Increased Susceptibility of Decay-Accelerating Factor Deficient Mice to Anti-Glomerular Basement Membrane Glomerulonephritis This information is current as of September 28, 2021. Hajime Sogabe, Masaomi Nangaku, Yoshitaka Ishibashi, Takehiko Wada, Toshiro Fujita, Xiujun Sun, Takashi Miwa, Michael P. Madaio and Wen-Chao Song J Immunol 2001; 167:2791-2797; ;

doi: 10.4049/jimmunol.167.5.2791 Downloaded from http://www.jimmunol.org/content/167/5/2791

References This article cites 54 articles, 23 of which you can access for free at: http://www.jimmunol.org/content/167/5/2791.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists by guest on September 28, 2021 • Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2001 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Increased Susceptibility of Decay-Accelerating Factor Deficient Mice to Anti-Glomerular Basement Membrane Glomerulonephritis1

Hajime Sogabe,* Masaomi Nangaku,2* Yoshitaka Ishibashi,* Takehiko Wada,* Toshiro Fujita,* Xiujun Sun,† Takashi Miwa,† Michael P. Madaio,‡ and Wen-Chao Song2†

To prevent complement-mediated autologous tissue damage, host cells express a number of membrane-bound complement in- hibitors. Decay-accelerating factor (DAF, CD55) is a GPI-linked membrane complement regulator that is widely expressed in mammalian tissues including the kidney. DAF inhibits the C3 convertase of both the classical and alternative pathways. Although DAF deficiency contributes to the human hematological syndrome paroxysmal nocturnal hemoglobinuria, the relevance of DAF Downloaded from in autoimmune tissue damage such as immune glomerulonephritis remains to be determined. In this study, we have investigated the susceptibility of knockout mice that are deficient in GPI-anchored DAF to nephrotoxic serum nephritis. Injection of a sub- nephritogenic dose of rabbit anti-mouse glomerular basement membrane serum induced glomerular disease in DAF knockout mice but not in wild-type controls. When examined at 8 days after anti-glomerular basement membrane treatment, DAF knockout -mice had a much higher percentage of diseased glomeruli than wild-type mice (68.8 ؎ 25.0 vs 10.0 ؎ 3.5%; p < 0.01). Morpho ؋ 3 ␮ 3 ؎ ؎

logically, DAF knockout mice displayed increased glomerular volume (516 68 vs 325 18 10 m per glomerulus; p < http://www.jimmunol.org/ and cellularity (47.1 ؎ 8.9 vs 32.0 ؎ 3.1 cells per glomerulus; p < 0.01). Although the blood urea nitrogen level showed (0.0001 no difference between the two groups, proteinuria was observed in the knockout mice but not in the wild-type mice (1.4 ؎ 0.7 vs mg/24 h albumin excretion). The morphological and functional abnormalities in the knockout mouse kidney were 0.01 ؎ 0.02 associated with evidence of increased complement activation in the glomeruli. These results support the conclusion that membrane C3 convertase inhibitors like DAF play a protective role in complement-mediated immune glomerular damage in vivo. The Journal of Immunology, 2001, 167: 2791Ð2797.

omplement is a form of natural immunity that plays an (MCP), and 1 (CR1) (3Ð5). DAF prevents important role in host defense (1). However, if not prop- the formation and accelerates the decay of C3 convertases, by guest on September 28, 2021 C erly controlled, activated complement can also cause by- whereas MCP and CR1 serve as cofactors for factor I-mediated stander injury to host tissues (1, 2). To prevent complement-me- cleavage of (3Ð6). CR1 also accelerates the decay of C3 con- diated autologous attack, host tissues express a number of fluid vertases as well as serving as an receptor (3, 4). phase and membrane-bound inhibitors (3Ð5). These inhibitors In rodents, a transmembrane known as Crry, which pos- work at different steps of the complement activation cascade, and sesses both DAF and MCP activities. has also been identified (7Ð collectively they ensure that inappropriate complement activation 9). In addition to regulation at the C3 cleavage step, autologous does not occur within normal tissues. Some of the membrane- complement damage can also be restricted at the terminal step by bound complement inhibitors act by inactivating C3/C5 converta- the GPI-linked membrane protein CD59 (4, 10, 11). ses (3Ð5). In humans, membrane C3 convertase inhibitors include The identification and study of human DAF have historically decay-accelerating factor (DAF),3 membrane cofactor protein been associated with the human hematological disorder paroxys- mal nocturnal hemoglobinuria (PNH) (12Ð14). PNH is caused by *Division of Nephrology and Endocrinology, University of Tokyo School of Medi- a combined deficiency of DAF and CD59 on the affected † cine, Tokyo, Japan; and Center for Experimental Therapeutics and Department cells of the patients (12). As a result of DAF and CD59 deficiency, of Pharmacology, and ‡Renal-Electrolyte and Hypertension Division, University of Pennsylvania School of Medicine, Philadelphia, PA 19104 blood cells of PNH patients are not protected from autologous Received for publication February 21, 2001. Accepted for publication June 25, 2001. complement attack. In addition to circulating blood cells, DAF is The costs of publication of this article were defrayed in part by the payment of page also expressed prominently on many other cell types such as en- charges. This article must therefore be hereby marked advertisement in accordance dothelial and epithelial cells (5, 15). For example, in the human with 18 U.S.C. Section 1734 solely to indicate this fact. kidney DAF has been detected in the glomerulus on mesangial and 1 This work was supported by National Institutes of Health Grants AI44970 (to W.C.-S.) and 53088 (to M.P.M.) and Grant in Aid for Scientific Research 11671030 epithelial cells (16, 17). from the Ministry of Education, Science and Culture (to M.N.). In principle, DAF should also protect these cells from comple- 2 Address correspondence and reprint requests to Dr. Wen-Chao Song, Center for ment-mediated inflammatory damage. This may be particularly Experimental Therapeutics, Department of Pharmacology, University of Pennsylva- true in an autoimmune disease setting in which either binding of nia School of Medicine,1351 BRBII/III, 421 Curie Boulevard, Philadelphia, PA 19104; or Dr. Masaomi Nangaku, Division of Nephrology and Endocrinology, Uni- autoantibodies to specific tissue Ags or formation of immune de- versity of Tokyo School of Medicine, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8655, posits in vital organs, such as the kidney, activates the classical Japan. complement pathway. Although in vitro studies have demonstrated 3 Abbreviations used in this paper: DAF, decay-accelerating factor; MCP, membrane cofactor protein; CR1, ; PNH, paroxysmal nocturnal hemo- that DAF expressed on these cells is functional as a C3 convertase globinuria; GBM, glomerular basement membrane; BUN, blood urea nitrogen. inhibitor (17, 18), thus far little direct evidence is available to

