bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
Live-cellmappingoforganelle-associatedRNAsviaproximitybiotinylation
combinedwithprotein-RNAcrosslinking
Authors/Affiliations
PornchaiKaewsapsak1,5,DavidM.Shechner2-3,5,WilliamMallard2-3,JohnL.Rinn2-3,
andAliceY.Ting1,4,6*
1DepartmentofChemistry,MassachusettsInstituteofTechnology,Cambridge,
Massachusetts,USA;
2DepartmentofStemCellandRegenerativeBiology,andMolecularandCellular
Biology,HarvardUniversity,Cambridge,Massachusetts,USA
3BroadInstituteofMassachusettsInstituteofTechnologyandHarvard,Cambridge,
Massachusetts,USA
4DepartmentsofGenetics,Biology,andChemistry,StanfordUniversity,Stanford,CA,
USA
5Co-firstauthors
6LeadContact
*Correspondence:[email protected]
Keywords
RNAlocalization,SubcellularTranscriptomics,SequencingTechnologies,
Peroxidase,APEX2,HorseradishPeroxidase,FISH,RNA-Seq,MammalianCellRNA
Analysis
Abstract
ThespatialorganizationofRNAwithincellsisacrucialfactorinawiderangeof
biologicalfunctions,spanningallkingdomsoflife.However,ageneral
understandingofRNAlocalizationhasbeenhinderedbyalackofsimple,highthroughputmethodsformappingthetranscriptomesofsubcellularcompartments.
Here,wedevelopsuchamethod,termedAPEX-RIP,whichcombinesperoxidasecatalyzed,spatiallyrestrictedinsituproteinbiotinylationwithRNA-protein
chemicalcrosslinking.Wedemonstratethat,usingasingleprotocol,APEX-RIPcan
isolateRNAsfromavarietyofsubcellularcompartments,includingthe
mitochondrialmatrix,nucleus,bulkcytosol,andendoplasmicreticulum(ER),with
higherspecificityandcoveragethandoconventionalapproaches.Wefurthermore
identifycandidateRNAslocalizedtomitochondria-ERjunctionsandnuclearlamina,
twocompartmentsthatarerecalcitranttoclassicalbiochemicalpurification.Since
APEX-RIPissimple,versatile,anddoesnotrequirespecialinstrumentation,we
envisionitsbroadapplicationinavarietyofbiologicalcontexts.
Introduction
SpatialcompartmentalizationofRNAiscentraltomanybiologicalprocessesacross
allkingdomsoflife,andenablesdiverseregulatoryschemesthatexploitbothcoding
andnoncodingfunctionsofthetranscriptome.Forexample,thelocalizationand
spatiallyrestrictedtranslationofmRNAplaysafundamentalroleinasymmetriccell
divisioninbacteriaandyeast,body-patternformationinDrosophilaandXenopus,
signalingatmammalianneuronalsynapses(Jungetal.2014),andawidevarietyof
1
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
otherbiologicalcontexts.Inanotherexample,thelocalizationofnoncodingRNAs
(ncRNAs)canplayanarchitecturalroleintheassemblyofsubcellularstructures,
includingshort-rangechromatinloops,higher-orderchromatindomains,andlarge
sub-nuclearstructureslikenucleoliandBarrbodies(RinnandGuttman2014;
Engreitz,Ollikainen,andGuttman2016).However,despitetheseexamples,our
generalunderstandingofthebreadthandbiologicalsignificanceofRNAsubcellular
localizationremainsinchoate.
TechniquesthatelucidatethesubcellularlocalizationofRNAsaretherefore
criticalforadvancingourunderstandingofRNAbiology.Classically,suchtechniques
relyeitheronimagingorbiochemicalapproaches.Imagingmethods–suchas
FluorescenceInSituHybridization(FISH)andartificialRNAreporterschemes–are
powerfultoolsforelucidatingthepositionsofasmallnumberoftargetRNAsatlowto-moderatethroughput(Wilketal.2016;K.H.Chenetal.2015;Paige,Wu,and
Jaffrey2011;Hocineetal.2013;Nellesetal.2016).Alternatively,unbiased
approachesforRNAdiscoverycouplebiochemicalmanipulationstodeep
sequencing.Forexample,theRNApartnersofproteinswithcharacteristic
subcellularlocalizationcanbeidentifiedthroughavarietyoftechniquesthatcouple
proteinimmunoprecipitationtoRNA-seq(Uleetal.2003;ChristopherGilbertetal.
2004).SuchmethodshaverevealedmanylocalizedmRNAs,inadditiontonovel
non-codingRNAsinvolvedinRNAsplicing(Chietal.2009)andRNAi(Motamediet
al.2004).Onabroaderscale,adeepsamplingofRNAsresidingwithinacellular
compartment—forexample,anintactorganelleofinterest,orpartitionsalonga
sucrosegradient,canbeidentifiedbycouplingsubcellularfractionationtoRNA-Seq
(“Fractionation-Seq”)(Sterne-Weileretal.2013;Merceretal.2011).
However,atechnologicalgapexistsamongthesecurrentmethodsfor
studyingRNAlocalization.Imagingapproachesareoflimitedthroughput,andmay
requirespecializedreagents,constructs,ormicroscopesthatareonlyaccessibletoa
handfuloflaboratories(Wilketal.2016;K.H.Chenetal.2015;Paige,Wu,and
Jaffrey2011;Hocineetal.2013;Nellesetal.2016).Theefficacyof
immunoprecipitation-basedapproachesishighlysensitivetotheantibodiesand
enrichmentprotocolsused(Hendricksonetal.2016)andcapturesonlyRNAsthat
aredirectlycomplexedwitheachtargetprotein.Fractionation-Seqisapplicableonly
toorganellesandsubcellularfractionsthatcanbepurified,andisfrequently
complicatedbycontaminants(falsepositives)andlossofmaterial(falsenegatives)
Therefore,anewtechnologyisneededforunbiasedandlarge-scalediscoveryand
characterizationofRNAneighborhoods,withhighspatialspecificity,andwithin
cellularstructuresthatcannotbeenrichedbybiochemicalfractionation.
Hereweintroducesuchatechnology–termedAPEX-RIP–thatenables
unbiaseddiscoveryofendogenousRNAsinspecificcellularlocales.APEX-RIP
mergestwoexistingtechnologies:APEX(engineeredascorbateperoxidase)catalyzedproximitybiotinylationofendogenousproteins(Rheeetal.2013),and
RNAImmunoPrecipitation(RIP)(ChristopherGilbertetal.2004).Wedemonstrate
thatAPEX-RIPisabletoenrichendogenousRNAsinmembrane-enclosedcellular
organelles,suchasthemitochondrionandnucleus,andinmembrane-abutting
cellularregionssuchasthecytosolicfaceoftheendoplasmicreticulum.The
specificityandcoverageofthisapproacharemuchhigherthanthoseobtainedby
2
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
traditionalFractionation-Seq.Moreover,byapplyingAPEX-RIPtomultiple
mammalianorganelles,wehavegeneratedhighqualitydatasetsof
compartmentalizedRNAsthatshouldserveasvaluableresourcesfortestingand
generatingnovelhypothesespertinenttoRNAbiology.
DevelopmentofAPEX-RIPmethodandapplicationtomitochondria
APEXisanengineeredperoxidasethatcanbetargetedbygeneticfusiontovarious
subcellularregionsofinterest(Rheeetal.2013)(Figure1A).Uponadditionofits
substrates,biotin-phenol(BP)andhydrogenperoxide(H2O2),tolivecells,APEX
catalyzestheformationofbiotin-phenoxylradicalsthatthendiffuseoutwardand
covalentlybiotinylatenearbyendogenousproteins.Moredistalproteinsarenot
significantlylabeledbecausethebiotin-phenoxylradicalhasahalf-lifeoflessthan1
millisecond(WishartandMadhavaRao2010).PreviousworkhasshownthatAPEXcatalyzedproximitybiotinylation,coupledtostreptavidinenrichmentandmass
spectrometry,cangenerateproteomicmapsofthemitochondrialmatrix,
intermembranespace,outermembrane,andnucleoid,eachwith<5nmspatial
specificity(Rheeetal.2013;Hungetal.2014;Hungetal.2017;Hanetal.2017).
BecausemostcellularRNAsexistincloseproximitytoproteins,wereasoned
thatAPEX-taggedsubcellularproteomescouldalsoprovideaccesstothenearby
RNAcontent,ifproteinsandRNAcouldbecrosslinkedtogetherinsitu,immediately
beforeorafterAPEXlabeling.Asourfirsttargetorganelle,weselectedthe
mitochondrionbecauseitsRNAcontent--derivedfromboththemitochondrial
genomeandfromimported,nuclear-encodedRNAs--hasbeenextensively
characterizedbyawidearrayofcomplementarymethods(Merceretal.2011;Alán
etal.2010;Piechotaetal.2006;Roetal.2013),henceprovidinga“gold-standard”
towhichwecancompareourresults.Themitochondrialmatrixwasalsothefirst
mammaliancompartmentmappedbyAPEXproteomicsmethodology(Rheeetal.
2013).AsaRNA-proteinchemicalcrosslinker,weoptedformildformaldehyde
treatment,whichcovalentlycapturesmostprotein-proteinandprotein-nucleicacid
interactions,andcanbeachievedwithminimaldisruptionofnativeinteractionsin
livecells.ItisforthesereasonsthatformaldehydeisusedforseveralRIP(Chris
GilbertandSvejstrup2006)technologiesforidentifyingtheRNApartnersofspecific
proteinsofinterest,includingourown“fRIP-Seq”protocol(Hendricksonetal.2016).
SinceitwasunclearaprioriwhetherAPEX-catalyzedbiotinylationshould
precedeorfollowtheformaldehydecrosslinkingstep,weexploredbothschemesin
parallel(FigureS1A;seemethods).Eachprotocol,appliedtoHEK293Tcells
expressingmitochondrially-localizedAPEX2(“mito-APEX2”,Figures1B-C),resulted
inclearenrichmentoffifteenmitochondrial-encodedRNAs—relativetothe
cytosolicmarkerGAPDH—asgaugedbyRT–qPCR(averageof49.3±3.5and
60.9±4.1-foldenrichment,respectively,FigureS1A).Assumingthatfixingcellsprior
tobiotinylationwouldbettercapturetransientorweakRNA–proteininteractions,
weselectedthecrosslinking-then-BPprotocolforRNA-Seqanalysis.Whilethis
confirmedthatmitochondrialmRNAswereenriched,asizeable“shoulder”of
conspicuousoff-targetRNAswerealsounexpectedlyenriched(FigureS1B).Thus,
were-examinedourlabelingandcrosslinkingprotocols,usingasamplingofthese
off-targetRNAmarkers(e.g.,theabundantnuclearRNAXIST,andcytosol-localized
3
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
RNAsHOOK2andMAN2C1).ThismorecomprehensiveanalysisrevealedthatAPEX
labelingfollowedbycrosslinkingprovidessuperiorspecificity(FigureS1C).We
hypothesizethatthemildformaldehydetreatmentcompromisesmembrane
integrity(Foxetal.1985),allowingBPradicalstoescapetoadjoiningcompartments
whenAPEXlabelingisperformedafterformaldehydetreatment.
WeusedtheoptimizedAPEXfollowedbycrosslinkingprotocoltomap
mitochondrialRNAsinmito-APEX2-expressingHEK293Tcells(Figure1D,Table1,
tab2).Gene-levelanalysis,comparingRNAcountsbeforeandafterstreptavidin
enrichment,revealedthatall13mRNAsencodedbythemitochondrialgenomewere
highlyenriched(greaterthan3.5fold)inthreeindependentreplicates(Figures1D
andS1E,Table1tab1).EnrichmentwasabsentinnegativecontrolswithH2O2
omitted(FigureS1F).Readdensityplotsmappedtothemitochondrialgenome
demonstratedthatmostofourcapturedRNAscorrespondtofully-processed
transcripts,includingmRNAs,interstitialtRNAs,andtheD-loopleadersequence
fromwhichmitochondrialtranscriptioninitiates(Figure1E).Intriguingly,mitomRNAreaddensitiesappearedtocorrelatewithpreviousmeasuresofmRNAhalflife(Nagao,Hino-Shigi,andSuzuki2008).Forexample,mRNAsencodingMTCO1-3
havelongerhalf-lives,andmorereadsfromAPEX-RIP,thanmRNAsencoding
MTND1-2.
APEX-RIPmappingofthenuclear-cytoplasmicRNAdistribution
HavingestablishedthatAPEX-RIPisbothspecificandsensitiveinthe
mitochondrion,wenextturnedourattentiontoamorechallengingcompartment:
themammaliannucleus.Thenucleusismorecomplexandhasalessdefined
transcriptomethanthemitochondrialmatrix,butpreviousFractionation-Seq
datasets,includingbyENCODE(Dunhametal.2012)(FiguresS2A–D),againprovide
areferencelisttowhichwecancompareourresults.
WegeneratedHEK293TcellsthatstablyexpressAPEX2inthenucleus
(APEX-NLS)orinthecytosol(APEX-NES;NESisanuclearexportsignal).The
specificityofinsitubiotinylationbytheseconstructswithineachcompartmentwas
confirmedbyimaging(Figure2A).Wholecelllysatespreparedfromeachcellline
alsoproduceddistinct“fingerprints”ofbiotinylatedproteins,asassayedby
streptavidinblotting(FigureS1D).
