INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Idiopathic pulmonary fibrosis (IPF) is a chronic interstitial lung disease characterized by the accumulation of inflammatory cells in the lower respiratory tract, including alveolar macrophages (AM) and neutrophils, followed by parenchymal cell injury and a progressive interstitial fibrosis (1). The mechanism of the fibrogenic process in IPF has been extensively investigated, and multiple cells, cytokines, and mediators have been found to be involved in this process (1, 2, 4). Of these cells, myofibroblasts are particularly important, since they are major producers of extracellular matrix (ECM) proteins (5, 6) and are associated with the deposition of ECM proteins (5, 7, 8). Myofibroblast hyperplasia has been demonstrated in areas of active fibrosis in patients with pulmonary fibrosis and in bleomycin (BLM)-induced pulmonary fibrosis (7, 9). Thus, myofibroblasts play an important role in the fibrogenic response of pulmonary fibrosis. Several cell types present in the lung are potential candidates as precursors of myofibroblasts, which appear to originate from fibroblasts under the influence of various factors (6, 12). Among these factors, transforming growth factor (TGF)-₁ is well known to induce the phenotypic modulation of fibroblasts to myofibroblasts (13), and increased expression of TGF-₁ associated with active fibrosis has been shown to occur in pulmonary fibrosis (14, 15). Therefore, it is important to clarify the mechanism responsible for the phenotypic modulation of human lung fibroblasts (HLF) to myofibroblasts upon TGF-₁ stimulation.

Many extracellular stimuli elicit specific biologic responses through activation of mitogen-activated protein kinase (MAPK) cascades (16, 17). Three subgroups of mammalian MAPKs have been molecularly characterized: extracellular signal-regulated kinase (Erk), p38 MAPK, and c-Jun-NH₂-terminal kinase (JNK). p38 MAPK and JNK are activated by environmental stresses such as hyperosmotic shock, heat shock, cold shock, ultraviolet (UV) irradiation, and inflammatory cytokines, and play important roles in apoptosis and cytokine expression (17), whereas Erk is activated by mitogenic stimuli and plays a central role in cell proliferation and differentiation (21). However, recent studies have suggested that Erk and JNK also play important roles in the signal cascades leading to the induction of various inflammatory mediators, including cytokines and chemical mediators (22, 23). Thus, the MAPK superfamily regulates a variety of cellular functions; however, the role of the MAPK superfamily in the TGF-₁-induced phenotypic modulation of HLF to myofibroblasts has not been determined.

In the present study, we therefore examined the role of JNK, p38 MAPK, and Erk in inducing the myofibroblastic phenotype characterized by -smooth-muscle actin expression (24) in TGF-₁-stimulated HLF. To this end, we examined -smooth-muscle actin expression and JNK, p38 MAPK, and Erk phosphorylation and activity in TGF-₁-stimulated HLF, and the effect of CEP-1347 as a specific inhibitor of the JNK-mediated signaling pathway (25), SB 203580 as a specific inhibitor of p38 MAPK (26), and PD 98059 as a specific inhibitor of MAPK-1 (MEK-1), which is upstream of Erk (27) in the induction of -smooth-muscle actin expression in TGF-₁-stimulated HLF.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Reagents

SB 203580, PD 98059, and anti-TGF-₁ antibody were obtained from Carbiochem-Novabiochem Corporation (La Jolla, CA), New England Biolabs Inc. (Beverly, MA), and Genzyme TECHNE (Minneapolis, MN), respectively. CEP-1347 was kindly provided by Cephalon Inc. (West Chester, PA). SB 203580, PD98059, and CEP-1347 were dissolved in dimethylsulfoxide (DMSO).

