Involvement of αvβ5 integrin–mediated activation of latent transforming growth factor β1 in autocrine transforming growth factor β signaling in systemic sclerosis fibroblasts

Authors


Abstract

Objective

To confirm the involvement of αvβ5 in the self-activation system in systemic sclerosis (SSc) fibroblasts.

Methods

Levels of αvβ5 expression were analyzed by immunoprecipitation. The promoter activity of the human α2(I) collagen gene was determined by transient transfection assay. Phosphorylation levels and DNA binding ability of Smad3 were investigated by immunoprecipitation and DNA affinity precipitation, respectively. The localization of active transforming growth factor β (TGFβ) was determined by coculture assay using TMLC cells (mink lung epithelial reporter cells that stably express a portion of the plasminogen activator inhibitor 1 promoter). The morphologic features of cells were determined by immunofluorescence analysis.

Results

Levels of αvβ5 expression were significantly elevated in SSc fibroblasts compared with normal fibroblasts. Treatment with anti-αvβ5 antibody or β5 antisense oligonucleotide significantly reduced human α2(I) collagen gene promoter activity in SSc fibroblasts. In SSc fibroblasts pretreated with TGFβ1 antisense oligonucleotide, the exogenous latent TGFβ1 stimulation significantly increased human α2(I) collagen gene promoter activity; this effect was significantly reduced in the presence of anti-αvβ5 antibody. Phosphorylation levels and DNA binding ability of Smad3 in SSc fibroblasts were significantly reduced by treatment with β5 antisense oligonucleotide. The luciferase activity of TMLC cells cocultured with SSc fibroblasts was significantly elevated compared with that of TMLC cells cocultured with normal fibroblasts and was significantly reduced in the presence of anti-αvβ5 antibody. Anti-αvβ5 antibody reversed the myofibroblastic features of SSc fibroblasts.

Conclusion

Up-regulated expression of αvβ5 contributes to the establishment of autocrine TGFβ signaling in SSc fibroblasts through activation of endogenous latent TGFβ1.

Scleroderma (systemic sclerosis [SSc]) is an acquired disorder that typically results in fibrosis of the skin and internal organs (1). Previous findings indicate that the pathogenesis of this disorder includes inflammation, autoimmune attack, and vascular damage, leading to fibroblast activation (2). However, cultured SSc fibroblasts, which are free of such environmental factors, continue to produce excessive amounts of extracellular matrix (ECM) proteins (3, 4), suggesting that once activated, these cells establish a constitutive self-activation system. One of the major cytokines involved in this process is transforming growth factor β1 (TGFβ1) (5). The principal effect of TGFβ on mesenchymal cells is its stimulation of ECM deposition, as evidenced by the findings that 1) SSc fibroblasts express elevated levels of TGFβ receptors, and this correlates with elevated levels of α2(I) collagen messenger RNA (mRNA) (6–9) and 2) blockade of TGFβ signaling with anti-TGFβ antibody or TGFβ1 antisense oligonucleotide abolishes the increased expression of human α2(I) collagen mRNA in SSc fibroblasts (7).

TGFβ1 is normally secreted as a complex composed of 3 proteins, including the bioactive peptide of TGFβ1, latency-associated peptide β1 (LAP-β1), and latent TGFβ (LTGFβ) binding protein 1 (LTBP-1). TGFβ1 forms a noncovalent complex with LAP-β1, which is called small latent complex (SLC), and in this configuration, TGFβ1 is unable to bind to its receptors. SLC is joined by LTBP-1, the N-terminal region of which is covalently crosslinked to ECM proteins by transglutaminase; the complex of all 3 proteins is called the large latent complex (10). The constitutive secretion of LTGFβ1 by many cell types in culture suggests that there are extracellular mechanisms that control the activity of this potent cytokine. Although these processes are not fully understood, studies have demonstrated that cell surface molecules or secreted extracellular molecules can activate LTGFβ1. Specifically, the αvβ6 integrin and thrombospondin 1 (TSP-1) have been implicated in activation of LTGFβ1 through nonproteolytic mechanisms (11, 12). In addition, it has been proposed that plasmin leads to activation of LTGFβ1 through proteolytic degradation of LAP-β1 (13). The ability of αvβ8 integrin to activate LTGFβ1 by membrane type 1 matrix metalloproteinase (MT1-MMP)–dependent degradation of LAP-β1 has also been demonstrated (14). Thus, normal TGFβ function is thought to be largely controlled by its activation from the latent state.

