There is evidence that interleukin-4 (IL-4) plays a major role in the induction of extracellular matrix protein synthesis in fibrotic disease. We therefore examined the effect of IL-4 on collagen synthesis in primary fibroblasts isolated from normal and TSK/+ mice, which spontaneously develop a scleroderma-like syndrome characterized by diffuse cutaneous hyperplasia.
Expression of the IL-4 receptor was determined by flow cytometry and Western blotting. The IL-4 signal transduction cascade was analyzed by Western blotting. We assessed the role of signal transducer and activator of transcription 6 (STAT-6) in IL-4 induction of α2(I) collagen promoter activity and message levels via luciferase reporter assay and real-time polymerase chain reaction. The activation status of the transcription factors activator protein 1 (AP-1) and Sp-1 upon stimulation with IL-4 in normal and TSK/+ fibroblasts was examined by electrophoretic mobility shift assay.
Flow cytometry and Western blotting showed that IL-4 receptor α expression was elevated in TSK/+ fibroblasts compared with normal fibroblasts. After IL-4 stimulation, janus-activated kinase 1 (JAK-1) and JAK-2 were phosphorylated to a greater degree in TSK/+ fibroblasts than in C57BL/6 fibroblasts. TSK/+ fibroblasts appeared to be hyperresponsive to IL-4, displaying increased synthesis of α1(I) collagen messenger RNA (mRNA), collagen protein, and activity of a luciferase reporter construct containing the −300 to +54 murine α2(I) collagen promoter. Overexpression of STAT-6 enhanced this effect, whereas expression of a dominant-negative STAT-6 abrogated the ability of IL-4 to induce α1(I) collagen mRNA in TSK/+ fibroblasts. Moreover, IL-4 induced increased DNA binding activity of transcription factors that are important for collagen synthesis.
Our observations indicate that IL-4 has a profound effect on several factors that have been identified as playing major roles in the regulation of collagen synthesis and suggest that IL-4 increases the expression of type I collagen through a mechanism involving the activation of transcription factors that bind to and activate collagen promoter.
Fibrosis is the major feature of many inflammatory disorders characterized by T cell infiltration and represents the end point of pathogenic events in systemic sclerosis (SSc), an autoimmune disease defined by limited-to-diffuse fibrosis of the connective tissues, autoimmunity, and vascular damage (1). While the causes of fibrosis remain unclear, it is evident that several cytokines and growth factors may play an important role in the development of fibrotic disease.
Interleukin-4 (IL-4) is a cytokine generally associated with immunoglobulin class switching, B cell activation, and T cell polarization to the Th2 phenotype. However, recent data suggest that it may be a major factor in the fibrotic process. For example, increased serum levels of IL-4 (2, 3) and IL-4–producing T cells were noted in patients with SSc (4) as well as in mice with bleomycin-induced skin sclerosis and lung fibrosis (5). In addition, fibroblasts have been shown to respond to IL-4 with increased synthesis of collagen (6, 7), sulfated glycosaminoglycans, and decorin (8). Interestingly, a study by Oriente et al (9) suggests that fibroblasts from fibrotic lesions have a heightened response to IL-4 and that the activated state induced by the cytokine is of longer duration, demonstrating that IL-4 induced collagen synthesis in keloid fibroblasts more rapidly and at greater magnitude than in normal fibroblasts. This conclusion is supported by the observation that fibroblasts from SSc patients are more sensitive to stimulation with IL-4 than are fibroblasts from healthy subjects, suggesting a hyperresponsiveness to the cytokine as a general consequence or cause of fibrotic disease (10).
A strong correlation between IL-4 and fibrosis has been demonstrated in several animal models. In transgenic mice, overexpression of IL-4 under control of the insulin promoter results in pancreatic fibrosis (11), and transgenic expression of the gene results in high levels of IL-4 and increased amounts of dermal collagen (12). Treatment with anti–IL-4 antibodies blocks fibrosis in the TSK mouse model of fibrotic disease (13). In addition, it was demonstrated that the disruption of IL-4 (14, 15), IL-4 receptor α (IL-4Rα) (16), or signal transducer and activator of transcription 6 (STAT-6) (14) genes prevents skin fibrosis in TSK/+ mice.
IL-4 stimulation activates a series of protein kinases via tyrosine phosphorylation. The phosphorylation is achieved through the activation of the janus-activated kinase (JAK) family of kinases that associate with the IL-4 receptor and become activated and phosphorylated after the binding of IL-4Rα to its cognate ligand. These phosphorylated proteins include the IL-4Rα chain, JAK-1, JAK-3, phosphatidylinositol 3-kinase (PI 3-kinase), and STAT-6 (17, 18). Although in murine fibroblasts, the phosphorylation of JAK kinases has not been described, in human fibroblasts, JAK-1 is activated after stimulation with IL-4 or IL-13 (19).
