Study on the expression and function of RAD54L in oral squamous cell carcinoma

Objective To investigate the expression and role of the DNA repair and recombination protein RAD54-like (RAD54L) in oral squamous cell carcinoma (OSCC). Methods Using OSCC-related data from The Cancer Genome Atlas (TCGA) database, the difference in RAD54L expression between OSCC and control samples was analysed using the Mann-Whitney rank sum test, and the potential value of RAD54L mRNA in OSCC diagnosis was assessed using the receiver operator characteristic curve. The correlation between RAD54L expression levels and clinicopathological data of OSCC patients was analysed using the chi-square test. Once OSCC samples were divided into two groups of high and low expression based on the median value of RAD54L mRNA expression, Cox regression analysis was used to compare the prognostic differences between the two groups. The differentially expressed genes between the groups were subsequently screened using the DESeq2 package, and KEGG pathway enrichment analysis was performed using the clusterProfiler package. The correlation between RAD54L mRNA and gene expression in the homologous recombination repair pathway was demonstrated by Spearman correlation analysis. After clarifying the bioinformatics significance of RAD54L, RAD54L knockdown experiments were performed in human oral squamous carcinoma cell line HSC-3, and the knockdown efficiency was verified through real-time quantitative polymerase chain reaction. After transfection, the changes in proliferation, migration, apoptosis, and cycle of HSC-3 cells were assessed by CCK-8, 5-ethynyl-2'-deoxyuridine (EdU) staining, wound healing, apoptosis, and cell cycle assays. Results Bioinformatics analysis showed that the expression of RAD54L mRNA was higher in OSCC than in normal controls (<i>P</i>&lt;0.001) and had a high value in predicting poor prognosis (<i>AUC</i> = 0.927). The high RAD54L expression group was associated with an increased proportion of male patients (<i>P </i>= 0.032), having a higher T-stage (<i>P </i>= 0.040), clinical stage (<i>P </i>= 0.027), and pathological grading (<i>P </i>= 0.013). Once OSCC samples were divided into two groups of high and low expression using the median value of RAD54L mRNA expression, the prognosis of the group with high expression of RAD54L was poorer than that of the group with low expression (<i>P </i>= 0.049). The differentially expressed genes between the high and low RAD54L expression groups two groups were mainly enriched in neuroactive ligand-receptor interactions, cytokine-cytokine receptor interactions, calcium signaling pathway, cell cycle, gastric cancer, extracellular matrix receptor interactions, chemical carcinogenesis-DNA adducts, DNA replication, homologous recombination, and mismatch repair pathways (<i>P</i>&lt;0.05). In the homologous recombination repair pathway, the expression of RAD54L was positively correlated with the expression of BRCA1, BLM, EME1, XRCC2, POLD1, TOPBP1, RAD51, BRIP1, RAD54B, BRCA2, and SYCP3 (<i>P</i>&lt;0.05), and was strongly positively correlated with the expression of BRCA1, BLM, and EME1 (<i>R</i>&gt;0.8, <i>P</i>&lt;0.05). The results of <i>in vitro</i> experiments showed that RAD54L expression was knocked down to approximately 25% in HSC-3 cells (<i>P</i>&lt;0.001). Compared with the control group, the RAD54L knockdown group showed a lower proliferation rate (<i>P</i>&lt;0.05), a lower proportion of EdU-positive cells (<i>P</i>&lt;0.001), a lower proportion of wound closure (<i>P</i>&lt;0.001), a higher proportion of G1-phase cells (<i>P</i>&lt;0.001), a lower proportion of S-phase cells (<i>P</i>&lt;0.001), and a higher proportion of apoptotic cells (<i>P</i>&lt;0.001). Conclusion RAD54L is highly expressed in OSCC and correlates with poor prognosis. Down-regulation of RAD54L expression inhibits the proliferation and migration of HSC-3 cells, promotes apoptosis, and impedes cell cycle progression.