Small-molecule MMP2/MMP9 inhibitor SB-3CT modulates tumor immune surveillance by regulating PD-L1

Abstract

Background: Immune checkpoint blockade (ICB) therapy has demonstrated considerable clinical benefit in several malignancies, but has shown favorable response in only a small proportion of cancer patients. Recent studies have shown that matrix metalloproteinases (MMPs) are highly associated with the microenvironment of tumors and immune cells. However, it is unknown whether MMPs are involved in immunotherapy.

Methods: Here, we used integrative analysis to explore the expression landscape of the MMP family and its association with immune features across multiple cancer types. We used T cell cytotoxicity-mediated tumor killing assay to determine the co-cultured T cell activity of SB-3CT, an MMP2/9 inhibitor. We then used in vitro assays to examine the regulating roles of SB-3CT on PD-L1. We further characterized the efficacy of SB-3CT, in combination with anti-PD-1 and/or anti-CTLA4 treatment in mouse models with melanoma and lung cancer.

Results: Our computational analysis demonstrated a strong association between MMP2/9 and immune features. We demonstrated that inhibition of MMP2/9 by SB-3CT significantly reduced the tumor burden and improved survival time by promoting anti-tumor immunity. Mechanistically, we showed that SB-3CT treatment significantly diminished both mRNA and protein levels of PD-L1 in cancer cells. Pre-clinically, SB-3CT treatment enhanced the therapeutic efficacy of PD-1 or CTLA-4 blockade in the treatment of both primary and metastatic tumors.

Conclusions: Our study unraveled novel molecular mechanisms regarding the regulation of tumor PD-L1 and provided a novel combination therapeutic strategy of SB-3CT and ICB therapy to enhance the efficacy of immunotherapy.

Keywords: Combination treatment; Immune checkpoint blockade; Matrix metalloproteinases (MMPs); SB-3CT.

Fig. 1

Dysregulation of MMPs and associations with cancer hallmarks. a Seven MMP groups based on their typical structures. b Differential score of MMP groups across 14 cancer types (Y-axis) compared to paired normal samples (fold change > 1.5; paired two-sided Student’s t test p < 0.05). Pie chart in the right panel shows the percentage of cancer types with significant upregulation (red), downregulation (blue), and non-significant alteration (gray). c, d MMP2/9 score correlated with c T cell–tumor infiltrating lymphocytes (TILs) and d regulatory T cells (Tregs). Pie charts show the percentage of cancer types with positive (red) and non-significant (gray) correlation. e Spearman’s correlation of MMP groups and inhibitory immune checkpoints. f, g T cell–mediated cancer cell killing assay. f SK-MEL-28 melanoma cells co-cultured with activated T cells for 48 h with or without SB-3CT (12.5, 25 μM) or IFNγ (200 ng/mL) were subjected to crystal violet staining. SK-MEL-28-to-T cell ratio, 1:3. g Statistical analysis. Results are mean ± s.d. NS, p > 0.05, *p < 0.05, **p < 0.01, and ***p < 0.001, as determined by one-way ANOVA and Dunnett’s multiple comparison test

Fig. 2

Synergistic therapeutic effect of combination treatments with MMP2/9 inhibitor, SB-3CT, and anti-PD-1. Images of a B16F10 tumor and b LLC tumor, collected from the combination strategies (control: isotype; PD1 inhibition: anti-PD1; MMP2/9 inhibition: SB-3CT; combination treatment: anti-PD-1+SB-3CT). The B16F10 tumor was captured at the ninth days, and the LLC tumor was captured at the fifth days after treatment. Tumor volumes of c B16F10 and d LLC tumor-bearing C57/BL6 mice treated with isotype, SB-3CT, anti-PD-1, or combination strategy. Kaplan–Meier survival analysis of e B16F10 or f LLC tumor-bearing C57/BL6 mice treated with isotype, SB-3CT, anti-PD-1, or combination strategy. Body weight of g B16F10 and h LLC tumor-bearing C57/BL6 mice treated with isotype, SB-3CT, anti-PD-1, or combination strategy. Sample size per group in one experiment is 5. Error bars represent s.d. of individual mice per group in one experiment. NS, p > 0.05, *p < 0.05, **p < 0.01, and ***p < 0.001, as determined by c, d two-way ANOVA, e, f two-sided log-rank test, and g, h one-way ANOVA

Fig. 3

Immune features in tumors for B16F10 xenograft mouse model with SB-3CT treatment. a Fluorescence expression and b quantification of CD8⁺ T cells. c Heatmap of Z-score normalized percentage of immune cell populations (d–h) in TILs for B16F10 tumor-bearing mice treated with anti-PD-1 and SB-3CT in combination or alone. d–h In the implanted B16F10 tumors from mice treated with or without SB-3CT and PD-1 blockade, fluorescence-activated cell sorting (FACS) was used to measure d CD8⁺ in CD3⁺ T cells, e CD8⁺ IFNγ⁺ in CD8⁺ T cells, f CD8⁺ GZMB⁺ in CD8⁺ T cells, g Gr-1⁺ CD11b⁺ MDSCs in CD45⁺ cells, and h CD25⁺ FOXP3⁺ Treg in CD4⁺ cells. i Fluorescence expression and j quantification of PD-L1 in B16F10 tumor-bearing wild-type C57/BL6 mice treated with isotype, SB-3CT, anti-PD-1, or combination strategy. Sample size is 5 in each cohort. Scale bars, 50 μm. Results are mean ± s.d. ns, p > 0.05,*p < 0.05, **p < 0.01, and ***p < 0.001, as determined by one-way ANOVA and Dunnett’s multiple comparison test

