Antigen spreading-induced CD8+T cells confer protection against the lethal challenge of wild-type malignant mesothelioma by eliminating myeloid-derived suppressor cells

Abstract

A key focus in cancer immunotherapy is to investigate the mechanism of efficacious vaccine responses. Using HIV-1 GAG-p24 in a model PD1-based DNA vaccine, we recently reported that vaccine-elicited CD8+ T cells conferred complete prevention and therapeutic cure of AB1-GAG malignant mesothelioma in immunocompetent BALB/c mice. Here, we further investigated the efficacy and correlation of protection on the model vaccine-mediated antigen spreading against wild-type AB1 (WT-AB1) mesothelioma. We found that this vaccine was able to protect mice completely from three consecutive lethal challenges of AB1-GAG mesothelioma. Through antigen spreading these animals also developed tumor-specific cytotoxic CD8+ T cells, but neither CD4+ T cells nor antibodies, rejecting WT-AB1 mesothelioma. A majority of these protected mice (90%) were also completely protected against the lethal WT-AB1 challenge. Adoptive cell transfer experiments further demonstrated that antigen spreading-induced CD8+ T cells conferred efficacious therapeutic effects against established WT-AB1 mesothelioma and prevented the increase of exhausted PD-1+ and Tim-3+ CD8+ T cells. A significant inverse correlation was found between the frequency of functional PD1-Tim3- CD8+ T cells and that of MDSCs or tumor mass in vivo. Mechanistically, we found that WT-AB1 mesothelioma induced predominantly polymorphonuclear (PMN) MDSCs in vivo. In co-cultures with efficacious CD8+ T cells, a significant number of PMN-MDSCs underwent apoptosis in a dose-dependent way. Our findings indicate that efficacious CD8+ T cells capable of eliminating both tumor cells and MDSCs are likely necessary for fighting wild-type malignant mesothelioma.

Keywords: CD8+T cells; Immune response; Immunity; Immunology and Microbiology Section; MDSCs; antigen spreading; mesothelioma; vaccination.

Conflict of interest statement

CONFLICTS OF INTEREST

No potential conflicts of interest were disclosed.

Figures

Figure 1. Complete protection of mice against three consecutive lethal challenges of AB1-GAG mesothelioma

A. Schematic representation of a prophylactic study on the sPD1-p24fc/EP DNA vaccine against repeated AB1-GAG mesothelioma challenges. B. After three consecutive lethal challenges of 5 × 105 AB1-GAG mesothelioma cells, tumor growth was assessed by BLI as compared with p24fc/EP-vaccinated mice. **P = 0.007. Data was plotted as mean ± s.e.m. C. Representative BLI of AB1-GAG tumor growth in vaccinated mice according to the experimental time schedule A.. Images were acquired at the indicated time points after tumor cell inoculation. The color scale indicates the BLI signal intensity. Two independent experiments generated the same results.

Figure 2. Tumor-specific immune responses after three consecutive lethal challenges of AB1-GAG mesothelioma

A. Antibody binding to WT-AB1 cells by flow cytometry. Representative histogram shows the binding of 1:100 diluted anti-sera to WT-AB1 cells with the fluorescent signal shifted to the right (unfilled peak) as compared to control sera (red filled peak). Pre-labeled anti-mouse IgG was used as the detection antibody (left panel). Using various serum dilutions, immune sera show consistent binding to WT-AB1 cells in a dose-dependent way as compared with the control sera (right panel). **P < 0.01; ***P < 0.001. B.In vitro CD8+ T cell cytotoxic assay. Representative bioluminescence images depict the luminescence intensity in each well at the E/T ratio of 20:1 (left panel). The degree of CTL-mediated killing of tumor cells was indicated by the decrease of luminescence activity using the IVIS100 luminescence imaging system (right panel). BLI signals were acquired for 20 second. **P = 0.0066; ***P = 0.00078. C.In vitro CD4+ T cell cytotoxic assay. Representative bioluminescence images depict the luminescence intensity in each well at the E/T ratio of 20:1 (left panel). There were no CD4+ T cytotoxic activities against tumor cells because of the lack of significant differences between experimental groups (right panel).

