C5 nephritic factors drive the biological phenotype of C3 glomerulopathies
2017; Elsevier BV; Volume: 92; Issue: 5 Linguagem: Inglês
10.1016/j.kint.2017.04.017
ISSN1523-1755
AutoresMaria-Chiara Marinozzi, Sophie Chauvet, Moglie Le Quintrec, Morgane Mignotet, Florent Petitprez, Christophe Legendre, Mathilde Cailliez, Georges Deschênes, Michel Fischbach, Alexandre Karras, François Nobili, Christine Piètrement, Marie‐Agnès Dragon‐Durey, Fádi Fakhouri, Lubka T. Roumenina, Véronique Frémeaux‐Bacchi,
Tópico(s)Phagocytosis and Immune Regulation
ResumoC3 Glomerulopathies, which include Dense Deposit Disease and C3 Glomerulonephritis, are associated with genetic and acquired dysregulation of the C3 convertase alternative pathway of complement. The potential role of the activation of the C5 convertase has not been studied extensively. Here we analyzed IgG samples from patients with C3 Glomerulopathies to identify circulating autoantibodies that stabilize the C3 alternative pathway (C3 Nephritic Factors) as well as C5 convertases (C5 Nephritic Factors), thus preventing decay of these enzyme complexes. Rare variants in alternative pathway genes were found in 28 of 120 tested patients. C3 and C5 Nephritic Factors were found in 76 of 101 (75%) and 29 of 59 (49%) of the patients, respectively. Therefore, we compared the results of the assays for the C3 and C5 nephritic factors functional activity: 29% were positive for C3 Nephritic Factors alone, 39% were positive for both C3 and C5 Nephritic Factors, and 10% were positive for C5 Nephritic Factors alone. We found that the addition of properdin-enhanced stabilization of C3 convertase in the presence of IgG doubly positive for both Nephritic Factors, while it did not modify the stabilization mediated by IgG solely positive for C3 Nephritic Factors. Both C3 and C5 Nephritic Factors correlated with C3 consumption, while only C5 Nephritic Factors correlated with sC5b9 levels. C5 Nephritic Factors–positive patients were more likely to have C3 Glomerulonephritis than Dense Deposit Disease. Thus, dysregulation of the C5 convertase contributes to C3 Glomerulopathies inter-disease differences and may have direct therapeutic implications. C3 Glomerulopathies, which include Dense Deposit Disease and C3 Glomerulonephritis, are associated with genetic and acquired dysregulation of the C3 convertase alternative pathway of complement. The potential role of the activation of the C5 convertase has not been studied extensively. Here we analyzed IgG samples from patients with C3 Glomerulopathies to identify circulating autoantibodies that stabilize the C3 alternative pathway (C3 Nephritic Factors) as well as C5 convertases (C5 Nephritic Factors), thus preventing decay of these enzyme complexes. Rare variants in alternative pathway genes were found in 28 of 120 tested patients. C3 and C5 Nephritic Factors were found in 76 of 101 (75%) and 29 of 59 (49%) of the patients, respectively. Therefore, we compared the results of the assays for the C3 and C5 nephritic factors functional activity: 29% were positive for C3 Nephritic Factors alone, 39% were positive for both C3 and C5 Nephritic Factors, and 10% were positive for C5 Nephritic Factors alone. We found that the addition of properdin-enhanced stabilization of C3 convertase in the presence of IgG doubly positive for both Nephritic Factors, while it did not modify the stabilization mediated by IgG solely positive for C3 Nephritic Factors. Both C3 and C5 Nephritic Factors correlated with C3 consumption, while only C5 Nephritic Factors correlated with sC5b9 levels. C5 Nephritic Factors–positive patients were more likely to have C3 Glomerulonephritis than Dense Deposit Disease. Thus, dysregulation of the C5 convertase contributes to C3 Glomerulopathies inter-disease differences and may have direct therapeutic implications. The C3 glomerulopathies (C3G) are chronic renal diseases characterized by glomerular lesions with predominant complement component C3 fragment deposition.1Pickering M.C. D'Agati V.D. Nester C.M. et al.C3 glomerulopathy: consensus report.Kidney Int. 2013; 84: 1079-1089Abstract Full Text Full Text PDF PubMed Scopus (399) Google Scholar, 2Cook H.T. Pickering M.C. Histopathology of MPGN and C3 glomerulopathies.Nat Rev Nephrol. 