Coexistence of different circulating anti-podocyte antibodies in membranous nephropathy

Abstract

Background and objectives: The discovery of different podocyte autoantibodies in membranous nephropathy (MN) raises questions about their pathogenetic and clinical meaning. This study sought to define antibody isotypes and correlations; to compare levels in MN, other glomerulonephritides, and controls; and to determine their association with clinical outcomes.

Design, setting, participants, & measurements: Serum IgG(1), IgG(3), and IgG(4) against aldose reductase (AR), SOD2, and α-enolase (αENO) were measured at diagnosis in 186 consecutive MN patients, in 96 proteinuric controls (36 with FSGS, and 60 with IgA nephropathy), and in 92 healthy people recruited in four Italian nephrology units. Anti-phospholipase A2 receptor (PLA2r) and anti-neutral endopeptidase (NEP) IgG(4) were titrated in the same specimens. Association with 1-year follow-up clinical parameters was studied in 120 patients.

Results: IgG(4) was the most common isotype for all antibodies; IgG(1) and IgG(3) were nearly negligible. IgG(4) levels were positive in a significant proportion of MN patients (AR, 34%; SOD2, 28%; αENO, 43%). Antibody titers were higher in MN than in healthy and pathologic controls (P<0.005). Anti-NEP IgG(4) did not differ from normal controls (P=0.12). Anti-PLA2r IgG(4) was detected in 60% of patients and correlated with anti-AR, anti-SOD2, and anti-αENO IgG(4) (P<0.001). In MN patients negative for the whole antibody panel (20%), 1-year proteinuria was lower compared with patients with at least one antibody positivity (P<0.05).

Conclusions: Our data suggest that IgG(4) is the prevalent isotype for antibodies against cytoplasmic antigens of podocytes (AR, SOD2, αENO). Their levels were higher than in other proteinuric glomerulonephritides and in normal controls and were correlated with anti-PLA2r. Only baseline negativity for all known antibodies predicted lower 1-year proteinuria.

Figure 1.

Serum antibodies against cytoplasmic antigens of podocytes are increased in a significant portion of MN patients. Circulating IgG4 (A) anti-AR, (B) anti-SOD2, and (C) anti-αENO in MN and control populations are shown. In all cases, we utilized a technique based on dot blot analysis with recombinant protein linked to nitrocellulose as an antigen (Supplemental Figure 1). Results are given as chemiluminescence OD arbitrary units that corresponds to one unit of signal intensity of chemiluminescence detected by VersaDoc and computed with QuantyOne software (Bio-Rad). The horizontal line is set at the 95th percentile of levels titrated in normal controls. AR, aldose reductase; αENO, α-enolase; MN, membranous nephropathy.

Figure 2.

ROC curves. The area under the ROC curves for each antibody was significantly greater in MN patients compared with (A) patients with other nephropathies or (B) normal participants, although to a lesser extent. ROC, receiver operating characteristic; MN, membranous nephropathy.

Figure 3.

Serum levels of circulating anti-PLA2r and anti-NEP IgG₄. (A) Anti-PLA2r was revealed utilizing a Western blot assay with podocyte extracts as a fixed antigen. PLA2r was previously recognized by specific antibodies in the area of the gel between 116 and 220 kD where PLA2r was the unique spot (Supplemental Figure 1). Anti-PLA2r serum positivity was validated with a semiquantitative immunofluorescence test (Supplemental Table 2). (B) Anti-NEP IgG₄ was determined with dot blot analysis (Supplemental Figure 1). The horizontal line is set at the 95th percentile of levels titrated in normal controls. Anti-PLA2r, anti-phospholipase A2 receptor; anti-NEP, anti-neutral endopeptidase.