DNMT3a/DNMT3b在CD133+干细胞驱动的肝癌发生中的意义

We have previously identified a liver cancer stem cell (CSC) / tumor-initiating cell (TIC) population that is marked by their CD133 surface phenotype and bearing features that include the ability to self-renew, differentiate, initiate tumors in vivo and resist chemotherapy. We have also recently found CD133 to play a functional role in regulating liver TIC properties. Clinical HCC samples demonstrated the presence of a small CD133+ population, while in contrast CD133 expression in non-tumor was significantly lower if at all present. Interestingly, CD133 is also abundantly expressed in human normal fetal liver and regenerating liver in a mouse model. Despite our growing understanding of the importance and existence of a CD133+ liver TIC population, there remains a lot unknown regarding the underlying molecular mechanism that governs normal CD133 gene regulation and initiation of CD133+ liver TICs in HCC. Though CD133 is known to be dynamically expressed in normal fetal liver to non-tumor liver to HCC, the reason to why only CD133 in liver TICs, but not CD133 in normal fetal liver or regenerating liver, possess tumorigenic potential is not known. By RNA-Seq profiling, we recently identified a significant up-regulation of catalytic DNA methyltransferases, DNMT3a and DNMT3b, in CD133+ liver TICs. In contrast, we have preliminary evidence to show that DNMT3a and DNMT3b expression is unaltered in CD133+ and CD133- cells isolated from normal fetal liver and in a mouse model of active regenerating (CD133+) vs. non-regenerating liver (CD133-). We hypothesize that the specific up-regulation of DNMT3a and DNMT3b in CD133+ liver TIC which results in the altered methylation status in these cells underlies the exclusive tumorigenic potential of the CD133+ subpopulation in liver TICs. We propose a study (1) to characterize the functional role and evaluate the clinical relevance of aberrant DNMT3a and DNMT3b expression in CD133+ liver TICs; (2) to delineate the specificity of DNMT3a and DNMT3b deregulation in CD133+ liver TICs through examining their effects in CD133+ normal fetal liver; and (3) to characterize the targets to which DNMT3a and DNMT3b binds and regulates; as well as to delineate the disparity in global differential DNA methylation patterns between CD133+ liver TICs and CD133+ normal fetal liver cells compared with its CD133- counterparts. Knowledge gained will contribute to the identification of new epigenetic markers and molecular targets, as well as the development of diagnostic and therapeutic targets towards CD133+ liver TICs.

肿瘤干细胞（TICs）被证明是肿瘤发生、转移、耐药及复发的根源，已在多种肿瘤中分离并确认。但其致瘤的机制尤其是TICs表观遗传学调控的特异性尚不明晰。我研究组最新结果表明：DNA甲基转移酶DNMT3a和DNMT3b表达改变及DNA甲基化异质性很可能是CD133+TICs驱动肝癌发生的潜在分子机制。本项目中，我们将深入探讨DNMT3a和DNMT3b高表达在肝癌发生中的临床意义，确认其在诱导肿瘤发生中的生物学功能，ChIP分析寻找其调控的特异靶基因。同时，比较CD133+ TICs、正常胎肝CD133+细胞及对应CD133-细胞间DNA甲基化的整体差异，在全基因组水平揭示CD133+ TICs特异甲基化谱，从而最终阐明CD133+ TICs中由甲基化异常调控介导肝癌发生的分子机制。本项目将对肿瘤干细胞及肿瘤发生机制研究提供新视角，为临床以CD133+TICs为特异靶标的肝癌治疗提供新契机。

本课题组前期研究发现并确认以CD133为标记的癌症干细胞（cancer stem cells, CSCs）细胞亚群在肝癌的发生发展中具有重要作用。随着这一标记的发现，不少研究已经揭示了关于CSCs亚群的相关通路及表观遗传学调控机制，但其DNA甲基化水平调控却鲜有报道。本项目旨在研究DNA甲基化异常对CD133＋CSCs驱动肝细胞癌的调控。我们利用Illumina Infinium Human Methylation 450 BeadChip（HuMet450 BeadChip）技术发现CD133＋和CD133－细胞的基因甲基化水平有明显区别。其中，Reduced expression protein 1 ［REX1, 也被称为鋅指蛋白42（ZFP42）］ 在CD133＋细胞中表现出显著过甲基化。在研究过甲基化对REX1表达的影响时，我们首先发现REX1的mRNA表达水平在CD133+细胞及肝细胞癌 （hepatocellular carcinoma, HCC）临床样本中均降低。经5-aza（DNA甲基化抑制剂）处理后，REX1的表达量在REX1低表达的细胞系中显著上升。经电脑模拟比对及预测，我们选定REX1启动子中一处的CpG island作为具体研究对象。对此CpG island的Pyrosequencing结果显示，REX1的表达与其在HCC细胞系中的甲基化水平呈负相关；且此CpG island在HCC临床样本中的甲基化水平明显高于非肿瘤样品。同时我们在Gene Expression Omnibus （GEO）上获取的另外两组HuMet450 BeadChip数据（GSE54503 and GSE60753）也同样显示REX1启动子CpG island甲基化水平在HCC临床样本中明显高于非肿瘤样品。我们已经建立了慢病毒介导的过表达和敲低HCC细胞系。REX1表达水平与细胞的迁移侵染及集落形成能力，和在体内实验中的肿瘤形成及转移能力呈负相关。我们正利用染色质免疫共沉淀（Chromatin immunoprecipitation sequencing，ChIP-seq）技术研究在HCC细胞中REX1的DNA结合特性。总而言之，本项目研究表明REX1在HCC癌症特性的调控中起重要作用。