Overexpression of Csk-binding protein contributes to renal ... - Nature

Oncogene (2009) 28, 3320–3331

& 2009 Macmillan Publishers Limited All rights reserved 0950-9232/09 $32.00 www.nature.com/onc

ORIGINAL ARTICLE

Overexpression of Csk-binding protein contributes to renal cell carcinogenesis X Feng1,2,5, X Lu1,2,5, X Man3, W Zhou1,2, LQ Jiang1,6, P Knyazev4, L Lei1,2, Q Huang1,2, A Ullrich4, Z Zhang1 and Z Chen1 1

State Key Laboratory of Molecular Biology, Institute of Biochemistry and Cell Biology, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, Shanghai, China; 2Graduate University of the Chinese Academy of Sciences, Chinese Academy of Sciences, Beijing, China; 3International Co-operation Laboratory on Signal Transduction, Eastern Hepotabiliary Surgical Institute, The Second Military Medical University, Shanghai, China and 4Department of Molecular Biology, Max-Planck-Institute of Biochemistry, Martinsried, Germany

C-terminal Src kinase (Csk)-binding protein (Cbp) is a transmembrane adaptor protein that localizes exclusively in lipid rafts, where it regulates Src family kinase (SFK) activities through recruitment of Csk. Although SFKs are well known for their involvement in cancer, the function of Cbp in carcinogenesis remains largely unknown. In this study, we reported overexpression of Cbp in more than 70% of renal cell carcinoma (RCC) specimens and in the majority of tested RCC cell lines. Depletion of Cbp in RCC cells by RNA interference led to remarkable inhibition of cell proliferation, migration, anchorage-independent growth as well as tumorigenicity in nude mice. Strikingly, silencing of Cbp negatively affected the sustaining of Erk1/2 activation but not c-Src activation induced by serum. Besides, the RhoA activity in RCC cells was remarkably impaired when Cbp was knocked down. Overexpression of wild-type Cbp, but not its mutant Cbp/DCP lacking C-terminal PDZ-binding motif, significantly enhanced RhoA activation and cell migration of RCC cells. These results provided new insights into the function of Cbp in modulating RhoA activation, by which Cbp might contribute to renal cell carcinogenesis. Oncogene (2009) 28, 3320–3331; doi:10.1038/onc.2009.185; published online 6 July 2009 Keywords: Csk-binding protein; renal cell carcinoma; carcinogenesis; RhoA; PDZ-binding motif

Introduction Renal cell carcinoma (RCC) is the most lethal of the common urologic cancers, accounting for approximately 3% of all adult malignancies (Lam et al., 2005). The major type of RCC, clear cell renal cell Correspondence: Dr Z Chen or Dr Z Zhang, Institute of Biochemistry and Cell Biology, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, 320 Yue Yang Road, Shanghai 200031, China. E-mails: [email protected] or [email protected] 5 These authors contributed equally to this work. 6 Current address: The Department of physiology and pharmacology, Section of Integrative Physiology, Karolinska Institute, Stockholm, Sweden. Received 1 December 2008; revised 23 May 2009; accepted 29 May 2009; published online 6 July 2009

carcinoma (ccRCC), represents 75% of all kidney tumors (Cohen and McGovern, 2005). RCC is one of the most challenging malignancies because of its resistance to cytotoxic chemotherapeutic agents and radiation therapy (D’Hondt et al., 2005). Recent investigations have been focused on the molecular mechanisms of potential targets such as von HippelLindau protein and hypoxia-inducible factors (HIFs) and their roles in oncogenesis and progression of RCC (Bratslavsky et al., 2007; Klatte et al., 2007; Rathmell and Chen, 2008). Despite these efforts, the precise molecular mechanisms of carcinogenesis, progression and metastasis of RCC remain largely unknown. The C-terminal Src kinase (Csk)-binding protein (Cbp), also known as PAG for phosphoprotein associated with glycosphingolipid-enriched microdomains, is a ubiquitously expressed transmembrane adaptor protein anchored in lipid rafts (Brdicka et al., 2000; Kawabuchi et al., 2000). Like other adaptor proteins, human Cbp has multiple protein–protein interacting motifs including 2 proline-rich domains, 10 putative phosphorylatable tyrosines and 1 PDZ (PSD95, DLG and ZO-1) domain-binding motif at the extreme carboxyl terminus (Brdicka et al., 2000; Kawabuchi et al., 2000). Previous studies reported that Cbp was involved in the regulation of Src family kinases (SFKs). SFKs phosphorylate multiple tyrosines of human Cbp including tyrosine 317, which recruits Csk, a negative regulator of SFKs, from the cytosol into lipid rafts where SFKs are localized, resulting in the inhibition of SFKs activation (Brdicka et al., 2000; Kawabuchi et al., 2000). This negative regulatory mechanism has been found in several systems, including T-cell activation, cell adhesion, spreading processes and growth factor signaling (Kawabuchi et al., 2000; Torgersen et al., 2001; Davidson et al., 2003; Shima et al., 2003; Matsuoka et al., 2004; Zhang et al., 2004). Besides recruiting Csk, Cbp also regulates SFKs activation by interacting directly with SFKs, such as Fyn and Lyn (Ingley et al., 2006; Davidson et al., 2007; Solheim et al., 2008). For Fyn, interaction with Cbp made it insensitive to Csk negative regulation and thus maintained its activity (Solheim et al., 2008). These studies indicate that Cbp both negatively and positively regulates SFKs.

