IgM Contributes to Glomerular Injury in FSGS

BASIC RESEARCH www.jasn.org

IgM Contributes to Glomerular Injury in FSGS

Derek Strassheim,* Brandon Renner,* Sarah Panzer,* Richard Fuquay,* Liudmila Kulik,*Danica Ljubanovi�c,† V. Michael Holers,* and Joshua M. Thurman*

*Department of Medicine, School of Medicine, University of Colorado Denver, Aurora, Colorado; and †Department ofPathology, University Hospital Dubrava, Zagreb, Croatia

ABSTRACTGlomerular IgM and C3 deposits frequently accompany idiopathic FSGS and secondary glomeruloscle-rosis, but it is unknown whether IgM activates complement, possibly contributing to the pathogenesis ofthese diseases. We hypothesized that IgM natural antibody binds to neoepitopes exposed in the glomer-ulus after nonimmune insults, triggering activation of the complement system and further injury. Weexamined the effects of depleting B cells, using three different strategies, on adriamycin-induced glomer-ulosclerosis. First, we treated wild-type mice with an anti-murine CD20 antibody, which depletes B cells,before disease induction. Second,we evaluated adriamycin-induced glomerulosclerosis in Jhmice, a strainthat lacks mature B cells. Third, we locally depleted peritoneal B cells via hypotonic shock before diseaseinduction. All three strategies reduced deposition of IgM in the glomerulus after administration of adria-mycin and attenuated the development of albuminuria. Furthermore, we found that glomerular IgM andC3 were detectable in a subset of patients with FSGS; C3 was present as an activation fragment andcolocalizedwith glomerular IgM, suggesting that glomerular IgMmay have bound a cognate ligand. Takentogether, these results suggest that IgM activates the complement system within the glomerulus in ananimal model of glomerulosclerosis. Strategies that reduce IgM natural antibody or that prevent comple-ment activation may slow the progression of glomerulosclerosis.

J Am Soc Nephrol 24: 393–406, 2013. doi: 10.1681/ASN.2012020187

Glomerular IgM deposition is observed in a widerange of primary and secondary renal diseases, al-thoughthesignificanceof thesedepositshas remainedelusive. Prominent mesangial deposits have beendescribed in patients with idiopathic nephroticsyndrome, including patients with minimal changedisease and idiopathic FSGS.1,2 In some patients,IgM is the primary immune factor detected in thekidney, and the term IgM nephropathywas coined todescribe kidneys with minimal changes by light mi-croscopy but with abundant mesangial depositionof IgM.3 Patients with mesangial expansion ormesangial proliferation by light microscopy fre-quently have abundant glomerular IgM deposition.4

Glomerular IgM is also observed in many secondaryforms of glomerulosclerosis, including diabeticnephropathy5 and hypertensive nephropathy.6 Im-portantly, C3 and C4 are frequently detected in theglomeruli of patients who have IgM deposits.4,7,8

In spite of its promiscuous presence in glomer-ular lesions, a pathogenic role for glomerular IgM

has not been demonstrated. Some series, however,have shown that the presence of IgM is of prognosticsignificance.9,10 Furthermore, recent reports sug-gest that rituximab, a mAb to CD20 that depletesmature B cells, may be beneficial in some patientswith the nephrotic syndrome.11–16 However, themechanisms by which rituximab alleviates the ne-phrotic syndrome are not yet known. Recent workhas demonstrated that “natural antibody” contributesto tissue injury in a number of different diseases.Natural antibody refers to Ig that reacts to certain

Received February 16, 2012. Accepted November 13, 2012.

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

Correspondence: Dr. Joshua M. Thurman, Division of Nephrol-ogy and Hypertension, Department of Medicine, School ofMedicine, University of Colorado Denver, B-115, P.O. Box 6511,Aurora, CO 80045. Email: [email protected]

Copyright © 2013 by the American Society of Nephrology

J Am Soc Nephrol 24: 393–406, 2013 ISSN : 1046-6673/2403-393 393

BASICRESE

ARCH

http://www.jasn.org

mailto:[email protected]

conserved epitopes, even without prior exposure to that epi-tope.17 In mice, natural antibody is primarily produced byCD5+CD11b+IgMhighB220low B-1a B cells, which are pre-dominantly located in the peritoneum.18–20 Studies havedemonstrated that natural antibody IgM binds to endogenousneoepitopes that are exposed after injury of the heart,21 in-testine,22,23 skeletal muscle,24 and kidney.25 Once bound, IgMactivates the complement system and can cause further tissueinflammation and injury. Strategies that deplete peritonealB-1a cells or that block binding of specific natural antibodyclones within injured tissues have proven successful for atten-uating injury in a number of different models.23,26

We hypothesized that natural antibody IgM binds toglomerular epitopes that are exposedafter adriamycin-inducedinjury of glomerular cells, and that bound IgM activates thecomplement system. To test this hypothesis, we utilized threestrategies to deplete B cells in mice with adriamycin nephrop-athy. We also analyzed biopsy tissue from patients withidiopathic FSGS to determine whether the pattern of C3deposition suggests complement activation by deposited IgM.

RESULTS

Treatment of Mice with a mAb to CD20 DepletesPeritoneal B-1a CellsBalb/cmice were injected intravenously with amurinemAb tomouse CD20 (clone 5D2, murine IgG2a) or with vehicle con-trol. Treatment with this antibody reduced splenic total B cellsas well as B-1a B cells (Table 1). As previously reported,20 anti-CD20 was not as effective at reducing peritoneal B cells. Twoweeks after treating mice with the anti-CD20 antibody, serumlevels of IgM were significantly reduced compared with vehicle-treated controls (Table 1).

Treatment with a mAb to CD20 Reduces GlomerularIgM Deposition and Reduces Complement Activationin the Glomeruli of Mice with Adriamycin NephropathyAdriamycinnephropathywas induced inBALB/cmice, and theabundance of glomerular IgM was examined (Figure 1, A andB). Injection of the mice with adriamycin caused a significant

increase in the abundance of glomerular IgM. Pretreatment ofthe mice with anti-CD20 prevented an increase in glomerularIgM after injection with adriamycin, and levels of glomerularIgM in mice that received both anti-CD20 and adriamycinwere similar to those seen in healthy controls. To determinewhether glomerular IgM deposition is an early event aftertreatment of the mice with adriamycin, a group of mice washarvested 1 week after injection of the adriamycin, a timepoint at which they have significant proteinuria but have notyet developed glomerulosclerosis.27,28 We found that glomer-ular IgM levels were significantly higher 1 week after injectionwith adriamycin than in control mice (Figure 1C).

