Cancer Stem Cells: Methods and Protocols (Methods in Molecular Biology, 2777) [2 ed.] 1071637290, 9781071637296

Table of contents :
Preface
Contents
Contributors
Chapter 1: Cancer Stem Cells: Current Challenges and Future Perspectives
1 Introduction
2 Challenges in CSC Research
2.1 The CSC Definition Is a Functional Trait Rather than a Specific Cell
2.2 Molecular Markers Used to Identify the CSC Do Not Specifically Nor Uniformly Mark CSCs
2.3 The Origins of CSCs
2.4 Drug Resistance
2.5 Drug Development
3 Future Perspectives
3.1 Single-Cell RNA Sequencing
3.2 Genome-Wide CRISPR Screens
3.3 Precision Medicine
3.4 Clinical Trial Design
References
Chapter 2: Immunohistochemistry for Cancer Stem Cell Detection: Principles and Methods
1 Introduction
2 Materials
2.1 Buffer, Diluents, and Antigen Retrieval Solutions
2.2 Substrates, Chromogens, and Counterstain Solutions
2.3 Endogenous Activity Blocking Solutions
3 Methods
3.1 Preanalytical Stage: Tissue Fixation and Processing of FFPE Samples
3.2 Antigen Retrieval Methods
3.3 Endogenous Activity Block
3.4 Signal Revelation Systems
3.5 Automated Immunohistochemical Systems
3.6 Advantages of the Automated Immunohistochemical Approach
3.7 Automated Immunohistochemistry Staining Protocol
3.8 Immunohistochemical Detection of Multiple Antigens
3.9 Main Markers of Cancer Stem Cells
3.9.1 Aldehyde Dehydrogenase 1A1 (ALDH1A1)
3.9.2 CD44
3.9.3 Prominin-1/CD133
3.10 Troubleshooting
3.10.1 Excessive Background Staining
3.10.2 Inadequate or No Staining of the Test Slide and Adequate Staining of the Positive Control Slide
3.10.3 Weak or No Staining of Positive Control and Weak or No Staining of Test Slides
3.10.4 No Staining of the Positive Control Slide and Adequate Staining of the Test Slide
4 Notes
5 Pitfalls in TMA Use to Detect CSCs by IHC
References
Chapter 3: Isolating Cancer Stem Cells from Solid Tumors
1 Introduction
2 Materials
2.1 Reagents
2.1.1 Cell Culture Reagents and Materials
2.1.2 Antibodies
2.1.3 Side Population Reagents
2.1.4 ALDH Activity Reagents
2.1.5 Equipment and Plates
3 Methods
3.1 Primary Cell Culture
3.1.1 Glioblastoma Cell Culture
3.1.2 Breast Cancer Cell Culture
3.1.3 Lung Cancer Cell Culture
3.1.4 Sarcoma Cell Culture
3.2 Spheres
3.2.1 Neurospheres from Glioblastoma
3.2.2 Mammospheres
3.2.3 Lung Tumorspheres
3.2.4 Sarcospheres
3.3 Detection and Isolation of CSCs by Flow Cytometry
3.3.1 Stemness Marker Expression
3.3.2 Side Population Detection
3.3.3 ALDH Activity
3.4 Flow Cytometer Setup and Data Acquisition
3.4.1 Cytometric Detection of Stemness Markers
3.4.2 Cytometric Analyses of Side Population
3.4.3 Cell Sorting
4 Notes
References
Chapter 4: Surface Markers for the Identification of Cancer Stem Cells
1 Introduction
1.1 CSC Surface Markers
1.2 Isolation and Detection
2 Materials
2.1 Cell Culture
2.2 Single-Cell Suspension
2.3 Isolation and Characterization of CSCs
2.3.1 Flow Cytometry
2.3.2 Aldefluor Assay Conjugated Flow Cytometry
2.3.3 Magnetic Separation
2.4 Detecting CSC Functions
2.4.1 Sphere-Forming Assay
2.4.2 Western Blotting
2.4.3 Immunostaining
2.4.4 Polymerase Chain Reaction (PCR)
3 Methods
3.1 Preparation of Single-Cell Suspension
3.2 Isolation and Detection
3.2.1 Flow Cytometry-Based Method
3.2.2 Aldefluor Assay Conjugated Flow Cytometry Method
3.2.3 Magnetic Separation-Based Method
3.3 In Vitro Growth Properties of CSCs
3.4 Molecular Expression of CSC Markers
3.4.1 PCR Technique
Extraction of Total RNA
cDNA Synthesis
Real-Time Polymerase Chain Reaction
3.4.2 Western Blot
3.4.3 Immunological Analysis
4 Notes
References
Chapter 5: CD44-Based Detection of CSCs: CD44 Immunodetection by Flow Cytometry
1 Introduction
2 Materials
2.1 Cell Dissociation from Tissue Using gentleMACS Dissociator
2.2 CD44 Staining for Flow Cytometry
3 Methods
3.1 Sample Preparation
3.2 CD44 Staining for Flow Cytometry
3.3 Sample Reading and Analysis
4 Notes
References
Chapter 6: ALDH Activity Assay: A Method for Cancer Stem Cell (CSC) Identification and Isolation
1 Introduction
2 Materials
2.1 Reagents
2.2 Sample Preparation
3 Methods
3.1 Aldefluor Assay
3.2 Flow Cytometer Setup and Data Acquisition
3.2.1 Prepare an Acquisition Template
3.2.2 To Set up Analyzer and Acquire Data
3.2.3 Cell Sorting
4 Notes
References
Chapter 7: In Vitro Tumorigenic Assay: A Tumor Sphere Assay for Cancer Stem Cells
1 Introduction
2 Materials
2.1 Reagents
2.2 Equipment
3 Methods
3.1 Isolation of Cell Suspensions from Tumor Tissues
3.2 Preparation of Cells from Prostate Cancer Cell Lines
3.3 Seeding Cells for Sphere Formation Assay
3.4 Propagating Spheres
4 Notes
References
Chapter 8: Multicellular Tumoroids for Investigating Cancer Stem-Like Cells in the Heterogeneous Tumor Microenvironment
1 Introduction
2 Materials
2.1 Cells and Culture Medium
2.2 General Tissue Culture
2.3 Lentiviral Transduction
2.4 Cell Labeling with CellTracker Dye
2.5 3D Hanging Drop Platform
2.6 Fluorescence-Activated Cell Sorting
2.7 RT-qPCR
2.8 Transwell Migration Assays
3 Methods
3.1 Lentiviral Transduction of Adherent Cells
3.2 Lentiviral Transduction of Nonadherent Cells
3.3 Labeling Adherent or Nonadherent Cells with Fluorescent Cell Tracker
3.4 Preparation of Hanging Drop Plates
3.5 Differentiation of U937 Monocytes to Macrophages
3.6 Generation of 3D Spheroids or Tumoroids with Single or Multiple Cell Types, Respectively
3.7 Maintenance of 3D Spheroids and Tumoroids
3.8 FACS to Separate Cell Types from Tumoroids
3.9 Gene Expression Analysis of Individual Cell Types
3.10 Flow Cytometry to Analyze Stem Markers on Cancer Cells
3.11 Flow Cytometry to Analyze Drug Response of Cancer Cells
3.12 Transwell Assays to Investigate Cancer Cell Migration
4 Notes
References
Chapter 9: Generation, Expansion, and Biobanking of Gastrointestinal Patient-Derived Organoids from Tumor and Normal Tissues
1 Introduction
2 Materials
2.1 Sample Collection and Preparation (or Handling)
2.2 Tissue Processing
2.3 Organoid Culture in BME
2.4 Organoid Passaging and Cryopreservation
2.5 Drug Sensitivity Assay of Tumor Organoids
3 Methods
3.1 Primary Tissue Processing
3.2 Organoid Establishment Protocol
3.3 Passaging and Expansion of Organoids
3.4 Cryopreservation and Thawing of Organoids
3.5 Drug Sensitivity Assay of Tumor Organoids
4 Notes
References
Chapter 10: Prostate Cancer Organoids for Tumor Modeling and Drug Screening
1 Introduction
2 Materials
2.1 Reagents
2.2 Equipment
3 Methods
3.1 Isolation of Single-Cell Suspensions from Prostate Tumor Tissues (See Fig. 1)
3.2 Preparation of Cells from Prostate Cancer Cell Lines
3.3 Seeding Cells for Organoid Formation Assay (See Fig. 1)
3.4 Propagation of Prostate Tumor Organoids (See Fig. 1)
3.5 Treatment of Prostate Tumor Organoids
4 Notes
References
Chapter 11: Mimicking the Tumor Niche: Methods for Isolation, Culture, and Characterization of Cancer Stem Cells and Multicell...
1 Introduction
1.1 Properties and Characterization of CSCs
1.2 Interplay Between TME and CSCs
2 Materials
2.1 CSC Enrichment from Cell Lines: Primary and Secondary Tumorsphere Generation
2.2 Soft Agar Colony Formation Assay
2.3 CSC Characterization by Flow Cytometry
2.3.1 Hoechst Exclusion Assay
2.3.2 CSC Marker Evaluation
2.4 In Vivo Tumorigenic Properties of CSCs (Orthotopic Limiting Dilution Assay)
2.5 Multicellular Spheroid Production
3 Methods
3.1 CSC Enrichment from Cell Lines: Primary and Secondary Tumorsphere Generation
3.2 Soft Agar Colony Formation Assay
3.3 Cancer Stem Cell Characterization by Flow Cytometry
3.3.1 Hoechst Exclusion Assay
3.3.2 Cancer Stem Cell Marker Evaluation
3.4 In Vivo Tumorigenic Properties of CSCs (Orthotopic Limiting Dilution Assay)
3.5 Multicellular Spheroids
3.5.1 Cell Culture
3.5.2 Formation of Multicellular Spheroids Following the Hanging-Drop Protocol
4 Notes
References
Chapter 12: Detection of Cancer Stem Cells in Normal and Dysplastic/Leukemic Human Blood
1 Introduction
2 Materials
3 Methods
3.1 RBC Lysis and Sample Preparation
3.2 Direct Staining Procedure
3.3 Flow Cytometric Acquisition and Parameter Setting
3.4 Blast Phenotyping
4 Notes
References
Chapter 13: Methods to Study the Role of Mechanical Signals in the Induction of Cancer Stem Cells
1 Introduction
2 Materials
2.1 Hydrogel Preparations
2.2 Isolation and Seeding of Primary Mammary Luminal Cells
3 Methods
3.1 Hydrogel Preparations
3.2 Mammary Gland Dissociation and Isolation of Luminal Differentiated Cells
3.3 Seeding and Infection of Mammary LD Cells
3.4 Clonogenic Assay
4 Notes
References
Chapter 14: Co-Delivery Polymeric Poly(Lactic-Co-Glycolic Acid) (PLGA) Nanoparticles to Target Cancer Stem-Like Cells
1 Introduction
1.1 Cancer Stem Cells Show Increased Chemoresistance, Metastasis, and ALDH Activity due to PDGF Signaling with Mesenchymal Ste...
1.2 Nanoparticles for Co-Delivery of Sunitinib and Paclitaxel
2 Materials
2.1 Nanoparticle Synthesis and Collection
2.1.1 Materials and Solvents
2.1.2 Supplies
2.2 2D and 3D Cell Culture
2.2.1 Cells and Culture Medium
2.2.2 General Tissue Culture
3 Methods
3.1 Nanoparticle Synthesis and Collection
3.1.1 Electrospray Setup
3.1.2 Electrospray Solution Formulation
3.1.3 Nanoparticle Production
3.1.4 Nanoparticle Collection
3.2 Nanoparticle Characterization
3.2.1 Scanning Electron Microscopy (SEM): Nanoparticle Size and Morphology
3.2.2 Nanoparticle Tracking Analysis (NTA): Nanoparticle Size, Polydispersity, and Quantity
3.2.3 Liquid Chromatography-Mass Spectrometry (LCMS): Drug Loading of Nanoparticles
3.3 Nanoparticle Drug Delivery and Cell Assays
3.3.1 Hetero-Spheroid Generation and Co-Delivery of Paclitaxel and Sunitinib
4 Notes
References
Chapter 15: Isolating Circulating Cancer Stem Cells (CCSCs) from Human Whole Blood
1 Introduction
2 Materials
2.1 Antibodies, Cells, and Kits
2.2 Equipment and Disposable Plastic/Glassware
3 Methods
3.1 Procurement of Human Whole Blood Samples
3.2 Depletion of Red Blood Cells and White Blood Cells from Human Whole Blood Samples
3.3 Analyzing Residual Cells in the Final Elution from the Human Blood Samples
3.4 Verifying Efficient Depletion Process from Human Blood Samples
3.5 Confirming Presence of CCSCs in the Final Elution
4 Notes
References
Chapter 16: Generation of Cancer Stem Cells by Co-Culture Methods
1 Introduction
2 Materials
2.1 Co-Culture of Cancer Cells with TAMs
2.2 Co-Culture of Cancer Cells with CAFs
3 Methods
3.1 Co-Culture of Cancer Cells with TAMs
3.1.1 Cell Culture
3.1.2 Differentiation of THP-1 to M1 TAMs Using External Stimuli in a Co-Culture of THP-1 and Cancer Cells
3.1.3 In Vitro Generation of Monocyte-Derived Macrophages (MDMs) Using THP-1 Cells
3.1.4 Enrichment of the CSCs by Incubating Cancer Cells with the CM
3.2 Co-Culture of Cancer Cells with CAFs
3.2.1 Cell Culture
3.2.2 Activation of CAFs in HFB Cells Using External Stimuli
3.2.3 Preparation of Chemokine-Enriched Conditioned Medium (CM)
3.2.4 Generation of CSCs by Co-Culturing CAFs and Cancer Cells
4 Notes
References
Chapter 17: Deep Learning of Cancer Stem Cell Morphology
1 Introduction
2 Materials
2.1 Typical PC Hardware, OS, and Software Framework
2.2 Microscopy and CSC Culture
3 Methods
3.1 Cell Image Acquisition from Cell Culture
3.2 Image Processing for pix2pix
3.3 Construction of an AI Model Predicting CSC in Phase Contrast Cell Image
3.3.1 Basic Concepts of AI Models
3.3.2 GAN, Conditional GAN, and pix2pix
3.4 Evaluation of CSC Image Created Using an AI Model
3.5 Cancer Stem Cell Image Classification and Visualization
3.5.1 Overview
3.5.2 Installation
3.5.3 Sample Data
3.5.4 Usage
4 Notes
References
Index

Citation preview

Methods in Molecular Biology 2777

Federica Papaccio Gianpaolo Papaccio  Editors

Cancer Stem Cells Methods and Protocols Second Edition

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.

Cancer Stem Cells Methods and Protocols Second Edition

Edited by

Federica Papaccio Department of Medicine, Surgery and Dentistry “Scuola Medica Salernitana”, University of Salerno, Baronissi, Italy

Gianpaolo Papaccio Department of Experimental Medicine, University of Campania “Luigi Vanvitelli”, Naples, Italy

Editors Federica Papaccio Department of Medicine, Surgery and Dentistry “Scuola Medica Salernitana” University of Salerno Baronissi, Italy

Gianpaolo Papaccio Department of Experimental Medicine University of Campania “Luigi Vanvitelli” Naples, Italy

ISSN 1064-3745 ISSN 1940-6029 (electronic) Methods in Molecular Biology ISBN 978-1-0716-3729-6 ISBN 978-1-0716-3730-2 (eBook) https://doi.org/10.1007/978-1-0716-3730-2 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This 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 This second edition of the book Cancer Stem Cells: Methods and Protocols follows the great success of its first edition, published in 2018. In this second edition, a great effort has been made to provide readers with all the scientific advances in methods and protocols related to cancer stem cells. They will, therefore, find both the most important updated chapters from the previous edition and new chapters on hot topics. The aim of the book is always to try to be a comprehensive collection of all methods, protocols, and procedures used for the identification, characterization, and selection of cancer stem cells. The new chapters focus on the latest technologies that have improved our knowledge in this field. Among the newly included chapters, we feature contributions on organoids, machine learning, nanoparticles, and other recent advances. Regarding the topics of the new chapters, we believe that organoids, in particular, represent a great challenge, and the related techniques are of great importance for cancer stem cells thanks to the many applications they can have. At this point, it is very important to be rigorous and understand the difference between spheroids and organoids. We hope that the hard work we put into producing this book will be useful to scientists who want to read it. Baronissi, Italy Naples, Italy

Federica Papaccio Gianpaolo Papaccio

v

Contents Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Contributors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

v ix

1 Cancer Stem Cells: Current Challenges and Future Perspectives . . . . . . . . . . . . . . 1 Muhammad Vaseem Shaikh, Stefan Custers, Alisha Anand, Petar Miletic, Chitra Venugopal, and Sheila K. Singh 2 Immunohistochemistry for Cancer Stem Cell Detection: Principles and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Giuseppa Zannini, Renato Franco, and Federica Zito Marino 3 Isolating Cancer Stem Cells from Solid Tumors . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Vitale Del Vecchio, Marcella La Noce, and Virginia Tirino 4 Surface Markers for the Identification of Cancer Stem Cells . . . . . . . . . . . . . . . . . . 51 Tasfik Ul Haque Pronoy, Farhadul Islam, Vinod Gopalan, and Alfred King-yin Lam 5 CD44-Based Detection of CSCs: CD44 Immunodetection by Flow Cytometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 Lornella Seeneevassen, Anissa Zaafour, and Christine Varon 6 ALDH Activity Assay: A Method for Cancer Stem Cell (CSC) Identification and Isolation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 Vitale Del Vecchio, Marcella La Noce, and Virginia Tirino 7 In Vitro Tumorigenic Assay: A Tumor Sphere Assay for Cancer Stem Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 Amani Yehya, Hisham Bahmad, and Wassim Abou-Kheir 8 Multicellular Tumoroids for Investigating Cancer Stem-Like Cells in the Heterogeneous Tumor Microenvironment . . . . . . . . . . . . . . . . . . . . . . 99 Kathleen M. Burkhard and Geeta Mehta 9 Generation, Expansion, and Biobanking of Gastrointestinal Patient-Derived Organoids from Tumor and Normal Tissues. . . . . . . . . . . . . . . . . 123 Manuel Cabeza-Segura, Blanca Garcia-Mico, Andre´s Cervantes, and Josefa Castillo 10 Prostate Cancer Organoids for Tumor Modeling and Drug Screening. . . . . . . . . 135 Amani Yehya, Fatima Ghamlouche, Sana Hachem, and Wassim AbouKheir 11 Mimicking the Tumor Niche: Methods for Isolation, Culture, and Characterization of Cancer Stem Cells and Multicellular Spheroids. . . . . . . . 145 ˜ a, Cristiano Farace, Andrea Pisano, Laura De Lara-Pen Julia Lopez de Andre´s, Grazia Fenu, Federica Etzi, ˜ a´n-Lison, Juan Antonio Marchal, and Roberto Madeddu Carmen Grin

vii

viii

12

13

14

15

16 17

Contents

Detection of Cancer Stem Cells in Normal and Dysplastic/Leukemic Human Blood . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Alessia De Stefano, Alessandra Cappellini, Irene Casalin, Stefania Paolini, Sarah Parisi, Maria Vittoria Marvi, Antonietta Fazio, Irene Neri, Foteini-Dionysia Koufi, Stefano Ratti, Carlo Finelli, Antonio Curti, Lucia Manzoli, Lucio Cocco, and Matilde Y. Follo Methods to Study the Role of Mechanical Signals in the Induction of Cancer Stem Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Alessandro Gandin, Paolo Contessotto, and Tito Panciera Co-Delivery Polymeric Poly(Lactic-Co-Glycolic Acid) (PLGA) Nanoparticles to Target Cancer Stem-Like Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . Catherine S. Snyder, Taylor Repetto, Kathleen M. Burkhard, Anish Tuteja, and Geeta Mehta Isolating Circulating Cancer Stem Cells (CCSCs) from Human Whole Blood . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Carla Kantara and Pomila Singh Generation of Cancer Stem Cells by Co-Culture Methods . . . . . . . . . . . . . . . . . . . Biswajit Das and Chanakya Nath Kundu Deep Learning of Cancer Stem Cell Morphology . . . . . . . . . . . . . . . . . . . . . . . . . . . Hiroyuki Kameda, Hiroaki Ishihata, and Tomoyasu Sugiyama

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

163

177

191

205 219 231 257

Contributors WASSIM ABOU-KHEIR • Department of Anatomy, Cell Biology, and Physiological Sciences, Faculty of Medicine, American University of Beirut, Beirut, Lebanon ALISHA ANAND • Department of Biochemistry and Biomedical Sciences, McMaster University, Hamilton, ON, Canada HISHAM BAHMAD • Arkadi M. Rywlin M.D. Department of Pathology and Laboratory Medicine, Mount Sinai Medical Center, Miami Beach, FL, USA KATHLEEN M. BURKHARD • Department of Biomedical Engineering, University of Michigan, Ann Arbor, MI, USA MANUEL CABEZA-SEGURA • Department of Medical Oncology, Hospital Clı´nico Universitario de Valencia, INCLIVA, Biomedical Research Institute, University of Valencia, Valencia, Spain ALESSANDRA CAPPELLINI • Cellular Signalling Laboratory, Department of Biomedical and Neuromotor Sciences, University of Bologna, Bologna, Italy IRENE CASALIN • Cellular Signalling Laboratory, Department of Biomedical and Neuromotor Sciences, University of Bologna, Bologna, Italy JOSEFA CASTILLO • Department of Medical Oncology, Hospital Clı´nico Universitario de Valencia, INCLIVA, Biomedical Research Institute, University of Valencia, Valencia, Spain; Centro de Investigacion Biome´dica en Red (CIBERONC), Instituto de Salud Carlos III, Madrid, Spain; Department of Biochemistry and Molecular Biology, Universitat de Valencia, Burjassot, Spain ANDRE´S CERVANTES • Department of Medical Oncology, Hospital Clı´nico Universitario de Valencia, INCLIVA, Biomedical Research Institute, University of Valencia, Valencia, Spain; Centro de Investigacion Biome´dica en Red (CIBERONC), Instituto de Salud Carlos III, Madrid, Spain LUCIO COCCO • Cellular Signalling Laboratory, Department of Biomedical and Neuromotor Sciences, University of Bologna, Bologna, Italy PAOLO CONTESSOTTO • Department of Molecular Medicine, University of Padua, Padua, Italy ANTONIO CURTI • IRCCS – Azienda Ospedaliero-Universitaria di Bologna, Istituto di ` gnoli”, Bologna, Italy Ematologia “Sera STEFAN CUSTERS • Department of Biochemistry and Biomedical Sciences, McMaster University, Hamilton, ON, Canada BISWAJIT DAS • Cancer Biology Division, School of Biotechnology, Kalinga Institute of Industrial Technology (KIIT)-Deemed to be University, Bhubaneswar, Odisha, India JULIA LO´PEZ DE ANDRE´S • Biopathology and Regenerative Medicine Institute (IBIMER), Centre for Biomedical Research (CIBM), University of Granada, Granada, Spain; Instituto de Investigacion Biosanitaria ibs.GRANADA, University Hospitals of GranadaUniversity of Granada, Granada, Spain; Excellence Research Unit ‘Modeling Nature’ (MNat), University of Granada, Granada, Spain; Department of Human Anatomy and Embryology, Faculty of Medicine, University of Granada, Granada, Spain; BioFab i3D Lab- Biofabrication and 3D (bio)printing Singular Laboratory, University of Granada, Granada, Spain

ix

x

Contributors

LAURA DE LARA-PEN˜A • Biopathology and Regenerative Medicine Institute (IBIMER), Centre for Biomedical Research (CIBM), University of Granada, Granada, Spain; Instituto de Investigacion Biosanitaria ibs.GRANADA, University Hospitals of GranadaUniversity of Granada, Granada, Spain; Excellence Research Unit ‘Modeling Nature’ (MNat), University of Granada, Granada, Spain; Department of Human Anatomy and Embryology, Faculty of Medicine, University of Granada, Granada, Spain; BioFab i3D Lab- Biofabrication and 3D (bio)printing Singular Laboratory, University of Granada, Granada, Spain ALESSIA DE STEFANO • Cellular Signalling Laboratory, Department of Biomedical and Neuromotor Sciences, University of Bologna, Bologna, Italy VITALE DEL VECCHIO • Department of Experimental Medicine, Histology and Embryology Section, University of Campania “L. Vanvitelli”, Naples, Italy FEDERICA ETZI • Department of Biomedical Sciences-Histology, University of Sassari, Sassari, Italy; I.N.B.B. National Institute of Biostructure and Biosystem, Rome, Italy CRISTIANO FARACE • Department of Biomedical Sciences-Histology, University of Sassari, Sassari, Italy; I.N.B.B. National Institute of Biostructure and Biosystem, Rome, Italy ANTONIETTA FAZIO • Cellular Signalling Laboratory, Department of Biomedical and Neuromotor Sciences, University of Bologna, Bologna, Italy GRAZIA FENU • Department of Biomedical Sciences-Histology, University of Sassari, Sassari, Italy CARLO FINELLI • IRCCS – Azienda Ospedaliero-Universitaria di Bologna, Istituto di ` gnoli”, Bologna, Italy Ematologia “Sera MATILDE Y. FOLLO • Cellular Signalling Laboratory, Department of Biomedical and Neuromotor Sciences, University of Bologna, Bologna, Italy RENATO FRANCO • Pathology Unit, Dipartimento di Salute Mentale, Fisica e Medicina ` degli Studi della Campania ‘Luigi Vanvitelli’, Naples, Italy Preventiva, Universita ALESSANDRO GANDIN • Department of Industrial Engineering (DII), University of Padua, Padua, Italy BLANCA GARCIA-MICO´ • Department of Medical Oncology, Hospital Clı´nico Universitario de Valencia, INCLIVA, Biomedical Research Institute, University of Valencia, Valencia, Spain FATIMA GHAMLOUCHE • Department of Anatomy, Cell Biology, and Physiological Sciences, Faculty of Medicine, American University of Beirut, Beirut, Lebanon VINOD GOPALAN • Cancer Molecular Pathology, School of Medicine and Dentistry, Griffith University, Gold Coast, QLD, Australia CARMEN GRIN˜A´N-LISO´N • Biopathology and Regenerative Medicine Institute (IBIMER), Centre for Biomedical Research (CIBM), University of Granada, Granada, Spain; Instituto de Investigacion Biosanitaria ibs.GRANADA, University Hospitals of GranadaUniversity of Granada, Granada, Spain; Excellence Research Unit ‘Modeling Nature’ (MNat), University of Granada, Granada, Spain; BioFab i3D Lab- Biofabrication and 3D (bio)printing Singular Laboratory, University of Granada, Granada, Spain; Department of Biochemistry and Molecular Biology 2, School of Pharmacy, University of Granada, Granada, Spain; GENYO, Centre for Genomics and Oncological Research, Pfizer/University of Granada/Andalusian Regional Government, Granada, Spain SANA HACHEM • Department of Anatomy, Cell Biology, and Physiological Sciences, Faculty of Medicine, American University of Beirut, Beirut, Lebanon HIROAKI ISHIHATA • School of Computer Science, Tokyo University of Technology, Tokyo, Japan

Contributors

xi

FARHADUL ISLAM • Department of Biochemistry and Molecular Biology, University of Rajshahi, Rajshahi, Bangladesh HIROYUKI KAMEDA • School of Bioscience and Biotechnology, Tokyo University of Technology, Tokyo, Japan; School of Computer Science, Tokyo University of Technology, Tokyo, Japan CARLA KANTARA • Departments of Neuroscience and Cell Biology and Biochemistry and Molecular Biology, The University of Texas Medical Branch, Galveston, TX, USA FOTEINI-DIONYSIA KOUFI • Cellular Signalling Laboratory, Department of Biomedical and Neuromotor Sciences, University of Bologna, Bologna, Italy CHANAKYA NATH KUNDU • Cancer Biology Division, School of Biotechnology, Kalinga Institute of Industrial Technology (KIIT)-Deemed to be University, Bhubaneswar, Odisha, India MARCELLA LA NOCE • Department of Experimental Medicine, Histology and Embryology Section, University of Campania “L. Vanvitelli”, Naples, Italy ALFRED KING-YIN LAM • Cancer Molecular Pathology, School of Medicine and Dentistry, Griffith University, Gold Coast, QLD, Australia ROBERTO MADEDDU • Department of Biomedical Sciences-Histology, University of Sassari, Sassari, Italy; I.N.B.B. National Institute of Biostructure and Biosystem, Rome, Italy LUCIA MANZOLI • Cellular Signalling Laboratory, Department of Biomedical and Neuromotor Sciences, University of Bologna, Bologna, Italy JUAN ANTONIO MARCHAL • Biopathology and Regenerative Medicine Institute (IBIMER), Centre for Biomedical Research (CIBM), University of Granada, Granada, Spain; Instituto de Investigacion Biosanitaria ibs.GRANADA, University Hospitals of GranadaUniversity of Granada, Granada, Spain; Excellence Research Unit ‘Modeling Nature’ (MNat), University of Granada, Granada, Spain; Department of Human Anatomy and Embryology, Faculty of Medicine, University of Granada, Granada, Spain; BioFab i3D Lab- Biofabrication and 3D (bio)printing Singular Laboratory, University of Granada, Granada, Spain MARIA VITTORIA MARVI • Cellular Signalling Laboratory, Department of Biomedical and Neuromotor Sciences, University of Bologna, Bologna, Italy GEETA MEHTA • Department of Biomedical Engineering, University of Michigan, Ann Arbor, MI, USA; Department of Materials Science and Engineering, University of Michigan, Ann Arbor, MI, USA; Macromolecular Science and Engineering, University of Michigan, Ann Arbor, MI, USA; Rogel Cancer Center, University of Michigan, Ann Arbor, MI, USA; Precision Health, University of Michigan, Ann Arbor, MI, USA GEETA MEHTA • Department of Materials Science and Engineering, University of Michigan, Ann Arbor, MI, USA; Department of Biomedical Engineering, University of Michigan, Ann Arbor, MI, USA; Macromolecular Sciences and Engineering, University of Michigan, Ann Arbor, MI, USA; Rogel Cancer Center, University of Michigan, Ann Arbor, MI, USA; Precision Health, University of Michigan, Ann Arbor, MI, USA PETAR MILETIC • Department of Surgery, McMaster University, Hamilton, ON, Canada IRENE NERI • Cellular Signalling Laboratory, Department of Biomedical and Neuromotor Sciences, University of Bologna, Bologna, Italy TITO PANCIERA • Department of Molecular Medicine, University of Padua, Padua, Italy STEFANIA PAOLINI • IRCCS – Azienda Ospedaliero-Universitaria di Bologna, Istituto di ` gnoli”, Bologna, Italy Ematologia “Sera SARAH PARISI • IRCCS – Azienda Ospedaliero-Universitaria di Bologna, Istituto di ` gnoli”, Bologna, Italy Ematologia “Sera

xii

Contributors

ANDREA PISANO • Department of Biomedical Sciences-Histology, University of Sassari, Sassari, Italy TASFIK UL HAQUE PRONOY • Department of Biochemistry and Molecular Biology, University of Rajshahi, Rajshahi, Bangladesh STEFANO RATTI • Cellular Signalling Laboratory, Department of Biomedical and Neuromotor Sciences, University of Bologna, Bologna, Italy TAYLOR REPETTO • Department of Materials Science and Engineering, University of Michigan, Ann Arbor, MI, USA LORNELLA SEENEEVASSEN • INSERM U1312 BRIC Bordeaux Institute of Oncology, University of Bordeaux, Bordeaux, France MUHAMMAD VASEEM SHAIKH • Department of Surgery, McMaster University, Hamilton, ON, Canada POMILA SINGH • Departments of Neuroscience and Cell Biology and Biochemistry and Molecular Biology, The University of Texas Medical Branch, Galveston, TX, USA SHEILA K. SINGH • Department of Surgery, McMaster University, Hamilton, ON, Canada; Department of Biochemistry and Biomedical Sciences, McMaster University, Hamilton, ON, Canada CATHERINE S. SNYDER • Department of Materials Science and Engineering, University of Michigan, Ann Arbor, MI, USA TOMOYASU SUGIYAMA • School of Bioscience and Biotechnology, Tokyo University of Technology, Tokyo, Japan VIRGINIA TIRINO • Department of Experimental Medicine, Histology and Embryology Section, University of Campania “L. Vanvitelli”, Naples, Italy ANISH TUTEJA • Department of Materials Science and Engineering, University of Michigan, Ann Arbor, MI, USA; Macromolecular Sciences and Engineering, University of Michigan, Ann Arbor, MI, USA; Department of Chemical Engineering, University of Michigan, Ann Arbor, MI, USA; Biointerfaces Institute, University of Michigan, Ann Arbor, MI, USA CHRISTINE VARON • INSERM U1312 BRIC Bordeaux Institute of Oncology, University of Bordeaux, Bordeaux, France CHITRA VENUGOPAL • Department of Surgery, McMaster University, Hamilton, ON, Canada AMANI YEHYA • Department of Anatomy, Cell Biology, and Physiological Sciences, Faculty of Medicine, American University of Beirut, Beirut, Lebanon ANISSA ZAAFOUR • INSERM U1312 BRIC Bordeaux Institute of Oncology, University of Bordeaux, Bordeaux, France GIUSEPPA ZANNINI • Pathology Unit, Dipartimento di Salute Mentale, Fisica e Medicina ` degli Studi della Campania ‘Luigi Vanvitelli’, Naples, Italy Preventiva, Universita FEDERICA ZITO MARINO • Pathology Unit, Dipartimento di Salute Mentale, Fisica ` degli Studi della Campania ‘Luigi Vanvitelli’, Naples, e Medicina Preventiva, Universita Italy

Chapter 1 Cancer Stem Cells: Current Challenges and Future Perspectives Muhammad Vaseem Shaikh, Stefan Custers, Alisha Anand, Petar Miletic, Chitra Venugopal, and Sheila K. Singh Abstract Despite major advances in health care including improved diagnostic tools, robust chemotherapeutic regimens, advent of precision, adjuvant and multimodal therapies, there is a major proportion of patients that still go on to experience tumor progression and recurrence. Cancer stem cells (CSCs) are shown to be responsible for tumor persistence and relapse. This subpopulation of cancer cells possess normal stem cell like traits of self-renewal, proliferation, and multilineage differentiation. Currently, they are isolated and enriched based on the cell surface markers that can be detected and sorted through fluorescence and magnetic-based cell sorting. In this chapter, we review the current challenges and limitations often encountered in CSC research, including the identification of universal markers, therapy resistance, and new drug development. Current and future perspectives are discussed to address these challenges including utilization of cutting-edge technologies such as next-generation sequencing to elucidate the genome, epigenome, and transcriptome on a single-cell level and genome-wide CRISPR-Cas9 screens to identify novel pathway-based targeted therapies. Further, we discuss the future of precision medicine and the need for the improvement of clinical trial designs. Key words Cancer stem cells, Tumor initiating cells, Standard of care, Intratumoral heterogeneity, Cell surface makers, Fluorescent activated cell sorting, CRISPR-Cas9 system, Single cell RNA sequencing, Minimal residual disease.

1 Introduction Cancer is the second leading cause of death worldwide, accounting for nearly 10 million deaths in 2020 [1]. Although there have been significant improvements in the prognosis of patients due to advancements in diagnostic tools, intensive chemotherapeutic regimens, and the advent of precision and multimodal therapies [2–4], most patients are faced with relapse and disease

Muhammad Vaseem Shaikh and Stefan Custers contributed equally with all other contributors. Federica Papaccio and Gianpaolo Papaccio (eds.), Cancer Stem Cells: Methods and Protocols, Methods in Molecular Biology, vol. 2777, https://doi.org/10.1007/978-1-0716-3730-2_1, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

1

2

Muhammad Vaseem Shaikh et al.

progression and ultimately death. Moreover, there have been little to no improvements in prognosis for some advanced and aggressive cancers [5]. This therapeutic failure can be partly attributed to extensive intratumoral heterogeneity (ITH) present at the genetic, transcriptomic, and functional levels [6, 7]. Several studies have shown correlation between tumor heterogeneity and poorer clinical outcomes [8–10]. ITH was traditionally explained by the clonal evolution model, wherein carcinogenesis was believed to be a result of sequential accumulation of mutations and the subsequent gradual selection of clones under various attractor states [11]. Recently, several studies indicate heterogeneity to be driven by distinct subpopulations of cells commonly referred as cancer stem cells (CSCs) that possess traits of normal stem cells, such as multilineage differentiation, self-renewal, and therapy evasion [12]. CSCs are believed to adapt and survive the intensive standard of care (SoC) regimen, become chemoradiotherapy-resistant, and seed the formation of therapy-resistant fulminant relapse [13]. CSCs are also referred to as tumor-initiating cells (TICs), as they are believed to initiate tumorigenesis in a hierarchical manner, with CSCs at the apex and rapidly proliferating cells forming the bulk of the tumor [14]. In 1997, Bonnet and Dick were the first to isolate and characterize cancer stem cell subpopulations from acute myeloid leukemia (AML) patients [15]. They demonstrated that a finite number of cells termed leukemic stem cells (LSCs), characterized by the presence of cell surface phenotype CD34+CD38-, when injected into immunodeficient mice were able to proliferate and differentiate, giving rise to disease similar to the patient donor. They were able to demonstrate the self-renewal capacity of CSCs by serial transplantation of cells into secondary recipient mice, resulting in re-establishment of AML. Moreover, they demonstrated an enhanced tumorigenic ability of these cells by limiting dilution assay. Later, Al-Hajj et al. demonstrated the presence of CSCs in solid tumors, when they demonstrated CD44+/CD24- cells in breast cancer tumors had enhanced tumorigenic capacity [16]. CSCs have since then been characterized in several cancers, viz., brain, lung, liver, pancreatic, and melanoma [17–21]. Understanding the biology of CSCs is instrumental for the development of novel therapeutics that can target CSCs. In doing so we recognize the challenges that accompany CSC research which include the origins of CSCs, inconsistent markers, drug resistance, and drug development. To circumvent these barriers, novel tools and techniques such as single-cell RNA sequencing, genome-wide CRISPR-Cas9 screens, and modifications in precision medicine and clinical trial designs are being incorporated into CSC research.

Cancer Stem Cells: Current Challenges and Future Perspectives

2

3

Challenges in CSC Research The concept of the CSC has progressed from being a highly speculative and intensively debated concept to a fairly defined and better understood biological entity in many tumor types, including AML, breast, brain, colon, lung, and other cancers. Despite the advances made in understanding CSCs, many challenges lie ahead and will be discussed below.

2.1 The CSC Definition Is a Functional Trait Rather than a Specific Cell

CSCs by definition are a subset of tumor cells that contain both self-renewal and multilineage differentiation capacities that fuel tumor initiation, treatment resistance, and recurrence [22]. This definition is based on the shared functional tumor-initiation trait among CSCs. Considering that CSC phenotypes are complex, vary between tumors, and are affected by abnormalities arising from neoplastic transformation, defining CSCs outside the scope of the shared functional trait of tumor initiation can be difficult [23]. Importantly, this functional CSC definition can make data interpretation complex as the gold standard assay for CSC identification is in vivo serial transplantation, a functional read-out which can underestimate total CSC numbers and has a bias towards more malignant CSC types. Moreover, serial implantation risks isolating artifacts and may limit true heterogeneity of the original tumor. Due to this limiting effect a more refined definition has been proposed: CSCs are tumor cells that can be proven to generate long-term progeny, either by direct engraftment or genetic tracing [22].

2.2 Molecular Markers Used to Identify the CSC Do Not Specifically Nor Uniformly Mark CSCs

Historically, CSCs have been isolated by Hoechst staining, utilizing the ATP-binding cassette (ABC) transporter protein family, which was later replaced and standardized by fluorescence-activated cell sorting (FACS) based on cell surface markers. Although markerbased CSC isolation is easier to perform, the markers used are not always reliable, leading to challenges in CSC identification and data interpretation [22]. First, the markers currently used do not specifically identify CSCs, as these markers play central roles in normal stem cell biology and are also expressed in other normal cell types. For some markers, CSC specificity can be achieved by targeting splice variants. For example, full-length CD144 is widely expressed, while the CD44v6 splice variant is more indicative of CSCs [24–26]. The use of marker combinations has alleviated this problem to some extent; however, the lack of specificity remains a hurdle for reliable CSC detection [22, 23]. On top of this, markers may have roles in other mechanisms, which may be favorably selected for during in vivo serial transplantation. For example, CD44, CD90, and CD34 are involved in cell adhesion and attachment and are thus

4

Muhammad Vaseem Shaikh et al.

favorably selected for during in vivo engraftment. Together these observations indicate that markers that support the CSC phenotype need to be identified [27]. Second, CSC markers are not universal for CSC identification due to inter- and intratumoral heterogeneity, as CSC biology differs substantially both within tumors and among cancer types [22, 23]. Proper use of existing markers is highly tumor dependent; for example, ALDH1 seems to be a reliable CSC marker in certain cancers such as breast cancer [28], but its validity is debated in other cancer types such as liver and pancreatic cancer [29]. However, there is evidence indicating that deletion of ALDH1 enhances stemness in breast cancer, making its validity as a CSC marker highly contested only in certain contexts [30]. The identification of consensus markers, capable of uniformly detecting CSCs across and within tumors, is warranted to improve CSC research. Third, CSC markers are not universally applicable, as antibodies targeting CSC markers may not be suitable for every research method [22]. This is the case for CD133, which can be used to sort CSCs by FACS, but its suitability for mRNA- or immunohistochemistry (IHC)-based applications is limited [31, 32]. Fourth, current CSC markers may not be constitutively expressed. CD133 is an example of a marker that is downregulated during the G0/G1 phase of the cell cycle [33]. Additionally, interand intratumor heterogeneity may make markers unreliable as they might be downregulated, potentially leaving certain subclones undetectable [22], as is the case for CD133 inactivation by CpG methylation in colorectal cancer and glioblastoma (GBM) [34]. In conclusion, consensus markers are needed that are functionally linked to stemness mechanisms and assure reliable and uniform expression on CSCs. Research into enzymatic and signaling mechanisms is a starting point on the functional CSC road of discovery. Additional work should be performed to identify cell surface markers that are selectively induced by signaling pathways involved in normal stem cell biology (Wnt, TGF-b, Hedgehog). Stringent validation in human primary tumor samples is warranted for these potential CSC markers. It is expected that these markers will be different between tumor types and potentially between clones [22]. 2.3 The Origins of CSCs

One of the most debated topics of the CSC is its cell of origin. Tumors are the result of genetic and epigenetic alterations accumulated over many years. The presence of stem cell markers seems indicative of a stem cell origin. Another possibility is that CSCs arise from transit-amplifying/progenitor cells that increase mutational burden during multistep tumorigenesis and introduce their acquired mutations into corresponding stem cell pools via a process of spontaneous dedifferentiation [23, 35, 36].

Cancer Stem Cells: Current Challenges and Future Perspectives

5

Realizing these two hypotheses may not exclude one another, the rigid hierarchical CSC model has progressed into the more flexible model of phenotypic reversion. Phenotypic reversion indicates that there is a balance between CSCs and differentiated cell states which can interconvert and hereby link stochastic and hierarchical properties [22]. Stemness is acquired via CSC intrinsic changes in gene expression programs and underlying regulatory mechanisms of selfrenewal and differentiation. On top of these intrinsic mechanisms, studies propose a prominent role of the cancer microenvironment, or CSC niche, generating feed-forward and feed-backward mechanisms. For example, gliomas can be initiated from terminally differentiated neurons by reverting back into the CSC state, a mechanism stimulated by Notch secreting endothelial cells [37, 38]. Additionally, studies suggest that epithelial-to-mesenchymal transitioning (EMT) signaling can induce CSCs from differentiated tumor cells [39]. Not all CSCs express EMT markers [40] and EMT signaling factors in some cases can reduce CSC qualities [41]. Molecular understanding of how CSC-intrinsic and environmental signaling dictate CSC hierarchies and how they govern effective interconversion between tumor states remains limited. In summary, current models capable of capturing microenvironmental interactions with CSCs remain suboptimal. Molecular understanding of these mechanisms will provide further insights in self-renewal and differentiation cues and thus tumor cell hierarchy dependencies on the CSC niche. Single cells would need to be followed in vitro and in vivo using genetic tracing and live imaging. Lineage tracing animal models are needed to determine whether CSC reversion occurs in vivo and to assess its frequency of occurrence. Moreover, this would help understand to what extent differentiation becomes irreversible. 2.4

Drug Resistance

A major hurdle in cancer treatment is CSC-mediated treatment resistance to SoC and initiation of relapse. These cells are causatively correlated with poor prognosis, treatment resistance, and subsequent relapse. Unraveling these mechanisms and effective targeting of CSCs will be a decisive victory in the treatment of cancer as a whole [42]. CSCs are mostly quiescent and abundantly express ABC2 transporter proteins and anti-apoptotic proteins. Together these properties aid in evading apoptosis and drug-induced cytotoxicity when compared to differentiated tumor cells [23, 43–45]. As described above, research has indicated the involvement of (partial) EMT programs in the initiation and sustainment of CSCs, and secreted EMT signaling molecules from stromal cells have been shown to induce chemo- and radiotherapy resistance [46–48]. Further insights into treatment resistance are warranted for more efficient therapeutic targeting of CSCs.

6

Muhammad Vaseem Shaikh et al.

2.5 Drug Development

3

Another challenge is the development of drugs targeting CSCs. Currently, there are two approaches considered: the direct targeting of CSCs and the targeting of the CSC niche. Targeting CSCs is dependent on its genetics and environment, resulting in context-dependent signaling pathways and unique challenges per tumor type. Drug screens can be technically difficult due to the rarity of CSCs in most tumors and limited means of propagating them in vitro. Means to circumvent these issues include the use of enrichment media used for brain cancers such as medulloblastoma and GBM [17, 49]. Another method that has been employed is genetically enforcing an EMT-like state in epithelial tumors [23]. Heterotypical signals arising from tumor-associated stroma are often responsible for activating EMT. Targeting paracrine signaling pathways that activate and sustain CSC programs should provide insights for targeting CSCs. There have been successes with antibodies targeting tropical signals. For example, macrophage M-CSF receptor antibodies have been developed and effectively reduced macrophage homing tumors in syngeneic mouse models [50]. In addition, stimulating paracrine signaling molecules favoring non-CSCs could be used to push the CSC/non-CSC balance towards the non-CSC side. Derivates of molecules such as secreted frizzled-related protein 1 (SFRP) and Dickkopf WNT signaling pathway inhibitor 1 (DKK) could be developed and employed to induce SoC susceptibility [23, 40]. CSCs exist in a niche formed by multiple cell types and distinct signaling molecules (growth factors, cytokines, extracellular matrix components). Studying biological properties of CSCs in isolation or performing screens in the absence of a surrounding relevant niche is unlikely to yield translatable results in vivo. Currently, in vivo models depend on immunodeficient mice or syngeneic mouse models. Immunodeficient models lack a full microenvironment as the adaptive immune system is lacking and syngeneic settings are not fully translational as they are artificially induced by knocking out one or two genes. Neither model recapitulates the clinical setting to a full extent and are therefore limited in translating therapies to the clinic.

Future Perspectives To overcome the challenges encountered in CSC research, an enormous amount of work is being done to better characterize the CSCs, understand resistance mechanisms, develop precision medicines and multimodal therapy regimens, and enhance clinical trial design (Fig. 1).

Cancer Stem Cells: Current Challenges and Future Perspectives

7

Fig. 1 Schematic representation of target discovery for the development of CSC-specific therapies 3.1 Single-Cell RNA Sequencing

Indeed, next-generation sequencing (NGS) technologies have helped us to sequence large amounts of DNA and RNA and have led to a better understanding of cancer biology and intertumoral heterogeneity [51]. However, cancer is a heterogenous ecosystem comprising bulk cancer cells, tumor immune microenvironment (TIME), and CSCs [52], and bulk sequencing obscures the accurate understanding of all the different cellular subpopulations [53]. Rapidly evolving methods of single-cell RNA sequencing (scRNA seq), which can profile genome, epigenome, and transcriptome of an individual cell, can help us better elucidate the drivers of tumor progression, metastasis, and therapy resistance [54, 55]. scRNA seq technology not only illuminates intratumoral heterogeneity, but also highlights the plasticity and dynamic nature of

8

Muhammad Vaseem Shaikh et al.

various cellular states. Neftel et al. profiled glioma stem cells (GSCs) using scRNA seq and classified GBM into four major subtypes: oligodendrocyte progenitor cells (OPC-like), astrocytes (AC-like), neural progenitor cells (NPC-like), and mesenchymal (MES-like) [7]. They discovered that a single GBM tumor comprises multiple cellular states in stark contrast to a singular cellular state as initially believed [7]. Through lineage tracing experiments, they further discovered that when cells of a particular type are engrafted in mice, they not only form the tumor of that specific state but re-establish the heterogeneity seen in the primary tumor through lineage tracing experiments [7]. Additionally, when they investigated the commonly known GSC marker and transcription factors, they found each of the four states was enriched for specific markers; for example, CD133 was enriched in OPC-like and CD44 was enriched in MES-like state [7]. Additionally, scRNA seq has aided in profiling circulating tumor cells, identifying prognostic markers, understanding the TIME, and studying tumor resistance mechanisms, thereby providing a new perspective in CSC research. Aceto et al. [56] identified oligoclonal circulating tumor clusters to have increased propensity for breast-to-lung metastasis, and abundance of circulating tumor cells (CTC) correlated with poorer outcomes. scRNA seq helped identify the cell junction component plakoglobin as differentially expressed, and knockdown of plakoglobin led to significant reduction in metastasis. Lei et al. [57] employed scRNA seq to profile breast cancer stem cells (BCSCs) isolated from two HER2+ breast cancer patients and compared the transcriptomic profile with breast cancer cells, mammary cells, and CD44+ mammary cells. They were able to identify 404 genes differentially expressed in BCSCs, and additionally, they identified CA12 as a novel prognostic marker in HER2+ breast cancer patients. Chung et al. [58] performed scRNA seq in 11 patients with primary breast cancer. At a single-cell resolution, the immune population was characterized in three distinct clusters of T lymphocytes, B lymphocytes, and macrophages. T lymphocyte and macrophages displayed T cell regulatory and M2 phenotype, respectively, which promotes an immunosuppressive and pro-tumorigenic environment. To understand resistance, Zhang et al. carried out scRNA seq in samples before and after treatment in high-grade serous ovarian cancer. They identified a stress-associated genetic signature present before treatment which was subsequently enriched posttreatment and is correlated with poor patient prognosis. In addition to stress-associated genetic signature genes associated with stemness, pro-survival and pro-inflammation were also enriched, further protecting the cells from chemotherapy [59]. Research over the past decade has concentrated on characterizing CSCs as their elimination might translate to clinical benefit. scRNA seq has helped to deconvolute the CSC theory, ITH, drug

Cancer Stem Cells: Current Challenges and Future Perspectives

9

resistance mechanisms, and identify prognostic markers. The vast amount of data generated can be used to identify vulnerabilities of CSCs and develop targeted therapies or multimodal therapies to synchronously attack the susceptible targets. 3.2 Genome-Wide CRISPR Screens

Comprehensive genetic studies in the past two decades have emphasized and helped us understand correlations between the disease phenotype and the mutations seen in cancers; however, gene editing technologies that could allow us to study the driving forces of cancer and correct mutations in the genome were needed to drive further breakthroughs in precision medicine. Discovery of zinc finger nucleases (ZFNs) [60], TALE domains in transcription activator-like effector nucleases (TALENs) [61], and CRISPR/Cas (clustered regularly interspaced palindromic repeats/CRISPRassociated) technology have allowed for the precise editing of the eukaryotic genome. The CRISPR/Cas system has revolutionized the gene editing technology due to its ability to precisely manipulate genes even if they are silenced or in a dense location and have allowed for ease of design, affordability, and scalability compared to other techniques [62, 63]. With the advent of genome-wide lossof-function (LOF) and gain-of-function (GOF) single-guide RNA (sgRNA) libraries, it is now possible to screen and discover new therapeutic vulnerabilities in CSCs. LOF screens are carried out by using genome-wide knockdown libraries, in the presence or absence of therapies in order to identify the fitness genes and biological processes that make the cells resistant. Shalem et al. [64] constructed a genome-wide CRISPR/Cas9 knockout library (GecKO) and employed it to mice for genes that confer resistance to a human melanoma cell line when treated with vemurafenib. Hart et al. [65] developed an improved CRISPR lossof-function library known as the Toronto KnockOut (TKO) library targeting over 18,000 genes, identifying ~2000 fitness genes which is 4–5 times more than RNAi screens. Chen et al. [66] employed a genome-wide CRISPR LOF screen (mGeCKOa) comprising 67,405 sgRNAs targeting 20,611 protein-coding genes and 1175 microRNA precursors to study the evolution of cancer and metastasis in a mouse model. LOF screens have been routinely used to discover novel and adjuvant therapeutic targets as well as studying the TIME. CRISPR LOF screen conducted on CSCs from patient-derived colon cancer lines identified 44 essential genes including genes involved in the cholesterol biosynthesis pathway. The authors validated these findings by demonstrating enhanced antitumor efficacy in vivo when treated with combinatorial treatment with cholesterol biosynthesis inhibitors and 5-FU [67]. Wei et al. [68] discovered loss of Regnase-1 led to increased fitness in CD8 + Tcells, prolonging their persistence and their ability to infiltrate and fight cancer cells. Recently, CRISPR Cas9 screens carried out on GSCs

10

Muhammad Vaseem Shaikh et al.

identified SOX transcription family, SOCS3, USP8, and DOT1L, and protein UFMylation as essential genes responsible for temozolomide resistance [69]. Xu et al. [70] identified that NMDAR as an adjuvant therapeutic target with tyrosine kinase inhibitors (TKIs) in hepatocellular carcinoma (HCC), co-treatment with an NMDAR antagonist, ifenprodil, with TKI, sorafenib, suppressed genes associated with the WNT signaling pathway and stemness and reduced self-renewal ability and proliferation of HCC cells. In contrast CRISPR activation (CRISPRa) screens are a potent tool for testing gain of function of endogenous genes by transcriptional upregulation of target genes through the selective activation using transcriptional activators targeting the promoter sites of target genes guided by sgRNA and a dead CAS9 system [71, 72]. CRISPRa screens have tremendous potential in the elucidation of drug resistance mechanisms, which are normally believed to be gain-of-function events in response to therapy. Recently, a reciprocal CRISPRa and CRIPR LOF screen was carried out on GSCs and chimeric antigen receptor-T (CAR-T) revealing RELA and NPLOC4 to be essential in GSCs for resistance against immunotherapy and also elucidated that the knockout of TLE4 and IKZF2 led to superior effector function and reduced the exhaustion of CAR-Ts [73]. Moreover, CRISPRa screens carried out on pancreatic ductal adenocarcinoma lines identified a cellular resistance mechanism to gemcitabine, 5-fluorouracil, irinotecan, and oxaliplatin, revealing overexpression of ABCG2, a drug efflux pump, as a common drug resistance mechanism to all the four drugs [74]. Taken together these studies suggest the immense promise for CRISPR technology in the development of new precision oncolytics by identifying novel targets and facilitating existing therapies through the elucidation of resistance mechanisms. 3.3 Precision Medicine

Pathway-based medicines have been developed using our deeper understanding of cancer biology from the genomic, proteomic, and functional data generated from CRISPR screens and single-cell sequencing. These therapies could not only target aberrant cellular populations but also help to spare healthy cells. Imatinib, which targets gain-of-function mutations in the genes encoding plateletderived growth factor receptor-α (PDGFRα) in gastrointestinal stromal tumors, was the first clinical success story for targeted cancer treatment [75]. Since then, several targeted therapies have been designed against CSCs owing to the increased clinical significance of targeting these functionally important cells. Hedgehog signaling pathway expression is correlated with chemoresistance, metastatic spread, and CSC maintenance. Hh regulates the target gene expression through the transmembrane protein Smoothened (Smo) [76]. Glasdegib, vismodegib, and sonidegib showed significant

Cancer Stem Cells: Current Challenges and Future Perspectives

11

activity in locally advanced and metastatic basal cell carcinoma and AML and were approved by the US Food and Drug Administration (FDA) [77, 78]. MK-0752 a γ-secretase inhibitor that targets the Notch signaling pathway, which is implicated for maintenance of CSCs [79], showed clinical benefit in refractory central nervous tumors in pediatric patients [80]. Gao et al. used CRISPR LOF screens on CSCs that identified 44 genes in the cholesterol biosynthesis pathway to be essential for stemness and chemoresistance in colon cancer. They also further demonstrated that lovastatin and 5-flurouracil combination treatment showed significant preclinical efficacy [67]. However, clinical approval of new drugs is a lengthy and an expensive process, on average taking 13 years of research and 2–3 billion USD for a drug to reach from bench to market [81]. Hence, recently, drug repurposing (“teaching old dogs new tricks”) is receiving lot of traction in the research community, wherein the strategy is to use next-generation biological data and leverage it for the usage of inhibitors that have already received regulatory approval. This pathway greatly reduces development costs and time as well as alleviates business risks as safety, toxicity, and pharmacokinetic profiles of the investigational drug have already been established, allowing drugs to be directly fast-tracked to Phase II and III of the clinical trial process [82]. Machine learning and data mining are vital to tap into the unprecedented resources of approved drugs. Connectivity Mapping (CMap) and the Library of Integrated Network-based Cellular Signatures (LINCS) are extensive databases of aberrant gene expression signatures and putative drugs that could potentially have an inhibitory or potentiating effect on these genetic profiles [83, 84]. These tools have greatly facilitated and accelerated the process for identification of new molecules. Using CMap, Singh et al. identified apomorphine, a dopamine antagonist, to prevent brain metastasis in mouse models by targeting premetastasis-associated genes KIF16B, SEPW1, and TESK2 in BTICs [85]. Recently, a clinical trial conducted in patients with pancreatic cancer showed significant survival benefit in patients treated with metformin, a common antidiabetic drug [86]. This strategy has also shown clinical benefit in elderly non-fit AML patients, and the addition of all-trans retinoic acid to decitabine showed significant higher remission rates as compared to the single treatment arm [87]. These studies highlight the window of opportunity that drug repurposing has opened. Carcinogenesis is believed to be strongly related to a dysfunctional immune system as well as immune surveillance [88]. Cancer’s cellular landscape evolves with cancer progression and cancer cells remodel the TIME to their advantage [89], while the dysregulated immune system activates stem cell signaling pathways responsible for therapy resistance, antigen escape, and recurrence

12

Muhammad Vaseem Shaikh et al.

[89, 90]. Based on our understanding of the TIME, over the past decade, novel immunotherapeutic strategies such as immune checkpoint inhibitors and CAR-T cell therapy have emerged. In a first of its kind study, Vora et al. carried out a head-to-head comparison of three immunotherapeutic modalities (antibodies, bispecific T cell engagers [BITEs], and CAR-T) targeting CD133 on GSCs and found the CAR-T-based treatment modality to be most efficacious [91]. Several immune checkpoint inhibitors including CTLA4 (ipilimumab) [92], PD-1 (cemiplimab) [93], and PDL-1 (avelumab) [94] have been evaluated clinically in various cancers and have shown efficacy. CD19 and CD22 CAR-T cell therapy have shown clinical success in liquid tumors [95–97] and several trials testing their efficacy are underway for solid tumors (NCT02915445, NCT02465983, NCT03585764). 3.4 Clinical Trial Design

Improved therapeutic regimens have allowed complete remission in patients with most cancers and thereby improving prognosis [98, 99]. Complete remission is defined as reduction in the tumor burden by 99% as measured by advanced radiological techniques (PET, CT, or MRI) [100]. However, small therapy-resistant subpopulations of tumor cells survive aggressive treatment regimens to drive tumor recurrence [101, 102]. These cells are traditionally called “minimal residual disease” (MRD) [103]. Recent studies indicate that tumor recurrence is not driven by innate mutational mechanisms [104, 105], rather by enrichment of a small population of cells (CSCs) that are inherently resistant to anticancer therapy [102, 106]. There are several conceptual advantages of treating patients at MRD rather than waiting for clinical relapse to initiate further therapy: first, the fact that cancer which is characterized by extensive ITH will be at its lowest subclonal complexity and most homogeneous state which would increase the likelihood of designing a tailored drug, targeting a rare minority of CSCs enriched at MRD [107]. Additionally, due to the decrease in tumor burden there is also reduction in the ability of the cells to remodel the surrounding environment, or reprogram the TIME leading to a chemoprotective milieu [108]. Moreover, at MRD patients are clinically well and generally off SoC therapy and hence more likely to tolerate any potential therapeutic side effects compared to being treated at relapse [109]. Identification, characterization, validation, and developing targeted therapies against MRD will allow for the development of precision treatment modalities that would most likely improve patient outcomes and may lead to a potential cure [110]. Unfortunately, it comes with a new set of clinical and technical challenges in sampling, especially in solid tumors wherein repeated invasive procedures would lead to patient discomfort and reduce clinical applicability [111]. To overcome this situation there is a need to develop

Cancer Stem Cells: Current Challenges and Future Perspectives

13

in vitro and in vivo models that would recapitulate human disease. Recently, Qazi et al. [109] used a unique model of human GBM recurrence along with barcoding technology to track the BTIC clonal population through SoC revealing that MRD can be driven by either prior fitness of the subclones or prior equipotency resulting in selection of subclones under therapeutic selection pressure. Additionally, they profiled GBM temporally by bulk sequencing and scRNA-seq which could be used for potential target mining as well as predicting and anticipating recurrence. Also, several patient-derived xenograft models have been used to study MRD for breast cancer, AML, ALL, and melanomas. Monitoring patients at MRD using iterative sampling especially in liquid cancers is being used as a standard practice to tailor the anticipation of relapse and to refine clinical decisions [112]. Detection of MRD above or below a threshold is used as a tool for escalating or tapering off therapy, though still it is not being used to select targeted therapies to truly exploit the vulnerabilities of MRD. Additionally, there are several studies indicating the use of adjuvant therapies after complete remission resulting in improved clinical outcomes [113–115]. Clinical trials in colorectal cancer are underway wherein MRD is being used to inform which patients should be enrolled for an adjuvant therapy. Targeting CSCs at MRD appears to be a tractable strategy for efficiently eliminating the surviving CSCs after SoC and preventing relapse.

References 1. Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A et al (2021) Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 71(3):209–249 2. Skaane P, Gullien R, Bjørndal H, Eben EB, Ekseth U, Haakenaasen U et al (2012) Digital breast tomosynthesis (DBT): initial experience in a clinical setting. Acta Radiol 53(5): 524–529 3. Romond EH, Perez EA, Bryant J, Suman VJ, Geyer CE, Davidson NE et al (2005) Trastuzumab plus adjuvant chemotherapy for operable HER2-positive breast cancer. N Engl J Med 353(16):1673–1684 4. Baselga J, Corte´s J, Kim SB, Im SA, Hegg R, Im YH et al (2012) Pertuzumab plus Trastuzumab plus Docetaxel for metastatic breast cancer. N Engl J Med 366(2):109–119 5. Oberstein PE, Olive KP (2013) Pancreatic cancer: why is it so hard to treat? Ther Adv Gastroenterol 6(4):321–337

6. El-Sayes N, Vito A, Mossman K (2021) Tumor heterogeneity: a great barrier in the age of cancer immunotherapy. Cancers 13(4):806 7. Neftel C, Laffy J, Filbin MG, Hara T, Shore ME, Rahme GJ et al (2019) An integrative model of cellular states, plasticity and genetics for glioblastoma. Cell 178(4):835–849.e21 8. Jamal-Hanjani M, Quezada SA, Larkin J, Swanton C (2015) Translational implications of tumor heterogeneity. Clin Cancer Res 21(6):1258–1266 9. Patel AP, Tirosh I, Trombetta JJ, Shalek AK, Gillespie SM, Wakimoto H et al (2014) Single-cell RNA-seq highlights intratumoral heterogeneity in primary glioblastoma. Science 344(6190):1396–1401 10. Mroz EA, Rocco JW (2013) MATH, a novel measure of intratumor genetic heterogeneity, is high in poor-outcome classes of head and neck squamous cell carcinoma. Oral Oncol 49(3):211–215

14

Muhammad Vaseem Shaikh et al.

11. Greaves M, Maley CC (2012) Clonal evolution in cancer. Nature 481(7381):306–313 12. Steinbichler TB, Duda´s J, Skvortsov S, Ganswindt U, Riechelmann H, Skvortsova II (2018) Therapy resistance mediated by cancer stem cells. Semin Cancer Biol 53:156–167 13. Baisiwala S, Hall RR, Saathoff MR, Shireman J, Park C, Budhiraja S et al (2020) LNX1 modulates Notch1 signaling to promote expansion of the glioma stem cell population during Temozolomide therapy in glioblastoma. Cancers 12(12):E3505 14. Walcher L, Kistenmacher AK, Suo H, Kitte R, Dluczek S, Strauß A et al (2020) Cancer stem cells—origins and biomarkers: perspectives for targeted personalized therapies. Front Immunol 11:1280 15. Bonnet D, Dick JE (1997) Human acute myeloid leukemia is organized as a hierarchy that originates from a primitive hematopoietic cell. Nat Med 3(7):730–737 16. Al-Hajj M, Wicha MS, Benito-Hernandez A, Morrison SJ, Clarke MF (2003) Prospective identification of tumorigenic breast cancer cells. Proc Natl Acad Sci 100(7):3983–3988 17. Singh SK, Hawkins C, Clarke ID, Squire JA, Bayani J, Hide T et al (2004) Identification of human brain tumour initiating cells. Nature 432(7015):396–401 18. Eramo A, Lotti F, Sette G, Pilozzi E, Biffoni M, Di Virgilio A et al (2008) Identification and expansion of the tumorigenic lung cancer stem cell population. Cell Death Differ 15(3):504–514 19. Ma S, Chan KW, Hu L, Lee TKW, Wo JYH, Ng IOL et al (2007) Identification and characterization of tumorigenic liver cancer stem/ progenitor cells. Gastroenterology 132(7): 2542–2556 20. Li C, Heidt DG, Dalerba P, Burant CF, Zhang L, Adsay V et al (2007) Identification of pancreatic cancer stem cells. Cancer Res 67(3):1030–1037 21. Schatton T, Murphy GF, Frank NY, Yamaura K, Waaga-Gasser AM, Gasser M et al (2008) Identification of cells initiating human melanomas. Nature 451(7176): 345–349 22. Medema JP (2013) Cancer stem cells: the challenges ahead. Nat Cell Biol 15(4): 338–344 23. Pattabiraman DR, Weinberg RA (2014) Tackling the cancer stem cells – what challenges do they pose? Nat Rev Drug Discov 13(7): 497–512 24. Dalerba P, Dylla SJ, Park IK, Liu R, Wang X, Cho RW et al (2007) Phenotypic

characterization of human colorectal cancer stem cells. Proc Natl Acad Sci U S A 104(24):10158–10163 25. Al-Hajj M, Wicha MS, Benito-Hernandez A, Morrison SJ, Clarke MF (2003) Prospective identification of tumorigenic breast cancer cells. Proc Natl Acad Sci U S A 100(7): 3983–3988 26. Snyder EL, Bailey D, Shipitsin M, Polyak K, Loda M (2009) Identification of CD44v6+/ CD24- breast carcinoma cells in primary human tumors by quantum dot-conjugated antibodies. Lab Investig 89(8):857–866 27. Kemper K, Grandela C, Medema JP (2010) Molecular identification and targeting of colorectal cancer stem cells. Oncotarget 1(6):387–395 28. Ginestier C, Hur MH, Charafe-Jauffret E, Monville F, Dutcher J, Brown M et al (2007) ALDH1 is a marker of normal and malignant human mammary stem cells and a predictor of poor clinical outcome. Cell Stem Cell 1(5):555–567 29. Deng S, Yang X, Lassus H, Liang S, Kaur S, Ye Q et al (2010) Distinct expression levels and patterns of stem cell marker, aldehyde dehydrogenase isoform 1 (ALDH1), in human epithelial cancers. PLoS One 5(4):e10277 30. Ginestier C, Wicinski J, Cervera N, Monville F, Finetti P, Bertucci F et al (2009) Retinoid signaling regulates breast cancer stem cell differentiation. Cell Cycle 8(20): 3297–3302 31. Kemper K, Sprick MR, de Bree M, Scopelliti A, Vermeulen L, Hoek M et al (2010) The AC133 epitope, but not the CD133 protein, is lost upon cancer stem cell differentiation. Cancer Res 70(2):719–729 32. Mak AB, Blakely KM, Williams RA, Penttil€a PA, Shukalyuk AI, Osman KT et al (2011) CD133 protein N-glycosylation processing contributes to cell surface recognition of the primitive cell marker AC133 epitope. J Biol Chem 286(47):41046–41056 33. Sun Y, Kong W, Falk A, Hu J, Zhou L, Pollard S et al (2009) CD133 (Prominin) negative human neural stem cells are clonogenic and tripotent. PLoS One 4(5):e5498 34. Yi JM, Tsai HC, Glo¨ckner SC, Lin S, Ohm JE, Easwaran H et al (2008) Abnormal DNA methylation of CD133 in colorectal and glioblastoma tumors. Cancer Res 68(19): 8094–8103 35. Chaffer CL, Brueckmann I, Scheel C, Kaestli AJ, Wiggins PA, Rodrigues LO et al (2011) Normal and neoplastic nonstem cells can spontaneously convert to a stem-like state.

Cancer Stem Cells: Current Challenges and Future Perspectives Proc Natl Acad Sci U S A 108(19): 7950–7955 36. Gupta PB, Fillmore CM, Jiang G, Shapira SD, Tao K, Kuperwasser C et al (2011) Stochastic state transitions give rise to phenotypic equilibrium in populations of cancer cells. Cell 146(4):633–644 37. Borovski T, Verhoeff JJC, ten Cate R, Cameron K, de Vries NA, van Tellingen O et al (2009) Tumor microvasculature supports proliferation and expansion of gliomapropagating cells. Int J Cancer 125(5): 1222–1230 38. Friedmann-Morvinski D, Bushong EA, Ke E, Soda Y, Marumoto T, Singer O et al (2012) Dedifferentiation of neurons and astrocytes by oncogenes can induce gliomas in mice. Science 338(6110):1080–1084 39. Vermeulen L, De Sousa E, Melo F, van der Heijden M, Cameron K, de Jong JH, Borovski T et al (2010) Wnt activity defines colon cancer stem cells and is regulated by the microenvironment. Nat Cell Biol 12(5): 468–476 40. Scheel C, Eaton EN, Li SHJ, Chaffer CL, Reinhardt F, Kah KJ et al (2011) Paracrine and autocrine signals induce and maintain mesenchymal and stem cell states in the breast. Cell 145(6):926–940 41. Bruna A, Greenwood W, Le Quesne J, Teschendorff A, Miranda-Saavedra D, Rueda OM et al (2012) TGFβ induces the formation of tumour-initiating cells in claudinlow breast cancer. Nat Commun 3(1):1055 42. Chen J, Li Y, Yu TS, McKay RM, Burns DK, Kernie SG et al (2012) A restricted cell population propagates glioblastoma growth after chemotherapy. Nature 488(7412):522–526 43. Roesch A, Fukunaga-Kalabis M, Schmidt EC, Zabierowski SE, Brafford PA, Vultur A et al (2010) A temporarily distinct subpopulation of slow-cycling melanoma cells is required for continuous tumor growth. Cell 141(4): 583–594 44. Grimm M, Krimmel M, Polligkeit J, Alexander D, Munz A, Kluba S et al (2012) ABCB5 expression and cancer stem cell hypothesis in oral squamous cell carcinoma. Eur J Cancer 48(17):3186–3197 45. Moitra K, Lou H, Dean M (2011) Multidrug efflux pumps and cancer stem cells: insights into multidrug resistance and therapeutic development. Clin Pharmacol Ther 89(4): 491–502 46. Gupta PB, Onder TT, Jiang G, Tao K, Kuperwasser C, Weinberg RA et al (2009) Identification of selective inhibitors of cancer

15

stem cells by high-throughput screening. Cell 138(4):645–659 47. Farmer P, Bonnefoi H, Anderle P, Cameron D, Wirapati P, Becette V et al (2009) A stroma-related gene signature predicts resistance to neoadjuvant chemotherapy in breast cancer. Nat Med 15(1):68–74 48. Kurrey NK, Jalgaonkar SP, Joglekar AV, Ghanate AD, Chaskar PD, Doiphode RY et al (2009) Snail and slug mediate radioresistance and chemoresistance by antagonizing p53-mediated apoptosis and acquiring a stem-like phenotype in ovarian cancer cells. Stem Cells 27(9):2059–2068 49. Venugopal C, Hallett R, Vora P, Manoranjan B, Mahendram S, Qazi MA et al (2015) Pyrvinium targets CD133 in human glioblastoma brain tumor–initiating cells. Clin Cancer Res 21(23):5324–5337 50. Sica A, Allavena P, Mantovani A (2008) Cancer related inflammation: the macrophage connection. Cancer Lett 267(2):204–215 51. Garziera M, Roncato R, Montico M, De Mattia E, Gagno S, Poletto E et al (2019) New challenges in tumor mutation heterogeneity in advanced ovarian cancer by a targeted next-generation sequencing (NGS) approach. Cell 8(6):E584 52. Sathe A, Grimes SM, Lau BT, Chen J, Suarez C, Huang RJ et al (2020) Single-cell genomic characterization reveals the cellular reprogramming of the gastric tumor microenvironment. Clin Cancer Res 26(11): 2640–2653 53. Sun G, Li Z, Rong D, Zhang H, Shi X, Yang W et al (2021) Single-cell RNA sequencing in cancer: applications, advances, and emerging challenges. Mol Ther Oncolytics 21:183–206 54. Tang F, Barbacioru C, Wang Y, Nordman E, Lee C, Xu N et al (2009) mRNA-Seq wholetranscriptome analysis of a single cell. Nat Methods 6(5):377–382 55. Navin N, Kendall J, Troge J, Andrews P, Rodgers L, McIndoo J et al (2011) Tumour evolution inferred by single-cell sequencing. Nature 472(7341):90–94 56. Aceto N, Bardia A, Miyamoto DT, Donaldson MC, Wittner BS, Spencer JA et al (2014) Circulating tumor cell clusters are oligoclonal precursors of breast cancer metastasis. Cell 158(5):1110–1122 57. Lei B, Zhang XY, Zhou JP, Mu GN, Li YW, Zhang YX et al (2016) Transcriptome sequencing of HER2-positive breast cancer stem cells identifies potential prognostic marker. Tumour Biol 37(11):14757–14764

16

Muhammad Vaseem Shaikh et al.

