Characterization and deep sequencing analysis of … Urine Exosome RNA Kit in Kidne… · Characterization and deep sequencing analysis of ... exosomal isolation and RNA yield for

Characterization and deep sequencing analysis ofexosomal and non-exosomal miRNA in human urineLesley Cheng1,2,3, Xin Sun1,2,3, Benjamin J. Scicluna1,2, Bradley M. Coleman1,2 and Andrew F. Hill1,2

1Department of Biochemistry and Molecular Biology, University of Melbourne, Melbourne, Victoria, Australia and 2Bio21 Molecular Scienceand Biotechnology Institute, Department of Biochemistry and Molecular Biology, University of Melbourne, Melbourne, Victoria, Australia

Micro RNAs (miRNAs) have been shown to circulate in

biological fluids and are enclosed in vesicles such as

exosomes; they are present in urine and represent a

noninvasive methodology to detect biomarkers for

diagnostic testing. The low abundance of RNA in urine

creates difficulties in its isolation, of which exosomal miRNA

is a small fraction, making downstream RNA assays

challenging. Here, we investigate methods to maximize

exosomal isolation and RNA yield for next-generation deep

sequencing. Upon characterizing exosomal proteins and total

RNA content in urine, several commercially available kits

were tested for their RNA extraction efficiency. We

subsequently used the methods with the highest miRNA

content to profile baseline miRNA expression using next-

generation deep sequencing. Comparisons of miRNA profiles

were also made with exosomes isolated by differential

ultracentrifugation methodology and a commercially

available column-based protocol. Overall, miRNAs were

found to be significantly enriched and intact in urine-derived

exosomes compared with cell-free urine. The presence of

other noncoding RNAs such as small nuclear and small

nucleolar RNA in the exosomes, in addition to coding

sequences related to kidney and bladder conditions, was also

detected. Our study extensively characterizes the RNA

content of exosomes isolated from urine, providing the

potential to identify miRNA biomarkers in human urine.

Kidney International advance online publication, 18 December 2013;

doi:10.1038/ki.2013.502

KEYWORDS: deep sequencing; exosomes; microRNA; urine

There is increased interest in detecting extracellular micro-RNA (miRNA) from biological fluids to identify biomarkersfor disease. miRNAs are a class of small noncoding RNAs(ncRNAs) that are 22–25 nucleotides long, which function toregulate mRNA processing at the transcriptional and post-transcriptional level. They are derived from mRNA hairpinscomprising precursor miRNAs that are further processedby endoribonucleases (Dicer and Drosha) to form maturemiRNA. The mature miRNA is incorporated into the RNA-induced silencing complex, which binds to complementarysites in the 30 untranslated region of their mRNA targets,resulting in the downregulation of gene expression.1

miRNA originating from specific tissues can be secretedinto the extracellular environment, including biological fluidssuch as urine.2 Urine is a sterile biological fluid comprisingend products generated by protein metabolism that issecreted by the kidneys and can be collected noninvasivelyand in a simple manner. The majority of urinary miRNAoriginates from renal and urethral cells, and analysis of thesecells can provide a measure of the health of the excretorysystem.3–5 Circulating extracellular miRNA from other tissueswithin the body can be delivered to renal epithelial cells andreleased into the urine bound to RNA-binding proteins6,7 orpackaged into microvesicles such as exosomes.8

Exosomes are 40- to 100-nm-diameter vesicles released bycells and are formed within multivesicular bodies in theendosomal system.9 They have been found to be enriched inRNA, including coding RNA and ncRNA species such asmiRNA, small nuclear RNA, transfer RNA, ribosomal RNA,and long intergenic RNA.8,10 The delivery of large clusters ofmiRNA in exosomes has the capacity to influence a largerdiversity of genes and regulate biological pathways. Profiles ofderegulated miRNA secreted into peripheral blood11 andserum12–14 have been generated and suggest that they havediagnostic potential for human disease such as gliobastomaand ovarian cancer.12,13

Currently, there are limited methodologies available toefficiently isolate extracellular miRNA from the cell-freecomponent of urine. Recent advances in next-generationsequencing (NGS) technologies have allowed the ability toprofile total RNA transcriptomes. The challenges faced usingNGS to profile small RNA transcriptomes in urine includeisolation of total RNA from urine, of which small RNAs make

http://www.kidney-international.org t e c h n i c a l n o t e s

& 2013 International Society of Nephrology

Correspondence: Andrew F. Hill, Department of Biochemistry and Molecular

Biology, Bio21 Molecular Science and Biotechnology Institute, University of

Melbourne, 30 Flemington Road, Melbourne, VIC 3010, Australia.

E-mail: [email protected]

3These authors contributed equally to this work.

Received 2 August 2013; revised 9 September 2013; accepted 3 October

2013

Kidney International 1

http://dx.doi.org/10.1038/ki.2013.502

http://www.kidney-international.org

mailto:[email protected]

up a fraction, in addition to the added difficulty of isolatingexosomal miRNA.

Here we have systematically characterized exosomal RNAprofiles in human urine. We investigated methodologies toobtain high exosomal yields from minimal urine volumesand tested RNA isolation efficiency by using six commerciallyavailable extraction kits in order to maximize RNA yieldsfrom these RNA-limited samples. These methods were thenanalyzed using NGS to profile baseline miRNA and otherncRNAs by small RNA deep sequencing. We present amethod that is suitable for analyzing circulating miRNAprofiles in urine that can be applied to biomarker discoverystudies in diseases affecting the kidney and bladder, such asurothelial and renal cell carcinoma.

RESULTSIsolation and characterization of exosomes isolated fromhuman urine

An optimized differential ultracentrifugation protocol wasestablished to isolate exosomes from urine samples.15,16

Tamm–Horsfall protein (THP) polymerization can lowerthe yield of exosomes by entrapping exosomes within theirprotein network.15 Dithiothreitol (DTT) has been used inproteomic studies to depolymerize disulphide bonds of THP,releasing entrapped exosomes that pellet upon re-ultracentri-fugation and improving the detection of exosomal proteins.15,17

To investigate the requirement of THP depletion from urinefor transcriptomic studies, exosome pellets were treated withor without DTT and re-ultracentrifugation (Figure 1).

Exosomes isolated from urine treated with DTT yielded aslight increase of exosomes compared with those treatedwithout DTT, as characterized by western immunoblotting ofexosomal markers (Figure 2a). Cell lysates and exosomesfrom SH-SY5Y cells were used as a positive control given thatthe methodology to isolate exosomes from cell line super-natant is well established (Figure 2a).18,19 Furthermore, nosignificant difference was observed upon comparing the yieldof RNA content extracted from exosomes treated with orwithout DTT (Figure 2b). To directly visualize the exosomes,transmission electron microscopy (TEM) was used. Vesiclesisolated by differential ultracentrifugation were found to havea diameter of 50–100 nm and morphology consistent withprevious reports (Figure 2c).20,21 Treatment with DTT signi-ficantly reduced the presence of the THP fibril network, asobserved in the TEM images (Figure 2c). We furthercharacterized the sizes of vesicles from cell-free urine andvesicles isolated by ultracentrifugation via the particle-to-particle qNano nanopore-based instrument. The majority ofvesicle particles were also found to be between 50 and 100 nmin diameter (Figure 2d). Given that the treatment with DTTdid not significantly improve exosomal yields (Figure 2a)or affect RNA extraction (Figure 2b), we considered thistreatment to be unnecessary for this transcriptomic study.

