Interleukin (IL)-1β induces cyclooxygenase (COX)-2 expression and prostanoid formation in cultured human airway smooth muscle (HASM) cells. In other cell types, IL-6 family cytokines induce COX-2 or augment IL-1β-induced COX-2 expression. The purpose of this study was to determine whether IL-6 family cytokines were involved in COX-2 expression in HASM cells. RT-PCR was used to demonstrate that the necessary receptor components for IL-6-type cytokine binding are expressed in HASM cells. IL-6 and oncostatin M (OSM) each caused a dose-dependent phosphorylation of signal transducer and activator of transcription-3, whereas IL-11 did not. IL-6, IL-11, and OSM alone had no effect on COX-2 expression. However, OSM caused dose-dependent augmentation of COX-2 expression and prostaglandin (PG) E2 release induced by IL-1β. In contrast, IL-6 and IL-11 did not alter IL-1β-induced COX-2 expression. IL-6 did increase IL-1β-induced PGE2formation in unstimulated cells but not in cells stimulated with arachidonic acid (AA; 10−5 M), suggesting that IL-6 effects were mediated at the level of AA release. Our results indicate that IL-6 and OSM are capable of inducing signaling in HASM cells. In addition, OSM and IL-1β synergistically cause COX-2 expression and PGE2 release.
- signal transducer and activator of transcription
cyclooxygenase (COX) is an enzyme that catalyzes the conversion of arachidonic acid (AA) to prostaglandins (PG) and thromboxane. There are two isoforms of the enzyme. COX-1 is constitutively expressed in most mammalian cells, whereas COX-2 is induced by certain cytokines, lipopolysaccharide, and mitogens depending on the cell type (8, 18, 19, 34). Interleukin (IL)-1β and tumor necrosis factor-α are among the cytokines with the most potent effects on COX-2 expression. In human airway smooth muscle (HASM) cells, IL-1β at concentrations as low as 0.2 ng/ml induces COX-2 expression and PGE2 release. In these cells, the induction of COX-2 by IL-1β has important functional consequences, including a decrease in β2-adrenergic receptor responsiveness (27).
In some cell types, members of the IL-6 family of cytokines, which includes IL-6, IL-11, and oncostatin M (OSM), increase COX-2 expression. For example, IL-6 alone and in conjunction with IL-1β has been demonstrated to augment COX-2 expression and PGE2production in mouse osteoblasts (40). IL-6 also has been shown to promote PGE2 release in a canine basilar artery model (32). OSM induces COX-2 expression both alone and synergistically with IL-1β in human vascular smooth muscle (9). Because IL-1 has been shown to induce large quantities of IL-6 family cytokines in HASM cells (11) and other human cell types (35) and because members of the IL-6 family are capable of inducing COX-2 in other cell types, we hypothesized that these cytokines might in part mediate the effect of IL-1β on COX-2 expression in HASM.
Members of the IL-6 cytokine family share a common receptor subunit, glycoprotein (gp)-130. IL-6 signaling occurs through binding of the cytokine to an IL-6 receptor (IL-6R) coupled to two gp130 subunits. Signaling through the IL-11 receptor (IL-11R) occurs in a similar manner, with IL-11 binding to the IL-11R coupled to two gp130 subunits. OSM acts through an OSM receptor coupled to a single gp130 subunit. Janus kinases associated with the gp130 component of each receptor type then become phosphorylated and in turn phosphorylate the gp130 subunit. The phosphorylated gp130 receptor is able to bind signal transducer and activator of transcription (STAT), particularly STAT3, through SH2 domains on STAT3. Once bound to the receptor, STAT3 itself is phosphorylated. The phosphorylated STAT3 is then released from the receptor, dimerizes with other phosphorylated STAT3 molecules, and translocates to the nucleus, where it activates target genes such as α2-macroglobulin in the rat and α1-antichymotrypsin in humans (7, 16, 22, 29,38). IL-6 also induces extracellular signal-regulated kinase (ERK) and p38 mitogen-associated protein kinase (MAPK) activation in some cell types (39). There is also evidence that ERK may phosphorylate STAT3 near its COOH terminus in some cell types, enhancing its activity (23).
