ENGINEERINGOFFUNCTIONALLYASYMMETRICSIRNADUPLEXESByPhillipAngartADISSERTATIONSubmittedtoMichiganStateUniversityinpartialentoftherequirementsforthedegreeofChemicalEngineering-DoctorofPhilosophy2017ABSTRACTENGINEERINGOFFUNCTIONALLYASYMMETRICSIRNADUPLEXESByPhillipAngartShortinterferingRNAs(siRNAs)areapromisingnucleicacid-basedtherapeuticstrat-egythatanalternativetotraditionalsmall-moleculebaseddrugsforthetreatmentofavarietyofotherwise-untreatablediseases.Thesesmall,chemicallysynthesizedRNAscantransientlyinhibittheexpressionoftargetgenesthroughtheuseofanativeeukary-oticpathwaycalledRNAinterference(RNAi).ThedesignofhighlyactivesiRNAsisnotstraightforward.OurstrategytoimprovesiRNAdesigncriteriafocusesontheideathatsiRNAsentertheRNAipathwayasduplexesbutonlybecomefunctionalonceoneofthesiRNAstrandshasbeenselected.Ensuringhighselectivityoftheintendedstrandisessen-tialforpropersiRNAfunctionality.Toaddressthischallenge,thisresearchaimedto(1)understandhowfeaturesofthesiRNAcontroltherelativeactivitiesofthetwostrandsofthesiRNAduplex,termedfunctionalasymmetryand(2)betterunderstandthemechanismofsiRNAasymmetricstrandselection.ToimproveourunderstandingofthecharacteristicsthatdrivesiRNAstrandselectionandstrandactivity,webeganourinvestigationwithtwocharacteristics,relativeterminalhybridizationstability(and50terminalnucleotide(TN)Rank.Thesecharacter-isticshavebeenpreviouslyshowntocooperativelypredictsiRNAactivity.OuranalysisindicatesthatthesecharacteristicsarealsopredictiveofsiRNAfunctionalasymmetry.AcomparativeanalysisbetweenindividualstrandloadingandactivityshowsthattheacombinationofduplexhybridizationstabilitiesthatareimportantintheformationofRISCanditskinetics.TheTNwasfoundtosiRNAstrandactivitypost-siRNAloading,suggestingaroleinmaximizingRISChalf-life.Takentogether,theseresultsindicatethatsiRNAactivityisbysiRNA-proteininteractionsthatoccurbothpre-andpost-siRNAstrandselection.ThemechanismofsiRNAloadingandtheprotein-RNAinteractionsresponsiblefordrivingtialstrandloadingarenotfullyunderstood.Basedonevidencefromothersystems,wehypothesizedthatasymmetricstrandselectionwasdriven,atleastinpart,bypreferentialbindingoftsegmentsofsiRNAsbyRNAipathwayproteins.HerewedemonstratedthatoneRNAiprotein,PACT,preferentiallylocalizestoonesiRNAterminusofaduplexknowntobefunctionallyasymmetric.ThisindicatesthatPACTmaytherebysiRNAasymmetricstrandloadingduringinitiationofRNAi.TheworkpresentedhereidencharacteristicsofsiRNAsduplexesthatwerefoundtodrivesiRNAstrandselectionandactivity.Collectively,theseresultsinformsiRNAdesignandaddressbiologicalquestionsaboutthefunctionalaspectsoftheRNAipathway.FutureworkinthisareawillcontinuetoinvestigatehowcharacteristicsofsiRNAsimpacttheirfunctionalitywithintheRNAipathway,ultimatelyleadingtowardsahighlysetofrulesforsiRNAdesign.AmoreguidedapproachtosiRNAdesignwillfacilitatethedevelopmentofsiRNAsasatherapeuticplatform.TomyfamilyivACKNOWLEDGMENTSIamgratefultoallofthepeoplewhohavehelpedinthepreparationofthisdissertationandmygraduateeducation.Thankyoutomyadvisor,ProfessorS.PatrickWaltonforguidance,support,andfriendship.IamgratefulfortheopportunityIhavebeengiven.Tomycommitteemembers,ProfessorsChristinaChan,CharlesHoogstraten,andTim-othyWhiteheadfortheirintegralpartindevelopingtheworkpresentedhere.Iappreciatebothyourtimeandinguidingmethroughthisprocess.TotheundergraduatesIhavehadthepleasureofworkingwithandmentoring:AmandaLindgren,RebeccaJacobs,KwasiAdu-Berchie,SarahThorwall,andRebeccaCarlson.Thankyouforyourwithoutwhichthisworkwouldnotbepossible.ThankyoutothepastandpresentmembersoftheCellularandBiomolecularEngi-neeringlab,whohavebeenfriendsaswellascoworkers.AspecialnoteofthankstoAritroNath,HyunCho,AmandaMalefytandBetulBilginforyourmentorshipinthebeginningofmygraduatecareer.Toallmyfriends,whohavelistenedtometalkaboutmyresearchandneverunderstoodit,atleastyoutried.YouaretoonumeroustonamebutIamthankfulforyoursupport.Finally,tomybrother,mother,andfather,noneofthiswouldhavebeenpossiblewith-outyourloveandsupport.˘Thanks.vTABLEOFCONTENTSLISTOFTABLES...................................viiiLISTOFFIGURES..................................ixKEYTOABBREVIATIONS............................xiChapter1Introduction..............................11.1...................................11.2Background...................................21.2.1SmallRNAsofRNAi..........................31.2.2DetailsoftheRNAInterferencePathway...............41.2.3Ago2...................................71.2.4TRBPandPACT............................91.3siRNADesign..................................101.3.1FunctionalAsymmetry.........................111.3.1.1RelativeTerminalHybridizationStability.....131.3.1.250TerminalNucleotideSequence(TN)...........141.3.2mRNATargetRegion.........................151.3.3Immunogenicity.............................151.3.4Non-Sp...........................161.3.5OthersiRNADesignCriteria.....................171.3.6Non-CanonicalsiRNAStructuralDesigns...............181.3.7IncorporationofChemicalMoations...............181.4ApproachandSpAims..........................20Chapter2TerminalduplexstabilityandnucleotideidentitytiallycontrolsiRNAloadingandactivityinRNAinterference222.1Abstract.....................................222.2Introduction...................................232.3Results......................................252.3.1TNRankand3nnsiRNAFunctionalAsymmetryinCulturedHeLaCells..........................252.3.2AsymmetricsiRNAStrandLoading..................292.3.3RISCSpActivity.........................352.4Discussion....................................382.5Conclusion....................................41Chapter3siRNAAsymmetrySensingofPACT...............423.1Abstract.....................................423.2Introduction...................................43vi3.3Results......................................463.3.1FunctionalCharacterizationofRecombinantlyExpresseddsRBPs.463.3.2NS3DoesNotAsymmetricBindofsiRNAs.............513.3.3PACTAsymmetricallyLocalizestosiRNATermini.........553.3.4PACTDimerizeswithssRNAandsiRNA...............573.4Discussion....................................583.5Conclusions...................................60Chapter4ConclusionsandFutureDirections................614.1Conclusions...................................614.2FutureDirections................................624.2.1ParsingwithLargerDatasets.....................634.2.2TRBPandPACT............................65APPENDICES.....................................68AppendixAMaterialsandMethodsforCh.2...................69AppendixBMaterialsandMethodsforCh.3...................77AppendixCExpressionofBioactiveBrain-DerivedNeurotrophicFactor(BDNF)inBrevibacilluschoshinensis.........................85BIBLIOGRAPHY...................................113viiLISTOFTABLESTable2.1:AkaikeWeightsforLinearRegressionModels................32Table3.1:PACTAsymmetricsiRNABindingFromTRBP..........57Table3.2:PredicitedAsymmetricLocalizationofR2D2,TRBP,andPACT.....60TableA.1:PKR-targetingsiRNASequences.......................70TableA.2:CloningandSequencingOligos........................71TableA.3:Stem-loopPrimerSequences.........................72TableC.1:ProteinRecoveredatStagesoftheBDNFProcess......101viiiLISTOFFIGURESFigure1.1:AnatomyofsiRNAsandmiRNAs.....................4Figure1.2:RNAiMechanism..............................6Figure1.3:DomainsofRNAiProteins.........................9Figure1.4:FunctionalDomainsofGuideRNAinRISC................9Figure1.5:G3nnandTNRank....................13Figure2.1:CharacteristicsofPKR-targetingsiRNAs.................26Figure2.2:SchematicofsiRNALuciferaseAssay...................27Figure2.3:siRNAFunctionalAsymmetry.......................28Figure2.4:IC50CurvesforPKR-targetingsiRNAs..................29Figure2.5:siRNAAsymmetricStrandLoading....................31Figure2.6:IndividualsiRNAStrandActivitiesandLoading.............33Figure2.7:RelationshipBetweensiRNAFunctionalAsymmetryandAsymmetricLoading....................................35Figure2.8:CorrelationofsiRNAStrandActivitywithAgo2Loading........36Figure2.9:EvaluationofsiRNALoadingandActivityPredictors..........37Figure3.1:ConofsiRNAsforCross-linkingExperiments..........46Figure3.2:FunctionalCharacterizationofNS3....................48Figure3.3:FunctionalCharacterizationofMBP-PACT................50Figure3.4:SymmetricalCross-linkingofNS3tosiRNA................52Figure3.5:SymmetricCross-linkingofNS3tossRNA.................54Figure3.6:AsymmetricalBindingofMBP-PACTtosiRNA.............56ixFigure3.7:rentBindingModesofTRBPandPACT...............58Figure4.1:TNcanAltertheRoleofinPredictingsiRNAActivity.......64FigureC.1:PurityofBDNFStandard..........................92FigureC.2:StandardCurveofAlamarBlueFluorescence...............94FigureC.3:B:choshinensisGrowthandBDNFExpression.............96FigureC.4:BDNFSizeIndicatesLossofSecretionTag................98FigureC.5:BDNFinCellularFractionisLargelyInsoluble..............99FigureC.6:ofBDNF............................100FigureC.7:BDNFPurityAfterIMACChromatography................102FigureC.8:BioactivityofBDNF.............................104FigureC.9:BDNFEC50Curves.............................105FigureC.10:IntervalsComparingBDNFSamplestoBioactiveBDNF..106FigureC.11:BDNFBioactivityImages..........................107FigureC.12:EvidenceofBDNFAggregation.......................109FigureC.13:B:choshinensisGrowthat20°C......................111xKEYTOABBREVIATIONSAadenineAgoArgonauteANOVAanalysisofvariancebpbasepairB:ChoshinensisBrevibacilluschoshinensisBDNFbrainderivedneurotrophicfactorBSAbovineserumalbuminCcytidineC3POcomponent3promoterofRISCcDNAcodingDNACHOChinesehamsterovaryclp1cleavageandpolyadenylationfactorIsubunitCVcolumnvolumerelativeterminalhybridizationstabilityhybridizationstabilityDGCR8DiGeorgesyndromechromosomalregion8dNTPdeoxynucleotidetriphosphatemixtureDNAreoxyribonucleicaciddsdouble-strandedDTTdiothiothreitolE:coliEscherichiacolixiEDTAethylenediaminetetraaceticacidEGFPenhancedgreentproteinEMSAelectrophoeticmobilityshiftassayGguanineHEKhumanembryonickidneyHishistidineHRPhorseradishperoxidaseHSPheatshockproteinIMACimmobilizedmetalychromatographyIPTGisopropyl-D-1-thiogalactopyranosideisomiRmiRNAisoformMBPmaltosebindingproteinMedipalMedipalMerlin-Dicer-PACTliaisondomainmiRNAmicroRNAmRNAmessengerRNANanynucleotideNDnotdeterminedNGFnervegrowthfactorNS3non-structuralprotein3ntnucleotideOAS120-50-oligoadenylatesynthetase1OD600opticaldensityat600nmxiip75loynervegrowthfactorreceptorPACTPKRactivatorPAGEpolyacylamidegelelectrophoresisPAZPiwi-Argonaute-ZwillePBSphosphatesalinePCRpolymerasechainreactionPIWIP-elementinducedwimpytestisPKRproteinkinaseRpp-lucPhotinuspyralisluciferasepre-miRNAprecursor-miRNApre-RISCprecursor-RISCRBDRNAbindingdomainRBPRNAbindingproteinRIG-Iretinoicacid-induciblegeneIRISCRNA-inducedsilencingcomplexRK13rabbitkidneyepithelialcellsRNAribonucleicacidRNAiRNAinterferenceRT-qPCRreversetranscriptase-quantitativepolymerasechainReactionSDstandarddeviationSDS-PAGEsodiumdodecylsulfate-PAGESF21SpodopterafrugiperdaovariancellsxiiishRNAsmallhairpinRNAsiRNAshort-interferingRNAsodCu,Znsuperoxidedismutasesssingle-strandedTARTrans-activatingresponseTLRToll-likereceptorTmmeltingtemperatureTN50terminalnucleotideTRBPTARRNAbindingproteinTrkBtropomyosin-relatedkinaseBUuridineUTRuntranslatedregionVEGFvascularendothelialgrowthfactorWAorUnucleotidexivChapter1Introduction1.1Biologicsareaclassofpharmaceuticalsthatutilizebiologicalmacromoleculestoachieveatherapeuticct.Asalternativestotraditionalsmallmoleculedrugs,biologicshavethepotentialtoprovidetreatmentsforpreviouslyuntreatableconditions.Antisensetech-nologiesarebiologicsthatutilizenucleicacidstotargetRNA(andoccasionallyDNA)withinacelltoaltergeneexpression.Oneantisensetechnology,RNAinterference(RNAi)therapeutics,leveragestheabilityofanativeeukaryoticregulatorymechanismtoincor-poratechemically-synthesized,smallinterferingRNAs(siRNAs)todirectgenesilencingagainstatherapeutictarget.siRNAsprovideatransient,highlytargetedmethodofpost-transcriptionalgenesilencing.siRNAsareroutinelyusedinresearchstudiesofeukaryoticbiologicalprocesses,thoughtransitioningthetechnologytotheclinichasprovenchalleng-ing.PrincipalamongthechallengesarethedeliveryofsiRNAstothecellsofinterestand,thefocusofthiswork,theselectionofhighlyactivesiRNAswithgoodspey.thasbeenspentidentifyingfactorsthatenhancesiRNAactivityandspy.siRNAactivityisby,amongotherfactors,strandselection,thestruc-tureofthemRNAtargetregion,basepreferences,overallsiRNAG/Ccontent,andsiRNAduplexthermodynamics.Incomparison,siRNAspydependsonstrandselection,1immunogenicity,anduniquenessofthetargetsequence.CurrentsiRNAdesigndependsuponachievingabalanceamongthesefactors;siRNAdesignalgorithmstypicallyweightthesefactorsbaseduponanalysesofsiRNAactivitydata(NaitoandUi-Tei,2012).Asitstands,therulesforselectingactive,spsiRNAscontinuetoevolve.siRNAsentertheRNAipathwayasduplexesbutmustbesingle-strandedinordertofunction,requiringselectionofonesiRNAstrandforincorporationintotheactivecom-plex.AsymmetricstrandselectioniscriticalforproperfunctionalityofsiRNAsagainstanintendedtargetandoccursbyrecognitionofspfeaturesofthesiRNAbythepro-teinsoftheRNAipathway.TheworkdescribedherefocusesonunderstandingthesiRNAcharacteristicsthatleadtoafunctionallyasymmetricandactivesiRNAduplex.Wealsoin-vestigatedtheProteinKinaseR(PKR)Activator(PACT),whichispostulatedtofacilitatesiRNAloading,foritsabilitytoactasanasymmetrysensorintheRNAipathway.1.2BackgroundRNAiisacentralpost-transcriptionalregulatoryanddefensemechanisminmanyeukary-oticorganismsandisessentialtounderstandingtheregulatorylandscapeofeukaryoticcells.StudiesofRNAihaveinformedmechanismsofdevelopmentaldisorders,diseaseparthenogenesis,andcancerprogression.ItisalsoremarkablefortherelativeeaseinwhichexogenoussiRNAs,oncedeliveredtothecytoplasm,canentertheRNAipathwayandconstitutetargetedgenesilencing,creatingatherapeuticRNAiwasobservedinpetuniaswhenthegenethatproducesanthocyanin(thecompoundgivingpetuniastheirpurplecolor)wasoverexpressed,inanattempttoenhancethewers'color(NapoliandLemieux,1990).Instead,manywersturnedwhiteinstead2ofdeeperpurple.ItwaslaterdeterminedthatthepetuniasturnedwhiteduetoactivationoftheRNAipathway,whichrecognizedtheoverexpressedgeneasnon-nativeandsilenceditsexpression,concomitantlysilencingthenativegene(NapoliandLemieux,1990).ThediscoveryofboththeRNAtriggerandexplanationofthepost-transcriptionalsilencingef-fectwassubsequentlypublished,earningtheseniorauthorstheNobelPrizeinPhysiologyandMedicinein2006(Fireetal.,1998).In2001,RNAiwascharacterizedinmammals,withthesimultaneousdiscoveryofmultiplenative,smallregulatoryRNAs(Elbashiretal.,2001a;Lagos-Quintana,2001).NativesmallRNAshavebeenfoundtobecentralinde-velopment,cancers,infections,andotherdiseasesandremainanactiveareaofresearch(CarthewandSontheimer,2009;WittrupandLieberman,2015).1.2.1SmallRNAsofRNAiTheRNAipathwayprocessesandutilizessmallRNAstoenactsequence-sppost-transcriptionalgenesilencingandtranslationalrepression(CarthewandSontheimer,2009).ThesmallRNAtriggerconfersspyoftheactiveprotein-RNAcomplextowardsamessengerRNA(mRNA)targetwithacomplementarysequence(Meisteretal.,2004).ForenactingRNAi,smallRNAscanbegroupedintotwocategories,siRNAsandmicroRNAs(miRNAs).siRNAsarefullybasepaired˘21nucleotide(nt)duplexeswith19internalbasepairs(bp),50phosphates,and30dinucleotideoverhangs(Figure1.1)(Bernsteinetal.,2001;Zamoreetal.,2000;anenetal.,2001;Elbashiretal.,2001c).miRNAs,whilesimilarinoverallarchitecture,typicallycontainbulgesandmismatcheswithintheduplexandonlypartiallybasepairwiththeirtarget(Figure1.1)(Bartel,2009).Thepathwayitselfismadeupofrelativelyfewproteins,butthediversityinnativesmallRNAsenables3theregulationof>60%ofhumangenes(Friedmanetal.,2008).Figure1.1:AnatomyofsiRNAsandmiRNAs.(Top)ThecanonicalsiRNAstructurewith˘19bp,and30dinucleotideoverhangs.(Bottom)ApproximatemiRNAstructure,sim-ilartosiRNAs,with˘19internalbasesand30dinucleotideoverhangs,butalsocontainingbulgesandmismatcheswithintheduplex.1.2.2DetailsoftheRNAInterferencePathwayInmammals,miRNAbiogenesisbeginsinthenucleuswiththetranscriptionofaprimary-miRNA,whichisthenprocessedbytheDrosha-DiGeorgesyndromechromosomalregion8(DGCR8)complextoformaprecursor-miRNAs(pre-miRNA)(Gregoryetal.,2004;Hanetal.,2004)(Figure1.2).Pre-miRNAsareexportedtothenucleusbyexportin-5(Yietal.,2003)wheretheyarefurtherprocessedintomiRNAsbytheDicercomplex,composedofDicer,and/orTRBPandPACT,bycleavageofthepre-miRNAhairpin(Yodaetal.,2010).miRNAsarethenloadedintooneofthe4Argonaute(Ago)proteins,formingpre-RNAinducedsilencingcomplex(pre-RISC)(Yodaetal.,2010).Oncethepassengerstrandisremoved,eitherbyduplexunwindingorbycleavageanddegradationofoneofthestrands,matureRISCisformed(Matrangaetal.,2005;Yodaetal.,2010;Kawamataetal.,2009).StrandselectiondependsonitsorientationwithinAgo,dictatingwhichstrandwillbe4loaded(guidestrand)andwhichstrandwillbedegraded(passengerstrand)(Matrangaetal.,2005;Randetal.,2005).TheguideRNA-AgocomplexcomposestheminimalRISC,capableoftargetingcomplementarymRNAforgeneknockdownviaWatson-Crickbasepairing(Rivasetal.,2005).miRNAprogrammedRISCbindwithpartialcomplementaritytothe30UTRoftheirtargetmRNA,which,throughavarietyofmechanisms,causeseithertranslationalrepressionortranscriptdecay(ReviewedinBartel(2009)andAmeresandZamore(2013)).EndogenoussiRNAbiogenesisbeginswithlongdsRNAorshort-hairpinRNA(shRNA),whichdirectlyentertheDicercomplexandarecleavedintosiRNAduplexes.WhilesiRNAsarepredominatelyloadedintoAgo1andAgo2,onlyAgo2iscapableoftlyformingmatureRISC(Suetal.,2009;Guetal.,2011;Yodaetal.,2010;Meisteretal.,2004)throughendonucleolyticcleavageandremovalofthefullycomplementarypassengerstrand(Matrangaetal.,2005;Randetal.,2005;Yodaetal.,2010).siRNAprogrammedRISCbindstocomplementarymRNAand,inasimilarmannertopassengerstrandcleavageandremoval,cleavesthemRNAbetweenthenucleotidescomplementarytothe10thand11thbasesoftheguidestrand(Wangetal.,2008;Elbashiretal.,2001b).ThecleavedmRNAisthenreleasedanddegraded,allowingRISCtotargetanothermRNA(HaleyandZamore,2004).Ago2istheonlyhumanAgoproteinwithanactiveribonucleasedomainmakingitessentialtosiRNA-mediatedknockdown(Yodaetal.,2010).ExogenoussiRNAsaretypicallydesignedtomimicthematureendogenoussiRNAstruc-ture(Elbashiretal.,2001a;Zamoreetal.,2000).OncesiRNAsaredeliveredtothecytoplasm,the50terminiarerapidlyphosphorylatedbytheproteinClp1(WeitzerandMartinez,2007)andarethenloadedintoaproteincomplextoformpre-RISC(Yodaetal.,52010;Sakuraietal.,2011).Theonlyproteinrequiredforpre-RISCisoneoftheAgopro-teins(Yodaetal.,2010),although,byitself,Ago2isunabletoformmatureRISC(Yeetal.,2011).Itisunclearastowhatproteinsformpre-RISCinlivingcells;however,recombinantproteinbasedassayshaveshownthatpre-RISCactivitycanbeformedinvitrowithAgo2andeitherTRBP,Dicer,HSP70/HSP90,orC3PO(Willkommetal.,2016;Bernardetal.,2015;Iwasakietal.,2010;Liuetal.,2009;Yeetal.,2011).Figure1.2:RNAiMechanism.EndogenousmiRNAbiogenesis(bluelines)andsiRNAbiogenesis(blacklines).ExogenoussiRNAsenterthepathwaythroughthepre-RISC(dashblacklines),thecomponentsofwhicharenotcompletelyknown.Notdepictedarepre-miRNAsthataredirectlyloadedintoAgo2constitutingpathway"cross-talk"(Yangetal.,2010;etal.,2010),thelocalizationofRNAiproteins(Stalderetal.,2013),reg-ulatoryelementsofRNAi(HaandKim,2014),andthemechanismsofmiRNAmediatedgenesilencing(Bartel,2009),andmechanismsofsiRNAentryintothecell(Malefytetal.,2012).61.2.3Ago2Ago2isthecorecatalyticproteinofsiRNA-mediatedRISC(Rivasetal.,2005).ItisoneoffourhumanAgoproteinsbutitistheonlyonecapableofloadingsiRNAsandcleavingitsmRNAtargetthroughitsactivePIWIdomain(Figure1.3)(Meisteretal.,2004;Yodaetal.,2010).Ago2interactsdirectlywiththesiRNAphosphatebackbone,50terminalnucleotide(TN),andthe2ntofthe30endoftheguidestrand,which,beforesiRNAstrandseparation,composedthe30overhangoftheguidestrand(Franketal.,2010;Maetal.,2004;Parkeretal.,2009).Ago2,aspartofRISC,splitstheguideRNAintodtdomains,witheachdomainhavingatroleinRISCfunction(Figure1.4).RISCexposesnucleotides2-4oftheguidestrandtothesolvent(1°seed),usingthesebasestoscanforcomplementarytargets.Onceacomplementary3ntsequenceisfound,bases5-8engagethemRNAtarget(2°seed)(Salomonetal.,2015;Chandradossetal.,2015).Nucleotides2-8oftheguidestrandaretheminimumsequencerequiredtomaintainastableassociationbetweenRISCanditstarget;thisregionoftheguidestrandisreferredtoastheseedregion(Parkeretal.,2009;Ameresetal.,2007;Lambertetal.,2011).Agoproteinshaveanadeninebindingpocket,which,incircumstanceswherethenucleotideoppositethe1stnucleotideoftheguidestrandisanadenine,alterstheminimumrequiredseedregiontonucleotides2-7(Bartel,2009).SeedregionbindinginitiatestconformationalchangestoallowbindingoftheremainingguidestrandtothetargetmRNAfollowedbycleavageofthemRNAbetweenthe10thand11thbasesoftheguidestrand(partofthecentraldomain)(Elbashiretal.,2001b;Deetal.,2013;Wangetal.,2008;SchirleandMacRae,2012).Onlynucleotides2-16areinvolvedinfulltargetrecognition,withnucleotides13-16actingtostabilizetheRISC-7mRNAinteraction(30supplement)(Weeetal.,2013).ThenucleotideoftheguideRNAinteractsdirectlywiththeMIDdomainofAgo2(Figure1.3)anddoesnotengageinbasepairingwiththetarget(TN;Figure1.4)(Franketal.,2010).The50nucleotidebindingpocketinteractsdirectlywiththe50phosphateandthephosphatebackboneofnucleotides1-4(Suzukietal.,2015;Franketal.,2010);italsocontainsaloopstructurethatdiscriminatesagainsttheTNsequence,showingpreferenceforuridineandadenineoverguanineandcytidinenucleotides.Functionally,thesebindingpreferencesthehalf-lifeofRISC(Deetal.,2013).ThissamebindingpocketfacilitatestheloadingofthermodynamicallyunstablesmallRNAtermini(Suzukietal.,2015)(SeeChapter2).The30tailoftheguidestrandinteractsdirectlywiththePAZandN-terminaldomains(Figure1.3)anddoesnotengageintargetbinding(Maetal.,2004;Lingeletal.,2004;KwakandTomari,2012).8Figure1.3:DomainsofRNAiProteins.DepictedarethedomainsoftheessentialRNAicytoplasmicproteinswiththenumberontheC-terminusdenotingthenumberofaminoacidsineachprotein.Figure1.4:FunctionalDomainsofGuideRNAinRISC.GuideRNA-mRNAinteractionsvarydependingontheregionoftheguidestrandandaremodulatedbyAgo2.TheseinteractionshavebeenmappedinDrosophilaandmouseAgo2.1.2.4TRBPandPACTTRBPandPACTaredsRBPswithsimilararchitectures,containingtwodoublestrandedRNAbindingdomains(dsRBDs)andathirddomainthatbindstoanumberofotherpro-teins(Figure1.3),includingDicer(Larakietal.,2008;Danielsetal.,2009).The2dsRBDs9ofTRBPandPACTactindependentlyofoneanother,withdsRBD2havinggreateryfordsRNAcomparedtodsRBD1(Davietetal.,2000).WithinthecontextoftheDicercomplex(Figure1.2),TRBPandPACTactbothindependentlyandcoordinatelytofa-cilitatethegenerationofsiRNAsandmiRNAs,controllingsubstrateloadingandisomiRgenerationin(LeeandDoudna,2012;Wilsonetal.,2015;Kimetal.,2014).Thesepro-teinsalsohaveinsubstratebindingpreferences,withPACTshowingyforpre-miRNAlikestructuresandTRBPhavingnityforbothdsRNAandpre-miRNAs(Leeetal.,2013).BothTRBPandPACTfunctionincontextsoutsideoftheRNAi,playingaroleintranslationalcontrol,cellgrowthandcellcycle,andPKRactivation/suppression(DanielsandGatignol,2012).Ineachcase,TRBPandPACTinteractwitheachother,aswellastheproteinsPKRandMerlin;collectivelytheseproteinsconstituteasignalingnetworkbetweencellularRNAandtranscription(DanielsandGatignol,2012).Theseproteinsarealsoregulatedpost-transcriptionallythroughphosphorylation,ubiquination,andSUMOylationtheirfunctionwithintheRNAipathway(Parooetal.,2009;Chenetal.,2015;Leeetal.,2004,2005).1.3siRNADesignsiRNAsaredesignedtobefullycomplementarywiththeirtargetRNA;still,sequenceselectionisnotnecessarilystraightforward.Thisisduetothenumberofcriticalinter-andintramolecularinteractionswithintheRNAipathwaythatdeterminetheeventualactivityofthesiRNAinsilencingitstarget.ThefollowingsectionsdiscussfactorsthatsiRNAsequenceselectionandadditionalmothatcanbemadetosiRNAsto10improve.1.3.1FunctionalAsymmetryForsiRNAstosilencetheintendedtarget,theymustbeproperlyorientedtoensurein-corporationoftheintendedguidestrandintoRISC(Schwarzetal.,2003;Tomarietal.,2004;Gredelletal.,2010;Nolandetal.,2011;Sakuraietal.,2011;Yodaetal.,2010).Incorporationoftheunintendedstrand(i.e.,theintendedpassengerstrand)leadstotheformationofaRISCthatcannotcleavetheintendedtarget(reducingtheactivityofthesiRNAtherapeutic)andpotentiallycausingte(reducingthespyofthesiRNA)(Schwarzetal.,2003;Khvorovaetal.,2003;Grimmetal.,2006).SelectionofansiRNAtargetregionlargelydictatesthesequenceofboththeguideandpassengersiRNAstrands;becauseofthisreductioninthedegreesoffreedom,itisusefultorefertosiRNAactivitybothintermsofitsabsoluteactivityandrelativeactivityoverthepassengerstrand.WewillrefertothediintheactivityofonesiRNAstrandrelativetotheotherasfunctionalasymmetry(Schwarzetal.,2003).Functionalasymmetryisafunctionofmultiplefactors,includingRISChalf-life,RISCturnoverrate,andbiasedincorporationofonesiRNAstrandintoRISC(Franketal.,2010;HaleyandZamore,2004;Tomarietal.,2004;Suzukietal.,2015;Deetal.,2013).TwofactorshypothesizedtosiRNAfunctionalasymmetryaretherelativeterminalthermodynamichybridizationstabilityG)andtherelative50terminalnu-cleotide(TN)sequence(Figure1.5A),whichcanberankedaccordingtotheirpredictedactivity,TNRank(Figure1.5Aand1.5B)(Waltonetal.,2010;Malefytetal.,2013).BothofthesecharacteristicsarelocalizedtotheterminiofthesiRNA.ThesiRNAterminiare11criticalforauthenticationofthesiRNAstructurenotonlywiththecharacteristic2nt30overhangsand50phosphates(Elbashiretal.,2001a;anenetal.,2001),butalsobe-causethesearetheonlyregionsofthesiRNAguideknowntointeractwithAgo2andDicer(Limaetal.,2009;KiniandWalton,2007)inamannerotherthanthegenericphosphatebackboneinteractions(Franketal.,2010;Lingeletal.,2004;Suzukietal.,2015;SchirleandMacRae,2012).12Figure1.5:G3nnandTNRank(A)siRNAasymmetrycharacteristicsmarkedonansiRNAwiththetopstrand(blue)asthedesiredguidestrand(Antisense)andthebottomstrand(red)asthedesiredpassengerstrand(Sense).(B)TNrankings(fromMalefytetal.(2013)).ArrowdenotesTNRankspredictedtofavorgreaterantisense(+)strandactivity(Blue)orfavorgreatersense(-)strandactivity(Red)(+/-NotationdescribedinFigure2.2).1.3.1.1RelativeTerminalHybridizationStability(G)BiasedstrandincorporationoccursbytherecognitionofintheterminiofthesiRNAsbyvariousproteinsoftheRNAipathway(Tomarietal.,2004;Schwarzetal.,2003;13Khvorovaetal.,2003;Gredelletal.,2010;Waltonetal.,2010;Nolandetal.,2011;BetancurandTomari,2012;Kohetal.,2013).BasedonearlystudiesoffunctionalasymmetryinDrosophila,itwasconcludedthata4nn,theinhybridizationenergybetweenthetwoendsofthesiRNA(similartoFigure1.5A),waspredictiveoffunctionalasymmetryandfurtherattributedtobiasedstrandloading(Tomarietal.,2004).Thesestudiesdeterminedthatthestrandwhose50terminusislessstablyhybridized(morepositivehybridizationfreeenergy)ispreferentiallyloadedintoRISC(Schwarzetal.,2003;Tomarietal.,2004;Gredelletal.,2010;Nolandetal.,2011).MorerecentanalyseshavefoundthatfunctionalasymmetryinmammaliansystemsismorestronglypredictedbytheTN,withthe4nnbeingasecondordertoTN(Waltonetal.,2010;Malefytetal.,2013;Suzukietal.,2015)(SeeChapter2).Whenbothfactorsarecombined,a3nnismoreinformativethana4nn(Waltonetal.,2010).1.3.1.250TerminalNucleotideSequence(TN)TheTNstronglypredictssiRNAactivity,with50uridineandadeninenucleotidesbeingstronglyfavoredover50cytidineandguaninenucleotides(Franketal.,2010;Seitzetal.,2011).TheTNcontributestoRISCstabilityviaadirectinteractionwiththenucleotidespyloopwithinAgo2(Franketal.,2010;Deetal.,2013).TheincreasedyofsomeguidestrandsrelativetoothersmayresultinanincreaseinRISCstability,thehalf-lifeofRISC,andsubsequentlyhigherstrandactivity(Franketal.,2010;Deetal.,2013).TNRank,anextensionofTN,isameasureofsiRNAactivitybasedontheTNofboththepassengerandguidestrands(Figure1.5Aand1.5B)(Waltonetal.,2010;Gredelletal.,2010;Malefytetal.,2013).DespitethepartialcollinearitybetweenTNRankand3nn,thesefeaturesareindependentpredictorsofsiRNAactivity(Waltonetal.,2010;14Malefytetal.,2013;Suzukietal.,2015).1.3.2mRNATargetRegionStrongmRNAsecondarystructureprecludessiRNAbindingthroughsterichindranceofRISCbindingandcleavage(Brownetal.,2005;Ameresetal.,2007).Insilicomethodsexisttopredict,relativelyaccurately,mRNAsecondarystructures(Vickersetal.,2003;Ameresetal.,2007;Bohulaetal.,2003;Ovetal.,2005;Schubertetal.,2005;Shaoetal.,2007;Yoshinarietal.,2004);however,mRNAstructureinalivingcellisdynamicandpredictiveapproachescanonlyprovidepartialmRNAstructuralinformation.SecondarystructureinformationcaninformsiRNAdesignand,ingeneral,regionsofthemRNAwithlowoverallsecondarystructureshouldbetargeted.(Ameresetal.,2007;Ovetal.,2005;Shaoetal.,2007).1.3.3ImmunogenicitysiRNAscancauseanumberofimmunologicandcytotoxicresponses,someofwhicharesequence-spandothersthatariseduetothedsRNAstructure(Sledzetal.,2003;Samuel-AbrahamandLeonard,2010;JacksonandLinsley,2010;Robbinsetal.,2007).Infact,immunogenicresponsescanbethecauseofthetherapeuticratherthanspsilencingmediatedbythesiRNA(Haussecker,2012;Schleeetal.,2006).Ingeneral,therecognitionofspsequencemotifsthatleadstoimmuneactivationismediatedbytoll-likereceptor(TLR)recognitionofpathogen-associatedmolecularpatterns(Heiletal.,2004;Judgeetal.,2005;Jacksonetal.,2006a;Dieboldetal.,2004,2006;Goodchildetal.,2009;Weberetal.,2012;Kleinmanetal.,2008;Reynoldsetal.,2006;Karikoetal.,2004;Forsbach15etal.,2008).Immunostimulatorymotifsinclude:GUCCUUCAA(Hornungetal.,2005),UGUGU(Judgeetal.,2005),UGUJudge2005,UCAJurk2011,GU-richsequences(Heiletal.,2004),AU-richsequences(Forsbachetal.,2008),andU-richsequences(Goodchildetal.,2009).UGGChasbeenshowntobecytotoxicvianon-immunemechanisms(Fedorovetal.,2006).OAS1,PKR,andRIG-IarecytoplasmicreceptorsthatinteractwithdsRNAs,likesiRNAs,toenactaninnateimmuneresponse(Kodymetal.,2009;Mancheetal.,1992;BevilacquaandCech,1996;Marquesetal.,2006;Katoetal.,2008;Samuel-AbrahamandLeonard,2010;GantierandWilliams,2007).ThesereceptorsdetectRNAstructuralfeaturesratherthansequencemotifs.TheonlyexceptionisOAS1,whichisactivatedbyaNNWW(N)9WGNmotif,whereWcanbeeitheraAorUnucleotideandNisanynucleotide(Kodymetal.,2009).1.3.4Non-SpmiRNA-liketargetingoccurswhensiRNAshaveunintendedseedregioncomplementaritywiththe30UTRofanmRNA,resultingintranslationalrepressionortranscriptionalsi-lencingofanuntargetedtranscript(Doenchetal.,2003;Lambertetal.,2011;Guetal.,2011;Bartel,2009;Lewisetal.,2003;Linetal.,2005;Lai,2002).mRNAsthatarena-tivelyregulatedbymiRNAsarealsomoresusceptibletomiRNA-likeduetoextended30UTRsandthepreexistenceofmiRNAtargetsites(Schultzetal.,2011).AvoidingmiRNA-liketargetingiscomplicatedbyaninabilitytoaccuratelypredictseedsequences(DidianoandHobert,2006;Schultzetal.,2011;Linetal.,2005).miRNA-liketargetingisbythesurroundingsequenceofthetarget,thepositionofthetargetinthemRNA,andtherepetitivenessofthetargetsequence(DoenchandSharp,162004;Linetal.,2005;Brodericketal.,2011;Bartel,2009).Assuch,thebestwaytoaccountformiRNA-liketargetingistoavoidseedsequencesthathavealreadybeeniden-(http://www.mirbase.org)(KozomaraandGr2011).miRNA-mediatedsilencingisdosedependent(Bartel,2009);assiRNApotencyincreasesandlowerdosesofsiRNAsarerequired,miRNA-likefromsiRNAsshouldconcomitantlydecrease.1.3.5OthersiRNADesignCriteriaOtherfactorsshowntosiRNAactivityincludetheG/Ccontent(i.e.,theoverallduplexstability)ofthesiRNA(Reynoldsetal.,2004;Vertetal.,2006;Khvorovaetal.,2003),thesecondarystructureoftheguidestrand(Patzeletal.,2005;oberleetal.,2006),internalrepeats(Reynoldsetal.,2004),palindromicsequences(Hossbachetal.,2006),andpositionalbasepreferencesalongthesiRNA(Reynoldsetal.,2004;Ui-Teietal.,2004;Jaglaetal.,2005;Hueskenetal.,2005;Ladunga,2006;Shabalinaetal.,2006;AmarzguiouiandPrydz,2004;Gongetal.,2006;Takasakietal.,2004;Holen,2006;Takasaki,2009;KatohandSuzuki,2007;Vermeulenetal.,2005).Morerecently,additionalstructuralcriteria,eventotheleveloftertiarystructure(Sciabolaetal.,2012),havebeenidenasvaluableinpredictingsiRNAactivity.WhileeachofthesefactorsmaybeimportantinsiRNAdesign,eithertheiroverallisthoughttobeminimalcomparedtotheotherselectioncriteriaorthereisalackofconsensusonhowtoimplementthefeatureasaselectioncriterion.Assuch,theuseofthementionedcriteriaaregenerallyconsideredasasecondordersetofrules.171.3.6Non-CanonicalsiRNAStructuralDesignsUsingnon-canonicalsiRNAstructuresinlargepartchangesthepointofentryintotheRNAipathwayorchangesthewayinwhichtheRNAiproteinsinteractwiththenon-canonicalsiRNA(SneadandRossi,2012).Fromthebottomup,ssRNAsarecapableofbeingloadedbyAgo2,albeittly,toformanactiveRISCinvitro(Martinezetal.,2002;Holenetal.,2003;Birminghametal.,2006;Rivasetal.,2005)andinvivowhenchemicallymo(Limaetal.,2012;Haringsmaetal.,2012).ByonlyprovidingonesiRNAstrandtoenterRISC,properstrandselectionisnolongeradesigncharacteristic.Otherstructuresaredesignedtobiasstrandincorporationbyalteringthelengthofthepassengerstrand(asymmetricinterferingRNAsandasymmetricshort-duplexsiRNAs(ChuandRana,2008;Sunetal.,2008;Changetal.,2009)),thelengthofthe30overhang(fork-siRNAs(Hohjoh,2004)),orbyassemblingaduplexwithasegmentedpassengerstrand(smallinternallysegmentedinterferingRNAs(Bramsenetal.,2007)).Amorerecentlytestedstructureutilizesabulgeatthesecondpositionoftheguidestrand,creatingaperturbationatthebaseoftheseedregion(Duaetal.,2011);theuseofthismotothecanonicalsiRNAstructurewasshowntodecreasebymiRNA-likeactivity.Othernon-canonicalstructuresexploittheabilityforlongerRNAstoenterRISCmoretly;thesestructuresarecalledDicersubstrateRNAs(Amarzguiouietal.,2006;Collingwoodetal.,2008;Tanudjietal.,2010;Fosteretal.,2012).1.3.7IncorporationofChemicalMotionsChemicalmoinsiRNAscanincreasetheirstability(spwithregardstonucleasedegradation),minimizeimmunogenicity,and,toacertainextent,improvethe18activityofthesiRNA(Turneretal.,2006;Volkovetal.,2009;Hongetal.,2010;Robbinsetal.,2007;Hornungetal.,2005;Goodchildetal.,2009;Allersonetal.,2005;Bramsenetal.,2009;Kenskietal.,2012).ChemicalmoarealsonecessaryfortheuseofsiRNAsinclinicalapplications.ChemicalmonsofthesiRNAfocusonchangingthephosphodiesterbackbone,ribosesugar,nucleotidebase,and20-OHribosegroup.euseofchemicalmorequiresthesubstitutionofthenewchemicalmoietyatapositionwithinthesiRNAwheretheadditionalgrouporstructuralalterationdoesnotinhibitnormalsiRNAfunction.Ingeneral,therationalebehindchemicalmoistoincorporatesmallperturbationsinthesiRNAstructuretopreventrecognitionand/orbindingofthesiRNAbynucleasesandtheimmunereceptorsforRNA.Themostcommonmoincludealteringthe20-OHgrouptoa20-O-CH3or20-FtopreventrecognitionoftheRNAbynucleasesandTLR7andTLR8(Braaschetal.,2003;ChiuandRana,2003;Allersonetal.,2005;Manoharanetal.,2011;Cekaiteetal.,2007;Robbinsetal.,2007;Tluketal.,2009;Fucinietal.,2012).TLR3interactswithdsRNA,and,whileTLR3activationbycanonicalunmosiRNAshasbeenshowninmice,itmaynotbeanissueinhumans(Reynoldsetal.,2006;Karikoetal.,2004;Kleinmanetal.,2008;Weberetal.,2012).Unfortunately,noabsoluterulesexistingwhichchemicalmonsaremostusefulandhowtheyarebestapplied.The30overhangsarealsoacommonlocationforchemicalmofortworeasons:i)theymayprovideasiteofattackforendoribonucleasesandii)chemicalmoevenbulkyones,aretypicallywelltoleratedatthesepositions(Haupenthaletal.,2006).PhosphorothioateandphosphorodithioatemotothebackboneofthesiRNAgeneratesiRNAswithnucleaseresistance,butthenumberandpositionsofmo19areimportantinretainingsiRNAactivity(Braaschetal.,2003;Amarzguioui,2003;Yangetal.,2012).1.4ApproachandSpAimsTheworkdescribedhereaimedtoimprovethedesignofsiRNAsforbothhigheractivityandspybyfocusingonthementofthefeaturesthatdeterminesiRNAfunc-tionalasymmetry.ThisdissertationdetailstwoapproachestakentowardsunderstandingsiRNAduplexcharacteristicsandmolecularinteractionsthatleadtoahighlyfunctionalasymmetricsiRNAduplex.ThespaimsaddressingthequestionofsiRNAfunctionalasymmetryareoutlinedbelow:1.UnderstandtheinterrelationofTNRankand3nnonsiRNAstrandloadingandsiRNAfunctionalasymmetry.TNRankandG3nnarecharacteristicsofsiRNAduplexesareepredictorsofsiRNAactivity.ThesecharacteristicswereshownheretoalsopredictofsiRNAfunctionalasymmetry.WefurtherexploredhowthesecharacteristicsimpactsiRNAstrandloadingandfoundthat3nnterm,whileusefulforpredictingsiRNAfunctionalasymmetry,ac-tuallycontainspredictiveinformationaboutthesiRNAduplexenergiesthatareimportantforsiRNAloadingandfordescribingsiRNA-mRNAinteractions.DiscriminationoftheTNwasalsofoundtooccurpost-loading.ThesendingsprovideamoreaccuratedescriptionofhowduplexthermodynamicsbothsiRNAloadingandactivity.2.CharacterizetheextenttowhichPACTcansenseasymmetriccharacteristicsofsiRNAduplexes20TheproteinsTRBPandPACTaremammalianhomologsoftheproteinR2D2,aproteincreditedwithsensingsiRNAduplexasymmetryinDrosophila.TRBPhasbeenpreviouslyshowntosensesiRNAduplexasymmetry(Gredelletal.,2010).HereweinvestigatedtheasymmetrysensingofthePACTprotein,showingthatittoocanasymmetricallylocalizetosiRNAterminioffunctionallyasymmetricsiRNAduplexes.Furthermore,wewereabletoshowthatPACTterminuslocalizationwasnotidenticaltoTRBPforallofthesequencesexamined,suggestingthatTRBPandPACTdonotfunctioninanidenticalmanner.Lastly,wefoundthatPACTsiRNAterminuslocalizationcanbepredictedby4nn.21Chapter2TerminalduplexstabilityandnucleotideidentitytiallycontrolsiRNAloadingandactivityinRNAinterference2.1AbstracttshortinterferingRNA(siRNA)-mediatedgenesilencingrequiresselectionofase-quencethatiscomplementarytotheintendedtargetandpossessessequenceandstructuralfeaturesthatencouragefavorablefunctionalinteractionswiththeRNAinterference(RNAi)pathwayproteins.Here,weinvestigatedhowterminalsequenceandstructuralcharacter-isticsofsiRNAscontributetosiRNAstrandloadingandsilencingactivity,andhowthesecharacteristicsultimatelyresultinafunctionallyasymmetricduplexinculturedHeLacells.OurresultsreiteratethatthemostimportantcharacteristicindeterminingsiRNAactivityisthe50terminalnucleotideidentity.OurfurthersuggestthatsiRNAloadingiscontrolledprincipallybythehybridizationstabilityofthe50terminus(Nucleotides:1-2)ofeachsiRNAstrand,independentoftheopposingterminus.Post-loading,RISCspe-22activitywasfoundtobeimprovedbylowerhybridizationstabilityinthe50terminus(Nucleotides:3-4)oftheloadedsiRNAstrandandgreaterhybridizationstabilitytowardsthe30terminus(Nucleotides:17-18).Concomitantly,sprecognitionofthe50termi-nalnucleotidesequencebyhumanAgo2improvesRISChalf-life.TheseindicatethatcarefulselectionofsiRNAsequencescanmaximizeboththeloadingandthespactivityoftheintendedguidestrand.2.2IntroductionTheRNAinterference(RNAi)pathwayisacentralregulatorymechanisminmammaliancells,whereendogenoussmallRNAs,onceincorporatedintoaribonucleoproteincomplex,targetcomplementarysequencesandcausepost-transcriptionalgenesilencing(SneadandRossi,2010;CarthewandSontheimer,2009).ShortinterferingRNAs(siRNAs),asubsetofsmallRNAsrecognizedbytheRNAipathway,areselectivelyloadedintotheArgonaute2(Ago2)protein(Yodaetal.,2010),anendoribonuclease(Liuetal.,2004),andenactgenesilencingthroughbindingandcleavingcomplementarymRNAs(Martinezetal.,2002;Elbashiretal.,2001c).TheabilitytodesignanddeliverexogenoussiRNAsfortargetedproteinknockdownanattractivetherapeuticstrategy(Angartetal.,2013).RNAiisinitiatedwhenthepathwayproteinsrecognizeRNAduplexeswiththechar-acteristicsiRNAstructure,˘19internalbasepairsand2nt30overhangs(Elbashiretal.,2001c;Zamoreetal.,2000).EitherstrandfromtheduplexcanbeprocessedandloadedintoAgo2toformthematureRNA-inducedsilencingcomplex(RISC)(Rivasetal.,2005).ThequantityofintendedRISCformedanditsactivitydeterminetheofansiRNAinsilencingitstarget.WhilegenerallyonlyonesiRNAstrand(theguidestrand)cantar-23getaspmRNA,insystemswheretargetsexistforbothsiRNAstrands,bothstrandscansilence(Yodaetal.,2010;Sakuraietal.,2011).Evenintheabsenceofasptarget,loadingandfunctionoftheunintendedstrand(thepassengerstrand)canresultintnon-sp(Jacksonetal.,2003,2006b).Thus,toachievemaximalactivityandspyofsilencing,itisessentialtodesignsiRNAsforloadingandactivityofonlytheintendedguidestrand.Todosorequiresanunderstandingofhowtocontroli)selectionandloadingoftheguidestrandandii)thehalf-lifeandactivityofRISC.Concomitantly,thesesamedesignrulesshouldbeappliedtominimizingtheloadingandactivityofthepassengerstrand,therebyachievingmaximalfunctionalasymmetry,whichweastheratioofthesilencingactivitiesofthesiRNAstrands.TheconceptofasymmetryhasbeenessentialtodesigningsiRNAssincetheirinitialstructure(double-stranded)andtheirfunctionalform(single-stranded)weredetermined(Schwarzetal.,2003).FactorsthattiatebetweenthetwosiRNAstrandshavebeeninvestigated,withtheobjectiveofdetermininghowtheRNAipathwayproteinsidentifytheguideandpassengerstrandofeachsiRNA.Thefeaturediscoveredforpredict-ingsiRNAasymmetrywasrelativeterminalhybridizationstability(Figure1.5A)(Schwarzetal.,2003),whichcapturestherelativestabilityoftheduplexstructuresatthesiRNAtermini.wasshowntopredictthefunctionalasymmetryofsiRNAs(Schwarzetal.,2003;LuandMathews,2008)andtheirasymmetricbindingbyproteinsoftheRNAipathway(Tomarietal.,2004;Nolandetal.,2011).Inourpriorwork,wesiRNAasymmetriesaccordingtothe50terminalnucleotide(TN)oneachstrand(TNRank)(Fig-ure1.5)(Gredelletal.,2010;Malefytetal.,2013;Waltonetal.,2010),showingthisfeaturetobemorepredictiveofsiRNAactivitythan24However,usingactivityasthereadoutminimizesthedegreetowhichidenfeaturescanbelinkedtospstepsintheRNAimechanism.OurgoalsinthepresentstudyweretounderstandwhetherTNRankandpredictsiRNAstrandselectioninadditiontoactivityandtoquantifythedegreetowhichstrandselectiondeterminestheactivityofansiRNAstrand.OurresultsshowedthatbothTNRankandpredictasymmetricstrandloading,albeitlessaccuratelythanpredictingsiRNAfunctionalasymmetry.WhiletheofTNonstrandactivityisprincipallypost-loading,presumablythroughinteractionsbetweentheguidestrandandAgo2,hybridizationstability,particularlyatthe50terminus,bothstrandselectionandpost-loadingevents(e.g.,interactionswiththemRNAtarget).ThepresentedhereincreaseourunderstandingofsiRNAstructure-functionrelationshipsandcanbeusefulintheselectionofactiveandspsiRNAs.2.3Results2.3.1TNRankandG3nnsiRNAFunctionalAsym-metryinCulturedHeLaCellsOurpreviousworkdescribedtheimportanceofTNRankand3nn(Figure1.5)inpredictingsiRNAactivity(Malefytetal.,2013).ThesetofsiRNAschosentoinvestigateTNRankand3nnwasselectedtopossessfavorable,neutral,andunfavorablerankingsofthetwofeatures,withfavorablereferringtoacharacteristicthatwouldbeassociatedwithhighersilencingofthesiRNAagainstitstargetmRNA,inthiscasePKR(Figure2.1)(Malefytetal.,2013).Here,wewishedtodetermineiftheseparameterscouldalso25predictsiRNAfunctionalasymmetry,asthemosthighlyactiveandspsiRNAswillhavefunctionalasymmetrieshighlybiasedinfavoroftheintendedguidestrand.Figure2.1:CharacteristicsofPKR-targetingsiRNAs.Featurepredictedtofavorgreaterantisense(+)strandactivity(Blue),featurepredictedtofavorgreatersense(-)strandactivity(Red).siRNAsarenamedbythe50positiononthePKRmRNAtargetedbythesiRNA(+)strand.siRNAsequencesareprovidedinTableA.1.WeinitiallymeasuredtheactivityofeachsiRNAstrandusingaluciferasereporteras-saywherethefull-lengthcodingsequenceofPKR(hereafterthe(+)target,targetedbythe(+)strandofeachsiRNA)orthefull-lengthcomplementsequenceofPKR(hereafterthe(-)target,targetedbythe(-)strandofeachsiRNA)wasclonedasafusionproductdownstreamoftheRenillaluciferasegene(Figure2.2).siRNAstrandIC50valuesweredeterminedbyco-transfectingHeLacellswithoneoftheplasmidsandoneofthePKR-targetingsiRNAs.FunctionalasymmetrywascalculatedastheratioofIC50valuesfor26eachpairofcomplementarystrands(Figure2.3A).TwosiRNAstrands,PKR816(-)andPKR928(+),didnotdisplayanymeasurableactivity(Figure2.4);forthesesiRNAsfunc-tionalasymmetrywasestimatedassuminganIC50of10nM,thehighestconcentrationtested,fortheinactivestrand.ExaminingourresultsforbothTNRankand3nn,wethat,aswefoundwithpredictingstrandactivity,TNRankisastrongerpredictoroffunctionalasymmetrythan3nn(Figures2.3Band2.3C),withtheinformationpro-videdby3nnimprovingthecorrelationbutbeingttoprovideatcorrelationalone(Figures2.3Cand2.3D).Theimportanceofthe3nnismostevi-dentforsequenceswithintermediateTNRank,wherepositive3nnsequencesfavor(+)strandactivitieswhilenegative3nnsequencesfavor(-)strandactivities(Figure2.3A).Figure2.2:SchematicofsiRNALuciferaseAssay.LuciferasereporterconstructsweredesignedwitheitherthePKRcodingsequence,(+)target(Blue),orPKRtemplatesequence,(-)target(Red),cloneddownstreamoftheRenillaluciferasegene.ReporterconstructsweretransfectedindependentlywitheachPKRtargetingsiRNAtodeterminetheactivityofeachsiRNAstrand.27Figure2.3:siRNAFunctionalAsymmetry.