of July 28, 2018.This information is current as

Bleomycin-Induced Pulmonary FibrosisFactor 5 in the Development of Opposing Regulatory Roles of Complement

Erin Addis-Lieser, Jörg Köhl and Mónica G. Chiaramonte

http://www.jimmunol.org/content/175/3/1894doi: 10.4049/jimmunol.175.3.1894

2005; 175:1894-1902; ;J Immunol 

Referenceshttp://www.jimmunol.org/content/175/3/1894.full#ref-list-1

, 23 of which you can access for free at: cites 71 articlesThis article

        average*  

4 weeks from acceptance to publicationFast Publication! •    

Every submission reviewed by practicing scientistsNo Triage! •    

from submission to initial decisionRapid Reviews! 30 days* •    

Submit online. ?The JIWhy

Subscriptionhttp://jimmunol.org/subscription

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/About/Publications/JI/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/alertsReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists All rights reserved.Copyright © 2005 by The American Association of1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on July 28, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

by guest on July 28, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

Opposing Regulatory Roles of Complement Factor 5 in theDevelopment of Bleomycin-Induced Pulmonary Fibrosis1

Erin Addis-Lieser,* Jorg Kohl,† and Monica G. Chiaramonte2*

The mechanisms of idiopathic pulmonary fibrosis pathogenesis, a chronic and progressive interstitial lung disease, remain elusive.The complement system, a crucial arm of the innate immune response, plays a pivotal role in several pathological disorders;however, the contribution of individual complement components to lung fibrosis has not yet been examined. Complement factor5 (C5) and its cleavage product C5a are critical mediators in inflammatory diseases. Thus, to evaluate the role of C5 in lungfibrosis, we compared congenic C5-sufficient and C5-deficient mice in a well-characterized murine model of bleomycin-inducedpulmonary fibrosis. C5-deficient mice had an exaggerated inflammatory phenotype compared with C5-sufficient mice during acutebleomycin-induced lung injury. These findings suggest a protective and anti-inflammatory role for C5, which was linked to theregulation of matrix metalloproteinases involved in cell migration. In contrast, C5 had a detrimental effect during chronic stagesof bleomycin-induced injury, indicating a profibrotic role for C5. This deleterious activity for C5 was associated with expressionof the fibrogenic cytokine TGF-�1 and matrix metalloproteinase-3, an important mediator in fibroblast contraction. Altogether,our data reveal novel and opposing roles for C5 in both inflammation and tissue repair. Furthermore, these findings provideinsight into the development of new therapeutic strategies for idiopathic pulmonary fibrosis patients. The Journal of Immunology,2005, 175: 1894–1902.

I diopathic pulmonary fibrosis (IPF)3 is a chronic interstitiallung disease of unknown etiology, characterized by diffuseand progressive fibrosis, which leads to a dramatic loss of

pulmonary function (1). This inflammatory and fibroproliferativedisorder has an incidence of �15 per 100,000/year in the U.S., andresults in death in 50–60% of the cases (2). The mechanisms un-derlying IPF pathogenesis remain largely unknown, although eti-ologic factors such as viral insults, smoking, and genetic predis-position have been associated in the process (1). Although severalstudies have implicated a variety of cytokines, chemokines, andextracellular matrix components in the development of pulmonaryfibrosis (3–5), the origin of tissue injury and the immunologicalcomponents involved in the abnormal repair process observed inIPF are still unclear. Moreover, IPF patients respond poorly to thecurrent therapeutic regimens (6); thus, a better understanding ofthis disorder is clearly warranted.

The experimental model of bleomycin-induced pulmonary fi-brosis has been used extensively to elucidate the basis of pulmo-nary fibrosis (7). Administration of bleomycin to rodents inducesan early inflammatory response, characterized by marked infiltra-

tion of neutrophils, eosinophils, and lymphocytes into the lung,followed by the development of interstitial fibrosis. This model hasprovided considerable insight into the immunological mechanismsinvolved in the pathogenesis of IPF, as well as other pulmonaryfibrotic diseases.

The complement system is a crucial arm of the innate immunesystem. Once activated, it results in enzymatic cleavage of severalproteins that form a complex network crucial for host defense. Theactivation of initial components of the cascade subsequently leadsto the formation of complement factor 5 (C5) convertase, gener-ating C5a anaphylatoxin and C5b from C5. Recent studies havedemonstrated a critical role for C5 in the pathogenesis of autoim-mune diseases, delayed-type hypersensitivity responses, experi-mental allergic asthma, and sepsis (8–11). Importantly, C5a hasproinflammatory effects, as well as immunomodulatory properties(12). C5 exerts a potent chemotactic activity for neutrophils, mac-rophages, basophils, mast cells, and B and T cells (13–17). Fur-thermore, C5 contributes to diverse biological functions rangingfrom early hemopoiesis, skeletal and vascular development, liverregeneration, cell survival, and apoptosis (18, 19).

A potential contribution of the complement system to IPF patho-genesis has been suggested by studies in human IPF patients. In-creased levels of immune complexes, which can directly activatethe complement cascade, as well as complement fragments of theclassical and alternative pathways, such as C3d, C4d, and Ba, havebeen detected in the circulation and bronchoalveolar fluid of IPFpatients (20–22), indicating extensive activation of complementpathways and a potential association with pulmonary pathogenesisin these patients. A potential role of the complement system inpulmonary fibrosis was previously suggested as complete deple-tion of downstream complement components attenuates bleomy-cin-induced fibrosis in mice (23). However, this experimental ap-proach does not allow the identification of any individualcomponent of the complement system potentially involved in theestablishment of fibrosis.

Thus, to evaluate the specific contribution of C5 in the pathogenesisof IPF, we compared the development of bleomycin-induced

*Division of Immunobiology and †Division of Molecular Immunology, CincinnatiChildren’s Hospital Research Foundation, Department of Pediatrics, University ofCincinnati, Cincinnati, OH 45229

Received for publication September 27, 2004. Accepted for publication May12, 2005.

The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby marked advertisement in accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.1 This work was supported in part by a research grant from the American LungAssociation (to M.G.C.).2 Address correspondence and reprint requests to Dr. Monica G. Chiaramonte, Divisionof Immunobiology (MLC 7038), Cincinnati Children’s Hospital Research Foundation,3333 Burnet Avenue, Cincinnati, OH 45229. E-mail address: monica.chiaramonte@cchmc.org3 Abbreviations used in this paper: IPF, idiopathic pulmonary fibrosis; BALF, bron-choalveolar lavage fluid; C5, complement factor 5; C5D, C5 deficient; C5S, C5 suf-ficient; Hp, hydroxyproline; MMP, matrix metalloproteinase; TIMP, tissue inhibitorof metalloproteinase.

The Journal of Immunology

Copyright © 2005 by The American Association of Immunologists, Inc. 0022-1767/05/$02.00

by guest on July 28, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

pulmonary fibrosis in congenic C57B10.D2 C5-sufficient (C5S)and C5-deficient (C5D) mice (24). Our findings reveal unique anddual roles for C5 in the establishment of bleomycin-inducedpulmonary fibrosis. Specifically, C5 appears to have a protectiverole during the initial inflammatory response by modulating matrixmetalloproteinases (MMPs) involved in cell migration. The profi-brotic role of C5, which is manifest during later stages of bleo-mycin-induced lung pathology, arises by regulation of the fibro-genic cytokine TGF-�1 and direct induction of MMP-3 productionin fibroblasts.

Materials and MethodsMice

Six- to 8-wk-old congenic C5S (B10.D2-Hc1H2d H2-T18c/0SnJ) and C5D(B10.D2-Hc0H2d H2-T18c/0SnJ) mice were obtained from The JacksonLaboratory. All mice were housed in Cincinnati Children’s Hospital Med-ical Center American Association for the Accreditation of Laboratory An-imal Care-approved animal facility in accordance with National Institutesof Health standards.