2015; 11: 14-22Crossref PubMed Scopus (74) Google Scholar The 2 major subgroups of C3G, dense deposit disease (DDD) (also known as membranoproliferative glomerulonephritis type 2) and C3 glomerulonephritis (C3GN), are caused by dysregulation of the alternative pathway (AP) of the complement system due to genetic and acquired factors.3Sethi S. Fervenza F.C. Membranoproliferative glomerulonephritis–a new look at an old entity.N Engl J Med. 2012; 366: 1119-1131Crossref PubMed Scopus (347) Google Scholar Although mutations have been identified in several complement genes, in >50% of the cases, the disease is linked to the presence of AP C3 convertase autoantibodies, the C3 nephritic factors (C3NeFs).4Servais A. Noel L.H. Roumenina L.T. et al.Acquired and genetic complement abnormalities play a critical role in dense deposit disease and other C3 glomerulopathies.Kidney Int. 2012; 82: 454-464Abstract Full Text Full Text PDF PubMed Scopus (371) Google Scholar The AP C3 convertase is formed after the binding of factor B (FB) to C3b in an Mg2+-dependent manner, followed by cleavage of FB by factor D (FD).5Merle N.S. Church S.E. Fremeaux-Bacchi V. et al.Complement System Part I - Molecular Mechanisms of Activation and Regulation.Front Immunol. 2015; 6: 262Crossref PubMed Scopus (833) Google Scholar The enzyme C3bBb has a very limited half-life. Properdin binds and stabilizes the C3bBb convertase (C3bBbP) and facilitates the switch of the C3 convertase to the C5 convertase.6Lesher A.M. Nilsson B. Song W.C. Properdin in complement activation and tissue injury.Mol Immunol. 2013; 56: 191-198Crossref PubMed Scopus (55) Google Scholar, 7Berends E.T. Gorham Jr., R.D. Ruyken M. et al.Molecular insights into the surface-specific arrangement of complement C5 convertase enzymes.BMC Biol. 2015; 13: 93Crossref PubMed Scopus (37) Google Scholar C3NeFs are IgGs that increase the AP activity and exert a stabilizing effect on both the fluid phase and the cell-bound C3 convertases.8Daha M.R. Fearon D.T. Austen K.F. C3 nephritic factor (C3NeF): stabilization of fluid phase and cell-bound alternative pathway convertase.J Immunol. 1976; 116: 1-7PubMed Google Scholar The complement biomarkers associated with C3NeFs are heterogeneous. In some patients, C3NeF is associated with C3 consumption in the fluid phase but does not always enhance C5 consumption and terminal pathway activation.9Mollnes T.E. Ng Y.C. Peters D.K. et al.Effect of nephritic factor on C3 and on the terminal pathway of complement in vivo and in vitro.Clin Exp Immunol. 1986; 65: 73-79PubMed Google Scholar, 10Zhang Y. Nester C.M. Martin B. et al.Defining the complement biomarker profile of C3 glomerulopathy.Clin J Am Soc Nephrol. 2014; 9: 1876-1882Crossref PubMed Scopus (60) Google Scholar, 11Corvillo F. Garcia-Morato M.B. Nozal P. et al.Serum properdin consumption as a biomarker of C5 convertase dysregulation in C3 glomerulopathy.Clin Exp Immunol. 2016; 184: 118-125Crossref PubMed Scopus (19) Google Scholar Despite published cases of the association of properdin-independent C3NeF with DDD and properdin-dependent C3NeF with MPGN type I,12Varade W.S. Forristal J. West C.D. Patterns of complement activation in idiopathic membranoproliferative glomerulonephritis, types I, II, and III.Am J Kidney Dis. 1990; 16: 196-206Abstract Full Text PDF PubMed Scopus (52) Google Scholar, 13Tanuma Y. Ohi H. Hatano M. Two types of C3 nephritic factor: properdin-dependent C3NeF and properdin-independent C3NeF.Clin Immunol Immunopathol. 