WeperformedAPEX-RIPonbothAPEX-NLSandAPEX-NEScells,usingthe
biotinylation-first/crosslinking-secondprotocolestablishedabove,withan
additionalone-minuteradical-quenchingstepinbetweentheAPEXandcrosslinking
steps(FigureS3A;seemethods).Encouragingly,“goldstandard”nuclearand
cytosolicRNAs(definedfromtheENCODEdataasthetop1000RNAsineach
compartment;seeTable2tab4)wereenrichedfromthecorrespondingcelllinesas
predicted(Figure2BhistogramsandFiguresS2E–F).Moreover,whendirectly
comparingthefold-enrichmentsfromeachcompartmenttooneanother,itwas
apparentthatAPEX-NLShadeffectivelyenrichedknownnuclear-localizedRNAs,
whileAPEX-NEShadenrichedknowncytosol-localizedRNAs(Figure2Bscatterplot,
Table2tab3).WecalculatedforeachRNAa“nuclearpreferencescore,"definedas
theminimumgeometricdistanceofeachpointtotheliney=x(correspondingtothe
setofgeneswhicharenotpreferentiallyenrichedfromeithercompartment).
4
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
ReceiverOperatorCharacteristic(ROC)analysisofthesenuclearpreferencescores
wasusedtofilterthedataandobtainfinaltranscriptlistsof5,467nuclearRNAsand
10,130cytosolicRNAsfromlivingHEK293Tcells(Table2tabs1and2).Thefalse
discoveryratesofthesetwolistsare<0.6%and<0.4%,respectively.
Whenplottedbynuclearpreferencescore,thehumantranscriptome
displayedanoverallbimodaldistribution,whereinthemajorityofspecieswere
cytoplasmic,appendedbyasmallerright-shiftedpopulaceofpredominantlynuclear
RNAs(Figure2C,left).Asmightbepredicted(Derrienetal.2012),manyofthislatter
groupwerelncRNAs,whichclearlyshowedpreferentialnuclearlocalization(Figure
2C,middle).MostmRNAsappearedtobecytosolicinourdata(Figure2C,right).
Notably,wealsoobservedsizeablepopulacesofRNAsexhibitingnoncanonical
nuclear–cytoplasmicpartitioning(Figure2D).3323mRNAs–includingC1orf63,for
example(Figure2D)–appearedpreferentiallynuclear.Manyofthesespecieshave
beenproposedtoplayaroleindampeninggeneexpressionnoise(BaharHalpernet
al.2015).Conversely,234lncRNAsappearedpreferentiallycytoplasmic;these
includetheknowncytoplasmiclncRNASNHG5,amodulatorofstaufen-mediated
decaythatinfluencescolorectaltumorgrowth(Derrienetal.2012;Damasetal.
2016)(Figure2D).
OurAPEX-RIPnuclearandcytosolicRNAlistsprovideanopportunityfora
head-to-headcomparisonwiththetraditionalFractionation-Seqmethodfor
mappingsubcellularRNAlocalization.ROCanalysisoftheENCODEFractionationSeqdatayieldedalistof3,056RNAsenrichedbynuclearfractionation(Table2tab
5).OftheseRNAs,81%(2469)werealsoenrichedinourAPEX-RIPnucleardataset,
implyinggeneralagreementbetweenthetwotechnologies(Figure2E).Notably,
APEX-RIPalsoenrichednearly3000additionaltranscripts.ThesemaybenuclearlocalizedRNAsthatwereopaquetotheENCODEprotocol,orcontaminantsenriched
byAPEX-RIP.Toaddressthispossibility,weexaminedeachdatasetforconspicuous
non-nuclearcontaminants:RNAsthatareknowntobelocalizedattheEndoplasmic
Reticulum(Jan,Williams,andWeissman2014).Satisfyingly,theAPEX-RIPnuclear
dataset,thoughlarger,containedfewerERcontaminantsthandidtheanalogous
fractionation-baseddataset,implyingthatAPEX-RIPproduceshigherspecificity
thanFractionation-Seq(Figure2F,left).
Tocomparethecoverage/sensitivityofeachmethod(sometimestermed
recall),weexaminedtheenrichmentineachdatasetoflncRNAs,whicharethought
tobepredominantlynuclear(Derrienetal.2012).Weassembledalistof827
annotatedlncRNAs(GENCODEhg19)withaverageFPKMpre-enrichmentgreater
than1.0(Table2tab7).OftheselncRNAs,71.7%areenrichedinourAPEX-RIPderivednucleardataset,whilenuclearFractionation-Seqfromthesamecellline
enrichedonly43.4%(Figure2E,right).WeconcludethatAPEX-RIPcanbesuperior
toFractionation-Seqintermsofbothspecificityandcoverage,foranalysisof
endogenousRNAsubcellularlocalization.
EnrichmentofRNAsproximaltotheERmembrane
TheaboveeffortsestablishthatAPEX-RIPcanenrichRNAsinmembrane-enclosed
cellularcompartments.Wenextsoughttoaddresswhetherthetechniquecould
successfullycapturethetranscriptomesof“open”subcellularregions.Previous
5
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
proteomicworkhasshownthatAPEXtaggingexhibitssufficientspatialspecificity
forsuchopencompartments,sincethistechnologyhasproducedhighlyspecific
proteomicmapsof,forexample,themammalianneuronalsynapticcleft(Lohetal.
2016),outermitochondrialmembrane(Hungetal.2017),mitochondrial
nucleoid(Hanetal.2017),andG-proteincoupledreceptorinteraction
network(Lobingieretal.2017;Paeketal.2017).Wewereunsure,however,ifthe
additionalformaldehydecrosslinkingstepwouldpreserveorblurtheestimated<10
nanometerspatialresolutionofAPEXlabeling(Rheeetal.2013).
AsatestcaseforthegeneralityofAPEX-RIPatsuchopencompartments,we
selectedtheEndoplasmicReticulum(ER).TheERisanappealingtargetforseveral
reasons.First,itishosttoaknownsetofcharacteristicRNAsthatwecanuseas
positivecontrols—theso-called“secretome”—whichcomprisesmRNAsencoding
secreted,glycosylated,and/ortransmembraneproteins,thataretranslatedby
ribosomesontheroughER.Second,theERprovidestheopportunitytocomparethe
efficacyofAPEX-RIPtoalternativeapproaches,sinceRNAsinthissubcellularlocale
havebeenpreviouslycharacterizedbothbyFractionation-Seq,andbyanewer
methodologytermedproximity-dependentribosomeprofiling(Jan,Williams,and
Weissman2014;Williams,Jan,andWeissman2014).Thislattertechniquemaps
activeproteintranslationattheERmembranebycombiningribosome
profiling(Ingoliaetal.2009)withproximity-restrictedsequence-specific
biotinylation,usinganER-targetedbiotinligaseandribosomesthataretaggedwith
thepeptidesubstrate(AviTag)ofthatligase.
SinceitwasinitiallyunclearwhichfaceoftheERmembrane(cytosolicor
luminal)wouldbemostamenabletotheAPEX-RIPmethod,wegeneratedfusion
constructsthatlocalizedtheperoxidasecatalyticcentertoeach(Figures3A–B).
ERM-APEX2targetsAPEX2totheERcytosolicsurfaceviaa27-aminoacidfragment
derivedfromthenativeERmembrane(ERM)proteincytochromeP450C1.HRPKDELtargetshorseradishperoxidase(HRP)totheERlumenviaanN-terminalERtargetingsignalandaC-terminalKDELER-retentionmotif(Martelletal.2012).We
haveshownthatHRPcatalyzesthesameproximity-dependentbiotinylation
chemistryasAPEX2(Lohetal.2016),buthashigherspecificactivitythanAPEX2in
theERlumen(Lametal.2014).WegeneratedHEK293Tcellsstablyexpressing
ERM-APEX2andHRP-KDEL,andconfirmedbymicroscopyandstreptavidinblotting
thateachproducedtheexpectedlabelingpatterns(Figures3CandD).Next,we
comparedtheefficacyofeachconstructfortargetRNAisolation,usingthe
biotinylation-first/crosslinking-secondAPEX-RIPprotocol,andanalyzingour
resultsviaRT-qPCRanalysisofestablishedsecretomeandnon-secretomemRNAs24.
ParallelexperimentswithAPEX2-NEScellsservedasnegativecontrols(Figure3E).
Intriguingly,whileAPEX-RIPfromHRP-KDELcellsefficientlyenrichedtarget
secretomemRNAsrelativetonon-targetcontrols(averagefoldenrichment=19.5,
pairedt-testp-value=0.00009),parallelexperimentsinERM-APEX2cellsexhibited
onlymodest,qualitativeenrichmentoftargetspecies(averagefoldenrichment=
1.49,pairedt-testp-value=0.0515).Indeed,resultsfromERM-APEX2cellswere
nearlyindistinguishablefromthoseacquiredfromAPEX2-NEScontrolcells(paired
t-testp-value=0.830,Figure3E,right).Thisissurprising,sinceproteomic
6
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
experimentsinHEK293TcellsexpressingtheidenticalERM-APEX2construct
yieldedhighlyspecificenrichmentofER-localizedproteins(Hungetal.2017).
OurdatastronglyimplythatAPEX-RIPdoesnothavethesamespatial
specificityasperoxidase-catalyzedproteomiclabeling,andmaybelimitedby
perturbationsinducedbyformaldehydecrosslinking.However,wewerehighly
encouragedbythedataobtainedwiththeHRP-KDELconstruct.Wehypothesizethat
APEX-RIPwiththisconstructiseffectivebecauseformaldehydecrosslinking
physicallycouplesRNAsonthecytosolicfaceoftheERtoproteincomplexesthatare
biotinylatedwithintheERlumen,therebyallowingtargetRNAstobeenrichedby
streptavidin(Figure3A).Furthermore,weobservedthatthetargetspecificityofthis
approachcouldbegreatlyimprovedbyadditionofaone-minuteradical-quenching
stepinbetweenthebiotinylationandcrosslinkingstepsinourprotocol(Figure
S3A).Wesurmisethatthisadditionalsteppreventsresidualperoxidase-generated
radicalsfromleakingintoadjoiningcompartmentswhentheintegrityoftheER
membraneiscompromisedduringformaldehydetreatment.
Usingthisimprovedprotocol,weperformedAPEX-RIPonHRP-KDELcells
(Table3tab2).Gene-levelanalysis,comparingRNAcountsbeforeandafter
streptavidinpulldown,revealedadistinctpopulationofsubstantiallyenrichedRNAs
(Figures3FandS3B).Encouragingly,themajority(63.4%)ofsecretomemRNAs
(definedbyERproximalRNAs(Jan,Williams,andWeissman2014)andPhobius
predictedmRNAswithexclusionofnuclearencodedmitochondrialmRNAs,see
methods)residedinthisset,whilemost(97.1%)mRNAsinatestsetofknownnonsecretedgeneswerenotenriched,thusdemonstratingtheabilityofAPEX-RIPto
isolateER-associatedtranscriptsfromthelargerpopulationofcellularRNAs(Figure
3F).UsinghistogramandROCanalysis,wedeterminedtheoptimallog2FKPM
significancethresholdcutoffforeachexperimentalreplicate(FigureS3C;see
methods),obtainingafinallistof2970ERM-associatedRNAsthatwere
independentlyenrichedinmultipleexperiments(Table3tab1).Thisdataset
exhibited94%specificity,basedonprevioussecretoryannotationasdefinedby
GOCC,SignalP,TMHMM,orPhobius(Ashburneretal.2000;Petersenetal.2011;
Kroghetal.2001;Käll,Krogh,andSonnhammer2004).Figure3Hshowsthatwe
alsode-enrichedmRNAslackingsuchsignals.Coveragewaslikewiseexceptional
(97%),asgaugedbytherecallof71literature-curatedwell-establishedERresident
proteins’mRNAs(Table3tab5;Figure3I,seemethods).
WenextcomparedtheERMAPEX-RIPdatasettoanalogousresultsobtained
bysubcellularbiochemicalfractionation(ReidandNicchitta2012),andby
proximity-dependentribosomeprofiling(Jan,Williams,andWeissman2014)(Table
3,tabs3and4,respectively).Encouragingly,APEX-RIPcapturesthemajorityof
RNAsenrichedbyeachofthesealternativetechniques(70%and93%,respectively,
Figure3J),implyingbroadagreementbetweenthedifferentmethodologies.To
examinethisfurther,wequantifiedthespecificityandcoverageofeachapproach,as
above(seemethods).SpecificityanalysisdemonstratedthatAPEX-RIPandribosome
profilingexhibitedsimilarlyhighspecificity(94%and98%,respectively).However,
Fractionation-Seqwassubstantiallynoisier,suchthatonly90%ofenrichedmRNAs
boreasecretoryannotation(Figure3H);theremaining10%comprisedsizeable
populationsofconspicuouscontaminants(FigureS3E).ThecoverageofER-localized
7
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
mRNAsretrievedbyAPEX-RIP(97%)wasalsoconsiderablyhigherthanthose
retrievedbybothFractionation-Seqandribosomeprofiling(73%and77%,
respectively,Figure3J).WeattributetheenhancedcoverageofAPEX-RIPtoits
highersensitivity,sincethismethodappearsbettersuitedforcapturingRNAswith
lowerabundancesthandothealternativeapproaches(FigureS3F-G).Suchhigher
sensitivitymayalsoexplainwhythesetofRNAsenrichedbyAPEX-RIPissomuch
largerthanthoseobtainedbyfractionationandribosome-profiling(Figure3H).
Excitingly,thisfurtherunderscorestheabilityofAPEX-RIPtorecoverRNAsthatare
opaquetoothermethods.Whilethevastmajority(88.7%)ofourenrichedRNAsare
mRNAs,wealsoenrichhundredsofnoncodingRNAspecies–includingantisense
RNAsandlincRNAs(Figure3G).TheseRNAsarenottranslated,andthuscannotbe
detectedbyribosomeprofiling,andtendtobelowlyexpressed,makingthem
difficulttargetsforeitherribosomeprofilingorFractionation-Seq.
