Characterization of human non-small cell lung cancer (NSCLC) cell lines for expression of MHC, co-stimulatory molecules and tumor-associated antigens.

A panel of 31 long-term non-small cell lung cancer (NSCLC) cell lines was examined for the expression of protein and/or mRNA transcripts for 11 distinct immune response related molecules or tumor associated antigens (TAA). To assess whether cytokine stimulation might up-regulate expression of the genes of interest, cells were cultured in 500 U/ml of gamma-interferon (gamma-IFN) for 48-72 h prior to analysis. Major histocompatibility complex (MHC) Class I antigens were detected by indirect immunofluorescence and were constitutively expressed on all of the cell lines. The average of the mean fluorescence intensity (MFI) measured 222+/-22. gamma-IFN stimulation produced a significant increase to 482+/-36. For MHC Class II only 7/31 cell lines (23%) exhibited constitutive expression, while gamma-IFN treatment had a dramatic effect and yielded 18/31 (58%) positive cell lines. The co-stimulatory molecules CD80 and CD86 were examined by direct immunofluorescence for cell surface expression and RT-PCR amplification for mRNA. CD80 protein was not detected at all, while an insignificant percentage of cells were positive (mean 2%) for CD86 in all cell lines tested. gamma-IFN had no apparent effect on CD80 or CD86 protein expression. Constitutive CD80 or CD86 mRNA levels were observed in 45 and 61% of the NSCLC lines, respectively. These percentages increased to 77% and 90% with gamma-IFN. Cell surface phenotypic analysis for TAA revealed positive populations in 28/31 cell lines (90%) for Her-2/neu, 18/31 (58%) for CEA and 8/31 (26%) for GD-2, with gamma-IFN having no effect. After gamma-IFN stimulation, RT-PCR amplification for Mage-1, -2, -3 and WT-1 detected mRNA in 33%, 33%, 44% and 70% of the cell lines, respectively. Overall, gamma-IFN stimulation led to the up-regulation of MHC Class I molecules and class II molecules as well as CD80 and CD86 mRNA transcripts. This survey represents the first comprehensive analysis of NSCLC cell lines for a variety of molecules that could play an important role in the generation of an NSCLC anti-tumor CD8+ cytotoxic T lymphocyte (CTL) response.

CD80 expression in an HLA-A2-positive human non-small cell lung cancer cell line enhances tumor-specific cytotoxicity of HLA-A2-positive T cells derived from a normal donor and a patient with non-small cell lung cancer.

To generate non-small cell lung cancer (NSCLC)-reactive lymphocytes, we transfected an HLA-A2-expressing human NSCLC line (1355) with the cDNA encoding the lymphocyte co-stimulatory molecule CD80. Following selection in G418, 1355.7 demonstrated stable cell-surface expression of CD80. Allogeneic mixed lymphocyte tumor cell cultures (MLTCs) were established in 600 IU/ml IL-2 using HLA-A2+ normal donor peripheral blood mononuclear cells (PBMCs) stimulated with 1355-P (parental), 1355.7 or IL-2 alone. In 7 of 9 MLTCs, those stimulated with 1355.7 demonstrated enhanced growth after 30 to 45 days of culture. The predominant lymphocyte to grow in all MLTCs was a CD3+alphabeta+CD4+ T cell. In one case, lymphocytes stimulated with 1355.7 (MLTC 2389.7) exhibited preferential lysis of 1355. MLTC 2389.7 was cloned by limiting dilution, and 2 resultant cloids were shown to be NSCLC-reactive and dependent on both MHC class 1 and CD3 in their recognition of tumor cells. Additionally, allogeneic MLTCs were established using three HLA-A2+ NSCLC patients' PBMC. The predominant lymphocyte to grow in these MLTCs was a CD3+ alphabeta+CD8+ T cell. In cytotoxicity studies, MLTC-UKY25.7 demonstrated preferential lysis of 1355-P, 1355.7 and an HLA-A2+ NSCLC cell line, 1650. Lymphocytes from this MLTC did not lyse K562, Daudi or an HLA-A2- NSCLC cell line, 647. Our data suggest that CD80-expressing NSCLC tumor cells may enhance the generation of specific CTLs in vitro. These CTLs could be important reagents for use in cellular immunotherapy and/or in isolating tumor antigens for potential tumor vaccine development.