DNA sequence-dependent heterochromatin microdomain formation. DNA sequence-dependent heterochromatin microdomain formation

Cell-type specific gene expression programs are established by distinct chromatin state patterns that involve thousands of heterochromatin microdomains of ~1-2 kb in size marked by di- and trimethylation of histone H3 at lysine 9 (H3K9me2/me3). However, no theoretical framework exists to predict the location and boundaries of such domains from the DNA sequence. Here, we compare H3K9me2/me3-heterochromatin microdomains in mouse embryonic stem cells (ESCs) that are dependent on the histone methylases SUV39H1/2 and GLP, transcription factor ADNP or chromatin remodeler ATRX. By applying a novel Ising-type chromatin hierarchical lattice (ChromHL) model, we identify two different microdomain types that are distinct with respect to their dependence on DNA sequence motifs or nucleosome interactions. ChromHL is able to predict microdomain location, extension and boundaries based on binding sites of PAX3, PAX9, ADNP, CTCF and repeat sequence motifs, the concentration of heterochromatin protein 1 (HP1) and the strength of nucleosome-nucleosome interactions. Thus, our result provides insight how distinct patterns of silenced heterochromatin states are implemented and regulated. Overall design: Wild type murine embryonic stem cell (ESC) wt26 and two ATRX knock out cell lines (KO1-40 and KO1-45) were constructed as described previously (Sadic et al., EMBO Rep, 2015). Cells were cultured on 0.2% v/v gelatine (in PBS) in high glucose DMEM (gibco 31053-028) supplemented with 1 mM sodium pyruvate and 4 mM L-Glutamine (PAA M11-006), 15% v/v FCS (Sigma F7524, lot: 091M3398), 1% v/v Penicillin-Streptomycin (PAN Biotech P06-07100), 100 µM β-mercaptoethanol (Sigma 63689), 1% v/v non-essential amino acids and 0.41% v/v LIF (self-made; supernatant from LIF-producing cells; batch: 7/26/14). Chromatin immunoprecipitation. Cells were fixed in 1% formaldehyde (Thermo Scientific 28906) in PBS for 10 minutes and stopped by addition of glycine to a final concentration of 125 mM. Cells were washed twice with PBS/0.5 mM PMSF, dissolved in swelling buffer (25 mM Hepes pH7.8, 1 mM MgCl2, 10 mM KCl, 0.1% NP-40, 1 mM DTT, 0.5 mM PMSF, protease inhibitor COMPLETE from Roche) and incubated on ice for 10 minutes. After pelleting they were digested with MNase at 37°C for 15 minutes in a buffer containing 25 mM KCl, 4 mM MgCl2, 1 mM CaCl2, 50 mM Tris/HCl pH7.4 and 1x protease inhibitor from Cell Signalling. Chromatin was fragmented with a Covaris S2 sonicator (parameters: 900 s, burst 200, cycle 20%, intensity 8) in sonication buffer (10 mM Tris pH 8.0, 200 mM NaCl, 1 mM EDTA, 0.5% N-lauroylsarcosine, 0.1% Na-deoxycholate) and centrifuged at 13000 rpm and 4°C for 15 minutes. The supernatant was mixed with 4 µg normal rabbit IgG (R&D Systems, AB-105-C, lot: ER1212071) and incubated with ChIP-grade protein G magnetic beads (Cell Signalling 9006S, 25 µl/sample) at 4°C for 1:45 h (pre-clearance). 1/20 of the supernatant was used as input sample and the remaining material was used for the IP reaction by adding 4 µg of anti-H3K9me3 antibody (Abcam, ab8898, lot: GR148830-2). The sample was incubated for 2 h at 4°C, protein G magnetic beads were added and incubated at 4°C overnight. On the next day, the IP samples were washed with sonication buffer, high-salt buffer (50 mM Hepes pH 7.9, 500 mM NaCl, 1 mM EDTA, 1% Triton X-100, 0.1% Na-deoxycholate, 0.1% SDS), Li buffer (20 mM Tris pH 8.0, 1 mM EDTA, 250 mM LiCl, 0.5% NP-40, 0.5% Na-deoxycholate) and finally twice with TE buffer (10 mM Tris pH 8.0, 1 mM EDTA). Subsequently, the DNA-antibody complexes were eluted by incubation with a buffer containing 50 mM Tris pH 8.0, 1 mM EDTA, 1% SDS, 50 mM NaHCO3 at 37°C for 15 minutes. IP samples and input control were treated with RNase A added to a final concentration of 10 µg/ml for one hour at 37°C. Then Proteinase K (Genaxxon Bioscience M3037) and NaCl were added to final concentrations of 80 µg/ml and 200 mM, respectively and samples were incubated over night at 65°C for protein digestion and reversal of crosslinks. DNA was precipitated and the input samples were used to assess the fragment size by gel electrophoresis on a 2% agarose E-Gel and on a DNA1000 bioanalyzer ChIP. Experiments were conducted for two replicates of each cell line.

