of August 19, 2016.This information is current as

and Reperfusion InjurySusceptibility of Steatotic Livers to Ischemia A Role for Complement in the Enhanced

and Stephen TomlinsonMark Southwood, Andrew Elvington, Kenneth D. Chavin Songqing He, Carl Atkinson, Zachary Evans, Justin D. Ellett,

http://www.jimmunol.org/content/183/7/4764doi: 10.4049/jimmunol.0900550September 2009;

2009; 183:4764-4772; Prepublished online 14J Immunol 

Referenceshttp://www.jimmunol.org/content/183/7/4764.full#ref-list-1

, 7 of which you can access for free at: cites 48 articlesThis article

Subscriptionshttp://jimmunol.org/subscriptions

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/ji/copyright.htmlSubmit copyright permission requests at:

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

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists, Inc. All rights reserved.Copyright © 2009 by The American Association of9650 Rockville Pike, Bethesda, MD 20814-3994.The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on August 19, 2016

http://ww

w.jim

munol.org/

Dow

nloaded from

by guest on August 19, 2016

http://ww

w.jim

munol.org/

Dow

nloaded from

A Role for Complement in the Enhanced Susceptibility ofSteatotic Livers to Ischemia and Reperfusion Injury1

Songqing He,2*‡ Carl Atkinson,2* Zachary Evans,*† Justin D. Ellett,*† Mark Southwood,§

Andrew Elvington,* Kenneth D. Chavin,*† and Stephen Tomlinson3*

Hepatic steatosis typically renders the donor organ unusable, as donor organs with >30% steatosis are more likely to developgraft failure. The mechanisms leading to failure are not well defined, but steatosis enhances hepatic susceptibility to ischemiareperfusion injury (IRI). We investigated the role of complement in hepatic IRI in lean and steatotic (diet-induced) mice.Steatotic mice were significantly more susceptible to total warm hepatic IRI than lean mice as determined by serum alanineaminotransferase, histopathologically assessed damage, and 24-h survival. C3 deficiency protected both lean and steatoticmice from IRI, as determined by all measured outcomes. Furthermore, treatment of wild-type mice with the complementinhibitor CR2-Crry provided protection equivalent to that seen in C3-deficient mice. Importantly, although steatotic liverswere much more susceptible to IRI than lean livers, by most measures there was no statistical difference between the levelof IRI to steatotic or lean livers when complement was inhibited. To investigate the clinical relevance of these findings in thecontext of transplantation, we treated recipients of lean or steatotic liver grafts with saline or CR2-Crry. There was a markedreduction in graft inflammation and injury and significantly improved 7-day survival in CR2-Crry-treated recipients ofeither lean or steatotic grafts. These data indicate that complement plays a key role in the enhanced susceptibility of steatoticlivers to IRI and suggest that complement inhibition represents a potential strategy to reduce the donor shortage by allowingthe more routine use of marginal steatotic donor livers. The Journal of Immunology, 2009, 183: 4764 – 4772.

L iver transplantation has become an accepted therapy forend-stage liver disease and although the number of pa-tients requiring liver transplants has risen significantly

in recent years, the number of available donors has decreased.This shortage of suitable donor livers has led to the increasinguse of so-called “marginal donors” (1–3). Transplantation of“marginal” donor livers is associated with an increased risk ofprimary nonfunction, increased retransplantation rates, and in-creased recipient morbidity and mortality (4, 5). Liver steatosis(hepatic lipid accumulation) is a primary factor in determiningthe suitability of donor livers for transplantation, since donorliver steatosis is associated with a significantly increased risk ofprimary graft nonfunction (6 – 8). Currently, donor livers with�30% fat content are considered to have a prohibitively highrate of primary nonfunction and �25% of donor livers are re-jected due to steatosis. The pathogenic mechanisms for graftfailure following transplantation of steatotic livers are incom-

pletely understood, but there is evidence that steatotic livers aremore susceptible to ischemia reperfusion injury (IRI)4 (6 – 8).

IRI is a major and unavoidable clinical problem associated withliver transplantation, whether donor livers are lean or steatotic.Activation of complement has been shown to occur in a number oforgan systems following ischemia and reperfusion (I/R), and com-plement inhibition can be protective in animal models. With regardto hepatic IRI, studies using rat models have shown that comple-ment depletion with cobra venom factor or complement inhibitionwith soluble CR, significantly reduce Kupffer cell activation, ox-idative stress, neutrophil accumulation, microvascular dysfunc-tion, and injury (9, 10). C1 inhibitor has also been shown to reducehepatic IRI in a rat model, indicating a role for the classical and/orlectin pathway of complement activation (10–13). In the singlereported study that uses a mouse model of hepatic IRI to investi-gate the role of complement, the classical/lectin pathway is simi-larly implicated since transgenic mice overexpressing C1 inhibitordisplay reduced endothelial injury following hepatic I/R (14). Arole for C5a in hepatic IRI is indicated by the finding that C5areceptor antagonist treatment attenuates tissue injury followingpartial hepatic I/R in rats (15). It is important to note that C3a andC5a also play important roles in liver regeneration, either throughdirect signaling or via their effects on cytokine production, andthese C3 and C5 activation products may also have protective ef-fects by promoting cellular regeneration and tissue repair follow-ing IRI (16–18).

In the current study, we investigated the role of complement inhepatic IRI and the increased susceptibility of steatotic livers toIRI and graft failure. For these studies, we used mice deficient inC3 and wild-type (wt) mice treated with the complement inhibitor

*Department of Microbiology and Immunology, Children’s Research Institute and†Surgery, Division of Transplant, Medical University of South Carolina, Charles-ton, South Carolina 29245; ‡Hepatic Surgery Center, Tongji Hospital, HuazhongUniversity of Science and Technology, Wuhan, China; and §Department of Pa-thology, Papworth Hospital National Health Service Trust, Papworth Everard,Cambridge, United Kingdom

Received for publication February 17, 2009. Accepted for publication August 5, 2009.

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 by National Institutes of Health Grant R01 HL86576 anda scientist development grant from The American Heart Association (to C.A.).2 S.H. and C.A. contributed equally to the preparation of this manuscript.3 Address correspondence and reprint requests to Dr. Stephen Tomlinson, Depart-ment of Microbiology and Immunology, Darby Children’s Research Institute,Medical University of South Carolina, Charleston, SC 29425. E-mail address:tomlinss@musc.edu

4 Abbreviations used in this paper: IRI, ischemia reperfusion injury; ALT, alanineaminotransferase; I/R, ischemia reperfusion; MPO, myeloperoxidase; wt, wild type.