(A)RelativesiRNAfunctionalasym-metrysortedlefttorightbyTNRankandthenbydecreasing3nn.ShownaretheratiosofsiRNA(-)to(+)strandIC50valuesdeterminedfromcurveto4biologicalreplicates.Errorbarsare1SD.(*)representsvalueswhereonesiRNAstrandIC50wasnotmeasurableandassumedtobe10nM(thehighestconcentrationtested)forcalculatingfunctionalasymmetry.(B)DistributionofsiRNAfunctionalasymmetrywithrespecttoTNRank.HorizontallinesrepresentthemeanactivitywithintheTNRank;overheadbarmarkedwith(*)representsastatisticallytpairwisecomparison(p<0.05)byone-wayANOVAwithTukey'sposthocanalysis.(C)Correlationoffunctionalasymmetrywith3nn.(B,C)siRNAsPKR816andPKR928hadoneestimatedIC50andweremarkedwith(+)andnotincludedinanystatisticalanalysis.(D)MultiplelinearregressionofsiRNAfunctionalasymmetrywithTNRankand3nn,whereYisthefunctionalasymmetry((-)IC50:(+)IC50)andXiarethevariables.CotswerenormalizedtoRank1.28Figure2.4:IC50CurvesforPKR-targetingsiRNAs.(+)siRNAstrandisshowninblueandthe(-)siRNAstrandinred.2.3.2AsymmetricsiRNAStrandLoadingTobetterunderstandthestepsofthemechanismthat,onaggregate,resultinfunctionalasymmetry,wemeasuredsiRNAstrandloadingintoAgo2bytransfectingHeLacellswithoursiRNAofinterestfor24h,immunoprecipitatingAgo2,andthenmeasuringitsasso-29ciatedRNAbystemloopRT-qPCR.WeobservedthatTNRankcorrelatedwithsiRNAasymmetricstrandloading(Figures2.5Aand2.5B).3nnwasnotcorrelatedwithsiRNAasymmetricstrandloading(similartoitslackofcorrelationwithfunctionalasym-metry)(Figure2.5C).Inthiscase,however,itprovidedconsiderablylesspredictiveinfor-mationwhencombinedwithTNRank(Figure2.5D)thanitdidforrelativestrandactivity,asshownbytherelativeinAkaikeweightsforthecombined2-factormodelsandtheirrespectivesinglefactormodels(i.e.,asmallerinAkaikeweightsindicatesasmallergaininmodelinformationcontent;Table2.1).30Figure2.5:siRNAAsymmetricStrandLoading.(A)RelativesiRNAstrandloadingintoAgo2sortedlefttorightbyTNRankandthenbydecreasing3nn.ShownaretheratiosofsiRNA(+)strandto(-)strandloading.N=3,errorbarsare1SD.(B)DistributionofsiRNAasymmetricloadingbyTNRank.HorizontallinesrepresentthemeanactivitywithintheTNRank;overheadbarsmarkedwith(*)representstatisticallytpairwisecomparisons(p<0.05)byone-wayANOVAwithTukey'sposthocanal-ysis.(C)CorrelationofsiRNAasymmetricstrandloadingwith3nn.(D)MultiplelinearregressionofasymmetricstrandloadingwithTNRankand3nn,whereYistherelativestrandloading((+)[RISC]:(-)[RISC])andXiarethevariables.CotswerenormalizedtoRank1.31Table2.1:AkaikeWeightsforLinearRegressionModels.andAkaikeweightcomparisonsof1-and2-factorlinearregressionsusingTNRankand3nn.ForanumberofsiRNAs,weobservedintherelativeloadingsandrelativeactivities.PKR1129,whichexhibitedsymmetricalstrandactivity(p=0.24,extrasum-of-squaresFtest),hadasymmetricstrandloading(p=0.02,two-tailedt-test)inlinewithitsTNRank.Interestingly,itisthe(-)strandofPKR1129thathasanunexpectedlyhighactivitycomparedtotheothersequenceswitha50Gandsimilarloading(PKR410(-)andPKR952(-);Figures2.6Aand2.6B).Thisispresumablybecausethissequencehasa3nnstronglyinfavorofthe(-)siRNAstrand,whichmaypartiallyitsunfa-vorable50nucleotide.Additionally,sequencesPKR336,PKR379,andPKR440displayedsymmetricalstrandloadingsandasymmetricactivities(Figures2.3Aand2.5A).AswithPKR1129,thesesiRNAshavelargeG3nnvalues(eitherhighlypositiveornegative),suggestingthat3nnmaytlysiRNAfunctiononlyafterexceedingathresholdmagnitude.32Figure2.6:IndividualsiRNAStrandActivitiesandLoading.(A)IndividualsiRNAstrandIC50values(pM);N=4;errorbarsare1SD.DottedlinerepresentsmaximumsiRNAconcentrationassayed,valuesabovethelineareextrapolatedfromex-perimentaldata.NoactivitywasobservedforsiRNAstrandPKR816(-)andPKR928(+);denotedwith(*).(B)IndividualsiRNAstrandloadingnormalizedtoaninternalstan-dard;N=3;errorbarsare1SD.TNRankscorrespondtonucleotidepairspresentedinFigure1.5B.siRNAstrandactivity(C)andloading(D)groupedbyTN.HorizontallinesrepresentthemeanactivitywithintheTN;overheadbarsmarkedwith(*)representstatisticallytpairwisecomparisons(p<0.05)byone-wayANOVAwithTukey'sposthocanalysisAdirectcomparisonofsiRNAstrandloadingwithitsactivitydemonstratesthatwhilesiRNAloadingandactivityarecorrelated,multiplestrandsdivergetlyfromthetrend(Figure2.7).Thesesuggestsomemodulationofactivitypost-siRNAloading.33Furthermore,thechangeinAkaikeweightsbetweentheindividualfactors(TNRankand3nn)andthecombinedmodelindicatesthatTNRankand3nnarecomplemen-tarypredictorsoffunctionalasymmetry(Table2.1).Incomparisontoactivity,G3nnaddslesscomplementaryinformationtothepredictionofasymmetricstrandloading(Table2.1),indicatingthat3nncontainsinformationmorerelevanttopost-loadingevents.Forclarity,aswediscusspost-loadingevents,wewilluseRISCspactivitytorefertotheenzymaticcharacteristicsofamatureRISCthatareindependentoftheamountofRISCgenerated.34Figure2.7:RelationshipBetweensiRNAFunctionalAsymmetryandAsym-metricLoading.FunctionalasymmetryvaluesarethesameasFigure2.3Aandasym-metricloadingvaluesarethesameasFigure2.5A.Solidblacklineisasemi-logtothedata.Dottedgreylinesarethe95%intervalsontheDiamonds(PKR816andPKR928)representpointsforwhichoneoftheIC50valuescouldnotbeandwasestimatedtobe10nM,thehighestconcentrationtested.2.3.3RISCSpActivityBecauseRISCspactivityisafeatureofthesiRNAaftertheseparationofthetwosiRNAstrands,weanalyzedthebehavioroftheindividualsiRNAstrandsindependentoftheircomplements(Figure2.8).Likewise,inlieuofTNRankand3nn,weexaminedTN(U,A,C,G)andtheenergiesofallnearest-neighborpairs(x)(y)nt)alongthelengthoftheduplex(Figure2.9).Whilesimilar,thenucleotidepreferencesforloading35(A>U>C>G)(Figure2.6D)aretfromthoseforactivity(U>A>C>G)(Figure2.6C),indicatingthatTNimpactsRISCspactivity,asothershavepreviouslyobserved(Deetal.,2013).However,TNalonedoesnotaccountforalloftheobservedvariationinactivities;forinstance,siRNAswith50adenineswithsimilarloadingvariedsubstantiallyintheiractivities(Figure2.8).Figure2.8:CorrelationofsiRNAStrandActivitywithAgo2Loading.XvaluesareIC50values;N=4;errorbarsare1SD.YvaluesarethenormalizedaveragesiRNAloading;N=3;errorbarsare1SD.Dottedlinesrepresentthemedianofeachdataset.36Figure2.9:EvaluationofsiRNALoadingandActivityPredictors.1-factorand2-factorcorrelationsofsiRNAloadingandactivitywithTNand(x)(y)ntalongthelengthoftheduplex.2-factorcorrelationsuseTNasonevariableanda(x)(y)ntvalueastheother.inthesecorrelationsindicatesthatthevariablepredictsloadingoractivityindependentofTN.ABonferronicorrectionwasappliedtocorrectformultiplecomparisons.ToidentifyregionsofthesiRNAdupleximportantforcontrollingRISCspactivity,wecorrelatedloadingandactivitywiththeTNandalongthelengthoftheduplex,lookingforregionsinthedupleximportantforsiRNAactivitybutnotimportantforRISCloading(Figure2.9).Wefoundthatonlythenearest-neighborparameterssurroundingtheTN,(2)1ntand12nt,correlatedwithactivity(Figure2.9).TodecoupletheoftheTNsequencefromweusedmultiplelinearregressionandre-evaluatedthecorrelationbetweenandbothsiRNAloadingandactivity.Wefoundthat34ntand1718ntwereonlypredictiveofsiRNAactivitynotloading(Figure2.9).ThisindicatesthatRISCspactivityisimprovedforsequenceswithweaker(lessnegative)34ntandstronger(morenegative)1718nt.Thesevaluesareincludedinthecalculationof3nnandlikelycontributetoitsutilityinpredictingsiRNAactivity.372.4DiscussionDesigningsiRNAsformaximalfunctionrequiresthedouble-strandedstructuretotlyentertheRNAipathwaysuchthatitcanbeeasilyprocessedtoRISCloadedwiththecor-rectsingle-strand.ItisthereforecriticaltounderstandhowtheRNAipathwayproteinssensespfeaturesofsiRNAduplexesthattiatethetwosiRNAstrands.siRNAfunctionalasymmetrywascharacterizedinDrosophilalysatesandattributedtothein4nn(Schwarzetal.,2003).ManyArgonauteproteinscontainanaddi-tionalstructurethatpreferentiallybinds50terminalnucleotides(Franketal.,2010).OurpreviousworksoughttoexplainhowthesetwoterminalfeaturescontributetosiRNAac-tivity(Waltonetal.,2010;Malefytetal.,2013).Here,weshowedhowTNRankand3nncontributetosiRNAfunctionalasymmetryandasymmetricstrandloading.Fur-thermore,wehaveshownthat,inculturedHeLacells,functionalasymmetryismodulatedbothbeforeandaftersiRNAloading,withTNRankand3nnexertingattstagesofthemechanism.Functionalasymmetryandasymmetricstrandloadingwerenotperfectlycorrelated,indicatingadditionalmodulationofsiRNAactivitypost-siRNAloading(Figure2.7).Ourtop-downanalysisidentfreeenergiesattwopositions,34ntand1718nt,tobepredictiveofsiRNAactivitybutnotsiRNAloading(Figure2.9).MatureRISCsplitssiRNA-targetmRNAinteractionsinto5domains,the50anchor(Nucleotide:1),theseedregion(Nucleotides:2-8),thecentralregion(Nucleotides:9-12),the30supplementalre-gion(Nucleotides:13-16),andthe30tail(Nucleotides:17-21)(Weeetal.,2013).Thepreferenceforweakerthermodynamicstabilitythroughnucleotides1-4ofthesiRNAsiswellestablished(Schwarzetal.,2003;Reynoldsetal.,2004;AmarzguiouiandPrydz,2004;38Hibioetal.,2012);oursuggestthatweakerhybridizationinthisregionleadstogreaterRISCspactivity(Figure2.9),inagreementwiththathighstabilityintheseedregiondecreasesRISCturnover(Salomonetal.,2015).RISC-mRNAinteractionsbeyondthe30supplementalregionhavebeenshowntobedispensablefortargetcleavageorinsomecasesevendeleterious(Deetal.,2013;Weeetal.,2013).Wecontendthatthepreferenceforastronger1718ntisalsoimportantinmaintainingastableinteractionbetweenRISCanditstarget,similartobasepairsinthe30supplementalregion(Weeetal.,2013).AfterdeliveryofthesiRNAtothecellcytoplasm,tguidestrandloadingrequiresnucleotide-independentphosphorylationofthesiRNA50terminalnucleotides(WeitzerandMartinez,2007),outcompetingnativemiRNAsforbindingtoAgo2(Kolleretal.,2006;Tanudjietal.,2010),conformationalchangeswithinAgo2facilitatedbyHSP90andpo-tentiallyotherproteins(Iwasakietal.,2010;Sakuraietal.,2011;Bernardetal.,2015;Willkommetal.,2016;Yeetal.,2011),andcleavageandremovalofthepassengerstrand(Matrangaetal.,2005;Leuschneretal.,2006).WeobservedalargevariationinthetotalamountofAgo2loadingevenamongsiRNAstrandswiththesameterminalnucleotide(Figure2.6B).Inanalyzingothercharacteristicsfortheirabilitytopredictloading,wefound(2)1ntand12ntofthe50endoftheguidestrandtobethemostpredic-tivefeaturesofsiRNAstrandloading(Figure2.9).Interestingly,therewasnocorrelationbetweensiRNAstrandloadingandthevaluesnearthe30terminus(i.e.,Nucleotides17-21),suggestingthattheinteractionsofAgo2areprimarilywiththe50endsofsiRNAs.Thus,parametersbasedonrelativestrandproperties(e.g.,TNRankandarelessusefulinpredictingabsolutesiRNAstrandloadingandactivitythaninpredictingrelative39strandactivitiesandloadings.TheestablishedbindingpreferencesfortheAgo2nucleotidespyloop(U>A>C>G)matchtheorderofsiRNAactivitiesinourdata(Figure2.6C)butfromtheload-ingpreferencesweobserved(A>U>C>G)(Figure2.6D).Thissuggeststhat50TNRISCspactivity,presumablythroughinRISChalf-life(Deetal.,2013),similartowhatwasobservedinDrosophilaAgo1(Kawamataetal.,2011).Thecorre-lationofsiRNAloadingwiththe50terminal(2)1ntand12ntindicatesthatsiRNAloadingiscontrolledinpartbythe50terminalduplexenergy(Figure2.9),thoughTNremainsatpredictorofsiRNAstrandloadingbecauseofitscollinearitywiththe(2)1ntand12ntterms.Recently,sensingofsiRNAstrandthermodynamicswasproposedtooccurthroughAgo2MIDdomaininteractionswiththephosphateback-boneofthefournucleotidesofthesiRNAguidestrand(Suzukietal.,2015),sensingtheaccessibilityandsingle-strandcharacterofweaklybasepairedtermini(Suzukietal.,2015).Collectively,ouragreewiththismodelbutsuggestthattheandsecondnucleotideofthesiRNAguidestrandarethemostcantindeterminingtotalstrandloading(Figure2.9).siRNAdesignalgorithmsshouldthusaccountforrelativesiRNAfea-tures(TNRankandtoensurecorrectstrandselectionandabsolutestrandfeatures(TNandtoensurehighstrandloadingandRISCspactivity.WhileourexperimentsandanalysiscouldnotcovertheentiresiRNAfeaturespace,thenumberofsiRNAstestedwaslargeenoughtoidentifyregionsofthesiRNAthatexplainsomeoftheerencesbetweenactivityandloading.50Cand50GsequencesareunderrepresentedinourindividualsiRNAstrands;thus,amorecomprehensivesetofsiRNAscouldpotentiallyhaveprovidedmoreinsightintohowfeaturesbeyondthe5040TNisiRNAbehaviorbothpre-andpost-loading.Nonetheless,theparametersidenindicatethatitispossibletodesignsiRNAswiththegoalofhighloading,highactivity,orboth.2.5ConclusionTakentogether,oursuggestthatsiRNAloadingandactivityarepartiallyinde-pendent,allowingforanadditionaldegreeoffreedominsiRNAselectionanddesign.ThebehaviorofsiRNAstrandsisstronglydependentonthe50nucleotide,buttheduplexhybridizationenergieshaveasecondordertonbothsiRNAloadingandRISCspe-activity.Sp,(2)1ntand12ntarepredictiveofsiRNAloading,and34ntand1718ntarepredictiveofRISCspcactivity.Lastly,wehaveshownthatsiRNAfunctionalasymmetryisaresultofmanycompetingfactorsthatultimatelycontroltheactivityofeachsiRNAstrand.AmoreaccurateunderstandingoftheinterplayamongthesefactorswillleadtobettersiRNAs.41Chapter3siRNAAsymmetrySensingofPACT3.1AbstractRNAinterference(RNAi)isapost-transcriptionalregulatorymechanismfoundinmanyeukaryotesthatutilizesasmallRNA,suchasamiRNA,todirectsilencingtowardsacom-plementarymRNAtarget.siRNAsaredesignedtoentertheRNAipathwayimmediatelybeforetheformationoftheactiveRNA-inducedsilencingcomplex(RISC).ThetransitionfromsiRNAtoanactivesiRNA-programmedRISCrequirestheselectionofonestrandoftheduplexandtheremovaloftheotherstrand,allowingtheselectedstrandtobasepairwithitstarget.TheproteinsinvolvedinsiRNAloadingarenotfullycharacterized,sp,inregardstohowproteinsoftheRNAipathwaycontributetothepreferentialselectionofonesiRNAstrand.Previously,weinvestigatedTRBPforitsabilitytoactasansiRNAasymmetrysensor.HereweinvestigatedPACT,ahomologueofTRBP,foritsabilitytoasymmetricallylocalizetothesiRNAtermini,abehaviorindicativeofsiRNAasymmetrysensing.WefoundthatlikeTRBP,PACT,alsoasymmetricallylocalizestosiRNAterminibutthatitspatternoflocalizationtosiRNAterminifromTRBPforthesetofsiRNAsstudied.PACTwasfoundtolocalizetothesiRNAterminuswithgreaterhybridizationstability,asbythe4nnmodelforsiRNAasymmetry.423.2IntroductionShortinterferingRNAs(siRNAs)provideamethodofpost-transcriptional,targetedgeneknockdownthatisofinterestasatherapeuticstrategyforthetreatmentofavarietyofdiseases(BobbinandRossi,2016).Geneknockdownorsilencingoccursthroughincorpo-rationofansiRNAintothenativeeukaryoticRNAinterference(RNAi)pathway,enablingthetargetinganddegradationofamRNAsharingcomplementarytothesiRNAsequence(Elbashiretal.,2001a;Zamoreetal.,2000).Beforethematurecomplexisformed,oneofthestrandsofthesiRNAduplexmustberemoved(Matrangaetal.,2005).Selectionofthedesiredstrandiscriticaltoensurepropertranscripttargeting(Schwarzetal.,2003).TheinitialprocessofstrandselectionoccursthroughtheasymmetricbindingofthesiRNAduplexbytheproteinsoftheRNAipathway,whichcanrecognizespfeaturesofthesiRNA(Franketal.,2010;Nolandetal.,2011;NolandandDoudna,2013;Gredelletal.,2010;Suzukietal.,2015).LoadingofsiRNAsintooneoftheAgoproteinswasoriginallyproposedtooccurbypositioningthesiRNAintoDicer'shelicasedomainwithintheDicer-TRBPcomplex(Wangetal.,2009;Maetal.,2008;Nolandetal.,2011),followedbythehandofthematuresiRNAtoanAgoprotein(Matrangaetal.,2005;MacRaeetal.,2008).siRNAsarethensortedtotAgoproteinsbasedontstructuralfeatures(Duecketal.,2012;Burroughsetal.,2011;CzechandHannon,2010),withthemajorityofsiRNAsendingupineitherAgo1orAgo2(Guetal.,2011).Ago2istheonlyAgowithanactivenucleasedomaincapableofcleavinganmRNAtarget(Meisteretal.,2004;Yodaetal.,2010),makingittheonlyAgoproteinthatcantlysilencewithasmallRNAthatishighlycomplementarytoitstarget,suchasansiRNA.43EvidencehasalsosuggestedthatsiRNAloadingintoAgo2canoccurwithouttheDicercomplex(Koketal.,2007;BetancurandTomari,2012;Kimetal.,2014),suggestingatmechanismofsiRNAloading.ThemechanismofsiRNAloadingremainsanareaofdebateastexperimentalmodels(i.e.,lossoffunctionmodelsusingknockoutcelllines(BetancurandTomari,2012;Kimetal.,2014)orsiRNAs(Takahashietal.,2014;KiniandWalton,2009;Chendrimadaetal.,2005)andrecombinantproteinreconstitutionmodels(Yeetal.,2011;Willkommetal.,2016;Bernardetal.,2015;Iwasakietal.,2010;Macraeetal.,2007))produceresults.Todate,C3PO,TRBP,PACT,Dicer,andHSP90haveallbeenshowntoindividuallyconstituteRISCactivitywithAgo2invitro(Yeetal.,2011;Willkommetal.,2016;Bernardetal.,2015;Iwasakietal.,2010).Whatremainsunclearishow,andunderwhatcircumstances,eachoftheseproteinsmaymodulatesiRNAfunctionwithinthecomplexcellularenvironment.OftheproteinsoftheRNAipathway,Dicer,TRBP,andAgo2havebeenshowntoasymmetricallyinteractwiththeterminiofsiRNAduplexesindicatingthattheseproteinsmaysiRNA(ormiRNA)asymmetry(Nolandetal.,2011;Gredelletal.,2010;KiniandWalton,2009;Franketal.,2010;Suzukietal.,2015).PACTisaproteinthatshares44%homologywithTRBP(Chendrimadaetal.,2005).PACT,likeTRBP,contains2dsRBDsandthedsRBD-likeMedipaldomain,throughwhich,itinteractswithDicer,inasimilarmannertoTRBP(Haaseetal.,2005;Larakietal.,2008).TRBPhasbeenshowntosensesiRNAasymmetrybypreferentiallylocalizingtothemorestablybase-pairedsiRNAtermini(Gredelletal.,2010).BasedonthesimilaritiesbetweenPACTandTRBP,weinvestigatedtheabilityofPACTtosensesiRNAasymmetrybyassessingitsabilitytotialylocalizetosiRNAtermini.Wewereabletodemonstrate44thatPACTshowstialterminuslocalization,similartoTRBP.Ingeneral,double-strandedRNAbindingproteins(dsRBPs)donotsenseinRNAduplexstructurethroughthetheirdouble-strandedRNAbindingdomains.WehavehypothesizedthatPACT,similartoTRBP,iscapableofsensinginRNAstruc-tureallowingittotiallylocalizetomorestablybasepairedsiRNAtermini.Ourmethodfordeterminingterminuslocalizationiswitha4-thiouracilphotoreactivecross-linkerincorporatedatthe20thpositionofthesiRNA(Figure3.1A).Giventhatthepro-posedmethodofRNArecognitionisnon-canonical,weinvestigatedsiRNAterminuslo-calizationwithanothergenericdsRBP,NS3,tothatcross-linkingpatternswerearesultofspcprotein-RNAinteractionsandnotaresultofcross-linker.ThecriteriafortheselectionofanegativecontroldsRBPincludedtheabilitytobindtobothdsRNAandssRNA.NS3,aviralsuppressorofRNAifromtheRicestripevirus,spbindstosiRNAduplexeswithhighyandisalsoknowntobindtossRNA,although,withlowery(Shenetal.,2010).Ourresultsdemonstratetheofthephotoreactive4-thiouracilcross-linkerbyshowingthatallofthesiRNAsinthisstudycross-linkedwithequaltotheRicestripevirusNS3protein.45Figure3.1:ofsiRNAsforCross-linkingExperiments.(A)Place-mentofphotoreactivecross-linkers(4-thiouracil)andradiolabels(32P)todeterminedsRBPterminuslocalizationaandb).cshouldbetheaverageofthecross-linkingobservedforaandb.(B)siRNAsusedtodeterminedsRBPterminallocalization.pp-lucandsod1siRNAsarehighlyfunctionallyasymmetricandsymmetric,respectively.ThefunctionalasymmetryofEGFP#274isunknown.3.3Results3.3.1FunctionalCharacterizationofRecombinantlyExpresseddsRBPsWeexpressedHis-taggedNS3fromtheRicestripevirusinRosetta2(DE3)cells,ittohomogeneity(assessedbySDS-PAGEgel,Figure3.2A),andvimmunoreactivitywithananti-6xHisantibody(Figure3.2B).TheapparentsizeofNS3bybothSDS-PAGE(Figure3.2A)andwesternblotting(Figure3.2B)wasnearthetheoreticalsizeof24.6kDa.FunctionalityofNS3wasbyassessingitsabilitytobindssRNAandsiRNA(Figures3.2Cand3.2D)viaelectrophoreticmobilityshiftassay(EMSA)withincreasingquantitiesofNS3.NS3didbindtosiRNAswithhighy(lownanomolarrange),46howeverweobservedasaturationinthequantityofcomplexformedwith˘60%bindingoflabeledsiRNA(Figure3.2D).NS3-siRNAcomplexformationsaturatedwithincreasingproteinconcentrationsuggestingthateither,complexesaredisturbedduringelectrophoresisorthatthesiRNAisinaconformationthatdiscouragesNS3binding.Ineithercase,thequantityofNS3-siRNAcomplexesformed,providedtsignaltoperformcross-linkingexperiments.NS3-ssRNAcomplexeswerenotwellresolvedbyEMSAwith<1%ofssRNAformingcomplexeswith1MofNS3(Figure3.2C).47Figure3.2:FunctionalCharacterizationofNS3.(A)SDS-PAGEanalysisofrecom-binantHis-NS3stainedwithGelCodeBluetotalproteinstain,theoreticalsize:24.57kDa.(B)WesternblotofrecombinantNS3withanti-6xHisantibody.(C)EMSAofNS3bindingto50-32Plabeledpp-lucASstrandresolvedwithnativePAGE.