Bleomycin treatment

Mice were treated with bleomycin (Blenoxane; Mead Johnson) by intra-tracheal instillation, as previously described (25). Doses of 0.1 U (5 U/kg)or 0.03 U (1.5 U/kg) of bleomycin in a volume of 50 �l were administeredper mouse according to the experiment. Control animals were treated withthe same volume of PBS. Mice were sacrificed at different time points forcollection of different tissues. Right lung lobes were collected and inflatedwith 4% paraformaldehyde for histological analysis of tissue sectionsstained with H&E. For localization of collagen fibers within the lung tis-sues, Masson’s trichrome-specific staining was used. Bronchoalveolar la-vage fluid (BALF) was collected by flushing the airways three times with1 ml of PBS, and BALF supernatants were obtained after centrifugation for15 min at 12,000 � g.

BB5.1 Ab treatment

The anti-mouse C5 mAb (BB5.1, IgG1) (26, 27) was administered intra-tracheally to C5S mice using a dose of 500 �g on days 3, 10, and 17 afterbleomycin exposure (0.03 U/mouse). Mice were sacrificed on day 21 afterbleomycin administration.

BALF gelatin zymography

The 10% SDS-polyacrylamide gels containing gelatin (Invitrogen LifeTechnologies) were used to identify proteins with gelatinolytic activityfrom BALF supernatants (containing 10 �g of protein). After electrophore-sis in nonreducing conditions, gels were washed in renaturing buffer (In-vitrogen Life Technologies) for 30 min at room temperature to removeSDS, followed by incubation in developing buffer overnight at 37°C opti-mized for proteolytic activity. Protease activity appeared as clear bandsafter staining with Coomassie blue R250 and destaining in solution of 10%acetic acid. Gels were photographed in GelDoc (Bio-Rad). PurifiedMMP-2 and -9 controls were obtained from Chemicon International.

RNA extraction/RT-PCR

Total RNA was extracted from lung left lobes collected in STAT-60 RNA-zol, and cDNA was prepared by reverse transcription, as described (28).mRNA levels were assessed by real-time RT-PCR, using iCycler (Bio-Rad). Values were normalized by mRNA levels of �-actin (housekeepinggene), and relative units were calculated as: (�-actin Ct � gene Ct)1.8 �100,000, where Ct is the cycle threshold. Primers for IL-1�; TNF-�; TGF-�1; collagen I; collagen III; MMP-2, -8, -9, -13, -3, and -12; and TIMP-1were previously published (29–31).

Cytokine ELISAs

Lung homogenates were prepared from left lungs homogenized in PBS/protease inhibitors by using Omni Homogenizer. IL-1� and TGF-�1 pro-tein levels were measured in lung homogenates by capture ELISA, usingR&D Systems reagents, following the manufacturer’s instructions. Levelsof TGF-�1 were expressed as total TGF-�1 (active � latent) per milliliterof homogenate, after acidification of the samples.

Hydroxyproline (Hp) assay

To quantify collagen deposition, left lung lobes were hydrolyzed in 6 N ofHCl for 18 h at 110°C, and Hp levels were quantified by colorimetric assayusing Ehrlich’s reagent, as described (32).

Isolation of lung fibroblasts

Lung fibroblasts were isolated from left lung lobes, minced, subsequentlytreated with Liberase (Roche) for 1 h at 37°C, and filtered through 40-�mcell strainers. Adherent cells were grown in RPMI 1640 medium supple-mented with 10% FBS, 2 mM L-glutamine, 1 mM sodium pyruvate, 0.1mM nonessential amino acids, 25 mM HEPES, and 50 �M 2–ME (com-plete medium), in a humidified 5% CO2 air incubator at 37°C and passedevery 3–4 days by dissociation of confluent cultures with 0.25% trypsinwith 1 mM EDTA. For experiments, fibroblasts were used between pas-sages 6 and 8 and seeded at 1.5 � 105 cells in 25-cm2 flasks. Confluentcultures were rendered quiescent by eliminating serum content from themedium for 24 h, and RNA was extracted by using STAT60-RNAzol forsubsequent RT-PCR, as described above.

In vitro fibroblast studies

NIH3T3 murine fibroblasts were cultured in complete medium in a humid-ified 5% CO2 air incubator at 37°C. Cells from passages 4–8 were seededat 5 � 104 in 24-well plates and cultured with medium without serum for24 h before incubation with different doses of human rC5a (Sigma-Al-drich). Cell lysates were extracted at different time points by using lysisbuffer (M-Per; Pierce Lab), according to manufacturer’s instructions.Twenty micrograms of protein from cell lysates were submitted to elec-trophoresis in 4–20% Tris-glycine gels (Invitrogen Life Technologies) un-der reducing conditions and transferred to nitrocellulose membranes. Blotswere probed with goat biotinylated anti-mouse MMP-3 (R&D Systems) in1/250 dilution in PBS/Tween 20 0.05/5% BSA, followed by avidin-perox-idase (HRP) in a 1/10,000 dilution. Bands were viewed using ECL reagent(Immun-Star HRP; Bio-Rad). Purified human MMP-3 (R&D Systems) wasused as a control. For control of protein load, sheep anti-�-actin Ab(Chemicon International) was used in a 1/2000 dilution, followed by anti-sheep IgG (Zymed Laboratories) as secondary Ab. Expression of C5aR(CD88) in the surface of NIH3T3 fibroblasts was determined by FACSanalysis, using the specific mAb 20/70 (33) (kindly provided by J. Zwirner,Georg August University, Gottinge, Germany), by using a BD BiosciencesFACScan and CellQuest software.

Statistical analysis

For the survival experiments, statistical analysis was performed by Graph-Pad PRISM software using log-rank test to compare two groups. ANOVAtest was used to compare multiple groups using STATView 4.5 software.Statistical significance was determined by p � 0.05.

ResultsC5 protects from bleomycin-induced lethal lung injury

To investigate the involvement of C5 in the development of bleo-mycin-induced pulmonary fibrosis, we compared the responses tobleomycin exposure in congenic C57B10.D2 C5S and C5D mice.These mice have been used to analyze the role of C5 in several invivo models (34, 35). First, we compared the response of bothstrains to intratracheal administration of a high dose of bleomycin(0.1 U/mouse), to evaluate their respective susceptibility to lethallung injury. The absence of C5 markedly enhanced mortality, and50% of C5D mice died as early as 5 days after treatment and nonesurvived after 14 days. In contrast, only 40% of C5S mice diedduring this period (Fig. 1). When using a lower dose of bleomycin(0.03 U/mouse), all C5S and C5D mice survived as long as 28 daysafter treatment (data not shown). Thus, the intense detrimental re-sponse triggered by a high dose of bleomycin observed in C5Danimals strongly suggests a protective role for C5 during bleomy-cin-induced acute lung injury.

Bleomycin-induced pulmonary inflammation is enhanced in C5Dmice

To evaluate the mechanisms involved in the development of acuteinflammation and chronic fibrosis induced by bleomycin, we used

1895The Journal of Immunology

by guest on July 28, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

a lower and nonlethal dose of bleomycin (0.03 U/mouse) and an-alyzed inflammatory and fibrotic parameters at days 3, 7, 14, and21 posttreatment. As expected, histological evaluation of the lungtissue of C5S mice showed a strong cellular infiltration 7 days afterbleomycin administration (Fig. 2A), compared with the PBS-treated controls. In contrast, the inflammatory response to bleo-mycin was more pronounced in congenic C5D mice, with exten-sive areas of cellular infiltration in the lung compared with C5Smice (Fig. 2A, middle and right panels). These results were con-firmed by enumeration of total lung cells prepared after enzymatic

tissue dissociation (C5S bleomycin, 6.4 � 0.6 � 106 vs C5D bleo-mycin, 9.2 � 0.9 � 106 cells per left lung; p � 0.05).

