Cyclooxygenase-2 (COX-2) mediates various inflammatory responses and is expressed in pancreatic tissue from patients with chronic pancreatitis. To examine the role of COX-2 in chronic pancreatitis, we investigated its participation in regulating functions of pancreatic stellate cells (PSCs), using isolated rat PSCs. COX-2 was expressed in culture-activated PSCs but not in freshly isolated quiescent PSCs. TGF-β1, IL-1β, and IL-6 enhanced COX-2 expression in activated PSCs, concomitantly increasing the expression of α-smooth muscle actin (α-SMA), a parameter of PSC activation. The COX-2 inhibitor NS-398 blocked culture activation of freshly isolated quiescent PSCs. NS-398 also inhibited the enhancement of α-SMA expression by TGF-β1, IL-1β, and IL-6 in activated PSCs. These data indicate that COX-2 is required for the initiation and promotion of PSC activation. We further investigated the mechanism by which cytokines enhance COX-2 expression in PSCs. Adenovirus-mediated expression of dominant negative Smad2/3 inhibited the increase in expression of COX-2, α-SMA, and collagen-1 mediated by TGF-β1 in activated PSCs. Moreover, dominant negative Smad2/3 expression attenuated the expression of COX-2 and α-SMA enhanced by IL-1β and IL-6. Anti-TGF-β neutralizing antibody also attenuated the increase in COX-2 and α-SMA expression caused by IL-1β and IL-6. IL-6 as well as IL-1β enhanced TGF-β1 secretion from PSCs. These data indicate that Smad2/3-dependent pathway plays a central role in COX-2 induction by TGF-β1, IL-1β, and IL-6. Furthermore, IL-1β and IL-6 promote PSC activation by enhancing COX-2 expression indirectly through Smad2/3-dependent pathway by increasing TGF-β1 secretion from PSCs.
- transforming growth factor-β
- pancreatic fibrosis
cyclooxygenase (COX) is the rate-limiting enzyme that catalyzes arachidonic acid conversion to prostaglandins. COX exists as two isoforms, COX-1 and COX-2. COX-1 is constitutively expressed in most types of cells and exerts such diverse physiological functions as gastric mucosal protection and platelet aggregation (9). In contrast, COX-2 is an inducible isoform in response to various proinflammatory and mitogenic stimuli. COX-2 induction leads to the increase in prostaglandin production that is responsible for the progression of inflammation. Recently, it has been shown that COX-2 is expressed in pancreatic tissue from patients with chronic pancreatitis (16, 26). The inflammation in the pancreatic tissue of such patients is caused by repeated acute pancreatic injury, mainly due to excessive alcohol intake (6). Although COX-2 expression in pancreatic tissue from patients with chronic pancreatitis suggests the participation of COX-2 in chronic pancreatic inflammation and fibrosis, the precise mechanism is unknown.
Pancreatic stellate cells (PSCs) have recently been identified, isolated, and characterized (2, 5). In the normal pancreas, PSCs possess fat droplets containing vitamin A and are quiescent (2). When cultured in vitro, PSCs are autoactivated (autotransformed), changing their morphological and functional features (5). PSCs commence losing vitamin A-containing lipid droplets, proliferate, express α-smooth muscle actin (α-SMA), and produce and secrete extracellular matrix components such as collagen and fibronectin. Thus PSCs are autotransformed to myofibroblast-like cells. In vivo, PSCs also are activated during pancreatic fibrosis in humans and animal models (12). Therefore, PSCs are thought to play an important role in pancreatic fibrogenesis in patients with chronic pancreatitis.
PSCs respond to various proinflammatory and profibrogenic cytokines including TGF-β1, IL-1β, and IL-6 (3, 22). Of these, a major profibrogenic cytokine, TGF-β1, regulates multiple functions of PSCs. For example, TGF-β1 1) stimulates extracellular matrix synthesis, 2) promotes PSC activation and increases α-SMA expression, 3) attenuates proliferation in an autocrine manner, and 4) reduces matrix metalloproteinase-3 and -9 expression (28). Furthermore, TGF-β1 expression is upregulated in chronic pancreatitis tissues (12). Thus TGF-β1 has been implicated in the etiology of pancreatic fibrosis.
