Molecular differential analysis of uterine leiomyomas and leiomyosarcomas through weighted gene network and pathway tracing approaches

Uterine smooth muscular neoplastic growths like benign leiomyomas (UL) and metastatic leiomyosarcomas (ULMS) share similar clinical symptoms, radiological and histological appearances making their clinical distinction a difficult task. Therefore, the objective of this study is to identify key genes and pathways involved in transformation of UL to ULMS through molecular differential analysis. Global gene expression profiles of 25 ULMS, 25 UL, and 29 myometrium (Myo) tissues generated on Affymetrix U133A 2.0 human genome microarrays were analyzed by deploying robust statistical, molecular interaction network, and pathway enrichment methods. The comparison of expression signals across Myo vs UL, Myo vs ULMS, and UL vs ULMS groups identified 249, 1037, and 716 significantly expressed genes, respectively (p ≤ 0.05). The analysis of 249 DEGs from Myo vs UL confirms multistage dysregulation of various key pathways in extracellular matrix, collagen, cell contact inhibition, and cytokine receptors transform normal myometrial cells to benign leiomyomas (p value ≤ 0.01). The 716 DEGs between UL vs ULMS were found to affect cell cycle, cell division related Rho GTPases and PI3K signaling pathways triggering uncontrolled growth and metastasis of tumor cells (p value ≤ 0.01). Integration of gene networking data, with additional parameters like estimation of mutation burden of tumors and cancer driver gene identification, has led to the finding of 4 hubs (JUN, VCAN, TOP2A, and COL1A1) and 8 bottleneck genes (PIK3R1, MYH11, KDR, ESR1, WT1, CCND1, EZH2, and CDKN2A), which showed a clear distinction in their distribution pattern among leiomyomas and leiomyosarcomas. This study provides vital clues for molecular distinction of UL and ULMS which could further assist in identification of specific diagnostic markers and therapeutic targets. Abbreviations UL: Uterine Leiomyomas; ULMS: Uterine Leiomyosarcoma; Myo: Myometrium; DEGs: Differential Expressed Genes; RMA: Robust Multiarray Average; DC: Degree of Centrality; BC: Betweenness of Centrality; CGC: Cancer Gene Census; FDR: False Discovery Rate; TCGA: Cancer Genome Atlas; BP: Biological Process; CC: Cellular Components; MF: Molecular Function; PPI: Protein–Protein Interaction子宫平滑肌源性肿瘤性增生，如良性子宫肌瘤（Uterine Leiomyomas, UL）与转移性子宫平滑肌肉瘤（Uterine Leiomyosarcoma, ULMS），二者具有相似的临床症状、影像学表现及组织学特征，导致临床鉴别诊断颇具难度。 因此本研究旨在通过分子差异分析，筛选出子宫肌瘤向子宫平滑肌肉瘤转化过程中涉及的关键基因与通路。本研究采用稳健统计、分子互作网络及通路富集分析方法，对基于Affymetrix U133A 2.0人类基因组微阵列平台生成的25例子宫平滑肌肉瘤、25例子宫肌瘤及29例子宫肌层（Myometrium, Myo）组织的全局基因表达谱进行了分析。 通过对比Myo vs UL、Myo vs ULMS以及UL vs ULMS三组的基因表达信号，分别鉴定出249、1037和716个显著差异表达基因（p ≤ 0.05）。针对Myo与UL组的249个差异表达基因（Differential Expressed Genes, DEGs）进行的分析证实，细胞外基质、胶原、细胞接触抑制及细胞因子受体等多条关键通路存在多阶段失调，该过程可将正常子宫肌层细胞转化为良性子宫肌瘤（p ≤ 0.01）。 UL与ULMS组间的716个DEGs则被证实可调控细胞周期、细胞分裂相关的Rho GTP酶（Rho GTPases）及磷脂酰肌醇3-激酶（PI3K）信号通路，进而触发肿瘤细胞的失控性生长与转移（p ≤ 0.01）。结合基因网络数据与肿瘤突变负荷估算、癌症驱动基因识别等额外参数，本研究最终筛选出4个核心枢纽基因（JUN、VCAN、TOP2A及COL1A1）与8个瓶颈基因（PIK3R1、MYH11、KDR、ESR1、WT1、CCND1、EZH2及CDKN2A），这些基因在子宫肌瘤与子宫平滑肌肉瘤中的分布模式存在显著差异。 本研究为子宫肌瘤与子宫平滑肌肉瘤的分子鉴别提供了关键线索，可为后续特异性诊断标志物与治疗靶点的开发提供重要参考。 缩写说明：UL：子宫肌瘤（Uterine Leiomyomas）；ULMS：子宫平滑肌肉瘤（Uterine Leiomyosarcoma）；Myo：子宫肌层（Myometrium）；DEGs：差异表达基因（Differential Expressed Genes）；RMA：稳健多阵列平均（Robust Multiarray Average）；DC：中心性度数（Degree of Centrality）；BC：中心性介数（Betweenness of Centrality）；CGC：癌症基因普查（Cancer Gene Census）；FDR：错误发现率（False Discovery Rate）；TCGA：癌症基因组图谱（Cancer Genome Atlas）；BP：生物过程（Biological Process）；CC：细胞组分（Cellular Components）；MF：分子功能（Molecular Function）；PPI：蛋白质-蛋白质相互作用（Protein–Protein Interaction）