AQP5 Is a Novel Prognostic Biomarker in Pancreatic Adenocarcinoma

Abstract

Background: Pancreatic adenocarcinoma (PAAD) is a highly malignant tumor with a poor prognosis. The identification of effective molecular markers is of great significance for diagnosis and treatment. Aquaporins (AQPs) are a family of water channel proteins that exhibit several properties and play regulatory roles in human carcinogenesis. However, the association between Aquaporin-5 (AQP5) expression and prognosis and tumor-infiltrating lymphocytes in PAAD has not been reported.

Methods: AQP5 mRNA expression, methylation, and protein expression data in PAAD were analyzed using GEPIA, UALCAN, HAP, METHSURV, and UCSC databases. AQP5 expression in PAAD patients and cell lines from our cohort was examined using immunohistochemistry and Western blotting. The LinkedOmics database was used to study signaling pathways related to AQP5 expression. TIMER and TISIDB were used to analyze correlations among AQP5, tumor-infiltrating immune cells, and immunomodulators. Survival was analyzed using TCGA and Kaplan-Meier Plotter databases.

Results: In this study, we investigated AQP5 expression in PAAD and determined whether the expression of AQP5 is a strong prognostic biomarker for PAAD. We searched and analyzed public cancer databases (GEO, TCGA, HAP, UALCAN, GEPIA, etc.) to conclude that AQP5 expression levels were upregulated in PAAD. Kaplan-Meier curve analysis showed that high AQP5 expression positively correlated with poor prognosis. Using TIMER and TISIDB, we found that the expression of AQP5 was associated with different tumor-infiltrating immune cells, especially macrophages. We found that hypomethylation of the AQP5 promoter region was responsible for its high expression in PAAD.

Conclusions: AQP5 can serve as a novel biomarker to predict prognosis and immune infiltration in PAAD.

Keywords: AQP5; methylation; pancreatic adenocarcinoma; prognosis; tumor microenvironment.

Figure 1

AQP5 is widely distributed among the human body. (A) AQP5 is widely distributed in the human body, such as in renal, digestive, and sensory organs. (B) Genomic regions, transcript, and products. (C) The AQP5 protein is composed of four monomers. (D) Topology map of the basic monomeric AQP5 fold.

Figure 2

The expression of gene AQP5 mRNA in different kinds of tumors and its relationship with clinical parameters in PAAD. (A) Human expression levels of AQP5 in 33 various malignant tumor types from The Cancer Genome Atlas (TCGA) database were analyzed by the Tumor Immune Estimation Resource (TIMER). (B) Venn diagram showing the common differential expression genes in three public databases (TCGA, GES16515, and GES71729). (C) The expression of AQP5 in PAAD and normal pancreatic tissues, respectively. (D, E) Association between AQP5 expression and clinicopathologic characteristics with PAAD patients’ cancer grade (D) and pathological T stage (E). The immunohistochemistry of AQP5 in PAAD and normal pancreatic tissue (F, G). (H–J) AQP5 mRNA and protein levels were overexpressed compared with real-time PCR (H), Western blotting (I), and immunofluorescence (J). ^*p < 0.05, ^**p < 0.01, ^***p < 0.001, NS, not significant.

Figure 3

The prognostic value of AQP5 expression in patients with PAAD. (A–C) Forest plot of the prognostic values in different tumor types of AQP5. (D–F) Kaplan–Meier survival analysis revealed that low AQP5 expression in patients with PAAD had a longer overall survival (D), progression-free interval (E), and disease-specific survival (F). (G) The ROC curve of the prognostic values in patients with PAAD of AQP5.

Figure 4

Correlation between AQP5 expression and immune cell infiltrations in PAAD. (A) The expression of AQP5 in PAAD tissues positively correlated with tumor immune infiltration levels of macrophage. (B) Relationship between AQP5 expression and 28 subtypes of tumor-infiltrating immune cells in different kinds of tumors. (C). The CD56 (bright) cell, CD56 (dim) cell, neutrophil, CD4⁺ central memory T cell, CD8⁺ central memory T cell, CD4⁺ effective memory T cell, and T helper cell 17 had significant Spearman’s correlation with AQP5 expression in PAAD. (D) Low AQP5 expression levels in PAAD in enriched B cell, natural killer cell, CD8⁺ T cell, and Th2 cell cohorts had a favorable prognosis. (E) The abundance of tumor-infiltrating immune cells with different somatic copy number aberrations of AQP5 in PAAD, *p < 0.05, **p < 0.01.

Figure 5

The heatmap demonstrated the mutation, CNV, and methylation analysis of AQP5 in PAAD.

Figure 6

DNA methylation levels of AQP5 and its prognostic value in PAAD. The difference of promoter methylation level of AQP5 between tumor stage (A) and grade (B) in PAAD. (C) The expression of AQP5 correlated with AQP5 promoter methylation. (D) The heatmap of DNA methylation at CpG sites of AQP5 in PAAD. (E) The survival analysis of AQP5 promoter methylation in patients with PAAD. (F) Relationship between AQP5 promoter methylation and 28 subtypes of tumor-infiltrating immune cells in different kinds of tumors. (G) Association between the methylation status of AQP5 with immune infiltrates in PAAD. *p < 0.05, **p < 0.01.

Figure 7

AQP5 co-expressed gene and functional enrichment analysis in PAAD. (A) The volcano map of AQP5 gene co-expression. (B, C) The heatmap of the top 50 positive and negative AQP5 gene co-expression. (D–F) The GO analysis of AQP5 gene co-expression in molecular function (MF), cellular component (CC), and biological process (BP). (G) AQP5 co-expression genes were annotated by Gene Ontology (GO) analysis. (H) The top 20 KEGG pathway analysis of AQP5 gene co-expression.

Figure 8

Top 10 positively correlated GSEA enrichment plots.

Figure 9

AQP5 knockdown inhibits migration and invasion of PAAD cells in vitro. Knockdown of AQP5 significantly impaired the migration and invasion capabilities of PANC-1 cells (A, B). ***p < 0.001.