Murine and Human Lupus Nephritis: Pathogenic Mechanisms and Theoretical

Strategies for Therapy

Hege Lynum Pedersen, PhD,* Kjersti Daae Horvei, MD,* Dhivya Thiyagarajan, PhD,*

Natalya Seredkina, MD, PhD,* and Ole Petter Rekvig, MD, PhD*,†

Summary: Lupus nephritis is one of the most serious manifestations of systemic lupus erythematosus, and

0270-9295/ -& 2015 Elsevhttp://dx.doi.o

Conflict of in

*RNA and MMedical BiTromsø, Tr

†DepartmentTromsø, No

Address reprMolecularBiology, FN-9037 Tro

Financial sufinancial m

Seminars in N

represents one of the criteria implemented to classify systemic lupus erythematosus. Although studied fordecades, no consensus has been reached related to the basic cellular, molecular, and immunologicmechanism(s) responsible for lupus nephritis. No causal treatments have been developed; therapy isapproached mainly with nonspecific immunosuppressive medications. More detailed insight into diseasemechanisms therefore is indispensable to develop new therapeutic strategies. In this review, contemporaryknowledge on the pathogenic mechanisms of lupus nephritis is discussed based on recent data in murine andhuman lupus nephritis. Specific focus is given to the effect of anti–double-stranded DNA/antinucleosomeantibodies in the kidneys and whether they bind exposed chromatin fragments in glomeruli or whether theybind inherent glomerular structures by cross-recognition. Overall, the data presented here favor the exposedchromatin model because we did not find any indication to substantiate the anti–double-stranded DNAantibody cross-reacting model. At the end of this review we present data on why chromatin fragments areexpressed in the glomeruli of patients with lupus nephritis, and discuss how this knowledge can be used todirect the development of future therapies.Semin Nephrol 35:427-438 C 2015 Elsevier Inc. All rights reserved.Keywords: Systemic lupus erythematosus, murine/human lupus nephritis, DNase I, heparin, chaperonemolecules, therapeutic strategies

Systemic lupus erythematosus (SLE) is a systemicautoimmune disease characterized by variousaberrant clinical and biological parameters.1–4

A characteristic phenomenon in SLE is the presence ofautoantibodies against double-stranded DNA (dsDNA),histones, nucleosomes, and chromatin.1,5,6 Renal accu-mulation of antinuclear antibodies by direct binding tointrinsic renal antigens, or in complex with chromatinantigens, induces severe kidney inflammation analogousto a type II or type III immune-mediated hypersensitivityreaction.7,8 To understand the nature of the processes thataccount for lupus nephritis and to develop new specifictreatment modalities, we need to determine the nature ofthe renal target structures for anti-dsDNA antibodies andthe processes that account for their exposure.

Since 1957, the year anti-dsDNA antibodies werediscovered in an autoimmune context,9–12 they have

see front matterier Inc. All rights reserved.rg/10.1016/j.semnephrol.2015.08.004

terest statement: none.

olecular Pathology Research Group, Department ofology, Faculty of Health Sciences, University ofomsø, Norway.of Radiology, University Hospital of North Norway,rway.

int requests to Hege Lynum Pedersen, RNA andPathology Research Group, Department of Medicalaculty of Health Sciences, University of Tromsø,msø, Norway. E-mail: hege.lynum.pedersen@uit.no

pport: Supported by the University of Tromsø asilieu support (O.P.R.).

ephrology, Vol 35, No 5, September 2015, pp 427–438

been linked to SLE3,4,13,14 and to lupus nephritis.7,15

International consensus has concluded that this auto-antibody family is central to the pathogenesis of lupusnephritis.16,17 However, how these antibodies partic-ipate in the pathogenesis of lupus nephritis has beenand remains controversial.7 The reason for this issimple. Many patients produce anti-dsDNA antibodies,however, of these patients, many do not develop lupusnephritis. Therefore, a unique property must existamong the antibodies that make them nephritogenic,16

or, as an alternative, all anti-dsDNA antibodies havenephritogenic potential, but this is manifest only inindividuals in whom the chromatin fragments, thetarget for anti-dsDNA antibodies, are exposed andaccessible in glomeruli.18

These alternatives have resulted in two main direc-tions in the study of the pathogenesis of lupus nephritis.One is dominated by evidence that the anti-dsDNAantibodies cross-react with intrinsic renal antigens, suchas phospholipids,19–21 laminin or the extracellularmatrix,22–25 entactin,26 α-actinin,27 annexin II,28 riboso-mal P protein,29 vimentin,30 or others. Whether theantilaminin antibodies detected in the urine of lupusnephritis patients22 really cross-reacts with DNA wasnot investigated, however, in other studies, such cross-reactions have been suggested.23,25 Lupus nephritis maydevelop only in patients with such cross-reacting anti-dsDNA antibodies.

In the alternative model, antibodies comprising thewhole spectrum of specificities of dsDNA, as found inchromatin fragments, such as elongated or highly bentDNA,31,32 may initiate lupus nephritis, but only when

427

H.L. Pedersen et al.428

chromatin fragments are exposed in the glomeruli. Thisis not considered in the Witebsky criteria that classify adisease as autoimmune in nature.33 These criteriarequire three types of evidence of the pathogenicityof an autoimmune factor: direct evidence from thetransfer of a pathogenic antibody and/or a T cell,indirect evidence based on replication of the auto-immune disease in experimental animals, and circum-stantial evidence from clinical parameters. If thealternative chromatin model is correct then an addi-tional Witebsky criterion may include showing thatchromatin fragments are exposed in affected organs.

Central to understanding these two models is iden-tifying the characteristics of a pathogenic anti-dsDNAantibody, and determining the origin and character-istics of cross-reacting renal antigens and/or chromatinfragments retained in the glomeruli and targeted byanti-dsDNA antibodies. Linked to this is defining theexact role of the silencing of renal DNase I inprogressive lupus nephritis.34–36 This is potentiallyimportant because silencing DNASE1 gene expressionis central to chromatin exposure in kidneys.37,38 Boththe cross-reacting and the chromatin models are attrac-tive, and provide a basis to explain how anti-dsDNAantibodies may initiate and maintain lupus nephritis.

In this review, we present data that favor theexposed chromatin model. Experimental and observa-tional data developed in the autoimmune, lupus-prone(NZB � NZW)F1 mouse strain will be compared withanalyses of renal biopsy specimens from patients withlupus nephritis. The data are discussed in the context ofnew causal therapy modalities that are expected toeventually confirm the chromatin model.

SYSTEMIC LUPUS ERYTHEMATOSUS: ONE ORSEVERAL DISEASES?

The etiology of SLE is unknown. Moreover, one mayraise the provocative question as to whether humanSLE is one disease entity, or a mixture of individual,etiologically unrelated organ manifestations as definedby the American college of Rheumatology3 or theSystemic Lupus International Collaborating Clinicsclassification criteria4 for SLE. The classification cri-teria do not appear to reflect a common pathogenicprocess, so it is not clear how genetic aberrancies andbiomarkers can be associated with SLE when SLErepresents such a divergent mixture of phenotypes. Forexample, evidence of autoimmunity to nucleosomes,particularly to the individual components of nucleo-somes, such as native (ds)DNA and histones, is animportant diagnostic criteria for SLE.3,4,39 In addition,autoantibodies to dsDNA have the potential to inducenephritis.40,41 However, although anti-dsDNA antibod-ies have a strong pathogenic potential in SLE, this

potential appears to be related only to lupus nephritis(see studies by Seredkina et al,7 Krishnan et al,26 VanBruggen et al,42 and Berden et al43), lupus dermatitis,44–46

and possibly certain forms of cerebral lupus.47–49

MURINE AND HUMAN LUPUS NEPHRITIS:NEW INSIGHTS

We studied the evolution of lupus nephritis by serialexamination of kidneys of lupus-prone (NZB � NZW)F1 and observed a two-step process in the pathogenesisof murine lupus nephritis. First, a mild mesangialnephritis developed simultaneously with the appear-ance of anti-dsDNA antibodies. Later, the diseaseprogressed to a membranoproliferative nephritis withchromatin-IgG immune-complex deposition in theglomerular basement membrane (GBM). As the dis-ease progressed, the DNASE1 gene was silenced,followed by a profound increase of proteinuria.34