Cell Cultures

HLF (Clonetics, San Diego, CA) were placed onto collagen-coated tissue culture plates (IWAKI, Tokyo, Japan) for Western blot analysis and in vitro kinase assay, or onto chamber slides (Nunc Inc., Naperville, IL) coated with collagen type I (Sigma Chemical Corp., St. Louis, MO) for immunocytochemical study, and were cultured in fibroblast growth medium (FGM-2; Clonetics) containing 0.2% fetal bovine serum (FBS), gentamycin and amphotericin B, fibroblast growth factor (FGF), and insulin at 37° C in a humidified 5% CO₂ atmosphere. When the cells were grown in subconfluent conditions, the culture medium was replaced with FGM-2 medium without FBS, FGF, or insulin (medium) for 16 h. The cells reached 80% confluence. The density of cell cultures did not change during the culture periods regardless of culture conditions.

Western Blot Analysis of -Smooth-Muscle Actin Expression

Expression of -smooth-muscle actin protein was analyzed by Western blotting with an anti--smooth-muscle actin antibody (American Research Products, Inc., Belmont, MA) as described previously (13).

Western Blot Analysis of JNK, p38 MAPK, and Erk

Phosphorylation of JNK, p38 MAPK, and Erk was analyzed by Western blotting according to the manufacturer's instructions (New England Biolabs, Inc.), using antibodies to phosphorylated threonine and tyrosine residues of JNK, p38 MAPK, and p42/p44 MAPK, as described previously (23). Each antibody reacts specifically with the corresponding kinase.

JNK, p38 MAPK, and Erk Assay

JNK, p38 MAPK, and Erk activities were analyzed with a commercially available SAPK/JNK assay kit, a p38 MAPK assay kit, and a MAPK assay kit (all from New England Biolabs, Inc.) according to the manufacturer's instructions as described previously (23).

Confocal Laser Microscopy

Fixed HLF cells were incubated with a specific antibody to -smooth-muscle actin (ZYMED Laboratories, Inc., San Francisco, CA), followed by incubation with fluorescein isothiocyanate-conjugated antimouse IgG antibody for 60 min. The cells were viewed with an Olympus confocal laser microscope (Olympus Kogyo Corp., Ltd., Tokyo, Japan).

Measurement of TGF-₁

The concentrations of TGF-₁ in the culture supernatants were measured with commercially available enzyme-linked immunosorbent assay kits (Genzyme TECHNE) according to the manufacturer's instructions. The minimum detectable concentration of TGF-₁ was 7 pg/ml.

Statistical Analysis

Statistical significance was analyzed through analysis of variance (ANOVA). Values of p < 0.05 were considered significant. When statistical significance was reached, post hoc tests (Fisher's protected least significant difference test and Scheff's F test) were performed.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

TGF-₁ Induces -Smooth-Muscle Actin Expression in HLF

First, we investigated for a dose-dependent effect of TGF-₁ on -smooth-muscle actin expression in HLF. To this end, cell lysates from HLF stimulated with various concentrations of TGF-₁ were immunoblotted with anti--smooth-muscle actin antibody on each of 6 d after cultivation (Figure 1a). The expression of -smooth-muscle actin in TGF-₁-stimulated HLF increased in a dose-dependent manner. Next, we examined the kinetics of the effects of TGF-₁ on -smooth-muscle actin expression. To this end, we stimulated HLF with 10 ng/ml of TGF-₁ and immunoblotted the cell lysates at 2, 4, and 6 d after cultivation (Figure 1c). The expression of -smooth-muscle actin in TGF-₁-stimulated HLF increased in a time-dependent manner. Figures 1a and 1c show representative results of each of three experiment. Figures 1b and 1d show the mean ± SD of quantities of -smooth-muscle actin proteins in each of three experiments. We measured the concentrations of TGF-₁ in the culture supernatants from the TGF-₁-unstimulated control HLF on Days 2, 4, and 6 after cultivation. TGF-₁ concentrations on Days 2, 4, and 6 were below the assay sensitivity limit, at 28 ± 7 pg/ml and 44 ± 8 pg/ml (mean ± SD of three different experiments), respectively. Anti-TGF-₁ antibody at sufficient concentrations for neutralizing 1 ng/ml of TGF-₁ abolished slight increases in -smooth-muscle actin expression in TGF-₁-unstimulated control HLF on Day 4 and Day 6 (data not shown). These results indicated that TGF-₁ constitutively produced by TGF-₁-unstimulated control HLF contributed to slight increases in -smooth-muscle actin expression on Days 4 and 6. 

