The double PHD finger domain of MOZ/MYST3 induces α-helical structure of the histone H3 tail to facilitate acetylation and methylation sampling and modification.

Histone tail modifications control many nuclear processes by dictating the dynamic exchange of regulatory proteins on chromatin. Here we report novel insights into histone H3 tail structure in complex with the double PHD finger (DPF) of the lysine acetyltransferase MOZ/MYST3/KAT6A. In addition to sampling H3 and H4 modification status, we show that the DPF cooperates with the MYST domain to promote H3K9 and H3K14 acetylation, although not if H3K4 is trimethylated. Four crystal structures of an extended DPF alone and in complex with unmodified or acetylated forms of the H3 tail reveal the molecular basis of crosstalk between H3K4me3 and H3K14ac. We show for the first time that MOZ DPF induces α-helical conformation of H3K4-T11, revealing a unique mode of H3 recognition. The helical structure facilitates sampling of H3K4 methylation status, and proffers H3K9 and other residues for modification. Additionally, we show that a conserved double glycine hinge flanking the H3 tail helix is required for a conformational change enabling docking of H3K14ac with the DPF. In summary, our data provide the first observations of extensive helical structure in a histone tail, revealing the inherent ability of the H3 tail to adopt alternate conformations in complex with chromatin regulators.

Differential effects of garcinol and curcumin on histone and p53 modifications in tumour cells.

BACKGROUND: Post-translational modifications (PTMs) of histones and other proteins are perturbed in tumours. For example, reduced levels of acetylated H4K16 and trimethylated H4K20 are associated with high tumour grade and poor survival in breast cancer. Drug-like molecules that can reprogram selected histone PTMs in tumour cells are therefore of interest as potential cancer chemopreventive agents. In this study we assessed the effects of the phytocompounds garcinol and curcumin on histone and p53 modification in cancer cells, focussing on the breast tumour cell line MCF7. METHODS: Cell viability/proliferation assays, cell cycle analysis by flow cytometry, immunodetection of specific histone and p53 acetylation marks, western blotting, siRNA and RT-qPCR. RESULTS: Although treatment with curcumin, garcinol or the garcinol derivative LTK-14 hampered MCF7 cell proliferation, differential effects of these compounds on histone modifications were observed. Garcinol treatment resulted in a strong reduction in H3K18 acetylation, which is required for S phase progression. Similar effects of garcinol on H3K18 acetylation were observed in the osteosarcoma cells lines U2OS and SaOS2. In contrast, global levels of acetylated H4K16 and trimethylated H4K20 in MCF7 cells were elevated after garcinol treatment. This was accompanied by upregulation of DNA damage signalling markers such as γH2A.X, H3K56Ac, p53 and TIP60. In contrast, exposure of MCF7 cells to curcumin resulted in increased global levels of acetylated H3K18 and H4K16, and was less effective in inducing DNA damage markers. In addition to its effects on histone modifications, garcinol was found to block CBP/p300-mediated acetylation of the C-terminal activation domain of p53, but resulted in enhanced acetylation of p53K120, and accumulation of p53 in the cytoplasmic compartment. Finally, we show that the elevation of H4K20Me3 levels by garcinol correlated with increased expression of SUV420H2, and was prevented by siRNA targeting of SUV420H2. CONCLUSION: In summary, although garcinol and curcumin can both inhibit histone acetyltransferase activities, our results show that these compounds have differential effects on cancer cells in culture. Garcinol treatment alters expression of chromatin modifying enzymes in MCF7 cells, resulting in reprogramming of key histone and p53 PTMs and growth arrest, underscoring its potential as a cancer chemopreventive agent.

Structural diversity in p160/CREB-binding protein coactivator complexes.

Ligand-induced transcription by nuclear receptors involves the recruitment of p160 coactivators such as steroid receptor coactivator 1 (SRC1), in complex with histone acetyltransferases such as CREB-binding protein (CBP) and p300. Here we describe the solution structure of a complex formed by the SRC1 interaction domain (SID) of CBP and the activation domain (AD1) of SRC1, both of which contain four helical regions (Calpha1, Calpha2, Calpha3, and Calpha3' in CBP and Salpha1, Salpha2', Salpha2, and Salpha3 in SRC1). A tight four-helix bundle is formed between Salpha1, Calpha1, Calpha2, and Calpha3 that is capped by Salpha3. In contrast to the structure of the AD1 domain of the related p160 protein ACTR in complex with CBP SID, the sequences forming Salpha2' and Salpha2 in SRC1 AD1 are not involved in the interface between the two domains but rather serve to position Salpha3. Thus, although the CBP SID domain adopts a similar fold in complex with different p160 proteins, the topologies of the AD1 domains are strikingly different, a feature that is likely to contribute to functional specificity of these coactivator complexes.

