Serine-phosphorylated STAT1 is a prosurvival factor in Wilms' tumor ...

Oncogene (2006) 25, 7555–7564

& 2006 Nature Publishing Group All rights reserved 0950-9232/06 $30.00 www.nature.com/onc

ORIGINAL ARTICLE

Serine-phosphorylated STAT1 is a prosurvival factor in Wilms’ tumor pathogenesis OA Timofeeva1, S Plisov1, AA Evseev1, S Peng1, M Jose-Kampfner1, HN Lovvorn2, JS Dome3, and AO Perantoni1 1

Laboratory of Comparative Carcinogenesis, Center for Cancer Research, National Cancer Institute, Frederick, MD, USA; Department of Pediatric Surgery, Vanderbilt University, Nashville, TN, USA and 3Department of Hematology/Oncology, St Jude Children’s Research Hospital, Memphis, TN, USA 2

Wilms’ tumor (WT), one of the most common pediatric solid cancers, arises in the developing kidney as a result of genetic and epigenetic changes that lead to the abnormal proliferation and differentiation of the metanephric blastema. As activation of signal transducers and activators of transcription (STATs) plays an important role in the maintenance/growth and differentiation of the metanephric blastema, and constitutively activated STATs facilitate neoplastic behaviors of a variety of cancers, we hypothesized that dysregulation of STAT signaling may also contribute to WT pathogenesis. Accordingly, we evaluated STAT phosphorylation patterns in tumors and found that STAT1 was constitutively phosphorylated on serine 727 (S727) in 19 of 21 primary WT samples and two WT cell lines. An inactivating mutation of S727 to alanine reduced colony formation of WT cells in soft agar by more than 80% and induced apoptosis under conditions of growth stress. S727-phosphorylated STAT1 provided apoptotic resistance for WT cells via upregulation of expression of the heat-shock protein (HSP)27 and antiapoptotic protein myeloid cell leukemia (MCL)-1. The kinase responsible for STAT1 S727 phosphorylation in WT cells was identified based upon the use of selective inhibitors as protein kinase CK2, not p38, MAP-kinase kinase (MEK)1/2, phosphatidylinositol 30 -kinase, protein kinase C or Ca/calmodulindependent protein kinase II (CaMKII). The inhibition of CK2 blocked the anchorage-independent growth of WT cells and induced apoptosis under conditions of growth stress. Our findings suggest that serine-phosphorylated STAT1, as a downstream target of protein kinase CK2, plays a critical role in the pathogenesis of WT and possibly other neoplasms with similar STAT1 phosphorylation patterns. Oncogene (2006) 25, 7555–7564. doi:10.1038/sj.onc.1209742; published online 26 June 2006 Keywords: serine 727; STAT1; protein kinase CK2; MCL-1; HSP27; Wilms’ tumor

Correspondence: Dr AO Perantoni, Laboratory of Comparative Carcinogenesis, National Cancer Institute, 1050 Boyless St., Bldg. 538, R. 205E, Frederick, MD 21702, USA. E-mail: [email protected] Received 9 March 2006; revised 4 May 2006; accepted 4 May 2006; published online 26 June 2006

Introduction The signal transducer and activator of transcription (STAT) proteins play a variety of roles in normal physiological cell processes, such as differentiation, proliferation, apoptosis and angiogenesis (Bromberg and Chen, 2001). STAT proteins become transiently phosphorylated at a conserved tyrosine on the Cterminus in response to several growth factors, for example, epidermal growth factor and platelet-derived growth factor, and cytokines, for example, members of the interkeukin-6 and interferon (IFN) family (Ihle, 2001). This facilitates dimerization and translocation to the nucleus, where STATs interact with proteins of the transcription machinery and with DNA to activate gene transcription. Until recently, tyrosine phosphorylation of STATs was believed to be necessary and sufficient for transcriptional activation. However, studies now suggest that phosphorylation at specific serine/threonine residues is also required for maximal transcriptional activation by cytokines and is possibly mediated by extracellular signal-regulated kinases (ERKs) or H7-sensitive kinase (Decker and Kovarik, 2000). In untransformed cells, activation of STAT proteins is transient and under tight control, whereas aberrant constitutive activation of STAT signaling is associated with uncontrolled growth, invasiveness and angiogenesis, and is frequently observed in cancers (Bromberg, 2002). Wilms’ tumor (WT), or nephroblastoma, afflicts one in 10 000 children and arises in undifferentiated metanephric mesenchyme as a result of, as yet, undefined genetic and epigenetic changes that perturb differentiation of the metanephric blastema (Ruteshouser and Huff, 2004), yielding variable percentages of blastemal, epithelial and stromal elements. WT is thought to have a strong genetic component to its etiology – the first WT gene, WT1, a second WT locus, WT2, and two familial WT loci, FWT1 and FWT2, were found to contribute to Wilms’ tumorigenesis (Dome and Coppes, 2002). However, WT1 is mutated in only 1–2% of familial tumors, and can be excluded as the predisposing gene in most WT families. FWT1 and FWT2 loci also lack linkage in some WT families what implies the existence of at least one additional familial WT locus or an epigenetic

