Immune rejection of mouse tumors expressing mutated self

Abstract

How the immune system recognizes and responds to mutations expressed by cancer cells is a critical issue for cancer immunology. Mutated self-polypeptides are particularly strong tumor-specific rejection antigens for natural tumor immunity, but we know remarkably little about T-cell responses to mutated self during tumor growth in vivo, including levels of response, kinetics, and correlates that predict tumor rejection. To address these questions, a mutated self-antigen, designated tyrosinase-related protein 1 (Tyrp1)-WM, derived from Tyrp1 was expressed in the poorly immunogenic, spontaneously arising B16 melanoma and the immunogenic, chemically induced LiHa fibrosarcoma. Syngeneic mice challenged with LiHa fibrosarcoma cells expressing Tyrp1-WM, but not native Tyrp1, induced specific CD8(+) and CD4(+) T-cell responses against defined mutated epitopes in tumor-draining lymph nodes and in tumors. Subsequently, specific CD8(+) T-cell responses contracted as a minority of tumors progressed. B16 melanomas expressing Tyrp1-WM induced minimal T-cell responses, and no tumor immunity was detected. Treatment with an agonist monoclonal antibody against glucocorticoid-induced tumor necrosis factor receptor family-related gene (GITR) increased the level of CD8(+) T cells recognizing a peptide derived from the Tyrp1-WM sequence and the proportion of mice rejecting tumors. These results show that B16 tumors expressing mutations that generate strongly immunogenic epitopes naturally induce T-cell responses, which are insufficient to reject tumors. Immune modulation, such as inducing GITR signaling, is required to enhance CD8(+) T-cell responses to specific mutations and to lead to tumor rejection.

Figure 1. LiHa fibrosarcoma expressing WM regresses

A, a schematic of mouse Tyrp1 and WM mutants. Mutations (vertical bars) were designed to improve binding to MHC I molecules. SP, signal peptide. TM, transmembrane domain. CD, cytoplasmic domain. B, 25,000 LiHa cells were used to challenge C57BL/6 mice. 25,000 LiHa-WM cells were also used to challenge CD8-depleted C57BL/6 mice (-CD8). 100,000 B16 cells were used to challenge C57BL/6 mice. Tumor growth was followed three times per week for >60 days. Ratios equal the number of mice with tumors over the total number of mice.

Figure 2. Tumor cells expressing WM, but not wild-type Tyrp1 induce CD8⁺ T cell responses

Mice were injected intradermally with 50,000 LiHa cells or with 100,000 B16 cells. Each symbol represents one mouse. A, proportion of CD8⁺ T cells to total viable cells in tumors (top row); proportion of tetramer-positive cells within the CD3⁺CD8⁺ gate of tumor-infiltrating cells (middle row); proportion of IFN-γ positive cells in the CD3⁺CD8⁺ gate after single-cell suspension from tumors was stimulated with 1 µg/ml WM_455–463 peptide overnight (bottom row). *p=0.008; # p= 0.03, compared to LiHa-Tyrp1. B, top row shows the proportions of tetramer-positive cells within the CD3⁺CD8⁺ gate in individual lymph nodes. The bottom row shows frequencies of functional CD8⁺ T cells by ELISPOT assay. LN, lymph node. DLN, tumor-draining lymph node. CLN, contralateral lymph node. Horizontal bars are means for individual mice. Error bars represent standard error of mean, SEM. * p=0.008, compared to LiHa-Tyrp1 or B16-Tyrp1.

Figure 3. Tumor cells expressing WM induce strong CD4⁺ T cell responses to a novel epitope

A, mice were challenged intradermally with 50,000 LiHa cells or 100,000 B16 cells. Percentages of CD4⁺ T cells in individual tumors are shown. B–C, mice were challenged with either 50,000 LiHa cells or 100,000 B16 cells. IFN-γ ELISPOT assays with CD4⁺ cells (B) or CD8⁺ cells (C) from DLN 10 days after tumor challenge. D, mice were challenged with 50,000 LiHa cells. 50,000 LiHa-WM455/521 cells were also used to challenge CD4-depleted mice, CD8-depleted mice, or NOD-SCID mice. Tumor growth was followed three times per week for >60 days. Error bars represent SEM. Ratios equal the number of mice with tumors over the total number of mice.

Figure 4. Kinetics of T cell responses and tumor infiltration

A, mice were challenged with 50,000 LiHa-WM cells. The line represents the growth curve of tumors. Square symbols, mean tumor diameters (n = 10). Bars, the proportions of tetramer-positive cells in blood. Tumor sizes are shown on the left Y-axis and the proportions of tetramer-positive cells are shown on the right Y-axis. * p< 0.05 compared to naive mice. B, C57BL/6 mice were challenged with 100,000 LiHa cells. At indicated days after tumor challenge, splenocytes were used for intracellular IFN-γ assays. C, two groups of mice were challenged intradermally with LiHa-Tyrp1 (filled bars) or LiHa-WM cells (open bars). At indicated days after tumor challenge, tumors were analyzed by flow cytometry. The proportions of Foxp3⁺CD25⁺ (upper left), CD11b⁺Gr-1⁺ cells (Upper right), NK1.1⁺CD3⁻ (lower left) and NK1.1⁺CD3⁺ (lower right) were shown. D, the proportions of Foxp3⁺CD25⁺CD4⁺ T cells (left panel) and CD11b⁺Gr-1⁺ cells (right panel) in B16 tumors. Error bars represent the SEM of individual mice.

Figure 5. Anti-GITR agonistic antibody treatment augments CD8⁺ T cell responses and tumor immunity

A–D, 100,000 B16-WM cells were injected intradermally in non-treated or anti-GITR-treated mice. A, mice were followed for tumor growth three times weekly for >100 days. B, frequencies of CD4⁺ (filled)/CD8⁺ (open) T cells in tumors. C, frequencies of tetramer-positive cells in tumors were calculated and plotted. D, day 11 after tumor challenge, tumor-draining lymph nodes were used for IFN-γ ELISPOT assays. Error bars represent the SEM of three mice.