SP223
URINARY PROTEOMICS ANALYSIS REVEALS MULTIPLE FEATURES ASSOCIATED WITH CHRONIC KIDNEY DISEASE PROGRESSION

Introduction and Aims: Existing clinical methods diagnose kidney dysfunction at an advanced stage. However, detection of early stages as well as prediction of kidney disease progression remains a challenge. To address this issue, the identification of biomarker candidates is pursued on multiple molecular levels. We investigated the urine proteome by high resolution mass spectrometry for the identification of protein biomarkers associated with Chronic Kidney Disease (CKD) progression. In addition, we aim at predicting mechanisms underlying disease pathophysiology through integration of proteomics with miRNA and peptidomics data.

Methods: Urine from 64 CKD patients (stages 2-4) with available follow-up data of at least 2 years was analyzed. CKD patients were divided into progressors (% slope/year of eGFR ≤ -4.3%; n=32) and non-progressors (-1.1% ≤ % slope/year of eGFR ≤ 5.5%; n=32), and matched for multiple variables including age, gender and CKD stage distribution. Patients with % slope/year of eGFR > 5.5% were excluded. Urinary proteome was analyzed using high-resolution LC-MS/MS. Pathway analysis was performed using Cytoscape ClueGO plugin with Reactome Pathway Database.

Results: We identified 142 differentially expressed proteins in CKD progressors compared to non-progressors: 34 were up-regulated (mainly abundant plasma proteins such as albumin, alpha-2-macroglobulin and haptoglobin), and 108 down-regulated (signaling molecules, proteases, etc. including previously reported CKD-associated proteins such as kallikrein-1, mucin-1, apolipoprotein E and others). Pathways related to heme scavenging and platelet activation were found to be up-regulated, while these associated with extracellular matrix organization or cell signaling down-regulated. Following literature mining and prioritization based on statistical significance, functional annotation and protein expression profiles in kidney, 26 protein candidates were chosen for future validation using multiple reaction monitoring (MRM) in a new set of 70 prospectively collected samples. Integration of proteomics and peptidomics data led to the prediction of several proteases with altered activity in CKD progressors.

Conclusions: Multiple proteomics changes in relation to CKD progression were identified in urine. Deregulation of several biological processes observed on both molecular and pathway levels reflect the known phenotypic manifestations of the disease. The data obtained and their correlation with other omics data collected in the context of the iMODE-CKD network (imodeckd.org) increase power of individual observations and coverage of molecular processes involved in CKD progression. This may lead to the discovery of molecular targets for therapeutic intervention.