In human lupus nephritis, it is not clear whetherclasses II through IV represent different directions oflupus nephritis, or if the natural course of lupusnephritis is to progress from one class to another,similar to the steady progression seen in the (NZB �NZW)F1 mouse. Mesangial proliferative nephritis(class II) generally has been considered a mild formwithout progression, and with a 10-year renal survivalrate of 100%.50 However, two recent studies assessingthe course of class II lupus nephritis showed progres-sion from class II to class III or IV despite treatment.Lee et al51 found progression from class II to class IIIor IV in 5 of 15 patients over a mean of 5 years, and,earlier, Tam et al52 described poor prognosis, reportedas progression in 9 of 19 patients originally diagnosedwith class II nephritis. Although these were smallstudies, and the exact progression rate of class II lupusnephritis has yet to be determined, these data supportthe continuous progressive model in at least somepatients with lupus nephritis.

Murine Lupus Nephritis

Given the fact that nephritis is a serious manifestation ofSLE,7,40,41,53 it is important to determine by whichpathways anti-dsDNA antibodies act as pathogenic fac-tors. Parameters that historically have been regarded asimportant in determining the nephrogenicity of anti-dsDNA antibody subpopulations are antibody avidityfor DNA, specificity for unique DNA or nucleosomalstructures,31,32,54–57 as well as cross-reactivity with inher-ent renal or non-nucleosomal DNA molecules.23,24,27,58–62

The murine data were reviewed recently.2,7,63 In onecentral study, we focused on the pathogenic processesin kidneys taken at time intervals from lupus-prone(NZB � NZW)F1 mice. For these studies we devel-oped high-resolution techniques that provided evidence

Murine and human lupus nephritis 429

that nephritogenic anti-dsDNA and antinucleosomeantibodies selectively recognize intraglomerular extrac-ellular chromatin fragments in vivo.18,34,64,65 Thesechromatin fragments appear as electron-dense struc-tures (EDS) by electron microscopy and are associatedwith the GBM. Autoantibody deposits detected in vivoare localized strictly to these structures.

EDS in glomeruli were first described by Comerfordet al66 in 1968, and confirmed in 1975 and 1979 byDillard et al67 and Ben-Bassat et al,68 respectively.These studies did not characterize the content of EDS.However, according to recent findings, it is fair tosuggest that the three pioneering studies performed byComerford et al,66 Dillard et al,67 and Ben-Bassatet al68 represent high-resolution studies on the natureof immune complexes seen in glomeruli in lupusnephritis. In line with this, in our studies IgG boundin vivo in EDS. This observation lead us to hypothe-size that EDS contained an ectopic target structure forthe nephritogenic anti-dsDNA antibodies because weassumed that intrinsic renal antigens are not electron-dense. This assumption derives from the fact that inGoodpasture syndrome, the nephritis component of thesyndrome is caused by binding of anticollagen auto-antibodies to collagen 4 within the GBM.69,70 Thisbinding appears linear by direct immunofluorescence,not granular as in lupus nephritis, and EDS have notbeen reported in Goodpasture syndrome. Thus, weassumed that the target for nephritogenic anti-dsDNAantibodies is not an intrinsic renal antigen that nor-mally is exposed, but rather is ectopically expressedglomerular chromatin.18 We confirm that anti-dsDNAantibodies form complexes with chromatin fragmentsin glomerular basement membranes and matrices, asdiscussed later.7,43,71–75

What Does IgG Target in the Glomeruliof Lupus-Prone Mice?

Our early analyses to elucidate the nature of the targetsfor nephritic anti-dsDNA antibodies were founded onthe hypothesis that only anti-dsDNA antibodies thatcross-react with inherent glomerular structures werepathogenic. To prove this we used high-resolutiontechniques to trace where and to what antigensnephritogenic anti-dsDNA antibodies bound in glo-meruli. These techniques included transmission elec-tron microscopy,76 immune electron microscopy(IEM),18 co-localization IEM,18 and co-localizationterminal deoxynucleotidyl transferase–mediated deox-yuridine triphosphate nick-end labeling (TUNEL)IEM,18 in addition to classic immunologic assays,measurements of apoptosis in renal cells, and studiesof renal endonucleases.17 Although we anticipatedfinding evidence for the cross-reacting model, theresults instead showed that IgG antibodies in lupus

nephritis targeted chromatin fragments irrespective ofwhether they were in the mesangium or GBM. In vivo–bound IgG co-localized with the following: structuresthat bound experimental dsDNA antibodies, histones,and transcription factors in vitro,18,76 and withTUNEL-positive material that contained nickedDNA.18 IgG molecules eluted from the kidneys werespecific for DNA, histones, and nucleosomes.23,25,71,76

We did not observe in vivo–bound IgG outside EDS,or in association with regular basement membranestructures, as is the case in Goodpasture syndrome.Because antibodies bound within EDS and not out-side,18 this argued against the cross-reaction model,and suggested that murine lupus nephritis is initiatedand maintained by antibodies recognizing structuresexposed in chromatin fragments.

Central Role of Anti-dsDNA Antibodies, Renal DNase I,Chromatin Fragments, and Matrix Metalloproteases inthe Evolution of Murine Lupus Nephritis

Anti-DNA antibodies,34 renal DNase I,34,35,37,77 andmatrix metalloprotease (MMP)78,79 messenger RNAlevels and enzyme activities are instrumental in earlyand late events in murine lupus nephritis.34,35,77 Earlyphases of nephritis are associated with chromatin–IgGcomplex deposition in the mesangial matrix, whichcorrelates with the appearance of anti-dsDNA antibodiesand clinically silent or mild mesangial nephritis.34,80

Subsequently, renal DNase I message and enzymeactivity decreases, whereas MMP2 message and enzymeactivity increase.34 A reduction of renal DNase I levelsis coincident with deficient fragmentation of chromatinfrom dead cells and the retention and accumulation ofchromatin in the GBM. Similar observations have notbeen described in other experimental nuclease deficien-cies.81–85 Increased expression of MMP2 may beexplained by chromatin stimulation of Toll-like recep-tors in, for example, dendritic cells.86–90 MMPs arecollagenases with the potential to disintegrate basementmembranes, and this could facilitate deposition of largeimmune complexes in the GBM.

HUMAN LUPUS NEPHRITIS: THE SAMEPATHOGENESIS AS FOR MURINELUPUS NEPHRITIS?

Our studies of human lupus nephritis were based onthese findings from murine disease. We analyzed biopsyspecimens from patients with lupus nephritis using thesame strategies that had been applied to murine lupusnephritis (Table 1 shows the comparative analyticapproaches in murine and human lupus nephritis).

Transmission electron microscopy and IEM showedthat EDS were present in the mesangial matrix and

Table 1. Analytic Approach to Study Loci and Composition of Immune Complexes in Murine and Human Lupus Nephritis

Assay Murine Lupus Nephritis Human Lupus Nephritis

Antibodies bound in glomeruli in vivo Confocal microscopy/ DIF/IHC*18,35 DIF/IHC94

Immune complex characterization TEM18,76 TEM64

IEM18 IEM64

Co-localization IEM18 Co-localization IEM64

Co-localization TUNEL IEM118 Co-localization TUNEL IEM64

Analyses of TLR7-TLR9, MMP2, and MMP9 Protein expression in situ/qPCR/IHC94 Protein expression in situ94

Proinflammatory cytokines qPCR37 qPCR37

Renal DNaseI and Trap 1 expression qPCR, IHC, WB35,37,77 qPCR, IHC94

Determination of affinity between chromatinfragments and membrane structures

Surface plasmon resonance122 Surface plasmon resonance45

Abbreviations: DIF, direct immunofluorescence; IHC, immunohistochemistry; TEM, transmission electron microscopy; qPCR, quanti-tative polymerase chain reaction; WB, Western blot; Trap 1, tumor necrosis factor–receptor–associated protein-1.