(D)EMSAofNS3with50-32Plabeledpp-lucsiRNAresolvedwithnativePAGE.TostudytheasymmetricbindingpreferencesofPACTwithsiRNAs,werecombinantlyexpressedPACTwiththemaltosebindingprotein(MBP)expressiontaginE:ColiBL21(DE3)-RIPLcells.TheMBP-tagwasusedtoincreasethesolubilityofPACTandhas48beenfoundnottointeractwithsiRNAsininvitrobindingassays(Gredelletal.,2010).Recombinantproteinwasto˘95%purity(assessedbySDS-PAGEgel,Figure3.3A).Proteinidentitywaswithananti-MBPantibodyandhadanobservedsizenearthetheoreticalproteinsizeof79.8kDa(Figure3.3B).FunctionalityofMBP-PACTwasbyverifyingitsabilitytobindssRNAandsiRNAviaEMSA(Figure3.3Cand3.3D,respectively).MBP-PACTyforsiRNAswascharacterizedwithincreasingproteinconcentrationsandalimitingsiRNAconcentration(˘1nMsiRNA)(Figure3.3D).siRNAbindingsaturatedat˘50%with1MMBP-PACT(Figure3.3D).AswithNS3,wesuspectthatlackofacompleteshiftoflabeledsiRNAdevelopsfromcomplexinstabilityduringelectrophoresisorthatsiRNAisinaconformationthatprecludesbindingbyPACT.siRNAcross-linkingsignalswithPACTwerestrongenoughtotiatesiRNAterminuslocalization.49Figure3.3:FunctionalCharacterizationofMBP-PACT.(A)SDS-PAGEanalysisofrecombinantMBP-PACTstainedwithGelCodeBluetotalproteinstain,theoreticalsize:79.8kDa.(B)WesternblotofrecombinantMBP-PACTwithanti-MBPantibody.(C)EMSAofMBP-PACTbindingto50-32Plabeledpp-lucASstrandresolvedwithnativePAGE.(D)EMSAofMBP-PACTwith50-32Plabeledpp-lucsiRNAresolvedwithnativePAGE.503.3.2NS3DoesNotAsymmetricBindofsiRNAsTocontrolforthepossibilitythatthemeasuredasymmetricbindingofsiRNAsbydsRBPsresultsfromabiasinthecross-linkingciencyofthesiRNAstrands,siRNAbindingandcross-linkingexperimentswereperformedwithNS3.Reactionswereperformedwith50nMofprotein,atwhichconcentrationapproximately50%ofthesiRNAiscomplexedwithNS3(Figures3.4Aand3.4B).Thefractionofcross-linkedNS3proteinwithsiRNA(a)and(b),wasnottlytfortheasymmetric(p=0.61),symmetric(p=0.70),orUnknown(p=0.72)siRNAs.Thesecross-linkingpatternsaretfromthosepublishedwiththeTRBPprotein,whichshowedatinthecross-linkingpatternbetweenthe(a)and(b)cfortheasymmetricandunknownsiRNAs(p<1x104)(Gredelletal.,2010).51Figure3.4:SymmetricalCross-linkingofNS3tosiRNA.(A)Nativegelshiftanal-ysisand(B)quanofNS3bindingtosiRNAswitheachcross-linkerinFigure3.1A.(C)Denaturinggelelectrophoresisofcross-linkedsiRNAswithNS3and(D)QuanoftheNS3siRNAcross-linking.Fractioncross-linkedaremean1SD;N=3.ComparisonofaandbforeachsiRNAbyone-wayANOVAwithTukey'sposthocanalysisshowednosigtdiNS3bindingtossRNAwasrelativelylowfortheconcentrationsofproteintested(Figure3.5A),howeverwewereabletodemonstrate>10%cross-linkingtoallsiRNAstrandswith1MNS3(Figure3.5C).SimilardisruptionofTRBP-ssRNAcomplexesduringnativePAGEwasobservedunderidenticalconditions(Gredelletal.,2010).Nativebindingof52NS3tossRNAvariedwiththetRNAsequences;however,thequantityofssRNAthatcross-linkedtoNS3wasnottlytbetweentheASandSSforthesequencestested(asymmetricp=0.31,symmetricp=0.28,andEGFP#274p=0.087,two-tailedt-test).Furthermore,cross-linkingwassimilaramongallstrands,exceptfortheunknownSSsequencewhichwascross-linkedtoNS3modestlylowerthantheotherssRNAs(Figure3.5Cand3.5D).ThessRNAcross-linkingpatternswithNS3varysubstantiallyfromthecross-linkingpatternsobservedwithMBP-TRBP(Figure3.5D),whichhadtincross-linkingbetweenmultiplessRNA(Gredelletal.,2010).CollectivelyboththessRNAandsiRNAcross-linkingresultssupportthehypothesisthatcross-linkingissimilarforthesequencestested,andthatanysincross-linkingareduetosplocalizationoftheproteinonthesiRNAstructure.53Figure3.5:SymmetricCross-linkingofNS3tossRNA.(A)Nativegelshiftanalysisand(B)quanofHis-NS3bindingtoeachssRNAinFigure3.1B.(C)Denaturinggelelectrophoresisofcross-linkedsiRNAswithHis-NS3and(D)QuanoftheHis-NS3siRNAcross-linkingtossRNA,eachwitha50-32Plabeland4-thiouracilphotoreactivecross-linkersubsititutedatthe20thnucleotide.Fractioncross-linkedaremean1SD;N=3.infractioncross-linkedforeachssRNAwerebytestedbyone-wayANOVAwithTukey'sposthocanalysis.Overheadbarsrepresentvalueswhicharesitlyt(p<0.05).543.3.3PACTAsymmetricallyLocalizestosiRNATerminiBasedonthehighdegreeofhomologybetweenTRBPandPACT(Chendrimadaetal.,2005),wehypothesizedthatPACTwouldpossesssimilarasymmetricterminallocalizationpreferencesdespiteitsloweryforsiRNAs.MBP-PACTbindingandcross-linkingexperimentswereperformedataconcentrationof3M,wherefractionalsiRNAbindingwas40%(Figure3.6Aand3.6B).MBP-PACTcross-linkedpreferentiallytothemorestableterminusofthefunctionallyasymmetricsiRNApp-luc(p=0.065,two-tailedt-test),andcross-linkedequallytobothterminiofthefunctionallyasymmetricsiRNAs(p=0.66,two-tailedt-test)andsiRNAwithunknownfunctionalasymmetry(EGFP#274;p=0.53,two-tailedt-test)(Figures3.6Cand3.6D,Table3.1).MBP-PACTcross-linkingpatternsweresimilartoTRBPandR2D2,theDrosophilahomologofTRBP,forthefunctionallyasymmetricandsymmetricsiRNAs(Table3.1).However,thesymmetriccross-linkingpatternobservedforEGFP#274withMBP-PACTfromtheasymmetriccross-linkingpatternpreviouslyobservedwithMBP-TRBP(Gredelletal.,2010).55Figure3.6:AsymmetricalBindingofMBP-PACTtosiRNA.(A)Nativegelshiftanalysisand(B)quanofMBP-PACTbindingtosiRNAswitheachcross-linkerinFigure3.1A.(C)Denaturinggelelectrophoresisofcross-linkedsiRNAswithMBP-PACTand(D)QuanoftheHis-NS3siRNAcross-linking.Fractioncross-linkedaremean1SD;N=3.ComparisonofaandbforeachsiRNAbyone-wayANOVAwithTukey'sposthoc,(*)representsap=0.065,noothercomparisonsofaandbarestatisticallyt.56Table3.1:PACTAsymmetricsiRNABindingFromTRBP.Asymmetriccross-linkingpatternsfordsRBPs,R2D2,DrosophilahomologofTRBP,(Tomarietal.,2004),TRBP(Gredelletal.,2010),andPACT.(+)Representshighercross-linkingwiththesiRNAantisensestrand,neutralmeansthatcross-linkingbetweenthesiRNAterminiwasthesame,andNDisnotdetermined.3.3.4PACTDimerizeswithssRNAandsiRNATRBP-siRNAbindingreactionsproducedtwocomplexesonanativegel(Figure3.7A,lane2).Thecomplexhasa1:1stoichiometricratioofsiRNA:MBP-TRBPandthesecond,a1:2stoichiometricratio(Gredelletal.,2010).Interestingly,PACTonlyformsonecomplexwithbothsiRNAandssRNA,andrunsatasimilarsizetothedimericMBP-TRBP(Figure3.7A,lanes3and4).Thecomplexformeddoesnotchangewithincreasingproteinconcentration(Figures3.3Cand3.3D),whichisinagreementwiththegsthatPACTnativelyexistsasadimer(Takahashietal.,2013).57Figure3.7:tBindingModesofTRBPandPACT.(A)EMSAofMBP-TRBP(Lanes1and2)andMBP-PACT(Lanes3and4)with50-32Plabeledpp-lucASssRNA(Lanes1and3)and50-32Plabeledpp-lucsiRNA(Lanes2and4).Proteinconcen-trationsforeachbindingreactionareasfollows,Lane1:1250nMMBP-TRBP,Lane2:350nMMBP-TRBP,Lane3:3000nM,Lane4:10000nM.3.4DiscussionSmallRNAasymmetrydevelopsfromthebindingpreferencesoftheRNAiproteinsforspe-featuresofthesmallRNAs.ThelocalizationororientationoftheRNAiproteinswheninteractingwithsmallRNAscanthendrivesmallRNAstrandselection.ThecumulativeofasymmetricinteractionsleadstotheasymmetricloadingofonesmallRNAstrandandexclusionoftheother.WewereabletodemonstratethatPACTtiallylocalizestosiRNAterminiwithsomesiRNAsbutnotothers,similartobutstilluniquefromthefunctionofTRBP(Figure3.6)(Gredelletal.,2010).Furthermore,wefoundthatNS3,aproteinwithnoknownpreferencesforspdsRNAstructuresornucleotides,didnot58asymmetricallycross-linktosiRNAtermini.Thus,incross-linkingbetweenthetterminiofthesiRNAwithPACTandTRBPareafunctionoftheprotein-RNAinteractionandarenotaresultofincross-linking.TheasymmetriclocalizationofbothTRBPandPACTwithfunctionallyasymmetricduplexesindicatesthattheseproteinsarecapableofdirectingtheasymmetricbindingofsiRNAduplexesandsiRNAasymmetricstrandloading.Whilenotexploredhere,itispossiblethattheinoligomericstatebetweenTRBPandPACT,im-pactedhoweachproteininteractswithsiRNAtermini(Figure3.7).ThefullrolesofTRBPandPACTinRNAiarenotunderstoodanditispossiblethateitherTRBPandPACTsiRNAloadingthroughtheirinvolvementwiththeDicercomplex(Leeetal.,2013),throughdirectinteractionwithAgo2(Willkommetal.,2016),ornotatall(Kimetal.,2014;Wilsonetal.,2015).Nonetheless,ourresultselucidateinTRBPandPACTbindingpreferences,demonstratethepossibilityofPACTtoactasansiRNAasymmetrysensor,andinformshowPACTmayinteractwithotherdsRNAstructures.TwomodelshavebeenproposedtopredictthelocalizationoftheR2D2andTRBPtosiRNAtermini,4nn(Tomarietal.,2004)andTNRank(Gredelletal.,2010).TRBP,R2D2,aDrosophilaofTRBP,andPACTbindingpatternswereallasymmetricwithpp-lucandsymmetricwithsod1siRNAs(Table3.1);consistentwithbothmodelsforsiRNAterminuslocalizationandtheknownfunctionalasymmetryofthesesiRNAs(Table3.2).TheasymmetricbindingpatternswerenotsimilarbetweenTRBPandPACTwithEGFP#274(Table3.1),norwerethepredictedbindingpatternconsistentbetweentheTNRankand4nnmodels(Table3.2).The4nnofEGFP#274is-0.1kcal/mol,arelativelysmallinhybridizationstability.Ifthe4nnofEGFP#274canbe59interpretedassymmetrical,thenthe4nnmodelpredictedtheasymmetriclocalizationofPACTforallthreesiRNAs,making4nnabettermodelfortheasymmetricbindingpatternofPACT,comparedtoTNRank.However,thevalidationofthesemodelsisverylimitedandmanymoresequencesareneededbeforethesemodelscouldbeusedwithanyutilityinpredictingterminuslocalizationofTRBPorPACTwithsiRNAs(SeeChapter4,FutureDirections:TRBPandPACT).Table3.2:PredicitedAsymmetricLocalizationofR2D2,TRBP,andPACT.PredictionofasymmetricbindingpatternontheAsymmetric,Symmetric,andUnknownsiRNAsusingtheTNRankand4nnmodels.(+)Representshighercross-linkingwiththesiRNAantisensestrand,neutralmeansthatcross-linkingbetweenthesiRNAterminiwasthesame,andNDisnotdetermined.3.5ConclusionsHerewedemonstratedthattheproteinPACT,anRNAipathwayproteinhypothesizedtoaidintheloadingofsmallRNAs,iscapableofasymmetricallylocalizingtosiRNAtermini.PACTbindstosiRNAsbutatamuchlowerythanitshomolog,TRBP.PACTalsodfromTRBPinitsasymmetricalbindingpatternwithoneofthesiRNAstested,andunlikeTRBP,alignedwellwitha4nnmodeltopredictterminilocalization.TheseresultsindicatethatPACTiscapableofsensingcharacteristicsofthesiRNAduplex,suggestingthatPACTcanserveasanasymmetrysensorwithintheRNAipathway.60Chapter4ConclusionsandFutureDirections4.1ConclusionsThepurposeofthesestudieswastobetterunderstandthesiRNAcharacteristicsandprotein-RNAinteractionsresponsibleformakingonesiRNAstrandofaduplexmoreactivethantheother.TheresultsfromthisworkprovidebothdetailsoftheRNAimechanismandinformssiRNAdesign.InordertobetterunderstandtherelativeimportanceoftwosiRNAduplexfeaturesthatarepredictiveofsiRNAactivity,TNRankand3nn,weinvestigatedhoweachcharacteristicpredictssiRNAstrandloadingandactivitywithsetofsiRNAsthatsam-pledbothfeatures.WefoundthatTNRankalonewasabletopredictsiRNAfunctionalasymmetry,whereas3nnpredictssiRNAfunctionalasymmetryonlywhenappliedasasecondordertoTNRank.Asimilartrendwasobservedwithasymmetricstrandloading,butthepredictiveabilityof3nnwasmuchweakercomparedtofunctionalasymmetry.Uponfurtherinvestigation,wefoundthatindividualsiRNAstrandloadingwasbetterpredictedbythe50terminalduplexenergies(2)1ntand12nt).TNRankwaspredictiveofsiRNAactivitybecauseofitscollinearitywiththe50terminaldu-plexenergies.Post-siRNAloading,lowhybridizationstabilityintheprimaryseedregionwasfoundtocorrelatewithgreaterRISCspactivity,possiblyindicatingthataweak61targetinteractiontoencouragesRISCturnover.Greaterhybridizationstabilitynearthe30termini1718nt)wasalsofoundtocorrelatewithRISCspactivity.Becausethisregionoftheguidestrandhasbeenshown,withrecombinantmouseAgo2,tobepartofthe30supplement(Weeetal.,2013),aregionoftheguidestrandnotinvolvedinRISC-targetinteractions,ourresultssuggestthatthecellularenvironmentmayaltertheRISCinteractionsatthe30endofthesiRNA.Theseresultsourunderstandingofhowthehighlevelparameters,TNRankandimpactsiRNAfunctionattstagesintheRNAipathway.ThemolecularmechanismsunderlyingsiRNAasymmetricstrandselectionarenotfullyknown.HerewecharacterizedPACT,ahomologtotheprotein,TRBP,anddemonstratedthatPACT,likeTRBP,asymmetricallyinteractswithsiRNAduplexesalthoughnotinanidenticalmannertoTRBP.WeproposethattheinduplexlocalizationisbecausePACTnativelyexistsasadimer,whereasTRBPexistsnativelyasamonomer,alteringthewayinwhichtheseproteinsinteractwithsiRNAsbutstillretainingstructuralpreferencesformorestablybasepairedtermini.Asymmetriclocalizationwasshownusingaphotoreactive4-thiouracilcross-linkersubstitutedinthe30overhangofsiRNAduplex;themechanismofcross-linkingwasrigorouslyvalidatedtocunbiasedcross-linking.TheseresultsprovidegeneralinsightintohowtheseproteinstiallyinteractwithRNAsequences.4.2FutureDirectionsAswebegintounderstandmoreabouttheRNAimechanism,siRNAactivitypredictionshaveimproved.Outlinedbelowarefuturedirectionsoftheworkpresentedhere.624.2.1ParsingwithLargerDatasetsOuranalysisinChapter2investigatedhowtheTNandparameterssiRNAstrandselectionandactivityusing2-factorlinearregressionmodeling.A2-factormodelwasnecessarytocontrolfortheofTN.WewerelimitedinthenumberofvariablesthatwecouldsimultaneouslyanalyzebythesizeofourdatasetandthecoverageoftsiRNAfeatures.AsaresultwewereunabletolookatinteractionsbetweenTNand(x)(y)nt).Inparticular,wewerecuriousiftheTNcouldtheofacorrelationbetweenandactivityatagivenlocationinthesiRNAduplex.Weperformedapreliminaryanalysisusingadatasetcontainingtheactivitydataof2431siRNAs(Hueskenetal.,2005),hereafterreferredtoastheNovartisdataset.Webeganouranalysisutilizingthesame1-factor(TNor(x)(y)nt))and2-factor(TNor(x)(y)nt))linearregressionmodels(Figure4.1A)usedinChapter2tocompareourresultstothoseoftheNovartisdataset.Becauseofthesizeofthedataset,manyofthewithinthesiRNAduplexwerefoundtobetinboththe1-factorand2-factormodels(Figure4.1A).ToprovideacomparisonbetweenourdatasetandtheNovartisdataset,wedecidedtoonlyevaluatethetoptwoparameters,becausewewereonlycapableofidentifyingtwoparametersinouractivitydataset.Thetoptwomosttparametersidenusing1-factormodel(TNand12nt)matchedtheparametersideninthe1-factoractivitymodelfromourdataset(Figure2.9).However,theresultsfromtheNovartis2-factormodelvariedfromour2-factormodel(Figure2.9),withtheNovartis2-factormodelretainingastrongcorrelationbetweensiRNAactivityand50terminusduplexenergies(2)(1)ntand12nt).Thesediscrepancieslikelydevelopfromthedesignofeachdataset,withourdatasethavinganunequalrepresentationof50TNs,potentiallybiasingtheimportance63ofaspfeatures,andtheNovartisdatasetusingasinglehighconcentrationofsiRNA,limitingthedynamicrangeofsiRNAactivities.Figure4.1:TNcanAltertheRoleofGinPredictingsiRNAActivity.(A)1-factorand2-factorcorrelationsofsiRNAactivitydatafromtheNovartisdatasetwithTNand(x)(y)ntalongthelengthoftheduplex.2-factorcorrelationsuseTNasonevariableanda(x)(y)ntvalueastheother.inthesecorrelationsindicatesthatthevariablepredictsactivityindependentofTN.ABonferronicorrectionwasappliedtocorrectformultiplecomparisons.TolookforinteractionsbetweenTNand(x)(y)nt,weparsedtheNovartisdatasetbyTN(Figure4.1B).AfterparsingthedatabyTN,weagainlookedforcorrelationsbetweenactivityand(x)(y)nt,usinga1-factorlinearregressionmodel.Weobservedapreferenceforlowhybridizationstabilityatthe50terminus12and23)forallTN.64The(2)1parameterwasnotuniqueafterparsingthedatabyTNandcouldnotbeincorporatedintothelinearmodel.Wealsoobservedapreferenceforgreaterhybridizationstabilityatthe30terminusofsiRNAswith50adenine,cytidine,andguanineTN,butnoturidinenucleotides.ThecorrelationforsiRNAsequenceswith50adenineTNwasmuchweakerthanthecorrelationsatthe30terminiforsiRNAsequenceswitha50cytidineorguanineTN.Theseresultssuggestantialdependenceon30terminalduplexenergies.Apreferenceforweakerhybridizationstabilityatnucleotides5-6and6-7wasalsoobservedforsiRNAsequenceswithuridineandadenineTN.ThetintheNovartisdatasetdonotperfectlyalignwiththeresultspresentedinChapter2,however,becauseofindatasetdesignitistomakedirectcomparisons.Theseresultsdo,however,indicatethattheTNcanalterhowduplexstabilityesiRNAactivity.GreaterconsiderationshouldbegiventounderstandingtheinteractionsamongsiRNAactivitypredictors.Futureworkinthisareashouldexploretheconditionalnatureofthe30duplexenergieswithtTN,lookingatwhythischaracteristicisimportant.4.2.2TRBPandPACTBoththeworkpresentedinChapter3andpreviousworksuggestthatTRBPandPACThavepreferencesforcertainfeaturesofthesiRNAduplex(Gredelletal.,2010),allowingthemtotiallylocalizetosiRNAtermini.Futureworkcanbesplitintotwoareas,i)understandingtheRNAstructuralpreferencesofTRBPandPACTandii)investigatingTRBPandPACTfunctioninsiRNAloadingusingantissueculturemodel.Inourpriorwork,onlythreesequenceswereusedtoevaluatetheterminuslocalizationofTRBPandPACT.TounderstandhowTRBPandPACTcandrivesiRNAasymmetric65strandloading,manymoresequenceswillneedtobeevaluated.ThecurrentexperimentaldesignisnotamenabletoscreenmorethanafewsiRNAsequences,soinordertomapRNAstructuralpreferencesofTRBPandPACTamorerobustexperimentalstrategywillberequired.SomehavebeenmadetomapthestructuralpreferencesofTRBPthroughDNAsubstitutions,whichperturbTRBPbinding(Takahashietal.,2014;Gredelletal.,2010),andthroughtheintroductionofmismatcheswithinduplexstructures(Acevedoetal.,2015).ThesestudieshavefoundthatTRBPbindingisperturbedbyinternalbulgesandmismatches(Acevedoetal.,2015),andthatTRBPmoststronglyassociateswithsiRNAtermini(Takahashietal.,2014;Koketal.,2011).Whilehelpful,thesestudieswerenotabletolookatatlycomplexpoolofsequencestofullymapTRBPorPACTinteractions.TRBPandPACTstructuralpreferencescouldbemappedwithaselectionexperimentusingapoolofuniqueRNAsequencesandevaluatingtheenrichmentofRNAsequences.Deep-sequencingallowstheparallelquanionofRNAsequencesandwouldbeideallysuitedfortheapplication.TheminimumlengthduplexthatTRBPcanbindtois˘15nt(Kohetal.,2013);usingacomplexpoolof15ntRNAfragmentswouldlimitthenumberofpossiblemotifsTRBPcouldrecognizeinanindividualsequence.TheminimumlengthduplexthatPACTcanbindtoisnotknownandwouldneedtobedeterminedbeforeexperimentation.FurthercharacterizationoftheindividualdomainsofTRBPandPACTwouldhelptounderstandhowTRBPandPACTrecognizetduplexstructuresandcouldalsobeperformedusingasimilarenrichmentassay.Tothesecondpoint,manystudiesofTRBPandPACT,includingthoseperformed66inChapter3,wereperformedinvitro.NoexperimenthaselydemonstratedthefunctionofTRBPorPACTinsiRNAloadingincells.Likewise,theonlyevidenceindicatingtheimportanceofTRBPinsiRNAmediatedsilencingistheobservationofadecreaseinsilencingwhenTRBPwasknockeddownwithansiRNA(Haaseetal.,2005;Chendrimadaetal.,2005;Takahashietal.,2014).MorerigorousexperimentslookingatthefunctionalaspectsofTRBPandPACTwillberequiredtounderstandtheroleoftheseproteinsinRNAi.TheproposedrolesofTRBPandPACTasproteinsinvolvedintheloadingofsiRNAsimpliesthattheirinteractionswithsiRNAsaretransient,therefore,thequantitiesofsiRNAinteractingwitheitheroftheseproteinsatasinglepointintimearelowandhardtodetect.Asaresult,functionalassaysarerequiredtoinvestigateTRBPandPACTinlivecells.AHeLacelllinewithbothTRBPandPACTknockedouthasrecentlybeendeveloped(Kimetal.,2014).Usingthiscellline,siRNAfunctionalasymmetryandsiRNAloadingcouldbeinvestigatedwithandwithoutrescuingofTRBPorPACTproteinexpressionusingthemethodsdevelopedinChapter2.Further,theroleofthe50aswasshowninChapter2topredictsiRNAstrandloading,canbeinvestigated.67APPENDICES68AppendixAMaterialsandMethodsforCh.2CellCultureandTransfectionHeLacellsweremaintainedinDMEMHighGlucose(LifeTechnologiescat#11965-092)supplementedwith10%FBS(LifeTechnologiescat#16000-044)and1%Pen-Strep(LifeTechnologiescat#15240062)andincubatedat37°Cand5%CO2inahincuba-tor.Beforetransfection,cellswereplatedat15,000cells/wellin96-wellplatesor350,000cells/wellin6-wellplatesinmediawithoutPen-Strepfor24h.ForwardtransfectionswereperformedwithLipofectamine2000(LifeTechnologiescat#11668019)atacon-centrationof2.33g/mL,basedonwellvolume.LipoplexeswerepreparedperthemanufacturersinstructionsinOpti-MEM(LifeTechnologiescat#31985-070);siRNAandplasmidconcentrationsarespforeachexperiment.NucleicAcidsProteinKinaseR(PKR)-targetingsiRNAswereorderedfromDharmaconasduplexes,de-signedwitha19bpantisensestrandtargetingthePKRgeneand30uridinedinucleotideoverhangs.siRNAsequencesarelistedinTableA.1.siRNAnearest-neighborenergypa-rameterswereobtainedfromtheDINAMeltWebServer(MarkhamandZuker,2005,2008)69andusedtocalculaterelativeterminalhybridizationstabilities(Figure1.5A).AllDNAprimersforthisstudywereorderedfromIntegratedDNATechnologies.TableA.1:PKR-targetingsiRNASequences.siRNAsequencestargetingthePKRgenewiththeircorrespondingTNRankand3nn.(+)siRNAstrandisinblueand(-)siRNAstrandisinred.AlgorithmrankisthepredictedrelativeactivitycomparedtoallsiRNAsequencestargetingthePKRgenewhentakingintoconsiderationtheirTNRankand3nn(Malefytetal.,2013).Somealgorithmrankingsand3nnvaluesfromourpriorworkduetoupdatednearest-neighborparametersandtstothealgorithm.ThelastsequenceisthesiRNAusedasaninternalstandardforRT-qPCRexperiments.Forluciferaseexperiments,thePKRgenewasPCRfrompET28a-PKR(Bevilac-quaandCech,1996)(kindlyprovidedbyDr.PhilipC.Bevilacqua)andsubclonedintothepsiCHECK2-AS34avector(Navarroetal.,2009)(Addgene#37099)intheforward70(psiCHECK2-PKR+)orreverse(psiCHECK2-PKR-)directionusingClontechIn-fusioncloning(Figure2.2).PlasmidswerevbySangersequencing.PrimersusedforcloningandsequencingarelistinTableA.2.TableA.2:CloningandSequencingOligos.ThefollowingprimerswereusedforthecloningofthePKRgeneintothepsiCHECK2vector,inboththeforwardandreversedirections,andforsequencingvPrimersforstem-loopRT-qPCRweredesignedandpreparedaccordingtopublishedsp(Varkonyi-GasicandHellens,2011),withaconstanthairpinregionanda6ntoverlapwiththetarget.AllsequencesusedarelistedinSupplementaryTableA.3.qPCRwasperformedwithasequence-spforwardprimerandauniversalreverseprimer.AllprimerswerevbyspikinginknownquantitiesofsiRNAandverifyingam.StandardcurvesweregeneratedforeachsiRNAsequencetoaccountforvariationin.71TableA.3:Stem-loopPrimerSequences.PrimerdesignfollowstheprotocoloutlinedbyChenC.etal.