Control of a neuronal morphology program byan RNA-binding zinc finger protein Unkempt

The MIT Faculty has made this article openly available Please share how this access benefits you Your story matters

Citation Murn Jernej Kathi Zarnack Yawei J Yang Omer Durak ElisabethA Murphy Sihem Cheloufi Dilenny M Gonzalez et al ldquoControl ofa Neuronal Morphology Program by an RNA-Binding Zinc FingerProtein Unkemptrdquo Genes Dev 29 no 5 (March 1 2015) 501ndash512

As Published httpdxdoiorg101101gad258483115

Publisher Cold Spring Harbor Laboratory Press

Version Final published version

Citable link httphdlhandlenet1721198393

Terms of Use Creative Commons Attribution

Detailed Terms httpcreativecommonsorglicensesby-nc40

Control of a neuronal morphologyprogram by an RNA-binding zinc fingerprotein Unkempt

Jernej Murn12 Kathi Zarnack34 Yawei J Yang5678910 Omer Durak1112 Elisabeth A Murphy5678910

Sihem Cheloufi1314 Dilenny M Gonzalez5678910 Marianna Teplova15 Tomaz Curk16

Johannes Zuber17 Dinshaw J Patel15 Jernej Ule18 Nicholas M Luscombe3192021 Li-Huei Tsai1112

Christopher A Walsh5678910 and Yang Shi12

1Department of Cell Biology Harvard Medical School Boston Massachusetts 02115 USA 2Division of Newborn MedicineBoston Childrenrsquos Hospital Boston Massachusetts 02115 USA 3European Molecular Biology Laboratory (EMBL) EuropeanBioinformatics Institute Hinxton Cambridge CB10 1SD United Kingdom 4Buchmann Institute for Molecular Life SciencesGoethe University Frankfurt 60438 Frankfurt am Main Germany 5Division of Genetics and Genomics 6Manton Center forOrphan Disease Research Boston Childrenrsquos Hospital Boston Massachusetts 02115 USA 7Department of Pediatrics8Department of Neurology Harvard Medical School Boston Massachusetts 02115 USA 9Broad Institute of MassachusettsInstitute of Technology and Harvard Cambridge Massachusetts 02142 USA 10Howard Hughes Medical Institute Chevy ChaseMaryland 20815 USA 11Picower Institute for Learning and Memory 12Department of Brain and Cognitive SciencesMassachusetts Institute of Technology Cambridge Massachusetts 02139 USA 13Cancer Center 14Center for RegenerativeMedicine Massachusetts General Hospital Boston Massachusetts 02114 USA 15Structural Biology Program Memorial Sloan-Kettering Cancer Center New York New York 10065 USA 16Faculty of Computer and Information Science University ofLjubljana 1000 Ljubljana Slovenia 17The Research Institute of Molecular Pathology Vienna Biocenter 1030 Vienna Austria18Department of Molecular Neuroscience University College London Institute of Neurology London WC1N 3BG UnitedKingdom 19UCL Genetics Institute Department of Genetics Environment and Evolution University College London LondonWC1E 6BT United Kingdom 20Cancer Research UK London Research Institute London WC2A 3LY United Kingdom21Okinawa Institute for Science and Technology Graduate University Onna-son Kunigami-gun Okinawa 904-0495 Japan

Cellular morphology is an essential determinant of cellular function in all kingdoms of life yet little is knownabout how cell shape is controlled Here we describe a molecular program that controls the early morphology ofneurons through a metazoan-specific zinc finger protein Unkempt Depletion of Unkempt in mouse embryosdisrupts the shape of migrating neurons while ectopic expression confers neuronal-like morphology to cells ofdifferent nonneuronal lineages We found that Unkempt is a sequence-specific RNA-binding protein and identifiedits precise binding sites within coding regions of mRNAs linked to protein metabolism and trafficking RNAbinding is required for Unkempt-induced remodeling of cellular shape and is directly coupled to a reducedproduction of the encoded proteins These findings link post-transcriptional regulation of gene expression withcellular shape and have general implications for the development and disease of multicellular organisms

