Rapid single-step method for flow cytometric detection of surface and intracellular antigens using whole blood

0 1996 Wiley-Uss, Inc. Cytometry 25:58-70 (1996)

Rapid Single-Step Method for Flow Cytometric Detection of Surface and Intracellular Antigens

Using Whole Blood Carol Francis and Mark Carle Connelly

Immunocytometry Department, Ortho Diagnostic Systems, Inc., Raritan, New Jersey

Received for publication July 15, 1994; accepted March 3, 19%

Fixationlpermeabilization methods used for the detection of intracellular antigens by flow cytometry often result in the destruction of cellular morphology and surface immunoreactivity, properties useful in flow cytometry for the characterization of cells in heterogeneous populations. In addition, a majority of these methods are incompatible with whole blood and require that peripheral blood leukocytes (PBLs) be puritied prior to fixation. This article describes a new technique for the

rapid detection of both intracellular and cell surface antigens, while preserving cell morphology, through the use of a single-step fixationlpermeabilization reagent, ORTHO PermeaFix (OPF). OPF is compatible with whole blood, allowing for the di- rect preparation of PBLs without prior cell separation. An additional red blood cell lysing reagent was not required because RBC lysis occurred upon re- suspension of OPF-treated whole blood samples in isotonic solution. Discrimination of leukocyte populations by light scatter after OPF treatment was comparable to matched unfixed live cells. In addition, absolute lymphocyte and white blood cell (WBC) counts were not sig&cantly affected when OPF-treated cells were compared with un8xed cells. Treatment of whole blood fkom 7 normal donors showed no significant difference in percentage of cells positive for CD2, CD3, CD4, CD8, CD16, or CD19 between fixed and unfixed samples when cells

were stained before fixation, and no difference in CD3, CD4, CD8, CD16, or CDl9 percentages when cells were stained following fhation. Monoclonal antibodies specific for intracellular antigens located at various sites within the cell were tested on fked samples. OPF-treated peripheral blood lymphocytes showed greater than 95% reactivity for the inner mitochondrial membrane protein bcl-2, and the cytoskeletal cytoplasmic protein vimentin. TIA-1, a cytolytic granule-associated protein, showed differential reactivity within lymphocyte subsets, from a low of 8 k 2% in CD4+ cells to 89 f 6% in CD16+ cells, when whole blood from five normal donors was fixed and stained. Reh cells treated with OPF showed greater than 95% reactivity for the internuclear protein TdT. A comparison of OPF with two other fixatiodpermeabilization procedures, 1% parafor- mddehyde followed by 45% ethanol and 0.25% paraformaldehyde followed by 0.2% Tween 20, showed that only OPF could be used both prior to or following cell s h c e staining with no effect on antigen detection while allowing optimal detection of all of the intracellular antigens tested. 0 19% Wey-Liss, Inc.

Key terms: PermeaFix, flow cytometry, permeabilization, intracellular antigen, whole blood, immu- nophenotyping, light scatter

Flow cytometry allows for the simultaneous measure- ment of multiple correlated parameters on a single cell. Flow cytometric analysis of heterogeneous cell populations, such as leukocytes in whole blood, allows cell types to be differentiated based upon cell size and morphology. When reacted with specific fluorescently labelled monoclonal antibodies, even cells with similar physical properties may be differentiated and quantitated based on cell surface antigen expression (5,15,23).

Although surface molecules provide important infor- mation about cell type, differentiation and activation, intracellular molecules, both cytoplasmic and nuclear, can

provide valuable insight into the regulation and function of cells. For example, the bcl-2 proto-oncogene has been found to encode a mitochondrial protein that blocks pro- grammed cell death (1).

Because flow cytometry has the capability of being able to analyze multiple fluorescent parameters simulta- neously, it lends itself especially well to analyzing multi-

Address reprint requests to Mark Carle Connelly, Ortho Diagnostic Systems Inc., 1001 US HWY 202, P.O. Box 350, Raritan, NJ 08869-0606.

CELL SURFACE AND INTRACELLULAR ANTIGEN DETECTION 59

ple intracellular (or cell surface) markers or both intracellular and cell surface markers of individual cells. Being able to detect both cell surface and intracellular proteins on a cell would allow one to analyze functional subsets of cells for the presence or absence of intracellular effector molecules, cytokines, cell cycle regulating proteins, or the presence of intracellular pathogens such as the Hu- man Immunodeficiency Virus (HIV) (2,8,10,13). Also, because flow cytometers have the ability to analyze large numbers of cells and can gate and collect events based on a particular cell subset, detection of positive cells present at low frequency is possible.

To label intracellular molecules the target cells must be fixed and made permeable to antibodies or other probes, and to their labels. Due to the stringent conditions required for permeabilizing cells, these treatments often destroy sigmficant numbers of cells and damage or eliminate the morphologic and antigenic properties of those cells that remain (14). This may severely limit the applications to which detection of intracellular molecules is relevant and remove many of the advantages or- dinarily gained by the use of flow cytometry. In addition, many of the procedures described for fixation and permeabilization of cells are not compatible with whole blood and require cells be first purified (18,20) or are lengthy and require multiple treatment steps (3,7,16).

We have developed a single-step fixatiodpermeabilization reagent, ORTHO PermeaFix (OPF) for the rapid detection of both cell surface and intracellular molecules in PBLs using whole blood. Development of this reagent was accomplished, in part, through the use of X-STAT, a commercially available reagent formulation software. Us- ing this software, combinations and concentrations of reagents were optimized so that the cell parameters most useful in flow cytometry were preserved. Samples treated in this way had intact light scatter profiles, low background fluorescence and little cell debris or aggregates. Detection of cell surface antigens was comparable to that of unfixed samples, regardless of whether the cells were stained prior to or following fixation. Pizzolo et al. (22) have found OPF useful in the detection of intracellular CD3, CD22, and TdT in leukemic cells. We have identi- fied several additional intracellular proteins, located at various sites within the cell, which were clearly detectable following treatment of normal PBLS with OPF. These proteins included vimentin, a cytoskeletal cytoplasmic protein (9); TdT, a nuclear protein (4,25); bcl-2, an inner mitochondrial membrane protein (1,12h and TIA-1 antigen, a cytolytic granule-associated protein detectable with TIA-1 monoclonal antibody (3). Also, we were able to show that cells treated in this way could be stained for both intracellular and cell surface antigens without com- promising subset specificity using combinations of TIA- 1, CD4, CD8, and CDl6 monoclonal antibodies. Finally, we compared OPFs performance with the performance of two other fixatiodpermeabilization methods: 1 % paraformaldehyde followed by 45% ethanol (PFE) and 0.25% paraformaldehyde followed by 0.2% Tween 20 (PFT) (24,26). An evaluation of cell surface and intracellular

staining showed that only OPF could be used both prior to or following cell surface staining with no effect on antigen detection while allowing optimal detection of all of the intracellular antigens tested.