The pronounced inflammatory response generated early afterbleomycin exposure in the absence of C5 was also associated withincreased expression of proinflammatory cytokines in the lung.Indeed, mRNA levels of IL-1� and TNF-� in C5D mice weresignificantly higher than those observed in C5S mice (Fig. 2B) asmeasured by RT-PCR in the lung tissue on day 7 posttreatment.C5D mice also exhibited significantly higher levels of IL-1� pro-tein in lung homogenates, when compared with C5S mice on day7 posttreatment (Fig. 2B). Overall, the exacerbated inflammatoryphenotype observed early after bleomycin exposure in C5D miceis strongly indicative of a protective role of C5 in acute lung injury.

Attenuation of bleomycin-induced pulmonary fibrosis in C5Dmice

To determine the effect of C5 on bleomycin-induced chronic fi-brosis, we assessed total lung collagen content by quantifying thelevels of Hp, an amino acid specifically present in collagen protein.Bleomycin-treated C5S mice showed a progressive increase in to-tal Hp levels in the lung, which was significantly higher duringchronic stages of the response (days 14 and 21) (Fig. 3A). C5Dmice also showed increased Hp levels compared with their con-trols, but surprisingly those levels were consistently reduced whencompared with C5S bleomycin-treated group on days 14 and 21

FIGURE 1. Survival rate of C5S and C5D mice after bleomycin treat-ment. C5S (f) and C5D (E) mice were treated by intratracheal instillationwith a single dose of bleomycin (0.1 U/mouse). Mice were monitored dailyfor survival. Ten mice were included in each group. Groups were signif-icantly different (p � 0.0001) analyzed by log-rank test.

FIGURE 2. Increased inflamma-tory phenotype in C5D mice early af-ter bleomycin administration. A,Lungs were collected 7 days aftertreatment with PBS or bleomycin(0.03 U/mouse), and sections werestained with Masson’s trichrome stain(blue) for visualization of collagen fi-bers (arrow), with H&E as counter-stain. Representative results (n � 4–6)are shown. Magnification � �100 (leftand middle panels) and �320 (rightpanels). B, Lungs from C5S and C5Dmice (n � 4–8 per group) were col-lected 7 days after treatment, and real-time RT-PCR was used to measuremRNA levels of proinflammatory cyto-kines IL-1� and TNF-�. �-actin genewas used to normalize all values. IL-1�protein levels were measured by ELISAin lung homogenates prepared from thedifferent groups, as described in Mate-rials and Methods. Results are ex-pressed as mean � SEM. �, Denotessignificant difference compared withPBS controls. †, Denotes significantdifference compared with C5S-bleomy-cin treated.

1896 DUAL ROLE OF C5 IN LUNG FIBROSIS

by guest on July 28, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

posttreatment (Fig. 3A). The difference in bleomycin-induced pul-monary fibrosis observed between the strains was also evident byhistological evaluation of lung tissue, which showed decreaseddeposition of collagen fibers in C5D than C5S mice on day 14 afterbleomycin administration (Fig. 3B, middle and right panels).These results were corroborated by analyzing mRNA expressionof the fibrillar collagens types I and III, because bleomycin-treatedC5D mice had significantly lower mRNA levels of both genescompared with bleomycin-treated C5S animals (Fig. 3C). Takentogether, these results suggest that C5 has a detrimental profibroticrole during later phases of the response to bleomycin, by enhanc-ing the level of collagen deposition.

In vivo blockade of C5 attenuates bleomycin-induced fibrosis

To confirm the specific role of C5 in the development of bleomy-cin-induced fibrosis, we decided to inhibit C5 activity in vivo byusing a specific Ab that prevents C5 cleavage into its fragmentsC5a and C5b (26, 27). Intratracheal treatment of C5S mice with theBB5.1 mAb prevented the development of bleomycin-induced fi-brosis, as evidenced by significantly decreased lung levels of Hp(Fig. 4). Importantly, these Hp levels were indistinguishable fromthose observed in C5S mice treated with PBS (Fig. 4). These re-sults, in conjunction with those obtained using C5D mice (Fig. 3),strongly support the important role of C5 in contributing to thelung fibrotic process induced by bleomycin.

C5 deficiency is associated with decreased expression ofTGF-�1

The cytokine TGF-�1 has been previously shown to be critical inthe establishment of bleomycin-induced pulmonary fibrosis (36),as well as other fibrotic processes (37). Therefore, we comparedthe expression of TGF-� induced in response to bleomycin admin-istration in C5S and C5D mice. RT-PCR analysis of total lungtissue showed that the TGF-�1 gene was significantly induced inC5S bleomycin-treated mice on days 14 and 21 after treatment,compared with their controls (Fig. 5A). In contrast, no increase inTGF-�1 mRNA was observed in C5D mice on days 14 and 21after bleomycin (Fig. 5A). More importantly, TGF-�1 mRNA lev-els in C5D mice were consistently lower than C5S mice on days 14and 21 after bleomycin exposure (Fig. 5A). Moreover, levels ofTGF-�1 protein in lung homogenates were significantly decreasedin C5D mice compared with C5S mice (Fig. 5B). These resultsindicate that the diminished fibrotic process observed in the ab-sence of C5 might be explained by the decreased expression of thefibrogenic mediator TGF-�1.

C5 modulates the expression of MMPs

To further investigate the effects of C5 on the fibrotic process, weevaluated the expression of matrix metalloproteinases (MMPs),

FIGURE 3. Attenuated collagendeposition during chronic stages ofthe response to bleomycin in C5Dmice. A, Levels of Hp were measuredby colorimetric assay in the left lungof C5S (f) and C5D (E) mice (n �8–10) collected at different timepoints after bleomycin administration(0.03 U/mouse). Values at each timepoint are representative of two tothree experiments and expressed asmean � SEM. Values at day 0 are in-dicative of average values for PBS-treated controls. �, Denotes signifi-cant difference compared with C5Sgroup. B, Lungs were collected 14days after treatment with PBS or bleo-mycin (0.03 U/mouse), and sectionswere stained with Masson’s trichromestain (blue) for visualization of colla-gen fibers (arrow). Representative re-sults (n � 4–6) are shown. Magnifi-cation � �100 (left and middlepanels) and �320 (right panels). C,Real-time RT-PCR was used to mea-sure interstitial collagens I and IIImRNA levels in right lungs of C5Sand C5D mice on day 14 after treat-ment with PBS or bleomycin. �-actingene was used to normalize all of thevalues (n � 4–6). Results are ex-pressed as mean � SEM. �, Denotessignificant difference compared withPBS controls. †, Denotes significantdifference when compared with C5S-bleomycin treated.

1897The Journal of Immunology

by guest on July 28, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

which have been extensively studied in several experimental mod-els (38). The MMP family of genes primarily controls extracellularmatrix degradation, which affects normal biological activities, aswell as a variety of pathological disorders (38). MMPs are regu-lated at level of gene transcription, by latent proenzyme activation(39), and are inhibited by a family of proteins known as tissueinhibitors of metalloproteinases (TIMPs), which bind to MMPs’active site and latent forms of MMPs (40).

Specifically, we compared the induction of MMP genes duringthe course of the response to bleomycin in both C5S and C5Dmice. First, we analyzed mRNA expression of collagenasesMMP-8 and MMP-13, which preferentially degrade fibrillar col-lagen types I, II, and III. C5S and C5D mice showed a similarpattern of expression for both genes in response to bleomycin,except on day 3 after exposure, when C5D mice exhibited signif-icantly higher mRNA levels of both MMP-8 and -13 comparedwith C5S mice (Table I).