To examine the hypothesis that IL-1β-induced formation of IL-6 family cytokines might, in part, mediate the effects of IL-1β on COX-2 formation and PGE2 production, we first determined whether IL-6, IL-11, and OSM were capable of acting on HASM cells by examining the expression of IL-6 family receptors. We also measured the phosphorylation of STAT3 in response to IL-6, IL-11, and OSM to ensure that these cytokines did indeed signal in these cells. Finally, we measured COX-2 expression and PGE2 production in HASM cells in response to IL-6, IL-11, and OSM in the presence and absence of IL-1β. Our results indicate that receptors for IL-6 family cytokines are present on HASM cells and that ligation of these receptors induces STAT3 activation. In addition OSM, but not IL-6 or IL-11, synergizes with IL-1β to augment COX-2 expression.
We have previously reported that activation of ERK by IL-1β is required for COX-2 formation and PGE2 release (25,26). Because our results indicated that OSM also induced ERK, we hypothesized that ERK might mediate the ability of OSM to augment IL-1β-induced COX-2 expression. To address this hypothesis, we measured ERK phosphorylation by OSM alone and in combination with IL-1β. We also examined the effect of U-0126, an inhibitor of MAPK/ERK (MEK) (12), the enzyme upstream of ERK, on the ability of OSM to augment IL-1β-induced COX-2 expression.
HASM cells were obtained from lung transplant donor tracheae in accordance with procedures approved by the University of Pennsylvania (Philadelphia) Committee on Studies Involving Human Beings. A segment of trachea just proximal to the carina was dissected under sterile conditions. The trachealis muscle was isolated, and the tissue was prepared as previously described (33). For culture, the cells were plated in plastic flasks or six-well plates at 104 cells/cm2 in Ham's F-12 medium supplemented with 10% fetal bovine serum, 100 U/ml of penicillin, 0.1 mg/ml of streptomycin, 2.5 mg/ml of amphotericin B, 12 mM NaOH, 1.6 μM CaCl2, 2 mM l-glutamine, and 25 mM HEPES. The medium was replaced every 3–4 days. The cells were passaged with 0.25% trypsin and 1 mM EDTA every 10–14 days. Confluent cells were serum deprived and supplemented with 5.7 μg/ml of insulin and 5 μg/ml of transferrin 24 h before use. Cells inpassages 4–7 from 12 different donors were used in the studies described below.
Western blotting for measurement of phosphorylated STAT3, phosphorylated ERK, phosphorylated p38 levels, and COX-2 expression.
HASM cells were grown to confluence in six-well plates and serum deprived for 24 h as described above. For measurement of phosphorylated STAT3, ERK, and p38, cells were treated with IL-6, IL-11, or OSM (20 ng/ml) for 5, 10, 15, 30, and 60 min. We also examined the effect of IL-6 and OSM at varying doses from 0.2 to 50 ng/ml for 15 min. The effect of IL-6 and OSM on IL-1β-induced ERK activation was measured using 20 ng/ml of IL-6 or OSM in the presence and absence of IL-1β (2 ng/ml). The medium was then removed, and the cells were washed with PBS and lysed with 100 μl of extraction buffer [10 mM Tris · HCl buffer with 50 mM NaCl, 10 mMd-serine, 1 mM EDTA, I mM EGTA, 1% sodium dodecyl sulfate (SDS), 1% Triton X-100, 0.2 mM phenylmethylsulfonyl fluoride, 5 μg/ml of leupeptin, 1 μg/ml of pepstatin, and 10−2U/ml of aprotinin]. The cells were scraped from the plates, passed through a 25 -gauge needle, and solubilized by 10 s of sonication. To examine the effect of IL-6 family cytokines alone or in conjunction with IL-1β on COX-2 expression, cells were treated with IL-6 (20 ng/ml), IL-11 (20 ng/ml), or OSM (at 0.5, 2, and 20 ng/ml) in the presence and absence of IL-1β at 0.2 or 2 ng/ml. The effect of OSM (20 ng/ml) on IL-1β (2 ng/ml)-induced COX-2 expression was also examined after 2 h of pretreatment with either the MEK inhibitor U-0126 (10 μM; Promega, Madison, WI) or vehicle (DMSO 0.01%). We have previously reported that U-0126 causes a marked inhibition of IL-1β-induced ERK phosphorylation (24), indicating its efficacy. Cytokines were added simultaneously and protein was extracted 24 h later.
