Glomerulonephritis With Isolated C3 Deposits andMonoclonal Gammopathy: A Fortuitous Association?Frank Bridoux,*†¶ Estelle Desport,*¶ Veronique Fremeaux-Bacchi,‡ Christine Fen Chong,* Jean-Marc Gombert,§¶

Corinne Lacombe,§¶ Nathalie Quellard, and Guy Touchard*¶

SummaryBackground and objectives Glomerular deposition of monoclonal Ig has been exceptionally described as thecause of membranoproliferative glomerulonephritis, through activation of the complement alternative path-way (CAP).

Design, setting, participants, & measurements We retrospectively studied six adults with monoclonal gam-mopathy and glomerulonephritis (GN) characterized by isolated C3 deposits.

Results All patients presented with hematuria, associated with chronic renal failure and proteinuria in fivepatients, three of whom had nephrotic syndrome. Five patients had monoclonal gammopathy of undeter-mined significance and one had smoldering myeloma. The serum monoclonal IgG (� four of six, � two ofsix) was associated with light chain (LC) proteinuria in five patients. Four patients had low serum C3and/or factor B levels. C4, factor H (CFH), and I protein levels were normal in five of five patients; nonehad detectable C3NeF. IgG anti-CFH activity was positive in one case. No mutations in CFH, CFI, and MCPgenes were identified in four of four patients. Deposits were intramembranous, subepithelial, and mesan-gial by electron microscopy, and stained positive for C3 (six of six), properdin, and CFH (two of two) butnegative for Ig LC and heavy chains, C4, and C1q (6/6) by immunofluorescence. Five patients progressedto end-stage renal disease over a median period of 47 months, despite chemotherapy in four patients. Inone patient, monoclonal �LC deposits were observed on a follow-up kidney biopsy after 4 years.

Conclusions GN with isolated glomerular C3 deposits might represent an unusual complication of plasmacell dyscrasia, related to complement activation through an autoantibody activity of the monoclonal Igagainst a CAP regulator protein.

Clin J Am Soc Nephrol 6: 2165–2174, 2011. doi: 10.2215/CJN.06180710

IntroductionA wide variety of glomerular diseases may occurduring the course of plasma cell disorders, resultingfrom deposition of a monoclonal Ig. Based on lightmicroscopy (LM), immunofluorescence (IF), and elec-tron microscopic (EM) studies of kidney biopsy, thesedisorders may be classified according to the nature,localization, and ultrastructural appearance of mono-clonal Ig deposits (1,2). Organized glomerular mono-clonal Ig deposits include immunoglobulinic amyloid-osis (2), microtubular/immunotactoid glomerulopathy(3,4), and type I cryoglobulinemic glomerulonephritis(GN) (5,6), whereas granular amorphous deposits aremostly represented by Randall-type monoclonal Igdeposition disease (MIDD). MIDD is characterized byperitubular, glomerular, and vascular deposition ofeither a single monoclonal light chain (LC) or a trun-cated heavy chain (HC) lacking the first constant do-main, or of both monoclonal Ig LC and HC (1,7).

Recently, a novel type of proliferative GN or mem-branoproliferative GN (MPGN) with monoclonal Ig

deposits was described (8–12). This entity mimicsimmune-complex GN and differs from classic MIDDby the absence of Ig deposition on tubular and vas-cular basement membranes, and by the nonlinear,granular appearance of glomerular deposits (8,9).Like other monoclonal Ig-related glomerulopathies,it is prone to recurrence after kidney transplanta-tion (11).

Although hypocomplementemia and glomerulardeposition of C3 or other complement componentsare common findings in most types of glomerulardisorders related to monoclonal Ig deposition (3,4,8–10,13,14), little attention has been paid to the role ofthe complement system in the pathogenesis of renallesions. Evidence of the nephrotoxic property of amonoclonal Ig, through activation of the complementalternative pathway (CAP) involving anticomplementfactor H (CFH) antibody activity, was first demon-strated in a patient with MPGN, dense deposits of C3,and monoclonal lambda LC (15,16). Rare cases ofdense deposit disease (DDD), also referred to as

*Department ofNephrology, CHUPoitiers, Universite dePoitiers, Poitiers,France; †CNRS UMR6101, Universite deLimoges, Limoges,France; ‡AssistancePublique-Hopitaux deParis, Laboratory ofImmunology, HopitalEuropeen GeorgesPompidou, Paris,France; §Laboratory ofImmunology,�Department ofPathology, CHUPoitiers, Universite dePoitiers, Poitiers, France;¶Centre national de re-ference maladies raresamylose AL et autresmaladies a depotsd’immunoglobulinesmonoclonales.

Correspondence: PrFrank Bridoux,Department ofNephrology and RenalTransplantation, HopitalJean Bernard, 2, rue dela Miletrie, CHUPoitiers, 86021 Poitiers,France. Phone: 33 5 4944 41 59; Fax: 33 5 4944 42 36; E-mail:f.bridoux@chu-poitiers.fr

www.cjasn.org Vol 6 September, 2011 Copyright © 2011 by the American Society of Nephrology 2165

Article

MPGN type II, have been reported in association withplasma cell disorders (17–19). We studied six patients withmonoclonal gammopathy and glomerular disease distinctfrom DDD, characterized by glomerular C3 deposits with-out monoclonal Ig deposits at presentation. Our data sug-gest that this entity might represent an unusual renal com-plication of plasma cell dyscrasia not related to monoclonalIg deposition but to local glomerular CAP activation by themonoclonal Ig, with or without hypocomplementemia.

