An improved flow cytometric assay for the determination of cytotoxic T lymphocyte activity

Journal of Immunological Methods 259 Ž2002. 159–169 www.elsevier.comrlocaterjim

An improved flow cytometric assay for the determination of cytotoxic T lymphocyte activity Karin Fischer, Reinhard Andreesen, Andreas Mackensen ) Department of Hematologyr Oncology, UniÕersity of Regensburg, Franz-Josef-Strauss-Allee 11, D-93042 Regensburg, Germany Received 29 May 2001; received in revised form 25 July 2001; accepted 27 August 2001

Abstract The cytotoxic activity of T lymphocytes, natural killer and lymphokine-activated killer cells is usually tested by radioactive assays, which detect the release of cytoplasmic contents after plasma membrane disintegration of dying cells. In contrast to this indirect evaluation of cytotoxicity, we describe here an improved fluorescence assay that is based on the direct quantitative and qualitative flow cytometric analysis of cell damage at a single cell level. Target cells are stained with PKH-26, a lipophilic dye that stably integrates into the cell membrane and permits distinction between target and effector cells. After 3 h of in vitro incubation, costaining with AnnexinV-FITC Žann-FITC. and propidium iodide ŽPI. permitted discrimination between vital, early apoptotic and necrotic cells. Data analysis is performed first by gating on PKH-26-positive target cells followed by the analysis of ann-FITC- and PI-positive subpopulations. The percentage of cytotoxicity in the PKH-26-gated cell population is calculated by subtracting non-specific ann-FITC- or PI-positive target cells, measured in appropriate controls without effector cells. Membrane staining of target cells such as primary melanoma cells or leukemic blasts revealed high and stable loading of PKH-26 without altering the viability or the immunogenicity of the cells. Using in vitro-generated antigen-specific cytotoxic T lymphocytes ŽCTL., we could demonstrate that this flow cytometric assay is sensitive and correlates well with the standard 51 Cr release assay. In conclusion, the improved fluorescence assay is a simple and highly reproducible procedure for evaluating the specific cytotoxicity of T cells. q 2002 Elsevier Science B.V. All rights reserved. Keywords: Cell-mediated cytotoxicity; Flow cytometry; PKH-26; Cytotoxic T lymphocytes; Apoptosis

1. Introduction Cellular cytotoxicity can be mediated by two major contact-dependent mechanisms. The first includes the granule exocytosis pathway which is predominantly used by CD8q cytotoxic T lymphocytes

)

Corresponding author. Tel.: q49-941-944-5580; fax: q49941-944-5502. E-mail address: [email protected] ŽA. Mackensen..

ŽCTL., natural killer or lymphokine-activated killer cells. After conjugation with target cells, cytotoxic granules migrate to the site of contact, and together with perforin-created pores, induce target cell death ŽHenkart, 1994.. The second pathway is based on receptor–ligand interactions Že.g. FasrFas-ligand. and is known to be the major mechanism by which CD4q T helper lymphocytes kill cells expressing the appropriate receptor ŽRouvier et al., 1993.. Effector cell-mediated cytolysis is usually determined by methods based on the release of different tracers from lysed target cells. In general, these

0022-1759r02r$ - see front matter q 2002 Elsevier Science B.V. All rights reserved. PII: S 0 0 2 2 - 1 7 5 9 Ž 0 1 . 0 0 5 0 7 - 5

160

K. Fischer et al.r Journal of Immunological Methods 259 (2002) 159–169

markers are compounds containing radioactive isotopes such as 51 Chromium Ž51 Cr., 75 Selenium Ž75 Se. or Tritium Ž3 H. ŽCerottini and Brunner, 1974; Mantovani et al., 1979.. The 51 Cr release assay is the most widely used method ŽBrunner et al., 1968. for measuring CTL responses in vitro due to its reliability and simplicity. Nevertheless, as the usage of a radioactive material leads to various potential hazards, several groups have searched for easy and reliable alternatives. Several non-radioactive assays based on labelling with markers such as Europium ŽEu3q . ŽPatel and Boyd, 1995., bisbenzamide dye ŽToka et al., 1996., Calcein-AM Žacetoxymethyl ester of calcein. ŽLichtenfels et al., 1994. have been developed. However, indirect measurement of the supernatant frequently fails to offer sufficient accuracy. More recently, novel assays have been developed using fluorescent dyes ŽMattis et al., 1997; Johann et al., 1995; Flieger et al., 1995.. The assessment of cell damage by flow cytometry aims to provide a more exact characterization of the death pathway via detection of the percentage of apoptotic and dead cells ŽAubry et al., 1999.. One of the characteristic processes of the apoptosis pathway is a change in the plasma membrane architecture. During early apoptosis, typical membrane compounds, such as phosphatidyl serine ŽPS. molecules, are redirected from the inner to the outer leaflet of the cell membrane without loss of membrane integrity ŽMartin et al., 1995; Vermes et al., 1995.. AnnexinV-FITC ŽannFITC., a molecule with high affinity for PS, can be used to label cells in the early apoptotic state, while propidium iodide ŽPI. indicates late apoptosis or cell death. We present here the development of a threecolor flow cytometric assay that uses the lipophilic membrane dye PKH-26 and permits a fine characterization of specific effector cell cytotoxicity. PKH-26 has previously been used both in proliferation analysis, cell tracking studies ŽYoung and Hay, 1995; Rosenblatt-Velin et al., 1997. and flow cytometric analysis ŽYamamura et al., 1995; Flieger et al., 1999; Sheehy et al., 2001. and has been found to be a stable tracer. While a recently described three-color flow cytometric assay ŽAubry et al., 1999. proposed the staining of effector cells with PKH-26 for further analysis of unstained target cells, we decided to label the target cells. Labelling of effector cells instead of

