ABOVE: Flow cytometry uses lasers to measure cell characteristics. © ISTOCK.COM, Meletios Verras

In the mid-twentieth century, Louis Kamentsky, an engineer at Columbia University at the time, searched for a convenient approach for differentiating cancerous and normal cells. He modified a cell counting device that arranged samples into a single-file line by mounting an oscilloscope to measure their absorption and scattering of light as the cells passed through a flow tube.1-4 

Around the same time, Mack Fulwyler, an engineer working at Los Alamos National Laboratory,  needed to separate particles, so he drew on existing techniques to create droplets to separate cells from a flow stream based upon charge that correlated to their volume.5,6 These approaches laid the foundation for flow cytometry, which is now a staple in biological research. 

“All of the methodology that existed before flow cytometry suddenly could be applied to the single cell,” said Thomas Jovin, a biophysicist at the Max Plank Institute who developed advancements to the instruments in the 1970s as flow cytometry emerged as a major player in the research space.

Flow Cytometry: Cell Analysis Speeds Up

Flow cytometry entered biomedical research in immunology and cancer labs out of initial interests in separating and counting cells in a mixed population, but groups also developed instruments purely to characterize cells.7,8 “The flow cytometer and the flow sorter are not separate instruments,” explained Jovin. “The flow sorter requires that it be a flow cytometer at the same time because you have to make the same measurement. It’s just that you’re using it to process the cell after it has gone through the detection system.” Today, instruments that both analyze and sort cells are referred to as flow sorters and those that do not are called flow analyzers.

Initially, flow measurements were based on fluorescent light emitted from dyes that researchers used to identify genetic material, but soon after, scientist also determined the cell’s size based on its light-scatter patterns.9, 10 These first instruments used lamps as their light source, but this soon changed. “The lasers came along very quickly,” Jovin said. “They were important because you could focus a laser down to microns, whereas you can’t do that with a large optical source like a lamp.” 

You can measure essentially anything in, on, or produced by a cell at a high rate of speed in a heterogeneous solution at a rapid rate.

 —Jonni Moore, University of Pennsylvania

Soon, researchers added more lasers to their instruments to expand the colors they could detect and developed methods to analyze and sort cells labeled with two fluorescent molecules.11,12 With the help of dichroic mirrors and bandpass filters that reflect and isolate, respectively, specific wavelengths of light to dedicated detectors, scientists could funnel the signal from multiple parameters to specific detectors to study more features of their samples.13 

As the parameters that flow systems used expanded, data poured out of labs globally. “You have a lot of signals that have been processed in real time, and you have to make decisions, in the case of the sorter, in real time, because otherwise your cells won’t be there anymore,” Jovin said. “The only way to do that was by computation.” Jovin and his team developed a computer-controlled flow cytometry instrument that facilitated the data analysis process.14 

When Flow Cytometry Outshined Microscopy 

With the ability to rapidly assay and separate cells of interest from a mixed population based on multiple parameters, flow cytometry rivaled its predecessor, microscopy, in the study of cells. Jonni Moore, an immunologist and the director of the shared resource laboratory at the University of Pennsylvania, recalled using a flow cytometer for the first time after only having used a fluorescent microscope during graduate school. “I thought I had died and gone to heaven,” she said. According to Moore, classifying T lymphocytes on the microscope took several hours longer than the seconds it took her to analyze thousands of cells by flow cytometry. “It really allowed me to ask a lot more questions in my research,” Moore said. 

While some research focused on the ability to analyze cell properties with flow systems, many groups used flow cytometry for its sorting capacity.15 However, as scientists developed new dyes, they could use flow cytometry to analyze more cellular parameters, such as mitochondrial activity and the quantity of particular receptors on cells.16-18 

Flow cytometry analysis expanded into the clinical setting by helping streamline the quantification of CD4+ T cells during the human immunodeficiency virus (HIV) epidemic. Compared to microscopy, flow cytometry analysis was faster and more reliable.19, 20 “Over the next 30 plus years, analytical cytometry exploded as we realized that we could measure virtually anything in, on, or produced by a cell, in multiple populations at the same time,” Moore said. 

Today, researchers still use flow cytometry to analyze a population of cells based on the presence of surface markers tagged with a fluorescent antibody or other probe. However, these analyzers can also use dyes and other techniques to investigate cellular functions, such as metabolism and protein secretion.21, 22 Researchers can assess cell proliferation and death with flow cytometry by measuring the dilution of dye or uptake of it.23, 24 While various individual methods exist that can measure the amount of protein or other mediators produced by cells or their activity, they require researchers to do them separately. “The technology of flow cytometry, as it exists today, allows you to do all of that together,” Moore said. 

