Molecular Basis for the Cell Type Specific Induction of SnoN Expression by Hepatocyte Growth Factor

Transcription

1 BASIC RESEARCH Molecular Basis for the Cell Type Specific Induction of SnoN Expression by Hepatocyte Growth Factor Ruoyun Tan,* Xianghong Zhang,* Junwei Yang, Yingjian Li,* and Youhua Liu* *Department of Pathology, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania; and Department of Medicine, The First Affiliated Hospital, Nanjing Medical University, Nanjing, China ABSTRACT Hepatocyte growth factor (HGF) is a potent antifibrotic cytokine that antagonizes the TGF- 1/Smad signaling in diverse types of kidney cells by different mechanisms. HGF is shown to induce Smad co-repressor Sloan-Kettering Institute proto-oncogene related novel gene, non Alu-containing (SnoN) expression in proximal tubular epithelial cells (HKC-8) but not in glomerular mesangial cells and interstitial fibroblasts. This study investigated the molecular mechanisms underlying the cell type specific induction of SnoN by HGF. Treatment of HKC-8 cells with actinomycin D completely abolished HGFmediated SnoN induction, suggesting its dependence on gene transcription. Although HGF activated several signal pathways in HKC-8 cells, blockade of extracellular signal regulated kinase-1 and -2 (Erk-1/2) activation but not Akt and p38 mitogen-activated protein kinase abolished SnoN induction. HGF rapidly activated both upstream and downstream signaling of Erk-1/2, which led to the activation of the camp response element binding protein (CREB). In the promoter region of human SnoN gene, two camp response elements were located in close proximity to Sp1 sites. Chromatin immunoprecipitation assay revealed that activated CREB and Sp1 bound to their cognate cis-acting elements in SnoN promoter in response to HGF stimulation. Ectopic expression of wild-type CREB promoted SnoN expression, whereas dominant negative mutant CREB abrogated SnoN induction by HGF. Likewise, chemical blockade of Sp1 binding abolished HGF-mediated SnoN induction. Furthermore, HGF selectively induced CREB phosphorylation in HKC-8 cells but not in mesangial cells and fibroblasts. In vivo, administration of HGF gene induced renal Erk-1/2 phosphorylation, CREB activation, and SnoN expression in obstructive nephropathy. Collectively, these results suggest that CREB activation, in concert with Sp1, constitutes a molecular switch that confers the cell type specific induction of SnoN in response to HGF stimulation. J Am Soc Nephrol 18: , doi: /ASN Hepatocyte growth factor (HGF) is a unique endogenous protein that possesses potent antifibrotic potential. 1,2 Extensive studies in the past several years have demonstrated that administration of exogenous HGF or its gene effectively ameliorates kidney dysfunction and fibrotic lesions in a wide variety of animal models of chronic kidney diseases. 3 9 Conversely, blockade of HGF signaling with neutralizing antibody markedly exacerbates the kidney function and accelerates the progression of renal fibrosis in different chronic kidney disease models. 10,11 Experimental evidence indicates that HGF is able to antagonize directly the profibrotic TGF- 1/Smad signaling in diverse types of kidney cells by different mechanisms. 1 In kidney tubular epithelial cells, HGF has been shown to block the epithelial-to-mesenchymal transition induced by TGF ,13 This action of HGF is primarily mediated by its ability to upregulate Sloan-Kettering In- Received January 30, Accepted May 2, Published online ahead of print. Publication date available at Correspondence: Dr. Youhua Liu, Department of Pathology, University of Pittsburgh, S-405 Biomedical Science Tower, 200 Lothrop Street, Pittsburgh, PA Phone: ; Fax: ; liuy@upmc.edu Copyright 2007 by the American Society of Nephrology 2340 ISSN : / J Am Soc Nephrol 18: , 2007

2 BASIC RESEARCH stitute proto-oncogene (Ski)-related novel gene, non Alucontaining (SnoN). 14 Consistently, HGF induces SnoN expression in vivo, and ectopic expression of SnoN impedes TGF- 1 action in tubular epithelial cells. 14 SnoN belongs to a family of proteins that are known as Smad transcriptional co-repressors, which include Ski, SnoN, and TG-interacting factor Through various mechanisms, SnoN can efficiently block TGF- 1/Smad signaling. Copious studies have shown that SnoN can interact physically with Smad proteins in the nuclei and sequester the ability of Smads to transactivate gene expression. 14,15 Furthermore, SnoN is instrumental in recruiting other transcriptional co-repressors, such as the nuclear hormone receptor co-repressor and mammalian Sin3 ortholog, and prevents Smads from binding to the transcriptional co-activator p300/cbp. 15,16 In normal kidney, SnoN is highly expressed, 18 which is presumably necessary for keeping the TGF- 1/Smad signaling tightly constrained. However, the abundance of SnoN protein is progressively reduced in the fibrotic kidney, primarily through an enhanced ubiquitindependent proteasomal degradation. 19,20 Therefore, TGF- 1/Smad signaling in the pathologic state is dramatically amplified by virtue of losing a crucial negative control mechanism. In this context, restoration of SnoN expression by HGF in the fibrotic kidney could be a primary pathway leading to inhibition of the profibrotic TGF- 1/Smad signaling. SnoN induction by HGF seems to occur in a cell type specific manner. In glomerular mesangial cells and interstitial fibroblasts, HGF is unable to induce SnoN gene expression, although it competently blocks TGF- 1/Smad signaling via other mechanisms. 