bioRxiv preprint first posted online Mar. 25, 2017; doi: http://dx.doi.org/10.1101/120634. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
AlternativemiRNAs:Humansequencesmisidentifiedasplant
miRNAsinplantstudiesandinhumanplasma
KennethW.Witwer
TheJohnsHopkinsUniversitySchoolofMedicine
DepartmentofMolecularandComparativePathobiology
DepartmentofNeurology
733N.Broadway
MillerResearchBuildingRm829
BaltimoreMD21205
USA
p1-410-955-9770
f1-410-955-9823
email:[email protected]
1
bioRxiv preprint first posted online Mar. 25, 2017; doi: http://dx.doi.org/10.1101/120634. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
Abstract
Arecentstudyreportedthat“PlantmiRNAsfoundinhumancirculatingsystemprovide
evidencesofcrosskingdomRNAi”1.Analysisoftwohumanbloodplasmasequencing
datasetswassaidtoprovideevidenceforuptakeofplantmiRNAsintohumanplasma.The
resultswerealsopurportedlyinconsistentwithcontamination1.However,areviewof
thesedatasuggeststhattheydonotsupportdietaryxenomiRuptake,butinsteadconfirm
previousfindingsthatdetectionofrareplantmiRNAsinmammaliansequencingdatasetsis
artifactual.OnlyoneputativeplantmiRNA(“peu-MIR2910)inthisstudymapped
consistentlyabovebackground,andthissequenceisfoundinahumanrRNA.Severalother
rarerbutconsistentlymappedplantmiRNAsalsohave100%ornear100%matchesto
humantranscriptsorgenomicsequences,andsomedonotmaptoplantgenomesatall.
Thesemisidentified“alternativemiRNAs”—includingMIR2910andMIR2911—emphasize
theneedforrigorousfilteringstrategieswhenassessingpossiblexenomiRNAs.
2
bioRxiv preprint first posted online Mar. 25, 2017; doi: http://dx.doi.org/10.1101/120634. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
INTRODUCTION
ReportsofplantorotherdietarymiRNAs,orxenomiRs,enteringmammaliancirculation
throughthediet2–5generatedexcitementforthexenomiRtransferhypothesis,yetnegative
resultsofreplicationandreproductionstudieshavecastdoubtonxenomiRtransferasa
generalmechanism6–12.AprominentclaimofxenomiRfunction2hasalsofailedrigorous
reproduction8,unmaskedastheresultofanuncontrolledvariableintheoriginal
experiment.AnalysesofpublicdatasetshaverevealedthatstudiesofxenomiRsandother
foreign-originnucleicacidsarefraughtwithartifacts:combinationsofcontamination,
amplificationorsequencingerrors,permissiveanalysispathways,andbatcheffects9,11,13–
17.AparticularlycomprehensivestudyrecentlyfoundthatforeignmiRNAsinhuman
biofluidsandtissuesdonotmatchhumanfoodconsumption,aremarkedbybatcheffects,
andarethusmostparsimoniouslyexplainedasartifacts14.Studiesoforganismswithno
exposuretoplantshavealsofoundevidenceofthesametypesofapparentplant
contaminationthatplaguesomemeasurementsofhumansamples9,18.Liuetal1mapped
sequencingdatafromtwostudiesofhumanplasmaandothersamples19,20tovariousplant
genomesusinga2010plantmiRNAdatabase,PMRD21,concludingthatpreviousreportsof
dietaryxenomiRtransferaresupported.Inthisbriefreport,theseresultsareexamined
critically.