[Keywords RNA-binding proteins cell morphology gene expression program neurons translation Unkempt]

Supplemental material is available for this article

Received January 9 2015 revised version accepted February 6 2015

Cellular shape is one of the most distinctive features ofsomatic cells in multicellular organisms and is inti-mately linked with cellular function Numerous descrip-tions of in vitro cell fate conversion experiments (Daviset al 1987 Vierbuchen et al 2010 Sato et al 2011) as wellas spontaneous morphogenesis of dissociated primarycells in culture (Dotti et al 1988) suggest that the basicinstructions for morphology of a particular cell type are

intrinsically encoded ie specified at the time of celllineage commitment However it is largely unknownhow cell shape is determined and to what extent it isprogrammedThe emergence and homeostasis of complex cellular

phenotypes including cell shape are critically dependent

2015 Murn et al This article is distributed exclusively by ColdSpring Harbor Laboratory Press for the first six months after the full-issue publication date (see httpgenesdevcshlporgsitemisctermsxhtml)After six months it is available under a Creative Commons License(Attribution-NonCommercial 40 International) as described at httpcreativecommonsorglicensesby-nc40

Corresponding authors yang_shihmsharvardedu murnjernejgmailcomArticle is online at httpwwwgenesdevorgcgidoi101101gad258483115

GENES amp DEVELOPMENT 29501ndash512 Published by Cold Spring Harbor Laboratory Press ISSN 0890-936915 wwwgenesdevorg 501

Cold Spring Harbor Laboratory Press on March 26 2015 - Published by genesdevcshlporgDownloaded from

on specific gene regulatory programs (Niehrs and Pollet1999) Spatiotemporal organization of gene expressionwhich is of particular relevance to cellular morphologyrelies heavily on the control of post-transcriptional eventsincluding mRNA export stability and translation tosustain cellular homeostasis RNA-binding proteins (RBPs)can synchronize the fates of multiple RNA molecules bybinding to particular secondary structures or sequencespresent in some RNAs but not others (Keene 2007 Li et al2010 Ray et al 2013) Moreover genome-wide studieshave found that by selective targeting individual RBPscoordinate post-transcriptional processing of whole co-horts of functionally related RNAs Such functionallycoherent proteinndashRNA units also known as lsquolsquoRNA op-eronsrsquorsquo (Keene 2007) substantially expand the regulatoryplasticity of the genomes endowing cells with tissue-specific functions and allowing for swift cellular responsesto the changing microenvironmentA group of CCCH-type zinc finger proteins has been

associated with different aspects of cellular asymmetryand several family members have shown the capacity torapidly alter gene expression programs Contrary to thegeneral notion that zinc finger domains bind DNA theCCCHmotif is thought to specialize in the recognition ofRNA (Hall 2005 Lunde et al 2007 Liang et al 2008)Accordingly the majority of the studied CCCH familymembers regulate different post-transcriptional processesacross species and several have been associatedwith humandisease including myotonic dystrophy (Miller et al 2000Wang et al 2012) autoimmune disorders (Vinuesa et al2005 Uehata et al 2013) and cancer (Rounbehler et al2012) However despite their coordinated control of phys-iologically related genes and their relevance to humandisease the roles of most of the CCCH zinc finger proteinsare still poorly understoodUnkempt is a conserved family member that harbors

a set of six tandem CCCHmotifs the largest such array inthe human genome (Fig 1A Liang et al 2008) Initiallydescribed in 1992 as an embryonically expressed gene infruit flies unkempt was shown to be essential for earlydevelopment its homozygous deletion led to larval lethal-ity while heterozygous flies carrying a hypomorphic alleledisplayed an lsquolsquounkemptrsquorsquo phenotype (Mohler et al 1992)Another recent report identified Unkempt as a neurogeniccomponent of the mTOR pathway suggesting that it mayact as a negative regulator of photoreceptor differentiationin fruit flies (Avet-Rochex et al 2014) However the exactfunction of Unkempt has remained obscure We hypoth-esized that Unkempt might regulate a gene expressionprogram with a critical role for a distinct aspect of cellularphysiology or development of specific cell lineages