细胞类型特异性基因表达程序由独特的染色质状态模式所确立，这类模式包含数以千计的~1-2kb大小的异染色质微结构域，这些微结构域以组蛋白H3赖氨酸9的二甲基化和三甲基化（H3K9me2/me3）作为标记。然而，目前尚无能够基于DNA序列预测此类结构域位置与边界的理论框架。
本研究针对依赖组蛋白甲基转移酶SUV39H1/2与GLP、转录因子ADNP或染色质重塑因子ATRX的小鼠胚胎干细胞（ESCs）中的H3K9me2/me3异染色质微结构域展开比较分析。通过应用新型伊辛型染色质层级晶格（ChromHL）模型，本研究鉴定出两类不同的微结构域，二者在对DNA序列基序或核小体相互作用的依赖特性上存在显著差异。
ChromHL模型能够基于PAX3、PAX9、ADNP、CTCF结合位点与重复序列基序、异染色质蛋白1（HP1）浓度以及核小体间相互作用强度，预测微结构域的位置、延伸范围与边界。因此，本研究结果阐明了沉默型异染色质状态的独特模式是如何建立并被调控的。
实验总体设计：按照此前研究（Sadic等，《EMBO Reports》，2015年）的方法，构建了野生型小鼠胚胎干细胞（ESC）wt26以及两株ATRX敲除细胞系（KO1-40与KO1-45）。
细胞培养于添加了以下组分的高糖杜氏改良伊格尔培养基（DMEM，赛默飞Gibco 31053-028）中：0.2%体积比的明胶（溶于磷酸盐缓冲液PBS）、1mM丙酮酸钠、4mM L-谷氨酰胺（L-Glutamine，PAA M11-006）、15%体积比胎牛血清（FCS，Sigma F7524，批号：091M3398）、1%体积比青霉素-链霉素（PAN Biotech P06-07100）、100μM β-巯基乙醇（β-mercaptoethanol，Sigma 63689）、1%体积比非必需氨基酸，以及0.41%体积比自制白血病抑制因子（LIF，由LIF分泌细胞上清制备，批次：7/26/14）。
染色质免疫沉淀（ChIP）实验：将细胞置于含1%甲醛（Thermo Scientific 28906）的PBS中固定10分钟，随后加入甘氨酸至终浓度125mM以终止固定反应。用含0.5mM苯甲基磺酰氟（PMSF）的PBS洗涤细胞两次，将细胞重悬于肿胀缓冲液（25mM羟乙基哌嗪乙硫磺酸（Hepes）pH7.8、1mM氯化镁、10mM氯化钾、0.1% NP-40、1mM二硫苏糖醇（DTT）、0.5mM PMSF、罗氏Complete蛋白酶抑制剂）中，冰上孵育10分钟。离心收集细胞后，在含25mM氯化钾、4mM氯化镁、1mM氯化钙、50mM Tris/HCl pH7.4以及1×Cell Signalling蛋白酶抑制剂的缓冲液中，用微球菌核酸酶（MNase）于37℃消化15分钟。使用Covaris S2超声破碎仪（参数：900秒，脉冲200，循环占比20%，强度8）在超声缓冲液（10mM三羟甲基氨基甲烷（Tris）pH8.0、200mM氯化钠（NaCl）、1mM乙二胺四乙酸（EDTA）、0.5% N-月桂酰肌氨酸钠、0.1%脱氧胆酸钠）中对染色质进行片段化，随后于4℃、13000rpm离心15分钟。取上清液，加入4μg正常兔IgG（R&D Systems，AB-105-C，批号：ER1212071），并与ChIP级蛋白G磁珠（Cell Signalling 9006S，每样品25μl）于4℃孵育1小时45分钟进行预澄清。取1/20体积的上清液作为Input样品，剩余上清液中加入4μg抗H3K9me3抗体（Abcam，ab8898，批号：GR148830-2）用于免疫沉淀（IP）反应。将样品于4℃孵育2小时后，加入蛋白G磁珠并于4℃过夜孵育。
次日，用超声缓冲液、高盐缓冲液（50mM Hepes pH7.9、500mM氯化钠（NaCl）、1mM EDTA、1% Triton X-100、0.1%脱氧胆酸钠、0.1%十二烷基硫酸钠（SDS））、氯化锂（LiCl）缓冲液（20mM Tris pH8.0、1mM EDTA、250mM氯化锂（LiCl）、0.5% NP-40、0.5%脱氧胆酸钠）洗涤IP样品，最后用TE缓冲液（Tris-EDTA缓冲液，10mM Tris pH8.0、1mM EDTA）洗涤两次。随后，将DNA-抗体复合物置于含50mM Tris pH8.0、1mM EDTA、1% SDS、50mM碳酸氢钠的缓冲液中，于37℃孵育15分钟进行洗脱。IP样品与Input对照样品均加入核糖核酸酶A（RNase A）至终浓度10μg/ml，于37℃孵育1小时。随后分别加入蛋白酶K（Proteinase K，Genaxxon Bioscience M3037）与氯化钠（NaCl）至终浓度80μg/ml和200mM，将样品于65℃过夜孵育以完成蛋白质消化与交联逆转。沉淀回收DNA，取部分Input样品通过2%琼脂糖E-Gel凝胶电泳以及DNA1000生物分析仪（ChIP试剂盒配套）检测片段大小。每个细胞系均设置两次生物学重复实验。