Identification of histone codes and crosstalk in fission yeast. Schizosaccharomyces pombe

Aims: To map histone modifications with unprecedented resolution both globally and locus-specifically, and to link modification patterns to gene expression. Materials & methods: Using correlations between quantitative mass spectrometry and chromatin immunoprecipitation/microarray analyses, we have mapped histone post-translational modifications in fission yeast (Schizosaccharomyces pombe). Results: Acetylations at lysine 9, 18 and 27 of histone H3 give the best positive correlations with gene expression in this organism. Using clustering analysis and gene ontology search tools, we identified promoter histone modification patterns that characterize several classes of gene function. For example, gene promoters of genes involved in cytokinesis have high H3K36me2 and low H3K4me2, whereas the converse pattern is found ar promoters of gene involved in positive regulation of the cell cycle. We detected acetylation of H4 preferentially at lysine 16 followed by lysine 12, 8 and 5. Our analysis shows that this H4 acetylation bias in the coding regions is dependent upon gene length and linked to gene expression. Our analysis also reveals a role for H3K36 methylation at gene promoters where it functions in a crosstalk between the histone methyltransferase Set2KMT3 and the histone deacetylase Clr6, which removes H3K27ac leading to repression of transcription. Conclusion: Histone modification patterns could be linked to gene expression in fission yeast. Overall design: The combined microarray strategy in this study was performed essentially as outlined in Wiren et al, 2005. We used the S. pombe spotted microarrays (Eurogentec, Belgium). For histone modification maps, ChIP-CHIP method was essentially used according to Robyr and Grunstein (2003). Antibodies against H3K9Ac, H3K14Ac, H3K18Ac, H3K23Ac, H3K27Ac, H3K56Ac , H4K5Ac, H4K8Ac, H4K12Ac, H4K16Ac (Suka, Suka et al. 2001; Xu, Zhang et al. 2007) and H3K36Me2 (Millipore) were used. The histone ?H3cter? antibody (Upstate) was used for ChIP according to Wiren et al, 2005. Spotted microarrays were hybridized using Cy3 and Cy5. For normalization of the data (in Series supplementary file) all the different modification channels were divided by average input value and then divided by ?H3cter? value to correct the histone loss calculations followed by 50th percentile normalization. For our cluster analysis we have used variance normalization method. To reduce the variations between different sample preparations we used same extracts for all IPs. The standard S.pombe laboratory strain Hu303(972h-) was used for this study. For Wt, set1D and set2D expression studies we have used Affymetrix genechip yeast Genome 2.0 microarray. Wt, set1D and set2D cells were grown to mid logarithmic phase (5*10 power 6 cells/ml) in rich medium. RNA was extracted and hybridized according to Affymetrix protocol. For each expression profile at least two independent cultures were analyzed with microarray experiments.研究目的：本研究旨在以空前分辨率在全基因组及位点特异性层面绘制组蛋白修饰图谱，并将修饰模式与基因表达建立关联。 材料与方法：本研究借助定量质谱与染色质免疫沉淀（chromatin immunoprecipitation, ChIP）/微阵列分析之间的相关性，对粟酒裂殖酵母（Schizosaccharomyces pombe）的组蛋白翻译后修饰进行了定位鉴定。 研究结果：在该模式生物中，组蛋白H3的赖氨酸9、18和27位点的乙酰化与基因表达呈现最佳正相关。通过聚类分析与基因本体（Gene Ontology, GO）检索工具，我们鉴定出了表征多类基因功能的启动子组蛋白修饰特征模式。例如，参与胞质分裂的基因启动子具有高H3K36me2与低H3K4me2的修饰特征，而参与细胞周期正调控的基因启动子则呈现完全相反的修饰模式。我们检测到H4的乙酰化优先发生于赖氨酸16位点，其次依次为赖氨酸12、8和5位点。本分析显示，编码区的这种H4乙酰化偏好性与基因长度相关，并与基因表达水平存在显著关联。此外，本研究还揭示了基因启动子区域H3K36甲基化的调控功能：该修饰可介导组蛋白甲基转移酶Set2KMT3与组蛋白去乙酰化酶Clr6之间的信号串扰，后者通过去除H3K27位点的乙酰化修饰，最终实现转录抑制。 研究结论：粟酒裂殖酵母中的组蛋白修饰模式可与基因表达建立明确关联。 整体实验设计：本研究采用的组合微阵列策略基本遵循Wiren等人2005年的实验方案。我们使用了比利时Eurogentec公司生产的粟酒裂殖酵母点样微阵列。在组蛋白修饰图谱绘制环节，本研究基本采用Robyr与Grunstein于2003年报道的ChIP-CHIP实验方法。本研究使用的特异性抗体包括：针对H3K9Ac、H3K14Ac、H3K18Ac、H3K23Ac、H3K27Ac、H3K56Ac、H4K5Ac、H4K8Ac、H4K12Ac、H4K16Ac的抗体（Suka等，2001；Xu、Zhang等，2007），以及针对H3K36Me2的抗体（Millipore）。根据Wiren等人2005年的方法，我们使用组蛋白H3羧基端抗体（Upstate）开展染色质免疫沉淀实验。点样微阵列采用Cy3与Cy5荧光染料进行杂交。关于数据归一化（详见数据集系列补充文件），我们首先将所有修饰通道的信号值除以输入样本的平均信号值，再除以组蛋白H3羧基端抗体的信号值以校正组蛋白丢失带来的计算偏差，随后进行50%分位数归一化处理。本研究的聚类分析采用方差归一化方法。为减少不同样本制备过程中的实验偏差，我们使用同一批提取物完成所有免疫沉淀（IP）实验。本研究使用的粟酒裂殖酵母标准实验室菌株为Hu303（972h-）。在野生型（Wt）、set1D与set2D的基因表达分析中，我们使用了Affymetrix酵母基因组2.0基因芯片微阵列。将野生型、set1D与set2D细胞在丰富培养基中培养至对数中期（5×10^6 个细胞/毫升），随后提取总RNA并按照Affymetrix标准实验流程进行杂交。每个基因表达谱至少通过两次独立培养的微阵列实验进行验证。