High Content Screening: Science, Techniques and Applications

Preface xix

Contributors xxi

Section I Essentials of High Content Screening 1

1. Approaching High Content Screening and Analysis: Practical Advice for Users 3
Scott Keefer and Joseph Zock

1.1 Introduction 3

1.2 What is HCS and Why Should I Care? 4

1.3 How does HCS Compare with Current Assay Methods? 5

1.4 The Basic Requirements to Implement HCS 8

1.4.1 Cell Banking 9

1.4.2 Plating, Cell Density, and the Assay Environment 10

1.4.3 Compound Addition and Incubation 11

1.4.4 Post-Assay Processing 11

1.4.5 HCS Imaging Hardware 12

1.4.6 HCS Analysis Software 13

1.4.7 Informatics 13

1.5 The Process 15

1.6 An Example Approach 16

1.7 Six Considerations for HCS Assays 18

1.7.1 Garbage In, Garbage Out (GIGO) 18

1.7.2 This is Not a Plate Reader 19

1.7.3 Understand Your Biology 20

1.7.4 Subtle Changes Can Be Measured and are Significant 20

1.7.5 HCS Workflow — Flexibility is the Key 21

1.7.6 HCS is Hard — How Do I Learn It and Become Proficient at It? 21

References 22

2. Automated High Content Screening Microscopy 25
Paul A. Johnston

2.1 Introduction 25

2.2 Automated HCS Imaging Requirements 26

2.3 Components of Automated Imaging Platforms 26

2.3.1 Fluorescence Imaging and Multiplexing 26

2.3.2 Light Sources 28

2.3.3 Optical Designs: Confocal Versus Wide-Field 28

2.3.4 Objectives 29

2.3.5 Detectors 29

2.3.6 Autofocus 29

2.3.7 Environmental Controls and On-Board Liquid Handling Capabilities 30

2.4 Imaging Platform Software 31

2.5 Data Storage and Management 32

2.6 Selecting an HCS Platform 32

2.7 Comparison of a SAPK Activation HCS Assay Read on an ArrayScan^Ⓡ 3.1, an ArrayScan^Ⓡ V^Ti, and an IN Cell 3000 Automated Imaging Platform 33

References 40

3. A Primer on Image Informatics of High Content Screening 43
Xiaobo Zhou and Stephen T.C. Wong

3.1 Background 43

3.2 HCS Image Processing 46

3.2.1 Image Pre-Processing 46

3.2.2 Cell Detection, Segmentation, and Centerline Extraction 48

3.2.2.1 Cell Detection 48

3.2.2.2 Particle Detection 50

3.2.2.3 Cell Segmentation 52

3.2.2.4 Centerline/Neurite Extraction 57

3.2.3 Cell Tracking and Registration 60

3.2.3.1 Simple Matching Algorithm 60

3.2.3.2 Mean Shift 62

3.2.3.3 Kalman Filter 62

3.2.3.4 Mutual Information 63

3.2.3.5 Fuzzy-System-Based Tracking 64

3.2.3.6 Parallel Tracking 66

3.2.4 Feature Extraction 66

3.2.4.1 Features Extracted from Markov Chain Modeling of Time-Lapse Images 67

3.3 Validation 67

3.4 Information System Management 69

3.5 Data Modeling 70

3.5.1 Novel Phenotype Discovery Using Clustering 70

3.5.2 Gene Function Study Using Clustering 72

3.5.3 Screening Hits Selection and Gene Scoring for Effectors Discovery 74

3.5.3.1 Fuzzy Gene Score Regression Model 75

3.5.3.2 Experimental Results 76

3.5.4 Metabolic Networks Validated by Using Genomics, Proteomics, and HCS 76

3.5.5 Connecting HCS Analysis and Systems Biology 77

3.5.6 Metabolic Networks 78

3.6 Conclusions 79

3.7 Acknowledgments 79

References 80

4. Developing Robust High Content Assays 85
Arijit Chakravarty, Douglas Bowman, Jeffrey A. Ecsedy, Claudia Rabino, John Donovan, Natalie D’Amore, Ole Petter Veiby, Mark Rolfe, and Sudeshna Das