Study Population and MethodsPatients

Six patients referred to five nephrology departmentsbetween 1990 and 2009 were retrospectively studied. In-clusion criteria were as follows: (1) isolated diffuse granu-lar glomerular C3 deposits, without evidence of Ig LC orHC deposits on IF study of kidney biopsy; (2) presence ofserum and/or urine monoclonal Ig; and (3) absence ofdetectable serum cryoglobulin.

Demographics and clinical and biologic data were re-corded at the time of the first kidney biopsy and at the lastfollow-up visit. Estimated GFR (eGFR) was calculated us-ing the modified MDRD (Modification of Diet in RenalDisease) equation (20).

Pathologic StudiesAll kidney biopsy samples were processed for light and

IF microscopy, as described previously (3). Sections weresystematically stained with Congo red and examined un-der polarized light. The extent of tubular atrophy, intersti-tial fibrosis, arteriosclerosis, and the abundance of glomer-ular deposits were graded on a scale from 0 to 3�. Fordirect IF, 3-�m cryostat sections were stained using poly-clonal FITC conjugates specific for �, �, and � Ig HC; � and� LC (Dakopatts, Glostrup, Denmark); and C3, C4, C1q,and albumin (Morphosys AbD, Dusseldorf, Germany). Intwo patients, indirect IF was performed using monoclonalantibodies to C5b-C9 membrane-attack complex, CFH andproperdin (Morphosys AbD), and a fluorochrome-conju-gated rabbit anti-mouse as secondary antibody (Dako-patts). In five patients, ultrathin sections were processedfor EM studies and examined under a JEOL JEM-1010electron microscope (JEOL, Tokyo, Japan) (3).

Hematologic and Immunologic StudiesBone marrow smears or biopsy (with IF studies in one

case) were performed in all patients. CH50 and plasmaconcentrations of C3, C4, factor B, CFH, and I were mea-sured, as described previously (21). CD46 surface expres-sion was determined on granulocytes using flow cytom-etry (22). C3 nephritic factor (C3NeF) was assessed by theability of tested plasma IgG to stabilize a preformed cell-bound C3bBb convertase, and anti-CFH Ab by ELISA(23,24). Complement genetic screening was performed infour patients, using direct sequencing of all CFH, IF, andMCP exons. The CFH H402 haplotype was tagged bygenotyping the single nucleotide polymorphism rs1061170(c.1204T�C; p. Tyr402His), as reported elsewhere (25).

Serum samples were collected, processed at 37°C, andtested for cryoglobulins. Serum and urine monoclonal Igwere detected by conventional electrophoretic and immu-

noelectrophoretic (IEL) analysis. In two patients, Westernblots were performed to determine serum monoclonal Igsubclasses, using the same specific monoclonal antibodiesas for IF studies (26).

ResultsClinical Data at Diagnosis

Six Caucasian patients—three women and three men(median age 67.5 years, range 40 to 74)—were included inthe study. At the time of kidney biopsy, five patients hadhypertension with significant proteinuria (median 3.3 g/d,range 2 to 5.1), and nephrotic syndrome in three patients.Hematuria was found in all patients, two of whom hadexperienced episodes of gross hematuria. Median serumcreatinine level was 150 �mol/L (range 70 to 298 �mol/L).Three patients had stage 3 chronic kidney disease(CKD), two had stage 4 CKD, and one had stage 5 CKD.Fundoscopic examination, performed in two patients,did not show Drusen or macular degeneration. Nonehad lipodistrophy or other extrarenal manifestations,and none had experienced any infectious episode beforeadmission (Table 1).

Hematologic and Immunologic FindingsAt the time of diagnosis, all patients had a serum mono-

clonal IgG (kappa four patients, lambda two patients), withLC proteinuria in five patients. Western blot showed thatthe IgG subclass was �1 in patients 1 and 2. Three patientshad abnormal serum free kappa (two patients) or lambda(one patient) LC levels. In patients 1, 3, and 5, monoclonalgammopathy had been diagnosed 2 to 10 years beforeadmission.

Bone marrow examination showed a 3% infiltration bydystrophic plasma cells, consistent with stage I multiplemyeloma, in patient 3. IgG1 �-positive plasma cells (1.5%)were found by IF in patient 1. None of the patients hadlytic bone lesions, lymphadenopathy, or spleen or liverenlargement (Table 1).

Tests for serum cryoglobulins, rheumatoid factor, hepa-titis B and C, and HIV infection were negative in all pa-tients. Antinuclear, anti-dsDNA, and M2 antimitochon-drial antibodies were detected in patient 4, without anysymptom of systemic autoimmune disease.

Complement StudiesPatients 2 and 3 had low serum C3 and factor B levels,

which suggested CAP activation. Patient 5 showed de-creased levels of factor B, with normal C3 levels. In patient1, low levels of C3 with normal C4 were observed, butfurther tests were not available. In the five other patients,serum levels of C4, CFH, and I proteins were normal.CD 46 surface expression on granulocytes was normal inthe three patients tested. None of the patients presentedwith a C3NeF. Anti-CFH IgG antibodies were detectedbefore treatment in patient 2. No genetic abnormality wasfound in CFH, CFI, and MCP genes in patients 2 to 4 and 6.Patients 3 and 6 carried at least one copy of CFH H402allele (Table 2).

