A Small-Volume Technique for Simultaneous Immunophenotyping and Apoptosis Detection in Rat Whole Blood by Four-Color Flow Cytometry

Transcription

1 2002 Wiley-Liss, Inc. Cytometry 47: (2002) A Small-Volume Technique for Simultaneous Immunophenotyping and Apoptosis Detection in Rat Whole Blood by Four-Color Flow Cytometry Jose Diaz-Romero, Gerard Vogt, and Gisbert Weckbecker* Department of Transplantation, Novartis Pharma AG, Basel, Switzerland Received 12 April 2000; Revision Received 24 January 2002; Accepted 30 January 2002 Background: Cell permeabilization for the detection of intracellular molecules by flow cytometry is usually incompatible with whole blood. This article describes a new technique for the simultaneous detection of surface antigens and DNA content in rat whole blood. Methods: In 20 l of rat whole blood, DNA staining is obtained by permeabilization of cells using a standard red blood cell lysing reagent (Erythrolyse). Immunophenotyping and apoptosis detection by flow cytometry are achieved by using a combination of three surface markers (CD3, CD4, and CD8 ) and a DNA binding dye (TO-PRO- 3). Results: After a 24-h incubation of whole blood with 1 M dexamethasone, apoptotic lymphocytes were clearly distinguishable from normal lymphocytes by their reduced size and DNA content. The dexamethasone-induced percentage of apoptotic cells was for CD4 and for CD8 T cells, compared with for CD4 and for CD8 T cells in the absence of dexamethasone (data from 10 animals with duplicate samples). Conclusions: We have developed a new technique to permeabilize nucleated cells in microsamples of rat whole blood. The methodology allows simultaneous immunophenotyping and apoptosis detection in rat whole blood. Cytometry 47: , Wiley-Liss, Inc. Key terms: rat; flow cytometry; whole blood; apoptosis; immunophenotyping; permeabilization; TO-PRO-3; dexamethasone Flow cytometry is used for the simultaneous measurement of multiple correlated parameters in a single cell. Flow cytometric analysis of heterogeneous cell populations, such as leukocytes in whole blood, allows cell types to be differentiated based on cell size and morphology. When reacted with specific fluorescently labeled monoclonal antibodies (mabs), even cells with similar physical properties can be differentiated and quantitated based on cell surface antigen expression. Although surface molecules provide important information about cell type, differentiation, and activation, intracellular molecules can provide valuable insight into the regulation and function of cells. To label intracellular molecules, the target cells must be permeabilized. However, permeabilizing treatments are time-consuming, often destroy a significant number of cells, damage antigenic properties of remaining cells, and usually are not compatible with whole blood (1 3). In flow cytometry, whole blood lysis methods have substituted for density gradient separation of peripheral blood mononuclear cells (PBMCs) as the routine preparation for immunophenotyping (4 7). The cells in whole blood are usually labeled first and then the red blood cells are lysed either with ammonium chloride or commercial whole blood lysing solutions. Several red blood cell lysing solutions permeabilize leukocytes (4), but these solutions were developed primarily and were commonly used for human samples. Therefore, it is necessary to evaluate red blood cell lysing solutions for animals. In this article, we evaluated the properties of three commercial red blood cell lysing solutions in Lewis rats. The results show that it is possible to lyse rat red blood cells and permeabilize nucleated cells in a single step by using a commercial reagent and standard procedures. The methodology described permits simultaneous detection of surface antigens and DNA content by flow cytometry in lysed rat whole blood samples. By using this methodology, we have developed a whole blood method for apoptosis detection by modifying the hypodiploid peak method (8) and by using multiparametric analysis of light scatter, DNA content, and surface marker expression of peripheral blood leukocytes. MATERIALS AND METHODS Animals and Blood Sampling Inbred Lewis/Crl BR (LEW) rats were obtained from Iffa Credo (St. Aubin les Elbeuf, France). All the rats were *Correspondence to: Gisbert Weckbecker, Department of Transplantation, Preclinical Research, Building 386/628, Novartis Pharma AG, Lichstrasse 35, CH-4002 Basel, Switzerland. gisbert.weckbecker@pharma.novartis.com Published online in Wiley InterScience ( DOI: /cyto.10081