Copyright © 2009 by The American Association of Immunologists, Inc. 0022-1767/09/$2.00

The Journal of Immunology

www.jimmunol.org/cgi/doi/10.4049/jimmunol.0900550

by guest on August 19, 2016

http://ww

w.jim

munol.org/

Dow

nloaded from

CR2-Crry. CR2-Crry is a murine C3 convertase inhibitor thatblocks all complement activation pathways and that targets to sitesof complement activation. The targeting and the complement in-hibitory activities of CR2-Crry have been previously characterized(19, 20). Initial characterizations were performed using a model ofwarm total hepatic I/R, a procedure similar to the Pringle maneu-ver that is used to manage a variety of severe intraoperative com-plications. Complement deficiency and complement inhibitionprotected both lean and steatotic livers from warm IRI. To extendthe study and place the findings in a clinically relevant context oftransplantation, lean and steatotic livers were orthotopically trans-planted and the therapeutic effect of complement inhibition in re-cipient mice was assessed.

Materials and MethodsMice

Male C57BL/6 mice (The Jackson Laboratory) and C57BL/6 C3�/� mice(in-house breeding colony) were housed in temperature- and light-con-trolled chambers on a 12-h light/dark cycle and provided with rodent chow(Purina 5001) and water ad libitum. At 4 wk of age, mice were eithercontinued on normal diet or placed on a high-fat diet (34 g% fat, 60% kcalfat; Research Diets) for 4 wk to induce hepatic steatosis. Mice described asbaseline did not undergo any surgery and were used for resting statemeasurements.

Surgical models

At 8 wk of age, following dietary treatment, wt lean, wt fat feed, C3�/�

lean, and C3�/� fat feed mice were subjected to total warm hepatic isch-emia as previously described (21). Ischemia was performed for 45 min andsurviving mice were sacrificed 24 h after reperfusion. Mouse orthotopicliver transplantation was performed as previously described (22). For trans-plant model, 8- to 10-wk-old male lean or steatotic size-matched C57BL/6mice were used. Time for the donor procedure averaged 30 min. The re-cipient operation averaged 45 min, and the portal vein clamp time averaged15 min. Donor livers were stored for 3 h in cold saline before transplan-tation. Grafts were harvested for analysis at 6 and 24 h after transplant.Survival was monitored for 7 days after transplant, at which time the an-imals were sacrificed. There was no significant difference in operation timeor ischemic times between groups. For complement inhibitor studies, wtrecipients of lean or steatotic grafts were treated with either 0.25 mg ofCR2-Crry or saline (vehicle control) by i.v. injection. All procedures wereapproved by the Medical University of South Carolina Committee for An-imal Research in accordance with the National Institutes of Health Guidefor Care and Use of Laboratory Animals. The CR2-Crry fusion protein wasproduced and purified as described previously (19).

Biochemical and immunological assays

Whole blood was collected from the right ventricle after thoractomy underanesthesia. Blood was allowed to clot at room temperature for 15 min andthen centrifuged at 3500 � g for 5 min at room temperature to collectserum. Serum levels of alanine aminotransferase (ALT) were determinedusing an autoanalyzer (Clinical Laboratory Services, Medical University ofSouth Carolina) or using an analytical kit from Sigma-Aldrich according tothe manufacturers’ instructions. Serum levels of TNF-� and IL-6 weremeasured by ELISA using kits from eBioscience. For quantitative assess-ment of neutrophil infiltration into the liver parenchyma, liver myeloper-oxidase (MPO) content was assessed using the Hbt mouse MPO ELISA Kitfrom Hycult Biotechnology according to the manufacturer’s instructions.

Histology

H&E staining of liver sections was performed as previously described (21).Sections stained with H&E were graded in a blinded fashion for centri-

lobular necrosis on a 0–3 scale as described by Neil and Hubscher. (23).Briefly, a grade of 0 indicated absent necrosis, 1 indicated individual he-patocyte dropout, 2 indicated small foci of missing hepatocytes up to twoto three cells across, and 3 indicated confluent foci of hepatocyte dropoutgreater than three cells across. Ten high-powered fields per section wereanalyzed in relation to the central vein. Necrotic indexing was performedin a blinded fashion.

Complement activation

Complement activation was assessed in liver samples after IRI by immu-nohistochemical localization of C3d and by Western blot analysis of thecomplement activation fragment C3a. For immunohistochemistry, paraf-fin-embedded sections of livers were stained for complement (C3d; Dako-Cytomation) as previously described (19). For Western blot analysis, liversamples were homogenized and protein was extracted as previously de-scribed (24). Protein samples were separated by electrophoresis on a4–20% gradient SDS-PAGE gel. Protein concentration was confirmed us-ing the Bradley method (24), and equal amounts of total protein were addedto each lane. To further verify that protein loading was equal, membraneswere additionally probed with GAPDH. Following transfer to nitrocellu-lose membranes, blots were blocked with TBS-Tween 20 containing 5%dry milk for 1 h at room temperature. Blots were washed three times withTBS-Tween 20 and then incubated with anti-C3a-biotinylated (BD Bio-sciences) or GADPH (Santa Cruz Biotechnology) primary Ab in TBS-Tween 20 for 2 h at room temperature. C3a blots were washed three timeswith TBS-Tween 20 and incubated with streptavidin-biotin-HRP second-ary Ab (BD Biosciences) for 1 h before visualization with ECL substratereagent (Bio-Rad). GADPH blots were washed three times with TBS-Tween 20 and incubated with a secondary Ab (Santa Cruz Biotechnology)conjugated with HRP for 1 h and then visualized with ECL substrate re-agent. Quantification of Western blots was performed using a Bio-RadMolecular Imager Gel Doc.

Liver steatosis

The degree of liver steatosis was quantified by Oil Red O staining and bymeasurement of total liver triglycerides. Oil Red O staining was performedas previously described (25) and the percentage of steatosis was calculatedusing computerized image analysis with the Leica Microsystems QwinImage analysis software. Total liver triglyceride levels were measured us-ing a triglyceride reagent kit according to the manufacturer’s instructions(Pointe Scientific).

Statistical analysis

Data are expressed as mean � SD. Statistical analysis was conducted usingSPSS software (version 6.0). Significant differences between groups weredetermined by ANOVA, and p � 0.05 was considered significant. Allhistological statistical analyses were conducted using nonparametric two-tail analysis and p � 0.05.

ResultsCharacterization of steatosis

C3 contributes to ethanol-induced steatosis in mice, and C3-defi-cient mice fed an ethanol diet do not develop steatosis (26–28).We therefore characterized the effect of a high fat diet on liversteatosis in wt and C3 �/� mice. Both wt and C3 �/� mice re-sponded comparably to the high fat diet and hepatic steatosis wassimilarly induced in both groups of mice as quantified by bodyweight indices, intrahepatic lipid, triglyceride content of livers, andserum ALT levels (results summarized in Table I). Qualitatively,the pattern and distribution of fat deposits was similar in wt andC3�/� mice fed a high fat diet, with moderate to severe macrove-sicular steatosis and mild microvesicular steatosis, as demonstrated

Table I. Strain-specific effects of high fat dieta

Genotype Body Weight Liver:Body Weight Ratio (%) Triglyceride (mg/g protein) Oil Red O (%)

C57BL/6 24.6 � 0.5 4.37 � 0.21 42.8 � 6.5 52.25 � 1.43C3�/� 25.8 � 0.6 4.68 � 0.18 46.9 � 8.5 59.6 � 1.31

a Comparisons of body weight, liver:body weight ratio, total hepatic triglyceride, and fat content by Oil Red O demonstratethat the response of wt and C3�/� animals to a high fat diet is not significantly different.