21,22 These observations indicate that HGF seems to elicit its antifibrotic effects in different kidney cells through distinctive modes of action. However, the molecular mechanism that governs the cell type specific induction of SnoN by HGF in tubular epithelial cells remains completely unknown. In this study, we show that induction of SnoN by HGF is operated at the gene transcriptional level through an extracellular signal regulated kinase-1 and -2 (Erk-1/2)-dependent pathway. HGF seems to trigger a cascade of signal transduction, which leads to the activation of the camp response element (CRE)-binding protein (CREB), a transcription factor that is selectively activated in tubular epithelial cells. Our data suggest that CREB activation, together with Sp1, plays an essential role in mediating the cell type specific induction of SnoN. RESULTS SnoN Induction by HGF Occurs at the Gene Transcriptional Level Figure 1, A and B, shows that HGF specifically induced SnoN mrna expression in human proximal tubular epithelial cells (HKC-8), consistent with a previous report. 14 To understand the mechanism that governs SnoN induction by HGF in tubular epithelial cells, we first determined whether SnoN induction occurs at the gene transcriptional or posttranscriptional level. To this end, HKC-8 cells were incubated with HGF in the absence or presence of actinomycin D, an inhibitor of gene transcription. As shown in Figure 1, C and D, in the absence of gene transcription after treatment with actinomycin D, HGF failed to induce SnoN mrna and protein expression. In contrast, HGF was able to induce SnoN expression in HKC-8 cells when the gene transcription was intact. Therefore, it seems clear that SnoN induction by HGF occurs at the gene transcriptional level. SnoN Induction Depends on the Activation of Erk-1/2 Signaling We next sought to determine which signal pathway triggered by HGF is responsible for mediating SnoN induction. As shown in Figure 2, HGF activated multiple signal pathways in tubular epithelial cells. When HKC-8 cells were treated with HGF for various periods of time, Erk-1/2 mitogen-activated protein kinase (MAPK) was rapidly phosphorylated and activated, as demonstrated by phospho-specific antibody (Figure 2A). Similarly, HGF rapidly induced Akt kinase (Figure 2B) and p38 MAPK phosphorylation (Figure 2C) in HKC-8 cells. We found that blockade of Erk-1/2 activation with PD98059 abolished SnoN induction by HGF (Figure 2E). However, blockade of Akt kinase and p38 MAPK activation had no or Figure 1. Induction of Sloan-Kettering Institute proto-oncogene related novel gene, non Alu-containing (SnoN) expression by hepatocyte growth factor (HGF) occurs at the gene transcriptional level. (A and B) Semiquantitative reverse transcriptase PCR (RT-PCR) analyses demonstrated SnoN mrna induction by HGF. Human proximal tubular epithelial (HKC-8) cells were treated with 20 ng/ml HGF for various periods of time as indicated. (A) Representative RT-PCR result. (B) Graphic presentation of the relative abundance of SnoN mrna at various periods of time as indicated after normalization with -actin. Data are means SEM of three experiments. *P 0.05 versus controls; **P versus controls. (C and D) RT-PCR and Western blot analyses demonstrated that the induction of SnoN expression by HGF occurred at the gene transcriptional level. HKC-8 cells were incubated in the absence or presence of actinomycin D (5 g/ml) for 1 h, followed by treatment with or without HGF (20 ng/ml) for 3 h (C) or 24 h (D). Total cellular RNA was subjected to RT-PCR (C), whereas whole-cell lysates were immunoblotted with antibodies against SnoN and actin (D). J Am Soc Nephrol 18: , 2007 HGF Upregulates SnoN Expression 2341

3 BASIC RESEARCH Figure 2. Induction of SnoN by HGF depends on the extracellular signal regulated kinase-1 and -2 (Erk-1/2) signaling. (A through C) HGF triggered various signaling pathways in tubular epithelial cells. HKC-8 cells were treated with 20 ng/ml HGF for various periods of time as indicated. Whole-cell lysates were immunoblotted with antibodies against either phospho-specific or total Erk-1/2 (A), Akt (B), and p38 mitogen-activated protein kinase (MAPK; C). (D and E) Blockade of Erk-1/2 activation abolished SnoN induction by HGF. HKC-8 cells were pretreated with either various chemical inhibitors or vehicle (DMSO) for 30 min, followed by incubation in the absence or presence of 20 ng/ml HGF for 1 (D) and 24 h (E). Special inhibitors for phosphatidylinositol 3-kinase (10 nm wortmannin), Mek1 (10 M PD98059), and p38 MAPK (20 M SC68376) were used. Whole-cell lysates were immunoblotted with antibodies against phospho-specific or total Erk-1/2, SnoN, and actin. little effect on SnoN induction. Of note, these chemical inhibitors at the concentrations used were able to block Akt and p38 MAPK activation in HKC-8 cells, as previously reported. 23,24 These results suggest that HGF induces SnoN expression primarily by an Erk-1/2 MAPK-dependent pathway in tubular epithelial cells. Consistent with the data on SnoN induction, we found that Erk-1/2 activation was also essential for HGF to elicit its anti TGF- 1 action. As shown in Figure 3, HGF abolished TGF- 1 induced E-cadherin suppression and -smooth muscle actin induction in HKC-8 cells; however, blockade of Erk-1/2 activation with PD98059 completely restored the profibrotic TGF- 1 action in tubular epithelial cells. Under the same conditions, blockade of Akt or p38 MAPK signaling had no or little effect on the ability of HGF to antagonize TGF- 1 action. HGF Activates CREB and Induces Its Binding to SnoN Promoter Figure 4 shows that both upstream and downstream signaling mediators of Erk-1/2 were activated by HGF in tubular epithelial cells. HGF was able to induce a rapid phosphorylation of Raf and Mek1, the upstream kinase of Erk-1/2 in HKC-8 cells. Similarly, the downstream effector kinase of Erk-1/2, p90rsk, was quickly activated. This led to the phosphorylation and activation of CREB, a transcription factor that is activated by Erk-1/2 p90rsk signaling. 