2 266 DIAZ-ROMERO ET AL. males weighing between 250 and 300 g. They were cared for in accordance with institutional guidelines. Five animals were housed in polycarbonate microisolation cages. A standard diet (Eberle Nafag AG, Gossau, Switzerland, catalog no. 890) and tap water were provided ad libitum. The temperature in the animal rooms was maintained at C with 45% humidity. Fluorescent lighting was provided for 12 h each day. Peripheral blood from rats (0.5 ml per rat) was collected from the sublingual vein in 1.7-ml tubes coated with heparin-lithium (Milian Geneva, Switzerland, catalog no. WHP-19) using a previously described refined method (9). The blood was kept at room temperature and used within 4 h. From each blood sample, two 20- l aliquots were immunostained immediately and two additional 20- l aliquots were immunostained after whole blood culture (see next sections). For some experiments, PBMC were isolated by density gradient centrifugation using Histopaque-1083 (Sigma, Buchs, Switzerland, catalog no ), according to the standard protocol supplied by the manufacturer. Reagents All mabs used in this study were directly conjugated. Fluorescein isothiocyanate (FITC)-conjugated mouse antirat CD3 mab (0.5 mg/ml catalog no D), phycoerythrin (PE)/Cy5 tandem-conjugated mouse anti-rat CD4 mab (0.2 mg/ml catalog no A), and PE-conjugated mouse anti-rat CD8 mab (0.2 mg/ml catalog no B) were obtained from Pharmingen (Basel, Switzerland). PEconjugated mouse anti-rat CD4 mab (100 test/ml catalog no. MCA55PE) was obtained from Serotec (Dottikon, Switzerland). TO-PRO-3 iodide (Molecular Probes, Lucerne, Switzerland, catalog no. T3605) is a monomeric cyanine nucleic acid stain with 642 and 661 nm of absorption and fluorescence maxima, respectively. A TO-PRO-3 working dilution of 20 M was prepared with phosphate-buffered saline (PBS; Sigma, catalog no. D5652) and stored in aliquots at -20 C. Three red blood cell lysing solutions were evaluated in this study: FACS-lysing (Becton Dickinson, Basel, Switzerland, catalog no ), CAL-lyse (Caltag, Zurich, Switzerland, catalog no. GAS-010), and Erythrolyse (Serotec, catalog no. BUF04C). FACS-lysing and Erythrolyse are supplied as 10 concentration solutions and must be diluted 1:10 with distilled water before use. CAL-lyse is a ready-to-use solution. Z-VAD-fmk (Sigma, catalog no. V116), a cell-permeant peptide fluormethylketone inhibitor of caspase activity (10,11), was dissolved in dimethylsulfoxide (DMSO; Fluka, Buchs, Switzerland, catalog no ) to give a final concentration of 50 mm. It was stored at -20 C until use. Aggregated mouse IgG was prepared by incubation at 63 C for 30 min of normal mouse IgG (Caltag, catalog no ). Staining buffer was prepared with PBS containing EDTA 0.02 M (Sigma, catalog no. E4884) and 2% bovine serum albumin (Fluka, catalog no ), ph 6.0. In our hands, this staining buffer containing a high concentration of EDTA was essential to obtain a good cell recovery from microplate wells after 24 h of whole blood culture (see below). Whole Blood Culture and Induction of Apoptosis Aliquots of 20 l heparinized whole blood were transferred into 96-well round-bottomed culture plates (Costar, Wallisellen, Switzerland, catalog no. 3799). Each sample was diluted 1:10 with Dulbecco s minimum essential medium (DMEM) high glucose (Amimed, Allschwil, Switzerland, catalog no. 1-26F01-l) supplemented with penicillin (100 U/ml), streptomycin (100 g/ml), L-glutamine (2 mm) (Sigma, catalog no. G6784), and 2-mercaptoethanol (50 M; Fluka, catalog no ). For induction of apoptosis, dexamethasone (Sigma, catalog no. D2915), an inducer of apoptosis in mature peripheral T lymphocytes (12), was included in the culture medium at a final concentration of 1 M (concentration based on preliminary studies). In one experiment, Z-VAD-fmk was used as an apoptosis inhibitor. Immediately prior to the addition to blood, the stock solution was diluted with culture medium and added at the beginning of the blood culture (final concentration of 200 M). The plates were incubated in a 5% CO 2 incubator at 37 C for 24 h. Magnetic Cell Separation of Apoptotic Cells Isolation of apoptotic cells from whole blood incubated with dexamethasone, as described before, was done by using annexin-v coated MACS microbeads (Miltenyi Biotec, Gladbach, Germany, catalog no ). The blood was washed once with staining buffer (see Reagents ), twice with annexin-v binding buffer (Miltenyi Biotec), and resuspended in a volume of 80 l. Annexin-V coated MACS microbeads (20 l) were added to the blood. After a 15-min incubation period at room temperature, the cells were separated using MS separation columns and Mini- MACS (Miltenyi Biotec) according to the standard protocol supplied by the manufacturer. Immunofluorescence Staining of Whole Blood All the incubation steps were done in 1.2-ml microdilution tubes in a 96-place rack (USA Scientific Plastics, Merenschwand, Switzerland, catalog no ), with shaking at room temperature. All washing steps were done by centrifugation (5 min, 400 g at room tempertaure) in a centrifuge with microplate carriers (Jouan BR4i, Instrumenten-Gesellschaft AG, Basel, Switzerland). The supernatant was removed by aspirating with pasteur pipettes connected to a vacuum. For staining of fresh blood, aliquots of 20 l were dispensed directly from heparin-coated tubes in the microdilution tubes. For staining of whole blood culture, the culture plate was centrifuged, the supernatant was discarded, and the pellet was resuspended in 200 l staining buffer and transferred to microdilution tubes containing 200 l staining buffer. After centrifugation of the tubes, the supernatant was again discarded. Prior to staining, blocking was done by a 30-min incubation with a mixture of 8 l staining buffer and 2 l aggregated mouse IgG. After blocking, samples were incubated for 15 min with a mixture of 2 l TO- PRO-3 (20 M), 0.1 l anti-cd3-fitc, 0.1 l anti-cd8 -PE, 0.5 l anti-cd4-pe/cy5, and 7.3 l staining buffer. Follow-