RESULTS
Dataevaluation
Across-checkofthesourcefilesandarticlesshowsthattheplasmadataevaluatedbyLiuet
alwerefrom198plasmasamples,not410asreported.Ninomiyaetalsequencedsix
humanplasmasamples,sixPBMCsamples,and11culturedcelllines19.Yuanetal
sequenced192humanplasmalibraries(preparedfrompolymer-precipitatedplasma
particles)20.Eachlibrarywassequencedonce,andthenasecondtimetoincreasetotal
reads.Countswerepresentedasreadspermillionmappedreads(rpm)20.Incontrast,Liu
etalappeartohavereportedtotalmappedreadsintheirdatatable1.Yuanetalalsosetan
expressioncutoffof32rpm(log2rpmof5orabove).Withanaverage12.5millionreads
persample(thesumofthetworunsperlibrary),and,onaverage,abouthalfofthe
sequencesmapped,the32rpmcutoffwouldtranslatetoaround200totalreadsinthe
averagesampleasmappedbyLiuetal1.
OnlyoneputativeplantmiRNAabovebackgroundlevels
ConsultingtheLiuetalmappingtable1andtheSequenceReadArchive(SRA),resultsfrom
duplicatesequencingrunsfromtheYuanetaldatasetwerecombined,andtwosamples
withoutreliablereplicateswereeliminated.Atotalof1294putativeplantmiRNAshadat
leastonemappedreadinatleastoneoftheremaining190samples.However,manyof
thesemiRNAswereidenticalorthologsorparalogs,andmostweremappedatoneorfewer
rpmonaverage,andinonlyasmallminorityofsamples.Acrossallsamples,onlyone
putativeplantmiRNAmappedaboveamedian200readcutoff,roughlycorrespondingto
the32rpmcutoffofYuanetal(Table1).AllotherRNAs,includingpreviouslyreported
3
bioRxiv preprint first posted online Mar. 25, 2017; doi: http://dx.doi.org/10.1101/120634. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
xenomiRssuchasMIR159a,MIR168a,andtheplantribosomaldegradationfragment
MIR291122–24,werethusbelowthelevelofbackgroundnoiseestablishedbytheoriginal
investigators.Indeed,previouslyreportedxenomiRsweremappedinfewsamplesand
below1rpm.TheabsenceoftheseRNAsisconfirmedbyLiuetal’sanalysisofthe
Ninomiya,etalstudy19,whereMIR159a,MIR168a,andMIR2911mappedinnoneofthe
plasmasamples.ThesingleputativeplantmiRNAthatmappedabovebackgroundlevelsin
thisstudywas,again,peu-MIR2910(Table2).
Loweringthethreshold:stillonlyahandfulofpossiblexenomiRs
SinceonlyoneplantmiRNAappearedtomapconsistentlyabovebackground,theinclusion
thresholdofYuanetalwasrelaxedtoincludeallmiRNAswiththreeormoremappedreads
(Liuetaldata)in10%ormoreofthesamplesfromeitherstudy.Thesearerather
permissivecriteriabutmayatleastscreenoutsomefalsepositivesduetoamplification
andsequencingerrors.AllsamplesfromtheNinomiyastudywereincluded,despitethefact
thatmostwerenotplasma.11miRNAssatisfiedthesecriteriafortheYuanetaldata(Table
1).(Onelow-mappingmiRNAwasexcludedbecauseitssequencecouldnotbefoundin
miRBase25,26,miRMaid27,miRNEST2.028orindeedthroughanysearchesattempted.)10
satisfiedthecriteriafromtheNinomiyastudy(Table2),includingonesequencethatwas
partofanother(compareath-MIRf10046-akrandath-MIRf10045-ak,Table3).However,if
onlytheplasmasamplesfromthelatterstudyareconsidered,threemiRNAsremain(Table
2).Intotal,15putativemiRNAssatisfiedthepermissiveinclusioncriteria,includingfive
(Yuanonly),four(Ninomiyaonly),andsix(both)(Table3).
TomiRornottomiR
AsmiRNAdiscovery,validation,andannotationhasadvanced,numerousreportedmiRNAs
havebeenreclassifiedasdegradationfragmentsofothernoncodingRNAs(ncRNAs).A
classicexampleisMIR2911,aplantrRNAdegradationfragmentthathasbeenmisidentified
asamicroRNA.Interestingly,only2ofthe15miRNAsidentifiedasplantmiRNAsinthis
studyareannotatedinmiRBase.Althoughsomeofthesesequencesmayrepresentrareor
unusuallystructuredmiRNAs,severalarepartofnon-miRNAncRNAsorothersequences
thatseemunlikely,atleastatfirstglance,togiverisetomicroRNAs.Amongtheapparently
misidentifiedmiRNAsisMIR2910,themostabundantplantmiRNAidentifiedbyLiuetal.
