of its HFD dimerization partner TAF10 and our results indicate that incorporation into TFIID is important for coactivation by TAF3 of endogenous and reporter genes. Replacement of its PHD by H3K4me0-binding PHDs drastically reduced TAF3 activation function, whereas replacing with H3K4me3-binding PHDs supported transcription activation. Taken together, the results with the TAF3 mutant proteins support the model that transcriptional activation via TAF3 depends on a stable incorporation of TAF3 into TFIID and specific recognition of H3K4me3-marked promoters. H3T3ph blocks TFIID association with chromatin RA Varier et al A TAF5 Merge TAF5/H2B 4 min TAF5 Merge TAF5/H2B 46 min TAF5 Merge TAF5/H2B 8 min TAF5 Merge TAF5/H2B 44 min Control siRNA Haspin siRNA 42 min 48 min 20 min 46 min 44 min 56 min 42 min 56 min B DAPI GFP H3T3ph Merge DAPI GFP H3T3ph Merge Control siRNA C DAPI GFP H3S10ph Merge Haspin siRNA D Chromosomal versus cytoplasmic GFP-TAF5 staining 1.2 1.0 0.8 0.6 0.4 0.2 0 Control siRNA Haspin siRNA Aurora B inhibitor Aurora B inhibitor DAPI GFP H3T3ph Merge & 2010 European Molecular Biology Organization H3T3ph blocks TFIID association with chromatin RA Varier et al Crosstalk between H3 phosphorylation and methylation Recognition of the H3K4me3 mark by TAF3 and TFIID is subjected to crosstalk at the histone H3 tail. We reported previously that asymmetric dimethylation of H3R2 selectively inhibits TFIID binding to H3K4me3 and that acetylation of H3K9 and H3K14 potentiates TFIID interaction, which most likely depends on the double bromo-domain of TAF1. H3R2me2a has also been shown to reduce the affinity of CHD1 and ING4 for H3K4me3. It is important to note that previous experiments indicated a mutual exclusion of H3K4me3 and H3R2me2a in promoter MedChemExpress Lypressin regions. In contrast to most H3K4me3-binding PHD domains, RAG2-PHD binding to H3K4me3 is enhanced rather than inhibited by dimethylation of H3R2. Similar to the effects of H3R2me2a, we tested whether H3T3ph adjacent to H3K4me3 modulates effector proteins for H3K4me3. Using nuclear extracts, we found that TFIID binding to H3K4me3 is inhibited by H3T3ph. Our H3 peptide-binding assays indicate the H3T3ph effect occurs at the level of the PHD of TAF3. H3T3ph also inhibits H3K4me3 binding of the PHDs of BPTF, PubMed ID:http://www.ncbi.nlm.nih.gov/pubmed/19828836 ING2 and ING4 and H3 tail binding of the PHD of BHC80. A recently reported screen using a PTM-randomized combinatorial H3 peptide library found that phosphorylation of H3T3 resulted in a considerable attenuation of H3 interaction of several PHD modules . Together, this indicates a general inhibitory role of H3T3ph and emphasizes the importance of H3T3ph in regulating the interaction of proteins with chromatin. An important issue for crosstalk between different histone modifications is whether these marks actually cooccur in vivo. Several observations support cooccurrence of H3T3ph and H3K4me3 during mitosis. First of all, it is clear that the H3K4me3 mark persists throughout mitosis. H3T3ph is a very abundant mark during mitosis and H3T3 phosphorylation is dependent on haspin. Structural and functional characterization of the haspin kinase indicate that H3K4me3 reduces substrate recognition and kinase activity, suggesting a regulatory role of H3K4 methylation in deposition of H3T3ph by haspin. However, even though in vitro studies revealed that H3K9 methylation antagonizes H3S10 phosphorylation by 
Aurora B kinase, H3K9me3 and H3S10ph marks have been observed together in vivo using both mass spe