1990; 56: 226-238Crossref PubMed Scopus (67) Google Scholar, 14Clardy C.W. Forristal J. Strife C.F. et al.A properdin dependent nephritic factor slowly activating C3, C5, and C9 in membranoproliferative glomerulonephritis, types I and III.Clin Immunol Immunopathol. 1989; 50: 333-347Crossref PubMed Scopus (42) Google Scholar the relationship between properdin, C3NeF, and C3G remains unclear.15Lesher A.M. Zhou L. Kimura Y. et al.Combination of factor H mutation and properdin deficiency causes severe C3 glomerulonephritis.J Am Soc Nephrol. 2013; 24: 53-65Crossref PubMed Scopus (68) Google Scholar, 16Ruseva M.M. Vernon K.A. Lesher A.M. et al.Loss of properdin exacerbates C3 glomerulopathy resulting from factor H deficiency.J Am Soc Nephrol. 2013; 24: 43-52Crossref PubMed Scopus (59) Google Scholar Numerous studies have explored mechanisms of action of C3NeF on the C3bBb, but few have investigated the reactivity of patient IgG to C3bBb and C3bBbP.17Paixao-Cavalcante D. Lopez-Trascasa M. Skattum L. et al.Sensitive and specific assays for C3 nephritic factors clarify mechanisms underlying complement dysregulation.Kidney Int. 2012; 82: 1084-1092Abstract Full Text Full Text PDF PubMed Scopus (71) Google Scholar, 18Zhang Y. Meyer N.C. Wang K. et al.Causes of alternative pathway dysregulation in dense deposit disease.Clin J Am Soc Nephrol. 2012; 7: 265-274Crossref PubMed Scopus (154) Google Scholar In this retrospective study, we confirmed that C3G is an acquired rather than a hereditary disease. We identified for the first time patients with circulating autoantibodies stabilizing the C5 convertase, namely, C5NeF, and we showed a correlation between the capacity of patients’ IgG to stabilize the C5 convertase and the sC5b-9 levels in patient plasma. We analyzed the autoantibodies profile against the C3 and C5 convertases and showed that C5NeFs are strongly associated with C3GN. The clinical and biological data in the 127 patients with C3G are summarized in Table 1. The age at onset ranged from 1 month to 72 years, and 57 patients had a pediatric onset (44.9%). Nephrotic syndrome was the most common feature at diagnosis (55.1%). Sixty-three percent of patients were receiving conventional immunosuppressive treatment (corticosteroids, cyclophosphamide, azathioprine, mycophenolate mofetil). At last follow up, end-stage renal disease (ESRD) had developed in 52%. No significant differences were observed in the type of clinical presentation between patients with pediatric and adult onset, but the incidence of ESRD or death was significantly higher in patients with adult onset. At the time of the first laboratory investigation, 53 of 127 patients (41.7%) had a low C3 level and a normal C4, suggesting AP consumption, and 71 of 118 patients (60.2%) had an sC5b-9 level above the normal range, suggesting activation of the terminal complement pathway. The frequency of patients with 1 (N = 27) or 2 (N = 1) variants was 23.3% (28/120 tested patients). We identified 10 novel and 16 rare variants with a minor allele frequency <1% in 6 complement genes (CFH, CFI, MCP, C3, CFB, CFHR5). Variants were demonstrated as pathogenic in 50% of the cases (13/26). Details of the detected novel or rare variants are shown in minor allele frequency (Supplementary Table S1). The frequency of variants was not significantly different between the patients with pediatric (11/55) and adult (17/65) onset of the disease. Using the cell-bound C3bBb stabilization assay, 75% were positive for C3NeF (76/101) (Supplementary Figure S1A). The stabilization of the C3 convertase was IgG dose