MATERIALS AND METHODS Blood Samples and Cell Lines

Blood from healthy donors was collected into ethyl- enediaminetetraacetic acid (EDTA) Vacutainer tubes (Becton Dickinson, Rutherford, NJ) for preparation of whole blood samples or sodium heparin Vacutainer tubes (BD) for preparation of Ficolled samples. Samples were held at room temperature (RT) and used within 24 h. For the fixatiodpermeabilization comparison experiments, blood mononuclear cells were isolated by density cen- trifugation on Ficoll-Hypaque (Pharmacia, Piscataway, NJ), washed twice with RPMI 1640 medium (GIBCO Lab- oratories, Grand Island, NY) then adjusted to 2 X lo7 cells/ml. The Reh acute lymphocytic leukemia cell line, positive for the intranuclear antigen TdT, and the Ramos, Burkitt lymphoma cell line negative for TdT, both available from ATCC, were maintained in continuous culture.

PermeaFix Reagent Development Optimization of OPF as a mononuclear cell fixation/

permeabilization reagent for use with whole blood was accomplished through the use of X-STAT sohare uohn Wiley, New York, NY). PermeaFix consists of several components in an isotonic buffer: formaldehyde for cell fixation, a detergent for membrane permeabilization, DMSO to facilitate the transport of molecules across the cell membrane, and dimethyl-dinitro benzoic sulfate for preservation of certain intracellular antigens. The amount of each of these components, in addition to the sample treatment conditions, was optimized using mathematical models based on the prospective reagent’s ability to preserve cell surface immunophenotype and light scatter while allowing for optimum intracellular antigen detection with minimum cell loss and autofluorescence.

Staining for Cell SurEace Antigens Whole blood or ficolled blood samples were stained

with monoclonal antibodies obtained from Ortho Diag- nostic Systems (Raritan, NJ). Three-color reagents were TRIO CD4/CDS/CD3, CD16/CD19/CD3, and isotype control. Single-color reagents were fluorescein isothiocy- anate (FITC)-conjugated OKTll (CD2), M14 (CD14), and M 15 (CDl5), and phycoerythrin (PE)-labelled OKT4 (CD4), OKT8 (CDS), and OK-NK (CDl6). Cell surface staining of whole blood was performed by incubating 100 p1 whole blood with 10 p1 of the appropriate antibody reagent for 20 min in the dark at RT. Staining of Ficolled cells was performed by incubating 100 pl Ficolled cells with 10 ~1 of the appropriate antibody reagent for 30 min in the dark at the temperature specified for that particular fixation/permeabilization procedure. Samples treated for intracellular antigen detection were further processed as described below. Unfixed whole blood samples were treated with 2 ml ORTHO-Mune Lysing Reagent (Ortho

60 FRANCIS AND CONNELLY

Diagnostic Systems) containing ammonium chloride (AmCl). These samples were vortexed and incubated for at least 10 min to allow for the lysis of the red blood cells. Samples were read immediately after lysing was complete or stored for up to 2 h at 2°C in the dark.

Fixation and Permeabilization Two milliliters of 1 X PermeaFix, (Ortho Diagnostic

Systems) was added to 100 pl cells or whole blood, either stained or unstained for cell surface antigens. Sam- ples were vortexed and incubated for 40 min at RT, un- less otherwise noted. Samples were then centrifuged at 400s for 3 min. The supernatant was aspirated and the cell pellets resuspended in 2 ml wash buffer containing 1.5% BSA, .0055% EDTA, and 5% goat serum in PBS. The whole blood samples were then incubated for 10 min in wash buffer at RT in order to lyse the red blood cells. Following the incubation, samples were centrifuged at 400g for 3 min. The supernatant was aspirated and the cell pellet was washed in 2 ml wash buffer. Cells were then either resuspended in 1 ml PBS containing 1 % formaldehyde (Polysciences, Warrington, PA) or stained for intracellular and/or cell surface antigens in which case they were resuspended in 100 pl wash buffer. The paraformaldehyde/ethanol (PFE) procedure was performed as described by Toba et al. (26). Briefly, 2 X lo6 mononuclear cells were fixed with 1 ml cold 1% paraformaldehyde (PF)(Sigma, St. Louis, MO) in PBS on ice for 15 min. Cells were pelleted and resuspended in 2 ml 45% ethanol in saline and incubated for 30 min at RT. Cells were then washed twice with the appropriate hypotonic buffer and either resuspended in 2 ml67% PBS, 0.5% PF, 0.01% sodium azide, 1% fetal bovine serum, or stained for intracellular and/or cell surface antigens, in which case they were resuspended in 100 pl hypotonic buffer. The paraformaldehyde/Tween 20 (PFT) protocol was performed as described by Schmid et al. (24). A total of 2 X 10" mononuclear cells were treated with 2 ml cold 0.25% PF at 4°C for 1 h. Cells were pelleted and gently resuspended in 2 ml 0.2% Tween 20 in PBS and incubated for 15 min at 37°C. Cells were washed with 2% normal calf serum, 0.1% sodium azide in PBS (PBSAz), and either resuspended in 1 ml 1% PF for flow cytometric analysis or resuspended in 100 pl PBSAz for additional staining.

Intracellular Staining

One hundred microliters of fixed samples were incubated with 10 pl of the following monoclonal antibodies directed against intracellular antigens: vimentin (clone V9, Sigma, St Louis, MO), bcl-2 (clone 124, Dako, Carpen- teria, CA), TIA- 1 (Coulter Immunology, Hialeah, FL), and TdT (anti-HTdT-6-FITC, Supertechs, Bethesda, MD). Each antibody was titered against the appropriate cell preparation to determine the optimum working concentration. Depending on the cell prep and antibody specificity, 0.1- 0.8 pg of antibody was added to each sample. For anti- TdT and anti-TIA- 1, the corresponding commercially available isotype controls were used to detect nonspe-

cific background fluorescence. For anti-vimentin and bc1.2, a matched concentration of mouse IgG, isotype control (Fisher Scientific, Pittsburgh, PA) was used. Sam- ples were incubated for 30 min at RT for OPF samples, 37°C for PFT samples and at 4°C for PFE samples. After washing 2-3 times in 2 ml of the appropriate buffer, cells reacted with unconjugated antibodies were then incubated with sheep F(ab' ), anti-mouse FITC conjugate (Sigma) for 30 min. at the above temperatures in the dark. Following incubation, cells were washed again 2-3 times in the appropriate buffer and resuspended in 1 ml PBS containing 1% formaldehyde for OPF and PFT samples or 67% PBS, 0.5% PF, 0.01% NaN,, 1% FBS for PFE samples.

Double Staining: Cell Surface and TIA-1 Antigens Cells were first fixed with OPF then stained for TIA-1

using the fixation, permeabilization, and intracellular staining methods described above. Following intracellular staining, the cells were resuspended in 100 pl wash buffer. These cells were then stained for cell surface markers, washed once following the incubation with antibody, then resuspended in 1 ml PBS containing 1% formaldehyde.