4765The Journal of Immunology

by guest on August 19, 2016

http://ww

w.jim

munol.org/

Dow

nloaded from

by H&E staining of sections (Fig. 1). Quantification of lipid con-tent by Oil Red O staining and by assay of total hepatic triglyceridewere also similar in fat-fed wt and C3�/� mice (Table I and Fig.1). There was no lipid detectable by Oil Red O staining in lean wtor C3�/� mice and hepatic triglyceride levels were �5% (data notshown). We also measured baseline levels of ALT in wt andC3�/� fat-fed animals, since liver steatosis is associated withraised levels of this enzyme. There was an apparent increase inbaseline ALT levels in fat-fed mice compared with lean mice, butthe increase was not statistically significant ( p � 0.05). Finally,body weight analysis revealed no significant difference in weightgained by either wt or C3�/� mice given a high fat diet, and liverbody weight ratios between wt and C3�/� mice were also notsignificantly different (Table I).

Warm hepatic I/R model and experimental groups

Animals were subjected to 45 min of total warm hepatic ischemiafollowed by 24 h of reperfusion, at which time survival, liver in-jury, and local inflammation were assessed. In all, six experimentalgroups were evaluated: lean and steatotic (fat-fed) wt mice, leanand steatotic C3�/� mice, and lean and steatotic wt mice treatedwith the complement inhibitor CR2-Crry (0.25 mg i.p.,10 min afterreperfusion). A model of total ischemia was used because of itsrelevance to liver transplantation.

Survival and liver function following I/R

All lean mice (wt, C3�/�, and CR2-Crry treated) survived for 24 hfollowing hepatic I/R. In contrast, only 33% (5 of 15) of wt stea-totic mice survived 24 h following I/R. However, C3 deficiencyconferred complete protection to steatotic mice in terms of survival(10 of 10 survived) and 87.5% (seven of eight) of CR2-Crry-treated steatotic mice survived (Fig. 2). Increased serum ALT lev-els that occur following hepatic I/R are indicative of liver injury,and ALT levels were analyzed 24 h after reperfusion in survivinganimals from all six groups. There was no significant differencebetween baseline levels of ALT (preischemic) in steatotic wt, stea-totic C3�/�, lean wt, and lean C3�/� mice ( p � 0.05). However,postreperfusion levels of ALT in wt lean mice were significantlyincreased over baseline (Fig. 3). Levels of ALT were also in-creased above baseline in C3�/� and CR2-Crry-treated lean mice,but the increase was significantly less than that seen in untreated

wt mice (Fig. 3). Postreperfusion levels of ALT were significantlyhigher in all steatotic animals compared with their lean counter-parts. However, as with lean mice, complement deficiency andcomplement inhibition in steatotic mice resulted in significantlyreduced ALT levels when compared with increases seen in stea-totic untreated wt mice. Although C3 deficiency/inhibition signif-icantly reduced ALT levels in steatotic mice compared with theirwt counterparts, levels were still significantly higher than thoseseen in wt lean animals. All ALT determinations were made at24 h after reperfusion to study survival in the same animal groups.Although the relative differences in ALT levels is the significantfinding here, it is of note that the actual levels are lower than wouldbe expected at 6 h when ALT levels peak in this model. Never-theless, the ALT levels at 24 h in the model used here are close topreviously reported ranges (21).

FIGURE 1. Assessment of hepatic steatosis in wt and C3-deficient micefed a high fat diet. H&E (upper panels) and Oil Red O staining (lowerpanels) in liver sections isolated from fat-fed wt and C3�/� mice. No fataccumulation was observed in lean animals from either group (data notshown). Figures are representative of five mice in each experimental group.

FIGURE 2. Effect of C3 deficiency and complement inhibition on sur-vival of mice following hepatic ischemia and reperfusion. Hepatic steatosisis associated with a significant reduction in post-I/R survival, and C3 de-ficiency and CR2-Crry treatment of wt mice significantly improves sur-vival (p � 0.02 wt for steatotic vs wt steatotic plus CR2-Crry, p � 0.05 forwt steatotic vs C3�/� steatotic) (n � 8–15).

FIGURE 3. Effect of C3 deficiency and complement inhibition on ALTlevels following hepatic ischemia and reperfusion. Serum ALT levels at24 h after reperfusion are shown, and ALT levels were significantly in-creased above baseline in all groups (p � 0.05). Average baseline ALTlevels (preischemic) were between 21 and 26.5 U/L for the different groups(see text). Compared with untreated wt mice, ALT levels were significantlylower in both lean (#, p � 0.05) and steatotic (��, p � 0.05) mice that werecomplement deficient (C3�/�) or complement inhibited (CR2-Crry).Mean � SD, n � 5–15/group.

4766 COMPLEMENT-DEPENDENT IRI TO STEATOTIC LIVERS

by guest on August 19, 2016

http://ww

w.jim

munol.org/

Dow

nloaded from

Histopathology following I/R

There was centrilobular necrosis in all lean groups at 24 h afterreperfusion, but complement deficiency and complement inhibi-tion with CR2-Crry was associated with a significantly lower in-jury score when compared with wt lean animals (Fig. 4A). Therewas also centrilobular necrosis in all steatotic groups, but againcomplement deficiency and complement inhibition were associ-ated with a significantly lower injury score when compared withwt animals (Fig. 4A). Wild-type steatotic livers were significantlymore susceptible to IRI than wt lean livers as assessed histologi-cally. Of note, however unlike ALT levels, there was no significantdifference between the extent of damage seen in lean and steatoticlivers when complement was deficient or inhibited. Representativeimages of H&E-stained sections from steatotic livers after reper-fusion are shown in Fig. 4, B–D. Liver samples from normal (pre-ischemic) mice showed no evidence of centrilobular necrosis (datanot shown).

