• Cindy M Yamamoto, Taku Murakami, Melanie L Oakes, Masato Mitsuhashi, Colleen Kelly, Robert R Henry, and Kumar Sharma.
• Hitachi Chemical Co. America, Ltd., R and D Center, Irvine, California, USA.
• Am. J. Nephrol. 2018 Jan 1; 47 (5): 283-291.

BackgroundExtracellular vesicles (EVs) enclose mRNA derived from their cell of origin and are considered a source of potential biomarkers. We examined urinary EV mRNA from individuals with diabetic kidney disease (DKD), chronic kidney disease, type 2 diabetes (T2DM), and obese and healthy controls to determine if such biomarkers had the potential to classify kidney disease and predict patients at higher risk of renal function decline.MethodsA total of 242 participants enrolled in this study. Urinary EV mRNA from all subjects were isolated by a filter-based platform, and the expression of 8 target genes were determined by quantitative polymerase chain reaction (qPCR). Changes in estimated glomerular filtration rate (eGFR) in 161 T2DM patients were evaluated for 2 consecutive years and compared with EV RNA profiles at baseline.ResultsWe observe that mild and severe DKD groups show a significant 3.2- and -4.4-fold increase in UMOD compared to healthy controls and expression increases linearly from healthy, diabetic, and DKD subjects. UMOD expression is significantly correlated to albumin creatinine ratio (ACR), eGFR, and HbA1c. Using linear discriminant analyses with mRNA from severe DKD and T2DM as training data, a multi-gene signature classified DKD and -non-DKD with a sensitivity of 93% and specificity of 73% with area under the receiver operating characteristic (ROC) curve (AUC) = 0.90. Although 6% of T2DM were determined to have a > 80% posterior probability of developing DKD based on this mRNA profile, eGFR changes observed within the 2-year follow-up did not reveal a decline in kidney function.ConclusionUrinary EV UMOD mRNA levels are progressively elevated from T2DM to DKD groups and correlate with widely used eGFR and ACR diagnostic criteria. An EV mRNA signature could identify DKD with greater than 90% sensitivity and 70% specificity.© 2018 S. Karger AG, Basel.

46 keywords...

hide…