58. Chung W, Eum HH, Lee HO, Lee KM, Lee HB, Kim KT et al (2017) Single-cell RNA-seq enables comprehensive tumour and immune cell profiling in primary breast cancer. Nat Commun 8:15081 59. Zhang K, Erkan EP, Jamalzadeh S, Dai J, Andersson N, Kaipio K et al (2022) Longitudinal single-cell RNA-seq analysis reveals stress-promoted chemoresistance in metastatic ovarian cancer. Sci Adv 8(8):eabm1831 60. Bibikova M, Beumer K, Trautman JK, Carroll D (2003) Enhancing gene targeting with designed zinc finger nucleases. Science 300(5620):764–764 61. Boch J, Scholze H, Schornack S, Landgraf A, Hahn S, Kay S et al (2009) Breaking the code of DNA binding specificity of TAL-type III effectors. Science 326(5959):1509–1512 62. Knight SC, Xie L, Deng W, Guglielmi B, Witkowsky LB, Bosanac L et al (2015) Dynamics of CRISPR-Cas9 genome interrogation in living cells. Science 350(6262):823–826 63. Zhang F, Wen Y, Guo X (2014) CRISPR/ Cas9 for genome editing: progress, implications and challenges. Hum Mol Genet 23 (R1):R40–R46 64. Shalem O, Sanjana NE, Hartenian E, Shi X, Scott DA, Mikkelson T et al (2014) Genomescale CRISPR-Cas9 knockout screening in human cells. Science 343(6166):84–87 65. Hart T, Chandrashekhar M, Aregger M, Steinhart Z, Brown KR, MacLeod G et al (2015) High-resolution CRISPR screens reveal fitness genes and genotype-specific cancer liabilities. Cell 163(6):1515–1526 66. Chen S, Sanjana NE, Zheng K, Shalem O, Lee K, Shi X et al (2015) Genome-wide CRISPR screen in a mouse model of tumor growth and metastasis. Cell 160(6): 1246–1260 67. Gao S, Soares F, Wang S, Wong CC, Chen H, Yang Z et al (2021) CRISPR screens identify cholesterol biosynthesis as a therapeutic target on stemness and drug resistance of colon cancer. Oncogene 40(48):6601–6613 68. Wei J, Long L, Zheng W, Dhungana Y, Lim SA, Guy C et al (2019) Targeting REGNASE1 programs long-lived effector T cells for cancer therapy. Nature 576(7787):471–476 69. MacLeod G, Bozek DA, Rajakulendran N, Monteiro V, Ahmadi M, Steinhart Z et al (2019) Genome-wide CRISPR-Cas9 screens expose genetic vulnerabilities and mechanisms of Temozolomide sensitivity in glioblastoma stem cells. Cell Rep 27(3):971–986.e9 70. Xu F, Tong M, Tong CSW, Chan BKC, Chu HY, Wong TL et al (2021) A combinatorial

CRISPR–Cas9 screen identifies ifenprodil as an adjunct to sorafenib for liver cancer treatment. Cancer Res 81(24):6219–6232 71. Decker CE, Young T, Pasnikowski E, Chiu J, Song H, Wei Y et al (2019) Genome-scale CRISPR activation screen uncovers tumorintrinsic modulators of CD3 bispecific antibody efficacy. Sci Rep 9(1):20068 72. Joung J, Kirchgatterer PC, Singh A, Cho JH, Nety SP, Larson RC et al (2022) CRISPR activation screen identifies BCL-2 proteins and B3GNT2 as drivers of cancer resistance to T cell-mediated cytotoxicity. Nat Commun 13(1):1606 73. Wang D, Prager BC, Gimple RC, Aguilar B, Alizadeh D, Tang H et al (2021) CRISPR screening of CAR T cells and cancer stem cells reveals critical dependencies for cellbased therapies. Cancer Discov 11(5): 1192–1211 74. Ramaker RC, Hardigan AA, Gordon ER, Wright CA, Myers RM, Cooper SJ (2021) Pooled CRISPR screening in pancreatic cancer cells implicates co-repressor complexes as a cause of multiple drug resistance via regulation of epithelial-to-mesenchymal transition. BMC Cancer 21(1):632 75. Dagher R, Cohen M, Williams G, Rothmann M, Gobburu J, Robbie G et al (2002) Approval summary: imatinib mesylate in the treatment of metastatic and/or unresectable malignant gastrointestinal stromal tumors. Clin Cancer Res 8(10):3034–3038 76. Zhao Y, Tong C, Jiang J (2007) Hedgehog regulates smoothened activity by inducing a conformational switch. Nature 450(7167): 252–258 77. Sekulic A, Migden MR, Oro AE, Dirix L, Lewis KD, Hainsworth JD et al (2012) Efficacy and safety of vismodegib in advanced basal-cell carcinoma. N Engl J Med 366(23): 2171–2179 78. Norsworthy KJ, By K, Subramaniam S, Zhuang L, Del Valle PL, Przepiorka D et al (2019) FDA approval summary: glasdegib for newly diagnosed acute myeloid leukemia. Clin Cancer Res 25(20):6021–6025 79. Quail DF, Taylor MJ, Postovit LM (2012) Microenvironmental regulation of cancer stem cell phenotypes. Curr Stem Cell Res Ther 7(3):197–216 80. Hoffman LM, Fouladi M, Olson J, Daryani VM, Stewart CF, Wetmore C et al (2015) Phase I trial of weekly MK-0752 in children with refractory central nervous system malignancies: a pediatric brain tumor consortium study. Childs Nerv Syst 31(8):1283–1289

Cancer Stem Cells: Current Challenges and Future Perspectives 81. Su M, Zhang Q, Bai X, Wu C, Li Y, Mossialos E et al (2017) Availability, cost, and prescription patterns of antihypertensive medications in primary health care in China: a nationwide cross-sectional survey. Lancet 390(10112): 2559–2568 82. Maxmen A (2016) Busting the billion-dollarmyth: how to slash the cost of drug development. Nature 536(7617):388–390 83. Lamb J, Crawford ED, Peck D, Modell JW, Blat IC, Wrobel MJ et al (2006) The connectivity map: using gene-expression signatures to connect small molecules, genes, and disease. Science 313(5795):1929–1935 84. Vidovic´ D, Koleti A, Schu¨rer SC (2014) Large-scale integration of small moleculeinduced genome-wide transcriptional responses, Kinome-wide binding affinities and cell-growth inhibition profiles reveal global trends characterizing systems-level drug action. Front Genet 5:342 85. Singh M, Venugopal C, Tokar T, McFarlane N, Subapanditha MK, Qazi M et al (2018) Therapeutic targeting of the premetastatic stage in human lung-to-brain metastasis. Cancer Res 78(17):5124–5134 86. Chaiteerakij R, Petersen GM, Bamlet WR, Chaffee KG, Zhen DB, Burch PA et al (2016) Metformin use and survival of patients with pancreatic cancer: a cautionary lesson. JCO 34(16):1898–1904 87. Lu¨bbert M, Grishina O, Schmoor C, Schlenk RF, Jost E, Crysandt M et al (2020) Valproate and retinoic acid in combination with Decitabine in elderly nonfit patients with acute myeloid leukemia: results of a multicenter, randomized, 2 × 2. Phase II Trial J Clin Oncol 38(3):257–270 88. Hiam-Galvez KJ, Allen BM, Spitzer MH (2021) Systemic immunity in cancer. Nat Rev Cancer 21(6):345–359 89. Prager BC, Xie Q, Bao S, Rich JN (2019) Cancer stem cells: the architects of the tumor ecosystem. Cell Stem Cell 24(1):41–53 90. Lytle NK, Barber AG, Reya T (2018) Stem cell fate in cancer growth, progression and therapy resistance. Nat Rev Cancer 18(11): 669–680 91. Vora P, Venugopal C, Salim SK, Tatari N, Bakhshinyan D, Singh M et al (2020) The rational development of CD133-targeting immunotherapies for glioblastoma. Cell Stem Cell 26(6):832–844.e6 92. Sakamuri D, Glitza IC, Betancourt Cuellar SL, Subbiah V, Fu S, Tsimberidou AM et al (2018) Phase I dose-escalation study of antiCTLA-4 antibody Ipilimumab and

17

lenalidomide in patients with advanced cancers. Mol Cancer Ther 17(3):671–676 93. Migden MR, Rischin D, Schmults CD, Guminski A, Hauschild A, Lewis KD et al (2018) PD-1 blockade with cemiplimab in advanced cutaneous squamous-cell carcinoma. N Engl J Med 379(4):341–351 94. Motzer RJ, Penkov K, Haanen J, Rini B, Albiges L, Campbell MT et al (2019) Avelumab plus axitinib versus sunitinib for advanced renal-cell carcinoma. N Engl J Med 380(12):1103–1115 95. Park JH, Rivie`re I, Gonen M, Wang X, Se´ne´chal B, Curran KJ et al (2018) Longterm follow-up of CD19 CAR therapy in acute lymphoblastic leukemia. N Engl J Med 378(5):449–459 96. Maude SL, Frey N, Shaw PA, Aplenc R, Barrett DM, Bunin NJ et al (2014) Chimeric antigen receptor T cells for sustained remissions in leukemia. N Engl J Med 371(16): 1507–1517 97. Fry TJ, Shah NN, Orentas RJ, StetlerStevenson M, Yuan CM, Ramakrishna S et al (2018) CD22-targeted CAR T cells induce remission in B-ALL that is naive or resistant to CD19-targeted CAR immunotherapy. Nat Med 24(1):20–28 98. Kahl B (2008) Chemotherapy combinations with monoclonal antibodies in non-Hodgkin’s lymphoma. Semin Hematol 45(2): 90–94 99. Stupp R, Mason WP, van den Bent MJ, Weller M, Fisher B, Taphoorn MJB et al (2005) Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 352(10):987–996 100. Adler S, Seidel J, Choyke P, Knopp MV, Binzel K, Zhang J et al (2017) Minimum lesion detectability as a measure of PET system performance. EJNMMI Phys 4(1):13 101. Bao S, Wu Q, McLendon RE, Hao Y, Shi Q, Hjelmeland AB et al (2006) Glioma stem cells promote radioresistance by preferential activation of the DNA damage response. Nature 444(7120):756–760 102. Liu G, Yuan X, Zeng Z, Tunici P, Ng H, Abdulkadir IR et al (2006) Analysis of gene expression and chemoresistance of CD133+ cancer stem cells in glioblastoma. Mol Cancer 5(1):67 103. Rambow F, Rogiers A, Marin-Bejar O, Aibar S, Femel J, Dewaele M et al (2018) Toward minimal residual disease-directed therapy in melanoma. Cell 174(4):843–855. e19

18

Muhammad Vaseem Shaikh et al.

104. Ravindran Menon D, Das S, Krepler C, Vultur A, Rinner B, Schauer S et al (2015) A stress-induced early innate response causes multidrug tolerance in melanoma. Oncogene 34(34):4448–4459 105. Sharma SV, Lee DY, Li B, Quinlan MP, Takahashi F, Maheswaran S et al (2010) A chromatin-mediated reversible drug-tolerant state in cancer cell subpopulations. Cell 141(1):69–80 106. Trumpp A, Wiestler OD (2008) Mechanisms of disease: cancer stem cells—targeting the evil twin. Nat Rev Clin Oncol 5(6):337–347 107. Rodriguez-Brenes IA, Wodarz D (2015) Preventing clonal evolutionary processes in cancer: insights from mathematical models. Proc Natl Acad Sci U S A 112(29):8843–8850 108. McGranahan N, Swanton C (2017) Clonal heterogeneity and tumor evolution: past, present, and the future. Cell 168(4):613–628 109. Qazi MA, Salim SK, Brown KR, Mikolajewicz N, Savage N, Han H et al (2022) Characterization of the minimal residual disease state reveals distinct evolutionary trajectories of human glioblastoma. Cell Rep 40(13):111420 110. Luskin MR, Murakami MA, Manalis SR, Weinstock DM (2018) Targeting minimal residual disease: a path to cure? Nat Rev Cancer 18(4):255–263

111. Cescon DW, Kalinsky K, Parsons HA, Smith KL, Spears PA, Thomas A et al (2022) Therapeutic targeting of minimal residual disease to prevent late recurrence in hormone-receptor positive breast cancer: challenges and new approaches. Front Oncol 1:667397 112. Dillon R, Potter N, Freeman S, Russell N (2021) How we use molecular minimal residual disease (MRD) testing in acute myeloid leukaemia (AML). Br J Haematol 193(2): 231–244 113. Andre´ T, Boni C, Navarro M, Tabernero J, Hickish T, Topham C et al (2009) Improved overall survival with oxaliplatin, fluorouracil, and leucovorin as adjuvant treatment in stage II or III colon cancer in the MOSAIC trial. J Clin Oncol 27(19):3109–3116 114. Winton T, Livingston R, Johnson D, Rigas J, Johnston M, Butts C et al (2005) Vinorelbine plus cisplatin vs. observation in resected non– small-cell lung cancer. N Engl J Med 352(25):2589–2597 115. Yu Z, Guo X, Jiang Y, Teng L, Luo J, Wang P et al (2018) Adjuvant endocrine monotherapy for postmenopausal early breast cancer patients with hormone-receptor positive: a systemic review and network meta-analysis. Breast Cancer 25(1):8–16

Chapter 2 Immunohistochemistry for Cancer Stem Cell Detection: Principles and Methods Giuseppa Zannini, Renato Franco, and Federica Zito Marino Abstract Cancer stem cells (CSCs) are rare immortal cells within tumors with capabilities of self-renewal, differentiation, and tumorigenicity. CSCs play a pivotal role in the tumor development, progression, relapse, and resistance of anticancer therapy. The technique of choice to detect CSCs in formalin-fixed and paraffinembedded (FFPE) samples is immunohistochemistry (IHC) since it is inexpensive and widespread in most laboratories. The main aims of this chapter are the description of the protocols and the automated immunohistochemical systems used for the identification of CSCs. Furthermore, a focus on the most common troubleshooting in CSC IHC is provided. Finally, an overview of the main markers of cancer stem cells in several cancer types will be provided. Key words Cancer stem cells (CSCs), Antigen retrieval, Detection methods, Fixation, Immunohistochemistry, Standardization, Troubleshooting

1

Introduction Cancer stem cells (CSCs) comprise a rare subpopulation of tumor cells quiescent and self-renewing that play a key role in tumor development, progression, and relapse. CSCs originate from differentiated cells or stem cells resident in adult tissue. CSCs can lead to possible metastasis and relapse of the disease since they have unlimited potential cell division [1, 2]. Several findings reported that CSCs have been identified in many tumor types, including colorectal, breast, ovary, pancreas, prostate, melanoma, head, and neck cancer [2–4]. CSCs survive to radio- and chemotherapy treatments; thus, they are frequently associated with resistance to the treatment and recurrence. CSCs could be identified on the cell surface or in the intracellular space of formalin-fixed and paraffin-embedded (FFPE) samples using both immunohistochemistry (IHC) or gene expression arrays [2, 5, 6]. Several CSC markers have been identified for many types of cancer, as shown in Table 1 [7–18]. CSC markers are

Federica Papaccio and Gianpaolo Papaccio (eds.), Cancer Stem Cells: Methods and Protocols, Methods in Molecular Biology, vol. 2777, https://doi.org/10.1007/978-1-0716-3730-2_2, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

19

20

Giuseppa Zannini et al.

Table 1 The main cancer stem cell markers Cancer type

Cancer stem cell markers

Bladder

Aldehyde dehydrogenase, 1-A1/ALDH1A1, CD44, CD47, CEACAM-6/ CD66c, Nanog, KRT14

Breast

Aldehyde dehydrogenase, 1-A1/ALDH1A1, BMI-1, CD24, CD44, CD133, connexin 3/GJA1, CXCR4, DLL4, EpCAM/TROP1, ErbB2/Her2, GLI-1, GLI-2, IL-1 α/IL-1F1, IL-6 Rα, CXCR1/IL-8 RA, integrin α6/CD49f, PON1, PTEN, CD29, CD61, CD70, CD90, Lgr5, ProC-R, Nanog, notch, Oct-3/4, SOX2, Wnt/β-catenin

Colon

ALCAM/CD166, aldehyde dehydrogenase, 1-A1/ALDH1A1, CD44, CD133, DPPIV/CD26, EpCAM/TROP1, GLI-1, Lgr5/GPR49, Musashi-1, CD24, Letm1, Nanog, Oct-3/4, Sall 4, SOX2

Gastric

CD44, DLL4, Lgr5/GPR49, CD24, CD90, CD133, ALDH, Letm1, Musashi-2, Nanog, Oct-3/4, SOX2

Glioma/ A20/TNFAIP3, ABCG2, aldehyde dehydrogenase, 1-A1/ALDH1A1, BMI-1, Medulloblastoma CD15/Lewis X, CD44, CD133, CX3CL1/Fractalkine, CX3CR1, CXCR4, HIF-2 α/EPAS1, IL-6 Rα, integrin α6/CD49f, L1CAM, c-Maf, Musashi-1, c-Myc, nestin, podoplanin, SOX2, SSEA1, A2B5, PDGFRA, EGFR Head and neck

ABCG2, aldehyde dehydrogenase, 1-A1/ALDH1A1, BMI-1, CD44, HGF R/cMET, Lgr5/GPR49, CD133, Nanog, Oct-4

Leukemia

BMI-1, CD34, CD38, CD44, CD47, CD96, CD117/c-kit, GLI-1, GLI-2, IL-3 Rα /CD123, MICL/CLEC12A, Musashi-2, TIM-3, CD33, CD123, CLL-1, ALDH, Nanog, Oct-3/4, SOX2, CD25, CD26, CD36, CD117, IL1RAP, JAK/STAT, Wnt/β-catenin, FOXO, hedgehog/Smo/Gli2

Liver

α-Fetoprotein/AFP, aminopeptidase N/CD13, CD45, CD45.1, CD45.2, CD90/Thy1CD90, NF2/merlin, CD24, CD44, CD133, EpCAM, Nanog, notch, Oct-3/4, SOX2, Wnt/β-catenin

Lung

ABCG2, aldehyde dehydrogenase, 1-A1/ALDH1A1, CD90/Thy1, CD117/ckit, EpCAM/TROP1, CD44, CD87, CD133, CD166, Nanog, Oct-3/4

Melanoma

ABCB5, ABCG2, ALCAM/CD166, CD133, MS4A1/CD20, nestin, NGF R/TNFRSF16, CD271, ALDH1A

Myeloma

ABCB5, CD19, CD27/TNFRSF7, CD38, MS4A1/CD20, Syndecan-1/ CD138, ALDH, CD24, SOX2, Oct-4, BTK, RARα, ROS, Wnt/β-catenin, hedgehog, notch, PI3K/AKT

Osteosarcoma

ABCG2, CD44, endoglin/CD105, nestin, STRO-1, SOX2, ALDH, CD133, CD117, CD271

Ovarian

CD133, aldehyde dehydrogenase, CD24, ROR1, CD326/EpCam, ALDH1, CD338/ABCG2, CD243/ABCB1/MDR1, Nanog, SOX2, Oct-4/POU5F1, c-Myc, CD184/CXCR4

Pancreatic

Aldehyde dehydrogenase, 1-A1/ALDH1A1, BMI-1, CD24, CD44, CXCR4, EpCAM/TROP1, PON1, CD133, nestin, notch, Jegged, ABCG2

Prostate

ABCG2, ALCAM/CD166, aldehyde dehydrogenase, 1-A1/ALDH1A1, α-methylacyl-CoA, racemase/AMACR, BMI-1, CD44, CD151, c-Maf, c-Myc, TRA-1-60(R), α2β1 integrin, α6 integrin, p63, Sca-1, CD133

Immunohistochemistry for Cancer Stem Cell Detection: Principles and Methods

21

generally tumor-specific and the IHC detection is preferentially used since it is inexpensive, widespread in most laboratories, and easy to apply clinically. Immunohistochemistry (IHC) is a technique that allows the visualization of an antigen and its location in tissue sections using a specific antigen–antibody reaction. Over time, many improvements have been made in the standardization of IHC methods in order to use this assay in various fields including immunology, histology, and oncology. As a highly sensitive and specific method, IHC is an essential tool for routine diagnosis and research. The application of IHC has been widely expanded in the pathology field, for example, it is used for differential diagnosis and the identification of the site of the primary tumor in metastatic patients. Two different IHC approaches could be used: indirect and direct methods. The indirect method involves an unlabeled primary antibody that binds the specific antigen, a labeled secondary antibody that reacts with the primary antibody. In particular, the secondary antibody is conjugated with an enzyme that reacts with the substrate yielding a chromogenic development. Conversely, the direct method uses a primary antibody directly conjugated to an enzyme, which is then activated by adding a substrate producing a detectable product. Direct detection does not require the use of a secondary antibody. To date, the IHC manual approach has been almost completely replaced by the automated one, since it is not operator-dependent and uses ready-to-use reagents, kits capable of amplifying the signal, and standardized protocols, allowing increased sensitivity and specificity. The main aims of this chapter are to explain materials and methods, as well as the standardization and troubleshooting in immunohistochemistry assay with the automated approach. Finally, we describe the pitfalls of the IHC on tissue microarrays (TMAs) to detect cancer stem cells.

2

Materials

2.1 Buffer, Diluents, and Antigen Retrieval Solutions

1. Reaction buffer, a washing solution applied between the various staining steps. 2. Target retrieval solution citrate, a pH 6 solution used to unmask the antigenic sites. 3. Target retrieval solution EDTA, a pH 9.0 solution used to unmask the antigenic sites (see Note 1). 4. Proteinase K, used to digest the proteic component of tissue sections. 5. Ez Prep Concentrate (10X), used to dewax tissue sections. 6. LCS (predilute), a liquid coverslip used to protect the slides from drying.

22

Giuseppa Zannini et al.

2.2 Substrates, Chromogens, and Counterstain Solutions

1. Hydrogen peroxide-DAB solution, used for peroxidase-based immunohistochemical methods (see Note 2). 2. Permanent red substrate—chromogen—used for alkaline phosphatase-based immunohistochemical methods. 3. Hematoxylin, used as a counterstain to highlight specific cellular morphologies or structures in order to better visualize the localization of the primary antibody (see Note 3). 4. Bluing reagent, used as a post-counterstain.

2.3 Endogenous Activity Blocking Solutions

1. Peroxidase and alkaline phosphatase blocking reagent (see Notes 4 and 5). 2. Protein block, serum-free. 3. Biotin blocking system (see Note 6).

3

Methods

3.1 Preanalytical Stage: Tissue Fixation and Processing of FFPE Samples

In order to maintain cell morphology, tissue architecture, and the antigenicity of target epitopes, the complete preparation and the correct handling of the sample from the appropriate fixation to adequate paraffin block preparation are very important for the good success of the immunohistochemical technique. Fixation chemically is critical because it cross-links proteins or reduces protein solubility, which can mask target antigens during prolonged or improper fixation. The gold standard fixative for routine histology and immunohistochemistry is formaldehyde, which is a semireversible, covalent cross-linking reagent that can be used for perfusion or immersion fixation for any length of time, depending on the level of fixation required. Minimal fixation of 1–2 days is recommended (see Note 7). After fixation, tissue samples are embedded in paraffin to maintain the natural shape and architecture of the sample during longterm storage and sectioning for IHC. Paraffin-embedded tissue samples are sectioned into slices as thin as 4–5 μm with a microtome. Then these sections are mounted on glass slides. These glass slides are coated with an adhesive, which is added by surfacetreating glass slides with 3-aminopropyltriethoxysilane (APTS) or poly-L-lysine, which both leave amino groups on the surface of the glass to which the tissue directly couples (see Note 8).

3.2 Antigen Retrieval Methods

The tertiary and quaternary structures of many antigens are modified by fixation with cross-linking agents, and this makes antigens undetectable by antibodies. In order to retrieve the loss of antigenicity, antigen retrieval (AR) methods are used and proteins return to their prefixation conformation. About 85% of antigens fixed in formalin need some type of AR to optimize the immunoreactions [19].

Immunohistochemistry for Cancer Stem Cell Detection: Principles and Methods

23

FFPE tissue sections always need a treatment to unmask the antibody epitopes, either by heat or enzymatic degradation. These steps represent a critical phase because, if the antibodies will not have complete access to the tissue, they will be unable to bind to the correct epitopes. The request of AR depends on both the antigen examined and the antibody used. Epitope recognition by the primary antibodies can be hampered by methylene bridges between proteins due to formaldehyde. To remove these bridges there are two methods, heat-induced epitope retrieval (HIER) and proteolytic-induced epitope retrieval (PIER). PIER was the most commonly used antigen retrieval method before the invention of heat-based antigen retrieval methods. The PIER approach uses the enzymatic activity of pronase, pepsin, ficin, trypsin, or proteinase K to partially digest proteins to unmask the antibody epitopes, and the efficacy of using PIER depends on enzyme concentration and incubation time. In this chapter, we support the utilization of proteinase K because it is active at room temperature. The method of PIER is a digestion of protein crosslinkages introduced during formalin fixation, but this cleavage is nonspecific and so some antigens may be negatively poked by this treatment. The effect of PIER is related to the concentration and type of enzyme and incubation parameters (time, temperature, and pH), as well as to the duration of fixation. HIER is to date the most common approach used for antigen retrieval and is based on the concept that the chemical reactions between proteins and formalin may be reversed by high temperature or strong alkaline hydrolysis [20]. Temperature, pH, and time of incubation are very important factors that must be optimized for proper antigen unmasking without causing morphological damage. Sodium citrate (pH 6) and Tris/EDTA (pH 9) buffers are commonly used with HIER in conjunction with the heat source. The grade of fixation can drastically change the response of antigens to antigen retrieval. Normally, unfixed proteins are denatured at temperatures of 70–90
C, while such proteins do not suffer denaturation at the same temperatures when they have been fixed in formaldehyde. 3.3 Endogenous Activity Block

Endogenous biotin, peroxidase, phosphatase, or other enzymatic activities must be blocked to avoid false positives and high background detection; physical or chemical blocking strategies are employed to achieve this. Antibodies bind with avidity to specific epitopes, but they may partially or weakly bind to reactive sites that are sites on nonspecific proteins and these sites are similar to the cognate-binding sites on the target antigen. This nonspecific binding causes high background staining that can mask the detection of the target antigen. In order to decrease background staining in

24

Giuseppa Zannini et al.

IHC, tissue samples are incubated with a buffer that blocks the reactive sites to which the primary or secondary antibodies may bind. Among common blocking buffers there are normal serum, nonfat dry milk, BSA, or gelatin, and commercial blocking buffers. 3.4 Signal Revelation Systems

The final intent of any immunohistochemical method is to identify the maximum quantity of antigen with the minimal possible background. The option of method will depend on the quantity of antigen present, the level of sensitivity required, and the technical capabilities of the laboratory. The protein of interest is detected by an enzyme that converts the substrate into a colored detectable product. Enzymes can be directly conjugated to primary antibodies (direct immunohistochemistry, less sensitive than the indirect approach), labeled with streptavidin, and combined with biotinylated secondary antibodies (avidin–biotin complex method, most heavily used in research labs), or directly conjugated to secondary antibodies (indirect immunohistochemistry, highly sensitive with low background). Enzymes that are mainly used in antigen detection methods are horseradish peroxidase (HRP) and alkaline phosphatase (AP). The HRP reacts with hydrogen peroxide and 3,30 -diaminobenzidine (DAB), which is a derivative of benzene commonly used in IHC staining as a chromogen. In the presence of hydrogen peroxide, the DAB is oxidized leading to the production of a brown precipitate visible using light microscopy. The DAB precipitate is insoluble in water, alcohol, and other organic solvents, such as xylene and isopropanol. This allows for great flexibility in the choice of counterstain and mounting medium. Moreover, DAB staining is heat-resistant and extremely stable for many years. However, DAB should be used relatively quickly once combined with hydrogen peroxide in order to avoid generating brown precipitate in the solution which could contribute to background staining. In fact, DAB staining kits are supplied with concentrated DAB substrate and a hydrogen peroxide buffer, to be combined shortly before use. The AP hydrolyzes its substrates into phenolic compounds and phosphates, which interact with colorless diazonium salts and produces a colored product at the reaction site on tissues and cells. Fast Red, Fast Blue, or NBT/BCIP is commonly used as chromogens with AP in IHC.

3.5 Automated Immunohistochemical Systems

In clinical practice, IHC is performed with an automated approach allowing to carry out simultaneously the stains of multiple samples with minimal operator intervention. Currently, different staining instruments are commercially available. The first staining platforms placed slides in a vertical orientation to apply and rinse reagents from the slide surface with capillary action using the force of gravity. Current systems put slides in trays or racks in a horizontal

Immunohistochemistry for Cancer Stem Cell Detection: Principles and Methods

25

orientation applying the reagents from above on the slide surface by air or capillary action. The newest IHC devices equipped with independent slots can stain each slide independently from the others, allowing the fastest throughput. All automated systems need a mechanism to dispense specific amounts of reagents to the slides. In the rotary system, which is the most commonly used, slides and reagents are placed in two different circular platters. Rotation of the upper platter allows the dispensing of the reagents directly on the corresponding slide. This is possible thanks to a barcode system that uniquely identifies the slide with the specific reagents. Moreover, a standard amount of reagent is dispensed for each slide with a single compression of a plunger. Staining systems that incorporate the heating slide function allow applying heat during the incubation of reagents in order to accelerate reaction times. Thus, slide baking, used to fix tissue sections on the slide, can be performed within the instrument itself, but many laboratories perform this step in an independent oven in order to free up the equipment for the other staining phases, like the HIER step. Moreover, the slides are protected from drying by covering them with other slides, cover tiles, or liquid coverslips. 3.6 Advantages of the Automated Immunohistochemical Approach

The automated immunohistochemical system has completely replaced the manual approach since it has several advantages, which can be summarized as follows: • Most of the staining instruments are equipped with computerized software used in reagent inventory management for tracking reagent stocks and their expiration dates. The barcode system avoids the possibility of cross-contamination between the different slides and allows the tracking of each staining reaction, recording all the information describing the process that leads to the immunohistochemical reaction. • Operator-dependent errors are minimized. The instrument allows the monitoring of some errors with alarm signals, such as inappropriate temperatures and reagent volumes and incorrectly added reagents identified with barcode tracking. • Most companies provide the use of prediluted “ready-to-use” antibodies, which is convenient because the antibody dilution step is avoided, but this is reflected in the higher cost of the reagent. • Use of standardized protocols for most of the antibodies detected in clinical practice. Indeed, the priority in the clinical setting is the reproducibility of the established protocols to obtain the same staining results in different laboratories. Conversely, “open” automated systems are preferred in the research setting, in which developing new staining protocols for the growing number of new biomarkers could be a hot issue.

26

Giuseppa Zannini et al.

Indeed, they are more flexible in protocol options and allow the choice of the reagents to use with different incubation times and temperatures as in manual protocols. • To avoid any staining errors not signaled by the instrument, some companies provide positive controls to be loaded with the samples into the instrument, or in some cases, the operator can place an archival positive control on the same slide of the sample. 3.7 Automated Immunohistochemistry Staining Protocol

Automated immunohistochemistry is optimized by signal amplification using polymers directly conjugated with enzymes that react with their substrates and produce a colored precipitate (Fig. 1). These little molecules minimize steric hindrance and easily penetrate tissue, increasing the specificity and sensitivity of detection reactions. Recent signal amplification methods provide the use of secondary antibodies conjugated to numerous copies of a synthetic non-endogenous HQ hapten. HRP multimers recognize the HQ haptens and exponentially multiply the number of signaling molecules, allowing for exceptional levels of staining intensity without an increase in the background. In detail, the automated staining method can be summarized with the following steps: 1. Dewaxing tissue with EZ prep at 72
C for 8 min. 2. HIER with sodium citrate (pH 6) or Tris/EDTA (pH 9) at 95
C for 44 min. 3. Blocking of endogenous proteins and peroxides at 37
C for 4 min.

Fig. 1 The automated approach of the immunohistochemical method

Immunohistochemistry for Cancer Stem Cell Detection: Principles and Methods

27

4. Incubation with primary antibody at 37
C for 30–60 min (see Note 9). 5. Incubation with HRP/AP-labelled secondary antibody for 16 min. 6. Revelation with DAB or Fast Red solutions for the identification of mouse and rabbit IgG and IgM antibodies for 4–8 min. 7. Counterstaining with hematoxylin for 4 min. 8. Post-counterstain with a bluing solution for 4 min. 9. Dehydration and mounting tissue sections. 3.8 Immunohistochemical Detection of Multiple Antigens

In commerce, there are detection kits that allow the detection of two or more antigens using two different enzymes as labels (e.g., peroxidase and alkaline phosphatase). The quality of the detection of antigens depends on their location (different or the same tissues, cells, or cellular compartments). An accurate selection of the chromogen for each antigen is also necessary to achieve optimal discrimination between antigens. The same localization of two antigens (e.g., both antigens are within the nucleus or the cytoplasm of the same cell type) does not allow their discrimination. Double immunodetection is complicated by the variety of AR methods used for different antigens; in other words, an AR necessary for one antigen might have deleterious effects for the second antigen to be detected [21].

3.9 Main Markers of Cancer Stem Cells

Until now, no universal markers are defined to identify CSC in specific tumor, since CSC markers are shared by both cancer and noncancerous cells, such as normal SC and various types of differentiated cells. Moreover, some CSC markers, called inclusive markers, are shared by a wide variety of cancer types, while exclusive markers are specific to only some of them [22]. Among the several CSC markers identified, we report a focus on three of the most commonly used in most cancers, including aldehyde dehydrogenase 1A1, CD44, and CD133.

3.9.1 Aldehyde Dehydrogenase 1A1 (ALDH1A1)

Aldehyde dehydrogenases (ALDHs) belong to a family of enzymes that catalyze the aldehyde oxidation in carboxylic acid. The main ALDH1 proteins are ALDH1A1, ALDH1A2, and ALDH1A3, which are highly conserved cytosolic isozymes localized in various adult tissues, such as brain, kidney, testis, lens, retina, eye, liver, and lungs. Historically, ALDH1A1 is the isoenzyme related to stem cell (SC) populations, playing a critical role as a marker of SC and CSCs. In particular, in normal tissues, ALDH1A1 is involved in retinoid signaling pathways, acetaldehyde metabolism and self-protection, differentiation, and expansion of SC populations. It is normally expressed in several tissues, such as liver, kidney, red blood cells, skeletal muscle, lung, breast, stomach, brain, pancreas, testis,

28

Giuseppa Zannini et al.

prostate, and ovary. However, ALDH1A1 is involved in the regulation of the growth and differentiation of normal and cancer cells. Its overexpression is correlated with both poor and favorable prognoses of different types of cancers. Moreover, ALDH1A1 is considered a CSC therapeutic target in tissue types that in physiological conditions do not express a high level of ALDH1A1, such as lung, colon, breast, esophagus, and stomach epithelium. IHC can be used for ALDH1A1 tissue expression using isotype-specific antibodies: positive samples have specific staining in the cytosol of cancer cells [23]. 3.9.2

CD44

CD44 is a transmembrane glycoprotein initially identified as the lymphocyte homing receptor. It is commonly expressed in hematopoietic, epithelial, and mesenchymal cells. Furthermore, CD44 is considered a surface marker of CSCs in human cancers. CD44 is widely distributed in fetal tissues and in normal adult tissues, such as the central nervous system, lung, and epidermis. It interacts with hyaluronic acid, extracellular matrix molecules, and growth factors by modulating signal transduction through posttranslational modifications. It is also involved in cell adhesion, migration, differentiation, and signaling transduction. CD44 is expressed in different cancer types, such as pancreatic, breast, prostate, head and neck squamous cell carcinoma, and gastrointestinal cancers [24]. In detail, it is significantly linked with cancer patients’ survival in both gastric and pancreatic cancer, while in pancreatic, colon adenocarcinoma, and gastric cancer CD44 is closely correlated with immune infiltration and immune-suppressive features. Moreover, CD44 overexpression is associated with invasiveness, drug resistance, epithelial–mesenchymal transition, and metastasis [25]. CD44 can be detected through the IHC method. Samples are considered positive with strong membranous staining in more than 5% of cancer cells, while samples are classified as negative with only cytoplasmatic staining and/or membranous staining in less than 5% of cancer cells [26].

3.9.3

Prominin-1/CD133

Promin-1, also known as CD133, is one of the most wellcharacterized biomarkers used for the detection of CSCs. It is a transmembrane glycoprotein expressed on endothelial progenitors, hematopoietic stem cells, and dermal-derived stem cells capable of differentiating into neural cells. CD133 activates the Wnt signaling pathway modulating the angiogenesis through the increased expression of VEGF-A and interleukin-8. Furthermore, it is responsible for tumorigenesis, metastasis, chemoresistance, cell proliferation, apoptosis inhibition, and upregulation of FLIP leading to the chemoresistance of cancer cells. CD133 is a biomarker that can be used for the detection of stem cells in both normal and cancer cells: it is commonly expressed in prostate, brain, kidney,

Immunohistochemistry for Cancer Stem Cell Detection: Principles and Methods

29

liver, bone marrow, pancreas, and skin, but also in pancreatic, prostate, ovarian, colon, brain, and liver cancers. Moreover, in adult tissues, CD133 can be found in the epididymis, seminal vesicle, ductus deferens, kidney, and prostate [27]. High expression of CD133 is a poor prognostic factor in colorectal, endometrial, melanoma, hepatocellular, neuroblastoma, and ovarian cancers. CD33 tissue expression can be detected through the IHC assay, in which specimens are classified as positive with cellular membrane staining [28]. 3.10 Troubleshooting 3.10.1 Excessive Background Staining

1. Pre-staining problems: (a) Insufficient fixation, necrosis, and autolysis. (b) Tissue sections allowed to dry out. Then, reduce incubation time and/or incubate in a humidified chamber. (c) If sections are not completely deparaffinized, use fresh dewaxing solutions. (d) Slide adhesive inappropriate or too thick. Make use of adhesives specific for IHC or positively charged slides. (e) Tissue section too thick, then prepare thinner sections. (f) Inappropriate antigen retrieval used, then upgrade antigen retrieval conditions. (g) Incubation temperature too elevated, then reduce temperature. 2. Blocking problems: (a) If endogenous enzyme activity is not suppressed, increase the concentration of the blocking agent. (b) Inappropriate protein blocking. Change blocking agent. (c) Inappropriate blocking of endogenous avidin-binding activity. Employ an avidin–biotin blocking step or use a non-avidin–biotin detection method. (d) Inappropriate blocking of endogenous biotin. Employ an avidin–biotin blocking step or a non-avidin–biotin detection method. (e) Blocking serum not from the same species. Employ blocking serum from the same species as the link (secondary) antibody. 3. Primary antibody problems: (a) Primary antibody too concentrated, then dilute the primary antibody. (b) Primary antibody incubation time too long, then decrease incubation time.

30

Giuseppa Zannini et al.

(c) If the primary antibody is from a similar or identical species as the test tissue (mouse on mouse, rat on mouse, etc.), use specific protocols or additional blocking steps. (d) Insufficient buffer washes (inappropriate buffer ion concentration), then change the ionic strength of the buffer solution. 4. Secondary antibody problems: (a) Secondary antibody and label concentration too high. (b) Secondary antibody and label incubation time too long. (c) Buffer washes inadequate. (d) Secondary antibody immunoglobulins.

identifies

endogenous

(tissue)

5. Chromogen and counterstain problems: (a) Chromogen concentration too elevated. Decrease concentration of chromogen. (b) Chromogen allowed to react too long, then decrease incubation time with chromogen. (c) Buffer washes are inadequate. Extend buffer washes. (d) Counterstain hides the IHC reaction. Use another type of counterstain that does not interfere with the immunohistochemical staining. 3.10.2 Inadequate or No Staining of the Test Slide and Adequate Staining of the Positive Control Slide

1. The test tissue does not present the antigen in question. 2. The antigen is present in the test tissue in a very low concentration. Use an amplification procedure, or increase the primary antibody concentration, incubation time or temperature, or a combination thereof. 3. There is an over- or under-fixation of the test tissue. Change antigen retrieval protocol. 4. If the test tissue is from a different species than the control tissue, it has different reactivity with the primary antibody. Then, validate the IHC test with the same species control and test tissues.

3.10.3 Weak or No Staining of Positive Control and Weak or No Staining of Test Slides

1. If all slides from some of the primary antibodies used in the run are affected, control inadequacy of the primary antibody, method incompatibility, primary and link antibody incompatibility, or inadequate antigen retrieval. 2. If the whole run has a negative result, control assay log, adequacy of reagent volumes, and sequence of reagent delivery to the slide. Assess whether the reagents have been passed on all slides (e.g., buffer, chromogen).

Immunohistochemistry for Cancer Stem Cell Detection: Principles and Methods

31

3. If it is hit-and-miss throughout the run, there are technical problems or problems with the tissues. Control inadequate sequence of reagents, unbalanced autostainer, and inadequate drop zone. 3.10.4 No Staining of the Positive Control Slide and Adequate Staining of the Test Slide

4

1. Review the checklist if there is a technical error in the staining or handling of the positive control slide, and repeat the assay if everything is in order. 2. Tissue photo-oxidation and dehydration during prolonged storage of tissue control sections can lead to tissue section aging [10–13]. Use a known positive case to test the tissue control section by IHC.

Notes 1. When using heat-induced epitope retrieval (antigen retrieval), employ the target retrieval solution more appropriate to detect that particular antigen of the two included in Materials. 2. The incubation time is related to the amount of antigen; in fact, the background increases with a longer incubation time. 3. In order to highlight the precipitates of the DAB–peroxidase reaction, hematoxylin can be used as a counterstain. 4. Pretreatment of the sections for antigen retrieval can follow or anticipate the endogenous peroxidase blocking step. 5. This treatment can damage the same antigens, in particular those in the cytoplasmic membrane, and for this reason, it should be done after the incubation with the primary antibody. 6. Add this solution before the incubation with the biotinylated antibody [14]. 7. Other fixatives. Many of the formalin substitutes are coagulating fixatives that precipitate proteins by breaking hydrogen bonds in the absence of protein cross-linking, and the typical non-cross-linking fixative is ethanol. Other fixatives employed in IHC are glyoxal (dialdehyde), a mixture of glyoxal and alcohol, 4% paraformaldehyde (PFA), and zinc formalin. 8. Other slide adhesives can be used (e.g., poly-L-lysine). 9. As necessary, incubation duration and incubation temperature (e.g., 37
C, 4
C) can be modified.

5

Pitfalls in TMA Use to Detect CSCs by IHC IHC is often used to study marker expression on TMA sections. TMA is a paraffin block in which several tissue cores are assembled in an array fashion with the aim of putting in place multiplex

32

Giuseppa Zannini et al.

histological analysis. Cantile et al. [17] performed IHC analysis on breast and lung cancer TMAs to assess the expression of CSC markers, specifically CD133, ALDH1, and CD44. The comparison of the results on TMA with single/whole sections of the same samples demonstrated that TMA is an inadequate technique for the detection of CSCs. A high heterogeneity of CSCs within tumor areas could explain the unsuitability of TMA. Finally, the use of a whole section is recommended for the detection of CSCs through IHC. References 1. Clarke MF, Dick JE, Dirks PB et al (2006) Cancer stem cells-perspectives on current status and future directions: AACR workshop on cancer stem cells. Cancer Res 66:9339–9344 2. Walcher L, Kistenmacher AK, Suo H et al (2020) Cancer stem cells-origins and biomarkers: perspectives for targeted personalized therapies. Front Immunol 7(11):1280 3. Krause M, Dubrovska A, Linge A et al (2016) Cancer stem cells: radioresistance, prediction of radiotherapy out come and specific targets for combined treatments. Adv Drug Deliv Rev 109:63–73 4. Vinogradov S, Wei X (2012) Cancer stem cells and drug resistance: the potential of nanomedicine. Nanomedicine (Lond) 7:597–615. https://doi.org/10.2217/nnm.12.22 5. Baumann M, Krause M, Thames H et al (2009) Cancer stem cells and radiotherapy. Int J Radiat Biol 85:391–402 6. Baumann M, Krause M, Hill R (2008) Exploring the role of cancer stem cells in radioresistance. Nat Rev Cancer 8:545–554 7. Cancer Stem Cell Markers. Available: https:// www.rndsystems.com/research-area/cancerstem-cell-markers 8. Yamada T, Toyoda T, Matsushita K et al (2021) Expression of stem cell markers as useful complementary factors in the early detection of urinary bladder carcinogens by immunohistochemistry for γ-H2AX. Arch Toxicol 95:715– 726 9. Siddiqui Z, Srivastava AN, Sankhwar SN et al (2020) Synergic effects of cancer stem cells markers, CD44 and embryonic stem cell transcription factor Nanog, on bladder cancer prognosis. Br J Biomed Sci 77:69–75 10. Suva` ML, Tirosh I (2020) The glioma stem cell model in the era of single-cell genomics. Cancer Cell 37:630–636 11. Fan Z, Li M, Chen X et al (2017) Prognostic value of cancer stem cell markers in head and

neck squamous cell carcinoma: a meta-analysis. Sci Rep 7:43008 12. Kozovska Z, Gabrisova V, Kucerova L (2016) Malignant melanoma: diagnosis, treatment and cancer stem cells. Neoplasma 63:510–517 13. Guo W, Wang H, Chen P et al (2021) Identification and characterization of multiple myeloma stem cell-like cells. Cancers 13:3523 14. Brown HK, Tellez-Gabriel M, Heymann D (2017) Cancer stem cells in osteosarcoma. Cancer Lett 386:189–195 15. Hatina J, Boesch M, Sopper S et al (2019) Ovarian cancer stem cell heterogeneity. Adv Exp Med Biol 1139:201–221 16. Burgos-Ojeda D, Rueda BR, Buckanovich RJ (2012) Ovarian cancer stem cell markers: prognostic and therapeutic implications. Cancer Lett 322:1–7 17. Vizio B, Mauri FA, Prati A et al (2012) Comparative evaluation of cancer stem cell markers in normal pancreas and pancreatic ductal adenocarcinoma. Oncol Rep 27:69–76 18. Takao T, Tsujimura A (2008) Prostate stem cells: the niche and cell markers. Int J Urol 15:289–294 19. Ramos-Vara JA, Beissenherz ME (2000) Optimization of immunohistochemical methods using two different antigen retrieval methods on formalin-fixed, paraffin-embedded tissues: experience with 63 markers. J Vet Diagn Investig 12:307–311 20. Ramos-Vara JA (2005) Technical aspects of immunohistochemistry. Vet Pathol 42:405– 426 21. Van der Loos CM (1999) Immunoenzyme multiple staining methods. Bios Scientific Publishers, Oxford 22. Nair M, Sandhu SS, Sharma AK (2018) Cancer molecular markers: a guide to cancer detection and management. Semin Cancer Biol 52:39– 55

Immunohistochemistry for Cancer Stem Cell Detection: Principles and Methods 23. Tomita H, Tanaka K, Tanaka T et al (2016) Aldehyde dehydrogenase 1A1 in stem cells and cancer. Oncotarget 7:11018–11032 24. Chen C, Zhao S, Karnad A et al (2018) The biology and role of CD44 in cancer progression: therapeutic implications. J Hematol Oncol 11:64 25. Williams K, Motiani K, Giridhar PV et al (2013) CD44 integrates signaling in normal stem cell, cancer stem cell and (pre)metastatic niches. Exp Biol Med 238:324–338 26. Horimoto Y, Arakawa A, Sasahara N et al (2016) Combination of cancer stem cell

33

markers CD44 and CD24 is superior to ALDH1 as a prognostic indicator in breast cancer patients with distant metastases. PLoS One 11:e0165253 27. Barzegar Behrooz A, Syahir A, Ahmad S (2019) CD133: beyond a cancer stem cell biomarker. J Drug Target 27:257–269 28. Yoshikawa S, Zen Y, Fujii T et al (2009) Characterization of CD133+ parenchymal cells in the liver: histology and culture. World J Gastroenterol 15:4896–4906

Chapter 3 Isolating Cancer Stem Cells from Solid Tumors Vitale Del Vecchio, Marcella La Noce, and Virginia Tirino Abstract Over the past 20 years, there has been a lot of interest in the study and investigation of cancer stem cells (CSCs) or tumor-initiating cells (TICs). CSCs are rare, dormant cells and able to self-renew and maintain tumor development and heterogeneity. A new age of basic and clinical cancer research, reclassification of human tumors, and the development of novel therapeutic approaches will undoubtedly result from a better knowledge of CSCs. In order to develop effective and therapeutic strategies to treat cancer, it is crucial to understand the basic characteristics of CSCs, their importance to cancer therapy, and methodologies to isolate, detect, and characterize them. Here, we outline the main methods and protocols to identify, isolate, and culture CSCs from primary tumors. Key words Cancer stem cells (CSCs), Spheres, CSC isolation, Side population

1

Introduction Cancer stem cells (CSCs) are a subpopulation of tumor cells with distinct stem-like properties, responsible for tumor initiation, invasive growth, and possibly dissemination to distant organ sites. These few cells have the ability to divide asymmetrically, resulting in identical daughter cells and more differentiated cells that produce the majority of the tumor bulk during successive divisions [1]. There have been other names used to refer to this subpopulation, but “cancer stem cells” has gained widespread recognition. According to the CSC model, a cancer stem cell is a cancer cell that has the ability to self-renew and to generate the diverse lineages of cancer cells in the tumor [2]. Through serial transplantations in immunodeficient nonobese diabetic (NOD)/severe combined immunodeficient (SCID) mice, this population is recognized experimentally by its capacity to re-establish tumor heterogeneity through the formation of new tumors [1]. John Dick [3] and colleagues conducted the first investigation on the discovery of

Federica Papaccio and Gianpaolo Papaccio (eds.), Cancer Stem Cells: Methods and Protocols, Methods in Molecular Biology, vol. 2777, https://doi.org/10.1007/978-1-0716-3730-2_3, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

35

36

Vitale Del Vecchio et al.

CSCs in human acute myeloid leukemia (AML). In 1997, they discovered a hierarchy of stem cells that resembled the normal hierarchy of hematopoietic stem cells. Dick and co-workers demonstrated that the immature CD34+CD38 subpopulation contained rare malignant cells with the ability to repopulate the entire original disease over serial transplantations, implying self-renewal and the capacity to differentiate [3]. This research served as the starting point for studies on CSCs in solid tumors as well as hematologic malignancies. Al-Hajj identified and isolated CSCs from human breast cancer using CD44 and CD24 markers in 2003 [4]. Since, CSCs have been identified and isolated in a variety of solid tumors such as glioblastomas [5], melanoma [6], osteosarcoma [7], prostate [8], ovarian [9], gastric [10], and lung cancers [11]. Although the therapies enhanced the life quality of the patient, metastasis formation remains the first issue in terms of recovery. This is due to the inefficient effect of standard treatments on quiescent CSCs remaining vital and retaining the ability to form second tumors [12]. Usually, the therapies are developed versus proliferant cells and not quiescent or dormant CSCs. Therefore, CSC recognition is the first challenge to block the tumor progression [12]. The identification of CSC-specific markers may help the development of specific target therapeutic strategies. Finally, multiple and specific pathways activated by CSCs also needs to be explored for their effective elimination. CSCs are more resistant to therapies, due to an increase of anti-apoptotic activities and levels of drug efflux pumps such as BCRP (breast cancer resistance protein) and MDR (multidrug resistance) complexes [13]. Future therapeutic strategies will need to integrate inhibition of these mechanisms with CSC killing components. Therefore, it is very important to clarify the CSC biology, in order to develop better diagnostic and therapeutic methodologies, with which to classify, treat, and cure cancer. Here, we provide an overview about the main methodologies currently used to detect and isolate CSCs. These are (1) isolation of CSCs by flow cytometry according to CSC-specific cell surface markers [3–11], (2) detection of side population phenotype by Hoechst 33342 exclusion [14, 15], (3) ability to grow as floating spheres in serum-free medium [16, 17], and (4) aldehyde dehydrogenase (ALDH) activity [18– 21]. None of the methods is exclusively used to isolate tumor CSCs, highlighting the imperative to use combinatorial markers and methodologies.

Methods in Cancer Stem Cells

2 2.1

37

Materials Reagents

2.1.1 Cell Culture Reagents and Materials

1. Type I or III or IV collagenase. Store at 4  C or expiration date.

20  C until

2. Accutase in DPBS without Ca2+ or Mg2+. Store at + 4  C or 20  C until expiration date. 3. TrypLE™ Express Enzyme (1X). Store at +4 expiration date.



C until

4. Trypan blue stain (0.4%) solution. Store at room temperature and protect from light. 5. Ammonium chloride-based buffered solution, erythrocyte lysing agent (without detergent or fixatives). 6. DNase I enzyme. Store at +2 to +8  C until expiration date. 7. B-27™ Supplement (50X), serum-free. Store in freezer ( 5 to 20  C) and protect from light. 8. N-2 Supplement (100X). Store in freezer ( 5 to protect from light.

20  C) and

9. Dispase® I (neutral protease, grade I) lyophilizate from Bacillus polymyxa, filtered through 0.2 μm pore size membrane. Store the lyophilizate at +2 to +8  C until expiration date. 10. Hyaluronidase. Lyophilized powder. Store at 20  C. Stable as supplied for 12 months from date of receipt. 11. Gentamicin solution, 50 mg/mL in deionized water. Store at room temperature until expiration date. 12. PBS w/o Ca2+, Mg2+. Store at 4  C until expiration date. 13. 70 μm Falcon strainers. 14. DMEM low glucose, DMEM, RPMI1640 culture media. Store at 4  C until expiration date. 15. DMEM–F12 and Advanced DMEM–F12 culture media. Store at 4  C until expiration date. 16. Antibiotic–antimycotic solution 100X (10,000 U/mL sodium penicillin, 10,000 μg/mL streptomycin sulfate, and 25 μg/mL amphotericin B). Store at 25 to 15  C. 17. Fetal bovine serum (FBS), South American Origin. Store at 20  C until expiration date. 18. L-Glutamine, 200 mM solution. Store at expiration date. 19. Amphotericin B, 250 μg/mL. Store at expiration date.

5 to

20



C until

20  C until

20. Pen/Strep (5000 units/mL penicillin and 5000 μg/mL streptomycin. Store at 5 to 20  C until expiration date.

38

Vitale Del Vecchio et al.

21. Bovine serum albumin (BSA), lyophilized powder. Store at 2–8  C until expiration date. 22. Trypsin–EDTA solution (0.5% trypsin/0.2% EDTA), stored at 4  C until expiration date. 23. Leukemia inhibitory factor (LIF). Store at

80  C.

24. Putrescine dihydrochloride, powder. Store at room temperature and protect from light, until expiration date. 25. Sodium selenite, powder. Store at room temperature and protect from light, until expiration date. 26. Progesterone, powder. Store at room temperature and protect from light, until expiration date. 27. Insulin solution. Store at 2–8  C until expiration date. 28. Transferrin, powder. Store at room temperature and protect from light, until expiration date. 29. Basic fibroblast growth factor, bFGF. Stored at expiration date. 30. Epidermal growth factor, EGF. Store at date expiration date. 2.1.2

Antibodies

20  C until

20  C until exposure

1. CD24 PE (Becton & Dickinson). Store in the dark, at 2–8  C until expiration date. 2. CD44 FITC (Becton & Dickinson). Store in the dark, at 2–8  C until expiration date. 3. CD45 APC-H7 (Becton & Dickinson). Store in the dark, at 2–8  C until expiration date. 4. CD133/2 PE (clone 293C3, Miltenyi Biotec). Store in the dark, at 2–8  C until expiration date. 5. CD326 (EpCAM) PERCP-Cy5.5 (Becton & Dickinson). Store in the dark, at 2–8  C until expiration date. 6. CD324 (e-cadherin) PERCP-Cy5.5 (Becton & Dickinson). Store in the dark, at 2–8  C until expiration date.

2.1.3 Side Population Reagents

1. Verapamil. Store at +4  C until expiration date. Verapamil can be prepared as a stock solution of 10 mM in DMSO and used at a final concentration of 50 μM (1:200 dilution). 2. Hoechst 33342. Store at +4  C until expiration date. Hoechst 33342 can be prepared as a stock solution of 1 mg/ml in H2O. 3. Propidium iodide (PI) solution. Store at +4 expiration date.



4. 7-Aminoactinomycin D (7-AAD). Store at +4 expiration date.



C until C until

Methods in Cancer Stem Cells

39

2.1.4 ALDH Activity Reagents

1. ALDEFLUOR kit (StemCell Technologies). Store at +4  C until expiration date.

2.1.5 Equipment and Plates

1. Flow cytometer equipped with a 488-nm blue laser, 633-nm red laser, and 405- or 375-nm UV laser for excitation. 2. Inverted phase-contrast microscope (Nikon TS 100, Nikon). 3. Ultralow-attachment plates (Corning). 4. 12  75 mm tubes compatible with the cytometer using low-speed centrifuge (capable of 250 g). 5. 37  C heating device (water bath or heat block). 6. Refrigerator (2–8  C) or ice.

3

Methods

3.1 Primary Cell Culture 3.1.1 Glioblastoma Cell Culture

1. Fresh glioblastoma tissue is minced with a scalpel in small fragments, and then incubate with Accutase and TrypLE (1: 1) or type I collagenase 0.3% for 10 min at 37  C (see Note 1). 2. After filtration through a 70 μm pore filter, perform a cell count and determine the viable cell concentration using trypan blue exclusion cell count. 3. Cells are washed with DMEM low glucose, centrifugate at 600 rpm for 8 min, and culture in 25 cm2 culture flasks in DMEM low glucose supplemented with L-glutamine 200 mM, antibiotic–antimycotic (10,000 U/mL sodium penicillin, 10,000 μg/mL streptomycin sulfate, and 25 μg/mL amphotericin B), and 10% fetal bovine serum. 4. Maintain flasks at 37  C and 5% CO2 and change the medium twice a week. 5. Cells are maintained and passaged as adherent cultures and aliquots are stored as frozen cells in liquid nitrogen.

3.1.2 Breast Cancer Cell Culture

1. Fresh breast cancer tissue is minced with a scalpel in small fragments and enzymatically digested in a 1:1 solution of type III collagenase/hyaluronidase at 37  C for 40–60 min on shaking bath (see Note 1). 2. After filtration through a 70 μm pore filter, perform a cell count and determine the viable cell concentration using trypan blue exclusion cell count. 3. Recovered cells are cultured in DMEM or RPMI1640 at 10% FBS in standard condition (see Note 2). 4. Maintain flasks at 37  C and 5% CO2 and change the medium twice a week.

40

Vitale Del Vecchio et al.

5. Cells are maintained and passaged as adherent cultures and aliquots are stored as frozen cells in liquid nitrogen. 6. If using samples containing blood and the erythrocyte to leukocyte ratio (RBC/WBC) of the specimen is >2:1, lyse the erythrocytes with an ammonium chloride-based buffered solution that does not contain detergents or fixatives to preserve the cell viability. 3.1.3 Lung Cancer Cell Culture

1. Fresh lung cancer tissue is minced with a scalpel in small fragments and enzymatically digested in 1 mg/ml type IV collagenase, 1 mg/ml Dispase, and 0.001% DNAse for 2 or 3 h at 37  C on shaking bath (see Note 1). 2. After filtration through a 70 μm pore filter, perform a cell count and determine the viable cell concentration using trypan blue exclusion cell count. 3. Recovered cells are cultured in DMEM or RPMI1640 at 10% FBS in standard condition (see Note 2). 4. Maintain flasks at 37  C and 5% CO2 and change the medium twice a week. 5. Cells are maintained and passaged as adherent cultures and aliquots are stored as frozen cells in liquid nitrogen. 6. If using samples containing blood and the erythrocyte to leukocyte ratio (RBC/WBC) of the specimen is >2:1, lyse the erythrocytes with an ammonium chloride-based buffered solution that does not contain detergents or fixatives to preserve the cell viability.

3.1.4 Sarcoma Cell Culture

1. Fresh sarcoma biopsies are dissected, minced, and washed in PBS. 2. After all visible clumps are removed, perform enzymatic digestion with 10 mg/ml collagenase type IV or I (soft tissue or bone sarcoma, respectively) and 3 mg/ml Dispase at 37  C overnight (see Note 1). 3. After filtration through a 70 μm pore filter, perform a cell count and determine the viable cell concentration using trypan blue exclusion cell count. 4. Recovered cells are cultured in DMEM or RPMI1640 at 10% FBS in standard condition (see Note 2). 5. Maintain flasks at 37  C and 5% CO2 and change the medium twice a week. 6. Cells are maintained and passaged as adherent cultures and aliquots are stored as frozen cells in liquid nitrogen.

Methods in Cancer Stem Cells

41

7. If using samples containing blood and the erythrocyte to leukocyte ratio (RBC/WBC) of the specimen is >2:1, lyse the erythrocytes with an ammonium chloride-based buffered solution that does not contain detergents or fixatives to preserve the cell viability. 3.2

Spheres

3.2.1 Neurospheres from Glioblastoma

1. Prepare tumor brain stem cell DMEM/F12 basal media adding 2 mM L-glutamine, 100 U/ml penicillin, and 100 U/ml streptomycin. 2. Prepare tumor brain stem cell DMEM/F12 complete media immediately before use adding 20 ng/ml recombinant human epidermal growth factor (EGF), 20 ng/ml recombinant human basic fibroblast growth factor (bFGF), leukemia inhibitory factor 10 ng/μL, N2, and 1x B27 supplement. 3. Pre-warm at 37  C the ultralow-attachment plates. 4. Seed cells in 24-well plates at 2  104 density in tumor brain stem cell DMEM/F12 complete media. 5. Incubate ultralow-attachment plates at 37  C and 5% CO2. 6. Replace the media or refresh growth factors twice a week. 7. Neurospheres are visible after 48–72 h of culture. 8. For serial passaging, disaggregate the neurospheres with 500 μl of pre-warmed Accutase for 2–3 min. 9. Add the FBS to neutralize the Accutase, wash, and count the single cells using trypan blue. 10. Seed the cells into a new ultralow-attachment plate at the same density used in the primary generation.

3.2.2

Mammospheres

1. Prepare mammosphere DMEM/F12 basal media adding 2 mM L-glutamine, 100 U/ml penicillin, and 100 U/ml streptomycin. 2. Prepare mammosphere DMEM/F12 complete media immediately before use adding 20 ng/ml recombinant human epidermal growth factor (EGF), 10 ng/ml recombinant human basic fibroblast growth factor (bFGF), and 1x B27 supplement. 3. Pre-warm at 37  C the ultralow-attachment plates. 4. Plate usually 500–4000 cells/cm2 cell per well in complete DMDE/F12 media. 5. Incubate ultralow-attachment plates at 37  C and 5% CO2. 6. Replace the media or refresh growth factors twice a week. 7. Mammospheres are visible after 48–72 h of culture. 8. For serial passaging, disaggregate the mammospheres with 500 μl of pre-warmed 0.5% trypsin/0.2% EDTA for 5–10 min.

42

Vitale Del Vecchio et al.

9. Add the FBS to neutralize the trypsin, wash, and count the single cells using trypan blue. 10. Seed the cells into a new ultralow-attachment plate at the same density used in the primary generation. 11. For differentiation, plate the mammospheres in standard culture condition, using DMEM/F12 supplemented with 10% FBS without growth factors and B27 supplement. 3.2.3 Lung Tumorspheres

1. Prepare lung tumorsphere Advanced DMEM/F12 basal media adding 2 mM L-glutamine, 100 U/ml penicillin, and 100 U/ ml streptomycin. 2. Prepare lung tumorsphere Advanced DMEM/F12 complete media immediately before use adding 0.4% bovine serum albumin (BSA), 50 μg/mL epidermal growth factor (EGF), 20 μg/ mL basic fibroblast growth factor (bFGF), 50 μg/mL insulin, 100 μg/mL transferrin, 0.03 mM sodium selenite, and 2% B-27 supplement. 3. Pre-warm at 37  C the ultralow-attachment plates. 4. Seed cells in 6-well plates at 60  104 density in lung tumorsphere Advanced DMEM/F12 complete media. 5. Incubate ultralow-attachment plates at 37  C and 5% CO2. 6. Replace the media or refresh growth factors twice a week. 7. Tumorspheres are visible after 48–72 h of culture. 8. For serial passaging, disaggregate mechanically the tumorspheres or use 500 μl of pre-warmed Accutase for 2–3 min. 9. Add the FBS to neutralize the Accutase, wash, and count the single cells using trypan blue. 10. Seed the cells into a new ultralow-attachment plate at the same density used in the primary generation. 11. For differentiation, seed the tumorspheres in standard culture condition, using DMEM or RPMI1640 with 10% FBS.

3.2.4

Sarcospheres

1. Prepare sarcosphere DMEM/F12 basal media adding 2 mM Lglutamine, 100 U/ml penicillin, and 100 U/ml streptomycin. 2. Prepare sarcosphere DMEM/F12 complete media immediately before use adding 1% methylcellulose, 10 μg/mL epidermal growth factor (EGF), 10 μg/mL basic fibroblast growth factor (bFGF), 10 μg/mL insulin, 10 nM progesterone, 50 μM putrescine dihydrochloride, 15 nm sodium selenite, and 13 μg/mL transferrin. 3. Pre-warm at 37  C the ultralow-attachment plates. 4. Seed cells in 6-well plates at 60  104 density in sarcosphere DMEM/F12 complete media.

Methods in Cancer Stem Cells

43

5. Incubate ultralow-attachment plates at 37  C and 5% CO2. 6. Replace the media or refresh growth factors twice a week. 7. Sarcospheres are visible after 48–72 h of culture. 8. For serial passaging, disaggregate mechanically the sarcospheres or use 500 μl of pre-warmed Accutase for 2–3 min. 9. Add the FBS to neutralize the Accutase, wash, and count the single cells using trypan blue. 10. Seed the cells into a new ultralow-attachment plate at the same density used in the primary generation. 11. For differentiation, seed the sarcospheres in standard culture condition, using DMEM or RPMI1640 with 10% FBS. 3.3 Detection and Isolation of CSCs by Flow Cytometry 3.3.1 Stemness Marker Expression

1. When the primary cells reach 70% of confluence, they are detached, counted, and resuspended in 2 ml media. 2. Spin cells at 250 g for 5 min. 3. Discard media and resuspend cells at 1  106 cells/mL in PBS 0.1% BSA. 4. Label one 12  75 mm tube for each of the following points: tube 1, control tube, and tube 2, test tube. 5. Transfer 200 μl (200,000 cells) into each tube. 6. Add 1 μg/ml of each antibody into the test tube and mix well. 7. Incubate the tubes at 4  C for 30 min in the dark, the reaction being light sensitive. 8. After a 30 min incubation, add 2 ml PBS 0.1% BSA into tubes and spin cells at 250 g for 5 min. 9. Discard PBS 0.1% BSA and add 400 μl of PBS 0.1% BSA. 10. After this, proceed with acquisition to a cytometer. 11. The same procedure is used for spheres. 12. Cells derived from fresh biopsies or heterogeneous cell populations can have different SSC or FSC characteristics and the gate must be set correctly. For example, if epithelia cells are used, it is a good practice to stain the cells with an epithelial marker (CD326 or CD324) and gated on epithelial cells.

3.3.2 Side Population Detection

1. When the primary cells reach 70% of confluence, they are detached, counted, and resuspended in 2 ml media. 2. Spin cells at 250 g for 5 min. 3. Discard media and resuspend cells at 2  106 cells/mL in PBS 0.1% BSA. 4. Label one 12  75 mm tube for each of the following points: tube 1, SP tube, and tube 2, verapamil tube. 5. Transfer 1 ml (1  106 cells) into each Falcon tube.

44

Vitale Del Vecchio et al.

6. Spin cells at 250 g for 5 min. 7. Add 5 μg/ml of Hoechst 33342 in media to both tubes (see Note 3). 8. Add 50 μM verapamil only in the verapamil tube and leave the SP tube untreated. 9. Mix well the tubes and place them in the 37  C water shaking bath for 90 min (see Note 4). 10. During incubation, mix the tubes for 15 min each. 11. After incubation, spin cells at 250 g for 5 min and resuspend in 1 ml cold PBS. 12. All further procedures must be carried out at 4  C to prohibit leakage of the Hoechst33342 probe. 13. Add 2 μg/ml PI or 7-AAD for dead cell discrimination and mix for 5 min before cytometer acquisition. 14. Hoechst 33342 probe is excited at 350 nm and the emitted fluorescence is measured at two wavelengths using 450/40 nm blue BP filter and 675/20 nm red BP filter and the dichroic mirror to separate the blue and red filters (580 LP). This is typical configuration of filters for SP acquisition. 15. Run the samples within 4 h of cell preparation. 3.3.3

ALDH Activity

3.4 Flow Cytometer Setup and Data Acquisition 3.4.1 Cytometric Detection of Stemness Markers

See chapter entitled “ALDH activity assay: a method for Cancer Stem Cells (CSCs) identification and isolation.” 1. Create a forward scatter (FCS) versus side scatter (SSC) dot plot, to select nucleated cell population in a gate (P1). 2. To ensure that a detected signal arises from single cells, cell doublets and aggregates are gated out based on their properties displayed on the FSC area (FSC-A) versus the height (FSC-H) dot plot (P2). 3. Create an FL-3 channel in log for PI detection versus FSC channel or FL-3 and or FL-2 channel for 7AAD detection versus FSC and select the viable nucleated cells negative to PI or 7AAD (P3). 4. Load a control tube (without antibodies) on the cytometer and adjust the photomultiplier voltage for each fluorescence corresponding to Abs used for phenotypic characterization. The voltages are placed at second log decade and gated on P3. 5. Remove the control tube and load the test tube. 6. Apply a compensation previously calculated on the basis of the specific tools of cytometer. For each cytometer, specific beads should be used to calculate the compensation. This is important to avoid a “false positive” due to the overlapping of emission spectrum of different fluorescensces used.

Methods in Cancer Stem Cells

45

7. Collect almost 500,000 events in P1 for each sample, without changing the instrument settings. 8. Glioblastoma cancer stem cells will be detected and selected by the following phenotype: CD133+CD44+CD45 . 9. Breast cancer stem cells will be detected and selected by the following phenotype: CD44+CD24low/-CD45 . 10. Lung cancer stem cells will be detected and selected by the following phenotype: CD133+CD44+CD45 . 11. Sarcoma cancer stem cells will be detected and selected by the following phenotype: CD133+CD45 (see Note 5). 3.4.2 Cytometric Analyses of Side Population

1. Create a forward scatter (FCS) versus side scatter (SSC) dot plot, to select in a gate (P1) nucleated cell population. 2. To ensure that a detected signal arises from single cells, cell doublets and aggregates are gated out based on their properties displayed on the FSC area (FSC-A) versus height (FSC-H) dot plot (P2). 3. Create an FL-3 channel in log for PI detection versus FSC channel or FL-3 and or FL-2 channel for 7AAD detection versus FSC and select the viable nucleated cells negative to PI or 7AAD (P3). 4. Based on Hoechst 33342 emitting properties, create a dot plot in linear for Hoechst 33342-Blue versus Hoechst 33342-Red in order to detect the presence of the SP on P3. 5. Load the SP tube on the cytometer and adjust the photomultiplier voltage for each fluorescence (red and blue) and gated on P3. 6. Given that the Hoechst 33342 dye is a DNA probe, a doubleemitting population (blue and red) placed in the center of the dot plot represents those cells in the G0/G1 phase of the cell cycle. 7. At the upper right corner of the dot plot, find cells in the S and G2/M phases of the cell cycle. 8. At the left from G0/G1 cells, the SP appears as a dim tail of fluorescence (red and blue) being the SP able to efflux the Hoechst 33342 dye (Fig. 1). 9. Collect a minimum of 100,000 live cell events to resolve specimens with scarcity of SP cells. 10. Remove the SP tube. The SP profile is reversed when staining is performed in the presence of verapamil, an ATP-binding cassette transporter inhibitor. 11. Load the verapamil tube on the cytometer without changing the parameters of photomultiplier voltages set during the acquisition of the SP tube and gated on P3.