Next, we analyzed bleomycin induction of gelatinases MMP-2and MMP-9, which degrade collagen IV, one of the main compo-nents of the basement membrane. MMP-2 mRNA levels were sig-nificantly lower in C5D mice compared with C5S mice only onday 14 after bleomycin, while no differences were observed atother time points (Table I). In the case of MMP-9, C5D mice hadsignificantly higher levels of this gene in the lung compared withC5S animals on day 3 after bleomycin exposure, but both strainsshowed similar expression later on in the response (Table I). Theseresults were confirmed at the protein level by zymography, which

allowed us to monitor gelanolytic activity in BALF. C5D micehave elevated MMP-9 enzymatic activity compared with onlyweak activity in C5S mice on day 3 after bleomycin exposure (Fig.6). No significant MMP-9 activity was observed in BALF in eitherC5S or C5D mice at later time points (data not shown). IncreasedMMP-2 enzymatic activity (latent and active) was detected asearly as day 3 (Fig. 6) and maintained during the course of theresponse to bleomycin in C5S and C5D mice (data not shown).

We analyzed the expression of MMP-12 (macrophage met-alloelastase), which degrades elastin and basement membranecomponents and plays a crucial role in chronic obstructive pulmo-nary disease (41). Bleomycin-treated C5S and C5D mice showedsimilar induction of MMP-12 mRNA throughout the course of theresponse, except on day 3 after bleomycin exposure, when C5Dmice had significantly higher levels than C5S mice (Table I).

MMP-3 (stromelysin-1) has been shown to play a relevant role inwound-healing process (42), as well as in the activation of other met-alloproteinase genes (43). At early phases of the response to bleomy-cin (days 3–7), both strains had a similar induction of MMP-3 com-pared with the PBS-treated mice (Fig. 7). C5S mice showed asignificant induction of MMP-3 at later time points after bleomycinexposure. However, the absence of C5 produced a significant de-crease in MMP-3 mRNA levels during later stages of the response tobleomycin (Fig. 7), on days 14 and 21 posttreatment compared withC5S mice. These results suggest that C5 contributes to MMP-3 mod-ulation during chronic stages of bleomycin-induced lung injury.

Finally, TIMP-1 mRNA levels were significantly higher in thelungs of C5D mice than C5S mice on day 3 after bleomycin ex-posure. However, no significant differences between the strainswere observed at other time points (Table I).

Reduced expression of interstitial collagens and MMP-3 in C5Dlung fibroblasts

Our results strongly suggest that C5 contributes to the establish-ment of chronic bleomycin-induced pulmonary fibrosis. To eval-uate whether C5 deficiency intrinsically affects one of the mainlung cell types involved in collagen deposition, we isolated lungfibroblasts from both C5S and C5D mice treated with bleomycin invivo and analyzed mRNA expression of collagens I and III. Ex-pression of these collagens in lung fibroblasts from C5D mice wassignificantly impaired compared with cells isolated from C5S an-imals (Fig. 8A). Furthermore, we found significantly reduced lev-els of MMP-3 mRNA (Fig. 8B) in C5D lung fibroblasts comparedwith cells from C5S mice. These results were confirmed by West-ern blot analysis, which showed increased production of MMP-3

FIGURE 5. TGF-�1 mRNA lung levels in C5S and C5D during the course of the response to bleomycin. Lungs were collected at several time pointsafter PBS or bleomycin (0.03 U/mouse) treatment. A, Real-time RT-PCR was used to measure TGF-�1 mRNA levels. �-actin was used as a housekeepinggene to normalize all of the values (n � 4–6). B, Total TGF-�1 protein levels (active � latent) were measured by ELISA in lung homogenates collectedon day 14 after treatment, as described in Materials and Methods. Values are expressed as mean � SEM. �, Denotes significant difference compared withPBS controls. †, Denotes significant difference when compared with C5S-bleomycin treated.

FIGURE 4. Attenuation of bleomycin-induced pulmonary fibrosis afterin vivo blockade of C5. C5S mice received 500 �g of mAb BB5.1 on days3, 10, and 17 after bleomycin exposure (0.03 U/mouse). Hp was measuredon left lungs collected on day 21 in the different groups (n � 3–5). Resultsare expressed as mean � SEM. Values of p between groups are indicated.NS � not significant.

1898 DUAL ROLE OF C5 IN LUNG FIBROSIS

by guest on July 28, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

protein in C5S fibroblasts, but no induction of this protein in C5Dcells (Fig. 8C). MMP-3, predominantly produced by fibroblasts(44), regulates fibroblast contraction and subsequently collagenproduction (45). Thus, our results strongly suggest that C5 regu-lates the tissue-remodeling process by directly affecting the activ-ity and functionality of lung fibroblasts.

C5a induces production of MMP-3 in murine fibroblasts

To directly prove the involvement of C5 on MMP-3 induction, weused an in vitro assay using murine NIH3T3 fibroblasts. Specifi-cally, we evaluated whether C5a has a direct effect on fibroblasts,because the activities of C5 are manifest through the generation ofthe C5a and C5b upon activation. C5a binds to a specific cellsurface receptor, C5aR (CD88), which belongs to a family of Gprotein-coupled receptors (46). Indeed, 3T3 fibroblasts expressC5aR on their surface, as determined by flow cytometric analysis(Fig. 9A). Western blotting analysis of cell lysates showed thattreatment with C5a for 24 h produced a dose-dependent increase inproduction of MMP-3 protein in fibroblasts (Fig. 9B). These re-sults strongly indicate that C5 can modulate fibrosis and tissueremodeling through a direct induction of MMP-3 production infibroblasts, thus affecting cell contraction and collagen production.

DiscussionTo date, the involvement of specific components of the comple-ment system in pulmonary fibrosis pathogenesis has not yet beenaddressed. Our studies using a murine model of bleomycin-in-duced pulmonary fibrosis reveal distinctive and novel roles for C5in the modulation of lung tissue injury and remodeling process.Remarkably, C5 appears to have both an early protective anti-inflammatory role as well as a later detrimental profibrotic effectduring bleomycin-induced lung injury.

The beneficial role of C5 was evident in lethal acute lung injury,as C5D mice showed significantly higher mortality after exposureto a high dose of bleomycin compared with C5S (Fig. 1). Becauseall C5D mice receiving this high dose of bleomycin succumbedbefore the onset of fibrosis, severe acute lung injury is the morelikely cause of the rapid mortality observed in these animals.Moreover, a similar protective effect for C5 was observed afterusing a lower and nonlethal dose of bleomycin, which produced aresponse to injury that progressed from early acute inflammation tolater chronic fibrosis. Indeed, early after exposure to this low doseof bleomycin, C5D mice exhibited an exacerbated inflammation,with marked cellular migration to the lung and high levels ofproinflammatory cytokines IL-1� and TNF-� on day 7, comparedwith C5S mice (Fig. 2). Thus, we postulate that C5 limits themagnitude of the early inflammatory phase triggered by bleomy-cin-induced injury.