H.L. Pedersen et al.430

GBM of class IV biopsy specimens.91,92 GlomerularIgG was strictly confined to these EDS.64,78 Co-localization IEM showed that when experimental anti-bodies against dsDNA, histones, transcription factor–associated TATA-binding protein, and cyclic adeno-sine monophosphate response element-binding proteinwere added to sections of human kidney, they co-localized with glomerular in vivo–bound IgG autoanti-bodies. One problem with the co-localization assays isthat the experimental antibodies, similar to true auto-antibodies, may be cross-reactive. The results from co-localization IEM were supported by an antibody-independent assay, co-localization TUNEL IEM(Table 1). This approach showed that in vivo–boundIgG was binding to structures that contained nickedDNA, similar to murine lupus nephritis.

One striking observation that emerged from themurine studies was the selective silencing of the renalDNase I gene.34,36,77,93 In human kidney biopsy speci-mens we observed an analogous silencing of renalDNase I,36,94 together with accumulation of chromatinfragments in the mesangium and GBM.64

DNase I is the dominant renal endonuclease andaccounts for 80% of the total endonuclease activity inthe kidneys. In mice, the loss of DNase I endonucleaseactivity was associated with a marked reduction inchromatin fragmentation, and the subsequent accumu-lation of large chromatin fragments in glomeruli.36,93

Because this event occurred in the kidneys, and DNaseI and other endonuclease activities were normal inother organs,35 it seemed reasonable to assume thatexposed chromatin fragments found in glomeruliderive from the kidneys themselves.

MURINE AND HUMAN LUPUS NEPHRITIS:TRANSLATING PATHOGENESIS TO THERAPY

The data obtained in our studies of murine and humanlupus nephritis provide an approach to understand basicmolecular and immunologic processes accounting for

antibody-mediated nephritis in human SLE. Notably, allelements of the mechanisms that account for murinelupus nephritis have been shown to be operational inhuman lupus nephritis. The pathogenesis of the chro-matin model may be simplified to a two-step process.First, early anti-dsDNA antibodies accumulate in themesangium and create a local and silent inflammation.After this, the renal DNase I gene is silenced, followedby reduced fragmentation of chromatin from dead anddying cells. Therefore, instead of silent clearance ofchromatin, chromatin is retained in glomeruli and bindsto basement membranes and matrices as a complex withIgG. This will promote more severe inflammation and,consequently, the progression of lupus nephritis intosevere organ failure.

This model suggests that preventing the silencing ofthe renal DNase I gene and/or preventing chromatinbinding to basement membranes and matrices may berelevant therapeutic approaches for human lupus neph-ritis. Maintaining renal DNase I activity may hamperprogressive lupus nephritis, whereas interfering withchromatin binding may prevent the development ofboth mesangial and progressive nephritis (Fig. 1).

THERAPY AIMED TO PREVENT SILENCING OF THERENAL DNASE I GENE: BACKGROUNDINFORMATION AND THERAPEUTIC STRATEGIES

Aberrations in DNase I expression have been associatedwith SLE and the production of antinuclear antibodies atleast since 1968.95 Since then, reduced expression of theDNase I enzyme or mutations in the DNASE1 gene havebeen linked to SLE.96,97 Our discovery of an acquiredsilencing of the renal DNASE1 gene in the context oflupus nephritis that was followed consistently bydeposition of large chromatin fragments in complexwith IgG antichromatin antibodies in the GBM andmesangial matrix18,34,35,76 is consistent with theseearly observations. Loss of renal DNase I may be the

Figure 1. The exposed chromatin model of lupus nephritis. Wepropose that the overall mechanism in the exposed chromatinmodel is similar for both human and murine lupus nephritis. Ingeneral, a two-step process occurs. (A) First, a mild mesangialnephritis develops together with the appearance of anti-dsDNAantibodies, (B) followed by membranoproliferative nephritis withchromatin–IgG complex deposition in the GBM. (B) As lupusnephritis progresses into membranoproliferative nephritis, therenal DNase I enzymatic activity is silenced, with subsequentaccumulation of chromatin fragments in the glomeruli, andincreased proteinuria. Therapy 1 and 2 in the figure refer tothe therapy models described in the text. Briefly, therapy 1 isaimed at preventing the down-regulation of DNase I geneexpression, whereas therapy 2 targets chromatin structure toprevent binding of chromatin fragments to the mesangial matrixand the GBM.

Murine and human lupus nephritis 431

adverse event that promotes progression of (class II)lupus nephritis.

DNase I is an endonuclease that cleaves DNA inchromatin. It exerts its dominant role in degradation ofchromatin in the context of cell death. Traditionally, ithas been linked to apoptosis,82,98,99 but it also plays arole in necrosis.98,100 In addition, DNase I also issecreted,101 so it could have a role in the digestion ofextracellular DNA.101,102 For example, Nagata andKawane103 point out that undigested DNA in macro-phages, whether autologous or viral, activates theinnate immune system and causes strong inflammation,resulting in inflammatory diseases.

One way to analyze the effect of DNase I in lupusnephritis would be to inject the enzyme and todetermine if this reduces exposure of chromatin inthe kidneys. This has been tested in both experimental

animals and in patients with lupus nephritis.7,97,104–106

One promising result was published by Macanovicet al.104 They injected DNase I in young prenephritic(NZB � NZW)F1 mice and observed that diseaseprogression was retarded. Treatment of nephritic micereduced levels of proteinuria and serum creatinine,however, these observations were not pursued. Inanother study, injection of DNase I in (NZB �NZW)F1 mice did not improve early or late stages ofmurine lupus nephritis.106 In a human study performedby Davis et al,107 patients with lupus nephritis classesIII and IV were given intravenous or subcutaneousrecombinant human DNase I. This treatment did notshow any effect on kidney function or disease activ-ity.107 These findings show that exogenously adminis-tered DNase I has little or no influence on chromatinfragmentation in dying renal cells, and suggest thatintracellular DNase I is responsible for the safe andeffective degradation of chromatin in dying cells.82,99

The reason why extracellular DNase I may not beeffective for degradation of extracellular chromatincould be explained by two facts. First, chromatinpossesses a fairly high affinity for laminin and colla-gens.45,108 Second, extrapolating from informationprovided by DNase footprinting assays,109,110 theinteraction of chromatin with proteins may protectchromatin from being degraded enzymatically.

This assumption is strengthened by the strongtherapeutic effect of DNase I in the chronic lungdisease cystic fibrosis. Extracellular chromatin frommassive neutrophil extracellular trap (NET)-osis ofrecruited neutrophils contributes to increased viscosityof purulent secretions in the respiratory tract. Inhala-tion of recombinant human DNase I reduces viscositywithin minutes, and is a standard chronic treatment forcystic fibrosis to reduce excerbations.111,112 The reasonwhy treatment with exogenous DNase I is efficient inclearing extracellular chromatin in cystic fibrosis, butnot beneficial in SLE and lupus nephritis, may lie inthe fact that chromatin in cystic fibrosis is embedded inmucous and are more accessible for enzymatic degra-dation, in contrast to basement membrane–boundchromatin in lupus nephritis.

