Many of the efforts at IntelliCyt the past few years have been focused on developing products that can be used in antibody discovery. In light of that focus, it is truly exciting when an antibody with potential to impact global health is discovered.
One such discovery was reported this week in Science. The discovery addresses a significant missing link in development of universal therapies and a vaccine. The research was conducted by a team of scientists at Crucell and Scripps Research, along with collaborators from the Center of Influenza Research at the University of Hong Kong.
The researchers report they have discovered a human antibody that protects against essentially all influenza A and B strains. This discovery may pave the way to a universal flu vaccine and an actual treatment for patients infected with the flu. Currently patients infected with flu viruses are given only supportive treatment while their bodies fight off the flu and the immunocompromised and elderly are particularly at risk of death from flu.
The potential health impact of this influenza antibody discovery is huge. The human risk in annual flu epidemics, as reported by a 2009 WHO report results in about three to five million cases of severe illness, and about 250,000 to 500,000 deaths per year. Additionally, the economic impact related to infections in livestock and poultry is enormous, reported in the billions.
“To develop a truly universal flu vaccine or therapy, one needs to be able to provide protection against influenza A and influenza B viruses,” said Ian A. Wilson, Hansen professor of structural biology at Scripps Research and the new study’s senior investigator. “With this report, we now have broadly neutralizing antibodies against both.
To find protective antibodies against the Influenza B virus, scientists at Crucell generated flu antibodies from immune cells of volunteers who were given seasonal vaccinations. They screened this collection for antibodies that could bind with influenza B strains. Three of these antibodies – CR 8033, CR8071 and CR9114 protected mice against lethal doses of two major strains of Influenza B. CR9114 also protected mice against Influenza A virus, including the H1N1 subtype that killed 17,000 people in a 2009.
As these antibodies protected against a variety of flu strains, it became clear to scientists that they bound themselves to functionally important portions of molecules or epitopes in the virus that were unchanging from one strain to another.
Using an electron microscope and from X-ray crystallography studies, scientists from Scripps Research found out that while the antibody CR 8033 bound to a highly conserved epitope on the head of the hemagglutinin protein found in the outer coat of the flu virus, antibody CR8071 bound to the base of the hemagglutinin protein. The antibodies prevented the virus particles from exiting infected cells and thus neutralized them.
While the potential for this antibody is great, researchers acknowledge that further research will need to be done to see if it performs as well in humans as in mice. Still, it is quite exciting to think that soon there may be a single vaccination against the flu that will protect season after season and potentially save a lot of lives-humans as well as other animals.
Download this Webinar: Phenotypic Screening using GFP Reporter Cells to Accelerate the Search for Novel Acute Myeloid Leukemia (AML) Therapies
David B. Sykes M.D., Ph.D., a hematology researcher and practicing physician who specializes in treating patients with myeloproliferative disorders, at Massachusetts General Hospital presented a very timely and interesting webinar on a very elegant assay that was developed in order to screen for compounds that could potentially treat AML (Acute Myeloid Leukemia).
Dr. Sykes presentation was highly educational by describing the problems with AML and how to develop a high-throughput flow cytometry-based screening solution to discover compounds to treat this leukemia. I watched it twice because it was really good.
Additionally, Dr. Sykes presentation outlines many of the steps that one should explore in developing a high-throughput, high content screening assay and compares several platforms that he investigated in order to optimize his screen. I have summarized below to identify major items discussed in the webinar. The webinar is posted on the IntelliCyt website Resource section so you can download and share with your colleagues.
NORMAL BLOOD CELL DEVELOPMENT VS ACUTE MYELOID LEUKEMIA
- 150,000 White Blood Cells are generated each second; that is >12Billion/day
- Normally, there is a 2 day life span for mature cells
- Mistakes along the way—in mature cells apoptosis takes care of them
- In AML, there is a proliferation of immature cells which accumulate causing disease
DEVELOPING AN ASSAY AND A MODEL CELL LINE:
- Developing a model that can be adapted to a high throughput screen
- Easier way- Incorporating a built in GFP reporter
- Let the cell tell us when differentiation occurs.
- GFP marker of differentiation from transgenic mouse\Are there any small molecules that stimulate cells for differentiation?
