Intratumoral delivery of the chitin-derived C100 adjuvant promotes robust STING, IFNAR, and CD8+ T cell-dependent anti-tumor immunity

Abstract

Stimulator of IFN genes (STING) is a promising target for adjuvants utilized in in situ cancer vaccination approaches. However, key barriers remain for clinical translation, including low cellular uptake and accessibility, STING variability necessitating personalized STING agonists, and interferon (IFN)-independent signals that can promote tumor growth. Here, we identify C100, a highly deacetylated chitin-derived polymer (HDCP), as an attractive alternative to conventional STING agonists. C100 promotes potent anti-tumor immune responses, outperforming less deacetylated HDCPs, with therapeutic efficacy dependent on STING and IFN alpha/beta receptor (IFNAR) signaling and CD8⁺ T cell mediators. Additionally, C100 injection synergizes with systemic checkpoint blockade targeting PD-1. Mechanistically, C100 triggers mitochondrial stress and DNA damage to exclusively activate the IFN arm of the cGAS-STING signaling pathway and elicit sustained IFNAR signaling. Altogether, these results reveal an effective STING- and IFNAR-dependent adjuvant for in situ cancer vaccines with a defined mechanism and distinct properties that overcome common limitations of existing STING therapeutics.

Keywords: CD8+ T cell; DNA sensing; STING; adjuvant; cancer immunotherapy; cancer vaccine; chitin derived polymer; chitosan; interferon.

Graphical abstract

Figure 1

C100 significantly improves the survival of mice bearing MC38 colon carcinoma or B16 melanoma (A) Schematic of B16 melanoma and MC38 carcinoma experimental design. In brief, C57BL/6 mice were injected subcutaneously (s.c.) on the flank with B16 melanoma or MC38 carcinoma cells. Tumor-bearing mice were randomized and injected i.t. with PBS or HDCPs on days 0, 4, and 8. (B and C) B16 melanoma growth rate (B) and Kaplan-Meier survival analysis (C) in mice treated with PBS or HCDPs (100 μg) on days 0, 4, and 8. (D and E) MC38 carcinoma growth rate (D) and Kaplan-Meier survival analysis (E) in mice treated with PBS or C100 (100 μg) on days 0, 4, and 8. (B and D) Data are expressed as mean tumor volume ± SEM. (B and D) Unpaired two-tailed t tests or (C and E) Mantel-Cox Tests were used to determine statistical significance between treatments, ∗∗p < 0.001 and ∗∗∗p < 0.001. (B and C) n = 7 mice per group. (D and E) PBS n = 6, C100 n = 5 mice per group.

Figure 2

STING and IFNAR are required for the therapeutic efficacy of C100 (A and B) C57BL/6 WT or Tmem173^−/− mice were injected s.c. on the flank with WT or Tmem173^−/− B16F10 cells. Tumor-bearing mice were randomized and injected i.t. with PBS or C100 (100 μg) on days 0, 4, and 8. (A) B16 melanoma growth rates. (B) B16 melanoma Kaplan-Meier survival analysis. (C‒F) C57BL/6 WT, Ifnar1^−/−, or Nlrp3^−/− mice were injected s.c. on the flank with MC38 carcinoma cells. Tumor-bearing mice were randomized and injected i.t. with PBS or C100 (100 μg) days 0, 4, and 8. (C) MC38 carcinoma growth rate in WT and Ifnar^−/− mice. (D) MC38 carcinoma Kaplan-Meier survival analysis in WT and Ifnar^−/− mice. (E) MC38 carcinoma growth rate in WT and Nlrp3^−/− mice. (F) MC38 carcinoma Kaplan-Meier survival analysis in WT and Nlrp3^−/− mice. (A, C, and E) Data are expressed as mean tumor volume ±SEM. (A and C) Unpaired two-tailed t test or (D) Mantel-Cox tests were used to determine statistical significance between treatments. (B) PBS vs. treatment. (C) WT vs. Ifnar1^−/−. (E) WT vs. Nlrp3^−/−. ∗p < 0.05, ∗∗p < 0.001, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001. (A and B) n = 5 mice per group. (C and D) WT PBS n = 8, C100 n = 12. Ifnar1^−/− PBS n = 6, C100 n = 6. (E and F) WT PBS + C100 n = 9. Nlrp3^−/− PBS n = 7, C100 n = 8.

Figure 3

C100-induced STING activation requires nuclear DNA damage (A) BMDCs from WT or Tmem173^−/− mice were treated with 8 μg/mL HDCPs for 2 h. Histogram overlay of MitoSOX fluorescence in single live cells. Data are representative of two independent experiments. (B) WT bone marrow precursors were cultured to BMDCs as usual (control [Ctrl]) or in low-dose ethidium bromide (EtBr). Ctrl and EtBr BMDCs were treated with 8 μg/mL HDCPs for 24 h or DMXAA for 3 h. Ifnb mRNA levels were calculated by qPCR with respect to Actb and Rps18. Data are expressed as mean ± SD for technical triplicates with respect to Actb and are representative of 3 independent experiments. Statistical analysis was performed by two-tailed unpaired Student’s t tests with the Holm-Sidak method for multiple comparisons. ∗∗p < 0.01 and ∗∗∗p < 0.001. (C) BMDCs were treated with 5 μg/mL HDCPs for 9 h or 100 μM etoposide for 24 h and then monitored for DNA fragmentation with the APO-BrdU (bromodeoxyuridine) TUNEL kit. Data are representative of two independent experiments. (D) Ctrl and EtBr BMDCs were left untreated or treated with 5 μg/mL C100 for 8 h and then monitored for DNA fragmentation. Data are representative of two independent experiments. (E and F) C57BL/6 mice were injected s.c. on the flank with B16F10 cells. Tumor-bearing mice were randomized and injected i.t. with C100 or intraperitoneally with cisplatin on days 0, 4, and 8. (E) Unpaired two-tailed t test or (F) Mantel-Cox tests were used to determine statistical significance between treatments. ∗p < 0.05 and ∗∗p < 0.001. (E and F) n = 7 mice per group.