Pathologic FindingsIn all patients, the light microscopic appearance of glo-

merular lesions was characterized by mesangial prolifera-

2166 Clinical Journal of the American Society of Nephrology

tion, neutrophilic leukocyte afflux in glomerular tufts, anddeposits in the mesangium and glomerular capillary walls(CW; Figure 1). Patients 2 and 6 had mild extracapillaryproliferation (Figure 1C), and microthrombi within capil-lary loops were present in patient 3. Few subepithelialdeposits (humps) were observed in five patients (Figure1D). Five patients had degenerative lesions, includingglobal glomerular sclerosis, tubular atrophy, interstitialfibrosis, and arteriosclerosis, of variable severity.

The main feature by direct IF was diffuse and brightgranular glomerular deposits of C3 in all patients (Figure2A, C, D). No significant staining was observed with anti-kappa, lambda, �, �, �, C4, and C1q conjugates. Indirect IFstudies, available in patients 2 and 3, did not show signif-icant C5b-C9 deposits; however, positive staining was ob-served with anti-CFH and antiproperdin antibodies (Fig-ure 2B). By EM, performed in five patients, nonextensiveamorphous electron-dense deposits, or deposits of inter-mediate density, with a “sausage-shaped” appearance,were observed within the lamina densa, which appeared to

be interrupted on the subendothelial surface (Figure 3).Deposits of lesser density were found in the subepithelialspace (humps) or in paramesangial areas (nodular pseu-dohumps) in most patients (Figure 3A, B).

In patient 1, two follow-up biopsies showed persistenceof glomerular C3 deposits with progression of interstitialfibrosis and glomerular sclerosis. In patient 2, a controlbiopsy at 4 years showed significant amounts of granularmonoclonal lambda LC deposits that colocalized with C3in the mesangium and glomerular CW (Figure 2D, inset;Table 3).

Treatment and Follow-UpFour patients were given chemotherapy with high-dose

dexamethasone, either alone (patient 2) or combined withmelphalan (patient 3) or cyclophosphamide (patient 4).Patient 5, who progressed to stage III multiple myelomaafter 11 years of follow-up, received three courses of bort-ezomib plus dexamethasone, relayed by melphalan plus

Table 1. Renal presentation at baseline and clinical outcome

Patient No. 1 2 3 4 5 6

Baseline characteristicsage (years)/gender

(M/F)40/F 59/M 74/M 71/F 73/F 64/M

hypertension No Yes Yes Yes Yes Yesserum creatinine

(�mol/L)70 150 150 298 130 280

GFR ml/min per 1.73 m2 85 44 42 14 37 21protids/albumin (g/L) 75/32 60/34 66/28 69/35 57/24 64/29proteinuria (g/24 h)/

Nephrotic syndrome0.5/No 2/No 3.3/Yes 2.5/No 5.1/Yes 3.5/Yes

haematuria Yesa Yes Yes Yes Yesa Yesserum monoclonal Ig IgG1� IgG1� IgG� IgG� IgG� IgG�MIg concentration (g/L) 30 7 16 18.9 21.3 4serum FLC (mg/L) NA 988 (�) NA 525 (�) 102 (�) NAkappa/lambda ratio NA 0.03 NA 53.2 5.0 NALC proteinuria � (U � S) � (U � S) � (U) � (U) � (U � S) Nobone marrow PC (%) 4 � 5 3 (dystrophic) 7.5 5 3

Treatmenttime from diagnosis

(months)47 36 7 20

serum creatinine(�mol/L)

450 180 500 250

GFR (ml/min per 1.73 m2) atthe onset of treatment

12 34 8 17

chemotherapy regimen No treatment Dex Mel � Dex CYC � Dex Bortezomib � Dex No treatmentOutcome

ESRD (months fromdiagnosis)

Yes (90) Yes (55) Yes (48) Yes (12) Yes (23) No

serum Cr (�mol/L) 290serum monoclonal

IgG (g/l)23 17 27 14 NA 6

serum FLC (mg/L) 8800 (�) NA NA 653 (�) 559 (�) NAkappa/lambda ratio 1392.4 NA NA 31.4 36.3 NAdeath Yes No No No No Nofollow-up time (months) 162 94 60 25 36 4

amacroscopic hematuria.S, serum; U, urine; NA, not available; ESRD, end stage renal disease; FLC, free light chains; GFR, glomerular filtration rate; LC, lightchains; MIg, monoclonal immunoglobulin, Dex, high-dose dexamethasone; Mel, melphalan; CYC, cyclophosphamide.

Clin J Am Soc Nephrol 6: 2165–2174, September, 2011 Glomerular C3 Deposits and Monoclonal Gammopathy, Bridoux et al. 2167

thalidomide and dexamethasone. Treatment was intro-duced less than 1 years after diagnosis in only one patient,and three had at least stage 4 CKD at the onset of therapy.None achieved hematologic response; however, in patient2, serum titers of IgG anti-CFH autoantibodies decreasedand became undetectable after 11 courses of dexametha-sone. Post-treatment kidney biopsies were not performedin any patient.

