Altered contractility of airway smooth muscle (SM) is one of the main causes of allergic asthma, in which the predominance of Th2 over Th1 cytokines plays a central role. In the present study, we examine the effects of Th2 cytokines on airway SM contraction. Treatment with a low concentration of IL-4 (0.2 ng/ml) for 6 h augmented, whereas higher concentrations (2–20 ng/ml) inhibited, agonist-induced contractions of collagen gels containing bovine tracheal SM cells. Another Th2 cytokine (IL-13) showed an augmentation of gel contraction in the concentration range of 20–200 ng/ml. IL-4 and IL-13 increased mRNA expression and protein secretion of matrix metalloproteinase (MMP)-1, but these cytokines did not affect Ca2+-mobilizing properties and phosphorylation levels of myosin L chain in bovine tracheal SM cells. These changes were sensitive to wortmannin, an inhibitor of PI3K, but not to leflunomide, an inhibitor of STAT6. Scanning electron microscope observation revealed that collagen fibers twining around SM cells were completely dissolved in 20 ng/ml IL-4-treated gels and reorganized into basket-like structure in 20 ng/ml IL-13-treated gels. Exogenous application of high and low concentrations of MMP-1 also induced the inhibition and augmentation of gel contraction, respectively. Furthermore, nonselective MMP inhibitor galardin suppressed the effects of IL-4 and IL-13 on gel contraction, and MMP-1-targeted small-interfering RNA reversed the inhibitory effects of IL-4 on gel contraction to the augmentation. This indicates that Th2 cytokines modulate airway contraction without affecting cellular contractility but by secreting MMP-1 from the SM cells via PI3K activation and changing cell-to-matrix interactions.
Airway hyperresponsiveness, an increased contractility of airway smooth muscles (SMs),3 plays a major role in the pathogenesis of bronchial asthma (1, 2, 3), and the predominance of Th2 over Th1 cytokines plays a central role in the allergic inflammation in bronchial asthma (4). Using receptor neutralization (5) and IL-4- and IL-13-knockout mice (6, 7, 8), it has been shown that these Th2 cytokines are implicated in the development of airway hyperresponsiveness. In nonhematopoietic cells, binding of IL-4 and IL-13 to their receptor subunits leads to heterodimerization with the other’s receptor subunit and activates the same intracellular signals, PI3K and STAT6 (9). It has therefore been suggested that STAT6 (10) and PI3K (11) might be involved in developing airway hyperresponsiveness.
Matrix metalloproteinase (MMP) family proteins are produced in various cells including SM cells and secreted into the extracellular space to alter the extracellular environment. Proinflammatory eicosanoid leukotriene D4 was shown to produce MMP-1, an interstitial collagenase (12), in cultured human airway SM cells (13), and MMP-1 was detected in asthmatic airway tissues with immunostaining (14). It has also been reported that MMP-1 mRNA was increased in the bronchial secretions of asthmatics, suggesting the involvement of MMP-1 in the pathogenesis of asthma (15). However, to our knowledge, the relationship between MMP-1 and Th2 cytokine pathogenesis and/or airway hypercontractility has not been established.
This study aimed to clarify the possible alteration of airway contraction by Th1/Th2 cytokines and its cellular mechanisms. We have previously shown that contractility of cultured bovine tracheal SM cells (BTSMCs) can be restored by embedding them into a collagen gel lattice (16). Because drugs can easily access the cells that are embedded into the collagen gel lattice, we used this technique in the present study to examine the effects of Th1 and Th2 cytokines on the contractility of BTSMCs. Our results suggest for the first time that IL-4 and IL-13 alter airway contraction without changing the intrinsic contractility of SM cells but by modulating the extracellular environment via MMP-1 secretion.
Materials and Methods
Cell culture and tissue preparation
Tracheas of 1-year-old calves were obtained from a local slaughterhouse, and SM cells were cultured in DMEM supplemented with 10% FBS by the explant method (17). Briefly, tracheal SM tissues were cut with scalpel blades into 1–2 mm3 pieces, attached to the bottom of culture dishes (100 × 20 mm), and cultured with 8 ml of culture medium at 37°C in 5% CO2 air. BTSMCs that were migrated out of the tissues were harvested after 2 wk by trypsin digestion and stored at −80°C after two-step subculture. The present study was performed using BTSMC obtained from 12 tracheas.
