Fibrosis Attenuates Bleomycin-Induced Pulmonary Macrophage ...

Neutralization of the CXC Chemokine, Macrophage Inflammatory Protein-2, Attenuates Bleomycin-Induced Pulmonary Fibrosis1 Michael P. Keane,* John A. Belperio,* Thomas A. Moore,* Bethany B. Moore,* Douglas A. Arenberg,* Robert E. Smith,† Marie D. Burdick,* Steven L. Kunkel,† and Robert M. Strieter2* Few studies have addressed the importance of vascular remodeling in the lung during the development of bleomycin-induced pulmonary fibrosis. For fibroplasia and deposition of extracellular matrix to occur, there must be a geometric increase in neovascularization. We hypothesized that net angiogenesis during the pathogenesis of fibroplasia and deposition of extracellular matrix during bleomycin-induced pulmonary fibrosis are dependent in part upon an overexpression of the angiogenic CXC chemokine, macrophage inflammatory protein-2 (MIP-2). To test this hypothesis, we measured MIP-2 by specific ELISA in whole lung homogenates in either bleomycin-treated or control CBA/J mice and correlated these levels with lung hydroxyproline. We found that lung tissue from mice treated with bleomycin, compared with that from saline-treated controls, demonstrated a significant increase in the presence of MIP-2 that was correlated to a greater angiogenic response and total lung hydroxyproline content. Neutralizing anti-MIP-2 Abs inhibited the angiogenic activity of day 16 bleomycin-treated lung specimens using an in vivo angiogenesis bioassay. Furthermore, when MIP-2 was depleted in vivo by passive immunization, bleomycin-induced pulmonary fibrosis was significantly reduced without a change in the presence of pulmonary neutrophils, fibroblast proliferation, or collagen gene expression. This was also paralleled by a reduction in angiogenesis. These results demonstrate that the angiogenic CXC chemokine, MIP-2, is an important factor that regulates angiogenesis/fibrosis in pulmonary fibrosis. The Journal of Immunology, 1999, 162: 5511–5518.

I

diopathic pulmonary fibrosis (IPF)3 is a chronic and often fatal pulmonary disorder. A recent study has shown a prevalence rate of 27–29 cases/100,000 and may even be as high as 250 cases/100,000 in individuals .75 yr of age (1). In certain parts of the world (United Kingdom, New Zealand, Germany) the incidence of IPF appears to be on the rise (2). Conventional treatment with immunosuppressive therapy in IPF has been disappointing, and 5-yr survival has been predicted to be 50%. The elucidation of mediators that orchestrate this aberrant tissue repair will allow the development of novel interventions to treat this disorder. The pathology of IPF demonstrates features of dysregulated and abnormal repair with exaggerated angiogenesis, fibroproliferation, and deposition of extracellular matrix, leading to progressive fibrosis and loss of lung function. The contribution of neovascularization to the progression of fibrosis in IPF has been largely ig-

*Department of Internal Medicine, Division of Pulmonary and Critical Care Medicine, and †Department of Pathology, University of Michigan Medical School, Ann Arbor, MI 48109 Received for publication September 10, 1998. Accepted for publication February 11, 1999. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 This work was supported in part by National Institutes of Health Grants HL03906 (to M.P.K.), P50HL56402 and P50HL60289 (to S.L.K.), and P50HL56402, CA66180, and P50HL60289 (to R.M.S.). 2 Address correspondence and reprint requests to Dr. Robert M. Strieter, Department of Internal Medicine, Division of Pulmonary and Critical Care Medicine, 6200 MSRB III, Box 0642, 1150 W. Medical Center Dr., University of Michigan Medical Center, Ann Arbor, MI 48109-0642. 3 Abbreviations used in this paper: IPF, idiopathic pulmonary fibrosis; ELR, GluLeu-Arg; MIP-2, macrophage inflammatory protein-2; IP-10, IFN-g-inducible protein; GRO, growth-related oncogene; MPO, myeloperoxidase.

Copyright © 1999 by The American Association of Immunologists

nored. The existence of neovascularization in IPF was originally identified by Turner-Warwick (3), who examined the lungs of patients with widespread IPF and demonstrated neovascularization leading to anastomoses between the systemic and pulmonary microvasculature. Further evidence of neovascularization during the pathogenesis of pulmonary fibrosis has been seen in a rat model of bleomycin-induced pulmonary fibrosis. Peao et al. perfused the vascular tree of rat lungs with methacrylate resin at a time of maximal pulmonary fibrosis after bleomycin treatment (4). Using scanning electron microscopy, these investigators demonstrated major vascular modifications that included neovascularization of an elaborate network of microvasculature located in the peribronchial regions of the lungs and distortion of the architecture of the alveolar capillaries. The location of neovascularization was closely associated with regions of pulmonary fibrosis, similar to the findings for human lungs. We have shown that members of the CXC chemokine family exert disparate effects in mediating angiogenesis as a function of the presence or the absence of three amino acid residues (Glu-LeuArg; the ELR motif) that immediately precedes the first cysteine amino acid of the primary structure of these cytokines (5, 6). While IL-8, an ELR CXC chemokine, was initially discovered on the basis of chemotaxis and activation of neutrophils, our laboratory and others have found that IL-8 in vitro has endothelial cell chemotactic activity and in vivo induces neovascularization in the cornea of rats and rabbits, without inflammation (6 – 8). Macrophage inflammatory protein-2 (MIP-2) is a murine functional homologue of IL-8 and a structural homologue of human GRO-b/g, both ELR CXC chemokines. Furthermore, similar to IL-8, it is chemotactic for endothelial cells and induces neovascularization in 0022-1767/99/$02.00

5512

NEUTRALIZATION OF MIP-2 INHIBITS PULMONARY FIBROSIS

the cornea of rats without associated inflammation (our unpublished observations). We have recently shown that the CXC chemokines, IL-8 and IP-10, regulate angiogenic activity in IPF (9). To demonstrate proof of this principle, we extended these studies to a murine model of bleomycin-induced pulmonary fibrosis. We hypothesized that net angiogenesis during the pathogenesis of fibroplasia and deposition of extracellular matrix during bleomycin-induced pulmonary fibrosis is dependent in part upon an overexpression of the angiogenic CXC chemokine, MIP-2. We found that lung tissue from mice treated with bleomycin, compared with that from saline-treated controls, demonstrated a significant increase in the presence of the CXC chemokine, MIP-2, that was correlated to a greater angiogenic response. In addition, this angiogenic activity was significantly attenuated in the presence of neutralizing Abs to MIP-2. Furthermore, pulmonary fibrosis was significantly reduced when bleomycin-exposed animals were passively immunized with anti-MIP-2 Abs. The reduction of pulmonary fibrosis was also associated with a reduction in angiogenesis. These results demonstrate that the angiogenic CXC chemokine, MIP-2, is an important factor that regulates angiogenesis in bleomycin-induced pulmonary fibrosis.