dependent (Supplementary Figure S1B/C).Table 1Clinical, biological, and genetic data according to pediatric or adult onsetClinical description and biological and genetic resultsAllPediatricAdultP ValueN (%)12757 (44.9)70 (55.1)Sex (M/F) (% M)72/55 (57)28/29 (49)44/26 (62)0.1504At diagnosis DDD (%)46 (36)aThe histologic characteristics have been reported in 85 patients4 (56 C3GN and 29 DDD patients).26 (57)20 (43)0.0633 C3GN (%)81 (64)aThe histologic characteristics have been reported in 85 patients4 (56 C3GN and 29 DDD patients).31 (38)50 (62)0.0633 Age (yr), mean ± SD23.2 ± 16.89.4 ± 4.334.3 ± 14.8 Nephrotic syndrome (%)59/117 (55)30/54 (55)29/63 (46)0.3556 Acute renal failure (<60 ml/min per 1.73 m2) (%)35/99 (35)16/46 (34.7)19/53 (35)1.0000 eGFR (ml/min per 1.73 m2)70.8 ± 33.2 (N = 98)74.2 ± 35.1 (N = 45)62.1 ± 30.7 (N = 53)0.0718 Immunosuppressive treatment (%)65/102 (63)35/47 (74)30/55 (54)0.0415 Follow-up, yr9.8 ± 8.88 ± 8.512 ± 8.80.0108 Time between diagnosis and screening, mo.124 ± 139138 ± 157113 ± 1240.3179At last follow-up Death505 Dialysis (%)66/127 (52)23/57 (40)43/70 (61)0.0211 Age at dialysis (yr), mean ± SD35.3 ± 16.423.7 ± 8.741.5 ± 16.2<0.0001 Duration of evolution until ESRD (yr) if ESRD, mean ± SD8 ± 7.712.4 ± 8.75.7 ± 6<0.0001Complement component assessment C3 (660–1250 mg/l)726 ± 343712 ± 358729 ± 3320.7822 Low C3 (<660 mg/l) (%)53/127 (41.7)24/57 (42)29/70 (41)1.0000 C4 (93–380 mg/l)232 ± 91228 ± 102238 ± 820.5413 Low C4 (<90 mg/l) (%)3/127 (2.4)3/57 (5.3)0/70 (0)0.0878 Low factor H (<338 mg/l) (%)6/127 (4.7)3/57 (5.3)3/70 (4.3)1.0000 Low factor I ( 440 μg/l) (%)71/118 (60.2)30/54 (55.5)41/64 (64.0)0.3470Antibodies against complement proteins Circulating C3NeF against the C3bBb80/125 (64)41/57 (71.9)39/68 (57.3)0.0970 Anti-FH Ab14/116 (12)9/53 (17)5/63 (8)0.1363 Anti-C3 Ab3/116 (2.6)1/53 (1.9)2/63 (3.2)0.6633 Anti-Fb Ab3/116 (2.6)1/53 (1.9)2/63 (3.2)0.6633 In association with C3NeF11/17 (64.7)7/8 (87.5)4/9 (44.4)0.6372Novel or rare variants in complement genes All28/120 (23.3)11/55 (20)17/65 (26.1)0.4271 With C3NeF12/28 (42.8)6/11 (54.5)6/17 (35.3)0.3147Ab, antibody; DDD, dense deposit disease; ESRD, end-stage renal disease; F, female; FH, factor H; M, male.Pediatric (younger than 18 years of age).P values in bold highlight a significant difference among groups.a The histologic characteristics have been reported in 85 patients4Servais A. Noel L.H. Roumenina L.T. et al.Acquired and genetic complement abnormalities play a critical role in dense deposit disease and other C3 glomerulopathies.Kidney Int. 2012; 82: 454-464Abstract Full Text Full Text PDF PubMed Scopus (371) Google Scholar (56 C3GN and 29 DDD patients). Open table in a new tab Ab, antibody; DDD, dense deposit disease; ESRD, end-stage renal disease; F, female; FH, factor H; M, male. Pediatric (younger than 18 years of age). P values in bold highlight a significant difference among groups. C3NeF activity was found in 72% and 57% of patients with pediatric and adult onset, respectively. Anti-CFH antibodies were detected in 12% (N = 14) of patients, but only 3 of these 14 patients had a titer >1000 UA at diagnosis. To evaluate the ability of a patient’s IgG to stabilize the cell-bound properdin–stabilized C3bBb convertases, we developed a first hemolytic assay using properdin added at the time of the convertase formation. In similar experimental conditions, the number of hemolytic sites was related to the input of C5.19Fischer E. Kazatchkine M.D. Surface-dependent modulation by H of C5 cleavage by the cell-bound alternative pathway C5 convertase of human complement.J Immunol. 