Antitumorigenic Effect of Wnt 7a and Fzd 9 in Non-small Cell Lung Cancer Cells Is Mediated through ERK-5-dependent Activation of Peroxisome Proliferator-activated Receptor γ*

The Wnt pathway is critical for normal development, and mutation of specific components is seen in carcinomas of diverse origins. The role of this pathway in lung tumorigenesis has not been clearly established. Recent studies from our laboratory indicate that combined expression of the combination of Wnt 7a and Frizzled 9 (Fzd 9) in Non-small Cell Lung Cancer (NSCLC) cell lines inhibits transformed growth. We have also shown that increased expression of peroxisome proliferator-activated receptor γ (PPARγ) inhibits transformed growth of NSCLC and promotes epithelial differentiation of these cells. The goal of this study was to determine whether the effects of Wnt 7a/Fzd 9 were mediated through PPARγ. We found that Wnt 7a and Fzd 9 expression led to increased PPARγ activity. This effect was not mediated by altered expression of the protein. Wnt 7a and Fzd 9 expression resulted in activation of ERK5, which was required for PPARγ activation in NSCLC. SR 202, a known PPARγ inhibitor, blocked the increase in PPARγ activity and restored anchorage-independent growth in NSCLC expressing Wnt 7a and Fzd 9. SR 202 also reversed the increase in E-cadherin expression mediated by Wnt 7a and Fzd 9. These data suggest that ERK5-dependent activation of PPARγ represents a major effector pathway mediating the anti-tumorigenic effects of Wnt 7a and Fzd 9 in NSCLC.

Wnts are a family of secreted glycoproteins that serve as extracellular signaling molecules controlling diverse morphogenic and developmental programs (1). Signaling is mediated by a family of distinct seven-membrane receptors known as Frizzled (Fzd) 2 (2), and is further regulated by co-receptors LRP 5/6 (3). Aberrant Wnt signaling has been implicated in a variety of cancers (4,5). Our laboratory has focused on the role of Wnt signaling in lung cancer. We have previously reported that the restoration of Wnt 7a and Fzd 9 signaling inhibited both cell proliferation and anchorage-independent growth, promoted cellular differentiation, and reversed the transformed phenotype in Non-small Cell Lung Cancer cells (NSCLC) (6). These findings unveil a novel tumor suppressor pathway in lung cancer and implicate Wnt 7a and Fzd 9 in the maintenance of epithelial cellular differentiation. The downstream effector pathways mediating these effects are not well understood.
Proliferator-activated receptor ␥ (PPAR␥) is a member of the PPAR family of ligand-activated nuclear receptors implicated in a wide variety of biological functions (7). Three PPAR isoforms have been identified, ␣, ␥, and ␤/␦, which all bind as heterodimers with the retinoic acid X receptor to specific regulatory elements in the promoter regions of their target genes. The role of PPAR␥ has been extensively studied in a variety of cancers including colon, breast, prostate, and lung (see Ref. 8 for review). Inactivating mutations in the PPAR␥ gene have been seen in colon cancers (9,10), suggesting that PPAR␥ behaves as a tumor suppressor gene. Pharmacological activators of PPAR␥ inhibit growth of NSCLC cells and induce apoptosis (11,12). In human lung tumors, decreased expression of PPAR␥ was correlated with poor prognosis (13). Our laboratories have demonstrated that overexpression of PPAR␥ in NSCLC inhibited transformed growth and metastasis of NSCLC and promoted epithelial differentiation (14,15).
Because both Wnt 7a/Fzd 9 and PPAR␥ appeared to have similar tumor-suppressive effects on NSCLC, we postulated that they represented components of a common signaling pathway and that the cell surface genes Wnt 7a/Fzd 9 were likely upstream of the nuclear receptor gene PPAR␥. The current study demonstrates that Wnt 7a/Fzd 9 signaling leads to the stimulation of PPAR␥ through activation of Extracellular Signal-Regulated Kinase 5 (ERK-5). Furthermore, pharmacological inhibition of PPAR␥ reversed the anti-tumorigenic effects of Wnt 7a/Fzd 9.