HnRNP G and Tra2beta: opposite effects on splicing matched by antagonism in RNA binding.

The hnRNP G family comprises three closely related proteins, hnRNP G, RBMY and hnRNP G-T. We showed previously that they interact with splicing activator proteins, particularly hTra2beta, and suggested that they were involved in regulating Tra2-dependent splicing. We show here that hnRNP G and hTra2beta have opposite effects upon the incorporation of several exons, both being able to act as either an activator or a repressor. HnRNP G acts via a specific sequence to repress the skeletal muscle-specific exon (SK) of human slow skeletal alpha-tropomyosin, TPM3, and stimulates inclusion of the alternative non-muscle exon. The binding of hnRNP G to the exon is antagonized by hTra2beta. The two proteins also have opposite effects upon a dystrophin pseudo-exon. This exon is incorporated in a patient to a higher level in heart muscle than skeletal muscle, causing X-linked dilated cardiomyopathy. It is included to a higher level after transfection of a mini-gene into rodent cardiac myoblasts than into skeletal muscle myoblasts. Co-transfection with hnRNP G represses incorporation in cardiac myoblasts, whereas hTra2beta increases it in skeletal myoblasts. Both the cell specificity and the protein responses depend upon exon sequences. Since the ratio of hnRNP G to Tra2beta mRNA in humans is higher in skeletal muscle than in heart muscle, we propose that the hnRNP G/Tra2beta ratio contributes to the cellular splicing preferences and that the higher proportion of hnRNP G in skeletal muscle plays a role in preventing the incorporation of the pseudo-exon and thus in preventing skeletal muscle dystrophy.

Bifunctional antisense oligonucleotides provide a trans-acting splicing enhancer that stimulates SMN2 gene expression in patient fibroblasts.

The multiplicity of proteins compared with genes in mammals owes much to alternative splicing. Splicing signals are so subtle and complex that small perturbations may allow the production of new mRNA variants. However, the flexibility of splicing can also be a liability, and several genetic diseases result from single-base changes that cause exons to be skipped during splicing. Conventional oligonucleotide strategies can block reactions but cannot restore splicing. We describe here a method by which the use of a defective exon was restored. Spinal muscular atrophy (SMA) results from mutations of the Survival Motor Neuron (SMN) gene. Mutations of SMN1 cause SMA, whereas SMN2 acts as a modifying gene. The two genes undergo alternative splicing with SMN1, producing an abundance of full-length mRNA transcripts, whereas SMN2 predominantly produces exon 7-deleted transcripts. This discrepancy is because of a single nucleotide difference in SMN2 exon 7, which disrupts an exonic splicing enhancer containing an SF2ASF binding site. We have designed oligoribonucleotides that are complementary to exon 7 and contain exonic splicing enhancer motifs to provide trans-acting enhancers. These tailed oligoribonucleotides increased SMN2 exon 7 splicing in vitro and rescued the incorporation of SMN2 exon 7 in SMA patient fibroblasts. This treatment also resulted in the partial restoration of gems, intranuclear structures containing SMN protein that are severely reduced in patients with SMA. The use of tailed antisense oligonucleotides to recruit positively acting factors to stimulate a splicing reaction may have therapeutic applications for genetic disorders, such as SMA, in which splicing patterns are altered.

Effects of interferon alpha on the expression of p21cip1/waf1 and cell cycle distribution in carcinoid tumors.

Interferon alpha (IFN-alpha) has been shown to produce antitumor effects in 50-80% of carcinoid tumor patients and has demonstrated anti-proliferative effects in carcinoid tumor cells, but the mechanism is not well established. This study presents evidence that in a carcinoid tumor cell line, Bon1, IFN-alpha increases the expression of p21 and promotes nuclear translocation of endogenous p21. Furthermore, immunoprecipitation experiments demonstrated that p21 formed immuno-complexes with Stat1 and Stat2 in the nucleus of cells. Interferon alpha can decrease G1- and G2-phase cells, but increase S-phase population. The p21 mRNA expression is inversely correlated to the G1 population (r = -0.933, P < 0.05) and positively correlated to the S-phase population (r = 0.901, P < 0.05). In addition, IFN-alpha inhibited cyclin dependent kinases (CDK), CDK2-, CDK3-, CDK4-, and cyclin E- but not cyclin A-associated kinase activities. Immunodepletion of p21 resulted in a significant enhancement of CDK3 kinase activity (approximately 1.6-fold increase). These results suggest that the mechanism of antitumor and cell cycle regulation of IFN-alpha in carcinoid tumors may, at least in part, be p21-dependent. Based on these results, we conclude that IFN-alpha exerts antitumor effects by increased p21 expression in neuroendocrine tumors.