Flow Cytometry is the measurement (meter) of characteristics of single cells (cyto) suspended in a flowing saline stream.
A basic flow cytometer consists of five main components:
- A source of light i.e. Laser;
- A flow cell;
- Optical components to focus light of different colours on to the detectors;
- Electronics to amplify and process the resulting signals; and
- A computer.
Most scientific procedures involve obtaining measurements as average values for the whole population, however, flow cytometric analysis involves measurements being made separately on each particle within the suspension in turn, and thus is a very important feature of flow cytometry.
In addition several parameters can be measured on tens of thousands of individual cells within a few minutes, and this is the power of flow cytometry. The properties measured include a particle�s relative size, granularity, and fluorescence intensity. Any suspended particle or cell from 0.2-150 micrometers in size is suitable for analysis.
There are two types of flow cytometer; the basic bench top type which only analyse cells, and more complex ones which can analyse cells and also physically sort them, called Cell Sorters.
Flow Cytometry is a very useful tool and is used worldwide. Its importance in scientific research and diagnosis is increasing. It is now used routinely in many labs to detect the presence of specific surface and intracellular markers; measure metabolic activity and DNA content, among many other uses. With cell sorting applications range from the separation of large numbers of cells for functional studies or chromosomes for preparing gene libraries to the direct cloning of single rare transfected or hybridoma cells into each well of a tissue culture plate.
Before going on to explain how a flow cytometer works, it is interesting to find out from where it originated and how it was developed. The flow cytometer evolved from the development of several fields:
- Dye chemistry;
- Electronics; and
Without advances in these fields it would not have been possible to develop a flow cytometer. Below is a brief run through history but further details can be obtained from several text books including “Practical Flow Cytometry” by Howard Shapiro.
16th Century: Leeuwenhoek built the first simple microscope.
1742: Lomonosov produced dark field illumination and performed light scatter measurements. At that time the only means of illumination was white light until,
1904: Kohler developed a microscope with an ultraviolet (UV) light source.
1934: Moldavan described a photoelectric technique for counting cells flowing through a capillary tube. This was the beginning of flow cytometry.
1938: Caspersson developed a primitive flow cytometer to measure cell properties in the UV and visible regions.
1940: Crosland and Taylor developed a blood cell counter using the sheath flow principle, light scatter and darkfield illumination.
After 1850: the textile industry required absorptive dye chemistry and so this drove the development of stains. These stains were sought to enhance the visibility of prokaryotic and eukaryotic material under the microscope.
Malachowski and Romanowski used stains for identifying parasites in blood cells and for identifying haematopoietic cells, by using acidic and basic dyes. These dyes gave rise to the Geimsa, Leishman, MacNeal and Wright stains.
1880s: Fluorescent dyes were developed. At this time Paul Ehrlich synthesized and used Fluorescein, and in order to resolve the internal structures of leukocytes, he mixed acidic and basic dyes.
1900: DNA dyes first used.
Late 1960s and 1970s: several new dyes were developed by the introduction of flow cytometry, and hence measurement of DNA content became one of the first major applications of flow cytometry.
1969: Ethidium bromide first used by Dittrich and Gohde.
1973: Crissman and Steinkamp introduced Propidium iodide.
1974: Crissman and Toby used mithramycin.
1976: Hoechst dyes introduced by Latt and Stetten.
1977: Stohr introduced DAPI.
Flow cytometry today commonly uses all these dyes.
1940: In order to detect microorganisms in tissues, Coons, Creech and Jones used antipneumococcal antibodies conjugated with anthracene.
1950: Immunophenotyping was born when Coons, Creech and Jones started using antibodies conjugated with fluorescein isothiocyanate (FITC).
So at this point for scientists instead of depending upon morphology they could identify cells by the specific proteins present on their cell membrane. This was because immunophenotyping allowed for the labelling of these specific proteins on cells, predominately leukocytes.
1975: Felix Milgrom introduced the process of producing monoclonal antibodies, therefore it became easy to manufacture monoclonal antibodies and features of cells could then start to be catalogued.
By 1945: the photomultiplier tube (PMT) had been developed to detect photons and convert them to electrical pulses. In addition, amplifiers and analogue-to-digital converters were developed.
1949: the first non-optical electronic blood cell counter was patented by Wallace Coulter.
1950s: Academic medical institutions began to realise that automated cytology could be very useful in clinical diagnosis. The revolution to automate was initiated by the ability to electronically count blood cells faster and more accurately than a haemocytometer.
