BASIC RESEARCH www.jasn.org

Soluble CR1 Therapy Improves ComplementRegulation in C3 Glomerulopathy

Yuzhou Zhang,* Carla M. Nester,*†‡ Danniele G. Holanda,§ Henry C. Marsh,|

Russell A. Hammond,| Lawrence J. Thomas,| Nicole C. Meyer,* Lawrence G. Hunsicker,†

Sanjeev Sethi,¶ and Richard J.H. Smith*†‡

*Molecular Otolaryngology and Renal Research Laboratories, Carver College of Medicine, University of Iowa, IowaCity, Iowa; †Division of Nephrology, Department of Internal Medicine, Carver College of Medicine, University of Iowa,Iowa City, Iowa; ‡Division of Nephrology, Department of Pediatrics, Carver College of Medicine, University of Iowa,Iowa City, Iowa; §Department of Pathology, Carver College of Medicine, University of Iowa, Iowa City, Iowa; |CelldexTherapeutics Inc., Needham, Massachusetts; and ¶Department of Laboratory Medicine and Pathology, Mayo Clinic,Rochester, Minnesota

ABSTRACTDense deposit disease (DDD) and C3 glomerulonephritis (C3GN) are widely recognized subtypes of C3glomerulopathy. These ultra-rare renal diseases are characterized by fluid-phase dysregulation of thealternative complement pathway that leads to deposition of complement proteins in the renal glomerulus.Disease triggers are unknown and because targeted treatments are lacking, progress to end stage renalfailure is a common final outcome. We studied soluble CR1, a potent regulator of complement activity, totest whether it restores complement regulation in C3 glomerulopathy. In vitro studies using sera frompatients with DDD showed that soluble CR1 prevents dysregulation of the alternative pathway C3 con-vertase, even in the presence of C3 nephritic factors. In mice deficient in complement factor H and trans-genic for human CR1, soluble CR1 therapy stopped alternative pathway activation, resulting innormalization of serum C3 levels and clearance of iC3b from glomerular basement membranes. Short-term use of soluble CR1 in a pediatric patient with end stage renal failure demonstrated its safety andability to normalize activity of the terminal complement pathway. Overall, these data indicate that solubleCR1 re-establishes regulation of the alternative complement pathway and provide support for a limitedtrial to evaluate soluble CR1 as a treatment for DDD and C3GN.

J Am Soc Nephrol 24: 1820–1829, 2013. doi: 10.1681/ASN.2013010045

Dense deposit disease (DDD) and C3 glomerulone-phritis (C3GN) are twowidely recognized subtypes ofC3 glomerulopathy (C3G).1,2 These ultra-rare renaldiseases are caused by fluid-phase dysregulation ofthe C3 convertase of the alternative pathway (AP)of complement, with variable concomitant dysregu-lation of the C5 convertase. Consistent with comple-ment-mediated disease acting through the AP, C3G isstrongly positive for C3 and notably negative for Igsby immunofluorescence microscopy.2 Electronmicroscopy distinguishes DDD from C3GN, withthe former characterized by pathognomonicelectron-dense transformation of the lamina densaof the glomerular basement membrane (GBM).3 InC3GN, the electron microscopy deposits are lighter

in color, and are more often mesangial and/or sub-endothelial, intramembranous, and subepithelial inlocation.4 In both diseases, mass spectroscopy of la-ser dissected glomeruli is highly enriched for proteinsof the AP and terminal complement cascade.4,5

Received January 10, 2013. Accepted May 14, 2013.

Y.Z. and C.M.N are co-first authors.

Published online ahead of print. Publication date available atwww.jasn.org.

Correspondence: Dr. Richard J.H. Smith, Molecular Otolaryn-gology and Renal Research Laboratories, Carver College ofMedicine, University of Iowa, 200 Hawkins Drive, Iowa City, IA52242. Email: richard-smith@uiowa.edu

Copyright © 2013 by the American Society of Nephrology

1820 ISSN : 1046-6673/2411-1820 J Am Soc Nephrol 24: 1820–1829, 2013

Although long-term outcome data are not available for C3GN,nearly half of all DDDpatients progress to end stage renal failure(ESRF) within 10 years of diagnosis.6,7 In virtually all cases ofDDD, transplantation is associated with histologic recurrence,explaining the 5-year graft failure rate of 50%.7,8 There are notarget-specific treatments for C3G; however, its pathophysiologysuggests that therapeutic approaches to restore C3 convertasecontrol, impair C3 convertase activity, or remove C3 breakdownproducts from the circulation warrant consideration.1,9

Similar to human DDD, the complement factor H (Cfh)–deficient mouse, reclassified recently as a murine model ofC3GN, accumulates C3 fragments within glomeruli as the firstpathologic abnormality to develop.10 Systemic administrationof either murine or human complement factor H (fH) to theCfh2/2 mouse rapidly reverses both the renal deposition ofC3 and its plasma depletion, suggesting that fH replacementtherapy may be an effective therapy to restore the underlyingprotein deficiency and correct the disease process in C3G pa-tients with CFHmutations.11,12 However, it is unlikely that fHadministration would be therapeutically successful in the ab-sence of CFH-inactivating mutations. For example, fH wouldbe ineffective in C3G patients carrying C3 mutations that ren-der C3 convertase fH resistant or in the presence of autoanti-bodies to C3 convertase that block regulation by fH.13,14