Insummary,APEX-RIPissuperiortoexistingmethodsformapping
endogenousRNAsproximaltotheERmembrane,andmaybeextensibletoother
membrane-abuttingsubcellularregionsaswell.
HypothesesfromERandnuclearAPEX-RIPdatasets
WewonderedifthehighlyspecificandcomprehensiveRNAsubcellularlocalization
datasetsproducedbyAPEX-RIPcouldbeminedfornewbiologicalhypotheses.We
firstobservedthat,ofthe2635mRNAsinourERMdataset,141codefor
mitochondrialproteins.Itisthoughtthatthatthebulkofthenuclear-encoded
mitochondrialproteomeistranslatedwithinthebulkcytosol,orinproximityto
mitochondriathemselves(Lesnik,Golani-Armon,andArava2015),raisingthe
possibilitythatthetranslationorsubsequentprocessingofthese141protein
productsrequiremachinerylocalizedtotheER.Additionally,thesemRNAsmaybe
translatedatmitochondria-ERcontactsites,someofwhichhavebeenobservedto
containribosomes(Csordásetal.2006).Togaininitialinsightintotheseunusual
RNAs,weanalyzedthese141genestoseewhether,relativetototalpoolofmRNAs
encodingmitochondrially-localizedproteins,theywereenrichedinparticular
properties(Table4tab1).Intriguingly,57.1%ofthesemRNAscodefor
transmembraneproteins(aspredictedbyTMHMM),comparedtoonly20.4%forall
mitochondrialproteinmRNAs(Figure4A).Mitochondrialsubcompartmentanalysis
showedthattheER-proximalpopulationisenrichedforproteinsdestinedforthe
innermitochondrialmembrane,andisdepletedforresidentmatrixproteins,
comparedtothetotalmitochondrialproteome(Figure4B).Interestingly,proximitydependentribosomeprofilingbyWilliamsetal.inyeastusingbiotinligasetargeted
totheoutermitochondrialmembrane(Williams,Jan,andWeissman2014)also
showedenrichmentofmRNAsencodingproteinsdestinedfortheinner
mitochondrialmembrane.Perhapsasubsetofinnermitochondrialmembranedestinedproteinsarelocallytranslatedatmitochondria-ERcontactsites.
Next,wetestedwhethernewinsightscouldbegainedbyexaminingRNAs
thatAPEX-RIPhadenrichedfrommorethanonesubcellularcompartment.Because
theERlumeniscontiguouswiththatofthenuclearenvelope,wehypothesizedthat
theHRP-KDELAPEX-RIPexperiment,inadditiontoenrichingRNAsproximaltothe
ER,mightalsoenrichRNAsproximaltothenuclearmembrane.Thisregionwithin
8
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
thenucleus,termedthenuclearlamina,iswidelythoughttoplayacriticalrolein
generepression(KindandvanSteensel2010),andinshapingtheglobalthreedimensionalarchitectureofchromatin(C.-K.Chenetal.2016).However,no
exclusivelylaminar-residentRNAshaveyetbeenidentified.Wehypothesizedthat
wemightidentifysuchlong-soughtlaminaRNAsbyintersectingourAPEX-RIP
nuclearandERMRNAlists(Figure4C).Encouragingly,weobserved673suchRNAs
intheintersectionlist,34ofwhicharelongnoncodingRNAs(Figure4D;Table4tab
2).ThissmalllistisacompellingstartingpointforexplorationofregulatoryRNAs
thatmayresideinthenuclearlamina.
Discussion
MethodsformappingRNAsubcellularlocalizationareconstrainedbythelimitsof
theirspatiotemporalprecision,thediversityofRNAspeciesthattheycan
simultaneouslyanalyze,theirgeneralityacrosscelltypesandcompartments,and
theireaseofuse.WebelievethatAPEX-RIPissuperiortoexistingimaging-and
sequencing-basedtechniqueswithregardtomanyofthesefactors.
Comparedtoimaging-basedtechnologies,APEX-RIPofferssuperiortarget
throughput,easeofuse,andtemporalcontrol.Forexample,althoughmodern
variantsofFISHcanachieveextremelyhighspatialprecision–evenenablingthe
visualizationofindividualRNAmolecules(Batish,Raj,andTyagi2011)–this
techniquerequiresthesynthesisandtestingofcustomizedfluorescentprobesfor
eachtranscriptunderenquiry,acumbersomeprocessthatlimitsthroughput(Cabili
etal.2015).AhighlymultiplexedFISHvariant,merFISH,substantiallyboosts
throughput—enablingthousandsoftranscriptstobesimultaneouslyvisualized—
butrequirescomplexprotocolsforprobesetdesignandimaging(K.H.Chenetal.
2015).Analternateapproach,FISSEQ,achievessimilartargetdepthwithoutthe
needforgene-specificprobes,butinsteadreliesoncustomizedinstrumentationand
arococoprocessofinsitusequencingandimaging(Leeetal.2014).Notably,without
incorporatingadditionalstainsormarkers,theseimaging-basedapproaches
providelittleinformationregardingthelocalenvironment(i.e.,proximalcellular
compartmentsorfeatures)neareachRNAtarget.Furthermore,thesetechniques
fundamentallylacktemporalprecision:eachrequiresextensivelyfixingand
permeabilizingcellspriortodatacollection,duringwhichtimediffusionortheloss
ofcellularintegritycanperturbendogenousRNAlocalization.Thislatterissuecan
becircumventedthroughavarietyoflive-cellimagingtechnologies,butthese
requiretheimplementationofcustomizedreagentsthatlimitthroughput,andmay
evendistortthelocalizationoftheRNAtargetsunderenquiry(Paige,Wu,andJaffrey
2011;Hocineetal.2013;Nellesetal.2016).Bycontrast,APEX-RIPisnot
encumberedbyanyoftheseconstraints.Itdoesnotrequirethedevelopmentof
target-specificexpressionconstructsorprobes;nordoesitrelyonspecialized
instrumentation.TheensembleofRNAtargetsanalyzed(and,forthatmatter,the
arrayofRNAclassesanalyzed)istheoreticallylimitedonlybythelibrarysynthesis
andsequencingprotocolsemployed.Moreover,sinceAPEX-RIPcapturesonlyRNAs
proximaltoaspecificsubcellularcompartment,anddoessoduringaone-minute
reaction,thetechniqueoffersbothgreaterinformationcontentandhighertemporal
resolutionthandoitsimaging-basedalternatives.
9
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
Comparedtofractionation-basedtechnologies,APEX-RIPofferssuperior
accuracy,easeofuse,andgeneralversatility.AsillustratedinthenucleusandER,
ourtechniqueoutperformsconventionalFractionation-Seqwithregardtoboth
targetspecificityandrecall,apparentlycircumventingthedualissuesoftargetloss
andoff-targetcontaminationthatcanplaguesuchapproaches(Figures2E-F).We
ascribethisperformanceboosttotwoprincipalfactors.First,thehigh
spatiotemporalprecisionaffordedbyinsitubiotinylation(Rheeetal.2013)allows
ustoefficientlyisolatetargetmaterialfromcontaminantsthatmightbedifficultto
removebyclassicalfractionation,therebyimprovingspecificity.Second,covalently
couplingtargetRNAstoaffinity-taggedproteinsallowsustorecoverlow-abundance
orweaklyaffiliatedtranscriptsthatmightotherwisebelostduringbiochemical
enrichment,therebyimprovingtargetrecall.Perhapsmoreimportantly,however,
wehaveachievedtheseresultsinavarietyofsubcellularcompartmentsusinga
commonprotocol,thusobviatingtheneedtodevelopcustomizedpurification
schemesforeachcompartment.ThisgeneralityshouldenableAPEX-RIPtoaccess
“unpurifiable”subcellularcompartmentsforwhichsuchpurificationschemeswould
beimpossible.Whilearelatedtechnology,proximity-dependentribosomeprofiling,
exhibitssimilarversatilitywithindiversesubcellularmilieus(Jan,Williams,and
Weissman2014),thisapproachislimitedtomRNAsactivelyundergoingtranslation.
Italsorequiresbiotinstarvationpriortotagging,whichistoxictomammaliancells.
Aswehavedemonstrated,APEX-RIPcanmapdiverseclassesofnoncodingRNAand
quiescentmRNA(Figure3G),andtoxicprotocolsstarvingcellsofessentialnutrients
forhoursarenotrequired.
TheAPEX-RIPmethodologydoeshavenotablelimitations.Cellstobe
analyzedmustbetransfectedwitharecombinantconstruct,incontrasttoFISHand
Fractionation-Seq,whichcanbeperformedonnativetissues.APEX-RIPalsogives
poorspatialspecificityinmembrane-freesubcellularregions.
TheAPEXperoxidaseusedherehasalsopreviouslybeenusedtogenerate
contrastforelectronmicroscopyinfixedcells(Martelletal.2012;Lametal.2014),
andforspatially-resolvedproteomicmappinginlivingcells(Rheeetal.2013;Hung
etal.2014;Lohetal.2016;Hanetal.2017;Hungetal.2017;Micketal.2015).This
studyextendsAPEXtoanewclassofapplicationsandtoanewbiopolymer.In
principle,itshouldbepossibletoutilizeasingleAPEX-expressingcelllineto
characterizeatargetsubcellularcompartmentbyelectronmicroscopic,proteomic,
andtranscriptomicmeans.Relatedmethodsforproteomicmapping,suchas
BioID(Rouxetal.2012),lackthisversatility,becausetheunderlyingchemistryisnot
asflexibleastheone-electronoxidationreactioncatalyzedbyAPEX.
Weanticipatethattheinitialsubcellulartranscriptomicmappresentedin
thiswork—probingthemitochondrialmatrix,cytosol,nucleus,andERmembraneof
HEK293Tcells—willserveasvaluableresourcesforcellbiologists.Analysisofthese
datahasalreadyyieldedpotentialinsightintonuclear-retainedmRNAs,cytosolic
lncRNAs,putativelamina-localizedRNAs,andgenesthatmaybetranslatedlocallyat
mitochondria-endoplasmicreticulumjunctions.ApplyingAPEX-RIPatother
subcellularcompartmentswillfurtherexpandthedepthandbreadthofthismap.
Furthermore,giventhehightemporalresolutionofAPEX-RIP,weimaginethatour
technologymightenableprofilingofsubcellularRNApoolsinresponsetoacute
10
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
stimuliordrugs,orthroughoutstagesofthecellcycleanddevelopment.
Collectively,suchstudieswouldyieldanunderstandingintothebiologyofRNA
subcellularlocalizationatunprecedentedscale.
Significance
RNAsubcellularlocalizationisacriticalfactorthatinfluencesawidearrayof
biologicalprocesses,rangingfromDrosophilaembryogenesistomammalian
neuronalsignaling.However,whilethisspatiallayeroftranscriptomeregulationhas
beencharacterizedinahandfulofcontexts,abroaderunderstandingofitsoverall
extent,thefactorsgoverningitsestablishment,anditsimpactonbiologicalfunction,
remaininchoate.Thelimitationshinderingthisunderstandinghavebeenlargely
technical,sinceconventionalmethods—suchasfluorescenceinsituhybridization
(FISH)andFractionation-Sequencing(“Frac-Seq”)—dependuponspecialized
reagentsandprotocolsthatcanlimitthroughputandgeneralapplicability.To
addressthisfundamentalneed,wehavedevelopedanewstrategy—APEX-RIP—
whichusesasimpletoolkitandworkflowtomapthetranscriptomesofdiscrete
subcellularcompartmentsathighdepthandspatiotemporalresolution.APEX-RIP
utilizestheengineeredascorbateperoxidaseAPEXtobiotinylateproteinswithina
targetsubcellularcompartmentinlivecells;theseaffinity-taggedproteinsarethen
chemicallycrosslinkedinsitutonearbyRNAs.Whenappliedtoavarietyof
membrane-enclosedandmembrane-adjacentcompartments,theAPEX-RIPstrategy
exhibitedhighertargetspecificityandcoveragethandoconventionalfractionationsequencing-basedapproaches,atadepthfarexceedingthoseattainablebyimagingbasedmethods.Furthermore,APEX-RIPcanbeappliedtocompartmentsthatare
recalcitranttoconventionalbiochemicalpurification.Giventhesuperiorprecision,
flexibility,andeaseofthisapproach,weanticipatethatAPEX-RIPwillprovidea
powerfultoolfordissectingRNAsubcellularlocalizationinabroadrangeof
biologicalcontexts.
Authorcontributions
Conceptualization,PK,DMS,AYT;Methodology,PK,DMS,AYT;Validation,PKand
DMS;FormalAnalysis,PKandWM;Investigation,PKandDMS;DataCuration,PK;
Writing–OriginalDraft,PK,DMS,andAYT;Writing–Review&Editing,PK,DMS,
JLR,andAYT;Visualization,PK,DMS,andAYT;Supervision,JLRandAYT;Project
Administration,AYT;FundingAcquisition,JLRandAYT.
Acknowledgements
WethankmembersoftheTinglaboratory,especiallyJeffreyMartellforvaluable
experimentaladviceandOzanAygunforthecuratedERproteinlist.Wethankthe
Rinnlaboratory,especiallyChinmayShukla,forvaluablecomputationaladvice.
FundingwasprovidedbytheNIH(R01-CA186568toA.Y.T.andU01DA040612to
J.L.R.)andStanford(toA.Y.T).
