INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

A striking feature of idiopathic pulmonary fibrosis (IPF) is irreversible fibrosis resulting from excessive remodeling of damaged alveoli (1). Among inflammatory cells accumulating in the alveoli, neutrophils are predominant during the initial phase, followed by infiltrations of lymphocytes and macrophages into the injured area. In the alveolar walls, fibroblasts migrate, proliferate, and remodel the respiratory space. These fibroblasts not only fill the respiratory space but also secrete collagen and matrix proteins in response to many cytokines (2). Although corticosteroids, immunosupressive agents (2), neutrophil elastase inhibitor (3), hepatocyte growth factor (4), and interferon gamma-1b (5) have been proposed as treatment agents for IPF, IPF remains a fatal disorder with a 3 to 6 yr median range of survival (6). Thus, the first line of treatment of IPF has not yet been established.

The small GTPase, RhoA, which belongs to the Rho subfamily, controls cell adhesion and its motility via actin-cytoskelton reorganization and actin-myosin filament bundles regulation. ROCK (Rho-associated coiled-coil-forming protein kinase) was isolated by T. Ishizaki and coworkers as a downstream target protein of Rho. ROCK is divided into two isoforms, ROCK-I and ROCK-II, corresponding to ROK and ROK, respectively. Although the functional differences between ROCK-I and ROCK-II are unknown, ROCK-I mRNA was ubiquitously expressed except in the brain and muscle, whereas ROCK-II mRNA was expressed abundantly in the brain, muscle, heart, lung, and placenta. The active form of RhoA binds to ROCK, which phosphorylates myosin phosphatase, resulting in an increase in myosin light chain (MLC) phosphorylation. The Rho/ROCK signaling pathway also contributes to Ca²⁺ sensitization of smooth muscle contraction. We have demonstrated that the Rho/ROCK signaling pathway plays a central role in Ca²⁺ sensitization of airway smooth muscle contraction through inhibition of myosin phosphatase (7). Recent studies have revealed that the Rho/ROCK-mediated pathway regulates not only smooth muscle contraction but also cell motility (8). Bleomycin (BLM), an antineoplastic agent, induces pulmonary fibrosis in mice, rats, and humans. BLM-induced pulmonary fibrosis has been established in several animal models (9). Y-27632, ([+]-[R]-trans-4-[1-aminoethyl]-N-[4-pyridyl] cyclohexane carboxamide dihydrochloride, monohydrate), a highly selective inhibitor of ROCKs both in vitro and in vivo (10), blocked human neutrophil migration in response to chemoattractants through suppression of MLC phosphorylation in vitro (11). Continuous delivery of Y-27632 using an osmotic pump reduced the rates of dissemination of tumor cells implanted into the peritoneal cavities of rats in vivo (12). Y-27632 is a powerful tool for investigating the role of the Rho/ROCK signaling pathway in vivo. We hypothesized that the Rho/ROCK-mediated pathway plays a role in infiltration of inflammatory cells and the fibloblasts into injured lungs resulting in the development of lung fibrosis.

To test this hypothesis, we employed a mouse lung fibrosis model with BLM, and examined the inhibitory effect of Y-27632. We evaluated histological findings of the BLM-induced lung fibrosis by counting fibrotic score, measuring hydroxyproline (OH-proline) content in the lungs and characterizing the inflammatory cells of bronchoalveolar lavage fluid (BALF). To investigate the mechanisms of Y-27632-induced inhibition of the lung fibrosis model, various in vitro chemotaxis assays were also performed. In addition, we verified the expression of ROCK-II mRNA and its protein levels, and measured total MLC20 phosphorylation levels of lung homogenates from the mice.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Study Design and Experimental Protocol

Pathogen-free 6-wk-old female C57BL/6 mice weighing 13-15 g, from Charles River (Yokohama, Japan) were used for the experiments according to guidelines set by the Council of Animal Care and Experimentation Committee, Gunma University, Showa Campus Japan. The animals were maintained under standard conditions with free access to water and rodent laboratory food. Figure 1 shows the protocol of our BLM-induced IPF model. Thirty-six mice were equally divided into six groups (group A to F) to determine fibrotic score and OH-proline content. BLM accumulates in the subpleural regions, resulting in the preferential development of lung fibrosis at subpleural lesions. This is very similar to the pathological features of human IPF (9). BLM (Bleo; Nippon Kayaku, Tokyo, Japan), Y-27632 (WelFide Corporation [Yoshitomi Pharmaceutical Industries Ltd], Fukuoka, Japan), and saline were administered via intraperitoneal injection. BLM diluted in 200 µl saline was injected intraperitoneally at 40 mg/kg body weight on Days 0, 2, 4, 6, and 8. To produce severe fibrosis in the murine IPF model, we administered a total of 200 mg of BLM. This is a high dose compared with those used in other investigations (13). Y-27632 at 0.1, 10, and 100 µg/kg body weight diluted in 100 µl sterile saline was injected into mice every other day from Day 0 to Day 40. Control mice were treated with saline in a similar manner (Figure 1). Body weights were measured before every administration of the compounds. Mice were sacrificed by inhalation of diethyl ether, and their thoraces were then exposed. The lungs were washed with cold phosphate-buffered saline (PBS) and surgically removed. The excised lungs were used for histopathological examination and assayed for OH-proline contents. The left lungs were used to evaluate the fibrotic score by histological examination, and the right lungs for measurement of OH-proline contents. The other 27 mice were also divided equally into three groups (group A, D, and F) to determine cell differentiation of BALF, expression of RhoA and ROCK-II protein, as well as expression of ROCK-II mRNA and MLC20 phosphorylation in lungs. Group A was treated with BLM (40 mg/kg × five times and saline every other day), group D with BLM (40 mg/kg × five times), and Y-27632 (100 µg/kg) every other day, and group F with saline alone every other day. BAL was performed on Days 7, 14, 21, and 40 after initial injection of BLM. Mice were anesthetized by inhalation of diethyl ether and BAL was performed. After receiving BAL, the lungs were perfused with cold PBS to wash out blood from the tissue, and the lungs were then homogenized to analyze the expression of RhoA and ROCK-II at mRNA and protein levels, and MLC20 phosphorylation. Another group of 20 mice was used for the analysis of Y-27632 pharmacokinetics.