Among the non-fH proteins that regulate C3 convertase issoluble CR1. CR1 is a cell-surface glycoprotein expressed onerythrocytes, monocytes, neutrophils, B cells, some T cells,follicular dendritic cells, and podocytes, and modulates thecomplement cascade at multiple levels (Supplemental Figures1 and 2).15–18 It is composed of 30 short consensus repeats(SCRs) linked to transmembrane and cytosolic domains.19 Onthe basis of SCR sequence homology, the 30 extracellular SCRscan be grouped into four long homologous repeats, which aresequentially lettered A through D. These long homologousrepeats give CR1 C3 and C5 convertase-controlling activity.CR1 is also the only cofactor of factor I (fI) to promote cleav-age of inactive C3b and inactive C4b into their smaller proteinfragments, C3dg and C4d.20–22

The soluble formofCR1 (sCR1) is present in the circulationat extremely low concentrations precluding any recognizedphysiologic significance. However, super-physiologic single-dose treatment with sCR1 has been used in .500 patients(infants and adults) undergoing a variety of cardiac surgeriesor after myocardial infarctions to arrest complement activa-tion.23–26 These studies showed that sCR1 was well toleratedwith no adverse events clearly or consistently associated withits use. Importantly, the possible development of anti-sCR1antibodies was monitored but never observed in .350patients.23,26

RESULTS

We assessed the efficacy of sCR1 as a complement regulator invitro in normal and DDD sera. In normal pooled human sera,

it prevented classic pathway (CP) and AP complement activa-tion (IC50 values of 2.5560.55 nM and 0.7160.08 nM, re-spectively; Figure 1A). Confirmatory hemolytic assays wereperformed with rabbit and sheep erythrocytes. Rabbit eryth-rocytes are a complement-activating surface in human sera;however, lysis could be prevented by the addition of sCR1(IC50=29.4664.64 nM). In comparison, fH did not preventhemolysis even at high concentrations (Figure 1B). Sheeperythrocytes do not activate complement in normal sera;however, hemolysis occurred when tested against DDD sera.The addition of sCR1 restored AP control in a dose-dependentmanner (Figure 1C) and prevented hemolysis evenwhenDDDsera contained C3 convertase-stabilizing autoantibodies calledC3Nefs (Figure 1D).

Wenextmeasured the systemic effect of sCR1 in vivo in bothCfh2/2 and Cfh2/2/huCR1-Tg mice (the latter express CR1on erythrocytes only) by introducing sCR1 by the tail vein orintraperitoneally, respectively, after first verifying that it pre-vented hemolysis of rabbit erythrocytes in mouse sera(IC50=6.1560.18 nM; Figure 2 and Supplemental Figure 3).After a single tail vein injection of sCR1 in Cfh2/2 mice(50mg/kg, n=5; Figure 2A, blue), serumC3 levels rose drama-tically within 24 hours from 12.03612.29 mg/L to219.77642.32 mg/L but fell to near preinjection levels by 48hours (41.84624.29 mg/L; Figure 2B, blue). Single-dose in-traperitoneal injections in Cfh2/2/huCR1-Tg mice at concen-trations of 5 mg/kg, 25 mg/kg, or 50 mg/kg showed thatchanges in serum C3 levels were sCR1 dose dependent (Figure2, A and B, red). In all sCR1-treated animals, glomerular C3deposition was markedly reduced (Figure 2C, top panel) withonly trace glomerular staining for sCR1 (Figure 2C, bottompanel). In PBS-treated and unmanipulated animals, glomer-ular C3 deposition was clearly evident (Figure 2C, top panel)and sCR1 was undetectable (Figure 2C, bottom panel).

We calculated the serum t1/2 of sCR1 in mice to be approx-imately 18 hours. On the basis of the dose-response curves tosingle intraperitoneal injections, we treated four Cfh2/2/huCR1-Tg mice with three intraperitoneal injections(25 mg/kg, n=2mice; 50 mg/kg, n=2 mice) at 24-hour intervals(Figure 3A). In all sCR1-treated animals, serum C3 levels roseto near normal by 12 hours and remained high throughout theexperiment (Figure 3B). Glomerular immunofluorescencestudies showed that there was a significant decrease in newC3 deposition, with clearance of old C3 demonstrated by themarked reduction in C3c staining. C3d staining also decreasedand sCR1 staining was absent (Figure 3D). To verify that con-trol of the C3 convertase had been restored and that it was anincrease in serum C3 and not its degradation products that wewere measuring, we performed Western blotting under reduc-ing conditions with a C3a-chain–specific antibody to docu-ment the appearance of the 106-kD C3a chain. In PBS-treatedand untreated animals, serum C3 levels remained nearly un-detectable and the 106-kDa C3a chain was not seen (Figure3C). Thus, sCR1 treatment restored fluid-phase AP regulationin these murine models of C3G.