11
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
Figure1.APEX-RIPinmitochondria.(A)OverviewoftheAPEX-RIPworkflow.Cells
expressingAPEX2(grey‘pacmen’)targetedtothecompartmentofinterest(here,the
mitochondrialmatrix.)areincubatedwiththeAPEXsubstratebiotin-phenol(BP;redB:
biotin).Aone-minutepulseofH2O2initiatesbiotinylationofproximalendogenous
proteins(Rheeetal.2013),whicharethencovalentlycrosslinkedtonearbyRNAsby0.1%
formaldehyde.Followingcelllysis,biotinylatedspeciesareenrichedbystreptavidin
pulldown,andcoelutingRNAsareanalyzedbyRT-qPCRorRNA-Seq.IMM:inner
mitochondrialmembrane.(B)ImagingAPEX2biotinylationinsitu.HEK293Tcells
expressingV5-taggedmito-APEX2werebiotinylatedandfixedasdescribedin(A)and
stainedasindicated.ThebottomrowisanegativecontrolinwhichH2O2treatmentwas
omitted.Scalebars,10µm.(C)Streptavidinblotanalysisofwholecelllysatespreparedas
describedin(A).Insitubiotinylation(lane1)isablatedintheabsenceoftheAPEX2protein,
H2O2,orBP.Anti-V5blotdetectsexpressionofmito-APEX2.(D-E)mito-APEX-RIPefficiently
recoversthemitochondrialtranscriptome.(D)Gene-levelRNA-Seqanalysisofmito-APEXRIPenrichmentforRNAslongerthan200nt.Datafromonebiologicalreplicateareshown.
(E)Nucleotide-levelRNA-Seqanalysisofmito-APEX-RIP,mappedtothehuman
mitochondrialgenome(innermostcircle).Outermostcircle:readsfromthefullAPEX-RIP
12
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
519
520
521
protocol;middlecircle:readsfromthenegativecontrol.Notetheenrichmentofseveral
mitochondrially-encodedtRNAsandtheD-loopleadertranscript.RibosomalRNAswere
removedduringlibrarypreparation(seemethods).Seealso:FigureS1.
13
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
522
523
524
525
526
527
528
529
530
531
Figure2.APEX-RIPmappingofthenuclear-cytoplasmicRNAdistribution.(A)
Fluorescenceimagingofnuclearandcytosol-targetedAPEX2fusionconstructs.HEK293T
cellsexpressingtheindicatedconstructs(“NLS,”nuclearlocalizationsignal;“NES,”nuclear
exportsignal)werelabeledwithbiotin-phenol,crosslinkedandstainedasindicated.DIC,
DifferentialInterferenceContrast.Scalebars,10µm.(B)CombinedanalysisofAPEX2-NLS
andNESexperimentsdistinguishesnuclearandcytoplasmicallylocalizedRNAs.Fold
enrichmentvalueswerecalculatedrelativetomatchedinputsamples;themedianvaluesof
14
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
threereplicatesareshown(seemethods).The1000RNAswiththehighestpredicted
nuclearandcytosoliclocalizationbyENCODE(Dunhametal.2012)arecoloredgreenand
red,respectively(seemethods).Histogramplots,summarizingtheseparationoftheseRNA
standardsbyeachAPEX2construct,areprojectedalongtheaxesofthescatterplotanduse
thesamescales.TheblackdottedlinemarksthecutoffbetweennuclearandcytosolicRNAs.
(C)APEX-RIPcapturestheestablishednuclear-cytoplasmicdistributionofmRNAsand
lncRNAs.Top:APEX2-NLSversusAPEX2-NESscatterplots,asin(B),forallRNAs(left),
lncRNAs(middle),andmRNAs(right).Dataarethemediansofthreereplicates.Dottedlines
markthecutoffbetweennuclearandcytosolicRNAs,asin(B).Bottom:histogramplotsof
nuclearpreferencescores(seemethods)foreachclassofRNA.Dottedredlines:theROCderivedsignificancethreshold(seemethods).Inset:thecompletedistribution.(D)Read
densityplotsofRNAswithstereotypicalandatypicallocalization.Foreachgene,acommon
y-scaleisusedforallreadtracks.SnoRNAsencodedintheSNHG5genebodyareindicated
asgrayrectangles.(E)VenndiagramcomparingAPEX-RIPandENCODEnuclearRNA
datasets(Dunhametal.2012).(F)NuclearAPEX-RIPismorespecificandsensitivethanis
biochemicalfractionation.Left:SpecificityoftheAPEX-RIPandENCODEnuclearRNA
datasets(Dunhametal.2012).Off-targetRNAsweredefinedactivelytranslatedERproximalmRNAs(Jan,Williams,andWeissman2014).Right:Recallofnuclearstandard
RNAs,definedasasetof827lncRNAsannotatedbyGENCODEhg19withaveragepreenrichmentFPKM≥1.0.Seealso:FigureS2.
15
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
552
553
16
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
Figure3.APEX-RIPmapsRNAsproximaltotheEndoplasmicReticulum.(A—B)
SchematicssummarizingalternateER-targetingstrategies.(A)HRP,targetedtotheERwith
aKDELsequence,biotinylatesproteinswithintheERlumen.RedB:biotin.RedXs:chemical
crosslinksinducedby0.1%Formaldehydetreatment.(B)APEX2,displayedontheER
membrane(ERM)byfusingittothetransmembranesegmentofrabbitP450C1,facesthe
cytosol.(C)ImagingHRP-KDEL-catalyzedbiotinylation.HEK293Tcellsstablyexpressing
HRP-KDELwerelabeledwithBP,fixedandimagedasinFigure1B.DIC,Differential
InterferenceContrast.Scalebars,10µm.(D)StreptavidinblotdetectionofresidentER
proteinsbiotinylatedbyHRP-KDEL,asinFigure1C.Arrowheadsdenoteendogenously
biotinylatedproteins(Chapman-SmithandCronan1999).(E)RT-qPCRanalysis,comparing
specificitiesofthelabelingschemesshownin(A–B).Targetandoff-targetgeneswere
selectedusingpreviously-reportedfoldenrichmentsattheERmembrane(Jan,Williams,
andWeissman2014).Dataarethemeanofthreereplicates,±onestandarddeviation.(F)
ScatterplotshowingRNAabundancebeforeandafterstreptavidinenrichment.Datashown
areforonereplicate.AdditionalreplicatesinFigureS3B.(G)ClassificationofAPEX-RIP
enriched,ER-associatedRNAs.Collectively,non-codingRNAsconstitute11.3%ofenriched
genes(335of2970RNAs).(H)Specificityanalysisforprotein-codingmRNAsinourERassociatedRNAlist.95%ofthe2635APEX-RIPER-enrichedmRNAsexhibitsomeformof
secretoryannotation(aspredictedbyPhobius,TMHMM,SignalP,orGOCC,seemethods),
whereasonly60.3%ofallhumanmRNAsaresimilarlyclassified(left).RNAsinourdataset
thatwerenotenrichedbyribosomeprofiling(1802RNAs)werealsopredominantly
secretory(90.9%).(I)TargetrecallofERAPEX-RIPexceedsthoseofproximity-restricted
ribosomeprofiling(Jan,Williams,andWeissman2014)andbiochemicalfractionation(Reid
andNicchitta2012).Seealso:TableS5.(J)ThegenesetenrichedbyERAPEX-RIPlargely
recapitulatesthoseenrichedbyalternativemethods.Seealso:FigureS3.
17
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
Figure4.APEX-RIPrevealsRNAswithpotentiallynovellocalization.(A)Many
mitochondrialtransmembraneproteinsappeartobetranslatedattheER.mRNAsencoding
mitochondrialproteinsdefinedbyGOCCandMitoCarta1.0(Pagliarinietal.2008;Ashburner
etal.2000),withpredictedtransmembranehelices(predictedbyTMHMM(Kroghetal.
2001);greendistribution)arepreferentiallyenrichedbyHRP-KDELAPEX-RIP,relativeto
mitochondrialmRNAslackingtransmembranedomains(reddistribution)(B)Predicted
localizationofmitochondrialproteinsencodedbymRNAsthatwereenrichedinER-and
bulkcytosol-APEX-RIPexperiments(leftandmiddle,respectively),andofallGOCCannotatedmitochondrialproteins.OMM:Outermitochondrialmembrane.IMS:
Intermembranespace.IMM:Innermitochondrialmembrane.(C)Computationalschemefor
identifyingputativelamina-associatedRNAs.SinceHRP-KDEL-enrichedRNAs(left)
comprisebothER-andlamina-associatedRNAs,candidatenuclearlamina-localizedRNAs,
wereidentifiedastheintersection(right)ofRNAsenrichedbybothAPEX2-NLS(middle)
andHRP-KDEL.Red:enrichedRNAs;greenpacmen:isAPEX2orHRPperoxidases.(D)Venn
diagramidentifyingputativelamina-associatedRNAs,definedastheoverlapbetweenHRPKDEL-andAPEX2-NLS-enrichedRNAs.Seealso:Table4,tab2.Thesignificanceofoverlap
betweenER-associatedRNAsandnuclear-enrichedRNAsbyAPEX-RIPismeasuredby
hypergeometrictestusingalltypeofRNAsoronlylncRNAsaspopulation,respectively.
18
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
599
600
19
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
SupportingFigure1.OptimizationofAPEX-RIPprotocolandadditionalmitochondrial
APEX-RIPdata(relatedtoFigure1).(A)Top:Alternatelabelingandcrosslinking
protocols.InprotocolI.,cellsarecrosslinkedwithformaldehyde(FA)andquenchedwith
Glycine(Gly)priortotheintroductionofbiotin-phenol(BP)andtheinitiationofAPEXcatalyzedbiotinylationwithH2O2.InProtocolII.,livecellsareincubatedinBP,andinsitu
biotinylationisinitiatedpriortoFAcrosslinking.Inbothcases,celllysis,streptavidin
enrichmentandRNApurificationproceedasdescribed(seemethods).Bottom:qRT-PCR
analysiscomparingProtocolsIandII.Negativecontrolexperimentsreplacemito-APEXwith
mito-GFP,omitBPoromitH2O2.AllconstructswereAPEX1derivatives,transiently
expressedinHEK293Tcells.Dataarethemeansofthreereplicates±onestandard
deviation.(B)RNA-SeqanalysisofRNAsenrichedbyprotocolI.Althoughall13
mitochondrially-encodedmRNAs(green)wereenriched,thesewereaccompaniedby
substantialcontaminatingRNAs,includingXISTandMAN2C1.(C)qRT-PCRanalysis
comparingProtocolsI.andII.,includingoff-targetcontrolsdesignedusingtheresultsfrom
RNA-Seq.NotethesuperiorenrichmentobtainedusingProtocolII.Cellsinthisexperiment
stably-expressedmito-APEX2.Thisprotocolandcelllinewereusedtocollectalldatain
(Figure1).Dataarethemeansofthreereplicates±onestandarddeviation.(D)
CharacterizationofallAPEX2fusionconstructsusedinthiswork.HEK293Tcellsstably
expressingtheindicatedconstructs(right)werelabeledandcrosslinkedviaProtocolII.,
lysedandanalyzedbySDS-PAGE,blottingwithstreptavidin-HRP,Anti-V5andanti-FLAG.L:
ladder;U:untransfectedHEK293Tcells.(E)Datafromadditionalreplicatesofthe
mitochondrialAPEX-RIPexperiment,depictedasin(Figure1D).(F)IntheabsenceofH2O2
treatment,mito-APEX-RIPfailstoenrichmitochondrialtargetgenes.Datafromindividual
replicatesareshown.
20
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
SupportingFigure2.IdentificationofENCODE-derivednuclearRNAstandards;
additionalrepresentationsoftheAPEX-RIPnuclear—cytoplasmicdata.(A)humancell
line(NHEK)fractionationdatafromENCODE(Dunhametal.2012),illustratingtherelative
enrichmentofgenesduringnuclearandcytosolicfractionation.RNAenrichmentsare
calculatedasin(Figure2B)(seemethods).Displayedaretheaveragevaluesofthree
replicates,forallgeneswithFPKMpreenrichment≥1.0.Dashedlinedenotesy=xline.(B)
Distributionofgenesbynuclearpreferencescore(seemethods).lncRNAs,predictedtobe
mostlynuclear,areplottedingreen,andmRNAs,predictedtobemostlycytosolic,are
plottedinred.(C–D)Selectionofnuclear-enrichedRNAstandardsusingENCODE
fractionationdata.(C)ROCanalysisappliedtohistogramin(B).Foreachnuclearpreference
cutoffvalue,theTruePositiveRate(TPR)–definedasthefractionoflncRNAsabovethe
cutoff–wasplottedagainsttheFalsePositiveRate(FPR)–definedasthefractionofmRNAs
abovethecutoff.(D).UsingtheoutputofROCanalysis,anuclearpreferencescorecut-off
valuewascalculated,definedasthefirstlocalmaximuminthegraphof(TPR-FPR)versus
NuclearPreferenceScore.Thisvaluewasappliedtoobtainalistof3056nuclear-enriched
standardRNAs.(E)DatafromindividualreplicatesoftheNLS-APEX-RIPexperiment.Genes
enrichedbynuclearandcytoplasmicfractionationintheENCODEdatasetarecoloredgreen
andred,respectively.(F)DatafromindividualreplicatesoftheNES-APEX-RIPexperiment,
depictedasin(E).
21
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
646
647
22
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
SupportingFigure3.FurtheroptimizationoftheAPEX-RIPprotocol;additionalHRPKDELdata(RelatedtoFigure3).(A)AdditionofaradicalquenchingstepbetweenAPEX2
labelingandformaldehydecrosslinkingimprovesthespecificityofRNAcapture.Top:
schematicoftherevisedlabeling–crosslinkingworkflow.Bottom:qRT-PCRanalysisofthe
ER-APEX-RIPexperimentwithandwithoutthisadditionalquenchingstep,asin(Figure
S1C).TheradicalquenchersusedwereTroloxandascorbicacid.(B–D)Qualityassessment
forindividualreplicatesoftheER-APEX-RIPexperiment.(B)Datafortwoadditional
replicatesoftheHRP-KDELAPEX-RIPexperiment,depictedasin(Figure3F).(C)
HistogramsshowingthedistributionofRNAenrichmentvalues–log2([FKPMpost
enrichment]/[FKPMpreenrichment])–formRNAsencodingknownsecretoryproteins(top
histograms,green),andmRNAsencodingnon-secretoryproteins(bottomhistograms,red).
