A novel peroxisome proliferator–activated receptor γ ligand, MCC-555, induces apoptosis via posttranscriptional regulation of NAG-1 in colorectal cancer cells

The peroxisome proliferator–activated receptor γ (PPARγ) is a ligand-activated transcription factor that belongs to the nuclear hormone receptor superfamily. PPARγ forms a heterodimeric complex with the retinoid X receptor and then binds to the PPAR response element. This interaction can be responsible for the regulation of cellular events ranging from glucose and lipid homeostasis to cell differentiation and apoptosis (1). PPARγ ligands include the natural prostaglandin 15-deoxy-Δ^12,14-prostaglandin J₂, the synthetic antidiabetic thiazolidinediones, and oxidative metabolites of polysaturated fatty acids. These ligands activate PPARγ and exhibit antitumorigenic effects in many types of tumor cells, including breast (2), lung (3), prostate (4), and pancreas (5). However, the most extensive investigations have focused on the colon (6), where PPARγ is highly expressed both in well-differentiated and poorly differentiated adenocarcinomas and in normal colonic mucosa (7). Treatment with rosiglitazone inhibits anchorage-independent growth of several colorectal carcinoma cancer cell lines and the degree of antiproliferative activity correlates with the levels of functional PPARγ, suggesting that activation of PPARγ by rosiglitazone results in the inhibition of cell growth (8). On the other hand, a large body of evidence shows that the antitumorigenic activity of PPARγ ligands, such as troglitazone and 15-deoxy-Δ^12,14-prostaglandin J₂, is independent of PPARγ activation in many cell types (9–11). Our laboratory has reported that troglitazone, independent of PPARγ, can induce the expression of early growth response 1 transcription factor and subsequently result in the increased expression of nonsteroidal anti-inflammatory drug–activated gene (NAG-1; refs. 12, 13). However, troglitazone was voluntarily withdrawn from the market in March 2000 because it caused severe idiosyncratic liver injury (14). A new class of thiazolidinediones was developed and subsequent studies established one member, MCC-555 (also known as RWJ-241947), as an antidiabetic drug in animal models of type II diabetes. Like other thiazolidinediones, MCC-555 binds to PPARγ and increases transcriptional activities, but its binding affinity for PPARγ is relatively weak. Interestingly, the antidiabetic potency of MCC-555 is more effective than that of the other thiazolidinediones such as rosiglitazone (15). Moreover, cardiac and hematopoietic side effects were reduced compared with other thiazolidinediones (16), which may be associated with the unique properties of MCC-555 in the activation of PPARγ (i.e., MCC-555 acts as a full agonist, partial agonist, or antagonist depending on the cell type or DNA binding site). In addition to its antidiabetic activity, MCC-555 has potent antiproliferative effects against prostate cancer in vivo and enhances the antitumor activity of As₂O₃ against myeloma cells (17). These results prompted us to investigate MCC-555 antitumorigenic activity in colorectal cancer cells.

NAG-1 has previously been identified by us and other groups (18–21). NAG-1 protein has broad activity in various organs but the molecular mechanisms responsible for these functions have not been determined in detail. However, NAG-1 has been shown to possess antitumorigenic and proapoptotic activity in some types of cancer cells (19, 22, 23). NAG-1 is up-regulated in several epithelial cancer cell lines by several nonsteroidal anti-inflammatory drugs (23), as well as by antitumorigenic compounds such as resveratrol (24), catechins (25), indole-3 carbinol (26), conjugated linoleic acid (27), PPARγ ligands (13), DIM derivatives (28), horehound extracts (29), and 5F-203 (30). In contrast to other transforming growth factor-β superfamily genes, NAG-1 contains a p53 binding site in the 5′ upstream region (19, 22, 23), and several dietary compounds induce NAG-1 expression via p53 (24, 25). Furthermore, NAG-1 expression is regulated by the tumor suppressor protein early growth response 1 and by the glycogen synthase kinase 3β pathway (13, 31). These previous reports provide evidence that supports a possible role of NAG-1 as a tumor suppressor.