3 ing incubation, red cells were lysed and leukocytes were permeabilized simultaneously by incubation in 400 l Erythrolyse for 10 min. Erythrolyse was used based on the results of preliminary work to determine which commercial red cell lysis solution was most adequate to use with rat blood. After removal of the supernatant by centrifugation, cells were washed once with 500 l CellWash (Becton Dickinson, catalog no ), resuspended with 100 l CellWash, and kept at 4 C before analysis by flow cytometry. Immunofluorescence Staining of PBMC PBMC were isolated from 400 l of heparin anticoagulated blood by layering onto 400- l Histopaque-1083 in a 1.2-ml microdilution tube. The volume of PBMC was brought up to 400 l with staining buffer and 20- l aliquots were stained with TO-PRO-3, anti-cd3-fitc, anti- CD8 -PE, and anti-cd4-pe/cy5. After staining, samples were treated for 10 min with 400 l Erythrolyse or with 400 l staining buffer. After this step, samples were processed as before. Whole Blood Lysis Evaluation For evaluation of potential interference in antibody labeling, 20 l fresh blood was stained with 2 l PE-conjugated mouse anti-rat CD4 mab for 15 min. For permeabilization studies, dexamethaxone-treated blood was stained with 2 l PE-conjugated mouse anti-rat CD4 mab and 2 l TO-PRO-3 (20 M). After staining, the red blood cell lysis solutions were evaluated according to the following procedures: (A) CAL-lyse, 25 l for 10 min plus 5 min with 200 l deionized water; (B) Erythrolyse 400 l for 10 min; (C) FACS-lysing, 500 l for 10 min; (D) FACS-lysing, 250 l for 5 min. After blood lysis, samples were processed as before. A, B, and C are based on manufacturer instructions after adjusting the volumes to our sample volume (20 l). D is a modified version of C. Flow Cytometry Cells were analyzed on a FACScalibur flow cytometer (Becton Dickinson) equipped with an air-cooled argon ion laser emitting light at 488 nm and a second red diode laser emitting light at 635 nm. For sample injection with 1.2-ml sample tubes, these were inserted in mm polystyrene tubes (Falcon catalog no. 2052, Becton Dickinson) that were used as adapters for the sample injection port. For data acquisition, FITC emission was measured at FL1 (band pass filter, 530/30 nm), PE emission was measured at FL2 (band pass filter, 585/42 nm), PE/Cy5 emission was measured at FL3 (long pass filter, 670 nm), and TO-PRO-3 emission was measured at FL4 (band pass filter, 661/16 nm). Data on scatter parameters were acquired in linear mode and data on fluorescence channels in log mode. Electronic compensation was used to remove spectral overlap. For each sample, a region for CD4 cells was defined. At least 10,000 CD4 cells were acquired and stored in list mode using Cellquest software (Becton Dickinson). Data were analyzed with WinMDI version 2.8 IMMUNOPHENOTYPING AND APOPTOSIS IN RAT BLOOD 267 software (J. Trotter, Scripps Research Institute, La Jolla, CA). RESULTS Reagents for Rat Whole Blood Lysis Interference with staining. Three red blood cell lysing reagents were tested for potential interference with antibody labeling and for compatibility with rat leukocytes. Figure 1 shows contour plots for forward scatter (FSC) versus CD4-PE of the whole blood lysis procedures evaluated in heparanized rat blood. When lysing whole blood samples with CAL-lyse (Fig.1A) or Erythrolyse (Fig.1B), it was possible to distinguish monocytes from lymphocytes based on FSC and CD4 staining: CD4 lymphocytes exhibited low FSC values and CD4 bright staining, whereas monocytes caused medium FSC values and CD4 dim staining (6). In contrast, the treatment with FACSlysing markedly affected the staining of CD4-labeled cells (Fig.1C), with the presence of a new population of CD4 cells with the FSC characteristic of lymphocytes but weaker CD4 staining and the almost complete disappearance of monocytes. This effect was reduced, but not completely suppressed, by decreasing the concentration and incubation time with FACS-lysing (Fig.1D). Based on these results, only CAL-lyse and Erythrolyse were tested further for permeabilization of leukocytes in whole blood. Permeabilization properties. Apoptotic cells exhibit several modifications that can be detected by flow cytometry in a very simple way: reduced FSC and increased side scatter (SSC) due to cell shrinkage, as well as low DNA stainability resulting in the presence of a