View larger version (31K): [in this window] [in a new window]   Figure 1.   Experimental protocol. All reagents were administered to the mice intraperitoneally. Mice were sacrificed on Days 0, 7, 14, 21, and 40. BLM (bleomycin) was administered 40 mg/kg × five times every other day in groups A through D. Each concentration of Y-27632 was administered every other day in groups B through D from Day 0 to Day 40. Pretreatment of Y-27632 was performed 30 min before injection with BLM. All compounds were administered by intraperitoneal injection.

Histologic Examination

Morphological evaluation of fibrotic changes in the lungs was performed on Day 40. The excised lungs were immediately fixed with 10% formaldehyde neutral buffer solution for 48 h, and then embedded in paraffin. Sagittal sections were cut at 2 µm thickness and stained with hematoxylin-eosin and Masson-trichrome. The total lung area of the sections was used for the fibrotic scale microscope evaluation (Olympus, BX50F4). Criteria for grading lung fibrosis were according to the method reported by Ashcroft and coworkers (14): Grade 0, normal lung; Grade 1, minimal fibrous thickening of alveolar or bronchiolar walls; Grade 3, moderate thickening of walls without obvious damage to the lung architecture; Grade 5, increased fibrous with definite damage to lung architecture and formation of fibrous bands or small fibrous masses; Grade 7, severe distortion of architecture and large fibrous area; Grade 8, total fibrous obliteration of the field. Severity of fibrotic changes in each lung section was assessed as the mean score for severity from the observed microscopic fields. The grade of lung fibrosis was scored on a scale from 0 to 8 by examining 20 randomly chosen regions per sample at a magnification of ×100. To minimize investigator variability, all histological specimens were randomly numbered and scored by another investigator in a single blinded fashion.

OH-Proline Assay

OH-proline contents of the lungs were measured objectively to estimate lung fibrosis (15). The right lungs of each mouse were dissected free from major bronchi, and the wet weights were measured. They were hydrolyzed in 500 µl of 12 N hydrochloric acid and in the same aliquot of distilled water at 110° C 20 h, in dry block. After the resultant hydrolysate was neutralized with sodium hydroxide, a 100-µl supernatant was mixed in 1.5 ml of 0.3 N lithium hydroxide solution. The OH-proline content was determined by high-performance liquid chromatography (HPLC) and expressed as micrograms per right lung.

Serial Changes in Serum Concentration of Y-27632

Plasma concentrations of Y-27632 were measured as follows: 6-wk-old female mice were administered Y-27632 1 or 100 µg/kg intraperitoneally. Serial blood samples were obtained from the orbital sinus of the mice in heparinized microhematocrit tubes at 5 and 15 min after the injection of the compound. The tubes were centrifuged for 15 min at 3,000 rpm, and sera were separated from pellets and stored at 80° C until use. The plasma concentrations of Y-27632 were determined by HPLC and expressed as micromolar.

Bronchoalveolar Lavage and Cell Counting

BAL was performed in groups A, D, and F. After anesthesia with diethyl ether, a tracheal cannula was inserted into the lumen of the cervical trachea after a tracheotomy. The lungs were lavaged five times with 400-µl aliquots of cold PBS after tracheotomy, and the lavage fluid was centrifuged 1,500 rpm for 5 min at 4° C. After removing the supernatants, the cells were suspended in PBS and then 30 µl of trypan blue was added to 30 µl of the cell suspension. Total cell counts were performed, followed by a standard hemocytometer measurement. To characterize the cells that had infiltrated the BAL fluid, we counted cells on smears prepared using a cytospin (Auto Smear CF-12D; Chiyoda seisakusyo, Tokyo, Japan), and May-Grünwald's staining. Two hundred cells were counted at ×200 magnification to identify macrophages, lymphocytes, neutrophils, eosinophils, and epithelial cells (16).