We previously found that polyclonal IgM purified frommouse serumdoesnotbindwithin the glomeruliwhen injectedsystemically.25 To determine whether IgM eluted specificallyfrom kidneys with adriamycin nephropathy would bind toglomerular epitopes when re-injected into mice, we purifiedIgM from the kidneys of adriamycin-treated mice. We theninjected the purified IgM into mice with adriamycin nephrop-athy that had previously undergone B cell depletion with theanti-CD20. This was repeated every week during the course ofthe study. A trend toward greater glomerular IgM was seen inthe reconstituted mice (Figure 1, A and B). Although the in-jection of anti-CD20 mice with purified IgM did not fullyrestore injury to the level seen in BALB/c mice that did notreceive anti-CD20, it is possible that isolation of the IgM fromkidney lysates diminishes the binding of the IgM, and it losesaffinity for its target ligands.

In mice with adriamycin nephropathy, low levels of glomer-ular C4d deposition were detected in control mice but signif-icantly increased in mice treated with adriamycin (Figure 2A).Treatment of mice with anti-CD20 reduced glomerular C4d,confirming that the glomerular C4d deposits are caused by Ig-mediated complement activation. Immunostaining of kidneysfor C3 deposits showed a similar pattern. C3 deposition in-creases in mice after treatment with adriamycin, and treatmentof the mice with anti-CD20 prevents this increase (Figure 2B).Kidney sections from mice with adriamycin nephropathy weredual-stained for IgM and C3 (Figure 2C). IgM and C3 colocal-ized within the glomerulus (appearing yellow in the overlay),supporting the hypothesis that the glomerular IgM activates

Table 1. Reduction in B cell subsets and Ig after intravenous injection with PBS or anti-CD20 mAb

Compartment CellsBefore Adriamycin (wk 0) After Adriamycin (wk 4)

PBS Anti-CD20 % Reduction PBS Anti-CD20 % Reduction

Spleen IgM+B220+ 22.61 0.76 96.64 20.7 0.09 99.57IgMHighB220+CD5+ 0.32 0.09 73.02 0.44 0.02 95.45

Peripheral blood IgM+B220+ 25.30 1.14 95.51 5.75 0 100IgMHighB220+CD5+ 0.49 0.62 226.53 0.45 0 100

Peritoneum IgM+B220+ 29.80 19.97 32.98 17.28 1.43 91.72IgMHighB220+CD5+ 0.41 0.23 44.44 0.605 0.4 33.88

Serum IgM (mg/ml) 7.8061.10 2.6060.30a 66.7Treatment was started 2 weeks before injection with adriamycin. Data are reported as the percentage of total cells.aP,0.001 versus PBS.

394 Journal of the American Society of Nephrology J Am Soc Nephrol 24: 393–406, 2013


complement. Tubulointerstitial C3 did not colocalize with IgM,consistent with our previous work demonstrating that comple-ment activation in the tubulointerstitium occurs via the alter-native pathway.28 C3d, the final cleavage product of activated C3was detected it the glomeruli of adriamycin-treatedmice (Figure2D). This indicates that the C3 deposits are due to complementactivation, not simply to passive trapping of circulating C3.

Treatment with a mAb to CD20 Reduces Proteinuriaand Attenuates Histologic Injury in Mice withAdriamycin NephropathyUrine albumin/creatinine ratios were measured as a markerof glomerular injury. Mice with adriamycin nephropathy weregrossly albuminuric (Figure 3A), and treatment of the mice with

anti-CD20 reduced the degree of albuminuria inmicewith adria-mycin nephropathy.When anti-CD20–treatedmice were treatedwith adriamycin and then re-injected with purified glomerularIgM, however, the degree of albuminuria was comparable withthat seen in mice that received adriamycin alone. This demon-strates that IgM is an importantmediator of injury in thismodel.

Glomerular matrix deposition was measured as an index ofglomerulosclerosis. Treatment with the anti-CD20 antibodyreduced glomerulosclerosis, and injectionof themicewith IgMrestored glomerulosclerosis (Figure 3, B and C). Mice injectedwith adriamycin demonstrated greater glomerular collagen IVdeposition than control mice, but treatment with anti-CD20did not significantly reduce glomerular collagen IV deposition(Figure 3D). Synaptopodin is an actin-associated protein

Figure 1. B cell depletion with an anti-CD20 antibody reduces glomerular IgM deposition in mice with adriamycin nephropathy. Miceare injected with anti-CD20 mAb to deplete their B cells before the induction of adriamycin nephropathy. (A) Immunofluorescencemicroscopy is performed on kidneys 4 weeks after induction of adriamycin nephropathy to assess the abundance of glomerular IgM.Kidneys from three to five mice per group are examined. Thirty glomeruli per section are visualized, and the average for each mouse isdetermined. Mice injected with adriamycin demonstrate a substantial increase in glomerular IgM deposition compared with controlanimals. Mice treated with anti-CD20 demonstrate less glomerular IgM. Mice with adriamycin nephropathy that had been injected withthe anti-CD20 but were also reconstituted with purified IgM eluted from the kidneys of mice with adriamycin nephropathy (Adria/anti-CD20/IgM) demonstrate an apparent increase in glomerular IgM. Glomeruli are indicated with arrowheads and areas of IgM depositionwithin the glomeruli are indicated with small arrows. After converting the images to grayscale, the brightness and contrast of the shownimages are adjusted. All images are adjusted equally. (B) Quantitative analysis of the glomerular IgM in the different treatment groupsconfirms that glomerular IgM deposition is increased after injection with adriamycin, but that this increase is attenuated in mice injectedwith anti-CD20 therapy. (C) Analysis of kidneys taken from mice 1 week after injection with adriamycin demonstrates that IgM depositsdevelop within the first week of disease. Original magnification, 3200. Scale bar, 100 mM.

J Am Soc Nephrol 24: 393–406, 2013 IgM and Glomerulosclerosis 395

www.jasn.org BASIC RESEARCH

Figure 2. Treatment with anti-CD20 reduces glomerular complement activation in mice with adriamycin nephropathy. Mice are injectedwith anti-CD20 mAb to deplete B cells before the induction of adriamycin nephropathy. Complement activation in the glomeruli isexamined by immunofluorescence microscopy 4 weeks after injection with adriamycin. (A) Staining for C4 demonstrates that glomerularC4 deposition increases after injection with adriamycin, but that treatment of the mice with anti-CD20 prevents this increase. (B) Stainingfor C3 demonstrates that glomerular C3 deposition increases after injection with adriamycin. Treatment of the mice with anti-CD20prevents this increase in C3 deposition. Glomeruli are indicated with arrowheads and areas of C3 and C4 staining within the glomeruli areindicated with small arrows. Kidneys from three to five mice per group are examined. Thirty glomeruli per section are visualized, and theaverage for each mouse is determined. After converting the images to grayscale, the brightness and contrast of the shown images are



expressed in podocytes. Staining of the kidneys for synapto-podin (Figure 3E) demonstrated that synaptopodin expres-sion was lost in mice with adriamycin nephropathy, indicatingpodocyte injury. The pattern of synaptopodin staining waspreserved in mice treated with anti-CD20. Electron micros-copy also demonstrated widespread foot process effacement inadriamycin-treated mice (Figure 3F). Although there wereareas of effacement in the anti-CD20 treated mice, the efface-ment was not as severe as in mice with adriamycin nephrop-athy that did not receive the anti-CD20.

