Aquaporin 9 inhibits growth and metastasis of hepatocellular carcinoma cells via Wnt/β-catenin pathway

AQP9 was down-regulated in HCC tissues and associated with the prognosis of patients

AQP9 was abundantly expressed on the cell membrane in para-carcinoma tissues (Figure 1A and 1C), while its expression level was significantly decreased in HCC samples (Figure 1B and 1D). Western blotting was also used to determine the expression of AQP9 protein in HCC and matched para-carcinoma tissues. The results revealed that the expression level of AQP9 was remarkably down-regulated in HCC samples compared with the control (Figure 1E and 1F). To further investigate the relationship between AQP9 expression and the progression of HCC, RT-qPCR was performed to examine the levels of AQP9 mRNA. The results indicated that AQP9 mRNA levels were significantly reduced in HCC tissues compared with paired para-carcinoma controls (Figure 2A). Furthermore, the levels of AQP9 were evaluated in patients with different clinical characteristics. The expression of AQP9 was significantly down-regulated in HCC patients with larger tumor diameter (≥5cm), lymph node metastasis and advanced TNM stage (III-IV; Figure 2B–2D). In univariate analysis, low expression of AQP9 was associated with tumor size/number, TNM stage and lymph node/distant metastasis within HCC patients (Table 1).

Figure 1. The expression of AQP9 was reduced in HCC samples. (A–D) The levels of AQP9 were examined in HCC and paired non-tumor tissues by immunohistochemistry analysis (magnificationx200 and x400). (E and F) The levels of AQP9 were also examined using western blotting. The results were represented as mean ± SD. P<0.05 vs. para-carcinoma control. Each experiment was repeated 3 times. AQP, aquaporin 9; HCC, hepatocellular carcinoma.

Figure 2. The levels of AQP9 in HCC tissues were associated with the prognosis of patients. (A) The expression of AQP9 was evaluated in HCC and matched para-carcinoma samples by RT-qPCR. The levels of AQP9 were examined in HCC tissues with various tumor diameter (B) lymph node metastasis (C) and different TNM stages (D). Survival rate of 50 HCC patients with different AQP9 expression levels (E) and those in the GEPIA database (F); 182 cases with high-/low-expression were analyzed. The results were represented as mean ± SD. P<0.05 vs. control group. Each experiment was repeated 3 times. AQP, aquaporin 9; HCC, hepatocellular carcinoma; RT-qPCR, reverse transcriptase-quantitative polymerase chain reaction.

A total of 50 HCC patients were followed up for five years. The results suggested that there was no significant difference in one-/three-year survival rates between AQP9 low- and high-expression group. However, the five-year survival rate of AQP9 low-expression group was 35.7% (10/28), which was significantly reduced compared with AQP9 high-expression group (63.6%, 14/22; Table 2). The log-rank test revealed that the overall survival rate of AQP9 low-expression group was remarkably poorer compared to high-expression counterpart (Figure 2E). To further determine the relationship between AQP9 expression and the prognosis of HCC patients, 182 patients with AQP9 high-expression and 182 patients with AQP9 low-expression were analyzed in GEPIA (Figure 2F). The data indicated that the survival rate of AQP9 low-expression group was notably decreased, suggesting the occurrence and progression of HCC was correlated with the expression of AQP9.

AQP9 was down-regulated in HCC cells

To further evaluate the expression of AQP9 in HCC cell lines, the levels of AQP9 mRNA were examined in normal liver cells L02 and HCC cell lines Huh-7, HLE, HepG2 and SMMC-7721 by RT-qPCR. The expression level of AQP9 was significantly reduced in HCC cells compared with normal hepatocytes L02 (Figure 3A). Additionally, Huh-7 cells exhibited the highest expression of AQP9, while the level in SMMC-7721 cells was the lowest among HCC cell lines. Therefore, Huh-7 and SMMC-7721 cells were used in further function study. Following the transfection with LV-AQP9, the expression level of AQP9 was significantly elevated in HCC cells (Figure 3B–3D).

Figure 3. Down-regulation of AQP9 in HCC cells. (A) The mRNA levels of AQP9 in HCC cell lines were determined compared with the control. (B–D) The protein levels of AQP9 in Huh-7 and SMMC-7721 cells transfected with LV-NC or LV-AQP9 were also examined using western blotting. The results were represented as mean ± SD. P<0.05 vs. LO2 or LV-NC. Each experiment was repeated 3 times. AQP, aquaporin 9; HCC, hepatocellular carcinoma; NC, negative control.

