Complement Deficiencies Result in Surrogate Pathways of Complement Activation in Novel Polygenic Lupus-like Models of Kidney Injury This information is current as of October 2, 2021. Sladjana Skopelja-Gardner, Lucrezia Colonna, Payton Hermanson, Xizhang Sun, Lena Tanaka, Joyce Tai, Yenly Nguyen, Jessica M. Snyder, Charles E. Alpers, Kelly L. Hudkins, David J. Salant, YuFeng Peng and Keith B. Elkon

J Immunol published online 1 April 2020 Downloaded from http://www.jimmunol.org/content/early/2020/03/31/jimmun ol.1901473

Supplementary http://www.jimmunol.org/content/suppl/2020/03/31/jimmunol.190147 http://www.jimmunol.org/ Material 3.DCSupplemental

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 October 2, 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 © 2020 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Published April 1, 2020, doi:10.4049/jimmunol.1901473 The Journal of Immunology

Complement Deficiencies Result in Surrogate Pathways of Complement Activation in Novel Polygenic Lupus-like Models of Kidney Injury

Sladjana Skopelja-Gardner,* Lucrezia Colonna,* Payton Hermanson,* Xizhang Sun,* Lena Tanaka,* Joyce Tai,* Yenly Nguyen,* Jessica M. Snyder,† Charles E. Alpers,‡ Kelly L. Hudkins,‡ David J. Salant,x YuFeng Peng,*,1 and Keith B. Elkon*,{,1

Lupus nephritis (LN) is a major contributor to morbidity and mortality in lupus patients, but the mechanisms of kidney damage remain unclear. In this study, we introduce, to our knowledge, novel models of LN designed to resemble the polygenic nature of human lupus by embodying three key genetic alterations: the Sle1 interval leading to anti-chromatin autoantibodies; Mfge82/2, leading to defective clearance of apoptotic cells; and either C1q2/2 or C32/2, leading to low complement levels. We report that Downloaded from proliferative glomerulonephritis arose only in the presence of all three abnormalities (i.e., in Sle1.Mfge82/2C1q2/2 and Sle1.Mfge82/2C32/2 triple-mutant [TM] strains [C1q2/2TM and C32/2TM, respectively]), with structural kidney changes re- sembling those in LN patients. Unexpectedly, both TM strains had significant increases in autoantibody titers, Ag spread, and IgG deposition in the kidneys. Despite the early complement component deficiencies, we observed assembly of the pathogenic terminal complement membrane attack complex in both TM strains. In C1q2/2TM mice, colocalization of MASP-2 and C3 in both the glomeruli and tubules indicated that the likely contributed to complement activation and tissue injury in this http://www.jimmunol.org/ strain. Interestingly, enhanced thrombin activation in C32/2TM mice and reduction of kidney injury following attenuation of thrombin generation by argatroban in a serum-transfer nephrotoxic model identified thrombin as a surrogate pathway for complement activation in C3-deficient mice. These novel mouse models of human lupus inform the requirements for nephritis and provide targets for intervention. The Journal of Immunology, 2020, 204: 000–000.

ystemic lupus erythematosus (SLE) is a complex polygenic SLE: 1) low serum complement levels, 2) accumulation of AC/AC autoimmune disease associated with more than 90 genetic debris in the germinal centers (GC) and kidney tissue, and 3) risk loci (1). Despite the large number of genes implicated, presence of autoantibodies (3–7). How these pathways interact S by guest on October 2, 2021 distinctive pathways of immune function, such as complement and lead to lupus nephritis (LN), the main contributor to morbidity activation, immune complex (IC) and apoptotic cell (AC) pro- and mortality in SLE patients (8, 9), remains unclear. cessing, B cell tolerance, and IFN, are affected (2). Involvement Both animal models and evidence in humans indicate that de- of these pathways is commonly reflected in three hallmarks of fective AC clearance is an important component of SLE devel- opment (4, 6, 10, 11). Cells from SLE patients show accelerated death ex vivo (12) and impaired phagocytosis of AC (13), both of *Division of Rheumatology, University of Washington, Seattle, WA 98109; which could be the contributing factors to increased accumulation †Department of Comparative Medicine, University of Washington, Seattle, WA of cellular debris found in the GC of SLE patients (14). Multiple 98109; ‡Department of Nephrology, University of Washington, Seattle, WA 98109; x { , including milk fat globule epidermal growth factor 8 Division of Nephrology, Boston University, Boston, MA 02215; and Department of Immunology, University of Washington, Seattle, WA 98109 (MFGE8), and cell surface receptors have been shown to bind AC 1Y.P. and K.B.E. are joint last authors. and enable their clearance (4, 6, 15). MFGE8 is a soluble glyco- ORCIDs: 0000-0001-6350-9582 (S.S.-G.); 0000-0002-1859-1772 (C.E.A.). protein that binds to phosphatidylserine on AC, facilitating the removal of AC through its interaction with the a b on Received for publication December 12, 2019. Accepted for publication March 15, v 3 2020. phagocytes (10, 16). Aberrant splicing of Mfge8 has been asso- This work was supported by National Institutes of Health (NIH)/National Institute of ciated with dysregulated MFGE8 function in SLE patients, as have Environmental Health Sciences Grant P30ES007033, Foundation for the NIH Grant genetic polymorphisms in Mfge8 (17, 18). MFGE8-deficient mice 5T32AR007108-40, and by the Lupus Research Alliance. accumulate AC in GC and develop a lupus-like disease on the Address correspondence and reprint requests to Dr. YuFeng Peng and Dr. Keith B. mixed C57BL/6 3 129 (10) but not on the pure C57BL/6 (B6) Elkon, Division of Rheumatology, University of Washington, 750 Republican Street E531, Seattle, WA 98109. E-mail addresses: [email protected] (Y.P.) and background (19). Therefore, defective AC clearance alone is not [email protected] (K.B.E.) sufficient to induce a lupus-like pathology but can fuel disease The online version of this article contains supplemental material. progression in a susceptible environment. Abbreviations used in this article: AC, apoptotic cell; DM, double-mutant; fibrin(ogen), The role of complement in kidney disease is complex: early fibrin/fibrinogen; GC, germinal center; IC, immune complex; IF, immunofluorescence; components (C1q, C2, C4, and C3) function to clear AC and ICs, LN, lupus nephritis; LP, lectin pathway; MAC, membrane attack complex; MASP-2, mannan-binding lectin-associated serine protease 2; MFGE8, milk fat globule epidermal whereas late components form the membrane attack complex growth factor 8; NTN, nephrotoxic nephritis; NTS, nephrotoxic serum; PAS, periodic (MAC) that acts as an effector of kidney injury in SLE (20–22). acid–Schiff; SLE, systemic lupus erythematosus; TF, tissue factor; TM, triple-mutant; Specifically, more than 90% of C1q-deficient individuals develop UACR, urine albumin/creatinine ratio. SLE, including LN in a subset of those patients (3), and anti-C1q Copyright Ó 2020 by The American Association of Immunologists, Inc. 0022-1767/20/$37.50 Abs in SLE patients are associated with more severe LN (23). C3

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1901473 2 SURROGATE PATHWAYS OF COMPLEMENT ACTIVATION IN LUPUS MODELS deficiency can also lead to SLE, characterized by LN in ∼26% of alkaline phosphatase–conjugated goat anti-mouse IgG (Sigma-Aldrich). C3-deficient patients (24). In addition, low circulating C3 levels For confirmation of anti-dsDNA IgG, Crithidia luciliae slides (Bio-Rad are strongly associated with active and recurrent renal disease in Laboratories) were used. Serum samples were diluted in PBS/FBS (1:50), and IgG binding was detected by Alexa 555–conjugated anti-mIgG Ab. To SLE patients, in part because of consumption (25, 26). Whereas analyze the autoantibody repertoire, pooled serum samples from six dif- previous studies have suggested that C1q and C3 play a role in ferent strains used in the spontaneous disease model (12-mo-old females, AC clearance and B cell tolerance (3, 27–29), the mechanisms n = 15–20) were examined for reactivity to a defined autoantigen array at that lead to kidney injury in low complement states in the presence the University of Texas Microarray Center (University of Texas South- western). Normalized data were clustered according to the multidimen- of autoantibodies and defective clearance of dead or dying cells sional principal component analysis and visualized in heatmap form using remain unknown. Qlucore Omics Explorer data analysis software; statistically significant To determine the mechanisms of kidney injury in a disease expression of strain-specific autoantigens was determined (Qlucore, Lund, model that closely mimics human SLE pathogenesis, we generated Sweden). SLE-like polygenic murine models that have defective clearance Immunohistochemistry and pathology evaluation of AC (Mfge8-deficiency), produce anti-chromatin autoantibodies (Sle1 interval), and have low complement levels (C1q-orC3- Formalin fixed kidney tissues sections (3 mm) were stained with periodic acid–Schiff (PAS) at the University of Washington Pathology Research deficiency). We demonstrate that spontaneous nephritis and pro- Service Laboratory. Kidneys were scored in a blinded fashion using a teinuria, associated with increased IgG deposition in the kidney and modification of previously described scoring systems (34, 35). Briefly, the enhanced B cell autoreactivity, occur only in the presence of all following were scored: glomerular proliferation/PAS-positive material, three abnormalities (i.e., in Sle1.Mfge82/2C1q2/2 [C1q2/2 triple- glomerular hypercellularity and/or cellular infiltration, tubular protein, and 2 2 2 2 2 2 interstitial inflammation. For glomerular mesangial expansion, the fol- mutant (TM)] or Sle1.Mfge8 / C3 / [C3 / TM] mice, both of

