Preventing tumor escape by targeting a post-proteasomal trimming independent epitope

Abstract

Adoptive T cell therapy (ATT) can achieve regression of large tumors in mice and humans; however, tumors frequently recur. High target peptide-major histocompatibility complex-I (pMHC) affinity and T cell receptor (TCR)-pMHC affinity are thought to be critical to preventing relapse. Here, we show that targeting two epitopes of the same antigen in the same cancer cells via monospecific T cells, which have similar pMHC and pMHC-TCR affinity, results in eradication of large, established tumors when targeting the apparently subdominant but not the dominant epitope. Only the escape but not the rejection epitope required postproteasomal trimming, which was regulated by IFN-γ, allowing IFN-γ-unresponsive cancer variants to evade. The data describe a novel immune escape mechanism and better define suitable target epitopes for ATT.

Figure 1.

Schematic representation of TCR-I and TCR-IV retroviral vectors with both TCR chains linked by a P2A element as indicated.

Figure 2.

Similar avidity for TCR-I and TCR-IV T cells. (A) Untransduced and TCR-I– or TCR-IV–transduced CD8⁺ P14/Rag^−/− T cells labeled with pI- or pIV-specific tetramers and gated on total lymphocytes. One representative of four experiments is shown. (B) The time needed for dissociation of 50% of bound pMHC complexes (T_1/2) from the surface of an individual T cell is indicated in seconds (TCR-I, n = 21; TCR-IV, n = 12). Combined data from two experiments. The two datasets were compared using a two-sided Wilcoxon rank sum test (P = 0.002). (C) IFN-γ values obtained by co-culturing TCR-I and TCR-IV T cells with equal numbers of RMA-S cells pulsed with different concentrations of pI or pIV. The values are given as percentage of maximum, which was calculated by taking the T cell response to the highest peptide concentration as 100% (maximal value), and then calculating the percentage of for each value as percentage of maximum. One representative experiment of two is shown.

Figure 3.

Similar Ag levels for Tet-TagLuc and 16.113p cells, and similar pMHC affinity for pI and pIV. (A) Histogram shows T-Ag expression measured in T-Ag–negative cells (MCA-205), Tet-TagLuc, and 16.113p tumor cells by intracellular T-Ag staining. One representative of two experiments is shown. (B) The concentration of pI and pIV necessary for yielding 50% binding inhibition (IC₅₀) of the indicator peptides to MHC-I (pI/H2-D^b, pIV/H2-K^b). One representative of three experiments is shown. (C) IFN-γ values obtained by 24-h co-culturing of 10⁵ TCR-I or TCR-IV T cells with indicated numbers of Tet-TagLuc or 16.113p tumor cells. The target cells were incubated with 100 ng/48 h IFN-γ before co-culture. The values are given as percentage of maximum, which was calculated by taking the T cell response to the highest tumor cell number as 100% (maximal value), and then calculating the percentage of for each value as percentage of maximum. One representative of two experiments is shown.

Figure 4.

TCR-I and TCR-IV T cells reject MHC-I high Tet-TagLuc tumors, but only TCR-I reject MHC-I low 16.113p, whereas TCR-IV T cells favor escape. (A) H2-D^b and H2-K^b surface expression for Tet-TagLuc and 16.113p tumor cells stimulated with IFN-γ or left untreated and corresponding isotype controls are shown in the histograms. One representative of three experiments is shown. (B) Tet-TagLuc tumor volumes over time for Rag^−/− mice receiving TCR-I, TCR-IV, or mock-transduced P14/Rag^−/− CD8⁺ T cells. Each line represents the mean value of indicated numbers of mice receiving the same therapy. Error bars correspond to SD. Total mouse numbers from four experiments are shown in the table. (C) 16.113p tumor volumes of indicated number of Rag^−/− mice are shown for TCR-I and mock (error bars correspond to SD), and each line represents a single mouse for TCR-IV (n = 4) treatment. Total mouse numbers from five experiments are shown in the table.

Figure 5.

TCR-IV T cells remain functional in mice with tumor escape. (A) Mean radiance of the signal emitted at the tumor site by localized T cells over time. T cell signal curves for mice receiving TCR-I– and TCR-IV–transduced ChRLuc/OT-1/Rag^−/− CD8⁺ T cells represent a mean of four mice, and one mouse for mock. Error bars correspond to SD. One representative mouse for each treatment is shown at days 1 and 10 after ATT. One representative experiment out of three is shown. (B) The histograms depict three peptide-pulsed CD45.1⁺ gated splenocyte populations (no peptide, pI, and pIV) with low, intermediate, and high CFSE concentration. C57BL/6 mice that were either immunized with 16.113 cells 1 wk before the experiment or left untreated, and a TCR-IV–treated Rag^−/− mouse with recurrent 16.113p tumor (TCR-IV), are shown. The specific pIV kill of individual mice for each group is shown in the chart below the histograms (n = 3 for C57BL/6; n = 4 for C57BL/6 imm; n = 5 for TCR-IV).