View larger version (21K): [in this window] [in a new window]   Figure 1.   TGF-1 induced -smooth-muscle actin expression. HLF were cultured either with medium or with various concentrations of TGF-1 for 6 d (a and b). HLF were cultured either with medium (open circles) or with 10 ng/ ml of TGF-1 (closed circles) for 2, 4, and 6 d (c and d ). The cell lysate containing 10 µg of protein separated by 15% sodium dodecylsulfate-polyacrylamide gel electrophoresis was electrophoretically transferred to a membrane and the membrane was incubated with antibody specific to - smooth-muscle actin. It was then incubated with a horseradish peroxidase-conjugated antirabbit IgG antibody and an HRP-conjugated antibiotin antibody to detect biotinylated protein markers. Blots were incubated with enhanced chemiluminescence (Amersham International, Aylesbury, UK) solution for 1 min and exposed on Kodak (Rochester, NY) XAR film. The amounts of -smooth-muscle actin proteins were quantified with the NIH Image Analyzer program and are presented as the amounts of -smooth-muscle actin proteins relative to those in control cells treated without TGF-1 (1.0). The multiplicities of increases in amounts of -smooth-muscle actin proteins are indicated beneath the blot films. Three identical experiments, performed independently, gave similar results. The results are expressed as mean ± SD (b and d ). *1: p < 0.05 compared with the amounts of -smooth-muscle actin proteins in the cells cultured with medium. *2: p < 0.01 compared with the amounts of -smooth-muscle actin proteins in the cells cultured with medium.

TGF-₁ Induces JNK, p38 MAPK, and Erk Phosphorylation

To determine whether TGF-₁ could stimulate the threonine and tyrosine phosphorylation of JNK, p38 MAPK, and Erk, we stimulated HLF with TGF-₁, and immunoblotted JNK, p38 MAPK, and Erk in the cells at the times indicated. Amounts of phosphorylated threonine and tyrosine of JNK increased at 30 min, were maximal at 60 min, and thereafter returned to nearly basal levels at 180 min (Figure 2a, upper panel ). Amounts of phosphorylated threonine and tyrosine of p38 MAPK increased at 30 min, were maximal at 60 min, and thereafter returned to nearly basal levels at 180 min (Figure 2b, upper panel ). Amounts of phosphorylated threonine and tyrosine of Erk also increased at 30 min, were maximal at 60 min, and thereafter returned to nearly basal levels at 180 min (Figure 2c, upper panel ). The lower panel of Figure 2a shows that equal amounts of JNK protein were immunoblotted with the phosphorylation-independent, JNK-specific antibody regardless of the culture period, indicating that TGF-₁-induced JNK phosphorylation occurred in the absence of changes in JNK protein levels. Similarly, TGF-₁-induced p38 MAPK phosphorylation occurred in the absence of changes in p38 MAPK protein levels (Figure 2b, lower panel ), and TGF-₁- induced Erk phosphorylation occurred in the absence of changes in Erk protein levels (Figure 2c, lower panel ).

