مقالات پذیرفته شده در هفتمین کنگره بین المللی زیست پزشکی
Identification of Potential and Novel Biomarkers Based on Drug Resistance in Two Different Type of Glioma Cell Lines.
Identification of Potential and Novel Biomarkers Based on Drug Resistance in Two Different Type of Glioma Cell Lines.
Pooria Salehi Sangani,1,*Yasin parvizi,2
1. Student Research Committee, Faculty of Medicine, Mashhad University of Medical Sciences, Mashhad, Iran 2. Cancer Research Center, Hamadan University of Medical Sciences, Hamadan, Iran
Introduction: Glioma is a common tumor originating in the brain. About one-third of all brain tumors are gliomas. Temozolomide resistance is an essential challenge in the management of patients with glioma. This study aimed to introduce hub genes that play a role in the treatment resistance of glioma, and further, reveal two different prognostiv models based on two different glioma cell lines.
Methods: Gene expression data of resistance and sensitive glioma cell lines U87 and U251, were downloaded from three separate datasets using the Gene Expression Omnibus(GEO) database, including GSE 193957(U87), GSE151680(U87 & U251) and GSE100736(U251). Differentially expressed genes(DEGs) were analyzed and obtained from each data set based on the cell line type, using limma and DESeq2 packages. Common DEGs for each cell line were obtained for constructing a protein-protein interaction(PPI) network using the STRING database. Cytoscape software and the cytohubba plugin were used to determine hub genes for common DEGs of each cell line. Then, functional enrichment analyses of hub genes were exerted. Further, RNA sequencing and clinical data of 1018 glioma patients were downloaded from the Chinese glioma genome atlas(CGGA). Univariate Cox analysis was performed for common DEGs of each cell line based on CGGA data and significant genes were extracted for multivariate Cox regression. The “Step” method for each cell line was applied to construct two models based on the common DEGs of U87 and U251 cell lines, and CGGA data. Finally, the risk score based on the last multivariate model, was evaluated for each CGGA patient using this formula: Σβi×ExpGenei (βi was the coefficient value and ExpGenei was the gene expression level). Finally, survival analysis was performed based on the risk scores.
Results: 28 and 24 hub genes were obtained for U87 and U251, respectively. Functional enrichment analyses showed that cell–cell interaction and adhesion could be involved in both U87 and U251 resistance to temozolomide. Two risk score models based on common DEGs, univariate and multivariate Cox regression for each cell line, were evaluated. Based on the obtained risk scores, patients were bifurcated into “High-risk” and “Low-risk” groups. Survival analyses and Kaplan Meier plots of two models revealed that high-risk patients experience lower survival in comparison to low-risk patients, significantly. Further, it was independent of other variables such as gender, Isodehyrogenase mutation, and grade.
Conclusion: We believe that differential gene expression between resistant and sensitive glioma cells can reveal new genes that could be involved in resistant mechanisms, and could be targeted by future studies. Further, by integrated bioinformatic analyses based on drug resistance, we potentially, demonstrated two models that could predict the survival of glioma patients independent of other variables.