46

Vitale Del Vecchio et al.

Fig. 1 A schematic dot plot for Hoechst Blue versus Hoechst Red simultaneous emission of a sample containing SP cells

12. At the left from G0/G1 cells, the dim tail of fluorescence (red and blue) disappears (Fig. 2) with all cells being stained with Hoechst 33342. Verapamil treatment blocks the ABC transporters, and Hoechst does not efflux from cells and bind DNA. This represents a control, and it ensures that the weak fluorescence detected from the SP tube is true and due to a presence of SP cells and not cell debris. 13. Collect a minimum of 100,000 live cell events without changing the instrument settings. 3.4.3 Cell Sorting

1. The sorting gates are established using as negative controls the cells stained with PI or 7-ADD only. 2. Before performing cell sorting, it is important exclude doublets. Doublets exclusion is to ensure counting of single cells and exclude doublets from analysis. If a doublet containing a fluorescence-positive and fluorescence-negative cell passes through the laser, it will produce a positive pulse leading to false positives in both analysis and sorting experiments. 3. Doublet exclusion is performed by plotting the height or width against the area for forward scatter or side scatter. Doublets will have double the area and width values of single cells, whilst the height is roughly the same. Therefore, disproportions between height, width, and area can be used to identify doublets.

Methods in Cancer Stem Cells

47

Fig. 2 A schematic dot plot showing characteristic Ho342 dye staining profiles in the presence of verapamil that blocks ATP transporters inducing disappearance of the dim tail of fluorescence (red and blue)

4. CSC and not-CSC fractions are sorted as the following: (i) CD133+CD44+ glioblastoma CSCs and CD133 CD44 glioblastoma not-CSCs. (ii) CD44+CD24low/ breast CSCs and CD44 CD24+ breast not-CSCs. (iii) CD133+CD44+ lung CSCs and CD133 CD44 not-CSCs.

lung

(iv) CD133+ sarcoma CSCs and CD133 sarcoma not-CSCs. (v) SP and not-SP fractions. 5. Aliquots of CSC and not-CSC sorted cells are evaluated for purity by flow cytometry. The purity must be almost 80%. 6. CSC and not-CSC sorted cell populations are cultured in standard medium, used for in vivo and in vitro experiments (see Note 6).

4 Notes 1. The enzymatic digestion time is dependent on the size and tissue texture. Moreover, optimal incubation times may vary among different cell types. Therefore, it is necessary to test different times. Suggested incubation times can be 15 min, 30 min, 45 min, 60 min, and 90 min.

48

Vitale Del Vecchio et al.

2. It is recommended to use different media (DMEM and RPMI1740) and different concentrations of FBS (5%, 10%, 20%) to establish the better culture conditions to promote cell growth and adhesion. 3. Since Hoechst 33342 is cytotoxic for cells, it is recommended to perform Hoechst saturation and toxicity dose response curves, with the optimal dye concentration lying within a plateau region, where the percentage of SP cells is stable. 4. When the staining process is over, the cells should be maintained at +4  C to prohibit further dye efflux. For best results, samples should be analyzed by flow cytometry shortly after staining. 5. For each antibody used, it is recommended to use positive control cell lines for antibody recognition and functionality. 6. Sorted cells are cultured in media supplemented with an excess of gentamicin to avoid contamination rendering the sorting semi-sterile. References 1. Tirino V, Desiderio V, Paino F et al (2013) Cancer stem cells in solid tumors: an overview and new approaches for their isolation and characterization. FASEB J 27:13–24 2. Shipitsin M, Polyak K (2008) The cancer stem cell hypothesis: in search of definitions, markers, and relevance. Lab Investig 88:459–463 3. Bonnet D, Dick JE (1997) Human acute myeloid leukemia is organized as a hierarchy that originates from a primitive hematopoietic cell. Nat Med 3:730–737 4. Al-Hajj M, Wicha MS, Benito-Hernandez A et al (2003) Prospective identification of tumorigenic breast cancer cells. Proc Natl Acad Sci USA 100:3983–3988 5. Singh SK, Clarke ID, Terasaki M et al (2003) Identification of a cancer stem cell in human brain tumors. Cancer Res 63:5821–5828 6. Fang D, Nguyen TK, Leishear K et al (2005) A tumorigenic subpopulation with stem cell properties in melanomas. Cancer Res 65: 9328–9337 7. Tirino V, Desiderio V, Paino F et al (2011) Human primary bone sarcomas contain CD133+ cancer stem cells displaying high tumorigenicity in vivo. FASEB J 25:2022– 2030 8. Collins AT, Berry PA, Hyde C et al (2005) Prospective identification of tumorigenic prostate cancer stem cells. Cancer Res 65:10946– 10951

9. Bapat SA, Mali AM, Koppikar CB et al (2005) Stem and progenitor-like cells contribute to the aggressive behavior of human epithelial ovarian cancer. Cancer Res 65:3025–3029 10. Takaishi S, Okumura T, Tu S et al (2009) Identification of gastric cancer stem cells using the cell surface marker CD44. Stem Cells 27: 1006–1020 11. Tirino V, Camerlingo R, Franco R et al (2009) The role of CD133 in the identification and characterization of tumor-initiating cells in non-small-cell lung cancer. Eur J Cardiothorac Surg 36:446–453 12. Dean M, Fojo T, Bates S (2005) Tumor stem cells and drug resistance. Nat Rev Cancer 5: 275–284 13. Ming Zhou HM, Zhang JG, Zhang X et al (2021) Targeting cancer stem cells for reversing therapy resistance: mechanism, signaling, and prospective agents. Signal Transduct Target Ther 6:62 14. Goodell MA, Brose K, Paradis G et al (1996) Isolation and functional properties of murine hematopoietic stem cells that are replicating in vivo. J Exp Med 183:1797–1806 15. Wu C, Alman BA (2008) Side population cells in human cancers. Cancer Lett 268:1–9 16. Herreros-Pomares A, De-Maya-Girones JD, ˜ as S et al (2019) Lung tumorCalabuig-Farin spheres reveal cancer stem cell-like properties and a score with prognostic impact in resected

Methods in Cancer Stem Cells non-small-cell lung cancer. Cell Death Dis 10: 660 17. Weiswald LB, Richon S, Massonnet G et al (2013) A short-term colorectal cancer sphere culture as a relevant tool for human cancer biology investigation. Br J Cancer 108:1720– 1731 18. Xu X, Chai S, Wang P et al (2015) Aldehyde dehydrogenases and cancer stem cells. Cancer Lett 369:50–57 19. Ma S, Chan KW, Lee TKW et al (2008) Aldehyde dehydrogenase discriminates the CD133

49

liver cancer stem cell populations. Cancer Res 6:1146–1153 20. Marcato P, Dean CA, Pan D et al (2011) Aldehyde dehydrogenase activity of breast cancer stem cells is primarily due to isoform ALDH1A3 and its expression is predictive of metastasis. Stem Cells 29:32–45 21. Desiderio V, Papagerakis P, Tirino V et al (2015) Increased fucosylation has a pivotal role in invasive and metastatic properties of head and neck cancer stem cells. Oncotarget 6:71–84

Chapter 4 Surface Markers for the Identification of Cancer Stem Cells Tasfik Ul Haque Pronoy, Farhadul Islam, Vinod Gopalan, and Alfred King-yin Lam Abstract Cancer stem cells have genetic and functional characteristics which can turn them resistant to standard cancer therapeutic targets. Identification of these cells is challenging and is done mainly by detecting the expression of antigens specific to stem cells. Currently, there is a significant number of surface markers available which can detect cancer stem cells by directly targeting the specific antigens present in cells. These markers possess differential expression patterns and sub-localizations in cancer stem cells compared to nonneoplastic and somatic cells. In addition to these biomarkers, multiple analytical methods and techniques, including functional assays, cell sorting, filtration approaches, and xenotransplantation methods, are used to identify cancer stem cells. This chapter will overview the functional significance of cancer stem cells, their biological correlations, specific markers, and detection methods. Key words Stem cell, Cancer, Detection, Markers, Methods Isolation

1

Introduction Cancer stem cells (CSCs) or tumor-initiating cells can be defined as a group of undifferentiated cells within a tumor that can self-renew and drive carcinogenesis [1]. These cells possess distinct genetic, epigenetic, and functional heterogeneity which in turn can resist therapy and may later initiate metastatic process [2–4]. Previous studies have proved that increased genetic instabilities in stem cells may lead to the formation of CSCs suggesting that stemness is attained from the additional genetic modifications [5, 6]. CSCs represent approximately 0.1% to 10% of all cancer cells, and they express characteristically low levels of their antigens compared to established cancer-associated antigens [7]. CSC antigens were identified not based on their overexpression but on their expression in a group of cancer cells with stem-cell-like properties [8–11]. The origin and development of CSCs are highly regulated by a key cellular process called epithelial–mesenchymal transition

Federica Papaccio and Gianpaolo Papaccio (eds.), Cancer Stem Cells: Methods and Protocols, Methods in Molecular Biology, vol. 2777, https://doi.org/10.1007/978-1-0716-3730-2_4, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

51

52

Tasfik Ul Haque Pronoy et al.

(EMT) which is a key determinant of cancer growth and metastasis [1, 4]. EMT makes the epithelial cells polarize and develop stemlike properties by losing their epithelial characteristics and gaining mesenchymal properties, leading to increased invasion and motility. Thus, detecting EMT changes during or before cancer development provides a valuable prognostic tool and therapeutic target [3]. Also, the increased expressions of some of these EMT markers in cancer tissues may be used to predict the response of targeted therapy developed against these groups of cancer cells. Thus, patients’ morbidity, quality of life, and survival rates could be lower by developing exclusive therapies that target CSCs, particularly for patients that suffer from distant metastatic disease. The existence of CSCs in vivo was first confirmed in 1997 in acute myeloid leukemia by Bonnet and Dick [8]. The authors have identified CD34+ CD38
subpopulation cells in those with leukemia, like normal hematopoietic stem cells, and were proficient in initiating acute myeloid leukemia once transplanted into nonobese diabetic/severe combined immune-deficient mice [8]. Since then, CSCs have been identified in several cancers using specific biomarkers. It was demonstrated that a general panel of biomarkers for detecting all CSCs are unlikely as CSCs are generally tissue specific [1]. Also, varied expression levels and co-expression of antigens from normal stem cells have made the detection of antigens specific for identifying or targeting CSCs difficult [7]. Every cell surface in the body is coated with specialized proteins, namely, receptors, which could selectively bind or adhere to other signaling molecules [12]. The identification of receptors is generally based on the increased expression of specific markers in different cell types [13]. These receptors are noted in specific cell types including cancer stem cells [12]. Expressions of specific cell surface antigens that enrich cells with CSC properties remain the most common approach to identifying CSCs [13]. Many of these antigens are primarily targeted due to their identified expressions on endogenous stem cells. Currently, no universal marker can identify CSCs in every cancer. Therefore, multiple genetic markers are used in combination to achieve high specificity in CSC detections. A brief overview of common markers used in multiple cancers is listed in Table 1. 1.1 CSC Surface Markers

In recent years, there has been a significant increase in the number of surface marker molecules to detect and characterize CSCs. Till now, CSCs and their specific markers have been identified from various cancers, including those from the breast, brain, blood (leukemia), prostate, skin, thyroid, lung, gastrointestinal tract, reproductive tract, and head and neck. Stemness activity or presence is generally defined by either increased expression or absence of certain markers [14]. Also, many markers expressed in nonneoplastic

Surface Markers for the Identification of Cancer Stem Cells

53

Table 1 Cell surface markers for the identification of cancer stem cells in common cancers Marker

Type

Cancers

Reference

CD13

Enzyme

Breast, liver, cervix

[31]

CD15

Carbohydrate antigen

Brain, stomach

[32, 34]

CD19

Membrane protein

Blood

[32]

CD20

Membrane protein

Blood, melanoma

[31, 32, 34]

CD24

Membrane protein

Breast, liver, lung pancreas, colon, ovary, head and neck, stomach, esophagus

[30–35]

CD26

Enzyme

Blood, colon

[30, 32, 33]

CD29

Adhesion molecule

Breast, colon, cervix

[30–34]

CD34

Membrane protein

Blood, lung

[30–34]

CD38

Membrane protein

Blood

[30, 32, 34]

CD44

Membrane protein

Breast, liver, lung, pancreas, prostate, colon, ovary, blood, brain, head and neck, bladder, stomach, cervix, esophagus

[30–35]

CD47

Membrane protein

Leukemia, bladder

[32]

CD49f

Adhesion molecule

Breast, liver, prostate, brain

[30–34]

CD61

Adhesion molecule

Breast

[31, 33]

CD66c

Adhesion molecule

Bladder

[32]

CD87

Receptor protein

Lung

[31, 33]

CD90

Membrane protein

Breast, liver, lung, blood, brain, head and neck, stomach, esophagus

[30–35]

CD105

Membrane protein

Ovary, cervix

[32]

CD117

Receptor protein

Ovary, lung, liver, blood

[30–34]

CD123

Receptor

Blood

[30, 32, 33]

CD133

Membrane protein

Brain, liver, lung, colon, pancreas, prostate, ovary, melanoma, head and neck, brain, stomach, kidney

[30–34]

CD151

Membrane protein

Prostate

[32]

CD166

Adhesion molecule

Lung, prostate, colon, melanoma

[30–33]

CD271

Receptor protein

Melanoma, esophagus, head and neck

[31, 34, 35]

EPCAM

Adhesion molecule

Breast, liver, lung, pancreas, colon, ovary, brain, stomach

[30–34]

FUT4

Enzyme

Brain

[30, 31]

LGR5

Receptor protein

Breast, stomach, colon, head and neck

[30–33] (continued)

54

Tasfik Ul Haque Pronoy et al.

Table 1 (continued) Marker

Type

Cancers

Reference

CXCR4

Receptor protein

Brain, breast, stomach, pancreas

[31–33]

ABCB5

Transporter

Melanoma

[31]

AFP

Binding protein

Liver

[31, 33]

TIM3

Receptor protein

Blood

[32, 33]

LINGO2

Binding protein

Stomach

[34]

EPCAM Epithelial cellular adhesion molecule, FUT4 Fucosyltransferase 4, LGR5 Leucine-rich repeat-containing G-protein coupled receptor 5, CXCR4 C-X-C chemokine receptor type 4, ABCB5 ATP-binding cassette sub-family B member 5, AFP Alpha-fetoprotein, TIM3 T cell immunoglobulin and mucin domain-containing protein 3

Table 2 Cell surface markers expresses on normal stem cells and cancer stem cells Marker

Normal expression

Cancer stem cell expression

CD19

B lymphocytes

B cell malignancies

CD20

B lymphocytes

Melanoma

CD24

B lymphocytes; neuroblasts

Pancreatic carcinoma, lung carcinoma

CD34

Haemopoietic cells; endothelial cells Haemopoietic malignancies

CD38

B and T lymphocytes

Negative on acute myeloid leukemia

CD44

Multiple tissues

Breast, liver, head and neck, pancreas carcinomas

CD90

T lymphocytes, neurons

Liver carcinoma

CD133

Multiple tissues

Brain, colorectal, lung, liver carcinomas

EpCAM/ESA Epithelial cells

Colorectal and pancreas carcinomas

ABCB5

Melanoma

Keratinocytes

ABCB5 ATP-binding cassette subfamily B member 5

stem cells are expressed in cancer stem cells (Table 2). The detection and specificity of these markers depend upon the type of stem cells they target. For example, some markers can specifically differentiate embryonic stem cells from adult stem cells or pluripotent from progenitor cells [14]. At present, there is no clear distinction between the expression patterns and detection of stem cell markers across nonneoplastic and cancer stem cells. Oncofetal stem cell markers, defined as markers undetectable in adult tissues but expressed in embryos or fetuses and with re-expression in cancers, are currently surfaced as the best option for the identification of cancer stem cells [14, 15].

Surface Markers for the Identification of Cancer Stem Cells

55

Majority of stem cell markers identified so far are proteins; a small group of markers are noted to be glycans (e.g., glycoprotein) bound to proteins or lipids. For example, CD15 is a stage-specific embryonic antigen that is carried on lipids or proteins in CSCs from glioblastoma [16]. These markers, through their various genetic modifications, could be physically isolated as distinct subpopulations of CSCs with differing biological features. Overall, these markers either are upregulated (e.g., ATP-binding cassette [ABC] transporter G subfamily [ABCG]) [17], are absent, are mutated (e.g., Notch homolog 1, translocation-associated [Drosophila] [NOTCH1]) [18], or show differential expression patterns with their isotypes (e.g., CD44v) [19]. Another interesting feature of cancer stem cell markers include their difference in physical localizations such as cell surface or intracytoplasmic (in nucleus/cytoplasmic). CSC markers, such as CD133, represent a cell surface antigen that is expressed by its corresponding adult stem cells. Meantime, aldehyde dehydrogenase 1 (ALDH1) is in the cytoplasm. It is a soluble protein and has been used to detect CSCs in various cancers, including leukemia [20], breast cancer [21], and colon cancer [22]. Hoechst 33342 is a nuclear dye, and it is used for the detection of nonneoplastic stem cells as well as CSCs in SP (side population) cells [23]. These markers’ differential expression patterns and sub-localizations imply that CSCs possess complex epigenetic and genetic alterations. 1.2 Isolation and Detection

Isolation and detection of CSCs depend on their specific biological and molecular features, such as the ability of sphere formation, generation of new tumors in mice, and positivity/differential expression patterns of stemness markers. The current analytical techniques for CSC detection focus on functional, image-based, molecular, cytological sorting, filtration approaches, and xenotransplantation [24–26]. A graphical illustration of CSC isolation and characterization is illustrated in Fig. 1. Functional differences between CSCs and non-CSCs are identified in colony formation, sphere formation, side population (SP) analysis, aldehyde dehydrogenase (ALDH) activity, and therapy resistance assays [27]. In addition, CSCs can be specifically isolated from non-CSCs by sorting these cells by cell sorting methods like flow cytometry and magnetic activated cell sorting (MACS) using CSC-specific cell surface markers [28]. Furthermore, molecular methods to identify the expression (mRNA and protein) profiling of CSC markers using quantitative polymerase chain reaction (PCR), immunocytochemistry, immunofluorescence, and immunohistochemistry are also widely used by researchers to confirm the localization of the surface markers [24]. Recent studies have demonstrated that the gold standard for the confirmation of CSCs is xenotransplantation into

56

Tasfik Ul Haque Pronoy et al.

Fig. 1 Isolation and characterization of CSCs using fluorescence-activated cell sorting (FACS), in vitro, and in vivo techniques. Surface markers play a key role in further characterizing or confirming CSCs. These cells can later undergo various in vitro and in vivo xenograft models to validate their biological characteristics

immune-deficient animals [29]. These methods are practiced in combination to achieve a sensitive and specific detection of CSC isolation and detection in various tumors. Detailed methodology and steps are described below.

2 Materials 2.1

Cell Culture

1. Primary cancer cell lines. 2. Culture medium: Roswell Park Memorial Institute (RPMI) Medium or, Dulbecco’s Modified Eagle Medium (DMEM). 3. 10% fetal bovine serum (FBS). 4. 1% of 100 g/mL streptomycin.

Surface Markers for the Identification of Cancer Stem Cells

57

5. 100 U/mL penicillin. 6. Trypsin–ethylenediaminetetraacetic acid (EDTA) solution. 7. L-Ascorbic acid. 8. L-Glutamine. 9. Bovine serum antigen. 10. Transferrin. 11. Insulin. 12. 2-Mercaptoethanol (see Note 1). 2.2 Single-Cell Suspension

1. Single-cell suspension kit. 2. Collagenase or hyaluronidase. 3. DNase I solution (1 mg/mL). 4. Ammonium chloride solution. 5. RPMI 1640 medium. 6. 35 mm culture dish. 7. 14 mL round-bottom tube. 8. 70 μm strainer. 9. 50 mL conical tube. 10. Razor blade. 11. Scalpel or scissor. 12. Pipettor. 13. Pipette tips. 14. Serological pipette controller. 15. Serological pipettes. 16. Recommended medium (see Note 2).

2.3 Isolation and Characterization of CSCs 2.3.1

Flow Cytometry

1. A single-cell suspension of cancer cells. 2. Sample buffer: (a) Phosphate-buffered saline (PBS) solution (pH 7.4). (b) 0.1% bovine serum albumin or 0.5% fetal bovine serum. 3. Fluorophore-conjugated desirable antibodies against specific surface markers. 4. Respective isotypic fluorophore-conjugated antibodies. 5. Cell viability staining kit (e.g., LIVE/DEAD™ fixable dead cell stain sampler kit). 6. UltraComp eBeads (compensation beads). 7. 5 mL polystyrene tubes. 8. Culture media.

58

Tasfik Ul Haque Pronoy et al.

9. Ice bath. 10. 0.8 μm cell strainer. 11. Flow cytometer (see Note 3). 2.3.2 Aldefluor Assay Conjugated Flow Cytometry

1. Aldefluor™ kit. 2. Fluorescent Note 4).

substrate

(BODIPY-aminoacetaldehyde)

(see

3. Diethylaminobenzaldehyde (DEAB). 4. Pan-aldehyde dehydrogenase inhibitor (see Note 5). 5. Dimethyl sulfoxide (DMSO). 6. 2 N hydrochloric acid (HCl). 7. Cell sorter (flow cytometry facility). 8. RPMI medium, supplemented with: • 20 ng/mL epidermal growth factor (EGF). • 20 ng/mL basic fibroblast growth factor (bFGF). • B27 supplement (see Note 6). 9. Labelled antibodies (see Note 7). 2.3.3 Magnetic Separation

1. Positive selection kit. 2. Single-cell suspension of cancer cells. 3. Mouse IgG against specific surface marker (e.g., mouse antihuman CD44 IgG, mouse anti-human CD24 IgG). 4. Phosphate-buffered saline (PBS) solution (pH 7.4). 5. 2% fetal bovine serum (FBS). 6. 1 mM EDTA (sample buffer). 7. Magnetic stand. 8. 5 mL polystyrene tubes. 9. Water bath or incubator. 10. Biosafety cabinet level 2.

2.4 Detecting CSC Functions

1. Isolated CSC population.

2.4.1 SphereForming Assay

3. Serum-free DMEM or, RPMI medium.

2. 6-well ultralow-attachment culture plates. 4. 20 ng/mL epidermal growth factor (EGF). 5. 20 ng/mL fibroblast growth factor (FGF). 6. B27 supplement. 7. Microscope (see Note 8).

Surface Markers for the Identification of Cancer Stem Cells 2.4.2

Western Blotting

59

1. 200–600 μL protein buffer comprised of 10% cell lysis buffer. 2. 10% protease inhibitor cocktail. 3. 80% distilled water. 4. Bicinchoninic acid (BCA) protein assay kit. 5. 10% 2D polyacrylamide gel electrophoresis (PAGE) gel. 6. Gel electrophoresis. 7. 1x Laemmli running buffer (100 mL of 10 Laemmli stock in 1 L of distilled water). 8. 40 ug of extracted protein from CSCs. 9. Coomassie brilliant blue dye mixture. 10. Protein standard ladder. 11. Polyvinylidene fluoride (PVDF) membrane. 12. 1 blotting buffer (Tris/lysate/10% methanol). 13. Milk power solution. 14. Tris-buffered saline and Tween 20 (TBST) buffer (see Note 9).

2.4.3

Immunostaining

1. 12-well culture plates, or desired plates. 2. Sterile coverslips. 3. 70% cold ethanol. 4. 5% of bovine serum albumin (BSA). 5. Phosphate-buffered saline (PBS) solution. 6. Antibody-coated immunofluorophore. 7. Fluorescence stain: 40 ,6-diamidino-2-phenylindole (DAPI). 8. Mounting media. 9. Confocal microscope.

2.4.4 Polymerase Chain Reaction (PCR)

1. mRNA/total RNA extraction kit: • Cell lysis buffer. • Wash buffers. • Filter columns. • Mini columns. • Collection tubes. • Elution tubes. 2. Beta-mercaptoethanol. 3. RNase-free DNase I solution. 4. RNase-free water. 5. Nano-drop to measure RNA quality and concentration. 6. Gene-specific forward and reverse primers (see Note 10).

60

Tasfik Ul Haque Pronoy et al.

7. cDNA synthesis kit. 8. qPCR/RT-PCR Master Mix. 9. Real-time PCR machine.

3

Methods

3.1 Preparation of Single-Cell Suspension

1. Prepare 5 mL of tumor digestion medium by combining 500 μL collagenase/hyaluronidase, 750 μL DNase I solution (1 mg/mL), and 3.75 mL RPMI 1640 medium. Mix thoroughly and warm at 15–25  C. 2. Harvest the tumor tissue into a dish. 3. Slice the tumor tissue into small pieces (2 mm) using a razor blade, scalpel, or scissor and transfer to a 14 mL round-bottom tube containing tumor digestion medium (prepared in step 1). 4. Incubate at 37  C for 25 min on a shaking platform. 5. Place a 70 μm nylon mesh strainer on a 50 mL conical tube and rinse with the recommended medium. Transfer the digested tumor tissue into the strainer. Using the rubber end of a syringe plunger, push digested tumor tissue through the strainer. Rinse the strainer with recommended medium, and then top up the tube to 50 mL with the recommended medium. 6. Spin at 300  g for 10 min at room temperature and discard the supernatant. 7. Add 10 mL of ammonium chloride solution to the cell pellet. Incubate at room temperature for 5 min. 8. Top up to 50 mL with the recommended medium. Centrifuge at 300  g for 10 min at room temperature and discard the supernatant. 9. Resuspend cells at 5  107 cells/mL in recommended medium (see Note 11).

3.2 Isolation and Detection 3.2.1 Flow CytometryBased Method

Flow cytometry with sorting capabilities is the technique for identifying and isolating cell populations based on the expression of cell surface markers that allow potentially positive and negative selection. The key steps involved in this method are listed below in sequential order. 1. Count the total number of cells in the single-cell suspension using preferred cell counting methods (manual or automated). In subsequent steps, samples should be handled on ice or maintained at 4  C. 2. Transfer 2 mL cell suspension (1  107 cell/mL) to a 15 mL centrifuge tube and add 4 μL of LIVE/DEAD stain. Mix gently and incubate for 10 min on ice.

Surface Markers for the Identification of Cancer Stem Cells

61

3. Add 8 mL of sample buffer to the tube, mix gently, and centrifuge at 1200 g for 3 min. Discard the supernatant. 4. Repeat step 3 and resuspend the cells in 1 mL of sample buffer. 5. Split the solution equally into two 5 mL tubes. 6. Add fluorophore-conjugated antibodies (e.g., fluorescein [FITC]-conjugated CD44 and phycoerythrin [PE]-conjugated CD24 antibodies) for CSC markers to one tube and isotype controls to another. Mix gently and incubate for 30–45 min at 4  C in the dark (see Note 12). 7. Add 4 mL of sample buffer to each tube and centrifuge at 1200  g for 3 min at 4  C. Discard the supernatant and repeat this step twice. 8. Resuspend the cells in an ice-cold sample buffer at a maximum of 1  107 cells/mL concentration and pass through a cell strainer to remove large cell clumps. 9. Run the comp-beads on the flow cytometer to adjust the fluorescent compensation. 10. Run samples on a flow sorter (see Note 13). 3.2.2 Aldefluor Assay Conjugated Flow Cytometry Method

The primary objective of this method is to identify and isolate the Aldefluor bright cells and CD44 or other desired stem cell markers in a high population from the selected cancer cell groups. The key steps involved in this method are listed below in sequential order. 1. Aldefluor reagent should be resuspended in 25 μL dimethyl sulfoxide (DMSO) and allowed to stand for 1 min at room temperature. Then add 25 μL of 2 N hydrochloric acid (HCl) and incubate at room temperature for 15 min after mixing. Add 360 μL of Aldefluor assay buffer to this mixture and store at 2 to 8  C. 2. Prepare the single-cell suspension (1  106 cells/mL) and resuspend this cell suspension with 1 mL of the Aldefluor assay buffer containing cell viability stain. Prepare the “test” and “control” tubes of selected cell populations. Add 1 mL of the cell suspension into each of the tubes. Add 5 μL of the Aldefluor DEAB reagent into the “control” tube and add the same volume of the earlier activated Aldefluor reagent to each of the “test” samples, mix, and transfer 0.5 mL of the mixture to the DEAB “control” tube. Incubate the samples for 30 to 45 min at 37  C. Then centrifuge for 5 min at 250 g and remove the supernatant. After washing, the cell pellets should be resuspended again in 0.5 mL of Aldefluor assay buffer and stored on ice for flow cytometry. 3. Antibody labelling: Dive each cell line into two tubes labelled as “test” and “control.” The test tube cells will be treated with CD44-FITC conjugated antibody, while the control test tube

62

Tasfik Ul Haque Pronoy et al.

cells should be treated with retinoic acid early inducible 1 (RAE-1) control (S)-FITC mouse isotype control. RAE are expressed in mouse embryos and cancer cell lines but absent from adult tissues. Incubate the cells on ice for 30 min and keep in the dark with proper wrapping in aluminum foil for better efficiency. Afterwards, centrifuge the cells at 800 rpm for 10 min at 4  C. Remove the supernatant and resuspend the cells in 1% bovine serum albumin in phosphate-buffered saline (PBS) solution (100 mL). 4. Cell sorting: Elute the Aldefluor tested cells into 50 mL falcon tubes with their recommended culture media, FBS, and 1% of penicillin–streptomycin antibiotic. During sorting, cells expressing high levels of aldehyde dehydrogenase (ALDH) should be isolated and cultured to a healthy confluence of ~85% before the CD44 antibody sorting. After CD44 sorting, cells will be eluted into 1.5 mL Eppendorf tubes based on their CD44 expression. Cells expressing high levels of CD44 can be labelled as CSCs while the cells expressing lower levels of CD44 can be used as control cells. To avoid the cells clumping during the cell sorting, they will be resuspended in a buffer comprising of PBS, 10% FBS, and 2 mM EDTA. 3.2.3 Magnetic Separation-Based Method

Magnetic separation techniques perform positive and negative selection, providing an alternative to flow cytometry-based CSC isolation. However, it cannot simultaneously identify and separate a combination of populations. A serial selection is required to employ to increase the separation time compromising cell viability. There are kit suppliers for magnetic beads and magnetic cell sorting techniques, including the EasySep system (Stem Cell Technologies), Dynabeads (Life Technologies), MojoSort (Biolegend), and magnetic-activated cell sorting (Miltenyi Biotech). The protocol involving immunomagnetic separation using EasySep Kit (Stem Cell Technologies) is listed sequentially: 1. Cocktail preparation: Add mouse IgG1 monoclonal antibody to a 1.5 mL polypropylene tube (not provided in the kit). Transfer 100 μL component A to the polypropylene tube and mix well. Add 100 μL component B. Mix gently and incubate at 37  C for 5 h overnight. Add 100 μL component C and mix well. Top up the polypropylene tube to 1 mL using sterile PBS solution. 2. Resuspended 1  107–1  108 cell in 0.25–2.5 mL PBS containing 2% FBS solution and transfer to 5 mL polystyrene tubes. 3. Add 100 μL cocktail per mL of sample. Mix gently and incubate for 15 min at room temperature.

Surface Markers for the Identification of Cancer Stem Cells

63

4. Add 50 μL rapid spheres per mL of sample and mix gently. Vortex rapid spheres before use. 5. Incubate the sample for 10 min at room temperature. 6. Top up the sample to 2.5 mL using PBS containing 2% FBS solution and mix gently. Place the tube (without lid) into the magnet for 5 min and discard the supernatant without removing the tube from the magnet, resulting in cell isolation. 7. Remove the tube from the magnet and repeat steps 6 two to three times to increase the purity. 8. Repeat the whole protocol for any of the additional markers. 9. Use an enrichment kit for immunomagnetic negative selection if a negative selection is required (e.g., CD44+, CD24-). 3.3 In Vitro Growth Properties of CSCs

Tumorsphere formation assay is the most performed method to confirm the enrichment of cancer stemness and self-renewal capacity. A brief working protocol is mentioned below: 1. After sorting, take isolated CSCs in ultralow-attachment plates at a density of ~1000 viable cells per well. 2. Growth of these cells should be maintained in serum-free medium, supplemented with 20 ng/mL EGF, 20 ng/mL bFGF, and B27 supplement (1:50 ratio) at 37  C in 5% carbon dioxide. 3. Tumorspheres can be collected carefully after 7–10 days of growth and centrifuged gently (800 rpm/min) for 5 min (see Note 14).

3.4 Molecular Expression of CSC Markers

Polymerase chain reaction (PCR), Western blot, and immunological staining are the commonly used methods to detect the expression and localization of surface markers in CSC after isolation. The key steps involved in each method are detailed below.

3.4.1

CSCs can be characterized at molecular level using gene expression analysis performed by sequential methods such as mRNA extraction from isolated CSCs, cDNA synthesis, and finally RT-PCR (real-time PCR). Protocols for these methods are briefed bellow.

PCR Technique

Extraction of Total RNA

1. Collect 1–5  106 cells by centrifuging from single-cell suspension at 300 g for 5 min at 4  C and discard the supernatant (see Note 15). 2. Then add 350 μL of FARB buffer and 3.5 μL of ß-mercaptoethanol to the cell pellet and vortex for 1 min (see Note 16). 3. Transfer the sample mixture to the filter column which is placed in a collection tube and then centrifuge at 18,000 g for 2 min.

64

Tasfik Ul Haque Pronoy et al.

4. Measure the volume of the supernatant and transfer it to a microcentrifuge tube (see Note 17). 5. Add an equal volume of 70% RNase-free ethanol solution and vortex for good mixing. 6. Transfer the ethanol mixed sample mixture (including any precipitate) to the mini-column that is placed into a collection tube and centrifuge at 18,000 g for 30 seconds. Discard the flow through and place the mini-column back into the collection tube. 7. DNase treatment: Add 250 μL of wash buffer 1 to the minicolumn and centrifuge at 18,000 g for 30 seconds. Discard the flow through and return the mini-column back to the collection tube. Add 50 μL of RNase-free DNase I solution (0.5 U/μL) to the center of the FARB mini-column and wait for 15 min. Then add 250 μL of wash buffer 1, centrifuge at 18,000 g for 30 seconds, and discard the flow through. 8. Add 750 μL of wash buffer 2 to the mini-column, centrifuge for 30 seconds at 18,000 g, discard flow through, and return the column back to the collection tube (see Note 18). 9. Wash again by repeating step 8. 10. Dry mini-column by centrifuging for 3 min at 18,000 g and place the column to an elution tube. 11. Add 30–50 μL of RNase-free water to the membrane center of the mini-column (see Note 19) and wait for 1 min at room temperature. 12. Centrifuge mini-column for 30 seconds at 18,000 g to elute RNA into the elution tube. Measure the eluted RNA quality and purity by quantifying absorbance at 260 nm (A260) and 280 nm (A280) with a spectrophotometer (NanoDrop) and store at
70  C. cDNA Synthesis

The first-strand cDNA can be converted up to 5 μg from extracted RNA using a Reverse Transcription System kit by the following protocol: 1. Mix up to 5 μg/reaction of experimental RNA and 10–20 pmol/reaction of gene-specific primer of the cell surface markers (e.g., CD44, CD133, etc.) and make the final volume of 5 μL using nuclease-free water (see Note 20). 2. Incubate at 70  C for 5 min and chill immediately in ice water for at least 5 min. Then store the mixture on ice after centrifuging for 10 s. 3. Preparation of 15 μL reverse transcription reaction mix: Mix 4.0 μL reaction buffer, 1.2–6.4 μL 1.5–5.0 mM conc. MgCl2,

Surface Markers for the Identification of Cancer Stem Cells

65

1.0 μL 0.5 mM PCR nucleotide mix, 20 units recombinant RNasin ® ribonuclease inhibitor, and 1.0 μL reverse transcriptase. Make 15 μL final volume using nuclease-free water. 4. Make 20 μL solution using 15 μL of reverse transcription mix and 5 μL of RNA and primer mix (prepared in step 1). 5. The annealing step should be performed at 25  C for 5 min and the extension step at 42  C for up to 1 h. Reactions should be stopped after the extension step and frozen at
20  C for storage until the quantities of cDNA are analyzed using qPCR Master Mix. Real-Time Polymerase Chain Reaction

Once RNA is converted to cDNA, the quantitative real-time PCR can be performed to detect the mRNA expression changes. PCR results of the CSC markers should be normalized to control genes such as alpha-tubulin, beta-actin, or glyceraldehyde 3-phosphate dehydrogenase (GAPDH). Detection of annealing and melting temperature can be obtained commercially. For example, in using qPCR Master Mix 25–30 cycles are enough for optimal amplification of desired products. The steps of denaturation, annealing, and extension are briefed below: 1. For the initial denaturation step, 2 min at 95  C is sufficient, and subsequent denaturation steps are performed between 15 s and 1 min. 2. The annealing temperature is below 5  C of the calculated melting temperature of primers and performed typically 15 s to 1 min. 3. The extension reaction is typically performed at 72–74  C, which is the optimal temperature for Taq DNA polymerase activity and allows approximately 1 min for every 1 kb of DNA to be amplified. A final extension is recommended for 5 min at 72–74  C.

3.4.2

Western Blot

This method requires the extraction of total proteins from tumorspheres and from control cells. Approximately 40 ug of the total protein should be used for this analysis. The key steps involved in this method are detailed below: 1. The total proteins were separated using 4–16% polyacrylamide gel and transferred to the polyvinylidene difluoride (PVDF) membrane. 2. Block the PVDF membrane using 5% of nonfat milk for 90 min at room temperature before incubating overnight with the CSC-specific antibodies at 4 C. 3. Wash the membrane thrice with PBS-T and incubate with a secondary antibody at room temperature for 30–90 min.

66

Tasfik Ul Haque Pronoy et al.

4. Detect the protein bands by enhanced chemiluminescence (ECL) and visualize them using an imaging system (see Note 21). 3.4.3 Immunological Analysis

Cellular distribution and qualitative expression of the stem cell markers can be confirmed with immunofluorescence analysis. The protocol is given below: 1. Firstly, seed the CSC spheres or cells directly on the sterile coverslips at a density of ~1  104 cm2 and incubate them at 37  C for 16 h. 2. After incubation, wash the cells with ice-cold PBS solution. 3. Add 70% cold ethanol and incubate for 30 min to fix the cells. 4. After incubation, subsequently rewash with PBS solution and block with 5% of BSA solution for 1 h at room temperature. 5. After washing, incubate the cells with desired CSC markers’ antibodies at 4 C overnight. 6. With PBS solution, cells should be rinsed again and incubated with a secondary antibody for 90 min at room temperature in the dark. 7. After a final rinse with PBS-T three times for 10 min on a shaker, remove the coverslips with sterilized forceps and affix them on a glass slide with a drop of DAPI (and mounting media). 8. Slides should dry for 48 h and then be observed by a confocal microscope.

4

Notes 1. Fetal bovine serum (FBS), streptomycin, penicillin, and trypsin–ethylenediaminetetraacetic acid (EDTA) solution are essential for cell passage in efficient cell culturing. 2. Recommended medium may vary depending on the downstream application. 3. Flow cytometer is required with the appropriate channels for employing fluorochromes. A sorter is necessary for sorting the CSCs. A sterile site must be provided if the sorted cells are to be cultivated. Before beginning an experiment, the flow cytometry equipment must be facilitated. 4. The aldehyde dehydrogenase activity is measured via the breakdown of a fluorescent substrate (BODIPYaminoacetaldehyde). 5. A pan-aldehyde dehydrogenase inhibitor is employed as a control to determine the specific number of cells with strong ALDH activity.

Surface Markers for the Identification of Cancer Stem Cells

67

Fig. 2 Tumorsphere formation of CSC after isolation from fluorescence-activated cell sorting (FACS)

6. B27 supplements can be aliquoted and stored at
20  C that can be used within 2 weeks. 7. Based on the sensitivity of this Aldefluor assay, one to three or more antibodies can be used for labelling the stem cells from somatic cancer cell lines. Antibodies commercially available include CD24, CD29, CD44, CD45, CD90, CD133, CD326, and cytokeratin. A FITC conjugated RAE (retinoid acid early inducible gene) control (S) antibody clone REA293 (a universal isotype control) is required for recognizing the cell surface antigens. 8. B27 supplement should be maintained at 1:50 ratio and other recommended growth factors or supplements will be needed for this assay. After 48–72 h of culture, spheres can be visualized with inverted phase contrast (Fig. 2). 9. The membrane will be developed using the VersaDoc-MP Imaging System. 10. The primers for the PCR experiment could be purchased commercially or made custom using primer BLAST tool from the National Center for Biotechnology Information. 11. For higher TIL purity, resuspend cells at 5–10  10^7 cells/ mL; for higher TIL recovery, resuspend cells at 1–2.5  10^7 cells/mL in the recommended medium. 12. A mixture of antibodies should be prepared in advance and titrate the antibody concentrations according to the manufacturer’s instructions. 13. The collection tubes must be 25% filled with culture media. To confirm the purity, run a small sample of the isolated cells through the cytometer after sorting. The cells should retain around 95% purity with some signal quenching.

68

Tasfik Ul Haque Pronoy et al.

14. Cells obtained from dissociation can be sieved through a 40 μm sieve and analyzed microscopically for single cellularity. If groups of cells were present at a frequency >150 for 10,000 single cells, mechanical dissociation and sieving can be repeated. 15. Too much sample will make incomplete cell lysis and resulting lower RNA yield and purity. 16. If the clump is still visible after vortexing, pipet the sample mixture up and down to break down the clump for proper cell suspension. 17. During transferring the supernatant, avoid any debris and pellet to pipet. 18. Check to ensure that ethanol has been added to the wash buffer 2 at initial usage. 19. Do not use less than 30 μL RNase-free water to elute RNA. It will result in a poor RNA yield. 20. Give each component a short spin/centrifuge before using. 21. High protein expression is correlated with a strong/thick band and a lower expression of proteins is represented as a thin/ weak band. References 1. Islam F, Gopalan V, Smith RA, Lam AK (2015) Translational potential of cancer stem cells: a review of the detection of cancer stem cells and their roles in cancer recurrence and cancer treatment. Exp Cell Res 335:135–147 2. Frank NY, Schatton T, Frank MH (2010) The therapeutic promise of the cancer stem cell concept. J Clin Invest 120:41–50 3. Zhou BB, Zhang H, Damelin M, Geles KG, Grindley JC, Dirks PB (2009) Tumourinitiating cells: challenges and opportunities for anticancer drug discovery. Nat Rev Drug Discov 8:806–823 4. Islam F, Qiao B, Smith RA, Gopalan V, Lam AK (2015) Cancer stem cell: fundamental experimental pathological concepts and updates. Exp Mol Pathol 98:184–191 5. Major AG, Pitty LP, Farah CS (2013) Cancer stem cell markers in head and neck squamous cell carcinoma. Stem Cells Int 2013:319489 6. Sayed SI, Dwivedi RC, Katna R, Garg A, Pathak KA, Nutting CM, Rhys-Evans P, Harrington KJ, Kazi R (2011) Implications of understanding cancer stem cell (CSC) biology in head and neck squamous cell cancer. Oral Oncol 47:237–243

7. Deonarain MP, Kousparou CA, Epenetos AA (2009) Antibodies targeting cancer stem cells: a new paradigm in immunotherapy? MAbs 1: 12–25 8. Bonnet D, Dick JE (1997) Human acute myeloid leukemia is organized as a hierarchy that originates from a primitive hematopoietic cell. Nat Med 3:730–737 9. Collins AT, Berry PA, Hyde C, Stower MJ, Maitland NJ (2005) Prospective identification of tumourigenic prostate cancer stem cells. Cancer Res 65:10946–10951 10. O’Brien CA, Pollett A, Gallinger S, Dick JE (2007) A human colon cancer cell capable of initiating tumour growth in immunodeficient mice. Nature 445:106–110 11. Dirks PB (2008) Brain tumour stem cells: bringing order to the chaos of brain cancer. J Clin Oncol 26:2916–2924 12. Zhao W, Ji X, Zhang F, Li L, Ma L (2012) Embryonic stem cell markers. Molecules 17: 6196–6236 13. Shah A, Patel S, Pathak J, Swain N, Kumar S (2014) The evolving concepts of cancer stem cells in head and neck squamous cell carcinoma. Sci World J 2014:842491

Surface Markers for the Identification of Cancer Stem Cells 14. Karsten U, Goletz S (2013) What makes cancer stem cell markers different? Springerplus 2:301 15. Hsu CC, Chiang CW, Cheng HC, Chang WT, Chou CY, Tsai HW, Lee CT, Wu ZH, Lee TY, Chao A, Chow NH, Ho CL (2011) Identifying LRRC16B as an oncofetal gene with transforming enhancing capability using a combined bioinformatics and experimental approach. Oncogene 30:654–667 16. Hennen E, Faissner A, Lewis X (2012) A neural stem cell specific glycan? Int J Biochem Cell Biol 44:830–833 17. Monzani E, Facchetti F, Galmozzi E, Corsini E, Benetti A, Cavazzin C, Gritti A, Piccini A, Porro D, Santinami M, Invernici G, Parati E, Alessandri G, LaPorta CA (2007) Melanoma contains CD133 and ABCG2 positive cells with enhanced tumourigenic potential. Eur J Cancer 43:935–946 18. Guo W, Lasky JL, Chang C-J, Mosessian S, Lewis X, Xiao Y, Yeh JE, Chen JY, IruelaArispe ML, Varella-Garcia M, Wu H (2008) Multi-genetic events collaboratively contribute to Pten-null leukemia stem-cell formation. Nature 453:529–533 19. Gu¨nthert U, Hofmann M, Rudy W, Reber S, Zo¨ller M, Haussmann I, Matzku S, Wenzel A, Ponta H, Herrlich P (1991) A new variant of glycoprotein CD44 confers metastatic potential to rat carcinoma cells. Cell 65:13–24 20. Cheung AM, Wan TS, Leung JC, Chan LY, Huang H, Kwong YL, Liang R, Leung AY (2007) Aldehyde dehydrogenase activity in leukemic blasts defines a subgroup of acute myeloid leukemia with adverse prognosis and superior NOD/ SCID engrafting potential. Leukemia 21:1423–1430 21. Ginestier C, Hur MH, Charafe-Jauffret E, Monville F, Dutcher J, Brown M, Jacquemier J, Viens P, Kleer CG, Liu S, Schott A, Hayes D, Birnbaum D, Wicha MS, Dontu G (2007) ALDH1 is a marker of normal and malignant human mammary stem cells and a predictor of poor clinical outcome. Cell Stem Cell 1:555–567 22. Carpentino JE, Hynes MJ, Appelman HD, Zheng T, Steindler DA, Scott EW, Huang EH (2009) Aldehyde dehydrogenase-expressing colon stem cells contribute to tumorigenesis in the transition from colitis to cancer. Cancer Res 69:8208–8215 23. Li R, Wu X, Wei H, Tian S (2013) Characterization of side population cells isolated from the gastric cancer cell line SGC-7901. Oncol Lett 5:877–883

69

24. Lianidou ES, Markou A (2011) Circulating tumor cells in breast cancer: detection systems, molecular characterization, and future challenges. Clin Chem 57:1242–1255 25. Tirino V, Desiderio V, Paino F, De Rosa A, Papaccio F, La Noce M, Laino L, De Francesco F, Papaccio G (2013) Cancer stem cells in solid tumors: an overview and new approaches for their isolation and characterization. FASEB J 27:13–24 26. Milne AN, Carneiro F, O’Morain C, Offerhaus GJ (2009) Nature meets nurture: molecular genetics of gastric cancer. Hum Genet 126: 615–628 27. Podberezin M, Wen J, Chang CC (2013) Cancer stem cells: a review of potential clinical applications. Arch Pathol Lab Med 137: 1111–1116 28. Greve B, Beller C, Cassens U, Sibrowski W, Go¨hde W (2006) The impact of erythrocyte lysing procedures on the recovery of hematopoietic progenitor cells in flow cytometric analysis. Stem Cells 24:793–799 29. Fulawka L, Donizy P, Halon A (2014) Cancer stem cells – the current status of an old concept: literature review and clinical approaches. Biol Res 47:66 30. Murar M, Vaidya A (2015) Cancer stem cell markers: premises and prospects. Biomark Med 9(12):1331–1342. https://doi.org/10.2217/ bmm.15.85. Epub 2015 Nov 19 31. Hadjimichael C, Chanoumidou K, Papadopoulou N, Arampatzi P, Papamatheakis J, Kretsovali A (2015) Common stemness regulators of embryonic and cancer stem cells. World J Stem Cells 7(9): 1150–1184 32. Zhao W, Li Y, Zhang X (2017) Stemnessrelated markers in cancer. Cancer Transl Med 3(3):87–95 33. Walcher L, Kistenmacher AK, Suo H, Kitte R, Dluczek S, Strauß A, Blaudszun AR, Yevsa T, Fricke S, Kossatz-Boehlert U (2020) Cancer stem cells-origins and biomarkers: perspectives for targeted personalized therapies. Front Immunol 11:1280 34. Atashzar MR, Baharlou R, Karami J, Abdollahi H, Rezaei R, Pourramezan F, Zoljalali Moghaddam SH (2020) Cancer stem cells: a review from origin to therapeutic implications. J Cell Physiol 235(2):790–803 35. Qian X, Tan C, Wang F, Yang B, Ge Y, Guan Z, Cai J (2016) Esophageal cancer stem cells and implications for future therapeutics. Onco Targets Ther 9:2247–2254

Chapter 5 CD44-Based Detection of CSCs: CD44 Immunodetection by Flow Cytometry Lornella Seeneevassen, Anissa Zaafour, and Christine Varon Abstract CD44 has been described in many malignancies as a marker of cancer stem cells (CSCs). Several techniques can be used to detect these cells. Here we detail CD44 detection by flow cytometry, a precise technique allowing to determine the percentage of positive cells and the mean fluorescent intensity reflecting the CD44 expression by cells in the samples. The protocol explained here can be used to detect CD44 from cell suspensions prepared from tissues or in vitro cell cultures. Key words Cancer stem cell, CD44, Immunofluorescence, Flow cytometry, Antibody, Cancer

1 Introduction Cancer stem cells are described as a small cell population having stem cell-like properties allowing them to self-renew, give rise to new heterogeneous tumors, resist conventional therapies, and cause metastasis disease and tumor relapse [1]. These cells have been identified till now in a variety of cancers like acute myeloid leukemia [2], breast [3], brain [4], colon [5], and gastric cancer [6], amongst others, and are considered at the origin of bad prognosis disease. The study of CSCs is thus at its peak since understanding these cells and the mechanisms behind their multiple properties will allow the development of therapies against these cells. One crucial aspect of their study is the ability to detect these cells. Several studies have taken advantage of their properties to purify them and characterize markers allowing their identification. Amongst the various markers, most of the time specific of the tumors from which these cells are isolated, CD44 was found to be

Authors Lornella Seeneevassen and Anissa Zaafour have equally contributed to this chapter. Federica Papaccio and Gianpaolo Papaccio (eds.), Cancer Stem Cells: Methods and Protocols, Methods in Molecular Biology, vol. 2777, https://doi.org/10.1007/978-1-0716-3730-2_5, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

71

72

Lornella Seeneevassen et al.

expressed in CSCs from various origins [1, 3, 6–8]. CD44 is usually described as being the receptor of hyaluronic acid [9] and for its various roles from lymphocyte activation, to recirculation, homing, and hematopoiesis [10–12]. Its identification as a functional CSC marker, CD44-negative cells being less tumorigenic than CD44positive cells [6], has given hope to the development of CD44based CSC targeting therapies [13]. In this method review, we describe a protocol allowing the detection of these CD44-positive CSCs in different types of samples, cell suspensions from in vitro cultures, or tissues by flow cytometry. We provide details on samples’ preparation, staining, and analysis.

2

Materials

2.1 Cell Dissociation from Tissue Using gentleMACS™ Dissociator

1. DMEM-F12 Glutamax culture medium. 2. Penicillin/streptomycin (P/S) antibiotics solution (10,000 U/mL, 100X). 3. Fetal bovine serum (FBS). 4. Collagenase type 1A, 1% (Sigma C9891). 5. Hyaluronidase type IV-S, 1% (Sigma, H3884). 6. NH4Cl 2M: 5.349 g in 100 mL of distilled water. 7. KHCO3 1M: 3 g in 30 mL distilled water. 8. EDTA 0.25M: 3.7 g in 40 mL distilled water. 9. Red blood cell (RBC) lysis buffer (for 1 L): NH4Cl 170 mM (85 mL from 2 M stock solution), KHCO3 2 mM (20 mL from 1 M stock solution), EDTA 0.1 mM (0.4 mL from 0.25 M stock solution). Make up to 1 L with distilled water (894.6 mL) and filter through a 0.22 μm filter; store remaining buffer at 4 °C for up to 3 months. 10. 10-cm petri dishes. 11. Scalpels. 12. GentleMACS™ C Tubes (Miltenyi Biotech, 130-096-334). 13. GentleMACS™ Dissociator (Miltenyi Biotech, 130-093-235). 14. 70 μm strainers.

2.2 CD44 Staining for Flow Cytometry

1. CD44-APC (BD Pharmingen, 559942) or CD44-PE (BD, 555479) mouse IgG2b primary antibodies. 2. Isotype antibodies: mouse IgG2b-APC (BD, 555745) or IgG2b-PE (BD, ref 555543). 3. SYTOX™ Blue (Fisher 10297242) DNA dye. 4. EDTA 0.5 M (Sigma, E7889).

CD44-Based Detection of CSCs by Flow Cytometry

73

5. BSA Fraction V 7.5% (Gibco, 15260-037). 6. Cytometry buffer (for 50 mL): EDTA 2 mM, BSA Fraction V 0.5%. Make up to 50 mL with PBS and filter through a 0.22 μm filter; store remaining buffer at 4 °C. 7. Glass FACS tubes. 8. Flow cytometry machine (BD flow cytometer Canto II or equivalent, with BD FACSDiva™ Software).

3

Methods

3.1 Sample Preparation

For CD44 detection in cells from in vitro cell cultures, cells are recovered according to the specific protocols required for the cells (e.g., trypsinization for adherent cells). For CD44 detection from tissues, cell dissociation from tissues should be carried out prior to CD44 staining. One cell dissociation protocol is suggested below. Cell Dissociation from Tissue 1. Prepare DMEM/F12 medium supplemented with 1% P/S (medium). 2. Prepare DMEM/F12 medium supplemented with 1% P/S and 10% FBS (medium-FBS). 3. Wash each tissue separately by adding 5 mL of medium from step 1 to tissue and discarding the medium. Repeat step 3 once. 4. Add 3.5 mL of medium to tissue and transfer to a 10-cm petri dish. 5. Using two scalpels, cut the tissue into small pieces and transfer to a 50 mL gentleMACS™ C tube. 6. Add 1:10 collagenase type 1A and 1:50 hyaluronidase type IV-S. 7. Dissociate using gentleMACS™ depending on the tissue type).

Dissociator

(program

8. Incubate 25 minutes at 37 °C under agitation. Repeat steps 7–8 once. 9. Filter samples using 70 μm strainers in 50 mL tubes. (a) Transfer sample onto a 70 μm strainer, placed on top of a 50 mL tube. (b) Using the piston of a 10 mL syringe, crush pieces of tissue on the strainer. (c) Add 7 mL of medium-FBS to rinse filter and collect maximum of tissue (see Note 1).

74

Lornella Seeneevassen et al.

10. Transfer sample to a 15 mL tube. 11. Centrifuge at 1200 rpm for 3 min to pellet down the cells and discard supernatant (see Note 2). 12. Proceed to RBC lysis (see Note 3). Repeat step if sample is still red. (a) Resuspend pellet in 8 mL of RBC lysis buffer, mix well with pipette, and incubate for 8 min at 4 °C. (b) Centrifuge at 1200 rpm for 3 min to pellet down cells and discard supernatant. 13. Optional (Ficoll gradient to eliminate debris and dead cells) (a) Resuspend sample in 5 mL of medium-FBS and slowly add 5 mL of Ficoll 100%. (b) Centrifuge for 30 min at 1000× g with acceleration and with no brake. (c) Recover the live cells, which will be found at the interface between the two liquids by pipetting. (d) Transfer the cells to a 15 mL tube, and add 5mL of PBS 1X to wash the cells. (e) Centrifuge at 1200 rpm for 3 min to pellet down the cells. (f) Discard supernatant. 3.2 CD44 Staining for Flow Cytometry

Start this part with cell pellet recovered from in vitro cultures or tissue dissociation. 1. Resuspend cell pellet in 1 mL of cytometry buffer. 2. Count the cells and distribute at least 100 000 cells in each FACS tube corresponding to the following conditions (see Note 4): (a) Unstained control cells. (b) Isotype control: cells will be stained with isotype antibodies APC (or isotype antibodies PE). (c) Single-color controls: cells will be stained with antiCD44APC (or anti-CD44PE) antibodies and with SYTOX™ Blue; in case of using several primary antibodies, prepare one tube for each single-color stain of all antibodies used in the experiment (see Note 5). (d) Samples (stained with all antibodies except isotype controls). 3. Add 1 mL of cytometry buffer to each tube. 4. Centrifuge at 1200 rpm for 3 min to pellet down cells and discard supernatant.

CD44-Based Detection of CSCs by Flow Cytometry

75

Table 1 Summary of the antibodies to be used according to the different conditions SYTOX™ Blue (or other viability marker)

Tubes

Conditions

IgG isotype CD44 (e.g., mouse (APC) APC here)

1

Unstained control

–

–

–

2

Isotype control IgG2b-APC antibodies

x

–

–

3

Single-color control CD44-APC antibodies

–

x

–

4

Single-color control SYTOX™ Blue

–

–

x

5

Sample 1 (CD44-APC antibodies + SYTOX™ Blue)

–

x

x

5. Resuspend each sample in corresponding antibody mix each at 1:25 concentration (Table 1) and incubate for 30 min at 4 °C (see Note 6). Keep samples away from light. 6. Add 1 mL of cytometry buffer to each tube. 7. Centrifuge at 1200 rpm for 3 min to pellet down cells and discard supernatant. Repeat steps 6–7 once. 8. Resuspend samples in 1:5000 SYTOX™ Blue and incubate for 10 min on ice before cytometry analysis. SYTOX™ Blue will stain dead cells. Keep samples waiting on ice throughout the analysis process (see Note 7). 3.3 Sample Reading and Analysis

1. Analysis is carried out on the available flow cytometry machine according to the machine-standardized protocols and software. We describe here the general parameters to take into consideration for analysis. The software used here, for example, is BD FACSDiva™ Software. Create a new experiment and new specimen on BD FACSDiva™ Software. 2. Choose appropriate lasers corresponding to the different colors present in the experiment. 3. Carry out compensation. This will allow us to calculate the compensation needed to overrule the possible overlapping of one channel read-out into another (Fig. 1). If more colors (antibodies) are used, single-color controls corresponding to each of the colors should be added. The controls can be prepared using only one sample, which would be representative of all the samples analyzed. (a) Create compensation controls. Several tabs corresponding to the different colors required for compensation will appear.

250

(x 1 000)

200 150

FSC-H

50

50

NCIN87-Unstained

100

250 200

(x 1 000)

SSC-H

C

NCIN87-Unstained

150

250

(x 1 000)

200 150

Cells

50

SSC-A

B

NCIN87-Unstained

100

A

Lornella Seeneevassen et al.

100

76

Singlet2

Singlet1 100

150

SSC-W

SSC-A

(x 1 000)

250

Live cells

200

250

50

100

150

FSC-W

(x 1 000)

200

250

(x 1 000)

NCIN87-Unstained 250

E

NCIN87-Unstained 150 200

(x 1 000)

50

50

SSC-A

250

(x 1 000)

100

D

200

150 200

150

FSC-A

CD44+ cells

100

100

50

50

102

103

104

105

Sytox Pacific Blue-A

102

103

104

105

CD44 APC-A

F

Fig. 1 Working space preparation and gating strategy on the population of interest using unstained cells. Representative experiment performed on NCI-N87 gastric cancer cell line. (a) All cells detected. (b, c) Singletcell selection. (d) Selection of live cells based on the incorporation of SYTOX™ Blue in dead cells. (e) Selection of the gate for CD44-positive cells stained with anti-CD44-APC antibodies. (f) Hierarchical table allowing to visualize the percentages of each cell population

(b) Run unstained sample and set laser powers so as the unstained cells appear completely to the left of each compensation dot plot. Record the unstained sample. (c) Run each single-color control. Each single-color control sample should appear to the right of its corresponding dot plot (positively stained). Record the single-stained samples.

CD44-Based Detection of CSCs by Flow Cytometry

77

(d) Use the calculated compensation module of BD FACSDiva™ Software to calculate compensations and select apply and link to experiment. 4. Prepare working space by plotting the following dot plots: (a) SSC-A (side scatter area) on the y-axis and FSC-A (forward scatter area) on the x-axis to select cells (cells) (Fig. 1a). (b) FSC-W on the x-axis and FSC-H on y-axis on one graph and SSC-W on the x-axis and SSC-H on the y-axis on another graph to select only single cells (Fig. 1b, c). (c) SSC-A on the y-axis and SYTOX™ Blue on the x-axis to select the live cells (Fig. 1d). (d) SSC-A on the y-axis against all the other markers of the sample to identify the cells of interest. Here we choose CD44-APC (Fig. 1e). (e) Population hierarchy table (Fig. 1f). 5. Identify population 1 (cells) correspondingly and set up unstained: (a) Load unstained sample and run. (b) Modulate the laser powers corresponding to the SSC and FSC until the cells of interest appear in the middle of the dot plot. (c) Adjust the gates on the dot plot to place it as close as possible to the unstained cells. This will correspond to the negative unstained populations. 6. Run and record all the samples (on at least 10,000 cells per sample). The stained population should appear inside the gate, to the right of the negative population on the x-axis. These cells should be negative for SYTOX™ Blue to be live and our CSC population here should appear as CD44 positive. 7. Record the percentage of CD44-positive cells and the mean fluorescence intensity (MFI) for each sample (see Notes 8 and 9) (Fig. 2).

4

Notes 1. When filtering sample, recover with a pipette the liquid just underneath the strainer which is concentrated with cells. 2. Discard supernatant by inverting tubes. Keep tubes inverted and press against a paper towel to discard as much supernatant as possible. Once reverted, be careful not to invert the tubes again. The pellet detaches as soon as the tube is reverted and might be lost.

Lornella Seeneevassen et al.

250 200

(x 1 000)

SSC-A

NCIN87-Sample 1

150

250

Live cells

100

(x 1 000)

200

Live cells

50

50

SSC-A

B

NCIN87-Unstained

150

A

100

78

102

103

104

105

102

Sytox Pacific Blue-A

104

105

250

(x 1 000)

200 150

SSC-A

NCIN87-Sample 1

CD44+ cells

50

100

250

(x 1 000)

200 150

CD44+ cells

50

SSC-A

D

NCIN87-Unstained

100

C

103

Sytox Pacific Blue-A

102

103

104

CD44 APC-A

105

102

103

104

105

CD44 APC-A

E

Fig. 2 CD44-based detection of CSCs by flow cytometry in NCI-N87 gastric cancer cell line. Gating strategy on unstained cells (a) allowing the visualization of SYTOX Blue-negative live cells in the tested sample (b). Gating strategy on unstained cells (c) allowing the visualization of CD44-positive cells in the tested sample 1 (d). (e) Detection of 77.8% of CD44-positive cells with an MFI of 868 in live cells of sample 1

CD44-Based Detection of CSCs by Flow Cytometry

79

3. Red blood cell lysis is an important step during dissociation when doing flow cytometry to eliminate red blood cells, which are autofluorescent. 4. More cells can be added to the unstained control tube, which is also used to set up the specific parameters during analysis. 5. Beads can also be used in place of samples for single-color controls. This is done especially when the specific population recognized by the antibody is too small to be used for compensation. Beads are then stained with the same antibody as the samples and negative unstained beads are added to the tubes to have a mix of stained and unstained beads. 6. Discard as much supernatant as possible. Consider that approximately 10 μL of buffer remains in the tube after washing. Staining can be done in 100 μL total volume by adding 90 μL of antibody solution (90 μL antibody solution + 10 μL residual volume). The total amount of antibody can be increased depending on the number of cells stained. 7. No washing step is required after SYTOX™ Blue staining. Note that other viability stains can be used, such as 7-amino actinomycin D (7-AAD) [1, 14–16] or 4′,6-diamidino-2-phenylindol (DAPI) [17]. 8. The MFI of positive cells corresponds to the expression level of a marker (here CD44). It is particularly useful for monitoring increase or decrease in markers’ expression between samples (upon some treatments, for instance). 9. An important point to keep in mind about CD44-based CSC detection is the fact that CD44 is not solely expressed in CSCs but also in a variety of different cells, related to its numerous physiological roles [10–12]. To cater for that, when analyzing samples derived from tissue, the use of EPCAM or any other epithelial cell surface marker is essential to distinguish epithelial cells from immune cells (expressing CD45) also expressing CD44 [1]. In addition, amongst the other markers that can be used to co-detect CD44 and ensure that CSCs are being analyzed is ALDH which is aldehyde dehydrogenase activity, found to be also elevated in CSCs [1]. Other cancer-specific CSC markers like CD166, CD24, and CD133 can be used in parallel with CD44 for the same purpose [1, 18]. Moreover, variants of CD44 like CD44v3, CD44v6, CD44v8-10, and CD44v9 have been described as being more precise of the subpopulation of CSC studied and can allow to refine analysis and CSC detection [19–22].

80

Lornella Seeneevassen et al.

References 1. Nguyen PH, Giraud J, Chambonnier L, Dubus P, Wittkop L, Belleanne´e G et al (2017) Characterization of biomarkers of tumorigenic and chemoresistant cancer stem cells in human gastric carcinoma. Clin Cancer Res 23:1586–1597. https://doi.org/10. 1158/1078-0432.CCR-15-2157 2. Bonnet D, Dick JE (1997) Human acute myeloid leukemia is organized as a hierarchy that originates from a primitive hematopoietic cell. Nat Med 3:730–737. https://doi.org/10. 1038/nm0797-730 3. Al-Hajj M, Wicha MS, Benito-Hernandez A, Morrison SJ, Clarke MF (2003) Prospective identification of tumorigenic breast cancer cells. Proc Natl Acad Sci U S A 100:3983– 3988. https://doi.org/10.1073/pnas. 0530291100 4. Singh SK, Clarke ID, Terasaki M, Bonn VE, Hawkins C, Squire J et al (2003) Identification of a cancer stem cell in human brain tumors. Cancer Res 63:5821–5828 5. Ricci-Vitiani L, Lombardi DG, Pilozzi E, Biffoni M, Todaro M, Peschle C et al (2007) Identification and expansion of human coloncancer-initiating cells. Nature 445:111–115. https://doi.org/10.1038/nature05384 6. Takaishi S, Okumura T, Tu S, Wang SSW, Shibata W, Vigneshwaran R et al (2009) Identification of gastric cancer stem cells using the cell surface marker CD44. Stem Cells Dayt Ohio 27:1006–1020. https://doi.org/10. 1002/stem.30 7. Lee CJ, Dosch J, Simeone DM (2008) Pancreatic cancer stem cells. J Clin Oncol 26:2806– 2812. https://doi.org/10.1200/JCO.2008. 16.6702 8. Meng E, Long B, Sullivan P, McClellan S, Finan MA, Reed E et al (2012) CD44+/ CD24- ovarian cancer cells demonstrate cancer stem cell properties and correlate to survival. Clin Exp Metastasis 29:939–948. https://doi.org/10.1007/s10585-0129482-4 9. Aruffo A, Stamenkovic I, Melnick M, Underhill CB, Seed B (1990) CD44 is the principal cell surface receptor for hyaluronate. Cell 61: 1303–1313. https://doi.org/10.1016/00928674(90)90694-a 10. Crosby HA, Lalor PF, Ross E, Newsome PN, Adams DH (2009) Adhesion of human haematopoietic (CD34+) stem cells to human liver compartments is integrin and CD44 dependent and modulated by CXCR3 and CXCR4.

J Hepatol 51:734–749. https://doi.org/10. 1016/j.jhep.2009.06.021 11. Funaro A, Spagnoli GC, Momo M, Knapp W, Malavasi F (1994) Stimulation of T cells via CD44 requires leukocyte-function-associated antigen interactions and interleukin-2production. Hum Immunol 40:267–278. https://doi.org/10.1016/0198-8859(94) 90026-4 12. Bourguignon LYW, Singleton PA, Diedrich F, Stern R, Gilad E (2004) CD44 interaction with Na+-H+ exchanger (NHE1) creates acidic microenvironments leading to hyaluronidase2 and cathepsin B activation and breast tumor cell invasion. J Biol Chem 279:26991–27007. https://doi.org/10.1074/jbc.M311838200 13. Yan Y, Zuo X, Wei D (2015) Concise review: emerging role of CD44 in cancer stem cells: a promising biomarker and therapeutic target. Stem Cells Transl Med 4:1033–1043. https://doi.org/10.5966/sctm.2015-0048 14. Nguyen PH, Giraud J, Staedel C, Chambonnier L, Dubus P, Chevret E et al (2016) All-trans retinoic acid targets gastric cancer stem cells and inhibits patient-derived gastric carcinoma tumor growth. Oncogene 35:5619–5628. https://doi.org/10.1038/ onc.2016.87 15. Seeneevassen L, Giraud J, Molina-Castro S, Sifre´ E, Tiffon C, Beauvoit C et al (2020) Leukaemia Inhibitory Factor (LIF) inhibits cancer stem cells tumorigenic properties through hippo kinases activation in gastric cancer. Cancers 12:2011. https://doi.org/10. 3390/cancers12082011 16. Giraud J, Molina-Castro S, Seeneevassen L, Sifre´ E, Izotte J, Tiffon C et al (2020) Verteporfin targeting YAP1/TAZ-TEAD transcriptional activity inhibits the tumorigenic properties of gastric cancer stem cells. Int J Cancer 146:2255–2267. https://doi.org/10. 1002/ijc.32667 ˜ a Amador LA, 17. Besse`de E, Staedel C, Acun Nguyen PH, Chambonnier L, Hatakeyama M et al (2014) Helicobacter pylori generates cells with cancer stem cell properties via epithelialmesenchymal transition-like changes. Oncogene 33:4123–4131. https://doi.org/10. 1038/onc.2013.380 18. Todaro M, Francipane MG, Medema JP, Stassi G (2010) Colon cancer stem cells: promise of targeted therapy. Gastroenterology 138:2151– 2162. https://doi.org/10.1053/j.gastro. 2009.12.063

CD44-Based Detection of CSCs by Flow Cytometry 19. Giraud J, Seeneevassen L, Rousseau B, Bouriez D, Sifre´ E, Giese A et al (2023) CD44v3 is a marker of invasive cancer stem cells driving metastasis in gastric carcinoma. Gastric Cancer 26:234–249. https://doi.org/ 10.1007/s10120-022-01357-y 20. Todaro M, Gaggianesi M, Catalano V, Benfante A, Iovino F, Biffoni M et al (2014) CD44v6 is a marker of constitutive and reprogrammed cancer stem cells driving colon cancer metastasis. Cell Stem Cell 14:342–356. https://doi.org/10.1016/j.stem.2014. 01.009

81

21. Tsugawa H, Kato C, Mori H, Matsuzaki J, Kameyama K, Saya H et al (2019) Cancer stem-cell marker CD44v9-positive cells arise from Helicobacter pylori-infected CAPZA1overexpressing cells. Cell Mol Gastroenterol Hepatol 8:319–334. https://doi.org/10. 1016/j.jcmgh.2019.05.008 22. Lau WM, Teng E, Chong HS, Lopez KAP, Tay AYL, Salto-Tellez M et al (2014) CD44v8-10 is a cancer-specific marker for gastric cancer stem cells. Cancer Res 74:2630–2641. https://doi.org/10.1158/0008-5472.CAN13-2309

Chapter 6 ALDH Activity Assay: A Method for Cancer Stem Cell (CSC) Identification and Isolation Vitale Del Vecchio, Marcella La Noce, and Virginia Tirino Abstract Cancer stem cells (CSCs) are a small tumor cell subpopulation, driving cancer initiation, progression, multidrug resistance, and metastasis. Several methods are used to detect and isolate CSCs by flow cytometry. Among these, measurement of aldehyde dehydrogenase (ALDH) activity within the cell is an assay widely used to identify and isolate CSCs from different types of solid tumors. The aldehyde dehydrogenase (ALDH) is a polymorphic enzyme responsible for the oxidation of aldehydes to carboxylic acids, overexpressed both in normal and cancer stem cells. In this chapter, it is described how CSCs are detected and isolated by using ALDH activity assay. Key words Aldehyde dehydrogenase (ALDH), Cancer, Cancer stem cells (CSCs), Aldefluor, Cell sorting, Flow cytometry

1

Introduction Aldehyde dehydrogenases (ALDHs) are a family of enzymes that catalyze the oxidation of endogenous aldehyde substrates to their corresponding carboxylic acids and conversion of retinol to retinoic acid. The human genome contains 19 ALDH genes that code for enzymes involved in detoxification from aldehydes produced by physiological metabolic processes or environmental agents and cytotoxic drugs [1]. For this reason, an increase in ALDH activity may confer cells more resistance to chemotherapeutic treatments [2]. ALDH1 has three isoforms (A1, A2, A3) and is considered a marker for normal and cancer stem cells (CSCs). Indeed, this subpopulation of stem-like tumor cells is responsible for tumor initiation, invasive growth, and metastasis formation and has high levels and high activity of this enzyme. Hilton [3] first showed the role of high ALDH activity in chemotherapeutic resistance to

Federica Papaccio and Gianpaolo Papaccio (eds.), Cancer Stem Cells: Methods and Protocols, Methods in Molecular Biology, vol. 2777, https://doi.org/10.1007/978-1-0716-3730-2_6, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

83

84

Vitale Del Vecchio et al.

cyclophosphamide, an alkylating agent, in leukemia stem cells. In addition to drug resistance, ALDH maintains low levels of reactive oxygen species (ROS) by preventing apoptosis of CSCs [4]. High ALDH activity has been demonstrated in CSCs of liver, lung, breast, colon, and head and neck cancers [5–9]. Generally, the ALDH1A1 isoform is commonly considered to be responsible for increasing ALDH activity in CSCs, although recent studies have shown that other isoforms contribute to increased activity, in particular the ALDH1A3 isoform [10]. However, the expression and activity of this enzyme can be considered as a reliable marker in tissues that normally do not express or express low levels of ALDH1A1 (breast, lung), but cannot be considered a tissue marker that already expresses high levels of ALDH1A1 (liver, pancreas) [1]. The method used to identify and select high ALDH activity cells is the ALDEFLUOR assay kit (StemCell Technologies, Durham, NC, USA). This assay is based on the principle that ALDH enzyme converts the ALDH substrate, BODIPY-aminoacetaldehyde (BAAA), into BODIPY-aminoacetate (BAA) which is retained inside the cells. The amount of BAA fluorescence in cells is proportional to ALDH activity and can be measured using a flow cytometer. Viable ALDH-bright (ALDHbr) cells can be isolated using a cell sorter. A specific inhibitor of ALDH, diethylaminobenzaldehyde (DEAB), is used as background fluorescence control. High ALDH activity in the ALDEFLUOR assay has been attributed to the expression of ALDH1, but it is not possible to understand which of the three isoforms of this enzyme (A1, A2, A3) is due. Here, we provide a staining protocol based on the evaluation of ALDH activity using flow cytometry.