TGF-β1 intracellular signaling is mediated and modulated primarily by Sma- and Mad-related proteins (Smads) (13, 21). Upon TGF-β1 binding to its receptor, Smad2 and Smad3 are phosphorylated by the receptor and form oligomeric complexes with Smad4; the complexes then translocate into the nucleus. These complexes subsequently activate the transcription of target genes. In addition to the Smad-dependent pathway, there exist Smad-independent TGF-β signaling pathways, for example, mitogen-activated protein kinases (MAPKs) such as extracellular signal-regulated kinase (4). Thus TGF-β1 intracellular signaling occurs via Smad-dependent and Smad-independent pathways.
In this study, we examined the role of COX-2 in the regulation of PSC functions and found that COX-2 plays a pivotal role in the functional regulation of PSCs by multiple cytokines. We further showed that Smad2/3-dependent TGF-β1 signaling is critical for the cytokine-mediated induction of COX-2 expression.
MATERIALS AND METHODS
TGF-β1, Nycodenz, pronase, and anti-α-SMA antibody were purchased from Sigma (St. Louis, MO). IL-1β, IL-6, and anti-TGF-β1 neutralizing antibody were obtained from R&D (Abrington, UK). DNase I was obtained from Roche (Basel, Switzerland). Collagenase P was obtained from Boehringer Mannheim (Mannheim, Germany). SB 203580, SP600125, and PD-98059 were obtained from Calbiochem (La Jolla, CA). NS-398, anti-COX-1, and anti-COX-2 antibodies were obtained from Cayman Chemical (Ann Arbor, MI). Anti-collagen 1 antibody was obtained from Rockland Immunochemicals (Gilbertsville, PA). Horseradish peroxidase (HRP)-conjugated donkey anti-goat IgG, HRP-conjugated donkey anti-mouse IgG, HRP-conjugated donkey anti-rabbit IgG, FITC-conjugated donkey anti-mouse IgG, and Cy3-conjugated donkey anti-rabbit IgG antibodies were obtained from Jackson ImmunoResearch (West Grove, PA).
Isolation and culture of rat pancreatic stellate cells.
Rat PSCs were prepared as described previously (2). Briefly, rat pancreas was digested in Grey’s balanced salt solution supplemented with 0.05% collagenase P, 0.02% pronase, and 0.1% DNase. After filtration through nylon mesh, cells were centrifuged in a 13.2% Nycodenz gradient at 1,400 g for 20 min. PSCs in the band just above the interface of the Nycodenz solution and the aqueous layer were collected, washed, and resuspended in Iscove’s modified Dulbecco’s medium containing 10% fetal calf serum, 100 U/ml penicillin, and 100 μg/ml streptomycin. PSCs were cultured in a 5% CO2 atmosphere at 37°C. The purity of isolated cells was confirmed to be >90% by observing their cytoplasmic droplets with vitamin A autofluorescence. We used PSCs between passages 2 and 3 as culture-activated PSCs in experiments, unless otherwise indicated.
Cells were fixed with 2% formaldehyde in PBS, treated with Triton-X in PBS for 5 min, and incubated sequentially with blocking ACE (Snow Brand Milk Products, Tokyo, Japan), primary antibodies, and secondary antibodies. For double staining using anti-COX-2 and anti-α-SMA antibodies, Cy3-conjugated donkey anti-rabbit IgG and FITC-conjugated donkey anti-mouse IgG antibodies were used as secondary antibodies, respectively. Samples were examined under an Olympus BX51 microscope (Tokyo, Japan). Images were digitized and then processed using Photoshop 5.0 software (Adobe System, Mountain View, CA).
Western blotting was performed as described previously (23). Briefly, 10 μg of protein from whole cell lysates were separated on 10% SDS minigels and transferred to nitrocellulose membranes. Membranes were then probed with primary antibodies, followed by detection with peroxidase-conjugated secondary antibodies and visualization by enhanced chemiluminescence. Western blotting using anti-α-tubulin antibody was carried out as an internal control.
Western blot images were captured and digitized with a Fluor-S Multi-Imager (Bio-Rad, Hercules, CA). Densitometric evaluation of Western blot data was then carried out using Quantity One software (Bio-Rad).
Evaluation of prostaglandin E2 production.
Production of prostaglandin E2 by PSCs was examined by determining the culture medium concentration of prostaglandin E2 released from PSCs, using a commercial enzyme immunoassay kit (Cayman Chemical, Ann Arbor, MI) according to the manufacturer’s instructions.