STAT1 is a prosurvival factor in Wilms’ tumor OA Timofeeva et al

7556

etiology (Ruteshouser and Huff, 2004). Only 10–15% sporadic tumors, accounting for 98% of all cases, carry mutations in WT1 and/or the b-catenin gene (Gessler et al., 1994; Koesters et al., 1999; Maiti et al., 2000). Thus, the genetics of WT development are complex and still unclear. In this study, we have sought to identify signaling defects that prevent normal differentiation of the metanephric blastema. Leukemia inhibitory factor and the associated transient activation of STAT proteins has been implicated in inducing conversion of metanephric mesenchyme/blastema to podocytes and epithelial tubules of the nephron (Barasch et al., 1999; Plisov et al., 2001). We, therefore, hypothesized that abnormal constitutive activation of STATs in metanephric mesenchyme may lead to the deregulation of the differentiation program, resulting in renal tumorigenesis. Here we report that STAT1 is constitutively phosphorylated on serine 727 (S727), but not on tyrosine 701 in WTs. Furthermore, we demonstrate that S727phosphorylated STAT1 promotes the neoplastic behavior of WT cells. These findings describe a novel mechanism in WT pathogenesis and provide new targets for the treatment of WT.

Results STAT1 is constitutively phosphorylated on S727 in WTs To determine if STAT proteins are constitutively activated in WTs, we examined the phosphorylation status of STAT1, STAT3 and STAT5 in WT cells. WTs with favorable or anaplastic pathology and normal tissue from WT-bearing kidneys lacked constitutive tyrosine phosphorylation of STAT1, STAT3 and STAT5 (Table 1). STAT3 was phosphorylated on S727 only in normal tissue but not in WTs (Table 1). In contrast, STAT1 was constitutively phosphorylated on S727 in 19 of 21 tumors and only in one of eight normal tissues (Table 1, Figure 1). Constitutive S727 phosphorylation of STAT1 was detected in SK-NEP-1 and WiT49 cells, which are both derived from WTs, and in human embryonic kidney cell line HEK293, an adenovirus-transformed line from human metanephric mesenchyme (Figure 1). Interestingly, STAT1 is phosphorylated on S727, but not tyrosine 701, in embryonic but not adult rat kidney (Table 1), suggesting it functions in normal metanephric growth, perhaps

ensuring survival of tissue progenitors. Phosphorylation of STAT1 on S727 in most WT samples suggests that this modification may be important in the pathogenesis of this disease. Loss of STAT1 serine phosphorylation blocks anchorageindependent growth of SK-NEP-1 cells Anchorage-independent growth is a hallmark of cell transformation. To examine the potential role of serinephosphorylated STAT1 on this behavior, SK-NEP-1 tumor cells were stably transfected with enhanced green fluorescent protein (eGFP)-(STAT1-wt) (wild-type), eGFP-(STAT1-S727A), eGFP-(STAT1-Y701F), eGFP(STAT1-S727E) or eGFP and suspended in soft agar. The dominant-negative mutant eGFP-(STAT1-S727A) caused a 60% reduction in the number of colonies as compared to eGFP, eGFP-(STAT1-wt), eGFP-(STAT1Y701F) or eGFP-(STAT1-S727E) mutants (Figure 2b). Comparable results were obtained using transiently transfected cells with pcDNA4.1 vectors expressing STAT1-wt, STAT1-Y701F, STAT1-S727A, STAT1S727E or LacZ. In these experiments, the dominantnegative mutant STAT1-S727A reduced colony numbers by more than 80%, whereas the constitutively active mutant STAT1-S727E increased colony numbers by 20–25% and STAT1-Y701F, STAT1-wt and LacZ had no effect (Figure 2a and b). To further analyze the role of serine-phosphorylated STAT1 on the tumorigenic phenotype of WT cells, we used RNA interference (RNAi) to ‘knockdown’ the level of endogenous STAT1. SK-NEP-1 cells were transfected with a vector, expressing short hairpin RNA (shRNA) against STAT1

Figure 1 STAT1 is constitutively phosphorylated on S727 in WT cells. Western blots were performed with antibodies against serinephosphorylated (pS727) STAT1 using whole-cell extracts from WTs with favorable histology, or normal kidney from WT-bearing patients, or from different cell lines. All blots were re-probed with antibodies against total STAT1.

Table 1 Phosphorylation status of STATs in WTs compared to normal adult and embryonic renal tissue Tissue type/cell line Normal Tumora Rat embryonic kidney SK-NEP-1 HEK293 WiT49Y

STAT1 pTyr701

STAT1 pSer727

STAT3 pTyr705

STAT3 pSer727

STAT5a,b pTyr694

0/8 0/21

1/8 19/21 + + + +

0/8 0/8

4/4 1/8

0/8 0/8 ND

Abbreviations: HEK293, human embryonic kidney cell line; ND, not determined; STAT, signal transducers and activators of transcription; WT, Wilms’ tumors. aWT samples included 17 with favorable histology and four with anaplastic histology. Oncogene


7557

Figure 2 A S727-to-alanine (S727A) mutant form of STAT1 inhibited anchorage-independent growth of SK-NEP-1 cells. Colony formation in soft agar: (a) representative experiment is shown; (b) columns are mean of three independent experiments for SK-NEP-1 cells stably expressing eGFP-tagged STAT1 proteins (light), or transiently expressing STAT1 proteins (dark); bars are s.d. SK-NEP-1 cells transiently transfected with control or shRNA-STAT1 expressing vectors were either suspended in soft agar (c) or subjected to Western blot with antibody against STAT1, pS727 STAT1 and b-ACTIN (d).