Exactly how DNase I expression is regulated is notunderstood, despite the fact that the enzyme wasdescribed in 1946 by McCarty.113 Nonetheless, thisinformation will be critical for developing DNase I as atherapy for lupus nephritis. We recently approachedthis problem based on the observation that silencing ofrenal DNase I generally follows mesangial nephritis.This suggested that proinflammatory cytokines couldregulate DNase I. We found that hypoxia and tumornecrosis factor α, but not interferon-γ, interleukin (IL)-1β, IL-6, or IL-10, possess the potential to up-regulateDNase I expression in a human tubular cell line

H.L. Pedersen et al.432

(Thiyagarajan et al, unpublished data), but the keymediators that down-regulate DNase I in tubular cellsremain unknown. This is important to pursue becausetubular cells are the dominant producers of DNase I inthe kidneys.

HEPARIN FUNCTIONS AS A CHAPERONE FORCHROMATIN AND ENHANCES ENZYMATICCHROMATIN DEGRADATION

Several factors are involved in chromatin remodeling.For example, it has been shown that chaperonemolecules such as nucleosome assembly protein1,114,115 nucleoplasmin,116,117 and Hsp90118,119 inducechanges in chromatin conformation through the proc-ess of assembly/disassembly of chromatin fragments,and in so doing increase the susceptibility of chromatinto enzymatic degradation. In the compact structure ofchromatin, digestion with DNase I can degrade linkerDNA, but not the core nucleosomal DNA, which isprotected by the histone core octamer and by nonhi-stone proteins.120 The DNA wrapped around thehistone core octamer therefore is not digested. Sim-ilarly, the core histones H2A, H2B, H3, and H4 havebeen shown to resist enzymatic degradation. Trypsini-zation removed only the N and C terminal histone tailsprotruding from the nucleosome surface, while the coreglobular parts and core nucleosome structure remainedlargely intact.120 Chaperone molecules, however, canopen even this restricted core structure, making itaccessible for proteinases and nucleases. As such,chaperone molecules could be therapeutic candidatesto enhance enzymatic degradation of ectopic chromatinin vivo, but may not be tolerated in doses necessary toachieve the desired effect.

As an alternative to chaperone therapeutics, it maybe possible to exploit the potential of heparin toenhance enzymatic degradation of chromatin frag-ments. Heparin has been shown to have an effect onchromatin structure similar to that of certain chaperonemolecule classes. Heparin binds tightly to histone tailsand changes the net surface-oriented charge of thenucleosome. It also may interfere with binding ofchromatin fragments, and even immune complexesconsisting of chromatin fragments and IgG, to struc-tures within the GBM and the mesangial matrix. Thesedata were obtained using plasmon resonance analysesof chromatin fragments’ affinity for laminins andcollagens.45,108 Isolated DNA did not bind to thesestructures.108 Heparin binds the trypsin-sensitive sol-vent-phase N and C terminal tails of core histones.121

The interaction of heparin and core histones increasesthe sensitivity of nucleosomes and chromatin fornucleases.102,121,122 Furthermore, heparin disrupts the

nucleosome structure, and, linked to this effect,increases enhancer–promoter communication123 bydisassembling the chromatin structure.124,125

In other words, heparin may have two beneficialeffects in lupus nephritis. First, by facilitating enzymaticdegradation of chromatin, heparin may preclude expo-sure of chromatin to the immune system in vivo, andthereby prevent or reduce the production of pathogenic,chromatin-specific autoantibodies and the accumulationof immune complexes in kidneys. In addition, thechaperone effect of heparin on chromatin structuremay preclude binding of chromatin fragment–IgG com-plexes to central structures in basement membranes andmatrices, such as laminin and collagen. The latterprocess has been tested experimentally by Van Bruggenet al,126 who showed that interaction of heparin orheparin analogs with immune complexes containingnucleosomal antigens prevents the binding of theseimmune complexes to the GBM. Importantly, theyfurther showed that nephritis in the MRL/lpr mousemodel of lupus nephritis was delayed by treatment withheparins.126 Their data are consistent with the idea thatheparin prevents binding of immune complexes to theGBM and also presumably to the mesangial matrix.126

We therefore performed in vitro and in vivo analy-ses to determine the effects of heparin on enzymaticchromatin degradation, the kinetics of anti-dsDNAantibody production, and on the progression of murinelupus nephritis.122 These studies showed that heparinincreased the sensitivity of chromatin to DNase Idigestion (Fig. 2A). Likewise, proteinase digestion ofchromatin was faster in the presence of heparin.122

These in vitro results then were compared with theeffects of heparin in (NZB � NZW)F1 mice. Thekinetics of anti-dsDNA antibody production, deposi-tion of chromatin fragment–IgG complexes in GBM,and proteinuria were followed up. The data showeddelayed anti-dsDNA antibody production and redu-ced antibody titers (Fig. 2B). This was accompanied byreduced chromatin fragment–IgG complex accum-ulation in the GBM and the mesangial matrix.122

Importantly, these effects occurred at doses belowthose needed for heparin’s anticoagulant activity.

PREVENTING THE BINDING OF CHROMATINFRAGMENTS TO GBM OR MESANGIAL MATRIX

In these studies we used surface plasmon resonance todetermine if heparin could prevent nucleosome bindingto laminin and collagen. The data showed that heparinnot only prevented binding of nucleosomes to GBMcomponents in vitro122 (Fig. 3), but that heparinpromoted dissociation of these complexes from lam-inin and collagen after stable binding.122 Similarly,binding of nucleosomes in complex with IgG

Nuc anti-DNA mAb- - - +- + - -

30" 30"

Nuc

Dnase1 digestion time: - - - 30" 1’ 2' 30" 1’ 2’ 30" 1’ 2'

1200bp

1000bp

500bp

100bp

Hep 100:1 - - + - - - - - - + + +Hep 10:1 - + - - - - + + + - - -

A

Hep treated from 12 weeks

15 20 25 300.0

0.5

1.0

1.5BW control

BW heparin treated

Age (weeks)

OD

492

at 1

/100

ser

um d

ilutio

n

B

Figure 2. Heparin renders Nucleosomes (NUC) more sensitive to DNase I degradation. (A) Nucleosomes withoutheparin appeared as mononucleosomes up to oligonucleosomes in agarose gel electrophoresis. In the presenceof heparin, these migrated faster, corresponding to mononucleosomes and dinucleosomes. DNase I digestion ofnucleosomes without heparin yielded a limited digest representing core nucleosomes, whereas digestion with 10molar excess of heparin resulted in complete degradation of DNA after 2 minutes of incubation. Increasing themolar excess of heparin to 100-fold, DNA is degraded completely after 30 seconds. (B) The effect of heparin in the(NZBxNZW)F1 (BW) murine lupus nephritis model on anti-dsDNA antibody production and progressive kidneydisease was tested by giving the mice heparin subepidermally with diffusion pumps. Each mouse received 50 μg ofenoxaparin (low-molecular-weight heparin) per day. Anti-dsDNA antibody production in vivo was delayed onaverage by 10 weeks (compared with optical density (OD) 0.2) in the heparin-treated group of (NZB � NZW)F1mice (in which heparin treatment was initiated at the age of 12 weeks) compared with the early appearance ofantibodies in the untreated control group. The difference was highly significant (P o .0001 by paired t test). mAb,monoclonal antibody. Modified from Hedberg et al.122

Murine and human lupus nephritis 433

antichromatin antibodies to membrane componentsalso was inhibited by heparin. In addition, correspond-ing results were obtained when comparing data fromthis experimental system using either of the laminin orcollagen structures. Notably, isolated DNA had noaffinity for these GBM structures, and sets of IgG-chromatin antibodies including anti-dsDNA antibodiesdid not bind to laminin, as has been reported pre-viously.127 These data are consistent with data fromother investigators who showed that nucleosomes bindGBM in kidneys in the context of lupus nephri-tis,73,128,129 and that this binding could be inhibitedby heparin and heparinoids.126,130

In summary, the beneficial effects of heparin onlupus nephritis are enhanced degradation and elimina-tion of chromatin fragments, reduced anti-dsDNAantibody production in vivo, and prevention of chro-matin–IgG complex binding to basement membranestructures. Taken together, all of these effects of

heparin may be regarded as ex juvantibus confirmationof the chromatin model of murine lupus nephritis.Because the role of chromatin fragment–IgG com-plexes in murine lupus nephritis seems to be highlyanalogous to early and late human lupus nephri-tis,36,38,64,94,131 the promising results of treatment ofmurine lupus nephritis with low-molecular-weightheparin should foster new therapeutic strategies andclinical trials along these lines (Fig. 1).

