KRAS G12V neoantigen specific T cell receptor for adoptive T cell therapy against tumors

Abstract

KRAS mutations are broadly recognized as promising targets for tumor therapy. T cell receptors (TCRs) can specifically recognize KRAS mutant neoantigens presented by human lymphocyte antigen (HLA) and mediate T cell responses to eliminate tumor cells. In the present study, we identify two TCRs specific for the 9-mer KRAS-G12V mutant neoantigen in the context of HLA-A*11:01. The TCR-T cells are constructed and display cytokine secretion and cytotoxicity upon co-culturing with varied tumor cells expressing the KRAS-G12V mutation. Moreover, 1-2C TCR-T cells show anti-tumor activity in preclinical models in female mice. The 9-mer KRAS-G12V mutant peptide exhibits a distinct conformation from the 9-mer wildtype peptide and its 10-mer counterparts. Specific recognition of the G12V mutant by TCR depends both on distinct conformation from wildtype peptide and on direct interaction with residues from TCRs. Our study reveals the mechanisms of presentation and TCR recognition of KRAS-G12V mutant peptide and describes TCRs with therapeutic potency for tumor immunotherapy.

Fig. 1. KRAS-G12V-9 specific TCR screening and binding validation in HEK-293T cells.

a TCR repertoire analysis of cloned KRAS-G12V-9/HLA-A*11:01 tetramer⁺ CD8⁺ T cells from Mus-T5, Mus-TF1, Mus-TF2, and Mus-TF6. Numbers below the pie charts represent the number of TCRs identified in the mouse. b The expression of TCRs in HEK-293T cells after co-transfection of chimeric TCR and human CD3-CD8 constructs was detected by staining with antibodies specific to human αβ TCR. Binding of wildtype KRAS-G12wt-9/HLA-A*11:01 tetramer (c) or mutated KRAS-G12V-9/HLA-A*11:01 tetramer (d) to HEK-293T cells transiently expressing the indicated TCRs as in (b) for each panel. The Y axis represents the staining events by the indicated tetramer, while the X axis represents the staining events of CD3-positive cells by the anti-CD3 antibody. Data in (b), (c), and (d) are representative of three independent experiments.

Fig. 2. Specific binding of the 1-2C and 3-2E TCRs to the KRAS-G12V mutant.

a Schematic of the binding assay in HEK-293T cells was created with BioRender.com. b Binding of pHLA tetramers loaded with wildtype or varied KRAS-G12 mutant peptides to the 1-2C or 3-2E TCR expressed on HEK-293T cells was analyzed by flow cytometry. The data shown are from one of three independent experiments. SPR assay characterization of the binding profiles of 1-2C (c) or 3-2E (d) with different KRAS-G12 mutant peptide pHLA proteins. The pHLA proteins were immobilized on the chip and serial dilutions of 1-2C or 3-2E TCR proteins were then flowed through. The figures represent measurements at equilibrium with serial 2-fold dilutions of 1-2C or 3-2E proteins with concentrations ranging from 100 to 6.25 μM. The mean value of the KD was recorded after repeating each experiment three times. Source data are provided as a Source Data file.

Fig. 3. Antigen sensitivity and KRAS-G12V specific tumor responsiveness of 1-2C and 3-2E TCR-T cells.

