Vascular endothelial growth factor (VEGF), a potent angiogenesis factor, likely contributes to airway remodeling in asthma. We sought to examine the effects and mechanism of action of IL-6 family cytokines on VEGF release from human airway smooth muscle (HASM) cells. Oncostatin M (OSM), but not other IL-6 family cytokines, increased VEGF release, and IL-1β enhanced OSM-induced VEGF release. OSM increased VEGF mRNA expression and VEGF promoter activity, whereas IL-1β had no effect. IL-1β did not augment the effects of OSM on VEGF promoter activity but did augment OSM-induced VEGF mRNA expression and mRNA stability. The STAT3 inhibitor piceatannol decreased both OSM-induced VEGF release and synergy between OSM and IL-1β, without affecting responses to IL-1β alone. Piceatannol also inhibited OSM-induced VEGF mRNA expression. In contrast, inhibitors of MAPK pathway had no effect on OSM or OSM plus IL-1β-induced VEGF release. OSM increased type 1 IL-1 receptor (IL-1R1) mRNA expression, as measured by real-time PCR, and piceatannol attenuated this response. Consistent with the increase in IL-1R1 expression, OSM markedly augmented IL-1β-induced VEGF, MCP-1, and IL-6 release. In summary, our data indicate OSM causes VEGF expression in HASM cells by a transcriptional mechanism involving STAT3. IL-1β also synergizes with OSM to increase VEGF release, likely as a result of effects of IL-1β on VEGF mRNA stability as well as effects of OSM on IL-1R1 expression. This is the first description of a role for OSM on IL-1R1 expression in any cell type. OSM may contribute to airway remodeling observed in chronic airway disease.
- interleukin-1 receptor 1
- signal transducer and activator of transcription 3
- monocyte chemoattractant protein-1
- vascular endothelial growth factor
it is likely that vascular endothelial growth factor (VEGF), a potent endothelial cell mitogen and angiogenesis factor (57), plays an important role in asthma. Angiogenesis is a feature of asthmatic airways (19, 22), and expression of VEGF is increased in asthmatic airways, especially upon allergen challenge (1, 6, 19). Indeed, VEGF expression in airway biopsies has been shown to correlate with decline in forced expiratory volume in 1 s (19). The pathophysiological importance of VEGF for asthma is emphasized by the observations that VEGF receptor inhibitors reduce airway hyperresponsiveness in a mouse model of toluene diisocyanate-induced asthma (35) and that overexpression of VEGF in the airways is sufficient to induce airway hyperresponsiveness (34).
Potential cellular sources of VEGF in the airways include airway epithelial cells (7), mast cells (5), and airway smooth muscle cells (26). Human airway smooth muscle (HASM) cells in culture produce VEGF constitutively, and VEGF expression is upregulated by the inflammatory cytokine, IL-1β, and by a number of other inflammatory mediators (24, 26, 60). Immunohistochemical staining of human and rat lungs for VEGF confirms its expression in airway smooth muscle (13).
The IL-6 family of cytokines, which includes IL-6 itself, leukemia inhibitory factor (LIF), cardiotrophin-1 (CT-1), IL-11, ciliary neurotrophic factor (CNTF), and oncostatin M (OSM), is a group of pleiotropic cytokines released during infection, inflammation, and injury. In other cell types, members of the IL-6 family have been shown to induce VEGF expression (14, 46, 55, 58, 59). Of these cytokines, OSM is the most potent (59). OSM has also been shown to induce angiogenesis (55), perhaps through effects on VEGF. Receptors for IL-6 family cytokines are present on HASM cells, and these cytokines, particularly OSM, induce signaling events in HASM cells typical of IL-6 family cytokines. These events include activation of signal transducer and activator of transcription 3 (STAT3), extracellular signal-regulated kinase (ERK), and c-Jun NH2-terminal kinase (JNK) (11, 25, 27, 39). Hence, the purpose of these experiments was to determine whether OSM and/or other IL-6 family cytokines also cause VEGF release from HASM cells and to examine the mechanistic basis for this effect.