25 As demonstrated in Figure 4, phospho-specific CREB but not total cellular CREB was dramatically increased shortly after HGF stimulation in HKC-8 cells. The band below phospho-creb was likely the phosphorylated activating transcription factor-1, a CREB-related protein that is also recognized by this antibody. To establish the relevance of CREB activation to SnoN induction, we analyzed the promoter region of human SnoN gene. As shown in Figure 5A, there were two putative camp response elements (CRE) in the SnoN promoter region. Chromatin immunoprecipitation (ChIP) assay revealed that HGF induced phosphorylated CREB binding to that region in the SnoN promoter in a time-dependent manner (Figure 5B). Quantitative analysis showed that HGF increased the interaction between phospho-creb and SnoN promoter by approximately 14-fold at 30 min after stimulation (Figure 5C). CREB Activation Is Necessary for Mediating SnoN Induction To examine the importance of CREB activation in SnoN expression, we sought to investigate whether overexpression of CREB mediates SnoN induction in tubular epithelial cells. To this end, HKC-8 cells were transiently transfected with wildtype CREB expression vector (pcmv-wt-creb), followed by incubation with HGF for 24 h. As shown in Figure 6A, ectopic expression of CREB induced SnoN expression in HKC-8 cells in the absence of HGF (Figure 6A, lane 1 versus lane 3). SnoN expression was further increased in the CREB-transfected cells after HGF stimulation (Figure 6A, lane 3 versus lane 4). Figure 6B shows that CREB activation was also required for HGF-mediated SnoN induction in tubular epithelial cells. HKC-8 cells were transfected with dominant negative mutant CREB expression vector (pcmv-dn-creb) that expresses a variant of the human CREB protein that contains a serine-toalanine mutation, which blocks phosphorylation of CREB, thereby preventing transcription in a dominant negative manner. We found that overexpression of mutant CREB completely abolished SnoN induction by HGF in HKC-8 cells (Fig Journal of the American Society of Nephrology J Am Soc Nephrol 18: , 2007

4 BASIC RESEARCH Figure 4. HGF activates both upstream and downstream signaling mediators of Erk-1/2 in tubular epithelial cells. HKC-8 cells were incubated with 20 ng/ml HGF for various periods of time as indicated, and the cell lysates were immunoblotted with antibodies against phosphorylated Raf, phosphorylated Mek1, phosphorylated ant total Erk-1/2, phosphorylated p90rsk, phosphorylated and total camp response element (CRE)-binding protein (CREB), and -tubulin. The band below phospho-creb was likely the phospho activating transcription factor-1 (ATF-1), a CREB-related protein that is also recognized by this antibody. Figure 3. Blockade of Erk-1/2 activation abrogates the HGFmediated inhibition of TGF- 1 action in tubular epithelial cells. (A) HKC-8 cells were pretreated with either various chemical inhibitors or vehicle (DMSO) for 30 min, followed by incubation without or with 2 ng/ml TGF- 1 and/or 20 ng/ml HGF for 48 h as indicated. Whole-cell lysates were immunoblotted with antibodies against E-cadherin, -smooth muscle actin ( -SMA), and actin. (B through I) Immunofluorescence staining for E-cadherin and -SMA in HKC-8 cells after various treatments as indicated. T, TGF- 1; H, HGF; PD, PD ure 6B), suggesting that CREB activation is essential for SnoN induction in tubular epithelial cells. Similar results were obtained when using immunofluorescence staining for SnoN (Figure 6C). HGF Selectively Activates CREB in Tubular Epithelial Cells Because HGF induces SnoN expression in a cell type specific manner, we next determined whether there is a variation in CREB activation by HGF in different types of kidney cells. Therefore, HKC-8 cells, human mesangial cells (HMC), and rat kidney interstitial fibroblast (NRK-49F) cells were treated with HGF for various periods of time, and CREB activation was assessed. As shown in Figure 7, HGF induced a dramatic phosphorylation and activation of CREB in HKC-8 cells but not in HMC and NRK-49F cells. However, Erk-1/2 phosphorylation was detected in all three cell types after HGF stimulation (Figure 7). These data suggest that there is a strong correlation between CREB activation and SnoN induction in different kidney cells. Sp1 Is Indispensable for Mediating the Cell Type Specific Induction of SnoN by HGF Sequence analysis also showed that there were two Sp1 binding sites in the proximity of CRE in the SnoN promoter (Figure 5A). This prompted us to examine whether Sp1 transcription factor participates in SnoN regulation. ChIP assay demonstrated that Sp1 could constitutively bind to its cognate sites in the SnoN promoter region (Figure 8A). Furthermore, HGF substantially augmented the interaction between Sp1 and SnoN promoter in a time-dependent manner (Figure 8, A and B). To test the functionality of this Sp1 binding, we examined the effect of chemical blockade of Sp1 binding on SnoN expression. Therefore, HKC-8 cells were pretreated with mithramycin A, a potent inhibitor of Sp1 binding, followed by incubation with HGF for 24 h. As shown in Figure 8, C and D, blockade of Sp1 binding suppressed SnoN expression at basal condition and completely abolished SnoN induction by HGF. J Am Soc Nephrol 18: , 2007 HGF Upregulates SnoN Expression 2343