Results

Unkempt is conserved across metazoans and isenriched in embryonic brains

Taking evolutionary conservation as a measure of func-tional significance we looked for bona fide orthologs ofhuman Unkempt protein and found them across theanimal kingdom but not in plants or fungi (Fig 1B

Figure 1 Evolutionary conservation and expression of Unkempt(A) Sketch of the human Unkempt protein showing the relativepositions of the six tandem CCCH zinc finger motifs and a RINGfinger domain Bar 100 amino acids (B) Unkempt protein isconserved across metazoans The cladogram shows the percentageof amino acid identities of Unkempt orthologs found in each of theindicated species compared with the full-length human orthologNote that no orthologs of Unkempt are found in nonmetazoanspecies Evolutionary relationships of animals shown are based onSrivastava et al (2010) (Adapted by permission from MacmillanPublishers Ltd 2010) RefSeq protein accession numbers of allof the indicated Unkempt orthologs are listed in SupplementalFigure S1A (C) Detection of Unkempt (UNK) in continuous celllines by immunoblotting (DH) Depletion of endogenous Un-kempt in SH-SY5Y cells detected by immunoblotting (D) andimmunofluorescence (H) using Unkempt-specific antibodies Bar10 mm (E) Immunohistochemistry of mouse E15 whole embryosagittal sections revealing highly enriched expression of Unkemptin the CNS The location of the CNS is highlighted by theexpression of the neuronal marker NeuN See also SupplementalFigure S1D (F) Expression of Unkempt in the cortex of E15 mouseembryos The cortical wall was probed with antibodies againstUnkempt (green) and neuronal marker Tuj-1 (red) and counter-stained with DAPI to label the nuclei (blue) The highlightedregion (orange rectangle) is shown magnified (middle) along witha close-up of the ventricular zone (right) together indicatinga pervasive expression of Unkempt throughout the cortical wallBars 10 mm (G) Immunocytochemistry of dissociated E15 neuronsin vitro showing the expression of Unkempt and the neuronalmarker Tuj-1 Bar 10 mm

Murn et al

502 GENES amp DEVELOPMENT

Cold Spring Harbor Laboratory Press on March 26 2015 - Published by genesdevcshlporgDownloaded from

Supplemental Fig S1A) Multiple sequence alignmentanalysis revealed a particularly deep evolutionary con-servation of all six tandemly arrayed CCCH zinc fingers(Supplemental Fig S1B) A query of the published expres-sion profiles of a wide range of mouse tissues and celllines found most abundant transcripts of Unkempt inmouse neuroblastoma cells consistent with its mRNAlocating to the CNS of a fly larva (Supplemental Fig S1CMohler et al 1992) This observation was confirmed inour survey of continuous cell lines and whole mouseembryos that revealed the highest expression of Unkemptprotein in a human cell line of neuronal origin (SH-SY5Y)and CNS respectively (Fig 1CndashE Supplemental Fig S1D)Unkempt appeared particularly abundant in mature neu-rons where it partitioned intomainly cytoplasmic punctasimilar to the pattern seen in SH-SY5Y cells (Fig 1FndashHSupplemental Fig S1EF) Whole mouse brains at differentstages of development showed induction of Unkempt atembryonic day 12 (E12) and a decline postnatally (Supple-mental Fig S1G) The rough temporal overlap with the

peak of neurogenesis and structuring of the brain sug-gested a broad regulatory role of Unkempt during theformation of the CNS

Control of early neuronal morphology and reshapingof nonneuronal cells by Unkempt

To examine the function of Unkempt in vivo we carriedout in utero electroporation of plasmids expressingshRNA and a fluorescent reporter to acutely silenceUnkempt in the developing CNS of mouse embryosImmunostaining of electroporated cortexes revealed a sig-nificant impact on neuronal migration and this effectpersisted postnatally (Fig 2AB Supplemental Fig S2AndashC) The observed defect in neuronal migration could berescued by coexpression of RNAi-resistant wild-typeUnkempt but not mutant Unkempt proteins lackingportions of the CCCH zinc finger domain (Fig 2AB seebelow) Upon a closer inspection of cellular morphologyas a key parameter in neuronal migration we noticed that

