Cyclooxygenase-2 (COX-2), which is expressed by cholangiocytes in biliary tract disorders, has recently been implicated in biliary tract carcinogenesis. The mechanisms responsible for this COX-2 expression remain unclear. In human diseases, bile contains oxygenated derivatives of cholesterol (oxysterols) which possess diverse biological properties. Therefore, we determined if oxysterols modulate COX-2 expression. The effect of an oxysterol (22(R)-hydroxycholesterol, 22-HC) on COX-2 expression in KMBC cells, a human cholangiocarcinoma cell line, was examined. 22-HC enhanced COX-2 protein expression. This oxysterol activated p42/44 and p38 MAPK, but not JNK 1/2. A p42/44 MAPK inhibitor did not block COX-2 induction, while p38 MAPK inhibitor effectively attenuated COX-2 induction. Although COX-2 mRNA levels were increased by 22-HC, this increase was not transcriptionally regulated, as 22-OH did not increase activity in a COX-2 promoter gene assay. In contrast, COX-2 mRNA stability was augmented by 22-HC treatment, and this effect was reversed by a p38 MAPK inhibitor. In conclusion, the results demonstrate that the oxysterol 22-HC stabilizes COX-2 mRNA via a p38 MAPK-dependent mechanism. This enhanced COX-2 protein expression by oxysterols may participate in the genesis and progression of cholangiocarcinoma. (HEPATOLOGY 2004;39:732–738.)
Cholangiocarcinoma is a highly malignant, generally fatal neoplasm, originating from the bile duct epithelial cells or cholangiocytes of the intra- and extrahepatic biliary system. A number of risk factors have been shown to be important in the development of cholangiocarcinoma; most of these factors share long-standing inflammation and chronic injury of the biliary epithelium.1, 2 For example, primary sclerosing cholangitis (PSC), a chronic inflammatory disease of the biliary system, is well-established to predispose to the development of cholangiocarcinoma, and the prevalence of cholangiocarcinoma in this disease syndrome has been estimated to range from 5–15%.3 The precise mechanisms linking inflammation to biliary carcinogenesis remain obscure. Likely, inflammatory mediators upset the balance between cell death by apoptosis and cellular replication. Cyclooxygenase 2 (COX-2) is induced by epithelial cells during inflammation and has been implicated in epithelial cell carcinogenesis.4, 5 The prostanoids generated from this enzyme can promote cell replication and angiogenesis and dysregulate apoptosis.6, 7 COX-2 has been convincingly shown to be present in human cholangiocarcinomas.8, 9 However, the mechanisms linking inflammation to COX-2 expression in inflammatory biliary diseases remain unclear.
Oxysterols, oxygenated derivatives of cholesterol, are produced in inflammatory environments and have recently been identified in bile obtained from patients with inflammatory biliary disorders.10 This observation raises the possibility that oxysterols may contribute to COX-2 expression by cholangiocytes in inflammatory biliary tract disorders. Indeed, oxysterols have diverse biological activities, including regulation of gene expression.11 This regulation is mainly controlled by oxysterol interactions with various intracellular receptors, such as the oxysterol binding protein and the liver X receptor (LXR) nuclear orphan receptors, and activation of various kinase-dependent signaling cascades.11 Oxysterols may also modulate cellular apoptosis by altering intracellular calcium and/or diverse kinase signalings.12, 13 These studies, however, have largely been performed in vascular endothelial and smooth muscle cells, suggesting oxysterol participation in the pathogenesis of atherosclerosis. The role of oxysterols in mediating signaling cascades in epithelial cells such as cholangiocytes is less well defined.
The objective of this study was to ascertain if oxysterols modulate COX-2 expression levels in human cholangiocytes. To address this objective, we formulated the following questions: 1) Do oxysterols induce COX-2 protein expression in a human cholangiocarcinoma cell line? 2) If so, which intracellular signaling mechanisms are responsible for this induction? and 3) What is the mechanism of COX-2 induction by these signaling processes? Collectively, the results of the current study demonstrate that oxysterols stabilize COX-2 mRNA via a p38 mitogen activated protein kinase (MAPK)-dependent mechanism resulting in COX-2 protein accumulation in a human cholangiocarcinoma cell line. This enhanced COX-2 protein expression by oxysterol may participate in the genesis and progression of cholangiocarcinomas.