B Cell–Deficient Mice Are Protected from AdriamycinNephropathyThere is evidence that anti-CD20 can reduce proteinuriathrough direct effects on podocytes.29 To confirm that B cellsare involved with the pathogenesis of this disease, we inducedadriamycin nephropathy in BALB/c mice, Rag12/2 mice(a strain of mice that lacks T cells and B cells), and Jh mice(a strain of mice deficient in mature B cells). There was a trendtoward lower albuminuria in the Rag12/2 mice comparedwithwild-type controlmice, but albuminuriawas significantlylower in Jh mice than in wild-type controls (Figure 4A). Elec-tron microscopy demonstrated patchy foot process efface-ment (Figure 4B), but the effacement was not as widespreadas in the adriamycin-treated BALB/cmice (Figure 3F). Inwild-type mice, IgM and C3 colocalize within the glomeruli (seeFigure 2C). As expected, no IgM was seen in the glomeruli ofthe Jh mice (Figure 4C). Glomerular C3 deposition was alsolower in Jh mice, indicating that glomerular IgM activates thecomplement system in wild-type mice. Tubulointerstitial C3deposition was still evident, however, consistent with our pre-vious observation that complement activation in this com-partment proceeds exclusively via the alternative pathway.28

Depletion of Peritoneal B cells by Hypotonic ShockPrevents Glomerular IgM Deposition in AdriamycinNephropathyNatural antibody IgM is predominantly produced by perito-nealB-1a cells.Asdescribed above, treatmentofmicewith anti-CD20 also depletes peritoneal B cells, although less effectivelythan it depletes splenic B cells (Table 1). To effectively reducethe peritoneal B cells without affecting splenic B cell function,we lysed peritoneal cells by hypotonic shock. This treatmentcauses lysis of all peritoneal cells, including macrophages, butit does not affect splenic B cells (Table 2). We did see somereduction in circulating B cells using this strategy, suggestingthat the peritoneal compartment is the source of some circu-lating B cells. Although all peritoneal cells are reduced by this

method, the treatment does not reduce the number of B-1acells as a percentage of total peritoneal cells. Consequently, theeffects of this treatment may not be specific to depletion ofB-1a cells. The level of total IgM in the serum was unaffectedby this treatment (Table 2).

To deplete the peritoneal B cells in mice with adriamycinnephropathy, the peritoneal cells were lysed by hypotonicshock every 3 days for 2weeks before injectionwith adriamycinin order to give sufficient time for preformed IgM to turnover.25 Depletion of the cells was maintained by intraperito-neal injections of distilled water every three days during thecourse of the study. Although this treatment is very effective atimmediately reducing the peritoneal B cell numbers,25 thecells do re-accumulate and the depletion at day 14 was notcomplete (Table 2). As controls, another group of mice re-ceived peritoneal injections with PBS according to the sameschedule. When reported as a percentage of total peritonealcells, the B-1a cells were not reduced by this treatment (Table2) because the treatment reduces all peritoneal cells equally.

In mice that underwent peritoneal B cell depletion, theaccumulation of glomerular IgM was significantly attenuatedafter injection of the mice with adriamycin (Figure 5A). De-pletion of the peritoneal cells showed a trend toward reductionof glomerular C4 deposition, and glomerular C3 depositionwas reduced by this treatment (Figures 5, B and C). Tubuloin-terstitial C3 deposition was unaffected by this treatment.

These results demonstrate that peritoneal B cells generate asignificant proportion of the IgMdeposited in the glomeruli inthis model. Levels of circulating IgM were not affected by theperitoneal cell depletion even though glomerular depositionwas reduced, suggesting that the glomerular IgM deposits arecomposed of IgM that binds to specific glomerular antigens. Aswith treatment of mice with anti-CD20, depletion of theperitoneal B cells reduced glomerular complement activation.

Depletion of Peritoneal B-1 Cells by Hypotonic ShockReduces Proteinuria and Histologic Injury in Mice withAdriamycin NephropathyUrine albumin/creatinine ratiosweremeasured in samples col-lected 1 or 4weeks after injectionwith adriamycin (Figure 6A).The level of albuminuria was significantly lower in mice thathad undergone peritoneal cell depletion at both the early andthe later time points. The reduction in albuminuria at the1-week time point indicates that glomerular IgM contributesto injury at an early phase of disease in this model. Depletion ofthe peritoneal B cells attenuated the overall degree of glomer-ulosclerosis (Figure 6, B and C), but it did not prevent collagenIV accumulation in adriamycin-treated mice (Figure 6D).

adjusted.All imagesareadjustedequally. (C)Dual stainingof kidneys frommicewith adriamycinnephropathy for IgMandC3demonstratescolocalization of IgM and C3 within glomeruli (appearing yellow in the overlay). Glomeruli are indicated with arrowheads and areas ofcolocalization are indicated with small arrows. (D) Staining of kidneys from mice with adriamycin nephropathy for C3d confirms that theglomerular deposits contain activated complement proteins. Original magnification, 3200 in A–C; 3400 in D. Scale bar, 100 mM.



Figure 3. Treatment with anti-CD20 reduces glomerular injury in mice with adriamycin nephropathy. Mice are injected with anti-CD20mAb to deplete B cells before the induction of adriamycin nephropathy. (A) Urine albumin/creatinine levels are measured. Treatmentof mice with adriamycin causes high-level albuminuria, but this is significantly attenuated by treatment with anti-CD20. Mice treatedwith anti-CD20 that are re-injected with IgM purified from diseased kidneys have levels of albuminuria similar to those seen in



Glomerular IgM and C3d Colocalize in Human FSGSPrevious series have reported that glomerular IgM and/or C3may be detected in up to 90% of patients with FSGS.1 Wereviewed the biopsy reports of 174 cases of FSGS evaluatedover the past 8 years by one of the authors of this study (D.L.).

Of those biopsies, approximately 23%demonstrated glomerular IgM withoutC3, approximately 2% demonstrated glo-merular C3 without IgM, and 7% of thebiopsies demonstrated glomerular IgMand C3 (Figure 7A). The percentage of bi-opsies demonstrating IgM and/or C3 inour patients is lower than that reported inother series, and indicates that natural IgMmay only contribute to a small subset ofFSGS patients. It is possible, however,that a focused examination of fresh tissuespecifically for IgM or C3 would have iden-tified these proteins in a greater percentageof the patients. We performed dual stainingfor IgM and for C3d on tissue in whichboth factors were detectable (Figure 7B).C3d and IgM showed a similar pattern ofdistribution throughout the glomerulus.Dual staining for IgM and C4 was alsoperformed (Figure 7C). This staining dem-onstrated similar colocalization of thesefactors within the glomerulus.

