Dual chromatin recognition by the histone deacetylase complex HCHC is required for proper DNA methylation in Neurospora crassa.

DNA methylation, heterochromatin protein 1 (HP1), histone H3 lysine 9 (H3K9) methylation, histone deacetylation, and highly repeated sequences are prototypical heterochromatic features, but their interrelationships are not fully understood. Prior work showed that H3K9 methylation directs DNA methylation and histone deacetylation via HP1 in Neurospora crassa and that the histone deacetylase complex HCHC is required for proper DNA methylation. The complex consists of the chromodomain proteins HP1 and chromodomain protein 2 (CDP-2), the histone deacetylase HDA-1, and the AT-hook motif protein CDP-2/HDA-1-associated protein (CHAP). We show that the complex is required for proper chromosome segregation, dissect its function, and characterize interactions among its components. Our analyses revealed the existence of an HP1-based DNA methylation pathway independent of its chromodomain. The pathway partially depends on CHAP but not on the CDP-2 chromodomain. CDP-2 serves as a bridge between the recognition of H3K9 trimethylation (H3K9me3) by HP1 and the histone deacetylase activity of HDA-1. CHAP is also critical for HDA-1 localization to heterochromatin. Specifically, the CHAP zinc finger interacts directly with the HDA-1 argonaute-binding protein 2 (Arb2) domain, and the CHAP AT-hook motifs recognize heterochromatic regions by binding to AT-rich DNA. Our data shed light on the interrelationships among the prototypical heterochromatic features and support a model in which dual recognition by the HP1 chromodomain and the CHAP AT-hooks are required for proper heterochromatin formation.

Single telomere length analysis in Ustilago maydis, a high-resolution tool for examining fungal telomere length distribution and C-strand 5'-end processing.

Telomeres play important roles in genome stability and cell proliferation. Telomere lengths are heterogeneous and because just a few abnormal telomeres are sufficient to trigger significant cellular response, it is informative to have accurate assays that reveal not only average telomere lengths, but also the distribution of the longest and shortest telomeres in a given sample. Herein we report for the first time, the development of single telomere length analysis (STELA) - a PCR-based assay that amplifies multiple, individual telomeres - for Ustilago maydis, a basidiomycete fungus. Compared to the standard telomere Southern technique, STELA revealed a broader distribution of telomere size as well as the existence of relatively short telomeres in wild type cells. When applied to blm∆, a mutant thought to be defective in telomere replication, STELA revealed preferential loss of long telomeres, whose maintenance may thus be especially dependent upon efficient replication. In comparison to blm∆, the trt1∆ (telomerase null) mutant exhibited greater erosion of short telomeres, consistent with a special role for telomerase in re-lengthening extra-short telomeres. We also used STELA to characterize the 5' ends of telomere C-strand, and found that in U. maydis, they terminate preferentially at selected nucleotide positions within the telomere repeat. Deleting trt1 altered the 5'-end distributions, suggesting that telomerase may directly or indirectly modulate C-strand 5' end formation. These findings illustrate the utility of STELA as well as the strengths of U. maydis as a model system for telomere research.