Retrotransposon reactivation to elicit innate immunity in squamous cancers while sparing normal tissues

Transposons make up over 40% of the human genome, and while most are inactive, a few retrotransposons remain capable of jumping, contributing to mutations and genetic variation. Squamous cell carcinomas (SCCs) including those from the head and neck, esophagus, lung, cervix, anogenital sites and skin show particularly high retrotransposon activity among all cancers analyzed, yet their pathogenic mechanisms and therapeutic potential remain poorly understood. We tackle these challenges by using a genetic model lacking a crucial epigenetic repressor of endogenous retroviruses (ERVs, a type of retrotransposons resembling retroviruses) in the adult murine skin. We saw in our model robust ERV induction, accompanied by retroviral peptides production and viral like particles assembly, leading to stem cell exhaustion and hair loss phenotypes, which can be reversed in vivo by antiviral drugs that inhibit reverse transcriptase (Lyu Y et al, Ge Y. Cell. 2024). In the current study, we dissect host mechanisms restraining and responding to ERV reactivations, and leverage these pathways to inhibit SCCs while sparing normal tissues. Retrotransposon reactivation elicits host antiviral responses in the tumors, known as viral mimicry, which we recapitulated in the skin. Interestingly, we observed two genetically and biochemically diverging pathways elicited by ERVs versus SINEs (short interspersed elements, another type of retrotransposons), that are responsible for skin pathogenesis and tumor progression. Moreover, while retrotransposon reactivations are widely associated with antiviral responses in cancer patients, such correlations remain context dependent, and their decoupling often occurs unexplained. We saw intriguing decoupling of retrotransposon induction and antiviral response in our model. We further noted that distinct heterochromatin repressors cooperate to not only dictate retrotransposon-independent antiviral responses, but also blunt antiviral responses despite the presence of retrotransposon reactivations. Finally, we propose retrotransposon related SCC specific vulnerabilities that can be tackled while sparing normal tissues.