or control vector. The level of STAT1 protein was decreased only in cells transfected with STAT1 shRNA. The levels of serine-phosphorylated STAT1 also decreased (Figure 2d). The expression of STAT1 shRNA inhibited anchorage-independent growth of SK-NEP-1 cells by more than 50% compared to control vector (Figure 2c). These results clearly demonstrate that serine-phosphorylated STAT1 facilitates the anchorage-independent growth of WT cells and thus may contribute to their tumorigenic behavior. Serine-phosphorylated STAT1 enhances survival of WT cells under conditions of growth stress To understand the basis for the effect of serinephosphorylated STAT1 on anchorage-independent growth, we evaluated the proliferation and cell death rates in SK-NEP-1 transfected with the various STAT1

mutants. No differences in rates of proliferation or apoptosis were discernible in monolayer cultures. As growth in soft agar represents a form of stress, where cells are growing under conditions of nutritional deprivation and hypoxia, we reasoned that serinephosphorylation may only be relevant under conditions of growth stress. Accordingly, SK-NEP-1 cells transfected with eGFP-(STAT1-wt) and eGFP-(STAT1S727A) were incubated in medium with low serum (0.5% fetal bovine serum (FBS)) to imitate nutritional deprivation and/or deferoxamine (DFO) to simulate hypoxia (Bianchi et al., 1999). For each stress or combination of stress, cells transfected with eGFP(STAT1-S727A) were significantly more sensitive to apoptosis (Figure 3), suggesting that serine phosphorylation provides resistance to apoptosis. These results suggest a prosurvival function for constitutively phosphorylated STAT1 in WT cells and that the reduction in Oncogene


7558

Figure 3 Dominant-negative mutant eGFP-(STAT1-S727A) increases apoptosis of SK-NEP-1 cells under conditions of growth stress. Fluorescence-activated cell sorting analysis of apoptosis using annexin-V staining of control cells, treated with 0.5% FBS, and 0.1 mM DFO is shown (a). (b) Quantitation summary of three experiments. The y axis is the percentage of annexin V-positive cells.

soft agar cloning efficiency caused by eGFP-(STAT1S727A) may be a consequence of increased cell death under conditions of hypoxia and nutritional stress. Serine-phosphorylated STAT1 promotes survival through the regulation of MCL-1 and HSP27 expression To identify possible transcriptional targets of serinephosphorylated STAT1 signaling, wild-type- or mutanttransfected cells were analyzed by microarray under conditions of growth stress. Genes, which were significantly down- or upregulated by eGFP-(STAT1S727A) compared to eGFP-(STAT1-wt), are listed in Supplementary Table 1. Two prosurvival genes, BCL-2 family member myeloid cell leukemia (MCL)-1 and the stress-induced molecular chaperone heat-shock protein (HSP27), were downregulated after treatment with DFO and low serum in cells expressing eGFP(STAT1-S727A). Expression of corresponding proteins was also reduced as confirmed by Western blot analysis (Figure 4a). To evaluate the ability of these genes to promote survival of SK-NEP-1 cells, RNAi specific to MCL-1 or HSP27 or control RNAi for BCL-2, HSP60 or a scrambled RNAi duplex were used. The downregulation of MCL-1 or HSP27 led to apoptosis as measured by poly-(ADP-ribose) polymerase (PARP) cleavage Oncogene

(Figure 4b). Downregulation of HSP60 or BCL-2 showed a minimal, if any, effect on PARP cleavage over the same time period (data not shown). Furthermore, knockdown of STAT1 not only induced apoptosis but also caused decreased expression of MCL-1 and HSP27 (Figure 4c). Finally, primary WTs show elevated expression of MCL-1, although the level of HSP27 was similar in tumors and normal tissues (Figure 5). All these observations suggest that serine-phosphorylated STAT1 upregulates MCL-1 expression in WT cells to enhance their survivability. Protein kinase CK2 mediates STAT1 phosphorylation on S727 in WT cells To identify the serine kinase(s) responsible for S727 phosphorylation of STAT1 and understand the mechanism for its constitutive activation in WTs, we used a panel of specific serine kinase inhibitors. None of the kinases previously implicated in STAT1 serine phosphorylation, such as extracellular signal-regulated kinase ERK1/2, p38, phosphatidylinositol 30 -kinase (PI3K), Ca/calmodulin-dependent protein kinase II (CaMKII) and protein kinase C (PKC), were responsible for phosphorylation in WT cells (data partially shown in Figure 6a). An analysis of the STAT1 amino-


7559

Figure 4 MCL-1 and HSP27 provide apoptotic resistance for WT cells. (a) SK-NEP-1 cells transfected with eGFP-(STAT1-wt) or eGFP-(STAT1-S727A) were incubated with 0.5% FBS, 0.1 mM DFO or both for 20 h, and then subjected to Western blot analysis with antibodies against MCL-1, HSP27 or b-ACTIN. (b) SK-NEP-1 cells were transfected with siRNA against MCL-1 and HSP27 for different times. Western blot analyses were performed with antibodies against HSP27, MCL-1, PARP or b-ACTIN. (c) SK-NEP-1 cells were transfected with vector expressing shRNA-STAT1 for different time points. Western blotting was performed with antibodies against STAT1, HSP27, MCL-1, BCL-2, HSP60, PARP or b-ACTIN.

