Complement Factor I Variants in Complement-Mediated Renal Diseases

Abstract

C3 glomerulopathy (C3G) and atypical hemolytic uremic syndrome (aHUS) are two rare diseases caused by dysregulated activity of the alternative pathway of complement secondary to the presence of genetic and/or acquired factors. Complement factor I (FI) is a serine protease that downregulates complement activity in the fluid phase and/or on cell surfaces in conjunction with one of its cofactors, factor H (FH), complement receptor 1 (CR1/CD35), C4 binding protein (C4BP) or membrane cofactor protein (MCP/CD46). Because altered FI activity is causally related to the pathogenesis of C3G and aHUS, we sought to test functional activity of select CFI missense variants in these two patient cohorts. We identified 65 patients (16, C3G; 48, aHUS; 1 with both) with at least one rare variant in CFI (defined as a MAF < 0.1%). Eight C3G and eleven aHUS patients also carried rare variants in either another complement gene, ADAMTS13 or THBD. We performed comprehensive complement analyses including biomarker profiling, pathway activity and autoantibody testing, and developed a novel FI functional assay, which we completed on 40 patients. Seventy-eight percent of rare CFI variants (31/40) were associated with FI protein levels below the 25^th percentile; in 22 cases, FI levels were below the lower limit of normal (type 1 variants). Of the remaining nine variants, which associated with normal FI levels, two variants reduced FI activity (type 2 variants). No patients carried currently known autoantibodies (including FH autoantibodies and nephritic factors). We noted that while rare variants in CFI predispose to complement-mediated diseases, phenotypes are strongly contingent on the associated genetic background. As a general rule, in isolation, a rare CFI variant most frequently leads to aHUS, with the co-inheritance of a CD46 loss-of-function variant driving the onset of aHUS to the younger age group. In comparison, co-inheritance of a gain-of-function variant in C3 alters the phenotype to C3G. Defects in CFH (variants or fusion genes) are seen with both C3G and aHUS. This variability underscores the complexity and multifactorial nature of these two complement-mediated renal diseases.

Keywords: C3 glomerulonephritis; C3 glomerulopathy; atypical hemolytic uremic syndrome; complement; dense deposit disease; factor I.

Figure 1

CFI variants identified in patients with C3G and aHUS. (A) Schematic view of factor I and the 45 rare variants (MAF < 0.1%) identified in 65 patients (red, aHUS; blue, C3G; brown, both C3G and aHUS). All variants were identified in heterozygosis except one homozygote (p.Ala210Thr) and two compound heterozygotes (brackets above variants indicate compound heterozygosity). Novel variants are marker with #. (L, leading sequence; FIMAC, Factor I Membrane Attack Complex; SRCR, Scavenger Receptor Cysteine-Rich; LDLR, Low-Density Lipoprotein Receptor). (B) FI levels in 40 patients (open circles, aHUS; filled blue circles, C3GN; filled red circles, DDD; open square, C3G/aHUS; dashed lines, reference values: red, upper and lower limits of normal; blue, percentile lines).

Figure 2

Complement biomarker profiling in aHUS and C3G patients. (A) Age distribution. (B) Factor I levels. (C) Factor H levels. (D–H) Biomarkers for alternative pathway (AP) activity including C3 and factor B, and associated activation products C3c and Bb. (I) AP fluid-phase activity by immunofixation eletrophoresis. (J–L) Biomarkers for terminal pathway activity including C5, properdin and soluble terminal complement complex (sC5b-9). (left column, aHUS; right column-red dots, DDD patients; black dots, C3GN patients; open circle, C3GN/aHUS patient; dashed line, lower limit of normal value for each assay; patients on eculizumab excluded from (J, L); *P < 0.05).

Figure 3

Assessing FI function. In the presence of a cofactor (A) FH 50 nM; (B) sCR1 25nM; (C) MCP 50nM, C3b cleavage activity by FI in patient serum (diluted 1:32) is measured using C3b-decorated sheep erythrocytes. Non-cleaved C3b is developed by the addition of both FB/FD to form C3 convertase and rat-EDTA serum as a source of C5-C9. Percent hemolysis is calculated and plotted as a function of FI concentration determined by ELISA. The grey line in each figure serves as a normal reference line calculated from a normal sample (FI concentration, 40 mg/L) with serial dilutions. Serum from a CFI c.1429+1G>C homozygote (FI is undetectable) serves as a positive control (black dot). (D) Circulating FI by a Western blot. Patient sample (1:80 diluted in PBS) with reducing reagent separated on a 4-15% polyacrylamide gel and transferred. FI was visualized by an antibody specific to the heavy chain (50k Da). CFI p.Arg336Gly results in an unprocessed single chain (88k Da).

Figure 4

Increases in FI, FB, C4 and FH in the acute phase. These complement proteins are positive acute-phase reactants, and their levels were significantly increased in the acute phase in fourteen patients in our C3G and aHUS registries. The increase in FI does not occur in isolation but rather in conjunction with increases in other complement biomarkers, especially FB, C4 and FH. This pattern of increase was not observed in the patients described in this study (left, complement levels in remission; right, complement levels in the acute phase).