(Chenetal.,2005),withastandardstem-loopsequenceanda30tailwitha6ntoverlapsptothe30endofeachsiRNAstrand.qPCRwasperformedwithauniversalreverseprimerandasiRNAstrandspeforwardprimer.G/Cnucletotideswereaddedtothe50endoftheforwardprimertoadjustTmto˘60°C.72Dual-LuciferaseAssayHeLacellswereseededina96-wellplateat15,000cells/wellin100Lofmedia24hbeforetransfection.After24h,themediawaschanged,andboththePKRandnon-targetingsiRNAsweredilutedinOpti-MEMandco-transfectedwith40ng/wellofeitherthe(+)or(-)psiCHECK2-AS34aplasmid,using0.35L/wellofLipofectamine2000foratotaltransfectionvolumeof50L.SerialdilutionsofPKRsiRNAsrangingfrom3.2-10,000pMweretransfectedinduplicate,andtransfectionswereperformedtwiceforeachsiRNA.ThePKRsiRNAwasdilutedinanon-targetingsiRNAtomaintaintotalsiRNAconcentrationat10nMperwell.24hpost-transfection,themediawasaspirated,replacedwith79LofDulbecco'sPBS(Invitrogencat#14040133),andlysedusing79LofDual-GloLuciferaseReagent(Promegacat#E2940).Afterincubatingatroomtemperaturefor15minonarocker,150Lofthesolutionwastransferredtoasolidwhite96-wellplate,andactivitywasmeasuredusingaSynergyH4microplatereader(Biotek).75LofDual-GloStop&GloReagent(Promegacat#E2940)wasadded,and,aftera10minincubationperiod,Renillaactivitywasmeasured.RelativesiRNAactivitywasdeterminedbysubtractingbackground,dividingRenillasignalbytheluciferasesignalwithinawell,andthennormalizingtheratiototheratiofromawelltreatedwithanon-targetingsiRNA.Valuesof0and1indicatecompletesilencingandnosilencing,respectively.IC50valuesweredeterminedbydatatothefollowingequation,Y=11+10XLog(IC50)73usingleast-squaresregression,whereXistheconcentrationoftargetingsiRNAandYistherelativesiRNAactivity.Stem-loopRT-qPCRHeLacellswereplatedina6-wellplateandtransfectedafter24hwith10nMPKRsiRNAand10nMofanon-targetingsiRNA(internalstandard).siRNA-Ago2co-immunoprecipitationwassimilartotheproceduredescribed(BeitzingerandMeister,2011),withseveralexcep-tions.Immunoprecipitationswerescaleddownto400gofcelllysate;celllysateproteinconcentrationsweredeterminedviaBradfordassay(Bio-Radcat#5000201)withaBSAstandard.ImmunoprecipitationofAgo2wasperformedwith3.375gofAgo2Antibody(Sigma-Aldrich,Clone11A9cat#SAB4200085-200UL)pre-boundto25LofProteinGmagneticbeads(NEBcat#S1430S).ProteinaseKDigestionerwasalsosupplementedwith˘400g/mLtRNA(Rochecat#10109541001),toactasacarrier,andAgo2-boundRNAsusingDirect-zolRNAkitaccordingtothemanufacturer'spro-tocol(ZymoResearchcat#R2051).TheprotocolforRT-qPCRquanwasdevelopedbasedonpublishedmethods(Varkonyi-GasicandHellens,2011;Chenetal.,2005;Kramer,2001).Stem-loopprimerswerefoldedbyheatingto95°Cfor10min,rampingto75°Cover10min,thenholdingat75°C,68°C,65°C,62°C,and60°Cfor30mineach,beforerampingto4°Cover2h.RNAsampleswereheatedto65°Cfor5minandsnapcooledonicefor5minimmediatelybeforereversetranscription.Pulsedreversetranscriptionreactionswereassembledwith1xFirst-Strand10mMDTT,0.25mMdNTPmix,20USuperscriptIII(LifeTechnologiescat#18080044),0.2UofSUPERaseIn(Ambioncat#AM2696),and1nM74stem-loopprimerin20Lwith1LofRNAandcycledasfollows:16°Cfor30minfollowedby60cyclesof30°Cfor30sec,42°Cfor30sec,and50°Cfor1sec,reversetranscriptasewasthenheatinactivatedat85°Cfor5min.qPCRswereassembledwith300nMofansiRNAspforwardprimer,300nMuniversalreverseprimer,1xIQSYBRGreenSupermix(Bio-Radcat#170882),and1.8LofcDNAina25Lreaction.qPCRcyclingwasperformedasfollows:95°Cfor10minfollowedby40cyclesof95°Cfor10sand60°Cfor10sonaMyiQThermocycler(Bio-Rad).(+)siRNAstrand,(-)siRNAstrand,andinternalstandardsiRNAstrandwerereversetranscribedandquantiseparately(TableA.3).spywasvbyperformingstem-loopRT-qPCRonsamplesthatweretransfectedwithansiRNAthatwasnotcomplementarytothestem-loopprimers.ControlPCRswereperformedonallsamplesfollowingreversetranscriptionintheabsenceofreversetranscriptase.NothatinterferedwithquanwasobservedinanyPCRcontrol.StatisticalMethodsLinearregressionanalysiswasperformedinpythonwiththepandasandstatsmodelspack-ages,usinganordinaryleastsquaresregression.,1-and2-factormodelswerebuiltwitheither,siRNAIC50valuesorsiRNAloadingintoAgo2astheresponsevariable,andcorrelatedwithTNRankand3nn(double-strandedanalyses)orwithTNand(single-strandedanalyses).pvaluesfor1-factormodelswereusedtotestiftheslopewasnon-zero.2-factormodelswerebuiltusingeitheri)TNRankwith3nnorii)TNandateachpositionalongtheduplexusingpvaluestotestiftheslopeofthe3nnorparameterwasnon-zero.ABonferronicorrectionwasappliedtopvaluestocorrectfor75multiplecomparisons.AllotherstatisticalanalyseswereperformedusingGraphpadPrism6,withtheexceptionofwhichwascalculatedusingRStudiowiththeAICcmo-davgpackage.andAkaikeweightswereusedtocomparelinearregressionmodels.Akaikeweightsproviderelativeprobabilitiesthatonemodelbetterthedata.andAkaikeweightscontainacorrectionformodelswithtnumbersofvariables,al-lowingdirectcomparisonamongmodels.WhenanadditionalvariableisaddedtoamodelandtheAkaikeweightfavorsthenewmodel,thatvariableisprovidingnew,explanatoryinformation.76AppendixBMaterialsandMethodsforCh.3RecombinantExpressionofMBP-PACTCloningThegeneencodingthePACTproteinwassubclonedintothepMBP-6His-TEV-MMS2vec-tor,agiftfromCynthiaWolberger(Addgene#25465),fromthepET15b-PACTplasmid(KindlyprovidedbyDr.GanesSen),usingIn-fusioncloning(Clontechcat#639684).,thePACTgenewasPCRusingAdvantageHDPolymerasemix(Clon-techcat#639241)accordingtothemanufacture'sprotocol,withaTmof57°Candanelongationtimeof1min.ThefollowingprimerswereusedforPCR:50-GTACTTCCAGGGATCCATGTCCCAGAGCAGGCACand50-GGCCAGTGCCAAGCTTTTACTTTCTTTCTGCTATTATCTTTAAATACTG.ThepMBP-6His-TEV-MMS2plasmidwassequentiallydigestedwithHindIIIandthenBamHI(NEBcat#R0104SandR0136S,respectively).HindIIIwasheatinactivatedbeforeBamHIdigestion.BoththeplasmidandPCRfragmentwerecleanedupusingaPCRCleanupkit(Qiagencat#28104),recombinedbyIn-fusioncloning,andthentransformedintoTop10cells(Invitrogencat#C404003).CloneswerescreenedbycolonyPCRandvbySangersequencingwiththefollowingprimers:50-AAGACGCGCAGACTAATTCand50-77GGCCTCTTCGCTATTACG.ExpressionandExpressionofPACTwaspreformedsimilarlytotheprocedureusedtoexpressTRBP(Gredelletal.,2010;Larakietal.,2008).AnovernightcultureoftheBL21(DE3)-RIPLcells(Agilentcat#230280)withthepMBP-6His-TEV-PACTplasmidwasdilutedtoOD6000.05andculturedinLBbrothuntiltheculturedreachedanOD600of0.6atwhichtimecultureswereinducedwith300mMIPTGfor2h.Cellswerecollectedbycentrifugationat4000rpmfor10minandpelletswerefrozenuntilCellpelletsweresuspendedincolumn(20mMTris-HClpH7.4,200mMNaCl,1mMEDTA)andsupplementedwith1mg/mLlysozyme,250UofDNaseIand1xCompleteproteaseinhibitor(Rochecat#11873580001).Cellswereincubatedfor30minwithgentlerockingfollowedbysonication.Thecelllysatewasthenclearedbycentrifugationat15,000gfor25minat4°C.Thecelllysatewaspassedthrougha0.22MandloadedontoanAKTAFPLCwithaMBPTrapHPcolumn(GEHealthcarecat#28-9187-79).Thecolumnwaswashedwith10columnvolumes(CV)ofcolumnandstepelutedwith5CVofcolumnsupplementedwith10mMMaltose.FractionswereanalyzedbySDS-PAGEandfrozenat-80°C.ProteinquanwasdonewithQuickStartBradfordDye(BioRadcat#5000205)withBSAstandardsandpurityassessedbySDS-PAGEgelstainedwithGelCodeBlue(ThermoFishercat#24590).78RecombinantExpressionofHis-andMBP-NS3CloningofMBP-NS3ThepET28a-NS3vectorwasagiftfromDr.KeqiongYe.ThevectorwaswithTop10cellsandusedforproteinexpressionwithoutanymotothevector.AnMBP-taggedversionofNS3wasgeneratedbyIn-fusioncloningintothepMBP-6His-TEV-MMS2vector.ThevectorwasdigestedandcleanedupsimilarlytothegenerationofthepMBP-6His-TEV-PACTvector.PCRwasperformedwithAdvantageHDpolymerasemastermixwithastep.ThefollowingprimerswereusedforPCR50-GTACTTCCAGGGATCCATGGGCAACGTGTTCACATCGTCand50-GGCCAGTGCCAAGCTTTTACAGCACAGCTGGAGAGCTGC.cationwasperformedbyrunning10cycles:98°Cfor10sec,64-59°Cfor15sec,and72°Cfor45sec.Thesecondstepintheprocedurewasrampeddownfrom64°Cto59°Coverthecourseofthe10cycles.ofthetargetwasthencontinuedbyrunning25cyclesasfollows:98°Cfor10sec,59°Cfor15sec,and72°Cfor45sec.ThePCRprod-uctwasDPNItreatedandcleanedupbeforeIn-fusionhomologousrecombinationwiththepMBP-6His-TEV-MMS2vector.ConstructsweretransformedintoTop10CompetentcellsandclonesvbySangersequencing.ExpressionandBothaHis-taggedandMBP-taggedconstructwereexpressedinE:coliRosetta2(DE3)cells(Novagencat#71397-3).ExpressionandwasperformedwithsimilarconditionstothoseusedbyShenetal.(2010).AnovernightcultureofexpressingE:coli79Rosetta2(DE3)expressingMBP-NS3wasusedtoinoculateculturestoanOD600of0.01andgrowntoanODof˘0.4atwhichtimetheincubatorwasturneddownfrom37°Cto16°C.OncetheculturesreachedanOD600of˘0.6cultureswereinducedwith400MofIPTGandculturedovernight.Cellswerecollectedbycentrifugationandfrozenat-80°C.ofHis-NS3requiredtheuseofaHisTrapHPcolumn(GEHealthcarecat#17-5247-01),followedbyaHeparinHPcolumntoremovesmallcontaminatingRNAs.,thecellpelletcontainingHis-taggedNS3wassuspendedinH300(20mMHEPES-KOHpH7.6,300mMKCl,5%Glycerol,and25mMImidazole)supplementedwith1xCompleteProteaseInhibitors,1mMEDTA,and5g/mLDNAseI(Roche).Cellswereincubatedwith1mg/mLLysozymefor10minandthensonicated.Cellsupernatantwascollectedbycentrifugationat15,000rpmandthenpassedthrougha0.22MThecelllysatewasthenpassedoveraHisTrapHPcolumnusingtheAKTAFPLCsystem.Thecolumnwasthenwashedwith10CVofthesameH300butwith50mMImidazole.Alineargradientgoingfrom50mMImidazoleto500mMImidazoleover5CVwasusedtoeluteHis-taggedNS3fromthecolumn.FractionscontainingthemostproteinwerepooledandloadedontoaHiTrapHeparinHP(GEHealthcarecat#17-0406-01)columnequilibriatedinH(20mMHEPES-KOHpH7.6,5%Glycerol)with100mMKCl.Thecolumnwaswashedwith5CVofHwith300mMKClandHis-NS3elutedwithalineargradientbetween300mMand1000mMKClover3CV.FractionswereanalyzedbySDS-PAGEandthefractionscontainingtheHis-NS3weresupplementedwithDTTtoaconcentrationof5mMandthenfrozenandstoredat-80°C.80NucleicAcidsand50-32PLabelingsiRNAsusedforcross-linkingassays(Figure2.1)arethesameasthoseusedGredelletal.(2010).siRNAsweresynthesizedassinglestrandswithandwithoutthesubstituted4-thiouracilbasebyDharmacon.siRNAstrandswereeitherHotlabeledwithATP-[32P](PerkinElmercat#BLU502A)orColdwithATP.Labelingreactionsweresetupwith3pmolofRNAand10pmolATPusingT4PolynucleotideKinase(NEBcat#M0201)ina25Lreaction.ComplementarysiRNAstrandsmixedandsupplementedwith1xSTE(10mMTris-HClpH7.4,100mMNaCl,and1mMEDTA),thenheatedto90°Cfor3min,andcooledto37°Cfor60min.ExcesslabelwasremovedusingG-25sephadexspincolumns(Rochecat#11273990001).GelShiftAssayNativegelshiftassayswereusedtofunctionalityoftheMBP-PACTandHis-NS3constructsaswellastodeterminetheirnityforbothsiRNAsandssRNAs.,30,000cpm(˘1nM)of50labeledsiRNAorssRNAwasincubatedwithanincreasingproteinina10Lreaction.MBP-PACTbindingreactionsweresupplementedwith20mMHEPESpH7.6,40mMKCl,1.5mMMgCl2,0.1%CA630,and10UofSUPERaseIn(Ambioncat#AM2696).His-NS3bindingreactionsweresupplementedwithHepes-KOHpH7.6,100mMKCl,2mMMgCl2,1mMDTT,and0.01%CA630.Bindingreactionswereincubatedfor30minatroomtemperatureafterwhich2Lof5xloadingwasaddedtoeachreaction,andthenloadedonanative1xTBEpolyacrylamidegel.Afterelectrophoresis,thegelwastransferredtoWhatmanpaperanddriedundervacuumat80°Cfor6081min.Thedriedgelwasexposedtoastoragephosphorscreenfor˘16handimagedonanAmershamStorm860scanner.BindingandCross-linkingAssayExperimentalDesignTRBPlocalizationwasmeasured,previously,withtheuseofasubstitutedphotoreactiveuracilbase(4-thiouracil)atthe20thpositionofthesiRNA(Gredelletal.,2010).Asym-metricsiRNAterminuslocalizationwasdemonstratedbyalternatingthelocationofthe4-thiouracilcross-linkerbetweenthetwoterminiandcomparingthefractionsofTRBPcross-linkedforeach(Figure3.1A).ThefractionofansiRNAstrandthatcross-linkedtoTRBPwasvisualizedwitha32P-label,enzymaticallyincorporatedonthe50terminusofthesiRNAstrandcontainingthe4-thiouracilbase.ThecomplexedandfreesiRNAstrandswereresolvedusingdenaturingSDS-PAGE,allowingtheseparationofthetwoRNAstrands,andvisualizedwitha32P-labelonthe50terminusofthesiRNAviaPhosphorimaging.Protein-siRNAbindingreactionswereincubatedfor30min,followedbycross-linkingofthe4-thiouracilbasetonearbyaminoacidsbyexposureto312nmlight(MeisenheimerandKoch,1997;Sontheimer,1994).The4-thiouracilcross-linkerformsonlyshort-rangecross-links,providingtheresolutionneededtodistinguishsincross-linkingbetweenthetwo-siRNAtermini(Gredelletal.,2010;Pellino,2007).Excitationatwavelengths>300nmlimitsnon-spRNAcross-linking,whichoccursatshorterwavelengths(250-270nm)(MeisenheimerandKoch,1997).TheindividualsiRNAstrandsareindistinguishable82bySDS-PAGE,thus,requiringcross-linkingreactionstobeperformedseparatelyforeachsiRNAstrand(Figure3.1A).interminuslocalizationwerecorrelatedwithsiR-NAswithknownasymmetric(pp-luc)andsymmetric(sod1)functionalasymmetries,aswellas,athirdsequencewithunknownasymmetry(EGFP#274;Figure3.1B)(Tomarietal.,2004;Gredelletal.,2010).TwotstrategiesareusedtomanipulatesiRNAforstudyofitsonsiRNAasymmetricbinding(Tomarietal.,2004;Nolandetal.,2011;Gredelletal.,2010;Sakuraietal.,2011),asymmetricstrandselection(Suzukietal.,2015)(Chapter2),andfunctionalasymmetry(Schwarzetal.,2003;Sakuraietal.,2011).Themethodistointroducemismatchesbetweenthe50nucleotideofonestrandandthe19thnucleotideofitscomplement(Nolandetal.,2011;Schwarzetal.,2003;Tomarietal.,2004;Gredelletal.,2010).ThesecondmethodistochangethesiRNAsequenceandmodelthechangeinduplexthermodynamicswithnearest-neighborparameters(Gredelletal.,2010;Tomarietal.,2004;Schwarzetal.,2003;Suzukietal.,2015).Whileterminalmismatchescansuccessfullyaltertheasymmetriccross-linkingpatternbetweenTRBPandsomeduplexes,notallsiRNAcross-linkingpatternscanbemanipulatedwithterminalmismatches(Gredelletal.,2010),astheyareinfunctionalsiRNAmediatedsilencingassays(Schwarzetal.,2003).Alternatively,byusingsiRNAswithvaryingsequence,thesiRNAstructuresaregeometricallymoreconsistentbutthereistheconcernthatthelocalRNAsequencecanchangetheof4-thiouracilcross-linking(Nolandetal.,2011).Forthisreason,weusedagenericdsRBP,NS3,todemonstratethattheofcross-linkingdoesnotchangewithlocalRNAsequence.83ExperimentalSetupSimilartothegelshiftassay,10Lproteinbindingreactionswith30,000cpm(˘1nM)ofsiRNA(with4-thiourcailbasesubstitution,Figure2.1A)wasincubated50nMofHis-NS3or1MofMBP-PACTfor30minatroomtemperature,allowingbindingreactionstoreachequilibrium.Bindingreactionsweretheplacedonanicecoldaluminumblock,coveredwithapetridishtoblockwavelengths<300nMandexposedtoUVlight(312nM)withaTransilluminator(FisherScienfor10min.Reactionswerethenmixedwith5L3xSDSLoadingDye(NEBcat#B7703S)andheatedto95°Cfor5minbeforebeingresolvedbySDS-PAGEanalysis.SDS-PAGEgelswereresolvedsimilartonativegels.84AppendixCExpressionofBioactiveBrain-DerivedNeurotrophicFactor(BDNF)inBrevibacilluschoshinensisThisworkwasproducedaspartofacollaborationandcontainsatamountofworkbutdoesnotthematicallywithintherestofmydissertationandisthusbeingincludedintheappendix.AbstractBackground:Brain-derivedneurotrophicfactor(BDNF)isamemberoftheneurotrophinfamilycriti-calforneuronalcellsurvivalanderentiation,makingcharacterizationofitsfunctionessentialforunderstandingnervecellbiologyandforthedevelopmentoftherapeuticsforneurologicaldisordersandspinalcordinjuries.NeurotrophinslikeBDNFfoldwithacharacteristiccystine-knotconformation,complicatingexpressionofsoluble,bioactiverecombinantproteininmosttraditionalmicrobialexpressionsystems.85Results:HereweinvestigatedtheproductionofBDNFusingBrevibacilluschoshinensis.Weevaluatedtheofmediatype(2SYandTM),temperature(25°Cand30°C),andculturetime(48-120h)ontheproductionofbioactiveBDNF.ProteinproductionwashigherinTMmedia,thoughrecoveriesbyNi2+chromatographywerecomparableto2SYmedia.Greaterbioactivity(perunitmass)wasobservedforBDNFproducedin2SYculturesatextendedtimes(96hat30°Cor>72hat25°C),whichwascomparabletobioactivityofcommercially-availableBDNF.Conclusion:ThisstudyprovidesthatB.choshinensisiscapableofproducingbiologicallyactiveBDNF.ThechoiceofcultureconditionsimpactedtheproductionrateandcationofbioactiveBDNF,andfurtheroptimizationispossible.Oftheconditionstested,theconditionthatledtothegreatestquantityofbiologicallyactiveproteinintheshortestculturetimewasin2SYmediaat25°Cfor72hresultingin26482g/LofbioactiveBDNF.IntroductionBrain-derivedneurotrophicfactor(BDNF)playsadirectroleintheregulationofmultipleprocessesinvolvedinneuronalcellgrowth,tiationandsurvival(Numakawaetal.,2010;Leibrocketal.,1989;BinderandScharfman,2004;a~nezandSimi,2012).BDNFisnativelyexpressedasproBDNF,whichreadilydimerizesintracellularly(Heymachand86Shooter,1995;Jungbluthetal.,1994)andmaturesuponproteolyticcleavageofthepropeptide(Leeetal.,2001;Mowlaetal.,2001;Lu,2003).BDNFsignalingismediatedbytworeceptors,tropomyosin-relatedkinaseB(TrkB)andp75(Soppetetal.,1991;Tengetal.,2005;Numakawaetal.,2010;NagaharaandTuszynski,2011).TheTrkBreceptorrecognizesthematuredimerizedformofBDNFandenhancesneuronalgrowthandsurvival(Numakawaetal.,2010;Luetal.,2005a),whilethep75receptorrecognizesproBDNFcausingneuronalcelldeath(Tengetal.,2005;Luetal.,2005a;Numakawaetal.,2010;a~nezandSimi,2012).DisruptioninBDNFsignalinghasbeenimplicatedinanumberofneurodegenerativediseases,includingAlzheimer's,Huntington's,andParkinson's(ZuccatoandCattaneo,2009;NagaharaandTuszynski,2011;Luetal.,2005a).BDNF,andothermembersoftheneurotrophinfamily,arealsonecessarytodirectneuronalregenerationpost-injury(Meneietal.,1998;Luetal.,2005b).Forthesereasons,studiesofBDNFfunctionarecriticalforunderstandingneuronalbiologyandforthedevelopmentofnewtherapeuticstrategiesforneurodegenerativediseases(NagaharaandTuszynski,2011)andspinalcordinjuries(Lynametal.,2015).However,studiesofBDNFfunctionarelimitedbythecostoftheproteinfromcommercialsources.BDNFisacystine-knotprotein(BinderandScharfman,2004),containingaseriesofnon-sequentialbondsthatmakeittoexpressproperlyfoldedproteinthatretainstheabilitytohomodimerize(Numakawaetal.,2010).AttemptstoexpressBDNFinavarietyofhostshavebeenlimitedbyaggregateformationandlowbioactivity(E:coli,B:subtilis,andS:cerevisiae)(Fukuzonoetal.,1995;Takeshitaetal.,1996;Hoshinoetal.,2002;ParkandShimizu,1996;Burnsetal.,2014,2016)orlowyields(Sf21,RK13,CHO,andHEKcells)(Meyeretal.,1994,1992;Burton,1993;Knuseletal.,1991),motivating87theneedforabetterhostforexpressingBDNFatusefulquantitieswithreasonablelaborandmaterialcosts.Brevibacilluschoshinensis,agram-positivebacterium,wasoriginallyisolatedfromsoilandfoundtosecretelargequantitiesofproteinwithlowextracellularproteaseactivity(Takagietal.,1989a).BecauseB:choshinensisisgram-positive,secretedproteincanbewithoutconcernforcontaminatingendotoxins(Ilketal.,2011).B:choshinensishasbeenusedtosuccessfullyexpressanumberofmammalianproteins,includingthegrowthfactorsVEGFandNGF,whichare,likeBDNF,cystine-knotproteins(SunandDavies,1995).Inthiswork,wehaveexpressedthematureBDNFsequenceusingB:choshinensis,preliminarilyinvestigatingtheoftemperature,mediacomposition,andculturetimeonbioactiveproteinproduction.WehavefoundthatcultureconditionsgreatlyimpactedtheproductionandofbioactiveBDNF.Thisstudydemon-stratesthatB:choshinensisiscapableofproducingusefulquantitiesofbioactiveBDNFinlaboratoryscalecultures.MethodsPlasmidConstructionandExpressionStrainBDNFcDNAencodingthematuresequence(aminoacids129-247)waspurchasedfromBioclon(SanDiego,CA)andsubclonedwithaC-terminal6xHis-tagintothepNCMO2B:choshinensisexpressionvector(Clontechcat#HB112)byIn-fusioncloning(Clontechcat#638909).,theBDNFgenewasPCRwiththefollowingprimers,50-TCCCATGGCTTTCGCTCACTCTGACCCGGCTCGCand8850-TACCGAATTCCTCGACAGCCGGATCTCAGTGGTGG,containing16bpand15bp50overlapswiththepNCMO2plasmid,respectively.Anadditionalnucleotidewasincludedintheforwardprimertomaintainthecorrecttranslationalframe.ThepNCMO2vectorwasdigestedwithPstI-HFandXhoIrestrictionenzymesovernight(NEBcat#R3140andR0146,respectively).BoththePCRproductandplasmidwerecleanedupusingaPCRcleanupkit(Qiagencat#28104),recombinedwithIn-fusioncloning,andtransformedintoJM109cells(Promegacat#L1001).ColonieswerescreenedbycolonyPCRandvbySangersequencingwiththefollowingprimers,50-CGGGCTTTAAAAAGAAAGATAand50-CCAAATGGTGTTACTTTGAGA.B:choshinensiscompetentcellswerepurchasedfromClontech(cat#HB116).Thestrainusedinthisstudyhasbeengeneticallymotoremoveanyremainingproteolyticgenesandthegenesresponsibleforsporulation(HanagataandNishijyo,2010).Transforma-tionofvplasmidswasperformedbytheTris-PEGmethod,perthemanufacturer'sinstructions(Clontech).Singlecolonieswereobtainedaftertransformation,andproteinexpressionwasvbywesternblotting(detailsbelow).B:choshinensisCultureConditionsProteinwasexpressedin50mLshakeineitherTMMedia(1%glucose,1%polypep-tone,0.5%meatextract,0.2%yeastextract,0.001%FeS04*7H2O,0.001%MnSO4*4H2O,0.0001%ZnSO4*7H2O,50g/mLneomycin)or2SYMedia(2%glucose,4%soytone,0.5%yeastextract,0.015%CaCl2*7H2O,50g/mLneomycin),perthemanufacturer'srecom-mendations(Clontech).CultureswereinoculatedwithasinglecolonyfromtransformedB:choshinensisandgrownovernightat30°Cand200RPMinTMmedia.50mLcultures89wereinoculatedtoanOD600of0.01andgrowneitherat25°Cor30°Cand200RPM.Whenneeded,culturesweredilutedinPBStoensureaccuracyoftheOD600measurements.Sam-pleswereremovedorculturescollectedatsptimepointsrangingupto96hat30°Cand120hat25°C.ProteinCellswereseparatedfromtheBDNF-containingsupernatantsbycentrifugationat8000gfor10min.Supernatantsweresupplementedwith10xMOPS(250mMMOPSpH7.4,5MNaCl,and40mMimidazole)toa1xconcentrationtocontrolpH,ionicconcentration,andnon-spbindingtoNi2+Sepharose6FastFlow(referredtoasNi2+sepharose;GEHealthcarecat#17-5318-06).Supernatantswerethencentrifugedat16,000gfor20minat4°Ctoremoveanyremaininginsolublematerial.35mLoftherecoveredsupernatantswerethenincubatedwith200LofNi2+sepharosefor1hwithend-over-endrotationandrecoveredbycentrifugationat1000gfor5min