Deciphering the function of histone modifications in centromere establishment in Caenorhabditis elegans

The centromere is a chromatin region specialized for chromosome segregation. Conserved H3-variant CENP-A and kinetochore proteins are found on all active centromeres. These proteins assemble to form a platform for microtubule binding during cell division. However, centromeric DNA varies greatly among species in terms of size and sequences. Moreover, neocentromeres can form on non-centromeric DNA in endogenous chromosomes and even on exogenously introduced DNA. This suggests that centromere function is not determined by DNA sequences alone. It is proposed that CENP-A and other epigenetic factors specify the centromere location and maintain its function. Several lines of evidence suggest that histone post-translational modifications are some of the important factors regulating centromere function and identity. In C. elegans, DNA injected into the gonad can concatemerize to form extrachromosomal arrays, or artificial chromosomes (ACs), in embryos produced from the injected gonad. These ACs gain segregation ability in a few cell cycles by forming a new centromere (neocentromere). Importantly, C. elegans ACs gain neocentromeres much easier than Human Artificial Chromosomes (HACs). I used a recently developed real-time AC visualization system in C. elegans to monitor AC segregation and studied the role of histone modifications in neocentromere formation. It is found that H3K9ac, H4ac and H3K27me3 are significantly enriched on newly formed C. elegans repetitive ACs when compared to endogenous chromosomes, whereas H3K4me2 on ACs and endogenous chromosomes have similar levels. It is hypothesized that the special histone modification pattern on first-generation ACs in C. elegans may have contributed to its relatively efficient neocentromere formation. By tethering the histone deacetylase, HDA-1, to ACs, H3K9ac and H4ac were depleted on ACs specifically. Neocentromere formation rate of histone acetylations-depleted ACs dropped significantly when compared to wild type. This indicates that histone acetylations may support efficient neocentromere formation in C. elegans. This finding is consistent with a previous study in HACs. We suggest that histone acetylations may promote neocentromere formation by maintaining an open chromatin environment to allow incorporation of CENP-A and other centromeric proteins. On the other hand, H3K27me3 is not related to open chromatin. It is usually correlated with repressed transcription. However, it was previously found to be correlated with CENP-A occurrence in C. elegans. Tethering of the H3K27me3-specific histone methyltransferase, MES-2, to ACs increased H3K27me3 level on both ACs and endogenous chromosomes. The neocentromere formation rate of H3K27me3-enriched ACs increased slightly, but was not statistically significant when compared to wild type. This suggests that H3K27me3 may be favourable for neocentromere formation, possibly by promoting ectopic CENP-A incorporation or facilitating centromere function. Further study is needed to verify this notion. This study demonstrated the use of C. elegans ACs and tethering of specific histone modifiers to ACs as a tool to study the role of histone modifications in neocentromere formation. Our results suggest that histone modifications can contribute to neocentromere formation.published_or_final_versionBiological SciencesMasterMaster of Philosoph