单细胞测序筛选原发和复发胶质母细胞瘤中恒定存在的胶质瘤细胞

Exploring the Difference between Newly Diagnosed and Recurrent Glioblastoma by Screening a Constant Tumor Cell Type via Single-Cell Sequencing

摘要: 目的：胶质母细胞瘤是成人最常见、侵袭性最强的原发性高致死性脑肿瘤。基因表达似乎是预测胶质母细胞瘤患者生存和治疗反应的重要因素。然而，不同患者之间肿瘤细胞的异质性可能会阻碍找到一种癌症的恒定治疗靶点，因为不仅ndGBM和rGBM之间的遗传谱不同，不同患者之间甚至同一患者的肿瘤细胞亚群之间的遗传谱也不同。因此，在不同的ndGBM和rGBM患者中发现一种恒定的肿瘤细胞类型可能是更好地治疗胶质母细胞瘤的一种潜在方法。本文通过单细胞测序(single-cell RNA sequencing, scRNA-seq)筛选恒定的肿瘤细胞类型，比较新诊断的胶质母细胞瘤(ndGBM)和复发性胶质母细胞瘤(rGBM)组的遗传机制。方法：从Gene Expression Omnibus (GEO)数据库(GSE182109)下载3例ndGBM患者(3例样本)和2例rGBM患者(7例样本)胶质母细胞瘤样本的基因表达谱，并使用R包进行scRNA-seq。结果：共有59,400个胶母细胞瘤组织细胞及12,172个胶质瘤细胞最终通过质控，本研究成功筛选出了恒定的胶质瘤细胞亚型。此外，本研究还进行了拟时序分析、基因本体(Gene Ontology, GO)和京都基因与基因组百科全书(Kyoto Encyclopedia of Genes and Genomes, KEGG)富集分析、细胞–细胞相互作用和单细胞调控网络推测聚类。本研究还鉴定了ndGBM和rGBM组的关键基因基因和各组的不同功能。结论：本文筛选了WHO IV级、IDH野生型和甲基化的胶质母细胞瘤中存在于每个样本和每个发展阶段的一类肿瘤细胞亚型。进一步的研究需要通过分子实验来验证分子机制，并针对这些靶点开发诊断方法和药物治疗ndGBM和rGBM。

Abstract: Objective: Glioblastoma is the most common and most aggressive primary highly lethal brain tumor in adults. Genetic expression seems to be an important factor to predict survival and treatment response in glioblastoma. Nevertheless, the heterogeneity of tumor cells among different patients may be an obstruction to find out a constant therapeutical target for one type of cancer, since the genetic profile is different not only between ndGBM and rGBM, but among different patients and even tumor cell subsets in one patient. Thus, finding a constant tumor cell type that exists across different patients in ndGBM and rGBM may be a potential approach to manage glioblastoma better. This article compares the genetic mechanisms between the newly diagnosed glioblastoma (ndGBM) and recurrent glioblastoma (rGBM) groups by screening a constant tumor cell type via single-cell RNA sequencing (scRNA-seq). Method: The gene expression profile of glioblastoma samples obtained in 3 patients (3 samples) with ndGBM, and 2 patients (7 samples) with rGBM was downloaded from the Gene Expression Omnibus (GEO) database (GSE182109), and scRNA-seq was conducted using R package. Result: Fifty-nine thousand and forty cells of glioblastoma tissue with 12,172 glioma cells were finally passed the quality control process, and a constant glioma cell type was screened. Pseudotime analysis, Gene Ontology (GO) and Kyoto Encyclopaedia of Genes and Genomes (KEGG) enrichment analysis, cell-cell interaction, and single-cell regulatory network inference and clustering for the constant tumor cell type were also conducted. Hub genes and different functions for the constant glioma cell type were also identified. Conclusion: Our article screened a kind of tumor cell subtype that exists across each sample and each development stage in IV level, IDH wild type, and methylated glioblastoma. Further studies are required to verify the resulting genes via molecular experiment, as well as to develop diagnostic methods and pharmaceutical therapies aimed at these targets in treating ndGBM and rGBM.