of March 25, 2018.This information is current as

Murine Ischemic StrokePropagates Inflammation and Injury in The Alternative Complement Pathway

TomlinsonTakahashi, Gregory L. Stahl, Mark S. Kindy and Stephen Andrew Elvington, Carl Atkinson, Hong Zhu, Jin Yu, Kazue

http://www.jimmunol.org/content/189/9/4640doi: 10.4049/jimmunol.1201904October 2012;

2012; 189:4640-4647; Prepublished online 1J Immunol 

Referenceshttp://www.jimmunol.org/content/189/9/4640.full#ref-list-1

, 17 of which you can access for free at: cites 53 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. All rights reserved.1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on March 25, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

by guest on March 25, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

The Journal of Immunology

The Alternative Complement Pathway PropagatesInflammation and Injury in Murine Ischemic Stroke

Andrew Elvington,* Carl Atkinson,* Hong Zhu,† Jin Yu,† Kazue Takahashi,‡

Gregory L. Stahl,x Mark S. Kindy,†,{ and Stephen Tomlinson*,{

There is mounting evidence indicating an important role for complement in the pathogenesis of cerebral ischemia-reperfusion

injury, or ischemic stroke. The role of the alternative complement pathway in ischemic stroke has not been investigated, and there

is conflicting data on the role of the terminal pathway. In this study, we show that comparedwith wild-typemice, mice deficient in the

alternative pathway protein factor B or mice treated with the alternative pathway inhibitor CR2-fH have improved outcomes after

60-min middle cerebral artery occlusion and 24-h reperfusion. Factor B-deficient or CR2-fH–treated mice were protected in terms

of improved neurologic function and reduced cerebral infarct, demyelination, P-selectin expression, neutrophil infiltration, and

microthrombi formation. Mice deficient in both the classical and lectin pathways (C1q/MBL deficient) were also protected from

cerebral ischemia-reperfusion injury, and there was no detectable C3d deposition in the ipsilateral brain of these mice. These data

demonstrate that the alternative pathway is not alone sufficient to initiate complement activation and indicate that the alternative

pathway propagates cerebral injury via amplification of the cascade. Deficiency of C6, a component of the terminal cytolytic

membrane attack complex, had no effect on outcome after ischemic stroke, indicating that the membrane attack complex is not

involved in mediating injury in this model. We additionally show that the protective effect of factor B deficiency and CR2-fH

treatment is sustained in the subacute stage of infarct development, adding to the clinical relevance of these findings. The Journal

of Immunology, 2012, 189: 4640–4647.

Transient ischemic stroke is a pathological process in whicha cerebral vessel is occluded creating an ischemic core,and once the blockage is resolved, either naturally or

through pharmaceutical intervention, the returned blood flowcreates a paradoxical secondary injury (reviewed in Refs. 1, 2).Stroke is a leading cause of death and disability, and the role ofcomplement in cerebral ischemia-reperfusion injury (IRI) has re-ceived much attention recently.Complement can be activated by either the classical, lectin, or

alternative pathways. The classical and lectin pathways can beactivated by Abs, certain polysaccharides, and apoptotic cells. Thealternative pathway can be activated spontaneously but alsofunctions to amplify the other pathways. All pathways convergewith the formation of a C3 convertase and the cleavage of C3 toproduce C3a and C3b. C3a is a soluble bioactive peptide, and C3bbecomes covalently bound to the activating surface, where it isfurther cleaved to produce iC3b and C3d opsonins. C3 cleavage

also leads to C5 cleavage to yield C5a and C5b. C5a has multipleinflammatory activities, including the recruitment and activationof leukocytes, whereas C5b initiates the terminal pathway andformation of the cytolytic membrane attack complex (MAC).Although there is good clinical evidence for a role of comple-

ment in ischemic stroke (3), our current understanding of thecomplement-dependent mechanisms involved in ischemic strokederives mostly from murine studies and the use of complement-deficient and complement-inhibited mice. In a seminal study, itwas shown that after middle cerebral artery occlusion (MCAO),ischemic neurons upregulate the classical complement pathwaymolecule C1q and that a soluble complement inhibitor, sCR1, isprotective (4). Furthermore, an sCR1 derivative with sialyl Lewisx

glycosylation inhibited both complement activation and selectin-mediated adhesion and demonstrated enhanced therapeutic activ-ity. Subsequent studies demonstrated that genetic deficiency of thecentral complement protein C3 (5, 6), as well as inhibition of C3convertases (5), reduced infarct volumes and inflammation andimproved neurologic deficit after MCAO. As noted above, theprinciple complement effector molecules that are produced afteractivation of any pathway are C3a, C5a, C3 opsonins, and theMAC. In the context of rodent ischemic stroke, both C3a and C5asignaling have been shown to contribute to ischemia-related in-flammation and injury, as both C3a receptor (6, 7) and C5a re-ceptor (8, 9) antagonism reduced cerebral inflammation, infarctvolume, and neurologic deficit. It should be noted, however, thatC5 deficiency is not protective after focal cerebral ischemia, afinding that is not consistent with a role for C5a (6).A recent study demonstrated an important role for natural self-

reactive IgM in activating complement after ischemic stroke (10).Complement activation by natural IgM has also been shown tooccur in other organ models of IRI, with IgM-mediated activationoccurring via the lectin pathway (11–13). A role for the lectinpathway is also indicated in cerebral IRI, as lectin pathway defi-ciency is protective in murine models (14, 15). Clinical data also

*Department of Microbiology and Immunology, Children’s Research Institute, Med-ical University of South Carolina, Charleston, SC 29425; †Department of Neurosci-ence, Neuroscience Institute, Medical University of South Carolina, Charleston, SC29425; ‡Department of Pediatrics, Massachusetts General Hospital, Harvard MedicalSchool, Boston, MA 02114; xDepartment of Anesthesiology, Perioperative and PainMedicine, Center of Experimental Therapeutics and Reperfusion Injury, HarvardInstitute of Medicine, Harvard Medical School, Boston, MA 02115; and {Ralph H.Johnson Veterans Affairs Medical Center, Charleston, SC 29401

Received for publication July 9, 2012. Accepted for publication September 5, 2012.

This work was supported by grants from the National Institutes of Health(HL082485, HL86576) and the Department of Veterans Affairs (Merit AwardNURC-051010F).

Address correspondence and reprint requests to Dr. Stephen Tomlinson, Department ofMicrobiology and Immunology, Children’s Research Institute, Medical University ofSouth Carolina, 173 Ashley Avenue BSB 201, Charleston, SC 29401. E-mail address:tomlinss@musc.edu

Abbreviations used in this article: ECA, external carotid artery; fB, factor B; GFAP,glial fibrillary acidic protein; IRI, ischemia-reperfusion injury; MAC, membraneattack complex; MCAO, middle cerebral artery occlusion; tPA, tissue-type plasmin-ogen activator; wt, wild-type.

