Givan A.L. Flow Cytometry. First Principles

Подождите немного. Документ загружается.

bead cluster. The intensity of the green ¯uorescence will, under care-

ful conditions, correlate with the concentration in solution of the

particular analyte that has been captured.

With this methodology, it is possible to assay for a large number of

cytokines simultaneously in one tube or to test for a complete selec-

tion of clinically relevant enzymes or for antibodies to a series of

allergens. In addition, by using one color bead type for each patient

and then mixing the samples together for analysis, many patients

can be assayed for a particular analyte in a single test tube. Each red/

orange cluster will mark the results from a particular patient.

REPRODUCTIVE TECHNOLOGY

Prenatal Diagnosis

Prenatal diagnosis of fetal abnormalities conventionally involves

testing of cells from either the amniotic ¯uid or the chorionic villus.

Both techniques involve sampling of tissues closely associated with

the fetus; they therefore pose some risk of causing miscarriage of the

pregnancy. For this reason, sampling fetal cells noninvasively has

been a long-held goal.

Small numbers of fetal erythrocytes that have escaped through the

placenta exist in the maternal circulation. These cells can be detected

by their fetal hemoglobin (which distinguishes them from most ma-

ternal erythrocytes). Their presence in signi®cant numbers (greater

than 0.6%) has been used to diagnose a fetal±maternal hemorrhage

across the placenta and to mandate clinical intervention to prevent

the development in the Rh-negative mother of anti-Rh antibodies (see

Chapter 10). Because fetal cells are accessible by drawing blood from

the mother, sampling of them from the mother's circulation involves

no hazard to the fetus and holds the possibility of noninvasive pre-

natal diagnosis of fetal abnormalities. However, prenatal diagnosis

requires DNA, and most of the fetal erythrocytes in the maternal

circulation are not nucleated. Over the past decade, work in this

®eld (led, to a great extent, by Diana Bianchi at Tufts University in

Boston, Massachusetts) has centered on the development of methods

for distinguishing nucleated fetal cells from maternal cells and from

non-nucleated fetal cells and then on methods for isolating these cells

Flow Cytometry220

for subsequent ¯uorescence in situ hybridization (FISH) analysis of

the fetal genome for disease-related sequences. Because the nucleated

fetal cells are very rare (estimated to be fewer than 20 nucleated fetal

erythrocytes per 20 ml of maternal peripheral blood; that is, a cell

ratio of about 1:10

), ¯ow sorting has been proposed as a method for

purifying or at least enriching the population before FISH analysis.

Such sorting requires a ¯ow-detectable method for classifying

nucleated fetal cells. Two protein markers, among many others, that

have been proposed to distinguish fetal cells from maternal cells

are the transferrin receptor and fetal hemoglobin. Antibodies against

fetal hemoglobin have been used (by Bianchi) in conjunction with a

Hoechst stain for DNA to detect only those cells that have fetal

hemoglobin and are also nucleated. As a model system for the relia-

bility of ¯ow sorting to enrich for fetal cells, blood from women

carrying male fetuses has been studied. The presumptive fetal cells

from the maternal blood are sorted (based on Hoechst and anti-fetal

hemoglobin staining) and then tested with chromosome-speci®c

FISH probes for the presence of a Y chromosome or for only a single

X-chromosome (proving that the sorted cells were, indeed, from the

male fetus and not from the mother). Results from these experiments

are promising. Although these techniques are still at early stages of

investigation, there is considerable optimism that ¯ow sorting of fetal

cells may ful®ll the need for prenatal diagnosis of fetal abnormalities

without the ethical problems posed by invasive sampling of low risk

pregnancies.

Determining Sex

In another facet of reproductive technology, ¯ow sorting has been

used to separate sperm bearing X and Y chromosomes. If you own a

dairy herd, you want your cows to produce more cows (and few

bulls). Conversely, beef farmers want males. Because bovine X chro-

mosomes are considerably larger than Y chromosomes, sperm bear-

ing X chromosomes have approximately 4% more DNA that their Y

bearing brothers. By staining with a Hoechst dye, the two classes of

sperm can be sorted, with a ¯ow cytometer, according to their ¯uo-

rescence intensity. Sex determination of the o¨spring, resulting from

insemination with the sorted sperm, is not perfect (think of the pos-

Research Frontiers 221

sibility of overlap between the intensities of sperm that are separated

by only 4%), but it is not bad. There is now a commercial ¯ow cy-

tometer that has been developed speci®cally for sorting animal sperm.

