Abstract
Over 20% of human X-linked genes escape from X-chromosome inactivation (XCI), and are important contributors to sex differences in gene expression. Candidate factors involved in escape have been identified through enrichment analyses and include both regional as well as promoter-proximal elements; however, functional testing is limited. Using both in vivo and in vitro mouse models, we refine a region of just 2.6 kb of the human escape gene KDM5C as able to drive escape from XCI. Transgenes of mouse Kdm5c escape XCI; however, human KDM5C is one of three escape genes in a more than 200 kb region, so we initially tested a BAC transgene containing a full-length version of the gene with a reporter insertion. Contrary to our expectation, this transgene failed to escape from XCI. To understand why, we moved to a mouse embryonic stem cell system and tested the BAC transgene without the reporter cassette. Despite being separated from other human escape genes, and also being tested in a different species, human KDM5C was able to escape from XCI, suggesting that the reporter integration disrupted or separated critical escape elements. We refined escape-essential sequences to only 2.6 kb including the promoter, exon 1 and contiguous 1.6 kb of the first intron, consistent with previous studies demonstrating local elements are sufficient for escape. Interestingly, dual copy insertions showed higher escape, suggesting that while local elements are important drivers for escape, the size or number of escape genes in a region can boost inactive X expression.