Quantitative PCR-Total RNA was extracted from cultured cells with the RNeasy mini kit (Qiagen Inc., Valencia, CA). Aliquots of the RNA (10 g) were converted to cDNA with Superscript II (Invitrogen) and random hexamers according to the manufacturer's specifications. Primer sets for the quantitative PCR of human PPAR␥ and GAPDH are as follows: PPAR␥ (forward, 5Ј-CAT AAT GCC ATC AGG TTT GG-3Ј; reverse, 5Ј-TCA GCG GAC TCT GGA TTC AG-3Ј). GAPDH (forward, 5Ј-GCC AAA TAT GAT GAC ATC AAG AAG G-3Ј; reverse, 5Ј-GGT GTC GCT GTT GAA GTC AGA G-3Ј. Aliquots (1 l) of reverse transcription reactions were subjected to PCR using the following conditions: 95°C ϫ 10 min, followed by 95°C ϫ 15 s and 60°C ϫ 1 min for 40 cycles in 50-l reactions with SYBR Green Jumpstart Taq Readymix (Sigma). Initial real-time PCR amplification products were resolved by electrophoresis on 5% polyacrylamide gels to verify that the primer pairs amplified a single product of the predicted size. GAPDH mRNA levels were measured by quantitative PCR in the samples as a control gene. The real-time PCR data were analyzed with the Smart Cycler software (version 1.2d) to calculate the threshold cycle values for the different samples and are presented as mRNA levels in arbitrary units.
Transient Transfections and Reporter Luciferase Assays-The reporter plasmids (PPAR-RE, PDRE, and E-cadherin promoter-luciferase), expression plasmids (Wnt 7a, Fzd 9, wild-type PPAR␥, wild-type PPAR␦, and DN-MEK-5), and ␤-galactosidase control plasmids were transfected into cells using Lipofectamine reagent (Invitrogen). The cells were transfected with either 2 g of PPAR-RE, PDRE (gift from Dr. Kenneth Kinzler, Johns Hopkins School of Medicine), dominant negative MEK-5, or E-cadherin promoterluciferase (gift from Dr. Eric Fearon, University of Michigan), and 1 g of pCMV-␤-galactosidase for determination of transfection efficiency. For studies employing pharmacological inhibitors, agents were added 24 h after transfection and cells were harvested 48 h later. Inhibitors used were the MEK inhibitors PD98059 (Sigma) and U0126 (Calbiochem/EMD Biosciences, San Diego, CA), and the PPAR␥ inhibitor SR202 (Ilex Oncology, San Antonio, TX). Cells were collected, washed once with ice-cold phosphate-buffered saline and resuspended in 250 l of Luciferase Reporter Lysis Buffer (Promega). The cell lysates were centrifuged in a microcentrifuge and aliquots (80 l) of the supernatants assayed for luciferase and ␤-galactosidase activity as previously described (6). The data are presented as relative light units/ milliunit of ␤-galactosidase and represent the mean of at least three independent experiments with the S.E. indicated.
Soft Agar Colony Formation-For measurement of anchorage-independent cell growth, 5,000 cells were plated in tripli-FIGURE 1. Transfected Wnt 7a and Fzd 9 stimulate PPAR-RE activity and have a synergistic effect when co-transfected with wild-type PPAR␥. H157 cells transfected with empty vector pcDNA3, Wnt 7a, Fzd 9, Wnt 7a and Fzd 9, wild-type PPAR␥, or both Wnt 7a/Fzd 9 and wild-type PPAR␥ were transiently transfected with PPAR-RE and pCMV-␤-galactosidase as described under "Experimental Procedures." Similarly, H2122 cells were transfected with pcDNA3, Wnt 7a, Fzd 9, wild-type PPAR␥, or both Fzd 9 and wild-type PPAR␥. The cells were incubated for 72 h, and luciferase and ␤-galactosidase activities were measured. The data are presented as relative light units/milliunit of ␤-galactosidase activity and represent the mean of at least three independent experiments with the S.E. indicated. *, p Ͻ 0.05 versus pcDNA-3, Wnt7a alone, or Fzd 9 alone in H157 cells; **, p Ͻ 0.05 versus Wnt7aϩFzd 9 in H157 cells or Fzd 9 in H2122 cells.
cate in 35-mm wells of a six-well plate in a volume of 1.5 ml of growth medium containing 0.3% nobel agar onto a base of 1.5 ml of growth medium containing 0.5% agar. The plates were incubated in a 37°C CO 2 incubator for 21 days. Colonies were stained for 5-12 h at 37°C with nitroblue tetrazolium chloride (1 mg/ml), visualized under a microscope, and counted.