Figure 2 Unkempt is required for the earlyneuronal morphology and is sufficient to polar-ize cells of nonneuronal origin (AB) Impairedmigration of Unkempt-deficient neurons (A)Cortical sections of mouse embryos electropo-rated at E145 with the indicated constructs andanalyzed at E19 Electroporated neurons are ingreen Unk(WT) Unk(23) and Unk(1ndash6) areRNAi-resistant wild-type Unkempt mutantmissing zinc fingers 2 and 3 and mutant missingzinc fingers 1ndash6 (see also Fig 3A) (IZ) Interme-diate zone (CP) cortical plate (B) Quantificationof electroporated neurons as shown in A Thedata are based on the evaluation of at least 1000cells per condition (CD) Morphological analysisof migrating neurons (C) Computationallyreconstructed shapes of GFP-positive neuronsfrom the lower cortical plate of E19 embryoselectroporated with the indicated constructs (D)Quantification of GFP-positive neurons as in C

by the number of primary neurites per cell Thenumber of cells quantified for each indicatedcondition is shown in parentheses above eachcolumn (E) Efficiency of the RNAi constructsused in explanted neurons in vitro (FG) Im-paired morphogenesis of Unkempt-depleted cor-tical neurons in vitro (F) with quantification ofprimary neurites per cell (G) () P lt 0001Studentrsquos t-test Bar 10 mm (H) Representativeimages of SH-SY5Y cells growing in clusters(left) or individually (right) Bars 50 mm (IJ)GFP-inducible or GFP and Unkempt-inducibleHeLa cells at 36 h (I) and U2OS cells at 72 h (J) oftreatment with Dox Bars 50 mm (K) Overlaidoutlines of GFP-inducible () or GFP and Un-kempt-inducible (+) HeLa U2OS and HepG2cells at 72 h of treatment with Dox Error barsrepresent SD Bar 25 mm

Neuronal morphology program controlled by Unkempt

GENES amp DEVELOPMENT 503

Cold Spring Harbor Laboratory Press on March 26 2015 - Published by genesdevcshlporgDownloaded from

the majority of the poorly migrating Unkempt-deficientneurons had abnormally round cell bodies and extendedshort and numerous neurites (Fig 2CD SupplementalFig S2D) This was in contrast to the control as well asknockdown neurons rescued with RNAi-resistant wild-type Unkempt both of which displayed a typical bipolarshape that normally allows the migrating neurons toreach their final positions in the brain (Fig 2CD Noctoret al 2004) These data suggest that Unkempt is manda-tory for the early morphology of neurons during embry-onic development of miceAs the shape of neurons is influenced by a variety of

intracellular and extracellular cues we next askedwhether Unkempt regulates neuronal morphology ina cell-autonomous manner To that end we ablated theexpression of Unkempt in explanted cortical neurons andfollowed their morphogenesis in vitro Similar to thephenotype observed in utero knockdown of Unkempt inisolated neurons led to a reduction in neurite length anda dose-dependent increase in the number of primaryneurites compared with control (Fig 2EndashG) Moreoversilencing of Unkempt in human SH-SY5Y cells convertedthe early neuronal-like cellular shape into a roundermorphology with shorter but more numerous processesakin to the change seen in vivo (Fig 2H) Together theseresults indicate a cell-autonomous role of Unkempt inthe establishment andmaintenance of the early morphol-ogy of cortical neuronsThe broad expression pattern suggested that Unkempt

might be required for shaping other types of neurons inthe CNS as well We thus wished to explore the possibil-ity that the morphogenetic effect of Unkempt might becell type-independent and limited solely by its expres-sion We selected a set of continuous cell lines of diversebut nonneuronal origin including HeLa U2OS HepG2and immortalized mouse embryonic fibroblasts (iMEFs)