lowing scores were assigned: 0 = absent; 1 = minimal, with thickening of Downloaded from which exhibit glomerular and tubular structural features akin to the mesangium and minimal diffuse glomerular loop staining; 2 = mild, those found in LN patients). Increased IgG2c levels, together with with PAS-positive material affecting ,10% of glomeruli; 3 = mild-to-moderate, C3/C3d, MASP-2, and MAC deposition in the glomeruli of with 10–25% of glomeruli; 4 = moderate-to-marked, with 25–50% of . C1q2/2TM mice suggest activation of the lectin complement glomeruli; and 5 = marked, with 50% of glomeruli. Other parameters were scored on a scale of 0–5, where 0 = absent, 1 = minimal (,5% glomerular or pathway in the absence of C1q. The unexpected discovery of MAC cortical area affected), 2 = mild (5–20%), 3 = mild-to-moderate (21–33%), 2/2 deposition in C3 TM mice introduces a novel C3-independent 4 = moderate (34–65%), and 5 = marked (.65%). Representative images mechanism of complement activation in low C3 states in the kidney. were taken using NIS-Elements BR 3.2 64-bit. Transmission electron http://www.jimmunol.org/ Using a model of IC-mediated nephritis, we uncover that thrombin microscopy was performed using whole kidney that was immersion-fixed in formalin. After removing the capsule, small samples of cortex were contributes to complement activation and kidney damage in postfixed in 2% osmium tetroxide and processed by standard procedures. the absence of C3. Enhanced fibrin deposition in the kidneys of Thin sections (0.1 mm) were stained with uranyl acetate and lead citrate, 2 2 2 2 C3 / TM but not C1q / TM mice reveals a role for thrombin then examined with a Jeol JEM-1230 electron microscope. Normal mouse activation in low C3 states. kidney was obtained in a separate experiment, fixed in half-strength Karnovsky Fixative (2% paraformaldehyde, 2.5% glutaraldehyde in sodium cacodylate buffer), then processed, stained, and examined in a similar Materials and Methods fashion. Mice and disease models Immunofluorescence staining by guest on October 2, 2021 Mfge82/2 mice on B6 background were kindly provided by Dr. Nagata 2/2 2/2 Frozen spleen sections (6 mm) were fixed in acetone and blocked in TBS (10). C1q mice were a gift from Dr. M. Botto (30). C3 mice (31) with 5% BSA and 0.05% Tween 20. Tissues were stained with peanut were originally from The Jackson Laboratory and were kindly provided by agglutinin and Alexa Fluor 488 anti-mouse CD169 Ab (BioLegend). Dr. J. Heinecke (University of Washington). B6.Sle1 mice were from Dr. L. Frozen kidney sections (5 mm) were fixed in cold acetone for 10 min and Morel (32). All mice were crossed onto the C57BL/6 background for at blocked (PBS/1% BSA/5% goat serum) for 30 min at room temperature. least seven generations. Twelve murine strains were examined across two Mouse IgG deposition in the kidney was detected with FITC-conjugated different disease models: 1) spontaneous disease (B6.Sle1, Sle1.Mfge82/2, 2/2 2/2 2/2 2/2 2/2 goat anti-mouse IgG Ab or Alexa 555–conjugated goat anti-mouse IgG2c Sle1.C1q , Sle1.C3 , Sle1.Mfge8 C1q [C1q TM] and Ab (Invitrogen). Sheep IgG deposition was evaluated using FITC-conjugated Sle1.Mfge82/2C32/2 [C32/2TM]) and 2) accelerated nephrotoxic 2/2 2/2 2/2 2/2 goat anti-sheep IgG Ab (Jackson ImmunoResearch). Deposition of com- nephritis (NTN) (Mfge8 , C3 , and Mfge8 C3 , on B6 back- plement component C3 in the kidneys was detected with FITC-conjugated ground). Mice subjected to accelerated NTN were immunized with sheep goat anti-mouse C3 Ab (1:100; CEDARLANE), whereas C3d staining was IgG(1mgs.c.;JacksonImmunoResearch)combinedwithFreund’s performed using rabbit anti-mouse C3d IgG (1:100; R&D systems) followed adjuvant (Sigma-Aldrich), and 5 d later they were challenged intra- by Cy5-conjugated goat anti-rabbit IgG (1:1000; Jackson ImmunoResearch). vascularly with one dose of nephrotoxic serum (NTS) or normal sheep Deposition of complement component C4 in the kidneys was detected with serum (75 ml–5 mg/ml). The preimmunization protocol generates anti- FITC-conjugated goat anti-mouse C4 Ab (1:100; CEDARLANE). Fibrin or sheep IgG Abs that bind to sheep IgG from NTS that has fixed on glo- fibrinogen staining was performed using FITC-conjugated goat anti-mouse merular Ags in the kidney. Proteinuria was quantified by Bradford assay fibrin/fibrinogen [abbreviated fibrin(ogen)] IgG (1:100; OriGene). Fibrin (Bio-Rad Laboratories) and urine albumin/creatinine ratio (UACR) was was distinguished from fibrinogen using biotin-labeled 59D8 anti-fibrin evaluated using Albuwell Albumin Assay and its Creatinine Companion IgG (15 mg/ml) provided by Dr. Esmon (Oklahoma Medical Research Kit (Exocell). Survival of different mouse strains in the spontaneous model Foundation); kidney sections were fixed in 4% paraformaldehyde, free was monitored over 12 mo, and data were presented by Kaplan–Meier aldehyde groups were blocked with 0.1 M glycine/PBS, and tissues were curve. All animal experiments were approved by the Institutional Animal blocked with avidin and biotin-blocking buffers (Vector Laboratories); Care and Use Committee of the University of Washington, Seattle. fibrin staining was detected using Cy5-labeled streptavidin (1:200). MAC Autoantibody detection deposition was detected with rabbit anti-mouse C5b-9 IgG Ab (clone B7, 1:100), developed and provided by Dr. P. Morgan (36), followed by Alexa Serum samples from different mouse strains were collected every month 647–conjugated goat anti-rabbit IgG (1:1000; Jackson ImmunoResearch). after birth and stored in 280˚C before analysis. To detect anti-dsDNA IgG, Mannan-binding lectin-associated serine protease 2 (MASP-2) deposition calf thymus DNA (10 mg/ml) was dissolved in DNA Binding Solution was detected with rabbit anti–MASP-2 IgG (1 mg/ml; Abcam), followed by (Pierce) and was used to coat ELISA plates at 4˚C overnight. To detect Alexa Cy5-conjugated goat anti-rabbit IgG (1:1000; Jackson ImmunoResearch); anti-chromatin Abs, 10 mg/ml purified histone from calf thymus (Sigma- kidney tissue from MASP-2 knockout mice was used as negative control Aldrich) was added to the plate to form histone/DNA complexes [artificial (37). For indirect immunofluorescence (IF) staining, nonspecific binding to chromatin (33)]. Pooled serum samples from 7-mo-old NZB/W F1 female Abs deposited in the kidney was prevented by preincubating both primary mice were used as a standard for anti-dsDNA and anti-chromatin IgG and secondary Abs with 5 mg/ml purified mouse IgG as well as with 5 mg/ml detection assays. For both assays, serum samples (1:100 in PBS/10% FBS) sheep IgG for kidneys from mice exposed to accelerated NTN. All IF were incubated overnight at 4˚C, and IgG binding was detected using staining was imaged using the digital fluorescence microscope EVOS FL, The Journal of Immunology 3 and fluorescence intensity and colocalization were quantified by Image (interstitial scores) (Fig. 1A, 1B, Supplemental Fig. 1A). Whereas J. AC in the spleen and kidney tissues were detected using Roche In Situ complement deficiency exacerbated nephritis in Sle1.Mfge82/2 Cell Death Detection Kit, Fluorescein (Sigma-Aldrich). The total number + mice, lack of MFGE8 was sufficient to trigger changes in the of TUNEL cells per 50 randomly selected glomeruli was determined using Image J. kidney pathology in Sle1 mice (Fig. 1A, 1B). However, only the TM strains (9–11 mo) spontaneously developed proteinuria Plasma thrombin activity and kidney tissue factor expression (Fig. 1C). Despite severe kidney pathology in both TM strains, 2 2 2 2 Thrombin activity in the plasma of C32/2TM, C1q2/2TM, and B6 mice was the C1q / but not the C3 / TM mice demonstrated reduced measured as previously described (38) using S-2238 (H-D-phenylalanyl-L- survival at 12 mo of age (Supplemental Fig. 1B). The reasons for pipecolyl-L-argininep-nitroaniline dihydrochloride; Diapharma) as a sub- the difference in survival is not certain but may be related to the strate. At a 1:1 ratio, plasma (25 ml) was incubated with the thrombin substrate and the amount of p-nitroaniline dihydrochloride formed was broader anti-inflammatory effects of C1q (42). measured by absorbance at 405 nm. A standard curve was generated using IgG autoantibodies are central to the pathogenesis of glo- bovine thrombin (Sigma-Aldrich), and data were presented as units of merulonephritis in SLE patients as well as in lupus mouse models 2/2 2/2 thrombin. RNA was isolated from C3 TM, C1q TM, and B6 kidneys (43–45). To what extent kidney damage is instigated by circu- (QIAGEN RNAeasy MidiPrep Kit). Tissue factor (TF) mRNA levels were detected by quantitative PCR (SYBER Green; Invivogen) using the lating Ab:Ag (immune) complexes versus accumulation of dead following primers: 59-TCAAGCACGGGAAAGAAAAC-39 (forward) cell debris in the kidney (planted Ag) targeted by IgG autoanti- and 59-CTGCTTCCTGGGCTATTTTG-39 (reverse). bodies (46–48) is controversial. Because, in addition to MFGE8, early complement components facilitate AC clearance (27, 49), Statistical analyses we first asked whether there were differences in AC deposition in Statistical analyses were performed using GraphPad Prism 7. Normality of the test strains. Although both complement-deficient TM strains had Downloaded from the data were determined using a Shapiro–Wilk test, and statistical dif- a significantly greater number of TUNEL+ AC in the glomeruli ference between two sample means (specified in the figures and figure 2/2 legends) was determined using a Student t test for normally distributed relative to their Mfge8-sufficient DM controls (Sle1.C1q and 2/2 2/2 data sets and a Mann–Whitney U test for not normally distributed data sets. Sle1.C3 ), these levels were comparable to Sle1.Mfge8 mice Differences between three or more data groups were analyzed by multiple (Fig. 1D), indicating that reduced clearance of cell debris in the comparison one-way ANOVA with a Bonferroni post hoc test. Correlation kidney was not sufficient to account for the more severe nephritis strength between two parameters was determined by Pearson correlation analysis. Statistical significance was determined according to the following in the TM strains. In contrast, in the TM models of spontaneous http://www.jimmunol.org/ p values: *p , 0.05, **p , 0.01, and ***p , 0.001. The Heatmap function nephritis, renal IgG deposition was significantly higher in both 2 2 in Qlucore was used for unsupervised clustering of autoantigen array data. C1q- and C3-deficient TM mice compared with Sle1.Mfge8 / controls (Fig. 1E). Sle1.Mfge82/2 mice had comparable levels of IgG deposition with B6.Sle1 and the complement-deficient DM Results 2 2 2 2 strains (Sle1.C1q / and Sle1.C3 / ) (Fig. 1E). C1q- and C3-deficient Sle1.Mfge82/2 mice develop Electron microscopic analysis of representative kidney samples proliferative glomerulonephritis associated with glomerular 2 2 from Sle1.Mfge8 / demonstrated increased leukocyte infiltration AC and IgG deposition in the glomeruli of this strain compared with the normal control Deficiencies in the early complement components are among the