Wnt 7a and Fzd 9 Activate PPAR␥ in NSCLC Cell Lines-H157
and H2122 cells were transiently transfected with empty vector, Wnt 7a, and/or Fzd 9 or wild-type PPAR␥ along with a vector encoding three copies of a PPAR response element (PPAR-RE) linked to firefly luciferase. In H157 cells, which lack both Wnt 7a and Fzd 9 (6), simultaneous expression of both proteins led to a marked increase in PPAR-RE activity ( Fig. 1, left panel). In H2122 cells, which express endogenous Wnt 7a, transient expression of Fzd 9 increased PPAR-RE activity 8-fold (right panel). Increases in PPAR␥ activity were also seen in both cell lines in response to expression of PPAR␥. Co-expression of Wnt 7a and Fzd 9 with PPAR␥ resulted in a synergistic increase in PPAR␥ activity (Fig. 1). By contrast, co-expression of other Wnts (1, 2, 3, 4, 5a, 5b, and 6) and Fzd 9 failed to significantly increase PPAR-RE (data not shown). Similarly, co-expression of Wnt7a with Fzd 1, 2, 3, 4, 6, 7, or 8 failed to increase PPAR-RE activity (data not shown). We also examined activity of PPAR-RE in cells stably expressing Wnt 7a/Fzd 9 or PPAR␥. As shown in Fig. 2A, similar increases were observed in these stable transfectants. Interestingly, stable expression of PPAR␦ failed to increase PPAR-RE activity in either cell line, confirming that this activity is due to PPAR␥. To confirm the specificity of Wnt 7a/Fzd 9 on PPAR isoforms, we measured the activity of a transiently transfected PPAR␦-specific promoter (PDRE) reporter (17). Expression of PPAR␦, but not PPAR␥, increased PDRE promoter activity (Fig. 2B), demonstrating the specificity of these PPAR-RE and PDRE luciferase reporters for PPAR␥ and FIGURE 2. Stably transfected Wnt 7a and Fzd 9, but not wild-type PPAR␦, stimulate PPAR-RE activity. A, H157 cells stably transfected with empty vectors, LNCX and LPCX, Wnt 7a and Fzd 9, wild-type PPAR␥, or wild-type PPAR␦ were transiently transfected with PPAR-RE and pCMV-␤-galactosidase as described under "Experimental Procedures." The cells were incubated for 72 h, and luciferase and ␤-galactosidase activities were measured. The data are presented as relative light units/milliunit of ␤-galactosidase activity. Similarly, H2122 cells were stably transfected with LPCX, Fzd 9, wild-type PPAR␥, or wild-type PPAR␦. B, H157 cells stably transfected with empty vectors LNCX and LPCX, Wnt 7a and Fzd 9, wild-type PPAR␦, or wild-type PPAR␥ were transiently transfected with PDRE and pCMV-␤-galactosidase as described under "Experimental Procedures." Similarly, H2122 cells were stably transfected with LPCX, Fzd 9, wild-type PPAR␦, or wild-type PPAR␥. The cells were incubated for 72 h, and luciferase and ␤-galactosidase activities were measured. The data are presented as relative light units/milliunit of ␤-galactosidase activity and represent the mean of at least three independent experiments with the S.E. indicated. *, p Ͻ 0.05 versus empty vector control. FIGURE 3. Wnt 7a/Fzd 9 do not increase expression of PPAR␥ in NSCLC. A, total RNA purified from the indicated H157 and H2122 cell lines with or without Wnt 7a/Fzd 9 expression were submitted to quantitative reverse transcription PCR using primers specific for PPAR␥ as described under "Experimental Procedures." Cells overexpressing PPAR␥ were used as a positive control. The relative mRNA abundance for PPAR␥ in the different samples was normalized to GAPDH (primers specific for GAPDH as described under "Experimental Procedures") measured by quantitative reverse transcription PCR in the same samples. B, aliquots of extracts containing equal protein as measured by Bradford assay from the indicated cells with and without Wnt 7a/Fzd 9 were resolved by SDS-PAGE and immunoblotted with an antibody to PPAR␥ (55 kDa; Cell Signaling). Cells overexpressing PPAR␥ were used as a positive control. The filters were stripped and re-immunoblotted for ␤-actin (42 kDa; Abcam).
