Emerging complexity of the HuD/ELAVl4 gene; implications for neuronal development, function, and dysfunction.

2013·0.47 MB·English

by  Bronicki, Lucas M.

REVIEW
Emerging complexity of the HuD/ELAVl4 gene; implications
for neuronal development, function, and dysfunction
LUCASM. BRONICKI and BERNARDJ. JASMIN1
DepartmentofCellularandMolecularMedicine,FacultyofMedicine,UniversityofOttawa,Ottawa,Ontario,CanadaK1H8M5
ABSTRACT
Precise control of messenger RNA (mRNA) processing and abundance areincreasingly being recognized as critical for proper
spatiotemporal gene expression, particularly in neurons. These regulatory events are governed by a large number of trans-
actingfactorsfoundinneurons,mostnotablyRNA-bindingproteins(RBPs)andmicro-RNAs(miRs),whichbindtospecificcis-
acting elements or structures within mRNAs. Through this binding mechanism, trans-acting factors, particularly RBPs, control
all aspects of mRNA metabolism, ranging from altering the transcription rate to mediating mRNA degradation. In this context
the best-characterized neuronal RBP, the Hu/ELAVl family member HuD, is emerging as a key component in multiple
regulatory processes—including pre-mRNA processing, mRNA stability, and translation—governing the fate of a substantial
amount of neuronal mRNAs. Through its ability to regulate mRNA metabolism of diverse groups of functionallysimilar genes,
HuDplaysimportant roles inneuronal developmentandfunction. Furthermore, compelling evidenceindicates supplementary
roles for HuD in neuronal plasticity, in particular, recovery from axonal injury, learning and memory, and multiple
neurological diseases. The purpose of this review is to provide a detailed overview of the current knowledge surrounding the
expression and roles of HuD in the nervous system. Additionally, we outline the present understanding of the molecular
mechanismspresidingoverthelocalization,abundance,andfunctionofHuDinneurons.
Keywords: HuD/ELAVl4;RNA-bindingprotein;mRNAmetabolism;post-transcriptionalregulation
INTRODUCTION post-transcriptional stages, thereby effectively coupling dis-
tinct regulatory events during mRNA metabolism (Dahan
TheflowofgeneticinformationfromDNAtoproteinrequires
etal.2011).ThecoordinationofgroupsofmRNAsalongsuc-
numerousstagesofintricatecontrol,withamyriadofregula-
cessiveprocessingandregulatoryeventsformedthepremiseof
toryeventsoccurringbetweentranscriptionandtranslation.
theRNAoperontheory,whichhelpstoclarifytheorganiza-
Theseregulatoryeventsareorchestratedbytrans-actingfac-
tion and dynamics of post-transcriptional networks (Keene
tors,agroupofregulatorymoleculescomprisingRNA-bind-
2007).Thesenetworks,collectivelyknownastheribonome,
ingproteins(RBPs)andmultipleclassesofnoncodingRNAs
confernumerousbenefitstocellsincludingincreasingprotein
(ncRNAs),thataggregateontomRNAsandformmessenger
diversityfromafixednumberofgenes,accuratespatialand
ribonucleoprotein (mRNP) complexes (Lukong et al. 2008;
temporalgeneexpression,andrapidmodificationofprotein
Carninci2010).Thepasttwodecadeshaveseentremendous
levels(MansfieldandKeene2009).
advancesinthediscoveryofnoveltypesoftrans-actingfactors
The advantages convened by post-transcriptional regu-
andprovidedagreaterunderstandingoftheirdiverseandvital
lation are particularly necessary in neurons since these cells
rolesinmRNAmetabolism.Bybindingtocis-actingelements
containcomplexcellulararchitectureduetoextensiveneurite
and/or structures in mRNAs, trans-acting factors control a
branching. In addition to playing critical roles in neuronal
broadspectrumofprocessesthatrangefromalternativesplic-
development, maintenance, and function, the importance
ingandpolyadenylationinthenucleustolocalization,stabili-
oftheseregulatoryeventsisfurtherhighlightedbythegrow-
zation,andtranslationinthecytoplasm(CarmodyandWente
ingnumberofdiseases associatedwithmutationsofcertain
2009; Licatalosi and Darnell 2010; Vazquez-Pianzola and
mRNAsordysregulationofRBPsorncRNAs,notablysmall
Suter2012;WuandBrewer2012).Sometrans-actingfactors,
ncRNAs termed micro-RNAs (miRs), that associate with
in particular, RBPs, have been shown to transition mRNAs
them(Lukongetal.2008;Cooperetal.2009;Esteller2011;
encoding proteins with similar functions through different
Im and Kenny 2012). For example, defects affecting the
normal expression of mRNAs and/or function of RBPs are
1Correspondingauthor centralintheetiologyandpathogenesisofabroadspectrum
[email protected]
of neuronal diseases, including the neurodegenerative dis-
Article is online at http://www.rnajournal.org/cgi/doi/10.1261/rna.
039164.113. ordersfragileXsyndrome,fragileXtremorataxiasyndrome
RNA19:1019–1037;©2013;PublishedbyColdSpringHarborLaboratoryPressfortheRNASociety 1019
BronickiandJasmin
(LiandJin2012),spinalmuscularatrophy
TABLE 1. Select AU-rich element targeting RNA-binding proteins in neurons and their
(SMA)(Fallinietal.2012b),Huntington’s
effectsonmRNAmetabolism
disease (Krzyzosiak et al. 2012), and
RNA-bindingprotein EffectonmRNA
amyotrophiclateralsclerosis(ALS)(Fie-
sel and Kahle 2011; Strong and Vol- HuB/ELAVl2,HuC/ELAVl3, Localtranscriptionrateenhancement(Zhouetal.2011)
kening2011). HuD/ELAVl4 Alternativesplicing(Zhuetal.2006,2008)
The nervous system encompasses an Alternativepolyadenylation(Zhuetal.2006)
Nuclearexport(Kasashimaetal.1999)
extensiveamountofRBPs,manyofwhich
Stabilization(Jainetal.1997;Andersonetal.2000;
arecriticalforneuronaldevelopment.In
Mobaraketal.2000)
support of this idea, one genome-wide Cytoplasmictransportation(Aronovetal.2002)
screenfoundthat85%ofthe380RBPsex- Translationrepression(Kullmannetal.2002)
aminedwereexpressedintheembryonic Translationenhancement(Anticetal.1999;Kullmann
etal.2002;Fukaoetal.2009)
(E13.5) and early postnatal (P0) mouse
HuR/ELAVl1 Localtranscriptionrateenhancement(Zhouetal.2011)
brain(McKeeetal.2005).Mostnotable
Alternativesplicing(Izquierdo2008)
arethewell-describedRBPsthatregulate Alternativepolyadenylation(Zhuetal.2006)
variousstagesofmRNAmetabolismand Nuclearexport(FanandSteitz1998)
areimportantforproperneuronaldiffer- Stabilization(FanandSteitz1998;Levyetal.1998;Peng
etal.1998)
entiationand/or function,such as buty-
Destabilization(Kimetal.2009)
rate-response factor 1 (BRF1) (Sanduja
Translationrepression(Kullmannetal.2002)
etal.2011),neuro-oncologicventralanti- Translationenhancement(Mazan-Mamczarzetal.2003;
gens 1/2 (NOVA 1/2) (Darnell 2006), A Mengetal.2005)
+U binding factor 1 (AUF1) (Gratacos AU-richelementbinding Mostlydestabilization(BrewerandRoss1989;Brewer
factor1(AUF1;p37,p40, 1991;Zhangetal.1993)
and Brewer 2010), tristetraprolin (TTP)
p42,andp45)/hnRNPD Translationenhancement(Liaoetal.2007)
(Sandujaetal.2011),Khomologysplic-
Khomology-typesplicing Transcriptionenhancement(Davis-Smythetal.1996)
ing regulatory protein (KSRP) (Briata regulatoryprotein(KSRP) Alternativesplicing(Minetal.1997)
etal.2012),andHu/embryoniclethalab- Destabilization(Chenetal.2001;Gherzietal.2004)
normal vision-like (ELAVl) (Table 1; Cytoplasmictransport(Sneeetal.2002)
Tristetraprolin(TTP) Transcriptionrepression(Liangetal.2009;Schichletal.