Recombinant adenoviral vector for dominant negative Smad2/3 (AdDNSmad2/3) was kindly provided by Dr. K. Miyazono (University of Tokyo, Japan). Cells were infected with a recombinant adenoviral vector at a dose of 10 plaque-forming units per cell in the culture medium described above. An adenoviral vector expressing β-galactosidase (AdLacZ) was used as an infection control.
Measurement of TGF-β1 peptide secretion.
TGF-β1 peptide secretion was measured by determining its concentration in the culture medium with a commercial ELISA kit (DRG International, Mountainside, NJ) according to the manufacturer’s instructions.
COX-2 expression in PSCs.
In our initial attempt to examine the role of COX-2 in the regulation of PSC functions, we examined the expression of COX-2 in both quiescent and culture-activated PSCs. Freshly isolated PSCs were used for experiments 12 h after isolation. Immunocytochemistry using anti-COX-2 antibody showed that COX-2 was not expressed in freshly isolated quiescent PSCs as determined by the presence of lipid droplets (Fig. 1A, a and b). In contrast, COX-2 was expressed in culture-activated PSCs as determined by their absence of droplets, fibroblast-like shape, and α-SMA expression, a parameter of PSC activation (5) (Fig. 1A, c–e). In addition, Western blotting revealed that freshly isolated PSCs did not express either COX-2 or α-SMA (Fig. 1B). However, culture-activated PSCs expressed both COX-2 and α-SMA (Fig. 1B). These data suggest that COX-2 expression is induced in PSCs during their activation.
TGF-β1, IL-1β, and IL-6 enhanced the expression and the activity of COX-2 in activated PSCs.
Since TGF-β1 plays a central role in regulating PSC functions such as increasing collagen and α-SMA expression (3), we next examined the effect of TGF-β1 on the expression and activity of COX-2 in activated PSCs. As shown in Fig. 2, A and B, TGF-β1 enhanced COX-2 expression in activated PSCs concurrently with an increase in α-SMA and collagen-1 expression in a dose-dependent manner. In contrast, the expression of COX-1, which occurs constitutively in various types of cells, and α-tubulin was not altered by TGF-β1. Furthermore, TGF-β1 augmented COX-2 activity in PSCs (Fig. 2C). In addition to responding to TGF-β1, PSCs also respond to various proinflammatory cytokines, including IL-1β and IL-6 (22). We therefore examined the effect of these cytokines on COX-2 expression and activity in PSCs. As shown in Fig. 3, both IL-1β and IL-6 increased COX-2 expression (A and B) and COX-2 activity (C) while concomitantly enhancing α-SMA expression (A and B) in activated PSCs. These data suggest the possibility that COX-2 may mediate the regulation of activated PSC functions by TGF-β1, IL-1β, and IL-6.
COX-2 inhibitor NS-398 repressed PSC culture activation.
To examine the role of COX-2 in the regulation of PSC functions, we investigated the effect of NS-398, which specifically inhibits COX-2 activity, on PSC culture activation using freshly isolated PSCs. Since the maximal inhibition of basal COX-2 activity in activated PSCs was observed with 100 μM NS-398 treatment (data not shown), which is consistent with a previous report (8), we utilized 100 μM NS-398 for the current study. As shown in Fig. 4, at 48 h after isolation, a very low-level α-SMA signal was observed in both NS-398-treated and control cells. In control cells, however, α-SMA expression was increased at 72 and 96 h after the isolation. In contrast, α-SMA expression did not increase at 72 h, and its signal in NS-398-treated cells was much smaller than that in control cells even at 96 h. These data indicate that blockade of COX-2 activity by NS-398 inhibited PSC activation, suggesting that COX-2 is necessary for PSC activation.
NS-398 inhibited the responses of activated PSCs to TGF-β1, IL-1β, and IL-6.
We next examined the role of COX-2 in the regulation of PSC functions by cytokines. For this purpose, we first investigated the effect of NS-398 on TGF-β1-mediated enhancement of α-SMA and collagen-1 expression in activated PSCs. As shown in Fig. 5A, TGF-β1 augmented COX-2 activity in PSCs. When PSCs were pretreated with 100 μM NS-394, basal COX-2 activity was decreased and TGF-β1 failed to increase it. Under this condition, moreover, TGF-β1 also failed to increase the expression of α-SMA and collagen-1 (Fig. 5, B and C). These data indicate that COX-2 mediates the stimulatory effect of TGF-β1 on α-SMA and collagen-1 expression in activated PSCs, suggesting that TGF-β1 increases α-SMA and collagen-1 expression by inducing COX-2 expression. We then examined the effect of NS-398 on the enhancement of α-SMA expression by IL-1β and IL-6. As shown in Fig. 6, both IL-1β and IL-6 enhanced COX-2 activity and α-SMA expression in activated PSCs but not when PSCs were pretreated with 100 μM NS-398 (Fig. 6). These data indicate that COX-2 also mediates the stimulatory effect of IL-1β and IL-6 on α-SMA expression in activated PSCs.