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1201904

by guest on March 25, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

suggest a role for the lectin pathway in human ischemic stroke(14, 16). And although IgM also binds C1q and is a powerfulactivator of the classical pathway, C1q deficiency is not protectiveafter murine focal cerebral ischemia (6, 17). The role of the al-ternative pathway of complement activation in ischemic strokehas not been investigated, although this pathway is importantfor propagating inflammation and injury in multiple pathologicalconditions (18, 19).A role of the terminal pathway in ischemic cerebral injury has

been investigated using neonatal rats that are developmentallydeficient in C9. Reconstitution of neonatal rats with C9 exacerbatedcerebral injury after carotid artery ligation with hypoxia, indicatingan important role for the MAC (20). In contrast, in a mouse modelof neonatal hypoxic-ischemic brain injury involving permanentischemia, C6 deficiency was not protective, indicating that theMAC does not play a role in cerebral injury in this model (21). Ina model of MCAO and transient ischemia, as used in the currentstudy, the effect of CD59a deficiency, a membrane-bound inhibitorof MAC formation, was dependent on gender and the period ofischemia (22).In the current study, we investigated the role of the alternative

pathway in propagating secondary inflammation and injury aftertransient focal cerebral ischemia and reperfusion. In addition, wefurther investigated the role of the terminal pathway in transientischemic stroke after MCAO by a direct approach by using micethat are deficient in C6 and that cannot form the MAC. We alsoinvestigated a therapeutic strategy for limiting cerebral inflamma-tion and injury using CR2-fH, which targets sites of complementactivation via CR2-mediated recognition of C3 opsonins (23). Ithas been shown previously that CR2-fH specifically inhibits thealternative pathway (23, 24). A human counterpart of CR2-fHtermed TT30, also shown to be specific for the alternative pathway(25), is currently in clinical trials.

Materials and MethodsMice

All mice used in these studies, including complement-deficient mice, wereon the C57BL/6 genetic background, male, and 8–9 wk old at the time ofstudy. Wild-type mice were obtained from The Jackson Laboratory (BarHarbor, ME). The following complement-deficient strains of mice wereused: fB2/2 (lack alternative pathway) (26); C1q/MBL2/2 (lack classicaland lectin pathway) (27); C62/2 (lack ability to form the MAC) (28, 29);CD59a2/2 (lack membrane inhibitor of MAC formation) (30). The C62/2

and CD59a2/2 mice were kindly provided by Dr. B. Paul Morgan (Uni-versity of Wales, Cardiff, U.K.). Deficient mice were bred and housed atthe Medical University of South Carolina. All procedures were approvedby the Institutional Animal Care and Use Committee at the MedicalUniversity of South Carolina, in accordance with the National Institutes ofHealth Guide for the Care and Use of Laboratory Animals.

Middle cerebral artery occlusion

Mice were subjected to 60-min MCAO, followed by a reperfusion period of24 h, 72 h, or 7 d before micewere sacrificed and brains isolated for analysis,as previously described (5, 10). Briefly, mice were anesthetized, the ex-ternal carotid artery (ECA) exposed, and the blunted tip of a 6-0 nylonsuture inserted through the ECA into the internal carotid artery. The suturewas advanced so as to block the middle cerebral artery. After 60 min ofischemia, the suture was withdrawn, the ECA stump ligated, and the in-cision closed. Laser Doppler flowmetry was used to assess cerebral bloodflow before, during, and after ischemia; mice not achieving reduction inblood flow to 20% of pre-ischemia levels were excluded from analysis.Temperature, blood pressure, and heart rate were also monitored before,during, and 10 min after ischemia.

CR2-fH and CR2-Crry preparation and treatment

The recombinant fusion protein CR2-fH, a site-targeted inhibitor of thealternative pathway (24), was prepared as previously described (23). CR2-Crry, an inhibitor of all complement pathways at the C3 activation step, was

prepared as previously described (31). Protein purity was assessed by SDSgel electrophoresis, and complement inhibitory activity was evaluated byzymosan assay (23). For complement inhibitor treatments, randomizedwild-type (wt) C57BL/6 mice received 0.4 mg CR2-fH, 0.25 mg CR2-Crry,or 100 ml vehicle control (PBS) intravenously via tail vein 30 min post-reperfusion (5). Inhibitor dosing was based on previously published dose-response data for CR2-fH and CR2-Crry in a model of intestinal IRI (23).

Twenty-four hour survival

Survival to 24 h after 60 min of MCAO was compared among groups.Survival was compared using Fisher’s exact 2 3 2 test.

Evaluation of infarct volume

After sacrifice of mice, brains were perfused with PBS, isolated, and placedin a rodent brain matrix (EMS, Hatfield, PA). Two-millimeter slices wereprepared and stained with 2% triphenyltetrazolium chloride (Sigma-Aldrich, St. Louis, MO). Total infarct area was assessed using ImageJAnalysis Software (National Institutes of Health) and calculated by sum-mation of infarcted areas of all slices for each hemisphere. Animals that didnot survive to the established endpoint (24 h, 72 h, or 7 d) were excludedfrom the determination of infarct volumes.

Neurologic deficit scoring

Neurologic deficits were determined as described (32) according to thefollowing scoring system: 0, normal function; 1, flexion of torso andcontralateral forelimb when lifted; 2, normal posture at rest with circling tocontralateral side when held by tail on flat surface; 3, leaning to contra-lateral side when at rest; 4, no spontaneous movement. Scoring was per-formed by an observer blinded to experimental groups.

Histopathology

At sacrifice, mice were perfused with PBS. Brains were then removed, fixedovernight in 4% paraformaldehyde in PBS, and either immersed in 20%sucrose for cryosectioning or processed to paraffin for Luxol fast blue/Nissl(American MasterTech, Lodi, CA) or H&E staining for morphologicalanalysis.

