EMBL-Australia Collaborating Group, Department of Genome Sciences, John Curtin School of Medical Research, Australian National University, Canberra, 2601, Australian Captial Territory, Australia.
CAS Key Laboratory of Computational Biology, CAS-MPG Partner Institute for Computational Biology, Shanghai Institute of Nutrition and Health, Shanghai Institutes for Biological Sciences, University of Chinese Academy of Sciences, Chinese Academy of Sciences, Shanghai, 200031, China.
BMC Biol. 2020 Apr 15;18(1):40. doi: 10.1186/s12915-020-00769-5.
5-Methylcytosine (mC) is a prevalent base modification in tRNA and rRNA but it also occurs more broadly in the transcriptome, including in mRNA, where it serves incompletely understood molecular functions. In pursuit of potential links of mC with mRNA translation, we performed polysome profiling of human HeLa cell lysates and subjected RNA from resultant fractions to efficient bisulfite conversion followed by RNA sequencing (bsRNA-seq). Bioinformatic filters for rigorous site calling were devised to reduce technical noise.
We obtained ~ 1000 candidate mC sites in the wider transcriptome, most of which were found in mRNA. Multiple novel sites were validated by amplicon-specific bsRNA-seq in independent samples of either human HeLa, LNCaP and PrEC cells. Furthermore, RNAi-mediated depletion of either the NSUN2 or TRDMT1 mC:RNA methyltransferases showed a clear dependence on NSUN2 for the majority of tested sites in both mRNAs and noncoding RNAs. Candidate mC sites in mRNAs are enriched in 5'UTRs and near start codons and are embedded in a local context reminiscent of the NSUN2-dependent mC sites found in the variable loop of tRNA. Analysing mRNA sites across the polysome profile revealed that modification levels, at bulk and for many individual sites, were inversely correlated with ribosome association.
Our findings emphasise the major role of NSUN2 in placing the mC mark transcriptome-wide. We further present evidence that substantiates a functional interdependence of cytosine methylation level with mRNA translation. Additionally, we identify several compelling candidate sites for future mechanistic analysis.
5- 甲基胞嘧啶（mC）是 tRNA 和 rRNA 中普遍存在的碱基修饰，但它也广泛存在于转录组中，包括 mRNA 中，在 mRNA 中，它具有尚未完全了解的分子功能。为了探索 mC 与 mRNA 翻译之间的潜在联系，我们对人 HeLa 细胞裂解物进行了多核糖体谱分析，并对所得馏分中的 RNA 进行了高效亚硫酸氢盐转化，随后进行 RNA 测序（bsRNA-seq）。设计了严格的位点调用生物信息学过滤器，以减少技术噪声。
我们在更广泛的转录组中获得了约 1000 个候选 mC 位点，其中大多数位于 mRNA 中。通过在独立的人 HeLa、LNCaP 和 PrEC 细胞样本中进行扩增子特异性 bsRNA-seq，验证了多个新的位点。此外，RNAi 介导的 NSUN2 或 TRDMT1 mC:RNA 甲基转移酶耗竭表明，在大多数测试的 mRNA 和非编码 RNA 中，大多数测试的位点都明显依赖于 NSUN2。mRNA 中的候选 mC 位点富含 5'UTR 并靠近起始密码子，并且嵌入在类似于 NSUN2 依赖的 tRNA 可变环中发现的 mC 位点的局部环境中。在多核糖体谱上分析 mRNA 位点发现，修饰水平，在总体和许多单个位点上，与核糖体结合呈负相关。
我们的发现强调了 NSUN2 在全转录组范围内放置 mC 标记的主要作用。我们进一步提供了证据，证明胞嘧啶甲基化水平与 mRNA 翻译之间存在功能上的相互依赖关系。此外，我们鉴定了几个有前途的候选位点，用于未来的机制分析。