5 BASIC RESEARCH Figure 5. HGF induces the binding of CREB to its cognate element in the SnoN gene promoter. (A) Partial sequence of human SnoN gene promoter region. Highlighted are two putative CRE and two Sp1 binding sites in this region. (B and C) Chromatin immunoprecipitation (ChIP) assay revealed that HGF induced phosphorylated CREB binding to SnoN promoter. HKC-8 cells were treated with 20 ng/ml HGF for various periods of time as indicated and subjected to ChIP assay. Shown are a representative ChIP assay (B) and the quantitative data of three independent experiments (C). * P Because Sp1 is predominantly expressed in epithelial cells in the kidney, 29 it seems that Sp1 also plays a critical role in mediating the cell type specific induction of SnoN by HGF. HGF Induces Erk-1/2 and CREB Activation In Vivo Earlier studies showed that HGF can restore SnoN expression that was repressed in the obstructed kidney in vivo. 14 To explore the mechanism by which HGF induces SnoN expression in vivo, we examined the Erk-1/2 and CREB activation in the obstructed kidney after delivery of HGF gene. As shown in Figure 9, A and B, administration of HGF gene induced Erk- 1/2 phosphorylation in the obstructed kidney after unilateral ureteral obstruction (UUO). The relative abundance of phosphorylated Erk-1/2 in the obstructed kidney after HGF administration was increased by approximately 12-fold over the controls. Similarly, phospho-creb was markedly induced by HGF and primarily localized in the nuclei of tubular epithelia (Figure 9C), a pattern that closely resembles SnoN expression in this model. 14 To confirm further the relevance of CREB activation to SnoN induction, we used a double immunostaining approach to examine the co-localization of phosphorylated CREB and SnoN. As shown in Figure 9D, co-localization of phosphorylated CREB and SnoN was clearly observable in the nuclei of the obstructed kidney after administration of HGF gene. Therefore, it seems that HGF also induces SnoN expression in an Erk-1/2 CREB dependent pathway in vivo. DISCUSSION HGF, a potent endogenous antifibrotic cytokine that is capable of antagonizing the profibrotic TGF- 1 signaling, specifically induces SnoN mrna and protein expression in kidney proximal tubular epithelial cells 14 but not in glomerular mesangial cells and interstitial fibroblasts. 21,22 However, the molecular basis for this cell type specific induction was completely unknown. The results presented herein demonstrate that this cell type specific induction of SnoN is mediated primarily by selective activation of CREB transcription factor in the tubular epithelial cells after HGF stimulation. HGF activates Erk-1/2 MAPK signaling that involves a cascade of signal transduction, which leads to phosphorylation and activation of CREB. Phosphorylated CREB then binds to the cis-acting CRE in the promoter region of human SnoN gene and presumably trans-activates SnoN transcription. Consistently, overexpression of CREB induces SnoN expression in tubular epithelial cells, and a dominant negative mutant CREB abrogates SnoN induction by HGF. These results establish that selective CREB activation may play a crucial role in conferring the cell type specific induction of SnoN in tubular epithelial cells. Our studies provide significant insights into understanding the mechanism underlying SnoN transcriptional regulation in kidney cells. CREB is a ubiquitously expressed nuclear transcription factor that contains the basic region/leucine zipper domain. It is originally isolated as a gene regulatory protein that binds to the CRE site of the somatostatin gene promoter region in response to an increase in the cellular camp level. 25,30 CREB forms a homodimer through its leucine zipper domain and binds to DNA through its basic region and is involved in regulating the expression of a wide range of genes that are important to cell proliferation, differentia Journal of the American Society of Nephrology J Am Soc Nephrol 18: , 2007

6 BASIC RESEARCH Figure 6. CREB is necessary for mediating SnoN induction in tubular epithelial cells. (A) Ectopic expression of wild-type CREB induced SnoN expression. HKC-8 cells were transiently transfected with empty vector pcdna3 or wild-type CREB expression vector (pcmv-wt-creb) as indicated, followed by incubation with 20 ng/ml HGF for 24 h. Whole-cell lysates were immunoblotted with antibodies against SnoN and -tubulin. (B) Overexpression of mutant CREB abolished SnoN induction by HGF. HKC-8 cells were transfected with either wild-type (pcmv-wt-creb) or mutant CREB expression vector (pcmv-dn-creb), followed by incubation with 20 ng/ml HGF for 24 h. Mutant CREB abrogated SnoN induction in a dominant negative manner. (C) Immunofluorescence staining for SnoN expression in tubular epithelial cells. HKC-8 cells were transfected with either pcmv-wt-creb or pcmv-dn-creb expression vectors, followed by incubation with or without HGF (20 ng/ml) for 24 h. Cell nuclear staining was carried out with 4,6-diamidino-2-phenylindole, HCl. tion, adaptation, and survival. 25,30,31 Activation of CREB requires phosphorylation of the serine residue at 133, 32 which is necessary and sufficient for transcriptional activation of target genes. This study demonstrates that HGF activates SnoN expression by an Erk-1/2 CREB dependent pathway and suggests that SnoN is a direct target of CREB in kidney cells. Because SnoN is intimately implicated in the regulation of TGF- 1/Smad signaling, this suggests that CREB may be an important transcriptional regulator that is capable of controlling the fibrogenic responses of tubular epithelial cells after chronic injury. Of interest, a recent report indicated that CREB activation also plays a critical role in mediating Erk-1/2 induced tubular epithelial cell survival after oxidant stress. 33 Therefore, it is conceivable that CREB activation may be vital for mediating the beneficial effects of HGF by virtue of its ability to induce SnoN expression and to promote tubular epithelial cell survival. Consistently, attenuation of renal fibrosis and induction of SnoN after administration of HGF gene in vivo are closely correlated with the activation of Erk-1/2 and CREB in obstructive nephropathy (Figure 9). 12,14 CREB could be activated by a wide range of extracellular stimuli through distinct signaling pathways. 