Copyright © 2001 by The American Association of Immunologists 0022-1767/01/$02.00 2792 DAF AND ANTI-GBM GLOMERULONEPHRITIS corroborate the expectation that DAF plays a protective role in Materials and Methods vivo on nonvascular cells from complement-mediated injury. In Animals humans, rare cases of complete DAF deficiency due to germline GPI-DAF-deficient mice were generated as described previously (23). In mutation of the DAF (Inab serological ) have been brief, the first three of the GPI-DAF gene were deleted and replaced identified (19, 20). These individuals differ from PNH patients in with the NEO gene. As expected, no GPI-DAF was expressed in the knock- that they lack DAF expression in all their tissues, whereas DAF out mouse tissues because the proximal sequence necessary for RNA and the first two short consensus repeats (encoded by deficiency in PNH patients, resulting from somatic mutations of a exons 2 and 3, respectively) were absent. This was confirmed by Northern gene critical to GPI-anchor biosynthesis (21), is limited to affected blotting analysis (23). In contrast to the total abrogation of GPI-DAF gene blood cells and occurs in conjunction with CD59 dysfunction (12). expression, TM-DAF in the knockout mouse testis was Furthermore, although DAF gene mutation apparently did not lead unaffected (23). These GPI-DAF-null mice could develop, grow, and re- produce normally (23). The original C57BL/6Ð129J knockout mice were to PNH-like disease, two of the five individuals had an intestinal backcrossed with C57BL/6 mice for four generations. Because the founder inflammatory disorder (19, 20). However, due to the rare nature of knockout mice had Ͼ50% C57BL/6 background (C57BL/6 blastocyst were DAF gene mutation in the human population, it has not been pos- used in the generation of chimera, and C57BL/6 females were used in the sible to determine whether such individuals are more susceptible to subsequent germline transmission breeding), the four times-backcrossed mice should have a predominantly C57BL/6 background (Ͼ97%). Litter- complement-mediated inflammatory tissue damage. mates of the backcrossed mice were intercrossed to obtain wild-type and

To address this issue and to aid the study of DAF biology in DAF knockout founder mice. Age- and sex-matched F1 mice were used in vivo, we cloned the mouse homologue of human DAF (22) and all experiments. Mice were housed in a specific pathogen-free facility and were confirmed to be negative for common murine viral pathogens by sera generated a knockout mouse that completely lacks the expression analysis. Experiments were conducted by following established guidelines Downloaded from of the GPI-DAF gene product (23). In the mouse, two DAF , for animal care and all protocols were approved by the appropriate insti- arranged in tandem on mouse 1, have been identified tutional committees. (22, 24, 25). One DAF gene, termed GPI-DAF, is equivalent to Preparation of rabbit anti-GBM serum human DAF in that it encodes a GPI-anchored protein and is ex- The anti-mouse GBM serum was prepared as described by Nagai et al. pressed broadly in mouse tissues (22, 24, 25). The second DAF (30). Briefly, glomeruli were isolated by differential sieving from the renal gene, termed TM-DAF, encodes a transmembrane DAF and is ex- cortex and were disrupted by sonication. The GBM fraction was collected http://www.jimmunol.org/ pressed only on mouse sperm (22, 24, 26). Additionally, the by centrifugation at 76,000 ϫ g for 60 min. Anti-GBM serum was raised mouse and the rat express in many of their tissues a rodent- in Japanese white rabbits by repeated (five times) immunization with pu- rified mouse GBM. For the immunization step, 1 mg GBM protein was specific transmembrane C3 regulator called Crry (7Ð9). Crry emulsified with 1 ml CFA (Difco Laboratories, Detroit, MI) and was ad- was established to possess both DAF and MCP activities (7Ð9) ministered to the rabbit by s.c. injection. and may play the substituting role for MCP in most mouse and Induction of anti-GBM nephritis rat tissues because the MCP gene is expressed only in the testes in these two species (27, 28). The fact that GPI-DAF knockout Anti-GBM nephritis was induced according to a previously described pro- tocol by Nagai et al. (30). Mice were immunized i.p. with 0.5 mg rabbit mice could survive and function normally (23) also suggests IgG per 20 g body weight emulsified with CFA. Five days after immuni- by guest on September 28, 2021 that Crry may be able to compensate DAF function in regulating zation, 0.05 ml anti-GBM serum per 20 g body weight, diluted with 5 parts spontaneous (alternative pathway) complement activation. of saline, was administered intravenously through the orbital plexus. In However, the relative role of DAF and Crry as membrane reg- preliminary experiments, a dose-response curve was established to deter- mine a dose that did not produce nephritis (see below) in normal mice ulators in preventing classical pathway complement activation despite glomerular deposition of IgG. in vivo is not clear. Experimental design In this study, we used the GPI-DAF knockout mice and inves- tigated their susceptibility to complement-mediated inflammatory Mice between 6 and 8 wk of age were used in this study. Preliminary damage in a well-established autoimmune disease model, anti- experiments indicated that male knockout mice were more sensitive than the females to disease induction, presumably reflecting the higher level of glomerular basement membrane (GBM) Ab induced glomerulone- complement activity in male mice in general (39, 40). Accordingly, only phritis in mice (29, 30). The involvement of the complement sys- male mice were used in experiments. Mice were sacrificed either at 8 h tem in anti-GBM glomerulonephritis has been well documented (knockout, n ϭ 5; wild-type, n ϭ 10) or on day 8 (knockout, n ϭ 8, wild-type, n ϭ 5) after administration of a subnephritogenic dose of anti- (31, 32). For example, mice deficient in the complement compo- GBM serum. The kidneys were processed and evaluated as described be- nents C3 and C4 were partially protected from anti-GBM glomer- low. Serum and urine samples were collected at day 8. For urine collection, ular injury (33), and depletion of complement by cobra venom the mice were housed in individual metabolic cages with free access to tap factor in rats and rabbits reduced the degree of inflammation and water. disease progression (34, 35). Furthermore, a soluble form of Northern blot and RT-PCR analyses recombinant Crry, either administered as a fusion protein or Total tissue RNAs were extracted with Trizol reagent (Life Technologies, produced in vivo through transgenic expression, has been demon- Gaithersburg, MD), fractionated in a 1% agarose gel, and transferred to ϩ strated to attenuate the development of Ab-induced glomerulo- Hybond-N nylon membranes. To detect GPI-DAF mRNAs, a 276-bp Ј nephritis in the mouse (36Ð38). Nevertheless, there are comple- 3 -cDNA-specific fragment (22) was used as a probe for hybridization in QuickHyb solution (Stratagene, La Jolla, CA). To detect TM-DAF mRNA, ment-dependent and independent inflammatory responses initiated the membrane was stripped and rehybridized with a 180-bp specific probe by nephrotoxic Abs that may in part be related to the Ab dosage corresponding to the 3Ј-cDNA of TM-DAF (22). Finally, the membrane (33, 34). We demonstrate in this study that a calibrated dose of was stripped again and hybridized with a control probe (GAPDH) to con- firm equal loading of RNAs. First strand cDNAs for RT-PCR were syn- anti-GBM serum caused severe glomerulonephritis in GPI-DAF thesized as previously described (22) using total RNAs from kidneys and knockout mice but not in wild-type controls. This result offers oligo(dT) as a primer. The following two primers were used to amplify direct evidence that DAF protects self-tissues from complement mouse GPI-DAF cDNA: 5Ј-CATACATGTTTAATAACCTTGA Ј Ј attack in an autoimmune disease setting and suggests that activity CAGTTTTG-3 (upstream) and 5 -AACAAACAACACTATTAAATT TATTGTATCC-3Ј (downstream). The following two primers were used to of membrane C3-regulatory is a critical determinant for amplify mouse Crry cDNA: 5Ј-CCAGCCCCATCACAGCTTCCTTCT-3Ј complement susceptibility in autoimmune tissue injury. (upstream); and 5Ј-CTTCCCTCTCGCATCAGTGTT-3Ј (downstream). The Journal of Immunology 2793