Overexpressed AQP9 inhibited proliferation, invasion, migration and EMT in HCC cells, but promoted cell apoptosis

To identify the regulatory functions of AQP9 on the biological behaviors of HCC cells, CCK8 assay was carried out. The data revealed that cell proliferation was remarkably suppressed in Huh-7 and SMMC-7721 cells transfected with LV-AQP9 (Figure 4A and 4B). In addition, the expression levels of proliferating cell nuclear antigen (PCNA) were examined. The results of RT-qPCR indicated that overexpression of AQP9 notably reduced the mRNA levels of PCNA in HCC cells (Figure 4C). Furthermore, immunocytochemistry analysis suggested that PCNA was predominantly expressed in the nucleus of HCC cells, and its expression levels were remarkably reduced following the transfection with LV-AQP9 (Figure 4D–4F). These findings revealed that overexpression of AQP9 was able to inhibit proliferation of HCC cells.

Figure 4. Overexpression of AQP9 inhibited proliferation of HCC cells. (A and B) The proliferative activity of Huh-7 and SMMC-7721 cells transfected with LV-AQP9 was determined by CCK-8 assay compared with the control. (C) The expression levels of PCNA in HCC cells were examined following the transfection with LV-NC or LV-AQP9. (D–F) The expression of PCNA in Huh-7 and SMMC-7721 cells transfected with LV-NC or LV-AQP9 were evaluated using immunocytochemistry analysis. The results were represented as mean ± SD. P<0.05 vs. LV-NC. Each experiment was repeated 3 times. AQP, aquaporin 9; HCC, hepatocellular carcinoma; NC, negative control.

The progression of HCC was also assessed by examining cell invasion, migration and EMT. The results indicated that the invasion (Figure 5A and 5C) and migration (Figure 5B and 5D) of HCC cells transfected with LV-AQP9 were suppressed compared with the control. Furthermore, the expression levels of EMT-related markers, such as N-cad and α-SMA were significantly decreased, while E-cad was upregulated in HCC cells treated with LV-AQP9 (Figure 5E and 5F). Additionally, flow cytometry was performed to determine cell apoptosis following the transfection with LV-AQP9. Our data suggested that the apoptosis rate of Huh-7 and SMMC-7721 cells was remarkably increased after the treatment with LV-AQP9 (Figure 5G–5I). In summary, these findings indicated that overexpressed AQP9 could suppress cell proliferation, invasion, migration and EMT, while promoting cell apoptosis in HCC cells, suggesting AQP9 expression is associated with the biological behavior changes of HCC cells.

Figure 5. Overexpressed AQP9 inhibited invasion, migration and EMT in HCC cells, but promoted cell apoptosis. (A and C) The invasive abilities in Huh-7 and SMMC-7721 cells transfected with LV-NC or LV-AQP9 were determined by Transwell assay (magnificationx100). (B and D) Cell migration were examined following the transfection of LV-AQP9 (magnificationx100). (E and F) The mRNA levels of E-cad, N-cad and α-SMA in transfected Huh-7 and SMMC-7721 cells were evaluated using RT-qPCR. (G–I) The cell apoptosis rate in HCC cells transfected with LV-NC or LV-AQP9 was analyzed by flow cytometry. Data are shown as mean ± SD based on at least three independent experiments. The results were represented as mean ± SD. P<0.05 vs. LV-NC. Each experiment was repeated 3 times. AQP, aquaporin 9; HCC, hepatocellular carcinoma; NC, negative control; RT-qPCR, reverse transcriptase-quantitative polymerase chain reaction.

Overexpression of AQP9 suppressed Wnt/β-catenin pathway in HCC cells

Wnt/β-catenin signaling pathway plays essential roles during the development of HCC. The expression levels of related molecules involved in Wnt/β-catenin pathway were examined using RT-qPCR and western blotting. The results revealed that both mRNA and protein levels of DVL2, p-GSK-3β, CyclinD1 and β-catenin were significantly decreased in Huh-7 and SMMC-7721 cells transfected with LV-AQP9 (Figure 6A–6D). To further investigate the functions of Wnt/β-catenin signaling on the progression of HCC, Huh-7 and SMMC-7721 cells were treated with Wnt/β-catenin signaling inhibitor XAV939 or activator SKL2001, respectively. The data indicated that β-catenin was down-regulated in Huh-7 cells treated with XAV939 (Figure 6E and 6G) and up-regulated in SMMC-7721 cells following the treatment with SKL2001 (Figure 6F and 6G). These findings suggested that overexpressed AQP9 could exert its anti-tumor effects by suppressing Wnt/β-catenin signaling.