4.1 Introduction 85

4.2 Overview of a Typical Immunofluorescence-Based High Content Assay 86

4.2.1 Staining Protocol 87

4.2.2 Sources of Variability 87

4.3 Identifying Sources of Variability in a High Content Assay 88

4.3.1 Verifying the Accuracy and Precision of Liquid Handling Procedures 89

4.3.2 Deconstruction of Immunofluorescence and Cell Culture Protocols 90

4.3.3 Control Experiments 90

4.3.4 Protocol Optimization 92

4.3.5 Antibody Optimization Using a Design of Experiments Framework 94

4.3.6 Addressing Sources of Variability in Microscopy 96

4.3.7 Optimization of Image Processing Parameters in a High Content Assay 99

4.4 From Immunofluorescence to High Content: Selecting the Right Metric 101

4.5 Validation of High Content Assays 102

4.5.1 Establishing SOPs and Reagent Stocks for Cell Culture and Immunofluorescence Staining 103

4.5.2 Linking Assay Variability to Assay Performance 104

4.5.3 Design of Assay Quality Control Measures 105

4.6 Conclusion 107

4.7 Acknowledgments 108

References 108

Section II Applications of HCS In Basic Science and Early Drug Discovery 111

5. HCS in Cellular Oncology and Tumor Biology 113
Steven A. Haney, Jing Zhang, Jing Pan, and Peter LaPan

5.1 Cancer Cell Biology and HCS 113

5.1.1 Oncology Research and the Search for Effective Anticancer Therapeutics 113

5.1.2 A General Protocol for Establishing HCS Assays Within Oncology Research 115

5.1.2.1 What is the Underlying Biology to be Evaluated in an HCS Assay? 115

5.1.2.2 What Resources are Immediately Available for Characterizing the Target or its Activity? 116

5.1.2.3 How Do the Available Reagents Perform Quantitatively? 118

5.1.2.4 What Multiplexing is Required for the Assay? 119

5.2 The Cell Biology of Cell Death 120

5.2.1 Cell Death Stimuli and Response Pathways 120

5.2.2 Induction of Cell Death Signals 121

5.2.2.1 Activation of Cell Death Receptors 121

5.2.2.2 Mitochondrial Damage 122

5.2.2.3 Mitotic Arrest, Replication Stress, and DNA Damage 122

5.2.2.4 ER Stress 122

5.2.3 Propagation of Cell Death Signals into Specific Cell Death Responses 123

5.2.3.1 Apoptosis 126

5.2.3.2 Mitotic Catastrophe 127

5.2.3.3 Autophagy 127

5.2.3.4 Necrosis 128

5.2.3.5 Senescence 129

5.2.4 Cytological and High Content Assays for Cancer Cell Death 129

5.2.4.1 Detection of Moderate and Severe ER Stress in Cancer Cells 130

5.2.4.2 Effects of Cytotoxic Therapeutics on Apoptosis and Necrosis of Cancer Cells 130

5.3 Cell Signaling Pathways in Cancer 133

5.3.1 Signal Transduction in Cancer 133

5.3.2 A Multiparametric Assay for the PI3K/AKT Pathway as Representative of Quantitative Measures of Signal Transduction in Cancer Cells 135

5.4 HCS in Tumor Biology 137

5.4.1 The Biology of Tumor Growth 137

5.4.2 An HCS Assay to Study Tumor Biology in vitro 137

5.5 Conclusions 139

References 139

6. Exploring the Full Power of Combining High Throughput RNAi with High Content Readouts: From Target Discovery Screens to Drug Modifier Studies 145
Christoph Sachse, Cornelia Weiss-Haljiti, Christian Holz, Kathrin Regener, Francoise Halley, Michael Hannus, Corina Frenzel, Sindy Kluge, Mark Hewitson, Benjamin Bader, Amy Burd, Louise Perkins, Alexander Szewczak, Stefan Prechtl, Claudia Merz, Peter Rae, Dominik Mumberg, and Christophe J. Echeverri

6.1 Background: The Convergence of High Content Analysis and RNAi 145

6.2 Integrating HT-RNAi and HCA in Drug Discovery: The Potential 146

6.2.1 Technology Platform, HCA, and HT-RNAi Methodologies 146

6.2.2 Key Applications of HT-RNAi Combined with HCA in Drug Discovery 148

6.2.2.1 Target Discovery Screens 148

6.2.2.2 Target Validation Studies 149

6.2.2.3 Drug Mechanism of Action Screens 149

6.3 Combining RNAi and HCA in One Assay — The Reality 150

6.3.1 General Considerations 150

6.3.1.1 Choice of the Right Cell Model 150

6.3.1.2 Establishment of an RNAi Delivery Protocol 150

6.3.1.3 Assay Optimization 151

6.3.2 Applications: Combining HCA with HT-RNAi to Integrate Functional Validation Directly Within Target Discovery Studies 151