Others have reported marked synergy between OSM and IL-1β in the induction of VEGF release from other cell types (46), and our results indicate that OSM, but not other IL-6 family cytokines, also synergizes with IL-1β to induce VEGF release in HASM cells. To determine whether the effects of OSM on VEGF expression are transcriptionally mediated, we measured effects of OSM and IL-1β alone and in combination on VEGF mRNA expression, VEGF promoter activity, and VEGF mRNA stability. Because others have reported a role for STAT3 in the effects of OSM on VEGF release in other cell types (46), we also examined the role of STAT3. Finally, to determine whether OSM might enhance responses to IL-1β by augmenting expression of the type 1 IL-1 receptor (IL-1R1), we measured OSM-induced changes in IL-1R1 mRNA expression using real-time PCR.
HASM cells were obtained from lung transplant donor tracheae in accordance with procedures approved by the University of Pennsylvania Committee on Studies Involving Human Beings and were cultured as previously described (32, 33, 44). The medium was replaced every 3–4 days. The cells were passaged with 0.25% trypsin and 1 mM EDTA every 10–14 days. Unless otherwise indicated, confluent cells were serum deprived and supplemented with 5.7 μg/ml of insulin and 5 μg/ml of transferrin 48 h before use. Cells in passages 4–7 were used in the studies described below.
HASM cells were treated with cytokines for 24 h. The supernatant was removed and stored at −20°C until subsequent assay. VEGF, IL-6, and monocyte chemoattractant protein-1 (MCP-1) were analyzed by ELISA (R&D Systems, Minneapolis, MN). The lower limit of detection for each of these ELISAs is <5 pg/ml. For experiments involving inhibition of MAPKs or STAT3, cells were treated with inhibitor 2 h before addition of cytokines. To assess the effects of p38, JNK, and ERK, we used SB-203580, SP-600125, or U0126, respectively. The concentrations of inhibitors used were those that we and others have previously shown to be effective in inhibiting activation of the various MAPKs without nonspecific effects on other kinases (3, 9, 12, 32, 33). To assess the effects of STAT3, we used the STAT3 inhibitor piceatannol (2, 37, 52). In all cases, drugs were dissolved in DMSO and diluted appropriately in PBS before addition to cells. Consequently, DMSO (0.01%) was added to control cells. Recombinant human cytokines were purchased from R&D Systems and Oncogene Research Products (San Diego, CA). MAPK and STAT3 inhibitors were obtained from Calbiochem (San Diego, CA).
Confluent cells were treated with cytokines for 15 min and then lysed as previously described (27). In cells treated with piceatannol, this STAT3 inhibitor was added 2 h before cytokine treatment. Solubilized proteins (30 μg/lane) were separated by SDS-PAGE (Invitrogen, Carlsbad, CA) and transferred to a BioTrace polyvinylidene fluoride membrane in transfer buffer (PALL Life Sciences, Pensacola, FL). The membranes were blocked and then probed with rabbit anti-phospho-Tyr705-STAT3, anti-phospho-Ser727-STAT3, or anti-STAT3 (Cell Signaling, Woburn, MA) overnight at 4°C with agitation. 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 X-Omat BluexB-1 film (Eastman Kodak, Rochester, NY) with enhanced chemiluminescent substrate (Cell Signaling).
RNA extraction and real-time PCR.
