNA in. . . 151

Time(min)
0
PMA+Iono

30
60
120
240
360
RNA1 RNA2 RNA3 RNA4 RNA5 RNA6 RNA7 ACTB

B C RNA1 RNA2
RNA1 RNA2

PLAYR (Mean Signal)

RNA3 RNA4
RNA3 RNA4

qPCR (a.u)

Fig. 2 Validation of PLAYR probe specificity. B-cell line MAVER1 treated with 100 nM phorbol 12-myristate
13-acetate (PMA) + 1 μM ionomycin (PMA + Iono) for up to 6 h was examined by PLAYR (mass cytometry) and
qPCR for expression changes in eight RNAs (including ACTB). (a) Overlayed histograms showing single-cell
expression for each of the eight transcripts at each timepoint. (b) Comparison of mean PLAYR signal (right-
hand y axis) with quantification by qPCR (left-hand y axis) for 4 RNAs shown in (a) over the time course with
PMA + Iono. (c) Linear regression of data shown in (b) showing concordance of relative signal for PLAYR
(y axis) and qPCR (x axis) at each timepoint; gray shading indicates 95% confidence intervals

last step of PLAYR where DDP is added. After RNA and internal
antigen staining, cells should be run immediately for analysis by
flow cytometry. In contrast, cells can be left in fixative and analyzed
up to 48 h later using mass cytometry. Instructions on how to
process samples for analysis by mass cytometry are described in
Methods Subheading 3.4 of the protocol. The protocol has the
potential to be adapted for non-suspension cells for quantification
of RNA in tissues or adherent cells using microscopy.

2 Materials

All buffers are made and stored at room temperature (RT) unless
stated otherwise. Filter tips should be used for all mass cytometry
and PLAYR methods. All reagents should be RNase and DNase
152 Andrew D. Duckworth et al.

Fig. 3 Schematic of PLAYR staining procedure. Schematic overview of the PLAYR assay, with the top panel
showing the stages (1–5) for sample preparation and the bottom panel illustrating the stages (6–10) for RNA
staining with PLAYR. Stages 1–4 are all optional and, if not required, can be excluded without affecting the
detection of RNA with PLAYR. For PLAYR-CYTOF, live cell barcoding can be performed first (stage 1), allowing
multiple samples to be pooled into one tube. Dead cells can then be stained at RT (stage 2) followed by surface
antigen staining on ice (stage 3). Covalent crosslinking of surface antibodies using BS3 reagent is then
performed (stage 4) on ice to help maintain antigen binding during the remaining stages of PLAYR. Cells are
then fixed in PFA and permeabilized with methanol in stage 5. Once in methanol, cells can be stored at -80 °C
allowing the assay to be paused. Stages 6–10 illustrate the sequential steps of oligonucleotide binding and
amplification used by PLAYR. Internal antibody staining can be included in the final stage alongside incubation
with DDPs

free. If performing PLAYR-CyTOF, then extra care must be taken

to ensure the reagents are metal free, particularly those used in the
final wash steps of the assay; Fluidigm supplies ddH2O, phosphate
buffer saline (PBS), and cell staining buffer (CSB) that are all
guaranteed to be metal free.

2.1 Labeling DPPs 1. Microcentrifuge.

with Metal Isotopes for 2. Water bath.
Mass Cytometry
3. Ice-cold 100% ethanol.
4. Ice-cold 75% ethanol.
5. 3 M Sodium acetate: Dissolve in ddH2O and adjust the pH to
5.2 with glacial acetic acid.
6. 500 mM Tris(2-carboxyethyl)phosphine hydrochloride
(TCEP) solution.
7. Maxpar X8 Antibody Labeling Kit (Fluidigm). The choice of
metal for each DDP depends upon your panel design.
8. NanoDrop spectrophotometer.
9. Centrifugal filter with 30 kDa molecular weight cut off
(MWCO).
10. Centrifugal filter with 3 kDa MWCO.
 Highly Multiplexed and Simultaneous Characterization of Protein and RNA in. . . 153

11. 250 μM DDP carrying a 5′-thiol-modifier C6 S-S: Purified by

HPLC and dissolved in ddH2O. Store at -20 °C. (See Table 2
for list of sequences.)

2.2 General 1. Refrigerated microcentrifuge (swing rotor is advised, see Note

Materials for PLAYR 1).
2. ddH2O.
3. PBS (RT and ice cold).
4. CSB: 0.5% bovine serum albumin dissolved in PBS. Store at
4 °C. This reagent is required if performing PLAYR-CyTOF or
surface antibody staining.
5. Saline Sodium Cirtrate (SSC, 20×).

2.3 LCB for PLAYR- 1. Metal-conjugated antibodies for LCB: Dilute antibodies in
CyTOF (Optional) CSB to make 100 μL barcoding solution per 3 × 106 cells
stained (see Note 3). Antibody solutions can be made 2 h
before use and stored at 4 °C. See Table 1 for an example of
preparing an LCB scheme for PBMC using cadmium-
conjugated anti-CD45 antibodies.

2.4 Staining to 1. Dead cell exclusion reagent: For mass cytometry we use Cell-
Identify Dead Cells ID cisplatin (Fluidigm, cat. no. 201064). For flow cytometry
(Optional) we use Live/Dead Fixable Dead Cell Stain Kits diluted 1:1000
in RT PBS (Invitrogen). Reagents should be diluted 1:1000
immediately before use in RT PBS.

2.5 Surface Antibody 1. Metal- (for mass cytometry [19]; Fluidigm) or fluorophore-
Staining (Optional) conjugated (for flow cytometry) antibodies for surface antigen
phenotyping (see Note 2): Dilute antibodies in CSB to make
100 μL of antigen phenotyping solution per 3 × 106 cells
stained (see Note 3). Antibody cocktails can be made up to
2 h before use and stored at 4 °C.
2. 5 mM Bis(sulfosuccinimidyl) suberate (BS3 crosslinker): Dis-
solve BS3 crosslinker in ice-cold PBS (see Note 4).

2.6 Cell Fixation and 1. Ice-cold 1.6% Paraformaldehyde (PFA): Dilute 16% PFA using
Permeabilization ice-cold PBS and place on ice. 1.6% PFA can be made up to
1 day before use.
2. Ice-cold methanol.

2.7 PLAYR and 1. Thermal mixer (e.g., Eppendorf ThermoMixer C, cat.

Internal Antibody no. 5382000031).
Staining 2. 100 μM PLAYR oligonucleotides: PLAYR probes should be
dissolved in ddH2O and aliquoted (see Note 5). Purification by
154 Andrew D. Duckworth et al.

standard desalting is sufficient for PLAYR probes, while 5′-

-phosporylated oligonucleotides (i.e., backbone and inserts)
should be purified by HPLC. See Table 2 for list of sequences
(see also Table 3). Oligonucleotides should be stored at -20 °C
for up to 3 years.
3. PLAYR dilution buffer (PBST): 0.1% Tween diluted in PBS.
Make fresh at the start of the PLAYR assay (Subheading 3.3,
Step 1) and place on ice.
4. PLAYR wash buffer (PBSTR): 4 U/mL RNasin Plus (Pro-
mega, cat. no. N2611/N2615) (see Note 6) diluted in PBST.
Make fresh at the start of the PLAYR assay (Subheading 3.3,
Step 1) and place on ice.
5. PLAYR wash buffer with extra RNasin (PBSTR+): 40 U/mL
RNasin Plus (Promega, cat. no. N2611/N2615) (see Note 6)
diluted in ice-cold PBST. Make fresh at the start of the PLAYR
assay (Subheading 3.3, Step 1) and place on ice.
6. Preliminary PLAYR probe hybridization buffer (see Note 7):
1× SSC, 1% Tween20 (see Note 8), 2.5% Polyvinylsulfonic acid,
and 100 μg/mL Salmon sperm DNA. Reagents are diluted in
the appropriate amount of ddH2O (see Note 7). Make fresh at
the start of the PLAYR assay (Subheading 3.3, Step 3) and
leave at RT.
7. Final PLAYR probe hybridization buffer (see Note 7): Prelimi-
nary PLAYR probe hybridization buffer plus 100 nM of each
PLAYR probe which has been preheated (see Note 5, preheated
in Subheading 3.3, Step 5), 20 mM Ribonucleoside vanadyl
complexes (see Note 9, preheated in Subheading 3.3, Step 2),
and 40 U/mL RNasin Plus (Promega, cat. no. N2611/
N2615) (see Note 6). Make this immediately before adding
to the cells in Subheading 3.3, Step 6.
8. PLAYR post-hybridization wash buffer: 4× SSC (see Note 10)
and 40 U/mL RNasin diluted in ice-cold PBSTR. Make fresh
up to 20 min before use in Subheading 3.3, Step 9.
9. PLAYR backbone/insert hybridization mix: 1× SSC, 40 U/
mL RNasin, 100 nM backbone oligonucleotide, and 100 nM
of each insert oligonucleotide. Dilute in ice-cold PBSTR. Make
fresh up to 20 min before use in Subheading 3.3, Step 12.
10. PLAYR T4 buffer wash: 1× T4 ligase buffer (as supplied by the
manufacturer) and 40 U/mL RNasin. Make fresh up to 20 min
before use in Subheading 3.3, Step 16.
11. PLAYR T4 ligation mix: 1× T4 ligase buffer (as supplied by the
manufacturer), 5 U/mL T4 ligase and 40 U/mL RNasin.
Dilute in ddH2O. Make fresh up to 20 min before use in
Subheading 3.3, Step 16.
 Highly Multiplexed and Simultaneous Characterization of Protein and RNA in. . . 155

12. PLAYR Phi29 buffer wash: 1× Phi29 buffer (as supplied by the
manufacturer) diluted in ddH2O. Make fresh up to 20 min
before use in Subheading 3.3, Step 20.
13. PLAYR Phi29 amplification mix: 1× Phi29 buffer (as supplied
by the manufacturer), 5 U/mL Phi29 DNA polymerase, and
800 μM dNTP. Dilute in RNase-free ddH2O. Make fresh up to
20 min before use in Subheading 3.3, Step 20.
14. PLAYR mass cytometry detection probe buffer: 10 nM of each
metal-labeled DDP diluted in cell staining buffer. Use this
buffer if performing analysis by mass cytometry. If internal
antigens are also being analyzed, an appropriate concentration
of metal-labeled antibodies should be added to this buffer.
Make fresh up to 2 h before use in Subheading 3.3, Step 24.
15. PLAYR flow cytometry detection probe buffer: 1 nM of each
fluorescent-labeled DDP diluted in CSB. Use this buffer if
performing analysis by flow cytometry. If internal antigens are
also being analyzed, an appropriate concentration of
fluorescent-labeled antibodies should be added to this buffer.
Make fresh up to 2 h before use in Subheading 3.3, Step 24.

2.8 Additional 1. 62.5 nM Cell-ID Intercalator-Ir buffer: Dilute Cell-ID Inter-

Reagents Specific to calator-Ir (Fluidigm, cat. no. 201192A) in PBS containing
Mass Cytometry 1.6% PFA.
2. EQ Four Element Calibration Beads (Fluidigm, cat.
no. 201078).
3. Cell acquisition solution (CAS, Fluidigm, cat no. 201240).
4. 5 mL Round-Bottom Polystyrene Test Tubes with 35 μm Cell
Strainer Cap.

2.9 Software 1. Fluidigm CyTOF system control software for mass cytometry
(www.fluidigm.com/software).
2. NxT software (Thermofisher; cat. no. A25554) for the Attune
flow cytometer (or equivalent).
3. Software appropriate for analysis of multiparametric mass cyto-
metry data files (e.g., Cytobank (https://www.cytobank.org/),
Cytosplore (https://www.cytosplore.org/), etc.).
4. R software (https://www.r-project.org/) for PLAYR probe
design.
5. PLAYRDesign software to design PLAYR probes for your
RNA of interest (detailed instructions for installation can be
found at https://github.com/nolanlab/PLAYRDesign).
156 Andrew D. Duckworth et al.

3 Methods

3.1 Labeling DDPs 1. Calculate or make a note of the starting number of nmol for
with Metal Isotopes for each detection probe as this will be required during calculations
Mass Cytometry in step 18.
2. Add 500 mM TCEP to the 250 μM DDP to a final concentra-
tion of 50 mM (1 in 10) and incubate at RT for 30 min.
3. If required, raise the volume to a minimum of 200 μL using
ddH2O and then add sodium acetate to a final concentration of
300 mM. Mix by vortexing.
4. Add three volumes of ice-cold 100% ethanol and mix by vortex.
5. Place the tube at -20 °C overnight to allow precipitation.
Alternatively, precipitation and longer-term storage can be
performed at -80 °C for up to 6 months.
6. From the Maxpar X8 Antibody Labeling Kit, centrifuge the X8
polymer tube 4000rcf/30 s/RT) to ensure the reagent is at the
bottom.
7. Dissolve the polymer in 95 μL of L-Buffer by pipetting.
8. Add 5 μL of lanthanide metal solution and mix by pipetting.
9. Incubate the polymer-metal mix at 37 °C for 30 min.
10. While the polymer-metal mix is incubating (step 9), centrifuge
the precipitated DDP from step 5 at 14,000rcf/4 °C/30 min.
11. Add 200 μL of L-Buffer to a 3 kDa MWCO filter.
12. After the incubation in step 9, add the metal-loaded polymer
to the filter containing the 200 μL L-Buffer.
13. Centrifuge metal-conjugated polymer at 12,000rcf/RT/
30 min (see Note 11).
14. Once the centrifuge step is completed from step 10, carefully
remove supernatant from DDP pellet. Resuspend in 400 μL
ice-cold 75% ethanol and centrifuge at 14,000rcf/4 °C/
10 min.
15. Retrieve 3 kDa filter from centrifuge (from step 13) and
discard flow through. Add 300 μL of C-Buffer to the filter.
Centrifuge at 12,000rcf/RT/30 min; leave at RT while com-
pleting detection probe purification.
16. Repeat step 14 to wash the DDP pellet a second time.
17. After the final wash of the DDP in step 16, remove as much
ethanol as possible (a brief centrifuge step (14,000rcf/4 °C/
30 s) may help to pool residual ethanol for removal). Air-dry
the pellet (the tube can be placed at 40 °C on a heat block to
decrease drying time, see Note 12).
 Highly Multiplexed and Simultaneous Characterization of Protein and RNA in. . . 157

18. Resuspend the dried pellet of reduced DDP in C buffer (from

the Maxpar X8 Antibody Labelling Kit) to a concentration of
1 nmol/25 μL, assuming a yield of 80% after reduction. For
example, if there was 10 nmol of DDP to start, then you would
assume a yield of 8 nmol after the reduction/purification steps
and therefore resuspend the pellet in 200 μL of C buffer.
19. Measure the reduced DDP concentration from step 18 by
absorption at 260 nM on a NanoDrop spectrophotometer
using C buffer as a blank.
20. Using the concentration from step 19, calculate the volume
required for 2 nmol of DDP (MW = 6789 g/mol) and transfer
this volume to a fresh Eppendorf tube.
21. Transfer the entire metal-loaded polymer from the 3 kDa filter
to the Eppendorf tube containing the 2 nmol DDP.
22. Make the volume up to 100 μL using C-Buffer and mix by
pipetting.
23. Incubate at RT for 2 h.
24. Add 500 mM TCEP 1 in 100 to a final concentration of 5 mM,
immediately mix by vortexing, and incubate at RT for 30 min.
25. Add 400 μL of ddH2O to a 30 kDa MWCO filter.
26. Transfer the metal-polymer-DDP mixture to the 30 kDa
MWCO filter containing the ddH2O and centrifuge at
14,000rcf/RT/12 min.
27. Discard the flow-through, add 500 μL of ddH2O to the filter,
and centrifuge at 14,000rcf/RT/12 min.
28. Repeat step 27.
29. Invert the filter into a new collection tube and centrifuge the
filter and tube at 1000rcf/RT/2 min.
30. Wash the filter with 250 μL of ddH2O and elute into the same
collection tube by inverting the filter and centrifuging the filter
and tube at 1000rcf/RT/2 min.
31. Measure the oligonucleotide concentration by NanoDrop
spectrophotometer using ddH2O as a blank.
32. Adjust the volume of ddH2O to obtain a final concentration of
1 μM metal-labeled DDP (assuming 6789 ng/mL = 1 μM).
33. Aliquot and store the metal-labeled DDP at -20 °C for up to
3 years.

3.2 Surface Antibody LCB for PLAYR-CyTOF, staining to identify dead cells, and sur-
Staining, Dead Cell face antibody staining are all independent staining procedures and,
Exclusion, and if not required, each section can be excluded without affecting one
Preparing Cells for another or the detection of RNA by PLAYR. To minimize changes
PLAYR in transcript expression during surface antibody staining, cells
158 Andrew D. Duckworth et al.

should be kept cold (between 1 °C and 4 °C) for all steps unless
stated otherwise. All incubation steps for flow cytometry during
and post binding of fluorophore-conjugated reagents should be
performed in the dark to reduce quenching.
1. Aliquot 3 × 106 cells into an Eppendorf tube on ice for each
sample (see Notes 13 and 14).
2. Pellet cells by centrifugation at 500rcf/4 °C/3 min. (See Notes
15 and 16.)
3. Wash in ice-cold PBS by centrifugation at 500rcf/4 °C/3 min.
LCB for PLAYR-CyTOF (Optional)
4. Resuspend each sample in its appropriate LCB antibody solu-
tion (3 × 106 cells per 100 μL) and incubate on ice for 30 min.
Resuspend cells by agitation after 15 min.
5. Pellet cells by centrifugation at 500rcf/4 °C/3 min.
6. Wash pellet twice in 200 μL ice-cold CSB.
7. Resuspend in 200 μL ice-cold PBS and pool barcoded samples
together into one tube.
8. Centrifuge at 500rcf/4 °C/5 min and remove supernatant.
Staining to Identify Dead Cells (Optional)
9. Resuspend cells in 1 mL dead cell exclusion reagent per 107
cells.
10. Incubate at RT for 5 min.
11. Add 4× volume of ice-cold CSB to quench unreacted dead cell
exclusion dye and place tubes on ice.
12. Pellet cells by centrifugation at 500rcf/4 °C/5 min and
remove supernatant.
Surface Antibody Staining (Optional)
13. Resuspend cells in surface antibody cocktail and incubate on
ice for 30 min. Resuspend cells by agitation after 15 min.
14. Wash cells twice in ice-cold CSB.
15. Wash cells in ice-cold PBS.
16. Resuspend cells in 5 mM BS3 crosslinker (1 mL per 107 cells)
and incubate on ice for 30 min (see Note 4). Resuspend cells
by agitation after 15 min.
17. Pellet cells by centrifugation at 500rcf/4 °C/3 min and
remove supernatant.
Cell Fixation and Permeabilization
18. Resuspend cells in ice-cold 1.6% PFA (1 mL per 3 × 106 cells)
and incubate on ice for 10 min.
 Highly Multiplexed and Simultaneous Characterization of Protein and RNA in. . . 159

19. Centrifuge cells at 800rcf/4 °C/3 min and remove

supernatant.
20. Flick pellet to resuspend cells in residual supernatant.
21. Add ice-cold methanol dropwise while vortexing the sample
(see Note 17).
22. Incubate on ice for 10 min. Alternatively, cells suspended in
methanol can be placed at -80 °C and the protocol can be
paused (see Note 18).
3.3 PLAYR and All centrifuge steps are performed at 800rcf/4 °C/3 min unless
Internal Antibody stated otherwise.
Staining
1. Prepare PBST, PBSTR, and PBSTR+ buffers and place on ice.
2. Heat RVC at 65 °C for 15 min with vigorous agitation to
dissolve any precipitation and then place at RT (do not place
on ice as it will cause the RVC to reprecipitate).
3. While RVC is heating, prepare the preliminary PLAYR probe
hybridization buffer.
4. Once the RVC has finished heating and been placed at RT,
pellet cells in methanol (see Note 19) and wash with 200 μL of
ice-cold PBSTR+ (see Notes 20 and 21).
5. While washing cells in the previous step, heat PLAYR probe
mixture at 90 °C for 5 min and then place immediately on ice
(see Note 22).
6. Quickly add remaining reagents (preheated RVC, preheated
PLAYR probes, and RNasin Plus) to make the final PLAYR
hybridization buffer and immediately use to resuspend cells at a
maximum of 3 × 107 cells/mL (minimum of 100 μL).
7. Incubate cells for 1 h at 40 °C (see Note 23) with vigorous
agitation (we use 1400 rpm).
8. Wash cells three times with 200 μL of ice-cold PBSTR.
9. While washing the cells in step 8 prepare PLAYR post-
hybridization wash buffer.
10. Resuspend cells in PLAYR post-hybridization wash buffer at a
maximum of 3 × 107 cells/mL (minimum of 100 μL) and
incubate at 40 °C (see Note 23) for 20 min under vigorous
agitation (we use 1400 rpm).
11. During this incubation prepare the PLAYR backbone/insert
hybridization mix.
12. Wash cells two times in ice-cold PBSTR.
13. Resuspend the cells in PLAYR backbone/insert hybridization
mix at 3 × 107 cells/mL (minimum of 100 μL).
160 Andrew D. Duckworth et al.

14. Incubate cells for 30 min at 37 °C with agitation (we use

1100 rpm).
15. While incubating prepare the 1× PLAYR T4 wash buffer and
PLAYR T4 ligation mix.
16. Wash cells two times in 200 μL ice-cold PBSTR.
17. After two washes in PBSTR, wash cells in 100 uL of 1× PLAYR
T4 wash buffer.
18. Resuspend the cells in PLAYR ligation mix at 6 × 107 cells/mL
(minimum of 50 μL).
19. Incubate cells for 30 min at RT (21 °C) with agitation (we use
1100 rpm).
20. While incubating prepare the Phi29 wash buffer and PLAYR
Phi29 amplification mix.
21. Wash the cells in Phi29 wash buffer.
22. Resuspend the cells in PLAYR Phi29 amplification mix at
6 × 107 cells/mL (minimum of 50 μL).
23. Incubate cells for 2 h (flow cytometry) or 4 h (mass cytometry)
at 30 °C with agitation (we use 1100 rpm). See Note 24.
24. Prepare either the PLAYR mass- or flow-cytometry detection
probe buffer.
25. Wash cells once in 200 μL PBSTR (flow cytometry) or CSB
(mass cytometry).
26. Resuspend cells in the appropriate PLAYR detection probe
buffer at 3 × 107 cells/mL (minimum of 100 μL).
27. Incubate cells for 30 min at 37 °C with agitation (we use
1100 rpm).
28. Wash cells twice in PBSTR (for flow cytometry) or CSB (for
mass cytometry).
29. For flow cytometry, resuspend cells in ice-cold PBS and analyze
immediately using laboratory-specific protocols. For mass
cytometry, resuspend cells in Cell-ID Intercalator-Ir buffer
and incubate for 1 h at room temperature or for up to
2 weeks at 4 °C (see Note 25).

3.4 Sample 1. Wash cells once in 1 mL CSB (800rcf/4 °C/3 min).

Preparation for Mass 2. Wash cells once in 1 mL CAS (800rcf/4 °C/3 min).
Cytometry
3. Resuspend in 1 mL CAS.
4. Filter through a 35 μM filter cap into a 5 mL test tube.
5. Count cells and adjust concentration to 5 × 105 cells/mL using
CAS (see Note 26).
6. Add 1/10 volume of EQ Four Element Calibration Beads and
mix by vortexing (see Note 27).
 Highly Multiplexed and Simultaneous Characterization of Protein and RNA in. . . 161

7. Collect data using the mass cytometer. Details on instrument

and panel-template setup are outside the scope of this protocol
but can be found in the user guides on the Fluidigm website.
8. Analyze data using software such as cytobank, flowjo, or rele-
vant R packages.

4 Notes

1. Due to the large number of centrifuge steps in the protocol,

every effort must be made to minimize cell loss. Swing rotor
centrifuges are highly recommended alongside the use of small
wash volumes (200 μL) and Polypropylene V-bottom 200 μL
PCR strips, or alternatively 1.5 mL Eppendorf tubes. Always
remove supernatant by aspiration (i.e., not by flicking the
liquid out) and use small pipette tips (200 μL or lower) to
increase accuracy. If the supernatant volume is high (i.e.,
>2 mL), it may be advised to remove down to ~300 μL,
resuspend the pellet again, and centrifuge a second time.
2. Metal- and fluorophore-conjugated antibodies used to stain
surface antigen for measurement by flow or mass cytometry
need to withstand conditions used during the remaining steps
of the protocol (i.e., PFA fixation, methanol treatment, and
PLAYR incubations). For flow cytometry, this generally means
the use of smaller fluorophores (e.g., FITC, Alexa Fluors)
which are not protein based and that are much more likely to
remain functional during the procedure. For PLAYR-CyTOF,
Bismuth-209 conjugated with X8 polymers is the only metal/
polymer conjugate found so far that cannot withstand the
PLAYR conditions. Metals and fluorophores employed at the
end of the protocol (i.e., conjugated to internal antigens or
DDPs) are not subjected to harsh conditions, and therefore
their choice is not limited at this later step.
3. Antibodies should be serially diluted to confirm specific and
efficient staining of target cells after PLAYR. Generally, final
concentrations of staining antibodies fall in the range of
10–1.25 μg/mL (most often 0.5 μg/mL is used). For LCB,
the signal produced by positively stained cells should be clearly
distinct from the signal generated by cells that have not been
stained. If the LCB antibodies target the same epitope, their
dilution should be tested in the barcoding combinations
(as using multiple antibodies that compete for binding to the
same epitope will reduce signal for each barcoding metal).
4. BS3 Crosslinker hydrolyzes in water. Allow the vial containing
BS3 to reach RT before opening to avoid condensation. BS3
162 Andrew D. Duckworth et al.

crosslinker powder should be weighed and dissolved in ice-cold

PBS immediately before addition to the cells.
5. To increase efficiency of PLAYR probe hybridization buffer
preparation we recommend that the 100 μM PLAYR probes
that target the same transcript are mixed in equal volumes, and
aliquots are made, so that the volume of the aliquots (in μL) is
equal to the number of probes being used for that transcript
(e.g., if four probe pairs are being used to target a transcript
then 8 μL aliquots would be made as there are eight PLAYR
probes). Each aliquot can then be frozen and will be sufficient
for 1 mL of PLAYR probe hybridization buffer. If the same set
panel of transcripts will be analyzed repeatedly, then to speed
the process up even further, each transcript-specific aliquot
made above can be pooled in advance of the PLAYR assay
and stored frozen (e.g., 10 transcripts are being analyzed,
each being targeted by 8 PLAYR probes, would produce an
80 μL aliquot of PLAYR probe mixture that can be heated and
added to the final PLAYR probe hybridization buffer).
6. We have tested other RNasin brands and found that RNasin
Plus (Promega) is the best at maintaining RNA integrity during
the PLAYR assay. Other RNasins may give suboptimal results.
7. To standardize the process (in terms of PLAYR probe and RVC
aliquot sizes) we recommend making up the PLAYR probe
hybridization buffer in 1 mL volumes (where 1 mL is enough
for 3 × 107 cells). The PLAYR probe hybridization buffer is
made up in two stages (termed the preliminary and final). The
preliminary buffer is made shortly before use (in Subheading
3.3, Step 3) but lacks the preheated PLAYR probes, preheated
RVC and RNasin plus that will need to be added immediately
before addition to the cells (Subheading 3.3., Step 6). The
volume of ddH2O required in the preliminary buffer is depen-
dent on the volume of probes added to the final buffer to give a
total volume of 1 mL.
8. 1% Tween20 can be replaced with lower concentrations of the
detergent (we have reduced this to 0.2% Tween20 without
noticing any difference in signal for our PLAYR panel). Speci-
ficity for your specific PLAYR probes should be confirmed
while using these milder conditions.
9. RVC precipitates at low temperatures and requires heating at
65 °C for 15 min to dissolve and fully activate the reagent. The
RVC will arrive frozen from the supplier and should be
defrosted and heated in a 65 °C water bath with frequent
agitation to allow it to dissolve. Once dissolved, aliquots can
be generated at RT and then frozen at -80 °C. For a 200 mM
Highly Multiplexed and Simultaneous Characterization of Protein and RNA in. . . 163

RVC solution, a 100 μL aliquot will be sufficient to make 1 mL

of PLAYR probe hybridization buffer which is enough for
1 × 107 cells or up to 10 samples (see Note 7 on making the
PLAYR probe hybridization buffer).
10. 4× SCC can be replaced with 2× SCC concentration to help
maintain surface antibody binding. Specificity for your specific
PLAYR probes should be confirmed while using these milder
conditions.
11. You should aim for a volume after centrifugation that is lower
than 20 μL. If it is higher, then centrifuge again for 5–10 min.
12. Failure to fully dry the pellet and remove all ethanol can affect
downstream applications. The pellet is dry when it turns from
opaque/white to bright white in color.
13. We encourage a minimum of 3 × 106 cells per sample because
of cell loss throughout the assay. This yields approximately
(depending on pipetting efficiency) 1 × 106 cells at the final
wash step to analyze by mass/flow cytometry. Lower cell num-
bers per sample can be used for mass cytometry if the experi-
mental plan includes LCB and pooling.
14. Flow cytometry experiments will require a negative control
sample to account for autofluorescence and background stain-
ing of the DDP. This could be a sample that is assayed without
probe 1 or 2 (i.e., only one of the probe pair is present), or
alternatively the backbone sequence is excluded from the back-
bone/insert hybridization mix.
15. Due to the considerable number of centrifuge steps during
antibody and RNA staining we highly recommend that cells
are placed in small tubes and pelleted in swing rotor centrifuges
using small volumes (200 μL or less) where possible to avoid
excessive cell loss. Small volumes (<20 μL) of supernatant may
be left during repeat wash steps to decrease cell loss.
16. Centrifugation details given are for primary PBMC cells and
can be modified for your cell of interest if required.
17. To avoid cell clumping, cells should be in a uniform suspension
while methanol is added.
18. Fixed cells can remain in methanol at -80 °C for several
months without noticeable depreciation in subsequent anti-
body or RNA signal.
19. Methanol has increased density when at -80 °C, which may
inhibit the pelleting of cells. Allow the methanol to warm up to
ice temperature for 5 min before centrifugation.
164 Andrew D. Duckworth et al.