CONCLUDING REMARKS

Deeper insight into processes that are operational inprogressive lupus nephritis is imperative to develop newcausal therapy strategies. Currently, no such treatmentmodalities are available. As discussed here, we haveincreased our insights into specific aspects of lupusnephritis, and from these insights we have proposedways to directly block specific pathogenic processes.

Anti-DNAmAbs 500nMnucleosomes 50 nMnucleosome: anti-DNA mAb 1:10nucleosome: anti-DNA mAb: Hep 1:10:100

RU

Time (s)

0

200

400

600

800

0 200 400 600 800 1000 1200

C Collagen IV chip

100

300

500

0 400 800 1200

RU

Time (s)

B Laminin chip

nucleosomes 500nMnucleosome: hep 1:1

nucleosome: hep 1:10nucleosome: hep 1:100

RU

100

300

500

700

0 200 400 600 800 1000 1200

RU

A Collagen IV chip

Time (s)nucleosomes 500nMnucleosome: hep 1:1nucleosome: hep 1:10nucleosome: hep 1:100

Figure 3. Heparin has the potential to inhibit in vitro binding of nucleosomes and nucleosome–IgG complexes toglomerular basement components. Surface plasmon resonance analyses show that nucleosomes (500 nmol/L)bind collagen IV and laminin. This binding is inhibited by heparin in a dose-dependent manner (A, collagen IV; B,laminin). (C) Nucleosomes in complex with anti-DNA monoclonal antibodies (mAbs) (1:10 molar ratio) bind laminin(not shown) and collagen IV, and the binding is inhibited by heparin. Similarly, binding of nucleosomes in complexwith antihistone mAbs to collagen IV also is inhibited by heparin.122 Similar results were observed whensubstituting the collagen IV with the laminin chip. The amount of the analytes in the mixtures is provided asmolar ratios. RU ¼ response unit. Modified from Hedberg et al.122

H.L. Pedersen et al.434

We suggest that studies to investigate these therapeuticapproaches in human lupus nephritis are warranted.

REFERENCES1. Rahman A, Isenberg DA. Systemic lupus erythematosus. N

Engl J Med. 2008;358:929-39.2. Rekvig OP, van der Vlag J. The pathogenesis and diagnosis of

systemic lupus erythematosus: still not resolved. SeminImmunopathol. 2014;36:301-11.

3. Tan EM, Cohen AS, Fries JF, Masi AT, McShane DJ,Rothfield NF, et al. The 1982 revised criteria for theclassification of systemic lupus erythematosus. ArthritisRheum. 1982;25:1271-7.

4. Petri M, Orbai AM, Alarcon GS, Gordon C, Merrill JT, FortinPR, et al. Derivation and validation of the Systemic LupusInternational Collaborating Clinics classification criteria forsystemic lupus erythematosus. Arthritis Rheum. 2012;64:2677-86.

5. van Bavel CC, Fenton KA, Rekvig OP, van der Vlag J,Berden JH. Glomerular targets of nephritogenic

autoantibodies in systemic lupus erythematosus. ArthritisRheum. 2008;58:1892-9.

6. Rahman A, Hiepe F. Anti-DNA antibodies–overview ofassays and clinical correlations. Lupus. 2002;11:770-3.

7. Seredkina N, van der Vlag J, Berden J, Mortensen E, RekvigOP. Lupus nephritis: enigmas, conflicting models and anemerging concept. Mol Med. 2013;19:161-9.

8. Jancar S, Sanchez CM. Immune complex-mediated tissueinjury: a multistep paradigm. Trends Immunol. 2005;26:48-55.

9. Ceppellini R, Polli E, Celada F. A DNA-reacting factor inserum of a patient with lupus erythematosus diffusus. ProcSoc Exp Biol Med. 1957;96:572-4.

10. Robbins WC, Holman HR, Deicher H, Kunkel HG. Comple-ment fixation with cell nuclei and DNA in lupus erythemato-sus. Proc Soc Exp Biol Med. 1957;96:575-9.

11. Seligmann M. [Demonstration in the blood of patients withdisseminated lupus erythematosus a substance determining aprecipitation reaction with desoxyribonucleic acid]. C R HebdSeances Acad Sci. 1957;245:243-5.

12. Miescher P, Strassle R. New serological methods for thedetection of the L.E. factor. Vox Sang. 1957;2:283-7.

Murine and human lupus nephritis 435

13. Hahn BH. Antibodies to DNA. N Engl J Med. 1998;338:1359-68.

14. Rekvig OP. Anti-dsDNA antibodies as a classification crite-rion and a diagnostic marker for SLE: critical remarks. ClinExp Immunol. 2015;179:5-10.

15. Lorenz G, Desai J, Anders HJ. Lupus nephritis: update onmechanisms of systemic autoimmunity and kidney immuno-pathology. Curr Opin Nephrol Hypertens. 2014;23:211-7.

16. Waldman M, Madaio MP. Pathogenic autoantibodies in lupusnephritis. Lupus. 2005;14:19-24.

17. Mortensen ES, Rekvig OP. Nephritogenic potential of anti-DNA antibodies against necrotic nucleosomes. J Am SocNephrol. 2009;20:696-704.

18. Kalaaji M, Mortensen E, Jorgensen L, Olsen R, Rekvig OP.Nephritogenic lupus antibodies recognize glomerular base-ment membrane-associated chromatin fragments released fromapoptotic intraglomerular cells. Am J Pathol. 2006;168:1779-92.

19. Lafer EM, Rauch J, Andrzejewski C Jr, Mudd D, Furie B,Furie B, et al. Polyspecific monoclonal lupus autoantibodiesreactive with both polynucleotides and phospholipids. J ExpMed. 1981;153:897-909.

20. Shoenfeld Y, Rauch J, Massicotte H, Datta SK, AndreSchwartz J, Stollar BD, et al. Polyspecificity of monoclonallupus autoantibodies produced by human-human hybridomas.N Engl J Med. 1983;308:414-20.

21. Carroll P, Stafford D, Schwartz RS, Stollar BD. Murinemonoclonal anti-DNA autoantibodies bind to endogenousbacteria. J Immunol. 1985;135:1086-90.

22. Ben-Yehuda A, Rasooly L, Bar-Tana R, Breuer G, Tadmor B,Ulmansky R, et al. The urine of SLE patients containsantibodies that bind to the laminin component of the extrac-ellular matrix. J Autoimmun. 1995;8:279-91.

23. Xie C, Liang Z, Chang S, Mohan C. Use of a novel elutionregimen reveals the dominance of polyreactive antinuclearautoantibodies in lupus kidneys. Arthritis Rheum. 2003;48:2343-52.

24. Amital H, Heilweil M, Ulmansky R, Szafer F, Bar-Tana R,Morel L, et al. Treatment with a laminin-derived peptidesuppresses lupus nephritis. J Immunol. 2005;175:5516-23.

25. Van Bruggen MC, Kramers C, Hylkema MN, Smeenk RJ,Berden JH. Significance of anti-nuclear and anti-extracellularmatrix autoantibodies for albuminuria in murine lupus neph-ritis; a longitudinal study on plasma and glomerular eluates inMRL/l mice. Clin Exp Immunol. 1996;105:132-9.

26. Krishnan MR, Wang C, Marion TN. Anti-DNA autoanti-bodies initiate experimental lupus nephritis by binding directlyto the glomerular basement membrane in mice. Kidney Int.2012;82:184-92.