While much is known about the mechanisms of IL-4 signaling in lymphocytes (for review, see ref. 20), there is little information concerning the mechanisms through which it increases collagen synthesis in fibroblasts. Recently, Doucet et al (21) reported unusual IL-4 and IL-13 signaling in human normal and tumor lung fibroblasts. They found that both cytokines activate insulin receptor substrate 2, JAK-1, and STAT-6, whereas Tyk-2 was constitutively phosphorylated. Murata et al (19) reported that while IL-4 induces the phosphorylation of JAK-1, JAK-2, and STAT-6 in human skin fibroblasts, JAK-3 was present but not phosphorylated. However, in those studies, the activation of IL-4 signaling events in human fibroblasts was not correlated with the activation of transcription factors that activate the expression of collagen genes or with the effect on collagen synthesis.
The goals of the present study were to examine the effects of IL-4 stimulation on transcription factors that are known to play a role in the regulation of collagen expression and to more closely examine the role of STAT-6 in IL-4–induced up-regulation of collagen synthesis. We found important differences between control and TSK fibroblasts. TSK/+ fibroblasts expressed higher levels of IL-4Rα compared with control C57BL/6 fibroblasts. While JAK-1 and JAK-2 were constitutively phosphorylated in TSK fibroblasts, they were phosphorylated in normal fibroblasts only after IL-4 stimulation. In addition, we found that IL-4 increased α2(I) collagen promoter activity, collagen gene transcription, and collagen protein synthesis. Likewise, overexpression of STAT-6 further increased collagen promoter activity and gene transcription. Electrophoretic mobility shift assay (EMSA) of transcription factors showed that IL-4 induced increased DNA binding activity of STAT-6 as well as activator protein 1 (AP-1) and Sp-1, transcription factors that are known to activate type I collagen promoter and collagen synthesis. Consistent with the observation of hyperresponsiveness by TSK/+ fibroblasts to IL-4, these factors appeared to exhibit elevated activity in fibroblasts isolated from animals with fibrotic disease. These data demonstrate the ability of IL-4 to enhance fibrosis with induction of collagen gene activity, and they suggest that the fibrosis observed in TSK/+ mice may be due to an overly reactive response to IL-4 that is present in the extracellular microenvironment.
MATERIALS AND METHODS
Primary fibroblast lines were obtained from C57BL/6 and TSK/+ mice as previously described (22). The fibroblasts exhibiting the TSK/+ mutation, namely, the partially duplicated gene, were genotyped by polymerase chain reaction (PCR), as previously described (23), using the forward primer of exon 40 and the reverse primer of intron 17, which cover the break-point on chromosome 2 of duplication of exons 17–40 in the fibrillin 1 gene. Primary fibroblasts were grown in 35-mm2 plates in Dulbecco's modified Eagle's medium (DMEM) supplemented with 100 units/ml of penicillin, 100 μg/ml of streptomycin, and 10% fetal calf serum (FCS; Mediatech, Herndon, VA).
Flow cytometry analysis.
Cells were scraped off the surface, washed once with phosphate buffered saline (PBS), and preincubated with or without IL-4 (1 μg/ml) on ice for 30 minutes in fluorescence-activated cell sorter (FACS) buffer (PBS with 3% FCS), followed by staining with rat anti–IL-4Rα monoclonal antibody (M1; Immunex, Seattle, WA) for another 30 minutes. The cells were then washed 3 times in FACS buffer and incubated with biotin-labeled anti-rat IgG. After further washing, cells were stained with streptavidin–phycoerythrin. Samples were analyzed on a FACScan (Becton Dickinson, Mountain View, CA) according to standard techniques.
Stimulation of fibroblasts with IL-4.
TSK/+ and C57BL/6 fibroblasts were serum starved for 24 hours and then stimulated (or not stimulated) with 400 units/ml of murine recombinant IL-4 (R&D Systems, Minneapolis, MN). This dose of IL-4 was determined in pilot experiments to be an optimum dose for the stimulation of collagen synthesis.
Fibroblasts were grown to confluence in 75-cm2 flasks, detached via trypsinization, and plated at a density of 105 cells/well in 24-well plates. At 24 hours after plating, the fibroblasts were washed with PBS, and the media were replaced with proline-free DMEM supplemented with 3 μCi/well of 3H-proline (Amersham, Arlington Heights, IL), 50 μg/ml of ascorbic acid, 50 μg/ml of β-aminopropionitrile, and 400 units/ml of IL-4. After a 24-hour incubation, the media were collected for determination of collagen synthesis, and the cell numbers were counted.