One potential mechanism for the effect of C5 on early bleomy-cin-induced inflammation can be found in the expression of a par-ticular group of MMPs. MMPs and TIMPs are important mediatorsin both normal and pathological processes involving extracellularmatrix remodeling (39). Both MMPs and TIMPs have been de-tected in IPF patients (47, 48) and the bleomycin-induced fibrosismodel (49, 50); however, their specific role in pathogenesis is stillunclear. Specifically, our studies showed that C5D mice exhibitedincreased levels of MMP-8, -13, -9, -12, and TIMP-1 comparedwith C5S mice on day 3 after bleomycin treatment (Table I), which

FIGURE 6. Levels of MMP-2 and -9 in BALF from C5S and C5D miceafter bleomycin treatment. BALF samples, collected on day 3 after bleo-mycin exposure, were run in 10% gelatin gels (10 �g of protein per lane).Zones of enzymatic activity appear as clear bands. C, MMP-2- and -9-purified controls. M, m.w. Lanes 1–4, C5S bleomycin treated; lanes 5–7,C5D bleomycin treated. Arrow, MMP-9 band (92 kDa).

FIGURE 7. Expression of MMP-3 in C5S and C5D mice. Lungs werecollected from both strains at different time points after PBS or bleomycintreatment, and mRNA levels of MMP-3 were measured by real-time RT-PCR. Values at day 0 are indicative of average values for PBS-treatedcontrols. Values are expressed as mean � SEM. †, Denotes significantdifference between Bleo-treated groups.

Table I. Expression of MMPs and TIMPs in C5S and C5D micea

Early Phase Chronic Phase

Day 3 Day 7 Day 14 Day 21

C5S C5D C5S C5D C5S C5D C5S C5D

MMP-8 27 � 3 54 � 5b 184 � 19 159 � 22 118 � 13 96 � 5 90 � 4 81 � 5MMP-13 59 � 8 134 � 18b 13 � 1 14 � 2 65 � 4 48 � 4 45 � 4 52 � 3MMP-2 4337 � 508 4365 � 424 2492 � 297 3077 � 224 5256 � 360 3539 � 329b 4765 � 332 4670 � 332MMP-9 70 � 7 180 � 21b 28 � 3 37 � 5 45 � 6 42 � 4 23 � 2 24 � 3MMP-12 809 � 118 1321 � 196b 1221 � 144 1478 � 150 2614 � 267 2033 � 219 1513 � 145 1667 � 122TIMP-1 1393 � 129 2431 � 488b 1893 � 199 1689 � 209 524 � 75 398 � 57 79 � 9 76 � 7

a Lungs were collected from both strains (n � 6–10) at different time points after bleomycin treatment and mRNA levels of MMPs were measured by real-time RT-PCR andexpressed as relative units. Data are mean � SEM. Mean (SEM) from PBS-treated controls were as follows (C5S vs C5D): MMP-8 � 34 (4) vs 37 (8); MMP-13 � 21 (6) vs26 (8); MMP-2 � 2610 (482) vs 2679 (263); MMP-9 � 68 (25) vs 77 (28); MMP-12 � 157 (52) vs 141 (30); TIMP-1 � 80 (17) vs 89 (12).

b Significant difference ( p � 0.05) compared to C5S-Bleo-treated group.

1899The Journal of Immunology

by guest on July 28, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

preceded the peak influx of inflammatory cells observed in thelungs of C5D on day 7 (Fig. 2A). Interestingly, our results are inagreement with previous findings demonstrating the role of MMPsin cellular migration, due to their collagenase activity (51), as wellas their modulation of chemokines (52). For instance, high expres-sion of MMP-8 correlates with increased migration of inflamma-tory cells in a murine model of hepatitis B (53). MMP-12 has beenassociated with the ability of macrophages to migrate through ma-trix (54). MMP-9 is a crucial mediator in cell migration, as shownin models of acute lung injury (55) and asthma (56). Furthermore,it has been shown that MMP-9-deficient mice exhibited a dimin-ished degree of alveolar bronchiolization (57) after bleomycin ad-ministration, indicating that MMP-9 is important in facilitatingmigration of bronchiolar cells into regions of alveolar injury. Col-lectively, our findings suggest that the protective role for C5 duringthe early response to bleomycin may arise from the inhibition ofMMP-8, -13, -9, -12, and TIMP-1, particularly crucial in cell mi-gration to the lung. Moreover, our results reveal for the first timethe involvement of C5 in the regulation of MMPs.

The fact that C5 prevents an excessive early cellular infiltrationin the lung is surprising considering the well-known chemoattrac-tant activities of the C5a fragment, which is generated from C5during complement activation. However, an anti-inflammatoryrole for C5 has also been revealed in recent studies. For example,deletion of C5 or C5aR resulted in worsening of both pancreatitisand pancreatitis-associated lung injury (58). Furthermore, admin-istration of C5a protects against glutamate-mediated neurodegen-erative disorder (59). Although the mechanisms underlying the anti-inflammatory role of C5 remain unclear, some studies havesuggested that C5a can produce a desensitization of different celltypes to secondary cell chemoattractants, such as IL-8 and leuko-

triene B4 (60). Therefore, it is possible that C5 would protectagainst a robust early inflammatory response by decreasing theresponse to secondary chemokines in the injury site.

Importantly, our studies revealed an opposing role of C5 duringthe chronic phase of bleomycin-induced injury. The aggressiveinflammatory response initially developed in C5D mice after bleo-mycin did not translate into chronic exaggerated fibrosis later on inthe response. Quite the opposite, the absence of C5 resulted in apartial reduction in collagen deposition during the chronic stagesof the response to bleomycin (Fig. 3). These results were con-firmed by using a specific anti-C5 mAb (26). Indeed, in vivoblockade of C5 significantly attenuated the development of bleo-mycin-induced pulmonary fibrosis (Fig. 4). Altogether, our resultssupport a critical role for C5 as a profibrotic mediator and suggestthat the early inflammation and the late fibrosis induced by bleo-mycin are dissociated.

One mechanism by which C5 might participate in the chronicinduction of bleomycin-induced pulmonary fibrosis is by modula-tion of TGF-� expression. Specifically, we found that the attenu-ated fibrosis observed in C5D mice was associated with decreasedlevels of TGF-�1 mRNA and protein (Fig. 5). The crucial role ofTGF-� has been extensively studied in several fibrotic processes(37), and increased expression of TGF-� mRNA and protein hasbeen detected in IPF patients (61) and in bleomycin-induced fi-brosis (36). Our studies reveal a link between C5 and TGF-� reg-ulation that could be potentially important for fibrotic processes inother tissues.

Several investigators have emphasized the role of MMPs andTIMPs in the structural tissue remodeling that characterizes thechronic fibrotic response observed in IPF patients (47, 48, 62) andthe bleomycin-induced fibrosis model (49, 50). We found no major

FIGURE 9. Induction of MMP-3 by C5a in murine fibroblasts. A, Detection of surface CD88 (C5aR) by flow cytometry in NIH3T3 fibroblasts. Thinline, isotype control; dark line, anti-C5aR mAb 20/70. MFI, mean fluorescence intensity. B, Western blotting of fibroblast cell lysates treated for 24 h withdifferent doses of rC5a and probed with anti-mouse MMP-3 Ab.

FIGURE 8. Expression of interstitial collagens andMMP-3 in C5S and C5D mice fibroblasts. Lung fibro-blasts were isolated from C5S and C5D mice after bleo-mycin treatment, as described in Materials and Meth-ods. mRNA expression of collagen I, III (A), andMMP-3 (B) was evaluated by real-time RT-PCR in trip-licate. Values are expressed as mean � SEM. �, De-notes significant difference. C, Western blotting of lungfibroblast lysates isolated as described in Materials andMethods from C5S-PBS mice (lane 1), C5S-bleomycinmice (lane 2), C5D-PBS mice (lane 3), and C5D-bleo-mycin mice (lane 4). Membranes were probed with anti-mouse MMP-3 Ab and anti-� actin for control of pro-tein load.

