Download The Principles of Flow Cytometry

왗your lab focus 왘
(4) trained employees to follow the
standard operating procedures manual.
Linden and Kaplan4 emphasized
that rapid identification of transfusion
errors (especially AHTR) as well as
properly identifying the cause of such
errors and implementation of corrective
actions are critical steps in the reduction
of transfusion-associated fatalities. They
cited previous work2 and recommended
a number of strategies that may decrease
the risk of AHTR. A modified version of
these recommendations, including suggestions put forth by Sazama,1 is provided in [T2].
Conclusion
Erroneous transfusion of ABO-incompatible blood is the most prevalent
transfusion error and almost always reflects a preventable breakdown in transfusion protocol and standard operating
procedure. These errors can have disastrous outcomes, accounting for significant
iatrogenic morbidity and mortality. While
most transfusion errors occur at the patient’s bedside and thus are remote from
the blood bank and the immediate oversight of the transfusion service director,
the hospital-wide implementation and
enforcement of transfusion policies can
help minimize the risk and occurrence of
transfusion errors.
1. Sazama K. Reports of 355 transfusion–associated
deaths: 1976 through 1985. Transfusion.
1990;30:583-590.
2. Linden JV, Paul B, Dressler KP. A report of 104
transfusion errors in New York State. Transfusion.
1992;32:601-606.
3. Tourault MA, Mummert TB. Review of
transfusion related fatalities: many preventable.
Hosp Technol Scanner. 1993;11:1-3.
4. Linden JV, Kaplan HS. Transfusion errors: causes
and effects. Transfus Med Rev. 1994;8:169-183.
CE update [cytology | hematology | generalist]
The Principles of Flow Cytometry
Antony C. Bakke, PhD
From the Department of Pathology, Oregon Health Sciences University, Portland, OR
On completion of this article, the reader should be familiar with the basic components of a flow cytometer, the
principles of its operation, specimen analysis, data presentation, and quality control.
Cytology exam 0102 questions and the corresponding answer form are located after the “Your Lab Focus” section, p 213.
왘
The uses of flow cytometry
왘
The basic components of flow
cytometers
왘
Flow cytometry specimen analysis
왘
Flow cytometry quality control
Flow cytometry is defined as the simultaneous measurement of multiple
physical characteristics of a single cell as
the cell flows in suspension through a
measuring device. It is a subset of the
field of cytometry based on the use of
specialized instruments.
Flow Cytometry Compared With
Other Techniques
The original flow cytometer was the
Coulter counter (Beckman Coulter M)
invented by Wallace Coulter in the 1950s.
All current automated, non–slide-based
hematology instruments are flow cytometers. These instruments require cells in
suspension and the ability to move cells
through the instrument. Blood is the ideal
tissue because the cells are already in suspension and can be easily analyzed. The
Coulter principle uses the resistance of a
cell in an electrical current as it passes
through an orifice to determine numbers
of particles and the size of the particle.
Flow cytometry has specifically come to
denote the use of fluorescence measurement, usually with a laser light source. In
laser flow cytometers, light scatter is used
to measure the intrinsic size and granularity of the cell. In addition, fluorescence
can be used to measure extrinsic features
such as specific protein expression and
nucleic acid content using added
reagents, such as fluorescent stains and
antibodies.
Flow cytometers may be compared
with fluorescence microscopes. With a
microscope, the light source is a lamp,
the objective lens focuses the image, and
the eye is the detector. With a flow cytometer, the light source is a laser, an ob-
©
laboratorymedicine> april 2001> number 4> volume 32
jective lens focuses the light, and the detector is a photomultiplier tube [F1]. With
a fluorescence microscope, the observer
views cells and recognizes their different
sizes. If a fluorescent probe is bound to
the cell, a specific-colored fluorescence is
emitted. Flow cytometry does not provide
a focused image as does a microscope,
but it does detect light and transform the
signal into statistical data that can be presented in various graphic formats.
Components of a Flow
Cytometer
Optics
A laser beam is used in most flow
cytometers as the light source to excite a
fluorescent tag (eg, a fluorescent antibody) bound to the cells in the specimen.
An argon laser, which produces blue
light, is the most commonly used,
although some instruments add a red helium-neon laser, and occasionally a mercury arc lamp is substituted. Before
207
왗your lab focus 왘
208
striking the cells, the light from either
source passes through focusing and shaping optics to form an elliptical beam with
a broad, maximal intensity in the center.
