Differential expression of urinary exosomal MicroRNAs miR-21-5p and miR-30b-5p in individuals with diabetic kidney disease

Background and aims: The earliest stages of diabetic kidney disease (DKD) are asymptomatic and may go unrecognised without regular screening for albuminuria and/or declining eGFR. Renal impairment develops in approximately one third of those with diabetes leading to increased comorbidity and premature mortality. Biomarkers for identification of DKD are needed as current tests lack sensitivity for detecting early kidney damage. MicroRNAs (miRNAs) are short, non-coding regulatory molecules commonly found in urinary exosomes and are differentially expressed as renal function declines. Urine is an ideal source for non-invasive miRNA profiling because it is produced in the kidneys and its constituents reflect renal function. Materials and methods: We evaluated urinary exosomal miRNA expression in persons with type 2 diabetes mellitus and DKD (T2DKD). 87 human urinary exosomal miRNAs were profiled in a discovery cohort of patients with T2DKD (n=14) and age and gender matched controls with type 2 diabetes mellitus and normal renal function (T2DNRF; n=15). Independent validation of differentially expressed target miRNAs was performed in a second cohort with T2DKD (n=22) and two control groups: T2DNRF (n=15) and controls with chronic kidney disease (CCKD) and poor renal function without diabetes (n=18). Receiver operating characteristics (ROC) were used to assess the potential of differentially expressed miRNAs to discriminate between individuals with good and poor renal function. Results: In the discovery cohort, urinary miR-21-5p, let-7e-5p and miR-23b-3p were significantly up-regulated in T2DKD compared to T2DNRF (p<0.05). Conversely, miR-30b-5p and miR-125b-5p expression was significantly lower in T2DKD (p<0.05). Independent validation confirmed up-regulation of miR-21-5p in a replication cohort in T2DKD (2.13-fold, p=0.006) and in CCKD (1.73-fold, p=0.024). In contrast, miR-30b-5p was down-regulated in T2DKD (0.82-fold, p=0.006) and in CCKD (0.66-fold, p<0.002). Independently, miR-21-5p had an area under the curve (AUC) of 0.791 (95% CI: 0.638-0.944, p=0.003), and miR-30b-5p had an AUC of 0.773 (95% CI: 0.610-0.935, p=0.005). Combined, both miRNAs improved the discriminatory power of the model further with an AUC of 0.833 (95% CI: 0.688-0.978, p=0.001). ROC analyses were also undertaken to discriminate poor renal dysfunction in T2DKD and CCKD combined; miR-21-5p: AUC: 0.785 (95% CI: 0.643-0.927, p=0.001); miR-30b-5p: AUC: 0.797 (95% CI: 0.656-0.938, p=0.001). When both miRNAs were included together in the model the discriminatory power increased to 0.840 (95% CI: 0.708-0.972, p<0.001). Conclusion: A profiling approach identified altered expression of urinary exosomal miRNAs miR-21-5p and miR-30b-5p in association with poor renal function. These changes were subsequently validated in a larger independent cohort. Urinary miR-21-5p was enriched in T2DKD and CKD patients compared with T2DM individuals with good renal function; while in contrast, the expression of miR-30b-5p was reduced in T2DKD and CKD patients. Both miRNAs were significantly correlated with serum creatinine levels. Urinary miR-21-5p and miR-30b-5p represent candidate biomarkers of renal function, although further clarification is necessary to determine the extent of this association more generally across individuals with other renal conditions. Urinary miRNAs offer a simple, non-invasive biomarker of kidney function measurable in an easily accessible liquid biopsy.