However, despite measuring an entire population of cells, flow cytometry is a single-cell technique. “Because you’ve dissociated tissues and you’ve put these objects into kind of single file, you’ve lost where they’re seated next to one another,” explained Lisa Nichols, the director of the flow cytometry facility at Stanford University. That level of spatial information requires microscopy. Nonetheless, flow cytometry produces high dimensional information on individual cells, and in contrast to other single cell techniques, does so more quickly on larger populations. “Flow cytometry can actually go through and get you the results from millions of cells in a matter of minutes,” Nichols said. 

Flow Cytometry: Scattering Light to Measure Cells

A high-throughput, single-cell method enables researchers to assess several cell parameters simultaneously with the help of lasers.

          Infographic showing how flow cytometry enables researchers to assess several cell parameters simultaneously at a single-cell level with the help of lasers.

(1) Sample Uptake

Scientists prepare samples as single cell suspensions and labels components of interest with fluorescent antibodies or other probes. The cytometer uses pumps to draw the sample through tubing to analyze it. 

(2) Cellular Alignment

Using hydrodynamic focusing the instrument injects the sample into a fast-moving stream of fluid that funnels the sample single file through a narrow channel. 

(3) Laser Interrogation

The channel leads to a point where the individual cells intersect with one or more lasers. The measured sample is deposited into a waste receptable after it passes this point. 

(4) Light Scatter and Detection

As a cell begins to cross the laser beam, it scatters light. Light that mostly crosses the cell is detected as forward scatter and measures the cell’s size. Light that encounters obstacles in the cell changes direction and is detected by a side scatter detector, indicating the granularity of the cell. If the lasers excite fluorescent molecules in the cell, the emitted light is channeled through dichroic mirrors and bandpass filters to isolate specific wavelengths that meet detectors specific for those wavelengths. 

© Ashleigh Campsall
See full infographic: WEB | PDF

Spectral Flow Cytometry and Imaging Flow Cytometry

Fluorescent probes have come a long way since the 1960s. Researchers have added lasers and probes that recognize the violet and infrared range, as well as expanded probes into quantum dots, or inorganic nanocrystals.25-27 These additions greatly expanded the available colors for researchers to use, but introduced new challenges, as more color parameters increased the likelihood of overlapping spectra from these probes. “As those overlaps increase, your ability to resolve very dim signals is compromised,” said Nichols.

In traditional cytometers, to minimize overlapping signals from multiple fluorescent probes, the instrument doesn’t use all of the light energy that a molecule emits. “We take that whole spectrum, and we take a slice of it. And we measure that slice,” said Timothy Bushnell, the flow cytometry core director at the University of Rochester. Mirrors and bandpass filters only permit a certain range of wavelengths to reach their detectors, which usually correspond to the peak emission spectra of commonly used probes.

While this method simplifies the problem of overlapping spectra in multiparameter experiments, it eliminates potentially valuable information. This prompted the development of spectral analyzers, which capture a fluorescent molecule’s full emission spectrum.28, 29 “We now get the whole picture of what that spectrum looks like,” Bushnell said. 

Using single-labeled and unlabeled controls, the instrument accesses the entire spectrum of these samples to calculate the distinct emission spectra of each color from the mixed readout. The introduction of spectral flow cytometry enabled researchers to conduct multidimensional analyses. “It lets you have more flexibility in what fluorochromes you use because you’re not confined to this one detector, one fluorochrome phenomenon,” Bushnell said. These advancements come in tandem with improved detector technology, such as swapping out current photomultiplier tubes for silica-based models that pick up longer wavelengths better.30 

While flow cytometry enables a high dimensional analysis of individual cells within a population, researchers cannot see where their target of interest is within or on the cell. “Our resolution is basically a dot on a plot,” Bushnell said. This type of resolution traditionally had to be done with microscopy, but at the expense of time and quantity of cells analyzed. The introduction of imaging cytometry is changing that.31 

Image flow cytometers capture an image of a cell as it flows through transit. “We could combine the power of knowing where something is, so seeing where it is in the cell, with the statistics that flow can give you,” Bushnell said.

Anything you can actually make into a particulate solution and put a fluorescent tag on, you can now measure.