the target cells may lead to an accumulation of unlabelled effector cells in the target cell gate at high effector to target cell Ž ErT . ratios. We demonstrate here that the loading of target cells with PKH-26 is stable and does not impair the viability and expression of surface molecules such as adhesion and major histocompatibilty ŽMHC. antigens. The PHK-26 assay correlates well with the conventional 51 Cr release assay and provides additional information on the pathway of target cell death induced by CTL.

2. Materials and methods 2.1. Media and monoclonal antibodies Cells were cultured in RPMI 1640 supplemented with 200 mmolrl L-glutamine, 50 mmolrl b-mercaptoethanol, 100 mmolrl sodium pyruvate, minimum essential medium ŽMEM. vitamins, 40 mgrml streptomycin and 40 Urml penicillin Žstandard medium MX .. For phenotypic and functional analysis of PKH26-labelled target cells, the following monoclonal antibodies ŽmAbs. were used: CD3, CD8, CD33, CD34, CD54, CD58, HLA-DR Žall from Becton Dickinson, Mountain View, CA.. Anti-HMW-MAA ŽEP-2. mAb, kindly provided by Dr. S. Ferrone ŽDepartment of Medicine, Roswell Park Cancer Institute, Buffalo, NY., detects the high molecular weight melanoma-associated antigen expressed on a large percentage of human malignant melanoma cells ŽGiacomini et al., 1987.. Anti-HLA-ABC ŽW6r32. mAb recognizes a monomorphic determinant of HLA class I gene products ŽDako, Glostrup, Denmark.. Anti-Mel-1 and anti-Mel-2 mAbs detect pigmentation-associated glycoproteins expressed by melanoma cells ŽSignet Laboratories, Dedham, MA.. 2.2. Target cells Tumor cell lines and T2 cells were maintained in MX medium supplemented with 10% fetal calf serum ŽFCS. ŽPAN Systems, Aidenbach, Germany.. Melanoma cell lines MeR190 and MeI493 were established from surgically excised melanoma metastasis. Expression of HLA-A2 and Melan-A was as-

K. Fischer et al.r Journal of Immunological Methods 259 (2002) 159–169

sessed by FACS analysis using an anti-HLA-A2specifc mAb ŽBB7.2, ATCC, Rockville, MD. and by RT-PCR, respectively, as described previously ŽMackensen et al., 2000.. Primary leukemic blasts from AML patients were obtained after ficoll separation and cryopreserved. T2 cells are HLA-A) 0201 human lymphoid cells that are defective in the antigen processing, but effectively present exogenously supplied peptides ŽSalter et al., 1985.. T2 cells were loaded with 30 mgrml of peptide and incubated overnight at 37 8C in serum-free MX medium. 2.3. Effector cells Generation and expansion of antigen-specific T cell lines have been described previously ŽOelke et al., 2000.. Briefly, CD8q T lymphocytes were enriched from PBMC by depletion of CD4q, CD11bq, CD16q, CD19q, CD20q and CD56q cells with magnetic cell sorting using a midiMACS device ŽMiltenyi Biotec, Bergisch Gladbach, Germany.. The resulting population consisted of ) 90% CD8q T cells and was used as a responder cell population. Autologous monocyte-derived DC used as stimulator cells were pulsed for 2 h at 37 8C with 30 mgrml of a modified Melan-A 26 – 35 peptide ŽELAGIGILTV. and human b 2-microglobulin Ž10 mgrml. in serumfree MX medium. The 10 4 responder cellsrwell and 5 = 10 3 peptide-pulsed autologous DCrwell were cocultured in 96-well round-bottom plates in 200 ml MX medium per well, supplemented with 5% autologous plasma and 3% T cell growth factor ŽTCGF.. Medium and TCGF were replenished twice a week. On day 7, T cells were harvested, counted, and replated at 10 4 T cellsrwell together with 5 = 10 3 peptide-pulsed autologous DCrwell in complete medium supplemented with 3% TCGF. Subsequent restimulations with peptide-pulsed DC were performed once a week, and a total of three to four stimulation cycles were conducted before functional analysis. 2.4. Fluorescence labelling Various target cells were transferred to polystyrene tubes and washed twice with serum-free medium before staining. The cells were then resuspended in a loading buffer Žan aqueous, osmolarity-regulating solution containing no Ca2q or other physiological salts; Sigma, St. Louis, MO. and incubated for 40