Figure 5 MCL-1 is overexpressed in WTs. Western blotting was performed with antibodies against MCL-1 and HSP27 on wholecell extracts from WT with favorable histology or normal kidneys from WT-bearing patients.

acid sequence using Generunner software identified a protein kinase CK2 motif at S727. To determine if CK2 can phosphorylate S727 of STAT1, we employed two selective inhibitors, apigenin and 4,5,6,7-tetrabromo-2azabenzamidazole (TBB). Both inhibitors suppressed S727 phosphorylation of STAT1 in SK-NEP-1, HEK293 and WiT49 cells (Figure 6b and c). These results demonstrate that CK2 causes STAT1 serine phosphorylation and identify STAT1 as a new target of CK2 signaling. Protein kinase CK2 is responsible for survival of WT cells under conditions of growth stress If protein kinase CK2 is responsible for the oncogenic activation of STAT1 in WT cells, then apigenin or TBB should inhibit anchorage-independent growth.

Both TBB and apigenin inhibited colony formation and growth of SK-NEP-1 cells in soft agar in a dosedependent manner (Figure 7a). To study the role of CK2 in the survival of WT cells, we treated SK-NEP-1 cells with TBB in the presence of low serum or DFO. MTT assay showed that under conditions of growth stress TBB decreased cell survival (data not shown) and also downregulated MCL-1 (Figure 7b). In all cases, when MCL-1 was decreased, enhanced PARP cleavage was observed. The expression level of HSP27 was downregulated by TBB in low serum and DFO (Figure 7b). These data suggest that CK2 is an essential factor in providing apoptotic resistance for WT cells under conditions of growth stress. To determine if CK2 level is elevated in primary tumors, we evaluated the expression of CK2a and CK2b subunits. Although the CK2a expression levels were similar between normal and tumor samples (data not shown), high levels of CK2b were observed in blastemal tumor cells relative to normal tissue by immunohistohemistry (Figure 8a and b) and by Western blotting (Figure 8c). It has been shown previously that overexpression of CK2b subunit increases activity of the enzyme (Heller-Harrison and Czech, 1991). Moreover, the activity of CK2 and expression of CK2b were found to be upregulated in kidney tumors (Stalter et al., 1994). We propose that protein kinase CK2 activity is dysregulated in Wilms’ tumorigenesis and that its Oncogene


7560

downstream target STAT1 regulates expression of MCL-1 and other prosurvival genes possibly via interaction with other targets of CK2. Discussion The role of STAT1 in tumorigenesis is complex and controversial. On the one hand, it has been shown to

Figure 6 Only protein kinase CK2 inhibitors decrease S727 phosphorylation of STAT1 in WT cells. (a) SK-NEP-1 cells were treated with 10–50 mM inhibitors for MEK1/2, p38 MAPK, PI3K, CaMKII or PKC for 30 min. (b) SK-NEP-1 cells were treated with DMSO, 10 mM apigenin or 10 mM TBB for 30 and 60 min. (c) HEK293 and WiT49 cells were treated with 10 mM apigenin or 10 mM TBB. Western blot analysis was performed using antibodies against pS727 STAT1. Membranes were re-blotted with antibodies against total STAT1.

Figure 7 Protein kinase CK2 promotes apoptotic resistance in WT cells under conditions of growth stress. (a) SK-NEP-1 cells were suspended in soft agar containing different concentrations of protein kinase CK2 inhibitors. The mean7s.d. of three experiments is shown. (b) SK-NEP-1 cells were treated with 50 mM TBB in the presence of 0.5% FBS, or 0.1 mM DFO or both. Western blot analysis was performed with antibody against pS727 STAT1, MCL-1, HSP27 or PARP.

Figure 8 Expression of protein kinase CK2b in WTs. Immunohistochemistry for CK2b in normal kidney (a) and primary tumor (b) showed high levels of expression in blastemal populations in the tumor. By immunoblotting (c), increased expression was observed for all tumors relative to matched normal tissues from the same patients. Oncogene


7561

suppress oncogenesis and metastatic growth (Bowman et al., 2000; Danial and Rothman, 2000; Huang et al., 2002), and STAT1 deficiency has been found in some tumors (Abril et al., 1998; Sun et al., 1998; Pansky et al., 2000). Although Stat1-deficient mice do not display a heightened incidence of spontaneous neoplasms, they are more sensitive to chemical induction of tumors (Kaplan et al., 1998; Levy and Gilliland, 2000). However, STAT1 expression is frequently upregulated in tumors (Frank et al., 1997; Spiekermann et al., 2001; Widschwendter et al., 2002; Benekli et al., 2003; Hudelist et al., 2005) and associated with prosurvival functions under conditions of growth stress, such as X-irradiation (Khodarev et al., 2004), or anticancer therapy (Friedberg et al., 2004; Roberts et al., 2005). Even ectopic expression of IFNg can support tumor development via STAT1 activation (Wang et al., 2000; Lin et al., 2004). Taken together, these data suggest that STAT1 activity is complex and probably context dependent and that STAT1 may indeed function in an oncogenic capacity. In the present study, we found that STAT1 is constitutively phosphorylated on serine in WT cells (Table 1). S727 STAT1 constitutive phosphorylation was previously found in B-CLL, and authors suggested that it might affect cell survival and/or proliferation (Frank et al., 1997). Here we showed that S727-phosphorylated STAT1 promotes anchorage-independent growth of WT cells (Figure 2), probably by increasing cell survival under conditions of growth stress, such as hypoxia and starvation (Figure 3). Potential proapoptotic and significant prosurvival effects of STAT1 have been described previously (Shen et al., 2001; Stephanou and Latchman, 2003; Terui et al., 2004). Both effects were attributed to the ability of STAT1 to activate expression of either proapoptotic (Kumar et al., 1997; Stephanou et al., 2000, 2001) or prosurvival proteins (Stephanou et al., 1999; Madamanchi et al., 2001; Terui et al., 2004). We found here that in WT cells serine-phosphorylated STAT1 promotes cell survival through the regulation of expression of two known prosurvival genes, MCL-1 and HSP27 (Kabakov et al., 2003; Michels et al., 2005). MCL-1, a BCL-2 family protein, promotes survival by preventing the release of cytochrome c from mitochondria (Packham and Stevenson, 2005). It is a short-lived highly regulated protein, induced by a wide range of survival signals and rapidly downregulated during apoptosis. It has been shown that knockdown of MCL-1 in highly metastatic cells increased sensitivity to hypoxia-induced apoptosis and decreased metastatic ability (Koshikawa et al., 2006). MCL-1 was downregulated in WT cells under hypoxic conditions when serine-phosphorylated STAT1 was decreased, and knockdown of MCL-1-induced apoptosis in WT cells. Moreover, MCL-1 is overexpressed in primary WT tumors. These observations suggest that serine-phosphorylated STAT1 can maintain expression of MCL-1 in WTs, thus providing survival preference to tumor cells under stress conditions.