Involvement of Cbp in renal cell carcinogenesis X Feng et al

3321

Besides the classical SFKs activities regulation, Cbp is implicated in some other pathways as yeast two-hybrid analysis demonstrated that Cbp binds to EBP50 (ezrin/ radixin/moesin (ERM)-binding phosphoprotein of 50 kDa), an cytosolic adaptor protein that associates with ERM proteins (Brdickova et al., 2001; Itoh et al., 2002). Although it has been shown that Cbp negatively regulates T-cell activation by interacting with EBP50 (Itoh et al., 2002), the precise mechanism and biological significance of this interaction in other systems remain to be clarified. Recently, several lines of evidence have shown that Cbp is involved in cell transformation. Cbp negatively regulates epidermal growth factor (EGF)-induced NIH3T3 cell transformation by downregulating c-Src activity (Jiang et al., 2006). The direct binding of Cbp to c-Src sequesters active c-Src in lipid rafts, inhibiting its oncogenic potential in Csk/ MEF cells and several colon cancer cell lines (Oneyama et al., 2008). Theileria transformation promotes B cell proliferation by downregulating Cbp expression, thereby excluding Csk from Hck-positive microdomains (Baumgartner et al., 2003). These data indicate that Cbp serves as a tumor suppressor for some cancer types. However, Cbp has also been implicated as a positive cancer regulator. It is reported to form a protein complex with Lyn and active Stat3 in B-non-Hodgkin lymphomas. In this example, it appears to promote the oncogenic role of Lyn, because knocking down Cbp in these tumor cells leads to cell apoptosis (Tauzin et al., 2008). Moreover, strong expression of Cbp has also been found in a subset of lymphomas, including B-non-Hodgkin lymphomas (Svec et al., 2005; Tedoldi et al., 2006; Svec, 2007). The reports on different Cbp functions in different tumors implicated that the role of Cbp in tumors might involve some unidentified signaling pathways, besides regulating SFKs activation. To explore the role of Cbp in cancers, we determined the expression of Cbp in several malignant tumors including kidney carcinoma. We found that Cbp was overexpressed in 73% of RCC (in particular, in 83% ccRCC). Knockdown of Cbp in 786-0 cells resulted in remarkable impairment of tumor cell motility, transforming ability and tumorigenicity in nude mice. Our data further suggested a novel function of Cbp in regulating RhoA activation possibly via its PDZbinding motif. Results Cbp was overexpressed in renal cell carcinoma To determine Cbp expression in RCC, we performed northern blot analysis on 14 pairs of RCC and their matched noncancerous tissue samples. An overexpression of Cbp was found in 71% (10/14) of tested RCC tissues (Figure 1a). To further verify this finding, western blot analysis was performed on another 27 pairs of RCC protein samples. Clearly, Cbp was overexpressed at protein level in 74% (20/27) of tested RCC specimens (Figure 1b). In addition, overexpression

of Cbp was also detected in three out of four human RCC lines tested here, 786-0, A498 and Caki-1, when normal human kidney tissues were used as control (Figure 1c). The correlation between Cbp overexpression and histological classification of kidney carcinoma was then analysed. As shown in Table 1, Cbp overexpression was detected in 10 out of 12 cases of ccRCC, but not in other types of kidney cancers except for one sample from chromophobe RCC. Statistical analysis demonstrated that Cbp overexpression significantly correlated with ccRCC when all cases were classified into either ccRCC or non-clear cell RCC (P ¼ 0.028). Suppressing Cbp expression inhibited cell proliferation To investigate the function of Cbp in ccRCC, 786-0 cell line was chosen for further study due to its overexpression of Cbp. Two small interfering RNA (siRNAs: Cbp-Ri-3 and Cbp-Ri-5) targeting human Cbp were able to efficiently suppress endogenous Cbp expression in 786-0 cells (Supplementary Figure 1a). Construct of Cbp-Ri-3 shRNA was also generated and stably transfected into 786-0 cells. Two selected transfectants, 786-0/B5 and 786-0/B6, showed about 90 and 80% inhibition of Cbp expression, respectively, whereas Cbp expression remained unaffected in vector-transfected cells (Figure 2a, right panel). To examine the effects of Cbp silencing on the proliferation of RCC cells, we monitored cell proliferation of the two transfectants using MTT assay and Brdu (5-Bromo-20 -deoxy-uridine) incorporation assay. As shown in Figures 2a and b, proliferation of 786-0/B5 and 786-0/B6 were notably impaired compared to control cell lines 786-0 and 786-0/U6. Transient expression of the Cbp siRNAs, Cbp Ri-3 and Cbp Ri-5, could also significantly inhibit 786-0 cell proliferation when compared with scrambled siRNA-transfected cells (Supplementary Figure 1a and b). These data indicated a role of Cbp in cell proliferation in 786-0 cells. Silencing of Cbp inhibited the motility and invasiveness of RCC cells To explore the involvement of Cbp in cell motility, wound-healing assay was performed. As shown in Figure 3a, wound-healing process was remarkably inhibited in 786-0/B5 and 786-0/B6 when compared to their control cells. Then, transwell assays were carried out in these cells to quantitatively determine the effect of Cbp on cell migration. As shown in Figure 3b, similar numbers of wild-type and vector-transfected cells migrated to the lower face of the transwell membrane (786-0: 170±27; 786-0/U6: 177±8), whereas the Cbp knockdown cells exhibited a strongly inhibited motility, with only 42 and 21% cells migrating (786-0/B5: 74±17; 786-0/B6: 37±22). Transient transfection with Cbp-Ri-3 or Cbp-Ri-5 siRNAs resulted in a similar effect (Supplementary Figure 1c). We also investigated whether Cbp affected the invasive ability of RCC cells. In vitro invasion assay was performed in chambers with the upper wells coated Oncogene


3322

Figure 1 C-terminal Src kinase (Csk)-binding protein (Cbp) expression in renal cell carcinoma. (a) Left panel: Representative northern blot results showing Cbp mRNA levels in renal cell carcinoma (RCC) samples (T) versus paired noncancerous tissues (N). Upper panel, Cbp; lower panel, b-actin. KC: renal cell carcinoma. Right panel: Quantification of the northern blot of 14 pairs of RCC samples. Cbp expressions were normalized to b-actin level. (b) Western blot of Cbp in RCC specimens (T) and paired noncancerous tissues (N). b-actin was used as loading control. (c) Cbp expression in RCC cell lines. The expression of Cbp in normal tissues and four RCC cell lines was detected by anti-Cbp antibody in western blot. b-actin was used as loading control. Table 1 Differential expression of Cbp in subtypes of kidney cancers Percentage of P value Cbp upregulation (upregulated cases/total cases; proved by IB and RT–PCR) Histological classification Clear cell renal cell carcinoma Non-clear cell renal cell carcinoma Papillary renal cell carcinoma Chromophobe renal cell carcinoma Granular renal cell carcinoma Transitional cell carcinoma in renal pelvis