161

min with freshly prepared 2 mM PKH-26 ŽSigma. at room temperature. The staining reaction was stopped by the addition of 500 ml human serum ŽPAN Biotech, Aidenbach, Germany. and 30-s incubation at RT. After centrifugation, the cell pellet was transferred to a fresh 50-ml tube ŽFalcon, France. and washed twice with 50 ml RPMI containing 10% human serum. Labelled target cells were seeded into 96 well-V-bottom plates ŽCostar, Corning, NY. and incubated with the effector cells at different ErT ratios. The dye stock is dissolved in ethanol and can be stored and protected from light at room temperature. It should be kept tightly capped to prevent evaporation. 2.5. Staining with ann-FITC and PI After coincubation of PKH-26-labelled target cells together with effector cells, the total cell population was harvested from the microtiter plates, collected into polystyrene tubes ŽFalcon., washed twice in high calcium-binding buffer ŽBender Med Systems, Vienna, Austria. and resuspended in 100 ml binding buffer. Ann-FITC staining was performed with 5 ml ann-FITC ŽBD PharMingen, USA. and an incubation time of 10 min at room temperature in the dark. Immediately before flow cytometric analysis 0.1 mg PI ŽCalbiochem, Germany. per 100 ml, the binding buffer was added. 2.6. FACS analysis FACS analysis was carried out on a FACScan Cytometer ŽBecton Dickinson, Immunocytometry Systems, San Jose, ´ USA.. By excitation at 488 nm with an argon laser, the emission of three fluorochromes was recorded through specific band pass filters: 530 nm for FITC ŽFL1., 585 nm for PKH-26 ŽFL2. and ) 675 nm for PI ŽFL3.. For compensation, unstained and PKH-26-stained cells should be prepared. The small amount of spontaneously apoptotic cells in the cell suspension was usually sufficient for the compensation. Since ann-FITC is known to increase non-specific fluorescence, compensation was carried out in the presence of ann-FITC. Compensation between FL1 and FL2: vial 1: unlabelled target cells stained with annFITC Žcompensation of specific ann-FITC-positive cells.

K. Fischer et al.r Journal of Immunological Methods 259 (2002) 159–169

162

vial 2: PKH-26-labelled cells with ann-FITC Žcompensation of ann-negative cells neglecting the small population of specific ann-FITC-positive cells. Compensation between FL2 and FL3: vial 3: unlabelled target cells stained with annFITC and PI Approximate values were: FL1–0.5% FL2; FL2– 78% FL1; FL2–47% FL3; FL3–17% FL2. Fluorescence analysis was quantified using the Cellquest software ŽBecton Dickinson.. 2.7.

51

with 2000 target cellsrwell. For spontaneous release, targets were plated without T cells in MX medium plus 5% autologous plasma. For maximum release, target cells in 100 ml of medium were plated with 100 ml of 0.15% Triton X-100 ŽSigma.. Triplicate wells were averaged and %specific cytotoxicity was calculated as follows:

Ž cpm sample y cpm spontaneous release . Ž cpm maximum release y cpm spontaneous release .

= 100

s %specific lysis.

Cr release assay

Conventional 51 Cr release assays were performed as previously described ŽMackensen et al., 1994.. Briefly, the cytotoxic activity of T cell lines was measured using triplicate cultures in V-bottomed plates. Target cells were labelled with 200 mCi for 1 h. Target cells were then washed twice and seeded into the plates at ErT ratios of 25:1, 5:1 and 1:1

2.8. Blocking of cytotoxicity with monoclonal antibodies The functional effects of the antibodies, either on effector cells wanti-CD3 and anti-CD8x or on target cells wanti-HLA-ABC and anti-HMW-MAAx, were tested by incubating each of them for 2 h at 37 8C

Fig. 1. Target cell-labelling kinetics. Panels represent either one color histogram ŽA and C. with the x-axis showing log scale red fluorescence intensity ŽFL2. and the y-axis showing relative cell number or a two-color dot plot ŽB.. ŽA. Leukemic blasts were incubated for 40 min at 37 8C with 2, 5 and 10 mM of PKH-26. Unstained cells were used as a control. ŽB. Overstaining of leukemic blasts resulting in an artificial light emission in FL1 Župper right quadrant. was observed after incubation with 10 mM PKH-26. ŽC. Retention of PKH-26 was evaluated with stained leukemic blasts Ž2 mM dye. at 1 and 72 h post staining. The results shown are representative of one experiment.