HSP27 was also significantly downregulated under stress conditions by the dominant-negative STAT1 mutant. It has been shown to prevent caspase-dependent cell death triggered by various stress stimuli (Garrido et al., 2003) by sequestering cytochrome c released from the mitochondria (Parcellier et al., 2003). The knockdown of HSP27 induced apoptosis in WT cells. Overexpression of HSP27 has been detected in many different types of tumors, including renal carcinomas (Sarto et al., 2004), but it was not elevated in WTs. Further evaluation is required to determine if posttranslational modifications of HSP27, such as phosphorylation or S-thiolation, which regulate protein function (Sarto et al., 2004), account for the difference. Interestingly, microarray analysis revealed that dominant-negative STAT1 mutant downregulates the expression of genes directly linked to tumorigenesis, like cyclin D2, which is upregulated in 80% of WTs (Rigolet et al., 2001), or thymosin b10, H-ras, cyclin-dependent kinase 10, which promote tumorigenesis in other organs (Santelli et al., 1999; Cox and Der, 2002; Nishiu et al., 2002), and genes that are involved in growth of the embryonic kidney, for example, Cux-1 (Vanden Heuvel et al., 1996). We observed serine-phosphorylated STAT1 in the developing rat kidney, and it may have a role in kidney development, for example, to promote progenitor cell survival before tissue vascularization. Although STAT1-deficient mice or mice expressing constitutively inactive STAT1-S727A mutant do not have a kidney phenotype (Kaplan et al., 1998; Varinou et al., 2003), probably because activated STAT3 can replace STAT1 to drive the transcription of certain genes (Qing and Stark, 2004), it may be more revealing to evaluate the renal response to overexpression of constitutively active STAT1 during metanephric differentiation or carcinogenesis. Previous findings indicated that S727 phosphorylated STAT1 acts as an integrator of signals from multiple pathways, resulting in chromatin remodeling and gene activation (Zhang et al., 1998; Ouchi et al., 2000; Kovarik et al., 2001). Therefore, it is conceivable that serine-phosphorylated STAT1 functions as a transcriptional co-activator interacting with other DNA-bound proteins, and its role in cellular processes can be different depending on other proteins expressed and/or activated under the same conditions in cells. In this case, the role of the protein kinase activating STAT1 on serine becomes critical. Previously different kinases were implicated in S727 STAT1 phosphorylation, including p38 kinase in response to lipopolysaccharide or ultraviolet (UV); p38 kinase, ERKs, c-Jun NH2-terminal kinases (JNKs), PI3K, 3-phosphoinositide-dependent protein kinase I (PDK1) and p90 ribosomal S6 kinase (p90RSK2) in the cellular response to UVB; p38 kinase, JNK, PKCd, CaMKII in response to INFg (Kovarik et al., 1999; Nair et al., 2002; Xu et al., 2005; Zykova et al., 2005); and PKC in B-CCL patients (Frank et al., 1997). In this report, for the first time, we provide evidence that CK2 is involved in STAT1 phosphorylation. Using a specific inhibitor of CK2, TBB, which selectively inhibits this enzyme among a panel of 33 other kinases (Sarno et al., 2001, 2002), we inhibited Oncogene


7562

serine phosphorylation of STAT1 in WT cells concomitantly with anchorage-independent growth (Figures 6b and 7a). We propose that STAT1 in WT functions as part of a CK2 signaling pathway behaving as a prosurvival factor possibly through interactions with other targets of CK2. CK2 is a ubiquitous and highly conserved serine/ threonine kinase essential for cell survival and proliferation (Unger et al., 2004). In normal cells, the level of CK2 is tightly regulated, however, in many different cancers, it is elevated, and our findings suggest its activity is increased in WT as well (Figure 8). If that is the case, the inhibition of CK2 in WTs may offer an opportunity to re-regulate these cells. This seems rather attractive, as disruption of CK2 by treatment of cells with antisense CK2 or selective inhibitors induced apoptosis (Wang et al., 2005; Zien et al., 2005), and, as we show here, selective inhibitors of CK2 also induce apoptosis in WT cells under conditions of growth stress (Figure 7). Considering that CK2 activates Wnt signaling (Seldin et al., 2005) and the SIX proteins (Ford et al., 2000; Yu et al., 2004), which are both important for kidney development and tumorigenesis, the role of CK2 in WT development should receive further attention. In conclusion, our results have demonstrated that constitutively serine-phosphorylated STAT1 can contribute to Wilms’ tumorigenesis through dysregulation of genes important for the proper growth and differentiation of metanephric blastemal cells. We propose that protein kinase CK2 activity in blastemal cells causes this phosphorylation, which in turn activates expression of MCL-1, providing WT cells with a growth/survival advantage. Although further studies are required to determine the precise mechanisms of STAT1 activation by protein kinase CK2 or induction of MCL-1 expression by STAT1, this work identifies CK2 kinase, MCL-1 and serine-phosphorylated STAT1 as important factors and potential drug targets in WT pathogenesis.