hypodiploid peak in the DNA histogram due to DNA fragmentation (8,13). In order to compare the permeabilization properties of CALlyse and Erythrolyse, samples of rat whole blood cultured with dexamethasone and stained for CD4 and DNA (TO- PRO-3) were treated with CAL-lyse or Erythrolyse. With both lysis reagents, it was possible to induce permeabilization of the leukocytes, as determined by DNA staining (Fig. 2C-E, G-I). Only Erythrolyse was able to distinguish between a typical diploid DNA peak and a hypodiploid peak (Fig. 2C). Further gating of the CD4 peak in cells of decreased size (R1, Fig. 2D) and in cells of normal size (R2, Fig. 2E) showed a direct correlation between hypodiploidy and cells of decreased size (presumably apoptotic cells). Based on these results, Erythrolyse was the lysing reagent of choice for further testing. Simultaneous DNA Staining and Immunophenotyping of Fresh Blood Immunophenotyping of heparanized rat blood was done by simultaneous staining of CD3, CD4, CD8, and DNA using Erythrolyse as the lysis reagent. After staining and red blood cell lysis, acquisition in the flow cytometer was done without the FSC threshold. Figure 3 illustrates the light scatter properties of a representative experiment and the DNA gating strategy used in the analysis. The DNA content of the cells was measured by the fluorescence intensity of TO-PRO-3. Two regions were established

4 268 DIAZ-ROMERO ET AL. FIG. 1. Comparison of light scattering and immunofluorescence signals of the different red blood lysing procedures. Rat whole blood was stained with a mab anti-cd4 conjugated to PE and erythrocytes were lysed according to the following protocols. A: CAL-lyse. B: Erythrolyse. C: FACS-lysing, 500 l for 10 min. D: FACS-lysing, 250 l for 5 min (see Materials and Methods). CD4 lymphocytes (FSC low, CD4 bright ) are indicated by arrows and CD4 monocytes (FSC medium, CD4 dim ) are indicated by arrowheads. A population with FSC of lymphocytes and CD4 dim is observed in C and D (dotted circles). based on DNA content: DNA low and DNA high (Fig. 3B). By gating on DNA high (Fig. 3D), the debris (events with very low values of FSC and SSC) present in the ungated FSC/ SSC contour plot (Fig. 3A) was excluded and nucleated cells (DNA labeled) were selected. It was possible to distinguish between two clusters of nucleated cells: a cluster with characteristic light scatter of granulocytes (FSC high and SSC high ) and a cluster with FSC from low to high and with SSC from low to medium for mononuclear cells (lymphocytes and monocytes). Not only events with very low values of FSC and SSC were present in the FSC/SSC contour plot gated on DNA low (Fig. 3C). There was also a cluster of events with apparent larger size. This population had an immunophenotype of monocytes: CD3-/CD4 /CD8 -. The presence of unpermeabilized leukocytes with larger FSC values after lysis with Erythrolyse

5 IMMUNOPHENOTYPING AND APOPTOSIS IN RAT BLOOD 269 FIG. 2. Comparison of CAL-lyse and Erythrolyse for permeabilization of leukocytes. Samples of whole blood incubated for 24 h with 1 M dexamethasone (DEX) were stained with anti-cd4 mab conjugated to PE and with TO-PRO-3 and treated with CAL-lyse or Erythrolyse (see Materials and Methods). An electronic gate was set on CD4 cells (A). Contour plots gated for CD4 of FSC versus SSC of Erythrolyse (B) or CAL-lyse (F) treated whole blood were used to define two regions: R1 and R2. TO- PRO-3 (DNA content) histograms gated for CD4 (C,G), CD4 and R1 (D,H), and CD4 and R2 (E,I) are shown for Erythrolyse-treated samples (C-E) and for CAL-lyse treated samples (G-I). The gating strategy is indicated in each contour plot and histogram. was described previously (4). In order to confirm that this population comprised unpermeabilized monocytes, we repeated the staining using PBMC, treated with or without Erythrolyse (Fig. 4). The FSC/SSC pattern of PBMC treated with Erythrolyse gated on DNA low (Fig. 4C) was very similar to the pattern obtained previously with whole blood (Fig. 3C), including the presence of a CD3-/CD4 /CD8 - population. A similar FSC/SSC pattern was obtained with PBMC not treated with Erythrolyse, gated simultaneously on DNA low (Fig. 4G) and on CD3-, CD4, and CD8 - (Fig. 4I). From this experiment, we concluded that permeabilization induced by Erythrolyse is correlated with an apparent decrease in FCS (cellular size), probably due to leakage of cytoplasmic content. FIG. 3. Light scatter properties and DNA staining of fresh blood treated with Erythrolyse. Rat whole blood was stained with CD3-FITC, CD8 -PE, CD4-PE/Cy5, and TO-PRO-3 and treated with Erythrolyse. Regions for DNA content were established based on TO-PRO-3 staining (B). Light scatter properties before (A) and after gating for low (C) or high DNA staining (D) are shown. The expression of CD3, CD4, and CD8 for a region of FSC high in C (dotted line) is indicated. Note that no FSC threshold was used during acquisition. Detection of Apoptotic Cells