Materials and Methods Reagents Polyclonal anti-murine MIP-2-specific antiserum was produced by the immunization of a rabbit with murine MIP-2 (R&D Systems, Minneapolis, MN) in multiple intradermal sites with CFA (10, 11). The specificity of the Ab was assessed by either Western blot analysis or ELISA against a panel of other recombinant cytokines. This Ab was found to neutralize 30 ng of MIP-2 at a dilution of 1/1000. Neutralizing capacity was assessed using two strategies: 1) blocking i.p. MIP-2-induced neutrophil chemotaxis in mice, and 2) inhibiting MIP-2-mediated angiogenesis using the rat corneal micropocket assay of neovascularization. Abs were specific in our sandwich ELISA without cross-reactivity to a panel of cytokines, including IL-1R antagonist protein, IL-1, IL-2, IL-4, IL-6, TNF-a, IFN-g, and other members of the CXC and CC chemokine families (10, 11). The antiprotease buffer for tissue homogenization consisted of 13 PBS with 2 mM PMSF and 1 mg/ml each of antipain, aprotinin, leupeptin, and pepstatin A. Rabbit anti-factor VIII-related Ag Abs were purchased from Biomeda (Foster City, CA).

Animal model of pulmonary fibrosis Female CBA/J mice (6 – 8 wk old; The Jackson Laboratory, Bar Harbor, ME) were used in all experiments. Mice were maintained in specific pathogen-free conditions and were provided with food and water ad libitum. To induce pulmonary fibrosis, mice were treated with intratracheal bleomycin (Blenoxane, a gift from Bristol Myers, Evansville, IN; 0.15 U/kg) on day 0 as previously described (12). Control animals received only sterile saline as previously described (12). Briefly, CBA/J mice were anesthetized with 250 ml of 12.5 mg/ml ketamine injected i.p. followed by intratracheal instillation of 0.025 U of bleomycin in 25 ml of sterile isotonic saline. At 2, 4, 8, 12, 16, and 20 days postinstillation, animals were euthanized, and both lungs were removed for homogenization as described below. In separate experiments bleomycin-treated mice were passively immunized by i.p. injection of 0.5 ml of anti-MIP-2 serum or preimmune serum on days 8, 10, 12, and 14, before maximal fibrosis.

Lung tissue preparation Bleomycin- or saline (control)-treated lungs were homogenized and sonicated in anti-protease buffer using a method previously described (11, 13). Specimens were centrifuged at 900 3 g for 15 min, filtered through 1.2-mm pore size Sterile Acrodiscs (Gelman Sciences, Ann Arbor, MI), and frozen at 270°C until thawed for assay by specific MIP-2 ELISA. A portion of the specimen was concentrated approximately 12-fold by lyophilization (Speed-Vac, Savant, Farmingdale, NY) and was used in the corneal micropocket model of neovascularization for analysis of angiogenic activity. Additionally, some lungs were fixed in 4% paraformaldehyde and embedded in paraffin for histologic and immunohistochemical analyses.

MIP-2 ELISA Antigenic MIP-2 was quantitated using an ELISA as previously described (10, 11). Briefly, flat-bottom 96-well microtiter plates (Nunc, Copenhagen, Denmark) were coated with 50 ml/well of the appropriate polyclonal Ab (1 ng/ml in 0.6 M NaCl, 0.26 M H3BO4, and 0.08 N NaOH, pH 9.6) for 24 h at 4°C and then washed with PBS and 0.05% Tween 20 (wash buffer). Nonspecific binding sites were blocked with 2% BSA. Plates were rinsed, and samples were added (50 ml/well), followed by incubation for 1 h at 37°C. Plates were then washed, and 50 ml/well of the appropriate biotinylated polyclonal Ab (3.5 ng/ml in wash buffer and 2% FCS) was added for 45 min at 37°C. Plates were washed three times, streptavidin-peroxidase conjugate (Bio-Rad, Richmond, CA) was added, and the plates were incubated for 30 min at 37°C. Chromogen substrate (Dako, Carpinteria, CA) was then added, and the plates were incubated at room temperature to the desired extinction. Plates were read at 490 nm in an automated microplate reader (Bio-Tek Instruments, Winooski, VT). Standards were 1⁄2-log dilutions of recombinant MIP-2 from 100 ng to 1 pg/ml (50 ml/well).

Immunohistochemistry of MIP-2 Paraffin-embedded tissue from control and bleomycin-treated lung was processed for immunohistochemical localization of MIP-2, using a method previously described (10). Briefly, tissue sections were dewaxed with xylene and rehydrated through graded concentrations of ethanol. Tissue-nonspecific binding sites were blocked using normal goat serum (BioGenex, San Ramon, CA). Tissue sections were overlaid with a 1/500 dilution of either control (rabbit) or polyclonal rabbit anti-MIP-2 Abs. The tissue sections were washed with Tris-buffered saline, incubated for 60 min with secondary goat anti-rabbit biotinylated Abs (BioGenex), washed in Trisbuffered saline, and incubated with alkaline phosphatase conjugated to streptavidin (BioGenex). Fast Red (BioGenex) reagent was used for chromogenic localization of MIP-2. After optimal color development, tissue sections were immersed in sterile water, counterstained with Lerner’s hematoxylin, and coverslipped using an aqueous mounting solution.

