Modeling complement activation on human glomerular microvascular endothelial cells

Abstract

INTRODUCTION: Atypical hemolytic uremic syndrome (aHUS) is a rare kidney disease caused by dysregulation of the complement alternative pathway. The complement dysregulation specifically leads to damage to the glomerular endothelium. To further understand aHUS pathophysiology, we validated an ex vivo model for measuring complement deposition on both control and patient human glomerular microvascular endothelial cells (GMVECs). METHODS: Endothelial cells were incubated with human test sera and stained with an anti-C5b-9 antibody to visualize and quantify complement depositions on the cells with immunofluorescence microscopy. RESULTS: First, we showed that zymosan-activated sera resulted in increased endothelial C5b-9 depositions compared to normal human serum (NHS). The levels of C5b-9 depositions were similar between conditionally immortalized (ci)GMVECs and primary control GMVECs. The protocol with ciGMVECs was further validated and we additionally generated ciGMVECs from an aHUS patient. The increased C5b-9 deposition on control ciGMVECs by zymosan-activated serum could be dose-dependently inhibited by adding the C5 inhibitor eculizumab. Next, sera from five aHUS patients were tested on control ciGMVECs. Sera from acute disease phases of all patients showed increased endothelial C5b-9 deposition levels compared to NHS. The remission samples showed normalized C5b-9 depositions, whether remission was reached with or without complement blockage by eculizumab. We also monitored the glomerular endothelial complement deposition of an aHUS patient with a hybrid complement factor H (CFH)/CFH-related 1 gene during follow-up. This patient had already chronic kidney failure and an ongoing deterioration of kidney function despite absence of markers indicating an aHUS flare. Increased C5b-9 depositions on ciGMVECs were observed in all samples obtained throughout different diseases phases, except for the samples with eculizumab levels above target. We then tested the samples on the patient's own ciGMVECs. The C5b-9 deposition pattern was comparable and these aHUS patient ciGMVECs also responded similar to NHS as control ciGMVECs. DISCUSSION: In conclusion, we demonstrate a robust and reliable model to adequately measure C5b-9-based complement deposition on human control and patient ciGMVECs. This model can be used to study the pathophysiological mechanisms of aHUS or other diseases associated with endothelial complement activation ex vivo.