 —Lisa Nichols, Stanford University

“You are limited by the fact that it is flow, so these things are moving,” Nichols said. “You’re never going to get the resolution you’re going to get with a microscope where it’s sitting still.” Although not in the resolution possible with microscopy, the photographs provide additional information about where signal originates from within and on a sample.

Additionally, having been available for flow cytometry analyzers for more than a decade, this imaging capacity is becoming available for flow cytometry sorters.32 One setback in this application is the ability to take an image rapidly and interpret that image to make a decision for a falling sample’s fate. “Things are moving so fast, you need to do one of two things,” Nichols said. “You either have to have a whole bunch of predetermined features that you’re looking for that can be matched to each individual cell, or you have to have AI and computing technologies.” 

Expanding the Capabilities of Flow Cytometry

Not only will the rapid computing power of machine learning be necessary for quick sorting decisions, but as flow cytometry becomes increasingly multiparametric, researchers forgo the traditional bivariate plots for computational analyses already used in single-cell sequencing analyses.33-35 “When you look at dot plots, two by twos, you only ever see the elephant foot. You can never see the whole elephant by doing that,” said Moore. This opens the opportunity to explore and interpret data in completely new ways, possibly by introducing previously overlooked findings in datasets. 

Beyond crunching the numbers in individual experiments, machine learning may offer the ability to account for variations between experiments, or batch effects. Even more broadly, these intelligent tools may be imperative for comparing and combining analyses between different institutions, confidently enabling collaborations.36 

Flow cytometry is not restricted to cells. “Anything you can actually make into a particulate solution and put a fluorescent tag on, you can now measure,” said Nichols. With the help of microfluidic technology,  instruments analyze everything from metal nanoparticles and microplastics to exosomes.37-40 These droplets have also paved the way for studying materials typically released from cells, including antibodies and other proteins and may soon be compatible with existing flow systems.41-43 Meanwhile, specially developed cytometers with the ability to more accurately measure the small scale of microparticles advance the research potential of this field. 44, 45 All of these developments aim to push flow cytometry to its next limit.  

