In lung injury and progressive lung diseases, the multifunctional cytokine transforming growth factor-β1 (TGF-β1) modulates inflammatory responses and wound repair. Heme oxygenase-1 (HO-1) is a stress-inducible protein that has been demonstrated to confer cytoprotection against oxidative injury and provide a vital function in maintaining tissue homeostasis. Here we report that TGF-β1 is a potent inducer of HO-1 and examined the signaling pathway by which TGF-β1 regulates HO-1 expression in human lung epithelial cells (A549). TGF-β1(1–5 ng/ml) treatment resulted in a marked time-dependent induction of HO-1 mRNA in A549 cells, followed by corresponding increases in HO-1 protein and HO enzymatic activity. Actinomycin D and cycloheximide inhibited TGF-β1-responsive HO-1 mRNA expression, indicating a requirement for transcription and de novo protein synthesis. Furthermore, TGF-β1 rapidly activated the p38 mitogen-activated protein kinase (p38 MAPK) pathway in A549 cells. A chemical inhibitor of p38 MAPK (SB-203580) abolished TGF-β1-inducible HO-1 mRNA expression. Both SB-203580 and expression of a dominant-negative mutant of p38 MAPK inhibited TGF-β1-induced ho-1 gene activation, as assayed by luciferase activity of an ho-1enhancer/luciferase fusion construct (pMHO1luc-33+SX2). These studies demonstrate the critical intermediacy of the p38 MAPK pathway in the regulation of HO-1 expression by TGF-β1.
- heme oxygenase
- mitogen-activated protein kinase
- oxidant stress
- transforming growth factor-β1
transforming growth factor-β1 (TGF-β1), a multifunctional cytokine, participates in numerous biological processes, including cell proliferation, differentiation, apoptosis, fibrosis, wound repair, and inflammation (25, 42). Recent evidence has suggested a fundamental role for TGF-β1 as a critical mediator of the tissue response to lung injury (4, 30,35). Elevated expression of TGF-β1 occurs in pulmonary epithelial cells during the progression of lung diseases, such as chronic obstructive airway disease, asthma, interstitial pulmonary fibrosis, and lung allograft rejection, which are disorders characterized by acute and chronic inflammation and cellular injury (7, 19, 37, 46). The mechanisms that govern these inflammatory responses and wound repair processes after lung injury remain incompletely understood. The critical role TGF-β1plays in inflammation and the resolution of tissue injury in multiple organs, including the lungs, has been well demonstrated by the observations that TGF-β1 null mice [TGF-β1(−/−)] have severe and generalized inflammatory disorders (20, 41). TGF-β1 can exert such protective effects to attenuate cellular injury and maintain tissue homeostasis in part by inducing cytoprotective proteins. Heme oxygenase (HO)-1 is an inducible protein activated in inflammatory conditions by oxidative stress that recently has fast gained notoriety as a key protective player in the cellular response to injury (33).
HO catalyzes the initial and rate-limiting step in the oxidative degradation of heme, generating equimolar amounts of biliverdin-IXα, carbon monoxide (CO), and ferrous iron (44). NAD(P)H-biliverdin reductase completes heme degradation by converting biliverdin-IXα to bilirubin-IXα, a lipid-soluble antioxidant. In addition to HO-1, the stress-inducible form, the HO system includes two constitutively expressed and genetically distinct isozymes (HO-2 and HO-3; see Ref. 28). The expression of HO-1 represents a ubiquitous stress response that confers protection against oxidative injury in cells and tissues. HO-1 expression responds to stimulation by a number of chemical and physical effectors, including its physiological substrate heme, as well as heavy metals, ultraviolet radiation, oxidants, nitric oxide, thiol-reactive substances, and hyperoxic or hypoxic states (1, 4, 24). Furthermore, HO-1 responds to a number of cytokines, hormones, and related substances, including proinflammatory mediators such as tumor necrosis factor-α (TNF-α), interleukin-1 (IL-1), and bacterial lipopolysaccharide, the thrombopoietic cytokine IL-11, and growth factors, such as platelet-derived growth factor and TGF-β1 (5, 16, 21, 24).