DISCUSSION

Glomerulosclerosis is a common histologicfinding in both primary and secondaryglomerular diseases. Glomerular IgM andcomplement proteins are frequently detec-ted within glomeruli that bear segmentalscars. The significance of these deposits hasbeen uncertain, and some investigators

have proposed that these large molecules are nonspecificallytrapped in areas of glomerular injury.30 In this study, we uti-lized three strategies to deplete B-1a B cells, the primary sourceof natural antibody IgM. Our first approach was to treat micewith a murine anti-murine CD20 antibody. Treatment of the

adriamycin-treatedmice. (B)Glomerulosclerosis in the different treatment groups is assessedon sections stainedwithperiodic acid–Schiff.Glomeruli are indicated with arrowheads and areas of sclerosis are indicated with small arrows. (C) The percentage of each glomerulusoccupied by matrix is determined as an index of glomerulosclerosis. The reduced glomerulosclerosis is observed in mice treated withadriamycin and anti-CD20 compared with adriamycin-treated mice, and mice that are reconstituted with IgM demonstrate an increase inglomerulosclerosis. (D) Staining of kidneys for collagen IV demonstrates that injection of mice with adriamycin causes an increase inglomerular collagen IV deposition, and this is not significantly affected by treatment with anti-CD20. Kidneys from three mice per groupare examined. Thirty glomeruli per section are visualized, and the average for each mouse is determined. Representative glomeruli frommice in each group are shown. Glomeruli are indicated with arrowheads and areas of collagen IV deposition are indicated with smallarrows. (E) Staining of kidneys for synaptopodin demonstrates that injection of mice with adriamycin disrupts synaptopodin expression inthe glomerulus. Some expression is detected in mice treated with adriamycin and anti-CD20, but the expression is lower than in controlmice. Glomeruli are indicatedwith arrowheads and areas of synaptopodin staining are indicatedwith small arrows. (F) Electronmicroscopydemonstrates severe foot process effacement in mice treated with adriamycin. Mice treated with adriamycin and anti-CD20 demonstratepatchy foot process effacement. Small arrows indicate areas of effacement, and the arrowhead indicates an area of preserved footprocesses. Original magnification, 3400 in B, D, and E. Scale bar, 100 mM.

Figure 4. B cell–deficient mice are protected from adriamycin nephropathy. Balb/c,Rag-1 deficient mice (which lack T cells and B cells) on a Balb/c background, and Jhmice (which lack mature B cells) on a Balb/c background are injected with adriamycin.(A) There is significantly less albuminuria in Jh mice than in Balb/c mice 4 weeks afterinjection with adriamycin. (B) Electron microscopy demonstrates areas of preservedfoot processes in the Jh mice. (C) No IgM is detected in the Jh mice. C3 is detected inthe tubulointerstitium and along Bowman’s capsule, but is not seen within the glo-merular tuft. Glomeruli are indicated with arrowheads and areas of C3 staining areindicated with small arrows.



mice with anti-CD20 reduced glomerular IgM deposition andglomerular complement activation in mice with adriamycinnephropathy. Treatment with the anti-CD20 reduced the de-gree of proteinuria that developed after injection with adria-mycin, demonstrating that B cells contribute to glomerularinjury in this model. Treatment with anti-CD20 preservedglomerular synaptopodin expression and reduced the degreeof foot process effacement, indicating that this treatment pro-tected the podocytes from injury in this model. Mice that re-ceived the same dose of anti-CD20 but that were reconstitutedwith IgM purified from kidneys with adriamycin nephropathyshowed a trend toward greater proteinuria thanmice that werenot reconstituted with the IgM, supporting the hypothesis thatIgM is an important mediator of glomerular injury in thismodel. Injection of the mice with purified IgM did not restoreinjury to the level seen in mice that did not receive anti-CD20,however, possibly because the purification process altered thebinding of the IgM with glomerular epitopes.

Next, we utilized mice with targeted genetic deletions thatprevent the formation of mature B cells. We utilized Rag12/2

and Jh mice. The Jh mice developed milder albuminuria thanBALB/c controls. As with the mice that received anti-CD20,there was less glomerular complement activation than BALB/C mice, indicating that IgM activates the complement systemin the injured glomeruli. The foot processes in the kidneys ofJh mice were also relatively preserved, indicating that podo-cyte injury is reduced in B cell–deficient mice. Interestingly,the Rag12/2 mice were not protected to the same degree asthe Jh mice, suggesting that the combined absence of T cellsand B cells is not as protective as the absence of B cells. Thisdifference suggests that some T cell populations, such as reg-ulatory T cells, are protective in this model.31 This also raisesthe possibility that therapies that selectively target B cells willbemore effective than those that target both B cells and T cells.It is worth noting, however, that B cells have additional im-munologic functions beyond Ig production. For example, theyaffect T cell function through the production of cytokines andB cells also function as antigen presenting cells. These other Bcell functions may contribute to their pathogenic role in thismodel.

Finally, we depleted peritoneal B cells by hypotonic shock.This treatment depletes all peritoneal cells (not just B cells),

but it has only aminor effect on splenicB cellsor serum levels of IgM. Depletion of theperitoneal B cells inmice reduced glomerularIgM deposition, glomerular complementactivation, albuminuria, and glomeruloscle-rosis. Furthermore, this strategy reducedglomerular IgM despite the fact that circu-lating levels of IgM were unchanged. Thisindicates that the glomerular IgM depositsare composed of specific clones of IgM, andare not simply the result of deposition of totalcirculating IgM.

More than 70%ofB-1aB cells reside in theperitoneum.18 The reduction in glomerular IgM in mice sub-jected to peritoneal hypotonic shock supports the concept thatthe nephritogenic IgM is natural antibody made primarily bythis B cell subset. Our findings indicate that glomerular IgMdeposition is due to the binding of specific clones of IgM toglomerular epitopes. Similarly, our results indicate that thedeposition of C4 and C3 fragments in the glomeruli is causedby Ig deposition and complement activation because therapiesthat depleted B cells reduced the degree of glomerular com-plement activation. Because IgM activates the classic pathwaywhen bound to a cognate ligand, complement activation byIgM within the glomeruli also indicates that the IgM hasbound to specific glomerular epitopes.