SignificancecutoffsweredeterminedusingROCanalysis(seebelow).Knownsecretoryand
non-secretorystandardmRNAswerecatalogedasdescribed(seemethods).(D)
Determinationofsignificancethresholdsratiocut-offsbyROCanalysis.Datawere
processedasdescribedinFiguresS2D–E,usingTruePositivesecretoryandFalsePositive
non-secretorystandardRNAs(seemethods).Giventheseanalyses,andthosein(B–C),only
datafromReplicates1and2wereusedtogenerateourfinalER-associatedRNAlist.Todo
this,theindividualROC-derivedthresholdswereusedtocalculatesignificantlyenriched
genesfromeachdataset,andthefinalER-associatedRNAlistwasdefinedasthe
intersectionofthesetwolists.(E)Geneontology(GO)analysisofmRNAsinERdatasets
lackingsecretoryannotation.Left:non-secretorymRNAsenrichedbybiochemical
fractionation(ReidandNicchitta2012)predominantlyexhibitnuclearandmitochondrial
annotation.Right:non-secretorymRNAsenrichedbyER-APEX-RIPhavepredominantly
cytosolicannotation.Alltermsshownhavep<0.05,asassessedusingDAVID(Huang,
Sherman,andLempicki2009).(F—G)APEX-RIPrecoverslow-abundancetargetswith
greaterefficiencythandoconventionalapproaches.Ineachcase,startingabundancesare
definedbyENCODE.(F)ComparisonofAPEX-RIPtobiochemicalfractionation.(G)
ComparisonofAPEX-RIPtoproximity-restrictedribosomeprofiling.
23
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
ExperimentalProcedures
Plasmidsandcloning
ThepCDNA3mito-APEXplasmidwaspublishedpreviously(Rheeetal.2013).The
Mito-APEX2constructwasclonedfromthisplasmidusingatwo-stepprotocol.First,
theA134Pmutation(Lametal.2014)wasintroducedintotheAPEXgeneitself,
usingQuikChangemutagenesis(Agilent),andthereaftertheAPEX2genewasmoved
tothelentiviralvectorpLX304viaGatewaycloning(Thermofisher),togenerate
plasmidpLX304mito-APEX2.OtherAPEX-fusionconstructs(pLX304APEX2-NLS,
pLX304APEX2-NES,andplx304ERM-APEX2)wereclonedbyGibsonassembly
(NEB),usingPCRtoaddtargetingsequencesandGibsonAssemblyhomologyarms
totheAPEX2gene,andjoiningtheresultinginsertintothepLX304vectordigested
byBstBIandNheI. ForHRP-KDEL,HRPCpreviouslypublished(Martelletal.2016)
wasusedasatemplatetomakeHRP-KDEL-IRES-PuromycinPCRfragment.Thenthe
insertwasclonedintoPCDNA3vectordigestedbyNotIandXbaI.Targeting
sequencesandrestrictionsitesforallconstructsarelistedin(TableS1).
Mammaliancellculture
Humanembryonickidney(HEK)293TcellswereobtainedfromATCC,andcultured
ingrowthmediaconsistingof1:1DMEM:MEM(Cellgro),supplementedwith10%
FetalBovineSerum(FBS),50units/mLpenicillin,and50μg/mLstreptomycin,at37
°Candunder5%CO .Cellswerediscardedat25passages,andwereperiodically
testedforMycoplasmacontaminationusingUniversalMycoplasmaDetectionkit
(ATCC).Forfluorescencemicroscopyimagingexperiments(Figures1B,2Aand3C),
cellsweregrownon7×7-mmglasscoverslipsin48-wellplates.Toimprovecell
adherence,coverslipswerepretreatedwith50μg/mLfibronectin(Millipore)for20
minat37°CandwashedoncewithDulbecco’sphosphate-bufferedsaline(DPBS),
pH7.4.CellsusedforgeneratinglentivirusweregrownonT25plates,inMEM
supplementedasabove,at37°Cunder5%CO2.
PreparationofcelllinesstablyexpressingAPEX-fusionconstructs
Topreparelentivirus,one~70%confluentT25plateofHEK293Tcells,grownas
above,wasco-transfectedwith2.5μgofAPEX2fusionplasmid,alongwith0.25μg
and2.25μg,respectively,ofthelentiviruspackagingplasmidsVSV-G,and
dR8.91(Pagliarinietal.2008).Transfectionmixesused10μLLipofectamine2000
(Invitrogen)andwerebroughttoafinalvolumeof2mLwithunsupplemented
MEM.Thecellsweretransfectedfor3hours,afterwhichmediawasreplacedwith2
mloffreshgrowthmediawithFBS.After48hours,thelentiviralsupernatantwas
collectedbyaspirationandfilteredthrougha0.45μmsyringe-mountedfilter.This
filteredsupernatantwasimmediatelyusedtoinfectcells.HEK293Tcells,grownin
6-wellplatesasdescribedabove,wereinfectedat~50%confluency,grownfor2
days,followedbyselectioningrowthmediumsupplementedwith8μg/mL
blasticidinfor7days,beforefurtheranalysis.
ForthecellsstablyexpressingHRP-KDEL,HEK293Tcellsat~60%
confluency,grownin6-wellplatesasdescribedabove,weretransfectedwiththe
2
24
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
mixtureof150μgofplasmidand10μLLipofectamine2000(Invitrogen)in
unsupplementedMEMfor3hours,afterwhichmediawasprelacedwith2mlof
freshgrowthmediawithFBS.After48hours,thecellsweretrypsinizedand
replatedinT25flaskingrowthmediumsupplementedwith1μg/mLpuromycinfor
7days,beforefurtheranalysis.
Insitubiotinylationandcrosslinking
Stable-expressionHEK293Tcellsweregrownto90%confluencyin6-wellplates,as
describedabove.Forthecrosslinking–then–BPbiotinylationprotocol(FigureS1A,
top),cellswerewashedoncewith5mLPBS,andcrosslinkedin5mL0.1%(v/v)
formaldehydeinPBSfor10minatroomtemperature,withgentleagitation.The
crosslinkingreactionwasquenchedbyadditionofglycine(1.2M,inPBS)tofinal
concentration125mM,andgentleagitationfor5minutesatroomtemperature.
CrosslinkedcellswerethenwashedthreetimeswithPBSandincubatedwith500
µMbiotin-phenol(BP)(Rheeetal.2013)inPBSatroomtemperature,for30min.
Thereafter,H2O2wasaddedtoafinalconcentration1mM,for1min.Theliquid
phasewasthenremovedbyaspiration,andcellswerewashedtwicewith2mL
quenchingsolution(5mMTrolox,10mMAscorbate,10mMsodiumazide,inPBS).
Crosslinked,labeledcellswerecollectedbyscraping,andpelletedbycentrifugation,
andeitherprocessedimmediatelyorflashfrozeninliquidnitrogenandstoredat–
80°Cbeforefurtheranalysis.
FortheBP–then–crosslinkingprotocol(FigureS1A,bottom)usedformitoAPEX2experiments(Figure1),cellgrowthmediawasreplacedwithfreshmedia
supplementedwith500µMBP.CellswereincubatedinBP-supplementedmediafor
30minutesat37°C,afterwhichH2O2wasaddedtoafinalconcentrationof1mM.
After1min,themediawasreplacedwith5mLcrosslink-quenchsolution(0.1%
(v/v)formaldehyde,10mMascorbate,and5mMTrolox,inPBS)foroneminute,to
simultaneouslyquenchtheAPEX2BPlabelingreactionandinitiateformaldehyde
crosslinking.Thereafter,cellswerewashedandincubatedin5mLoffreshcrosslinkquenchfortwoadditional1-minuteincubationsteps,followedbyathird,8-minute
wash.Thereafter,crosslinkingwasterminatedbytheadditionofGlycine,andcells
wereharvestedasdescribedabove.
TheBP–quench–then–crosslinkingprotocol(FigureS3A)usedforallother
subcellularcompartmentswasidenticaltotheBP–then–crosslinkingprotocol,
exceptthat,followingBP-labling,andpriortotheadditionofcrosslink-quench
solution,cellswereincubatedin2mLazide-freequenchingsolution(10mM
ascorbateand5mMTrolox,inPBS)foroneminute.Subsequently,cellswere
subjectedtoonlytwo(1and9minute)treatmentsincrosslink-quenchsolution.
Thereafter,crosslinkingwasterminatedbytheadditionofGlycine,andcellswere
harvestedasdescribedabove.
Immunofluorescencestainingandmicroscopy
Forimmunofluorescenceexperiments(Figures1B,2A,and3C),stableAPEX-or
HRP-expressingcellswereBP-labeledandcrosslinked,asabove,andsubsequently
fixedwith4%(v/v)paraformaldehydeinPBSatroomtemperaturefor10min.Cells
25
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
werethenwashedwithPBSthreetimesandpermeabilizedwithcoldmethanolat–
20°Cfor5min.Cellswerewashedagainthreetimeswithroom-temperaturePBS
andthenincubatedwithprimaryantibodiesinPBS–supplementedwith1%(w/v)
BovineSerumAlbumin(BSA)–for1hatroomtemperature.Afterwashingthree
timeswithPBS,cellswereincubatedwithsecondaryantibodiesandneutravidinAlexaFluor647(1:1000dilution)inBSA-supplementedPBSfor30min.Cellswere
thenwashedthreetimeswithPBSandimagedbyconfocalfluorescencemicroscopy,
orinPBSat4°Cinlight-tightcontainerspriortoimaging.Primaryandsecondary
antibodiesusedwerelistedinTableS2.
FluorescenceconfocalmicroscopywasperformedwithaZeissAxioObserver
microscopewith63×oil-immersionobjectives,outfittedwithaYokogawaspinning
diskconfocalhead,aCascadeII:512camera,aQuad-bandnotchdichroicmirror
(405/488/568/647),405(diode),491(DPSS),561(DPSS)and640nm(diode)
lasers(all50mW).AlexaFluor488(491laserexcitation,528/38emission),Alexa
Fluor568(561laserexcitation,617/73emission),andAlexaFluor647(640laser
excitation,700/75emission)anddifferentialinterferencecontrast(DIC)images
wereacquiredthrougha63xoil-immersionlens.Acquisitiontimesrangedfrom100
to1,000ms.Forimagingquantitationandanalysis,weusedtheSlideBook6.0
software(IntelligentImagingInnovations)toprocessandnormalizetheimages.
Thedatainthesefigures(Figure1B,2A,and3C)arerepresentativeofthree
independentexperimentswith³5fieldsofvieweach.
WesternandStreptavidinblotting
Forblottingexperiments(Figures1C,3DandS1D),stableAPEX-orHRP-expressing
cellsweregrownin6-wellplates.Afterlabeling,thecellswereharvestedby
scraped,pelletedbycentrifugationat3,000×gfor10min,andstoredat–80°Cprior
touse.ThawedpelletswerelysedbygentlepipettinginRIPAlysisbuffer(50mM
Tris,150mMNaCl,0.1%SDS,0.5%sodiumdeoxycholate,1%TritonX-100,5mM
EDTA),supplementedwith1×proteasecocktail(SigmaAldrich),1mMPMSF
(phenylmethylsulfonylfluoride),for5minat4°C.Lysateswerethenclarifiedby
centrifugationat15,000×gfor10minat4°Cbeforeseparationonhomemade8%
SDS-PAGEgels.Gelsweretransferredtonitrocellulosemembranes,stainedby
PonceauS(0.1%(w/v)PonceauS,5%(v/v)aceticacid,inwater)for10minat
roomtemperature,andimaged.Theblotswerethenblockedwithblockingbuffer
(3%(w/v)BSA,0.1%(v/v)Tween-20inTris-bufferedsaline)for1hatroom
temperature,andincubatedwithprimaryantibodiesinblockingbufferfor1hmore.
Thedilutionsoftheantibodiesareasfollowed:Mouseanti-V5antibody(Life
Technologies)1:1000dilutionandMouseanti-FLAGantibody(LifeTechnologies)
1:800dilution.Blotswererinsedfourtimesfor5minwithwashbuffer(0.1%
Tween-20inTris-bufferedsaline),andthenimmersedinblockingbuffer
supplementedwithGoatanti-MouseIgGH+L-HRPConjugate(1:3,000dilution,BioRad),for1hatroomtemperature.Blotswererinsedfourtimesfor5minwithwash
buffer,anddevelopedwiththeClarityreagent(Bio-Rad)andimagedonanAlpha
Innotechgelimagingsystem.Processingofstreptavidinblotswassimilar.
FollowingPonceauimaging,blotswereblockedinblockingbufferfor30minat
roomtemperature,immersedinblockingbuffersupplementedwithstreptavidin
26
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
HRP(1:3,000dilution,ThermoFisherScientific)atroomtemperaturefor15min,
rinsedwithblockingbufferfivetimesfor5mineach,developedandimagedusing
theClarityreagentandanAlphaInnotechgelimagingsystem.
Thedataintheseexperiments(Figures1C,3DandS1D)werealso
reproducedforqualitycontrolpriortoquantitativePCRandsequencing.
