Send to printer »

Feature Articles : Nov 15, 2009 (Vol. 29, No. 20)

Flow Cytometry's Expanding Niche

Evolving Tool Has Applications in Medical Science, Drug Design, Food Microbiology, and Population Biology
  • Richard A. Stein, M.D., Ph.D.

Cell population heterogeneity consistently emerges as an interesting and biologically relevant feature of biological systems. Nevertheless, many experimental tools are informative only about entire populations. While this level of inquiry is fundamental to our understanding of many phenomena governing the living world, only a limited number of approaches are available to dissect the differential behavior of selected individual cells within a group.

Among the techniques able to characterize cells from heterogeneous populations based on discrete properties, flow cytometry has experienced a significant expansion in recent years. At CHI’s flow cytometry meeting held recently in Boston, Peter O. Krutzik, Ph.D., senior research scientist in the department of microbiology and immunology at the Baxter laboratory of genetic pharmacology, Stanford University, talked about a new tool that he and collaborators recently developed.

“The phospho-specific flow cytometry platform enables us to analyze not only the surface markers of different cells but, at the same time, the phosphorylation level of intracellular proteins,” said Dr. Krutzik. “We can now perform complex biochemistry at the single-cell level in primary cell populations.”

Intracellular signaling pathways are perturbed in specific ways during disease states such as malignant tumors, viral infections, or autoimmune conditions, and surveying phosphorylation changes at the single-cell level by flow cytometry provides diagnostic, therapeutic, and prognostic benefits. “Not only can we stratify patients but, for the same patient, or for a particular tumor sample, we can identify heterogeneous responses that might point to different progenitor cells leading to that tumor,” he explained.

In addition, this approach could help characterize different signaling responses  such as the ones correlating to more aggressive behaviors. “This also enables us to identify patients who could have a subset of cells that is more resistant to classical treatment and who might immediately require more robust or second-stage therapy.”

During high-throughput flow cytometry experiments, the time needed for data acquisition and the high cost of antibodies represent two of the major limiting factors.  

One way to address these limitations is by combining several samples into one tube, a technique known as multiplexing. Dr. Krutzik and collaborators recently developed a multiplexing technique, known as fluorescent cell barcoding, in which each individual sample is treated with different concentrations and combinations of fluorescent dyes that covalently attach to cellular proteins and establish specific fluorescence signatures, conferring thus an additional dimension.

For example, 64–216 different signatures can be generated with only three fluorophores. When used for a 96-well plate, fluorescent cell barcoding decreased antibody costs 100-fold and reduced the acquisition time to 5–15 minutes, while providing excellent resolution.

“The cytometer is no longer a bottleneck,” emphasized Dr. Krutzik. “Fluorescent cell barcoding enables us to perform larger screens that we typically would not have done before. In addition, the data is quantitatively more robust because we can include both positive and negative controls in the same tube with the drug-treated samples.”

At the same meeting, Jonathan R. Fromm, M.D., Ph.D., assistant professor of laboratory medicine and associate director of the hematopathology laboratory at the University of Washington, talked about recent advances in using flow cytometry to detect classical Hodgkin lymphoma.  Traditionally, the diagnosis of this condition has relied on the use of immunocytochemistry, a technique in which paraffin-embedded tissue sections are stained and examined microscopically for the presence of specific cellular proteins.

A few years ago, Dr. Fromm and collaborators developed the first flow cytometry-based approach and were able to reliably identify Hodgkin and Reed-Sternberg cells, the anatomopathological hallmark of this hematopoietic malignancy. Guided by previous observations, which reported that Hodgkin and Reed-Sternberg cells appear surrounded by T cells in paraffin sections, the authors hypothesized that this phenomenon, known as “rosetting”, could explain the difficulties accompanying their flow cytometric detection.

When mixtures of Hodgkin lymphoma cell lines and T lymphocytes were examined by flow cytometry, the investigators noticed that, although the cells looked like the original cell lines, they were also expressing T cell-specific markers such as CD3 and CD45, suggesting that they were binding to surrounding T cells.

Adhesion molecules were another antigen class observed on the cell surface, and by preincubating the cells with monoclonal antibodies directed against them, the authors were able to disrupt these interactions. When used on clinical samples, this approach allowed the identification of malignant cells with a sensitivity of 89% and a specificity of 100%, Dr. Fromm said. “In no case did we see a Hodgkin lymphoma cell population that was not Hodgkin lymphoma, suggesting that this could potentially be a clinically useful assay.” 

More recently, Dr. Fromm and collaborators developed a single-tube, nine-color flow cytometry assay. “We looked at 420 tissues that were independently evaluated by morphology and examined the data without knowing what the morphology shows.” The assay sensitivity and specificity were similar to the ones reported in the previous study, and in October 2007 the authors started routinely using this approach as a diagnostic tool for classical Hodgkin lymphoma in the clinical laboratory.

Challenges

For years, flow cytometry has represented a valuable instrument to examine single cells, but the analysis of large numbers of samples is still technically and computationally challenging. Bruce S. Edwards, Ph.D., research professor of pathology and director of the shared flow cytometry resource, together with Larry A. Sklar, Ph.D., Regents professor of pathology and director of the center for molecular discovery at the University of New Mexico, and other collaborators, recently developed HyperCyt®, a high-throughput flow cytometry platform that creates temporal gaps in the collection process by introducing air bubbles between distinct samples, before delivering them to the flow cytometer.

