Effect of Superoxide Dismutase on Bleomycin-InducedDermal Sclerosis: Implications for the Treatment of SystemicSclerosis

Toshiyuki Yamamoto, Shinsuke Takagawa, Ichiro Katayama* , Yutaka Mizushima† and Kiyoshi NishiokaDepartment of Dermatology, Tokyo Medical and Dental University School of Medicine, Tokyo, Japan; *Department of Dermatology, NagasakiUniversity School of Medicine, Nagasaki, Japan; †Division of Drug Delivery Systems, Institute of Medical Science, St Marianna University School ofMedicine, Kanagawa, Japan

Bleomycin has a chemical toxicity capable of inducing the hydroxyproline content in the skin was significantlyreduced, as compared with mice treated with bleomy-superoxide radicals, which are suggested to play ancin only or bleomycin and 5% mannitol. In a separateimportant part in bleomycin-induced pulmonary fib-experiment, after the development of dermal sclerosisrosis. We have recently established a mouse model forfollowing treatment with bleomycin for 3 wk, PC-scleroderma induced by repeated local injections ofsuperoxide dismutase was administered for 2 wk.bleomycin. In this study, we examined the inhibitoryHistologic examination again revealed a reduction ofeffect of superoxide dismutase on the development ofdermal sclerosis, followed by a significant associate indermal sclerosis induced by bleomycin using thisthe number of both mast cells and eosinophils. Themouse model. PC-superoxide dismutase, which is a hydroxyproline content in the skin was not significantlylecithinized superoxide dismutase with high tissue decreased, however, even after injections of high

accumulation and long half-life in blood, was adminis- amounts of PC-superoxide dismutase (30,000 U pertered (3000 U per kg; dissolved in 5% mannitol) 3 h kg). These results support the involvement of oxygenbefore the injection of bleomycin in C3H mice for free radicals in bleomycin-induced dermal sclerosis,3 wk. Systemic PC-superoxide dismutase markedly and also indicate that administration of superoxideinhibited the development of dermal sclerosis, which dismutase may be effective in the therapeutic approachwas also accompanied by a decrease in the number of in systemic sclerosis. Key words: animal model/bleomycin/

scleroderma/superoxide. J Invest Dermatol 113:843–847, 1999infiltrating mast cells and eosinophils. Furthermore,

S in the immune system, including T cells, macrophages, and mastcells, may contributed to extracellular matrix deposition and vasculardamage in SSc.

Peripheral neutrophils produce high levels of reactive oxygenspecies (ROS) in association with the severity and extent of skin

ystemic superoxide (SSc) is a connective tissue diseasewhich involves not only the skin but also various internalorgans leading to fibrosis with excessive collagen depos-ition. Although the pathogenesis of SSc is still unknown,the primary event is supposed to be an injury to endothel-

ial cells (LeRoy et al, 1989), and it is well known that inflammatory involvement in patients with SSc (Maslen et al, 1987; Law et al,1992). Recent studies have focused on the interaction of circulatingcells infiltrate in the dermis during the early phase of SSc (Fleisch-phagocytes, mainly (PMN), with endothelial cells and the potentiallymajer et al, 1977). These inflammatory cells are seen in associationcytotoxic effects of phagocyte products, including ROS (Harlan,with signs of endothelial injury with enhanced expression of1987). Increased free radical production is supposed to be involvedadhesion molecules. The most likely explanation for the prolifera-in the pathogenesis of SSc as an important initiator of tissue damage.tion of the intimal vascular wall is a sequence of events that follows