2 2.1

Materials Reagents

1. ALDEFLUOR™ Kit by STEMCELL Technologies. 2. 12
75 mm tubes compatible with the cytometer using a low-speed centrifuge (capable of 250
g). 3. 37  C heating device (water bath or heat block). 4. Flow cytometer equipped with a 488-nm blue laser for excitation and an optical filter set to detected 530-nm (green FITC) fluorescence. 5. Refrigerator (2–8  C) or ice. 6. Erythrocyte lysing agent (without detergent or fixatives).

CSC Detection Using ALDH Activity Assay

2.2 Sample Preparation

85

1. Prepare fresh or frozen test cells according to standard procedures for the cell type. 2. If using samples containing blood and the erythrocyte to leukocyte ratio (RBC/WBC) of the specimen is >2:1, lyse the erythrocytes with an ammonium chloride-based buffered solution that does not contain detergents or fixatives to preserve the cell viability. 3. If using sphere-forming cells to separate them and obtain a single-cell solution, centrifuge the sample for 5 min at 250
g, remove the supernatant and suspend cells in 1
trypsin – EDTA and incubate at 37  C for 10 min, and then neutralize the trypsin with medium supplemented with serum. 4. Perform a cell count and determine the viable cell concentrations using trypan blue exclusion cell count.

3 3.1

Methods Aldefluor Assay

1. Obtain a single viable cell suspension. 2. Spin cells at 250
g for 5 min. 3. Discard media and resuspend cells at 1
106 cells/mL in ALDEFLUOR assay buffer (see Note 1). 4. Label one 12
75 mm Falcon tube for each of the following points: tube 1, with 1 μg/ml propidium iodide (PI) or 7-actinoaminomycin-D (7-AAD); tube 2, DEAB CONTROL; and tube 3, ALDH TEST. If a cell sort is required, label an additional ALDH TEST SORT tube for every 500,000 cells (500 μl) of sample to be sorted (see Note 2). 5. Transfer 500 μl (500,000 cells) into each Falcon tube. Due to the nature of the bioassay, no tube may contain more than 500 μl of volume (500,000 cells), as it is a very concentration-specific assay. 6. Add 5 μl of DEAB (1,5 mM stock solution) into the tube labeled DEAB CONTROL and mix well. The final DEAB concentration should be 15 μM (see Note 3). 7. Add 2.5 μl of ALDEFLUOR substrate into the DEAB CONTROL AND ALDH TEST tubes. Mix all tubes well (see Note 4). 8. Place all tubes (including PI or 7-ADD only tube) into a water bath preset to 37  C, making sure that there is a lid as the reaction is light-sensitive. 9. Incubate the tubes at 37  C for 30–60 min (see Note 5). Do not exceed 60 min (see Note 6).

86

Vitale Del Vecchio et al.

10. After 30–60 min of incubation, the tubes can be removed from the water bath and brought back to the flow hood. If there is no sorting required, and there are only three tubes (PI or 7-AAD only, DEAB CONTROL, ALDH TEST), proceed to step 12 (see Note 7). 11. If the cells are going to be sorted, pipette all the volume in the tubes labeled ALDH into a single tube labeled ALDH TEST SORT and centrifuge for 5 min at 250
g. Resuspend the sample in ALDEFLUOR assay buffer to a concentration of 2,000,000 cells/mL (this is half the volume used during the incubation) and proceed to step 12. 12. Add 2 μl of 1 μg/ml dye probe to the PI or 7-AAD tube, DEAB CONTROL tube, and the ALDH TEST tube if for analysis only. If the ALDH TEST SORT tube is to be sorted and has more than 500,000 cells in it, add 20 μl per ml of volume in the ALDH TEST SORT tube (see Note 8). 3.2 Flow Cytometer Setup and Data Acquisition

1. Create a forward scatter (FCS) vs. side scatter (SSC) dot plot, to select in a gate (P1) nucleated cell population based on scatter excluding RBCs and debris.

3.2.1 Prepare an Acquisition Template

2. Create a fluorescence channel 1 (FL1) vs. SSC dot plot, gated on P1 (see Note 9).

3.2.2 To Set up Analyzer and Acquire Data

1. Load the DEAB CONTROL tube on the cytometer. 2. Adjust the FL1 photomultiplier voltage so that the stained population is placed at second log decade and gated on P1. 3. Remove the tube. 4. Load the ALDH TEST tube on the cytometer. Create a gate (P2) to encompass the ALDHbright cell population using the same instrument settings (see Note 10) (Fig. 1). ALDH 250

(x 1 000)

200 150

SSC-A

6,9%

P2

50

100

250 200 150

P2

50

SSC-A

1,6%

100

(x 1 000)

DEAB

102

103

104

ALDH FITC-A

105

102

103

104

105

ALDH FITC-A

Fig. 1 A representative plot for chondrosarcoma cell line. The gate (P2) is set on the DEAB sample to include ALDH-positive cells (5.3%)

CSC Detection Using ALDH Activity Assay

87

5. ALDHbright cells derived from fresh biopsies or heterogeneous cell populations can have different SSC characteristics and the gate must be set correctly. For example, if epithelial cells are used, it is a good practice to stain the cells with an epithelial marker and gate on epithelial cells. 6. To exclude dead cells, create an FSC vs. viability stain dot plot gated on P1, and gate on cells within the first log decade on the viability stain axis. 7. Sometimes, depending on the DNA probe used, it is necessary to perform a compensation setting. PI dye emits in PE fluorescence and ALDH in FITC fluorescence. PE emission is largely detected in the detector specific for PE, but the emission tail lies within the range of the band-pass filter used for detection of FITC. This will be seen as “false-positive” signals in the FITC channel and fluorescence compensation is needed to correct this overlap. 8. To perform a compensation, run a sample stained only with a PE-labeled dye such as PI. Observe the signal in both PE and FITC channels. 9. Adjust the compensation settings until no PE signal is seen in the FITC channel: (FITC + PE overlap)  (PE) ¼ accurate FITC results. 10. Collect almost 100,000 events in P1 for each sample, without changing the instrument settings. 3.2.3

Cell Sorting

1. The sorting gates are established using the cells stained with PI or 7-ADD only as negative controls. 2. Before performing cell sorting, it is important to exclude doublets. Doublet exclusion is to ensure the count of single cells and the exclusion of doublets from analysis. If a doublet containing a fluorescence-positive and fluorescence-negative cell passes through the laser, it will produce a positive pulse leading to false positives in both analysis and sorting experiments. 3. Doublet exclusion is performed by plotting the height or width against the area for forward scatter or side scatter. Doublets will have double the area and width values of single cells whilst the height is roughly the same. Therefore, disproportions between height, width, and area can be used to identify doublets. 4. ALDHbright and ALDH fractions are sorted. 5. Aliquots of ALDHbright and ALDH sorted cells are evaluated for purity by flow cytometry. The purity must be almost 80%. 6. ALDHbright and ALDH sorted cell populations are cultured in standard medium, used for in vivo and in vitro experiments, and analyzed for stemness markers and sphere formation assay (see Note 11).

88

4

Vitale Del Vecchio et al.

Notes 1. Optimal cell concentration may vary among different cell types. Therefore, it is necessary to identify the cell concentration that gives the strongest fluorescence intensity of ALDHbright cells and the highest signal to background ratio and, for heterogeneous cell samples, the best distinction between ALDHbright and ALDHlow cells. Suggested concentrations of cells per mL of ALDEFLUOR™ samples can be as follows: 1
105, 2
105, 5
105, 1
106, and 2
106. Moreover, it is good practice to use cells that have a high ALDH activity as positive control. Usually, the A549 cell line can be used as a positive control cell line for ALDH activity. 2. If the samples contain fewer than 90% viable cells, it is recommended to stain the cells with a DNA dye, such as propidium or 7-actinoaminomycin-D, to stain dead cells. 3. DEAB is diluted in 95% ethanol. Therefore, recap immediately to prevent evaporation. 4. The ALDH reaction starts immediately after ALDEFLUOR substrate addition to cell suspension. 5. The cells will settle to the bottom of the tube and each tube must be mixed every 15 min. 6. Optimal incubation times may vary among different cell types. Therefore, it is necessary to test different times. Suggested incubation times can be 15 min, 30 min, 45 min, and 60 min. 7. It is imperative that samples are kept on ice once the ALDEFLUOR substrate incubation is finished in order to prevent fluorescent product efflux. For best results, samples should be analyzed by flow cytometry shortly after staining. 8. To perform double staining with a cell surface marker (for example, ALDHbright/CD44+), after step 13, incubate samples with antibody for 15–30 min, at 2–8  C. Centrifuge test control tube with antibody, DEAB CONTROL, and ALDH TEST at 250
g for 5 min, and then remove the supernatant for each tube. Resuspend each cell pellet in 0.5 mL ALDEFLUOR Assay Buffer. 9. FL1 is assumed to correspond with green fluorescence signal. 10. Cells with high ALDH activity are identified in comparison with DEAB-negative control sample. 11. Sorted cells are cultured in media supplemented with an excess of gentamicin to avoid contamination rendering the sorting semi-sterile.

CSC Detection Using ALDH Activity Assay

89

References 1. Tomita H, Tanaka K, Tanaka T et al (2016) Aldehyde dehydrogenase 1A1 in stem cells and cancer. Oncotarget 7:11018–11032 2. Tirino V, Desiderio V, Paino F et al (2013) Cancer stem cells in solid tumors: an overview and new approaches for their isolation and characterization. FASEB J 27:13–24 3. Hilton J (1984) Role of aldehyde dehydrogenase in cyclophosphamide-resistant L1210 leukemia. Cancer Res 44:5156–5160 4. Xu X, Chai S, Wang P et al (2015) Aldehyde dehydrogenases and cancer stem cells. Cancer Lett 369:50–57 5. Ma S, Chan KW, Lee TKW et al (2008) Aldehyde dehydrogenase discriminates the CD133 liver cancer stem cell populations. Cancer Res 6:1146–1153 6. Jiang F, Qiu Q, Khanna A et al (2009) Aldehyde dehydrogenase 1 is a tumor stem cell-

associated marker in lung cancer. Mol Cancer Res 7:330–338 7. Marcato P, Dean CA, Pan D et al (2011) Aldehyde dehydrogenase activity of breast cancer stem cells is primarily due to isoform ALDH1A3 and its expression is predictive of metastasis. Stem Cells 29:32–45 8. Giraud J, Failla LM, Pascussi JM et al (2016) Autocrine secretion of progastrin promotes the survival and self-renewal of colon cancer stem– like cells. Cancer Res 76:3618–3628 9. Desiderio V, Papagerakis P, Tirino V et al (2015) Increased fucosylation has a pivotal role in invasive and metastatic properties of head and neck cancer stem cells. Oncotarget 6:71–84 10. Marcato P, Dean CA, Giacomantonio CA et al (2011) Aldehyde dehydrogenase: its role as a cancer stem cell marker comes down to the specific isoform. Cell Cycle 10:1378–1384

Chapter 7 In Vitro Tumorigenic Assay: A Tumor Sphere Assay for Cancer Stem Cells Amani Yehya, Hisham Bahmad, and Wassim Abou-Kheir Abstract Cancer stem cells (CSCs) represent a subpopulation of tumor cells that are thought to be responsible for therapy resistance, recurrence, and metastasis through their capacity to self-renew and differentiate into heterogeneous downstream lineages of cancer cells. Understanding the features of CSCs is crucial for managing cancer disease and establishing potential targeted therapeutics. Tumor sphere formation assay is a widely used in vitro method that selects and enriches the CSC subpopulation from the total population of cancer cells, based on their inherent ability to grow and clonally expand in serum-free and nonadherent culture conditions. Here we provide a detailed methodology to generate and propagate spheres from isolated cell suspensions of tumor tissues and cell lines using a semisolid MatrigelTM-based threedimensional (3D) culture system. Key words Cancer stem cells, Self-renewal, Differentiation, Tumor sphere assay, MatrigelTM

1

Introduction Cancer stem cells (CSCs), also known as tumor-initiating cells (TICs), represent a subpopulation of tumor cells that replicate the unique characteristics of noncancerous stem/progenitor cells, including the self-renewal capability and multilineage differentiation capacity [1, 2]. Having such distinctive “stemness” features, CSCs are able to preserve the diversity and heterogeneity of the cancer cells that constitute the tumor bulk [3]. Accumulating evidence suggests that CSCs instigate a pivotal role in sustaining cancer survival via harboring cancer hallmarks that contribute to cancer growth and progression, treatment resistance, recurrence, and metastasis [4–6]. Isolating and enrichment of this subpopulation of CSCs in vitro enables us to study their biological properties

Federica Papaccio and Gianpaolo Papaccio (eds.), Cancer Stem Cells: Methods and Protocols, Methods in Molecular Biology, vol. 2777, https://doi.org/10.1007/978-1-0716-3730-2_7, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

91

92

Amani Yehya et al.

and understand their role in carcinogenesis. Eventually, developing potential targeted therapies against CSCs warrants us to overcome possible tumor relapse and chemoresistance through more effective therapeutic strategies [7]. There are different approaches to isolate and characterize CSCs based on distinguishable properties and biomarkers [8, 9]. One property includes their ability to form spheres in nonadherent in vitro culture conditions. The tumor sphere-forming assay has been developed as an in vitro method to isolate CSCs and study their properties, replacing the more time-consuming and complicated in vivo tumorigenicity assays [10]. This assay shows that CSCs can be designated, under certain experimental conditions, to survive and clonally expand forming tumor spheres. The concept relies on the CSCs’ remarkable ability to grow in serum-free and nonadherent conditions, whereas the anchorage-dependent non-CSCs undergo programmed cell death. The ability of a single cell to generate a spheroid is considered indicative of self-renewal ability and is hence consistent with a CSC phenotype [10–12]. The sphere formation assay provides a useful tool to assess and characterize the CSCs’ population and screen for drugs specifically targeting CSCs [13–17]. Here, we report the detailed approach for generating and propagating tumor spheres from tumor tissues and cell lines’ isolated cell suspensions, using a semisolid MatrigelTM-based 3D culturing system. This approach allows for the efficient enrichment of CSCs and overcomes the limitations encountered with other culturing methods.

2 2.1

Materials Reagents

1. Rho-associated coiled coil containing protein kinase (ROCK) inhibitor Y-27632. 2. Dulbecco’s PBS, without calcium and magnesium (D-PBS). 3. 0.05% trypsin/EDTA. 4. TrypLE™. 5. Dispase. 6. ACK lysing buffer. 7. 0.4% trypan blue solution. 8. Growth factor-reduced MatrigelTM. 9. Culturing media: appropriate growth media (see Note 1) supplemented with 5 μg/mL plasmocin® prophylactic, 1X penicillin–streptomycin, 1X nonessential amino acid solution, and 1 mM sodium pyruvate.

In Vitro Tumorigenic Assay: A Tumor Sphere Assay for Cancer Stem Cells

93

10. Complete media: culturing media supplemented with 10% fetal bovine serum (FBS). 11. Dissociation media: culturing media with 5 mg/mL collagenase type II (see Note 2) and 10 μM ROCK inhibitor Y-27632. 2.2

Equipment

1. 15 mL conical tubes. 2. 50 mL conical tubes. 3. Microtubes (1.5 mL). 4. Tissue culture dish (100 mm). 5. Tissue culture plate (12 well). 6. Scalpel and forceps. 7. Syringes and needles 8. 70 μm and 40 μm cell strainers sized to fit 50 mL conical tubes. 9. Bright-LineTM Hemacytometer. 10. Micropipettes. 11. Pipette aid. 12. Ice container. 13. Biosafety cabinet. 14. Cell culture humidified incubator set at 37
C, supplied with 5% CO2. 15. Water bath set at 37
C. 16. Centrifuge machine. 17. Inverted microscope. 18. Vortex mixer.

3

Methods All the following procedures should be performed in a primary cell culture laboratory under a biosafety cabinet using sterile reagents and equipment to minimize the risk of culture contaminations. For the sphere formation assay, cell suspensions can be generated from tumor tissues and cell lines.

3.1 Isolation of Cell Suspensions from Tumor Tissues

1. Tumor tissues can be obtained surgically from patient specimens or xenograft murine tumors (see Note 3). 2. Transfer the tumor tissue to a 100 mm tissue culture dish while keeping it submerged in growth media, and then mince the tissue mechanically using a sterile scalpel blade and forceps into very small pieces of around 0.3–0.5 mm in size.

94

Amani Yehya et al.

3. For enzymatic digestion, transfer the minced tissue into a 15 mL conical tube containing 3 mL of the dissociation media filtered with a 0.22 μm filter unit. Incubate the mixture in a humidified incubator at 37
C for a minimum of 1 h with intermittent vortexing (see Note 4). 4. Spin down the minced tissue at 200 g for 5 min at 4
C. Aspirate supernatant and add 2 mL of TrypLE™ solution with 10 μM ROCK inhibitor Y-27632. Resuspend the pellet and place it in the incubator at 37
C for a minimum of 5 min (see Note 5). 5. Neutralize the TrypLE™ solution by adding 2 mL of complete media and then spin down at 200 g for 5 min at 4
C. Aspirate the supernatant and then resuspend the pellet in 3 mL of culturing media. 6. Optional: To further dissociate the tissue lysate, pass it gently through a series of needles with different gauges. You start by using 19 gauge, then 21 gauge, then 23 gauge, and then 25 gauge. Make sure cells are properly dissociated and of appropriate morphology (see Note 6). 7. Optional: If using samples containing blood, lyse the red blood cells using the ACK lysis buffer (see Note 7). 8. Strain with a 70 μm cell strainer into a 50 mL conical tube to remove tissue debris and obtain single-cell suspensions. 9. Strain flow-through again with a 40 μm cell strainer into a new 50 mL conical tube. 10. Recover cells by centrifugation at 200 g for 5 min at 4
C and then aspirate the supernatant. 11. Resuspend the cells in 1 mL of serum-free culturing media (see Note 8). Quantify the number of viable cells on a hemocytometer using the trypan blue dye (see Note 9). 3.2 Preparation of Cells from Prostate Cancer Cell Lines

1. Aspirate the growth media from a 100-mm dish of adherent monolayer tumor cells. 2. Wash once with pre-warmed D-PBS. 3. Add 1 mL of pre-warmed 0.05% trypsin–EDTA solution. Incubate at 37
C for at least 3 min. 4. Observe the cells under the microscope for detachment. When 80–90% of cells are detached, neutralize the trypsin with 1 mL of complete media (see Note 10). 5. Transfer into a 15 mL conical tube and spin down the cells at 200 g for 5 min at room temperature. 6. Resuspend the cells in 1 mL of serum-free culturing media (see Note 8). Quantify the number of viable cells on a hemocytometer using the trypan blue dye (see Note 9).

In Vitro Tumorigenic Assay: A Tumor Sphere Assay for Cancer Stem Cells

95

Fig. 1 Schematic illustrating the strategy of the sphere formation assay. After isolating single-cell suspension from tumor cell lines or murine/patient tissues, a mix of cold cell suspension and cold MatrigelTM (ratio 1:1) is seeded around the bottom rim of the well. Media with or without drugs is added to every generation in the sphere formation assay. Spheres are serially propagated to a minimum of five generations of spheres (G5) to verify the extensive self-renewal ability of cancer stem cells (CSCs). The sphere formation efficiency has to be calculated for each generation to assess the self-renewal ability of sphere-forming cells 3.3 Seeding Cells for Sphere Formation Assay

1. To generate spheres from primary tumor cells, plate 2000–10,000 cells/well in a 12-well plate. To generate spheres from tumor cell lines, plate 1000–5000 cells/well in a 12-well plate (see Note 10). 2. Prepare a total of 100 μL mix (50 μL cold cell suspension/ 50 μL cold MatrigelTM) for each well of a 12-well plate. Use a 1:1 ratio of cell suspension in serum-free culturing media to MatrigelTM (see Note 11) (see Fig. 1). 3. Slowly pipet 100 μL around the bottom rim of a well in a 12-well plate uniformly by going around in a circular manner (see Note 12). 4. Allow the MatrigelTM to solidify in a humidified incubator at 37
C for 45–60 min. 5. After solidification, slowly add 1 mL of pre-warmed serum-free culturing media or low-serum (1–5% FBS) culturing media, with or without a specific drug, in the middle of each well (see Note 13). 6. Replenish with warm media as in the original plating (with or without a specific drug) every 2–3 days (see Note 14). 7. Spheres will generally start to appear by day three or four. Each cell type exhibits a distinct pattern in the process of sphere formation and growth, which must be monitored to optimize the follow-up schedule (see Note 15).

96

Amani Yehya et al.

8. Spheres can be counted starting day 8–10 and the sphere formation efficiency (SFE) or sphere formation unit (SFU) can be calculated based on this simple formula: SFE ¼ (number of spheres counted  number of seeded cells)  100 (see Note 16). 3.4 Propagating Spheres

1. Aspirate the medium from the center of the well and add 1 mg/ mL of dispase solution dissolved in culturing media. Incubate for a minimum of 30 min in a humidified incubator at 37
C. Dispase will break down the MatrigelTM and will release the spheres into the media (see Fig. 1). 2. Collect the media with the released spheres and centrifuge at 200 g for 5 min at 4
C. 3. Resuspend the pellet in 1 mL of 0.05% trypsin–EDTA and incubate in a humidified incubator at 37
C for 10 min. 4. Neutralize trypsin by adding 1 mL of complete media and transfer into a 15 mL conical tube. 5. Centrifuge the cells at 200 g for 5 min at room temperature, aspirate the supernatant, and then resuspend the pellet in 1 mL of culturing media. 6. Start the mechanical dissociation process by passing them through a series of needles with different gauges, starting with the 21 gauge (see Note 6). 7. Centrifuge cells at 200 g for 5 min at room temperature, resuspend in fresh serum-free culturing media, and then count the cells using a hemocytometer and trypan blue. 8. Seed the cells as described in Subheading 3.3. 9. Repeat the propagation steps to get a minimum of five generations of spheres (G5) to verify the extensive self-renewal ability of CSCs (see Note 17).

4

Notes 1. Growth media vary by cell type as different cell types have distinct nutrient and supplement requirements. 2. Collagenase II can be substituted by other collagenase types including IV, V, and XI as they display equivalent activity on different tumor and tissue types [18]. 3. Patient tumors should be collected under Institutional Review Board (IRB)-approved protocols. Murine tumor should be carried out in accordance with the recommendations of the NIH Guide for Use and Care of Animals. Tumor samples can be temporarily stored in the appropriate growth media at 4
C but should be processed within 24 h of surgical resection for maximal cell viability.

In Vitro Tumorigenic Assay: A Tumor Sphere Assay for Cancer Stem Cells

97

4. The incubation time in the dissociation media varies (1–24 h) depending on the type of tissue being dissociated and its sensitivity to the collagenase enzyme. 5. The incubation in TrypLE™ varies (5–15 min) depending on the type of tissue being dissociated and its sensitivity to the TrypLE™ enzyme. 6. At this stage, you can take a 10 μL sample of the mixture and place it on a glass slide to check the status of the cells under a light microscope; you should have a clear single-cell suspension. If there are still cell clumps, continue by passing the cells through a 27-gauge needle, followed by a 30-gauge needle. 7. Incubate the cell pellet for 3–5 min with ACK buffer at 37
C, and then centrifuge at 200 g for 5 min at 4
C. Discard the supernatant and resuspend the cell pellet with 1 mL of culturing media. 8. For spheres initiated with CSCs, the use of a serum-free medium is strongly commended. 9. After counting the freshly isolated single-cell suspension, cells can be kept on ice for up to 1 h before use. 10. If cells are less than 90% detached, increase the incubation time a few more minutes, checking for dissociation every 30 s. You may also tap the flask to expedite cell detachment. The incubation in 0.05% trypsin–EDTA solution varies (5–15 min) depending on the type of cell line being detached and its sensitivity to the trypsin enzyme. 11. Growth factor-reduced MatrigelTM should be thawed on ice before you start the cell isolation procedure. Always keep the mix of cells and MatrigelTM on ice in order to prevent the solidification of MatrigelTM. 12. Plating the matrix around the rim of the well is mostly recommended [13]. Avoid forming bubbles as bubbles create gaps in the MatrigelTM and lead to its breakage. To get rid of a bubble, carefully prick it with the needle of a 27-gauge syringe. A uniform and bubble-free plating can be confirmed under light microscopy. 13. It is very important to use warm media as cold media will lead to MatrigelTM dissociation. If you are interested in assessing the effects of a specific drug on the sphere-forming ability, a proper amount of the drug must be used in specific wells and compared to control wells with the appropriate media [13]. 14. It is very important to aspirate the media from the center of the well very gently so as not to disturb the MatrigelTM. 15. It is very important to periodically acquire bright-field images, using an inverted light microscope, of developing spheres to assess for any changes in morphology.

98

Amani Yehya et al.

16. The SFE has to be calculated through manual counting of the spheres. The spheres that are formed in the primary sphere formation assay, referred to as generation 1 spheres (G1), do not all account for stem cells, as progenitor cells can form short-lived spheres too. 17. Assess the sphere-forming efficiency in secondary, tertiary, quaternary, and quinary passages using the same formula described above (Subheading 3.3, step 8). References 1. Guo W, Lasky JL, Wu H (2006) Cancer stem cells. Pediatr Res 59(4):59–64 2. Reya T, Morrison SJ, Clarke MF, Weissman IL (2001) Stem cells, cancer, and cancer stem cells. Nature 414(6859):105–111 3. Tang DG (2012) Understanding cancer stem cell heterogeneity and plasticity. Cell Res 22(3):457–472 4. Islam F, Gopalan V, Lam AK-Y (2019) Chapter 6 – Cancer stem cells: role in tumor progression and treatment resistance. In: Dammacco F, Silvestris F (eds) Oncogenomics. Academic, pp 77–87 5. Ayob AZ, Ramasamy TS (2018) Cancer stem cells as key drivers of tumour progression. J Biomed Sci 25(1):20 6. Marzagalli M, Fontana F, Raimondi M, Limonta P (2021) Cancer stem cells-key players in tumor relapse. Cancers (Basel) 13(3) 7. Shibata M, Hoque MO (2019) Targeting cancer stem cells: a strategy for effective eradication of cancer. Cancers (Basel) 11(5) 8. Duan JJ, Qiu W, Xu SL, Wang B, Ye XZ, Ping YF et al (2013) Strategies for isolating and enriching cancer stem cells: well begun is half done. Stem Cells Dev 22(16):2221–2239 9. Akbarzadeh M, Maroufi NF, Tazehkand AP, Akbarzadeh M, Bastani S, Safdari R et al (2019) Current approaches in identification and isolation of cancer stem cells. J Cell Physiol 234(9):14759–14772 10. Weiswald L-B, Bellet D, Dangles-Marie V (2015) Spherical cancer models in tumor biology. Neoplasia 17(1):1–15 11. Cianciosi D, Ansary J, Forbes-Hernandez TY, Regolo L, Quinzi D, Gracia Villar S et al (2021) The molecular basis of different approaches for the study of cancer stem cells and the advantages and disadvantages of a

three-dimensional culture. Molecules 26(9): 2615 12. Chatzinikolaidou M (2016) Cell spheroids: the new frontiers in in vitro models for cancer drug validation. Drug Discov Today 21(9): 1553–1560 13. Bahmad HF, Cheaito K, Chalhoub RM, Hadadeh O, Monzer A, Ballout F et al (2018) Sphere-formation assay: three-dimensional in vitro culturing of prostate cancer stem/progenitor sphere-forming cells. Front Oncol 8: 347 14. Monzer A, Wakimian K, Ballout F, Al Bitar S, Yehya A, Kanso M et al (2022) Novel therapeutic diiminoquinone exhibits anticancer effects on human colorectal cancer cells in two-dimensional and three-dimensional in vitro models. World J Gastroenterol 28(33):4787–4811 15. Al Bitar S, Ballout F, Monzer A, Kanso M, Saheb N, Mukherji D et al (2022) Thymoquinone radiosensitizes human colorectal cancer cells in 2D and 3D culture models. Cancers (Basel) 14(6) 16. Bodgi L, Al-Choboq J, Araji T, Bou-Gharios J, Azzi J, Challita R et al (2022) Radiation treatment timing and dose delivery: effects on bladder cancer cells in 3D in vitro culture. Radiation 2(4):318–337 17. Bou-Gharios J, Assi S, Bahmad HF, Kharroubi H, Araji T, Chalhoub RM et al (2021) The potential use of tideglusib as an adjuvant radio-therapeutic treatment for glioblastoma multiforme cancer stem-like cells. Pharmacol Rep 73(1):227–239 18. Leelatian N, Doxie DB, Greenplate AR, Mobley BC, Lehman JM, Sinnaeve J et al (2027) Single cell analysis of human tissues and solid tumors with mass cytometry. Cytometry B Clin Cytom 92(1):68–78

Chapter 8 Multicellular Tumoroids for Investigating Cancer Stem-Like Cells in the Heterogeneous Tumor Microenvironment Kathleen M. Burkhard and Geeta Mehta Abstract Cancer stem-like cells (CSC) are a major contributing factor to chemoresistance, tumor recurrence, and poor survival outcomes in patients across cancer types. Signaling from non-tumor cells in the tumor microenvironment (TME) enriches for and supports CSC. This complex cell–cell signaling in the heterogeneous TME presents a challenge for patient survival; however, it also presents an opportunity to develop new targeted therapies that can inhibit survival of CSC. In this chapter, we report a multicellular tumoroid model which can be used to investigate the interactions between cancer cells and non-tumor cells in the TME to better understand the contribution of various cell types to cancer cell phenotypes, as well as the underlying mechanisms involved. The following methods allow for each cell type to be distinguished using FACS and studied individually. Gene expression can be analyzed for cancer cells, as well as the other non-tumor cells using qPCR following sorting. The response to chemotherapeutic agents and expression of stem markers can be determined for cancer cells using flow cytometry, excluding the other cell types to get an accurate view of the cancer cells. Furthermore, the viability of non-tumor cells can be analyzed as well to determine if there are cytotoxic effects of the drugs on non-tumor cells. Thus, the multicellular tumoroid model will reveal the interactions between the CSC and non-tumor cells in the heterogenous TME, resulting in discoveries in the fields of cancer biology, novel targeted therapies, and personalized drug screening for precision medicine. Key words Spheroids, Organoids, Tumoroids, Cancer stem-like cells, CSC, Heterogeneity, Tumor cells, Tumor microenvironment, Immune microenvironment, Mesenchymal stem cells, MSC, Carcinoma-associated fibroblasts, CAF, Macrophages, Carcinoma associated mesenchymal stem cells, CA-MSC, Alternately activated macrophages, AAM

1

Introduction Cancer stem-like cells (CSC) have the ability to self-renew and generate tumors. CSC are correlated closely with metastasis, are able to promote an immunosuppressive environment, and contribute towards developing a heterogeneous tumor microenvironment (TME) cell population [1]. Even though CSC comprise a rare subpopulation of all tumor cells, they are capable of repopulating an entire tumor [2]. Many studies in various solid cancers have

Federica Papaccio and Gianpaolo Papaccio (eds.), Cancer Stem Cells: Methods and Protocols, Methods in Molecular Biology, vol. 2777, https://doi.org/10.1007/978-1-0716-3730-2_8, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

99

100

Kathleen M. Burkhard and Geeta Mehta

implicated CSC in metastasis and resistance to chemo-, biologic, and immune therapies, thereby resulting in reduced survival across a variety of cancer types [3–6]. Although CSC have demonstrated their contributions to the pro-tumoral phenotypes mentioned above, they do not produce these phenotypes by themselves. Indeed, they reprogram non-tumor cells in the TME to promote CSC programs [7– 9]. Cells such as mesenchymal stem cells (MSC), carcinomaassociated fibroblasts (CAF), and macrophages stimulate CSC and cancer cells such that they enrich for pro-tumoral stem-like phenotypes [9–11]. As we decouple the various sources that provide CSC their survival, quiescence, proliferative, chemoresistant, and immune-evasive signatures, there is a critical need for in vitro models that can facilitate interaction studies between CSC and non-tumor TME cells. In fact, recent probing of these cell–cell interactions has identified many signaling pathways that are directly involved in maintaining the CSC niche, including TGF-β, Hedgehog, BMP4, IL-6, LIF, osteopontin, WNT, and PDGF [10– 16]. These signaling pathways are promising targets to inhibit the signals that promote CSC. Consequently, many new therapies have been developed with the goal of improving patient outcomes by targeting the tumor microenvironment [17–19]. These therapies have the potential to improve patient outcomes by reducing the likeliness of recurrence, drug resistance, and metastasis, and several are currently in active clinical trials or already approved for clinical practice. However, challenges remain due to drug toxicity, re-emergent chemoresistance, as well as the incomplete understanding of the complex signaling in the tumor microenvironment. In the campaign to expose the CSC interactions with non-tumor TME cells, many in vivo and in vitro approaches have been taken, for example, patient-derived xenografts (PDX) and organoids (PDO), respectively. These models incorporate heterogeneous cellular compositions of the TME and recapitulate the tissue architecture and gene expression of primary tumors and can predict the response of a patient to treatment [20–24]. However, the heterogeneity and patient-to-patient variability pose challenges in decoupling the specific mechanisms involved in the CSC-to-nontumor cell signaling between cancer cells and each of the other cell types. Therefore, in this chapter we describe an in vitro tumoroid model which can be used to study the interaction of cancer cells with non-tumor cells of the TME in a highly controllable manner. We coined the word “tumoroid” to distinguish the aggregates that contain other non-tumor cells from the cancer cell-only spheroids. We utilize cells labeled with distinct fluorescent proteins via lentivirus transduction in order to sort the cells from the tumoroids and study each cell type separately. Flow cytometry can be used to analyze stem marker expression on cancer cells, as well as viability

Multicellular Tumoroids for Investigating Cancer Stem-Like Cells in the. . .

101

in response to drug treatment. Furthermore, sorted cells can be analyzed using qPCR to study gene expression of each distinct population. Although we have utilized ovarian cancer tumoroids containing MSC and macrophages to illustrate the types of data that can be obtained from tumoroids, the methods that follow can be easily adapted to model other solid tumors including breast, prostate, and liver and incorporate other non-tumor cells such as carcinoma-associated fibroblasts, endothelial cells, and other immunocytes [25–27]. With this tumoroid model and protocols described below, the contribution of non-tumor cells to the CSC fates can be examined precisely in the heterogenous TME.

2

Materials

2.1 Cells and Culture Medium

1. Cancer cells: There are commercially available high-grade serous ovarian cancer cell lines, which include OVCAR3, OVCAR4, OVCAR5, CAOV3, Kuramochi, and OVSAHO. 2. MSC: There are commercially available cell lines for adiposederived human MSC, and bone marrow-derived MSC, as well as patient-derived carcinoma-associated MSC (CA-MSC) from matched patient tumor samples. 3. Monocytes, macrophages, and peripheral blood mononuclear cells (PBMC): Examples of human monocyte cell lines include U937 and THP1, from which macrophages can be derived. Viably frozen PBMC from healthy donor, as well as matched patient-derived PBMC, with limited proliferation potential can be used. 4. Cancer cell culture medium: RPMI 1640 supplemented with 2 mM L-glutamine, 10% fetal bovine serum (FBS), and 1× antibiotics/antimycotics. 5. U937/PBMC culture medium: RPMI 1640 supplemented with 2 mM L-glutamine, 10% heat-inactivated FBS, and 1× antibiotics/antimycotics. 6. MSC culture medium: Adipose-derived stem cell basal medium (ADSCBM) supplemented with 2 mM L-glutamine, 10% FBS, and 1× antibiotics/antimycotics. 7. Mϕ macrophage differentiation medium: U937/PBMC culture medium (see item 5) supplemented with 5 ng/mL phorbol myristate acetate (PMA). 8. M2-AAM polarization medium: U937/PBMC culture medium (see item 5) supplemented with 200 ng/mL macrophage colony-stimulating factor (M-CSF) and 200 ng/mL IL-4.

102

Kathleen M. Burkhard and Geeta Mehta

2.2 General Tissue Culture

1. T-75 tissue culture flasks. 2. 15 mL centrifuge tubes. 3. 1.5 mL microcentrifuge tubes. 4. Aspirator pipette tips. 5. 5 mL and 10 mL serological pipette tips. 6. P1000, P200, P20, and P2.5 pipettes and tips. 7. Hemacytometer. 8. 1x phosphate-buffered saline (PBS). 9. 0.25% trypsin–EDTA. 10. 0.05% trypsin–EDTA 11. Trypan blue.

2.3 Lentiviral Transduction

1. Six-well plates. 2. Lentivirus stock solution. 3. 8 μg/mL polybrene.

2.4 Cell Labeling with CellTracker Dye

1. CellTracker dye (Invitrogen) in colors of interest.

2.5 3D Hanging Drop Platform

1. Injection molded 384-well hanging drop plates.

2. Sterile DMSO.

2. Water bath sonicator. 3. 0.1% pluronic acid in deionized water. 4. Sterile deionized water. 5. UV light for sterilization. 6. Six-well plates. 7. Parafilm.

2.6 Fluorescence-Activated Cell Sorting

1. DAPI (40,6-diamidino-2-phenylindole, dihydrochloride). 2. APC-isotype IgG2b. 3. CD133-APC IgG2b. 4. FACS tubes. 5. ALDEFLUOR assay kit (Stem Cell): 50 μg dry ALDEFLUOR reagent, N,N-diethylaminobenzaldehyde (DEAB) reagent 1.5 mM in 95% ethanol, hydrochloric acid, dimethylsulfoxide, ALDEFLUOR Assay Buffer, and the ALDEFLUOR Quick Reference Guide. 6. AldeRed ALDH Detection Assay (MilliporeSigma): 50 μg of AldeRed reagent powder, 1 mL diethylaminobenzaldehyde (DEAB) reagent, 1.5 mL of 2 normal hydrochloric acid, 1.5 mL dimethylsulfoxide, 25 mL ×4 AldeRed Assay Buffer, and 615 μg ×4 verapamil powder.

Multicellular Tumoroids for Investigating Cancer Stem-Like Cells in the. . .

103

7. FACS buffer: 1x PBS supplemented with 2% FBS. 8. DAPI buffer: FACS buffer supplemented with 300 μM DAPI. 2.7

RT-qPCR

1. RNeasy Mini Kit (QIAGEN). 2. Nanodrop spectrophotometer. 3. High-capacity cDNA reverse transcription kit. 4. Nuclease-free water. 5. MicroAmp Optical 384-well reaction plate. 6. Power SYBR Green PCR master mix. 7. Primers for genes of interest.

2.8 Transwell Migration Assays

1. 24-well plate with Transwell inserts with a polyester membrane and 8 μm pore size (Corning). 2. Serum-free cell culture medium (RPMI 1640 with 2 mM Lglutamine, and 1× antibiotics/antimycotics). 3. 10% formalin (w/v). 4. DI water. 5. 100% methanol. 6. 0.5% crystal violet in 20% methanol.

3

Methods

3.1 Lentiviral Transduction of Adherent Cells

1. Aspirate cell culture medium from the adherent cells to be transduced. 2. Wash away excess medium with 1x PBS, and then aspirate. 3. Add enough trypsin–EDTA to cover the bottom of the cell culture dish (2 mL for T-75 flask). Use 0.25% trypsin–EDTA for cancer cells and 0.05% trypsin–EDTA for more sensitive cells such as mesenchymal stem cells (MSC). 4. Incubate for 5 min at 37 °C. 5. Check that the cells have detached from the flask using a light microscope. If many cells are still attached, place them in the incubator for more time. The incubation can be increased to a total of 10 min. 6. Once cells have detached, add FBS-containing medium equivalent to four times the volume of trypsin (8 mL media for T-75 flask) to deactivate the trypsin. Using a serological pipette, repeatedly pipette over the surface of the flask to ensure all cells are collected. 7. Add cells to a 15 mL conical tube and ensure they are uniformly suspended by pipetting a few times.

104

Kathleen M. Burkhard and Geeta Mehta

8. Collect 20 μL of the cell suspension in a 1.5 mL tube to count. 9. Add 2 μL trypan blue dye to the cells in the 1.5 mL tube (1:10 dilution) and pipette up and down to create a homogeneous solution. 10. Pipette 10 μL of cell + dye solution into each side of a manual hemacytometer. 11. Count the cells under 10× magnification and record the number of live, bright, round cells that have excluded the trypan blue dye. 12. Using the number of live cells found in step 11, calculate the concentration of the original cell suspension according to the instructions for your hemacytometer. 13. Plate the cells to be transduced in a six-well plate, at a density of approximately 10,000–20,000 cells per well. A minimum of one well for control and one well for transduction is needed. 14. The cells must be adhered to the tissue culture plate and healthy (40–50% confluent) before proceeding with transduction (see Note 1). 15. Aspirate the old medium and replace with 1.35 mL of fresh medium, 3 μL of 8 μg/mL polybrene, and 150 μL of 10× lentiviral stock. Adjust the concentration of the virus according to your cell type. Cancer cells maintain viability with a 10% (v/v) virus concentration, whereas MSC will not maintain their viability at more than 5% (v/v) virus concentration. 16. Incubate the cells with virus for 24 h at 37 °C. After 24 h, aspirate the virus into a container of 10% bleach and soak contaminated tubes and pipette tips in 10% bleach for 20 min before disposing into a biohazard waste container. 17. Add fresh medium and allow the cells to recover for 48 h at 37 °C. 18. Check for transduction efficiency (fluorescence) 48 h after transduction on a microscope. If transduction efficiency is greater than 80%, collect the successfully transduced cells using fluorescence-activated cell sorting (FACS). If not, leave the cells for another 24–48 h and check again. 19. To improve transduction efficiency, re-transduce the cells with virus after FACS using steps 1–18. 3.2 Lentiviral Transduction of Nonadherent Cells

1. Transfer the nonadherent cells to be transduced to a 15 mL tube and mix to ensure a homogeneous cell suspension. 2. Count the cells according to steps 8–12 in Subheading 3.1. 3. Plate the cells to be transduced in a six-well plate, with approximately 100,000 cells/mL and 1.35 mL per well. A minimum of one well for control and one well for transduction is needed.

Multicellular Tumoroids for Investigating Cancer Stem-Like Cells in the. . .

105

4. Add 3 μL of 8 μg/mL polybrene and 150 μL of 10× lentiviral stock. Virus concentration may have to be adjusted according to your cell type. 5. Incubate the cells with virus for 24 h at 37 °C. After 24 h, transfer the cells to a 15 mL conical tube and centrifuge at 400–800× g, depending on cell size, for 5 min. Aspirate the supernatant with virus into a container of 10% bleach and soak contaminated tubes and pipette tips in 10% bleach for 20 min before disposing into a biohazard waste container. 6. Resuspend the cells in fresh growth medium, and place in a new six-well plate. Allow the cells to recover for 48 h at 37 °C. 7. Check for transduction efficiency (fluorescence) 48 h after transduction on a microscope. If transduction efficiency is greater than 80%, collect the successfully transduced cells using FACS. If not, leave the cells for another 24–48 h and check again. 8. To improve transduction efficiency, re-transduce with virus after flow sorting. 3.3 Labeling Adherent or Nonadherent Cells with Fluorescent Cell Tracker

1. Dissolve the CellTracker dye powder in sterile DMSO to a final concentration of 10 mM. 2. Dilute the 10 mM CellTracker stock solution to a working concentration of 5 μM in cell culture medium (see Note 2). 3. Collect the cells to be labeled as in steps 1–7 in Subheading 3.1 for adherent cells, or following step 1 in Subheading 3.2 for nonadherent cells (see Note 3). 4. Transfer enough cells for the entire experiment into a new 15 mL tube. The number of cells needed for each hanging drop experimented will be calculated below, in Subheading 3.5 Generation of 3D Spheroids or Tumoroids with Single or Multiple Cell Types, Respectively. 5. Centrifuge the cells at 400–800× g, depending on cell size, for 5 min and remove the supernatant. 6. Gently add the pre-warmed CellTracker working solution to the pellet. 7. Incubate the cell + CellTracker solution in the 15 mL tube for 30 min at 37 °C and 5% CO2 for 30 min. 8. Centrifuge the cells at 400–800× g for 5 min and aspirate the supernatant to remove the CellTracker working solution. 9. Wash the cells in 1 mL PBS to ensure all excess CellTracker dye is removed (see Note 4). 10. Centrifuge the cells at 400–800× g for 5 min and aspirate the supernatant to remove the PBS.

106

Kathleen M. Burkhard and Geeta Mehta

11. The cells are now ready to be used for experiments. The pellet can be resuspended in 1 mL of media to be distributed into multiple tubes if that is necessary for the experiment. 3.4 Preparation of Hanging Drop Plates

1. Sonicate the 384-well hanging drop plates in the water bath sonicator for 20 min to dislodge any dust or debris in the wells and on the surface of the plates. 2. Rinse each side of the hanging drop plates well with sterile DI water to remove any dust and debris remaining on the surface. Shake the plates to remove water trapped in the wells. 3. Place the plates in 0.1% pluronic acid baths. Tap the plates around to remove air bubbles trapped in the wells so that the pluronic acid is in full contact with the wells. Let the plates soak for at least 8 h, or overnight. The pluronic acid reduces adhesion of cells to the plastic wells, allowing spheroid formation. 4. Remove the plates from the pluronic acid and rinse well with sterile DI water. Shake the plates to remove excess water. 5. Bring the plates to a sterile biosafety cabinet. Tap the plates hard on the surface, which will make a loud noise. Continue tapping each side of the plate until no more water leaves the wells and lands on the biosafety cabinet surface. Wipe up the excess water periodically to avoid splashing. 6. Once all the plates have been tapped, wipe up any remaining water and decontaminate the biosafety cabinet using 70% ethanol. 7. Lay out the hanging drop plates in a single layer on the biosafety cabinet surface. Make sure there is space between each plate to ensure there are no shadows that will block the UV light. 8. Turn the UV light on for at least 30 min, ideally 1 h, to sterilize the first side of the hanging drop plates. 9. Flip the plates over from the edge, taking care not to touch the area where the drops will be. Sterilize the second side with UV light for another 30–60 min. 10. Stack the hanging drop plates together and sandwich them between the base and lid of a 6-well plate until they are needed. 11. Before plating the hanging drops with cells, fill a sterile 6-well plate with 5 mL sterile DI water per well (see Note 5). Place a hanging drop plate on top of the base of the 6-well plate and add 2 mL of sterile DI water around the rim of the hanging drop plate. The water will help minimize evaporation of the drops. Place the lid of the 6-well plate on top of the hanging drop plate to protect it.

Multicellular Tumoroids for Investigating Cancer Stem-Like Cells in the. . .

3.5 Differentiation of U937 Monocytes to Macrophages

107

1. To differentiate U937 monocytes into Mϕ macrophages, transfer the U937 cells to a 15 mL conical tube and pipette up and down to create a homogeneous cell suspension. 2. Collect 20 μL of the U937 suspension to count, following steps 8–12 in Subheading 3.1. 3. Calculate the number of cells needed for the differentiation. Plate the U937 at a density of 2500 cells per drop in a hanging drop plate. Each plate will have 308 wells filled when one row is left empty around the edges to avoid evaporation. Round up to 325 wells per plate to account for pipetting error. For example, let A = the number of U937 cells needed and let B = the number of plates you are plating. Calculate the U937 cells needed (A) by multiplying the number of cells per drop by the number of drops per plate and the number of plates (B). (a) A = 2500 cells/drop * 325 drops/plate * B 4. Calculate the volume of cell suspension needed for the differentiation. For example, let C = the total plating volume. Determine the total volume (C) by multiplying the number of plates (B) by the volume per well and the number of wells per plate. (a) C = 20 μL/drop * 325 drops/plate * B 5. Transfer the volume of cell suspension calculated in step 3 into a new 15 mL conical tube. 6. Centrifuge the cells at 800× g for 5 min, and then aspirate the supernatant. 7. Resuspend the cells in the appropriate volume of Mϕ macrophage differentiation medium, calculated in step 4. 8. Plate 20 μL of the cell suspension in each well of a hanging drop plate (leaving one row of wells empty along each side) and incubate for 24 h to allow for monocytes to differentiate into unpolarized Mϕ macrophages in suspension. 9. To polarize Mϕ macrophages into M2-like alternately activated macrophages (AAM), add 2 μL of M2-AAM polarization medium to each drop at the 24-h time point. Incubate for another 24–48 h. To create Mϕ macrophages, do not add anything at the 24-h time point and continue to incubate for another 24–48 h. 10. After differentiation and polarization, macrophages can be collected from the hanging drop plate with a 1 mL pipette. The resulting macrophage spheroids are compact and will require trypsin to break apart. 11. Centrifuge the collected macrophages at 400× g for 5 min. 12. Aspirate the supernatant and resuspend the cells in 250 μL of 0.05% trypsin–EDTA.

108

Kathleen M. Burkhard and Geeta Mehta

13. Incubate for 5 min at 37 °C. 14. Deactivate the trypsin by adding 1 mL HI RPMI to the cells. 15. Pipette well to ensure all the spheroids have broken apart and there is a single-cell suspension. 16. This suspension can then be counted for addition to the tumoroids, as in steps 8–12 in Subheading 3.1. 3.6 Generation of 3D Spheroids or Tumoroids with Single or Multiple Cell Types, Respectively

1. Collect and count macrophages as in steps 10–16 in Subheading 3.5 if they will be used in the tumoroids (see Note 6). 2. Collect and count cancer cells and any other adherent cells that will be used in the tumoroids (such as MSC) following steps 1–12 in Subheading 3.1. 3. Collect and dye cells that will be labeled with CellTracker as in Subheading 3.3 if they will be used in the tumoroids. 4. Calculate the amount of cell suspensions needed to create the spheroids or tumoroids. For example, let D = the total number of cells (of a given cell type) needed. Let E = the desired number of cells (for a given cell type) per spheroid or tumoroid. Let F = the number of wells to be plated for a given condition. Calculate the number of cells for a given cell type needed (D) by multiplying the number of cells per well (E) by the number of wells (F). (a) D = E*F 5. Repeat step 4 for each cell type to be added to the spheroids or tumoroids. 6. Determine the volume of each cell suspension needed (G) by dividing the total amount of cells (of a given cell type) needed (D) by the concentration of the cell suspension (H) calculated in steps 1–3. (a) G = D/H. 7. Repeat step 6 for each cell type to be added to the spheroids or tumoroids. 8. Label 15 mL conical tubes for each spheroid/tumoroid condition you will have. 9. Add the volumes (G) calculated in step 6 for each cell type into the appropriate tubes. All tubes receiving cell type 1 will receive volume G1 of cell type 1 suspension. All tubes receiving cell type 2 will receive volume G2 of cell type 2 suspension. Tubes can receive G1 and G2 if the desired tumoroids will contain both cell types. 10. Centrifuge the tubes at 400× g for 5 min.

Multicellular Tumoroids for Investigating Cancer Stem-Like Cells in the. . .

109

11. Calculate the plating volume needed. Let I = the total plating volume. Multiply the number of wells (F) by the size of each drop, which is 20 μL/well. (a) I = F*20 μL/well 12. Aspirate the supernatant and resuspend the cells in volume I of cell culture medium (see Note 7). Tumoroids containing cancer cells with MSC or macrophages maintain cell viability using complete RPMI 1640. The appropriate cell culture medium for tumoroids may have to be optimized for other cell types. 13. Deposit 20 μL of well-mixed plating solution into each well of a prepared hanging drop plate. Mix the plating suspension periodically to ensure the cells are uniformly suspended while depositing solution in wells (see Note 8). 14. After all cells have been deposited in wells, place the six-well plate lid on top of the hanging drop plate and carefully wrap the edges with Parafilm before gently placing the plate in the incubator (see Note 9). 3.7 Maintenance of 3D Spheroids and Tumoroids

1. Once the hanging drops are plated, growth of the spheroids and tumoroids can be monitored by imaging them periodically using a phase contrast or fluorescent microscope. 2. Before imaging a plate, transfer it from the incubator to a clean biosafety cabinet. 3. Carefully remove and discard the Parafilm surrounding the plate. Take care during this step so that the Parafilm does not get stuck on the plate and jostle it, causing drops to fall. 4. Gently lift the hanging drop plate and lid off the base of the six-well plate. 5. Gently lower the hanging drop plate into place on the microscope stage and take images of the spheroids (see Note 10). 6. After imaging, bring the hanging drop plate back into the biosafety cabinet and replace it on the base of the six-well plate. 7. Add 2 μL of fresh medium to each well to feed the spheroids and tumoroids (see Note 11). 8. After feeding, carefully wrap the edges of each plate with new Parafilm and place in the incubator. 9. Hanging drops should be fed every 2–3 days.

3.8 FACS to Separate Cell Types from Tumoroids

1. Collect tumoroids and make single-cell suspensions for each culture condition according to steps 10–15 in Subheading 3.5 (see Note 12). 2. Centrifuge the single-cell suspensions at 400× g for 5 min. 3. Aspirate the supernatant and resuspend each pellet in 400 μL of FACS 4′,6-diamidino-2-phenylindole (DAPI) buffer (see Note 13).

110

Kathleen M. Burkhard and Geeta Mehta

4. Prepare collection tubes for each sample to be collected. Add approximately 2 mL of appropriate culture medium to each 15 mL conical tube. If two cell types will be collected from the same tumoroid condition, two collection tubes will be needed. 5. Sort for live cells (DAPI negative) that are positive for the fluorescent marker of one type and collect in one tube, and sort for live cells that are positive for the other fluorescent marker and collect in another tube. 6. After collection, centrifuge collection tubes at 400× g for 5 min. 7. Aspirate supernatant. 8. At this point, cells can be frozen and analyzed later, or resuspended and used immediately for further assays. 3.9 Gene Expression Analysis of Individual Cell Types

1. Collect and sort the spheroids for your cell types of interest as in Subheading 3.7 (see Note 14). 2. Freeze the collected cell pellets at -80 °C to store for RT-qPCR analysis. 3. Use the RNeasy Mini Kit and follow the instructions to isolate the RNA from your samples, including cancer cells, MSC, macrophages, or any other cell type you used. 4. Quantify the RNA concentration and evaluate purity using a Nanodrop spectrophotometer (see Notes 15 and 16). 5. Use a High-Capacity Reverse Transcriptase cDNA Transcription Kit to transcribe the RNA to cDNA. 6. Dilute the resulting cDNA in sterile nuclease-free water to a final concentration of 3.33 ng/μL. 7. Prepare a MicroAmp Optical 384-well reaction plate for qPCR. Include three housekeeping genes such as 18S, ACTB, and GAPDH. Up to 21 other genes can be investigated at the same time. Include triplicate reactions for each gene of each condition to ensure accuracy. 8. Dilute primer pairs to 2 μM in nuclease-free water. Create enough volume to add 2 μL of the solution to each well for triplicate of each condition. Let V = the volume of primer solution. Let W = the number of conditions. Multiply W by three for the triplicate and add 3 extra wells for pipetting error. Multiply this number of wells by 2 μL/well. (a) V = (3 W + 3)*2 μL/well 9. To the diluted primer pairs, add sybr green. Each well receives 5 μL sybr green, so multiply the number of wells used in step 8 by 5 μL/well. Let X = the amount of sybr green to add to each primer solution. (a) X = (3 W + 3)*5 μL/well

Multicellular Tumoroids for Investigating Cancer Stem-Like Cells in the. . .

111

Fig. 1 Changes in CSC-specific genes due to the interactions with mesenchymal stem cells and alternately activated macrophages in the tumoroids. The mRNA expression of genes associated with (a) stemness and (b) epithelial to mesenchymal transition was analyzed using RT-qPCR and 2-ΔΔCT. Ovarian cancer cell line OVCAR3 was used to create control spheroids, as well as tumoroids with OVCAR3 + MSC or OVCAR3 + M2-AAM. Gene expression was calculated as a fold change difference from OVCAR3 control spheroids. The dashed line shows gene fold of 1 for comparison with OVCAR3 control; downregulated genes have a fold change below 0.5 and upregulated genes have a fold change above 2.0

10. Add 7 μL of primer + sybr solution per well of the qPCR plate. Add the primer + sybr solution for one gene to all the wells of one column and make each column a different gene of interest (see Note 17). 11. Add 3 μL of cDNA to the wells of the appropriate experimental conditions. 12. The total volume in each well should equal 10 μL. Add 10 μL of nuclease-free water to any empty wells in the plate. 13. Cover the plate with an optical adhesive cover. 14. Perform RT-qPCR. 15. To determine the degree of change in expression of genes, use the 2-ΔΔCT method and compare the co-culture cells to their corresponding monoculture controls (Fig. 1a, b).

112

Kathleen M. Burkhard and Geeta Mehta

3.10 Flow Cytometry to Analyze Stem Markers on Cancer Cells

1. Create tumoroids using cancer cells that are transduced to express a fluorescent protein or labeled with CellTracker dye. 2. Collect the spheroids and create single-cell suspensions as in steps 10–15 in Subheading 3.5. 3. Collect cells that are not fluorescent as a negative control for gating, as in steps 1–7 in Subheading 3.1. 4. Centrifuge the single-cell suspensions at 400× g for 5 min. 5. Determine the number of tubes you will need for each condition (see Note 18). 6. Aspirate the supernatant and resuspend the cells in ALDEFLUOR assay buffer, with a volume of 100 μL per tube in each condition (see Note 19). For example, if you have two markers of interest and will need six tubes per condition, resuspend the cells in 600 μL of ALDEFLUOR assay buffer. 7. Set out 1.5 mL tubes so that there are enough for the number you determined in step 5 for each condition. Add 100 μL of the cell + buffer suspension to each 1.5 mL tube for the appropriate condition (see Note 20). 8. Add the correct volume of antibody/stain to the corresponding tubes based on manufacturer recommendations. 9. Vortex all tubes on highest power for 2–3 s to ensure the cells are mixed well with the antibody. 10. Incubate the cells for 30–45 min at 37 °C. 11. Vortex all the samples again. 12. Centrifuge the samples for 5 min at 400× g (see Note 21). 13. Aspirate the supernatant. 14. Resuspend the cells in 400 μL of FACS buffer for the unstained sample and 400 μL of DAPI buffer for the rest of the samples. 15. Add the 400 μL samples of stained cells + buffer to the FACS tubes and put them on ice. 16. Read the data on a flow cytometer and save the FCS files. 17. Import the FCS files into an analysis software such as FlowJo. 18. Open the non-transduced, unstained sample and change the axes to a SSC-H versus FSC-H plot. Make a polygon gate around the cells, avoiding debris and the QC beads, and label it “Cells” (Fig. 2a). 19. Within the “Cells” gate created in step 18, change the axes to FSC-H versus FSC-W. Make a rectangular gate around the cells, excluding 1% of cells that are far to the right, which are the doublets and triplets. Label this “Single Cells” (Fig. 2b).

Multicellular Tumoroids for Investigating Cancer Stem-Like Cells in the. . .

113

Fig. 2 Gating scheme to isolate the CSC marker expression in the tumor cell compartment of the multicellular tumoroids. CSC tagged with GFP were incorporated into tumoroids with CA-MSC and evaluated for ALDH expression after 7 days. The represented data are from a negative control tumor sample with a low number of cells with ALDH activity. (a) A polygon gate was drawn around the cell population, avoiding debris in the lower left. (b) The single cells were selected within this population. (c) Then DAPI exclusion was used to select the live cells within the single cell population. (d) Next, GFP was used to identify the GFP-tagged CSC within the live single-cell population. (e) Finally, ALDH activity could be quantified using AldeRed exclusively for live single CSC cells

20. Within the “Single Cells” population, change the axes to SSC-A versus GFP-A. Create a polygon gate to the right of the cells and containing approximately 0.50% of the cells to account for background fluorescence. Label this gate “GFP Cancer Cells.” 21. Copy the “Cells” and “Single Cells” gates into the tube containing DAPI only from the same spheroid condition. Within the “Single Cells” gate, change the axes to a histogram of the DAPI-A channel. Create a gate for live cells from the left side of the graph to cover the first peak. The DAPI+ cells usually begin around 103 on the x-axis, so this is a good area to end the gate around. Label this gate “Live Cells” (Fig. 2c). 22. Copy the “GFP Cancer Cells” gate from step 20 and paste it within the “Live Cells” population to select the single, live cancer cells (Fig. 2d).

114

Kathleen M. Burkhard and Geeta Mehta

23. Copy the “Cells,” “Single Cells,” “Live Cells,” and “GFP Cancer Cells” gates into the first isotype tube from the same spheroid condition. Within the “Live Cells” gate, change the axes to SSC-A versus the isotype channel-A. Create a polygon gate to the right of the cells containing 0.50% of the cells to account for nonspecific staining. Label this gate “Isotype.” 24. Copy the “Cells,” “Single Cells,” “Live Cells,” “GFP Cancer Cells,” and “Isotype” gates into the next tube with the marker corresponding to the isotype from the same spheroid condition. The cells in the isotype gate are the cells positive for the marker of interest (Fig. 2e). 25. Repeat steps 23–24 for all isotype–marker pairs within the spheroid condition. 26. Repeat steps 18–19 and 21–25 for all spheroid conditions. 27. The percentages of cancer cells positive for the cancer stem cell markers of interest, such as ALDH, CD133, and CD24, can be compiled in a spreadsheet and analyzed using a statistical and graphing software such as GraphPad Prism. 3.11 Flow Cytometry to Analyze Drug Response of Cancer Cells

1. Create spheroids and tumoroids using cells that are transduced to express a fluorescent protein or labeled with CellTracker dye, as in Subheading 3.6. 2. On day 3, feed 2 μL of cell culture medium to half of the wells in the plate as normal. Add 2 μL of drug solution to the other half of the wells in the plate. The drug will be diluted by a factor of 10 when it is added to the drop, so make sure the drug stock created is 10 times the final concentration you wish to test (see Notes 22 and 23). 3. Let the drug incubate with the spheroids and tumoroids for 48 h. 4. Collect the spheroids and tumoroids and create single-cell suspensions as in steps 10–15 in Subheading 3.5. 5. Centrifuge the single-cell suspensions at 400× g for 5 min. 6. Resuspend the cells in 400 μL DAPI buffer and transfer to a FACS tube. 7. Keep the samples on ice. 8. Collect cells from 2D that are not fluorescent, as well as the cells that are fluorescent, as in steps 1–7 in Subheading 3.5. 9. Transfer 100,000 of the unlabeled cells into each of two new 15 mL conical tubes. Transfer 100,000 of the fluorescent cells into one new 15 mL conical tube. Do this for each cell type present in the tumoroids. These will be control samples for setting up gates. 10. Centrifuge the cells at 400× g for 5 min. Aspirate the supernatant.

Multicellular Tumoroids for Investigating Cancer Stem-Like Cells in the. . .

115

Fig. 3 Gating scheme to analyze the cancer cell viability after drug treatment in the multicellular tumoroids. CSC tagged with GFP were incorporated into tumoroids with CA-MSC, and the data are from the DMSO control sample (no treatment). Drug or DMSO was added on day 5 and flow cytometry was performed on day 7 to analyze the viability of the cancer cells. (a) A polygon gate was drawn to select the cell population, avoiding debris. (b) Then single cells were selected using a rectangle gate. (c) Next, GFP expression was used to identify the cancer cells. (d) Finally, DAPI exclusion was used to determine the percentage of single cancer cells that were viable after drug treatment

11. Resuspend one of the unlabeled cell tubes and the fluorescent cell tube in FACS buffer. Resuspend the other unlabeled cell tube in DAPI buffer. 12. Transfer the cells from step 11 into FACS tubes and place on ice. 13. Run the samples on the flow cytometer and save the FCS files. 14. Begin the analysis as in steps 17–19 in Subheading 3.10 (Fig. 3a, b).

116

Kathleen M. Burkhard and Geeta Mehta

15. Copy the “Cells” and “Single Cells” gates into the tube containing fluorescent cancer cells only. Within the “Single Cells” gate, change the axes to a SSC-A versus the fluorescent channel-A. Create a polygon to the right of the cells containing 0.50% of the cells. Label this gate “Cancer Cells” (Fig. 3c). 16. Copy the “Cells,” “Single Cells,” and “Cancer Cells” gates into the tube containing DAPI only. Within the “Cancer Cells” gate, change the axes to a histogram of the DAPI-A channel. Create a gate for live cells from the left side of the graph to cover the first peak. The DAPI+ cells usually begin around 103 on the x-axis, so this is a good area to end the gate around. Label this gate “Live Cancer Cells” (Fig. 3d). 17. Copy the “Cells,” “Single Cells,” “Cancer Cells,” and “Live Cancer Cells” gates into all of the spheroid tubes. The percentage of cells in the live “Live Cancer Cells” gate reveals the viability specifically of the cancer cells and excludes viability information on any other cell type in the spheroids and tumoroids. 18. Repeat steps 14 through 17 using a different fluorophore to create gates for the other cell types present in the tumoroid. This will provide viability and cytotoxicity information for the non-tumor cell. 19. The percentages of viable cancer cells in response to a therapeutic agent can be compiled in a spreadsheet and analyzed using a statistical and graphing software such as GraphPad Prism. 3.12 Transwell Assays to Investigate Cancer Cell Migration

1. Create tumoroids using cancer cells that are transduced to express a fluorescent protein, or labeled with CellTracker dye. 2. Collect the spheroids and create single-cell suspensions as in steps 10–15 in Subheading 3.5. 3. Use FACS to collect the fluorescent cancer cells from the tumoroids, as in Subheading 3.8 (see Note 24). 4. Resuspend the resulting cancer cells in serum-free cell culture medium (see Note 25). 5. Count the cells following steps 8–12 in Subheading 3.1. 6. Plate 200,000 cells in 200 μL serum-free medium into the transwell membrane for each condition (see Note 26). 7. Add 600 μL of complete medium to the bottom well so that the medium just touches the bottom of the membrane. 8. Incubate the plate for 48 h at 37 °C and 5% CO2. Note: The incubation time may need to be adjusted for your experimental conditions.

Multicellular Tumoroids for Investigating Cancer Stem-Like Cells in the. . .

117

9. After the 48-h incubation, carefully aspirate the cell culture medium from the top and bottom chambers of the transwell plate. 10. Gently rinse each membrane and well with 1x PBS, and then aspirate to remove. 11. Bring the samples to the chemical fume hood and add 1 mL 10% formalin to each well, covering the bottom of the well, the bottom of the membrane, and the top of the membrane to fix all the cells. 12. Let the wells incubate in the formalin at room temperature for 10 min. Then, remove the formalin from the plates into a chemical waste container. 13. Add 1 mL DI water to each well to wash. Then, remove the DI water into the waste container. 14. Add 1 mL 100% methanol to each well and incubate at room temperature for 30 min. Then, remove the 100% methanol into the waste container. 15. Add 1 mL DI water to each well to wash. Remove the DI water into the waste container. 16. Add 1 mL 0.5% crystal violet to each well and incubate at room temperature for 20 min. 17. Remove the crystal violet from the wells. The crystal violet can be reused, so it can be returned to a tube for storage, protected from light. 18. Add 1 mL DI water to each well to wash. Remove the DI water into the waste container. 19. Repeat step 18 until the water is clear and there is no crystal violet residue left behind on the well bottom or membrane. Typically, 3 to 4 washes are needed. 20. While the membranes are still wet, use a cotton swab to carefully remove all the cells from the top of the membrane. Do not use too much pressure, as it could cause the membrane to warp. If the cells are not coming off, add a little bit of water to the cotton swab to make it slightly damp. Be careful not to touch or rub the bottom of the membrane, since this is where the migrated cells are. 21. Leave the transwell plate(s) in the hood with the lid slightly cracked so they can fully dry. 22. The transwell membranes can be imaged directly using a color bright-field microscope to visualize the cells on the bottom of the membrane that successfully migrated through (see Notes 27 and 28). Take three to five representative images of each membrane.

118

Kathleen M. Burkhard and Geeta Mehta

23. Count the number of cells in each image to quantify cell migration. Automatic cell quantification techniques can confuse the pores of the transwell membrane with cells, so manual counting is the most accurate method.

4

Notes 1. Transduction efficiency will be reduced if cells are overconfluent. Avoid plating cells at a high density and consider the growth rate of your particular cells and the incubation time needed for the transduction efficiency so that the cells are not overconfluent by the end. 2. The CellTracker working concentration can be increased up to 25 μM if necessary, but cell viability is typically better at lower concentrations and worsens as the concentration increases. 3. The CellTracker dye is divided between daughter cells, which results in lower fluorescence intensity with each cell division. Therefore, the CellTracker dye is more effective for long-term tracking (over several days) on cells that have lower proliferation rates. It may be helpful to choose cells with lower proliferation rates as candidates for CellTracker dye labeling and cells with higher proliferation rates as candidates for labeling with lentivirus transduction. 4. Washing cells with PBS after CellTracker labeling is a critical step to ensure that no dye is transferred to other cell types that will be added to the spheroids. 5. Six-well plates used to sandwich hanging drop plates can be reused to reduce waste. If no drops have fallen into the 6-well plate, it can be sterilized using UV light and then used again. However, it is best to use a new sterile 6-well plate if drops have fallen to avoid microbe growth and contamination. 6. It takes some time to image and collect macrophages (approximately 6 min per plate depending on the person). It is best to image the plates first and then collect the macrophages to reduce the amount of time the cells are out of the incubator. It is also best to collect the macrophages before the other adherent or nonadherent cells because the macrophages take the most time and this order will reduce the amount of time the other cells spend outside of the incubator to preserve maximum viability. 7. Be careful while transferring the tubes from the centrifuge to the biosafety cabinet because bumping or shaking the tubes can dislodge the cell pellet. Aspirate the supernatant carefully by slowly following the meniscus down the tube and take care not to get the aspirator tip too close to the cell pellet or else it could remove the cells.