Dominant negative Smad2/3 expression blocked the enhancement of COX-2 and α-SMA expression by TGF-β1, IL-1β, and IL-6 in activated PSCs.
Since COX-2 mediates the effects of multiple cytokines on activated PSC functions, we tried to elucidate the mechanism of COX-2 induction by cytokines in activated PSCs. Since TGF-β1 is critical for the regulation of PSC functions, we first examined the signaling pathway through which TGF-β1 increases COX-2 expression. TGF-β signaling occurs by Smad2/3-dependent and Smad2/3-independent pathways (4, 21). To identify the pathway through which TGF-β1 enhances COX-2 expression in PSCs, we examined the effect of dominant negative Smad2/3 expression with an adenovirus vector (AdDNSmad2/3) on COX-2 expression in PSCs. The dominant negative Smad2/3 mutant was generated by substituting Glu for Asp-407 of Smad3, which renders it defective in TGF-β receptor-dependent phosphorylation (11). Nevertheless, this mutant possesses a dominant negative effect on both Smad2 and Smad3 (11). We utilized an adenovirus vector expressing β-galactosidase (AdLacZ) as an infection control. We previously reported that >98% of PSCs are infected with these adenovirus vectors and express the corresponding proteins (24). As shown in Fig. 7, TGF-β1 increased COX-2 expression and consequently enhanced α-SMA and collagen-1 expression in activated PSCs infected with AdLacZ. In contrast, TGF-β1 did not increase COX-2 expression and failed to enhance α-SMA and collagen-1 expression in cells infected with AdDNSmad2/3. These data indicate that TGF-β1 increases COX-2 expression through a Smad2/3-dependent pathway and that the Smad2/3-dependent pathway is important for TGF-β1 regulation of PSC functions. We further examined the effect of dominant negative Smad2/3 expression on the enhancement of COX-2 expression by IL-1β and IL-6. Both IL-1β and IL-6 increased COX-2 and α-SMA expression in activated PSCs infected with AdLacZ (Fig. 8). Surprisingly, the promoting effect of IL-1β and IL-6 on COX-2 expression was inhibited by infection with AdDNSmad2/3. Moreover, IL-1β and IL-6 failed to increase α-SMA expression in cells infected with AdDNSmad2/3. These data suggest that IL-1β and IL-6 increase COX-2 expression, at least in part, through a Smad2/3-dependent pathway and that COX-2 is critical for the enhancement of α-SMA expression in PSCs by IL-1β and IL-6.
IL-6 increased TGF-β1 expression and secretion by PSCs.
Although the current results suggest the participation of a Smad2/3-dependent pathway in the expression of COX-2 and α-SMA induced by IL-1β and IL-6, it is unlikely that the Smad2/3-dependent pathway directly mediates the stimulatory effect of IL-1β and IL-6 in PSCs, since Smad proteins are intracellular signaling molecules specific to the TGF-β family. In the regulation of activated PSC functions, TGF-β1 acts in an autocrine manner (17, 28). We recently reported that IL-1β enhances TGF-β1 expression and secretion by activated PSCs (1). Therefore, we hypothesized that IL-1β and IL-6 may increase COX-2 expression through an Smad2/3-dependent pathway by increasing TGF-β1 secretion from PSCs. To test this hypothesis, we first investigated whether IL-6 enhances TGF-β1 secretion from activated PSCs. As shown in Fig. 9, IL-6 increased TGF-β1 peptide secretion by activated PSCs in a dose-dependent manner. These data reinforce the idea that IL-6, as well as IL-1β, enhances COX-2 and α-SMA expression in PSCs by increasing autocrine TGF-β1.
Anti-TGF-β neutralizing antibody blocked the enhancement of COX-2 and α-SMA expression by IL-1β and IL-6.