Western blot analysis Statistical analysis Membrane proteins of kidneys were solubilized as previously described Values are presented as mean Ϯ SEM. Statistical comparisons and corre- (41). In brief, kidneys were washed with PBS twice; homogenized in a lation analysis were performed with the StatView program (Abacus Con- 20ϫ volume of PBS, pH 7.2, containing 10 mM EDTA, 1% Nonidet P-40, cepts, Berkeley, CA) using the Mann-Whitney U test or Student’s t test as 0.1 mM PMSF, 1 ␮g/ml leupeptin, and 1 ␮g/ml pepstatin A with a Polytron appropriate. A p value of Ͻ0.05 was considered statistically significant. (Polytron, Paterson, NJ); and then solubilized for 30 min at 4¡C. Nuclei and cytoplasmic debris was then pelleted at 14,000 rpm using a microcentrifuge Results for 10 min at 4¡C. Expression of GPI-DAF but not TM-DAF gene in the SDS-PAGE was performed with 300 ␮g/lane of solubilized kidney sam- ples under nonreducing condition. Western blot to detect Crry expression mouse kidney was performed using 1F2 (10 ␮g/ml; PharMingen, San Diego, CA), a rat As alluded to earlier, two DAF genes are known to exist in the anti-mouse Crry mAb (8). The bound Ab was then detected with alkaline phosphatase-conjugated goat anti-rat IgG, (1/500; Promega, WI). 5-Bro- mouse (22, 24). To examine the respective expression of the GPI- mo-4-chloro-3-indolyl phosphate/NBT (Sigma, St. Louis, MO) was used as DAF and TM-DAF genes in the wild-type mouse kidney and the a substrate for alkaline phosphatase. possibility that compensatory expression of the TM-DAF gene might have occurred in the GPI-DAF knockout mouse kidney, Histology Northern blot analysis using cDNA probes specific to either the Renal cortical tissue for microscopy was fixed in methyl Carnoy’s GPI-DAF mRNA or the TM-DAF mRNA were performed. Fig. 1A ␮ solution and embedded in paraffin. Sections of 4 m were cut and stained demonstrates that two GPI-DAF mRNA species were detected in with periodic acid-Schiff and counterstained with hematoxylin. To accu- rately perform quantitative analysis, computer-assisted morphometry was the wild-type but not in the GPI-DAF gene knockout mouse kid- used (42). For each individual animal, nine photographs of randomly cho- ney. The pattern of two distinct mRNA species, presumably a re- Downloaded from sen areas of a cortex were taken at ϫ200 magnification. Each photograph sult of alternative splicing, is similar to that observed in other included one to four glomeruli, and its genotype was blinded to eliminate mouse tissues (23), although the relative abundance of these mes- potential experimental bias. The photographs were converted to picture sages does vary from tissue to tissue (23). No TM-DAF mRNA files using a scanner at a resolution of 300 dots per inch and were evaluated on a video screen using Photoshop software (Adobe Systems, San Diego, was detected in either the wild-type or the GPI-DAF gene knock- CA). Only equatorially sectioned glomeruli were evaluated. With this out mouse kidney. This result suggests that the TM-DAF gene is