To measure collagen synthesis, 100-μl aliquots of media from labeled fibroblasts were incubated at 37°C for 18 hours in the presence or absence of highly purified bacterial collagenase (Sigma, St. Louis, MO) at a concentration of 200 units/aliquot. The samples were then extensively dialyzed against PBS at 4°C in microdialysis chambers (Pierce, Rockford, IL) to remove unincorporated 3H-proline and digested collagen. The total counts per minute per sample were then determined using a liquid scintillation counter. Biosynthetic labeling of collagen was estimated by subtracting the cpm of the collagenase-digested aliquot from the cpm of the aliquot without collagenase.
Reverse transcription–PCR (RT-PCR) and messenger RNA (mRNA) quantification.
Confluent cultures of fibroblasts in 60-mm2 plates were serum starved for 12 hours and then stimulated with 400 units/ml of IL-4 for 12 hours. Then, mRNA was isolated from the cultures using an mRNA isolation kit (Boehringer Mannheim, Indianapolis, IN) according to the manufacturer's directions. The isolated mRNA was quantified via spectrophotometer, and 1 μg of mRNA was subjected to RT-PCR. The forward primer for amplification of α1(I) collagen mRNA was 5′-AAAGGCTGGAGAGCGA-3′, and the reverse primer was 5′-AGCAGGACCTGGGGGA-3′. The forward primer for the amplification of murine β-actin was 5′-ATGGATGACGATATCGCT-3′, and the reverse primer 5′-ATGAGGTAGTCTGTCAGGT-3′.
The RT-PCR reaction was performed using a SYBR Green RT-PCR kit and a LightCycler PCR machine (both from Boehringer Mannheim) according to the manufacturer's directions. Briefly, the cycle conditions were 1 cycle at 55°C for 10 minutes, then 45 cycles at 95°C for 10 seconds, 50°C for 5 seconds, and 72°C for 12 seconds. The α1(I) collagen mRNA was quantified using the LightCycler analysis software (Boehringer Mannheim) according to the manufacturer's directions. The PCR products for α1(I) collagen and β-actin were analyzed for specificity by agarose gel electrophoresis. The amount of α1(I) collagen was normalized to β-actin in the sample.
Generation of FLAG-tagged STAT-6– and STAT-6 phosphorylation–deficient protein–expressing plasmids.
PCR was performed on 0.3 μg of pcDNA3 containing complementary DNA (cDNA) for murine STAT-6 (a gift from Dr. Curt Horvath, Mount Sinai School of Medicine, New York, NY) utilizing a mixture of Taq and Pfu II polymerase (Boehringer Mannheim) with the following cycle conditions: 95°C for 5 minutes, then 30 cycles at 95°C for 30 seconds, 45°C for 30 seconds, and 72°C for 1.5 minutes, followed by a 10-minute extension at 72°C. For the initial PCR, 2 reactions were cycled. The first contained the forward primer 5′-CTGGGGGACCGGATCCGGCAT-3′ and the reverse primer 5′-ATGTAGAGACAAAACCCCTCC-3′, generating a 2,000-bp fragment. The second reaction contained the forward primer 5′-CGGGAGGGGTTTTGTCTCTACT-3′ and the reverse primer 5′-CAGGAGAAGCTTTCATTTATCGTCATCGTCTTTGTAGTCGGTGAGGTCCTG-3′, generating a 600-bp fragment. The fragments were resolved by agarose gel electrophoresis and purified using a gel extraction kit (Boehringer Mannheim) according to the manufacturer's directions.
A second set of reactions was performed using either 0.1 μg of each of the gel-purified products from the previous reactions or 0.3 μg of the wild-type STAT-6 expression vector, with the forward primer 5′-CTGGGGGACCGGATCCGGCAT-3′ and the reverse primer 5′-CAGGAGAAGCTTTCATTTATCGTCATCGTCTTTGTAGTCGGTGAGGTCCTG-3′. This generated the full-length STAT-6 cDNA with or without a mutation converting Tyr641 to Phe with a Not I site on the 5′ terminus and a FLAG epitope sequence and a Xho I site on the 3′ terminus. The wild-type and mutant PCR products were cloned into the mammalian expression vector pCMV-Script (Stratagene, La Jolla, CA) to generate the plasmids pST6Y641 and pST6F641, respectively. The sequences of the 2 cloned vectors were verified at the DNA Sequencing facility of Rockefeller University, and the expression of the 2 proteins was verified by immunofluorescence staining with mouse monoclonal anti-FLAG IgG (Sigma) in NIH3T3 fibroblasts transfected with either pST6Y641 or pST6F641 (data not shown).
Transfection and luciferase assay.