With Defined Immunophenotype in Whole Blood The proposed method to measure apoptosis in whole blood consists of a modification of Nicoletti s technique (8) and is based on multiparametric FACS analysis of light scatter, DNA content, as well as on the CD3, CD4, and CD8 expression of peripheral blood T cells. Representative light scatter patterns and DNA content, as measured by TO-PRO-3 labeling, for fresh rat blood T cells are shown in Figure 5. By gating on CD3, CD4, and CD8, it is possible to analyze separately CD4 and CD8 T cells. In fresh blood T cells, changes associated with apoptosis, such as decreased FSC, increased SSC, and the presence of a hypodiploid peak, are not observed. In order to induce apoptosis, rat whole blood samples were cultured for 24 h in DMEM medium supplemented with or without dexamethasone. Light scatter patterns and DNA content for these samples are shown in Figure 6. Figure 6A,D shows ungated FCS/SSC cytograms and DNA

6 270 DIAZ-ROMERO ET AL. FIG. 4. The population with immunophenotype of monocytes with DNA low staining are nonpermeabilized cells during Erythrolyse treatment. PBMC from rat were stained as in Figure 3 and treated with (A-D) or without Erythrolyse (E-I). Contour plots of FSC versus SSC without gating (A,E), DNA gating strategy (B,F), and contour plots of FSC versus SSC after gating for DNA low (C,G) and DNA high (D,H) are displayed. The DNA low /CD3-/CD4 /CD8 -/FSC high population in Erythrolyse-treated PBMC (C, dotted line) corresponds to the same light scatter characteristics of a population gated on CD3-/ CD4 /CD8 - in nontreated (nonpermeabilized) PBMC (G, dashed line, and I). histograms for untreated and dexamethasone-treated leukocytes. Dexamethaxone increased the presence of a subpopulation smaller and more granular, as determined indirectly by light scatter properties, both in CD4 (Fig. 6E) and CD8 T cells (Fig. 6F). This subpopulation was not as pronounced in control blood cultures without dexamethasone (Fig. 6B,C). The same samples analyzed for DNA content presented the typical emission of a hypodiploid peak in the case of dexamethasone-treated blood, as expected from apoptotic cells (Fig. 6E,F). No hypodiploid peak was observed in T cells not treated with dexamethasone (Fig. 6B,C). By combining simultaneously size and DNA content (Fig. 7), apoptotic T cells were distinguishable from normal T cells because of their reduced cell size (FSC low ) and DNA content (low TO-PRO-3 labeling). Based on these properties, regions were defined for apoptotic (APO) and nonapoptotic (No-APO) T cells. Apoptotic cells from dexamethasone-treated blood isolated by magnetic enrichment using annexin-v microbeads (Fig. 7, lower panel) exhibit FSC and TO-PRO-3 labeling patterns similar to the cells present in the APO region. Figure 8A shows the quantitative results of spontaneous (no-treatment) and dexamethasone-induced apoptosis of CD4 and CD8 T cells after 24 h of incubation at 37 C of whole blood. Data of the analyses were expressed as percentage of apoptotic cells: 100 [APO/(APO No APO)]. The spontaneous apoptosis percentage was 12.6

7 IMMUNOPHENOTYPING AND APOPTOSIS IN RAT BLOOD 271 FIG. 5. No apoptosis changes (decreased FSC, increased SSC, and hypodiploid peak) are observed in fresh blood T cells. Rat whole blood was stained and treated with Erythrolyse as in Figure 3. Contour plots of FSC versus SSC (left) and DNA staining histograms (right) are displayed gated for CD3 /CD4 cells (upper panels) and for CD3 /CD8 cells (lower panels). 2.7 for CD4 and for CD8 T cells and the dexamethasone-induced apoptosis percentage was for CD4 and for CD8 T cells. To investigate whether a caspase inhibitor would prevent dexamethasone-induced apoptosis, whole blood was treated with dexamethasone in the presence or absence of the peptide fluormethylketone inhibitor, Z-VAD-fmk, a known inhibitor of glucocorticoid-induced apoptosis (10,11). Figure 8B shows a significant but limited effect of Z-VAD-fmk in preventing dexamethasone-induced apoptosis in CD4 and CD8 T cells. It has been reported that in order to effectively prevent dexamethasone-induced apoptosis, Z-VAD-fmk must be added at periodic intervals throughout the 24-h incubation period (10). The cumulative concentration of DMSO obtained using this approach caused cellular toxicity, thus masking the effect of the inhibitor (data not shown). DISCUSSION We have established a method for the simultaneous analysis of DNA content and immunophenotyping in rats by using small volumes of lysed blood and four-color flow cytometry. The rat is the species of choice for several autoimmune disease models and experimental organ transplantation. In rodent models, access to cells of the immune system is usually achieved by terminal removal of the spleen or local lymph nodes. Therefore, long-term studies are limited. These problems could be reduced to