0.028a 83% 20% 0% 100% 0% 0%

(10/12) (1/5) (0/1) (1/1) (0/1) (0/2)

Abbreviations: Cbp, C-terminal Src kinase (Csk)-binding protein; IB, immunoblot; RT–PCR, reverse transcription-PCR. a P value was calculated using Fisher’s exact test. Po0.05 was regarded as significant difference.

with Matrigel to mimic the extracellular matrix. In sharp contrast to control cells, Cbp knockdown cells showed dramatically reduced invasive ability (Figure 3c). The Oncogene

number of invaded 786-0/B5 and B6 cells was 33% and 42% of that of the control cells, respectively (786-0: 83±7; 786-0/U6: 107±33; 786-0/B5: 27±3; 786-0/B6: 35±12). To exclude the possibility that the above inhibitory effects were off-target effects of the shRNA, scrambled Cbp shRNA was used as an additional control. We did not observed any differences regarding Cbp expression, cell shape, rates of cell proliferation and migration between parental 786-0, 786-0/U6 (empty vector control) and 786-0/3-neg (scrambled Cbp shRNA control) cells (Supplementary Figure 2 and data not shown). Depletion of Cbp suppressed tumorigenicity of RCC cells in nude mice To substantiate the role of Cbp in RCC cell malignancy, soft agar assay was employed. As shown in Figure 4a, the anchorage-independent growth was almost depleted in Cbp knockdown cells. Only few small pinpoint colonies consisting of two or three cells were formed.


3323

Figure 2 Knockdown of C-terminal Src kinase (Csk)-binding protein (Cbp) inhibited cell proliferation. (a) Left panel: Results of MTT assay performed in control cells and Cbp knockdown cells. 786-0: wild type; 786-0/U6: mock transfected; 786-0/B5 and 786-0/B6: two clones showing high efficiency of Cbp suppression. Right panel: Western blot analysis showing the efficiency of Cbp knockdown in the indicated cells. b-actin was used as loading control. (b) Brdu (5-Bromo-20 -deoxy-uridine) assay. Upper panel: representative images showing the Brdu incorporation in the indicated cells. Lower panel: statistical analysis of Brdu incorporation. Data were shown as the mean value from three independent experiments±s.d. (standard deviation). *Po0.05, **Po0.01.

In contrast, control cells formed large, visible colonies consisting of at least 50 cells. Statistical analysis of colony number (X50 cells) revealed a significant difference between the control cells and the Cbp knockdown cells (Figure 4a, right panel). To determine the effect of Cbp on tumorigenicity in vivo, control cells, 786-0/B5, 786-0/B6 and 786-0/Ri-3 pool, the pool cells of Ri-3 shRNA stable transfectants which showed a relatively low Cbp silencing efficiency (Figure 4b, right panel), were injected subcutaneously into nude mice. Tumor formation was monitored for 70 days. The control cells were able to form tumors in nude mice, but 786-0/B5 and 786-0/B6 cells failed to grow

(Figure 4b, left panel). Although 786-0/Ri-3 pool cells were able to form tumors, the tumor sizes were significantly smaller than those of the control cells (Figure 4b, left panel). These data suggested that tumor formation of 786-0 cells in nude mice was Cbp dependent. Cbp knockdown caused alterations in cell shape and F-actin cytoskeleton While generating the transfectants, we noticed that the cell shape of Cbp knockdown cells was dramatically changed. The parental and vector-transfected cells showed typical cobblestone epithelial-like or keratocyte-like cell morphology with apparent large lamelliOncogene


3324

Figure 3 Cell motility was impaired by knocking down C-terminal Src kinase (Csk)-binding protein (Cbp) expression. (a) Woundhealing assay. Control cells (786-0 and 786-0/U6) and Cbp knockdown cells (786-0/B5 and 786-0/B6) were grown to full confluence and then wounded with a pipette tip. Recovered area was measured after 10 h. Right panel: statistical analysis of relative wound recovery of each transfectant comparing to that of control cells. The result was expressed as the mean value±s.d. from three independent experiments. (b) Migration assay in transwell chambers. Cells that migrated from the upper well of a transwell chamber into the lower well were stained, photographed and counted. Data were presented as the mean value from three different fields±s.d. (c) Invasion assay of control and Cbp knockdown 786-0 cells in extracellular matrix (ECM)-coated transwell chambers. The experiment was performed as described in b. Data were presented as the mean value from three independent experiments±s.d. *Po0.05, **Po0.01.

podia, whereas the cell body of Cbp-knockdown cells was thinner and longer with small lamellipodia. Multiple long cell processes branched from Cbp knockdown cells, extending for 3–6 times the body length (Figure 5a, arrow). Furthermore, phalloidin staining revealed that 786-0/B5 and 786-0/B6 cells bore a few short, smeared, thin stress fibers, whereas control cells had long and clear F-actin bundles parallel to the long axis of the cell (Figure 5b), indicating that the morphological changes of the cells might be associated with disorganization of the actin cytoskeleton caused by Cbp knockdown. The mean fluorescence intensity (MFI) analysis confirmed that the amount of F-actin in Cbp knockdown cells was significantly decreased compared with that of control cells (Figure 5c). Oncogene

The effects of Cbp knockdown on the activation of c-Src, MAPK and Akt To explore the working mechanism of Cbp in RCC cells, we examined the activities of several important signaling transducers including c-Src, Akt and Erk1/2 in the steady state. To our surprise, we did not observe obvious alteration of c-Src, Akt and Erk1/2 activities either in 786-0/B5 and 786-0/B6 cells, or in the cells transiently transfected with the siRNAs (Supplementary Figures 3a, b and c and data not shown). We then investigated the serum-induced activation of those kinases in RCC cells. Although compared to control cells, an increase of c-Src kinase activity (indicated by pY 416 of c-Src) at basal level was observed in the Cbp-knockdown cells, the kinetics and intensity of