LAP-β1 contains an RGD motif that is recognized by αv-containing integrins, including αvβ1, αvβ3, αvβ5, αvβ6, and αvβ8 (11, 14–16). Although all of these αv-containing integrins bind to LAP-β1 and have the potential to modulate the localization and possibly activation of SLC, only αvβ6 and αvβ8, neither of which is expressed in dermal fibroblasts, have been demonstrated to be able to activate SLC (11, 14). In particular, αvβ6-mediated activation of SLC has been shown to play an important role in response to tissue injury, since epithelium-restricted β6−/− mice exhibited only a minor lung fibrotic response to bleomycin administration, compared with wild-type mice (11). Although there have been no reports of activation of SLC by other αv-containing integrins (αvβ1, αvβ3, and αvβ5), we recently demonstrated that αvβ5 is up-regulated in SSc dermal fibroblasts and that transient overexpression of αvβ5 induces increased transcriptional activity of the human α2(I) collagen gene in normal dermal fibroblasts (17). These observations suggest that up-regulated expression of αvβ5 contributes to the establishment of autocrine TGFβ signaling in SSc fibroblasts through the activation of SLC.

The present study was undertaken to investigate the above hypothesis. We first studied the effect of anti-αvβ5 antibody or β5 antisense oligonucleotide on the expression of the human α2(I) collagen gene and the activation state of TGFβ signaling in normal and SSc fibroblasts. We then investigated whether SLC is activated on the cell surface of normal and SSc fibroblasts. We further determined the effect of anti-αvβ5 antibody on the morphologic features of SSc fibroblasts. Our results suggest that the up-regulation of αvβ5 contributes to the establishment of autocrine TGFβ signaling in SSc fibroblasts and that blockade of this integrin reverses the myofibroblastic phenotype of these cells. To our knowledge, this is the first report of findings indicating the possibility that fibrotic disorders, especially SSc, might be regulated by targeting this integrin.

MATERIALS AND METHODS

Reagents.

Recombinant human TGFβ1 and SLC were obtained from R&D Systems (Minneapolis, MN). Antibodies to β-actin and α-smooth muscle actin (α-SMA) were purchased from Sigma (St. Louis, MO). Antibodies to the β5 subunit (E-19), αv subunit (Q-20), Smad2/3 (N-19), and Smad3 (FL-425) were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). Anti-Smad2/3 antibody (S66220) was obtained from Transduction Laboratories (Lexington, KY). Functional blocking antibody against αvβ5 (P1F6) was purchased from Chemicon (San Francisco, CA). FuGENE 6 was obtained from Roche Diagnostics (Indianapolis, IN).

Cell cultures.

Human dermal fibroblasts were obtained from skin biopsy samples from the affected area (dorsal forearm) in 8 patients with diffuse cutaneous SSc (duration of skin thickening <2 years). Control fibroblasts from 8 healthy donors were obtained by skin biopsy. Institutional approval was obtained, and all subjects provided informed consent. Each control donor was matched with an SSc patient for age, sex, and biopsy site, and control and patient samples were processed in parallel. Primary explant cultures in modified Eagle's medium (MEM) with 10% fetal calf serum (FCS), 2 mML-glutamine, and 50 μg/ml of amphotericin were established in 25-cm2 culture flasks, as described previously (9). Fibroblast cultures isolated independently from different individuals were maintained as monolayers at 37°C in 95% air, 5% CO2, and studied between the third and sixth subpassages. TMLC cells (mink lung epithelial reporter cells that stably express a portion of the plasminogen activator inhibitor 1 promoter) (kindly provided by Dr. Daniel B. Rifkin, New York University School of Medicine, New York, NY) were cultured in MEM with 10% FCS until assayed.

Immunoblotting.

Confluent quiescent fibroblasts were treated with appropriate reagents. After incubation for the indicated periods of time, cells were washed 3 times with cold phosphate buffered saline (PBS) and lysed in lysis buffer (1% Triton X-100 in 50 mM Tris HCl [pH 7.4], 150 mM NaCl, 3 mM MgCl2, 1 mM CaCl2 containing 10 μg/ml each of leupeptin, pepstatin, and aprotinin, and 1 mM phenylmethylsulfonyl fluoride [PMSF]). Protein extracts were subjected to sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to nitrocellulose membranes. Membranes were incubated overnight with primary antibody, washed, and incubated for 1 hour with secondary antibody. After washing, visualization was performed by enhanced chemiluminescence according to the recommendations of the manufacturer (Amersham Pharmacia Biotech, Buckinghamshire, UK). The densities of bands were measured with a densitometer.

Immunoprecipitation.