Complement deposition and activation following I/R

The deposition and distribution of C3 was assessed in lean andsteatotic wt mice 24 h after reperfusion by immunohistochemistry.C3d was deposited on hepatocyte membranes, sinusoidal endothe-lium, and Kupffer cells within livers that had undergone I/R (Fig.5, A and B). Staining for C3d was more intense in centrilobularareas, corresponding to areas of increased liver damage as deter-mined by H&E staining (above). The distribution of C3d was sim-ilar in steatotic and lean livers, but C3d deposition in centrilobular

areas of steatotic livers appeared to involve more hepatocytes andto spread deeper into the liver away from the centrilobular area.However, although C3d staining of viable cells appeared to begreater in steatotic compared with lean livers, this finding shouldbe interpreted with caution. Staining of lean and steatotic liverswith isotype control Abs, as well as staining in C3�/� animals,resulted in some nonspecific staining, and nonspecific staining wasmore intense in steatotic livers. Therefore, we cannot rule out thepossibility that the increase in C3d staining intensity seen in stea-totic livers was due not only to an increase in centrilobular dam-age, but was also associated with an increase in nonspecific stain-ing associated with fat deposition. We therefore assessed whethersteatosis was associated with increased complement activation byassay of the C3 activation fragment C3a in liver homogenates.Samples from lean and steatotic livers isolated from both control-treated and CR2-Crry-treated mice were analyzed by Western blotand densitometry. C3a was barely detectable in liver samples fromsham-operated mice, but was readily detected in postreperfusionsamples from lean and steatotic livers, with significantly higherlevels in steatotic livers (Fig. 5, C and D). Furthermore, treatmentof either lean or steatotic animals with CR2-Crry resulted in asignificant reduction in C3a levels compared with their PBS-treated counterparts, a result that correlates with decreased C3ddeposition seen in CR2-Crry-treated mice (Fig. 5, C and D). AllC3a semiquantitative densitometry results are normalized toGADPH density.

Orthotopic liver transplantation model and experimental groups

The foregoing data demonstrate that complement plays an impor-tant role in hepatic injury following warm I/R and that complementcontributes to the enhanced susceptibility of steatotic livers towarm I/R. To extend these studies and investigate the role of com-plement in IRI in a clinical setting of transplantation, we investi-gated the effect of complement inhibition in recipients of lean andsteatotic liver grafts. Following transplantation of livers, lean re-cipient mice were treated with either saline or CR2-Crry (0.25 mg)immediately after implantation of liver. Grafts were isolated ateither 6 or 24 h after transplantation and analyzed for liver injuryand inflammation. Survival of recipient mice was also monitoredfor 7 days. In all, four experimental groups were evaluated: leangraft-to-lean recipient, saline treated; lean-to-lean, CR2-Crry treat-ed; steatotic-to-lean, saline treated; and steatotic-to-lean, CR2-Crry treated.

Graft injury following liver transplantation

Following transplantation, there was a significant increase in re-cipient serum ALT levels in all groups compared with baseline,and recipients of steatotic grafts had significantly higher levels ofALT than recipients of lean grafts (Fig. 6A). However, comparedwith saline-treated controls, CR2-Crry treatment resulted in a sig-nificant reduction of ALT levels in recipients of both lean andsteatotic grafts, both at 6 and 24 h after transplantation (Fig. 6A).Of note, complement inhibition in recipients of steatotic graftsreduced serum ALT to levels at or below those seen in untreatedrecipients of lean grafts.

In keeping with our observations in the IRI model, centrilobularhepatocyte necrosis was evident in both lean and steatotic grafts at6 and 24 h after transplantation. However, compared with the IRImodel, the degree of hepatic damage was increased in the trans-plant model, with necrosis expanding beyond the centrilobular ar-eas. Quantitative assessment of the histological damage demon-strated a marked protective effect associated with complementinhibition; there was excessive hepatic necrosis in both lean andsteatotic grafts isolated from saline-treated recipients, but not in

FIGURE 4. Histological assessment of hepatocellular necrosis follow-ing hepatic ischemia and reperfusion. A, Centrilobular necrosis at 24 h afterreperfusion scored on an injury scale of 0–3 (see Materials and Methods).Injury scores were significantly lower in C3�/� mice and CR2-Crry-treatedmice compared with wt mice for both lean (�, p � 0.05) and steatotic (#,p � 0.05) groups. There was no significant difference in injury scoresbetween C3-deficient or CR2-Crry-treated mice, whether lean or steatotic(p � 0.05). Mean � SD, n � 5–15/group. B–D, Representative H&Eimages of steatotic liver sections from wt (B), C3�/� (C), and CR2-Crry-treated animals (D). Notice the disruption of endothelial cells and presenceof necrotic hepatocytes in wt steatotic liver (B, arrows). In C3�/� mice andCR2-Crry-treated mice, hepatocytes appear well preserved without any ob-vious disruption of the endothelium or hepatocyte morphology. Scale bar,20 �m.

4767The Journal of Immunology

by guest on August 19, 2016

http://ww

w.jim

munol.org/

Dow

nloaded from

FIGURE 5. Complement activa-tion assessed by immunohistochem-istry for C3d and by Western blotanalysis for C3a. A, Immunohisto-chemistry staining of livers from leanmice. Note the intense C3d deposition(brown) seen associated with hepato-cytes and endothelial cells within acentrilobular region in wt lean ani-mals, with specific staining absent inlivers from C3�/� mice and signifi-cantly reduced in wt mice treated withCR2-Crry. B, Immunohistochemistrystaining of livers from steatotic mice.C3d deposition is similarly localizedto centrilobular areas in wt steatoticmice and is associated with hepato-cyte damage. In steatotic C3 �/�

mice, a low level of nonspecific C3dstaining was associated with fat accu-mulation. Staining is reduced in CR2-Crry-treated steatotic mice comparedwith wt mice. Images representative,n � 5, and magnification indicated byscale bars. C, Western blot analysis ofC3a and GADPH in liver homoge-nates 24 h after I/R. Shown is a rep-resentative blot and densitometricanalysis (mean intensity per mm2 �SD, n � 3). C3a levels were higher inwt steatotic livers compared with leanlivers, and CR2-Crry treatment signif-icantly reduced C3a levels in bothlean and steatotic mice. Results areexpressed as intensity per mm2 nor-malized to GADPH expression. #,p � 0.05 vs wt counterparts and �,p � 0.05 vs lean. Mean � SD, n � 3.

4768 COMPLEMENT-DEPENDENT IRI TO STEATOTIC LIVERS

by guest on August 19, 2016

http://ww

w.jim

munol.org/

Dow

nloaded from

grafts isolated from CR2-Crry-treated recipients (Fig. 6B). Of note,CR2-Crry afforded a similar level of protection to both lean andsteatotic grafts.

Inflammation in graft following liver transplantation

Complement activation products are known to play important rolesin leukocyte recruitment, vascular adherence, and cytokine pro-duction, and a role for neutrophils and cytokines (particularlyTNF-� and IL-6) in hepatic IRI is well documented (29–32). Toassess the effect of complement activation on neutrophil infiltra-tion, MPO levels in liver homogenates were determined. Levels ofMPO were elevated in all posttransplant samples compared with

baseline and sham-operated controls but, in correlation with mea-surements of hepatic injury, complement inhibition reduced MPOlevels in recipients of both lean and steatotic grafts (Fig. 7). TNF-�and IL-6 was undetectable in lean and steatotic sham controls, butboth cytokines were elevated in all recipient mice at 6 and 24 hafter transplant, and cytokine levels were higher in recipients ofsteatotic grafts compared with recipients of lean grafts. Comple-ment inhibition, however, significantly reduced the levels of bothcytokines when compared with their lean and steatotic counter-parts at both time points (Fig. 8). Of note, in saline-treated recip-ients of lean grafts, MPO and cytokine levels at 24 h after trans-plant were similar or decreased compared with levels at 6 h aftertransplant. However, in saline-treated recipients of steatotic grafts,MPO and cytokine levels were increased at 24 h after transplant,indicating exacerbated and prolonged inflammation in the steatoticgrafts. Overall, CR2-Crry treatment reduced the inflammatory pro-file in recipients of steatotic grafts to a level similar to that insaline-treated recipients of lean grafts.