Immunolocalization of factor VIII-related Ag Paraffin-embedded tissue from control and bleomycin-treated lung was processed for immunohistochemical localization of factor VIII-related Ag as previously described (10). Briefly, tissue sections were dewaxed with xylene and rehydrated through graded concentrations of ethanol. Slides were blocked with normal goat serum (BioGenex) and overlaid with a 1/500 dilution of either control (rabbit) or polyclonal rabbit anti-factor VIII-related Ag Abs. Slides were then rinsed and overlaid with secondary biotinylated goat anti-rabbit IgG (1/35) and incubated for 60 min. After washing with Tris-buffered saline, slides were overlaid with a 1/35 dilution of alkaline phosphatase conjugated to streptavidin (BioGenex) and incubated for 60 min. Fast Red (BioGenex) reagent was used for chromogenic localization of factor VIII-related Ag. After optimal color development, sections were immersed in sterile water, counterstained with Lerner’s hematoxylin, and coverslipped using an aqueous mounting solution.

Western blot analysis of factor VIII-related Ag Total protein extracts were made by homogenizing lungs in lysis buffer (20 mM Tris-HCl (pH 8), 150 mM NaCl, 1% Nonidet P-40, and 2.5 mM EDTA) supplemented with 2 ng/ml aprotinin and 35 ng/ml PMSF. Cell extracts were incubated on ice for 30 min, followed by centrifugation at 4°C for 30 min. Supernatants were then removed and assayed for total protein content using BCA protein assay reagents (Pierce, Rockford, IL) and comparison to known amounts of BSA. One microgram of total protein was loaded in each well of a 12% polyacrylamide gel, and extracts were subjected to SDS-PAGE. The separated proteins were transferred to a polyvinylidene fluoride membrane (Pierce) by electrophoretic transfer overnight in Tris-glycine buffer (20 mM Tris and 150 mM glycine (pH 8.0) with methanol added to a final concentration of 20% (v/v)). Blots were blocked in 5% skim milk in TBST buffer (10 mM Tris-HCl (pH 8.0), 150 mM NaCl, and 0.05% Tween-20) for 2 h at room temperature, followed by incubation in rabbit primary Ab serum against factor VIII-related Ag (Sigma, St. Louis, MO) diluted 1/200 in blocking solution for 2 h at room temperature. Blots were washed for three 10-min washes in TBST and were incubated for 1 h at room temperature in goat anti-rabbit horseradish peroxidase-conjugated secondary Ab (Bio-Rad, Hercules, CA) at a 1/20,000 dilution. Blots were again washed four times, 10 min each time, in TBST, and proteins were visualized following incubation of the blots in SuperSignal chemiluminescent substrate solution according to the manufacturer’s protocol (Pierce) and exposure to XAR-5 film (Eastman Kodak, Rochester, NY).

The Journal of Immunology FACS analysis of factor VIII-related Ag Single lung cell suspension preparations were made using a method previously described (14). Briefly, lungs were harvested on day 16 from bleomycin-treated animals who had been treated with either normal rabbit serum or anti-MIP-2 Abs. Lungs were minced with scissors to a fine slurry in 15 ml of digestion buffer (RPMI, 5% FCS, 1 mg/ml collagenase (Boehringer Mannheim, Indianapolis, IN), and 30 mg/ml DNase (Sigma)). Lung slurry was enzymatically digested for 45 min at 37°C. Any undigested fragments were further dispersed by drawing the solution up and down through the bore of a 10-ml syringe. The total lung cell suspension was pelleted, resuspended, and spun through a 20% Percoll gradient. Cell counts and viability were determined using trypan blue exclusion on a hemocytometer. Single cell suspensions were stained with anti-CD45 Tricolor (Caltag, South San Francisco, CA) and primary rabbit anti-factor VIII-related Ag Abs (Sigma) followed by FITC-labeled goat anti-rabbit Abs (PharMingen) to allow live gating on CD45-negative cells, thus selecting out the nonleukocyte population for further analysis of expression of factor VIII-related Ag. Cells were analyzed on a FACScan flow cytometer (Becton Dickinson, San Jose, CA) using CellQuest software (Becton Dickinson).

Hydroxyproline assay Total lung collagen was determined by analysis of hydroxyproline as previously described (12). Briefly, lungs were harvested on days 2, 4, 8, 12, 16, and 20 postbleomycin administration and homogenized in 2 ml of PBS, pH 7.4, with a Tissue Tearor. One-half milliliter of each sample (both lungs) was then digested in 1 ml of 6 N HCl for 8 h at 120°C. Five microliters of citrate/acetate buffer (5% citric acid, 7.24% sodium acetate, 3.4% sodium hydroxide, and 1.2% glacial acetic acid, pH 6.0) and 100 ml of chloramine-T solution (282 mg chloramine-T, 2 ml of n-propanol, 2 ml of H2O, and 16 ml of citrate/acetate buffer) were added to 5 ml of sample, and the samples were left at room temperature for 20 min. Next, 100 ml of Ehrlich’s solution (2.5 g of 4-(dimethylamino)benzaldehyde; Aldrich, Milwaukee, WI), 9.3 ml of n-propanol, and 3.9 ml of 70% perchloric acid (Eastman Kodak)) were added to each sample, and the samples were incubated for 15 min at 65°C. Samples were cooled for 10 min and read at 550 nm on a Beckman DU 640 spectrophotometer (Fullerton, CA). Hydroxyproline (Sigma) concentrations from 0 to 100 mg/ml were used to construct a standard curve.

Myeloperoxidase assay Pulmonary neutrophil sequestration was quantitated using a myeloperoxidase (MPO) assay as previously described (15). Briefly, at the time of sacrifice lungs were perfused free of blood with 10 ml of 0.9% saline via the spontaneously beating right ventricle. The lungs were excised and placed in a 50 mM potassium phosphate buffer solution (pH 6.0) with 5% hexadecyltrimethyl ammonium bromide (Sigma). The lung tissue was homogenized, sonicated, and centrifuged at 12,000 3 g for 15 min at 4°C. The supernatant was then assayed for MPO activity using a spectrophotometric reaction with o-dianisidine hydrochloride (Sigma) at 460 nm.