3325

Figure 4 Knockdown of C-terminal Src kinase (Csk)-binding protein (Cbp) in 786-0 cells inhibited anchorage independent growth and tumorigenicity in nude mice. (a) Soft agar assay. Cells were grown in soft agar culture medium for 10 days and colonies (X50 cells) were counted under a microscope. Left panel: Representative images showing the colony formation in soft agar. Right panel: Colony numbers were presented as the mean value±s.d.; **Po0.01. (b) Tumor formation in nude mice. An equal number of cells were inoculated subcutaneously into 5-week-old female nude mice. Tumor volume was estimated according to the following formula: volume ¼ 1/2 (largest diameter) (smallest diameter)2. The mean value±s.d. were from four animals in each group. Right panel: Cbp expression level in the indicated cells was assessed by western blot. b-actin was used as internal control. *Po0.05.

serum-induced c-Src activation remained unchanged (Figure 6a). In contrast to c-Src, the duration of Erk1/ 2 activation was notably shortened though its maximum activation achieved at 10 min serum stimulation was not affected by Cbp knockdown (Figure 6b). Silencing of Cbp had no effect on serum-induced Akt activation (Figure 6c). Depletion of Cbp led to a reduction of RhoA activation As small Rho GTPases are essential for the regulation of cell morphology and actin cytoskeleton, we examined whether RhoA, Cdc42 and Rac1 might be involved in Cbp knockdown-induced morphological changes. Interestingly, we found that RhoA activation was significantly suppressed in 786-0/B5 and B6 cells (Figure 7a), whereas the activation of Cdc42 and Rac1 remained unchanged (Supplementary Figure 4). Negative effect of Cbp downregulation on RhoA activation could also be observed in cells transiently transfected with the siRNAs (Figure 7b). These suggested a role of Cbp in regulating RhoA activation. The PDZ-binding motif of Cbp may mediate the regulation of RhoA activation Cbp is known to interact via its C-terminal PDZbinding motif with EBP50, which in turn binds to ERM family (Brdickova et al., 2001; Itoh et al., 2002). ERM proteins have been implicated in the regulation of cytoskeleton organization and Rho GTPases activation

(Bretscher, 1999; D’Angelo et al., 2007; Niggli and Rossy, 2008). Indeed, the interaction between Cbp and EBP50 was detected in 786-0 cells (Supplementary Figure 5a). To test whether the PDZ-binding motif of Cbp plays a role in RhoA activation, Cbp/DCP, a mutant lacking its C-terminal PDZ-binding motif, was constructed. This mutant failed to interact with EBP50 (Supplementary Figure 5b). As expected, overexpression of Cbp/WT, but not Cbp/DCP, resulted in a significant increase of RhoA activation in 786-0 cells (Figure 8a). To further study the role of PDZ-binding motif of Cbp in cell migration, Cbp/WT and Cbp/DCP were stably transfected into 786-0 cells. Expression of Cbp/WT, but not Cbp/DCP, could significantly enhance 786-0 cells migration (Figures 8b and c).

Discussion Cbp belongs to a protein family named TRAPs (transmembrane adaptor proteins). One of the characteristics of Cbp that distinguishes it from other TRAP members is its ubiquitous expression rather than restricted expression in lymphoid cells (Lindquist et al., 2003). In this study, we found that Cbp was overexpressed in RCC, particularly in ccRCC. Knockdown of Cbp expression strongly inhibited transforming ability and tumorigenicity of RCC cells in nude mice. Oncogene


3326

Figure 5 Changes in cell morphology and stress fiber formation in C-terminal Src kinase (Csk)-binding protein (Cbp) knockdown cells. (a) Altered cell morphology of Cbp knockdown cells. Arrows indicated the long branches in Cbp knockdown cells. (b) Representative pictures showing phalloidin-stained F-actin in control and Cbp knockdown cells. (c) Quantitation of mean fluorescence intensity of phalloidin-stained F-actin. Data were presented as the mean value±s.d. **Po0.01.

These data indicated the diagnostic and therapeutic value of Cbp for RCC. Involvement of Cbp in cancers has been recently concerned. High expression of Cbp has been seen in follicle-center-derived lymphoma and some diffuse large B-cell lymphomas, where it appears to promote the oncogenic role of Lyn (Tauzin et al., 2008). A human protein atlas for normal and cancer tissues based on antibody proteomics was newly developed (Uhlen et al., 2005; Mathivanan et al., 2008; http://www.proteinatlas. org/). This database revealed an overexpression of Cbp in kidney cancer and some lymphoma, but not in other tested cancer tissues. Consistent with this, a recently published paper has reported that Cbp expression is downregulated in some colon cancers (Oneyama et al., 2008). These data matched well with our results. Among the three cancer types we checked in this study, only RCC exhibited overexpression of Cbp. In non-small cell lung cancer, Cbp was downregulated, and in hepatocelOncogene

lular carcinoma, Cbp expression varied rather randomly (Supplementary Figure 6). These data suggest that overexpression of Cbp is not a universal phenomenon observed in cancers. An interesting hypothesis would be that Cbp plays diverse roles in different types of tumors depending on its expression level. Previous studies have shown the negative regulation of SFKs by Cbp in breast cancer cells and colon cancer cells where Cbp expression was reduced (Jiang et al., 2006; Oneyama et al., 2008). Yet, in our study, no effect of Cbp depletion on serum-induced c-Src activation was observed up to 60 min of serum stimulation (Figure 6a). Interestingly, Cbp knockdown did enhance the basal level c-Src activity (in serum starvation) indicating that the canonic Cbp function to negatively regulate c-Src activity still existed, but might not be a dominant regulatory mechanism in RCC cells cultured with serum. As serum is a mixed pool of many growth factors, its final effect on c-Src would be a balance of all the forces.


