Selected miRNAs in urinary extracellular vesicles show promise for non-invasive diagnostics of diabetic kidney disease

Abstract

Diabetic kidney disease is a growing health burden that lacks specific early non-invasive diagnostic procedures. To approach a solution for this clinical need, we sequenced microRNAs of urinary extracellular vesicles and performed biomarker discovery by small RNA sequencing in a type 1 diabetes cohort including males with and without albuminuria. The results were replicated by sequencing or qPCR in two independent cohorts and four previously published datasets including type 1 and 2 diabetes as well as both sexes. Non-diabetic and prostate cancer cohorts were used as additional controls and miRNAs changed due to preanalytical urine collection variables were excluded. Using these data, we additionally validated previously identified reference candidate miRNAs. Correlations with clinical parameters, receiver operating characteristic analysis, targeted mRNAs and pathways including integrated single cell data, and targeted circulating proteins from type 1 and 2 diabetes cohorts were analyzed. We pinpointed 6 stable microRNAs, 11 differentially expressed microRNAs, 9 target proteins and 16 DKD-associated pathways in individuals with diabetic kidney disease. Replication showed that the differentially expressed miRNAs in DKD were partly shared between diabetes subtypes and sexes with overall strongest evidence for miR-192-5p, miR-146a-5p, miR-486-5p, and miR-574-5p. Combination of these miRNAs with clinical variables showed potential to classify individuals with the fastest kidney function decline (sensitivity 0.75-1.00 and specificity 0.83-1.00) even in the normoalbuminuria group, thus holding the potential as early diagnostic markers. Altogether, the candidate microRNAs show specificity for diabetic kidney disease, identify declining kidney function, and target key kidney cell types, mRNAs, proteins, and pathogenic mechanisms.