July 23, 2018
The August 2018 SLAS Discovery Special Issue on High-Throughput Flow Cytometry in Drug Discovery captures recent advances that inform us about the current state of automated workflows, the expanding range of applications and the creativity in some novel assays. The issue is timely as “high-throughput flow cytometry (HTFC) is gaining acceptance and people are using it in innovative ways,” says Co-Guest Editor Bruce Edwards, Ph.D., of the University of New Mexico (UNM, Albuquerque, NM, USA).
SLAS Discovery Editorial Board member and Co-Guest Editor Mei Ding, Ph.D., of AstraZeneca (Gothenburg, Sweden) and Edwards teamed up last fall. “We put out the call for papers and were excited by the wide range of screening applications, sample preparation/acquisition and data analysis solutions for high-throughput flow cytometry (HTFC) represented in these reports,” Edwards says.
The power of flow cytometry is playing an increasing role in in high-throughput screening (HTS) assays used in drug discovery programs, Edwards explains. “Flow cytometry has been used in clinical and basic research as a low-throughput device for decades, but recent advances in automation of sample handling and data analysis are allowing this technology to contribute to the early stages of drug discovery.” A recent SLAS Discovery review article by Edwards and Larry A. Sklar of UNM highlighting flow cytometry’s capability to rapidly and accurately measure multiple parameters of single cells and potential applications to HTS screening assays provides the background for appreciating the latest innovations and solutions highlighted in the August 2018 special issue. The selected articles contain detailed information about automated systems used in the screening pipeline, a wide range of applications in drug and biologics discovery programs and use in response profiling of primary cells.
Edwards and his group at UNM use evolving systems to speed throughput for flow cytometry. Edwards says the newer instruments are getting smaller and are more user friendly and points out that the flow cytometers that work well in HTS systems are not the ones conventionally thought of, which require some expertise to run well. Simplified instruments for HTS retain the wide dynamic range and important features of high-end flow cytometers but limit the user’s ability to change most settings. Edwards feels “the simpler systems that don’t allow users to muck around with many adjustments are the ones that are going to be the most useful from a screening perspective because the instrument will be performing exactly the same way every time.”
Flow cytometers can measure multiple parameters on tens of thousands of cells very quickly, but the front and back end processes were not automated or integrated and had to be optimized for HTS. “The three main challenges to using flow cytometry for HTS are sample acquisition, sample preparation and data analysis,” says Edwards.
As with other technologies adapted to HTS, miniaturizing sample sizes to conserve reagents and enable rapid automated sample acquisition from 384-well and 1536-well formats was essential. Edwards was involved in the development of the HyperCyt sampling technology, used in IntelliCyt systems (Albuquerque, NM, USA), that loads 1-5 µL samples serially with air bubbles separating them. “This enables all the data from a plate to be acquired in a single data file which you are able to parse into the individual wells,” says Edwards.
Working with cells or particles in suspension and complex multiplexed assays makes sample preparation complicated, and in many instances, involves repeated aspirating, dispensing, centrifuging and mixing.
Joslin et al. present detailed information about the first fully automated HTFC system designed and built at the Genomics Institute of the Novartis Research Foundation (GNF, San Diego, CA, USA). The GNF system is highly customized as it was developed before integrated systems were commercially available. The article describes use of their system with several phenotypic screening assays achieving throughput of 50,000 wells per day using 384-well or 1536-well plates. Joslin et al. state phenotypic assays requiring analysis of multiple surface markers drove their development of a custom automated liquid handling device that deals with the multiple dispensing and aspirating steps in sample preparation needed to produce robust data. “Even though flow cytometry is pretty good at handling background fluorescence, there are certain assays where you have to centrifuge the samples, remove supernatant and add another reagent. I was impressed with the GNF automation of these processes with very small volumes,” says Edwards.
AstraZeneca also incorporates automated HTFC into its drug discovery programs. As described by Ding et al., her research team adopts systems commercialized by IntelliCyt and can screen a 500,000 small molecule library using 384-well or 1536-well plates or run phenotypic assays in the 384-well format. The AstraZeneca article explains how they have integrated these systems into their programs and some of the practical challenges and best practices for HTFC including details about sample preparation prior to sample acquisition.
According to Edwards, “Impressive amounts of data are generated with flow cytometry, and unlike imaging in which you throw away most of the data, most of these data are useful. Particularly in highly multiplexed assays in 1536-well formats, huge amounts of data are generated, and to put it mildly, analyzing it is a challenge, especially for the raw data. Many of the articles in this issue rely on pre-processed data by the IntelliCyt system that parses the data into spreadsheet-compatible formats. Third party screening software, for example, Genedata Screener for HT Flow (Basal, Switzerland) has also been adapted to evaluate the data for discrepancies across the plate and artifacts due to edge effects.” Even with some preprocessing there may be multiple post-processing steps and software programs used to translate the acquired data into something usable. Edwards feels “the software to manage these data is improving but is not quite there yet.”
The original research articles in this issue highlight a number of applications and novel approaches to gathering data with HTFC.
Edwards admits to finding the article from Martinez et al. “the most exciting because it highlights field of immuno-oncology, which is a very hot area these days and one for which flow cytometry is going to be particularly useful. Also, I liked it because they are working with solid (adherent) tumor cell lines rather than cells from blood tumors that are naturally in suspension. The trick with adherent cells is getting them into suspension without physically damaging them, disrupting the membrane or destroying the proteins of interest. And, of course, there are concerns about altering them physiologically by detaching them from a substrate.” Martinez et al. described a high-throughput multiparameter HTFC assay to characterize the CAR-T cell function and cytotoxity against solid tumor cells in a single assay and suggest that additional measureable parameters could be added to learn even more.