3327

Figure 6 Kinetics of c-Src, Erk1/2 and Akt activation in C-terminal Src kinase (Csk)-binding protein (Cbp) knockdown cells. Activation of c-Src (a), Erk1/2 (b) and Akt (c) on serum stimulation in Cbp knockdown cells was monitored at the indicated time. Total Erk1/2, c-Src, or Akt was used as loading control. Quantification of western blot signals was shown in the right panels. The y axis represented the relative activity of c-Src (a), Erk1/2 (b) or Akt (c) in indicated cells compared with that in control cells (786-0/U6) at the same time point. The activity of indicated molecules was normalized to the corresponding total protein level. Data were presented as the mean value from three independent experiments±s.d. *Po0.05, **Po0.01.

The inhibitory effect of Cbp on c-Src induced (for example) by EGF might be well masked by the effect of other players, which might sequester/inhibit Cbp or promote c-Src without activating Cbp-Csk negative feedback loop. This could explain why in our case, under serum stimulation, Cbp knockdown did not affect c-Src activation. Indeed, we did not observe obvious effects of Cbp knockdown on activation of c-Src, FAK, Paxillin (Supplementary Figure 3d), Akt and Erk1/2 (Supplementary Figure 3c) under normal cell culture condition. However, when we looked over time courses of Erk1/2 activation on serum stimulation, Cbp knockdown resulted in a notable acceleration of Erk1/2 inactivation, though it had no effect on the maximum intensity of Erk1/2 activation at 10-min, which might suggest that an unknown signaling mediated by Cbp influenced the sustaining of Erk1/2 activation in RCC cells. This result may partially explain the inhibited cell proliferation of Cbp knockdown cells. Obviously, the change in Erk1/2 activation alone is unlikely to account for the many functions of Cbp in RCC. We found RhoA as another potential downstream effector of Cbp. Previous studies have implicated a role of RhoA in RCC carcinogenesis. For instance, active RhoA was required for the accumulation of

HIF1a, one of the most important regulators for RCC (Klatte et al., 2007), in trophoblast (Hayashi et al., 2005) and RCC cells (Turcotte et al., 2003) under hypoxia condition. By promoting RhoA activation, overexpressed plasma membrane associated sialidase could increase RCC cell motility (Ueno et al., 2006). Here, we found that the PDZ-binding motif of Cbp might be associated with RhoA activation. Cbp is well known to bind to EBP50 via this PDZ-binding motif (Brdickova et al., 2001; Itoh et al., 2002). EBP50 is a cytosolic protein that links membrane proteins with ERM proteins, which directly interact with actin cytoskeletons once activated (Reczek et al., 1997; Bretscher, 1999). Proteins complexed with ERM have been reported to be involved in the regulation of Rho GTPases. For instance, CD44 could associate with the N terminus of Radixin to remove RhoGDI from Rho complexes, therefore, enhancing RhoA activity (Takahashi et al., 1997). Transmembrane protein Type I phosphatidylinositol 4-phosphate 5-kinase isoform-b activated RhoA through interaction with EBP50, which enabled further interactions with ERM and Rho GDI (Lacalle et al., 2007). Based on these data and our findings, we propose that Cbp might form a protein complex with EBP50 to regulate RhoA activation via ERM proteins in RCC Oncogene


3328

Figure 7 Knockdown of C-terminal Src kinase (Csk)-binding protein (Cbp) inhibited RhoA activity. (a) GST-RBD pull-down assay in Cbp stable knockdown cells. Left panel: Western blot showing active RhoA was pulled down by GST-RBD and detected by antiRhoA in western blot. Right panel: Quantification of active RhoA in control cells and Cbp knockdown cells. y axis represented the relative RhoA activity in indicated cells compared to that in control cells. The activity of RhoA was normalized to its total protein level. Data were presented as the mean value from three independent experiments±s.d. (b) GST-RBD pull-down assay in siRNA transiently transfected 786-0 cells. Left panel: Western blot showing active RhoA pulled down at 48 h post-siRNA transfection. Right panel: Quantification of active RhoA in scrambled siRNA or Cbp siRNA transfected cells. y axis represented the relative RhoA activity in indicated cells compared to that in control cells. Relative RhoA activity was calculated as described in (a). *Po0.05.

cells. However, we could not exclude the possibility that other PDZ domain containing proteins might interact with Cbp to regulate RhoA activation. PDZ-RhoGEF and LARG (leukemia associated Rho-specific GEF), two RhoA-specific GEFs that activate RhoA through interacting with lysophosphatidic acid receptors are good candidates because they might bind Cbp via their PDZ domains (Yamada et al., 2005). In addition, Cbp consists of multiple motifs that putatively mediate protein interactions (Brdicka et al., 2000); and it is extensively localized in lipid rafts, where many signaling molecules are enriched (Quest et al., 2004; Pike, 2005). It is also possible that there are yet unidentified signaling molecules that associate with Cbp in lipid rafts, through which Cbp may regulate RhoA activation or participate into yet unknown signaling pathways to contribute to renal cell carcinogenesis. These possibilities should also be investigated. In conclusion, we found that Cbp overexpression was associated with cancerous transformation of RCC cells. The novel function of Cbp in regulating RhoA activation sheds light on the molecular mechanism(s) of how Cbp overexpression is involved in renal cell carcinogenesis. The specific overexpression pattern of Cbp in ccRCC makes Cbp a potential biomarker for RCC diagnosis. Whether Cbp can be used as a potential therapeutic target for kidney carcinoma remains to be further investigated. Oncogene