.TheNi2+sepharosebeadswerewashed3timesin1xMOPSandeluted3timesin1xMOPSwith300mMimidazole,usingcentrifugationat1000gfor5mintoseparatetheNi2+sepharosefromthewashorelutionAllthreeelutionswerecombinedandsterilizedwitha0.22mPVDFlowbinding(Milliporecat#SLGV013SL).SDS-PAGEAnalysis,WesternBlotting,andBDNFQuanSamplesforSDS-PAGEandwesternblottingweresupplementedwith3xSDSLoad-ingDye(CellSignalingcat#7722)and30x1.25MDTT,andthenheatedto95°Cfor5minbeforebeingresolvedona4-20%SDS-PAGEgel.Gelstainingwasperformedusing90GelCodeBlue(ThermoFishercat#24590)accordingtothemanufacturer'sinstructions.Separategelswererunforwesternblotanalysis.Gel-separatedproteinsweretransferredontoa0.2mnitrocellulosemembrane(Bio-Radcat#161-0112)at90Vfor70min.BDNFwasdetectedwitheitherananti-BDNF(N20)(SantaCruzcat#sc-546),anti-BDNF(H117)(SantaCruzcat#sc-20981),orHRPconjugateanti-6xHis(CellSignalingcat#9991)anti-body.Detectionwiththeanti-BDNF(N20)antibodywasperformedasfollows,membraneswereblockedfor1hatRTin5%Milk-TBST,incubatedwithprimaryantibodyat1:2500dilutionovernightat4°Cin5%Milk-TBST,washed3timesfor15mininTBST,incubatedinsecondaryHRP-linkedanti-Rabbitantibodyata1:1000dilutioninTBSTatRTfor2h,washed3timesfor15mininTBST,andthenresolvedwithSuperSignalWestFemtoMaxi-mumSensitivitySubstrate(ThermoFishercat#34094).Detectionwithanti-BDNF(H117)antibodywasperformedwithsimilarconditions,except5%BSAwasusedinsteadof5%Milk,primaryantibodydilutionwasat1:200,andsecondaryantibodyincubationwasin5%BSA-TBST.Abiotinylatedladder(CellSignalingcat#7727)wasusedforsizeiden-onwesternblots;HRP-linkedanti-biotinantibodywasaddedat1:5000dilutionwiththesecondaryantibody.Fordetectionwiththeanti-6xHisantibody,membraneswereblockedwith5%BSA-TBST(0.1%Tween-20)for3h,incubatedwithprimaryantibodyat1:5000dilutionovernightat4°Cin5%BSA-TBST(0.1%Tween-20),washed3timesfor15min,washed3timesfor15minandresolvedwiththesamechemiluminescentsubstrate.ATween-20concentrationof0.05%wasusedunlessotherwisespBDNFconcentrationsweredeterminedbywesternblottingwiththeanti-6xHisanti-bodyandquanrelativetoasetofdilutonsofaBDNFstandard.TheBDNFstandardwasdtohomogeneityfromTMculturesupernatantunderdenaturingandreducing91conditionsandquandbyBradfordAssayincomparisontoaBSAstandard(FigureC.1).DuetothedenaturingconditionsrequiredfortheBDNFstandardretainednobioactivityandthuscouldonlybeusedasawesternblotstandard.FigureC.1:PurityofBDNFStandard.ABDNFstandardwastohomogene-ityfromB:choshinensisculturesusingNi2+sepharoseandquanbyBradfordassayusingaBSAstandard.ationrequireddenaturingandreducingconditions,whichdisruptedtheBDNFstructureandprecludedbioactivitytesting.Proteinconcentrationswere59g/mLand100g/mLforelutions1and2,respectively.ProliferationAssayforBDNFBioactivityBDNFbioactivitywasassayedviacellproliferationinmouse3T3cellsoverex-pressingtheTrkBreceptor(3T3-TrkB;generouslyprovidedbySakamoto)(McCartyandFeinstein,1998).Forthebioactivityassay,black,96-well,clearbottomplates(Costarcat#3904)werecoatedwith50Lof50g/mLpoly-Llysine(Sigma,cat#P9155)for1h,washed3timeswithwater,air-dried,andthencoatedwith50Lof2.5g/mL(Sigmacat#F1141)overnight.PlateswerewashedwithDMEM(LifeTechcat#11965092)beforecellplating.Cellswereseededat2000cells/wellinmedia92(3:1DMEM11965:HamsF12(LifeTechcat#11965092and11765047,respectively),15mMHEPES,4MMgCl2,3mML-histidine,10Methanolamine,1xITS+1(Sigmacat#I2521),2Mhydrocortisone,150g/mLG418(ThermoFishercat#10131035)for3h.Tonormalizeforcellnumber,10L(1/10thwellvolume)ofAlamarBlue(ThermoFishercat#DAL1285)wasalsoincludedinthemedia.AlamarBlueuorescencewasmeasuredonaBiotekH4platereader(ex:570nM,em:590nM,9.0)after3h.Initialreadingswereusedtonormalizeforincellplating.MediawasthenchangedtoBDNF-containingmedia,adding150Lofmediaandexchanging100Levery24hfor4days.Dilutionsofeachproteinpreparationweretestedstartingwith1%v/vproteinandseriallydiluted1:3to4.6x104%v/v.Forconsistencyinthemediacompositionandtocontrolforoftheelutiona1%v/vcompositionofelutionwasusedinalltests,equivalenttotheconcentrationofinthemostconcen-tratedproteinsamples.CommercialBDNFservedasapositivecontrolforproteinactivitycat#14-8366-62).ThisproteindidnotpossessaHis-tagandthereforewasnotusedasastandardinthewesternblottingstudiesdescribedabove.AmockproteinpreparationofB:choshinensiscarryingtheemptypNCMO2vectorundereachcultureconditionwasusedasanegativecontrol.After4days,AlamarBluewasmea-suredagain,asbefore,andinproliferationwascalculatedafterbackgroundsubtractionusingthefollowingequation:FoldDifference=FFinal;Sample=FInitial;SampleFFinal;Control=FInitial;Control:whereFistherelativeeintensity.AlamarBlueuorescencewasvtoincreaselinearlywithcellnumber(FigureC.2).Imagesofcellproliferationwereobtained93byplating20,0003T3-TrkBcells/wellina12-wellplate.Platecoatingwasperformedasdescribedabovewithvolumesofpoly-Llysineandadjustedto400L/well.TheassaywasperformedasdescribedbutwiththeomissionofAlamarBluereadings.Imagesweretakenpriortoeachmediachange.FigureC.2:StandardCurveofAlamarBlueFluorescence.3T3-TrkBcellswereplatedinblack-walled,clear-bottomedplatesfor3h.AlamarBluewasthenaddedandincubatedfor1h.AlamarBluewasthenmeasuredonaBiotekH4platereaderEx:570nm/Em:590nm;N=3.StatisticalAnalysis2-tailedt-testand2-wayANOVAanalyseswereperformedusingGraphpadPrism6.BDNFbioactivitieswerecomparedbyequivalencetestusing90%intervalstocomparepreparedBDNFtocommercialBDNF.Theseanalysesweredoneusing1-wayANOVAwithaDunnettposthoctestinGraphpadPrism6.94ImageCaptureandAlignmentImageswerecapturedusingaSpotRTColorcamera(DiagnosticInstrument,Inc.)onaLeicaMicrosystemsDMILinvertedmicroscope.ImageswerevisuallyalignedtomarksonthebottomoftheplateeachdayandalignmentwasadjustedusingImageJwiththealignslicesinstackplugin.Imageswerethencroppedto800x1000usingtheTransformJplugin(Meijeringetal.,2001)andcontrastnormalizedthroughaStackContrastAdjustmentplugin(Capeketal.,2006).ImageswerearrangedusingthepackageinR.ResultsandDiscussionSecretionofSolubleBDNFfromB:ChoshinensisHis-taggedBDNFwasconstitutivelyexpressedundertheP2promoterinB:choshinensisusingthepNCMO2vectorwithanN-terminalsecretiontag.TwotypesofgrowthmediaaresuggestedforexpressioninB:choshinensis,TMand2SY.WemonitoredB:choshinensisgrowthandBDNFexpressioninbothmediatypesover72hat30°Cand96hat25°C(FigureC.3A,C.3C,C.3E,C.3G).Supernatantandcellularfractionswereseparatedbycentrifugationandanalyzedbywesternblotting(FiguresC.3B,C.3D,C.3F,andC.3H),theexpressionofBDNF,thepresenceofaHis-tag,andthepresenceofsolubleBDNFintheculturesupernatants.95FigureC.3:B:choshinensisGrowthandBDNFExpression.GrowthofB:choshinensisandevaluationofBDNFsecretionusingtwotmediatypes,TM(A,C)and2SY(E,G)eachattwottemperatures,25°C(C,G)and30°C(A,E).GrowthwasassayedusingmeasurementsofOD600(solidcurves).Proteinexpressionwasmeasuredbywesternblotting(dashedcurves).N=3;errorbarsare1SD.(B,D,F,H)Anti-BDNF(toppanel)andanti-6xHis(bottompanel)antibodiesagainstsamplesfromB:choshinensisculturestransformedwithanemptyvector(e)ortheBDNFvector(b).Resultsareshownforthesupernatant(Sup.)andcellpellet(Cell)fractions.96ProteinrecoveredinboththesupernatantandcellularfractionsappearedatthesizeofmatureBDNF,indicatingremovaloftheB:choshinensissecretiontag(withtag:17.7kDaandwithouttag:14.5kDa;FigureC.4).AccumulationofBDNFwithoutthesecretiontaginthecellularfractionindicatesthatBDNFissuccessfullysecretedthroughtheplasmamembranebuttheneitherremainsassociatedwiththecellwallorprecipitatesoutofsolu-tion;ineithercase,itremainsassociatedwiththecellularfractionduringcentrifugation.ThesecretiontagisderivedfromthemiddlewallproteinofB:choshinensis(Mizukamietal.,2010;Yamagataetal.,1987)andmediatessecretionofasmuchas1.5g/Lofrecom-binant,ondedprotein(Takagietal.,1989b).Thus,weconcludethattheabsenceofdetectableproteinwithsecretiontagisduetotherobustnessofthesecretionsystem(i.e.,minimalintracellularaccumulationofBDNF).BDNFassociatedwiththepelletwasfoundtobelargelyinsoluble(FigureC.5)andnotfurtherortestedforbioactivity.Thatsaid,mediatypeathedistributionofBDNFbetweenthesupernatantandcellfractions,withagreaterfractionofBDNFremaininginthesupernatantintheTMculturesrelativetothe2SYcultures(TM-454%vs2SY-132%;p<0.05,2-tailedt-test;N=2;FigureC.3).97FigureC.4:BDNFSizeIndicatesLossofSecretionTag.RepresentativewesternblotofculturesupernatantsandcellpelletfractionstakenfromB:choshinensisculturesexpressingeitheranemptypNCMO2vector(e)orBDNFexpressingpNCMO2-BDNFvector(b)at30°C(A,B)or25°C(C,D)after48hand72hofculture,respectively.Sampleswereprobedwithananti-BDNF(N20)antibody(A,C)andananti-6xHisantibody(B,D).ArrowindicatesbandsptoBDNF(theoreticalsize=15kDa).LaneL-molecularweightladder(kDa).98FigureC.5:BDNFinCellularFractionisLargelyInsoluble.Cellpelletsfrom30°CTMand2SYculturesweresuspendedin1xBugBuster(EMDMilliporecat#70584-3)celllysisreagentandcentrifugedtoremoveinsolublematerial.Equivalentvolumesofcellsupernatantandthesolubilizedfractionfromthecellpelletwereloadedineachwell.ofBDNFbyNi2+IMACforBioactivityTestingCulturesweregrownintriplicate,andBDNFwasfromculturesupernatantsbyimmobilizedmetalychromatography(IMAC)usingNi2+sepharose.Withthisone-stepprotocol,wewereabletopurifytquantitiestotestBDNFbioactivity,butthelowstringencymethodresultedinvariableproteinrecovery(FigureC.6A,TableC.1)andpurity(FigureC.7).HigheryieldswereobtainedforlowerinitialproteinloadingsindicatingsaturationoftheNi2+sepharose;however,themaximumquantityofBDNFrecovered(18g)waswellbelowthetheoreticalbindingcapacityoftheNi2+sepharose(8mg).Nonetheless,someBDNFdidnotbindtotheNi2+sepharose(indicatedbypresenceofBDNFwthrough,FigureC.6B)orboundanddidnotelutewithimidazole(FigureC.62B).BothoftheseindicatethepresenceofaggregatesthatcanprecludeHis-tagaccessibilityandprecipitateontheNi2+sepharose,potentiallyfoulingtheNi2+sepharoseandlimitingitsbindingcapacityintheprocess.99FigureC.6:ofBDNF.(A)QuantityofBDNFrecoveredbyIMACcationdeterminedbywesternblottingwithastandardcurve.N=3;errorbarsrepresent1SD.(B)Representativewesternblotsofsamplestakenduringtheprocessincluding:input(35mL),w-through(35mL),combinedelutions(600L),andNi2+sepharosesuspendedinelution(600L);volumesarethetotalvolumesateachstepintheprocess.6.67Lwereloadedperlane.TheNi2+sepharosesuspendedinelutionwasanalyzedtodetectBDNFthatremainedassociatedwiththeNi2+sepharoseafterelution.Samplesforeachculturecondition(eachrow)wererunonasinglewestern.Imagesshowthefractionsforsamplesfromtheearliesttimeatwhichmaximalproteinbioactivitywasmeasuredforthegivencondition(seeFigureC.8).100TableC.1:ProteinRecoveredatStagesoftheBDNFProcess.QuantityofBDNFfromculturesupernatants(ProteinLoaded)andfollowingproteinpu-(Proteinwasquanbywesternblotting.N=3(exceptfor(a)whereN=2);errorbarsrepresent1SD.101FigureC.7:BDNFPurityAfterIMACChromatography.BDNFbyIMACchromatographywasanalyzedona4-20%denaturingSDS-PAGEgelandstainedwithGelCodeBluetotalproteinstain.Eachlaneisdenotedwiththemediatype,culturetemperature,andtimeofculturegrowth(h).LaneL-molecularweightladder(kDa).Byimageanalysis,BDNFpurityafterthisstepisestimatedtobeonaverage50%.RecombinantBDNFisBioactiveBDNFbioactivitywasmeasuredviaaproliferationassay,usingmouseNIH3T3asts,whichdonotnaturallyexpresstheTrkBreceptor,andwereinsteadengineeredtorespondtoBDNFthroughoverexpressionofTrkB(McCartyandFeinstein,1998).ActivationofTrkBsignalingwithBDNFcausesamitogenicresponseinthe3T3-TrkBcells(McCartyandFeinstein,1998).IntheabsenceofserumandBDNF,3T3-TrkBcellsdonotproliferatebutremaingenerallyhealthyandmetabolicallyactive(FigureC.8C,(-)Control).CellproliferationinthepresenceorabsenceofBDNFinserum-freemediawasmonitoredoverthecourseof4dayswiththemediachangeddaily.After4days,AlamarBluewasused102toassesscellnumber,andproliferationwasdeterminedbytheincreaseincellnumberofBDNF-treatedwellsrelativetocontrolwells(FigureC.8A).BecauseBDNFsampleswerenotdialyzedafterelutionfromtheNi2+sepharose,theimidazoleconcentrationpresentat1%v/vofBDNF(3mM)wasmadeupinculturesexposedtodilutedBDNFsamples.AsingleEC50curvewastobioactivitydataofBDNFproteinpreparedfromthreereplicateB:choshinensisculturesforeachcultureconditionandtime(FigureC.9).103FigureC.8:BioactivityofBDNF.BDNFspcactivitywasmeasuredbyprolif-erationassay.(A)EC50valuesmeasuredforBDNFpreparedfromthreereplicateB:choshinensisculturesforeachexpressionconditionandwithasingledose-responsecurve;(^)EC50wasnotreachedatthehighestconcentrationtested(1%v/v);errorbarsin-dicate1SDofthedose-responsecurve(*)bioactivityiscomparabletocommercially-availableBDNFbyequivalencetestusing90%intervals(FigureC.10).(B)Comparisonofbioactivityofproteinfromculturestransfectedwithanemptyvec-tor(e)ortheBDNFvector(b).(C)edproteinsampleswereaddedat1%v/vtoserum-freemedium.RecombinantBDNFat100ng/mLandwellswithoutBDNFwereusedas(+)and(-)controls,respectively.Shownarerepresentativeimagesfromdays1and5ofthesamelocationwithineachwell.ImagesareshownforcellsexposedtoproteinfromtheearliestB:choshinensisculturesatwhichmaximalproteinbioactivitywasmeasuredforthegivenmediaandtemperatureconditions.Scalebarrepresents100m.ImagesfromalldaysareincludedinFigureC.11.104FigureC.9:BDNFEC50Curves.(A)EC50curvesweretoproliferationdatafromthreereplicateculturesof3T3-TrkBcells,eachtreatedwithBDNFpreparedinaseparateB:choshinensisculture.BDNFproteinconcentrationsweredeterminedbywesternblot-ting(TableC.1).(B)BioactivityofcommercialBDNFtoasingledose-responsecurvefromthreereplicates.Errorbarsindicate1SDofthedose-responsecurve105FigureC.10:IntervalsComparingBDNFSamplestoBioactiveBDNF.BDNFbioactivitywascomparedtocommercialBDNFusinganequivalencetestwith90%intervals,calculatedusing1-wayANOVAwithDunnett'sposthocanalysis.Sampleswhoseerrorbarsarewhollycontainedbetween-1and+1areconsideredequivalenttocommercially-availableBDNF.106FigureC.11:BDNFBioactivityImages.3T3-TrkBcellsweretreatedwithBDNFthreehoursafterplating(Day1)and,subsequently,every24hthereafter(Days2-5).proteinfromeachcultureconditionwasaddedat1%v/vtotheserum-freemedium.RecombinantBDNFat100ng/mLandnoBDNFwereusedaspositiveandnegativecontrols,respectively.Threeimagesweretakenineachwellateachtimepoint;onlyonesetofimagesareshown.ImagesareshownforcellsexposedtoproteinfromtheearliestB:choshinensisculturesatwhichmaximalproteinbioactivitywasmeasuredforthegivenmediaandtemperatureconditions.Changesinbioactivitywithculturetimewereinvestigatedforeachmediatype,culturingforaminimumof48handupto120hat25°Cor96hat30°C.Webeganwith48hculturesastheseweretheearliestculturesforwhichBDNFlevelsweremaximalundersomeconditions(FigureC.3).Wetooksamplesevery24huntilatleast3ttime-pointsweretakenafterBDNFlevelshadreachedamaximum,resultinginthelongersamplingperiodforculturesat25°Crelativeto30°C.BDNFbioactivitywascomparedtoabioactive,107commercialBDNF(FigureC.8A).Onlyproteinobtainedfromculturesgrownin2SYmediahadcomparablebioactivitytothecommercialprotein(FigureC.8A),indicatingthat,atleastwhenusingourcurrentprotocol,culturesgrowninTMmediaproducedalowerfractionofbioactiveBDNFinthesupernatants.Furthermore,in2SYmediaweobservedthatEC50valuesdecreasedwithculturetime(FigureC.8A).Thissuggeststhat,aswithTM,notalloftheproteinproducedin2SYmediaisbioactiveandthatthefractionofbioactiveproteinpresentinthesupernatantvarieswithtime.WethatbioactivitywasduesptoproteinproducedfromtheBDNF-containingvector,asshownbyouremptyvectorcontrols(FiguresC.8BandC.8C).Decreasingcellnumberswithtimeintheempty-vectorcontrolssomeeventualcytotoxicityduetotheabsenceofserumandBDNF.DiscussionInthiswork,wedemonstratedthatBDNFpurfromB:choshinensiscultureswasbio-logicallyactive.Intestingavarietyofcultureconditionsandtimes,weidenconditionsthatprovidedtyieldsforsubsequentbioactivitystudies.UseofB:choshinensisasthehostmitigatedtheprincipalyinachievingexpressionofbioactiveBDNF,itscystine-knotstructure,whichcanleadtointermolecularlinkages,incorrectintramolecularlinkages,andaggregateformation(BinderandScharfman,2004;Burnsetal.,2014;Meyeretal.,1994;Knuseletal.,1991).Nonetheless,despiteouryieldsofbioactiveprotein,considerableamountsofBDNFproducedstillformedondedaggregates(FigureC.12).Additionaloptimizationofexpressionconditionsshouldfocusonreducingtheformationofaggregates,eitherduringtheexpressionprocessorafterac-108cumulationintheculturesupernatant,whichwilllikelyhavetheconcomitantbeofimprovingrecoveryofBDNFandreducingtheconcentrationofcontaminantsthatco-purifywithBDNF.FigureC.12:EvidenceofBDNFAggregation.RepresentativeBDNFwesternblotofthecellsupernatantfromeachcultureconditionwheresampleswere(+DTT)orwerenot(-DTT)treatedwithDTTtoevaluatethepresentofintermolecularbonds(aggregates).Shownareresultsfromsamplestakenwhenproteinconcentrationsreachedamaximumforthegivenmediaandtemperatureconditions.Whileaggregateswereobservedinallconditions,thequantityofnon-aggregatedBDNFinthesupernatantwasgreaterinTMcultures,suggestingthatfurtheroptimizationtomin-imizeaggregationmaybebestfocusedonculturesusingTMmedia.WehypothesizethattheaggregationhinderedaccessibilityoftheHis-tagandfouledtheNi2+sepharose,bothcontributingtoreducedyieldsofnon-aggregatedBDNF.However,totalBDNFexpression,includingboththesupernatantandcellpelletfractions,wascomparablebetweenbothmediatypes.Thissuggeststhataggregatesarelesssolublein2SYmediathanTMmedia(FigureC.3),resultinginahigherfractionofbiologicallyactiveBDNFin2SYsupernatants.Amoreetmeansofremovingaggregatescouldbeachievedbyleveragingthesta-bilityofBDNFatpHvaluesaslowas4(TanakaandKumano,2000);attheselowpHconditions,aggregatesshouldprecipitateandcanthenberemovedbycentrifugation,sim-ilartowhathasbeendonewithEGF(Miyauchietal.,1999).Non-aggregatedBDNFcouldthenbeusingavarietyofothertechniques,includingIMAC,aswe109performedhere,orbycationicexchange,anionicexchange,orreversed-phasechromato-graphicmethods,ashavebeenstudiedelsewhere(Jungbluthetal.,1994;Burton,1993;RosenfeldandBenedek,1993).Reducingtheculturetemperaturefrom30°Cto25°Cresultedinagreaterfractionofbi-ologicallyactiveBDNFatanearlierculturetimein2SYmedia(72hvs96h,FigureC.8A).Inotherexpressionsystems(i.e.,E:coli),reductionsinculturetemperatureareknowntoimprovetheexpressionofrecombinantproteins,byreducingtherateoftranslation(VasinaandBaneyx,1997)andthepropensityforproteinaggregation(Baldwin,1986;Schellman,1997).In2SYmedia,thereductioninculturetemperaturealsoincreasedthestationaryphaseOD600anddecreasedtherateofBDNFaccumulationintheculturesupernatant,indicatingamarkedreductioninthepercellexpressionofBDNF.Whilenotexploredexplicitlyhere,thelowerrateofBDNFexpression(byreductionintemperature)doescor-respondwiththeabilitytoproducebiologicallyactiveBDNFandcouldbeexploredmorecompletelybyuseofalowerstrengthpromoter(Onishietal.,2013).Furtherreductionintemperature(20°C)resultedintlylongerlagphase,especiallyinTMmedia,andwasnotexploredfurtherduetopracticalculturetimelimitations(FigureC.13).However,alteringthecultureconditions(i.e.,beginningwithamoreconcentratedinoculum)tore-duceculturedurationwouldallowinvestigationofexpressionattemperaturesbelow25°CandcouldbebintheproductionandrecoveryofbiologicallyactiveBDNF.110FigureC.13:B:choshinensisGrowthat20°C.B:choshinensisgrowth(OD600)wasexaminedat20°Cover6daysinTM(A)and2SY(B)media.N=3,errorbarsare1SD.Withthebestconditionsexaminedhere(2SYmediaat25°Cfor72h),wewereabletopurify26482g/(Lofculture)ofbiologicallyactiveBDNF.Weconsidertheseourbestconditionsastheyresultedin:i)BDNFwiththehighestbioactivity,ii)thehighestyieldofBDNF,andiii)theminimumculturetime.ThequantityofrecoverableBDNFdidnottlyincreaseattimes>72hin2SYmediaat25°C,butwedidobserveanincreaseinthevariabilityofBDNFwithtime,withseveralculturesproducing>500g/LofbiologicallyactiveBDNF.TheyieldofBDNFisgreaterthanorofthesamemagnitudeasotherexpressionsystems(Fukuzonoetal.,1995;Meyeretal.,1994;Knuseletal.,1991;Rosenthaletal.,1991).Inaddition,theuseofamicrobialsystembothcost(<$10/Lmedia)andtechnicaladvantages(standardmicrobialculturingtechniques)overhighercomplexityexpressionsystems(Meyeretal.,1994;Knuseletal.,1991;Rosenthaletal.,1991)ormethodsthatrequirerefoldingBDNF(Fukuzonoetal.,1995).Recently,˘1mg/LexpressionofBDNFwasdemonstratedinyeastaftertheengineeringofBDNFtofacilitateproteinfolding(Burnsetal.,2014,2016);itisconceivablethatusingamoBDNFsequenceforexpressioninB:choshinensiscouldalsosigtlyincreaseexpressionofbiologicallyactiveBDNFinthishost.111ConclusionsBDNFisahighlyvaluableprotein,bothforitspotentialclinicalrelevanceandasatoolforresearchers.TheinproducingbioactiveBDNFproteintlyhinderitswidespreaduse.Here,wehaveshownthatB:choshinensiscanproduce26482g/LofbioactiveBDNF,whenculturedat25°Cin2SYmediafor72h.Similartootherexpressionsystems,expressionofBDNFinB:choshinensisproducesaggregatesthatcomplicatetheofBDNFbyIMACchromatographyandforminactiveprotein.Thisworkdemonstratestheutilityofthegram-positivehostB:choshinensistoproducebioactiveBDNFatusefulscalesandprovidesastartingpointforadditionaloptimizationstudies.112BIBLIOGRAPHY113BIBLIOGRAPHYAcevedo,R.,Orench-Rivera,N.,Quarles,K.A.,andShowalter,S.A.(2015).HelicaldefectsinmicroRNAproteinbindingbyTARRNAbindingprotein.PLoSOne,10(1):e0116749.Allerson,C.R.,N.,Jarres,R.,Prakash,T.P.,Naik,N.,Berdeja,A.,Wanders,L.,,R.H.,Swayze,E.E.,andBhat,B.(2005).Fully2'-mooligonucleotideduplexeswithimprovedinvitropotencyandstabilitycomparedtounmosmallinterferingRNA.JournalofMedicinalChemistry,48(4):901{904.Amarzguioui,M.(2003).ToleranceformutationsandchemicalmoinasiRNA.NucleicAcidsResearch,31(2):589{595.Amarzguioui,M.,Lundberg,P.,Cantin,E.,Hagstrom,J.,Behlke,M.A.,andRossi,J.J.(2006).RationaldesignandinvitroandinvivodeliveryofDicersubstratesiRNA.NatureProtocols,1(2):508{517.Amarzguioui,M.andPrydz,H.