Absolute Count Determinations Blood from five normal donors was prepared in one of

two ways: Unfixed RBC-lysed samples were prepared by adding 2 ml AmCl RBC lysing reagent to 100 pl whole blood. Samples were incubated at least 10 min prior to analysis. Fixed samples were prepared by incubating 100 p1 whole blood with 2 ml OPF for 40 min at RT. Samples were centrifuged and RBCs were lysed following resus- pension in wash buffer. Cells were washed once and r e suspended in 2.1 ml 1% formaldehyde. For both treatment groups, triplicate tubes were prepared.

Flow Cytometry Five parameter analysis was performed using a Cytor-

OnAbsoZute flow cytometer equipped with a 15 mW air- cooled 488 nm argon-ion laser and absolute count capability. Sample acquisition and analysis were performed using Immunocount software v. 1 or v. 2 (Ortho Diag- nostic Systems) ( 19). Lymphocytes [forward scatter (FSC) vs. right scatter (RSC)] and positive subsets of lymphocytes (fluorescence vs. RSC) were gated using Immu nocount's autogating algorithm. Absolute lymphocyte counts were determined by autogating on the lymphocyte cluster. Calibration and daily verification of the absolute count were performed using the Ortho-Count Cal- ibration kit. Optical alignment and gain setting stability were verified daily with QC3 TM fluorescent microspheres (Flow Cytometry Standards Corp.,Research Triangle Park, NC).

RESULTS Preservation of Light Scatter Properties of PBLs in Whole Blood Following F i t i o n / P e n n e a b i n

With PermeaFix Fresh whole blood was treated with OPF and the light

scatter properties of the PBLs compared with those from


A B 250

200

150

100

0- 0 50 100 150 200 250

. . 250 I

. _ 0

0 50 100 150 200 250

Right Scatter Right Scatter

Total Region A Events Events %Total 19994 5910 29.6

Total Region A Events Events %Total 19948 6393 32.0

FIG. 1. Light scatter profiles of fresh whole blood from a normal donor treated with either A) AmCl for RBC lysis or B) PermeaFix for fixation and permcabilization of PBLS. The purity of the lymphocytes in Region A was confirmed with antibodies against CD14 and CD15.

unfixed AmCI-lysed preparations from the same donor. The light scatter of PBLs following fixation and permeabilization with OPF was not adversely affected as shown by the example in Figure 1. In order to determine the purity of the lymphocyte population defined by light scatter, in fixed samples, whole blood drom seven donors was stained with anti-CD 14 or anti-CD 15 antibodies, then fixed with OPF. Anti-CD14 antibodies detect CD14 on greater than 90% peripheral monocytes, whereas anti- CD15 antibodies detect CD15 on greater than 90% peripheral granulocytes. Both antigens are detected on less than 5% of peripheral blood lymphocytes. Analysis of OPF-fiued samples showed no decrease in purity of cells within the lymphocyte gate when compared with unfixed AmC1-lysed samples. For OPF-treated samples, 0.6 0.4% of the cells within the lymphocyte gate were CD14 positive compared to 0.8 f 0.6% for the unfixed AmC1- lysed samples. In the case of CD 15,OPF-treated samples showed 0.9 2 0.8% of the cells within the lymphocyte gate were positive, compared to 0.8 f 0.7% for the unfixed AmCI-lysed samples. Paired t-test analysis showed no sigruficant difference in the amount of CD14- and CDl5-positive cells within the lymphocyte gate of OPF- treated samples compared with unfixed AmC1-lysed samples. The ability of these antibodies to detect antigen on fixed cells was confirmed by gating on the appropriate cell clusters; 90 f 4% of monocytes and 99 f 0.4% granulocytes were positive for CD14 and CD15, respec-

tively. These results were similar to those obtained for the unfived AmCI-lysed samples, in which 91 f 4% of the monocytes and 99 2 0.5% of the granulocytes, respectively, expressed these antigens. Paired t-test analysis of CD14 monocyte staining and CD15 granulocyte staining showed no sigmficant difference between unfixed AmCl- lysed and OPF-fixed samples.

Red Blood Cell L y s h Occurs Following OPF Treatment Without the Need for Lysing Reagents We found that RBCs lysed following OPF treatment.

After the 40-min fixation step, OPF-treated whole blood was spun down and resuspended in isotonic buffer, such as PBS or PBS with serum. The RBCs lysed, by visual inspection, within 10 min at RT. The degree of RBC lysis was such that the lymphocyte, monocyte, and granulocyte populations were clearly discernible on a plot of FSc vs. RtSc (Fig. 18) without having to subject fixed whole blood to an additional RBC lysing reagent.

Absolute Cell Count Determinations U s h g Flow Cytometry

The lysis of intact red blood cells following treatment with OPF made it necessary for us to determine whether any cell loss was occurring with leukocytes. Samples drom five donors were treated either as unfixed AmCI- lysed or OPF-fixed. As determined by paired t-test andy- sis, the absolute lymphocyte and total white blood cell


Table 1 Absolute Cell Count Determinattom’

Lymphocytes Total WBC Donor Unfixed OPF Unfixed OPF 1 2615 +- 211b 2590 f 129 9067 *792 8701 2 288 2 1913 2 30 1851 f 210 5641 ? 348 5204 * 260 3 1397 2 199 1378 f 72 5872 f 767 5214 rf: 165 4 2373 * 224 2149 +- 47 8310 f 613 7112 2 118 5 3817 f 429 3480 +- 45 7766 f 933 6802 rf: 168

”Cell count per uL whole blood. bResults are expressed as mean f SD of triplicate samples.

( W C ) counts were not significantly different between the two groups for all of the donors (Table 1).

Preservation of Cell Surface Staining of OPF-Treated PBLs

The process of fixing and permeabilizing cells can often destroy the immunoreactivity of cell surface markers. Treatment with OPF, however, preserved the immunoreactivity of cell surface proteins. Whole blood was stained with three-color cocktails containing anti-CD4, anti-CD8, and anti-CD3 or anti-CDl6, anti-CD19, and anti-CD3; or with single-color anti-CD2, then fixed. The reactivity of samples stained prior to OPF treatment was compared with that of unfixed AmCI-lysed controls (Fig. 2). For each marker, the percentage of positive cells was not significantly different between OPF-fixed and unfixed samples (Table 2).