6.3.2.1 Multipass Strategies for Systematic Screens 151

6.3.2.2 Hurdles and Caveats 152

6.3.2.3 Example: A Multiparametric Oncology Assay Platform 155

6.3.3 RNAi Target Validation Studies 159

6.3.3.1 Functional Profiling 160

6.3.3.2 Transcriptional Profiling 160

6.3.3.3 Cytological Profiling 160

6.3.3.4 Pathway Profiling 160

6.3.3.5 Target Titration 160

6.3.4 RNAi Drug Modifier Screens 161

6.4 HCA-Based RNAi Studies — The Future 164

6.5 Acknowledgments 166

References 166

7. Leveraging HCS in Neuroscience Drug Discovery 169
Myles Fennell, Beal McIlvain, Wendy Stewart, and John Dunlop

7.1 High Content Screening and Drug Discovery 169

7.2 The Neuron and Neuronal Morphology 169

7.2.1 What Does Morphology Tell Us About Neuronal Function? 170

7.3 Methods for Measuring Neuronal Morphology 172

7.3.1 Traditional Methods 172

7.3.2 Available HCS Systems for Neuronal Morphology Measurements and Evolution of Technology 173

7.3.3 Methods for Imaging Neurons and Types of Morphologic Measurement 176

7.4 Small Molecule Screening for Neurite Outgrowth 178

7.5 RNAi in Neuroscience and HCA 179

7.6 Measurement of Signal Transduction in Neurons 180

7.7 High Content Screening in Complex CNS Models 181

7.8 Methods Used in Neuronal HCS 182

7.8.1 Preparation of Neuronal Culture Samples for HCS Morphology Analysis 182

7.8.2 Culture Fixing 183

7.8.3 Immunocytochemistry 183

7.8.4 Neurite Morphology Measurement and Analysis 184

References 185

8. Live Brain Slice Imaging for Ultra High Content Screening: Automated Fluorescent Microscopy to Study Neurodegenerative Diseases 189
O. Joseph Trask, Jr., C. Todd DeMarco, Denise Dunn, Thomas G. Gainer, Joshua Eudailey, Linda Kaltenbach, and Donald C. Lo

8.1 Introduction and Background 189

8.2 Live Brain Slice Model to Study Huntington’s Disease 191

8.3 Imaging Platforms 191

8.4 Center of Well (COW) for Image Processing 194

8.5 Generic Protocol for the Cellomics ArrayScan VTI 197

8.6 Data and Results 197

8.7 Discussion 201

References 203

9. High Content Analysis of Human Embryonic Stem Cell Growth and Differentiation 205
Paul J. Sammak, Vivek Abraham, Richik Ghosh, Jeff Haskins, Esther Jane, Patti Petrosko, Teresa M. Erb, Tia N. Kinney, Christopher Jefferys, Mukund Desai, and Rami Mangoubi

9.1 Introduction 205

9.2 Cell Culture Methods 206

9.2.1 Maintaining Pluripotency 206

9.2.2 Cardiomyocyte Differentiation 207

9.2.3 Neuronal Differentiation 207

9.3 Statistical Wavelet-Based Analysis of Images for Stem Cell Classification 207

9.3.1 Motivation for Algorithm Development 207

9.3.2 Measuring Amorphous Biological Shapes 209

9.3.3 Texture and Borders as Biological Features 210

9.3.4 Texture Analysis 210

9.4 Molecular Analysis of Pluripotency and Cell Proliferation in Undifferentiated Stem Cells 214

9.4.1 Methods 215

9.4.2 Analysis of Pluripotency and Cell Proliferation in Undifferentiated Stem Cells 215

9.5 Analysis of Cardiomyocyte Differentiation 218

9.6 Analysis of Neuronal Differentiation 219

9.6.1 Methods 219

9.6.2 Analysis of Neurectodermal Intermediates in Early Differentiated hESC 220

9.6.3 Analysis of Neuronal Processes 221

References 221

Section III HCS In Drug Development 225

10. HCS for HTS 227
Ann F. Hoffman and Ralph J. Garippa

10.1 Introduction 227

10.2 HCS for Orphan GPCRS and Transfluor 228

10.3 HCS for Multiparameter Cytotoxicity Screening 236

10.4 Discussion 243

10.5 Summary 246

References 246

11. The Roles of High Content Cellular Imaging in Lead Optimization 249
Jonathan A. Lee, Karen Cox, Aidas Kriauciunas, and Shaoyou Chu