20. Once cells are removed from methanol RNA degradation can
set in. Speed in processing during the PBSTR+ wash step and
fast transfer into the PLAYR probe hybridization buffer is
essential to maintain RNA integrity.
21. Due to the small volumes used, ensure that all wash buffer is
removed from the pellet before the addition of PLAYR hybri-
dization buffer to the cells in Subheading 3.3 Step 6.
22. Cool on ice to avoid denaturing RNasin in the PLAYR hybri-
dization buffer.
23. Incubation temperatures of PLAYR hybridization buffer and
post hybridization wash can be reduced to 37 °C to help
maintain surface antibody binding. Specificity for your specific
PLAYR probes should be confirmed while using these milder
conditions.
24. The RCA times using Phi29 can be made longer to increase the
signal from the assay. However, we have found that longer
RCA times (past the 2 h and 4 h suggested) do not increase
sensitivity of the assay (i.e., negative cells remain negative). As
mentioned in the introduction, the best way to increase sensi-
tivity is by increasing the number of probe pairs used to target
the RNA.
25. PLAYR-CyTOF signals will slowly depreciate the longer you
leave the samples at 4 °C. We recommend running the samples
within 48 h of finishing the assay to achieve maximum signal.
26. Cell concentrations may be adjusted depending upon how
readily they clog the mass cytometer. Larger or more sticky
cells may require running at lower concentrations.
27. Vigorously shake the EQ Four Element Calibration Bead bot-
tle for 30 s to make sure the contents are evenly resuspended
before adding to the sample.

References
1. Bauman JG, Bayer JA, van Dekken H (1990) 3. Lalli E, Gibellini D, Santi S, Facchini A (1992)
Fluorescent in-situ hybridization to detect cel- In situ hybridization in suspension and flow
lular RNA by flow cytometry and confocal cytometry as a tool for the study of gene
microscopy. J Microsc 157(Pt 1):73–81. expression. Anal Biochem 207(2):298–303.
https://doi.org/10.1111/j.1365-2818.1990. https://doi.org/10.1016/0003-2697(92)
tb02948.x 90015-y
2. Porichis F, Hart MG, Griesbeck M, Everett 4. Patterson BK, Till M, Otto P, Goolsby C, Fur-
HL, Hassan M, Baxter AE, Lindqvist M, Miller tado MR, McBride LJ, Wolinsky SM (1993)
SM, Soghoian DZ, Kavanagh DG, Reynolds S, Detection of HIV-1 DNA and messenger
Norris B, Mordecai SK, Nguyen Q, Lai C, RNA in individual cells by PCR-driven in situ
Kaufmann DE (2014) High-throughput hybridization and flow cytometry. Science
detection of miRNAs and gene-specific 260(5110):976–979. https://doi.org/10.
mRNA at the single-cell level by flow cytome- 1126/science.8493534
try. Nat Commun 5:5641. https://doi.org/ 5. Belloc F, Durrieu F (1994) Detection of
10.1038/ncomms6641 mRNA species by flow cytometry. Methods
 Highly Multiplexed and Simultaneous Characterization of Protein and RNA in. . . 165

Cell Biol 42(Pt B):59–69. https://doi.org/10. 995–1000. https://doi.org/10.1038/

1016/s0091-679x(08)61068-7 nmeth947
6. Borzi RM, Piacentini A, Monaco MC, 13. Lizardi PM, Huang X, Zhu Z, Bray-Ward P,
Lisignoli G, Degrassi A, Cattini L, Santi S, Thomas DC, Ward DC (1998) Mutation
Facchini A (1996) A fluorescent in situ hybri- detection and single-molecule counting using
dization method in flow cytometry to detect isothermal rolling-circle amplification. Nat
HIV-1 specific RNA. J Immunol Methods Genet 19(3):225–232. https://doi.org/10.
193(2):167–176. https://doi.org/10.1016/ 1038/898
0022-1759(96)00070-1 14. Brodin P (2019) The biology of the cell –
7. Larsson C, Grundberg I, Soderberg O, Nilsson insights from mass cytometry. FEBS J 286(8):
M (2010) In situ detection and genotyping of 1514–1522. https://doi.org/10.1111/febs.
individual mRNA molecules. Nat Methods 14693
7(5):395–397. https://doi.org/10.1038/ 15. Bendall SC, Simonds EF, Qiu P, Amir el AD,
nmeth.1448 Krutzik PO, Finck R, Bruggner RV,
8. Player AN, Shen LP, Kenny D, Antao VP, Kol- Melamed R, Trejo A, Ornatsky OI, Balderas
berg JA (2001) Single-copy gene detection RS, Plevritis SK, Sachs K, Pe’er D, Tanner
using branched DNA (bDNA) in situ hybridi- SD, Nolan GP (2011) Single-cell mass cytome-
zation. J Histochem Cytochem 49(5): try of differential immune and drug responses
6 0 3 – 6 1 2 . h t t p s : // d o i . o r g / 1 0 . 1 1 7 7 / across a human hematopoietic continuum. Sci-
002215540104900507 ence 332(6030):687–696. https://doi.org/
9. Frei AP, Bava FA, Zunder ER, Hsieh EW, Chen 10.1126/science.1198704
SY, Nolan GP, Gherardini PF (2016) Highly 16. Muftuoglu M, Li L, Liang S, Mak D, Lin AJ,
multiplexed simultaneous detection of RNAs Fang J, Burks JK, Chen K, Andreeff M (2021)
and proteins in single cells. Nat Methods Extended live-cell barcoding approach for mul-
13(3):269–275. https://doi.org/10.1038/ tiplexed mass cytometry. Sci Rep 11(1):12388.
nmeth.3742 https://doi.org/10.1038/s41598-021-
10. Duckworth AD, Gherardini PF, Sykorova M, 91816-w
Yasin F, Nolan GP, Slupsky JR, Kalakonda N 17. Hartmann FJ, Simonds EF, Bendall SC (2018)
(2019) Multiplexed profiling of RNA and pro- A universal live cell barcoding-platform for
tein expression signatures in individual cells multiplexed human single cell analysis. Sci
using flow or mass cytometry. Nat Protoc Rep 8(1):10770. https://doi.org/10.1038/
14(3):901–920. https://doi.org/10.1038/ s41598-018-28791-2
s41596-018-0120-8 18. Mei HE, Leipold MD, Schulz AR, Chester C,
11. Fredriksson S, Gullberg M, Jarvius J, Olsson C, Maecker HT (2015) Barcoding of live human
Pietras K, Gustafsdottir SM, Ostman A, Land- peripheral blood mononuclear cells for multi-
egren U (2002) Protein detection using plexed mass cytometry. J Immunol 194(4):
proximity-dependent DNA ligation assays. 2022–2031. https://doi.org/10.4049/
Nat Biotechnol 20(5):473–477. https://doi. jimmunol.1402661
org/10.1038/nbt0502-473 19. Han G, Spitzer MH, Bendall SC, Fantl WJ,
12. Soderberg O, Gullberg M, Jarvius M, Nolan GP (2018) Metal-isotope-tagged
Ridderstrale K, Leuchowius KJ, Jarvius J, monoclonal antibodies for high-dimensional
Wester K, Hydbring P, Bahram F, Larsson mass cytometry. Nat Protoc 13(10):
LG, Landegren U (2006) Direct observation 2121–2148. https://doi.org/10.1038/
of individual endogenous protein complexes in s41596-018-0016-7
situ by proximity ligation. Nat Methods 3(12):
 Chapter 11

Array-Based Comparative Genomic Hybridization

for the Detection of Copy Number Alterations in Single Cells
Giancarlo Feliciello, Zbigniew Tadeusz Czyz, and Bernhard M. Polzer

Abstract
Comprehensive genome-wide analyses of single cells represent an important tool for clinical applications,
such as pre-implantation diagnostic and prenatal diagnosis, as well as for cancer research purpose. For the
latter, studies of tumor heterogeneity, circulating tumor cells (CTCs), and disseminated cancer cells
(DCCs) require the analysis of single-cell genomes. Here we describe a reliable and robust array-based
comparative genomic hybridization (aCGH) protocol based on Ampli 1™ whole genome amplification
that allows the detection of copy number alterations (CNAs) in single cancer cells as small as 100 kb.

Key words aCGH, Single cells, PCR labeling, Array hybridization, CNAs

1 Introduction

Array comparative genomic hybridization (aCGH) is a technique

that enables genome-wide high-resolution screening of genomic
copy number alterations (CNAs) [1, 2]. It allows for comprehen-
sive analysis at multiple genomic loci by detecting DNA copy
number gains and losses. Nowadays, aCGH is an essential tool in
pre-implantation diagnostics [3, 4] and prenatal diagnosis
[5]. These diagnostic applications require the analysis of copy
number changes at the single-cell level. Similarly, in cancer research,
single-cell technologies are more and more demanded for the
characterization of circulating tumor cells (CTCs) isolated from
blood and disseminated cancer cells (DCCs) found in a secondary
distant site (e.g., bone marrow or lymph node). Multiple studies
have shown that the presence of these cells is an independent
prognostic factor of poor outcome for all tested tumor types [6–
8]. For this reason, the genomic characterization of these cells
becomes more and more a clinical need. CNAs in single cells have
been previously analyzed by Klein et al. [9] by gene metaphase-
based comparative genomic hybridization (mCGH) on amplified
single-cell genomes. Although the application of this method has

Miodrag Gužvić (ed.), Single Cell Analysis: Methods and Protocols, Methods in Molecular Biology, vol. 2752,
https://doi.org/10.1007/978-1-0716-3621-3_11, © Springer Science+Business Media, LLC, part of Springer Nature 2024

167
168 Giancarlo Feliciello et al.

proved to be reliable and robust, it had several limitations. Among

these, the most important are the low resolution of only 5–10 Mb,
as well as the laborious and time-consuming procedure. In this
context, aCGH offers a higher sensitivity and specificity
[10, 11]. Using arrays composed of highly purified BAC clones,
resolutions of 8.3 Mb for single copy deletion could be achieved
[12], as well as the detection of aberrant regions as small as 1–2 Mb
in cell lines and 4.8 Mb in DCCs [13]. Other studies indicated that
using high-density oligonucleotide microarrays the detection limit
of single-cell aCGH can be reduced to 1 Mb or less in freshly
isolated cells [14, 15]. Here we describe a protocol that enables
the detection of aberration down to 100 kb applicable to freshly
picked cells, DCCs and CTCs isolated, respectively, from secondary
distant site (e.g., bone marrow or lymph nodes) and blood of
clinical patients [16]. Additionally, cells derived from other body
fluids (e.g., cerebrospinal fluid) or from different isolation proce-
dures, e.g., CellSearch® cartridge, Parsortix™, DEParray™, are
well suited for the analysis.

2 Materials

2.1 Reagents In this protocol high-quality reagents are necessary to execute the
aCGH procedure. Utilization of the materials included in the kits,
reagents, or instrumentations indicated maximizes the likelihood of
a successful experiment. If using third-party materials, the user is
recommended to prepare all solutions using ultrapure water
(HPLC purified, DNAse-free) and PCR analytical grade reagents.
Store all reagents and enzymes according to the guidelines
specified by the manufacturer. Reagent expired or subjected to
improper storage may compromise the outcome of the aCGH
experiment. Always store the reagents separately from the DNA
samples and avoid repeated freeze-thaw cycles of solutions contain-
ing DNA or enzymes. When preparing frozen reagent solutions
thaw completely, mix briefly on a vortex mixer and spin in a micro-
centrifuge to collect the contents off the wall and lid, and pre-chill
to 4 °C or on ice until use.

2.2 Laboratory Setup To assure optimal condition for the aCGH, it is recommended to
perform all steps under a dedicated vertical laminar flow hood.
Make sure that all the pipetting aids are calibrated and handled
properly. Always wear powder-free laboratory gloves and use dedi-
cated solutions and pipettes with nuclease-free aerosol-resistant tips
to avoid contamination of reagents with nucleases and cross-
contaminations of samples. Maintain a clean working area by wip-
ing surfaces with Ethanol 80% or bleaching solutions (e.g., DNA-
Zap™ or DNA AWAY*) and by UV-irradiation of PCR bench, tips,
tubes, and tube racks.
 Array-Based Comparative Genomic Hybridization for the Detection of Copy. . . 169

2.3 Safety Notes Cyanine reagents are considered hazardous and may be harmful if
swallowed. Avoid contact with eyes, skin, and clothing.
2× HI-RPM Hybridization Buffers are considered hazardous
and may be harmful if swallowed. Avoid contact with eyes, skin, and
clothing.
Acetonitrile is considered toxic to blood, kidneys, liver, lungs,
gastrointestinal tract, central nervous system, skin, eyes, and upper
respiratory tract. Avoid contact with eyes, skin, and clothing.
Always refer to the Material Safety Data Sheet (MSDS) of your
specific country.
Carefully follow all waste disposal regulations when disposing
waste materials.

2.4 DNA Templates aCGH application uses a “two-color” process to measure DNA
copy number alterations in a Test Sample relative to a Reference
Sample.
1. Test Sample: The chapter describes the recommendations to
label and hybridize amplified DNA from tumor cells, in partic-
ular single DCCs and CTCs, but its spectrum of application is
wider and may include other applications where the amount of
input DNA material is very limited. Therefore, the method is
also well suited for the analysis of cell pools (≥2 cells), speci-
mens extracted from FFPE tissue, as well as cells or DNA
material collected with the use of various isolation procedures
(e.g., micromanipulators mounted on microscopes, the
DEParray™ System). A fundamental prerequisite to the use
of this protocol is the amplification with Ampli 1™ WGA Kit
that has been already described in detail in this book series
[17]. Moreover, aCGH of single cells demands for samples
with high-quality amplified DNA. Always assess the Genomic
Integrity Index (GII) of the samples before proceeding to
expensive molecular assays [18]. The use of DNA samples
that do not fulfill these quality criteria may result in the failure
of the labeling reactions or low quality of the aCGH results
(high background noise) that may impair the reliable detection
of copy number alterations.
2. Reference Sample: A pool of peripheral blood lymphocytes
isolated from a healthy male/female donor and amplified with
Ampli 1™ WGA Kit. Sex-matched or sex-mismatched controls
can be used in different settings. If the gender of the sample
donor is known, selection of reference DNA is based on the
gender of the donor of the test sample. Sex-matched designs
are more commonly used when the technology is well estab-
lished and may offer better sensitivity for detecting gains or
losses of regions on the X and Y chromosomes.
170 Giancarlo Feliciello et al.

Sex-mismatched designs are used when array CGH approaches

are first implemented because the dose differences between the
X and Y chromosomes provide useful information on assay
quality.
3. Controls: During procedures generating DNA templates for
hybridization, each reaction run should include a no-template
control (NTC) where the sample is substituted with PCR grade
water.

2.5 Fluorescent 1. Expand Long Template PCR System (Roche) containing: 10×
Labeling of Amplified Expand Long Template Buffer 1, 2, and 3 and Expand Long
DNA Template Enzyme Mix (5 U/μL).
2. Bovine Serum Albumin from bovine serum, 20 mg/mL
(Roche).
3. Fluorescent-labeled nucleotides 25 nmol each
(GE-Healthcare): Cy3-dCTP and Cy3-dUTP for the Reference
Sample, Cy5-dCTP and Cy5-dUTP for the Test Sample.
4. 9/10 mixes Deoxynucleotide triphosphate-dNTPs 100 mM
each (VWR International): 10 μL dATP, 10 μL dGTP, 9 μL
dCTP, 9 μL dTTP, and 62 μL of H2O.
5. Lib1 primer 5′- TAGTGGGATTCCTGCTGTCAGT -3′
(20 μM): 20 μL of Lib1 (100 μM) and 80 μL of H2O.

2.6 Purification of 1. Tru1I (MseI), HC (50 U/μL) restriction enzyme (Thermo-

the PCR/Digestion Fisher Scientific).
Product and 2. Amicon Ultra-0.5 Centrifugal Filter Unit with Ultracel-100
Quantification membrane (Merck).
3. Nanodrop instrument with microarray modus (ThermoFisher
Scientific).

2.7 Hybridization 1. Oligo aCGH/ChIP-on-chip Hybridization Kit (Agilent) con-

taining: 2× Hi-RPM Hybridization Buffer and 10× Oligo
aCGH/ChIP-on-Chip Blocking Agent.
2. 25% (v/v) Tween20 for molecular biology (Sigma): dissolve
250 μL of Tween20 in PCR grade water.
3. 25% (v/v) Igepal CA-630 for molecular biology (Sigma): dis-
solve 250 μL of Igepal CA-630 in 750 μL of PCR grade water.
4. Cot Human DNA from human placenta DNA (1 μg/μL),
enriched for repetitive sequences (Roche).
5. Thermomixer (Eppendorf).
6. Hybridization Gasket Slides Kit – 4 microarrays per slide for-
mat (Agilent) containing gasket slides.
 Array-Based Comparative Genomic Hybridization for the Detection of Copy. . . 171

7. SurePrint G3 Human CGH Microarray Kit, 4x180K (Agilent)

containing glass slides each formatted with four SurePrint G3
180K arrays.
8. Hybridization Chamber Kit – SureHyb enabled, Stainless (Agi-
lent) containing: stainless steel chamber (top and base), thumb
clamp and screw, plastic tweezers, user’s guide.
9. Microarray Hybridization Oven (Agilent).

2.8 Washing 1. Slide-staining jar and slide-staining dishes (alternative

providers).
2. 50 mL pyrex jar (alternative providers).
3. Oligo aCGH/ChIP-on-chip Wash Buffer Kit (Agilent) con-
taining: Oligo aCGH/ChIP-on-Chip Wash Buffer 1 and Oligo
CGH ChIP-on-Chip Wash Buffer 2.
4. Oscillator Horizontal Rotor (alternative providers).
5. Acetonitrile >99.9% (VWR).

2.9 Scanning 1. Ozone Barrier Slide Cover Kit (Agilent).

2. Slide Holders for Agilent DNA Microarray Scanner CA and BA
(Agilent).
3. DNA Microarray Scanner (Agilent).
4. Feature Extraction Software (Agilent).
5. Genomic Workbench Software (Agilent).

3 Methods

The overview of this protocol is given in Fig. 1.

3.1 PCR-Based Purpose In this step Test and Reference samples are differentially
Labeling Using labeled with dUTPs and dCTPs nucleotides conjugated to different
Incorporation of Dye- Cyanine dyes, Cy5 and Cy3 that have a red and green fluorescence,
Conjugated dNTPs and respectively. Cycle after cycle, the polymerase will incorporate the
Tru1I Digestion fluorescent nucleotides into the newly synthetized DNA strand
generating amplicons marked with two different colors. The result-
ing PCR products are then subjected to digestion with Tru1I
restriction endonuclease to cleave off the PCR-adaptors prior to
the aCGH hybridization, thereby avoiding unspecific cross-
hybridization of Test sample with Reference sample and oligonucle-
otide probes on the array.

Duration
1. Hands-on time: 45 min.
2. PCR program: 1 h 25 min.
3. Digestion program: 3 h.
172 Giancarlo Feliciello et al.

DNA Labeling
Hands-on: 30 min Hybridization
PCR: 1h Hands-on: 30 min
Denaturation : 3 min
Reagents
Renaturation: 30 min
DNA
Incubation: 24 h
Expand Long Template PCR
System Reagents
Cy5 dUTP/dCTP Cot1 DNA
Cy3 dUTP/dCTP Oligo aCGH/ChIP-on –chip
Lib1 Primer Hybridization kit
9/10 dNTPs Mix Tween20
BSA Igepal
Nuclease-free Water Output
Output DNA Hybridized on array
Cy5 and Cy3 labeled DNA

Short-term storage 4-8°C

Wash Slides
Lib1 Removal Hands-on: 30 min
Hands-on: 10 min
Reagents
Digestion: 3 h Oligo aCGH/ChIP-on-chip Wash
Buffer Kit
Reagents
Buffer R (10X)
Output
Tru1I Array with unhybridized DNA
removed
Output
DNA without Lib1 primer

Short-term storage 4-8°C

Scan Slides
Hands-on: 10 min
Purification and
Measurement Laser scanner measures the spot
Hands-on: 1 h intensities for Cy5 and Cy3 and
Reagents generates an image necessary for
data extraction
Amicon Ultra 0.5-30KDa columns
Nuclease-free Water Output
Output Array .tif image
Purified Cy5 and Cy3 labeled DNA

Long-term storage -20°C Data Analysis

Hands-on: 10 min

Agilent Genomic Workbench

transforms the generated images
in chromosomal profiles where
amplifications/deletions can be
detected
Output
SAFE STOPPING POINT Chromosomes plot

Fig. 1 Overview of the single cell aCGH workflow described in this chapter
 Array-Based Comparative Genomic Hybridization for the Detection of Copy. . . 173

Table 1
Cy5/Cy3 labeling master mix

Reagents Labeling mix Cy3 volume Labeling mix Cy5 volume

Expand long template buffer 1 (10×) 5.0 μL 5.0 μL
Lib1 (20 μM) 6.0 μL 6.0 μL
9/10 mix dNTPs 1.5 μL 1.5 μL
Cy5 dCTP (25 nmol) – 1.75 μL
Cy5 dUTP (25 nmol) – 1.75 μL
Cy3 dCTP (25 nmol) 1.75 μL –
Cy3 dUTP (25 nmol) 1.75 μL –
BSA (20 mg/mL) 0.5 μL 0.5 μL
Expand long template DNA polymerase 1.5 μL 1.5 μL
(5 U/μL)
DNA template 0.5 μL 0.5 μL
H2O 31.25 μL 31.25 μL
Total 50.0 μL 50.0 μL

Table 2
PCR program for DNA labeling

Temperature Time Cycles

94 °C 15 s
60 °C 30 s 10 (steps 1–3)
65 °C 3:30 min
94 °C 15 s
60 °C 30 s 2 (steps 4–6)
65 °C 3:30 min + 10 s/cycle
65 °C 7 min
4 °C 1

Procedure (see Notes 1 and 2)

1. Prepare the PCR labeling reaction mix (see Notes 3 and 4): For
each Test and Reference sample set up in parallel two 50 μL PCR
reactions according to the recipe outlined in Table 1.
2. Set the PCR program and run the DNA amplification accord-
ing to the program indicated in Table 2.
174 Giancarlo Feliciello et al.

Table 3
DNA digestion mix

Reagent Volume
Buffer R (10×) 5.6 μL
Tru1I 1.0 μL

Table 4
Thermal profile of DNA digestion

Temperature Time
65 °C 3h
4 °C 1

3. For short-term storage, labeled DNA can be kept at 4–8 °C

overnight.
4. Prepare the digestion reaction according to Table 3. Collect the
samples from the cycler, briefly centrifuge, and add 6.5 μL of
the digestion mix into each labeling reaction. Incubate the
DNA digestion using the program outlined in Table 4.
5. For short-term storage, digested DNA can be kept at 4–8 °C
overnight.

3.2 Purification of Purpose In this step the excess labeled nucleotides that have not
the PCR/Digestion been incorporated, as well as PCR and digestion reagents, are
Product and removed by column purification. Additionally, the cleaved
Measurement of the PCR-adaptors generated after digestion are also discarded in the
DNA Yield and purification process. Only when the samples are purified the mea-
Incorporation Rate surement can take place on the Nanodrop instrument.

Duration
1. Hands-on time: 30 min.
2. Purification: 30 min.
3. Quantification: 10 min.

Procedure
1. Assemble the Amicon Ultra 0.5 column as described by the
manufacturer.
2. Pool the resulting products of the same sample and fill up the
volume to 480 μL with H2O. Centrifuge at 14,000 RCF for
10 min at room temperature.
 Array-Based Comparative Genomic Hybridization for the Detection of Copy. . . 175

3. Discard the flow-through and reassemble the purification col-

umn assembly. Add 480 μL H2O and centrifuge again at
14,000 RCF for 10 min at room temperature.
4. Transfer the column in an upside-down (inverted) position to a
new 1.5 mL collection tube and centrifuge at 1000 RCF for
1 min at room temperature to elute the purified product (see
Note 5).
5. Discard the purification column and assess the volume of the
eluted sample with a pipette. If necessary, fill up the volume to
21 μL with H2O (see Note 6).
6. Assess the DNA yield and dye incorporation rates using a
NanoDrop instrument.
7. Start the Nanodrop Software and select the “Microarray”
modus from the menu.
8. Allow the instrument to perform a self-test, select “DNA-50”
from the “Sample Type” menu and blank with H2O.
9. Select Cy5 and Cy3 dyes for the measurement and keep default
absorption/emission settings unchanged.
10. Pipette 2.0 μL of DNA solution directly on the sensor avoiding
formation of air bubbles and start the measurement.
11. Wipe the sensor carefully after each measurement with a soft
KimTech wipe.
12. Once all the samples have been measured, record the .xls report
file in the designated experiment folder.
13. Calculate the
ng
Total DNA yield : Concentration DNA μL * Sample volume ðμLÞ.
Expected yield of the amplified DNA after labeling and clean-
up is between 9 and 16 μg.
340*N ½ ]dye
pmol

14. Calculate the Degree of labeling = CDNA ngμL*1000 * 100%

½μL]
where N is the dye concentration and CDNA is the DNA
concentration.
15. Calculate the Specific activity = Degree0:034
of labeling
. The specific
activity of Cy3 labeled samples is expected to be more than
8 pmol/μg while for Cy5 more than 11 pmol/μg.
16. Labeled DNA can be stored up to 1 month at -20 °C in the
dark.

3.3 Hybridization Purpose During this step the labeled genomic DNA from the Test
and the Reference samples are mixed in equal amounts and
co-hybridized to an array containing immobilized 60-mers of
nucleotides. Cot-1 DNA is used to reduce the cross-hybridization
of repetitive DNA sequences to the array preventing the nonspecific
binding of a labeled probe of interest to the repetitive DNA
sequence (see Note 7).
176 Giancarlo Feliciello et al.

Table 5
Hybridization mix

Reagent Volume
Cot1-DNA (1 μg/μL) 5.0 μL
Agilent blocking reagent (10×) 12.0 μL
HI-RPM hybridization buffer (2×) 60.0 μL
Tween20–25% (v/v) 5.2 μL
Igepal – 25% 5.2 μL
Total 87.4 μL

Duration
1. Hands-on time: 15–20 min (see Note 8).
2. Processing time of the Thermomixer 1: 3 min.
3. Processing time of the Thermomixer 2: 30 min.
4. Processing time of the Agilent Microarray Oven: 24 h.
Procedure
. Step 1: Denaturation /Renaturation of the hybridization
solution.
1. Turn on Thermomixer 1 and 2 and set the following para-
meters: 95 °C for 3 min at 350 rpm and 37 °C for 30 min at
350 rpm, respectively.
2. Turn on the Agilent Microarray Oven and set the temperature
to 65 °C (see Note 9).
3. Thaw the necessary reagents and the desired amount of Test
and Reference samples at room temperature (see Note 10).
4. Assemble the Hybridization Mix as indicated in Table 5 (see
Note 11).
5. In a new labeled 1.5 mL Eppendorf tube, mix Test and Refer-
ence samples (19 μL each) (see Note 12).
6. Add 87.4 μL of the Hybridization Mix into the premixed
samples for a total of 125.4 μL.
7. Mix gently by pipetting slowly up and down (see Note 11) and
quickly spin to drive contents to the bottom of the
reaction tube.
8. Denature the samples in Thermomixer 1: transfer the samples
and incubate at 95 °C for 3 min at 350 rpm (see Note 13).
9. Renature the samples in Thermomixer 2: transfer the samples
swiftly from Thermomixer 1 into Thermomixer 2 and incubate
at 37 °C for 30 min at 350 rpm (see Note 14).
Array-Based Comparative Genomic Hybridization for the Detection of Copy. . . 177

10. Remove the samples from the Thermomixer 2 and spin 1 min
at 6000 RCF in a minicentrifuge.
11. Samples are ready and must be hybridized immediately. Keep
the samples at 37 °C until the Hybridization starts (see Note
15).

. Step 2: Hybridization on the array slide.

1. Place the base of the Agilent SureHyb Chamber on a clean flat

surface.
2. Remove the Gasket Slide from its casing and place it into the
Agilent SureHyb Chamber with the rubber seals of the gasket
wells facing up. Do not touch the surface of the gasket cham-
bers at any time (see Note 16).
3. In a drag-and-drop manner dispense 110 μL of the hybridiza-
tion sample mixture into each gasket well, recording the order
of the samples (see Note 17).
4. Carefully place the Agilent microarray slide on the top of the
Gasket Slide with the “Agilent” labeling bar code (active side)
facing down and the numeric barcode side facing up. Do not
touch the active side of the slide at any time (see Note 18).
5. Put the SureHyb Chamber Cover onto the gathered slides and
slide the clamp assembly on both.
6. Tighten the clamp firmly onto the chamber.
7. Control the assembled chambers for any signs of spillage and
stationary air bubbles (see Note 19).
8. Place the assembled chambers in the rotator rack of the Agilent
Microarray Oven, close and set the rotation speed to 20 rpm
(see Note 20).
9. Hybridize at 65 °C for 22–24 h (see Note 21).
10. For the washing section clean all the necessary glassware equip-
ment before use.
11. Rinse glass slide-staining jar and slide-staining dishes with
copious amount of distilled water.
12. Empty out the water collected in the dishes at least five times.
13. Repeat steps until all traces of contaminating material are
removed (see Note 22).
14. It is essential to equilibrate the temperature of the Agilent
Oligo aCGH/ChIP-on-Chip Wash Buffer 2 at 37 °C for opti-
mal performance (see Note 23).
15. Add Agilent Oligo aCGH/ChIP-on-Chip Wash Buffer 2 to a
50 mL pyrex jar and warm overnight in a water bath (or incu-
bator) set to 37 °C.
16. Put a glass slide-staining jar secured with parafilm and warm
overnight in a water bath (or incubator) set to 37 °C.
178 Giancarlo Feliciello et al.