method, total glomerular nuclear counts, percentage of glomeruli exhibit- not expressed in the mouse kidney and that the GPI-DAF knockout http://www.jimmunol.org/ ing glomerular injury, and glomerular tuft volume were determined. Glo- mouse is completely deficient of DAF expression in the kidney. merular injury was defined by evidence of segmental increases in glomer- ular matrix, segmental collapse and obliteration of capillary lumina and Comparison of Crry and GPI-DAF expression in wild-type and accumulation of hyaline which was frequently associated with synechial attachments to Bowman’s capsule. To measure the glomerular tuft volume, knockout male and female mouse kidneys mean planar glomerular area was first determined and was used to calculate To determine whether there was compensatory expression of Crry glomerular tuft volume by a previously described formula (43). The mean in the DAF knockout mouse kidney, we performed Western blot planar glomerular area was determined by manually tracing the outer edges of all glomerular tufts in each kidney section and then calculating and analysis of membrane protein extracts of kidneys using a rat anti- summarizing the encircled areas by computerized morphometry (44). mouse Crry mAb. Fig. 1B shows that there is no significant dif- ference in Crry protein levels between wild-type and knockout by guest on September 28, 2021 Immunofluorescence mouse kidneys or between male and female mouse kidneys. Be- To examine the deposition of C3, 4-␮m frozen sections were stained with cause we observed in our preliminary experiments a gender dif- FITC-conjugated goat Abs against mouse C3 (1/100; Cappel, ICN Phar- ference in the sensitivity of glomerulonephritis development, we maceuticals, Aurora, OH). For fibrinogen, FITC-conjugated goat Abs also investigated by RT-PCR whether there is a sex difference in against rat fibrinogen, which have cross-reactivity with the mouse fibrin- ogen (1/400; Cappel, ICN Pharmaceuticals) was used. For semiquantitative the expression of GPI-DAF. As shown in Fig. 1C, RT-PCR anal- analysis of rabbit IgG deposition in glomeruli, indirect immunofluores- ysis revealed no appreciable difference in GPI-DAF expression cence using biotinylated goat anti-rabbit IgG (1/400; Vector Laboratories, between the male and female kidneys. Consistent with the Western Burlingame, CA) and NeutrAvidin-Oregon Green 488 conjugate (Molec- blot data (Fig. 1B), no gender difference was detected by RT-PCR ular Probes, Eugene, OR) was used. Fluorescence-positive glomeruli were counted, and the percentage of positive glomeruli in each sample was cal- in Crry expression between male and female kidneys (Fig. 1C). culated. Deposition of autologous (mouse) IgG in the glomeruli was assessed by using biotinylated goat anti-mouse IgG (1/400; Vector) and DAF knockout mice are more susceptible to nephrotoxic NeutrAvidin-Oregon Green 488 conjugate. serum nephritis Measurement of circulating IgG and IgM levels Preliminary experiments established the minimum and subnephri- togenic doses required for nephritis (0.05 ml/20g body weight). Circulating levels of total IgG and IgM were determined with the Mouse We also observed in preliminary experiments that using this dos- IgG ELISA and Mouse IgM ELISA kit (Bethyl Laboratories, Mont- Ϯ ϭ gomery, TX) according to the manufacturer’s protocol. Circulating levels age, 72 10% of glomeruli in knockout male mice (n 3) of rabbit IgG-specific mouse IgG were measured by ELISA using the pro- showed damage, whereas 3 Ϯ 3% of glomeruli in wild-type male cedure detailed below. Ninety-six-well ELISA plates coated with rabbit mice (n ϭ 3) were injured. Under the same conditions, 43 Ϯ 32% IgG (Organon Teknika, Durham, NC) were incubated with test plasma that of glomeruli in knockout female mice (n ϭ 3) demonstrated dam- was diluted to 1/2000. After being washed extensively with PBS containing age, whereas 3 Ϯ 6% of glomeruli in wild-type female mice (n ϭ 0.05% Tween 20, the plates were incubated with HRP-conjugated anti- mouse IgG (Vector) diluted to 1/1000. For the development, the wells were 3) were injured. Thus, although there was a trend of increased incubated with the reaction solution containing 3,3Ј,5,5Ј-tetramethylbenzi- sensitivity in the knockout mice of both sexes, knockout males dine (Sigma). The reaction was stopped by addition of H2SO4, and the appeared to be more sensitive to nephritis development. Because OD was determined and taken as a measurement of anti-rabbit IgG Abs. 450 male mice have higher complement activity than female mice (39, Measurement of urinary albumin and blood urea 40), we decided to use male mice only in our experiments to sim- nitrogen (BUN) plify the study. Eight days after anti-GBM serum administration, glomeruli of Urinary albumin excretion was measured by a mouse albumin ELISA quantitation kit (Bethyl Laboratories) according to the manufacturer’s pro- DAF knockout mice appeared enlarged and showed global glo- tocol. BUN was measured by the urease-indophenol method with a Urea merular hypercellularity and segmental sclerosis at the periphery NB kit (Wako Pure Chemical Industries, Tokyo, Japan). of the tuft (Fig. 2). Furthermore, the knockout mice had significant 2794 DAF AND ANTI-GBM GLOMERULONEPHRITIS

FIGURE 1. Analysis of GPI-DAF, TM-DAF, and Crry gene expression in wild-type and GPI-DAF knockout mouse kidneys. A, Northern blot analysis of DAF expression in the wild-type (ϩ/ϩ) and GPI- DAF knockout (Ϫ/Ϫ) mouse kidneys. GPI-DAF is expressed in the wild-type but not the knockout mouse kidney. TM-DAF is not expressed in either the wild-type or the knockout mouse kidney. Equal loading of total RNA in the two lanes is indicated by using GAPDH cDNA as a control probe. The same membrane was used in all three hybridizations. Left ordinate, Positions of the 18S and 28S ribosomal RNAs. B, Western blot analysis of Crry expression in wild-type (lanes 1 and 3) and knockout (lanes 2 and 4) male (lanes 1 and 2) or female (lanes 3 and Downloaded from 4) kidneys. Left ordinate, Positions of molecular mass markers. C, RT-PCR analysis of GPI-DAF and Crry expression in male (M) or female (F) mouse kidneys. Reactions were conducted either with (ϩ) or without (Ϫ) reverse transcriptase (RT) during the first strand cDNA synthesis. http://www.jimmunol.org/

accumulation of mesangial matrix (Fig. 2C), and in some cases ysis of the morphological changes. Both glomerular volume and there were completely collapsed lobules and enlarged epithelial the number of cells per glomerulus were significantly increased in by guest on September 28, 2021 cells (Fig. 2D). By contrast, the kidneys of the wild-type mice were the DAF knockout mice as compared with controls. Even more largely intact, with patent capillaries without increased matrix or striking is the increased percentage of glomeruli that showed signs cellularity (Fig. 2A). Fig. 3 shows the results of quantitative anal- of injury in the knockout mice (Fig. 3C). Approximately 70% of knockout glomeruli showed signs of injury whereas relatively few in the wild-type sustained similar degree of damage (Fig. 3). On day 8 after nephrotoxic serum administration, when histo- logical evidence of nephritis was apparent in the knockout mice, the BUN level remained normal in the knockout mice (26.4 Ϯ 1.0 and 27.6 Ϯ 1.6 mg/dl in knockout and wild-type, respectively), but the urinary albumin level was elevated in the knockout mice (1.4 Ϯ 0.7 mg/24 h for the knockout compared with 0.02 Ϯ 0.01 mg/24 h for the wild-type mice; Table I). No wild-type mice de- veloped abnormal albuminuria (Ͻ0.05 mg/day). However, three of the knockout mice with the most severe glomerular damage ex- creted Ͼ2 mg/day of albumin, and there was a significant corre- lation between albuminuria and glomerular damage as assessed by the percentage of glomeruli showing histological signs of injury under light microscopy (r ϭ 0.88, p ϭ 0.0022).