Fibroblast proliferation Fibroblasts were cultured as previously described (9). Briefly, fibroblasts were isolated from CBA/J mouse lungs. Pulmonary fibroblasts were grown to 80% confluence and passaged. At the time of the fourth passage, pulmonary fibroblast purity was .99% as determined by the absence of nonspecific esterase, factor VIII-related Ag, or cytokeratin immunostaining. The cells were .90% positive for vimentin, laminin, and fibronectin and were .90% negative for a-smooth muscle actin and desmin. This technique allowed the establishment of pulmonary fibroblast cell lines. On the day of use, pulmonary fibroblasts were plated out in 96-well plates at a concentration of 5000 cells/well and were allowed to adhere overnight. Fibroblasts were then washed free of serum and cultured for 24 h under serum-free conditions. At 24 h serum was again added along with varying concentrations of MIP-2 or platelet-derived growth factor, and fibroblasts were cultured for a further 24, 48, and 72 h. [3H]thymidine was added 12 h before harvesting using a cell harvester (Brandel, Gaithersburg, MD). The percent incorporation of [3H]thymidine was assessed using a Beckman LS 1801 scintillation counter (Beckman, Schaumburg, IL).

RT-PCR for collagen gene expression Fibroblasts were isolated as described above. At passage 4, fibroblasts were plated into 60-mm dishes. When they reached 80% confluence they were stimulated in the presence of 1-30 ng/ml of MIP-2 or 20 ng/ml of TGF-b. Total RNA was extracted from fibroblasts using Trizol reagent (Life Technologies, Grand Island, NY) according to the manufacturer’s instructions.

5513 RT-PCR was performed using an Access RT-PCR kit (Promega, Madison, WI). b-Actin was used as a housekeeping gene. For b-actin the sense primer used was 59-GTGGGGCGCCCCAGGCACCA; the antisense primer was 59-GCTCGGCCGTGGTGGTGAAGC. For type I collagen the sense primer used was 59-TGGTGCCAAGGGTCTCACTGGC; the antisense primer was 59-GGACCTTGTACACCACGTTCACC. For type III collagen the sense primer was 59-GCAGTCCAACGTAGATGAATTGG; the antisense primer was 59-GAAGGCCTGGTGGACCAGCTGG. PCR products were visualized by agarose gel electrophoresis.

Corneal micropocket assay of angiogenesis Angiogenic activity of lung homogenates was assayed in vivo in the avascular cornea of hooded Long-Evans rat eyes, as previously described (5, 9 –11, 13). Briefly, equal volumes of lyophilized lung tissue specimens normalized to total protein were combined with sterile Hydron (Interferon Sciences, New Brunswick, NJ) casting solution. Five-microliter aliquots were pipetted onto the flat surface of an inverted sterile polypropylene specimen container and polymerized overnight in a laminar flow hood under UV light. Before implantation, pellets were rehydrated with normal saline. Animals were anesthetized with ketamine (150 mg/kg) and atropine (250 mg/kg) i.p. Rat corneas were anesthetized with 0.5% proparacaine hydrochloride ophthalmic solution followed by implantation of the Hydron pellet into an intracorneal pocket (1–2 mm from the limbus). Six days after implantation, animals received 1000 U of heparin and ketamine (150 mg/ kg) i.p., followed by a 10-ml perfusion of colloidal carbon via the left ventricle. Corneas were harvested and photographed. No inflammatory response was observed in any of the corneas treated with the above specimens. Positive neovascularization responses were recorded only if sustained directional ingrowth of capillary sprouts and hairpin loops toward the implant was observed. Negative responses were recorded when either no growth was observed or only an occasional sprout or hairpin loop displaying no evidence of sustained growth was detected. All animals were handled in accordance with the University of Michigan unit for laboratory animal medicine.

Statistical analysis Data were analyzed on a Macintosh IIci computer using the StatView 4.5 statistical package (Abacus Concepts, Berkeley, CA). Comparisons were made using the unpaired t test. Data were considered statistically significant at p # 0.05.

Results Lung tissue from mice treated with bleomycin express greater levels of MIP-2 We obtained lung tissue from bleomycin-treated mice (n 5 6) at each time point or from saline-treated controls (n 5 6) at each time point and measured MIP-2 by specific ELISA standardized per lung. Lung tissue from bleomycin-treated animals demonstrated greater levels of MIP-2 compared with that from saline-treated controls on days 16 and 20 ( p , 0.05; Fig. 1). Furthermore, these greater levels of MIP-2 continued in parallel with the time of maximal pulmonary fibrosis (days 16 and 20) as determined by total lung hydroxyproline (Fig. 2). These results suggest a temporal relationship between elevated levels of MIP-2 and the development of pulmonary fibrosis. Lung tissue from mice treated with bleomycin induces greater angiogenic activity To substantiate that this CXC chemokine may be modulating lung tissue-derived angiogenic activity, we next assessed the in vivo angiogenic activity of six random pooled samples of either bleomycin-treated lung tissue or saline control lung tissue at 16 days in the presence or the absence of preimmune (control) or neutralizing MIP-2 Abs, using the rat corneal micropocket model of neovascularization (Fig. 3). These Abs did not contain significant quantities of LPS contamination as assessed by Limulus assay, and all samples were normalized to total protein. We found that lung tissue from bleomycin-treated mice (Fig. 3B) induced a greater angiogenic response than lung tissue from saline-treated controls (Fig. 3A; n 5 6 for each manipulation). Neutralizing Abs to MIP-2

5514


FIGURE 1. Time course of MIP-2 levels from lung tissue of bleomycinor saline-treated mice (n 5 6 lungs at each time point).