Further characterization of exosomes isolated withoutDTT treatment by western immunoblotting of exosomalmarkers (Figure 3a) and non-exosomal markers (nucleopor-

in, GM130, and Bcl-2) were observed to be highly enrichedupon differential ultracentrifugation compared with thevarious fractions collected during the procedure (wholeurine, cell pellet, cell-free urine, and exosome-depletedsupernatant, Figure 3a). In addition, upon analyzing thesmall RNA species (o200 nt) extracted from individualcomponents of urine, small RNA was found to be enriched inthe exosomal pellets (Figure 3b). Although the cell pelletobtained from whole urine contained significant amounts ofsmall RNA, the majority was found to be degraded RNA(Figure 3c). Large RNA species (4200 nt), in particular 18Sand 28S ribosomal RNA, were not detected in the cell pellet,as determined by Bioanalyser analysis using an RNA Nanochip (RNA integrity number¼ 0, Figure 3c), suggestingdegradation of the cellular RNA.

Maximization of RNA recovery from urine samples

A systematic comparison between different RNA extractionmethods was performed on urine samples uniformly pooledfrom four healthy control subjects (Supplementary Table S1online) to obtain sufficient volume for three independent

Whole urine sample (>20ml)

Vortex 90 sec2000 x g,

10 min

17,000 x g,45 min

200,000 x g,65 min

DTT treatment200,000 x g,

65 min

Cell pellet

Pellet(large MV’s)

Exosome-depletedsupernatant (SN)

Supernatant

Cell-free urine

Supernatant

Exosomes (+ THP)

Exosomes (– THP)

Figure 1 | Work flow representing the isolation of exosomes fromurine by differential ultracentrifugation. The steps indicated withblack arrows involve isolating exosomes without dithiothreitol(DTT) treatment (� ). The work flow indicated by the gray arrowsshows exosomes treated with DTT to remove Tamm–Horsfall protein(THP, þ ).

t e c h n i c a l n o t e s L Cheng et al.: Characterization of exosomal miRNA in urine

2 Kidney International

2.5

2

1.5

ng R

NA

/ml u

rine

1

+ THPSY5YUrine

49

62

49

62+ – Exo Cell

Tsg101

Flotillin

DTTkDa

– THP

DTT –

DTT +

16

14

12

10

% P

opul

atio

n (b

y co

unt)

8

6

60 120 180Particle diameter (nm)

240

UC exosomes

Cell-free urine

3000

4

2

0

NS

0.5

0

Scale bars = 200 nm

200 nm 200 nm 200 nm

200 nm200 nm50 nm

Figure 2 | The characterization of exosomes treated with dithiothreitol (DTT) to deplete Tamm–Horsfall protein (THP). (a) Westernimmunoblotting of exosomes isolated by ultracentrifugation using Tsg101 and flotillin. Exosomes from SH-SY5Y cells and total cell lysates wereloaded as positive controls. (b) Exosomal micro RNA (miRNA) yields quantified by Bioanalyser small RNA assay. The result displayed is the meanvalue across three independent experiments performed on the same urine sample. Error bars indicate±s.e.m. NS¼ no significant difference(P¼ 0.18 in a paired two-tail Student’s t-test). (c) Transmission electron microscope (TEM) images of exosome isolations. Arrows indicate thelocations of exosomes within the protein network of THP. Bars¼ 200 nm, as indicated in the images. Samples were treated with DTT (þ ) orwithout DTT (� ). (d) Size distribution of exosomes analyzed by the qNano instrument. Exosomes are untreated with DTT.


L Cheng et al.: Characterization of exosomal miRNA in urine t e c h n i c a l n o t e s

experiments (Figure 4). Six commercial RNA isolation kitswere tested using the pooled urine samples. The kitscompared in this section were miRNeasy (Qiagen), miRNeasywith RNeasy MinElute Cleanup Kit (Qiagen), mirVanaPARIS (Ambion), Trizol LS reagent (Life Technologies) withmirVana (Ambion), miRCURY (Exiqon), and the Urine ExosomeRNA Isolation kit (Norgen Biotek). Several modifications ofRNA extraction were investigated, including phenol/chloro-form extraction and further miRNA enrichment, to deter-mine the best methodology for maximal RNA yield.

With the exception of the Norgen Urine ExosomeIsolation kit, the remaining kits involve the isolation ofexosomes by differential ultracentrifugation followed by RNAextraction. The majority of these kits use guanidine andphenol extraction methods owing to their combined highefficiency in separating protein and RNA into an interphaseand aqueous phase, respectively. One exception is themirVana kit (Life Technologies), which supplies a phenol-free lysis reagent. Therefore, a minor modification of themirVana kit was also tested whereby the manufacturer’s lysis/binding reagent was replaced with Trizol LS reagent. Inaddition, enrichment and separation of miRNA (18–200 nt)from total RNA (4200 nt) was also investigated by using themiRNeasy kit and the RNeasy MinElute Cleanup kit. TheUrine Exosome RNA Isolation kit from Norgen Biotek is acentrifugation-independent kit that isolates exosomes bybinding urinary exosomes to a proprietary resin andenriching exosomal RNA by lysing the bound exosomes.The RNA is released from the exosomes, which is transferredto a column and contaminants are then removed by washesbefore the elution of exosomal RNA. We evaluated theNorgen Biotek kit to determine whether a similar result couldbe achieved for deep sequencing without using the lengthyultracentrifugation procedure.

RNA extractions were analyzed using a Bioanalyser withthe small RNA assays normalized to ng miRNA per ml ofurine (Figure 5). The Norgen Biotek kit isolated more than afourfold increase of small RNA compared with the otherisolation kits (Po0.001, Figure 5a). Among the five kitsdependent on the use of differential ultracentrifugation to

kDa

Urine SY5Y

ExoExo

Exo

Cell

Cell

WUWU

20100

0100

200

0

4 20 40

40

80 150 [nt]

4 20 40 80 150 [nt]

4 20

25 200 1000 4000

40 80 150 [nt]

[nt]

4

4

20

20

40 80

10060

150

150

[nt]

[nt]

CF

CF

SNSN

Cell

Cell

Tsg101

Flotillin

CD63

Nucleo-porin

GM130

Bcl-2

62

49

62

49

62

49

3828

62

49

98

28

[FU]

[FU]

[FU]

[FU]

[FU]

20

0

40

20

0

4020

0

105

[FU]

Figure 3 | Characterization of urinary exosomes by western immunoblotting. (a) Whole urine (WU), cell pellet lysates (Cell), cell-free urine(CF), exosome-depleted supernatant of the 200,000 g spin (SN), and exosomes (Exo) (left to right) from the same urine sample were analyzed byimmunoblotting with antibodies against exosomal proteins Tsg101, flotillin, and CD63 (under non-reducing conditions), and non-exosomalproteins nucleoporin, GM130, and Bcl2. Exosomes and whole-cell lysates from SH-SY5Y cells were loaded as positive control samples.(b) Bioanalyser chronographs of small RNA profiles obtained from the various urine components as above. (c) Total RNA profile of cell pelletsfrom urine, as determined by an RNA Nano chip.