(2004).AnalgorithmforselectionoffunctionalsiRNAsequences.BiochemicalandBiophysicalResearchCommunications,316(4):1050{1058.Ameres,S.L.,Martinez,J.,andSchroeder,R.(2007).MolecularbasisfortargetRNArecognitionandcleavagebyhumanRISC.Cell,130(1):101{112.Ameres,S.L.andZamore,P.D.(2013).DiversifyingmicroRNAsequenceandfunction.NatureReviewsMolecularCellBiology,14(8):475{488.Angart,P.,Vocelle,D.,Chan,C.,andWalton,S.P.(2013).DesignofsiRNAtherapeuticsfromthemolecularscale.Pharmaceuticals,6(4):440{68.Baldwin,R.L.(1986).Temperaturedependenceofthehydrophobicinteractioninproteinfolding.ProceedingsoftheNationalAcademyofSciences,83(21):8069{8072.Bartel,D.(2009).MicroRNAs:Targetrecognitionandregulatoryfunctions.Cell,136(2):215{233.Beitzinger,M.andMeister,G.(2011).Experimentalidenofmicrornatargetsbyimmunoprecipitationofargonauteproteincomplexes.InDalmay,T.,editor,MicroRNAsinDevelopment,volume732,pages153{67.HumanaPress,Totowa,NJ.Bernard,M.A.,Wang,L.,andTachado,S.D.(2015).DICER-ARGONAUTE2complexincontinuousassaysofRNAinterferenceenzymes.PlosOne,10(3):e0120614.Bernstein,E.,Caudy,A.A.,Hammond,S.M.,andHannon,G.J.(2001).RoleforabidentateribonucleaseintheinitiationstepofRNAinterference.Nature,409(6818):363{366.114Betancur,J.G.andTomari,Y.(2012).DicerisdispensableforasymmetricRISCloadinginmammals.RNA,18(1):24{30.Bevilacqua,P.C.andCech,T.R.(1996).Minor-grooverecognitionofdouble-strandedRNAbythedouble-strandedRNA-bindingdomainfromtheRNA-activatedproteinki-nasePKR.Biochemistry,35(31):9983{9994.Binder,D.K.andScharfman,H.E.(2004).Brain-derivedneurotrophicfactor.GrowthFactors,22(3):123{131.Birmingham,A.,Anderson,E.M.,Reynolds,A.,Ilsley-Tyree,D.,Leake,D.,Fedorov,Y.,Baskerville,S.,Maksimova,E.,Robinson,K.,Karpilow,J.,Marshall,W.S.,andKhvorova,A.(2006).30UTRseedmatches,butnotoverallidentity,areassociatedwithRNAiNatureMethods,3(3):199{204.Bobbin,M.L.andRossi,J.J.(2016).RNAinterference(RNAi)-basedtherapeutics:Deliveringonthepromise?AnnualReviewofPharmacologyandToxicology,56(1):103{122.Bohula,E.A.,Salisbury,A.J.,Sohail,M.,Playford,M.P.,Riedemann,J.,Southern,E.M.,andMacaulay,V.M.(2003).TheofsmallinterferingRNAstargetedtothetype1insulin-likegrowthfactorreceptor(IGF1R)isbysecondarystructureintheIGF1Rtranscript.JournalofBiologicalChemistry,278(18):15991{15997.Braasch,D.A.,Jensen,S.,Liu,Y.,Kaur,K.,Arar,K.,White,M.A.,andCorey,D.R.(2003).RNAinterferenceinmammaliancellsbychemically-moRNA.Biochemistry,42(26):7967{7975.Bramsen,J.B.,Laursen,M.B.,Damgaard,C.K.,Lena,S.W.,Babu,B.R.,Wengel,J.,andKjems,J.(2007).ImprovedsilencingpropertiesusingsmallinternallysegmentedinterferingRNAs.NucleicAcidsResearch,35(17):5886{5897.Bramsen,J.B.,Laursen,M.B.,Nielsen,A.F.,Hansen,T.B.,Bus,C.,Langkjaer,N.,Babu,B.R.,Hojland,T.,Abramov,M.,VanAerschot,A.,Odadzic,D.,Smicius,R.,Haas,J.,Andree,C.,Barman,J.,Wenska,M.,Srivastava,P.,Zhou,C.,Honcharenko,D.,Hess,S.,Muller,E.,Bobkov,G.V.,Mikhailov,S.N.,Fava,E.,Meyer,T.F.,Chattopadhyaya,J.,Zerial,M.,Engels,J.W.,Herdewijn,P.,Wengel,J.,andKjems,J.(2009).Alarge-scalechemicalmoscreenidendesignrulestogeneratesiRNAswithhighactivity,highstabilityandlowtoxicity.NucleicAcidsResearch,37(9):2867{2881.Broderick,J.A.,Salomon,W.E.,Ryder,S.P.,Aronin,N.,andZamore,P.D.(2011).ArgonauteproteinidentityandpairinggeometrydeterminecooperativityinmammalianRNAsilencing.RNA,17(10):1858{1869.Brown,K.M.,Chu,C.Y.,andRana,T.M.(2005).TargetaccessibilitydictatesthepotencyofhumanRISC.NatureStructural&MolecularBiology,12(5):469{470.115Burns,M.L.,Malott,T.M.,Metcalf,K.J.,Hackel,B.J.,Chan,J.R.,andShusta,E.V.(2014).Directedevolutionofbrain-derivedneurotrophicfactorforimprovedfold-ingandexpressioninSaccharomycescerevisiae.AppliedEnvironmentalMicrobiology,80(18):5732{5742.Burns,M.L.,Malott,T.M.,Metcalf,K.J.,Puguh,A.,Chan,J.R.,andShusta,E.V.(2016).Pro-regionengineeringforimprovedyeastdisplayandsecretionofbrainderivedneurotrophicfactor.BiotechnologyJournal,11(3):425{36.Burroughs,A.M.,Ando,Y.,deHoon,M.J.L.,Tomaru,Y.,Suzuki,H.,Hayashizaki,Y.,andDaub,C.O.(2011).Deep-sequencingofhumanArgonaute-associatedsmallRNAsprovidesinsightintomiRNAsortingandrevealsArgonauteassociationwithRNAfragmentsofdiverseorigin.RNABiology,8(1):158{177.Burton,L.E.(1993).Productionofrecombinantneurotrophicfactors.InABolton,GBaker,F.H.,editor,NeurotrophicFactors,volume25,pages155{176.SpringerScience.Capek,M.,acek,J.,andKubva,L.(2006).Methodsforcompensationofthelightat-tenuationwithdepthofimagescapturedbyaconfocalmicroscope.MicroscopyResearchandTechnique,69(8):624{635.Carthew,R.W.andSontheimer,E.J.(2009).OriginsandmechanismsofmiRNAsandsiRNAs.Cell,136(4):642{655.Cekaite,L.,Furset,G.,Hovig,E.,andSioud,M.(2007).Geneexpressionanalysisinbloodcellsinresponsetounmodand20-mosiRNAsrevealsTLR-dependentandindependentJournalofMolecularBiology,365(1):90{108.Chandradoss,S.D.,Schirle,N.T.,Szczepaniak,M.,MacRae,I.J.,andJoo,C.(2015).AdynamicsearchprocessunderliesMicroRNAtargeting.Cell,162(1):96{107.Chang,C.I.,Yoo,J.W.,Hong,S.W.,Lee,S.E.,Kang,H.S.,Sun,X.,H.A.,Ban,C.,Kim,S.,Li,C.J.,andLee,D.(2009).Asymmetricshorter-duplexsiRNAstructurestriggertgenesilencingwithreducednonspMolecularTherapy,17(4):725{732.S.,Santos,C.O.D.,Chong,M.M.W.,andHannon,G.J.(2010).Adicer-independentmiRNAbiogenesispathwaythatrequiresAgocatalysis.Nature,465(7298):584{589.Chen,C.,Zhu,C.,Huang,J.,Zhao,X.,Deng,R.,Zhang,H.,Dou,J.,Chen,Q.,Xu,M.,Yuan,H.,Wang,Y.,andYu,J.(2015).SUMOylationofTARBP2regulatesmiRNA/siRNA.NatureCommunications,6:8899.Chen,C.F.,Ridzon,D.A.,Broomer,A.J.,Zhou,Z.H.,Lee,D.H.,Nguyen,J.T.,Barbisin,M.,Xu,N.L.,Mahuvakar,V.R.,Andersen,M.R.,Lao,K.Q.,Livak,K.J.,andGuegler,K.J.(2005).Real-timequanationofmicroRNAsbystem-loopRT-PCR.NucleicAcidsResearch,33(20):e179.116Chendrimada,T.,Gregory,R.,Kumaraswamy,E.,Norman,J.,Cooch,N.,Nishikura,K.,andShiekhattar,R.(2005).TRBPrecruitstheDicercomplextoAgo2formicroRNAprocessingandgenesilencing.Nature,436(7051):740{744.Chiu,Y.L.andRana,T.M.(2003).siRNAfunctioninRNAi:Achemicalmoanalysis.RNA,9(9):1034{1048.Chu,C.andRana,T.M.(2008).PotentRNAibyshortRNAtriggers.RNA,14(9):1714{1719.Collingwood,M.A.,Rose,S.D.,Huang,L.,Hillier,C.,Amarzguioui,M.,Wiiger,M.T.,Soifer,H.S.,Rossi,J.J.,andBehlke,M.A.(2008).ChemicalmoionpatternscompatiblewithhighpotencyDicer-substratesmallinterferingRNAs.Oligonucleotides,18(2):187{200.Czech,B.andHannon,G.J.(2010).SmallRNAsorting:MatchmakingforArgonautes.NatureReviewsGenetics,12(1):19{31.Daniels,S.M.andGatignol,A.(2012).ThemultiplefunctionsofTRBP,atthehubofcellresponsestoviruses,stress,andcancer.MicrobiologyandMolecularBiologyReviews,76(3):652{66.Daniels,S.M.,Melendez-Pena,C.E.,Scarborough,R.J.,Daher,A.,Christensen,H.S.,ElFar,M.,Purcell,D.F.J.,Laine,S.,andGatignol,A.(2009).CharacterizationoftheTRBPdomainrequiredfordicerinteractionandfunctioninRNAinterference.BMCMolecularBiology,10(1):38.Daviet,L.,Erard,M.,Dorin,D.,Duarte,M.,Vaquero,C.,andGatignol,A.(2000).Analy-sisofabindingbetweenthetwodsRNA-bindingdomainsinTRBPrevealsthemodularfunctionofaKR-helixmotif.EuropeanJournalofBiochemistry,267(8):2419{2431.De,N.,Young,L.,Lau,P.W.,Meisner,N.C.,Morrissey,D.V.,andMacRae,I.J.(2013).HighlycomplementarytargetRNAspromotereleaseofguideRNAsfromhumanArgonaute2.MolecularCell,50(3):344{355.Didiano,D.andHobert,O.(2006).PerfectseedpairingisnotagenerallyreliablepredictorformiRNA-targetinteractions.NatureStructural&MolecularBiology,13(9):849{851.Diebold,S.S.,Kaisho,T.,Hemmi,H.,Akira,S.,andReiseSousa,C.(2004).InnateantiviralresponsesbymeansofTLR7-mediatedrecognitionofsingle-strandedRNA.Science,303(5663):1529{1531.Diebold,S.S.,Massacrier,C.,Akira,S.,Paturel,C.,Morel,Y.,andReiseSousa,C.(2006).NucleicacidagonistsforToll-likereceptor7arebythepresenceofuridineribonucleotides.EuropeanJournalofImmunology,36(12):3256{3267.Doench,J.G.,Petersen,C.P.,andSharp,P.A.(2003).siRNAscanfunctionasmiRNAs.Genes&Development,17(4):438{442.117Doench,J.G.andSharp,P.A.(2004).SpyofmicroRNAtargetselectionintrans-lationalrepression.Genes&Development,18(5):504{511.Dua,P.,Yoo,J.W.,Kim,S.,andLee,D.K.(2011).MosiRNAstructurewithasinglenucleotidebulgeovercomesconventionalsiRNA-mediatedsilencing.MolecularTherapy,19(9):1676{1687.Dueck,A.,Ziegler,C.,Eichner,A.,Berezikov,E.,andMeister,G.(2012).microR-NAsassociatedwiththethumanArgonauteproteins.NucleicAcidsResearch,40(19):9850{9862.Elbashir,S.M.,Harborth,J.,Lendeckel,W.,Yalcin,A.,Weber,K.,andTuschl,T.(2001a).Duplexesof21-nucleotideRNAsmediateRNAinterferenceinculturedmammaliancells.Nature,411(6836):494{498.Elbashir,S.M.,Lendeckel,W.,andTuschl,T.(2001b).RNAinterferenceismediatedby21-and22-nucleotideRNAs.Genes&Development,15(2):188{200.Elbashir,S.M.,Martinez,J.,Patkaniowska,A.,Lendeckel,W.,andTuschl,T.(2001c).FunctionalanatomyofsiRNAsformediatingntRNAiinDrosophilamelanogasterembryolysate.TheEMBOjournal,20(23):6877{6888.Fedorov,Y.,Anderson,E.M.,Birmingham,A.,Reynolds,A.,Karpilow,J.,Robinson,K.,Leake,D.,Marshall,W.S.,andKhvorova,A.(2006).bysiRNAcaninducetoxicphenotype.RNA,12(7):1188{1196.Fire,A.,Xu,S.,Montgomery,M.K.,Kostas,S.A.,Driver,S.E.,andMello,C.C.(1998).Potentandspgeneticinterferencebydouble-strandedRNAinCaenorhabditisel-egans.Nature,391(6669):806{811.Forsbach,A.,Nemorin,J.-G.,Montino,C.,Muller,C.,Samulowitz,U.,Vicari,A.P.,Jurk,M.,Mutwiri,G.K.,Krieg,A.M.,Lipford,G.B.,andVollmer,J.(2008).IdenofRNAsequencemotifsstimulatingsequence-spTLR8-dependentimmuneresponses.JournalofImmunology,180(6):3729{3738.Foster,D.J.,Barros,S.,Duncan,R.,Shaikh,S.,Cantley,W.,Dell,A.,Bulgakova,E.,O'Shea,J.,Taneja,N.,Kuchimanchi,S.,Sherrill,C.B.,Akinc,A.,Hinkle,G.,White,A.C.S.,Pang,B.,Charisse,K.,Meyers,R.,Manoharan,M.,andElbashir,S.M.(2012).ComprehensiveevaluationofcanonicalversusDicer-substratesiRNAinvitroandinvivo.RNA,18(3):557{568.Frank,F.,Sonenberg,N.,andNagar,B.(2010).Structuralbasisfor50-nucleotidebase-sprecognitionofguideRNAbyhumanAGO2.Nature,465(7299):818{822.Friedman,R.C.,Farh,K.K.,Burge,C.B.,andBartel,D.P.(2008).MostmammalianmRNAsareconservedtargetsofmicroRNAs.GenomeResearch,19(1):92{105.118Fucini,R.V.,Haringsma,H.J.,Deng,P.,Flanagan,W.M.,andWillingham,A.T.(2012).AdenosinemomaybepreferredforreducingsiRNAimmunestimulation.Nu-cleicAcidTherapeutics,22(3):205{210.Fukuzono,S.,Fujimori,K.,andShimizu,N.(1995).Productionofbiologicallyactivematurebrain-derivedneurotrophicfactorinEscherichiacoli.Bioscience,Biotechnology,andBiochemistry,59(9):1727{1731.Gantier,M.P.andWilliams,B.R.G.(2007).Theresponseofmammaliancellstodouble-strandedRNA.Cytokine&GrowthFactorReviews,18(5-6):363{371.Gong,W.,Ren,Y.,Xu,Q.,Wang,Y.,Lin,D.,Zhou,H.,andLi,T.(2006).IntegratedsiRNAdesignbasedonsurveyingoffeaturesassociatedwithhighRNAieness.BMCBioinformatics,7:516.Goodchild,A.,Nopper,N.,King,A.,Doan,T.,Tanudji,M.,Arndt,G.M.,Poidinger,M.,Rivory,L.P.,andPassioura,T.(2009).SequencedeterminantsofinnateimmuneactivationbyshortinterferingRNAs.BMCImmunology,10(40).Gredell,J.A.,Dittmer,M.J.,Wu,M.,Chan,C.,andWalton,S.P.(2010).RecognitionofsiRNAasymmetrybyTARRNAbindingprotein.Biochemistry,49(14):3148{3155.Gregory,R.I.,pingYan,K.,Amuthan,G.,Chendrimada,T.,Doratotaj,B.,Cooch,N.,andShiekhattar,R.(2004).ThemicroprocessorcomplexmediatesthegenesisofmicroRNAs.Nature,432(7014):235{240.Grimm,D.,Streetz,K.L.,Jopling,C.L.,Storm,T.A.,Pandey,K.,Davis,C.R.,Marion,P.,Salazar,F.,andKay,M.A.(2006).FatalityinmiceduetooversaturationofcellularmicroRNA/shorthairpinRNApathways.Nature,441(7092):537{541.Gu,S.,Jin,L.,Zhang,F.,Huang,Y.,Grimm,D.,Rossi,J.J.,andKay,M.A.(2011).ThermodynamicstabilityofsmallhairpinRNAshighlytheloadingprocessoftmammalianArgonautes.ProceedingsoftheNationalAcademyofSciences,108(22):9208{9213.Ha,M.andKim,N.(2014).RegulationofmicroRNAbiogenesis.NatureReviewsMolecularCellBiology,15:509{524.Haase,A.,Jaskiewicz,L.,Zhang,H.,Laine,S.,Sack,R.,Gatignol,A.,andFilipowicz,W.(2005).TRBP,aregulatorofcellularPKRandHIV-1virusexpression,interactswithDicerandfunctionsinRNAsilencing.EMBOReports,6(10):961{967.Haley,B.andZamore,P.D.(2004).KineticanalysisoftheRNAienzymecomplex.NatureStructural&MolecularBiology,11(7):599{606.Han,J.,Lee,Y.,Yeom,K.,Kim,Y.,Jin,H.,andKim,V.N.(2004).TheDrosha-DGCR8complexinprimarymicroRNAprocessing.Genes&Development,18(24):3016{3027.119Hanagata,H.andNishijyo,T.(2010).Brevibacilluschoshinensisandprocessforprod-cuingproteinwithuseofthemicrobeashost.USPatent7,655,452.Haringsma,H.J.,Li,J.J.,Soriano,F.,Kenski,D.M.,Flanagan,W.M.,andWilling-ham,A.T.(2012).mRNAknockdownbysinglestrandRNAisimprovedbychemicalmoNucleicAcidsResearch,40(9):4125{4136.Haupenthal,J.,Baehr,C.,Kiermayer,S.,Zeuzem,S.,andPiiper,A.(2006).InhibitionofRNAseAfamilyenzymespreventsdegradationandlossofsilencingactivityofsiRNAsinserum.BiochemicalPharmacology,71(5):702{710.Haussecker,D.(2012).ThebusinessofRNAitherapeuticsin2012.MolecularTherapy|NucleicAcids,1(2):e8.Heil,F.,Hemmi,H.,Hochrein,H.,Ampenberger,F.,Kirschning,C.,Akira,S.,Lipford,G.,Wagner,H.,andBauer,S.(2004).Species-sprecognitionofsingle-strandedRNAviatoll-likereceptor7and8.Science,303(5663):1526{1529.Heymach,J.V.,J.andShooter,E.M.(1995).Thebiosynthesisofneurotrophinheterodimersbytransfectedmammaliancells.JournalofBiologicalChemistry,270(20):12297{304.Hibio,N.,Hino,K.,Shimizu,E.,Nagata,Y.,andUi-Tei,K.(2012).StabilityofmiRNA5'terminalandseedregionsiscorrelatedwithexperimentallyobservedmiRNA-mediatedsilencing.Reports,2(996).Hohjoh,H.(2004).EnhancementofRNAiactivitybyimprovedsiRNAduplexes.FEBSletters,557(1-3):193{198.Holen,T.(2006).tpredictionofsiRNAswithsiRNArules1.0:Anopen-sourceJAVAapproachtosiRNAalgorithms.RNA,12(9):1620{1625.Holen,T.,Amarzguioui,M.,Babaie,E.,andPrydz,H.(2003).Similarbehaviourofsingle-strandanddouble-strandsiRNAssuggeststheyactthroughacommonRNAipathway.NucleicAcidsResearch,31(9):2401{2407.Hong,J.,Huang,Y.,Li,J.,Yi,F.,Zheng,J.,Huang,H.,Wei,N.,Shan,Y.,An,M.,Zhang,H.,Ji,J.,Zhang,P.,Xi,Z.,Du,Q.,andLiang,Z.(2010).Comprehensiveanalysisofsequence-spstabilityofsiRNA.TheFASEBJournal,24(12):4844{4855.Hornung,V.,Guenthner-Biller,M.,Bourquin,C.,Ablasser,A.,Schlee,M.,Uematsu,S.,Noronha,A.,Manoharan,M.,Akira,S.,deFougerolles,A.,Endres,S.,andHartmann,G.(2005).Sequence-sppotentinductionofIFN-byshortinterferingRNAinplasmacytoiddendriticcellsthroughTLR7.NatureMedicine,11(3):263{270.Hoshino,K.,Eda,A.,Kurokawa,Y.,andShimizu,N.(2002).Productionofbrain-derivedneurotrophicfactorinEscherichiacolibycoexpressionofDsbproteins.Bioscience,Biotechnology,andBiochemistry,66(2):344{350.120Hossbach,M.,Gruber,J.,Osborn,M.,Weber,K.,andTuschl,T.(2006).GenesilencingwithsiRNAduplexescomposedoftarget-mRNA-complementaryandpartiallypalin-dromicorpartiallycomplementarysingle-strandedsiRNAs.RNABiology,3(2):82{89.Huesken,D.,Lange,J.,Mickanin,C.,Weiler,J.,Asselbergs,F.,Warner,J.,Meloon,B.,Engel,S.,Rosenberg,A.,Cohen,D.,Labow,M.,Reinhardt,M.,Natt,F.,andHall,J.(2005).Designofagenome-widesiRNAlibraryusinganneuralnetwork.NatureBiotechnology,23(8):995{1001.a~nez,C.F.andSimi,A.(2012).p75neurotrophinreceptorsignalinginnervoussysteminjuryanddegeneration:Paradoxandopportunity.TrendsinNeurosciences,35(7):431{440.Ilk,N.,Schumi,C.-T.,Bohle,B.,Egelseer,E.M.,andSleytr,U.B.(2011).Expressionofanendotoxin-freeS-layer/allergenfusionproteiningram-positiveBacillussubtilis1012forthepotentialapplicationasvaccinesforimmunotherapyofatopicallergy.MicrobialCellFactories,10:6.Iwasaki,S.,Kobayashi,M.,Yoda,M.,Sakaguchi,Y.,Katsuma,S.,Suzuki,T.,andTomari,Y.(2010).Hsc70/Hsp90chaperonemachinerymediatesATP-dependentRISCloadingofsmallRNAduplexes.MolecularCell,39(2):292{299.Jackson,A.L.,Bartz,S.R.,Schelter,J.,Kobayashi,S.V.,Burchard,J.,Mao,M.,Li,B.,Cavet,G.,andLinsley,P.S.(2003).ExpressionrevealsgeneregulationbyRNAi.NatureBiotechnology,21(6):635{637.Jackson,A.L.,Burchard,J.,Leake,D.,Reynolds,A.,Schelter,J.,Guo,J.,Johnson,J.M.,Lim,L.,Karpilow,J.,Nichols,K.,Marshall,W.,Khvorova,A.,andLinsley,P.S.(2006a).Position-spchemicalmoofsiRNAsreducestranscriptsilencing.RNA,12(7):1197{1205.Jackson,A.L.,Burchard,J.,Schelter,J.,Chau,B.N.,Cleary,M.,Lim,L.,andLinsley,P.S.(2006b).WidespreadsiRNAtranscriptsilencingmediatedbyseedregionsequencecomplementarity.RNA,12(7):1179{1187.Jackson,A.L.andLinsley,P.S.(2010).RecognizingandavoidingsiRNAfortargetidenationandtherapeuticapplication.NatureReviewsDrugDiscovery,9(1):57{67.Jagla,B.,Aulner,N.,Kelly,P.D.,Song,D.,Volchuk,A.,Zatorski,A.,Shum,D.,Mayer,T.,DeAngelis,D.A.,Ouerfelli,O.,Rutishauser,U.,andRothman,J.E.(2005).SequencecharacteristicsoffunctionalsiRNAs.RNA,11(6):864{872.Judge,A.D.,Sood,V.,Shaw,J.R.,Fang,D.,McClintock,K.,andMacLachlan,I.(2005).Sequence-dependentstimulationofthemammalianinnateimmuneresponsebysyntheticsiRNA.NatureBiotechnology,23(4):457{462.121Jungbluth,S.,Bailey,K.,andBarde,Y.A.(1994).andcharacterisationofabrain-derivedneurotrophicfactor/neurotrophin-3(BDNF/NT-3)heterodimer.EuropeanJournalofBiochemistry,221(2):677{685.Kariko,K.,Bhuyan,P.,Capodici,J.,andWeissman,D.(2004).SmallinterferingRNAsme-diatesequence-independentgenesuppressionandinduceimmuneactivationbysignalingthroughtoll-likereceptor3.JournalofImmunology,172(11):6545{6549.Kato,H.,Takeuchi,O.,Mikamo-Satoh,E.,Hirai,R.,Kawai,T.,Matsushita,K.,Hiiragi,A.,Dermody,T.S.,Fujita,T.,andAkira,S.(2008).Length-dependentrecognitionofdouble-strandedribonucleicacidsbyretinoicacid-induciblegene-Iandmelanomatiation-associatedgene5.JournalofExperimentalMedicine,205(7):1601{1610.Katoh,T.andSuzuki,T.(2007).SpresiduesateverythirdpositionofsiRNAshapeitstRNAiactivity.NucleicAcidsResearch,35(4):e27.Kawamata,T.,Seitz,H.,andTomari,Y.(2009).StructuraldeterminantsofmiRNAsforRISCloadingandslicer-independentunwinding.Nature,16(9):953{960.Kawamata,T.,Yoda,M.,andTomari,Y.(2011).MultilayercheckpointsformicroRNAauthenticityduringRISCassembly.EMBOReports,12(9):944{949.Kenski,D.M.,Butora,G.,Willingham,A.T.,Cooper,A.J.,Fu,W.,Qi,N.,Soriano,F.,Davies,I.W.,andFlanagan,W.M.(2012).siRNA-optimizedmosforenhancedinvivoactivity.MolecularTherapy-NucleicAcids,1:e5.Khvorova,A.,Reynolds,A.,andJayasena,S.D.(2003).FunctionalsiRNAsandmiRNAsexhibitstrandbias.Cell,115(2):209{216.Kim,Y.,Yeo,J.,Lee,J.H.,Cho,J.,Seo,D.,Kim,J.S.,andKim,V.N.(2014).Deletionofhumantarbp2revealscellularMicroRNAtargetsandcell-cyclefunctionofTRBP.CellReports,9(3):1061{1074.Kini,H.andWalton,S.P.(2007).Invitrobindingofsingle-strandedRNAbyhumanDicer.FEBSLetters,581(29):5611{6.Kini,H.andWalton,S.P.(2009).ofsiRNAterminalmismatchesonTRBPandDicerbindingandsilencing.FEBSJournal,276(22):6576{85.Kleinman,M.E.,Yamada,K.,Takeda,A.,Chandrasekaran,V.,Nozaki,M.,J.Z.,Albuquerque,R.J.C.,Yamasaki,S.,Itaya,M.,Pan,Y.,Appukuttan,B.,Gibbs,D.,Yang,Z.,o,K.,Ambati,B.K.,Wilgus,T.A.,DiPietro,L.A.,Sakurai,E.,Zhang,K.,Smith,J.R.,Taylor,E.W.,andAmbati,J.(2008).Sequence-andtarget-independentangiogenesissuppressionbysiRNAviaTLR3.Nature,452(7187):591{597.Knusel,B.,Winslow,J.W.,Rosenthal,A.,Burton,L.E.,Seid,D.P.,Nikolics,K.,andHefti,F.(1991).Promotionofcentralcholinergicanddopaminergicneurontiationbybrain-derivedneurotrophicfactorbutnotneurotrophin3.ProceedingsoftheNationalAcademyofSciences,88(3):961{5.122oberle,C.,Kaufmann,S.H.E.,andPatzel,V.(2006).SelectingesiRNAsbasedonguideRNAstructure.NatureProtocols,1(4):1832{1839.Kodym,R.,Kodym,E.,andStory,M.D.(2009).2'-5'-Oligoadenylatesynthetaseisacti-vatedbyaspRNAsequencemotif.BiochemicalandBiophysicalResearchCommu-nications,388(2):317{322.Koh,H.R.,Kidwell,M.A.,Ragunathan,K.,Doudna,J.A.,andMyong,S.(2013).ATP-independentofdouble-strandedRNAbindingproteins.ProceedingsoftheNationalAcademyofSciences,110(1):151{156.Kok,K.,Ng,M.,andChing,Y.(2007).HumanTRBPandPACTdirectlyinteractwitheachotherandassociatewithdicertofacilitatetheproductionofsmallinterferingRNA.JournalofBiologicalChemistry.Kok,K.H.,Lui,P.Y.,Ng,M.H.J.,Siu,K.L.,Au,S.W.N.,andJin,D.Y.(2011).Thedouble-strandedRNA-bindingproteinPACTfunctionsasacellularactivatorofRIG-Itofacilitateinnateantiviralresponse.CellHostMicrobe,9(4):299{309.Koller,E.,Propp,S.,Murray,H.,Lima,W.,Bhat,B.,Prakash,T.P.,Allerson,C.R.,Swayze,E.E.,Marcusson,E.G.,andDean,N.M.(2006).CompetitionforRISCbindingpredictsinvitropotencyofsiRNA.NucleicAcidsResearch,34(16):4467{4476.Kozomara,A.andS.(2011).miRBase:IntegratingmicroRNAannotationanddeep-sequencingdata.NucleicAcidsResearch,39(Databaseissue):D152{7.Kramer,M.F.(2001).Stem-loopRT-qPCRformiRNAs.InCurrentProtocolsinMolecularBiology,volume95,pages15.10:15.10.1{15.10.15.JohnWiley&Sons,Inc.Kwak,P.B.andTomari,Y.(2012).TheNdomainofArgonautedrivesduplexunwindingduringRISCassembly.Nature,19(2):145{151.Ladunga,I.(2006).MorecompletegenesilencingbyfewersiRNAs:Transparentoptimizeddesignandbiophysicalsignature.NucleicAcidsResearch,35(2):433{440.Lagos-Quintana,M.(2001).IdennofnovelgenescodingforsmallexpressedRNAs.Science,294(5543):853{858.Lai,E.C.(2002).MicroRNAsarecomplementaryto3'UTRsequencemotifsthatmediatenegativepost-transcriptionalregulation.NatureGenetics,30(4):363{364.Lambert,N.J.,Gu,S.G.,andZahler,A.M.(2011).TheconformationofmicroRNAseedregionsinnativemicroRNPsisprearrangedforpresentationtomRNAtargets.NucleicAcidsResearch,39(11):4827{4835.Laraki,G.,Clerzius,G.,Daher,A.,andMelendez-Pe~na,C.(2008).Interactionsbetweenthedouble-strandedRNA-bindingproteinsTRBPandPACTtheMedipaldomainthatmediatesprotein-proteininteractions.RNABiology,29(1):254{265.123Lee,H.Y.andDoudna,J.A.(2012).TRBPaltershumanprecursormicroRNAprocessinginvitro.RNA,18(11):2012{9.Lee,H.Y.,Zhou,K.,Smith,A.M.,Noland,C.L.,andDoudna,J.A.(2013).tialrolesofhumanDicer-bindingproteinsTRBPandPACTinsmallRNAprocessing.NucleicAcidsResearch,41(13):6568{6576.Lee,J.Y.,Kim,H.,Ryu,C.H.,Kim,J.Y.,Choi,B.H.,Lim,Y.,Huh,P.W.,Kim,Y.H.,Lee,K.H.,Jun,T.Y.,Rha,H.K.,Kang,J.K.,andChoi,C.R.