StreptavidinbeadenrichmentofbiotinylatedmaterialandRNAisolation
Unlessotherwisenoted,allbuffersusedduringRNAisolationweresupplementedto
0.1U/µLRNaseOUT(ThermoFisher),1×EDTA-freeproteinaseinhibitorcocktail
(ThermoFisher)and0.5mMDTT,final.APEX-orHRP-expressingstablecellswere
grown,labeled,crosslinkedandharvestedasdescribedabove.Labeledcellpellets
werelysedbyincubationin1mLice-coldRIPAbuffer,supplementedwith10mM
ascorbateand5mMTrolox,for5minat4°Cwithend-over-endagitation.Samples
werethenshearedasdescribedpreviously(Hendricksonetal.2016)usinga
BransonDigitalSonifier250(EmersonIndustrialAutomation)at10%amplitudefor
three30-sintervals(0.7son+1.3soff),with30-srestingstepsbetweenintervals.
Sampleswereheldinice-coldmetalthermalblocksthroughoutsonication.Lysates
werethenclarifiedbycentrifugationat15,000×gfor5minat4°C,movedtofresh
tubesanddilutedwith1mLNativelysisbuffer(NLB:25mMTrispH7.4,150mM
KCl,0.5%NP-40,5mMEDTA),supplementedwithascorbateandtrolox),each.For
eachsample,20%wasremovedas“input;”totheremainderwasadded50µLof
streptavidin-coatedmagneticbeadslurry(Pierce)thathadbeenequilibratedbytwo
washesin1:1RIPA:NLB.Sampleswereincubatedfor2hat4°Cwithend-over-end
agitation.Beadsweresubsequentlywashedwiththefollowingseriesofbuffers(1
mLeach,5minperwash,4°C,withgentleend-over-endagitation):(1)RIPAbuffer,
supplementedwithtroloxandascorbate,(2)RIPAbufferwithoutradicalquenchers,
(3)highsaltbuffer(1MKCl,50mMTris,pH8.0,5mMEDTA),(4)ureabuffer(2M
Urea,50mMTris,pH8.0,5mMEDTA),(5)RIPABuffer,(6)1:1RIPA:NLB,(7)NLB,
and(8)TE(10mMTris,pH7.4,1mMEDTA).
EnrichedRNAswerereleasedfromthebeadsbyproteolysisin100µLof
ElutionBuffer(2%N-laurylsarcoside,10mMEDTA,5mMDTT,in1XPBS,
supplementedwith200µgproteinaseK(Ambion)and4URNaseOUT)at42°Cfor
1h,followedby55°Cfor1h,aspreviouslydescribed(Hendricksonetal.2016).
ElutedsampleswerecleanedupusingAgencourtRNACleanXPmagneticbeads
(BeckmanCoulter),followingthemanufacturer’s1.5mLtubeformatprotocol,and
elutedinto85µLH2O.Thereafter,contaminatingDNAwasremovedbydigestion
with5URQ1RNase-freeDNaseI(Promega)in100µLofthemanufacturer’s
suppliedbuffer(1Xfinalconcentration)at37°Cfor30min.PurifiedRNAswere
againcleanedupusingAgencourtRNACleanXPmagneticbeads,asabove,and
elutedinto30µLH2O.Theconcentrationandintegrityofallsampleswasmeasured
usinganAgilent2100Bioanalyzer,followingthe“RNANano”or“RNAPico”
protocols,whereappropriate.Sampleswerenotheat-cooledpriortoloading
Bioanalyzerchips.
QuantitativeRT–PCR
27
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
ForquantitativeRT–PCR(qRT–PCR,FiguresS1A,S1C,3E,andS3A)RNAsamples
werereversetranscribedusingtheSuperScriptIIIReverseTranscriptasekit
(ThermoFisherScientific),primingwithrandomhexamers(ThermoFisher
Scientific)accordingtothemanufacturer’sprotocol.Samplesweredilutedwith
water,mixedwithgenespecificprimers(TableS3),andRox-normalizedFastStart
UniversalSYBRGreenMasterMix(Roche),andaliquottedinto384-wellplates.qRT–PCRwasperformedonanAppliedBiosystems7900HTFastrealtimePCR
instrument,inquadruplicate.Allthresholdcycles(Ct,calculatedperwell)and
efficiencies(ε,calculatedperprimerpair),werecalculatedfrom“clipped”data,
usingRealtimeqPCRMiner(ZhaoandFernald2005).RawCtvalueswerecorrected
toaccountforthedifferencesinsamplevolume,andpercentyieldswerecalculated
viatheDCtmethod:
𝑦𝑖𝑒𝑙𝑑 = 100×(1 + 𝜀)∆/0 …wherein,∆𝐶2 = 89::𝐶234560 − 89::𝐶2<=> Experimentaluncertaintieswerecalculatedasdescribedpreviously(Shechneretal.
2015).GivenD=A–B,uncertainlywascalculatedusingtheformula:
𝜎@ = (𝜎A )B + (𝜎C )B …whereinsAandsBarethemeasurementerrorsofAandB,respectively.ForP,the
productorquotientofvaluesAandB,uncertaintywascalculatedusingtheformula:
𝜎A B
𝜎C B
+
𝐴
𝐵
Theuncertaintiesofotherfunctions,f(x),werecalculatedusingthefirstderivative
approximation:
𝜎H(I) = 𝜎I ×𝑓′(𝑥)
Samplesizesweredeterminedinaccordancewithstandardpracticesusedinsimilar
experimentsintheliterature;nosample-sizeestimateswereperformedtoensure
adequatepowertodetectaprespecifiedeffectsize.Experimentswereneither
randomizednorblindedtoexperimentalconditions.Eachsamplescontainedfour
replicatesandnosampleswereexcludedfromanalysis.
TheexperimentsforFiguresS1A,S1C,3E,andS3Awereperformedonce.For
statisticalanalysisonFigure3E,percentyieldof6targetgeneswerecompared
againstpercentyieldof6non-targetgenesusingpairedt-testforbothHRP-KDEL
andERM-APEX2.ForcomparisonbetweenERM-APEX2andAPEX2-NES,12target
andnon-targetgeneswerecomparedagainsteachotherusingpairedt-test.
Librarypreparation,sequencing,andquantification
𝜎D = 𝑃×
28
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
PurifiedRNAsamplesweredepletedofribosomalRNAusingtheRibo-ZeroGold
rRNAremovalkit(Illumina),using1µLor2µLofRibo-ZerorRNARemovalSolution
forsampleswith≤16ngor≥50ngtotalinputRNAmass,respectively,in20µLfinal
reactionvolume.ElutedRNAwascleanedupwithAgencourtRNACleanXPbeads
andelutedwith19.5µLofElute,Prime,FragmentmixfromtheTruSeqRNAsample
preparationkit,v2(Illumina).Thereafter,librarieswerepreparedusingtheTruSeq
RNAsamplepreparationkit,accordingtothemanufacturer’sinstructions,starting
from“IncubateRFP”step.Eachlibrarywasgivenauniqueindexduringsynthesis.
LibraryconcentrationandqualitywereconfirmedonanAgilent2100Bioanalyzer,
using“DNAHighSensitivity”kits.
Indexedlibrarieswerepooledinequimolarconcentrations,withnomore
thantenlibrariesperpool,andsubjectedto50cyclesofpairedendsequencing,
followedindexing,ontwolanesofIlluminaHiSeq2500flowcells,runinrapidmode
(GenomicsCore,BroadInstituteofHarvardandMIT).
Ingeneral,theexperimentsforeachconstructwereperformedinthree
biologicalreplicates.Themito-APEXexperimentinFigureS1Bandthemito-APEX2
negativecontrolexperiment(omitH2O2)inFigureS1Fwasperformedintwo
biologicalreplicates.
QuantificationofRNAAbundancesandFoldsEnrichment;AssemblyofTrue
positiveandFalsepositivelists
Deepsequencingreadsweremappedtohumangenomeassemblyhg19andUCSC
knowngenesusingTopHat2,settodefaultoptions(Kimetal.2013).Transcript-and
gene-levelabundanceswerequantifiedandCuffdiff2,settodefaultoptions(Trapnell
etal.2013),and0.01wasaddedtoallquantifiedvaluesforENCODEdata.
Enrichmentanalysis(e.g.Figures1D,3F,S1B,E,F,S2E,F,andS3B),wasrestrictedto
RNAswithFPKMs≥1.0followingstreptavidinpulldown.Foldenrichmentswere
calculatedasfollows:
𝐹𝑃𝐾𝑀D9Z2Z2:[\2]^_`_a[a:_8bc[a2
𝐹𝑜𝑙𝑑𝐸𝑛𝑟𝑖𝑐ℎ𝑚𝑒𝑛𝑡 = 𝑙𝑜𝑔B [
]
𝐹𝑃𝐾𝑀D:[Z2:[\2]^_`_a[a:_8bc[a2
Tocallsignificantlyenrichedgenesfromourdata,ENCODEdata(Dunhamet
al.2012),andER-fractionationdata(ReidandNicchitta2012),thresholdcutoffs
weredeterminedusingReceiverOperatorCharacteristic(ROC)analysis(Fawcett
2006),employingsetsoftrue-positiveandfalse-positivegenesidentifiedas
describedbelow.Ateachfoldenrichmentvalueofthedata,truepositiverate(TPR–
fractionofdetectedtruepositivegenesabovethefoldenrichmentvalue)andfalse
positiverate(FPR–fractionofdetectedfalsepositivegenesabovethefold
enrichmentvalue)arecalculated.ThefoldenrichmentvaluethatmaximizesTPRFPRischosenasthefoldenrichmentcutoff.InmitochondrialandER-associated
APEX-RIPexperiments,ROCanalysiswasbasedonfoldenrichmentvalues;inthe
nuclear-cytoplasmicpartitioningexperiment,itwasbasedoncalculatednuclear
preferencescores.
29
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
Nuclearpreferencescoreswerecalculatedasfollows.Toeachgene,i,we
assignedcoordinates,(x,y)=(log2Ni,log2Ci),whereNiandCidenotetothefoldenrichmentsfromnuclear-(NLS)andcytosolic-(NES)APEX-RIP,respectively.In
thisspace,theliney=x(orrather,log2N=log2C)correspondstoallgenesthatwere
equallyenrichedinbothnuclearandcytosolicAPEX-RIP,andhencedonot
preferentiallyresideineithercompartment.Thenuclearpreferencescoreforgenei
(NPSi),isthereforedefinedastheminimumdistancebetweenitscoordinatesand
thelinelog2N=log2C.Thisisequivalenttocalculatingthedistancebetweenpoints
(x1,y1)=(log2Ni,log2Ci)and(x2,y2)=(0.5(log2Ni+log2Ci),0.5(log2Ni+log2Ci)).Hence:
𝑁𝑃𝑆_ = (𝑙𝑜𝑔B 𝑁_ − 0.5(𝑙𝑜𝑔B 𝑁_ + 𝑙𝑜𝑔B 𝐶_ ))B + (𝑙𝑜𝑔B 𝐶_ − 0.5(𝑙𝑜𝑔B 𝑁_ + 𝑙𝑜𝑔B 𝐶_ ))B …whichreducesto:
1
𝑁𝑃𝑆_ = ( )(𝑙𝑜𝑔B 𝑁_ − 𝑙𝑜𝑔B 𝐶_ )
2
ThetrueandfalsepositivegenesetsneededforROCanalysisweredefinedas
follows:
(1)FormitochondrialAPEX-RIP,truepositivescorrespondedtothethirteen
mitochondrial-encodedmRNAs;falsepositiveRNAscorrespondedtonuclearencodedlongnon-codingRNAs.
(2)Forthenuclearandcytosolicpartitioningexperiment,trueandfalse
positivegenelistswerecompiledusingavailableENCODEhumancellline(NHEKNormalHumanEpidermalKeratinocytes)nuclear–cytoplasmicfractionation
data(Dunhametal.2012).Wecalculatedfold-enrichmentsforRNAsineach
compartment(scaledrelativetothewholecellRNA,FigureS2A),andusedthese
valuestoderiveNuclearPreferenceScores,asdescribedabove.Truepositiveand
truenegativenuclearRNAswerethendefinedasthe1000transcriptswiththe
highestandlowestNPSs,respectively(Figure2B;TableS2tab4).Usingthesegene
liststoperformROCanalysisontheoriginalENCODEdataproducedasignificance
thresholdcutoffatanNPSof1.107(FigureS2B–D),andlistsofthe5467and10130
RNAscalledasbeingenrichedinthenucleusandcytoplasm,respectively(TableS2,
tabs4and5).
(3)ForER-APEX-RIP,mosttruepositivegenesweredefinedusingdatafrom
ER-localizedproximity-dependentribosomeprofiling(Jan,Williams,andWeissman
2014),correspondingtoallRNAswithinputRPKM≥5.0,inputcount≥12,and
log2(foldenrichment)≥0.904(TableS3tab4).Additionaltruepositivegeneswere
predictedbyPhobiusashavingsecretorysignals,andwereabsentin
MitoCarta(Pagliarinietal.2008).(Rheeetal.2013)FalsepositiveRNAsincludedall
geneslackingsecretorysignals,aspredictedbyPhobius,SignalP,andTMHMM.
CoverageandSpecificityanalysisofnuclear,cytosolic,andER-proximalRNAs
Toestimatethecoverage(recall)andspecificityofAPEX-RIPateachsubcellular
compartment,weassembledlistsofestablishedtargetandoff-targetgenestailored
forthatcompartment.
30
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
Foranalysisofthenuclear–cytosolicdatasets(Figure2F),ourreference
nucleargenelistcomprised827lncRNAswithaverageRNApre-enrichment
abundancesof1.0orgreater.Ourreferenceoff-targetlistcomprisedthesetof1260
ER-proximalRNAsdefinedusingproximity-restrictedribosomeprofiling(Jan,
Williams,andWeissman2014).
ForanalysisoftheER-proximaldataset(Figure3I–J),ourreferencegenelist
comprised71mRNAsencodingER-residentproteins(TableS3tab5).Ourreference
off-targetlistcomprised(7589)RNAslackingsecretoryannotation,asassessed
usingPhobius(Käll,Krogh,andSonnhammer2004),TMHMM(Kroghetal.2001),
SignalP(Petersenetal.2011),andwhichlackedtheGOCCterms“Endoplasmic
reticulum,”“Golgi,”“membrane,”and“extracellular”(Ashburneretal.2000).