(Fig. 2A, arrows). Subendothelial electron-dense immune deposits by guest on October 2, 2021 strongest disease-susceptibility genes in SLE patients (3). Most were detected in the glomerular basement membranes of C32/2TM C1q (.90%) and 26% of C3-deficient (24) patients develop severe mice, resembling “wire loop” deposits seen in human LN (50) disease. However, single-complement deficiencies in mice do not (Fig. 2B). C1q2/2 TM kidneys demonstrated features characteristic lead to spontaneous disease, and ∼30–42% of C1q-deficient SLE of human membranoproliferative glomerulonephritis, including patients develop glomerulonephritis (21, 39), which indicates that mesangial immune deposits with focal extensions into the sub- other genes and/or defects in immunologic pathways contribute endothelial space of the capillary walls, duplication of basement to the development of and nephropathic effects in membrane matrices, and deposits interposed between duplicated SLE. Similarly, the presence of anti-chromatin Abs (B6.Sle1)or membrane matrices (Fig. 2C, first and second panels). Some deposits 2/2 impaired AC clearance (B6. Mfge8 ), both important abnor- showed a fibrillar-organized substructure, similar to that seen in pa- malities contributing to lupus pathogenesis, are insufficient to tients with LN (51) (Fig. 2C, third panel). Additionally, we observed drive spontaneous disease in B6 mice (19, 40, 41). Therefore, to severe damage of proximal tubules, characterized by cellular dis- determine whether and how complement deficiencies exacerbated solution and loss of interdigitating cell borders (Fig. 2C, fourth the SLE phenotype in a truly polygenic SLE-like environment panel). In sum, these substructural abnormalities further illustrate (i.e., in the presence of autoantibodies and impaired AC clearance the similarities between these novel lupus strains and human SLE. 2/2 [Sle1.Mfge8 ]), we first generated two double-mutant (DM) strains, 2/2 2/2 2/2 mice that lack C1q or C3 develop Sle1.C1q mice and Sle1.C3 mice. We then crossed these DM Sle1.Mfge8 strains to the Sle1.Mfge82/2 strain to create Sle1.Mfge82/2C1q2/2 marked Ag spread and Sle1.Mfge82/2C32/2 TM mice. Histologic analyses of the B6.Sle1 congenic mice spontaneously produce anti-chromatin but kidneys of both TM strains showed membranoproliferative glomer- only produce anti-DNA Abs when additional genetic intervals ulonephritis with PAS-positive deposits, glomerular hypercellularity, containing lupus susceptibility loci (sle2 and sle3) are crossed to and mesangial expansion, as well as multifocal-to-coalescing cellular B6.Sle1 (32). Similarly, many mouse strains deficient in opsonins, infiltrates in the kidney interstitium, which tended to be more severe such as MFGE8 (19) or C1q (52), fail to develop anti-dsDNA Abs in animals with more pronounced glomerular pathology (Fig. 1A, and overt clinical disease on the B6 genetic background, indi- Supplemental Fig. 1A). Significantly higher pathology scores cating that AC accumulation is insufficient to induce high-affinity were observed in Sle1.Mfge82/2 mice that lacked either C1q pathogenic anti-dsDNA Abs in these strains. Because we observed or C3 (C1q2/2 or C32/2 TM mice, respectively), compared enhanced renal IgG deposition in both TM strains (Fig. 1E), we with the DM controls (Fig. 1B). The higher pathology scores in next asked whether complement deficiency in the presence of TM strains were due to moderate-to-marked glomerular mesangial autoantibodies and AC accumulation alters the autoantibody expansion and glomerular hypercellularity (glomerular scores), response. Indeed, at 5 mo of age, both TM strains developed as well as marked-to-moderate tubulointerstitial inflammation significantly higher titers of anti-chromatin and anti-dsDNA 4 SURROGATE PATHWAYS OF COMPLEMENT ACTIVATION IN LUPUS MODELS Downloaded from http://www.jimmunol.org/

FIGURE 1. Complement deficiency in Sle1.Mfge82/2 mice leads to nephritis associated with AC accumulation and IgG deposition in the kidney. (A and B) Representative kidney pathology following PAS staining (original magnification 340). (A) The asterisk symbol (*) indicates glomerular hypercellularity and (B) shows pathology scores: cumulative (left panel, determined based on tubular, glomerular, mesangial, and interstitial protein deposition and cel- by guest on October 2, 2021 lularity), glomerular (middle panel, mesangial matrix, glomerular hypercellularity), and interstitial (right panel, interstitial inflammation)(n = 10 per group, 12 mo old). Tubular protein scores are not included in the glomerular and interstitial scores, as protein presence in the tubules may be derived from either glomerular or tubular injury (C) Proteinuria measurements in urine samples collected from 10–12-mo-old mice (n = 6–9 per group). (D) Levels of IgG deposition in the kidney tissues were determined by immunofluorescence staining (original magnification 340) and fluorescence intensity was quantified by ImageJ (n = 7–13 per group). (E) AC were detected by TUNEL staining of kidney sections (n = 4–6, 12-mo-old female mice per group; original mag- nification 340; scale bar, 100 mm). The number of TUNEL+ cells in the glomeruli (white dashed outline) was quantified by ImageJ. Data represent the mean 6 SEM. *p , 0.05, **p , 0.01, ***p , 0.001 by one-way ANOVA with Bonferroni post hoc test.

IgG compared with the DMs (Sle1.C1q2/2, Sle1.C32/2,and previously reported in murine models and SLE patients with Sle1.Mfge82/2; Fig. 3A, 3B). Only female 5-mo-old Sle1.Mfge82/2 LN (Fig. 3D) (53–55). Although many autoantibody specificities mice had increased anti-dsDNA Ab titers compared with B6.Sle1 were shared between both TM strains, C1q2/2TM mice developed controls (Fig. 3A, 3B). Elevated levels of IgG anti-dsDNA were strain-selective autoreactivity to Sm, SmD, ribosomal P protein, confirmed in 60% of C1q2/2TM and 80% of C32/2TM mice by hemocyanin, vimentin, and laminin (Fig. 3D, Supplemental Fig. 2A). C. luciliae (representative images in Fig. 3C). At the same age Autoreactivity in C32/2TM mice was selective for a-actinin, (9 mo), ∼30% of Sle1.Mfge82/2 mice developed anti-dsDNA collagens, aggrecan, and chondroitin sulfate (Fig. 3D, Supplemental autoantibodies. Fig. 2B). Surprisingly, anti-C1q Abs were detected in C1q2/2TM and To examine the effects of complement component deficiencies anti-C3 Abs were detected in C32/2TM mice (Fig. 3D). These could on the autoantibody repertoire in complement-deficient Sle1.Mfge82/2 be explained by low affinity cross-reactivities [e.g., anti-DNA Abs mice, we analyzed the specificities of mouse sera in an autoantigen in SLE patients cross-react with C1q (56) and/or polyclonal B cell microarray panel (53). As shown in Fig. 3D, the deletion of Mfge8 activation, for example, generation of anti-complement Abs fol- had a significant effect on broadening the autoantibody specificities lowing virus infection (57, 58)]. of either Sle1 or DM strains containing Sle1 and a complement The increase in autoantibody titers and repertoire in TM mice deficiency. Strikingly, the combination of Sle1 with Mfge82/2 and was accompanied by significant splenomegaly in 10–12-mo-old either C1q or C3 deficiencies in TM mice not only increased Ab mice compared with the DM strains (Supplemental Fig. 1C). titers and led to an earlier breaking of tolerance (5 mo) but also This increase was associated with an expansion in TUNEL+ AC enabled Ab production against numerous autoantigens found in in both the red pulp and the follicles of the spleen (Supplemental other autoimmune and/or rheumatic diseases (e.g., histones, col- Fig. 1D), the organ shown to play a prominent role in the lagens, aggrecan, a-actinin, cytochrome C, hemocyanin, vimentin, clearance of AC from the circulation (10). Together, these complement factors, SRP54, and SP100), many of which were findings suggest that both C1q and C3 play a role in suppressing The Journal of Immunology 5 Downloaded from

FIGURE 2. Transmission electron microscopy of kidney tissues demonstrating (A) representative glomeruli from normal (left panel) and Sle1.Mfge82/2 (right panel) mice, with the latter showing increased leukocyte infiltration (arrows). (B) Large, somewhat confluent subendothelial electron-dense immune deposits (arrows), resembling wire loop deposits seen in human LN, in the glomerular basement membranes of C32/2 TM mice. (C) glomerular and tubular damage in C1q2/2 TM mice: first panel, glomerular mesangial immune deposits (arrows), which segmentally extend into adjacent capillary walls (double http://www.jimmunol.org/ arrows) and which show deposits interposed between duplicated membrane matrices, characteristic of a membranoproliferative pattern of injury; second panel, high-power view of area demarcated within the box in the leftmost panel showing electron-dense immune deposits in a subendothelial location in the glomerular capillary wall between duplicated basement membrane matrices (double arrows); third panel, fibrillar-organized substructure within some deposits, similar to that seen in patients with LN (arrow); fourth panel, acute tubular injury with dissolution of cell cytoplasm and loss of interdigitating cell borders. The abundance of mitochondria indicates these cells are proximal tubular cells. No immune deposits are found in tubular basement membranes or in the surrounding interstitium (inset scale bar, 500 nm).

autoantibody responses to AC, and their absence enables Ag Immunofluorescence staining revealed deposition of MASP-2 in by guest on October 2, 2021 spread. the tubules and glomeruli of C1q2/2TM kidneys (Fig. 4D). MASP-2 colocalized with C3 (Fig. 4D), indicating complement activation via Nephritis in C1q-deficient TM mice is associated with IgG2c the lectin pathway (LP). Because we did not detect C4 in C1q2/2TM deposition and lectin pathway complement activation kidneys (data not shown), LP activation occurred independently of Although IC injury in the kidney is thought to be initiated by IgG2c C4, as previously reported in the ischemia-reperfusion model of fixing C1q and triggering classical complement pathway activation kidney injury (60). Some tubular and glomerular MASP-2 depo- (59), whether and how complement activation contributes to ne- sition was also observed in Sle1.C1q2/2 mice, however, to a phritis in the absence of C1q is not known. We first examined the significantly lower degree than in C1q2/2TM kidneys and with IgG subclass deposited in the kidney. Although no difference was minimal colocalization with C3 (Supplemental Fig. 3). observed in the intensity of renal IgG3 staining between strains (data not shown), IgG2c was detected in both the glomeruli and Nephritis in the absence of C3 is associated with the tubular interstitium of C1q2/2TM mice at significantly higher thrombin-mediated complement activation levels compared with the DM controls and the C32/2TM strain Whereas the absence of C3 in the C32/2TM strain would suggest (Fig. 4A). To determine whether a complement-mediated pathway that nephritis in these mice is complement independent, surpris- contributed to nephritis in the absence of C1q, we tested whether ingly, we detected MAC deposition in the glomeruli of C32/2TM the MAC was present in the kidneys of C1q2/2TM mice. Of mice and at a significantly higher fluorescence intensity than in considerable interest, we detected abundant staining of MAC in the Sle1.Mfge82/2DM controls (Fig. 5A, 5B), indicating that a the kidneys of C1q2/2TM mice, thus confirming terminal com- C5 convertase had been generated. Because a handful of studies plement pathway activation in the absence of C1q (Fig. 4B). Glo- have identified thrombin as a surrogate C5 convertase (38, 61–63), merular MAC expression was significantly higher in the C1q2/2TM we looked for evidence of thrombin activation in the kidneys of compared with the Sle1.Mfge82/2 DM strain (Fig. 4B), suggesting C32/2TM mice. Indeed, immunofluorescence staining revealed that complement activation contributes to the kidney injury seen in fibrin(ogen) deposition in C32/2TM glomeruli (Fig. 5C). Because C1q2/2TM mice. Consistent with activation of complement pathway deposition of fibrinogen could be a relatively nonspecific finding despite absence of C1q, we detected abundant C3 accumulation in in tissue damage, we used an mAb, 59D8, that distinguishes fi- the glomeruli and tubules of C1q2/2TM mice, with tubular C3 brinogen from fibrin (the cleavage product of thrombin) as it binds levels significantly greater compared with Sle1.Mfge82/2 controls to the N terminus of cross-linked fibrin b-chain, a site not exposed (Fig. 4C). Moreover, C1q2/2TM kidneys demonstrated enhanced in fibrinogen (64–66). Positive staining for fibrin was observed tubular and glomerular deposition of C3 breakdown product only in C32/2TM but not in Sle.Mfge82/2 DM nor in C1q2/2TM C3d relative to Sle1.Mfge82/2 controls, which indicates activa- mice (Fig. 5D). Moreover, we detected a significantly higher tion of the alternative or lectin complement pathways (Fig. 4C). number of glomeruli staining positive for fibrin in C3-deficient 6 SURROGATE PATHWAYS OF COMPLEMENT ACTIVATION IN LUPUS MODELS Downloaded from http://www.jimmunol.org/ by guest on October 2, 2021