For the collagen promoter assay, TSK/+ primary fibroblasts were grown to 90% confluence in 35-mm2 plates in DMEM supplemented with 100 units/ml of penicillin, 100 μg/ml of streptomycin, and 10% FCS (Mediatech). The cells were transfected with 1 μg of a luciferase reporter construct under control of the −300 to +54-bp α2(I) collagen promoter (a gift from Benoit deCrombrugghe, University of Texas, MD Anderson Cancer Center, Houston, TX) and either with 0.5 μg of pEGFP-C1 (Clontech, Palo Alto, CA) alone or with 1 μg of pST6Y641 or pST6F641 using FuGENE 6 transfection reagent (Boehringer Mannheim) according to the manufacturer's directions. Six hours after transfection, the media were replaced with fresh serum or with DMEM (alone or containing 400 units/ml of IL-4). Forty-eight hours later, the samples were measured for levels of luciferase activity using the dual luciferase assay (Promega, Madison, WI) according to the manufacturer's directions.
Western blotting assay.
Confluent fibroblasts in 100-mm2 plates were serum starved for 16 hours and then stimulated with 200 units of recombinant IL-4 for 30 minutes at 37°C. After stimulation, cells were lysed with detergent lysis buffer (Tris buffered saline [TBS], 1% Nonidet P40, 0.1% sodium dodecyl sulfate [SDS], 100 μg/ml of phenylmethylsulfonyl fluoride, 1 mM sodium orthovanadate, 1:100 dilution of protease and phosphatase inhibitor cocktails; Sigma), and 200 μg of the cell lysate was incubated overnight with 1 μg/ml of polyclonal rabbit anti–JAK-1, anti–JAK-2, anti–JAK-3, or STAT-6 antibodies (Santa Cruz Biotechnology, Santa Cruz, CA). The antigen–antibody complexes were precipitated with 50 μl of protein A–agarose (Santa Cruz Biotechnology) for 2 hours at 4°C, and the proteins were separated on a 7.5% SDS–polyacrylamide gel electrophoresis (SDS-PAGE) gel.
The proteins were electrotransferred to a polyvinylidene difluoride (PVDF) membrane (Millipore, Bedford, MA). The PVDF membrane was then blocked overnight at 4°C with TBS–0.05% Tween 20 (Sigma) containing 3% bovine serum albumin. Western blot analysis of tyrosine phosphorylation of the immunoprecipitated JAK-1/2/3 was then performed using a mouse anti-phosphotyrosine monoclonal antibody (Cell Signaling Technology, Beverly, MA) according to the manufacturer's directions. Coprecipitated p300 was detected using the methods described above and by probing the blot with rabbit anti-p300 antibodies (Santa Cruz Biotechnology).
In other experiments, Western blot analysis was used to quantify the IL-4Rα protein level in C57BL/6 and TSK/+ fibroblasts and in CD4+ T cells primed under Th2 conditions. Cell lysates obtained from 107 cells per sample as described above were mixed with 4× SDS-PAGE loading buffer. The samples were boiled and separated on 8% pre-made acrylamide gels (Novex, San Diego, CA) and transferred onto Immobilon-P membranes (Millipore). Membranes were then probed with anti–IL-4Rα (Santa Cruz Biotechnology) followed by horseradish peroxidase–labeled secondary antibodies (Jackson ImmunoResearch, Avondale, PA) and visualized with SuperSignal West Dura Extended Duration Substrate (Pierce). The probed membranes were stripped with stripping buffer (2% SDS, 62.5 mM Tris HCl, pH 6.8, 100 mM 2-mercaptoethanol) at 60°C for 30 minutes and then reprobed with anti–β-actin (clone AC-74; Sigma).
EMSA was performed according to protocols described previously (24). Briefly, confluent 100-mm2 plates of fibroblasts were incubated for 16 hours in 3 ml of serum-free DMEM. The fibroblasts were then stimulated for 30 minutes with IL-4. At the end of this period, nuclear extracts were generated as previously described by Funk et al (24). Then, 10 μg of nuclear extract was incubated for 10 minutes at room temperature in binding buffer containing 20 mM HEPES, 83.3 mM NaCl, 0.1 mM EDTA, 10% glycerol, 1.9 mM MgCl2, 0.5 mM dithiothreitol, 0.01% Triton X-100, and 25 μg/ml of poly(dI-dC) (Sigma), in the presence or absence of 40 ng of the unlabeled competitor oligonucleotide containing a wild-type or mutated consensus binding site for AP-1, nuclear factor 1 (NF-1), and Sp-1 (Santa Cruz Biotechnology).
For the STAT-6 binding reaction, a nucleotide containing either the AP-1 consensus site or the OCT-1 consensus site 5′-ATGCAAAT-3′ was used as a nonspecific inhibitor. One nanogram of 32P-radiolabeled oligonucleotide (200,000 cpm) containing the Sp-1 consensus sequence 5′-GGGGCGG-3′, the NF-1 consensus sequence 5′-TGGATTGAAGCCAAT-3′, the AP-1 consensus sequence 5′-CGCTTGATGACTCAGCCGGAA-3′, or the STAT-6 consensus sequences 5′-TTT-CCC-AGA-AA-3′ (Santa Cruz Biotechnology) was added and incubated for 20 minutes at room temperature. The samples were then loaded on a 4% nondenaturing polyacrylamide gel and subjected to electrophoresis for 3 hours at 200V. The shifted bands were visualized after autoradiographic exposure for 6 hours at −70°C.