Protein extracts (1 μg) were preadsorbed with protein G–agarose beads (Santa Cruz Biotechnology) for 2 hours at 4°C, and incubated with 2 μg of immunoprecipitating antibody for 1 hour and then with protein G–agarose beads for 2 hours. The beads were spun down, washed 3 times with lysis buffer, resuspended in the sample buffer for electrophoresis, boiled for 3 minutes, and spun briefly. The supernatants were subjected to immunoblotting.

Biotinylation and immunoprecipitation.

Confluent quiescent fibroblasts were incubated with membrane-impermeable NHS-LC-biotin (Pierce, Rockford, IL) dissolved at 0.5 mg/ml in PBS at 4°C for 30 minutes. The cells were washed with cold PBS, harvested into lysis buffer, and immunoprecipitation was performed. Each immunoprecipitate was subjected to SDS-PAGE, and Western blots were prepared. The Western blots were probed with horseradish peroxidase–conjugated streptavidin and visualized as described above.

Immunofluorescence.

Quiescent cells cultured in 4-well Lab Tek chambers (Nunc, Naperville, IL) were treated with 10 μg/ml anti-αvβ5 antibody (P1F6) or preimmune mouse IgG for 48 hours. Cells were then fixed with 3.7% formaldehyde, permeabilized with 0.5% Triton X-100 in PBS, and blocked with 10% FCS in PBS containing 0.5% Triton X-100. Cells were stained with anti–α-SMA antibody, washed, and incubated with fluorescein isothiocyanate–conjugated rabbit anti-mouse IgG (Sigma). A Zeiss microscope was used to visualize fluorescence. Myofibroblasts were defined as cells with features such as cellular hypertrophy and well-formed α-SMA fibers. To determine the percentage of cells that differentiated into myofibroblasts, 100 cells per cell strain were examined microscopically.

DNA affinity precipitation.

Two oligonucleotides containing biotin on the 5′ nucleotide of the sense strand were prepared as described previously (18). The sequences of these oligonucleotides were as follows: 1) 3 × CAGA oligonucleotide 5′-TCGAGAGCCAGACAAGGAGCCAGACAAGGAGCCA-GACACTCGAG, which is a trimer of the CAGA motif, and 2) 3 × CAGA-M oligonucleotide 5′-TCGAGAGCTACATAAAAAGCTACATATTTAGCTACATACTCGA, which is a trimer of the mutated CAGA motif. Nuclear protein was extracted using lysis buffer containing 10 mM Tris HCl (pH 7.4), 150 mM NaCl, 1 mM EDTA, 1% Nonidet P40, 50 mM NaF, 1 mM PMSF, 1 mM Na3VO4, and 1 μg/ml each of leupeptin, aprotinin, and pepstatin. Poly(dI-dC) competitor (5 μg) was incubated with 500 μg of nuclear protein for 30 minutes at 4°C, followed by a 1-hour incubation with 500 pmoles of each double-stranded oligonucleotide. Then 65 μl of streptavidin–agarose (Sigma) was added to the reaction mixture and incubated overnight at 4°C. The protein–DNA–streptavidin–agarose complex was washed 3 times with lysis buffer, resuspended in the sample buffer for electrophoresis, boiled for 3 minutes, and spun briefly. The supernatants were subjected to immunoblotting with anti-Smad2/3 antibody (S66220). The specific binding of Smad3 to 3 × CAGA oligonucleotide was confirmed in experiments using 3 × CAGA-M oligonucleotide. The binding of Smad3 to 3 × CAGA-M oligonucleotide was not observed in the presence or absence of TGFβ1 (results not shown).

Plasmid construction.

A −772 COL1A2/chloramphenicol acetyltransferase (CAT) construct consisting of the human α2(I) collagen gene fragment (+58 to −353 bp relative to the transcription start site) linked to the CAT reporter gene was generated as previously described (19). Plasmids were purified twice on CsCl gradients. At least 2 different plasmid preparations were used for each experiment.

Transient transfection.

Cells were grown to 50% confluence in 100-mm dishes, and the medium was replaced with serum-free medium; after 4 hours, cells were transfected with the −772 COL1A2/CAT construct, using FuGENE 6. To control for minor variations in transfection efficiency, 1 μg of pSV–β-galactosidase vector was included in all transfections. After 48 hours of incubation, cell lysates were prepared using Reporter Lysis Buffer (Promega, Madison, WI). Extracts, normalized for protein content, were incubated with butyl–coenzyme A and 14C-chloramphenicol for 90 minutes at 37°C. Butylated chloramphenicol was extracted using an organic solvent (a 2:1 mixture of tetramethylpentadecane and xylene) and quantitated by scintillation counting. Each experiment was performed in duplicate.