ForanalysisofcontaminantsinERdatasets(FigureS3E),theRNAsthat
encodeproteinswithnopredictedsecretoryannotationbyPhobius,TMHMM,and
SignalPandlackedGOCCterms“Endoplasmicreticulum,”“Golgi,”“membrane,”and
“extracellular”weresubmittedtoDAVIDBioinformaticsanalysis(Huang,Sherman,
andLempicki2009)tofindGeneontologytermenrichmentagainsthuman
background.Onlytermswithp-valueslessthan0.05wereshown.
31
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
1007
SupplementaryTableS1:Geneticconstructsusedinthisstudy
Name
Mito-V5-APEX
Features
NotI-mito-BamHIV5-APEX-XhoI
Promoter/Vector
CMV/pCDNA3
mito-V5APEX2
Mito-GFP
mito-BamHI-V5APEX2-NheI
NotI-mito-BamHIBFP-XhoI
NotI-V5-APEX2EcoRI-3xNLS-NheI
BstBI-FLAGAPEX2-NES-XhoI
NotI-IgK-HRP-V5KDEL-IRESpuromycin-XbaI
CMV/pLX304
Details
Mitoisa24-aminoacid
mitochondrialtargeting
sequence(MTS)derivedfrom
COX4.
V5:GKPIPNPLLGLDST
CMV/pCDNA3
V5-APEX2-NLS
FLAG-APEX2NES
HRP-V5-KDEL
ERM-APEX2V5
1008
1009
BstBI-ERM-APEX2V5-NheI
CMV/pLX304
CMV
CMV/pLX304
NLS:DPKKKRKV
NES:LQLPPLERLTLD
IgKisN-terminalsignaling
sequencethatbringsprotein
toER
(METDTLLLWVLLLWVPGSTG
D).
KDELisER-retainingsequence
ERMisERmembrane
targetingsequencederived
fromN-terminal27amino
acidsofrabbitP450C1
(MDPVVVLGLCLSCLLLLSLWK
QSYGGG)
SupplementaryTable2.Antibodiesusedforimmunofluorescence
Antibody
Source Company
Catalog
number
Dilution AssociatedFigure
AntiV5
Mouse
R960-25
1:1000
Figure1B,2A,and3A
1:500
1:1000
Figure2A
Figure1B,2A,and3A
AntiFLAG
Mouse
AntiMouseGoat
AlexaFluor488
1010
CMV
Life
Technologies
Agilent
Life
Technologies
200472
A-11029
32
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
1011
SupplementaryTable3.qRT-PCRprimersusedinthisstudy
Primer/probename
Sequence(5’-3’)
MT-ND1forward
MT-ND1reverse
MT-ND2forward
MT-ND2reverse
MT-ND3forward
MT-ND3reverse
MT-ND4forward
MT-ND4reverse
MT-ND4Lforward
MT-ND4Lreverse
MT-ND5forward
MT-ND5reverse
MT-ND6forward
MT-ND6reverse
MT-CYTBforward
MT-CYTBreverse
MT-COX1forward
MT-COX1reverse
MT-COX2forward
MT-COX2reverse
MT-COX3forward
MT-COX3reverse
MT-ATP6forward
MT-ATP6reverse
MT-ATP8forward
MT-ATP8reverse
MT-RNR1forward
MT-RNR1reverse
MT-RNR2forward
MT-RNR2reverse
GAPDHforward
GAPDHreverse
XISTforward
XISTreverse
EMC10forward
EMC10reverse
PCSK1Nforward
PCSK1Nreverse
SSR2forward
SSR2reverse
TMX1forward
TMX1reverse
SFT2D2forward
SFT2D2reverse
EPT1forward
EPT1reverse
DRAP1forward
DRAP1reverse
FAUforward
FAUreverse
CACCTCTAGCCTAGCCGTTT
CCGATCAGGGCGTAGTTTGA
CTTAAACTCCAGCACCACGAC
AGCTTGTTTCAGGTGCGAGA
CCGCGTCCCTTTCTCCATAA
AGGGCTCATGGTAGGGGTAA
ACAACACAATGGGGCTCACT
CCGGTAATGATGTCGGGGTT
TCGCTCACACCTCATATCCTC
AGGCGGCAAAGACTAGTATGG
TCCATTGTCGCATCCACCTT
GGTTGTTTGGGTTGTGGCTC
GGGTTGAGGTCTTGGTGAGT
ACCAATCCTACCTCCATCGC
TCTTGCACGAAACGGGATCA
CGAGGGCGTCTTTGATTGTG
TCCTTATTCGAGCCGAGCTG
ACAAATGCATGGGCTGTGAC
AACCAAACCACTTTCACCGC
CGATGGGCATGAAACTGTGG
CTAATGACCTCCGGCCTAGC
AGGCCTAGTATGAGGAGCGT
TTCGCTTCATTCATTGCCCC
GGGTGGTGATTAGTCGGTTGT
ACTACCACCTACCTCCCTCAC
GGCAATGAATGAAGCGAACAGA
CATCCCCGTTCCAGTGAGTT
TGGCTAGGCTAAGCGTTTTGA
CAGCCGCTATTAAAGGTTCGT
AAGGCGCTTTGTGAAGTAGG
TTCGACAGTCAGCCGCATCTTCTT
GCCCAATACGACCAAATCCGTTGA
CCCTACTAGCTCCTCGGACA
ACACATGCAGCGTGGTATCT
TTCATTGAGCGCCTGGAGAT
TTCATTGAGCGCCTGGAGAT
GAGACACCCGACGTGGAC
AATCCGTCCCAGCAAGTACC
GTTTGGGATGCCAACGATGAG
CTCCACGGCGTATCTGTTCA
ACGGACGAGAACTGGAGAGA
ATTTTGACAAGCAGGGCACC
CCATCTTCCTCATGGGACCAG
GCAGAACACAGGGTAAGTGC
TGGCTTTCTGCTGGTCGTAT
AATCCAAACCCAGTCAGGCA
ACATCCCACCTGAAGCAGTG
GATGCCACCAGGTCCTTCAA
TCCTAAGGTGGCCAAACAGG
GTGGGCACAACGTTGACAAA
33
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
SUB1forward
SUB1reverse
LSM6forward
LSM6reverse
COPS2forward
COPS2reverse
CGGBP1forward
CGGBP1reverse
BCA53forward
BCA53reverse
CEP128forward
CEP128reverse
MAD1L1forward
MAD1L1reverse
RAD51Bforward
RAD51Breverse
RBPMSforward
RBPMSreverse
TCF7forward
TCF7reverse
HOOK2forward
HOOK2reverse
MAN2C1forward
MAN2C1reverse
1012
CGTCACTTCCGGTTCTCTGT
TGATTTAGGCATCGCTTCGC
CGGACGACCAGTTGTGGTAA
CCAGGACCCCTCGATAATCC
AGGAGGACTACGACCTGGAAT
GCCGCTTTTGGGTCATCTTC
GCCTCGTCCACTTTCCCTAA
TCATGCCTTTACGTAGGATCGAG
TCTTGCCTGCTCCACAGTTT
CAAACACCAAGGAGGGGTCT
TACAGTAATGGACAGGCGGG
TCCGGAGTTGGTCGATTGAT
CGAGTCTGCCATCGTCCAA
GCACTCTCCACCTGCTTCTT
TTTGGACGAAGCCCTGCAT
CACAACCTGGTGGACCTGTA
ACAGTCGCTCAGAAGCAGAG
CGAAGCGGATGCCATTCAAA
TCAACAGCCCACATCCCAC
AGAGGCCTGTGAACTTGCTT
TTTGCTGAAAAGGAAGCTGGA
GCAACTCCAGATCTGCCTCA
ATGAGGCCCACAAGTTCCTG
TCTCATAGGTGGCCTGGGAA
34
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
References
Alán,Lukáš,JaroslavZelenka,JanJežek,AndreaDlasková,andPetrJežek.2010.
“FluorescentinSituHybridizationofMitochondrialDNAandRNA.”Acta
BiochimicaPolonica57(4):403–8.
http://www.ncbi.nlm.nih.gov/pubmed/21125028.
Ashburner,Michael,CatherineA.Ball,JudithA.Blake,DavidBotstein,Heather
Butler,J.MichaelCherry,AllanP.Davis,etal.2000.“GeneOntology:Toolforthe
UnificationofBiology.”NatureGenetics25(1):25–29.doi:10.1038/75556.
BaharHalpern,Keren,InbalCaspi,DoronLemze,MaayanLevy,ShanieLanden,Eran
Elinav,IgorUlitsky,andShalevItzkovitz.2015.“NuclearRetentionofmRNAin
MammalianTissues.”CellReports13(12):2653–62.
doi:10.1016/j.celrep.2015.11.036.
Batish,Mona,ArjunRaj,andSanjayTyagi.2011.“SingleMoleculeImagingofRNAIn
Situ.”InRNADetectionandVisualization,3–13.doi:10.1007/978-1-61779-0058_1.
Cabili,MoranN,MargaretCDunagin,PatrickDMcClanahan,AndrewBiaesch,Olivia
Padovan-Merhar,AvivRegev,JohnLRinn,andArjunRaj.2015.“Localization
andAbundanceAnalysisofHumanlncRNAsatSingle-CellandSingle-Molecule
Resolution.”GenomeBiology16(1):20.doi:10.1186/s13059-015-0586-4.
Chapman-Smith,A,andJECronan.1999.“MolecularBiologyofBiotinAttachmentto
Proteins.”TheJournalofNutrition129(2SSuppl):477S–484S.
http://www.ncbi.nlm.nih.gov/pubmed/10064313.
Chen,Chun-Kan,MarioBlanco,ConstanzaJackson,ErikAznauryan,NoahOllikainen,
ChristineSurka,AmyChow,AndreaCerase,PatrickMcDonel,andMitchell
Guttman.2016.“XistRecruitstheXChromosometotheNuclearLaminato
EnableChromosome-WideSilencing.”Science(NewYork,N.Y.),August.
doi:10.1126/science.aae0047.
Chen,K.H.,A.N.Boettiger,J.R.Moffitt,S.Wang,andX.Zhuang.2015.“Spatially
Resolved,HighlyMultiplexedRNAProfilinginSingleCells.”Science348(6233):
aaa6090-aaa6090.doi:10.1126/science.aaa6090.
Chi,SungWook,JulieB.Zang,AldoMele,andRobertB.Darnell.2009.“Argonaute
HITS-CLIPDecodesmicroRNA–mRNAInteractionMaps.”Nature,June.
doi:10.1038/nature08170.
Csordás,György,ChristianRenken,PéterVárnai,LudivineWalter,DavidWeaver,
KarolynF.Buttle,TamásBalla,CarmenA.Mannella,andGyörgyHajnóczky.
2006.“StructuralandFunctionalFeaturesandSignificanceofthePhysical
LinkagebetweenERandMitochondria.”TheJournalofCellBiology174(7):
915–21.doi:10.1083/jcb.200604016.
Damas,NkeroremaDjodji,MichelaMarcatti,ChristopheCôme,LiseLotte
Christensen,MortenMuhligNielsen,RolandBaumgartner,HeleneMaria
Gylling,etal.2016.“SNHG5PromotesColorectalCancerCellSurvivalby
CounteractingSTAU1-MediatedmRNADestabilization.”Nature
Communications7(December):13875.doi:10.1038/ncomms13875.
Derrien,T.,R.Johnson,G.Bussotti,A.Tanzer,S.Djebali,H.Tilgner,G.Guernec,etal.
2012.“TheGENCODEv7CatalogofHumanLongNoncodingRNAs:Analysisof
TheirGeneStructure,Evolution,andExpression.”GenomeResearch22(9):
35
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1775–89.doi:10.1101/gr.132159.111.
Dunham,Ian,AnshulKundaje,ShelleyF.Aldred,PatrickJ.Collins,CarrieA.Davis,
FrancisDoyle,CharlesB.Epstein,etal.2012.“AnIntegratedEncyclopediaof
DNAElementsintheHumanGenome.”Nature489(7414):57–74.
doi:10.1038/nature11247.
Engreitz,JesseM.,NoahOllikainen,andMitchellGuttman.2016.“LongNon-Coding
RNAs:SpatialAmplifiersThatControlNuclearStructureandGeneExpression.”
NatureReviewsMolecularCellBiology17(12):756–70.
doi:10.1038/nrm.2016.126.
Fawcett,Tom.2006.“AnIntroductiontoROCAnalysis.”PatternRecognitionLetters
27(8):861–74.doi:10.1016/j.patrec.2005.10.010.
Fox,CH,FBJohnson,JWhiting,andPPRoller.1985.“FormaldehydeFixation.”
JournalofHistochemistry&Cytochemistry33(8):845–53.
doi:10.1177/33.8.3894502.
Gilbert,Chris,andJesperQSvejstrup.2006.“RNAImmunoprecipitationfor
DeterminingRNA-ProteinAssociationsinVivo.”CurrentProtocolsinMolecular
Biology/EditedbyFrederickM.Ausubel...[etAl.]Chapter27(August):Unit
27.4.doi:10.1002/0471142727.mb2704s75.
Gilbert,Christopher,ArnoldKristjuhan,GSebastiaanWinkler,andJesperQ
Svejstrup.2004.“ElongatorInteractionswithNascentmRNARevealedbyRNA
Immunoprecipitation.”MolecularCell14(4):457–64.
http://www.ncbi.nlm.nih.gov/pubmed/15149595.