Pascale et al. 2008). The majority of
2009)
RBPscanbeclusteredaccordingtotheir
Destabilization(Carballoetal.1998)
similar molecular functions, such as Translationrepression(PfeifferandBrooks2012;Qietal.
those belonging to the translation and 2012;Tiedjeetal.2012)
turnover regulatory (TTR)-RBP group. Butyrate-responsefactor-1 3′Endprocessinginhibition(Desroches-Castanetal.
(BRF1) 2011)
Althoughnotlimitedtothesefunctions,
Destabilization(Stoecklinetal.2002)
TTR-RBPsareinvolvedinstabilizing/de-
TINO/Mex3D/RKHD1 Destabilization(Donninietal.2004)
stabilizing mRNAs and promoting/ Polyadenylate-bindingprotein- Stabilization(Onestoetal.2004)
inhibitingtranslation,andcomprisevari- interactingprotein2(PAIP2) Translationrepression(Khaleghpouretal.2001)
ous members, including TTP, KSRP, CUG-bindingprotein1 Alternativesplicing(Philipsetal.1998)
(CUG-BP1) Destabilization(Moraesetal.2006)
AUF, and Hu/ELAVl proteins (Table 1;
Translationenhancement(Timchenkoetal.1999)
Pullmann et al. 2007). Among all the
CUG-BP2 Alternativesplicing(Zhangetal.2002)
RBPsexpressedinneurons,thebestchar- RNAediting(Anantetal.2001)
acterized is the predominantly neuron- Stabilization(Mukhopadhyayetal.2003)
specific Hu member HuD. This review Translationrepression(Mukhopadhyayetal.2003)
Nucleolin Transcriptionrepression(Yangetal.1994)
will focus on the multifunctional roles
Stabilization(Chenetal.2000)
anddiverseregulatoryeventscontrolling
Destabilization(ZaidiandMalter1995)
theexpressionofHuDinneurons. Translationrepression(Takagietal.2005)
Translationenhancement(Fahlingetal.2005)
THE HuD GENE, mRNA, AND
PROTEIN
RNA-binding protein 9 (rbp9) genes. These four Drosophila
ThenameofHu/ELAVlgenesoriginatedfromthefirstinitials genesencodeRBPsthatdisplaysimilarstructuresandoverlap-
(Hu)ofthepatientinwhichtheywereidentifiedandtheclose pingfunctionsinneuronstoHuD(Belletal.1988;Robinow
relation of these genes to the Drosophila embryonic lethal etal.1988;KimandBaker1993;SamsonandChalvet2003;
abnormal vision (elav) gene (Graus et al. 1987; Okano and Colombritaetal.2013).Inmammals,Huproteinswerefirst
Darnell1997).Inaddition,Hugenesarealsocloselyrelated discoveredasantigenstargetedbyauto-antibodiesinpatients
totheDrosophilasexlethal(sxl),foundinneurons(fne),and with the neurological syndrome paraneoplastic encephalo-
1020 RNA,Vol.19,No.8
Expression,function,andregulationofHuD
myelitis and sensory neuropathy (PEM/SN) (Dalmau et al. A
~2.3kb ~23kb ~13kb ~1.6kb
1990, 1992). This syndrome arises in a small subset of
a a1 a2 a3a4 b c c1 2 3 4 5 67 8
patients when theectopicexpression ofHu proteins in cer-
~59kb~1.8kb~0.4kb~0.3kb~0.4kb ~12kb ~25kb ~1.5kb ~3kb
tain types of cancers, typically small cell lung carcinomas B
(SCLCs), provokes an autoimmune attack on the nervous a
system. a1 67
a2
OutofthefourHumembers,HuDwasthefirsttobecloned
a3 2 3 4 5 6 8 AAAAA
and characterized, followed by HuC/ELAVl3, HuB/ELAVl2, a4
andHuR/ELAVl1 (Szaboetal.1991;King etal.1994;Sakai b
c
etal.1994;Maetal.1996).TheexpressionofthreeHumem-
c c1
bers—HuB, HuC and HuD—is predominantly restricted
toneurons(nELAVs),whileHuRisfoundinmostcelltypes C
PP P P P PPMe P P
(Szabo et al. 1991; King et al. 1994; Good 1995; Ma et al.
RRM1 RRM2 67 RRM3 HuDpro
1996;OkanoandDarnell1997).Inaddition, nELAVsshare
a higher degree of amino acid sequence identity (>80%) RRM1 RRM2 6 RRM3 HuD
with each other compared to HuR (72.5%–73.6%) (Okano
RRM1 RRM2 RRM3 HuDmex
andDarnell1997).Despitethehighdegreeofsequenceiden-
tity between nELAVs, variation of amino acid composition FIGURE1. OrganizationofthemouseHuDgene,mRNA,andprotein.
inkeyregions,particularlytheN-terminalandlinkerregions (A)MouseHuDgeneorganizationwithnoncodingexonsdepictedas
(seebelow),andspatiotemporalexpressionsuggestthatHuD orangerectangles,codingexonsasgrayandgreenrectangles,andin-
trons as black lines. (B) Alternative splicing/processing of HuD pre-
carriesoutsomeuniquerolesandissubjectedtopartiallydis- mRNAproducesdifferentHuDisoforms.5′/3′UTRsandcodingexons
tinctregulation(OkanoandDarnell1997;Claytonetal.1998; arerepresentedbyorangeandgrayrectangles,respectively.Greenrect-
Hambardzumyanetal.2009). anglesdenotealternativelysplicedexonsnearthe3′endofthegene.Also
Overall,theHuDgeneiswellconservedamonghigherver- shown are the approximate locations of the Hu/ELAVl (white star),
miR-375(blackstar),andputativeARE(graystars)bindingsites.(C)
tebratesandislocatedonchromosome1inhumansandchro-
ThethreemajorHuDproteinvariantsfoundinneuronswithposition
mosome 4 in mice (Fig. 1A). The human and mouse HuD ofRRMs(gray)andaminoacidextensionsinthelinkerregion(green)
genespans∼146kbofDNAandisdividedintosevencoding shown.“P”and“Me”indicateapproximatelocationsofserine/threo-
exons(E2toE8)thatencompass∼44kbofDNA(Sekidoetal. nine and arginine residues that are subjected to phosphorylation by
′ PKCandmethylationbyCARM1,respectively.