Measurement of TGFβ1 levels.

Cells were grown to confluence in 6-well plates. The culture medium was removed, washed with MEM to remove excess TGFβ1, and replaced with serum-free medium. After incubation for 24 hours, levels of active and total (active plus latent) TGFβ1 in the supernatants were measured using a TGFβ1 enzyme-linked immunosorbent assay system (Amersham Pharmacia Biotech) (20). This system is designed to measure active TGFβ1. Total TGFβ1 was assayed after acid activation by addition of 1N HCl to samples.

TGFβ bioassay.

To determine TGFβ activation, TMLC cells were cocultured with test cells in the presence or absence of anti-TGFβ antibody (10 μg/ml) or anti-αvβ5 antibody (10 μg/ml) as described previously (11, 14), with minor modifications. TMLC and test cells were mixed at a ratio of 1:1 and suspended at 1 × 106 cells/ml in MEM containing 10% FCS. These cells were plated at 200 μl/well in 12-well plates and allowed to attach for 1 hour. Medium was replaced with 200 μl/well of new medium with or without antibodies and cultured for 24 hours. Cell lysates were prepared using Reporter Lysis Buffer, and luciferase activity was determined using the Promega luciferase assay system. Similar cocultures were performed in 24-well plates with inserts designed for attachment-dependent cell culture (Millicell-PCF 3-μm filter; Millipore, Bedford, MA), but with 1.5 × 105 TMLC and test cells added to the upper or lower chambers.

Statistical analysis.

The mean ± SD results from at least 5 independent experiments were calculated. Statistical analysis was performed using the Mann-Whitney U test. P values less than 0.05 were considered significant.

RESULTS

Cell surface expression levels of integrin αvβ5 in SSc fibroblasts.

In a previous study (17), we demonstrated that levels of αv and β5 subunit protein and mRNA were elevated in SSc fibroblasts compared with normal fibroblasts, by immunoblotting and Northern blotting, respectively. We also showed, by immunoprecipitation, that cell surface expression levels of αvβ5 integrin were elevated in SSc fibroblasts. We further confirmed, by immunohistochemistry analysis, that expression levels of αv and β5 subunit proteins were elevated in SSc fibroblasts. These previous experiments were performed using 5 strains of normal and 5 strains of SSc fibroblasts. In this study, 8 additional strains of normal and 8 of SSc fibroblasts were newly prepared, and all experiments were performed using these cells. In an initial experiment, we compared the cell surface expression levels of αvβ5 integrin in normal and SSc fibroblasts by immunoprecipitation. Since the αv subunit is the only subunit found to be associated with the β5 subunit but it can bind to several additional β subunits (β1, β3, β6, and β8) (21), we performed immunoprecipitation using anti-β5 antibody. As shown in Figure 1, consistent with our previous results, the cell surface expression levels of αvβ5 integrin were significantly elevated in SSc fibroblasts (by 4.2-fold; P < 0.05). These bands were confirmed to be the αv or β5 subunit by immunoblotting with anti–αv integrin or β5 subunit antibody (results not shown).

Figure 1.

Cell surface expression levels of αvβ5 integrin in normal skin (NS) and systemic sclerosis (SSc) fibroblasts. A, Cell surface proteins were labeled with biotin, and whole cell lysates were immunoprecipitated using anti-β5 antibody. Precipitated proteins were subjected to sodium dodecyl sulfate–polyacrylamide gel electrophoreses, and Western blots were prepared. The Western blots were probed with horseradish peroxidase–conjugated streptavidin. Results shown are representative of 5 independent experiments. B, Protein levels were quantitated by scanning densitometry. The level of αvβ5 was defined as the mean of the densitometric values of the αv subunit and the β5 subunit. Levels relative to those in normal fibroblasts (100 arbitrary units [AU]) are shown. Values are the mean and SD from 5 independent experiments. ∗ = P < 0.05 versus normal fibroblasts.

Effect of anti-αvβ5 antibody or β5 antisense oligonucleotide on COL1A2 promoter activity in normal and SSc fibroblasts.

To determine whether up-regulated expression of αvβ5 contributes to the activation of SSc fibroblasts, we investigated the effect of anti-αvβ5 antibody, which was shown to block the function of αvβ5 specifically and not to cross-react with other αv-containing integrins, on COL1A2 promoter activity in normal and SSc fibroblasts. As shown in Figure 2A, anti-αvβ5 antibody significantly reduced COL1A2 promoter activity in a dose-dependent manner in SSc fibroblasts, while it did not affect COL1A2 promoter activity in normal fibroblasts that were either untreated or treated with TGFβ1.