27. Mostoslavsky G, Fischel R, Yachimovich N, Yarkoni Y,Rosenmann E, Monestier M, et al. Lupus anti-DNA autoanti-bodies cross-react with a glomerular structural protein: a casefor tissue injury by molecular mimicry. Eur J Immunol.2001;31:1221-7.

28. Yung S, Cheung KF, Zhang Q, Chan TM. Anti-dsDNAantibodies bind to mesangial annexin II in lupus nephritis.J Am Soc Nephrol. 2010;21:1912-27.

29. Sun KH, Hong CC, Tang SJ, Sun GH, Liu WT, Han SH, et al.Anti-dsDNA autoantibody cross-reacts with the C-terminalhydrophobic cluster region containing phenylalanines in theacidic ribosomal phosphoprotein P1 to exert a cytostatic effecton the cells. Biochem Biophys Res Commun. 1999;263:334-9.

30. Andre-Schwartz J, Datta SK, Shoenfeld Y, Isenberg DA,Stollar BD, Schwartz RS. Binding of cytoskeletal proteins bymonoclonal anti-DNA lupus autoantibodies. Clin ImmunolImmunopathol. 1984;31:261-71.

31. Widom J. A relationship between the helical twist of DNAand the ordered positioning of nucleosomes in all eukaryoticcells. Proc Natl Acad Sci U S A. 1992;89:1095-9.

32. Griffith J, Bleyman M, Rauch CA, Kitchin PA, Englund PT.Visualization of the bent helix in kinetoplast DNA by electronmicroscopy. Cell. 1986;46:717-24.

33. Rose NR, Bona C. Defining criteria for autoimmune diseases(Witebskyʼs postulates revisited). Immunol Today. 1993;14:426-30.

34. Fenton K, Fismen S, Hedberg A, Seredkina N, Fenton C,Mortensen ES, et al. Anti-dsDNA antibodies promote initia-tion, and acquired loss of renal Dnase1 promotes progressionof lupus nephritis in autoimmune (NZBxNZW)F1 mice. PLoSOne. 2009;4:e8474.

35. Seredkina S, Rekvig OP. Acquired loss of renal nucleaseactivity is restricted to DNaseI and is an organ-selectivefeature in murine lupus nephritis. Am J Pathol. 2011;179:1120-8.

36. Zykova SN, Tveita AA, Rekvig OP. Renal Dnase1 enzymeactivity and protein expression is selectively shut down inmurine and human membranoproliferative lupus nephritis.PLoS One. 2010;5:8.

37. Thiyagarajan D, Fismen S, Seredkina N, Jacobsen S, Elung-Jensen T, Kamper AL, et al. Silencing of renal DNaseI inmurine lupus nephritis imposes exposure of large chromatinfragments and activation of Toll like receptors and the Clec4e.PLoS One. 2012;7:e34080.

38. Hedberg A, Mortensen ES, Rekvig OP. Chromatin as a targetantigen in human and murine lupus nephritis. Arthritis ResTher. 2011;13:214.

39. Tan EM. Antinuclear antibodies: diagnostic markers and cluesto the basis of systemic autoimmunity. Pediatr Infect Dis J.1988;7(Suppl):S3-9.

40. Balow JE, Boumpas DT, Ausin HA. Systemic lupus eryth-ematosus and the kidney. In: Lahita RG editor, Systemic lupuserythematosus, 3rd ed. San Diego: Academic Press; 1999.p.657-85.

41. Shlomchik MJ, Madaio MP. The role of antibodies and B cellsin the pathogenesis of lupus nephritis. Springer SeminImmunopathol. 2003;24:363-75.

42. Van Bruggen MC, Kramers C, Walgreen B, Elema JD,Kallenberg CG, van den Born J, et al. Nucleosomes andhistones are present in glomerular deposits in human lupusnephritis. Nephrol Dial Transplant. 1997;12:57-66.

43. Berden JH, Licht R, Van Bruggen MC, Tax WJ. Role ofnucleosomes for induction and glomerular binding of autoanti-bodies in lupus nephritis. Curr Opin Nephrol Hypertens.1999;8:299-306.

44. Grootscholten C, Van Bruggen MC, van der Pijl JW, ElemaJD, Kallenberg CG, van den Born J, et al. Deposition ofnucleosomal antigens (histones and DNA) in the epidermalbasement membrane in human lupus nephritis. ArthritisRheum. 2003;48:1355-62.

45. Fismen S, Hedberg A, Fenton K, Jacobsen S, Krarup E,Kamper A, et al. Circulating chromatin-anti-chromatin anti-body complexes bind with high affinity to dermo-epidermalstructures in murine and human lupus nephritis. Lupus.2009;18:597-607.

46. Fismen S, Rekvig OP, Mortensen E. Pathogenesis of SLEdermatitis – a reflection of the process in SLE nephritis? CurrRheumatol Rev. 2007;3:1-7.

47. Diamond B, Volpe BT. A model for lupus brain disease.Immunol Rev. 2012;248:56-67.

48. Faust TW, Chang EH, Kowal C, Berlin R, Gazaryan IG,Bertini E, et al. Neurotoxic lupus autoantibodies alter brainfunction through two distinct mechanisms. Proc Natl Acad SciU S A. 2010;107:18569-74.

H.L. Pedersen et al.436

49. Huerta PT, Kowal C, DeGiorgio LA, Volpe BT, Diamond B.Immunity and behavior: antibodies alter emotion. Proc NatlAcad Sci U S A. 2006;103:678-83.

50. Appel GB, Cohen DJ, Pirani CL, Meltzer JI, Estes D. Long-term follow-up of patients with lupus nephritis. A study basedon the classification of the World Health Organization. Am JMed. 1987;83:877-85.

51. Lee SG, Cho YM, So MW, Kim SS, Kim YG, Lee CK, et al.ISN/RPS 2003 class II mesangial proliferative lupus nephritis:a comparison between cases that progressed to class III or IVand cases that did not. Rheumatol Int. 2012;32:2459-64.

52. Tam LS, Li EK, Lai FM, Chan YK, Szeto CC. Mesangiallupus nephritis in Chinese is associated with a high rate oftransformation to higher grade nephritis. Lupus. 2003;12:665-71.

53. Berden JH. Lupus nephritis. Kidney Int. 1997;52:538-58.54. Marion TN, Krishnan MR, Steeves MA, Desai DD. Affinity

maturation and autoimmunity to DNA. Curr Dir Autoimmun.2003;6:123-53.

55. Radic MZ, Weigert M. Genetic and structural evidence forantigen selection of anti-DNA antibodies. Annu Rev Immu-nol. 1994;12:487-520.

56. Shlomchik MJ, Marshak-Rothstein A, Wolfowicz CB, Roth-stein TL, Weigert MG. The role of clonal selection andsomatic mutation in autoimmunity. Nature. 1987;328:805-11.

57. Richmond TJ, Davey CA. The structure of DNA in thenucleosome core. Nature. 2003;423:145-50.

58. Raz E, Ben Bassat H, Davidi T, Shlomai Z, Eilat D. Cross-reactions of anti-DNA autoantibodies with cell surface pro-teins. Eur J Immunol. 1993;23:383-90.

59. DʼAndrea DM, Coupaye Gerard B, Kleyman TR, Foster MH,Madaio MP. Lupus autoantibodies interact directly withdistinct glomerular and vascular cell surface antigens. KidneyInt. 1996;49:1214-21.

60. Termaat RM, Assmann KJ, Dijkman HB, van Gompel F,Smeenk RJ, Berden JH. Anti-DNA antibodies can bind to theglomerulus via two distinct mechanisms. Kidney Int. 1992;42:1363-71.

61. Deocharan B, Qing X, Lichauco J, Putterman C. Alpha-actininis a cross-reactive renal target for pathogenic anti-DNAantibodies. J Immunol. 2002;168:3072-8.