Increased complement activation and fibrin deposition in DAF knockout mouse kidneys Potential mechanisms of increased nephritis were then explored. Eight hours after anti-GBM serum treatment, there was prominent FIGURE 2. Representative light microscopy (periodic acid-Schiff) of deposition of rabbit as well as mouse IgG in the wild-type and glomeruli from wild-type (A) and DAF knockout mice (BÐD) 8 days after knockout kidneys (Fig. 4, top and bottom rows). By contrast, dep- administration of nephrotoxic serum. The wild-type mouse kidneys showed minimal pathological change. By contrast, glomeruli from DAF knockout osition of C3 in capillary loops of glomeruli was observed only in mice were enlarged and hypercellular (B), and segmental sclerosis with the knockout mouse kidneys (Fig. 4, second row, and Table II). accumulation of mesangial matrix was evident (C). D, Inflamed glomerulus This was associated with fibrinogen deposition and thrombus for- with a completely collapsed lobule and enlarged epithelial cells from a mation, which was detected only in the glomeruli of the knockout knockout mouse. Original magnification, ϫ200. mice (Fig. 4, third row, and Table II). Analysis of circulating total The Journal of Immunology 2795 Downloaded from

FIGURE 4. Immunofluorescence staining of rabbit IgG, mouse C3, fi- brinogen, and mouse IgG deposition in glomeruli of normal (nontreated wild-type (column A), anti-GBM-treated wild-type (column B), or anti- GBM-treated DAF knockout (column C) mice. Equal deposition of rabbit

IgG (8 h after treatment) and mouse IgG (7 days) was observed in the http://www.jimmunol.org/ treated mice regardless of the genotype (columns B and C, top and bottom rows). However, C3 and fibrinogen deposition, mainly along the glomer- ular capillary loops was observed only in the treated knockout mouse glo- meruli (8 h after treatment) (columns B and C, middle two rows; see also Table II). In the middle two rows, background staining of C3 and fibrin in the glomeruli is indicated by arrows and specific staining is indicated by FIGURE 3. Quantitative analysis of glomerulonephritis in wild-type arrowheads. Background staining of C3 was distinct, mostly restricted to (n ϭ 5) and DAF knockout (n ϭ 8) mice after anti-GBM treatment. At least Bowman’s capsules, and may indicate local C3 synthesis (45) or nonspe- cific activity of the anti-mouse C3 Ab used. nine sections, each containing one to four glomeruli, from each mouse by guest on September 28, 2021 were examined. Compared with wild-type animals, DAF knockout mouse kidneys had larger glomerular volumes (A, 516 Ϯ 68 vs 325 Ϯ 18 ϫ 103 ␮ 3 Ϯ Ϯ m per glomerulus), displayed higher cellularity (B, 47.1 8.9 vs 32.0 glomerulonephritis in rodents has been often used as an animal 3.1 cells per glomerulus) and contained more injured glomeruli (C, 68.8 Ϯ model to dissect the roles of complement and other inflammatory 25.0 vs 10.0 Ϯ 3.5%). p Ͻ 0.05 for all. pathways in immune glomerulonephritis (29, 30, 46). However, the role of complement-regulatory proteins has not been ade- IgG, IgM at day 1 and rabbit IgG-specific mouse IgG at day 7 after quately addressed. In this study, we used GPI-DAF knockout mice, disease induction (i.e., anti-GBM administration) also revealed no generated in our laboratory (23), to evaluate the role of an endog- significant difference between wild-type and knockout mice (total enous membrane complement-regulatory protein, DAF, in the IgG 0.47 Ϯ 0.11 mg/ml for wild-type, n ϭ 5, 0.50 Ϯ 0.11 mg/ml pathogenesis of nephrotoxic serum glomerulonephritis. Our results for knockout, n ϭ 8; total IgM 0.21 Ϯ 0.02 mg/ml for wild-type, n ϭ show that DAF knockout mice are more susceptible to nephrotoxic 5, and 0.17 Ϯ 0.02 mg/ml for knockout n ϭ 8; ELISA readings of serum glomerulonephritis than the wild-type controls. Injection of Ϯ a subnephritogenic (in normal mice) rabbit anti-mouse GBM se- rabbit IgG-specific mouse IgG assay 480 201 OD450 for wild-type, ϭ Ϯ ϭ ϭ rum caused significant nephritis in DAF knockout mice. n 5, and 589 442 OD450 for knockout, n 8; p 0.62). DAF is a central membrane complement regulator which in hu- Discussion mans has been mostly studied in the context of PNH syndrome and The plays paradoxical roles in the pathogen- xenotransplantation experiments (12, 47). Although a number of esis and manifestations of autoimmune diseases (1, 2). Although studies have examined the expression of DAF protein in both nor- complement, particularly its early components, is recognized to mal and diseased human kidneys (16, 17), the role of DAF in facilitate clearance (1, 2), full activation of the immune glomerulonephritis has not been adequately addressed. complement system in an autoimmune disease setting can generate inflammatory and cytolytic mediators (31, 32). Anti-GBM-induced Table II. C3 and fibrinogen deposition in glomeruli of wild-type and DAF knockout mice Table I. Urinary albumin and BUN levels in wild-type and DAF knockout mice Knockout Wild-Type (n ϭ 5) (n ϭ 10) Knockout (n ϭ 8) Wild-Type (n ϭ 5) a 0 Ϯ 0ءC3 deposition (%) 18.1 Ϯ 15.4 Ϯ 0 0 ءUrinary albumin (mg/24 h) 1.4 Ϯ 0.7 0.02 Ϯ 0.01 Fibrinogen deposition (%) 4.7 Ϯ 2.2 BUN (mg/dl) 26.4 Ϯ 1.0 27.6 Ϯ 1.6 .(p Ͻ 0.05 vs wild-type group (Mann-Whitney U test ,ء 2796 DAF AND ANTI-GBM GLOMERULONEPHRITIS