and engineered them to inducibly express a reporter-traceable exogenous Unkempt upon doxycycline (Dox)treatment (Supplemental Fig S2E) Remarkably as earlyas 12 h after the addition of Dox the inducible cells beganto showmorphological changes and some adopted a spin-dle-like shape (Supplemental Fig S2F) This phenotypebecame progressively more evident upon longer periodsof induction with the cells displaying an overt bipolarmorphology (Fig 2IndashK Supplemental Fig S2FndashM) Noneof the established polarity components including PARcomplex proteins CDC42 Smurf2 and others that wetested in this system was able to recapitulate the Un-kempt-induced phenotype signifying the unique role ofUnkempt in cell morphogenesis (Supplemental Fig S2NO) We observed no induction of neuronal markersincluding NeuN Tuj-1 Pax6 vimentin or nestin evenafter extended periods of treatment with Dox arguingagainst transdifferentiation of cells toward neuronal fate(data not shown) The fact that ectopic Unkempt inducesa similar phenotype in cells of unrelated origin suggeststhat Unkempt engages components of a specific mor-phology program that are not endemic to neurons but areexpressed ubiquitously

Unkempt-driven cell morphogenesis requires mRNAbinding

Given the highest degree of sequence conservationwithinthe CCCH zinc finger domain of Unkempt (SupplementalFig S1B) we hypothesized that this structural elementmight present a critical determinant of Unkempt-drivenshaping of cells We performed a structurendashfunctionanalysis in which we deleted different segments of theinducible Unkempt protein and investigated the impactof the created mutants on cellular shape (Fig 3A Sup-plemental Fig S2K) Unkempt lacking any portion of thezinc finger domain failed to induce a bipolar phenotype

Figure 3 Structurendashfunction analysis and iden-tification of Unkempt as an RBP (A) A series ofdeletion mutants of Unkempt protein examinedfor their capacity to impact cellular morphologyInternal deletions are indicated by bracketedregions (B) The morphologies of HeLa cellsinducibly expressing GFP alone or GFP andeither of the indicated Unkempt mutants werequantified by calculating their axial ratios (seeSupplemental Fig S2K) The results are com-pared with GFP control () P = 00007 () P lt

00001 (C) Outline of the Unkempt iCLIP ex-periment coupled with deep sequencing (see alsoSupplemental Table S1) (D) Binding of Unkemptto RNA in SH-SY5Y cells (Top) AutoradiogramThe bottom three stripes are immunoblotsof UNK and Actin (WT) Wild-type (KD)knockdown

Murn et al

504 GENES amp DEVELOPMENT

Cold Spring Harbor Laboratory Press on March 26 2015 - Published by genesdevcshlporgDownloaded from

whereas mutants lacking the C-terminal portions inducedthe bipolar morphology in a manner similar to full-lengthUnkempt (Fig 3B Supplemental Fig S3A) These datahighlight an essential role for the CCCH zinc fingersa putative RNA-binding domain in the Unkempt-drivenmorphological transformation of cellsTo investigate the RNA-binding capacity of Unkempt

we made use of iCLIP (individual nucleotide resolutionUV cross-linking and immunoprecipitation) a stringentmethod that allows for genome-wide detection as well asprecise mapping of proteinndashRNA interactions in livingcells (Fig 3C Konig et al 2010) Affinity purification ofendogenous Unkempt from UV-irradiated SH-SY5Y cellsfollowed by SDS-PAGE separation revealed the presenceof a single complex that corresponded in size to Unkemptbound to labeled RNA (Fig 3D) The signal produced bythis complex was dependent on the dose of the addedRNase I diminished upon knockdown of Unkempt andundetectable when either UV irradiation or Unkempt-specific antibodies were omitted A similar signal wasobtained with Unkempt protein overexpressed in HeLacells or endogenous Unkempt inmouse embryonic brains(see below) demonstrating the occurrence of UnkemptndashRNA interactions in vivo Importantly in contrast to thewild-type Unkempt the inactive Unkempt mutants lack-ing either the entire zinc finger domain or a part thereofshowed no appreciable affinity for RNA (Supplemental FigS3B) Moreover unlike the wild-type Unkempt RNA-binding-deficient mutant proteins were unable to rescuethe aberrant neuronal migration or the morphologicalabnormalities of Unkempt-deficient neurons (Fig 2AndashDSupplemental Fig S2AndashD) Taken together these resultsindicate a requirement for RNA binding by Unkempt inthe establishment of bipolar cell morphology in vitro andin vivoTo identify the RNA species targeted by Unkempt we

carried out RTndashPCR amplification of the UV cross-linkedRNA followed by high-throughput sequencing of theprepared cDNA libraries (Supplemental Fig S3C) Weperformed iCLIP experiments in three to five replicates ineach of the three biological contexts in which we observed