(2004).Merlin,atumorsuppressor,interactswithtransactivation-responsiveRNA-bindingproteinandinhibitsitsoncogenicactivity.JournalofBiologicalChemistry,279(29):30265{30273.Lee,J.Y.,Moon,H.J.,Lee,W.K.,Chun,H.J.,Han,C.W.,Jeon,Y.W.,Lim,Y.,Kim,Y.H.,Yao,T.P.,Lee,K.H.,Jun,T.Y.,Rha,H.K.,andKang,J.K.(2005).Mer-linfacilitatesubiquitinationanddegradationoftransactivation-responsiveRNA-bindingprotein.Oncogene,25(8):1143{1152.Lee,R.,Kermani,P.,Teng,K.K.,andHempstead,B.L.(2001).Regulationofcellsurvivalbysecretedproneurotrophins.Science,294(5548):1945{8.Leibrock,J.,Lottspeich,F.,Hohn,A.,Hofer,M.,Hengerer,B.,Masiakowski,P.,Thoenen,H.,andBarde,Y.(1989).Molecularcloningandexpressionofbrain-derivedneurotrophicfactor.Nature,341(6238):149{152.Leuschner,P.J.F.,Ameres,S.L.,Kueng,S.,andMartinez,J.(2006).CleavageofthesiRNApassengerstrandduringRISCassemblyinhumancells.EMBOReports,7(3):314{320.Lewis,B.P.,Shih,I.H.,Jones-Rhoades,M.W.,Bartel,D.P.,andBurge,C.B.(2003).PredictionofmammalianmicroRNAtargets.Cell,115(7):787{798.Lima,W.F.,Murray,H.,Nichols,J.G.,Wu,H.,Sun,H.,Prakash,T.P.,Berdeja,A.R.,Gaus,H.J.,andCrooke,S.T.(2009).HumanDicerbindsshortsingle-strandanddouble-strandRNAwithhighyandinteractswithtregionsofthenucleicacids.JournalofBiologicalChemistry,284(4):2535{2548.Lima,W.F.,Prakash,T.P.,Murray,H.M.,Kinberger,G.A.,Li,W.,Chappell,A.E.,Li,C.S.,Murray,S.F.,Gaus,H.,Seth,P.P.,Swayze,E.E.,andCrooke,S.T.(2012).Single-strandedsiRNAsactivateRNAiinanimals.Cell,150(5):883{894.Lin,X.,Ruan,X.,Anderson,M.G.,Mcdowell,J.A.,Kroeger,P.E.,Fesik,S.W.,andShen,Y.(2005).siRNA-mediatedtargetgenesilencingtriggeredbya7ntcomplementation.NucleicAcidsResearch,33(14):4527.Lingel,A.,Simon,B.,Izaurralde,E.,andSattler,M.(2004).Nucleicacid3'-endrecognitionbytheArgonaute2PAZdomain.NatureStructural&MolecularBiology,11(6):576{577.124Liu,J.D.,Carmell,M.A.,Rivas,F.V.,Marsden,C.G.,Thomson,J.M.,Song,J.J.,Hammond,S.M.,Joshua-Tor,L.,andHannon,G.J.(2004).Argonaute2isthecatalyticengineofmammalianRNAi.Science,305(5689):1437{1441.Liu,Y.,Ye,X.,Jiang,F.,Liang,C.,Chen,D.,Peng,J.,Kinch,L.N.,Grishin,N.V.,andLiu,Q.(2009).C3PO,anendoribonucleasethatpromotesRNAibyfacilitatingRISCactivation.Science,325(5941):750{753.Lu,B.(2003).Pro-regionofneurotrophins:Roleinsynapticmodulation.Neuron,39(5):735{8.Lu,B.,Pang,P.T.,andWoo,N.H.(2005a).Theyinandyangofneurotrophinaction.NatureReviewsNeuroscience,6(8):603{614.Lu,P.,Jones,L.,andTuszynski,M.H.(2005b).BDNF-expressingmarrowstromalcellssupportextensiveaxonalgrowthatsitesofspinalcordinjury.ExperimentalNeurology,191(2):344{360.Lu,Z.J.andMathews,D.H.(2008).tsiRNAselectionusinghybridizationther-modynamics.NucleicAcidsResearch,36(2):640{647.Lynam,D.A.,Shahriari,D.,Wolf,K.J.,Angart,P.A.,J.,Tuszynski,M.H.,Chan,C.,Walton,P.,andSakamoto,J.(2015).Brainderivedneurotrophicfactorreleasefromlayer-by-layercoatedagarosenerveguidanceActaBiomaterialia,18:128{131.Ma,E.,Macrae,I.J.,Kirsch,J.F.,andDoudna,J.A.(2008).AutoinhibitionofhumanDicerbyitsinternalhelicasedomain.JournalofMolecularBiology,380(1):237{243.Ma,J.B.,Ye,K.,andPatel,D.J.(2004).Structuralbasisforoverhang-spsmallinterferingRNArecognitionbythePAZdomain.Nature,429(6989):318{322.MacRae,I.J.,Ma,E.,Zhou,M.,Robinson,C.V.,andDoudna,J.A.(2008).InvitroreconstitutionofthehumanRISC-loadingcomplex.ProceedingsoftheNationalAcademyofSciences,105(2):512{517.Macrae,I.J.,Zhou,K.,andDoudna,J.A.(2007).Structuraldeterminantsofrnarecog-nitionandcleavagebydicer.NatureStructural&MolecularBiology,14(10):934{940.Malefyt,A.P.,Walton,S.P.,andChan,C.(2012).Endocytosispathwaysfornucleicacidtherapeutics.NanoLife,2(3):1241005.Malefyt,A.P.,Wu,M.,Vocelle,D.B.,Kappes,S.J.,Lindeman,S.D.,Chan,C.,andWalton,S.P.(2013).ImprovedasymmetrypredictionforsiRNAs.FEBSJournal,281(1):320{330.Manche,L.,Green,S.R.,Schmedt,C.,andMathews,M.B.(1992).Interactionsbetweendouble-strandedRNAregulatorsandtheproteinkinaseDAI.MolecularandCellularBiology,12(11):5238{5248.125Manoharan,M.,Akinc,A.,Pandey,R.K.,Qin,J.,Hadwiger,P.,John,M.,Mills,K.,Charisse,K.,Maier,M.A.,Nechev,L.,Greene,E.M.,Pallan,P.S.,Rozners,E.,Rajeev,K.G.,andEgli,M.(2011).Uniquegene-silencingandstructuralpropertiesofsiRNAs.AngewandteChemie,50(10):2284{2288.Markham,N.R.andZuker,M.(2005).DINAMeltwebserverfornucleicacidmeltingprediction.NucleicAcidsResearch,33(WebServerissue):W577{81.Markham,N.R.andZuker,M.(2008).UNAFold:Softwarefornucleicacidfoldingandhybridization.InKeith,J.M.,editor,Bioinformatics,VolumeII:Structure,Function,andApplications,volume453ofMethodsinMolecularBiology,pages3{31.HumanaPress,Totowa,NJ.Marques,J.T.,Devosse,T.,Wang,D.,Zamanian-Daryoush,M.,Serbinowski,P.,Hart-mann,R.,Fujita,T.,Behlke,M.A.,andWilliams,B.R.(2006).Astructuralbasisfordiscriminatingbetweenselfandnonselfdouble-strandedRNAsinmammaliancells.NatureBiotechnology,24(5):559{565.Martinez,J.,Patkaniowska,A.,Urlaub,H.,Luhrmann,R.,andTuschl,T.(2002).Single-strandedantisensesiRNAsguidetargetRNAcleavageinRNAi.Cell,110(5):563{574.Matranga,C.,Tomari,Y.,Shin,C.,Bartel,D.,andZamore,P.(2005).Passenger-strandcleavagefacilitatesassemblyofsiRNAintoAgo2-containingRNAienzymecomplexes.Cell,123(4):607{620.McCarty,J.H.andFeinstein,S.C.(1998).ActivationlooptyrosinescontributevaryingrolestoTrkBautophosphorylationandsignaltransduction.Oncogene,16(13):1691{1700.Meijering,E.H.,Niessen,W.J.,andViergever,M.A.(2001).Quantitativeevaluationofconvolution-basedmethodsformedicalimageinterpolation.MedicalImageAnalysis,5(2):111{126.Meisenheimer,K.M.andKoch,T.H.(1997).Photocross-linkingofnucleicacidstoasso-ciatedproteins.CriticalReviewsinBiochemistryandMolecularBiology,32(2):101{140.Meister,G.,Landthaler,M.,Patkaniowska,A.,Dorsett,Y.,Teng,G.,andTuschl,T.(2004).HumanArgonaute2mediatesRNAcleavagetargetedbymiRNAsandsiRNAs.MolecularCell,15(2):185{197.Menei,P.,Montero-Menei,C.,Whittemore,S.R.,Bunge,R.P.,andBunge,M.B.(1998).SchwanncellsgeneticallymotosecretehumanBDNFpromoteenhancedaxonalregrowthacrosstransectedadultratspinalcord.EuropeanJournalofNeuroscience,10(2):607{621.Meyer,M.,Matsuoka,I.,Wetmore,C.,Olson,L.,andThoenen,H.(1992).Enhancedsynthesisofbrain-derivedneurotrophicfactorinthelesionedperipheralnerve:tmechanismsareresponsiblefortheregulationofBDNFandNGFmRNA.JournalofCellBiology,119(1):45{54.126Meyer,S.L.,Lang,D.M.,Forbes,M.E.,Knight,Jr,E.,Hirsch,J.D.,Trusko,S.P.,andScott,R.W.(1994).Productionandcharacterizationofrecombinantmousebrain-derivedneurotrophicfactorandratneurotrophin-3expressedininsectcells.JournalofNeurochemistry,62(3):825{833.Miyauchi,A.,Ozawa,M.,Mizukami,M.,Yashiro,K.,Ebisu,S.,Tojo,T.,Fujii,T.,andTakagi,H.(1999).Structuralconversionfromnon-nativetonativeformofrecombinanthumanepidermalgrowthfactorbyBrevibacilluschoshinensis.Bioscience,Biotechnol-ogy,andBiochemistry,63(11):1965{1969.Mizukami,M.,Hanagata,H.,andMiyauchi,A.(2010).Brevibacillusexpressionsystem:Host-vectorsystemfortproductionofsecretoryproteins.CurrentPharmaceuticalBiotechnology,11(3):251{8.Mowla,S.J.,Farhadi,H.F.,Pareek,S.,Atwal,J.K.,Morris,S.J.,Seidah,N.G.,andMurphy,R.A.(2001).Biosynthesisandpost-translationalprocessingoftheprecursortobrain-derivedneurotrophicfactor.TheJournalofBiologicalChemistry,276(16):12660{12666.Nagahara,A.H.andTuszynski,M.H.(2011).PotentialtherapeuticusesofBDNFinneurologicalandpsychiatricdisorders.NatureReviewsDrugDiscovery,10(3):209{219.Naito,Y.andUi-Tei,K.(2012).siRNAdesignsoftwareforatargetgene-spRNAinterference.FrontiersinGenetics,3:102.Napoli,C.andLemieux,C.(1990).Introductionofachimericchalconesynthasegeneintopetuniaresultsinreversibleco-suppressionofhomologousgenesintrans.ThePlantCell,2:279{289.Navarro,F.,Gutman,D.,Meire,E.,Caceres,M.,Rigoutsos,I.,Bentwich,Z.,andLieber-man,J.(2009).miR-34acontributestomegakaryocytictiationofK562cellsindependentlyofp53.Blood,114(10):2181{2192.Noland,C.L.andDoudna,J.A.(2013).MultiplesensorsensureguidestrandselectioninhumanRNAipathways.RNA,19(5):639{648.Noland,C.L.,Ma,E.,andDoudna,J.A.(2011).siRNArepositioningforguidestrandselectionbyhumanDicercomplexes.MolecularCell,43(1):110{121.Numakawa,T.,Suzuki,S.,Kumamaru,E.,Adachi,N.,Richards,M.,andKunugi,H.(2010).BDNFfunctionandintracellularsignalinginneurons.Histologyandhistopathol-ogy,25(2):237{258.anen,A.,Haley,B.,andZamore,P.D.(2001).ATPrequirementsandsmallinterferingRNAstructureintheRNAinterferencepathway.Cell,107(3):309{321.127Onishi,H.,Mizukami,M.,Hanagata,H.,Tokunaga,M.,Arakawa,T.,andMiyauchi,A.(2013).tproductionofanandanti-lysozymeassingle-chainan-tibodyfragments(scfv)byBrevibacillusexpressionsystem.ProteinExpressionandation,91(2):184{91.OvM.,Alken,M.,Far,R.K.-K.,Lemaitre,M.,Lebleu,B.,Sczakiel,G.,andRobbins,I.(2005).LocalRNAtargetstructuresirnaAsystematicglobalanalysis.JournalofMolecularBiology,348(4):871{881.Park,K.andShimizu,N.(1996).Expressionofbrain-derivedneurotrophicfactorbyBacillussubtilis.JournalofFermentationandBioengineering,82(6):585{588.Parker,J.S.,Parizotto,E.A.,Wang,M.,Roe,S.M.,andBarford,D.(2009).Enhancementoftheseed-targetrecognitionstepinRNAsilencingbyaPIWI/MIDdomainprotein.MolecularCell,33(2):204{214.Paroo,Z.,Ye,X.,Chen,S.,andLiu,Q.(2009).PhosphorylationofthehumanmicroRNA-generatingcomplexmediatesMAPK/Erksignaling.Cell,139(1):112{122.Patzel,V.,Rutz,S.,Dietrich,I.,oberle,C.,ScA.,andKaufmann,S.H.E.(2005).DesignofsiRNAsproducingunstructuredguide-RNAsresultsinimprovedRNAinterference.NatureBiotechnology,23(11):1440{1444.Pellino,J.L.(2007).Dicer'sroleinRISCassemblyduringRNAsilencing.PhDthesis,NorthwesternUniversity.Rand,T.A.,Petersen,S.,Du,F.,andWang,X.(2005).Argonaute2cleavestheanti-guidestrandofsiRNAduringRISCactivation.Cell,123(4):621{629.Reynolds,A.,Anderson,E.M.,Vermeulen,A.,Fedorov,Y.,Robinson,K.,Leake,D.,Karpilow,J.,Marshall,W.S.,andKhvorova,A.(2006).InductionoftheinterferonresponsebysiRNAiscelltype-andduplexlength-dependent.RNA,12(6):988{993.Reynolds,A.,Leake,D.,Boese,Q.,Scaringe,S.,Marshall,W.S.,andKhvorova,A.(2004).Rationalsirnadesignforrnainterference.NatureBiotechnology,22(3):326{330.Rivas,F.V.,Tolia,N.H.,Song,J.J.,Aragon,J.P.,Liu,J.D.,Hannon,G.J.,andJoshua-Tor,L.(2005).Argonaute2andansiRNAformrecombinanthumanRISC.NatureStructural&MolecularBiology,12(4):340{349.Robbins,M.,Judge,A.,andLiang,L.(2007).20-O-methyl-moRNAsactasTLR7antagonists.MolecularTherapy,15(9):1663{9.Rosenfeld,R.andBenedek,K.(1993).Conformationalchangesofbrain-derivedneu-rotrophicfactorduringreversed-phasehigh-performanceliquidchromatography.JournalofChromatographyA,632(1-2):29{36.128Rosenthal,A.,Goeddel,D.V.,Nguyen,T.,Martin,E.,Burton,L.E.,Shih,A.,Laramee,G.R.,Wurm,F.,Mason,A.,Nikolics,K.,andWinslow,J.W.(1991).Primarystruc-tureandbiologicalactivityofhumanbrain-derivedneurotrophicfactor.Endocrinology,129(3):1289{94.Sakurai,K.,Amarzguioui,M.,Kim,D.,Alluin,J.,Heale,B.,Song,M.,Gatignol,A.,Behlke,M.A.,andRossi,J.J.(2011).AroleforhumanDicerinpre-RISCloadingofsiRNAs.NucleicAcidsResearch,39(4):1510{1525.Salomon,W.E.,Jolly,S.M.,Moore,M.J.,Zamore,P.D.,andSerebrov,V.(2015).Single-moleculeimagingrevealsthatArgonautereshapesthebindingpropertiesofitsnucleicacidguides.Cell,162(1):84{95.Samuel-Abraham,S.andLeonard,J.N.(2010).Stayingonmessage:DesignprinciplesforcontrollingnonspresponsestosiRNA.TheFEBSJournal,277(23):4828{4836.Schellman,J.A.(1997).Temperature,stability,andthehydrophobicinteraction.Biophys-icalJournal,73(6):2960{2964.Schirle,N.T.andMacRae,I.J.(2012).ThecrystalstructureofhumanArgonaute2.Science,336(6084):1037{40.Schlee,M.,Hornug,V.,andHartmann,G.(2006).siRNAandisRNA:Twoedgesofonesword.MolecularTherapy,14(4):463{470.Schubert,S.,Grunweller,A.,Erdmann,V.A.,andKurreck,J.(2005).LocalRNAtargetstructuresiRNASystematicanalysisofintentionallydesignedbindingregions.JournalofMolecularBiology,348(4):883{893.Schultz,N.,Marenstein,D.R.,DeAngelis,D.A.,Wang,W.Q.,Nelander,S.,Jacobsen,A.,Marks,D.S.,Massague,J.,andSander,C.(2011).dominatealarge-scaleRNAiscreenformodulatorsoftheTGF-betapathwayandrevealmicroRNAregulationofTGFBR2.Silence,2:3.Schwarz,D.,Hutvagner,G.,Du,T.,Xu,Z.,Aronin,N.,andZamore,P.(2003).AsymmetryintheassemblyoftheRNAienzymecomplex.Cell,115(2):199{208.Sciabola,S.,Cao,Q.,Orozco,M.,Faustino,I.,andStanton,R.V.(2012).ImprovednucleicaciddescriptorsforsiRNAprediction.NucleicAcidsResearch,41(3):1383{94.Seitz,H.,Tushir,J.S.,andZamore,P.D.(2011).A5'-uridinemiRNA/miRNA*asymmetryinDrosophilabypromotingRNA-inducedsilencingcomplexformation.Si-lence,2:4.Shabalina,S.A.,Spiridonov,A.N.,andOgurtsov,A.Y.(2006).ComputationalmodelswiththermodynamicandcompositionfeaturesimprovesiRNAdesign.BMCBioinfor-matics,7:65.129Shao,Y.,Chan,C.Y.,Maliyekkel,A.,Lawrence,C.E.,Roninson,I.B.,andDing,Y.(2007).oftargetsecondarystructureonRNAi.RNA,13(10):1631{1640.Shen,M.,Xu,Y.,Jia,R.,Zhou,X.,andYe,K.(2010).Size-independentandnoncoop-erativerecognitionofdsRNAbytheRicestripevirusRNAsilencingsuppressorNS3.JournalofMolecularBiology,404(4):665{679.Sledz,C.A.,Holko,M.,deVeer,M.J.,Silverman,R.H.,andWilliams,B.R.G.(2003).Activationoftheinterferonsystembyshort-interferingRNAs.NatureCellBiology,5(9):834{839.Snead,N.M.andRossi,J.J.(2010).BiogenesisandfunctionofendogenousandexogenoussiRNAs.WileyInterdisciplinaryReviews:RNA,1(1):117{131.Snead,N.M.andRossi,J.J.(2012).RNAinterferencetriggervariants:GettingthemostoutofRNAforRNAinterference-basedtherapeutics.NucleicAcidTherapeutics,22(3):139{146.Sontheimer,E.J.(1994).Site-spRNAcrosslinkingwith4-thiouridine.MolecularBiologyReports,20(1):35{44.Soppet,D.,Escandon,E.,Maragos,J.,Middlemas,D.S.,Reid,S.W.,Blair,J.,Bur-ton,L.E.,Stanton,B.R.,Kaplan,D.R.,Hunter,T.,Nikolics,K.,andParada,L.F.(1991).Theneurotrophicfactorsbrain-derivedneurotrophicfactorandneurotrophin-3areligandsforthetrkBtyrosinekinasereceptor.Cell,65(5):895{903.Stalder,L.,Heusermann,W.,Sokol,L.,Trojer,D.,Wirz,J.,Hean,J.,Fritzsche,A.,Aeschimann,F.,Pfanzagl,V.,Basselet,P.,Weiler,J.,Hintersteiner,M.,Morrissey,D.V.,andMeisner-Kober,N.C.(2013).TheroughendoplasmaticreticulumisacentralnucleationsiteofsiRNA-mediatedRNAsilencing.EMBOJournal,32(8):1115{1127.Su,H.,Trombly,M.I.,Chen,J.,andWang,X.(2009).EssentialandoverlappingfunctionsformammalianArgonautesinmicroRNAsilencing.Genes&Development,23(3):304{317.Sun,P.D.andDavies,D.R.(1995).Thecystine-knotgrowth-factorsuperfamily.AnnualReviewsofBiophysicsandBiomolecularStructure,24:269{291.Sun,X.,H.A.,andLi,C.J.(2008).AsymmetricRNAduplexesmediateRNAinterferenceinmammaliancells.NatureBiotechnology,26(12):1379{1382.Suzuki,H.I.,Katsura,A.,Yasuda,T.,Ueno,T.,Mano,H.,Sugimoto,K.,andMiyazono,K.(2015).Small-RNAasymmetryisdirectlydrivenbymammalianArgonautes.NatureStructural&MolecularBiology,22(7):512{521.Takagi,H.,Kadowaki,K.,andUdaka,S.(1989a).Screeningandcharacterizationofprotein-hyperproducingbacteriawithoutdetectableexoproteaseactivity.AgriculturalandBio-logicalChemistry,53(3):691{699.130Takagi,H.,Miyauchi,A.,Kadowaki,K.,andUdaka,S.(1989b).PotentialuseofBacillusbrevisHPD31fortheproductionofforeignproteins.AgriculturalandBiologicalChem-istry,53(8):2279{2280.Takahashi,T.,Miyakawa,T.,Zenno,S.,Nishi,K.,Tanokura,M.,andUi-Tei,K.(2013).DistinguishableinvitrobindingmodeofmonomericTRBPanddimericPACTwithsiRNA.PLoSONE,8(5):e63434.Takahashi,T.,Zenno,S.,Ishibashi,O.,Takizawa,T.,Saigo,K.,andUi-Tei,K.(2014).In-teractionsbetweenthenon-seedregionofsiRNAandRNA-bindingRLC/RISCproteins,AgoandTRBP,inmammaliancells.NucleicAcidsResearch,42(8):5256{5269.Takasaki,S.(2009).SelectingesiRNAtargetsequencesbyusingBayes'theorem.ComputationalBiologyandChemistry,33(5):368{372.Takasaki,S.,Kotani,S.,andKonagaya,A.(2004).AnemethodforselectingsiRNAtargetsequencesinmammaliancells.CellCycle,3(6):788{793.Takeshita,T.,Fujimori,K.,andShimizu,N.(1996).Synthesisofbiologicallyactivehumannervegrowthfactorandbrain-derivedneurotrophicfactorbyacell-freesystem.JournalofFermentationandBioengineering,81(1):13{17.Tanaka,K.andKumano,M.(2000).StablepharmaceuticalcompositionofBDNF.USPatent6,077,829.Tanudji,M.,Machalek,D.,Arndt,G.M.,andRivory,L.(2010).Competitionbe-tweensiRNAduplexes:ImpactofRNA-inducedsilencingcomplexloadingandcomparisonbetweenconventional-21bpandDicer-substratesiRNAs.Oligonucleotides,20(1):27{32.Teng,H.K.,Teng,K.K.,Lee,R.,Wright,S.,Tevar,S.,Almeida,R.D.,Kermani,P.,Torkin,R.,Chen,Z.Y.,Lee,F.S.,Kraemer,R.T.,Nykjaer,A.,andHempstead,B.L.(2005).ProBDNFinducesneuronalapoptosisviaactivationofareceptorcomplexofp75NTRandsortilin.JournalofNeuroscience,25(22):5455{63.Tluk,S.,Jurk,M.,Forsbach,A.,Weeratna,R.,Samulowitz,U.,Krieg,A.M.,Bauer,S.,andVollmer,J.(2009).Sequencesderivedfromself-RNAcontainingcertainnatu-ralmoactassuppressorsofRNA-mediatedimmuneresponses.InternationalImmunology,21(5):607{619.Tomari,Y.,Matranga,C.,Haley,B.,Martinez,N.,andZamore,P.D.(2004).AproteinsensorforsiRNAasymmetry.Science,306(5700):1377{1380.Turner,J.J.,Jones,S.W.,Moschos,S.A.,Lindsay,M.A.,andGait,M.J.(2006).MALDI-TOFmassspectralanalysisofsiRNAdegradationinserummsanRNAseA-likeactivity.MolecularBioSystems,3(1):43{50.131Ui-Tei,K.,Naito,Y.,Takahashi,F.,Haraguchi,T.,Ohki-Hamazaki,H.,Juni,A.,Ueda,R.,andSaigo,K.(2004).GuidelinesfortheselectionofhighlyesiRNAsequencesformammalianandchickRNAinterference.NucleicAcidsResearch,32(3):936{948.Varkonyi-Gasic,E.andHellens,R.P.(2011).Quantitativestem-loopRT-PCRfordetectionofmicroRNAs.InKodama,H.andKomamine,A.,editors,RNAiandPlantGeneFunctionAnalysis:MethodsandProtocols,volume744ofMethodsinMolecularBiology,pages145{157.Vasina,J.A.andBaneyx,F.(1997).Expressionofaggregation-pronerecombinantproteinsatlowtemperatures:AcomparativestudyoftheEscherichiacolicspAandtacpromotersystems.ProteinExpressionandation,9(2):211{8.Vermeulen,A.,Behlen,L.,Reynolds,A.,Wolfson,A.,Marshall,W.S.,Karpilow,J.,andKhvorova,A.(2005).ThecontributionsofdsRNAstructuretoDicerspyand.RNA,11(5):674{682.Vert,J.P.,Foveau,N.,Lajaunie,C.,andVandenbrouck,Y.(2006).AnaccurateandinterpretablemodelforsiRNAprediction.BMCBioinformatics,7(520).Vickers,T.A.,Koo,S.,Bennett,C.F.,Crooke,S.T.,Dean,N.M.,andBaker,B.F.(2003).treductionoftargetRNAsbysmallinterferingRNAandRNaseH-dependentantisenseagents.Acomparativeanalysis.JournalofBiologicalChemistry,278(9):7108{7118.Volkov,A.A.,Kruglova,N.S.,Meschaninova,M.I.,Venyaminova,A.G.,Zenkova,M.A.,Vlassov,V.V.,andChernolovskaya,E.L.(2009).Selectiveprotectionofnuclease-sensitivesitesinsiRNAprolongssilencingOligonucleotides,19(2):191{202.Walton,S.P.,Wu,M.,Gredell,J.A.,andChan,C.(2010).DesigninghighlyactivesiRNAsfortherapeuticapplications.TheFEBSjournal,277(23):4806{4813.Wang,B.,Li,S.,Qi,H.,andChowdhury,D.(2009).DistinctpassengerstrandandmRNAcleavageactivitiesofhumanArgonauteproteins.NatureStructural&MolecularBiology,16(12):1259{66.Wang,Y.,Sheng,G.,Juranek,S.,Tuschl,T.,andPatel,D.J.(2008).Structureoftheguide-strand-containingArgonautesilencingcomplex.Nature,456(7219):209{213.Weber,C.,Muller,C.,Podszuweit,A.,Montino,C.,Vollmer,J.,andForsbach,A.(2012).Toll-likereceptor(TLR)3immunemodulationbyunformulatedsmallinterferingRNAorDNAandtheroleofCD14(inTLR-mediatedImmunology,136(1):64{77.Wee,L.M.,Flores-Jasso,C.F.,Salomon,W.E.,andZamore,P.D.(2013).ArgonautedividesitsRNAguideintodomainswithdistinctfunctionsandRNA-bindingproperties.Cell,152(1-2):1055{1067.Weitzer,S.andMartinez,J.(2007).ThehumanRNAkinasehClp1isactiveon3'transferRNAexonsandshortinterferingRNAs.Nature,447(7141):222{6.132Willkomm,S.,Deerberg,A.,Heidemann,J.,Flugge,F.,Meine,J.,Hu,R.,Kretschmer-KazemiFar,R.,andRestle,T.(2016).RecombinanthTRBPandhPACTmod-ulatehAgo2-catalyzedsiRNA-mediatedtargetRNAcleavageinvitro.PloSOne,11(1):e0146814{e0146814.Wilson,R.C.,Tambe,A.,Kidwell,M.A.,Noland,C.L.,Schneider,C.P.,andDoudna,J.A.(2015).Dicer-TRBPcomplexformationensuresaccuratemammalianmicroRNAbiogenesis.MolecularCell,57(3):397{407.Wittrup,A.andLieberman,J.(2015).Knockingdowndisease:AprogressreportonsiRNAtherapeutics.NatureReviewsGenetics,16(9):543{552.Yamagata,H.,Adachi,T.,Tsuboi,A.,Takao,M.,Sasaki,T.,Tsukagoshi,N.,andUdaka,S.(1987).Cloningandcharacterizationofthe5'regionofthecellwallproteingeneoperoninBacillusbrevis47.JournalofBacteriology,169(3):1239{1245.Yang,J.,Maurin,T.,Robine,N.,Rasmussen,K.D.,,K.L.,Chandwani,R.,Papa-petrou,E.P.,Sadelain,M.,O'Carroll,D.,andLai,E.C.(2010).Conservedvertebratemir-451providesaplatformforDicer-independent,Ago2-mediatedmicroRNAbiogene-sis.ProceedingsoftheNationalAcademyofSciences,107(34):15163{15168.Yang,X.,Sierant,M.,Janicka,M.,Peczek,L.,Martinez,C.,Hassell,T.,Li,N.,Li,X.,Wang,T.,andNawrot,B.(2012).GenesilencingactivityofsiRNAmoleculescontainingphosphorodithioatesubstitutions.ACSChemicalBiology,7(7):1214{1220.Ye,X.,Huang,N.,Liu,Y.,Paroo,Z.,Huerta,C.,Li,P.,Chen,S.,Liu,Q.,andZhang,H.(2011).StructureofC3POandmechanismofhumanRISCactivation.NatureStructural&MolecularBiology,18(6):650{657.Yi,R.,Qin,Y.,Macara,I.,andCullen,B.(2003).Exportin-5mediatesthenuclearexportofpre-microRNAsandshorthairpinRNAs.Genes&Development,17(24):3011{3016.Yoda,M.,Kawamata,T.,Paroo,Z.,Ye,X.,Iwasaki,S.,Liu,Q.,andTomari,Y.(2010).ATP-dependenthumanRISCassemblypathways.NatureStructural&MolecularBiol-ogy,17(1):17{23.Yoshinari,K.,Miyagishi,M.,andTaira,K.(2004).onRNAiofthetightstructure,sequenceandpositionofthetargetedregion.NucleicAcidsResearch,32(2):691{699.Zamore,P.D.,Tuschl,T.,Sharp,P.A.,andBartel,D.P.(2000).RNAi:Double-strandedRNAdirectstheATP-dependentcleavageofmRNAat21to23nucleotideintervals.Cell,101(1):25{33.Zuccato,C.andCattaneo,E.(2009).Brain-derivedneurotrophicfactorinneurodegenera-tivediseases.NatureReviewsNeurology,5(6):311{22.133