Activation of an LTR12D repeat by HIV and other stress inducers leads to DHRS2-mediated cellular senescence

Abstract

Human endogenous retroviruses (HERVs), remnants of ancient viral infections, constitute approximately 8% of the human genome and are typically epigenetically silenced. However, specific HERV-derived long terminal repeats (LTRs) can be reactivated by endogenous or exogenous stimuli, functioning as cis-regulatory elements that influence host gene expression. Notably, HIV-1 infection induces the activation of distinct HERV loci. Previous work from my host laboratory identified a HERV-derived LTR12D element on chromosome 14 (hg38: chr14:23636315–23636374) that is specifically activated by HIV-1, leading to transcriptional induction of the nearby gene DHRS2. DHRS2 stabilizes p53 by inhibiting MDM2, thereby promoting CDKN1A (p21) expression and triggering cellular senescence. HIV-infected cells exhibit increased LTR12D_DHRS2 transcripts, activation of the p53 pathway, elevated senescence-associated β-galactosidase activity, and enhanced secretion of pro-inflammatory SASP factors such as IL-6, IL-8, and TNF-α. In my thesis, I validated this LTR12D-DHRS2-driven senescence cascade across multiple models, including primary CD4⁺ T cells, monocyte-derived macrophages, and immortalized cell lines infected with HIV-1 or HIV-2. CRISPR/Cas9-mediated DHRS2 knockout significantly reduced senescence markers and SASP factor release, confirming its functional role. Furthermore, macrophage-tropic HIV-1 strains activated DHRS2 in macrophages, underlining the physiological relevance of this pathway. Interestingly, LTR12D_DHRS2 activation also occurred in response to oxidative, mitochondrial, and ER stress, but not in oncogene- or syncytia-induced senescence, suggesting specificity toward certain damage-associated stimuli. However, DHRS2 induction alone was insufficient to fully trigger senescence, indicating a requirement for co-factors or threshold activation levels. Lastly, I identified HSF1, LEF1, and TCF1 as transcriptional repressors of the LTR12D_DHRS2 promoter, implying that their inhibition may permit stress- or infection-induced activation. Together, this work identifies a novel LTR12D-DHRS2-p53-p21 axis driving virus- and stress-induced senescence and highlights the broader role of reactivated endogenous retroelements in the pathology of HIV-associated aging. These findings open new avenues for targeting retroelement-driven dysfunctions in viral infection and age-related diseases.