HASM cells were treated with OSM (2–20 ng/ml for 6 h) or left untreated. In experiments involving U0126 or piceatannol, cells were treated with the compound for 2 h before addition of OSM. Total RNA was extracted with TRIzol reagent (Invitrogen, Life Technologies) in accordance with the manufacturer's instructions. Samples were aliquoted and stored at −80°C. Reverse transcription (RT) was performed with 2 μg total RNA, 1 μg random hexamers (Invitrogen, Life Technologies), 50 nmol dNTP, 25 units of RNaseOUT RNase inhibitor (Invitrogen, Life Technologies), and 200 units of Moloney murine leukemia virus RT (Promega) in a 25-μl total volume consisting of 50 mM Tris·HCl (pH 8.3), 75 mM KCl, 3 mM MgCl2, and 10 mM DTT at 37°C for 60 min. RT reactions were diluted in 30 μl of water and stored at −80°C. Quantitative real-time RT-PCR was performed on RNA from HASM cell supernatants using an iCycler iQ Real Time Detection System and iQ SYBR Green Supermix in accordance with the manufacturer's instructions (Bio-Rad, Hercules, CA). The primer set forward: 5′-GCT TCA GAT GGA AAG ACC TA-3′ and reverse: 5′-TCC AGC TCA AGC AGG ACA A-3′ was used to amplify a 338-bp fragment of human IL-1R1 mRNA. The primer set forward: 5′-GAG GGC AGA ATC ATC ACG AA-3′ and reverse: 5′-CAT GGT GAT GTT GGA CTC CT-3′ was used to amplify a 215-bp fragment of human VEGF mRNA. Note that VEGF exists in several alternatively spliced forms, of which VEGF121, 165, 189, and 206 are expressed in airway smooth muscle (24, 26). Primers for VEGF were chosen to amplify a region common to all isoforms, which all contain exons 1–4 (54), to yield a single product suitable for real-time PCR. Similarly, the primer set forward: 5′-GAGTCCACTGGCGTCTTCA-3′ and reverse: 5′-TTCAGCTCAGGGATGACCTT-3′ was used to amplify a fragment of GAPDH. The products were gel purified, cloned into TOPO-TA vectors (Invitrogen, Life Technologies), and used as positive controls to construct standard curves for real-time PCR. For each of these genes, melting curve analysis yielded a single peak consistent with one PCR product. Changes in IL-1R1 or VEGF mRNA transcript copy number were assessed relative to changes in GAPDH mRNA transcript copy number.
Transfection of HASM cells.
After passage into six-well plates, HASM cells were grown in serum containing medium for 72 h (60–80% confluence). Before transfection, the medium was changed to serum-free hormone-supplemented media as described above. HASM cells were transfected with 2 μg of a 2.5-kb human VEGF promoter-firefly luciferase reporter construct (courtesy of Dr. Debabrata Mukhopadhyay, Mayo Clinic). The construct was transfected with Fugene 6 (Roche, Indianapolis, IN) according to the manufacturer's protocol. Twenty-four hours after transfection, the cells were either left untreated or were treated with cytokines for 24 h. The cells were then lysed with reporter lysis buffer (Promega, Madison, WI), harvested, and assayed for luciferase activity by scintillation counting using the Luciferase Reporter Assay System (Promega).
Analysis of mRNA stability.
Confluent HASM cells were treated with OSM (20 ng/ml) alone or in combination with IL-1β (10 ng/ml) for 6 h. Thereafter, actinomycin D (Sigma) was added (10 μg/ml), and total mRNA was isolated at 0, 1, and 2 h. Total mRNA was extracted and analyzed for VEGF mRNA by colorimetric quantitation (Quantikine mRNA, R&D Systems) following the manufacturer's instructions. According to the manufacturer's information, all known human VEGF mRNA splice variants (VEGF121, 145, 165, 183, 189, and 206) are detectable with this kit, although the splice variants cannot be distinguished.
Differences in VEGF, MCP-1, or IL-6 release or in VEGF mRNA or protein expression were assessed by ANOVA or factorial ANOVA using Statistica software (Statsoft, Tulsa, OK). A P value <0.05 was considered statistically significant.
OSM caused a concentration (Fig. 1A) - and time (Fig. 1B)-related increase in release of VEGF from HASM cells. In contrast, in cells from the same donors, none of the other IL-6 family cytokines had any significant effect, even at higher concentrations (Fig. 1A). IL-1β enhanced the ability of OSM to induce release of VEGF (Fig. 1C), leading to a marked upward shift in the OSM dose-response curve. In contrast, there was no synergy between IL-1β and any of the other IL-6 family cytokines (Fig. 1D). IL-11, LIF, IL-6, CT-1, and CNTF still failed to increase VEGF release even in the presence of IL-1β.