To further examine the involvement of autocrine TGF-β1 in the enhancement of COX-2 and α-SMA expression by IL-1β and 1L-6, we investigated the effect of anti-TGF-β neutralizing antibody on COX-2 and α-SMA expression in PSCs. When anti-TGF-β neutralizing antibody was added in the culture medium, the stimulatory effects of IL-1β and IL-6 on the COX-2 expression in activated PSCs were attenuated (Fig. 10). In addition, IL-1β and IL-6 did not enhance α-SMA expression in PSCs in the presence of anti-TGF-β neutralizing antibody. These data indicate that IL-1β and IL-6 enhance COX-2 expression in PSCs, at least in part, through autocrine TGF-β1, and that autocrine TGF-β1 plays a pivotal role in the enhancement of α-SMA expression by IL-1β and IL-6.
MAPK inhibitors did not block TGF-β1 enhancement of COX-2 expression in activated PSCs.
Our present data suggest that TGF-β1 plays a central role in cytokine-induced COX-2 expression through Smad2/3-dependent pathway. However, TGF-β1 intracellular signaling is mediated by MAPKs as well. In addition, in other types of cells, MAPKs function in COX-2 expression (7, 9). Therefore, we finally examined the participation of MAPKs in TGF-β1-induced COX-2 expression in activated PSCs. For this purpose, we pretreated activated PSCs with SP600125 (10 μM, JNK inhibitor), PD-98059 (10 nM, ERK inhibitor), or SB 203580 (25 μM, p38MAPK inhibitor). It has been previously reported that these dosages of inhibitors successfully block each MAPK-dependent pathway in rat activated PSCs (19, 20, 25). As shown in Fig. 11, TGF-β1 increased COX-2 expression in cells pretreated with SP600125, PD-98059, or SB 203580. These data reinforce the evidence that TGF-β1 enhances COX-2 expression in activated PSCs mainly through Smad2/3-dependent pathway.
In this study, we demonstrated that 1) COX-2 is important in both PSC activation and the cytokine regulation of activated PSC functions, 2) TGF-β1 induces COX-2 expression through a Smad2/3-dependent pathway, and 3) IL-1β and IL-6 indirectly increase COX-2 and α-SMA expression in activated PSCs, at least partly, through a Smad2/3-dependent pathway by enhancing autocrine TGF-β1.
COX-2 plays a pivotal role in the progression of acute and chronic inflammatory responses in various tissues and organs. Although two groups have shown that COX-2 is expressed in pancreatic tissue from patients with chronic pancreatitis (16, 26), the function of COX-2 in chronic pancreatitis is uncertain. Since PSCs are activated during chronic pancreatitis and promote pancreatic fibrosis by producing and secreting multiple extracellular matrix components, our current observation that COX-2 is expressed in activated PSCs but not in quiescent ones strongly suggests COX-2 participation in the progression of chronic pancreatitis, especially in pancreatic fibrosis. Moreover, the fact that profibrogenic cytokine TGF-β1 enhanced COX-2 expression and the COX-2 inhibitor NS-398 blocked TGF-β1 enhancement of α-SMA and collagen-1 expression implies a role for COX-2 in promoting pancreatic fibrosis.
To date, studies on the regulation of COX-2 expression have shown the involvement of multiple intracellular signaling mediators. For instance, nuclear factor-κB, a transcription factor that plays a central role in inflammatory and immune responses, is involved in COX-2 induction by lipopolysaccharide in macrophages and colonic epithelial cells (9, 10, 15). In addition, MAPKs, including ERK, JNK, and p38MAPK also play a role in COX-2 expression (9). The signaling pathway through which TGF-β1 induces COX-2 expression also has been elucidated; the COX-2 gene promoter region contains a TGF-β response element (27, 29). Sheares et al. (27) reported that TGF-β1 induces COX-2 expression through p38MAPK in human pulmonary artery muscle cells. Indeed, MAPK is one mediator of TGF-β1 signaling (4). However, our present study showed that MAPK inhibitors did not inhibit TGF-β1-induced COX-2 or α-SMA expression in activated PSCs (Fig. 11). Masamune et al.(19) previously reported that p38MAPK inhibitor SB 203580 inhibited α-SMA expression in PSCs during the culture activation. Consisting with their data, basal α-SMA expression in SB 203580-treated cells was relatively lower than in cells with other treatment (Fig. 11B). However, TGF-β1 successfully increased both COX-2 and α-SMA expression in PSCs treated with SB 203580. Therefore, we suggest that TGF-β1 enhances COX-2 and α-SMA expression in activated PSCs through a pathway independent of p38MAPK.