Materials and methods Cancer specimens and cell lines Specimens of RCC and paired noncancerous tissues were obtained from 41 patients including 27 pairs of protein samples and 14 pairs of mRNA samples. All tissues were snap frozen in liquid nitrogen immediately after resection. Cell lines used in this study are kind gifts from Professor Axel Ullrich (Martinsried, Germany). All culture media were supplemented with 10% fetal bovine serum (FBS; Invitrogen, Carlsbad, CA, USA); 786-0 and 769-P cells were maintained in RPMI 1640 medium (Invitrogen); A498 cells were maintained in MEM medium (Invitrogen) supplemented with nonessential amino acids. Caki-1 cells were maintained in McCoy’s 5A medium (Invitrogen). All cell cultures were incubated at 37 1C in 5% CO2/95% air. Silencing of Cbp by small interfering RNA Two siRNAs targeting position 269–289 (GGGACAUUCUUU CAGAGGACA; named Cbp-Ri-3) and position 1174–1194 (AAGCGAUACAGACUCUCAACA; named Cbp-Ri-5; Shima et al., 2003; Tauzin et al., 2008) of human Cbp mRNA were selected. The scrambled siRNA of Cbp-Ri-3 (3-neg) (CUAUAG GAUGAUCUGGCGAAC) was used as a negative control. The siRNAs were transiently transfected into 786-0 cells using lipofactamine 2000 (Invitrogen) according to the manufacture’s instruction. Assays were performed 48 h after transfection. Cell proliferation assay For MTT assay, indicated cells were plated in 96-well plates at a density of 500 cells per well and were cultured for 3 days for siRNA-transfected cells or 5 days for Cbp stably transfected


3329

Figure 8 C-terminal Src kinase (Csk)-binding protein (Cbp) regulated RhoA activation and cell migration via its PDZ-binding motif. (a) Cbp enhanced RhoA activity via its PDZ domain-binding motif. The indicated plasmids were transiently transfected into 786-0 cells by electroporation. Active RhoA was pulled down by GST-RBD and detected by western blot using anti-RhoA. Vec: pRK5 vector; WT: wild-type rat Cbp; DCP: mutant rat Cbp lacking PDZ-binding motif. Relative RhoA activity was calculated as described in Figure 7 and presented as mean value from three independent experiments±s.d. in the right panel. (b) Wound-healing assay in wildtype and DCP-Cbp stably overexpressed cells. Left panel: Representative images of wound-healing assay. Right upper panel: The expression level of Cbp assessed by western blot. Right lower panel: Statistical analysis of relative wound recovery. The analysis method was the same as in Figure 3a. (c) Migration assay in transwell chambers. Cells that migrated from the upper well of a transwell chamber into the lower well were stained, photographed and counted. Right panel: Statistical analysis of migration assay. Data were presented as the mean value from three different fields±s.d. *Po0.05.

cells. The cells were incubated with 20 ml 3-(4,5)-dimethylthiahiazo (-z-y1)-3,5-di-phenytetrazolium bromide (MTT; 5 mg/ml) for 3 h. The medium was then removed and the cells were incubated with 50 ml dimethyl sulfoxide (DMSO) in the dark for 30 min to dissolve violet crystals. The color intensity was quantified by measuring the absorbance at 570 nm using a microplate reader (Bio-Rad Laboratories Inc., Hercules, CA, USA). DMSO alone was used as control. For Brdu incorporation assay, cells were grown on the glass coverslips to 50% confluence. Brdu was added to a final concentration of 10 mM and cells were cultured for another 1 h.

Cells were then fixed and stained with Brdu antibody or 40 ,6-diamidino-2-phenylindole dihydrochloride (DAPI) using Brdu labeling and detection kit I (Roche, Pensberg, Germany). At least 500 cells were counted per group. Cell motility assay For wound-healing assay, cells were allowed to grow to full confluence. Wounds were made by scraping the monolayer of cells with a 200-ml tip. Cells were then washed with the culture medium three times and incubated for the indicated time before being photographed. Recovered wound area was Oncogene


3330 measured by Image J software (National Institutes of Health, Bethesda, MD, USA) for each indicated cell line and the wound recovery was compared to that of control cells. For transwell migration assay, indicated cells were serum starved for 20 h, then 1.5 104 cells in 100 ml serum-free RPMI-1640 culture medium were seeded into the upper well of the transwell chamber (Coring Incorporated, NY, USA; 8-mm pore size polycarbonate membrane, 6.5 mm diameter). An aliquot of 600 ml of culture medium supplemented with 10% FBS was added into the lower well of the chamber. After 24 h, cells in the upper well were removed with a cotton swab. Cells that migrated into the lower well were washed with phosphatebuffered saline (PBS), fixed in 3.7% paraformaldehyde at room temperature and stained by 0.2% crystal violet in 2% ethanol. Cells were photographed and counted under microscopy. The number of migrated cells was expressed as mean value±standard deviation (s.d.). For in vitro invasion assay, the upper well of the transwell was coated with 50 ml (1 mg/ml) growth factor reduced Matrigel (BD Biosciences, San Jose, CA, USA). The Matrigel was allowed to harden at 37 1C in a 5% CO2 incubator for 1 h before cells were added into the upper well. The rest of the assay was performed as described above (see migration assay). Colony formation in soft agar A portion of 2 ml of 0.5% agar (Sigma, St Louis, MO, USA) in RPMI-1640 medium supplemented with 10% FBS was added into 6-well plates and allowed to harden; 1 104 cells in 2 ml of RPMI-1640 medium supplemented with 10% FBS and 0.25% agar were seeded on the polymerized agar. The cell solution was allowed to set for 30 min at room temperature before being moved into a 37 1C, 5% CO2 incubator. Cell colonies (X50 cells) formed in soft agar were counted and photographed 10 days after inoculation. Ten fields were counted for each cell line. In vivo tumorigenicity assay For each cell line, 6 106 cells were resuspended in prewarmed PBS and inoculated subcutaneously into the left flank of each of the four nude mice. Tumor formation was monitored weekly by measuring the largest and the smallest diameter of the formed tumors, and the volume of the tumors was calculated using the following formula: volume ¼ 1/2 (largest diameter) (smallest diameter)2. The experiment was performed under the Institute’s guidelines for animal experiments. Microscopy and measurement of mean fluorescence intensity Cells were grown to similar confluence, and phase contrast photographs were taken for cell morphology. For staining stress fibers, cells were seeded onto glass coverslips. After 24 h, cells were washed with PBS, fixed with 3.7% paraformaldehyde, permeabilized with 0.1% Triton X-100 and blocked in 1% BSA for 1 h. Alexa fluo 488 phalloidin (Invitrogen) was used to stain F-actin. Nuclei were counterstained with DAPI. Images were photographed with a fluorescent confocal microscope (Leica TCS SP2, Germany).