Isolation of Murine Neutrophils and Lung Fibroblasts

To obtain murine neutrophils, 2 ml of 12% (wt/vol) casein sodium (Tokyo Kasei Kogyo Co., Ltd, Japan) was injected intraperitoneally into the 6-wk-old mice (n = 6) (17). Six hours after the injection, the mice were sacrificed using diethyl ether. Five milliliters of cold PBS was injected into mice intraperitoneally, followed by aspiration of the fluids. The cell suspensions from the abdomen were harvested through a cell strainer (40-µm nylon; FALCON 2340) and washed once with PBS and centrifuged 500 × g for 5 min. The collected cells were subsequently treated with 1.5% dextran (Dextran T500; Pharmacia Biotech, Tokyo, Japan) to remove erythrocytes. Separation of neutrophils from the mononuclear cells was carried out using Ficoll Hypaque (Dainippon Pharmaceutical Co., Ltd, Tokyo, Japan). The purity and viability of the isolated neutrophils were greater than 96 and 90%, respectively. To obtain murine lung fibroblasts (MLF), the lungs were removed from mice on Day 21 after the first injection of BLM. Three mice were used for this experiment. Mice were sacrificed by inhalation of diethyl ether, and then their thoraces were exposed. The lungs were washed with cold PBS and surgically removed. MLF were isolated by trypsin digestion of tissues minced to fragments, and MLF strains were established in Dulbecco's modified Eagle medium (DMEM; NIPRO, Tokyo, Japan), supplemented with NaHCO₃, L-glutamine, 1,000 mg/L D-glucose, 1% penicillin-streptomycin, and 10% fetal bovine serum (FBS). Incubations were performed in 5% CO₂-95% air at 37° C in a humidified atmosphere. Only early passage cell cultures (passage 3-6) were used for the experiments (18). The isolated cells were observed microscopically, and all the cell strains used in this study were characterized as fibroblasts by Western blot analysis. The anticytoskeletal antibodies used were anti- smooth muscle actin (sigma), antivimentin (Sigma, St. Louis, MO), and anticytokeratin (Progen).

Cell Culture

MH-S murine alveolar macrophage cells were grown in RPMI medium (RPMI Medium 1640; GIBCO, Grand Island, NY) supplemented with L-glutamine, 1% penicillin-streptomycin (Penicillin-Streptomycin; GIBCO), 10% FBS (EQUITECH-BIO, Inc., TX), and NIH3T3 fibroblast cells were grown in DMEM (NIPRO), supplemented with NaHCO₃, L-glutamine, 1,000 mg/L D-glucose, 1% penicillin-streptomycin, and 10% FBS. Incubations were performed in 5% CO₂-95% air at 37° C in a humidified atmosphere.

Chemotaxis Assay

Chemotaxis assays with MH-S, neutrophils, and MLF were performed using the Boyden chamber technique using 48-microwell chemotaxis chambers with minor modifications. The pore size of the nucleopore polycarbonate filter for MH-S and neutrophil was 5 µm, and that of MLF was 8 µm (Chemotaxicell, Kurabo, Japan). MH-S and MLF were detached with 0.25% Trypsin-EDTA (GIBCO) diluted in PBS and counted using a hemocytometer. Trypsin was promptly inactivated by dilution into RPMI or DMEM containing 0.1% bovine serum albumin (BSA) (Daiichi Pure Chemicals Co., Ltd.), and they were then washed twice in RPMI for MH-S or DMEM for MLF without FBS and resuspended in each medium. Neutrophils were washed twice in RPMI without FBS. Then, 50 µl of the cell suspensions was added to the upper compartment at a final concentration of 1.0 × 10⁶ cells/ml, and medium alone was added to the lower compartment. After incubation for 30 min at 37° C in 5% CO₂, various concentrations of Y-27632 were added to the upper compartments, and then incubated at 37° C in 5 % CO₂ for another 30 min. Lipopolysaccharides (LPS, E. coli 0111:B-4; Sigma) were diluted with FBS-free RPMI at a final concentration of 1 µg/ml for MH-S (19), platelet-derived growth factor-BB human recombinant homodimer (PDGF-BB; UBI, Lake Placid, NY) was diluted in a FBS-free DMEM at a final concentration of 10 ng/ml for MLF (20), recombinant mouse interleukin-1 (IL-1; Genzyme/Techne USA) diluted in FBS-free RPMI at a final concentration of 5 ng/ml, and N-formyl-Met-Leu-Phe (fMLP; Wako, Osaka, Japan) was diluted in FBS-free RPMI at a final concentration of 10⁷ M for neutophils (21). To verify the specificity of the PDGF-BB effect, anti-PDGF neutralizing antibody (R&D Systems) was added at a final concentration of 10 ng/ml to the bottom chamber in the MLF experiment. MH-S and MLF were allowed to migrate for 120 min at 37° C in 5 % CO₂. In the experiments with neutrophils, cells were allowed to migrate for 90 min for IL-1 and 60 min for fMLP. At the end of the incubation periods, the cells on the membranes were fixed in 10% formaldehyde, stained with Giemsa (Muto, Co., Ltd., Japan), and mounted on glass slides. Migrating cells were quantified by counting 10 high-power fields (HPF) at a magnification of ×200 and expressed as mean ± SEM. Each sample was assayed in duplicate and the experiments were repeated six times.