Pooled urine

5 ml

NG-Exo RNA prep UC-Exo prep

Exosomeisolation

Keep cell pelletand 2.5 ml cell-free urine miRNeasy kit

miRNeasy - RNeasy kit

mirVana kit

TRIzol LS + mirVana kit

miRCURY kit

Agilent Bioanalysersmall RNA assay

Samples weredivided equally andRNA was extracted

95 ml

Figure 4 | Study design for comparing different RNA extractionkits.



isolate exosomes, similar yields of small RNA were obtained.Regardless of the kit used, there was high reproducibility andconsistency of RNA yield. To confirm an enrichment of intactmiRNA in exosomes, RNA extraction was also performed onthe cell pellet (Figure 5b) and the cell-free component ofurine (Figure 5c). As expected, there was less consistency andmiRNA extracted from the cell pellet, as this contains mostlycellular debris and degraded RNA (Figure 3c), although mostkits were able to extract nucleic acids of less than 200 nt.Overall, a higher yield of total small RNA was extracted fromcell-free urine compared with the pellet of isolated exosomes(Figure 5c). This is not surprising, as the cell-free urineanalyzed in Figure 5c had not been depleted of exosomes. It isimportant to note that the integrity of the small RNA isolatedin cell-free urine cannot be analyzed owing to the absence ofribosomal RNA. Deep sequencing would provide an insightinto whether there is an intact population of miRNAfragments extracted from cell-free urine.

In addition to the amount of RNA obtained per sample,the percentage of miRNA recovered from the total RNAextracted was also analyzed by using a Bioanalyser small RNAassay. The highest percentage of miRNA was extracted fromexosomes compared with the cell pellet and cell-free urine,suggesting that there is an enrichment of miRNA inexosomes. The most efficient kit that isolated the highestpercentage of miRNA was the miRNeasy kit through theseparation and enrichment of miRNA from large RNA usingthe additional RNeasy MinElute column (80%). The secondmost efficient was the miRNeasy kit (without miRNAenrichment; 70%), followed by the other three kits(approximately 60%). Although miRNA comprised 50% ofthe small RNA extracted by the Norgen kit, overall the

miRNA yield was considerably higher compared with theother methods of extraction (summarized in Table 1).Although considerable percentages of miRNA were obtainedfrom the cell pellet and cell-free urine, it cannot bedetermined whether these fragments between 10 and 40 ntare miRNA or degraded RNA.

On comparing preparation time of the differentialultracentrifugation procedure (4 h), it was found that theNorgen Biotek kit eliminates almost 3 h of preparation timeand uses minimal sample volume (5 ml) to isolate exosomalRNA for downstream assays (Table 1). Unfortunately, proteincharacterization of the exosomes cannot be performed withthe samples extracted from the Norgen Biotek kit. Owing tothe high efficiency of the Norgen Biotek kit, we decided tocontinue the use of this kit to profile exosomal miRNA bydeep sequencing and compare the miRNA profiles withexosomes isolated using the differential ultracentrifugationmethod.

miRCURY

miRNeasy-RNeasy

mirVana 46%

81%

53%

58%

74%

0 2 4 6ng miRNA/ml urineng miRNA/ml urine

8 10 12 14

TRIzol LS+ mirVana

miRNeasy

0 1 2 4 53

47%

36%

46%

63%

47%

ng miRNA/ml urine

miRCURY

miRNeasy-RNeasy

mirVana

TRIzol LS+ mirVana

miRNeasy

miRCURY

miRNeasy-RNeasy

mirVana

TRIzol LS+ mirVana

miRNeasy

Norgen

0 5 10

50%

70%

59%

61%

80%

61%

NS

*

15 25 3520 30

Figure 5 | miRNA yields obtained by different RNA isolation kits. (a) Exosomal miRNA, (b) cell pellet miRNA, (c) cell-free miRNA. Eachsample was measured by Bioanalyser small RNA assay in three independent experiments. Percentage of miRNA is obtained by the Bioanalyser,which gates RNA fragments between 10 and 40 nt. Two-tailed Student’s t-test was carried out between the two kits. *P¼ 0.001.

Table 1 | Performance of six commercial RNA extraction kitsin extracting RNA from urinary exosomes

Kit UCSample

volume (ml) Time (h) Efficiency%

miRNA

Norgen No 5–10 1.5 Very high MediummiRNeasy Yes X20 4 High HighmiRNeasy-RNeasy 4.5 Medium Very highmirVana 4 Medium MediumTrizol LS þ mirVana 4.5 High MediummiRCURY 4 Medium Medium

Abbreviations: miRNA, micro RNA; UC, ultracentrifuge.



Baseline miRNA profiles of exosomal preparations andcell-free urine

Urine samples from three different subjects (SupplementaryTable S1 online) were sequenced using the Ion TorrentPersonal Genome Machine (Ion PGM, Life Technologies)sequencing platform. Small RNA libraries were constructedusing RNA extracted from exosomes isolated from the ultra-centrifugation method and the Norgen Biotek kit. To observewhether there was a selective profile of exosomal miRNA andenrichment by using these methods, deep sequencing wasalso performed on cell-free urine. Deep sequencing was notperformed on cell pellets, as the small RNA libraries con-structed were found to contain a high abundance of smallcomplementary DNA (cDNA) inserts, most likely of degradedRNA. miRNA enrichment using the RNeasy MinElutecolumn was not performed, as cDNA library preparationfor deep sequencing involves size selection of small RNA, andhence the additional miRNA enrichment step was unneces-sary. In addition, the majority of RNA species containedin the exosomes were small RNAs. In all, nine librarieswere constructed (n¼ 3) with identifiable indexes andsequenced on three 318 Ion Torrent chips using the IonPGM platform.

A total of 7,075,206 reads were obtained with an averageof 786,134 single-end reads produced per sample. Rawsequences were aligned to the human genome (HG19), andreads were mapped to miRBase V.1922 (Figure 6).

Of these, 535 known miRNAs were identified, whichcomprised 35% of total reads (Figure 7a, SupplementaryData SD1 online). Upon removing sequences mapped tomiRNA, the remaining sequences were mapped to otherncRNAs of the human genome using Ensembl annotations(Homo_sapiens.GRCh37.72).23 Less than 1% was found tomap to other ncRNAs such as small nuclear RNA (0.02%),snoRNA (0.04%), LncRNA (0.02%), and LinRNA (0.17%;Figure 7a). Coding RNA (3%) mapped to the human genomewas also detected in all samples (Supplementary Data SD2and SD3 online).

Low abundant miRNA sequences containing less than fivenormalized reads per million were removed. The removal oflow abundant miRNA reads eliminates possible artifactsobtained from the normalization of miRNA that are notpresent in one or more samples. Therefore, miRNA with highread counts are considered abundant and consist of actualraw reads before and after post-processing of NGS data. Fromour analyses, only 12 miRNAs were abundantly expressed incell-free urine compared with 66 and 184 miRNAs detectedin RNA isolated using the Norgen Biotek kit and by ultracen-trifugation of exosomes, respectively (Figure 7b, Supplemen-tary Data SD4 online). For each miRNA identified, themajority of miRNA was present in all three subjects(Figure 8).

Upon analyzing common and unique miRNA across allsamples, seven miRNAs were found to be common in allsamples (Figure 7b). Surprisingly, there was a low abundanceof miRNA in cell-free urine. Although a high yield of small

RNA was extracted from cell-free urine (Figure 5c), it ispossible that non-exosomal RNA circulating in cell-free urineis substantially degraded and consequently does not mapto miRNA annotations in miRBase. Only two miRNAs(hsa-miR-3648 and hsa-miR-4516) were found to bespecifically detected in cell-free urine. A total of 182 miRNAsidentified were exosomal specific. By comparing the twoexosome preparations, 5 miRNAs were specifically isolatedusing the Norgen Biotek kit (Table 2) and 51 miRNAs weredetected in both samples.

