SNX10 and PTGDS are associated with the progression and prognosis of cervical squamous cell carcinoma

Abstract

Background: Cervical cancer (CC) is the primary cause of death in women. This study sought to investigate the potential mechanism and prognostic genes of CC.

Methods: We downloaded four gene expression profiles from GEO. The RRA method was used to integrate and screen differentially expressed genes (DEGs) between CC and normal samples. Functional analysis was performed by clusterprofiler. We built PPI network by Search Tool for the Retrieval of Interacting Genes Database (STRING) and selected hub modules via Molecular COmplex Detection (MCODE). CMap database was used to find molecules with therapeutic potential for CC. The hub genes were validated in GEO datasets, Gene Expession Profiling Interactive Analysis (GEPIA), immunohistochemistry, Cox regression analysis, TCGA methylation analysis and ONCOMINE were carried out. ROC curve analysis and GSEA were also performed to describe the prognostic significance of hub genes.

Results: Functional analysis revealed that 147 DEGs were significantly enriched in binding, cell proliferation, transcriptional activity and cell cycle regulation. PPI network screened 30 hub genes, with CDK1 having the strongest connectivity with CC. Cmap showed that apigenin, thioguanine and trichostatin A might be used to treat CC(P < 0.05). Eight genes (APOD, CXCL8, MMP1, MMP3, PLOD2, PTGDS, SNX10 and SPP1) were screened out through GEPIA. Of them, only PTGDS and SNX10 had not appeared in previous studies about CC. The validation in GEO showed that PTGDS showed low expression while SNX10 presented high expression in tumor tissues. Their expression profiles were consistent with the results in immunohistochemistry. ROC curve analysis indicated that the model had a good diagnostic efficiency (AUC = 0.738). GSEA analysis demonstrated that the two genes were correlated with the chemokine signaling pathway (P < 0.05). TCGA methylation analysis showed that patients with lowly-expressed and highly-methylated PTGDS had a worse prognosis than those with highly-expressed and lowly-methylated PTGDS (p = 0.037). Cox regression analysis showed that SNX10 and PTGDS were independent prognostic indicators for OS among CC patients (P = 0.007 and 0.003).

Conclusions: PTGDS and SNX10 showed abnormal expression and methylation in CC. Both genes might have high prognostic value of CC patients.

Keywords: Cervical squamous cell carcinoma; Methylation; PTGDS; Prognosis; Risk model; SNX10.

Fig. 1

Flow chart of the present study

Fig. 2

LogFC heatmap of the image data of each expression microarray. Notes: The abscissa is the geo ID, and the ordinate is the gene name. Red represents logFC>0, green represents logFC<0, and the values in the box represent the logFC values

Fig. 3

GO and KEGG analysis of the DEGs. a GO analysis of upregulated genes associated with cervical cancer. b GO analysis of downregulated genes associated with cervical cancer. c KEGG analysis of upregulated genes associated with cervical cancer. d KEGG analysis of downregulated genes associated with cervical cancer

Fig. 4

PPI network analysis. a Using the STRING online database, a total of 147 DEGs were filtered into the DEGs PPI network. b Degree of the top 30 nodes in the PPI network. All these nodes are upregulated genes

Fig. 5

Three-dimensional diagram of the three most significant drugs. a Apigenin b Thioguanosine c Trichostatin A

Fig. 6

Validation of PTGDS and SNX10. a Immunohistochemistry of PTGDS. b Immunohistochemistry of SNX10. c Expression of PTGDS in ONCOMINE. d Expression of SNX10 in ONCOMINE

Fig. 7

Validation of PTGDS and SNX10. a ROC curve analysis of the two genes. b GSEA of PTGDS and SNX10. c The methylated expression and gene expression of PTGDS. d The methylated expression and gene expression of SNX10. e Survival analysis of patients with methylated PTGDS expression

Fig. 8

Survival prognosis model of the 5 hub genes. a Survival analysis showed that the patients in the high risk group had statistically significantly worse overall survival than those in low risk group in TCGA cohort. b ROC analysis was performed to find out the most optimal cutoff value to divide the CC patients into high risk and low risk group. c-d The risk scores for all patients in TCGA cohort are plotted in ascending order and marked as low risk (blue) or high risk (red), as divided by the threshold (vertical black line). e Eight expression and risk score distribution in TCGA cohort by z-score, with red indicating higher expression and light blue indicating lower expression