1994; Inman et al. 1998). For several years, the 5 region of
HuDwasthoughttocontainthreepresumablyuntranslated
exon1variants(termedE1a,E1b,andE1c)thatarealterna- andE7encodethelinkerregion(Inmanetal.1998).TheHuD
tively spliced to the 5′ end of exon 2 (Inman et al. 1998). linkerregioncontainsmultipleresiduesthatarepost-transla-
Recent findings in our laboratory, however, have identified tionally modified and housesthe nuclear export (NES) and
the expression of five additional HuD E1 variants, resulting putativenuclearlocalization(NLS)signals(Kasashimaetal.
in a total of eight (Fig. 1A,B; Bronicki et al. 2012). We also 1999).
found that most E1 variants house an in-frame methionine HuD utilizes its RRMs to recognize and bind specific
codon(withtheexceptionofE1a4,sinceitharborstwostop targetmRNAs.DeletionandmutationalanalysisoftheHuD
codons downstream from the AUG), suggesting that they proteinrevealedthatthefirsttwoRRMsbindtothecis-acting
′
mayencodealternateHuDNtermini. elements,typicallyinthe3 UTR,whilethethirdRRMinter-
Inadditiontothe5′ end,HuDpre-mRNAissubjectedto actswiththepoly(A)tailoftranscriptsandstabilizestheRBP-
alternativesplicingofexons(E)6and7,resultinginthreead- mRNAcomplex(Chungetal.1996,1997;Maetal.1997;Ross
ditional transcript isoforms (HuD, HuD , and HuD ) etal.1997;Andersonetal.2000;Parketal.2000;Wangand
mex pro
(Fig. 1B,C), two of which are predominantly expressed in TanakaHall2001;Beckel-Mitcheneretal.2002).Bindingto
thepostnatalmousebrain(HuDandHuD )(Inmanetal. thepoly(A)tailseemstodependonthetaillength,sinceresults
pro
1998). The combinatorial complexity of HuD alternative 5′ from in vitro binding assays demonstrated that the longer
and3′exonsmayproduceupto24differentmRNAvariants. thepoly(A)tail,thehighertheaffinityofHuproteinbinding
TheHuDprotein,likeallHumembers,hasamolecularweight toitstargetmRNA(Beckel-Mitcheneretal.2002).Another
of ∼40 kDa and contains three RNA recognition motifs functionofthethirdRRMofHuDinvolvesinteractingwith
(RRMs)withalinkerregionseparatingthesecondandthird other proteins, including homo- and hetero-multimerizing
RRM (Fig. 1C; Good 1995; Liu et al. 1995; Okano and withHumembers(Kasashimaetal.2002).BoththeDroso-
Darnell1997).TheRRMisthemostabundant(foundinup phila ELAV and other mammalian Hu proteins have also
to1%ofhumangenes)andwell-characterizedRNA-binding beenshowntoformhomo-multimersontargetmRNAs,sug-
domain,composedof∼90aminoacids(Cleryetal.2008).In gestingthatcomplexformationisanevolutionarilyconserved
theHuDgenelocus,E2andE3,E4andE5,andE8encodethe mechanism necessary forefficient function of Hu members
threeRRMs,respectively,whereasthealternativelysplicedE6 (Levine et al. 1993; Gao and Keene 1996; Kasashima et al.
www.rnajournal.org 1021
BronickiandJasmin
2002;Devauxetal.2006;Fialcowitz-Whiteetal.2007;Toba 2011;Fallinietal.2012b).Inadditiontobeingexpressedin
andWhite2008). variousneuronalsubpopulationsintheCNSandPNS,quan-
titativeRT-PCRandWesternblotanalysisrevealedrelatively
minor,althoughsignificant,HuDmRNAandproteinlevels
TEMPORALAND SPATIALEXPRESSION OF HuD
in several nonneuronal tissues including lung, testes, liver,
Converging findings suggest that HuD is one of the earliest pituitary gland, pancreatic βcells, and fibercells of the lens
markersoftheneuronalphenotype(Good1995;Okanoand (Abdelmohsenetal.2010;Biteletal.2010;Leeetal.2012).
Darnell 1997; Wakamatsu and Weston 1997; Hambard- Theserecentfindingsindicatethat,inparalleltoitskeyfunc-
zumyanetal.2009).Multiplelaboratorieshavedemonstrated tionsinneurons,HuDlikelyplaysimportantrolesinseveral
viaNorthernblotsandquantitativeRT-PCRthatHuDmRNA othernonneuronalcelltypes.
levels are detectable by day 10 of mouse and rat embryonic
brain development, peak at E16, and slowly decline until
Cis-ELEMENTAND TARGET mRNAs OF HuD
birth (Clayton et al. 1998; Hambardzumyan et al. 2009;
Abdelmohsen et al. 2010). This temporal pattern of mRNA The most extensively studied HuD-targeted cis-acting ele-
expression is similar to that of HuB and HuC, although ments are adenosine/uridine (A/U)-rich elements (AREs)
′
there is some evidence that the relative abundance of each commonly found in the 3 UTR of almost 10% of cellular
nELAVmembervariesatdifferenttimepointsofdevelopment mRNAs(Haleesetal.2008).AREsusuallyrangefrom50to
(Good 1995; Okano and Darnell 1997; Hambardzumyan 150ntinsizeand,astheirnameindicates,consistofA-and
etal.2009). U-richstretches.Thesemotifsvaryconsiderablyinsequence,
InsituhybridizationanalysesofnELAVexpressioninthe andforsimplification,theyarecategorizedintothreemajor
mousebrainatday14ofembryogenesis(E14)haverevealed classes;classIcontainsonetothreeAUUUApentamersina
that the spatial expression of HuD mRNA is also partially U-richcontext,classIIincludesoverlappingUUAUUUAUU
unique compared to HuB and HuC (Okano and Darnell nonamers,andclassIIIconsistsofU-richsequenceswithout
1997), thus further supporting the notion that HuD is not AUUUApentamers(Bakheetetal.2003,2006).AREmotifs
completelyfunctionallyredundant.Inthedevelopingmouse are typically found in short-lived transcripts encoding pro-
andratcerebralcortex,HuDmRNAispresentinproliferating teins with diverse functions such as cellular proliferation,
neuronalstem/progenitorcellsintheventricularzone,neuro- differentiation, transcription, RNA metabolism, inflamma-
blastsmigratingintheintermediatezones,andterminallydif- tion,andstress-response(Bakheetetal.2006;Khabar2010).
ferentiatingneuronsinthecorticalplate(OkanoandDarnell TheseelementsaffectthestabilityofmRNAsbyservingastar-
1997;Claytonetal.1998).HuDtranscriptsarealsopresentin getsitesforARE-bindingRBP(s),alsoknownasAU-binding
severalregionsofthedevelopinghippocampalformation,ol- proteins (AUBPs). Out of the roughly 20 currently known
factorybulb,retinaandspinalcord,suchassensoryandmo- AUBPs,themajorityareexpressedinneuronsandpromote
tor neurons (Okano and Darnell 1997; Clayton et al. 1998; mRNAdecayincluding,forexample,AUF1,TTP,andKSRP
Bronickietal.2012).Intheadultmouseandrodentnervous (Table 1; Wu and Brewer 2012). Nevertheless, there are a
system,expressionofHuDmRNAismostlyrestrictedtospe- few known AUBPs that enhance mRNA stability, with Hu
cific neuronal populations, including large pyramidal-like proteins being the most notable (Brennan and Steitz 2001;
neurons in layer V of the neocortex and cerebellar cortex HinmanandLou2008;Pascaleetal.2008).
(Purkinje cells), the four Cornu Amonis (CA1-4) regions of ThecrystalstructureofHuDboundtoc-fosortumorne-
the hippocampus, dorsal root ganglia, and motor neurons crosis factor α (TNFα) mRNAs illustrates that specific resi-
inthespinalcord,mitralcellsintheolfactorybulb,ganglion dues within its first two RRMs preferentially bind the
and internal plexiform layers in the retina, and enteric ner- pyrimidine-rich X-U/C-U-X-X-U/C?-U-U/C consensus se-
vous system neurons (Okano and Darnell 1997; Clayton quencewherethe“?”denotes ambiguity whetheracytosine
etal.1998;D’Autreauxetal.2011). in this position is tolerated (Wang and Tanaka Hall 2001).
