Fig 2: The in vivo persistence and proliferation of CAR-T cells were evaluated. (A and B) 1 × 106 sorted CAR-T cells per mouse were administrated by intravenous injection. After 2 weeks of metabolism, the abilities to limit tumor growth of the infused CAR-T cells were assessed by re-challenge with A549–19luc cells. BLI images (a) and statistical data (b) both indicated that the residual SCR-CART19 cells limited tumor growth more significantly. c Raji-luc xenograft model was established by injection of 1 × 106 tumor cells per mouse inoculated into abdominal cavity 1 week before the CAR-T treatment. And then, the tumor bearing mice were treated with different CAR-T cells or non-infected T cells as control. The tumor burdens were measured by bioluminescence imaging. d The percentage of CD3 positive T cells was used to evaluate the number of CAR-T cells. And the data for each mouse (up) and the mean value of each group (down) were presented here within 4 weeks after CAR-T reinfusion. e Average fluorescence intensity in each mouse was measured to study the tumor burden, and the changes in tumor burden within 4 weeks after CAR-T reinfusion were presented here. f Survival curves were analyzed using the log-rank test, and the percent survival for each group was presented. These results showed that the in vivo anti-tumor function and proliferation ability of CAR-T cells were impaired by PD-1 knockdown. g The copy numbers of tumor infiltrating CAR-T cells were examined by qRT-PCR to evaluate the intratumoral expansion. And PD-1 silenced CAR-T cells were shown to have impaired intratumoral proliferation. 0.01 < *P < 0.05; **P < 0.01. Statistical significance was determined using the ANOVA method for multiple comparisons. Data represent the mean ± SEM of triplicates and are representative of at least 3 independent experiments or are plotted as individual points

Fig 3: Reactivity of HZ6 and HZ8 TCR engineered Jurkat T cells to NY-ESO-1157-165 WT-SCT-K562 or alanine-substituted pMHC-SCTs (SCT-K562 mutants) target cells. (A) Schematic of the functional evaluation assay of the TCRs in Jurkat T cells. TCR engineered Jurkat T cells co-transduced with human CD3-CD8 construct were co-cultured with K562 target cells that were transduced with NY-ESO-1157-165/HLA-A2 single-chain trimer (SCT-K562 cells). The reactivity of TCR engineered Jurkat T cells was evaluated by detection of the IL-2 secretion after co-culturing of effector cells and target cells. (B) Flow cytometry histogram comparison of NY-ESO-1157-165/HLA-A2 expression on the surface of K562 cells transduced with NY-ESO-1157-165 WT-SCT (WT-SCT-K562) and K562 cells without transduction. (C) Flow cytometry analysis of TCR expression on Jurkat T cells transduced with TCR and CD3-CD8 before and after14 days in vitro culture in the presence of 2 µg/ml puromycin. (D) ELISA measuring secretion of IL-2 from untransduced, HZ6-CD8, and HZ8-CD8 Jurkat effector cells following 48 h co-incubation with WT-SCT-K562 target cells or K562 cells as control. (E) Alanine scanning approach for HZ6 and HZ8 TCR in Jurkat T cells. HZ6-CD8-Jurkat cells or HZ8-CD8-Jurkat cells were co-cultured with WT-SCT-K562 or alanine-substituted pMHC-SCTs K562 (SCT-K562 mutants) cells. IL-2 concentrations in the supernatant were measured by ELISA. For (D,E), the data is a representative of 3 independent experiments with two technical replicates. Means ± SD for a representative experiment are shown.

Fig 4: Specific binding of HZ6 and HZ8 TCRs with NY-ESO-1157-165/HLA-A2 in HEK-293T cells system. (A) Verification strategy for the detection of TCR binding specificity. Cloned TCRs and human CD3-CD8 complex were co-transfected into 293T cells. Binding of NY-ESO-1157-165/HLA-A2 tetramer-PE to TCRs expressed on 293T cells was analyzed with flow cytometry. (B) Binding of NY-ESO-1157-165/HLA-A2 tetramer with CD3-expressing 293T cells co-transfected with TCR and human CD3-CD8 complex constructs. Detection with HPV-E6/HLA-A2 tetramer-PE was enrolled as negative control to verify the binding specificity of HZ6 and HZ8 TCRs (lower flow chart).

Fig 5: Treatment response of 6 different groups (control, DDP, anti‐PD‐1, DDP + anti‐PD‐1, DDP + anti‐PD‐1 + anti‐TGF‐β, DDP + SHR‐1701) in an orthotopic mouse model. (A) CMT167 cells were transplanted into the lungs of C57BL/6 mice. Follow‐up treatment (injected with DDP, anti‐PD‐1, anti‐TGF‐β, SHR‐1701, or isotype control antibody) was then performed in six groups as shown. (B) Images and quantification of in vivo bioluminescence imaging of the six groups (control, DDP, anti‐PD‐1, DDP + anti‐PD‐1, DDP + anti‐PD‐1 + anti‐TGF‐β, DDP + SHR‐1701) at the indicated time points. The P‐value represents a comparison of bioluminescence intensity on the 15th day. (C) Effect of different treatments (control, DDP, anti‐PD‐1, DDP + anti‐PD‐1, DDP + anti‐PD‐1 + anti‐TGF‐β, DDP + SHR‐1701) on survival of CMT167 orthotopic tumor‐bearing mice. (D‐E) Representative images of IHC staining for tumor‐infiltrating Foxp3+ Treg cells (D) and CD8+ T cells (E). Scale bars: 100 μm, upper panel; 40 μm, lower panel. (F) Orthotopic tumors were analyzed by flow cytometry for GITR (left panel), IFN‐γ (middle panel), and CD69 (right panel) expression on CD8+ T cells after stimulation with CD3 and CD28 (both 1 mg/mL) for 48 h. (G) IFN‐γ (left panel) and CD69 (right panel) expression on CD8+ T cells in spleens were analyzed. Bar graphs represent the means ± SEM for each group. B‐G: n = 5 for each group. One‐way ANOVA statistical tests were adopted for more than two groups (B, D, E, F, and G). Log‐rank tests were used in C. *, P < 0.050; **, P < 0.010; ***, P < 0.001. Abbreviations: n.s., not significant; PD‐1, programmed cell death‐protein 1; DDP, cisplatin; TGF‐β, transforming growth factor‐beta; Treg, regulatory T cell; IFN‐γ, interferon‐γ; SEM, standard error of the mean; BL, bioluminescence