1983; 130: 2821-2824PubMed Google Scholar For this study, we used the same purified IgG samples as those used for the C3NeF assays to evaluate the ability of IgG to stabilize the properdin-C3bBb enzyme. The capacity of patients’ IgG to stabilize the C5 convertases was first examined using rat sera to trigger lysis. Less than 20% of residual lytic sites per cell was observed at 30 minutes with 400 μg IgG purified from healthy donors (N = 30). Residual lysis of 20% is the threshold for positive screening of circulating antibodies for stabilization of the C5 convertase. Of the 59 tested patients’ IgG, 29 (49%) were positive in this assay (Figure 1a). The capacity of patients’ IgG to stabilize the C3bBbP was IgG dose dependent (Figure 1b). The direct stabilizing effect of IgG on cell-bound C5 convertase was further demonstrated using purified human C5 to C9 to trigger lysis. In these experimental conditions, the number of hemolytic sites was related to C5 as no further C3 could be cleaved (data not shown). Cell-bound C5 convertase decayed with time after formation because no lysis occurs after 30 minutes in the presence of IgG from healthy donors. We tested the IgG of 6 representative patients in both assays. The positive IgG in the first assay stabilized the C5 convertase in the second assay. No stabilization of the C5 convertase occurred for IgG that was negative in the first test (Supplementary Table S2). We used 4 representative protein G–eluted IgGs to test their capacity to bind the convertases C3bBb formed with or without properdin in real time by surface plasmon resonance (SPR). The patients’ IgGs P1 and P2 are positive only in a C3 convertase stabilization assay (C3/C5 NeF +/−), and the patients’ IgGs P3 and P4 are positive in C3 and C5 convertase stabilization assays (C3/C5 NeF +/+). Protein G–eluted IgGs from 2 healthy donors were used as controls. C3b was immobilized on the chip, and convertase was formed by flowing either FD and FB or FD, FB, and properdin. After convertase formation, the IgGs were injected onto the chip, and their binding to the complex was followed. The binding signal was higher when the patients’ IgGs (P1 and P2) were flowed over the C3bBb convertase, but no binding signal was observed in the presence of C3bBbP convertase (Figure 2a). The binding signal was higher on both the C3bBb and C3bBbP convertases in the presence of IgGs from P3 and P4 (Figure 2b). As properdin stabilizes the C3bBb convertase,20Hourcade D.E. The role of properdin in the assembly of the alternative pathway C3 convertases of complement.J Biol Chem. 2006; 281: 2128-2132Crossref PubMed Scopus (183) Google Scholar we studied the residual C3 convertase sites after the addition of properdin at the time of the dissociation step in the presence of an increased amount of patient IgG. We tested 2 representative patients’ IgG samples positive only for C3NeF (P6 and P7) and 2 positive for both C3NeF and C5NeF (P4 and P5). Increasing the amount of patient IgG increased the percentage of stabilization in a dose-dependent manner. Fifty percent of convertase stabilization was obtained with variable amounts of patient IgG ranging from 50 to 400 μg of IgG. In our experimental conditions, 60% of residual lytic sites were obtained 20 minutes after the addition of 800 ng of properdin on the preformed C3bBb convertase. The addition of increasing amounts of IgG from patients positive for C3NeF only (C3/C5 NeF +/−) did not change the stabilization of the convertase induced by properdin (Figure 3a). IgG from patients positive for C3 and C5NeF (double positives C3/C5 NeF +/+) caused a significant increase in the residual lytic sites in the presence of properdin (Figure 3b). Patients with C3G can be grouped into 4 distinct biological subsets based on the results of testing for the C3 and C5 convertase stabilization assays (Figure 4): (i) the C3NeF screening was negative in the 2 assays (13/59, 22%); (ii) the patient’s IgG stabilized