3.4 Washing Purpose Washing steps are meant to remove the excess of unbound
probes or probes loosely attached to non-complementary DNA
while leaving the matched probe target intact. Additionally, they
clean the corner spots of the main grid – necessary to position
correctly the grid template – from contaminating materials (i.e.,
dust or dirty particles).

Duration
1. Hands-on time: 5 min.
2. Disassembly in Agilent Oligo aCGH/ChIP-on-Chip Wash
Buffer 1: 1 min.
3. Agilent Oligo aCGH/ChIP-on-Chip Wash Buffer 1, wash step
1: 2 min 30 s.
4. Agilent Oligo aCGH/ChIP-on-Chip Wash Buffer 1, wash step
2: 2 min 30 s.
5. Agilent Oligo aCGH/ChIP-on-Chip Wash Buffer 2, wash step
1: 30 s.
6. Agilent Oligo aCGH/ChIP-on-Chip Wash Buffer 2, wash step
2: 30 s.
7. Acetonitrile wash: 5 s.
Procedure
. Step 1: Disassembly of the Agilent Hybridization Assembly.

1. Fill a slide-staining jar with 50 mL of room temperature Agi-

lent Oligo aCGH/ChIP-on-Chip Wash Buffer 1 used only for
the disassembly (see Note 24).
2. Remove the Agilent Hybridization Assembly from the Agilent
Hybridization Oven (see Note 25).
3. Untighten the clamp and disassemble the Agilent Hybridiza-
tion Assembly (see Note 26).
4. Remove the slide sandwich (aCGH and Gasket Slide) from the
base of the Agilent SureHyb Chamber and transfer it into the
slide-staining jar containing the Agilent Oligo aCGH/ChIP-
on-Chip Wash Buffer 1 (see Note 27).
5. Using a plastic forceps, carefully, peel open the slide sandwich
by exerting a slight pressure between the two-stacked glasses
(see Note 28).
. Step 2: Washing Steps (see Note 29).

1. Before starting the washing steps, under a fume hood, fill a

slide-staining dish with Acetonitrile in sufficient amount to
cover an aCGH slide.
 Array-Based Comparative Genomic Hybridization for the Detection of Copy. . . 179

2. Transfer the aCGH slide into a new slide-staining jar filled with
Agilent Oligo aCGH/ChIP-on-Chip Wash Buffer 1 and wash
on the horizontal oscillator at room temperature for 2 min and
30 s, 120 rpm (see Note 30).
3. Invert the aCGH slide and place it back again in the same slide-
staining jar, wash on the horizontal oscillator at room temper-
ature for 2 min and 30 s, 120 rpm (see Note 31).
4. Shortly before the end of the wash prepare the slide-staining jar
and the Agilent Oligo aCGH/ChIP-on-Chip Wash Buffer
2 that have been equilibrated overnight at 37 °C.
5. Transfer the aCGH slide into the slide-staining jar filled with
Agilent Oligo aCGH/ChIP-on-Chip Wash Buffer 2 and wash
on the horizontal oscillator at 37 °C for 30 s, 120 rpm (see
Note 27).
6. Invert the aCGH slide and place it back again in the same slide-
staining jar, wash on the horizontal oscillator at 37 °C for 30 s,
120 rpm.
7. Dip the aCGH slide shortly (3–5 s) into the slide-staining dish
filled with Acetonitrile.
8. Using a KimTech tissue, wipe gently the remaining drops on
the side of the slide (see Note 32).
9. Place the aCGH slide in the Slide Holder with the “Agilent”
side of the slide facing up.
10. Place the Ozone barrier slide on top of the aCGH slide (see
Note 33).
11. Close the Slide Holder and push on the tab end until you hear
it click.
12. The aCGH slide is now ready for scanning (see Note 34).

3.5 Scanning and Purpose The aCGH slides are scanned into image files using a
Extraction specific microarray scanner. During this operation the spot inten-
sities for the two colors are measured and a .tif image, necessary for
the data extraction, is generated. In the Feature extraction software,
the fluorescence data are translated into log ratio, allowing the
identification of aberrations in the Test Sample.

Duration
1. Hands-on time: 5 min.
2. Scanning time: Depending on the number of aCGH slides
analyzed.
3. Extraction time: Depending on the number of aCGH slides
analyzed.
180 Giancarlo Feliciello et al.

Procedure
. Step 1: Scanning.

1. Switch on the Agilent Microarray Scanner 30 min before to

allow the laser to warm up until the light of the scanner turns
from “orange” to “green.”
2. Start the program and wait until the instrument is connected to
the software.
3. Put the assembled Slide Holder with the Ozone Barrier Slide
cover into the scanner carousel always starting from position
number 1.
4. Select the Start Slot m and End Slot n representing, respec-
tively, the first and last aCGH slide to be scanned.
5. Select Profile Agilent G3_aCGH and verify the following
settings:
– Dye channel: R + G (red and green).
– Scan region: Agilent HD (61 × 21.6 mm).
– Scan resolution: 3 μm.
– Tiff file dynamic range: 16 bit.
– Red PMT gain: 100%.
– Green PMT gain: 100%.
– XDR: no XDR.
6. Select the desired location to save the .tif file in the Output Path
Browse.
7. Verify that the Scanner status in the main window says Scanner
Ready.
8. Click Scan Slot m-n on the Scan Control main window.
. Step 2: Analyze microarray image.

1. After the scanning is complete, extract the feature with Agilent

Feature Extraction Software.
2. A QC metrics file generated after data extraction is used to
evaluate the relative data quality of the hybridization set and if
potential errors have happened during the analysis (i.e., align-
ment grid not properly placed).
3. The thresholds that can be observed when analyzing single cells
according to this protocol are the following:
– BGNoise: <10
– Signal Intensity: >50
– Signal to Noise: >30
– Reproducibility: <0.2
– DLRS: <1
 Array-Based Comparative Genomic Hybridization for the Detection of Copy. . . 181

4 Notes

1. Alternative protocol: Random-primed DNA labeling approach.

Test and Reference DNA samples are labeled using SureTag
DNA Labeling Kit (Agilent Technologies) according to the
instruction provided by the supplier (Agilent Oligonucleotide
Array-Based CGH for Single Cell Analysis, August 2015).
Briefly, 1.0 μg of the purified input DNA (see Note 3) diluted
in 26 μL of H2O is supplemented with 5 μL of Random Primer
Mix to a final volume of 31 μL. Samples are denatured at 95 °C
for 3 min and then kept at 4 °C or on ice. To the 31 μL of
denatured DNA add the following ingredients: 10 μL of 5×
Reaction Buffer, 5 μL of 10× dNTP Mix, 3 μL of Cy5-dUTP
(Test) or Cy3-dUTP (Reference) and 1 μL of Exo(-) Klenow
fragment. Incubate the labeling reaction at 37 °C for 2 h,
followed by an inactivation step at 65 °C for 10 min. Purify
the labeled DNA using Amicon Ultra 0.5 purification system
with a size cut-off of 30 kDa. Quantify DNA yields and dye
incorporation rates as described in the main protocol. Typical
specific activities are 20–50 pmol/μg for Cyanine 3 and
15–40 pmol/μg for Cyanine 5, with a final DNA yield of
9–15 μg.
2. Take care to keep all the reagents on ice during the reaction
preparation, unless otherwise indicated. Return enzymes,
dNTPs, and Cyanine nucleotides at -20 °C as soon as possible.
Check that the correct program is set in the thermocycler and
that evaporation of the samples is avoided by using a heated lid.
Make sure that all the pipette aids are calibrated. All these
conditions, if improperly managed, can result in a suboptimal
specific activity and degree of labeling of the samples.
3. Sometimes the amount of DNA that can be available is limited,
especially in the case of clinical samples (CTCs or DCCs). For
this reason, the original WGA can be re-amplified for a limited
number of cycles in order to generate sufficient amount of
DNA (especially needed for the enzymatic labeling, see Note
1). The short re-amplification does not introduce any signifi-
cant PCR bias that could affect the reliability of the aCGH in
terms of aberrations detection. Run three reactions in parallel
for each sample as follows: 5 μL Expand Long Template Buffer
1 (Roche Diagnostic), 1 μM of the Lib1 primer, 1.75 μL
dNTPs (10 mM), 1.25 μL BSA (Roche Diagnostic), 2.5 U of
Expand-Long-Template DNA Polymerase (Roche Diagnos-
tic), and 1 μL of the template DNA. Set the thermocycler as
follows: 1 cycle of 94 °C for 60 s, 60 °C for 30 s, 65 °C for
2 min, 10 cycles of 94 °C for 30 s, 60 °C for 30 s, 65 °C for
2 min (extended by 20 s/cycle). Pool the reactions after
182 Giancarlo Feliciello et al.

completion and use directly for PCR-based labeling using

incorporation of dye-conjugated dNTPs as described in the
main procedure or, alternatively, purify the reamplified DNA
using Amicon Ultra 0.5 purification system with a size cut-off
of 30 kDa and proceed with the Random-primed DNA label-
ing approach (see Note 1).
4. In order to reduce pipetting errors of small volumes of DNA
always add 1 μL of DNA to 99 μL of PCR reaction, mix well by
short vortexing, and spin down to drive contents off the walls
and lid. Divide the reaction into two tubes of 50 μL each. This
procedure ensures a proper homogeneous mixing of the
DNA-reaction solution minimizing PCR biases in the individ-
ual reactions.
5. Remember to change the centrifugation speed to 1000 RCF. If
the speed is too high, this will cause perforation of the column
membrane and consequent loss of the sample.
6. The volume per sample will be approximately 12–20 μL.
7. The efficient suppression of repeated sequences by the Cot-1
DNA during hybridization represents a relevant factor for the
success of array-CGH. Repetitive freeze and thawing of Cot-1
DNA tend to reduce the quality and may alter the suppression
efficiency. The use of lower-quality Cot-1 DNA may result in
incomplete repeat suppression, altered log ratios, and increased
background.
8. All the indicated hands-on processing times were estimated for
experiments involving four arrays (e.g., 16 samples in total).
Accordingly, hands-on processing times may deviate, when
processing higher or lower number of samples in one run. We
recommend to get acquainted with the method before proces-
sing a higher number of samples. To practice hybridization,
prepare a 1:1 2× HI-RPM Hybridization Buffer and water mix
and use a microscope slide or used microarray slide, and a
gasket slide. You can use the same slide to practice wash and
placement of slide in the slide holder.
9. Switch on in advance the Agilent Microarray Oven and the
Thermomixers. It may take a while before the exact tempera-
tures will be reached. Check always that the correct tempera-
ture is reached and make sure that the oven is calibrated. Too
high temperatures of the Agilent Microarray Oven may be
responsible for post-labeling signal loss due to hybridization
conditions that are too stringent. Too low temperatures may
decrease the stringency and increase the BGNoise and DLRSD.
10. For this and all future steps we recommend working in
dimmed light and covering your samples with aluminum foil
to slow down any possible photo-bleaching.
Array-Based Comparative Genomic Hybridization for the Detection of Copy. . . 183

11. The Hybridization Mix is viscous and foams very intensively

due to the presence of Tween20 and Igepal. Avoid to vortex
intensively, mix by pipetting slowly up and down or by gently
inverting the tube. Care must be taken when pipetting to
ensure that the correct volumes are used. Using an incorrect
volume will change the stringency of the hybridization solution
and could lead to poor probe performance. Always prepare the
Hybridization Mix with an excess of one sample.
12. After pairing the Cy5 labeled Test and Cy3 labeled Reference,
the sample should look purple in color. If it is too pink or too
blue, it will indicate, respectively, inefficient Cy5 or Cy3 label-
ing. Inefficient labeling can result from WGA with low GII
index, many cycles of freeze-thaw of dNTPs mix and Cyanine
nucleotides, light and/or air exposure of the dyes, incorrect
volume pipetting, wrong temperatures or times, and degraded
enzymes not properly stored on ice.
13. Always protect the samples from direct sunlight exposure by
covering the Thermomixer with an aluminum foil. Keep
strictly the incubation time because this is a time-sensitive step.
14. The duration of this step can be extended, if necessary, until the
moment of the hybridization. Nevertheless, extend only for
the necessary time because at 37 °C part of the DNA may start
to renature, rendering it unavailable for the hybridization on
the array. This process could cause loss of information.
15. Keep the temperature of hybridization sample mixtures as close
as possible to 37 °C. To do this, process the samples one by one
by keeping them in the Thermomixer 2.
16. Make sure that the Gasket Slide is lying flat at the bottom of the
Agilent SureHyb Chamber. Help you in the process by using a
plastic forceps to align the slide at one side of the rectangular
section of the chamber base. Ensure that the gasket slide is
flush with the chamber base and is not unlatched from the
borders. This will help to keep the slide stable, avoiding acci-
dental moving, during the dispensing phase of the samples.
Keep in mind that this phase is also very sensitive to cross-
contamination, especially when samples are loaded onto each
of the four adjacent Gasket wells.
17. Start to dispense the samples from the “Subarray 1” (the
nearest to the “Agilent” label) to the “Subarray 4” (the most
far). Slowly dispense your sample at the center of the well
without touching the glass slide. To prevent leakage of the
hybridization solution, avoid touching the O-ring of the gasket
slide with the hybridization mixture. Proceed with this step
swiftly avoiding that the samples loaded on the array are cool-
ing down for too long. Take care to not form too many
184 Giancarlo Feliciello et al.

bubbles on the array. In addition, avoid pipetting the last drop

of Hybridization Mix on the Gasket Slide because it will form a
bubble.
18. Hold the Agilent aCGH slide on the sides or on the barcode
side only. Approach it, as near as possible, to the Gasket Slide
and release when you reach a short distance between them, in
order to avoid spreading the Hybridization-Sample mixture
outside of the Gasket well and causing material loss and sample
cross-contamination. Once the aCGH slide is placed on the
Gasket Slide, do not try to readjust the position as this may
cause spillage and cross-contamination of samples.
19. Control, by rotating the chamber, if there are stationary bub-
bles. Knock the Hybridization Assembly with your hand or on
a hard surface until the stationary bubbles become mobile or
they merge to form a large mobile bubble. It is important that
multiple small bubbles are not stationary at the same place.
This would compromise the hybridization of the DNA in the
affected area of the array.
20. Be always sure that each position in the Agilent Oven is bal-
anced, as you would do for a centrifuge. Use an empty assem-
bled Hybridization Chamber to balance the loaded
Hybridization Chambers on the rack to prevent unnecessary
overburden on the oven motor.
21. The hybridization step can be shortened or elongated accord-
ing to the necessity. Do not hybridize less than 20 h because
this may impact the quality of the hybridization.
22. Make sure that the staining jar and dishes are very clean. High
BGNoise values can be caused by glassware not properly
washed. Always use MilliQ water to wash equipment, do not
use tap water or any detergents. If high BGNoise persists, the
glassware should be washed with acetonitrile followed by rinses
with MilliQ water to remove build-up of unincorporated dye
molecules.
23. Post-labeling signal loss of Cy5 can be due to the incorrect
temperature of Agilent Oligo aCGH/ChIP-on-Chip Wash
Buffer 2. Always ensure that the temperature is set correctly
at 37 °C.
24. We recommend preparing a staining jar filled with wash buffer
for each sample or, alternatively, washing the staining jar in
between each array disassembly with copious amounts of
MilliQ water and refill with fresh buffer. In this way, cross-
contamination between slides can be minimized.
25. We recommend washing array slides one by one. If there are
arrays left in the Hybridization Oven, keep them until you are
finished washing the first slide. Control the level of the fluid
among the four Gasket Chambers to ensure that there was no
Array-Based Comparative Genomic Hybridization for the Detection of Copy. . . 185

loss of the samples due to leakage. Cross-contamination can

arise during the overnight incubation due to gasket leakage. If
you notice a loss of fluid, you may consider that the presence of
contamination can result in altered and inaccurate patient data.
26. This step may take 2–3 min because the Hybridization Assem-
bly is too hot and it is not possible to untighten the clamp. Wait
until it reaches a lower temperature to avoid burns of the
hands.
27. We recommend wearing multiple layers of gloves that can be
quickly changed between each step. Change them after the
disassembly of the Hybridization Assembly as you can notice
a black smear on your gloves. Change your gloves when you
step from Agilent Oligo aCGH/ChIP-on-Chip Wash Buffer
1 to Wash Buffer 2 and to Acetonitrile. This procedure will
avoid transporting traces of contaminants during the different
washing steps.
28. Do not rush at this stage. Exert a light pressure to avoid the
harsh separation of the Gasket Slide from the aCGH slide that
may cause the loss of the DNA-probe complex from the surface
of the array and/or the break of the glass. Let the Agilent
Oligo aCGH/ChIP-on-Chip Wash Buffer 1 slowly penetrate
between the two slides. This will reduce adherence and facili-
tate the separation.
29. Wash slides in an ozone-controlled environment. Ozone can
affect the fluorescent signal during the washing step of the
arrays. You may notice Cy5 signal red signal loss around the
edges of the feature because after the last wash buffer, the outer
edges will dry before the center, and, as a consequence, they
will become exposed earlier to the degrading effect of ozone.
Signal degradation can potentially compromise data quality.
We recommend using small batches of slides that can be
washed and scanned in a short time to minimize exposure
to air.
30. Keep strictly the incubation time and rpm of the Horizontal
Oscillator. High BGNoise is often introduced if uncorrected
wash steps are performed. The indicated rpm is necessary to
remove all non-uniformities from the surface of the array and
excess of unbound labeled DNA.
31. By inverting the aCGH slide, it is ensured a more homoge-
neous wash throughout the entire array, in the upper and in the
lower part.
32. Do not wipe with the KimWipe the active surface of the aCGH
slide (Agilent site) if there are residual drops. Rather, submerge
again shortly after the aCGH slide in Acetonitrile. Limit your-
self to clean only the edges from eventual small drops. In
general, Acetonitrile is very volatile and it does not leave the
aCGH slide wet, indeed it leaves very dry.
186 Giancarlo Feliciello et al.

33. The second stage where ozone can cause a problem occurs
when the slide is being scanned. The increased air circulation
at the open end of the slide holder (left side) causes more
ozone to contact the Cy5 thereby inducing a visible degrada-
tion gradient. To prevent degradation during scanning we
recommend using a barrier slide, which sits over the microar-
ray. This seals the array in a small chamber and limits the extent
of ozone degradation.
34. Scanning the slides immediately after this step is highly recom-
mended to minimize the reduction in the fluorescence signal
intensity due to environmental oxidants and possible photo-
bleaching. In case immediate scanning is not possible keep the
slides in black light-sealed boxes (you can use the boxes the
array slides come in) or slide-holding boxes wrapped in
aluminum foil.

References
1. Pinkel D, Segraves R, Sudar D, Clark S, disseminated cancer cell--the unit of selection.
Poole I, Kowbel D, Collins C, Kuo WL, Adv Cancer Res 89:35–67
Chen C, Zhai Y, Dairkee SH, Ljung BM, 7. Riethdorf S, Wikman H, Pantel K (2008)
Gray JW, Albertson DG (1998) High resolu- Review: biological relevance of disseminated
tion analysis of DNA copy number variation tumor cells in cancer patients. Int J Cancer
using comparative genomic hybridization to 123(9):1991–2006. https://doi.org/10.
microarrays. Nat Genet 20(2):207–211. 1002/ijc.23825
https://doi.org/10.1038/2524 8. Cohen SJ, Punt CJ, Iannotti N, Saidman BH,
2. Solinas-Toldo S, Lampel S, Stilgenbauer S, Sabbath KD, Gabrail NY, Picus J, Morse M,
Nickolenko J, Benner A, Dohner H, Mitchell E, Miller MC, Doyle GV, Tissing H,
Cremer T, Lichter P (1997) Matrix-based Terstappen LW, Meropol NJ (2008) Relation-
comparative genomic hybridization: biochips ship of circulating tumor cells to tumor
to screen for genomic imbalances. Genes Chro- response, progression-free survival, and overall
mosomes Cancer 20(4):399–407 survival in patients with metastatic colorectal
3. Fiorentino F (2012) Array comparative geno- cancer. J Clin Oncol 26(19):3213–3221.
mic hybridization: its role in preimplantation https://doi.org/10.1200/JCO.2007.15.
genetic diagnosis. Curr Opin Obstet Gynecol 8923
24(4):203–209. https://doi.org/10.1097/ 9. Klein CA, Schmidt-Kittler O, Schardt JA,
GCO.0b013e328355854d Pantel K, Speicher MR, Riethmuller G (1999)
4. Munne S (2012) Preimplantation genetic diag- Comparative genomic hybridization, loss of
nosis for aneuploidy and translocations using heterozygosity, and DNA sequence analysis of
array comparative genomic hybridization. Curr single cells. Proc Natl Acad Sci U S A 96(8):
Genomics 13(6):463–470. https://doi.org/ 4494–4499
10.2174/138920212802510457 10. Greshock J, Feng B, Nogueira C, Ivanova E,
5. Fiorentino F, Caiazzo F, Napolitano S, Perna I, Nathanson K, Protopopov A, Weber
Spizzichino L, Bono S, Sessa M, Nuccitelli A, BL, Chin L (2007) A comparison of DNA copy
Biricik A, Gordon A, Rizzo G, Baldi M (2011) number profiling platforms. Cancer Res
Introducing array comparative genomic hybri- 67(21):10173–10180. https://doi.org/10.
dization into routine prenatal diagnosis prac- 1158/0008-5472.CAN-07-2102
tice: a prospective study on over 1000 11. Haraksingh RR, Abyzov A, Gerstein M, Urban
consecutive clinical cases. Prenat Diagn AE, Snyder M (2011) Genome-wide mapping
31(13):1270–1282. https://doi.org/10. of copy number variation in humans: compara-
1002/pd.2884 tive analysis of high resolution array platforms.
6. Klein CA (2003) The systemic progression of PLoS One 6(11):e27859. https://doi.org/10.
human cancer: a focus on the individual 1371/journal.pone.0027859
 Array-Based Comparative Genomic Hybridization for the Detection of Copy. . . 187

12. Fiegler H, Geigl JB, Langer S, Rigler D, One 8(6):e67031. https://doi.org/10.1371/

Porter K, Unger K, Carter NP, Speicher MR journal.pone.0067031
(2007) High resolution array-CGH analysis of 16. Czyz ZT, Hoffmann M, Schlimok G, Polzer B,
single cells. Nucleic Acids Res 35(3):e15. Klein CA (2014) Reliable single cell array CGH
https://doi.org/10.1093/nar/gkl1030 for clinical samples. PLoS One 9(1):e85907.
13. Fuhrmann C, Schmidt-Kittler O, Stoecklein https://doi.org/10.1371/journal.pone.
NH, Petat-Dutter K, Vay C, Bockler K, 0085907
Reinhardt R, Ragg T, Klein CA (2008) High- 17. Czyż ZT, Klein CA (2015) Deterministic
resolution array comparative genomic hybridi- whole-genome amplification of single
zation of single micrometastatic tumor cells. cells. In: Kroneis T (ed) Whole genome ampli-
Nucleic Acids Res 36(7):e39. https://doi. fication methods and protocols. Springer,
org/10.1093/nar/gkn101 New York
14. Bi W, Breman A, Shaw CA, Stankiewicz P, 18. Polzer B, Medoro G, Pasch S, Fontana F,
Gambin T, Lu X, Cheung SW, Jackson LG, Zorzino L, Pestka A, Andergassen U, Meier-
Lupski JR, Van den Veyver IB, Beaudet AL Stiegen F, Czyz ZT, Alberter B, Treitschke S,
(2012) Detection of >/=1 Mb microdeletions Schamberger T, Sergio M, Bregola G,
and microduplications in a single cell using Doffini A, Gianni S, Calanca A, Signorini G,
custom oligonucleotide arrays. Prenat Diagn Bolognesi C, Hartmann A, Fasching PA, San-
32(1):10–20 dri MT, Rack B, Fehm T, Giorgini G,
15. Mohlendick B, Bartenhagen C, Behrens B, Manaresi N, Klein CA (2014) Molecular
Honisch E, Raba K, Knoefel WT, Stoecklein profiling of single circulating tumor cells with
NH (2013) A robust method to analyze copy diagnostic intention. EMBO Mol Med 6(11):
number alterations of less than 100 kb in single 1371–1386. https://doi.org/10.15252/
cells using oligonucleotide array CGH. PLoS emmm.201404033
 Chapter 12

Single Cell Micro RNA Sequencing Library Preparation

Sarah M. Hücker and Stefan Kirsch

Abstract
Micro RNAs represent important post-transcriptional regulators in health and are involved in the onset of
many diseases. Therefore, the further characterization of physiological miRNA functions is an important
basic research question, and miRNAs even have high potential as biomarkers both for prognosis and
diagnosis. In order to exploit this potential, it is mandatory to accurately quantify the miRNA expression
not only in bulk but also on the single-cell level. Here, we describe a protocol, which facilitates miRNA
sequencing library preparation of very low input samples, single cells, and even clinical samples such as
circulating tumor cells. The protocol can be combined with different single-cell isolation methods (e.g.,
micromanipulation and FACS sorting). After cell lysis, sequencing adapters are ligated to the miRNAs,
other ncRNA species, and adapter dimers are reduced by exonuclease digest, the miRNA library is reverse
transcribed, amplified, and purified. Furthermore, quality controls are described to select only high-quality
samples for sequencing.

Key words miRNA, small RNA, ncRNA, Single-cell sequencing, NGS, CTC

1 Introduction

Micro RNAs govern mRNA abundance and represent important

post-transcriptional regulators in many biological processes
[1]. The expression levels of miRNAs are very heterogeneous
between developmental stages and different tissue or cell types
[2]. Also, in diseases like cancer the expression level of some miR-
NAs is altered [3, 4]. In order to investigate the role of miRNAs in
health and disease in further detail, accurate detection and quanti-
fication methods are required. Compared to mRNA sequencing,
miRNA sequencing poses some additional challenges: First, miR-
NAs neither have a polyA tail nor any other specific sequence
feature, which discriminates them from other RNA types inside
the cell. Second, mature miRNAs have a size of only 20–24 nt,
which hinders the discrimination and separation of adapter dimers
from libraries with insert. Ideally, the library preparation method
already avoids the formation of adapter dimers. Third, the quantifi-
cation accuracy is only moderate due to ligation and amplification

Miodrag Gužvić (ed.), Single Cell Analysis: Methods and Protocols, Methods in Molecular Biology, vol. 2752,
https://doi.org/10.1007/978-1-0716-3621-3_12, © Springer Science+Business Media, LLC, part of Springer Nature 2024

189
190 Sarah M. Hücker and Stefan Kirsch

biases [5–8]. For bulk miRNA sequencing, several commercial kits

and protocols are available [9, 10]. However, for single-cell miRNA
sequencing only very few protocols have been published [11–
15]. Therefore, we compared 19 miRNA sequencing protocol
variants and selected the best-performing protocol regarding the
amount of adapter dimers, the amount of reads mapping to anno-
tated miRNA loci, the number of different detected miRNAs, the
reproducibility, and the quantification accuracy. The best-
performing protocol showed reliable results with very low amounts
of miRNA input, single cells of different cell lines, and circulating
tumor cells from the blood of small cell lung cancer patients. In
addition, the protocol captured other non-coding RNAs like
lncRNAs, rRNAs, snoRNAs, and piRNAs [16]. The protocol can
be performed in 2–3 days and the hands-on-time is about 8 h. Up
to 20 samples can be processed in parallel. First, reaction vessels
with lysis buffer are prepared. Then, a single-cell suspension of the
target cells has to be made. Single cells are picked using microma-
nipulation into the lysis buffer and could be stored at -80 °C for up
to 6 months. During the library preparation, cells are lysed, 5.8S
rRNA is depleted, chemically modified adapters are ligated to the 3′
and 5′ miRNA ends, remaining unligated adapter is removed,
reverse transcription is performed, and the cDNA is amplified in
two sequential PCRs. In order to enrich for fragments with miRNA
inserts, a bead size selection is performed. Finally, the concentra-
tion and fragment length distribution of the library is controlled.
Figure 1 summarizes the protocol workflow.

2 Materials

To avoid sample contamination or RNA degradation, only sterile,

DNase- and RNase-free reaction tubes should be used, and all
pipetting steps should be conducted using sterile filter/barrier
tips. Furthermore, all plastic labware should be low retention to
reduce sample loss.

2.1 Oligonucleotides All oligonucleotides should be purified by HPLC. The lyophilized

oligonucleotides are dissolved in nuclease-free water to a stock
concentration of 100 μM. DNA oligonucleotides can be stored at
-20 °C. RNA oligonucleotides are aliquoted and stored at -80 °C.
The aliquots are thawed only once and then discarded. The names
and sequences of the required oligonucleotides are listed in Table 1.

2.2 Reagents If not stated differentially, the reagents are stored at -20 °C and
thawed on ice.
1. 1% Triton X-100 (Sigma-Aldrich) in nuclease-free water, store
at RT.
 miRNA Sequencing Library Preparation 191

1. Cell isolation & lysis

rRNA
2. 5.8S rRNA removal
block

miRNA rApp-MP 3‘ adapter ddC 3. 3‘ Adapter ligation

3‘ ad 3‘ ad 4. Removal of non-ligated 3‘ adapter

RT RT

5‘ adapter 2‘OMe 3‘ ad 5. 5‘ Adapter ligation

RT

5‘ ad 3‘ ad
6. Reverse transcription
cDNA

i5 i7
i5 i7
7. PCR amplification

Fig. 1 Workflow of the single-cell miRNA sequencing protocol

Table 1
Names and sequences of the required oligonucleotides

Name Sequence (5′ → 3′) Note

3′ adapter rApp-T(MP)GGAATTCTCGGGTGCCAAGG-ddC rApp = 5′ preadenylation (see Note
1)
MP = methyl-phosphonate
ddC = dideoxy-cytidine
5′ adapter rGrUrUrCrArGrArGrUrUrCrUrArCrArGrUrCrCrG r = ribonucleotides
rArCrGrArUrCrHrHrHrHrHrHrHrHrCrA-2′OMe H = rA, rC or rU
2′OMe = ribose 2′ O methylation
RT primer Biotin-CCTTGGCACCCGAGAATTCCrA r = ribonucleotides
5.8S ATCGGCAAGCGACGCTCAGACAGGCGTAG TEG = triethyleneglycol spacer
rRNA CCCCGGGAGGAACCCGGGGCCGCAAGTGC
mask GTTCGAAGTGTCGATGAT-TEG-biotin
RP1 AATGATACGGCGACCACCGAGATCTACAC
primer GTTCAGAGTTCTACAGTCCGA
Index- CAAGCAGAAGACGGCATACGAGATXXXXXX XXXXXX = index sequence 1–48,
RPI 1– GTGACTGGAGTTCCTTGGCACCCGAGAATTC use the indices of the Illumina
48 CA TruSeq small RNA library
preparation kit
192 Sarah M. Hücker and Stefan Kirsch

2. 40 U/μL Recombinant RNase Inhibitor (Takara).

