How deregulation of chromatin modifiers causes malignancies is of general interest. Here, we show that histone H2A T120 is phosphorylated in human cancer cell lines and demonstrate that this phosphorylation is catalyzed by hVRK1. Cyclin D1 was one of ten genes downregulated upon VRK1 knockdown in two different cell lines and showed loss of H2A T120 phosphorylation and increased H2A K119 ubiquitylation of its promoter region, resulting in impaired cell growth. In vitro, H2A T120 phosphorylation and H2A K119 ubiquitylation are mutually inhibitory, suggesting that histone phosphorylation indirectly activates chromatin. Furthermore, expression of a phosphomimetic H2A T120D increased H3 K4 methylation. Finally, both VRK1 and the H2A T120D mutant histone transformed NIH/3T3 cells. These results suggest that histone H2A T120 phosphorylation by hVRK1 causes inappropriate gene expression, including upregulated cyclin D1, which promotes oncogenic transformation.

Histone acetylation plays a pivotal role in transcriptional regulation, and ATP-dependent nucleosome remodeling activity is required for optimal transcription from chromatin. While these two activities have been well characterized, how they are coordinated remains to be determined. We discovered ATP-dependent histone H2A acetylation activity in Drosophila nuclear extracts. This activity was column purified and demonstrated to be composed of the enzymatic activities of CREB-binding protein (CBP) and SMARCAD1, which belongs to the Etl1 subfamily of the Snf2 family of helicase-related proteins. SMARCAD1 enhanced acetylation by CBP of H2A K5 and K8 in nucleosomes in an ATP-dependent fashion. Expression array analysis of S2 cells having ectopically expressed SMARCAD1 revealed up-regulated genes. Using native genome templates of these up-regulated genes, we found that SMARCAD1 activates their transcription in vitro. Knockdown analysis of SMARCAD1 and CBP indicated overlapping gene control, and ChIP-seq analysis of these commonly controlled genes showed that CBP is recruited to the promoter prior to SMARCAD1. Moreover, Drosophila genetic experiments demonstrated interaction between SMARCAD1/Etl1 and CBP/nej during development. The interplay between the remodeling activity of SMARCAD1 and histone acetylation by CBP sheds light on the function of chromatin and the genome-integrity network.

No abstract

In mouse embryonic stem (mES) cells, ubiquitylation of histone H2A lysine 119 represses a large number of developmental genes and maintains mES cell pluripotency. It has been suggested that a number of H2A ubiquitin ligases as well as deubiquitylases and related peptide fragments contribute to a delicate balance between self-renewal and multi-lineage differentiation in mES cells. Here, we tested whether known H2A ubiquitin ligases and deubiquitylases are involved in mES cell regulation and discovered that Dzip3, the E3 ligase of H2AK119, represses differentiation-inducible genes, as does Ring1B. The two sets of target genes partially overlapped but had different spectra. We found that Dzip3 represses gene expression by orchestrating changes in 3D organization, in addition to regulating ubiquitylation of H2A. Our results shed light on the epigenetic mechanism of transcriptional regulation, which depends on 3D chromatin reorganization to regulate mES cell differentiation.

Regulation of the expression of diverse genes is essential for making possible the complexity of higher organisms, and the temporal and spatial regulation of gene expression allows for the alteration of cell types and growth patterns. A critical component of this regulation is the DNA sequence-specific binding of transcription factors (TFs). However, most TFs do not independently participate in gene transcriptional regulation, because they lack an effector function. Instead, TFs are thought to work by recruiting cofactors, including Mediator complex (Mediator), chromatin-remodeling complexes (CRCs), and histone-modifying complexes (HMCs). Mediator associates with the majority of transcribed genes and acts as an integrator of multiple signals. On the other hand, CRCs and HMCs are selectively recruited by TFs. Although all the pairings between TFs and CRCs or HMCs are not fully known, there are a growing number of established TF-CRC and TF-HMC combinations. In this review, we focused on the most important of these pairings and discuss how they control gene expression.