1900 DUAL ROLE OF C5 IN LUNG FIBROSIS

by guest on July 28, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

differences between C5S and C5D mice regarding the expressionof collagenases, gelatinases, or TIMPs during chronic stages of theresponse to bleomycin. However, C5D mice exhibited MMP-3mRNA levels lower than C5S mice on days 14 and 21 after bleo-mycin exposure, suggesting that the role of C5 in chronic tissuerepair is associated with its ability to regulate MMP-3 expression.Indeed, we demonstrated that the fragment C5a directly inducesMMP-3 protein production in murine fibroblasts (Fig. 9B). MMP-3belongs to the group of stromelysins, which degrade laminin, fi-bronectin, elastin, and nonhelical collagens IV and IX (63).MMP-3 has been shown to play a crucial role in osteo- and rheu-matoid arthritis (64, 65) and coronary heart disease (66). Thesefindings reveal for the first time a direct effect of the complementsystem in the intricate process of tissue repair and remodeling.

Another relevant role of MMP-3 in fibrosis has been discoveredusing models of wound healing, in which MMP-3 deficiency wasdirectly linked to a defect in fibroblast contraction (42, 45). Re-markably, fibroblasts from IPF patients show increased contractil-ity compared with normal patients (67), thus contributing to thedistortion of the lung architecture. Therefore, we hypothesize thatthe profibrotic role of C5 is directly linked to the ability of MMP-3to promote fibroblast contraction. Indeed, our results show thatlung fibroblasts from bleomycin-treated C5D mice have signifi-cantly diminished levels of collagens I/III (Fig. 8A) and MMP-3compared with fibroblasts from C5S mice (Fig. 8, B and C). As theexpression of these genes is closely related to fibroblasts’ contrac-tile capacity, these results strongly validate the essential contribu-tion of C5 to the tissue-remodeling process by direct modulation ofMMP-3 expression. Finally, it is notable that MMP-3 is preferen-tially expressed in the lungs of susceptible C75BL/6 mice com-pared with resistant BALB/c mice after bleomycin administration(data not shown), strongly suggesting a crucial role for MMP-3 inlung fibrosis pathogenesis.

Our studies revealing opposing regulatory roles for C5 in theestablishment of bleomycin-induced pulmonary fibrosis are in ac-cordance with recent studies identifying divergent roles for severalmediators in the fibrotic response. Specifically, IL-4 has beenshown to have an early anti-inflammatory role for IL-4 as well asa profibrotic role during chronic phases of the response in thebleomycin model (68). Moreover, mice deficient in integrin �v�6

developed exaggerated lung inflammation, but are protected frombleomycin-induced chronic pulmonary fibrosis (69). These find-ings challenge the notion that increased inflammation in responseto injury directly correlates with increased fibrosis during chronicstages. The dissociation between inflammation and fibrosis is sup-ported by the lack of efficacy of anti-inflammatory treatments inIPF patients, which emphasizes the relevance of fibrogenesis inboth lung injury and the repair process in pulmonary fibrosis (1).

In summary, our results highlight for the first time a significantcontribution for C5 in the development of chronic fibrosis anddisclose a new level of involvement of the innate immunity systemin tissue repair. Our studies propose that C5’s modulation of bleo-mycin-induced inflammation and fibrosis occurs in a biphasic fash-ion. At the beginning of the response to an acute injury, C5 has ananti-inflammatory role by modulating cell migration most likelythrough regulation of MMPs. At later time points of the response,C5 promotes fibrosis by regulating TGF-� and MMP-3 expression.As an effective therapy for pulmonary fibrosis is critically needed,our studies suggest that the manipulation of the complement sys-tem, particularly as inhibitors of C5/C5a have become available(70, 71), is a plausible therapeutic alternative for IPF treatment, aswell as other devastating fibrotic disorders.

AcknowledgmentsWe thank Dr. Tom Wynn for providing NIH3T3 fibroblasts;Dr. Heiko Hawlisch for his help on the flow cytometry analysis of murinefibroblasts; Drs. Brigitta Stockinger and Ivana Cecic for providing theBB5.1 hybridoma; Jennifer Best for her help in the production and puri-fication of the BB5.1 Ab; and Drs. Marsha Wills-Karp and Claire Chougnetfor helpful discussions and critically reviewing the manuscript.

DisclosuresThe authors have no financial conflict of interest.

References1. American Thoracic Society. 2000. Idiopathic pulmonary fibrosis: diagnosis and

treatment: International consensus statement: American Thoracic Society (ATS),and the European Respiratory Society (ERS). Am. J. Respir. Crit. Care Med. 161:646–661.

2. Coultas, D. B., and M. P. Hughes. 1996. Accuracy of mortality data for interstitiallung diseases in New Mexico, USA. Thorax 51: 717–720.

3. Zhang, Y., T. C. Lee, B. Guillemin, M. C. Yu, and W. N. Rom. 1993. EnhancedIL-1� and tumor necrosis factor-� release and messenger RNA expression inmacrophages from idiopathic pulmonary fibrosis or after asbestos exposure.J. Immunol. 150: 4188–4196.

4. Ziegenhagen, M. W., S. Schrum, G. Zissel, P. F. Zipfel, M. Schlaak, andJ. Muller-Quernheim. 1998. Increased expression of proinflammatory chemo-kines in bronchoalveolar lavage cells of patients with progressing idiopathic pul-monary fibrosis and sarcoidosis. J. Invest. Med. 46: 223–231.

5. Martinet, Y., O. Menard, P. Vaillant, J. M. Vignaud, and N. Martinet. 1996.Cytokines in human lung fibrosis. Arch. Toxicol. Suppl. 18: 127–139.

6. Richeldi, L., H. R. Davies, G. Ferrara, and F. Franco. 2003. Corticosteroids foridiopathic pulmonary fibrosis. Cochrane Database Syst. Rev. 3:CD002880.

7. Bowden, D. H. 1984. Unraveling pulmonary fibrosis: the bleomycin model. Lab.Invest. 50: 487–488.

8. Riedemann, N. C., R. F. Guo, T. A. Neff, I. J. Laudes, K. A. Keller, V. J. Sarma,M. M. Markiewski, D. Mastellos, C. W. Strey, C. L. Pierson, et al. 2002. In-creased C5a receptor expression in sepsis. J. Clin. Invest. 110: 101–108.

9. Kohl, J., and J. E. Gessner. 1999. On the role of complement and Fc�-receptorsin the Arthus reaction. Mol. Immunol. 36: 893–903.

10. Karp, C. L., and M. Wills-Karp. 2001. Complement and IL-12: yin and yang.Microbes Infect. 3: 109–119.

11. Drouin, S. M., J. Kildsgaard, J. Haviland, J. Zabner, H. P. Jia, P. B. McCray, Jr.,B. F. Tack, and R. A. Wetsel. 2001. Expression of the complement anaphylatoxinC3a and C5a receptors on bronchial epithelial and smooth muscle cells in modelsof sepsis and asthma. J. Immunol. 166: 2025–2032.

12. Hugli, T. E. 1986. Biochemistry and biology of anaphylatoxins. Complement 3:111–127.

13. Lett-Brown, M. A., and E. J. Leonard. 1977. Histamine-induced inhibition ofnormal human basophil chemotaxis to C5a. J. Immunol. 118: 815–818.

14. Schmid, E., R. L. Warner, L. D. Crouch, H. P. Friedl, G. O. Till, T. E. Hugli, andP. A. Ward. 1997. Neutrophil chemotactic activity and C5a following systemicactivation of complement in rats. Inflammation 21: 325–333.

15. Richards, S. W., P. K. Peterson, H. A. Verbrugh, R. D. Nelson,D. E. Hammerschmidt, and J. R. Hoidal. 1984. Chemotactic and phagocytic re-sponses of human alveolar macrophages to activated complement components.Infect. Immun. 43: 775–778.

