Identification Of Immunometabolic Subtypes Of Cervical Cancer And Study On The Mechanism Of IMPDH1 Mediated Progression

Objective:To explore the existence of immune-metabolism related molecular subtypes of cervical cancer through high-throughput sequencing data and to analyze the biological and prognostic heterogeneity of different molecular subtypes.To construct a prognostic risk prediction model based on immune-metabolism related genes and uncover novel therapeutic targets through external validation of key molecules at single cell resolution and in vitro experiments.Methods:Bulk sequencing data of cervical cancer was downloaded from The Cancer Genome Alta(TCGA)database,with clinical data collected.Immune and metabolism related genes were downloaded from Immport database and Kyoto Encyclopedia of Genes and Genomes(KEGG)database respectively.Univariate Cox regression analysis was performed for every gene in the geneset to select the genes with prognostic significance.Unsupervised consensus clustering was performed to identify immune-metabolism related subtypes.Differential expressed genes(DEGs),gene set enrichment analysis(GSEA)and gene set variation analysis(GSVA)were performed among the subtypes.Cibersort and GSVA were used for evaluation of the immune cells infiltration status and metabolic characteristics respectively.Multivariate Cox regression analysis was applied in model construction and the key molecules were validated at single cell resolution.In vitro experiments were conducted based on clinical cervical cancer specimens and cell lines.High throughput sequencing was performed to explore the potential molecular mechanism.Results:Three immune-metabolism related subtypes of cervical cancer with distinct prognosis(OS,PFS and DSS)were identified.631 DEGs were found among three subtypes.Immune cells infiltration and metabolism pathway enrichment analysis were performed.C1 was characterized with immunosuppression and cytochrome P450 metabolism activation.C2 was an immune desert subtype with aminoacyl-tRNA biosynthesis activation.C3 is characterized with immune inflammation and activation of aromatic amino acid metabolism.Ten genes were selected for subsequent model construction with Lasso-Cox regression analysis.The AUC of the model was 0.8.Validation results at single cell resolution showed that IMPDH1 was mainly expressed in cancer cells.Immunohistochemistry of clinical samples revealed significant upregulation of IMPDH1 in cervical cancer.The results of in vitro experiments suggested that targeting IMPDH1 can inhibit the growth of cervical cancer,induced apoptosis and downregulated PD-L1 through inhibition of NF-κB signaling.Conclusions:There were three immune-metabolism related subtypes in cervical cancer with different prognosis.The immune-metabolism related risk model showed good performance and applicability in clinical use.In vitro experiments suggested that targeting IMPDH1 can effectively suppress cervical cancer and downregulate the expression of PD-L1.