After a median follow-up of 47 months (range 4 to 162months), five patients had progressed to (ESRD) over amedian time of 48 months (range 12 to 90 months). Patient1 died from sepsis 14 years after diagnosis and 6 years afterthe onset of hemodialysis.

DiscussionThe association of proliferative glomerular disease with

isolated C3 deposits is an extremely rare condition inadults (27). Isolated intramembranous diffuse C3 depositsis characteristic of DDD (27,28), but disseminated granularglomerular CW and mesangial C3 deposits, without IgGdeposits, are sometimes observed in late stages of post-streptococcal GN. These two conditions result from CAPactivation. C3NeF, an autoantibody with anti-CAP C3 con-vertase activity, is found in more than 80% of DDD casesand in some cases of poststreptococcal GN (23,27). Re-cently, Servais et al. (29) introduced the term glomerulone-phritis C3 (GNC3) to describe glomerular disease in aseries of 19 patients, mostly adults, with isolated glomer-ular C3 deposits distinct from classical DDD and poststrep-tococcal GN. Thirteen patients displayed features of type IMPGN, whereas five patients had mesangial and epimem-

branous deposits without mesangial proliferation and sub-endothelial deposits. Circulating C3NeF and low serum C3levels, indicative of systemic CAP activation, were fre-quent in patients with MPGN features. By contrast, C3levels were normal in most patients without MPGNfeatures who displayed mutations in CFH or factor I intwo thirds of cases. Mutations in the C3 gene have beendescribed in DDD (30), whereas mutations in CFH gene,or in complement factor H-related protein 5 gene, havebeen identified in DDD and GNC3 (21,31,32). These datasuggest that different mechanisms of CAP activationmight result in various patterns of glomerular damage.In the present series, all six patients had similar glomer-ular lesions, with isolated glomerular C3 deposits andoverlapping features of type III MPGN and DDD. Clinicalmanifestations were homogeneous, including heavy pro-teinuria, with nephrotic syndrome and progressive renalfailure in most cases. Three patients had low serum C3levels (with low factor B levels in two), and one showedlow factor B levels with normal C3 levels. Serum levels ofCFH and I proteins were normal in five of five patients, aswas CD46 expression in three of three patients, and nomutations in CFH, CFI, and MCP genes were identified infour of four patients. Strikingly, all had evidence of mono-clonal gammopathy. None of the patients had circulatingC3NeF, indicating that an autoantibody activity against theCAP C3 convertase was not involved in the pathogenesisof glomerular lesions.

The prevalence of monoclonal gammopathy in adultswith isolated glomerular C3 deposits appears to largelyexceed that of the general population, which is around

Table 2. Complement studies

Patient No. 1

2

3 4 5 6BeforeChemo

DuringChemo

AfterChemo

Complementcomponents

CH50 (%) NA 55 86 NA 91 82 68 106C3 antigen (mg/L) 550 433 433 539 583 915 672 901C4 antigen (mg/L) 109 239 218 168 117 224 186 319

CAP proteinsFactor B antigen

(mg/L)NA 55 48 43 66 94 82 142

H protein (%) NA 88 73 89 138 177 117 165I protein (%) NA 131 117 125 111 130 88 148CD46 (MFI) NA 714 NA 791 584 NA 879 NAC3 NeF NA Negative NA NA Negative Negative Negative Negativeanti-factor H IgG NA Positive Positive Negative Negative Negative Negative Negative

Genetic testingCFH NA No

mutation2 copies of

H402No

mutationNA 1 copy of

H402CFI NA No

mutationNo

mutationNo

mutationNA No

mutationMCP NA No

mutationNo

mutationNo

mutationNA No

mutation

Chemo, chemotherapy; MFI, mean fluorescence index; C3Nef, C3 nephritic factor; NA, not available.Normal values: CH50, 70% to 130%; C3 antigen, 660 to 1250 mg/L; C4 antigen, 93 to 380 mg/L; Factor B, 90 to 320 mg/L; H protein, 65%to 140%; I protein, 70% to 130%; MCP, 600 to 1400 MFI.

2168 Clinical Journal of the American Society of Nephrology

1% to 2% in adults and increases from 3.2% to 7.5% inpatients aged over 50 years and over 85 years, respec-tively (33,34). Since the first description of MPGN andisolated C3 deposits with monoclonal gammopathy (17),few similar cases have been described. Nasr et al. (18)reported that four adults (22%) out of a series of 32 DDDcases had a history of plasma cell disorder. All hadtypical ultrastructural features of DDD and sole or dom-inant C3 deposition without LC restriction. The type ofthe circulating monoclonal component was not detailed.In a recent series of 81 hepatitis-negative MPGN pa-tients, 28 had evidence of monoclonal gammopathy,three of whom had isolated C3 deposits (12). Recently,Sethi et al. found that among 14 patients with DDD, 10(71.4%) had a serum monoclonal IgG (19). By IF, only C3deposits were observed. Two patients with monoclonalgammopathy and either DDD (19) or GNC3 (35) carried

one or two copies of the CFH H402 allele variant, whichhas been found to be associated with increased risk ofDDD (36). However, as the frequency of the H402 alleleis high in the general population, and as it was notpresent in two patients from the present series, it isunclear whether a genetic permissive background is re-quired for the development of DDD or GNC3 in patientswith monoclonal gammopathy.

