Background
Mitochondrial fusion is a vital cellular process in eukaryotic cells, playing a significant role in preserving cell function. The impact of mitochondrial fusion extends to various aspects of cellular activity, including energy production, stress response, and cell survival. A growing body of research is delving into the role of mitochondrial fusion in the emergence and advancement of cancer. As the runner-up in terms of prevalence among renal cell carcinoma types, kidney renal papillary cell carcinoma (KIRP) exhibits a diverse array of prognostic outcomes. Uncovering new prognostic biomarkers for KIRP is essential to enhancing patient recovery and tailoring individualized therapeutic approaches.
Objective
In this study, our goal is to examine the gene expression associated with mitochondrial fusion and construct a novel gene signature model for predicting the prognosis of kidney renal papillary cell carcinoma (KIRP), thereby contributing to improved clinical outcomes.
Methods
We gathered RNA sequencing information and associated clinical data for 285 individuals with kidney renal papillary cell carcinoma (KIRP) from The Cancer Genome Atlas (TCGA) database. In order to create a gene signature panel for risk identification, we utilized LASSO regression analysis and multivariate Cox regression analysis on differentially expressed genes (BNIP3, GDAP1, MIEF2, and PRKN) associated with mitochondrial fusion. To predict immunotherapeutic responses in KIRP tumors, we conducted an array of assessments including scores for checkpoint inhibitor immunotherapy, tumor mutation burden (TMB), TIDE, and the tumor microenvironment (TME). This was integrated with our work predicting chemotherapeutic responses based on RNA-sequencing expression profiles and related clinical data from the TCGA dataset. By utilizing the GDSC database and the R package "prophetic", we estimated each sample's IC50 via ridge regression, considered combat batch effects and tissue types, and summarized duplicate gene expression as mean values. All computations were conducted within the R foundation's version 4.0.3 for statistical computing. To uncover the relationship between the gene signature and Cisplatin, we performed the correlation analysis between them and selected MIEF2 for further in vitro. Both loss-of- and gain-of-function research was performed to examine the impact of MIEF2 on therapeutic response to Cisplatin using KIRP cell line Caki-2 and ACHN.
Results
We identified 31 potential genes related to mitochondrial fusion. Four mitochondrial fusion-related genes (BNIP3, GDAP1, MIEF2, and PRKN) showed a significant correlation with overall survival. We constructed a risk score model predicated on the expression levels of these genes, which categorized patients into high- and low-risk groups showing significant differences in overall survival. The area under the ROC curve (AUC) for the risk score was 0.782, indicating its robust predictive performance. The RNA signature related to mitochondrial fusion was validated as an independent predictor of prognosis (P = 0.011, HR = 1.063, and 95% CI = 1.014–1.114). Additionally, our findings suggest that this model demonstrates significant potential in predicting cisplatin sensitivity in KIRP. By loss-of- and gain-of-function research targeting MIEF2 in vitro, we further confirmed that patients in the high-risk group who showed lower expression of MIEF2 were more sensitive to Cisplatin compared to the patients in the low-risk group.
Conclusion
We developed a novel mitochondrial fusion RNA signature that effectively predicts the prognosis of KIRP patients. This signature could serve as a valuable tool for guiding personalized treatment and follow-up strategies in clinical practice.