To determine whether the effect of OSM and the synergy between IL-1β and OSM occurred at the transcriptional level, we measured VEGF mRNA expression by real-time PCR. OSM (for 6 h) caused a dose-dependent increase in VEGF mRNA expression relative to GAPDH mRNA expression (Fig. 2A). OSM (20 ng/ml) also increased VEGF promoter activity (Fig. 2B), suggesting that the effect of OSM on VEGF mRNA was transcriptionally mediated. In contrast, IL-1β (10 ng/ml for 6 h) had no statistically significant effect on either VEGF promoter activity (Fig. 2B) or VEGF mRNA expression (Fig. 2C), suggesting that the effect of IL-1β alone was mediated at the translational level. A more detailed kinetic study also failed to show any effect of IL-1β on VEGF mRNA expression after 30 min or 1, 6, 8, 18, or 24 h (data not shown). The combination of IL-1β and OSM did not increase VEGF promoter activity above that induced by OSM alone (Fig. 2B). Cells treated with IL-1β and OSM had greater VEGF mRNA, as determined by real-time PCR, than cells treated with OSM alone (Fig. 2C), although the effect failed to reach statistical significance.
To determine whether synergy between OSM and IL-1β might result from effects of IL-1β on VEGF mRNA stability, we examined VEGF mRNA in cells treated with OSM alone or OSM in combination with IL-1β for 6 h. Actinomycin D was then added to the cells to inhibit transcription, and VEGF mRNA was measured 0, 1, and 2 h after addition of actinomycin D. In cells treated with OSM alone, VEGF mRNA was relatively short lived, as reported by others (36, 51). OSM (20 ng/ml for 6 h) significantly increased VEGF mRNA over values obtained in untreated control cells, but VEGF mRNA returned to control untreated values within 2 h after the addition of actinomycin D (Fig. 3). Before addition of actinomycin D, VEGF mRNA was greater in cells treated with OSM plus IL-1β than in cells treated with OSM alone (P < 0.05). In addition, the stability of VEGF mRNA was increased in cells treated with OSM plus IL-1β. In these cells, even 2 h after addition of actinomycin D, VEGF mRNA had not declined from pre-actinomycin D values (Fig. 3).
To examine the role of STAT3 in the effects of OSM, we used the STAT3 inhibitor piceatannol (Fig. 4) (2, 37, 52). At concentrations up to 3 μg/ml, there was no effect of piceatannol on total cell numbers or cell viability measured by trypan blue exclusion (data not shown). Piceatannol (3 μM) significantly decreased OSM-induced STAT3 phosphorylation by 47% (P < 0.05), as measured by densitometric analysis of Western blots from four smooth muscle cell donors (data not shown). Piceatannol caused a dose-dependent inhibition of VEGF release in cells treated with OSM (Fig. 4A) (P < 0.01). Piceatannol (3 μM) also significantly reduced VEGF release in unstimulated cells (P < 0.01), consistent with our previous observations of some degree of constitutive STAT3 phosphorylation in these cells even under basal, unstimulated conditions (27). Piceatannol had no effect on IL-1β-induced VEGF release, but it significantly inhibited VEGF release caused by the combination of IL-1β plus OSM (Fig. 4B). We also observed a decrease in OSM-stimulated VEGF mRNA expression in cells treated with piceatannol. In OSM-treated cells, piceatannol (3 μg/ml) caused a 38 ± 11% decrease in VEGF mRNA expression (normalized for GAPDH; P < 0.05).
Activation of STAT3 is initiated by OSM-dependent dimerization of glycoprotein (gp)130 and the OSM receptor (OSMR), resulting in STAT3 phosphorylation at Tyr705 (27). STAT3 can also be phosphorylated at Ser727, an event that does not impact dimerization or nuclear translocation of STAT3 but does increase its transcriptional activity (50). Because ERK is capable of causing Ser727 phosphorylation of STAT3 that is already phosphorylated on Tyr705 (61) and because OSM and IL-1β each cause ERK activation in HASM cells (27, 32), we examined the effect of OSM and IL-1β on Ser727 phosphorylation. OSM caused robust Tyr705 phosphorylation of STAT3 but failed to cause phosphorylation of STAT3 at Ser727 (Fig. 5), most likely because OSM does not cause a sufficient increase in ERK activation. IL-1β on its own failed to induce either Tyr705 or Ser727 phosphorylation of STAT3. However, combined treatment with OSM and IL-1β did result in Ser727 phosphorylation of STAT3 that was prevented by inhibiting ERK phosphorylation with the MEK inhibitor U0126. It is likely that this was the result of Tyr705 phosphorylation of STAT3 by OSM, which is required for subsequent Ser727 phosphorylation (61), in conjunction with the very robust ERK activation that is induced by IL-1β in HASM cells (32). To investigate the role of ERK-dependent Ser727 phosphorylation in the synergism between OSM and IL-1β, we examined the effect of U0126 on VEGF release. U0126 had no effect on OSM (5 ng/ml)-induced VEGF release, and it did not alter VEGF release in cells treated with IL-1β alone (10 ng/ml) or cells treated with IL-1β plus OSM (data not shown). Thus whereas OSM and IL-1β together, but not separately, do induce ERK-dependent Ser727 STAT3 phosphorylation, Ser727 phosphorylation does not appear to be required for the synergism observed between the two cytokines. Rather, it is phosphorylation of STAT3 Tyr705 that is important for the effects of OSM on VEGF expression. The p38 inhibitor SB-203580 and the JNK inhibitor SP-600125 were also without effect on VEGF release induced by OSM either alone or in combination with IL-1β (data not shown).