the C3bBb convertase but did not stabilize the C5 convertase (17/59, 28.8%); (iii) the patient’s IgG stabilized both C3 and C5 convertases (23/59, 39%); (iv) the patient’s IgG stabilized the C5 convertase but not the C3 convertase (6/59, 10.2%). No significant difference was noted in the clinical presentation, but in children, double-positive patients have a better outcome than the double-negative patients (P = 0.0002), and an opposite trend is observed in adult (Supplementary Figure S2 and Supplementary Table S3). Double-positive or C5NeF-positive patients were more frequently in the patients with C3GN than DDD patients (88% vs. 37%, P = 0.001) (Table 2).Table 2Clinical and biological data of the subcohort of 59 tested patients for C3NeF and C5NeF according to the status of the nephritic factorsC3NeFC5NeFNegativeNegativePositiveNegativePositivePositiveNegativePositiveNegativeNAPositiveNAP valueNo.1317236636Sex, no. (% male)8 (62)8 (47)9 (39)4 (67)1 (17)25 (69)0.07At diagnosis Age, yr (range)28 (8–72)14 (1–34)17 (2–64)28 (6–67)27 (19–53)20 (5–63)0.22 Adult, >18 yr, no. (%)7 (48)9 (53)8 (35)4 (67)6 (100)16 (44)0.09 Nephrotic syndrome (n/number of available data; %)3/11 (27)10/16 (63)16/23 (70)5/6 (83)1/5 (20)17/31 (55)0.07 eGFR, ml/min per 1.73 m2 (range)89 (20–100)72 (20–100)75 (10–100)75 (54–100)62 (25–100)92 (10–100)0.1 CKD 3/4 or 50/8 (0)8/16 (50)7 (30)1/2 (50)3 (50)8 (22)0.1Immunosuppressive treatment, %5/10 (50)13 (76)18 (78)4 (67)0/5 (0)12/34 (35)0.007 C3 convertase stabilization, % (median, IQR)2 (0; 6)50 (35; 75)51 (35; 70)7 (3; 10)5 (2; 9)30 (21; 39)<0.0001 C5 convertase stabilization, % (median, IQR)5 (1; 8)8 (4; 11)58 (40; 100)54 (47; 88)<0.0001 C3, mg/l (median, IQR)1010 (933; 1110)849 (123; 1020)393 (188; 823)498 (298; 753)729 (602; 894)695 (562; 962)0.0016 Low C3, <660 mg/l (%)1 (8)8 (47)14 (61)4 (67)2 (33)14 (39)0.0408 sC5b-9, ng/ml, median (IQR)449 (376; 488)352 (278; 432)548 (362; 887)847 (540; 1620)442 (320; 478)449 (313; 610)0.0004 High sC5b-9, ng/ml (%)8 (62)5 (29)17 (74)6 (100)4 (67)21 (58)0.0261At follow-up Follow-up, mo113 (1–360)114 (13–368)81 (21–278)102 (56–205)39 (3–339)92 (20–427)0.47CKD <3 (%)0/7 (0)3/7 (43)2/12 (17)2/4 (50)1/4 (25)1 (3)0.27 ESRD (%)8 (62)9 (53)8 (35)2 (33)2 (33)18 (50)0.58 Adult-onset ESRD (%)3/7 (43)7/9 (78)7/8 (88)1/4 (25)2 (33)13/16 (81)0.08 Pediatric-onset ESRD (%)5/6 (83)2/8 (25)1/15 (7)1/2 (50)—5/20 (25)0.01 Mean (range) time to ESRD, yr8.4 (0.1–25)5.8 (1.5–25)5 (2.4–11)21 (17–25)0.5 (0–1)8 (2–27)0.18 Age at ESRD stage, yr (mean)33 (14–74)24 (8–48)36 (13–66)39 (34–46)34 (24–43)36 (36–65)0.825-yr ESRD, n (%)5/13 (38)6/17 (35)5/23 (22)0 (0)2 (33)13 (36)0.46Pathologic pattern, no. (%) DDD1 (8)12 (70)5 (22)2 (33)0 (0)13 (36)0.033 C3GN12 (92)5 (30)18 (78)4 (67)6 (100)23 (64)0.033C3GN, C3 glomerulonephritis; CKD, chronic kidney disease; DDD, dense deposit disease; eGFR, estimated glomerular filtration rate; ESRD, end-stage renal disease; FH, factor H; IQR, interquartile range; NA, not available; NeF, nephritic factor.The Fisher exact test was used to compare the characteristics by ESRD status. The percentage of those with ESRD was significantly higher in patients with adult-onset ESRD (7/8, 88%) compared with those with pediatric-onset ESRD (1/15, 7%) in the double-positive group (P = 0.0002).Using logistic regression models, the results on adults showed a trend toward a higher probability of ESRD in patients who were double positives (odds ratio: 9.33, P = 0.089). Conversely, the model for children showed a significant impact toward a lower ESRD probability for patients who were double positive compared with double-negative patients (odds ratio: 0.014, P = 0.0048).P values in bold highlight a significant difference among groups. Open table in a new tab C3GN, C3 