For STAT3, ERK (p42/p44), p38, and COX-2 Western blots, supernatants of cell lysates were mixed with equal volumes of loading buffer [0.062 M Tris · HCl (pH 6.8), 10% glycerol, 2% SDS, 5% 2-mercaptoethanol, and 0.01% (wt/vol) bromphenol blue] and boiled for 5 min. Solubilized proteins (30 μg/lane) were separated by SDS-polyacrylamide gel electrophoresis on 12% Tris-glycine gel (Invitrogen, Carlsbad, CA) under nonreducing conditions and transferred electrophoretically to a nitrocellulose membrane in transfer buffer (Pierce, Rockford, IL). For STAT3, ERK, and p38 Western blots, the membrane was blocked with 5% nonfat dry milk in Tris-buffered saline containing 0.1% Tween 20 overnight at 4°C. The blots were probed with rabbit anti-phospho-STAT3, anti-phospho-p44/p42, or anti-phospho-p38 (New England Biolabs, Beverly, MA) for 2 h at room temperature. For COX-2, the blot was probed with rabbit polyclonal antibody to COX-2 (Oxford Biomedical Research, Oxford, MI). The blots were washed and incubated in Tris-buffered saline containing 0.1% Tween 20 and 5% nonfat dry milk with horseradish peroxidase-conjugated goat anti-rabbit IgG for 2 h. The proteins were visualized by light emission on film with enhanced chemiluminscence substrate (Pierce; New England Biolabs). The bands visualized at 89 kDa for phospho-STAT3, 42 and 44 kDa for phospho-p42/p44, 38 kDa for phospho-p38, and 70 kDa for COX-2 were quantified with a laser densitometer. Band density values are expressed as arbitrary optical density units.
To examine the ability of IL-6 family cytokines alone or in combination with IL-1β to induce PGE2 formation, cells were grown to confluence in 24-well plates and serum deprived for 24 h. Wells were either left untreated or were treated with IL-6, IL-11, or OSM (all at 20 ng/ml) alone, IL-1β alone (0.2 and 2.0 ng/ml), or the combination of either IL-6, IL-11, OSM, and IL-1β (both concentrations). After incubation at 37°C for 20 h, the medium was replaced with 0.5 ml of fresh serum-free medium and incubated at 37°C for 15 min in the presence and absence of AA (10−5M) and the supernatant was then harvested. Supernatants were stored at −20°C until assayed with a PGE2 enzyme immunoassay kit (Cayman Chemical, Ann Arbor, MI).
IL-6 and OSM release.
To determine whether IL-1β-induced prostanoids contributed to IL-1β-induced IL-6 and OSM release, HASM cells were grown to confluence in 24-well plates and serum deprived for 24 h. Wells were either left untreated or treated with IL-1β (2 ng/ml) for 20 h. One-half of the IL-1β-treated cells were pretreated with indomethacin (10−6 M) 2 h before addition of IL-1β. Some cells not treated with cytokines were also treated with indomethacin. After incubation for 20 h, the supernatants were harvested and stored at −20°C until assayed with a human IL-6 enzyme immunoassay kit (Cayman Chemical). The assay had 100% specificity for IL-6 and <0.01% specificity for IL-1α and IL-1β per manufacturer's specifications. A human OSM enzyme immunoassay kit (R&D Systems, Minneapolis, MN) was used to assay OSM. The minimum detectable concentration of OSM was less than 6 pg/ml per manufacturer's specifications.
PCR for IL-6R, IL-11R, OSM receptor, and gp130.