Immunohistochemistry revealed C3d deposition in a similar dis-tribution to that seen in the warm IRI model, with C3d depositedon hepatocytes, sinusoidal endothelium, and Kupffer cells. Thestaining intensity was more prominent in steatotic livers than inlean livers and was undetectable in livers from CR2-Crry-treatedmice at both 6 and 24 h after transplantation (data not shown).

Recipient survival following liver transplantation

Clinical data has shown that steatotic grafts are associated with anincreased risk of primary nonfunction and increased recipient mor-bidity and mortality. In agreement with clinical findings, steatoticlivers were more susceptible to graft failure than lean livers, andthis was regardless of whether recipient mice were treated withCR2-Crry (Fig. 9). However, complement inhibition in recipientsof both lean and steatotic grafts was associated with a significantimprovement in posttransplant survival when compared with sa-line-treated counterparts. These data show that the level of com-plement-dependent inflammation and injury observed in lean and

FIGURE 6. Effect of CR2-Crry treatment on liver injury in recipients oflean and steatotic grafts. A, ALT levels at 6 and 24 h after transplantation.Recipients of either lean or steatotic donor livers were treated with salineor 0.25 mg of CR2-Crry immediately after surgery. ALT levels were sig-nificantly increased in all steatotic groups compared with lean counterparts(##, p � 0.05). Compared with saline treatment, CR2-Crry treatment sig-nificantly reduced ALT levels in recipients of either lean or steatotic grafts(�, p � 0.05). Mean � SD, n � 6 in each group. B, Histological injuryscore at 6 and 24 h after transplantation. There is increased injury to stea-totic grafts compared with lean grafts in saline-treated recipients at 24 hafter transplant (##, p � 0.05), but not at 6 h after transplant. CR2-Crrytreatment reduced injury to both lean and steatotic grafts at both time pointsof analysis (�, p � 0.05), and there was no significant difference betweeninjury scores for lean and steatotic grafts at either time point. There was noevidence of damage in baseline donor animals. Mean � SD, n � 6 in eachgroup.

FIGURE 7. Neutrophil infiltration of grafts. Lean or steatotic graftswere isolated from saline- or CR2-Crry-treated recipients at 6 or 24 h aftertransplantation. Neutrophil activity in liver homogenates was estimated byassessing MPO activity. MPO levels in steatotic grafts were increased com-pared with lean grafts (#, p � 0.05). Treatment of recipients with CR2-Crryresulted in a significant reduction in MPO levels in both lean and steatoticgrafts compared with saline-treated recipients (�, p � 0.05). Mean � SD,n � 6 in each group.

4769The Journal of Immunology

by guest on August 19, 2016

http://ww

w.jim

munol.org/

Dow

nloaded from

steatotic grafts following transplantation (as shown above) corre-lates with graft failure and support the concept that complementinhibition may represent a strategy to permit expanded use of mar-ginal steatotic livers.

DiscussionIschemia and reperfusion of a donor organ following transplanta-tion results in an inflammatory response that causes acute injuryand may also increase graft immunogenicity and alloresponsive-ness of the recipient (33, 34). Steatosis significantly increases thesusceptibility of livers to IRI. The pathogenic mechanisms in-volved in IRI are complex and multifaceted, but studies using an-imal models have shown that activation of complement is a keyinitiating event in IRI of various organs and tissues (35–38). In our

initial studies, we demonstrated that C3 deficiency or complementinhibition at the C3 activation step protects against hepatic IRI ina mouse model of warm I/R. Furthermore, using the same modelwith fat-fed mice, we showed that the enhanced susceptibility ofsteatotic livers to IRI was reversed by complement deficiency orinhibition. We used a model of total hepatic ischemia because ofits relevance to liver transplantation, although we have also shownthat complement deficiency and inhibition protects against IRI in amodel of partial hepatic ischemia that does not carry the risk ofvenous portal congestion and endotoxemia (our unpublished data).

Thus, data obtained using the total hepatic IRI model suggeststhat complement inhibitor therapy may represent a strategy to in-crease the donor pool, since �25% of donor livers are currentlyrejected due to steatosis (livers with �30% steatosis are usuallyrejected as donor organs). Therefore, to put our findings in a moreclinically relevant context, we investigated the effect of comple-ment inhibition in a model of orthotopic liver transplantation inwhich lean recipients received either lean or steatotic grafts. Therewas increased injury and increased and prolonged inflammation insteatotic grafts compared with lean grafts, and in keeping with thebulk of clinical data, steatosis was associated with increased graftfailure. CR2-Crry administered to recipients protected both leanand steatotic grafts from injury and graft failure, and the protectiveeffect of complement inhibition correlated with a reduced inflam-matory profile.

Since the extent of steatosis is positively associated with theextent of IRI (7, 8, 39, 40), and the C3 activation productC3adesArg (also known as acylation- stimulating protein) is in-volved in the regulation of lipid metabolism (41, 42), a possibleexplanation for the reduced level of hepatic IRI seen in C3�/�

mice is that C3�/� mice develop less liver steatosis. However,following a 4-wk high fat diet (60% calories from fat), we detectedno significant quantitative or qualitative differences in steatosisbetween wt and C3�/� mice. Also, the fact that complement in-hibition in wt steatotic mice is protective in the IRI model furtherargues against any inherent increased susceptibility of C3�/� miceto the high fat diet. It is of note, nevertheless, that C3�/� mice donot develop ethanol-induced hepatic steatosis (26–28), and therole of complement in the development of steatosis can thus bedependent on the inducing factor. There is also a report that a highfat diet does not produce steatosis in wt (C3�/�) C57BL/6 mice,but a different diet was used (28). Indeed, numerous previous stud-ies have shown that the high fat diet used in the current studyresults in hepatic steatosis in C57BL/6 mice. We show that mac-rovesicular steatosis predominates in both wt and C3�/� fat-fedmice, and although not completely without controversy, macrove-sicular steatosis rather than microvesicular steatosis is considereda risk factor for primary graft dysfunction following transplanta-tion (5, 39, 43). Livers with macrovesicular steatosis have also

FIGURE 8. Effect of complement inhibition on expression of inflam-matory cytokines. A, IL-6 serum levels at 6 and 24 h after transplantation.IL-6 is significantly increased in recipients of steatotic livers comparedwith recipients of lean livers (##, p � 0.05). Compared with saline, CR2-Crry treatment significantly reduced IL-6 levels in both lean (�, p � 0.001and 0.001 at 6 and 24 h, respectively) and steatotic (�, p � 0.008 and 0.001at 6 and 24 h, respectively) grafts. Mean � SD, n � 6 in each group. B,TNF-� serum levels at 6 and 24 h after transplantation. TNF-� is signif-icantly increased in recipients of steatotic livers compared with recipientsof lean livers (##, p � 0.05). Compared with saline, CR2-Crry treatmentsignificantly reduced TNF-� levels in both lean (�, p � 0.007 and 0.004 at6 and 24 h, respectively) and steatotic (�, p � 0.007 and 0.001 at 6 and24 h, respectively) mice. Mean � SD, n � 6 in each group.