glomerulonephritis; CKD, chronic kidney disease; DDD, dense deposit disease; eGFR, estimated glomerular filtration rate; ESRD, end-stage renal disease; FH, factor H; IQR, interquartile range; NA, not available; NeF, nephritic factor. The Fisher exact test was used to compare the characteristics by ESRD status. The percentage of those with ESRD was significantly higher in patients with adult-onset ESRD (7/8, 88%) compared with those with pediatric-onset ESRD (1/15, 7%) in the double-positive group (P = 0.0002). Using logistic regression models, the results on adults showed a trend toward a higher probability of ESRD in patients who were double positives (odds ratio: 9.33, P = 0.089). Conversely, the model for children showed a significant impact toward a lower ESRD probability for patients who were double positive compared with double-negative patients (odds ratio: 0.014, P = 0.0048). P values in bold highlight a significant difference among groups. To evaluate the influence of the presence of different types of NeFs on the level of complement activation in patients, we analyzed the C3 and sC5b-9 plasmatic levels. C3 levels were significantly lower in patients positive for C3NeF (101 tested samples) or C5NeF (59 tested samples) compared with patients negative for these tests (data not shown). sC5b-9 levels were significantly higher in C5NeF-positive patients than in C5NeF-negative patients. In the group of patients positive for C3NeF only, plasma C3 levels were lower in 47% (8/17) and sC5b-9 was in the normal range in 71% (12/17). Four of the 5 patients with higher sC5b-9 levels were close to the threshold of the normal ranges. In the group of patients double positive for C3NeF and C5NeF, plasma C3 levels were lower in 61% (14/23), and sC5b-9 was higher in 74% (17/23). All the patients with C5NeF and without C3NeF had a high sC5b-9 level (Table 2 and Supplementary Figure S3). The capacity of C3NeF and C5NeF to stabilize the C3 convertase was negatively correlated with C3 consumption (P < 0.0001 and P = 0.0044, respectively). The capacity of C5NeF to stabilize the C5 convertase was positively correlated with the sC5b-9 (P = 0.0014). No correlation was observed between the level of sC5b-9 and the capacity of C3NeF to stabilize the C3 convertase (P = 0.9181) (Figure 5). We previously reported a French retrospective cohort of 56 C3GN and 29 DDD patients of whom two-thirds had C3NeF directed against the C3bBb convertase and/or pathogenic variants in 1 of the complement genes involved in the AP.4Servais A. Noel L.H. Roumenina L.T. et al.Acquired and genetic complement abnormalities play a critical role in dense deposit disease and other C3 glomerulopathies.Kidney Int. 2012; 82: 454-464Abstract Full Text Full Text PDF PubMed Scopus (371) Google Scholar This new retrospective study of genetic abnormalities and autoantibodies in 127 patients with C3G is, to our knowledge, the largest of its kind published to date. Our data confirm and expand insights on the antibodies that stabilize the C3 and C5 convertases. We showed that C3NeF and C5NeF have distinct functional consequences and pathologic associations. C3 and sC5b-9 are the 2 main biomarkers reflecting the activation of the alternative and terminal pathways of complement. In our series of patients with C3NeF, the plasma levels of C3 were below the normal range in 53% (40 of 76 patients) and plasma levels of sC5b-9 above the normal range in 62% (47 of 76 patients). Another study investigating a C3G population also found that the subsequent activation of the terminal pathway in the presence of C3NeF was inconsistent.10Zhang Y. Nester C.M. Martin B. et al.Defining the complement biomarker profile of C3 glomerulopathy.Clin J Am Soc Nephrol. 