Different groups of cells was photographed in the exactly same condition and analysed using the Leica TCS SP2 AOBS confocal system. Leica confocal software (LAS AS Lite) was used to quantify the MFI of F-actin in cells. At least 50 cells per group were measured. RhoA, Cdc42 and Rac1 pull-down assay For pull-down assays, cells were lysed in CLB buffer (50 mM Tris, pH 7.4, 10 mM MgCl2, 500 mM NaCl, 2 mM Na3VO4 1% NP-40, 10% Glycerol, 1 mM PMSF). Equal amounts of cell lysates were incubated at 4 1C for 1 h with 40 mg beads coupled GST-RBD recombinant protein (GST-fused to the RhoAbinding domain of Rhotekin) to pull down RhoA or 40 mg bead-coupled GST-PBD recombinant protein (GST fused to the Cdc42 and the Rac1-binding domain of PAK1) to pull down Cdc42 or Rac1. After the incubation, beads were washed three times with CLB buffer. Bound proteins were fractionated by 12% SDS–polyacrylamide gel electrophoresis and immunoblotted with antibodies against RhoA, Cdc42 or Rac1. For each group, active RhoA as a percentage of total RhoA was compared to that of control cells. Statistical analysis Autoradiographs and western blots were scanned using a GS800 Calibrated Densitometer (Bio-Rad Laboratories Inc.) and the band density was quantified using Quantity One software (Bio-Rad Laboratories Inc.). Excel 2003 software (Microsoft Corp., Redmond, WA, USA) was used for statistical analysis. The data was presented as the mean value±s.d. The paired, two-tailed Student’s t-test was used to analyse the significance of difference between groups. P value o0.05 was considered to be statistically significant. For antibodies and reagents, plasmid construction, northern blot, western blot, cell transfection, and stable clones selection, detailed description is given in the Supplementary information.

Conflict of interest The authors declare no conflict of interest. Acknowledgements We thank Dr Vaclav Horejsi (Institute of Molecular Genetics, Academy of Sciences of the Czech Republic, Videnska 1083, 14220 Prague 4, Czech Republic) for kindly and generously giving us the monoclonal antibody against Cbp. This work was supported by grants from National Natural Science Foundation of China (30730055 and 30623002), National Key Scientific Program of China (2007CB914504), the Chinese Academy of Sciences (KSCX1-YW-R-67, KSCX2-YW-R108), Program of Shanghai Subject Chief Scientist (08XD14051) and the National High Technology Research and Development Programme of China (2006AA02A308).

References Baumgartner M, Angelisova P, Setterblad N, Mooney N, Werling D, Horejsi V et al. (2003). Constitutive exclusion of Csk from Hckpositive membrane microdomains permits Src kinase-dependent proliferation of Theileria-transformed B lymphocytes. Blood 101: 1874–1881. Oncogene

Bratslavsky G, Sudarshan S, Neckers L, Linehan WM. (2007). Pseudohypoxic pathways in renal cell carcinoma. Clin Cancer Res 13: 4667–4671. Brdicka T, Pavlistova D, Leo A, Bruyns E, Korinek V, Angelisova P et al. (2000). Phosphoprotein associated with glycosphingolipid-


3331 enriched microdomains (PAG), a novel ubiquitously expressed transmembrane adaptor protein, binds the protein tyrosine kinase csk and is involved in regulation of T cell activation. J Exp Med 191: 1591–1604. Brdickova N, Brdicka T, Andera L, Spicka J, Angelisova P, Milgram SL et al. (2001). Interaction between two adapter proteins, PAG and EBP50: a possible link between membrane rafts and actin cytoskeleton. FEBS Lett 507: 133–136. Bretscher A. (1999). Regulation of cortical structure by the ezrinradixin-moesin protein family. Curr Opin Cell Biol 11: 109–116. Cohen HT, McGovern FJ. (2005). Renal-cell carcinoma. N Engl J Med 353: 2477–2490. D’Angelo R, Aresta S, Blangy A, Del Maestro L, Louvard D, Arpin M. (2007). Interaction of ezrin with the novel guanine nucleotide exchange factor PLEKHG6 promotes RhoG-dependent apical cytoskeleton rearrangements in epithelial cells. Mol Biol Cell 18: 4780–4793. Davidson D, Bakinowski M, Thomas ML, Horejsi V, Veillette A. (2003). Phosphorylation-dependent regulation of T-cell activation by PAG/Cbp, a lipid raft-associated transmembrane adaptor. Mol Cell Biol 23: 2017–2028. Davidson D, Schraven B, Veillette A. (2007). PAG-associated FynT regulates calcium signaling and promotes anergy in T lymphocytes. Mol Cell Biol 27: 1960–1973. D’Hondt V, Gil T, Lalami Y, Piccart M, Awada A. (2005). Will the dark sky over advanced renal cell carcinoma soon become brighter? Eur J Cancer 41: 1246–1253. Hayashi M, Sakata M, Takeda T, Tahara M, Yamamoto T, Minekawa R et al. (2005). Hypoxia up-regulates hypoxia-inducible factor-1alpha expression through RhoA activation in trophoblast cells. J Clin Endocrinol Metab 90: 1712–1719. Ingley E, Schneider JR, Payne CJ, McCarthy DJ, Harder KW, Hibbs ML et al. (2006). Csk-binding protein mediates sequential enzymatic down-regulation and degradation of Lyn in erythropoietin-stimulated cells. J Biol Chem 281: 31920–31929. Itoh K, Sakakibara M, Yamasaki S, Takeuchi A, Arase H, Miyazaki M et al. (2002). Cutting edge: negative regulation of immune synapse formation by anchoring lipid raft to cytoskeleton through Cbp-EBP50-ERM assembly. J Immunol 168: 541–544. Jiang LQ, Feng X, Zhou W, Knyazev PG, Ullrich A, Chen Z. (2006). Csk-binding protein (Cbp) negatively regulates epidermal growth factor-induced cell transformation by controlling Src activation. Oncogene 25: 5495–5506. Kawabuchi M, Satomi Y, Takao T, Shimonishi Y, Nada S, Nagai K et al. (2000). Transmembrane phosphoprotein Cbp regulates the activities of Src-family tyrosine kinases. Nature 404: 999–1003. Klatte T, Seligson DB, Riggs SB, Leppert JT, Berkman MK, Kleid MD et al. (2007). Hypoxia-inducible factor 1 alpha in clear cell renal cell carcinoma. Clin Cancer Res 13: 7388–7393. Lacalle RA, Peregil RM, Albar JP, Merino E, Martinez AC, Merida I et al. (2007). Type I phosphatidylinositol 4-phosphate 5-kinase controls neutrophil polarity and directional movement. J Cell Biol 179: 1539–1553. Lam JS, Leppert JT, Figlin RA, Belldegrun AS. (2005). Role of molecular markers in the diagnosis and therapy of renal cell carcinoma. Urology 66: 1–9. Lindquist JA, Simeoni L, Schraven B. (2003). Transmembrane adapters: attractants for cytoplasmic effectors. Immunol Rev 191: 165–182. Mathivanan S, Ahmed M, Ahn NG, Alexandre H, Amanchy R, Andrews PC et al. (2008). Human Proteinpedia enables sharing of human protein data. Nat Biotechnol 26: 164–167. Matsuoka H, Nada S, Okada M. (2004). Mechanism of Csk-mediated down-regulation of Src family tyrosine kinases in epidermal growth factor signaling. J Biol Chem 279: 5975–5983.