Bovine tracheal tissue extract was prepared by homogenizing tracheal SM tissues with Polytron (Kinematica) according to a previously described method (18), and used as a control for Western blot analysis of SM marker proteins. Bovine aortic endothelial cells (BAECs) were prepared as described previously (19), and used as a negative control for the same experiment.
Gel contraction assay
Contractility of cultured BTSMCs was examined with a gel contraction assay (20). Collagen solution was prepared by mixing 7 volumes of ice-cold type IA collagen (3 mg/ml) with 2 volumes of 5× concentrated DMEM and 1 volume of 200 mM HEPES solution (resulting in 20 mM HEPES buffer), and pH was adjusted to 7.4 with NaOH. BTSMCs suspension was centrifuged at 1000 rpm for 5 min and the pellet was resuspended in collagen solution at a density of 4 × 105 cells/ml; 0.5 ml of cell suspension per well was poured into a 24-well culture plate. The plate was kept at 37°C for 10 min to form a gel, and 1 ml of culture medium was then added. The gels were cultured for 3 days and used for the contraction assay after each pretreatment. The lateral surface of the gel was carefully detached from the culture well with a fine needle. The culture plate was then placed on a hotplate (HP-19U300; KPI) and kept at 37°C. The gel surface images were captured with a digital camera (QV-800SX; Casio) every 1 min throughout the experiment. Contraction of the gel was then evaluated by measuring its surface area with image analysis software (Adobe Photoshop; Adobe Systems). Because the degree of contraction in control gels varied between batches, data after each treatment were compared with the control data obtained in the same batch.
Measurement of intracellular calcium concentrations ([Ca2+]i)
[Ca2+]i was measured in nonconfluent BTSMCs with fura-2 by using an Attofluor digital fluorescence microscopy system (Atto Instruments), as previously described (21). For the statistical analysis of [Ca2+]i, results from 20 to 30 cells on a coverslip were averaged and treated as one data point.
Expressions of α-SM actin, SM myosin, calponin, myosin L chain (MLC), phosphorylated MLC (p-MLC), RhoA, STAT6, phosphorylated STAT6 (p-STAT6), and MMP-1 were assessed with ECL Western blotting. Whole cell lysates were prepared after each pretreatment and separated by electrophoresis except for the analysis of RhoA and MMP-1. For the assessment of the membrane translocation of RhoA activation, cell lysate was centrifuged for 1 h at 100,000 × g, and the pellet was harvested as membrane fraction. Extracellular secretion of MMP-1 was examined with harvested culture medium. Confluent BTSMCs were cultured for 6 h in serum-free DMEM with or without cytokines, and the culture medium was concentrated using an ultrafiltration membrane with a 30-kDa cutoff (Amicon Ultra; Millipore). Same amount of proteins (2 mg) was applied to SDS-PAGE.
Western blot analysis was then conducted with these samples by using relevant Ab. In each experiment, the bands were detected with a chemiluminescence system (SuperSignal West Dura; Pierce) and analyzed with a Lumino Image Analyzer (FAS-1000; Toyobo).
RT-PCR and real-time PCR analysis of MMPs and tissue inhibitor of metalloprotease 1 (TIMP-1) mRNA expression
Expressions of MMP-1, -2, -3, -9, -13, TIMP-1, and GAPDH mRNA were examined with RT-PCR. Confluent BTSMCs were treated with cytokines for 6 h or left untreated, and cellular total RNA was extracted using a commercial kit (RNAqueous 4PCR kit; Ambion) and converted to first-strand cDNA using reverse transcriptase (Superscript II; Invitrogen Life Technologies). Qualitative RT-PCR was then performed for 32 cycles with a thermal cycler (PC-8000; Astec) by using a commercial kit (Ready-To-Go RT-PCR beads; GE Healthcare Life Sciences). The resulting PCR products were analyzed with agarose gel electrophoresis, after which the cDNA bands were excised and extracted with a spin column (Quantum Prep; Bio-Rad) as standards for real-time PCR.
