Immunology Letters 91 (2004) 93–97

Short communication

Susceptibility of mast cell-deficient W/Wv mice to pristane-inducedexperimental lupus nephritis

Ling Lin a, Andrea J. Gertha, Stanford L. Penga,b,∗a Department of Internal Medicine/Rheumatology, Washington University School of Medicine, Campus Box 8045,

CSRB 6617, 660 S. Euclid Ave., St. Louis, MO 63110, USAb Department of Pathology and Immunology, Washington University School of Medicine, St. Louis, MO 63110, USA

Received 12 November 2003; received in revised form 23 November 2003; accepted 23 November 2003

Abstract

In many models of organ-specific autoimmune diseases, mast cells provide a critical cellular link between autoantibodies and end-organinflammation, both initiating and propagating disease. However, their role in systemic autoimmunity remains speculative. We thereforeexamined the role of mast cells in a murine model of systemic immune complex-related autoimmune disease, lupus nephritis, expectingto observe the development of humoral autoimmunity in the absence of end-organ disease. Surprisingly, not only did mast cell-deficientanimals develop characteristic humoral features of lupus, including hypergammaglobulinemia and autoantibodies, they also developed immunecomplex glomerulonephritis, as evidenced by renal immune deposits, glomerular disease, and proteinuria. These findings implicate the presenceof distinct effector pathways to end-organ damage in humoral autoimmune diseases: one involving the interaction between autoantibodies andmast cells to recruit inflammation in organ-specific autoimmunity, and another involving a more direct—mast cell-independent—interactionbetween autoantibodies and circulating inflammatory mediators in systemic autoimmunity.© 2003 Elsevier B.V. All rights reserved.

Keywords: Lupus; Mast cells; Glomerulonephritis

Although the pathogenesis of systemic lupus erythemato-sus remains elusive, substantial evidence indicates a centralrole for autoantibodies in the induction of end-organ disease,particularly, glomerulonephritis. Here, anti-dsDNA antibod-ies in both animals and humans may form pathogenic im-mune complexes, directly binding to glomerular antigens,and/or penetrating cells and causing cellular toxicity. Manyrecent studies have begun to identify key cytokines andchemokines that may participate in the effector mechanismsof lupus kidney disease. However the cellular link, if any,between pathogenic autoantibodies and end-organ inflam-mation remains unclearly defined[1].

Recent work has strongly implicated mast cells as po-tential initiators of autoimmune damage. For instance,mast cell-deficient mice are protected from experimentalautoimmune encephalomyelitis[2], experimental bullouspemphigoid[3], and the Arthus reaction[4]. Even more

∗ Corresponding author. Tel.:+1-314-747-3609;fax: +1-314-454-1091.

E-mail address: speng@im.wustl.edu (S.L. Peng).

impressively, mast cells have been found to play criticalroles in the effector arm of autoantibody-mediated induc-tion of vasculitis and arthritis[5,6]. Such findings suggestthat mast cells could provide the cellular link between au-toantibodies and end-organ disease in lupus. Indeed, mastcells can be found in both murine[7] and human[8,9] lupusrenal lesions, where they are speculated to participate in thegeneration and recruitment of inflammatory mediators ofnephritis. We therefore examined the ability of lupus-relatedantibodies to elicit renal disease in mast cell-deficient mice.

Female mast cell-deficient WBB6F1/J-KitW /KitW-v

(W/Wv) and control congenic WBB6F1/J (WT) micewere purchased from the Jackson Laboratory (Bar Harbor,ME). To induce autoantibodies and glomerulonephritis,0.5 ml of pristane (ICN Biomedicals Inc., Irvine, CA)or phosphate-buffered saline (PBS) was administered in-traperitoneally to 8-weeks-old animals[10]. Animals weremaintained in a specific pathogen-free environment at theWashington University School of Medicine. Serum levelsof IgM, IgG1, IgG2b, IgG3, IgA, and IgE were deter-mined by isotype-specific ELISA (Southern Biotechnology

0165-2478/$ – see front matter © 2003 Elsevier B.V. All rights reserved.doi:10.1016/j.imlet.2003.11.014

94 L. Lin et al. / Immunology Letters 91 (2004) 93–97

Fig. 1. Lupus-related hypergammaglobulinemia in mast cell-deficient mice. Shown are serum titers of specific Ig isotypes in mast cell-deficient (W/Wv)or congenic control (WT) mice treated 12 weeks previously with PBS (n = 3 in each group) or pristane (n = 12 in each group).

Associates, Birmingham, AL). IgG2aa was not significantlydetectable in the serum of WBB6F1 animals (our unpub-lished data), so the IgG2a referred to here reflects IgG2ab

(IgG2c) titers as determined by isotype-specific ELISA(BD Pharmingen, San Diego, CA). Antinuclear antibodyimmunofluorescence was determined using HEp-2 cell sub-strates (INOVA Diagnostics, Inc., San Diego, CA), followedby FITC-conjugated anti-mouse IgG (Pierce Biotechnol-ogy, Inc., Rockford, IL), and subsequent visualization on aNikon Eclipse fluorescent microscope (Nikon InstrumentsInc., Melville, NY). Fluorescence intensity was rated as0–4+, as determined by the estimated required exposuretime for 10–12 cells per high-powered field, using the ac-quisition software (Magnafire®; Optronics, Goleta, CA:>2 s, 0; 1–2 s, 1+; 0.5–1 s, 2+; 0.25–0.5 s, 3+; <0.25 s,4+). IgG anti-dsDNA antibodies were assayed by ELISAusing calf thymus DNA (Sigma, St. Louis, MO), and/orby Crithidia luciliae indirect immunofluorescence (Anti-bodies Incorporated, Davis, CA) on sera at a 1:10 dilution.