25,34 However, it seems that HGF-mediated CREB activation in tubular epithelial cells is mainly dependent on Erk-1/2 signaling, because blockade of Erk-1/2 activation completely abolishes SnoN induction. Of note, TGF- 1 itself does not activate Erk-1/2 signaling in HKC-8 cells. 14 Although p38 MAPK and Akt may potentially stimulate CREB phosphorylation, 25 blockade of p38 MAPK and Akt signaling displays little effect on SnoN induction by HGF (Figure 2). p90rsk, one of the CREB kinases downstream of Erk-1/2, 35,36 is activated and may be directly responsible for CREB activation after HGF stimulation (Figure 4). The importance of CREB activation in mediating SnoN induction by HGF is corroborated by several lines of evidence. First, CREB is rapidly activated in tubular epithelial cells in response to HGF stimulation. Second, SnoN gene promoter harbors two CRE and phosphorylated CREB binds to the CRE-containing region of SnoN promoter. Third, ectopic expression of CREB induces SnoN expression and dominant negative CREB abrogates SnoN induction by HGF. Fourth, CREB is selectively activated in tubular epithelial cells but not in mesangial cells and fibroblasts after HGF stimulation, suggesting an intrinsic coupling of SnoN induction to CREB activation. Finally, SnoN induction by HGF in vivo is associated with Erk-1/2 and CREB activation in obstructive nephropathy. Therefore, it seems plausible that selective CREB activation after HGF stimulation is indispensable for conferring the cell type specific induction of SnoN. Besides CREB, HGF induction of SnoN is completely dependent on proper binding and function of a ubiquitous transcription factor Sp1, as chemical ablation of Sp1 binding abrogates HGF-induced SnoN expression in tubular epithelial cells (Figure 8). Sp1 is the prototype of a family of proteins that consists of four members with distinct expression pattern and diverse functions in different types of cells. 37 The Figure 7. HGF induces CREB phosphorylation and activation in a cell type specific manner. HKC-8 cells, human mesangial cells (HMC), and rat kidney interstitial fibroblasts (NRK-49F) cells were treated with 20 ng/ml HGF for various periods of time as indicated. Whole-cell lysates were immunoblotted with antibodies against phosphorylated CREB, phosphorylated Erk-1/2, and actin. Despite Erk-1/2 activation in all three cell types, significant CREB phosphorylation was found only in HKC-8 cells. J Am Soc Nephrol 18: , 2007 HGF Upregulates SnoN Expression 2345

7 BASIC RESEARCH Figure 8. HGF promotes Sp1 binding to SnoN promoter, and blockade of Sp1 binding abolishes SnoN induction by HGF in tubular epithelial cells. (A and B) ChIP assay revealed that HGF promoted Sp1 binding to its cognate sites in the SnoN promoter. HKC-8 cells were treated with 20 ng/ml HGF for various periods of time as indicated and subjected to ChIP assay. Shown are a representative ChIP assay (A) and the quantitative data of three independent experiments (B). *P (C and D) Blockade of Sp1 binding abolished SnoN induction by HGF. HKC-8 cells were pretreated with 10 7 M mithramycin A for 16 h, followed by incubation with 20 ng/ml HGF for 24 h. (C) Representative Western blot showing SnoN abundance after various treatments. (D) Graphic presentation of relative abundance of SnoN after normalization with glyceraldehyde-3-phosphate dehydrogenase (GAPDH) in different groups. Relative levels (fold induction over control) are means SEM of three experiments. *P versus normal control; P versus HGF group. participates in controlling either suppressive or inducible expression of a single gene under different circumstances, often through interacting with different partners. 26,39 For instance, Sp1 mediates HGF receptor (c-met) suppression after oxidative stress by interacting with Egr-1, leading to its sequestration. 40 Sp1 is also crucial for c-met induction triggered by TGF- 1 through interacting with Smad proteins. 28 The involvement of both Sp1 and CREB in SnoN induction is also highlighted by their binding to the cognate sites in SnoN promoter in a HGF-dependent manner. SnoN is an imperative transcriptional co-repressor that makes the TGF- 1/Smad signaling tightly controlled in normal conditions. Given its eminent role in constraining TGF- 1 signaling, SnoN expression is stringently regulated at both transcriptional and posttranscriptional levels. Earlier studies demonstrated that SnoN protein is progressively reduced in obstructive nephropathy, 18 and such downregulation of SnoN is primarily mediated by an enhanced ubiquitin-dependent degradation. 19,20 However, as shown in this study, SnoN induction by HGF occurs at the transcriptional level and is mediated by transcription factors CREB and Sp1. Although more studies are needed, these observations have set a foundation for better understanding of the mechanism that governs the regulation of SnoN under different circumstances. abundance of Sp1 protein displays tremendous variations in different types of cells, 29 albeit ubiquitously expressed. In kidney, proximal tubular epithelial cells express a high level of Sp1, whereas glomerular mesangial cells and interstitial fibroblasts barely produce this protein. 29 These observations establish a tight correlation between SnoN inducibility by HGF and the levels of Sp1 in various types of kidney cells. Hence, cellular Sp1 level probably is another key molecular determinant for SnoN induction in diverse types of kidney cells in response to HGF stimulation. In support of this, there are numerous Sp1 sites in SnoN promoter, and Sp1 binds to SnoN promoter after HGF stimulation, as demonstrated by ChIP assay. Cooperation between CREB and Sp1 may be operative in mediating SnoN induction in response to HGF stimulation. Sp1 sites are located in close proximity to CRE in SnoN promoter, suggesting a physical availability for Sp1 and CREB to interact. Of note, this type of cooperation between Sp1 and CREB for gene regulation is not without precedent. In the promoters of the Na-K-ATPase 1-subunit, a synergy between Sp1 and CREB for gene activation is documented. 