FIGURE 9. Graft survival. recipients of lean or steatotic grafts weretreated with saline or CR2-Crry immediately after implantation. CR2-Crrytreatment significantly improved 7-day survival for recipients of both leanand steatotic grafts (p � 0.0023 for lean CR2-Crry vs saline, and p � 0.047for steatotic CR2-Crry vs saline). n � 8 in each group.

4770 COMPLEMENT-DEPENDENT IRI TO STEATOTIC LIVERS

by guest on August 19, 2016

http://ww

w.jim

munol.org/

Dow

nloaded from

been shown to be more susceptible to injury than livers with mi-crovesicular steatosis in a mouse model of hepatic IRI, a findingthat correlated with significantly lower hepatic perfusion and por-tal vein flow in macrosteatotic livers (40). Macrovesicular steato-sis, as induced by diet and insulin, is more common within thedonor population, whereas, microvesicular is more commonly as-sociated with drugs, toxins, and metabolic disorder (44).

Previous studies have demonstrated the presence of complementin the ischemically injured liver, with complement deposition onhepatocytes and sinusoidal endothelial cells, cells that are reportedto express low levels of complement regulatory proteins (12, 45).We similarly demonstrated the presence of C3d on hepatocytesand the sinusoidal endothelium in wt lean and steatotic mice afterI/R and after transplantation. The pattern of staining was similar inboth lean and steatotic livers, although the frequency and intensityof complement deposition appeared to be increased in steatoticliver samples. The increased complement deposition is likely re-lated to the increased damage seen in steatotic livers after reper-fusion. We did not attempt to quantify the increase in stainingintensity in steatotic livers, since there was an increase in nonspe-cific staining in these livers as evidenced by staining for C3d inC3�/� and CR2-Crry-treated animals. Furthermore, previous stud-ies have reported that steatotic livers produce increased levels ofcomplement proteins de novo (46), which may also contribute tothe apparent increase in C3d immunostaining. Nevertheless, CR2-Crry treatment resulted in a significant reduction in immunostain-ing for C3d in livers from both lean and steatotic mice after I/R andposttransplantation. Taken together, these data suggest that steato-sis results in exacerbated complement activation and may contrib-ute to an increased susceptibility to IRI. We therefore quantifiedcomplement activation in lean and steatotic livers by measuringthe complement activation fragment C3a. Following IRI, steatoticlivers were associated with significantly higher hepatic concentra-tions of C3a than lean livers.

Steatotic livers were similarly more susceptible to injury thanlean livers in the transplant and warm ischemia models, as deter-mined by ALT levels and histological injury score. In the trans-plant model, we analyzed two time points after IRI (6 and 24 h)and demonstrated that ALT levels were significantly elevated atboth time points in both lean and steatotic livers compared withsham controls, although ALT levels were reduced at 24 h com-pared with 6 h. This trend was not seen for histological injury scoreand there was evidence of elevated necrosis in steatotic grafts at24 h compared with 6 h. It is difficult to reconcile these oppositetrends seen for ALT levels and histological injury. It is possiblethat this may be related to the dynamics of injury, with cellularinjury that leads to the release of ALT occurring early after IRI,and histological evidence of hepatic necrosis occurring progres-sively, a process that may be delayed in with hepatic steatosis. Inthis regard, hepatic steatosis has been shown to affect the meta-bolic functions of the liver and it is possible that metabolic con-straints delay the appearance of histologically detectable necrosis.

Data from the transplantation model show that the increasedfailure of steatotic grafts compared with lean grafts is associatedwith a significantly higher inflammatory burden (C3a, MPO,TNF-�, and IL-6 levels). Furthermore, the improved survival rateof recipients treated with CR2-Crry was associated with a signif-icant reduction in inflammatory burden, whether the transplantedgraft was lean or steatotic. Together with the activation of Kupffercells and neutrophil infiltration, the inflammatory cytokines TNF-�and IL-6 have been implicated in mediating hepatic injury follow-ing I/R via various mechanisms (for reviews, see Refs. 47–49).However, TNF-� and IL-6 also play key roles in hepatocyte pro-liferation and thus appear to have dual functions by participating in

both inflammation/injury and regeneration. Complement activa-tion products C3a and C5a have been shown to play an essentialrole in the early priming phase of hepatocyte regeneration via in-duction of TNF-� and IL-6 expression and subsequent NF-�B andSTAT-3 activation (17). Indeed, complement deficiency or com-plement inhibition significantly impairs liver regeneration and in-creases mortality in models of partial hepatectomy or toxic injury(16, 17). It is not clear to what extent the regenerative response isnecessary for cellular repair and recovery following IRI, althoughstudies in transplant models indicate that the regenerative responseand expression of TNF-� and IL-6 become more important to graftrecovery as cold storage time is increased (16, 17). However, in thecurrent study, in a setting of full-size liver transplantation (and 3 hcold storage), complement inhibition has clear benefits regardinggraft function and survival. Although the cytokine data is onlycorrelative, a contributing protective mechanism of complementinhibition may be the modulation of TNF-�- and IL-6-dependentpathways of hepatocellular injury. Regardless, the protective effectof complement appears to outweigh any potential benefit that in-creased expression of the cytokines may have on hepatocyte re-generation and graft recovery (for both lean and steatotic grafts).

In conclusion, the current data demonstrate an important role forcomplement in the enhanced susceptibility of steatotic livers to IRIand suggest that complement inhibition represents a therapy thatmay enable the expanded use of so-called marginal steatotic liversfor transplantation. In this context, the CR2-mediated targetingstrategy used here for complement inhibition not only significantlyincreases the efficacy of the linked complement inhibitor, but alsominimizes immune suppression that is associated with systemiccomplement inhibition (19), a potentially important benefit for atransplant recipient.

AcknowledgmentsWe acknowledge the support of Efrain Martinez for protein purificationand Emily Pauling for assistance with histology.

DisclosuresStephen Tomlinson owns stock in Taligen Therapeutics Inc., a companydeveloping complement inhibitors.

References1. Chiari, P., V. Piriou, G. Hadour, C. Rodriguez, J. Loufouat, J. J. Lehot, M. Ovize,

and R. Ferrera. 2002. Preservation of ischemia and isoflurane-induced precondi-tioning after brain death in rabbit hearts. Am. J. Physiol. 283: H1769–H1774.

2. Samarasinghe, D., and C. Tasman-Jones. 1992. The clinical associations withhepatic steatosis: a retrospective study. N. Z. Med. J. 105: 57–58.