Figure 2.

Comparison of the effect of anti-αvβ5 antibody or β5 antisense oligonucleotide on human α2(I) collagen promoter activity in normal skin (NS) and systemic sclerosis (SSc) fibroblasts. A, The −772 COL1A2 promoter activity in SSc fibroblasts in the presence of anti-αvβ5 antibody or preimmune IgG, and in normal fibroblasts in the presence of anti-αvβ5 antibody or preimmune IgG with or without transforming growth factor β (TGFβ) treatment, was determined and is shown as chloramphenicol acetyltransferase (CAT) activity relative to that in normal fibroblasts treated with preimmune IgG (100 arbitrary units [AU]). ∗ = P < 0.05 versus SSc fibroblasts treated with preimmune IgG. B, Cells were treated with β5 sense (S) and antisense (AS) oligonucleotides (S–β5-1, S–β5-2, AS–β5-1, and AS–β5-2) for 48 hours, and whole cell lysates were subjected to immunoblotting using anti-β5 antibody. Results shown are representative of 5 independent experiments. C, The −772 COL1A2 promoter activity in SSc fibroblasts in the presence of AS–β5-2 or S–β5-2, and in normal fibroblasts in the presence of AS–β5-2 or S–β5-2 with or without TGFβ treatment, was determined and is shown as CAT activity relative to that in normal fibroblasts treated with S–β5-2 (100 AU). ∗ = P < 0.05 versus SSc fibroblasts treated with S–β5-2. Values in A and C are the mean and SD from 5 independent experiments.

To further confirm this, we investigated the effect of β5 antisense (AS) oligonucleotide on the COL1A2 promoter activity in SSc fibroblasts. For this purpose, we established 2 β5 antisense oligonucleotides (AS–β5-1 [5′-GGCTTGCCCTCTGCCTCCCGGC-3′] and AS-β5-2 [5′-GCAGACATAGGTCCCGCTCCCGTTGC-3′]) according to previously described methods (22). As shown in Figure 2B, both AS–β5-1 and AS–β5-2 significantly reduced the expression levels of β5 subunit proteins in SSc fibroblasts to levels similar to those in normal fibroblasts, but the effect was stronger with AS–β5-2. We therefore used AS–β5-2 in the subsequent experiments. As shown in Figure 2C, AS–β5-2 significantly reduced COL1A2 promoter activity in SSc fibroblasts in a dose-dependent manner, while it did not affect COL1A2 promoter activity in normal fibroblasts with or without TGFβ1 treatment. Taken together, these results suggest that up-regulated expression of αvβ5 contributes to the activation of SSc fibroblasts.

Treatment with β5 antisense oligonucleotide reduces the constitutive phosphorylation and the increased DNA binding ability of Smad3 in SSc fibroblasts.

To further confirm the findings described above, we investigated the effect of AS–β5-2 on levels of Smad3 phosphorylation in normal and SSc fibroblasts. In normal fibroblasts (Figure 3A), Smad3 was strongly phosphorylated by TGFβ1 stimulation, and pretreatment with AS–β5-2 or S–β5-2 had no effect on the phosphorylation levels of Smad3. In contrast, as seen in Figure 3B, Smad3 was constitutively phosphorylated in SSc fibroblasts. AS–β5-2 treatment reduced the Smad3 phosphorylation levels in SSc fibroblasts by ∼60%, whereas treatment with S–β5-2 had no effect on Smad3 phosphorylation levels in SSc fibroblasts. We also confirmed the effect of AS–β5-2 on the DNA binding ability of Smad3. As shown in Figure 3C, treatment with AS–β5-2 or S–β5-2 did not affect Smad3 DNA binding levels in normal fibroblasts treated with TGFβ1. In contrast, as shown in Figure 3D, the DNA binding ability of Smad3 in SSc fibroblasts was reduced by ∼45% with AS–β5-2 treatment. These results support the notion that αvβ5 up-regulation has a role in the activation of SSc fibroblasts.

Figure 3.