significantly attenuated the angiogenic activity of bleomycintreated lung tissue (Fig. 3C) compared with control Abs (Fig. 3B). These findings suggested that MIP-2 is a significant angiogenic factor in bleomycin-induced pulmonary fibrosis. Immunolocalization of MIP-2 is predominantly associated with areas of pulmonary fibrosis Since MIP-2 was significantly elevated in bleomycin-treated lung tissue and was temporally related to the development of pulmonary fibrosis, we next assessed whether MIP-2 was associated with areas of pulmonary fibrosis. Using immunohistochemistry, we found that MIP-2 expression was localized to areas of pulmonary fibrosis (Fig. 4). Interestingly, the areas associated with MIP-2 localization were essentially devoid of infiltrating neutrophils and demonstrated significant vascular remodeling as evidenced by the immunolocalization of factor VIII-related Ag (Fig. 5). Taken together these results are strong evidence for a significant role for MIP-2 in the development of the pulmonary fibrotic response to bleomycin. Passive immunization with neutralizing MIP-2 Abs reduces bleomycin-induced pulmonary fibrosis Based on our findings of increased angiogenic activity from bleomycin-treated lung tissue that was significantly attributable to MIP-2, which itself correlated to the development of fibrosis, we next assessed whether neutralization of MIP-2 in the in vivo model

FIGURE 3. Representative corneal micropocket assay for neovascularization of lung-derived angiogenic activity from mice treated intratracheally with saline (A), bleomycin normalized to total protein in the presence of normal rabbit serum (B), or bleomycin normalized to total protein in the presence of neutralizing anti-MIP-2 Abs (C). The data shown are representative of six corneas in each group.

of bleomycin-induced pulmonary fibrosis would attenuate the fibrotic response. Passive immunization with neutralizing MIP-2 Abs on days 8, 10, 12, and 14 led to reduced total lung hydroxyproline compared with control values (Fig. 6). Control serum-treated mice had similar hydroxyproline levels as mice treated with bleomycin alone (data not shown). We chose these time points as they were the times of maximum MIP-2 expression in our model. In addition, these time points proceed the time of maximum pulmonary fibrosis. Passive immunization with neutralizing Abs reduces in vivo angiogenesis

FIGURE 2. Time course of total lung hydroxyproline from lung tissue of bleomycin- or saline-treated controls (n 5 6 lungs at each time point).

Having shown that the neutralization of MIP-2 led to a reduction in bleomycin-induced fibrosis we next assessed the effect of passive immunization and neutralization of MIP-2 on in vivo angiogenesis. We demonstrated that in vivo neutralization of MIP-2 led to a reduction in lung-derived angiogenic activity as assessed by the corneal micropocket assay (Fig. 7). Furthermore, neutralization of MIP-2 led to a reduction in the total number of endothelial cells in the lung as assessed by 1) expression of factor VIII-related Ag using FACS analysis (Table I) and 2) Western blot analysis of

The Journal of Immunology

FIGURE 4. Representative photomicrograph of the immunolocalization of MIP-2 in bleomycin-treated lung tissue. A, Bleomycin-treated lung tissue immunostained with control Abs demonstrating the lack of nonspecific staining (3312). B, C, and D, Same lung specimen immunostained for MIP-2; B represents a relatively spared area showing less intense staining (3312).

expression of factor VIII-related Ag compared with that in control treated mice (Fig. 8). Neutralization of MIP-2 is not associated with reduced neutrophil presence Having shown that MIP-2 has an important role in the development of the fibrotic response to bleomycin our next step was to substantiate that neutralization of MIP-2 was attenuating fibrosis through an antiangiogenic mechanism by assessing its effect on total lung neutrophil infiltration. As shown in Fig. 9, neutralization of MIP-2 did not lead to a reduction in neutrophil presence in the lungs of treated animals as assessed by MPO activity. These results support the contention that MIP-2 has an important role in the development and maintenance of pulmonary fibrosis that is independent of neutrophil chemotactic or activating properties.

FIGURE 5. Representative photomicrograph of the immunolocalization of factor VIII-related Ag in bleomycin-treated lung tissue. A, Bleomycintreated lung tissue immunostained with control Abs demonstrating the lack of nonspecific staining (3400). B, C, and D, Same lung specimen immunostained for factor VIII-related Ag (3400).

5515

FIGURE 6. Total lung hydroxyproline on day 16 from saline-treated control mice or bleomycin-treated mice passively immunized with either anti-MIP-2 or control Abs (n 5 6 lungs in each group).

MIP-2 does not stimulate fibroblast proliferation or collagen gene expression Having shown that MIP-2 reduces fibrosis through a neutrophilindependent mechanism we were next interested to determine whether MIP-2 had any direct action on pulmonary fibroblasts. Pulmonary fibroblasts were isolated from CBA/J mouse lungs and stimulated with various concentrations of MIP-2, whereas plateletderived growth factor was used as a positive control. Proliferation

FIGURE 7. Representative corneal micropocket assay for neovascularization of lung-derived angiogenic activity from mice who received intratracheal saline alone (A), bleomycin and control serum (B), or bleomycin and neutralizing anti-MIP-2 Abs (C; n 5 6 lungs in each group).

5516


Table I. Factor VIII-related Ag-expressing cells, as assessed by FACS analysis (p , 0.05) Condition

Factor VIII-Positive Cells

Bleomycin 1 anti-MIP-2 Bleomycin 1 NRSa

1.7 6 0.48 3 106 2.58 6 0.31 3 106

a

NRS, normal rabbit serum.

was measured using incorporation of tritiated thymidine. As shown in Fig. 10, MIP-2 had no effect on fibroblast proliferation. Furthermore, when isolated pulmonary fibroblasts were stimulated with either MIP-2 or TGF-b as a positive control, there was no increase in either collagen I or collagen III gene expression (Fig. 11). Taken together these findings are further evidence for an alternative biological role for MIP-2 in the pathogenesis of bleomycin-induced pulmonary fibrosis other than through an effect on fibroblast proliferation, collagen gene expression, or neutrophil recruitment or activation.