KMBC cells, a human cholangiocarcinoma cell line,14 were grown in DMEM supplemented with 10% fetal bovine serum, penicillin 100,000 U/L, streptomycin 100 mg/L, and gentamycin 100 mg/L. Cells were serum-starved for 24 hours prior to oxysterol treatment in order to avoid the confounding variable of serum-induced signaling.
Cells were lysed for 20 minutes on ice with lysis buffer (50 mM Tris-HCl, pH 7.4; 1% Nonidet P-40; 0.25% sodium deoxycholate; 150 mM NaCl; 1 mM EDTA; 1 mM PMSF; 1 μg/mL aprotinin, leupetin, pepstatin; 1 mM Na3VO4; 1 mM NaF) and centrifuged at 14,000g for 10 minutes at 4°C. Samples were resolved by 10% or 12% SDS-PAGE, transferred to nitrocellulose membrane, and blotted with appropriate primary antibodies at a dilution of 1:1,000. Peroxidase-conjugated secondary antibodies (Biosource International, Camarillo, CA) were incubated at a dilution of 1:2,000. Bound antibodies were visualized using chemiluminescent substrate (ECL; Amersham, Arlington Heights, IL) and exposed to Kodak X-OMAT film. Primary antibodies: Goat anti-COX-2 and goat anti-actin were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). Rabbit anti-ACTIVE MAPK, anti-ACTIVE p38, and anti-ACTIVE JNK specific for the phosphorylated forms of p42/p44 MAPK, p38 MAPK, and JNK (c-Jun-N-terminal kinase), respectively, were obtained from Promega (Madison, WI).
Real-Time Polymerase Chain Reaction (PCR).
Total RNA was extracted from the cells using the Trizol Reagent (Invitrogen, Carlsbad, CA). The cDNA template was prepared using oligo-dT random primers and MLV (Moloney Murine Leukemia Virus) reverse transcriptase as previously described in detail.15 After the reverse transcription reaction, the cDNA template was amplified by PCR with Taq polymerase (Invitrogen). COX-2 mRNA was quantitated using real-time PCR technology and the following primers: forward, 5′-TGAAACCCACTCCAAACACA-3′, reverse, 5′-CCCATGGGCATTCAATAAAC-3′. Universal 18S primers (Ambion, Austin, TX) were used as a control for RNA integrity and as a “housekeeping gene.” For quantification, we used real-time PCR (LightCycler, Roche Molecular Biochemicals, Mannheim, Germany) using SYBR green as the fluorophore (Molecular Probes, Eugene, OR). After electrophoresis in 1% agarose gel, the portion of the gel containing the expected PCR product of interest was cut out and the product eluted into Tris-HCl using a DNA elution kit (Gel extraction kit, Qiagen, Valencia, CA). The eluted and purified PCR product was quantitated using a spectrophotometer (Beckman DU 7400) at 260 nm. The inverse linear relationship between copy number and cycle number was then determined. The resulting standard curve was then used to calculate the copy number per ml in experimental sample.
Reporter Gene Assay.
The generation of the reporter gene constructs for COX-2 has been previously described.16, 17 Cells were cotransfected with 20 ng of TK-Renilla-CMV and 1 μg of phPES2(-1432/+59) or phPES2(-327/+59). Twenty-four hours after the transfection, cells were incubated with oxysterol or media (control) for 4 hours. Both firefly and Renilla luciferase activities were quantitated using the dual luciferase reporter assay system (Promega) according to the manufacturer's instructions and data were expressed as the ratio of firefly/Renilla luciferase activity.
Materials and Reagents.
22(R)-hydroxycholesterol (22-HC), 3,5,6-cholestanetriol, docosahexaenoic acid, all trans-retinoic acid, actinomycin D, and cycloheximide were obtained from Sigma Chemicals (St. Louis, MO), and TO-901317 was from Cayman Chemical (Ann Arbor, MI). The MAPK inhibitors, U0126 for MAP or extracellular signal-regulated kinase and SB203580 for p38 MAPK, were from Calbiochem (La Jolla, CA). Recombinant human TNF-α was from R&D Systems (Minneapolis, MN).
Do Oxysterols Induce COX-2 Expression?
