All primary RCC and pericarcinous tissues were obtained from 123 patients with appropriate informed consent at the Department of Urology of the First Affiliated Hospital of Nanjing Medical University from February 2008 to August 2011. The follow-up deadline was January 2016. The specimens were assessed by immunohistochemistry and the diagnosis was verified by histopathological examination. The study was approved by the Institutional Research Ethics Committee of the First Affiliated Hospital of Nanjing Medical University.

Total RNA was extracted from cultured cell lines and clinical samples using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) and cDNA was synthesized using Primescript RT Reagent (Takara, Otsu, Japan) according to the manufacturer's instructions. The qRT-PCR was performed by using StepOne Plus Real-time PCR system (Applied Biosystems, Foster City, CA, USA) with SYBR® Premix Ex Taq™ Reagent (Takara). The following primers were used for qRT-PCR: Annexin A5, forward: 5′-AGCGGGCTGATGCAGAAAC-3′, reverse: 5′-ACTTCGGGATGTCAACAGAGT-3′; β-actin, forward: 5′-CCTGGCACCCAGCACAAT-3′, reverse: 5′-GCTGATCCACATCTGCTGGAA-3′. Data analysis was performed with ABI Step One Software version 2.1 and the relative mRNA level was calculated using 2^−ΔΔCt method.

Cells or frozen tissues were lysed in cell lysis buffer for 30 min on ice and centrifuged at 14,000 × g at 4°C for 15 min. The total protein concentration was calculated by the BCA Protein Assay kit (Pierce, Rockford, IL, USA). Proteins were separated by 10% SDS-PAGE gel and transferred onto a polyvinylidene difluoride membrane (Millipore, Billerica, MA, USA). Western blot analysis followed a standard procedure. The primary antibody Annexin A5 was obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA, USA). N-cadherin, vimentin, β-catenin, MMP2, MMP9, PI3K, phospho-PI3K, AKT, phospho-AKT, mTOR, phospho-mTOR were obtained from Cell Signaling Technology, Danvers, MA, USA. The anti-mouse and anti-rabbit secondary antibodies were also from Cell Signaling Technology.

The human renal cancer cell lines (Caki-1, Caki-2, ACHN, 769P) and normal epithelial cells of renal tubule (HK2) were obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA). The cells were cultured in McCoy's 5A or DMEM media supplemented with 10% fetal bovine serum (Gibco, Carlsbad, CA, USA) and 1% penicillin/streptomycin (Invitrogen) at 5% CO₂ and 37°C incubator. LY294002, a phosphatidylinositol 3-kinase (PI3K) inhibitor, was obtained from Selleck Chemical (Houston, TX, USA) (no. S1105).

Lentivirus packaging cells were transfected with LV3-pGLV-h1-GFP-puro vector (GenePharma, Shanghai, China) containing either the Annexin A5 knockdown (shA5-1 and shA5-2) or Annexin A5 overexpression (A5) and a negative control sequence (NC), respectively. Lentiviral transduction was performed in Caki-1 and Caki-2 cell lines. Pools of stable transductants were generated by selection using puromycin (4 μg/ml) for 2 weeks.

A Cell Counting Kit-8 assay (Dojindo Laboratories, Kumamoto, Japan) was used to estimate the proliferation potential. Cells were seeded in 96-well plates with 3000 cells/well. CCK-8 reagents were added into wells after cells grew for 1, 2, 3, and 4 days, respectively, and the absorbance was measured at 450 nm using a micro-plate reader at 2 h after CCK-8 addition.

Cells were seeded into 6-well plates (600 cells/well) and cultured in media containing 10% FBS for 2 weeks. Then the colonies were fixed with paraform and stained with 0.1% crystal violet. The colonies were counted and each group were repeated three times.