Figure 3. Protection of mice against the lethal challenge of wild-type mesothelioma

A. Schematic representation of the study design. B. After a lethal dose of 5×105 WT-AB1 challenge, tumor growth was assessed by BLI overtime. Statistical analysis was done on day 21 between groups I and II or III (n = 5 in I and II, 3 in III). Data was plotted as mean ± s.e.m. **P = 0.0098; *P = 0.0376. C. Survival curve. The majority of mice (90%, 9/10) in group I survived after the WT-AB1 challenge and remained tumor free. Results were combined from two independent experiments with n = 5 each. ***P = 0.0002. D. Representative BLI depicts the growth of WT-AB1 mesothelioma in one of two independent experiments. Images were acquired at the indicated time points after the inoculation of WT-AB1.

Figure 4. Adoptive transfer of tumor-specific CD8+ T cells eliminates established WT-AB1 malignant mesothelioma

A. Time schedule of the adoptive transfer of CD8+ T cells. SCID mice were pre-inoculated s.c. with 5 × 105 WT-AB1 cells, and tumors were left to grow for 7 days to reach approximately 5 mm in length. Cyclophosphamide (Cy) was injected intraperitoneally (150 mg/kg) one day before the adoptive transfer of CD8+ T cells. 2 × 106 of isolated CD8+ T cells in 100 μl PBS were then injected via the tail vein into each mouse. B. Tumor growth was assessed by BLI overtime. CD8+ T cells of groups I, II and III were adoptively transferred i.v. into SCID mice of groups I’, II’ and III’, correspondingly. Intravenous transfer of CD8+ T cells of group I significantly eliminated established WT-AB1 tumor in SCID mice of group I’ (n = 5). Data represent mean ± s.e.m. *P = 0.0194. C. Representative BLI of WT-AB1 tumor growth in three groups of SCID mice. Images were acquired at the indicated time points after WT-AB1 tumor inoculation.

Figure 5. Frequency and correlation analysis of CD8+ T cells and MDSC in spleens of SCID mice of groups I’, II’ and III’

A. Frequencies of MDSC, PD1+ CD8+ T cells, and Tim3+ CD8+ T cells were measured at the experimental endpoint (Figure 4A). Data represent mean ± s.e.m. *P < 0.05; ** P < 0.01; ***P < 0.001. B. The frequency of PD1−Tim3− CD8+ T cells was inversely correlated with that of MDSC (left panel) and with tumor mass (right panel), respectively.

Figure 6. In vitro analysis of MDSCs and CD8+ T cells

A. Flow cytometry analysis of MDSC subpopulations isolated from spleens of WT-AB1-bearing mice (left panel). Different levels of Fas and DR5 expression were found between PMN-MDSCs and M-MDSCs (right panel). B. The expression of active caspase-3 and annexin V overtime. The majority of M-MDSCs undergo apoptotic cell death faster than PMN-MDSCs during the in vitro culture. C. Flow cytometry analysis of FasL (left panel) and TRAIL (right panel) expression on CD8+ T cells isolated from spleens of group I and control mice. D. IL-2 and IFN-γ production from co-cultures of WT-AB1-induced MDSCs with group I CD8+ T cells but not with control CD8+ T cells at different MDSC:CD8+ ratios of 1:1, 1:2 and 1:5. Data represent mean ± s.e.m. Results are representative of three independent experiments. E. By measuring the expression of caspase-3 and annexin V, significant proportions of apoptotic PMN-MDSCs were detected in co-cultures with group I CD8+ T cells as compared with control CD8+ T cells at 24 h. Data represent mean ± s.e.m. Results are representative of three independent experiments. *P < 0.05; **P < 0.01; ***P < 0.001.