Back

Transcription-dependent heterochromatin at the Xist promoter shapes the random choice of the inactive X chromosome

Kanata, E.; Pelaez-Conde, I.; Noviello, G.; Dunkel, I.; Milanowska, L.; Schwaemmle, T.; Bothe, M.; Gjaltema, R. A.; Schulz, E. G.

2026-07-26 genetics
10.64898/2026.07.22.737064 bioRxiv
Show abstract

In female mammals, Xist, the master regulator of X-chromosome inactivation (XCI), is expressed monoallelically. This pattern is established during early embryonic development, when the active Xist allele is chosen at random in each cell. How this choice is made remains incompletely understood. Combining knock-down and overexpression strategies in differentiating mouse embryonic stem cells, which recapitulate the onset of random XCI, we identify a role for the repressive chromatin mark H3K9me3 in the XCI initiation. We show that H3K9me3 accumulates at the promoter-proximal region of the silent Xist allele in female cells as monoallelic expression is established. Unexpectedly, this accumulation requires prior transcription of Xist itself, likely during the initial phase of upregulation, when Xist is frequently transcribed in male cells and from both X chromosomes in females. A repressive function of Xist-dependent H3K9me3 accumulation is supported by our finding that premature, transient Xist overexpression primes an allele for future silencing and skews the choice of the inactive X. Xist-dependent H3K9me3 recruitment does not require its antisense transcript Tsix, which can nonetheless enhance subsequent maintenance of the mark. In addition, the X-linked Xist activator RNF12 counteracts H3K9me3 formation, independently of its known target REX1. Our results thus point to facultative heterochromatin formation as a key contributor to choice at the onset of XCI, where activating and repressing mechanisms are intertwined to establish monoallelic Xist expression.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.