Treatment of KMBC cells with 22-HC, a common oxysterol present in bile, increased COX-2 cellular protein levels in a concentration-dependent manner (Fig. 1A). This enhanced COX-2 protein expression was augmented when the cells were simultaneously treated with TNF-α, a proinflammatory cytokine (Fig. 1A). When cells were treated with another oxysterol, 3,5,6-cholestanetriol, which is also present in hepatic bile, a similar induction of COX-2 protein was identified (Fig. 1B). These observations demonstrate that oxysterols induce COX-2 expression in vitro.
Does 22-HC Activate MAPK Signaling Cascades Which Are Responsible for COX-2 Induction?
Because COX-2 induction may be modulated by MAPK signaling cascades, the ability of 22-HC to activate MAPKs in KMBC cells was examined. Activation of p42/44 MAPK, p38 MAPK, and JNK was assessed by immunoblot analysis using phosphorylation-specific antisera. Both p42/44 MAPK and p38 MAPK were phosphorylated following treatment of the cells with 22-HC, while JNK phosphorylation was not enhanced (Fig. 2). To explore the role of p42/44 and p38 MAPK in COX-2 induction, cells were next treated with 22-HC either in the presence or absence of selective MAPK pharmacologic inhibitors. The inhibitor, U0126 for p42/44 MAPK, did not block COX-2 induction, while SB203580 for p38 MAPK effectively attenuated COX-2 induction (Fig. 3). Therefore, these data suggest that 22-HC activation of p38 MAPK leads to COX-2 induction.
Does 22-HC Alter COX-2 mRNA Expression or Stabilization?
COX-2 mRNA levels were increased 1.5-fold by 22-HC treatment (Fig. 4), a value similar to the 2-fold increase observed with TNF-α used as a positive control. Since oxysterols can bind the LXR nuclear receptor and thereby transcriptionally induce a variety of genes, we next examined whether agonists for retinoid X receptor, a binding partner of LXR, or an LXR agonist can induce COX-2. The maximal doses reported to induce LXR/RXR-dependent responses were selected and employed in this study.18 However, neither docosahexaenoic acid or all trans-retinoic acid, ligands for the retinoid X receptor (Fig. 5A), nor TO-901317, a potent and selective agonist for LXR (Fig. 5B), induced COX-2 expression. Consistent with these data, 22-HC did not increase COX-2 promoter activity in a reporter gene assay (Fig. 6). Taken together, these data suggest 22-HC does not enhance transcription of COX-2.
Because mRNA stabilization will also alter COX-2 mRNA levels, the effect of 22-HC on COX-2 mRNA half-life was next assessed by quantitating COX-2 mRNA over time in cells treated with the transcriptional inhibitor, actinomycin D. As shown in Figure 7, COX-2 mRNA stability was increased by oxysterol treatment, and this was inhibited by the p38 MAPK inhibitor. Taken together, these data suggest that oxysterol stabilizes COX-2 mRNA via p38 MAPK activation, resulting in COX-2 protein accumulation.
Does 22-HC Alter COX-2 Protein Stability?
To examine if the stability of COX-2 protein is also affected by oxysterol, we next performed protein stability analysis. COX-2 protein levels decreased following cycloheximide treatment (Fig. 8A). However, 22-HC treatment did not increase COX-2 protein stability irrespective of the pretreatment of p38 MAPK inhibitor (Fig. 8B). Therefore, COX-2 protein stability is likely not affected by oxysterol.
The principal findings of this study relate to the effect of oxysterols on COX-2 expression in a human cholangiocarcinoma cell line. The results demonstrate that the oxysterol 22-HC stabilizes COX-2 mRNA via a p38 MAPK-dependent mechanism resulting in COX-2 protein accumulation in cholangiocarcinoma cells. These results provide new information regarding the growth-modifying potential of oxysterols in cholangiocytes and cholangiocarcinoma. In particular, this enhanced COX-2 protein expression by oxysterol may participate in the genesis and progression of cholangiocarcinomas. Each of these findings will be discussed below.