In the present series, the diagnosis of DDD was unlikely,as characteristic glomerular patterns (“tram tracks,” mes-angial rings, continuous glomerular intramembranousdense deposits) were not observed, and because circulatingC3NeF, a common finding in DDD, was absent. Whereas,in previous reports, typical ultrastructural features of DDDwere mostly observed in patients with monoclonal gam-mopathy (17–19), C3GN has been also described as in thepresent patients (35).

Figure 1. | Light microscopic findings. (A) Patient 2, toluidine blue staining (original magnification: �400): diffuse membrano proliferativeglomerulonephritis with mesangial and glomerular capillary wall deposits, and numerous neutrophils within capillary lumens. (B) Patient 3,Marinozzi’s silver staining (original magnification: �400): atypical segmental membranoproliferative lesions with nodular appearance of themesangium and duplication of glomerular basement membranes. (C) Patient 6, light green trichrome (original magnification: �400):glomerular crescentic proliferation with ribbon-like basement membrane thickening of the underlying tuft. (D) Patient 6, light green trichrome(original magnification: �400): numerous round mesangial (pseudohumps) (arrowhead), associated with fibrinoid deposits in glomerularcapillary walls, and some voluminous humps (arrow).

Clin J Am Soc Nephrol 6: 2165–2174, September, 2011 Glomerular C3 Deposits and Monoclonal Gammopathy, Bridoux et al. 2169

The frequency of complement activation in patients withplasma cell disorders is likely to be underestimated. Defi-cient alternative and terminal complement pathways havebeen reported in 30% of patients with multiple myeloma(37). In the present series, a role of monoclonal Ig in thelocal or systemic activation of CAP, leading to subse-quent glomerular proliferative lesions and C3 deposits,is questionable. Such a hypothesis is strongly suggestedin patient 2, in whom MPGN with C3 deposits, but noevidence of monoclonal Ig deposition, was diagnosedsimultaneously with IgG1� monoclonal gammopathyand circulating anti-CFH IgG autoantibody. A controlbiopsy at 4 years, before treatment, revealed colocaliza-tion of glomerular C3 and lambda LC deposits, suggest-ing that isolated C3 deposits might be an initial stepbefore subsequent monoclonal Ig deposition. This case isreminiscent of the MPGN case LOI, in which a dimericmonoclonal V�3 LC was shown to behave as a mini

autoantibody to CFH. By binding to the short consensusrepeat domain 3 at the N-terminal end of CFH, the LOIdimer blocked the interaction between CFH and C3b,thereby inhibiting the activity of CFH and inducinguncontrolled CAP activation (15,16). We later reported asimilar case (LOP) with MPGN and lambda LC, C3,CFH, and C5b-C9 deposits, without evidence of C3NeF.A serum fraction enriched in monoclonal lambda LCinduced C3 conversion by the AP, with cleavage offactor B and increased Bb level, while the C4 level re-mained normal (2). The association of isolated glomer-ular C3 deposits with MGUS, rather than with high-mass myeloma (12,18,19,35), might suggest thatprolonged CAP activation by the monoclonal Ig is re-quired for the development of glomerular lesions.

In the present case, and in previously reported cases, ofglomerular C3 deposits with monoclonal gammopathy,renal outcome was poor, with progression to ESRD in most

Figure 2. | Immunofluorescence findings. (A) Patient 3, anti-C3 FITC-conjugate (original magnification: �400): bright disseminated granularand semilinear glomerular deposits, without ring forms in mesangium or dual tonality. (B) Patient 3, anticomplement factor H FITC-conjugate(original magnification: �400): similar, although weaker, fluorescence pattern of glomerular capillary walls. (C) Patient 6, anti-C3 FITCconjugate (original magnification: �400): monotonal diffuse granular capillary wall and mesangial bright deposits. (D) Patient 2, first kidneybiopsy, anti C3-FITC conjugate (original magnification: �400): similar staining pattern as in Fig. 2A. Inset, second kidney biopsy, anti-� FITCconjugate: few mesangial and capillary wall deposits associated with persistent colocalized bright C3 deposits (not shown).

2170 Clinical Journal of the American Society of Nephrology

Figure 3. | Electron microscopic findings. (A) Patient 2, second kidney biopsy (original magnification: �4000): electron-dense voluminoussubepithelial deposits (humps) (asterisks). (B) Patient 2, second kidney biopsy (original magnification: �6000): round, nodular (pseudohumps)electron-dense mesangial deposits (asterisk) and interrupted intramembranous dense deposits (arrow). (C) Patient 1, third kidney biopsy(original magnification: �3300): glomerular basement membrane thickened by highly electron-dense interrupted intramembranous depositsdisplaying a sausage-shaped (arrow) or bead-like (arrowhead) pattern. (D) Patient 1, third kidney biopsy, original magnification: �3000):intramembranous interrupted electron dense deposits (asterisks). Note the disruption of glomerular basement membrane, with protrusion ofpodocyte epithelium (between short arrows) into the capillary lumen (white arrow) and neoproduction of basement membrane-like material(arrowheads). US, urinary space; CL, capillary lumen.