- Screening overview-needs to be fast, simple and good window of detection
- Non adherent cells: # live cells, % mature cells
- Tricking cells into maturing (positive control)
FOUR PLATFORMS (INSTRUMENT/PROBES) EVALUATED FOR A HIGH THROUGHPUT SCREEN
Laser Scanning Acumen Explorer GFP Reporter Assay:
- Experienced edge effects
- Cells clumping on top of each other
- Not evenly dispersed
- Couldn’t differentiate if cells are there and negative, or cells are dead
Typical Flow Cytometry with Mab surface markers:
- MAC-1, GR-1 – antibody staining provided good resolution but with wash steps-not amenable to a large screen
- 384 files per microplate = 384,000 files to manage in 1000 plate screen
High Content Image-Image Express: Functional Assay for Phagocytosis:
Results were good, but slow =45 min/plate
Observed Edge Effects
HyperCyt-enabled Flow Cytometry Screening
Easy to distinguish live/dead cells by label free light scatter properties
No edge effects
No issue of clumping with cells
GFP-non overlapping populations-pos/neg
Z’ statistics indicate this is a very good assay (z’ =0.89)
Multiplexing-MAC-1 APC no wash identifies real cell effects
Phenotypic Screening GFP reporter by HTFC – HyperCyt-enabled flow cytometry
- Are there any small molecules that stimulate cells for differentiation?
- Non-adherent cells measuring # live cells, % mature cells
- Let the cell tell us when differentiation occurs with GFP reporter.
- Great window of detection – flow cytometry based
- Fast- each plate is a single file
The initial screen was performed in conjunction with the flow cytometry research core laboratory at Mass General and with researchers at the Broad Institute. An ongoing project, in collaboration with the University of New Mexico, is utilizing the GFP reporter assay to screen 350,000 compounds and results will be published to the NIH PubChem site. It is very exciting to see what results and medical innovations will result from this research, Our sincere gratitude to Dr. Sykes for sharing this research with us and the drug discovery community Enjoy the webinar!
A recent article in GEN reviewing CHI’s High Content Analysis meeting reports: “The initial resistance that the high-content/high-throughput community had to face has been overcome, and now most in the field acknowledge that flow cytometry is enormously high throughput and very high content.” You may have heard a little chuckling in the background as often happens when the obvious is stated dramatically. For a technology that has been able to perform population phenotyping with a combination of label free and multiple fluorescent probes at rates of nominally 10,000 cells per second and up – it’s about time someone acknowledged that indeed this is high content and high throughput. More to the point, I would suggest that, semantics aside, while flow is definitely high content and with HyperCyt technology, high throughput, there have been some significant advances in how flow cytometry technology is being integrated into automated processes that is making the difference in the utilization of flow cytometry screening assays. I will highlight a few comments from speakers at the HCA meeting that discuss those advances.
Pushing the boundaries of current technology: Larry Sklar, Ph.D., director, University of New Mexico Center for Molecular Discovery, continues to push the boundaries on the application of flow cytometry for drug discovery, exploring the molecular interactions at cell surfaces that lead to intracellular signaling. In his presentation, Dr. Sklar described his group’s use of a 384-well plate-based HyperCyt® high-content flow cytometry screening system from IntelliCyt. In development is a 1,536-well system capable of generating 250,000–500,000 data points/day. Their research focuses on the regulation of integrin, an adhesion molecular present on the surface of white blood cells.
Miniaturizing assays: Typically assays run on flow cytometers require a considerable volume per well in order to perform analysis. Dr. Sklar’s lab has routinely performed assays with less than 10ul total volume per well and actually sampling less than 2ul/well. This reduction in sample volume translates into massive cost savings that enable primary screens of large libraries.
Automating Sample Preparation and Data Analysis: Also speaking at the HCA conference was J. Paul Robinson, Ph.D., SVM professor of cytomics and professor of biomedical engineering at Purdue University. Dr. Robinson talked about the ability to automate flow cytometry-based high-content screens and data processing to perform multiplexed high-content analysis in real-time, running multiple assays and samples in parallel, and to miniaturize the technology to conserve resources and reduce costs, utilizing HyperCyt technology. Dr. Robinson explained that the main challenge in using flow cytometry in drug discovery is data extraction issues. ” Eliminating operator error is the main challenge in sample prep and multiparametric data analysis and automating the entire process makes it possible to perform functional phenotypic profiling of cell populations efficiently.”
