Forty-seven human RCC and 11 matched normal kidney specimens were obtained from patients who underwent surgical resection with the approval of the Institutional Review Board (IRB). Immunohistochemistry was performed as described previously (16). Briefly, paraffin-embedded sections were subjected to deparaffinization, rehydration, and heat-induced antigen retrieval. After blocking of endogenous peroxidase with 3% hydrogen peroxide, sections were subsequently incubated with primary SPOP antibody (Santa Cruz Biotechnology), horseradish-peroxidase-labeled dextran polymer (Dako EnVision™) and developed with 3,3′-diaminobenzidine chromogen followed by counter staining with hematoxylin. All stains were assessed by an independent pathologist according to the histologic scoring system (H-score) based on the product of staining intensity (0, no staining; 1, weak; 2, moderate; and 3, strong) and percentage of stained cells (0, 0%; 1, 1–30%; 2, 31+70%; and 3, 71–100%). The expression of SPOP in each tissue was considered either negative (H-score, <2) or positive (H-score, >2).

Human normal kidney cell HK-2, RCC cells 786-0, A498, RCC4 and 769-P were maintained in RPMI-1640 medium (Gibco) supplied with 10% fetal bovine serum (FBS), RCC cells ACHN, CAKI-1, CAKI2 and A498 were maintained in Dulbecco’s modified Eagle’s medium (DMEM) (Gibco) supplied with 10% FBS. All the cells were cultured in a humidified incubator containing 5% CO₂ at 37°C.

Human SPOP cDNA cloned into pDONR221 vector was obtained from the DNASU Plasmid Repository. Control siRNA (si-NC) and siRNA specifically targeting SPOP (si-SPOP) were from Guangzhou RiboBio Co., Ltd. For transfections, 2×10⁵ cells were seeded in 6-well plate and cultured overnight, SPOP plasmids or siRNAs were transfected into cells by Xfect™ transfection reagent (Clontech) according to the manufacturer’s instructions. Twenty-four hours (h) after the transfection, cell protein or RNA was collected for further assays.

Matrigel-coated Transwell chambers were applied to examine RCC cell in vitro invasive ability. RCC cells were pre-transfected with the indicated plasmids or siRNAs, 100 μl 0.5% FBS medium of 3×10⁴ cell suspension was then planted into the upper chamber, and 600 μl of 10% FBS medium was supplied to the lower chamber. Cells were cultured at 37°C in 5% CO₂ for 24 h. Invaded cells onto the lower surface of the upper chambers were stained with 0.5% crystal violent (Sigma) and photographed and counted.

Microslide cultured cells were fixed with 4% paraformaldehyde, permeabilized with 0.3% Triton X-100 and blocked with 5% bovine serum albumin. Cells were incubated with β-catenin primary antibody (Cell Signaling Technology) overnight at 4°C and subsequently incubated with AlexaFluor 488-conjugated secondary antibody (Sigma) for 1 h at room temperature, followed by nuclear staining with 4,6-diamidino-2-phenylindole and fluorescence was visualized by fluorescence microscopy (Olympus Optical Co.)

Cell total RNA was extracted with RNeasy mini kit (Qiagen) and reverse transcribed with cDNA synthesis kit (Invitrogen). Real-time PCR analysis was set up with SYBR Green qPCR Supermix kit (Invitrogen) supplied with commercial primers specific for the indicated genes, and carried out in the iCycler thermal cycler (Bio-Rad). The relative level of mRNA expression of each gene was determined by normalizing with an internal control gene GAPDH.

Western blotting was performed as previously described (17). Cells were first lysed and total proteins were collected, equivalent amounts of protein were separated on 10% NuPAGE Bis-Tris gels (Invitrogen) and transferred to nitrocellulose membranes. Membranes were blocked with 3% skim-milk (w/v), and incubated with primary antibodies overnight at 4°C. After washing, membranes were incubated with appropriate secondary antibodies conjugated with horseradish peroxidase and signals were then detected by chemiluminescence (Pierce). Primary SPOP, vimentin, pan-cytokeratin (Pan-CK), TCF4 and GAPDH antibodies were purchased from Santa Cruz Biotechnology; E-cadherin and α-SMA antibodies were from BD Biosciences, ZEB1 and MMP-2 antibodies were from Cell Signaling Technology.

The RNA-sequencing-based mRNA expression data for SPOP, TCF4, ZEB1 genes and the reverse phase protein array-based protein expression data for β-catenin of human clear cell RCC samples were all retrieved from The Cancer Genome Atlas (TCGA) Data Portal (18). Gene microarray data for SPOP of human normal kidney and clear cell RCC tissues were retrieved from the GEO datasets (GSE14994, GSE781 and GSE15641). SPOP gene microarray data and DNA copy number data of multiple types of cancer cell lines were retrieved from the Cancer Cell Line Encytopedia (CCLE) datasets. The Kaplan-Meier analysis (long-rank test) was performed to analyze recurrence-free survival. Pearson’s correlation coefficient was used to test the association between genes. Data from in vitro assay are presented as the mean ± SEM from three independent experiments, and the differences between two groups were compared by Student’s two-tail t-test. All statistical analyses were performed by GraphPad Prism6 (GraphPad Software).

Previous studies suggest that overexpression of SPOP may lead to dysregulation of pathways involved in tumorigenesis (8,9). We assessed SPOP expression in urological tumors including prostate cancer, bladder cancer, RCC and normal kidney tissues by immunohistochemistry. Interestingly, we found SPOP was negative in prostate cancer, bladder cancer and normal kidney tissues, but it was highly expressed in RCC tissues (Fig. 1A). RCC is a heterogeneous group of tumors with distinct histological subtypes, including clear cell, papillary, chromophobe, and other rare subtypes in addition to oncocytoma (19). When RCC subtypes were stratified, we found that the papillary, chromphobe or oncocytoma RCC were weak or negative for SPOP, but clear cell RCC was positively stained with this protein (Fig. 1B). In total we analyzed 47 clear cell RCC and matched 11 normal kidney tissues, and the results showed that 83% clear cell RCC were positive for SPOP, while only 18% normal tissues were positive (Table I). This indicates that SPOP is highly expressed in clear cell RCC and may serve as a specific biomarker for this type of RCC.