2014; 9: 1876-1882Crossref PubMed Scopus (60) Google Scholar This heterogeneity suggests the presence of nephritic factors with different functional activities and different consequences in vivo, a hypothesis that had not yet been assessed in depth. Because of the functional similarities with C3NeF, the autoantibody that prolongs the half-life of the C5 convertase is defined as a C5 nephritic factor (C5NeF). Although the functional role of C3NeF has been well described, few studies to date have focused on the role of autoantibodies in stabilizing the AP C5 convertase.17Paixao-Cavalcante D. Lopez-Trascasa M. Skattum L. et al.Sensitive and specific assays for C3 nephritic factors clarify mechanisms underlying complement dysregulation.Kidney Int. 2012; 82: 1084-1092Abstract Full Text Full Text PDF PubMed Scopus (71) Google Scholar For this purpose, we set up functional hemolytic assays to study the capacity of patients’ IgGs to stabilize the C5 convertase. The C3 convertase (C3bBb) is converted to C5 convertase (C3bBbC3b) via its interaction with an additional molecule of C3b.21Bajic G. Degn S.E. Thiel S. et al.Complement activation, regulation, and molecular basis for complement-related diseases.EMBO J. 2015; 34: 2735-2757Crossref PubMed Scopus (215) Google Scholar A minimal inherent activity of the C3 convertase is sufficient for the switch from C3 to C5 convertase activity under the influence of properdin at the cell surface.7Berends E.T. Gorham Jr., R.D. Ruyken M. et al.Molecular insights into the surface-specific arrangement of complement C5 convertase enzymes.BMC Biol. 2015; 13: 93Crossref PubMed Scopus (37) Google Scholar, 20Hourcade D.E. The role of properdin in the assembly of the alternative pathway C3 convertases of complement.J Biol Chem. 2006; 281: 2128-2132Crossref PubMed Scopus (183) Google Scholar Therefore, we built a C5 convertase with FB, FD, and properdin on the surface of C3b-coated sheep erythrocytes. The assay followed the same experimental procedure as that used for the screening of C3NeF except that the C3 convertase was formed in the presence of properdin. In these experimental conditions, the C5 convertase spontaneously decays, resulting in decreased hemolysis at later time points. In presence of C5NeF, the convertase will be stabilized. We further confirmed the direct capacity of patients’ IgGs to stabilize the C5 convertase by using a second functional assay performed with purified C5 to C9 proteins. We showed that 29 of 59 tested patients’ IgGs were positive for the C5NeF. We isolated some antibodies that only stabilized the C3 convertase and had a limited effect on the C5 convertase (C3NeF) and, on the contrary, others that only stabilized the C5 convertase (C5NeF). In 50% of the cases, the patient IgG stabilized both the C3 and the C5 convertases (called C3/C5NeF +/+ or double positive). More interestingly, we showed for the first time that the sC5b-9 level is directly correlated with the percentage of stabilization of the C5NeF, confirming the dysregulation of the C5 convertase. C5NeFs are also related to C3 consumption due to the fact that some NeFs were able to stabilize both the C3 and C5 convertases. Furthermore, we provide evidence that binding of properdin to C3bBb has an additive effect on the convertase stabilization with C5NeF but not with C3NeF. This remarkable differential effect of properdin and the epitopes of the antibodies on the convertases may influence the functional consequences of the nephritic factors. We decided to further explore these findings by evaluating the concordance between the results of the nephritic factor assays and the histologic classification. Among the patients with isolated C3NeF, 71% were diagnosed with DDD. On the contrary, among patients with C5NeF, renal pathology was suggestive of C3GN in 72% of the cases
Referência(s)