SUN-025 The role of fumarate in PLA2R-associated membranous nephropathy: urine metabolomic analysis and experimental validation

Abstract

The clinical remission of phospholipase A2 receptor (PLA2R)-associated membranous nephropathy (MN) does not always coincide with its immunological remission, suggesting that complex biochemical processes influenced by host adaptation or environmental factors may modulate the pathogenesis. However, the underlying mechanisms remain elusive. In the present study, we aimed to characterize the urinary metabolomic profiles of PLA2R-associated MN, assess the prognostic value of potential metabolomic biomarkers identified, and validate the role of these biomarkers via in vitro experiments mimicking the PLA2R-associated MN milieu. We used nuclear magnetic resonance spectroscopy to compare the urinary metabolomic profiles of PLA2R-associated MN patients (n=57) with those of patients with minimal change disease (MCD, n=73) and healthy controls (n=82). Urine samples for PLA2R-associated MN and MCD were collected with written consent at the time of kidney biopsy in Seoul National University Hospital between October 2009 and March 2016. PLA2R-associated MN was defined as patients having positive serum levels of anti-PLA2R or glomerular PLA2R antigen. We also performed an experimental validation by using primary cultured human podocytes and purified IgG from a patient with PLA2R-associated MN and a healthy control. Among fourteen metabolites differentially expressed between PLA2R-associated MN patients and healthy controls, fumarate was the only differentially expressed metabolite in PLA2R-associated MN compared to MCD. High urinary fumarate levels could predict the outcome of PLA2R-associated MN. Fumarate hydratase, which catalyzes the hydrolysis of fumarate, was colocalized with podocalyxin, and its expression was lower in kidney sections from PLA2R-associated MN patients than in those from the controls. IgG purified from serum with a high anti-PLA2R titer (MN-IgG) bound the same 185-kD podocyte lysate proteins as did commercial anti-PLA2R. MN-IgG stimulation of podocytes led to decreased fumarate hydratase, WT1 and ZO-1 expression and increased fumarate, Snail and fibronectin expression. The MN-IgG treatment also increased albumin flux across the podocyte layer. The activation of fumarate hydratase ameliorated these alterations. In summary, these findings provide evidence of the role played by fumarate in podocyte injury during PLA2R-associated MN through the acquisition of the mesenchymal phenotype and the loss of cell-cell adhesion and indicate that fumarate may be a potential target for therapeutic intervention in PLA2R-associated MN.