3. 200 U/μL T4 RNA Ligase 2 truncated KQ, 10× Reaction
Buffer and 50% w/v PEG 8000 (NEB).
4. 5 U/μL Lambda Exonuclease (NEB).
5. 50 U/μL 5′ Deadenylase (NEB).
6. 10 U/μL T4 RNA Ligase 1 ssRNA Ligase (NEB).
7. 200 U/μL SuperScript II Reverse Transcriptase and 100 mM
DTT (Invitrogen).
8. 2 U/μL Phusion Hot Start II DNA Polymerase and 5× Phu-
sion HF Buffer (Thermo Fisher).
9. 100 mM Tris-buffered ATP (Thermo Fisher).
10. 10 mM each PCR Deoxynucleotide Mix (Roche).
11. 10× Taq DNA Polymerase Buffer (Roche).
12. Ampure XP Beads (Beckman Coulter), store at 4 °C.
13. 80% ethanol solution, store at 4 °C.
14. Nuclease-free water, store at RT.
15. Phosphate buffered saline, PBS, store at RT.
16. Bioanalyzer High Sensitivity DNA Kit (Agilent).
17. Qubit™ 1× dsDNA HS Assay Kit (Invitrogen).

2.3 Equipment 1. Inverted microscope.

2. Micromanipulator Patchman with CellTram pump
(Eppendorf).
3. AdcellTM diagnostic slides (Thermo Fisher).
4. Micro Hematocrit capillary (Brand).
5. DMZ Universal Puller (Zeitz Instruments).
6. Thermo cycler.
7. 2100 Bioanalyzer Instrument (Agilent).
8. Qubit 2.0 Fluorometer (Invitrogen).

3 Methods

Single-cell miRNA-Seq is very sensitive to contamination with

environmental miRNA or to RNA degradation caused by environ-
mental RNases. Therefore, the experimenter should wear a lab coat
and gloves. All pipetting steps should be conducted under a laminar
flow hood. Ideally, the hood and the lab equipment (e.g., tube
racks, ice boxes) are not commonly used to process amplified
nucleic acids.
 miRNA Sequencing Library Preparation 193

The protocol Subheadings 3.1, 3.2, and 3.3 should be per-

formed on ice. Subheading 3.4 is performed at room temperature.
The required reagent volumes for one sample are given. If more
samples are processed, the volumes have to be increased respec-
tively including some excess. We do not recommend to process
more than 20 samples in parallel (see Note 2).

3.1 Lysis Buffer 1. Prepare a 1% Triton X-100 solution with water.

Preparation 2. Prepare the lysis buffer master mix: 0.39 μL 1% Triton X-100,
0.1 μL 40 U/μL RNase inhibitor, and 1.51 μL water.
3. Vortex lysis buffer master mix, pipet 2 μL per sample into
0.2 mL reaction tubes, and spin down.
4. Store tubes with lysis buffer on ice.

3.2 Single Cell 1. Harvest cells and resuspend cell pellet in PBS (see Notes 3 and
Isolation 4).
2. Pick single cells by micromanipulation (see Notes 5 and 6).
3. Aspirate single cell into 1 μL PBS, pipet into lysis buffer, and
centrifuge down.
4. Store the tubes with the isolated cell on dried ice while isolating
additional single cells.
5. Long-term storage of isolated cells in lysis buffer is possible at
-80 °C.

3.3 miRNA-Seq 1. Process samples from Subheading 3.2 immediately or thaw in

Library Preparation ice samples that were stored at -80 °C.
2. Dilute 5.8S rRNA Mask oligo from 100 μM to 5 μM using
water.
3. Pipet 2 μL of 5 μM 5.8S rRNA Mask oligo to the sample and
spin down (Note 7).
4. Incubate at 72 °C for 20 min in a thermo cycler.
5. Prepare the 3′ Adapter ligation mix: 0.16 μL 100 μM 3′
Adapter oligo, 1.28 μL 50% (w/v) PEG 8000, 0.64 μL 10×
T4 RNA Ligase buffer, 0.25 μL T4 RNA Ligase 2 truncated
KQ, 0.1 μL 40 U/μL RNase inhibitor, and 0.57 μL water.
6. Mix 3′ Adapter ligation mix, pipet 3 μL to the sample and spin
down (see Note 8).
7. Incubate at 30 °C for 6 h, then at 4 °C for 10 h (overnight) in a
thermo cycler.
8. Prepare the adapter removal mix: 0.5 μL 100 μM RT primer,
0.5 μL 5 U/μL Lambda Exonuclease, 0.5 μL 50 U/μL 5′
Deadenylase, and 0.5 μL water.
194 Sarah M. Hücker and Stefan Kirsch

9. Mix adapter removal mix, pipet 2 μL to the sample and

spin down.
10. Incubate at 30 °C for 15 min, then at 37 °C for 15 min in a
thermo cycler.
11. Prepare the 5′ Adapter ligation mix: 0.12 μL 100 μM 5′
Adapter oligo, 0.08 μL 100 μM ATP, 0.3 μL 10× T4 RNA
Ligase buffer, 1.35 μL 10 U/μL T4 RNA Ligase 1 ssRNA
Ligase, and 0.15 μL water.
12. Mix 5′ Adapter ligation mix, pipet 2 μL to the sample, and
spin down.
13. Incubate at 37 °C for 1 h in a thermo cycler.
14. Prepare reverse transcription mix: 2.2 μL 10× Taq DNA Poly-
merase Buffer, 1.35 μL 100 mM DTT, 0.85 μL 10 mM
dNTPs, 0.5 μL 200 U/μL SuperScript II Reverse Transcrip-
tase, and 0.1 μL 40 U/μL RNase Inhibitor.
15. Mix reverse transcription mix, pipet 5 μL to the sample, and
spin down.
16. Incubate at 42 °C for 1 h, then at 70 °C for 15 min in a thermo
cycler.
17. Prepare first PCR mix: 0.3 μL 100 μM RP1 primer, 0.45 μL
10 mM dNTPs, 6 μL 5× Phusion HF Buffer, 0.5 μL 2 U/μL
Phusion Hot Start II DNA Polymerase, and 5.75 μL water.
18. Mix first PCR mix, pipet 13 μL to the sample, and spin down.
19. Perform the PCR program shown in Table 2 in a thermo cycler.
20. Prepare the indexing PCR mix: 0.2 μL 100 μM RP1 primer,
0.5 μL 10 mM dNTPs, 5 μL 5× Phusion HF Buffer, 0.25 μL
2 U/μL Phusion Hot Start II DNA Polymerase, and 17.55 μL
water.

Table 2
Program of the first PCR

Step Temperature (°C) Time (s) Additional

1 98 30
2 98 10
3 60 30
4 72 30
5 Go to step 2, 12 times
6 72 300
7 4 Forever
 miRNA Sequencing Library Preparation 195

Table 3
Program of the indexing PCR

Step Temperature (°C) Time (s) Additional

1 98 30
2 98 10
3 67 30
4 72 30
5 Go to step 2, 12 times
6 72 300
7 4 Forever

21. Mix indexing PCR mix, pipet 23.5 μL into a fresh 0.2 mL
reaction tube, and spin down.
22. Mix the sample after first PCR (step 19) and pipet 1 μL each
into the indexing PCR mix.
23. Pipet 0.5 μL of 100 μM Index RPI oligo into the indexing PCR
mix. Use a different Index RPI oligo for every sample (see
Note 9).
24. Mix sample, spin down, and perform the PCR program shown
in Table 3 in a thermo cycler.

3.4 Bead Size 1. Bring Ampure XP Beads to room temperature.

Selection 2. Vortex beads, pipet 25 μL of beads to 25 μL sample after the
indexing PCR, and mix by pipetting up and down 10 times (see
Note 10).
3. Incubate 10 min at RT.
4. Place tube into a magnetic rack and incubate 4 min.
5. Pipet the supernatant into a fresh 0.2 mL reaction tube (see
Note 11).
6. Vortex beads, pipet 15 μL of beads to the supernatant and mix
by pipetting up and down 10 times.
7. Incubate 10 min at RT.
8. Place tube into a magnetic rack and incubate 4 min.
9. Remove and discard supernatant.
10. Pipet 200 μL of 80% ethanol to every sample and incubate 30 s
(see Note 12).
11. Remove and discard supernatant.
12. Repeat steps 10 and 11 once.
13. Remove any residual liquid with a small pipetting tip.
196 Sarah M. Hücker and Stefan Kirsch

14. Incubate with open lid for 2 min (Note 13).

15. Take tube from magnetic rack and resuspend beads in 16 μL of
water by pipetting up and down.
16. Incubate 2 min at RT.
17. Place tube into a magnetic rack and incubate 4 min.
18. Pipet 15 μL of the supernatant into a fresh 0.2 mL reaction
tube and discard beads.
19. The sample can be stored at -20 °C for at least 1 year.

3.5 Library Quality 1. Determine the dsDNA concentration using the Qubit™ 1×
Control dsDNA HS Assay Kit: Prepare standards (10 μL standard +
190 μL Qubit working solution) and samples (1 μL sam-
ple + 199 μL Qubit working solution), vortex, incubate for
2 min in the dark, and measure dsDNA concentration
(Note 14).
2. If the DNA concentration is higher than 2 ng/μL, dilute 1 μL
sample to 2 ng/μL with water. If the DNA concentration is
lower than 2 ng/μL, just use 1 μL undiluted sample.
3. Determine the fragment length distribution on a Bioanalyzer
using a High Sensitivity DNA Kit. Pipet 1 μL of (diluted)
sample on the chip (see Notes 15–17).

4 Notes

1. An already preadenylated 3′ Adapter can be ordered or the

preadenylation can be performed using the 5′ DNA Adenyla-
tion Kit (NEB).
2. We recommend to include two controls: A positive control
consisting of a cell pool or a synthetic miRNA standard (e.g.,
10 pg miRXplore Universal Reference from Miltenyi) and a
negative control without a cell to detect potential
contaminations.
3. The protocol is working not only with cultured cell lines but
also with single cells of body fluids and tissues. In this case, the
target cell population has to be enriched and converted into a
single-cell suspension. It is important to check in advance if
these additional processing steps influence RNA quality (e.g.,
total RNA could be isolated and the RNA integrity score could
be determined on the Bioanalyzer).
4. The time between cell harvest and the transfer into lysis buffer
should be as short as possible because long processing times
could influence miRNA expression.
 miRNA Sequencing Library Preparation 197

5. If a fluorescence microscope is used, staining of surface markers

could be applied to identify target cells. It should be controlled
in advance so that the staining protocol does not influence
RNA quality.
6. Other methods than micromanipulation can be used for single-
cell isolation (e.g., flow cytometry or single-cell printing).
However, the total volume after cell isolation has to be exactly
3 μL.
7. Always pipet on the edge of the tube at the border of the liquid.
Do not dip the pipetting tip into the liquid, this can lead to
sample loss.
8. PEG 8000 is very viscous. It should be pipetted very slowly and
the tip should be controlled for air bubbles and remaining
liquid on the outside. The master mix should be mixed at
least 30 s and controlled for homogenous appearance of the
liquid. Mixing by pipetting up and down might be required to
get a homogenous liquid.
9. If the samples should be sequenced together, choose compati-
ble indices regarding the Illumina pooling guide.
10. Do not vortex, only mix beads by pipetting up and down.
11. Control the supernatant in the pipetting tip for beads. If brown
beads are visible, pipet the whole liquid back into the tube,
incubate again on the magnet, and repeat the removal of the
supernatant.
12. The 80% ethanol solution should be prepared freshly
every time.
13. The beads should not dry completely, and on the other hand,
there should be no remaining liquid visible.
14. The dsDNA concentration varies between different cell types.
Libraries of single cells usually have concentrations of
0.5–10 ng/μL.
15. A high-quality library should show the main product at
140–155 bp. Products around 125 bp indicate adapter dimers
and fragments larger than 155 bp are caused by nucleic acids
other than miRNAs (Fig. 2).
16. We recommend to perform qPCR quantification of the
libraries before sequencing. The qPCR is an additional quality
control (the Cp value should be lower than 20) as well as
necessary for equimolar library pooling.
17. The miRNA libraries can be sequenced on any Illumina
sequencer. In the sample sheet, the Illumina TruSeq Small
RNA assay and index adapters have to be selected. A read
length of 50 nt is sufficient. We usually pool 25–40 libraries
for one MiSeq run.
198 Sarah M. Hücker and Stefan Kirsch

Fig. 2 Example Bioanalyzer profiles of three miRNA libraries. (a) This library shows the expected fragment
length distribution with the main product at about 150 bp. (b) This library contains a large amount of adapter
dimers visible by the main product at about 120 bp. (c) The main product is larger than 150 bp indicating
contamination by larger RNA classes

References

1. Cai Y, Yu X, Hu S, Yu J (2009) A brief review 6. Fuchs RT, Sun Z, Zhuang F, Robb GB (2015)
on the mechanisms of miRNA regulation. Bias in ligation-based small RNA sequencing
Genomics Proteomics Bioinformatics 7(4): library construction is determined by adaptor
147–154. https://doi.org/10.1016/S1672- and RNA structure. PLoS One 10(5):
0229(08)60044-3 e0126049. https://doi.org/10.1371/journal.
2. Ludwig N, Leidinger P, Becker K, Backes C, pone.0126049
Fehlmann T, Pallasch C, Rheinheimer S, 7. Giraldez MD, Spengler RM, Etheridge A,
Meder B, Stahler C, Meese E, Keller A (2016) Godoy PM, Barczak AJ, Srinivasan S, De Hoff
Distribution of miRNA expression across PL, Tanriverdi K, Courtright A, Lu S,
human tissues. Nucleic Acids Res 44(8): Khoory J, Rubio R, Baxter D, Driedonks
3865–3877. https://doi.org/10.1093/nar/ TAP, Buermans HPJ, Nolte-’t Hoen ENM,
gkw116 Jiang H, Wang K, Ghiran I, Wang YE, Van
3. Di Leva G, Garofalo M, Croce CM (2014) Keuren-Jensen K, Freedman JE, Woodruff
MicroRNAs in cancer. Annu Rev Pathol 9: PG, Laurent LC, Erle DJ, Galas DJ, Tewari M
287–314. https://doi.org/10.1146/annurev- (2018) Comprehensive multi-center assess-
pathol-012513-104715 ment of small RNA-seq methods for quantita-
4. Horita M, Farquharson C, Stephen LA (2021) tive miRNA profiling. Nat Biotechnol 36(8):
The role of miR-29 family in disease. J Cell 746–757. https://doi.org/10.1038/nbt.
Biochem 122(7):696–715. https://doi.org/ 4183
10.1002/jcb.29896 8. Wright C, Rajpurohit A, Burke EE, Williams C,
5. Baran-Gale J, Kurtz CL, Erdos MR, Sison C, Collado-Torres L, Kimos M, Brandon NJ,
Young A, Fannin EE, Chines PS, Sethupathy P Cross AJ, Jaffe AE, Weinberger DR, Shin JH
(2015) Addressing bias in small RNA library (2019) Comprehensive assessment of multiple
preparation for sequencing: a new protocol biases in small RNA sequencing reveals signifi-
recovers microRNAs that evade capture by cur- cant differences in the performance of widely
rent methods. Front Genet 6:352. https://doi. used methods. BMC Genomics 20(1):513.
org/10.3389/fgene.2015.00352 https://doi.org/10.1186/s12864-019-
5870-3
 miRNA Sequencing Library Preparation 199

9. Dard-Dascot C, Naquin D, d’Aubenton- 1264–1266. https://doi.org/10.1038/nbt.

Carafa Y, Alix K, Thermes C, van Dijk E 3701
(2018) Systematic comparison of small RNA 13. Hagemann-Jensen M, Abdullayev I,
library preparation protocols for next- Sandberg R, Faridani OR (2018) Small-seq
generation sequencing. BMC Genomics for single-cell small-RNA sequencing. Nat Pro-
19(1):118. https://doi.org/10.1186/ toc 13(10):2407–2424. https://doi.org/10.
s12864-018-4491-6 1038/s41596-018-0049-y
10. Herbert ZT, Thimmapuram J, Xie S, Kershner 14. Isakova A, Neff N, Quake SR (2021) Single-
JP, Kolling FW, Ringelberg CS, LeClerc A, cell quantification of a broad RNA spectrum
Alekseyev YO, Fan J, Podnar JW, Stevenson reveals unique noncoding patterns associated
HS, Sommerville G, Gupta S, Berkeley M, with cell types and states. Proc Natl Acad Sci
Koeman J, Perera A, Scott AR, Grenier JK, U S A 118(51). https://doi.org/10.1073/
Malik J, Ashton JM, Pivarski KL, Wang X, pnas.2113568118
Kuffel G, Mesa TE, Smith AT, Shen J, 15. Wang N, Zheng J, Chen Z, Liu Y, Dura B,
Takata Y, Volkert TL, Love JA, Zhang Y, Kwak M, Xavier-Ferrucio J, Lu YC, Zhang M,
Wang J, Xuei X, Adams M, Levine SS (2020) Roden C, Cheng J, Krause DS, Ding Y, Fan R,
Multisite evaluation of next-generation meth- Lu J (2019) Single-cell microRNA-mRNA
ods for small RNA quantification. J Biomol co-sequencing reveals non-genetic heterogene-
Tech 31(2):47–56. https://doi.org/10. ity and mechanisms of microRNA regulation.
7171/jbt.20-3102-001 Nat Commun 10(1):95. https://doi.org/10.
11. Alberti C, Manzenreither RA, Sowemimo I, 1038/s41467-018-07981-6
Burkard TR, Wang J, Mahofsky K, Ameres 16. Hucker SM, Fehlmann T, Werno C, Weidele K,
SL, Cochella L (2018) Cell-type specific Luke F, Schlenska-Lange A, Klein CA,
sequencing of microRNAs from complex ani- Keller A, Kirsch S (2021) Single-cell micro-
mal tissues. Nat Methods 15(4):283–289. RNA sequencing method comparison and
https://doi.org/10.1038/nmeth.4610 application to cell lines and circulating lung
12. Faridani OR, Abdullayev I, Hagemann-Jensen- tumor cells. Nat Commun 12(1):4316.
M, Schell JP, Lanner F, Sandberg R (2016) https://doi.org/10.1038/s41467-021-
Single-cell sequencing of the small-RNA tran- 24611-w
scriptome. Nat Biotechnol 34(12):
 Chapter 13

Immunoblot Analysis from Single Cells Using Milo™

Single-Cell Western Platform
Prashant V. Thakkar

Abstract
In this new era of precision medicine, characterization of single-cell subpopulations to better understand
disease etiology is paramount. It is thus an opportune time to explore techniques that allow molecular
analysis of single cells and to better understand the basis of pathogenesis of diseases like cancer. Single-cell
western blotting is one such method that allows analysis of single cells at the protein level. In contrast to
traditional western blotting, which relies heavily on bulk analysis of lysates generated from tissues and is
often indicative of the population average, this technique allows analysis of lysates from single-cell sub-
populations thereby providing a glimpse into cell heterogeneity. The method entails the use of a chip
containing 30 μm thick photoactivated polyacrylamide gel spotted with nearly 6400 microwells. Single cells
loaded on the chip are captured in the microwells by passive gravity and are then lysed and electrophoresed
using the MILO™ single-cell western platform. This method forgoes the use of transfer of proteins on a
PVDF and a nitrocellulose membrane, as performed in traditional western blotting, and all other steps
including probing of primary and fluorescent secondary antibodies against the protein of interest are
performed directly on the chip. The proteins of interest can then be visualized by scanning a chip with
the use of a microarray scanner. The entire procedure can be performed in as less as 4–6 h, and thus this
method provides several advantages over traditional western blotting.

Key words Single-cell western blot, Immunoblot, Precision medicine, Tumor heterogeneity, Cell
heterogeneity

1 Introduction

In the past decade, we have witnessed huge advances in biomedical

research, owing to innovations in genomic technologies, diagnos-
tics, and therapeutic biologics. The plethora of biological informa-
tion has given us important clues for understanding local and
global regulation at various levels of genetic and metabolic hierar-
chy, thereby boosting applied therapeutics [1–3]. These advances
have led to a new era of precision medicine, whereby diseases such
as cancer are no longer perceived as a single disease. The traditional
“one-size-fits-all” approach to cancer therapy included treating all
cancer patients with the same therapy regimen. It is now becoming

Miodrag Gužvić (ed.), Single Cell Analysis: Methods and Protocols, Methods in Molecular Biology, vol. 2752,
https://doi.org/10.1007/978-1-0716-3621-3_13, © Springer Science+Business Media, LLC, part of Springer Nature 2024

201
202 Prashant V. Thakkar

increasingly clear that the treatment of cancer needs to be custo-

mized at an individual level following molecular analysis at the level
of single cells [4].
Cellular heterogeneity is an intrinsic characteristic of all multi-
cellular organisms. Both normal and tumor tissues exhibit cellular
heterogeneity and different cells within the same tissue can differ in
function [5]. These functional differences in various cells are a
manifestation of variations in the DNA sequence and/or differ-
ences in the expression of RNA and protein. Elucidating these
molecular differences is critical to identifying various subpopula-
tions that may play an important role in disease etiology and
prognosis. Furthermore, diseases like cancer often show clonal
selection that further results in a subpopulation of cells that are
either resistant to therapy or show metastatic potential and are
responsible for disease progression [6]. Investigation of these cell
subpopulations is inherently limited by the fact that the readout of
any pooled assay that uses bulk tissue represents a weighted average
of that population’s cellular constituents [7]. Therefore, it is more
important than ever to employ methods that target molecular
analyses at a single-cell level. Recent technological innovations
now allow one to perform single-cell analysis at the level of DNA,
RNA, as well as proteins. Such techniques include capture of CTCs
from liquid biopsies with the use of microfluidic devices [6], RNA
seq followed by picking cells by Cell Celector [8], single-cell RNA
seq using Chromium 10× genomics platform [9], etc.
In the quest for further refining techniques that allow molecu-
lar analysis at a single-cell level, a new technique referred to as the
single-cell western blot/immunoblot has been developed. This
method, as the name suggests, allows analysis of proteins at a single
cell level using a modified method based on the traditional western
blotting of proteins. This method entails capture of single cells into
individual microwells present on a chip coated with photoactivated
polyacrylamide (scWest chips from ProteinSimple). These cells are
then lysed and electrophoresed directly on the chip using the
MILO™ single-cell western blot instrument (available from Pro-
teinSimple) [10]. The separated proteins of interest are then
probed for primary antibodies followed by fluorescent secondary
antibodies. The proteins of interest can then be visualized as fluo-
rescent spots on the chip as visualized using a microarray scanner.
The schematic for the single-cell western blot method-using
MILO™ single-cell western platform is presented in Fig. 1. This
chapter further discusses the details of the method and hopefully
serves as a reference for all new and existing users of this technique.
 Single-Cell Western Blot 203

Fig. 1 Schematic of single-cell western blot protocol. Steps a–e highlight different steps in the single-cell
western blot protocol. (a) The scWest chip is divided into 16 different grids (displayed over two rows and eight
columns). (b) Each grid contains 400 microwells thus giving us a total of 6400 microwells per chip. (c) The
chip is rehydrated with 1× Suspension buffer and a single-cell suspension (approximately 100,000 cells) is
loaded on the cells. (d) The cells are allowed to settle and nearly 30% of the total microwells are occupied with
cells. The excess number of cells is then washed off. (e) The cells in the microwells are then lysed and
electrophoresed in the same step using the MILO™ single-cell western platform. The chip is washed twice
with 1× wash buffer followed by one rinse with deionized water. (f) The chip is then probed with primary
antibodies against the protein of interest followed by fluorescent secondary antibody. (g) The chip is then
scanned using a microarray scanner and the resulting image can be analyzed using the Scout software.
(Figure adapted from [12])

2 Materials

2.1 Materials 10× Suspension Buffer.

Provided with 5× Wash Buffer.
Standard scWest Kit
Antibody Diluent 2.
[10]
Lysis Buffer.
scWest chips.

2.2 Other Required Clean 10 cm petri dishes.

Materials Aluminum foil/dark incubation boxes.
Antibody probing chamber (ProteinSimple).
204 Prashant V. Thakkar

MILO™ single-cell western blot instrument (ProteinSimple).

Metal forceps.
Deionized water/Molecular-grade water.
Primary antibodies.
Fluorescent Secondary antibodies.
Microarray scanner.
Scout software for data analysis (ProteinSimple).
Vortex.
Tabletop centrifuge.

3 Methods

3.1 Preparation of 1. The 10× suspension buffer and 5× wash buffer are provided
Buffers and Reagents with the standard scWest kit.
2. Prepare 1× suspension buffer and 1× wash buffer using the
above-mentioned 10× and 5× stock solutions respectively
using deionized water or molecular grade water [10].

3.2 Priming and 1. Take a scWest chip and place it in the center of a clean 10 cm
Rehydrating petri dish with the gel-side facing up. It is important that the
scWest Chips dish is clean and sterile since any dust particles that settle on the
chip can lead to a high background during the imaging steps.
Touching the gel surface of the chip must be avoided since that
can add to more dust particles on the chip.
2. Add 10 mL of 1× suspension buffer to the petri dish to
completely cover the chip.
3. Incubate the chip for at least 10 min, with slow shaking
(<100 rpm) on an orbital shaker before use, so that the chip
is completely rehydrated. Alternatively, the chip can also simply
be allowed to stand in 1× suspension buffer, although, it is
important to make sure that the chip is completely immersed.
4. The chip can be kept in the 1× suspension buffer until the
single-cell suspension is ready to be loaded on the chip.

3.3 Preparation of 1. A single-cell suspension of cell lines can be generated using

Single-Cell standard mammalian cell culture procedures.
Suspension 2. Briefly, remove the cell growth medium and gently wash the
dish with 1× PBS (without MgCl2 and CaCl2).
3. Remove 1× PBS and add 2 mL Trypsin solution in a T75 flask
(0.05% Trypsin, 0.53 mM EDTA, without sodium bicarbon-
ate) and incubate the flask in a 37 °C incubator for approxi-
mately 1–2 min. Incubation times and volume of trypsin can
vary depending on the cell lines and size of the flask.
 Single-Cell Western Blot 205

4. Gently tap the flask to detach all the cells and neutralize the
trypsin with at least equal volume of growth medium contain-
ing 10% FBS.
5. Count cells using hemocytometer or other cell counting equip-
ment and prepare a 5 mL suspension of 10,000 to 100,000
cells/mL in growth medium with 1× suspension buffer.
6. Centrifuge cells at 300 × g for 5 min. Remove the growth
medium and wash the cells once with 5 mL of 1× PBS (without
MgCl2 and CaCl2).
7. Centrifuge cells at 300 × g for 5 min. Remove 1× PBS solution
and resuspend cells in 5 mL of 1× suspension buffer. The
single-cell suspension is now ready for use on the single-cell
western blot.

3.4 Loading scWest 1. Remove the scWest chip from the petri dish containing 1×
Chips with Single-Cell suspension buffer and tilt it on one side to remove excess
Suspension liquid. A paper tissue can be placed underneath to help drain
off the excess liquid ensuring that paper tissue does not touch
the top of the gel. Place the chip in a new and clean 10 cm petri
dish. For selecting appropriate scWest chips, see Note 1.
2. Load 1 mL (approximately 100,000 cells) of the single-cell
suspension dropwise on the scWest chip [10]. Ensure that all
areas of the chip are fully covered (Fig. 1c). For using lower
number of cells, see Note 2.
3. Allow the cells to settle for 5–20 min. This step allows the cells
to enter and settle in the wells (Fig. 1d). For optimizing
appropriate settling time for different cells, see Note 3.
4. Once the cells are allowed to settle for the desired time, tilt the
slide to drain off the liquid and remove the uncaptured cells
from the top of the slide. To efficiently remove all the cells from
the top of the chip, tilt the petri dish containing the chip at a
45° angle and carefully run the 1× suspension buffer from top
to bottom using a 1000 uL pipette.
5. At this step, the chip can be viewed under a bright field micro-
scope to visualize settling of the cells in microwells. Immerse
the slide back into the 10 cm petri dish containing 1× suspen-
sion buffer. This chip is now ready to be run on MILO.

3.5 Running scWest 1. Turn on the MILO instrument and use the touch screen to
Chip Using MILO™ create optimized user settings. Most typically the conditions
Instrument listed below can serve as a good reference point [10]. Further
optimization may be required for determining optimal
conditions.
Lysis time: 10–15 s.
Electrophoresis run time: 60–90 s.
206 Prashant V. Thakkar

scWest chip

Electrophoresis Gel side up

Gradually lower the chip to
chamber allow the lysis buffer to
spread evenly on the bottom
surface of the chip

Trough

300 ml lysis buffer

loaded one on end the
trough

Fig. 2 Methodology to load the chip in the MILO™ instrument to ensure efficient lysis: 300 μL of lysis buffer is
loaded on the shorter end of the trough in the electrophoresis. Tilt the chip at a 45° angle and align the shorter
end of the chip to the shorter end of the trough where lysis buffer was added. Gradually lay the chip from the
other end so that the chip now lays flat and perfectly aligned in the trough. This will cause the lysis buffer to
spread throughout the bottom surface of the chip

Electrophoresis voltage: 240 V.

