Establishment of Optimal Housekeeping Genes for Urinary Extracellular Vesicle Based Biomarker Development: A Step Towards Non-Invasive Diagnostics.

Recent years have witnessed a growing interest in employing urine as a clinical source of renal pathology biomarkers. Urinary extracellular vesicles(UEVs) hold cellular RNAs, including small RNA and micro RNA. Quantitative polymerase chain reaction (qPCR) is one of the most sensitive methods for evaluating gene expression, which depends on comparative analysis with reference/house-keeping genes. However, reliable interpretation of UEVs gene expression data is biased due to the lack of reported ideal house-keeping reference-genes in UEVs. UEVs were isolated from 40 healthy human controls using Polyethylene glycol (P.E.G. Mn6000) based precipitation. UEVs were characterized by biophysical and biochemical assays. At the molecular biology level, the expression and stability of ve commonly used housekeeping genes B2M, RPL13A, PPIA, HMBS, and GAPDH, were considered for comparing and nding out ideal reference gene. Data were analyzed using four practical algorithmic approaches, including Norm Finder, GeNorm, Best Keeper, and the Delta Ct for reference gene evaluation integrated with RefFinder. The nal ranking of stable genes is derived from the weighted geometric means of all the above algorithms. 12% PEG isolated UEVs were round and cup-shaped, ranging from 30 - 100nm, as per electron microscopy, nanoparticle-tracking-analysis, and dynamic-light-scattering prole. The functional purity of UEVs was determined with their acetylcholine esterase and Dipeptidyl peptidase-IV activity. RefFinder established the stability index of housekeeping genes. B2M and RPL13A genes were identied as stable genes with a mean stability score of 1.5(Genorm) and below 1 (Norm nder), indicating a reduced gene expression variation. The comprehensive ranking analysis identied B2M and RPL13A as optimal reference genes for UEVs based gene expression studies. 96% H2SO4 was added to 40µl of the lipid standard, DOPC (1,2-Dioleoyl-sn-glycero-3-phosphocholine), or the UEV samples and evaporated at 90°C on a dry bath for 20 min. After the tubes were cooled to RT, 120µl of Phosphovanillin reagent (200µg vanillin in 17% H3PO4) was added. The reaction is allowed to occur by incubating for 1 h at 37°C, and then absorbance was recorded at 540nm. a signicant subset of extracellular vesicles derived from the and urogenital tract 26). They hold a concentrated source of biomarkers (RNA, protein, and lipid), protected from urinary ribonucleases and proteases degradation(2). Easy access and non-invasiveness have made urine an ideal source for biomarker analysis. This study established a clinically adaptable protocol for UEVs isolation and characterization and dened a normalization strategy for UEVs based gene expression studies. Hydroxymethylbilane synthase, NTA Nanoparticle tracking analysis, DLS Dynamic scattering.

pairwise comparison, the Delta Ct method assigns stability values based on C q standard deviation differences (19). RefFinder (https://www.heartcure.com.au/re nder/); a web-based tool that evaluates and screens reference genes from experimental data sets integrating the weighted average from the above four algorithms. This study established an optimal UEVs isolation protocol and identi ed the most stable housekeeping genes among the ve conventional reference genes from the gene pool of UEVs.