Neither PPAR␥ mRNA nor Protein Expression Is Induced in NSCLC Cells as a Result of Wnt 7a and Fzd 9 Expression-We sought to determine whether increased PPAR␥ activity in response to expression of Wnt 7a/Fzd 9 was mediated through increased expression of the protein. Our quantitative reverse transcription PCR experiments revealed that stable expression of Wnt 7a and Fzd 9 in either H157 or H2122 had no significant effect on PPAR␥ mRNA levels (Fig. 3A). This was confirmed by immunoblotting, where stable expression of Wnt 7a and Fzd 9 failed to cause a significant increase in the PPAR␥ expression (Fig. 3B). As a positive control, stable expression of wild-type PPAR␥ resulted in a marked increase in PPAR␥ protein and mRNA expression. These data indicate that Wnt 7a and Fzd 9 activation of PPAR␥ is not mediated through increased expression but by activation of PPAR␥.
Wnt 7a and Fzd 9 Activation of PPAR␥ Is Mediated through Activation of ERK5-Previous studies have demonstrated that ERK5 is capable of mediating PPAR␥ transcriptional activation in endothelial cells (18). To test the hypothesis that ERK5 stimulates activation of PPAR␥ in NSCLC, we examined the effect of Wnt 7a and Fzd 9 expression in both H157 and H2122 cell lines on ERK5 phosphorylation by immunoblotting with a phospho-ERK5-specific antibody. Expression of Wnt 7a/Fzd 9 resulted in a 5-to 10-fold increase in pERK5 relative to the empty vector control cell lines (Fig. 4A). The levels of total ERK5 were not changed (Fig. 4A, lower panels). Under these FIGURE 4. Stimulation of PPAR␥ by Wnt 7a/Fzd 9 is mediated through activation of ERK5. A, the various cell lines were exposed for 48 h with either nothing, 25 M PD98059, or 10 M U0126 and then immunoblotted. Extracts were prepared from the indicated stable transfectants using MAPK lysis buffer, and aliquots containing 100 g of protein were resolved on 10% polyacrylamide SDS gels, transferred to nitrocellulose, and probed with an antibody to phospho-ERK5 (115 kDa; Cell Signaling) or phospho-ERK1,2 (42, 44 kDa; Cell Signaling). The filters were stripped and re-immunoblotted for total ERK5 or total ERK1,2 as loading controls. B, the indicated NSCLC cell lines were transiently transfected with PPAR-RE, empty vector pCDNA3, Wnt 7a/Fzd 9 (H157), or Fzd 9 (H2122) along with CMV-␤-galactosidase to normalize for transfection efficiency. After an overnight incubation, cells were exposed for 48 h with either 25 M PD98059 or 10 M U0126. Extracts were prepared and promoter activity determined as luciferase units normalized to ␤-galactosidase. Results represent the mean of three independent experiments with the S.E. indicated. *, p Ͻ 0.05 versus pCDNA-3; **, p Ͻ 0.05 versus Control. C, the indicated NSCLC cell lines were transiently transfected with PPAR-RE and either empty vector (pCDNA3), Wnt 7a/Fzd 9 (H157), or Fzd 9 (H2122), either with or without dominant negative MEK5, along with CMV-␤-galactosidase to normalize for transfection efficiency. As described under "Experimental Procedures" the cells were incubated for 72 h and luciferase and ␤-galactosidase activities were measured. The data are presented as relative light units/milliunit of ␤-galactosidase activity. The results represent the mean of three independent experiments with the S.E. indicated. *, p Ͻ 0.05 versus empty vector; **, p Ͻ 0.05 versus Wnt7a/Fzd 9 in H157 cells or Fzd 9 in H2122 cells.
conditions we did not observe any significant increase in ERK1/2 activity, either by blotting with a phospho-ERK antibody (Fig. 4A) or by direct measurement of ERK1/2 kinase activity (data not shown).