As previously reported in several case series,1,2,32 we foundthat IgM is present in the glomeruli of a subset of patientsdiagnosed with primary FSGS. Dual staining of the tissue forIgM and C3d showed colocalization of these immune factorswhen they were both present. C3d is a cleavage fragment of C3that is generated during complement activation, and IgM ac-tivates the complement system when in complex with a targetligand. Thus, the biopsy results indicate that IgM binds toglomerular epitopes in some patients with FSGS. Measure-ment of complement activation fragments in plasma frompatients with FSGS further demonstrates that the complementsystem is activated in this disease (J.M. Thurman andH. Trachtman, unpublished data).33

Glomerular IgM and C3 are only detected in a subset ofpatients with primary and secondary forms of FSGS. FSGS is avery heterogeneous disease, however, both clinically and mor-phologically.34,35 Although B cell depletion and complementdeficiency contribute to the early development of albuminuriain the adriamycin model,28,36 Ig- and complement-mediatedinjury may be secondary phenomena that occur after a primaryrenal insult. We have observed glomerular IgM deposition afterischemic and chemical injury of the kidneys,25 suggesting thatdiverse insults may generate glomerular neoepitopes. The de-gree to which the IgM-complement axis is engaged in a givenpatient, however, is likely influenced by the abundance of glo-merular epitope and also by the degree to which local comple-ment regulation remains intact. Nevertheless, the presence ofthese same immune factors in kidneys affected by systemic dis-eases such as hypertensive nephrosclerosis6 and diabetic

Table 2. Reduction in B cell subsets and Ig after intraperitoneal hypotonic shockwas performed every 3 days for a total of 14 days

Compartment Cells PBS H2O % Reduction

Spleen IgM+B220+ 26.34 24.45 7.19IgMHighB220+CD5+ 0.15 0.16 23.33

Peripheral blood IgM+B220+ 25.165 11.72 53.45IgMHighB220+CD5+ 0.355 0.09 76.06

Peritoneum IgM+B220+ 26.44 7.86 70.29IgMHighB220+CD5+ 0.21 0.19 11.90

Serum IgM (mg/ml) 13.7663.01 16.6762.54 —

Data are reported as the percentage of total cells.



nephropathy5 raises the possibility that recognition of glomer-ular epitopes by natural IgM is a common downstream mech-anism for a wide range of renal insults.

As previously reported,25,37 we found that there are lowlevels of IgM and C4 in the glomeruli of normal mice. In un-manipulated mice, the IgM colocalized with C4, but only lowlevels of C3 were observed within the glomeruli (Figure 2).This may be because complement regulatory proteins ex-pressed within the glomerulus effectively limit complementactivation by the IgM.25 Studies of normal human tissue also

demonstrate low levels of IgM and C4 deposition within theglomeruli,38,39 and mesangial C4d is so reliably present that itis often used as an internal staining control for C4d staining oftransplant biopsies. Glomerular injury by adriamycinmay im-pair local complement regulation, thus permitting greatercomplement activation by bound IgM. IgM is a classical path-way activator, but we and others previously demonstrated arole for alternative pathway amplification in this model.28,36 Itis possible, therefore, that glomerular injury simultaneouslyincreases classical pathway activation by natural IgM, which

Contro

Adria

Figure 5. Depletion of peritoneal B cells reduces the glomerular deposition of IgM, C3, and C4 in mice with adriamycin nephropathy.Peritoneal cells are depleted with hypotonic shock, starting 2 weeks before inducing adriamycin nephropathy. (A) Immunofluorescencemicroscopy demonstrates that depletion of the peritoneal cells attenuates the glomerular deposition of IgM. (B) Immunofluorescencemicroscopy for C4 demonstrates that depletion of the peritoneal cells also reduces C4 deposition within the glomeruli, although thisdoes not reach statistical significance. (C) Immunofluorescence microscopy for C3 demonstrates that depletion of the peritoneal cellsprevents complement C3 activation within the glomerulus. Kidneys from three to four mice per group are examined. Thirty glomeruli persection are visualized, and the average for each mouse is determined. Glomeruli are indicated with arrowheads, and areas of positivestaining within the glomeruli are indicated with small arrows. Original magnification, 3200. Scale bar, 100 mM.



Figure 6. Depletion of peritoneal B cells reduces albuminuria in mice with adriamycin nephropathy. Peritoneal cells are depleted withhypotonic shock, starting 2 weeks before the induction of adriamycin nephropathy. Urine albumin/creatinine levels are measured. (A)Peritoneal depletion of B cells significantly attenuates the level of albuminuria 1 week and 4 weeks after injection of the adriamycin. (B)Representative glomeruli from mice in each group are shown. The kidneys are stained with periodic acid–Schiff, and glomeruli areindicated with arrowheads. (C) The percentage of each glomerulus occupied by matrix is determined as an index of glomerulosclerosis.Peritoneal cell depletion reduces the degree of glomerulosclerosis compared with PBS-treated mice with adriamycin nephropathy. (D)Staining of kidneys for collagen IV demonstrates that injection of mice with adriamycin causes an increase in glomerular collagen IVdeposition, and this is not significantly affected by depletion of peritoneal B cells. Thirty glomeruli per section are visualized, and theaverage for each mouse is determined. Representative glomeruli from mice in each group are shown. Original magnification, 3400.Scale bar, 100 mM.



binds to injury associated epitopes, while also decreasing theability of complement regulatory proteins within the glomer-ulus to control amplification of complement activationthrough the alternative pathway.

The primary treatments for idiopathic FSGS are immu-nosuppressive drugs, yet it has remained uncertain whetherFSGS is an immune-mediated disease. Indeed, studies haveindicated that the efficacy of some of these drugs may be due,in part, to nonimmune effects. Cyclosporine, for example,may reduce proteinuria by stabilizing the podocyte cyto-skeleton.40 Even rituximab may confer benefit from direct(nonimmune) effects on the podocytes.29 We cannot ex-clude off-target effects of either of the treatments we haveutilized. Taken together, however, the current studies dem-onstrate that B cells and IgM contribute to glomerularinjury. Our previous study demonstrating that comple-ment-deficient mice are also protected in this model furtherimplicates the IgM-complement axis in the development ofglomerulosclerosis.28

Conventional immunosuppressive treatments, such ascorticosteroids and calcineurin inhibitors, have limited effi-cacy in reducing natural antibody formation and complementactivation.41–43 Thus, even when this system is engaged, thestandard immunomodulatory therapies used to treat FSGSmay not be of benefit. Newer drugs that target B cells or thecomplement system may be more effective at reducing nat-ural antibody deposition and complement activation withinthe glomeruli of patients with FSGS. Our findings raise thepossibility that identification of glomerular IgM and C3 de-posits may identify patients likely to benefit from treatmentwith rituximab or eculizumab (a mAb to the complement C5protein).

Thecurrent treatment approach for patientswith secondaryglomerulosclerosis is treatment of the primary disease andnonspecific measures, such as BP control.44 CKD often pro-gresses, however, even if the initiating process becomes quies-cent. This suggests that common downstream mechanisms ofdisease progression may be engaged.45 The prevalence of glo-merular IgM and C3 in diverse toxic, metabolic, hemody-namic, and autoimmune diseases of the kidney raises thepossibility that IgM natural antibody binds epitopes exposedor generated by all of these insults. If so, strategies that blocknatural IgM in the kidney could be beneficial in awide range ofchronic, progressive renal diseases. Competitive antagonistshave been used to block the binding of natural antibodies totissue-specific epitopes.46

In conclusion, this study demonstrates that therapeuticstrategies that deplete B cells ameliorate glomerular injuryin adriamycin nephropathy. All three of the strategies wetested reduced the deposition of IgM in the glomeruli andprevented glomerular complement activation. These find-ings indicate that the deposition of IgM in this model is dueto specific binding of IgM to glomerular epitopes, andtherapies that deplete B-1a cells may be a useful therapy forglomerulosclerosis.