8 272 DIAZ-ROMERO ET AL. FIG. 6. Light scatter and DNA staining of apoptotic T cells. Whole blood was incubated for 24 h at 37 C with (D-F) or without 1 M dexamethasone (A-C), stained for CD3, CD4, CD8, and DNA content, and treated with Erythrolyse. Contour plots of FSC versus SSC (left) and DNA staining histograms (right) are displayed without gating (A,D), gating for CD3 /CD4 cells (B,E), and gating for CD3 /CD8 cells (C,F). Dexamethaxone increases a FSC low population with increased SSC and induces the presence of a sub-g1 peak (arrow), both in CD3 /CD4 and in CD3 /CD8 cells. some extent by the development of techniques to study peripheral blood lymphocytes in rodent models using volumes of blood small enough for repeated sampling (5). Our method uses 20 l of anticoagulated whole blood per sample, a volume that could allow long-term studies of rats and mice. The direct leukocyte staining method with lysis of red blood cells has been adopted widely in clinical practice (1,4). Because a density gradient separation step for lymphocytes or granulocytes is avoided, the technique is faster and less prone to loss of populations within the separation medium. However, whole blood labeling techniques in rodents are not as well established as in humans and the lysis procedures for these species need to be validated (7,14). In this study, we compared the effects on rat lymphocytes of three lysis reagents commonly used for human blood. Several reports have suggested the efficient use of FACS-lysing with rat peripheral blood (2,5,14). Our results have shown that FACS-lysing interferes markedly with surface staining. The other two tested reagents, CALlyse and Erythrolyse, lysed erythrocytes and permeabilized leukocytes in rat blood, without interfering significantly with surface staining. Unexpectedly, a remarkable property of Erythrolyse was observed during these experi-

9 IMMUNOPHENOTYPING AND APOPTOSIS IN RAT BLOOD 273 FIG. 7. Combination of light scatter and DNA content analysis for the detection of apoptotic T cells. Contour plots of FSC versus TO-PRO-3 (DNA content) of the samples described in Figure 6. Apoptotic cells (dashed line, APO) have decreased FSC (size) and DNA content, compared with nonapoptotic cells (dotted line, No-APO). The lower panel (dashed line) shows contour plots of whole blood treated with dexamethaxone after magnetic isolation with annexin-v coated microbeads (see Materials and Methods). ments. In combination with a DNA dye (TO-PRO-3; 15,16), Erythrolyse treatment allowed discrimination between two populations with different DNA content in rat blood incubated with an apoptotic stimulus (dexamethasone). Due to these specific characteristics, Erythrolyse was our choice to develop an apoptosis detection assay based on simultaneous DNA staining and surface antigens in rat whole blood. First, we evaluated the effectiveness of Erythrolyse for simultaneous DNA staining and immunophenotyping of fresh rat blood. We used a combination of three surface markers (CD3, CD4, and CD8 ) and one DNA dye (TO- PRO-3). The obvious advantage of this methodology is the possibility to select only nucleated (DNA stained) cells for analysis. We found, in line with a previous report (4), that a proportion of monocytes is not permeabilized by Erythrolyse, and thus the nucleus of these cells is unstained. However, this population can be differentiated from noncellular unstained debris by light scatter (Fig. 3). The reason for the presence of unpermeabilized monocytes

10 274 DIAZ-ROMERO ET AL. FIG. 8. A: Percentage of apoptotic CD4 and CD8 T cells in whole blood after 24 h of incubation without (Medium) or with 1 M dexamethaxone (DEX). The data represent the mean response obtained from duplicate samples of 10 animals. Error bars represent the SD. B: Percentage of apoptotic CD4 and CD8 T-cells in whole blood after 24 h of incubation with 1 M dexamethaxone (DEX) in the presence of 200 M Z-VAD-fmk or DMSO (vehicle control). The data represent the mean response obtained from quadruplicate samples of a pool of blood from five animals. Error bars represent the SD. Values with Z-VAD-fmk were compared for statistical significance with controls by student t-test analysis (*P 0.001). both in blood and in PBMC treated with Erythrolyse is unclear. One appealing possibility is that this population has characteristic cell membrane properties, different from other leukocytes, as in the case of basophils. Basophils are distinguished from other leukocytes because of their resistance to cytoplasmic stripping when exposed to a lytic agent at low ph, as applied for basophil counting in some automated cell counters (17). By using this DNA surface markers technology, we wanted to develop an apoptotic assay in whole blood. Apoptosis was described as a morphological phenomenon many years ago. However, only recently has its pivotal role in the regulation of cell number in a wide variety of circumstances been recognized (18). In contrast to necrosis, apoptosis is an active, signal-dependent cellular suicide that is associated with characteristic