Figure 6. Overexpression of AQP9 was able to suppress Wnt/β-catenin signaling. (A–D) The mRNA and protein levels of DVL2, GSK-3β, CyclinD1 and β-catenin in Huh-7 and SMMC-7721 cells transfected with LV-NC or LV-AQP9 were examined by RT-qPCR and western blotting. (E) The expression of β-catenin in HCC cells treated with XAV939 were evaluated using RT-qPCR. (F) The mRNA levels of β-catenin in SMMC-7721 cells were determined following the treatment with SKL2001. (G) The protein levels of β-catenin in HCC cells were assessed by western blot analysis. The results were represented as mean ± SD. P<0.05 vs. NC or LV-NC. Each experiment was repeated 3 times. AQP, aquaporin 9; HCC, hepatocellular carcinoma; NC, negative control; RT-qPCR, reverse transcriptase-quantitative polymerase chain reaction.

Blockage of Wnt/β-catenin signaling inhibited the proliferation, invasion and migration of HCC cells, but promoted cell apoptosis

To further identify the regulatory roles of AQP9 in HCC via Wnt/β-catenin signaling, further function studies were conducted. The results revealed that cell proliferation was inhibited (Figure 7A), and EMT was suppressed (Figure 7B) in Huh-7 cells treated with XAV939. Furthermore, the expression levels of PCNA were downregulated in HCC cells following the treatment with XAV939 (Figure 7C and 7D). Additionally, cell invasion (Figure 7E and 7F) and migration (Figure 7G and 7H) were suppressed in Huh-7 cells treated with XAV939. Flow cytometry also indicated that the apoptosis rate of Huh-7 cells was remarkably decreased after the treatment with XAV939 (Figure 7I and 7J). Taken all together, as Wnt/β-catenin signaling plays key roles in the pathogenesis of liver cancer, AQP9 could affect the proliferation, invasion, migration and apoptosis of HCC cells through Wnt/β-catenin pathway.

Figure 7. Blockage of Wnt/β-catenin signaling inhibited growth and metastasis of HCC cells. (A) The proliferation of Huh-7 cells treated with XAV939 was determined by CCK-8 assay. (B) The mRNA levels of E-cad, N-cad and α-SMA in transfected HCC cells were examined using RT-qPCR. (C and D) The expression of PCNA in Huh-7 cells were evaluated following the treatment with XAV939. (E to H) The invasive and migrative abilities of HCC cells treated with XAV939 were assessed by Transwell assay (magnificationx100). (I and J) Cell apoptosis following the treatment with XAV939 was determined using flow cytometry. The results were represented as mean ± SD. P<0.05 vs. NC. Each experiment was repeated 3 times. HCC, hepatocellular carcinoma; RT-qPCR, reverse transcriptase-quantitative polymerase chain reaction.

Activation of Wnt/β-catenin pathway reversed the effects of AQP9 on HCC

In order to confirm that overexpressed AQP9 could suppress the progression in HCC in a Wnt/β-catenin-dependent manner, SMMC-7721 cells were co-transfected with LV-AQP9 and SKL2001. The results revealed that cell proliferative activity (Figure 8A) and the expression of PCNA (Figure 8B) were significantly downregulated in HCC cells transfected with LV-AQP9, which was abolished by SKL2001 treatment. Similarly, EMT was inhibited in SMMC-7721 cells transfected with LV-AQP9, and these effects were reversed following co-transfection SKL2001 (Figure 8C and 8D). Furthermore, Transwell assays indicated that cell invasion and migration were suppressed by overexpressed AQP9, which were abrogated by the treatment with SKL2001. Flow cytometry also revealed that overexpression of AQP9 promoted cell apoptosis, and these effects were abolished by co-transfection with SKL2001. Our findings revealed that SKL2001 could reverse the effects caused by overexpressed AQP9 during the development of HCC, suggesting AQP9 may function as a tumor suppressor via suppressing Wnt/β-catenin signaling.