UV exposure time: 4 min.
In our experience, shorter lysis times lead to sharper and more
concentrated spots at the visualization step as opposed to more
diffused spots seen with longer lysis time.
2. Open the top lid of the MILO instrument and add 300 μL of
the lysis buffer provided in the Standard scWest kit along one of
the shorter ends of the trough of the white electrophoresis
chamber. Do not add more than 300 μL of the lysis buffer
since the excess buffer will overflow on top of the gel causing
lysis to begin immediately [10].
The lysis buffer provided with the scWest kits from Protein-
Simple does not contain reducing agents. Prior to performing
lysis, reducing agents can be added to the lysis buffer. For more
information regarding addition of reducing agents, see Note 4.
3. Tilt the chip at a 45° angle and align the shorter end of the chip
to the shorter end of the trough where lysis buffer was added in
step 2 above. Gradually lay the chip from the other end so that
the chip now lays flat and perfectly aligned in the trough. This
will cause the lysis buffer to spread throughout the bottom
surface of the chip (Fig. 2).
This step is important since any bubbles or gaps in the layer of
lysis buffer underneath the chip may result in inconsistent electro-
phoresis of the gel. Handle the chip by only touching it on the side
and not the top of the gel. Alternatively, forceps may also be used to
avoid physical contact with the gel.
4. Pour the rest of the lysis buffer into the chamber on top of the
chip, close the lid, and hit run immediately [10]. It is important
 Single-Cell Western Blot 207

to run the instrument immediately at this step since any delays

at this step can cause higher than desired incubation times.
5. Once the run is complete, remove the scWest chip from
MILO™ and perform two washes with 1× wash buffer with
gentle shaking on an orbital shaker for 10 min each.
6. At the end of the second wash, remove wash buffer and repeat
wash with 10 mL of deionized water with gentle shaking on an
orbital shaker for 10 min. This step removes the excess salt
from the wash buffer and thus will result in low background
during the imaging steps.
7. At this step, one can proceed with probing the chip with the
primary and secondary antibodies of interest. Alternatively, the
chip can be dried overnight and stored at room temperature for
months until one is prepared to probe the chip with primary
and secondary antibodies.

3.6 Probing scWest 1. Dilute the primary antibody to the final concentration of
Chips with Primary 0.2 μg/μL to 0.05 μg/μL using the “antibody diluent 2”
and Fluorescent solution provided in the kit [10].
Secondary Antibodies The protocol is designed to use as little as a total volume of
80 μL of the antibody at a dilution of 1:5 to 1:20, that is, for a 1:
20 dilution add 4 μL of the antibody with 76 μL of the “antibody
diluent 2” solution. Most commercially available antibodies
should work just fine within this concentration range. This con-
centration range can serve as a good reference point, although
further optimization may be required. Primary antibodies of two
different species can also be multiplexed in the same step. Thus, to
make an antibody solution of two different primary antibodies at
1:20 dilution each, add 4 μL of each of the antibodies with 72 μL
of the antibody diluent. Alternatively, the antibodies against two
different proteins of interest, even though from two different
species, can also be probed sequentially.
2. A special probing chamber (available through ProteinSimple)
[10] is used for antibody incubation. Pipette 80 μL of antibody
solution on one end of the probing chamber and place the chip
with gel side facing down so that the antibody solution spreads
evenly and covers the entire gel surface of the chip. To ensure
that the antibody covers the entire gel surface, the chip should
be held steadily on one end where the antibody solution was
loaded and gradually lowering the chip from the other end
(Fig. 3).
3. Avoid bubbles since no antibody signal will be detected in the
spot of the bubble. Adjust the chip to eliminate the bubbles
if any.
4. Incubate the chip with primary antibody at room temperature
for 1 h.
208 Prashant V. Thakkar

scWest chip

Gradually lower the chip to

allow the antibody solution to
spread evenly and thus cover
Gel side down the entire gel surface

Probing chamber

80 ml antibody
solution loaded on
one end the
probing chamber

Fig. 3 Methodology to load the chip on antibody probing chamber while probing
it with primary and fluorescent secondary antibodies: Pipette 80 μL of secondary
antibody solution on one end of the probing chamber and chip is placed with gel
side facing down so that the antibody solution spreads evenly and covers the
entire gel surface of the chip. The chip should be held steadily on one end where
the antibody solution was loaded and gradually lowering the chip from the
other end

5. At the end of 1-h incubation, carefully lift the chip from one
end while holding the other end so as not to tear the gel. Invert
it with the gel side facing up in a 10 cm petri dish.
6. Wash it three times with 1× wash buffer, while gently shaking
on an orbital shaker for 10 min each.
7. At the end of the third wash, prepare the fluorescent secondary
antibody solution using the “Antibody diluent” at 1:20 to 1:40
dilution [10].
Similar to primary antibody incubation, secondary antibo-
dies against primary antibodies from two different species can be
multiplexed. It is important that the two secondary antibodies
against the two different primary antibodies fluoresce in differ-
ent spectral channels to ensure specificity of the signal.
8. For secondary antibody incubation, pipette 80 μL of secondary
antibody solution on one end of the probing chamber and
place the chip with gel side facing down so that the antibody
solution spreads evenly and covers the entire gel surface of the
chip. To ensure that the antibody covers the entire gel surface,
the chip should be lowered slowly keeping one end steadily
rested on the probing chamber and gradually lowering the chip
from the other end (Fig. 3).
9. Incubate the secondary antibody solution for 1 h, this time
covering it with either a black box or aluminum foil so as to
protect it from light.
 Single-Cell Western Blot 209

10. At the end of the 1-h incubation, carefully lift the chip from
one end while holding the other end so as not to tear the gel.
Invert it to with the gel side facing up in a 10 cm petri dish.
11. Wash it three times with 1× wash buffer, while gently shaking
on an orbital shaker for 15 min each this time also protecting
from light [10].
12. After the last wash, proceed to incubating primary antibody for
the next protein interest. If all the proteins of interest have
been probed with their respective primary and secondary anti-
bodies, air dry the chip in the dark for about 40–45 min.
13. The chip is now ready for imaging.

3.7 Imaging of 1. scWest chip can be scanned using many of the recommended
scWest Chips fluorescence microarray scanners recommended by Protein-
Simple [10]. In case where the microarray scanner is not avail-
able, see Note 5 for the use of Zeiss spinning disc microscope
for scanning images.
2. We have personally used GenePix4000B from Molecular
Devices [11], which in our experience is very user-friendly to
use. However, this instrument can only allow for scanning of
up to two different fluorescent channels (excitation wave-
lengths of 532 nm and 635 nm with emission wavelengths in
greenish yellow to red/far red regions of the spectrum)
whereas other microarray scanners can allow up to four or
even five different channels thus allowing for probing of more
proteins per chip.
3. To start scanning the slide using the GenePix4000B, open the
door and lift the slide holder. Push the slide clamp to place the
slide in its position with gel side facing down. Lower the slide
holder back to its original position and close the door.
4. Close the instrument door and start the “Genepix pro image
analysis software” on the computer attached to the instrument.
5. Once the software interface is open, click on “Hardware set-
tings” to select the wavelength the chip would be scanned
at. For each wavelength, set the scan pixel resolution to 5 μm
and adjust the PMT gain to maximize the brightness of the
signal without saturating the detector. Change the power of
each laser to 100%.
6. Once all the settings are set, one can hit preview scan to get a
quick preview of image the scan would generate. If the preview
scan looks good, one can then hit “Data Scan” for scanning the
chip using both wavelengths of “Single wavelength scan”
option to scan the chip only using one wavelength.
7. The instrument allows simultaneous scans at two wavelengths
and scans the standard scWest chip (approximately
210 Prashant V. Thakkar

25 mm × 75 mm) at a resolution of 5 μm in approximately

12 min. The resulting image is saved as a 16-bit TIFF file,
which can be easily exported into the Scout software for further
analysis.

3.8 Data Analysis 1. The images generated using microarray scanners for any given
Using Scout Software chip are saved as separate file per channel. These images are
saved in .tiff format, and this format can be easily used for
analysis using Scout software. Export/Add all the images for
a given chip from “File” menu in the Scout software [10].
2. The software requires that each exported image be registered.
To do this, select the first block of the first row and the eight
block in the second row (i.e., two diagonally opposite grids).
Then click on the first well in the first block and the last well in
the last block. This step helps ensure that the image alignment
is correct and any misalignment can be corrected [10].
3. Following step 2, the software will then automatically detect
the lanes with a positive signal/peaks. This detection can be
then manually inspected, and these parameters can be adjusted
to define the true signal under “Scan properties” table [10].
4. Once all the peaks are detected, go to the peak table to identify
and name the peak of interest as the appropriate molecular
weight size peak.
5. The software can then provide statistics such as total number of
microwells occupied, total number of wells that are positive in
each spectral channel, etc. A representative image of the results
obtained from this single-cell western blot method is presented
in Fig. 4.

3.9 Limitations Although this technique shows incredible promise and is huge
advancement in allowing molecular analysis of proteins at the level
of single cells, it is not without limitations. Some of these limita-
tions are highlighted below:
1. The method currently does not incorporate the use of molecu-
lar weight markers that can be run simultaneously with the
sample. This limitation then requires one to probe for one of
the housekeeping genes such as Tubulin or GAPDH to be used
as a size marker. One can then predict the size of the protein of
interest based on the size and migration of the housekeeping
gene. In addition, since one has to always probe for a
housekeeping gene, it expends one of the fluorescent channels,
leaving one less channel that can be used for probing an addi-
tional protein of interest.
2. Although migration of protein occurs over a short distance of
900 μm, and while the migration of proteins of different
molecular sizes follows a linear range in that short distance,
 Single-Cell Western Blot 211

Fig. 4 Representative image of the result obtained from a single-cell western

blot: Single-cell western blot protocol was performed using the Standard scWest
chip and was probed for two antibodies against protein A and protein B. Two
isoforms that differ in size by 30 kDa can be visualized in the top panel and the
housekeeping gene (Protein B) can be visualized in the bottom panel

deducing molecular weights of proteins of interest in reference

to the housekeeping gene can sometimes still be inconclusive.
In particular, some of the proteins migrate as dimers and multi-
mers since the migration of proteins occurs under
non-reducing, non-denaturing conditions. For example: Pro-
teins such as cytokeratin and GAPDH, which would normally
migrate as 67 kDa and 38 kDa monomers respectively on a
western blot, these proteins migrate as 130 kDa dimer and
approximately.
3. Currently, the system only allows for analysis of proteins in the
molecular weight size range from 175 kDa to 15 kDa. Hence,
proteins higher than 175 kDa cannot be studied using this
system and will need to be studied by traditional western
blotting. Furthermore, the scWest chips available are only
available in single gel composition of 8% polyacrylamide. As
of now, no gradient gels are available, and hence proteins with
molecular sizes at the extreme ends of the above-mentioned
range can only be resolved with short run times, which then
would compromise on the resolution of proteins in the same
spectral channel. Even while resolving proteins similar in size,
the system requires that the proteins differ in size by at least
10% if viewed in different spectral channels and at least 30%
when viewed in the same spectral channel. For example: two
isoforms of the same protein, and thus viewed in the same
spectral channel, cannot be resolved on the same scWest chip
212 Prashant V. Thakkar

unless they differ from each other in molecular weight size by

at least 30%.

4 Notes

1. Choosing appropriate scWest Chips for the cells of interest

There are three different types of scWest chips available
from ProteinSimple, namely (1) Small scWest kit for small
cells, (2) Standard scWest kit, and (3) Large scWest kit for
large cells. These kits differ in the diameter of microwells, and
hence appropriate kit must be chosen depending on the diam-
eter of the cells of interest. Small scWest kit can be used for cells
with diameter less than 10 μm, standard kits can be used for
cells with diameter in the range of 10–20 μm, and large kits can
be used for cells with 20–30 μm [10].
2. Using the appropriate number of cells
The scWest chips are divided into a total of 16 grids over
two rows and eight columns. Each block has 400 microwells
divided over 10 rows of 40 microwells each. This layout gives a
total of 6400 total microwells per scWest chip. When cells are
loaded on the chip, the cells settle into the microwells by
passive gravity. This process follows a Gaussian distribution
where only 20–30% of the microwells are occupied, with nearly
10–15% of the occupied microwells contain single cells and the
remaining occupied microwells contain two or three cells, that
is, doublets or triplets [10]. This percentage occupancy is even
lower with lower number of cells. In situations where the
starting material is not a limiting factor, for example, cell
lines, 100,000 cells should be loaded on the chip. The recom-
mended minimum number of cells to be loaded on the chip is
10,000 cells in 1 mL volume over the entire chip. We have tried
loading as little as 1000 cells and were able to capture only
15–20 cells on the chip when loaded in 1 mL of suspension
buffer. Alternatively, loading a lower number of cells in as little
as 100–200 μL of suspension buffer can help improve the
capture efficiency of cells. This is especially important in cases
where a specific subset of cells, isolated using sorting by a flow
cytometer or concentrated by means of specific cell surface
markers, needs to be analyzed, and thus the starting material
is limited.
3. Optimizing settling time for the cells on scWest chip
Since settling of the cells into microwells is driven by
passive gravity, 10 min incubation leads to capture of 0–4
cells/microwells and occupancy of approximately 30% of the
microwells. Longer incubation times result in capture of higher
number of cells but also lead to a higher percentage of
 Single-Cell Western Blot 213

microwells with more than one cell per microwell, that is,
doublets and triplets per microwell [10].
4. Use of reducing agents to completely denature proteins
The lysis buffer provided in the kit currently does not
contain any reducing agents. Furthermore, since the lysis step
is performed directly on the chip once cells are loaded in the
microwells, there is no boiling step, which may result in incom-
plete denaturation of proteins as well as incomplete dissocia-
tion of oligomers. Reducing agents can be easily incorporated
in the protocol by using 1× SDS buffer for lysis of cells in the
microwells. Although this buffer is highly ionic, the conduc-
tance of 1× SDS buffer is higher compared to the lysis buffer
provided in the kit. Hence, the voltage should be reduced from
240 V to 150 V and run time should be increased by approxi-
mately 50–60%. Alternatively, 10 mM TCEP (tris-(2-carbox-
yethyl)phosphine) (pH 7.0) can be added to the lysis buffer
provided to the kit. This is important since the increased con-
ductance by the lysis buffer can lead to heating up of the gel,
which may in turn affect the resolution.
5. Use of Zeiss spinning disc microscope for scanning scWest chips
In cases where microarray scanners are hard to find or are
not available in one’s institution, we have also successfully
employed Zeiss spinning disc microscope where one can scan
the entire chip in “tiles” and then “stitch” all the images to
create a single image of the entire chip. This image can then be
exported to the Scout software for further analysis.

References

1. Ashley EA, Butte AJ, Wheeler MT, Chen R, Bhattacharjee A, Eichler EE, Bamshad M,
Klein TE, Dewey FE, Dudley JT, Ormond Nickerson DA, Shendure J (2009) Targeted
KE, Pavlovic A, Morgan AA, Pushkarev D, capture and massively parallel sequencing of
Neff NF, Hudgins L, Gong L, Hodges LM, 12 human exomes. Nature 461(7261):
Berlin DS, Thorn CF, Sangkuhl K, Hebert JM, 2 7 2 – 2 7 6 . h t t p s : // d o i . o r g / 1 0 . 1 0 3 8 /
Woon M, Sagreiya H, Whaley R, Knowles JW, nature08250
Chou MF, Thakuria JV, Rosenbaum AM, Zar- 4. Ashley EA (2016) Towards precision medicine.
anek AW, Church GM, Greely HT, Quake SR, Nature reviews. Genetics 17(9):507–522.
Altman RB (2010) Clinical assessment incor- https://doi.org/10.1038/nrg.2016.86
porating a personal genome. Lancet 5. McGranahan N, Swanton C (2017) Clonal
375(9725):1525–1535. https://doi.org/10. heterogeneity and tumor evolution: past, pres-
1016/S0140-6736(10)60452-7 ent, and the future. Cell 168(4):613–628.
2. Choi M, Scholl UI, Ji W, Liu T, Tikhonova IR, https://doi.org/10.1016/j.cell.2017.01.018
Zumbo P, Nayir A, Bakkaloglu A, Ozen S, 6. Ignatiadis M, Dawson SJ (2014) Circulating
Sanjad S, Nelson-Williams C, Farhi A, tumor cells and circulating tumor DNA for
Mane S, Lifton RP (2009) Genetic diagnosis precision medicine: dream or reality? Ann
by whole exome capture and massively parallel Oncol 25(12):2304–2313. https://doi.org/
DNA sequencing. Proc Natl Acad Sci U S A 10.1093/annonc/mdu480
106(45):19096–19101. https://doi.org/10.
1073/pnas.0910672106 7. Yuan GC, Cai L, Elowitz M, Enver T, Fan G,
Guo G, Irizarry R, Kharchenko P, Kim J,
3. Ng SB, Turner EH, Robertson PD, Flygare Orkin S, Quackenbush J, Saadatpour A,
SD, Bigham AW, Lee C, Shaffer T, Wong M, Schroeder T, Shivdasani R, Tirosh I (2017)
214 Prashant V. Thakkar

Challenges and emerging directions in single- 10. MILO™ user guide from ProteinSimple.
cell analysis. Genome Biol 18(1):84. https:// https://www.proteinsimple.com/documents/
doi.org/10.1186/s13059-017-1218-y Milo_User_Guide_R5.pdf
8. Zhu S, Qing T, Zheng Y, Jin L, Shi L (2017) 11. Genepix4000B Microarray scanner user guide.
Advances in single-cell RNA sequencing and its http://mdc.custhelp.com/euf/assets/con
applications in cancer research. Oncotarget tent/GenePix_4000B_UserGuide.pdf
8(32):53763–53779. https://doi.org/10. 12. Hughes AJ, Spelke DP, Xu Z, Kang CC, Schaf-
18632/oncotarget.17893 fer DV, Herr AE (2014) Single-cell western
9. Haque A, Engel J, Teichmann SA, Lonnberg T blotting. Nat Methods 11(7):749–755.
(2017) A practical guide to single-cell RNA-- https://doi.org/10.1038/nmeth.2992
sequencing for biomedical research and clinical
applications. Genome Med 9(1):75. https://
doi.org/10.1186/s13073-017-0467-4
 Chapter 14

Imaging of Subcellular Distribution of Platinum in Single

Cells Using Laser Ablation Inductively Coupled Plasma Mass
Spectrometry
Amy J. Managh and Calum J. Greenhalgh

Abstract
Laser ablation inductively coupled plasma-mass spectrometry (LA-ICP-MS) is a well-established and
sensitive analytical technique, which provides high-resolution imaging of endogenous elements, element
tagged-markers, metal-containing nanoparticles, and metallodrugs within cells. Here we describe a proto-
col for imaging the subcellular distribution of platinum within A549 cells, following their incubation with
the platinum-based anticancer agent, Oxaliplatin. We outline the essential steps in sample preparation and
instrumental setup and discuss how the current generation of low-dispersion instruments facilitates new
approaches to data acquisition and image processing. The protocol described herein can be easily adapted
for other cell lines and metal-containing labeling agents.

Key words Laser ablation, ICP-MS, Imaging, Oxaliplatin, A549 cells

1 Introduction

Laser ablation inductively coupled plasma-mass spectrometry

(LA-ICP-MS) is a sensitive elemental analysis technique, which
enables highly specific sampling and imaging of solid materials
with micrometer scale resolution. Although traditionally used for
geological analysis, the technique is now well-established for a
range of biological applications. Examples include visualization of
drugs and tagged markers in tissue sections [1, 2], as well as
quantitative measurements of endogenous elements and isotope
ratio distributions [3, 4]. Due to its high spatial resolution, the
technique is particularly suited to cellular and subcellular imaging.
The first use of LA-ICP-MS for subcellular histology was reported
in 2011 [5]. Geisen and co-workers used a 4 μm diameter laser
beam to differentiate the nucleus and cytoplasmic regions of iodi-
nated fibroblast cells. Subsequent studies have reported the use of
lanthanide-tagged complexes and iridium intercalators to visualize

Miodrag Gužvić (ed.), Single Cell Analysis: Methods and Protocols, Methods in Molecular Biology, vol. 2752,
https://doi.org/10.1007/978-1-0716-3621-3_14, © Springer Science+Business Media, LLC, part of Springer Nature 2024

215
216 Amy J. Managh and Calum J. Greenhalgh

the protein distribution, DNA content, cell volume, and cell cycle
phase of individual cells [6, 7]. LA-ICP-MS has also been used to
determine nanoparticle and drug uptake by cells. Studies have
reported both quantitative determination of Au nanoparticle
uptake by whole cells [8] and subcellular imaging of their localiza-
tion to different cellular compartments [9]. Furthermore, the tech-
nique may be vital for monitoring the fate of rare cell populations
used in cell-based treatments. In a 2013 study, individual regu-
latory T cells that were labeled with the MRI contrast agent Omnis-
can were successfully identified in mouse peritoneal lavage samples
at 10 days post-administration [10]. Subsequent publications have
also reported LA-ICP-MS identification of individual therapeutic
cells within ex vivo tissue sections in similar cell tracking experi-
ments [11, 12].
Despite the success of the above applications, LA-ICP-MS has
not yet gained full traction within the molecular biology commu-
nity. This is mainly due to inefficiencies associated with the previous
generation of instruments, in particular in the interface between the
LA and ICP-MS units, which resulted in broadening of signals and
hence lengthy analysis times compared to conventional imaging
methods. In recent years a number of new interfaces have been
developed to combat this problem [13–15]. These new technolo-
gies provide vastly improved aerosol transport characteristics, lead-
ing to faster signal responses, higher absolute sensitivity, and
improved signal-to-noise ratios. Importantly, these new develop-
ments are now starting to reach the market. In this protocol we
describe a method for qualitative single-cell analysis and imaging of
platinum distribution in Oxaliplatin spiked A549 cells, using one of
these new laser ablation platforms. We focus on how the improved
aerosol transport characteristics have created an opportunity to
move away from the traditional “rastered” method of imaging,
toward an approach that preserves maximum image resolution for
individual cell analysis.

2 Materials

2.1 Equipment 1. Element XR sector-field ICP-MS (Thermo Fisher Scientific).

2. NWR image laser ablation system (Elemental Scientific Lasers).
3. Hettich Universal 320R centrifuge, fitted with 4 place rotor
with fixed angle chambers.
4. Laminar flow cabinet.
5. Microscope.
6. Hemocytometer.
 LA-ICP-MS of Single Cells 217

2.2 Reagents 1. A549 cell line.

2. RPMI-1640 medium, containing L-glutamine and sodium
bicarbonate, supplemented with 10% Fetal Bovine Serum
(FBS).
3. Dulbecco’s modified phosphate-buffered saline (PBS).
4. Trypsin/EDTA.
5. Trypan blue.
6. High purity (99.996%) helium and argon supply.

2.3 Other 1. T-25 cell culture flasks.

Consumables 2. Glass microscope slides (see Note 1).
3. Angle cyto chamber (DJB Labcare).
4. Filter cards for 1, 2, and 4 mL Cyto chambers (DJB Labcare).
5. Conical tubes (15 mL; polypropylene or polystyrene, e.g., from
Falcon).

3 Methods

3.1 Preparation of Cell culture and labeling conditions will vary depending on the
Cells for Analysis nature of the cell line being studied. The instructions below are
based on the culture of A549 cells, which are an adherent human
lung adenocarcinoma cell line, and their labeling with the antican-
cer drug, Oxaliplatin. For other cell lines and labeling approaches,
you should follow your usual procedures for that cell line, and then
proceed to step 9.
1. Seed cells at an initial density of 1 × 106 cells in 5 mL of
complete media. Incubate at 37 °C in 5% CO2 overnight.
2. Add 50 μM Oxaliplatin to the culture media and leave to
incubate for a further 24 h.
3. Aspirate the used media. Add 5 mL PBS to the flask (see Note
2) and gently swirl for 1 min ensuring the entire surface is
covered. Aspirate the PBS into a waste container. Repeat 2–4
times (see Note 3).
4. Add 1 mL of warm trypsin-EDTA solution to the flask and
incubate at 37 °C for 3 min.
5. Inspect the flask under a microscope to check that the cells have
successfully detached from the surface (see Note 4), then add
4 mL of complete media to quench the trypsin.
6. Set aside a 100 μL aliquot of the cell suspension for cell
counting.
7. Transfer the remaining cells into a 15 mL falcon tube and
centrifuge at 400 rpm for 5 min.
218 Amy J. Managh and Calum J. Greenhalgh

8. Mix the 100 μL aliquot of the cell suspension with an equal

volume of 0.4% trypan blue solution. Dispense 10 μL of this
solution into a hemocytometer and count the cells (see Note
5).
9. Remove the supernatant from the centrifuged cells (see Note 6)
and dilute the cell pellet in phosphate-buffered saline to a
concentration of 100,000 cells/mL. Mix the solution well.
10. Disperse the cells onto a glass microscope slide using a Cyto-
centrifuge (see Note 7). Place a filter card onto a clean glass
microscope slide, then clamp an angled cyto chamber in place
over the filter card, using the locking clip provided. Place the
cyto chamber into the fixed-angled chambers in the centrifuge,
with the funnel inlet facing upwards. Transfer 0.5 mL of the
diluted cells to the funnel. Ensure that the centrifuge is bal-
anced, then centrifuge at 35 rcf for 3 min to disperse the cells
across the slide. Allow the slide to air-dry in a dust-free envi-
ronment for half an hour before performing the analysis. If
necessary the slides can be stored for several weeks in a cool,
dry, and dust-free environment.

3.2 Setting Up the 1. Insert the slide into the laser ablation chamber and secure it in
LA-ICP-MS System place (see Note 8). Ensure that the chamber is appropriately
sealed afterwards by fastening the screws on the outside of the
chamber (finger-tight).
2. Connect the inlet of the laser ablation chamber to a helium
supply and connect the outlet of the chamber to the ICP-MS
torch, using suitable tubing, e.g., Tygon. Allow for the intro-
duction of an argon make-up flow between the ablation cham-
ber and the ICP torch (see Note 9).
3. If available, connect a trigger cable between SyncOut on the
laser ablation system and the external trigger in port on the
mass spectrometer. This is to synchronize the firing of the laser
with the acquisition cycle of the mass spectrometer.
4. Purge the chamber with helium for around 10 min by selecting
the purge button in the laser software. While the instrument is
purging it is useful to move the inner cup around the chamber
to aid removal of residual air from the corners of the chamber.
This can be done using the x,y controls.
5. Light the plasma (see Note 10). Once the plasma has lit set the
helium flow through the laser ablation chamber to around
1.4 L/min. It is useful to do this at a low ramp rate of around
20 mL/min in order to avoid extinguishing the plasma. Allow
the plasma conditions to stabilize for approximately an hour
before performing any analysis.
6. A sample to cup distance of 0.2–0.3 mm is optimal for efficient
material transport out of the chamber. To set this use the z
 LA-ICP-MS of Single Cells 219

Fig. 1 Left image: Microscopic image showing A549 cells that were dosed with 50 μM Oxaliplatin. The
Cytospun cells are spaced an adequate distance for single-cell targeting. Right: LA-ICP-MS signal for the
analysis of the Pt in the cell on the far left. Data points in raw counts (shown as horizontal lines) are spaced at
1 ms intervals allowing the full peak profile to be visualized

controls within the laser software to focus on the lower inner

edge of the sample cup, and then focus down a further
0.2–0.3 mm. Use the adjustment screws on the sample drawer
to bring the glass slide up into focus (see Note 11).
7. Use the x,y controls to move to the location of interest on the
slide. You may need to refocus in the z direction afterwards if
the slide is not perfectly level. When in focus the cells should
appear similar in nature to the image in Fig. 1. If necessary,
re-center the cup so that the crosshairs on the screen are
positioned in the center of the aperture in the cup.
8. Warm up the laser before performing any analysis by firing the
laser for a few minutes at a high energy and repetition rate, with
the shutter closed. The Emission button can also be used to
reinitialize the laser after periods of downtime.
9. If the sample chamber needs to be opened while the plasma is
lit, use the bypass button to avoid entrainment of air. Once
closed, the chamber will need to be re-purged before proceed-
ing to online mode (see Note 12).

3.3 LA-ICP-MS Here, analyzing the platinum content of whole cells is used to
Analysis of Pt Content qualitatively assess the heterogeneity in Oxaliplatin uptake across
in Whole Cells the cell population.
1. In the LA software select the spot tool, and then left-click on
the center of one of the cells of interest.
2. Right-click the above spot and amend its properties. To ensure
the entire cell is ablated, set a spot size that is at least 5–10 μm
larger than the diameter of cells of interest (for this analysis a
spot size of 60 μm was used). Set the repetition frequency as
220 Amy J. Managh and Calum J. Greenhalgh

1 Hz. Select a laser energy that is sufficient to ablate through

the entire depth of the cell, while minimizing ablation of the
glass substrate (in this work a laser energy of ~5 J/cm2, was
used—see Note 13). Check the box to set these parameters as
default.
3. Click on the center of the remaining cells of interest. To avoid
accidental ablation of neighboring cells, only select cells that
are separated by a distance equivalent to at least the diameter of
the laser beam.
4. In the LA software, set a laser warm-up time of 1 s. In the Run
Experiment window, select All Patterns and click the check-
boxes Manually Control Valves and Enable Laser During Scans.
In the laser SyncOut options, set the trigger to Active During
Pattern Scan.
5. Ensure that the ICP-MS is in speed or time-resolved mode and
set up an appropriate ICP-MS method and sequence. First,
select the mass values to be detected (in this study the 195Pt
isotope was measured). Ensure that the segment durations of
the method (i.e., the data intervals of the output signal) allow
more than one data point per ablation event (see Note 14).
Determine the analysis time for the selected patterns and
ensure that the length of the ICP-MS method is equivalent to
this duration, plus an extra 5–10 s.
6. Click Run in the ICP-MS software and wait until the “awaiting
external trigger” message appears in the sequence window.
7. Click Run in the laser software.
8. Export the data as a time-resolved data file. Plot the graphs in
Excel, then find and integrate the area under each of the peaks
(see Note 15).