We also examined the effects of pharmacological inhibitors of the ERK family of MAP kinases. PD 98059 (25 M), a known MEK inhibitor, decreased expression of phospho-ERK1, 2, and 5 on immunoblot (Fig. 4A), consistent with what has been previously reported (19,20). Exposure of either H157 or H2122 cells to this concentration of PD98059 had no effect on basal PPAR␥ activity in control cells transfected with empty vector but potently inhibited the increase seen in cells expressing Wnt 7a/Fzd 9 (Fig. 4B). Because these results do not rule out a role for ERK1/2 in regulation of PPAR␥ activity, we employed a second MEK inhibitor, U0126. This agent inhibited MEK1/2 but did not affect MEK5 activity at the concentration used in this study (10 M) (Fig. 4, A and B) (21). The addition of U1026 to the various cell lines did not lead to a reduction of phospho-ERK5 protein expression or PPAR-RE reporter activity in either the H157 Wnt 7a/Fzd 9 or H2122 Fzd 9 cell lines (Fig. 4, A and B). From these studies, we propose that increased PPAR␥ activity is mediated through activation of MEK5/ERK5. This finding is consistent with the lack of an effect of Wnt7a/Fzd 9 on ERK1/2 activity (6).
To determine the role of ERK5 in activating PPAR␥, we coexpressed dominant negative MEK5 an inhibitor of ERK5 with Wnt7a and Fzd 9. Expression of dominant negative MEK5 had no effect on basal PPAR-RE activity but inhibited the activation stimulated by expression of Wnt 7a and Fzd 9 in both the H157 and H2122 cell lines. In H157 cells, PPAR␥ activity was decreased by ϳ60%, whereas in H2122 cells the decrease was Ͼ90% (Fig. 4C). Wnt 7A and Fzd 9 Expression Results in Reduced Expression of the PPAR␥ Co-repressor SMRT-Past studies have shown that PPAR␥ activity is inhibited by the co-repressors silencing mediator of retinoid and thyroid hormone receptor (SMRT) and NCor (18,22). Activation of PPAR␥ by ligands is associated with displacement of co-repressors by coactivators, which is mediated through ubiquitination of co-repressors, and subsequent degradation via the proteasome pathway. To determine whether this mechanism was involved in activation of PPAR␥ by Wnt 7a/Fzd9, we examined expression of SMRT. Co-expression of Wnt 7a and Fzd 9 in NSCLC significantly reduced the protein expression of SMRT (Fig. 5).
A PPAR␥ Inhibitor SR 202 Blocks the Effects of Wnt 7a/Fzd 9 on Transformed Growth of NSCLC-The role of PPAR␥ in mediating the effects of Wnt 7a/Fzd 9 on transformed growth of NSCLC was assessed FIGURE 5. Wnt 7a/Fzd9 decreases expression of the PPAR␥ co-repressor SMRT. Extracts were prepared from the indicated transfectants using MAPK lysis buffer, and aliquots containing 100 g of protein were resolved on 10% polyacrylamide SDS gels, transferred to nitrocellulose, and probed with an antibody to SMRT (275 kDa; Santa Cruz). The filters were stripped and re-immunoblotted for ␤-catenin (80 kDa; BD Transduction Laboratories) as a loading control. using a recently described pharmacological antagonist, SR202 (23,24). To confirm the efficacy of this drug, H2122 cells, either stably expressing Fzd 9 or empty vector, were transiently transfected with the PPAR-RE-luciferase construct in the presence or absence of SR202 (20 M). Exposure to SR202 completely prevented the increase in PPAR␥ activity seen with Fzd 9 expression (Fig. 6A), confirming our earlier conclusion that Wnt 7a and Fzd 9 selectively activate PPAR␥. We next assessed the effects of this agent on colony formation in soft agar, a measure of transformed growth. Consistent with our earlier findings, expression of Fzd 9 in H2122 cells (6) strongly inhibited colony formation (Fig. 6B). Treatment with SR202 had no effect on colony formation in control cells transfected with empty vector. However, exposure to SR202 reversed the inhibition of colony formation seen with expression of Fzd 9 and resulted in ϳ50% of the number of colonies seen in control cells (Fig. 6, B and C). These data indicate that blocking PPAR␥ acti-vation partially reverses the inhibitory effects of Wnt 7a/Fzd 9, suggesting that PPAR␥ is a major effector pathway downstream from Wnt 7a/Fzd 9.