Effect of β5 subunit antisense oligonucleotide on Smad3 phosphorylation and DNA binding in untreated and TGFβ1-treated normal fibroblasts and in SSc fibroblasts. Confluent quiescent fibroblasts were treated with AS–β5-2 or S–β5-2 for 48 hours. In some experiments, cells were stimulated with 2 ng/ml of TGFβ1 for the last hour. A and B, Whole cell lysates were subjected to immunoprecipitation (IP) using anti-Smad3 antibody, and phospho-Smad3 was detected by immunoblotting using anti–phospho-serine antibody. The same membrane was stripped and reprobed with anti-Smad2/3 antibody. Results shown in the upper panels are representative of 5 independent experiments. Lower panels show relative phospho-Smad3 levels, quantitated by scanning densitometry and corrected for the level of total Smad3 in the same samples relative to those in untreated cells (100 AU). C and D, Nuclear extracts were incubated with biotin-labeled oligonucleotides. Proteins bound to these nucleotides were isolated with streptavidin–agarose beads, and Smad3 was detected by immunoblotting. Results shown in the upper panels are representative of 5 independent experiments. Lower panels show Smad3 levels relative to those in untreated cells (100 AU), quantitated by scanning densitometry. Values in the lower panels of AD are the mean and SD from 5 independent experiments. ∗ = P < 0.05 versus SSc fibroblasts treated with S–β5-2. See Figure 2 for other definitions.

Involvement of up-regulated αvβ5 expression in the increased inducibility of COL1A2 gene expression by exogenous SLC stimulation in SSc fibroblasts.

To further elucidate the detailed mechanisms involved in the association of up-regulated αvβ5 with activation of SSc fibroblasts, we compared the ability of SSc versus normal fibroblasts to activate SLC. For this purpose, we first investigated the effect of exogenous SLC stimulation on COL1A2 promoter activity in normal and SSc fibroblasts. To suppress endogenous TGFβ1 production, cells were pretreated with 10 μM of TGFβ1 antisense oligonucleotide (AS-TGFβ1) for 48 hours prior to stimulation by TGFβ1 or SLC.

As shown in Figure 4, AS-TGFβ1 significantly reduced the increased levels of COL1A2 promoter activity in SSc fibroblasts but caused little reduction in the basal levels of COL1A2 promoter activity in normal fibroblasts, consistent with previous reports (7). This indicates that the activation of SSc fibroblasts may be a result of stimulation by autocrine TGFβ1. Exogenous active TGFβ1 stimulation caused a significant increase in COL1A2 promoter activity in normal and SSc fibroblasts pretreated with AS-TGFβ1. In contrast, with exogenous SLC stimulation, there was a significant and dose-dependent increase in COL1A2 promoter activity in SSc fibroblasts treated with AS-TGFβ1, but the same treatment had no significant effect on normal fibroblasts treated with AS-TGFβ1. This effect of SLC on SSc fibroblasts was significantly reduced by pretreatment with anti-αvβ5 antibody. These results indicated that αvβ5 up-regulation might play a role in the activation of SLC on SSc fibroblasts.

Figure 4.

Comparison of the inducibility of the COL1A2 gene by exogenous small latent complex (SLC) stimulation, in normal versus SSc fibroblasts treated with TGFβ1 antisense oligonucleotide. Cells were cultured for 48 hours in modified Eagle's medium containing 10 μM TGFβ1 antisense oligonucleotide. The medium was then changed to serum-free medium containing 10 μM TGFβ1 antisense oligonucleotide, and cells were transfected with 2 μg of −772 COL1A2/CAT construct. Cells were stimulated with TGFβ1 or SLC at the indicated concentrations for the last 24 hours, and CAT activity was determined. In some experiments, SLC stimulation was performed in the presence of anti-αvβ5 antibody. Levels of CAT activity relative to that in normal fibroblasts treated with TGFβ1 sense oligonucleotide (100 AU) are shown. Values are the mean and SD from 5 independent experiments. ∗ = P < 0.05 versus normal fibroblasts treated with TGFβ1 sense oligonucleotide; ∗∗ = P < 0.05 versus SSc fibroblasts treated with TGFβ1 antisense oligonucleotide and 10 ng/ml SLC. See Figure 2 for other definitions.

Activation of SLC on the cell surface of SSc fibroblasts.

We next investigated the levels of total (active plus latent) and active TGFβ1 protein in normal and SSc fibroblast cultures. Levels of total TGFβ1 proteins were slightly decreased in SSc fibroblasts compared with normal fibroblasts (mean ± SD 0.445 ± 0.098 ng/ml versus 0.505 ± 0.110 ng/ml), but the difference was not significant. There was also no significant difference in the levels of active TGFβ1 (0.081 ± 0.024 ng/ml in SSc fibroblasts and 0.073 ± 0.017 ng/ml and normal fibroblasts). These results contradict the notion that SSc fibroblasts efficiently activate SLC. One possible explanation for this discrepancy is that self-secreted SLC is activated on the cell surface of SSc fibroblasts, but that at least a small amount of the active TGFβ formed is freely diffusible. To investigate this, we cocultured either normal or SSc fibroblasts with TMLC cells. Since the luciferase activity of TMLC cells is hypersensitive to stimulation with TGFβ, this assay can be used to determine the localization of active TGFβ in normal and SSc fibroblasts. When active TGFβ is localized on the cell surface of fibroblasts, the luciferase activity of TMLC cells cocultured in the presence of cell contact will be significantly elevated compared with that cocultured in the absence of cell contact. In contrast, in situations in which active TGFβ is freely diffusible in cell culture medium, there will be no significant difference in luciferase activity between TMLC cells cocultured in the presence of cell contact and those cocultured in the absence of cell contact.