1957: The first Model A Coulter Counter was developed, primarily to count erythrocytes and leukocytes from blood.
1961: Model B Coulter Counter was developed. This model also enabled the size distribution of these cells to be determined.
The disadvantage of the Coulter Counter was that it couldn’t identify what was being counted. This led to the need for an automated microscopic identification process. By this time Computers had newly been introduced. In an attempt to achieve this automated process computers along with relevant software seemed to be the way forward.
Marylou Ingram (a haematologist at the University of Rochester) and Kendall Preston (a biomedical engineer at Perkin-Elmer) together built the Cytoanalyser, a microscopic-based instrument, in order to automate the microscopic identification of leukocytes in stained smears.
At the same time another group, consisting of Mortimer Mendelssohn, Brian Mayall and Judith Prewitt, began studies to extend the image analysis of cells by using an automated instrument called CYDAC. These were the first examples of automated image cytometry.
The real beginning of modern flow cytometry started when Fulwyler (U.S. Los Alamos National Laboratories) built a cell sorter using the Coulter principle to size cells and electrostatic charging of droplets to sort them.
The development of the impulsecytophotometer (ICP) came next by Dittrich and Gohde. It became known as the “phywe”. Cells were introduced into a flowing sheath stream located under a high power microscope objective that provided the optics for scatter measurements and fluorescence detection.
By combining the measurement of volume, light scatter and fluorescence into a single instrument, Paul Mullaney (also at Los Alamos) introduced multiparameter flow cytometry.
The ability to measure side scatter was added by extensive experiments carried out by Gary Salzman.
By the mid 1970s, flow cytometers were entering the marketplace, and Leonard Herzenberg (Stanford) came up with the term, Fluorescence Activated Cell Sorter, or FACS.
Advances in Immunophenotyping
In the early 1980s, the discovery that HIV or AIDS killed CD4 T cells led to the advancement of immunophenotyping. Antibodies to the membrane proteins CD3, CD4 and CD8 expressed by two subsets of these T cells were among the first characterised monoclonal antibodies (MAb).
The first major clinical application of immunophenotyping by flow cytometry resulted as AIDS began to spread through the populations of the world. A test to monitor the progression of AIDS was developed by the measurement of CD4 T cells using MAb’s.
With this development, and the successful measurement of DNA content, it became evident that flow cytometry would be a vital automated approach to cell analysis, and hence other clinical applications began to be created. These applications included ploidy and S phase fraction measurement in solid tumours, diagnosis and follow-up of haematopoietic malignancy and paroxysmal nocturnal haemoglobinuria, and monitoring of transplant rejection and haematopoietic regeneration.
It soon became apparent that as well as measuring membrane surface markers on cells, although a slightly longer process, intracellular markers could also be measured. The ability to identify a cell population by membrane markers and at the same time determine the cell’s function had now become a powerful application of flow cytometry.
Chromosome staining and sorting for cloning of DNA to produce chromosome-specific sequences was another application of flow cytometry. This was pioneered by Larry Deaven and Scott Cram (Los Alamos) and Joe Gray and Ger van den Engh (Lawrence Livermore National Laboratory). This new application led to the construction of the first high-speed cell sorter for the purpose of sorting each human chromosome and consequently this was the first step in the Human Genome Project.
To use a flow cytometer, you need a sample of cells. It is important therefore to get some background information as to the process occurring before the sample is presented to the flow cytometer and data can be obtained from it. So this section will give you a brief background on:
Antigens and Antibodies,
Fluorochromes and Fluorescence,
Antigens and Antibodies
The cell is the basic building block of all tissues and microorganisms. Around the cell is the plasma membrane which anchors surface proteins. These surface proteins are called antigens (Ag’s) and can serve as cell identifiers by defining characteristics about a cell, such as function, lineage and developmental stage. A classification system based on assigned cluster of differentiation (CD) numbers has been adopted because there are so many of these antigens. For example, T-cells have the antigens CD3, CD4 etc. on their surface. These distinguish it from other cell types, such as B-cells which have the antigens CD19, CD20 etc. on their surface.
Antibodies (Ab’s) are proteins that bind very tightly to their targets, the antigens. Each type of antibody has a distinct antigen-binding site. Each antibody recognises its antigen with great specificity.
They are produced in vertebrates as a defence against infection. Each antibody molecule is made of two identical light chains and two identical heavy chains, so the two antigen binding sites are identical.
