Background: Cutaneous lupus erythematosus (CLE) is a common manifestation of systemic lupus erythematosus (SLE), affecting up to 75% of SLE patients and serving as the initial clinical sign in approximately 25% of cases. CLE is two to three times more prevalent than overt systemic disease. Ultraviolet (UV) radiation, particularly UVB from sunlight or artificial sources (e.g., fluorescent lights), is the most potent and consistently identified environmental trigger for inducing or exacerbating CLE lesions. Molecularly, CLE pathogenesis is driven by aberrant nucleic acid sensing, with the cyclic GMP-AMP Synthase (cGAS)–stimulator of interferon genes (STING) pathway playing a central role. UV-induced keratinocyte apoptosis or DNA damage leads to accumulation of cytoplasmic DNA triggering STING-dependent signaling and robust production of type I interferons (IFNs). This IFN signature drives a self-amplifying inflammatory network involving keratinocytes, plasmacytoid dendritic cells, cytotoxic CD8 ⁺ T cells, and lesional B cells, sustaining chronic skin inflammation. Conventional SLE-prone mouse models (e.g., MRL/lpr, NZB/W) have been used to induce cutaneous lupus-like phenotypes via UVB irradiation; however, their unclear autoimmune mechanisms and incomplete alignment with current therapeutic targets limit their utility for drug evaluation. Trex1 knockout (Trex1 KO) mice provide a genetically defined alternative. TREX1 (three prime repair exonuclease 1) is the major cytosolic exonuclease responsible for degrading aberrant single-stranded and double-stranded DNA, thereby preventing inappropriate activation of the cGAS-STING pathway. In the absence of functional TREX1, accumulation of undegraded self-DNA leads to chronic cGAS-STING activation, sustained type I IFN production, and systemic lupus-like autoimmunity (high anti-dsDNA antibodies, and organ inflammation). Importantly, unchallenged Trex1 KO mice lack spontaneous skin lesions. Based on this, we aimed to establish UVB-induced CLE models with clear mechanistic relevance.

Objectives: This study aimed to develop and characterize UVB-induced cutaneous lupus erythematosus models in Trex1 KO mice that closely recapitulate the key clinical, histological, and molecular features of human CLE on a mechanistically well-defined cGAS-STING/type I interferon hyperactivation background, and to assess the utility of these models for preclinical evaluation of targeted therapies, including JAK inhibition and type I interferon receptor blockade.

Methods: Two UVB irradiation protocols were applied to Trex1 KO mice to generate short-term and medium-term CLE models. In the short-term model, mice received 450 mJ/cm ² UVB daily on days 0, 1, and 2, with disease monitored for at least 15 days. In the medium-term model, the same initial three-day induction (days 0–2) was followed by additional doses of 450 mJ/cm ² every 7 days for four additional weeks (total duration ~37 days). Clinical disease was assessed using a composite dorsal skin score (erythema, edema, scaling, crusting) and skin fold thickness. Therapeutic interventions included oral upadacitinib (JAK1/JAK2 inhibitor, 10 mg/kg BID) and intraperitoneal anti-mouse IFNAR-1 neutralizing antibody (250 μg BIW). Lesional skin was analyzed by quantitative RT-PCR for interferon-stimulated genes (CXCL10, IFN-β) and by H&E histology for inflammatory cell infiltration.

Results: In the short-term model, UVB induced reproducible skin inflammation persisting ≥15 days, with composite scores peaking on day 8, remaining maximal for ~6 days, then gradually declining from day 14. The medium-term model with weekly boosters extended pathology reliably to day 37, though late-stage hair regrowth/melanin provided partial photoprotection. Both models showed significant upregulation of ISGs (CXCL10, IFN-β) and increased dermal inflammatory infiltrates compared with non-irradiated Trex1 KO controls.

Treatment with upadacitinib or anti-IFNAR-1 significantly reduced skin thickness and disease scores in both models. Anti-IFNAR-1 potently suppressed ISG expression across both short- and medium-term models. Upadacitinib, however, showed a marked reduction in ISG expression specifically in the medium-term model, which is consistent with the dependence on sustained JAK inhibition. Histologically, inflammatory infiltrates were significantly decreased by both agents in the medium-term model, whereas their effects were more modest in the short-term model.

Conclusions: We established two UVB-induced CLE models in Trex1 KO mice with distinct temporal profiles, built on a mechanistically defined cGAS-STING/type I IFN hyperactivation background. These models recapitulate clinical, histological, and molecular features of human CLE and respond robustly to pathway-targeted therapies. The short-term model enables rapid efficacy screening, while the medium-term model better captures sustained therapeutic effects on IFN signature, inflammation, and tissue resolution. These Trex1 KO-based models offer a valuable preclinical platform for evaluating novel therapies directed against cGAS-STING signaling, type I interferon production, or downstream JAK-STAT pathways in cutaneous lupus erythematosus.

REFERENCES: NIL.

Acknowledgments: NIL.

Disclosure of Interests: None declared.