Real Time Quantitative RT-PCR Analysis

The assay of quantitation of mRNA levels of the mice lungs was carried out using a real-time fluorescence detection method as described previously (22). BLM 40 mg/kg every other day in group A, BLM plus Y-27632 100 µg/kg every other day in group D, and saline every other day in group F were used for this experiment after receiving BAL. The left lower lungs were removed from the mice after BAL treatment and were perfused with cold PBS. Lungs were cut with scissors into fragments of 0.5-2.0 mm³ and washed with cold PBS three times. Total mRNA was isolated from the lung tissue using guanidinium thiocyanate solution and oligo(dT)-cellulose spin-column according to the manufacturer's protocol (Quick Prep Micro mRNA Purification Kit; Pharmacia Biotech, Tokyo, Japan). After mRNA isolation, cDNA was prepared from each sample. The RT-PCR amplification was performed using 96-well optical trays and caps with a 25-µl final reaction mixture consisting of the following; dH₂O, × TaqMan EZ buffer, manganase (3 mM), dATP (300 µM), dCTP (300 µM), dGTP (300 µM), dUTP (600 µM), forward primer (200 nM), reverse primer (200 nM), TaqMan fluorescent probe (100 nM), rTth DNA polymerase (2.5 U/µl), AmpErase UNG (1 U/µl), and isolated RNA samples. Specific cDNA (ROCK-II cDNA) was PCR amplified separately using an oligonucleotide probe with a 5' fluorescent reporter dye, 6 FAM (6-calboxyl-fluorescein), and a 3' quencher dye, TAMRA (6-calboxyl-tetramethyl-rhodamine). The 5' to 3' nuclease activity of Taq DNA polymerase cleaved the probe and released the reporter, of which fluorescence could be detected using the laser detector of the ABI Prism 7700 Sequence Detection System (Perkin-Elmer Corp., Japan). After crossing a fluorescence detection threshold, the PCR amplification resulted in a fluorescent signal proportional to the amount of PCR product generated. Relative gene expression was determined based on the threshold cycle of the target gene and of the internal reference gene. The PCR conditions were 50° C for 2 min, 95° C for 5 min for 1 cycle each followed by 30 cycles each of 94° C for 20 s and 62° C for 90 s. Primers were obtained from Biochemical Division, Kurabo (Osaka, Japan) and oligonucleotide hybridization probes were obtained from Applied Biosystems Division, Perkin Elmer (Foster City, CA). Primer and probe sequences were as follows; ROCK-II forward CATGGTGCATTGCGACACA, reverse TCGCCCATAGTAACCATCACCT, TaqManProbe ACATATCGCCCGAGGTTCTGAAATCACAAG, GAPDH forward CTTCACCACCATGGAGAAGGC, reverse GGCATGGACTGTGGTCATGAG, and TaqMan probe, CCTGGCCAAGGTCATCCATGACAACTTT.

Western Blot Analysis

Western blot analyses for RhoA and ROCK-II were performed according to a previously described method (8). BLM 40 mg/kg every other day in group A, BLM plus Y-27632 100 µg/kg every other day in group D, and saline every other day in group F were used for this experiment. The right lower lungs were removed from the same mouse used in the real time quantitative RT-PCR experiments. The lungs were cut with scissors into fragments of 0.5-2.0 mm³, washed with cold PBS three times, and then rinsed in lysis buffer containing 1% sodium dodecyl sulfate, 10% glycerol, and 20 mM dithiothreitol. The lung tissue was then ground with 5 ml of glass homogenizer on ice. MH-S and MLF cells were washed with cold PBS three times and homogenized in a similar manner. All lysates were incubated with lysis buffer at 4° C for 60 min and centrifuged at 16,000 rpm. The supernatants were subjected to SDS polyacrylamide gel for electrophoresis according to Laemli's SDS-PAGE (7.5% gel for ROCK-II, 12.5% gel for RhoA). Separated proteins were transferred onto nitrocellulose membrane (Amersham Life Science Ltd, Buckinghamshire, UK) and the membrane was blocked with 5% skim milk in Tris-buffered saline. The membrane was probed with primary antibodies, anti-ROK (ROCK-II) diluted 1:1,000 for 1 h (Transduction Laboratories, R54520), monoclonal anti-RhoA diluted 1:1,000 overnight (Santa Cruz Biotechnology, 26C4), and monoclonal anti--actin (Sigma, AC74) diluted 1:2,000 for 1 h. Primary antibodies were detected with a horseradish peroxidase (HRP)-conjugated goat anti-mouse IgG antibody (Amersham) diluted 1:2,000 for 1 h and visualized according to standard protocols for the ECL detection system (Amersham Life Science Ltd). Protein concentrations were determined according to the manufacturer's instructions for the Bio-Rad Protein assay with BSA as the reference protein (Bio-Rad, South Richmond, CA). The loading volume of each sample was normalized by -actin on a reprobing membrane. The methods of stripping and reprobing membranes were performed according to the technical instructions for the ECL detection system (Amersham). The RhoA and ROCK-II protein contents relative to the -actin were quantified by scanning an NIH image on a Macintosh computer G3.

