Defining the complement biomarker profile of C3 glomerulopathy

Abstract

Background and objectives: C3 glomerulopathy (C3G) applies to a group of renal diseases defined by a specific renal biopsy finding: a dominant pattern of C3 fragment deposition on immunofluorescence. The primary pathogenic mechanism involves abnormal control of the alternative complement pathway, although a full description of the disease spectrum remains to be determined. This study sought to validate and define the association of complement dysregulation with C3G and to determine whether specific complement pathway abnormalities could inform disease definition.

Design, setting, participants, & measurements: This study included 34 patients with C3G (17 with C3 glomerulonephritis [C3GN] and 17 with dense deposit disease [DDD]) diagnosed between 2008 and 2013 selected from the C3G Registry. Control samples (n=100) were recruited from regional blood drives. Nineteen complement biomarkers were assayed on all samples. Results were compared between C3G disease categories and with normal controls.

Results: Assessment of the alternative complement pathway showed that compared with controls, patients with C3G had lower levels of serum C3 (P<0.001 for both DDD and C3GN) and factor B (P<0.001 for both DDD and C3GN) as well as higher levels of complement breakdown products including C3d (P<0.001 for both DDD and C3GN) and Bb (P<0.001 for both DDD and C3GN). A comparison of terminal complement pathway proteins showed that although C5 levels were significantly suppressed (P<0.001 for both DDD and C3GN) its breakdown product C5a was significantly higher only in patients with C3GN (P<0.05). Of the other terminal pathway components (C6-C9), the only significant difference was in C7 levels between patients with C3GN and controls (P<0.01). Soluble C5b-9 was elevated in both diseases but only the difference between patients with C3GN and controls reached statistical significance (P<0.001). Levels of C3 nephritic factor activity were qualitatively higher in patients with DDD compared with patients with C3GN.

Conclusions: Complement biomarkers are significantly abnormal in patients with C3G compared with controls. These data substantiate the link between complement dysregulation and C3G and identify C3G interdisease differences.

Keywords: C3 glomerulopathy; C3GN; alternative pathway; complement; dense deposit disease.

Figure 1.

Electron microscopic distinction between DDD and C3GN. (A) Representative electron microscopy from three patients in the DDD cohort shows the characteristic extremely electron-dense well defined osmiophilic deposits that are typically located in the lamina densa of the glomerular basement membrane. (B) By contrast, representative electron microscopy from three patients in the C3GN cohort shows less well defined deposits that may lead to massive expansion of both the glomerular basement membrane and the mesangium. The reason for these underlying differences is not known. C3GN, C3 glomerulonephritis; DDD, dense deposit disease.

Figure 2.

The alternative and terminal pathways of complement. The alternative pathway is constitutively active due to spontaneous C3 hydrolysis, a process known as tick-over that leads to the generation of C3(H₂O) (also known as iC3). C3(H₂O)Bb then forms, leading to creation of the C3 convertase of the alternative pathway. The C3bBb amplification loop is represented by the pink circular arrow. If positive feedback continues unchecked or if C3bBb is stabilized by an autoantibody like C3Nef (shown in purple), the alternative pathway components are consumed and their breakdown products are generated. In the presence of FI and its cofactors (like FH and CR1), C3b is broken down to iC3b and then by additional cleavages, to C3c, C3d and C3g. Excessive C3 convertase promotes the formation of C5 convertase by the recruitment of additional C3b molecules. Note the similarity between the C3 and C5 convertases. C5 convertase cleaves C5 to C5a and C5b and initiates the assembly of the membrane attack complex by the sequential addition of C6, C7, C8, and C9 or of soluble (C5b-9) in the fluid phase. AT, alternative pathway; C3Nef, C3 nephritic factor; Cof, cofactors; CR1, complement receptor 1; FH, factor H; FI, factor I; MAC, membrane attack complex; TP, terminal pathway.

Figure 3.

Patient recruitment. Thirty-four patients with biopsy-proven DDD or C3GN were included in this study. Because 19 different complement assays were performed and each assay was completed at least three times, the initial inclusion criterion was the availability of sufficient sera and plasma. We estimated the sera requirement at 1 ml/patient and the plasma requirement at 3 ml/patient. Next, light microscopy, immunofluorescence, and electron microscopy results were reviewed to confirm the pathologic diagnosis of C3G. In 52 patients, the immunofluorescence and/or electron microscopy results were not available and thus these patients were excluded. In 12 patients, the diagnosis was equivocal (C3GN versus DDD) and thus these patients were also excluded. EM, electron microscopy; IF, immunofluorescence.

Figure 4.

Complement biomarker assessments in the C3G cohort. (A) Eight alternative pathway biomarkers (C3, FB, properdin, and their associated degradation products C3a, C3c, C3d, Ba and Bb) were assessed in 17 patients with C3GN, 17 patients with DDD, and 100 controls (normal). Serum C3 and FB levels were significantly lower in patients with C3G than in controls, whereas the breakdown products C3d, Ba, and Bb were significantly elevated. Properdin serum levels were reduced in both DDD and C3GN although properdin levels were generally lower in C3GN compared with DDD. (B) Seven terminal pathway biomarkers (C5–C9 and the activation products C5a and sC5b-9) were assessed in the same cohorts. C5 levels were reduced compared with controls in both DDD and C3GN. Although C5a levels tended to be elevated in both diseases, only the elevation in C3GN versus controls was significant. Whereas C6, C8, and C9 levels were not reduced in DDD or C3GN, serum levels of C7 were decreased in C3GN compared with both controls and DDD. Soluble C5b-9 was elevated in both diseases, with the elevation being greater in C3GN. (C) C3Nefs were measured by two direct assays (C3CSA and C3CSAP) and one indirect assay (IFE) (8). No control was positive for any C3Nef assay. In all assays, C3Nef activity was greater in DDD compared with C3GN. For all box-and-whisker plots, the first and third quartiles are represented by the bottom and top ends of the box, the median is represented by the horizontal line in the interior of the box, and the whiskers represent 1.5× the interquartile range. Open circles are outliers. *P<0.05; **P<0.01; *** P<0.001. C3CSA, C3 convertase stabilizing assay; C3CSAP, C3 convertase stabilizing assay with properdin; FB, factor B; IFE, immunofixation electrophoresis; NL, normal; sC5b-9, soluble C5b-9.