Downregulation of SAV1 plays a role in pathogenesis of high-grade clear cell renal cell carcinoma

The renal cell carcinoma cell lines 786-O (#CRL-1932), 769-P (#CRL-1933) and Caki-2 (#HTB-47), and the human cell line HK2 (#CRL-1427), were purchased from the American Type Culture Collection (ATCC) (Rockville, MD). KMRC-1, KMRC-2, KMRC-3, and KMRC-20 were purchased from JCRB (Osaka, Japan), and TUHR4TKB was provided by RIKEN BRC through the National Bio-Resource Project of MEXT, Japan. RPTEC was purchased from Lonza Walkersville Inc. (Walkersville, MD). All cell lines were maintained in accordance with the supplier's instructions.

Primary ccRCCs were surgically resected at Oita University Hospital, and diagnosed histopathologically as described previously [7]. Information on the other 98 patients is summarized in Additional file 1: Table S1. Use of the tissue samples for all experiments was approved by all the patients and by Oita University Ethics Committee (Approval No. P-05-05).

Immunohistochemistry of paraffin-embedded RCC tissue sections was performed in a similar way to that used in our previous work [9]. Anti-SAV1 antibody (clone 3B320; Abnova, Taipei, Taiwan) and anti-YAP antibody (Cell Signaling Technology, Danvers) were used as the primary antibodies.

Quantitative RT-PCR was performed with a Universal probe library (Roche Diagnostics, Mannheim, Germany) and LightCycler 480 probe master (Roche Diagnostics) by the Taqman method as described previously [10]. Levels of messenger RNA expression relative to KPNA6 were obtained from a standard curve. We used KPNA6 as a control in this study because KPNA6 expression was not statistically different among the ccRCC and non-neoplastic tissues, although expression levels of known markers such as beta-2-microgloblin and glyceraldehydes 3-phosphate dehydrogenase (GAPDH) were variable across the samples [11].

The full-length SAV1 cDNA was obtained from NBRC (NITE Biological Resource Center, Chiba, Japan), cloned into pLenti7.3/V5-DEST (Invitrogen, Carlsbad, CA, USA) using the Gateway system^™ in accordance with the manufacturer's instructions, and used to generate the vector SAV1-pLenti7.3/V5-DEST (pLenti-SAV1). Transient transfections of the pLenti7.3/V5-DEST empty vector and pLenti-SAV1 with ViraPower Packaging Mix (Invitrogen) into HEK293T cells were performed using Lipofectamine2000 (Invitrogen) in accordance with the manufacturer's instructions. After 48 h, the viral supernatants were collected, filtered, and enriched by ultracentrifugation. Subsequently, transduction of target cells was performed at an optimized multiplicity of infection (MOI) of 5 with Polybrene at a final concentration of 6.0 μg/ml. To identify stable clones of SAV1-re-expressing cells, the full-length SAV1 cDNA was cloned into pLenti6.3/V5-DEST (Invitrogen). Lentivirus expressing SAV1 was generated as described above. 786-O cells were transduced with SAV1-pLenti6.3/V5-DEST or the pLenti6.3/V5-DEST vector, respectively. After transduction and selection with 0.5 μg/ml blasticidin (Invitrogen), two independent clones for SAV1 were obtained, and designated as SAV1-1 and SAV1-2. Control cells were also generated using the pLenti6.3/V5-DEST empty vector.

One day after transduction of 786-O or 769-P cells with lentivirus expressing SAV1, one fiftieth of the cells were reseeded on 100-mm dishes and cultured. When colonies became visible, they were stained with Giemsa for 30 min, rinsed with water, and counted.

Cells were transfected with Stealth^™ RNAi oligonucleotide or the Stealth^™ RNAi Negative Control Duplex with the corresponding GC content (Invitrogen) at a final concentration of 10 nM using Lipofectamine RNA-MAX (Invitrogen) in accordance with the manufacturer's instructions.

After transfection of siRNAs in a 96-well plate, MTS assay was carried out using the CellTiter 96^RAQ_ueous One Solution Cell Proliferation Assay Kit (Promega, Madison, WI, USA), and the optical density was measured at 492 nm using a fluorescence reader (Tecan SpectraFluor) (Tecan, Crailsheim, Germany). In addition, BrdU incorporation was performed using a Cell Proliferation ELISA Kit (Roche Applied Science, Mannheim, Germany) in accordance with the manufacturer's protocols.