Multicellular Tumoroids for Investigating Cancer Stem-Like Cells in the. . .

119

8. Cells will settle to the bottom of tubes and reservoirs quickly since they are denser than the cell culture medium. Periodic mixing of the cell suspension is essential to ensure spheroids and tumoroids are consistent in size and cell number. 9. Wrapping the hanging drop plate sandwich in Parafilm is one of the steps that is most likely to result in drops falling. Parafilm does not always stretch the same amount before tearing, so it is safest to pre-stretch the Parafilm in your hands and then wrap it around the edges of the hanging drop plate sandwich while applying minimal tension. Using enough Parafilm also makes this step easier. Usually, using one piece of Parafilm that is approximately 2 inches by 5 inches for the front of the plate and another piece for the back of the plate is a sufficient size to easily cover all the seams without resulting in excess waste. This reduces the risk of the Parafilm breaking and jostling the hanging drop plate which would result in fallen drops. 10. The hanging drop plate will fit in the stage for a standard well plate. 11. Approximately 2 μL will evaporate from each drop over the course of 2–3 days in the incubator and the time outside of the humidified 6-well plate for imaging and feeding. Adding 2 μL of fresh medium will replenish nutrient and maintain the total drop volume at 20 μL. 12. Some tumoroid conditions may form more compact spheroids with tighter cell–cell connections than other tumoroid conditions depending on the cell types used. To avoid introducing confounding variables, try to pipette approximately the same amount for each condition while breaking up the spheroids. Excessive pipetting introduces stress on the cells and uneven pipetting could therefore affect cell viability or other results. 13. The volume of DAPI buffer may need to be adjusted depending on the number of cells in the sample. The chance of clogs occurring in the flow cytometer is high once the concentration is above one million cells/mL, so you may need to add more DAPI buffer to reduce the concentration. Conversely, a very dilute sample will take a long time to run, so avoid adding excess DAPI buffer. 14. The number of spheroids required to generate sufficient RNA for analysis will vary depending on the proliferation rate of the cell type and the number of cells used in each spheroid. 15. The 260/280 absorbance ratio should be about 2.0 for pure RNA. 16. The RNA concentration should be greater than 70 ng/μL to generate enough cDNA for RT-qPCR analysis, but this may vary depending on the extent of your analyses.

120

Kathleen M. Burkhard and Geeta Mehta

17. It is strongly recommended to print a template of a 384-well plate and make note of which primers and samples are in each well. 18. Each spheroid condition will need the following tubes: one that is unstained, one with DAPI only, one for each marker of interest, and one isotype control tube for each type of antibody used (most likely you will need one isotype tube for each marker you use). The non-transduced cells will only need two tubes: unstained and DAPI. 19. If the cancer cells express GFP, use the AldeRed (488- and 532-nm excitation/615-nm emission) assay buffer to avoid ALDEFLUOR (488-nm excitation/535-nm emission) overlap with GFP (488-nm excitation/510-nm emission). 20. Add the cell + buffer solution to the tubes receiving antibody first and do the control tubes last. This will ensure that each tube being stained with antibody has 100 μL and that the antibody concentration is the same across conditions. Pipetting errors can result in not having a full 100 μL for each tube, but if the control tubes are less than 100 μL, the analysis will not be impacted. 21. Place all the tubes in the centrifuge with the same orientation so that the cell pellet is in the same place in each tube after centrifugation. This will make it easier to find the pellet especially if the samples contain small numbers of cells. 22. Many anticancer drugs are poorly soluble in water and are dissolved in organic solvents, commonly DMSO. Organic compounds are toxic to cells, so the amount in the final solution should be minimized. Additionally, a vehicle control should be used, especially if a large amount of organic solvent will end up in the final solution to control for the effects of the drug versus the organic solvent. 23. The solubility of the drug in cell culture medium for the desired concentration should also be tested prior to the experiment. It is possible for the drug to precipitate out of solution when the stock is added to the cell culture medium and this can result in inaccurate and inconsistent drug concentrations. 24. It is also possible to plate the single-cell suspensions from the tumoroids directly into the transwell membrane. This may be a good option if your cells are sensitive and lose viability from the flow sorting process. With this method, the same number of spheroids should be added to the membranes for each condition to ensure consistency. Fluorescence imaging can then be used to image the cells that migrated based on the color they are marked with (ex: GFP cancer cells).

Multicellular Tumoroids for Investigating Cancer Stem-Like Cells in the. . .

121

25. The absence of FBS in the cell culture media will provide a driving force for the cells to migrate through the transwell membrane to the bottom chamber. 26. The concentration of cells added to each membrane may need to be adjusted for your experimental conditions. 27. The migrated cells stained with crystal violet can also be imaged using fluorescence microscopy in the Texas red channel. 28. The membranes can be imaged directly, or they can be mounted on microscope slides first. If imaging directly and using fluorescence microscopy, add some DI water or PBS to the well to cover the membrane. This reduces the background signal from the pores of the transwell membrane. Funding Acknowledgements This work is supported primarily by the American Cancer Society Research Scholar Award RSG-19-003-01-CCE (G.M.), the Office of the Assistant Secretary of Defense for Health Affairs through the Ovarian Cancer Research Program under award no. W81XWH-16-1-0426 (G.M.), W81XWH-13-1-0134 (G.M.), DOD Investigator Initiated award W81XWH-18-0346 (G.M.), Rivkin Center for Ovarian Cancer (G.M.), and the Michigan Ovarian Cancer Alliance (G.M.), and NSF EFRI DChem (award number 2029139). Research reported in this publication was supported by the National Cancer Institute under award number P30CA046592 and by the National Institutes of Health under grant number UL1TR002240. Author Contributions Conceptualization, G.M.; formal analysis, K.M.B., G.M.; writing–original draft preparation, K.M.B., G.M.; writing–review and editing, K.M.B., G.M.; supervision, G.M.; funding acquisition, G.M. All authors have read and agreed to this version of the manuscript.

References 1. Atashzar MR, Baharlou R, Karami J et al (2020) Cancer stem cells: a review from origin to therapeutic implications. J Cell Physiol 235: 790 2. Kleinsmith LJ, Pierce GB (1964) Multipotentiality of single embryonal carcinoma cells. Cancer Res 24:1544 3. Lathia J, Liu H, Matei D (2020) The clinical impact of cancer stem cells. Oncologist 25: 1 2 3 . h t t p s : // d o i . o r g / 1 0 . 1 6 3 4 / t h e oncologist.2019-0517 4. Abdullah LN, Chow EK (2013) Mechanisms of chemoresistance in cancer stem cells. Clin

Transl Med 2. https://doi.org/10.1186/ 2001-1326-2-3 5. He L, Wick N, Germans SK, Peng Y (2021) The role of breast cancer stem cells in chemoresistance and metastasis in triple-negative breast cancer. Cancers (Basel) 13:6209 6. Motohara T, Yoshida GJ, Katabuchi H (2021) The hallmarks of ovarian cancer stem cells and niches: exploring their harmonious interplay in therapy resistance. Semin Cancer Biol 77:182. https://doi.org/10.1016/j.semcancer.2021. 03.038

122

Kathleen M. Burkhard and Geeta Mehta

7. Oshimori N, Guo Y, Taniguchi S (2021) An emerging role for cellular crosstalk in the cancer stem cell niche. J Pathol 254:384 8. Das M, Law S (2018) Role of tumor microenvironment in cancer stem cell chemoresistance and recurrence. Int J Biochem Cell Biol 103: 115. https://doi.org/10.1016/j.biocel.2018. 08.011 9. Zheng X, Yu C, Xu M (2021) Linking tumor microenvironment to plasticity of cancer stem cells: mechanisms and application in cancer therapy. Front Oncol 11:678333 10. Raghavan S, Mehta P, Xie Y et al (2019) Ovarian cancer stem cells and macrophages reciprocally interact through the WNT pathway to promote pro-tumoral and malignant phenotypes in 3D engineered microenvironments. J Immunother Cancer 7:190. https://doi.org/ 10.1186/s40425-019-0666-1 11. Raghavan S, Snyder CS, Wang A et al (2020) Carcinoma-associated mesenchymal stem cells promote chemoresistance in ovarian cancer stem cells via pdgf signaling. Cancers (Basel). https://doi.org/10.3390/cancers12082063 12. Kan T, Wang W, Ip PP et al (2020) Single-cell EMT-related transcriptional analysis revealed intra-cluster heterogeneity of tumor cell clusters in epithelial ovarian cancer ascites. Oncogene 39:4227. https://doi.org/10.1038/ s41388-020-1288-2 13. Cascio S, Chandler C, Zhang L et al (2021) Cancer-associated MSC drive tumor immune exclusion and resistance to immunotherapy, which can be overcome by Hedgehog inhibition. Sci Adv 7:eabi5790. https://doi.org/10. 1126/sciadv.abi5790 14. Coffman LG, Choi YJ, McLean K et al (2016) Human carcinoma-associated mesenchymal stem cells promote ovarian cancer chemotherapy resistance via a BMP4/HH signaling loop. Oncotarget 7:10.18632/oncotarget.6870 15. McLean K, Tan L, Bolland DE et al (2019) Leukemia inhibitory factor functions in parallel with interleukin-6 to promote ovarian cancer growth. Oncogene 38:1576. https://doi.org/ 10.1038/s41388-018-0523-6 16. Kuo MC, Kothari AN, Kuo PC, Mi Z (2018) Cancer stemness in bone marrow micrometastases of human breast cancer. Surgery (United States) 163:330. https://doi.org/10.1016/j. surg.2017.07.027

17. Zhong S, Jeong JH, Chen Z et al (2020) Targeting tumor microenvironment by smallmolecule inhibitors. Transl Oncol 13:57 18. Skorupan N, Dominguez MP, Ricci SL, Alewine C (2022) Clinical strategies targeting the tumor microenvironment of pancreatic ductal adenocarcinoma. Cancers (Basel) 14:4209 19. Faraj JA, Al-Athari AJH, Mohie SED et al (2022) Reprogramming the tumor microenvironment to improve the efficacy of cancer immunotherapies. Med Oncol 39:239 20. Nero C, Vizzielli G, Lorusso D et al (2021) Patient-derived organoids and high grade serous ovarian cancer: from disease modeling to personalized medicine. J Exp Clin Cancer Res 40:116 21. Rae C, Amato F, Braconi C (2021) Patientderived organoids as a model for cancer drug discovery. Int J Mol Sci 22:3483 22. Maru Y, Hippo Y (2019) Current status of patient-derived ovarian cancer models. Cell 8: 505 23. Mehta P, Novak C, Raghavan S et al (2018) Self-renewal and CSCs in vitro enrichment: growth as floating spheres. In: Papaccio G, Desiderio V (eds) Methods in molecular biology: cancer stem cells: methods and protocols, pp 61–75 24. Bregenzer M, Horst E, Mehta P et al (2022) The role of the tumor microenvironment in CSC enrichment and Chemoresistance: 3D co-culture methods. Methods Mol Biol. https://doi.org/10.1007/978-1-0716-19568_15 25. Neuwirt H, Bouchal J, Kharaishvili G et al (2020) Cancer-associated fibroblasts promote prostate tumor growth and progression through upregulation of cholesterol and steroid biosynthesis. Cell Commun Signal 18:11. https://doi.org/10.1186/s12964-0190505-5 26. Hsiao AY, Tung YC, Qu X et al (2012) 384 hanging drop arrays give excellent Z-factors and allow versatile formation of co-culture spheroids. Biotechnol Bioeng 109: 1293. https://doi.org/10.1002/bit.24399 27. Raghavan S, Mehta P, Horst EN et al (2016) Comparative analysis of tumor spheroid generation techniques for differential in vitro drug toxicity. Oncotarget. 10.18632/oncotarget.7659 7:16948

Chapter 9 Generation, Expansion, and Biobanking of Gastrointestinal Patient-Derived Organoids from Tumor and Normal Tissues Manuel Cabeza-Segura, Blanca Garcia-Mico´, Andre´s Cervantes, and Josefa Castillo Abstract Patient-derived organoids (PDOs) generated from adult stem cells present in tissues are invaluable tools for translational cancer research (Drost, Clevers, Nat Rev Cancer 18(7):407–418, 2018). The generation of this 3D cultures is not trivial and requires dedicated procedures. Despite the rapid increase in the use of organoids in cancer research, it is noteworthy that published procedures regarding their generation often lack critical information and standardized protocols remain elusive. Addressing these limitations, the protocol described in this chapter offers an in-depth and comprehensive guide to establishing, expanding, and freezing gastrointestinal PDOs obtained from normal and tumor tissue biopsies. Notably, it also provides valuable insights in the form of tips and tricks to guide and overcome potential challenges that may arise during the procedure. Key words Patient-derived organoids, Tumour tissue, Normal epithelial tissue, Adult stem cells, Colorectal cancer, Gastric cancer, Organoid establishment, Organoid expansion, Organoid cryopreservation, Drug sensitivity assay

1

Introduction The establishment of patient-derived organoids from tumor tissue biopsies has gained an enormous interest as a valuable approach for modelling cancer diseases. These organoids hold immense potential for discovering new biomarkers and predicting drug responses in patients turned them into an invaluable tool in translational research [1–6]. Gastrointestinal organoids, which are simplified 3D in in vitro epithelial models derived from adult stem cells found in tissue biopsies, can be expanded in vitro while retaining

Authors Manuel Cabeza-Segura and Blanca Garcia-Mico´ have equally contributed to this chapter. Federica Papaccio and Gianpaolo Papaccio (eds.), Cancer Stem Cells: Methods and Protocols, Methods in Molecular Biology, vol. 2777, https://doi.org/10.1007/978-1-0716-3730-2_9, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

123

124

Manuel Cabeza-Segura et al.

their intrinsic features. This makes them robust personalized models [7–12]. However, the generation of these models presents a substantial challenge. In this chapter we present our protocol and share our experiences in establishing, expanding, and cryopreserving organoid cultures derived from colon and stomach tissue biopsies of cancer patients. While there are informative articles available that very helpfully describe similar protocols for gastrointestinal organoids [7, 13, 14], we think there is a need to not only provide the protocols but also discuss the challenges associated with establishing such models. Moreover, we recognize the importance of sharing practical tips and tricks to overcome potential difficulties that may arise during these procedures. We have successfully developed a standardized protocol for generating gastric and colon organoids, with minimal variations, encompassing both healthy and tumor tissue sources (Fig. 1). It is important to highlight that only in the case of colon can we use a selective medium that avoids the overgrowth of contaminant normal organoids [7]. This chapter is divided into four major procedures starting with how to process human biopsy tissue (protocol 1), establishment of organoids (protocol 2), organoid maintenance and passaging to expand the culture (protocol 3), and cryopreservation and thawing (protocol 4). We have tried to explain in detail the tricks to successfully establish PDOs which are not usually found in the literature. Additionally, we have supplemented this chapter with a protocol to perform drug sensitivity assays since it is one of the main applications of PDOs (protocol 5). It is important to point out that the generation of PDOs requires written informed consent from patient donor and their management must comply with the institutional regulation and

Fig. 1 Flowchart of gastrointestinal PDO generation, showing representative images of gastric and colon PDOs (on the right)

Gastrointestinal Patient-Derived Organoids: Generation and Biobanking

125

legal and ethical normative. Moreover, a highly interconnected sample collection circuit should be established, involving the participation of several professionals: oncologists, surgeons, endoscopists, pathologists, and researchers. Alternatively, PDOs could be also obtained from a biobank or the research community (go directly to protocols 3 and 4).

2

Materials All materials used should be sterile. Tissue processing and cell culture procedure should be executed with proper aseptic technique to prevent cell culture contamination inside a Class II biosafety cabinet. Other equipment needed include a 5% CO2 incubator at 37 °C, inverted light microscope, centrifuge, and microcentrifuge. Waste material, i.e., biological residues, must be conveniently wrapped in appropriate containers following the legislative framework. All the solutions and reagent need to be prepared using Ultrapure water (18.2 MΩ·cm) and stored at indicated temperature. The plastic material used for handling and culturing organoids must be nonstick (see Note 1).

2.1 Sample Collection and Preparation (or Handling)

1. PBS (phosphate-buffered saline) (1X, pH 7.4).

2.2 Tissue Processing

1. Prepare the stock solution of Liberase TL (thermolysin low) at 13 U/mL. Subsequently, the working concentration should be 0.3 U/mL diluted in PBS (see Note 4).

2. PBS + 2X antibiotics: PBS supplemented with 2X of penicillin/ streptomycin (P/S) and 200 μg/mL of Normocin (see Notes 2 and 3).

2. DNAse I solution: 30 mg/mL diluted in sterile H2O. 3. ROCK inhibitor (RocKi) Y-27632: 10 mM in DMSO (see Note 5). 4. 70 μm cell strainer. 5. 60 mm petri dishes. 6. Scalpel. 7. 50 mL and 15 mL Falcon tube. 8. Red blood cell lysis buffer (see Note 6). 9. Basal medium (BM): DMEM/F12, HEPES 10 mM, L-glutamine 2 mM, P/S 1.5X, Normocin 150 μg/mL, N2 supplement 1X, and NeuroCultTM SM1 1X. Filter media through a 0,22 μm Stericup vacuum filtration system (see Note 7).

126

Manuel Cabeza-Segura et al.

2.3 Organoid Culture in BME

1. Neubauer chamber. 2. Trypan blue solution. 3. Basement membrane extract (BME): thaw and keep on ice (see Note 8). 4. Colon tumor medium (CTM): BM supplemented with N-acetyl-L-cysteine 1 mM, nicotinamide 10 mM, EGF 50 ng/mL, Noggin 100 ng/mL, A-8301 0,5 μM, SB202190 10 μM, gastrin 10 nM, R-Spondin1 500 ng/mL, FGF10 10 ng/mL, PGE2 10 nM, and RocKi Y-27632 10 μM (see Note 9). 5. StemCell medium (STM): IntestiCult OGM Human Basal Medium, organoid supplement, P/S 1.50X, and Normocin 150 μg/mL. The composition of the ST medium is basically IntesticultTM Organoid Growth Medium Human (StemCell Technologies) supplemented with antibiotics (see Note 10). 6. StemCell medium alternative without organoid supplement (STA): IntestiCult OFM Human Basal Medium from stem cell medium, DMEM/F12, P/S (1.50X), and Normocin (150 μg/mL) (see Note 11). 7. 24-well cell culture plates.

2.4 Organoid Passaging and Cryopreservation

1. TrypLE Express enzyme. 2. Bambanker. 3. Cryo Falcon tubes. 4. CoolCell. 5. -80 °C freezer. 6. N2 liquid tank.

2.5 Drug Sensitivity Assay of Tumor Organoids

1. CellTiter Glo 3D® reagent. 2. 96-well round flour cell culture plate. 3. Luminescence plate reader. 4. Drugs to be tested and the vehicle solution controls. 5. Multichannel pipette. 6. Solution basins.

3

Methods The described procedures are carried out inside the culture hood unless otherwise specified. Once organoid culture is established, mycoplasma testing should be performed regularly. Short tandem repeat (STR), single-nucleotide polymorphism, or NGS-based methods are recommended to authenticate patient-derived organoids, which is essential to build and maintain a high-quality biobank.

Gastrointestinal Patient-Derived Organoids: Generation and Biobanking

3.1 Primary Tissue Processing

127

1. Before starting preheat the 24-well plate in the CO2 incubator at 37°C. Prepare a 50-mL Falcon tube with 10 mL of cold PBS + 2X antibiotics to receive the tissue sample. Clean the culture hood and put inside all the plastic material needed: 60 mm Petri dishes, sterile tips, and serological pipettes with filter, 15 mL and 50-mL Falcon tubes, scalpels, and cryotubes. Leave the UV light on for 15 min (see Note 12). Cool the centrifuge at 4 °C. 2. Collect the sample in the PBS Falcon tube. Place the tube in an ice bucket. 3. Out of the hood decant the PBS in a sterile Falcon tube, being careful not to throw the sample and perform 6–7 washes with cold PBS, each with ≈10 mL (see Note 13). 4. Add the sample to a new Falcon tube with PBS + 2X antibiotics and incubate 20 min at room temperature (RT) (see Note 14). 5. Wash three times for 5 min with sterile PBS without antibiotics. In the last wash we leave approximately 1 mL. 6. Place the fragments in a 60 mm Petri dish with the 1 mL of PBS (used to decant the piece of the tube to the plate). 7. Cut/crush with a scalpel into very small fragments and recover everything in a 15 mL Falcon tube (see Note 15). 8. Digest the tissue with Liberase at a final concentration: 0.3 U/ mL. From a stock solution of 13 U/mL, 23 μL of Liberase is added for each mL of sample. 9. Incubate 45–60 min at 37 °C (bath) and pipetting with a P1000 tip inside the hood every 15 min (see Note 16). 10. After mechanical and enzymatic dissociation of tissue, filter tube content through a 70 μm cell strainer into a new 50-mL Falcon tube. Use 1 mL of PBS to wash the strainer well to recover most of the released tissue cells. 11. Centrifuge at 500× g for 5 min, 4 °C. Discard the supernatant very carefully (see Note 6). 12. Resuspend the final cell pellet in 500 μL of cold BM (see Note 17). 13. Count the number of cells with Neubauer chamber taking 5 μL of tube content and adding 45 μL of trypan blue (see Note 18).

3.2 Organoid Establishment Protocol

1. Plate cells at a density between 200,000 and 300,000 cells in 50 μL of 50% BME (diluted in BM) per well of 24-well plate (see Notes 19 and 20). To do that, calculate the volume of cells according to the number of wells that will be plated. Transfer this volume to a 1.5 mL Eppendorf. 2. Centrifuge at 500× g for 5 min, 4 °C. Remove the supernatant carefully.

128

Manuel Cabeza-Segura et al.

3. Calculate the total volume of 50% BME required (number of wells × 50 μL per well) (see Note 21). 4. Remove the 24-well plate from the incubator just before plating. Try to keep its temperature at 37 °C all the time. 5. Resuspend the cells in the volume of cold BM needed and add the same volume of 100% BME to reach the concentration of 50%. Quickly, plate 50 μL per well in the tempered 24-well plate, forming a dome, without introducing bubbles (see Note 22). 6. Incubate the plate for 20–30 min in the incubator (37 °C, 5% CO2) to allow the BME to solidify. 7. Once the BME has solidified, add 500 μL of appropriate 37 °Ctempered medium in each well. Add medium ST for normal colon organoids and gastric organoids. Add selective CTM medium for tumor colon organoids. In all cases RocKi Y-27632 must be added (see Note 23). 8. Place the plates in the cell culture incubator at 37 °C, 5% CO2. 9. Refresh the medium every 2–3 days. Remove the medium from the well by slightly tilting the plate and bringing the tip closer to the edge, be careful not to touch the BME dome. Add 500 μL of appropriate medium plus Rocki Y-27632 tempered to 37 °C in each well only in the first change of medium; in the subsequent changes of medium it is not necessary to maintain the Rocki Y-27632. 3.3 Passaging and Expansion of Organoids

1. When organoid culture reaches high confluence inside the dome, it must be splitted and expanded to more wells (see Fig. 1 as an example) (see Note 24). 2. Remove the medium of each well with pipette and perform a wash with PBS tempered at 37 °C. Remove the PBS with the pipette. 3. Add 1 mL of cold PBS and disrupt mechanically the BME dome. Centrifuge at 500× g for 5 min at 4 °C and discard the supernatant. 4. Add 250 μL of trypsin Express to the organoid sediment. Resuspend with pipette and incubate at 37 °C in a bath no more than 5 min (see Note 25). 5. Add 2 mL of cold BM and mix well. 6. Centrifuge 500× g for 5 min, 4 °C. Remove the supernatant. 7. Follow steps 3–9 of the organoid establishment protocol.

3.4 Cryopreservation and Thawing of Organoids

1. Follow steps 2–6 of the organoid passaging procedure. 2. Add 1 mL of Bambanker freezing medium per cryotube that is going to be frozen.

Gastrointestinal Patient-Derived Organoids: Generation and Biobanking

129

3. Resuspend the sediment in the freezing medium and transfer 1 mL per cryovial. Store the cryotubes in a CoolCell (-1 °C/ min) for 24 h in a -80 °C freezer. 4. After that time, store the sample in a liquid N2 tank. 5. To thaw a cryotube of organoids, prepare previously a 15 mL Falcon tube with 5 mL of cold BM and defrost an aliquot of BME on ice. 6. Preheat a 24-well plate in the incubator at 37 °C. 7. Thaw the cryotube quickly and transfer its contents to the 15 mL Falcon tube containing 5 mL of cold BM. Mix by inversion. 8. Centrifuge at 500× g for 5 min, 4 °C. Remove supernatant. 9. Follow steps 3–9 of the organoid establishment protocol. 3.5 Drug Sensitivity Assay of Tumor Organoids

1. Before starting, preheat the 96-well plate in the CO2 incubator at 37 °C. 2. When organoid culture reaches high confluence, follow steps 2–6 of the organoid passaging procedure (Subheading 3.3), extending the trypsin time incubation to 10 min (see Note 26). 3. Resuspend the final cell pellet in 500 μL of cold BM and count the number of cells with Neubauer chamber (see Note 18). 4. Plate cells at a density of 3000 cells in 5 μL of 50% BME (diluted in BM) per well of 96-well plates. To do that, calculate the volume of cells according to the number of wells that will be plated. Transfer this volume to a 1.5 mL Eppendorf. 5. Remove the 96-well plate from the incubator just before plating. 6. Resuspend the cells in the needed volume of cold BM and add the same volume of 100% BME to reach the concentration of 50%. Quickly, plate 5 μL per well in the tempered 96-well plate, forming a dome, without introducing bubbles. 7. Incubate the plate for 15 min in the incubator (37 °C, 5% CO2) to allow the BME to solidify. 8. Once the BME has solidified, add 100 μL per well of appropriate 37 °C-tempered medium. In all cases RocKi Y-27632 must be added (see Note 5). 9. Incubate the plates in the cell culture incubator at 37 °C, 5% CO2 for 48 h to allow organoid formation. 10. Remove the medium from the wells by slightly tilting the plate and bringing the tips closer to the edge with the help of a multichannel pipette. Be careful not to touch the BME domes.

130

Manuel Cabeza-Segura et al.

11. Add 100 μL per well of 37 °C-tempered appropriate medium with a specific concentration of a selected drug. A vehicletreated condition and at least 5 increasing doses per drug need to be plated in triplicates (see Note 27). 12. Place the plates in the cell culture incubator at 37 °C, 5% CO2 for 120 h. 13. After this time, add the same volume per well of CellTiter Glo 3D® reagent (CTG) previously tempered at RT (see Note 28). Mix well by pipetting, avoiding bubble generation. 14. Incubate the plates for 25 min at RT in darkness. 15. Readout the plates with a luminescence scanner to measure cell viability.

4

Notes 1. To avoid the adhesion of cells or organoids to the inside of pipette tips, culture plates, and tube walls, nonstick plastic material should be used. 2. The incubation of collected tissue with PBS with a double concentrated solution of antibiotics (2X) is essential before processing the sample to prevent bacterial, yeast, mycoplasma, and fungi contamination of organoid culture. Subsequently, it is possible to diminish them to 1X concentration when the culture was established. Freshly prepared PBS or medium containing antibiotics is recommended. 3. If the sample is not going to be processed the same day, it can be stored in PBS + 1X antibiotics at 4 °C overnight. Although this could decrease efficiency in generating organoids, it is possible to establish them. 4. Liberase TL is a mixture of enzymes, collagenase isoforms I and II, and a neutral protease. To follow an appropriate reconstitution schedule and storage, is important to maintain enzyme stability. Reconstitute the entire vial of lyophilized enzyme at 13 U/mL with sterile water. Due to possible lot-to-lot differences in enzyme-specific activity, always use collagenase concentration in terms of U/mL instead of mg/mL. Store the stock solution in aliquots at -20 °C. Use an aliquot to prepare working concentration solution. 5. ROCKi (Rho kinase inhibitor) Y-27632 used at a final concentration of 10 μM enhances survival of single cell in suspension preventing dissociation-induced apoptosis [15]. 6. Sometimes, after filtering and centrifuging the digested tissue, it is possible to observe a red halo in the pellet due to the presence of red blood cells (RBCs). In this case a RBC lysis

Gastrointestinal Patient-Derived Organoids: Generation and Biobanking

131

buffer (1 mL) is added to the pellet and incubated up to 5 min at room temperature. After incubation add 5 mL of BM and centrifuge at 500× g for 5 min, 4 °C. 7. Filter sterilized 10 mL aliquots must be stored at -20 °C for up to 3 months. Once thawed, do not freeze again. 8. Store this product at -80 °C for long term. When needed, thaw the entire vial on ice and dispensed into working aliquots. BME aliquots can be stored at 4 °C up to one week or stored at -20 °C in a defrost freezer for a maximum of 3 months. It is very important to work always at lower temperature or on ice to avoid gelling of matrix. 9. Medium without WNT3A, mostly used with colon tumor samples and metastasis to select tumor cell growth. More than 90% of CRC acquire Wnt pathway independence growth throughout different genetic alterations [16]. 10. Mostly used with gastric tumor and healthy samples. 11. Some of the colon tumor samples grow better in this medium. Use only if no organoid generation is observed after one week in CTM medium. 12. Do not put the 24-well plates under the UV light in the culture hood. 13. Decanted PBS to a sterile 50 mL tube in case any tissue fragment should be recovered. 14. If the sample is not to be processed the same day, it can be stored in PBS +1X antibiotics at 4 °C until use. Not more than 24 h after collection is recommended to have a high efficiency. Freezing the tissue results in cell death, making organoid generation from it not possible. 15. It is advisable to cut quickly and move to enzymatic digestion. If the pieces are very large and do not fit on the P1000, cut the tip with the help of the scalpel. 16. The idea of pipetting is to help enzymatic digestion by mechanical disaggregation; therefore, if the pieces do not fit on the P1000 tip, cut the tip with the help of the scalpel. To reduce viscosity, DNase I (0.3 mg/mL) can be added to the tissue dissociation medium to digest DNA leaked as a result of cell damage. 17. The volume depends on the size of the starting tissue fragment and the subsequent tissue processing performance. Therefore, if the cell pellet after centrifugation is poorly seen, add 200 μL of BM to the sediment. On the other hand, add 500 μL if the sediment is seen and 1 mL if the sediment is huge. It is

132

Manuel Cabeza-Segura et al.

important to note that if the starting fragment is very small and the sediment is not visible at all, we do not count cells. We go directly to the plated step (Subheading 3.2, step 1; see Note 20). 18. Trypan blue stains dead cells because live cells have an intact cell membrane so do not count blue-stained cells. 19. Different cell densities and different concentrations of BME have been tested and these conditions have been established as optimal for the generation of organoids. 20. If the total number of cells counted is less than 200,000, pellet is not seen, or the starting fragment is very small, plate the whole cell pellet in a single well with a smaller amount of 50% BME (30 μL instead or 50 μL) to achieve high cell density. It can be plated on a 48-well plate and even further reduce the amount of BME: drops of 10–20 μL on 48-well plates. 21. For example, to plate 4 wells you need 200 μL of 50% BME (prepare 100 μL of BM + 100 μL of 100% BME). 22. Following the previous example, resuspend the cells in 100 μL of cold BM (stored on ice) and then add the 100 μL of 100% BME. It is important to keep the aliquot of BME always on ice, avoid heating the Eppendorf with the fingers, carefully pipette, and avoid blowing bubbles when resuspended. 23. It is important to add the medium slowly, letting it fall to the wall of the well carefully so as not to change the shape of the BME drop. 24. The time needed for each PDO culture to be expanded is very variable and must be evaluated with a follow-up every three days. Be careful not to pass the organoids too soon (at low confluence) as their growth may be severely affected and they may stop growing. 25. Depending on the size of the organoids, this time could be less than 5 min. The idea is to obtain fragments of organoids that could be reassembled as new organoids more than obtaining an isolated cell suspension after trypsin treatment. 26. Depending on the size of the organoids, this time could be increased up to 15 min. Bigger organoids will need more time to be completely dissociated until a single cell. 27. At least three independent experiments will be necessary to validate the obtained results. 28. CTG should be defrosted gradually, putting the reagent on ice or in the fridge overnight and tempered at RT at least 25 min before being added.

Gastrointestinal Patient-Derived Organoids: Generation and Biobanking

133

References 1. Drost J, Clevers H (2018) Organoids in cancer research. Nat Rev Cancer 18(7):407–418 2. Busslinger GA, Lissendorp F, Franken IA, van Hillegersberg R, Ruurda JP, Clevers H et al (2020) The potential and challenges of patient-derived organoids in guiding the multimodality treatment of upper gastrointestinal malignancies. Open Biol 10(4):190274 3. Ooft SN, Weeber F, Dijkstra KK, McLean CM, Kaing S, van Werkhoven E et al (2019) Patientderived organoids can predict response to chemotherapy in metastatic colorectal cancer patients. Sci Transl Med 11(513):eaay2574 4. de Witte CJ, Espejo Valle-Inclan J, Hami N, ˜ hmussaar K, Kopper O, Vreuls CPH et al Lo (2020) Patient-derived ovarian cancer organoids mimic clinical response and exhibit heterogeneous inter- and intrapatient drug responses. Cell Rep 31(11):107762 5. Bleijs M, van de Wetering M, Clevers H, Drost J (2019) Xenograft and organoid model systems in cancer research. EMBO J 38(15): e101654 6. Cabeza-Segura M, Garcia-Mico` B, La Noce M, Nicoletti GF, Conti V, Filippelli A et al (2023) How organoids can improve personalized treatment in patients with gastro-esophageal tumors. Curr Opin Pharmacol 69:102348 7. Sato T, Stange DE, Ferrante M, Vries RGJ, van Es JH, van den Brink S et al (2011) Long-term expansion of epithelial organoids from human colon, adenoma, adenocarcinoma, and Barrett’s epithelium. Gastroenterology 141(5): 1762–1772 8. van de Wetering M, Francies HE, Francis JM, Bounova G, Iorio F, Pronk A et al (2015) Prospective derivation of a living organoid biobank of colorectal cancer patients. Cell 161(4): 933–945 9. Bartfeld S, Bayram T, van de Wetering M, Huch M, Begthel H, Kujala P et al (2015) In

vitro expansion of human gastric epithelial stem cells and their responses to bacterial infection. Gastroenterology 148(1):126–136.e6 10. Seidlitz T, Merker SR, Rothe A, Zakrzewski F, von Neubeck C, Gru¨tzmann K et al (2019) Human gastric cancer modelling using organoids. Gut 68(2):207–217 11. Yan HHN, Siu HC, Law S, Ho SL, Yue SSK, Tsui WY et al (2018) A comprehensive human gastric cancer organoid biobank captures tumor subtype heterogeneity and enables therapeutic screening. Cell Stem Cell 23(6): 882–897.e11 12. Papaccio F, Garcı´a-Mico B, Gimeno-Valiente F, Cabeza-Segura M, Gambardella V, Gutie´rrezBravo MF et al (2023) Proteotranscriptomic analysis of advanced colorectal cancer patient derived organoids for drug sensitivity prediction. J Exp Clin Cancer Res 42(1):8 13. Pleguezuelos-Manzano C, Puschhof J, den Brink S, Geurts V, Beumer J, Clevers H (2020) Establishment and culture of human intestinal organoids derived from adult stem cells. Curr Protoc Immunol 130(1) 14. Broutier L, Andersson-Rolf A, Hindley CJ, Boj SF, Clevers H, Koo BK et al (2016) Culture and establishment of self-renewing human and mouse adult liver and pancreas 3D organoids and their genetic manipulation. Nat Protoc 11(9):1724–1743 15. Watanabe K, Ueno M, Kamiya D, Nishiyama A, Matsumura M, Wataya T et al (2007) A ROCK inhibitor permits survival of dissociated human embryonic stem cells. Nat Biotechnol 25(6):681–686 16. The Cancer Genome Atlas Network (2012) Comprehensive molecular characterization of human colon and rectal cancer. Nature 487(7407):330–337

Chapter 10 Prostate Cancer Organoids for Tumor Modeling and Drug Screening Amani Yehya, Fatima Ghamlouche, Sana Hachem, and Wassim Abou-Kheir Abstract Prostate cancer (PCa) is the second most common malignancy and the fifth leading cause of cancer death in men worldwide. Despite its prevalence, the highly heterogenic PCa has shown difficulty to establish representative cell lines that reflect the diverse phenotypes and different stages of the disease in vitro and hence hard to model in preclinical research. The patient-derived organoid (PDO) technique has emerged as a groundbreaking three-dimensional (3D) tumor modeling platform in cancer research. This versatile assay relies on the unique ability of cancer stem cells (CSCs) to self-organize and differentiate into organ-like mini structures. The PDO culture system allows for the long-term maintenance of cancer cells derived from patient tumor tissues. Moreover, it recapitulates the parental tumor features and serves as a superior preclinical model for in vitro tumor representation and personalized drug screening. Henceforth, PDOs hold great promise in precision medicine for cancer. Herein, we describe the detailed protocol to establish and propagate organoids derived from isolated cell suspensions of PCa patient tissues or cell lines using the 3D semisolid Matrigel™-based hanging-drop method. In addition, we highlight the relevance of PDOs as a tool for evaluating drug efficacy and predicting tumor response in PCa patients. Key words Prostate cancer, Patient-derived organoids, Three-dimensional, Preclinical model, Matrigel™-based hanging-drop method, Drug screening

1

Introduction Despite the significant advances in prostate cancer (PCa) research and the considerable efforts that have been made to elucidate new therapeutic strategies, PCa still constitutes an alarming public health problem among men globally [1]. Much of the complexity of PCa management is rooted in the multifocal nature and pervasive heterogeneity of the disease [2]. A compelling body of evidence suggests that the unique subpopulation of cancer stem cells (CSCs)

Amani Yehya and Fatima Ghamlouche contributed equally with all other contributors. Federica Papaccio and Gianpaolo Papaccio (eds.), Cancer Stem Cells: Methods and Protocols, Methods in Molecular Biology, vol. 2777, https://doi.org/10.1007/978-1-0716-3730-2_10, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

135

136

Amani Yehya et al.

present within PCa tumors is chiefly involved in the maintenance of heterogeneity [3–5]. These cells are characterized by self-renewal and differentiation abilities that allow them to generate cancerous cells of multiple lineages. Accordingly, CSCs play a main role in treatment failure and disease relapse as they can reconstitute the tumor bulk and re-establish its phenotypic diversity [6, 7]. One major hurdle in the development of effective and personalized therapeutic modalities for PCa patients is that the currently available in vitro and in vivo preclinical models do not adequately reflect the heterogeneity of the patient tumors [8, 9]. Indeed, several anticancer treatments that have been shown to be effective in in vitro cancer models fail to achieve the same desired outcomes in clinics. This is due to the structural and functional discrepancies found between cancer patients and the in vitro cancer systems [9– 11]. Thus, it is important to rely on new preclinical cancer models to bridge the gap between the traditional two-dimensional (2D) cell culture and the in vivo models, paving the way to effectively translate the results of basic research into the clinical level. Patient-derived organoid (PDO) is a revolutionary cancer modeling technology that relies on the self-renewal ability of CSCs to generate a miniature collection of self-organized cells and differentiate into three-dimensional (3D) organ-like structures. Organoid technology was first established by Sato et al. through the isolation of Lrg5+ stem cells from intestinal crypts that were then expanded ex vivo and led to the formation of crypt–villus organoids [12]. Later, they could generate the first PDO model from colorectal cancer tissues [13]. Following that, PDOs of other cancer types were established [14–20]. The organoid formation assay is considered a reliable technique to recreate mini tumors in a dish. These PDOs can mimic the intrinsic pathohistological and pathophysiological properties of the parental tumor [21] and, hence, provide a more accurate preclinical cancer model as they faithfully reflect the patient diversity and cellular heterogeneity of the tumors [22, 23]. In addition, PDOs can be sustained in culture for long periods without the risk of acquiring genetic mutations. They also allow for multidrug screening due to their ability to generate many clones in short time intervals [24]. Thus, PDOs constitute a promising tool for personalized drug screening and precision therapy [25, 26]. Herein, we report the detailed approach for generating/propagating PCa cell line- and PDOs using the semisolid Matrigel™-based hanging-drop method. In addition, we detail the preparation and care required for cell plating in hanging drop and maintenance through the feeding and image tracking of organoid formation. Finally, the use of organoids for quantifying drug sensitivity response is described. We have chosen PCa as the model system for these demonstrations; however, this approach can be applied to other tumor types with minor modifications.

Prostate Cancer Organoids for Tumor Modeling and Drug Screening

2

137

Materials

2.1

Reagents

1. Rho-associated coiled-coil-containing protein kinase (ROCK) inhibitor Y-27632. 2. Dulbecco’s PBS, without calcium and magnesium (D-PBS). 3. 0.05% trypsin/EDTA. 4. TrypLE™. 5. Dispase. 6. Ammonium–chloride–potassium (ACK) lysing buffer. 7. 0.4% trypan blue solution. 8. Growth factor-reduced Matrigel™. 9. Culturing media: Advanced DMEM/F-12 (Dulbecco’s Modified Eagle Medium/Ham’s F-12) supplemented with 5 μg/ mL Plasmocin® prophylactic, 1X penicillin–streptomycin, 1X nonessential amino acid solution, and 1 mM sodium pyruvate. 10. Growth media: culturing media supplemented with a mixture of growth factors (see Table 1). 11. Dissociation media: culturing media supplemented with 5 mg/ mL collagenase type II and 10 μM ROCK inhibitor Y-27632. 12. Complete media: culturing media with 10% fetal bovine serum (FBS).

2.2

Equipment

1. 15 mL conical tubes. 2. 50 mL conical tubes. 3. 1.5 mL microtubes. 4. 100-mm tissue culture dish. 5. 24-well culture plate. 6. Scalpel and forceps.

Table 1 Essential components and concentrations of human prostate growth medium Component

Stock concentration

Solvent

Final concentration

B27

50.0X

–

1.00X

Nicotinamide

1.0 M

PBS

10.00 mM

NAC

500.0 mM

PBS

1.25 mM

A83–01

5.0 mM

DMSO

500.00 nM

NOG

100.0 μg/ml

PBS + 0.1% BSA

50.00 ng/ml

Y-27632

10.0 mM

PBS + 0.1% BSA

10.00 μM

Modified from Cheaito et al. Abbreviations: NAC N-acetylcysteine, NOG noggin

138

Amani Yehya et al.

7. Syringes and needles. 8. 70 μm and 40 μm cell strainers. 9. Bright-Line™ Hemacytometer. 10. Ice container. 11. Micropipettes. 12. Pipette aid. 13. Vortex mixer. 14. Inverted microscope. 15. Centrifuge machine. 16. Water bath set at 37  C. 17. Biosafety cabinet. 18. Cell culture humidified incubator set at 37  C, supplied with 5% CO2.

3

Methods The following procedures should be performed in a primary cell culture laboratory under a biosafety cabinet. All reagents and equipment should be sterile to minimize/prevent the risk of cell culture contamination.

3.1 Isolation of Single-Cell Suspensions from Prostate Tumor Tissues (See Fig. 1)

1. Put the freshly resected human prostate tissues directly in a 15 mL conical tube containing culturing media (see Notes 1 and 2). 2. Transfer the tissue into a 100 mm culture dish and mince it mechanically using sterile scalpel blades and forceps into approximately 0.1–0.5 mm diameter pieces (see Note 3). 3. Digest the tissue fragments in dissociation media in a 15 mL conical tube overnight in a humidified incubator containing 5% CO2 at 37  C. 4. After incubation, wash by adding culturing media and spin down at 200  g for 5 min at 4  C. Aspirate the supernatant. 5. Add 2 mL TrypLE™ with 10 μM ROCK inhibitor Y-27632 and incubate for ~15 min at 37  C (see Note 4). Inactivate the enzyme by adding complete media. 6. Optional: To further dissociate the tissue lysate, pass it gently through a series of needles with different gauges until a clear single-cell suspension can be observed under the microscope. 7. Pass the digested tissue suspension through a sterile 70 μm cell strainer and collect the filtrate into a new sterile 50 mL conical tube. Strain flow-through again with a 40-μm cell strainer.

Prostate Cancer Organoids for Tumor Modeling and Drug Screening

139

Fig. 1 Schematic diagram representing the approach for (a) generating, (b) propagating, and treating patientderived organoids (PDOs) using the semisolid Matrigel™-based hanging-drop method. Abbreviations: RI Rho-associated coiled-coil-containing protein kinase (ROCK) inhibitor Y-27632, G generation, D day, PDOs patient-derived organoids, MG Matrigel, EDTA ethylenediaminetetraacetic acid, OFC organoid formation count

8. Centrifuge the filtrate at 200  g for 5 min at 4  C and aspirate the supernatant. 9. Optional: If using samples containing blood, lyse the red blood cells using the ACK lysis buffer following the manufacturer’s protocol. 10. Resuspend the cells in serum-free culturing media and quantify the number of viable cells on a hemocytometer using the trypan blue dye. 3.2 Preparation of Cells from Prostate Cancer Cell Lines

1. Aspirate the growth media from a 100 mm dish of adherent monolayer prostate tumor cell lines. 2. Wash once with pre-warmed D-PBS and aspirate. 3. Detach the cells by adding 0.05% trypsin–EDTA solution and incubate at 37  C for at least 3 min. Inactivate the trypsin by adding complete media. 4. Collect the cell suspension and centrifuge at 1500  g for 5 min at room temperature.

140

Amani Yehya et al.

5. Resuspend the cells in serum-free culturing media and quantify the number of viable cells on a hemocytometer using the trypan blue dye. 3.3 Seeding Cells for Organoid Formation Assay (See Fig. 1)

1. Prepare a master mix at a ratio of 20,000 of primary PCa patient tumor cells or 5000 of PCa cell lines/40 μL of 90% Matrigel™/each well of a 24-well plate (see Notes 5 and 6). 2. Plate the cells in a drop in the middle of each well (see Note 7). 3. Invert the plate upside down and place it at 37  C for 30–45 min for the Matrigel™ to solidify (see Note 8). 4. Gently add 500 μL of pre-warmed growth media with or without treatment to each well (see Subheading 3.5 steps 1 and 2 and Note 9). Replenish every 2–3 days (see Note 10). 5. Assess the organoid-forming efficiency over time by visualizing and counting the formed organoids (see Subheading 3.5 step 3 and Note 11).

3.4 Propagation of Prostate Tumor Organoids (See Fig. 1)

1. Aspirate the medium from the wells and incubate with 1 mg/ mL of dispase dissolved in culturing media for a minimum of 30 min in a humidified incubator at 37  C (see Note 12). 2. Collect the organoids from each well/condition separately into a 15 ml tube and centrifuge at 200  g for 5 min at 4  C. Discard the supernatant. 3. Resuspend the pellet in 0.05% trypsin–EDTA and incubate at 37  C for ~10 min. Inactivate the trypsin by adding complete media. 4. Start the mechanical dissociation process by passing them through a series of needles with different gauges to obtain single-cell suspension. 5. Centrifuge the cells at 200  g for 5 min at 4  C and aspirate the supernatant. 6. Resuspend the cells in serum-free culturing media and quantify the number of viable cells on a hemocytometer using the trypan blue dye. 7. Repeat the cell seeding as aforementioned in Subheading 3.3 to get a minimum of 5 generations of organoids (G5).

3.5 Treatment of Prostate Tumor Organoids

1. For treatment on day 0: prepare the different drug concentrations diluted in growth media. Add the prepared drugs gently to each corresponding well. Example: G2 of primary PCa-derived organoids from patient 1 were treated with 0.05 nM of docetaxel (Doc) which is a commonly used chemotherapeutic to manage PCa (see Fig. 2).

Prostate Cancer Organoids for Tumor Modeling and Drug Screening

141

Fig. 2 Treatment response of chemotherapy on G2 patient-derived prostate cancer organoid growth. (a) Representative bright-field images of G2 prostate cancer organoids grown in the presence or absence of 0.05 nM docetaxel (Doc). Scale bar, 100 μm. (b) OFC for patient 1. Data are presented as the mean  SD of duplicate wells. (c) Average diameter of 75 organoids from patient 1. Data are presented as the mean  SEM of duplicate wells. Abbreviations: CTRL control, Doc docetaxel, G2 generation 2, OFC organoid formation count

2. For treatment on day 1: Add growth media to all the wells after Matrigel™ solidification and leave the plate in a humidified incubator at 37  C. After 24 h, gently aspirate the media and treat (see Note 13). Example: G2 of primary PCa-derived organoids from patient 2 were subjected to different irradiation doses including 2 Gy, the clinically achievable dose, on day 1 after cell seeding (see Fig. 3). 3. Assess the effect of the treatment(s) on the number, shape, and size of the organoids formed compared to the control. Manually count the total number of organoids in each well under bright-field light microscopy (20X objective). Then, determine the organoid formation count (OFC) by calculating the average number of formed organoids from the duplicate wells per condition. Assess the size and shape of the organoids and calculate the mean diameter of at least 30 organoids using Carl Zeiss AG ZEN 2013 image software.

142

Amani Yehya et al.

Fig. 3 Treatment response to different doses of irradiation of G2 patient-derived prostate cancer organoid growth. (a) Representative bright-field images of G2 prostate cancer organoids. Scale bar, 100 μm. (b) OFC for patient 2. Data are presented as the mean  SD of duplicate wells. (c) Average diameter of 30 organoids from patient 2. Data are presented as the mean  SEM of duplicate wells. Abbreviations: CTRL control, Gy gray (unit of ionizing radiation dose), G2 generation 2, OFC organoid formation count

Example: The drug response of G2 prostate organoids derived from two patients treated with different regimens, Doc and irradiation, is illustrated in Figs. 2 and 3. Doc slightly reduced the count and size of PDOs derived from patient 1 at 0.05 nM concentration (see Fig. 2). As for patient 2, increasing doses of irradiation from 1 to 4 Gy potently reduced the OFC as well as the organoids’ diameter (see Fig. 3).

4

Notes 1. The prostate tumor tissues used in this protocol are obtained with informed consent from distinct stages of human prostate adenocarcinoma from treatment-naı¨ve patients undergoing radical prostatectomy at the American University of Beirut– Medical Center (AUB–MC, Beirut, Lebanon). All protocols are performed in accordance with the Code of Ethics of the World Medical Association (Declaration of Helsinki) for experiments involving human subjects. 2. A sample is taken from each patient from the most likely affected area (from the core of the cancerous lesion) and another one from the unaffected area (far from the tumor

Prostate Cancer Organoids for Tumor Modeling and Drug Screening

143

site) according to the urologist’s and pathologist’s recommendation. Tumor samples are kept at 4  C until processed within 24 h of surgical resection for maximal cell viability and reliability of organoid generation. 3. Part of the fragments are used for the organoid formation assay and the rest are used for protein/RNA/DNA extraction or histology embedding. 4. Pipetting/vortexing the tissue suspension every 15 min ensures efficient digestion and further dissociates the tissue to obtain single-cell suspensions. 5. A 1:9 ratio of cell suspension is prepared in serum-free culturing media to Matrigel™. Keep the master mix on ice and work quickly to prevent the solidification of the Matrigel™. 6. Based on the conditions to be tested, calculate the number of wells and accordingly the total number of cells needed. Take into consideration that each condition should be seeded at least in duplicate. 7. Avoid forming bubbles as they create gaps in the Matrigel™ and lead to its breakage. To get rid of a bubble, quickly prick it with the needle of a 27-gauge syringe. A uniform and bubblefree plating can be confirmed under light microscopy. 8. The plate is placed upside down to prevent the adherence of the cells to the plate bottom. 9. The ROCK inhibitor Y-27632 at 10 μM is added to the growth media strictly on the first week post-seeding. 10. The day of the seeding whether from the dissociated tissue or upon propagation is considered as day 0. 11. Organoids are generated after 18 to 22 days or more depending on the sample. The organoids that are formed in the primary organoid formation assay are referred to as generation 1 (G1) organoids. 12. Dispase breaks down the Matrigel™ and releases the organoids into the media. 13. Using this assay, several different treatment regimens may be tested including radiation therapy. To do so, subject the plate to irradiation at the desired dose(s) on day 1 post-plating. References 1. Siegel RL, Miller KD, Wagle NS, Jemal A (2023) Cancer statistics, 2023. CA Cancer J Clin 73(1):17–48 2. Haffner MC, Zwart W, Roudier MP, True LD, Nelson WG, Epstein JI et al (2021) Genomic

and phenotypic heterogeneity in prostate cancer. Nat Rev Urol 18(2):79–92. https://doi. org/10.1038/s41585-020-00400-w 3. Batlle E, Clevers H (2017) Cancer stem cells revisited. Nat Med 23(10):1124–1134

144

Amani Yehya et al.

4. Kreso A, Dick JE (2014) Evolution of the cancer stem cell model. Cell Stem Cell 14(3): 275–291 5. Najafi M, Mortezaee K, Ahadi R (2019) Cancer stem cell (a) symmetry & plasticity: tumorigenesis and therapy relevance. Life Sci 231: 116520 6. Atashzar MR, Baharlou R, Karami J, Abdollahi H, Rezaei R, Pourramezan F et al (2020) Cancer stem cells: a review from origin to therapeutic implications. J Cell Physiol 235(2):790–803 7. Escudero-Lourdes C, Alvarado-Morales I, Tokar EJ (2022) Stem cells as target for prostate cancer therapy: opportunities and challenges. Stem Cell Rev Rep 18(8):2833–2851 8. Kamb A (2005) What’s wrong with our cancer models? Nat Rev Drug Discov 4(2):161–165 9. Namekawa T, Ikeda K, Horie-Inoue K, Inoue S (2019) Application of prostate cancer models for preclinical study: advantages and limitations of cell lines, patient-derived xenografts, and three-dimensional culture of patient-derived cells. Cell 8(1):74 10. Antunes N, Kundu B, Kundu SC, Reis RL, Correlo V (2022) In vitro cancer models: a closer look at limitations on translation. Bioengineering 9(4):166 11. Weeber F, Ooft SN, Dijkstra KK, Voest EE (2017) Tumor organoids as a pre-clinical cancer model for drug discovery. Cell Chem Biol 24(9):1092–1100 12. Sato T, Vries RG, Snippert HJ, Van De Wetering M, Barker N, Stange DE et al (2009) Single Lgr5 stem cells build crypt-villus structures in vitro without a mesenchymal niche. Nature 459(7244):262–265 13. Sato T, Stange DE, Ferrante M, Vries RG, Van Es JH, Van den Brink S et al (2011) Long-term expansion of epithelial organoids from human colon, adenoma, adenocarcinoma, and Barrett’s epithelium. Gastroenterology 141(5): 1762–1772 14. Cheaito K, Bahmad HF, Hadadeh O, Msheik H, Monzer A, Ballout F et al (2022) Establishment and characterization of prostate organoids from treatment-naı¨ve patients with prostate cancer. Oncol Lett 23(1):1–16 15. Mazzucchelli S, Piccotti F, Allevi R, Truffi M, Sorrentino L, Russo L et al (2019)

Establishment and morphological characterization of patient-derived organoids from breast cancer. Biol Proced Online 21:1–13 16. Kim M, Mun H, Sung CO, Cho EJ, Jeon H-J, Chun S-M et al (2019) Patient-derived lung cancer organoids as in vitro cancer models for therapeutic screening. Nat Commun 10(1): 3991 17. Aberle MR, Burkhart RA, Tiriac H, Olde Damink S, Dejong CH, Tuveson DA et al (2018) Patient-derived organoid models help define personalized management of gastrointestinal cancer. J Br Surg 105(2):e48–e60 18. Demyan L, Habowski AN, Plenker D, King DA, Standring OJ, Tsang C et al (2022) Pancreatic cancer patient-derived organoids can predict response to neoadjuvant chemotherapy. Ann Surg 276(3):450–462 19. Yakoub AM, Sadek M (2018) Development and characterization of human cerebral organoids: an optimized protocol. Cell Transplant 27(3):393–406 20. Monzer A, Wakimian K, Ballout F, Al Bitar S, Yehya A, Kanso M et al (2022) Novel therapeutic diiminoquinone exhibits anticancer effects on human colorectal cancer cells in two-dimensional and three-dimensional in vitro models. World J Gastroenterol 28(33):4787–4811 21. Rae C, Amato F, Braconi C (2021) Patientderived organoids as a model for cancer drug discovery. Int J Mol Sci 22(7):3483 22. Driehuis E, Kretzschmar K, Clevers H (2020) Establishment of patient-derived cancer organoids for drug-screening applications. Nat Protoc 15(10):3380–3409 23. Xu H, Lyu X, Yi M, Zhao W, Song Y, Wu K (2018) Organoid technology and applications in cancer research. J Hematol Oncol 11(1): 1–15 24. Shiihara M, Furukawa T (2022) Application of patient-derived cancer organoids to personalized medicine. J Pers Med 12(5):789 25. Takahashi T (2019) Organoids for drug discovery and personalized medicine. Annu Rev Pharmacol Toxicol 59:447–462 26. Yang H, Wang Y, Wang P, Zhang N, Wang P (2021) Tumor organoids for cancer research and personalized medicine. Cancer Biol Med 19(3):319–332

Chapter 11 Mimicking the Tumor Niche: Methods for Isolation, Culture, and Characterization of Cancer Stem Cells and Multicellular Spheroids Laura De Lara-Pen˜a, Cristiano Farace, Andrea Pisano, Julia Lo´pez de Andre´s, Grazia Fenu, Federica Etzi, Carmen Grin˜a´n-Liso´n, Juan Antonio Marchal, and Roberto Madeddu Abstract Cancer stem cells (CSCs) play a significant role in driving several tumor hallmarks. Their behavior and tumor progression are strictly related to the tumor microenvironment (TME). The dynamic interplay between CSCs and TME drives metastasis, chemoresistance, and disease relapse. In this chapter, we describe different techniques and protocols for isolating, culturing, and characterizing CSCs and we explain the methodology for the culture of multicellular spheroids comprising CSCs. Key words Tumour, Cancer stem cells (CSCs), CSC niche, 3D cultures, Tumorspheres, CSC characterization, CSC markers

1

Introduction Cancer stem cells (CSCs) play a significant role in tumor metastasis, therapy resistance, and tumor relapse. They exhibit stem cell-like characteristics, including self-renewal and differentiation capacities. CSCs have been identified in various cancer types and are associated with genomic instability and the development of tumor heterogeneity [1–6]. The tumor microenvironment (TME) plays a crucial role in the behavior of CSC behavior and tumor progression. CSCs interact with the TME, which comprises diverse cell types and signaling molecules. Furthermore, CSCs shape the TME through the secretion of factors that modulate angiogenesis, recruit stromal cells, and remodel the extracellular matrix (ECM) [6–10]. This dynamic interplay between CSCs and the TME drives metastasis, therapy resistance, and disease relapse [6–8]. Targeting the communication networks between CSCs and the TME presents

Federica Papaccio and Gianpaolo Papaccio (eds.), Cancer Stem Cells: Methods and Protocols, Methods in Molecular Biology, vol. 2777, https://doi.org/10.1007/978-1-0716-3730-2_11, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

145

146

Laura De Lara-Pen˜a et al.

promising therapeutic opportunities to disrupt CSC functions and improve treatment outcomes [4]. Understanding this intricate relationship guides the development of innovative strategies that effectively target CSCs within their complex microenvironment [11]. In this chapter, we describe different techniques and protocols for isolating, culturing, and enriching CSCs and spheroids [12– 15]. 1.1 Properties and Characterization of CSCs

Characterization of CSCs to obtain a well-defined population of cells with known phenotypes [12], through the use of intracellular [13] and specific surface markers depending on the cancer type or cell line [13–15], as well as the soft agar assay for colony formation and the side population assay [15, 16]. We explain these techniques and assays in the methods section.

1.2 Interplay Between TME and CSCs

In recent years, researchers have been striving to refine in vitro study models, attempting to closely resemble the real tumor contexts. These models provide valuable tools for conducting translational experiments in personalized oncology and high-throughput screening of novel antitumor agents. We explain the methodology for culturing multicellular spheroids in Subheading 3.5. These spheroids are composed of tumor cells, fibroblasts (FBs), and mesenchymal stem cells (MSCs) [10, 17]. By utilizing in vitro models, researchers can recreate the complex cellular interactions found in tumors and investigate the influence of the TME on cancer cell behavior [18].

2

Materials

2.1 CSC Enrichment from Cell Lines: Primary and Secondary Tumorsphere Generation

1. Supplemented DMEM: Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% of fetal bovine serum (FBS) and 1% of penicillin–streptomycin, or other appropriate medium for the cell line used in CSC enrichment. 2. CSC medium: DMEM-F12 supplemented with 1x B-27, 4 ng/mL heparin (heparin sodium cell culture tested), 10 μg/mL insulin (insulin–transferrin–selenium), 1 μg/mL hydrocortisone, 10 ng/mL epidermal growth factor (EGF), 10 ng/mL fibroblast growth factor (FGF), 10 ng/mL interleukin 6 (IL-6), and 10 ng/mL hepatocellular growth factor (HGF). Basal CSC medium is stable for a month at 4 °C; growth factors and cytokines (EGF, FGF, IL-6, and HGF) should be added to the medium just before the cell seeding. 3. T75 flasks and 6-well and 24-well ultralow-attachment plates for cell culture. 4. 10 mL pipettes and Pasteur pipettes.

Mimicking the Tumor Niche: Methods for Isolation, Culture, and. . .

147

5. 1x phosphate-buffered saline (PBS). 6. Trypsin–EDTA solution: 0.53 mM EDTA.

0.25%

w/v

trypsin

and

7. 15 mL conical tubes. 8. Automated cell counter or counting chamber. 9. Optical microscope and dissecting microscope. 10. Incubator at 37 °C and 5% CO2. 11. Laminar flow cabin. 12. Water bath at 40 °C. 13. 1 mL syringe with 0.5 × 16 mm needle. 2.2 Soft Agar Colony Formation Assay

1. Supplemented DMEM: Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% of fetal bovine serum (FBS) and 1% of penicillin–streptomycin, or other appropriate medium for the cell line used in CSC enrichment. 2. CSCs medium: DMEM-F12 supplemented with 1x B-27, 4 ng/mL heparin (heparin sodium cell culture tested), 10 μg/mL insulin (insulin–transferrin–selenium), 1 μg/mL hydrocortisone, 10 ng/mL epidermal growth factor (EGF), 10 ng/mL fibroblast growth factor (FGF), 10 ng/mL interleukin 6 (IL-6), and 10 ng/mL hepatocellular growth factor (HGF). Basal CSC medium is stable for a month at 4 °C, growth factors, and cytokines (EGF, FGF, IL-6, and HGF) should be added to the medium just before the cell seeding. 3. T75 flasks and 6-well and 24-well ultralow-attachment plates for cell culture. 4. 10 mL pipettes and Pasteur pipettes. 5. 1x phosphate-buffered saline (PBS). 6. Trypsin–EDTA solution: 0.53 mM EDTA.

0.25%

w/v

trypsin

and

7. 15 mL conical tubes. 8. Automated cell counter or counting chamber. 9. Optical microscope and dissecting microscope. 10. Incubator at 37 °C and 5% CO2. 11. Laminar flow cabin. 12. Microwave. 13. Water bath at 40 °C. 14. 1 mL syringe with 0.5 × 16 mm needle. 15. 1.6% ultrapure agar solution: 1.6 g of ultrapure agar dissolved in 100 mL of 1x PBS. 16. Iodonitrotetrazolium chloride solution: 10g of iodonitrotetrazolium dissolved in 10 mL of sterile H2O.

148

Laura De Lara-Pen˜a et al.

2.3 CSC Characterization by Flow Cytometry 2.3.1 Hoechst Exclusion Assay

1. Supplemented DMEM: Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% of fetal bovine serum (FBS) and 1% of penicillin–streptomycin, or other appropriate medium for the cell line used in CSC enrichment. 2. CSC medium: DMEM-F12 supplemented with 1x B-27, 4 ng/mL heparin (heparin sodium cell culture tested), 10 μg/mL insulin (insulin–transferrin–selenium), 1 μg/mL hydrocortisone, 10 ng/mL epidermal growth factor (EGF), 10 ng/mL fibroblast growth factor (FGF), 10 ng/mL interleukin 6 (IL-6), and 10 ng/mL hepatocellular growth factor (HGF). CSC medium is stable for a month at 4 °C; growth factors and cytokines (EGF, FGF, IL-6 and HGF) should be added to the medium just before the cell seeding. 3. T75 flasks and 6-well and 24-well ultralow-attachment plates for cell culture. 4. 10 mL pipettes and Pasteur pipettes. 5. 1x phosphate-buffered saline (PBS) 6. 15 mL conical tubes. 7. 1.5 mL Eppendorf tubes. 8. Automated cell counter or counting chamber. 9. Optical microscope and dissecting microscope. 10. Incubator at 37 °C and 5% CO2. 11. Laminar flow cabin. 12. Water bath at 40 °C. 13. 1 mL syringe with 0.5 × 16 mm needle. 14. Hoechst 33342 to a final concentration of 5 μg/mL. 15. 5 μM verapamil. 16. Flow cytometry instrument.

2.3.2 CSC Marker Evaluation

1. T75 flasks and 6-well and 24-well ultralow-attachment plates for cell culture. 2. 10 mL pipettes and Pasteur pipettes. 3. 1x phosphate-buffered saline (PBS). 4. Trypsin–EDTA solution: 0.53mM EDTA.

0.25%

w/v

trypsin

and

5. Permeabilization buffer: 0.01% saponin or other commercially available. 6. Blocking buffer: 1% bovine serum albumin (BSA) in 1x PBS. 7. Staining buffer with appropriate set of markers. 8. 15 mL conical tubes.

Mimicking the Tumor Niche: Methods for Isolation, Culture, and. . .

149

9. Automated cell counter or counting chamber. 10. Optical microscope and dissecting microscope. 11. Incubator at 37 °C and 5% CO2. 12. Laminar flow cabin. 13. Water bath at 40 °C. 14. Flow cytometry instrument 2.4 In Vivo Tumorigenic Properties of CSCs (Orthotopic Limiting Dilution Assay)

1. To establish xenograft tumors, six- to eight-week-old male NOD scid gamma mice (NOD.Cg-Prkdcscid Il2rgtm1Wjl/ SzJ, NSG) were used. They were housed and maintained at 20–24 °C, 50% RH, a light–dark cycle of 14–10 h with food and water ad libitum. 2. 26-gauge needles. 3. Isoflurane. 4. Corning Matrigel matrix. 5. CSC medium.

2.5 Multicellular Spheroid Production

1. CellTracker solution: Add 20 μL of DMSO per 50 μg of CellTracker stock directly to the vial and mix well by pipetting. Solution is stable for 12 months at -20 °C. 2. Trypsin–EDTA solution: 0.53mM EDTA.

0.25%

w/v

trypsin

and

3. Supplemented DMEM: Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% of fetal bovine serum (FBS) and 1% of penicillin–streptomycin. 4. Methylcellulose stock: To prepare 50 mL of methylcellulose stock, weigh 600 mg of methylcellulose in an autoclave bottle with a stirrer bar inside for further dissolution. Autoclave methylcellulose. Heat 50 mL of supplemented DMEM in a water bath at 60 °C (higher or lower temperature may precipitate some components of the medium). Once the temperature is reached, first add 25 mL of hot medium to the methylcellulose and stir at 60 °C for 10–15 min until dissolved. Once dissolved, add the remaining 25 mL of medium. Leave in stirring O/N at 4 °C. The next day, centrifuge the methylcellulose stock at 3000 g for 1 h 30 min. Carefully collect 45 mL of supernatant (in order not to recover the remains deposited at the bottom of the tube). Methylcellulose stock is stable for 3 months at 4 °C. 5. Methylcellulose medium: Prepare the methylcellulose medium in which the spheroids will be seeded at a ratio of 20% methylcellulose stock and 80% supplemented DMEM.

150

3

Laura De Lara-Pen˜a et al.

Methods

3.1 CSC Enrichment from Cell Lines: Primary and Secondary Tumorsphere Generation

The initial steps for establishing primary tumorspheres are similar to a regular cell passage, but during the final steps, it is necessary to resuspend the cells in the CSC medium and seed them onto low-attachment plates. Primary tumorspheres are established 72 h after the seeding. The procedure involves dissociating the primary tumorspheres and initiating the culture again. The resulting secondary tumorspheres will exhibit enhanced stemness properties and can be used for experimental purposes. 1. Growth the cells in adherent conditions at 37 °C and 5% CO2 using an appropriate medium for the cell line (i.e., supplemented DMEM). 2. At 80% of confluence, aspirate the medium using a sterile 10 mL pipette. 3. Add 5 mL of 1x PBS and wash the cells. 4. Add 4 mL of trypsin–EDTA solution to detach the cells and incubate for 5 min at 37 °C. 5. Once the cells are detached, add 4 mL of medium for trypsin inactivation. 6. Add 6 mL of 1x PBS and collect cells into a 15 mL conical tube. 7. Centrifuge at 300 g for 5 min and remove the supernatant. 8. Resuspend the pellet with 10 mL of 1x PBS. 9. Repeat steps 7 and 8. 10. Resuspend the pellet in an appropriate volume of 1x PBS based on the size of the pellet. 11. Perform cell counting. 12. Add the desired volume of CSC medium to the cells, depending on the desired cell number per flask or well. For each well of 6-well ultralow-attachment plates, seed 2 mL of cell suspension. 13. Incubate at 37 °C for 72 h. 14. Observe the primary tumorspheres at the optic microscope. 15. Transfer the primary tumorspheres using a Pasteur pipette to a new 15 mL conical tube filled with 10 mL of 1x PBS. 16. Centrifuge at 300 g for 5 min and remove the supernatant. 17. Resuspend the tumorsphere pellet with 1 mL of 1x PBS. 18. Disaggregate the primary tumorspheres using a 1 mL syringe with 0.5 × 16 mm needle by aspirating and releasing the solution approximately 30 times.

Mimicking the Tumor Niche: Methods for Isolation, Culture, and. . .

151

19. Centrifuge at 300 g for 5 min and remove the supernatant. 20. Repeat steps 10–13 to obtain CSC-enriched secondary tumorspheres (see Note 1). 3.2 Soft Agar Colony Formation Assay

1. Melt 10 mL of 1.6% ultrapure agar in a microwave and cool to 40 °C using a water bath. 2. Warm 10 mL of supplemented DMEM to 40 °C in a water bath for 30 min. 3. For the base layer, prepare a 0.8% agar solution by mixing equal parts of 1.6% agar solution and supplemented DMEM (or other appropriate medium used for the cell lines). 4. Dispense 500 μL of 0,8% agar solution into each well of a 24-well ultralow-attachment plate avoiding bubble formation on the surface and allow it to solidify at RT (12 mL total). 5. Transfer the secondary tumorspheres using a Pasteur pipette to a new 15 mL conical tube filled with 10 mL of 1x PBS. 6. Centrifuge at 300 g for 5 min and remove the supernatant. 7. Resuspend the pellet of secondary tumorspheres with 1 mL of 1x PBS. 8. Disaggregate the tumorspheres using a 1 mL syringe with 0.5 × 16 mm needle by aspirating and releasing the solution approximately 30 times. 9. Resuspend the cells in an appropriate volume of 1x PBS based on the size of the pellet. 10. Perform cell counting. 11. Centrifuge at 300 g for 5 min and remove the supernatant. 12. Resuspend the cell pellet with appropriate volume of DMEM. 13. For the top layer solution, prepare a 0.4% agar solution by mixing 1:4 of 1.6% agar and supplemented DMEM and add the cells at a concentration of 20,000 cells/mL. 14. Dispense 500μl of top layer solution containing 10,000 cells into each well of a 24-well ultralow-attachment plate avoiding bubble formation on the surface, and allow it to solidify at RT. 15. Add 500 μL of supplemented DMEM to each well. 16. Incubate plates at 37 °C and 5% of CO2 for 10 up to 30 days. 17. Feed the cells 1–2 times per week with supplemented DMEM. 18. Stain the plates with 500 μL of iodonitrotetrazolium chloride for at least 1 h. 19. Wash the wells with 1x PBS. 20. Count the colonies using a dissecting microscope.

152

Laura De Lara-Pen˜a et al.