Immunohistochemistry

Paraffin processed sections cut at 8 mm were subjected to Ag retrieval bycitrate steam for 30 min (IHC World, Ellicot City, MD), and primary Abswere then applied in PBS for 2 h at room temperature. Primary Abs usedwere goat anti-mouse C3d (1:20; R&D Systems, Minneapolis, MN), goatanti-mouse IgM (1:50; Sigma-Aldrich), and rat anti-mouse Ly6G+Ly6C(Gr-1, 1:20; BD Biosciences, San Jose, CA). Brain sections from micesacrificed at 72 h were stained with rabbit anti-mouse glial fibrillary acidicprotein (GFAP) (1:1000; Dako, Carpinteria, CA) in PBS for 2 h at roomtemperature. Immunohistochemistry to detect P-selectin expression wasdone on 8-mm cyrosections stained with rat anti-mouse CD62P (1:10; BDBiosciences) overnight at 4˚C. Primary Abs were detected using the ap-propriate ImmPress kit (Vector Laboratories, Burlingame, CA), developedusing NovaRed (Vector Laboratories), and sealed with CytoSeal-60(Richard-Allan Scientific, Kalamazoo, MI). Primary Abs were omittedfor negative controls. A blinded observer imaged the slides by light mi-croscopy (Nikon Eclipse E600).

Assessment of complement deposition, microthrombiformation, and neutrophil infiltration

Brain tissue sampled was located between +2 mm and 22 mm relative tothe bregma, with injury confirmed with Luxol fast blue/Nissl or counter-staining. C3d deposition within the penumbra was scored as describedpreviously (33): 0, no observable staining; 1, mild staining; 2, moderatestaining; 3, severe staining. To assess microthrombi formation, H&E sectionswere scored as described previously (5): 0, no observable thrombi; 1, modestRBC attachment; 2, focal RBC attachment and fibrin; 3, 50% occlusion byRBC and fibrin; 4, total occlusion. To quantify neutrophil infiltration in thepenumbra, 10 random fields, at 340 magnification, from Gr-1–stained sec-tions were counted. All assessments were performed by an observer blindedto experimental groups.

Evaluation of apoptosis

Paraffin brain sections from mice sacrificed at 72 h were evaluated forapoptosis by TACS 2 TdT-DAB in situ apoptosis detection kit (Trevigen,Gaithersburg, MD). The staining protocol was followed according to themanufacturer’s instructions. Negative and positive controls were includedwith each staining batch. Apoptotic cells in the peri-infarct region of the

The Journal of Immunology 4641

by guest on March 25, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

penumbra were counted from 10 random fields at 340 magnification bylight microscopy. The observer was blinded to the experimental groups.

Statistical analysis

Statistical analyses were performed using Prism 4 software (GraphPad, LaJolla, CA). Parametric data (infarct volumes) were analyzed for statisticalsignificance with one-way ANOVA test. Non-parametric data (neurologicdeficits, C3d deposition scores, microthrombi scores, and cell counts) wereanalyzed for statistical significance with Kruskal–Wallis test. Comparisonsbetween two groups were conducted using Student t test (parametric) orMann–Whitney U test (non-parametric). Differences were consideredsignificant when p , 0.05.

ResultsAlternative pathway deficiency and inhibition protects againstcerebral IRI

Neurologic deficit was assessed 24 h after 60 min of cerebral is-chemia. Compared to wt controls, mice deficient in either factor B(fB; no alternative pathway) or C1q/MBL (only alternative path-way) displayed significant improvements in median neurologicdeficit (Fig. 1A). There was no significant difference in neurologicdeficits between fB- and C1q/MBL-deficient mice. To investigatethe role of the alternative pathway in a more clinically relevantparadigm, wt mice were treated with either CR2-fH, a targetedinhibitor of the alternative pathway, or CR2-Crry, a similarly tar-geted inhibitor that blocks all complement activation pathways.Treatments were given at 30 min after reperfusion. Both inhibitorssignificantly improved median neurologic deficit compared with wtcontrol mice, and there was no significant difference either between

CR2-fH– and CR2-Crry–treated mice or between inhibited miceand fB- or C1q/MBL-deficient mice (Fig. 1A). Focal cerebralinfarct volumes were also analyzed 24 h after reperfusion. Com-pared to wt controls, mean infarct volumes were significantlydecreased in both fB- and C1q/MBL-deficient mice, althoughinfarct volumes were significantly smaller in C1q/MBL-deficientmice compared with fB-deficient mice (Fig. 1B). Furthermore,consistent with the effect of CR2-fH and CR2-Crry on neurologicdeficit, both inhibitors significantly reduced mean infarct volumescompared with wt controls, and although there was a trend towardimproved outcome with CR2-Crry compared with CR2-fH, thedifference was not significant (Fig. 1B).

The terminal complement pathway does not play a role incerebral injury after 60-min transient ischemia

The terminal complement pathway andMAC formation is commonto all activation pathways and has been implicated as a key me-diator of IRI in various organs. However, a previous study dem-onstrated that deficiency of CD59a, an endogenous membrane-bound inhibitor of MAC formation, exacerbated cerebral IRIafter a 30-min but not after a 60-min ischemic period (22). Toinvestigate more directly the role of the terminal pathway and theMAC in cerebral IRI, we determined the effect of C6 deficiency,a component protein of the MAC, in the more severe model ofcerebral IRI incorporating 60 min of MCAO. C6 deficiency wasnot protective in terms of either neurologic function or infarctvolume compared with wt control mice (Fig. 2), indicating that theterminal pathway does not contribute to IRI in this model. We alsoconfirmed the previous data demonstrating that CD59a deficiencywas without effect on cerebral injury after 60 min of MCAO (Fig.

FIGURE 1. Effect of complement deficiency and CR2-fH or CR2-Crry

treatment on neurologic deficit and infarct volume after 60-min MCAO

and 24-h reperfusion. (A and B) Neurologic deficit (A) and infarct volume

(B) in C1q/MBL2/2 mice, fB2/2 mice, and wt mice treated with CR2-fH

or CR2-Crry 30 min after reperfusion. Horizontal bar indicates median

neurologic score and mean infarct volume, n = 12–15 for neurologic deficit

and n = 8–11 for infarct volume. *p , 0.05, **p , 0.01, ***p , 0.001.

FIGURE 2. Effect of terminal pathway deficiency and terminal pathway

inhibitor deficiency on neurologic deficit and infarct volume after 60-min

MCAO and 24-h reperfusion. (A) Neurologic deficit in C62/2 and

CD59a2/2 mice. Horizontal bar represents median; n = 10–11. No sta-

tistical difference between groups. (B) Infarct volume in C62/2 and

CD59a2/2 mice. Horizontal bar indicates mean, n = 6–8. No statistical

difference between groups.

4642 COMPLEMENT AND ISCHEMIC STROKE

by guest on March 25, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

2). Subsequent studies focused on the role of the alternativepathway in this model of ischemic stroke.