a Schematic of the functional evaluation assay of the TCRs in Jurkat T cells, created with BioRender.com. Jurkat T cells were transduced to express 1-2C (b) or 3-2E (c) engineered Jurkat T cells were subsequently co-cultured with K562-HLA-A11 target cells and serially diluted peptides for 24 h. Co-cultured supernatants were analyzed by ELISA for secreted IL-2. d Schematic of the functional evaluation assay of the TCR-T cells engineered with primary T cells, created with BioRender.com. e, f 1-2C or 3-2E TCR-T cells were co-cultured with PANC-1 target cells and serially diluted peptides for 24 h. Co-cultured supernatants were analyzed by ELISA for secreted IFN-γ. Each data point represents the mean concentration of IL-2/IFN-γ for each sample run in triplicate wells in (b), (c), (e) and (f), and the data are representative of n = 3 independent experiments. g The 1-2C TCR-T cells were co-cultured with PANC-1 cells stably expressing the wildtype KRAS genes or KRAS G12V, G12C, or G12D mutants. Responses of 1-2C TCR-T cells with wildtype CFPAC-1 cells or CFPAC-1 cells stably expressing HLA-A*11:01 (h), or with wildtype SW-620 cells or SW-620 cells stably expressing HLA-A*11:01 (i). j–l Reactivity of 3-2E TCR-T cells against PANC-1, CFPAC-1, and SW-620 as indicated in (g–i). Each dot of (g–l) represents one technical replicate (n = 1 experiment). The responses were evaluated with 1-2C or 3-2E TCR-T cells prepared with T cells from n = 3 separate donors run independently (Donor 1-3) from (g) to (l). The columns show means of the three technical replicates. (m-r) Luciferase-transduced cell lines were co-cultured with mock-T or TCR-T for 48 h at various E: T ratios. The % specific lysis of the wild type tumor cell lines (black) and over-expression tumor cell lines (blue) obtained by bioluminescence assay is plotted against multiple E:T ratios. Dots of (m–r) represents three technical replicates (n = 1 experiment), data shown are representative of responses from n = 3 separate donors run independently (Donor S1-S3). Source data are provided as a Source Data file.

Fig. 4. Specificity of 1-2C and 3-2E TCR-T cells to combinatorial peptide library screening of KRAS-G12V-9.

a Amino acid positions of KRAS-G12V-9. (b-s) 9-mer combinatorial peptide library screening with PANC-1 and KRAS-G12V-9 peptide reactive 1-2C or 3-2E TCR-T cells, showing the reactive amino acid residue landscape (KRAS-G12V-9 peptide shown with a black star). Responses of the 1-2C (b–j) and 3-2E (k–s) TCR-T cells to the combinatorial peptide library were presented as indicated. Secreted IFN-γ was analyzed by ELISA with the triplicate wells of co-cultured supernatants of TCR-T cells and PANC-1 cells in the presence of the indicated peptides. The dots represent three technical replicates (n = 1 experiment) of the responses from one representative donor. The columns show means of the three technical replicates. Data are representative of the responses from n = 3 separate donors run independently (Donor S1-S3). The responses were investigated with 1-2C or 3-2E TCR-T cells prepared with T cells from Donor S1. Source data are provided as a Source Data file.

Fig. 5. Tumor inhibition efficacy of 1-2C TCR-T cells in a tumor-bearing mouse model.

a Schematic of the PANC-1-G12V mouse model experimental process, created with BioRender.com. NCG mice were inoculated with PANC-1 cells stably expressing KRAS-G12V (4 × 10⁶) (PANC-1-G12V) subcutaneously at day 0 (D0), and three dose of TCR-T cells (1 × 10⁷, 1 × 10⁶, or 1 × 10⁵) were intravenously injected on day 2 (D2). Tumor weights were monitored at the end of the experiment after sacrificing the mice. b The tumor weights of each tumor group from sacrificed mice at the end of the experiment are shown, n = 6 mice per group. c Tumor volumes of five groups of mice treated with 1 × 10⁷ mock-T, 1 × 10⁷ TCR-T, 1 × 10⁶ TCR-T, or 1 × 10⁵ TCR-T. d–g Individual follow-up of tumor sizes is presented for each experimental group with each line showing the changes of the tumor size of each mouse. N = 6 mice per group. h Schematic of the SW-620-A11 mouse model experimental process, created with BioRender.com. Four doses of PD-1 antibodies at 5 mg/kg were administrated through peritoneal injection twice a week in two mouse groups. i The tumor weights of each sacrificed mice from SW-620-A11 tumor groups at the end of the experiment were shown. j Tumor volume of five groups of mice treated with PBS, mock-T cell, mock-T cell plus anti-PD-1 antibody, 1-2C TCR-T cells, 1-2C TCR-T cells plus anti-PD-1 antibody. Mock-T cells without TCR transduction were expanded in parallel with TCR-T cells as a negative control. k–o Individual follow-up of tumor sizes is presented for each experimental group, with each line showing the changes of the tumor size of each mouse. The mean tumor weights or tumor volumes of each group in (b), (c), (i) and (j) were shown as black lines while the standard deviations were represented by the error bars. Statistical analyses utilized two-tailed Student’s t test and the P values were presented as indicated, ns, P > 0.05. Data are shown as means ± SD. Source data are provided as a Source Data file.