Discussion Our findings support the idea that a proangiogenic environment exists during the pathogenesis of bleomycin-induced pulmonary fibrosis, and that MIP-2 is an important factor in the regulation of this angiogenic environment. Studies directed at understanding the pathogenesis of IPF have primarily focused on mechanisms related to fibroplasia and deposition of extracellular matrix. However, these investigations have often been based on the evaluation of IPF in a “snap-shot” manner, not in a temporal fashion, and usually at end-stage fibrosis when the cellular phase of IPF has been replaced by extracellular matrix and nonviable scar. While a number of eloquent studies have delineated mechanisms of fibroplasia and deposition of extracellular matrix in IPF, few have addressed the importance of angiogenesis in the lung during injury and subsequent fibrosis. While angiogenesis has been shown to play a role in the evolution of tissue repair and fibroplasia associated with acute lung injury and sarcoidosis (16, 17), the contribution of neovascularization to the pathogenesis of fibrosis in IPF has been largely ignored. The existence of morphological neovascularization in IPF was originally identified by Turner-Warwick (3), who examined the lungs of patients with widespread IPF and demonstrated neovascularization/vascular remodeling that was often associated with anastomoses between the systemic and pulmonary microvasculature. Further evidence of neovascularization during the pathogenesis of pulmonary fibrosis has been seen in a rat model of bleomycininduced pulmonary fibrosis (4). Peao and associates perfused the vascular tree of rat lungs with methacrylate resin at a time of maximal pulmonary fibrosis (4). Using scanning electron microscopy, these investigators demonstrated major vascular modifications that included neovascularization of an elaborate network of microvasculature located in the peribronchial regions of the lungs and dis-

FIGURE 8. Western blot analysis of factor VIII-related Ag from bleomycin-treated mice who received either neutralizing anti-MIP-2 Abs or control serum (n 5 2 in each group). The first lane represents a normal lung from an untreated mouse. The data shown are representative of two separate experiments.

FIGURE 9. Lung MPO activity on day 16 from the lungs of mice passively immunized with either anti-MIP-2 Abs or control Abs (n 5 6 lungs in each group). NS Þ, not statistically significant.

tortion of the architecture of the alveolar capillaries. The location of neovascularization was closely associated with regions of pulmonary fibrosis, similar to the findings for human lungs (3), and this neovascularization appeared to lead to the formation of systemic-pulmonary anastomoses (4). We have recently shown that the CXC chemokines, IL-8 and IP-10, are important factors that regulate angiogenic activity in IPF and that an imbalance exists in their expression that favors net angiogenesis in this disease (9). We found that levels of IL-8 were greater from tissue specimens of IPF patients than in those from controls. In contrast, IP-10 levels were higher from tissue specimens obtained from control subjects compared with IPF patients. When IL-8 or IP-10 was depleted from IPF tissue specimens, tissue-derived angiogenic activity was markedly reduced or enhanced, respectively. These findings support the idea that IL-8 and IP-10 are important factors that regulate angiogenic activity in IPF. To effectively assess the relevance of these mechanisms during the pathogenesis of pulmonary fibrosis in vivo, it is necessary to use an animal model of fibrotic lung disease. Bleomycin sulfate has been used in rodents to initiate fibrotic lung lesions, which have many of the histologic components of IPF (18, 19). Bleomycin administration results in a route-, dose-, and strain-dependent pulmonary inflammatory response characterized by increases in

FIGURE 10. Fibroblast proliferation at 72 h in response to varying concentrations of MIP-2 or a single concentration of platelet-derived growth factor (PDGF). Results from six individual experiments and are representative of 24 and 48 h data.

The Journal of Immunology

FIGURE 11. Representative RT-PCR of procollagen mRNA from pulmonary fibroblasts. MIP-2 fails to induce gene expression of either procollagen I or II. TGF-b, our positive control, was found to increase gene expression of procollagen III. Experimental n 5 9.

leukocyte accumulation, fibroblast proliferation, and collagen content (18 –23). In addition, the lungs of these animals typically exhibit necrosis of type I pneumocytes within the first 24 h postchallenge, acute alveolitis 2–3 days postchallenge, and intense interstitial inflammation 4 –12 days postchallenge (12, 19, 24, 25). Moreover, fibroblast proliferation and extracellular matrix synthesis are initiated 4 –14 days postbleomycin challenge, with collagen content elevated approximately twofold at 3 wk postchallenge (12, 19 –26). Although these pathologic changes occur in a more rapid fashion than those in human IPF, the rodent pulmonary inflammatory response to intratracheal bleomycin challenge constitutes a representative model of human IPF. Our studies have demonstrated that during the pathogenesis of bleomycin-induced pulmonary fibrosis there is increased expression of the CXC chemokine, MIP-2, which favors augmented net angiogenic activity. Lung tissue from bleomycin-treated animals had elevated levels of MIP-2 compared with control values and demonstrated in vivo angiogenic activity that could be significantly attenuated in the presence of neutralizing MIP-2 Abs. Immunolocalization of MIP-2 showed it to be associated with areas of fibrosis, and areas of MIP-2 expression were essentially devoid of neutrophil infiltration. The diffuse nature of the staining is consistent with the fact that CXC chemokines are heparin binding proteins that bind glycosoaminoglycans and is evidence for a significant presence of MIP-2 in the extracellular matrix (27). These findings support our previous observations of increased angiogenic activity in IPF lung tissue, which was significantly attributable to IL-8 (9). Similarly, areas of IL-8 expression were essentially devoid of neutrophil infiltration (9). Moreover, while neutralization of MIP-2 attenuated the fibrotic response, this did not appear to be mediated by its effect on neutrophil recruitment as there was no difference in MPO activity. This supports an alternative biologic role for MIP-2 in the pathogenesis of bleomycin-induced pulmonary fibrosis other than neutrophil recruitment. Further support for this idea is seen in the study by Thrall et al. (28). They showed that neutrophil depletion augmented the fibrotic response to bleomycin, suggesting a beneficial role for neutrophils in bleomycin-induced pulmonary fibrosis. Our results demonstrate that neutralization of MIP-2 leads to a reduction in the expression of factor VIII-related Ag in the lung, which correlates with the reduction in fibrosis. This supports the idea that MIP-2 is an important angiogenic factor during the pathogenesis of bleomycin-induced pulmonary fibrosis. In further support of the role of MIP-2 as an angiogenic factor is the role of the human functional homologue, IL-8, as an angiogenic factor in psoriasis and in the development of neovascularization of the pannus of rheumatoid arthritis (29 –31). Passive immunization with neutralizing MIP-2 Abs attenuated the fibrotic response to bleomycin. The magnitude of reduction in