As seen in Figure 5, luciferase activity was significantly elevated in TMLC cells cocultured with SSc fibroblasts compared with those cocultured with normal fibroblasts (∼3.5-fold increase; P < 0.05). This increase was significantly reduced by treatment with anti-αvβ5 antibody and was almost completely abolished by treatment with anti-TGFβ antibody. We also performed coculture assays with inserts to separate TMLC cells and fibroblasts while allowing soluble molecules to pass. In the absence of contact, SSc fibroblasts showed no significant induction of luciferase activity. Taken together, these results indicate that SLC is activated on the cell surface of SSc fibroblasts and that αvβ5 is involved in this process.

Figure 5.

Contribution of up-regulated αvβ5 to small latent complex (SLC) activation on the cell surface of SSc fibroblasts. Normal or SSc fibroblasts were mixed with TMLC cells at a ratio of 1:1 and cocultured under confluent conditions. In experiments without cell–cell contact, either normal or SSc fibroblasts were cocultured with TMLC cells separately by using inserts. After incubation for 24 hours, luciferase activity was determined. In some experiments, assays were performed in the presence of anti-αvβ5 antibody (10 μg/ml) or anti-TGFβ antibody (10 μg/ml). Luciferase activity relative to that of TMLC cells cocultured with normal fibroblasts in the absence of contact (100 AU) is shown. Values are the mean and SD from 5 independent experiments. ∗ = P < 0.05 versus normal fibroblasts; ∗∗ = P < 0.05 versus SSc fibroblasts treated with preimmune mouse IgG (a control for anti-αvβ5 antibody) or preimmune rabbit IgG (a control for anti-TGFβ antibody). See Figure 2 for other definitions.

Reversal of the morphologic myofibroblastic features of SSc fibroblasts by αvβ5 integrin blockade.

In immunofluorescence studies using anti–α-SMA antibody, we investigated the effect of anti-αvβ5 antibody on the morphologic myofibroblastic features of SSc fibroblasts (Figure 6). Approximately 60% of SSc fibroblasts exhibited the morphologic changes of cellular hypertrophy and well-formed α-SMA fibers, which are characteristic of myofibroblasts. However, after the treatment with anti-αvβ5 antibody, the percentage of cells with these features was reduced to ∼25%. In contrast, <5% of normal fibroblasts exhibited these features, in the presence or absence of anti-αvβ5 antibody. These results indicate that blockade of αvβ5 integrin can reverse the myofibroblastic phenotype of SSc fibroblasts.

Figure 6.

Effect of anti-αvβ5 antibody on the morphology and the levels of α-smooth muscle actin (α-SMA) expression in normal and SSc fibroblasts. Confluent quiescent cells were treated with anti-αvβ5 antibody or preimmune IgG for 48 hours. Levels of α-SMA expression were determined by immunofluorescence. Results shown are representative of 8 independent experiments with normal fibroblasts and 8 with SSc fibroblasts. See Figure 2 for other definitions.

DISCUSSION

This study was undertaken to clarify the involvement of αvβ5 in the self-activation system in SSc fibroblasts. We confirmed our previous observation that αvβ5 is up-regulated in SSc fibroblasts and demonstrated that blockade of αvβ5 by anti-αvβ5 antibody or β5 antisense oligonucleotide significantly reduced the COL1A2 promoter activity and the Smad3 phosphorylation level in these cells. Since previous reports demonstrated that all αv-containing integrins bind to LAP-β1 and have the potential to modulate the localization and possibly activation of SLC, these results suggest that αvβ5 may recruit and activate SLC on the cell surface of SSc fibroblasts. To further reinforce the validity of this hypothesis, we performed coculture assays using TMLC cells and demonstrated that active TGFβ is localized on the cell surface of SSc fibroblasts and that αvβ5 may be involved in this process. Finally, we confirmed that blockade of αvβ5 reversed the myofibroblastic phenotype of SSc fibroblasts as well as their COL1A2 gene expression. To our knowledge, this is the first report of findings indicating that the up-regulated expression of αvβ5 contributes to the establishment of autocrine TGFβ signaling in SSc fibroblasts through activation of endogenous LTGFβ1.