Bleomycin-induced lung injury is a well-known mouse modelendothelial injury, platelet adhesion as well as aggregation, andwhich resembles human pulmonary fibrosis histologically andrelease of platelet-derived growth factor (PDGF) and transformingbiochemically (Aso et al, 1976; Adamson, 1984; Chandler, 1990).growth factor-β (TGF-β) (Gibbons and Dzau, 1994). TGF-β,It is widely accepted that pulmonary fibrosis is in part due to anPDGF, tumor necrosis factor-α (TNF-α), or several interleukinsimbalance of oxidants and antioxidants (Strausz et al, 1990; Wallaert(IL-1, IL-2, IL-4, IL-6) are suggested to play a part in theet al, 1990). Oxygen free radicals are important initiators of tissuepathogenesis of SSc. These cytokines produced by activated cellsdamage (McCord, 1985), and there is also evidence that oxygenradicals are involved in bleomycin-induced lung injury (Hakkinen,1982; Frank, 1983). Bleomycin is known to generate ROS, such

Manuscript received June 17, 1999; revised July 30, 1999; accepted for as superoxide and hydroxyl radicals (Oberley and Buettner, 1979;publication August 5, 1999. Ekimoto et al, 1980; Guttering and Xiai-Chang, 1981). TheReprint requests to: Dr Toshiyuki Yamamoto, Department ofhypothesis that bleomycin leads to the generation of oxygenDermatololgy, Tokyo Medical and Dental University, School of Medicine,metabolites is supported by the fact that hypoxia protects against1–5-45 Yushima, Bunkyo-ku, Tokyo 113, Japan.the effects of intratracheal bleomycin and that hyperoxia vice versaAbbreviations: ROS, reactive oxygen species, SOD, superoxide

dismutase; SSc, systemic sclerosis augments bleomycin-induced injury (Tryka et al, 1982; Berend,

0022-202X/99/$14.00 · Copyright © 1999 by The Society for Investigative Dermatology, Inc.

843

844 YAMAMOTO ET AL THE JOURNAL OF INVESTIGATIVE DERMATOLOGY

1984). The exact mechanism of bleomycin-induced lung damageis not completely understood; however, it is believed that the initialinjury caused by this agent involves binding to DNA and Fe21,and this complex is thought to generate ROS under aerobicconditions (Giri and Wang, 1989).

Superoxide dismutase (SOD) is a family of enzymes that catalyzethe dismutation of O2

– to H2O2 and O2. Several different SODhave evolved to inactivate both intracellular and extracellularsuperoxide (Fridovich, 1995; Tsan, 1997). Human CuZn–SODhas been shown to be effective in a variety of animal models. Wehave recently established a mouse model for scleroderma inducedby bleomycin (Yamamoto et al, 1999). In this model, dermal sclerosismimicking human SSc both histologically and biochemically wasinduced by repeated local injections of bleomycin for 4 wk inBALB/c mice and for 3 wk in C3H mice. In this study, we haveexamined the therapeutic effect of PC-SOD, which has beenmodified in order to have a higher accumulation in the tissue anda longer half-life in peripheral blood (Igarashi et al, 1992, 1994),in the formation of bleomycin-induced dermal sclerosis in mice.

Figure 1. Representative photomicrographs of hematoxylin andeosin (H&E) staining showing effects of SOD on bleomycin-MATERIALS AND METHODSinduced dermal sclerosis. Simultaneous treatment of bleomycin and

Mice Specific pathogen-free female, 6 wk old C3H mice (purchased intravenous administration of PC-SOD (3000 U per kg) for 2 wkfrom Sankyo Co., Tokyo, Japan), weighing 20–25 g, were used. Mice demonstrated a significant decrease of sclerotic lesions, with slight fibrosiswere kept in separate clean rooms ad libitum. and mild cellular infiltration of mononuclear cells (A). In contrast, mice

treated with bleomycin and intravenous administration of 5% mannitolInduction of dermal sclerosis by bleomycin and administration of showed a definite sclerotic skin with thickened collagen bundles andPC-SOD As previously described (Yamamoto et al, 1999), mice were homogeneous deposition (B). (Original magnification, 3170.)treated by local bleomycin injections on the back. C3H mice had beendemonstrated to develop dermal sclerosis by subcutaneous daily bleomycininjections for 3 wk. First, mice were randomly classified into three groups RESULTS(n58 in each group) for the bleomycin treatment. One group received