E-cadherin Is Reduced in Cell Lines Expressing Wnt 7a and Fzd 9 by the PPAR␥ Inhibitor Compound SR 202-E-cadherin is a well known marker of epithelial differentiation (25,26). We have previously shown that E-cadherin is induced by both Wnt 7a/Fzd 9 and PPAR␥ expression in NSCLC (6,14,15). We evaluated the effects of the PPAR␥ antagonist SR202 on NSCLC cell lines expressing Wnt 7a/Fzd 9. Expression of Wnt 7a/Fzd 9 resulted in a 5-to 10-fold increase in E-cadherin expression relative to the empty vector control cell lines, consistent with previous findings (Fig.  7A). The addition of SR202 to the cell lines expressing Wnt 7a/Fzd 9 resulted in a marked reduction of E-cadherin in both H157 and H2122 cells. Total ␤-catenin was used as a loading control (Fig. 7A, bottom  panels).
We confirmed these findings by examining effects on the E-cadherin promoter (27). H157 and H2122 cells, either stably expressing Wnt 7a/Fzd 9 or empty vector, were transiently transfected with the E-cadherin-luciferase construct in the presence or absence of SR202. Wnt7a and Fzd 9 signaling stimulated E-cadherin promoter activity 5-to 10-fold, and SR202 inhibited this increase by Ͼ50% (Fig. 7B). These data indicate that regulation of E-cadherin is controlled by transcriptional regulation.

DISCUSSION
Deregulation of the Wnt pathway plays a key role in the development of various tumors (1,5,28). Whereas the canonical Wnt pathway has been demonstrated to be pro-tumorigenic, Wnt 7a is critical for the formation of normal epithelium in development and plays a significant role in maintaining a normal epithelial phenotype in the lungs of adults (29). Wnt 7a is frequently lost in NSCLC (6,30), and its re-expression in the context of Fzd 9 has been shown to restore increased differentiation and decrease the transformed phenotype in cancer cell lines (6). In separate studies, we (14) and others have shown that thiazolidinediones, which are pharmacological activators of PPAR␥, inhibit growth and promote increased differentiation in various types of tumors, including NSCLC (31,32). We have also shown FIGURE 7. E-cadherin protein expression and promoter activity induced by Wnt 7a/Fzd 9 are reduced by the PPAR␥ antagonist SR202. A, extracts were prepared from the indicated transfectants using MAPK lysis buffer, and aliquots containing 100 g of protein were resolved on 10% polyacrylamide SDS gels, transferred to nitrocellulose, and probed with an antibody to E-cadherin (125 kDa; BD Transduction Laboratories). The filters were stripped and re-immunoblotted for ␤-catenin (80 kDa; BD Transduction Laboratories) as a loading control. B, the indicated NSCLC cell lines were transiently transfected with the E-cadherin promoter construct, along with CMV-␤-galactosidase to normalize for transfection efficiency. After an overnight incubation, cells were treated for 48 h with 20 M SR 202 or vehicle. Extracts were prepared and promoter activity determined as luciferase units normalized to ␤-galactosidase. Results represent the mean of three independent experiments with the S.E. indicated. *, p Ͻ 0.05 versus empty vector; **, p Ͻ 0.05 versus Wnt7a and Fzd 9 in H157 cells or Fzd 9 in H2122 cells.
that increased PPAR␥ activity, induced by overexpression of PPAR␥, inhibited transformed growth and increased expression of epithelial markers in H2122 and H157 cells (15). In the current study we have demonstrated that these two pathways are linked. Specifically, expression of Wnt 7a/Fzd 9 leads to a marked increase in PPAR␥ activity. Furthermore, blocking PPAR␥ with a pharmacological inhibitor reverses both the inhibition of transformed growth and the induction of E-cadherin expression seen with Wnt 7a/Fzd 9, implying that activation of PPAR␥ is a major effector pathway in mediating the anti-tumorigenic effects of Wnt 7a/Fzd 9 in NSCLC.