FIGURE 3. C1q or C3 complement deficiency in MFGE8-deficient Sle1 mice results in increased autoantibody titers and Ag spread. (A and B) The titers of anti-chromatin (A) and anti-dsDNA IgG (B) in 5-mo-old male and female mice of indicated genotypes were determined by ELISA (n = 20–30 per group). Data in (A) and (B) represent the mean 6 SEM. *p , 0.05, **p , 0.01, ***p , 0.001 by one-way ANOVAwith multiple comparison and Bonferroni post hoc test. (C) Serum samples from 9-mo-old female mice were tested for high-affinity IgG binding to dsDNA by C. lucilea staining. Representative images of negative and positive staining are shown. (D) Pooled serum from 12-mo-old female mice from different groups (n = 15–20 per group) were tested on autoantigen array. The heatmap of IgG binding is shown. Red asterisk (*) denotes targets previously reported as renal Ags in murine and human LN (54, 55, 88–90). The Journal of Immunology 7 Downloaded from http://www.jimmunol.org/

FIGURE 4. IgG2c deposition and complement activation in C1q-deficient TM mice. (A) Representative staining (left panel) and summary data of IgG2c deposition in the kidneys by immunofluorescence (original magnification 340; scale bar, 100 mm) (n = 3–12, 12-mo-old female mice per group). (B)MAC deposition was detected by IF (original magnification 340) in the glomeruli of C1q2/2TM mice; Fluorescence Intensity was compared with kidney tissues of Sle1.Mfge82/2 mice (n = 9). (C) C3 and C3d deposition was detected in glomeruli and tubules of C1q2/2 TM mice by IF (original magnification 320; scale bar, 200 mm); Fluorescence intensity in both glomeruli and tubules was compared with kidney tissues of Sle1.Mfge82/2 mice (n = 5–7). (D) MASP-2 by guest on October 2, 2021 and C3 deposition in two representative C1q2/2TM mouse kidneys (original magnification 340) was detected by IF in tubules (arrows) and glomeruli (dashed circles). MASP-2 knockout (KO) mouse kidney tissue was used as negative control (right panel). Colocalization of MASP-2 (red) and C3 (green) is detected in tubules and glomeruli (yellow). Data represent the mean 6 SEM. *p , 0.05, **p , 0.01 by (A) Mann–Whitney nonparametric U test or (B and C) Student t test.

TM mice compared with Sle1.Mfge82/2 and C1q2/2TM controls C32/2TM mice, we detected greater glomerular deposition of (Fig. 5D), indicating that fibrin deposition was selective for in- both MAC and fibrin(ogen) in C3-deficient mice treated with flammation in the absence of C3. Significantly higher thrombin NTS (C32/2 and Mfge82/2C32/2) compared with B6 and Mfge82/2 activity in the plasma of C32/2TM mice compared with C1q2/2TM controls, respectively (Fig. 6C, 6D). The levels of renal MAC and controls (Fig. 5E) confirmed increased systemic activation of the fibrin(ogen) deposition in NTS-treated C32/2 mice were comparable coagulation pathway in the absence of C3. At the kidney level, we to those found in Mfge82/2C32/2 kidneys (Fig. 6C, 6D), indicating detected greater TF expression in C32/2TM relative to C1q2/2TM that increased thrombin activity associated with MAC assembly mice (Fig. 5E), which could play a role in thrombin activation in in Ab-mediated kidney injury arises specifically in the absence of the kidneys of the C3-deficient TM strain (67). C3, independent of the disease model and consistent with obser- Because our spontaneous model takes ∼9 mo for disease to vations in IC-mediated lung injury (38). However, worse kidney develop, we turned to a well-described short-term model of kidney disease in Mfge82/2C32/2 mice compared with C32/2 controls injury instigated by Ab binding to planted glomerular Ags, accelerated indicates that the increased AC burden because of MFGE8 NTN (68), to investigate whether thrombin-triggered complement deficiency contributed to additional inflammatory processes in activation contributes to kidney disease in the absence of C3 (Fig. 6A). NTS-mediated injury and increased UACR. Analogous to the findings in TM mice that developed worse ne- To verify that thrombin contributes to NTS-driven kidney injury phritis in the absence of C3, generation of NTN in C3-deficient in C3-deficient mice, we compared kidney function in the presence mice (C32/2 and Mfge82/2C32/2) led to worse kidney disease, or absence of the thrombin-specific inhibitor argatroban (69) characterized by an increase in the UACR, compared with their (Fig. 6E). Indeed, treatment of Mfge82/2C32/2 mice with arga- respective controls (B6 and Mfge82/2) (Fig. 6B). Kidney injury troban led to a decrease in NTS-triggered UACR compared with was independent of the glomerular deposition of sheep or mouse vehicle-treated controls (Fig. 6F). The improvement in kidney IgG or circulating Ab titers, as similar levels were detected in function in argatroban-treated mice was accompanied by a de- all NTS-treated strains (Supplemental Fig. 4A–D). Neither Ab crease in glomerular deposition of fibrin(ogen) (Fig. 6G) as well as deposition nor kidney disease were observed in normal serum– in diminished glomerular MAC deposition (Fig. 6H). Argatroban- treated controls (data not shown). Parallel to the findings in the mediated effects were independent of the levels of sheep or mouse 8 SURROGATE PATHWAYS OF COMPLEMENT ACTIVATION IN LUPUS MODELS Downloaded from http://www.jimmunol.org/ by guest on October 2, 2021

FIGURE 5. Glomerular MAC and fibrin deposition in C32/2 TM mice. (A) MAC deposition was detected by IF in the glomeruli of C32/2TM mice. Left panel shows representative staining and the right panel shows fluorescence intensity (FI). (B) MAC staining in C32/2TM versus DM mice is shown; representative images on the left and FI summary data on the right (n = 9). (A and B) Original magnification 320; scale bar, 100 mm. (C) Fibrin(ogen) deposition was detected by IF in the glomeruli of C32/2TM mice (left panel). Fibrin deposition in C32/2TM kidneys was confirmed using fibrin-specific mAb 59D8 (middle panel). Total number of fibrin-positive glomeruli was compared between three groups of kidney tissues: Sle1.Mfge82/2 versus C32/2TM versus C1q2/2TM (n = 6). Original magnification 320; scale bar, 100 mm. (D) Thrombin activity in the plasma was compared between C32/2TM, C1q2/2TM, and B6 mice (n = 7–9) in a chromogenic assay. (E) TF mRNA expression levels were detected in the kidneys of C32/2TM, C1q2/2TM, and B6 mice (n = 7–8). Data are presented as mean 6 SEM. *p , 0.05, **p , 0.01, ***p , 0.001 by Mann–Whitney U test (A), Student t test (B) or one-way ANOVA with Bonferroni post hoc test (C–E).

IgG deposition in the kidney or circulating Ab levels, as similar SLE in humans. We have therefore generated mouse strains that titers were detected in drug- versus control-treated animals (data contain combinations of abnormalities in three common pathways not shown). Together, these findings strongly suggest that, in the affected in SLE: defective clearance of AC, low complement levels, absence of C3, Ab-mediated renal damage leads to thrombin and B cell dysfunction. Significantly, we observed that, whereas activation in the kidney and is responsible for generating a C5 anti-chromatin autoantibodies and defective clearance of AC convertase that subsequently triggers MAC formation and kidney (Sle1.Mfge82/2 mice on the B6 background) were not sufficient to damage. induce clinical nephritis, the addition of either C1q or C3 deficiency promoted spontaneous glomerulonephritis in Sle1.Mfge82/2 mice Discussion with “subclinical lupus” (19). Two major differences in the TM Significant limitations to current mouse models of lupus exist. compared with DM mice were identified (Fig. 7). First, serum In the spontaneous models, the genetic factors that contribute to autoantibodies in both C1q2/2 and C32/2 TM mice were higher, disease are largely unknown, rendering the ability to tease apart showed significantly greater Ag spread, and were associated with cause and effect a challenge. In addition, single-gene knockout increased glomerular IgG deposition compared with their matched models do not accurately reflect the complexity of polygenic Sle1.Mfge82/2 counterparts. Second, although Sle1.Mfge82/2 as The Journal of Immunology 9 Downloaded from http://www.jimmunol.org/ by guest on October 2, 2021