HASM cells from two different donors were serum deprived and hormone supplemented for 24 h. Total RNA was isolated using RNeasy spin columns (QIAGEN, Valencia, CA) according to the manufacturer's specifications. For each sample, ∼0.5 μg of total RNA was reverse transcribed using Advantage RT-for-PCR (Clontech, Palo Alto, CA) according to the manufacturer's specifications. PCR was then performed to assess the expression by HASM cells of IL-6R and gp130. gp130 was amplified using two sets of primers. The first set, 5′-TGACGTTGCAGACTTGGGTA-3′ (forward primer) and 5′-TTCTGTTCAAGCTGTCCGAA-3′ (reverse primer), yielded a 337-bp product. The second set, 5′-TGGAGTGAAGAAGCAAGTGG-3′ (forward primer) and 5′-AACAGCTGCATCTGATTTGC-3′ (reverse primer), yielded a 303-bp product. Each PCR contained 200 ng of cDNA, 0.5 μl of Taq polymerase (Promega, Madison, WI), 200 μM dNTPs (Clontech), and 20 pmol of each primer in a total volume of 50 μl of PCR buffer (Promega). Conditions for PCR were 94°C for 3 min, followed by 35 cycles for 45 s at 94°C, 45 s at 60°C, and 2 min at 72°C, with a final extension time of 7 min at 72°C. Amplication of the human IL-6R was accomplished using the amplimer set from Clontech. The PCR was performed with the same conditions as described previously for gp130.
PCR was also performed to assess the expression of IL-11R and OSM receptor (OSM-R) by HASM. IL-11R was amplified using the following primers: 5′-CCAAACCTGTAGAGGACCCA-3′ (forward primer) and 5′-CGTTCCTTGAGCAGAACTCC-3′ (reverse primer), yielding a 224-bp product. OSM-R was amplified using 5′-TCACGTGCTGGTGGATACAT-3′ (forward primer) and 5′-TGAATCAGCATCGAGGAGTG-3′ (reverse primer), yielding a 342-bp product. The PCR conditions for both the IL-11R and OSM-R were as follows: 94°C for 3 min, followed by 35 cycles for 45 s at 94°C, 45 s at 58°C, and 1 min at 72°C, with a final extension time of 7 min at 72°C.
Drugs and reagents for tissue culture used in this study were obtained from Sigma (St. Louis, MO), with the exception of the following. Amphotericin B and trypsin-ETDA solution were purchased from GIBCO BRL (Life Technologies, Grand Island, NY). Recombinant human IL-1β, IL-6, IL-11, and OSM were purchased from R&D Systems.
Data analysis and statistics.
ANOVA was used to examine the statistical significance among the changes in PGE2 release induced by IL-1β, IL-6, IL-11, and OSM, using treatment and experimental days as main effects.P < 0.05 was considered significant.
HASM cells express IL-6R, gp130, IL-11R, and OSM-R.
To determine whether IL-6 family cytokines were able to act on HASM, we used RT-PCR to examine the expression of IL-6 family receptor components. As shown in Fig. 1, both components of the IL-6 receptor, gp130 and IL-6R (gp80), are expressed in HASM cells from two different donors. The gp130 products are visible in lanes 2–5 (Fig. 1 A). Two different sets of primers were used to examine the expression of gp130. Lanes 2 and 3 show a band consistent with the 337-bp product expected using the first set of primers, whereas lanes 4 and5 show a slightly smaller molecular mass band consistent with the 303-bp product expected using the second set of primers. Figure 1 B shows bands in lanes 2 and 3consistent with the 251-bp product representing the IL-6R. The presence of the IL-11R and OSM-R is displayed in Fig. 1 C. The 224-bp product in lanes 2 and 3 represents the IL-11R, whereas the higher molecular mass bands in lanes 4 and5 are consistent with the 334-bp product predicted for the OSM-R. In all these instances, PCR performed without cDNA or without the RT step yielded no discernible bands.
IL-6 and OSM activate STAT3.
To determine whether ligation of the IL-6 receptor is capable of inducing signaling in HASM cells, we measured STAT3 phosphorylation after the addition of IL-6 (20 ng/ml). Maximal phosphorylation of STAT3 by IL-6 occurred at 15 min (Fig.2 A). By 1 h, the effect of IL-6 had markedly diminished. Treatment with IL-11 (20 ng/ml) did not produce a significant increase in phosphorylated STAT3 (Fig.2 B), whereas OSM at 20 ng/ml induced a large signal that peaked at 10 min and remained elevated even 1 h later (Fig.2 C). Similar results were obtained in cells from three different donors (Fig. 3) and confirmed that OSM induced a more substantial and prolonged activation of STAT3 than IL-6. Fifteen-minute treatment with IL-6 caused a dose-dependent phosphorylation of STAT3 beginning at 2 ng/ml and peaking at 20 ng/ml (Fig. 4 A). A similar dose response was observed for OSM-induced phospho-STAT3 production (Fig.4 B). IL-6 did not induce phosphorylation of ERK in HASM. However, OSM (20 ng/ml) caused an increase in ERK phosphorylation, which was maximal at 15 min and markedly waned by 1 h. (Fig.5). Whereas both OSM (20 ng/ml) and IL-1β (0.2 or 2 ng/ml) induced ERK phosphorylation at 15 min, there were no additive or synergistic effects observed when both were given in combination (data not shown). None of the IL-6 family cytokines tested induced phosphorylation of p38 (data not shown).