Real-time PCR was performed for a quantitative analysis of MMP-1, -2, -3, and TIMP-1 mRNA expression. First-strand cDNA was mixed with primers and a reaction reagent (Full Velocity SYBR Green QPCR Master Mix; Stratagene), and real-time PCR was performed with MX3000P (Stratagene) to obtain the threshold cycle numbers at which the amplified fluorescent PCR products become detectable. The threshold cycle values were then converted to the equivalent amount of template mRNA using standard curves obtained with extracted RT-PCR bands. Data were expressed as relative to the amount of GAPDH in the same volume of first-strand cDNA. The primers used for these measurements are listed in Table I.
|.||Forward Primer .||Reverse Primer .||Application .|
|MMP-2||5′-ctggtgtccagaaggtggat-3′||5′-taggcgcccttgaagaagta-3′||RT- and real time|
|MMP-3||5′-tgtgctcagcctatccactg-3′||5′-agctttcctgtcacctccaa-3′||RT- and real time|
|.||Forward Primer .||Reverse Primer .||Application .|
|MMP-2||5′-ctggtgtccagaaggtggat-3′||5′-taggcgcccttgaagaagta-3′||RT- and real time|
|MMP-3||5′-tgtgctcagcctatccactg-3′||5′-agctttcctgtcacctccaa-3′||RT- and real time|
Scanning electron microscopy
Microstructure of collagen fibers and embedded SM cells was observed with scanning electron microscope. Collagen gels were fixed with 2.5% glutaraldehyde and 2% paraformaldehyde, and freeze-dried. The samples were cut with a fine razor and the section was coated with osmium and observed with scanning electron microscope (JXA-8600MX; JEOL).
Gene silencing with small-interfering RNA (siRNA)
The following siRNA sequences were used to target MMP-1: 5′-CAGCAAUUUCCAAGAUUAUAACUTT-3′ (sense) and 5′-AGUUAUAAUCUUGGAAAUUGCUGTT-3′ (antisense). BTSMCs (3 × 105 cells in 600 μl) were transfected with siRNAs (final concentration 30 nM) by electroporation (350V, 975μF) using Gene Pulser XCell (Bio-Rad). Control cells were also electroporated without siRNAs. Cells were then cultured on plates for 2 days for real-time PCR or embedded into collagen gels for 3 days for gel contraction assay.
Krebs’s solution was used in the gel contraction assay and Ca2+ measurements contained (in millimoles): NaCl 132.4, KCl 5.9, CaCl2 1.5, MgCl2 1.2, glucose 11.5, HEPES 11.5, and pH was adjusted to 7.4 by NaOH.
Anti-MLC (clone MY-21), anti-β-actin (clone AC-15), anti-calponin (clone hCP), anti-SM myosin (clone HSM-V), and anti-MMP-1 Abs were purchased from Sigma-Aldrich. Anti-α-SM actin (clone 1A4) was purchased from DakoCytomation. Anti-p-MLC (Thr18/Ser19), anti-STAT6, and anti-p-STAT6 (Tyr641) Abs were purchased from Cell Signaling Technology. Anti-RhoA Ab was purchased from Cytoskeleton. MMP-1 was purchased from Biomol International. Collagen type IA was purchased from Nitta Gelatin. Fura-2/AM was purchased from Wako Pure Chemicals. Human recombinant IL-4 and IL-13, and all other reagents were purchased from Sigma-Aldrich.
Data were expressed as mean ± SEM values. Statistical significance in gel contraction assay was examined with repeated measures ANOVA with the Bonferroni post-hoc test, using data points from 10 to 70 min, and that in [Ca2+]i assay, real-time PCR, and Western blotting with the Student unpaired t test, by using StatView (SAS Institute) for both analyses. Probability below 0.05 (p < 0.05) was considered as a significant difference.