At the subcellular level, HuD is predominantlyexpressed Thisreportalsoshowedthattheaminoacidresiduesinteract-
in the cytoplasm, with minor levels detected in the nucleus ing with RNA nucleotides arewell conserved among all Hu
(Kasashimaetal.1999).Indevelopingneurons,HuDisfound proteins and the Drosophila ELAV protein. Several studies
ingrowthconesofextendingneurites,whileinmatureneu- haveutilizedinvitroandinvivoapproachessuchasRNAelec-
rons it is present in axons and dendrites, including at the trophoreticmobilityshiftassays(REMSAs)andRNA-immu-
pre-andpost-neuriteterminals,respectively(Aranda-Abreu noprecipitation (IP), respectively, to demonstrate that HuD
etal.1999;Aronovetal.2002).Closerinspectionofintraneu- binds to ARE sequences in mRNAs (Liu et al. 1995; Chung
ronalcompartmentsrevealedthatHuDislocalizedtogran- et al. 1996, 1997; Deschenes-Furry et al. 2003; Ratti et al.
ules, along with other proteins, such as kinesin-like protein 2008).However,itisimportanttopointoutthatthepresence
(KIF3A)andsurvivalofmotoneuron(SMN),withinthecyto- ofanAUUUAsequenceisnotalwayssufficientforHuDbind-
plasm and neurites (Aronov et al. 2002; Smith et al. 2004; ing,indicatingthatHuDdisplayspreferenceforspecificARE
Tiruchinapallietal.2008a,b;Hubersetal.2010;Aktenetal. motifs(Tobaetal.2002).
1022 RNA,Vol.19,No.8
Expression,function,andregulationofHuD
Interestingly,amicroarrayscreenofHuD-boundmRNAs rest(p21cip1/waf1)(Josephetal.1998),neuroblastproliferation
found that less than half of its targets contain an ARE (MSI-1) (Ratti et al. 2006), neuron migration (MARCKS)
(Bolognani et al. 2010). Instead, the investigators identified (Wein et al. 2003), neurite extension ([GAP-43] [Chung
three novel consensus motifs that are present in ∼80% of et al. 1997], Tau [Aranda-Abreu et al. 1999], and AChE
HuD-targeted mRNAs expressed in the mouse forebrain. [Cuadradoetal.2003;Deschenes-Furryetal.2006;Bronicki
ThefirstmotifconsistsofC-andU-richstretches,withaprev- and Jasmin 2012]), synapse formation (NOVA-1) (Ratti
′ ′
alence of the former, and is located mostly in the 5 and 3 etal.2008),andneuronalgrowthandsurvival(NGF,BDNF,
UTRs of mRNAs. The second and third motifs are U-rich andNT-3)(Table2;LimandAlkon2012).
with interspersed guanines (G) or adenosines (A), respec- InadditiontotheseHuD-targetedmRNAs,threeindepen-
tively. These latter two motifs are predominantly found in dent high-throughput studies have unravelled functionally
′ ′
the3 UTRandoccasionallyinthe5 UTRandcodingregions diverseclassesofgenesregulatedbyHuD.Oneofthesestud-
ofmRNAs.Recenthigh-throughputanalyses,usingvariations ies purified HuD-containing mRNP complexes from HuD
of cross-linking and immunoprecipitation (CLIP) such as overexpressing (HuD-Tg) adult mouse forebrains and sub-
Photoactivatable-Ribonucleoside-EnhancedCLIP(PAR-CLIP) sequently pulled down target mRNAs with GST-HuD for
and individual-nucleotide resolution CLIP (iCLIP), of HuR microarray examination. Gene ontology analysis identified
andnELAVtargetmRNAscorroboratethesefindingsbydem- several mRNAs whose proteins are involved in regulating
onstratingthatallfourHuproteinspreferentiallybindU-rich RNAprocessing,nuclearexport,translation,cell-to-cellsig-
sequences interspersed withGsorAs(Lebedeva etal.2011; nalling,andvesicletrafficking(Bolognanietal.2010).Ofpar-
Mukherjee et al. 2011; Uren et al. 2011; Ince-Dunn et al. ticularinterest,thisstudyalsofoundthatHuDbindsmRNAs
2012). These studies also found that Hu binding sites were of several RBPs, including its own, HuB, HuR, Cugbp2,
′
mostlylocatedin3 UTRsofmRNAs,withasignificantpor- Musashi 2, and Staufen 2, indicating a complex regulatory
′
tion in introns and relatively few in 5 UTRs. In terms of network between HuD and other components of the ribo-
HuD,thevarioustypesofcis-actingmo-
tifs that it targets are found in mRNAs
encodingproteinswithdiversefunctions
TABLE2. Establishedfunction(s)ofHuDonneuronaltargetmRNAs
inneuronsrangingfromcellcyclearrest
to synapse formation (Bolognani et al. Gene EffectofHuDonmRNA
2010). Growthassociatedprotein Stabilization(Liuetal.1995;Chungetal.1997;Tsaietal.
Evidence that Hu proteins bind AREs 43(GAP-43) 1997;Andersonetal.2000;Mobaraketal.2000)
originated from a study demonstrating Transportationintoneurites(Smithetal.2004)
p21cip1/waf1 Stabilization(Josephetal.1998;Fujiwaraetal.2006)
preferentialbindingofHuBtoA/GUUU
′ Tau Stabilization(Aronovetal.2002)
A/G/U sequences located within the 3
Transportationintoneurites(Aranda-Abreuetal.1999)
UTRs of mRNAs (Levine et al. 1993). Translationenhancement(Aronovetal.2002)
This fundamental finding enabled the p27KIP Translationrepression(Kullmannetal.2002)
subsequentidentificationofthefirsttar- Neuroserpin Stabilization(Cuadradoetal.2002)
N-myc Stabilization(Manoharetal.2002)
get mRNA of HuD, c-fos, a member of
Myristoylatedalanine-richC Stabilization(Weinetal.2003)
the immediate early (IE) gene family of
kinasesubstrate(MARCKS)
transcription factors (TFs) that encode Acetylcholinesterase(AChE) Stabilization(Deschenes-Furryetal.2003)
proteins involved in cell proliferation Musashi1(Msi1) Stabilization(Rattietal.2006)
and differentiation (Liu et al. 1995). Calcitonin/CGRP Alternativesplicing(exonexclusion)(Zhuetal.2006)
Alternativepolyadenylation(Zhuetal.2006)
UsingREMSAs,theauthorsshowedthat
Ikaros Alternativesplicing(Bellaviaetal.2007)
HuD binds specifically to an ARE se-
Neurofibromatosis(NF-1) Alternativesplicing(exonexclusion)(Zhuetal.2008)
′
quence in the c-fos 3 UTR. Numerous Localtranscriptionrateenhancement(Zhouetal.2011)
other ARE-harboring transcripts, such Neuro-oncologicalventral Stabilization(Rattietal.2008)
asN-mycandc-myc,werelateridentified antigen1(NOVA1) Translationenhancement(Rattietal.2008)
LIMdomaintranscription Stabilization(Chenetal.2007)
as targets of HuD using similar in vitro
factor(LMO4)
approaches, indicating that HuD binds
HuD Alternativesplicing(exoninclusion)(Wangetal.2010)
short-livedmRNAswhoseproteinprod- HuR Alternativepolyadenylation(MansfieldandKeene2011)
uct functions in cell proliferation (Liu Glutaminase(Gls1/Gls) Alternativesplicing(Ince-Dunnetal.2012)
etal.1995;Rossetal.1997).Inaddition Stabilization(Ince-Dunnetal.2012)
Translationenhancement(Ince-Dunnetal.2012)
tothesetranscripts,HuDhasbeenshown
Nervegrowthfactor(NGF) Stabilization(LimandAlkon2012)
tobindandregulateseveralothermRNAs
Neurotrophin3(NT-3) Stabilization(LimandAlkon2012)
invitroandinvivo,manyofwhichencode Brain-derivedgrowthfactor Stabilization(LimandAlkon2012)
proteinswithimportantfunctionsinneu- (BDNF)
ronaldifferentiationsuchascellcyclear-
www.rnajournal.org 1023
BronickiandJasmin
nome (Pullmann et al. 2007; Mansfield and Keene 2009; specific exons remained hyperacetylated, which resulted in
Bolognanietal.2010).Asecondstudyalsoemployedamicro- an increased RNAPII transcription rate and decreased exon
arrayapproachtodetermineHuD-targetedmRNAsindentate inclusion atthesegenomicloci.ExperimentswithHuBand
granulecellsofHuD-Tgmice(Perrone-Bizzozeroetal.2011). HuC deletion mutants indicated that the linker region and
The authors found that the data set is mostly enriched in RRM3arenecessaryfornELAVregulationofalternativesplic-
mRNAsencodingproteinsinvolvedinneuronaldevelopment ing(Zhuetal.2008;Zhouetal.2011).