Effect of IL-6, IL-11, and OSM on IL-1β-induced COX-2 expression and PGE2 release.
To determine whether IL-1β-induced production of IL-6 family cytokines contributes to the induction of COX-2 by IL-1β, we examined the effect of IL-6, IL-11, and OSM alone on COX-2 expression by Western blotting (Fig. 6). IL-6 (20 ng/ml for 24 h) treatment alone did not result in COX-2 expression, whereas treatment of HASM cells with IL-1β at concentrations of 0.2 and 2 ng/ml resulted in significant COX-2 expression, as previously described (Fig. 6 A) (16). No further increase in COX-2 expression was observed when cells were treated with both IL-1β and IL-6. Similar results were obtained in cells from four donors. IL-11 also failed to induce COX-2 expression or to augment IL-1β-induced COX-2 expression (Fig. 6 B). OSM alone did not induce COX-2 expression. However, there was marked synergy between IL-1β (2 ng/ml) and OSM (Fig. 6 B). These results were reproduced in cells from three donors (Fig. 7). The effect of OSM was dose dependent and observed at concentrations as low as 0.5 ng/ml (Fig. 6 C).
To examine the role of ERK in the ability of OSM to augment IL-1β-induced COX-2 expression, we examined the effect of U-0126, a MEK inhibitor (Fig. 8). As previously described (24), U-0126 reduced but did not abolish IL-1β-induced COX-2 expression. U-0126 also attenuated COX-2 expression in cells treated with IL-1β and OSM, but the ability of OSM to augment IL-1β-induced COX-2 expression was not reduced by U-0126. In cells treated with U-0126, OSM still caused an augmentation of COX-2 expression to 178 ± 100% of that induced by IL-1β alone, compared with 155 ± 51% in cells treated with vehicle (n = 4 in each group).
IL-6 (20 ng/ml for 24 h) had no effect on basal release of PGE2, whereas IL-1β (0.2 and 2 ng/ml) caused a marked increase in PGE2 release. IL-6 did not enhance the ability of IL-1β at 0.2 ng/ml to induce PGE2 release. However, IL-1β at a concentration of 2 ng/ml together with IL-6 yielded higher amounts of PGE2 (1,076 ± 186 pg/ml) compared with IL-1β alone at 2 ng/ml (599 ± 126 pg/ml, P < 0.005). Compared with unstimulated cells, AA (10−5 M) caused a marked increase in PGE2 release. IL-6 alone had no effect on PGE2 release in cells treated with AA, whereas IL-1β did, consistent with previous reports. In contrast to the ability of IL-6 to augment IL-1β-induced basal PGE2release, IL-6 did not augment IL-1β-induced PGE2 release induced by AA. These results suggest that the ability of IL-6 to enhance IL-1β PGE2 release in unstimulated cells is the result of increased availability of AA rather than increased COX-2 expression.
IL-11 alone (20 ng/ml) did not induce PGE2 release and did not enhance the effects of IL-1β in either unstimulated or AA-stimulated cells (data not shown). OSM (20 ng/ml) alone also had no effect on PGE2 release in HASM cells. However, OSM produced a significant increase in IL-1β (0.2 and 2.0 ng/ml)-induced PGE2 production in both unstimulated (Fig.9 A, P < 0.005 and P < 0.05, respectively, for OSM + IL-1β at 0.2 ng/ml and OSM + IL-1β at 2 ng/ml) and AA-stimulated cells (Fig. 9 B, P < 0.001 for both OSM + IL-1β at 0.2 and 2 ng/ml), consistent with the ability of OSM to augment COX-2 expression induced by IL-1β (Figs. 6, B andC, and 7).