TheMIR2910sequence,UAGUUGGUGGAGCGAUUUGUC,isfoundinthehighlyconserved
andexpressedlargesubunit(LSU)rRNAofplants,andhasbeenspecificallyremovedfrom
miRBaseasanon-miRNA.EventhetwoidentifiedmiRNAsthatremaininmiRBase,
MIR2916andMIR894,arenotabovequestion.A20nucleotidestretchofMIR2916mapto
rRNA,whilethefullMIR894sequenceappearstobefoundinavarietyofplanttranscripts.
Humansequencesintheplantdatabaseandvice-versa
Curiously,severalsequencesdidnotmaptothespeciestowhichtheywereascribedbythe
PMRD21.Unfortunately,thePMRDcouldnotbeaccesseddirectlyduringthisstudy;
however,otherdatabasesappeartoprovideaccesstoitscontents.Specifically,ptcMIRf12412-akrandptc-MIRf12524-akrdidnotmaptoPopulusortootherplants.The
poplartreeisalsonotacommondietarystapleofhumanpopulations.Incontrast,both
sequencesmappedwith100%identityandcoveragetonumeroushumansequences(Table
3).ptc-MIRf10804-akrhadnumerous100%identityhumanmatches,plusa1-mismatch
4
bioRxiv preprint first posted online Mar. 25, 2017; doi: http://dx.doi.org/10.1101/120634. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
alignmenttothehumanmiR-3929precursor.OthermiRNAs,includingMIR2911,also
displayedsomelesserdegreeofmatchingtohumantranscriptsorthegenome.Strikingly,
theputativeMIR2910sequenceisnotonlyafragmentofplantrRNA;ithasa100%
coverage,100%identitymatchinthehuman18SrRNA(seeNR_003286.2inGenBank;
Table3).ThesematchesofputativeplantRNAswithhumansequencesaredifficultto
reconcilewiththestatementofLiuetalthatBLASTofputativeplantmiRNAs"resultedin
zeroalignmenthit"1,suggestingthatperhapsamistakewasmade,andthattheBLAST
procedurewasperformedincorrectly.
CONCLUSION
Inmammalianstudies,mappingofMIR2910andotherdubiousplantmiRNAsisbest
explainedasmappingofhumandegradomefragmentstoplantRNAsthatareinsomecases
genuinesequencesbutnotmiRNAs,andinothercases,humansequencesthathave
contaminatedplantRNAsamplesanddatabases.Re-analysisoftheresultsofLiuetal1thus
echoestherecentfindingsofKang,Bang-Berthelsen,andcolleagues14,aswellasprevious
negativefindingssurroundingdietaryxenomiRs,summarizedabove.Astringentdata
analysisprocedure,suchasfilteringallreadsagainsttheingestingorganism
genome/transcriptomewithoneortwomismatches,thenrequiringperfectmatchesof
remainingreadsagainstplantorotherforeignorganisms,wouldengenderhigher
confidencethat“foreign”RNAsarenotsimplyamplificationorsequencingartifacts.Indeed,
pre-mappingtotheingestingorganism’sgenomemaynotbesufficient;asshown14,the
largestnumberofxenomiRsinsomehumanstudiesarefromrodents,likelybecauseof
proximityinresearchlaboratories.Therefore,itmaybebesttoscreenagainstmammalian
sequencesingeneral,andperhapsalsoagainstwidespreadmicrobecontaminants.Of
course,eventhemoststringentanalysisprocedurescannotdistinguishaphysical
contaminantfroma“real”read;thereforestrictprocesscontrolsarealsoneededtoassess
possiblecontamination.Ingeneral,suchcontrolshavenotbeendoneinexistingstudies.