and axogenesis, such as Notch3 and Neurog2 (Perrone- Incontrasttothisfunction,evidencethatHuDandthetwo
Bizzozeroetal.2011).Last,arecentstudyusedCLIPandmi- other nELAVs enhance exon inclusion also exists. For in-
croarraytechnologiestorevealthatnELAVsregulatemultiple stance,allfourHumemberswereshowntopromoteinclusion
transcripts,someofwhichencodecomponentsoftheamino ofHuDexon6bybindingtotwoAREsequencesinthedown-
acidglutamatebiosyntheticpathway,suchasglutaminase1, streamintron(Wangetal.2010).Inaddition,arecentreport
which indicatesthatanimportantfunction ofnELAVs isto identifiedintronicnELAVbindingsitesflankinganalternative
′
controlneuronalactivity(Ince-Dunnetal.2012). 3 SSintheGls1/GlsgeneanddemonstratedthatnELAVsgen-
eratealongerGlsmRNAvariant,indicatingthattheymayen-
′
hanceinclusionofaGls3 exon(s)(Ince-Dunnetal.2012).
MULTILEVEL CONTROLOFmRNA
Thesamereportalsousedgenomewideanalysistomapthe
METABOLISM BY HuD
binding sites in nELAV targeted pre-mRNAs and revealed
that these proteins preferentially bind introns surrounding
Alternativesplicing
cassette exons, especially near exon/intron splice junctions
FollowingbindingtoatargetmRNA,HuDhasbeenshown (Ince-Dunn et al. 2012). In line with previous studies,
to control one or multiple regulatory events during the nELAVswerefoundtoenhanceexoninclusionwhenbound
transcript’s metabolism, with alternative splicing being one to5′SSsinthedownstreamintronandinhibitexoninclusion
′ ′
oftheearliest.ThefirstcluesthatHuDplaysaroleinalter- when bound to adjacent 3 and 5 SSs of alternative exons.
nativesplicingwerebasedonitssignificanthomologytothe Altogether,thesestudiesclearlydemonstratethatHuD,along
Drosophilasplicingfactorsxl(Szaboetal.1991)andexperi- withtheothernELAVproteins,canpromoteorsuppressexon
mentsdemonstratingthattheDrosophilaELAVproteinpro- inclusionbyinteractingwith(orantagonizing)splicing,tran-
motes neuron-specific alternative splicing (Koushika et al. scription,andchromatincomponents(Fig.2A).
1996).Followingtheseseminalfindings,amoredirectindica-
tionthatHuDandtheotherthreeHuproteinsregulatealter-
Alternativepolyadenylation
nativesplicinginmammalsemergedfromastudythatshowed
HuDbindingtobothexonicandintronicregionsofN-myc Inadditiontoalternativesplicing,boththeDrosophilaELAV
pre-mRNA(Lazarovaetal.1999).Afewyearslater,aseries andthemammalianHuproteinsweredemonstratedtocon-
of papers predominantly fromthe Hua Lou laboratory cor- trolalternativepolyadenylationspecificallyinneurons(Soller
roborated thatall Hu proteins function as auxiliary splicing andWhite2003).TruncationofthemouseHuBproteinre-
factors(HinmanandLou2008).Usingvariouscomplementa- vealed that all three RRMs are necessary for Hu proteins
rymethods,suchasREMSAandUVcross-linking/IPassays, toperformthisfunction(Zhuetal.2006).Employingmainly
onereportdemonstratedthatnELAVscompetewithTIA-1/ REMSAs and in vitro cleavage and polyadenylation assays,
TIARRBPsforaU-richintronicsequenceinthecalcitonin/ one study determined that Hu proteins bind U-rich se-
calcitoningene-relatedpeptide(CGRP)pre-mRNAtoblock quencesdownstreamfromtheexon4poly(A)signalincalci-
inclusionofexon4.Throughthismechanism,nELAVspro- tonin/CGRPtoblockbindingofcleavagestimulationfactor
moteexpressionoftheneuron-specificCGRPisoforminneu- 64 (CstF64) and cleavage-polyadenylation specificity factor
rons(Zhuet al.2006). nELAVs werealso showntointeract (CPSF),essentialcomponentsofthecleavageandpolyadeny-
withintronicAREssurroundingexon23aoftheneurofibro- lationmachinery.ThroughthismechanismHuproteinspre-
matosistype1(NF1)pre-mRNAtoblockitsinclusionbyin- vent cleavage and polyadenylation atthe nonneuronal exon
hibiting U1/U6 snRNP and U2AF auxiliary splicing factor 4 poly(A) signal, thereby promoting the neuron-specific
′ ′
binding to the 5 and 3 splice sites (SSs), respectively (Zhu CGRPpathway(Zhuetal.2006).Moreover,allfourHupro-
etal.2008).Furthermore,overexpressionanddown-regula- teins were recently demonstrated to differentially regulate
tionofHuDinT-celllinesdemonstratedthatitmediatesalter- alternative polyadenylation of HuR in neurons (Mansfield
nativesplicingoftheIkarosgene,possiblybysuppressingexon and Keene 2011; also see Dai et al. 2012). Mansfield and
4inclusion(Bellaviaetal.2007;HinmanandLou2008).In KeeneshowedthattherearethreedifferentHuRmRNApoly-
′
additiontobindingintronicsequences,co-IPassaysrevealed adenylationvariants(1.5,2.4,and6.0kb),whichdifferin3
thatHuproteinsinteractwithRNApolymeraseIIandhistone UTR length. All four Hu members were found to bind and
deacetylase2(HDAC2)toregulatealternativesplicing(Zhou blockU-richsequencesnearthe2.4-kbpolyadenylationsite
et al. 2011).In vitrotranscription elongationassays showed and thereby promote expression of the 6.0-kb transcript.
that, by blocking HDAC2 activity, histones surrounding Since the 6.0-kb variant is less stable and not as efficiently
1024 RNA,Vol.19,No.8
Expression,function,andregulationofHuD
mRNA stability
The initial demonstration that Hu proteins regulate mRNA
?
? E stability was based on the observation that HuB enhances
GLUT1 mRNA half-life and, incidentally, translation (Jain
A 7mG C A7AmAGA D A7AmAGA A7AmAGA etal.1997).Followingthislandmarkstudy,otherreportsde-
scribedsimilarrolesforHuR(FanandSteitz1998;Levyetal.