Indomethacin does not alter IL-1β-induced IL-6 production.
In human cancer cells (14), human osteoblasts (41), and mouse macrophages (29), PGE2 has been found to increase IL-6 production. Because IL-1β induces both marked increases in PGE2 release (see above) as well as increases in IL-6 production (11), we hypothesized that IL-1β-induced IL-6 formation might be mediated by IL-1β-induced PGE2. To test this hypothesis, we treated HASM cells with IL-1β and measured IL-6 release in the presence and absence of the COX inhibitor indomethacin. IL-1β (2 ng/ml) caused a marked increase in IL-6 release into the supernatant in HASM, consistent with previous reports (11). This increase in IL-6 was not affected by pretreatment with the cyclooxygenase inhibitor indomethacin at 10−6 M (Fig.10). In contrast, indomethacin virtually abolished IL-1β-induced PGE2 formation, indicating that the lack of effect on IL-6 production was not due to a lack of efficacy. Production of PGE2 averaged 16 ± 5 pg/ml (n = 3), 649 ± 96 pg/ml (n= 6), and 59 ± 22 pg/ml (n = 6) in control cells, cells treated with IL-1β (2 ng/ml for 24 h), and cells treated with both IL-1β and indomethacin, respectively. OSM was below the limit of detection both in unstimulated HASM cells and in cells treated with IL-1β (2 ng/ml).
Our data indicate that IL-6 family cytokines are capable of activating HASM cells and altering their function; the components necessary for IL-6, IL-11, and OSM to bind to the cell membrane, namely gp130 and IL-6R, IL-11R, and OSM-R, are expressed, and both IL-6 and OSM cause a dose- and time-dependent phosphorylation of STAT3. Furthermore, both IL-6 and OSM synergize with IL-1β to increase PGE2 release, although their mechanisms of action differ.
Activation of STAT3 by IL-6 family cytokines has been described in other cell types. The phosphorylation of STAT3 by IL-6 has been well documented in mouse hepatocytes (28), rat Sertoli cells (20), and rat cerebral cortex (10). Significant activation of STAT3 in response to IL-6 also has been described in human breast carcinoma (42) and hepatoma (39) cells. It is therefore not surprising that we found increased STAT3 phosphorylation in response to IL-6. Although IL-11 also has been shown to result in phosphorylation of STAT3 in other cell types, such as human umbilical vein endothelium (30), there was no significant STAT3 effect seen in our cells. Whereas the presence of the IL-11R was established by PCR, the actual receptor number was not quantified. It is possible that the level of receptor expression was too low to induce detectable signaling. We found that OSM induced a more potent phosphorylated STAT3 signal compared with either IL-6 or IL-11 in HASM cells (Fig. 3). Similar results were found in human brain tumor cells, with OSM producing an earlier, more potent activation of STAT3 compared with the other IL-6 family cytokines (37).
IL-6 does have a role in the expression of COX-2 in some cell types. In mouse osteoblasts, IL-6 both alone and in conjunction with IL-1β was found to stimulate PGE2 production and COX-2 gene transcription (40). Similar effects were seen in a canine basilar artery model, although the degree to which IL-6-induced PGE2 was less than with IL-1β (32). Ferreira et al. (13) have also demonstrated that IL-6 synergistically acts with IL-1β to induce COX-2 expression in a rat hyperalgesia model. The expression of COX-2 induced by serum from preeclamptic subjects was inhibited by anti-IL-6 antibody in uterine endothelial cells (1). However, IL-6 alone did not induce COX-2 expression in HASM cells, and IL-6 did not synergize with IL-1β to augment COX-2. These findings also have been observed in rat uterus, as well as in rat microglial cells and hepatocytes (5, 6,31).
PGE2 is produced when membrane phospholipids are converted by phospholipase A2 to AA. AA becomes a substrate for COX, resulting in the release of PGE2. Even though IL-6 did not augment COX-2 expression caused by IL-1β, it did increase IL-1β-induced PGE2 release. The increase in PGE2 by IL-6 was seen only in unstimulated but not in AA-stimulated cells. Therefore, the increased production of PGE2 by IL-6 in unstimulated cells may result from effects of the cytokine on the expression or activation of phospholipase A2. If the effect of IL-6 occurred at the level of COX-2 expression, an increase in the amount of PGE2 release would have been expected after the administration of AA.
