荟萃分析鉴定上皮衍生转录组预测卵巢癌不良临床结局和免疫浸润。

Meta-analysis identifying epithelial-derived transcriptomes predicts poor clinical outcome and immune infiltrations in ovarian cancer.

机构信息

Department of Pharmacy, First Affiliated Hospital of Xinjiang Medical University, Urumqi 830011, China.

Department of Oncology, First Affiliated Hospital of Xinjiang Medical University, Urumqi 830011, China.

出版信息

Math Biosci Eng. 2021 Jul 30;18(5):6527-6551. doi: 10.3934/mbe.2021324.

DOI:10.3934/mbe.2021324

PMID:34517544

Abstract

BACKGROUND

Previous studies revealed that the epithelial component is associated with the modulation of the ovarian tumor microenvironment (TME). However, the identification of key transcriptional signatures of laser capture microdissected human ovarian cancer epithelia remains lacking.

METHODS

We identified the differentially expressed transcriptional signatures of human ovarian cancer epithelia by meta-analysis of GSE14407, GSE2765, GSE38666, GSE40595, and GSE54388. Then we investigated the enrichment of KEGG pathways that are associated with epithelia-derived transcriptomes. Finally, we investigated the correlation of key epithelia-hub genes with the survival prognosis and immune infiltrations. Finally, we investigated the genetic alterations of key prognostic hub genes and their diagnostic efficacy in ovarian cancer epithelia.

RESULTS

We identified 1339 differentially expressed genes (DEGs) in ovarian cancer epithelia including 541upregulated and 798 downregulated genes. We identified 21 (such as E2F4, FOXM1, TFDP1, E2F1, and SIN3A) and 11 (such as JUN, DDX4, FOSL1, NOC2L, and HMGA1) master transcriptional regulators (MTRs) that are interacted with upregulated and the downregulated genes in ovarian tumor epithelium, respectively. The STRING-based analysis identified hub genes (such as CDK1, CCNB1, AURKA, CDC20, and CCNA2) in ovarian cancer epithelia. The significant clusters of identified hub genes are associated with the enrichment of KEGG pathways including cell cycle, DNA replication, cytokine-cytokine receptor interaction, pathways in cancer, and focal adhesion. The upregulation of SCNN1A and CDCA3 and the downregulation of SOX6 are correlated with a shorter survival prognosis in ovarian cancer (OV). The expression level of SOX6 is negatively correlated with immune score and positively correlated with tumor purity in OV. Moreover, SOX6 is negatively correlated with the infiltration of TILs, CD8+ T cells, CD4+ Regulatory T cells, cytolytic activity, T cell activation, pDC, neutrophils, and macrophages in OV. Also, SOX6 is negatively correlated with various immune markers including CD8A, PRF1, GZMA, GZMB, NKG7, CCL3, and CCL4, indicating the immune regulatory efficiency of SOX6 in the TME of OV. Furthermore, SCNN1A, CDCA3, and SOX6 genes are genetically altered in OV and the expression levels of SCNN1A and SOX6 genes showed diagnostic efficacy in ovarian cancer epithelia.

CONCLUSIONS

The identified ovarian cancer epithelial-derived key transcriptional signatures are significantly correlated with survival prognosis and immune infiltrations, and may provide new insight into the diagnosis and treatment of epithelial ovarian cancer.

摘要

背景

先前的研究表明，上皮成分与卵巢肿瘤微环境（TME）的调节有关。然而，用于鉴定激光捕获微解剖的人类卵巢癌上皮的关键转录特征的方法仍有待确定。

方法

我们通过对 GSE14407、GSE2765、GSE38666、GSE40595 和 GSE54388 的元分析，确定了人类卵巢癌上皮的差异表达转录特征。然后，我们研究了与上皮衍生转录组相关的 KEGG 途径的富集情况。最后，我们研究了关键上皮枢纽基因与生存预后和免疫浸润的相关性。最后，我们研究了关键预后枢纽基因的遗传改变及其在卵巢癌上皮中的诊断功效。

结果

我们在卵巢癌上皮中确定了 1339 个差异表达基因（DEGs），其中包括 541 个上调和 798 个下调基因。我们确定了 21 个（如 E2F4、FOXM1、TFDP1、E2F1 和 SIN3A）和 11 个（如 JUN、DDX4、FOSL1、NOC2L 和 HMGA1）主转录调节因子（MTR），它们分别与卵巢肿瘤上皮中的上调和下调基因相互作用。基于 STRING 的分析确定了卵巢癌上皮中的枢纽基因（如 CDK1、CCNB1、AURKA、CDC20 和 CCNA2）。鉴定的枢纽基因的显著聚类与包括细胞周期、DNA 复制、细胞因子-细胞因子受体相互作用、癌症途径和焦点粘附在内的 KEGG 途径的富集相关。SCNN1A 和 CDCA3 的上调和 SOX6 的下调与卵巢癌（OV）的较短生存预后相关。SOX6 的表达水平与免疫评分呈负相关，与 OV 中的肿瘤纯度呈正相关。此外，SOX6 与 OV 中 TILs、CD8+T 细胞、CD4+调节性 T 细胞、细胞毒性、T 细胞激活、pDC、中性粒细胞和巨噬细胞的浸润呈负相关。SOX6 还与各种免疫标志物呈负相关，包括 CD8A、PRF1、GZMA、GZMB、NKG7、CCL3 和 CCL4，表明 SOX6 在 OV 的 TME 中具有免疫调节效率。此外，SCNN1A、CDCA3 和 SOX6 基因在 OV 中发生遗传改变，SCNN1A 和 SOX6 基因的表达水平在卵巢癌上皮中具有诊断功效。