Cell migration or invasion assay were performed using a 24-well Transwell chamber (Costar, Corning, NY, USA) with or without Matrigel (Invitrogen). Cells (2×10⁴) were suspended in serum-free medium and seeded into the upper chambers which were inserted in the 24-well plate. Medium containing 10% FBS was added to the lower chamber. Cells were incubated at 37°C for 48 h and then cells on the surface of the upper chambers that did not migrate through the pores were removed with a cotton swab. Cells which migrated to the bottom surface of the chamber were stained with 0.1% crystal violet. Numbers of migratory and invasive cells were counted in five randomly selected fields and the presented data represent three individual experiments.

Mouse studies were approved by the Animal Research Ethics Committee of Nanjing Medical University. The 5-week-old female nude mice were randomly divided into two groups consisting of five mice each. The stable cells (7×10⁶) shA5-Caki-1 and the control cells (NC-Caki-1) were suspended in 150 μl PBS and injected subcutaneously into the flank of each mouse. Tumor size was calculated (length × width² ×0.52) once a week. After 6 weeks, tumors were removed, weighed, fixed, and embedded for immunohistochemical staining.

Immunohistochemistry was performed on tissue microarray to evaluate Annexin A5 protein expression. Tissue microarray was incubated with a primary antibody against Annexin A5 at 4°C overnight and incubated with HRP conjugated secondary antibody followed by DAB staining. Immunohistochemistry staining was assessed by two experienced pathologists. To evaluate the expression of Annexin A5 in RCC tissues, a semi-quantitative scoring system (0–3) was used based on the staining intensity of the tumor tissue: 0, negative; 1, weak positive; 2, moderate positive; 3, strong positive. Annexin A5 high expresion refers to scores 2–3 and Annexin A5 low expresion refers to scores 0–1. For assessing the association of Annexin A5 expression with clinicopathological characteristics of the RCC patients, following parameters were included: age (≤60 and >60), gender, tumor size (≤4 and >4), Histology (clear cell carcinoma and others), histological stage (grades I, II, III, and IV) and tumor stage (TNM stages I, II, III, and IV).

Statistical analyses were performed with SPSS 22.0 software. All the data were presented as the mean ± SD from three independent experiments. Student's t-test and the Chi-square test were used to analyze the differences between groups and survival curves were drawn by the Kaplan-Meier method. P<0.05 was considered to indicate a statistical significant difference.

To explore the protein and mRNA expression of Annexin A5 in RCC cell lines, western blot and qRT-PCR were performed in four human RCC cell lines (Caki-1, Caki-2, ACHN and 769P) and one normal epithelial cell of renal tubule (HK2). As shown in Fig. 1A, Annexin A5 mRNA was upregulated by 61.2- to 92.4-fold in all RCC cell lines compared to HK2 and western blot analysis showed similar results. To assess whether Annexin A5 was also highly expressed in RCC tissues, 22 pairs of RCC tissues and matched adjacent non-cancerous tissue were selected for qRT-PCR and western blot, the results suggested that Annexin A5 expression was markedly upregulated in both protein and mRNA level compared to the adjacent tissue (Fig. 1B–D).

To validate the association between Annexin A5 expression and clinicopathologic features, 123 cases of RCC were evaluated by IHC. Patients were divided into a low expression group (n=56) and a high expression group (n=67) according to the Annexin A5 expression. As shown in Table I, although Annexin A5 expression and age (P=0.628), gender (P=0.576), tumor size (P=0.192) or tumor histology (P=0.637) did not significantly correlate, Annexin A5 exhibited a signifcantly positive correlation with histological grade (P=0.007) and TNM stage (P=0.013). High expression group was correlated with higher histological grade and TNM stage compared with the low expression group. As shown in Fig. 2, RCC patients with high Annexin A5 expression had a significantly shorter overall survival time than those with low Annexin A5 expression (P=0.031). These data suggested that the upregulated expression of Annexin A5 might have an important role during RCC progression.

