Characterization of Tumor-Infiltrating Lymphocyte-Derived Atypical TCRs Recognizing Breast Cancer in an MR1-Dependent Manner

Abstract

The MHC class I-related 1 (MR1) molecule is a non-polymorphic antigen-presenting molecule that presents several metabolites to MR1-restricted T cells, including mucosal-associated invariant T (MAIT) cells. MR1 ligands bind to MR1 molecules by forming a Schiff base with the K43 residue of MR1, which induces the folding of MR1 and its reach to the cell surface. An antagonistic MR1 ligand, Ac-6-FP, and the K43A mutation of MR1 are known to inhibit the responses of MR1-restricted T cells. In this study, we analyzed MR1-restricted TCRs obtained from tumor-infiltrating lymphocytes (TILs) from breast cancer patients. They responded to two breast cancer cell lines independently from microbial infection and did not respond to other cancer cell lines or normal breast cells. Interestingly, the reactivity of these TCRs was not inhibited by Ac-6-FP, while it was attenuated by the K43A mutation of MR1. Our findings suggest the existence of a novel class of MR1-restricted TCRs whose antigen is expressed in some breast cancer cells and binds to MR1 depending on the K43 residue of MR1 but without being influenced by Ac-6-FP. This work provides new insight into the physiological roles of MR1 and MR1-restricted T cells.

Keywords: MR1; T-cell receptors; breast cancer; tumor-infiltrating lymphocytes.

Figure 1

MR1 tetramer staining. BW-hCD8⍺β+ cells transfected with MR1-restricted TCRs were stained with APC-conjugated MR1 tetramers (hMR1/6-FP or hMR1/5-OP-RU) and APC Cy-7 conjugated anti-mouse CD3 antibody. Right upper quadrant is showing CD3+Tetramer+ TCR-expressing BW-hCD8⍺β+ cells. TCR-negative (TCR-ve) BW-hCD8⍺β+ cells were used as a negative control. The plots are representative of two independent experiments.

Figure 2

The reactivity of TCR-transduced mouse splenic T cells and human PBMCs. (A) TCR-expressing mouse splenic T cells were cocultured with IFN-γ-stimulated MCF7 cells. The reactivity against the MCF7 cells was measured by mouse IFN-γ ELISA in triplicate. TCR-negative (TCR-ve) T cells were used as a negative control. (A single experiment is representative of three independent experiments). (B) Human PBMCs from healthy donor A were transduced with TCRs and cocultured with IFN-γ-stimulated MCF7 cells. IFN-γ-production in the supernatant was measured by ELISA in triplicate. TCR-negative (TCR-ve) T cells were used as a negative control. A single experiment is representative of 3 independent experiments. The mean and SD values from technical triplicate cultures are indicated. Unpaired, two-tailed t-tests were performed (** p <0.01 and *** p <0.001).

Figure 3

The reactivity of TCR10-59 to MCF7 WT and various KO cells. (A,B) TCR10-59 and TCR2-15 were transduced into human PBMCs from healthy individuals, and the cells were cocultured with IFN-γ-stimulated MCF7 cells. TNF-⍺ (A) and IFN-γ (B) production in the supernatant was measured by ELISA. TCR-negative (TCR-ve) T cells were used as a negative control. T cells stimulated with PMA + ionomycin were used as a positive control. (A single experiment is representative of three independent experiments). (C) TCR2-78- or TCR10-59-expressing BW-hCD8⍺β+ cells were cocultured with MR1-overexpressing MCF7WT cells or MCF7ΔMR1 cells, and mouse IL-2-secretion was analyzed by ELISA. The mean and SD values from technical triplicate cultures are indicated. (D) PBMCs were retrovirally transduced with conventional (TCR2-15) or unconventional (TCR10-59) TCR genes and cocultured with luciferase-expressing IFN-γ-stimulated MCF7WT, MCF7 WT+B*59:01, and MR1-deleted MCF7 cells at different effector-to-target (E:T) ratios. TCR-negative (TCR-ve) T cells were used as a negative control. The cytotoxic effects were observed by identifying the viable cells by luciferase activity. A single experiment is representative of three independent experiments with similar results. The mean and SD values from technical triplicate cultures are indicated. Unpaired, two-tailed t-tests were performed (** p < 0.01 and *** p < 0.001).