ThisreportunderlinesthedangerinassumingthatxenomiRsinmammalianmaterial
originatefromthediet.Whenthespeciesandrolesarereversed—forexample,withthe
findingofhumansequencesinalistofpoplartreemiRNAs—fewanalystswouldconclude
thatpoplartreesconsumehumans.Thesimplestexplanationisthatthesequencedplant
materialwascontaminatedwithhumannucleicacid.Inthesameway,theextremelylowlevel,variable,andbatch-effectproneconcentrationsofseveralplantsequencesinhuman
plasmaandtissuecouldbeduetouptakefromthediet,albeitatlevelsfartoolowtoaffect
physiologicprocesses.However,artifactremainsthesimplestexplanation.
METHODS
PlantmappingresultsfromLiuetal1(totalmappedcounts)weredownloadedfromthe
BMCGenomicswebsite.Accessionnumbersofsequencingdatasetswerecheckedagainst
5
bioRxiv preprint first posted online Mar. 25, 2017; doi: http://dx.doi.org/10.1101/120634. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
thepublicationsofNinomiyaetal19andYuanetal20,aswellastheSequenceReadArchive
(SRA,https://www.ncbi.nlm.nih.gov/sra).DataweresortedandanalyzedinMicrosoft
Excel.PlantmiRNAsequenceswereobtainedfrommiRBase(http://mirbase.org/)29.
BecausecertainplantsequenceshavebeenremovedfrommiRBasebecausetheyhavebeen
identifiedasncRNAdegradationartifacts,theplantmicroRNAdatabase(PMRD)21was
consulted;however,repeatedattemptstoaccessthesite
(http://bioinformatics.cau.edu.cn/PMRD)wereunsuccessful,soinformationwasretrieved
insteadfrommiRMaid(http://140.mirmaid.org/home)27ormiRNEST2.0
(http://rhesus.amu.edu.pl/mirnest/copy/home.php)28.Allanalysisfilesareavailableon
request([email protected]).
6
bioRxiv preprint first posted online Mar. 25, 2017; doi: http://dx.doi.org/10.1101/120634. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
References
1.
Liu,Y.-C.,Chen,W.L.,Kung,W.-H.&Huang,H.-D.PlantmiRNAsfoundinhuman
circulatingsystemprovideevidencesofcrosskingdomRNAi.BMCGenomics18,112
(2017).
2.
Zhang,L.etal.ExogenousplantMIR168aspecificallytargetsmammalianLDLRAP1:
evidenceofcross-kingdomregulationbymicroRNA.CellRes22,107–126(2012).
3.
Chin,A.R.etal.Cross-kingdominhibitionofbreastcancergrowthbyplantmiR159.
CellRes.26,217–28(2016).
4.
Liang,H.etal.Effectivedetectionandquantificationofdieteticallyabsorbedplant
microRNAsinhumanplasma.JNutrBiochem26,505–512(2015).
5.
Baier,S.R.,Nguyen,C.,Xie,F.,Wood,J.R.&Zempleni,J.MicroRNAsAreAbsorbedin
BiologicallyMeaningfulAmountsfromNutritionallyRelevantDosesofCowMilkand
AffectGeneExpressioninPeripheralBloodMononuclearCells,HEK-293KidneyCell
Cultures,andMouseLivers.JNutr(2014).doi:jn.114.196436
[pii]10.3945/jn.114.196436
6.
Snow,J.W.,Hale,A.E.,Isaacs,S.K.,Baggish,A.L.&Chan,S.Y.Ineffectivedeliveryof
diet-derivedmicroRNAstorecipientanimalorganisms.RNABiol10,1107–1116
(2013).
7.
Witwer,K.W.,McAlexander,M.A.,Queen,S.E.&Adams,R.J.Real-timequantitative
PCRanddropletdigitalPCRforplantmiRNAsinmammalianbloodprovidelittle
evidenceforgeneraluptakeofdietarymiRNAs:Limitedevidenceforgeneraluptake
ofdietaryplantxenomiRs.RNABiol10,1080–1086(2013).