Unkempt-dependent cell morphology namely in SH-SY5Y cells whole brains of E15 mouse embryos and HeLacells ectopically expressing Unkempt (Supplemental TableS1) Genomic annotation of cDNA sequences found up to90 of all Unkempt-binding events mapping to mRNAswith more than half of all binding sites mapping to codingsequences (CDSs) (Fig 4AB) Importantly we observedlittle correlation between the number of iCLIP tags pertranscript and either transcript length or abundance in-dicating a highly selective manner in which Unkemptbinds its RNA targets (Supplemental Fig S4AB)To determine the identities of mRNAs bound by Un-

kempt we only considered iCLIP tags that mapped toregions within mature transcripts and could be unambig-uously assigned to a specific gene By taking into accounta particular binding pattern of Unkempt (see below)reproducibility of binding and the combined number ofunique iCLIP tags we identified 1186 1020 and 649Unkempt mRNA targets that repeatedly scored in SH-SY5Y cells HeLa cells and embryonic brains respectively(Supplemental Table S2AndashC) Comparison of the threedata sets defined a core subset of 263 genes that werebound by Unkempt in all three sample types displayingUnkempt-dependent cellular morphology (SupplementalFig S4C Supplemental Table S2D) While this list con-tains bona fide mRNA targets of Unkempt it is likelyincomplete due to the conservative selection criteriarestricting identification of low-abundance targetsFunctional annotation of Unkempt mRNA targets

strongly pointed to their involvement in general processesrelated to protein metabolism and trafficking (Fig 4CSupplemental Table S3) This was in contrast to severalpreviously studied RBPsmdashincluding NOVA proteinsfragile X mental retardation protein (FMRP) and neuro-nal Elav-like proteinsmdashthat primarily regulate neural-specific largely synapse-related transcripts in the brain(Ule et al 2005 Darnell et al 2011 Ince-Dunn et al 2012Wagnon et al 2012) A separate analysis of Unkempttargets interrogating canonical pathways revealed a strongenrichment ofmolecules implicated in translation initiationand p70S6K signaling the protein ubiquitination pathway

Figure 4 Unkempt-bound RNA species and themode of RNA binding (A) Distribution of UnkemptiCLIP tag clusters among different RNA segmentsin SH-SY5Y cells Similar distributions of clusterswere observed in HeLa cells and embryonic brains(data not shown) (B) Metatranscript analysis show-ing the positional frequency of Unkempt bindingsites along the length of all target mRNAs in SH-SY5Y cells (C) Gene ontology (GO) analysis ofUnkempt target transcripts in SH-SY5Y cells Thetop 10 GO terms are ranked by their P-values (seealso Supplemental Table S3) (D) Unkempt contactsits target transcripts commonly at just one domi-nant binding site Snapshots from the University ofCalifornia at Santa Cruz Genome Browser (httpgenomeucscedu human assembly GRCh37hg19and mouse assembly GRCm38mm10 Kent et al2002) of different Unkempt target genes depictbinding positions in the indicated sample types(see also Supplemental Fig S6A)

Neuronal morphology program controlled by Unkempt

GENES amp DEVELOPMENT 505

Cold Spring Harbor Laboratory Press on March 26 2015 - Published by genesdevcshlporgDownloaded from

and signaling by the Rho and Ran families of GTPases(Supplemental Fig S4D Supplemental Table S4) To-gether these analyses indicate that much of the Un-kempt-bound transcriptome is represented by mRNAsfunctionally linked to protein turnover in addition totranscripts encoding proteins with direct roles in regulat-ing cellular shapeTo probe the functional dependence of Unkempt-