In the present study, we report that MCC-555, a novel PPARγ ligand, induces apoptosis in human colorectal cancer cells, and the apoptosis seems to be mediated by NAG-1. Proapoptotic protein NAG-1 is posttranscriptionally up-regulated by MCC-555 at the mRNA level, in contrast to transcriptional regulation observed for other antitumorigenic compounds. The mechanisms by which MCC-555 induces apoptosis may be, in part, associated with the up-regulation of NAG-1 independent of PPARγ activation.

Human colorectal cancer cells HCT-116 and LoVo were purchased from American Type Culture Collection (Manassas, VA). HCT-116 p53−/− cells were generously provided by Dr. Bert Vogelstein (Johns Hopkins University, Baltimore, MD). Ciglitazone, rosiglitazone, MCC-555, and GW9661 were purchased from Cayman Chemical Co. (Ann Arbor, MI). Troglitazone was obtained from Calbiochem (San Diego, CA). Antihuman NAG-1 antibody was previously described (23). Antibodies against actin, extracellular signal–regulated kinase 1 (ERK1), and ERK2 were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Phospho-ERK1/2, p38 mitogen-activated protein kinase (MAPK), phospho-p38 MAPK, c-jun NH₂-terminal kinase, and phospho-c-jun NH₂-terminal kinase antibodies were obtained from Cell Signaling Technology (Beverly, MA).

For cloning a luciferase-NAG-1 3′-untranslated region (3′UTR) hybrid construct, the 3′UTR of NAG-1 was amplified by PCR using primers containing XbaI recognition sequences: 5′-gctctagagcagtcctggtccttccactgt-3′ (top strand; XbaI site is underlined) and 5′-gctctagagctatcaccagagttttaaata-3′ (bottom strand; XbaI site is underlined). The amplified products were cloned into the pCR2.1 TOPO vector (Invitrogen, Carlsbad, CA); amplified fragment was then released by digestion with XbaI and was cloned into pGL3 promoter vector (Promega, Madison, WI) downstream of the luciferase gene (pSV40-LUC-NAG3′UTR). To delete AU-rich element (ARE) sequences on this construct, the QuickChange II site-directed mutagenesis kit (Stratagene, Cedar Creek, TX) and the following primers were used: 5′-ggaactgtgtaaactctggtgatagctcta-3′ (top strand) and 5′-accagagtttacacagttccatcagaccag-3′ (bottom strand). Several NAG-1 promoter clones linked to luciferase have previously been reported (32). All plasmids were sequenced for verification.

HCT-116 cells were plated in 12-well plates at 10⁵ per well and were grown for 16 hours. Plasmid mixtures containing 0.5 μg of reporter vector and 0.05 μg of pRL-null (Promega) were transfected by LipofectAMINE (Invitrogen) according to the protocol of the manufacturer. To assess the transactivation of PPARs, 0.25 μg of pMH100x4-TK-LUC plasmid and 0.25 μg of pCMX-Gal-mPPARα-, γ-, or δ-LBD were transfected into HCT-116 cells using LipofectAMINE. Plasmids for the PPAR study were kindly provided by Dr. Ronald M. Evans (Howard Hughes Medical Institute, La Jolla, CA). After transfection, the media were replaced with serum-free media and indicated reagents were added. For inhibitor experiments, the transfected cells were pretreated with PD98059 (Calbiochem), U0126 (Cell Signaling Technology), or SB203580 (Sigma, St. Louis, MO) for 30 minutes and subsequently treated with MCC-555 for 12 hours. Cells were harvested in 1× luciferase lysis buffer and luciferase activity was determined and normalized to the pRL-null luciferase activity using a dual luciferase assay kit (Promega).

Cells were treated with MCC-555 in the absence of serum and lysed with radioimmunoprecipitation assay buffer (1× PBS, 1% NP40, 0.5% sodium deoxycholate, 0.1% SDS) supplemented with protease and phosphatase inhibitors (1 μg/mL aprotinin, 1 μg/mL leupeptin, 1 mmol/L phenylmethylsulfonyl fluoride, 1 mmol/L Na₃VO₄, and 25 mmol/L NaF). The soluble protein concentrations were determined by the bicinchoninic acid protein assay kit (Pierce Biotechnology, Rockford, IL). Proteins were separated by SDS-PAGE and transferred onto nitrocellulose membranes (Osmonics, Minnetonka, MN). The membranes were incubated first with primary antibody at 4°C overnight and then with horseradish peroxidase–conjugated secondary antibody for 1 hour. The signal was detected by the enhanced chemiluminescence system (Amersham Biosciences, Piscataway, NJ) and quantified by Scion Image Software (Scion Corp., Frederick, MD).