Investigation using animal models has been hampered by the ini- that female mice may have less circulating complement activity tial slow pace in identifying the animal orthologues of human DAF (39, 40). A similar gender bias in complement-dependent pheno- in nonprimate species (7). It is now known, however, that both the types was also observed in a CD59 gene knockout mouse (40). rat and the mouse contain DAF genes (4), as well as a related and The present results that DAF-deficient mice were more sensitive rodent-specific membrane complement regulator Crry (4). In fact, to glomerulonephritis provide further evidence for the role of com- two DAF genes, possibly resulting from a recent gene duplication plement in immune-mediated glomerular damage. Although it has event, have been identified in the mouse (22, 24). We showed here been generally accepted based on clinical and animal experimental that the GPI-DAF gene but not the TM-DAF gene is expressed in data that complement plays a detrimental role in immune glomer- the mouse kidney. Furthermore, although the TM-DAF gene is still ular damage (31, 32), some recent experiments using Fc receptor intact in the GPI-DAF knockout mouse (23), no compensatory knockout mice have questioned the relevance of the complement expression of TM-DAF in the GPI-DAF knockout mouse kidney pathway in autoimmune glomerulonephritis (55, 56). In contrast, was observed. Thus, our GPI-DAF gene knockout mouse provides studies of C3- and C4-deficient mice have clearly demonstrated a an appropriate animal model to examine the consequence of DAF role of complement in anti-GBM glomerulonephritis and showed deficiency in immune glomerulonephritis development. that the relative contribution in the disease pathogenesis by the A number of studies have demonstrated a protective role for two complement pathway is dependent on the dosage of the nephro- other membrane regulators of complement activation, the mem- toxic Abs used (33). Results presented here suggest that the func- brane attack complex inhibitor CD59 and the rodent-specificC3 tion of membrane complement-regulatory proteins may be another inhibitor Crry. For example, it has been shown that neutralization critical variable in determining the degree of complement-medi- of CD59 function exacerbated complement-mediated kidney in- ated inflammatory damage in an immune disease setting. Downloaded from jury in the rat (48). Similarly, in a rat model of thrombotic mi- croangiopathy induced by anti-endothelial cell Abs, perfusion of Acknowledgments animals with anti-CD59 led to more severe glomerular endothelial We thank Dr. Takeshi Sugaya (Tanabe Seiyaku, Osaka, Japan) and Dr. damage with enhanced and fibrin deposition (49). In a Norio Hanafusa (University of Tokyo School of Medicine, Tokyo, Japan) related study, neutralization of rat Crry caused tubulointerstitial for their technical support. We also thank Drs. Reiko Inagi, Toshio Miyata, injury in the kidney (50). Much evidence in support of a protective and Kiyoshi Kurokawa (Tokai University School of Medicine, Kanagawa, http://www.jimmunol.org/ role of Crry in immune glomerular damage has also accumulated Japan) for their generous support. from studies using soluble Crry as a systemic fluid phase inhibitor. In this regard, both administration of a recombinant, soluble Crry References systemically or overexpression of Crry in transgenic mice was pro- 1. Volanakis, J. E., and M. Frank. 1998. The Human Complement System in Health tective from Ab-induced glomerular injury (36Ð38). Thus, it is and Disease. Dekker, New York. especially remarkable that DAF knockout mice were more sensi- 2. Morgan, B. P. 1995. Physiology and pathophysiology of complement: progress and trends. Crit. Rev. Clin. Lab. Sci. 32:265. tive to nephrotoxic serum, despite normal Crry expression in the 3. Hourcade, D., V. M. Holer, and J. P. Atkinson. 1989. The regulators of comple- DAF knockout mouse kidney (Fig. 1). Although our finding does ment activation (RCA) gene cluster. Adv. Immunol. 45:381. not exclude a protective role of membrane-anchored Crry in this 4. Miwa, T., and W.-C. Song. 2001. Membrane complement regulatory proteins: by guest on September 28, 2021 insight from animal studies and relevance to human diseases. Int. Immunophar- disease model, it does suggest that the function of DAF in the macology 1:445. mouse kidney cannot be completely compensated by Crry in this 5. Lublin, D. M., and J. P. Atkinson. 1989. Decay-accelerating factor: biochemistry, molecular biology, and function. Annu. Rev. Immunol. 7:35. model. The relative efficacy of protection by membrane-anchored 6. Nangaku, M. 1998. Complement regulatory proteins in glomerular diseases. Kid- Crry and DAF in this disease model remains to be defined. How- ney Int. 54:1419. ever, recent in vitro assays have established that, at least in the 7. Holers, V. M., T. Kinoshita, and H. Molina. 1992. The evolution of mouse and human complement C3-binding proteins: divergence of form but conservation of fluid phase, recombinant mouse DAF is more active on a molar function. Immunol. Today 13:231. basis than recombinant Crry as a classical pathway complement 8. Li, B., C. Sallee, M. Dehoff, S. Foley, H. Molina, and V. M. Holers. 1993. Mouse inhibitor, whereas the reverse is true concerning their potency as Crry/p65: characterization of monoclonal and the tissue distribution of a functional homologue of human MCP and DAF. J. Immunol. 151:4295. alternative pathway complement inhibitors (51). 9. Sakurada, C., H. Seno, N. Dohi, H. Takizawa, M. Nonaka, N. Okada, and The principal mediators responsible for the increased sensitivity H. Okada. 1994. Molecular cloning of the rat complement regulatory protein, 5I2 of DAF knockout mice to nephrotoxic serum are not yet known. It . Biochem. Biophys. Res. Commun. 198:819. 10. Okada, N., R. Harada, T. Fujita, and H. Okada. 1989. A novel membrane gly- is well recognized that immune glomerular damage can occur as a coprotein capable of inhibiting membrane attack by homologous complement. result of and C5a generation or of C5b-9 deposition on kidney Int. Immunol. 1:205. cells (52, 53). Because DAF inhibits C3 convertases, thus acting at 11. Davies, A., D. L. Simmons, G. Hale, R. A. Harrison, H. Tighe, P. J. Lachmann, and H. Waldmann. 1989. CD59, an Ly-6-like protein expressed in human lym- an early step of the complement activation cascade, deficiency of phoid-cells, regulates the action of the complement membrane attack complex on DAF is expected to increase local C3a and C5a biosynthesis after homologous cells. J. Exp. Med. 170:637. nephrotoxic serum binding to their target Ags. Moreover, human 12. Rosse, W. F., and C. J. Parker. 1985. Paroxysmal nocturnal hemoglobinuria. Clin. Haematol. 14:105. DAF can also inhibit the activity of C5 convertase (5, 54). Thus, 13. Nicholson-Weller, A., J. P. March, S. I. Rosenfeld, and K. F. Austen. 1983. it is possible that increased deposition of the C5b-7 complex on Affected erythrocytes of patients with paroxysmal nocturnal haemoglobinuria are resident glomerular cells will overwhelm the protective effect of deficient in the complement regulatory protein decay-accelerating factor. Proc. Natl. Acad. Sci. USA 80:5066. CD59, resulting in increased membrane attack complex deposition 14. Medof, M. E., T. Kinoshita, R. Silber, and V. Nussenzweig. 1985. Amelioration and damage in the knockout kidney. Consistent with increased of lytic abnormalities of paroxysmal nocturnal hemoglobinuria with decay-ac- local C3 activation, our immunofluorescence experiments showed celerating factor. Proc. Natl. Acad. Sci. USA 82:2980. 15. Medof, M. E., E. I. Walter, J. L. Rutgers, D. M. Knowles, and V. Nussenzweig. enhanced C3b deposition in the knockout mouse glomeruli. Fi- 1987. Identification of the complement decay-accelerating factor (DAF) on epi- brinogen deposition, indicating glomerular thrombosis and vascu- thelium and glandular cells and in body fluids. J. Exp. Med. 165:848. 16. Abe, K., M. Miyazaki, T. Koji, A. Furusu, Y. Ozono, T. Harada, H. Sakai, lar damage, is also a hallmark of complement-mediated kidney P. K. Nakane, and S. Kohno. 1998. Expression of decay accelerating factor damage and was observed in the knockout mouse glomeruli. In our mRNA and complement C3 mRNA in human diseased kidney. Kidney Int. 54: pilot experiments, we found that DAF knockout females were not 120. 17. Shibata, T., F. G. Cosio, and D. J. Birmingham. 1991. Complement activation as sensitive to nephrotoxic serum glomerular damage as the DAF induces the expression of decay-accelerating factor on human mesangial cells. knockout males. One possible explanation for this phenomenon is J. Immunol. 147:3901. The Journal of Immunology 2797