62. DeGiorgio LA, Konstantinov KN, Lee SC, Hardin JA, VolpeBT, Diamond B. A subset of lupus anti-DNA antibodies cross-reacts with the NR2 glutamate receptor in systemic lupuserythematosus. Nat Med. 2001;7:1189-93.

63. van der Vlag J, Berden JH. Lupus nephritis: role of anti-nucleosome autoantibodies. Semin Nephrol. 2011;31:376-89.

64. Kalaaji M, Fenton KA, Mortensen ES, Olsen R, Sturfelt G,Alm P, et al. Glomerular apoptotic nucleosomes are centraltarget structures for nephritogenic antibodies in human SLEnephritis. Kidney Int. 2007;71:664-72.

65. Mjelle JE, Kalaaji M, Rekvig OP. Exposure of chromatin andnot high affinity for dsDNA determines the nephritogenicimpact of anti-dsDNA antibodies in (NZBxNZW)F1 mice.Autoimmunity. 2009;42:104-11.

66. Comerford FR, Cohen AS, Desai RG. The evolution of theglomerular lesion in NZB mice. A light and electron micro-scopic study. Lab Invest. 1968;19:643-51.

67. Dillard MG, Tillman RL, Sampson CC. Lupus nephritis.Correlations between the clinical course and presence ofelectron-dense deposits. Lab Invest. 1975;32:261-9.

68. Ben-Bassat M, Rosenfeld J, Joshua H, Hazaz B, Gura V.Lupus nephritis. Electron-dense and immunofluorescentdeposits and their correlation with proteinuria and renalfunction. Am J Clin Pathol. 1979;72:186-93.

69. Borza DB, Hudson BG. Molecular characterization of thetarget antigens of anti-glomerular basement membrane anti-body disease. Springer Semin Immunopathol. 2003;24:345-61.

70. Borza DB, Neilson EG, Hudson BG. Pathogenesis of Good-pasture syndrome: a molecular perspective. Semin Nephrol.2003;23:522-31.

71. Winfield JB, Faiferman I, Koffler D. Avidity of anti-DNAantibodies in serum and IgG glomerular eluates from patientswith systemic lupus erythematosus. Association of highavidity antinative DNA antibody with glomerulonephritis.J Clin Invest. 1977;59:90-6.

72. Berden JH, Grootscholten C, Jurgen WC, van der Vlag J.Lupus nephritis: a nucleosome waste disposal defect?J Nephrol. 2002;15(Suppl 6):S1-10.

73. Van Bruggen MC, Kramers C, Berden JH. Autoimmunityagainst nucleosomes and lupus nephritis. Ann Med InterneParis. 1996;147:485-9.

74. Van Bruggen MC, Walgreen B, Rijke TP, Tamboer W,Kramers K, Smeenk RJ, et al. Antigen specificity of anti-nuclear antibodies complexed to nucleosomes determinesglomerular basement membrane binding in vivo. Eur JImmunol. 1997;27:1564-9.

75. van Bavel CC, van der Vlag J, Berden JH. Glomerular bindingof anti-dsDNA autoantibodies: the dispute resolved? KidneyInt. 2007;71:600-1.

76. Kalaaji M, Sturfelt G, Mjelle JE, Nossent H, Rekvig OP.Critical comparative analyses of anti-alpha-actinin andglomerulus-bound antibodies in human and murine lupusnephritis. Arthritis Rheum. 2006;54:914-26.

77. Seredkina N, Zykova SN, Rekvig OP. Progression of murinelupus nephritis is linked to acquired renal Dnase1 deficiencyand not to up-regulated apoptosis. Am J Pathol. 2009;175:97-106.

78. Tveita AA, Rekvig OP, Zykova SN. Increased glomerularmatrix metalloproteinase activity in murine lupus nephritis.Kidney Int. 2008;74:1150-8.

79. Tveita A, Rekvig OP, Zykova SN. Glomerular matrix metal-loproteinases and their regulators in the pathogenesis of lupusnephritis. Arthritis Res Ther. 2008;10:229.

80. Fenton KA, Tommeras B, Marion TN, Rekvig OP. Pure anti-dsDNA mAbs need chromatin structures to promote glomer-ular mesangial deposits in BALB/c mice. Autoimmunity.2010;43:179-88.

81. Frisoni L, McPhie L, Kang SA, Monestier M, Madaio M,Satoh M, et al. Lack of chromatin and nuclear fragmentationin vivo impairs the production of lupus anti-nuclear anti-bodies. J Immunol. 2007;179:7959-66.

82. Kawane K, Nagata S. Nucleases in programmed cell death.Methods Enzymol. 2008;442:271-87.

83. Kawane K, Fukuyama H, Yoshida H, Nagase H, Ohsawa Y,Uchiyama Y, et al. Impaired thymic development in mouseembryos deficient in apoptotic DNA degradation. Nat Immu-nol. 2003;4:138-44.

84. Kawane K, Ohtani M, Miwa K, Kizawa T, Kanbara Y,Yoshioka Y, et al. Chronic polyarthritis caused by mammalianDNA that escapes from degradation in macrophages. Nature.2006;443:998-1002.

85. Napirei M, Karsunky H, Zevnik B, Stephan H, Mannherz HG,Moroy T. Features of systemic lupus erythematosus inDnase1-deficient mice. Nat Genet. 2000;25:177-81.

86. Krieg AM, Vollmer J. Toll-like receptors 7, 8, and 9: linkinginnate immunity to autoimmunity. Immunol Rev. 2007;220:251-69.

87. Christensen SR, Shupe J, Nickerson K, Kashgarian M, FlavellRA, Shlomchik MJ. Toll-like receptor 7 and TLR9 dictateautoantibody specificity and have opposing inflammatory and

Murine and human lupus nephritis 437

regulatory roles in a murine model of lupus. Immunity.2006;25:417-28.

88. Christensen SR, Shlomchik MJ. Regulation of lupus-relatedautoantibody production and clinical disease by Toll-likereceptors. Semin Immunol. 2007;19:11-23.

89. Lim EJ, Lee SH, Lee JG, Chin BR, Bae YS, Kim JR, et al.Toll-like receptor 9 dependent activation of MAPK and NF-kBis required for the CpG ODN-induced matrix metalloproteinase-9 expression. Exp Mol Med. 2007;39:239-45.

90. Merrell MA, Ilvesaro JM, Lehtonen N, Sorsa T, Gehrs B,Rosenthal E, et al. Toll-like receptor 9 agonists promotecellular invasion by increasing matrix metalloproteinaseactivity. Mol Cancer Res. 2006;4:437-47.

91. Weening JJ, Dʼagati VD, Schwartz MM, Seshan SV, AlpersCE, Appel GB, et al. The classification of glomerulonephritisin systemic lupus erythematosus revisited. J Am Soc Nephrol.2004;15:241-50.

92. Markowitz GS, Dʼagati VD. Classification of lupus nephritis.Curr Opin Nephrol Hypertens. 2009;18:220-5.

93. Zykova SN, Seredkina N, Benjaminsen J, Rekvig OP.Reduced fragmentation of apoptotic chromatin is associatedwith nephritis in lupus-prone (NZB x NZW)F(1) mice.Arthritis Rheum. 2008;58:813-25.

94. Fismen S, Thiyagarajan D, Seredkina N, Nielsen H, Jacobsen S,Elung-Jensen T, et al. Impact of the tumor necrosis factorreceptor-associated protein 1 (Trap1) on renal DNaseI shut-down and on progression of murine and human lupus nephritis.Am J Pathol. 2013;182:688-700.

95. Frost PG, Lachmann PJ. The relationship of desoxyribonu-clease inhibitor levels in human sera to the occurrence ofantinuclear antibodies. Clin Exp Immunol. 1968;3:447-55.

96. Yasutomo K, Horiuchi T, Kagami S, Tsukamoto H, Hashimura C,Urushihara M, et al. Mutation of DNase1 in people with systemiclupus erythematosus. Nat Genet. 2001;28:313-4.