Intriguing are our present data indicating that both dominant negative Smad2/3 expression and anti-TGF-β neutralizing antibody attenuated the enhancement of the COX-2 expression by IL-1β and IL-6. These data imply that IL-1β and IL-6 indirectly increase COX-2 expression in PSCs through autocrine TGF-β1. We further reinforced this idea by showing that IL-6 as well as IL-1β (1) augmented TGF-β1 secretion by PSCs (Fig. 9). However, we do not rule out a direct effect of IL-1β and IL-6 on COX-2 expression in PSCs. It is well known that IL-1β directly modulates PSC functions through its specific intracellular signaling pathway (18). In the present study, dominant negative Smad2/3 expression did not completely inhibit IL-1β- or IL-6-induced COX-2 expression (Fig. 8). In addition, although anti-TGF-β neutralizing antibody abolished IL-6-mediated enhancement of COX-2 expression in PSCs, it partially attenuated the increase of COX-2 expression mediated by IL-1β (Fig. 10). Thus IL-1β and IL-6 could directly enhance COX-2 expression in PSCs. However, we propose that IL-1β and IL-6 indirectly enhance COX-2 expression in PSCs, at least partly, through an Smad2/3-dependent pathway by increasing autocrine TGF-β1.
COX-2 participation in gastrointestinal fibrosis has been intensively studied in the liver. Consistent with our current data, for example, studies by Cheng et al. (8) using an immortalized human hepatic stellate cell (HSC) line showed that COX-2 inhibitor NS-398 attenuated α-SMA expression in HSCs, indicating a role for COX-2 in promoting HSC activation. On the other hand, Hui et al. (14), using another immortalized human HSC line, showed that NS-398 increased basal and TGF-β1-stimulated collagen-1 expression and that COX-2-derived prostaglandin E2 inhibited both basal and TGF-β1-stimulated collagen-1 expression, suggesting an inhibitory role for COX-2 on activated HSC function (14). In the present study, we examined COX-2 participation concurrently in basal and TGF-β1-stimulated expression of α-SMA and collagen-1. We demonstrated that COX-2 promotes both α-SMA and collagen-1 expression. However, there exists an apparent discrepancy between our data and that of Hui et al. (14). One possible explanation for the discrepancy is that the role of COX-2 in PSCs may be distinct from that in HSCs. Another possibility is that the COX-2 function might be variable in different immortalized HSC lines, since Hui et al. (14) and Cheng et al. (8) utilized different ones. We have extended the knowledge obtained from studies utilizing HSCs by elucidating the mechanisms of COX-2 induction in PSCs through the demonstration that TGF-β1/Smad signaling is critical for COX-2 expression in PSCs.
It is assumed that PSCs are activated by various cytokines in chronic pancreatitis (3, 22) and that activated PSCs promote pancreatic fibrosis by producing and secreting various extracellular matrix components. Thus a therapeutic strategy that inhibits PSC activation or increases PSC quiescence could be useful for the treatment of pancreatic fibrosis. We showed that COX-2 inhibitor NS-398 blocked the expression of α-SMA in freshly isolated quiescent PSCs during primary culture (Fig. 4). In addition, densitometry analyses of Western blotting results showed that the enhancement of COX-2 expression by TGF-β1, IL-1β, and IL-6 preceded or accompanied the increase in α-SMA expression (Figs. 2 and 3). Furthermore, NS-398 blocked the enhancement of α-SMA expression by TGF-β1, IL-1β, and IL-6 in activated PSCs (Figs. 5 and 6). These data suggest that COX-2 is essential for the increase of α-SMA expression mediated by these cytokines. Since α-SMA expression is a parameter of PSC transformation from quiescent to activated myofibrobast-like cells, these observations indicate that COX-2 inhibitor blocked the initiation of PSC transformation and attenuated its promotion by these cytokines. These data suggest that COX-2 inhibitors could be effective in the therapy of pancreatic fibrosis. Further in vivo studies on the effect of COX-2 inhibitors on pancreatic fibrosis are warranted.
In conclusion, we have shown that COX-2 plays a pivotal role in the response of PSCs to multiple cytokines and that TGF-β1/Smad signaling is critical for COX-2 expression. These data provide insights for understanding the mechanism of pancreatic fibrosis and for developing novel therapeutic strategies for its treatment.
This work was supported by grants-in-aid from the Ministry of Education, Culture, Sports, Science and Technology of Japan.
We are grateful to Dr. Kohei Miyazono (University of Tokyo) for dominant negative Smad2/3 adenovirus vector.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 the American Physiological Society