induced cellular morphology we carried out a focusedloss-of-function screen in inducible HeLa cells by silenc-ing the expression of 34 of the strongest Unkempt targetsand as a control an equal number of nontargeted mole-cules with known roles in morphology-related processes(Supplemental Fig S5AB Supplemental Table S5) Al-though depletion of no single molecule except Unkemptitself completely suppressed the reshaping of cells uponinduction of Unkempt several of the tested componentsexhibitedmilder phenotypic effects Notably only a smallproportion of the nontargeted molecules (11 of 34 32)affected the Unkempt-induced morphology despite theirknown roles in cell shape control In contrast knock-down of close to half (16 of 34 47) of the strongly boundtarget genes interfered with cell morphogenesis many ofthem (eg S100A11 PSMD12 CCT5 and HSPA8) with-out a prior record in cell shape regulation (SupplementalFig S5B) Because of the particular regulatory function ofUnkempt and its mode of RNA binding (see below) wefurther examined the effects of several of the identifiedmolecules by perturbing their levels through overexpres-sion of the corresponding cDNAs lackingUnkempt-bindingsites (Supplemental Fig S5CD see below) Notably onlythose target molecules that scored in the loss-of-functionscreen moderately suppressed the reshaping of cells uponoverexpression suggesting that expression levels of theseUnkempt target genes play a critical role in Unkempt-induced cell morphogenesis (Supplemental Fig S5E) Theseresults point to the complexity of the process and suggestthat Unkempt acts as a hub to coordinate not one but ratheramultitude ofmolecular pathways to bring about amorpho-logical transformation of cells

A unique mode of RNA binding and the Unkemptrecognition element

Global analysis of all binding sites showed their ubiquitousdistribution within the CDS as the most densely populatedmRNA segment (Fig 4B) However the examination ofoccupancy of Unkempt on individual transcript targetsrevealed unique narrowly defined sites of contact commonlywith just one dominant binding site on the message (Fig 4DSupplemental Fig S6A) This pattern differsmarkedly fromthose reported for other CDS-bindingRBPs (including FMRPand LIN28) that exhibit a broad distribution of binding sitesalong the coding regions ofmRNAs (Darnell et al 2011 Choet al 2012) The precise positions of Unkempt target siteswere well conserved between SH-SY5Y cells and the in-ducible HeLa cells indicating the maintained specificity forRNA binding regardless of whether Unkemptwas expressedendogenously or forcefully introduced into cells (Fig 4DSupplemental Fig S6A)

To examine the RNA sequence specificity of Unkemptwe searched for enrichment of all possible pentamers inthe vicinity of the cross-link sites (Fig 5AB) With somevariation betweendifferent sample types twomotifs emergedfrom all iCLIP data sets a U-rich motif and a UAG-containing motif (Fig 5B) To resolve this dichotomy wealigned sequences spanning the cross-link sites of severalstrongly bound transcripts and found a U-rich regionalmost invariably located precisely at the cross-link sites(Fig 5AC) A further manual inspection identified a com-mon UAG trimer at a distance of just a few nucleotides 59to the cross-link site (Fig 5C) A global analysis of allUnkempt target transcripts confirmed amarked enrichmentof the UAG motif consistently occurring just upstream ofthe cross-link sites in all three sample types (Fig 5D) Thisfixed linear arrangement of the UAG triplet and the U-richstretch suggested that bothmotifsmight contribute to targetrecognition by UnkemptTo evaluate the functional relevance of these results

we first performed electrophoretic mobility shift assay(EMSA) using short synthetic RNAs as substrates fora full-length recombinant mouse Unkempt protein(Supplemental Fig S6B) Wild-type sequences encom-passing binding sites of two Unkempt target mRNAsmdashhumanHSPA8 andmouse Ptnmdashbound to Unkempt witha dissociation constant in the nanomolar range (Fig 5E)Markedly mutating the UAG trimer essentially abolishedany detectable affinity of RNA for Unkempt (Fig 5E)Randomization of the nucleotides outside of either de-duced motif did not affect the binding while alterationsof the UAG trimer in the context of a randomer includingsingle nucleotide substitutions substantially reduced theaffinity of Unkempt for RNA (Fig 5E Supplemental FigS6C) The U-rich region displayed a smaller but noticeableeffect on binding replacement of Ursquos with Arsquos preservedthe affinity while substitutions with Crsquos or Grsquos resulted indecreased binding Together these data identify a consen-sus Unkempt recognition element consisting of a manda-tory UAG trimer upstream of a UA-rich motif Globallywe found this element present within binding sites of56ndash72 of mRNAs targets indicating its dominant roleas a specificity determinant for binding by Unkempt(Supplemental Fig S6D Supplemental Table S2) Thecross-link sites of Unkempt thus appear shifted to onlyone of the two binding motifs within the Unkempt-binding element likely due to a slight uridine bias of theUV light (Sugimoto et al 2012)We further examined the RNA-binding affinity of