Administration of PC-SOD exhibits inhibitory effects ononly bleomycin (100 µg per ml; diluted in phosphate-buffered saline)treatment; one group received intravenous injections of 100 µl of PC- the development of dermal sclerosis induced bySOD (generous gift from Seikagaku Kogyo Co. Ltd, Tokyo, Japan) (3000 U bleomycin At first, we examined the inhibitory effect of PC-per kg; diluted in 5% mannitol and filtrated by 0.2 µm filter) (decided in SOD administered together with bleomycin on the induction ofa preliminary experiment) 3 h before treatment with bleomycin. The third dermal sclerosis. In preliminary experiments, it was demonstratedgroup received injections with bleomycin and 5% mannitol in the same that administration of bleomycin for 3 wk induced dermal sclerosisamounts as the second group. In a separate experiment, PC-SOD was

in C3H mice. Therefore, PC-SOD was also applied for 3 wk.intravenously administered after the onset of dermal sclerosis induced byDaily administration of intravenous injections of PC-SOD (3000 Urepeated injections of bleomycin for 3 wk in C3H mice. Mice were killedper kg) with subcutaneous injections of bleomycin (100 µg perby cervical dislocation the next day after the final treatment, and the backml) for 3 wk attenuated the dermal sclerosis in C3H mice (Figskin was harvested for further examination.1A). A mild inflammatory infiltrate of mononuclear cells wasdetected in the lower dermis. In contrast, control mice treatedHistopathologic analysis Tissue specimens were fixed in 10% formalin

solution and embedded in paraffin. The general histologic appearance of with bleomycin alone, and bleomycin together with mannitol,the tissue was examined by routine hematoxylin and eosin staining. both showed a marked dermal sclerosis (Fig 1B). Analysis ofSkin specimens were assessed and scored to provide a semiquantitative the histologic results showed that administration of PC-SODmeasurement of dermal inflammation (0, none; 1, little; 2, mild; 3, significantly reduced the degree of dermal thickness and thickenedmoderate; 4, severe), thickened collagen bundles (0, normal; 1, little; 2, collagen bundles, whereas the degree of cellular infiltrates in themild; 3, moderate; 4, severe), and dermal thickness compared with normal

dermis was not significantly decreased (Fig 2). Hyalinization andskin (0, ,125%; 1, 125–150%; 2, 150–175%; 3, 175–200%; 4, .200%).sclerosis of the vascular wall was also diminished. Mast cells areDermal thickness was measured after taking photographs.suggested to play an important part in SSc, because the number ofmast cells is increased in early scleroderma (Nishioka et al, 1987).Morphometric analysis Mast cells containing metachromatic granulesEosinophils are also occasionally seen in scleroderma skin, andwere identified by Toluidine Blue staining at a pH of 2.5, 4.1, and 7.0.suggested to be one of the responsible cells for TGF-β (Zang et al,Eosinophils were identified by their polymorphonuclear morphology and

characteristic staining by Giemsa or 1% Congo Red. The numbers of mast 1995). Based on these findings, we examined the numbers of thesecells and eosinophils were quantified by counting 10 randomly chosen cells after SOD treatment. Mast cells were significantly decreasedhigh power fields (HPF; 3400) using an ocular grid of a light microscope, (17.261.6/HPF in mannitol treatment vs. 8.062.2/HPF in PC-and the mean number of cells was calculated. SOD treatment, p,0.05), and eosinophils were also significantly