However, the majority of miRNAs (184) identified in thisstudy were extracted from exosomes isolated by using theultracentrifuge, comprising 126 specific miRNAs (Figure 7b).The 10 most abundant miRNAs detected in the three samples(n¼ 3) were pooled and presented in Table 3. Those that arecommon across the three samples are outlined in bold. Thistable illustrates that the Norgen Biotek kit isolates miRNAthat can be found both in cell-free and ultracentrifugeexosomes.

DISCUSSION

The advantages of using urine in clinical tests are that it iscollected noninvasively, and the procedure is relatively fastand cost-efficient compared with other clinical samples suchas blood and cerebrospinal fluid. A number of recent studieshave pooled patient samples in order to perform microarrayor quantitative PCR analysis, as it is difficult to obtainenough RNA to successfully construct cDNA librariesfor high-throughput deep sequencing.24,25 Most studiesuse the urine cellular sediment obtained after low-speedcentrifugation for miRNA analysis.4,26–30 However, wedetected a large proportion of low-quality and degradedRNA in the cell pellets of urine. This is not surprising, asthere is a high presence of RNase activity in the kidneys,bladder, and urinary tract to maintain sterility andprotect the excretory system from microbial infections.31,32

False-positive reads can be uncovered upon analyzing deepsequencing data whereby degraded RNA sequences align tothe human genome and map to small RNA annotations.

Fewer studies have used cell-free urine or exosomes toisolate miRNAs as we have done in this study.24,29,33 Apossible reason could be because of the technical difficultiesof handling large liquid volumes during RNA extraction.However, once separated, exosomes isolated from cell-freeurine may contain genetic material originating from othertissues from the body apart from the excretory system.Irrespective of the RNA extraction method, quantification ofmiRNA yield by small RNA analysis demonstrated that intactmiRNAs are enriched in exosomes compared with the cellpellet and cell-free component of urine. The abundance ofmiRNA in exosomes further supports the function of theprotective membrane enclosure of exosomes that allows theshuttling of genetic material in biological fluids otherwise,degraded by RNase activity. The notion that urinaryexosomes contain a stable source of miRNA makes urinean appealing biological specimen to discover biomarkers.



21

Px 1 Cell free

Px 2 Cell free

Px 3 Cell free

Px 1 UC Exosomes

Px 2 UC Exosomes

Px 3 UC Exosomes

Px 1 NG Exosomes

Px 2 NG Exosomes

Px 3 NG Exosomes

A C G T NBase colors

Chromosome 17

Chromosome 8

Num

ber

of r

aw r

eads

Num

ber

of r

aw r

eads

Num

ber

of r

aw r

eads

Chromosome 22

hsa-let-7b-5p

hsa-miR-320a

465095824650957646509570 46509588 46509594

221024992210249322102487 22102505 22102511

27188432

hsa-miR-451a

2718842627188420q21.32 c11.2

c18.81q13.32

q24.23 p21.3

p12

p13

27188438 27188444

0

26

0

7

0

0

0

0

1276

247

199

119

0

0

0

548

42

p22

Figure 6 | A map of chromosomes showing locations of micro RNAs (miRNAs) identified in this study. An example of three highlyabundant miRNAs mapped to chromosomes (HG19) based on the coordinates of miRBase version 19. The number of raw reads mapped to therespective miRNA is displayed using Partek Genomic Suites genome browser. NG, Norgen Biotek Urine Exosomal RNA kit; UC, ultracentrifuge.



Here, we have successfully characterized exosomes isolatedby differential ultracentrifugation and optimized conditionsto increase exosomal yield for transcriptome studies. Thedepletion of THP was found to be unwarranted in theisolation procedure, as exosomal yield was sufficient from20 ml of urine without DTT treatment. To extract exosomalRNA, the most efficient commercially available kits tomaximize RNA extraction from exosomes were the miRNeasykit (Qiagen) and mirVana Kit (Ambion).

To gain a thorough insight into exosome-specific miRNA,comparisons of exosomal profiles were analyzed againstmiRNA profiles from cell-free urine. Unlike miRNA profilessequenced from plasma, cell-free urine does not contain ahigh abundance of miRNA. From 2.5 ml of cell-free urine,only 12 miRNAs were abundantly expressed. In contrast, 1 mlof plasma or serum can contain more than 500 miRNAs.34,35

The low abundance of miRNA-detected cell-free urinefurther supports the presence of high RNase activity in thebladder compared with serum.36

The volume (20 ml) of urine required to isolate a suffi-cient quantity of exosomes was far greater than the volume ofcell-free urine (2.5 ml) required to isolate circulating miRNA.However, from the small pellet of exosomes isolated by usingthe ultracentrifuge, 184 exosomal miRNAs were detected.Only 7 miRNAs of the 184 were common to the cell-freeurine, indicating the selective packaging of miRNA or aresult reflective of the lack of miRNA detected in cell-freeurine. Exosomal-associated miRNAs were detected, suchas hsa-Let-7b, hsa-miR-29b, hsa-miR-200c, and hsa-miR-21.10,37,38 Microarray analysis of urine exosomes detected 194miRNAs in 11 samples of pooled urine, and it was performedin a study investigating miRNA expression response to saltintake and blood pressure.25 This demonstrates the power ofunbiased high-throughput deep sequencing, which was ableto detect 184 miRNAs from only three samples using theprocedures outlined in this study.

We utilized a commercial exosome to RNA isolation kit byNorgen Biotek to investigate whether the protocol would besuitable for NGS. The advantage of the Norgen Biotek kit isthat it does not require an ultracentrifuge and is suitable forclinical laboratories. Interestingly, the Norgen Biotek kitisolated almost four times the amount of miRNA comparedwith ultracentrifugation methods. Upon further analysis, adifferent miRNA profile was obtained from the NorgenBiotek kit compared with that profiled from characterizedexosomes isolated by ultracentrifugation. The majority ofmiRNAs (51) extracted using the Norgen Biotek kit were alsodetected in exosomes isolated by the ultracentrifugationmethod. However, 4 of the 10 most abundant miRNAs (hsa-miR-451a, hsa-miR-125a-5p, hsa-miR-191–5p, and hsa-miR-223-3p) were also found in the cell-free urine, indicatingsome potential non-exosomal miRNA contamination withthe Norgen kit procedure. Nonetheless, applications thatrequire a time-efficient protocol that delivers a consistentyield of highly abundant intact miRNA from urine wouldvalue the Norgen Biotek kit to be suitable for downstreamassays such as deep sequencing and quantitative PCR.

In addition to miRNA, less than 1% of reads were mappedto ncRNA annotations. Most of the reads (61%) mapped toregions in between genes. Coding RNA was also detected(3%). The most abundant coding RNA detected in cell-freeurine was thrombomodulin (NM_000361), a protein cofac-tor expressed on endothelial cell surfaces and detected inthe glomerulus of the kidneys.39 In exosomes isolated byultracentrifugation, CTC1 (NM_025099) was frequentlyexpressed. CTC1 is a CTS telomere maintenance complexcomponent 1 that protects telomeres from degradation. Theelevation of telomere cofactors has been detected in bladdercancer.40 Finally, STMN1 (NM_203401) was frequentlyexpressed in exosomes isolated using the Norgen Biotek kit,followed by THBP. STMN1 encodes a ubiquitous cytosolicphosphoprotein proposed to function as an intracellularfeedback signal relating to the cellular environment. Anincrease of the STMN1 protein has been associated withurothelial carcinoma of the bladder.41 Other ncRNAs (61%)

snRNA, snoRNA,LinRNA, lincRNA, >1%

miRNA, 35%

Coding RNA, 3%

Cell-free urine (12)

2

3 0

1265

NG Exosomes (66) UC Exosomes (184)

51

7

Other non-codingRNA, 61%

Figure 7 | Mapped known noncoding RNA identified by deepsequencing. (a) Percentage of total reads mapped to noncodingRNA and coding RNA identified by deep sequencing. (b) Venndiagram showing unique and common micro RNAs (miRNAs) indifferent samples. miRNA with read counts 45 reads per million wereshown for comparison.