Multiplexing populations: Peter Krutzik, Ph.D., a senior scientist in Gary Nolan’s group at Stanford University, described the use of phosphor flow cytometry to profile perturbations in signaling networks when peripheral blood cells are exposed to drug compounds. Dr. Krutzik stated, “This method allows for multiplexed analysis of cell types and signaling pathways, and has applications in drug target identification and throughout the drug development workflow, including in the analysis of patient samples during clinical testing.”
The overriding message of these pioneers of high content/high throughput flow cytometry is that it is not only possible, but practical to use flow cytometry for primary and secondary screens and evaluate the effects of drugs on population-specific cell function. Finally we are able to “to ask the what-if questions” and begin to integrate pathways
For the last decade and a half I have worked to take High Content Imaging (HCI) from an idea to a useful tool that is changing the way researchers approach the workflow of drug discovery and, more broadly, life science research. Turning fluorescence microscopy into a detection mode for screening cells took a leap of faith. Believing the hurdles of automating the manual process of finding the cells, reproducibly acquiring cell images across the visible spectrum, and algorithmically extracting biologically relevant information, could be overcome, we pushed on. Eventually, through the cross functional efforts of many gifted scientists and engineers, High Content Screening was born. The main characteristics of the high content approach that make it so compelling and unique compared to high throughput methods are:
- Individual, intact cell measurements
- Multiplexed assays per well and cell
- Morphological phenotyping
- Artifact rejection and subpopulation isolation
These features and their utility formed the basis for countless novel assays, creating new ways of assessing biology from cell inception to cell death and everything in between. I am extremely proud to have been a part of it.
There were, however, limitations to High Content Imaging. The underlying engine of microscopy meant that only cells associated with the bottom of the well, either growing on it like adherent cells or associated with it by settling, centrifuging, or immobilizing in a matrix (like suspension cells) could be focused on accurately enough to acquire the images needed. For suspension cells which are biologically sensitive to contact with the plate, this caused a great deal of variation, so adherent cells have become the norm in HCI assay formats. Another limitation of the approach lies in the battle between resolution of the objects and the number of objects collected. With imaging, you can’t have both, so you either end up with a compromise to collect a lot of low resolution objects (if your assay does not require higher resolution) or you end up collecting a lot of high resolution images in order to get enough objecst to statistically represent your biology. Unfortunately, there is a cost associated with storage and analysis of large number of images and many of you have multiple terabyte server farms of images and distributed processing engines whirring away to analyze those images.
But ‘high content’ is still fantastic because of the kinds of data that can be generated and cross correlated in the well and sometimes to the cell. But not having a high content screening tool optimized for suspension cells always bugged me…and by the number of inquiries about sticking cells down that I would get from the high content community, I guess it bugged you too.
But there is high content technology, as defined by the bullets above, for suspension cells that has been in use for the last 30+ years, and of course I am talking about flow cytometry. Flow cytometry has all the attributes of a high content approach, but was designed and optimized for suspension cells, beads, microbes, and mixtures of these objects. I have read many articles on our industry over the years and flow cytometry is always mentioned as a great technology, but lacking the capacity and simplicity to be a true screening tool. Imagine my delight when I found out, through the grapevine, that IntelliCyt, headed up by one of my early Cellomics colleagues, was commercializing technology that would overcome the limitations of current flow cytometry by combining their novel approach to high speed sample loading with a revolutionary, and widely accepted flow cytometer that dramatically reduced complexity, adjustments, and cost. As I discovered more about the potential and future directions this technology was about to take, it became apparent to me that I had found the missing link— a technology to fill the gap in the high content toolbox. All the projects to stick cells down, all the times we treated cells like “nails” because we only had a high content “hammer”, could be solved by the introduction of a high content suspension screening platform! After a lot of soul searching, I decided that this was too compelling of an opportunity to ignore, and I moved by family out to the desert of New Mexico and joined IntelliCyt as the Director of Product Management.
So, this new high content tool needed a name, something short and sweet that linked the technology to the high content family of tools but still retained the essence of what makes it special. High Capacity Flow (HCF) is the name we chose, because “capacity” is the value we bring to a technique that was high content years before we invented high content. Our vision at IntelliCyt is that HCF and HCI become a complementary set of tools used by the high content community, where suspension screening releases an entire class of cells, beads, microbes, and mixtures from the shackles of compromise.