Cell proliferation and 35S incorporation.
To determine the effects of IL-4 on cell growth, 104 fibroblasts were plated in 96-well plates. After a 4-hour incubation, murine IL-4 was added at a concentration of 300, 400, or 1,000 units/ml (R&D Systems), and the cells were allowed to grow at 37°C for 24 or 48 hours. The number of cells present was then determined using the Cell Titer Proliferation Assay (Promega) according to the manufacturer's directions. Results are expressed as the mean ± SEM of triplicate wells.
The relative intensities of the bands identified by EMSA were determined on digital image files. Quantity One software (Bio-Rad, Richmond, CA) was used according to the manufacturer's instructions.
Statistical significance was determined by Student's unpaired t-test. Differences between groups were considered statistically significant when the P value was less than 0.05.
Stimulation of collagen synthesis by IL-4.
TSK/+ fibroblasts, like human SSc fibroblasts (6, 7) or keloid fibroblasts (9), exhibit elevated collagen production as compared with collagen production by normal fibroblasts. We previously showed that incubation with IL-4 or transforming growth factor β (TGFβ) induced severalfold increases in collagen synthesis by normal and TSK/+ fibroblasts (16). Results of an independent experiment performed in the present study (Figure 1) were consistent with those previously reported, demonstrating that IL-4 induces a substantial increase in collagen synthesis. This effect appeared to be much greater in TSK/+ fibroblasts than in normal fibroblasts.
Expression of IL-4Rα and JAK kinases in C57BL/6 and TSK/+ fibroblasts.
The expression of IL-4Rα on C57BL/6 and TSK/+ fibroblasts was studied by flow cytometry and Western blotting assay. Flow cytometry was carried out on fibroblasts that had been preincubated with or without IL-4 and then stained with anti–IL-4Rα monoclonal antibody M1. As shown in Figure 2A, TSK/+ fibroblasts stained much more strongly than normal fibroblasts (difference in mean fluorescence intensity [ΔMFI] compared with cultures without IL-4 29.7 in TSK/+ fibroblasts and 5.5 in C57BL/6 fibroblasts). This staining was specific, since preincubation with IL-4 prevented the binding of M1 monoclonal antibody to fibroblasts. Consistent with this finding, Western blotting showed basal levels of IL-4Rα to be higher in TSK/+ fibroblasts than in C57BL/6 fibroblasts (Figure 2B). Lysates from murine CD4+ T cells primed under Th2 conditions were used as controls.
It is unlikely that increased expression of IL-4R in TSK/+ fibroblasts is due to autocrine synthesis of IL-4 for 2 reasons. First, no IL-4 transcript was detected by real-time PCR in fibroblast culture regardless of whether they were treated with IL-4. Second, the ΔMFI of staining of the C57BL/6 fibroblasts was not increased upon incubation with IL-4. However, it is worth noting that the intensity of staining of IL-4Rα in TSK/+ fibroblasts in the absence of IL-4 (ΔMFI 6.9) was similar to that observed in C57BL/6 fibroblasts (ΔMFI 5.5), which raises the possibility that in vivo, IL-4 has an effect on the expression of IL-4Rα.
Among the early events in the IL-4 signaling cascade in lymphocytes is the rapid phosphorylation of JAK kinases, which phosphorylate the IL-4 receptor (23). Using lysates prepared from C57BL/6 and TSK/+ fibroblasts, we observed that whereas in TSK/+ fibroblasts, JAK-1 and JAK-2 were constitutively phosphorylated, in normal fibroblasts, IL-4 induced the phosphorylation of JAK-1 and JAK-2. However, it appeared that IL-4 stimulation of TSK/+ fibroblasts did not further increase the phosphorylation of JAK-2, indicating that in TSK/+ fibroblasts, the phosphorylation of JAK-2 may already be induced to maximum levels (Figure 2C). This difference in phosphorylation was not due to changes in the levels of expression of the proteins, since the Western blot showed similar levels of protein in all samples. No phosphorylation of JAK-3 was observed in fibroblasts incubated with or without IL-4, despite the fact that JAK-3 protein was detected by Western blotting assay (data not shown). Taken together, these results show that IL-4Rα is more highly expressed in TSK/+ fibroblasts and that JAK-1 and JAK-2 are autophosphorylated in TSK/+ fibroblasts, whereas they are activated only upon exposure to IL-4 in normal fibroblasts.