Clin J Am Soc Nephrol 6: 2165–2174, September, 2011 Glomerular C3 Deposits and Monoclonal Gammopathy, Bridoux et al. 2171

patients (12,18,19). The demonstration of improvement ofrenal function with reduction in proteinuria following che-motherapy would reinforce the role of monoclonal gam-mopathy in the pathogenesis of isolated glomerular C3deposits. In our series, treatment was introduced too late inthe course of the disease to draw any valid conclusion onits potential benefit. However, as the anti-CFH autoanti-body disappeared after dexamethasone therapy in patient2, early chemotherapy aimed at controlling the underlyingclonal disorder should be discussed with regard to pre-venting progression toward ESRD (12). Eculizumab ther-apy, which may improve renal symptoms in DDD (38),also represents an interesting alternative (39).

ConclusionsMonoclonal gammopathy should be considered in adult

patients with MPGN. Isolated C3 deposits might representan unusual complication of plasma cell dyscrasia related toCAP activation by the monoclonal Ig. Further studies,based on testing the ability of purified monoclonal Igs toactivate CAP through an autoantibody activity againstCFH or another complement regulator protein, are neededto confirm this hypothesis.

AcknowledgmentsPortions of this work were previously presented in abstract

form at the 2007 World Congress of Nephrology, Rio de Janeiro,Brazil, and at the 9th Reunion Commune de la Societe de Neph-rologie et de la Societe Francophone de Dialyze, 2007, Lyon,France.

This work was supported by a grant from AREN Poitou-Char-entes.

We gratefully acknowledge the colleagues who participated

in this study: R. Makdassi; C. H. U. Amiens; D. Le Chapois andA. Nony; C. H. Bourges; J. L. Mahe; C. H. Abbeville; L. Mer-cadal, Hopital La Pitie-Salpetriere, Paris; P. Siohan; C. H. Quim-per; S. Le Coz; and C. H. Saint Nazaire. We are indebted to Dr.Winston Hutchinson, who carefully reviewed the manuscript.

DisclosuresNone.

References1. Preud’homme JL, Aucouturier P, Touchard G, Striker L,

Khamlichi AA, Rocca A, Denoroy L, Cogne M: Monoclonalimmunoglobulin deposition disease (Randall type). Relation-ship with structural abnormalities of immunoglobulin chains.Kidney Int 46: 965–972, 1994

2. Touchard G: Ultrastructural pattern and classification of renalmonoclonal immunoglobulin deposits. In: Monoclonal Gam-mopathies and the Kidney, edited by Touchard G, Aucoutu-rier P, Hermine O, and Ronco P, Dordrecht, Kluwer Aca-demic Publishers, 2003, 95–117

3. Bridoux F, Hugue V, Coldefy O, Goujon JM, Bauwens M,Sechet A, Preud’Homme JL, Touchard G: Fibrillary glomeru-lonephritis and immunotactoid (microtubular) glomerulopathyare associated with distinct immunologic features. Kidney Int62: 1764–1775, 2002

4. Rosenstock JL, Markowitz GS, Valeri AM, Sacchi G, AppelGB, D’Agati VD: Fibrillary and immunotactoid glomerulone-phritis: Distinct entities with different clinical and pathologicfeatures. Kidney Int 63: 1450–1461, 2003

5. Galea HR, Bridoux F, Aldigier JC, Paraf F, Bordessoule D,Touchard G, Cogne M: Molecular study of an IgG1kappacryoglobulin yielding organized microtubular deposits andglomerulonephritis in the course of chronic lymphocytic leu-kaemia. Clin Exp Immunol 129: 113–118, 2002

6. Bridoux F, Sirac C, Jaccard A, Abou Ayache R, Goujon JM,Cogne M, Touchard G: Renal disease in croglobulinemic vas-culitis. In: The Kidney in Systemic Autoimmune Diseases,

Table 3. Renal biopsy findings

Patient No. 1 23 4 5 6

Kidney Biopsy No. 1 2 3 1 2

Light microscopyno. of glomeruli 13 8 15 32 12 12 32 23 10sclerotic glomeruli (%) 0 0 20 9 83 16 62 26 20crescents (%) 0 0 0 10 0 0 0 10tubular atrophy � �� ��� � ��� �� � ��interstitial

inflammation0 � a �� �� � b � � �� �� a

interstitial fibrosis � �� ��� �� �� �� �� � ��arteriosclerosis 0 � �� � ��� ��� �� ��

ImmunofluorescenceIg heavy chain � � � � � � � � �Ig light chain � � � � �� (�) � � � �complement C3��� C3��� C3��� C3��� C3��� C3��� C3��� C3��� C3���electron microscopy A NA A A NA NA A A A

Topography of depositsmesangial deposits � � � ��� �� �� ��� ��� ��subepithelial deposits � � � � � � �� � �subendothelial deposits � � � ��� �� � � � �GBM � � � �� � ��

a Infiltrate containing numerous plasma cells; b infiltrate containing predominant �-positive lympho plasmacytoid cells. � : absent, � :mild, �� : moderate, ��� : diffuse.A, available; NA, not available; GBM, glomerular basement membrane.