More to come on the advantages of HCF and where it might be used to optimize your high content workflow….
A friend sent me a link to a blog a couple of days ago: Phenotype Reigns Supreme by Derek Lowe. I am familiar with phenotyping, as it has been a mainstay application for flow cytometry for decades and so I definitely wanted to know what exactly it was reigning over. First, I googled “phenotype”(results=14 million) just to make sure that I hadn’t missed the formation of a new political party this week.
OK, now I was ready to delve into Derek’s blog about the phenotype. Derek reviewed an article entitled: How were new medicines discovered?. The paper’s authors, David Swinney and Jason Anthony, looked back at 259 drugs approved between 1999 and 2008, and noted of the 75 first-in-class drugs, 28 were assigned to phenotypic assays and only 17 to target-based approaches. Now, I am having a little problem with the math. What’s the story with the other 30 first- in-class drugs? So I googled the original article (results=1) and found out that of the 75 first-in-class drugs, 50 were small molecules and 25 were biologics. Interesting, and the math works. The article compares efficiency with effectiveness of the two approaches and provides some great insights into the high attrition rates and low productivity in pharmaceutical research.
However, I am still intrigued about the evolution of the phenotype. I googled flow cytometry phenotype. (Results = 3,790,000) Flow cytometry excels at phenotyping. In my personal experience in the lab, we routinely performed phenotypic analysis of peripheral blood using monoclonal antibodies to identify subpopulations of cells in patient samples to identify or quantify disease states back in the 80’s. So why has it taken so long for flow cytometry to move from phenotyping to phenotypic screening in drug discovery? I googled flow cytometry phenotypic screening (Results = 270,000), then googled phenotypic screening (Results= 2,770.000)
Now, I had to know more and found a review article that provides some great insights. In Multiparameter phenotypic profiling: using cellular effects to characterize small molecule compounds, published in 2009, the authors state that…”cell based screens tend to have higher hit rates than biochemical screens, presumably reflecting the existence of many potential targets” and propose multiparameter phenotypic profiling as a promising solution for evaluating cellular efficacy and to pinpoint side effects earlier in the drug discovery pipeline. The authors compare several profiling technologies and listed the advantages and disadvantages of each. For flow cytometry, the advantages were: 1) Allows single cell response to be measured, 2) High signal to noise ratio and 3) Can be multiplexed. The disadvantages listed were: Requires many cells and low throughput. However, they go on to say, “Technology improvements-such as the continuous-flow autosampler” (referencing the paper Flow cytometry for high-throughput, high content screening by Bruce Edwards et. al. describing the HyperCyt)…has enabled high-throughput flow cytometry experiments for profiling purposes.”
I googled HyperCyt (Results = 41,900) So, after emerging from this rabbit hole, I can see why phenotypic screening by flow hasn’t been more widely used in the past, but now with tools like the HyperCyt or HTFC Screening System, phenotypic screening by flow cytometry is a valid and practical option for researchers in drug discovery and elsewhere to perform multiplexed, suspension cell screening in order to create better drugs faster to treat and cure disease. Just go with the flow.
With great pleasure I announce that Joe Zock has joined IntelliCyt as the Senior Director of Product Management. I still can’t believe it is true. I first met Joe at SBS in 2007 when Terry Dunlay, our CEO, and I went to the conference with our fledgling company and presented our first poster. We were wandering through the exhibit hall, greeting old friends and running into former co-workers, noting that cytometry is a field of increasing mobility. One year, a person is working in flow cytometry, a few years later, image cytometry and that the edges between technologies and employees in each field were beginning to blur. We talked to Joe for a while, and as we walked around the exhibit hall, Terry shared some stories of how he and Joe made things happen in the early days of Cellomics—defining the field of High Content Screening and developing a new market.
Over the next few years, we would run into Joe at increasing numbers of meetings as we grew the company and started attending more conferences. I learned that Joe is a passionate scientist who knew a lot and cared even more about the role and importance of cell-based screening in drug development. Every now and then, it would come up in our internal discussions how we could use a guy like Joe to help us close the gap between flow cytometry and image cytometry by moving flow into screening with suspension assays. I read an article that Joe wrote in 2009, where he said, “The search is on for better tools that can automate discovery processes and yet provide the innovation needed to explore therapies for existing and emerging diseases.” Yep, we thought Joe would fit right in at IntelliCyt.