Antibodies are made by a class of white blood cells called B lymphocytes, or B cells. Each resting B cell carries a different membrane-bound antibody molecule on its surface that serves as a receptor for recognising a specific antigen. When antigen binds to this receptor, the B-cell is stimulated to divide and to secrete large amounts of the same antibody in a soluble form.
Monoclonal antibodies (MAb�s) are antibodies of a single specificity. They are created in the lab by mixing spleen cells from a mouse which has been immunized with the desired antigen, with tumour cells. These can be grown indefinitely. Antibodies bind to three dimensional molecular structures on antigens called epitopes and each antigen contains several hundred different epitopes. A monoclonal antibody is specific for a single epitope, while polyclonal antibodies are actually the natural pool of several hundred MAb�s produced within the animal, each one binding to its unique epitope.
MAb�s are produced by many different manufacturers and they now come conjugated to a fluorescent marker. Most commercial manufacturers use in vitro cell culture systems to produce MAb�s, hence large and, at least theoretically, unlimited quantities of MAb�s of specific characteristics can be produced.
Fluorochromes and Fluorescence
A fluorochrome is a fluorescent marker. It is a molecule that fluoresces (i.e. gives out a particular colour) when a laser light hits it. Fluorescence occurs when a molecule excited by light of one wavelength loses its energy by emitting light of a longer wavelength. The exciting and emitting light, being of different wavelengths, can be separated from one another using optical filters.
The most common fluorophores are FITC (fluorescein isothiocyanate) and PE (Phycoerythrin), but there are many others available. FITC is a small, charged molecule that can be easily conjugated to protein through an isothiocyanate group. It has an excitation maximum at 495 nm and emits green light (520 nm). It can be excited by a 488 nm argon-ion laser. R-phycoerythrin is a phycobiloprotein extracted from red algae. Its excitation maxima are 564 and 495 nm, and emission maximum is 576 nm. This is also excited by a 488 nm argon-ion laser.
The fluorescence intensity emitted is proportional to the quantity of binding sites for the fluorescent compound on the cell/particle. So the more binding sites there are, the more fluorescence is emitted, and therefore the higher on the fluorescence intensity scale the data appears.
For example, if we have cells stained with CD3-FITC Ab�s, the cells that have the CD3 Ag will bind the CD3-FITC Ab�s. These will be CD3+ve cells. The FITC fluorescence intensity of cells with CD3-FITC Ab�s bound will be higher than cells without CD3-FITC binding, therefore appear further to the right on the FITC scale.
As more and more different fluorochromes are being developed, multi-colour flow cytometry, using several different lasers, is becoming possible, giving a great deal more information about a cell population, using only a small quantity of sample.
One of the main uses of flow cytometry today is for immunophenotyping. Immunophenotyping is the term applied to classification of cells by identifying specific proteins on their surface, using antibodies.
When a fluorescent marker is conjugated to a MAb, it can be used to identify a particular cell type based on the individual antigenic markers of the cell. This can be done by adding the conjugated Ab�s to a mixture of cells/particles. This process is called staining or labelling.
Antigens on a cell can be numerous or very few. Using Ab�s with fluorescent markers attached we can bind to these Ag�s and calculate how many there are on a cell, or we can identify cells that have a certain type of Ag as apposed to some other Ag e.g. CD4 or CD8 Ag�s.
Immunophenotyping using flow cytometry has become the method of choice in identifying and sorting cells within complex populations. Applications of this technology have occurred in both basic research and clinical laboratories. Immunophenotyping has broadened our understanding of diseases such as the progression of HIV infection, the origin and nature of leukaemia and lymphoma, and the regulation of haematopoietic cell differentiation and maturation.
Fluidics, optics and electronics are the three main systems that make up a flow cytometer.
The basic operation of a flow cytometer is that a tube or other type of sample vessel containing the prepared cells under investigation is placed in the flow cytometer. The sample is drawn up from the sample vessel and pumped into the flow chamber (flow cell) through tubing. Cells flow through the flow chamber one at a time very quickly and are presented to one or more light sources (Lasers). The laser beam hits the cells as they pass through the flow chamber. The way the light bounces off each cell gives information about the cell�s physical characteristics. Light scatter and/or fluorescence (if fluorescent molecules are present on the particle) are captured, spectrally filtered and directed to appropriate photodetectors for conversion to electrical signals.