Cell free

hsa-miR-16-5p hsa-miR-301a-3p hsa-miR-664a-3p hsa-miR-1307-5phsa-miR-378c

hsa-miR-3065-5phsa-miR-664a-5p

hsa-miR-320c

hsa-miR-320bhsa-miR-934

hsa-miR-877-5phsa-miR-708-5phsa-miR-891a

hsa-miR-501-3phsa-miR-500a-5phsa-miR-340-5phsa-miR-362-3phsa-miR-361-3phsa-miR-140-3phsa-miR-10a-3p

hsa-miR-30c-2-3phsa-miR-31-3phsa-miR-22-5phsa-let-7b-3p

hsa-miR-769-5phsa-miR-454-3phsa-miR-542-3phsa-miR-660-5phsa-miR-590-5phsa-miR-584-5phsa-miR-582-5phsa-miR-501-5phsa-miR-500a-3phsa-miR-146b-5p

hsa-miR-489hsa-miR-20b-5phsa-miR-18b-5phsa-miR-186b-5phsa-miR-324-3phsa-miR-324-5phsa-miR-135b-5phsa-miR-151a-3phsa-miR-378a-3phsa-miR-374a-5phsa-miR-363-3phsa-miR-362-5p

hsa-miR-301a-3phsa-miR-106b-5p

hsa-miR-190a

hsa-miR-9-5p

hsa-miR-152hsa-miR-138-5phsa-miR-222-3phsa-miR-183-5phsa-miR-181a-5phsa-miR-34a-5p

hsa-miR-196a-5phsa-miR-106a-5p

hsa-miR-101-3p

hsa-miR-98-5phsa-miR-96-5p

hsa-miR-95hsa-miR-33a-5phsa-miR-32-5phsa-miR-28-5phsa-miR-19a-3phsa-miR-17-3p

hsa-miR-374c-3phsa-miR-374b-3phsa-miR-365b-3p

hsa-miR-320d

hsa-miR-320bhsa-miR-455-3phsa-miR-532-3phsa-miR-502-3phsa-miR-29c-5p


hsa-miR-28-3phsa-miR-421

hsa-miR-615-3phsa-miR-532-5phsa-miR-503-5p

hsa-miR-200a-5phsa-miR-429

hsa-miR-424-5phsa-miR-148b-3phsa-miR-188-5phsa-miR-186-5phsa-miR-185-5p

hsa-miR-9-5phsa-miR-140-5phsa-miR-138-5phsa-miR-135a-5p

hsa-miR-210hsa-miR-182-5phsa-miR-181a-5p

hsa-miR-25-3p

hsa-miR-1180hsa-miR-744-5p


hsa-let-7d-3phsa-miR-425-5phsa-miR-652-3phsa-miR-574-3phsa-miR-339-5phsa-miR-30e-3phsa-miR-194-5phsa-miR-149-5phsa-miR-126-3phsa-miR-216-5p

hsa-miR-196a-5phsa-miR-92a-3phsa-miR-22-3phsa-miR-17-5p

hsa-miR-3184-3p

hsa-miR-103b

hsa-miR-423-5p

hsa-miR-103a-3p

hsa-miR-551b-3phsa-miR-193b-3phsa-miR-423-3phsa-miR-345-5p

hsa-miR-365a-3phsa-miR-30e-5p

hsa-miR-135a-5phsa-miR-128hsa-miR-107

hsa-miR-31-5phsa-miR-32b-3phsa-miR-361-5phsa-miR-320a

hsa-miR-141-3p

hsa-miR-221-3phsa-miR-187-3phsa-miR-148a-3phsa-miR-20a-5phsa-miR-19b-3phsa-miR-15a-5p

hsa-miR-484hsa-miR-200a-3phsa-miR-15b-5phsa-miR-218-5phsa-miR-93-5p

hsa-miR-92a-3phsa-miR-29a-3phsa-miR-194-5p

hsa-miR-128hsa-miR-197-3phsa-miR-100-5p

hsa-let-7d-5phsa-miR-204-3phsa-miR-335-5phsa-miR-192-5phsa-miR-24-3phsa-miR-21-5p

hsa-miR-99b-5phsa-miR-24-3p

hsa-miR-205-5phsa-miR-181b-5phsa-miR-30a-3phsa-miR-16-5p


hsa-miR-200b-3phsa-miR-29b-3phsa-miR-26b-5phsa-miR-16-5p

hsa-miR-19b-3p

hsa-miR-27a-3phsa-miR-342-3phsa-miR-29c-3phsa-miR-23b-3phsa-miR-23a-3p

hsa-miR-181b-5p

hsa-miR-29b-3phsa-miR-30d-5phsa-miR-200c-3phsa-miR-99a-5p

hsa-miR-125b-5p

hsa-miR-30c-5p

hsa-miR-125b-5phsa-let-7a-5phsa-miR-451a

hsa-miR-30a-5phsa-miR-30b-5phsa-miR-203a

hsa-miR-26a-5p

hsa-let-7b-5phsa-miR-10b-5phsa-miR-10a-5phsa-miR-191-5p

hsa-miR-125a-5phsa-miR-204-5p

hsa-miR-223-3p

hsa-let-7f-5phsa-let-7e-5phsa-let-7g-5p

hsa-let-7f-5p

hsa-let-7a-5p

hsa-let-7c

hsa-let-7a-5p

hsa-let-7i-5p

hsa-miR-323b-5phsa-miR-124-3phsa-miR-145-5p

hsa-miR-124-3p

hsa-miR-486-5phsa-miR-126-3phsa-miR-29a-3phsa-miR-3065-5p

hsa-let-7g-5p

hsa-miR-16-5p

hsa-miR-130a-3phsa-miR-26b-3phsa-miR-335-5phsa-miR-99b-5phsa-miR-23b-3phsa-miR-19b-3phsa-miR-26a-5phsa-miR-30c-5phsa-miR-26a-5phsa-miR-30c-5phsa-miR-30b-5phsa-miR-451ahsa-miR-21-5p

hsa-miR-148a-3phsa-miR-24-3p

hsa-miR-194-5p

hsa-miR-29b-3p

hsa-let-7f-5p


hsa-miR-125a-5phsa-miR-191-5phsa-miR-223-3phsa-miR-451ahsa-miR-134hsa-miR-320ahsa-miR-4516hsa-miR-4488hsa-miR-3648hsa-miR-6087