5517 hydroxyproline was of a similar magnitude, as has previously been shown with other cytokine neutralization studies (12, 32–34). The lack of complete reduction of fibrosis suggests that there are potentially other angiogenic factors present in bleomycin-induced fibrosis. Our laboratory has previously shown that CXC chemokines other than MIP-2, such as GRO-a (human structural homologue of KC) display angiogenic activity (5). Furthermore, there are a variety of non-CXC chemokine molecules that can induce angiogenesis, and we cannot exclude their potential contribution to neovascularization in bleomycin-induced pulmonary fibrosis. Moreover, blocking angiogenesis may not be a complete way of blocking fibrosis, as a variety of mediators do have a direct effect on fibroblast proliferation and collagen gene expression (35–38). Further indirect support for the role of MIP-2 as an angiogenic factor during the pathogenesis of bleomycin-induced pulmonary fibrosis is the lack of effect of MIP-2 on fibroblast proliferation or on fibroblast collagen gene expression. This supports the contention that MIP-2 has no direct effect on pulmonary fibroblasts, suggesting that the beneficial effects of neutralization of MIP-2 are mediated through the regulation of angiogenesis. In conclusion, we have shown that in the context of bleomycininduced pulmonary fibrosis, a proangiogenic environment exists. This proangiogenic environment may be important in supporting fibroplasia and deposition of extracellular matrix during the pathogenesis of bleomycin-induced pulmonary fibrosis. This persistent proangiogenic environment contrasts with the normal process of tissue repair in which angiogenesis is usually rapidly expressed, transient, and tightly regulated (39 – 41). In addition, passive immunization with neutralizing MIP-2 Abs attenuates the fibrotic response to bleomycin through a mechanism that is independent of fibroblast proliferation, fibroblast collagen gene expression, or neutrophil recruitment or activation. Our findings support the idea that MIP-2 supports fibroplasia and deposition of extracellular matrix by regulating angiogenesis. These results suggest that targeting the regulation of angiogenesis may represent a viable therapeutic option for the treatment of pulmonary fibrosis.

References 1. Coultas, D. B., R. E. Zumwalt, W. C. Black, and R. E. Sobonya. 1994. The epidemiology of interstitial lung diseases. Am. J. Respir. Crit. Care Med. 150: 967. 2. du Bois, R. M. 1997. Diffuse lung disease: a view for the future. Sarcoidosis Vasc. Diffuse Lung Dis. 14:23. 3. Turner-Warwick, M. 1963. Precapillary systemic-pulmonary anastomoses. Thorax 18:225. 4. Peao, M. N. D., A. P. Aguas, C. M. DeSa, and N. R. Grande. 1994. Neoformation of blood vessels in association with rat lung fibrosis induced by bleomycin. Anat. Rec. 238:57. 5. Strieter, R. M., P. J. Polverini, S. L. Kunkel, D. A. Arenberg, M. D. Burdick, J. Kasper, J. Dzuiba, J. Van Damme, A. Walz, D. Marriott, et al. 1995. The functional role of the ELR motif in CXC chemokine-mediated angiogenesis. J. Biol. Chem. 270:27348. 6. Koch, A. E., P. J. Polverini, S. L. Kunkel, L. A. Harlow, L. A. DiPietro, V. M. Elner, S. G. Elner, and R. M. Strieter. 1992. Interleukin-8 (IL-8) as a macrophage-derived mediator of angiogenesis. Science 258:1798. 7. Hu, D. E., Y. Hori, and T. P. D. Fan. 1993. Interleukin-8 stimulates angiogenesis in rats. Inflammation 17:135. 8. Strieter, R. M., S. L. Kunkel, V. M. Elner, C. L. Martonyl, A. E. Koch, P. J. Polverini, and S. G. Elner. 1992. Interleukin-8: a corneal factor that induces neovascularization. Am. J. Pathol. 141:1279. 9. Keane, M. P., D. A. Arenberg, J. P. Lynch, R. I. Whyte, M. D. Iannettoni, M. D. Burdick, C. A. Wilke, S. B. Morris, M. C. Glass, B. DiGiovine, et al. 1997. The CXC chemokines, IL-8 and IP-10, regulate angiogenic activity in idiopathic pulmonary fibrosis. J. Immunol. 159:1437. 10. Arenberg, D. A., S. L. Kunkel, P. J. Polverini, M. Glass, M. D. Burdick, and R. M. Strieter. 1996. Inhibition of interleukin-8 reduces tumorigenesis of human non-small cell lung cancer in SCID mice. J. Clin. Invest. 97:2792. 11. Arenberg, D. A., S. L. Kunkel, P. J. Polverini, S. B. Morris, M. D. Burdick, M. Glass, D. T. Taub, M. D. Iannetoni, R. I. Whyte, and R. M. Strieter. 1996. Interferon-g-inducible protein 10 (IP-10) is an angiostatic factor that inhibits human non-small cell lung cancer (NSCLC) tumorigenesis and spontaneous metastases. J. Exp. Med. 184:981.