The NAG-1 small interfering RNA (siRNA) vectors were constructed using a pSuper-retro-puro (OligoEngine, Seattle, WA) as previously described (33). HCT-116 cells were transfected with 2 μg of the siRNA vectors using LipofectAMINE 2000 (Invitrogen) for 24 hours. The cells were grown and treated with MCC-555 in the presence of serum. For suppressing ERK1/2 expressions, ERK1, ERK2, and control siRNAs were purchased from Cell Signaling Technology. HCT-116 cells were transfected with the siRNAs (ERK1, 150 nmol/L and ERK2, 80 nmol/L) using TransIT-TKO transfection reagent (Mirus, Madison, WI). After transfection for 24 hours, the media were replaced with serum-free media containing vehicle or MCC-555 (5 μmol/L). The cells were incubated for 24 hours in the absence of serum.

Total RNA was isolated with TRIzol (Invitrogen) according to the instructions of the manufacturer. NAG-1 cDNA was labeled with biotin-N4-dCTP using Biotin Random Primer Kit (Pierce Biotechnology). Total RNA (10 or 20 μg) was separated on 1.2% agarose gels containing formaldehyde and transferred to nylon membranes. Hybridization and chemiluminescent signal detection were done using the North2South Chemiluminescent Hybridization and Detection Kit (Pierce Biotechnology).

For detection of apoptotic cells, cells were stained with FITC-labeled Annexin V using the Annexin V-FITC apoptosis detection kit (BD Biosciences PharMingen, San Diego, CA) according to the instructions of the manufacturer. A total of 10,000 cells were examined by flow cytometry using a Beckman Coulter Epixs XL equipped with ADC and ModFit LT software. Cells were gated on side scatter and forward scatter to exclude debris. Doublets were eliminated using peak versus integral analysis. Annexin V–positive/propidium iodide–positive and Annexin V–positive/propidium iodide–negative cell populations were determined as apoptotic populations from the total gated cells. For detection of nuclear fragmentation, cells were stained with 4′,6-diamidino-2-phenylindole. The results were analyzed using a Nikon Eclipse E600 (Nikon, Tokyo, Japan). The images were captured with a MicroPublisher (QImaging, Burnaby, British Columbia, Canada) and processed with QCapture Pro software.

HCT-116 cells were treated with different thiazolidinediones, including ciglitazone (5 μmol/L), rosiglitazone (5 μmol/L), troglitazone (5 μmol/L), and MCC-555 (5 μmol/L), or with vehicle for 24 hours. The cell lysates were harvested and Western blot analysis was done. A dramatic induction of NAG-1 expression was observed in troglitazone- and MCC-555-treated HCT-116 cells although the degree of induction by MCC-555 was not as high as observed for troglitazone. On the other hand, ciglitazone and rosiglitazone treatment slightly increased the expression of NAG-1 (Fig. 1A). We also examined NAG-1 expression in other colorectal cancer LoVo cells, and MCC-555 treatment induced NAG-1 expression in those LoVo cells (data not shown). Because some antitumorigenic phytochemicals such as resveratrol (24) evoke induction of NAG-1 expression through p53, we were prompted to analyze the effects of p53 on thiazolidinedione-induced NAG-1 expression. NAG-1 expression in response to thiazolidinediones was thus examined in p53-null HCT-116 (HCT-116 p53−/−) cells. These thiazolidinediones increased NAG-1 expression in p53-null cells, indicating that thiazolidinedione-induced NAG-1 expression is p53 independent (Fig. 1A). Next, NAG-1 expression was analyzed after treatment of HCT-116 cells with various concentrations of MCC-555 (0.1–10 μmol/L). As shown in Fig. 1B, potent expression was observed at a concentration of 5 μmol/L. The induction of NAG-1 was also dependent on the time of incubation, with the maximal increase occurring at 24 hours (Fig. 1C).