The beam then passes through the center
of a flow chamber, which is a quartz cuvette with an approximately 200-µm
inner diameter. The specimen is also injected through the middle of this flow
chamber, where the laser beam strikes the
cell [F2].
When the laser strikes the cell, light
is diffracted around the edges of the cell,
producing a diffraction pattern along the
path of the laser beam. This scattered
light is approximately equivalent to the
cell circumference and is the same wavelength as the exciting laser light (usually
blue). Termed forward scatter (FSc) or
forward angle light scatter (FALS), it is
collected along the same axis as the laser
beam. Because this is a very strong signal, a simple photodiode can convert the
forward-scattered light into an electrical
pulse. However, the direct laser beam
could enter this photodiode, giving an
overwhelming signal. A physical barrier
eliminates the direct laser beam by blocking about a 2-degree angle of light along
the path of the laser, allowing only the
light diffracted at greater angles to pass.
In addition to the forward-scattered
light, blue laser light is also reflected off
the cell and off internal structures in the
cell. This is termed side scatter (SSc),
right-angle light scatter (RALS), or 90degree light scatter (90LS) because it is
detected by a collection objective set at
90 degrees to the incident laser light. It is
a measure of granularity of the cell.
Granulocytes have a much larger sidescattered light signal than do lymphocytes
([F1]; see also [F5]). After being focused
by the objective lenses, the blue light is
reflected by a dichroic mirror. These mirrors direct specific wavelengths of light
into the correct detector. The detector is a
photomultiplier tube, which converts the
light signal into an electrical signal.
Fluorescent probes can also be
added to the cell for detecting specific
molecules in the cell. These probes are
typically antibodies to cellular antigens
with a covalently attached fluorochrome.
The fluorochrome, excited by the laser
1. Rapid analysis; rates of 50-4,000 cells/s
2. Total number analyzed can be very high; 5K100K+ (for rare events and accurate statistics)
3. Lack of information about intracellular probe
distribution
4. Excellent quantitation of fluorescence per cell;
range of 105× change
5. Single cell suspension critical
6. Correlated multiparameter analysis
1. Labor intensive; rates of 100 cells/min
2. Total number analyzed usually low,
1000 or fewer
3. Morphology and intracellular probe
distribution; tissue sections
4. Quantilation of fluorescence limited; eye
sensitivity approximately equal to 10× change
[F1] Light microscopy compared with flow cytometry for the detection of cells and cellular probes.
Reproduced with permission from Becton-Dickinson M.
[F2] Light scatter measurements in a flow cytometer. The blue laser beam intersects the cell in the middle
of the flow chamber. Forward scatter or forward angle light scatter (FALS) is measured along the axis of
the laser beam. Blue laser light reflected off the cell and off internal structures in the cell is termed side
scatter or 90-degree light scatter (90LS). Reproduced and altered with permission from Purdue University
Cytometry Laboratories. The copyright for the original figure cannot be sold or transferred.
light, will fluoresce at a longer wavelength. Fluorescein gives a green fluorescence, phycoerythrin (an algal
protein) gives orange, and peridinin
chlorophyll protein or phycoerythrin
coupled to cyanin 5 produces red fluorescence. Some flow cytometers have a
second laser that can excite other fluorochromes. Like the side-scattered blue
laboratorymedicine> april 2001> number 4> volume 32
light, all of these fluorescent signals pass
through the objective set at 90 degrees to
the incident laser light. Additional dichroic
mirrors reflect specific wavelengths of
light, allowing separation of the signals
from each fluorochrome with minimal
overlap. In front of each photomultiplier, a
final bandpass filter precisely defines the
wavelength of light entering that detector.
©
왗your lab focus 왘
point of the laser, which is crucial for
collecting accurate information about
each cell in the specimen.
Fluidics
The flow cytometer has 2 sources
of fluid passing through it: a sheath
fluid that is usually a buffered saline,
and the specimen containing the cells to
be analyzed. The specimen forms a
core, which is injected into the middle
of the sheath fluid, and both pass
through the flow chamber [F3]. The
outer sheath fluid causes hydrodynamic
focusing, which compresses the specimen core into a small area. Cells are
forced to pass singly through the focal
Electronics
The photodiode detectors or photomultipler tubes first collect the light signal from the cells. In both cases, the
photons produce a pulse of electrical
current. Photomultiplier tubes have a
voltage applied across a series of plates.