Sex-selected calves are being produced more or less routinely by ar-

ti®cial insemination with sorted sperm. The technology has also been

applied to the high-stakes world of race horse breeding.

Speaking of high-stakes worlds, there is now at least one commer-

cial clinic that provides sperm sorting for humans (``for prevention of

genetic disorders'' or ``to increase their probability of having a child

of the less represented sex in the family''). Although human sperm

bearing X and Y chromosomes di¨er in DNA content (and ¯uores-

cence intensity) by only 2.8%, the technique does appear to work

with some reliability. In a 1998 summary from the clinic, 14 preg-

nancies from insemination with X-bearing sperm were reported:

92.9% of those were female as desired. The procedure is not inex-

pensive, it may have arguable ethical implications, and it is not per-

fect. The question remains as to whether it could be a glimpse of the

future. If it is, then we need to ponder the fact that most American

families seem to want daughters. . . .

FURTHER READING

Section 9 in Current Protocols in Cytometry provides methods and some

theory on many ¯ow cytometric functional assays (including those men-

tioned here and many others). Chapter 20 in Darzynkiewicz is a discussion

of methods related to staining lymphocytes for activation antigens.

A special issue of Cytometry (Vol. 10, No. 5, 1989) is devoted to ``Cy-

tometry in Aquatic Sciences.'' A more recent issue of Scientia Marina

(Vol. 64, No. 2, 2000) is also devoted to the same subject: ``Aquatic Flow

Cytometry: Achievements and Prospects.'' In addition, Chapter 31 of

Melamed et al. discusses the application of ¯ow cytometry to higher plant

systems.

Section 4 in Diamond and DeMaggio has several methodological chapters

discussing reporter genes, GFP, and cell tracking dyes. The classic reference

on GFP is by Chal®e M, et al. (1994). Green ¯uorescent protein as a marker

for gene expression. Science 263:802±805. A good review of the uses of ¯ow

cytometry for GFP analysis is by Galbraith DW, Anderson MT, Herzen-

berg LA (1999). Flow cytometric analysis and FACS sorting of cells based

on GFP accumulation. Methods Cell Biol. 58:315±339.

Flow Cytometry222

Discussion of the WACS methodology can be found in Krasnow MA,

Cumberledge S, Manning G, et al. (1991). Whole animal cell sorting of

Drosophila embryos. Science 251:81±85.

The gel microdroplet technique is described in articles by Powell KT,

Weaver JC (1990). Gel microdroplets and ¯ow cytometry: Rapid determi-

nation of antibody secretion by individual cells within a cell population.

Bio/Technology 8:333±337; and Weaver JC, Bliss JG, Powell KT, et al.

(1991). Rapid clonal growth measurements at the single cell level. Bio/

Technology 9:873.

Research on ¯ow cytometry of microorganisms is reviewed in Chapter 29

of Melamed et al. and in Section XI in Darzynkiewicz. Microbiological

methods are described in Section 11 of Current Protocols in Cytometry.A

useful book edited by D Lloyd (1993) is Flow Cytometry in Microbiology

(Springer-Verlag, London).

A good reference on the ¯ow cytometry of DNA fragments is Kim Y, Jett

JH, Larson EJ, et al. (1999). Bacterial ®ngerprinting by ¯ow cytometry:

Bacterial species discrimination. Cytometry 36:324±332. The proposed

technique for sequencing DNA is described in a paper by Jett JH, Keller

RA, Martin JC, et al. (1989). High-speed DNA sequencing: An approach

based upon ¯uorescence detection of single molecules. J. Biomol. Struct.

Dynamics 7:301±309.