Materials And Methods
Urine collection and processing Urine samples from forty healthy individuals (aged 20-72 years) were collected from Apollo Hospital, Hyderabad. Demographic characteristics are represented in Table 1. Female 5 Table 1: Urine samples were collected from 40 healthy individuals whose age is represented as Mean ± SD.
Urine was collected in collection containers, and 4.2ml protease inhibitor (1mg/ml Leupeptin (G Biosciences, MO, USA), 10mM Sodium Azide, 50mM PMSF ( both from Sigma, MO, USA) per 50ml of urine was added. Before proceeding with UEV isolation, all urine samples were subjected to urine tests using the urinalysis strips (Siemens Multistix 10 SG, Siemens, Munich, Germany). The urinalysis strips were dipped into the urine containing tube, and excess urine was drained. The colour changes on the strips were observed and quanti ed using Siemens instrument CLINITEK Advantus for the parameters shown in Table 2. The collected urine samples were immediately processed for UEV isolation using the modi ed salt precipitation method described previously (5). Brie y, 30-50 ml of rst void urine samples were centrifuged at 1800*g for 10 min at 4°C to clarify the cell and cellular debris. The cell pellet was washed with 1x PBS (Hyclone GE Healthcare, IL, USA) and resuspended in RNA later solution (Sigma) for RNA isolation. The supernatant was carefully transferred to a fresh tube containing 20mMTris-EDTA (Sigma) pH 6.8, and the pH was adjusted to 4. The supernatant mixture was vortexed for 90sec and further centrifuged at 8000*g for 15min at 4°C. After the centrifugation step, the supernatant was collected in a fresh tube, an equal volume of 2x PEG solution was added ( nal 12% PEG in 0.5M NaCl (Sisco Research Laboratories (SRL), Mumbai, India) and mixed thoroughly. The UEV pellet obtained was then treated with 500µl of 100mM DTT (Sigma) and incubated for 10 min at 37°C for Tamm-Horsfall Glycoprotein (THP) removal. Subsequently, the DTT treated pellet was centrifuged at 17000*g for 15 min at 4°C with zero brakes. The supernatant obtained here was added to the previously derived PEG-supernatant mixture and incubated overnight at 4°C. The next day, the PEG-supernatant mixture was centrifuged at 10,000*g for 60 min at 4°C. Finally, the pellet containing UEVs was resuspended in 1x PBS and stored at -80°C for further use.

Urinary Extracellular Vesicle Characterisation
The isolated UEVs were subjected to biophysical and biochemical characterization to ensure the quality of isolated vesicles.

Morphological characterization TEM Analysis
Transmission electron microscopic analysis was performed to examine the size and con rm the presence of intact exosomes and size determination.
The grid was washed twice with water, and further stained with 2% uranyl acetate, and then air-dried at room temperature. Images of the isolated UEVs were captured using a transmission electron microscope (JEM-2100, JEOL Ltd. Tokyo, Japan).

Nanoparticle Tracking Analysis
Nanoparticle tracking analysis is a powerful technique that utilizes laser light scattering microscopy combined with Brownian motion to determine the size, concentration, and morphology of extracellular vesicles [45]. We performed NTA on the isolated UEVs using NanoSight LM10 instrument (NanoSight, Amesbury, UK), where the laser light-scattering was measured at 488 nm, and the relative concentration and size was determined.

Dynamic Light Scattering
The size of UEVs was further determined using the dynamic light scattering (DLS) technique that measures the size based on the Brownian motion of the dispersed particle as described previously [46]. 3ml homogenous suspension of UEVs (diluted 1:100 in 1X PBS) was transferred to a cuvette, and the hydrodynamic diameter of exosomes was measured in a DLS instrument (Nicomp Z3000, Entegris, MA, USA).

Biochemical Characterisation
Protein estimation and SDS-PAGE analysis The total protein content in isolated UEVs was quanti ed using a bicinchoninic (BCA) protein assay kit (G-Biosciences) following the manufacturer's instructions. Samples were stored at -80°C until further analysis. For SDS-PAGE, 10µg of UEV protein was lysed using RIPA lysis and extraction buffer (G-Biosciences) including protease inhibitor cocktail (Roche, Basel, Switzerland) followed by incubation at 4°C for 15min. The lysed protein samples were then mixed with reducing sample buffer and denatured for 10 mins at 70°C. The exosomal proteins were resolved on 10% SDS PAGE for 1.5 h at 120 V, and the gels were stained using 2% silver nitrate.

Lipid Estimation
Total lipid content of the UEVs was quanti ed using Phospho Vanillin Assay, as reported previously (35). Brie y, 200µl of 96% H2SO4 was added to 40µl of the lipid standard, DOPC (1,2-Dioleoyl-sn-glycero-3-phosphocholine), or the UEV samples and evaporated at 90°C on a dry bath for 20 min. After the tubes were cooled to RT, 120µl of Phosphovanillin reagent (200µg vanillin in 17% H3PO4) was added. The reaction is allowed to occur by incubating for 1 h at 37°C, and then absorbance was recorded at 540nm.

AChE activity
Acetylcholinesterase (AChE) activity assay was carried out to con rm the presence of acetylcholinesterase (Sigma), which is considered as a marker enzyme for exosomes [21].20ul UEV fraction was added to 96-well at-bottomed microplate. Acetylthiocholine iodide ( nal 1.25mM) and 5', 5'-dithio-bis (2nitrobezoic acid) ( nal 0.1mM) (both from Sigma) was then added to each well at a nal volume of 300ul. The absorbance was recorded at 412nm every 5 min for 30 min. The amount of AChE activity in the exosome fraction was determined from the AChE enzyme's standard curve.