Figure 6.

TCR-IV escape variants are unresponsive to IFN-γ stimulation. (A) Shown in the histograms is the MHC-I expression by double staining with biotin-α-H2-K^b/D^b and streptavidin Ab, including the corresponding isotype controls for untreated or IFN-γ–pretreated (100 ng/48 h) 16.113p and 999 (escape variant) cells. One representative staining of two is shown. All seven escape variants were analyzed with similar results (Fig. S1 B). (B) Shown in the histograms is the IFNγR expression on 16.113p and 999 cells by double staining with biotin-α-CD119 Ab and streptavidin Ab, including the corresponding isotype control. The signal was amplified using FASER kit. Shown is one representative staining of two. All seven escape variants were analyzed with similar results (Fig. S1 C). (C) A WB for indicated proteins that were isolated from 16.113p and 999 cells, with or without IFN-γ pretreatment (100 ng/48 h). The same blot was repeatedly stripped and rehybridized. One representative of two is shown, and the blots for the other escape variants are shown in Fig. S2 (A and B).

Figure 7.

Recognition of 16.113p escape variants by TCR-IV T cells is IFN-γ-dependent as well as the rejection of Tet-TagLuc tumors. (A) Levels of secreted IFN-γ by TCR-I– or TCR-IV–transduced CD8⁺ OT-1/Rag^−/− T cells upon 24-h co-culture with 16.113p and 999 cells (with [+] or without [−] 100 ng/48 h IFN-γ pretreatment) measured by an IFN-γ ELISA assay (E:T/2:1). Combined data from two experiments are shown, and error bars indicate SD. The remaining escape variants from the same experiments are shown in Fig. S2 C. (B) Tumor volumes over time from Rag^−/− mice bearing Tet-TagLuc tumors treated with TCR-I, TCR-IV, or mock T cells derived from OT-1/IFN-γ^−/−/Rag^−/− mice. Each line reflects an individual mouse and the bottom table shows total mice numbers from two experiments.

Figure 8.

Epitope I is produced in sufficient amounts by the proteasome, whereas epitope IV requires postproteasomal trimming. Shown are the relevant cleavage products detected in standard proteasome digestions of T-Ag_196-221 (A) and T-Ag_398-417 (B) in three independent experiments. Irrelevant peptide-products are indicated in black, and the epitope and epitope precursor peptides are indicated in red and blue, respectively. For pIV, peptide precursors for an alternative epitope with methionine on the C terminus are shown in green. Thickness of the cleavage products indicates relative cleavage intensity.

Figure 9.

Expression of ERAP1 in escape variants restores recognition by TCR-IV T cells. (A) 1575 cells were transduced with either pMP71-ERAP1-IRES-GFP or pMP71-IRES-GFP and GFP sorted. Histograms indicate GFP expression at days 10 and 20 after sort. (B) The picture shows ERAP1, pSTAT1, LMP2, LMP7, MECL-1, and β-actin WB (single blot) for proteins from pMP71-ERAP1-IRES-GFP stably transduced 1575 GFP-selected cell clones (Cl1, Cl2, and Cl4) compared with 1575 and 16.113p cells (±IFN-γ pretreatment; 100 ng/48 h). One representative of two experiments is shown. (C) The graph shows levels of secreted IFN-γ by TCR-I– and TCR-IV–transduced OT-1/Rag^−/− T cells after co-culture with parental 16.113p cells, 1575, the ERAP1-overexpressing 1575 clones (Cl1 and Cl4), and the ERAP1-negative clone (Cl2) measured by ELISA (E:T/2:1). Combined data from three experiments are shown, and error bars indicate SD.

Figure 10.

Escape variants develop only in large established tumors, independent of endogenous T cells. (A) Tumor volumes over time from Rag^−/− mice bearing small (∼10 d) and large (∼30 d) 16.113p tumors that were treated on the same day with TCR-IV or mock CD8⁺ OT-1/Rag^−/− T cells. A single experiment is shown, and each line represents mean tumor volume of indicated mice receiving the same treatment. (B) Tumor volumes over time from LoxP-TagLuc-pA/CM2 mice bearing 16.113p tumors that received either irradiation (5 Gy) or combination of irradiation and TCR-IV T cell therapy (derived from OT-1/Rag^−/− mice). A single experiment is shown, where each line reflects an individual mouse. (C) MHC-I expression by double staining with biotin-α-H2-K^b/D^b and streptavidin Ab, including the corresponding isotype control for untreated or IFN-γ–pretreated (100 ng/48 h) 16.113p, 29839, 29836, 29840, and 29846 (escape variant) cells. One representative staining of two is shown.