NAG-1 has antitumorigenic and proapoptotic activities (19, 22, 23). As shown in Fig. 1, MCC-555 significantly increased NAG-1 expression in human colorectal cancer cells. Therefore, we investigated whether NAG-1 plays an important role in MCC-555-induced apoptosis. First, HCT-116 cells were treated with MCC-555 and apoptosis was detected by 4′,6-diamidino-2-phenylindole staining. The treated cells revealed the typical morphologic features of apoptotic cells, with condensed and fragmented nuclei (Fig. 2A). Consistent with the 4′,6-diamidino-2-phenylindole staining, flow cytometric analysis also revealed increased apoptotic cells in MCC-555-treated cells (Fig. 2C). To determine the contribution of NAG-1 to MCC-555-induced apoptosis, the endogenous NAG-1 gene was down-regulated by siRNA and apoptosis was analyzed by Annexin V staining in the absence or presence of MCC-555. RNA interference reduced MCC-555-induced NAG-1 by 80% (Fig. 2B). Moreover, knockdown of NAG-1 expression significantly suppressed apoptosis induced by MCC-555 (Fig. 2C). These data indicate that MCC-555 induces apoptosis in HCT-116 cells, and NAG-1 mediates, at least in part, MCC-555-induced apoptosis.

Three PPAR isoforms have been identified: PPARα, PPARγ, and PPARδ. Binding activities of thiazolidinediones for PPARs were determined using Gal4-PPAR fusion proteins. The luciferase reporter construct and expression vector (Fig. 3A, top) were cotransfected into HCT-116 cells. The cells were then treated with 1 μmol/L of indicated PPARγ ligands for 24 hours and examined for luciferase activity. The PPAR activity of thiazolidinediones was compared with vehicle. As expected, all the PPAR ligands showed binding capacity to PPARγ whereas none of the ligands exhibited binding capacity to PPARδ. Interestingly, MCC-555 exhibited significant PPARα activation, unlike other thiazolidinediones (Fig. 3A). However, PPARα may not be involved in the mechanism of MCC-555-induced NAG-1 expression because WY14643, a PPARα agonist, did not induce NAG-1 expression (data not shown) and HCT-116 cells do not express PPARα (12). MCC-555 strongly activated transcription by Gal4-PPARγ although the activation by MCC-555 was less than that by rosiglitazone and troglitazone (Fig. 3A). Moreover, these profiles of PPARγ transactivities by thiazolidinediones could lead us to expect that NAG-1 induction by MCC-555 is independent of PPARγ because the potency of PPARγ transactivity and NAG-1 induction are not correlated (Figs. 1A and 3A). To confirm the PPARγ-independent mechanism by which MCC-555 induces NAG-1 expression, a specific PPARγ antagonist, GW9661, was used. Concomitant treatment of several concentrations of GW9661 with MCC-555 resulted in the NAG-1 expression, which supports the evidence of a PPARγ-independent mechanism (Fig. 3B). In addition, the activity of GW9661 was confirmed to inhibit PPARγ transactivation by MCC-555 (Fig. 3C).

To investigate the molecular mechanism involved in MCC-555-induced NAG-1 expression, we initially examined the transcriptional activity of MCC-555 on the NAG-1 promoter using 3.5-kb NAG-1 promoter and the deletion clones pNAG3500/LUC, 1086/LUC, and 133/LUC as previously described (32). Treatment with MCC-555 slightly transactivated the NAG-1 promoter compared with vehicle treatment (data not shown), indicating that a transcriptional mechanism may not play a major role in MCC-555-induced NAG-1 expression. Posttranscriptional control in eukaryotic gene expression can also provide a regulatory process by determining the abundance of a particular protein (34). To determine the effects of MCC-555 on NAG-1 RNA stability, the half lives of mRNAs were estimated by Northern blot analysis using RNA from actinomycin D (5 μg/mL)–treated HCT-116 cells (Fig. 4A). As a result, the half lives of NAG-1 mRNA from vehicle- and MCC-555-treated cells were estimated to be around 58 and 102 minutes, respectively (Fig. 4A). This result suggests that MCC-555 may alter NAG-1 mRNA stability. A cycloheximide experiment determined further the property of stabilization of NAG-1 mRNA by MCC-555. HCT-116 cells were pretreated with or without 10 μg/mL cycloheximide for 30 minutes, followed by treatment with vehicle or MCC-555. The increased mRNA expression of NAG-1 was found to be insensitive to cycloheximide, suggesting that the mechanism of stabilization by MCC-555 is not involved in de novo synthesis (Fig. 4B).