PresentAbsent

Px

2P

x 3

Px

1

Px

2P

x 3

Px

1

Px

2P

x 3

Px

1

Px

2P

x 3

Px

1

NG exosomes UC exosomes

hsa-let-7a-5p

hsa-miR-30d-5phsa-miR-10a-5p

hsa-miR-484hsa-miR-6087


hsa-miR-19b-3phsa-miR-30a-5phsa-miR-181b-5phsa-miR-425-5p

hsa-let-7a-5phsa-miR-100-5phsa-miR-10b-5phsa-miR-4488

hsa-miR-181b-5phsa-miR-200b-3p


hsa-miR-185-5phsa-miR-320c

hsa-miR-423-3phsa-miR-125b-5p

hsa-miR-4497hsa-miR-28-3phsa-miR-320c

hsa-miR-200c-3phsa-miR-191-5phsa-miR-320bhsa-miR-378c

hsa-miR-205-5phsa-miR-193a-5phsa-miR-204-5phsa-miR-125b-5p

hsa-miR-320bhsa-miR-223-3p

hsa-let-7b-5phsa-miR-99a-5p

hsa-miR-134hsa-miR-125a-5phsa-miR-192-5phsa-miR-378a-3phsa-miR-204-3phsa-miR-203a

hsa-miR-222-3phsa-miR-34a-5phsa-miR-320a

hsa-let-7c

hsa-miR-18a-5p

Figure 8 | The presence of the highly abundant micro RNAs (miRNAs) identified across each subject. Heat maps demonstrating thepresence or absence of each miRNA identified in cell-free or exosomal preparations across each subject. NG, Norgen Biotek Urine ExosomalRNA kit; UC, ultracentrifuge.



were found to be mapped in between gene regions. The highpercentage of reads mapped in between introns may bebecause of the biological nature of urine and function of thekidneys to excrete waste. Deep sequencing of exosomesisolated from other biological fluids such as plasma and salivatypically do not contain a high percentage of reads mappedin between genes, as seen in our laboratory and otherstudies.42,43 In addition, a small percentage of these readswere those mapped to transcripts resulting from manualcuration of genome annotations with unknown functions.The methods described here can be used to identify miRNAbiomarkers in a number of diseases such as cancer and usedto monitor patients in therapeutic drug trials.

MATERIALS AND METHODS

Exosome isolationMorning urine was collected from four healthy volunteers withinformed consent and stored at � 80 1C until use. Ethical approvalwas obtained from the University of Melbourne Health SciencesHuman Ethics Sub-committee (#1136882). Upon thawing on ice,urine was vortexed for 90 s. Urine samples (20 ml) were centrifugedat 500 g for 15 min to pellet the cells and cellular debris. The cell-freesupernatant was transferred into Ti70 centrifuge tubes andcentrifuged at 17,000 g for 45 min at 4 1C (70Ti rotor, BeckmanCoulter, Gladesville, NSW, Australia) to remove large membranevesicles. The supernatant was transferred to a new tube andcentrifuged at 200,000 g for 65 min at 4 1C to pellet exosomes, whichwere resuspended in 30 ml of phosphate-buffered saline (PBS).

Table 2 | Exosomal-specific miRNA detected in UC and NG exosome samples

NG exosomes Reads (r.p.m.) UCa and NG exosomes UC reads (r.p.m.) NG reads (r.p.m.) UC exosomesa Reads (r.p.m.)

hsa-miR-124-3p 6 hsa-miR-30a-5p 31 3594 hsa-miR-29c-3p 560hsa-miR-145-5p 5 hsa-miR-30b-5p 41 3583 hsa-miR-200a-3p 510hsa-miR-323b-5p 9 hsa-miR-10b-5p 53 3041 hsa-let-7e-5p 453hsa-miR-4497 5 hsa-miR-10a-5p 30 2076 hsa-miR-141-3p 345hsa-miR-486-5p 7 hsa-miR-30d-5p 13 1568 hsa-miR-429 218

hsa-miR-26a-5p 19 1467 hsa-let-7d-5p 182hsa-miR-30c-5p 21 1341 hsa-miR-20a-5p 174

hsa-miR-200c-3p 28 1180 hsa-miR-26b-5p 170hsa-miR-99a-5p 54 792 hsa-miR-17-5p 169

hsa-let-7b-5p 62 779 hsa-miR-30e-5p 157hsa-miR-19b-3p 13 718 hsa-miR-93-5p 155

hsa-miR-200b-3p 8 703 hsa-miR-27b-3p 155hsa-let-7a-5p 7 609 hsa-miR-27a-3p 139

hsa-miR-29b-3p 21 476 hsa-miR-660-5p 136hsa-miR-203a 39 454 hsa-miR-221-3p 129hsa-miR-21-5p 16 446 hsa-miR-30a-3p 119hsa-let-7f-5p 7 425 hsa-miR-19a-3p 108hsa-let-7g-5p 8 358 hsa-miR-103a-3p 105

hsa-miR-29a-3p 9 310 hsa-miR-103b 105hsa-miR-125b-5p 29 243 hsa-miR-23a-3p 103

hsa-let-7c 10 201 hsa-miR-135b-5p 97hsa-miR-23b-3p 11 200 hsa-miR-424-5p 94hsa-miR-24-3p 7 174 hsa-miR-374b-5p 84

hsa-miR-99b-5p 8 170 hsa-miR-374c-3p 84hsa-miR-335-5p 7 139 hsa-miR-106a-5p 76

Abbreviations: miRNA, micro RNA; NG, Norgen Biotek Urine Exosomal RNA kit; UC, ultracentrifuge.All miRNAs identified and accession numbers in Supplementary Data SD6 online.Data uploaded on Vesiclepedia (http://www.microvesicles.org/exp_summary?exp_id=357).aRNA extracted by the miRNeasy kit.

Table 3 | The 10 most highly expressed miRNA

Cell-free urinea Reads (r.p.m.) NG exosomes Reads (r.p.m.) UC exosomesa Reads (r.p.m.)

hsa-miR-6087 51 hsa-miR-320a 197 hsa-miR-30a-5p 3594hsa-miR-451a 22 hsa-miR-125a-5p 100 hsa-miR-30b-5p 3583hsa-miR-223-3p 17 hsa-miR-451a 68 hsa-miR-10b-5p 3041hsa-miR-191-5p 15 hsa-miR-191-5p 67 hsa-miR-10a-5p 2076hsa-miR-3648 15 hsa-let-7b-5p 62 hsa-miR-30d-5p 1568hsa-miR-4488 9 hsa-miR-223-3p 61 hsa-miR-26a-5p 1503hsa-miR-204-5p 9 hsa-miR-192-5p 55 hsa-miR-30c-5p 1343hsa-miR-125a-5p 7 hsa-miR-34a-5p 55 hsa-miR-200c-3p 1180hsa-miR-134 7 hsa-miR-99a-5p 54 hsa-miR-99a-5p 792hsa-miR-16-5p 7 hsa-miR-10b-5p 53 hsa-let-7b-5p 779

Abbreviations: miRNA, micro RNA; NG, Norgen Biotek Urine Exosomal RNA kit; UC, ultracentrifuge.Bold font indicates common miRNA.All miRNAs identified and accession numbers in Supplementary Data SD7 online.aRNA extracted by the miRNeasy kit.



Where indicated, exosome pellets were resuspended in 200 mg/mlDTT and incubated at 37 1C for 10 min. After incubation with DTT,exosomes were transferred to an ultracentrifuge tube and resuspendedwith cold 0.2-mm-filtered PBS buffer. The sample was centrifuged at200,000 g for 65 min to pellet the DTT-treated exosomes.