TGF-β1 exerts its diverse cellular effects via heteromeric interactions of type I and type II serine/threonine kinase receptors, which in turn activate downstream intracellular signaling pathways (26). Initiation of signaling requires binding of TGF-β1 to TGF-β type II receptor, a constitutively active serine/threonine kinase that subsequently transphosphorylates and activates the TGF-β type I receptor (TβR-I). Activated TβR-I in turn phosphorylates and activates the Smads, a family of structurally related proteins that represent the human analogs of theDrosophila protein MAD (Mothers Against Decapentaplegic) and the Caenorhabditis elegans protein SMA (Small body size; see Refs. 26 and 27). However, in addition to the Smad proteins, recent studies indicate that TGF-β1 signaling also involves the mitogen-activated protein kinase (MAPK) pathways, which either act independent of, or converge with, the Smad signaling pathways (9, 26, 38, 47, 50).
The MAPKs are a family of structurally related signal-transducing enzymes displaying a high degree of evolutionary conservation that connect cell surface receptors to critical regulatory targets within cells (6). MAPKs participate in the regulation of many essential cellular processes such as growth, differentiation, apoptosis, and adaptive responses to environmental stress (6, 40). The three major subgroups of the MAPK family identified to date include the extracellular signal-regulated kinases (ERK1/ERK2), the c-Jun NH2-terminal kinase (JNK), and the p38 MAPKs (12, 40). MAPK-dependent signal transduction involves a coordinated three-tiered cascade of protein kinase reactions that sequentially phosphorylate and activate the downstream kinase in their respective pathways. Our laboratory has recently reported the activation of ERK1/ERK2 by TGF-β1 in cultured macrophages and demonstrated a critical role of the ERK pathway in signaling the anti-apoptotic effects of TGF-β1 in these cells (10). We have also reported that TGF-β1 is capable of activating the ERK1/ERK2 and the p38 MAPK in glomerular mesangial cells (9). In both studies, the rapid kinetics of phosphorylation of ERK1/ERK2 and p38 MAPK after TGF-β1treatment indicate direct activation of the MAPKs by TGF-β1 (9, 10).
In the present study, we explored the hypothesis that TGF-β1 regulates the expression of HO-1, a potent cytoprotective molecule that can modulate inflammatory damage and wound repair to attenuate cellular injury, in human pulmonary epithelial cells, a major target cell type in a variety of lung injury. We show that TGF-β1 is capable of strong induction of HO-1 mRNA and protein and HO enzymatic activity in A549 cells. We also show the first demonstration that TGF-β1 rapidly activates the p38 MAPK in A549 cells. Furthermore, blockade of the p38 pathway by either specific chemical inhibitors or transfection of dominant-negative p38 MAPK mutant prevented the induction of HO-1 by TGF-β1, indicating the requirement of the p38 MAPK signal transduction pathway in the regulation of HO-1 expression by TGF-β1.
MATERIALS AND METHODS
Recombinant human TGF-β1 was obtained from GIBCO-BRL Life Technologies (Gaithersburg, MD). The polyclonal rabbit anti-rat HO-1 antibody was from Affinity Bioreagents (Golden, CO). The (anti)phospho-p38 MAPK (Thr180/Tyr182), p38 MAPK, phospho-p44/42 MAPK (Thr202/Tyr204), p44/42 MAPK, phospho stress-activated protein kinase (SAPK)/JNK (Thr183/Tyr185), SAPK/JNK, phospho-activating transcription factor 2 (ATF2) and ATF2 rabbit polyclonal antibodies, and the mitogen/extracellular signal-regulated kinase (MEK) inhibitors PD-098059 and U-0126 were purchased from Cell Signaling Technology, New England Biolabs (Beverly, MA). The specific inhibitor of p38 MAPK, SB-203580, was from Calbiochem (San Diego, CA). The transfection reagent Fugene6 was from Roche Molecular Biochemicals (Indianapolis, IN). The dual luciferase reporter system was from Promega (Madison, WI). Actinomycin D, cycloheximide, and all other reagent chemicals were obtained from Sigma (St. Louis, MO).
Human pulmonary epithelial cells, A549, derived from a lung carcinoma were obtained from American Type Culture Collection (ATCC, Manassas, VA). The cells were maintained in Ham's F-12 Nutrient Mixture Medium (GIBCO-BRL) supplemented with 10% FBS (HyClone) and gentamicin (50 μg/ml) in a humidified incubator containing an atmosphere of 5% CO2 and 95% air at 37°C. Cells grown to subconfluence were rendered quiescent in medium containing 0.5% FBS before the addition of human TGF-β1 (GIBCO-BRL). In experiments using inhibitors of transcription and translation, the cells were pretreated for 1 h with actinomycin D (5 μg/ml) or with cycloheximide (10 μg/ml) followed by stimulation with exogenous TGF-β1. For experiments using p38 and MEK inhibitors, cells were preincubated for 1 h with SB-203580, PD-098059, or U-0126 (10–30 μM) before treatment with exogenous TGF-β1. The concentrations used for the kinase inhibitors fall within the optimal range for inhibiting respective MAPK pathways without cytotoxicity.