Figure 4

C100 preferentially activates IRF and not NF-κB signaling (A) BMDCs were stimulated with C100 (5 μg/mL) or DMXAA (10 μg/mL) for indicated times. Cells were lysed and IκBα levels determined by immunoblot. (B) BMDCs were stimulated with C100 (5 μg/mL), DMXAA (10 μg/mL), or 2′3′-cGAMP (1 μM) for indicated times. mRNA levels were calculated by qPCR for il6 and tnfa with respect to actb. Data show technical triplicate mRNA levels with respect to actb and are representative of three independent experiments. (C‒F) C57BL/6 mice were injected s.c. on the flank with 5 × 10⁵ B16F10 cells. Tumor-bearing mice (volume: 40–60 mm³) were randomized and injected i.t. with PBS, DMXAA, or C100. 24 h later, tumors were isolated with RNA subsequently extracted for gene expression analysis via NanoString. (C) Schematic of study. (D) Heatmaps displaying the relative induction of expression of IRF or NF-κB target genes in treatment vs. PBS Ctrl groups (log2 fold change relative to PBS) measured via NanoString. (E) Representative induction of IRF and NF-κB genes from NanoString dataset. (F) Overall IRF and NF-κB gene fold induction (log₂ fold change relative to PBS − C100 values/DMXAA values) from NanoString dataset. (C‒F) n = 5 mice per group.

Figure 5

i.t. C100 therapy synergizes with systemic α-PD-1 therapy (A‒C) C57BL6/J mice were injected s.c. with 3.5 × 10⁵ MC38 tumor cells. Tumor-bearing mice (volume: 40–60 mm³) were randomized and injected i.t. with PBS or C100 (100 μg) and received either PBS or 250 μg of a monoclonal anti-PD-1 antibody on days 0, 4, and 8. (A and B) MC38 tumor growth rates. (C) MC38 Kaplan-Meier survival analysis. (D‒E) C57BL6/J mice were injected s.c. with 5 × 10⁵ B16F10 melanoma cells. Tumor-bearing mice (volume: 40–60 mm³) were randomized and injected i.t. with PBS or C100 (100 μg) and received either PBS or 250 μg of a monoclonal anti-PD-1 antibody on days 0, 4, and 8. (D) B16 tumor growth rates. (E) B16 Kaplan-Meier survival analysis. An unpaired two-tailed t test was used to determine statistical significance between treatments. (B) Unpaired two-tailed t tests were used to determine statistical significance between treatments. ∗p < 0.05, ∗∗p < 0.001, ∗∗∗p < 0.001, and ∗∗∗∗p <0 .0.0001. (E) A Mantel-Cox test was used to determine statistical significance between the survival curves of α-PD-1 alone vs. C100- and α-PD-1-treated mice, where ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001. (B and C) PBS, PBS/PD-1 n = 9. C100, C100/PD-1 n = 7. (D and E) PBS, C100, C100 + PD-1 n = 7. PBS + PD-1 n = 6.

Figure 6

The therapeutic efficacy of i.t. C100 monotherapy or in combination with systemic α-PD-1 therapy is dependent on CD8⁺ T cells (A) C57BL6/J mice were injected s.c. with 3.5 × 10⁵ MC38 tumor cells. Tumor-bearing mice (volume: 40–60 mm³) were randomized and injected i.t. with PBS or C100 (100 μg) and received either PBS or 250 μg of a monoclonal anti-PD-1 antibody on days 0, 4, and 8. In addition, mice were injected intraperitoneally with either an anti-CD8α-targeted monoclonal antibody to deplete CD8α⁺ T cells or a relevant isotype Ctrl. (B) CD8⁺ T cells were successfully depleted from experimental mice. (C) MC38 tumor growth rates. (D) MC38 Kaplan-Meier survival analysis. (E) Median survival time of experimental mice post-initial treatment. (F) THP-1 cells were stimulated with media, rotenone (5 μM), C100 (5 μg/mL), cGAMP (CDN) (1 μM), or lipopolysaccharide (LPS; 10 ng/mL), for 4 h. Cells were assayed for induction of mitochondrial stress via MitoSOX, with a histogram overlay of MitoSOX fluorescence in single live cells displayed. (G and H) Cells were lysed and measured for STING and β-actin levels by immunoblot. (G) Mean fold STING densitometry compared to β-actin densitometry, n = 3. (C and D) Iso PBS, Iso C100, α-CD8α C100 + PD-1 n = 7. Iso C100 + PD-1, CD8α PBS n = 8. CD8α C100 n = 9.