Typically, the fluorescence intensities of stained, then OPF-treated cells appeared brighter than those of stained, unfixed AmC1-lysed samples. CDl6 was the only reactivity tested where the fluorescence intensity appeared to be lower on fixed cells than unfixed. Although fixed samples could have been read using the same fluorescent gain settings as for unfixed AmC1-lysed samples in gen- eral, the fluorescent gains were reoptimized for each treatment group in order to place the negative cell population appropriately within the first decade. The apparent increase in intensity could, therefore, have been due to the changes in the fluorescent gain settings introduced between OPF-fixed and unfixed AmCl-lysed samples. In order to compare objectively the fluorescence intensities of unfixed, AmC1-lysed samples with OPF-treated samples, stained prior to or after fixation, the separation between the positive and negative populations was calcu- lated for each antibody tested. The mean channel fluorescence (MCF) intensity of the negative population was subtracted from the MCF intensity of the positive population and expressed as the “delta MCF”. The delta MCF of the samples stained prior to OPF fixation were greater than those of the unfixed, AmC1-lysed samples for all antibodies except anti-CDl6 (Table 2).

To determine whether cell surface phenotypes could also be stained after OPF treatment, cells were fixed with OPF, washed, then stained with antibodies. Cells were successfully stained with all antibodies tested, including antibodies to CD2, CD3, CD4, CD8, CD16, and CD19.

The same percentage of cells were positive for each marker when staining was done post-OPF treatment as were positive in the matched unfixed, AmCI-lysed controls except for CD2 (Table 2). The delta means for cells that were stained post-OPF were lower than those of un- fmed, AmC1-lysed cells for CD4/3, CD19, and CD2, although the lower fluorescent intensity did not interfere with the ability to clearly distinguish positive from negative cell populations.

Optimization of Fixation Time To Preserve Cell Surhce Staining While Allowing for the Maximal

Detection of Intracellular Antigens Samples were OPF fixed for various times then stained

for intracellular or cell surface antigens. Whole blood samples were fixed for various times between 0 and 120 min, then immediately centrifuged and resuspended in wash buffer. RBC lysis was complete within 10 min for all samples except for the 0 time OPF treatment, which required ammonium chloride lysing reagent. Light scatter properties of PBLs remained intact even after 2 h of OPF treatment. Samples stained for cell surface antigens following OPF fixation showed no difference in the percent positive events, compared with unfixed, AmCl-lysed samples, for any of the cell surface markers at all timepoints. Fluorescence intensities did show a decrease in the delta MCF with increasing fixation time (Fig. 3). Samples fixed for up to 40 min showed no more than a 10% decrease in delta mean channel fluorescence for all cell surface markers except CD16. CDl6 showed a 19% decrease in delta MCF compared with time 0.

The ability to detect intracellular molecules was mon- itored by immunoreactivity with anti-vimentin. The anti- vimentin reactivity of cells increased with time, reaching a plateau after approximately 40 min of room-temperature OPF treatment (Fig. 3). Over 90% of the gated lymphocytes were positive for vimentin staining after 40 min (Fig. 3). Although the optimum conditions may differ depending upon the molecules of interest and the cells being studied, an OPF fixation time of 40 min at room temperature gave the greatest intracellular access with the least disruption to cell surface markers (Fig. 3).

Detection of Intracellular Antigens Following OPF Treatment

In order to determine whether OPF fixation for 40 min at RT could be used to detect other intracellular antigens, fixed PBLs were reacted with monoclonal antibodies against bcl-2 (an inner mitochondrial membrane protein), T cell intracellular antigen-1 (TIA-1, a cytoplasmic granule-associated protein), and vimentin (a cytoskeletal cytoplasmic protein). Greater than 95% of the gated lymphocytes stained positive for bcl-2, and vimentin (Fig. 4). In contrast, a discrete population representing only 25% of gated lymphocytes stained positive for TIA-1, consistent with its expression being restricted to a subset of lymphocytes (3). TdT staining was performed to determine whether OPF treatment is capable of permeabilizing the nuclear membrane. Reh cells, positive for TdT,

CELL SURFACE AND INTRACELLULAR ANTIGEN DETECTION

A 0 C

1000 1000 I000

v) 100 100 100

c1 cc 2 2

10 10 10

1 1 1 0 50 100 150 200 250 0 50 100 150 200 250 1 10 100 1000

Right Scatter Right Scatter CD4-Fl L'C

Total Region B Total Region B Gated by: B

19994 4111 20.6 19948 4431 22.2 Reg %Tot Events MeanX MeanY -~ 3.0 123 13.2 8.6 -+ 22.1 YO8 23.2 200.2 +- 74.1 3047 148.8 5.Y ++ 0.8 33 121.1 190.4

Events - O / r T o t a l Events - O / r T o t a l

D

1000

100

10

1

CD4-PITC

Catcd by: B Reg %,To1 Events McanX MeanY - - 2.1 91 28.9 19.3 FIG. 2. Detection of T-cell antigens on the surface of PBLs. Whole blood from a normal -+ 22.5 loo0 42.5 224.8 donor was stained with the three color reagent CD4KDWCD3 and A,C) exposed to AmCl +- 74.3 3295 157.8 23.1 or B,D) fmred and permeabilized with OPF. Percentages of CD4- and CD8-positive T-cells ++ 1.1 51 118.8 214.1 were determine by gating on CD3-positive events and analyzing for CD4 and CD8.

and Ramos cells, negative for TdT, were treated with OPF and reacted with a monoclonal antibody against the intranuclear antigen TdT. Greater than 95% of the Reh cells stained positive for TdT, whereas less than 2% of the Ramos cells stained positive. In all experiments, gated cells (lymphocytes or cell line clusters) gave less than 1% positive cells with isotype controls. Unfixed cells gave less than 3% positive cells when reacted with monoclonal antibodies specific for intracellular antigens.

Shnultaneous Detection of Multiple Antigens, Cell Suiface and Intracellular, on OPF-Treated Cells TIA-1 has been described by Anderson et al. (3) as an

intracellular antigen restricted to a subpopulation of CD 8+ T-lymphocytes and NK cells. To determine whether TIA- 1 retains its proper subset distribution following OPF

fixation, cells were fixed then stained for the intracellular TIA-1 antigen followed by staining for cell surface CD4 and CDS. Two-color analysis of lymphocytes showed a difference in TIA-1 staining within subtypes (Fig. 5 ) . Analysis of bright CD8+ lymphocytes from five normal donors showed an average of fifty-seven 10% staining positive for TIA-1, whereas only 8 2 2% of CD4+ cells stained positive. A substantial number of dim CD8+ cells, suggestive of natural killer (NK) cells, were also TIA-I positive. In order to determine the proportion of NK cells which were TIA-1 positive, fixed cells were stained for both TIA-1 and CDl6 (Fig. 6). Analysis of five donors showed 89 k 6% of CD16+ cells were positive for TIA-1. These results agree with the findings of previously published reports (3,16) while offering additional insight into the subset distribution of TIA-1. Furthermore, these


Table 2 Effect of OPF Treatment on Cell Surface Immunoreactivity

Treatment CD 413 CD 813 CD 3 CD 16 Cd l V CD 2

Unfixed control % 48 2 9" 2 6 + 11 7 6 t 6 8 + 5 11 + 4 79 * 5 DeltaMCFb 1 1 5 k 6 1 7 7 & 4 1 5 9 2 9 1 2 8 * 5 1 3 4 5 3 1 0 6 k 7

DeltaMCF 1 2 3 ? 4 * 1 8 0 2 4 ' 1 6 9 + V * 1 1 7 + 1 1 * 138+ 1' 1 1 0 * 8 * Staining after OPF treatment % 48 2 8 2 6 t 10 7 4 + 6 8 + 5 12 5 5 75 2 6

DeltaMCF 1 0 6 2 4 ' 179+ 3 1 5 4 5 3 1 2 9 * 4 1 2 9 5 2' 75 k 2'

Staining prior to OPF treatment % 48 2 8 262 11 7 4 * 7 7 * 4 12 + 4 78 2 4

'Results are expressed as mean t SD (n = 7). bDelta mean channel fluorescence. Mean fluorescence positive population-mean fluorescence negative population. 'Significantly different, P < 0.05 as determined by paired t-test. All results compared to unfixed control.