Results using multiplexed bead ¯ow cytometry have been reported by

Carson RT, Vignali DAA (1999). Simultaneous quantitation of 15 cytokines

using a multiplexed ¯ow cytometric assay. J. Immunol. Methods 227:41±52.

A review of progress in developing ¯ow methods for prenatal diagnosis is by

Bianchi DW (1999). Fetal cells in the maternal circulation: Feasibility for

prenatal diagnosis (review). Br. J. Haematol. 105:574±583. In addition,

Chapter 20 by JF Leary in Darzynkiewicz, 1994 discusses general issues of

rare cell cytometry as well as speci®c reference to the sorting of fetal cells.

A provocative, nontechnical article on the ability to choose, with the aid of

a ¯ow cytometer, the sex of our children was written by L Belkin (``Getting

the Girl'' in The New York Times Magazine, July 25, 1999, pp 26±55).

Results from a clinic that sells this technique to prospective parents have

been reported by Fugger EF, Black SH, Keyvanfar K, Schulman JD (1998).

Births of normal daughters after MicroSort sperm separation and intra-

uterine insemination, in-vitro fertilization, or intracytoplasmic sperm injec-

tion. Hum. Reprod. 13:2367±2370.

Research Frontiers 223

Flowing On: The Future

In the Preface, I referred to the way ¯ow cytometry has moved over

the past three decades in directions that have surprised even those

intimately involved in the ®eld. It is obvious that attempts to predict

the future of ¯ow cytometry should be left to informal discussions

among friends. The temptation to speculate is, however, an impos-

sible one to resist. This chapter should therefore be read in the spirit

in which it was written: with a large measure of skepticism (prefera-

bly among friends, after a good dinner and a good bottle of wine).

The capabilities of state-of-the-art research cytometers should,

over the next years, continue to move toward easier analysis of par-

ticles of greater size range with more sensitivity and greater speed

than they do now. Although currently including particles from over

100 mm (like plant protoplasts) down to bacteria and DNA fragments

on research instruments, this range may soon become possible on

benchtop cytometers. Exploitation of this ¯exibility may ®nally lead

to greater use of ¯ow cytometers for looking at plant cells and also at

microbiological systems. It may also be possible to use cytometers to

look at the properties of isolated organelles such as mitochondria and

chloroplasts. It is likely that some type of ¯ow cytometry will begin

to be used more routinely in bacteriology and environmental labo-

ratories. Recent developments with multiplexed bead-based assays

lead many to guess that ¯ow cytometry is also poised to enter the

biochemistry laboratory, as an alternative to the ELISA technique

for assaying soluble components.

High-speed drop sorters may continue to increase in speed, al-

though the high pressures required in these sorters may have reached

their limit in terms of the viability of the sorted cells. Optical sorting

(by using a laser to inactivate unwanted cells or chromosomes) may

225

Flow Cytometry: First Principles, Second Edition. Alice Longobardi Givan

 2001 by Wiley-Liss, Inc.

ISBNs 0-471-38224-8 (Paper); 0-471-22394-8 (Electronic)

begin to provide a practical, rapid alternative to drop sorting. An

obvious method for allowing more e¨ective use of time and sample

on conventional drop sorters would be to sort four or more popu-

lations (instead of just two) simultaneously. Although slow to catch

on, four-population sorting is now available on commercial instru-

ments. This involves charging some drops more than others so that

drops can be de¯ected slightly to the left and to the right as well as

strongly to the left and to the rightÐgiving four options and allowing

four populations to be sorted during each run. Four populations is

not the technical limitÐif biologists can think of experiments that

demand more.

Extra lasers have been appearing even now on benchtop cytometers;

excitation lines may continue to become less limiting in the choice of

¯uorochromes and the design of experiments. Developments in ¯uo-

rochrome technology, in conjunction with extra lasers, have already

facilitated super-multiparameter analysis by allowing the use of wider

regions of the spectrum without unsurmountable problems of spectral

overlap. Whereas routine instruments today look at forward and side

scatter as well as three or four ¯uorescence parameters, some high-tech

instruments already quite happily look at 11 ¯uorescence parameters

(plus forward and side scatter), and, with demand, the electronics are

capable of handling even more. The question really concerns, then,

the ability of the human intellect to plan rational experiments and to

cope with the information that can be obtained in a short period of

time from particles that are analyzed for large numbers of parameters.