In addition to the determination of SPOP protein status in clear cell RCC, we checked SPOP mRNA expression by analysis of gene microarray data of normal kidney and clear cell RCC from the GEO datasets. Results from three independent datasets consistently showed that SPOP mRNA was significantly upregulated in clear cell RCC compared to normal kidney (Fig. 2A). To further explore whether SPOP upregulation is due to genomic abnormality, we analyzed the association of SPOP mRNA level and its DNA copy number in multiple types of cancer cell lines including 1,014 samples from the CCLE datasets, and found there was a positive correlation between SPOP mRNA level and its DNA copy number (Fig. 2B), and in RCC cell lines, a similarly positive correlation was also found (Fig. 2C). These data suggest that the upregulation of SPOP in RCC may be due to the genomic variation.

Previous studies indicate that SPOP plays important roles during tumor cell apoptosis and proliferation (7,10,15), we further investigated the potential functions of SPOP in tumor progression. The expression of SPOP in human clear cell RCC with local invasion (tumor cell invaded into perirenal fat, renal capsule or regional lymph node) was detected by immunohistochemistry, notably, the results showed that almost all the RCC with local invasion were SPOP-positive (Fig. 3A). We compared SPOP expression in RCC with different pathological stages according to the 2010 AJCC TNM classfication (20), and found RCC in T3–4 stages (primary tumors with local invasion) showed much high frequency of SPOP positive staining compared to RCC in T1–2 stages (without local invasion). Similarly, RCC with lymph node invasion (N1) or distant metastasis (M1) showed very high frequency of SPOP positive staining compared to RCC in N0 (without lymph node invasion) or M0 (without distant metastasis) (Table II). These date suggest SPOP is associated with clear cell RCC invasion and metastasis. We further analyzed the association of SPOP expression and tumor recurrence-free survival, and the result demonstrated that SPOP was negatively correlated with RCC recurrence-free survival (Fig. 3B), indicating SPOP as novel prognostic marker for RCC patients.

To confirm the biological function of SPOP in RCC invasion, in vitro cell line based assays were performed. Profile of SPOP expression in a series of RCC cell lines by RT-PCR showed ACHN cells with low SPOP expression, thus, it was applied for SPOP overexpression model, while A498 cells with high SPOP was applied for SPOP silencing model (Fig. 4A and B). In vitro Transwell invasion assays demonstrated that overexpression of SPOP promoted ACHN invasion (Fig. 4C), while silencing of SPOP by siRNA in A498 cells suppressed cell invasion (Fig. 4D). It has been well documented that epithelial-mesenchymal transition (EMT) is a process thought to initiate metastasis, and it is characterized by the gain of mesenchymal markers (e.g., vimentin, α-SMA, MMP2) and the loss of epithelial markers (e.g., E-cadherin, cytokeratins), as well as increased motility and invasion of cancer cells (21,22). We examined the expression of EMT markers in SPOP overexpressing or silencing cells, and the results showed that SPOP downregulated epithelial markers, such as E-cadherin and Pan-CK, and upregulated mesenchymal makers, such as vimentin, α-SMA and MMP-2 (Fig. 4E). These data indicate SPOP as an inducer for the invasiveness of RCC cells.

We further dissected the molecular mechanisms of SPOP in regulation of RCC invasion. Firstly, the associations between SPOP and a panel of EMT markers and EMT-inducing transcription factors (EMT-TFs) were analyzed in human clear cell RCC samples from TCGA datasets by Pearson correlation analyses. The results showed negative correlations between SPOP and epithelial marker CDH1 (E-cadherin), and positive correlations between SPOP and mesenchymal makers VIM (vimentin), ACTA2 (α-SMA) and MMP-2 (Table III). Importantly, SPOP was positively correlated with many critical EMT-TFs, such as TCF4 and ZEB1 (Table III and Fig. 5A). Further cell line based assays confirmed that overexpression of SPOP upregulated TCF4 and ZEB1 expression (Fig. 5B). ZEB1 is well known as one of the most critical transcriptional factors driving EMT in many cancer cells (23), and β-catenin/TCF4 complex has been demonstrated to bind ZEB1 gene promoter region and promote its transcription (24). Our data showed that silencing of TCF4 in RCC cells suppressed ZEB1 expression (Fig. 5C), and there was an extremely positive correlation between TCF4 and ZEB1 mRNA in clear cell RCC samples (Fig. 5D), these data again confirmed that TCF4 is an upstream regulator for ZEB1 expression. Additionally, we observed that SPOP could also upregulate cytosolic β-catenin protein expression and promote its nuclear translocation (Fig. 5E). Silencing of β-catenin suppressed ZEB1 expression (Fig. 5F), and there was a positive correlation between β-catenin protein and ZEB1 expression in clear cell RCC samples (Fig. 5G), suggesting β-catenin as upstream regulator for ZEB1. Furthermore, although overexpression of SPOP upregulated ZEB1 expression as well as cell invasion, co-transfection of TCF4 siRNA or β-catenin siRNA could ablate the effects of SPOP on ZEB1 gene expression and cell invasive ability (Fig. 5H and I). Taken together, our results indicate that SPOP promotes ZEB1 to drive RCC cell invasion via activating the β-catenin/TCF4 complex (Fig. 6).