18. Quigg, R. J., A. Nicholson-Weller, A. V. Cybulsky, J. Badalamenti, and 37. Quigg, R. J., C. He, A. Lim, D. Berthiaume, J. J. Alexander, D. Kraus, and D. J. Salant. 1989. Decay accelerating factor regulates complement activation on V. M. Holers. 1998. Transgenic mice overexpressing the complement inhibitor glomerular epithelial cells. J. Immunol. 142:877. Crry as a soluble protein are protected from -induced glomerular injury. 19. Telen, M. J., S. E. Hall, A. M. Green, J. I. Moulds, and W. F. Rosse. 1988. J. Exp. Med. 188:1321. Identification of human erythrocyte blood group on decay-accelerating 38. Schiller, B., P. N. Cunningham, J. J. Alexander, L. Bao, V. M. Holers, and factor (DAF) and an erythrocyte phenotype negative for DAF. J. Exp. Med. R. J. Quigg. 2001. Expression of a soluble complement inhibitor protects trans- 167:1993. genic mice from antibody-induced acute renal failure. J. Am. Soc. Nephrol. 12:71. 20. Wang, L., M. Uchikawa, H. Tsuneyama, K. Tokunaga, K. Tadokoro, and T. Juji. 39. Beurskens, F. J. M., J. D. M. Kuenen, F. Hofhuis, A. C. Fluit, D. M. Robins, and 1998. Molecular cloning and characterization of decay-accelerating factor defi- H. van Dijk. 1999. Sex-limited protein: in vitro and in vivo functions. Clin. Exp. ciency in Cromer blood group Inab phenotype. Blood 91:680. Immunol. 116:395. 21. Miyata, T., N. Yamada, Y. Iida, J. Nishimura, J. Takeda, T. Kitani, and 40. Holt, D. S., M. Botto, A. E. Bygrave, N. K. Rushmere, M. J. Walport, and T. Kinoshita. 1994. Abnormalities of PIG-A transcripts in granulocytes from B. P. Morgan. 2000. Knocking out mouse CD59 using a PCR derived gene patients with paroxysmal nocturnal hemoglobinuria. N. Engl. J. Med. 330:249. targeting vector. Immunopharmacology 49:67 (Abstr.). 22. Song, W.-C., C. Deng, K. Raszmann, R. Moore, R. Newbold, J. A. McLachlan, 41. Nangaku, M., R. J. Quigg, S. J. Shankland, N., Okada, R. J. Johnson, and and M. Negishi. 1996. Mouse decay-accelerating factor, selective and tissue- W. G. Couser. 1997. Overexpression of Crry protects mesangial cells from com- specific induction by estrogen of the gene encoding the glycosylphosphatidyl- plement-mediated injury. J. Am. Soc. Nephrol. 8:223. inositol-anchored form. J. Immunol. 157:4166. 42. Quigg, R. J., A. Lim, M. Haas, J. J. Alexander, C. He, and M. C. Carroll. 1998. 23. Sun, X., C. D. Funk, C. Deng, A. Sahu, J. D. Lambris, and W.-C. Song. 1999. Immune complex glomerulonephritis in C4- and C3-deficient mice. Kidney Int. Role of decay-accelerating factor in regulating complement activation on the 53:320. erythrocyte surface as revealed by gene targeting. Proc. Natl. Acad. Sci. USA 43. Hirose, K., R. Osterby, M. Nakazawa, H. Jorgensen, and G. Gumderson. 1982. 96:628. Development of glomerular lesions in experimental long-term diabetes in the rat. 24. Spicer, A. P., M. F. Seldin, and S. J. Gendler. 1995. Molecular cloning and Kidney Int. 21:689. chromosomal localization of the mouse decay-accelerating factor genes: dupli- 44. Floege, J., B. Hackmann, V. Kliem, W. Kriz, C. E. Alpers, R. J. Johnson, cated genes encode glycosylphosphatidylinositol-anchored and transmembrane K. W. Kuhn, K. W., and R. Brunkhorst. 1997. Age-related glomerulosclerosis forms. J. Immunol. 155:3079. and interstitial fibrosis in Milan normotensive rats. Kidney Int. 51:230. Downloaded from 25. Fukuoka, Y., A. Yasui, N. Okada, and H. Okada. 1996. Molecular cloning of 45. Tang, S., W. Zhou, N. S. Sheerin, R. W. Vaughan, and S. H. Sacks. 1999. murine decay-accelerating factor by immunoscreening. Int. Immunol. 8:379. Contribution of renal secreted complement C3 to the circulating pool in humans. 26. Miwa, T., X. Sun, R. Ohta, N. Okada, C. Harris, B. P. Morgan, and W.-C. Song. J. Immunol. 162:4336. 2001. Characterization of GPI-DAF and TM-DAF expression in wild-type and 46. Feith, G. W., K. J. Assmann, M. J. Bogman, A. P. van Gompel, J. Schalkwijk, GPI-DAF gene knockout mice using polyclonal and monoclonal antibodies with and R. A. Koene. 1996. Different mediator systems in biphasic heterologous dual or single specificity. Immunology. In press. phase of anti-GBM nephritis in mice. Nephrol. Dial. Transplant. 11:599. 27. Miwa, T., M. Nonaka, N. Okada, S. Wakana, T. Shiroishi, and T. Okada. 1998. 47. Storck, M., D. Abendroth, R. Prestel, G. Pino-Chavez, J. Muller-Hoker, Molecular cloning of rat and mouse membrane cofactor protein (MCP, CD46): D. J. White, and C. Hammer. 1997. Morphology of hDAF (CD55) transgenic pig preferential expression in testis and close linkage between the mouse Mcp and http://www.jimmunol.org/ kidneys following ex-vivo hemoperfusion with human blood. Transplantation Cr2 genes on distal . Immunogenetics 48:363. 63:304. 28. Tsujimura, A., K. Shida, M. Kitamura, M. Nomura, J. Takeda, H. Tanaka, M. Matsumoto, K. Matsumiya, A. Okuyama, Y. Nishimune, M. Okabe, and 48. Matsuo, S., H. Nishikage, F. Yoshida, A. Nomura, S. J. Piddlesden, and T. Seya. 1998. Molecular cloning of a murine homologue of membrane cofactor B. P. Morgan. 1994. Role of CD59 in experimental glomerulonephritis in rats. protein (CD46): preferential expression in testicular germ cells. Biochem. J. 330: Kidney Int. 46:191. 163. 49. Nangaku, M., C. E. Alpers, J. Pippin, S. J. Shankland, K. Kurokawa, S. Adler, 29. Schrijver, G., K. J. M. Assmann, J. J. T. Bogman, J. C. M. Robben, B. P. Morgan, R. J. Johnson, and W. G. Couser. 1998. CD59 protects glomerular R. M. W. de Waal, and J. C. M. Robben. 1988. Antiglomerular basement mem- endothelial cells from immune-mediated thrombotic microangiopathy in rats. brane nephritis in the mouse: study on the role of complement in the heterologous J. Am. Soc. Nephrol. 9:590. phase. Lab. Invest. 59:484. 50. Nomura, A., K. Nishikawa, Y. Yuzawa, H. Okada, N. Okada, B. P. Morgan, 30. Nagai, H., T. Takizawa, T. Nishiyori, and A. Koda. 1982. Experimental glomer- S. J. Piddlesden, M. Nadai, T. Hasegawa, and S. Matsuo. 1995. Tubulointerstitial