HO enzyme activity assay.
For measurement of HO activity, cells were scraped in ice-cold PBS, centrifuged at 1,000 g, and resuspended in 0.25 M Tris · HCl (pH 7.4) homogenization buffer containing protease inhibitors (Roche), as previously described (22). Cell suspensions were sonicated on ice 2 × 15 s and centrifuged for 20 min at 18,000 g. The resulting supernatant was centrifuged at 100,000 g to isolate microsomal pellets. Protein concentrations were determined by Coomassie blue dye-binding assay (Bio-Rad, Hercules, CA). The HO activity was measured by the spectrophotometric determination of bilirubin production. Final reaction concentrations were 25 μM heme, 2 mM glucose 6-phosphate, 2 units glucose-6-phosphate dehydrogenase, 1 mM β-NADPH, 1 mg/ml microsomal extract, and 2 mg/ml partially purified bovine liver biliverdin reductase preparation. Reaction mixtures were incubated in the dark at 37°C for 60 min. The reactions were terminated by addition of chloroform (2:1 vol/vol). Bilirubin concentration in the chloroform extracts was determined on a Hitachi model U-2000 scanning spectrophotometer (Hitachi Instruments, St. Louis, MO) by measuring optical density (464–530 nm). HO activity was reported as picomoles bilirubin per milligram protein per hour, assuming an extinction coefficient of 40 mM−1 (cm−1 for bilirubin). Partially purified biliverdin reductase was prepared from bovine liver according to the protocol of Tenhunen et al. (43). Three independent experiments were performed, each with triplicate determinations. Statistical significance was determined by the Student's t-test for paired data points comparing treated with control samples. P values <0.01 were considered significant. Data are presented as means ± SE of triplicate determinations.
Northern blot analysis.
Total RNA was isolated by cell lysis with TRIzol (GIBCO-BRL) according to the manufacturer's instructions and size-fractionated (10 μg/lane) on a 1% agarose, 2% formaldehyde gel in 20 mM MOPS, 5 mM sodium acetate, and 1 mM EDTA, pH 7.2. mRNA was transferred and ultravioletly (UV) linked to a nylon membrane (Gene Screen Plus; Dupont). The blots were prehybridized for 2 h at 65°C in 1% BSA, 7% SDS, 1 mM EDTA, and 0.5 M phosphate buffer, pH 7.0, and hybridized overnight in the same solution containing the appropriate32P-labeled probe at 65°C. The blots were then washed two times in solution containing 0.5% BSA, 5% SDS, 1 mM EDTA, and 40 mM phosphate buffer, pH 7.0, for 30 min each at 65°C, followed by four 15-min washes with 1% SDS, 1 mM EDTA, and 40 mM phosphate buffer, pH 7.0, at 65°C. The blots were exposed to Kodak X-AR 5 film. The HO-1 cDNA probe was the previously described 0.9-kb cDNA fragment of the rat HO-1 cDNA labeled with [α-32P]dCTP using a random primer kit from Boehringer Mannheim (Indianapolis, IN) (22). To control for relative equivalence of RNA loading, the blots were hybridized with 18S rRNA oligonucleotide labeled with [α-32P]dATP at the 3′-end with terminal deoxynucleotidyltransferase (GIBCO-BRL).
Western blot analysis.
Total cellular extracts were obtained for the Western analyses by lysis of cells in buffer containing 1% Nonidet P-40, 150 mM NaCl, 20 mM Tris, pH 8.0, 1 mM Na3VO4, 1 mM phenylmethylsulfonyl fluoride, and 10 μg/ml aprotinin. Protein concentrations of the cell lysates were determined by Coomassie blue dye-binding assay (Bio-Rad). An equal volume of 2× SDS loading buffer (0.125 mM Tris · HCl, pH 7.4, 4% SDS, and 20% glycerol) was added, and the samples were boiled for 5 min. Protein samples (100 μg) were resolved on a 12% SDS-PAGE and then electroblotted on nitrocellulose membranes (Bio-Rad). The membranes were incubated with polyclonal anti-HO-1 antibody (1:500) or with phosphospecific and nonphosphospecific rabbit polyclonal antibodies to p38 MAPK, p44/42 MAPK, SAPK/JNK, or ATF2 (1:1,000) for 1.5 h, followed by incubation with horseradish peroxidase-conjugated anti-rabbit antibody (1:5,000) for 1.5 h. The filters were treated with LumiGLO (New England Biolabs) for signal development and then exposed to X-ray film. All gels were repeated in triplicate.