Dermal fibroblast activation is an important process in the pathogenesis of SSc (2–5). We have reported that the activation of SSc fibroblasts may be a result of stimulation by autocrine TGFβ1 and that up-regulation of TGFβ receptors may contribute to this process (6–9). However, the biologic effect of cytokines, including TGFβ1, is mainly determined by the occurrence of cytokine–receptor interaction, which is modulated by the concentration and the activity of cytokines and/or their receptors. Therefore, the concentration and/or activity of TGFβ1, as well as the expression levels of its receptors, are important factors in the pathogenesis of SSc. Although previously (7) and in the present study we found that there was no significant difference in the levels of total (latent plus active) and active TGFβ1 protein in conditioned media from cultured normal and SSc fibroblasts, recruitment and/or activation of LTGFβ1 in the pericellular region may enhance the incidence of interaction between active TGFβ1 and its receptors, leading to the activation of SSc fibroblasts. In this study, we demonstrated that one of molecules mediating this process is αvβ5, and this finding greatly furthers understanding of the establishment of autocrine TGFβ signaling in SSc fibroblasts.

Two mechanisms in the activation of SLC have been reported to date. One is the proteolysis of LAP, which results in the release of active TGFβ from SLC. Proteases such as plasmin, metalloproteinases, aspartic proteases, cysteine proteases, and serine proteases have been reported to be involved in this process (13, 15, 23). A previous study demonstrated that αvβ8 activates SLC via an MT1-MMP–dependent proteolytic pathway (14).

The other mechanism is the conformational change of LAP, leading to the activation of SLC. This nonproteolytic process is thought to be dependent on an intrinsic ability of LAP to adopt different conformations (24). TSP-1 and αvβ6 have been demonstrated to be involved in this process (11, 12, 23). SLC binds to TSP-1 through the N-terminus of LAP, and this interaction induces a conformational change and subsequent activation of SLC, although the active TGFβ molecules remain bound to TSP-1. SLC also interacts with αvβ6 through the C-terminus of LAP, but this interaction is not sufficient for its activation. Following binding, αvβ6 must interact with actin cytoskeleton to activate bound SLC. The detailed mechanism has not previously been elucidated, but the present results demonstrate that αvβ5 activates SLC. Given the previous finding that, of β1, β3, β5, and β8, only β5 has a cytoplasmic domain that is highly homologous with those of the β6 subunit, which can interact with the αv subunit (25), αvβ5 may activate SLC via a nonproteolytic pathway. Further studies are needed to clarify this.

Since autocrine TGFβ signaling may play a central role in the fibroblast activation in SSc, TGFβ and its receptors could be an attractive target for treatment of this disease. However, because TGFβ is one of the major regulators of the immune system, TGFβ1-null animals develop massive autoimmune inflammation affecting multiple organs (26). Furthermore, selective expression of kinase-deficient TGFβ receptor type II in fibroblasts leads to paradoxical ligand-dependent activation of TGFβ signaling and causes skin and lung fibrosis in transgenic mice (27).

Given the above findings, molecules that can modulate the activation state of LTGFβ and/or its interaction with TGFβ receptors, such as αvβ5, could be the next therapeutic target in SSc. The evidence that blockade of αvβ5 inhibits autocrine TGFβ signaling in SSc fibroblasts without affecting the responsiveness to exogenous TGFβ stimulation in normal fibroblasts suggests that functional blocking of αvβ5 can selectively diminish the pathologically important signaling, i.e., the αvβ5-dependent activation of SLC, without inducing any undesirable effects on the normal TGFβ signaling that is indispensable for normal biologic activities. In a previous study it was demonstrated that β5-knockout mice develop, grow, and reproduce normally and show no abnormality in wound healing or susceptibility to adenovirus infection, which are major biologic processes in which αvβ5 participates (28). These observations suggest that most roles of αvβ5 can be compensated for by other αvβ5-independent pathways. Taken together, these findings indicate that this functional redundancy in αvβ5 makes pharmacologic interference with αvβ5 functions a promising approach to the treatment of SSc.

In summary, the results of the present study demonstrate that up-regulation of αvβ5 contributes to the establishment of autocrine TGFβ signaling in SSc fibroblasts. Although in vivo studies using animal models will be needed in the future, the present data suggest that this integrin could be an attractive therapeutic target in SSc.

Ancillary