2172 Clinical Journal of the American Society of Nephrology

Vol. 7, edited by Mason J, Pusey C, Amsterdam, Elsevier,2008, 215–239

7. Lin J, Markowitz GS, Valeri AM, Kambham N, Sherman WH,Appel GB, D’Agati VD: Renal monoclonal immunoglobulindeposition disease: The disease spectrum. J Am Soc Nephrol12: 1482–1492, 2001

8. Bridoux F, Binaut R, Zanetta G, Mougenot B, Goujon JM,Vanhille P, Bauwens M, Chevet D, Ronco P, Preud’hommeJL, Touchard G: Glomerulopathy with non-organized andnon-Randall type monoclonal immunoglobulin deposits: Arare entity [Abstract]. J Am Soc Nephrol 12: 94 A, 2001

9. Nasr SH, Markowitz GS, Stokes MB, Seshan SV, ValderramaE, Appel GB, Aucouturier P, D’Agati VD: Proliferative glo-merulonephritis with monoclonal IgG deposits: A distinct en-tity mimicking immune-complex glomerulonephritis. KidneyInt 65: 85–96, 2004

10. Nasr SH, Satoskar A, Markowitz GS, Valeri AM, Appel GB,Stokes MB, Nadasdy T, D’Agati VD: Proliferative glomerulo-nephritis with monoclonal IgG deposits. J Am Soc Nephrol20: 2055–2064, 2009

11. Lorenz EC, Sethi S, Leung N, Dispenzieri A, Fervenza FC,Cosio FG: Recurrent membranoproliferative glomerulonephri-tis after kidney transplantation. Kidney Int 77: 721–728, 2010

12. Sethi S, Zand L, Leung N, Smith RJ, Jevremonic D, HerrmannSS, Fervenza FC: Membranoproliferative glomerulonephritissecondary to monoclonal gammopathy. Clin J Am Soc Neph-rol 5: 770–782, 2010

13. Soma J, Sato K, Sakuma T, Saito H, Sato H, Sato T, Abbas A,Aucouturier P: Immunoglobulin gamma3-heavy-chain deposi-tion disease: Report of a case and relationship with hypoco-mplementemia. Am J Kidney Dis 43: E10–E16, 2004

14. Mougenot B, Plaisier E, Goujon JM, Bridoux F: Atteinteglomerulaire au cours des dysglobulinemies et des cryo-globulinemies. In: Atlas de pathologie renale, edited byNoel LH. Paris, Medecine Sciences Flammarion, 2008,310 –344

15. Meri S, Koistinen V, Miettinen A, Tornroth T, Seppala IJ: Acti-vation of the alternative pathway of complement by mono-clonal lambda light chains in membranoproliferative glomer-ulonephritis. J Exp Med 175: 939–950, 1992

16. Jokiranta TS, Solomon A, Pangburn MK, Zipfel PF, Meri S:Nephritogenic lambda light chain dimer: A unique humanminiautoantibody against complement factor H. J Immunol163: 4590–4596, 1999

17. Sepandj F, Trillo A: Dense deposit disease in association withmonoclonal gammopathy of unknown significance. NephrolDial Transplant 11: 2309–2312, 1996

18. Nasr SH, Valeri AM, Appel GB, Sherwinter J, Stokes MB, SaidSM, Markowitz GS, D’Agati VD: Dense deposit disease: Clin-icopathologic study of 32 pediatric and adult patients. ClinJ Am Soc Nephrol 4: 22–32, 2009

19. Sethi S, Sukov WR, Zhang Y, Fervenza FC, Lager DJ, MillerDV, Cornell LD, Krishnan SG, Smith RJ: Dense deposit dis-ease associated with monoclonal gammopathy of undeter-mined significance. Am J Kidney Dis 56: 977–982, 2010

20. Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D: Amore accurate method to estimate glomerular filtration ratefrom serum creatinine: A new prediction equation. Modifica-tion of Diet in Renal Disease Study Group. Ann Intern Med130: 461–470, 1999

21. Dragon-Durey MA, Fremeaux-Bacchi V, Loirat C, Blouin J,Niaudet P, Deschenes G, Coppo P, Herman Fridman W,Weiss L: Heterozygous and homozygous factor h deficien-cies associated with hemolytic uremic syndrome or mem-branoproliferative glomerulonephritis: Report and geneticanalysis of 16 cases. J Am Soc Nephrol 15: 787–795, 2004

22. Fremeaux-Bacchi V, Moulton EA, Kavanagh D, Dragon-Du-rey MA, Blouin J, Caudy A, Arzouk N, Cleper R, Francois M,Guest G, Pourrat J, Seligman R, Fridman WH, Loirat C, Atkin-son JP: Genetic and functional analyses of membrane cofac-tor protein (CD46) mutations in atypical hemolytic uremicsyndrome. J Am Soc Nephrol 17: 2017–2025, 2006

23. Fremeaux-Bacchi V, Weiss L, Demouchy C, May A, Palom-era S, Kazatchkine MD: Hypocomplementaemia of poststrep-tococcal acute glomerulonephritis is associated with C3 ne-

phritic factor (C3NeF) IgG autoantibody activity. NephrolDial Transplant 9: 1747–1750, 1994

24. Dragon-Durey MA, Loirat C, Cloarec S, Macher MA, Bl-ouin J, Nivet H, Weiss L, Fridman WH, Fremeaux-BacchiV: Anti-Factor H autoantibodies associated with atypicalhemolytic uremic syndrome. J Am Soc Nephrol 16: 555–563, 2005

25. Fremeaux-Bacchi V, Kemp EJ, Goodship JA, Dragon-DureyMA, Strain L, Loirat C, Deng HW, Goodship TH: The de-velopment of atypical haemolytic-uraemic syndrome is in-fluenced by susceptibility factors in factor H and mem-brane cofactor protein: Evidence from two independentcohorts. J Med Genet 42: 852– 856, 2005