38 Hence, cooperation between Sp1 and CREB may represent a general regulatory mechanism for conferring induction of many genes. It should be stressed that although Sp1 is a ubiquitous protein that regulates the constitutive expression of many genes, it also CONCISE METHODS Cell Culture and Treatment Human proximal tubular epithelial cells (HKC-8) were cultured in DMEM and Ham s F12 medium (1:1) supplemented with 5% FBS (Invitrogen, Carlsbad, CA), as described previously. 12,41 HMC were purchased from ScienCell Research Laboratories (San Diego, CA). 42 Normal rat kidney interstitial fibroblast cells (NRK-49F) were obtained from American Type Culture Collection (Manassas, VA). HMC and NRK-49F cells were maintained in DMEM/F12 medium supplemented with 10% FBS. Cells were typically serum-starvedfor16h,followedbyincubationwithhgfforvariousperiodsof time as indicated. In some experiments, cells were pretreated with either various inhibitors at given concentrations or vehicle (0.1% DMSO) 0.5 h before incubationwithdifferentcytokines.humanrecombinanthgfandtgf- 1 were obtained from R&D Systems (Minneapolis, MN). PD98059 (Mek1 inhibitor), wortmannin (phosphatidylinositol 3-kinase inhibitor), and SC (p38 MAPK inhibitor) were purchased from Calbiochem (La Jolla, CA). Actinomycin D and mithramycin A were purchased from Sigma (St. Louis, MO). For chemical blockade of Sp1 binding, HKC-8 cells were pretreated with mithramycin A at 10 7 M for 16 h. 28 Animal Model and HGF Treatment Animal studies were approved by the Institutional Animal Care and Use Committee at the University of Pittsburgh and carried out as 2346 Journal of the American Society of Nephrology J Am Soc Nephrol 18: , 2007

8 BASIC RESEARCH Figure 9. HGF induces Erk-1/2 and CREB phosphorylation and SnoN expression in vivo. (A and B) Administration of HGF gene induced Erk-1/2 phosphorylation in the obstructed kidney. Whole-tissue homogenates from the obstructed kidney of different groups as indicated were immunoblotted with antibodies against phosphorylated and total Erk-1/2. Numbers (1, 2, and 3) indicate each individual animal in a given group. Relative abundance of phosphorylated Erk-1/2 after normalization with total Erk-1/2 in each group is shown in B. *P 0.05 versus the pcdna3 control. (C) Immunohistochemical staining showed that HGF induced CREB phosphorylation in the obstructed kidney. Phosphorylated CREB was predominantly localized in the nuclei of renal tubules. (D) Double immunofluorescence staining for phosphorylated CREB (red) and SnoN (green) in the obstructed kidneys. Co-localization of phospho-creb and SnoN (arrowheads) was evident in the obstructed kidney after HGF administration. Nuclear staining in kidney tissue was carried out with4,6-diamidino-2-phenylindole, HCl. described previously. 43 Briefly, male CD-1 mice that weighed 18 to 22 g were obtained from Harlan Sprague-Dawley (Indianapolis, IN). They were housed in the animal facilities of the University of Pittsburgh Medical Center, with free access to food and water. UUO was performed using an established procedure. 12 On the day of surgery, mice received a single intravenous injection of naked HGF expression plasmid (pcmv-hgf) or empty vector pcdna3 at 1 mg/kg body wt. At day 7 after UUO, mice were killed and the kidneys were removed. One part of the kidneys was frozen immediately in OCT compound for cryosection to perform immunofluorescence studies. The remaining kidneys were snap-frozen in liquid nitrogen and used for Western blot analyses. RNA Isolation and Reverse Transcriptase PCR Total RNA isolation, reverse transcription of the RNA, and PCR amplification were performed as described previously. 44 Briefly, the firststrand cdna synthesis was carried out by using a Reverse Transcription System kit according to the instructions of the manufacturer (Promega, Madison, WI). PCR amplification was performed using HotStar TaqMaster Mix Kit (Qiagen, Valencia, CA). The primer sequences were as follow: SnoN 5 -TTTCTGCCTCTTCCATCACC-3 (sense) and 5 -GACTTGGGGCAAACAGAGTC-3 (antisense) and -actin, 5 -TCAAGATCATTGCTCCTCCTGAGC-3 (sense) and 5 - TGCTGTCACCTTCACCGTTCCAGT-3 (antisense). Relative levels of SnoN mrna (fold induction over the controls) were calculated after normalization with housekeeping gene -actin. Western Blot Analysis Detection of protein expression by Western blot was carried out according to the established protocols described previously. 12 The primary antibodies used were as follows: Anti-SnoN (sc-9595), anti-sp1 (sc-59), and anti-actin (sc-1616) (Santa Cruz Biotechnology, Santa Cruz, CA); anti -smooth muscle actin (clone 1A4) and anti -tubulin (T-9026l Sigma); anti-glyceraldehyde-3-phosphate dehydrogenase (Ambion, Austin, TX); anti E-cadherin (clone 36; BD Biosciences, San Diego, CA); antibodies against phospho-specific (Thr202/Tyr204) and total Erk-1/2, phospho-specific (Ser473) and total Akt, phospho-specific (Thr183/Tyr182) and total p38 MAPK, phospho-specific (Ser133) and total CREB, phospho-specific Raf (Ser338), phospho-specific Mek1/2 (Ser217/221), and phospho-specific p90 ribosomal protein S6 kinase (p90rsk; Ser380; Cell Signaling Technology, Beverly, MA). Quantitative analysis of Western blot data was performed by measurement of the intensity of the band signals with the use of National Institutes of Health Image analysis software. J Am Soc Nephrol 18: , 2007 HGF Upregulates SnoN Expression 2347