3.4 LA-ICP-MS Here, subcellular imaging is used to qualitatively assess the distri-
Imaging of Pt bution of platinum within individual cells, in order to assess
Distribution Across whether the drug is reaching its target site of action (binding to
Cells DNA in the nucleus).
1. In the LA software, use the line tool to draw a single horizontal
or vertical line that is long enough to span the region of interest
at its widest point. If necessary, this line can be subsequently
resized or repositioned to the edge of the region of interest,
using the Edit Endpoints function.
2. Right-click the above line and amend its properties. Choose a
laser beam diameter that is smaller than the dimensions of the
cells of interest (in this work a 2 μm spot size was used). Set the
highest possible repetition frequency that enables baseline res-
olution between the peaks from consecutive ablation events
(in this work 20 Hz was used). Select a scan rate that is
 LA-ICP-MS of Single Cells 221

equivalent to the spot size multiplied by the repetition fre-

quency of the laser (e.g., for the settings above, a scan speed
of 40 μm/s was selected). These settings avoid “over-sam-
pling” and preserve image resolution by ensuring that the
data from adjacent, non-overlapping locations is resolved in
the time-domain. Finally, set a laser energy that is sufficient to
ablate through the entire depth of the cell, while minimizing
ablation of the glass substrate (in this work a laser energy of
~4 J/cm2 was used—see Note 13).
3. Duplicate the above line an appropriate number of times to
cover the region of interest. Ensure that these duplicate lines
are offset by a space equivalent to the size of the laser beam.
4. Ensure that the ICP-MS is in speed or time-resolved mode and
set up an appropriate ICP-MS method. First, select the mass
values to be detected (in this study the 195Pt isotope was
measured). Ensure that the segment durations of the method
(i.e., the data intervals of the output signal) are compatible with
the repetition rate of the laser (see Note 14). Determine the
analysis time for a single ablated line (the length of the line
divided by the laser scan speed). The length of the ICP-MS
method should be equal to this time, plus an extra 1–2 s.
5. Set up a sequence in the ICP-MS software. Ensure that the
number of samples in the ICP-MS sequence is equal to the
number of lines in the laser method. Set the method to trigger
the acquisition when a signal from the laser system is received.
In the Element XR software the relevant settings file for this
(ext_trigger.inl) can be found under Sampling in the Sequence
window. Other instruments will have a similar option.
6. In the LA software, set a laser warm-up time of 5 s and a
washout time of 5 s. The washout time is a buffer to ensure
that the ICP-MS is waiting to receive the trigger signal when
ablation of the next line begins. In the Run Experiment win-
dow, select All Patterns and click the checkboxes Manually
Control Valves and Enable Laser During Scans. In the laser
SyncOut options, set the trigger to Active During Pattern
Scan (see Note 16).
7. Click Run in the ICP-MS software and wait until the “awaiting
external trigger” message appears in the sequence window.
8. Click Run in the laser software. It is advisable to stay with the
instrument for the first few ablated lines in order to check that
the ICP-MS is triggering correctly.

3.5 Generation of LA- 1. The data output from the imaging experiment will consist of a
ICP-MS Images series of files, each file containing the data for one ablated line.
Assuming sufficient instrument sensitivity, the data will consist
of peaks separated by background (or close to background)
222 Amy J. Managh and Calum J. Greenhalgh

Fig. 2 Schematic depiction of the data processing approach required for a pixel-by-pixel imaging experiment.
The instrumental conditions are set such that every laser shot samples an adjacent, non-overlapping area and
the signal is allowed to wash out before the next location is sampled. In this example, data points in raw
counts (shown as horizontal lines) are spaced at 10 ms intervals in the ICP-MS method, allowing 5 data points
to be collected per laser shot (laser at 20 Hz). Markers are then placed between each laser shot and the data
between the markers is summed to give a value for each pixel in the final image

Fig. 3 Left image: Microscopic image of a clump of A549 cells, which were dosed with 50 μM Oxaliplatin.
Right: LA-ICP-MS image showing the distribution of Pt across the cells. Cells were imaged using a 2 μm laser
beam, with a repetition frequency of 20 Hz and a scan speed of 40 μm/s. It appears that Pt is preferentially
accumulated in the nucleus of these cells

signal. An example of the typical output signal is shown in

Fig. 2. First, check the data in the acquired files to get a feel
for the magnitude of the signals in your data.
2. Sum the signal for each ablated shot, as shown in Fig. 2, to get a
list of pixel intensities for each ablated line. Compile these
values into a 2D matrix, such that the columns in the matrix
 LA-ICP-MS of Single Cells 223

correspond to the ablated lines (placed in order of ablation).

This processing can be manually performed within Excel
(extremely time consuming) or automated using a custom
Excel macro or Matlab script. Alternatively we use a custom
app, the LA-ICP-MS Image Tool, to find the peaks in our LA-
ICP-MS data, position markers between the peaks and sum the
signals between the markers. The app then generates a preview
of the final image as a 2D color map (see Fig. 3). The app is
freely available, has a detailed instruction manual [16], and can
process the raw data into images in less than 5 min.
3. Finally, if higher quality images are required, you may wish to
export the data to a graphics package for further processing.

4 Notes

1. We recommend Polysine-coated microscope slides (Thermo

Fisher Scientific, UK). The surface coating minimizes back-
ground signal by providing a barrier between the sample and
glass surface.
2. Take care not to touch the cell surface. We recommend adding
the liquid down the side of the flask, before gently tipping it
onto the cells.
3. Washing the cells several times is required to remove unbound
label from the surface of the cells. A higher number of washes
may be required depending on the nature of the labeling agent
and the concentration of it used.
4. Cells should appear round when detached. Tap the flask to
dislodge any unbound cells. If the cells have not been trypsi-
nized, return the flask to the incubator for a further 1 min.
5. The initial cell concentration per mL can be calculated by
multiplying the average viable cell count per square by the
trypan blue dilution factor (2) and then dividing by the volume
(in mL) of the square.
6. Check that a pellet has formed at the bottom of the tube. Take
care not to disturb the cell pellet when removing the
supernatant.
7. Centrifuging the cells disperses the cells across the slide and
helps to remove traces of PBS and culture media, which are
absorbed into a filter card. The aim of this step is to obtain
sufficient spatial separation between cells to enable single-cell
targeting (Fig. 1).
8. It is advisable to fix the slide securely in place inside the cham-
ber to prevent it from moving during the analysis. The NWR
image chamber is equipped with a suitable locking clip.
224 Amy J. Managh and Calum J. Greenhalgh

However, for chambers without this, a piece of double-sided

sticky tape placed on the back of the slide is an effective substi-
tute to temporarily fix the slide in place. Before fastening down
the slide we find that drawing the outline of the cell containing
region on the rear side of the slide is helpful in order to help us
to move quickly to the location of the cells once the slide is in
the chamber.
9. The argon make-up flow can be introduced through a simple
T-piece located in the transfer line between the ablation cham-
ber and the ICP torch. However, we prefer to use a Dual
Concentric Injector (Elemental Scientific Lasers, USA) to min-
imize aerosol dispersion and provide improved sensitivity.
Regardless of the mechanism of argon introduction, minimiz-
ing the total path length between the ablation chamber and the
ICP torch is recommended.
10. ICP-MS parameters will vary depending on the ICP-MS
instrument used. We find that the optimum Ar sample gas
flow is usually much lower with the NWR image system, com-
pared to our normal ICP-MS operation. A sample gas flow of
around 0.5 L/min and an RF power of 1100 W are usually
optimum for our instrument.
11. If your sample tray does not enable sample height adjustment,
skip setting the sample-to-cup distance and use the z controls
to focus on the sample instead.
12. The Dual Concentric Injector bypasses the valve system on the
laser unit. If using the DCI, it is advisable to clamp the trans-
port tubing with a temporary clip if opening the chamber or
purging the system while the plasma is lit.
13. The degree of laser coupling with the glass slide will vary
depending on the wavelength of the laser system used. We
have found that the 266 nm wavelength is efficient at ablating
through biological material, but produces negligible ablation
of the glass at the energies described above. For other common
wavelengths, e.g., 213 nm and 193 nm, the energy may need
to be tuned more extensively due to more efficient coupling of
the deep-UV wavelengths with the glass matrix.
14. Ideally the temporal resolution of the ICP-MS method will be
such that there is more than one data point per cell event. For
example in the peak in Fig. 1 (right hand image) data points are
spaced at 1 ms intervals resulting in around 50 data points over
the ~50 ms event. Note that the data acquisition speed of some
of the older ICP-MS systems is not always sufficient to enable
the required resolution for high-speed LA signals. This issue
has largely been rectified in the current generation instru-
ments, but plug-in detector boards are a solution to improve
the temporal resolution of older systems [17].
 LA-ICP-MS of Single Cells 225

15. The peaks can be manually located within the Excel data or this
can be automated through the use of in-house Excel macros or
Matlab scripts. We use a custom app, the LA-ICP-MS Image
Tool, to find the peaks in our LA-ICP-MS data [16]. Once the
peak areas have been found, the %RSD of the data can be
calculated to provide information on the variability in label
uptake across the cell population. We find that plotting a
histogram of the integrated signals is a useful way of visually
displaying this information in publications and conference pre-
sentations. For an example histogram, see reference [10].
16. Some modern systems have updates that allow the option of
sending a trigger pulse before every laser shot. This approach
may simplify your data processing. It is advisable to speak to
your relevant LA and ICP-MS service engineers to discuss
whether this is a possible option for your instrumentation.

References
1. Giesen C, Wang HA, Schapiro D, Zivanovic N, ablation inductively coupled plasma mass spec-
Jacobs A, Hattendorf B, Schuffler PJ, trometry. J Anal Atom Spectrom 26(11):
Grolimund D, Buhmann JM, Brandt S, 2160–2165. https://doi.org/10.1039/
Varga Z, Wild PJ, Gunther D, Bodenmiller B c1ja10227c
(2014) Highly multiplexed imaging of tumor 6. Herrmann AJ, Techritz S, Jakubowski N,
tissues with subcellular resolution by mass Haase A, Luch A, Panne U, Mueller L (2017)
cytometry. Nat Methods 11(4):417–422. A simple metal staining procedure for identifi-
https://doi.org/10.1038/nmeth.2869 cation and visualization of single cells by
2. Theiner S, Van Malderen SJM, Van Acker T, LA-ICP-MS. Analyst 142(10):1703–1710.
Legin A, Keppler BK, Vanhaecke F, Koellen- https://doi.org/10.1039/c6an02638a
sperger G (2017) Fast high-resolution laser 7. Mueller L, Herrmann AJ, Techritz S, Panne U,
ablation-inductively coupled plasma mass spec- Jakubowski N (2017) Quantitative characteri-
trometry imaging of the distribution of zation of single cells by use of immunocyto-
platinum-based anticancer compounds in mul- chemistry combined with multiplex
ticellular tumor spheroids. Anal Chem 89(23): LA-ICP-MS. Anal Bioanal Chem 409(14):
12641–12645. https://doi.org/10.1021/acs. 3667–3676. https://doi.org/10.1007/
analchem.7b02681 s00216-017-0310-1
3. de Vega RG, Fernandez-Sanchez ML, 8. Wang M, Zheng LN, Wang B, Chen HQ, Zhao
Pisonero J, Eiro N, Vizoso FJ, Sanz-Medel A YL, Chai ZF, Reid HJ, Sharp BL, Feng WY
(2017) Quantitative bioimaging of Ca, Fe, Cu (2014) Quantitative analysis of gold nanopar-
and Zn in breast cancer tissues by ticles in single cells by laser ablation inductively
LA-ICP-MS. J Anal Atom Spectrom 32(3): coupled plasma-mass spectrometry. Anal Chem
6 7 1 – 6 7 7 . h t t p s : // d o i . o r g / 1 0 . 1 0 3 9 / 86(20):10252–10256. https://doi.org/10.
c6ja00390g 1021/ac502438n
4. Willmes M, Glessner JJG, Carleton SA, Gerrity 9. Drescher D, Giesen C, Traub H, Panne U,
PC, Hobbs JA (2016) Sr-87/Sr-86 isotope Kneipp J, Jakubowski N (2012) Quantitative
ratio analysis by laser ablation MC-ICP-MS in imaging of gold and silver nanoparticles in sin-
scales, spines, and fin rays as a nonlethal alter- gle eukaryotic cells by laser ablation ICP-MS.
native to otoliths for reconstructing fish life Anal Chem 84(22):9684–9688. https://doi.
history. Can J Fish Aquat Sci 73(12): org/10.1021/ac302639c
1852–1860. https://doi.org/10.1139/cjfas- 10. Managh AJ, Edwards SL, Bushell A, Wood KJ,
2016-0103 Geissler EK, Hutchinson JA, Hutchinson RW,
5. Giesen C, Waentig L, Mairinger T, Drescher D, Reid HJ, Sharp BL (2013) Single cell tracking
Kneipp J, Roos PH, Panne U, Jakubowski N of gadolinium labeled CD4(+) T cells by laser
(2011) Iodine as an elemental marker for imag- ablation inductively coupled plasma mass
ing of single cells and tissue sections by laser
226 Amy J. Managh and Calum J. Greenhalgh

spectrometry. Anal Chem 85(22): spectrometry. Anal Chem 87(22):

10627–10634. https://doi.org/10.1021/ 11285–11294. https://doi.org/10.1021/acs.
ac4022715 analchem.5b02466
11. Managh AJ, Hutchinson RW, Riquelme P, 14. Van Malderen SJM, van Elteren JT, Vanhaecke
Broichhausen C, Wege AK, Ritter U, F (2015) Development of a fast laser ablation-
Ahrens N, Koehl GE, Walter L, Florian C, inductively coupled plasma-mass spectrometry
Schlitt HJ, Reid HJ, Geissler EK, Sharp BL, cell for sub-mu m scanning of layered materials.
Hutchinson JA (2014) Laser ablation- J Anal Atom Spectrom 30(1):119–125.
inductively coupled plasma mass spectrometry: https://doi.org/10.1039/c4ja00137k
an emerging technology for detecting rare cells 15. Wang HAO, Grolimund D, Giesen C, Borca
in tissue sections. J Immunol 193(5): CN, Shaw-Stewart JRH, Bodenmiller B, Gun-
2600–2608. https://doi.org/10.4049/ ther D (2013) Fast chemical imaging at high
jimmunol.1400869 spatial resolution by laser ablation inductively
12. Reifschneider O, Wentker KS, Strobel K, coupled plasma mass spectrometry. Anal Chem
Schmidt R, Masthoff M, Sperling M, Faber C, 85(21):10107–10116. https://doi.org/10.
Karst U (2015) Elemental bioimaging of thu- 1021/ac400996x
lium in mouse tissues by laser ablation-ICPMS 16. Reid P, Managh AJ (2017) LA-ICP-MS Ima-
as a complementary method to heteronuclear g e To o l . h t t p s : // w w w. r e i d - i t . c o . u k /
proton magnetic resonance imaging for cell downloads/down_la_icp_ms_it.htm
tracking experiments. Anal Chem 87(8): 17. Managh AJ, Douglas DN, Cowen KM, Reid
4225–4230. https://doi.org/10.1021/ HJ, Sharp BL (2016) Acquisition of fast tran-
ac504363q sient signals in ICP-MS with enhanced time
13. Douglas DN, Managh AJ, Reid HJ, Sharp BL resolution. J Anal Atom Spectrom 31(8):
(2015) High-speed, integrated ablation cell 1688–1692. https://doi.org/10.1039/
and dual concentric injector plasma torch for c6ja00140h
laser ablation-inductively coupled plasma mass
 Chapter 15

Patch-seq: Multimodal Profiling of Single-Cell Morphology,

Electrophysiology, and Gene Expression
Cathryn R. Cadwell and Andreas S. Tolias

Abstract
Cells exhibit diverse morphologic phenotypes, biophysical and functional properties, and gene expression
patterns. Understanding how these features are interrelated at the level of single cells has been challenging
due to the lack of techniques for multimodal profiling of individual cells. We recently developed Patch-seq,
a technique that combines whole-cell patch clamp recording, immunohistochemistry, and single-cell
RNA-sequencing (scRNA-seq) to comprehensively profile single cells. Here we present a detailed step-
by-step protocol for obtaining high-quality morphological, electrophysiological, and transcriptomic data
from single cells. Patch-seq enables researchers to explore the rich, multidimensional phenotypic variability
among cells and to directly correlate gene expression with phenotype at the level of single cells.

Key words Patch-seq, Electrophysiology, Single-cell RNA-sequencing, Transcriptomics, Morphol-

ogy, Cell type, Whole-cell recording, Patch clamp

1 Introduction

Advances in single-cell RNA sequencing [1, 2] have enabled unbi-

ased high-throughput molecular cell type classification in complex
tissues. However, linking molecular cell types with their morpho-
logical and electrophysiological phenotypes has been difficult due
to the limited availability of cell subtype–specific tools, such as viral
vectors and transgenic mouse lines. Many cell types cannot be
clearly defined by the differential activation of a single gene or
gene regulatory element, and so new technologies that correlate
transcriptomic and phenotypic data at the level of single cells pro-
vide a powerful and complementary approach to better understand
the molecular underpinnings of cell type diversity.
Early attempts to combine whole-cell patch clamp recording
with single-cell gene expression included reverse transcription poly-
merase chain reaction (RT-PCR) of the patch pipette contents
following whole-cell recording [3–5]. However, this approach
requires a priori selection of a small number of genes, providing

Miodrag Gužvić (ed.), Single Cell Analysis: Methods and Protocols, Methods in Molecular Biology, vol. 2752,
https://doi.org/10.1007/978-1-0716-3621-3_15, © Springer Science+Business Media, LLC, part of Springer Nature 2024

227
228 Cathryn R. Cadwell and Andreas S. Tolias

only a glimpse of the full transcriptome and limiting the identifica-

tion of new genes that may be important in defining cell types.
Single-cell microarray was also used to assess genome-wide expres-
sion after patch recording [6], but microarray has a limited dynamic
range, poor sensitivity, and poor specificity compared to sequenc-
ing-based approaches, and cannot detect novel transcripts or splice
variants. Initial attempts to perform scRNA-seq on patched neu-
rons yielded poor quality sequencing data [7], but we and others
have shown that high-quality scRNA-seq data, comparable to that
obtained from dissociated cells, can be obtained from patch clamp-
recorded neurons [8, 9]. We have since further refined our meth-
odology to improve direct morphological recovery of cells after
Patch-seq [10–12]. Here we present the most up-to-date version
of our Patch-seq protocol. We assume the user has a basic under-
standing of the patch-clamp technique and focus our instructions
on modifications that are needed to facilitate Patch-seq (Fig. 1).

2 Materials

All reagents for RNase-free solutions should be ordered at the

highest purity available. All glassware, spatulas, stir bars, pipettes,
and pipette tips should be certified RNase-free and/or be cleaned
thoroughly with RNase Zap prior to preparing any RNase-free
solution.

2.1 Electro- 1. Juvenile (postnatal day 15–19) or adult wild-type C57Bl/6

physiology and Single- mice (see Note 1).
Cell RNA Sample 2. Isoflurane.
Collection
3. 0.2 M phosphate buffer (PB): Dissolve 23.004 g sodium phos-
phate dibasic (Na2HPO4, 142.0 g/mol) and 4.56 g sodium
phosphate dibasic (NaH2PO4, 120.0 g/mol) in 1 L distilled
water (dH2O). Solution can be stored at room temperature
(25 °C) for up to a month.
4. 0.01 M phosphate-buffered saline (PBS): Dissolve 9 g NaCl in
50 mL 0.2 M PB and 950 mL dH2O. Solution can be stored at
room temperature for up to a month.
5. RNase Zap and DNA-OFF. Wear protective gloves and eye
protection when using.
6. 70% (vol/vol) ethanol: Dissolve non-molecular biology grade
ethanol in dH2O to a final concentration of 70% (vol/vol).
Store in RNase Zap-treated spray bottle for up to a week.
7. Nonstick, RNase-free 1.5 mL microfuge tubes.
8. Thin-walled, RNAse-free 0.2 mL PCR tubes.
 Patch-seq of Single Cells 229

Fig. 1 Schematic of Patch-seq technique. (Adapted from [10] with permission from Springer)

Fig. 2 Custom equipment used in Patch-seq. (a) 1 mL syringe with 0.2 μm syringe filter and tapered pipette
tip, used for backfilling glass pipettes. (b) Patch pipette marked at approximately 0.3 μL volume. (c) Positive
pressure device used to eject pipette contents into PCR tube after aspirating cell contents. (d) Staining
chamber used for immunohistochemistry. (Reproduced from [10] with permission from Springer)

9. RNase-free oligo-dT30VN (sequence: 5′- AAGCAGTGGTAT

CAACGCAGAG
TACTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTVN -3′,
where V represents A, C, or G and N represents any nucleotide;
HPLC purified). Dissolve in molecular biology–grade Tris-
EDTA (TE) buffer at 100 μM. Store 50 μL aliquots at -20 °
C for up to a year.
10. RNase-free ERCC spike-in stock solution: Thoroughly vortex
ERCC RNA spike-in mix. Dilute original ERCC mix 1:4 × 104
in RNase-free sterile water. Store 15 μL aliquots in RNase-free
1.5 mL tubes for up to 6 months at -20 °C. Avoid freeze/
thaw cycles by using a fresh aliquot for each experiment.
11. 96-well polypropylene deep well plates.
12. RNase-free triton X-100 stock: Make 33% (vol/vol) stock
solution in RNase-free sterile water. Store at room temperature
for up to a year. Dilute stock solution 1:100 to make working
dilution 0.33% (vol/vol). Store working dilution at -20 °C for
up to 6 months.
13. RNase-free 0.5 M KOH: Dissolve in RNase-free sterile water at
0.5 M. Store at room temperature for up to a year.
230 Cathryn R. Cadwell and Andreas S. Tolias

14. RNase-free internal solution stock: 111 mM potassium

D-gluconate (K-gluconate), 4 mM potassium chloride (KCl),
10 mM HEPES, 0.2 mM ethylene glycol-bis(2-aminoethy-
lether)-N,N,N′,N′-tetraacetic acid (EGTA), 4 mM adenosine
5′-triphosphate magnesium salt (Mg-ATP), 0.3 mM guanosine
5′-triphosphate sodium salt hydrate (Na-GTP), 5 mM phos-
phocreatine disodium salt hydrate (Na2-phosphocreatine),
13.4 mM biocytin. First, dissolve K-gluconate KCl, HEPES,
and EGTA in ~40 mL RNase-free water in a 125 mL Erlen-
meyer flask. Autoclave, allow to cool to room temperature, and
then add the Mg-ATP, Na-GTP, Na2-phosphocreatine, and
biocytin. Adjust the pH to 7.25 with RNase-free 0.5 M KOH
using a dedicated pH meter that is cleaned with RNase Zap
prior to each use. Adjust osmolarity to 235–240 mOsm using
RNase-free sterile water. Filter the internal solution using an
RNase-free 60 mL syringe (cleaned with RNase Zap) and
0.45 μm syringe filter. Aliquot into 1 mL samples in RNase-
free 1.5 mL tubes. Store at -20 °C for up to 3 weeks.
15. Oxygenated artificial cerebrospinal fluid (ACSF): Dissolve the
following salts in distilled water: 125 mM sodium chloride
(NaCl), 2.5 mM potassium chloride (KCl), 1.25 mM
NaH2PO4, 25 mM sodium bicarbonate (NaHCO3), 1 mM
magnesium chloride (MgCl2), 11.1 mM glucose, 2 mM cal-
cium chloride (CaCl2). Oxygenate for 30 min before adding
the MgCl2 and CaCl2. This solution can be stored at 4 °C for
up to 2 weeks.
16. RNase-free lysis buffer: 0.1% (vol/vol) Triton X-100, 5 mM
(each) dNTPs, 2.5 μM oligo dT30VN, 1 U/μL recombinant
RNase inhibitor, 1.6 × 10-5 ERCC (diluted from 1:4 × 104
ERCC stock) in nuclease-free sterile water. Thaw all reagents
completely and vortex thoroughly before use. Prepare enough
lysis buffer for all samples and controls immediately prior to the
experiment (see Note 2).
17. 4% Paraformaldehyde (PFA): Add 40 μl of 1 N NaOH to
7.5 mL distilled water in a 50 mL Erlenmeyer flask. Cover
with parafilm, and stir on a heating stir plate set on
low-medium heat for 5–10 min. When the solution is warm,
add 0.8 g paraformaldehyde (PFA) and re-cover with parafilm.
As soon as PFA dissolves remove the solution from heat and let
it cool to room temperature. Add 2 mL of 25% (wt/vol)
glutaraldehyde and 10 mL of 0.2 M phosphate buffer. Swirl
gently to combine. Store the PFA/glutaraldehyde solution at
4 °C. For best morphological recovery, use it the same day.
18. Recombinant RNase inhibitor (Takara Bio).
 Patch-seq of Single Cells 231

19. Slice electrophysiology rig equipped with an upright light

microscope with differential interference contrast and a CCD
camera.
20. Vibrating blade microtome.
21. Water bath.
22. Flaming/Brown micropipette puller with 3.0 mm square box
filament. Adjust the pull setting to achieve 3–7 MΩ resistance
pipettes.
23. Borosilicate glass (O.D. 2.0 mm, I.D. 1.6 mm, 10 cm length,
without filament). Autoclave prior to use.
24. 1 mL syringe with tapered pipette tip, for backfilling pipettes
(Fig. 2a). Clean with RNase Zap prior to use.
25. Positive pressure device for ejecting samples from pipette into
PCR tubes (Fig. 2c).
26. RNase-free 50 mL conical tubes.
27. RNase-free filter pipette tips.
28. 13 mm syringe filter with 0.2 μm PTFE membrane.

2.2 cDNA Library 1. RNase-free template switching oligonucleotide (LNA-TSO;

Preparation and sequence: 5′- AAGCAGTGGTATCAACGCAGAGTACrGrG
Sequencing +G-3′, where rG indicates riboguanosines and +G indicates a
locked nucleic acid (LNA)-modified guanosine; RNase-free;
HPLC purified). Dissolve in molecular biology–grade Tris-
EDTA (TE) buffer at 100 μM. Store 20 μL aliquots at -80 °
C for up to a year.
2. RNase-free 5 M betaine: Dissolve in RNase-free sterile water at
5 M. Store at -20 °C for up to 6 months.
3. RNase-free reverse transcription (RT) mix: 1.75× Superscript
II first strand buffer, 17.5 U/μL Superscript II reverse tran-
scriptase, 1.75 M betaine, 8.77 mM dithiothreitol (DTT),
1.75 U/μL recombinant RNase inhibitor, 1.75 μM
LNA-TSO, and 10.5 mM MgCl2 in nuclease-free sterile
water. Be sure that each component is completely thawed and
thoroughly vortexed before adding to mix. Prepare immedi-
ately prior to the experiment or while the samples are in the
thermal cycler for denaturation.
4. ISPCR Oligo (sequence: 5′- AAGCAGTGGTATCAACGCA
GAGT -3′, HPLC purified). Dissolve in molecular biology–
grade Tris-EDTA (TE) buffer at 100 μM. Store 50 μL aliquots
at -20 °C for up to a year.
5. PCR mix: 1.67× KAPA HiFi HotStart Ready Mix, 0.17 μM
ISPCR Oligo. Prepare at the time of sample processing.
6. Nonstick, RNase-free 1.5 mL microfuge tubes.
232 Cathryn R. Cadwell and Andreas S. Tolias

7. Nextera XT primers. Dilute the primers from the Nextera XT

index kit 1:4 with sterile water. They can be stored at –20C for
up to six months.
8. Tn5 transposase (produced according to protocol outlined in
[13]).
9. 0.2% (wt/vol) sodium dodecyl sulfate solution (SDS): Dilute
20% (wt/vol) SDS 1:100 (vol/vol) with sterile water to make a
working dilution of 0.2% (wt/vol). The solution can be stored
at room temperature for up to six months.
10. 200 mM TAPS-NaOH stock solution: Dissolve TAPS in sterile
molecular biology grade water and adjust pH to 8.5 (at room
temperature). The solution can be stored at room temperature
for up to six months.
11. 10× Tagmentation buffer: 10 mM TAPS-NaOH, 50 mM
MgCl2. The solution can be stored at room temperature for
up to six months.
12. Tagmentation mix: 2.5× tagmentation buffer, 25% (wt/vol)
polyethylene glycol solution (PEG-8000), and 3.125 μM Tn5
transposase. Prepare immediately before use.
13. Enrichment PCR mix: 3.3× KAPA HiFi Buffer, 1 mM each
dNTPs, 0.067 U/μL KAPA enzyme. Prepare immediately
prior to use.
14. Thermal cycler.
15. Axygen AxyPrep mag PCR clean-up kit.
16. Ambion magnetic stand – 96.
17. 96-well polypropylene deep well plates.
18. 80% (vol/vol) ethanol: Dilute molecular biology grade ethyl
alcohol in sterile water. Prepare immediately prior to use.
19. Buffer EB.
20. Agilent Bioanalyzer and High Sensitivity DNA kit.
21. Qubit 3.0 quantitation kit.
22. A compatible sequencing machine.