Dipeptidyl peptidase-IV activity
Dipeptidyl peptidase-IV (DPPIV), a membrane-associated peptidase, is associated with kidney diseases and is secreted by tubular epithelial cells in the kidney. Hence, it is considered to be a component of urinary microvesicles in the urine. We used a colorimetric assay previously used to determine serum DPPIV activity by quantifying DPPIV in UEVs as additional criteria to assess exosome quality [31]. Brie y, 10µl of UEV sample was mixed with 50µl of 71mmol/l glycine/NaOH (pH 8.3) buffer in 96 well plate. 50µl of 0.5mg/ml substrate Gly-Pro-p-nitroanilide (Sigma) was added to all wells, including blank wells, and incubated for 60 min at 37°C. The substrate is cleaved by DPPIV in the sample which releases free 4-nitroaniline, a chromogenic, whose absorbance is measured at 405nm in a plate reader. DPPIV activity in the sample was calculated against the standard plot of p-nitroaniline standard (Sigma).

Antibody Array
Exo-Check exosome antibody array (SBI, Systems Biosciences, USA ) was performed according to the manufacturer's instructions. The array contains 8 antibodies for known exosome markers including (CD63, CD81, ALIX, FLOT1, ICAM1, EpCam, AnXA5, and TSG101) and 4 controls, including two positive controls, blank and gm130 cis-Golgi marker which monitors for any cellular contamination. Brie y, 100µg of UEV protein was treated with lysis buffer; 1µl of Labelling reagent was then added and incubated at RT for 30mins with constant mixing. The column provided in the kit was equilibrated with column buffer before the sample was eluted. The eluted lysate was mixed with blocking buffer, transferred on to membrane, and incubated at 4°C overnight on a rocker.
The next day, the membrane was washed thrice (5 mins each) with washing buffer and incubated with detection buffer for 30mins at RT on a shaker. The membrane was washed thrice again with washing buffer before development. Super Signal West Femto Maximum sensitivity substrate (Thermo Fisher Scienti c, US) was used for chemiluminescence detection of the membrane in the imaging system (ChemiDoc, BioRad, USA).

RNA isolation and quanti cation
Total RNA from UEV was isolated using TRIzol™ LS Reagent (Invitrogen, California, USA) following the manufacturer's instructions. Brie y, the exosome sample was mixed with Trizol Reagent (1:3) and incubated for 5 min at RT. Chloroform was added (1:5) to the exosomal mixture and incubated for 15 min at RT, followed by centrifugation at 12000g for 15 min at 4˚C. The aqueous phase was carefully removed and transferred to a fresh tube. To all the tubes equal volume of isopropanol, 500µg/ml Glycogen (GlycoBlue™ co-precipitant Invitrogen), and 0.5µg yeast tRNA (Invitrogen) was added and incubated overnight at -20˚C. The samples were then centrifuged at 12000g for 15 min at 4˚C, followed by an ethanol wash. Finally, the air-dried RNA pellet was resolved in nuclease-free water and stored at -80°C for further analysis. The RNA content was determined using Qubit™ RNA HS Assay Kit (Invitrogen). Additionally, the UEV RNA quality and quantity were analysed in Agilent Bioanalyzer 2.1 instrument using RNA Pico kit for high sensitivity analysis of total RNA and mRNA samples (Agilent Technologies, California, USA).

Reverse Transcription
Exosomal RNA was reverse transcribed with High capacity cDNA Reverse Transcription kit (Invitrogen) using 10µL exosomal RNA and 10µl cDNA master mix prepared according to the manufacturer's instruction. cDNA was pre-ampli ed using Sapphire Amp fast PCR master mix (Takara Bio Inc. Shiga Prefecture, Japan) Pre-ampli cation with pooled primers was carried out for 20 cycles at 95˚C for 5 min, 95˚C for 1 min, 60˚C for 30 sec, 72˚C for 1 min, and 72˚C for 5 min and hold 4˚C. The pre-ampli ed transcript was then used as a template for PCR and RT PCR analysis.

