Centromere Innovations with a Mouse Species

Abstract

Mammalian centromeres direct faithful genetic inheritance at cell division and are typically found in regions of highly repetitive DNA, such as a‐satellite in humans and minor satellite in many mouse species. In the house mouse M. musculus, the 120 bp DNA monomers are nearly identical (~95%) to each other, and CENP‐A nucleosomes containing the centromeric histone H3 variant, CENP‐A, are positioned with their center position, the nucleosomal dyad, within the DNA recognition element for the sequence‐specific DNA‐binding protein, CENP‐B. We find that CENP‐B‐/‐ mice have precisely the same positioning of CENP‐A nucleosomes as CENP‐B+/+ mice, indicating a potential role for DNA sequence in the positioning of CENP‐A nucleosomes. M. pahari diverged from M. musculus ~ 4 M years ago and has been reported to have a different general DNA sequence organization at centromeres, so we sought to determine the features of centromeric chromatin that were conserved between these species. We found that, on most M. pahari centromeres, CENP‐B is of very low abundance but is extremely enriched on a single chromosome. Using multiple genomic strategies, we identified the most abundant centromere repeat in M. pahari that we term pi‐satellite, which is ~189 bp long and organized into tandem repeat arrays. Chromosomes harboring pi‐satellite lack CENP‐B boxes entirely. Additionally, we employed a more sophisticated strategy to identify a less abundant centromere satellite that contains the CENP‐B box. This variant exhibits greater homogenization, is 188 bp long, and is found exclusively on the single centromere containing extremely high levels of CENP‐B. This centromere also recruits more constitutive heterochromatin, marked by H3K9me3, consistent with a role for CENP‐B in driving pericentromeric heterochromatin. Together, these findings reveal that a mouse species is undergoing dramatic molecular differences directed by evolutionary innovations in centromere DNA sequence.