View larger version (58K): [in this window] [in a new window]   Figure 2.   TGF-1 induced the threonine and tyrosine phosphorylation of JNK, p38 MAPK, and Erk. HLF were stimulated with 10 ng/ml of TGF-1 for the times indicated. The cell lysate containing 10 µg of protein separated by 15% sodium dodecylsulfate-polyacrylamide gel electrophoresis was electophoretically transferred to a nitrocellulose membrane and the membrane was blotted with an antibody to phosphorylated threonine and tyrosine residues of JNK Ab (phospho-JNK; a, upper panel ), an antibody to phosphorylated threonine and tyrosine residues of p38 MAPK (phospho-p38 MAPK; b, upper panel ) or an antibody to phosphorylated threonine and tyrosine residues of p42/p44 MAPK (phospho-Erk; c, upper panel ). It was then incubated with an HRP-conjugated antirabbit IgG antibody and HRP-conjugated antibiotin antibody to detect biotinylated protein markers. Blots were incubated with enhanced chemiluminescence solution for 1 min and exposed on Kodak XAR film. Blots were stripped and reprobed with a phosphorylation-state-independent anti-JNK antibody (JNK; a, lower panel ), phosphorylation-state-independent anti-p38 MAPK antibody (p38 MAPK; b, lower panel ), and antibody phosphorylation-state-independent anti-p42/p44 MAPK antibody (Erk; c, lower panel ) to determine total JNK, p38 MAPK antibody, and Erk levels, respectively. Lanes P of a, b, and c represent protein-positive preparations from 293 cells treated with UV light for phosphorylated threonine and tyrosine of JNK, protein-positive preparations from C-6 glioma cells stimulated with anisomycin for phosphorylated threonine and tyrosine of p38 MAPK, and protein-positive preparations from Escherichia coli expressing active p42/p44 MAPK as a result of coexpression with a constitutively active form of its activator, MEK2 for phosphorylated threonine and tyrosine of Erk (New England Biolabs, Inc.), respectively. Lanes N of a, b, and c represent protein-negative preparations from 293 cells untreated with UV light for JNK, protein-negative preparations from C-6 glioma cells unstimulated with anisomycin for p38 MAPK and bacterially expressed, and a kinase-inactive p42 MAPK protein-negative control (New England Biolabs, Inc.), respectively. When the cells were cultured with medium instead of TGF-1, amounts of phosphorylated JNK, p38 MAPK, and Erk did not increase over the time of study. The amounts of phosphorylated JNK, p38 MAPK, and Erk proteins were quantified with the NIH Image Analyzer program and are presented as the amounts of phosphorylated JNK, p38 MAPK, and Erk proteins relative to those in control cells treated without TGF-1 (1.0). The multiplicities of increases in amounts of phosphorylated JNK, p38 MAPK, and Erk proteins are indicated beneath the blots. Three identical experiments independently performed gave similar results.

JNK, p38 MAPK, and Erk Activity, and the Effect of CEP-1347, SB 203580, and PD 98059 on JNK, p38 MAPK, and Erk Activity, Respectively

Activation of JNK, p38 MAPK, and Erk is mediated by dual phosphorylation of the threonine and tyrosine residues of JNK, p38 MAPK, and Erk (16, 19), respectively. Increases in threonine and tyrosine phosphorylation of JNK, p38 MAPK, and Erk in TGF-₁-stimulated cells, shown in Figure 2, reflect the activation state of JNK, p38 MAPK, and Erk. In addition to analysis of the threonine and tyrosine phosphorylation of JNK, p38 MAPK, and Erk, we examined whether TGF-₁ could induce JNK, p38 MAPK, and Erk activity, and the effect of CEP-1347 on JNK activity, the effect of SB 20380 on p38 MAPK activity, and the effect of PD 98059 on Erk activity in TGF-₁-stimulated HLF. TGF-₁ induced JNK activity, as demonstrated by increased phosphorylation of c-Jun (Figure 3a); p38 MAPK activity, as demonstrated by the increased phosphorylation of its substrate, ATF-2 (Figure 3d); and Erk activity, as demonstrated by the increased phosphorylation of its substrate, Elk-1 (Figure 3g). CEP-1347, SB 203580, and PD 98059 attenuated TGF-₁-induced increases in JNK activity (Figure 3a), p38 MAPK activity (Figure 3d), and Erk activity (Figure 3g), respectively. CEP-1347 did not inhibit p38 MAPK or Erk activity (Figures 3b and 3c), SB203580 did not inhibit JNK or Erk activity (Figures 3e and 3f), and PD98059 did not inhibit p38 MAPK or JNK activity (Figures 3h and 3i) in TGF-₁-stimulated cells. These results verified the specificity of CEP-1347, SB 203580, and PD 98059 in their inhibitory effect on TGF-₁-induced JNK, p38 MAPK, and Erk activity. Addition of DMSO vehicle alone did not attenuate cytokine- induced increases in JNK, p38 MAPK, or Erk activity (data not shown).