Endogenous gene expression in Urinary EV
In the present study, we selected 5 reference genes, namely GAPDH, B2M, RPL13A, PPIA, and HMBS, for normalization of quantitative real time PCR in UEV samples from healthy individuals. All primer sequences were custom synthesized by the supplier (Bio serve, India), a detail of which is listed in Table 3. The pre-ampli ed samples mentioned in the above section were diluted 1:5 with nuclease-free water, and 1µl of the diluted product was used as a template for semi-quantitative as well as quantitative PCR. Brie y, a 10µl reaction was performed in Applied Biosystem 7500 Real-Time System using 5µl TB Green Premix Ex Taq (Takara Bio Inc.), 0.1µl 10µM primer pair, 1µl template cDNA, and 3.8µl nuclease-free water. Real-time PCR was conducted for 30s at 95˚C, followed by 35 cycles of 5s at 95˚C and 30s at 60˚C. All samples were evaluated in duplicates, and for every gene analyzed, a non-template control (NTC) was also included. PCR product purity was monitored from melting curve analysis and 2.0% agarose gel electrophoresis. Normalization of housekeeping genes To analyze the candidate gene expression stability, normalization analysis was performed using online available software RefFinder. RefFinder determines the best stably expressed candidate reference gene by generating a comprehensive ranking for each gene based on the combined expression stability data from four statistical algorithms-Genorm, Norm nder, Bestkeeper, and Delta Ct. These mathematical algorithms use threshold cycle (Ct) values to calculate expression stability of each candidate reference gene. Precisely, Norm nder implements an ANOVA based algorithm to determine inter and intragroup variations in the reference gene expression, which is represented in terms of stability value. The gene expression stability score in Genorm is determined by pairwise comparison, while Bestkeeper applies pairwise correlation analysis. Delta Ct method calculates variations in delta Ct by comparing relative transcription of pairs of genes. Ranking of genes was done based on the stability score, where genes with the lowest stability score were considered more stable.

Statistics
For the comparative analysis of exosomal characteristics using the PEG vs. Kit method, a paired student's t-test was done to test the statistical signi cance using GraphPad software version 8.3 was used. Only P < 0.05 was considered statistically signi cant. All experimental data are shown as mean ± SEM unless otherwise mentioned. Gene stability analysis of housekeeping genes was performed in RefFinder. The stability value or SD data obtained using different algorithms were used for the ranking of genes.

Results
Urinary EV isolation from healthy subjects UEVs were isolated from forty healthy individuals using the PEG-based precipitation method described previously(8) with modi cations. For reference, we performed a comparative analysis of the UEVs isolated using a commercially available kit (n=11). An overview of the experimental procedure is shown in Figure 1. We further determined the characteristic features of the isolated UEVs for their biochemical, biophysical, and molecular characteristics.
Optimization of the exosome isolation procedure Initially, we had optimized the UEV isolation method by altering PEG concentrations, i.e. (6%, 12%, 24%). The total protein content and yield were maximum using 12% PEG compared to 6% or 24% (Supplementary Figure 1). To con rm the purity of the isolated EVs, we have compared their biochemical properties with those isolated, using the standard commercially available kit as a reference. Quantitative analysis of the UEV protein and lipid content showed moderate differences in the yield, Figure 2A-B. The ratio of protein to lipid content, a commonly used parameter to assess the quality of vesicles (20), was comparable ( Figure 2C). Also, acetylcholine esterase (AChE) enzyme activity was quanti ed in the UEVs to assess the purity of isolation as previously reported (21). AChE enzyme activity of the PEG-isolated UEVs was slightly lower (1.7-fold) when compared to the Kit -method ( Figure 2D). As the biochemical properties of the vesicle-derived using both methods were comparable, we chose to isolate using the PEG-based method further.