Figure 8. Activation of Wnt/β-catenin pathway reversed the effects of AQP9 on HCC progression. (A) The proliferation of SMMC-7721 cells transfected with LV-NC, LV-AQP9 or LV-AQP9+SKL2001 was determined. (B–D) The expression levels of PCNA, E-cad, N-cad and α-SMA in transfected HCC cells were examined using RT-qPCR. (E and F) The invasive and migration activities of SMMC-7721 cells were evaluated following the treatment with LV-NC, LV-AQP9 or LV-AQP9+SKL2001 (magnificationx100). (G) The apoptosis rate of transfected HCC cells was assessed by flow cytometry. The results were represented as mean ± SD. P<0.05 vs. LV-NC. Each experiment was repeated 3 times. AQP, aquaporin 9; HCC, hepatocellular carcinoma; NC, negative control; RT-qPCR, reverse transcriptase-quantitative polymerase chain reaction.

Overexpression of AQP9 suppressed tumor growth in vivo

To identify the novel functions of AQP9 in vivo, SMMC-7721 cells transfected with LV-NC or LV-AQP9 were injected into nude mice subcutaneously. Tumor volumes were measured weekly until day 42. Six weeks post-injection, the mice were sacrificed and the isolated tumors were examined. Mean value of tumor volume in LV-AQP9 group was significantly reduced compared with the control (Figure 9A and 9B). In addition, the tumor weight in LV-AQP9 group was notably decreased than LV-NC group (Figure 9C). Furthermore, the protein levels of AQP9, DVL2, GSK-3β, CyclinD1 and β-catenin in xenograft tumors were determined using western blot analysis. Our results revealed that the expression of AQP9 was significantly increased, while the levels of Wnt/β-catenin-associated molecules were remarkably reduced in LV-AQP9 group (Figure 9D). Moreover, EMT was inhibited in LV-AQP9 group compared with the control (Figure 9E). These findings suggested that AQP9 was a putative tumor suppressor during the development of HCC, which could affect the growth and metastasis of HCC cells by suppressing Wnt/β-catenin signaling. More importantly, AQP9/Wnt/β-catenin axis could be a potential therapeutic target during the treatment of HCC.

Figure 9. Overexpression of AQP9 suppressed tumor growth of HCC in vivo. (A) The growth curves of tumors from nude mice in LV-NC and LV-AQP9 group. (B and C). Orthotopic tumor volumes and weights at day 42 post-injection were calculated. (D) The protein levels of AQP9, DVL2, GSK-3β, CyclinD1 and β-catenin in isolated tumors were examined using western blotting. (E) The mRNA levels of E-cad, N-cad and α-SMA were determined by RT-qPCR. The results were represented as mean ± SD. P<0.05 vs. LV-NC. AQP, aquaporin 9; HCC, hepatocellular carcinoma; NC, negative control; RT-qPCR, reverse transcriptase-quantitative polymerase chain reaction.

Patient samples

A total of 50 paired HCC specimens (42 male and 8 female; average age, 53.26±10.22 years) were obtained from the patients who underwent surgical resection at the Second Affiliated Hospital of Chongqing Medical University (Chongqing, China). None of the patients received immuno-, chemo- or radio-therapy prior to the operation. The clinicopathological characteristics of HCC patients were reviewed by two independent pathologists. All the samples were snap-frozen using liquid nitrogen until further use. The patients were divided into two groups according to the mean expression level of AQP9 (AQP9 high/low expression group). Informed written consents were signed by all the patients. The protocol of this study was approved by the Ethics Committee of the Second Affiliated Hospital of Chongqing Medical University.

Cell culture and transfection

Normal human hepatocyte (LO2) as well as HCC cells (HLE, Huh-7, HepG2 and SMMC-7721) were purchased from the Cell Culture Center, Chinese Academy of Medical Sciences (Shanghai, China). The cells were cultured using Dulbecco’s modified Eagle’s medium (DMEM; Gibco, USA) supplemented with 10% fetal bovine serum (FBS; Gibco, USA) and incubated at 37°C in a humidified atmosphere with 5% CO₂. Huh-7 and SMMC7721 cells were transfected with AQP9 overexpressing vector (LV-AQP9) or the control (LV-NC). Subsequently, the cells were selected with 1 μg/mL puromycin for four weeks. The cells were harvested 48 h post-transfection for use in further experiments. To inhibit or induce Wnt/β-catenin pathway, Huh-7 cells were treated with 10 mM XAV939 (MedChemExpress LLC, Monmouth Junction, NJ, USA), and SMMC-7721 were treated with 40 μM SKL2001 (MedChemExpress LLC, Monmouth Junction, NJ, USA), respectively.