CONCISE METHODS

Adriamycin Nephropathy ModelBALB/c and Rag1 deficient mice (C.129S7(B6)-Rag1tm1Mom/J; defi-

cient in mature T cells and B cells) on a BALB/c background were

obtained from the Jackson Laboratory (Bar Harbor, ME). Mice with

targeted deletion of the Jh gene, which are deficient in mature B cells,

were obtained fromTaconic (Hudson,NY). Allmice were housed and

maintained at the University of Colorado Center for Laboratory

Animal Care in accordance with the National Institutes of Health

Guidelines for the Care and Use of Laboratory Animals. Adriamycin

nephropathy was induced in Balb/c mice as previously described.27,28

Disease was induced in 8- to 10-week-old male mice with a single

intravenous injection of 11 mg/kg of adriamycin, (doxorubicin; Bed-

ford Laboratories, Bedford, OH). Mice were followed for 4 weeks

after disease induction.

Urine Albumin MeasurementUrine albumin was measured by an ELISA, according to the

manufacturer’s instructions (Bethyl Laboratories, Montgomery,

TX). Urine creatinine was measured using an Alfa Wasserman ACE

Chemistry Analyzer. To normalize urine albumin excretion, the val-

ues are reported as micrograms albumin per milligram creatinine.

Renal MorphologyAfter the kidneys were removed from the mice, sagittal sections were

fixed, embedded in paraffin, and 4-mm sections were cut and stained

with periodic acid–Schiff. To quantitate the degree of glomerular in-

jury, glomeruli were examined using an Aperio digital microscope.

The degree of fibrosis was quantified with the Scanscope digital pro-

gram. Regions of interest were drawn around glomeruli, and the

percentage of each glomerulus occupied by matrix was determined.

An average percentage for each slide was calculated. Electron micros-

copy was performed at the University of Colorado Denver School of

Medicine electron microscopy core facility.

Depletion of B Cells by CD20 AntibodyMice were injected via the tail vein with 250 mg of Ab to CD20 (mu-

rine IgG2a [clone 5D2], generously provided by Genentech). This

treatment was performed 2 weeks before disease initiation, and every

2 weeks for the duration of the study. Control mice received an equal

volume of PBS as a vehicle control.

Depletion of Peritoneal B Cells by Hypotonic ShockPeritoneal B cells were depleted, as previously described.25 Mice re-

ceived intraperitoneal injections of distilled water (0.5 ml) starting 12

days before disease induction, and this treatment was continued every

3 days for the duration of the experiment. Control mice were injected

with 0.5 ml of PBS according to the same schedule.

Flow CytometryPeritoneal cells were obtained by lavage, and were then washed and

resuspended in 500 ml of PBS. To stain splenocytes, the spleen was

mechanically dissociated and suspended in 10 ml of PBS, and the cell

suspension was passed through a 70-mM nylon strainer. After



Figure 7. IgM and C3d colocalize in glomeruli of patients with FSGS. (A) The biopsy reports of 174 patients with FSGS are reviewed. Ofthese, 23% are noted to have IgM deposition, 7% have both IgM and C3, and 2% have detectable C3 in the absence of IgM. Availabletissue from 19 patients with FSGS is dual-stained for IgM and C3d (B) or for IgM and C4d (C). In biopsies that contain IgM andcomplement deposits, the two immune factors colocalize within the glomeruli. Nonspecific binding of the FITC-conjugated secondaryantibody is examined (control) in order to confirm specificity of the staining. The brightness and contrast of the shown images areadjusted to improve localization of the factors in the overlay. Original magnification, 3400. Scale bar, 100 mM.



washing in the cells in PBS, they were resuspended in 1 ml of 36%

Percoll, gently applied to 72% Percoll, and centrifuged at 10003g for

30 minutes at 6°C, and the interphase was collected. The cells were

washed in PBS and resuspended in PBS containing 1% FCS. The cells

were stained with antibodies to murine B220, CD5, and CD3 (all

antibodies from Biolegend, San Diego, CA). Antibodies to murine

CD19 (BD Pharmingen, San Diego, CA), IgM, and IgG (both from

Novus Biologicals, Littleton, CO) were also used. Flow cytometry was

performed using a FACS Calibur machine (BD Biosciences, San Jose,

CA), and analyzed with CellQuest Pro software (BD Biosciences).

Immunofluorescence MicroscopySagittal sections of the kidneys were snap-frozen in OCT compound

(Sakura Finetek, Torrance, CA). Four-micrometer sections were cut

with a cryostat and stored at270°C. At the time of staining, the slides

were fixed with acetone and then incubated with the primary anti-

body. Sections were stained with FITC-conjugated anti-mouse C3

(MP Biomedicals, Solon, OH), an antibody to human C3d that

cross-reacts with murine C3d (Dako, Carpinteria, CA),47 FITC-

conjugated antibodies tomouse IgGand IgM(Jackson ImmunoResearch

Laboratories,West Grove, PA), anti-synaptopodin (Sigma-Aldrich, St.

Louis, MO), and anti-mouse C4 (Hycult Biotech, Uden, The Nether-

lands). Secondary antibodies were obtained from Jackson Immuno-

Research Laboratories (West Grove, PA). Sections were counterstained

with hematoxylin (Vector Laboratories, Burlingame, CA). Staining

was visualized using an Olympus BX51 microscope with a digital

camera (Pixera, Santa Clara, CA) and Studio 3.0.1 software (Pixera).

For quantitative analysis, high-powered images of 30 glomeruli were

obtained per section at room temperature. The fluorescence intensity

was determined using ImageJ software, and the mean value for each

treatment group was determined. Image overlays were created using

Adobe Photoshop CS3 software (San Jose, CA).

Serum IgM Measurement by ELISABriefly, ELISA plates (Corning Glass, Corning, NY) were coated with

100 ng of anti-mouse IgM (Southern Biotechnology Associates,

Birmingham, AL). After the plates were washed and blocked with 1%

BSA (Sigma-Aldrich) for 2 hours, serum samples were diluted 1:50

and added to the wells. Standard curves were generated using serial

dilutions of purified IgM (7.8125–500 ng/ml). The plates were

then washed, and IgM was detected with a horseradish peroxidase–

conjugated anti-mouse IgM Ab (Southern Biotechnology Associates).