FIGURE 6. MAC and fibrin(ogen) deposition in the kidneys of C3-deficient mice in NTN is attenuated by thrombin inhibition. (A) Accelerated NTN was induced by priming with sheep IgG and a subsequent challenge with NTS or control normal serum (NS), according to the timeline shown. (B) UACR ratio in urine samples collected on the third day after NTS or NS challenge were examined. Results combined from two independent experiments (n = 3–6 per group). (C) Glomerular MAC deposition was examined by IF staining (original magnification 340) of tissues obtained on day 7 after the NTS or NS challenge (n = 4–6 per genotype per treatment). Representative images are shown, and fluorescence intensity (FI) was quantified by ImageJ and represented as fold induction relative to NS-treated controls. (D) Fibrin(ogen) deposition in the kidneys of NTS-treated mice was examined by IF, and number of positive glomeruli per tissue section was quantified (n = 4–6). Original magnification 320; scale bar, 100 mm. (B–D) Data (Figure legend continues) 10 SURROGATE PATHWAYS OF COMPLEMENT ACTIVATION IN LUPUS MODELS well as both TM strains had increased AC debris in the kidneys, associate with renal disease exacerbations in human SLE and to the TM mice showed significantly greater activation of the MAC possibly cross-react with other renal targets (83–85). The quality of the complement pathway. Examination of the mechanisms in- of the IgG Abs was also different, particularly in C1q2/2TM volved in generation of the MAC revealed that activation occurred mice [a greater deposition of IgG2c, the IgG subtype with potent through different pathways in C1q2/2 and C32/2 TM mice. In effector functions for both complement fixation and engagement C1q2/2TM mice, the alternative and/or LP were likely activated, of the FcgR on myeloid cells (86, 87)]. Regarding differences in Ag leading to generation of abundant glomerular and tubular C3d. targeting, many of the identified Ab specificities in the TM strains, When C3 was absent, thrombin activation was observed locally including anti–a actinin, anti-aggrecan, anti-phosphatidylinositol, and systemically, and functional inhibition of thrombin signifi- anti-proteoglycans, anti-collagens, anti-vimentin, anti-laminin, anti- cantly attenuated kidney injury in the NTN model, strongly sug- fibrinogen, and anti-hemocyanin autoantibodies, have been associ- gesting that thrombin acted as an alternative C5 convertase. Our ated with severe nephritis in both murine (54) and/or human studies novel disease model demonstrates that the three pathways com- (55, 88–90). Similar nephritogenic properties have been attrib- monly abnormal in SLE combine to promote disease by enhanc- uted to anti-vimentin autoantibodies, which were shown to target ing B cell function and, at the target tissue level, enabling multiple vimentin in situ and to associate with tubulointerstitial inflam- nonconventional pathways of terminal complement pathway mation in LN patients (88). In our model, anti-vimentin autoan- activation, which is relevant in human disease. tibodies were selective for C1q2/2 TM mice, which demonstrated The increased Ag spread in the absence of the complement C3d deposition similar to some SLE patients (91) and signifi- components C1q or C3 in addition to Sle1.Mfge82/2 (C1q2/2TM cant tubular damage by electron microscopy. Indeed, microarray and C32/2TM strains) was surprising because of the previously analysis of kidney tissues confirmed elevated levels of IgG tran- described indispensable role that complement and, particularly, C3 scripts in situ in C1q2/2TM mice, similar to what was observed Downloaded from plays in foreign Ag processing and lowering the threshold of the in human tubular interstitial inflammation LN (88). Therefore, in B cell immune response (70–73). Instead, our findings suggest addition to anti-DNA Abs, autoantibodies reactive to local kidney that C1q and C3 prevent or modulate production of Abs to self- Ags likely played a role in the nephropathic effect in the TM mice antigens arising from AC (Fig. 7A). Complement proteins lacking either C1q or C3. could protect against Ag spread by several mechanisms. First, Besides the consequences of complement component deficien- we found significantly greater levels of AC in the spleens of cies on B cell tolerance in the presence of excess AC, the significant http://www.jimmunol.org/ C1q2/2TM and C32/2TM mice compared with Sle1.Mfge82/2 increase in nephritis in the TM strains could result from the cu- controls, which supports the important role that both C1q and mulative effects that MFGE8 and C1q/C3 deficiencies have on AC C3 play in the clearance of AC (27, 42, 49). Therefore, one pos- and IC clearance. However, AC accumulation was similar in sibility is that absence of complement enables a greater quantity and Sle1.Mfge82/2 and the TM mice, indicating that AC accumulation persistence of self-antigens, which leads to necrosis and activation in the kidney alone was not sufficient to account for the nephritis of phagocytes and adaptive immune cells (74). C1q has also been observed in the TM mice. These findings suggested that the early shown to promote a tolerogenic state by modulating both the den- complement components play a role in preventing nephritis. This dritic cells and macrophages toward anti-inflammatory functions could occur through a variety of mechanisms that include direct by guest on October 2, 2021 (75, 76) as well as by enhancing presentation of self-antigens to engagement by C1q of inhibitory receptors, such as LAIR-1 and immature B cells, thus promoting negative selection by BCR CD33 on immune cells (92–95), as well as C3-mediated damp- editing, anergy, or death (77–79). Similarly, C3 bound to self-antigens ening of innate immunity (27, 96–98). Unexpectedly, we observed from dying cells likely facilitates immune tolerance through swift that absence of C1q or C3 allowed surrogate mechanisms to ac- clearance and enhancement of anti-inflammatory cytokines by tivate complement and lead to the assembly of the MAC. The phagocytes (27, 80). Baudino et al. (81) have proposed that C3 MAC not only mediates tissue (kidney) injury through perforin modulates intracellular trafficking and processing of opsonized lysis but also by activation of inflammatory signaling pathways and phagocytosed AC Ags, thus controlling the immune responses (99–101). Besides MAC, terminal complement activation gener- to self-antigens. Together with excess dying cells in splenic ates another potent inflammatory mediator, C5a, which could follicles, the absence of C1q- and C3-mediated tolerogenic contribute to kidney injury. In fact, C5a/C5aR signaling has been mechanisms might be contributing to Ag spread in C1q2/2TM found instrumental in mediating nephritis in Mrllpr mice by driv- and C32/2TM mice. Finally, lack of C1q and C3 could influ- ing recruitment of inflammatory cells and production of Th1 cy- ence the nature and availability of B cell epitopes. For example, tokines (102). Moreover, direct engagement of C5Ra on kidney C1q appears to be required to drive C1r/C1s toward proteolytic endothelial cells could trigger cellular proliferation and cytokine activity of nucleolar Ags (82), and C3 deficiency could lead to production, further exacerbating the kidney pathology (103, 104). aberrant proteolysis by thrombin or other proteases, yielding In C1q2/2TM mice, abundant deposition of C3 and its break- peptides that are more immunogenic. Future studies will address down product, C3d, in the glomeruli and tubules was associated these critical questions. with MASP-2 deposits, indicating that C3 was activated by the Because we demonstrated an expansion in the autoantibody lectin complement pathway. Because C3d was detected in the response in TM mice relative to their DM controls, the development tubules as well as glomeruli, the role of local tubular C3 pro- of nephritis in these strains could be attributed to the quality or duction (105, 106) and activation in the generation of inter- quantity of these autoantibodies. The higher titers of autoanti- stitial nephritis will be important to explore further. Moreover, bodies, particularly anti-DNA, in the two TM strains may well the significantly greater MASP-2 deposition in C1q2/2TM contribute to the disease, as these autoantibodies are known to compared with Sle1.C1q2/2 mice, supports the activation of

represent the mean 6 SEM. *p , 0.05, **p , 0.01 by one-way ANOVA. (E) Mfge82/2C32/2 mice were treated with argatroban or vehicle throughout the course of accelerated NTN treatment, according to the outlined timeline. (F) UACR was measured in urine collected on day 3 after the NTS challenge (n = 7–8 per treatment, two independent experiments). (G) Fibrin(ogen) and (H) MAC deposition were examined by IF staining (original magnification 340) (n = 6–8 per treatment, two independent experiments). Data in (F–H) represent the mean 6 SEM. *p , 0.05, **p , 0.01 by Student t test. The Journal of Immunology 11 Downloaded from http://www.jimmunol.org/ by guest on October 2, 2021

FIGURE 7. Proposed models of kidney injury in low complement states in the presence of autoantibodies and increased AC burden. (A) Low C1q and C3 levels in Sle1.Mfge82/2 mice lead to increased AC accumulation in the spleen which, in the presence of anti-chromatin autoantibodies (Sle1 interval), results in marked Ag spread and increased Ab deposition in the kidney. (B) In mice with autoantibody spread and defective clearance of AC, low C1q and C3 states allow alternative mechanisms of MAC assembly and complement activation in the kidneys. Where C1q is lacking (left side), C3 cleavage and MAC assembly likely occur through the alternative or LP. In the absence of all the three conventional complement pathways in low C3 states (right side),an alternative convertase mediated by thrombin enables MAC assembly and tissue injury. Image created via BioRender. mannose-binding lectin in the presence of increased tissue generation of the MAC in C3-deficient kidneys was confirmed by damage (107, 108). functional attenuation of this protease using the thrombin-specific Whereas C1q insufficiency is strongly associated with SLE inhibitor argatroban in the NTN model. Together, these findings development, homozygous deficiency in C3, the convergence point reveal a novel mechanism of kidney injury in lupus-prone mice of all three complement pathways, has a lower penetrance (∼26%) with low C3. Although, to our knowledge, this is the first study to (24). In practice, low levels of C3 resulting from disease activity identify a role for thrombin in terminal complement activation or the presence of autoantibodies to C3 (∼30% of SLE patients) in the kidney in C3 deficiency, our findings are consistent with (109) are encountered frequently in active SLE and are associated previous reports of increased expression and enzymatic activity with more severe LN, often characterized by increased glo- of thrombin leading to surrogate C5 convertase activity in the merular MAC deposition (110–112). Greater MAC accumula- liver and blood of C3-deficient mice (38, 63) (Fig. 7B). What tioninthekidneysofC32/2TM mice, compared with their triggers the compensatory mechanism in the absence of C3 is matched Sle1.Mfge82/2 counterparts, raised the question as to unknown. Interestingly, we detected increased expression of TF, a how MAC is assembled in the kidneys in the absence of C3. known initiator of the extrinsic coagulation pathway, in C32/2TM Because thrombin was previously identified as an alternative but not C1q2/2 TM kidneys. The elevated expression of TF in the C5 convertase in IC-mediated lung injury in C3-deficient mice absence of C3 was also reported in C32/2 mice with IC-mediated (38), we sought evidence of thrombin activation in C32/2TM lung injury (38) as well as in a model of allograft rejection in mice kidneys. Detection of fibrin in the glomeruli of C32/2TM but not lacking C3 (113). This latter study demonstrated that excessive C1q2/2TM mice implicated thrombin as a surrogate C5 convertase production of C5a in the absence of C3 led to increased activation responsible for MAC generation. A direct role for thrombin in the of TF in endothelial cells and subsequent thrombin generation. 12 SURROGATE PATHWAYS OF COMPLEMENT ACTIVATION IN LUPUS MODELS

It is plausible that tissue damage in the kidney, possibly because models, in Ab-mediated arthritis it was reported either that acti- of Ab deposition, coupled with lack of opsonization by C3 drives vation of both the complement pathway and FcgR were necessary TF expression in C32/2TM mice. Alternatively, IgG-dependent (127) or that inflammation was independent of FcgR and C3, yet activation of thrombin in autoantibody-mediated arthritis has been dependent on the downstream complement activation (C5) and reported in C32/2-deficient mice (63) and could also play a role. thrombin generation (63, 127). Complement and coagulation cascades coevolved as necessary homeostatic pathways rich in serine proteases, and numerous in- Acknowledgments teractions between these pathways have been documented (5, 114). We thank Drs. Paul Morgan and Charles Esmon for providing the MAC Studies published decades ago revealed that C5 could be cleaved and the fibrin Abs, respectively, as well as Drs. Dudler and Yaseen from by proteases in the coagulation pathway, including thrombin and Omeros (Seattle, WA) for providing MASP-2 knockout mouse tissues. We plasmin (61), a result confirmed in vivo in models of IC-mediated also thank Dr. Christian Lood for insightful discussion. lung disease and Ab-mediated arthritis (38, 63). Detailed molec- ular studies revealed that during TF-induced clotting of plasma, Disclosures thrombin efficiently cleaved C5 at the highly conserved R947 site The authors have no financial conflicts of interest. rather than the conventional site of cleavage by C5 convertase R751, generating the intermediates C5T and C5bT, which were References more effective than C5b in generating MAC (67). However, the 1. Ramos, P. S., A. M. Shedlock, and C. D. Langefeld. 2015. Genetics of efficiency of thrombin-mediated C5 cleavage in these in vitro autoimmune diseases: insights from population genetics. J. Hum. Genet. studies was lower than that of the conventional C5 convertase. 60: 657–664. Whereas this might not be the case in vivo, the modest reduction 2. Harley, I. T., K. M. Kaufman, C. D. Langefeld, J. B. Harley, and J. A. Kelly. Downloaded from 2009. Genetic susceptibility to SLE: new insights from fine mapping and in MAC deposition and kidney injury on thrombin inhibition in genome-wide association studies. Nat. Rev. Genet. 10: 285–290. our model does suggest that other proteases besides thrombin 3. Manderson, A. P., M. Botto, and M. J. Walport. 2004. The role of complement in the development of systemic lupus erythematosus. Annu. Rev. Immunol. 22: might act as surrogate C5 convertase in the absence of C3. These 431–456. could be other proteases in the coagulation pathway, such as 4. Nagata, S., R. Hanayama, and K. Kawane. 2010. Autoimmunity and the factors Xa, IXa, and Xia (5), in the fibrinolysis pathway, such clearance of dead cells. Cell 140: 619–630.