B
1998; Peng et al. 1998) and HuC/D (Anderson et al. 2000;
AAAA Gm7 Mobaraketal.2000)inregulatingmRNAstability.Sincethese
Legend: microtubules initialfindings,HuDhasemergedasthebest-describedstabi-
7mG AAAAmRNA miR lizingRBPinneurons,asevidencedbythenumerousstudies
destabilizing RBPs ribosomes
outlining itsrole inincreasingthehalf-livesofvarious neu-
splicing factors HuD
ronaltranscripts(Fig.2C).AmongalloftheHuD-regulated
mRNP
mRNAs, the most prominent is the GAP-43 mRNA, which
′
housesaclassIIIAREinits3 UTR(Chungetal.1997;Tsai
FIGURE2. Modeldepictingtheintracellularlocalizationsandmultiple
post-transcriptionalfunctionsofHuD.(A)Inthenucleus,HuDregu- et al. 1997). HuD-dependent control of GAP-43 expression
latesalternativesplicingandpolyadenylationbycompetingforspecific isparticularlyinterestingbecauseGAP-43playsmultipleroles
pre-mRNAbindingsiteswithothertrans-actingfactors.(B)HuDforms in neurons, including functioning as a critical component
partofanmRNPcomplex,anditlikelyfacilitatesmRNAexportintothe
in neural development (Maier et al. 1999; Mani et al. 2001;
cytoplasm. (C) In the cytoplasm, HuD and destabilizing RBPs bind
competitively or cooperatively to mRNAs. HuD might also prevent Shen et al. 2008) and synaptic remodeling during learning
bindingofmiRstothemRNAorantagonizetheirfunction.Through and memory (Routtenberg et al. 2000). Multiple studies
thesemechanisms,HuDincreasesthehalf-livesofmRNAs.(D)HuD have shown that overexpression or down-regulation of
transportsmRNAstodifferentcompartmentsofneurons,mostnotably
HuDresultsinincreasedordecreased,respectively,GAP-43
neurites,alongmicrotubules.(E)Atthesynapticterminal,HuDmay
promoteorrepresstranslationoftranscripts. mRNAlevelsand/orstabilityinculturedcellsandthemouse
nervous system (Anderson et al. 2000, 2001; Mobaraket al.
2000; Bolognani et al. 2006, 2007a). Interestingly, HuD is
translatedasthe2.4-kbtranscript,theauthorspostulatedthat more effective at stabilizing GAP-43 mRNAs that contain a
the reduction of HuR expression prevents HuR’s ability to longer poly(A) tail, indicating that HuD binds to both the
promoteproliferationandtherebypermitsneuronstotermi- ARE and poly(A) tail to increase GAP-43 mRNA half-life
nally differentiate (Mansfield and Keene 2011). The role of (Beckel-Mitcheneretal.2002).InadditiontoGAP-43tran-
Hu proteins in controlling alternative polyadenylation is scripts,HuDhasalsobeenshowntopromotethestabilityof
likely extended to several of their target mRNAs, given that several other ARE-containing mRNAs in neurons, such as
the ELAV protein in Drosophila was found to coordinate AChE (Deschenes-Furry et al. 2003) and Nova1 (Ratti et al.
′
3 UTR extension of multiple mRNAs in neuronal cells 2008),bothinvitroandinvivo(Table2;Perrone-Bizzozero
(Hilgersetal.2012). and Bird 2013). Despite mRNA stabilization being one of
thebetter-knownrolesofHuD,themoleculareventsbehind
thisfunctionarenotwellunderstood.
Nucleocytoplasmic shuttling
OnelikelymechanismthroughwhichHuDandotherHu
Another chief function of HuD involves shuttling of target members stabilize mRNAs involves antagonizing destabi-
mRNAsintothecytoplasm.Aspreviouslymentioned,HuD lizingRBPssuchasAUF1frombindingtocis-actingelements
containsaNESandputativeNLSlocatedinthevariablelink- (Barreau et al. 2006). Once bound to a target mRNA, des-
erdomain,betweenthesecondandthirdRRM,whichpermit tabilizingRBPsmaypromotedeadenylationthroughavariety
nucleocytoplasmic shuttling of HuD (Fig. 2B; Kasashima ofmannerssuchasblockingpoly(A)bindingprotein(PABP)
etal.1999).Overexpressionofwild-typeanddeletionmutant binding and/or recruitment of decapping, deadenylation,
HuD proteins revealed that cytoplasmic shuttling of HuD and exosome components (Wu and Brewer 2012). Thus, it
requirestheNESregion (Kasashimaetal.1999).Moreover, is feasible that, by binding to AREs or other motifs, HuD
in vitro binding assays showed that the first two RRMs of blocks accesstothebindingsite andconsequently prolongs
HuDassociatewiththeprimarygeneralmRNAexportadap- mRNA half-life (Fig. 3). However, there is evidence indi-
tor complex (TAP-p15; also known as NXF1/NXT1) in a cating that this mechanism is more complex since both
RNA-independent mannerand thatthis interaction maybe stabilizing and destabilizing RBPs can bind the same tran-
necessary for shuttling of HuD (Saito et al. 2004). Further scriptsimultaneously.Fluorescenceresonanceenergytrans-
studies are necessary to establish whether TAP-p45 and/or fer (FRET) and biochemical techniques revealed that HuR
another export pathway, such as the karyopherin CRM1, is interactswithAUF1,KSRP,andTIA-1inthenucleusandcy-
used by HuD for nuclear export of mRNAs (Carmody and toplasmic stress granules (David et al. 2007; David Gerecht
Wente2009). etal.2010).Moreover,theseinteractionsrequirethepresence
www.rnajournal.org 1025
BronickiandJasmin
A B ingprotein(PABP),translationinitiation
factor 4E (eEF4E) (Tiruchinapalli et al.
2008b) and the microtubule-associat-
ed component MAP1B (Fujiwara et al.
2011b). Various conditions, including
KCldepolarizationofprimaryhippocam-
palneurons,cocaine,orseizuretreatment
RNAation RNAation of mice and phorbol esteror bryostatin-
increased mstability/transl decreased mstability/transl 1(drNiateanBspc),pecaloienclfadlstH,iiorutnesDsui,tnolitctrsihenruaeasmdendiasanturaisgbnsmuoeutceiirnoaotntbeiodltanosatbnwoueminutha--
mRNAsandmRNPcomponents(Pascale
etal.2005;Tiruchinapallietal.2008a,b).
ComplementarytoitsroleinmRNAlocal-
C
ization, evidence for the direct involve-
m7Gppp AAAAAn m7Gppp AAAAAn
ment of HuD in regulating translation
wasunveiledwhenFukaoandcolleagues
D
showed that HuD binds to eIF4A and
m7Gppp AAAAAn m7Gppp AAAAAn
the poly(A) tail of transcripts, via its
linkerregionandthirdRRM,toenhance
E cap-dependent protein synthesis (Fukao
m7Gppp AAAAAn m7Gppp AAAAAn et al. 2009). Moreover, this mechanism
wasdeterminedtobeimportantforHuD-
FIGURE3. RelativelevelsofRBPsandtheircooperativeandcompetitivebindingtomRNAs
controlsgeneexpression.TherelativeexpressionofRBPsthat(A,B)promote(e.g.,Hu/ELAVl dependent neurite extension in PC12
proteins;green)orinhibit(e.g.,AUF1;red)mRNAstability/translationinfluencesgeneexpres- cells.
sion.FunctionallyantagonisticRBPscanalsobind(C)competitivelyor(D,E)cooperativelyto
In addition to enhancing translation,
regulatemRNAfate.Cooperativebindingcouldpotentiallyallowforaquickerresponsetoextra-
HuDiscapableofrepressingproteinsyn-
cellularcuesindeterminingmRNAfate.