3.3 Cancer Stem Cell Characterization by Flow Cytometry

1. Transfer the secondary tumorspheres using a Pasteur pipette to a new 15 mL conical tube filled with 10 mL of 1x PBS.

3.3.1 Hoechst Exclusion Assay

3. Resuspend the pellet of tumorspheres with 1 mL of 1x PBS.

2. Centrifuge at 300 g for 5 min and remove the supernatant. 4. Disaggregate the tumorspheres using a 1 mL syringe with 0.5 × 16 mm needle by aspirating and releasing the solution approximately 30 times. 5. Centrifuge at 300 g for 5 min and remove the supernatant. 6. Resuspend the pellet with an appropriate volume of 1x PBS based on the size of the pellet. 7. Perform cell counting. 8. Resuspend the cells in an appropriate volume of supplemented DMEM to achieve a concentration of 1 × 106 cells/mL and transfer to a 1.5 mL Eppendorf tube. 9. Add Hoechst 33342 to a final concentration of 5 μg/mL. 10. Add 5 μM verapamil for negative controls. 11. Incubate in the dark at 37 °C for 90 min. 12. Centrifuge at 300 g for 5 min at 4 °C and resuspend the cells in 1 mL of cold 1x PBS. 13. Transfer the cells to flow cytometry tubes. 14. Perform the flow cytometry run collecting up to 100,000 events and analyze the Hoechst 33342-positive cells in Hoechst blue (440/40) and Hoechst red (695/40) channels.

3.3.2 Cancer Stem Cell Marker Evaluation

Each cell line has its own sets of markers (listed in Tables 1 [19] and 2 [20]). It is recommended to compare the expression of these markers between the CSC-enriched culture and the normal culture. 1. To detach cells from the bottom of the flask: (a) Aspirate the medium using a sterile 10 mL pipette. (b) Add 5 mL of 1x PBS and wash the cells. (c) Add 4 mL of trypsin–EDTA solution to detach the cells and incubate for 5 min at 37 °C. (d) Once the cells are detached, add 4 mL of supplemented DMEM for trypsin inactivation. (e) Collect the cells into a 15 mL conical tube. (f) Centrifuge at 300 g for 5 min and remove the supernatant. 2. To disaggregate the tumorspheres: (a) Transfer the secondary tumorspheres using a Pasteur pipette to a new 15 mL conical tube. (b) Centrifuge the cells at 300 g for 5 min and remove the supernatant.

Mimicking the Tumor Niche: Methods for Isolation, Culture, and. . .

153

Table 1 CSC intracellular markers [19] CSC intracellular marker

Origin and function

ALDH1A1 (aldehyde dehydrogenase 1 family member A1)

Alcohol metabolism and synthesis of retinoic Colorectal, gastric, acid lung, breast, glioblastoma, prostate

LETM1 (leucine zipper and EF- Mitochondrial proton/calcium exchanger hand-containing protein transmembrane protein 1)

Expression in CSCs

Colorectal, gastric

NANOG

DNA binding homeobox transcription factor Colorectal, gastric, involved in embryonic stem cell liver, lung, breast, proliferation, renewal, and pluripotency glioblastoma

POU5F1 (POU class 5 homeobox 1)

Transcription factor involved in embryonic development and stem cell pluripotency

Colorectal, gastric, liver, lung, breast, glioblastoma

SALL4 (Spalt-like transcription Transcription factor involved in the factor 4) development of abducens motor neurons

Colorectal, glioblastoma

SOX2 (SRY-box transcription factor 2)

Transcription factors involved in embryonic development and cell fate

Colorectal, gastric, liver, breast, glioblastoma

AFP (alpha fetoprotein)

Major plasma protein during fetal life

Liver

NOTCH1 (Notch receptor 1)

Notch signaling pathway

Liver, breast

NOTCH3 (Notch receptor 3)

Notch signaling pathway

Liver, breast

CTNNBL1 (catenin beta-like 1) It is a part of spliceosome and involved in pre-mRNA splicing

Liver, breast

BMI-1 (BMI1 proto-oncogene) Polycomb ring finger—epigenetic regulation, Breast embryonic development, self-renewal FOXO3 (forkhead box O3)

Transcription factor involved in apoptosis

Breast

KLF4 (Kruppel-like factor 4)

Transcription factor involved in G1-to-S Glioblastoma transition control following DNA damage

NES (Nestin)

Member of the intermediate filament protein Glioblastoma family—brain and eye development

TGM2 (transglutaminase 2)

Involved in apoptosis

Prostate

(c) Add 10 mL of 1x PBS and wash the cells. (d) Repeat steps b and c three times, centrifuge the cells at 300 g for 5 min, and remove the supernatant. (e) Add 200 μL of trypsin–EDTA solution to detach the cells and incubate for 5 min at 37 °C. (f) Observe the disaggregation status under an optical microscope.

154

Laura De Lara-Pen˜a et al.

Table 2 CSC surface markers [20] CSC surface marker

Origin and function

Expression in CSCs

ABCB5

ABC transporter

Melanoma

ABCG2

ATP-binding cassette transporter

Lung, breast, brain

CD10 (neprilysin)

Metallo-endopeptidase, FDA-approved Breast, head and neck target

CD105 (endoglin)

Coreceptor for TGF-β

Renal

CD114 (CSF3R)

Colony-stimulating factor 3 receptor, FDA-approved target

Neuroblastoma

CD117 (c-KIT)

Receptor for stem cell factor, FDA-approved target

Ovary

CD123 (IL-3R)

Receptor for IL-3

Leukemia

CD13 (alanine aminopeptidase)

Marker for kidney disease

Liver

CD133 (AC133)

Marker for hematopoietic stem cells

Breast, prostate, colon, glioma, liver, lung, ovary

CD146 (MCAM)

Melanoma cell adhesion molecule

Rhabdoid tumor, sarcoma

CD166 (ALCAM)

Cell–cell/cell–matrix interaction

Colorectal, lung

CD20 (MS4A1)

B cell lineage, FDA-approved target

Melanoma

CD24

B cell proliferation

Breast, gastric, pancreas

CD26 (DPP-4)

Dipeptidyl peptidase iv, FDA-approved Colorectal, leukemia target

CD271

Nerve growth factor receptor

Melanoma, head and neck

CD29 (integrin β1)

Cell adhesion, FDA-approved target

Breast, colon

CD326 (EpCAM)

Cell adhesion, signal transduction

Colon, pancreas, liver

CD34

Cell adhesion

Leukemia, squamous cell carcinoma

CD44 variants

Hyaluronic acid receptor, FDA-approved target

HNSCC, breast, colon, liver, ovarian, pancreas, gastric

CD49f (integrin α6)

Cell adhesion

Glioma

CD54 (ICAM-1)

Cell adhesion, FDA-approved target

Gastric

CD55 (DAF)

Inhibitor of complement

Breast

CD56 (NCAM)

Cell adhesion

Lung

CD9

Cell adhesion

Leukemia

CD90 (Thy-1)

Signal transduction/cell adhesion

Brain, liver

CD96

T cell-specific receptor

Leukemia (continued)

Mimicking the Tumor Niche: Methods for Isolation, Culture, and. . .

155

Table 2 (continued) CSC surface marker

Origin and function

Expression in CSCs

Cripto-1 (TDGF1)

Self-renewal/survival in esc

Breast, colon, lung

CXCR1, 2

Receptor for chemokine

Breast, pancreas

CXCR4

Receptor for chemokine, FDA-approved target

Breast, brain, pancreas

DLL4 (delta-like ligand 4)

Notch ligand

Colorectal, ovarian

LGR5

Cell adhesion

Intestinal, colorectal

Notch2

Signal transduction

Pancreas, lung

Notch3

Signal transduction

Pancreas, lung

PODXL-1 (podocalyxin- Ligand for L-selectin like protein 1)

Leukemia, breast, pancreas, lung

SSEA1

Mouse ESC marker

Teratocarcinoma, renal, lung

SSEA3

hESC marker

Teratocarcinoma, breast

SSEA4

hESC marker

Teratocarcinoma, breast

TIM-3 (HAVCR2)

Immune checkpoint receptor

Leukemia

TRA-1-60

hESC marker

Teratocarcinoma, breast, prostate

TRA-1-81

hESC marker

Teratocarcinoma, breast

(g) If the disaggregation is complete, add 500 μL supplemented DMEM for trypsin inactivation. (h) Collect the cells into a 15 mL conical tube. (i) Centrifuge at 300 g for 5 min and remove the supernatant. (j) Repeat steps b and c three times, centrifuge the cells at 300 g for 5 min, and remove the supernatant to remove any traces of FBS. 3. Resuspend the cells in fixation buffer by briefly vortexing. 4. Incubate at room temperature for at least 30 min, or overnight at 4 °C. 5. Centrifuge at 300 g for 5 min and remove the supernatant. 6. If the staining is not carried out immediately, resuspend the cells in blocking buffer and store at 4 °C until use. 7. When proceeding with the experiment, centrifugate at 300 g for 5 min and remove the blocking buffer.

156

Laura De Lara-Pen˜a et al.

For intracellular markers (Table 1), proceed to steps 10 and 11. For surface markers (Table 2), go to step 13. 8. Add permeabilization buffer and incubate for 10 min at room temperature. 9. Centrifuge at 300 g for 5 min, and remove the permeabilization buffer. 10. Add 10 mL of 1x PBS and wash the cells. 11. Centrifugate at 300 g for 5 min and remove the supernatant. 12. Repeat steps 10 and 11 for three times. 13. Perform cell counting and bring the cells to a concentration of 0.5–1 × 106 cells per tube. 14. Add staining buffer containing the antibody and incubate according to the manufacturer’s instructions. 15. Rinse the cells in FACS buffer. 16. Centrifuge at 300 g for 5 min. 17. Repeat steps 15 and 16 three times. 18. Resuspend the cells in 500 μL of 1x PBS and store in the dark until use. 19. Proceed with the flow cytometry run (see Note 2) Lists of stemness markers can be found in Tables 1 and 2. 3.4 In Vivo Tumorigenic Properties of CSCs (Orthotopic Limiting Dilution Assay)

1. Aspirate the medium using a sterile 10 mL pipette. 2. Add 5 mL of 1x PBS and wash the cells. 3. Add 4 mL of trypsin–EDTA solution to detach the tumor cell monolayer and incubate for 5 min at 37 °C. 4. Once the cells are detached, add 4 mL of supplemented DMEM for trypsin inactivation. 5. Collect the cells into a 15 mL conical tube. 6. Centrifuge at 300 g for 5 min and remove the supernatant. 7. Add 10 mL of 1x PBS and wash the cell pellet. 8. Centrifuge at 300 g for 5 min and remove the supernatant. 9. Resuspend the pellet in 1 mL of PBS. 10. Perform cell counting. 11. Seed 5000 cells/well in 6-well ultralow-attachment plates with CSC medium. 12. Collect the primary tumor-spheres using a Pasteur pipette in a new 15 mL conical tube filled with 10 mL of 1x PBS. 13. Centrifuge at 300 g for 5 min at 4 °C and remove the supernatant.

Mimicking the Tumor Niche: Methods for Isolation, Culture, and. . .

157

14. Disgregate and separate CSC spheres with 200 μL of trypsin– EDTA for 3–5 min at 37 °C. 15. Neutralize with 5 mL of supplemented DMEM 16. Resuspend the tumorsphere pellet with 1 mL of 1x PBS. 17. Centrifuge at 300 g for 5 min at 4 °C. 18. Add 10 mL of 1x PBS to wash the cell pellet. 19. Centrifuge at 300 g for 5 min at 4 °C. 20. Resuspend the pellet in 1 mL of 1x PBS. 21. Perform cell counting. 22. Prepare tubes of 1.5 mL with the dilution to 1000, 100, and 10 viable cells in 100 μL of Matrigel per mice and keep them on ice. 23. Prepare mice in a laminar flow cabinet to keep the highest possible sterility. 24. Anesthetize the mice with 4% isoflurane–oxygen mixture for induction and 2.5% for the maintenance of anesthesia with 100% oxygen flow. 25. Inject subcutaneously 100 μL of Matrigel with cells in each NSG male. 26. Assess tumor growth twice weekly using a digital caliper. Calculate the tumor volume by the formula V = (length) 2 × width × π/6. 3.5 Multicellular Spheroids 3.5.1

Cell Culture

3.5.2 Formation of Multicellular Spheroids Following the HangingDrop Protocol

Mesenchymal stem cell (MSC) and fibroblast (FB) cell lines are cultured with supplemented DMEM in culture flasks under controlled conditions (37 °C, 5% CO2). At 80% of confluence, cells are subcultured using trypsin–EDTA solution. 1. Aspirate the medium using a sterile 10 mL pipette. 2. Add 5 mL of 1x PBS and wash the cells. 3. Add 4 mL of trypsin–EDTA solution to detach the tumor cell monolayer and incubate for 5 min at 37 °C. 4. Once the cells are detached, add 4 mL of supplemented DMEM for trypsin inactivation. 5. Collect the cells into a 15 mL conical tube. 6. Centrifuge at 300 g for 5 min and remove the supernatant. 7. Add 10 mL of 1x PBS to the cell pellet. 8. Centrifugate at 300 g for 5 min and remove the supernatant. 9. Resuspend the pellet in 1 mL of supplemented DMEM. 10. Perform cell counting.

158

Laura De Lara-Pen˜a et al.

11. Prepare the amount of methylcellulose medium required according to the number of spheroids to be seeded (25 μL/ spheroid) including the cell suspension (500 cells per spheroid layer). 12. Seed 500 cells per 25 μL droplet of methylcellulose medium. Create the droplets using a multichannel micropipette on the lid of a Petri dish, with sufficient spacing between the droplets to prevent them from merging together, and close the dish to keep the droplets in suspension. 13. After 24 h of culture, the cells will have aggregated at the center of the droplet, forming a cellular spheroid (see Note 3). To add a subsequent cell layer, repeat steps 1–12 with FBs and/or MSCs after 24 h, using the same concentrations of medium and cells for the second layer (500 cells/25 μL of medium). Carefully, seed the 25 μL of methylcellulose medium with cell suspension over the drop seeded the day before, so that the second cell layer can be seeded over the spheroid already formed from the first layer. 14. To maintain the hanging drop culture for up to 4 days, add 10–15 μL of fresh methylcellulose medium to each drop every 24 h. 15. Optionally, the cells can be stained prior to visualization under the confocal microscope. To do so, after step 5, follow these additional steps: (a) Stain the pellet with 2 microliters of CellTracker Solution for 30 min. (b) Add 10 mL of 1x PBS and wash the cells. (c) Centrifuge at 300 g for 5 min at 4 °C. (d) Repeat steps b and c twice more (perform two additional washes).

4

Notes 1. Tumorsphere appearance Spheres of CSCs may present different morphologies depending on the cell line. For example, the A375 tumor line exhibits cluster-like spheroids, while the BXPC3 line tends to form more compact tumorspheres, but if the culture has been performed correctly, there will be few or no individual cells in suspension. Incorrect sphere formation may indicate that the culture medium is not well prepared, or that the cells are not viable at the beginning of the experiment. If the spheres have formed too large, the core will appear black and visibly obscured; in that case, reduce the number of cells per well at seeding.

Mimicking the Tumor Niche: Methods for Isolation, Culture, and. . .

159

2. Potential problems related with flow cytometry (a) Low or absent signal, low event rate: – Signal could be not correctly compensated: Verify if positive single-color control is set up correctly on the flow cytometer, gated, and compensated correctly to capture all the events. – Lasers could be wrongly aligned: Ensure the correct laser alignment by running flow check beads. – The antibody concentration could be too low: Increase antibody concentration. – The cell concentration could be too low: Increase cell concentration. – The cells may form aggregate, blocking the machine: Before running the experiment be sure to obtain a single-cell suspension by accurate pipetting. – The intracellular target could be inaccessible: Verify and optimize the permeabilization protocol for the experiment. – The target protein could be non-expressed or expressed at too low concentration: Check literature; furthermore, when targets are less expressed, a co-stain may be included to differentiate less expressed target or less frequent cell population. – The reagents may be expired or damaged: Check the expiration dates and, in case of problems, perform test experiments – The setting may be wrong: Set the offset and gain parameters using a positive control. (b) High signal intensity or high event rate: – The antibody quantity may be too high: Reduce the amount of antibody. – In case of intracellular staining the antibody could be trapped into the cell: Perform washing with detergents, such as Triton or Tween. – The number of cells may be too high: Adjust the cell concentration. (c) High or nonspecific background: – The amount of antibody may be too high, inducing nonspecific conjugation: Reduce the amount of antibody. – The use of secondary antibodies may cause a crossreaction between primary and secondary antibodies: Check the specificity of the secondary antibody.

160

Laura De Lara-Pen˜a et al.

– The setting may be wrong: Set the offset and gain parameters using a positive control. – If in the results a high side scattered background is present, it could be due to cell debris or bacterial contamination: Ensure that the cells have not been damaged during preparation; ensure that the culture has not been polluted. (d) Result variability day to day: – The cell viability may be changed: Verify that cell viability is constant in repeats of experiments. – Calibration problems of the instrument: Check the adequate instrument calibration. – The antibody’s batch may be different: Be sure to use the same batch of antibody in the experimental repetitions. 3. Multicellular spheroid appearance If the nucleus of the multicellular spheroid appears darker, this may be due to necrosis inside the spheroid; in this case, reduce the number of cells in the seeding. For co-culture of different cell types, it is necessary to optimize the culture medium in advance; this protocol describes co-culture of tumor cells with MSCs and FBs growing optimally in supplemented DMEM. References 1. Walcher L, Kistenmacher AK, Suo H, Kitte R, Dluczek S, Strauß A et al (2020) Cancer stem cells-origins and biomarkers: perspectives for targeted personalized therapies. Front Immunol 11:1280 2. Nassar D, Blanpain C (2016) Cancer stem cells: basic concepts and therapeutic implications. Annu Rev Pathol. 11:47–76 3. Herna´ndez-Camarero P, Jime´nez G, Lo´pezRuiz E, Barungi S, Marchal JA, Pera´n M (2018) Revisiting the dynamic cancer stem cell model: importance of tumour edges. Crit Rev Oncol Hematol. 131:35–45 4. Toledo B, Gonza´lez-Titos A, Herna´ndezCamarero P, Pera´n M (2023) A brief review on chemoresistance; targeting cancer stem cells as an alternative approach. Int J Mol Sci 24(5):4487 5. Najafi M, Farhood B, Mortezaee K (2019) Cancer stem cells (CSCs) in cancer progression and therapy. J Cell Physiol 234(6):8381–8395

6. Batlle E, Clevers H (2017) Cancer stem cells revisited. Nat Med 23(10):1124–1134 ˜a´n-Liso´n C, 7. Lo´pez de Andre´s J, Grin Jime´nez G, Marchal JA (2020) Cancer stem cell secretome in the tumor microenvironment: a key point for an effective personalized cancer treatment. J Hematol Oncol 13(1):136 8. Jime´nez G, Lo´pez de Andre´s J, Marchal JA (2020) Stem cell-secreted factors in the tumor microenvironment. Adv Exp Med Biol 1277: 115–126 9. Babaei G, Aziz SG, Jaghi NZZ (2021) EMT, cancer stem cells and autophagy; the three main axes of metastasis. Biomed Pharmacother 133:110909 10. Paolillo M, Colombo R, Serra M, Belvisi L, Papetti A et al (2019) Stem-like cancer cells in a dynamic 3D culture system: a model to study metastatic cell adhesion and anti-cancer drugs. Cells 8(11):1434 11. Herna´ndez-Camarero P, Lo´pez-Ruiz E, Marchal JA, Pera´n M (2021) Cancer: a

Mimicking the Tumor Niche: Methods for Isolation, Culture, and. . . mirrored room between tumor bulk and tumor microenvironment. J Exp Clin Cancer Res 40(1):217 12. Jime´nez G, Hackenberg M, Catalina P, ˜ a´n-Liso´n C et al (2018) MesBoulaiz H, Grin enchymal stem cell’s secretome promotes selective enrichment of cancer stem-like cells with specific cytogenetic profile. Cancer Lett 429: 78–88 13. Mele L, Liccardo D, Tirino V (2018) Evaluation and isolation of cancer stem cells using ALDH activity assay. Methods Mol Biol 1692: 43–48 14. Tirino V, Desiderio V, Paino F, De Rosa A, Papaccio F et al (2013) Cancer stem cells in solid tumors: an overview and new approaches for their isolation and characterization. FASEB J 27(1):13–24 ˜ a´n-Liso´n C, Olivares-Urbano MA, 15. Grin Jime´nez G, Lo´pez-Ruiz E, Del Val C et al (2020) miRNAs as radio-response biomarkers for breast cancer stem cells. Mol Oncol 14(3): 556–570

161

˜ an-Lison C, Farace C, Fiorito G, 16. Pisano A, Grin Fenu G et al (2020) The inhibitory role of miR-486-5p on CSC phenotype has diagnostic and prognostic potential in colorectal cancer. Cancers (Basel) 12(11):3432 17. Giolito MV, Claret L, Frau C, Plateroti M (2021) A three-dimensional model of spheroids to study colon cancer stem cells. J Vis Exp. 167 18. Lo´pez de Andre´s J, Ruiz-Toranzo M, Antich C, Chocarro-Wrona C, Lo´pez-Ruı´z E et al (2023) Biofabrication of a tri-layered 3D-bioprinted CSC-based malignant melanoma model for personalized cancer treatment. Biofabrication 15(3) 19. Atlas Antibodies (2022) Cancer stem cells markers. Accessed 6 June 2023 20. Kim WT, Ryu CJ (2017) Cancer stem cell surface markers on normal stem cells. BMB Rep 50(6):285–298

Chapter 12 Detection of Cancer Stem Cells in Normal and Dysplastic/Leukemic Human Blood Alessia De Stefano, Alessandra Cappellini, Irene Casalin, Stefania Paolini, Sarah Parisi, Maria Vittoria Marvi, Antonietta Fazio, Irene Neri, Foteini-Dionysia Koufi, Stefano Ratti, Carlo Finelli, Antonio Curti, Lucia Manzoli, Lucio Cocco, and Matilde Y. Follo Abstract The hierarchical organization of the leukemic stem cells (LSCs) is identical to that of healthy counterpart cells. It may be split into roughly three stages: a small number of pluripotent stem cells at the top, few lineage-restricted cells in the middle, and several terminally differentiated blood cells at the bottom. Although LSCs can differentiate into the hematopoietic lineage, they can also accumulate as immature progenitor cells, also known as blast cells. Since blast cells are uncommon in healthy bloodstreams, their presence might be a sign of cancer. For instance, a 20% blast cutoff in peripheral blood or bone marrow is formally used to distinguish acute myeloid leukemia from myelodysplastic neoplasms, which is essential to plan the patients’ management. Many techniques may be useful for blast enumeration: one of them is flow cytometry, which can perform analyses on many cells by detecting the expression of cell surface markers. Leukemic and non-leukemic blast cells might indeed be characterized by the same surface markers, but these markers are usually differently expressed. Here we propose to use CD45, in combination with CD34 and other cell surface markers, to identify and immunophenotype blast cells in patient-derived samples. Key words Cancer, Progenitor cells, Blasts, Flow cytometry, Immunophenotyping

1

Introduction Cancer stem cells, which are characterized by their ability of selfrenewal, are responsible for tumor development, progression, and relapse [1]. Self-renewal ability does not belong to cancer stem cells only. Indeed, the normal hematopoietic tissue has a hierarchical organization divided in a sort of three stages: on the top, a few pluripotent stem cells that can self-renew; in the middle,

Authors Alessia De Stefano and Alessandra Cappellini have equally contributed to this chapter. Federica Papaccio and Gianpaolo Papaccio (eds.), Cancer Stem Cells: Methods and Protocols, Methods in Molecular Biology, vol. 2777, https://doi.org/10.1007/978-1-0716-3730-2_12, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

163

164

Alessia De Stefano et al.

Fig. 1 Schematic representation of normal and malignant hematopoiesis. In normal hematopoiesis (green box), hematopoietic stem cells (HSCs) with self-renewal ability, give rise to multipotent progenitors (MPPs). These cells can differentiate into common lymphoid and myeloid progenitors (CLP and CMP, respectively). On the bottom, mature, nonproliferative, terminally differentiated blood cells are shown both for lymphoid and myeloid lineage. Mutations in HSCs, MPPs, or lineage-restricted progenitors give rise to preleukemic stem cells (grey box). Pre-LSCs can be further transformed, with mutation accumulation, into leukemic stem cells (LSCs). Myelodysplastic neoplasms (MDS) and acute myeloid leukemia (AML) originate from transformation of normal HSCs into LSCs, that subsequently can give rise to leukemic blast cells in circulation

some lineage-restricted cells with reduced self-renewal capability; and on the bottom, several mature, nonproliferative, terminally differentiated blood cells [2] (Fig. 1). Hematopoietic stem cells (HSCs) with oncogenic mutations maintain the same hierarchical organization as their healthy counterpart. They can differentiate into the hematopoietic lineage, carrying the alterations, or they can accumulate as immature progenitor cells, also known as blast cells [3]. The “blast” term, therefore, indicates an immature hematopoietic cell that is normally present in less than 5% of healthy bone marrow (BM). Therefore, blasts are not commonly found in healthy bloodstreams and their presence may indicate a disease state [4]. Before the transformation into myeloid malignancies, such as myelodysplastic neoplasms (MDS) or acute myeloid leukemia (AML), other blast-transforming events may give rise to preleukemic stem cells (pre-LSCs), characterized by the acquisition of specific gene mutations [5]. These preleukemic mutations in HSCs occur in 10–20% of healthy adults over the age of 70 and may become dominant in a limited number of HSC-derived clones, a process known as clonal hematopoiesis of indeterminate potential (CHIP) [6].

Flow Cytometry to Detect Blood Cancer Stem Cells

165

Even preleukemic mutations affecting the epigenome regulation, such as DNMT3A, ASXL1, IDH2, and TET2, are recurrently early associated with leukemogenesis. Interestingly, these mutations were detected even in B and T lymphocytes from myeloid neoplasms at diagnosis, indicating a clear contribution of pre-LSCs to hematopoiesis [7]. These pre-LSCs can endure chemotherapy and continue to exist in BM even after remission, acting as a potential reservoir for the relapse of leukemia [8, 9]. LSCs are exceedingly uncommon and often represent less than 0.1% of the total BM [10]. In 1997, Bonnet and Dick demonstrated that a subpopulation of LSCs was able to initiate tumors in NOD/SCID mice. They isolated this subpopulation using the expression of surface markers. In fact, although LSCs may show an immunophenotype similar to the one of normal HSCs in many aspects, the expression of some markers is peculiar. For instance, both LSCs and HSCs express CD34, but not CD38 (i.e., becoming CD34+CD38-) [11]. However, the expression of some other cell surface antigens on LSCs differs from their healthy counterpart. In particular, CD90 and CD117 are normally expressed on HSCs but not on LSCs [12, 13], whereas CD123 represents a unique marker for LSCs [14]. Both LSCs and the normal residual HSCs are retrieved in the hematopoietic tissues of patients diagnosed with myeloid neoplasms, and the ability of self-renewal is a crucial trait shared by both populations. The delicate balance between selfrenewal and differentiation in HSC and LSCs is regulated by several genes and signaling pathways. Some genes are not activated in normal HSCs, while others show differences in their activation mechanisms when in healthy or cancer cells. For instance, nuclear factor-κB (NF-κB) is a transcription factor that plays a crucial role in cellular proliferation, survival, and differentiation, is associated with anti-apoptotic activity in cancer and is not expressed in normal HSCs [2, 3]. Therefore, Nf-kB activation specifically indicates the presence of LSCs [15, 16]. Even the Wnt/β-catenin signaling pathway is dysregulated in various types of cancer [17], including myeloid leukemias, and is crucial for LSC self-renewal, tumor occurrence, development, and relapse. In fact, β-catenin is strongly expressed in normal HSCs but is downregulated during their myeloid differentiation [15]. Moreover, in LSCs, β-catenin accumulates in the nucleus and promotes the transcription of several oncogenes [16, 18]. The identification of the specific LSC features, as well as the differentially expressed genes or the surface markers, may therefore represent a potential therapeutic target that can contribute to eradicate LSCs in myeloid neoplasms [19]. According to several studies, LSCs are less sensitive to chemotherapy than blast cells and may lead to disease relapse [20]. In myeloid neoplasms, such as AML or MDS, the percentage of circulating blast cells has a critical role in defining the disease.

166

Alessia De Stefano et al.

According to the 5th edition of the World Health Organization Classification of Haematolymphoid Tumors, published in 2022 (WHO 2022), MDS are currently distinguished between MDS with low blasts (MDS-LB), with blast percentage less than 5% in BM and less than 2% in peripheral blood (PB), and MDS with increased blasts (MDS-IB), with blast percentage between 5% and 9% in BM or 2% and 4% in PB for MDS-IB type 1 (MDS-IB1) and between 10% and 19% in BM or 5% and 19% in PB for MDS-IB type 2 (MDS-IB2) [21]. Blast cutoff was eliminated for most AML types with the addition of defining genetic alterations, but a 20% blast cutoff in PB or BM is formally retained to distinguish AML from MDS [21, 22]. Although genetic alterations are becoming more and more important for the disease diagnosis, blast enumeration remains useful for therapeutic consideration and clinical trial design [23]. Depending on the techniques used, blasts are identified and quantified using various features. Cytomorphologic-based blast count is the current gold standard to assess blast percentage over the total nucleated cells in BM aspirate [24]. Although the blast morphology shows some differences, and standardization is still far from being achieved, some features may be considered common and useful for the diagnosis. Blast features include a large cell size, a large nucleus, a scant cytoplasm, a high nuclear to cytoplasmic ratio, immature chromatin, few cytoplasmic granules, prominent nucleoli, and, sometimes, needlelike cytoplasmic structures (i.e., Auer rods) [25]. However, blast enumeration by manual microscopy is affected by various limitations, such as the BM collection methods or the preparation of slides. These variables may affect the percentage composition, resulting in inaccurate blast quantification, even considering the low number of cells analyzed [26]. Therefore, alternative methods, such as flow cytometry, can minimize sampling errors by analyzing a larger number of cells [24]. In fact, blast enumeration using flow cytometry versus other direct count methods usually produces higher counts. Although myeloid malignancy diagnosis does not currently require the blast quantification by flow cytometry, it is necessary to pay attention to the sample collection, because the quantification can be compromised by hemodilution [27]. CD34, CD45, CD117, and CD123 are all markers found in normal myeloid progenitor cells and allow to distinguish them from other cell populations, such as HSCs, B cell precursors, and other lineage progenitors [28]. In MDS, the decreased expression, or even the absence, of CD45 and CD117 may uncover an underlying dysplasia of immature progenitors [28]. Even the altered expression of CD34 and CD33 is also associated with dysplasia [29]. Moreover, in the CD34-positive population, an abnormal expression of CD11b, CD15, CD2, CD5, CD7, and CD56 may be observed in blast cells [30].

Flow Cytometry to Detect Blood Cancer Stem Cells

167

In AML, the leukemic blast cells are CD34 positive and, generally, CD38 negative, which represents the most chemotherapyresistant blast cell compartment, recapitulating the features of quiescent LSCs [31]. In most AML patients, the CD33, CD44, CD45, CD99, and CD123 markers are all expressed [32], but several studies showed that these markers are expressed in the typical blast cell as well [33, 34]. Other markers, such as CD2 [32], CD11b, and CD56 [34], are expressed only on the leukemic blast cell surface, but, unfortunately, they are expressed in less than 50% of AML patients [35]. All in all, it is currently impossible to distinguish between leukemic blast cells and normal blast cells based on the analysis of a single marker. Therefore, to discriminate between the leukemic blast cells and their non-leukemic counterparts, it might be useful to combine these markers by building more complex antigen panels. Here, we propose to use flow cytometry to identify blast cells, using cell size and a limited panel of antibodies, in normal or cancer BM and PB samples, to perform a molecular characterization of patients’ cells at baseline, during therapy and follow-up.

2

Materials 1. Peripheral blood or bone marrow patients’ derived samples collected using EDTA as an anticoagulant. Optimal volume starts from 1 mL, of total blood. 2. Refrigerator (2–8 °C). 3. 5 mL polystyrene round-bottom tubes (12 × 75 mm, compatible with cytofluorimeter and low-speed centrifuge). 4. Red blood cell (RBC) lysis buffer (see Note 1): 8.29 g/L NH4Cl, 1 g/L KHCO3, 10 ml/L EDTA (100×) (EDTA [100×]: 10-2 M EDTA pH: 7.2 ÷ 7.4). 5. Phosphate-buffered saline (PBS): 8 g/L NaCl, 0.2 g/L KCl, 1.44 g/L Na2HPO4 (anhydrous), 0.2 g/L KH2PO4. 6. Cytofluorimeter equipped with 488 nm blue laser for excitation and with an optical system characterized by emission filters: FL1 530/30; FL2 574/26; FL3 695/40; FL4 780/60 (e.g., Attune NxT, Thermo Fisher). 7. Primary conjugated antibodies: CD45, CD14, CD11b, CD33, CD15, CD123, CD34, CD56, and CD62L. 8. Centrifuge. 9. Vortex.

168

3

Alessia De Stefano et al.

Methods

3.1 RBC Lysis and Sample Preparation

1. Keep BM or PB samples preferentially at room temperature or otherwise at 4 °C before starting the staining procedure and perform cytofluorimetric analysis within 24 h from the sample collection. 2. Perform RBC lysis by aliquoting 1,5 mL of RBC lysis buffer into four individual polystyrene round-bottom tubes (12 × 75 mm). Label the tubes indicating the chosen antibody combination (see Note 2). 3. Add 17 μL (see Note 3) of BM or PB sample into prefilled polystyrene round-bottom tubes. 4. Incubate for 20 min at room temperature and vigorously vortex every 5 min (see Note 4). 5. Centrifuge at 300× g for 5 min. 6. Remove supernatant and briefly vortex tubes to completely resuspend cells in residual solution. 7. Add 1 mL of PBS to each tube.

3.2 Direct Staining Procedure

1. Centrifuge the tubes (containing 1 mL of PBS + cells) at 300× g for 5 min. 2. Remove supernatant and briefly vortex tubes to completely resuspend cells in residual solution. 3. Add 100 μL of PBS in each tube. 4. Add 50 ng of each antibody (without any previous dilution), directly in the PBS prefilled tubes, for the surface staining, and briefly vortex tubes (see Note 5). 5. Incubate all tubes for 30 min, at room temperature, avoiding light exposure. 6. After incubation, wash all the samples, adding 2 mL of PBS in each tube. 7. Centrifuge at 300× g for 5 min. 8. Remove supernatant and briefly vortex tubes to completely resuspend cells in residual solution. 9. Add 2 mL of PBS in each tube and briefly vortex. 10. Perform the flow cytometric acquisition.

3.3 Flow Cytometric Acquisition and Parameter Setting

1. Set the acquisition protocol: create four dot plots and four histograms: • 1st dot plot: forward scatter–high (FCS-H) versus side scatter–high (SSC-H)

Flow Cytometry to Detect Blood Cancer Stem Cells

169

• 2nd dot plot: side scatter–area (SSC-A) versus SSC-H • 3rd dot plot: FCS-H versus SSC-H • 4th dot plot: SSC-H versus green fluorescence (FL1) • 1st univariate histogram: green fluorescence (FL1) • 2nd univariate histogram: orange fluorescence (FL2) • 3rd univariate histogram: red fluorescence (FL3) • 4th univariate histogram: deep red fluorescence (FL4) 2. Collect at least 800 μL of each sample (see Note 6) at an acquisition speed of 1000 μL/min (see Note 7), starting from the autofluorescence sample. 3. Pre-run an autofluorescence sample to properly visualize all populations of nucleated cells in the sample. In the first dot plot (FCS-H vs. SSC-H) work on “all events” to establish a gate (R1) of nucleated cells based on physical parameters (size and cellular complexity) excluding residual RBCs and debris (e.g., FSC photomultiplier voltage 188 mV; SSC photomultiplier voltage 388 mV) (see Note 8). An FSC-H threshold was applied around 35,000 (linear scale) to allow a better exclusion of red cell debris. The lymphocyte and erythroid precursors’ populations can be visualized at lower FCS-H and SSC-H channel values compared with the other leukocyte populations. Monocytes and granulocytes have similar FCS-H values, but granulocytes have a higher SSC-H (Fig. 2a).

Fig. 2 Steps for nucleated cell population identification. (a:) All the collected events are shown on the dot plot (y-axis, FCS-H; x-axis, SSC-H). The R1 gate (red) was created to circumscribe the leukocyte and erythroid precursors’ populations and exclude them from RBCs and debris. (b) R1-gated events are shown on the dot plot (y-axis, SSC-A; x-axis, SSC-H). The R2 gate (blue) was created to exclude doublets. (c) R2-gated events are shown on the dot plot (y-axis, FCS-H; x-axis, SSC-H). The shown gates are: R3, lymphocyte and erythroid precursors (yellow); R4, monocyte (light blue); and R5, granulocyte (red)

170

Alessia De Stefano et al.

Fig. 3 Leukocytes autofluorescence histograms based on R2-gated cells. Cell count was set up on the y-axis, and fluorescence parameters, related to each used antibody, were set up on the x-axis (FL1, FITC-H; FL2, R-PE-H; FL3, PerCP-H; FL4, PE-Cy7-H). A linear marker, in a different color, is shown for each histogram

4. In the 2nd dot plot (SSC-A vs. SSC-H), using only cells contained in the R1 gate, create a new polygonal gate (R2) to exclude doublets (Fig. 2b). 5. In the 3rd dot plot (FCS-H vs. SSC-H), the R2 gate is used to draw three gates to circumscribe each subpopulation and assign a specific identification color and label (Fig. 2c). 6. Set the photomultipliers’ voltage to place histogram peaks into the lowest decade according to the instrument sensitivity. The used voltages for each fluorescence channel should be the same as the ones used during the compensation procedure, to apply the right sample compensation during the analyses (see Note 9). A linear marker is created on each histogram above the subsequent decades to detect the samples’ positiveness (Fig. 3). Using the described setting, the fourth dot plot (SSC-H vs. FL1-H) represents the pivotal plot to discriminate the blast population using the chosen antibody (Fig. 4a). 7. Proceed with the stained sample acquisition as suggested in step 2. 8. For CD45, being a leukocyte common antigen, green fluorescence positivity is expected to be found in all the leukocyte populations. A new polygonal gate (R6) is created to circumscribe the area (CD45dim/int SCCDlow/int) where blasts are typically located. The blast population can be visualized at the same SSC-H value as the lymphocyte population (Fig. 4b). 3.4 Blast Phenotyping

Blast cells within the R6 gate can be further characterized for surface marker expression. In this chapter, the blast population can be stained with both CD45 and CD34, the latter being essential for the identification of blasts, along with other specific hematological markers (CD14, CD11b, CD33, CD15, CD123, CD56, and CD62L). Data were represented in histograms from the R6 region

Flow Cytometry to Detect Blood Cancer Stem Cells

171

Fig. 4 Nucleated cells populations (y-axis, SSC-H; x-axis, green fluorescence parameter; FL1, FITC-H). (a) R2-gated events, from the autofluorescence sample, are shown on the dot plot (lymphocytes and erythroid precursors: yellow; monocytes: light blue; and granulocytes: red). (b) The R6 gate (green) CD45dim/int SSClow/int is shown

selected. The events localized within the linear marker (set up in step 6) can be interpreted as the positive dataset for the considered surface marker, whereas events localized below the linear marker represent the negative dataset. As shown in Fig. 5a, multiple peaks are observed, due to markers’ different expressions within the same population in the analysis. Moreover, it is possible to perform a subsequent bivariate analysis on a population set in the positive range of a specific histogram. For instance, events included in the R11 linear marker can be plotted versus further fluorescence parameters (Fig. 5b). Events in the R6 gate can also be used for subsequent analyses using different dual-fluorescence parameter plots. These plots are created by drawing four quadrants to determine single positive cells (upper left [UL] quadrant and/or low right [LR] quadrant for each marker) and both double-negative (low left [LL] quadrant) and double-positive cells (upper right [UR] quadrant) (Fig. 5c). The inclusion of additional tubes in the analysis panel, maintaining the presence of both CD45 and CD34, might be useful. This approach enables the evaluation of CD34 co-expression with other recommended surface marker outlined in this protocol.

172

Alessia De Stefano et al.

Fig. 5 Sample images of blast phenotyping. (a) Events from the R6 region were visualized as a histogram (y-axis, cell count; x-axis, orange fluorescence parameters [FL2: R-PE-H]). A linear marker R11 (orange) is shown. (b) Events from the R11 linear marker were visualized as a dot plot (FL2, R-PE-H vs. FL3, PerCP-H). It can be observed that, within the FL2-positive population (R11 marker), two subpopulations coexist, expressing different PerCP fluorescences (UR quadrant) and a subpopulation FL3-negative (LR quadrant). (c) Four-color staining blast phenotyping sample images

Flow Cytometry to Detect Blood Cancer Stem Cells

4

173

Notes 1. Store the RBC lysis buffer at 4 °C for a maximum of 1 week. 2. According to the optical system of the instrument and antibodies used in this chapter, the four-polystyrene round-bottom tubes (12 × 75 mm) were labeled as follows: • Autofluorescence • CD45FITC/CD123PE/CD34PerCP/CD117PE-Cy7 • CD45FITC/CD11bPE/CD56PerCP/CD14PE-Cy7 • CD45FITC/CD15PE/CD33PerCP/CD62LPE-Cy7 3. A volume of 17 μL was identified as optimal both for hyperand hypocellularity samples. Indeed, the sample cellularity depends on the patients’ condition during the blood sampling. 4. Twenty minutes was identified as the maximum time for the RBC lysis procedure to avoid leukocyte lysis. 5. The antibody amount was chosen considering the different samples’ cellularity. For each used antibody, on a healthy whole blood sample, a titration curve was created. For each antibody, 50 ng represents an optimal amount to saturate all binding sites in patient-derived samples, which generally show cytopenia. Autofluorescence sample was prepared performing the same steps without adding the antibodies. 6. Collect at least a double volume for samples with cytopenia (1600 μL). 7. Here, the acquisition speed is optimal due to the sample high dilution and thanks to the acoustic focalization of the Attune NxT cytofluorimeter. For instruments equipped only with hydrodynamic focalization, run the sample at the highest acquisition speed. 8. During the analysis, some samples may show an incomplete RBC lysis (Fig. 6a). To overcome this trouble, a reduced number of events were visualized within the dot plot, to better delimitate the lymphocyte/erythroid precursors’ line during the creation of the R1 gate (Fig. 6b), and to incorporate the entire region where blasts may be inside the gate. 9. An accurate compensation setting is crucial for effective multiparameter analysis in flow cytometry. Here, a commercial Compensation Bead Kit, based on polystyrene microspheres, was used to set the flow cytometry compensation. Compensate for spectral overlap using the same antibodies’ microliters used for the samples’ staining. The compensation matrix is shown in Table 1.

174

Alessia De Stefano et al.

Fig. 6 Sample images of ungated cells (y-axis, FCS-H; x-axis, SSC-H). (a) 100% of the collected events were visualized in the dot plot. (b) Thirty percent of the collected events were visualized in the dot plot. The lymphocyte/erythroid precursors’ line was indicated by the red arrow

Table 1 Compensation matrix for four-color immunostaining of human BM and PB cells. BL1-BL4-H (blue laser 1-4) are detectors that measure the output from the 488 nm laser BL1-H

BL2-H

BL3-H

BL4-H

BL1-H

100.00

21.89

3.67

1.11

BL2-H

0.67

100.00

19.45

4.78

BL3-H

0.26

0.25

100.00

25.79

BL4-H

0.06

0.92

0.49

100.00

References 1. Yu Z, Pestell TG, Lisanti MP, Pestell RG (2012) Cancer stem cells. Int J Biochem Cell Biol 44(12):2144–2151 2. Wang JC, Dick JE (2005) Cancer stem cells: lessons from leukemia. Trends Cell Biol 15(9): 494–501 3. Jordan CT (2007) The leukemic stem cell. Best Pract Res Clin Haematol 20(1):13–18 4. Veiga CB, Lawrence EM, Murphy AJ, Herold MJ, Dragoljevic D (2021) Myelodysplasia

syndrome, clonal hematopoiesis and cardiovascular disease. Cancers (Basel) 13(8):1968 5. Jan M, Snyder TM, Corces-Zimmerman MR, Vyas P, Weissman IL, Quake SR, Majeti R (2012) Clonal evolution of preleukemic hematopoietic stem cells precedes human acute myeloid leukemia. Sci Transl Med 4(149):149ra18 6. Bowman RL, Busque L, Levine RL (2018) Clonal hematopoiesis and evolution to hematopoietic malignancies. Cell Stem Cell 22(2): 157–170

Flow Cytometry to Detect Blood Cancer Stem Cells 7. Corces-Zimmerman MR, Hong WJ, Weissman IL, Medeiros BC, Majeti R (2014) Preleukemic mutations in human acute myeloid leukemia affect epigenetic regulators and persist in remission. Proc Natl Acad Sci U S A 111(7): 2548–2553 8. Shlush LI, Zandi S, Mitchell A, Chen WC, Brandwein JM, Gupta V, Kennedy JA, Schimmer AD, Schuh AC, Yee KW, McLeod JL, Doedens M, Medeiros JJF, Marke R, Kim HJ, Lee K, McPherson JD, Hudson TJ, Brown AMK, Yousif F, Trinh QM, Stein LD, Minden MD, Wang JCY, Dick JE (2014) Identification of pre-leukaemic haematopoietic stem cells in acute leukaemia. Nature 506(7488):328–333 9. Stauber J, Greally JM, Steidl U (2021) Preleukemic and leukemic evolution at the stem cell level. Blood 137(8):1013–1018 10. Eppert K, Takenaka K, Lechman ER, Waldron L, Nilsson B, van Galen P, Metzeler KH, Poeppl A, Ling V, Beyene J, Canty JA, Danska JS, Bohlander SK, Buske C, Minden MD, Golub TR, Jurisica I, Ebert BL, Dick JE (2011) Stem cell gene expression programs influence clinical outcome in human leukemia. Nat Med 17(9):1086–1093 11. Bonnet D, Dick JE (1997) Human acute myeloid leukemia is organized as a hierarchy that originates from a primitive hematopoietic cell. Nat Med 3(7):730–737 12. Blair A, Hogge DE, Ailles LE, Lansdorp PM, Sutherland HJ (1997) Lack of expression of Thy-1 (CD90) on acute myeloid leukemia cells with long-term proliferative ability in vitro and in vivo. Blood 89(9):3104–3112 13. Blair A, Sutherland HJ (2000) Primitive acute myeloid leukemia cells with long-term proliferative ability in vitro and in vivo lack surface expression of c-kit (CD117). Exp Hematol 28(6):660–671 14. Jordan CT, Upchurch D, Szilvassy SJ, Guzman ML, Howard DS, Pettigrew AL, Meyerrose T, Rossi R, Grimes B, Rizzieri DA, Luger SM, Phillips GL (2000) The interleukin-3 receptor alpha chain is a unique marker for human acute myelogenous leukemia stem cells. Leukemia 14(10):1777–1784 15. Luis TC, Ichii M, Brugman MH, Kincade P, Staal FJ (2012) Wnt signaling strength regulates normal hematopoiesis and its deregulation is involved in leukemia development. Leukemia 26(3):414–421 16. Staal FJ, Famili F, Garcia Perez L, Pike-Overzet K (2016) Aberrant Wnt signaling in leukemia. Cancers (Basel) 8(9):78

175

17. Reya T, Morrison SJ, Clarke MF, Weissman IL (2001) Stem cells, cancer, and cancer stem cells. Nature 414(6859):105–111 18. Wang Y, Krivtsov AV, Sinha AU, North TE, Goessling W, Feng Z, Zon LI, Armstrong SA (2010) The Wnt/beta-catenin pathway is required for the development of leukemia stem cells in AML. Science 327(5973): 1650–1653 19. Ding Y, Gao H, Zhang Q (2017) The biomarkers of leukemia stem cells in acute myeloid leukemia. Stem Cell Investig 4:19 20. van Gils N, Denkers F, Smit L (2021) Escape from treatment; the different faces of leukemic stem cells and therapy resistance in acute myeloid leukemia. Front Oncol 11:659253 21. Khoury JD, Solary E, Abla O, Akkari Y, Alaggio R, Apperley JF, Bejar R, Berti E, Busque L, Chan JKC, Chen W, Chen X, Chng WJ, Choi JK, Colmenero I, Coupland SE, Cross NCP, Jong DD, Elghetany MT, Takahashi E, Emile JF, Ferry J, Fogelstrand L, Fontenay M, Germing U, Gujral S, Haferlach T, Harrison C, Hodge JC, Hu S, Jansen JH, Kanagal-Shamanna R, Kantarjian HM, Kratz CP, Li XQ, Lim MS, Loeb K, Loghavi S, Marcogliese A, Meshinchi S, Michaels P, Naresh KN, Natkunam Y, Nejati R, Ott G, Padron E, Patel KP, Patkar N, Picarsic J, Platzbecker U, Roberts I, Schuh A, Sewell W, Siebert R, Tembhare P, Tyner J, Verstovsek S, Wang W, Wood B, Xiao W, Yeung C, Hochhaus A (2022) The 5th edition of the World Health Organization classification of Haematolymphoid Tumours: myeloid and histiocytic/dendritic neoplasms. Leukemia 36(7):1703–1719 22. Swerdlow SH, Campo E, Harris NL, Jaffe ES, Pileri SA, Stein H, Thiele J (2017) WHO classification of Tumours of Haematopoietic and lymphoid tissues. IARC 23. Zhang Y, Wu J, Qin T, Xu Z, Qu S, Pan L, Li B, Wang H, Zhang P, Yan X, Gong J, Gao Q, Gale RP, Xiao Z (2022) Comparison of the revised 4th (2016) and 5th (2022) editions of the World Health Organization classification of myelodysplastic neoplasms. Leukemia 36(12): 2875–2882 24. Chen X, Fromm JR, Naresh KN (2022) “Blasts” in myeloid neoplasms – how do we define blasts and how do we incorporate them into diagnostic schema moving forward? Leukemia 36(2):327–332 25. Zini G (2021) How I investigate difficult cells at the optical microscope. Int J Lab Hematol 43(3):346–353

176

Alessia De Stefano et al.

26. DiFrancesco T, Boychuk DJ, Lafferty JD, Crowther MA (2012) Bone marrow aspirate collection and preparation–a comparison of three methods. Clin Invest Med 35(3):E114– E116 27. van de Loosdrecht AA, Alhan C, Be´ne´ MC, Della Porta MG, Dr€ager AM, Feuillard J, Font P, Germing U, Haase D, Homburg CH, Ireland R, Jansen JH, Kern W, Malcovati L, Marvelde GJT, Mufti GJ, Ogata K, Orfao A, Ossenkoppele GJ, Porwit A, Preijers FW, Richards SJ, Schuurhuis GJ, Subira´ D, Valent P, van der Velden VHJ, Vyas P, Westra AH, de Witte TM, Wells DA, Loken MR, Westers TM (2009) Standardization of flow cytometry in myelodysplastic syndromes: report from the first European LeukemiaNet working conference on flow cytometry in myelodysplastic syndromes. Haematologica 94(8): 1124–1134 28. Westers TM, Ireland R, Kern W, Alhan C, Balleisen JS, Bettelheim P, Burbury K, Cullen M, Cutler JA, Della Porta MG, Dr€ager AM, Feuillard J, Font P, Germing U, Haase D, Johansson U, Kordasti S, Loken MR, Malcovati L, te Marvelde JG, Matarraz S, Milne T, Moshaver B, Mufti GJ, Ogata K, Orfao A, Porwit A, Psarra K, Richards SJ, Subira´ D, Tindell V, Vallespi T, Valent P, van der Velden VHJ, de Witte TM, Wells DA, Zettl F, Be´ne´ MC, van de Loosdrecht AA (2012) Standardization of flow cytometry in myelodysplastic syndromes: a report from an international consortium and the European LeukemiaNet Working Group. Leukemia 26(7):1730–1741 29. Alhan C, Westers TM, Cremers EM, Cali C, Witte BI, Ossenkoppele GJ, de Loosdrecht AA (2014) High flow cytometric scores identify adverse prognostic subgroups within the

revised international prognostic scoring system for myelodysplastic syndromes. Br J Haematol 167(1):100–109 30. Wells DA, Benesch M, Loken MR, Vallejo C, Myerson D, Leisenring WM, Deeg HJ (2003) Myeloid and monocytic dyspoiesis as determined by flow cytometric scoring in myelodysplastic syndrome correlates with the IPSS and with outcome after hematopoietic stem cell transplantation. Blood 102(1):394–403 31. Hanekamp D, Cloos J, Schuurhuis GJ (2017) Leukemic stem cells: identification and clinical application. Int J Hematol 105(5):549–557 32. Zeijlemaker W, Kelder A, Oussoren-Brockhoff YJ, Scholten WJ, Snel AN, Veldhuizen D, Cloos J, Ossenkoppele GJ (2016) A simple one-tube assay for immunophenotypical quantification of leukemic stem cells in acute myeloid leukemia. Leukemia 30(2):439–446 33. Ehninger A, Kramer M, Ro¨llig C, Thiede C, Bornh€auser M, von Bonin M, Wermke M, Feldmann A, Bachmann M, Ehninger G, Oelschl€agel U (2014) Distribution and levels of cell surface expression of CD33 and CD123 in acute myeloid leukemia. Blood Cancer J 4(6):e218 34. Chen Z, Li Y, Tong Y, Gao Q, Mao X, Zhang W, Xia Z, Fu C (2016) Stepwise discriminant function analysis for rapid identification of acute promyelocytic leukemia from acute myeloid leukemia with multiparameter flow cytometry. Int J Hematol 03(3):306–315 35. Dalal BI, Mansoor S, Manna M, Pi S, Sauro GD, Hogge DE (2012) Detection of CD34, TdT, CD56, CD2, CD4, and CD14 by flow cytometry is associated with NPM1 and FLT3 mutation status in cytogenetically normal acute myeloid leukemia. Clin Lymphoma Myeloma Leuk 12(4):274–279

Chapter 13 Methods to Study the Role of Mechanical Signals in the Induction of Cancer Stem Cells Alessandro Gandin, Paolo Contessotto, and Tito Panciera Abstract Microenvironmental mechanical signals are fundamental regulators of cell behavior both in physiological and in pathological context, particularly in the induction and maintenance of tumorigenic properties. It is thus of utmost importance to experimentally recreate conditions that mimic the physical attributes of real tissues to study their impact on cell behavior and in particular on the induction of cancer stem cell (CSC) properties. Here we present protocols to investigate the role of mechanical stiffness on reprogramming of primary mammary gland cells into CSCs, including the synthesis of hydrogel substrates of the desired stiffness, the isolation and culture of primary differentiated normal cells derived from the human mammary gland, and the assessment of their CSC attributes after oncogene-mediated transformation. Key words Mechanotransduction, Hydrogels, Cancer stem cells, Mammary gland

1

Introduction The mechanical signals that cells receive from their microenvironment are recognized as overarching regulators of cell behavior, in a process referred to as mechanotransduction [1–3]. The transduction of mechanical cues is fundamental to ensure correct tissue development and regeneration and to control stemness properties [3–5], and aberrant mechanotransduction has been associated to a number of diseases, including tumor induction and progression [6, 7]. It thus appears increasingly important to adopt experimental strategies that allow to retro engineer physical attributes found in real tissues to study their impact on biological processes in vitro or ex vivo, particularly in the context of cancer research. The definition of cancer stem cells (CSCs), i.e. that fraction of cells specifically equipped with self-renewal and tumor-seeding potential [8], is typically based on functional assays that assess self-renewal ability in vitro and tumor-seeding capacity in vivo, starting from cell lines that have long been cultured on glass or

Federica Papaccio and Gianpaolo Papaccio (eds.), Cancer Stem Cells: Methods and Protocols, Methods in Molecular Biology, vol. 2777, https://doi.org/10.1007/978-1-0716-3730-2_13, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

177

178

Alessandro Gandin et al.

tissue culture plastic [9]. It should be noted, however, that these substrates provide to cells a mechanical rigidity in the range of various GPa, which vastly exceeds the stiffness physiologically observed in real tissues [10]. As such, they convey physical cues very different from those present in real cellular microenvironments [11]. Moreover, tissue culture plastic substrates fail to support the growth of most primary cells explanted from animal models or from patient-derived tissues [10]. Here we provide methods to assess the contribution of physiological-range microenvironmental mechanical cues in the induction of CSC attributes in primary differentiated mammary gland cells derived from patients. In the first part, we present a protocol to synthetize and functionalize polyacrylamide-based hydrogels to be used as synthetic substrates that provide physiologically relevant mechanical cues to cells; in the second part we provide protocols to isolate, culture ex vivo, and genetically manipulate primary mammary gland cells of human origin and to perform three-dimensional clonogenic assays to assess cancer stem cell properties after these cells have been conditioned by culturing on hydrogels of different stiffness. We recently demonstrated that the experimental setup here detailed is ideally suited to evaluate the impact of physical cues emanating from the microenvironment on the ability of oncogenic lesions to confer cancer stem cell properties to otherwise healthy normal cells [12]. Breast cancers emerge from cells of the luminal lineage that acquired cancer stem cell properties after oncogene transformation [13], and here we refer to the example of freshly explanted luminal differentiated (LD) cells from healthy human mammary glands, which are postmitotic cells [14]. Those cells can be reprogrammed into cells displaying CSC properties by the expression of a constitutively active form of HER2 (HER2-CA), but only when seeded on rigid hydrogels (40 kPa). This reprogramming does not occur when the same HER2-CA-transduced cells are plated on soft adhesive hydrogels of 0.5 kPa [12], phenocopying the compliance of the normal mammary gland [15].

2

Materials

2.1 Hydrogel Preparations

1. 40% (w/V) acrylamide solution in sterile water. 2. 2% (w/V) bis-acrylamide solution in sterile water. 3. 50 mM sodium acetate buffer pH 4.5. 4. 70% (V/V) ethanol solution in sterile water. 5. 50 mM sodium acetate, pH 4.5. 6. Kapton foil. 7. Polydimethylsiloxane (PDMS) gaskets (see Note 1). 8. Ammonium persulfate (APS).

CSC Mechanoassay

179

9. Dumont forceps. 10. Bunsen burner. 11. Vacuum chamber. 12. 24 mm glass coverslips. 13. Prepare under a chemical hood the coverslip functionalization solution: 95% V/V ethanol, 4% V/V acetic acid, and 1% V/V TMSPM (3-(trimethoxysilyl)propyl methacrylate). 2.2 Isolation and Seeding of Primary Mammary Luminal Cells

1. Dissociation medium: Advanced DMEM/F12 supplemented with 1% HEPES, 1% GlutaMAX, 600 U/ml collagenase I, 400 U/ml hyaluronidase. Prepare fresh at each experiment. 2. Wash medium: HBSS supplemented with 2% PenStrep (of a 10000 U/mL stock). Keep ice cold (4  C). 3. Human mammary epithelial growth medium (HMGM): Advanced DMEM/F12 supplemented with 1% HEPES, 1% GlutaMAX(R), 0.5% FBS, 4 μl/ml BPE, 10 ng/ml hEGF, 10 mM Y27632, 10 mM forskolin, and 1% PenStrep. Prepare fresh at each experiment. 4. Sorting solution: PBS 1X (Ca++/Mg++ free) supplemented with 1 mM EDTA, 25 mM HEPES pH 7.0, 0.1% BSA. Sterile filter 0.2 μm and store at 4  C. 5. Hemolytic solution (0.64% NH4Cl); prepare solution A, 7.1 g/L NH4Cl in sterile water, adjust pH to 7.65; and solution B, 20.6 g/L TrisBase in sterile water. Obtain hemolytic solution by mixing 1 volume of solution A with 9 volumes of solution B and adjust pH to 7.2. 6. Mammary clonogenic suspension medium (MCSM): Advanced DMEM/F12 supplemented with 1% HEPES, 1% GlutaMAX, 5% Matrigel, 2% FBS, 10 ng/ml human EGF, 4 μl/ml BPE, 10 mM forskolin, 2 μg/mL doxycycline, and antibiotics. Prepare fresh at each experiment on the day of clonogenic assay (step 1, Subheading 3.4) and always keep on ice to avoid Matrigel(R) gelification. 7. 0.25% trypsin/EDTA solution. 8. 0.05% trypsin/EDTA solution. 9. 5 mg/mL Dispase solution in HBSS. 10. 1 ug/mL DNAse-I solution in sterile water. 11. Disposable scalpels. 12. 40-μm cell strainers. 13. Antibodies: CD31-APC/Cy7 (BioLegend), CD45-APC/Cy7 (BD Biosciences), CD49f-PE (BD Biosciences), and EpCAMFITC (BD Biosciences); store according to manufacturer’s instructions.

180

Alessandro Gandin et al.

14. Growth factor-reduced Matrigel (Corning); stock at 20  C; thaw overnight at +4  C. Never leave at room temperature. 15. 50 mg/mL doxycycline in sterile water; store at

20  C.

16. 24-well multiwell ultralow-attachment tissue culture plates. 17. Plasmids: FUdeltaGW-rtTA (Addgene no. 19780); FUWtetO-EGFP (Addgene no. 84041); FUW-tetO-MCS (Addgene no. 84008); FUW-tetO-HER2-CA (as in Panciera et al. [12]). 18. Normal mammary gland tissue can be collected from healthy women undergoing reduction mastoplasty surgery (see Notes 2 and 3).

3

Methods

3.1 Hydrogel Preparations

Hydrogel preparations follow the protocols reported in [12, 16], with minor modifications. Hydrogel meshes are polymerized on a glass coverslip that serves both as adhesion surface and as carrier for the hydrogel (steps 1–16); hydrogels are then functionalized to optimize cell adhesion (steps 17–25) (see Note 4). 1. Wash coverslips by a quick immersion in acetone followed by a wash in 100% ethanol. 2. Flame one 24 mm glass coverslip on a Bunsen burner and place it on a clean glass petri dish with flamed side facing up. 3. Place a drop of coverslip functionalization solution on each glass coverslip, close the petri dish, and cover with aluminum foil; incubate 15 min at room temperature. 4. Wash coverslips three times in and out in acetone and place the coverslips on blotting paper with functionalized side facing up. 5. Bake the Kapton foil in an oven set at 110  C for 30 min to ensure smoothening of the surface. 6. Wash PDMS gaskets with 100% ethanol, blot the ethanol away, and place each PDMS gasket on the Kapton foil (see Fig. 1a); leave for a few minutes under the chemical hood to ensure complete ethanol evaporation (see Note 5). 7. Lift the Kapton foil and, maintaining it in a vertical position, rinse the Kapton foil and the adherent gaskets with abundant 70% ethanol once and with sterile water twice; settle the Kapton foil with gaskets facing up and leave for a few minutes under the chemical hood to dry. 8. Under a chemical hood mix acrylamide (AA) and bis-acrylamide (BA) solutions according to the required formulations to obtain stiff (40 kPa) or soft (0.5 kPa) hydrogels (see Table 1). Keep the solutions prepared in step 8 warm by placing in a water bath at 37  C for 5 min (see Note 5).

CSC Mechanoassay

181

Fig. 1 (a) Representative scheme of the setup used for hydrogel polymerization (step 13, Subheading 3.1). Kapton foil is shown in orange, and PDMS gaskets are shown as clear rings. The first row of the setup displays the recovery of polymerized hydrogel attached to glass coverslips. (b) Representative image of a 6-well plate with adhesion protein-functionalized (in green) polymerized hydrogels placed inside (step 22, Subheading 3.1)

Table 1 Hydrogel formulations, as in Panciera et al. [12] Hydrogel stiffness 0,5 kPa

40 kPa

Acrylamide (AA) w/V

4%

8%

Bis-acrylamide (BA) w/V

0.03%

0.48%

N,N,N,N-Tetramethylethylenediamine (TEMED) V/V

1:1000

1:1000

Ammonium persulfate (APS) w/V

0.1%

0.2%

9. Degas the solutions prepared in step 8 in a vacuum chamber by lowering internal pressure to 0.1 bar for 15 min (see Note 7). 10. During the degassing step, weight ammonium persulfate (APS) powder to prepare a stock solution of 20% w/V APS in sterile water (see Note 8). 11. Add N,N,N,N-tetramethylethylenediamine (TEMED) to the acrylamide (AA) and bis-acrylamide (BA) solution degassed in step 9, according to the formulations of Table 1; mix three times with a micropipette taking care not to introduce bubbles (see Note 9). 12. Add APS to the solution of step 11, according to the formulations of Table 1; mix five times with a micropipette taking care not to introduce bubbles. This is a complete hydrogel solution and polymerization will start from this moment; immediately proceed to step 13.

182

Alessandro Gandin et al.

13. Immediately place a drop of 140 μL hydrogel solution inside each gasket and cover it with a functionalized coverslip (from step 4), with functionalized side facing the hydrogel (Fig. 1a) (see Note 10). 14. Incubate at room temperature 8 min for 40 kPa and 17 min for 0.5 kPa hydrogels, respectively, to allow completion of polymerization, then remove the hydrogels from the Kapton foil by pulling on the coverslip side with Dumont forceps, and place the hydrogels in a petri dish in sterile water with hydrogel facing up (see Note 11). 15. After all rounds of polymerization have been completed, leave the hydrogels in sterile water overnight at room temperature to allow complete swelling and washout of unreacted reagents. 16. To recover Kapton foils and PDMS gaskets for future use, under a chemical hood remove residual unpolymerized reagents with blotting paper, and wash Kapton foil and gaskets twice with sterile water. Steps 17–25 describe the procedure for collagen coating of hydrogels, following the protocol of Damljanovic et al. [17] with minor modifications, which is required for proper adhesion of primary mammary gland cells, as in Panciera et al. [12]. For other types of coating with other adhesion proteins that might be required for other cell types, see Gandin et al. [16]. 17. Activate hydrogels by submersion in hydrazine hydrate overnight at room temperature under a chemical hood. 18. Stop the reaction by immersion of the hydrogels in 5% acetic acid V/V in sterile water. 19. Wash hydrogels by immersion in sterile water at room temperature for 1 h. 20. Dilute rat tail collagen 1 (Corning) to a concentration of 0.1 mg/mL in 50 mM sodium acetate buffer pH 4.5 (see Note 12). 21. Add sodium periodate to a final concentration of 3.6 mg/mL to the collagen solution of step 20, and incubate for 30 min at room temperature. 22. Pick up the hydrogels by the coverslip, clean the bottom of the coverslip with a drop of 70% ethanol, and place the hydrogels in a 6-well multiwell plate, one per well, with hydrogel facing up (see Fig. 1). 23. Sterilize the hydrogels in a tissue culture hood by exposing to UV radiation for 15 min.

CSC Mechanoassay

183

24. Place a drop of 50 μL collagen solution from step 21 on the hydrogel surface to cover the entire hydrogel and incubate for 1 h at room temperature. 25. Wash the hydrogels twice with PBS 1x, let the hydrogels dry for 5 min (see Note 13) under a tissue culture hood, and plate cells on top from step 3, Subheading 3.3 (see below). 3.2 Mammary Gland Dissociation and Isolation of Luminal Differentiated Cells

1. Under a tissue culture hood, place the mammary gland surgical sample on a strainer over a Becker jar and wash the tissue with abundant ice-cold wash solution to drain away as much blood as possible. 2. Place the mammary gland tissue in an empty non-cell adhesive dish of 150 mm diameter and mince the tissue with disposable scalpels approximately to 1 mm3 pieces. Proper mincing of the tissue is crucial for an efficient digestion. 3. Resuspend the minced tissue in 35 mL of tissue dissociation solution, making sure it is homogeneously distributed in the dish. 4. Leave the dish in tissue culture incubator (37  C, 5% CO2) overnight (14 h) to ensure complete dissociation. 5. After 14–16 h harvest the digested tissue and place in 50 mL conical tubes. 6. Shake the tubes vigorously until the solution is completely homogenized and place the tubes vertically in a tissue culture incubator for 30 min to let the tissue decant. 7. After 30 min tubes should appear as a solution separated in three layers (Fig. 2): The dissociated epithelial components of the mammary gland (fragmented mammary gland ductal tree) will have sedimented on the bottom of the tube, overlaid by tissue dissociation medium (mainly containing red blood cells and debris), with a layer of adipose tissue on top.

Adipose tissue

Dissociation medium

Fragmented mammary gland ductal tree

Fig. 2 Representative scheme of the three-layered suspension obtained after sedimentation of the overnight-digested human mammary gland tissue sample (step 7, Subheading 3.2)

184

Alessandro Gandin et al.

8. Taking care not to invert the tubes, aspirate the adipose tissue and tissue dissociation medium (the first two top layers) with a serological pipette and discard it; keep the mammary gland ductal tree in the bottom of the tube. 9. Wash the dissociated tissue with 10 mL of wash medium, followed by centrifugation (5 min at 350 RCF) and removal of the supernatant. Repeat this step twice. 10. Optional: If red blood cells are still visible in the pellet, resuspend the pellet in 4 mL of hemolytic solution and incubate for 3 min on ice, and then wash repeating step 9. 11. After the last wash step ensure that proper digestion was achieved by placing a few microliters of digested tissue under a microscope: The solution should appear enriched in mammary ductal tree fragments with minimal single-cell dissociation. Single-cell dissociation will be obtained with steps 12–15. 12. Resuspend the pellet in 3 mL of 0.25% trypsin/EDTA solution and incubate at 37  C for 5 min. 13. Without removing trypsin solution add on top 7.5 mL of Dispase solution (5 mg/mL in HBSS) supplemented with 1 ug/mL DNAse-I, pipette thoroughly, and incubate at 37  C for further 10 min. 14. Stop the digestion by adding 10 mL of HBSS supplemented with 5% FBS. 15. Filter the cells through a 40-μm cell strainer to remove residual tissue fragments and cell aggregates. 16. Spin down 5 min at 350 RCF and remove supernatant. 17. Resuspend the cell pellet in 300 μL of antibody solution (sorting solution supplemented with 5 μL CD31-APC/Cy7, to a final concentration of 3.33 μg/mL; 5 μL CD45 APC/Cy7, to a final concentration of 420 ng/mL; 12.5 μL EpCAM-FITC, to a final concentration of 125 ng/mL and 10 μL of CD49fPE) (see Notes 14 and 15). 18. Incubate with antibodies 30 min on ice in the dark. 19. Dilute antibody solution with 5 mL of wash solution, spin down 5 min at 350 RCF, and remove supernatant. 20. Resuspend the pellet in 2 mL of sorting solution and proceed to FACS (see Note 16). 21. To isolate human mammary gland luminal differentiated cells, after excluding CD31+CD45+(Lin+) cells, mammary cells should be sorted into four populations: LD cells, CD49f EpCAM+; LP cells, CD49f+EpCAM+; basal cells, CD49f+EpCAM ; and stromal cells, CD49f EpCAM , as represented in Fig. 3 (see Note 17).

185

LP

104

LD

103

-195

0 102

Epcam FITC-A

105

CSC Mechanoassay

-286

Stromal -102 0 102

103

Basal 104

105

CD49f PE-Cy5-A

Fig. 3 Representative FACS plot of mammary gland cells stained and sorted as in steps 17–21, Subheading 3.2 3.3 Seeding and Infection of Mammary LD Cells

1. Wash LD cells recovered from FACS procedure with 10 mL of HMGM, spin down 5 min at 350 RCF and eliminate supernatant. 2. Resuspend LD cell pellet in a total of 250 μL/well mix of lentiviral supernatant and HMGM as follows: FUdeltaGWrtTA/FUW-tetO-HER2-CA/HMGM in a 1:1:1 volume ratio. FUW-tetO-HER-CA can be substituted with FUWtetO-EGFP as a negative control or with any desired ORF cloned into the multiple cloning site of the tetracyclineinducible lentiviral vector FUW-tetO-MCS (see Note 18). 3. Plate on hydrogels (prepared as in steps 1–25 of Subheading 3.1) making sure that the cell suspension is plated as a drop only on top of the hydrogel surface (see Notes 19 and 20). 4. Let the cells sit in a cell culture incubator for 48 h to allow for proper cell attachment and infection. 5. After 48 h remove the drop of medium containing the lentiviral mix, taking care not to dry completely the hydrogels and cover the hydrogels with 1 mL of HMGM supplemented with 2 μg/ mL doxycycline to trigger tetracycline-dependent gene expression. 6. Let the cells sit in a cell culture incubator for 5 days to allow cell reprograming to a tumor-like state. Monitor tissue culture on a daily basis.

3.4 Clonogenic Assay

1. After 7 days of culture remove HMGM and detach cells from the hydrogels by incubation with 250 μL of 0.05% trypsin/ EDTA solution for 10 min at 37  C.

186

Alessandro Gandin et al.