Instead of collecting and saving each data point individually during the analysis, HyperCyt continuously collects data for the whole plate and stores the information as a single file. This approach eliminates delays that are otherwise caused when separately collecting data for each sample of large datasets. 

“This system provides a new way to analyze the data. The technology that we developed decreased the time requirement about 20 fold, and we can now analyze 40 samples in a minute,” Dr. Edwards said. “We can routinely sample an entire 384-well plate without stopping and only save the data at the end.”

By using this approach, Dr. Edwards and collaborators have developed a high-throughput phenotypic assay to screen for small molecules that induce intracellular granularity, which is one of the known markers of cellular apoptosis, and from over 24,000 compounds that were screened they identified 95 molecules of interest.

“The biggest challenge will be to develop software that is able to handle the huge volume of data generated during the screening process,” predicted Dr. Edwards. “In some experiments, each well of a 384-well plate performs up to seven assays, and each of those sets of data can tell us about 10,000–50,000 cells, which represents a lot of data to analyze and archive, and this information overload is the real challenge of the future.”

The University of New Mexico flow-cytometry screening program was recently selected into the Molecular Libraries Probe Production Centers Network, a nationwide consortium of nine small molecule screening centers that proposes to use small chemical compounds to understand different cellular processes and identify new therapeutic targets.

“Our groups and some others have been proposing to use much more sophisticated classification tools to separate signals in a more sophisticated way,” said J. Paul Robinson, Ph.D., professor of immunopharmacology and biomedical engineering and director of the Purdue University cytometry laboratories.

At the Boston meeting, Dr. Robinson emphasized the need to develop more advanced hardware and software applications to facilitate high-throughput assays  such as the ones needed for drug screening and functional studies, which often require an additional level of sophistication. “The analytical tools that we have are not adequate for the needs. I think this is a fundamental area that needs change, and I think it will be changed.”

One of the recent advances has been the development of multispectral flow cytometry, which provides several advantages over existing approaches. “Another thing that I think will happen over the next few years is an approach toward multispectral analysis,” Dr. Robinson noted. “I don’t think multispectral instruments will replace current systems, I think they represent a different tool to supplement existing technologies. By using hyperspectral fingerprinting, we will be able to identify certain aspects of cells that are more difficult to extract.”

An important application of flow cytometry is to monitor CD4 lymphocytes, the primary targets of HIV, during the course of the infection. CD4 levels are informative about the stage of the infection in seropositive individuals, help monitor the response to antiretroviral therapy, and recently were also shown to predict the risk to develop malignant tumors during the course of the infection.

While the level of CD4 lymphocytes needs to be regularly monitored during therapy, ideally once every three to four  months, in many locations worldwide this is difficult or impossible to accomplish due to high costs. It has been estimated that only approximately 2–3% of the approximately 35 million HIV-positive individuals from developing countries receive adequate treatment.

The Cytometry for Life Program, founded in 2006 by Dr. Robinson and his colleague Gary Durack, intends to construct robust, user-friendly instruments that would reduce costs and make individual CD4 tests affordable in underserved regions worldwide.

To bring attention to the importance of this issue, Dr. Robinson recently climbed the world’s tallest mountain. “It was easier to climb Mount Everest than to solve this problem,” he said. “We need an instrument that is easy to use and has low costs. The only solution is the ultimate solution. And the ultimate solution might well be something ridiculously simple and extremely cheap.”

Purifying Cell Populations

Flow cytometry represents an irreplaceable tool that allows researchers to identify and purify cell populations to homogeneity. This aspect is of particular interest in stem cell biology, where cell population heterogeneity poses major challenges for downstream applications, which often require pure cell populations. At the International Society for Stem Cell Research meeting held in Barcelona, Nil Emre, Ph.D., stem cell scientist at BD Biosciences, presented a recent approach that she and collaborators developed to define a cell-surface signature for neurons derived from human embryonic stem cells and subsequently separate different cell populations on this basis.

“I predict that there will be defined cell signatures for numerous cell types, and there will be methods to enable their purification, and this will help develop better and more robust in vitro assays,” said Christian Carson, Ph.D., senior author on the study. “I also think that using flow cytometry to define cell-surface signatures will provide great tools for quality control, enabling us to quantify purity of cell preparations.”

While flow cytometry allows high-quality data to be obtained in a more time-efficient and cost-effective manner as compared to other techniques, for certain applications it represents one of the few available methodologies. One example is protein glycosylation, a well-known post-translational modification during malignant development. Overexpression, loss of expression, and the incomplete synthesis of glycosylated structures have long been described in various tumors, and some of these changes represent valuable progression markers with prognostic value.

Glycosylation can be examined with two main approaches—mass spectrometry and flow cytometry. Flow cytometry can be used for routine clinical laboratory applications since it is possible to have antibodies that recognize specific glycosylation epitopes and differentiate cells on this basis. “It’s hard to beat flow cytometry for this approach if the cells are amenable to it and the antibodies are available,” emphasized Dr. Carson.

Flow cytometry has undergone exciting changes in its approximately 40 years of existence. While initially used mostly to quantitate the DNA content of malignant tumor cells, this approach is increasingly becoming an indispensable tool in medical sciences, food microbiology, ecology, drug design, and population biology and, in all likelihood, the next few years will witness many novel and more sophisticated applications.