Expression of SM marker proteins in cultured BTSMCs
First, we examined the expression of SM marker proteins, i.e., α-SM actin, SM myosin, and calponin, in BTSMCs. Although lower than tissue extract, BTSMCs showed a significant amount of expression of these proteins compared with endothelial cells (Fig. 1). Thus, we consider that BTSMCs used in the present study sufficiently retained SM nature.
Effects of IL-4, IL-13, and IFN-γ on the contraction of BTSMCs
Next, we examined the effects of Th2 and Th1 cytokines on the contractility of BTSMCs. ATP (10 μM) induced a reversible contraction of untreated control collagen gels that contained BTSMCs (Fig. 2 A). A treatment of the gels with IL-4 (20 ng/ml) for 6 h resulted in an inhibition, whereas a treatment with the same concentration of IL-13 (20 ng/ml) for 6 h induced a marked augmentation of the ATP-induced gel contraction. The Th1 cytokine IFN-γ (100 U/ml), however, did not affect the ATP-induced gel contraction. A similar degree of IL-4 (20 ng/ml)-induced inhibition of gel contraction was observed after 1 and 24 h of treatment (data not shown). In contrast, IL-13 (20 ng/ml) induced a less potent augmentation of gel contraction after 1 and 24 h of treatment (data not shown).
IL-4 showed a less potent inhibition of gel contraction with a lower concentration (2 ng/ml), whereas it augmented the ATP-induced gel contraction with a much lower concentration (0.2 ng/ml; Fig. 2,B). In contrast, a lower concentration of IL-13 (2 ng/ml) did not alter ATP-induced gel contraction, and with a higher concentration (200 ng/ml) IL-13 augmented the contraction to a significantly lower degree than that with 20 ng/ml (Fig. 2 C). Simultaneous treatment with 20 ng/ml IL-4 and IL-13 for 6 h inhibited the ATP-induced gel contractions, and the degree of inhibition was to the same extent as in gels treated with 20 ng/ml IL-4 alone. In the following experiments, we treated BTSMCs and the gels for 6 h with 20 ng/ml IL-4 and 20 ng/ml IL-13 to explore the cellular mechanisms of the inhibition and the augmentation of gel contraction, respectively.
IL-4 and IL-13 do not affect Ca2+-mobilizing properties in BTSMCs
Because muscle contraction is triggered by an elevation of [Ca2+]i, we then examined the effects of IL-4, IL-13, and IFN-γ on ATP-induced Ca2+ transients. Treatment of BTSMCs with IL-4 (20 ng/ml), IL-13 (20 ng/ml), and IFN-γ (100 U/ml) for 6 h did not alter their basal level of [Ca2+]i (Fig. 3,Aa). Furthermore, the peak increase of [Ca2+]i during ATP-induced Ca2+ transients was not significantly different between control and IL-4 (20 ng/ml)-, IL-13 (20 ng/ml)-, or IFN-γ (100 U/ml)-treated cells (Fig. 3,Ab). Also, the time course and shape of the Ca2+ traces induced by 10 μM ATP in IL-4- and IL-13-treated BTSMCs were comparable to those in control cells (Fig. 3 A, c–e).
IL-4 and IL-13 do not affect MLC phosphorylation and RhoA activation in BTSMCs
We then examined the effects of 20 ng/ml IL-4 and 20 ng/ml IL-13 on MLC phosphorylation, which leads to contraction of SM cells (22). ATP increased the amount of p-MLC not only in control BTSMCs, but also in IL-4- and IL-13-treated cells, and the time course and maximal level of phosphorylation under these conditions were not significantly different from control (Fig. 3,B). Furthermore, activation of the RhoA/Rho-kinase pathway, which increases the Ca2+ sensitivity of the contractile proteins in airway SM cells (16), is unlikely because neither IL-13 nor IL-4 affected the membrane translocation of RhoA, which is considered as a hallmark of RhoA activation (23) (Fig. 3 C).
