4.1 Cells and culture conditions
The human alveolar type II epithelial cell line A549 was cultured in DMEM supplemented with 10 mM Hepes, 5 mM L-glutamine, 100 μg/ml gentamycin and 10% FCS (all from GIBCO BRL, Eggenstein, Germany) in 5% CO2 at 37°C. If not otherwise indicated, A549 cells were removed from the wells by trypsinization (trypsin-EDTA; GIBCO BRL) and collected. The human alloreactive CD8+ T cell line T221 was kindly provided by Dr. Elisabeth Märker-Hermann (Johannes Gutenberg University of Mainz, Mainz, Germany) and was maintained in RPMI 1640 medium (GIBCO BRL) containing 10 mM Hepes, 2 mM L-glutamine, 100 μg/ml gentamycin, 20 U/ml human IL-2 (kindly provided by Dr. Hansjörg Hauser, German Research Center for Biotechnology, Braunschweig, Germany) and 10% human AB serum (ICN, Meckenheim, Germany) in 5% CO2 at 37°C. T221 CTL were stimulated with PHA (1 μg/ml; Sigma-Aldrich, Deisenhofen, Germany) every 3 weeks in the presence of the irradiated EBV-immortalized B cell lines PLH and HMY-2 [kindly provided by Dr. Reinhardt Schwartz-Albiez and Dr. Gerd Moldenhauer (German Cancer Research Center, Heidelberg, Germany), respectively] and irradiated PBMC. T221 CTL were separated from feeder cells by density gradient centrifugation before use in each experiment.
4.2 Immunostaining and flow cytometry
A549 cells that were either left untreated or treated with 1,300 U/ml human IFN-γ (Upstate Biotechnology, Lake Placid, NY) for 24 h were collected by rubber policemen rather than by trypsinization. Both T221 CTL and A549 cells were stained with mouse anti-human HLA-DR-PE (BD PharMingen, Hamburg, Germany) or isotype-matched mouse IgG2a-PE (Dianova, Hamburg, Germany). For detection of CD95, TRAIL, and TRAIL-R, cells were stained with mouse IgG1 Ab against human CD95 44, TRAIL, TRAIL-R1, -R2, -R3, -R4 (Alexis Biochemicals, San Diego, CA), or isotype-matched mouse IgG1 as a control, followed by biotinylated goat anti-mouse IgG1 and then streptavidin-PE (all from BD PharMingen). Two-color staining of T221 CTL was done using mouse anti-human CD3-FITC and CD8-PE mAb or mouse anti-human CD4-FITC and CD8-PE mAb (all from BD PharMingen). Surface staining was determined on a FACScan flow cytometer (Becton Dickinson, Heidelberg, Germany), and data were analyzed with CellQuest software (Becton Dickinson).
4.3 Cytotoxicity assay
A549 cells were cultured at a concentration of 2×105 cells/ml in 100 μl DMEM containing 10% FCS in flat-bottom 96-well plates in the presence or absence of 1,300 U/ml human IFN-γ for 24 h. Cells were then washed twice and cultured for a further 7 h with the agonistic anti-CD95 Ab anti-APO-1 (IgG3) (which recognizes an epitope on the extracellular domain of human CD95 13), leucine zipper (LZ)-TRAIL (which was produced essentially as described 45 and which is a stable trimer of TRAIL that induces apoptosis upon binding to TRAIL-sensitive cells 45), or with TNF-α (Biomol, Hamburg, Germany), either alone or in the presence of CHX (Boehringer Mannheim, Mannheim, Germany). Cells were collected, washed twice with cold PBS and double-stained with FITC-conjugated annexin V (Boehringer Mannheim) and 50 μg/ml PI (Sigma-Aldrich), according to the manufacturer's instructions. Apoptosis of gated (R1 gate in Fig. 2A) cells in side scatter (SSC) versus forward scatter (FSC) dot blots was then analyzed by FACS. In one experiment, A549 cells treated or untreated with human IFN-γ as described above were cultured for a further 20 h in medium alone or medium supplemented with anti-APO-1, protein A (Sigma-Aldrich), or a combination of the two. Apoptosis was verified by FACS analysis using the method of Nicoletti et al. 46. Briefly, cells were collected, washed and resuspended in a buffer containing 0.1% (v/v) Triton X-100, 0.1% (w/v) sodium citrate (both from Sigma-Aldrich), and 50 μg/ml PI. After incubation at 4°C in the dark for at least 14 h, apoptotic nuclei were quantified by FACS.
In co-culture assays, A549 cells were seeded into 100 μl DMEM containing 10% FCS in U-bottom 96-well plates at a density of 0.4×105 cells/ml. After 24 h, the cell number in arepresentative well was counted, and T221 CTL were added to each well at the indicated E/T ratios in 100 μl DMEM containing 10% FCS and 20 U/ml human IL-2 after two washes of A549 cells. Plateswere centrifuged at 800 rpm for 2 min, and cells were further incubated in the presence or absence of SEB (Sigma-Aldrich). Cells were collected at the indicated times of incubation, washed twice with cold PBS, double-stained with FITC-labeled annexin V and PI and analyzed by FACS as described above. To assay for contamination of T221 CTL in the R1 gate (see Fig. 2A), A549 cells were labeled with 2.5 μM CFSE (Molecular Probes, PoortGebouw, The Netherlands) at room temperature for 10 min, washed four times and cultured at a density of 0.4×105 cells/ml in 100 μl DMEM with 10% FCS in U-bottom 96-well plates for 24 h. T221 CTL equivalent to ten times the number of A549 cells were added to some wells, and cells were further cultured for 12 h in the presence of 0.1 μg/ml SEB as described above. The percentage of CFSE dye-positive cells in the R1 gate was analyzed by FACS.