3. Merritt, W. T. 2005. Issues affecting liver transplantation. Best Pract. Res. Clin.Anaesthesiol. 19: 17–34.

4. Ferraz-Neto, B. H., M. P. Zurstrassen, R. Hidalgo, L. E. Fonseca, T. D. Motta,F. L. Pandullo, M. B. Rezende, S. P. Meira-Filho, J. R. Sa, and R. C. Afonso.2007. Donor liver dysfunction: application of a new scoring system to identify themarginal donor. Transplant. Proc. 39: 2516–2518.

5. McCormack, L., H. Petrowsky, W. Jochum, B. Mullhaupt, M. Weber, andP. A. Clavien. 2007. Use of severely steatotic grafts in liver transplantation: amatched case-control study. Ann. Surg. 246: 940–946; discussion 946–948.

6. Birsner, J. H., C. Wan, G. Cheng, Z. P. Evans, C. C. Polito, R. N. Fiorini,G. Gilbert, J. K. Haines, M. G. Schmidt, and K. D. Chavin. 2004. Steatotic livertransplantation in the mouse: a model of primary nonfunction. J. Surg. Res. 120:97–101.

7. Chavin, K. D., R. N. Fiorini, S. Shafizadeh, G. Cheng, C. Wan, Z. Evans,D. Rodwell, C. Polito, J. K. Haines, G. M. Baillie, and M. G. Schmidt. 2004.Fatty acid synthase blockade protects steatotic livers from warm ischemia reper-fusion injury and transplantation. Am. J. Transplant. 4: 1440–1447.

8. Fiorini, R. N., S. F. Shafizadeh, C. Polito, D. W. Rodwell, G. Cheng, Z. Evans,C. Wan, S. Belden, J. K. Haines, J. Birsner, et al. 2004. Anti-endotoxin mono-clonal antibodies are protective against hepatic ischemia/reperfusion injury insteatotic mice. Am. J. Transplant. 4: 1567–1573.

9. Chavez-Cartaya, R. E., G. P. DeSola, L. Wright, N. V. Jamieson, and D. J. White.1995. Regulation of the complement cascade by soluble complement receptortype 1: protective effect in experimental liver ischemia and reperfusion. Trans-plantation 59: 1047–1052.

4771The Journal of Immunology

by guest on August 19, 2016

http://ww

w.jim

munol.org/

Dow

nloaded from

10. Lehmann, T. G., T. A. Koeppel, S. Munch, M. Heger, M. Kirschfink, E. Klar, andS. Post. 2001. Impact of inhibition of complement by sCR1 on hepatic micro-circulation after warm ischemia. Microvasc. Res. 62: 284–292.

11. Lehmann, T. G., M. Heger, S. Munch, M. Kirschfink, and E. Klar. 2000. In vivomicroscopy reveals that complement inhibition by C1-esterase inhibitor reducesischemia/reperfusion injury in the liver. Transplant. Int. 13(Suppl. 1):S547–S550.

12. Heijnen, B. H., I. H. Straatsburg, N. D. Padilla, G. J. Van Mierlo, C. E. Hack, andT. M. Van Gulik. 2006. Inhibition of classical complement activation attenuatesliver ischaemia and reperfusion injury in a rat model. Clin. Exp. Immunol. 143:15–23.

13. Bergamaschini, L., G. Gobbo, S. Gatti, L. Caccamo, P. Prato, M. Maggioni,P. Braidotti, R. Di Stefano, and L. R. Fassati. 2001. Endothelial targeting withC1-inhibitor reduces complement activation in vitro and during ex vivo reperfu-sion of pig liver. Clin. Exp. Immunol. 126: 412–420.

14. Inderbitzin, D., G. Beldi, I. Avital, G. Vinci, and D. Candinas. 2004. Local andremote ischemia-reperfusion injury is mitigated in mice overexpressing humanC1 inhibitor. Eur. Surg. Res. 36: 142–147.

15. Arumugam, T. V., T. M. Woodruff, S. Z. Stocks, L. M. Proctor, S. Pollitt,I. A. Shiels, R. C. Reid, D. P. Fairlie, and S. M. Taylor. 2004. Protective effectof a human C5a receptor antagonist against hepatic ischaemia-reperfusion injuryin rats. J. Hepatol. 40: 934–941.

16. Strey, C. W., M. Markiewski, D. Mastellos, R. Tudoran, L. A. Spruce,L. E. Greenbaum, and J. D. Lambris. 2003. The proinflammatory mediators C3aand C5a are essential for liver regeneration. J. Exp. Med. 198: 913–923.

17. Markiewski, M. M., D. Mastellos, R. Tudoran, R. A. DeAngelis, C. W. Strey,S. Franchini, R. A. Wetsel, A. Erdei, and J. D. Lambris. 2004. C3a and C3bactivation products of the third component of complement (C3) are critical fornormal liver recovery after toxic injury. J. Immunol. 173: 747–754.

18. Yuceturk, H., M. C. Yagmurdur, G. Gur, M. Demirbilek, B. Bilezikci, M. Turan,H. Karakayali, and M. Haberal. 2007. Role of heparin on TNF-� and IL-6 levelsin liver regeneration after partial hepatic resection. Eur. Surg. Res. 39: 216–221.

19. Atkinson, C., H. Song, B. Lu, F. Qiao, T. A. Burns, V. M. Holers, G. C. Tsokos,and S. Tomlinson. 2005. Targeted complement inhibition by C3d recognitionameliorates tissue injury without apparent increase in susceptibility to infection.J. Clin. Invest. 115: 2444–2453.

20. Atkinson, C., H. Zhu, F. Qiao, J. C. Varela, J. Yu, H. Song, M. S. Kindy, andS. Tomlinson. 2006. Complement-dependent P-selectin expression and injuryfollowing ischemic stroke. J. Immunol. 177: 7266–7274.

21. Evans, Z. P., J. D. Ellett, M. G. Schmidt, R. G. Schnellmann, and K. D. Chavin.2008. Mitochondrial uncoupling protein-2 mediates steatotic liver injury follow-ing ischemia/reperfusion. J. Biol. Chem. 283: 8573–8579.

22. Theruvath, T. P., C. Czerny, V. K. Ramshesh, Z. Zhong, K. D. Chavin, andJ. J. Lemasters. 2008. C-Jun N-terminal kinase 2 promotes graft injury via themitochondrial permeability transition after mouse liver transplantation.Am. J. Transplant. 8: 1819–1828.

23. Neil, D. A., and S. G. Hubscher. 2001. Are parenchymal changes in early post-transplant biopsies related to preservation-reperfusion injury or rejection? Trans-plantation 71: 1566–1572.

24. Juan, M. E., M. Maijo, and J. M. Planas. 2009. Quantification of trans-resveratroland its metabolites in rat plasma and tissues by HPLC. J. Pharm. Biomed. Anal.Epub ahead of print.