97. Tsukumo S, Yasutomo K. DNaseI in pathogenesis of systemiclupus erythematosus. Clin Immunol. 2004;113:14-8.

98. Samejima K, Earnshaw WC. Trashing the genome: the role ofnucleases during apoptosis. Nat Rev Mol Cell Biol. 2005;6:677-88.

99. Oliveri M, Daga A, Cantoni C, Lunardi C, Millo R, Puccetti A.DNase I mediates internucleosomal DNA degradation in humancells undergoing drug-induced apoptosis. Eur J Immunol.2001;31:743-51.

100. Napirei M, Gultekin A, Kloeckl T, Moroy T, Frostegard J,Mannherz HG. Systemic lupus-erythematosus: deoxyribonu-clease 1 in necrotic chromatin disposal. Int J Biochem CellBiol. 2006;38:297-306.

101. Ludwig S, Mannherz HG, Schmitt S, Schaffer M, Zentgraf H,Napirei M. Murine serum deoxyribonuclease 1 (Dnase1)activity partly originates from the liver. Int J Biochem CellBiol. 2009;41:1079-93.

102. Napirei M, Ludwig S, Mezrhab J, Klockl T, Mannherz HG.Murine serum nucleases–contrasting effects of plasmin andheparin on the activities of DNase1 and DNase1-like 3(DNase1l3). Febs J. 2009;276:1059-73.

103. Nagata S, Kawane K. Autoinflammation by endogenousDNA. Adv Immunol. 2011;110:139-61.

104. Macanovic M, Sinicropi D, Shak S, Baughman S, Thiru S,Lachmann PJ. The treatment of systemic lupus erythematosus(SLE) in NZB/W F1 hybrid mice; studies with recombinantmurine DNase and with dexamethasone. Clin Exp Immunol.1996;106:243-52.

105. Macanovic M, Lachmann PJ. Measurement of deoxyribonu-clease I (DNase) in the serum and urine of systemic lupuserythematosus (SLE)-prone NZB/NZW mice by a new radialenzyme diffusion assay. Clin Exp Immunol. 1997;108:220-6.

106. Verthelyi D, Dybdal N, Elias KA, Klinman DM. DNAsetreatment does not improve the survival of lupus prone (NZBx NZW)F1 mice. Lupus. 1998;7:223-30.

107. Davis JC Jr, Manzi S, Yarboro C, Rairie J, Mcinnes I,Averthelyi D, et al. Recombinant human Dnase I (rhDNase)in patients with lupus nephritis. Lupus. 1999;8:68-76.

108. Mjelle JE, Rekvig OP, Fenton KA. Nucleosomes possess ahigh affinity for glomerular laminin and collagen IV and bindnephritogenic antibodies in murine lupus-like nephritis. AnnRheum Dis. 2007;66:1661-8.

109. Brenowitz M, Senear DF, Shea MA, Ackers GK. QuantitativeDNase footprint titration: a method for studying protein-DNAinteractions. Methods Enzymol. 1986;130:132-81.

110. Hsieh M, Brenowitz M. Quantitative kinetics footprinting ofprotein-DNA association reactions. Methods Enzymol.1996;274:478-92.

111. Shak S, Capon DJ, Hellmiss R, Marsters SA, Baker CL.Recombinant human DNase I reduces the viscosity of cysticfibrosis sputum. Proc Natl Acad Sci U S A. 1990;87:9188-92.

112. Mogayzel PJ Jr, Naureckas ET, Robinson KA, Mueller G,Hadjiliadis D, Hoag JB, et al. Cystic fibrosis pulmonaryguidelines. Chronic medications for maintenance of lunghealth. Am J Respir Crit Care Med. 2013;187:680-9.

113. McCarty M. Purification and properties of desoxyribonu-clease isolated from beef pancreas. J Gen Physiol. 1946;29:123-39.

114. Mazurkiewicz J, Kepert JF, Rippe K. On the mechanism ofnucleosome assembly by histone chaperone NAP1. J BiolChem. 2006;281:16462-72.

115. Kepert JF, Mazurkiewicz J, Heuvelman GL, Toth KF, Rippe K.NAP1 modulates binding of linker histone H1 to chromatin andinduces an extended chromatin fiber conformation. J BiolChem. 2005;280:34063-72.

116. Banuelos S, Omaetxebarria MJ, Ramos I, Larsen MR, Arregi I,Jensen ON, et al. Phosphorylation of both nucleoplasmindomains is required for activation of its chromatin decondensa-tion activity. J Biol Chem. 2007;282:21213-21.

117. Laskey RA, Mills AD, Philpott A, Leno GH, Dilworth SM,Dingwall C. The role of nucleoplasmin in chromatin assemblyand disassembly. Philos Trans R Soc Lond B Biol Sci.1993;339:263-9.

118. Zhao R, Davey M, Hsu YC, Kaplanek P, Tong A, ParsonsAB, et al. Navigating the chaperone network: an integrativemap of physical and genetic interactions mediated by thehsp90 chaperone. Cell. 2005;120:715-27.

119. Zhao R, Houry WA. Hsp90: a chaperone for proteinfolding and gene regulation. Biochem Cell Biol. 2005;83:703-10.

120. Rekvig OP, Hannestad K. The specificity of human autoanti-bodies that react with both cell nuclei and plasma membranes:the nuclear antigen is present on core mononucleosomes.J Immunol. 1979;123:2673-81.

121. Villeponteau B. Heparin increases chromatin accessibility bybinding the trypsin-sensitive basic residues in histones. Bio-chem J. 1992;288:953-8.

122. Hedberg A, Fismen S, Fenton KA, Fenton C, Osterud B,Mortensen ES, et al. Heparin exerts a dual effect on murinelupus nephritis by enhancing enzymatic chromatin degrada-tion and preventing chromatin binding in glomerular mem-branes. Arthritis Rheum. 2011;63:1065-75.

123. Polikanov YS, Rubtsov MA, Studitsky VM. Biochemicalanalysis of enhancer-promoter communication in chromatin.Methods. 2007;41:250-8.

124. Conde E Silva, Black BE, Sivolob A, Filipski J, Cleveland DW,Prunell A. CENP-A-containing nucleosomes: easier disassem-bly versus exclusive centromeric localization. J Mol Biol.2007;370:555-73.

H.L. Pedersen et al.438

125. Simeonova M, Tchacarov E. Prefixation chromosome bandingwith heparin. Hum Genet. 1980;56:63-6.

126. Van Bruggen MC, Walgreen B, Rijke TP, Corsius MJ,Assmann KJ, Smeenk RJ, et al. Heparin and heparinoidsprevent the binding of immune complexes containing nucle-osomal antigens to the GBM and delay nephritis in MRL/lprmice. Kidney Int. 1996;50:1555-64.

127. Naparstek Y, Ben-Yehuda A, Madaio MP, Bar-Tana R,Schuger L, Pizov G, et al. Binding of anti-DNA antibodiesand inhibition of glomerulonephritis in MRL-lpr/lpr mice byheparin. Arthritis Rheum. 1990;33:1554-9.

128. Van Bruggen MC, Kramers C, Hylkema MN, Smeenk RJ,Berden JH. Pathophysiology of lupus nephritis: the role ofnucleosomes. Neth J Med. 1994;45:273-9.

129. Kramers C, Hylkema MN, Van Bruggen MC, van deLagemaat R, Dijkman HB, Assmann KJ, et al. Anti-nucleosome antibodies complexed to nucleosomal antigensshow anti-DNA reactivity and bind to rat glomerular basementmembrane in vivo. J Clin Invest. 1994;94:568-77.

130. Berden JH. Lupus nephritis: consequence of disturbedremoval of apoptotic cells? Neth J Med. 2003;61:233-8.

131. Nowling TK, Gilkeson GS. Mechanisms of tissue injury inlupus nephritis. Arthritis Res Ther. 2011;13:250.