Western immunoblottingThe cell pellet, cell-free and isolated exosomes of urine were lysed inlysis buffer (50 mmol/l Tris pH 7.4, 150 mmol/l sodium chloride, 1%Triton X-100, 1% sodium-deoxycholic acid, and complete ULTRAprotease inhibitors (Roche, Hawthorn, VIC, Australia)) on ice for1 h and then centrifuged at 17,000 g for 20 min. The supernatant wascollected and transferred to a new tube. Samples (20 mg) werediluted in SDS loading buffer resolved on 4–12% Bis-Tris NUPAGESDS-PAGE gels (Life Technologies, Mulgrave, VIC, Australia) andtransferred onto Immobilon-P PVDF membrane (Millipore,Darmstadt, Germany). The membranes were blocked with 5% skimmilk buffer in PBS containing 0.05% tween, followed by incubationwith primary antibodies to anti-Tsg101, anti-flotillin, anti-CD63,anti-Bcl2, anti-nucleoporin, and anti-GM130. Immunoreactivebands were visualized by using the enhanced Chemiluminescencekit (Amersham Biosciences, Rydalmere, NSW, Australia) followed byexposure to Hyperfilm (Amersham Biosciences) and then developedby the Exomat automated developing system.

Electron microscopyExosomes were fixed with 2% glutaraldehyde/PBS for 30 min atroom temperature. A volume of 6 ml was applied to a glow-discharged 200-mesh Cu grid coated with carbon-Formvar film(ProSciTech, Kirwan, QLD, Australia) and allowed to absorb for5 min. Grids were washed twice with MilliQ water and contrastedwith 1.5% uranyl acetate. TEM was performed on a Tecnai G2 F30(FEI, Eindhoven, NL) TEM operating at 300 kV across �15,000 to�36,000 magnification. Electron micrographs were captured with aGatan UltraScans 1000 2 k�2 k CCD camera (Gatan, Pleasanton,CA).

Size distribution analysisScanning ion occlusion sensing analysis was performed using theqNano system (Izon, Christchurch, New Zealand); by applying avoltage and pressure, single particles are passed through apolyurethane nanopore, resulting in a change in ionic current,termed a ‘blockade event.’ Exosomes were resuspended in PBS with0.025% Tween-20 to reduce aggregation. Before measurement,exosomes and cell-free urine were passed through a 0.22-mm filter.Blockade events were calibrated against particles of a known sizemeasured under identical settings.

RNA extractionsSix commercially available kits were compared for total RNAextraction from the cell pellet, cell-free pellets, and exosome pelletsfrom urine: miRNeasy (Qiagen, Limburg, NL), miRNeasy withRNeasy MinElute Cleanup Kit (Qiagen), mirVana PARIS (Ambion,Mulgrave, VIC, Australia), Trizol LS reagent (Life Technologies) withmirVana (Ambion), miRCURY (Exiqon, Vedbaek, Denmark), andUrine Exosome RNA Isolation kit (Norgen Biotek, Thorold, ON,Canada). The manufacturers’ protocol was followed and total RNAextractions (418 nt) including the recovery of small RNA wereperformed on the same pool of urine samples in triplicate for eachkit. RNA extractions were eluted in 100 ml and stored at � 80 1C

until use. The quantity and quality of the RNA extractions weredetermined by the Agilent Bioanalyser 2100 with a small RNA Chipfor exosomal and cell-free miRNA, and an RNA Nano Chip forcellular RNA of the urine (Agilent Technologies, Mulgrave, VIC,Australia). miRNA and small RNA yields were normalized to ngRNA per ml of urine.

Small RNA library construction and sequencingDeep sequencing was performed on three individual urine samples,which were processed separately for exosomal and cell-free RNAisolations (n¼ 3). For each library, 1 ng of RNA was ligated toadaptors containing a unique index bar code (Ion Xpress RNA-SeqBarcode 1–16 Kit; Life Technologies). RNA samples were reversetranscribed to cDNA using adaptor-specific primers. Using theMagnetic Bead Purification Module (Life Technologies), cDNAsamples were size-selected from 94 to 130 nt (the length of themiRNA insert including the 30 and 50 adaptors). PCR amplificationwas then performed, followed by a library clean-up using nucleicacid beads (Life Technologies) and eluted in 10 ml of nuclease-freeH2O. The yield and size distribution of the small RNA libraries wereassessed using the Agilent 2100 Bioanalyzer instrument with theHigh-sensitivity DNA Assay (Agilent Technologies). Equally pooledlibraries were clonally amplified onto Ion Sphere Particles suppliedin the Ion OneTouch 200 Template Kit v2 DL kit (LifeTechnologies). Ion Sphere Particle templates were produced byusing the OneTouch 2 Instrument and ES system (Life Technolo-gies). Ion Sphere Particles loaded with libraries were sequenced onthe Ion Torrent PGM using Ion 318 chips (Life Technologies) andthe Ion PGM 200 Sequencing Kit with three bar-coded librariesprocessed per chip.

Sequencing data analysisFollowing sequencing on the Ion Torrent PGM, preprocessing ofreads, removal of adaptors, and bar codes were performed by theTorrent Suite (Life Technologies). Sequences were analyzed forquality control (FASTQC) and aligned to the Human genome(HG19) using the Torrent Suite. Output files (*.bam) were uploadedand analyzed using the Partek Genomic Suite software (Partek,Helios, Singapore). Sequences aligned to the human genome weremapped to miRBase_v.19 (ref. 22) and Emsembl release 17 (ref. 23).Reads were normalized to reads per million calculated asfollows: Number of sequenced reads/total reads�1,000,000.miRNA identified in this study were uploaded to Vesiclepedia(microvesicles.org)4

DISCLOSUREAll the authors declared no competing interests.

ACKNOWLEDGMENTSThis work was supported by grants from the Australian NationalHealth and Medical Research Council (NHMRC) (628946) and theAustralian Research Council (ARC) (FT100100560). BMC is therecipient of an NHMRC Postgraduate Scholarship, BJS is the recipientof an Australian Postgraduate Award, and AFH is an ARC FutureFellow. We thank the Advanced Microscopy Facility at Bio21Molecular Science and Biotechnology Institute, the University ofMelbourne, for electron microscopy facilities.

SUPPLEMENTARY MATERIALTable S1. Demographics of healthy control subjects in this study.Supplementary Data. Deep sequencing workbook spreadsheets.



Supplementary material is linked to the online version of the paper athttp://www.nature.com/ki

REFERENCES1. Bartel DP. MicroRNAs: genomics, biogenesis, mechanism, and function.

Cell 2004; 116: 281–297.2. Weber JA, Baxter DH, Zhang S et al. The microRNA spectrum in 12 body

fluids. Clin Chem 2010; 56: 1733–1741.3. Wang G, Kwan BC, Lai FM et al. Urinary miR-21, miR-29, and miR-93: novel

biomarkers of fibrosis. Am J Nephrol 2012; 36: 412–418.4. Wang G, Kwan BC, Lai FM et al. Expression of microRNAs in the urinary

sediment of patients with IgA nephropathy. Dis Markers 2010; 28:79–86.

5. Cheng L, Quek CY, Sun X et al. The detection of microRNA associated withAlzheimer’s disease in biological fluids using next-generation sequencingtechnologies. Front Genet 2013; 4: 150.

6. Arroyo JD, Chevillet JR, Kroh EM et al. Argonaute2 complexes carry apopulation of circulating microRNAs independent of vesicles in humanplasma. Proc Natl Acad Sci USA 2011; 108: 5003–5008.

7. Vickers KC, Remaley AT. Lipid-based carriers of microRNAs andintercellular communication. Curr Opin Lipidol 2012; 23: 91–97.