Because Wnt 7a/Fzd 9 did not significantly increase expression of either PPAR␥ mRNA or protein, our data support a mechanism in which Wnt 7a and Fzd 9 activate PPAR␥ through a post-translational mechanism involving activation of ERK5. ERK5 is a member of the MAP kinase family that is activated by redox and hyperosmotic stress, growth factors, and pathways involving certain G-protein-coupled receptors (20,33). ERK5 has a dual phosphorylation site like ERK1/2 but has a unique loop-12 structure and a large carboxyl-terminal, suggesting that its regulation and function may be different from that of ERK1/2 (34).
In endothelial cells, flow-mediated activation of ERK5 has been shown to lead to activation of PPAR␥ by direct binding to the hinge-helix region of PPAR␥ (18). This activation requires phosphorylation of ERK5 by the upstream activating kinase MEK5. Our data are consistent with this model, because expression of dominant negative MEK5 and pharmacological inhibition of MEK5 blocked the activation of PPAR␥ activity by Wnt 7a/Fzd 9. It has been demonstrated that ERK1/2 phosphorylates PPAR␥ on serine 82, which is in the AF1 region of the molecule (35,36). This phosphorylation inhibits transcriptional activation by decreasing the affinity for ligand (37). Therefore, inhibition of ERK1/2 would not be expected to decrease PPAR␥ activity. The effects of ERK5 could be mediated by controlling the association of PPAR␥ with co-repressors. Studies have shown that activation of MEK5/ERK5 caused displacement of silencing mediator of retinoid and thyroid hormone receptors (SMRT), a potent transcriptional co-repressor of PPAR␥ (38), resulting in PPAR␥ transcriptional activation (18). Similarly we have observed that Wnt 7a and Fzd 9 expression leads to reduced SMRT protein expression by immunoblotting (Fig. 5). This is an attractive model of activation of PPAR␥, and in the absence of increases in PPAR␥ mRNA and protein expression could potentially account for PPAR␥ activation.
Wnt activation of PPAR␥ has been demonstrated in colon cancer cells (28). In these studies, pro-tumorigenic signaling through Wnts resulted in increased expression of PPAR␥ by a post-transcriptional effect. Overexpression of ␤-catenin also increased PPAR␥, indicating that this effect was mediated through the canonical Wnt pathway (28). The biological effects of PPAR␥ in colon cancer remain controversial, with studies supporting both pro-and anti-tumorigenic roles. In contrast, studies in lung cancer, including data from our laboratory, support an anti-tumorigenic role for PPAR␥. Furthermore, we have shown that Wnt 7a/Fzd 9 signaling is not mediated through the canonical Wnt pathway (6), indicating that the mechanisms of PPAR␥ induction are cell context specific.
Our studies using the pharmacological PPAR␥ inhibitor do not show a complete reversal of the effects of Wnt 7a/Fzd 9, and this may be due to other effector pathways that are independent of PPAR␥. In this regard, we have previously demonstrated that Wnt 7a/Fzd 9 expression leads to activation of the JNK (c-Jun NH 2 -terminal kinase) pathway and increased expression of Sprouty-4 in NSCLC (6). It is likely that these pathways cooperate with activation of PPAR␥ in reversing the transformed phenotype. In summary, these data demonstrate a novel connection between Wnt signaling and activation of PPAR␥ in lung cancer. Our data as well as those of others raise the possibility that PPAR␥ activators may represent an effective treatment of patients with lung cancer. As with epidermal growth factor receptor inhibitors, these agents may be targeted to specific subpopulations of lung cancer cells that lack either Wnt 7a or Fzd 9. Our data would suggest that more specific pharmacological activators of ERK5 or PPAR␥ might have utility in the treatment of lung cancer.