Hematoxylin and eosin staining of formalin-fixed tissuesections, immunofluorescent studies on OCT-embeddedfrozen sections, and assays for serum antibodies and au-toantibodies were performed essentially as described[11].Severity of nephritis was evaluated on a 0–4+ scale by lightmicroscopy in glomeruli. Ig deposition, as determined byFITC-conjugated goat anti-mouse IgG (Pierce), was scoredby immunofluorescence microscopy on a 0–4+ scale, simi-lar to that for antinuclear antibodies (above). Sections werephotographed using identical exposure times. Proteinuriawas assessed by determining excreted protein concentrationby the Bradford method (Bio-Rad, Hercules, CA).

WBB6F1/J-KitW /KitW-v (W/Wv) and congenic controlWBB6F1/J (WT) animals were assessed 12 weeks after pris-tane administration, when evidence of both humoral au-toimmunity and immune-complex glomerulonephritis is typ-ically evident [10]. WT animals developed hypergamma-globulinemia, particularly, of IgM, IgG1, IgG2a and IgG3isotypes, specifically in response to pristane, but not PBS

L. Lin et al. / Immunology Letters 91 (2004) 93–97 95

Fig. 2. Antinuclear antibody development in mast cell-deficient mice. Sera from pristane-treated W/Wv or WT animals (n = 12 each) were assessed forthe presence of antinuclear antibodies by indirect immunofluorescence and anti-DNA antibodies by ELISA. Example staining patterns are shown on theright: combined specked and nucleolar pattern by a WT serum (3+, top), and a nucleolar pattern by a W/Wv serum (2+, bottom). Both panels are at40× magnification. PBS-treated WT and W/Wv animals were uniformly nonreactive in this assay (not shown).

(Fig. 1). In addition, they developed typical lupus-related au-toantibodies, including antinuclear antibodies as assessed byimmunofluorescence and anti-DNA antibodies as assessedby ELISA (Fig. 2). Interestingly, these latter specificitieslikely reflect anti-ssDNA activity, since all sera failed totest positive byCrithidia immunofluorescence (not shown),which specifically assesses anti-dsDNA activity; however,this observation agrees with previous findings in this nephri-tis model[10,12].

Strikingly, mast cell-deficient W/Wv animals developedhypergammaglobulinemia and autoantibodies comparableto, if not exceeding, their WT counterparts (Fig. 1). For in-stance, most pristane-exposed WT animals failed to developsignificant hypergammaglobulinemia of IgG2b, whereassome pristane-exposed W/Wv animals developed titers ashigh as 5–6 mg/ml. In addition, mast cell-deficient W/Wvanimals developed significant antinuclear antibodies withactivity comparable to WT counterparts, including the de-velopment ofCrithidia-negative, ELISA-positive anti-DNAantibodies (Fig. 2 and data not shown). Together, thesefindings clearly indicate that mast cells are dispensable inthe development of humoral autoimmunity.

At the same time, though, the development of humoralautoimmunity in the absence of mast cells is not surprisinggiven their prior implications in the effector, rather than af-ferent, pathogenesis of autoantibody-mediated end-organ in-flammation[6]. Given the importance of Fc� receptors in theability of mast cells to mediate inflammatory effects of anti-bodies, we speculated that mast cell deficiency might uncou-ple humoral and end-organ autoimmunity, similar to findings

in Fc� receptor-deficient lupus-prone mice[13]. However,mast cell-deficient W/Wv animals developed renal diseasecomparable to their WT counterparts, including glomeru-lonephritis, immune deposits, and proteinuria (Table 1andFig. 3). In fact, a higher proportion of W/Wv compared toWT kidneys examined contained evidence of diffusely pro-liferative glomerulonephritis, albeit with comparable levelsof proteinuria (4 of 5 versus 1 of 5,Table 1), suggestingthat perhaps at least some aspects of the renal disease were

Table 1Renal disease in pristane-treated mast cell-deficient mice

Genotype Treatment Glomerularlesion

Magnitude Immunedeposits

Proteinuria(mg/dl)

WT PBS none – – <3WT PBS none – – <3WT PBS None – – <3WT Pristane DPGNa 3+ 2+ 37WT Pristane FSGNb 3+ 2+ 34WT Pristane FSGN 3+ 2+ 31WT Pristane FSGN 1+ 1+ 11WT Pristane FSGN 2+ 2+ 19W/Wv PBS None – – <3W/Wv PBS None – – <3W/Wv PBS None – – <3W/Wv Pristane DPGN 2+ 3+ 15W/Wv Pristane DPGN 2+ 3+ 20W/Wv Pristane DPGN 2+ 3+ 22W/Wv Pristane DPGN 3+ 3+ 35W/Wv Pristane FSGN 1+ 2+ 12

a DPGN, diffuse proliferative glomerulonephritis.b FSGN, focal segmental glomerulonephritis.