To examine the possibility that OSM synergizes with IL-1β by increasing expression of the IL-1R1, using real-time PCR we compared IL-1R1 mRNA expression in untreated cells and cells treated with OSM. OSM (20 ng/ml for 6 h) caused a significant, approximately sixfold increase in the ratio of IL-1R1 mRNA:GAPDH mRNA (Fig. 6). Significant increases in IL-1R1 mRNA:GAPDH mRNA were also observed at 2 and 5 ng/ml of OSM (data not shown). Piceatannol did not alter basal IL-1R1 expression, but it did cause a significant, ∼50% reduction in IL-1R1 expression in cells treated with OSM (Fig. 6). In contrast, the MEK inhibitor U0126 had no effect on OSM-induced IL-1R1 expression (data not shown).
To evaluate possible functional implications of OSM-induced changes in IL-1R1 expression, we examined the effects of OSM on IL-1β-induced VEGF, MCP-1, and IL-6 release. IL-1β caused a dose-dependent increase in VEGF, MCP-1, and IL-6 release from HASM cells (Fig. 7), as described by others (10, 56, 60). In the presence of OSM (2 ng/ml), increases in VEGF, MCP-1, and IL-6 were observed at lower doses of IL-1β, and the magnitude of the increase induced by IL-1β was greatly enhanced.
As reported for other cell types (4, 8, 37, 49), OSM also caused a dose-related increase in MCP-1 and IL-6 release in HASM cells (Fig. 8, A and B), and IL-1β increased the dose-related effects of OSM. Other IL-6 family cytokines failed to cause either MCP-1 or IL-6 release even in the presence of IL-1β (data not shown). As was the case for VEGF, the STAT3 inhibitor piceatannol reduced OSM-induced MCP-1 and IL-6 release (Fig. 9, A and B). Piceatannol also attenuated OSM plus IL-1β-induced MCP-1 and IL-6 release, without any effect on responses to IL-1β alone.
Our results indicate that OSM, but not other IL-6 family cytokines, causes release of VEGF from HASM cells (Fig. 1). Other IL-6 family cytokines have been shown to induce VEGF in other cell types (14, 46, 55, 58, 59), but in studies that compared VEGF release by the various IL-6 family members, OSM was the most potent (59). OSM was also unique among IL-6 family members in its capacity to induce MCP-1 and IL-6 expression in HASM cells (Fig. 8). Similarly, OSM, but not LIF or IL-6, causes MCP-1 and IL-6 release from mouse synovial fibroblasts (29). Members of the IL-6 cytokine family share a common receptor subunit, gp130 (17). OSM binding to the OSMR results in the formation of an OSMR/gp130 heterodimer, whereas other IL-6 family members result in either gp130 homodimers or LIF receptor (LIFR)/gp130 heterodimers. Thus the increased ability of OSM to induce VEGF release in HASM cells may be related to its unique ability to activate OSMR. Other unique proinflammatory effects of OSM have been reported. For example, OSM, but not other IL-6 family members, increases expression of the neutrophil chemotactic factor epithelial cell-derived neutrophil activator-78 in endothelial cells (40). OSM, but not IL-6, LIF, or CNTF, also promotes growth of vascular smooth muscle cells (15).