ulonephritis in mice as a model for immunopharmacological studies. Jpn. injury induced in rats by a monoclonal antibody that inhibits function of a mem- by guest on September 28, 2021 J. Pharmacol. 32:1117. brane inhibitor of complement. J. Clin. Invest. 96:2348. 31. Couser, W. G., M. Nangaku, S. J. Shankland, and R. J. Johnson. 1997. Molecular 51. Kraus, D., J. M. Guthridge, H. C. Marsh, Jr., and V. M. Holers. 2000. A direct mechanisms of experimental glomerulonephritis: an overview. Nephrology comparison of complement inhibitory capacities of the GPI- and transmembrane 3:633. forms of mouse DAF to mouse Crry and human rsCR1. Immunopharmacology 32. Hebert, L. A., Cosio, F. G., Birmingham, D. J. 1992. The role of the complement 49:64 (Abstr.). system in renal injury. Semin. Nephrol. 12:408. 52. Schrijver, G., M. J. J. T. Bogman, K. J. M. Assmann, R. M. W. de Waal, 33. Sheerin, N. S., T. Springall, M. C. Carroll, B. Hartley, and S. H. Sacks. 1997. H. C. M. Robben, H. Gasteren, and R. A. P. Koene. 1990. Anti-GBM nephritis Protection against anti-glomerular basement membrane (GBM)-mediated nephri- in the mouse: role of granulocytes in the heterologous phase. Kidney Int. 38:86. tis in C3- and C4-deficient mice. Clin. Exp. Immunol. 110:403. 53. Nangaku, M., J. Pippin, and W. G. Couser. 1999. Complement membrane attack 34. Boyce, N. W., and S. R. Holdsworth. 1985. Anti-glomerular basement membrane complex (C5b-9) mediates interstitial disease in experimental nephrotic syn- antibody-induced experimental glomerulonephritis: evidence for dose-dependent, drome. J. Am. Soc. Nephrol. 10:2323. direct antibody and complement-induced, cell-independent injury. J. Immunol. 54. Medof, M. E., T. Kinoshita, and V. Nussenzweig. 1984. Inhibition of comple- 135:3918. ment activation on the surface of cells after incorporation of decay-accelerating 35. Lefkowith, J. B., T. Nagamatsu, J. Pippin, and G. F. Schreiner. 1991. Role of factor (DAF) into their membranes. J. Exp. Med. 160:1558. leukocytes in metabolic and functional derangements of experimental glomeru- 55. Park, S. Y., S. Ueda, H. Ohno, Y. Hamano, M. Tanaka, T. Shiratori, T. Yamazaki, lonephritis. Am. J. Physiol. 261:F213. H. Arase, N. Arase, A. Karasawa, et al. 1998. Resistance of Fc receptor-deficient 36. Quigg, R. J., Y. Kozono, D. Berthiaume, A. Lim, D. J. Salant, A. Weinfeld, mice to fatal glomerulonephritis. J. Clin. Invest. 102:1229. P. Griffin, E. Kremmer, and V. M. Holers. 1998. Blockade of antibody-induced 56. Clynes, R., C. Dumitru, and J. V. Ravetch. 1998. Uncoupling of immune complex glomerulonephritis with Crry-Ig, a soluble murine complement inhibitor. J. Im- formation and kidney damage in autoimmune glomerulonephritis. Science munol. 160:4553. 279:1052.