Light scatter is collected at two angles: Forward Scatter (FSC) and side scatter (SSC).
Forward scatter measures scattered light in the direction of the laser path and measures the size of the cell.
Side scatter measures scattered light at 90 degrees to the laser path and measures the granularity of the cell.
As the cells pass through the laser the fluorochromes attached to them are excited and they fluoresce, i.e. emit a specific colour of light depending on the type of fluorochromes. Filters direct the light emitted by the fluorochromes to the colour detectors. The colour detectors (PMTs) collect the different colours of light emitted by the fluorochromes.
The electronics in the cytometer amplify and process the resulting data. They convert analogue data to digital data which is stored in the computer.
This data can be analysed to provide information about subpopulations within the sample.
The flow cell lies at the heart of the instrument. The purpose of the fluidic system of a flow cytometer is to deliver particles of a random three-dimensional suspension in an orderly stream in single file past a specific point intersected by the illuminating beam. This is achieved by injection of the sample into the centre of a stream of liquid (sheath fluid). The design of the flow cell is such that the sheath fluid hydrodynamically focuses the sample stream delivering the cells to the point of detection with an accuracy of �1 or better.
As the sample enters the flow cell chamber, the sample core becomes focused in the centre of the sheath fluid which is faster flowing, and presents a single file of particles to excitation sources. This process is known as hydrodynamic focusing.
The sample pressure and the sheath fluid pressure are different from each other, the sample pressure being the greater one. The sample flow rate is controlled by the sample pressure regulator. This is achieved by changing the sample pressure relative to the sheath pressure. Most benchtop cytometers have fixed sample pressure settings of Low, Medium or High.
Increasing the sample pressure increases the width of the sample core and hence increases the flow rate. A higher flow rate is generally used for qualitative measurements e.g. immunophenotyping, whereas a lower flow rate is used in applications where greater resolution is important, e.g. DNA analysis.
For particles to properly intercept the laser beam, correct operation of fluidic components is critical. This means that the fluidics system must always be free of debris and air bubbles, and is properly pressurised at all times.
The optical system is made up of excitation and collection optics.
The laser and lenses (used to shape and focus the laser beam) are part of the excitation optics.
A collection lens (collects light emitted from the particle-laser beam interaction) and a system of optical mirrors and filters (route specified wavelengths of the collected light to designated optical detectors) make up the collections optics.
As a particle passes through the laser beam, light is scattered and if there are fluorescent markers attached to the molecule they are excited and emit a specific colour of light depending on the type of fluorochromes. The way the light bounces off each cell gives information about the cell�s physical characteristics. As mentioned before, light scatter is collected at two angles: Forward Scatter (FSC) and side scatter (SSC). A photodiode collects the FSC signals. Photomultiplier tubes (PMTs) collect the emitted SSC and fluorescence signals. A system of mirrors and optical filters routes these signals to their detectors.
Bandpass (BP) filters are placed in front of PMTs. This optimises the specificity of the detector for a particular fluorescent dye by allowing only a narrow range of wavelengths to reach the detector.
There are also longpass (LP) and shortpass (SP) filters present. LP filters permit longer wavelength transmission and SP filters permit shorter wavelength transmission.
The purpose of the electronics is to monitor and control the operation of the flow cytometer. After pre-amplification, the signal from a photomultiplier undergoes further processing. An electrical pulse is generated as particles pass through the laser beam and generate light signals. This electronic signal (voltage) is presented to the signal processing electronics and assigned a channel number on a data plot. The instrument is set to respond to signals derived from the particle of interest and to ignore debris and electronic noise. Thus a threshold (or discriminator) is set and only signals with an intensity greater than or equal to the threshold channel value will be processed and sent to the computer.
When a particle enters the laser beam and starts to scatter light or fluoresce, a voltage pulse is created. The light signals (photons) hit one side of the PMT or the photodiode, and are converted into a proportional number of electrons. These electrons are multiplied, hence creating a greater electrical current, which travels to the amplifier and is converted to a voltage pulse. When the particle is in the centre of the laser light where the maximum amount of scatter or fluorescence is achieved, you get the highest point of the pulse. The pulse returns to the baseline as the particle leaves the laser beam.
Pulse shape analysis is based on particle speed and size, width of the illumination beam, and with fluorescence, distribution of the fluorochromes within the particle. Different measurements can be made from the same signal, e.g. peak height, integral (area), width, skew, etc. All these will give different information and characteristics of the cell.
