Insertion of short L1 sequences generates inter-strain histone acetylation differences in the mouse

Beverly Ann G Boyboy,Kenji Ichiyanagi
DOI: https://doi.org/10.1186/s13100-024-00321-0
2024-05-10
Abstract:Background: Gene expression divergence between populations and between individuals can emerge from genetic variations within the genes and/or in the cis regulatory elements. Since epigenetic modifications regulate gene expression, it is conceivable that epigenetic variations in cis regulatory elements can also be a source of gene expression divergence. Results: In this study, we compared histone acetylation (namely, H3K9ac) profiles in two mouse strains of different subspecies origin, C57BL/6 J (B6) and MSM/Ms (MSM), as well as their F1 hybrids. This identified 319 regions of strain-specific acetylation, about half of which were observed between the alleles of F1 hybrids. While the allele-specific presence of the interferon regulatory factor 3 (IRF3) binding sequence was associated with allele-specific histone acetylation, we also revealed that B6-specific insertions of a short 3' fragment of LINE-1 (L1) retrotransposon occur within or proximal to MSM-specific acetylated regions. Furthermore, even in hyperacetylated domains, flanking regions of non-polymorphic 3' L1 fragments were hypoacetylated, suggesting a general activity of the 3' L1 fragment to induce hypoacetylation. Indeed, we confirmed the binding of the 3' region of L1 by three Krüppel-associated box domain-containing zinc finger proteins (KZFPs), which interact with histone deacetylases. These results suggest that even a short insertion of L1 would be excluded from gene- and acetylation-rich regions by natural selection. Finally, mRNA-seq analysis for F1 hybrids was carried out, which disclosed a link between allele-specific promoter/enhancer acetylation and gene expression. Conclusions: This study disclosed a number of genetic changes that have changed the histone acetylation levels during the evolution of mouse subspecies, a part of which is associated with gene expression changes. Insertions of even a very short L1 fragment can decrease the acetylation level in their neighboring regions and thereby have been counter-selected in gene-rich regions, which may explain a long-standing mystery of discrete genomic distribution of LINEs and SINEs.
What problem does this paper attempt to address?