To further investigate the functions of Annexin A5 in RCC cell lines, we transfected Caki-1 and Caki-2 cells with lentivirus to silence or overexpress the expression of Annexin A5. The silenced cells were named as shA5-1 and shA5-2 and the overexpressed cell lines were called A5, while the matched control cells were named as NC. The expression levels were confirmed by western blotting (Fig. 3A and E) and qRT-PCR (Fig. 3B and F, P<0.01).

The growth curve analysis demonstrated that the downregulation of Annexin A5 in Caki-1 and Caki-2 cells significantly inhibited cell growth compared with the control cells (Fig. 3C and D, P<0.05). Whereas, upregulated Annexin A5 expression enhanced cell growth markedly in Caki-1 and Caki-2 cells (Fig. 3G and H, P<0.05). In addition, a colony formation assay was also performed to further investigate the effect of Annexin A5 on the proliferation. The results indicated that Annexin A5 knockdown contributed significantly to decrease cell colony formation efficiency, whereas inducing Annexin A5 expression markedly enhanced the ability of Caki-1 and Caki-2 cells to form colonies (Fig. 4A and B, P<0.05).

To further study the mechanism by which Annexin A5 overexpression or knockdown affected proliferation, we investigated the effects of Annexin A5 overexpression or knockdown on the PI3K/AKT/mTOR pathway. Western blot analysis suggested that the levels of phospho-PI3K, phospho-AKT and phospho-mTOR decreased in the Annexin A5 knockdown cells, while increased in the Annexin A5 over-expressed cells. However, the total levels of PI3K, AKT and mTOR had no obvious change in Annexin A5 overexpression or knockdown cells (Fig. 4C and D, P<0.05).

To further confirm our hypothesis, we used PI3K inhibitor (LY294002) at a dose of 20 μM to observe the role of PI3K/AKT/mTOR pathway in Annexin A5-regulated cell proliferation. The dose of LY294002 selected was the dose of IC₅₀ to Caki-1 cells. After treated with LY294002 for 48 h, Caki-1 and Caki-2 with Annexin A5 overexpression (A5) and control cells (NC) were used for CCK8 analysis. As shown in Fig. 4E, Annexin A5 overexpression increased the cell proliferation while LY294002 reversed the cell growth promoting trend by Annexin A5. Overall, these results suggested that Annexin A5 positively regulated cell proliferation in RCC and might be mediated by the phosphorylation of PI3K/AKT/mTOR pathway.

To assess the potential role of Annexin A5 in regulating the migration and invasion ability of RCC cells, Transwell migration and invasion assays were performed in Caki-1 and Caki-2 cells with Annexin A5 overexpression or knockdown. Transwell migration assays revealed that Annexin A5 knockdown reduced the migration capability of Caki-1 and Caki-2 cells and overexpressing Annexin A5 signifcantly increased the ability of migration. The Transwell invasion assays showed similar results (Fig. 5A–D, P<0.01).

To further explore the mechanism by which Annexin A5 affects cell migration and invasion, western blotting was performed to determine the expression levels of N-cadherin, vimentin, β-catenin, MMP2 and MMP9, which are related to regulating the cell adhesion and metastasis. The results demonstrated that the expression of these proteins was significantly downregulated in Annexin A5 knockdown cells while they were upregulated in Annexin A5 overexpressing cells (Fig. 5E and F). These data suggested that Annexin A5 could promote migration and invasion potential in RCC cells and the underlying mechanism might be the upregulation of adhesion and metastasis-related molecules.

To study the effect of Annexin A5 expression on tumor growth in vivo, we constructed subcutaneous xenograft tumor model using Caki-1 cells with Annexin A5 knockdown and control cells in the female nude mice. As shown in Fig. 6A and B, tumors from the Annexin A5 knockdown group (shA5) grew markedly slower than those from control group (NC). The mean tumor weight of shA5 group was lower compared to the NC group (Fig. 6C) and Ki-67, a proliferation marker of tumor, was markedly decreased in tumors from shA5 group (Fig. 6D and E). Collectively, these results showed that Annexin A5 knockdown could significantly impede tumor growth in vivo.