reduced (7.061.1/HPF in mannitol treatment vs. 2.360.9/HPFHydroxyproline assay Six millimeter punch biopsy specimens were in PC-SOD treatment, p,0.05) after treatment with SOD for 3 wk.excised from the shaved back skin of each mouse, and stored at –80°C.Collagen deposition was estimated by determining the total hydroxyproline Hydroxyproline content is increased in the skin after bleomy-content of the skin. The biopsies were hydrolyzed with 6 M hydrochloric cin treatment, which is abrogated after administration ofacid under 110°C for 18 h as previously described (Woessener, 1961). PC-SOD and bleomycin As we have previously observedAfter neutralization with sodium hydroxide, the hydrolysates were diluted (Yamamoto et al, 1999), hydroxyproline contents in the skinwith distilled water. The amount of hydroxyproline in the hydrolysates

are increased in bleomycin-induced scleroderma skin. Next, wewas assessed colorimetrically at 560 nm with p-dimethylaminobenzaldehyde.examined whether the amount of hydroxyproline is decreased afterResults were expressed as micrograms hydroxyproline per skin.administration of PC-SOD, in parallel with histologic improvement.Treatment with injections of bleomycin and mannitol for 3 wkStatistical analysis Results were expressed as mean6SD. Significanceinduced an increase of the hydroxyproline content in the skin fortesting was analyzed using Mann–Whitney U-test. A p,0.05 was considered

to be significant. up to 177615% as compared with normal skin, whereas bleomycin

VOL. 113, NO. 5 NOVEMBER 1999 THERAPEUTIC ROLE OF SOD FOR BLEOMYCIN-INDUCED SCLERODERMA 845

Figure 2. Summary of the effects of SOD on cutaneoushistopathology. Shown are the mean6SD of the scores determined asdescribed in Materials and methods. *Groups that were significantly different(p,0.05) from the control mice treated with bleomycin only for eachhistopathologic parameter.

Figure 4. Effects of administration of SOD after onset of dermalsclerosis on the cutaneous histopathology. Intravenous administra-tion of PC-SOD (3000 U per kg) for 2 wk reduced the dermal sclerosis(A), compared with intravenous mannitol treatment (B). (Original magni-fication, 3170.)

Figure 3. Effects of SOD on bleomycin-induced dermal sclerosis.Cutaneous sclerosis was assessed by measurement of skin hydroxyprolinecontent after 3 wk of treatment. Bleomycin-induced cutaneous sclerosiswas significantly inhibited by treatment with simultaneous administrationof intravenous PC-SOD for 3 wk, compared with that in mice treatedwith intravenous mannitol (*p,0.05).

treatment together with PC-SOD showed an increase of only130611%. This difference reached a statistical significance (p,0.05)(Fig 3).

Intravenous injections of PC-SOD after onset of bleomycin-induced scleroderma does not improve sclerosis The post-onset situation is a much more stringent test of the ability of PC-SOD to inhibit scleroderma, as dermal thickness with thickenedcollagen bundles, collagen deposition, and cellular infiltrates arealready going on when the treatment is started. Furthermore, post-onset effects more closely reflect a clinical situation where symptomsare already apparent. To test the effects of PC-SOD on existing

Figure 5. Effects of post-onset treatment with SOD on cutaneousdermal sclerosis, we first treated C3H mice with bleomycin (100 µghistopathology. Shown are the mean6SD of the assessment values forper ml) for 3 wk. Then, after the development of dermal sclerosis, the indicated parameters. *Groups that were significantly different (p,0.05)

PC-SOD was given intravenously at a dose of 3000 and 30,000 U from the control mice treated with bleomycin only for eachper kg daily for 2 wk. Histologic evaluation revealed a reduction histopathologic parameter.of dermal sclerosis (Fig 4A,B). Analysis of the histologic changesindicates that the group receiving PC-SOD showed significantly DISCUSSIONlower collagen thickness and dermal thickness in the post-onsetadministration of PC-SOD experiment (Fig 5). A decrease of In this study, we demonstrated a therapeutic effect of PC-SOD on

bleomycin-induced dermal sclerosis in an experimental mousehyalinization and sclerosis of the vascular wall was found as well.Hydroxyproline content in the skin was slightly decreased model. Intravenous administration of SOD inhibited sclerosis not

only when given simultaneously with bleomycin injection but alsobut this decrease failed to show a significant difference afteradministration of PC-SOD 3000 or 30,000 U per kg (Fig 6). when given after the development of bleomycin-induced sclerosis,