A 250

200

Delta 150 Mean Channel Fluorescence

100

50

0

0 10 20 30 40 50 60 90 120

Time (min.)

B 120 -1 I

J 8 0 -

% Anti-Mmenlin Positive

Lymphocytes 60 -

40 -I /

2o -I/ 0 1 , I I I I I I I I '

0 10 20 30 40 50 60 90 120

Time (min.)

FIG. 3. The effects of fixation time on cell surface and intracellular staining Whole blood from a normal donor was fixed with OPF for up to 120 min then stained for cell surface antigens and intracellular vimentin. Although the percentage of cells positive for cell surface antigens remained unchanged for all of the timepoints, A) the delta means CD8 (open square), CD3 (filled circle), CD19 (filled diamond), CD16 (open circle), and CD4 (filled square) decreased with time. B: The percentage of cells positive for intracellular vimentin increased.

results indicate the restricted distribution of TIA- 1 within lymphocyte subsets was preserved following treatment with OPF.

Comparison of OPF Treatment With Other Fixation M e t h o d s

Ficolled PBMCs were treated in one of three ways: OPF treatment as previously described, 1% PF and 45% EtOH

(PFE), or 0.25% PF and 0.20% Tween 20 (PFT). The light scatter; cell surface staining, both pre- and post-fixation; and intracellular staining results of these three methods were compared. Evaluation of the light scatter results for the OPF and PFE samples showed that lymphocytes and monocytes were discernible as two defined populations, whereas the PFT-treated samples showed a diminished ability to distinguish these cell types (Fig. 7). In addition, the OPF and PFE forward scatter gain settings were the same as that for unfixed cells, whereas PIT required a two-fold higher FSC gain setting, consistent with the observations of Schmid et al. (24).

Cell surface staining analysis, using three-color reagents (for detection of CD4/3, CD8/3, CD3, CD16, and CDl9) showed that cells could be stained prior to fixation using any of the above methods; however, the CD3 and CD4/3 percent positive cells were decreased by approximately 10% and 20% of the total, respectively, in the PET-treated group when compared with unfixed controls (Table 3). Delta mean channel fluorescence values were consistently within 2 5 channels o f the unfiied control samples for the OPF group. PJT samples showed delta MCF values that were 10 to 25 channels lower than unfixed samples, whereas PFE samples showed a reduction of twenty to sixty channels for all of the lymphocyte subsets tested, with the exception of CD4/3 where a three channel increase was observed.

Evaluation of staining on cells post-PFE fixation showed only CDl6 could be detected. A comparison between cells OPF-fixed and then stained and unfixed cells showed that the percent positive cells for all of the markers tested was preserved post-fixation in the OPF-treated samples (Table 3). Samples PIT-fixed then stained again showed a reduction of approximately 10% in CD3- and CD3/4-positive cells. Fixation with either OPF or PFT prior to staining resulted in samples with similar delta MCF values for CD 4/3, CD 8/3, and CD19. However, the PIT samples showed lower delta MCF values for CD3 and CD 16.

In order to compare intracellular staining capabilities of the three methods, Ficolled PBMCs were fixed and permeabilized by each of the three methods, then reacted with antibodies against vimentin, bc1.2, and TIA-1. Reh and Ramos cells were also treated and reacted with anti-TdT antibody. All the methods detected the cytoplas-

A


Anti-TdT

D

I 250

24 7%

150 I 99 4% i 1 10 100 1000

TIA-1-FITC

B

99.6% 150

97.0%

100

50

0 1 10 100 1000

Anti- hcl-2-FITC

C

250

200

2 150

2 = 100

50

0

c

E s

1

65

98.76

10 100 1000

Anti-vimentin-FITC

FIG. 4. Fixation of cell line cultures or whole blood samples with OPF allowed for the detection of several intracellular antigens. Samples were treated with OPF then reacted with monoclonal antibodies against A) TdT, a nuclear protein, in Reh cells; B) blc-2, an inner mitochondrial protein, C) vimentin, a cytoskeletal protein; and D) TIA-1, a cytoplasmic granule associated antigen. B through D are examples of results obtained in periphelal blood lymphocytes. Mouse isotype control antibody (shown in outline) was adjusted to the same concentration for each test antibody.

mic antigens vimentin and TIA-1 equally well with some variability in fluorescence intensity (Table 4). Subopti- mal results were obtained when PFE-treated cells were stained for bc1.2 and TdT. Only 67% of cells permeabilized by the PFE method stained positive for bc1.2 compared to greater than 90% bcl.2-positive cells using the PFE and OPF methods. No significant staining was observed with anti-TdT following PFE treatment of Reh cells. In contrast, PFT- and OPF-treated cells showed nearly 100% positive staining for this antigen. All of the intracellular antigens tested were detected equally well in the PFT and OPF samples with only some differences in the fluorescence intensity of staining between these methods for vimentin and TdT.

DISCUSSION We have evaluated a reagent, PermeaFix (OPF), devel-

oped for the fixation and permeabilization of cells for analysis by flow cytometry. Cell lines, Ficolled PBMCs, or whole blood samples have all been successfully treated with this reagent for the detection of cell surface and

intracellular markers. The procedure is simple with min- imal handling of samples; the reagent is added directly to cells with optimal fixation and permeabilization occurring concurrently during a single 40-min incubation. Cells treated in this way are ready for both cell surface and intracellular staining. In the case of whole blood, a brief incubation in PBS following fixation results in the lysis of red cells. The reason for RBC lysis, even under isotonic conditions, is not known but may be due to the high protein content within fixed RBCs. Although red blood cell membranes become permeabilized when exposed to OPF, hemoglobin remains fixed within the cell and does not leach out. When these cells are then placed in “isotonic” buffers, the high protein concentration within these cells may cause an influx of buffer which causes the cells to burst.