Cells were transfected with siRNA in a 96-well plate and cultured for 72 h. Nuclear morphology was then evaluated for apoptosis. DNA fragmentation was detected in the 96-well format using Cell Death Detection ELISA plus (Roche Applied Science) in accordance with the manufacturer's instructions. The activity of caspase 3 and caspase 7 was detected in the 96-well format by using the caspase-Glo 3/7 Assay (Promega) according to the manufacturer's instructions.

Western blot analysis was performed similarly to that in our previous study [12]. In each analysis, 10 μg of cell lysate was used. The primary antibodies employed were: anti-human GAPDH antibody (Sigma-Aldrich), anti-SAV1 antibody (Cell Signaling Technology), anti-phosphorylated YAP antibody (Ser127; Cell Signaling Technology), and anti-YAP antibody (Santa Cruz Biotechnology). Detection was performed with ECL Western Blotting Detection Reagents (Amersham Biosciences, Piscataway, NJ, USA) in accordance with the manufacturer's instructions.

For the luciferase reporter assay, stable clones of 786-O cells (5 × 10⁴) were seeded in 6-well plates, and then co-transfected with 0.5 μg of pGL4.35 (9xUAS Gal4) and GAL4-TEADs, respectively, using Lipofectamine Plus (Invitrogen). At 48 h after transfection, the cells were lysed and luciferase activity was assayed using a Dual-luciferase Reporter Assay kit (Promega) in accordance with the manufacturer's instructions. All luciferase activities were normalized to the Renilla luciferase reporter pRL-CMV plasmid (Promega).

Quantitative RT-PCR data were analyzed statistically by the Mann-Whitney U-test. The two-sided Student's t test and Fisher's exact test was used to analyze differences in experimental data obtained from the cell lines. Differences at P < 0.05 were considered statistically significant.

Additional methods are described in Supplementary methods (Additional file 2: Methods S1)

We performed array CGH analysis and gene expression analysis of 8 RCC cell lines (786-O, 769-P, KMRC-1, KMRC-2, KMRC-3, KMRC-20, TUHR4TKB, and Caki-2) in order to identify downregulated genes located in the region of 14q loss (Additional file 3: Figure S1a, [7]). All the data obtained in the array CGH and expression microarray analysis are available at DDBJ via CIBEX (http://​cibex.​nig.​ac.​jp/​index.​jsp), under accession numbers CBX97 for array CGH, and CBX96 for expression-microarray. Copy number loss at 14q was detected in all the cell lines analyzed. Detailed analysis of the region of copy number loss at 14q revealed two different homozygous deletions at 14q in 3 (786-O, 769-P, KMRC-1) of the 8 cell lines.

In 786-O, a homozygous loss was detected at 14q22.1 (Additional file 3: Figure S1b), and this encompassed three genes: SPG, SAV1, and MAP4K5 (Additional file 3: Figure S1c). On the other hand, in 769-P and KMRC-1 cells, a homozygous loss was detected at 14q23.1, where only the HIF-1a gene is located. Next, using array CGH data and transcriptome data for these 8 cell lines, we analyzed whether the expression levels of the SPG, SAV1, MAP4K5, and HIF-1a genes were correlated with their gene copy number status. As shown in Figure 1a, the SAV1 gene showed a positive correlation between the copy number change and the expression ratio (R = 0.8871; p < 0.005) (Figure 1a). In contrast, copy number loss was not significantly correlated with the level of mRNA expression for the SPG, MAP4K and HIF-1a genes (Additional file 4: Figure S2), suggesting that SAV1 downregulation is correlated with gene copy number loss at 14q in these cell lines. To confirm this trend in ccRCC cases, quantitative RT-PCR (qRT-PCR) analysis of ccRCC samples was performed, and this revealed that the levels of SAV1 mRNA in cases showing 14q loss (n = 10) were significantly lower than those without 14q loss (n = 12) (p < 0.005), although they were similar between cases without 14q loss and normal kidney (Figure 1b). Furthermore, qRT-PCR analysis of ccRCC samples revealed that the levels of SAV1 mRNA in high-grade CCCs (n = 7) were significantly lower than those in low-grade CCCs (n = 8) and normal kidney (n = 5) (p < 0.005) (Figure 1c). In addition, protein level of SAV1 in ccRCC cases was evaluated by immunohistochemistry. SAV1 protein was detected in proximal renal tubules, podocytes, and endothelial cells in normal tissues (Additional file 5: Figure S3). The specificity of SAV1 antibody was confirmed by immunocytochemistry of 786-O cells transduced with lentivirus expressing SAV1 or control virus (Additional file 6: Figure S4). In tumor cells, SAV1 downregulation at the protein level was observed in 64 of the 98 ccRCC cases, and it was correlated with tumor grade (Table 1). These findings suggest that SAV1 is downregulated in RCC cell lines as well as ccRCC samples and that its downregulation occurs preferentially in high-grade ccRCCs. To evaluate the SAV1 gene mutation, we performed PCR for 6 ccRCC cases with SAV1 downregulation using 5 pairs of primers within open reading frame and we could not detect any mutations (data not shown), suggesting that gene mutation does not contribute to SAV1 inactivation in RCC.