16. Ottonello, L., A. Corcione, G. Tortolina, I. Airoldi, E. Albesiano, A. Favre, R.D’Agostino, F. Malavasi, V. Pistoia, and F. Dallegri. 1999. rC5a directs the invitro migration of human memory and naive tonsillar B lymphocytes: implica-tions for B cell trafficking in secondary lymphoid tissues. J. Immunol. 162:6510–6517.

17. Werfel, T., M. Oppermann, J. H. Butterfield, G. Begemann, J. Elsner, O. Gotze,and J. Zwirner. 1996. The human mast cell line HMC-1 expresses C5a receptorsand responds to C5a but not to C5a(desArg). Scand. J. Immunol. 44: 30–36.

18. Mastellos, D., and J. D. Lambris. 2002. Complement: more than a ‘guard’ againstinvading pathogens? Trends Immunol. 23: 485–491.

19. Cole, D. S., and B. P. Morgan. 2003. Beyond lysis: how complement influencescell fate. Clin. Sci. 104: 455–466.

20. Dreisin, R. B., M. I. Schwarz, A. N. Theofilopoulos, and R. E. Stanford. 1978.Circulating immune complexes in the idiopathic interstitial pneumonias. N. Engl.J. Med. 298: 353–357.

21. Haslam, P. L., F. Allan, A. F. Watling, C. Barrett, L. Morris, andM. Turner-Warwick. 1982. Impaired antibody-dependent cell-mediated cytotox-icity in cryptogenic fibrosing alveolitis (synonym: idiopathic pulmonary fibrosis).Clin. Exp. Immunol. 49: 59–66.

22. Jansen, H. M., A. J. Schutte, J. D. Elema, M. V. Giessen, R. P. Reig,M. A. Leeuwen, H. J. Sluiter, and T. H. The. 1984. Local immune complexes andinflammatory response in patients with chronic interstitial pulmonary disordersassociated with collagen vascular diseases. Clin. Exp. Immunol. 56: 311–320.

23. Phan, S. H., and R. S. Thrall. 1982. Inhibition of bleomycin-induced pulmonaryfibrosis by cobra venom factor. Am. J. Pathol. 107: 25–28.

24. Wheat, W. H., R. Wetsel, A. Falus, B. F. Tack, and R. C. Strunk. 1987. The fifthcomponent of complement (C5) in the mouse: analysis of the molecular basis fordeficiency. J. Exp. Med. 165: 1442–1447.

1901The Journal of Immunology

by guest on July 28, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

25. Chen, E. S., B. M. Greenlee, M. Wills-Karp, and D. R. Moller. 2001. Attenuationof lung inflammation and fibrosis in interferon-�-deficient mice after intratrachealbleomycin. Am. J. Respir. Cell Mol. Biol. 24: 545–555.

26. Wang, Y., S. A. Rollins, J. A. Madri, and L. A. Matis. 1995. Anti-C5 monoclonalantibody therapy prevents collagen-induced arthritis and ameliorates establisheddisease. Proc. Natl. Acad. Sci. USA 92: 8955–8959.

27. Frei, Y., J. D. Lambris, and B. Stockinger. 1987. Generation of a monoclonalantibody to mouse C5 application in an ELISA assay for detection of anti-C5antibodies. Mol. Cell Probes 1: 141–149.

28. Chiaramonte, M. G., M. Mentink-Kane, B. A. Jacobson, A. W. Cheever,M. J. Whitters, M. E. Goad, A. Wong, M. Collins, D. D. Donaldson,M. J. Grusby, and T. A. Wynn. 2003. Regulation and function of the interleukin13 receptor �2 during a T helper cell type 2-dominant immune response. J. Exp.Med. 197: 687–701.

29. Boutin, H., R. A. LeFeuvre, R. Horai, M. Asano, Y. Iwakura, and N. J. Rothwell.2001. Role of IL-1� and IL-1� in ischemic brain damage. J. Neurosci. 21:5528–5534.

30. Overbergh, L., D. Valckx, M. Waer, and C. Mathieu. 1999. Quantification ofmurine cytokine mRNAs using real time quantitative reverse transcriptase PCR.Cytokine 11: 305–312.

31. Vaillant, B., M. G. Chiaramonte, A. W. Cheever, P. D. Soloway, and T. A. Wynn.2001. Regulation of hepatic fibrosis and extracellular matrix genes by the Thresponse: new insight into the role of tissue inhibitors of matrix metalloprotein-ases. J. Immunol. 167: 7017–7026.

32. Bergman, I., and R. Loxley. 1963. Two improved and simplified methods for thespectrophotometric determination of hydroxyproline. Anal. Biochem. 35: 1961–1965.

33. Zwirner, J., A. Fayyazi, and O. Gotze. 1999. Expression of the anaphylatoxin C5areceptor in non-myeloid cells. Mol. Immunol. 36: 877–884.

34. Barton, P. A., and J. S. Warren. 1993. Complement component C5 modulates thesystemic tumor necrosis factor response in murine endotoxic shock. Infect. Im-mun. 61: 1474–1481.

35. Mastellos, D., J. C. Papadimitriou, S. Franchini, P. A. Tsonis, and J. D. Lambris.2001. A novel role of complement: mice deficient in the fifth component of com-plement (C5) exhibit impaired liver regeneration. J. Immunol. 166: 2479–2486.

36. Wang, Q., Y. Wang, D. M. Hyde, P. J. Gotwals, V. E. Koteliansky, S. T. Ryan,and S. N. Giri. 1999. Reduction of bleomycin induced lung fibrosis by trans-forming growth factor � soluble receptor in hamsters. Thorax 54: 805–812.

37. Letterio, J. J., and A. B. Roberts. 1998. Regulation of immune responses byTGF-�. Annu. Rev. Immunol. 16: 137–161.

38. Parks, W. C., C. L. Wilson, and Y. S. Lopez-Boado. 2004. Matrix metallopro-teinases as modulators of inflammation and innate immunity. Nat. Rev. Immunol.4: 617–629.

39. Parks, W. C., and S. D. Shapiro. 2001. Matrix metalloproteinases in lung biology.Respir. Res. 2: 10–19.

40. Gomez, D. E., D. F. Alonso, H. Yoshiji, and U. P. Thorgeirsson. 1997. Tissueinhibitors of metalloproteinases: structure, regulation and biological functions.Eur. J. Cell Biol. 74: 111–122.

41. Suzuki, R., Y. Miyazaki, K. Takagi, K. Torii, and H. Taniguchi. 2004. Matrixmetalloproteinases in the pathogenesis of asthma and COPD: implications fortherapy. Treat. Respir. Med. 3: 17–27.

42. Bullard, K. M., L. Lund, J. S. Mudgett, T. N. Mellin, T. K. Hunt, B. Murphy,J. Ronan, Z. Werb, and M. J. Banda. 1999. Impaired wound contraction instromelysin-1-deficient mice. Ann. Surg. 230: 260–265.

43. Unemori, E. N., M. J. Bair, E. A. Bauer, and E. P. Amento. 1991. Stromelysinexpression regulates collagenase activation in human fibroblasts: dissociable con-trol of two metalloproteinases by interferon-�. J. Biol. Chem. 266: 23477–23482.

44. Pilcher, B. K., M. Wang, X. J. Qin, W. C. Parks, R. M. Senior, and H. G. Welgus.1999. Role of matrix metalloproteinases and their inhibition in cutaneous woundhealing and allergic contact hypersensitivity. Ann. NY Acad. Sci. 878: 12–24.

45. Bullard, K. M., J. Mudgett, H. Scheuenstuhl, T. K. Hunt, and M. J. Banda. 1999.Stromelysin-1-deficient fibroblasts display impaired contraction in vitro. J. Surg.Res. 84: 31–34.