Transient transfection and luciferase activity assay.
A549 cells were seeded on 12-well plates (Falcon; BD Biosciences, Bedford, MA) at a density of 2 × 104 cells/well ∼24 h before the start of transfection. Cells were transfected with the luciferase reporter construct pMHO1luc-33+SX2 (1 μg/well plasmid DNA) in serum-free medium for 6 h using Fugene6 (Roche) according to the manufacturer's instructions. The cells were then restored to complete media and cultured for an additional 18 h. The luciferase reporter construct used in these studies was the plasmid pMHO1luc (kindly provided by Jawed Alam, Ochsner Medical Foundation), which contains the HO-1 minimal promoter sequences (33 bp) and theho-1 −4-kb distal enhancer fragment (268 bp), SX2. In the experiments involving cotransfection of a dominant-negative mutant of p38 (p38-DN), cells were transfected with 500 ng of pMHO1luc-33+SX2 and 500 ng of either p38-DN or empty vector pcDNA3 (3). The mutant p38 construct used was a kinase-deficient, dominant-negative mutant of p38 MAPK (kindly provided by Roger Davis, University of Massachusetts). Control reporter vector, pRL-TK (30 ng/well) (Promega), was included in all transfections as an internal control to normalize transfection efficiency. After transfection (24 h), cells were stimulated with exogenous TGF-β1 (10 ng/ml) for 6 h. Luciferase activities in cell lysates were determined using the Dual-Luciferase reporter assay system (Promega) and a TD-20/20 luminometer from Turner Designs (Sunnyvale, CA). The results are expressed as the ratio of measured relative light units of Firefly luciferase activity (pMHO1luc-33+SX2) to Renilla luciferase (pRL-TK) activity. Three independent experiments were performed, each with triplicate determinations. Statistical significance was determined by the Student's t-test for paired data points.P values <0.01 were considered significant. Data are presented as means ± SE of triplicate determinations.
TGF-β1 induces HO-1 mRNA, HO-1 protein expression, and HO activity in A549 cells.
The effects of TGF-β1 treatment on HO-1 mRNA expression were examined in cultured A549 cells by Northern blot analyses. As shown in Fig. 1 A, exposure to exogenous TGF-β1 (2 ng/ml) increased HO-1 mRNA levels in A549 cells in a time-dependent manner. Western blot analyses using polyclonal anti-HO-1 antibodies confirmed that the increase in HO-1 mRNA levels by TGF-β1 was accompanied by an increase in HO-1 protein expression (Fig. 1 B). Microsomal extracts from A549 cells treated with exogenous TGF-β1 (2 ng/ml) were assayed for HO activity by the spectrophotometric determination of bilirubin. Figure 1 C shows significant (2- to 3-fold) time-dependent increases in HO enzyme activity in A549 cells after exposure to exogenous TGF-β1 (2 ng/ml) compared with untreated controls (P < 0.01).
TGF-β1-mediated HO-1 induction is inhibited by actinomycin D and cycloheximide.
To determine if TGF-β1-induced HO-1 mRNA expression required transcription or translation, A549 cells were stimulated with exogenous TGF-β1 (2 ng/ml) in the absence or presence of actinomycin D or cycloheximide followed by measurement of HO-1 mRNA abundance by Northern blot analyses. As shown in Fig.2, TGF-β1-mediated induction of HO-1 mRNA was completely abolished in the presence of actinomycin D (5 μg/ml), a potent inhibitor of transcription. Moreover, treatment with cycloheximide (10 μg/ml), an inhibitor of translation, also markedly attenuated HO-1 mRNA induction by TGF-β1. Taken together, these findings demonstrate that the induction of HO-1 by TGF-β1 depends on de novo mRNA synthesis and partially on de novo protein synthesis.
TGF-β1 activates MAPK pathways in A549 cells.