thesis,asshownforp27andpreproinsulin
(Ins2)mRNAs(Kullmannetal.2002;Lee
ofRNA,sinceHuRandAUF1wereshowntosimultaneously etal.2012).Eventhoughthemechanismoftranslationalsi-
orcompetitivelybindontoexclusiveorcommon,respective- lencingisunclear,itwasdemonstratedtorequireHuDbind-
ly,cis-actingelementsinthe3′UTRsofp21cip1/waf1,CyclinD, ing toan internal ribosomeentrysite (IRES) in the5′ UTR
TNFα, and β -adrenergic receptor mRNAs (Lal et al. 2004; of p27 mRNA (Kullmann et al. 2002) and a short ∼22-nt
2
′
Davidetal.2007).Together,thesefindingssuggestthatregu- sequenceinthe5 UTRofIns2mRNA(Leeetal.2012).The
lationofmRNAstability,andlikelyothermRNAmetabolism reportbyKullmanetal.suggeststhatinhibitionofp27trans-
events,byHuproteinsisatleastpartiallydependentonabal- lationbyHuproteinsisaneventthatsustainscellsinaprolif-
ancebetweenstabilizinganddestabilizingRBPs(Fig.3). erativestate(Kullmannetal.2002).Ontheotherhand,the
studybyLeeetal.showedthattherepressionofIns2translation
byHuDislikelyrequiredtomaintaininsulinhomeostasis(Lee
mRNA localizationand translation etal.2012).Takingintoaccountthesefindings,itisplausible
thatHuDinhibitstranslationofselecttargetmRNAsinneuro-
AlongwithpromotingmRNAstability,compellingevidence
nal stem/precursor cells or in response to a neuronal stress
supports a role for HuD in regulating both localization and
suchashypoxia.Altogether,thesestudiesindicatethatHuD
translationoftranscriptsinthecytoplasmandpossiblyinneu-
recruitsmRNAstomicrotubulesandmaytransportthemto
rites(Fig.2D,E).Forexample,twotargetmRNAsofHuDin-
neuritestocontroltheirlocaltranslationinresponsetoextra-
volvedinaxonal outgrowth, namely GAP-43 andTau,were
cellularcues,particularlyduringneuronaldevelopmentand
found to colocalize with HuD protein and polysomes in
function.
growth cones during neuronal differentiation (Smith et al.
2004;Atlasetal.2007).Co-IPandimmunocytochemistryex-
perimentsrevealedthatHuDinteractswithahostofmRNP
MOLECULAR MECHANISMS REGULATING HuD
componentsinvolvedintransportandtranslationinthecyto-
plasmandneuritessuchasthemotorproteinKIF3A(Aronov ThekeyrolesofHuDinneuronaldevelopmentandplasticity
etal.2002),IGF-IImRNAbindingprotein1(IMP-1)(Atlas anditsimplicationsinneuronaldiseasesraiseimportantques-
etal.2004,2007),survivalofmotorneuron(SMN)(Hubers tions regarding the molecular mechanisms that regulate its
etal.2010;Aktenetal.2011;Fallinietal.2011),poly(A)bind- expression.Todate,onlyafewstudieshavestartedtodescribe
1026 RNA,Vol.19,No.8
Expression,function,andregulationofHuD
the post-translational, post-transcriptional, and transcrip- to increase nELAV levels and provoke their nuclear export
tionalmechanismsthatcontroltheexpressionandfunction followingtreatmentofSH-SY5YNBcellswiththePKC-acti-
of HuD (Fig. 4). However, given the key functions of HuD vating compounds phorbol esters and bryostatin-1 (Pascale
inneurons,moresystematicinvestigationsofthe molecular et al. 2005). Moreover, stimulation of PKCα promotes its
eventsthatregulateitsexpressionarenecessary. interaction with nELAV proteins resulting in their phos-
phorylationatthreonineresidues,redistributiontotheneu-
ronalcytoskeletalandmembranefractions,andstabilization
Post-translationalcontrolofHuD
ofGAP-43mRNA.Inaddition,arecentstudydemonstrated
expressionand function
that PKCε also interacts with HuD, and activation of both
Althoughsomeconditions,suchasapplicationofgrowthfac- PKC isoforms by bryostatin-1 results in HuD phosphoryla-
tors (Aranda-Abreu et al. 1999; Abdelmohsen et al. 2010), tion on nine residues (Lim and Alkon 2012). Importantly,
pharmacological agents (Pascale et al. 2005; Tiruchinapalli HuD is a focal downstream target of this neurite-extending
et al. 2008a,b), and cellular stress (Burry and Smith 2006), pathway since nerve growth factor (NGF) or phorbol ester
have been documented to alter the expression or function treatmentofPC12cellsfollowingHuDknockdownprevents
ofHuD,themolecularpathwaysanddownstreamregulators neuriteoutgrowth(Mobaraketal.2000).
involved areonly beginning to emerge. Currently, thereare Another pathway that has been documented to regulate
only three known pathways that regulate the abundance or HuD function involves the methyltransferase CARM1 (also
function of HuD: protein kinase C (PKC) (Mobarak et al. knownasproteinargininemethyltransferase4[PRMT4]),a
2000;Pascaleetal.2005;LimandAlkon2012);coactivator- memberofthePRMTfamilythatperformsavarietyofcellular
associated arginine methyltransferase1(CARM1)(Fujiwara functionssuchasregulatingtranscriptionandRNAprocess-
etal.2006;Hubersetal.2010);andproteinkinaseB(PKB/ ing by methylating arginine residues in substrate proteins
AKT)(Fujiwaraetal.2011a). (Bedford and Clarke 2009). Using a variety of comple-
The ubiquitously expressed serine/threonine PKC family mentary techniques, including an in vitro protein methyla-
consistsofatleast10isoformswithdiversefunctionsranging tion assay, CARM1 was demonstrated to interact with and
from controlling cell proliferation to synaptic remodeling methylate an amino acid located in the hinge region of
(Amadioetal.2006).TheclassicalPKCαisoformwasshown rat (Arg236) (Fujiwara et al. 2006) and mouse (Arg248)
(Hubersetal.2010)HuD(Fig.1C).
Moreover, methylation of HuD de-
creased p21cip1/waf1 mRNA stability
mRNA processing Legend: DNA and maintained neuronal precursor
mmRRNNAA strtaanbsilliaztaiotinon ? RBPs ribosomes cells in a proliferative state. On the
P phosphorylation mRNP otherhand,inhibitionofHuDmeth-
HuD interactor AAAA mRNA
B miR-375 proteins 7mG ylation or knockdown of CARM1
A PKC resulted in increased p21cip1/waf1
PHuD AAAA ? mRNA levels, cell-cycle withdrawal,
7mG
CARM1 and neuronal differentiation. Sur-
cytoplasm prisingly,althoughreducedCARM1
nucleus levelsalsoincreasedGAP-43mRNA
expression, they had no effect on
7mG otherknownHuD-targetedmRNAs,
T3+TR FoxO1 C including Tau and p27, suggesting
D
Hu thatstabilizationoftheseothertarget
Ngn2 E47 E1c E2 PolII mRNAs requires additional post-
HuD gene translational modifications of HuD
and/or other trans-acting factors
FIGURE4. SimplifiedmodelofthemoleculareventsgoverningHuDexpression,processing,and
(Fujiwara et al. 2006; Hubers et al.