Niggli V, Rossy J. (2008). Ezrin/radixin/moesin: versatile controllers of signaling molecules and of the cortical cytoskeleton. Int J Biochem Cell Biol 40: 344–349. Oneyama C, Hikita T, Enya K, Dobenecker MW, Saito K, Nada S et al. (2008). The lipid raft-anchored adaptor protein cbp controls the oncogenic potential of c-Src. Mol Cell 30: 426–436. Pike LJ. (2005). Growth factor receptors, lipid rafts and caveolae: an evolving story. Biochim Biophys Acta 1746: 260–273. Quest AF, Leyton L, Parraga M. (2004). Caveolins, caveolae, and lipid rafts in cellular transport, signaling, and disease. Biochem Cell Biol 82: 129–144. Rathmell WK, Chen S. (2008). VHL inactivation in renal cell carcinoma: implications for diagnosis, prognosis and treatment. Expert Rev Anticancer Ther 8: 63–73. Reczek D, Berryman M, Bretscher A. (1997). Identification of EBP50: A PDZ-containing phosphoprotein that associates with members of the ezrin-radixin-moesin family. J Cell Biol 139: 169–179. Shima T, Nada S, Okada M. (2003). Transmembrane phosphoprotein Cbp senses cell adhesion signaling mediated by Src family kinase in lipid rafts. Proc Natl Acad Sci USA 100: 14897–14902. Solheim SA, Torgersen KM, Tasken K, Berge T. (2008). Regulation of FynT function by dual domain docking on PAG/Cbp. J Biol Chem 283: 2773–2783. Svec A. (2007). Expression of transmembrane adaptor protein PAG/ Cbp in diffuse large B-cell lymphoma: immunohistochemical study of 73 cases. Pathol Res Pract 203: 193–198. Svec A, Velenska Z, Horejsi V. (2005). Expression pattern of adaptor protein PAG: correlation between secondary lymphatic follicle and histogenetically related malignant lymphomas. Immunol Lett 100: 94–97. Takahashi K, Sasaki T, Mammoto A, Takaishi K, Kameyama T, Tsukita S et al. (1997). Direct interaction of the Rho GDP dissociation inhibitor with ezrin/radixin/moesin initiates the activation of the Rho small G protein. J Biol Chem 272: 23371–23375. Tauzin S, Ding H, Khatib K, Ahmad I, Burdevet D, van EchtenDeckert G et al. (2008). Oncogenic association of the Cbp/PAG adaptor protein with the Lyn tyrosine kinase in human B-NHL rafts. Blood 111: 2310–2320. Tedoldi S, Paterson JC, Hansmann ML, Natkunam Y, Rudiger T, Angelisova P et al. (2006). Transmembrane adaptor molecules: a new category of lymphoid-cell markers. Blood 107: 213–221. Torgersen KM, Vang T, Abrahamsen H, Yaqub S, Horejsi V, Schraven B et al. (2001). Release from tonic inhibition of T cell activation through transient displacement of C-terminal Src kinase (Csk) from lipid rafts. J Biol Chem 276: 29313–29318. Turcotte S, Desrosiers RR, Beliveau R. (2003). HIF-1alpha mRNA and protein upregulation involves Rho GTPase expression during hypoxia in renal cell carcinoma. J Cell Sci 116: 2247–2260. Ueno S, Saito S, Wada T, Yamaguchi K, Satoh M, Arai Y et al. (2006). Plasma membrane-associated sialidase is up-regulated in renal cell carcinoma and promotes interleukin-6-induced apoptosis suppression and cell motility. J Biol Chem 281: 7756–7764. Uhlen M, Bjorling E, Agaton C, Szigyarto CA, Amini B, Andersen E et al. (2005). A human protein atlas for normal and cancer tissues based on antibody proteomics. Mol Cell Proteomics 4: 1920–1932. Yamada T, Ohoka Y, Kogo M, Inagaki S. (2005). Physical and functional interactions of the lysophosphatidic acid receptors with PDZ domain-containing Rho guanine nucleotide exchange factors (RhoGEFs). J Biol Chem 280: 19358–19363. Zhang SQ, Yang W, Kontaridis MI, Bivona TG, Wen G, Araki T et al. (2004). Shp2 regulates SRC family kinase activity and Ras/Erk activation by controlling Csk recruitment. Mol Cell 13: 341–355.

Supplementary Information accompanies the paper on the Oncogene website (http://www.nature.com/onc)

Oncogene