Recent studies, including ours, have reported that TGF-β1exerts its biological effects via the MAPK signaling pathway in several cell culture systems (12, 40). The present study investigated the role of the MAPKs in TGF-β1-mediated HO-1 induction in A549 cells. The levels of protein expression of p38 MAPK, ERK1/ERK2, and JNK in A549 cells treated with exogenous TGF-β1 (2 ng/ml) were assessed by Western analyses, using the corresponding phosphospecific and nonphosphospecific antibodies. The phosphospecific antibodies detect the phosphorylated forms of p38 MAPK, ERK1/ERK2, and JNK, whereas the corresponding nonphosphospecific antibodies detect total protein, independent of phosphorylation state. Increases in the phosphorylation of p38 MAPK (Fig.3 A) and ERK1/ERK2 (Fig.3 B) were detected within 15–30 min of stimulation with exogenous TGF-β1. In contrast, TGF-β1treatment did not appreciably increase the activation of JNK within the same time period (Fig. 3 C). To further examine whether TGF-β1 can activate the downstream molecule of the p38 MAPK in A549 epithelial cells, we performed similar experiments using antibodies for phospho-ATF2. As shown in Fig. 3 D, we observed that TGF-β1 can activate ATF2 in a time-dependent manner.
SB-203580 inhibits TGF-β1-mediated HO-1 induction in A549 cells.
Given that TGF-β1 treatment activated both p38 MAPK and ERK1/ERK2 in A549 cells, the possibility that MAPK signaling pathway(s) mediated the induction of HO-1 expression by TGF-β was examined, using selective chemical inhibitors of the ERK and p38 MAPK pathways. SB-203580 (10–30 μM), a selective inhibitor of p38 MAPK, inhibited the induction of HO-1 mRNA by TGF-β1 in A549 cells (Fig. 4 A) in a dose-dependent fashion (Fig. 4 B). In contrast, equivalent concentrations of a specific MEK1 inhibitor (PD-098059) that prevents the activation of the ERK1/ERK2 pathway failed to inhibit TGF-β1-induced HO-1 mRNA expression (Fig. 4 A). Similarly, U-0126, an inhibitor of MEK1/MEK2, also failed to prevent TGF-β1-induced HO-1 mRNA expression (Fig. 4 A). Taken together, our findings implicate the p38 MAPK pathway, but not the ERK pathway, in mediating TGF-β1-induced HO-1 expression in A549 cells. U-0126, however, exerted a modest inductive effect on HO-1 mRNA expression by TGF-β1.
Inhibition of the p38 MAPK pathway downregulates TGF-β1-induced HO-1 promoter activity in A549 cells.
To confirm the involvement of the p38 MAPK in TGF-β1-induced HO-1 expression, the effects of p38 MAPK pathway inhibition on TGF-β1-inducible ho-1promoter activity were determined. The p38 MAPK pathway was blocked either by the selective chemical inhibition of p38 MAPK with SB-203580 or by genetic blockade with a dominant-negative mutant of p38 MAPK. HO-1 promoter activity was assessed by transfection of a luciferase reporter construct (pMHO1luc-33+SX2) containing the 33-bp HO-1 minimal promoter sequences and the 268-bp HO-1 enhancer SX2, previously identified as a mediator of ho-1 gene activation by a variety of stimuli (2, 3). As shown in Fig.5, treatment with exogenous TGF-β1 resulted in an over twofold increase in the luciferase activity in A549 cells transfected with pMHO1luc-33+SX2 compared with the control unstimulated cells. However, the presence of increasing concentrations of SB-203580 (10–30 μM) dose dependently inhibited TGF-β1-inducible luciferase activity. Figure 6 demonstrates that cotransfection of a kinase-deficient, dominant-negative mutant of p38 (p38-DN) with pMHO1luc-33+SX2 attenuated TGF-β1-inducible luciferase activity in A549 cells. Thus, in agreement with the findings with SB-203580, these results strongly suggest that the transcriptional regulation of the ho-1 gene by TGF-β1 requires signaling via the p38 MAPK pathway.