Qualitative assessment of Urinary EVs
Following UEVs isolation, THP contaminant proteins were removed by treating the fraction with reducing agent DTT (22). SDS-PAGE analysis con rmed the reduction of THP protein in the DTT treated fraction compared to the untreated fraction, Figure 3A. THP removal from the UEV fractions enhanced the yield by 1.5-fold when compared to the unremoved fractions ( Figure 3B, n=3). Hence, 12% PEG-based isolation following DTT treatment was considered optimal for UEVs isolation. Following this procedure, UEVs were isolated from healthy individuals (n=12) and were characterized as described in the methods section. The exosomes' average total protein and lipid content was 646µg/ml and 1240µg/ml, respectively, as shown in Figure 3C. A semi-quantitative exosome array was performed, Figure 3D, results showed intense positive staining for various genes including in the exosome array -Intercellular Adhesion Molecule 1 (ICAM), Tumor Susceptibility Gene 101 (TSG101), and tetraspanins CD63 and CD81. A weak signal was observed for Flotillin 1 (FLOT1) and programmed cell death 6 interacting protein (ALIX) and Annexin A5 (ANXA5), while no signal was detected for Epithelial Cell Adhesion Molecule (EpCAM). cis-Golgi matrix protein (GM130) was employed as a cellular protein contamination spot and negative control (blank spot). The quality of UEVs was assessed from the ratio of protein to lipid content (P/L), which ranged from 0.3-1.6, Figure 3E, indicating the integrity of extracellular vesicles. The average total UEV yield was 2.45 mg per 100 ml of urine. The average AChE activity was 550mU per mg of protein, con rming the purity, as shown in Figure 3F. Furthermore, DDPIV activity in the isolate was quanti ed in Figure 3G to verify that the activity was indeed in the microvesicular components of urine (22). These observations con rmed the quality of UEVs isolated from urine.

Morphology and size distribution of Urinary Exosomes
UEVs appeared round cup-shaped when visualized under the transmission electron microscope (Figure 4 A). Nanoparticle tracking (NTA) analysis revealed the size distribution pro le of UEV with a mean particle size of 59 ± 22nm (Figure 4 B). The concentration of the UEV particles was 3.19 E8 particles per ml as quanti ed by NTA. Also, the size was determined from the intensity distribution calculated by DLS (Figure 4 C). Results of DLS measurements are shown as a number weighted distribution curve. At least three independent aliquots were measured in triplicate. The average size of the UEVs was 78 ± 56 nm based on the number distribution. Thus, using a combination of biophysical parameters, we veri ed that our isolated UEV fractions were in the reported size range and morphology of exosomes (23).

Selection of optimal housekeeping genes
Urinary EV RNA was extracted and quanti ed using the Agilent Pico RNA detection kit. The RNA was quanti ed to be 137pg/µl, n=6. The quality of isolated RNA was assessed against a standard marker; a sharp 5S peak was identi ed in between 25-200 nucleotides in RNA from UEV fractions, which con rmed the presence of exosome RNA, as shown in Figure 5A. There were no detectable 18S or 28S peaks. An overlay of urinary cell pellet RNA with the exosomal RNA shows degraded cellular RNA in the pellet RNA (24).
RNA was extracted from UEVs isolated from forty healthy urine samples, and the expression levels of ve selected reference genes were quanti ed using qPCR. The expression levels of all genes were normally distributed, as shown in Figure 5B, with an average Ct value of 10.26 ± 1.33. To identify the most stably expressed housekeeping genes, we have normalized the gene expression levels using the online software RefFinder. Each gene's stability score was determined using different tools included in the software such as Genorm, Norm nder, Bestkeeper, and the comparative delta Ct method to minimize the error associated with a single software-based evaluation. As different statistical algorithms were employed to evaluate the stability score, the ranking of genes slightly varied in different tools (details listed in Table 5).
Nevertheless, B2M and RPL13 A were identi ed to be consistently stable with a lower mean stability score below 1.5 (Genorm) and below 1 (Norm nder), indicating a reduced variation of the expressed genes shown in Figure 5C. The recommended comprehensive ranking values for the average gene expression data (n=40) showed a similar trend with B2M and RPL13A in top-ranking positions ( Figure 5D). The above two genes may be used as the most stable housekeeping genes for accurate gene expression analysis, particularly in urine samples. [r] Pearson correlation coe cient, SD-standard deviation