K. Fischer et al.r Journal of Immunological Methods 259 (2002) 159–169

before the assay at the predetermined saturating concentration. The percentage inhibition of lysis was calculated as: 1y

%specific lysis in mAb-treated wells %specific lysis in control wells

Table 1 Loading efficiency of primary tumor cells and cell lines stained with PKH-26 Cell type

= 100.

3. Results 3.1. Loading of target cells with PKH-26 Flow cytometric analysis of cell-mediated target cell damage is based on the stable loading of target cells combined with a low percentage of spontaneous release. We tested the membrane dye PKH-26 Žred fluorescence. for its ability to stain different primary tumor cells such as melanoma cells and leukemic blasts. As shown in Fig. 1A, the optimal PKH-26 concentration was found to be close to about 2 mM. Overstaining of the target cells leads to the accumulation of intensely stained single cells resulting in an artificial light emission in channel 1 ŽFig. 1B, upper right quadrant.. This would interfere with the real double-positive target cells and may lead to the false-positive results in the cytotoxic assay. The concentration of 2 mM used in the experiments permitted complete compensation in the absence of overstained cells. As shown in Table 1, the staining of different primary tumor cells with PKH-26 resulted in a loading efficiency of ) 95%. Interestingly, the loading of leukemic blasts, which generally shows a rather weak uptake of compounds from the external medium, revealed a high staining intensity with PKH-26. Labelling with PKH-26 remained stable in all vital cells tested for a minimum of 72 h ŽFig. 1C. without any loss of fluorescence intensity. In contrast, the labelling of primary leukemic blasts with 51 Cr revealed a high spontaneous release Ž) 25%. after 4 h of incubation Ždata not shown.. 3.2. Analysis of Õitality and surface marker expression in PKH-26-labelled target cells Next, we tested whether loading with PKH-26 may alter the viability or the cell surface phenotype.

163

Primary AML blasts Patient 1 Patient 2 Patient 3

% of vital cells stained with PKH-26 99.6 98.0 97.5

Primary melanoma cells MeI493 99.6 MeR190 99.9 Cell lines K562 THP1 T2

98.4 99.7 98.9

Primary tumor cells or cell lines were incubated for 40 min at 37 8C with freshly prepared 2 mM PKH-26. The loading efficiency was determined by flow cytometric analysis. Results represent the percentage of PKH-26-positive cells in a live gate. Unstained cells served as controls.

Apoptotic cell death of the target cells measured by ann-FITCrPI staining before and 4 h after loading with PKH-26 did not show any changes in the percentage of either ann-FITC Ždata not shown. or PI-positive cells ŽFig. 2.. We next assessed whether staining with PKH-26 could alter the expression of cell surface antigens required for functional effectorrtarget interactions. Target cells were stained before and 4 h after PKH-26 labelling with fluorescence-conjugated mAbs against CD54, CD58, HLAABC and HLA-DR. In addition, we used mAbs against CD33 and CD34, and Mel-1 and Mel-2, for leukemic blasts and melanoma cells, respectively. As shown in Fig. 2, staining with PKH-26 did not affect the expression of any of the antigens tested. 3.3. Data analysis and cytotoxicity quantification The calculation of specific effector cell mediated lysis is demonstrated in Fig. 3. In order to distinguish between target and effector cells, the gating of PKH-26-positive target cells was performed ŽFig. 3A.. Ann-FITC single-positive cells Žlower right quadrant. represented target cells in an early apoptotic state, ann-FITCrPI double-positive cells Župper

164 K. Fischer et al.r Journal of Immunological Methods 259 (2002) 159–169 Fig. 2. Effect of PKH-26 staining on the viability and expression of surface markers of target cells. Phenotypic analysis of target cells was performed after 4 h of incubation at 37 8C with or without the dye using flow cytometry. Target cells were stained with PI and with FITC-conjugated mAbs against HLA-ABC, HLA-DR, CD54, CD58, CD33rCD34 Žfor AML blasts. or Mel-1rMel-2 Žfor melanoma cells.. Results of one representative experiment show the phenotypic analysis of unlabelled and PKH-26-labelled primary AML blasts ŽA. or primary melanoma cells ŽB..