J Am Soc Nephrol 24: 1820–1829, 2013 Soluble CR1 in C3 Glomerulopathy 1821

www.jasn.org BASIC RESEARCH

On the basis of these data, we sought permission fromthe US Food and Drug Administration (FDA) for a limitedmultidose safety trial with sCR1 when presented with a C3Gpatient in ESRF. The patientwas an 8-year-old girl with biopsy-proven DDD (Figure 4A) who had presented 3 months earlierwith lower extremity swelling, nephritic urine sediment, ne-phrotic-range proteinuria, positive C3Nefs, and a low serumC3 level (21.6 mg/dl; normal range, 90–180 mg/dl). She tran-sitioned to rapidly progressive GN (eGFR ,10 ml/min per1.73 m2) necessitating dialysis. Approval for a seven-dose trialof sCR1was obtained from the FDA and from theUniversity ofIowa Institutional Review Board.

The patient received a 10 mg/kg loading dose of sCR1followed by six maintenance doses at 5 mg/kg every 48 hours.

During treatment, serum C3 rose to a peak of 48 mg/dl butdropped to pretreatment levels by the end of the trial. SolubleC5b-9 levels were elevated before treatment, normalized afterthe sixth dose of sCR1, and became abnormal after terminatingtherapy (Figure 4D). Biopsies taken before and after treatmentwere similar (Figure 4, B and C, and Table 1). The patientcontinues on dialysis at this time.

Although the trial was very short, there were no adverseeffects associated with the administration of sCR1 and im-munogenicity was not detected. Assays for anti-sCR1 anti-bodies on plasma obtained before dosing and at days 2, 4, 6, 8,10, 12, and 14 and 1month after dosing remained consistentlynegative (A450 values #0.076 at dilutions of 1:50 at all timepoints; positive control, A450 1.10, 0.68, and 0.40 at 1:100,

Figure 1. sCR1 prevents complement activation. (A) The ability of sCR1 to prevent activation of the alternative pathway (AP, red) andclassical pathway (CP, blue) was measured by Wieslab complement assay using pooled normal human serum (PNHS). Results are expressedas percent activation by PNHS in presence of increasing concentrations of sCR1. Half-maximal inhibitory concentration (IC50) of sCR1 for theCP and AP are 2.556 0.55 nM and 0.716 0.08 nM, respectively. (B) Rabbit erythrocytes are normally an activating surface for the AP. sCR1(red) protects rabbit erythrocytes from AP-mediated complement lysis (IC50 = 29.46 6 4.64 nM) while FH (blue) does not. (C) Sheeperythrocytes normally do not activate the human complement system, but DDD patient serum (20% v/v) causes hemolysis due to thepresence of C3 convertase-stabilizing C3 nephritic factors. sCR1 suppresses C3Nef activity in a dose-dependent manner (serum frompatient DDD-10). (D) In all DDD sera (20% v/v) tested, sheep erythrocyte hemolysis was reduced by sCR1. All patients except DDD-07 and08 were positive for C3Nefs. Data represent mean 6 SD of triplicates (dark blue, 0nM sCR1; light blue, 160nM sCR1).

1822 Journal of the American Society of Nephrology J Am Soc Nephrol 24: 1820–1829, 2013

BASIC RESEARCH www.jasn.org

Figure 2. Human sCR1 restores serum C3 levels and reduces C3 glomerular staining in Cfh2/2 and Cfh2/2/huCR1-Tg mice. (A) Plasmalevels of sCR1 dropped rapidly following a single tail-vein injection in 5 Cfh2/2 mice (blue lines; t1/2 approximately 18 hours). Aftera single IP injection of sCR1 (5 mg/kg, 25 mg/kg or 50 mg/kg) in Cfh2/2/huCR1-Tg mice, plasma levels rose and then fell slowly (redlines). Each line represents a single mouse. (B) Serum C3 levels in Cfh2/2mice (blue lines) increased 24 hours after a single tail-veininjection of sCR1 (50 mg/kg) but returned to baseline 48 hours later. Serum C3 levels in Cfh2/2/huCR1-Tg mice (red lines) also in-creased 12 hours after a single IP injection (5 mg/kg, 25 mg/kg, or 50 mg/kg) and closely followed the sCR1 concentration curves in A.All data (A and B) represent the mean of triplicate assays. (C) C3 glomerular deposition decreased after 48 hours in sCR1-injected Cfh2/2

mice (50 mg/kg by tail-vein injection) but not in PBS-injected or non-injected controls (top panel). Trace glomerular staining for human CR1(CD35) was seen in sCR1-injected mice 48 hours after treatment (bottom).

