Peptide Modification Diminishes HLA Class II-restricted CD4+ T Cell Recognition of Prostate Cancer Cells

Abstract

Prostate cancer poses an ongoing problem in the western world accounting for significant morbidity and mortality in the male population. Current therapy options are effective in treating most prostate cancer patients, but a significant number of patients progress beyond a manageable disease. For these patients, immunotherapy has emerged as a real option in the treatment of the late-stage metastatic disease. Unfortunately, even the most successful immunotherapy strategies have only led to a four-month increase in survival. One issue responsible for the shortcomings in cancer immunotherapy is the inability to stimulate helper CD4⁺ T cells via the HLA class II pathway to generate a potent antitumor response. Obstacles to proper HLA class II stimulation in prostate cancer vaccine design include the lack of detectable class II proteins in prostate tumors and the absence of defined class II specific prostate tumor antigens. Here, for the first time, we show that the insertion of a lysosomal thiol reductase (GILT) into prostate cancer cells directly enhances HLA class II antigen processing and results in increased CD4⁺ T cell activation by prostate cancer cells. We also show that GILT insertion does not alter the expression of prostate-specific membrane antigen (PSMA), an important target in prostate cancer vaccine strategies. Our study suggests that GILT expression enhances the presentation of the immunodominant PSMA₄₅₉ epitope via the HLA class II pathway. Biochemical analysis showed that the PSMA₄₅₉ peptide was cysteinylated under a normal physiologic concentration of cystine, and this cysteinylated form of PSMA₄₅₉ inhibited T cell activation. Taken together, these results suggest that GILT has the potential to increase HLA class II Ag presentation and CD4⁺ T cell recognition of prostate cancer cells, and GILT-expressing prostate cancer cells could be used in designing cell therapy and/or vaccines against prostate cancer.

Keywords: CD4+ T cells; GILT; HLA class II; antigen presentation; cysteinylation; prostate cancer; prostate-specific membrane protein.

Figure 1

Induction of GILT in human prostate cancer cells by treatment with IFN-γ and insertion of GILT in prostate cancer cells with GILT cDNA. (A) Human prostate cancer cell lines CWR22Rv1 and PC-3 were treated with IFN-γ for three days, followed by washing and Western blot analysis for GILT protein expression. β-actin was used as a loading control. (B) Quantitative analysis of protein bands from (A) by ImageJ software. (C) CWR22Rv1 and PC-3 cells were transfected with GILT cDNA and analyzed by Western blotting for GILT protein expression. A melanoma cell line previously transfected with GILT cDNA (DM-331.GILT) was used as a control. β-actin was used as a loading control. (D) Quantitative analysis of protein bands from (C) by ImageJ software. Data showed that IFN-γ-treated and GILT-transfected CWR22Rv1 and PC-3 cells contained detectable levels of GILT proteins.

Figure 2

GILT insertion did not influence PSMA protein expression in human prostate cancer cells. (A) The expression of PSMA was tested in CWR22Rv1 and PC-3 prostate cancer cell lines by immunofluorescent staining (green) with the anti-PSMA antibody, followed by confocal microscopy. (B) CWR22Rv1 and PC-3 cells transfected with either empty vector or GILT cDNA and analyzed by Western blotting for PSMA protein expression. β-actin was used as a loading control. (C) Densitometric analysis of protein bands obtained from CWR22Rv1.vec, CWR22Rv1.GILT, PC-3.vec, and PC-3.GILT showed no significant changes in PSMA protein expression following GILT transfection. Data are representative of three separate experiments.

Figure 3

GILT insertion did not significantly alter HLA class II protein expression in human prostate cancer cells. (A) The expression of HLA-DR was tested in CWR22Rv1 and PC-3 prostate cancer cell lines by immunofluorescent staining (red) with the anti-HLA-DR antibodies (L243), followed by confocal microscopy. (B) CWR22Rv1, CWR22Rv1.GILT, PC-3, and PC-3.GILT cells were transduced with HLA-DR4 by retroviral transduction and stained with anti-DR4 antibodies (359-F10). Cells were then analyzed by flow cytometry for HLA-DR4 expression. IN-1 was used as an isotype control. Data are representative of three separate experiments.

Figure 4

GILT expression enhances CD4⁺ T cell recognition of prostate cancer cells. HLA-DR-expressing prostate cancer cell lines CWR22Rv1.vec and CWR22Rv1.GILT were cocultured with PSMA-specific CD4⁺ T cells raised from an HLA-DR-positive healthy donor in the presence or absence of anti-class II antibodies (L243) for 48 h. Human IL-2 ELISA was performed to quantitate the amount of IL-2 present in the culture supernatant and was expressed as pg/mL. Data are representative of at least three separate experiments.

Figure 5

CD4⁺ T cell recognition of PSMA peptides on prostate cancer cells. CWR22Rv1 and CWR22Rv1.GILT cells were retrovirally transduced with HLA-DR4 as described in the methods. CWR22Rv1.DR4 and CWR22Rv1.GILT.DR4 cells were then incubated with the PSMA_(206–220) (A) and PSMA_(459–473) (B) peptides (20 μm) overnight at 37 °C and were washed and fixed with 1% paraformaldehyde. Cells were washed again to remove paraformaldehyde and were cocultured with the PSMA_206–220 and PSMA_459–473 peptide-specific CD4⁺ T cells for 48 h. The production of IFN-γ was quantitated using ELISA and expressed as pg/mL. Data are representative of three separate experiments.

Figure 6

Cysteinylation of the PSMA459 peptide reduces CD4⁺ T cell recognition of prostate cancer cells. HLA-DR4-expressing CWR22Rv1.DR4 and PC-3.DR4 cells were incubated with the PSMA_459–473 peptide (20 µm) in either HBSS buffer or IMDM media overnight at 37 °C and were washed and fixed with 1% paraformaldehyde. Cells were washed again and were cocultured with the PSMA_459–473 peptide-specific CD4⁺ T cells for 48 h. The production of IFN-γ was quantitated using ELISA and expressed as pg/mL as described. Data are representative of three separate experiments.

Figure 7

Mass spectroscopic analysis of the PSMA459 (NYTLRVDCTPLMYSL) peptide. The PSMA_(459–473) peptide was dissolved in PBS and culture medium (IMDM) which contains biological levels of cystine (0.29 mM). The PSMA459 peptide fragmentation in PBS was consistent with the calculated molecular mass of 1789.7 (upper panel). Fragmentation of the PSMA₄₅₉ peptide after incubation in IMDM media indicated a mass of 1908.7, consistent with cysteinylation. MSMS analyses in PBS and IMDM are shown in the lower panels. These data suggest that the PSMA₄₅₉ peptide becomes cysteinylated in media containing a physiological concentration of cystine.

Figure 7

Mass spectroscopic analysis of the PSMA459 (NYTLRVDCTPLMYSL) peptide. The PSMA_(459–473) peptide was dissolved in PBS and culture medium (IMDM) which contains biological levels of cystine (0.29 mM). The PSMA459 peptide fragmentation in PBS was consistent with the calculated molecular mass of 1789.7 (upper panel). Fragmentation of the PSMA₄₅₉ peptide after incubation in IMDM media indicated a mass of 1908.7, consistent with cysteinylation. MSMS analyses in PBS and IMDM are shown in the lower panels. These data suggest that the PSMA₄₅₉ peptide becomes cysteinylated in media containing a physiological concentration of cystine.