K. Fischer et al.r Journal of Immunological Methods 259 (2002) 159–169

165

Fig. 3. Analysis of specific cytotoxicity using the PKH-26 assay. ŽA. Target cells were selected by gating on the PKH-26-positive cell population and further analyzed for different subpopulations. ŽB. Dot plot of ann-FITCrPI flow cytometry of PKH-26 gated Melan-A-expressing target cells after coincubation with Melan-A-specific CTL at different ErT ratios from 0 to 20:1. The lower left quadrant of each panel shows the ann-FITCyrPIy viable cells. The upper left quadrant contains ann-FITCyrPIq non-viable necrotic cells. The right quadrants represent the apoptotic target cells, being either ann-FITCqrPIy Žlower. early or ann-FITCqrPIq Župper. late apoptotic cells. The results shown are representative of one experiment. ŽC. Percentage of different apoptotic target cell subpopulations calculated from ŽB. according to the formula indicated in the text.

right quadrant. included cells in the later apoptotic state, and PI single-positive cells Župper left quadrant. represented deadrnecrotic target cells Žsee Fig. 3B.. With escalating ErT ratios, increasing numbers of ann-FITC single-positive or ann-FITCrPI double-positive cells were observed ŽFig. 3B.. For the quantification of specific cytotoxicity, the population of non-specific positive cells Ždetermined in

the control sample without effector cells. in the corresponding quadrant has to be subtracted. The percentage of non-specific positive cells should not be subtracted at the end of the calculation because otherwise, 100% cytotoxicity, in principle, could not be reached. The calculation was carried out separately for each quadrant and is exemplified for the ann-FITC single-positive fraction.

166

K. Fischer et al.r Journal of Immunological Methods 259 (2002) 159–169

The proportion of corrected ann-FITC-positive events Žs ann-FITC-positive events sample y annFITC-positive events control . to corrected total events Žs total events sample y ann-FITC-positive events control . = 100 reflects the real percentage of annFITC-positive cytotoxicity. Note, if the number of acquired total events differs at various ErT ratios, the number of non-specific positive events has to be

determined separately for each ErT ratio in the formula: %specific ann-positive cells s

w ann-positive cells sample yannypositive cells control x w total events sample yannypositive cells control x

=100. As shown in Fig. 3C, the analysis of Melan-Aspecific CTL revealed a predominant population of early apoptotic ŽannqrPIy . target cells. Only small numbers of target cells were found to be in a late apoptotic state ŽannqrPIq .. We did not observe any accumulation of PI single-positive target cells after 4 h of incubation with specific effector cells. For comparison with the standard 51 Cr release assay, we calculated all ann-positive ŽannqrPIy together with annqrPIq . cells. 3.4. Comparison of specific target cell lysis between the PKH-26 and 51Cr release assays The PKH-26 flow cytometric assay was directly compared to the standard 51 Cr release assay. A representative example of T cell-mediated cytotoxicity is presented in Fig. 4A. Melan-A-specific CTL generated in vitro were tested simultaneously by both methods against HLA-A2q Melan-A-expressing target cells and HLA-A2q Melan-Ay control cells using ErT ratios of 25:1, 5:1 and 1:1. The two different assays showed comparable results at all ErT ratios. In addition, the cytotoxicity against the

Fig. 4. ŽA. Comparison of antigen-specific lysis determined by both the PKH-26 and 51 Cr release assays. In each assay, the same cultured effector cells and target cells were used at Er T ratios of 25:1, 5:1 and 1:1. Assays were performed simultaneously using a 4-h incubation period for the 51 Cr release assay and a 3.5-h incubation time for the PKH-26 assay. ŽB. Correlation of PKH-26 and 51 Cr release assays. The percentages of ann-FITC-positive cells detected by the PKH-26 assay are represented on the x-axis and correlated with the results obtained by the 51 Cr release assay represented on the y-axis. ŽC. Inhibition of specific cytotoxicity measured with the PKH-26 assay using monoclonal antibodies. Either effector cells Žanti-CD3 and anti-CD8. or target cells Žanti-HLA-ABC and anti-HMW-MAA. were preincubated with the respective antibodies for 30 min before being studied in the assay. Cytotoxicity was determined in a 3.5-h PKH-26 assay at an Er T ratio of 5:1, and the percentage of inhibition was calculated as indicated in the text.