5. Amara, U., M. A. Flierl, D. Rittirsch, A. Klos, H. Chen, B. Acker, http://www.jimmunol.org/ as plasmin (115), or independent of the coagulation cascade, U. B. Bru¨ckner, B. Nilsson, F. Gebhard, J. D. Lambris, and M. Huber-Lang. such as macrophage and proteases (116, 117). 2010. Molecular intercommunication between the complement and coagulation Of relevance to LN, studies published more than 20 y ago systems. J. Immunol. 185: 5628–5636. 6. Colonna, L., C. Lood, and K. B. Elkon. 2014. Beyond apoptosis in lupus. Curr. reported increased activation of the coagulation pathway in the Opin. Rheumatol. 26: 459–466. kidneys of SLE patients with active disease (118, 119) and ef- 7. Mohan, C., and C. Putterman. 2015. Genetics and pathogenesis of systemic lupus erythematosus and lupus nephritis. Nat. Rev. Nephrol. 11: 329–341. ficacy of fibrinolytics on attenuating kidney disease in SLE 8. Almaani, S., A. Meara, and B. H. Rovin. 2017. Update on lupus nephritis. Clin. (120). Recently, elevated levels of coagulation cascade proteins J. Am. Soc. Nephrol. 12: 825–835. in urine strongly predicted renal disease in patients with LN 9. Bennett, M., and H. I. Brunner. 2013. Biomarkers and updates on pediatrics lupus nephritis. Rheum. Dis. Clin. North Am. 39: 833–853. (121). Interestingly, TF was one of the proteins that correlated

10. Hanayama, R., M. Tanaka, K. Miyasaka, K. Aozasa, M. Koike, Y. Uchiyama, by guest on October 2, 2021 with clinical parameters of SLE (121), suggesting the role for and S. Nagata. 2004. Autoimmune disease and impaired uptake of apoptotic extrinsic activation of the coagulation pathway in the LN kidney. cells in MFG-E8-deficient mice. Science 304: 1147–1150. 11. Gaipl, U. S., A. Kuhn, A. Sheriff, L. E. Munoz, S. Franz, R. E. Voll, J. R. Kalden, Recent systematic integrated (genomic, proteomic, and metabolomic) and M. Herrmann. 2006. Clearance of apoptotic cells in human SLE. Curr. Dir. analysis has shown that cleavage of complement components asso- Autoimmun. 9: 173–187. 12. Georgescu, L., R. K. Vakkalanka, K. B. Elkon, and M. K. Crow. 1997. ciated with activation of the coagulation cascade in SLE patients, Interleukin-10 promotes activation-induced cell death of SLE lymphocytes likely because of enhanced inflammatory cytokine production (122). mediated by Fas ligand. J. Clin. Invest. 100: 2622–2633. Although the relationship between low C3 levels and thrombin 13. Herrmann, M., R. E. Voll, O. M. Zoller, M. Hagenhofer, B. B. Ponner, and J. R. Kalden. 1998. Impaired phagocytosis of apoptotic cell material by activation has not been studied at the tissue level in LN, terminal monocyte-derived macrophages from patients with systemic lupus erythematosus. complement activation in the kidney has been reported in a single Arthritis Rheum. 41: 1241–1250. C3-deficient SLE patient (123), implicating alternative pathways 14. Baumann, I., W. Kolowos, R. E. Voll, B. Manger, U. Gaipl, W. L. Neuhuber, T. Kirchner, J. R. Kalden, and M. Herrmann. 2002. Impaired uptake of apo- of C5 cleavage in the absence of C3. These clinical associations, ptotic cells into tingible body macrophages in germinal centers of patients coupled with our findings, encourage further investigation of with systemic lupus erythematosus. Arthritis Rheum. 46: 191–201. 15. Gordon, S., and A. Plu¨ddemann. 2018. Macrophage clearance of apoptotic thrombin activation and therapeutic intervention in SLE patients cells: a critical assessment. Front. Immunol. 9: 127. with low C3 levels. 16. Hanayama, R., M. Tanaka, K. Miwa, A. Shinohara, A. Iwamatsu, and In summary, we propose the following model for LN in our novel S. Nagata. 2002. Identification of a factor that links apoptotic cells to phago- cytes. Nature 417: 182–187. polygenic mice constructed to resemble pathway abnormalities in 17. Yamaguchi, H., T. Fujimoto, S. Nakamura, K. Ohmura, T. Mimori, F. Matsuda, human SLE. In the presence of increased self-antigens and hy- and S. Nagata. 2010. Aberrant splicing of the milk fat globule-EGF factor 8 perresponsive B cells (124), lack of early complement components (MFG-E8) gene in human systemic lupus erythematosus. Eur. J. Immunol. 40: 1778–1785. increases autoantibody production and Ag spread. Some of these 18. Hu, C. Y., C. S. Wu, H. F. Tsai, S. K. Chang, W. I. Tsai, and P. N. Hsu. 2009. autoantibodies (e.g., anti-DNA, anti-a actinin, and anti-vimentin) Genetic polymorphism in milk fat globule-EGF factor 8 (MFG-E8) is associ- ated with systemic lupus erythematosus in human. Lupus 18: 676–681. (45, 85) deposit in the kidney, causing cell injury and augmenting 19. Peng, Y.-F., and K. B. Elkon. 2011. Autoimmunity in MFG-E8-deficient mice is the damage-associated molecular pattern molecules arising from associated with altered trafficking and enhanced cross-presentation of apoptotic defective clearance of cell debris. These damage-associated mo- cell antigens. [Published erratum appears in 2012 J. Clin. Invest. 122: 782.] J. Clin. Invest. 121: 2221–2241. lecular pattern molecules, which may include mannose-binding 20. Morgan, B. P. 1999. Regulation of the complement membrane attack pathway. lectin, collectins, and ficolins and may also induce TF expression Crit. Rev. Immunol. 19: 173–198. (125, 126), activate downstream complement pathways (lectin and 21. Macedo, A. C., and L. Isaac. 2016. Systemic lupus erythematosus and deficiencies of early components of the complement classical pathway. Front. Immunol. 7: 55. alternative), as well as surrogate pathways such as thrombin (refer 22. Nowling, T. K., and G. S. Gilkeson. 2011. Mechanisms of tissue injury in lupus to Fig. 7). Assembly of the MAC leads to necrosis and also nephritis. Arthritis Res. Ther. 13: 250. 23. Yin, Y., X. Wu, G. Shan, and X. Zhang. 2012. Diagnostic value of serum anti- increases inflammatory responses in the kidney. Whereas FcgR C1q antibodies in patients with lupus nephritis: a meta-analysis. Lupus 21: engagement on myeloid cells may play a contributory role in these 1088–1097. The Journal of Immunology 13