MLC20 Phosphorylation Assay

To verify whether BLM treatment increased the MLC phosphorylation level of the total lungs, we performed an MLC20 phosphorylation assay by glycerol-PAGE and Western blot analysis (23). BLM 40 mg/kg every other day in group A, BLM plus Y-27632 100 µg/kg every other day in group D, and saline every other day in group F were used for this experiment. After washing the lungs immediately in ice cold PBS and cutting with scissors, 5% trichloroacetic acid (TCA) was added to the tissue to terminate the reaction in the lungs. To evaluate the maximum extent of MLC20 phosphorylation levels, normal lungs were removed from 8-wk-old C57BL/6 mice and incubated with calyculin A, a potent myosin phosphatase inhibitor (Calbiochem, La Jolla, CA) at a final concentration of 1 µM in a Ca²⁺ (30 µM) containing buffer for 120 min. To evaluate the minimum extent of MLC20 phosphorylation levels, the lungs were incubated with 50 µM 1,2-bis(2-aminophenoxy)ethane-N,N,N'N'- tetraacetic acid (BAPTA)-AM (Calbiochem) in a Ca²⁺-free solution (at the 10 mM final concentration EGTA buffered) for 120 min. TCA was added to the tissue to terminate the reaction in the lungs. To remove TCA, the lungs were washed with acetone containing 10 mM dithiothreitol (DTT) and dried. The dried lung pellets were homogenated in a 300 µl buffer containing 20 mM Tris base, 22 mM glycine, 10 mM DTT, 8 M deionized urea, and 0.1% bromophenol blue. The samples were centrifuged at 16,000 rpm for 10 min, and the supernatants were subjected to polyacrylamide ureagel containing glycerol. The phosphorylated and nonphosphorylated MLC20 were detected by Western blot (see above). The membrane was probed with monoclonal antimyosin antibody (MY-21; Sigma) diluted 1:200 overnight. Bound antibody was detected with HRP-conjugated goat anti-mouse IgG antibody diluted 1:2,000 for 1 h followed by development with an ECL detection system. The fraction of phosphorylated MLC relative to the MLC plus phosphorylated MLC was quantified by scanning an NIH image on a Macintosh G3 computer.

Reagents

Y-27632 was from WelFide Corporation (Yoshitomi Pharmaceutical Industries Ltd). Bleomycin was from Nipponkayaku Industries Ltd, Tokyo, Japan.

Statistical Analysis

All values are shown as mean ± SEM. Statistical analyses of the results were performed using parametric statistics with one-way ANOVA followed by Fisher's least significance test, which was used to detect differences between groups composed of the same number of animals. Kruskal-Wallis test followed by Scheffe's F test was used for nonparametric analysis to detect differences between groups composed of different numbers of animals.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Histopathological Examination

BLM (total of 200 mg BLM/mice) induced widespread subpleural thickening and marked fibrotic leisions. Edema of alveolar septa was apparent and accompanied by inflammatory cells including macrophages, lymphocytes, and neutrophils (Figure 2a and 2a'). These morphological changes were not observed in the saline-treated group F (Figures 2f and 2f'). Y-27632 dramatically reduced edema of alveolar walls and thickening of subpleura. Although the number of macrophages was decreased, they still infiltrated the lungs (Figures 2d and 2d'). The Y-27632 alone group E (1,000 µg/kg) did not induce any morphological changes (Figures 2e and 2e'). The fibrotic score in the BLM alone group A was 3.5 ± 0.4 (n = 6) (Figure 3a). The fibrotic scores in the BLM plus Y-27632 at 0.1, 10 and 100 µg/kg were 2.69 ± 0.4, 1.74 ± 0.3, and 1.98 ± 0.4, respectively. Scores of the Y-27632 alone (1,000 µg/kg) group and saline group were 1.33 ± 0.2 and 1.12 ± 0.3, respectively. The 10 and 100 µg/kg doses of Y-27632 decreased the fibrotic score in the BLM-induced IPF mice model.

View larger version (104K): [in this window] [in a new window]   Figure 2.   Histopathological examination of the murine right lung fields on Day 40 after the first bleomycin administration. Removed lungs were stained with Hematoxilin-eosin (a, d, e, f ) or Masson-trichrome stain (a', d', e', and f'). (a and a') BLM 40 mg/kg × five times and saline every other day (group A). (d and d'): BLM 40 mg/kg × five times and Y-27632 100 µg/kg every other day (group D). (e and e') Y-27632 1,000 µg/kg alone every other day (group E). (f and f') saline alone injections every other day (group F).

View larger version (34K): [in this window] [in a new window]   Figure 3.   Evaluation of fibrotic changes by Ashcroft score in the right lung and OH-proline content in the left lungs removed from the same mice on Day 40. Group A mice were treated with BLM and saline (n = 6), group B mice with BLM and Y-27632 0.1 µg/kg (n = 4), group C mice with BLM and Y-27632 10 µg/kg (n = 5), group D mice with BLM and Y-27632 100 µg/kg (n = 6), group E mice with Y-27632 1,000 µg/kg alone (n = 6), and group F mice with saline alone (n = 6). (a) Fibrotic scale was used to place the quantitative histological analysis of fibrotic changes induced by BLM in numerical order. Criteria for grading lung fibrosis were classified from 0 to 8. Values are expressed as mean ± SEM. Statistical analysis was performed using the Kruskal-Wallis test followed by Scheffe's F test (*p < 0.05 versus the group A). (b) OH-proline contents. Relative contents are expressed as a percentage of the saline-treated control (mean ± SEM). Statistical analysis was performed using one-way ANOVA followed by Scheffe's F test (* p < 0.05 versus group A ).

The Amounts of OH-proline in the Lungs

The OH-proline contents in groups A, B, C, D, and E were 214.8 ± 6.2 (n = 6), 191.7 ± 10.3 (n = 4), 183.1 ± 7.1 (n = 5), 170.7 ± 5.4 (n = 6), and 139.3 ± 6.9 (n = 6) µg/right lung, respectively. The OH-proline content in the saline (without BLM) group was 132.9 ± 2.3 (n = 6) µg/right lung. The relative increases in OH-proline content are shown in Figure 3b. Consistent with the fibrotic scores, Y-27632 decreased the amounts of OH-proline induced by BLM at both 10 and 100 µg/kg. A 54% reduction in OH-proline was achieved at 100 µg/ kg of Y-27632 (p < 0.05). Administration of Y-27632 1,000 µg/ kg alone did not affect the OH-proline content.