This increases their sensitivity so that
fewer photons are required to produce
an electrical current. The initial small
current requires amplification. The amplified current ranges from 0 to 10 V
depending on the strength of the signal.
The voltage signal is an analog signal
and cannot be recognized by the computer. It is converted by an analog-todigital converter into a digital signal,
called a channel, which can range from
0 to 1,023 channels. The channel is a
counter, and all signals of a specific
voltage are counted in a channel corresponding to the strength of the signal.
The amplification of the initial signal can be either linear or logarithmic.
Linear amplification provides a direct
visual relationship and is useful for scatter signals and fluorescent measurements
of DNA. Linear amplification allows a
dynamic range of approximately 200- to
1,000-fold. Logarithmic amplification is
used for most other biologic signals, including immunofluorescence, owing to
a
b
[F3] Flow cell showing hydrodynamic focusing
of the specimen by the surrounding sheath
fluid. Reproduced with permission from BectonDickinson M.
the extreme dynamic range of these signals. Logarithmic amplification gives a
dynamic range of 2,000- to 10,000-fold.
The fluorescent signals have been
discussed above. Although these signals
appear mainly as a single color under
the microscope, actually the emission
from any fluorochrome is broad.
Although fluorescein isothiocyanate
(FITC) appears apple-green, it also has
lower-level emissions in the orange area
of the spectrum. Hence, any green
FITC-stained cell will also generate a
signal in the orange channel, making the
cell appear to be labeled by 2
fluorochromes. To allow multicolor fluorescence, this apparent overlap must be
eliminated. This is done by fluorescence
compensation. A percentage of any
bright signal, such as FITC, is
subtracted from any other associated
signal. The instrument is set up at the
beginning of a run with pure signals of
each fluorochrome, and any associated
fluorescence at the other wavelengths is
compensated [F4].
Specimen Analysis
Specimen Preparation
Specimens are often whole blood
but may be other tissues. For analysis,
the specimen must contain greater than
95% single cells for accurate interpretation of the data. Whole blood is a single
cell suspension, but tissues must be disaggregated into single cells before analy-
c
209
[F4] Electronic compensation corrects spectral emission overlaps. Two-parameter contour plots of fluorescein vs phycoerythrin fluorescence. a, No
compensation produces results with many cells apparently expressing both fluorochromes, although these are mutually exclusive stains (CD4 and CD8). b,
Overcompensation leaves many cells on the axes and may decrease dual expression when present. c, Correct compensation produces results in which
mutually exclusive stains appear to be pure green or orange and few dual-labeling cells are present. FL1-H, x-axis, green; FL2-H, y-axis, orange.
©
laboratorymedicine> april 2001> number 4> volume 32
왗your lab focus 왘
a
b
c
0%
24%
Ungated total particles
M1
Gated
lymphocytes
11%
65%
100
101
102
103
FL1-H
[F5] Data can be displayed on several plots and electronically gated to analyze pure populations. a, Dot plot of lysed whole blood showing forward
scatter vs side scatter with a gate surrounding lymphocytes. b, Overlaid histograms of fluorescein fluorescence showing the difference with total and
gated population analysis. The positive cells within the markers are 1% of the total particles, including debris, but 89% of the gated lymphocytes. c, Dot
plot of fluorescein vs phycoerythrin fluorescence showing quadrant statistics. FL1-H, x-axis, green, fluorescein; FL2-H, y-axis, orange, phycoerythrin.
sis. Often antibodies are used to stain
cells and select subpopulations or evaluate protein expression. In most cases,
these antibodies are commercial products
with robust staining properties. However,
the staining patterns should be confirmed
in the laboratory on each new lot of antibody using appropriate control materials.
Unknown antibodies must be carefully
evaluated for specificity and efficacy of
staining. This is best done under a microscope initially because nonspecific staining or unusual patterns can be better
distinguished with a microscope than
with a flow cytometer. Once approximate
staining characteristics are determined,
the procedure can be fine-tuned for the
flow cytometer.