The cytoplasmic domains of OSMR, gp130, and LIFR are constitutively associated with janus kinases (JAKs), and receptor binding and dimerization leads to JAK phosphorylation of gp130 and LIFR or OSMR resulting in activation of STAT3, ERK, and JNK (18, 27, 30). We have previously reported that OSM is more effective than the other IL-6 family cytokines in causing ERK, JNK, and STAT3 phosphorylation in HASM cells (11, 28). Differences in ERK and JNK activation do not appear to contribute to differences in the ability of OSM vs. other IL-6 family cytokines to induce VEGF release, since neither the MEK inhibitor U0126, which inhibits ERK phosphorylation and subsequent activation, nor the JNK inhibitor SP-600125 had any effect on OSM-induced VEGF release. In contrast, OSM not only increased STAT3 activation (Fig. 5) but the STAT3 inhibitor piceatannol caused a marked and dose-dependent inhibition of OSM-induced VEGF release in HASM cells (Fig. 4), suggesting that STAT3 is required for OSM-induced VEGF expression in these cells. STAT3 is also required for release of VEGF by OSM and other IL-6 family members in other cell types (14, 46). Others have demonstrated a STAT3 binding site 848–841 bp upstream of the transcription start site that is important for VEGF promoter activity in other cell types. STAT3 binds to this site, mutation of this site abolishes VEGF promoter activity induced by a constitutively active form of STAT3, and promoter reporter constructs that do not contain this site are not activated by STAT3 (41, 57). A direct effect of STAT3 at the level of the VEGF promoter is consistent with our observations that OSM caused an increase in VEGF promoter activity (Fig. 2). Piceatannol also inhibited OSM-induced MCP-1 and IL-6 release from HASM cells (Fig. 9), suggesting a role for STAT3 in those events. A role for STAT3 in OSM-induced MCP-1 release has also been demonstrated in osteoblasts (37).
Piceatannol has been shown to effectively inhibit STAT3 phosphorylation and DNA binding in a number of cell types (2, 37, 52), and it is likely that its effects on responses to OSM are mediated via STAT3 inhibition. Indeed, piceatannol (3 μg) significantly inhibited OSM-induced STAT3 activation in our cells. At the doses used, we did not see any effect of piceatannol on cell viability, and it did not affect responses to IL-1β (Figs. 4B and 9, A and B), suggesting that the effects observed were not the result of any toxicity to the cells. Nevertheless, it is possible that effects of piceatannol other than inhibition of STAT3 may have contributed to its ability to reduce responses to OSM. For example, piceatannol also inhibits Syk tyrosine kinase (43). However, we are unaware of any data indicating a role for Syk in responses to any IL-6 family cytokines or of any role for Syk in HASM cells.
Our results indicate that IL-1β causes VEGF release from HASM cells (Fig. 1, C and D). Others have also reported increased VEGF release from IL-1β stimulated HASM cells (60), but the mechanistic basis for this effect has not previously been described. Our results indicate that IL-1β does not increase VEGF promoter activity or mRNA expression in HASM cells (Fig. 2), even though it does increase VEGF protein (Fig. 1, C and D), suggesting that the regulation of VEGF by IL-1β occurs by a posttranscriptional mechanism. A similar increase in VEGF protein but not VEGF mRNA occurs following IL-1β stimulation in astroglial cells (47). In contrast, IL-1β does increase VEGF mRNA in other cell types through effects on both transcription rate and mRNA stability (23, 53). Others have reported that in HASM cells, release of VEGF by bradykinin also occurs via a posttranscriptional mechanism (26).
We observed greater VEGF release in cells treated with OSM and IL-1β than in cells treated with either cytokine alone (Figs. 1 and 7). Because our data indicated a role for STAT3 in these events (Fig. 4), and because others have reported increased transcriptional activation of STAT3 following Ser727 phosphorylation (50), we considered the possibility that IL-1β and OSM might be synergizing at the level of Ser727 STAT3 phosphorylation. Our results indicate that OSM and IL-1β together do indeed cause Ser727 phosphorylation of STAT3 but that this phosphorylation is not required for the effects of these cytokines on VEGF release: the MEK inhibitor virtually abolished OSM plus IL-1β-induced Ser727 phosphorylation of STAT3 but did not alter OSM plus IL-1β-induced VEGF release.