Serial Concentrations of Y-27632

Sera from mice were obtained at 5 and 15 min after Y-27632 administration intraperitoneally. The concentrations of Y-27632 both 5 and 15 min after injection of Y-27632 (1 µg/kg, intraperitoneally) were under the detectable limit. However, the concentrations of Y-27632 were detectable: 1.2 ± 0.1 µM and 0.81 ± 0.1 µM (mean ± SEM, n = 5) at 5 and 15 min after injection of Y-27632 (100 µg/kg), respectively.

Analysis of Inflammatory Cells of Bronchoalveolar Lavage Fluid

To evaluate the effect of Y-27632 on BLM-induced infiltration of the inflammatory cells, we counted the number of total cells, alveolar macrophages, lymphocytes, and neutrophils in BAL fluids on Days 7, 14, 21, and 40 (Figures 4a-4d). Although lymphocyte numbers did not differ significantly among the three groups, Y-27632 significantly decreased total cell numbers from Day 7 to Day 21, macrophages from Day 14 to Day 21, and neutrophils from Day 7 to Day 14. 

View larger version (26K): [in this window] [in a new window]   Figure 4.   Total cell number and their characterization in bronchoalveolar lavage fluid (BALF). BAL was performed on Days 7, 14, 21, and 40 after the initial injection of BLM. Group A: BLM and saline every other day (open square); group D: BLM and Y-27632 100 µg/kg every other day (open circle); and group F: saline alone every other day (open triangle) were administered BAL. These mice were different from the mice used for fibrotic scoring and OH-proline measurement. (a) Total cell numbers, (b) macrophage numbers, (c) lymphocyte numbers, and (d) neutrophil numbers, respectively. Results are expressed as mean ± SEM (n = 5). Statistical analysis was performed using one-way ANOVA followed by Fisher's least significance test (* p < 0.05: group A versus group D; § p < 0.05: group A versus group F; +p < 0.05; group D versus group F).

Chemotaxis Assay

LPS-induced migration of MH-S cells reached a peak at a concentration of 1 µg/ml of LPS at 120 min (data not shown). Pretreatment with Y-27632 at 0.1, 1, 10, and 30 µM for 30 min significantly inhibited the LPS-induced migration dose dependently (p < 0.05). Y-27632 (30 µM) completely blocked the LPS response with IC₅₀ of 4.8 ± 2.0 µM (n = 6) (Figure 5a). The IL-1-induced neutrophil migration reached a peak at a concentration of 5 ng/ml at 90 min (data not shown). Similarly, pretreatment with Y-27632 dose dependently inhibited IL-1-induced neutrophil migration with IC₅₀ of 8.4 ± 2.1 µM (n = 6) (Figure 5b). Identical findings were obtained when fMLP (10⁷ µM) was used to migrate the neutrophils (data not shown). The recombinant PDGF-BB-induced MLF migration reached a peak at 10 ng/ml of PDGF at 120 min (data not shown), and Y-27632 inhibited this migration with IC₅₀ of 1.4 µM ± 0.5 (n = 6) (Figure 5c). We verified that anti-PDGF neutralizing antibody abolished migration of MLF induced by PDGF-BB. Identical findings were obtained when NIH3T3 fibroblasts were used for PDGF-BB-induced chemotaxis, and Y-27632 inhibited the migration with IC₅₀ of 1.6 µM ± 0.5 (n = 6) (data not shown). Rho and ROCK were expressed in all preparations (mice lungs, the cells employed in the chemotaxisis experiments) at approximately 21 and 180 kD, respectively (Figures 5d and 5e). Thus, Y-27632 inhibited migration of the cells tested regardless of the stimuli. To exclude the possibility that Y-27632 affects fibroblast growth, we examined the effect of Y-27632 on MLF cell growth by MTT assays. Cells were grown in DMEM with 10% FBS and various concentrations of Y-27632. Cell growth was not affected by Y-27632 at the concentrations from 1 to 10 µM for 6 d culture (data not shown).

View larger version (45K): [in this window] [in a new window]   Figure 5.   Effect of Y-27632 on chemotactic response in vitro. All cells were pretreated with various concentrations of Y-27632 for 30 min, and then each chemoattractant was added. Cell migration was determined using the modified Boyden chamber technique and the cells were observed under 10 HPF (×200) and the number of cells/10HPF was determined. (a) Migration of MH-S cells number after 120 min incubation with LPS (1 µg/ml); (b) migration of neutrophil number after 90 min incubation with IL-1 (5 ng/ml); (c) migration of MLF cell number after 120 min incubation with PDGF-BB (10 ng/ml). A polyclonal anti-PDGF antibody was used as a neutralizing antibody. Data are expressed as mean ± SEM (n = 6). Statistical analysis was performed using the one-way ANOVA test followed by Fisher's least significance test (* p < 0.05; versus the Y-27632 0 µM plus each chemoattractant, LPS [1 µg/ ml], IL-1 [5 ng/ml], and PDGF-BB [10 ng/ml]). (d) Expression of RhoA and (e) ROCK-II protein by immunoblot analysis. Lanes 1, 2, 3, 4, and 5 indicate lungs of C57BL/6 mice, MLF, MH-S, mice peritoneal neutrophils, and NIH3T3, respectively. RhoA expression was approximately 21 kD, and ROCK-II was approximately 180 kD.