846 YAMAMOTO ET AL THE JOURNAL OF INVESTIGATIVE DERMATOLOGY

indicates that monocytes contribute to the increased production ofsuperoxide anion in patients with SSc (Sambo et al, 1999). ROScan cause several abnormalities such as endothelial cell damage orenhanced platelet activation, leading to an upregulation of theexpression of adhesion molecules or the secretion of inflammatoryor fibrogenic cytokines including PDGF and TGF-β. In this study,the application of SOD together with bleomycin from the initialstage resulted in a significant reduction of dermal sclerosis, incontrast to post-onset administration of SOD after the inductionof sclerosis by bleomycin. Therefore, we hypothesized that ifbleomycin-related oxidative stress triggers sclerosis, then perhapsSOD protects from initiation. Our results in fact suggest that it isimportant to administer SOD before the development of sclerosis,and post-onset administration is insufficient to prevent the pro-gression.

Superoxide radicals produced as by-products of metabolic oxida-tion can cause extensive cellular injury, and the elimination ofFigure 6. Effects of post-onset administration of SOD on skinexcess superoxide may be important for the control of oxygenhydroxyproline content. Intravenous administration of PC-SOD showedradical related diseases. Because of the likely important role fora mild reduction of hydroxyproline content; however, the difference didreactive oxidants in causing injury to cellular constituents, anti-not reach any significance even when the dose of PC-SOD was increasedoxidant therapy has been used in animal models of lung injury.for up to 30,000 U per kg.Several studies have shown modulation of bleomycin-inducedpulmonary fibrosis with anti-oxidant therapies such as Mn–SOD,taurine, niacin, N-acetylcysteine, and amifostine (Parizada et al,which was also histologically apparent with reduced thickened

collagen bundles. This effect lasted at least 4 wk after the final 1991; Shahzeidi et al, 1991; Wang et al, 1991; Nici et al, 1998).SOD has evolved to inactivate both intracellular and extracellulartreatments (data not shown). Cellular infiltrates of the sclerotic skin

were rather mild both before and after the administration of superoxide (Fridovich, 1995; Tsan, 1997). CuZu–SOD and Mn–SOD are normally present in cells in sufficient amounts to scavengeSOD, although variations were found between individual mice.

Administration of PC-SOD after the development of dermal intracellular O2–. Although extracellular SOD has been reported

to protect the extracellular space and endothelial cell surface, thesclerosis could not significantly reduce skin hydroxyproline content,even when the dose was increased up to 30,000 U per kg. activity of extracellular SOD is low (Marklund, 1984; Karlsson and

Marklund, 1989), which suggests that exogenous SOD may be ofHydroxyproline is a major constituent of collagen. Dermal sclerosiswas histologically reduced, but still retained, by post-onset adminis- clinical value by scavenging excess O2

– at the cell membrane. PC-SOD has been shown to have a high affinity to the cell membrane,tration of PC-SOD. This may explain the discrepancy between

histologic and biochemical evaluation, and the results indicate that a high cellular uptake for both endothelial cells and neutrophils,and delayed plasma disappearance compared with unmodified SOD,early administration is important to obtain sufficient effects. In

contrast, local injections of PC-SOD did not show any effects (data and the pharmacologic activities are over 200 times more thanthose of unmodified SOD (Igarashi et al, 1992, 1994). The SODnot shown).