Involvement of STAT-6 in IL-4–mediated up-regulation of collagen synthesis.
STAT-6 is a major signal transduction molecule associated with IL-4–stimulated activation of gene expression in lymphocytes (25). There are a few observations on the ability of IL-4 to activate the STAT-6 signal transduction pathway in fibroblasts. For example, Murata et al (19) showed the activation of STAT-6 in human fibroblasts by IL-4 and IL-13. In the present study, we examined the ability of IL-4 stimulation to induce phosphorylation of STAT-6 in TSK/+ and C57BL/6 fibroblasts and the potential effect of STAT-6 on type I collagen promoter activity and mRNA levels. As shown in Figure 3A, the level of phosphorylation of STAT-6 was higher in TSK/+ fibroblasts than in wild-type fibroblasts upon stimulation with IL-4. This finding further indicates that the magnitude of the response to IL-4 is much greater in fibroblasts that exhibit a fibrotic phenotype. In addition, it was observed that p300 coprecipitated with STAT-6. This association was slightly higher in TSK/+ fibroblasts than in the wild-type fibroblasts under basal conditions and was not increased upon IL-4 stimulation, indicating that the association is not activation dependent (Figure 3A).
EMSA analysis revealed an elevated interaction of STAT-6 with DNA in the basal state in TSK/+ fibroblasts compared with normal fibroblasts. IL-4 stimulation resulted in higher STAT-6 DNA binding ability in TSK/+ fibroblasts (Figure 3B). The specificity of binding was demonstrated by the complete inhibition of binding by a 10 times molar excess of cold STAT-6 consensus oligonucleotide and the lack of inhibition in the presence of an AP-1 or OCT-1 cold competitor oligonucleotide.
We examined the role of STAT-6 in IL-4–induced activation of collagen genes in TSK/+ fibroblasts and found that they are hyperresponsive to IL-4. A luciferase reporter under control of the −300 to +54-bp α2(I) collagen promoter was introduced into TSK/+ fibroblasts alone or together with wild-type STAT-6 or p300. We found that IL-4 stimulation caused a 4-fold increase in luciferase expression in cells transfected with the reporter construct alone. Cells in which the wild-type STAT-6 was overexpressed had substantially higher basal levels of promoter activity and this was further increased in the presence of IL-4. Co-expression of p300 in cells into which the wild-type STAT-6 had been introduced had slight increases in luciferase expression but showed a very striking increase in response to IL-4 stimulation (Figure 4A). Thus, these data suggest that IL-4 can up-regulate type I collagen promoter activity and that this activity is enhanced by overexpression of STAT-6.
The effect of IL-4 and STAT-6 on the level of type I collagen message in fibroblasts was examined by real-time RT-PCR. IL-4 stimulation induced a 3–4-fold increase in α1(I) mRNA over basal levels in TSK/+ fibroblasts. Similar to the results shown in Figure 4A, introduction of a wild-type STAT-6 construct caused an even greater increase in α1(I) mRNA expression, with a further increase if the cells were also stimulated with IL-4 (Figure 4B). In contrast, cells transfected with a dominant-negative STAT-6 construct failed to respond to IL-4 with increased α1(I) collagen mRNA levels. These results clearly demonstrate the importance of STAT-6 in the induction of collagen biosynthesis in fibroblasts.
IL-4 activation of transcription factors associated with collagen synthesis.
The α1(I) collagen promoter contains 2 NF-1 and Sp-1 sites in tandem, between −174 and −84 bp; these sites are important for transcription in scleroderma fibroblasts (26). It has been shown that AP-1 and Sp-1 bind in a region spanning −442 to +1,697 bp in the human α2(I) collagen promoter (27, 28). Furthermore, it has been reported that increased α1 procollagen expression in TSK/+ myocardial fibroblasts is due to diminished interaction of AP-1 with a negative regulatory element located in the region between −675 and −804 bp of the α1(I) procollagen promoter (29).
Using EMSA, we observed that IL-4 increases the capacity of lysates from TSK/+ fibroblasts to bind to oligonucleotides containing AP-1–specific (Figure 5A) and Sp-1–specific (Figure 5B) sequences. The specificity of this binding was verified by inhibition with excess unlabeled specific oligonucleotide, but not by mutant oligonucleotides. It is interesting to note that lysates from TSK/+ fibroblasts appeared to have high basal binding activity for AP-1 that was greatly increased after a 30-minute exposure to IL-4. Conversely, lysates from unstimulated C57BL/6 fibroblasts showed very low DNA binding activity for AP-1, and IL-4 stimulation resulted in virtually no increase in DNA binding activity. This suggests that at level of transcription factors, there are striking differences between the response to IL-4 by normal and TSK/+ fibroblasts. Lysates from TSK/+ fibroblasts had strong binding activity for oligonucleotides specific for NF-1; this was not changed by stimulation with IL-4 (data not shown). Thus, IL-4–enhanced production of collagen is associated with increased expression or increased DNA binding activity of transcription factors that can regulate collagen biosynthesis, indicating a potential mechanism through which IL-4 can regulate collagen transcription.
