目的 本研究在前期生物信息学筛选的基础上,验证含FERM结构域蛋白4A(RMD4A)及纤连蛋白1(FN1)在葡萄膜黑色素瘤(UM)组织中的蛋白表达特征,并初步探讨FRMD4A在UM进展中的潜在生物学功能及调控通路,为阐明UM转移机制提供实验依据。方法 收集经病理确诊的转移性UM、非转移性UM及葡萄膜色素痣石蜡包埋标本,采用免疫组织化学法检测FRMD4A与FN1的蛋白表达水平,比较3组间的表达差异。基于癌症基因组图谱(TCGA)UVM队列的转录组测序数据,采用生物信息学策略模拟FRMD4A基因CRISPR敲除效应,利用DESeq2算法筛选差异表达基因,并进行基因本体论(GO)生物过程富集分析与京都基因与基因组百科全书(KEGG)信号通路富集分析,初步构建FRMD4A的潜在调控网络。结果 免疫组化结果显示,FRMD4A蛋白主要定位于肿瘤细胞胞质;转移性UM组织中FRMD4A阳性表达率显著高于非转移性UM及葡萄膜色素痣组织,组间差异有统计学意义(P<0.05);FN1在3组组织中均未见明显胞质阳性表达,组间表达差异无统计学意义(P>0.05)。FRMD4A基因模拟敲除后共筛选出8 593个显著差异表达基因,其中下调基因数量约为上调基因的1.9倍;功能富集分析提示,差异基因主要涉及免疫细胞功能调控、细胞黏附与迁移、代谢重编程三大模块,核心通路包括细胞黏附分子、T细胞受体信号、NF-κB信号及抗原加工与呈递等。结论 FRMD4A蛋白表达水平在转移组中呈明显升高趋势,其可能通过调控细胞黏附稳态、炎症信号通路及抗肿瘤免疫应答,参与UM的转移进程及免疫抑制微环境的形成,有望成为评估UM恶性程度的潜在生物标志物;FN1在组织蛋白水平的表达差异未获证实,其作为UM转移标志物的临床价值尚需进一步验证。

Objective Based on previous bioinformatics screening, this study aimed to verify the protein expression characteristics of FERM domain-containing protein 4 A(FRMD4 A) and fibronectin 1(FN1) in uveal melanoma(UM) tissues, and to preliminarily explore the potential biological functions and regulatory pathways of FRMD4 A in UM progression, thereby providing experimental evidence for elucidating the mechanisms of UM metastasis. Methods Paraffin-embedded specimens of metastatic UM, non-metastatic UM, and uveal nevus, confirmed by pathology, were collected. Immunohistochemistry(IHC) was employed to detect the protein expression levels of FRMD4 A and FN1, and the expression differences among the three groups were compared. Based on transcriptome sequencing data from The Cancer Genome Atlas(TCGA) UVM cohort, a bioinformatics strategy was adopted to simulate the CRISPR knockout effect of the FRMD4A gene. Differentially expressed genes(DEGs) were screened using the DESeq2 algorithm, followed by Gene Ontology(GO) biological process enrichment analysis and Kyoto Encyclopedia of Genes and Genomes(KEGG) pathway enrichment analysis to preliminarily construct the potential regulatory network of FRMD4A. Results IHC results showed that FRMD4 A protein was mainly localized in the cytoplasm of tumor cells. The positive expression rate of FRMD4 A in metastatic UM tissues was significantly higher than that in non-metastatic UM and uveal nevus tissues, with statistically significant differences among the groups(P<0.05). FN1 showed no obvious cytoplasmic positive expression in the three groups of tissues, and there was no statistically significant difference in expression among the groups(P>0.05). After simulated knockout of the FRMD4 A gene, a total of 8,593 significantly DEGs were screened, with the number of downregulated genes being approximately 1.9 times that of upregulated genes. Functional enrichment analysis suggested that the DEGs were mainly involved in three major modules: immune cell function regulation, cell adhesion and migration, and metabolic reprogramming. Core pathways included cell adhesion molecules, T cell receptor signaling, NF-κB signaling, and antigen processing and presentation. Conclusion The protein expression level of FRMD4 A shows a significant increasing trend in the metastatic group. It may participate in the metastatic process of UM and the formation of an immunosuppressive microenvironment by regulating cell adhesion homeostasis, inflammatory signaling pathways, and anti-tumor immune responses, suggesting its potential as a biomarker for evaluating the malignancy of UM. The difference in FN1 expression at the tissue protein level was not confirmed, and its clinical value as a metastatic marker for UM requires further verification.