Han,Shuo,NamrataDUdeshi,ThomasJDeerinck,TanyaSvinkina,MarkHEllisman,
StevenACarr,andAliceYTing.2017.“ProximityBiotinylationasaMethodfor
MappingProteinsAssociatedwithmtDNAinLivingCells.”CellChemicalBiology
24(3):404–14.doi:10.1016/j.chembiol.2017.02.002.
Hendrickson,DavidG,DavidRKelley,DanielleTenen,BradleyBernstein,andJohnL
Rinn.2016.“WidespreadRNABindingbyChromatin-AssociatedProteins.”
GenomeBiology17(February):28.doi:10.1186/s13059-016-0878-3.
Hocine,Sami,PascalRaymond,DanielZenklusen,JeffreyAChao,andRobertH
Singer.2013.“Single-MoleculeAnalysisofGeneExpressionUsingTwo-Color
RNALabelinginLiveYeast.”NatureMethods10(2):119–21.
doi:10.1038/nmeth.2305.
Huang,DaWei,BradTSherman,andRichardALempicki.2009.“Systematicand
IntegrativeAnalysisofLargeGeneListsUsingDAVIDBioinformatics
Resources.”NatureProtocols4(1):44–57.doi:10.1038/nprot.2008.211.
Hung,Victoria,StephanieSLam,NamrataDUdeshi,TanyaSvinkina,GaelenGuzman,
VamsiKMootha,StevenACarr,andAliceYTing.2017.“ProteomicMappingof
Cytosol-FacingOuterMitochondrialandERMembranesinLivingHumanCells
byProximityBiotinylation.”eLife6(April).doi:10.7554/eLife.24463.
Hung,Victoria,PengZou,Hyun-WooRhee,NamrataD.Udeshi,ValentinCracan,
TanyaSvinkina,StevenA.Carr,VamsiK.Mootha,andAliceY.Ting.2014.
“ProteomicMappingoftheHumanMitochondrialIntermembraneSpaceinLive
CellsviaRatiometricAPEXTagging.”MolecularCell55(2):332–41.
doi:10.1016/j.molcel.2014.06.003.
Ingolia,NicholasT,SinaGhaemmaghami,JohnRSNewman,andJonathanS
36
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
Weissman.2009.“Genome-WideAnalysisinVivoofTranslationwith
NucleotideResolutionUsingRibosomeProfiling.”Science(NewYork,N.Y.)324
(5924):218–23.doi:10.1126/science.1168978.
Jan,CalvinH,ChristopherCWilliams,andJonathanSWeissman.2014.“Principlesof
ERCotranslationalTranslocationRevealedbyProximity-SpecificRibosome
Profiling.”Science(NewYork,N.Y.)346(6210):1257521.
doi:10.1126/science.1257521.
Jung,Hosung,ChristosG.Gkogkas,NahumSonenberg,andChristineE.Holt.2014.
“RemoteControlofGeneFunctionbyLocalTranslation.”Cell157(1):26–40.
doi:10.1016/j.cell.2014.03.005.
Käll,Lukas,AndersKrogh,andErikLLSonnhammer.2004.“ACombined
TransmembraneTopologyandSignalPeptidePredictionMethod.”Journalof
MolecularBiology338(5):1027–36.doi:10.1016/j.jmb.2004.03.016.
Kim,Daehwan,GeoPertea,ColeTrapnell,HaroldPimentel,RyanKelley,andSteven
LSalzberg.2013.“TopHat2:AccurateAlignmentofTranscriptomesinthe
PresenceofInsertions,DeletionsandGeneFusions.”GenomeBiology14(4):
R36.doi:10.1186/gb-2013-14-4-r36.
Kind,Jop,andBasvanSteensel.2010.“Genome–nuclearLaminaInteractionsand
GeneRegulation.”CurrentOpinioninCellBiology22(3):320–25.
doi:10.1016/j.ceb.2010.04.002.
Krogh,A,BLarsson,GvonHeijne,andELSonnhammer.2001.“Predicting
TransmembraneProteinTopologywithaHiddenMarkovModel:Applicationto
CompleteGenomes.”JournalofMolecularBiology305(3):567–80.
doi:10.1006/jmbi.2000.4315.
Lam,StephanieS,JeffreyDMartell,KimberliJKamer,ThomasJDeerinck,MarkH
Ellisman,VamsiKMootha,andAliceYTing.2014.“DirectedEvolutionofAPEX2
forElectronMicroscopyandProximityLabeling.”NatureMethods12(1):51–
54.doi:10.1038/nmeth.3179.
Lee,JeHyuk,EvanRDaugharthy,JonathanScheiman,RezaKalhor,RyojiAmamoto,
DerekTPeters,BrianMTurczyk,etal.2014.“HighlyMultiplexedSubcellular
RNASequencinginSitu.”Science(NewYork,N.Y.)343(March):1360–63.
doi:10.1126/science.1250212.
Lesnik,Chen,AdiGolani-Armon,andYoavArava.2015.“LocalizedTranslationnear
theMitochondrialOuterMembrane:AnUpdate.”RNABiology12(8):801–9.
doi:10.1080/15476286.2015.1058686.
Lobingier,BradenT.,RuthHüttenhain,KelsieEichel,KennethB.Miller,AliceY.Ting,
MarkvonZastrow,andNevanJ.Krogan.2017.“AnApproachto
SpatiotemporallyResolveProteinInteractionNetworksinLivingCells.”Cell
169(2):350–360.e12.doi:10.1016/j.cell.2017.03.022.
Loh,KenH.,PhilippS.Stawski,AustinS.Draycott,NamrataD.Udeshi,EmilyK.
Lehrman,DanielK.Wilton,TanyaSvinkina,etal.2016.“ProteomicAnalysisof
UnboundedCellularCompartments:SynapticClefts.”Cell166(5):1295–1307.
doi:10.1016/j.cell.2016.07.041.
Martell,JeffreyD,ThomasJDeerinck,YaseminSancak,ThomasLPoulos,VamsiK
Mootha,GinaESosinsky,MarkHEllisman,andAliceYTing.2012.“Engineered
AscorbatePeroxidaseasaGeneticallyEncodedReporterforElectron
37
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
Microscopy.”NatureBiotechnology30(11).NaturePublishingGroup:1143–48.
doi:10.1038/nbt.2375.
Martell,JeffreyD,MasahitoYamagata,ThomasJDeerinck,SébastienPhan,CarolynG
Kwa,MarkHEllisman,JoshuaRSanes,andAliceYTing.2016.“ASplit
HorseradishPeroxidasefortheDetectionofIntercellularProtein–protein
InteractionsandSensitiveVisualizationofSynapses.”NatureBiotechnology34
(7):774–80.doi:10.1038/nbt.3563.
Mercer,TimR,ShaneNeph,MarcelEDinger,JoannaCrawford,MartinaSmith,
Anne-MarieJShearwood,EricHaugen,etal.2011.“TheHumanMitochondrial
Transcriptome.”Cell146(4).ElsevierInc.:645–58.
doi:10.1016/j.cell.2011.06.051.
Mick,DavidU,RachelBRodrigues,RyanDLeib,ChristopherMAdams,AllisSChien,
StevenPGygi,andMaxenceVNachury.2015.“ProteomicsofPrimaryCiliaby
ProximityLabeling.”DevelopmentalCell35(4):497–512.
doi:10.1016/j.devcel.2015.10.015.
Motamedi,MohammadR,AndréVerdel,SerafinUColmenares,ScottAGerber,
StevenPGygi,andDaneshMoazed.2004.“TwoRNAiComplexes,RITSand
RDRC,PhysicallyInteractandLocalizetoNoncodingCentromericRNAs.”Cell
119(6):789–802.doi:10.1016/j.cell.2004.11.034.
Nagao,Asuteka,NarumiHino-Shigi,andTsutomuSuzuki.2008.“Chapter23
MeasuringmRNADecayinHumanMitochondria.”In,489–99.
doi:10.1016/S0076-6879(08)02223-4.
Nelles,DavidA,MarkYFang,MitchellRO’Connell,JiaLXu,SebastianJMarkmiller,
JenniferADoudna,andGeneWYeo.2016.“ProgrammableRNATrackingin
LiveCellswithCRISPR/Cas9.”Cell165(2):488–96.
doi:10.1016/j.cell.2016.02.054.
Paek,Jaeho,MarianKalocsay,DeanP.Staus,LauraWingler,RobertaPascolutti,Joao
A.Paulo,StevenP.Gygi,andAndrewC.Kruse.2017.“Multidimensional
TrackingofGPCRSignalingviaPeroxidase-CatalyzedProximityLabeling.”Cell
169(2):338–349.e11.doi:10.1016/j.cell.2017.03.028.
Pagliarini,DavidJ.,SarahE.Calvo,BettyChang,SunilA.Sheth,ScottB.Vafai,Shao-En
Ong,GeoffreyA.Walford,etal.2008.“AMitochondrialProteinCompendium
ElucidatesComplexIDiseaseBiology.”Cell134(1):112–23.
doi:10.1016/j.cell.2008.06.016.
Paige,J.S.,K.Y.Wu,andS.R.Jaffrey.2011.“RNAMimicsofGreenFluorescent
Protein.”Science333(6042):642–46.doi:10.1126/science.1207339.
Petersen,ThomasNordahl,SørenBrunak,GunnarvonHeijne,andHenrikNielsen.
2011.“SignalP4.0:DiscriminatingSignalPeptidesfromTransmembrane
Regions.”NatureMethods8(10):785–86.doi:10.1038/nmeth.1701.
Piechota,Janusz,RafałTomecki,KamilGewartowski,RomanSzczesny,Aleksandra
Dmochowska,MarekKudła,LienDybczyńska,PiotrPStepien,andEwaBartnik.
2006.“DifferentialStabilityofMitochondrialmRNAinHeLaCells.”Acta
BiochimicaPolonica53(1):157–68.
Reid,DavidW,andChristopherVNicchitta.2012.“PrimaryRoleforEndoplasmic
Reticulum-BoundRibosomesinCellularTranslationIdentifiedbyRibosome
Profiling.”TheJournalofBiologicalChemistry287(8):5518–27.
38
bioRxiv preprint first posted online Jun. 21, 2017; doi: http://dx.doi.org/10.1101/153098. The copyright holder for this preprint (which was not
peer-reviewed) is the author/funder. It is made available under a CC-BY 4.0 International license.
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
doi:10.1074/jbc.M111.312280.
Rhee,Hyun-Woo,PengZou,NamrataDUdeshi,JeffreyDMartell,VamsiKMootha,
StevenaCarr,andAliceYTing.2013.“ProteomicMappingofMitochondriain
LivingCellsviaSpatiallyRestrictedEnzymaticTagging.”Science(NewYork,
N.Y.)339(6125):1328–31.doi:10.1126/science.1230593.
Rinn,J.,andM.Guttman.2014.“RNAandDynamicNuclearOrganization.”Science
345(6202):1240–41.doi:10.1126/science.1252966.
Ro,Seungil,Hsiu-YenMa,ChanjaePark,NicoleOrtogero,RuiSong,GrantWHennig,
HuiliZheng,etal.2013.“TheMitochondrialGenomeEncodesAbundantSmall
NoncodingRNAs.”CellResearch23(6):759–74.doi:10.1038/cr.2013.37.
Roux,KyleJ,DaeInKim,ManfredRaida,andBrianBurke.2012.“APromiscuous
BiotinLigaseFusionProteinIdentifiesProximalandInteractingProteinsin
MammalianCells.”TheJournalofCellBiology196(6):801–10.
doi:10.1083/jcb.201112098.
Shechner,DavidM,EzgiHacisuleyman,ScottTYounger,andJohnLRinn.2015.
“Multiplexable,Locus-SpecificTargetingofLongRNAswithCRISPR-Display.”
NatureMethods12(7):664–70.doi:10.1038/nmeth.3433.
Sterne-Weiler,T.,R.T.Martinez-Nunez,J.M.Howard,I.Cvitovik,S.Katzman,M.A.
Tariq,N.Pourmand,andJ.R.Sanford.2013.“Frac-SeqRevealsIsoform-Specific
RecruitmenttoPolyribosomes.”GenomeResearch23(10):1615–23.
doi:10.1101/gr.148585.112.
Trapnell,Cole,DavidGHendrickson,MartinSauvageau,LoyalGoff,JohnLRinn,and
LiorPachter.2013.“DifferentialAnalysisofGeneRegulationatTranscript
ResolutionwithRNA-Seq.”NatureBiotechnology31(1):46–53.
doi:10.1038/nbt.2450.
Ule,Jernej,KirkBJensen,MatteoRuggiu,AldoMele,AljazUle,andRobertBDarnell.
2003.“CLIPIdentifiesNova-RegulatedRNANetworksintheBrain.”Science
(NewYork,N.Y.)302(5648):1212–15.doi:10.1126/science.1090095.
Wilk,Ronit,JackHu,DmitryBlotsky,andHenryMKrause.2016.“Diverseand
PervasiveSubcellularDistributionsforBothCodingandLongNoncoding
RNAs.”Genes&Development30(5):594–609.doi:10.1101/gad.276931.115.
Williams,ChristopherC,CalvinHJan,andJonathanSWeissman.2014.“Targeting
andPlasticityofMitochondrialProteinsRevealedbyProximity-Specific
RibosomeProfiling.”Science(NewYork,N.Y.)346(6210):748–51.
doi:10.1126/science.1257522.
Wishart,JF,andBSMadhavaRao.2010.RecentTrendsinRadiationChemistry.
Singapore:WorldScientific.
Zhao,Sheng,andRussellDFernald.2005.“ComprehensiveAlgorithmfor
QuantitativeReal-TimePolymeraseChainReaction.”JournalofComputational
Biology :AJournalofComputationalMolecularCellBiology12(8):1047–64.
doi:10.1089/cmb.2005.12.1047.
39