Wortmannin, but not leflunomide, suppressed IL-4- and IL-13-induced alteration of gel contraction
Next, we examined the effects of the inhibitors of the intracellular signals that are activated by IL-4 and IL-13 on gel contraction. Pretreatment of the BTSMC-embedded gels with wortmannin (1 μM), a PI3K inhibitor (24), did not affect the control gel contraction, but reduced the IL-4-induced decrease and the IL-13-induced increase in gel contractions in a concentration-dependent manner (Fig. 4 A).
IL-4 and IL-13 increased MMP-1 mRNA and protein expression in BTSMCs
Because IL-4 and IL-13 did not alter [Ca2+]i and MLC phosphorylation of BTSMCs, next we examined the possible alteration of the relationship between SM cells and extracellular matrix.
RT-PCR revealed that BTSMCs expressed mRNAs of MMP-1, -2, -3, and TIMP-1 but not MMP-9 and MMP-13 (Fig. 5,Aa). Absence of MMP-9 and MMP-13 mRNA expression was confirmed with two other sets of primers (data not shown), and the expression of these mRNAs was not evoked by IL-4 or IL-13 (Fig. 5,Aa). A quantitative analysis of MMP-1, -2, -3, and TIMP-1 mRNAs with real-time PCR revealed that the expression of MMP-1 mRNA was significantly increased by a treatment for 6 h with IL-4 (20 ng/ml) and IL-13 (20 ng/ml) but not by IFN-γ (100 U/ml). IL-4 and IL-13, however, did not alter the expressions of MMP-2, -3, and TIMP-1 mRNA (Fig. 5 Ab).
Because MMP-1 protein synthesized in the cells is secreted into the extracellular space (12), we also estimated the amount of MMP-1 protein secreted into the culture medium in control, IL-4-, IL-13-, and IFN-γ-treated BTSMCs. As shown in Fig. 5,B, MMP-1 protein in the culture medium was markedly increased by 20 ng/ml IL-4. IL-13 (20 ng/ml) also significantly increased the amount of MMP-1 in the culture medium, but to a lesser extent than in IL-4-treated medium. IFN-γ did not increase the amount of MMP-1 protein secretion. As in case of the gel contraction assay, 1 μM wortmannin, but not 100 μM leflunomide, reversed the IL-4- and IL-13-induced MMP-1 secretion (Fig. 5 B).
Scanning electron microscopic observations of BTSMC-embedded collagen gels
Scanning electron microscopic observations of control gels containing BTSMCs revealed a dense network of collagen fibers twined around the SM cells (Fig. 6,A). After a pretreatment with 20 ng/ml IL-4 for 6 h, these collagen fibers around the cells were almost completely removed (Fig. 6,B). In the 20 ng/ml IL-13-treated gels, this meshwork was partially dissolved, and the collagen fibers became reorganized into basket-like structures surrounding the SM cells (Fig. 6 C).
Effects of exogenous MMP-1 on collagen fibers and gel contraction
The results above indicate the central role of MMP-1 in the modulation of contraction of BTSMC-embedded gel. We then examined the effects of exogenous MMP-1 on the ultrastructure of collagen fibers and gel contraction.
Scanning electron microscopic observation revealed that collagen fibers were fine in untreated gels, and were fused and reorganized into thick fibers in 10 ng/ml MMP-1-treated gels (Fig. 7 A). Collagen fibers became further thicker and coarse after the treatment with 100 ng/ml MMP-1 (data not shown).
Effects of MMP inhibitor and MMP-1 gene silencing on gel contraction
We finally examined the effects of MMP inhibitor and MMP-1-targeted siRNA on IL-4- and IL-13-induced changes in gel contraction.
Galardin, a nonspecific MMP inhibitor (26), reversed the IL-4-induced inhibition and IL-13-induced augmentation of gel contraction, respectively, in a concentration-dependent manner (Fig. 8 A). Galardin (10 nM), however, did not affect the control gel contraction (data not shown).