  1. Moldavan A. Photo-electric technique for the counting of microscopical cells. Science. 1934;80(2069):188-189
  2. Brecher G, et al. Evaluation of electronic red blood cell counter. Am J Clin Pathol. 1956;26(12):1439-1449
  3. Crosland-Taylor PJ. A device for counting small particles suspended in a fluid through a tube. Nature. 1953;171:37-38
  4. Kamentsky LA, et al. Spectrophotometer: New instrument for ultrarapid cell analysis. Science. 1965;150(3696):630-631
  5. Sweet RG. High frequency recording with electrostatically deflected ink jets. Rev Sci Instrum. 1965;36:131-136
  6. Fulwyler MJ. Electronic separation of biological cells by volume. Science. 1965;150(3698):910-911
  7. Hulett HR, et al. Cell sorting: Automated separation of mammalian cells as a function of intracellular fluorescence. Science. 1969;166(3906):747-749
  8. Holm DL, Cram LS. An improved flow microfluorometer for rapid measurement of cell fluorescence. Exp Cell Res. 1973;809(1):105-110
  9. Van Dilla MA, et al. Cell microfluorometry: A method for rapid fluorescence measurement. Science. 1969;163(3872):1213-1214
  10. Mullaney PF, et al. Cell sizing: A light scattering photometer for rapid volume determination. Rev Sci Instrum. 1969;40:1029-1032
  11. Shapiro HM, et al. Cytomat-R: A computer-controlled multiple laser source multiparameter flow cytophotometer system. J Histochem Cytochem. 1977;25(7):836-844
  12. Loken MR, et al. Two-color immunofluorescence using a fluorescence-activated cell sorter. J Histochem Cytochem. 1977;25(7):899-907
  13. Steinkamp JA, et al. Multiparameter analysis and sorting of mammalian cells. Exp Cell Res. 1974;84(1-2):15-23
  14. Arndt-Jovin DJ, Jovin TM. Computer-controlled multiparameter analysis and sorting of cells and particles. J Histochem Cytochem. 1974;22(7):622-625
  15. Cram LS, Brunsting A. Fluorescence and light-scattering measurements on hog cholera-infected PK-15 cells. Exp Cell Res. 1973;78(1):209-213
  16. Darzynkiewicz Z et al. Increased mitochondrial uptake of rhodamine 123 during lymphocyte stimulation. Proc Natl Acad Sci. 1981:78(4):2383-2387
  17. Darzynkiewicz Z et al. Interactions of rhodamine 123 with living cells studied by flow cytometry. Cancer Res. 1982;42(3):799-806
  18. Bohn, B. High-sensitivity cytofluorometric quantitation of lectin and hormone binding to surfaces of living cells. Exp Cell Res. 1976;103(1):39-46
  19. Newmann JT, et al. Lymphocyte phenotyping by fluorescence microscopy and flow cytometry: results in homosexual men and heterosexual controls. AIDS Res Hum Retroviruses. 1983;1(2):127-134
  20. Chattophadyay PK, Roederer M. Good cell, bad cell: Flow cytometry reveals T-cell subsets important in HIV disease. Cytom A. 2010;77A(7):614-622
  21. Robinson JP, et al. Measurement of intracellular fluorescence of human monocytes relative to oxidative metabolism. J Leukoc Biol. 1988;43(4):304-310
  22. Jung T, et al. Detection of intracellular cytokines by flow cytometry. J Immunol Methods. 1993;159(1-2):197-207
  23. Lyons AB, Parish CR. Determination of lymphocyte division by flow cytometry. J Immunol Methods. 1994;171(1):131-137
  24. Koopman G, et al. Annexin V for flow cytometric detection of phosphatidylserine expression on B cells undergoing apoptosis. Blood;1994(84(5):1415-1420
  25. Chattopadhyay PK, et al. Brilliant violet fluorphores: A new class of ultrabright fluorescent compounds for immunofluorescence experiments. Cytom Part A. 2012;81A(6):456-466
  26. Stewart CC, et al. Flow cytometer in the infrared: Inexpensive modifications to a commercial instrument. Cytom Part A. 2005;67A(2):104-111
  27. Chattopadhyay PK, et al. Quantum dot semiconductor nanocrystals for immunophenotyping by polychromatic flow cytometry. Nat Med. 2006; 12: 972-977
  28. Goddard G, et al. Single particle high resolution spectral analysis flow cytometry. Cytom Part A. 2006;69A(8):842-851
  29. Gauci MR, et al. Observation of single-cell fluorescence spectra in laser flow cytometry. Cytom Part A. 1996;25(4):388-393
  30. Lawrence WG, et al. Enhanced red and near infrared detection in flow cytometry using avalanche photodiodes. Cytom Part A. 2008;73A(8):767-776
  31. George TC, et al. Distinguishing modes of cell death using the ImageStream multispectral imaging flow cytometer. Cytom Part A. 2004;59A(2):237-245
  32. Schraivogel D, et al. High-speed fluorescence image-enabled cell sorting. Science. 2022;375(6578):315-320
  33. Ota S, et al. Implementing machine learning methods for imaging flow cytometry. Microsc. 2020;69(2):61-68
  34. Qiu P, et al. Extracting a cellular hierarchy from high-dimensional cytometry data with SPADE. Nat Biotechnol. 2011;29:886-891. 
  35. Becht E, et al. Dimensionality reduction for visualizing single-cell data using UMAP. Nat Biotechnol. 2019;37:38-44
  36. Hu Z, et al. MetaCyto: A tool for automated meta-analysis of mass and flow cytometry data. Cell Rep. 2018;24(5):1377-1388
  37. Mazutis L, et al. Single-cell analysis and sorting using droplet-based microfluidics. Nat Protoc. 2013;8:870-891
  38. Zucker RM, et al. Characterization, detection, and counting of metal nanoparticles using flow cytometry. Cytom Part A. 2015;89(2):169-183
  39. Li J, et al. Separation and flow cytometry analysis of microplastics and nanoplastics. Front Chem. 2023;11
  40. Liu H, et al. Analysis of extracellular vesicle DNA at the single-vesicle level by nano-flow cytometry. J Extracell Vesicles. 2022;11(4):e12206
  41. El Debs B, et al. Functional single-cell hybridoma screening using droplet-based microfluidics. Proc Natl Acad Sci. 2012;109(29):11570-11575
  42. Shahi P, et al. Abseq: Ultrahigh-throughput single cell protein profiling with droplet microfluidic barcoding. Sci Rep. 2017;7:44447
  43. Brower KK, et al. Double emulsion flow cytometry with high-throughput single droplet isolation and nucleic acid recovery. Lab Chip. 2020;20(12):2062-2074
  44. Nielsen MH, et al. A flow cytometric method for characterization of circulating cell-derived microparticles in plasma. J Extracell Vesicles. 2014;3(1):20795
  45. Woud WW, et al. An imaging flow cytometry-based methodology for the analysis of single extracellular vesicles in unprocessed human plasma. Commun Biol. 2022;5:633