Fig. 4 Representative bright-field image of colonies emerging from HER2-CA transduced LD cells after 14 days of culture in MCSM. Scale bar, 190 μm

2. Recover the detached cells in HMGM and count them using an automated cell counter (Countess(TM) II FL Cell Counter, or equivalent). 3. Spin down 5 min at 350 RCF, remove supernatant, and resuspend cells in the appropriate volume of ice-cold MCSM to obtain 2000 cells/mL and seed in 24-well ultralow-attachment plates, 1 mL per well. 4. After 14 days of culture in MCSM colonies should appear as compact spheres growing in suspension (see Fig. 4) and can be counted to assess the induction of cancer stem-cell-like properties. 5. To show self-renewal capacity of primary colonies, recover cells by incubating the suspension in an excess volume of ice-cold HBSS (ideal dilution is a 1:3 ratio) and incubate on ice for 1 h to allow Matrigel to dissolve. 6. Spin down 5 min at 350 RCF, remove the supernatant, and incubate for 5 min at 37  C in 250 μL of 0.05% trypsin/EDTA solution to obtain a single-cell suspension. 7. Stop trypsin by dilution of 1:10 in HBSS and count cells using an automated Countess II FL Cell Counter. 8. Spin down for 5 min at 350 RCF, remove supernatant, and resuspend cells in the appropriate volume of ice-cold MCSM to obtain a 2000 cells/mL suspension and seed in 24-well ultralow-attachment plates, 1 mL per well. 9. Secondary colonies can be counted 14 days after seeding, as in step 4.

CSC Mechanoassay

4

187

Notes 1. Gaskets of PDMS (250-μm thick, internal diameter 22 mm, external diameter 24 mm) can be easily prepared from a 250-μm height PDMS sheet (Specialty Manufacturing, Inc.) using hollow punches of the appropriate diameter. 2. Care should be taken in assuring to collect tissues from anonymized patients that provided informed consent and according to proper institutional guidelines approved by an authorized ethics committee. 3. Surgical-derived tissue should always be handled with care in Biosafety Level 2 tissue culture conditions. 4. Hydrogel synthesis must be started 24 h prior to mammary gland dissociation. To ensure proper handling timing, it is crucial to carefully consider the total number of hydrogels to be polymerized on the same day; an experienced operator should be able to confidently handle up to 50 hydrogels on the same day, by performing several rounds of polymerization of up to 5 hydrogels simultaneously. 5. After ethanol evaporation, PDMS gaskets will adhere to the Kapton foil; make sure that a tight seal is generated. 6. Acrylamide (AA) and bis-acrylamide (BA) solutions should be prepared in aliquots of 1 mL, which is sufficient to polymerize five hydrogels simultaneously, avoiding the risk of polymerization before the hydrogel solution is properly set in the gaskets and sealed with functionalized coverslips. 7. It is crucial to degas solutions properly, as molecular oxygen acts as inhibitor of acrylamide polymerization reaction. 8. APS stock solution should be prepared fresh at each hydrogel synthesis, as APS is very unstable in water. The same solution must be used within 30 min after resuspension in sterile water. In case a larger number of hydrogels are to be polymerized (e.g., more than 10), consider preparing more than one aliquot of APS powder. 9. Add TEMED to each aliquot of 1 mL acrylamide (AA) and bis-acrylamide (BA) solution just before adding APS (step 12, Subheading 3.1); do not add TEMED to a new aliquot before completion of the sealing of PDMS gaskets filled with the preceding aliquot (step 13, Subheading 3.1). 10. Make sure that no air bubbles are trapped within the gasket, as the presence of oxygen will cause inhomogeneous polymerization. 11. Take care not to let the hydrogels dehydrate by direct air contact, as this can create cracks on the hydrogel surface.

188

Alessandro Gandin et al.

12. Handle the collagen stock solution always on ice. 13. Take care not to exceed 5 min of drying, as overdying can render the hydrogel surface nonhomogeneous. 14. These antibody staining conditions are optimized for cell sorting with a FACS Aria III (BD Biosciences) cell sorter. Antibody conditions might need to be optimized for use with a different instrument. 15. From this step onwards, always operate in the dark, to avoid bleaching of the fluorescently labeled antibodies. 16. For optimal isolation, make sure to correct for the possible spillover of fluorochromes by incubating each fluorophoreconjugated antibody separately with the cell suspension and measuring spectral overlaps of single-color controls, to create a compensation matrix. 17. The identity of the epithelial cell populations can be validated by gene expression analysis as in Panciera et al. [12], by assessing the expression level of prototypical markers of basal (e.g., TP63, SNAI2, CD10, aSMA, K14) and luminal (e.g. K8, K18, K19, MUC1, E-CAD, CLDN4) cell identity. 18. Lentiviral supernatants should be prepared as in Panciera et al. [12]; lentiviral titers should be in the range of 2  108 viral particles/ml (quantified as in Panciera et al. [18]). 19. The ideal volume for drop plating should be 200 μL; this might be however slightly optimized by each operator to minimize the possibility of spilling the drop outside of the hydrogel. 20. Care should be taken to coordinate experimental timings, such that hydrogels are coated with collagen and washed during the FACS procedure, so that substrates are available as soon as cells have been isolated.

Acknowledgements This work was supported by a PRIN 2017 grant (no. 2017L8FWY8_004) and a CARIPARO Starting Grant (no. C94I19001680001) to T.P. References 1. Eyckmans J, Boudou T, Yu X, Chen CS (2011) A hitchhiker’s guide to mechanobiology. Dev Cell 21(1):35–47 2. Humphrey JD, Dufresne ER, Schwartz MA (2014) Mechanotransduction and extracellular matrix homeostasis. Nat Rev Mol Cell Biol 15(12):802–812

3. Panciera T, Azzolin L, Cordenonsi M, Piccolo S (2017) Mechanobiology of YAP and TAZ in physiology and disease. Nat Rev Mol Cell Biol 18(12):758–770 4. Discher DE, Mooney DJ, Zandstra PW (2009) Growth factors, matrices, and forces combine and control stem cells. Science 324(5935): 1673–1677

CSC Mechanoassay 5. Vining KH, Mooney DJ (2017) Mechanical forces direct stem cell behaviour in development and regeneration. Nat Rev Mol Cell Biol 18(12):728–742 6. Northey JJ, Przybyla L, Weaver VM (2017) Tissue force programs cell fate and tumor aggression. Cancer Discov 7(11):1224–1237 7. Piccolo S, Panciera T, Contessotto P, Cordenonsi M (2023) YAP/TAZ as master regulators in cancer: modulation, function and therapeutic approaches. Nat Cancer 4(1):9–26 8. Visvader JE, Lindeman GJ (2008) Cancer stem cells in solid tumours: accumulating evidence and unresolved questions. Nat Rev Cancer 8(10):755–768 9. Gupta PB, Chaffer CL, Weinberg RA (2009) Cancer stem cells: mirage or reality? Nat Med 15(9):1010–1012 10. Brusatin G, Panciera T, Gandin A, Citron A, Piccolo S (2018) Biomaterials and engineered microenvironments to control YAP/TAZdependent cell behaviour. Nat Mater 17(12): 1063–1075 11. Schwartz MA, Chen CS (2013) Cell biology. Deconstructing dimensionality. Science 339(6118):402–404 12. Panciera T, Citron A, Di Biagio D, Battilana G, Gandin A, Giulitti S et al (2020) Reprogramming normal cells into tumour precursors requires ECM stiffness and oncogenemediated changes of cell mechanical properties. Nat Mater 19(7):797–806

189

13. Van Keymeulen A, Lee MY, Ousset M, Brohee S, Rorive S, Giraddi RR et al (2015) Reactivation of multipotency by oncogenic PIK3CA induces breast tumour heterogeneity. Nature 525(7567):119–123 14. Linnemann JR, Miura H, Meixner LK, Irmler M, Kloos UJ, Hirschi B et al (2015) Quantification of regenerative potential in primary human mammary epithelial cells. Development 142(18):3239–3251 15. Acerbi I, Cassereau L, Dean I, Shi Q, Au A, Park C et al (2015) Human breast cancer invasion and aggression correlates with ECM stiffening and immune cell infiltration. Integr Biol (Camb) 7(10):1120–1134 16. Gandin A, Torresan V, Ulliana L, Panciera T, Contessotto P, Citron A et al (2022) Broadly applicable hydrogel fabrication procedures guided by YAP/TAZ-activity reveal stiffness, adhesiveness, and nuclear projected area as checkpoints for Mechanosensing. Adv Healthc Mater 11(3):e2102276 17. Damljanovic V, Lagerholm BC, Jacobson K (2005) Bulk and micropatterned conjugation of extracellular matrix proteins to characterized polyacrylamide substrates for cell mechanotransduction assays. BioTechniques 39(6): 847–851 18. Panciera T, Azzolin L, Fujimura A, Di Biagio D, Frasson C, Bresolin S et al (2016) Induction of expandable tissue-specific stem/ progenitor cells through transient expression of YAP/TAZ. Cell Stem Cell 19(6):725–737

Chapter 14 Co-Delivery Polymeric Poly(Lactic-Co-Glycolic Acid) (PLGA) Nanoparticles to Target Cancer Stem-Like Cells Catherine S. Snyder, Taylor Repetto, Kathleen M. Burkhard, Anish Tuteja, and Geeta Mehta Abstract Nanoparticle drug delivery has been promoted as an effective mode of delivering antineoplastic therapeutics. However, most nanoparticle designs fail to consider the multifaceted tumor microenvironment (TME) that produce pro-tumoral niches, which are often resistant to chemo- and targeted therapies. In order to target the chemoresistant cancer stem-like cells (CSCs) and their supportive TME, in this chapter we describe a nanoparticle-based targeted co-delivery that addresses the paracrine interactions between CSC and non-cancerous mesenchymal stem cells (MSCs) in the TME. Carcinoma-activated MSCs have been shown to increase the chemoresistance and metastasis of CSC. Yet their contributions to protect the CSC TME have not yet been systematically investigated in the design of nanoparticles for drug delivery. Therefore, we describe the fabrication of degradable poly(lactic-co-glycolic acid) (PLGA) nanoparticles (120–200 nm), generated with an electrospraying process that encapsulates both a conventional chemotherapeutic, paclitaxel, and a targeted tyrosine kinase inhibitor, sunitinib, to limit MSC interactions with CSC. In the 3D hetero-spheroid model that comprises both CSCs and MSCs, the delivery of sunitinib as a free drug disrupted the MSC-protected CSC stemness and migration. Therefore, this chapter describes the co-delivery of paclitaxel and sunitinib via PLGA nanoparticles as a potential targeted therapy strategy for targeting CSCs. Overall, nanoparticles can provide an effective delivery platform for targeting CSCs and their TME together. Forthcoming studies can corroborate similar combined therapies with nanoparticles to improve the killing of CSC and chemoresistant cancer cells, thereby improving treatment efficiency. Key words Nanoparticle drug delivery, PLGA, Degradable, Electrospray, Cancer stem cells, Codelivery, Paclitaxel, Sunitinib, Mesenchymal stem cells, Stromal cells

Authors Taylor Repetto and Kathleen M. Burkhard have equally contributed to this chapter. Federica Papaccio and Gianpaolo Papaccio (eds.), Cancer Stem Cells: Methods and Protocols, Methods in Molecular Biology, vol. 2777, https://doi.org/10.1007/978-1-0716-3730-2_14, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

191

192

1

Catherine S. Snyder et al.

Introduction

1.1 Cancer Stem Cells Show Increased Chemoresistance, Metastasis, and ALDH Activity due to PDGF Signaling with Mesenchymal Stem Cells

Ovarian cancer stem-like cells (CSCs) have been linked to increased chemoresistance, metastasis, and proliferation and are proposed to be the main driver in patient relapse following clinical treatment [1– 6]. CSCs have been identified to have increased aldehyde dehydrogenase (ALDH) activity and high CD133 expression, among other elevated markers (such as CD44, CD24, CD117), and detection of these cells is a predictor of poor clinical outcomes [7–12]. In the ovarian cancer microenvironment, CSCs interact dynamically with other noncancer cells, some of which become CSC-supportive, including mesenchymal stem cells (MSCs) [13, 14]. MSC reprogramming arises from bidirectional signaling between CSCs and MSCs through IL-6, SDF-1, TGF-β, BMP2, BMP4, CCL5, and platelet-derived growth factor (PDGF) [13–18]. Such reprogramming activates MSCs to carcinoma-associated MSC (CA-MSC). Previously, PDGF signaling between CSC and MSC demonstrated an increase in CSC platinum resistance, migration, EMT markers, and ALDH activity [18]. Additionally, sunitinib, a tyrosine kinase inhibitor which interrupts CSC-MSC communication through PDGF, negated the increase in platinum resistance and was suggested as a potential therapy to also reduce ALDH activity, EMT markers, and migration potential [18].

1.2 Nanoparticles for Co-Delivery of Sunitinib and Paclitaxel

Encapsulation of chemotherapeutics in a polymer nanoparticle increases their solubility, circulation residence time, and targeting of the delivered drugs [19, 20]. Paclitaxel is one of the ovarian cancer chemotherapeutic, and previous studies have shown that paclitaxel-encapsulated nanoparticles increase solubility and cytotoxicity [21–24]. While sunitinib has also been delivered as a nanoformulation for cancer therapy and found to be effective, it has not been studied as extensively as paclitaxel [25–27]. Sunitinib and paclitaxel have been delivered together in an open-label case study involving three cancer patients, where paclitaxel was delivered in the form of an albumin nanoparticle and sunitinib was delivered as a free drug. All three of the lung cancer patients experienced a reduction in tumor mass and symptoms with this co-delivery treatment, thereby providing strong premise to our work [28]. Codelivery, where two (or more) compounds are encapsulated in the nanoparticles, has generally been found to be more effective, and the first co-delivery nanoparticle formulation was approved by the FDA in 2017 [29]. Therefore, in this methods chapter, we demonstrate fabrication of a polymer nanoparticle which encapsulates both paclitaxel and sunitinib. Our protocols also showcase all of the steps of the nanoparticle fabrication and characterization processes. The cell-based assays can then be utilized to test the effectiveness of nanoparticle co-delivery. Although we present the

Co-Delivery Polymeric Poly(Lactic-Co-Glycolic Acid) (PLGA) Nanoparticles. . .

193

methods for encapsulation of paclitaxel and sunitinib in the PLGA nanoparticles here, these techniques can be applied for the co-delivery of many other compounds to target CSC and their heterogeneous interactions with non-tumor cells within the TME. Overall, the nanoparticle co-delivery method described here has the potential to improve cure rates for many cancers where CSCs play a major role in therapy resistance and tumor relapses. The example presented in this chapter focuses on ovarian cancers. However, the same techniques can be utilized to target CSCs in other solid epithelial cancers.

2

Materials

2.1 Nanoparticle Synthesis and Collection

1. Acid-capped poly(D,L-lactide-co-glycolide) (PLGA) with a ratio of 75:25 (lactide/glycolide) (Evonik RG753H).

2.1.1 Materials and Solvents

3. Triethylammonium formate solution, 1 M, pH 6.0 for HPLC (TEAF).

2. 0.01% Tween 20 in deionized water (DIW).

4. N,N-Dimethylformamide (DMF). 5. Acetonitrile (ACN). 6. Paclitaxel (abbreviated as Pac, APExBIO A4393). 7. Sunitinib malate (sunitinib; abbreviated as Sun, MedChemExpress HY-10255). 8. Formic acid. 9. Ultrapure grade water. 2.1.2

Supplies

1. Enclosed, nonconductive electrospray station. 2. Syringe pump. 3. Voltage source. 4. 1 mL syringes. 5. 30-gauge blunt-tip industrial needles. 6. Plastic forceps or ones with a rubber grip. 7. Collection pan: 9 inch disposable aluminum pie pans with a smooth bottom. 8. 1.5 inch plastic razor blades. 9. 40 μm nylon mesh cell strainers. 10. 50 mL centrifuge tube. 11. 13 mm, 0.2 μm PTFE syringe filters.

194

Catherine S. Snyder et al.

2.2 2D and 3D Cell Culture 2.2.1 Cells and Culture Medium

All cells were used within 10 passages to avoid genetic drift. 1. Ovarian cancer stem cells (CSCs): Epithelial ovarian cancer cells, OVCAR3 (American Type Culture Collection HTB-161), were transduced to express green fluorescent protein (GFP+). The cells were then sorted via fluorescenceactivated cell sorting (FACS) to isolate the subpopulation which highly expressed CD133 and had high ALDH activity to represent the CSC [18]. 2. CSC culture medium: Gibco RMPI 1640 cell medium with L-glutamine and phenol red and supplemented with 10% fetal bovine serum and 1% Gibco antibiotic–antimycotic (100×). 3. Human adipose-derived mesenchymal stem cells (MSCs) were purchased from Lonza (PT-5006) and were used without modification. 4. MSC culture medium: Lonza ADSC basal cell medium (PT-3273) supplemented with 10% fetal bovine serum, 1% Gibco antibiotic–antimycotic (100×), and 2 mM of Gibco’s 200 mM L-glutamine. 5. CSC-MSC culture medium: a mixture of 50% MSC medium and 50% CSC medium.

2.2.2 General Tissue Culture

1. 10 cm tissue culture dishes. 2. 384-well hanging drop plates (Xcentric Mold & Engineering Inc., Quickparts). 3. 15 mL centrifuge tubes. 4. 1.5 mL microcentrifuge tubes. 5. 10 mL serological pipette tips. 6. Hemacytometer. 7. Water bath sonicator. 8. Parafilm. 9. 1X phosphate-buffered saline (PBS). 10. Fetal bovine serum (FBS). 11. 0.25% trypsin–EDTA (1×). 12. 0.05% trypsin–EDTA (1×). 13. Trypan blue. 14. Pluronic acid. 15. Sterile deionized water.

Co-Delivery Polymeric Poly(Lactic-Co-Glycolic Acid) (PLGA) Nanoparticles. . .

3

Methods

3.1 Nanoparticle Synthesis and Collection 3.1.1

195

Electrospray Setup

1. Construct or purchase an enclosed, nonconductive electrospray station. The constructed station in Fig. 1 is made of an aluminum frame with sheets of polycarbonate to reduce drafts and to contain the drug-loaded nanoparticles that are produced. 2. Place a syringe pump on top of the station, such that the loaded syringe would point down. 3. Cut a hole through the top plastic sheet to allow the syringe tip to extend into the station. 4. Center a clean collection pan underneath the hole for the syringe on top of a nonconductive stand. 5. Ground the collection pan and thread the positive lead from the voltage source into the electrospray station so that the syringe needle can have a voltage applied to it when it is loaded.

Fig. 1 An enclosed, nonconductive electrospray station constructed in-house. The syringe pump is placed inside an enclosure made of an aluminum frame and polycarbonate siding. At the bottom of the polycarbonate enclosure, there is an aluminum pan placed on a nonconductive stand. The collection pan is connected to ground and the needle on the syringe with electrospray solution is connected to the positive lead of the voltage source

196

Catherine S. Snyder et al.

Table 1 PLGA co-delivery nanoparticle formulations Nanoparticle

DMF (μL) ACN (μL) TEAF (μL) Paclitaxel (mM) Sunitinib (mM) PLGA (mg)

Control

735

255

10

–

–

31.6

Paclitaxel

735

255

10

6.6

–

25.3

Sunitinib

735

255

10

–

10.0

25.3

Paclitaxel + sunitinib 735

255

10

6.6

10.0

25.3

3.1.2 Electrospray Solution Formulation

1. Prepare paclitaxel and sunitinib stock solutions in DMF. 2. Measure out the volumes of solvents and drug needed for your desired nanoparticle formation and combine them in a glass vial. See Table 1 for exact formulations used in this chapter (see Notes 1 and 2). 3. Add the PLGA and stir vigorously to ensure a homogeneous solution.

3.1.3 Nanoparticle Production

1. Prepare the electrospray Subheading 3.1.2.

solution

according

to

2. Aspirate 1 mL of the polymer solution using a 1 mL plastic syringe and then aspirate some air. 3. Point the syringe up such that no solution can drip out. Carefully clean all solution residue from the tip of the syringe and then tap all bubbles out. 4. Firmly press on the 30-gauge blunt-tip needle onto the syringe and then slowly press the plunger until solution starts to expel from the tip. Wipe the tip thoroughly with a dry paper wipe. 5. Load the syringe into the syringe pump and clamp the voltage line to the metal needle. Ensure that the needle remains straight and pointed down. 6. Place a dry paper wipe below the syringe to collect drips and then start the syringe pump until the solution begins to drip from the needle. Stop the syringe pump and thoroughly wipe the needle with a dry paper wipe. 7. Ensure the ground wire is connected to the collection pan. 8. Close the electrospray station, restart the syringe pump at 100 μL/h, and turn on the voltage source. The flow rate can be adjusted to alter nanoparticle characteristics (see Fig. 2). 9. Closely monitor the polymer droplet on the needle and adjust the voltage as needed to maintain a steady Taylor cone, approximately 7–10 kV [30]. 10. Allow the electrospray to dispel solution for as long as desired (see Note 3).

Co-Delivery Polymeric Poly(Lactic-Co-Glycolic Acid) (PLGA) Nanoparticles. . .

197

Fig. 2 SEM images of alterations to the electrospray solution and set up. Slight alterations in nanoparticle size and morphology are observed from changes to electrospray solution, syringe rate, and additional air flow. (a) Standard electrospray solution (25% ACN, 75% DMF and 10 μL TEAF) with the standard electrospray setup (no air flow, solution flow rate of 100 μL/h, (b) a reduction of 50 μL of ACN to 205 μL with the standard electrospray setup, (c) a 50% reduction of TEAF to 5 μL and the standard electrospray setup, (d) no TEAF and the standard electrospray setup, (e) 1.5 vol% PLGA instead of 2 vol% and the standard electrospray setup, (f) the standard electrospray solution sprayed at 50 μL/h instead of 100 μL/hr., (g) reduced ACN and reduced flow rate during spraying, (h) reduced TEAF and reduced flow rate, (i) no TEAF and reduced flow rate, and (j) reduced vol% PLGA and reduced flow rate. Scale bar is 1 μm 3.1.4 Nanoparticle Collection

1. After electrospraying onto the collection pan, remove the pan from the electrospray station and place in a fume hood to allow for any residual solvents to evaporate for ~8 h. 2. After drying, use a pipette to transfer 5 mL of 0.01% Tween 20 in water to the collection pan. Use a plastic razor blade to scrape the particles off the aluminum pan while submerged in the 0.01% Tween 20. 3. Use a pipette to remove the 0.01% Tween 20 containing nanoparticles and collect in a 50 mL centrifuge tube. 4. Repeat steps 2 and 3 at least twice, collecting the nanoparticle suspension into the same tube each time. Add deionized water to the tube to achieve a total volume of 50 mL. 5. Use a probe sonicator briefly to break up the nanoparticles and then pour the suspension through a 40-μm filter basket into a new 50 mL tube to remove large pieces of polymer. 6. Centrifuge for 15 min at 10,000 rcf to pellet particles larger than 1 μm. Carefully transfer the supernatant to a fresh centrifuge tube to keep the sub-1 μm particle population. 7. Pellet the remaining nanoparticles by centrifuging for 60 min at 10,000 rcf. Carefully remove the supernatant and then resuspend the nanoparticles in either ultrapure water or cell culture medium for characterization or use.

198

Catherine S. Snyder et al.

3.2 Nanoparticle Characterization 3.2.1 Scanning Electron Microscopy (SEM): Nanoparticle Size and Morphology

1. During a standard electrospray run, place a small piece of silicon or aluminum foil on the collection pan a few centimeters from where the center of the spray is deposited (see Note 4). 2. After waiting for enough particles to land on the SEM collection sample, use nonconductive forceps to remove it from the collection pan. Electrospraying can continue after removing the SEM collection sample. 3. Mount the SEM collection sample on a SEM stub using conductive carbon tape. 4. Sputter coat the nanoparticles with gold to prevent charging (see Note 5). 5. Image using a low voltage such as 3 kV (see Note 6). See Fig. 3b for representative SEM images of the nanoparticles.

3.2.2 Nanoparticle Tracking Analysis (NTA): Nanoparticle Size, Polydispersity, and Quantity

1. After nanoparticles have been electrosprayed, collect the nanoparticles and suspend them in water or cell culture medium, as described in Subheading 3.1.3. 2. Prepare four tenfold serial dilutions of the suspension and mix vigorously to ensure a homogenous suspension. 3. Analyze 1 mL of each diluted suspension with NanoSight NS300 (Malvern Analytical) and its NTA software to determine particle size and concentration. See Fig. 3a for representative results. Based on the efficiency of electrospraying and nanoparticle collection, the amount of nanoparticles can vary.

Fig. 3 Representative (a) nanoparticle size NTA results of (i) control nanoparticles, (ii) paclitaxel nanoparticles, (iii) sunitinib nanoparticles, and (iv) combination paclitaxel and sunitinib co-delivery nanoparticles. Representative (b) SEM images of (i) control nanoparticles, (ii) paclitaxel nanoparticles, (iii) sunitinib nanoparticles, and (iv) combination paclitaxel and sunitinib co-delivery nanoparticles

Co-Delivery Polymeric Poly(Lactic-Co-Glycolic Acid) (PLGA) Nanoparticles. . .

199

Run the four serially diluted samples on NTA. At least one sample should fall within the optimal range required by the Malvern NanoSight NS300. This will ensure accurate measurements of the size and concentration of nanoparticles. 4. Multiply the nanoparticle concentration obtained in Subheading 3.2.2, step 3 by the appropriate dilution factor from Subheading 3.2.2, step 2 to determine the concentration of particles in the stock solution. 3.2.3 Liquid Chromatography–Mass Spectrometry (LCMS): Drug Loading of Nanoparticles

1. Prepare an ACN solution consisting of 95% ACN, 5% ultrapure grade water, and 0.1% formic acid. 2. Create solutions of paclitaxel in ACN solution and sunitinib in ACN solution at specific concentrations to form a standard curve for each drug. 3. Using the particle concentration determined from NTA, pellet a known number of particles and remove the supernatant. 4. Dissolve the particles in ACN solution by vortexing or sonication and then filter with a 0.2 μm syringe filter to remove impurities for LCMS. 5. Perform LCMS using an Agilent 6520 Accurate Mass Q-TOF LC/MS with a 250-nm wavelength detector and Agilent Zorbax Eclipse Plus C18 (2.1 × 50 mm with 1.8 μm) to measure the paclitaxel and sunitinib solutions for the standard curve and the nanoparticle drug solution. 6. Quantify the amount of drug in the sample using the standard curve created in Subheading 3.2.3, step 1 (see Note 7). 7. Determine the nanoparticle drug loading by dividing the amount of drug quantified in Subheading 3.2.3, step 5 by the number of nanoparticles used in Subheading 3.2.3, step 2. See Table 2.

Table 2 Diameter, count, and drug loading of PLGA co-delivery nanoparticles

Nanoparticle

SEM diameter (nm)

NTA diameter (nm)

NTA nanoparticle Count

Drug loading (μg/NP)

Control

149 ± 29

133 ± 52

2.0 × 1010

–

Paclitaxel

135 ± 24

131 ± 43

2.3 × 1010

3.4 × 10-10

Sunitinib

139 ± 32

133 ± 48

7.3 × 1010

5.6 × 10-11

Paclitaxel + sunitinib

109 ± 36

145 ± 45

5.0 × 1010

1.0 × 10-9 + 2.9 × 10-11

200

Catherine S. Snyder et al.

3.3 Nanoparticle Drug Delivery and Cell Assays 3.3.1 Hetero-Spheroid Generation and Co-Delivery of Paclitaxel and Sunitinib

1. To prepare 384-well hanging drop plates, sonicate them in a bath sonicator for 20 min to remove dust, rinse with DI water, and place in a 0.1% pluronic acid bath for at least 8 h. Rinse with DI water and UV for 30–60 min on each side of the plate to sterilize [31–33]. 2. Remove MSC from 2D culture with Gibco 0.05% trypsin– EDTA (1X), neutralize with complete medium, and count cells with a hemacytometer. 3. Remove CSC from 2D culture with Gibco 0.25% trypsin– EDTA (1X), neutralize with complete medium, and count cells with a hemacytometer. 4. After cell counting and calculating cell density, resuspend the cells at the appropriate dilutions in the MSC-CSC cell culture medium to achieve 400 cells per 20 μL drop. (a) Mono-culture CSC will consist of 400 CSC per drop. (b) The hetero-spheroid co-culture will consist of 300 MSC and 100 CSC per drop. 5. Prepare 6-well plates with 5 mL DI water in each well. Place a hanging drop plate on top of the 6-well plate and add 1–2 mL DI water in the reservoir around the perimeter to prevent evaporation of hanging drops. 6. Plate 20 μL of the cell suspension prepared in Subheading 3.3.1, step 4 in each well of the hanging drop plate, leaving one row empty around the perimeter due to increased evaporation around the edges. 7. Place the lid of the 6-well plate on top of the hanging drop plate and carefully seal the edges of the 6-well/84-well plate sandwich with Parafilm to prevent evaporation. Place in an incubator at 37 °C with 5% CO2. 8. On day 2, image and feed spheroids by adding 2 μL of cell culture medium. See Fig. 4a, b. 9. On day 5, image the spheroids and add 2 μL of control (cell culture medium only) or co-delivery nanoparticles or free drug treatment solution to each well (see Note 8). See Fig. 4a, b. Drug treatment conditions can include the following: (a) Control: cell culture medium. (b) 10 μM paclitaxel (free drug): paclitaxel dissolved in cell culture medium (c) 5 μM sunitinib (free drug): sunitinib dissolved in cell culture medium (d) Combination (free drug): 10 μM paclitaxel and 5 μM sunitinib dissolved in cell culture medium. (e) Control nanoparticles: nanoparticles without drug suspended in cell culture medium.

Co-Delivery Polymeric Poly(Lactic-Co-Glycolic Acid) (PLGA) Nanoparticles. . .

201

Fig. 4 Optical microscope bright-field images of 3D CSC and CSC-MSC hetero-spheroids, demonstrating the size, shape, and tightly knitted microtissue structures. Day 3 images of (A) (i) CSC and (B) (ii) CSC-MSC hetero-spheroids. Day 5 images of (A) (iii) CSC and (B) (iv) CSC-MSC hetero-spheroids. (C) Day 7 images of (a) control CSC and (D) (e) CSC-MSC hetero-spheroids, 10 μM paclitaxel-treated (Pac) (C) (b) CSC and (D) (f) CSC-MSC hetero-spheroids, 5 μM sunitinib-treated (Sun) (C) (c) CSC and (D) (g) CSC-MSC hetero-spheroids and combination 10 μM paclitaxel + 5 μM sunitinib-treated (C) (d) CSC and (D) (h) CSC-MSC spheroids. Scale bar in all images is 400 μm

(f) 10 μM paclitaxel (nanoparticle): paclitaxel containing nanoparticles suspended in cell culture medium (g) 5 μM sunitinib (nanoparticle): sunitinib containing nanoparticles suspended in cell culture medium (h) Combination (co-delivery nanoparticle): paclitaxel and sunitinib containing co-delivery nanoparticles suspended in cell culture medium. 10. On day 7 (48 h after adding drug solution), image the different hetero-spheroid samples a final time. See Fig. 4c, d for representative images. These hetero-spheroids are now available to investigate changes in cell phenotypes due to co-delivery nanoparticles or other treatments using various molecular biology assays. Some assays of interest may include live/dead cytotoxicity assay, qPCR for relative mRNA expression, and scratch assay for migration potential.

4 Notes 1. There are many parameters that impact nanoparticle size and morphology, including polymer volume fraction, solution conductivity, solution viscosity, solution evaporation rate, syringe flow rate, and needle collection distance. To achieve the desired characteristics for a given application, these parameters will need to be adjusted [34].

202

Catherine S. Snyder et al.

2. The amount of nanoparticles produced with 1 mL of electrospray solution varies based on experimental conditions, but it is typically on the order of 1010 nanoparticles/mL (see Table 2). 3. Do not overspray nanoparticles such that they begin to land on top of each other in the collection pan; this will cause them to fuse. 4. If it is too close to the center, large polymer particles could cover it, and if it is too far away, a very low density of particles could be available for imaging. 5. SEM images with large areas of extreme brightness are due to buildup of charge that cannot be dissipated, which distorts the image. A thicker gold coating or a lower density of polymer nanoparticles when preparing SEM samples can reduce charging. 6. Tilting the stage can help visualize nanoparticles compared to when the stage is perpendicular to the electron beam. 7. Create a standard curve for LCMS by running several solutions of individual drug in solvent at known concentrations. This will identify the retention time of the drug, and a corresponding area under the curve (AUC) for each concentration. A calibration curve of concentration versus AUC will allow conversion of the AUC of an unknown drug concentration to a known value. 8. Drug delivery solutions for treatment in the hanging drop plates need to be an order of magnitude more concentrated than the final required dosage. This allows a 2 μL delivery of 100 μM paclitaxel to have a 10 μM final concentration within the 20 μL droplet. Nanoparticle drug delivery solutions will also need to be concentrated the same for single or co-delivery, but the number of nanoparticles might need to be adjusted based on the loading efficiency of the particles. Conduct NTA and LCMS on portions of the stock solution prior to treating spheroids with nanoparticles.

Acknowledgements This work is supported primarily by the American Cancer Society Research Scholar Award RSG-19-003-01-CCE (G.M.), the Office of the Assistant Secretary of Defense for Health Affairs through the Ovarian Cancer Research Program, under award no. W81XWH13-1-0134 (G.M.), W81XWH-16-1-0426 (G.M.), DOD Investigator Initiated award W81XWH-18-0346 (G.M.), Michigan Ovarian Cancer Alliance (G.M.), and NSF EFRI DChem (award number 2029139, G.M. A.T.). Research reported in this

Co-Delivery Polymeric Poly(Lactic-Co-Glycolic Acid) (PLGA) Nanoparticles. . .

203

publication was supported by the National Cancer Institute under award number P30CA046592 and by the National Institutes of Health under grant number UL1TR002240. Author Contributions Conceptualization, G.M., A.T.; formal analysis, C.S.S., K.M.B., T.R., G.M., A.T.; writing–original draft preparation, C.S.S., K.M.B., T.R., G.M.; writing–review and editing, C.S.S., K.M.B., T.R., G.M., A.T.; supervision, G.M., A.T.; funding acquisition, G.M., A.T. All authors have read and agreed to this version of the manuscript.

References 1. Ghoneum A, Afify H, Salih Z et al (2018) Role of tumor microenvironment in ovarian cancer pathobiology. Oncotarget 9:22832–22849 2. Al-Alem LF, Pandya UM, Baker AT et al (2019) Ovarian cancer stem cells: what progress have we made? Int J Biochem Cell Biol 107:92–103 3. Lupia M, Cavallaro U (2017) Ovarian cancer stem cells: still an elusive entity? Mol Cancer 16:64 4. Stack MS, Nephew KP, Burdette JE, Mitra AK (2018) The tumor microenvironment of high grade serous ovarian cancer. Cancers (Basel) 11:21 5. Steg AD, Bevis KS, Katre AA et al (2012) Stem cell pathways contribute to clinical chemoresistance in ovarian cancer. Clin Cancer Res 18: 869–881 6. Zhang B, Chen F, Xu Q et al (2018) Revisiting ovarian cancer microenvironment: a friend or a foe? Protein Cell 9:674–692 7. Silva IA, Bai S, McLean K et al (2011) Aldehyde dehydrogenase in combination with CD133 defines Angiogenic ovarian cancer stem cells that portend poor patient survival. Cancer Res 71:3991–4001 8. Landen CN, Goodman B, Katre AA et al (2010) Targeting aldehyde dehydrogenase cancer stem cells in ovarian cancer. Mol Cancer Ther 9:3186–3199 9. Zhang J, Guo X, Chang DY et al (2012) CD133 expression associated with poor prognosis in ovarian cancer. Mod Pathol 25:456 10. Curley MD, Therrien VA, Cummings CL et al (2009) CD133 expression defines a tumor initiating cell population in primary human ovarian cancer. Stem Cells 27:2875–2883 11. Sharrow AC, Perkins B, Collector MI et al (2016) Characterization of aldehyde dehydrogenase 1 high ovarian cancer cells: towards

targeted stem cell therapy. Gynecol Oncol 142:341–348 12. Kryczek I, Liu S, Roh M et al (2012) Expression of aldehyde dehydrogenase and CD133 defines ovarian cancer stem cells. Int J Cancer 130:29–39 13. Coffman LG, Choi Y-J, McLean K et al (2016) Human carcinoma-associated mesenchymal stem cells promote ovarian cancer chemotherapy resistance via a BMP4/HH signaling loop. Oncotarget 7:6916–6932 14. Cho JA, Park H, Lim EH et al (2011) Exosomes from ovarian cancer cells induce adipose tissue-derived mesenchymal stem cells to acquire the physical and functional characteristics of tumor-supporting myofibroblasts. Gynecol Oncol 123:379–386 15. Karnoub AE, Dash AB, Vo AP et al (2007) Mesenchymal stem cells within tumour stroma promote breast cancer metastasis. Nature 449: 557–563 16. Ding D-C, Liu H-W, Chu T-Y (2016) Interleukin-6 from ovarian mesenchymal stem cells promotes proliferation, sphere and Colony formation and tumorigenesis of an ovarian cancer cell line SKOV3. J Cancer 7:1815–1823 17. McLean K, Gong Y, Choi Y et al (2011) Human ovarian carcinoma–associated mesenchymal stem cells regulate cancer stem cells and tumorigenesis via altered BMP production. J Clin Invest 121:3206–3219 18. Raghavan S, Snyder CS, Wang A et al (2020) Carcinoma-associated mesenchymal stem cells promote chemoresistance in ovarian cancer stem cells via PDGF signaling. Cancers (Basel) 12:2063 19. Mitchell MJ, Billingsley MM, Haley RM et al (2021) Engineering precision nanoparticles for drug delivery. Nat Rev Drug Discov 20:101

204

Catherine S. Snyder et al.

20. Begines B, Ortiz T, Pe´rez-Aranda M et al (2020) Polymeric nanoparticles for drug delivery: recent developments and future prospects. Nanomaterials 10:1403 21. Danhier F, Lecouturier N, Vroman B et al (2009) Paclitaxel-loaded PEGylated PLGAbased nanoparticles: in vitro and in vivo evaluation. J Control Release 133:11–17 ˜ es S, Gaspar R (2002) 22. Fonseca C, Simo Paclitaxel-loaded PLGA nanoparticles: preparation, physicochemical characterization and in vitro anti-tumoral activity. J Control Release 83:273–286 23. Li M, Czyszczon EA, Reineke JJ (2013) Delineating intracellular pharmacokinetics of paclitaxel delivered by PLGA nanoparticles. Drug Deliv Transl Res 3:551–561 24. Mu L, Feng SS (2003) A novel controlled release formulation for the anticancer drug paclitaxel (Taxol®): PLGA nanoparticles containing vitamin E TPGS. J Control Release 86: 33–48 25. Alshetaili AS, Anwer MK, Alshahrani SM et al (2018) Characteristics and anticancer properties of Sunitinib malate-loaded poly-lactic-coglycolic acid nanoparticles against human colon cancer HT-29 cells lines. Trop J Pharm Res 17:1263–1269 26. Joseph JJ, Sangeetha D, Gomathi T (2016) Sunitinib loaded chitosan nanoparticles formulation and its evaluation. Int J Biol Macromol 82:952–958 27. Saber MM, Bahrainian S, Dinarvand R, Atyabi F (2017) Targeted drug delivery of Sunitinib

Malate to tumor blood vessels by cRGDchiotosan-gold nanoparticles. Int J Pharm 517:269–278 28. Dudek AZ, Nguyen S (2008) Safety of nab-paclitaxel plus Sunitinib: analysis of three cases. Anticancer Res 28:3099–3105 29. Anselmo AC, Mitragotri S (2019) Nanoparticles in the clinic: an update. Bioeng Transl Med 4:e10143 30. Zhang L, Huang J, Si T, Xu RX (2012) Coaxial electrospray of microparticles and nanoparticles for biomedical applications. Expert Rev Med Devices 9:595–612 31. Mehta P, Novak C, Raghavan S et al (2018) Self-renewal and CSCs in vitro enrichment: growth as floating spheres. In: Papaccio G, Desiderio V (eds) Methods in molecular biology: cancer stem cells: methods and protocols, pp 61–75 32. Bregenzer ME, Davis C, Horst EN et al (2019) Physiologic patient derived 3d spheroids for anti-neoplastic drug screening to target cancer stem cells. J Vis Exp 2019. https://doi.org/ 10.3791/59696 33. Bregenzer M, Horst E, Mehta P et al (2022) The role of the tumor microenvironment in CSC enrichment and Chemoresistance: 3D co-culture methods. Methods Mol Biol 2424: 217–245 34. Almerı´a B, Gomez A (2014) Electrospray synthesis of monodisperse polymer particles in a broad (60nm–2μm) diameter range: guiding principles and formulation recipes. J Colloid Interface Sci 417:121–130

Chapter 15 Isolating Circulating Cancer Stem Cells (CCSCs) from Human Whole Blood Carla Kantara and Pomila Singh Abstract Measuring circulating tumor cells (CTCs) or circulating cancer stem cells (CCSCs) in blood, which shed from primary tumors, is a noninvasive method to screen and/or diagnose patients with a high risk of developing metastatic cancers or relapse. Here, we describe an optimized and standardized laboratory method for isolating CCSCs from human blood samples, using cancer-specific stem cell biomarkers (Kantara et al., Lab Invest 95:100–112, 2015). When performing this technique, 0–1 circulating epithelial tumor cells/mL blood should be expected in patients free of malignant adenocarcinomas compared to over 3 circulating cancer stem cells/mL blood in patients positive for malignant adenocarcinomas (Kantara et al., Lab Invest 95:100–112, 2015). Key words Circulating cancer stem cells, Cancer stem cell biomarkers, Human blood, Buffy coat, Red blood cells, White blood cells, Depletion, Immunofluorescence, Spheroids

1

Introduction Cancer stem cells (CSCs) are believed to be resistant to currently available therapies. Measuring circulating tumor cells (CTCs) in blood of patients has emerged as a noninvasive diagnostic procedure for screening patients [1]. A number of noninvasive diagnostic assays/procedures have been, or are being, developed in order to either diagnose patients with metastatic disease or screen patients for relapse after treatment. Since CSCs are required to initiate metastatic tumors, our goal was to develop a method for identifying circulating CSCs in patients’ blood samples, using an epithelial CSC-specific biomarker [1]. This assay is based on the notion that metastatic cancer cells, including CSCs, going through epithelial–mesenchymal–transition (EMT), invade surrounding stroma and lymphatic/blood vessels by intravasation. Therefore, detection of CSCs in circulation will be diagnostic for ongoing metastasis and/or relapse [1, 2].

Federica Papaccio and Gianpaolo Papaccio (eds.), Cancer Stem Cells: Methods and Protocols, Methods in Molecular Biology, vol. 2777, https://doi.org/10.1007/978-1-0716-3730-2_15, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

205

206

Carla Kantara and Pomila Singh

Developing reproducible and accurate methods for detecting and isolating CTCs has remained a challenge. Epithelial cell membrane markers used for identification and separation of the rare population of CTCs may provide erroneous results, since CTCs undergoing EMT during metastasis are downregulated for the expression of epithelial cell markers [1, 3]. The rare CTCs represent a minute percentage of total blood cells in circulation, ranging from only 1 to 100 CTCs/mL blood, amongst >109 red and white blood cells [1, 4, 5]. Thus, a highly sensitive and specific bioassay is required to accurately detect and isolate the miniscule population of CTCs/CCSCs, which are expected to be present in the circulation of patients with metastatic cancer disease. We now know that CSCs are critically required for initiating and maintaining the growth of primary and metastatic tumors [1, 6–8]. In here, we present for the first time, a method for identifying circulating cancer stem cells (CCSCs) in the blood of colon cancer patients, using a combination of epithelial cell markers and cancer-specific stem cell markers, such as DCLK1, especially short isoform of DCLK1 [9–13], LGR5 [14, 15], and CD44 [16, 19]. Epithelial cell surface marker CK19 [17, 18] and epithelial cell adhesion membrane protein, EpCAM [19] were used to confirm epithelial origin of the isolated CCSCs; combining surface markers has been shown to increase accuracy of detecting CTCs [17, 18]. Using the methods and markers described in here, we report the development of relatively simple, but robust and reproducible assay(s) for detecting CCSCs in the blood of colon cancer patients [1]. CCSCs isolated from the blood of patients, positive for colonic adenocarcinomas (AdCAs), formed tumorospheres (spheroids) in vitro, confirming that the CCSCs, isolated by the method developed by our laboratory and outlined in this chapter, represented tumor-initiating cells (CSCs)which can potentially grow into metastatic tumors in vivo [1].

2

Materials

2.1 Antibodies, Cells, and Kits

1. Antibodies: anti-CD44 (Cell Signaling Technology, Danvers, MA), anti-DCLK1, anti-CD45, anti-EpCAM, anti-CK19, and anti-GPCR GPR49 (Lgr5) (Abcam, Cambridge, MA) (Sigma, St Louis, MO). 2. DAPI (4′,6-diamidino-2-phenylindole) (Sigma, St Louis, MO) for identifying nucleated cells. 3. Human colorectal cancer cells, induced to express green fluorescent protein (HCT-116-GFP cells) to analyze sensitivity and accuracy of isolating CSCs from blood samples.

Isolating Circulating Cancer Stem Cells from Human Blood

207

4. Alexa Fluor-594 and Alexa Fluor-488 coupled to secondary IgGs from Invitrogen (Carlsbad, CA) for purposes of dual labelling of cells during immunofluorescence analysis of cells, as described in methods. 5. HetaSep™ cocktail (Catalog #07806). 6. EasySep™ Human Whole Blood CD45 Depletion Kit (#18289). 7. RosetteSep™ Human CD45 Depletion Cocktail (#1522) may also be used as an alternative to the EasySep™ kit as needed. 2.2 Equipment and Disposable Plastic/ Glassware

1. Superfrost®/Plus B microscope slides. 2. CytoSpin III cytocentrifuge. 3. Cellometer™ Auto T4. 4. Epifluorescent microscope and the Metamorph, V6.0 software. 5. Ultralow-attachment surface culture dish. 6. BD Vacutainer EDTA Tubes (Lavender, 10.0 Draw Volume, 18 mg, BD Hemogard Closure tubes). 7. 5 mL Corning™ Falcon™ round-bottom polystyrene test tubes. 8. CO2 incubator. 9. Poly-L-lysine. Cover glass 0.13–0.17 mm Pearl. 10. Cellometer™ Auto T4 from Nexcelom Bioscience.

3

Methods

3.1 Procurement of Human Whole Blood Samples

This chapter highlights the example of isolating cancer stem cells in the blood of patients positive for colorectal adenocarcinomas. This method could be applied to blood from patients with other adenocarcinomas, using CSC-specific biomarkers as identified in the literature for the indicated cancers. 1. Obtain blood samples from consented patients scheduled for surgical removal of their colonic adenocarcinomas (AdCA). Control blood samples may be collected from consented patients scheduled for screening colonoscopy, who are negative for adenomatous growths. 2. Collect approximately 30–45 mL of blood in Vacutainer EDTA lavender-top tubes (10.0 mL). EDTA serves as an anticoagulant. You may need several tubes. 3. Invert blood tubes immediately upon collection up to ten times to avoid coagulation of blood samples. Keep tubes on ice until use. You may also store the samples up to 24 h at 2–4 °C (see Note 1).

208

Carla Kantara and Pomila Singh

3.2 Depletion of Red Blood Cells and White Blood Cells from Human Whole Blood Samples

1. First, deplete the red blood cells (RBCs) from the whole blood samples obtained from the patients. Use the HetaSep™ cocktail to separate the RBCs from nucleated cells. The gravity sedimentation, followed by centrifugation and processing steps, is to be conducted as per the manufacturer’s recommendations (see Note 2). 2. Briefly, add 1-part HetaSep™ solution to 5 parts of whole blood. The process is quick and within 20 min you will see three layers forming in the tube. The top yellow layer is known as the plasma, the middle blurry layer is known as the buffy coat, and the bottom red layer is where the RBCs will have aggregated (Fig. 1) (see Note 3). 3. Allow the blood sample, treated as in step 2, to settle until approximately half the tube is composed of the plasma layer. You may expedite the sedimentation process by incubating the tube at 37 °C or centrifuging at 90 g at room temperature (15–25 °C) with the centrifuge brake off. For detailed explanation, please refer to the manufacturer’s sedimentation processing steps.

Fig. 1 Separation of whole blood into three layers: plasma, buffy coat, and red blood cells. Approximately 20 min after adding 1-part HetaSep™ solution to 5 parts of whole blood, RBCs will aggregate at the bottom of the tube. The plasma and buffy coat will contain the nucleate cells, including the CCSCs

Isolating Circulating Cancer Stem Cells from Human Blood

209

4. Gently transfer the plasma and buffy coat layers, by pipetting out from the processed tubes (as described in steps 2 and 3), into new tubes. The two layers thus transferred into new tubes contain all the nucleated cells, including CCSCs. 5. Next, deplete white blood cells (WBCs) from the plasma and buffy coat layers using the negative selection EasySep™ kit Human CD45 Depletion Kit, as briefly described in step 6 (see Note 4). The CD45 depletion kit removes CD45+ cells, a known biomarker for WBCs, from the samples. The processing steps are to be conducted as per the manufacturer’s recommendations. See Note 5 for a potential alternative to the EasySep™ kit if needed. 6. Add the depletion EasySep™ cocktail to the plasma + buffy coat sample tubes. This cocktail contains CD45 antibody conjugated to magnetic particles. 7. Mix and incubate at room temperature for 3 min, to allow the antibody complexed magnetic beads to bind the WBCs in the sample. 8. Place the sample tube without a lid in the EasySep™ magnet (provided with the kit). Allow the tube to sit in the magnet, at room temperature for 5 min. 9. In a single continuous motion, invert the magnet and tube to pour the final elution in a new 5 mL round-bottom polystyrene test tubes. The final elute will contain enriched nucleated epithelial cells, including CCSCs if any, free of RBCs and WBCs. 3.3 Analyzing Residual Cells in the Final Elution from the Human Blood Samples

The human whole blood samples, processed for removing RBCs and WBCs, as described above, are subjected to cytocentrifugation for analyzing the presence of CCSCs in the samples. 1. Prior to cytospinning, coat glass slides with poly-L-lysine to ensure that the cytospun cells remain attached to the slides. 2. For the purpose of cytocentrifugation of the samples, cytospin the final elution containing epithelial cells onto Superfrost®/ Plus B microscope slides using a CytoSpin III cytocentrifuge using no more than 200 μL of the elute per slide, and spin at 300 g for 5 min. 3. Fix cells, spun onto slides, using an ice-cold solution of methanol and ethanol at 1:1 ratio, for 10 min at room temperature. 4. Process slides containing the fixed cells for immunofluorescence (IF) staining of the stem cell biomarkers of interest (see Note 6), as described [1, 11, 20]. Briefly, after fixing the cells, stain the cells with primary antibodies against the stem cell markers of interest, as per the manufacturer’s instructions. To

210

Carla Kantara and Pomila Singh

Fig. 2 Presence of CD44+/CD45- cells in the blood of patients with colon adenocarcinomas. Bar graphs illustrating (a) the number of recovered CD45+ cells/mL in the human blood, who were either free of colonic growths (normal) or positive for adenocarcinomas (AdCAs); IF images of remaining CD45+ cells collected in the final elution are shown above the bar graphs. (b) Number of CD44+ cells/mL in the blood of normal versus AdCA cancer samples. (c) Number of cells co-localizing for CD45/CD44 per mL of blood. Representative IF staining of circulating cells stained with indicated markers is shown above bar graphs, after negative selection of the samples. (d) Number of CD44+ cells/mL, which do not co-localize with CD45 in the blood of normal vs AdCA cancer samples; representative IF images are shown above bar graphs. Each bar graph = mean ± SEM of data from 5 to 7 patients analyzed in triplicate as described in [1]. * = P < 0.05 versus normal values (Repurposed from ref. 1 with permission)

detect the stained cells via immunofluorescence microscopy, use secondary antibodies tagged to green or red fluorescent proteins [1, 11, 20]. Use CD45 antibody for IF staining of slides, to detect residual WBCs, and use DAPI (4′,6-diamidino-2-phenylindole), as a counterstain for identifying nucleated (epithelial/WBC) cells (Fig. 2) [1]. 5. Use coverslips after fixing and staining the cells on the slides as described in step 5. The isolation methodology for CCSC does not guarantee 100% depletion of RBCs and WBCs (see Note 7). Therefore, in addition to staining for CCSC such as CD44 (see Note 8). DCKL1 and LGR5 (see Note 9), we recommend co-staining the slides with CD45 antibody to avoid falsepositive readings, given that CD45 serves as a robust marker for WBCs.

Isolating Circulating Cancer Stem Cells from Human Blood

211

Fig. 3 Comparative tables enumerating (a) the average number of CTCs/mL or CCSC/mL sample from normal versus AdCA cancer blood samples with contaminating CD45, (b) average number of CTCs/mL or CCSC/mL sample from normal versus AdCA cancer blood samples without contaminating CD45, and (c) average number of CTCs/mL or CCSC/mL co-expressing the specific markers in normal versus AdCA cancer blood (Repurposed from ref. 1 with permission)

6. Analyze IF staining using an epifluorescent microscope and the Metamorph, V6.0 software, as described [11, 20] to assess the number and composition of CCSCs present in the final elute (see Note 10) (Fig. 3). 3.4 Verifying Efficient Depletion Process from Human Blood Samples

To ensure the efficiency of the EasySep™ human whole blood CD45 depletion kit, you can conduct a small experiment whereby a normal blood sample is spiked with a predefined number of tagged epithelial cells. For example, you can spike a normal human blood sample with 10–500 colon cancer HCT-116-GFP cells. The green fluorescent protein (GFP) is a biological marker which when excited with 488-nm laser beam is seen as green fluorescence under the microscope. One may use any other labeled cell type as needed. This method will allow you to assess the percentage of GFP-positive colon cancer epithelial cells, which are

212

Carla Kantara and Pomila Singh

retained in the final elution after depleting RBCs and WBCs from the blood samples, as described above. Briefly, the steps are as follows: 1. Harvest HCT-116-GFP cells from culture plates by scraping and transferring the culture medium containing the scraped cells into 15–50 mL plastic centrifuge tubes. Centrifuge the cells down at 500 rpm and add 1 mL culture medium (as recommended by ATCC or the source of the cancer cell line). Gently rotate the cells in the tube and mix the cells by gently pipetting the solute (medium containing the tagged cells), to resuspend the tagged cells homogenously in a defined unit of the medium. 2. Remove an aliquot to confirm viability of cells by using the trypan blue exclusion method using a cell counter (as an example) and determine the number of cells present per mL of the resuspended solution. 3. Within 2–4 h of collecting and resuspending the tagged cell samples, use an aliquot containing 0–500 HCT-116-GFP cells to spike normal blood samples (which were collected no more than 4 h before spiking). The concentration of tagged cells in the spiked blood samples should range from zero (control group) to 500 cells/mL. 4. Gently resuspend the tagged cells homogenously in the spiked blood samples and subject the samples to negative selection for RBCs using the HetaSep™ kit (as described in the Methods section). 5. Subject spiked blood samples, free of RBCs, to negative selection for CD45+ WBCs using the EasySep™ kit, as per the manufacturer’s recommendations. 6. Once the spiked human blood sample is negatively selected for RBCs and WBCs, cytospin the residual cells in the final elute at 300 g for 5 min onto Superfrost®/Plus B microscope slides using a CytoSpin III cytocentrifuge as described in Methods. 7. Fix cells and process cells for immunofluorescence (IF) staining and use DAPI (4′,6-diamidino-2-phenylindole), as a counterstain for identifying nucleated (epithelial/WBC) cells. 8. Analyze slides using a fluorescent microscope and assess the presence of HCT-116-GFP cells by measuring the number of green fluorescent GFP-positive cells on the slides and extrapolate to calculate the total number of HCT-116-GFP cells per mL of spiked blood samples, to determine percent recovery of the tagged cells from the spiked blood samples [1]. Immunofluorescence images in Fig. 4 show the recovered cells in spiked versus non-spiked normal blood samples (Fig. 4).

Isolating Circulating Cancer Stem Cells from Human Blood

213

Fig. 4 Efficiency and optimization of the EasySep Kit™ for isolation of circulating tumor cells. (i) Recovery of GFP-expressing epithelial colon cancer cells (HCT116-GFP) from indicated blood samples, after spiking the samples with the cells as described in [1]. Spiked samples were depleted of RBCs and WBCs using the EasySep Kit™, before analyzing % recovery, and are shown as bar graphs. (ii) IF images of recovered cells in non-spiked and spiked blood samples. Images were taken at 10× and 40× magnification. (Repurposed from Ref. [1] with permission)

3.5 Confirming Presence of CCSCs in the Final Elution

To confirm the presence of tumor-initiating CCSCs in the final elution, the eluted cells can be seeded in low-attachment culture plates to grow as spheroids to evaluate the stemness of the cells as an indirect measure of tumor-initiating potential of the cells positive for the stem cell biomarker used for IF analysis [1] (see Note 10).

4 Notes 1. At no point in the experiment was the blood diluted. It is important to note that the common practice in the field is to express the number of tumor cells/mL of blood, and we have followed this tradition in order to be able to compare our results with that of others. 2. The HetaSep™ protocol requires only 50 μL of blood and takes about 20 min, but you will need up to 45 mL of whole blood collection to isolate enough CCSCs in the final elution.

214

Carla Kantara and Pomila Singh

3. The middle, buffy coat layer will be less distinct in older samples. Please note that the age of the blood sample affects the speed and extent of the RBC sediment. 4. The EasySep™ Human CD45 Depletion Kit (catalog #18259) was recently replaced with the updated EasySep™ Human CD45 Depletion Kit II (catalog #17989). The processing steps remain the same. 5. The RosetteSep™ Human CD45 Depletion Cocktail kit may also be used to deplete hematopoietic/white blood cells from the blood sample. The RosetteSep™ Human CD45 Depletion Cocktail kit is designed to enrich for epithelial circulating tumor cells (CTCs) from whole blood by depleting CD45+ cells. However, the combined use of EasySep™ Human CD45 Depletion Kit and HetaSep™ gave the best results for human whole blood [1]. 6. Immunofluorescence staining by its very nature is an extremely sensitive method for detecting the presence of an antigen on the cell surface, or within the cell, and can be altered by changing the dilution of the antibody working solution. The dilutions used were determined to be optimal, using cells positive or negative for the expression of the corresponding transcripts for the markers used in this experiment. For more information on the specific antibodies and specific dilutions used to demonstrate the presence of the specific stem cell markers, please refer to our reference [20]. 7. The isolation methodology for CCSC does not guarantee 100% depletion of RBCs and WBCs. For example, you will be analyzing the cells in the final elute on the slides with specific antibodies against stem cell biomarkers (purchased from commercial sources, and pretested for specificity and sensitivity as described in [1]), using IF methodology. If you find a cell positive for both CD44 and CD45, you will not count it as a CCSC, since it represents a WBC. However, a cell positive for CD44 and negative for CD45 will be considered a CCSC. Similarly, mature RBCs are known to lack a nucleus. DAPI staining is known to stain for cell nucleus, and therefore, if cells in the final elute are found to be negative for DAP1 staining, the cell(s) will be considered residual RBCs. 8. CD44 was found to be a robust marker for detection of CCSCs, as long as the cells were CD45 negative. When performing this isolation process, we found, on average, 82 ± 10 CD45-positive cells present per mL of cancer blood sample, compared to 70 ± 10 CD45-positive cells present per

Isolating Circulating Cancer Stem Cells from Human Blood

215

mL of normal blood sample, suggesting that a minute population of CD45-positive WBCs remains as a contaminant in the final elution. Since CD44 is also expressed by CD45-positive WBCs [21], we found that on an average, 50 CD44-positive cells/mL was present in the blood of normal patients compared to 85 CD44-positive cells/mL in the blood samples of adenocarcinoma patients [1]. Additionally, approximately 98% of CD44-positive cells in the blood from normal patients co-stained with CD45, suggesting that almost all the CD44positive cells present in the blood of normal patients were WBCs. However, only approximately 70% of CD44-positive cells co-stained with CD45 in the cancer blood samples suggesting that approximately 30% of CD44-positive cells circulating in the cancer blood samples represents CCSCs. Numerically, an average of 9 cells/mL were CD44-positive/ CD45-negative cells in adenocarcinoma cancer blood samples, while 0 cells/mL were CD44-positive and CD45-negative in normal blood samples [1]. 9. We found that DCKL1 and LGR5 were robust markers for detection of CCSCs in the blood, while EpCAM was not an ideal marker for detection of CCSCs [1]. Our experiments showed that CCSCs isolated from the cancer blood samples were positive for CCSC markers Lgr5/DCLK1, while normal samples were negative. This confirms that normal blood lacks CCSCs, while cancer samples are positive for cancer stem cells which likely have the potential for initiating metastatic tumors in vivo [1, 22]. Additionally, epithelial cells enriched from the cancer blood samples started growing as distinct spheroids by day 14, while cells remaining in the elute from normal samples did not form spheroidal structures (Fig. 5a), even after 1 month of culture [1]. Furthermore, on day 25 of culture, normal and cancer cells were collected from the respective cell culture plates, pelleted by centrifugation, and processed for Western blot analysis. The spheroidal cells from cancer blood sample were positive for Lgr5/DCLK1, while normal samples were negative (Fig. 5b). 10. Analyses of CCSC numbers are based on antibody staining of the cells (Fig. 4). The specificity and sensitivity of the antibodies are provided by the commercial sources of the antibody and can be obtained online, based on the catalog numbers.

216

Carla Kantara and Pomila Singh

Fig. 5 Formation of spheroids, in vitro, from circulating tumor cells (CTCs/CCSCs), isolated from AdCA blood samples. (a) Representative images of spheroids formed from enriched circulating epithelial cells, isolated from blood of normal and AdCA samples. Images were taken at 4× and 40× magnification at days 3, 14, and 25 after seeding the cells. (b, (i)) Western blot analysis demonstrating increased expression of LGR5 and DCLK1 in spheroids from CTCs/CCSCs isolated from a representative AdCA blood sample, compared to that obtained from a representative normal blood sample. (b, (ii)) Bar graphs demonstrating percent change in ratio of target proteins to β-actin (n = 3 normal blood samples; n = 3 AdCA blood samples). (Repurposed from ref. 1 with permission)

Isolating Circulating Cancer Stem Cells from Human Blood

217

Acknowledgements This work was supported by NIH grants CA97959 and CA114264 to Pomila Singh and NASA grants NNX09AM08G and NNJ04HD83G to Robert Ullrich. References 1. Kantara C, O’Connell MR, Luthra G, Gajjar A, Sarkar S, Ullrich RL, Singh P (2015) Methods for detecting circulating cancer stem cells (CCSCs) as a novel approach for diagnosis of colon cancer relapse/metastasis. Lab Investig 95:100–112 2. King MR (2012) Rolling in the deep: therapeutic targeting of circulating tumor cells. Front Oncol 2:184 3. Bonnomet A, Syne L, Brysse A, Feyereisen E, Thompson EW, Noe¨l A, Foidart JM, Birembaut P, Polette M, Gilles C (2012) A dynamic in vivo model of epithelial-to-mesenchymal transitions in circulating tumor cells and metastases of breast cancer. Oncogene 31:3741–3753 4. Allard WJ, Matera J, Miller MC, Repollet M, Connelly MC, Rao C, Tibbe AG, Uhr JW, Terstappen LW (2004) Tumor cells circulate in the peripheral blood of all major carcinomas but not in healthy subjects or patients with nonmalignant diseases. Clin Cancer Res 10: 6897–6904 5. Miller MC, Doyle GV, Terstappen LW (2010) Significance of circulating tumor cells detected by the cellsearch system in patients with metastatic breast colorectal and prostate cancer. J Oncol 2010:617421 6. Reya T, Morrison SJ, Clarke MF, Weissman IL (2001) Stem cells, cancer, and cancer stem cells. Nature 414:105–111 7. Li F, Tiede B, Massague´ J, Kang Y (2007) Beyond tumorigenesis: cancer stem cells in metastasis. Cell Res 17:3–14 8. Marotta LL, Polyak K (2009) Cancer stem cells: a model in the making. Curr Opin Genet Dev 19:44–50 9. May R, Riehl TE, Hunt C, Sureban SM, Anant S, Houchen CW (2008) Identification of a novel putative gastrointestinal stem cell and adenoma stem cell marker, doublecortin and CaM kinase-like-1, following radiation injury and in adenomatous polyposis coli/

multiple intestinal neoplasia mice. Stem Cells 26:630–637 10. Nakanishi Y, Seno H, Fukuoka A, Ueo T, Yamaga Y, Maruno T, Nakanishi N, Kanda K, Komekado H, Kawada M, Isomura A, Kawada K, Sakai Y, Yanagita M, Kageyama R, Kawaguchi Y, Taketo MM, Yonehara S, Chiba T (2013) Dclk1 distinguishes between tumor and normal stem cells in the intestine. Nat Genet 45:98–103 11. Sarkar S, Kantara C, Ortiz I, Swiercz R, Kuo J, Davey R, Escobar K, Ullrich R, Singh P (2012) Progastrin overexpression imparts tumorigenic/metastatic potential to embryonic epithelial cells: phenotypic differences between transformed and nontransformed stem cells. Int J Cancer 131:E1088–E1099 12. O’Connell M, Sarkar S, Luthra G, Okugawa Y, Toiyama Y, Gajjar A, Qiu S, Goel A, Singh P (2015) Epigenetic changes and alternate promoter usage by human colon cancers for expressing DCLK1-isoforms: clinical implications. Sci Rep 5:14983 13. Sarkar S, O’Connell M, Okugawa Y, Lee B, Toiyama Y, Kusunoki M, Daboval R, Goel A, Singh P (2017) FOXD3 regulates CSC marker, DCLK1-S, and invasive potential: prognostic implications in colon cancer. Mol Cancer Res 15:1678–1691 14. Schepers AG, Snippert HJ, Stange DE, van den Born M, van Es JH, van de Wetering M, Clevers H (2012) Lineage tracing reveals Lgr5+ stem cell activity in mouse intestinal adenomas. Science 337:730–735 15. Kemper K, Prasetyanti PR, De Lau W, Rodermond H, Clevers H, Medema JP (2012) Monoclonal antibodies against Lgr5 identify human colorectal cancer stem cells. Stem Cells 30:2378–2386 16. Park YS, Huh JW, Lee JH, Kim HR (2012) shRNA against CD44 inhibits cell proliferation, invasion and migration, and promotes apoptosis of colon carcinoma cells. Oncol Rep 27:339–346

218

Carla Kantara and Pomila Singh

17. Wang L, Wang Y, Liu Y, Cheng M, Wu X, Wei H (2009) Flow cytometric analysis of CK19 expression in the peripheral blood of breast carcinoma patients: relevance for circulating tumor cell detection. J Exp Clin Cancer Res 28:57 18. Katseli A, Maragos H, Nezos A, Syrigos K, Koutsilieris M (2013) Multiplex PCR-based detection of -circulating tumor cells in lung cancer patients using CK19, PTHrP, and LUNX specific primers. Clin Lung Cancer 14: 513–520 19. Gires O, Klein CA, Baeuerle PA (2009) On the abundance of EpCAM on cancer stem cells. Nat Rev Cancer 9:143

20. Kantara C, O’Connell M, Sarkar S, Moya S, Ullrich R, Singh P (2014) Curcumin promotes autophagic survival of a subset of colon cancer stem cells, which are ablated by DCLK1siRNA. Cancer Res 74:2487–2498 21. Johnson P, Ruffell B (2009) CD44 and its role in inflammation and inflammatory diseases. Inflamm Allergy Drug Targets 8:208–220 22. Pizon M, Zimon D, Carl S, Pachmann U, Pachmann K, Camara O (2013) Heterogeneity of circulating epithelial tumour cells from individual patients with respect to expression profiles and clonal growth (sphere formation) in breast cancer. Ecancermedicalscience 7:343

Chapter 16 Generation of Cancer Stem Cells by Co-Culture Methods Biswajit Das and Chanakya Nath Kundu Abstract Cancer stem cells (CSCs) exhibit intricate regulatory dynamics within the tumor microenvironment, involving interactions with various components like mesenchymal stem cells (MSCs), adipocytes, cancerassociated fibroblasts (CAFs), endothelial cells, tumor-associated macrophages (TAMs), and other immune cells. These interactions occur through complex networks of cytokines, inflammatory factors, and several growth factors. Diverse techniques are employed to generate CSCs, including serum-free sphere culture, chemotherapy, and radiation therapy. A novel approach to generate CSCs involves co-culturing, wherein recent research highlights the role of secreted factors such as inflammatory cytokines from MSCs, CAFs, and TAMs in inducing CSC-like characteristics in cancer cells. While the co-culture method shows promise in generating CSCs, further investigations are needed to comprehensively establish this process. This chapter focuses on establishing a co-culture-based technique for generating CSCs by combining cancer cells with TAMs and CAFs, elucidating the intricate mechanisms underlying this phenomenon. Key words Cancer cells, Tumor associated macrophages, Cancer associated fibroblasts, Co-culture, Cancer stem cells

1

Introduction Cancer stem cells (CSCs) are the key player of tumor initiation, progression, resistance, and recurrence [1]. These CSCs maintain a delicate balance with their surrounding microenvironments, and the expression of the CSC phenotype is under intricate control by both intrinsic cellular factors and extrinsic signals originating from neighboring cells or the stromal environment [2]. Leukemia as well as various solid tumors such as glioma, pancreatic ductal adenocarcinoma (PDAC), breast cancer, non-small cell lung cancer (NSCLC), hepatocellular carcinoma, and colorectal cancer (CRC) has been shown to harbor CSC populations [2–7]. From an experimental standpoint, CSC enrichment techniques encompass the serum-free sphere culture method, chemotherapy-based techniques, and radiation therapy [8]. Additionally, CSCs can be isolated based on distinctive surface markers like CD-133 or CD-44 using

Federica Papaccio and Gianpaolo Papaccio (eds.), Cancer Stem Cells: Methods and Protocols, Methods in Molecular Biology, vol. 2777, https://doi.org/10.1007/978-1-0716-3730-2_16, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

219

220

Biswajit Das and Chanakya Nath Kundu

techniques such as fluorescence-activated cell sorting (FACS) [9]. A more recent advancement in the pursuit of CSC enrichment involves co-culturing cancer cells with tumor-associated macrophages (TAMs), cancer-associated fibroblasts (CAFs), and mesenchymal stem cells (MSCs) [2, 10–12]. Key factors that enhance stemness, including interleukin-6 (IL-6), interleukin-8 (IL-8), and C-C motif chemokine ligand 5 (CCL-5), play crucial roles in both cancer stem cell (CSC) regulation and the metastatic behavior of tumors [11, 13]. Recent research has demonstrated that cancer-associated fibroblasts (CAFs) drive the generation of CSCs, tumor growth, and invasion through the production of stemness factors like IL-6, IL-8, CXCL1 and 12, HGF, and TGFβ [2]. When introduced into the tumor microenvironment (TME), MSCs can induce cancer stemness property by releasing some specific paracrine factors or altering into pro-stemness CAFs [2]. The tumor microenvironment (TME) is a specialized milieu that fosters the enrichment of the CSC population [1]. This microenvironment consists of chemokines, cytokines, and various cell types, including endothelial cells, fibroblasts, and immune cells, with TAMs being a pivotal component [1, 14]. TAMs, the most prevalent immune cells within the tumor cell population, exhibit distinct polarization states—pro-inflammatory stage, M1, and suppressive alternatively activated stage, M2—dictated by the local microenvironment [14]. Their polarization is influenced by cytokines, with interferon-c, bacterial lipopolysaccharide (LPS), and TNF-a triggering classical activation (M1), while macrophagecolony-stimulating factor (M-CSF), IL-4, and IL-13 induce alternative activation stage, M2 [15, 16]. Studies indicate that TAMs can induce the generation of CSCs, encouraging their proliferation and promoting angiogenesis and metastasis [14] (Fig. 1). Numerous reports suggest that co-culturing cancer cells with TAMs, MSCs, CAFs, and other TME components increases CSC populations or stimulates cancer stemness (Table 1). In this chapter, we elaborate on two prevalent co-culture methods involving cancer cells and TAMs, as well as cancer cells and CAFs, aiming to enhance CSC enrichment.

2

Materials

2.1 Co-Culture of Cancer Cells with TAMs

1. Cancer cell line (e.g., oral cancer cell line H-357, Sigma Aldrich, USA, cat # 06092004). 2. Human leukemic monocyte cell line (e.g., THP-1, ATCC, cat # ATCC®TIB-202 TM). 3. Basic cell culture reagents, DMEM-F12 (50:50, v/v) supplemented with (a) 2 mM glutamine, (b) 10% fetal bovine serum

Generation of Cancer Stem Cells by Co-Culture Methods

221

Fig. 1 TAMs and CAFs secrete several pro-stemness factors to stimulate cancer stem-like properties of cancer cells and self-renewal of cancer stem-like cell

(FBS), (c) 1% antibiotic (10 mg/mL of streptomycin and 100 U/mL of penicillin, in 0.9% normal saline), and (d) 0.5 mg/mL of hydrocortisone (only for the culture of oral cancer cells). 4. Phorbol-12-myristate-13-acetate (PMA) (e.g., cat no. p8139]. 5. Lipopolysaccharide (LPS) obtained from Escherichia coli (cat no. L2880; lot no. 025M4040V). 2.2 Co-Culture of Cancer Cells with CAFs

1. Oral cancer cell lines (e.g., H-357, cat no. 06092004) can be obtained from Sigma Aldrich, USA. 2. The human dermal fibroblast cells (HFBs) can be obtained from HIMEDIA (cat no. CL011). 3. All the basic cell culture reagents as mentioned in the above section. HFB should be grown in DMEM supplemented with the abovementioned reagents (2.1). 4. Platelet-derived growth factor-B (PDGF-B) purified protein (e.g., cat no. CYT-501). 5. Transforming growth factor (TGF-β) purified protein (e.g., cat no. CYT-716).