functioninneurons.(A)CARM1methylationofHuDpreventsitsbindingtospecificmRNAtargets
andmaintainsneuronalprecursorsinanundifferentiatedstate.Ontheotherhand,anincreasein 2010). These findings provide evi-
unmethylatedHuDproteinlevels,itspost-translationalmodification(suchasphosphorylationby dence that methylation of HuD by
PKCisoforms),andinteractionwithotherproteins(e.g.,AKT1)enableHuDtoregulatemRNAme-
CARM1 may be a universal switch
tabolismofproneuralgenesandpromoteneuronaldifferentiation/neuriteextension.(B)Thestabil-
inneuronalprecursorsthatregulates
ityandtranslationofmatureHuDmRNAisnegativelycontrolledbymiR-375andpotentiallyother
trans-actingfactors.(C)HuDpre-mRNAundergoesalternativesplicing,includinginclusionofE6 the transition from proliferation to
mediatedbyHuproteins.(D)DuringneuronaldevelopmenttranscriptionoftheHuDgeneispos- neuronaldifferentiation.
itivelyregulatedbytheNgn2-E47heterodimerbindingtoE-boxesupstreamofE1c.SynthesisofHuD
Arecentreportfoundadirectlink
mRNAisnegativelyregulatedbythyroidhormone(presumablyboundtothyroidhormonereceptor)
inneuronsandFoxO1inpancreaticβcells.Greenandredlinesrepresentpathwaysthatpromoteand between the antagonistic effects of
inhibitneuronaldifferentiation,respectively. PKCandCARM1onHuDfunction
www.rnajournal.org 1027
BronickiandJasmin
andcellfate(LimandAlkon2012).Inhippocampalneurons, AnalysisoftheXenopusHuDhomologelrDsequenceillus-
thestudydemonstratedusingpathway-specificinhibitorsthat tratedthatthe3′UTRcontainsan∼100-ntstretchofnucleo-
′
activatedPKCisoformsinteractandphosphorylateCARM1, tideswith90%identitytothehumanHuD3 UTR,implying
whichleadstodecreasedCARM1methyltransferaseactivity. that this fragment encompasses an important evolution-
Furthermore,activatingtheclassicalPKCisoformsandsimul- arily conserved regulatory structure and/or cis-element(s)
taneouslyblockingCARM1activitypotentiatesHuDbinding (Good 1995). Indeed, analysis of HuD-targeted mRNAs in
to target mRNAs and hippocampal neurite extension. This themouseforebrainshowedthatHuDbindstoitsowntran-
studyclearlyillustratesthatPKCactivationnegativelycontrols script,althoughtheroleofthisinteractionremainsunknown
CARM1-mediatedmethylationofHuDtopromoteneuronal (Bolognanietal.2010).Onefunctionalconsequenceofthis
differentiation. auto-regulatorybindingispresumablyincontrollingalterna-
The phosphatidylinositol 3-kinase (PI3K)/AKT1 pathway tivesplicing of its own pre-mRNA (see above) (Wang et al.
wasalso found to regulate HuD function, specifically at the 2010). Another possible significance of thisinteraction may
translational level (Fujiwara et al. 2011a). This finding is in besimilartotheonefoundfortheDrosophilaELAVhomolog,
agreementwithPI3K/AKT1beingoneoftheprominentsig- whichwasdemonstratedtonegativelyregulateitsowntran-
nallingpathwaysregulatingtranslation(RouxandTopisirovic script (Samson 1998; Borgeson and Samson 2005). This
2012).InthestudybyFujiwaraetal.,HuDmutantswereused auto-regulatorymechanismisthoughttooccurwhenELAV
to demonstrate direct binding of active (phosphorylated) levels pass a certain threshold, resulting in increased ELAV
′
AKT1 to a sequence in the linker region of HuD and that bindingtoanAU-richsiteinanoncoding3-terminalexon
thisinteractionisrequiredforHuD-dependentneuriteout- and production of a noncoding transcript (Borgeson and
growth(Fujiwaraetal.2011a).InadditiontobindingHuD, Samson 2005). Given the multifunctional roles of HuD on
chronic (7-day) activation of AKT1 enhances HuD protein mRNA metabolism and the numerous putative ARE signa-
′
levels,suggestingthatAKT1alsopositivelycontrolsHuDex- turesinits3 UTR(LBronickiandBJJasmin,unpubl.),itis
pression(Tiruchinapalli etal.2008a).Theseresultsindicate conceivable that HuD also controls other regulatory events
thatAKT1increasestheabundanceofHuDproteinandin- suchaspolyadenylation,stability,localization,andtranslation
teracts with the HuD-mRNA complex to promote neurite ofitsownmRNA.
elongation, possibly through regulation of HuD-mediated Furthersupportofthenotionthattrans-actingfactorsreg-
translation.Basedonthesefindings,anemergingmodelsug- ulate HuD mRNA metabolism surfaced when miR-375 was
′
geststhat,throughpost-translationalcontrolofHuDexpres- demonstratedto target the3 UTRof HuD and decrease its
sionandfunction,activationofdifferentsignallingpathways abundancebyreducingbothitsmRNAstabilityandtransla-
canmaintainneuronalprecursor/stemcellsinaproliferative tion(Abdelmohsen etal.2010). KnockdownofmiR-375in
state (CARM1) or promote neuronal differentiation (PKC culturedcellsandmousebrainrevealedthatthismicroRNA
andAKT1)(Fig.4A). prevents neurite outgrowth through negative control of
HuD expression. Altogether, these studies stress that HuD
transcriptsarelikelysubjectedtointricatepost-transcription-
Post-transcriptional controlofHuD
alcontrolinvolvingRBPsandmiRs.
expression
Recentstudieshavestartedtodecipherthemolecularmecha-
Transcriptional controlofHuD expression
nismscontrollingHuDmRNA(Fig.4B)andpre-mRNA(Fig.
4C)expression.ThemammalianHuDmRNAcontainsatleast Despiteafewstudiesdemonstratingdirectorindirectregula-
threealternate5′UTRsandalong3′UTR(∼2.6kb),charac- tionofHuDatpost-translational,translational,andpost-tran-
teristicsthatareastrongindicationofpredispositiontopost- scriptional levels, little was known regarding the molecular
transcriptional regulation. As mentioned above, previous mechanismsthatcontroltranscriptionofthisgeneinneurons.
studies, along with datafrom our laboratory, revealed eight As discussed above, a previous study suggested that at least
HuDE1variants,indicatingtheexistenceofatleasteightdif- threeHuDE1variantsexist,pointingtothepotentialpresence
′
ferent5 UTRs(Fig.1A,B),withmostcontaininganin-frame ofmultiplepromoters(Inmanetal.1998).Workinourlabo-
translation start site (Abe et al. 1994; Inman et al. 1998; ratoryhasshownthatthemammalianHuDgeneactuallycon-
Bronickietal.2012).RT-PCRanalysisrevealedthatalleight tainseightE1variantsthatarewellconservedamonghigher
exons are expressed in embryonic and adult mouse brain vertebrates,addingtothespeculationthatmultiplepromoters
and areinduced during neuronaldifferentiation. Moreover, controlHuDexpression(Bronickietal.2012).Inlinewiththis
′
ourdatasuggestthatE1cisthepredominantlyexpressedE1 possibility,the5 genomicregionofelrDwasdemonstratedto
variant (Bronickietal.2012).AlthoughthefunctionofE1c containtwodistinctpromotersthatdriveneuron-specificex-
′
andtherestoftheE1variantsisunknown,itistemptingto pressionoftwoE1variants,E1andE1(NassarandWegnez
′
speculatethattheyencodeHuD5 UTRsand/orN-terminal 2001). These variants encode transcripts exhibiting unique
peptidesthathaveimportantfunctionssuchasintranslational temporalexpressionduringXenopusnervoussystemdevelop-
′
controlorintracellularlocalization(Davulurietal.2008). ment (Nassar 2011). Remarkably,E1 and E1partiallyalign
1028 RNA,Vol.19,No.8