24. Online Mendelian Inheritance in Man. 2018. C3 deficincy. Johns Hopkins 52. Mitchell, D. A., M. C. Pickering, J. Warren, L. Fossati-Jimack, J. Cortes- University, Baltimore, MD. Available at: https://www.omim.org/entry/613779. Hernandez, H. T. Cook, M. Botto, and M. J. Walport. 2002. C1q deficiency and Accessed: November 20, 2019. autoimmunity: the effects of genetic background on disease expression. 25. Thurman, J. M. 2015. Complement in kidney disease: core curriculum 2015. J. Immunol. 168: 2538–2543. Am. J. Kidney Dis. 65: 156–168. 53. Zhu, H., H. Luo, M. Yan, X. Zuo, and Q.-Z. Li. 2015. Autoantigen microarray 26. Ho, A., S. G. Barr, L. S. Magder, and M. Petri. 2001. A decrease in complement for high-throughput autoantibody profiling in systemic lupus erythematosus. is associated with increased renal and hematologic activity in patients with Genomics Proteomics Bioinformatics 13: 210–218. systemic lupus erythematosus. Arthritis Rheum. 44: 2350–2357. 54. Li, Q. Z., C. Xie, T. Wu, M. Mackay, C. Aranow, C. Putterman, and C. Mohan. 27. Mevorach, D., J. O. Mascarenhas, D. Gershov, and K. B. Elkon. 1998. 2005. Identification of autoantibody clusters that best predict lupus disease Complement-dependent clearance of apoptotic cells by human macrophages. activity using glomerular proteome arrays. [Published erratum appears in 2006 J. Exp. Med. 188: 2313–2320. J. Clin. Invest. 116: 548.] J. Clin. Invest. 115: 3428–3439. 28. Botto, M., M. Kirschfink, P. Macor, M. C. Pickering, R. Wu¨rzner, and 55. Haddon, D. J., V. K. Diep, J. V. Price, C. Limb, P. J. Utz, and I. Balboni. 2015. F. Tedesco. 2009. Complement in human diseases: lessons from complement Autoantigen microarrays reveal autoantibodies associated with proliferative deficiencies. Mol. Immunol. 46: 2774–2783. nephritis and active disease in pediatric systemic lupus erythematosus. Arthritis 29. Quigg, R. J., A. Lim, M. Haas, J. J. Alexander, C. He, and M. C. Carroll. 1998. Res. Ther. 17: 162. Immune complex glomerulonephritis in C4- and C3-deficient mice. Kidney Int. 56. Franchin, G., M. Son, S. J. Kim, I. Ben-Zvi, J. Zhang, and B. Diamond. 2013. 53: 320–330. Anti-DNA antibodies cross-react with C1q. J. Autoimmun. 44: 34–39. 30. Botto, M., C. Dell’Agnola, A. E. Bygrave, E. M. Thompson, H. T. Cook, 57. Csorba, K., L. A. Schirmbeck, E. Tuncer, C. Ribi, P. Roux-Lombard, F. Petry, M. Loos, P. P. Pandolfi, and M. J. Walport. 1998. Homozygous C1q C. Chizzolini, U. Huynh-Do, D. Vanhecke, and M. Trendelenburg. 2019. Anti- deficiency causes glomerulonephritis associated with multiple apoptotic bodies. C1q antibodies as occurring in systemic lupus erythematosus could be induced Nat. Genet. 19: 56–59. by an epstein-barr virus-derived antigenic site. Front. Immunol. 10: 2619. 31. Prodeus, A. P., X. Zhou, M. Maurer, S. J. Galli, and M. C. Carroll. 1997. 58. Koma, T., V. Veljkovic, D. E. Anderson, L.-F. Wang, S. L. Rossi, C. Shan, P.- Impaired -dependent natural immunity in complement C3-deficient Y. Shi, D. W. Beasley, N. Bukreyeva, J. N. Smith, et al. 2018. Zika virus in- mice. Nature 390: 172–175. fection elicits auto-antibodies to C1q. Sci. Rep. 8: 1882. 32. Morel, L., B. P. Croker, K. R. Blenman, C. Mohan, G. Huang, G. Gilkeson, and 59. Bruhns, P. 2012. Properties of mouse and human IgG receptors and their E. K. Wakeland. 2000. Genetic reconstitution of systemic lupus erythematosus contribution to disease models. Blood 119: 5640–5649. Downloaded from immunopathology with polycongenic murine strains. Proc. Natl. Acad. Sci. 60. Asgari, E., C. A. Farrar, N. Lynch, Y. M. Ali, S. Roscher, C. Stover, W. Zhou, USA 97: 6670–6675. W. J. Schwaeble, and S. H. Sacks. 2014. Mannan-binding lectin-associated 33. Mohan, C., E. Alas, L. Morel, P. Yang, and E. K. Wakeland. 1998. Genetic serine protease 2 is critical for the development of renal ischemia reperfusion dissection of SLE pathogenesis. Sle1 on murine chromosome 1 leads to a se- injury and mediates tissue injury in the absence of complement C4. FASEB J. lective loss of tolerance to H2A/H2B/DNA subnucleosomes. J. Clin. Invest. 28: 3996–4003. 101: 1362–1372. 61. Wetsel, R. A., and W. P. Kolb. 1983. Expression of C5a-like biological ac- 34. Wang, B., Y. Yamamoto, N. S. El-Badri, and R. A. Good. 1999. Effective tivities by the fifth component of human complement (C5) upon limited di- treatment of autoimmune disease and progressive renal disease by mixed bone- gestion with noncomplement enzymes without release of polypeptide fragments. marrow transplantation that establishes a stable mixed chimerism in BXSB J. Exp. Med. 157: 2029–2048. http://www.jimmunol.org/ recipient mice. Proc. Natl. Acad. Sci. USA 96: 3012–3016. 62. Kalowski, S., E. L. Howes, Jr., W. Margaretten, and D. G. McKay. 1975. Effects 35. Miyazaki, T., M. Ono, W. M. Qu, M. C. Zhang, S. Mori, S. Nakatsuru, of intravascular clotting on the activation of the : the role of Y. Nakamura, T. Sawasaki, Y. Endo, and M. Nose. 2005. Implication of allelic the platelet. Am. J. Pathol. 78: 525–536. polymorphism of osteopontin in the development of lupus nephritis in MRL/lpr 63. Auger, J. L., S. Haasken, and B. A. Binstadt. 2012. Autoantibody-mediated mice. Eur. J. Immunol. 35: 1510–1520. arthritis in the absence of C3 and activating Fcg receptors: C5 is activated by 36. Goutaudier, V., H. Perrochia, S. Mucha, M. Bonnet, S. Delmas, F. Garo, the coagulation cascade. Arthritis Res. Ther. 14: R269. V. Garrigue, S. Lepreux, V. Pernin, J.-E. Serre, et al. 2019. C5b9 deposition in 64. Hui, K. Y., E. Haber, and G. R. Matsueda. 1983. Monoclonal antibodies to a glomerular capillaries is associated with poor kidney allograft survival in synthetic fibrin-like peptide bind to human fibrin but not fibrinogen. Science antibody-mediated rejection. Front. Immunol. 10: 235. 222: 1129–1132. 37. Schwaeble, W. J., N. J. Lynch, J. E. Clark, M. Marber, N. J. Samani, Y. M. Ali, 65. Chen, F., E. Haber, and G. R. Matsueda. 1992. Availability of the B beta(15-21) T. Dudler, B. Parent, K. Lhotta, R. Wallis, et al. 2011. Targeting of mannan- epitope on cross-linked human fibrin and its plasmic degradation products. binding lectin-associated serine protease-2 confers protection from myocardial Thromb. Haemost. 67: 335–340. by guest on October 2, 2021 and gastrointestinal ischemia/reperfusion injury. Proc. Natl. Acad. Sci. USA 66. Matsueda, G. R., and M. N. Margolies. 1986. Structural basis for the species 108: 7523–7528. selectivity of a fibrin-specific monoclonal antibody. Biochemistry 25: 1451– 38. Huber-Lang, M., J. V. Sarma, F. S. Zetoune, D. Rittirsch, T. A. Neff, 1455. S. R. McGuire, J. D. Lambris, R. L. Warner, M. A. Flierl, L. M. Hoesel, et al. 67. Krisinger, M. J., V. Goebeler, Z. Lu, S. C. Meixner, T. Myles, E. L. Pryzdial, 2006. Generation of C5a in the absence of C3: a new complement activation and E. M. Conway. 2012. Thrombin generates previously unidentified C5 pathway. Nat. Med. 12: 682–687. products that support the terminal complement activation pathway. Blood 120: 39. Walport, M. J. 2002. Complement and systemic lupus erythematosus. Arthritis 1717–1725. Res. 4(Suppl. 3): S279–S293. 68. Du, Y., Y. Fu, and C. Mohan. 2008. Experimental anti-GBM nephritis as an 40. Morel, L., C. Mohan, Y. Yu, B. P. Croker, N. Tian, A. Deng, and E. K. Wakeland. analytical tool for studying spontaneous lupus nephritis. Arch. Immunol. Ther. 1997. Functional dissection of systemic lupus erythematosus using congenic Exp. (Warsz.) 56: 31–40. mouse strains. J. Immunol. 158: 6019–6028. 69. Kassel, K. M., B. P. Sullivan, W. Cui, B. L. Copple, and J. P. Luyendyk. 2012. 41. Miyanishi, M., K. Tada, M. Koike, Y. Uchiyama, T. Kitamura, and S. Nagata. 2007. Therapeutic administration of the direct thrombin inhibitor argatroban reduces Identification of Tim4 as a phosphatidylserine . Nature 450: 435–439. hepatic inflammation in mice with established fatty liver disease. Am. J. Pathol. 42. Carroll, M. C. 2004. A protective role for innate immunity in systemic lupus 181: 1287–1295. erythematosus. Nat. Rev. Immunol. 4: 825–831. 70. Carrasco, Y. R., and F. D. Batista. 2007. B cells acquire particulate antigen in a 43. Koffler, D., V. Agnello, R. Thoburn, and H. G. Kunkel. 1971. Systemic lupus macrophage-rich area at the boundary between the follicle and the subcapsular erythematosus: prototype of immune complex nephritis in man. J. Exp. Med. sinus of the lymph node. Immunity 27: 160–171. 134: 169s–179s. 71. Phan, T. G., I. Grigorova, T. Okada, and J. G. Cyster. 2007. Subcapsular en- 44. Mannik, M., C. E. Merrill, L. D. Stamps, and M. H. Wener. 2003. Multiple counter and complement-dependent transport of immune complexes by lymph autoantibodies form the glomerular immune deposits in patients with systemic node B cells. Nat. Immunol. 8: 992–1000. lupus erythematosus. J. Rheumatol. 30: 1495–1504. 72. Phan, T. G., J. A. Green, E. E. Gray, Y. Xu, and J. G. Cyster. 2009. Immune 45. Davidson, A. 2016. What is damaging the kidney in lupus nephritis? Nat. Rev. complex relay by subcapsular sinus macrophages and noncognate B cells drives Rheumatol. 12: 143–153. antibody affinity maturation. Nat. Immunol. 10: 786–793. 46. Wagrowska-Danilewicz, M., and M. Danilewicz. 1998. Apoptosis in lupus 73. Carroll, M. C., and D. E. Isenman. 2012. Regulation of humoral immunity by SLE-N IV and non-lupus mesangiocapillary glomerulonephritis type I MCGN. complement. Immunity 37: 199–207. I. A comparative study. J. Nephrol. 11: 44–49. 74. Elkon, K. B. 2018. Review: cell death, nucleic acids, and immunity: inflam- 47. Kalaaji, M., E. Mortensen, L. Jørgensen, R. Olsen, and O. P. Rekvig. 2006. mation beyond the grave. Arthritis Rheumatol. 70: 805–816. Nephritogenic lupus antibodies recognize glomerular basement membrane- 75. Hosszu, K. K., F. Santiago-Schwarz, E. I. Peerschke, and B. Ghebrehiwet. associated chromatin fragments released from apoptotic intraglomerular cells. 2010. Evidence that a C1q/C1qR system regulates monocyte-derived dendritic Am. J. Pathol. 168: 1779–1792. cell differentiation at the interface of innate and acquired immunity. Innate 48. Fenton, K. 2015. The effect of cell death in the initiation of lupus nephritis. Immun. 16: 115–127. Clin. Exp. Immunol. 179: 11–16. 76. Hosszu, K. K., A. Valentino, Y. Ji, M. Matkovic, L. Pednekar, N. Rehage, 49. Walport, M. J., K. A. Davies, and M. Botto. 1998. C1q and systemic lupus N. Tumma, E. I. Peerschke, and B. Ghebrehiwet. 2012. Cell surface expression erythematosus. Immunobiology 199: 265–285. and function of the macromolecular on the surface of human 50. Weening, J. J., V. D. D’Agati, M. M. Schwartz, S. V. Seshan, C. E. Alpers, monocytes. Front. Immunol. 3: 38. G. B. Appel, J. E. Balow, J. A. Bruijn, T. Cook, F. Ferrario, et al. 2004. The 77. Prodeus, A. P., S. Goerg, L. M. Shen, O. O. Pozdnyakova, L. Chu, E. M. Alicot, classification of glomerulonephritis in systemic lupus erythematosus revisited. C. C. Goodnow, and M. C. Carroll. 1998. A critical role for complement in J. Am. Soc. Nephrol. 15: 241–250. maintenance of self-tolerance. Immunity 9: 721–731. 51. Alpers, C. E., and J. Kowalewska. 2008. Fibrillary glomerulonephritis and 78. Ferry, H., P. K. Potter, T. L. Crockford, A. Nijnik, M. R. Ehrenstein, immunotactoid glomerulopathy. J. Am. Soc. Nephrol. 19: 34–37. M. J. Walport, M. Botto, and R. J. Cornall. 2007. Increased positive selection of 14 SURROGATE PATHWAYS OF COMPLEMENT ACTIVATION IN LUPUS MODELS