Even after more than a decade of multiparameter benchtop cytometry,

it is not clear that our ability to design meaningful experiments with

multiple measured parameters has yet caught up with our ability to

measure those parameters. Despite the ability to look simultaneously

at three or four ¯uorescent stains quite routinely on many instru-

ments, most people still seem quite happy to stick to two. Whether

this is related to general conservatism, the high cost of additional anti-

bodies, problems with cross-reactivity in complex staining systems, or

some inherent human discomfort with forays into multidimensional

reasoning is a question that may be answered in the future.

There may well be increasing interest in analysis of the intrinsic

characteristics of cells that lead to alterations in auto¯uorescence.

John Steinkamp, Harry Crissman, and others at the Los Alamos Lab-

oratory have been using ¯ow systems to study the time character-

Flow Cytometry226

istics (that is, the decay kinetics) of ¯uorescence emission, as a handle

for studying alterations in the microenvironment of ¯uorochromes

bound to molecules in the cell; their results are exciting and might

be used to study auto¯uorescence in the future. On a larger time

scale, biochemical kinetics experiments may become more popular

with increasing use of functional probes. It may also happen that ¯ow

cytometrists and microscopists will begin, at last, to talk to each other

when they discover that they share common interests in image analysis

microscopy. Laser-scanning cytometry already marks the beginning

of methodological communication between the analysis strategies

typical of ¯ow cytometry and the microscope-based comfort zone

required by pathologists. A cytometry laboratory of the future may

well be equipped with a collection of ¯uorescence and confocal mi-

croscopes, laser-scanning cytometers, and ¯ow cytometersÐall being

used, as appropriate, by the same people.

The extra information collected by rapid cytometers looking at

many parameters will certainly increase the demand on data storage

systems. Ten years ago, when writing the ®rst edition of this book, I

said that, against general scienti®c practice, ¯ow cytometrists might

need to begin to learn to wipe out data that they no longer need. Less

expensive media for data storage have now made it possible to avoid

such measures. Zip cartridges and CD-ROMs are the media of choice

now; DVDs may be the choice in the near future (and the next ``per-

fect'' storage medium may already be on someone's drawing board).

Given the easier availability of inexpensive storage media, I see no

reason to expect that our demands for data storage capacity will do

anything other than continue to increase. With regard to software,

the trend may continue toward a healthy proliferation of varied

packages for ¯ow analysis with full compatibility of data acquired on

any and all systems.

I said at the beginning of this book that ¯ow cytometry is currently

moving in two directions at once: Technological advances provide, in

one direction, increasingly rapid, sensitive, complex, but precarious

analysis and, in the other direction, increasingly stable, fool-proof,

and automated capabilities. Thirty years ago, all ¯ow cytometry was

at the complex, but precarious level of development. Now, many of

those early precarious developments have been incorporated into

routine cytometers, and new techniques are appearing in research in-

struments. Over the next few years, this progression will continue:

The Future 227

The vast majority of future ¯ow cytometric analysis will be done

using routine, black-box instruments of increasingly greater capabili-

ties. In association with this trend, expert (arti®cial intelligence) systems

may begin to have an impact. Robotic systems, incorporating stain-

ing, lysing, and centrifugation steps, may allow sta¨ to put blood or

other crude material in at one end and get formatted print-outs of

¯ow data from the other. A remaining question is whether ¯uidics

stability, reagent quality control, and subjectivity in gating will ever

allow us to relax with this level of automation.

And what does the future hold for the spirit of ¯ow cytometry?

Although there will continue to be many new technical developments

(some predicted and some surprising), I suspect that the sense of ad-

venture may be lost from the ®eld as the balance continues to shift

from innovative technology to routine use. The trade-o¨ for that loss

of excitement could be the satisfaction that people who understand

¯ow cytometry may feel from having played a part in the comfort-

able acceptance of this powerful technique by the broad scienti®c

community.