By Edgar Wex
Flow cytometry is a laboratory technique that involves the processing and analysis of cells from a sample, one cell at a time. This, combined with various dyes and chemicals used for identification, can yield a lot of information about the cells; from the basics, like whether the cell is alive, and more complicated information, such as what proteins are being expressed. This makes it a very useful technique in diagnosis of disease, as it can pick up the expression of very specific molecules on defined populations of cells, which can help understand what is going wrong and see details that would never be seen otherwise.
The fundamental concept of flow cytometry involves the use of lasers to pass light through cells in a solution, and using the data generated from the scattering patterns of light to make inferences about the cell’s structure. This means cytometers can be used for immunophenotyping (finding out what types of immune cell are present), cell sorting (analysing cells and then splitting them into groups), cell cycle analysis (to find what phase the cells are in), apoptosis assays (to determine cell viability), and many more. While this is all very cool for looking at cells, there are also clinical applications for this - for example, the effects of a certain treatment on cells can be quantified.
While flow cytometry used to be found only in basic research laboratories, its use has started expanding into clinical laboratories due to developments improving the speed of the process, making it a viable option for diagnoses. In some contexts, it has replaced immunohistochemistry (IHC) as the main form of cell analysis (IHC involves staining a cell sample and looking at it under a microscope). This is because immunohistochemistry is only useful when looking at one stain at a time, as that one stain can only assess the expression of one molecule, so multiple stains and rounds of testing must be used to infer meaningful information. Flow cytometry can simply use one sample, and in seconds, produce multiple pages of graphs giving the exact, quantifiable expression of tens of different molecules and biomarkers inside the cell.
So how does it work? First of all, cells must be adequately prepared before analysis; harvesting, decanting, centrifugation, straining, and various other processes that would usually be done to look at cells under a microscope. Cells can then be stained with various chemicals and put in a suspension (liquid that dilutes the cells), which is then taken up by the cytometer for analysis. Through the precise design of the machine, the cells can then be funneled into an incredibly thin stream, where cells enter the laser area one by one. This is done through the use of sheath fluid, which flows through a nozzle and narrows the incoming cell sample.
Then, the cells pass by multiple lasers, and due to their irregular shapes, light is scattered at various angles. Electrons in different types of fluorochromes (added to the cells beforehand) are excited by the lasers, gain energy, and drop an energy level, emitting light. These light emissions from the fluorochromes can then be filtered with optical filters. A complicated set of lenses, mirrors, and filters are used to collect data that is then processed by a computer. This would take too long to explain in detail, however the basic premise is that there are multiple fluorochromes (small, fluorescent tags) added to the cell samples, which bind to specific target molecules on or in the cells. These are then targeted by specific lasers, and the emission characteristics (i.e. wavelength, strength, etc) are collected. Filters are used to ensure that only emissions from the desired fluorochrome are picked up, and not from others, so that multiple molecules can be tested for at once, and expression data from one molecule doesn’t ‘spill over’ into another's.
Forward and side scatter are collected as basic data to determine if the cell is single or joined to another and give information about its size and general internal structure. For example, given an FSC-SSC (forward scatter vs side scatter) plot, you would be able to tell that the low side scatter cell population is made of lymphocytes, while the high side scatter population is made of granulocytes (due to their different shapes and sizes).
Using computer software, FSC and SSC, as well as other dyes e.g. viability dye (this allows you to see which cells are alive or not based on how much they emit this fluorochrome), you can then isolate the population of cells you want to look at and look at their expression of various molecules you have tested for. The data will look something like this (4 of, say, 30 graphs you could get). [*graphs are linked below*]
Along the x-axis for all these graphs you have CD4, whose expression was determined by the emission levels of the A594 (alexa fluor 594) fluorochrome. While it isn’t important for analysing the results, this tells you that the fluorochrome used was a bright dye, with red emission, excitable by 594 nanometre lasers (594 nanometers being the wavelength of light). On the y-axis, you see various other biomarkers. For example, the second graph has CD8, which was tested for using the PerCP fluorochrome. What the graph tells you is that there are 3 general populations of cells; CD4+/CD8-, CD8+/CD4-, and CD4-/CD8-, or in English, cells that express CD4 protein but not CD8, CD8 but not CD4, or neither. And that then tells you that there are Th cells (CD4+), Tc cells (CD8+), and other cells, which could be determined by looking at other graphs. Populations of these cells are measured, and their percentages can be seen in the corners of the graphs. These can then be lined up against the values that would be seen in a healthy individual, and clinical inference can be made, for example, that there is an abnormally low population of Th cells indicative of a certain disease.
Looking at other graphs, you may also find cells that match the immunophenotypes of various cancers. This is why you see things like “Follicular Lymphoma cells are CD19+/CD20+/CD22+/CD79a+/HLA-DR+/CD5-/CD43-” and so if you are looking at cytometry data as a pathologist, and see cells which line up to these phenotypes, you can tell that these cells are cancer cells, and a diagnosis can be made. This is not only very precise and accurate, but it can also deal with abnormal cases where the patient has a unique or very uncommon condition - specific biomarkers can be quantified and inferences can be made about what this means about the cells and how they function.
Images:
Further images and explanations of flow cytometry:
References:
What did you learn?
How does flow cytometry work?
Flow cytometry passes a sample of cells in suspension through a nozzle and across lasers. Lasers excite electrons in fluorochromes/dyes attached to biomarkers within cells, and emit light. More emission means more of that fluorochrome, which means more of that biomarker is present. The emitted light is passed through a network of mirrors and optical filters, and then collected and analysed by computer software to produce graphs.
Why is flow cytometry useful in clinical diagnosis?
Flow cytometry is very accurate and precise, and can provide expression data for a multitude of biomarkers at once, and give quantifiable results. It is fast and automatic, except for the use of software at the end, leaving less down to the subjectivity of the pathologist.