Repetitive Element Transcripts are Associated with Inflammation in Older Adults.

Growing evidence shows that expression of non-coding repetitive sequences, or repetitive element (RE) transcripts, increases with age and may be related to biological aging itself. In fact, RE transcript levels can be used to accurately predict age, and they are reduced by healthy aging interventions. However, the exact mechanistic links between RE dysregulation and aging are unknown. Some evidence indicates that RE transcripts may stimulate age-related inflammation (via formation of RE-derived double-stranded RNA, an innate immune activator) but there is currently no evidence of this in human subjects. Therefore, we determined if serum interleukin 6 (IL-6), a key circulating marker of inflammation, was associated with RE transcript expression (assessed via RNA-seq) in 92 healthy, middle aged to older adults (64 ± 7 years, 47M/ 45F). We found that serum IL-6 was positively related to expression of every major type of RE (LINEs, SINEs, DNA transposons, LTRs, and Satellites; mean r value = 0.4, p < 0.0005). To examine transcriptome/gene expression differences associated with high vs. low serum IL-6, we analyzed differential gene expression in subjects with the highest vs. lowest IL-6 levels (6 per group, age/sex-matched). We then performed gene ontology analysis to determine biological processes associated with differentially expressed genes. Interestingly, we found that higher IL-6 was associated with reduced gene expression signatures of immune function and increased gene expression signatures related to cell death and differentiation. To assess a potential upstream link between RE expression and inflammation, we looked for correlations among RE and innate immune sensor transcript levels. We found a positive relationship between RE expression and levels of IFIH1, the gene that encodes MDA5, a proinflammatory sensor of double-stranded RNA (p < 0.0005, mean r value = 0.5). MDA5 expression correlated weakly with LINE expression (r = 0.36), but more strongly with satellite and LTR expression (r = 0.66), suggesting that these RE types may be involved in age-associated inflammation. These are the first data to show a relationship between IL-6 and global RE expression in older adults, and to characterize the possible mechanisms underlying this relationship.