Transfection of MMP-1-targeted siRNA significantly inhibited the expression of MMP-1 mRNA in control, IL-4-treated and IL-13-treated BTSMCs (Fig. 8,Ba). In IL-4-treated cells, siRNA inhibited MMP-1 mRNA to the similar level as that in IL-13-treated control cells. As expected from the amount of MMP-1 mRNA, the ATP-induced contraction of the gels containing siRNA-transfected BTSMCs was significantly augmented by IL-4, and was not affected significantly by IL-13 (Fig. 8 Bb).
We have shown in this study that Th2 cytokines show dual actions on the ATP-induced contraction of BTSMC-embedded collagen gels (Fig. 2,A). IL-4 augmented the gel contraction at 0.2 ng/ml and inhibited it at 2 and 20 ng/ml (Fig. 2,B). IL-13 only induced an augmentation of gel contractions (Fig. 2,C), but the effects were less potent at higher concentrations (200 ng/ml). Furthermore, simultaneous application of IL-4 and IL-13 did not lead to the summation of their solo effects, but inhibited the gel contraction as in the case of IL-4 alone (Fig. 2 D). It is known that the IL-4Rα chain binds IL-4 and dimerizes with IL-13Rα1 to form a type II IL-4R in nonhematopoietic cells (27). IL-13 binds to IL-13Rα1 and induces heterodimerization with IL-4Rα to form a complex identical with the type II IL-4R (28). IL-13 also binds to IL-13Rα2, a decoy receptor, with greater affinity than IL-13Rα1 but this does not induce a signal (29), and from this point of view quantitative rather than qualitative differences may cause the differing functional responses of IL-4 and IL-13. Therefore, we hypothesize that the limited activation of the IL-4/13 receptors on BTSMCs augments the gel contraction, and that the inhibition is caused by the more potent receptor activation.
Previous studies showed that IL-4 (50 ng/ml) inhibited (30) and IL-13 (50 ng/ml) augmented (31) agonist-induced Ca2+ transients, but we could not detect significant differences in the ATP-induced Ca2+ signals in control, IL-4-treated, and IL-13-treated BTSMCs (Fig. 3,A). Furthermore, the ATP-induced MLC phosphorylation was not affected by IL-4 or IL-13 (Fig. 3,B). Activation of RhoA, which would inhibit myosin phosphatase and thereby increase muscle contractility (32), is also unlikely, because we failed to detect an increased membrane translocation of RhoA by IL-4 and IL-13 (Fig. 3 C). Therefore, the present results strongly indicate that inhibition and augmentation of BTSMC-gel contractions induced by IL-4 and IL-13 were not due to an altered intrinsic contractility of the embedded SM cells.
Binding of IL-4 and/or IL-13 to the type II IL-4R leads to the activation of PI3K and STAT6 as the two major intracellular immediate signals (9). STAT6 phosphorylation in IL-4-treated cells was not different from that in IL-13-treated cells (Fig. 4,B), and leflunomide did not alter the actions of these cytokines on gel contraction (Fig. 4,C). So, we concluded that STAT6 is not responsible for the opposite effects of IL-4 and IL-13 on gel contraction. In contrast, wortmannin inhibited the effects of both IL-4 and IL-13 on gel contractions (Fig. 4,A). Several recent reports on the role of PI3K activation in migration and proliferation of stimulated airway SM cells (33, 34)—and also in Th2 cytokine secretion, airway inflammation, and airway hyperresponsiveness in asthma model mice (11, 35)—point to the possible involvement of PI3K in the pathogenesis of asthma. However, these studies did not provide more detailed cellular mechanisms of these PI3K-induced changes (33, 34, 35). IL-4 and IL-13 increased the expression level of mRNA and the protein secretion of MMP-1 in BTSMCs (Fig. 5). The increased secretion of MMP-1 protein was also inhibited by the PI3K inhibitor wortmannin, but not by leflunomide, which shows for the first time that MMP-1 secretion might be involved in the IL-4- and IL-13-induced and PI3K-mediated pathogenesis of airway SM cells. MMP-1 secretion induced by IL-4 was significantly larger than that by IL-13 (Fig. 5,B), which correlates with the observation that IL-4 showed stronger effects on gel contraction than IL-13 (Fig. 2 D). A similar increased expression level of MMP-1 by activation of PI3K has been observed in vascular endothelial cells and prostate cancer cells stimulated with FGF-1 (36) and integrin α5β1 ligand (37), respectively.