Using bleomycin, a number of studies have been performed to concentrations are assessed by enzyme-linked immunosorbent assayin the 6 mm punch-biopsied skins, which were all below thegenerate models of pulmonary fibrosis in animals. Bleomycin

upregulates the synthesis of type I collagen in lung and dermal detectable limits through the course of the experiments. Oneexplanation is due to the small size of the samples, because dermalfibroblasts (Kaelin et al, 1983). Other data, however, showed that

bleomycin in culture with lung slices significantly reduces the sclerosis is induced limited to the injected sites.A recent report has demonstrated that several autoantigensuptake of labeled proline into lung tissue, and inhibits hydroxylation

of proline into collagen, suggesting that the bleomycin-induced targeted by scleroderma autoantibodies are fragmented in thepresence of ROS and specific metals such as iron or copperincrease in collagen synthesis in vivo is not directly mediated (Phan

et al, 1991). In addition, bleomycin stimulates endothelial cells to (Casciola-Rosen et al, 1997). They therefore suggest that themetabolism of the metals is important for the formation of freesynthesize TGF-β at the protein and messenger levels, as well as

to upregulate collagen synthesis in fibroblasts (Phan et al, 1991). radical damage and abnormal accumulation of cellular metal mightbe of critical pathogenic importance in scleroderma. These findingsBleomycin is known to generate ROS, such as superoxide and

hydroxyl radicals. Several studies have shown the mechanism by support the notion that scleroderma is also characterized by isch-emia–reperfusion injury; overproduction of ROS is commonlywhich oxygen radicals are formed, suggesting a participation of the

Fenton reaction in which the iron acts as a redox agent (Chandler found in scleroderma patients with active disease, and that sclerod-erma is occasionally associated with a variety of environmentalet al, 1985, 1988a,b). Bleomycin complexes with free iron intracellu-

larly and, when bound to DNA, targets Fenton-driven OH toxins, such as silica or organic solvents. In conclusion, our resultssuggest that ROS is involved in the formation of bleomycin-formation to the DNA. Iron has been shown to be a critical

cofactor for the activity of prolyl hydroxylase, which is a key induced dermal sclerosis in this mouse model, and the cellularinfiltrate observed in this model probably plays a crucial part in theenzyme in collagen synthesis, and the involvement of free radical

was shown during the formation of hydroxylated proline by prolyl development of the skin lesions. Our data also suggest the possibilityof a therapeutic role for SOD in scleroderma.hydroxylase (Bhatnagar and Liu, 1972). Therefore, a reduction of

free radical formation may contribute to the decrease of collagencontent by inhibition of proline hydroxylation. Increased freeradical production is supposed to be involved in the pathogenesis

This work was supported in part by a grant of the Scleroderma Research of theof SSc as an important initiator of tissue damage (Murrel, 1993;Japanese Ministry of Health and Welfare.Herrick et al, 1994; Stein et al, 1996; Sambo et al, 1999). The exact

species produced, the mechanisms and sites of production, and theconsequences of these ROS, however, are still unknown. Peripheral

REFERENCESneutrophils produce high levels of ROS in patients with SSc, thatare associated with the severity and extent of skin involvement Adamson IVR: Drug induced pulmonary fibrosis. Environ Health Perspect 55:25–

36, 1984(Maslen et al, 1987; Law et al, 1992). In contrast, a recent paper

VOL. 113, NO. 5 NOVEMBER 1999 THERAPEUTIC ROLE OF SOD FOR BLEOMYCIN-INDUCED SCLERODERMA 847

Aso Y, Yoneda K, Kikkawa Y: Morphological and biological study of pulmonary determining the pathogenesis of systemic sclerosis. Is transforming growthfactor β the answer? Arthritis Rheum 32:817–825, 1989changes induced by bleomycin in mice. Lab Invest 35:558–568, 1976

Berend N: Protective effect of hyperoxia on bleomycin lung toxicity in the rat. Am Marklund SL: Extracellular superoxide dismutase in human tissue and human celllines. J Clin Invest 74:1398–1403, 1984Rev Respir Dis 130:307–308, 1984