46. Gerard, C., and N. P. Gerard. 1994. C5a anaphylatoxin and its seven transmem-brane-segment receptor. Annu. Rev. Immunol. 12: 775–808.

47. Suga, M., K. Iyonaga, T. Okamoto, Y. Gushima, H. Miyakawa, T. Akaike, andM. Ando. 2000. Characteristic elevation of matrix metalloproteinase activity inidiopathic interstitial pneumonias. Am. J. Respir. Crit. Care Med. 162:1949–1956.

48. Selman, M., V. Ruiz, S. Cabrera, L. Segura, R. Ramirez, R. Barrios, andA. Pardo. 2000. TIMP-1, -2, -3, and -4 in idiopathic pulmonary fibrosis: a pre-vailing nondegradative lung microenvironment? Am. J. Physiol. Lung Cell Mol.Physiol. 279: L562–L574.

49. Swiderski, R. E., J. E. Dencoff, C. S. Floerchinger, S. D. Shapiro, andG. W. Hunninghake. 1998. Differential expression of extracellular matrix remod-eling genes in a murine model of bleomycin-induced pulmonary fibrosis.Am. J. Pathol. 152: 821–828.

50. Corbel, M., C. Belleguic, E. Boichot, and V. Lagente. 2002. Involvement ofgelatinases (MMP-2 and MMP-9) in the development of airway inflammation andpulmonary fibrosis. Cell Biol. Toxicol. 18: 51–61.

51. Werb, Z., and J. R. Chin. 1998. Extracellular matrix remodeling during morpho-genesis. Ann. NY Acad. Sci. 857: 110–118.

52. Van Den Steen, P. E., A. Wuyts, S. J. Husson, P. Proost, J. Van Damme, andG. Opdenakker. 2003. Gelatinase B/MMP-9 and neutrophil collagenase/MMP-8 pro-cess the chemokines human GCP-2/CXCL6, ENA-78/CXCL5 and mouse GCP-2/LIX and modulate their physiological activities. Eur. J. Biochem. 270: 3739–3749.

53. Sitia, G., M. Isogawa, M. Iannacone, I. L. Campbell, F. V. Chisari, andL. G. Guidotti. 2004. MMPs are required for recruitment of antigen-nonspecificmononuclear cells into the liver by CTLs. J. Clin. Invest. 113: 1158–1167.

54. Shipley, J. M., R. L. Wesselschmidt, D. K. Kobayashi, T. J. Ley, andS. D. Shapiro. 1996. Metalloelastase is required for macrophage-mediated pro-teolysis and matrix invasion in mice. Proc. Natl. Acad. Sci. USA 93: 3942–3946.

55. Warner, R. L., L. Beltran, E. M. Younkin, C. S. Lewis, S. J. Weiss, J. Varani, andK. J. Johnson. 2001. Role of stromelysin 1 and gelatinase B in experimental acutelung injury. Am. J. Respir. Cell Mol. Biol. 24: 537–544.

56. Lee, K. S., S. M. Jin, H. J. Kim, and Y. C. Lee. 2003. Matrix metalloproteinaseinhibitor regulates inflammatory cell migration by reducing ICAM-1 andVCAM-1 expression in a murine model of toluene diisocyanate-induced asthma.J. Allergy Clin. Immunol. 111: 1278–1284.

57. Betsuyaku, T., Y. Fukuda, W. C. Parks, J. M. Shipley, and R. M. Senior. 2000.Gelatinase B is required for alveolar bronchiolization after intratracheal bleomy-cin. Am. J. Pathol. 157: 525–535.

58. Bhatia, M., A. K. Saluja, V. P. Singh, J. L. Frossard, H. S. Lee, L. Bhagat,C. Gerard, and M. L. Steer. 2001. Complement factor C5a exerts an anti-inflammatory effect in acute pancreatitis and associated lung injury.Am. J. Physiol. Gastrointest. Liver Physiol. 280: G974–G978.

59. Osaka, H., P. Mukherjee, P. S. Aisen, and G. M. Pasinetti. 1999. Complement-derived anaphylatoxin C5a protects against glutamate-mediated neurotoxicity.J. Cell. Biochem. 73: 303–311.

60. Foxman, E. F., J. J. Campbell, and E. C. Butcher. 1997. Multistep navigation andthe combinatorial control of leukocyte chemotaxis. J. Cell Biol. 139: 1349–1360.

61. Broekelmann, T. J., A. H. Limper, T. V. Colby, and J. A. McDonald. 1991.Transforming growth factor �1 is present at sites of extracellular matrix geneexpression in human pulmonary fibrosis. Proc. Natl. Acad. Sci. USA 88:6642–6646.

62. Henry, M. T., K. McMahon, A. J. Mackarel, K. Prikk, T. Sorsa, P. Maisi,R. Sepper, M. X. Fitzgerald, and C. M. O’Connor. 2002. Matrix metalloprotein-ases and tissue inhibitor of metalloproteinase-1 in sarcoidosis and IPF. Eur. Re-spir. J. 20: 1220–1227.

63. Welgus, H. G. 1991. Stromelysin: structure and function. Agents Actions Suppl.35: 61–67.

64. Massarotti, M., B. Marasini, A. Marchesoni, M. Arreghini, and M. L. Biondi.2003. Polymorphism in the stromelysin 1 (matrix metalloproteinase 3) promotergene and severity of rheumatoid arthritis: comment on the article by Constantinet al. Arthritis Rheum. 48: 2695–2696.

65. Tchetverikov, I., L. R. Lard, J. DeGroot, N. Verzijl, J. M. TeKoppele,F. C. Breedveld, T. W. Huizinga, and R. Hanemaaijer. 2003. Matrix metallopro-teinases-3, -8, -9 as markers of disease activity and joint damage progression inearly rheumatoid arthritis. Ann. Rheum. Dis. 62: 1094–1099.

66. Humphries, S. E., S. Martin, J. Cooper, and G. Miller. 2002. Interaction betweensmoking and the stromelysin-1 (MMP3) gene 5A/6A promoter polymorphismand risk of coronary heart disease in healthy men. Ann. Hum. Genet. 66: 343–352.

67. Fireman, E., I. Shahar, S. Shoval, G. Messer, S. Dvash, and J. Grief. 2001.Morphological and biochemical properties of alveolar fibroblasts in interstitiallung diseases. Lung 179: 105–117.

68. Huaux, F., T. Liu, B. McGarry, M. Ullenbruch, and S. H. Phan. 2003. Dual rolesof IL-4 in lung injury and fibrosis. J. Immunol. 170: 2083–2092.

69. Sheppard, D. 2001. Integrin-mediated activation of transforming growth factor-�(1) in pulmonary fibrosis. Chest 120: 49S–53S.

70. Hillmen, P., C. Hall, J. C. Marsh, M. Elebute, M. P. Bombara, B. E. Petro,M. J. Cullen, S. J. Richards, S. A. Rollins, C. F. Mojcik, and R. P. Rother. 2004.Effect of eculizumab on hemolysis and transfusion requirements in patients withparoxysmal nocturnal hemoglobinuria. N. Engl. J. Med. 350: 552–559.

71. Otto, M., H. Hawlisch, P. N. Monk, M. Muller, A. Klos, C. L. Karp, and J. Kohl.2004. C5a mutants are potent antagonists of the C5a receptor (CD88) and ofC5L2: position 69 is the locus that determines agonism or antagonism. J. Biol.Chem. 279: 142–151.

1902 DUAL ROLE OF C5 IN LUNG FIBROSIS

by guest on July 28, 2018http://w

ww

.jimm

unol.org/D

ownloaded from