morphological and biochemical features (13). Apoptosis plays an important role in the interplay between the graft and the recipient after allograft transplantation. Some of the immunosuppressive drugs currently in clinical use might exert their activity, at least in part, through effects on apoptotic pathways (19). A more precise understanding of the rules governing the induction and prevention of apoptosis may form the basis for further development of new strategies for immunosuppressive therapy and tolerance induction (20,21). As the majority of the events that characterize apoptotic death can be revealed easily by multiparameter flow cytometry, a lot of methods have been proposed to analyze and quantify the apoptotic process by this type of analysis (18,22,23). Flow cytometry presents many advantages. It requires only a small number of cells, fluorescent probes are now available for the analysis of many aspects of cell proliferation, differentiation, and activation, and it also permits the simultaneous investigation of cells of different phenotypes within the same sample. Among the more common methods for measuring apoptosis by flow cytometry are those based on cell size and granularity and on DNA content analysis (8,13,24,25). Loss of volume and an increase in density are characteristic features accompanying apoptosis and they are detected easily by flow cytometry (25). During apoptosis, there is an initial increase in SSC (probably due to the chromatin condensation) with a reduction in FSC (due to the cell shrinkage). Simplicity, its low cost, and the possibility of its combination with surface immunofluorescence to identify the phenotype of apoptotic cells in a heterogeneous population are the main advantages of this analysis. However, this is a nonspecific method and it requires accurate controls for eliminating mechanically broken cells and nuclear debris. DNA analysis is another common method used to detect apoptosis by flow cytometry. During apoptosis, as the DNA of the cell begins to break up, small molecular weight fragments are released from the nucleus. Following alcohol fixation, these fragments may be eluted by washing in phosphate/citrate buffer. Therefore, following DNA staining, these cells will show a reduced fluorescence and show a sub-g1 peak (8). The appearance of this peak is a specific marker of apoptosis: necrosis induced by metabolic poisons or lysis induced by complement did not induce any sub-g1 peak in the DNA. The sub-g1 peak method is simple, quantitative, and applicable to all cell types, but has a main limitation: it is not adequate for simultaneous immunophenotyping due to the ethanol-related alterations of the lipid membrane and protein structure (24). The method we propose here for apoptosis detection combines simultaneous DNA analysis, light scatter analysis, and immunophenotyping in the same assay. Although several methods have been developed for the simultaneous examination of apoptosis and surface phenotype of lymphocytes (24,26,27), the main advantage of our method is the use of unfractionated whole blood as the source of cells for testing susceptibility to apoptosis. Usually, the first step in performing a cytometric evaluation of apoptosis is the separation of PBMC through a density gradient (22). This procedure allows the removal of granulocytes and red blood cells, but has several drawbacks. It

11 is time- consuming, there is selective loss of cell subpopulations, cells are activated by mechanical stimulation, and it is not suitable for small blood volumes. The use of whole blood minimizes cell manipulation and saves time. Also, by using the proper combination of surface markers, it allows simultaneous testing of the susceptibility to apoptosis of different subsets of leukocytes present in a single sample. Finally, the present study supports the existence of glucocorticoid-sensitive mature T lymphocytes. Immature thymocytes are particularly sensitive to apoptotic death induced by glucocorticoids (25), but contradictory results have been published in relation to the glucocorticoidinduced apoptosis of mature T lymphocytes. One study using human mature T cells reported that freshly isolated, by contrast to short-term cultured, T cells were not sensitive to glucocorticoid-induced apoptosis (28). On the other hand, apoptosis induced by glucocorticoids in vitro in freshly isolated human (13) and mouse (12,24) lymphocytes has been reported. The higher sensitivity for glucocorticoid-induced apoptosis of CD8 T cells in relation to CD4 T cells observed in our study has been described previously in the context of isolated human T cells (28). ACKNOWLEDGMENTS We are grateful to Dr. Kurt Mueller for his critical reading of the manuscript and for all his helpful comments and suggestions. We thank Ms. Valerie Caballero for her helpful assistance with blood sampling. LITERATURE CITED 1. Francis C, Connelly MC. Rapid single-step method for flow cytometric detection of surface and