The stability of mRNA is determined in many cases by interactions between specific RNA-binding proteins and cis-acting sequences located in the 3′UTR of the mRNA. The ARE, which targets the mRNA for rapid degradation, is well known as characterized cis-acting sequences (34–36). We have compared the 3′UTRs of human and mouse NAG-1 sequences and found that four copies of the ARE sequence (ARE1–ARE4) are conserved between human and mouse sequences (Fig. 5A). The 3′UTR of the NAG-1 gene was cloned into a pGL3 promoter vector, downstream of the luciferase reporter gene (pSV40-LUC-NAG3′UTR; Fig. 5B), and luciferase activity was measured. As shown in Fig. 5C, luciferase activity was reduced by ∼70% with insertion of the 3′UTR of the NAG-1 gene, compared with that of the reverse orientation of the 3′UTR as a control. These results suggest that the ARE-containing region plays an important role in destabilizing NAG-1 mRNA in the cells. The insertion of the reverse orientation sequences of 3′UTR did not affect luciferase activity compared with the pGL3 control promoter (data not shown). To further identify the role of ARE sequences in the RNA degradation, ARE3 and ARE4 were deleted on pSV40-LUC-NAG3′UTR. The reduced reporter activity in NAG-1 3′UTR-transfected cells was clearly restored by the deletion of AREs, supporting that ARE sequences cause rapid degradation of NAG-1 mRNA. The effect of MCC-555 on NAG-1 mRNA destabilization caused by AREs was assessed by luciferase activity. As shown in Fig. 5D, MCC-555 treatment substantially increased the luciferase activity in the cells transfected with pSV40-LUC-NAG3′UTR in a concentration-dependent manner, but not in that of reverse orientation. These data indicate that MCC-555 may inhibit NAG-1 RNA degradation, thereby inducing NAG-1 expression.

Three subgroups of the MAPK family have been identified and they function as mediators of intracellular signals in response to various stimuli. The MAPK pathway can play a role in the regulation of mRNA stability. Thus, we examined the effects of MAPK pathway inhibitors on MCC-555-induced NAG-1 mRNA stability. Cells transfected with pSV40-LUC-NAG3′UTR were pretreated with PD98059 (MAPK/ERK kinase inhibitor; 20 μmol/L), U0126 (MAPK/ERK kinase inhibitor; 2 μmol/L), or SB203580 (p38 MAPK inhibitor; 10 μmol/L) for 30 minutes and treated with MCC-555 for 12 hours. As shown in Fig. 6A, the addition of PD98059 with MCC-555 completely abolished MCC-555-induced reporter activity whereas SB203580 did not affect it. Similarly, MAPK/ERK kinase inhibitor U0126 also inhibited luciferase activity, suggesting that the MAPK/ERK kinase-ERK signaling pathway may be involved in the posttranscriptional regulation of NAG-1 by MCC-555. Consistent with the reporter experiment, we found that MCC-555 phosphorylates ERK1/2, but not p38 MAPK and c-jun NH₂-terminal kinase (Fig. 6B). Finally, to obtain the evidence that ERK is involved in MCC-555-induced NAG-1 expression, ERK1/2 was down-regulated by RNA interference. Down-regulation of ERK1/2 expression suppressed NAG-1 induction by MCC-555 (Fig. 6C). These results suggest that the ERK1/2 pathway mediates MCC-555-induced NAG-1 expression at the posttranscriptional level.