The plates were then developed with ABTS (Sigma-Aldrich) and read

on a SpectramaxPLUSplate reader (Molecular Devices, Sunnyvale, CA).

Purification IgM from KidneysIgM was purified from mouse kidneys using immobilized anti-IgM

beads (rat anti-mouse IgM Sepharose 4B; Invitrogen, Grand Island,

NY). Mouse kidneys were solubilized and IgM was dissociated from

antigen in standard RIPA buffer. The supernatant was centrifuged at

20,0003g for 15minutes and the soluble fractionwas used for affinity

purification. The IgM content was determined by ELISA for mouse

IgM, and the purity was assessed by Coomassie staining and Western

blot analysis. Approximately 3 mg of IgM was injected per mouse per

week via the tail vein starting 1 week after disease initiation. The

weekly injections were continued for the duration of the experiment.

Natural IgM is generally polyreactive, and the specificity of the eluted

IgM is not known. Therefore, control IgM was not used in these

experiments.

Statistical AnalysesData were analyzed using GraphPad Prism software (La Jolla, CA).

Comparison between groups was performed with unpaired t testing

and by ANOVA when multiple groups were compared. A P value

,0.05 was considered significant.

ACKNOWLEDGMENTS

Murine anti-CD20 antibody was generously provided by Dr. Andrew

Chan at Genentech.

This work was supported in part by the National Institutes of

Health (R01 DK076690) and the National Institute of Allergy and

Infectious Disease (pilot project grant to J.M.T. and R01 AI311052 to

V.M.H.).

DISCLOSURESV.M.H. and J.M.T. are paid consultants for Alexion Pharmaceuticals Inc.

REFERENCES

1. Gephardt GN, Tubbs RR, Popowniak KL, McMahon JT: Focal andsegmental glomerulosclerosis. Immunohistologic study of 20 renal bi-opsy specimens. Arch Pathol Lab Med 110: 902–905, 1986

2. Habib R, Girardin E, Gagnadoux MF, Hinglais N, Levy M, Broyer M:Immunopathological findings in idiopathic nephrosis: Clinical signifi-cance of glomerular “immune deposits”. Pediatr Nephrol 2: 402–408,1988

3. Cohen AH, Border WA, Glassock RJ: Nehprotic syndrome with glo-merular mesangial IgM deposits. Lab Invest 38: 610–619, 1978

4. Lawler W, Williams G, Tarpey P, Mallick NP: IgM associated primarydiffuse mesangial proliferative glomerulonephritis. J Clin Pathol 33:1029–1038, 1980

5. Ainsworth SK, Hirsch HZ, Brackett NC Jr, Brissie RM, Williams AV Jr,Hennigar GR: Diabetic glomerulonephropathy: Histopathologic, im-munofluorescent, and ultrastructural studies of 16 cases. Hum Pathol13: 470–478, 1982

6. Mujais SK, Emmanouel DS, Kasinath BS, Spargo BH: Marked pro-teinuria in hypertensive nephrosclerosis. Am J Nephrol 5: 190–195,1985

7. Cavallo T, Johnson MP: Immunopathologic study of minimal changeglomerular disease with mesangial IgM deposits. Nephron 27: 281–284, 1981

8. Hsu HC, Chen WY, Lin GJ, Chen L, Kao SL, Huang CC, Lin CY: Clinicaland immunopathologic study ofmesangial IgM nephropathy: Report of41 cases. Histopathology 8: 435–446, 1984

9. Zeis PM, Kavazarakis E, Nakopoulou L, Moustaki M, Messaritaki A, ZeisMP, Nicolaidou P: Glomerulopathy withmesangial IgMdeposits: Long-term follow up of 64 children. Pediatr Int 43: 287–292, 2001

10. Alexopoulos E, Papagianni A, StangouM, Pantzaki A, PapadimitriouM:Adult-onset idiopathic nephrotic syndrome associated with pure diffusemesangial hypercellularity. Nephrol Dial Transplant 15: 981–987, 2000

11. Fernandez-Fresnedo G, Segarra A, González E, Alexandru S, DelgadoR, RamosN, Egido J, PragaM; Trabajo de Enfermedades Glomerulares



de la Sociedad Española de Nefrología (GLOSEN): Rituximab treat-ment of adult patients with steroid-resistant focal segmental glomer-ulosclerosis. Clin J Am Soc Nephrol 4: 1317–1323, 2009

12. Gulati A, SinhaA, JordanSC,Hari P, DindaAK, SharmaS, Srivastava RN,Moudgil A, Bagga A: Efficacy and safety of treatment with rituximab fordifficult steroid-resistant and -dependent nephrotic syndrome: Multi-centric report. Clin J Am Soc Nephrol 5: 2207–2212, 2010

13. Pescovitz MD, Book BK, Sidner RA: Resolution of recurrent focal seg-mental glomerulosclerosis proteinuria after rituximab treatment.N Engl J Med 354: 1961–1963, 2006

14. Ravani P, Magnasco A, Edefonti A, Murer L, Rossi R, Ghio L, Benetti E,Scozzola F, Pasini A, Dallera N, Sica F, Belingheri M, Scolari F, GhiggeriGM: Short-term effects of rituximab in children with steroid- and cal-cineurin-dependent nephrotic syndrome: A randomized controlledtrial. Clin J Am Soc Nephrol 6: 1308–1315, 2011

15. Sethna C, Benchimol C, Hotchkiss H, Frank R, Infante L, Vento S,Trachtman H: Treatment of recurrent focal segmental glomerulo-sclerosis in pediatric kidney transplant recipients: Effect of rituximab.J Transplant 2011: 389542, 2011

16. Suri M, Tran K, Sharma AP, Filler G, Grimmer J: Remission of steroid-resistant nephrotic syndrome due to focal and segmental glomerulo-sclerosis using rituximab. Int Urol Nephrol 40: 807–810, 2008

17. Carroll MC, Holers VM: Innate autoimmunity. Adv Immunol 86: 137–157, 2005

18. Baumgarth N: The double life of a B-1 cell: Self-reactivity selects forprotective effector functions. Nat Rev Immunol 11: 34–46, 2011

19. Hayakawa K, Hardy RR, Herzenberg LA, Herzenberg LA: Progenitors forLy-1 B cells are distinct from progenitors for other B cells. J Exp Med161: 1554–1568, 1985

20. Hamaguchi Y, Uchida J, Cain DW, Venturi GM, Poe JC, Haas KM,Tedder TF: The peritoneal cavity provides a protective niche for B1 andconventional B lymphocytes during anti-CD20 immunotherapy inmice.J Immunol 174: 4389–4399, 2005

21. Haas MS, Alicot EM, Schuerpf F, Chiu I, Li J, Moore FD, Carroll MC:Blockade of self-reactive IgM significantly reduces injury in a murinemodel of acutemyocardial infarction.Cardiovasc Res 87: 618–627, 2010