Fig. 6. Structure of the 1-2C and 3-2E TCRs bound to KRAS-G12V-9/HLA-A*11:01.

a–d Structure of 1-2C bound to KRAS-G12V-9/HLA-A*11:01. a Overall structures of the 1-2C TCR in complex with the KRAS-G12V-9/HLA-A*11:01 pHLA ligand. The 1-2C TCR is shown in cyan (α chain) and green (β chain). HLA-A*11:01 is shown in pale yellow (heavy chain) and blue (β₂m), and the peptide is shown in pink (KRAS-G12V-9). b The footprints of 1-2C on the KRAS-G12V-9/HLA-A*11:01 pHLA complex. The three CDRαs of 1-2C are represented as ribbons in cyan, while the three CDRβs are in green. HLA-A*11:01 is depicted as a surface in gray and the KRAS-G12V-9 peptide is presented as a surface in pink. c The distribution (%) of HLA-A*11:01 and KRAS-G12V-9 peptide for the recognition of 1-2C TCR. d The distribution (%) of the CDR loops of 1-2C for the interaction with HLA-A*11:01. e–h The binding of the 3-2E TCR with the KRAS-G12V-9 peptide is presented similar to that of 1-2C in (a–d). i Comparison of the CDRs of the 3-2E TCR to that of 1-2C when bound with KRAS-G12V-9/HLA-A*11:01. The CDRs of the 3-2E were colored in gray, whereas the three CDRα of 1-2C are presented in cyan and the three CDRβ in green. j The binding orientation of 1-2C and 3-2E TCR with KRAS-G12V-9/HLA-A*11:01 pHLA. k, l The distribution (%) of the CDR loops of 1-2C for the interaction with the KRAS-G12V-9 peptide.

Fig. 7. Structural basis for the binding specificity to the KRAS-G12V mutation.

a Structure of KRAS-G12V-9 peptides in complexed with 1-2C (magenta) or 3-2E (yellow) TCR. b Top view of the detailed interactions of the KRAS-G12V-9 peptide with residues in HLA-A*11:01, with hydrogen bond interactions depicted as black lines. c Comparison of the structure of KRAS-G12wt peptide (orange) with KRAS-G12V-9 mutant peptide (magenta) presented by HLA-A*11:01. d Comparison of the 9-mer KRAS-G12wt (orange) and G12V mutant (magenta) peptides, and the 10-mer KRAS-G12wt (green) and G12D mutant (cyan) peptides. Detailed interactions of the 1-2C (e) and 3-2E (f) TCRs with the KRAS-G12V-9 peptide. Residues of the TCR are represented in letter-number format. Detailed interactions of the mutated V5 residue of the KRAS-G12V-9 peptide with residues from the 1-2C (g) or 3-2E (h) TCR or residues from HLA-A*11:01. Hydrogen bonds are represented by yellow dashed lines, and Van der Waals contacts are represented by black dashed lines. i The formation of pHLA complex of the wild type or KRAS G12 mutant (G12V, G12D and G12C) peptides with HLA-A*11:01 were evaluated with size exclusion analysis. j DSC evaluated the thermal stabilities of varied pHLA complex proteins loaded with the 9-mer wildtype (G12wt) or KRAS G12 mutant (G12V, G12D and G12C) peptides. Source data are provided as a Source Data file.