Epigenetic engineering shows that a human centromere resists silencing mediated by H3K27me3/K9me3

Nuno M C Martins, Jan H Bergmann, Nobuaki Shono, Hiroshi Kimura, Vladimir Larionov, Hiroshi Masumoto, William C Earnshaw

Research output: Contribution to journalArticlepeer-review


Centromeres are characterized by the centromere-specific H3 variant CENP-A, embedded in chromatin with a pattern characteristic of active transcription that is required for centromere identity. It is unclear how centromeres remain transcriptionally active despite being flanked by repressive pericentric heterochromatin. To further understand centrochromatin's response to repressive signals, we nucleated a polycomb-like chromatin state within the centromere of a HAC (Human Artificial Chromosome) by tethering the methyltransferase EZH2. This led to deposition of the H3K27me3 MARK and PRC1 repressor binding. Surprisingly, this state did not abolish HAC centromere function or transcription, and this apparent resistance was not observed on a non-centromeric locus, where transcription was silenced. Directly tethering the READER/repressor PRC1 bypassed this resistance, inactivating the centromere. We observed analogous responses when tethering the heterochromatin EDITOR Suv39h1-methyltransferase-domain (centromere resistance) or READER HP1α (centromere inactivation), respectively. Our results reveal that the HAC centromere can resist repressive pathways driven by H3K9me3/H3K27me3 and may help to explain how centromeres are able to resist inactivation by flanking heterochromatin.

Original languageEnglish
Pages (from-to)177-96
Number of pages20
JournalMolecular Biology of the Cell
Issue number1
Early online date12 Nov 2015
Publication statusPublished - 1 Jan 2016


Dive into the research topics of 'Epigenetic engineering shows that a human centromere resists silencing mediated by H3K27me3/K9me3'. Together they form a unique fingerprint.

Cite this