96 L. Lin et al. / Immunology Letters 91 (2004) 93–97

Fig. 3. Glomerulonephritis in mast cell-deficient mice. (A) Normal glomerular appearance in a PBS-treated W/Wv animal; (B) mild focal segmentalglomerulonephritis including mesangial proliferation in a pristane-treated WT animal (1+); (C) moderate diffuse glomerulonephritis including mesangialhypercellularity and proliferation in a pristane-treated W/Wv animal (3+). Glomerular immune deposits were assessed by indirect immunofluorescencefor IgG in (D) WT (1+) or (E) W/Wv (3+) pristane-treated animals. All panels are at 40× magnification.

more severe in the setting of mast cell deficiency. Regard-less of such potential quantitative differences, though, wewere nonetheless left with the somewhat surprising obser-vation that mast cells are not required for the effector arm ofautoantibody-mediated glomerulonephritis in systemic au-toimmunity.

As such, autoantibodies likely utilize distinct mecha-nisms to evoke end-organ inflammation, depending uponwhether the target is organ-specific or organ-nonspecific.Notably, in settings in which mast cells have been demon-strated to play a role, autoantibodies elicit inflammation insingular, distinct anatomical sites, typically requiring theextravascular recruitment of cellular inflammatory media-tors, such as to the joint[6] or skin [3]. In contrast, lupusnephritis appears to reflect a more direct antibody-mediated

toxicity to the glomerular unit, in which antibodies and ortheir immune complexes may bind directly to glomerularantigens, and recruit inflammatory mediators, such as com-plement, that do not require extravasation for pathogeniceffects[14]. In addition, lupus autoantibodies may requireother cellular components of the kidney, such as mesangialor other parenchymal cells that express Fc receptors, whichare directly accessible to blood-borne autoantibodies—incontrast to synovial or dermal tissue—to mediate inflamma-tion via the subsequent expression of pathogenic cytokinesand chemokines like IFN-�, members of the IL-12 family,and MCP-1[1,13]. Thus, mast cells may be specificallyrequired in autoantibody-mediated reactions that requireextravascular inflammation, whereas lupus nephritis is mastcell-independent because it primarily reflects a unique

L. Lin et al. / Immunology Letters 91 (2004) 93–97 97

situation in which the fluid phase of autoantibodies andother mediators of inflammation directly interfaces with thetarget organ. As such, it will clearly be of future interestto examine the role of mast cells in spontaneous animalmodels of lupus, such as the NZB/W or MRL strains; how-ever, because both the c-kit and c-kit ligand mutations thatresult in mast cell deficiency reside near known murinelupus susceptibility loci, and do not appear to confer mastcell deficiency on lupus genetic backgrounds ([15], andour unpublished data), alternative cellular and/or geneticapproaches will be required.

Acknowledgements

We thank M. Robert Marshall for assistance withhistopathology, as well as David M. Lee and Alec M. Chengfor helpful comments. This work was supported in partby grants from the NIH (AI01803 and AI057471) and theLupus Research Institute. S.L.P. is supported in part by anArthritis Investigator Award from the Arthritis Foundation.

References

[1] Shlomchik MJ, Craft JE, Mamula MJ. Nat Rev Immunol 2001;1:147.[2] Secor VH, Secor WE, Gutekunst CA, Brown MA. J Exp Med

2000;191:813.[3] Chen R, Ning G, Zhao ML, Fleming MG, Diaz LA, Werb Z, Liu

Z. J Clin Invest 2001;108:1151.[4] Sylvestre DL, Ravetch JV. Immunity 1996;5:387.[5] Johnston B, Burns AR, Kubes P. Am J Pathol 1998;152:555.[6] Lee DM, Friend DS, Gurish MF, Benoist C, Mathis D, Brenner MB.

Science 2002;297:1689.[7] Ohmori J. Autoimmunity 1994;17:279.[8] Hiromura K, Kurosawa M, Yano S, Naruse T. Am J Kidney Dis

1998;32:593.[9] Kondo S, Kagami S, Kido H, Strutz F, Muller GA, Kuroda Y. J Am

Soc Nephrol 2001;12:1668.[10] Satoh M, Reeves WH. J Exp Med 1994;180:2341.[11] Peng SL, Madaio MP, Hughes DP, Crispe IN, Owen MJ, Wen L,

Hayday AC, Craft J. J Immunol 1996;156:4041.[12] Satoh M, Kumar A, Kanwar YS, Reeves WH. Proc Natl Acad Sci

USA 1995;92:10934.[13] Clynes R, Dumitru C, Ravetch JV. Science 1998;279:1052.[14] Grande JP. Lupus 1998;7:604.[15] Kelly JA, Moser KL, Harley JB. Genes Immun 2002;3(Suppl 1):S71.