microRNA-25 drives initial resistance to immune checkpoint therapy by repressing innate and humoral immunity via Syndecan3 [single_clone_RNA_seq_B16]

Immune Checkpoint Therapy (ICT) has demonstrated durable responses and long-lasting immunologic memory in cancer treatment. However, overcoming primary and acquired resistance remains a major challenge. Here, we show that CRISPR-Cas9-mediated deletion of miRNA-25 (miR-25) sensitizes tumors to ICT across three syngeneic mouse tumor models. Single-cell RNA sequencing (scRNA-seq) of the tumor microenvironment (TME) revealed that miR-25 deficiency induces innate immunity by upregulating major histocompatibility complex class II (MHC II) in antigen-presenting M1-like macrophages and enhances the classical complement cascade in cancer-associated fibroblasts (CAFs) to drive a humoral immune response. The complement activation polarizes CAFs from myofibroblastic CAFs (myCAFs) toward inflammatory CAFs (iCAFs) while simultaneously reduces immune-suppressive interactions between CAFs and tumor associated macrophages (TAMs). This shift results in a reduced macrophage population and fosters a pro-inflammatory, anti-tumor TME. Syndecan-3 (Sdc3), a membrane proteoglycan expressed in tumors, is repressed by miR-25 through miRISC (microRNA induced silencing complex) upon IFN-? exposure. Using an adenine base editor (ABE8e) to mutate the miR-25 binding site in the 3' untranslated region (3' UTR) of Sdc3 effectively overcomes the resistance. The repression of SDC3 by miR-25 is further validated in five human cancer cell lines upon IFN-? exposure but remains unaffected in non-cancerous cells. These findings identify miR-25 as a key driver of initial resistance through the repression of SDC3 and demonstrate that miR-25 deletion or stabilization of SDC3 could transform immune resistant "cold" tumors into immune responsive "hot" tumors, offering therapeutic avenues to enhance cancer immunotherapy. Overall design: We selected independent single-clone B16 cells harboring homozygous miR-25 knockout and compared their transcriptomic profiles with wild-type B16 cells and pooled miR-25 knockout cells.