intracellular antigens using whole blood. Cytometry 1996;25: Morris DL, Komocsar WJ. Immunophenotyping analysis of peripheral blood, splenic, and thymic lymphocytes in male and female rats. Pharmacol Toxicol Methods 1997;37: Mandy FF, Bergeron M, Minkus T. Evolution of leukocyte immunophenotyping as influenced by the HIV/AIDS pandemic: a short history of the development of gating strategies for CD4 T-cell enumeration. Cytometry 1997;30: Tiirikainen MI. Evaluation of red blood cell lysing solutions for the detection of intracellular antigens by flow cytometry. Cytometry 1995;20: Peakman M, Tredger JM, Davies ET, Davenport M, Dunne JB, Williams R, Vergani D. Analysis of peripheral blood mononuclear cells in rodents by three-colour flow cytometry using a small-volume lysed whole blood technique. J Immunol Methods 1993;158: Westermann J, Walter S, Nagahori Y, Heerwagen C, Miyasaka M, Pabst R. Blood leucocyte subsets of the rat: expression of adhesion molecules and localization within high endothelial venules. Scand J Immunol 1996;43: Webster GA, Pockley AG. Effect of red cell lysis protocols on the expression of rat peripheral blood lymphocyte subset and activation antigens. J Immunol Methods 1993;163: Nicoletti I, Migliorati G, Pagliacci MC, Grignani F, Riccardi C. A rapid and simple method for measuring thymocyte apoptosis by propidium IMMUNOPHENOTYPING AND APOPTOSIS IN RAT BLOOD 275 iodide staining and flow cytometry. J Immunol Methods 1991;139: Zeller W, Weber H, Panoussis B, Burge T, Bergmann R. Refinement of blood sampling from the sublingual vein of rats. Lab Anim 1998;32: McColl KS, He H, Zhong H, Whitacre CM, Berger NA, Distelhorst CW. Apoptosis induction by the glucocorticoid hormone dexamethasone and the calcium-atpase inhibitor thapsigargin involves Bc1-2 regulated caspase activation. Mol Cell Endocrinol 1998;139: Miyashita T, Nagao K, Krajewski S, Salvesen GS, Reed JC, Inoue T, Yamada M. Investigation of glucocorticoid-induced apoptotic pathway: processing of caspase-6 but not caspase-3. Cell Death Differ 1998;5: Migliorati G, Nicoletti I, Nocentini G, Pagliacci MC, Riccardi C. Dexamethasone and interleukins modulate apoptosis of murine thymocytes and peripheral T-lymphocytes. Pharmacol Res 1994;30: Matteucci C, Grelli S, De Smaele E, Fontana C, Mastino A. Identification of nuclei from apoptotic, necrotic, and viable lymphoid cells by using multiparameter flow cytometry. Cytometry 1999;35: Pelegri C, Rodriguez-Palmero M, Morante MP, Comas J, Castell M, Franch A. Comparison of four lymphocyte isolation methods applied to rodent T cell subpopulations and B cells. J Immunol Methods 1995;187: Van Hooijdonk CA, Glade CP, Van Erp PE. TO-PRO-3 iodide: a novel HeNe laser-excitable DNA stain as an alternative for propidium iodide in multiparameter flow cytometry. Cytometry 1994;17: Glade CP, Van Erp PE, Boezeman JB, van de Kerkhof PC. Multiparameter flow cytometry as a tool to evaluate antipsoriatic therapy. Br J Dermatol 1997;137: Kabata J, Gratwohl A, Tichelli A, John L, Speck B. Hematologic values of New Zealand white rabbits determined by automated flow cytometry. Lab Anim Sci 1991;41: Darzynkiewicz Z, Juan G, Li X, Gorczyca W, Murakami T, Traganos F. Cytometry in cell necrobiology: analysis of apoptosis and accidental cell death (necrosis). Cytometry 1997;27: Kabelitz D. Apoptosis, graft rejection, and transplantation tolerance. Transplantation 1998;65: Wells AD, Li XC, Li Y, Walsh MC, Zheng XX, Wu Z, Nunez G, Tang A, Sayegh M, Hancock WW, Strom TB, Turka LA. Requirement for T-cell apoptosis in the induction of peripheral transplantation tolerance. Nat Med 1999;5: Li Y, Li XC, Zheng XX, Wells AD, Turka LA, Strom TB. Blocking both signal 1 and signal 2 of T-cell activation prevents apoptosis of alloreactive T cells and induction of peripheral allograft tolerance. Nat Med 1999;5: Lecoeur H, Ledru E, Prevost MC, Gougeon ML. Strategies for phenotyping apoptotic peripheral human lymphocytes comparing ISNT, annexin-v and 7-AAD cytofluorometric staining methods. J Immunol Methods 1997;209: Ormerod MG. The study of apoptotic cells by flow cytometry. Leukemia 1998;12: Douglas RS, Tarshis AD, Pletcher CHJ, Nowell PC, Moore JS. A simplified method for the coordinate examination of apoptosis and surface phenotype of murine lymphocytes. J Immunol Methods 1995; 188: Swat W, Ignatowicz L, Kisielow P. Detection of apoptosis of immature CD4 8 thymocytes by flow cytometry. J Immunol Methods 1991;137: Hardin JA, Sherr DH, DeMaria M, Lopez PA. A simple fluorescence method for surface antigen phenotyping of lymphocytes undergoing DNA fragmentation. J Immunol Methods 1992;154: Lecoeur H, Ledru E, Gougeon ML. A cytofluorometric method for the simultaneous detection of both intracellular and surface antigens of apoptotic peripheral lymphocytes. J Immunol Methods 1998;217: Brunetti M, Martelli N, Colasante A, Piantelli M, Musiani P, Aiello FB. Spontaneous and glucocorticoid-induced apoptosis in human mature T lymphocytes. Blood 1995;86: