Ocular disease-associated mutations diminish the mitotic chromosome retention ability of PAX6.

Transcription factors play a key role in maintaining cell identity. One mechanism of such cell memory after multiple rounds of cell division cycles is through persistent mitotic chromosome binding, although how individual transcription factors achieve mitotic chromosome retention is not completely understood. Here we show that PAX6, a lineage-determining transcription factor, coats mitotic chromosomes. Using deletion and point mutants associated with human ocular diseases in live-cell imaging analysis, we identified two regions, MCR-D1 and MCR-D2, that were responsible for mitotic chromosome retention of PAX6. We also identified three nuclear localization signals (NLSs) that contributed to mitotic chromosome retention independent of their nuclear import functions. Full mitotic chromosome retention required the presence of DNA-binding domains as well as NLSs within MCR-Ds. Furthermore, disease-associated mutations and NLS mutations changed the distribution of intrinsically disordered regions (IDRs) in PAX6. Our findings not only identify PAX6 as a novel mitotic chromosome retention factor but also demonstrate that the mechanism of mitotic chromosome retention involves sequence-specific DNA binding, NLSs, and molecular conformation determined by IDRs. These findings link mitotic chromosome retention with PAX6-related pathogenesis and imply similar mechanisms for other lineage-determining factors in the PAX family.

M33 condenses chromatin through nuclear body formation and methylation of both histone H3 lysine 9 and lysine 27.

Heterochromatin, a type of condensed DNA in eukaryotic cells, has two main categories: Constitutive heterochromatin, which contains H3K9 methylation, and facultative heterochromatin, which contains H3K27 methylation. Methylated H3K9 and H3K27 serve as docking sites for chromodomain-containing proteins that compact chromatin. M33 (also known as CBX2) is a chromodomain-containing protein that binds H3K27me3 and compacts chromatin in vitro. However, whether M33 mediates chromatin compaction in cellulo remains unknown. Here we show that M33 compacts chromatin into DAPI-intense heterochromatin domains in cells. The formation of these heterochromatin domains requires H3K27me3, which recruits M33 to form nuclear bodies. G9a and SUV39H1 are sequentially recruited into M33 nuclear bodies to create H3K9 methylated chromatin in a process that is independent of HP1α. Finally, M33 decreases progerin-induced nuclear envelope disruption caused by loss of heterochromatin. Our findings demonstrate that M33 mediates the formation of condensed chromatin by forming nuclear bodies containing both H3K27me3 and H3K9me3. Our model of M33-dependent chromatin condensation suggests H3K27 methylation corroborates with H3K9 methylation during the formation of facultative heterochromatin and provides the theoretical basis for developing novel therapies to treat heterochromatin-related diseases.

Discovering chromatin motifs using FAIRE sequencing and the human diploid genome.

BACKGROUND: Specific chromatin structures are associated with active or inactive gene transcription. The gene regulatory elements are intrinsically dynamic and alternate between inactive and active states through the recruitment of DNA binding proteins, such as chromatin-remodeling proteins. RESULTS: We developed a unique genome-wide method to discover DNA motifs associated with chromatin accessibility using formaldehyde-assisted isolation of regulatory elements with high-throughput sequencing (FAIRE-seq). We aligned the FAIRE-seq reads to the GM12878 diploid genome and subsequently identified differential chromatin-state regions (DCSRs) using heterozygous SNPs. The DCSR pairs represent the locations of imbalances of chromatin accessibility between alleles and are ideal to reveal chromatin motifs that may directly modulate chromatin accessibility. In this study, we used DNA 6-10mer sequences to interrogate all DCSRs, and subsequently discovered conserved chromatin motifs with significant changes in the occurrence frequency. To investigate their likely roles in biology, we studied the annotated protein associated with each of the top ten chromatin motifs genome-wide, in the intergenic regions and in genes, respectively. As a result, we found that most of these annotated motifs are associated with chromatin remodeling, reflecting their significance in biology. CONCLUSIONS: Our method is the first one using fully phased diploid genome and FAIRE-seq to discover motifs associated with chromatin accessibility. Our results were collected to construct the first chromatin motif database (CMD), providing the potential DNA motifs recognized by chromatin-remodeling proteins and is freely available at http://syslab.nchu.edu.tw/chromatin.