Bhatnagar RS, Liu TZ: Evidence for free radical involvement in the hydroxylation Maslen CL, Hall ND, Woolf AD, Maddison PJ: Enhanced oxidative metabolism ofneutrophils from patients with systemic sclerosis. Br J Rheumatol 26:113–of proline: inhibition by nitro blue tetrazolium. FEBS Lett 26:32–34, 1972

Casciola-Rosen L, Wigley F, Rosen A: Scleroderma autoantigens are uniquely 117, 1987McCord JM: Oxygen-derived free radicals in post ischemic tissue injury. N Engl Jfragmented by metal-catalyzed oxidation reactions: implications for

pathogenesis. J Exp Med 185:71–79, 1997 Med 312:159–163, 1985Murrel DF: A radical proposal for the pathogenesis of scleroderma. J Am AcadChandler DB: Possible mechanisms of bleomycin-induced fibrosis. Clin Chest Med

11:21–30, 1990 Dermatol 28:78–85, 1993Nici L, Santos-Moore A, Kuhn C, Calabresi P: Modulation of bleomycin-inducedChandler DB, Hyde DM, Giri SN: The effect of deferoxamine on bleomycin-

induced lung fibrosis in the hamster. Am Rev Respir Dis 131:596–598, 1985 pulmonary toxicity in the hamster by the antioxidant amifostine. Cancer83:2008–2014, 1998Chandler DB, Barton JC, Briggs DD, Butler TW, Kennedy JI, Grizzle We, Fulmer

JD: Effect of iron deficiency on bleomycin-induced lung fibrosis in the hamster. Nishioka K, Kobayashi Y, Katayama I, Takijiri C: Mast cell numbers in diffusescleroderma. Arch Dermatol 123:205–208, 1987Am Rev Respir Dis 137:85–89, 1988a

Chandler DB, Butler TW, Briggs DD III, Grizzle We, Barton JC, Fulmer JD: Oberley LW, Buettner GR: The production of hydroxyl radicals by bleomycin andiron(II). FEBS Lett 97:47–49, 1979Modulation of the development of bleomycin-induced pulmonary fibrosis by

deferoxamine. Toxicol Appl Pharmacol 92:358–367, 1988b Parizada B, Werber MM, Nimrod A: Protective effect of human recombinantMnSOD in adjuvant arthritis and bleomycin-induced lung fibrosis. Free RadEkimoto H, Kuramochi H, Takahashi K, Matsuda A, Umezawa H: Kinetics of the

reaction of bleomycin Fe(II) –O2 complex with DNA. J Antibiot (Tokyo) 33: Res Commun 15:297–301, 1991Phan SH, Gharaee-Kermani M, Wolber F, Ryan US: Stimulation of rat endothelial426–434, 1980

Fleischmajer R, Perlish JS, Reeves JRT: Cellular infiltrates in scleroderma skin. cell transforming growth factor-β production by bleomycin. J Clin Invest87:148–154, 1991Arthritis Rheum 20:975–984, 1977

Frank L: Superoxidedismutase and lung toxicity. Trend Pharmacol Sci 4:124–136, 1983 Sambo P, Jannino L, Candela M, et al: Monocytes of patients with systemic sclerosis(scleroderma) spontaneously release in vitro increased amounts of superoxideFridovich I: Superoxide radical and superoxide dismutase. Annu Rev Biochem 64:97–

112, 1995 anion. J Invest Dermatol 112:78–84, 1999Shahzeidi S, Sarnstrand B, Jeffrey PK: Oral N-acetylcysteine reduces bleomycin-Gibbons GH, Dzau VJ: The emerging concept of vascular remodeling. New Engl J

Med 330:1431–1438, 1994 induced collagen deposition in the lungs of mice. Eur Respir J 4:845–852, 1991Strausz J, Muller-Quernheim J, Steppling H, Forlinz R: Oxygen radical productionGiri SN, Wang Q: Mechanisms of bleomycin-induced lung injury. Comments Toxicol

