Aquaporin 3 promotes epithelial-mesenchymal transition in gastric cancer

Abstract

Background: Gastric carcinoma (GC) is a common and lethal malignancy, and epithelial-mesenchymal transition (EMT) is believed to contribute to invasive and metastatic tumor growth. Aquaporin 3 (AQP3) is overexpressed in human GC tissues, while human epidermal growth factor (EGF) and hepatocyte growth factor, which can induce EMT, are able to up-regulate AQP3 expression, subsequently promoting GC cell migration and proliferation. The purpose of this study was to investigate the effects of AQP3 on EMT in human GC.

Methods: AQP3 and EMT-related proteins were detected by immunohistochemistry in human GC specimens and their clinical significance evaluated. AQP3 knockdown was attempted using small interfering RNAs, while EGF was used to up-regulate AQP3 expression. Western blotting, real-time quantitative polymerase chain reaction assays and immunofluorescence were used to evaluate changes in expression of AQP3 and EMT-related proteins in the SGC7901 and MGC803 human GC cell lines.

Results: AQP3 up-expression was associated with EMT-related proteins in human GC specimens, which correlated with poor prognosis for GC. AQP3 modulated GC cell proliferation, migration and invasion in vitro, and induced E-cadherin repression. AQP3 also up-regulated the expression of vimentin and fibronectin in vitro. The PI3K/AKT/SNAIL signaling pathway was likely involved in the induction of EMT by AQP3 in GC.

Conclusions: AQP3 promotes EMT in human cases of GC, allowing us to understand the mechanisms of AQP3 in GC progression, thus providing a potential strategy for its treatment.

Figure 1

Detection of AQP3, E-cadherin, and vimentin expression in GC tissue and adjacent normal tissue by IHC. Strong AQP3 immunoreactivity was identified in poorly differentiated adenocarcinomas. E-cadherin expression was observed in normal gastric glands but not in GC tissue. Vimentin expression was not seen in normal tissue but was observed in GC tissue (red arrows). Original magnification, ×400.

Figure 2

Expression of AQP3 and associated EMT proteins predict poor prognosis of GC. Patients that overexpressed AQP3 demonstrated shorter OS than those in the low expression group (P = 0.038). Patients with lower expression levels of E-cadherin had a worse OS than those with high E-cadherin expression levels (P = 0.008). Patients that were positive for vimentin expression exhibited poor survival rates compared with those who were negative for vimentin (P = 0.048). Patients with high expression levels of AQP3 but lacked E-cadherin and vimentin had a worse OS (P = 0.028).

Figure 3

AQP3 promotes cell proliferation of GC cells. Cell proliferation of SGC7901 (A and B) and MGC803 (A and C) was significantly increased after treatment with EGF and decreased after treatment with RNAi AQP3. Data are expressed as the mean ± SE from three independent experiments. #P < 0.05 compared with the untreated group (UNTR); *P < 0.05 compared with the RNAi AQP3 group.

Figure 4

AQP3 facilitates GC cell migration and invasion. GC cell migration and invasion were detected using transwell migration and invasion assays. The number of cancer cells migrating through the Matrigel decreased significantly after treatment with RNAi AQP3 compared with the UNTR group, while treatment with EGF had the opposite effect (A and B). AQP3-silenced GC cells invaded significantly slower when compared with the UNTR group and over-expression of AQP3 accelerated cell invasion (C and D). Data are expressed as the mean ± SE from three independent experiments. #P < 0.05 compared with the untreated group (UNTR); *P < 0.05 compared with the RNAi AQP3 group. Original magnification × 100.

Figure 5

AQP3 promotes EMT induction in human gastric adenocarcinoma cells. (A) Expression levels of AQP3, E-cadherin, vimentin and fibronectin in SGC7901 and MGC803 cells were determined using western blots. GAPDH was used as an internal control. The relative accumulation of proteins in different groups was compared with those in the untreated group (UNTR). (B) mRNA expression levels of AQP3 and EMT-related proteins were assayed using qPCR. Data are expressed as the mean ± SE from three independent experiments. *P < 0.05 compared with the UNTR group; ^#P < 0.05 compared with the RNAi AQP3 group. (C) Immunofluorescence assays for the detection of AQP3 and three EMT-related proteins. Target proteins were detected using the appropriate antibodies (green), and nuclei were stained with Hoechst33342 (blue).

Figure 6

AQP3 regulates EMT via the PI3K/AKT/Snail pathway. SGC7901and MGC803 cells were treated with control siRNA, RNAi AQP3 and EGF, with or without a PI3K/AKT inhibitor. Proteins were analyzed by western blotting assay. GAPDH was used as an internal control. The relative accumulation of proteins was compared with the untreated group.