Neuronal tissue integrity

Neuronal integrity, as demonstrated by myelin staining with Luxolfast blue/Nissl staining, was improved in C1q/MBL2/2mice, fB2/2

mice, and CR2-fH–treated mice compared with wt controls. Therewas no apparent improvement in C62/2 mice (Fig. 3). Thus, im-proved neuronal integrity correlates with improvements in infarctvolumes and neurologic deficits for the different groups of miceshown in Fig. 1. Taken together, these data indicate that the alter-native pathway alone is not sufficient to cause injury after cerebralischemia-reperfusion but that it plays a key role in propagating in-jury after complement activation by either the classical or lectinpathway, and published data have shown dependence on the lectinpathway (14, 15).

Twenty-four hour survival and physiological measurements

C1q/MBL deficiency significantly improved the survival to 24 h (18of 19) compared with wt (12 of 18, p = 0.03 by Fisher’s exact 2 32 test). Mice deficient in fB or treated with CR2-fH appeared tohave increased survival rates to 24 h compared with wt mice (15of 18 and 17 of 20, respectively), but the difference did not reachsignificance. There was also no significant difference between24-h survival rates of wt mice and mice deficient in C6 or CD59a(11 of 16 and 8 of 11, respectively).To confirm that cerebral blood flow was interrupted by the

MCAO procedure, blood flow was measured by laser Doppler, andthere were no significant differences in cerebral blood flow betweenany of the groups before, during, or 10 min after ischemia (data notshown). Changes in blood pressure and body temperature cansignificantly influence the outcome after stroke, and we thereforemeasured blood pressure, heart rate, and temperature before, during,and after ischemia. There were no differences between the groups(data not shown). Also, as expected, there was no damage observedin the contralateral hemisphere in any group.

Complement deposition is reduced in complement-deficientand -inhibited mice after MCAO

Penumbral deposition of the C3 activation product, C3d, was ex-amined on sections of brain isolated 24 h after reperfusion, at whichtime complement deposition is maximal in the 60-min MCAO

model (34). C3d staining was seen in the penumbra throughout theipsilateral hemisphere in wt mice but was virtually absent in C1q/MBL mice (Fig. 4). Mice deficient in fB or treated with CR2-fHhad significantly reduced C3d staining compared with wt mice, andthe staining that was observed was associated predominantly withthe vasculature in the penumbra (Fig. 4). Brain sections from C6-deficient mice and wt mice had similar levels of C3d staining,and contralateral tissue did not show any C3d staining for any ofthe experimental groups (data not shown). These data support theabove outcome measurements in terms of the conclusion that thealternative pathway amplifies complement activation and drivescerebral injury subsequent to classical/lectin pathway initiation.

Reduced complement deposition is not due to reduced IgMbinding in ischemic brain

Natural self-reactive IgM Abs that recognize neoepitopes exposedon ischemic tissue have been shown to play an important role inactivating complement and propagating IRI, including cerebral IRI(10). We therefore wanted to confirm that the effect of the dif-ferent complement deficiencies on cerebral IRI was not due todifferences in the binding of IgM after reperfusion. Immunohis-tochemical analysis of brain sections isolated 24 h after reperfu-sion revealed similar levels of IgM staining in all groups, and IgMstaining was absent in sections isolated from the contralateralhemisphere (Fig. 5).

Microthrombi formation is reduced in alternativepathway-deficient and -inhibited mice

Secondary thrombus formation in the penumbra can perpetuatefurther ischemic injury and is a significant complication after is-chemic stroke. We previously demonstrated fibrinogen deposition

FIGURE 3. Neuronal tissue integrity in complement-deficient and CR2-

fH–treated mice after 60-min MCAO and 24-h reperfusion. Brain sections

from the ipsilateral hemisphere were stained with Luxol fast blue/Nissl.

Section from contralateral hemisphere shown as undamaged control. Repre-

sentative images shown, n = 4/group. Original magnification 310; scale bar,

200 mm.

FIGURE 4. C3d deposition in penumbral tissue in complement-deficient

and CR2-fH–treated mice after 60-min MCAO and 24-h reperfusion. (A)

Representative immunohistochemistry images taken at original magnifi-

cation340. Scale bar, 50 mm. There was no observable C3d staining in the

contralateral hemispheres in any group. (B) Quantification of C3d staining.

Horizontal bar represents median, n = 5. *p , 0.05, **p , 0.01.

The Journal of Immunology 4643

by guest on March 25, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

and microthrombi formation in the cerebrovasculature afterMCAO, and we therefore investigated whether the alternativepathway contributes to this postischemic complication. Brainsections were analyzed for the presence of RBC accumulation andfibrin deposition as previously described (5). There were signifi-cantly lower numbers of microthrombi and partially occludedvessels in fB-deficient and CR2-fH–treated mice compared with wtmice (Fig. 6). Microthrombi formation appeared to be even lower inC1q/MBL-deficient mice compared with fB-deficient and CR2-fH–treated mice, although the difference did not reach significance.

P-selectin expression and neutrophil infiltration is reduced inalternative pathway-deficient and -inhibited mice

Infiltrating leukocytes are important mediators of postischemicinflammation and injury, and complement activation products areinvolved in their recruitment, either directly or indirectly via upreg-ulation of inflammatory mediators such as the adhesion moleculeP-selectin. The expression of P-selectin and the infiltration ofneutrophils are directly linked to cerebral inflammation and injuryafter ischemic stroke (4, 35), and we therefore examined howmodulation of the alternative pathway affects these two parameters.Compared to wt mice, fB-deficient and CR2-fH–treated mice bothhad significantly reduced P-selectin expression and neutrophil in-filtration in the cerebral vasculature 24 h after MCAO (Fig. 7).Furthermore, P-selectin expression was undetectable in C1q/MBL-

deficient mice, and neutrophil infiltration was significantly lowerin C1q/MBL-deficient mice compared with fB-deficient and CR2-fH–treated mice. There was no detectable expression of P-selectinin the contralateral hemisphere from any experimental group. Thus,blocking or inhibiting the alternative pathway reduces cerebralinflammation, and this correlates with the above data showing re-duced cerebral injury, reduced complement deposition, and re-duced microthrombi formation. The enhanced effect of C1q/MBLdeficiency on the above measured parameters compared with fBdeficiency is indicative of the alternative pathway exerting its effectin propagating pathogenesis after ischemic stroke via its function inthe complement cascade amplification loop.