26. Beaume A, Brizard A, Dreyfus B, Preud’homme JL: High inci-dence of serum monoclonal Igs detected by a sensitive im-munoblotting technique in B-cell chronic lymphocytic leuke-mia. Blood 84: 1216–1219, 1994

27. Appel GB, Cook HT, Hageman G, Jennette JC, Kashgarian M,Kirschfink M, Lambris JD, Lanning L, Lutz HU, Meri S, RoseNR, Salant DJ, Sethi S, Smith RJ, Smoyer W, Tully HF, TullySP, Walker P, Welsh M, Wurzner R, Zipfel PF: Mem-branoproliferative glomerulonephritis type II (dense depositdisease): An update. J Am Soc Nephrol 16: 1392–1403, 2005

28. Sethi S, Gamez JD, Vrana JA, Theis JD, Bergen HR, 3rd, Zip-fel PF, Dogan A, Smith RJ: Glomeruli of Dense Deposit Dis-ease contain components of the alternative and terminalcomplement pathway. Kidney Int 75: 952–960, 2009

29. Servais A, Fremeaux-Bacchi V, Lequintrec M, Salomon R,Blouin J, Knebelmann B, Grunfeld JP, Lesavre P, Noel LH,Fakhouri F: Primary glomerulonephritis with isolated C3 de-posits: a new entity which shares common genetic risk fac-tors with haemolytic uraemic syndrome. J Med Genet 44:193–199, 2007

30. Martinez-Barricarte R, Heurich M, Valdes-Canedo F,Vazquez-Martul E, Torreira E, Montes T, Tortajada A, Pinto S,Lopez-Trascasa M, Morgan BP, Llorca O, Harris CL, Rodri-guez de Cordoba S: Human C3 mutation reveals a mecha-nism of dense deposit disease pathogenesis and provides in-sights into complement activation and regulation. J ClinInvest 120: 3702–3712, 2010

31. Gale DP, de Jorge EG, Cook HT, Martinez-Barricarte R, Had-jisavvas A, McLean AG, Pusey CD, Pierides A, Kyriacou K,Athanasiou Y, Voskarides K, Deltas C, Palmer A, Fremeaux-Bacchi V, de Cordoba SR, Maxwell PH, Pickering MC: Identi-fication of a mutation in complement factor H-related protein5 in patients of Cypriot origin with glomerulonephritis. Lan-cet 376: 794–801, 2010

32. Servais A, Noel LH, Dragon-Durey MA, Gubler MC, RemyP, Buob D, Cordonnier C, Makdassi R, Jaber W, BoulangerE, Lesavre P, Fremeaux-Bacchi V: Heterogeneous patternof renal disease associated with homozygous Factor H de-ficiency [published online ahead of print March 9, 2011].Hum Pathol.

33. Saleun JP, Vicariot M, Deroff P, Morin JF: Monoclonal gam-mopathies in the adult population of Finistere, France. J ClinPathol 35: 63–68, 1982

34. Kyle RA, Therneau TM, Rajkumar SV, Larson DR, Plevak MF,Offord JR, Dispenzieri A, Katzmann JA, Melton LJ, 3rd: Preva-lence of monoclonal gammopathy of undetermined signifi-cance. N Engl J Med 354: 1362–1369, 2006

35. Sethi S, Fervenza FC, Zhang Y, Nasr SH, Leung N, Vrana J,Cramer C, Nester CM, Smith RJ: Proliferative glomerulone-phritis secondary to dysfunction of the alternative pathway ofcomplement [published online ahead of print March 17,2011]. Clin J Am Soc Nephrol.

36. Abrera-Abeleda MA, Nishimura C, Smith JL, Sethi S, McRaeJL, Murphy BF, Silvestri G, Skerka C, Jozsi M, Zipfel PF,Hageman GS, Smith RJ: Variations in the complement regula-tory genes factor H (CFH) and factor H related 5 (CFHR5) areassociated with membranoproliferative glomerulonephritistype II (dense deposit disease). J Med Genet 43: 582–589,2006

37. Lugassy G, Platok I, Schlesinger M: Hypocomplementemia inmultiple myeloma. Leuk Lymphoma 33: 365–370, 1999

Clin J Am Soc Nephrol 6: 2165–2174, September, 2011 Glomerular C3 Deposits and Monoclonal Gammopathy, Bridoux et al. 2173

38. Vivarelli M, Emma F. Treatment of a patient with dense de-posit disease with eculizumab (Soliris) [abstract]. J Am SocNephrol 20: 631A–632A, 2009

39. Smith RJ, Alexander J, Barlow PN, Botto M, Cassavant TL,Cook HT, de Cordoba SR, Hageman GS, Jokiranta TS, Kim-berling WJ, Lambris JD, Lanning LD, Levidiotis V, Licht C,Lutz HU, Meri S, Pickering MC, Quigg RJ, Rops AL, SalantDJ, Sethi S, Thurman JM, Tully HF, Tully SP, van der Vlag J,Walker PD, Wurzner R, Zipfel PF: New approaches to the

treatment of dense deposit disease. J Am Soc Nephrol 18:2447–2456, 2007

Received: July 19, 2010 Accepted: May 27, 2011

The first two authors contributed equally to the work.

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

2174 Clinical Journal of the American Society of Nephrology