Reverse transcription-quantitative polymerase chain reaction (RT-qPCR)

Total RNA was extracted from tissues or cells using TRIzol^® reagent (Invitrogen; Thermo Fisher Scientific, Inc., Carlsbad, CA, USA) according to the manufacturer's protocols. Extracted RNA was reverse transcribed by PrimeScript RT kit with gDNA Eraser (Takara Biomedical Technology Co., Ltd., Beijing, China) according to the manufacturer's protocols. Quantitative PCR was performed using SYBR Green PCR Master Mix (Takara Biomedical Technology Co., Ltd., Beijing, China) according to the manufacturer’s protocols. Relative expression was normalized to endogenous Glyceraldehyde 3-phosphate dehydrogenase (GAPDH). The forward and reverse primer sequences were as follows: AQP9 forward, 5′-AAATAAACCTCCTTGGCCTGA-3′ and reverse, 5′-GCAACAAACATCACCACACC-3′; E-Cad forward, 5′-CAATGGTGTCCATGGAACA-3′ and reverse, 5′-CCTCCTACCCTCCTGTTCG-3′; α-SMA forward, 5′-TCCCTTGAGAAGAGTTACGAGTTG-3′ and reverse, 5′-ATGATGCTGTTGTAGGTGGTTTC-3′; N-Cad forward, 5′-CAGTATCCGGTCCGATCTGC-3′ and reverse, 5′-GTCCTGCTCACCACCACTAC-3′; DVL2 forward, 5′-AGTCAGCTCTCATGTTGAGGGT-3′ and reverse, 5′-TACCCAGCCCACACCTTCTT-3′; GSK-3β forward, 5′-CCGACTAACACCACTGGAAGCT-3′ and reverse, 5′-AGGATGGTAGCCAGAGGTGGAT-3′; CyclinD1 forward, 5′- GAGACCATCCCCCTGACGGC-3′ and reverse, 5′-TCTTCCTCCTCCTCGGCGGC-3′; β-catenin forward, 5′-TGCAGTTCGCCTTCACTATG-3′ and reverse, 5′-ACTAGTCGTGGAATGGCACC-3′; and GAPDH forward, 5′-GCAAGAGCACAAGAGGAAGA-3′ and reverse, 5′-ACTGTGAGGAGGGGAG ATTC-3′. Relative expression levels were calculated using 2^-ΔΔCq method.

Western blotting

Total protein was extracted from tissues or cells using RIPA buffer (Beyotime Biotechnology, Shanghai, China). Protein concentrations were determined using a BCA Kit (Beyotime Biotechnology). Equal amounts (40 μg) of proteins were separated by 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred onto polyvinylidene fluoride membranes (Millipore, Billerica, MA, USA). Subsequently, the membranes were blocked with 5% (w/v) non-fat milk in TBST buffer (Beyotime Biotechnology) at 37°C for 1 h. The membranes were then incubated with anti-AQP9 (1:1000; cat. no. ab15129), anti-E-Cad (1:5000; cat. no. ab40772), anti-α-SMA (1:2000; cat. no. ab124964), anti-N-cad (1:5000 dilution; cat. no. ab18203), anti-β-catenin (1:4000; cat. no. ab6302), anti-CyclinD1 (1:100; cat. no. ab16663), anti-p-GSK-3β (1:10000; cat. no. ab75814), anti-DVL2 (1:1000; cat. no. ab22616), anti-β-actin (1:5000; cat. no. ab179467) antibodies (Abcam, Cambridge, MA, USA) at 4°C overnight. Following three washes in TBST for 10 min, the membranes were incubated with horseradish peroxidase-conjugated goat anti-mouse IgG (1:1,000; cat. no. 7076) or goat anti-rabbit IgG (1:1,000; cat. no. 7074) antibodies (Cell Signaling Technology, Danvers, MA, USA) at 37°C for 1 h. The protein bands were visualized by an enhanced chemiluminescence kit and quantified by densitometric analysis using ImageJ software (NIH, Bethesda, MD, USA).

Immunohistochemistry (IHC) staining

Patient samples were fixed and embedded in paraffin. The sections were deparaffinized and hydrated, then blocked with endogenous peroxidase and pretreated with citrate buffer for antigen retrieval. The sections were rinsed with PBS and then incubated with anti-AQP9 (1:500; cat. no. ab15129; Abcam, Cambridge, MA, USA) at 4°C overnight, followed by incubation with secondary antibody at room temperature for 30 min. After three washes with PBS, staining intensity was measured by selecting five random fields per section and analysed using Image-Pro Plus software (Media Cybernetics, Rockville, MD, USA).