K. Fischer et al.r Journal of Immunological Methods 259 (2002) 159–169

control cell line was negative in both assays. As shown in Fig. 4B, there was a good correlation between the results obtained with the PKH-26 assay and those obtained using the standard 51 Cr release assay Ž r 2 s 0.9777.. This indicates that T cellspecific immune responses can be detected by the PKH-26 assay. Of interest, results obtained with the PKH-26 assay revealed a slightly higher percentage of cytotoxicity as compared to the 51 Cr release assay. We next evaluated whether the addition of blocking mAbs could inhibit the percentage of specific cytotoxicity measured with the PKH-26 assay. As shown in Fig. 4C, blocking mAbs either against effector cells ŽCD3 and CD8. or target cells ŽHLAABC. were able to inhibit the cytotoxicity of MelanA-specific CTL. Incubation of target cells with antiHMW-MAA mAb recognizing an irrelevant epitope on melanoma cells had no effect. In an attempt to determine an appropriate targetr effector cell incubation time, the kinetics of the PKH-26 assay was evaluated. We analyzed the cytotoxic activity of Melan-A-specific CTL measured by PKH-26rann-FITC double-positive target cells at three different time points, 1.5, 2.5 and 3.5 h of coincubation. As shown in Fig. 5, the cytotoxic activity at higher ErT ratios Ž) 5:1. was found to be comparable from 1.5 to 3.5 h, reaching a plateau at a ratio of 20:1. In contrast, at lower ErT ratios

Fig. 5. PKH-26 assay kinetics. PKH-26-labelled Melan-A-expressing target cells were incubated with Melan-A antigen-specific CTL for either 1.5, 2.5 or 3.5 h at various effectorrtarget ratios prior to the PKH-26 assay. Data are presented as mean cytotoxicity"S.E. Ž ns 3. of Melan-A-specific CTL as determined by %ann-FITC-positive cells.

167

Ž0.2–5:1., the percentage of cytotoxicity correlated well with the time of incubation. At an ErT ratio of 1:1, the mean percentage of PKH-26qrann-FITCq cells was shown to be 14.9% after 1.5 h of coincubation compared to 45.4% after 3.5 h of coincubation. In our hands, 3 h of incubation was optimal for the evaluation of cell-specific cytotoxicity.

4. Discussion Analysis of the cytolytic activity of CTL is the center of monitoring antigen-specific immune responses. Traditionally, the cytotoxicity of T cells has been measured using 51 Cr release assays ŽBrunner et al., 1968.. Here, we have described a flow cytometry-based killing assay, which permits the fine evaluation of different stages of target cell death Žearly apoptosis and membrane damage. occurring during antigen-specific CTL responses. Target cells can be distinguished from effector cells using the red fluorescent membrane dye PKH-26. This dye has been shown to label efficiently the cell membrane of target cells without transferring to non-labelled cells in the same culture ŽSlezak and Horan, 1989.. Costaining with ann-FITC and PI permits the identification of target cells at different stages of cell death wearly apoptosis Žann-FITCqrPIy . andror necrosis ŽPIq.x . These results may provide further information on the death pathway of the target cells recognized by antigen-specific effector cells. Moreover, direct analysis of the effectorrtarget cell suspension in the PKH-26 assay instead of the supernatant in the 51 Cr release assay allows further phenotypic evaluation of effector cells Že.g. percentage of antigenspecific tetramer-positive CTL. or viable target cells Že.g. MHC expression.. Staining of several primary tumor cells with 2 mM PKH-26 revealed high and stable fluorescence intensity for a minimum of 3–4 days, permitting an exact analysis of the cytotoxic activity using both short-term as well as long-term assays. Of note, primary leukemic blasts showing inefficient labelling and high spontaneous release after incubation with 51 Cr ŽWeidmann et al., 1995. demonstrated high and stable staining of the dye without increased spontaneous apoptosis Žsee Table 1 and Fig. 2.. Analysis of the surface markers re-

168

K. Fischer et al.r Journal of Immunological Methods 259 (2002) 159–169

quired for functional effectorrtarget interactions did not show any significant changes after PKH-26 staining. The assessment of cytolytic activity against PKH26-gated target cells by ann-FITC versus PI revealed different populations: live cells Žann-FITCyrPIy ., early apoptotic cells Žann-FITCqrPIy ., late apoptoticrnecrotic cells Žann-FITCqrPIq andror annFITCyrPIq .. Since AnnexinV may bind to phosphatidylserine located either at the outer or the inner leaflet of the plasma membrane, the direct discrimination between late apoptotic and necrotic cells is critical. With increasing ErT ratios, we observed an accumulation of two different populations of PKH26-labelled target cells Žsee Fig. 3B.: the majority consisted of ann-FITCqrPIy early apoptotic cells while the minority were found to be ann-FITCqrPIq late apoptotic target cells. No significant increase in ann-FITCyrPIq necrotic cells was observed after 3.5 h of incubation. These data suggest that most of the target cells recognized by Melan-A-specific CTL die via the apoptosis pathway. Our results demonstrate that staining of both effector and target cells ŽFlieger et al., 1995. is not necessary for fine discrimination between the two cell types. Aubry et al. Ž1999. have recently shown that labelling of the effector instead of the target cells may represent an alternative method for the assessment of cellular cytotoxicity. However, the staining of effector cells may be critical since small percentages of cells may remain unstained. This is especially the case if high ErT ratios Ž50:1 and 100:1. are used since unstained effector cells are then counted as target cells and this may lead to false-positive or false-negative results. Our results demonstrate strong lysis of Melan-Aexpressing target cells by Melan-A-specific CTL lines, while maintaining excellent specificity and HLA restriction. A definitive comparison of the PKH-26 assay with the standard 51 Cr release assay was performed and showed a very good correlation between the two methods. However, the results obtained with the flow cytometric assay showed a higher percentage of cytotoxicity as compared to the 51 Cr release assay. The PKH-26 assay has several advantages over the 51 Cr release assay. First, the assay incubation time can be longer since the spontaneous release is

