Flow cytometry (FCM)

Flow cytometry offers a rapid, objective, and quantitative method for analysis and purification of cells in suspension. The fundamental concept of flow cytometry is simple. Cells or other particles interact with a light beam as they pass by single file in a liquid stream. Interaction with the light is generally measured as light scatter and fluorescence according to staining of the cells. If a fluorochrome is specifically and stoichiometrically bound to a cellular component, the fluorescence intensity will ideally represent the amount of that particular component.

Multiparameter flow cytometry allows one to estimate, with high accuracy, relative quantities of a variety of cell constituents simultaneously. When the measurements are recorded in a list mode, it is possible to attribute each of the several measured features to a particular cell and thus to obtain correlated measurements of these features on a cell by cell basis. Cellular heterogeneity can thus be estimated and subpopulations with distinct characteristics can be discriminated. Thus, multiparameter flow cytometry offers improved opportunities to describe the complex relationships between cell activation, proliferation, differentiation, maturation and decomposition within heterogeneous cell populations as e.g., blood and bone marrow where different differentiation lineages are mixed together, or tumors where malignant cells may be discriminated by clonal characteristics.

Flow sorting (fluorescence activated cell sorting, FACS)

In a flow sorter, individual cells can be physically separated from larger populations, based on a composite of parameters. Any set of criteria derived from the flow cytometric analysis can be used to activate the sorting decision for the single cell. No other method than flow sorting separates living cells by their quantitative expression of molecules or their combination of predefined properties. On this basis, flow sorting can be used as part of an analytical strategy. In order to increase the efficiency for preparative purposes, flow sorting (serial events) may be combined with bulk methods (parallel events), e.g. applied subsequent to immunomagnetic separation.

Image cytometry

In analysis of tissues, quantitative image cytometry based on scanning of stationary specimens is an important counterpart to flow cytometry. And in sorting of cells from tissues, laser microdissection is an important counterpart to flow sorting.



Flow cytometric analysis of cell proliferation and aneuploidy


Cell kinetics

Cell proliferation is defined as the increase in cell number resulting from completion of the cell cycle, as contrasted to cell growth which is the increase in cell mass. The difference between the rates of cell production by mitotic division and cell loss by cell death makes up the net proliferation rate of a particular cell population. In the tissues of an organism this balance may be influenced by cell migration. Most populations of cells consist of a mixture of three different subpopulations: continuously cycling cells, non-cycling (quiescent, G0) cells, and terminally differentiated cells. The cell cycle is traditionally considered to be composed of 4 phases: the gap before DNA replication (G1), the DNA synthetic phase (S), the gap after DNA replication (G2), and the mitotic phase (M), which culminates in cell division. There are important regulatory checkpoints at the G1S and G2M transitions.

Traditional cell kinetic analysis was based on countings of tritiated thymidine labelled S phase cells and mitotic figures. At now a variety of techniques are available for fluorescent labeling not only of DNA synthesizing cells and mitotic cells, but also of apoptotic cells. This makes it possible by flow cytometry to estimate mitotic and apoptotic rates, cell cycle phase durations, potential doubling time, growth fraction, cell loss factors etc. Markers of phenotypic subpopulations may be applied in combination with cell cycle markers to make cell kinetic analysis of particular subpopulations possible.


DNA aneuploidy

Flow cytometric analysis of nuclear DNA content reveals the distribution of cells into G0/1, S, and G2/M phase fractions. The S phase fraction (SPF) is often taken as a measure of proliferative activity. Be aware that from a single univariate DNA measurement you cannot derive kinetic information. The snapshot of the DNA distribution does not tell you whether those cells having an S phase DNA content were cycling or noncycling or apoptotic (they might even include emerging aneuploid subclones). A time series of DNA measurements from the same population may disclose a perturbation of the cell cycle.

Complete stoichiometry between fluorescence and DNA content is difficult to obtain. Methods for cell preparation are supposed to counteract any staining deviations associated with the membrane permeability of the dye, chromatin structure dependency , adverse RNA staining, AT/CG selectivity, etc. The staining method of Vindeløv et al., based on propidium iodide (PI) staining of unfixed nuclear suspensions after detergent/trypsin/RNase treatment and including the measurement of internal DNA reference standards, provides optimal opportunities for high resolution DNA aneuploidy detection.

Adjustment of the flow cytometer for maximum precision is necessary (preferably CV’s of 1-2 %). The method for deconvolution of the DNA histogram into ploidy subpopulations and their cell cycle phase fractions is critical. Assumption of a uniform S phase distribution may be meaningful in detection of aneuploid subclones in heteroploid tumor biopsies, whereas a high degree polynomium may be required for fitting a perturbed S phase distribution of a drug treated cell line. The DNA index (DI) is defined as the G1 cell DNA content of the subclone relative to the G1 cell DNA content of normal, diploid cells. Accurate DI determination for an aneuploid subclone depends on the use of internal DNA reference standards, e.g. similarly stained chicken and trout erythrocytes, for control of the measurement precision and stability as well as the DNA histogram offset and linearity. For detection and quantification of DNA aneuploid subclones in solid tumors, multiple biopsies may be required because of the tissue heterogeneity.


DNA synthesis

Cycling S phase cells may be detected based on their ability to incorporate halogenated deoxyuridines, e.g. bromodeoxyuridine (BrdU), in vivo or vitro. Using various schemes for pulse-chase, continuous or double labelling essential cell kinetic parameters such as S influx, S duration and potential doubling time can be assessed. There are two approaches for flow cytometric analysis of BrdU incorporation: 1) BrdU alters the stainability of some DNA fluorochromes; it quenches AT-DNA staining (Hoechst 33342, acridine orange) and enhances CG-DNA staining (mithramycin, 7-AAD, TO-PRO-3). 2) Immunocytochemical staining with anti-BrdUrd antibody, after partial denaturation of DNA to single-stranded state using treatment with HCl, HCl/pepsin, or DNase-1. Denaturation with HCl or HCl/pepsin is suitable for combined analysis of DNA and BrdU. Denaturation with DNase-1 is applied for simultaneous analysis of additional antigens, e.g. cell surface markers or intracellular markers such as cytokines.



Mitotic cells can be detected by 1) altered stainability with some DNA dyes (acridine orange after acid treatment, mithramycin/propidium iodide after formaldehyde) or 2) immunochemically with antibody against phosphorylated histone H3 (H3-P). The mitotic rate (cell population birth rate) can be estimated by arresting the mitosis with spindle poisons (stathmokinesis).

Tracking of cells in subsequent generations according to the dilution of a label by cell division is a different approach. Assuming that the label does not interfere with specific cell functions, cells are diffusely labeled with a fluorescent lipophilic membrane anchoring reagent (PKH) or carboxyfluorescein diacetate succinimidyl ester (CFSE).

The growth fraction of a cell population can be assessed immunochemically by discriminating cycling versus non-cycling cells according to the expression of proliferation associated antigens such as PCNA and Ki-67. Alternatively, the growth fraction can be assessed by the accumulation of BrdU labeled cells during continuous BrdU exposure. Cycling and non-cycling cells may also differ by RNA content (pyronin Y, acridine orange metachromasia).



Cells with damaged membranes are selectively stained with impermeant DNA dyes such as propidium iodide, 7-AAD and SYTOX, whereas permeant DNA dyes such as Hoechst 33342, DRAQ5 and SYTO also stain the intact, live cells.

Apoptotic cells may be discriminated by several markers, representing an entire sequence of degradation processes. These markers are associated to mitochondrial membrane functions (JC-1, DiOC2(3), bcl/bax), signal transduction pathways (caspases, FLICA), cell membrane disorder (annexin-5), DNA strand breaks (TUNEL), and DNA fragmentation (sub-G1 peak). The apoptotic rate can be assessed on basis of apoptotic arrest (stathmoapoptosis) with caspase inhibitors (FLICA).


Multivariate analysis

For quantitative staining of intracellular antigens such as the proliferation associated antigens PCNA, Ki-67, H3-P and cyclins it is necessary to permeabilize the cells so that antibodies and dyes can reach the target molecules in the cell interior. At the same time the antigens must be preserved in their natural conformation and leakage should be prevented. In general, cells are fixed with formaldehyde and permeabilized with detergent, or simply fixed with cold methanol, before staining of intracellular antigens.

In a multivariate analysis several parameters can be correlated. Phenotypic markers for cell lineage, differentiation, activation, functions, viability, etc. can be correlated to DNA for aneuploidy and cell cycle distribution and to BrdU for cell cycle progression. With the instrumental development into flow cytometers with a large number of fluorescence detectors and several excitation wavelengths the options are rapidly increasing. Anyway, for each added parameter we meet increasingly technical difficulties in ensuring the specificity of staining, accessibility of the molecular targets, adequate compensation for spectral overlap, etc.



o   Shapiro HM: Practical Flow Cytometry. 4th ed. Wiley-Liss, New York, 2003.

o   Haugland RP: Handbook of Fluorescent Probes and Research Products, 9th ed., Molecular Probes, 2002, Chapters 8.1 & 15.1-5 (free access at www.probes.com).

o   Ormerod M. (ed): Flow cytometry - A Practical Approach. 3rd ed. Oxford University Press, Oxford, UK, 2000.

o   Darzynkiewicz Z, Crissman HA & Robinson JP (eds): Cytometry. 3rd ed. Part A-B. Methods Cell Biol 63-64, 2001.

o   Darzynkiewicz Z, Robinson JP & Crissman HA (eds): Flow Cytometry. 2nd ed. Part A-B. Methods Cell Biol 41-42, 1994.

o   Robinson JP et al (eds): Current Protocols in Cytometry, Chapter 7 (www.currentprotocols.com).

o   Gray JW & Darzynkiewicz Z (eds): Techniques in Cell Cycle Analysis. Human Press, Clifton, NJ, 1987.

o   Vindeløv LL & Christensen IJ: A review of techniques and results obtained in one laboratory by an integrated system of methods designed for routine flow cytometric DNA analysis. Cytometry 11: 753-770, 1990.

o   Ottesen GL et al: DNA ploidy analysis in breast cancer. Comparison of unfixed and fixed tissue analyzed by image and flow cytometry. Anal Quant Cytol Histol 19: 413-422, 1997.

o   Flyger H et al: DNA ploidy in colorectal cancer, heterogeneity within and between tumors and relation to survival. Cytometry 38: 293-300, 1999.


Revised, 25 Oct 2004 /JKL