Materials and methods Reagents Chemical inhibitors wortmannin, LY2940002, PD98059, U0126, SB203580 and H7 were from Calbiochem (San Diego, CA, USA), apigenin was from Sigma (St Louis, MO, USA) and TBB was a gift from Dr D Shugar (Institute of Biochemistry and Biophysics, Polish Academy of Sciences, Warsaw, Poland). Anti-STAT1 and anti-phospho-serine STAT1 antibodies were from Upstate (Charlottesville, VA, USA); anti-phospho-tyrosine STAT1, anti-phospho-tyrosine STAT3, anti-phospho-serine STAT3, anti-STAT3, anti-phospho-tyrosine STAT5 and anti-STAT5 antibodies were from Cell Signaling Technology (Danvers, MA, USA); anti-MCL-1 and anti-BCL-2 antibodies were from Santa Cruz Biotechnology (Santa Cruz, CA, USA); and anti-HSP27, anti-HSP60, anti-CKIIa, anti-CKIIb and b-ACTIN antibodies from Abcam (Cambridge, MA, USA). Vectors The STAT1 full-size coding region was amplified using total cDNA from SK-NEP-1 cells and cloned into a pCRII-TOPO Oncogene

TA vector (Invitrogen, Carlsbad, CA, USA). Individual STAT1 mutants were generated using a QuikChange mutagenesis kit (Stratagene, La Jolla, CA, USA) and subcloned into pEGFP-C1 (Clontech, Mountain View, CA, USA) and pcDNA4/HisMax (Invitrogen, Carlsbad, CA, USA) vectors. Cell lines and tissue Tumor samples and normal kidneys were obtained from the Children’s Oncology Group (COG) Renal Tumor Biological Samples Bank via the Cooperative Human Tissue Network (CHTN). The SK-NEP-1 and HEK293 cell lines were from American Type Culture Collection (ATCC). The WiT49 cell line was a gift from Dr H Yeger (Hospital for Sick Children, Toronto, Canada). SK-NEP-1 cells were transiently transfected with pcDNA4.1-based vectors expressing LacZ, STAT1-wt, STAT1-S727A, STAT1-S727E or STAT1-Y701F; GenEclipse STAT1 shRNA or GenEclipse control vector was from Chemicon International (Temecula, CA, USA). SKNEP-1 cells stably transfected with vectors expressing eGFP, eGFP-(STAT1-wt), eGFP-(STAT1-Y701F), eGFP-(STAT1S727A) and eGFP-(STAT1-S727E) were sorted by eGFP intensity at 488 nm laser emission using a Becton Dickinson FACS DiVa (BD Biosciences, San Jose, CA, USA) with a 530/28 optical filter. Western blot analysis SK-NEP-1 cells or human tissue samples were homogenized in lysis buffer (20 mM Tris-HCl, pH 7.5, 1 mM ethylenediaminetetraacetic acid, 100 mM NaCl, 1% NP-40, 1% Triton X-100), containing protease inhibitor cocktail (Roche Applied Science, Indianapolis, IN, USA), 1 mM NaVaO4 and 1 mM NaF. Protein extract (50 mg) was loaded into 10% sodium dodecylsulfate–polyacrylamide gel electrophoresis gels. Following electrophoresis, proteins were transferred to polyvinylidene difluoride membranes and blotted with antibodies of interest. Soft agar assay SK-NEP-1 cells, stably or transiently transfected (3 104 or 6 104 per plate, respectively) were resuspended in medium containing 0.3% low-melting agarose and layered on top of a solidified 0.6% agarose bottom layer in 60-mm tissue culture plates in triplicate. In other experiments, SK-NEP-1 cells (3 104 per plate) were plated in soft agar, containing dimethyl sulfoxide (DMSO) or different concentrations of apigenin or TBB. After 2 weeks of incubation, the colonies were stained with P-iodonitrotetrazolium violet, and colonies were counted using ColoniCounting LabWork software (UVP Laboratory Products). Experiments were repeated at least three times. Cell survival assay under serum depletion and hypoxic conditions SK-NEP-1 cell were grown in normal medium to 60–70% confluence, then incubated in 0.5% FBS medium and/or treated with 0.1 mM DFO (Sigma, St Louis, MO, USA) to mimic hypoxia for different times (Bianchi et al., 1999). The percentage of cells undergoing early stages of apoptosis were identified with an annexin V–phycoerythrin-conjugated apoptosis detection kit (BD Bioscience) (Koopman et al., 1994) according to the manufacturer’s protocol, using a Becton Dickinson FACS Calibur system. The results were quantified from three independent experiments using WinMDI software (http://facs.scripps.edu/software.html). RNA preparation and microarray analysis SK-NEP-1 cells transfected with eGFP-(STAT1-wt) or eGFP(STAT1-S727A) were grown in the presence of 0.5% FBS and 0.1 mM DFO for 20 h. Total RNA was extracted using an


7563 RNeasy kit (Qiagen, Valencia, CA, USA). Synthesis of cDNA and labeled cRNA, hybridizations to the human 133A chips and measurement of hybridization intensities were carried out according to manufacturer’s protocols (Affymetrix Inc., San Diego, CA, USA). The hybridization was performed twice for each sample. Data were analyzed using Data Mining Tool software (Affymetrix Inc., San Diego, CA, USA).