Signals can be amplified by applying a voltage to the PMTs, or by increasing the amplification gain. There is a choice between linear or logarithmic amplifiers. Linear amplification is usually used to amplify scatter and fluorescent parameters, e.g. for DNA measurement. For immunofluorescence, log amplification increases the dynamic range so that weak and strong signals can be recorded on the same scale. This is done by expanding weak signals and compressing strong signals.
Next the signals undergo analogue to digital conversion before being transmitted to the computer. The voltage pulse is given a digital value by the Analogue-to-Digital Converter (ADC). The resolution of the data will depend on the ADC. For example, an 8-bit converter will yield 256 channels, and a 10-bit converter gives 1024.
The data is stored in a flow cytometry standard (FCS) format developed for the International Society for Analytical Cytology (ISAC). When a computer saves data from the cytometer, it is saved as list-mode data, which is a listing of cell parameters and their measurements on a cell-by-cell basis.
Once a data file has been saved, the data can be displayed in a number of different plot types. These range from histograms, to 2-D plots such as dotplots, contour and density plots, to 3-D plots such as a tomogram plot.
A histogram is a single parameter plot where the x-axis represents the parameter�s signal value in channel numbers and the y-axis represents the number of events per channel number. Each event is placed in the channel that corresponds to its signal value. Signals with identical intensities accumulate in the same channel. Brighter signals are displayed in channels to the right of the dimmer signals.
A dot plot will take two parameters and plot them against each other. Each event measurement is marked on the plot. As each event is plotted, populations start to emerge. (Each event represents 1 cell or particle).
Plotting FSC against SSC on a linear scale will give you information about the morphology of a cell population:
Plotting two fluorescent parameters on a log scale gives you information about the fluorescent properties of a cell population. Statistical information is also available:
Three parameters can be displayed simultaneously in a 3-D plot:
Gating is used to identify subsets of data or populations. Gate defined populations can be used to generate statistics and limit the number of events collected or saved.
For example, if you are interested in analysing lymphocytes in a sample of blood containing a mixed population of cells, you can place a gate on the lymphocytes and this will therefore show only data from within that gate i.e. the lymphocytes.
No gate has been applied to the FSC vs. SSC plot:
The plot below shows fluorescent data from all the cells in the population:
A region (A) has been applied to the lymphocyte population and all subsequent plots will be gated on region A.
The plot below shows only the fluorescent data from region A thus excluding the rest of the population of cells, giving a clearer picture:
A feature of flow cytometry when performing multi-colour experiments is the presence of spectral overlap. Since fluorochromes have a wide emission spectrum, if these fluorochromes are excited by the same laser then there will inevitably be some overlap in the emission spectra for each one. This is called spectral overlap.
For example, fluorochromes such as FITC and PE are excited by a 488 nm argon ion laser. They will fluoresce optimally at 520 and 576 nm, respectively. Therefore some of the FITC emission will be collected by the PE detector and vice-versa.
As a result of the spectral overlap that occurs between the two signals, the FITC- and PE-positive populations will appear offset, making it difficult to accurately distinguish true populations. This spectral overlap is usually corrected by subtracting a fraction of the FITC signal from the PE signal and vice-versa. This process is called Compensation. Compensation involves specific software or hardware manipulations that mathematically remove fluorescence overlap to simplify multicolour data interpretation and distinguish populations on a dual-parameter histogram.
Below is a dot plot showing uncompensated data. The FITC-positive population is offset and appears to be expressing PE:
The same plot after compensation to remove the spectral overlap i.e. the false PE-positivity, gives a more accurate result:
Flow sorting has greatly extended the research and diagnostic potential of the flow cytometer. This is because it enables subpopulation of cells or particles to be separated from the sample suspension with a high degree of purity. The purified fractions (usually >95% pure) are then made available for morphological or genetic examination, as well as functional assays and therapeutics.
The most common method of sorting cells is by electrostatic deflection of charged droplets. A gate is placed around the cells of interest during acquisition so that the cytometer knows which cells to sort. Using a conductive sheath fluid, as the sample flows through the flow cell, a piezoelectric transducer vertically vibrates the flow cell causing the fluid emerging from the exit nozzle to break up into droplets. The flow cell is charged at the moment a cell of interest is inside the droplet currently being formed. The stream of droplets passes through a pair of charged plates so that droplets which are charged are deflected and collected. The cells can be collected in tubes, slides or multiwell plates. The remaining sample goes to the waste tank.