J Am Soc Nephrol 24: 1820–1829, 2013 Soluble CR1 in C3 Glomerulopathy 1823

www.jasn.org BASIC RESEARCH

Figure 3. Multi-dose sCR1 treatment clears iC3b glomerular deposition in mouse. (A) Daily dosing (arrows) maintained sCR1 serumlevels (blue, 25mg/kg; red, 50mg/kg, black, PBS). (B) Serum C3 levels in Cfh2/2/huCR1-Tg mice increased and remained high with dailyIP dosing (0, 24, and 48 hours) of 25 (blue) or 50 (red) mg/kg sCR1 (black, PBS; normal range: 300;1500 mg/L). Data represent themean of triplicate assays. (C) Intact C3a (992 amino acids) was detected by Western blotting in reducing conditions using a murine C3a-specific antibody. The 106-kDa band (equivalent to the C3a chain in wild-type mice) appeared 12 hours after sCR1 treatment but wasabsent before treatment and in untreated controls. (D) IF of glomeruli harvested 60 hours after injection with relative fluorescence

1824 Journal of the American Society of Nephrology J Am Soc Nephrol 24: 1820–1829, 2013

BASIC RESEARCH www.jasn.org

1:200, and 1:400, respectively; negative control, 0.05 at 1:50).Pharmacokinetic data from measured sCR1 serum concentra-tions in this patient were in good agreement with dose-basedsimulations using pharmacokinetic data from adult patients un-dergoing high-risk cardiac surgery (Supplemental Figure 4).23

DISCUSSION

The in vitro activity of sCR1 in complement inhibition is de-pendent on the specific conditions of the assay and varies withthe concentration and hemolytic potential of the serum com-plement source and the efficiency of the complement-activat-ing mechanisms. In the assays we used, sCR1 prevented C3convertase activity in normal (rabbit erythrocyte hemolyticassay) and pathologic conditions (sheep erythrocyte hemo-lytic assay with DDD sera). In vivo experiments in two mousemodels of C3G confirmed the in vitro data—sCR1 stopped APdysregulation and restored plasma C3 levels to normal. Thesechanges were accompanied by reduced deposition of new iC3band clearance of old iC3b in the GBM.

Our results are consistent with the known role that CR1plays as a central complement regulator of both the C3 and C5convertases. In addition to regulating these convertases, CR1 isthe only cofactor of fI that can cleave iC3b into smallerfragments (C3c and the thioester-containing fragment C3dg),thus explaining the rapid clearance of iC3b from theGBM.Theslower clearance of C3d is consistent with its surface-bindingproperties (Figure 3D). Thus, by bringing complement dysreg-ulation under control, the deposition of new iC3b is arrested(Figure 3).

These observations provide strong evidence that iC3b isdeposited in the GBM and is an important component of theglomerular dense deposits, a conclusion consistent withwork done by Pickering and colleagues demonstrating thatthe presence of fI is an absolute requirement inCfh2/2mice forthe development of a C3GN renal phenotype. In mice defi-cient in both fH and fI (Cfh2/2.Cfi2/2 mice), although excessC3b is present in the circulation, no iC3b forms and GBMdense deposits are not seen.27 The Cfh2/2/huCR1-Tg mousealso provides an excellent animal model in which to studylong-term gene therapy to assess the effect of constitutive he-patic expression of sCR1 on the renal phenotype.

The patient with DDD who we treated received only sevendoses of sCR1; however, we demonstrated that multidosetreatment was safe and nonimmunogenic in this limited study.We did not expect to, nor did we, see a histologic improvementin the renal biopsy results. sCR1 administrationwas associatedwith a transient improvement in serum C3 and soluble C5b-9normalized. These biomarker changes were expected based on

in vitro and murine data and support a longer trial using sCR1in a carefully selected patient population to evaluate the po-tential of sCR1 as a specific therapy for C3G. Therapeuticentities that incorporate select domains of sCR1 also warrantinvestigation.

CONCISE METHODS

PatientAn 8-year-old patient with rapidly progressive DDD was enrolled

under an FDA-approved compassionate-use investigational new drug

studydesigned toadminister sevendosesof sCR1under IRB-approved

guidelines. DDD was diagnosed by renal biopsy.

Patient and Normal SeraTen patients with biopsy-proven DDD consented to provide sera

under IRB-approved guidelines. All patients were positive for C3Nefs

except for patients DDD-07 and DDD-08. Patient DDD-07 carries a

genetic rearrangement between CFHR2 and CFHR5, and patient

DDD-06 carries a genetic variant in CFHR5 (c.1541 T.G; p.Met514-

Arg). No other genetic mutations were identified in CFH, CFI, CFB,

MCP, or C3. Pooled normal human sera were obtained from Inno-

vative Research (Novo, MI); pooled murine sera were collected from

five 2-month-old C57BL/6 animals by cardiac puncture.

sCR1sCR1 (also called TP10 or CDX-1135) was produced by recombinant

DNA technology usingChinese hamster ovary cells and purified using

standard filtration and chromatography methods as a single-chain

polypeptide of 1931 amino acids with a protein molecular mass of

212 kD (Celldex Therapeutics, Needham, MA). Due to N-linked glyco-

sylation, the final total molecular mass was 247 kD.28

Complement Activity AssaysAP and CP activity were evaluated using the appropriate Wieslab

complement assay (Wieslab AB, Lund, Sweden) following the

manufacturer’s protocols (serum dilution 1:34 for AP and 1:100 for

CP). To measure complement inhibitory effects, increasing amounts

of sCR1 were added to diluted sera before conducting these assays.