3:145–176, 1989 by alveolar inflammatory cells in idiopathic pulmonary fibrosis. Am Rev RespirDis 141:124–128, 1990Gutteridge JMC, Ziai-Chang F: Protection of iron-catalyzed free radical damage to

DNA And lipids by copper(II) bleomycin. Biochem Biophys Res Commun Stein CM, Tanner SB, Awad JA, Roberts LI, Morrow JD: Evidence for free radical-mediated injury (isoprostane overproduction) in scleroderma. Arthritis Rheum99:1354–1360, 1981

Hakkinen PJ, Whiteley JW, Witschi HP: Hyperoxia, but not thoracic X-irradiation, 37:1146–1150, 1996Tryka AF, Skornik WA, Godleski JJ, Brain JD: Potentiation of bleomycin-inducedpotentiates bleomycin- and cyclophosphamide-induced lung damage in mice.

Am Rev Respir Dis 126:281–297, 1982 lung injury by exposure to 70% oxygen. Am Rev Respir Dis 126:1074–1079, 1982Harlan JH: Neutrophil-mediated vascular injury. Acta Med Scand 715(Suppl.):123–

129, 1987 Tsan M-F: Superoxide dismutase and pulmonary oxygen toxicity. Proc Soc Exp BiolMed 214:107–113, 1997Herrick AL, Rieley F, Schofield D, Hollis S, Braganza JM, Jayson MIV: Micronutrient

antioxidant status in patients with primary Raynaud’s phenomenon and systemic Wallaert B, Lassalle P, Fortin F, Aerts C, Bart F, Fournier E, Voisin C: Superoxideanion generation by alveolar inflammatory cells in simple pneumoconiosis andsclerosis. J Rheumatol 21:1477–1483, 1994

Igarashi R, Hoshino J, Takenaga M, et al: Lecithinization of superoxide dismutase in progressive massive fibrosis of nonsmoking coal workers. Am Rev Respir Dis141:129–133, 1990potentiates its protective effect against forsman antiserum-induced elevation in

guinea pig airway resistance. J Pharmacol Exp Ther 262:1214–1219, 1992 Wang QJ, Giri SN, Hyde DM: Amelioration of bleomycin-induced pulmonaryfibrosis in hamsters by combined treatment with taurine and niacin. BiochemIgarashi R, Hoshino J, Ochiai A, Morizawa Y, Mizushima Y: Lecithinized superoxide

dismutase enhances its pharmacologic potency by increasing its cell membrane Pharmacol 42:1115–1122, 1991Woessener JF: The determination of hydroxyproline in tissue and protein samplesaffinity. J Pharmacol Exp Ther 271:1672–1677, 1994

Kaelin RM, Center DM, Bernardo J, Grant M, Snider GL: The role of macrophage- containing small proportions of this amino acid. Arch Biochem Biophys 93:440–447, 1961derived chemoattractant activities in the early inflammatory events of

bleomycin-induced pulmonary injury. Am Rev Respir Dis 128:132–137, 1983 Yamamoto T, Takagawa S, Katayama I, Yamazaki K, Hamazaki Y, Shinkai H,Nishioka K: Animal model of sclerotic skin. I: Local injections of bleomycinKarlsson K, Marklund SL: Binding of human extracellular superoxide dismutase C

in cultured cell line and blood cells. Lab Invest 60:659–666, 1989 induce sclerotic skin mimicking scleroderma. J Invest Dermatol 112:456–462, 1999Law CS, Sridges AB, Muir A, Scott N, Bancroft A, Belch JJF: Further evidence

of increased polymorphonuclear cell activity in patients with Raynaud’s Zang K, Flanders KC, Phan SH: Cellular localization of transforming growth factor-β expression in bleomycin-induced pulmonary fibrosis. Am J Pathol 147:352–phenomenon. Br J Rheumatol 31:375–380, 1992

LeRoy EC, Smith EA, Kahaleh B, Trojanowska M, Silver RM: A strategy for 361, 1995