Although red cells were lysed following OPF treatment, leukocytes remained intact. Comparison of light scatter between unfixed ammonium chloride-lysed samples and OPF-treated samples shows that OPF preserved the light scatter of PBh sufficiently to allow the differ-

66

A B

FRANCIS AND CONNELLY

C

ir:

P V

%

D

Ig GI-FITC TIA-1-FITC

Reg %Tot Events MeanX MeanY Keg %To1 Events MeanX Meany ~~ 49.6 1887 27.8 23.3 -- 22.2 1184 354 26.1 -+ 50.2 19 I I 23.8 157.0 -+ 46.6 2489 31.6 156.5 +- 0.2 8 85.8 22.5 +- 27.7 1479 116.3 29.5 ++ 0.0 0 0.0 0.0 ++ 3.5 186 105.6 148.5

1 1 10 100 1000

Ig (;,-FITC

Keg %Tot Events MeanX MeanY -- 71.2 3862 24.2 13.9 -+ 28.5 1549 32.4 197.2 +- 0.2 10 94.3 27.6 ++ u.1 5 75.1 212.4

1 1 10 100 1000

TIA-1-FITC

Reg %Tot Events McanX MeanY ~~ 59.9 3243 34.0 19.0 -+ 8 8 475 38.9 214.4 FIG. 5. T-cell subset analysis of intracellular TIA-1 following fmtion with OPF. Whole

blood was fwed with OPF then reacted with A) CD4 and TIA-1 isotype control, B) CD4 and +- 10.9 593 101.9 16.0

TIA-1, C) CDS and TIA-1 isotype control, or D) CDS and TIA-1. ++ 20.4 1107 120.0 195.5

entiation of lymphocytes, monocytes and granulocytes (Fig. 1). Leukocyte numbers were also preserved following OPF fixation. Flow cytometric absolute count determinations of OPF-fixed samples showed no decrease in the number of lymphocytes or total WBCs when compared with unfixed AmCI-lysed samples (Table 1). The Eact that no cell loss occurred with OPF treatment is an important consideration when only small quantities of sample are available, or when preparing samples for rare event analysis where cell loss during sample preparation could lead to false negative results. In addition, very little or no cell debris or cell aggregates were present in OPF- fixed samples as determined by light scatter analysis and microscopic examination.

Although a number of methods exist for permeabilizing cells for immunofluorescent staining of intracellular antigens, most of these methods result in sigmficant changes to the light scatter profiles of PBLs such that

whole blood samples must be ficolled prior to treatment (24,26). This added step results in extended processing time, and more critically, the potential for cell loss. Of the three fixationlpermeabilization methods evaluated- OPF, 1% paraformaldehyde and 45% ethanol, 0.25% paraformaldehyde and 0.2% Tween 20-only OPF could be used on whole blood samples.

All of the cell surface markers evaluated in this study were clearly detectable whether staining was done prior to or following OPF treatment. HLA-DR, CD38, CD45, and IL-2 receptor have also been successfully detected when stained cells were treated with OPF (data not shown). No differences were observed in the percentage of cells staining positive for CD2, CD3, CD4/3, CD8/3, CD14, CD15, CD16, or CDl9 when staining was done prior to fixation (Table 2). Post OPF-fixation staining with three-color reagents CD16/CD19/CD3 and CD4 /CD8/ CD3 showed no difference in the percent positive

A

1000

100 c w r(

n 10


1 1 10 100 1000

lg GI-FITC

B

1000

100

10

1 1 10 100 1000

TIA-1-FITC

67

Reg %Tot Events MeanX MeanY Reg %Tot Events MeanX MeanY -- 85.9 4674 24.3 26.8 -- 68.7 3707 34.7 25.2 -+ 13.9 756 27.6 140.5 -+ 2.4 128 36.8 137.9 +- 0.2 9 84.2 26.7 +- 20.0 1081 109.7 24.9 ++ 0.0 2 117.5 164.5 ++ 8.9 481 125.6 148.5

FIG. 6. Detection of TIA-1 antigen in NK cells following fixation of whole blood with OPF. OPF was fixed then stained with A) CD16 and TIA-1 isotype control or B) CD16 and TIA-1.

values for CD4/3, CD W3, CD16, CD19, or CD3 when compared with unfixed, AmC1-lysed cells. However, for CD2, cells stained after the 40 min fixation did show a statistically signtficant decrease in CD2-positive cells, suggesting some antigens may be more sensitive to OPF treatment than others, and that the treatment conditions may need to be optimized for the antigen( s) and antibodies being studied.

An important objective for this fixation process was to preserve cell surface antigens for subset evaluation of intracellular antigens. Both OPF and PFT fixation, prior to or following staining, maintained the ability to detect and quantitate CD4/3, CD8/3, CD3, CD16, or CD19 lymphocyte subsets. However, although detectable, PFT fixation resulted in fewer CDb3-positive and CD3-positive cells, when compared to unfixed controls, regardless of whether fixation was done before or after staining. In contrast, when staining was done following PFE-fixation only CDl6-positive cells could be clearly detected and quantitated.

Fixing cells with OPF prior to staining, in some cases, resulted in a slightly decreased fluorescence intensity. This was easily overcome by staining cells for cell sur- Eace antigen detection prior to fixation. This approach not only preserved the fluorescence signal but in many cases actually improved it (Table 2). In the case of CD16, a significant decrease in fluorescence intensity was ob-

served with the FITC conjugate, even when cells were stained prior to fixation. However, it is interesting to note that with the PE conjugate this decrease was not observed. This suggests the decreased staining intensity observed with CDl6 may be due to the conjugate used and not due to damage of the CDl6 antigen on the cell surface. Decreases in fluorescence intensity were observed with both PFT and PFE, regardless of whether staining was done pre- or post-fixation.

An important property of OPF is its ability to fix and permeabilize cells in a single step for intracellular antigen detection. Although no single fixatiodpermeabilization technique may be expected to be optimal for all intracellular markers, all of the intracellular antigens tested in this study, located at various regions in the cell, were clearly detectable following OPF treatment and staining with specific antibody. OPF was shown to provide access to numerous intracellular compartments including the nucleus ( TdT), cytoplasmic organelles such as mitochondria (bcl-2) and cytoplasmic granules (TIA- l), in addition to cytoplasmic proteins such as vimentin (Fig. 4). Cytoplasmic CD3, CD22, and nuclear TdT have also been successfully detected in OPF-treated PBLs from leukemic patients (22), and cytoplasmic immunoglobu- lins have been detected in bone marrow mononuclear cells following permeabilization with OPF (27). A comparison of OPF fixation with alternate fixatiodpermeabil-


A 250 I

B 250

200

150

100

50

0 0 50 100 150 200 250 0 50 100 150 200 250

C 250

200 - * s 2

150

100 k.

50

0 0 50 100 150 200 250

Right Scatter Right Scatter Right Scatter

Total Rcgion A

IO.O(Kl 6653 06.5 E v e n t s -

Total Region A Events C v e n t s m 10,000 6870 687

Total Region A

10,002 7089 70 9 E v e n t ! , -

Right Scatter

Total Region A

10.000 6808 68.1 maldehyde and 45% ethanol, or D) OPF.