MCC-555 is characterized as a potent antidiabetic thiazolidinedione with reduced binding affinity for PPARγ compared with other thiazolidinediones (15). Recently, the antitumorigenic activity of MCC-555 has been shown in prostate cancer cell lines (LNCaP, PC-3, and DU145; ref. 17). However, MCC-555 has not been examined in other cancer cells, including colorectal cells, and further molecular mechanisms by which MCC-555 exhibits antitumorigenesis are not known. Here we report for the first time that MCC-555 induces apoptosis in human colorectal cancer cells, which is accompanied by up-regulation of NAG-1 via a PPARγ-independent mechanism.

Thiazolidinediones are a class of synthetic PPARγ ligands that regulate a number of genes. We showed that thiazolidinediones induced antitumorigenic and/or proapoptotic protein NAG-1 expression in several types of human colorectal cancer cell lines (Fig. 1). As a result, we confirmed that MCC-555 can cause a significant increase in apoptosis of human colorectal cancer cells (Fig. 2A and C). Importantly, down-regulation of the NAG-1 gene by RNA interference revealed involvement of NAG-1 function in MCC-555-induced apoptosis in HCT-116 cells (Fig. 2B and C). Mutations of the p53 tumor suppressor gene are frequently observed in human colon tumors (37, 38), and these mutations result in a loss of tumor suppressor function and thereby abolish the ability of such agents to prevent cancer from developing via p53-dependent mechanisms. Interestingly, MCC-555 could induce NAG-1 expression in p53-null cells, suggesting that p53 is not required for MCC-555-induced NAG-1 expression. Because troglitazone has been found to cause serious liver dysfunction (14), the novel PPARγ ligand MCC-555 may be able to provide the alternative therapeutic benefits on antitumorigenesis.

Thiazolidinediones have been shown to function as activating ligands for PPARγ, and the abilities of thiazolidinediones to activate PPARγ and to exhibit antitumorigenic effects are well correlated in many types of tumor cells (2, 4, 5, 8). However, thiazolidinediones also possess a PPARγ-independent mechanism to inhibit gastrointestinal tumor cell growth (10). Levels of PPARγ are not correlated with the rates of cell proliferation in human colon cancer cell lines (39), and troglitazone induces early growth response 1 and NAG-1 via a PPARγ-independent pathway (12, 13). We have shown that LY294002, a phosphatidylinositol 3-kinase inhibitor, affects antitumorigenic activity in both phosphatidylinositol 3-kinase–dependent and –independent manners (31, 40). Thus, the anticancer activity of a specific compound should be explored independently of the levels and/or genetic status of target proteins. In this report, we found that MCC-555 induces NAG-1 expression in a PPARγ-independent manner because pretreatment with PPARγ antagonist GW9661 did not affect MCC-555-induced NAG-1 expression (Fig. 3B and C). Moreover, results from the reporter activity of thiazolidinediones for PPARγ support the independent mechanism because the degree of thiazolidinedione-induced Gal4-PPARγ transcription activity is not related to NAG-1 expression. Taken together with NAG-1 siRNA results, these data provide evidence that the apoptotic activity of MCC-555 may be, in part, responsible for a PPARγ-independent mechanism in human colorectal cancer cells.

To investigate the molecular mechanism involved in MCC-555-induced NAG-1 expression, the effect of MCC-555 on NAG-1 promoter activity was analyzed. However, treatment with MCC-555 had little effect on the promoter activity (data not shown). It has been known that regulation of gene expression by posttranscriptional modification of mRNA stability and translation is also an important mechanism used in the control of cell growth (41). Namely, the rate of mRNA degradation determines the extent and duration of gene expression. The 3′UTR of some mRNAs contains AREs, which regulate degradation of mRNAs and provide an effective way to control protein expression by regulation of mRNA half-life and translation. In particular, cellular events such as differentiation, proliferation, apoptosis, and inflammation are known to be related to modifications of the rate of RNA degradation mediated by the ARE motif (34, 35, 42, 43). Examination of sequences revealed that the 3′UTR of NAG-1 mRNA contains highly conserved AREs (Fig. 5A), and these AREs should implicate the rapid degradation of mRNA. Indeed, insertion of the 3′UTR of NAG-1 into the reporter vector (Fig. 5B) caused a dramatic reduction in luciferase activity, and deletion of two ARE sequences restored the reduction (Fig. 5C). These results indicate that ARE sequences in the 3′UTR play a pivotal role in the degradation of NAG-1 mRNA. To determine whether MCC-555 protects against decay of NAG-1 mRNA, Northern blot analysis was done. MCC-555 induced a significant stabilization of NAG-1 mRNA in the presence of actinomycin D, an inhibitor of transcription (Fig. 4A). In fact, MCC-555 can substantially restore luciferase activity in a cell transfected with pSV40-LUC-NAG3′UTR (Fig. 5D), suggesting that loss of mRNA destabilization by MCC-555 is responsible for MCC-555-induced NAG-1 expression.