In response to injury, repair processes are initiated, and TGF-β1 plays a critical role in the inflammatory responses and the resolution of tissue injury in multiple organs, including the lungs (6). TGF-β1 may exert such protective effects that attenuate cellular injury and maintain tissue homeostasis in part by inducing cytoprotective proteins such as HO-1. In this study, we first determined that TGF-β1 is capable of strongly inducing the expression of HO-1 in human pulmonary epithelial cells (A549). We show that the levels of both HO-1 mRNA and HO-1 protein increased upon stimulation with exogenous TGF in A549 cells. We further confirmed that the TGF-β1-induced increases in HO-1 mRNA and protein were accompanied by corresponding increases in HO enzymatic activity in the microsomal extracts from A549 cells treated with exogenous TGF-β1, as assayed by the spectrophotometric determination of bilirubin generation. Thus our data provide clear evidence that TGF-β1 is a potent inducer of HO-1 in A549 pulmonary epithelial cells.
The stimulation of HO-1 expression by most inducers has been shown to occur primarily as a consequence of transcriptional regulation of theho-1 gene, and cis-acting DNA sequences involved in induction by various agents have been identified in theho-1 gene from several species (3, 11). Here we determined the ability of TGF-β1 to activate the transcription of an ho-1 promoter/luciferase reporter fusion gene containing the 33-bp HO-1 minimal promoter sequences and the 268-bp HO-1 enhancer SX2, previously identified as a mediator ofho-1 gene activation by a variety of stimuli in A549 cells (2, 3). Furthermore, in A549 cells, treatment with actinomycin D, which inhibits transcription, attenuated the ability of TGF-β1 to induce HO-1 mRNA. However, treatment with cycloheximide, a potent inhibitor of translation, also markedly reduced TGF-β1-induced HO-1 mRNA expression. Thus, in the A549 cells, a requirement exists for both de novo mRNA and protein synthesis for HO-1 induction by TGF-β1.
Although heme is the major substrate of HO-1, a wide variety of agents, including heavy metals, cytokines, hormones, endotoxin, and heat shock, has been identified as strong inducers of HO-1 expression (1, 5,24, 33). HO-1 is highly induced by various inflammatory stimulants and agents that cause oxidative stress, such as hydrogen peroxide, glutathione depletors, UV radiation, hyperoxia, and ischemia-reperfusion injury (5, 18, 22). The diversity of HO-1 inducers has provided support for the speculation that HO-1, besides its role in heme degradation, may also play a vital function is maintaining cellular homeostasis and act as a key mediator of cellular adaptation and/or defense against oxidative stress. A number of laboratories, including ours, have demonstrated that induction of endogenous HO-1 or exogenous administration of HO-1 via gene transfer provides protection against tissue injury and oxidative stress in many in vivo and in vitro models (5, 13, 22,34). Relative to the lung, we have previously reported that overexpression of HO-1 in pulmonary epithelial cells provided protection against hyperoxic injury (22). Moreover, we recently reported that exogenous administration of HO-1 by gene transfer also conferred protection against hyperoxia-induced lung injury in vivo (34).
The precise molecular mechanisms by which HO-1 confers cytoprotection remain unclear but likely relate to the biological activities of its enzymatic reaction byproducts, and there is ample evidence in the literature to support the role of its byproducts in imparting the cytoprotective properties of HO-1. For instance, HO-1-dependent effects on intracellular iron homeostasis, through augmentation of cellular iron efflux, have been described as an important mechanism controlling cell survival after stress (15). Biliverdin-IXα and its metabolite bilirubin-IXα have also been demonstrated to possess in vitro antioxidant and radical scavenging properties (24). More recently, considerable interest has been generated regarding the role of CO, the third major catalytic byproduct of HO-1 reaction. CO is a gaseous molecule released from heme degradation by HO-1 that can serve key physiological functions such as a chemical messenger in neuronal transmission and modulation of vasomotor tone (45,29). But, in addition, recent reports assert that, via its antioxidant and anti-apoptotic mechanisms, HO-derived CO protects against inflammatory injury in cardiac xenografts and that CO mediated the anti-inflammatory effects of IL-10 (23. 39). We have shown that the anti-inflammatory effects of HO-1 are mediated by CO-dependent downregulation of synthesis of proinflammatory cytokines (TNF-α, IL-1β, and MIP-1β), while increasing production of the anti-inflammatory cytokine IL-10, and that CO mediates these anti-inflammatory effects via the p38 MAPK pathway (32,33).