CR2-fH treatment is protective at 72 h after MCAO

To address better the clinical relevance of our findings, we in-vestigated the effect of a single post-reperfusion injection of CR2-fH, as well as fB deficiency, on infarct maturation and longer-termoutcome after MCAO. Compared to wt mice, CR2-fH–treated orfB-deficient mice had significantly reduced median neurologicdeficits at 3 d after reperfusion (Fig. 8A). Furthermore, the median

FIGURE 5. IgM deposition in penumbral tissue in complement-deficient

and CR2-fH–treated mice after 60-min MCAO and 24-h reperfusion. There

were no discernible differences in the levels or distribution of IgM staining

between any groups. No IgM staining was observed in the contralateral

hemispheres in any group. Representative images, n = 3. Original mag-

nification 340; scale bar, 50 mm.

FIGURE 6. Microthrombi formation in penumbral tissue in complement-

deficient and CR2-fH–treated mice after 60-min MCAO and 24-h reperfu-

sion. Microthrombi formation was assessed in H&E-stained brain sections.

Horizontal bars represent median microthrombi scores, n = 5–7. *p , 0.05.

FIGURE 7. P-selectin expression and neutrophil infiltration in penumbral

tissue in complement-deficient and CR2-fH–treated mice after 60-min

MCAO and 24-h reperfusion. (A) P-selectin expression as detected by im-

munohistochemistry. Expression is associated with the vasculature. Repre-

sentative images, n = 3. Original magnification 310; scale bar, 200 mm. No

staining was detected in contralateral hemisphere of any group. (B) Neutrophil

infiltration as detected by immunohistochemical staining of Gr-1+ cells.

Horizontal bar represents median, n = 4–5. **p , 0.01, ***p , 0.001.

4644 COMPLEMENT AND ISCHEMIC STROKE

by guest on March 25, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

neurologic deficit in CR2-fH–treated and fB-deficient mice waslower at 72 h after reperfusion than it was at 24 h after reperfusion.Infarct volumes also remained significantly smaller at 72 h afterreperfusion in CR2-fH–treated and fB-deficient mice comparedwith wt mice (Fig. 8B). Because astrocyte responses can bemodulated by complement activation products (36), we also de-termined how alternative pathway inhibition with CR2-fH effectsastrogliosis, a hallmark of brain injury (37). As determined byGFAP expression, CR2-fH treatment dampened the reactive as-trocyte presence associated with the injury border seen in wt miceat 3 d after reperfusion (Fig. 8C). Reduction in the GFAP reac-tivity was also observed in fB-deficient mice. There were nodifferences between groups in GFAP reactivity in the corre-sponding contralateral sections (data not shown). To determine ifCR2-fH treatment affects delayed cell death, we determined ap-optosis in the peri-infarct region of the penumbra at 72 h afterMCAO. As shown in Fig. 8D, there were fewer apoptotic cells inthe penumbral region of brains from mice treated with CR2-fHcompared with wt control mice. The number of apoptotic cells inbrains from fB-deficient mice was similar to that seen in CR2-fH–treated mice. There were no observed differences in the contra-lateral hemispheres between groups (data not shown).

CR2-fH–mediated protection from cerebral injury persists to7 d after reperfusion

Whereas C1-INH (38) and C3aR antagonist (39) have been shownto be neuroprotective in stroke models, with improved outcomemaintained for 7 d after ischemia, lectin pathway deficiency hasbeen shown to provide only acute (24 h) protection after MCAO(40). To determine if CR2-fH–mediated protection persists intothe subacute phase of stroke, neurologic deficit and infarct volumewas assessed at 7 d after MCAO. Compared to vehicle-treatedcontrols, mice receiving a single dose of CR2-fH at 30 min af-ter reperfusion had significantly reduced mean infarct volumes

and displayed a strong trend toward neurologic improvement(p = 0.07) 7 d after MCAO (Fig. 9). These data indicate that CR2-fH treatment provides sustained neuroprotection, rather than justdelaying cerebral ischemic injury.

FIGURE 8. Effect of fB deficiency and CR2-fH treatment on outcomes after 60-min MCAO and 72-h reperfusion. (A) Neurologic deficit. Horizontal bar

represents median, n = 5–6. *p , 0.05. (B) Infarct volume. Horizontal bar represents mean, n = 5–6. ***p , 0.001. (C) Astrocyte GFAP reactivity as

assessed by GFAP immunohistochemistry. Representative images shown, n = 3. Original magnification34, scale bar, 500 mm; original magnification 310,

scale bar, 200 mm. Contralateral GFAP staining was not different between groups. (D) Apoptosis as evaluated by counting 10 random 340 fields from the

peri-infarct region of the penumbra. Horizontal bar represents median, n = 3–4. *p , 0.05.

FIGURE 9. Effect of CR2-fH treatment on neurologic deficit and infarct

volume after 60-min MCAO and 7-d reperfusion. (A) Neurologic deficit.

Horizontal bar represents median, n = 7. p = 0.07. (B) Infarct volume.

Horizontal bar indicates mean, n = 7. **p , 0.01.

The Journal of Immunology 4645

by guest on March 25, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

DiscussionThis study demonstrates an important role for the alternative com-plement pathway in the pathogenesis of murine ischemic stroke.

Although the alternative pathway can be spontaneously activated,

the current data demonstrate that a functioning alternative pathway

is not alone sufficient to initiate complement activation and prop-

agate murine cerebral IRI. Rather, the alternative pathway serves to

amplify complement activation initiated by the classical and/or

lectin pathway. Previous data demonstrating that lectin pathway-

deficient (14, 15) but not classical pathway-deficient (6, 17) mice

are protected from cerebral IRI, indicating that it is alternative

pathway-mediated amplification of the lectin pathway that drives

injury in this model of ischemic stroke. Further, we have previously

shown an important role for self-reactive natural IgM in propa-

gating cerebral IRI (10), and others have shown that IgM mediates

IRI via activation of the lectin pathway, at least after ischemia and

reperfusion of the intestine and heart (11–13).In this study, we also show that mice deficient in C6, a com-

ponent protein of the MAC, are not protected from cerebral injury

after 60-min MCAO and 24-h reperfusion. This finding indicates

that the terminal complement pathway and the MAC do not play

a role in murine cerebral IRI after 60 min MCAO, and is in accord

with a previous study that demonstrated mice deficient in CD59a,

an inhibitor of the MAC, do not display exacerbated cerebral injury

in the same model (22). It is interesting that in contrast to this

finding, the MAC is implicated as an important mediator of IRI

in various other organs and tissues (41, 42), as well as in other

neuroinflammatory conditions such as spinal cord injury (33) and

traumatic brain injury (43). Nevertheless, it should be noted that

C6 deficiency is protective in a neonatal model of hypoxic-

ischemic brain injury involving permanent ischemia, implicating

a role for the MAC in a different stroke model (21). Furthermore,

CD59a-deficient mice show worse outcomes after 30-min MCAO,

as opposed to 60-min MCAO, indicating that the terminal pathway

may contribute to cerebral IRI under conditions of less severe

injury/ischemia (22). Also, although outcomes were not different

between wt and CD59a-deficient mice after 60-min MCAO, the

level of apoptosis was increased in CD59a-deficient mice, and the

authors suggested that MAC formation after severe ischemic

stroke may be redundant and secondary to injury propagated by

resident and infiltrating inflammatory cells. In this context, al-

though our data indicate that the MAC is not essential for the

progression of complement-dependent cerebral injury after 60-min

MCAO, we demonstrated that the protective effect of upstream

blockade of the alternative pathway was associated with re-

duced levels of apoptosis and inflammation (discussed later).