To further evaluate the mechanistic basis for the interaction between OSM and IL-1β, we evaluated the combined effects of these cytokines on VEGF mRNA expression, VEGF promoter activity, and VEGF mRNA stability. Our results indicate that IL-1β increases OSM-induced VEGF mRNA expression (Fig. 3) and protein expression (Fig. 1) without affecting OSM-induced VEGF promoter activity (Fig. 2B). Our data demonstrating increased VEGF mRNA stability in cells treated with OSM plus IL-1β (Fig. 3) compared with cells treated with OSM alone suggest that IL-1β may be acting, at least in part, to increase the stability of mRNA induced by OSM-induced VEGF promoter activation.
In addition to changes in VEGF mRNA stability, it may be that effects of OSM on IL-1R1 expression contribute to the synergy between IL-1β and OSM. OSM caused a marked increase in IL-1R1 mRNA relative to GAPDH (Fig. 6). We did not measure IL-1R1 protein expression, and we do not know for certain that IL-1R1 expression limits responses to IL-1β in HASM cells. However, the marked leftward and upward shifts in the IL-1β dose-response curves that were induced by OSM suggest that the observed changes in IL-1R1 mRNA expression may be functionally relevant, especially since increased responses to IL-1β were observed for all three outcome indicators examined (VEGF, IL-6, and MCP-1 release). The ability of OSM and IL-1β to synergize in the induction of MCP-1 and IL-6 has been reported in other cell types (4, 31), but the role of changes in IL-1R1 expression in these events has not been previously explored. The effects of the STAT3 inhibitor piceatannol on OSM-induced IL-1R1 expression are also consistent with the hypothesis that changes in IL-1R1 expression might play a role in the synergy between OSM and IL-1β. Piceatannol reduced OSM-induced IL-1R1 expression (Fig. 6) and also reduced synergy between OSM and IL-1β that was observed for VEGF, IL-6, and MCP-1 expression (Figs. 4 and 9). These data are consistent with the hypothesis that OSM augments IL-1R1 expression by a STAT3-dependent mechanism. To our knowledge, this is the first report of effects of OSM on IL-1R1 expression in any cell type and the first report of a role for STAT3 in IL-1R1 expression.
The IL-6 family of cytokines is released during infection, inflammation, and injury and play important roles in the acute phase response. There are currently no data regarding the expression of OSM in the lungs and airways. However, OSM is present at sites of inflammation in other organs (20, 21, 40), and inflammatory cells that infiltrate the lungs and airways in diseases such as asthma and chronic obstructive pulmonary disease, including macrophages, T lymphocytes, neutrophils, eosinophils, and dendritic cells, have been shown to produce OSM (21, 38, 40, 45). Hence, it is likely that OSM is expressed in diseases characterized by airway infiltration with inflammatory cells and contributes to the VEGF expression observed in those airways. The observation that OSMR is expressed in abundance on airway smooth muscle of asthmatics (42) increases that likelihood. The ability of OSM to augment MCP-1 expression may also be very relevant to asthma. MCP-1 is chemotatic for monocytes and lymphocytes but also plays a crucial role in mast cell and basophil activation (48). MCPs are upregulated in asthmatic airways and have been demonstrated to contribute to polarization of T cells toward a Th2 phenotype (16).
In summary, the current study demonstrates that OSM causes VEGF, MCP-1, and IL-6 release from cultured HASM cells. Moreover, our data suggest that these effects of OSM are mediated through STAT3 and that, at least in the case of OSM-induced VEGF release, STAT3 is likely to be mediating its effects at the level of the VEGF promoter. OSM also synergizes with IL-1β to increase VEGF, at least in part, as a result of IL-1β-induced increases in VEGF mRNA stability. We speculate that OSM-induced increases in IL-1R1 expression may also contribute. The observation that OSM has the capacity to drive and amplify production of VEGF by airway smooth muscle, in conjunction with the presence of cell types capable of synthesizing OSM in inflamed airways, indicates that OSM could contribute to airway remodeling observed in asthma.
This work was supported by National Institutes of Health Grants HL-67664, HL-33009, and ES-00002.
The authors thank Igor Schwartzman for excellent technical assistance.
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 © 2005 the American Physiological Society