STING agonists drive recruitment and intrinsic type I interferon responses in monocytic lineage cells for optimal anti-tumor immunity

The cyclic GMP-AMP synthase-stimulator of interferon genes (cGAS-STING) pathway, a sensor of cytosolic DNA, orchestrates the production of proinflammatory cytokines, chemokines, and type I interferons (IFN-Is), thereby contributing to spontaneous tumor surveillance. Intratumoral delivery of synthetic STING agonists induces IFN-I-dependent tumor regression in preclinical cancer models and is being tested clinically. In this study, we investigate the role of monocytic lineage cells (MCs) in response to STING agonist induced IFN-I signaling. We show that CCR2-deficient mice, lacking inflammatory MCs in the periphery, or Lyz2-Cre-IFNAR1fl/fl mice in which IFN-I signaling in monocytes is reduced, exhibit impaired responses to STING agonist therapy of MC38 and/or B16F10 tumors. STING agonist treatment induced CCR5-dependent migration of MCs carrying tumor antigen from the tumor to the lymph nodes. Single-cell RNA sequencing of CD45+ cells from lymph nodes and tumors of mice in which half the hematopoietic cells lack the IFNAR1 (interferon alpha/beta receptor 1) revealed that STING agonist therapy induces intrinsic IFNAR1-dependent acquisition of an inflammatory monocytic cell phenotype distinct from inflammatory classical dendritic cells (cDC) and a reduction in macrophages with a protumor TGFβ/angiogenesis transcriptome. Interleukin (IL)-18-IL-18R1 interaction was the top predicted interaction between monocytic lineage cells and CD8+ T cells or natural killer cells. Blocking IL-18 reduced IFN-γ production by CD8 T cells in lymph nodes and decreased the therapeutic efficacy of STING agonist treatment in Ccr2+/+ but not in Ccr2-/- mice. These findings support a pivotal role for IL-18 producing inflammatory monocytic lineage cells in CD8+ T cell control of melanoma following STING agonist treatment. Mice were reconstituted with a 1:1 mix of Ifnar1+/+ CD45.1 and Ifnar1-/- CD45.2 bone marrow cells, and chimerism in the blood was confirmed after 90 days. Ifnar1 mixed bone marrow chimeras were inoculated with 10^6 B16.OVA cells intradermally and a single dose of ADU-S100 was administered intratumorally on day 12. LNs and tumors were harvested at day 1 and day 2 post-treatment, respectively. For both the tumor and LN, equal amounts of cell suspension from 3 vehicle or 3 ADU-S100-treated mice were pooled into one vehicle sample and one ADU-S100 sample. Cells were stained with fluor-conjugated antibodies (for FACS) and TotalSeq-A antibodies (Biolegend) in D-PBS 2% FBS for 25 min on ice. Vehicle and ADU-S100 treated samples were labeled with distinct hashtag antibodies, while anti-CD45.1 and anti-CD45 barcoded antibodies were used to label Ifnar1+/+ and Ifnar1-/- cells. Cells were sorted on a BD FACS Aria IIu (tumor: live CD45+ cells; LN: live CD3- CD19- CD11b+ and/or MHCII+ cells). Vehicle and ADU-S100-treated samples were pooled together at a 1:1 ratio immediately before partitioning on the 10X chromium controller. The Chromium Single Cell 3’ Reagent Kits v3.1 were used according to the manufacturer’s user guide in conjunction with CITE-seq & Cell Hashing Protocol from the New York Genome Center Technology Innovation Lab.