Hemolytic AssaysAP hemolytic activity was measured using rabbit or sheep erythro-

cytes. Rabbit erythrocytes activate AP-mediated lysis in human and

mouse sera; sheep erythrocytes do not. For the rabbit erythrocyte

hemolytic assay, increasing concentrations of sCR1 (0.2 nM to

200 nM) or fH (0.2 nM to 2000 M) were mixed with normal human

or mouse sera (20% v/v) and incubated for 30 minutes at 37°C with

13108 rabbit erythrocytes in the presence of 10 mM EGTA/0.15 mM

Mg2+/gelation veronal buffer (GVB) or EDTA-GVB as a

quantitation (averagedcontrol value set to100%with theexceptionofCD35whereaveragedC3controlwas set to100%; scalebar: averageof 10 glomeruli6 SD). From top to bottom: C3 was dramatically decreased; CD35 (CR1) was absent; C3d decreased significantly with thehigher dose of sCR1; C3c was greatly reduced. *P,0.05, **P,0.001; controls versus experiments.

J Am Soc Nephrol 24: 1820–1829, 2013 Soluble CR1 in C3 Glomerulopathy 1825

www.jasn.org BASIC RESEARCH

Figure 4. sCR1 restores soluble MAC levels in a DDD patient. (A) Biopsy at presentation showed a proliferative-appearing glomerulus withcellular crescents, endocapillary and mesangial hypercellularity and capillary loop “double contours” (light microscopy [LM] 3400). Noglobal glomerulosclerosis was noted, and interstitial fibrosis and tubular atrophy were minimal. Glomerular “ribbon-like” electron-densedeposits were present on electron microscopy (EM,38200) with discrete, coarsely granular C3 immunofluorescence of the mesangium and

1826 Journal of the American Society of Nephrology J Am Soc Nephrol 24: 1820–1829, 2013

BASIC RESEARCH www.jasn.org

nonhemolytic control. The reaction was stopped by adding 150 ml of

20 mM EDTA-GVB. After centrifugation, the supernatant was trans-

ferred to a clean 96-well plate and absorbance was recorded at 415

nm. Percent lysis was calculated as a fraction of the maximum A415

absorbance in the absence of sCR1 [(A415sCR1-EGTA- A415EDTA)/

(A415EGTA) 3 100].

To test whether sCR1 prevents C3Nef stabilization of C3 con-

vertase, 10 ml of patient serum was added to 10 ml of sheep erythro-

cytes (13109/ml) coated with preformed C3 convertase in a total 50

ml (EGTA-Mg-GVB) reaction as described.29 The mixture was al-

lowed to decay at 30°C (water bath) for 20 minutes. Hemolysis was

assayed by adding 50 ml of rat serum (1:9 diluted in GVB-EDTA

buffer) as a source of C5-9. Sheep red blood cells were lysed in the

presence of nondecayed C3 convertase stabilized by C3Nefs. The re-

action was stopped and the percentage of hemolysis was measured as

described above. In parallel tests, varying concentrations of sCR1

(0 nM, 40 nM, 120 nM, 160 nM) were added to patient serum before

mixing with sheep erythrocytes and incubating on ice for 15minutes.

Repetition of this experiment was done with sera from 10 patients

with DDD.

MiceThe Cfh2/2 mutant mouse, obtained from Dr. Matthew Pickering,

was generated as described.10 On this background, human CR1 was

introduced to create the Cfh2/2/huCR1-Tg mutant.30 The latter

mouse line expresses CR1 only on mouse erythrocytes. The line

was used to avoid any murine immunoresponse. All procedures

were approved by the Institutional Animal Care and Use Committee

at the University of Iowa.

sCR1 and C3 ConcentrationssCR1 concentrations were measured by sandwich ELISA in a micro-

titer plate format using twomouse mAbs specific for CR1. Microtiter

plates were coated with mAb 6B1.H12; the detection antibody was

horseradish peroxidase–conjugated mAb 4D6.1.31 Murine total C3

serum levels weremeasured using a two-site ELISA (Kamiya Biomed-

ical, Seattle, WA); human total C3 was measured by radial immuno-

diffusion (The Binding Site, San Diego, CA). Data were collected in

three independent assays and expressed as the mean for each group.