Another report that impresses Edwards addresses epitope binning, because Edwards appreciates an “elegant solution to a complex problem.” Chan et al. inform readers that in the therapeutic antibody arena, information about the epitopes to which an antibody binds provides clues about the antibody’s functional capacity, which is an innate property that cannot be engineered. Edwards explains, “There is great interest in being able to profile epitopes of the antigens that antibodies are binding to because antibodies that bind similar target epitopes often share a similar function. This allows grouping antibodies by the epitopes they bind to, which can be a useful step in early drug discovery.” The technique created by Chan et al. leverages the multiplexing capabilities of HTFC by using cell barcoding and spectrally distinct beads to rapidly generate competitive binding profiles of antibodies. They group antibodies with similar binding profiles together, which can speed up identification of targets early in the discovery process.
Edwards also appreciates the clever use of bar coding in the Zhao et al. article. “Instead of using the bar codes to analyze different types of cells in suspension, they barcode the cells and add them to the assay at different time points. This allows them to determine the time course of the biological response in a single well, reducing the chance of missing the response by choosing only an early or late time point to evaluate. I thought it was a good idea to potentially enhance what you can achieve during screening.”
“Researchers in the pharmaceutical industry are very interested in being able to study primary cells but there is a limited number of cells that you can get from a single patient, so being able to work with small numbers using 384-well and 1536-well plates cells is a big advantage. These formats decrease waste and enable researchers to analyze almost every single cell that they can get. Profiling the response of primary cells to different conditions or treatments in an efficient manner is important to moving personalized medicine forward,” says Edwards. Fan et al. report on a miniaturized HTFC 384-well format assay that evaluates immune cell functionality and treatment effect using only one-tenth the number of patient cells and measuring multiple functional readouts in one well. Perez et al. report on an HTFC-based dose-response testing scheme to quickly identify small molecule inhibitors that are efficacious in a heterogeneous population of primary cells.
The SLAS Discovery Special Issue highlights a wide range of advances and applications of HTFC technology. Edwards feels “great strides have been made in alleviating the bottlenecks on the front and back end of the process but there are still challenges to overcome.” He points out that Javarappa et al. highlights one of the current limitations of HTFC—large cell numbers. Because they evaluate subpopulations of cells that are rare, they analyze a large number of cells in each sample and this limits their rate of throughput.
“Managing the volume of data efficiently feels like a chronic issue,” Edwards emphasizes. “At this time there are multiple steps needed to process the data into forms usable by researchers. Making those processes smoother and taking data analysis to the next level, with more combinations of analysis, are things to work toward and look forward to. What we have today is really not where it needs to be.
“Another issue inherent in today’s flow cytometers is size limitations on both the upper and lower ends of the scale. On the low end, there are biological particles of extreme interest right now—exosomes—that are very difficult to study due to their small size. On the high end, multicellular structures—spheroids, for instance—are handled only on very low-throughput instruments,” Edwards explains.
Edwards sees HTFC making important contributions in the areas of inflammatory diseases, immuno-oncology, viral diseases and exploration of microbiomes that are incredibly complex. Looking further ahead, Edwards sees the realm of microfluidic systems as promising with the use of acoustic focusing and also the development of novel probes to interrogate RNA.
The SLAS Discovery Special Issue on High-Throughput Flow Cytometry in Drug Discovery features 11 original research reports, one perspective, two technical notes and one application note from Finland, Japan, Sweden and the U.S. The issue is now available at SLAS Discovery Online for SLAS Premier members, SLAS Discovery subscribers and pay-per-view readers. Free public access becomes available one year after final publication. The following papers are included in the special issue:
High-Throughput Screening Approach for Identifying Compounds That Inhibit Nonhomologous End Joining
Bredemeyer et al.
A High-Throughput Flow Cytometry Screen Identifies Molecules That Inhibit Hantavirus Cell Entry
Buranda et al.
Automated High-Throughput Flow Cytometry for High-Content Screening in Antibody Development
Wang et al.
Expediting Antibody Discovery with a Cell and Bead Multiplexed Competition Assay
O’Rourke and Liu
Comparing Flow Cytometry QBeads PlexScreen Assays with Other Immunoassays for Determining Multiple Analytes
Ding et al.
A Multiplexed Screening Assay to Evaluate Chemotherapy-Induced Myelosuppression Using Healthy Peripheral Blood and Bone Marrow
Javarappa et al.
A Fully Automated High-Throughput Flow Cytometry Screening System Enabling Phenotypic Drug Discovery
Joslin et al.
A Scalable Pipeline for High-Throughput Flow Cytometry
Wilson et al.
Application of High-Throughput Flow Cytometry in Early Drug Discovery: An AstraZeneca Perspective
Ding et al.
High-Throughput Flow Cytometry Drug Combination Discovery with Novel Synergy Analysis Software, SynScreen
Perez et al.
High-Throughput Flow Cytometry Identifies Small-Molecule Inhibitors for Drug Repurposing in T-ALL
Perez et al.
Flow Cytometry: Impact on Early Drug Discovery
High-Throughput Flow Cytometry for Drug Discovery: Principles, Applications and Case Studies
Automated Blood Sample Preparation Unit (ABSPU) for Portable Microfluidic Flow Cytometry
IntelliCyt Testimonial by Steve Rees, AstraZeneca
Ecological Stability Properties of Microbial Communities Assessed by Flow Cytometry