8.
Dickinson,B.etal.LackofdetectableoralbioavailabilityofplantmicroRNAsafter
feedinginmice.NatBiotechnol31,965–967(2013).
9.
Tosar,J.P.,Rovira,C.,Naya,H.&Cayota,A.Miningofpublicsequencingdatabases
supportsanon-dietaryoriginforputativeforeignmiRNAs:underestimatedeffectsof
contaminationinNGS.RNA20,754–757(2014).
10. Micó,V.,Martín,R.,Lasunción,M.A.,Ordovás,J.M.&Daimiel,L.Unsuccessful
DetectionofPlantMicroRNAsinBeer,ExtraVirginOliveOilandHumanPlasmaAfter
anAcuteIngestionofExtraVirginOliveOil.PlantFoodsHum.Nutr.71,102–8(2016).
11. Auerbach,A.,Vyas,G.,Li,A.,Halushka,M.&Witwer,K.Uptakeofdietarymilk
miRNAsbyadulthumans:avalidationstudy.F1000Research5,721(2016).
12. Witwer,K.W.ContaminationorartifactsmayexplainreportsofplantmiRNAsin
humans.J.Nutr.Biochem.26,1685(2015).
13. Lusk,R.W.Diverseandwidespreadcontaminationevidentintheunmappeddepths
ofhighthroughputsequencingdata.PLoSOne9,e110808(2014).
14. Kang,W.etal.Surveyof800+datasetsfromhumantissueandbodyfluidreveals
XenomiRsarelikelyartifacts.RNArna.059725.116(2017).
doi:10.1261/rna.059725.116
15. Zhang,Y.etal.Analysisofplant-derivedmiRNAsinanimalsmallRNAdatasets.BMC
Genomics13,381(2012).
16. Bağcı,C.&Allmer,J.OneStepForward,TwoStepsBack;Xeno-MicroRNAsReported
inBreastMilkAreArtifacts.PLoSOne11,e0145065(2016).
17. Witwer,K.W.&Hirschi,K.D.Transferandfunctionalconsequencesofdietary
microRNAsinvertebrates:Conceptsinsearchofcorroboration.BioEssays36,394–
7
bioRxiv preprint first posted online Mar. 25, 2017; doi: http://dx.doi.org/10.1101/120634. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
406(2014).
Zheng,L.-L.etal.Exo-miRExplorer:AComprehensiveResourceforExploringand
ComparativelyAnalyzingExogenousMicroRNAs.Front.Microbiol.8,126(2017).
Ninomiya,S.etal.PotentialsmallguideRNAsfortRNaseZLfromhumanplasma,
peripheralbloodmononuclearcells,andculturedcelllines.PLoSOne10,e0118631
(2015).
Yuan,T.etal.PlasmaextracellularRNAprofilesinhealthyandcancerpatients.Sci.
Rep.6,19413(2016).
Zhang,Z.etal.PMRD:plantmicroRNAdatabase.NucleicAcidsRes.38,D806–D813
(2010).
Zhou,Z.etal.Honeysuckle-encodedatypicalmicroRNA2911directlytargets
influenzaAviruses.CellRes.25,39–49(2015).
Yang,J.etal.Anomalousuptakeandcirculatorycharacteristicsoftheplant-based
smallRNAMIR2911.Sci.Rep.6,26834(2016).
Yang,J.,Kongchan,N.,PrimoPlanta,C.,Neilson,J.R.&Hirschi,K.D.Theatypical
genesisandbioavailabilityoftheplant-basedsmallRNAMIR2911:bulkingupwhile
breakingdown.Mol.Nutr.FoodRes.1600974(2017).doi:10.1002/mnfr.201600974
Kozomara,A.&Griffiths-Jones,S.miRBase:annotatinghighconfidencemicroRNAs
usingdeepsequencingdata.NucleicAcidsRes.42,D68-73(2014).