Discussion
UEVs constitute a signi cant subset of extracellular vesicles derived from the kidney and urogenital tract (25,26). They hold a concentrated source of biomarkers (RNA, protein, and lipid), protected from urinary ribonucleases and proteases degradation (2). Easy access and non-invasiveness have made urine an ideal source for biomarker analysis. This study established a clinically adaptable protocol for UEVs isolation and characterization and de ned a normalization strategy for UEVs based gene expression studies.
Several techniques, including ultracentrifugation, ultra-ltration, size exclusion chromatography, and precipitation, are currently used for the isolation of UEVs (23). In this study, owing to the lower cost and easy adaptability of PEG-based precipitation of UEVs in clinical laboratories, we optimized this procedure with modi cations to isolate EVs from urine (8). Comparison of biochemical properties like total protein, lipid and P/L ratio, and acetylcholine esterase between exosomal preparations obtained by PEG-based precipitation and commercial kit showed similar yields. Therefore, our results emphasize that PEG-based UEVs precipitation can be used in clinical labs for microvesicle-based downstream applications.
Reduced vesicle load and the presence of contaminant protein Tamm-horse fall protein (THP) or uromodulin in the urine leads to reduced UEVs yields. THP is a protein that is abundantly found in urine under physiological conditions. It is known to oligomerize into long polymers and compromises the vesicular yield at lower temperatures by forming a mesh that entraps the vesicles (22). It is also often a common contaminant that co-precipitates out along with exosomes (27). A previous study emphasized employing a DTT treatment step to release THP-entrapped exosomes (28). Similarly, we observed a decrease of THP protein density and an increase in the microvesicular yield after treating the fraction with dithiothreitol (DTT), suggesting a dissociation of THP mesh entrapping UEVs, as reported earlier (8).
Protein/Lipid ratio represents pure exosome preparations and can also be used as one of the quality control measures for distinguishing soluble proteins and protein aggregates in the EVs preparations (20). Acetyl-CoA choline esterase (AChE) was shown to be concentrated in exosomes and can thus be used to measure the quality of exosome preparations. (21,23,29). Thus, to characterize our isolated UEV fractions' purity, we measured both the total protein content and lipid content by using BCA kit and Sulpho-phospho-vanilin assay, respectively (30). Dipeptidyl peptidase-IV (DPP IV) is another pan membraneassociated peptidase, widely expressed in tissues and is also reported in the kidney cortex and the brush border and microvillus fraction (31). Both AChE and DPP IV are used to measure the bioactivity, purity, and functional activity of exosome preparations (32). The protein determination of the exosome antibody array con rmed the purity of vesicular preparation (33,34). Tetraspanins CD63 and CD81, along with TSG101 and ALIX are pan exosome markers present on urinary exosomes (35,36). Generally, the majority of vesicles present in urine are believed to be exosomes (37).
Biophysical characterization involving DLS, NTA, and EM estimates quality check of exosome preparations (38). In this study, we also used Transmission Electron Microscopy to characterize the morphological structure of the UEV fraction isolated. The size pro le by DLS and NTA were in accord with the performance of exosomes(3). Thus, the above three methods estimate size, shape, morphology and agree with the published literature(3).
The need for ensuring quality assurance measures for qPCR is well acknowledged. MIQE guidelines provide an operational framework for qPCR-based gene expression studies (39). The above guidelines enable an e cient design of qPCR experiments, ascertain technical quality of gene expression studies, and reliable replication of experiments. For accurate interpretation of qPCR results, data normalization is a prerequisite step (19). Currently, there is a de cit of reliable housekeeping genes for conducting UEVs gene expression studies. Lack of ideal housekeeping genes leads to misinterpretation of nal results.
Candidate reference genes analyzed showed a narrow Cq range among all experimental series ranging from 5 to 15. According to Genorm, the candidate genes with the lowest stability values are B2M (1.1) and RP13A (1.1) in the study. Based on Norm nder analysis, B2M (0.5) and RP13A (1.0) were top stable candidates. Genes with stable expression are indicated by low average expression stability values by Norm nder (40). B2M (1.7) and RP13A (1.5) can be considered as stable gene candidates. Best keeper calculates the CP standard deviation (SD) and coe cient of variation (CV) for each gene (41). Genes with SD values <1 are considered stable and are categorized as reference genes.
RefFinder integrates the above four programs' outcomes and derives the most stable gene pro le (42). RefFinder analysis reduced the individual bias of each software and yielded a cumulative score. Absolute ranking values of B2M (1.2) and RP13A (1.6) de ned them as universal housekeeping genes in UEVs gene expression studies. Following the recent guidelines, two or more genes have to be used at arriving at a normalized score (43).
Taken together, we have developed an easily adoptable lab protocol for isolation and characterization of UEVs. We established a quality control pro le to characterize the urinary EVs preparations at the biochemical, biophysical, and molecular levels. To our knowledge, this is the rst study reporting stable housekeeping genes in UEVs. The optimal reference genes derived from this study could provide meaningful choices for target gene expression and functional studies based on UEVs, thereby strengthening UEVs applications, including diagnostics and biomarker studies.