210
Data Processing
The initial light signal is first converted by the detectors into an electric
current, which is amplified and measured
as an analog quantity, such as 5 V of a
maximum of 10 V. Because this signal is
analog, it must be converted into a digital
signal for computer processing. The signal then becomes an event with a specific
digital channel value such 512 on a
1,024-channel scale. The digital data can
be displayed as 1-parameter histograms
[F5b] or as 2-parameter plots [F4, F5a,
F5c]. The information on a 1-parameter
histogram is described either as the per-
centage of cells within a set of markers
or as the mean fluorescence intensity of a
population. For logarithmically amplified
signals, the geometric mean fluorescence
is used, since it is less influenced by the
extreme ends of the distribution. A 2parameter plot is usually divided into 4
quadrants, each containing a percentage
of the total population [F5c]. This division is used to distinguish between fluorescent and nonfluorescent cells. It also
defines expression and nonexpression of
a cell molecule marked by a fluorescent
antibody or other fluorochrome.
Because the specimen that is analyzed may be complex with several populations plus debris, such as lysed whole
blood, analysis of the total specimen
would give variable and confusing results. Therefore, it is often necessary to
analyze a single population, such as
lymphocytes. Electronic gating allows
separation of the total population into
parts, which can be individually
analyzed [F5a].
Quality Control
There are several kinds of quality
controls. First the flow cytometer itself
must be evaluated for proper function.
This is usually accomplished with standardized fluorescent beads. These give
very precise, reproducible patterns,
which quickly assess instrument func-
laboratorymedicine> april 2001> number 4> volume 32
©
tion. In addition, the beads are fairly stable and can be used daily with little
change in values. The bead values for
instrument settings and signal intensity
are used to generate Levy-Jennings plots
and determine instrument errors.
A second quality control material is
used to set up the appropriate instrument
settings for the type of staining used.
These can be beads or antibody-stained
cells. One of the primary functions of
this second control is to establish the
compensation settings for the instrument;
therefore, either beads or cells should
mimic the staining that will be used on
actual specimens.
The third level of quality control is
a control substance that mimics actual
specimens. These controls are available
commercially and usually consist of
stabilized blood, sometimes with added
tissue-culture cells that mimic a specific cancer cell. These controls can be
used to assess the sensitivity of the instrument. They will have published
ranges for many of the antibodies used
in the laboratory. Because they usually
contain normal, stabilized blood cells,
some proteins characteristic of specific
cancer cells will not be present. For this
reason, some tissue-culture cells may
be included because they express specific proteins that can be used to evaluate other antibodies.
Conclusion
Flow cytometry is used in the clinical
laboratory for a variety of analyses including diagnostic immunophenotyping of lymphomas, leukemias, and congenital
immunodeficiencies; detection of minimal
residual disease in leukemias; monitoring
CD4+ T-lymphocyte counts in HIV
patients; DNA content analysis for prognosis of malignancies; screening of hematologic disorders such as paroxysmal
nocturnal hemoglobinuria and idiopathic
thrombocytopenic purpura; evaluation of
peripheral stem cell products for transplantation; monitoring monoclonal antibody
therapy; and determination of the extent of
maternal/fetal bleeding by detection of hemoglobin F.
Flow cytometry remains a laborintensive area of laboratory medicine
requiring extensive training. Learning
the fundamental principles of flow
cytometry enhances both patient care
and work satisfaction. In addition,
these principles apply to other areas
using cytometry techniques, such as
hematology. For more details and theoretical discussion, the reader is referred to several books 1-4 and
Internet sites.5,6
1. Givan AL. Flow Cytometry: First Principles. New
York, NY: John Wiley & Sons; 1992.
2. Bauer KD, Duque RE, Shankey TV, eds. Clinical
Flow Cytometry: Principles and Application.
Baltimore, MD: Williams and Wilkins, 1993.
3. Owens MA, Loken MR. Flow Cytometry Principles
for Clinical Laboratory Practice, Quality
Assurance for Quantitative Immunophenotyping.
New York, NY: Wiley-Liss; 1995.
4. Melamed MR, Lindmo T, Mendelsohn ML. Flow
Cytometry and Sorting. 2nd ed. New York, NY:
Wiley-Liss; 1990.
5. Purdue University Cytometry Laboratories.
http://flowcyt.cyto.purdue.edu/. Accessed January
8, 2001.
6. Hematopathologic phenotypes made mockingly
simple. Available at: http://www.neosoft.com/
~uthman/cdphobia/cdphobia.html. Accessed
January 8, 2001.
PostScript Picture
211
laboratorymedicine> april 2001> number 4> volume 32