Biotechnology Inc. (Santa Cruz, CA, USA), a negative control for siRNA was from Ambion (Austin, TX, USA). SK-NEP-1 cells were transfected with 10 nM of siRNA using 4 ml of TransIT-TKO (Mirus, Madison, WI, USA). Protein lysates were prepared 2, 4 and 24 h after transfection. Protein lysates from cells transfected with STAT1 shRNA or control vector were prepared 24, 48 and 72 h after transfection.

Treatment of cells with inhibitors of serine/threonine kinases SK-NEP-1 cells were incubated for 30 or 60 min with or without specific protein kinase inhibitors, including wortmannin (100 nM), LY2940002 (50 mM), PD98059 (20 mM), U0126 (20 mM), SB203580 (20 mM), KN63 (50 mM), H7 (50 mM), apigenin and TBB (10–50 mM).

Acknowledgements

RNA interference Short interfering RNA (siRNA) duplexes specific for human MCL-1, HSP27, HSP60 and BCL-2 were from Santa Cruz

We thank Dr Nadya Tarasova for critical review of the manuscript, Dr Elizabeth Perlman for providing the tissue slides, and Kathleen Noer and Roberta Matthai from CCR Flow Cytometry Core and Barbara Kasprzak from Pathology Histotechnology Laboratory for excellent assistance. The research was supported by the Intramural Research Program of the NIH, National Cancer Institute, Center for Cancer Research.

References Abril E, Real LM, Serrano A, Jimenez P, Garcia A, Canton J et al. (1998). Cancer Immunol Immunother 47: 113–120. Barasch J, Yang J, Ware CB, Taga T, Yoshida K, ErdjumentBromage H et al. (1999). Cell 99: 377–386. Benekli M, Baer MR, Baumann H, Wetzler M. (2003). Blood 101: 2940–2954. Bianchi L, Tacchini L, Cairo G. (1999). Nucleic Acids Res 27: 4223–4227. Bowman T, Garcia R, Turkson J, Jove R. (2000). Oncogene 19: 2474–2488. Bromberg J. (2002). J Clin Invest 109: 1139–1142. Bromberg J, Chen X. (2001). Methods Enzymol 333: 138–151. Cox AD, Der CJ. (2002). Cancer Biol Ther 1: 599–606. Danial NN, Rothman P. (2000). Oncogene 19: 2523–2531. Decker T, Kovarik P. (2000). Oncogene 19: 2628–2637. Dome JS, Coppes MJ. (2002). Curr Opin Pediatr 14: 5–11. Ford HL, Landesman-Bollag E, Dacwag CS, Stukenberg PT, Pardee AB, Seldin DC. (2000). J Biol Chem 275: 22245–22254. Frank DA, Mahajan S, Ritz J. (1997). J Clin Invest 100: 3140–3148. Friedberg JW, Dong DA, Li S, Kim H, Stephans K, Noonan K et al. (2004). Leukemia Res 28: 139–147. Garrido C, Schmitt E, Cande C, Vahsen N, Parcellier A, Kroemer G. (2003). Cell Cycle 2: 579–584. Gessler M, Konig A, Arden K, Grundy P, Orkin S, Sallan S et al. (1994). Hum Mutat 3: 212–222. Heller-Harrison RA, Czech MP. (1991). J Biol Chem 266: 14435–14439. Huang S, Bucana CD, Van Arsdall M, Fidler IJ. (2002). Oncogene 21: 2504–2512. Hudelist G, Czerwenka K, Singer C, Pischinger K, Kubista E, Manavi M. (2005). Cancer Genet Cytogenet 158: 35–42. Ihle JN. (2001). Curr Opin Cell Biol 13: 211–217. Kabakov AE, Budagova KR, Bryantsev AL, Latchman DS. (2003). Cell Stress Chaperones 8: 335–347. Kaplan DH, Shankaran V, Dighe AS, Stockert E, Aguet M, Old LJ et al. (1998). Proc Natl Acad Sci USA 95: 7556–7561. Khodarev NN, Beckett M, Labay E, Darga T, Roizman B, Weichselbaum RR. (2004). Proc Natl Acad Sci USA 101: 1714–1719. Koesters R, Ridder R, Kopp-Schneider A, Betts D, Adams V, Niggli F et al. (1999). Cancer Res 59: 3880–3882. Koopman G, Reutelingsperger CP, Kuijten GA, Keehnen RM, Pals ST, van Oers MH. (1994). Blood 84: 1415–1420.