So, after two weeks working with Joe, I am happy to say that we have a remarkably capable and knowledgeable addition to our team and I think our customers and the scientific community will benefit with the addition of such a champion of cell-based screening to the IntelliCyt team. Go Joe!
As most of you know, we here at IntelliCyt have developed the HTFC Screening System; which is a high throughput screening platform using cell based assays. What is unique about our system is that it uses flow cytometry to measure cellular fluorescence. What you might not know is that our founder, Terry Dunlay, was also the co-inventor of High Content Screening and a co-founder of Cellomics, which pioneered the field of imaging based high content screening. This gives us a unique perspective, having developed both imaging and flow cytometry based screening products.
So back to the question; imaging or flow cytometry? As you might have guessed, I see these platforms as complementary technologies. Let’s look at flow cytometry first. Very simply put, this is a very powerful technology that measures fluorescence signals associated with cells in suspension as they pass through a laser based detection system, one cell at a time. This enables a couple of important features. The first is speed; thousands of cells can be analyzed in one second. This has obvious advantages in terms of statistical robustness. What’s more, multiple colors of fluorescence are detected on each cell simultaneously. This means that it doesn’t take additional time to measure additional parameters. Another valuable feature is that flow is perfect for suspension cell analysis. Suspension cells, including primary cells of the blood and bone marrow system and cell lines representing the immune/inflammatory system have largely been underutilized in high throughput screening environments. With the introduction of high throughput flow cytometry, screening assays employing these important cells are now being incorporated into drug discovery screening campaigns.
Now let’s look at imaging. High content screening platforms utilizing fluorescence microscopy have revolutionized the drug discovery industry. HCS platforms use fluorescence microscopy to capture images of cells in microplates, and then employ sophisticated imaging algorithms to interpret the fluorescence signals present in each cell. Analyzing cells by imaging in microplates thus enables the combination of throughput with spatial measurements. Imaging, then, is great for testing the effects of compounds (or other treatments) on readouts like subcellular translocation of molecules, cell morphology and even cell motility. Because the majority of these platforms work best with immobilized cells, they are perfectly suited for analyzing adherent cells.
This is an exciting time in cell based screening. Each year new platforms are becoming available for high throughput analysis of cells, which gives scientists powerful tools for drug discovery and systems biology. Now scientists have a choice in selecting the most appropriate cells –whether suspension or adherent- in which to perform small molecule or biologics screens, or when performing high throughput systems biology experiments.
Congratulations to Peter Lopez, president of MetroFlow, and all of the organizers who created a really strong educational program about current and potential future uses for flow cytometry in the area of drug development. The Fall MetroFlow meeting was hosted by Regeneron Pharmaceuticals, located in Tarrytown, NY.
Virginia Litwin, Ph.D., from Covance moderated the program and began with an Introduction to the Role of Flow Cytometry in Drug Development. Virginia also gave a touching tribute to Phil Marder, who passed away earlier this year and will be missed by all of us who knew him. Phil and Virginia co-edited a book, Flow Cytometry in Drug Discovery and Development that was just published this month. Some of the contributors to the book presented at this meeting, providing additional insight into Biomarkers, Small Molecule Discovery and Oncology Drug Discovery.
The talks were organized to provide some insight into how flow cytometry is being used in the drug discovery process. Dimitris Skokos, Ph.D., from Regeneron, spoke on New Target Discovery, describing how combining multicolor flow cytometry with high-throughput genomic engineering has allowed them to characterize the full hematopoietic cell profile of genetically engineered mice lacking Delta-like ligand 4.
Addressing Target Validation, Carmen Raventos-Suarez, PhD., from Bristol Myers-Squibb, presented data which clearly illustrated how, using multicolor flow cytometry, “cells within discrete populations can be identified before and after treatment using cell cycle profiles and molecular markers, thus providing evidence that treatment is within a mechanism and modulating the signaling cascade.” And, on a personal note, I was happy to hear her report that she has found the HyperCyt to be a valuable tool in expediting their research.