2.3 Immunohisto- 1. 24-well tissue culture plates.

chemistry and 2. Custom staining chamber (Fig. 2d).
Morphological
3. Clinical lab rotator set to low (~60–80 rpm).
Recovery
4. 0.01 M phosphate-buffered saline (PBS): Dissolve 9 g NaCl in
50 mL 0.2 M PB and 950 mL dH2O. Solution can be stored at
room temperature for up to a month.
5. 3% (vol/vol) hydrogen peroxide (H2O2): Combine 45 mL of
0.01 M PBS with 5 mL of 30% (vol/vol) H2O2. Vortex briefly.
Prepare immediately prior to use.
 Patch-seq of Single Cells 233

6. Vectastain ABC-HRP Elite Kit (Standard).

7. A/B mix: Combine 10 mL of 0.01 M PBS, 100 μL Triton
X-100, 4 drops of reagent A, and 4 drops of reagent B (from
ABC Elite Kit) in a 50 mL conical tube. Vortex briefly. Prepare
immediately prior to use.
8. 3,3′-diaminobenzidine (DAB) stock solution: Dissolve DAB
powder at 20 mg/mL in DMSO. Store aliquots at -20 °C for
up to 1 year.
9. DAB mix: Combine 10 mL 0.01 M PBS, 0.25 mL DAB stock
solution, and 3.3 μL 30% (vol/vol) H2O2 in a 50 mL conical
tube. Vortex briefly. Prepare immediately prior to use.
10. 0.2 M Tris buffer: Dissolve 2.4228 g Trizma base (121.14 g/
mol) in 35–50 mL dH2O. Adjust pH to 8.5 using 1 M HCl.
Add dH2O to total volume of 100 mL. Solution can be stored
at room temperature for up to 6 months.
11. Mowiol mounting medium: Weigh out 6 g glycerol in a 50 mL
conical tube. Add 2.4 g mowiol 4–88 and vortex thoroughly.
Add 6 mL dH2O, vortex, and leave at room temperature for
2 h. Add 12 mL 0.2 M Tris buffer and incubate at ~53 °C,
vortexing occasionally, until the mowiol dissolves. Clarify by
centrifugation at 2000–3000g for 20 min. Store 1 mL aliquots
in 1.5 mL tubes at -20 °C for up to 1 year. Once thawed, the
solution is stable at room temperature for up to a month.
12. Superfrost plus gold microscope slides.
13. Cover slips.
14. Light microscope for visualizing cell morphology.

3 Methods

3.1 Electro- 1. Prepare all work surfaces and equipment that may come in
physiology and Single- contact with RNase-free solutions by cleaning with RNase
Cell RNA Sample Zap, followed by 70% (vol/vol) ethanol (see Note 3). Any
Collection equipment that may have come into contact with post-PCR
products (tube holders, pipettes) should also be cleaned thor-
oughly with DNA-OFF before RNase Zap. Items can be
cleaned the day before the experiment.
2. Thaw an aliquot of RNase-free internal solution stock, and add
recombinant RNase inhibitor to a final concentration of 1 U/μ
L and vortex well. Addition of RNase inhibitor should increase
the osmolarity of the internal solution to the desired range of
315–320 mOsm. Adjust the osmolarity of the ACSF to be
18–20 mOsm lower than the internal solution, adding small
amounts of sucrose as needed to increase the osmolarity of the
ACSF (see Note 4).
234 Cathryn R. Cadwell and Andreas S. Tolias

Fig. 3 Ideal patching approach for aspiration of cell contents. (a) The target cell is approached from the side
rather than the top to provide the most direct path for cell contents to enter the pipette. As the cell slowly
swells, it may be helpful to move the pipette back a few microns to release the tension on the membrane and
prevent current leakage. During aspiration of the cell contents, the cell is continuously monitored to ensure
that no extracellular contents are entering the pipette. Following aspiration, the pipette is slowly pulled away
from the cell membrane, leaving the shrunken cell body intact in the tissue slice to preserve axodendritic
morphology. (b) Examples of collapse of the cell body after aspiration of cell contents, visualized using
two-photon imaging (2PI). Scale bars, 10 μm. (c) Examples of collapse of the cell body after aspiration of cell
contents, visualized using differential interference contrast (DIC) imaging. Scale bar, 10 μm. (Reproduced from
[10] with permission from Springer)

3. Draw the internal solution with RNase inhibitor into a 1 mL

syringe, and push through a 0.2 μm syringe filter into the
tapered pipette tip (Fig. 2a, see Note 5).
4. Prepare sample tubes by adding 4 μL of lysis buffer to each
RNase-free 0.2 mL PCR tube. Store tubes in a clean (RNase
Zap–treated) tube holder on ice.
5. Deeply anesthetize the mouse with isoflurane and cut acute
brain slices according to standard protocols for the brain region
of interest (e.g., [11, 14]).
6. Allow the slices to recover in oxygenated ACSF for 1 h at room
temperature (~25 °C).
7. Place a tissue slice in the recording chamber and select a cell to
target using differential interference contrast (DIC) and/or
wide-field fluorescence light microscopy.
 Patch-seq of Single Cells 235

8. Pull an autoclaved glass pipette and backfill with 0.1–0.3 μL of

internal solution (Fig. 2b). Tap or shake to remove any bubbles
(see Note 6).
9. Secure the glass pipette in the pipette holder and lower it into
the circulating ACSF. Immediately apply a small amount of
positive pressure and push out any air bubbles. Measure the
resistance (see Note 7).
10. Advance the pipette up to the target cell, maintaining a small
amount of positive pressure, until the cell membrane is
observed to dimple under the tip and the resistance across
the pipette slightly increases.
11. Release the positive pressure and apply a small amount of
suction, if necessary, to achieve a GΩ seal on the cell mem-
brane. Set the holding potential to -70 mV and compensate
the fast capacitance. Reposition the pipette if needed so that
the cell membrane is not under tension, and the pipette
approaches from the side of the cell rather than the top (Fig. 3).
12. Apply brisk suction to achieve whole-cell access (see Note 8).
Compensate the slow capacitance.
13. Run the desired electrophysiology protocol, typically a series of
hyperpolarizing, subthreshold, and suprathreshold current
steps to record the intrinsic biophysical properties and firing
pattern of the cell.
14. If morphological recovery is desired, hold the cell in current
clamp mode for 30–60 min (see Note 9).
15. Aspirate the cell contents into the patch pipette by gently
applying negative pressure (0.7–1.5 psi, see Note 10). Monitor
the holding currents and size of the cell body on DIC. If the
cell becomes unstable, discontinue the suction and wait a few
minutes to see if it recovers. Continue until the cell body has
significantly decreased in size (typically 2–10 min) or the cell
cannot tolerate more suction.
16. Release the suction and remove the pipette (see Note 11).
17. Immediately eject the contents of the pipette into an RNase-
free 0.2 mL PCR tube containing 4 μL lysis buffer (prepared in
step 4) by applying a small amount of positive pressure (using
the device depicted in Fig. 2c, or something similar) and gently
breaking the tip on the bottom of the tube. Avoid making any
bubbles. Cap the PCR tube and store on ice until all samples
have been collected.
18. If morphological recovery is desired, immediately remove the
slice from the recording chamber and place it in one well of a
24-well culture dish. Remove the excess ACSF and add ~1 mL
of cold, fresh PFA. Cover the plate with parafilm underneath
the lid to form an airtight seal (see Note 12).
236 Cathryn R. Cadwell and Andreas S. Tolias

3.2 cDNA Library 1. Remove any remaining bubbles by gently tapping the side of
Preparation and each tube. Spin down the samples at 700 g for 10 s at 4 °C and
Sequencing place back on ice.
2. Denature the RNA by incubating the samples in a thermal
cycler for 3 min at 72 °C.
3. Add 5.7 μL of RT mix to each tube and pipette up and down
four times to mix. Gently tap each tube as needed to remove
bubbles and spin down at 700 g for 10 s at 4 °C. Immediately
return to ice.
4. Incubate the samples in a thermal cycler according to the
following program: step 1: 42 °C for 90 min; step 2: 50 °C
for 2 min; step 3: 42 °C for 2 min (repeat steps 2–3 ten times);
step 4: 70 °C for 10 min; step 5: Hold at 4 °C (see Note 13).
5. Add 15 μL of PCR mix to each tube, vortex to mix, and spin
down at 700 g for 10 s at 4 °C.
6. Incubate the samples in a thermal cycler according to the
following program: step 1: 98 °C for 3 min; step 2: 98 °C
for 20 s; step 3: 67 °C for 15 s; step 4: 72 °C for 6 min (repeat
steps 2–4 eighteen times); step 5: 72 °C for 5 min; step 6:
Hold at 4 °C (see Notes 14 and 15).
7. Allow the Axygen AxyPrep mag PCR beads to come to room
temperature for 15 min. Vortex well.
8. Add 17.5 μL of beads to each sample and mix thoroughly by
pipetting up and down at least 10 times.
9. Transfer samples to a 96-well plate, cover with lid, and incubate
for 8 min at room temperature.
10. Place the 96-well plate on the magnetic stand and incubate for
an additional 5 min.
11. Carefully remove the clear solution without disturbing the
beads.
12. Wash the beads two times with 80% ethanol for 30 s each time.
Carefully remove the ethanol without touching the beads after
each wash.
13. Cover the plate with a lid and leave at room temperature until
the beads are completely dry and small cracks appear on the
surface (typically 5–10 min).
14. Take the plate off the magnetic stand, add 17.5 μL of Buffer EB
to each well, and pipette up and down ~10 times to resuspend
the beads.
15. Incubate the plate off the magnetic stand for 2 min.
16. Place the plate back on the magnetic stand and incubate for an
additional 2–3 min. Without touching the beads, collect 15 μL
 Patch-seq of Single Cells 237

Fig. 4 Quality check of full-length single-cell cDNA libraries. Bioanalyzer profiles of single-cell amplified cDNA
from several cells from a typical experiment (a) showing variable cDNA yield, as well as a negative control
from the same experiment that consisted of ERCC spike-in RNA only. Profiles from degraded (b) or
contaminated (c) samples are also shown. The low molecular weight fragments in (c) were traced back to
bacterial contamination of a reagent. (Reproduced from [10] with permission from Springer)

Fig. 5 Quality check of pooled sequencing library. Example of bioanalyzer profile

showing the size distribution of the final pooled sequencing library. (Reproduced
from [10] with permission from Springer)

of the clear liquid from each well into a new 0.2 mL PCR tube
(see Note 16).
17. Check the concentration and size distribution of the full-
length cDNA library on an Agilent Bioanalyzer using the
Agilent High Sensitivity DNA Kit according to the manufac-
turer’s instructions (Fig. 4; see Note 17).
18. Dilute each sample to 50 pg/μL.
19. Add 4 μL of tagmentation mix to 6 μL of each sample.
Vortex well.
238 Cathryn R. Cadwell and Andreas S. Tolias

Fig. 6 Post-sequencing quality control. The quality of the samples can be assessed by analyzing the
sequencing depth (a), number of detected genes (b), the maximum spearman correlation between each cell
and all other cells (c), and fraction of reads mapped to exons, introns, and intergenic segments (d). For this
data set approximately 10% of the samples were discarded post-sequencing based on pre-determined quality
control criteria (dashed lines in (a) and (b) represent the cutoffs below which samples were discarded). (e)
Histogram of mean normalized expression across cells for all genes in the example dataset. Genes with a
mean expression of less than one (dashed line) can be excluded to minimize the impact of gene dropout on
downstream analyses. (Reproduced from [10] with permission from Springer)

20. Incubate the samples in a thermal cycler at 55 °C for 8 min.

21. Add 2.5 μL of 0.2% SDS to each sample and incubate at room
temperature for 5 min. Return to ice.
22. Add 2.5 μL each of index 1 (N7XX) and index 2 (S5XX)
primers, ensuring that each sample receives a unique combina-
tion of indices.
23. Add 7.5 μL of enrichment PCR mix to each tube, vortex and
spin down at 700 g for 10 s at 4 °C.
24. Incubate the samples in a thermal cycler according to the
following program: step 1: 72 °C for 3 min; step 2: 95 °C
for 30 s; step 3: 95 °C for 10 s; step 4: 55 °C for 30 s; step 5:
72 °C for 30 s (repeat steps 3–5 12 times); step 6: 72 °C for
5 min; step 7: Hold at 4 °C.
25. Pool 2.5 μL in a 1.5 mL tube.
26. Perform PCR purification on the pooled, amplified cDNA as
described above (steps 7–16) with the following changes: 1)
use a 1:1 ratio of beads to cDNA, 2) resuspend the beads in
 Patch-seq of Single Cells 239

0.875 μL × the number of samples, 3) collect a final volume of

0.75 μL × the number of samples.
27. Determine the mean size and concentration of the final cDNA
library using an Agilent Bioanalyzer and Qubit, respectively
(Fig. 5; see Note 18). Dilute the library to ~3 nM.
28. Sequence the library on a compatible machine, such as Illumina
NextSeq 500, according to the manufacturer’s instructions.
29. Demultiplex the reads and align to the appropriate reference
genome.
30. Assess the quality of the samples and exclude poor quality
samples using predetermined criteria (Fig. 6).
31. Perform additional exploratory analyses as desired (see Note
19).

3.3 Immunohisto- 1. After 10–14 days of fixation, transfer the tissue slices to a
chemistry and staining chamber with mesh-bottom wells (Fig. 1d, see Note
Morphological 20).
Recovery 2. Carefully remove any trace of PFA and wash the slices three
times with 0.01 M PBS. Cover and leave on the rotator for
10 min at room temperature during each wash.
3. Remove the PBS and submerge slices in the 3% (vol/vol)
H2O2. Cover and leave on the rotator at room temperature
for 30 min.
4. Remove the 3% (vol/vol) H2O2 and wash the slices three times
with 0.01 M PBS. Cover and leave on the rotator for 10 min at
room temperature during each wash.
5. Remove the PBS and add the A/B mix. Cover and store
overnight at 4 °C.
6. Remove the A/B mix and wash the slices four times with
0.01 M PBS. Cover and leave on the rotator for 10 min at
room temperature during each wash.
7. Remove the PBS and add the DAB mix. Leave on the rotator at
room temperature for 7–8 min.
8. Remove the DAB mix and wash the slices three times with
0.01 M PBS. Cover and leave on the rotator for 10 min at
room temperature during each wash.
9. Mount the slices on microscope slides in Mowiol mounting
medium and cover with a glass coverslip (see Note 21).
10. Visualize cell morphology under a light microscope (Fig. 7, see
Note 22).
240 Cathryn R. Cadwell and Andreas S. Tolias

Fig. 7 Direct morphological recovery of Patch-seq cells. Examples of two excitatory (a, c) and two inhibitory
(b, d) neurons with successful characterization of morphologic and electrophysiological profiles as well as
acquisition of full-length cDNA. Scale bars for histochemical staining, 50 μm. Scale bars for electrophysiologi-
cal traces, 100 ms and 50 mV/500 pA. (Adapted from [10] with permission from Springer)

4 Notes

1. Experiments involving animals should be carried out in accor-

dance with the National Research Council’s “Guide for the
Care and Use of Laboratory Animals” and with approval from
the appropriate Institutional Animal Care and Use Committee.
2. Inclusion of appropriate positive and negative controls early in
the sample collection process, and at various intermediate
stages of processing, is critical to ensure sample quality and
can help tremendously to localize problems if sample quality
becomes compromised. Endogenous (within the tissue itself)
or exogenous (from the environment or experimenter) RNase,
RNA from cells other than the one targeted, and cross-
contamination with amplified cDNA from prior experiments
are all important potential sources of contamination that
should be eliminated to the extent possible and controlled for
by experimental design. Another important consideration
when designing a Patch-seq experiment is technical variability
and bias that can be introduced during the library preparation,
PCR amplification, and sequencing. To control for this, it is
helpful to collect and process samples in a randomized fashion
if possible and to assess results for batch effects.
 Patch-seq of Single Cells 241

3. Great care should be taken to reduce or eliminate any source of

RNase contamination during sample collection and solution
preparation. Items should be re-cleaned during the experiment
if they may have become contaminated (e.g., if the glass pipette
breaks and the wire electrode comes into contact with ACSF).
In addition to using sterile techniques, certified RNase-free
solutions, tubes, and pipette tips should be used whenever
possible. Any reagents not available as certified RNase-free
should be ordered at the highest purity available. Glassware,
pipettes, counters, and all other equipment should be cleaned
thoroughly with RNase Zap and autoclaved if possible. Ideally,
a dedicated RNase-free positive pressure room or dedicated
RNase-free bench stocked with RNase-free equipment (pip-
ettes, tips, glassware, tubes, vortex, tabletop centrifuge) should
be used exclusively for pre-RT sample handling steps.
4. Let the ACSF and internal solution come to room temperature
and alternate measuring the osmolarity of the two solutions to
obtain the most accurate comparative measurements. Add
small amounts of water or sucrose and stir thoroughly before
each measurement. Obtain at least three measurements when
approaching the desired osmolarity (within 10 mOsm of the
target osmolarity) to ensure the readings are stable before
making further adjustments.
5. Carefully tap the syringe to remove any air bubbles before
pushing the internal solution through the filter. If not
removed, air bubbles will destroy the integrity of the syringe
filter and can lead to clogging of the pipette during patching.
6. Using too much internal solution will inhibit the downstream
reverse transcription reaction.
7. In general, pipettes of 3–4 MΩ are typically ideal for larger cells
such as layer V pyramidal neurons, and pipettes of 5–7 MΩ are
ideal for smaller cells such as interneurons.
8. Discontinue suction immediately when the cell membrane
opens to prevent losing the seal between the cell membrane
and the pipette. Signs of a broken seal include an unstable
resting membrane potential, a large current required to hold
the cell at -70 mV (>200–300 pA), or blunted/absent action
potentials when stimulated in the current clamp. If the record-
ing seems unstable or there is a large amount of current leak-
age, try adjusting the pipette position or wait a few minutes
and, if there is no improvement, then the sample should be
discarded, and a new cell should be targeted.
9. Injecting the cell with large depolarizing currents (e.g.,
15 × 100 ms pulses of 10–20 nA delivered at 1 Hz) during
this holding period may help to diffuse biocytin throughout
the fine axonal processes and improve morphological recovery.
242 Cathryn R. Cadwell and Andreas S. Tolias

10. Be cautious to apply the suction gently to avoid losing the seal.
Use a large (20–30 mL), new syringe that is partially with-
drawn, and ensure smooth contact between the rubber piston
and syringe wall. If the cell is lost (i.e., there is a sudden increase
in the amount of current required to hold the cell at -70 mV),
immediately discontinue suction and remove the pipette. The
RNA may still be salvageable, but endogenous RNases from
the extracellular space will degrade it quickly. If a substantial
amount of extracellular material has entered the pipette, the
sample should be discarded.
11. If attempting to preserve cell morphology, first move the
pipette back slowly from the cell and allow the membrane to
slowly retract from the pipette before removing the pipette
from the tissue.
12. The fixed tissue slice can be stored at 4 °C for 10–14 days
before proceeding with immunohistochemistry.
13. Samples can be left in the PCR machine overnight while run-
ning the RT reaction.
14. PCR Products can be stored at -20 or -80 °C for up to
6 months before proceeding.
15. Amplified cDNA should be handled and stored in a separate
post-PCR area away from the pre-PCR samples to prevent
cross-contamination of new samples with amplified cDNA.
16. Purified full-length cDNA can be stored at -20 or -80 °C for
up to 6 months before proceeding with tagmentation and
sequencing.
17. The full-length cDNA should have a broad peak at
1500–2000 bp and very few fragments <500 bp.
18. The mean length of fragments in the final sequencing library
should be <500 bp.
19. A complete overview of possible analytical approaches is
beyond the scope of this chapter. In addition to the repertoire
of scRNA-seq analytical tools, additional resources specific to
analyzing Patch-seq data sets include [15–17].
20. If a specialized staining chamber is not available, staining can
be performed in a 24-well plate by changing the solution in
each well separately.
21. The microscope slides can be stored indefinitely at room tem-
perature, protected from light.
22. Staining quality may continue to improve for 48–72 h after
mounting on slides.
 Patch-seq of Single Cells 243

References
1. Picelli S, Bjorklund AK, Faridani OR, 10. Cadwell CR, Scala F, Li S, Livrizzi G, Shen S,
Sagasser S, Winberg G, Sandberg R (2013) Sandberg R, Jiang X, Tolias AS (2017) Multi-
Smart-seq2 for sensitive full-length transcrip- modal profiling of single-cell morphology,
tome profiling in single cells. Nat Methods electrophysiology, and gene expression using
10(11):1096–1098. https://doi.org/10. Patch-seq. Nat Protoc 12(12):2531–2553.
1038/nmeth.2639 https://doi.org/10.1038/nprot.2017.120
2. Tang F, Barbacioru C, Wang Y, Nordman E, 11. Scala F, Kobak D, Bernabucci M, Bernaerts Y,
Lee C, Xu N, Wang X, Bodeau J, Tuch BB, Cadwell CR, Castro JR, Hartmanis L, Jiang X,
Siddiqui A, Lao K, Surani MA (2009) mRNA- Laturnus S, Miranda E, Mulherkar S, Tan ZH,
Seq whole-transcriptome analysis of a single Yao Z, Zeng H, Sandberg R, Berens P, Tolias
cell. Nat Methods 6(5):377–382. https://doi. AS (2021) Phenotypic variation of transcrip-
org/10.1038/nmeth.1315 tomic cell types in mouse motor cortex. Nature
3. Eberwine J, Yeh H, Miyashiro K, Cao Y, Nair S, 598(7879):144–150. https://doi.org/10.
Finnell R, Zettel M, Coleman P (1992) Analy- 1038/s41586-020-2907-3
sis of gene expression in single live neurons. 12. Scala F, Kobak D, Shan S, Bernaerts Y,
Proc Natl Acad Sci U S A 89(7):3010–3014. Laturnus S, Cadwell CR, Hartmanis L,
https://doi.org/10.1073/pnas.89.7.3010 Froudarakis E, Castro JR, Tan ZH,
4. Sucher NJ, Deitcher DL (1995) PCR and Papadopoulos S, Patel SS, Sandberg R,
patch-clamp analysis of single neurons. Neuron Berens P, Jiang X, Tolias AS (2019) Layer 4 of
14(6):1095–1100. https://doi.org/10.1016/ mouse neocortex differs in cell types and circuit
0896-6273(95)90257-0 organization between sensory areas. Nat Com-
5. Sucher NJ, Deitcher DL, Baro DJ, Warrick mun 10(1):4174. https://doi.org/10.1038/
RM, Guenther E (2000) Genes and channels: s41467-019-12058-z
patch/voltage-clamp analysis and single-cell 13. Picelli S, Bjorklund AK, Reinius B, Sagasser S,
RT-PCR. Cell Tissue Res 302(3):295–307. Winberg G, Sandberg R (2014) Tn5 transpo-
https://doi.org/10.1007/s004410000289 sase and tagmentation procedures for massively
6. Subkhankulova T, Yano K, Robinson HP, Live- scaled sequencing projects. Genome Res
sey FJ (2010) Grouping and classifying 24(12):2033–2040. https://doi.org/10.
electrophysiologically-defined classes of neo- 1101/gr.177881.114
cortical neurons by single cell, whole-genome 14. Jiang X, Shen S, Cadwell CR, Berens P, Sinz F,
expression profiling. Front Mol Neurosci 3:10. Ecker AS, Patel S, Tolias AS (2015) Principles
https://doi.org/10.3389/fnmol.2010.00010 of connectivity among morphologically
7. Qiu S, Luo S, Evgrafov O, Li R, Schroth GP, defined cell types in adult neocortex. Science
Levitt P, Knowles JA, Wang K (2012) Single- 350(6264):aac9462. https://doi.org/10.
neuron RNA-Seq: technical feasibility and 1126/science.aac9462
reproducibility. Front Genet 3:124. https:// 15. Kobak D, Berens P (2019) The art of using
doi.org/10.3389/fgene.2012.00124 t-SNE for single-cell transcriptomics. Nat
8. Cadwell CR, Palasantza A, Jiang X, Berens P, Commun 10(1):5416. https://doi.org/10.
Deng Q, Yilmaz M, Reimer J, Shen S, 1038/s41467-019-13056-x
Bethge M, Tolias KF, Sandberg R, Tolias AS 16. Kobak D, Bernaerts Y, Weis MA, Scala F, Tolias
(2016) Electrophysiological, transcriptomic AS, Berens P (2021) Sparse reduced-rank
and morphologic profiling of single neurons regression for exploratory visualisation of
using patch-seq. Nat Biotechnol 34(2): paired multivariate data. J R Stat Soc Ser C
199–203. https://doi.org/10.1038/nbt. Appl Stat 70(4):980–1000. https://doi.org/
3445 10.1111/rssc.12494
9. Fuzik J, Zeisel A, Mate Z, Calvigioni D, 17. Tripathy SJ, Toker L, Bomkamp C, Mancarci
Yanagawa Y, Szabo G, Linnarsson S, Harkany BO, Belmadani M, Pavlidis P (2018) Assessing
T (2016) Integration of electrophysiological transcriptome quality in patch-seq datasets.
recordings with single-cell RNA-seq data iden- Front Mol Neurosci 11:363. https://doi.org/
tifies neuronal subtypes. Nat Biotechnol 34(2): 10.3389/fnmol.2018.00363
175–183. https://doi.org/10.1038/nbt.
3443
 Chapter 16

Single-Nucleus ATAC-seq for Mapping Chromatin

Accessibility in Individual Cells of Murine Hearts
Michail Yekelchyk, Xiang Li, Stefan Guenther, and Thomas Braun

Abstract
During the last decade a wide range of single-cell and single-nucleus next-generation sequencing techni-
ques have been developed, which revolutionized detection of rare cell populations, enabling creation of
comprehensive cell atlases of complex organs and tissues. State-of-the-art methods do not only allow
classical transcriptomics of individual cells but also comprise a number of epigenetic approaches, including
assessment of chromatin accessibility by single-nucleus Assay for Transposase Accessible Chromatin ATAC-
seq (snATAC-seq). The snATAC-seq assay detects “open chromatin,” a term for low nucleosome occu-
pancy of genomic regions, which is a prerequisite for effective transcription factor binding. Information
about open chromatin at the single-nucleus level helps to recognize epigenetic changes, sometimes before
transcription of respective genes occurs. snATAC-seq detects cellular heterogeneity in otherwise still
transcriptionally and/or morphologically homogeneous cell populations. Chromatin accessibility assays
may be used to detect epigenetic changes in cardiac lineages during heart development, chromatin
landscape changes during aging, and epigenetic alterations in heart diseases. Here, we provide an optimized
protocol for snATAC-seq of murine hearts. We describe isolation of single nuclei from snap-frozen hearts,
provide hints for preparation of libraries suitable for snATAC-seq next-generation sequencing (NGS) using
the Chromium 10× platform, and give general recommendations for downstream analysis using conven-
tional bioinformatic pipelines and packages. The protocol should serve as a beginner’s guide to generate
high-quality snATAC-seq datasets and to perform chromatin accessibility analysis of individual heart-
derived cell nuclei.

Key words Heart, Single-cell, Single-nucleus ATAC-seq, Transposase, Sequencing, Chromatin acces-
sibility, NGS, Chromium 10x™

1 Introduction

Chromatin accessibility depends on positions and density of

nucleosomes, which are composed of DNA wrapped around dimers
of the four core histone proteins, H2A, H2B, H3, and H4. The
chromatin is often referred to as “closed” and “open,” while in
reality there is a gradient of chromatin accessibility from inaccessi-
ble heterochromatin to accessible euchromatin, based on the
degree of nucleosome condensation [1]. The position of

Miodrag Gužvić (ed.), Single Cell Analysis: Methods and Protocols, Methods in Molecular Biology, vol. 2752,
https://doi.org/10.1007/978-1-0716-3621-3_16, © Springer Science+Business Media, LLC, part of Springer Nature 2024

245
246 Michail Yekelchyk et al.

nucleosomes is not stationary. Nucleosomes are able to slide along

genomic DNA, thereby rapidly altering chromatin accessibility,
either opening or closing certain regulatory regions. Nucleosome
sliding or eviction allows rapid modulation of transcription factor
(TF) binding, modifying gene expression according to cellular
needs [2]. Different cell types acquire distinct chromatin accessibil-
ity profiles during development, which support specific transcrip-
tional signatures [3]. Not all accessible regulatory elements are
constitutively active at basal conditions. Some genes are “primed”
for rapid expression when specific external or internal stimuli
emerge.
Heart development goes along with alterations in chromatin
accessibility as demonstrated by Jia, Preussner, and colleagues, who
analyzed promoter regions of Nkx2-5 and Isl1, critical genes for
cardiac progenitor cells [4]. Accessibility of the Isl1 gene primes
cardiac progenitor cells to become cardiomyocytes, smooth mus-
cles, and endothelial cells, while accessibility of the Nkx2-5 gene
restricts cardiac progenitor cells to the cardiomyocyte lineage.
Single-cell (and single-nucleus) resolution allowed the authors to
decipher early cell commitment steps toward different cardiac
lineages in otherwise undistinguishable cardiac progenitor cells.
In a nutshell, single-nucleus resolution of chromatin accessibility
allows separation of distinct cell types before transcriptional
changes occur [5]. It is now possible to detect branching of cell
lineages at very early stages, well before the expression of specific
markers of distinct cell populations commences.
The ATAC-seq method was initially developed by Jason Buen-
rosto in 2013 [6]. The method is based on a hyperactive Tn5
transposase, which reaches open chromatin in cell nuclei, where it
cuts double-stranded DNA and simultaneously attaches specific
primers to the ends of the respective fragments (“Tagmentation”)
[3]. Individual nuclei keep their integrity during Tagmentation and
retain DNA fragments within the nuclei. This feature allows sorting
of tagmented nuclei [5] and encapsulation into droplets for
droplet-based single-cell/single-nucleus NGS approaches
[7, 8]. In this manuscript, we provide guidance that allows users
to avoid potential pitfalls and set up the snATAC-seq protocol with
relatively little effort.