5518


12. Smith, R. E., R. M. Strieter, S. H. Phan, N. W. Lukacs, G. B. Huffnagle, C. A. Wilke, M. D. Burdick, P. Lincoln, H. Evanoff, and S. L. Kunkel. 1994. Production and function of murine macrophage inflammatory protein-1a in bleomycin-induced lung injury. J. Immunol. 153:4704. 13. Smith, D. R., P. J. Polverini, S. L. Kunkel, M. B. Orringer, R. I. Whyte, M. D. Burdick, C. A. Wilke, and R. M. Strieter. 1994. IL-8 mediated angiogenesis in human bronchogenic carcinoma. J. Exp. Med. 179:1409. 14. Huffnagle, G. B., M. F. Lipscomb, J. A. Lovchik, K. A. Hoag, and N. E. Street. 1994. The role of CD41 and CD81 T cells in the protective inflammatory response to a pulmonary cryptococcal infection. J. Leukocyte Biol. 55:35. 15. Colletti, L. M., S. L. Kunkel, A. Walz, M. D. Burdick, R. G. Kunkel, C. A. Wilke, and R. M. Strieter. 1995. Chemokine expression during hepatic ischemia/reperfusion-induced lung injury in the rat: the role of epithelial neutrophil activating protein. J. Clin. Invest. 95:134. 16. Henke, C., V. Fiegel, M. Peterson, M. Wick, D. Knighton, J McCarthy, and P. Bitterman. 1991. Identification and partial characterization of angiogenesis bioactivity in the lower respiratory tract after acute lung injury. J. Clin. Invest. 88:1386. 17. Marinelli, W. A., C. A. Henke, K. R. Harmon, M. I. Hertz, and P. B. Bitterman. 1990. Mechanisms of alveolar fibrosis after acute lung injury. Clin. Chest. Med. 11:657. 18. Adamson, I. Y., and D. H. Bowden. 1974. The pathogenesis of bleomycin-induced pulmonary fibrosis in mice. Am. J. Pathol. 77:185. 19. Chandler, D. B. 1990. Possible mechanisms of bleomycin-induced fibrosis. Clin. Chest Med. 11:21. 20. Bowden, D. H. 1984. Unraveling pulmonary fibrosis: the bleomycin model. Lab. Invest. 50:487. 21. Bedrossian, C. M. W., S. D. Grennberg, D. H. Yawn, and R. M. O’Neil. 1977. Experimentally induced bleomycin sulfate pulmonary toxicity. Arch. Pathol. Lab. Med. 101:248. 22. Aso, Y., and Y. Yoneda. 1976. Morphological and biochemical study of pulmonary changes induced by bleomycin in mice. Lab. Invest. 35:558. 23. Schrier, D. J., R. G. Kunkel, and S. H. Phan. 1983. The role of strain variation in murine bleomycin-induced pulmonary fibrosis. Am. Rev. Respir. Dis. 127:63. 24. Smith, R. E., R. M. Strieter, S. H. Phan, and S. L. Kunkel. 1996. CC chemokines: novel mediators of the profibrotic inflammatory response to bleomycin challenge. Am. J. Respir. Cell Mol. Biol. 15:693. 25. Chandler, D. B., D. M. Hyde, and S. N. Gin. 1983. Morphometric estimates of infiltrative cellular changes during the development of bleomycin-induced pulmonary fibrosis in hamsters. Am. J. Pathol. 112:170. 26. Piguet, P., C. Vesin, G. Grau, and R. C. Thompson. 1993. Interleukin 1 receptor antagonist (IL-1ra) prevents or cures pulmonary fibrosis elicited in mice by bleomycin or silica. Cytokine 5:57. 27. Webb, L. M. C., M. U. Ehrengruber, I. Clark-Lewis, M. Baggiolini, and A. Rot. 1993. Binding to heparan sulfate or heparin enhances neutrophil responses to interleukin 8. Proc. Natl. Acad. Sci. USA 90:7158.

28. Thrall, R. H., S. H. Phan, J. R. McCormick, and P. A. Ward. 1981. The development of bleomycin-induced pulmonary fibrosis in neutrophil-depleted and complement-depleted rats. Am. J. Pathol. 105:76. 29. Koch, A. E., P. J. Polverini, and S. J. Leibovich. 1986. Stimulation of neovascularization by human rheumatoid synovial tissue macrophages. Arthritis Rheum. 29:471. 30. Nickoloff, B. J., R. S. Mitra, J. Varani, V. M. Dixit, and P. J. Polverini. 1994. Aberrant production of interleukin-8 and thrombospondin-1 by psoriatic keratinocytes. Am. J. Pathol. 144:820. 31. Koch, A. E., S. L. Kunkel, J. L. Burrows, H. L. Evanoff, G. K. Haines, R. M. Pope, and R. M. Strieter. 1991. The synovial tissue macrophage as a source of the chemotactic cytokine interleukin-8. J. Immunol. 147:2187. 32. Piguet, P. F., C. Vesin, G. E. Grau, and R. C. Thompson. 1993. Interleukin 1 receptor antagonist (IL-1ra) prevents or cures pulmonary fibrosis elicited in mice by bleomycin or silica. Cytokine 5:57. 33. Piguet, P. F., C. Vesin, and F. Thomas. 1995. Bombesin down modulates pulmonary fibrosis elicited in mice by bleomycin. Exp. Lung Res. 21:227. 34. Gurujeyalakshmi, G., and S. N. Giri. 1995. Molecular mechanisms of antifibrotic effect of interferon g in bleomycin-mouse model of lung fibrosis: downregulation of TGF-b and procollagen I and III gene expression. Exp. Lung Res. 21:791. 35. Raines, E. W., D. F. Bowen-Pope, and R. Ross. 1990. Platelet-derived growth factor. In Handbook of Experimental Pharmacology, Vol. 95: Peptide Growth Factors and Their Receptors. M. B. Sporn and A. B. Roberts, eds. SpringerVerlag, Heidelberg, pp. 173–262. 36. Remmers, E. F., H. Sano, and R. L. Wilder. 1991. Platelet-derived growth factors and heparin-binding (fibroblast) growth factors in the synovial tissue pathology of rheumatoid arthritis. Semin. Arthritis Rheum. 21:191. 37. Postlethwaite, A. E., J. Keski-Oja, H. L. Moses, and A. H. Kang. 1987. Stimulation of the chemotactic migration of human fibroblasts by transforming growth factor b. J. Exp. Med. 165:251. 38. Leof, E. B., J. A. Proper, A. S. Goustin, G. D. Shipley, P. E. Dicorleto, and H. L. Moses. 1986. Induction of c-sis mRNA and activity similar to plateletderived growth factor by transforming growth factor b: a proposed model for indirect mitogenesis involving autocrine activity. Proc. Natl. Acad. Sci. USA 83:2453. 39. Clark, R. A. 1985. Cutaneous tissue repair: basic biologic considerations. I. Am. Acad. Dermatol. 13:701. 40. Clark, R. A. 1993. Basics of cutaneous wound repair. J. Dermatol. Surg. Oncol. 19:693. 41. Clark, R. A. 1993. Regulation of fibroplasia in cutaneous wound repair. Am. J. Med. Sci. 306:42.