Figure 3

The reactivity of TCR10-59 to MCF7 WT and various KO cells. (A,B) TCR10-59 and TCR2-15 were transduced into human PBMCs from healthy individuals, and the cells were cocultured with IFN-γ-stimulated MCF7 cells. TNF-⍺ (A) and IFN-γ (B) production in the supernatant was measured by ELISA. TCR-negative (TCR-ve) T cells were used as a negative control. T cells stimulated with PMA + ionomycin were used as a positive control. (A single experiment is representative of three independent experiments). (C) TCR2-78- or TCR10-59-expressing BW-hCD8⍺β+ cells were cocultured with MR1-overexpressing MCF7WT cells or MCF7ΔMR1 cells, and mouse IL-2-secretion was analyzed by ELISA. The mean and SD values from technical triplicate cultures are indicated. (D) PBMCs were retrovirally transduced with conventional (TCR2-15) or unconventional (TCR10-59) TCR genes and cocultured with luciferase-expressing IFN-γ-stimulated MCF7WT, MCF7 WT+B*59:01, and MR1-deleted MCF7 cells at different effector-to-target (E:T) ratios. TCR-negative (TCR-ve) T cells were used as a negative control. The cytotoxic effects were observed by identifying the viable cells by luciferase activity. A single experiment is representative of three independent experiments with similar results. The mean and SD values from technical triplicate cultures are indicated. Unpaired, two-tailed t-tests were performed (** p < 0.01 and *** p < 0.001).

Figure 4

The breast cancer cell-specific reactivity of TCR10-59. (A) TCR 2-15- and TCR10-59-expressing PBMCs were cocultured with various IFN-γ-stimulated human cancer cell lines (MDA-MB-231 and ZR-75-1 breast cancer cells, Colo205 colon cancer cells, A549 lung cancer cells, K562 myeloid cells, and Karpas 299 lymphoid cells) for 24 h. TCR-untransduced PBMCs were used as a negative control. T cells stimulated with PMA + ionomycin were used as a positive control. ELISA was performed to measure IFN-γ secretion in the supernatants. The mean and SD values from technical triplicate cultures are indicated. (B) TCR-transduced human PBMCs from healthy individuals were cocultured with IFN-γ-stimulated MCF7 cells as well as normal breast cells (HMECs). IFN-γ production in the supernatant was measured by ELISA. TCR-negative (TCR-ve) T cells were used as a negative control. For the positive control, T cells stimulated with PMA and ionomycin (P/I) were used. (A single experiment is representative of 2 independent experiments). The mean and SD values from technical triplicate cultures are indicated. Unpaired, two-tailed t-tests were performed (** p < 0.01 and *** p < 0.001).

Figure 5

The effects of known microbial MR1 antigens and dependency on K43 MR1-restricted TCRs. TCR2-78- and TCR10-59-expressing BW cells were cocultured with IFN-γ-stimulated MCF7 cells. For stimulation, 5-OP-RU (A), and for inhibition, Ac-6-FP (B) and anti-MR1 antibody (C) were added to the cultures at the indicated concentrations. The dose-dependent responses were assessed by analyzing the IL-2 production in the supernatant. The experiment was performed three times with similar results. (D) TCR2-78- and TCR10-59-expressing BW cells were cocultured with IFN-γ-stimulated MCF7WT, MCF7ΔMR1-MR1^K43A, and MCF7ΔMR1-MR1⁺⁺ cells. 5-OP-RU (125 nM) and Ac-6-FP (40 μg mL⁻¹) were added to the cultures for the stimulation and inhibition of TCR-expressing BW cells, respectively. The IL-2 production in the supernatant was measured as the assessment of responses. The mean and SD values from technical triplicate cultures are indicated. Unpaired, two-tailed t tests were performed (** p < 0.01, *** p < 0.001, **** p < 0.0001 and ns—not significant).