Wang Ke, Zhong Weibo, Long Zining, Guo Yufei, Zhong Chuanfan, Yang Taowei, Wang Shuo, Lai Houhua, Lu Jianming, Zheng Pengxiang, Mao Xiangming
Department of Urology, Zhujiang Hospital, Southern Medical University, Guangzhou, China.
Department of Urology, The Hospital of Trade-Business in Hunan Province, Changsha, China.
Front Mol Biosci. 2021 Dec 1;8:775304. doi: 10.3389/fmolb.2021.775304. eCollection 2021.
The effects of 5-methylcytosine in RNA (m5C) in various human cancers have been increasingly studied recently; however, the m5C regulator signature in prostate cancer (PCa) has not been well established yet. In this study, we identified and characterized a series of m5C-related long non-coding RNAs (lncRNAs) in PCa. Univariate Cox regression analysis and least absolute shrinkage and selector operation (LASSO) regression analysis were implemented to construct a m5C-related lncRNA prognostic signature. Consequently, a prognostic m5C-lnc model was established, including 17 lncRNAs: , , , , , , , , , , , , , , , , and . The high m5C-lnc score calculated by the model significantly relates to poor biochemical recurrence (BCR)-free survival ( < 0.0001). Receiver operating characteristic (ROC) curves and a decision curve analysis (DCA) further validated the accuracy of the prognostic model. Subsequently, a predictive nomogram combining the prognostic model with clinical features was created, and it exhibited promising predictive efficacy for BCR risk stratification. Next, the competing endogenous RNA (ceRNA) network and lncRNA-protein interaction network were established to explore the potential functions of these 17 lncRNAs mechanically. In addition, functional enrichment analysis revealed that these lncRNAs are involved in many cellular metabolic pathways. Lastly, was selected for experimental validation; it was upregulated in PCa and probably promoted PCa proliferation and invasion . These results offer some insights into the m5C's effects on PCa and reveal a predictive model with the potential clinical value to improve the prognosis of patients with PCa.
近年来，RNA中的5-甲基胞嘧啶（m5C）在各种人类癌症中的作用得到了越来越多的研究；然而，前列腺癌（PCa）中的m5C调节因子特征尚未完全确立。在本研究中，我们鉴定并表征了一系列PCa中与m5C相关的长链非编码RNA（lncRNA）。采用单因素Cox回归分析和最小绝对收缩与选择算子（LASSO）回归分析构建与m5C相关的lncRNA预后特征。因此，建立了一个预后m5C-lnc模型，包括17个lncRNA： ， ， ， ， ， ， ， ， ， ， ， ， ， ， ， 和 。该模型计算出的高m5C-lnc评分与无生化复发（BCR）生存率差显著相关（ <0.0001）。受试者工作特征（ROC）曲线和决策曲线分析（DCA）进一步验证了预后模型的准确性。随后，创建了一个将预后模型与临床特征相结合的预测列线图，并显示出对BCR风险分层有良好的预测效果。接下来，建立竞争性内源性RNA（ceRNA）网络和lncRNA-蛋白质相互作用网络，从机制上探索这17个lncRNA的潜在功能。此外，功能富集分析表明这些lncRNA参与了许多细胞代谢途径。最后选择进行实验验证；它在PCa中上调，可能促进PCa的增殖和侵袭 。这些结果为m5C对PCa的影响提供了一些见解，并揭示了一个具有潜在临床价值的预测模型，以改善PCa患者的预后。