minimal. This could be of interest if the frequency of effector cells is low. Second, the assay may measure either apoptotic or necrotic cell death of the target cells. Perforins are known to create channels in the cell membrane through which granzymes released by T cells can induce the apoptotic cascade ŽJaneway, 1997.. Therefore, some targets die a necrotic death from perforin lysis, while other cells die a slow, apoptotic death. The 51 Cr release assay measures the 51 Cr released by necrotic cells, whereas apoptotic cells may not release 51 Cr. Finally, a striking advantage of the PKH-26 assay is the shortened amount of time needed to produce reliable results. The sensitivity of standard 51 Cr release assays requires an incubation time of 4 h. Since optimal killing is generally reached soon after the effector and target cells are brought into contact ŽHiserodt et al., 1982., a reliable measurement of killing should be detectable at an earlier time point. As shown in Fig. 5, optimal killing in the PKH-26 assay was observed within 1.5–2.5 h of incubation period and correlated well with the 4-h 51 Cr release assay. To obtain reproducible results, the preparative steps after coincubation Žharvesting and ann-FITCrPI labelling of cells. should be standardized. In conclusion, the PKH-26 assay is a quantitative method for the detection of cell-mediated cytotoxicity and does not use radioactive labelling. The reliability and reproducibility of this assay make it a useful alternative to the standard 51 Cr release assay, providing a greater insight into the mechanisms of cell-mediated cytolysis by different effector populations.

Acknowledgements We gratefully acknowledge the excellent technical assistance of Sandra Vogl. This work was supported by the Wilhelm Sander Foundation and by the German Jose´ Carreras Leukemia Foundation. References Aubry, J.-P., Blaecke, A., Lecoanet-Henchoz, S., Jeannin, P., Herbault, N., Caron, G., Moine, V., Bonnefoy, J.-Y., 1999. AnnexinV used for measuring apoptosis in the early events of cellular cytotoxicity. Cytometry 37, 197–204.

K. Fischer et al.r Journal of Immunological Methods 259 (2002) 159–169 Brunner, K.T., Mauel, J., Cerottini, J.C., Chapuis, B., 1968. Quantitative assay of the lytic action of immune lymphoid cells on 51 Cr-labelled allogeneic target cells in vitro: inhibition by isoantibody and drugs. Immunology 14, 181–196. Cerottini, J.-C., Brunner, K.T., 1974. Cell mediated cytotoxicity, allograft rejection and tumor immunity. Adv. Immunol. 18, 67–132. Flieger, D., Gruber, R., Schlimok, G., Reiter, C., Pantel, K., Riethmuller, G., 1995. A novel non-radioactive cellular cytotoxicity test based on the differential assessment of living and killed target and effector cells. J. Immunol. Methods 180, 1–13. Flieger, D., Spengler, U., Beier, I., Keinschmidt, R., Hoff, A., Varvenne, M., Sauerbruch, T., Schmidt-Wolf, I., 1999. Enhancement of antibody-dependent cellular cytotoxicity ŽADCC. by combination of cytokines. Hybridoma 18, 63–68. Giacomini, P., Segatto, O., Natali, P.G., 1987. Multiple epitope recognition: an approach to improved radioimmunodetection of tumor-associated antigens. Int. J. Cancer 39, 729–736. Henkart, P.A., 1994. Lymphocyte-mediated cytotoxicity: two pathways and multiple effector molecules. Immunity 1, 343–346. Hiserodt, J.C., Britvan, L.J., Targan, S.R., 1982. Characterization of the cytolytic reaction mechanism of the human natural killer ŽNK. lymphocyte: resolution into binding, programming, and killer cell-independent steps. J. Immunol. 129, 1782–1787. Janeway Jr., C.A., 1997. Immunobiology: The Immune System in Health and Disease. Garlan Publ., New York. Johann, S., Blumel, J.S., Lipp, M., Forster, R., 1995. A versatile flow cytometry-based assay for the determination of short- and long-term natural killer cell activity. J. Immunol. Methods 185, 209–216. Lichtenfels, R., Biddison, W.E., Schulz, H., Vogt, A.B., Martin, R., 1994. CARE-LASS Žcalcein-release-assay., an improved fluorescence-based test system to measure cytotoxic T lymphocyte activity. J. Immunol. Methods 172, 227–239. Mackensen, A., Carcelain, G., Viel, S., Raynal, M.C., Michalaki, H., Triebel, F., Bosq, J., Hercend, T., 1994. Direct evidence to support the immunosurveillance concept in a human regressive melanoma. J. Clin. Invest. 93, 1397–1401. Mackensen, A., Herbst, B., Chen, J.L., Kohler, G., Noppen, C., ¨ Herr, W., Spagnoli, G.C., Cerundolo, V., Lindemann, A., 2000. Phase I study in melanoma patients of a vaccine with peptide-pulsed dendritic cells generated in vitro from CD34q hematopoietic progenitor cells. Int. J. Cancer 86, 385–392. Mantovani, A., Jerrells, T.R., Dean, J.H., Herberman, R.B., 1979. Cytolytic and cytostatic activity on tumor cells of circulating human monocytes. Int. J. Cancer 23, 18–27. Martin, S.J., Reutelingsperger, C.P., McGahon, A.J., Rader, J.A., Van-Schie, R.C., LaFace, D.M., Green, D.R., 1995. Early redistribution of plasma membrane phosphatidyl serine is a general feature of apoptosis regardless of the initiating stimulus: inhibition by overexpression of Bcl-2 and Abl. J. Exp. Med. 182, 1545–1556.