Immunohistochemical AnalysesKidneys were harvested at the time of euthanasia and imbedded in

Tissue-Tek OCT medium for routine processing (Sakura Finetek,

Torrance, CA). For glomerular C3 deposition, sections were stained

with FITC-conjugated mouse C3 antibody (MP Biomedicals, Solon,

OH) at a dilution of 1:800 for 1 hour. For glomerular sCR1, C3c,

and C3d staining, sections were treated with goat anti-human CR1

antibody (1:400; Santa Cruz Biotechnology, Santa Cruz, CA), rabbit

anti-human C3c (1:8000; Abcam, Cambridge, MA), or goat anti-

mouse C3d (1:400; R&D Systems, Minneapolis, MN) for 1 hour, re-

spectively, followed by staining with the corresponding Alexa Fluor

488 (1:500; Molecular Probes, Eugene, OR) for 1 hour at room tem-

perature. Fluorescence images were acquired on a Leica model TCS

SP5 laser system (Leica Microsystems, Buffalo Grove, IL). Relative

fluorescence intensities (average of 10 glomeruli; 6 SD) were quan-

tified by ImageJ software (http://rsbweb.nih.gov/ij/; National Insti-

tutes of Health, Bethesda, MD).

Immunoblot AnalysesMouse sera were diluted 1:40 in Laemmli buffer, separated by 7.5%

SDS-PAGE, and transferred to nitrocellulose membranes (Whatman

Schleicher and Schuell Inc., Dassel, Germany). Membranes were

blockedwith5%skimmilk inPBSat roomtemperature for 30minutes

and incubated with anti-mouse C3 antibody (1:500; in-house

antibody targeting the fI cleavage site on C3). Each membrane was

washed three times with 0.05% Tris-buffered saline Tween-20

followed by incubation with horseradish peroxidase–conjugated

anti-mouse secondary antibody (Jackson ImmunoResearch, Inc.,

West Grove, PA). Specific protein bands were visualized with an En-

hanced Chemiluminescent System (GE Healthcare, Piscataway, NJ).

sCR1 Dose SimulationsPharmacokinetic data fromadult cardiac surgery patients treatedwith

sCR1 were used to estimate sCR1 serum concentrations over time

using empirical calculations that assumed concentrations from

multiple doses to be additive because sCR1 pharmacokinetics has

capillary loops (IF, 3400). (B) The pre-sCR1 biopsy, 6 weeks after the initial biopsy, showed persistent glomerular hypercellularity andcapillary loop “double contours” (first LM, 3400), and a glomerular fibrocellular crescent (second LM, 3400) consistent with diseaseprogression to chronicity; moderate interstitial fibrosis and tubular atrophy were present. IF showed “chunky” C3 staining in the glo-merular mesangium and capillary loops (3400). (C) Post-sCR1 biopsy, 6 days after the final dose, showed glomerular fibrocellular andfibrous crescents with periglomerular fibrosis (LM,3200). The ribbon-like capillary loop andmesangial electron dense deposits remainedunchanged (EM,38200). IF (3400) continued to show bright C3 staining because the biopsy was done 6 days after sCR1 was stopped. (D)Immediately before receiving sCR1 serum C3 was low and sMAC was high. Although the C3 level initially doubled during treatment ittrended back to the pre-treatment level. SolubleMAC normalized after the sixth dose (,30mg/cL) but within 96 hours after termination oftherapy was abnormal (asterisks, levels measured at 48 and 96 hours after the last dose of sCR1; red dotted line, normal level of sMAC;green dotted line low normal of plasma C3; sMAC in mg/cl, C3 in mg/dl, sCR1 in mg/ml to allow representation on same graph; sCR1,1 mg/ml = 4nM).

Table 1. Patient biopsies at presentation and before andafter treatment with sCR1

Histology PresentationBefore

TreatmentAfter

Treatment

Crescents (n) 21 of 25 6 of 18 3 of 13Glomerularsclerosis (n)

0 of 25 9 of 18 3 of 13

Interstitial fibrosis Minimal Moderate ModerateC3 immunofluorescence 2–3+ 2–3+ 2–3+

J Am Soc Nephrol 24: 1820–1829, 2013 Soluble CR1 in C3 Glomerulopathy 1827

www.jasn.org BASIC RESEARCH

been shown to be dose proportional (dose versus area under the curve

are linear).23 Averaged data from 26 cardiac surgery patients

receiving a single 10 mg/kg bolus of sCR1 were used to simulate

the dosing regimen used in the single DDDpatient. The dose regimen

was designed to keep trough levels of sCR1 above 20 mg/ml through-

out the dosing period.

Immunogenicity AssaysPatient plasma samples were screened for anti-sCR1 antibodies using

a microtiter plate ELISA format. Microtiter plates were coated with

sCR1 and blocked with BSA in PBS. Patient samples were diluted 1:50

in PBS 10 mM EDTA 0.5% Tween and incubated in coated plates.

Antibodies were detected using horseradish peroxidase-conjugated

F(ab)2 goat anti-human IgG antibody (KPL Inc., Gaithersburg, MD).

Normal human serum was used as a negative control. Knops-McCoy

blood group antiserum, which is specific for CR1, served as a positive

control.32 Positive patient samples were confirmed specific for sCR1

by preincubating with excess sCR1 and rerunning the assay.

Statistical AnalysesIC50 values were calculated by GraphPad Prism 5 software (GraphPad

Software, Inc., La Jolla, CA). Statistical significance was calculated

with the t test.