11.1 Introduction 249

11.2 Statistical Validation of Assays 250

11.3 High Content Cellular Imaging is a Diverse Assay Platform 251

11.4 Use of High Content Cellular Imaging for Oncology Research at Eli Lilly 255

11.4.1 Cell Cycle and High Content Cellular Imaging 255

11.4.2 Advantages of High Content Cellular Imaging 256

11.4.2.1 Rare Cell Populations 256

11.4.2.2 End Point Multiplexing 257

11.4.2.3 Advantages of Multiplexing 259

11.5 The Future of High Content Cellular Imaging in Lead Optimization 261

11.6 Acknowledgments 264

References 264

12. Using High Content Analysis for Pharmacodynamic Assays in Tissue 269
Arijit Chakravarty, Douglas Bowman, Kristine Burke, Bradley Stringer, Barbara Hibner, and Katherine Galvin

12.1 Introduction 269

12.1.1 Preclinical Models 269

12.1.2 Pharmacokinetics/Pharmacodynamics (PK/PD) 270

12.1.3 PK/PD Approaches in Practice 271

12.2 Designing a High Content Assay for Use in Tissues 272

12.2.1 Preliminary Biomarker Characterization 272

12.2.2 Development and Validation of HC Assays in Tissue 273

12.3 Technical Challenges in Establishing High Content Assays for Tissue 274

12.3.1 Logistical Challenges in Tissue Staining and Acquisition 274

12.3.2 Plane-of-Focus and Plane-of-Section Issues 275

12.3.3 Heterogeneity in Tissue Samples 277

12.3.4 Automated Detection of Areas of Interest 279

12.3.5 Segmentation and Background Issues in High Content Assays 282

12.3.6 Variability in Staining 284

12.4 Case Study: Design and Validation of a High Content Assay for Biomarker X 286

12.5 Conclusions 289

12.6 Acknowledgments 290

References 290

13. High Content Analysis of Sublethal Cytotoxicity in Human HepG2 Hepatocytes for Assessing Potential and Mechanism for Chemicaland Drug-Induced Human Toxicity 293
Peter J. O’Brien

13.1 Introduction 293

13.1.1 Past Failure of Cytotoxicity Assessments 293

13.1.2 Development of a Novel Cellomic Cytotoxicity Model 295

13.1.3 Parameters Monitored in the Cellomic Cytotoxicity Model 296

13.1.4 Materials and Methods 300

13.2 Results from High Content Analysis of Human Toxicity Potential 301

13.3 Discussion 307

13.3.1 Applications of the Cellomic Cytotoxicity Model 307

13.3.2 Limitations of the Cellomic Cytotoxicity Model 308

13.3.3 Future Studies 309

13.4 Acknowledgments 309

13.5 Appendix: Detailed Methods 309

13.5.1 Materials 309

13.5.2 Methods: Cell Culture 310

13.5.3 Subculture of HepG2 Cells 310

13.5.4 Poly-D-Lysine Coating 311

13.5.5 Drug Treatment Protocol for Three-Day Plates 311

13.5.6 Drug Solubility 312

13.5.7 Preparing the Drug Plate 312

13.5.8 Indicator Dye Loading Procedure 312

13.5.9 KSR Protocol: Fluorescence Settings 313

13.5.9.1 Data Capture 314

13.5.9.2 Assay Protocol Settings 314

13.5.9.3 Plate Protocol Settings 315

13.5.9.4 Quality Control 315

References 315

Section IV Data Management, Data Analysis, and Systems Biology 317

14. Open File Formats for High Content Analysis 319
Jason R. Swedlow, Curtis Rueden, Jean-Marie Burel, Melissa Linkert, Brian Loranger, Chris Allan, and Kevin W. Eliceiri