View larger version (37K): [in this window] [in a new window]   Figure 3.   TGF-1 induced JNK, p38 MAPK, and Erk activity, and CEP-1347, SB 203580, and PD 98059 attenuated JNK, p38 MAPK, and Erk activity, respectively. HLF that had been incubated with medium, CEP-1347 (1,000 nmol), SB 203580 (10 nmol), or PD98059 (50 nmol) for 1 h were stimulated with 10 ng/ml of TGF-1. JNK, p38 MAPK, and Erk activity were analyzed at 30 min after stimulation. The inhibitors were present throughout the incubation periods. (a through c) Cells cultured with medium (lane 1), CEP-1347 (lane 2), TGF-1 (lane 3), and TGF-1 and CEP-1347 (lane 4). (d through f  ) Cells cultured with medium (lane 1), SB 203580 (lane 2), TGF-1 (lane 3), and TGF-1 and SB 203580 (lane 4). (g through i ) Cells cultured with medium (lane 1), PD98059 (lane 2), TGF-1 (lane 3) and TGF-1 and PD98059 (lane 4). Lanes P of a, e, and h; lanes P of b, d, and i; and g, c, and f represent phosphorylated JNK, p38 MAPK, and Erk controls for positive protein, respectively. We used cell lysate containing 200 µg of protein for the kinase assays. The SAPK/JNK Assay Kit employs an N-terminal c-Jun fusion protein bound to Sepharose beads to selectively remove JNK from the cell lysate, after which the kinase reaction is conducted in the presence of unlabeled adenosine triphosphate. c-Jun phosphorylation is selectively measured with antiphospho-specific c-Jun antibody that measures JNK-induced phosphorylation of c-Jun. p38 MAPK activity and Erk activity were analyzed by specific immunoprecipitation with an antiphospho-specific p38 MAPK antibody followed by an in vitro kinase assay of its substrate, ATF-2, and by specific immunoprecipitation with an antiphospho-specific p42/p44 MAPK antibody followed by an in vitro kinase assay of its substrate, Elk-1, respectively. The samples were separated by 15% sodium dodecylsulfate-polyacrylamide gel electrophoresis, transferred to membranes, and blotted with antiphospho-specific ATF-2 antibody or antiphospho-specific Elk-1 antibody. The membrane was incubated with HRP-conjugated antirabbit IgG antibody and the membrane was then incubated with the enhanced chemiluminescence solution and exposed on Kodak XAR film for 1 min. The amounts of phosphorylated c-Jun, ATF-2, and Elk-1 proteins were quantified with the NIH Image Analyzer program and are presented as the amounts of phosphorylated c-Jun, ATF-2, and Elk-1 proteins relative to those in control cells treated without TGF-1 (1.0). The multiplicities of increases in amounts of phosphorylated c-Jun, ATF-2, and Elk-1 proteins are indicated beneath the blots. Three identical experiments performed independently gave similar results.

CEP-1347, but not SB 203580 or PD 98059, Attenuates -Smooth-Muscle Actin Expression in HLF

The increases in -smooth-muscle actin expression and the increase in JNK, p38 MAPK, and Erk phosphorylation and activity in HLF were caused by TGF-₁. These observations suggested that -smooth-muscle actin expression might be mediated through JNK-, p38 MAPK- and Erk-dependent pathways. To test this possibility, we stimulated HLF that had been incubated with various concentrations of CEP-1347, SB 203580, or PD 98059 for 1 h with TGF-₁, and examined the expression of -smooth-muscle actin on each of 6 d after cultivation. CEP-1347 attenuated -smooth-muscle actin expression in a dose-dependent manner (Figure 4a), whereas SB 203580 and PD 98059 did not attenuate -smooth-muscle actin expression (Figures 4b and 4c). Figures 4a and 5 show representative results of three experiments and the mean ± SD of relative amounts of -smooth-muscle actin proteins in the three experiments, respectively. To further examine the effect of TGF-₁ on -smooth-muscle actin expression in HLF, we immunocytochemically stained HLF with an antibody specific to -smooth-muscle actin and viewed the expression of -smooth-muscle actin with confocal laser microscopy. HLF stimulated with TGF-₁ were intensely stained with antibody to -smooth-muscle actin (Figure 6c), whereas unstimulated HLF were homogeneously and very weakly stained (Figure 6a). CEP-1347 attenuated TGF-₁-induced increases in staining of -smooth-muscle actin (Figure 6d). These results indicated that -smooth-muscle actin expression as determined immunocytochemically was correlated with its expression as determined by immunoblotting. Addition of DMSO vehicle alone did not attenuate -smooth-muscle actin expression in either the control cells or the TGF-₁-stimulated cells (data not shown). The total number of cells, cell viability as determined by trypan blue exclusion, and total cell protein at the end of the culture period of each experiment, shown in Figures 1-6, did not differ with culture conditions (data not shown).