25. Fiorini, R. N., J. Kirtz, B. Periyasamy, Z. Evans, J. K. Haines, G. Cheng,C. Polito, D. Rodwell, S. F. Shafizadeh, X. Zhou, et al. 2004. Development of anunbiased method for the estimation of liver steatosis. Clin. Transplant. 18:700–706.

26. Pritchard, M. T., M. R. McMullen, A. B. Stavitsky, J. I. Cohen, F. Lin,M. E. Medof, and L. E. Nagy. 2007. Differential contributions of C3, C5, anddecay-accelerating factor to ethanol-induced fatty liver in mice. Gastroenterology132: 1117–1126.

27. Bykov, I., M. Jauhiainen, V. M. Olkkonen, S. T. Saarikoski, C. Ehnholm,S. Junnikkala, A. Vakeva, K. O. Lindros, and S. Meri. 2007. Hepatic gene ex-pression and lipid parameters in complement C3�/� mice that do not developethanol-induced steatosis. J. Hepatol. 46: 907–914.

28. Bykov, I., S. Junnikkala, M. Pekna, K. O. Lindros, and S. Meri. 2006. Comple-ment C3 contributes to ethanol-induced liver steatosis in mice. Ann. Med. 38:280–286.

29. Galloway, E., T. Shin, N. Huber, T. Eismann, S. Kuboki, R. Schuster,J. Blanchard, H. R. Wong, and A. B. Lentsch. 2008. Activation of hepatocytes byextracellular heat shock protein 72. Am. J. Physiol. 295: C514–C520.

30. Langdale, L. A., V. Hoagland, W. Benz, K. J. Riehle, J. S. Campbell,D. H. Liggitt, and N. Fausto. 2008. Suppressor of cytokine signaling expressionwith increasing severity of murine hepatic ischemia-reperfusion injury. J. Hepa-tol. 49: 198–206.

31. Ishizaki, M., M. Kaibori, Y. Uchida, T. Hijikawa, H. Tanaka, T. Ozaki,K. Tokuhara, K. Matsui, A. H. Kwon, Y. Kamiyama, et al. 2008. Protective effectof FR183998, a Na�/H� exchanger inhibitor, and its inhibition of iNOS induc-tion in hepatic ischemia-reperfusion injury in rats. Shock 30: 311–317.

32. Kaizu, T., A. Ikeda, A. Nakao, A. Tsung, H. Toyokawa, S. Ueki, D. A. Geller,and N. Murase. 2008. Protection of transplant-induced hepatic ischemia/reperfu-sion injury with carbon monoxide via MEK/ERK1/2 pathway downregulation.Am. J. Physiol. 294: G236–G244.

33. Sacks, S. H., and W. Zhou. 2003. Locally produced complement and its role inrenal allograft rejection. Am. J. Transplant. 3: 927–932.

34. Fuller, T. F., B. Sattler, L. Binder, F. Vetterlein, B. Ringe, and T. Lorf. 2001.Reduction of severe ischemia/reperfusion injury in rat kidney grafts by a solubleP-selectin glycoprotein ligand. Transplantation 72: 216–222.

35. Chan, R. K., S. I. Ibrahim, N. Verna, M. Carroll, F. D. Moore, Jr., andH. B. Hechtman. 2003. Ischaemia-reperfusion is an event triggered by immunecomplexes and complement. Br. J. Surg. 90: 1470–1478.

36. Chan, R. K., N. Verna, J. Afnan, M. Zhang, S. Ibrahim, M. C. Carroll, andF. D. Moore, Jr. 2006. Attenuation of skeletal muscle reperfusion injury withintravenous 12 amino acid peptides that bind to pathogenic IgM. Surgery 139:236–243.

37. Riedemann, N. C., and P. A. Ward. 2003. Complement in ischemia reperfusioninjury. Am. J. Pathol. 162: 363–367.

38. Zhang, M., W. G. Austen, Jr., I. Chiu, E. M. Alicot, R. Hung, M. Ma, N. Verna,M. Xu, H. B. Hechtman, F. D. Moore, Jr., and M. C. Carroll. 2004. Identificationof a specific self-reactive IgM antibody that initiates intestinal ischemia/reperfu-sion injury. Proc. Natl. Acad. Sci. USA 101: 3886–3891.

39. El-Badry, A. M., W. Moritz, C. Contaldo, Y. Tian, R. Graf, and P. A. Clavien.2007. Prevention of reperfusion injury and microcirculatory failure in macrostea-totic mouse liver by omega-3 fatty acids. Hepatology 45: 855–863.

40. Selzner, N., M. Selzner, W. Jochum, B. Amann-Vesti, R. Graf, and P. A. Clavien.2006. Mouse livers with macrosteatosis are more susceptible to normothermicischemic injury than those with microsteatosis. J. Hepatol. 44: 694–701.

41. Paglialunga, S., A. Fisette, Y. Yan, Y. Deshaies, J. F. Brouillette, M. Pekna, andK. Cianflone. 2008. Acylation-stimulating protein deficiency and altered adiposetissue in alternative complement pathway knockout mice. Am. J. Physiol. Endo-crinol. Metab. 294: E521–E529.

42. Maslowska, M., H. W. Wang, and K. Cianflone. 2005. Novel roles for acylationstimulating protein/C3adesArg: a review of recent in vitro and in vivo evidence.Vitam. Horm. 70: 309–332.

43. El-Badry, A. M., R. Graf, and P. A. Clavien. 2007. Omega 3-omega 6: what isright for the liver? J. Hepatol. 47: 718–725.

44. Vetelainen, R., A. van Vliet, D. J. Gouma, and T. M. van Gulik. 2007. Steatosisas a risk factor in liver surgery. Ann. Surg. 245: 20–30.

45. Scoazec, J. Y., G. Borghi-Scoazec, F. Durand, J. Bernuau, B. N. Pham,J. Belghiti, G. Feldmann, and C. Degott. 1997. Complement activation after isch-emia-reperfusion in human liver allografts: incidence and pathophysiological rel-evance. Gastroenterology 112: 908–918.

46. Gregoire, F. M., Q. Zhang, S. J. Smith, C. Tong, D. Ross, H. Lopez, andD. B. West. 2002. Diet-induced obesity and hepatic gene expression alterationsin C57BL/6J and ICAM-1-deficient mice. Am. J. Physiol. Endocrinol. Metab.282: E703–E713.

47. Okaya, T., and A. B. Lentsch. 2003. Cytokine cascades and the hepatic inflam-matory response to ischemia and reperfusion. J. Invest Surg. 16: 141–147.

48. Serracino-Inglott, F., N. A. Habib, and R. T. Mathie. 2001. Hepatic ischemia-reperfusion injury. Am. J. Surg. 181: 160–166.

49. Teoh, N. C., and G. C. Farrell. 2003. Hepatic ischemia reperfusion injury: patho-genic mechanisms and basis for hepatoprotection. J. Gastroenterol. Hepatol. 18:891–902.

4772 COMPLEMENT-DEPENDENT IRI TO STEATOTIC LIVERS

by guest on August 19, 2016

http://ww

w.jim

munol.org/

Dow

nloaded from