Further studies are needed fully to address the role of the MAC

in ischemic stroke, which in experimental models appears to be

dependent on the model and/or severity of injury.The alternative pathway plays an important role in driving in-

flammation and propagating injury in many pathological conditions

(reviewed in Ref. 18), including neuroinflammatory conditions

such as traumatic spinal cord injury (33), traumatic brain injury

(44), and laser-induced choroidal neovascularization (45). Hall-

marks of cerebral IRI (and inflammation in general) are the ex-

pression of endothelial adhesion molecules and the infiltration of

leukocytes (35, 46, 47), both of which can be directly and indi-

rectly modulated by complement activation products such as C3a

and C5a. We demonstrated previously that C3 deficiency or CR2-

Crry treatment (inhibits all complement pathways at C3 activa-

tion) protects against cerebral IRI and decreases the level of P-

selectin expression and neutrophil infiltration and microthrombi

formation after 60-min MCAO (5). In this study, we demonstrate

that fB deficiency or specific inhibition of the alternative pathwayhas a similar effect, and we show no significant difference betweenCR2-fH and CR2-Crry treatment outcomes in terms of neurologicdeficit and infarct volume at 24 h after reperfusion. In addition, thereduction in infarct volume and improvement in neurologic scoreseen in CR2-fH–treated mice at 24 h after reperfusion was furtherimproved at 72 h, with improvements persisting for 7 d afterreperfusion. Because complement has many important physio-logical functions, specific inhibition of the alternative pathwaywill likely be a beneficial therapeutic approach compared withinhibition of all complement pathways and is less likely to haveundesirable side effects. In this regard, complement activationproducts have been shown to have protective roles against certaintypes neuronal injury (reviewed in Ref. 3), and the classical andlectin pathways are known to play important roles in host defenseagainst some pathogens, as well as provide some important reg-ulatory and homeostatic functions (48, 49). For example, theclassical and lectin pathways play a role in the removal of apo-ptotic and injured cells, important for the resolution of inflam-mation and tissue repair (50).Recombinant tissue-type plasminogen activator (tPA) is the only

approved pharmacological agent for the treatment of ischemicstroke, and in general it must be administered within 3 h of symptomonset. tPA promotes reperfusion by enhancing the dissolution ofblood clots, but there is the risk of uncontrollable intracranialhemorrhage, and for this reason physicians are often reluctant touse this drug. There is currently no intervention that can reverse theeffects of tPA on acute hemorrhage. It is noteworthy that throm-bolytic therapy has been shown to activate complement (51), andcomplement inhibition may thus have the potential to synergizewith thrombolytic therapy, possibly even extending the windowfor tPA treatment. In this context, a human counterpart of CR2-fH,named TT30, is currently in phase 1 clinical trial for nocturnalparoxysmal hemoglobinuria (http://clinicaltrials.gov/ct2/results?term=tt30). TT30 has demonstrated neuroprotection in murine laser-induced choroidal neovascularization (a model of age-related mac-ular degeneration) (52) and has been shown to prevent alternativepathway activity, C3 opsonin accumulation, and MAC formation onRBCs, without interfering with the classical and lectin activationpathways (53).

AcknowledgmentsWe thank Dr. B. Paul Morgan for providing breeding pairs of C62/2 and

CD59a2/2 mice.

DisclosuresS.T. has received research support from Alexion Pharmaceuticals, a com-

pany developing complement inhibitory therapeutics. The other authors

have no financial conflicts of interest.

References1. Dirnagl, U., C. Iadecola, and M. A. Moskowitz. 1999. Pathobiology of ischaemic

stroke: an integrated view. Trends Neurosci. 22: 391–397.2. Schaller, B., and R. Graf. 2004. Cerebral ischemia and reperfusion: the patho-

physiologic concept as a basis for clinical therapy. J. Cereb. Blood Flow Metab.24: 351–371.

3. Brennan, F. H., A. J. Anderson, S. M. Taylor, T. M. Woodruff, andM. J. Ruitenberg. 2012. Complement activation in the injured central nervoussystem: another dual-edged sword? J. Neuroinflammation 9: 137.

4. Huang, J., L. J. Kim, R. Mealey, H. C. Marsh, Jr., Y. Zhang, A. J. Tenner,E. S. Connolly, Jr., and D. J. Pinsky. 1999. Neuronal protection in stroke by ansLex-glycosylated complement inhibitory protein. Science 285: 595–599.

5. 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.

6. Mocco, J., W. J. Mack, A. F. Ducruet, S. A. Sosunov, M. E. Sughrue,B. G. Hassid, M. N. Nair, I. Laufer, R. J. Komotar, M. Claire, et al. 2006.

4646 COMPLEMENT AND ISCHEMIC STROKE

by guest on March 25, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

Complement component C3 mediates inflammatory injury following focal ce-rebral ischemia. Circ. Res. 99: 209–217.

7. Ducruet, A. F., B. G. Hassid, W. J. Mack, S. A. Sosunov, M. L. Otten,D. J. Fusco, Z. L. Hickman, G. H. Kim, R. J. Komotar, J. Mocco, andE. S. Connolly. 2008. C3a receptor modulation of granulocyte infiltration aftermurine focal cerebral ischemia is reperfusion dependent. J. Cereb. Blood FlowMetab. 28: 1048–1058.

8. Arumugam, T. V., S. C. Tang, J. D. Lathia, A. Cheng, M. R. Mughal,S. Chigurupati, T. Magnus, S. L. Chan, D. G. Jo, X. Ouyang, et al. 2007. In-travenous immunoglobulin (IVIG) protects the brain against experimental strokeby preventing complement-mediated neuronal cell death. Proc. Natl. Acad. Sci.USA 104: 14104–14109.

9. Kim, G. H., J. Mocco, D. K. Hahn, C. P. Kellner, R. J. Komotar, A. F. Ducruet,W. J. Mack, and E. S. Connolly, Jr. 2008. Protective effect of C5a receptor inhibitionafter murine reperfused stroke. Neurosurgery 63: 122–125, discussion 125–126.