Because MMP-1 functions as a collagenase, the enhanced MMP-1 secretion in the presence of IL-4 or IL-13 probably accounts for the degradation of the dense collagen fiber networks twining around SM cells observed in control gels (Fig. 6). It is conceivable that removal of collagen fibers from the SM cell surface as observed in the presence of IL-4 might detach these SM cells from the extracellular collagen fibers, which would reduce the gel contraction even if the cellular contractility remains intact. This degradation of the collagen network might also provide space for the cells to migrate and/or proliferate, so that MMP-1 production may be partially responsible for the development of IL-4-induced airway remodeling (38). In contrast, the basket-like redistribution of collagen fibers surrounding the cells in IL-13-treated gels might induce a more efficient gel contraction than in the control gels with an intact collagen network. Basket-like structure in IL-13-treated gels probably resulted from the partial degradation and redistribution of collagen fibers, as was observed with the exogenous application of 10 ng/ml MMP-1 (Fig. 7,A). Furthermore, exogenous MMP-1 (10 ng/ml) augmented the ATP-induced gel contraction as in the case of IL-13-treated gels, and a much higher concentration of MMP-1 (100 ng/ml) reduced the gel contraction (Fig. 7,B). Inhibition of MMP with galardin (26) reversed the effects of IL-4 and IL-13 (Fig. 8,A), and partial inhibition of MMP-1 mRNA expression with gene silencing reversed the IL-4-induced inhibition of gel contraction to the augmentation (Fig. 8 B). Therefore, we have concluded that the inhibitory and augmentative effects of Th2 cytokines on the contraction of BTSMC-containing gels were due to MMP-1-induced removal and reorganization of extracellular collagen networks, respectively.
There are some marked histological differences between airway SM tissue and other SM tissues, such as blood vessels; i.e., airway SM cells are arranged into bundles separated by collagen and connective tissue cells, and are loosely connected to each other (39). These histological characteristics of airway SM are conserved in BTSMC-embedded collagen gels, i.e., randomly oriented and not connected to each other, as demonstrated previously (16). Recently, a role of the interrelationship between airway SM cells and extracellular matrix in airway diseases has been discussed (40). Our results with the model gel system are consistent with such a contention as they provide some evidence that MMP-1 production and the ensuing collagen redistribution represent a potential therapeutic target for bronchial asthma. The increased contractility induced by IL-13 has been reported previously as a main pathogenic factor of bronchial asthma (7, 41), and spontaneous contraction of airway mesenchymal cell-embedded gels as a model of airway remodeling was also increased by IL-4 and IL-13 (42). However, so far, no studies have suggested MMP-1 as a cause of hyperresponsiveness. MMPs have been proposed as a potential cause of airway remodeling (43), but the present study has clearly demonstrated that MMP-1 production might also contribute to altered contractions induced by Th2 cytokines. Thus, although Ca2+ mobilization and the phosphorylation state of MLC are the main determinants of the contractile state of SM cells (22, 44), it is likely that the interaction between SM cells and the extracellular collagen network is another important regulatory factor of the contractile response especially in allergic situations.
In conclusion, we have shown in the present study that IL-4 and IL-13 affect airway SM gel contraction via a PI3K-mediated endogenous production of MMP-1 that alters the interaction between SM cells and collagen fibers.
We thank Dr. Guy Droogmans for critical reading and comments on the manuscript.
The authors have no financial conflict of interest.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
This work was supported in part by a Grant-In-Aid from the Japan Society for the Promotion of Science (No. 18390075).
Abbreviations used in this paper: SM, smooth muscle; MMP, matrix metalloproteinase; BTSMC, bovine tracheal SM cell; BAEC, bovine aortic endothelial cell; MLC, myosin L chain; [Ca2+]i, intracellular calcium concentration; MLC, myosin L chain; TIMP, tissue inhibitor of metalloprotease 1; siRNA, small-interfering RNA.