TGF-β1, like HO-1, possesses potent anti-inflammatory properties and performs a critical function to modulate the initiation, progression, and resolution of inflammatory responses. These TGF-β1 actions are best exemplified by the observations that TGF-β1 null mice [TGF-β1(−/−)] exhibit severe and generalized inflammatory disorders (20,41). These mice succumb to a multitissue inflammatory disease, produce autoimmune antibodies, and die in the first several weeks of life. Despite such dramatic evidence for anti-inflammatory actions of TGF-β1, the underlying mechanisms remain incompletely understood. Given the growing evidence for the anti-inflammatory and cytoprotective effects of HO-1, and our present findings that HO-1 is strongly induced by TGF-β1, it appears quite plausible that the anti-inflammatory effects of TGF-β1 are at least in part mediated through the induction of HO-1. Interestingly, similar to TGF-β1(−/−) null mice, HO-1 null mice [HO-1(−/−)] and the first reported case of HO-1-deficient humans have the phenotype of an increased inflammatory state (36,48).
It is becoming increasingly clear that the complex and multiple TGF-β1 actions and signaling mechanisms are cell-specific and context-specific. Kutty et al. (21) previously demonstrated that TGF-β1 induced HO-1 mRNA and protein levels in human retinal pigment epithelial cells and bovine choroid fibroblasts but not in HeLa, HEL, or bovine corneal fibroblasts. Studies of Hill-Kapturczak et al. (16) have suggested the involvement of Smad proteins in the TGF-β1-mediated activation of HO-1 in human proximal tubule cells, which could be antagonized by overexpression of an inhibitory Smad (Smad7). In our studies with pulmonary epithelial cells, we show evidence that TGF-β1 induces HO-1 through the MAPK signaling pathway, in particular the p38 MAPK. Our findings demonstrate that TGF-β1 induced the activation of both ERK1/ERK2 and p38 MAPK but not the JNK pathway in A549 cells. The rapid kinetics of their activation within 15–30 min of stimulation by TGF-β1suggest that these TGF-β1 effects are likely direct. Furthermore, inhibition studies using two distinct strategies to block the MAPK signaling pathways, first using specific chemical inhibitors of the MAPK pathways and second by dominant-negative inhibition of p38 MAPK, revealed that TGF-β1-mediated activation of HO-1 expression in A549 cells involved specifically the activation of the p38 MAPK signaling pathway.
Numerous other stimuli that induce HO-1 expression also have been reported to enhance the activity of MAPKs, which directly participate in signaling pathways that induce HO-1 gene expression. However, the complexity of the differential signaling pathways (e.g., MAPK) regulating HO-1 gene expression is further highlighted by cell- and inducer-specific effects. For example, the ERK1/ERK2 and p38 MAPK pathways mediated the sodium arsenite-dependent induction ofho-1 gene transcription in avian hepatoma cells (14) and in HeLa cells after exposure to nitric oxide donors (8). The p38 MAPK but not ERK1/ERK2 pathways mediated the regulation of ho-1 transcription by CdCl2 in MCF-7 cells (3) and by hypoxia in cardiomyocytes (17). On the other hand, the induction of HO-1 expression in rat liver by phorone, a glutathione-depleting agent, was associated with JNK activation (31). Finally, overexpression of the TGF-β-activated kinase induced ho-1transcription in hepatoma cells, which was dependent on the transcription factor Nrf2 (49).
In summary, our data provide strong evidence that TGF-β1is a potent inducer of ho-1 transcription, leading to corresponding increases in HO-1 mRNA and protein expression, as well as HO enzyme activity in pulmonary epithelial cells. Our present studies demonstrate that the p38 MAPK mediates TGF-β1-stimulated HO-1 induction. We propose that the p38 MAPK pathway may play a dual role in mediating both inducer (TGF-β1)-dependent activation of HO-1, and in turn, HO-1- and CO-dependent cytoprotection (32, 33).
This work was supported in part by National Institutes of Health (NIH) Grant R01 DK-57661-01; Grant-in-Aid no. 0051319T from the American Heart Association (AHA); and a Veterans Affairs Career Development Award to M. E. Choi. A. M. K. Choi was supported by NIH Grants R01 HL-55330, R01 HL-60234, and R01 AI-42365, and an AHA Established Investigator Award.
Address for reprint requests and other correspondence: A. M. K. Choi, Pulmonary, Allergy and Critical Care Medicine, MUH 628 NW, 3459 Fifth Ave., Univ. of Pittsburgh, Pittsburgh, PA 15213 (E-mail:).
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.
August 2, 2002;10.1152/ajplung.00151.2002
- Copyright © 2002 the American Physiological Society