8. Valadi H, Ekstrom K, Bossios A et al. Exosome-mediated transfer of mRNAsand microRNAs is a novel mechanism of genetic exchange between cells.Nat Cell Biol 2007; 9: 654–659.

9. Bellingham SA, Guo BB, Coleman BM et al. Exosomes: vehicles for thetransfer of toxic proteins associated with neurodegenerative diseases?Front Physiol 2012; 3: 124.

10. Bellingham SA, Coleman BM, Hill AF et al. deep sequencing reveals adistinct miRNA signature released in exosomes from prion-infectedneuronal cells. Nucleic Acids Res 2012; 40: 10937–10949.

11. Jin P, Wang E, Ren J et al. Differentiation of two types of mobilizedperipheral blood stem cells by microRNA and cDNA expression analysis.J Transl Med 2008; 6: 39.

12. Taylor DD, Gercel-Taylor C. MicroRNA signatures of tumor-derivedexosomes as diagnostic biomarkers of ovarian cancer. Gynecol Oncol2008; 110: 13–21.

13. Skog J, Wurdinger T, van Rijn S et al. Glioblastoma microvesicles transportRNA and proteins that promote tumour growth and provide diagnosticbiomarkers. Nat Cell Biol 2008; 10: 1470–1476.

14. Chen X, Ba Y, Ma L et al. Characterization of microRNAs in serum: a novelclass of biomarkers for diagnosis of cancer and other diseases. Cell Res2008; 18: 997–1006.

15. Fernandez-Llama P, Khositseth S, Gonzales PA et al. Tamm-Horsfallprotein and urinary exosome isolation. Kidney Int 2010; 77: 736–742.

16. Gonzales PA, Zhou H, Pisitkun T et al. Isolation and purification ofexosomes in urine. Methods Mol Biol 2010; 641: 89–99.

17. Pisitkun T, Shen RF, Knepper MA. Identification and proteomic profiling ofexosomes in human urine. Proc Natl Acad Sci USA 2004; 101:13368–13373.

18. Vella LJ, Sharples RA, Lawson VA et al. Packaging of prions into exosomesis associated with a novel pathway of PrP processing. J Pathol 2007; 211:582–590.

19. Sharples RA, Vella LJ, Nisbet RM et al. Inhibition of gamma-secretasecauses increased secretion of amyloid precursor protein C-terminalfragments in association with exosomes. FASEB J 2008; 22: 1469–1478.

20. Coleman BM, Hanssen E, Lawson VA et al. Prion-infected cells regulatethe release of exosomes with distinct ultrastructural features. FASEB J2012; 26: 4160–4173.

21. Thery C, Zitvogel L, Amigorena S. Exosomes: composition, biogenesis andfunction. Nat Rev Immunol 2002; 2: 569–579.

22. Griffiths-Jones S. The microRNA Registry. Nucleic Acids Res 2004; 32:D109–D111.

23. Flicek P, Ahmed I, Amode MR et al. Ensembl 2013. Nucleic Acids Res 2013;41: D48–D55.

24. Alvarez ML, Khosroheidari M, Kanchi Ravi R et al. Comparison of protein,microRNA, and mRNA yields using different methods of urinary exosomeisolation for the discovery of kidney disease biomarkers. Kidney Int 2012;82: 1024–1032.

25. Gildea JJ, Carlson JM, Schoeffel CD et al. Urinary exosome miRNomeanalysis and its applications to salt sensitivity of blood pressure. ClinBiochem 2013; 1131–1134.

26. Bryant RJ, Pawlowski T, Catto JW et al. Changes in circulatingmicroRNA levels associated with prostate cancer. Br J Cancer 2012; 106:768–774.

27. Snowdon J, Boag S, Feilotter H et al. A pilot study of urinary microRNA asa biomarker for urothelial cancer. Can Urol Assoc J 2012; 15: 1–5.

28. Mengual L, Lozano JJ, Ingelmo-Torres M et al. Using microRNA profiling inurine samples to develop a non-invasive test for bladder cancer. Int JCancer 2013; 133: 2631–2641.

29. Wang G, Chan ES, Kwan BC et al. Expression of microRNAs in theurine of patients with bladder cancer. Clin Genitourin Cancer 2012; 10:106–113.

30. Miah S, Dudziec E, Drayton RM et al. An evaluation of urinary microRNAreveals a high sensitivity for bladder cancer. Br J Cancer 2012; 107:123–128.

31. Spencer JD, Schwaderer AL, Dirosario JD et al. Ribonuclease 7 is a potentantimicrobial peptide within the human urinary tract. Kidney Int 2011; 80:174–180.

32. Mizuta K, Awazu S, Yasuda T et al. Purification and characterization ofthree ribonucleases from human kidney: comparison with urineribonucleases. Arch Biochem Biophys 1990; 281: 144–151.

33. Cheng Y, Wang X, Yang J et al. A translational study of urine miRNAs inacute myocardial infarction. J Mol Cell Cardiol 2012; 53: 668–676.

34. Williams Z, Ben-Dov IZ, Elias R et al. Comprehensive profiling ofcirculating microRNA via small RNA sequencing of cDNA libraries revealsbiomarker potential and limitations. Proc Natl Acad Sci USA 2013; 110:4255–4260.

35. Yang Q, Lu J, Wang S et al. Application of next-generation sequencingtechnology to profile the circulating microRNAs in the serum ofpreeclampsia vs. normal pregnant women. Clin Chim Acta 2011; 412:2167–2173.

36. Naskalski JW, Celinski A. Determining of actual activities of acid andalkaline ribonuclease in human serum and urine. Materia Med Pol 1991;23: 107–110.

37. Hu G, Yao H, Chaudhuri AD et al. Exosome-mediated shuttling ofmicroRNA-29 regulates HIV Tat and morphine-mediated neuronaldysfunction. Cell Death Dis 2012; 3: e381.

38. Ng YH, Rome S, Jalabert A et al. Endometrial exosomes/microvesicles inthe uterine microenvironment: a new paradigm for embryo-endometrialcross talk at implantation. PLoS ONE 2013; 8: e58502.

39. Modde F, Agustian PA, Wittig J et al. Comprehensive analysis ofglomerular mRNA expression of pro- and antithrombotic genes inatypical haemolytic-uremic syndrome (aHUS). Virchows Archiv 2013; 462:455–464.

40. Eissa S, Motawi T, Badr S et al. Evaluation of urinary human telomerasereverse transcriptase mRNA and scatter factor protein as urine markersfor diagnosis of bladder cancer. Clin Lab 2013; 59: 317–323.

41. Bhagirath D, Abrol N, Khan R et al. Expression of CD147, BIGH3 andStathmin and their potential role as diagnostic marker in patientswith urothelial carcinoma of the bladder. Clin Chim Acta 2012; 413:1641–1646.

42. Ogawa Y, Taketomi Y, Murakami M et al. Small RNA transcriptomes of twotypes of exosomes in human whole saliva determined by next generationsequencing. Biol Pharm Bull 2013; 36: 66–75.

43. Huang X, Yuan T, Tschannen M et al. Characterization of humanplasma-derived exosomal RNAs by deep sequencing. BMC Genomics2013; 14: 319.

44. Kalra H, Simpson RJ, Ji H et al. Vesiclepedia: a compendium forextracellular vesicles with continuous community annotation. PLOSBiology 2012; 12: e1001450.



http://www.nature.com/ki

Characterization and deep sequencing analysis of … Urine Exosome RNA Kit in Kidne… · Characterization and deep sequencing analysis of ... exosomal isolation and RNA yield for

Documents