B1 cells and reduced B cell tolerance to intracellular antigens in c1q-deficient 103. Braun, M., and A. E. Davis, III. 1998. Cultured human glomerular mesangial mice. J. Immunol. 178: 2916–2922. cells express the . Kidney Int. 54: 1542–1549. 79. Chatterjee, P., A. F. Agyemang, M. B. Alimzhanov, S. Degn, S. A. Tsiftsoglou, 104. Park, K. W., M. Tofukuji, C. Metais, M. E. Comunale, H. B. Dai, M. Simons, E. Alicot, S. A. Jones, M. Ma, and M. C. Carroll. 2013. Complement C4 G. L. Stahl, A. Agah, and F. W. Sellke. 1999. Attenuation of endothelium- maintains peripheral B-cell tolerance in a myeloid cell dependent manner. Eur. dependent dilation of pig pulmonary arterioles after cardiopulmonary bypass is J. Immunol. 43: 2441–2450. prevented by monoclonal antibody to complement C5a. Anesth. Analg. 89: 42–48. 80. Fossati-Jimack, L., G. S. Ling, A. Cortini, M. Szajna, T. H. Malik, 105. Sheerin, N. S., P. Risley, K. Abe, Z. Tang, W. Wong, T. Lin, and S. H. Sacks. J. U. McDonald, M. C. Pickering, H. T. Cook, P. R. Taylor, and M. Botto. 2013. 2008. Synthesis of complement protein C3 in the kidney is an important me- Phagocytosis is the main CR3-mediated function affected by the lupus- diator of local tissue injury. FASEB J. 22: 1065–1072. associated variant of CD11b in human myeloid cells. PLoS One 8: e57082. 106. Li, K., S. H. Sacks, and W. Zhou. 2007. The relative importance of local and 81. Baudino, L., A. Sardini, M. M. Ruseva, L. Fossati-Jimack, H. T. Cook, D. Scott, systemic complement production in ischaemia, transplantation and other pa- E. Simpson, and M. Botto. 2014. C3 opsonization regulates endocytic handling thologies. Mol. Immunol. 44: 3866–3874. of apoptotic cells resulting in enhanced T-cell responses to cargo-derived 107. Nauta, A. J., N. Raaschou-Jensen, A. Roos, M. R. Daha, H. O. Madsen, M. C. Borrias- antigens. Proc.Natl.Acad.Sci.USA111: 1503–1508. Essers,L.P.Ryder,C.Koch,andP.Garred.2003. Mannose-binding lectin engagement 82. Cai, Y., S. Y. K. Wee, J. Chen, B. H. D. Teo, Y. L. C. Ng, K. P. Leong, and J. Lu. with late apoptotic and necrotic cells. Eur. J. Immunol. 33: 2853–2863. 2017. Broad susceptibility of nucleolar proteins and autoantigens to comple- 108. Kravitz, M. S., and Y. Shoenfeld. 2006. Autoimmunity to protective molecules: ment C1 protease degradation. J. Immunol. 199: 3981–3990. is it the perpetuum mobile (vicious cycle) of autoimmune rheumatic diseases? 83. de Leeuw, K., L. Bungener, C. Roozendaal, H. Bootsma, and C. A. Stegeman. Nat. Clin. Pract. Rheumatol. 2: 481–490. 2017. Auto-antibodies to double-stranded DNA as biomarker in systemic lupus 109. Vasilev, V. V., R. Noe, M.-A. Dragon-Durey, S. Chauvet, V. J. Lazarov, erythematosus: comparison of different assays during quiescent and active B. P. Deliyska, V. Fremeaux-Bacchi, J. D. Dimitrov, and L. T. Roumenina. disease. Rheumatology (Oxford) 56: 698–703. 2015. Functional characterization of autoantibodies against complement com- 84. LJP 394 Investigator Consortium. 2005. Relationship between anti-double-stranded ponent C3 in patients with lupus nephritis. J. Biol. Chem. 290: 25343–25355. DNA antibodies and exacerbation of renal disease in patients with systemic 110. Pickering, M. C., and M. J. Walport. 2000. Links between complement abnor- lupus erythematosus. Arthritis Rheum. 52: 1129–1137. 85. Yung, S., and T. M. Chan. 2015. Mechanisms of kidney injury in lupus nephritis malities and systemic lupus erythematosus. Rheumatology (Oxford) 39: 133–141. - the role of anti-dsDNA antibodies. Front. Immunol. 6: 475. 111. Walport, M. J., K. A. Davies, M. Botto, M. A. Naughton, D. A. Isenberg, 86. Oishi, K., N. L. Koles, G. Guelde, and M. Pollack. 1992. Antibacterial and D. Biasi, R. J. Powell, N. T. Cheung, and G. R. Struthers. 1994. C3 nephritic Downloaded from protective properties of monoclonal antibodies reactive with Escherichia coli factor and SLE: report of four cases and review of the literature. QJM 87: 609–615. O111:B4 lipopolysaccharide: relation to antibody isotype and complement- 112. Wang, S., M. Wu, L. Chiriboga, B. Zeck, and H. M. Belmont. 2018. Membrane fixing activity. J. Infect. Dis. 165: 34–45. attack complex (mac) deposition in lupus nephritis is associated with hyper- 87. Nimmerjahn, F., P. Bruhns, K. Horiuchi, and J. V. Ravetch. 2005. Fcgam- tension and poor clinical response to treatment. Semin. Arthritis Rheum. 48: maRIV: a novel FcR with distinct IgG subclass specificity. Immunity 23: 41–51. 256–262. 88. Kinloch,A.J.,A.Chang,K.Ko,C.J.Henry Dunand, S. Henderson, M. Maienschein- 113. Khan, M. A., C. Maasch, A. Vater, S. Klussmann, J. Morser, L. L. Leung, Cline, N. Kaverina, B. H. Rovin, M. Salgado Ferrer, D. Wolfgeher, et al. 2014. C. Atkinson, S. Tomlinson, P. S. Heeger, and M. R. Nicolls. 2013. Targeting

Vimentin is a dominant target of in situ humoral immunity in human lupus complement component 5a promotes vascular integrity and limits airway http://www.jimmunol.org/ tubulointerstitial nephritis. Arthritis Rheumatol. 66: 3359–3370. remodeling. Proc. Natl. Acad. Sci. USA 110: 6061–6066. 89. Silvarin˜o, R., O. Noboa, and R. Cervera. 2014. Anti-glomerular basement 114. Oikonomopoulou, K., D. Ricklin, P. A. Ward, and J. D. Lambris. 2012. Inter- membrane antibodies. Isr. Med. Assoc. J. 16: 727–732. actions between coagulation and complement--their role in inflammation. 90. Seret, G., F. Can˜as, L. Pougnet-Di Costanzo, C. Hanrotel-Saliou, S. Jousse- Semin. Immunopathol. 34: 151–165. Joulin, Y. Le Meur, A. Saraux, A. Valeri, C. Putterman, P. Youinou, et al. 2015. 115. Foley, J. H., B. L. Walton, M. M. Aleman, A. M. O’Byrne, V. Lei, M. Harrasser, Anti-alpha-actinin antibodies are part of the anti-cell membrane antibody spectrum K. A. Foley, A. S. Wolberg, and E. M. Conway. 2016. Complement activation that characterize patients with lupus nephritis. J. Autoimmun. 61: 54–61. in arterial and venous thrombosis is mediated by plasmin. EBioMedicine 5: 91. Wang, H., J. Xu, X. Zhang, Y. L. Ren, M. Cheng, Z. L. Guo, J. C. Zhang, 175–182. H. Cheng, G. L. Xing, S. X. Wang, et al. 2018. Tubular basement membrane 116. Huber-Lang, M., E. M. Younkin, J. V. Sarma, N. Riedemann, S. R. McGuire, immune complex deposition is associated with activity and progression of lupus K. T. Lu, R. Kunkel, J. G. Younger, F. S. Zetoune, and P. A. Ward. 2002. nephritis: a large multicenter Chinese study. Lupus 27: 545–555. Generation of C5a by phagocytic cells. Am. J. Pathol. 161: 1849–1859.

92. Kouser, L., S. P. Madhukaran, A. Shastri, A. Saraon, J. Ferluga, M. Al-Mozaini, 117. Ward, P. A., and J. H. Hill. 1970. C5 chemotactic fragments produced by an by guest on October 2, 2021 and U. Kishore. 2015. Emerging and novel functions of complement protein enzyme in lysosomal granules of . J. Immunol. 104: 535–543. C1q. Front. Immunol. 6: 317. 118. Miranda, J. M., R. Garcia-Torres, L. J. Jara, F. Medina, H. Cervera, and 93. Son, M., B. Diamond, B. T. Volpe, C. B. Aranow, M. C. Mackay, and A. Fraga. 1994. Renal biopsy in systemic lupus erythematosus: significance of F. Santiago-Schwarz. 2017. Evidence for C1q-mediated crosslinking of CD33/ glomerular thrombosis. Analysis of 108 cases. Lupus 3: 25–29. LAIR-1 inhibitory immunoreceptors and biological control of CD33/LAIR-1 119. Pollak, V. E. 1995. Glomerular thrombosis predicts progression of glomeru- expression. Sci. Rep. 7: 270. lonephritis: can we prevent progression? Am. J. Kidney Dis. 26: 535–540. 94. Son, M., F. Santiago-Schwarz, Y. Al-Abed, and B. Diamond. 2012. C1q limits 120. Kant, K. S., V. E. Pollak, A. Dosekun, P. Glas-Greenwalt, M. A. Weiss, and dendritic cell differentiation and activation by engaging LAIR-1. Proc. Natl. H. I. Glueck. 1985. Lupus nephritis with thrombosis and abnormal fibrinolysis: Acad. Sci. USA 109: E3160–E3167. effect of ancrod. J. Lab. Clin. Med. 105: 77–88. 95. Son, M., B. Diamond, and F. Santiago-Schwarz. 2015. Fundamental role of C1q 121. Qin, L., S. Stanley, H. Ding, T. Zhang, V. T. T. Truong, T. Celhar, A.- in autoimmunity and inflammation. Immunol. Res. 63: 101–106. M. Fairhurst, C. Pedroza, M. Petri, R. Saxena, and C. Mohan. 2019. Urinary 96. Quartier, P., P. K. Potter, M. R. Ehrenstein, M. J. Walport, and M. Botto. 2005. pro-thrombotic, anti-thrombotic, and fibrinolytic molecules as biomarkers of Predominant role of IgM-dependent activation of the classical pathway in the lupus nephritis. [Published erratum appears in 2019 Arthritis Res. Ther. 21: clearance of dying cells by murine bone marrow-derived macrophages in vitro. 185.] Arthritis Res. Ther. 21: 176. Eur. J. Immunol. 35: 252–260. 122. Liang, Y., S.-B. Xie, C.-H. Wu, Y. Hu, Q. Zhang, S. Li, Y.-G. Fan, R.-X. Leng, 97. Amarilyo, G., I. Verbovetski, M. Atallah, A. Grau, G. Wiser, O. Gil, Y. Ben- H.-F. Pan, H.-B. Xiong, and D.-Q. Ye. 2017. Coagulation cascade and com- Neriah, and D. Mevorach. 2010. iC3b-opsonized apoptotic cells mediate a distinct anti-inflammatory response and transcriptional NF-kappaB-dependent plement system in systemic lupus erythematosus. Oncotarget 9: 14862–14881. blockade. Eur. J. Immunol. 40: 699–709. 123. Kitamura, H., H. Nishimukai, Y. Sano, and K. Nagaki. 1984. Study on C3-like 98. Reed, J. H., M. Jain, K. Lee, E. R. Kandimalla, M. H. Faridi, J. P. Buyon, factor in the serum of a C3-deficient subject. Immunology 51: 239–245. V. Gupta, and R. M. Clancy. 2013. 3 influences toll-like 124. Kumar, K. R., L. Li, M. Yan, M. Bhaskarabhatla, A. B. Mobley, C. Nguyen, receptor 7/8-dependent inflammation: implications for autoimmune diseases J. M. Mooney, J. D. Schatzle, E. K. Wakeland, and C. Mohan. 2006. Regulation characterized by antibody reactivity to ribonucleoproteins. J. Biol. Chem. of B cell tolerance by the lupus susceptibility gene Ly108. Science 312: 1665– 288: 9077–9083. 1669. 99. Cole, D. S., and B. P. Morgan. 2003. Beyond lysis: how complement influences 125. Yang, A., F. Chen, C. He, J. Zhou, Y. Lu, J. Dai, R. B. Birge, and Y. Wu. 2017. cell fate. Clin. Sci. (Lond.) 104: 455–466. The procoagulant activity of apoptotic cells is mediated by interaction with 100. Mead, R. J., S. K. Singhrao, J. W. Neal, H. Lassmann, and B. P. Morgan. 2002. factor XII. Front. Immunol. 8: 1188. The membrane attack complex of complement causes severe demyelination 126. Greeno, E. W., R. R. Bach, and C. F. Moldow. 1996. Apoptosis is associated associated with acute axonal injury. J. Immunol. 168: 458–465. with increased cell surface tissue factor procoagulant activity. Lab. Invest. 75: 101. Morgan, B. P. 2016. The membrane attack complex as an inflammatory trigger. 281–289. Immunobiology 221: 747–751. 127. Ji, H., K. Ohmura, U. Mahmood, D. M. Lee, F. M. Hofhuis, S. A. Boackle, 102. Wenderfer, S. E., B. Ke, T. J. Hollmann, R. A. Wetsel, H. Y. Lan, and K. Takahashi, V. M. Holers, M. Walport, C. Gerard, et al. 2002. Arthritis M. C. Braun. 2005. C5a receptor deficiency attenuates T cell function and renal critically dependent on innate immune system players. Immunity 16: 157– disease in MRLlpr mice. J. Am. Soc. Nephrol. 16: 3572–3582. 168.