Cancer Diagnosis Program | Biorepositories and Biospecimen Research Branch

Search BRD Papers

Selected miRNAs in Urinary Extracellular Vesicles Show Promise for Early and Specific Diagnostics of Diabetic Kidney Disease.

Author(s): Barreiro K, Karttunen J, Valo E, Viippola E, Quintero I, Käräjämäki A, Rannikko A, Holthöfer H, Ganna A, Sandholm N, Thorn LM, Groop PH, Tuomi T, Dwivedi OP, Puhka M

Publication: J Extracell Biol, 2025, Vol. 4, Page e70089

PubMed ID: 41098313 PubMed Review Paper? No

Purpose of Paper

Conclusion of Paper

Studies

1. Study Purpose

   This study compared the uEV morphology, protein expression, and RNA size among normoalbuminuric, microalbuminuric, and macroalbuminuric urine. Additionally, the paper aimed to identify microRNA (miRNA, miR) biomarkers of diabetes in urinary extracellular vesicle (uEVs) that met the following criteria: (i) stable for a given case when controlled for specimen type and time of collection (24 h versus overnight collections); (ii) comparable between two cohorts (24 h specimens frozen without centrifugation versus overnight specimens centrifuged prior to freezing); and (iii) comparable between RNA samples that were and were not treated with DNase.  Identified miRNAs were validated in multiple urine specimen cohorts, including those collected from males and females, and patients with type I and type II diabetes. Potential correlations with clinical measurements and the sensitivity for predicting eGFR decline were also investigated.  The discovery cohort (men) included 24 h and overnight urine specimens from males in the FinnDiane and DIREVA cohorts (7 Non-diabetic, 40 normoalbuminuric, 15 microalbuminuria, 19 macroalbuminuric), the type I diabetes replication cohort included 24 h specimens from females with type I diabetes in the FinnDiane cohort (23 normoalbuminuric, 4 microalbuminuric and 4 macroalbuminuric) and the type 2 diabetes replication cohort included overnight urine from males and female (14 with microalbuminuria and 9 with macroalbuminuria). All urine was stored at -80°C. uEVs were isolated from filtered urine by ultracentrifugation (100,000 g for 1.5 h) and analyzed by immune electron microscopy for CD9, CD81 and CD63 and nanoparticle tracking analysis. RNA was isolated with the miRNEasy mini (or micro for replication cohort) Kit, and a DNase treatment step was included for specimens from the type 1 diabetes replication cohort. Sequencing libraries were constructed with the Lexogen Small RNA-Seq Library Kit with no size selection for the discovery and type 2 diabetes replication cohorts; BluePippin size selection was performed for specimens in the type 1 diabetes replication cohort. Identified miRNAs were validated by real-time PCR.  To identify effects associated with collection method, miRNA levels in matched 24 h and overnight urine collected from 12 patients in the discovery cohort (8 with macroalbuminuria and 4 non-diabetic) on the same day were compared. To assess the effect of DNase treatment, uEVs from four female patients were divided and treated with DNase or left untreated. To better understand the effects of collection method and centrifugation, levels of miRNAs in uEVs from the FinnDiane (24 h specimens frozen without centrifugation) and DIREVA cohorts (overnight specimens, centrifuged before freezing) were compared.

   Summary of Findings:

   Urine from normoalbuminuric, microalbuminuric, and macroalbuminuric patients had 1.6×10⁸ to 3.1×10⁹ cup-shaped particles/mL, with a mean particle size of 100 nm. There were more particles that expressed the EV markers CD9 or CD63 than CD81; the diameter of particles expressing an EV marker was similar among the markers evaluated. While the percentages of particles expressing CD9, CD9 and CD81 or CD63, CD81 and CD9 were comparable between normoalbuminuric and albuminuric urine, normoalbuminuric urine had a higher percentage of particles that were positive for both CD9 and CD63 than albuminuric urine.  Isolated urinary EV RNA had a peak just below 200 nt..  Isolated urinary EV RNA had a peak just below 200 nt.

   The number of reads and mapped reads was unaffected by DNase treatment of uEVs. PCA based on the expression of the 50 highest expressed miRNAs did not separate specimens based on whether collection was an overnight versus 24 h collection, and only two miRNAs were found to be differentially expressed between overnight and 24 h specimens from the same patients (miR-375, P=6.01 x10^-5 and miR-200b-3p, P=0.021). In contrast, PCA based on the expression of the 50 highest expressed miRNAs did show some clustering of specimens from the FinnDiane (24 h specimens from men frozen without centrifugation) and DIREVA (overnight specimens from men, centrifuged before freezing) cohorts, but there was no clear separation. Further, 20 miRNAs were differentially expressed between the two cohorts, which were also differentially expressed when analysis was limited to normoalbuminuria specimens.

   The authors identified miRNAs that were unaffected by the collection methods evaluated and inclusion/exclusion of DNase treatment that were also differentially expressed between normoalbuminuric and albuminuric specimens from men with type I diabetes. Among the identified miRNAs, significant correlations were observed between miR-143-3p and eGFR, HbA1c and systolic blood pressure, miR-146a-5p and waist-hip ratio, miR-486-5p and eGFR, miR-196b-5p and waist-hip ratio, and miR-222-3p and diastolic blood pressure. In the validation cohort of specimens from women, 10 miRNAs were differentially expressed, including miR-192-5p, miR-146a-5p, miR-486-5p and miR-574-5p from the discovery cohort. While the 10 miRNAs that were differentially expressed in the validation cohort (women) separated normoalbuminuric specimens from the albuminuric specimens, the 11 differentially expressed miRNAs from the discovery cohort (men) did not properly separate specimens from the validation cohort.  Real-time PCR confirmed differential expression of miR-143-3p, miR-192-5p, miR-222-3p, miR-23b-3p and miR-24-3p between normoalbuminuric and albuminuric specimens in the discovery cohort and miR-192-5p in the validation cohort. The differentially expressed miRNAs did not cluster specimens by other disease states except for patients with chronic kidney disease.  Levels of miR-146a-5p, miR-192-5p, miR-486-5p, and miR-574-5p had an area under the curve (AUC) of >0.7 for predicting eGFR decline in one or more cohorts, and this was better than other clinical markers (HbA1c, systolic blood pressure). The combination of all four miRNAs had an AUC of 0.82 for  ypredicting eGFR decline and was 0.81 when eGFR, HbA1c, and systolic blood pressure were included. Pathways analysis found that the differentially regulated miRNAs targeted 54 mRNAs in 26 pathways, 16 of which were diabetic kidney disease (DKD)-associated.

    

   Biospecimens

   • Bodily Fluid - Urine

   Preservative Types

   • Frozen

   Diagnoses:

   • Diabetes Type 1
   • Diabetes Type 2
   • Normal

   Platform:

   Analyte	Technology Platform
   Protein	Electron microscopy
   Morphology	Electron microscopy
   RNA	Real-time qRT-PCR
   RNA	Northern blot
   Cell count/volume	Light scattering

   Pre-analytical Factors:

   Classification	Pre-analytical Factor	Value(s)
   Preaquisition	Patient gender	Female Male
   Preaquisition	Diagnosis/ patient condition	Normal Type I diabetes Type 2 diabetes
   Biospecimen Aliquots and Components	Centrifugation	Centrifuged Not centrifuged
   Biospecimen Acquisition	Method of fluid acquisition	Voided urine (24-h collection) Voided urine (overnight collection)
   Analyte Extraction and Purification	Nucleic acid digestion	DNase treated Not treated
   Preaquisition	Prognostic factor	Normoalbuminuric Microalbuminuric Macroalbuminuric

   View more