Fig. 4

Downregulation of MMP2/9 by SB-3CT treatment reduced PD-L1 expression. a Spearman’s correlation of group4 score and PD-L1 mRNA expression across 33 cancer types. b–g Evaluation of PD-L1 expression derived from SK-MEL-28 melanoma cell lines and A549 lung cancer cell lines treated with DMSO, SB-3CT (25 μM), IFNγ (200 ng/mL), and IFNγ/SB-3CT in combination for 24 h. b, c PD-L1 expression by RT-PCR in b SK-MEL-28 melanoma cell lines and c A549 lung cancer cell lines. d, e Western blot (left panel: representative images, right panel: quantification) of PD-L1 protein levels in d SK-MEL-28 and e A549. f, g Flow cytometry of PD-L1⁺ membrane level in f SK-MEL-28 and g A549. h–m PD-L1 expression of SK-MEL-28 melanoma cell line transfected with shMMP2 (h–j) and shMMP9 (k–m) or the scrambled negative control shRNA (shNC). Western blot (h, k) quantification of MMP2, MMP9, and PD-L1 protein expression (i, l), and RT-PCR analysis of MMP2, MMP9, and PD-L1 mRNA expression (j, m)

Fig. 5

Regulation of PD-L1 expression through MMP2/9 has an anti-tumor effect. a–f Analysis of PD-L1 expression of SK-MEL-28 melanoma cell line with overexpression (oe) MMP2 (a–c), oeMMP9 (d–f). Western blot (a, d), quantification (b, e), and RT-PCR analysis (c, f) of MMP2, MMP9, and PD-L1 protein or mRNA expression. g–l PD-L1 expression of shMMP2 (g–i) and shMMP9 (j–l) SK-MEL-28 melanoma cell line treated with SB-3CT. Western blot (g, j); quantification of MMP2, MMP9, and PD-L1 protein (h, k); and mRNA expression (i, l). m Western blot showed the protein expression of PD-L1 for SK-MEL-28 melanoma cells with overexpression of PD-L1, treated with or without SB-3CT. n Z-scale normalization expression of differentially expressed genes (fold change > 1.5 and two-sided Student’s t test p < 0.05) between A375 melanoma cell lines treated with IFNγ/SB-3CT in combination and IFNγ. o Enriched signaling pathways for genes downregulated in A375 melanoma cell lines treated with SB-3CT and IFNγ in combination (Fisher’s exact test p < 0.05). All experiments were repeated three times independently. Results are mean ± s.d. NS, p > 0.05, *p < 0.05, **p < 0.01, and ***p < 0.001, as determined by one-way ANOVA and Dunnett’s multiple comparison test

Fig. 6

Synergistic therapeutic effect of SB-3CT treatment combined with CTLA-4 blockade. Tumor volumes of a B16F10 or b LLC tumor-bearing C57/BL6 mice treated with isotype, SB-3CT, anti-CTLA-4, or combination strategy. Kaplan–Meier survival analysis of c B16F10 or d LLC tumor-bearing C57/BL6 mice treated with isotype, SB-3CT, anti-CTLA-4, or combination strategy. e–i Immune features in B16F10 tumor-bearing C57/BL6 mice treated with isotype, SB-3CT, anti-CTLA-4, or combination strategy. e Fluorescence expression and f quantification of PD-L1 expression in tumor. g Fluorescence expression and h quantification of PD1⁺CD8⁺ T cells in tumor. i Heatmap represents Z-score normalized percentage of immune cell populations. Sample size per group in one experiment is 5. Error bars represent s.d. of individual mice per group in one experiment. NS, p > 0.05, *p < 0.05, **p < 0.01, and ***p < 0.001, as determined by a, b two-way ANOVA, c, d two-sided log-rank test, and f, h one-way ANOVA and Dunnett’s multiple comparison test

Fig. 7

SB-3CT significantly reduces melanoma lung metastasis in combination with checkpoint inhibition. a Spearman’s correlation between MMP2/9 score and epithelial-to-mesenchymal transition (EMT) score. b Heatmap shows the relative expression of mesenchymal genes differentially expressed between B16F10 tumor with vehicle (n = 4) and SB-3CT treatment (n = 4). c EMT score between SB-3CT treatment and vehicle group. p value is determined by two-sided Student’s t test. d Representative images and e quantification of lung metastasis in wild-type C57/BL6 mice with B16F10 cells intravenously treated with isotype, SB-3CT, anti-PD-1, or combination strategy. f Kaplan–Meier survival curve for mice over time. g Representative images and h quantification of lung metastasis in wild-type C57/BL6 mice with B16F10 cells intravenously treated with isotype, SB-3CT, anti-PD-1, or combination strategy. i Kaplan–Meier survival curve for mice over time. Sample size per group in one experiment is 5. Error bars represent s.d. of individual mice per group in one experiment. NS, p > 0.05, *p < 0.05, **p < 0.01, and ***p < 0.001, as determined by b, c two-way ANOVA, f, i two-sided log-rank test, and e, h one-way ANOVA and Dunnett’s multiple comparison test