Expression of ROCK-II mRNA in the Lungs of BLM-treated Mice

We examined the effect of Y-27632 on ROCK-II mRNA expression in lungs in groups A, D, and F from Day 7 to Day 40 by real time quantitative RT-PCR. As shown in Figure 6, BLM increased expression of ROCK-II mRNA from Day 7 to Day 21 (p < 0.05). The mRNA levels showed a reduced peak on Day 21 with an approximately 9-fold increase compared with group F. However, the mRNA levels decreased from Day 21 to Day 40, a trend that corresponded to the fibrotic stage. Y-27632 did not affect the increase in expression of ROCK-II mRNA induced by BLM.

View larger version (27K): [in this window] [in a new window]   Figure 6.   Changes in ROCK-II mRNA expression in murine left lungs from Day 0 to Day 40 by real time quantitative RT-PCR. Group A: BLM and saline every other day (closed bar), group D: BLM and Y-27632 100 µg/kg every other day (hatched bar), and group F: saline alone every other day (open bar) were used. ROCK-II mRNA expressions were given in comparison with GAPDH mRNA expressions. All data are expressed as mean ± SEM (n = 4). Statistical analysis was performed using the one-way ANOVA test followed by Fisher's least significance test (* p < 0.05; versus the group A Day 7; § p < 0.05: versus the group D Day 7; +p < 0.05; versus the group F).

Western Blot Analysis of Rho and ROCK-II Protein Expressed in BLM-treated Mouse Lungs

In groups A, D, and F, the expression of RhoA in lungs did not change from Day 0 to Day 40 (Figure 7a). ROCK-II protein levels showed a tendency to increase only in the BLM-IPF model (Figures 7b and 7c). Y-27632 had no effect on ROCK-II (mRNA and protein levels) and RhoA expression in response to BLM.

View larger version (34K): [in this window] [in a new window]   Figure 7.   (a) Changes in RhoA expression from Day 0 to Day 40. Group A (BLM and saline every other day), group D (BLM and Y-27632 100 µg/kg every other day), and group F (saline alone every other day) were used. (b) Changes in ROCK-II expression from Day 0 to Day 40. (c) Relative amounts of ROCK-II expression are given in comparison with -actin expressions on the reprobing membrane. Bars indicate group A (closed bar), group D (hatched bar), and group F (open bar). All data are expressed as mean ± SEM (n = 4). Statistical analysis was performed using the one-way ANOVA test followed by Fisher's least significance test (* p < 0.05).

MLC20 Phosphorylation Levels in Mouse Lungs

The level of MLC20 phosphorylation was elevated by BLM treatment from 35.3 ± 7.6% (saline control) to 56.1 ± 5.4% (mean ± SEM, n = 4). Y-27632 decreased MLC20 phosphorylation levels by 37.2 ± 3.8% in BLM-induced lungs (Figures 8a and 8b). Calyculin A in a Ca²⁺ (30 µM) containing buffer increased by 60.4 ± 5.1%. In contrast, following depletion of Ca²⁺ by BAPTA-AM in a Ca²⁺-free solution, the level of MLC20 phosphorylation decreased to 17.4 ± 4.1% (mean ± SEM, n = 3) (Figure 8c).

View larger version (32K): [in this window] [in a new window]   Figure 8.   MLC20 phosphorylation assay in lungs on Day 21. (a) Nonphosphorylated (MLC) and phosphorylated (MLC-P) myosin light chain were resolved using glycerol-PAGE and detected with a monoclonal antibody raised against the MLC20. Group A (BLM and saline every other day), group D (BLM and Y-27632 100 µg/kg every other day), and group F (saline alone every other day) were used. (b) The relative amounts of MLC phosphorylation are given in MLC-P/MLC + MLC-P in groups A, D, and F. (c) Maximum and minimum levels of MLC20 phosphorylation in mice lungs. Lungs were treated with calyculin A in a Ca2+ (30 µM)-containing buffer or BAPTA-AM in Ca2+-free buffer. The relative amounts of MLC20 phosphorylation are expressed as MLC-P/ MLC + MLC-P. All data are expressed as mean ± SEM (n = 4). Statistical analysis was performed using the one-way ANOVA test followed by Fisher's least significance test (* p < 0.05: versus the group A).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The main finding of this study was that treatment with Y-27632 significantly inhibited BLM-induced lung fibrosis in mice. Y-27632 improved histopathological features, decreased OH-proline content in the lungs, and reduced inflammatory cell numbers in the BALF (Figures 2, 3, and 4).

BLM-induced pulmonary fibrosis in mice develops after acute lung inflammation with infiltration of inflammatory cells. In humans, neutrophilia in the lungs worsens the prognosis of IPF and blunts the response to corticosteroids (2). In the present study, Y-27632 significantly decreased the number of neutrophils and macrophages in the BALF of BLM-treated mice. These findings were supported by the chemotaxis assays in vitro. Migration of MH-S cells, an alveolar macrophage cell line, induced by LPS, was completely inhibited by Y-27632 (Figure 5a). IL-1 is a pivotal inflammatory cytokine secreted from both monocytes and lymphocytes. Binding of IL-1 to IL-1 receptor (IL-1R) activates RhoA, which in turn evokes Rho-dependent cytoskeletal reorganization in vitro (24). IL-1 secretion in BALF obtained from patients with IPF is elevated (1). This is supported by the findings in BLM-induced IPF models of C57BL/6 mice (3) and from isolated neutrophils from C57BL/6 mice that possess the IL-1 receptor (17).