ACKNOWLEDGMENTS

We are grateful to those patients with DDD whose participation

made this research possible. The Cfh2/2 and Cfh2/2/huCR1-Tg mu-

tant mice were kindly provided by Drs. Matthew Pickering and Rick

Quigg.

This work was supported in part by National Institutes of Health

Grant DK074409 to R.J.H.S. and S.S.

DISCLOSURESH.C.M., R.A.H., and L.J.T. own stock and/or options and are employees of

Celldex Therapeutics.

REFERENCES

1. Smith RJ, Harris CL, Pickering MC: Dense deposit disease. Mol Im-

munol 48: 1604–1610, 20112. Sethi S, Nester CM, Smith RJ: Membranoproliferative glomerulone-

phritis and C3 glomerulopathy: Resolving the confusion. Kidney Int 81:434–441, 2012

3. Smith RJ, Alexander J, Barlow PN, Botto M, Cassavant TL, Cook HT, deCórdoba SR, Hageman GS, Jokiranta TS, Kimberling WJ, Lambris JD,Lanning LD, Levidiotis V, Licht C, Lutz HU, Meri S, PickeringMC, QuiggRJ, Rops AL, Salant DJ, Sethi S, Thurman JM, Tully HF, Tully SP, van derVlag J, Walker PD, Würzner R, Zipfel PF; Dense Deposit Disease FocusGroup: New approaches to the treatment of dense deposit disease.J Am Soc Nephrol 18: 2447–2456, 2007

4. Sethi S, Fervenza FC, Zhang Y, Zand L, Vrana JA, Nasr SH, Theis JD,Dogan A, Smith RJH: C3 glomerulonephritis: Clinicopathological

findings, complement abnormalities, glomerular proteomic profile,treatment, and follow-up. Kidney Int 82: 465–473, 2012

5. Sethi S, Gamez JD, Vrana JA, Theis JD, BergenHR3rd, Zipfel PF, DoganA, Smith RJ: Glomeruli of dense deposit disease contain components ofthe alternative and terminal complement pathway. Kidney Int 75: 952–960, 2009

6. Lu DF, McCarthy AM, Lanning LD, Delaney C, Porter C: A descriptivestudy of individuals with membranoproliferative glomerulonephritis.Nephrol Nurs J 34: 295–302, quiz 303, 2007

7. Lu DF, Moon M, Lanning LD, McCarthy AM, Smith RJ: Clinical featuresand outcomes of 98 children and adults with dense deposit disease.Pediatr Nephrol 27: 773–781, 2012

8. Braun MC, Stablein DM, Hamiwka LA, Bell L, Bartosh SM, Strife CF:Recurrence of membranoproliferative glomerulonephritis type II in re-nal allografts: The North American Pediatric Renal Transplant Co-operative Study experience. J Am Soc Nephrol 16: 2225–2233, 2005

9. Nester CM, Smith RJH: Treatment options for C3 glomerulopathy.CurrOpin Nephrol Hypertens 22: 231–237, 2013

10. Pickering MC, Cook HT, Warren J, Bygrave AE, Moss J, Walport MJ,Botto M: Uncontrolled C3 activation causes membranoproliferativeglomerulonephritis in mice deficient in complement factor H. Nat

Genet 31: 424–428, 200211. Fakhouri F, de Jorge EG, Brune F, Azam P, Cook HT, Pickering MC:

Treatment with human complement factor H rapidly reverses renalcomplement deposition in factor H-deficient mice. Kidney Int 78: 279–286, 2010

12. Paixão-Cavalcante D, Hanson S, Botto M, Cook HT, Pickering MC:Factor H facilitates the clearance of GBMbound iC3b by controlling C3activation in fluid phase. Mol Immunol 46: 1942–1950, 2009

13. Martínez-Barricarte R, Heurich M, Valdes-Cañedo F, Vazquez-Martul E,Torreira E, Montes T, Tortajada A, Pinto S, Lopez-Trascasa M, MorganBP, Llorca O, Harris CL, Rodríguez de Córdoba S: Human C3 mutationreveals a mechanism of dense deposit disease pathogenesis and pro-vides insights into complement activation and regulation. J Clin Invest

120: 3702–3712, 201014. Zhang Y, Meyer NC, Wang K, Nishimura C, Frees K, Jones M, Katz LM,

Sethi S, Smith RJ: Causes of alternative pathway dysregulation in densedeposit disease. Clin J Am Soc Nephrol 7: 265–274, 2012

15. Fang Y, Xu C, Fu YX, Holers VM, Molina H: Expression of complementreceptors 1 and 2 on follicular dendritic cells is necessary for the gen-eration of a strong antigen-specific IgG response. J Immunol 160:5273–5279, 1998

16. Kazatchkine MD, Fearon DT: Deficiencies of human C3 complementreceptors type 1 (CR1, CD35) and type 2 (CR2, CD21). Immunodefic

Rev 2: 17–41, 199017. Rødgaard A, Christensen LD, Thomsen BS, Wiik A, Bendixen G:

Complement receptor type 1 (CR1, CD35) expression on peripheral Tlymphocytes: Both CD4- and CD8-positive cells express CR1. Com-

plement Inflamm 8: 303–309, 199118. Weiss L, Fischer E, Haeffner-Cavaillon N, Jouvin MH, Appay MD,

Bariety J, KazatchkineM: The human C3b receptor (CR1). Adv Nephrol

Necker Hosp 18: 249–269, 198919. Klickstein LB,WongWW,Smith JA,Weis JH,WilsonJG,FearonDT:Human

C3b/C4b receptor (CR1). Demonstration of long homologous repeatingdomains thatarecomposedof theshort consensus repeats characteristicsofC3/C4 binding proteins. J Exp Med 165: 1095–1112, 1987

20. Morgan HP, Schmidt CQ, Guariento M, Blaum BS, Gillespie D, HerbertAP, Kavanagh D, Mertens HD, Svergun DI, Johansson CM, Uhrín D,Barlow PN, Hannan JP: Structural basis for engagement by complementfactor H of C3b on a self surface. Nat Struct Mol Biol 18: 463–470, 2011

21. Medof ME, Nussenzweig V: Control of the function of substrate-boundC4b-C3b by the complement receptor Cr1. J Exp Med 159: 1669–1685, 1984

22. Masaki T, Matsumoto M, Nakanishi I, Yasuda R, Seya T: Factor I-de-pendent inactivation of human complement C4b of the classical

1828 Journal of the American Society of Nephrology J Am Soc Nephrol 24: 1820–1829, 2013

BASIC RESEARCH www.jasn.org

pathway by C3b/C4b receptor (CR1, CD35) and membrane cofactorprotein (MCP, CD46). J Biochem 111: 573–578, 1992

23. Lazar HL, Bokesch PM, van Lenta F, Fitzgerald C, Emmett C, Marsh HCJr, Ryan U; OBE and the TP10 Cardiac Surgery Study Group: Solublehuman complement receptor 1 limits ischemic damage in cardiac sur-gery patients at high risk requiring cardiopulmonary bypass. Circula-tion 110[Suppl 1]: II274–II279, 2004

24. Lazar HL, Keilani T, Fitzgerald CA, Shapira OM, Hunter CT, Shemin RJ,Marsh HC Jr, Ryan US; TP10 Cardiac Surgery Study Group: Beneficialeffects of complement inhibition with soluble complement receptor 1(TP10) during cardiac surgery: Is there a gender difference?Circulation116[Suppl]: I83–I88, 2007

25. Li JS, Jaggers J, Anderson PA: The use of TP10, soluble complementreceptor 1, in cardiopulmonary bypass. Expert Rev Cardiovasc Ther 4:649–654, 2006

26. Li JS, Sanders SP, Perry AE, Stinnett SS, Jaggers J, Bokesch P, ReynoldsL, Nassar R, Anderson PA: Pharmacokinetics and safety of TP10, solublecomplement receptor 1, in infants undergoing cardiopulmonary by-pass. Am Heart J 147: 173–180, 2004

27. Rose KL, Paixao-Cavalcante D, Fish J, Manderson AP, Malik TH, BygraveAE, Lin T, Sacks SH,WalportMJ, CookHT, BottoM, PickeringMC: FactorI is required for the development of membranoproliferative glomerulo-nephritis in factor H-deficient mice. J Clin Invest 118: 608–618, 2008

28. Weisman HF, Bartow T, Leppo MK, Marsh HC Jr, Carson GR, ConcinoMF, Boyle MP, Roux KH, Weisfeldt ML, Fearon DT: Soluble humancomplement receptor type 1: In vivo inhibitor of complement sup-pressing post-ischemic myocardial inflammation and necrosis. Science249: 146–151, 1990

29. West CD: A hemolytic method for the measurement of nephritic factor.J Immunol Methods 335: 1–7, 2008

30. Alexander JJ, Hack BK, JacobA, ChangA, HaasM, Finberg RW,QuiggRJ: Abnormal immune complex processing and spontaneous glo-merulonephritis in complement factor H-deficient mice with humancomplement receptor 1 on erythrocytes. J Immunol 185: 3759–3767,2010

31. Nickells M, Hauhart R, KrychM, Subramanian VB, Geoghegan-Barek K,Marsh HC Jr, Atkinson JP: Mapping epitopes for 20 monoclonal anti-bodies to CR1. Clin Exp Immunol 112: 27–33, 1998

32. Moulds JM,NickellsMW,Moulds JJ, BrownMC, Atkinson JP: The C3b/C4b receptor is recognized by the Knops, McCoy, Swain-langley, andYork blood group antisera. J Exp Med 173: 1159–1163, 1991

This article contains supplemental material online at http://jasn.asnjournals.org/lookup/suppl/doi:10.1681/ASN.2013010045/-/DCSupplemental.

J Am Soc Nephrol 24: 1820–1829, 2013 Soluble CR1 in C3 Glomerulopathy 1829

www.jasn.org BASIC RESEARCH