View larger version (44K): [in this window] [in a new window]   Figure 4.   CEP-1347, but not SB 203580 or PD 98059, attenuated TGF-1-induced -smooth-muscle actin expression. HLF that had been incubated with medium, various concentrations of CEP-1347 (a), SB 203580 (b), or PD98059 (c) for 1 h were cultured with medium or 10 ng/ml of TGF-1 for 6 d, and the expression of -smooth-muscle expression was analyzed by Western blotting with an antibody to -smooth-muscle actin. Three identical experiments performed independently gave similar results.

View larger version (47K): [in this window] [in a new window]   Figure 6.   Immnunocytochemi-cal analysis of -smooth-muscle actin expression. HLF that had been incubated either with medium or with CEP-1347 (1,000 nM) for 1 h were cultured with medium or 10 ng/ml of TGF-1 for 6 d, and the cells were immunocytochemically stained with an antibody specific to -smooth-muscle actin. The expression of -smooth-muscle actin was viewed with confocal laser microscopy. The cells were cultured with medium (a), CEP-1347 (b), TGF-1 (c), and TGF-1 and CEP1347 (d ). Three identical experiments performed independently gave similar results.

View larger version (13K): [in this window] [in a new window]   Figure 5.   CEP-1347 attenuated TGF-1-induced -smooth-muscle actin expression. The amounts of -smooth-muscle actin proteins in cells that had been incubated either with medium or with various concentrations of CEP-1347 for 1 h were cultured with medium (open circles) or with 10 ng/ml of TGF-1 (closed circles) for 6 d were quantified with the NIH Image Analyzer program. The relative amounts of -smooth-muscle actin proteins were calculated as the amounts of -smooth-muscle actin proteins relative to those in control cells treated without TGF-1 (1.0). The results are expressed as mean ± SD of the relative amounts of -smooth-muscle actin proteins in three different experiments. *1: p < 0.01 compared with the amounts of -smooth-muscle actin proteins in TGF-1-stimulated cells cultured without CEP-1347.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In the present study, we analyzed the intracellular, signal-inducing, myofibroblastic phenotype of HLF characterized by the expression of -smooth-muscle actin. The results showed that: (1) TGF-₁ caused the phenotypic modulation of HLF to myofibroblasts in a dose- and a time-dependent manner; (2) TGF-₁ caused increases in JNK, p38 MAPK, and Erk phosphorylation and activity; (3) CEP-1347, SB 203580, and PD 98059 attenuated TGF-₁-induced JNK, p38 MAPK, and Erk activity, respectively; and (4) CEP-1347, but not SB 203580 or PD 98059, attenuated the TGF-₁-induced phenotypic modulation of HLF to myofibroblasts in a dose-dependent manner. These results indicate that TGF-₁ is capable of inducing the myofibroblastic phenotype of HLF, and that JNK regulates the phenotypic modulation of TGF-₁-stimulated HLF to myofibroblasts.