COX-2 may participate in carcinogenesis by promoting cell replication and angiogenesis and dysregulating apoptosis. In biliary epithelial cells, COX-2 can be induced by a variety of proinflammatory cytokines and has been implicated in cholangiocarcinogenesis.19 Indeed, we have recently demonstrated that bile acids induce COX-2 expression in a human cholangiocarcinoma cell line.20 In the current study, oxysterols also induced COX-2 protein expression in a malignant human cholangiocyte cell line. This finding is, to our knowledge, the first demonstration of COX-2 induction by oxysterols. In addition to 22-HC, we also treated cells with a different oxysterol, 3,5,6-cholestanetriol, and similarly identified COX-2 induction. Because these oxysterols are present in hepatic bile, these observations likely have in vivo relevance. Moreover, the oxysterol concentrations used in the current study are relevant to pathophysiologic disease states, since the concentrations of oxysterols are between 10–30 μM in inflammatory conditions of the biliary tract.21 Collectively, these findings implicate that oxysterols may also participate in cholangiocarcinogenesis through the induction of COX-2.
COX-2 expression can be regulated by MAPK signaling pathways. In this respect, 22-HC activated p42/44 and p38 MAPK, but not JNK 1/2. However, only the p38 MAPK inhibitor, SB203580, attenuated COX-2 induction, suggesting that this MAPK was specifically associated with 22-HC enhancement of cellular COX-2 protein levels. Given that SB203580 also inhibits the PI3K/AKT pathway, it is possible that AKT could also potentially contribute to COX-2 induction. However, 22-HC does not activate AKT in these cells (data not shown). Collectively, these data suggest that oxysterol activation of p38 MAPK induces COX-2 expression.
Maintenance of cellular protein levels is complex and includes the interplay of transcriptional, translational, and posttranslational regulatory processes. We further explored the potential mechanisms of oxysterol-mediated COX-2 induction. Since oxysterols can bind the LXR nuclear receptor and thereby transcriptionally induce a variety of genes,11 we next examined whether oxysterols increase COX-2 mRNA levels. As assessed by real-time PCR, oxysterol increased COX-2 mRNA levels. However, neither ligands for the retinoid X receptor, a binding partner of LXR, nor a selective agonist for LXR induced COX-2 expression. Moreover, oxysterol did not increase COX-2 promoter activity in a reporter gene assay. This lack of COX-2 transcriptional induction by oxysterol is also consistent with the present observations linking MAPK activation and COX-2 induction. COX-2 expression can be regulated by MAPKs via transcriptional processes.22 Activated p42/44 MAPK can directly phosphorylate transcription factors, such as Elk-1 and Sap-1, which then bind to the serum-responsive element of the c-fos promoter. This increases the expression of c-fos and c-jun and the binding activity of AP-1 (a transcription factor consisting of c-fos and c-jun protein complexes).23, 24 The JNK signal transduction cascade also results in transcriptionally active AP1 complexes.25 The AP-1 complexes transactivate the COX-2 promoter, increasing expression of COX-2 mRNA and protein.22 In the present study, p42/44 MAPK was activated following oxysterol treatment, while oxysterol did not activate JNK, and the inhibition of p42/44 MAPK did not block COX-2 induction. These data collectively implicated a posttranscriptional process as an explanation for oxysterol-mediated COX-2 induction. Indeed, COX-2 mRNA stability was increased by oxysterol treatment and this was inhibited by the p38 MAPK inhibitor. Our findings are quite similar to the other observation that p38 MAPK enhances COX-2 protein expression by stabilizing COX-2 mRNA in lipopolysaccharide-treated human monocytes.26 The 3′ untranslated region of COX-2 has been reported to contain AU-rich motifs important for its mRNA stability.27 Additionally, sequences that determine the stability of an mRNA may also reside in the 5′ untranslated region or the coding region in addition to the 3′ untranslated region.28 Therefore, it is plausible that oxysterol-induced COX-2 mRNA stabilization depends on pathways linking p38 MAPK to alterations of binding proteins for these mRNA regions.
Collectively, our current study demonstrates that oxysterols, which are present in bile during disease states, stabilize COX-2 mRNA via p38 MAPK activation, resulting in COX-2 protein accumulation in human cholangiocarcinoma cells. Since COX-2 induction has been compellingly shown to promote cellular replication and inhibit apoptosis, this enhanced COX-2 protein expression by oxysterol likely participates in the genesis and progression of cholangiocarcinomas. These studies provide further links between inflammatory diseases of the biliary tract and bile duct carcinogenesis. Inhibition of oxysterol generation, utilization of a p38 MAPK inhibitor, which are now in clinical trials, and/or COX-2 inhibitors may ultimately prove useful in chemoprevention or treatment of bile duct cancers.