Koshikawa N, Maejima C, Miyazaki K, Nakagawara A, Takenaga K. (2006). Oncogene 25: 917–928. Kovarik P, Mangold M, Ramsauer K, Heidari H, Steinborn R, Zotter A et al. (2001). EMBO J 20: 91–100. Kovarik P, Stoiber D, Eyers PA, Menghini R, Neininger A, Gaestel M et al. (1999). Proc Natl Acad Sci USA 96: 13956–13961. Kumar A, Commane M, Flickinger TW, Horvath CM, Stark GR. (1997). Science 278: 1630–1632. Levy DE, Gilliland DG. (2000). Oncogene 19: 2505–2510. Lin W, Kemper A, McCarthy KD, Pytel P, Wang JP, Campbell IL et al. (2004). J Neurosci 24: 10074–10083. Madamanchi NR, Li S, Patterson C, Runge MS. (2001). Arterioscler Thromb Vasc Biol 21: 321–326. Maiti S, Alam R, Amos CI, Huff V. (2000). Cancer Res 60: 6288–6292. Michels J, Johnson PW, Packham G. (2005). Int J Biochem Cell Biol 37: 267–271. Nair JS, DaFonseca CJ, Tjernberg A, Sun W, Darnell Jr JE, Chait BT et al. (2002). Proc Natl Acad Sci USA 99: 5971–5976. Nishiu M, Yanagawa R, Nakatsuka S, Yao M, Tsunoda T, Nakamura Y et al. (2002). Jpn J Cancer Res 93: 894–901. Ouchi T, Lee SW, Ouchi M, Aaronson SA, Horvath CM. (2000). Proc Natl Acad Sci USA 97: 5208–5213. Packham G, Stevenson FK. (2005). Immunology 114: 441–449. Pansky A, Hildebrand P, Fasler-Kan E, Baselgia L, Ketterer S, Beglinger C et al. (2000). Int J Cancer 85: 720–725. Parcellier A, Gurbuxani S, Schmitt E, Solary E, Garrido C. (2003). Biochem Biophys Res Commun 304: 505–512. Plisov SY, Yoshino K, Dove LF, Higinbotham KG, Rubin JS, Perantoni AO. (2001). Development 128: 1045–1057. Qing Y, Stark GR. (2004). J Biol Chem 279: 41679–41685. Rigolet M, Faussillon M, Baudry D, Junien C, Jeanpierre C. (2001). Pediatr Nephrol 16: 1113–1121. Roberts D, Schick J, Conway S, Biade S, Laub PB, Stevenson JP et al. (2005). Br J Cancer 92: 1149–1158. Ruteshouser EC, Huff V. (2004). Am J Med Genet C 129: 29–34. Santelli G, Califano D, Chiappetta G, Vento MT, Bartoli PC, Zullo F et al. (1999). Am J Pathol 155: 799–804. Sarno S, Moro S, Meggio F, Zagotto G, Dal Ben D, Ghisellini P et al. (2002). Pharmacol Ther 93: 159–168. Sarno S, Reddy H, Meggio F, Ruzzene M, Davies SP, Donella-Deana A et al. (2001). FEBS Lett 496: 44–48. Oncogene


7564 Sarto C, Valsecchi C, Magni F, Tremolada L, Arizzi C, Cordani N et al. (2004). Proteomics 4: 2252–2260. Seldin DC, Landesman-Bollag E, Farago M, Currier N, Lou D, Dominguez I. (2005). Mol Cell Biochem 274: 63–67. Shen Y, Devgan G, Darnell Jr JE, Bromberg JF. (2001). Proc Natl Acad Sci USA 98: 1543–1548. Spiekermann K, Biethahn S, Wilde S, Hiddemann W, Alves F. (2001). Eur J Haematol 67: 63–71. Stalter G, Siemer S, Becht E, Ziegler M, Remberger K, Issinger OG. (1994). Biochem Biophys Res Commun 202: 141–147. Stephanou A, Brar BK, Scarabelli TM, Jonassen AK, Yellon DM, Marber MS et al. (2000). J Biol Chem 275: 10002–10008. Stephanou A, Isenberg DA, Nakajima K, Latchman DS. (1999). J Biol Chem 274: 1723–1728. Stephanou A, Latchman DS. (2003). Int J Exp Pathol 84: 239–244. Stephanou A, Scarabelli TM, Brar BK, Nakanishi Y, Matsumura M, Knight RA et al. (2001). J Biol Chem 276: 28340–28347. Sun WH, Pabon C, Alsayed Y, Huang PP, Jandeska S, Uddin S et al. (1998). Blood 91: 570–576. Terui K, Haga S, Enosawa S, Ohnuma N, Ozaki M. (2004). Biochem J 380: 203–209.

Unger GM, Davis AT, Slaton JW, Ahmed K. (2004). Curr Cancer Drug Targets 4: 77–84. Vanden Heuvel GB, Bodmer R, McConnell KR, Nagami GT, Igarashi P. (1996). Kidney Int 50: 453–461. Varinou L, Ramsauer K, Karaghiosoff M, Kolbe T, Pfeffer K, Muller M et al. (2003). Immunity 19: 793–802. Wang D, Moriggl R, Stravopodis D, Carpino N, Marine JC, Teglund S et al. (2000). EMBO J 19: 392–399. Wang G, Unger G, Ahmad KA, Slaton JW, Ahmed K. (2005). Mol Cell Biochem 274: 77–84. Widschwendter A, Tonko-Geymayer S, Welte T, Daxenbichler G, Marth C, Doppler W. (2002). Clin Cancer Res 8: 3065–3074. Xu W, Nair JS, Malhotra A, Zhang JJ. (2005). J Interferon Cytokine Res 25: 113–124. Yu J, McMahon AP, Valerius MT. (2004). Curr Opin Genet Dev 14: 550–557. Zhang JJ, Zhao Y, Chait BT, Lathem WW, Ritzi M, Knippers R et al. (1998). EMBO J 17: 6963–6971. Zien P, Duncan JS, Skierski J, Bretner M, Litchfield DW, Shugar D. (2005). Biochim Biophys Acta 1754: 271–280. Zykova TA, Zhang Y, Zhu F, Bode AM, Dong Z. (2005). Carcinogenesis 26: 331–342.

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

Oncogene