Griffiths-Jones,S.,Grocock,R.J.,vanDongen,S.,Bateman,A.&Enright,A.J.miRBase:
microRNAsequences,targetsandgenenomenclature.NucleicAcidsRes34,D140-4
(2006).
Jacobsen,A.,Krogh,A.,Kauppinen,S.&Lindow,M.miRMaid:aunifiedprogramming
interfaceformicroRNAdataresources.BMCBioinformatics11,29(2010).
Szcześniak,M.W.&Makałowska,I.miRNEST2.0:adatabaseofplantandanimal
microRNAs.NucleicAcidsRes.42,D74–D77(2014).
Kozomara,A.&Griffiths-Jones,S.miRBase:annotatinghighconfidencemicroRNAs
usingdeepsequencingdata.NucleicAcidsRes42,D68-73(2014).
8
bioRxiv preprint first posted online Mar. 25, 2017; doi: http://dx.doi.org/10.1101/120634. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
PutativemiRNA
Samplesw/
reads≥3
Average
total
counts
Median
total
counts
Max
Est.median
rpminavg
sample
peu-MIR2910
190
1143.4
1072.5
2020
180.6
peu-MIR2916
190
119.7
115
315
19.4
tae-MIR2005
178
8.7
9
23
1.5
peu-MIR2914
169
14.3
9
348
1.5
tae-MIR2018
167
5.9
6
14
1.0
ath-MIRf10482-akr
161
7.3
6
29
1.0
ppt-MIR896
147
4.6
4
15
0.7
ptc-MIRf12412-akr
47
2.7
2
18
0.3
peu-MIR2911
42
2.6
2
10
0.3
ppt-MIR894
39
2.3
2
18
0.3
ptc-MIRf12524-akr
28
1.8
1
5
0.2
Table1.SummaryofthemostfrequentlymappingputativeplantmiRNAsintheLiuetal
analysisofdatasetsfromYuanetal.Here,datafromonly190of192plasmasampleswere
included,sinceallbuttheexcluded2samplesweresuccessfullysequencedtwice.miRNA
inclusioncriteriawere:1)ThreetotalmappedreadsaccordingtoLiuetal’sdatainatleast10%
ofthesamplesand2)discoverableputativematuresequencethroughmiRBase,miRMaid,or
miRNEST2.0.An“estimatedmedianrpm”valuewascalculatedbasedonmediantotalcounts,
averagereads,andthemidointofthereportedmappingpercentagerange.miRNAswith
perfecthumanmatchesareinred,althoughmost(seeTable1).NotethatonlyMIR2910
consistentlyexceedstherpmthresholdsetbyYuanetal.
bioRxiv preprint first posted online Mar. 25, 2017; doi: http://dx.doi.org/10.1101/120634. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
PutativemiRNA
peu-MIR2910
Cells(n=17)andplasma(n=6)
Samples
w/
reads
≥3
21
Average
Median
370
41.5
Max
Plasmaonly(n=6)
avg
rpm
5369
21.3
Samples
w/
reads
≥3
6
Average
Median
1210
480.5
Max
avg
rpm
5369
83.2
ptc-MIRf10804-akr
17
26
26
60
1.2
0 0
0
ptc-MIRf12412-akr
17
63
61
156
2.8
0 0
0
tae-MIR2018
17
39
29
163
1.7
0 0
0
ptc-MIRf12524-akr
10
13
10.5
31
0.4
2
31
0.4
tae-MIR2005
7
17
11
69
0.4
0 0
0
ath-MIRf10045-akr
4
5
4
10
0.1
0 0
0
ath-MIRf10046-akr
4
5
4
10
0.1
0 0
0
peu-MIR2914
3
26
11
81
0.3
3
81
1.2
peu-MIR2915
3
5
4.5
8
0.0
0 0
0
16
16
33.7
19
Table2.PutativeplantmiRNAmappingfromtheNinomiyaetaldataset.Thisdataset
consistedofbothcellularandplasmamiRNA.Here,allresultsareshowninthelefthalfofthe
table,andplasmaresultsontheright.miRNAinclusioncriteriawere:1)Threetotalmapped
readsaccordingtoLiuetal’sdatainatleast10%ofthesamples(cellsandplasmatogether)and
2)discoverableputativematuresequencethroughmiRBase,miRMaid,ormiRNEST2.0.“Avg
rpm”iscalculatedfromthetotalmappedreadsandtotalreadspersample(notmappedreads).
