Immune response to DNA and RNA: structural insights, molecular mechanisms, and therapeutic targeting

Abstract

The recognition of mislocalized DNA and RNA by cGAS–STING, RIG-I/MDA5, the OAS–RNase L axis, and endosomal TLR3/7/8 has emerged as a unifying paradigm linking cancer, autoinflammation, and antiviral immunity. Counterbalancing these sensors is a structurally heterogeneous nuclease repertoire whose distinct substrate specificities, subcellular compartments and pH optima constrain ligand availability in space and time. Disruption of this equilibrium drives disease through two mirror-image mechanisms. In cancer, DNASE1 is inactivated by tumor-derived G-actin, DNASE1L3 is transcriptionally silenced in hepatocellular, colorectal and lung adenocarcinomas, and DNASE2 is upregulated in immunologically “cold” tumors, together permitting neutrophil-extracellular-trap–mediated exclusion of cytotoxic T cells and suppression of cytosolic DNA sensing. In autoimmunity, biallelic loss of DNASE1L3, TREX1, RNase H2, ADAR1 or RNase T2 produces the interferonopathies of systemic lupus and Aicardi–Goutières syndrome. Diagnostically, nuclease-specific cleavage signatures have matured into cell-free DNA fragmentomics validated across 13 cancer types in a 3,021-patient cohort; therapeutically, the field now spans engineered actin-resistant DNASE1/DNASE1L3 biologics, selective TREX1 and ADAR1 inhibitors entering first-in-human evaluation, STING-activating nanomedicines, RNase Fc-fusions such as RSLV-132 in phase 2a lupus, and JAK1/2 inhibition as standard of care in Aicardi–Goutières syndrome. We synthesize this evidence as a two-fate problem: whether an endogenous nucleic acid accumulates at these sensors to drive autoinflammation or is cleared by nucleases before detection is set by the balance between sensor engagement and clearance capacity. The therapeutic corollary acts on the ligand rather than the enzyme, restoring ligand availability where disease is malignant and restoring clearance where disease is self-directed.