IL-4 is extremely pleiotropic. It activates genes in a variety of cells, including B cells, CD4+ T cells, macrophages, natural killer cells, endothelial cells, and fibroblasts (20). In this study, we showed that IL-4 up-regulates the collagen genes, as illustrated by enhanced collagen synthesis in TSK/+ fibroblasts compared with normal fibroblasts. This finding is consistent with the findings of several other studies showing that fibroblasts isolated from SSc patients and from mouse models of fibrosis produced elevated levels of collagen compared with normal controls, both in the basal state and after stimulation with IL-4 (5–7, 9).
IL-4 may enhance the expression of collagen genes directly, by activation of transcription factors, which bind to collagen promoter, or indirectly, by epistatic interaction with the TGFβ1 gene. The concept of a direct effect is supported by our preliminary results showing that collagen synthesis by a fibroblast line obtained from Smad3−/− mice (IIS3 knockout) was quite similar compared with the IIS3 wild-type line obtained from normal mice, as assessed by the incorporation of labeled hydroxyproline (mean ± SD incorporation by IIS3 wild-type fibroblasts 4,890 ± 58 cpm and 12,411 ± 622 cpm and by IIS3 knockout fibroblasts 2,564 ± 178 cpm and 7,042 ± 75 cpm in medium alone and in the presence of IL-4, respectively). Although IL-4 can affect fibroblast proliferation and potentially contribute to fibrosis, this is most likely not the case, since IL-4 appears to induce a weak proliferation of both TSK/+ and C57BL/6 fibroblasts in vitro (Table 1). IL-4 may also have an indirect effect on collagen promoter activity by activating the transcription of the TGFβ gene encoding another powerful profibrogenic cytokine. This concept is supported by previous observations showing that IL-4 enhances the expression of the TGFβ gene in fibroblasts and that TSK/TSK or TSK/+ mice with a disrupted IL-4 gene had reduced levels TGFβ gene transcription and did not develop skin fibrosis (15).
Table 1. Effect of interleukin-4 on the proliferation of fibroblasts from C57BL/6 and TSK/+ mice*
Values are the mean ± SD optical density in triplicate samples, as described previously (36). This experiment was performed twice, and the results were similar.
24 hours of incubation
0.818 ± 0.02
0.713 ± 0.05
0.816 ± 0.01
0.709 ± 0.03
0.823 ± 0.04
0.730 ± 0.03
0.946 ± 0.03
0.837 ± 0.04
48 hours of incubation
1.134 ± 0.009
1.216 ± 0.04
1.141 ± 0.01
1.230 ± 0.02
1.138 ± 0.02
1.241 ± 0.05
1.406 ± 0.03
1.437 ± 0.03
We investigated IL-4 signaling and regulation of collagen genes in murine fibroblasts and, in particular, in fibroblasts from TSK mice which develop a scleroderma-like syndrome, since little is known of these mechanisms in this cell type. The binding of IL-4 to its receptor triggers a cascade of events. Among the early events in the IL-4 signaling cascade is the rapid phosphorylation of JAK kinases. We found that in TSK/+ fibroblasts, JAK-1 and JAK-2 were constitutively phosphorylated, whereas in normal fibroblasts, they were activated only after IL-4 stimulation. The phosphorylation pattern of JAK-1 and JAK-2 in both cell types resembles that observed in normal human fibroblasts stimulated with IL-4 and CD40 ligation (30). It is noteworthy that in contrast to lymphocytes (25), JAK-3 in fibroblasts was not phosphorylated either in the basal state or after stimulation with IL-4. This finding is consistent with other reports suggesting that JAK-3 is activated mainly in cells of hematopoietic origin (25).
It is well known that STAT-6 represents one of the major signal transduction molecules in lymphocytes exposed to IL-4. However, the role of STAT-6 in the activation of collagen genes is unknown. We observed that overexpression of wild-type STAT-6 enhances collagen promoter activity and collagen mRNA expression. In contrast, introduction of a dominant-negative form of STAT-6 blocked IL-4–mediated elevation of type I collagen mRNA expression, suggesting a critical role of STAT-6 in IL-4–induced collagen gene expression. After stimulation with IL-4, there was a much greater increase in STAT-6 tyrosine phosphorylation in TSK/+ fibroblasts compared with normal fibroblasts. This finding was supported by the observation of a greater increase in STAT-6–DNA interactions in IL-4–stimulated TSK/+ fibroblasts than normal fibroblasts.