22. Fleming SD, Shea-Donohue T, Guthridge JM, Kulik L, Waldschmidt TJ,Gipson MG, Tsokos GC, Holers VM: Mice deficient in complementreceptors 1 and 2 lack a tissue injury-inducing subset of the naturalantibody repertoire. J Immunol 169: 2126–2133, 2002

23. Zhang M, Austen WG Jr, Chiu I, Alicot EM, Hung R, Ma M, Verna N, XuM, Hechtman HB, Moore FD Jr, Carroll MC: Identification of a specificself-reactive IgM antibody that initiates intestinal ischemia/reperfusioninjury. Proc Natl Acad Sci U S A 101: 3886–3891, 2004

24. Weiser MR, Williams JP, Moore FD Jr, Kobzik L, Ma M, Hechtman HB,Carroll MC: Reperfusion injury of ischemic skeletal muscle is mediatedby natural antibody and complement. J ExpMed 183: 2343–2348, 1996

25. Renner B, Strassheim D, Amura CR, Kulik L, Ljubanovic D, GlogowskaMJ, Takahashi K, Carroll MC, Holers VM, Thurman JM: B cell subsetscontribute to renal injury and renal protection after ischemia/reperfusion.J Immunol 185: 4393–4400, 2010

26. Peterson LK, Tsunoda I, Fujinami RS: Role of CD5+ B-1 cells in EAEpathogenesis. Autoimmunity 41: 353–362, 2008

27. Wang Y, Wang YP, Tay YC, Harris DC: Progressive adriamycin ne-phropathy in mice: sequence of histologic and immunohistochemicalevents. Kidney Int 58: 1797–1804, 2000

28. Lenderink AM, Liegel K, Ljubanovi�c D, Coleman KE, Gilkeson GS,Holers VM, Thurman JM: The alternative pathway of complement isactivated in the glomeruli and tubulointerstitium of mice with adria-mycin nephropathy. Am J Physiol Renal Physiol 293: F555–F564, 2007

29. Fornoni A, Sageshima J, Wei C, Merscher-Gomez S, Aguillon-Prada R,Jauregui AN, Li J, Mattiazzi A, Ciancio G, Chen L, Zilleruelo G, Abitbol

C, Chandar J, Seeherunvong W, Ricordi C, Ikehata M, Rastaldi MP,Reiser J, Burke GW 3rd: Rituximab targets podocytes in recurrent focalsegmental glomerulosclerosis. Sci Transl Med 3: 85ra46, 2011

30. Schwartz MM: Focal Segmental Glomerulosclerosis. In: Heptinstall’sPathology of the Kidney, edited by Jennette JC, Olson JL, SchwartzMM, Silva FG, 6th Ed., Philadelphia, Lippincott Williams & Wilkins,2007, pp 155–204

31. Wang Y, Wang Y, Feng X, Bao S, Yi S, Kairaitis L, Tay YC, Rangan GK,Harris DC: Depletion of CD4(+) T cells aggravates glomerular and in-terstitial injury in murine adriamycin nephropathy. Kidney Int 59: 975–984, 2001

32. Ji-Yun Y, Melvin T, Sibley R, Michael AF: No evidence for a specific roleof IgM in mesangial proliferation of idiopathic nephrotic syndrome.Kidney Int 25: 100–106, 1984

33. Morita Y, Ikeguchi H, Nakamura J, Hotta N, Yuzawa Y, Matsuo S:Complement activation products in the urine from proteinuric patients.J Am Soc Nephrol 11: 700–707, 2000

34. D’Agati VD, Fogo AB, Bruijn JA, Jennette JC: Pathologic classificationof focal segmental glomerulosclerosis: A working proposal. Am J Kid-ney Dis 43: 368–382, 2004

35. D’Agati V: The many masks of focal segmental glomerulosclerosis.Kidney Int 46: 1223–1241, 1994

36. Turnberg D, Lewis M, Moss J, Xu Y, Botto M, Cook HT: Complementactivation contributes to both glomerular and tubulointerstitial damagein adriamycin nephropathy in mice. J Immunol 177: 4094–4102, 2006

37. Park P, Haas M, Cunningham PN, Bao L, Alexander JJ, Quigg RJ: Injuryin renal ischemia-reperfusion is independent from immunoglobulinsand T lymphocytes. Am J Physiol Renal Physiol 282: F352–F357, 2002

38. Velosa J, Miller K, Michael AF: Immunopathology of the end-stagekidney. Immunoglobulin and complement component deposition innonimmune disease. Am J Pathol 84: 149–162, 1976

39. Zwirner J, Felber E, Herzog V, Riethmüller G, Feucht HE: Classicalpathway of complement activation in normal and diseased humanglomeruli. Kidney Int 36: 1069–1077, 1989

40. Faul C, DonnellyM,Merscher-Gomez S, Chang YH, Franz S, DelfgaauwJ, Chang JM, Choi HY, Campbell KN, Kim K, Reiser J, Mundel P: Theactin cytoskeleton of kidney podocytes is a direct target of the anti-proteinuric effect of cyclosporine A. Nat Med 14: 931–938, 2008

41. Atkinson JP, Frank MM: Effect of cortisone therapy on serum comple-ment components. J Immunol 111: 1061–1066, 1973

42. Martin F, ChanAC: B cell immunobiology in disease: Evolving conceptsfrom the clinic. Annu Rev Immunol 24: 467–496, 2006

43. Lemercier C, Julen N, Coulpier M, Dauchel H, Ozanne D, Fontaine M,Ripoche J: Differential modulation by glucocorticoids of alternativecomplement protein secretion in cells of the monocyte/macrophagelineage. Eur J Immunol 22: 909–915, 1992

44. Jafar TH, Stark PC, Schmid CH, Landa M, Maschio G, de Jong PE, deZeeuw D, Shahinfar S, Toto R, Levey AS; AIPRD Study Group: Pro-gression of chronic kidney disease: The role of blood pressure control,proteinuria, and angiotensin-converting enzyme inhibition: A patient-level meta-analysis. Ann Intern Med 139: 244–252, 2003

45. Kriz W, LeHir M: Pathways to nephron loss starting from glomerulardiseases-insights from animal models. Kidney Int 67: 404–419, 2005

46. Kulik L, Fleming SD, Moratz C, Reuter JW, Novikov A, Chen K, AndrewsKA, Markaryan A, Quigg RJ, Silverman GJ, Tsokos GC, Holers VM:Pathogenic natural antibodies recognizing annexin IV are required todevelop intestinal ischemia-reperfusion injury. J Immunol 182: 5363–5373, 2009

47. Sargsyan SA, Serkova NJ, Renner B, Hasebroock KM, Larsen B, StoldtC, McFann K, Pickering MC, Thurman JM: Detection of glomerularcomplement C3 fragments by magnetic resonance imaging in murinelupus nephritis. Kidney Int 81: 152–159, 2012



Documents

IgM Contributes to Glomerular Injury in FSGS