9 BASIC RESEARCH Immunofluorescence and Immunohistochemical Staining Indirect immunofluorescence staining was performed using an established procedure. 45 Briefly, HKC-8 cells cultured on coverslips, or kidney cryosections at 5- m thickness were washed twice with cold PBS and fixed with cold methanol:acetone (1:1) for 10 min at 20 C. After three extensive washes with PBS containing 0.5% BSA, the cells were blocked with 20% normal donkey serum in PBS buffer for 30 min at room temperature and then incubated with specific primary antibodies described previously. For visualization of the primary antibodies, slides were stained with cyanine Cy2- or Cy3-conjugated secondary antibodies (Jackson ImmunoResearch Laboratories, West Grove, PA). Cells were double-stained with 4,6-diamidino-2-phenylindole, HCl to visualize the nuclei. As a negative control, the primary antibody was replaced with non-igg, and no staining occurred. For immunohistochemical staining of kidney sections, paraffin-embedded slides were stained with anti phospho-creb antibody using the Vector M.O.M. immunodetection kit by the protocol specified by the manufacturer (Vector Laboratories, Burlingame, CA). Stained slides were mounted with anti-fade mounting medium (Vector Laboratories) and viewed with a Nikon Eclipse E600 Epi-fluorescence microscope (Melville, NY). ChIP Assay ChIP assay was performed to analyze in vivo interactions of transcription factors and their cognate cis-acting elements in SnoN promoter. This assay was essentially carried out according to the protocols specified by the manufacturer (ChIP assay kit; Upstate, Charlottesville, VA). Briefly, after treatment with HGF for various periods of time, HKC-8 cells were cross-linked with 1% formaldehyde for 10 min, washed twice with PBS containing protease inhibitors and scraped, and then resuspended in SDS lysis buffer containing protease inhibitors. The chromatin solution was sonicated. After centrifugation, the supernatant was diluted 10-fold in ChIP dilution buffer and precleared with protein A agarose containing salmon sperm DNA for 1 h. The antibodies against phospho-specific CREB or Sp1 were added and incubated at 4 C overnight, followed by incubation with protein A agarose for 1 h. The precipitates were washed, and chromatin complexes were eluted. After reversal of the cross-linking at 65 C for 4 h, the DNA was purified, and 1 l of input control or ChIP samples were used as a template for PCR using the primer sets for the SnoN promoter regions (from 1838 to 1639) containing two putative CRE and two Sp1 sites. The sequences of primers used for ChIP assay were as follows: Forward 5 -GGTCGGGGCCAGAGG-3 and reverse 5 - CGGACCGGGTCGAGA-3. The relative abundance of the PCR amplified product was calculated after normalization with the intensities of input DNA. Plasmid Transfection For transient transfection, HKC-8 cells were seeded in six-well plates at cells/well. The cells were then transfected with either human wild-type CREB (pcmv-wt-creb) or dominant negative mutant CREB (pcmv-dn-creb) expression vectors (Clontech Laboratories, Mountain View, CA) using Lipofectamine 2000 reagent according to the instructions specified by the manufacturer (Invitrogen). The pcmv-dn- CREB vector expresses a mutant variant of the human CREB protein that contains a serine-to-alanine mutation corresponding to amino acid residue The empty pcdna3 vector was also transfected as a negative control. After transfection for 24 h, the cells were incubated with 20 ng/ml HGF for an additional 24 h and then subjected to Western blot analyses and immunofluorescence staining. Statistical Analyses Statistical analysis was performed using SigmaStat software (Jandel Scientific Software, San Rafael, CA). Comparisons between groups were made using one-way ANOVA, followed by the Student-Newman-Keuls test. P 0.05 was considered significant. ACKNOWLEDGMENTS This work was supported by National Institutes of Health grants DK054922, DK061408, DK064005, and DK DISCLOSURES None. REFERENCES 1. Liu Y: Hepatocyte growth factor in kidney fibrosis: Therapeutic potential and mechanisms of action. Am J Physiol Renal Physiol 287: F7 F16, Matsumoto K, Mizuno S, Nakamura T: Hepatocyte growth factor in renal regeneration, renal disease and potential therapeutics. Curr Opin Nephrol Hypertens 9: , Mizuno S, Nakamura T: Suppressions of chronic glomerular injuries and TGF-beta 1 production by HGF in attenuation of murine diabetic nephropathy. Am J Physiol Renal Physiol 286: F134 F143, Azuma H, Takahara S, Matsumoto K, Ichimaru N, Wang JD, Moriyama T, Waaga AM, Kitamura M, Otsuki Y, Okuyama A, Katsuoka Y, Chandraker A, Sayegh MH, Nakamura T: Hepatocyte growth factor prevents the development of chronic allograft nephropathy in rats. JAm Soc Nephrol 12: , Mizuno S, Kurosawa T, Matsumoto K, Mizuno-Horikawa Y, Okamoto M, Nakamura T: Hepatocyte growth factor prevents renal fibrosis and dysfunction in a mouse model of chronic renal disease. J Clin Invest 101: , Herrero-Fresneda I, Torras J, Franquesa M, Vidal A, Cruzado JM, Lloberas N, Fillat C, Grinyo JM: HGF gene therapy attenuates renal allograft scarring by preventing the profibrotic inflammatory-induced mechanisms. Kidney Int 70: , Dworkin LD, Gong R, Tolbert E, Centracchio J, Yano N, Zanabi AR, Esparza A, Rifai A: Hepatocyte growth factor ameliorates progression of interstitial fibrosis in rats with established renal injury. Kidney Int 65: , Dai C, Yang J, Bastacky S, Xia J, Li Y, Liu Y: Intravenous administration of hepatocyte growth factor gene ameliorates diabetic nephropathy in mice. J Am Soc Nephrol 15: , Cruzado JM, Lloberas N, Torras J, Riera M, Fillat C, Herrero-Fresneda I, Aran JM, Alperovich G, Vidal A, Grinyo JM: Regression of advanced diabetic nephropathy by hepatocyte growth factor gene therapy in rats. Diabetes 53: , Mizuno S, Matsumoto K, Kurosawa T, Mizuno-Horikawa Y, Nakamura T: Reciprocal balance of hepatocyte growth factor and transforming 2348 Journal of the American Society of Nephrology J Am Soc Nephrol 18: , 2007