Speaking on Small Molecule Discovery, Larry Sklar, Ph.D., from the University of New Mexico Center for Molecular Discovery, presented a number of examples of screening assays using the HyperCyt flow cytometry platform in his talk, “High-throughput Flow Cytometry, Small Molecule Discovery and the NIH Roadmap Molecular Libraries Initiative.” Larry also presented on grant opportunities that are available to researchers who want to collaborate on projects.
Pre-clinical Toxicology was discussed by Dave McFarland, from iCyte in: “The Search for a Biomarker of Drug-induced Vascular Injury.” He also spoke about flow cytometric analysis of blood and how it holds the potential to provide such a biomarker.
There were too many outstanding presentations at the Fall Meeting to discuss in this post, but they will all be available soon for you to view on the MetroFlow meeting archive page.
Efforts are currently underway by many groups in the pharmaceutical industry to move from historical observation-based toxicity approaches to more predictive, mechanism-based approaches (e.g. EPA’s ToxCast™, the US National Toxicology Program, and National Center for Toxicogenomics).
In order to accomplish this, models and tools are required that allow for characterization of critical toxicity pathways such as common signaling pathways (e.g. p53, Nrf2), apoptotic pathways, etc. Much is already known about these common toxicity pathways and can be exploited through the use of commercial kits and published assays. However, in order to build a predictive model, more than just endpoints and assays are required—context needs to be given.
Context can relate to the organ that is being modeled (e.g. liver, kidneys), the whole organism (e.g. human, mouse), or even the compound that is being assessed (e.g. metabolism, interactions). The definition, and ultimately the models and tools, need to be defined up front in order to make successful predictions. Inherently, the difficulty is in both choosing the right systems and also in putting together the information in a way that is meaningful to the original question.
Given that there are many fairly straight-forward fluorescent based assays (e.g. Live/Dead®, Mito-Tracker®) which can be applied to the problem of toxicity testing, there is a lack of integration between these endpoints in assembling a clear picture of cell function and biochemistry.
What is sorely needed is a baseline using relevant cell types that helps to establish the “typical” function of a cell system under a defined set of conditions. Key to establishing this suite of baseline assays is the use of pathway-specific positive controls. Once this is established, perturbations can be made and compared against this baseline.
While many kit assays are available and fairly easy, they are typically used for quick checks of relevant endpoints, or to look at acute effects similar to a maximum tolerated dose (MTD). Many times these single readouts are then used to predict cytotoxic behavior. For example, there is ample evidence that cells that are undergoing necrosis will upregulate many other death pathways and processes, including enzymes such as caspases and release activators like cytochrome C.
With a single-point averaged readout, how is it possible to assemble the correct picture of toxicity? Bringing together multiple endpoints along key pathways is the next frontier as it relates to cell-based assay work.
Flow cytometry has long been used as a powerful cell analysis tool to examine complex populations such as immune cell distributions in the blood stream. With the ability to perform screens using flow cytometry, another interesting possibility exists in the application of this technique for drug screening. Specifically, given a single well in a multi-well plate, we can probe what the cell population distribution is in that well based on the responses of those cells.
Consider the treatment of a single cell line, such as Jurkat cells, with a chemical substance known to be toxic. Even at no concentrations of the chemical there will be cells that are in their death spirals among many healthy and viable cells. In other words, no cell population is completely homogenous. Increasing the incubation time as well as dose will cause more of these cells to enter these death spirals.
However, in every case where some cells exist, there are subpopulations of cells that are completely viable and intact. Any quadrant analysis will easily highlight this distribution. In fact, in this toxicity example, these cells may not have any upregulated stress pathways.
What then is it about these cells that allow them to be non-responders? Measurements on standard plate readers do not allow for such a detailed understanding of cell population –in short, plate readers provide a total overall or summed response of a population.
Since each cell within a population may respond slightly differently to its environment, an understanding of these differences is crucial to building a new molecular approach. A technique like flow cytometry, with single cell resolution, can provide a better mechanistic understanding of how a compound or chemical affects individual cells.
For too long the industry has focused on isolated biochemical endpoints, ignoring the fact that a so-called cell line is not homogeneous and contains information that may be pertinent to the development of more effective, and safer, drugs.
Let’s take advantage of the fact that we can use whole cell models and powerful techniques like flow cytometry to probe these questions. The promise of analyzing interconnected pathways may be right around the corner.