We next re-expressed SAV1 in 786-O cells, in which SAV1 is homozygously deleted, by transducing them with SAV1 (Figure 2a), and analyzed their colony-forming activity. As shown in Figure 2b, colony-forming activity was significantly reduced in SAV1-transduced 786-O cells. Similar observations were obtained for 769-P cells (Figure 2a, b). Furthermore, this trend was also confirmed by MTS assay (Additional file 7: Figure S5).

We next transfected small interfering RNAs (siRNAs), designed to target distinct sites of SAV1 mRNA, into the kidney epithelial cell line HK2 and renal proximal tubule epithelial cells (RPTECs), and analyzed their proliferation (Figure 3a). We found that cell proliferation was promoted in HK2 and RPTECs transfected with SAV1-siRNA (Figure 3b), whereas the cell cycle of HK2 cells was not affected by SAV1-siRNA transfection (Additional file 8: Figure S6). However, the apoptotic activity of HK2 cells transfected with SAV1-siRNA was significantly reduced in cells transfected with SAV1-siRNA in comparison with those transfected with the control (Figure 3c), suggesting that SAV1 downregulation inhibits apoptosis in renal tubule cells.

It has been reported that SAV1 may function as a component of the Hippo signaling pathway, and the Hippo pathway targets the transcriptional co-activator YAP1 [13‐17]. YAP1 is phosphorylated by LATS1/2, and the phosphorylated YAP1 is then kept localized in the cytoplasm. On the other hand, when Hippo signaling is inhibited, unphosphorylated YAP1 accumulates in the nucleus, where it forms a transcriptionally active complex with the transcription factor, TEAD [18, 19]. To determine whether Hippo signaling is actually inhibited in RCCs in which SAV1 is frequently downregulated, we analyzed the phosphorylation status of YAP1 in 786-O and 769-P cells in which SAV1 are downregulated and HK2 cells in which SAV1 is expressed at a comparable level (Figure 4a). As shown in Figure 4a, phosphorylated YAP1 was significantly decreased in 786-O and 769-P cells, relative to HK2 cells. Furthermore, YAP1 phosphorylation was enhanced when SAV1 was re-expressed in 786-O cells (Figure 4b).

It has been reported that unphosphorylated YAP1 binds to members of the TEAD family, including TEAD1, 2, 3 and 4, and plays a functional role as a transcriptional co-activator [17]. Therefore, we next examined whether SAV1 affects the transcriptional activity of the YAP1-TEAD complex in RCCs. SAV1-transduced 786-O cells were transfected with expression plasmids encoding a Gal4 DNA binding domain fused to TEAD1, TEAD2, and TEAD3 (Gal4-TEADs), a Gal4-9x UAS luciferase reporter (pGL4.31) and pRL-CMV (Figure 4c) (Additional file 9: Figure S7). In control cells, co-transfection of the Gal4-TEAD3 and Gal4-9x UAS luciferase reporter led to an increase in luciferase activity. On the other hand, in SAV1-transduced 786-O cells, the luciferase activity eventually decreased, suggesting that, in RCCs with suppression of Hippo signaling, YAP1 activity is significantly elevated (Figure 4c).

To further confirm whether intracellular localization of YAP1 is under the control of Hippo signaling, we immunohistochemically analyzed the 98 cases of ccRCC for expression of YAP1. We compared the intracellular localization of YAP1 between ccRCCs showing downregulation of SAV1 and those without SAV1 downregulation. As shown in Figure 5 and Table 2, nuclear localization of YAP1 was more frequently detectable in ccRCCs with SAV1 downregulation than in those without, suggesting that SAV1 downregulation leads to nuclear translocation of YAP1. When we evaluated the correlation between nuclear localization of YAP1 and tumor grade, nuclear localization of YAP1 was preferentially detected in high-grade (Table 3), suggesting that the suppression of Hippo signaling through SAV1 downregulation tends to occur in high-grade ccRCCs.