FIG. 7. Light scatter profiles of ficolled fresh PBMs from a normal donor: A) untreated, B) fixed and permeabilized with 0.25% paraformaldehyde and 0.2% Tween 20, C) 1% parafor- Evcnrs - " / o T o t a l

ization methods on intracellular staining showed that PFT treatment also permitted detection of TdT, TIA-1, vimentin, and bc1.2, whereas PFE treatment permitted detection of only the cytoplasmic antigens vimentin and TIA- 1, with suboptimal detection of bc1.2 and an inability to detect TdT. The PFT method has been used previously to detect BrdUrd incorporation into DNA using anti-BrdUrd antibody, suggesting it has the ability to permeabilize the nuclear membrane (26). Furthermore, Aiello et al. re- ported a decreased ability to detect bc1.2 when methanol fixation was used (1). Based upon these observations, treatment of cells with alcohol, while allowing permeabilization of intranuclear membranes such as the nuclear and mitochondrial membranes, may result in the extraction, masking, or destruction of epitopes on bc1.2 and TdT antigens. In addition to staining for intracellular antigens following OPF incubation, it was possible to stain for intracellular markers during the fixatiodpermeabilization step (data not shown). Addition of anti-vimentin antibody directly to cells in OPF during the fixation

step showed similar values in percent positive events when compared with samples fmed prior to intracellular staining.

The true potential of OPF may be realized when staining for both cell surface and intracellular antigens on individual cells. Because OPF-treatment preserves the light scatter and cell surface immunoreactivity of cells, leucocytes can be gated on the basis of light scatter (FSC vs. RSC), fluorescence of single or multiple cell surface antigens, or a combination of light scatter and fluorescence (F1 vs RSC). Analysis of gated cells for intracellular antigens can then be performed. In this way, we were able to determine the percentage of total lymphocytes positive for TIA-1, and also the subset distribution of this intracellular antigen. TIA-1 was found in only 20-30% of gated lymphocytes. Two-color analysis of TIA- 1 vs. T-cell subset antigens CD4 and CD8 clearly showed the subset distribution of TIA-1 to be restricted primarily to CD 8+ cells (Fig. 5) . A total of 7-1 1% CD4+ and 40-60% bright CD8+ cells stained positive for TIA-1, confirming the


Table 3 Evaluation of FixationlPerme~iliration Metboa3 on Pre- and Post-Fixation Cell Surface Staining

Treatment CD 4/3 CD 813 CD 3 CD 16 CD 19 Unfixed % 48 19 72 18 6

Delta MCF’ 110 159 125 138 122 Staining prior to PFE treatment % 47 20 71 18 8

Delta MCF 113 120 92 78 103 Staining prior to PFT treatment % 40 18 62 19 9

Delta MCF 96 148 110 113 106 Staining prior to OPF treatment % 48 20 72 17 8

Delta MCF 11; 161 122 133 121 - 21 59

- - - Staining after PFE treatment %

Staining after PFT treatment % 38 18 GO 21 9 Delta MCF 95 152 81 74 101

Staining after OPF treatment % 47 20 67 22 8 Delta MCF 91 154 119 110 101

- - - - Delta MCF

“Delta mean channel fluorescence. Mean fluorescence positive population-mean fluorescence negative population. bDescreet positive populations could not be detected.

Table 4 Evaluation of FixationfPermeabilization Methods on

Inhacellular Staining

Treatment Vimentin bc1.2 TIA-1 TdT PFE % 1 0 0 67 46 1

MCF 167 66 126 77 96

49 64 PFT % 100 95

MCF 205 88 134 OPF % 100 98 48 100

MCF 165 93 135 124

findings of previously published reports (516). From these results it was also apparent that dim CD8+ cells, suggestive of NK cells, were TIA-1 reactive. This was confirmed by two-color analysis of TIA-1 vs. CD16. 80- 90% of CD16+ cells stained positive for TIA-1. These results extend the findings by Anderson et al. (3) to in- clude not only NK clones but also peripheral blood cells.

Although the function of the TIA-1 protein has not been determined, it has been shown to be associated with cytolytic granules in CD8 T cells and CDl6 N U , cells with cytolytic capabilities. In HIV-infected individ- uals, a functional depletion of NK activity, along with a sigmficant reduction in CD16CD56 cells, has been observed ( 17). Subset analysis of TIA- 1 expression may provide insight into this depleted NK activity. In particular, it may be possible to determine whether the decrease in NK activity correlates with decreased expression of TIA- 1 by NKs, fewer NK cells expressing TIA- 1, or simply lower numbers of NK cells.

The differential reactivity of lymphocyte subsets, high CDl6 reactivity with low CD4 reactivity, shows that OPF-treatment fixes intracellular components sufficiently to prevent their leaching from the interior of the cell. In this respect, OPF could be especially effective in the analysis of HIV-infected samples, where it has been found that other permeabilization methods such as methanol or methano1:acetone result in the “leaching” of HIV proteins from infected to uninfected cells, causing these cells to be detected as HIV positive (7,ll).

Because of OPF’s ability to permeabilize the nuclear membrane, as demonstrated by TdT staining of treated Reh cells, it was necessary to next determine whether DNA analysis of OPF-treated cells could be performed. Ficolled PBLs were treated with OPF for up to 40 min at 4”C, 22OC, or 3?C, then stained with propidium iodide (PI) using the procedure provided by Clevenger (6). Al- though PI staining of DNA was possible with this procedure, DNA distributions with increased CVs and reduced mean channel fluorescence were obtained when compared with cells treated with 0.5% paraformaldehyde and 0.1% Triton X-100 (data not shown). Because OPF was developed for optimal detection of cell surface and intracellular antigens, DNA quantification on OPF-permeabi- Iized cells would require additional optimization studies. Although DNA detection with DNA staining dyes is not optimal using OPF, detection of specific DNA sequences by performing in situ hybridization has been demonstrated using OPF-treated PBLs ( 2 1 ). Patterson et al. were able to determine the percentage of CD4-positive cells containing HIV-1 DNA, and the degree of cell surface CD4 modulation in these infected cells, using specific HIV-1 DNA probes on OPF-treated PBMCs.

PermeaFix has been optimized for use with whole blood samples for the detection of cell surface and intracellular markers while preserving light scatter properties. In addition to detection of TdT in Reh cells, OPF has also been successfully used with cell lines in the following applications: detection of cytoplasmic CD3 in CEMs and detection of intracellular p24 antigen in HIV- 1-infected HUT 78s (data not shown). Based on these findings, the potential applications of OPF may be far-reaching in both flow cytometry and cell biology.

ACKNOWLEDGMENTS We thank Dr. Utpal Chakraborty and Dr. Houston

Brooks for their help in the development of ORTHO Per- meaFix, Gioacchino De Chirico, Dr. Tom Mercolino, Dr. Kesh Prakash, and Tim Nolan for their support in OPF


Commercialization and Dr. George Janossy for the critical reading of this manuscript.