In the present study, Y-27632 inhibited IL-1-induced neutrophil migration in a dose-dependent manner (Figure 5b). Fibroblasts also migrated into the BLM-induced lungs and proliferated in alveolar walls. Activated fibroblasts are known to secrete collagen and other matrix proteins (1). As a result, respiratory space is occupied by the products of fibroblasts, leading to alveolar-arterial oxygen differences (2). Because migration of fibroblasts plays an important role in IPF, and clinically PDGF-BB levels in BALF are elevated in human IPF (1), we investigated the extent of MLF migration in response to recombinant PDGF-BB as a chemoattractant. We confirmed by Western blot analysis that PDGF-BB bound to the PDGF receptor of MLF with phosphorylation. MLF appeared as typical fibroblastic spindle shape and expressed vimentin and -smooth muscle actin but not cytokeratin by Western blot analysis. MLF revealed characteristic of myofibroblast (data not shown). Y-27632 inhibited PDGF-BB-induced MLF migration with IC₅₀ of 1.4 µM in a dose-dependent manner (Figure 5c). The IC₅₀ values for Y-27632 to inhibit ROCK activity are usually seen at concentrations of 0.1-10 µM. At these doses of Y-27632, the actin cytoskelton is disrupted without influencing cell growth (25, 26). Niggli also showed that Y-27632 inhibits MLC phospholylation and chemotactic peptide-induced cell polarity and locomotion with similar potency (ED ₅₀ of 0.5-1.1 µM) (11). We verified that all cell migrations in this study were inhibited by Y-27632 at similar concentrations. Therefore, it is likely that the inhibitory effect of Y-27632 on cell migration in this study was due to the inhibition of ROCK activation in the cells in vitro. To determine whether the concentrations of Y-27632 were sufficient in vivo, we measured serum concentrations of Y-27632 (100 µg/kg, intraperitoneally), and found them to be above 1 µM at 5 min after injection into the mice. In addition, lungs removed from C57BL/6 mice and all cells used in the chemotaxis assays expressed RhoA and ROCK-II proteins (Figures 5d and 5e). According to histopathological examinations, BLM plus Y-27632-treated mice showed a marked reduction in the number of infiltrated cells. Thus, one possible mechanism underlying the Y-27632 effects on BLM-induced lung fibrosis appears to be a blockade of cell infiltration through inhibition of Rho/ROCK-mediated pathway of the cells in vivo.

It has been reported that mRNAs of RhoA and ROCK were up-regulated in the rat myometrium during pregnancy (27). Furthermore, RhoA was overexpressed in tumors in human lung (28). Therefore, we investigated the changes in RhoA and ROCK expressions in the lungs of our IPF model. The expression of RhoA protein did not change from Day 0 to Day 40 in all groups (Figure 7a). Although the expression of ROCK-II mRNA markedly increased from Day 7 to Day 21, which corresponded to the acute inflammatory stage (Figure 6), the expression of ROCK-II protein only showed a tendency to increase (Figures 7b and 7c). The BLM plus Y-27632-treated group presented similar results of ROCK-II expression regarding mRNA and protein levels. We showed that MLC20 phosphorylation of total lung homogenates was increased in BLM-induced lung fibrosis, which were inhibited by Y-27632 (Figures 8a and 8b). These findings indicate that the Rho/ROCK-mediated pathway may contribute to the development of pulmonary fibrosis, and further that Y-27632 inhibits ROCK function at protein levels, resulting in inhibition of muscle and nonmuscle MLC20 phosphorylation. Moreover, changes in the expression of Rho and ROCK proteins were not responsible for the development of pulmonary fibrosis in this animal model.

There have been many proposed approaches to prevent pulmonary fibrosis. However, the first line of treatment of IPF has not yet been established. Although corticosteroids and cyclophosphamide are widely used as standard protocols of current therapy in human IPF, their effects are not satisfactory. Indeed, high-dose corticosteroids did not influence the appearance of inflammatory cells in BAL by a single tracheal injection of BLM, or the increased lung water content in rat (29). Furthermore, corticosteroids and cyclophosphamide are not only insufficient, but also sometimes even worsen acute lung damage, and increase collagen synthesis in animal models (30, 31). In the present study, inhibition of ROCK by Y-27632 suppressed not only acute inflammation but also fibrotic changes. Our findings indicate that inhibition of the Rho/ ROCK-mediated pathway may become a novel treatment approach for IPF. In addition, because the pathway exists ubiquitously, this approach might be applicable to other types of organ fibrosis.

Finally, although we showed marked inhibition of lung fibrosis by Y-27632, the effect was not complete. We cannot exclude the possibility that continuous delivery of Y-27632 using an osmotic pump may provide better results. However, it is more likely that a Rho/ROCK-independent mechanism might also contribute to IPF, because continuous delivery of Y-27632 using an osmotic pump did not inhibit completely the dissemination of tumor cells in rats (16). Further studies are required to clarify whether the Rho/ROCK-independent pathway is involved in IPF.