2 Materials

See Table 1 for the list of suggested equipment, and Table 2 for the
list of necessary chemicals, reagents, and enzymes.
 snATAC-seq of Individual Heart Cells 247

Table 1
Laboratory equipment

Equipment Manufacturer
GentleMACS dissociator, M tubes Miltenyi Biotec, Germany
40 μm cell strainer No preference
Centrifuge (for 15 mL conical tubes and for 1.5 mL tubes) No preference
15 mL conical tubes, 1.5 mL tubes, 0.2 mL tubes No preference
Fluorescent-activated cell sorting (FACS) machine No preference
Chromium Controller & Next GEM Accessory Kit 120223 or 1000202, 10xGenomics, USA
Polymerase Chain Reaction (PCR) thermocycler No preference
Magnetic separator for SPRI bead cleanup No preference
Qubit measurement device and HS DNA kit Q32851, Thermo, USA
Capillary electrophoresis instrument Agilent, USA or PerkinElmer, USA
Next Generation Sequencer Illumina, USA

Table 2
Chemicals and enzymes

Final concentration in
Chemicals and enzymes the protocol Manufacturer
Dulbecco’s phosphate-buffered saline (DPBS) – Gibco / Thermo Fisher
Scientific, USA
Anti-PCM-1 antibody 1:1000 HPA023374, Sigma-
Aldrich, Germany
Anti-rabbit secondary antibody conjugated to 1:1000 A11011, Life
Alexa 594 Technologies
Chromium Next GEM Chip H Single Cell Kit – 1000162, 10xGenomics,
USA
Chromium Next GEM Single Cell ATAC – 1000176, 10xGenomics,
Library & Gel Bead Kit v1.1 USA
Single Index Kit N Set A – 1000212, 10xGenomics,
USA
Nuclease-free water – Elut
Elution Buffer (EB) – Qiagen, Netherlands
Solid Phase Reversible Immobilization (SPRI) – AC-60050, MagBio,
magnetic beads Switzerland
High-75 NextSeq500 cartridge – 20024906, Illumina, USA
248 Michail Yekelchyk et al.

Buffer Recipes
1. Lysis buffer: 5 mM CaCl2, 3 mM Mg(CH3COO)2, 2 mM
EDTA, 0.5 mM Ethylene glycol-bis(β-aminoethyl ether)-N,
N,N′,N′-tetraacetic acid (EGTA), and 10 mM Tris-HCl
(pH 8).
2. Triton buffer I: 0.4% Triton X-100 solution: 40 μL Triton
X-100 in 10 mL PBS.
3. Triton buffer II: 0.2% Triton X-100 solution: 20 μL Triton
X-100 in 10 mL PBS.
4. Sucrose buffer: 1 M sucrose, 3 mM Mg(CH3COO)2, 10 mM
Tris-HCl (pH 8).
5. Staining buffer: DPBS, 1% BSA, 0.2% Igepal CA-630,
1 mM EDTA.
6. DAPI staining solution: 2 ng/mL DAPI in nuclease-free water.
7. Tagmentation Mix: prepared in accordance with the Next
GEM scATAC-seq kit manual (contains ATAC Buffer B
(PN2000193) and ATAC Enzyme (PN2000123 or
PN2000138)).
8. Barcoding Mastermix: prepared in accordance with the Next
GEM scATAC-seq kit manual (contains scATAC Gel Beads
v.1.1 (PN2000210), Reducing Agent B (PN2000087), Bar-
coding Reagent B (PN2000194), Barcoding Enzyme
(PN2000125 or PN2000139), Partitioning Oil
(PN2000190), and 10× controller accessory parts as per
manual).
9. Elution solution: prepared in accordance with the Next GEM
scATAC-seq kit manual (contains Buffer EB, 0.1% Tween
20 (10 μL Tween 20 in 10 mL PBS), and Reducing Agent B
(PN2000087)).

3 Methods

3.1 A Brief Protocol The following protocol describes isolation of nuclei from murine
for Cardiac Nuclei hearts and optional identification of nuclei with PCM1 (pericen-
Isolation triolar material 1), which specifically labels cardiomyocyte nuclei
[9]. Combined with fluorescence-activated cell sorting (FACS),
this approach not only allows purification of all nuclei from hearts
for snATAC-seq but also separation of cardiomyocytes nuclei from
non-cardiomyocytes, which has advantages for some applications.
1. Sacrifice a mouse following approved procedures and dissect
the heart. Place the heart into ice-cold PBS buffer to remove
residual blood (see Note 1). Snap-freeze the heart in liquid
 snATAC-seq of Individual Heart Cells 249

nitrogen. The frozen heart can be stored at -80 °C up to

a year.
2. Thaw the frozen mouse heart in 3 mL of Lysis buffer. Homog-
enize the heart tissue utilizing Miltenyi gentleMACS™ disso-
ciator with M tubes and the protocol “protein_01.”
3. Add 3 mL of lysis buffer supplemented with Triton buffer I into
the M tube and keep the sample on ice for 10 min.
4. Filter the cell suspension using a 40 μm cell strainer. Wash the
filter with 2 mL lysis buffer, supplemented with Triton buffer
II. Centrifuge the suspension with 1000g for 5 min at 4 °C.
5. Resuspend the pellet and overlay the suspension on 2 mL of
Sucrose buffer, then centrifuge (1000g, 5 min, 4 °C).
6. (Optional) Discard the supernatant and resuspend the pellet in
500 μL of Staining buffer.
7. (Optional) Stain the nuclei with anti-PCM-1 antibody (1:
1000) for 30 min at room temperature (RT).
8. (Optional) Centrifuge the suspension with 1000g for 5 min, at
4 °C, and wash once with Staining buffer.
9. (Optional) Subsequently, incubate nuclei with an anti-rabbit
secondary antibody conjugated with Alexa 594 (1:1000) for
30 min at RT. Centrifuge (1000g, 5 min, 4 °C) and wash once
with Staining buffer.
10. In case the sample is not stained with the PCM1 cardiomyocyte
marker (steps 6–9), directly resuspend the pellet from step 5 in
500 μL Staining buffer. Then centrifuge the solution once with
1000g for 5 min at 4 °C, discard the supernatant, and resus-
pend the pellet again in the Staining buffer.
11. Stain the nuclei with DAPI staining solution for 10 min at RT.
12. The samples are ready for FACS. Use gating for forward and
side scatter (FCC/SCC) to exclude cell debris and aggregates.
Additionally, use a gate for nuclear DAPI stain (sorted nuclei
should be DAPI+) (see Note 2).
13. Nuclei are sorted using the same staining buffer as in step 10.

3.2 Brief Description The following protocol briefly describes the snATAC-seq library
of the Chromium Next preparation using the Chromium 10x™ kit and controller. For a
GEM Single-Cell ATAC detailed protocol, refer to the manufacturer’s instructions
Protocol [10, 11]. Here, we provide tips for library preparation from cardiac
nuclei.
1. The total number of nuclei isolated from whole murine hearts
is way too high for direct loading onto the Chromium 10×
snATAC-seq cartridge. The nuclei suspension should be
diluted. Depending on the efficiency of nuclei isolation, con-
secutive dilution steps may be necessary. The cartridge allows
250 Michail Yekelchyk et al.

loading of different numbers of nuclei, which adjusts the total

number of data points in the analysis (see Note 3).
2. Dilute nuclei and mix them with the Tagmentation Mix. Incu-
bate the mixture for 1 h at 37 °C (lid temperature = 50 °C).
3. When the incubation is done, mix tagmented nuclei with the
Barcoding Mastermix. Next, load the Chromium 10× snATAC-
seq cartridge according to the manufacturer’s instructions. The
separate rows are filled with gel beads, transposed nuclei in the
Mastermix, and partitioning oil. Empty wells are filled with
glycerol. Cover the cartridge with a rubber gasket and place
in the Chromium 10× controller.
4. When the droplet generation is done, immediately transfer the
acquired Gel Bead-in-Emulsion (GEMs) into a PCR strip, and
run PCR cycles, using the following protocol:
1. 72 °C 5 min
2. 98 °C 30 s
3. 98 °C 30 s
4. 59 °C 30 s
5. 72 °C 1 min
6. go back to step 3, 11 times (12 cycles total)
7. hold at 15 °C
5. After completion of the PCR, a recovery agent (see manufac-
turer’s protocol) is added to the tubes, which breaks the dro-
plets and forms a biphasic mixture (aqueous and pink oil
fractions). The pink (oil) lower fraction is removed and the
leftover solution is cleaned with Dynabeads™ MyOne
SILANE beads. Finally, the DNA fragments are eluted with
an Elution solution.
6. Clean the DNA fragments using SPRI beads (see manufac-
turer’s protocol).
7. Cleaned DNA fragments are used for library construction.
Briefly, the i5 and i7 uniquely barcoded primers, as well as the
PCR Amplification mix are added to the tubes. Amplify frag-
ments using the following PCR protocol:
1. 98 °C 45 s
2. 98 °C 20 s
3. 67 °C 30 s
4. 72 °C 20 s
5. go back to step 2, 8 times (9 cycles total)
6. 72 °C 1 min
7. hold at 4 °C
 snATAC-seq of Individual Heart Cells 251

8. Clean the library using SPRI beads according to the manufac-

turer’s protocol.

3.3 NGS Library To check quality (QC) and size of the final snATAC-seq library,
Quality Assessment measure concentration and size distribution. We suggest to use the
Qubit ™ HS DNA kit for concentration measurements (normal
range for such NGS libraries is 1–50 ng/μL). Capillary electropho-
resis should be employed to determine size distribution of the
library (Agilent Bioanalyzer, Agilent Fragment-Analyzer, PerkinEl-
mer Lab-Chip, or similar). We do not recommend using agarose gel
electrophoresis, since its sensitivity and resolution are not sufficient
for this purpose. Often snATAC-seq libraries have a “wavy struc-
ture” of size distribution, reflecting the nucleosome pattern
(Fig. 1). Although a “wavy structure” is a good sign, it is not
mandatory before moving forward and may depend on the cell
type and state of chromatin (see Note 4).

3.4 Sequencing of Sequencing should be performed using the Illumina platform. For
the snATAC-seq example, Illumina NextSeq500 and “High-75” cartridges can be
Libraries used, which allows to obtain 400–500 M of sequencing reads,
accommodating one snATAC-seq library (up to 10k data points)
(see Note 5). The following setup should be used for NextSeq500
(paired-end sequencing; bp):
Read 1–34;
Read 2–34;
Index 1–8;
Index 2–16.

3.5 Description of 1. After sequencing, the acquired BCL output folder is used for
the Conventional demultiplexing of sequencing reads. Demultiplexing is done
Bioinformatical using the “cellranger-atac mkfastq” pipeline, which is part of
Analysis Steps the Cell Ranger ATAC analysis pipeline developed by
10xGenomics [11].
2. After the sequencing reads are demultiplexed, read mapping,
peak detection, and count are accomplished via the “cellranger-
atac count” pipeline (see Note 6).
3. (Optional) To combine multiple snATAC-seq runs, the “cell-
ranger-atac aggr” pipeline can be used. Note, that individual
runs should be first demultiplexed and counted individually (see
Note 7).
4. After the counting (and aggregating, if applicable) is done, we
recommend to continue the analysis in the R programming
environment. We suggest to use “Signac” and “Seurat” R
packages as a basis [12–14].
5. Counts and metadata are imported into R using the
“Read10x_h5” function. The metadata and the fragments
252 Michail Yekelchyk et al.

Fig. 1 Quality control of snATAC-seq libraries. A “wavy” pattern of the snATAC-seq library is a good indicator
of quality. A “wavy” pattern reflects nucleosome occupancy, which may vary among different cell types. In
general, a consistent “wavy” pattern is observed in cells with tight chromatin organization (e.g., various stem
cell populations). Some fully differentiated somatic cells may have a less well-defined, or even absent “wavy”
pattern

Fig. 2 Identification of different cell types in snATAC-seq datasets. (a) Heatmap showing chromatin accessi-
bility (“gene activity”) of top marker genes in different cell types of the heart. (b) The scatter plots overlay
marker gene accessibility (“gene activity”) in different colors over UMAP clustering (continuous scale)
 snATAC-seq of Individual Heart Cells 253

Fig. 3 Visualization of differentially accessible genes in snATAC-seq datasets. (a) Knowledge about marker
genes accessibility profiles and their distribution across nuclei in UMAP clusters allows identification of
clusters corresponding to different cell types. (b) Example of DAG visualization as a color overlay over UMAP
clustering. (c) Visualization of chromatin accessibility at a gene of interest. Coverage plots are presented that
show distribution of raw snATAC-seq reads across the genomic region in different UMAP clusters. (d) GSEA of
DAGs between cardiomyocyte nuclei and all other nuclei (Reactome database)

files are read from the respective .csv and .tsv.gz files from the
output folder of “cellranger-atac count” (or “cellranger-atac
aggr” if applicable) pipeline.
6. The Seurat-class object is created using “CreateSeuratObject”
function of the Signac/Seurat packages. The respective peak
and gene annotations are acquired from the “EnsDb.Mmuscu-
lus.v79” (use the latest available version) package. The follow-
ing QC includes the detection of nucleosome signals,
transcription starting sites (TSS) enrichment, percentages of
reads in the peaks, and others [15]. Based on QC evaluation,
outliers are removed from the dataset.
254 Michail Yekelchyk et al.

7. Next, the dataset has to undergo normalization (based on the

total number of reads per nucleus) and linear dimensional
reduction (principal component analysis (PCA); or latent
semantic indexing (LSI)) [16]. After completing linear dimen-
sional reduction, nonlinear dimensional reduction has to be
calculated (uniform manifold approximation and projection
(UMAP)).
8. After the dimensional reduction step, the composition of the
dataset and the distribution of individual nuclei can be visua-
lised using the “DimPlot” function. The resulting UMAP plot
will likely contain multiple “clouds” of single nuclei represent-
ing different cell types and/or cell states. An initial cluster
analysis can now be performed, in particular assigning common
labels to individual nuclei that share similar chromatin accessi-
bility signatures (and hence belonging to the same UMAP
“cloud”), using “FindNeighbours” and “FindClusters” func-
tions. The clusters are therefore labeled as numbers, and
many—visually— undistinguishable clusters may receive multi-
ple internal clustering labels (see Note 8).
9. After initial clustering, a gene activity matrix is calculated. The
matrix represents accessibility counts, associated with respec-
tive genes (we recommend to use only promoter regions:
TSS + 5kb upstream). This resulting matrix represents “gene
activities” or pseudo-expression. Of course, this only gives a
hint of which genes are expressed, since gene accessibility does
not accurately reflect expression.
10. After calculation of the gene activities matrix, we suggest to
create a matrix of top “active” genes for each cluster (Fig. 2a).
This allows to identify top accessibility signatures for each
nuclei cluster and subsequent labeling of clusters. Accessibility
of top “active” genes may be overlaid on the UMAP plot to aid
definition of clusters (Fig. 2b). In our murine heart dataset, we
were able to distinguish nuclei derived from cardiomyocytes,
endothelial cells, epicardial cells, fibroblasts, pericytes, and
Purkinje cells (Fig. 3a). The UMAP clusters are therefore
renamed (and merged, if necessary) accordingly (see Note 9).
11. We recommend to apply differential accessibility analysis to
obtain a list of differentially accessible genes (DAGs). This is
possible using the “MAST” package, for example [17]. You
may select the clusters of interest (e.g., cardiomyocytes vs
fibroblasts, or pericytes vs the rest of the nuclei, or “young”
vs “old” cardiomyocytes, or “control” vs “treated” endothelial
cells, etc.) to identify DAGs among the groups (Fig. 3b).
12. DAGs are visualized as violin plots (using the “VlnPlot” func-
tion), or as color overlays on UMAP plots (using the “Feature-
Plot” function). Additionally, it is possible to generate coverage
 snATAC-seq of Individual Heart Cells 255

plots, which indicate the distribution of reads across the gene

region of interest (Fig. 3c).
13. Finally, it is recommended to perform gene-set enrichment
analysis (GSEA), using the list of previously identified DAGs,
which allows to evaluate the general pattern of gene accessibil-
ity between selected clusters and/or conditions (Fig. 3d). We
suggest to use different databases as reference to generate a
more comprehensive picture of the results (Gene Ontology,
BioCyc, Reactome, PANTHER, KEGG, etc.) [18] (see Note
10).

4 Notes

1. If the focus of the study is only on a specific part of the heart

(e.g., only ventricles), parts that are not required should be
removed by dissection before freezing. Avoid repetitive freez-
ing and thawing, since each cycle disrupts chromatin
organization.
2. Note that some cardiac nuclei (from cardiomyocytes) are poly-
ploid, therefore a wide range of DAPI+ signals will occur. A
wave-like pattern is expected (with peaks for 2N, 4N, 8N, etc.
nuclei). A cutoff after 3–4 peaks is recommended, since other-
wise it is difficult to distinguish polyploid nuclei from nuclei
aggregates.
3. We suggest to aim for the maximum allowed load of the
cartridge (10k data points). In our hands, the maximum capac-
ity corresponds to 3080–7700 nuclei/μL concentration of the
loading solution.
4. In case your library does not show a “wavy” pattern, you may
sequence an aliquot at low sequencing depth (5–10 M reads) as
a further quality control test. Since the library has a unique
index, you can add a test run to a “normal” sequencing run
together with another snATAC-seq sample. Evaluate the peak
distribution and decide whether you want to proceed with the
remaining sample.
5. If you use a high capacity Illumina sequencer (e.g., a Nova-
Seq™ machine), you can combine multiple snATAC-seq
libraries in one run, since individual snATAC-seq libraries are
uniquely indexed.
6. We recommend to use a Unix-operated server with >12 cores
and >75 Gb RAM for these calculations. The longest step is
the “cellranger-atac count,” which may run for up to a few
days, depending on the number of data-points and server
processing power.
256 Michail Yekelchyk et al.

7. In comparison to scRNA-seq, where it is relatively straightfor-

ward to combine multiple runs in one dataset, such an opera-
tion is more difficult for snATAC-seq. The reason behind this is
that a transcript annotation is fixed and therefore merging of
datasets is a simple merger of matrixes, sharing the same rows
(= transcripts; of note, some datasets may have transcripts that
are not present in others). In case of snATAC-seq, accessibility
peaks of merged fragments are being re-mapped and
re-counted during the merger. Thus, new custom reference
peaks are created for the combined dataset. Due to this prob-
lem, we suggest to use the “cellranger-atac aggr” pipeline,
instead of simply merging datasets in R.
8. Automatically detected clusters often do not fully recapitulate
biological differences among cell populations in the dataset.
For example, multiple clusters might be identified inside one—
visually—homogeneous large cluster. We suggest to analyze
marker gene accessibility profiles across such potential
sub-clusters and merge them into one cluster, if all cells/nuclei
share the same marker gene accessibility landscape and do not
own significant DAGs.
9. It is important to remember that “gene activity” does not mean
“expression.” Potential gene activity should not be confused
with active transcription, which is detected by scRNA-seq.
Some genes may be accessible, but not expressed.
10. The GSEA input is a list of genes or transcripts, which will vary
according to your peak-to-gene association strategy. A broad
peak-to-gene association will affect the precision of the DAGs
lists and therefore GSEA. We suggest to use a relatively narrow
range to associate peaks to gene promoter regions (TSS + 5 kb
upstream).

Acknowledgments

This work was supported by the Excellence Initiative “Cardiopul-

monary Institute” (CPI), the DFG collaborative research center
SFB1213, Transregional Collaborative Research Centre 267, the
DFG Transregional Collaborative Research Centre 81, and the
DFG Clinical Research Unit FKO 309.

References

1. Mayran A, Drouin J (2018) Pioneer transcrip- https://doi.org/10.1038/s41576-018-

tion factors shape the epigenetic landscape. J 0089-8
Biol Chem 293(36):13795–13804. https:// 3. Buenrostro JD, Wu B, Chang HY, Greenleaf
doi.org/10.1074/jbc.R117.001232 WJ (2015) ATAC-seq: a method for assaying
2. Klemm SL, Shipony Z, Greenleaf WJ (2019) chromatin accessibility genome-wide. Curr
Chromatin accessibility and the regulatory epi- Protoc Mol Biol 109:21.29.1–21.29.9.
genome. Nat Rev Genet 20(4):207–220.
 snATAC-seq of Individual Heart Cells 257

https://doi.org/10.1002/0471142727. Greenleaf WJ, Chang HY (2019) Massively

mb2129s109 parallel single-cell chromatin landscapes of
4. Jia G, Preussner J, Chen X, Guenther S, human immune cell development and intratu-
Yuan X, Yekelchyk M, Kuenne C, Looso M, moral T cell exhaustion. Nat Biotechnol 37(8):
Zhou Y, Teichmann S, Braun T (2018) Single 925–936. https://doi.org/10.1038/s41587-
cell RNA-seq and ATAC-seq analysis of cardiac 019-0206-z
progenitor cell transition states and lineage set- 11. 10xGenomics (2021) Chromium Single Cell
tlement. Nat Commun 9(1):4877. https:// ATAC. https://www.10xgenomics.com/
doi.org/10.1038/s41467-018-07307-6 products/single-cell-atac
5. Chen X, Miragaia RJ, Natarajan KN, Teich- 12. Butler A, Hoffman P, Smibert P, Papalexi E,
mann SA (2018) A rapid and robust method Satija R (2018) Integrating single-cell tran-
for single cell chromatin accessibility profiling. scriptomic data across different conditions,
Nat Commun 9(1):5345. https://doi.org/10. technologies, and species. Nat Biotechnol
1038/s41467-018-07771-0 36(5):411–420. https://doi.org/10.1038/
6. Buenrostro JD, Giresi PG, Zaba LC, Chang nbt.4096
HY, Greenleaf WJ (2013) Transposition of 13. Chen S, Lake BB, Zhang K (2019) High-
native chromatin for fast and sensitive epige- throughput sequencing of the transcriptome
nomic profiling of open chromatin, and chromatin accessibility in the same cell.
DNA-binding proteins and nucleosome posi- Nat Biotechnol 37(12):1452–1457. https://
tion. Nat Methods 10(12):1213–1218. doi.org/10.1038/s41587-019-0290-0
https://doi.org/10.1038/nmeth.2688 14. Stuart T, Butler A, Hoffman P, Hafemeister C,
7. Scott RW, Arostegui M, Schweitzer R, Rossi Papalexi E, Mauck WM 3rd, Hao Y,
FMV, Underhill TM (2019) Hic1 defines qui- Stoeckius M, Smibert P, Satija R (2019) Com-
escent mesenchymal progenitor subpopula- prehensive integration of single-cell data. Cell
tions with distinct functions and fates in 177(7):1888–1902. e1821. https://doi.org/
skeletal muscle regeneration. Cell Stem Cell 10.1016/j.cell.2019.05.031
25(6):797–813 e799. https://doi.org/10. 15. lab S Analyzing PBMC scATAC-seq. https://
1016/j.stem.2019.11.004 satijalab.org/signac/articles/pbmc_vignette.
8. Macosko EZ, Basu A, Satija R, Nemesh J, html
Shekhar K, Goldman M, Tirosh I, Bialas AR, 16. Tsuyuzaki K, Sato H, Sato K, Nikaido I (2020)
Kamitaki N, Martersteck EM, Trombetta JJ, Benchmarking principal component analysis
Weitz DA, Sanes JR, Shalek AK, Regev A, for large-scale single-cell RNA-sequencing.
McCarroll SA (2015) Highly parallel genome- Genome Biol 21(1):9. https://doi.org/10.
wide expression profiling of individual cells 1186/s13059-019-1900-3
using nanoliter droplets. Cell 161(5): 17. Finak G, McDavid A, Yajima M, Deng J,
1202–1214. https://doi.org/10.1016/j.cell. Gersuk V, Shalek AK, Slichter CK, Miller HW,
2015.05.002 McElrath MJ, Prlic M, Linsley PS, Gottardo R
9. Gilsbach R, Schwaderer M, Preissl S, Gruning (2015) MAST: a flexible statistical framework
BA, Kranzhofer D, Schneider P, Nuhrenberg for assessing transcriptional changes and char-
TG, Mulero-Navarro S, Weichenhan D, acterizing heterogeneity in single-cell RNA
Braun C, Dressen M, Jacobs AR, Lahm H, sequencing data. Genome Biol 16:278.
Doenst T, Backofen R, Krane M, Gelb BD, https://doi.org/10.1186/s13059-015-
Hein L (2018) Distinct epigenetic programs 0844-5
regulate cardiac myocyte development and dis- 18. Subramanian A, Tamayo P, Mootha VK,
ease in the human heart in vivo. Nat Commun Mukherjee S, Ebert BL, Gillette MA,
9(1):391. https://doi.org/10.1038/s41467- Paulovich A, Pomeroy SL, Golub TR, Lander
017-02762-z ES, Mesirov JP (2005) Gene set enrichment
10. Satpathy AT, Granja JM, Yost KE, Qi Y, analysis: a knowledge-based approach for inter-
Meschi F, McDermott GP, Olsen BN, Mum- preting genome-wide expression profiles. Proc
bach MR, Pierce SE, Corces MR, Shah P, Bell Natl Acad Sci U S A 102(43):15545–15550.
JC, Jhutty D, Nemec CM, Wang J, Wang L, https://doi.org/10.1073/pnas.0506580102
Yin Y, Giresi PG, Chang ALS, Zheng GXY,
 INDEX

A H
A549 cell..............................................216, 217, 219, 222 Heart................................ 4, 26, 246, 248, 249, 252–254
Array comparative genomic hybridization Heterogeneity................................... 54, 66, 71, 120, 202
(aCGH).................. 167–172, 177–181, 184, 185
I
B
Imaging.................................................68, 132, 133, 137,
Breast cancer (BC) .................................. 43–45, 102, 120 141, 204, 207–210, 215, 216, 220–222, 234
Immunoblot .................................................................. 202
C Immunofluorescence (IF)....................................... 44, 46,
Cell heterogeneity ................................................ 120, 202 48, 50, 66, 128, 130–136
Cell type................................ 7, 26, 28, 29, 40, 189, 197, Inductively coupled plasma-mass spectrometry
227, 228, 246, 250, 252, 253 (ICP-MS) ...... 215, 216, 218, 220–222, 224, 225
CellSearch ................................ 11–15, 17–21, 28, 30–33, In situ hybridization ............................................ 127–141
35, 36, 44, 65–67, 101, 168
L
Chromatin accessibility .......................245, 246, 252, 253
Chromium 10x™................................................. 202, 249 Laser ablation .............................................. 215, 216, 218
Circulating tumor cell (CTC) .................................11–15, Liquid biopsies ..................................................... 119, 202
17–19, 21, 25, 26, 28–30, 32, 36, 38, 40, 43, 44, LncRNAs ....................................128, 132, 138, 144, 190
46–50, 65–68, 84, 101–109, 111, 112, 116, Low-pass whole genome sequencing........................... 108
119–124, 167–169, 181, 190, 202
Circulating tumor cell (CTC) clusters ................ 101–117 M
Copy number alteration (CNA)......................... 102–104, Mass spectrometry ............................................... 215–225
108–111, 116, 167–186 Mechanical dissociation ................................................ 2, 8
CTC line culture .................................................. 120, 122 Mesenchymal phenotype ................................................ 44
Metal ..................................................................... 156–157
D
Metals ................................ 144–146, 152, 156, 161, 204
DEParray ........................................................66, 168, 169 Metastasis...........................................................43, 53, 54,
Dielectrophoresis (DEP) ................................................ 12 65, 71, 102, 119, 120
Dissociation ................................ 1–8, 54, 55, 57, 58, 213 Microarray hybridization .............................................. 171
Micromanipulation ...........................................48, 54, 55,
E 59, 60, 190, 193, 197
Electrophysiology................................................. 227–242 Microscopy ......................8, 56, 128, 132, 137, 151, 234
Epithelial-mesenchymal transition (EMT) ..............43, 44 Morphology..........................................34, 38, 58, 62, 66,
233, 234, 237, 242
F
N
Flow cytometry ...................................... 8, 120–122, 124,
149–151, 153, 155, 158, 160, 161, 163, 197 Next generation sequencing (NGS) ...........................109,
116, 246, 250
G
O
Genomic analysis ................................. 103–104, 106–109
Oxaliplatin ...........................................216, 217, 219, 222

Miodrag Gužvić (ed.), Single Cell Analysis: Methods and Protocols, Methods in Molecular Biology, vol. 2752,
https://doi.org/10.1007/978-1-0716-3621-3, © Springer Science+Business Media, LLC, part of Springer Nature 2024

259
 SINGLE CELL ANALYSIS: METHODS AND PROTOCOLS
260 Index
P Single cell nucleus ATAC-seq.............................. 245–256
Single-cell RNA-sequencing (scRNA-seq) ........ 228, 242,
Patch clamp .......................................................... 227, 228 256
Patch-seq .............................................228, 229, 240, 242 Single-cell suspension .............................. 2, 5, 8, 57, 190,
PCR labeling ................................................112, 171–174 196, 203–205
Precision medicine ........................................................ 201 Single cell western blot ........................................ 201–213
Primary cells .................................................................. 1, 2 Single-molecule RNA fluorescence in situ hybridization
Proximity ligation assay for RNA (PLAYR) ....... 143–164 (smRNA FISH) ....................................... 128, 131,
Puncher......................................................................66–69 132, 134, 136–138
Stress granule (SG) .................... 127–129, 133, 138, 140
R
RNA ..................................................54, 71, 91, 143–164, T
189–197, 202, 227, 229–231, 233, 234, 237, Transcriptome analysis .................................................... 71
240, 242 Transcriptomic ................................................... 8, 48, 227
Transposase........................................................... 232, 246
S
Tumor fraction prediction............................................ 102
Sequencing .............................................................. 12, 66,
71, 72, 90, 91, 98, 102–104, 108–110, 116, 117, V
143, 189–197, 227–229, 232, 234, 237, 238, VyCAP .......................................................................66–68
240, 242, 250, 253
Single cell......................................... 2, 12, 48, 57, 66, 71, W
128, 143, 167, 190, 202, 227, 247
Single cell analysis ............... 8, 48, 66, 71, 181, 202, 216 Whole-cell recording..................................................... 227
Single cell clones .................................................. 120, 121
Z
Single cell isolation.................................. 12, 19, 193, 197
Single cell ICP-MS........................................................ 219 Zebrafish .......................................................53–57, 61, 62
Single cell multi-omics.................................................... 71