Myofibroblasts play an important role in the fibrogenic response of pulmonary fibrosis. Several cell types present in the lung are potential candidates for being the origin of myofibroblasts, and myofibroblasts appear to originate from fibroblasts under the influence of various factors (6, 12). Among these factors, TGF-₁ is well known to induce the phenotypic modulation of fibroblasts to myofibroblasts (13). An increased expression of TGF-₁ has been shown in lung tissues and AM from patients with pulmonary fibrosis and in animal models of pulmonary fibrosis (14, 15), and TGF-₁ may therefore be involved in the phenotypic modulation of lung fibroblasts to myofibroblasts in the fibrogenic process of pulmonary fibrosis. In the present study, we analyzed the role of the MAPK superfamily in the TGF-₁-induced phenotypic modulation of HLF to myofibroblasts characterized by -smooth-muscle actin expression. TGF-₁ activated JNK, p38 MAPK, and Erk. Specific inhibitors of the p38 MAPK and Erk signaling pathways have been identified, providing effective tools for investigating the role of p38 MAPK and Erk in cellular signaling (26, 27). CEP-1347 is a novel inhibitor of the JNK signaling pathway (25). In the present study, we used these inhibitors to elucidate the biologic functions of different p38 MAPK, Erk, and JNK in -smooth-muscle actin expression in TGF-₁-stimulated HLF. Concentrations up to 10 µM SB 203580 and up to 50 µM PD 98059 were used in the study to examine the effect of these inhibitors on -smooth-muscle actin expression, since previous studies of the role of p38 MAPK and Erk in eliciting various biologic responses had shown that 10 µM SB 203580 and 50 µM PD 98059 almost completely inhibited such expression (26). A concentration of 1 µM CEP-1347 can rescue motoneurons undergoing apoptosis and inhibit JNK activity (25). Consequently, the concentrations of 10 µM SB 203580, 50 µM PD 98050, and 1 µM CEP-1347 used in the present study were sufficient concentrations to examine the signal-transduction pathway. TGF-₁ simultaneously increased JNK, p38 MAPK, and Erk activity. Concentrations of 1 µM CEP-1347, 10 µM SB 203580, and 50 µM PD 98059 almost completely attenuated TGF-₁-induced increases in JNK, p38 MAPK, and Erk activity, respectively. A concentration of 1 µM of CEP-1347, but not 10 µM SB 203580 or 50 µM PD 98059, almost completely attenuated TGF-₁-induced -smooth-muscle actin expression. These results indicated that the JNK-dependent pathway, but not the p38 MAPK or Erk-dependent pathways, participate in regulating TGF-₁-induced phenotypic modulation of HLF to myofibroblasts.

TGF-₁ belongs to the TGF- superfamily, which regulates various cell functions, such as cell proliferation, cell differentiation, apoptosis, cell adhesion and motility, and ECM production (29). TGF-₁ exerts its effects through heteromeric receptor complexes composed of type I and type II serine/threonine receptors (29, 30). Upon ligand binding, the type II receptor phosphorylates the type I receptor to activate its kinase activity (29, 30). The Smad family was recently identified as mediating the intracellular signaling downstream of these receptor complexes (29, 30). Other intracellular signaling pathways have also been shown to be activated by TGF-₁, including those of the MAPK superfamily, but this has not previously been tied to the increased expression of -smooth-muscle actin caused by TGF-₁. TGF-₁ activates JNK, p38 MAPK, and Erk in a variety of cell systems (31). Although the biologic functions of these kinases are not fully understood, it has been recently reported that TGF-₁-activated JNK mediates fibronectin production (34). In the present study, we showed that TGF-₁-activated JNK mediates the phenotypic modulation of HLF to myofibroblasts. These results indicate a novel biologic function of JNK in the TGF-₁-activated intracellular signal-transduction pathway in fibroblasts.

From the data presented here, we conclude that the TGF-₁- activated JNK pathway regulates the phenotypic modulation of HLF to myofibroblasts. The pathogenesis of IPF is complex, and there is no established treatment to halt the development of fibrosis (3, 35). Our results for the role of a JNK-dependent pathway in the TGF-₁-induced phenotypic modulation of HLF to myofibroblasts are important in understanding the fibrogenic process of pulmonary fibrosis, and suggest that a strategy for attenuating phenotypic modulation of HLF to myofibroblasts with specific inhibitors of the JNK cascade may have beneficial effects in preventing the development of fibrosis.