14.1 Introduction 319

14.2 The Data Problem in Biology: Why is it so Hard? 319

14.3 High Content Data in Biology: A Definition 320

14.4 The Difference Between a File Format and a Minimum Specification 321

14.5 File Formats: Open vs Closed 321

14.6 File Formats: Balancing Flexibility with Standards 323

14.7 Supporting a Successful File Format 323

14.8 Commercial Realities: How Users and Developers Can Define File Formats 324

14.9 OME-XML and OME-TIFF: Moving Towards a Standard Format For High Content Biological Data 324

14.9.1 Metadata Support for High Throughput Assays 326

14.10 Data Model and File Format Integration: Towards Usable Tools 327

14.11 Conclusions 327

14.12 Acknowledgments 328

References 328

15. Analysis of Multiparametric HCS Data 329
Andrew A. Hill, Peter LaPan, Yizheng Li, and Steven A. Haney

15.1 Cytological Classification and Profiling 329

15.1.1 Multiparametric HCS Data and Cytological Profiling 329

15.1.2 Cytological Features 330

15.1.3 Using Cytological Features in Assays 331

15.2 Setting Up Cytological Profiling Studies 333

15.2.1 Planning for a Cytological Classification Experiment 333

15.2.2 Feature Extraction by Image Analysis and Export of Data for Analysis 335

15.2.3 Example Studies that Use Cytological Profiling to Study Small Molecule Inhibitors and siRNAs 336

15.3 Sources of Variability and Corrections 336

15.3.1 Detection and Elimination of Plate and Sample Outliers from a Data Set 336

15.3.2 Visualization of Plate-Level Features to Assess Data Quality 337

15.3.3 Normalization and Scaling of Data 340

15.3.4 Post-Normalization Analysis of Data Quality 341

15.4 General Analysis Considerations 341

15.4.1 Choosing the Appropriate Analysis Level: Well or Cell 342

15.4.1.1 Cell Cycle Analysis 342

15.4.1.2 Perturbations Where the Cell is an Effective Experimental Block 342

15.4.2 Statistical Summaries for Cell-Level Features 343

15.4.3 Feature Relationships, Redundancy, and Selection 343

15.5 Data Analysis Methods 346

15.5.1 Feature Transformation 346

15.5.2 Linear Modeling of Feature Responses 347

15.5.3 Unsupervised Clustering Methods 348

15.5.4 Supervised Classification Methods 351

15.6 Software for HCS Data Analysis 352

15.7 Conclusions 352

References 353

16. Quantitative and Qualitative Cellular Genomics: High Content Analysis as an End Point for HT-RNAi Phenotype Profiling Using GE’s IN Cell Platform 355
David O. Azorsa, Christian Beaudry, Kandavel Shanmugam, and Spyro Mousses

16.1 Cellular Genomics 355

16.2 Enabling Technologies to Facilitate Cellular Genomics: RNA Interference 357

16.3 High Throughput RNAi (HT-RNAi) 358

16.3.1 Platforms and Screening Infrastructure 358

16.3.2 Establishing Methods for Successful HT-RNAi 358

16.4 High Content Analysis (HCA) for High Throughput Phenotype Profiling 362

16.4.1 IN Cell Analyzer 1000 363

16.4.2 IN Cell Analyzer 3000 363

16.4.3 HCA Assay Suites 363

16.4.4 Fixed-Cell Assays 366

16.4.5 Live-Cell Assays 366

16.5 Future Directions 368

References 368

17. Optimal Characteristics of Protein–Protein Interaction Biosensors for Cellular Systems Biology Profiling 371
Kenneth A. Giuliano, David Premkumar, and D. Lansing Taylor

17.1 Introduction 371

17.2 Challenge of Cellular Systems Biology (CSB) 372

17.3 Optimal Characteristics of Protein–Protein Interaction Biosensors (PPIBs) 373

17.4 Example of a PPIB and Cellular Systems Biology Profiling 375

17.4.1 Testing a First-Generation p53–HDM2 PPIB Based on Full Length and Protein Fragments 375

17.4.2 Overexpression of a Labeled p53 Fusion Protein Modulates Multiple Cellular Systems: Testing a Critical Potential Problem 379

17.4.3 An Optimized p53–HDM2 PPIB 380

17.5 Summary and Prospects 384

17.6 Acknowledgments 385

References 385