We observed a marked synergy between OSM and IL-1β in the induction of COX-2 (Figs. 6 B and 7). Synergy between OSM and IL-1β has also been observed in human aortic smooth muscle cells (9). However, in the aortic cells, COX-2 was induced by OSM alone, whereas in HASM cells, it was not. Differences in signaling may account for these differences in COX-2 expression. In contrast to our results, STAT1, not STAT3, was the major STAT involved in the OSM signaling pathway in aortic smooth muscle.
It is possible that differences in STAT3 phosphorylation by OSM and IL-6 (Fig. 3) may be responsible for the differences in their effects on COX-2. STAT3 has been shown to activate the junB promoter in a hepatoma cell line (22) and α2-macroglobulin in human breast carcinoma cells (42) as well as c-Fos (16, 24). The transcription factor AP-1 is composed of dimers of Fos and Jun and there are AP-1-like elements in the promoter of the COX-2 gene (4).
It is also possible that the differences in effects of IL-6 and OSM on IL-1β-induced COX-2 expression are due to differences in their effects on ERK phosphorylation. Whereas both OSM and IL-6, to a lesser degree, have been shown to phosphorylate ERK in HepG2 cells and or hepatoma cells (21, 39), OSM, but not IL-6, caused ERK phosphorylation in HASM cells. ERK phosphorylation is known to be important in IL-1β-induced COX-2 expression in HASM cells (25). However, there was no synergism between IL-1β and OSM in their ability to activate ERK. The MEK inhibitor U-0126 did appear to decrease COX-2 expression in cells treated with IL-1β and OSM, but this effect is likely due to the effects of U-0126 on IL-1β-induced effects. In fact, the ability of OSM to enhance IL-1β-induced COX-2 expression was not altered in cells treated with U-0126, suggesting that ERK is not involved in the mechanism of action of OSM.
Because PGE2 has been shown to be capable of inducing IL-6 in some cell types and because IL-1β evokes the release of large amounts of PGE2 and IL-6, we sought to determine whether IL-1β-induced prostanoids might contribute to the release of IL-6 by IL-1β. The ability of IL-1β to induce production of IL-6 was not affected by the nonsteroidal anti-inflammatory drug indomethacin (Fig.10). This suggests that in HASM cells PGE2 is not primarily responsible for IL-6 production. In contrast, studies on human osteoblasts have revealed a suppression of IL-1β-induced IL-6 production by treatment with the specific COX-2 inhibitor NS-398 (41). Similar results were seen in human macrophages, suggesting regulation of IL-6 by COX-2 (44). Animal models of inflammation also demonstrated that COX-2 was involved in the regulation of IL-6 production (3, 29).
No production of OSM was observed in either control or IL-1β-treated HASM cells, whereas both IL-6 and IL-11 were produced (11). OSM is produced by macrophages and neutrophils (15, 36). Hence, it is possible that OSM might contribute to COX-2 expression and PGE2 release with illnesses characterized by airway inflammation, such as asthma and other chronic obstructive pulmonary diseases. In these disease states, the above-implicated cells, macrophages and neutrophils, are known to have a major role.
In summary, we found that members of the IL-6 cytokine family activate STAT3 and play a role in COX-2 expression and/or PGE2release in HASM cells. OSM, synergistically with IL-1β, induced a significant increase in both COX-2 expression and PGE2release. Although OSM also phosphorylated ERK, ERK did not appear to have a role in the ability of OSM to enhance IL-1β-induced COX-2 expression. Although IL-6 was found to increase IL-1β-induced PGE2 release, this effect did not appear to occur at the level of COX-2 expression.
We acknowledge Igor Schwartzman and Trudi Church for the technical assistance they provided in performing these experiments.
This work was supported by the National Heart, Lung, and Blood Institute Grant HL-56383. J. L. Laporte and P. E. Moore were supported by an American Lung Association (ALA) Fellowship and an ALA grant.
Address for reprint requests and other correspondence: S. Shore, Physiology Program, Harvard School of Public Health, 665 Huntington Ave, Boston, MA 02115.
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 © 2001 the American Physiological Society