PutativemiRNAsthatmetinclusioncriteriaintheYuanetalstudyareitalicized,andsequences
withperfecthumanmatchesareinred.
PutativemiRNA
Yuanet
alor
Sequence
Ninomiya
etal
peu-MIR2910
Both
UAGUUGGUGGAGCGAUUUGUC
No
rRNAfragment
18srRNA
NR_003286.2
peu-MIR2916
Yuan
UGGGGACUCGAAGACGAUCAUAU
Yes
Possible,but20ntmapstorRNA
Partial
tae-MIR2005
Both
GGGUGUAUAGCUCAGUUGG
No
Partial
peu-MIR2914
Both
CAUGGUGGUGACGGGUGACGGAG
No
Unlikely:plantmitochondrial
genome
rRNAfragment
Partial
tae-MIR2018
Both
GCCCGUCUAGCUCAGUUGGU
No
Partial
ath-MIRf10482-akr
Yuan
UCUACUCGACUAGGUGGUCGAGUGG
No
Unlikely;mapstomanyplant
transcripts
MapstoArabidopsis
Partial
ppt-MIR896
Yuan
GUCAAUUUGGCCGAGUGGUUAAGGC
No
tRNAfragment
Partial
ptc-MIRf12412-akr
Both
GCUGGGAUUACAGGCGUGAGCCACC
No
DoesnotmaptoPopulus
Manyexact
XR_001736898.1
peu-MIR2911
Yuan
GGCCGGGGGACGGGCUGGGA
No
rRNAfragment
NC_018921.2
ppt-MIR894
Yuan
CGUUUCACGUCGGGUUCACC
Yes
ptc-MIRf12524-akr
Both
CCUGUAAUCCCAGCUACUCGGG
No
Yes,butothertranscriptome
matches
DoesnotmaptoPopulus
1mismatchin
a20-ntstretch
No
Manyexact
NG_053018.1
ath-MIRf10045-akr
Ninomiya
UCUACUCGACCUGGUGGUCGAGUGGU
No
Unlikely;chromosomalregion
Partial
ath-MIRf10046-akr
Ninomiya
CUCGACCUGGUGGUCGAGUGGU
No
Partofabovesequence
Partial
peu-MIR2915
Ninomiya
CCCGUCUAGCUCAGUUGGUA
No
tRNAfragment;sequencefoundin
manytranscripts
Partial
ptc-MIRf10804-akr
Ninomiya
CCUGUAAUCCCAGCACUUUGG
No
Unlikely;microsatellitesequence
Manyexact,
alsohsa-miR3929precursor
(1mismatch)
XR_001736898.1
(antisense)
miRBase
miRNAstatusinplant
Human
matches?
Examplematch
Table3.PutativeplantmiRNAsmappedbyLiuetalfromtheYuanetalorNinomiyaetalstudies.Inclusioncriteriawere:1)Three
totalmappedreadsaccordingtoLiuetal’sdatainatleast10%ofthesamplesintherespectivestudiesand2)discoverableputative
maturesequencethroughmiRBase,miRMaid,ormiRNEST2.0.miRNAstatuswasconsideredunlikelyifmiRBaselistedthemiRNAas
anon-miRNAorifthesequencemappedtonon-miRNAregions.Humanmatcheswereexact(withanexamplegiven),“partial”(at
least15ntstretcheswith100%identity),orasotherwisedescribed.Notethattheath-MIRf10046-akrisfoundwithintheathMIRf10045-akrsequence.