Characterization of novel antigens in membranous nephropathy
INTRODUCTION: Membranous nephropathy is an autoimmune disease that targets glomeruli of the kidney. Previous discoveries in membranous nephropathy include the discovery of megalin as an antigen in the proximal tubular brush border fraction (Fx1A) and glomeruli of Heymann nephritis rats, identification of neutral endopeptidase in alloimmune neonatal nephropathy, and discovery of PLA2R and THSD7A as causal antigens in approximately 80-85% of primary membranous nephropathy cases. It was then recognized that there must be other antigens responsible for the remaining 15-20% of cases. OBJECTIVES: The current study aims to screen membranous nephropathy patient serum samples via Western blotting for reactivity with potential antigens in protein extracts of normal human glomeruli, purify potential membranous nephropathy antigens, identify them with mass spectrometry, and validate these identifications with immunoprecipitation and immunohistochemical analysis. Previously, work had been done to identify one novel, 58 kDa antigen. A second novel antigen had been shown in the proximal tubule brush border. Finally, a third protein, CR1, was shown to contain corresponding antibodies in the antibody preparation used in the rat model of membranous nephropathy, making the antigen a protein of interest in human primary membranous nephropathy. METHODS: Using human glomeruli obtained by detergent extraction, we isolated extracellular domains and identified two novel antigens, called 58-kDa and brush-border, with patient serum. We attempted to further purify the 58-kDa antigen with lectin binding columns and partition phase separation. Upon the identification of a small cohort of cases associated with autoimmune tubulointerstitial nephritis, we set out to determine if these sera recognized a novel antigen. Prior to screening human glomerular extract with these sera, we exposed it to partial proteolysis with trypsin, reducing agent β-mercaptoethanol, and tubular elements to further characterize the antigen before it was pulled down with anti-brush-border antigen+ and control IgG4 and analyzed by mass spectrometry. The third and final antigen we investigated was CR1, which we screened with membranous nephropathy sera and immunoblotted its antibody against different protein preparations. RESULTS: Labeling of the extracellular portions of the 58-kDa and brush-border antigens with biotin was successful. It was determined that the 58-kDa antigen was not glycosylated due to its inability to bind lectin columns. The 58-kDa antigen was present in the hydrophilic layer when separated with tritonX-114 detergent. Partial proteolysis of the brush-border antigen with trypsin yielded bands at 140 kDa, 120 kDa and 95 kDa. The brush-border antigen was destroyed under reducing conditions. Candidate proteins for the brush-border antigen as determined by mass spectrometry include megalin and SVEP1. Membranous nephropathy sera were shown to be negative for anti-CR1+ antibody, and anti-CR1+ antibody was reactive with glomeruli and the TBS supernatant fraction. CONCLUSIONS: This study suggests that the 58-kDa antigen which has antibodies in some human primary membranous nephropathy sera contains extracellular portion(s), is not glycosylated, but is membrane-associated. The data indicate that there is also potential for a membranous nephropathy antigen in the tubular brush border with an immunoreactive element around 95 kDa in size, that is sensitive to reducing conditions. Preliminary mass spectrometry information points toward megalin as the identification of this antigen. CR1 does not appear to be a causal antigen in human primary membranous nephropathy.