To analyze the roles of several death systems in the killing of A549 cells by activated CD8+ T cells, NO synthase inhibitor L-NMMA (Sigma-Aldrich), vacuolar H+-ATPase inhibitor CMA (Calbiochem, Bad Soden, Germany), neutralizing Ab against CD95L NOK1 (BD PharMingen; 47), TRAIL-R2-Fc protein 45, TNFR2-Fc protein (EnbrelTM; obtained by the Heidelberg University, Pharmacy, Heidelberg, Germany), or isotype-matched control mouse IgG1 (BD PharMingen) or human IgG1 (Sigma-Aldrich) were preincubated with T221 CTL for 30 min. A549 cells were also preincubated with the broad-spectrum caspase inhibitor BD-fmk (Enzyme System, Livermore, CA) or L-NMMA for 30 min. In the apoptosis inhibition assay, A549 cellswere first labeled with 2.5 μM CFSE, then washed and subsequently cultured with T221 CTL and SEB in the presence or absence of several apoptosis-inhibiting reagents to minimize contamination ofA549 cells by T221 CTL. Collected cells were washed, stained with annexin V-PE (BD PharMingen) according to the manufacturer's instructions and then analyzed by FACS without gating in SSC versus FSCdot blots.
To examine whether activated CD8+ T cells induce DNA fragmentation in A549 cells, a JAM test was performed as described previously 48. Briefly, A549 cells were labeled by incubation with 10 μCi/ml [3H]thymidine (NEN, Neu-Isenburg, Germany) for 12 h at a density of 3×105 cells/ml. Cells were collected, washed, cultured at a density of1×105 cells/ml in 100 μl DMEM with 10% FCS in flat-bottom 96-well plates for 7 h, and then T221 CTL were added to each well at an E/T ratio of 5:1. The final volume per well was 200 μl DMEM with 10% FCS. Cells were centrifuged at 800 rpm for 2 min and co-cultured for 20 h. Cells were then harvested using a LKB/Wallace 1295–001 cell harvester (LKB Wallace, Turku, Finland).To collect all adherent cells, cells were trypsinized and re-harvested. Radioactivity bound to the filter was measured by scintillation counting in a 1205 betaplate counter (LKB Wallace). Percentages of DNA fragmentation were calculated using the following formula: DNA fragmentation (%) = (cpmspontaneous–cpmexperimental)/cpmspontaneous × 100. The value for cpmspontaneous was determined by incubating A549 cells with medium only.
4.4 Reverse transcription-PCR analysis
Total RNA was isolated using the RNeasy kit (Qiagen, Hilden, Germany). cDNA was synthesized using Superscript II reverse transcriptase (GIBCO BRL) according to the manufacturer's instruction. PCR amplification was performed for 23–33 cycles with the following primer pairs: β-actin (sense: 5′-GAGGCCCAGAGCAAGAGAGG-3′; antisense: 5′-TCACCGGAGTCCATCACGAT-3′),TCR Vβ14 (sense: 5′-GTCTCTCGAAAAGAGAAGAGGAAT-3′; antisense: 5′-TTCTGATGGCTCAAACAC-3′), perforin (sense: 5′-CAGCCAACTTTGCAGCCCAGAAG-3′; antisense: 5′-GCAGGTCGTTAATGGAGGTGTGA-3′), granzyme A (sense: 5′-AGAGCTGCTCACTGTTGGG-3′; antisense: 5′-AGGAGACAATGCCCTGGG-3′), granzyme B (sense: 5′-TGCAGGAAGATCGAAAGTGCG-3′; antisense: 5′-GAGGCATGCCATTGTTTCGTC-3′), CD95L (sense: 5′-ATGTTTCAGCTCTTCCACCTACAGAAGGA-3′; antisense: 5′-ACCAGAGAGAGCTCAGATACGTTGACATA-3′), TRAIL (sense: 5′-CCCAATGACGAAGAGAGTATGAA-3′; antisense 5′-ACCATTTGTTTGTCGTTCTTTGTG-3′), TNF-α (sense: 5′-ACAAGCCTGTAGCCCATGTT-3′; antisense: 5′-AAAGTAGACCTGCCCAGACT-3′), IFN-γ (sense: 5′-GCAGAGCCAAATTGTCTCCT-3′; antisense: 5′-ATGCTCTTCGACCTCGAAAC-3′). To ensure that PCR analyses were in the linear range, reactions were performedusing various numbers of cycles for each primer pair, and products were analyzed by electrophoresis on 2% agarose gels. β-Actin expression levels were used to normalize the amounts of cDNA template. In all reverse transcription (RT)-PCR analyses, negative controls in which cDNA synthesis was done without reverse transcriptase showed no specific band after PCR.
4.5 Preparation of supernatants
A549 cells were plated (2×105 cells/well) in 24-well plates in DMEM with 10% FCS in the presence or absence of human IFN-γ (1,300 U/ml) and cultured for 24 h. After three washes, cells were cultured for a further 24 h in DMEM with 0.5% FCS and 40 U/ml human IL-2 in the presence or absence of SEB (0.3 μg/ml) and T221 CTL (E/T ratio = 8:1). Supernatants were collected, centrifuged for 10 min at 5,000 rpm, filtered through 0.22-μm filters and either used immediately or frozen at –80°C until further use. The supernatant from T221 CTL cultured for 24 h in the presence or absence of SEB (0.3 μg/ml) at the same cell density and in the same medium as described above was also collected.