222

Biswajit Das and Chanakya Nath Kundu

Table 1 Stemness-inducing factor secreted by the components of TME such as TAMs, MSCs, and CAFs Pro-stemness Sl. factors secreted no by TAMs Cancer types

Role

Refs.

Shown to increase cancer stemness in esophageal cancer cells by secreting CCL2

[17]

1

CCL2

Esophageal cancer

2

MFG-E8

Colorectal cancer, breast Activates Stat3 and sonic hedgehog pathways in CSCs, contributing to cancer, melanoma, and anticancer drug resistance lung cancer

3

IL-6

Oral squamous cell carcinoma

Regulates cancer stemness in oral squamous cell carcinoma through the JAK/STAT-3 signaling axis

4

IL-10

Ovarian cancer

Upregulates WNT and increase cancer stemness.

[19]

Non-small cell lung cancer

Upregulates JAK-1/STAT1/NF-κβ/ Notch-1 signaling

[20]

5

[18]

6

CCL5

Prostate cancer

Enhances cancer stemness by upregulating [21] β-catenin and STAT3 signaling

7

CCL8

Glioblastoma

Upregulates CCR1/CCR5/ERK1/2

[22]

8

CCL-17

Hepatocellular carcinoma

Upregulates WNT/β-catenin signaling

[23]

9

CXCL-8

Breast cancer

Increases the cancer stemness by regulating CXCR-2

[21, 24]

10 IL-1β

Glioblastoma

Upregulates the expression of P38 MAPK [21]

11 TGF-β

Pancreatic cancer

Upregulates TGFβ-1/Smad2/3 axis to increase cancer stemness

[25]

12 TNF-α

Renal cell carcinoma

Increases the expression of NF-kβ

[26]

13 EGF

Breast cancer

Regulates the EGFR/STAT3/Sox2 signaling axis

[27]

14 IGF4

Thyroid cancer

Upregulates PI3K/AKT/mTOR axis

[28]

Sl. Pro-stemness factors no secreted by MSCs

Cancer types

Role

Refs.

1

CCL-5

Breast cancer

Induces cancer stemness and tumor dissemination

[29]

2

CXCL-7

Breast cancer

[30] Increases the number of CSCs and tumor growth through the IL-6-mediated pathway (continued)

Generation of Cancer Stem Cells by Co-Culture Methods

223

Table 1 (continued) Sl. Pro-stemness factors no secreted by MSCs

Cancer types

Role

Refs.

3

CXCL-10

Pancreatic ductal adenocarcinoma

Promotes CSC-like properties and increases the number of MSCs through the CXCR-3 pathway

[31]

4

IL-6

Breast cancer

Maintains the CSC population and stimulates cancer progression through the β-catenin pathway

[32]

5

JAG-1

Pancreatic ductal adenocarcinoma

Maintains CSC population and growth [33]

6

PGE-2

Colorectal cancer

[2, 32] Promotes the generation of CSCs by increasing the expressions of IL-6, IL-8, and CXCL-1

Pro-stemness Sl. factors secreted no by CAFs

Cancer types

Role

Refs.

1

CXCL-1

Pancreatic ductal adenocarcinoma

Increases the cancer stemness property by upregulating the IL-1α/JAK/STAT pathway

[2, 34]

2

CXCL-2

Colorectal cancer

Upregulates the expression of CSC markers through upregulation of the IL-1α/JAK/STAT pathway

[34]

3

CXCL-12

Colorectal cancer

Promotes the expression of CSC markers through regulating the Wnt/CD44v6 and PI3K pathway

[2, 35]

4

HGF

Colorectal cancer

Promotes CSC-like properties and increases the expression of CSC-specific markers

[2, 35]

5

IGF-II

Non-small cell lung cancer

Promotes the conversion of cancer cells into CSC-like cells through the PI3K, Wnt, hedgehog, IGF1R, EMT, and TGF-β pathway

[36]

6

IL-17A

Colorectal cancer

Enhances the self-renewal property of CSCs and cancer progression

[37]

7

IL-6

Pancreatic ductal adenocarcinoma, breast cancer

Promotes and maintains CSCs

[2]

8

IL-8

Breast cancer, pancreatic ductal adenocarcinoma

Enhances stemness property

[2]

(continued)

224

Biswajit Das and Chanakya Nath Kundu

Table 1 (continued) Pro-stemness Sl. factors secreted no by CAFs

Cancer types

Role

Refs.

Pancreatic ductal adenocarcinoma

Activates the CSCs related signaling such as Wnt, hippo, and STAT-3

[38]

10 Nodal

Pancreatic ductal adenocarcinoma

Directly binds to Alk-4/-7 to induce the stemness property through upregulation of the nodal/activin pathway

[2, 39, 40]

11 OPN

Colorectal cancer

Induces expression of CSC marker through the WNT /CD44v6 and PI3K pathway

[2, 41]

12 PGE-2

Breast cancer

Increases the secretion of IL-6 and promotes the generation of CSCs through upregulation of the NF-κB pathway

[42]

13 TGF-β

Colorectal cancer

Enhances the expression of CSC-specific markers through activating the WNT/ CD44v6 and PI3K pathway

[35]

14 TGF-β2

Colorectal cancer

Promotes trans-differentiation of the cancer cells into CSCs through the hedgehog/ GLI-2 pathway

[43]

15 WNT16B

Prostate cancer

Increases the CSC population and induces CSC properties such as increased proliferation and invasion of cancer cells through the WNT pathway

[44]

16 CCL-2

Breast cancer

Promotes CSC-like properties by regulating Notch-1 expression through the notch signaling pathway

[45]

9

3

LIF

Methods

3.1 Co-Culture of Cancer Cells with TAMs 3.1.1

Cell Culture

1. Oral cancer cell line, like H-357 (can be procured from Sigma Aldrich, USA, cat no. 06092004), should be cultured in DMEM-F12 (50:50, v/v) supplemented with (a) 2 mM glutamine, (b) 10% FBS, (c) 1% antibiotic (100 U/mL of penicillin, 10 mg/mL of streptomycin in 0.9% normal saline), and (d) 0.5 mg/mL of hydrocortisone (in case of oral cancer cells only) at 37 °C in an incubator with humidified atmosphere and the supply of 5% CO2 [46].

Generation of Cancer Stem Cells by Co-Culture Methods

225

2. Monocyte cell line, e.g., THP-1 cells, needs to be maintained in RPMI-1640 media supplemented with the abovementioned reagents and conditions [46] (see Note 1). 3.1.2 Differentiation of THP-1 to M1 TAMs Using External Stimuli in a CoCulture of THP-1 and Cancer Cells

1. Approximately 1 × 105 H-357 cells are seeded in a culture plate and allowed to grow for 24 h. 2. After that remove the medium, and add 1 mL RPMI media containing 1 × 103 THP-1 cells. Then add another 3–4 mL of culture media (DMEM-F12/RPMI-1640, in 50:50 ratio) to the co-cultured cells. 3. After 24 h, treat the co-cultured cells with 80 nM of PMA and incubate for another 96 h. 4. Subsequently, the cells are exposed to LPS (500 ng/mL) and incubated for an additional 48 h. 5. Collect the supernatant media. Then centrifuge at a speed of 1800 rpm in 4 °C for 5 mins. 6. Supernatant media is concentrated using a “liquid concentrator machine” (e.g., Eppendorf Concentrator Plus, Hamburg, Germany). 7. The resulting concentrated supernatant is designated as conditioned media (CM). 8. The protein content of the CM should be measured (e.g., concentration of the protein content can be measured by using UV–VIS spectrophotometer) and stored at -20 °C for further experimentation of CSC generation.

3.1.3 In Vitro Generation of Monocyte-Derived Macrophages (MDMs) Using THP-1 Cells

1. THP-1 cells (1 × 105) are grown in 60 mm tissue culture dishes. 2. Differentiation of THP-1 monocyte cells is induced by the treatment of 80 nM PMA for 96 h. 3. Next, the cells are exposed to 500 ng/mL LPS for an additional 48 h to obtain polarized (M1) macrophages (see Note 2). 4. The supernatant media are collected by the same procedure mentioned above. 5. The protein content of the CM was measured and utilized for CSC enrichment.

3.1.4 Enrichment of the CSCs by Incubating Cancer Cells with the CM

1. CSCs can be generated from cancer cells treating cancer cells with CM (which contains several stemness stimuli) derived from PMA and LPS-stimulated co-culture of H-357 + THP1 cells or PMA and LPS-stimulated M1 macrophages (see Note 3).

226

Biswajit Das and Chanakya Nath Kundu

2. The characterization of M1 macrophage generation from THP-1 monocyte cells was conducted using various bioassays (see Note 4). 3. The 500 μg/mL of CM from PMA and LPS-stimulated H-357 + THP-1 cells or PMA and LPS-stimulated M1 macrophages are added to cancer cells (e.g., H-357 cells) and incubated for additional 96 h. 4. Several secreted components from TAMs (M1 macrophages generated from THP-1 monocytes upon incubation with PMA and LPS) increased the number of CSCs in a heterogeneous cancer cell population. 5. CSC properties of the cells were verified using multiple bioassays (see Note 5). 3.2 Co-Culture of Cancer Cells with CAFs 3.2.1

Cell Culture

1. Cancer cells, e.g., H-357 cells, are cultured in DMEM F12 (50: 50, v/v) supplemented with (a) 10% FBS, (b) 2 mM glutamine, (c) 1% antibiotic (10 mg/mL of streptomycin and 100 U/mL of penicillin in 0.9% normal saline), and (d) 0.5 mg/mL of hydrocortisone at 37 °C in a mammalian cell culture incubator [12]. 2. Human dermal fibroblast (HFB) cells are cultured and maintained in DMEM supplemented with the same regents and conditions as mentioned above [12].

3.2.2 Activation of CAFs in HFB Cells Using External Stimuli

1. HFB (1 × 105) cells are seeded in a 60 mm culture plate and allowed to adhere for 24 h. 2. Subsequently, the cells are exposed to 50 ng/mL of PDGF-B followed by 10 μg/mL of TGF-β and incubated for an additional period of 96 h. 3. The PDGF-B and TGF-β stimulated the HFB cells to become activated CAFs. Generated CAFs should be characterized by using various bioassays (see Note 6).

3.2.3 Preparation of Chemokine-Enriched Conditioned Medium (CM)

1. HFB (1 × 106) cells are seeded into 60 mm cell culture dishes. 2. Treat the cells with 50 ng/mL of PDGF-B and 10 μg/mL of TGF-β and incubated for an additional 96 h. 3. The media is being collected and centrifuged at 1800 rpm at 4 ° C for 10 min. 4. The supernatant is then concentrated using the same process mentioned above. 5. The resulting concentrated supernatant is designated as conditioned media (CM). 6. The protein content of the CM is being measured and stored at -20 °C for further experimentation of CSC generation.

Generation of Cancer Stem Cells by Co-Culture Methods 3.2.4 Generation of CSCs by Co-Culturing CAFs and Cancer Cells

227

1. Approximately 1 × 103 CAF cells are introduced to a population of 1 × 105 H-357 cells and cultured for 96 h. 2. The secretory components released by activated CAFs contributed to the enhancement of CSCs within the cancer cell population. 3. Similarly, CSC-enriched oral cancer cells could be generated by subjecting H-357 cells to CM-CAF treatment (see Note 7). 4. H-357 cells were exposed to CM-CAFs (500 μg/mL) and cultured for 96 h. 5. The CSC properties were evaluated through a range of bioassays (see Note 8).

4

Notes 1. THP-1 cells, when activated by external stimuli (such as PMA and LPS), differentiate into M1 macrophages, a type of tumorassociated macrophage. 2. Stimulation with PMA alone induces basal macrophage levels (M0), while both PMA and LPS stimulate the formation of M1-type macrophages from THP-1 monocytes. 3. Co-culturing THP-1 cells (precursors of M1 macrophages) with cancer cells, followed by PMS and LPS treatment, leads to the polarization of THP-1 into M1 macrophages. Numerous secreted cytokines contribute to the gaining of CSC-like properties in cancer cells. A similar phenomenon occurs when cancer cells are treated with CM from M1 macrophages without direct co-culture. This suggests that during co-culture of cancer cells and TAMs, TAMs stimulate CSC properties in cancer cells by releasing various cytokines and inflammatory factors. 4. Enzyme-linked immunosorbent assay (ELISA) can be employed to assess representative cytokine markers (TNF-α, IL-1β, IL-6, and iNOS) specific for M1 macrophages, while immunocytochemical staining can determine specific cell surface markers of M0 (CD14-), M1 (CD68+), and M2 (CD163-) macrophages [46]. 5. To check the cancer stemness property representative CSC’s markers can be assessed by CD133, NONOG, ALDH-1, and Oct-4 using ELISA. Proliferation can be checked by cell viability assay using MTT and invasion potential cancer assessed by Matrigel invasion assay [46]. 6. Immunocytochemical staining and ELISA can be used to check specific quiescent fibroblast marker such as CD39 and CAF markers such as FAP and α-SMA [12].

228

Biswajit Das and Chanakya Nath Kundu

7. The CSC enrichment can be performed in types of cancer cells too; here the oral cancer cell line H-357 was represented as a model. 8. The stimulation of cancer stemness properties by activated CAFs in cancer cells can be confirmed by evaluating stemness markers CD133, ALDH1, and NANOG using methods like western blotting and ELISA. Proliferation assay and invasion assay also can confirm the true stemness property of the cells [12]. References 1. Ning Y, Cui Y, Li X, Cao X, Chen A, Xu C et al (2018) Co-culture of ovarian cancer stem-like cells with macrophages induced SKOV3 cells stemness via IL-8/STAT3 signaling. Biomed Pharmacother 103:262–271 2. Chan TS, Shaked Y, Tsai KK (2019) Targeting the interplay between cancer fibroblasts, mesenchymal stem cells, and cancer stem cells in desmoplastic cancers. Front Oncol 9:688 3. Singh SK, Hawkins C, Clarke ID, Squire JA, Bayani J, Hide T et al (2004) Identification of human brain tumour initiating cells. Nature 432(7015):396–401 4. Ginestier C, Hur MH, Charafe-Jauffret E, Monville F, Dutcher J, Brown M et al (2007) ALDH1 is a marker of normal and malignant human mammary stem cells and a predictor of poor clinical outcome. Cell Stem Cell 1(5): 555–567 5. Li C, Heidt DG, Dalerba P, Burant CF, Zhang L, Adsay V et al (2007) Identification of pancreatic cancer stem cells. Cancer Res 67(3):1030–1037 6. Van den Hoogen C, van der Horst G, Cheung H, Buijs JT, Lippitt JM, Guzma´nRamı´rez N et al (2010) High aldehyde dehydrogenase activity identifies tumor-initiating and metastasis-initiating cells in human prostate cancer. Cancer Res 70(12):5163–5173 7. O’Brien CA, Pollett A, Gallinger S, Dick JE (2007) A human colon cancer cell capable of initiating tumour growth in immunodeficient mice. Nature 445(7123):106–110 8. Wang L, Huang X, Zheng X, Wang X, Li S, Zhang L et al (2013) Enrichment of prostate cancer stem-like cells from human prostate cancer cell lines by culture in serum-free medium and chemoradiotherapy. Int J Biol Sci 9(5): 472–479 9. Torre-Healy LA, Berezovsky A, Lathia JD (2017) Isolation, characterization, and

expansion of cancer stem cells. Methods Mol Biol (Clifton, N.J.) 1553:133–143 10. Pasquier J, Rafii A (2013) Role of the microenvironment in ovarian cancer stem cell maintenance. Biomed Res Int 2013:630782 11. Korkaya H, Liu S, Wicha MS (2011) Breast cancer stem cells, cytokine networks, and the tumor microenvironment. J Clin Invest 121(10):3804–3809 12. Pradhan R, Paul S, Das B, Sinha S, Dash SR, Mandal M et al (2023) Resveratrol nanoparticle attenuates metastasis and angiogenesis by deregulating inflammatory cytokines through inhibition of CAFs in oral cancer by CXCL12/IL-6-dependent pathway. J Nutr Biochem 113:109257 13. Iliopoulos D, Hirsch HA, Wang G, Struhl K (2011) Inducible formation of breast cancer stem cells and their dynamic equilibrium with non-stem cancer cells via IL6 secretion. Proc Natl Acad Sci U S A 108(4):1397–1402 14. Sawa-Wejksza K, Kandefer-Szerszen´ M (2018) Tumor-associated macrophages as target for antitumor therapy. Arch Immunol Ther Exp (Warsz) 66(2):97–111 15. Storz P (2023) Roles of differently polarized macrophages in the initiation and progression of pancreatic cancer. Front Immunol 14: 1237711 16. Svensson J, Jenmalm MC, Matussek A, Geffers R, Berg G, Ernerudh J (2011) Macrophages at the fetal-maternal interface express markers of alternative activation and are induced by M-CSF and IL-10. J Immunol (Baltimore, Md.: 1950) 187(7):3671–3682 17. Wang D, Yue DL, Wang D, Chen XF, Yin XY, Wang YP et al (2018) Aspirin inhibits cell stemness of esophageal cancer by downregulation of chemokine CCL2. Chinese J Oncol 40(10): 744–749 18. Jinushi M, Chiba S, Yoshiyama H, Masutomi K, Kinoshita I, Dosaka-Akita H

Generation of Cancer Stem Cells by Co-Culture Methods et al (2011) Tumor-associated macrophages regulate tumorigenicity and anticancer drug responses of cancer stem/initiating cells. Proc Natl Acad Sci U S A 108(30):12425–12430 19. Raghavan S, Mehta P, Xie Y, Lei YL, Mehta G (2019) Ovarian cancer stem cells and macrophages reciprocally interact through the WNT pathway to promote pro-tumoral and malignant phenotypes in 3D engineered microenvironments. J Immunother Cancer 7(1):190 20. Yang L, Dong Y, Li Y, Wang D, Liu S, Wang D et al (2019) IL-10 derived from M2 macrophage promotes cancer stemness via JAK1/ STAT1/NF-κB/Notch1 pathway in non-small cell lung cancer. Int J Cancer 145(4):1099–1110 21. Luo S, Yang G, Ye P, Cao N, Chi X, Yang WH (2022) Macrophages are a double-edged sword: molecular crosstalk between tumorassociated macrophages and cancer stem cells. Biomol Ther 12(6):850 22. Zhang X, Chen L, Dang WQ, Cao MF, Xiao JF, Lv SQ et al (2020) CCL8 secreted by tumor-associated macrophages promotes invasion and stemness of glioblastoma cells via ERK1/2 signaling. Lab Investig 100(4): 619–629 23. Zhu F, Li X, Chen S, Zeng Q, Zhao Y, Luo F (2016) Tumor-associated macrophage or chemokine ligand CCL17 positively regulates the tumorigenesis of hepatocellular carcinoma. Med Oncol (Northwood, London, England) 33(2):17 24. Nie G, Cao X, Mao Y, Lv Z, Lv M, Wang Y et al (2021) Tumor-associated macrophagesmediated CXCL8 infiltration enhances breast cancer metastasis: suppression by Danirixin. Int Immunopharmacol 95:107153 25. Zhang B, Ye H, Ren X, Zheng S, Zhou Q, Chen C (2022) Macrophage-expressed CD51 promotes cancer stem cell properties via the TGF-β1/smad2/3 axis in pancreatic cancer. Cancer Lett 548:215897 26. Ma C, Komohara Y, Ohnishi K, Shimoji T, Kuwahara N, Sakumura Y (2016) Infiltration of tumor-associated macrophages is involved in CD44 expression in clear cell renal cell carcinoma. Cancer Sci 107(5):700–707 27. Yang J, Liao D, Chen C, Liu Y, Chuang TH, Xiang R (2013) Tumor-associated macrophages regulate murine breast cancer stem cells through a novel paracrine EGFR/Stat3/ Sox-2 signaling pathway. Stem Cells (Dayton, Ohio) 31(2):248–258 28. Wei X, Yang S, Pu X, He S, Yang Z, Sheng X (2019) Tumor-associated macrophages increase the proportion of cancer stem cells in

229

lymphoma by secreting pleiotrophin. Am J Transl Res 11(10):6393–6402 29. Karnoub AE, Dash AB, Vo AP, Sullivan A, Brooks MW, Bell GW et al (2007) Mesenchymal stem cells within tumour stroma promote breast cancer metastasis. Nature 449(7162): 557–563 30. Liu S, Ginestier C, Ou SJ, Clouthier SG, Patel SH, Monville F et al (2011) Breast cancer stem cells are regulated by mesenchymal stem cells through cytokine networks. Cancer Res 71(2): 614–624 31. Timaner M, Letko-Khait N, Kotsofruk R, Benguigui M, Beyar-Katz O et al (2018) Therapy-educated mesenchymal stem cells enrich for tumor-initiating cells. Cancer Res 78(5):1253–1265 32. Li HJ, Reinhardt F, Herschman HR, Weinberg RA (2012) Cancer-stimulated mesenchymal stem cells create a carcinoma stem cell niche via prostaglandin E2 signaling. Cancer Discov 2(9):840–855 33. Kabashima-Niibe A, Higuchi H, Takaishi H, Masugi Y, Matsuzaki Y, Mabuchi Y (2013) Mesenchymal stem cells regulate epithelialmesenchymal transition and tumor progression of pancreatic cancer cells. Cancer Sci 104(2): 157–164 34. Biffi G, Oni TE, Spielman B, Hao Y, Elyada E, Park Y (2019) IL1-induced JAK/STAT signaling is antagonized by TGFβ to shape CAF heterogeneity in pancreatic ductal adenocarcinoma. Cancer Discov 9(2):282–301 35. Todaro M, Gaggianesi M, Catalano V, Benfante A, Iovino F, Biffoni M (2014) CD44v6 is a marker of constitutive and reprogrammed cancer stem cells driving colon cancer metastasis. Cell Stem Cell 14(3):342–356 36. Chen WJ, Ho CC, Chang YL, Chen HY, Lin CA, Ling TY (2014) Cancer-associated fibroblasts regulate the plasticity of lung cancer stemness via paracrine signalling. Nat Commun 5:3472 37. Lotti F, Jarrar AM, Pai RK, Hitomi M, Lathia J, Mace A (2013) Chemotherapy activates cancer-associated fibroblasts to maintain colorectal cancer-initiating cells by IL-17A. J Exp Med 210(13):2851–2872 38. Shi Y, Gao W, Lytle NK, Huang P, Yuan X, Dann AM (2021) Targeting LIF-mediated paracrine interaction for pancreatic cancer therapy and monitoring. Nature 600(7889):E18 39. Lonardo E, Hermann PC, Mueller MT, Huber S, Balic A, Miranda-Lorenzo I (2011) Nodal/Activin signaling drives self-renewal and tumorigenicity of pancreatic cancer stem

230

Biswajit Das and Chanakya Nath Kundu

cells and provides a target for combined drug therapy. Cell Stem Cell 9(5):433–446 40. Lonardo E, Frias-Aldeguer J, Hermann PC, Heeschen C (2012) Pancreatic stellate cells form a niche for cancer stem cells and promote their self-renewal and invasiveness. Cell Cycle (Georgetown, Tex) 11(7):1282–1290 41. Lenos KJ, Miedema DM, Lodestijn SC, Nijman LE, van den Bosch T, Romero Ros X (2018) Stem cell functionality is microenvironmentally defined during tumour expansion and therapy response in colon cancer. Nat Cell Biol 20(10):1193–1202 42. Rudnick JA, Arendt LM, Klebba I, Hinds JW, Iyer V, Gupta PB (2011) Functional heterogeneity of breast fibroblasts is defined by a prostaglandin secretory phenotype that promotes expansion of cancer-stem like cells. PLoS One 6(9):e24605 43. Tang YA, Chen YF, Bao Y, Mahara S, Yatim SMJM, Oguz G (2018) Hypoxic tumor microenvironment activates GLI2 via HIF-1α and

TGF-β2 to promote chemoresistance in colorectal cancer. Proc Natl Acad Sci U S A 115(26):E5990–E5999 44. Sun Y, Campisi J, Higano C, Beer TM, Porter P, Coleman I et al (2012) Treatmentinduced damage to the tumor microenvironment promotes prostate cancer therapy resistance through WNT16B. Nat Med 18(9): 1359–1368 45. Tsuyada A, Chow A, Wu J, Somlo G, Chu P, Loera S (2012) CCL2 mediates cross-talk between cancer cells and stromal fibroblasts that regulates breast cancer stem cells. Cancer Res 72(11):2768–2779 46. Pradhan R, Chatterjee S, Hembram KC, Sethy C, Mandal M, Kundu CN (2021) Nano formulated resveratrol inhibits metastasis and angiogenesis by reducing inflammatory cytokines in oral cancer cells by targeting tumor associated macrophages. J Nutr Biochem 92: 108624

Chapter 17 Deep Learning of Cancer Stem Cell Morphology Hiroyuki Kameda, Hiroaki Ishihata, and Tomoyasu Sugiyama Abstract Knowledge regarding cancer stem cell (CSC) morphology is limited, and more extensive studies are therefore required. Image recognition technologies using artificial intelligence (AI) require no previous expertise in image annotation. Herein, we describe the construction of AI models that recognize the CSC morphology in cultures and tumor tissues. The visualization of the AI deep learning process enables insight to be obtained regarding unrecognized structures in an image. Key words Cancer stem cell, Morphology, Segmentation, Classification, CGAN, Google colaboratory, Jupyter Notebook, Pix2pix, Convolutional neural network, Grad-CAM

1

Introduction The concept of cancer stem cell (CSC) producing progenitor cells for differentiated cancer cells makes the morphology of CSC different from that of other cells. Indeed, CSC biologists are aware that some CSCs lose their pluripotency in a cell culture when cell is regularly checked under a microscope. Although microscopic image processing has been an important task allowing cell biologists to analyze objects in an image, there have been few reports on the morphology of CSCs [1]. However, image recognition technology has greatly advanced in recent years. The use of artificial intelligence (AI) regarding cell biology has been increasing [2]. The segmentation of intended objects and image classification are fields in which AI has exhibited the same expertise as trainees when enough datasets are available. We incorporated new technologies into a CSC morphological study. Pix2pix is an image generative AI using a conditional generative adversarial network (CGAN) [3] and has the potential to segment objects in an image that are difficult even for experts to find. Grad-CAM is a tool for analyzing the AI classification process

Federica Papaccio and Gianpaolo Papaccio (eds.), Cancer Stem Cells: Methods and Protocols, Methods in Molecular Biology, vol. 2777, https://doi.org/10.1007/978-1-0716-3730-2_17, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

231

232

Hiroyuki Kameda et al.

Fig. 1 Training of CGAN to map the CSC morphology ! stemness image. (a) Overview of workflow (reproduced from [16]/CC BY ND 4.0). (b) CGAN was trained using pairs of bright-field images for the morphology and dark-field images for labeling the cells with the stemness (reproduced from [18]/CC BY 4.0)

when using a convolutional neural network (CNN) [4]. It has potential to indicate objects recognized by AI, suggesting a structural view that differentiates the objects from other parts. Herein, we demonstrate that CSCs can be segmented in a CSC morphology without expertise by using a cell image and deep learning (DL). Although the methodology still requires improvement, it indicates the capability of constructing an AI model without expertise. We were able to build AI models using pix2pix (Fig. 1). After applying DL to images numbering from several hundred to thousands, an AI model was able to depict CSCs from each corresponding phase contrast image (Fig. 2). The image output by AI closely resembled the original stemness image obtained using fluorescent microscopy. An evaluation of the machine learning was conducted based on the recall, precision, and F-measure between the AI-output images and the corresponding stemness images (Fig. 3). CSC objects recognized through DL were visualized based on a Gad-CAM analysis during CNN classification (Fig. 4).

Deep Learning of Cancer Stem Cell Morphology

output

b

target

input

output

target

1471

input

No.

No.

2627

2798

2545

2782

2505

a

233

Fig. 2 CSC segmentation in bright-field image using AI models. (a) CSC in culture (reproduced from [17]/CC BY 4.0). The “Input” image was subjected to an AI “output” production for indicating the CSC. “Target” referred to stemness image based on microscopy. Bars ¼ 100 μm. (b) CSC in tumor tissue (reproduced from [17]/CC BY 4.0). Bars ¼ 100 μm stemness image by microscopy b1

d

a1

c1

a = 6an b = 6bn c = 6cn

Output image by AI b2

AI

a2 c2

a a+c a Recall = a+b d Spesificity = c+d

Microscopy

Pos. Pos. a Neg. c

Neg. b d

Precision = Evaluation functions

F-measure = 2 ∙

Precision ∙ Recall Precision + Recall

Fig. 3 Evaluation functions for CSC segmentation. Binarized images of stemness and the corresponding AI output were overlayed to calculate each area

Thus, we have made several AI models for recognizing a morphology characteristic of CSCs. The AI construction process does not require expertise in CSC morphology and is appropriate for evaluating the quality of AI using several values. A visualization technology was applied to the AI process to find structures in CSC images.

234

Hiroyuki Kameda et al.

b

a

1-day

1-day-center

Classification by CNN Grad-CAM Guided Grad-CAM 0

0

1

Grad-CAM

cell image sets

highlighting regions in image for predicting the concept

Guided Grad-CAM

classified 1

Fig. 4 Visualization of CSC classification using a CNN. (a) Overview of process (reproduced from [17] / CC BY 4.0). (b) Visualization of CNN image classification into 1-day image and 1-day center image (reproduced from [17] / CC BY 4.0). Regions where AI was referenced to classify the images are shown using the heat map generated by Grad-CAM and the highlighted region, arrow, and arrowhead generated by guided Grad-CAM. Region indicated by an arrowhead is a zoomed-in inset. Bars ¼ 100 μm

2

Materials

2.1 Typical PC Hardware, OS, and Software Framework

1. Hardware systems (prerequisite). As is well known, because the computational complexity of DL is quite high and the data size is usually extremely large, DL software generally requires numerous resources. For these reasons, hardware should be cautiously prepared. One option is to prepare a computer with a high-spec GPU. Another option is to use Google Colaboratory, which is a cloud-based computing environment that can run programs on high-spec machines without charge. We recommend the latter to AI beginners (see Note 1). 2. Software systems (prerequisite). Theoretically, it is feasible to write and run AI programs in any programming language. However, deep learning software programs are usually written in Python or C/C++. Among them, Python is an extremely easy to learn but powerful programming language. Although there are many other options, we therefore recommend Python, allowing AI programs to easily operate. The following are needed to run AI programs: (a) Python execution environment, for example, Google Colaboratory. Using such an environment, Python programs can be written and run with a software development tool, such as Jupyter Notebook, on Google Colaboratory.

Deep Learning of Cancer Stem Cell Morphology

235

(b) Web browser execution environment, for example, Microsoft Edge or Google Chrome. Using them, Google Colaboratory can be accessed and used free of charge. There are also other options, including Anaconda for running Python programs and text editors for writing programs. Although any option can be chosen, in this section, we describe only the use of Jupyter Notebook. 3. Skills and knowledge required. Experience using a PC is sufficient. 4. Practical solution (Google Colaboratory). As we previously stated, we recommend Google Jupyter to beginners of AI programming. 2.2 Microscopy and CSC Culture

1. Microscope: fluorescence microscope BZ-X800 equipped with a CFI Plan Fluor DL 10 lens, a 20 lens, and a digital camera with a controller PC. 2. CSC: miPS-LLCcm derived from iPS-MEF-Ng-20D-17 [5]. 3. Lewis lung carcinoma cell: LLC. 4. miPS-LLCcm culture medium: This is a mixed medium with M1 and M2 at a ratio of 1:1. M1 refers to Dulbecco’s Modified Eagle Medium high glucose, 15% fetal bovine serum (FBS), 1 nonessential amino acids, 1% penicillin/streptomycin (P/S), 1 L-glutamine, and 100 μM of 2-mercaptoethanol. M2 refers to a conditioned medium of LLC prepared through the cultivation of confluent LLC for a 1-day period. 5. LLC culture medium: DMEM high glucose, 10% FBS, 1  NEA, and 1% P/S. 6. Cell culture dish/multiwell plate: 6-cm dish, 6-well plate, and 12-well plate for cell culture grade. 7. Cell culture dissociation reagent: trypsin containing 0.25% ethylenediaminetetraacetic acid (trypsin–EDTA). 8. Cell wash solution: phosphate-buffered saline (PBS). 9. CO2 incubator.

3

Methods

3.1 Cell Image Acquisition from Cell Culture

1. Semi-confluently grown cells in culture dish are washed twice with PBS. 2. Cells are incubated with trypsin–EDTA for 5 min at 37  C. 3. Cells are collected from the dish and diluted with culture medium. 4. One-twentieth of cell solution is added to a new culture dish containing culture medium.

236

Hiroyuki Kameda et al.

5. Cells grown in culture dish for 1–3 days were ready for microscope image analysis. 6. Cells cultured on dish is set to the fluorescence microscope. 7. Image acquisition: Using the function of the microscope, obtain pairs of phase contrast and GFP fluorescence images in tiff format with a pixel resolution of 1920  1440 (see Note 2). 8. Image folder: Using the function of the microscope, indicate the place to save image file. Folder naming is typically in a ddmm-yy format, but this is not restricted. Each pair of images are held in a directory named with the well or dish number placed in a dd-mm-yy holder. Fluorescent images should be saved as Image_CH1.tif. Phase contrast images should be saved as Image_CH2.tif. 3.2 Image Processing for pix2pix

1. Number of images: Tens of thousands of image pairs are expected but not required. 2. PC for deep learning: The image folder in dd-mm-yy format is copied to the computer user directory. The directory should contain a tool’s holder where Python scripts are held. 3. Processing of images: Original images will be split to create images of 256 pixels  256 pixels in size using the Python script cropmix.py placed in the tool’s holder. The script can be obtained from GitHub. To do so, one must open the website https://github.com/KatakenW505/microscopy_image_tool, click the “code” button, and select “Download ZIP.” Newly made small image pairs are combined into an image in which the fluorescence image is placed to the left. The combined images are saved in image folder images_out, which will be made in the same user directory. For execution of the script, the following is typed in the same directory of the tool’s holder (see Note 3). $ python tools=cropmix:py dd  mm  yy

3.3 Construction of an AI Model Predicting CSC in Phase Contrast Cell Image

In this section, we start with an introduction. If attempting to run a pix2pix program, jump to Subheading 3.3.2. However, to become familiar with the basic ideas of AI, read the following Subheading 3.3.1.

3.3.1 Basic Concepts of AI Models

1. Basic neural network structure for understanding of deep neural network (DNN) We start from the basics of neural network technology, which is a base of modern AI. The origin of research on neural networks is the study by McCulloch and Pitts [6], after which perceptron Rosenblatt [7] and neocognitron Kunihiko Fukushima [8, 9] were proposed. In addition, in 2012, a

Deep Learning of Cancer Stem Cell Morphology

237

modern AI technology, that is, the DL-based AlexNet, was proposed [10]. Various types of ideas have since been proposed. A generative adversarial network (GAN) is an authentic representative technology of modern AI [11]. A GAN achieves a higher performance than humans, particularly in the field of image recognition/understanding/creation. Later, we describe an upgraded version of the original GAN, pix2pix (a type of conditional GAN) [12]. In this subsection, we first take a brief and compact tour of neural network-based AI software (see Note 4). The program codes in Fig. 5 are given below. This program learns the relation between the input and output and creates a simple neural-network-type prediction model. This program consists of five parts: – Part 1: framework settings. – Part 2: settings of the learning model type and learning data. – Part 3: network construction settings. – Part 4: network training, i.e., learning to build a model. – Part 5: prediction based on the trained network. This AI program learns and creates a neural network model for predicting a value (ideally, 2n + 3) from a given value n. Although we describe neither the details in the writing of programs using Python nor the algorithm applied by this program, we do show how to execute the program, allowing readers to learn how to run programs developed by Jupyter Notebook on Google Colaboratory, while also checking the computation environment at the same time. The following are the steps used to run the code: Step 1: Go to the website https://colab.research.google.com/ and open a new file. When you arrive at the website, a window pops up allowing you to select a file. Click the [Cancel] button in the lower right corner (Fig. 6) and then select the [File] button at the top left to create a new file for your programming (Fig. 7). Step 2: On a new notebook, add a new code area by clicking the [+ Code] button (Fig. 8). You will then receive a code area on the notebook (Fig. 9). If you are unaccustomed to using Jupyter Notebook as proposed by Project Jupyter, ask your colleagues to help you, or check the websites, for example, Project Jupyter (https://jupyter.org/). The program is slightly complicated but not too difficult. Step 3: Copy and paste the code in Fig. 5 to the Jupyter code area.

238

Hiroyuki Kameda et al.

#Part 1: Framework settings # library setting from tensorflow import * from keras import * #Part 2: Settings of learning model type and learning data # Setting to the type of learning network model model = Sequential() #Setting to learning data; # This program learns to output 5 when 1 is inputted, # 7 when 2, 9 when 3, and so on. input = [1, 2, 3, 4, 5] output = [5, 7, 9, 11, 13] #Part 3: Network construction settings # input layer model.add(layers.Dense(units=3,input_shape=[1])) #中間 layer model.add(layers.Dense(units=2)) # output layer model.add(layers.Dense(units=1)) #Part 3: Building an initial AI network model.compile(loss='mean_squared_error',optimizer='sgd') #Part 4: Network training model.fit(x=input,y=output,epochs=1000) #Part 5: Prediction based on the trained network # output is 5 print(round(model.predict([1]))) # output is 15 print(round(model.predict([6]))) print(round(model.predict([7]))) # output is 17 # output is 19 print(round(model.predict([8]))) # output is 23 print(round(model.predict([10]))) Fig. 5 A sample code of a neural network to test the computing environment

Step 4: Click the run button (the triangle button shown on the left in Fig. 9) to run the code. Step 5: Wait and check the outputs from the AI. In this case, the AI program outputs 5 to 1, 7 to 2, 9 to 3, 11 to 4, and 13 to 5. These outputs suggest that this AI program learned the function F : n ! 2n + 3 from only a pair of input and output lists.

Deep Learning of Cancer Stem Cell Morphology

239

Fig. 6 A pop-up window just after the Google Colaboratory starts

Fig. 7 Snapshot of the top left corner of the Colaboratory web page, where you click the [File] button and select [New notebook] to create your workspace, i.e., Jupyter notebook

240

Hiroyuki Kameda et al.

Fig. 8 A pop-up menu to create a new notebook by clicking the [New notebook] button

Fig. 9 Code area to write code to run, where you describe your code in the text box, and click the triangle button to run the programs

Random Noise

Dataset Real images

Generator

Real images

Fake images

Discriminator

Judge

Fig. 10 Structure of a GAN, which learns to discriminate what is real or fake by the generator and the discriminator fighting against each other, where the generator generates fake images from noise, and the discriminator judges if an input is a real or fake image

Now that you obtained the expected outputs from the AI system, you are ready to move on to the next phase, i.e., creating an AI program for predicting CSC. Congratulations! 3.3.2 GAN, Conditional GAN, and pix2pix

1. What is a GAN? A GAN is one of the epoch-making AI algorithms in the twenty-first century that has shown a surprisingly high performance in image processing. A GAN consists mainly of two modules (Fig. 10):

Deep Learning of Cancer Stem Cell Morphology

241

I. Generator: This module generates images like those in the training dataset by taking noise as an input. II. Discriminator: This module accepts the output image of the generator and an image from the training dataset to judge whether the image is real or fake. As the point of a GAN, the generator and discriminator effort to achieve their own best results, that is, the generator creates fake images to deceive the discriminator, and the discriminator sensitively detects fake images. As a result, the GAN reaches a learning stage to gain the ability to detect fake images more precisely than a human. To learn more, go to a tutorial video given by Ian Goodfellow [13]. 2. Conditional GAN and pix2pix As described before, a GAN is extremely powerful; however, we cannot control a GAN to detect images under certain conditions. For example, although a GAN can create highquality fake images of human faces [14], we cannot ask the original GAN to create a human face, for example, that of a young 27-year-old smiling Asian female. A conditional GAN can be used to achieve this [15]. Although this type of GAN is extremely interesting, we now delve further into a more revised version of a conditional GAN, that is, pix2pix, that we applied to detect iPS cancer cells in our research [16]. The pix2pix approach is an upgraded version of a conditional GAN, and its main application area is image-to-image translations. That is, pix2pix can be applicable to building a new AI system. It accepts both original cell images and processed images with markers and then creates a prediction model that translates images with no marks into those with marks. The pix2pix AI system outputs fake images with markers as if the cells were processed by inspection reagents. In the next section, we describe how to run the pix2pix AI program, allowing you to make a model for detecting iPS cancer cells. 3. How to run pix2pix There are currently many refined versions of pix2pix programs on the web. Although we used programs [17] in our research [16], it is somewhat obsolete and now we can choose many better versions of pix2pix. In this section, we present one of them to the readers. Now, we will start with obtaining resources. I. Getting resources from a website, e.g., GitHub Step 1: Go to the web page https://github.com/junyanz/ pytorch-CycleGAN-and-pix2pix.git, which looks like the image in Fig. 11. Step 2: Click the green [Code] button to show a pulldown menu (see Fig. 12).

242

Hiroyuki Kameda et al.

Fig. 11 A snapshot of the GitHub page (https://github.com/junyanz/pytorch-CycleGAN-and-pix2pix)

Fig. 12 A pulldown menu for you to select [Download Zip] and download a zipped file onto your desktop

Step 3: Select [Download zip] to download all resources from this web page onto your PC, for example, on your desktop. Step 4: Unzip the file.

Deep Learning of Cancer Stem Cell Morphology

243

II. Software deployment settings Step 5: Evoke a web browser and log into Google Drive. If you do not yet have a Google ID, obtain one in advance. Step 6: Upload the folder with all files onto Google Drive through a drag-and-drop approach. Step 7: Open the folder on your Google Drive. You may change the folder name if necessary. You are now ready to run the pix2pix program. III. Run the pix2pix program Step 8: Double-click the file titled “pix2pix.ipynb” in the uploaded folder. A Jupyter Notebook page will then open on Google Colaboratory. You are now in Google Colaboratory, which is an extremely useful and powerful application development software. The file with the suffix “.ipynb” is Jupyter Notebook. To learn more about Jupyter Notebook, search and check the web pages. Before running pix2pix, you must do one extremely important task, i.e., the GPU mode is set to “On” in advance to run programs faster than using only a CPU. Step 9: Set Google Colaboratory to run with a GPU by clicking [Runtime]-[Change runtime type] and select GPU (Figs. 13 and 14). Now, we must execute the following commands in order, i.e.: 1. Install the computing Colaboratory.

environment

on

Google

2. Apply the dataset settings: Download and deploy them on Google Colaboratory. 3. Run pix2pix to learn a model from the datasets. 4. Test the learned model. 5. Visualize some outcomes on-screen. Let us execute them individually. Step 10: Installing the computing environment. Click the triangle button to issue a command to install and deploy resources on the Google Colaboratory. Substep 1:!git clone https://github.com/junyanz/pytorchCycleGAN-and-pix2pix

244

Hiroyuki Kameda et al.

Fig. 13 The pulldown menu to select [Change runtime type] to set the GPU to on

Fig. 14 Menu to select GPU as a hardware accelerator

Deep Learning of Cancer Stem Cell Morphology

245

All resources you have to prepare are downloaded and deployed on the Google Colaboratory. Subsetp 2: import os os.chdir(‘pytorch-CycleGAN-and-pix2pix/’) The first line imports a library called “os” to use a Linux-like command environment. Here, change the working directory to ‘pytorch-CycleGAN-and-pix2pix/’. Substep 3:!pip install -r requirements.txt This line issues a pip command to set the requirements. Step 11: Dataset settings. Substep 4:!bash facades

./datasets/download_pix2pix_dataset.sh

This command organizes the datasets. In this case, a “facades” parameter is given to the shellScript command, download_pix2pix_dataset.sh. When you try other datasets, use the cityscapes, night2day, edges2handbags, edges2shoes, facades, or maps parameters in place of “facades,” which are prepared by the GitHub page author. Step 12: Official pretrained modes (this is an option, but is recommended to save time). Substep 5:!bash ./scripts/download_pix2pix_model.sh facades_label2photo. The line ShellScript command obtains and deploys the official pretrained model. Step 13: Run pix2pix to learn a model from the datasets. Substep 6:!python train.py --dataroot ./datasets/facades \. --name facades_pix2pix \ --model pix2pix \ --use_wandb This command runs a Python program to train a model, where the parameters --dataset, -name, and -model assign the dataset, experiment, and model names, respectively. Details are written in the program train.py. It takes 1 to 2 hours to finish this command. After that, go to step 14. Note that an input prompt appears while executing the program once (Fig. 15). Input “3” is as shown in Fig. 15. Step 14: Test the learned model. Substep 7: !python test.py --dataroot ./datasets/facades \. --model pix2pix \ --name facades_label2photo_pretrained\

246

Hiroyuki Kameda et al.

Fig. 15 An input prompt that appears while executing. Input “3”, for example

Fig. 16 Codes to show images on the screen

--use_wandb A Python program is used to test the learned model by applying images in the test folder under the name. / datasets/facades. This time you are also asked to select (1)–(3), like Fig. 15, while executing this program. Again, input “3”. It is now time to check the outcomes. Step 15: Visualize some outcomes on the screen.

Substep 8: Execute the following. Three images show up on the screen: a fake image and two real images (Fig. 16). These inputs and outputs show that a leaned AI model can produce a marked CSC image when an original non-marked CSC image is given as input to pix2pix (Fig. 17). As you have seen thus far, some parameter settings or dataset preparations might be necessary, which are difficult or tedious to handle. Except for such inevitable but tedious work, you only have to collect and deploy the datasets on an appropriate folder and run pix2pix. That is, if you have studied using AI pix2pix, you only need to prepare your own datasets (a collection of paired images, i.e., the original CSC

Deep Learning of Cancer Stem Cell Morphology

247

Fig. 17 Examples of inputs and outputs. Images (a) and (b) are paired input, and image (c) is one of the outputs based on the learned AI model using the pix2pix program

image and processed CSC images) and then deploy them in the appropriate directory and run the pix2pix program on the Google Colaboratory free of charge. Go to the original paper [12] if you want to know more about pix2pix, or read our original study [16, 17]. In short, pix2pix programs learn a model to translate images into other types of images. For example, black and white images into color images, or sketch drawings into natural images. More details are written in the original paper [12] and can be found at GitHub (https://github.com/junyanz/pytorchCycleGAN-and-pix2pix.git). Of course, again, some parameter settings or dataset preparations are necessary, which can be difficult to handle. Except for such inevitable but tedious work, you must only collect and deploy the datasets in an appropriate folder and run pix2pix. In our previous studies [16, 17], we prepared a number of paired images, that is, phase contrast and fluorescent images, and applied a modified version of pix2pix, the original Japanese version of which is open at GitHub (https://github.com/GINK03/pytorch-pix2pix) (see Note 5). In the example code described in this section, the structure of the dataset file can be seen in Fig. 18. 3.4 Evaluation of CSC Image Created Using an AI Model

The AI system described above is for prediction. The AI using a trained model reads in an image (a picture of iPS cancer cell) as an input and creates as an output an image with marks on it suggesting where cancer cells exist on the image. In this type of case, the prediction outputs are grouped into four categories: (a) true positive, (b) true negative, (c) false positive, and (d) false negative.

248

Hiroyuki Kameda et al.

Fig. 18 An example of a pix2pix application that accepts image (a) as an input to the AI and then produces image (b) as an output by using the learned AI model, whereas image (c) is a correct one corresponding to (a) (reproduced from Fig. 4 in [16] / CC BY ND 4.0) Table 1 Confusion matrix based upon which a prediction evaluation is conducted Prediction Input

Positive

Negative

Positive examples

A

C

Negative examples

D

B

Total number N ¼ A + B + C + D

Categories (a) and (b) are correct prediction results, and categories (c) and (d) are incorrect prediction results. They are usually described as a table or confusion matrix (Table 1). Based on a confusion matrix, the following three indices are defined: 1. Error rate ¼ (C + D) / N 2. True positive rate ¼ A / (A + C) 3. False-positive rate ¼ D / (B + D) By contrast, we have three types of data: input image A, input image B that corresponds to an experimentally human processed image, and a fake image created by the AI pix2pix model. In addition, we want to know how real the fake one appears to be. One method to check this is to calculate the correlation between a real input image and a fake one created by the AI model using the real image. This calculation is easily achieved using Python or R. Check the web pages if you want to know the definitions of the expressions used or how to calculate the program by machine. You can also refer to our studies as well [16, 17].

Deep Learning of Cancer Stem Cell Morphology

3.5 Cancer Stem Cell Image Classification and Visualization 3.5.1

Overview

249

In this section, we describe a method of image classification using machine learning. We will apply the classification of 1- and 2-day CSC cultures, as described in previous studies [3, 4, 17, 18]. The procedure consists of the following four steps: 1. Preparation of the software environment The authors used a PC (Ubuntu OS) with a GPU. Although it should be possible to run on Windows or other operating systems with different installation procedures, we do not discuss such procedures here. To use Grad-CAM, install Python3; the necessary libraries (i.e., NumPy, Matplotlib, and PIL); PyTorch and torchvision, which are deep learning frameworks; and PyTorch-Grad-CAM. 2. Construction of the dataset We use images from micrographs taken on 1- and 2-day cultures of CSCs. Place the image files cropped to an appropriate size (256  256) in a folder for each class (day1, day2). From these images, create a torchvision dataset called CSCDataset that is inherited from torch.utils.data.Dataset. To mitigate the small number of data, we define an image transformation pipeline (e.g., rotation, cropping, and flipping) for data augmentation. 3. Training a CNN We train CNN models for binary classification using either a pre-trained model or a model developed from scratch. Among the various CNN models, VGG16 [19] and ResNet50 [20] are used. For both, PyTorch has made not only the model available but also the pre-trained model parameters. In many cases, the training time will be reduced when using the pre-trained model. 4. Visualize where the CNN is focused. Grad-CAM is a method used to visualize which part of an image is focused on by the CNN. From the last convolution layer of the CNN, the focused areas of the CNN are extracted and displayed in the form of a heat map. The areas that contribute the most to the classification results are displayed in red, and the areas that contribute the least are displayed in blue.

3.5.2

Installation

1. Python3 Install Python3 and the necessary libraries (NumPy, Matplotlib, PIL). Because there are various ways to install Python3, the libraries must be installed in an appropriate manner. 2. PyTorch Install PyTorch and torchvision by referring to the official PyTorch page [21]. There are various combinations depending on the OS, Python version, and CUDA (GPU driver) version, and therefore, select the appropriate one.

250

Hiroyuki Kameda et al.

3. PyTorch-Grad-CAM Install PyTorch-Grad-CAM using pip as follows. Note that it does not exist in the Anaconda repository, and thus, if you are using anaconda, install it carefully. pip install PyTorch  Grad  CAM 4. Download the csc classification program Go to the GitHub site at https://github.com/hiroishihata/cancer-stem-cell-classification. You can see the “code” button at the upper right area of the page. Click it and select “Download ZIP.” The downloading will start. The source files of dataset.py, train.py, and gcam.py will be included. 3.5.3

Sample Data

Download (413.5 MB) the zip file from https://drive.google. com/file/d/1EXsWd8inxnQWs-6wxmOIVfrWRE75qVGS/ view?usp¼sharing. Extract and place it directly under the source code directory. This file contains the sample CSC data (images of 1-day and 2-day cultured). The image files of a 1-day culture (day1) are numbered from 1 to 3357, and the images of a 2-day culture (day2) are numbered from 3358 to 6620. All images are monochrome with a pixel resolution of 256  256 and are placed in separate directories for each class. Figures 19 and 20 show a single image from day1 and day2, respectively. In dataset.py, a dataset that can be used by PyTorch is constructed. All images are read out and stored in memory with labels. The labels are 0 for 1-day incubation and 1 for 2-day incubation.

Fig. 19 1-input.png

Deep Learning of Cancer Stem Cell Morphology

251

Fig. 20 6620-input.png

The data are randomly divided into five groups, four groups are used for training, and one is used for evaluation. When initializing the CSC dataset, the fold argument specifies the group numbers (0–4) to be used in a list format. The same value should be used for the random seed number when you make the training dataset and test dataset (see Note 5). The function get_transform () in dataset.py is used to obtain the transform function of the data augmentation. The function returns different transformations depending on the context in which it is used, as described in the following. • During training: 1. Random rotation within a range of 30 degrees. 2. From a random location, crop a square region with a size of 224  224. 3. Randomly flip horizontally. 4. Randomly flip vertically. 5. Convert into PyTorch tensor (pixel values are converted such that they are between 0 and 1.0). 6. Normalize using the mean ([0.485, 0.456, 0.406]) and std. ([0.229, 0.224, 0.225]). • During the evaluation: 1. Crop out the 224  224 center portion of the image. 2. Convert into PyTorch Tensor. 3. Normalize using the mean ([0.485, 0.456, 0.406]) and std. ([0.229, 0.224, 0.225]).

252

Hiroyuki Kameda et al.

• Image display: 1. Crop out the 224  224 center portion of the image. 2. Convert into PyTorch Tensor. The method getitem () in CSCDataset returns data-augmented image data and their labels upon request from the training program. The showim () function in dataset.py is used to display images from a list of PyTorch tensor images. 3.5.4

Usage

Execute commands to train the CNN and display Grad-CAM. 1. Training CNN. The following command input will train the selected CNN model. Usage: python train.py --model MODEL [--pretrained] [--outd DIR]

Example: python train.py --model VGG16 --outd result # VGG16 training for scratch python train.py --model ResNet50 --outd result # ResNet50 training for scratch python train.py --model VGG16 --pretrained --outd result # VGG16 transfer learning python train.py --model ResNet50 --pretrained --outd result # Resnet50 transfer learning

MODEL is either “VGG16” or “ResNet50”. By default, training is applied from scratch. If the --pretrained argument exists, transfer learning is conducted (see Note 6). A stochastic gradient decent (SGD) is used as the optimizer with a momentum of 0.5. The hyperparameters of the learning, learning rate, and number of epochs are set to 0.01 and 30, respectively. The data in the dataset are trained through a mini-batch process, i.e., the batch size of the data is also trained. A single epoch of training using all data in the dataset is called 1 epoch, and in this case, 30 epochs are trained. The results are stored under DIR (or “result” if omitted) in the execution directory. A file of the learned parameters and a graph of the training results are saved. The file names are MODEL[_pre]_model.pth and MODEL[_pre] _train.png. Table 2 shows an example of the training results. The graph in Fig. 21 shows the progress of the loss and accuracy per epoch for both training and evaluation.

Deep Learning of Cancer Stem Cell Morphology

253

Table 2 Example training results Model

Train accuracy

Test accuracy

VGG16_model.pth

0.973

0.936

ResNet50_model.pth

0.949

0.924

VGG16_pre_model.pth

0.884

0.906

ResNet50_pre_model.pth

0.920

0.926

Fig. 21 Example of ResNet50 training progress

2. Grad-CAM. The following command will display the Grad-CAM and Grad-CAM++ results using the trained model. Usage: python gcam.py --imgs n1[,n2,....] --model [--pretrained]

Example: show gradcam for data 1 2 3 4 6617 6618 6619 6620. python gcam.py --imgs 1 2 3 4 6617 6618 6619 6620 --model VGG16# use VGG16 model python gcam.py --imgs 1 2 3 4 6617 6618 6619 6620 --model ResNet50 python gcam.py --imgs 1 2 3 4 6617 6618 6619 6620 --model VGG16 --pretrained python gcam.py --imgs 1 2 3 4 6617 6618 6619 6620 --model ResNet50 --pretrained

254

Hiroyuki Kameda et al.

Fig. 22 Example of gcam.pyresult by running ‘python gcam.py --model ResNet50 --imgs 1 2 3 46,617 6618 6619 6620 --pretrained’

Here, --imgs specifies the image number in the dataset. MODEL and --pretrained specify the trained model used for visualization. The ground truth and predicted value for the specified image are printed to a standard output. At the same time, the input image, the Grad-CAMoutput image, and the Grad-CAM++ output image are displayed. The ground truth and predicted values are displayed in the upper left corner of each image. Figure 22 shows an example of a display using the ResNet50_pre_model.pth model. Four input images (1, 2, 3, and 4) from day1 and four input images (6617, 6618, 6619, and 6620) from day2 were used. The top row shows the input image, the middle row shows the Grad-CAM output, and the bottom row shows the Grad-CAM++ output. For other methods, see Note 7.

4 Notes 1. Future systems: Computer technology has been developing extremely quickly. Both hardware and software, their use, and their environment are all changing. Readers should catch up with these changes in technology. In computer science, new technologies, e.g., microservices and Docker containers (a kind of virtualization technology) are becoming more practical. We recommend you learn such technologies, as well. 2. The quality of the images largely affects the deep learning. The light exposure time should be constant among images of bright and dark fields. When an object is scanned automatically to obtain multiple pairs of images in a culture dish, the images

Deep Learning of Cancer Stem Cell Morphology

255

may be affected by the surface flatness of the dish. It might be necessary to consider using autofocusing for each scan and high-quality dishes. 3. Script cropmix.py requires some modules of Python to be installed. Please see the content to find the name of the module, which can be installed according to Python. 4. The steps required to run the code depend on the software environment you use. In this section, we again describe a commonly used software environment to run the code, i.e., Google Colaboratory. We provide a minimal and compact explanation. Visit the Colaboratory website (Project Jupyter; https://jupyter.org/) for more detailed information. 5. PyTorch provides several pre-trained models for images, including the following: • Image size of 224  224. • Pixel values between 0.0 and 1.0. • Normalize to the mean [0.485, 0.456, 0.406], std. [0.229, 0.224, 0.225]. The dataset module converts monochrome image data into color and normalizes the values accordingly. Normalization may not be necessary if a pre-trained model is not used. It is also possible to randomly change the brightness and color as a data augmentation. A better method called random erasing has recently been used. 6. Training a CNN from scratch requires a large number of data and takes time. For example, a PyTorch pre-trained model is trained using more than 1,000,000 images. When using a pre-trained model with the --pretrain argument, only the parameters of the last stage of the pre-trained model are trained. It therefore takes less time than training from scratch. The default batch size for training is 64. The learning rate of 0.01 is the same as the default for PyTorch. Depending on the GPU used, there may be a shortage of memory. In this case, reduce the batch size and learning rate (e.g., batchsize ¼ 32, lr ¼ 0.005). 7. Grad-CAM is computed using the output of the final convolution layer of the CNN. The final convolution layer is specified for VGG16 and ResNet50 individual (the target layer variable in the source code). Although another Grad-GAMmodule is available [1, 22], which implements more algorithms, the API is different, and thus, this program does not work as is.

256

Hiroyuki Kameda et al.

References 1. Wang J, Guo L-P, Chen L-Z, Zeng Y-X, Lu SH (2007) Identification of cancer stem cell–like side population cells in human nasopharyngeal carcinoma cell line. Cancer Res 67(8): 3716–3724. https://doi.org/10.1158/ 0008-5472.Can-06-4343 2. Gupta A, Harrison PJ, Wieslander H, Pielawski N, Kartasalo K, Partel G et al (2019) Deep learning in image cytometry: a review. Cytometry A 95(4):366–380. https:// doi.org/10.1002/cyto.a.23701 3. Isola P, Zhu J, Zhou T, Efros AA (2017) Image-to-image translation with conditional adversarial networks. IEEE, Honolulu 4. Selvaraju RR, Cogswell M, Das A, Vedantam R, Parikh D, Batra D (2020) GradCAM: visual explanations from deep networks via gradient-based localization. Int J Comput Vis 128(2):336–359. https://doi.org/10. 1007/s11263-019-01228-7 5. Chen L, Kasai T, Li Y, Sugii Y, Jin G, Okada M et al (2012) A model of cancer stem cells derived from mouse induced pluripotent stem cells. PLoS One 7(4):e33544. https://doi. org/10.1371/journal.pone.0033544 6. McCulloch WS, Pitts W (1943) A logical calculus of the ideas immanent in nervous activity. Bull Math Biophys 5(4):115–133. https://doi. org/10.1007/BF02478259 7. Rosenblatt F (1957) The perceptron – a perceiving and recognizing automaton. Cornell Aeronautical Laboratory, Ithaca/New York 8. Fukushima K (1980) Neocognitron: a selforganizing neural network model for a mechanism of pattern recognition unaffected by shift in position. Biol Cybern 36(4):193–202. https://doi.org/10.1007/BF00344251 9. Fukushima K (1979) Neocognitron: Neural network model for a mechanism of pattern recognition unaffected by shift in position. IEICI J62-A(10):658–665 10. Krizhevsky A, Sutskever I, Hinton GE (2017) ImageNet classification with deep convolutional neural networks. Commun ACM 60(6):84–90. https://doi.org/10.1145/ 3065386 11. Goodfellow I, Pouget-Abadie J, Mirza M, Xu B, Warde-Farley D, Ozair S et al (2014) Generative adversarial nets. Paper presented at

advances in neural information processing systems, 2014 12. Isola P, Zhu J-Y, Zhou T, Efros AA (2018) Image-to-image translation with conditional adversarial networks. arXiv:1611.07004v07003 [cs.CV], https://ui. adsabs.harvard.edu/abs/2016arXiv16110 7004I 13. Goodfellow I (2016) NIPS 2016 tutorial: generative adversarial networks. arXiv:1701.00160, https://www.youtube. com/watch?v¼HGYYEUSm-0Q 14. Karras T, Laine S, Aila T (2019) A style-based generator architecture for generative adversarial networks 15. Mirza M, Osindero S (2014) Conditional generative adversarial nets. arXiv:1411.1784, https://ui.adsabs.har vard.edu/abs/2014 arXiv1411.1784M 16. Aida S, Kameda H, Nishisako S, Kasai T, Sato A, Sugiyama T (2020) Conditional generative adversarial networks to model iPSCderived cancer stem cells. J Adv Comput Intell Intell Inf 24(1):134–141. https://doi.org/10. 20965/jaciii.2020.p0134 17. Aida S, Okugawa J, Fujisaka S, Kasai T, Kameda H, Sugiyama T (2020) Deep learning of cancer stem cell morphology using conditional generative adversarial networks. Biomol Ther 10(6):931. https://doi.org/10.3390/ biom10060931 18. Hanai Y, Ishihata H, Zhang Z, Maruyama R, Kasai T, Kameda H et al (2022) Temporal and locational values of images affecting the deep learning of cancer stem cell morphology. Biomedicine 10(5):941. https://doi.org/10. 3390/biomedicines10050941 19. Simonyan K, Zisserman A (2014) Very Deep Convolutional Networks for Large-Scale Image Recognition. https://arxiv.org/ abs/1409.1556 20. He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image Recognition. pp 770–778 21. PyTorch installation (2022) https://pytorch. org/get-started/locally/ 22. Gildenblat J (2021) PyTorch library for CAM methods. https://github.com/jacobgil/ pytorch-grad-cam

INDEX A Adult stem cells ................................................54, 55, 123 Aldefluor ............................................................39, 58, 61, 62, 67, 84–86, 88, 102, 112, 120 Aldehyde dehydrogenase (ALDH) ........................ 20, 27, 36, 39, 44, 53, 62, 66, 79, 83–88, 102, 113, 114, 153, 192, 194 Alternately activated macrophage (AAM) .......... 107, 111 Antibodies ..........................................4, 6, 12, 21–31, 38, 43, 44, 48, 57, 58, 61, 62, 65–67, 72, 74–76, 79, 88, 112, 120, 156, 159, 160, 167, 168, 170, 173, 179, 184, 188, 206, 209, 210, 214, 215 Antigen retrieval (AR) .......................... 21–23, 27, 29–31

B Blast ..................................................... 164–167, 169–172 Buffy coat .................................................... 208, 209, 214

C Cancer.............................................. 1, 19, 35, 51, 71, 84, 91, 99, 123, 136, 145, 165, 177, 192, 205, 219, 231 Cancer-associated fibroblasts (CAFs)........ 220–222, 224, 226–228 Cancer cells......................................................7, 8, 11, 28, 35, 39, 40, 51, 52, 56–58, 61, 62, 67, 76, 78, 91, 94, 100, 101, 103, 104, 108–116, 120, 139, 146, 165, 194, 205, 206, 212, 213, 215, 220–228, 231, 241 Cancer stem cells (CSCs)...........................................2, 19, 35, 51, 71, 83, 91, 99, 135, 145, 163, 177, 192, 205, 219, 231 Cancer stem like cells ................... 99–121, 192–202, 221 Carcinoma-associated mesenchymal stem cells (CA-MSC) ..............101, 113, 115, 192 Carcinoma-associated fibroblasts (CAF) ..................... 100, 101, 220, 221, 226–228 CD44 ............................................. 2, 3, 8, 20, 27, 28, 32, 36, 38, 45, 47, 53, 54, 58, 61–64, 67, 71–79, 88, 154, 167, 192, 206, 210, 214, 215 Cell sorting ..........................................46, 53, 62, 87, 188 Cell surface markers .............................................. 3, 4, 36, 53, 54, 64, 79, 88, 206, 227

Circulating cancer stem cell (CCSC) ........ 206, 208–211, 213–216 Classification............................................... 166, 231, 232, 234, 249–253 Co-culture ......................................... 111, 160, 200, 220, 221, 224–227 Co-delivery ................................. 192, 193, 196, 198–202 Colorectal cancer (CRC) ..................................4, 13, 131, 136, 206, 219, 222–224 Conditional generative adversarial network (CGAN) .................................................... 231, 232 Convolutional neural network (CNN) .............. 232, 234, 249, 251, 255 CRISPR-Cas9 system ................................................... 2, 8 CSC biomarkers ...................................................... 52, 92, 205, 207, 209, 213, 214 CSC isolation .................................................3, 53, 56, 62 CSC markers...........................................4, 19, 27, 32, 53, 55, 61, 63–66, 72, 79, 113, 148, 223, 224 CSC niche............................................................. 5, 6, 100 CSCs characterization (Characterization of CSCs).................. 56–58, 146

D Depletion ......................................................207–210, 214 Detection ................................................ 3, 13, 19–32, 36, 43–45, 52–56, 60–63, 65, 71–79, 87, 102, 192, 205, 214, 215 Detection method.....................................................24, 29 Differentiation......................................2, 3, 5, 27, 28, 42, 43, 91, 101, 107, 136, 145, 165, 225 Drug screening..................................................... 135–143 Drug sensitivity assays................................. 124, 126, 129

E Electrospray .........................................193, 195–198, 202

F Fixation ...................................................... 22, 23, 29, 155 Flow cytometry .......................................... 36, 43–44, 47, 48, 53, 57, 58, 60, 61, 66, 71–79, 84, 87, 88, 100, 112, 113, 115, 148–149, 152–153, 156, 159, 166, 167, 173

Federica Papaccio and Gianpaolo Papaccio (eds.), Cancer Stem Cells: Methods and Protocols, Methods in Molecular Biology, vol. 2777, https://doi.org/10.1007/978-1-0716-3730-2, © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

257

CANCER STEM CELLS: METHODS AND PROTOCOLS

258 Index

Fluorescence activated cell sorting (FACS) .......................................... 3, 4, 56, 67, 73, 74, 104, 105, 109, 112, 113, 115, 156, 184, 185, 188, 194, 220

G

N Nanoparticle drug delivery.................................. 200–202

O

Gastric cancer .............................................. 28, 71, 76, 78 Google Colaboratory.......................................... 234, 235, 237, 239, 243, 245, 247, 255 Grad-CAM ........................ 231, 234, 249, 251, 253–255

Organoid cryopreservation......................... 124, 126, 128 Organoid establishment....................................... 127–129 Organoid expansion............................................. 123–132 Organoids ................................................... 123–126, 128, 129, 131, 132, 135–143

H

P

Heterogeneity................................ 2–4, 8, 32, 35, 51, 91, 100, 135, 136, 145 Human blood............................. 163–173, 209, 210, 212 Hydrogels .......................... 178, 180–183, 185, 187, 188

Paclitaxel ..................................... 192, 193, 196, 198–202 Patient-derived organoids (PDOs) .................... 123–132, 136, 139, 142 Pix2pix ........................................................ 231, 232, 236, 237, 241, 243, 245–248 Poly(lactic-co-glycolic acid) (PLGA) .................. 192–202 Preclinical models.......................................................... 136 Progenitor cells ..................................................... 4, 8, 54, 91, 98, 164, 166, 231 Prostate cancer ............................................. 94, 135, 139, 141, 142, 222, 224

I Immune microenvironment .......................................7, 11 Immunofluorescence (IF).............. 53, 66, 207, 209–214 Immunohistochemistry (IHC)..................................4, 19, 21–26, 28–32, 53 Immunophenotyping.................................................... 165 Intratumoral heterogeneity (ITH)................2, 4, 7, 8, 12 Isolation .................................................. 6, 36, 43–44, 53, 56–58, 60–63, 67, 83–88, 93–94, 97, 136, 138, 145–160, 179, 181, 188, 210, 213, 214

J Jupyter Notebook ......................234, 235, 237, 239, 243

M Macrophages .............................................. 6, 8, 100, 101, 107–110, 117, 220, 225–227 Mammary glands.................................178, 180–185, 187 Markers ...................................................2, 19, 36, 52, 71, 83, 100, 146, 165, 188, 192, 206, 219, 223, 224, 227, 228, 241 MatrigelTM ......................................................... 92, 95–97, 136, 137, 139–141, 143 MatrigelTM-based hanging-drop method........... 136, 139 Mechanotransduction ................................................... 177 Mesenchymal stem cells (MSCs).................................100, 101, 103, 104, 108–111, 146, 157, 158, 160, 192, 194, 200, 220, 222 Methods................................................ 4, 6, 7, 19–32, 36, 39–47, 53, 56, 60–66, 72–77, 83–88, 92–96, 101, 103–118, 120, 126–130, 138–142, 145–160, 166, 168–172, 177–188, 192, 195–201, 205–207, 210, 212, 214, 235–253, 255 Minimal residual disease (MRD)..............................11, 13 Morphologies ........................................... 22, 94, 97, 158, 166, 197, 198, 201, 231–233

R Red blood cell (RBC) ............................................. 27, 40, 41, 72, 74, 79, 85, 86, 94, 125, 129, 139, 167–169, 173, 174, 181, 184, 208–210, 212–214

S Segmentation ....................................................... 231, 233 Self-renewal .................................................. 2, 3, 5, 8, 36, 63, 91, 92, 95, 96, 136, 145, 153, 155, 163–165, 177, 186, 221, 223 Side population ..................... 36, 38, 43, 45, 53, 55, 146 Single-cell RNA sequencing (scRNA-seq) .................. 7, 8 Spheres........................................... 36, 41–43, 53, 63, 66, 67, 87, 91–98, 157, 158, 186, 219 Spheroids ............................................................... 92, 100, 105–117, 119, 120, 145–160, 200–202, 206, 213, 215, 216 Standardization ...................................................... 21, 166 Standard of care (SoC) ..................................................... 2 Stem cells ............................................2–5, 11, 19, 27, 28, 36, 51, 52, 54, 61, 66, 67, 98, 145, 153, 154, 164, 184, 206, 209, 213, 214 Sunitinib ..................................... 192, 193, 196, 198–201

T Three-dimensional (3D)...................................... 136, 178 Troubleshooting ............................................................. 21 Tumor .............................................. 2, 19, 35, 83, 91, 99, 136, 145, 163, 177, 192, 205, 219, 233

CANCER STEM CELLS: METHODS Tumor-associated macrophages (TAMs) ........... 220–222, 224–227 Tumor cells ............................................... 3, 5, 11, 19, 35, 83, 91, 94, 95, 99, 113, 131, 139, 140, 146, 153, 157, 160, 205, 213, 214, 216, 220 Tumor-initiating cells (TICs) ..................... 2, 51, 91, 206 Tumor microenvironment (TME) ....................... 99–121, 145, 146, 193, 220, 222 Tumoroids ...............................................................99–121 Tumor sphere assay ...................................................91–98

AND

PROTOCOLS Index 259

Tumorspheres............................................. 42, 63, 65, 67, 91–98, 146, 150–153, 157, 158 Tumor tissue ............................................ 60, 92, 93, 123, 124, 138–139, 142, 233

W White blood cell (WBC) .......................... 40, 41, 85, 206, 208–210, 212–215