1.

2.

3.

4.

5

6.

7.

8.

9.

10.

11.

12.

13

14

LITERATURE CITED Aiello A, Delia D, Borrello M.G, Biassoni D, Giardini R, Fontanella E, Pezzella F, Pulford K, Pierotti M, Della Porta G: Flow cytometric detection of the mitochondrial bcl-2 protein in normal and neoplas- tic human lymphoid cells. Cytometry 13502-509, 1992. Anderson P, Blue ML, OBrien C, Schlossman SF: Monoclonal antibodies reactive with the T cell receptor chain: Production and characterization using a new method. J Immunol 143:1899-1904, 1989. Anderson P, Nagler-Anderson C, O’Brien C, Levine H, Watkins S, Slayter HS, Blue ML, Schlossman SF: A monoclonal antibody reactive with a 15-kDa cytoplasmic granule-associated protein d e f i e s a subpopulation of CD8+ T lymphocytes. J Immunol 144:574-582, 1990. Bardales RH, Carrato A, Fleischer M, Schwartz MK, Koziner B: De- tection of terminal deoxynucleotidyl transferase ( TdT) by flow cytometry in leukemic disorders. J Histochem Cytochem 37509-513, 1989. Bass HZ, Hardy WD, Mitsuyasu RT, Wang YX, Cumberland W, Fahey JL: Eleven lymphoid phenotypic markers in HIV infection: Selective changes induced by zidovudine treatment. J Acq Imm Def Syn

Clevenger C.V, Bauer KD, Epstein A. L: A method for simultaneous nuclear immunofluorescence and DNA content quantitation using monoclonal antibodies and flow cytometry. Cytometry 6:208-2 14, 1985. Cory JM, Ohlsson-Wilhelm BM, Brock E, Sheaffer NA, Steck ME, Eyster ME, Rapp F Detection of human immunodeficiency virus- infected lymphoid cells at low frequency by flow cytometry. J Im- munol Methods 105:71-78, 1987. Costigliola P, Tumietto F, Ricchi E, Chiodo F Detection of circulating p24 antigen-positive CD4 + cells during HIV infection hy flow cytometry. AIDS 6~121-1125, 1992. Franke WW, Schmid E, Osborn M, Weber K: Different intermediate- sized filaments distinguished by immunofluorescence microscopy. Proc Natl Acad Sci 10:5034-5038, 1978. Hayden GE, Walker KZ, Miller JFAP, Wotherspoon JS, Raison RL: Simultaneous cytometric analysis for the expression of cytoplasmic and surface antigens in activated T cells. Cytometry 944-51, 1988. Heynen CA, Holzer TJ: Evaluation of a flow cytometric model for monitoring HIV antigen expression in vitro. J Immunol Methods 152:25-33, 1992. Hockenbery D, Nunez G, Milliman C, Schreiber RD, Korsmeyer SJ: Bcl-2 is an inner mitochondrial membrane protein that blocks pro- grammed cell death. Nature 348334-336. 1990. Jacob MC, Favre M, Bensa JC: Membrane cell permeabilisation with saponin and multiparametric analysis by flow cytometry. Cytometry 12:550-558, 1991. Jacobberger J, Fogleman D, Lehman JM: Analysis of intracellular antigens by flow cytometry. Cytometry 7356-364, 1986.

5WK-897, 1992.

15. Kestens L, Vanham G, Gigase P, Young G, Hannet I, Vanlangendonck F, Hulstaert F, Bach BA Expression of activation antigens, HLA-DR and CD 38, on CD8 lymphocytes during HIV-1 infection. AIDS 6:793-797, 1992.

16. landay A,JeMings C, Forman M, Raynor R Whole blood method for simultaneous detection of surface and cytoplasmic antigens by flow cytometry. Cytometry 14:433-440, 1993.

17. Ianday A, Ohlsson-Wilhelm B, GiorgiJV: Application of flow cytometry to the study of HIV infection. AIDS 4:479-497, 1990.

18. McSharry JJ, Costantino R, Robbiano E, Echols R, Stevens R, Lehman JM: Detection and quantitation of human immunodeficiency virus- infected peripheral blood mononuclear cells by flow cytometry. J Clin Micro 28:724-733, 1990.

19. Mercolino TJ, Connelly MC, Meyer EJ, Knight MD, ParkerJW, Stelzer GT, DeChirico G: Immunologic differentiation of absolute lymphocyte count with an integrated flow cytometric system: A new con- cept for absolute T cell subset determinations. Cytometry 22:48- 59, 1995.

20. Ohlsson-Wilhelm BM, Cory JM, Kessler HA, Eyster ME, Rapp F, landay A Circulating human immunodeficiency virus (HIV) p24 antigen-positive lymphocytes: A flow cytometric measure of HIV infection. J Infect Dis 162:1018-1024, 1990.

21. Patterson BY Goolsby C, Hodara V, Lohman KL, Wolinsky SM: De. tection of CD4+ T cells harboring human immunodeficiency virus type I DNA by flow cytometry using simultaneous immunopheno- typing and pcr-driven in situ hybridization: Evidence of epitope masking of the CD4 cell surface molecule in vivo. J Virol 69:4316- 4322, 1995.

22. Pizzolo G, Vincenzi C, Nadali G, Veneri D, Vinante F, Chilosi M, Basso G: Connelly MC, Janossy G: Detection of membrane and intracellular antigens by flow cytometry following ORTHO PermeaFix fixation. Leukemia 8:672-676, 1994.

23. Prince HE, Jensen ER: Three-color cytofluorometric analysis of CD8 cell subsets in HIV-I infection. J Acq Imm Def Syn 41227-1232, 1991.

24. Schmid I, Uittenbogaart CH, Giorgi JV: A gentle f i t i o n and permeabilization method for combined cell surface and intracellular staining with improved precision in DNA quantification. Cytometry 12:

25. Slaper-Cortenbach ICM, Admiraal LG, Ken JM, van Leeuwen EF, Kr. von dem Borne AEG, T e t t e m PAT: Flow-cytometric detection of terminal deoxynucleotidyl transferax and other intracellular antigens in combination with memebrane antigens in acute lymphatic leukemias. Blood 72:1639-1644, 1988.

26. Toba K, Winton EF, Bray RA: Improved staining method for the simultaneous flow cytofluorometric analysis of DNA content, Sphase fraction, and surface phenotype using single laser instru- mentation. Cytometry 13:60-67, 1992.

27. Van Zaanen HCT, Vet RJWM, DeJong CM, Kr. von dem Borne AEG, Van Oers MHJ: A simple and sensitive method for determining plasma cell siotype and monoclonality in bone marrow using flow cytometry. Brit Jour Haem 1955-59, 1995.

279-285, 1991.

Documents

Rapid single-step method for flow cytometric detection of surface and intracellular antigens using whole blood