The regulation of ARE-binding proteins is a major mechanism in ARE-mediated RNA degradation (35). The MAPK pathway could play an important role in the posttranscriptional regulation of genes because a number of ARE-binding proteins require phosphorylation by MAPK for activation (44). The specific p38 inhibitor SB203580 had no influence on NAG-1 mRNA stabilization by MCC-555 (Fig. 6A), although p38 is a well-characterized mediator with a major role in regulating mRNA stability. The ERK pathway is also linked to posttranscriptional regulation. Indeed, the ERK pathway controls tumor necrosis factor-α, interleukins, and β-adrenergic receptor expressions at the posttranscriptional level (45–48). Additional treatment of PD98059 or U0126 with MCC-555 inhibited the protective effect of MCC-555 against NAG-1 mRNA decay (Fig. 6A). Consistent with the MAPK inhibitor experiment, ERK, but not p38 or c-jun NH₂-terminal kinase, is phosphorylated by MCC-555 treatment. In addition, a cycloheximide experiment determined that the mechanism of stabilization by MCC-555 does not require de novo protein synthesis, supporting the importance of protein modification, such as phosphorylation (Fig. 4B). Recent studies have shown that ERK activation is associated with cell growth arrest and apoptosis, although the ERK pathway is linked to cell proliferation and tumorigenic activity as well. In fact, downstream targets of ERK1/2, such as the Fas/APO-1 receptor and p53, have been identified as mediators for ERK1/2-triggered apoptosis (49, 50). Furthermore, activation of the ERK pathway is likely important in the mediation of some genotoxic reagent-induced apoptosis. For instance, ERK siRNA suppressed apoptosis induced by asiatic acid, a plant-derived compound, in breast cancer cell lines MCF-7 and MDA-MB-231 (51). l-Ascorbic acid and diallyl disulfide phosphorylate ERK1/2 in myeloid leukemia and prostate cancer cells, respectively, and pharmacologic inhibition of the ERK pathway prevents compound-dependent apoptosis (52, 53). Our data also showed that ERK1/2 siRNA suppressed proapoptotic protein NAG-1 induction by MCC-555 (Fig. 6C). Therefore, this evidence supports our hypothesis that MCC-555 activates the ERK pathway to cause loss of NAG-1 mRNA decay, followed by induction of apoptosis (Fig. 7). However, ERK pathway–related ARE-binding proteins have not been characterized in detail. Identification of these ARE-binding proteins may provide further understanding of how the ERK pathway plays an important role in RNA stability.

In summary, our data show for the first time that MCC-555 induces apoptosis in human colorectal cancer cells. MCC-555-induced apoptosis could be PPARγ dependent because MCC-555 strongly activates PPARγ. However, our results showed that the PPARγ-independent pathway is also responsible for apoptosis. In addition, a posttranscriptional mechanism of regulation of NAG-1 via the ERK pathway is responsible for the MCC-555 induced NAG-1 expression. Our study provides new insight into the pharmacologic approach against diseases caused by alterations to the rate of RNA degradation.

We thank Nichelle C. Whitlock for her technical assistance, Misty R. Bailey for her critical reading of manuscript, and Drs. Xuemin Xu and Mei-Zhen Cui (University of Tennessee) for helping with 4′,6-diamidino-2-phenylindole staining.