Immunofluorescence analysis

Transfected Huh-7 and SMMC-7721 cells were seeded onto a 6-well culture plate at the density of 2x10⁵ cells/well. After the incubation for 12 h, the culture medium was replenished. The cells were rinsed with PBS for five min and fixed with 4% paraformaldehyde solution for 15 min. Then the cells were treated with 2.5% Triton for 10 minutes to permeabilize the membrane and blocked with goat serum at 37°C for 1 h. Subsequently, the cells were incubated with anti-PCNA (1:100; cat. no. ab92552; Abcam, Cambridge, MA, USA) antibody at 4°C overnight, followed by the incubation with secondary antibody at room temperature for five min. DAPI nuclear staining solution (1 μg/mL) was added at room temperature for five min. The cells were immersed with fluorescent quencher and subjected to fluorescence microscope.

Cell proliferation assay

Transfected Huh-7 and SMMC-7721 cells were seeded at a density of 5x10³ cells/well and were examined on day 1, 2, 3 and 4 post-transfection according to the manufacturer’s protocols. Cell viability was evaluated using Cell Counting kit-8 assay (CCK-8; Beyotime, Shanghai, China). Briefly, CCK8 solution was added to each well and the cells were incubated for an additional 2 h. The absorbance at 450 nm was measured using a microplate reader (9200, Bio-Rad Laboratories, Hercules, CA, USA) according to the manufacturer's protocols.

Transwell migration and invasion assay

Two days post-transfection, Huh-7 and SMMC-7721 cells were harvested. The migration and invasion of cells were examined using a Transwell assay. For cell migration assay, a total of 5x104 cells in serum free medium were placed onto the upper chamber (pore size=8μm; BD Biosciences, Franklin Lakes, New Jersey, USA). For cell invasion assay, 1x105 cells were added into the upper chamber which was pre-coated with Matrigel^® (Sigma-Aldrich, St. Louis, MO, USA). Then, 500 μl of culture medium containing 10% FBS was added into the lower chamber. Following overnight incubation, non-migratory/-invasive cells were removed using a cotton swab. The migrated/invaded cells in the lower chamber were fixed using 4% paraformaldehyde and stained with 0.5% crystal violet. The numbers of cells were counted within five randomly selected fields by an inverted light microscope (magnificationx100, Olympus Corporation, Tokyo, Japan).

Apoptosis analysis

Transfected Huh-7 and SMMC-7721 cells were placed onto a 6-well culture plate at the density of 2x10⁵ cells/well. After 12 h, the medium was replenished. The cells were centrifugated and resuspended with pre-chilled 75% ethanol prior to overnight fixation at -20°C. Subsequently, the cells were resuspended using 450μL PBS supplemented with 50μL propidium iodide (PI, 0.5 mg/mL; Beyotime Biotechnology) and incubated at 37°C for 30 min. Then, 5 μL Annexin-V-fluorescein isothiocyanate (FITC; Beyotime Biotechnology) was used to stain the cells. The apoptosis was analyzed using a flow cytometer (BD Biosciences, USA).

Animal study

A total of eight female BALB/C nude mice (4-6 weeks old) with a weight of 17-22 g were obtained from the Experimental Animal Center of Chongqing Medical University (Chongqing, China). The protocols were reviewed and approved by the Research Ethics Committee of Chongqing Medical University (Chongqing, China). The mice were routinely housed under a temperature- (22±2°C) and humidity-controlled (80%) atmosphere, with a 12-h dark/light cycle and libitum access to food/water for at least three days before the operation. SMMC-7721/LV-AQP9 or SMMC-7721/LV-NC cells (2 x 10⁶) were suspended in 200 μl PBS and injected subcutaneously into right side of the armpit regions of mice. Six weeks post-injection, the mice were sacrificed by decapitation, and the tumors were isolated and examined. Tumor volume was calculated using the following formula: V (mm³) = 0.5 x length x width². The tumor tissues were snap-frozen in liquid nitrogen until further use.

Statistical analysis

Data were presented as the means ± standard deviation and were analyzed using Graphpad Prism v7.0 software (Graphpad Software Inc., La Jolla, CA, USA). A t-test (two-sided) or one-way analysis of variance (ANOVA) was used for statistical analysis. A student-Newman-Keuls test was performed as a post-hoc test following ANOVA. Count data were processed by chi-square test. Overall survival was analyzed using Kaplan-Meier survival test. All the experiments were performed at least three times. P<0.05 was considered to indicate a statistically significant difference.