Activating miRNA-mRNA network in gemcitabine-resistant pancreatic cancer cell associates with alteration of memory CD4(+) T cells

BACKGROUND: To identify key microRNAs (miRNAs) and their target mRNAs related to gemcitabine-resistant pancreatic cancer (PC) and investigate the association between gemcitabine-resistant-related miRNAs and mRNAs and immune infiltration. METHODS: Expression profiles of miRNAs and mRNAs were obtained from the Gene Expression Omnibus (GEO) database. The differentially expressed miRNAs and mRNAs (referred to as “DEmiRNAs” and “DEmRNAs”, respectively) were distinguished between gemcitabine-resistant PC cells and its parental cells. The DEmRNAs targeted by the DEmiRNAs were retrieved using miRDB, microT, and Targetscan. Furthermore, GO and KEGG pathway enrichment analysis and GSEA were performed. The Kaplan-Meier plotter was used to analyze the prognosis of key DEmiRNAs and DEmRNAs on PC patients. The relationship between the key DEmRNAs and tumor-infiltrating immune cells in PC was investigated using CIBERSORT method using the LM22 signature as reference. Key infiltrating immune cells were further analyzed for the associations with prognosis of TCGA PAAD patients. RESULTS: Four DEmiRNAs, including hsa-miR-3178, hsa-miR-485-3p, hsa-miR-574-5p, and hsa-miR-584-5p, were identified to target seven DEmRNAs, including MSI2, TEAD1, GNPDA1, RND3, PRKACB, TRIM68, and YKT6, individually, in gemcitabine-resistant PC cells versus parental cells. Gemcitabine-resistant PC cells were enriched in proteasome-related, immune-related, and memory CD4(+) T cell-related pathways, indicating a gemcitabine therapeutic effect on PC cells. All four DEmiRNAs and almost all DEmRNAs had an impact on the prognosis of PC patients. All seven DEmRNAs had remarkable effects on CD4(+) memory T cells, which were affected by the gemcitabine therapeutic effect. Effector memory CD4(+) T cells rather than central memory CD4(+) T cells predicted a good prognosis according to the TCGA PAAD dataset. CONCLUSIONS: Gemcitabine resistance can alter the fraction of memory CD4(+) T cells via hsa-miR-3178, hsa-miR-485-3p, hsa-miR-574-5p and hsa-miR-584-5p targeted MSI2, TEAD1, GNPDA1, RND3, PRKACB, TRIM68, and YKT6 network in PC.