169

Mattis, A.E., Bernhardt, G., Lipp, M., Forster, R., 1997. Analyzing cytotoxic T lymphocyte activity: a simple and reliable flow cytometry-based assay. J. Immunol. Methods 204, 135– 142. Oelke, M., Moehrle, U., Chen, J.L., Behringer, D., Cerundolo, V., Lindemann, A., Mackensen, A., 2000. Generation and purification of CD8q Melan-A-specific cytotoxic T lymphocytes for adoptive transfer in tumor immunotherapy. Clin. Cancer Res. 6, 1997–2005. Patel, A.K., Boyd, P.N., 1995. An improved assay for antibodydependent cellular cytotoxicity based on time resolved fluorometry. J. Immunol. Methods 184, 29–38. Rosenblatt-Velin, N., Arrighi, J.F., Dietrich, P.Y., Schnuriger, V., Masouye, I., Hauser, C., 1997. Transformed and nontransformed human T lymphocytes migrate to skin in a chimeric human skinrSCID mouse model. J. Invest. Dermatol. 109, 744–750. Rouvier, E., Luciani, M.F., Golstein, P., 1993. Fas involvement in calciumŽ2q.-independent T cell-mediated cytotoxicity. J. Exp. Med. 177, 195–200. Salter, R.D., Howell, D.N., Cresswell, P., 1985. Genes regulating HLA class I antigen expression in T–B lymphoblast hybrids. Immunogenetics 21, 235–246. Sheehy, M.E., McDermott, A.B., Furlan, S.N., Klenerman, P., Nixon, D.F., 2001. A novel technique for the fluorometric assessment of T lymphocyte antigen-specific lysis. J. Immunol. Methods 249, 99–110. Slezak, S.E., Horan, K.P., 1989. Cell mediated cytotoxicity: a highly sensitive and informative flow cytometric assay. J. Immunol. Methods 117, 205–214. Toka, F.N., Niemialtowski, M.G., Spohr de Faundez, I., Gierynska, M., 1996. Cytotoxic T lymphocyte control during ectromelia Žmousepox. virus infection: interaction between MHC-restricted cells analyzed by non-radioactive fluorometry. Acta Virol. 40, 239–244. Vermes, I., Haanen, C., Steffens-Nakken, H., Reutelingsperger, C., 1995. A novel assay for apoptosis: flow cytometric detection of phosphatidyl serine expression on early apoptotic cells using fluorescein-labelled Annexin V. J. Immunol. Methods 184, 39–51. Weidmann, E., Brieger, J., Jahn, B., Hoelzer, D., Bergmann, L., Mitrou, P.S., 1995. Lactate dehydrogenase-release assay: a reliable, nonradioactive technique for analysis of cytotoxic lymphocyte-mediated lytic activity against blasts from acute myelocytic leukemia. Ann. Hematol. 70, 153–158. Yamamura, Y., Rodriguez, N., Schwartz, A., Eylar, E., Bagwell, B., Yano, N., 1995. A new flow cytometric method for quantitative assessment of lymphocyte mitogenic potentials. Cell. Mol. Biol. 41, 121–132. Young, A.J., Hay, J.B., 1995. Rapid turnover of the recirculating lymphocyte pool in vivo. Int. Immunol. 7, 1607–1615.