10 BASIC RESEARCH growth factor-beta1 in renal fibrosis in mice. Kidney Int 57: , Liu Y, Rajur K, Tolbert E, Dworkin LD: Endogenous hepatocyte growth factor ameliorates chronic renal injury by activating matrix degradation pathways. Kidney Int 58: , Yang J, Liu Y: Blockage of tubular epithelial to myofibroblast transition by hepatocyte growth factor prevents renal interstitial fibrosis. JAm Soc Nephrol 13: , Liu Y: Epithelial to mesenchymal transition in renal fibrogenesis: Pathologic significance, molecular mechanism, and therapeutic intervention. J Am Soc Nephrol 15: 1 12, Yang J, Dai C, Liu Y: A novel mechanism by which hepatocyte growth factor blocks tubular epithelial to mesenchymal transition. JAmSoc Nephrol 16: 68 78, Luo K: Ski and SnoN: Negative regulators of TGF-beta signaling. Curr Opin Genet Dev 14: 65 70, Liu X, Sun Y, Weinberg RA, Lodish HF: Ski/Sno and TGF-beta signaling. Cytokine Growth Factor Rev 12: 1 8, Bottinger EP, Bitzer M: TGF-beta1 signaling in renal disease. JAmSoc Nephrol 13: , Yang J, Zhang X, Li Y, Liu Y: Downregulation of Smad transcriptional corepressors SnoN and Ski in the fibrotic kidney: An amplification mechanism for TGF-beta1 signaling. J Am Soc Nephrol 14: , Tan R, Zhang J, Tan X, Zhang X, Yang J, Liu Y: Downregulation of SnoN expression in obstructive nephropathy is mediated by an enhanced ubiquitin-dependent degradation. J Am Soc Nephrol 17: , Fukasawa H, Yamamoto T, Togawa A, Ohashi N, Fujigaki Y, Oda T, Uchida C, Kitagawa K, Hattori T, Suzuki S, Kitagawa M, Hishida A: Ubiquitin-dependent degradation of SnoN and Ski is increased in renal fibrosis induced by obstructive injury. Kidney Int 69: , Yang J, Dai C, Liu Y: Hepatocyte growth factor suppresses renal interstitial myofibroblast activation and intercepts Smad signal transduction. Am J Pathol 163: , Dai C, Liu Y: Hepatocyte growth factor antagonizes the profibrotic action of TGF-beta1 in mesangial cells by stabilizing Smad transcriptional corepressor TGIF. J Am Soc Nephrol 15: , Liu Y: Hepatocyte growth factor promotes renal epithelial cell survival by dual mechanisms. Am J Physiol 277: F624 F633, Dai C, Yang J, Liu Y: Transforming growth factor-beta1 potentiates renal tubular epithelial cell death by a mechanism independent of Smad signaling. J Biol Chem 278: , Ichiki T: Role of camp response element binding protein in cardiovascular remodeling: Good, bad, or both? Arterioscler Thromb Vasc Biol 26: , Poncelet AC, Schnaper HW: Sp1 and Smad proteins cooperate to mediate transforming growth factor-beta 1-induced alpha 2(I) collagen expression in human glomerular mesangial cells. J Biol Chem 276: , Ray R, Snyder RC, Thomas S, Koller CA, Miller DM: Mithramycin blocks protein binding and function of the SV40 early promoter. J Clin Invest 83: , Zhang X, Yang J, Li Y, Liu Y: Both Sp1 and Smad participate in mediating TGF-beta1-induced HGF receptor expression in renal epithelial cells. Am J Physiol Renal Physiol 288: F16 F26, Zhang X, Li Y, Dai C, Yang J, Mundel P, Liu Y: Sp1 and Sp3 transcription factors synergistically regulate HGF receptor gene expression in kidney. Am J Physiol Renal Physiol 284: F82 F94, Mayr B, Montminy M: Transcriptional regulation by the phosphorylation-dependent factor CREB. Nat Rev Mol Cell Biol 2: , Conkright MD, Montminy M: CREB: The unindicted cancer co-conspirator. Trends Cell Biol 15: , Gonzalez GA, Montminy MR: Cyclic AMP stimulates somatostatin gene transcription by phosphorylation of CREB at serine 133. Cell 59: , Arany I, Megyesi JK, Reusch JE, Safirstein RL: CREB mediates ERKinduced survival of mouse renal tubular cells after oxidant stress. Kidney Int 68: , Deak M, Clifton AD, Lucocq LM, Alessi DR: Mitogen- and stressactivated protein kinase-1 (MSK1) is directly activated by MAPK and SAPK2/p38, and may mediate activation of CREB. EMBO J 17: , Bohm M, Moellmann G, Cheng E, Alvarez-Franco M, Wagner S, Sassone-Corsi P, Halaban R: Identification of p90rsk as the probable CREB-Ser133 kinase in human melanocytes. Cell Growth Differ 6: , Pearson G, Robinson F, Beers Gibson T, Xu BE, Karandikar M, Berman K, Cobb MH: Mitogen-activated protein (MAP) kinase pathways: Regulation and physiological functions. Endocr Rev 22: , Suske G: The Sp-family of transcription factors. Gene 238: , Matlhagela K, Taub M: Regulation of the Na-K-ATPase beta(1)-subunit promoter by multiple prostaglandin-responsive elements. Am J Physiol Renal Physiol 291: F635 F646, Jungert K, Buck A, Buchholz M, Wagner M, Adler G, Gress TM, Ellenrieder V: Smad-Sp1 complexes mediate TGFbeta-induced early transcription of oncogenic Smad7 in pancreatic cancer cells. Carcinogenesis 27: , Zhang X, Liu Y: Suppression of HGF receptor gene expression by oxidative stress is mediated through the interplay between Sp1 and Egr-1. Am J Physiol Renal Physiol 284: F1216 F1225, Racusen LC, Monteil C, Sgrignoli A, Lucskay M, Marouillat S, Rhim JG, Morin JP: Cell lines with extended in vitro growth potential from human renal proximal tubule: characterization, response to inducers, and comparison with established cell lines. J Lab Clin Med 129: , Li Y, Wen X, Spataro BC, Hu K, Dai C, Liu Y: Hepatocyte growth factor is a downstream effector that mediates the antifibrotic action of peroxisome proliferator-activated receptor-gamma agonists. JAmSoc Nephrol 17: 54 65, Yang J, Dai C, Liu Y: Hepatocyte growth factor gene therapy and angiotensin II blockade synergistically attenuate renal interstitial fibrosis in mice. J Am Soc Nephrol 13: , Yang J, Shultz RW, Mars WM, Wegner RE, Li Y, Dai C, Nejak K, Liu Y: Disruption of tissue-type plasminogen activator gene in mice reduces renal interstitial fibrosis in obstructive nephropathy. J Clin Invest 110: , Yang J, Liu Y: Dissection of key events in tubular epithelial to myofibroblast transition and its implications in renal interstitial fibrosis. Am J Pathol 159: , 2001 J Am Soc Nephrol 18: , 2007 HGF Upregulates SnoN Expression 2349