The p300 and cAMP response element binding protein are large anchor factors that have been shown to interact with a number of transcription factors, including Sp-1, Smad2, Smad3, and STAT-6 (31, 32). While it is not known how p300 enhances the promoter-activating ability of other transcription factors, its function is probably related to its intrinsic histone acetylase activity. Also, p300 association has been shown to be essential for STAT-6–DNA interactions (32). Therefore, p300 association is most likely a crucial factor in STAT-6–mediated promoter activation, allowing STAT-6 to gain access to responsive elements. The results depicted in Figures 3A and 4A support this notion and demonstrate that STAT-6 associates with p300 in fibroblasts and that this association may be important in the IL-4–mediated induction of type I collagen genes.
No STAT-6–responsive elements have been identified in the type I collagen promoter. Nonetheless, the increased collagen promoter activity and transcription of collagen genes in fibroblasts transfected with STAT-6 indicates that it can induce type I collagen production in fibroblasts. It is possible that STAT-6 participates in the initiation of type I collagen gene transcription by recruiting anchor factors such as p300, allowing for the formation of a more stable or efficient transcription complex without directly interacting with the promoter DNA. This hypothesis is supported by the fact that p300 enhances α2(I) collagen promoter activity in fibroblasts overexpressing STAT-6, which is further increased by stimulation with IL-4. Alternatively, the STAT-6 effect may be indirect, through the activation of factors that regulate the expression of the TGFβ gene and subsequent autocrine synthesis of this cytokine, leading to higher expression of collagen genes.
EMSA analysis showed that IL-4 induced the activation of Sp-1 and AP-1. Sp-1 is a transcription factor known to bind to the α2(I) collagen promoter at several sites from −342 to −265 bp (28). These sites have been shown to play an important role in the response to other profibrogenic cytokines, such TGFβ. Shiraski et al (33) showed that an Sp-1 complex activates the human tenascin-C promoter in cooperation with the transcription factor Ets. Tenascin is also overproduced by scleroderma fibroblasts. Thus, Sp-1 may represent a common transcription factor associated with a preinitiaton complex after stimulation of fibroblasts with profibrogenic cytokines.
AP-1 is also an important transcription factor in the regulation of collagen synthesis. There are several AP-1 binding motifs in the human α2(I) collagen promoter (27). Chung et al (34) showed that the −265 to −241-bp region is crucial in the minimal promoter segment of α2(I) collagen for binding AP-1 after stimulation of fibroblasts with TGFβ. There is little evidence linking AP-1 and IL-4 stimulation. However, there is an AP-1 site in the −124 to −99-bp region of the Igε promoter, and Chen and Stavnezer (35) showed that IL-4–mediated activation of the Igε promoter in B cells is dependent on the presence of AP-1 and that IL-4 elevated the DNA binding capacity of AP-1. Apparently, the AP-1 complex acted synergistically with STAT-6 to enhance the synthesis of IgE. Therefore, AP-1 may be an important link in IL-4–mediated activation of collagen synthesis.
We found only a low level of AP-1 DNA binding activity in IL-4–stimulated C57BL/6 fibroblasts. However, in TSK/+ fibroblasts, there was a strong elevation of AP-1 binding, indicating IL-4 inducibility of the AP-1 complex (Figure 5A). This suggests that TSK/+ fibroblasts have elevated IL-4–mediated activation of transcription factors, potentially linking AP-1, IL-4, and the overexpression of collagen in the TSK mouse model of scleroderma.
Taken together, these observations suggest that IL-4 stimulation activates transcription factors, which leads to increased promoter activity of both collagen and TGFβ genes. The increased transcription of the TGFβ gene following IL-4 stimulation may further contribute to increased synthesis of collagen by fibroblasts.
In the present study, we also showed that TSK/+ fibroblasts, like human scleroderma fibroblasts, were hyperresponsive to IL-4 stimulation. This can be attributed to either constitutive phosphorylation of molecules involved in the IL-4 signaling pathway, constitutively increased activation of transcription factors, or differences in the density of IL-4R. Our results suggest that the hyperresponsiveness of TSK/+ fibroblasts might be related to increased expression of IL-4R, the activation of JAK kinases, and the activation of some transcription factors that interact with both collagen and TGFβ promoters.
In conclusion, in this study we sought to establish a role for IL-4 in the synthesis of collagen and, potentially, in fibrotic disease. We found that IL-4 activates intermediate proteins involved in the growth and proliferation of fibroblasts and up-regulates type I collagen genes in a STAT-6–dependent manner. The level of activation of these molecules was consistently higher in TSK/+ fibroblasts than in C57BL/6 fibroblasts, which suggests that hyperresponsiveness to IL-4 may be a principal cause of excessive collagen production associated with fibrotic disease.