10. Elvington, A., C. Atkinson, L. Kulik, H. Zhu, J. Yu, M. S. Kindy, V. M. Holers,and S. Tomlinson. 2012. Pathogenic natural antibodies propagate cerebral injuryfollowing ischemic stroke in mice. J. Immunol. 188: 1460–1468.

11. McMullen, M. E., M. L. Hart, M. C. Walsh, J. Buras, K. Takahashi, andG. L. Stahl. 2006. Mannose-binding lectin binds IgM to activate the lectincomplement pathway in vitro and in vivo. Immunobiology 211: 759–766.

12. Zhang, M., K. Takahashi, E. M. Alicot, T. Vorup-Jensen, B. Kessler, S. Thiel,J. C. Jensenius, R. A. Ezekowitz, F. D. Moore, and M. C. Carroll. 2006. Acti-vation of the lectin pathway by natural IgM in a model of ischemia/reperfusioninjury. J. Immunol. 177: 4727–4734.

13. Busche, M. N., V. Pavlov, K. Takahashi, and G. L. Stahl. 2009. Myocardial is-chemia and reperfusion injury is dependent on both IgM and mannose-bindinglectin. Am. J. Physiol. Heart Circ. Physiol. 297: H1853–H1859.

14. Cervera, A., A. M. Planas, C. Justicia, X. Urra, J. C. Jensenius, F. Torres,F. Lozano, and A. Chamorro. 2010. Genetically-defined deficiency of mannose-binding lectin is associated with protection after experimental stroke in mice andoutcome in human stroke. PLoS ONE 5: e8433.

15. Morrison, H., J. Frye, G. Davis-Gorman, J. Funk, P. McDonagh, G. L. Stahl, andL. Ritter. 2011. The contribution of mannose binding lectin to reperfusion injuryafter ischemic stroke. Curr. Neurovasc. Res. 8: 52–63.

16. Osthoff, M., M. Katan, F. Fluri, P. Schuetz, R. Bingisser, L. Kappos, A. J. Steck,S. T. Engelter, B. Mueller, M. Christ-Crain, and M. Trendelenburg. 2011.Mannose-binding lectin deficiency is associated with smaller infarction size andfavorable outcome in ischemic stroke patients. PLoS ONE 6: e21338.

17. De Simoni, M. G., E. Rossi, C. Storini, S. Pizzimenti, C. Echart, andL. Bergamaschini. 2004. The powerful neuroprotective action of C1-inhibitor onbrain ischemia-reperfusion injury does not require C1q. Am. J. Pathol. 164:1857–1863.

18. Thurman, J. M., and V. M. Holers. 2006. The central role of the alternativecomplement pathway in human disease. J. Immunol. 176: 1305–1310.

19. Holers, V. M. 2008. The spectrum of complement alternative pathway-mediateddiseases. Immunol. Rev. 223: 300–316.

20. Imm, M. D., P. W. Feldhoff, R. C. Feldhoff, and H. A. Lassiter. 2002. The ad-ministration of complement component C9 augments post-ischemic cerebralinfarction volume in neonatal rats. Neurosci. Lett. 325: 175–178.

21. Ten, V. S., J. Yao, V. Ratner, S. Sosunov, D. A. Fraser, M. Botto, B. Sivasankar,B. P. Morgan, S. Silverstein, R. Stark, et al. 2010. Complement component c1qmediates mitochondria-driven oxidative stress in neonatal hypoxic-ischemicbrain injury. J. Neurosci. 30: 2077–2087.

22. Harhausen, D., U. Khojasteh, P. F. Stahel, B. P. Morgan, W. Nietfeld, U. Dirnagl,and G. Trendelenburg. 2010. Membrane attack complex inhibitor CD59a pro-tects against focal cerebral ischemia in mice. J. Neuroinflammation 7: 15.

23. Huang, Y., F. Qiao, C. Atkinson, V. M. Holers, and S. Tomlinson. 2008. A noveltargeted inhibitor of the alternative pathway of complement and its therapeuticapplication in ischemia/reperfusion injury. J. Immunol. 181: 8068–8076.

24. Banda, N. K., B. Levitt, M. J. Glogowska, J. M. Thurman, K. Takahashi,G. L. Stahl, S. Tomlinson, W. P. Arend, and V. M. Holers. 2009. Targeted in-hibition of the complement alternative pathway with complement receptor 2 andfactor H attenuates collagen antibody-induced arthritis in mice. J. Immunol. 183:5928–5937.

25. Fridkis-Hareli, M., M. Storek, I. Mazsaroff, A. M. Risitano, A. S. Lundberg,C. J. Horvath, and V. M. Holers. 2011. Design and development of TT30, a novelC3d-targeted C3/C5 convertase inhibitor for treatment of human complementalternative pathway-mediated diseases. Blood 118: 4705–4713.

26. Matsumoto, M., W. Fukuda, A. Circolo, J. Goellner, J. Strauss-Schoenberger,X. Wang, S. Fujita, T. Hidvegi, D. D. Chaplin, and H. R. Colten. 1997. Abro-gation of the alternative complement pathway by targeted deletion of murinefactor B. Proc. Natl. Acad. Sci. USA 94: 8720–8725.

27. Banda, N. K., K. Takahashi, A. K. Wood, V. M. Holers, and W. P. Arend. 2007.Pathogenic complement activation in collagen antibody-induced arthritis in micerequires amplification by the alternative pathway. J. Immunol. 179: 4101–4109.

28. Bhole, D., and G. L. Stahl. 2004. Molecular basis for complement component 6(C6) deficiency in rats and mice. Immunobiology 209: 559–568.

29. Lewis, R. D., C. L. Jackson, B. P. Morgan, and T. R. Hughes. 2010. Themembrane attack complex of complement drives the progression of athero-sclerosis in apolipoprotein E knockout mice. Mol. Immunol. 47: 1098–1105.

30. Holt, D. S., M. Botto, A. E. Bygrave, S. M. Hanna, M. J. Walport, andB. P. Morgan. 2001. Targeted deletion of the CD59 gene causes spontaneousintravascular hemolysis and hemoglobinuria. Blood 98: 442–449.

31. 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.

32. Hata, R., G. Mies, C. Wiessner, K. Fritze, D. Hesselbarth, G. Brinker, andK. A. Hossmann. 1998. A reproducible model of middle cerebral artery occlu-sion in mice: hemodynamic, biochemical, and magnetic resonance imaging. J.Cereb. Blood Flow Metab. 18: 367–375.

33. Qiao, F., C. Atkinson, M. S. Kindy, A. Shunmugavel, B. P. Morgan, H. Song, andS. Tomlinson. 2010. The alternative and terminal pathways of complementmediate post-traumatic spinal cord inflammation and injury. Am. J. Pathol. 177:3061–3070.

34. Mack, W. J., M. E. Sughrue, A. F. Ducruet, J. Mocco, S. A. Sosunov,B. G. Hassid, J. Z. Silverberg, V. S. Ten, D. J. Pinsky, and E. S. Connolly, Jr.2006. Temporal pattern of C1q deposition after transient focal cerebral ischemia.J. Neurosci. Res. 83: 883–889.

35. Soriano, S. G., A. Coxon, Y. F. Wang, M. P. Frosch, S. A. Lipton, P. R. Hickey,and T. N. Mayadas. 1999. Mice deficient in Mac-1 (CD11b/CD18) are lesssusceptible to cerebral ischemia/reperfusion injury. Stroke 30: 134–139.

36. Jauneau, A. C., A. Ischenko, P. Chan, and M. Fontaine. 2003. Complementcomponent anaphylatoxins upregulate chemokine expression by human astro-cytes. FEBS Lett. 537: 17–22.

37. Pekny, M., and M. Nilsson. 2005. Astrocyte activation and reactive gliosis. Glia50: 427–434.

38. Heydenreich, N., M. W. Nolte, E. Gob, F. Langhauser, M. Hofmeister, P. Kraft,C. Albert-Weissenberger, M. Brede, C. Varallyay, K. Gobel, et al. 2012. C1-inhibitor protects from brain ischemia-reperfusion injury by combined antiin-flammatory and antithrombotic mechanisms. Stroke 43: 2457–2467.

39. Ducruet, A. F., B. E. Zacharia, S. A. Sosunov, P. R. Gigante, M. L. Yeh,J. W. Gorski, M. L. Otten, R. Y. Hwang, P. A. DeRosa, Z. L. Hickman, et al.2012. Complement inhibition promotes endogenous neurogenesis and sustainedanti-inflammatory neuroprotection following reperfused stroke. PLoS ONE 7:e38664.

40. Ducruet, A. F., S. A. Sosunov, B. E. Zacharia, J. Gorski, M. L. Yeh, P. Derosa,G. Cohen, P. R. Gigante, and E. S. Connolly, Jr. 2011. The neuroprotective effectof genetic mannose-binding lectin deficiency is not sustained in the sub-acutephase of stroke. Transl. Stroke Res. 2: 588–599.

41. Zhou, W., C. A. Farrar, K. Abe, J. R. Pratt, J. E. Marsh, Y. Wang, G. L. Stahl, andS. H. Sacks. 2000. Predominant role for C5b-9 in renal ischemia/reperfusioninjury. J. Clin. Invest. 105: 1363–1371.

42. Zhang, J., W. Hu, W. Xing, T. You, J. Xu, X. Qin, and Z. Peng. 2011. Theprotective role of CD59 and pathogenic role of complement in hepatic ischemiaand reperfusion injury. Am. J. Pathol. 179: 2876–2884.

43. Stahel, P. F., M. A. Flierl, B. P. Morgan, I. Persigehl, C. Stoll, C. Conrad,B. M. Touban, W. R. Smith, K. Beauchamp, O. I. Schmidt, et al. 2009. Absenceof the complement regulatory molecule CD59a leads to exacerbated neuropa-thology after traumatic brain injury in mice. J. Neuroinflammation 6: 2.

44. Leinhase, I., V. M. Holers, J. M. Thurman, D. Harhausen, O. I. Schmidt,M. Pietzcker, M. E. Taha, D. Rittirsch, M. Huber-Lang, W. R. Smith, et al.2006. Reduced neuronal cell death after experimental brain injury in micelacking a functional alternative pathway of complement activation. BMCNeurosci. 7: 55.

45. Rohrer, B., B. Coughlin, K. Kunchithapautham, Q. Long, S. Tomlinson,K. Takahashi, and V. M. Holers. 2011. The alternative pathway is required, butnot alone sufficient, for retinal pathology in mouse laser-induced choroidalneovascularization. Mol. Immunol. 48: e1–e8.

46. Ishikawa, M., D. Cooper, T. V. Arumugam, J. H. Zhang, A. Nanda, andD. N. Granger. 2004. Platelet-leukocyte-endothelial cell interactions after middlecerebral artery occlusion and reperfusion. J. Cereb. Blood Flow Metab. 24: 907–915.

47. Jin, A. Y., U. I. Tuor, D. Rushforth, J. Kaur, R. N. Muller, J. L. Petterson,S. Boutry, and P. A. Barber. 2010. Reduced blood brain barrier breakdown in P-selectin deficient mice following transient ischemic stroke: a future therapeutictarget for treatment of stroke. BMC Neurosci. 11: 12.

48. Lu, J. H., B. K. Teh, L. Wang, Y. N. Wang, Y. S. Tan, M. C. Lai, and K. B. Reid.2008. The classical and regulatory functions of C1q in immunity and autoim-munity. Cell. Mol. Immunol. 5: 9–21.

49. Takahashi, K. 2011. Mannose-binding lectin and the balance between immuneprotection and complication. Expert Rev. Anti Infect. Ther. 9: 1179–1190.

50. Bohlson, S. S., D. A. Fraser, and A. J. Tenner. 2007. Complement proteins C1qand MBL are pattern recognition molecules that signal immediate and long-termprotective immune functions. Mol. Immunol. 44: 33–43.

51. Oikonomopoulou, K., D. Ricklin, P. A. Ward, and J. D. Lambris. 2012. Inter-actions between coagulation and complement—their role in inflammation.Semin. Immunopathol. 34: 151–165.

52. Rohrer, B., B. Coughlin, M. Bandyopadhyay, and V. M. Holers. 2012. SystemicHuman CR2-Targeted Complement Alternative Pathway Inhibitor AmelioratesMouse Laser-Induced Choroidal Neovascularization. J. Ocul. Pharmacol. Ther.28: 402–409.

53. Risitano, A. M., R. Notaro, C. Pascariello, M. Sica, L. del Vecchio,C. J. Horvath, M. Fridkis-Hareli, C. Selleri, M. A. Lindorfer, R. P. Taylor, et al.2012. The complement receptor 2/factor H fusion protein TT30 protects par-oxysmal nocturnal hemoglobinuria erythrocytes from complement-mediatedhemolysis and C3 fragment. Blood 119: 6307–6316.

The Journal of Immunology 4647

by guest on March 25, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from