Unkempt by using isothermal titration calorimetry(ITC) to monitor the binding of the zinc finger domainencompassing all six CCCH zinc fingers (ZF1ndash6) orseparately either of the two sets of three highly con-served zinc fingers (ZF1ndash3 and ZF4ndash6) to the HSPA8-binding site (Fig 5F) While each set of three zinc fingersbound to the 18-mer RNA with a moderate affinity(Kd

ZF1ndash3 = 50 mM and KdZF4ndash6 = 16 mM) the entire

CCCH domain bound to the same oligonucleotide witha significantly lower dissociation constant (Kd

ZF1ndash6 = 02mM) suggesting a cooperative binding of RNA by thezinc fingers of Unkempt

Murn et al

506 GENES amp DEVELOPMENT

Cold Spring Harbor Laboratory Press on March 26 2015 - Published by genesdevcshlporgDownloaded from

Unkempt reduces translational efficiency of targetmRNAs

To understand the functional significance of mRNA bind-ing by Unkempt we first considered the possibility thatUnkempt may regulate the stability of its target tran-scripts analogous to some other CCCH family members(Carballo et al 1998 Matsushita et al 2009 Leppek et al2013) However differential expression analysis upon de-pletion of Unkempt in SH-SY5Y cells showed little corre-lation with the number of iCLIP tags per transcript andsuggested that overall the RNA binding did not have animpact on the steady-state levels of the targeted messages(Supplemental Fig S7AB)The binding preference for coding regions led us to

speculate that Unkempt may regulate translation of itsbound messages analogous to some other CDS-bindingRBPs (Fig 4AB Abdelmohsen et al 2011 Darnell et al2011 Cho et al 2012 Brummer et al 2013) To determinewhether Unkempt associates with polyribosomes we

fractionated the lysates of SH-SY5Y cells and mouse embry-onic brains on linear sucrose density gradients and examinedthe sedimentation pattern of Unkempt (Fig 6A Supplemen-tal Fig 8A) A significant proportion of Unkempt cosedi-mented with fractions containing heavy polyribosomes andshowed a distribution akin to FMRP a knownpolyribosome-associated protein (Stefani et al 2004) Treatment of lysateswith EDTA to dissociate the large from the small ribosomalsubunits disrupted the polysomes and shifted Unkempt tolighter fractions Binding of Unkempt to polyribosomes wasalso RNA-dependent since digestion of the pooled heavysucrose fractions with RNase I or micrococcal nuclease(MNase) quantitatively released Unkempt from ribosomesinto the soluble fraction similar to poly(A)-bindingprotein (PABP) while a component of the 40S ribosomalsubunit (RPS3) was resistant to RNA cleavage (Fig 6B)Taken together these results suggest that the association ofUnkempt with polyribosomes is dependent on Unkemptbinding to both the ribosomes and the target mRNA

Figure 5 The RNA recognition element of Un-kempt (A) Magnified regions of two Unkempttarget genes revealing DNA sequences and theencoded amino acids at each major cross-linksite (B) k-mer analysis in the vicinity of all cross-link sites for each sample type The most highlyenriched RNA pentamers are ranked by the Z-score (C) Alignment of several target mRNAregions harboring the major cross-link sites (blue)and the identification of the conserved UAGmotif (red) (D) Global enrichment of the UAGmotif in the vicinity of Unkempt-binding sites ontarget mRNAs (E) EMSA demonstrating a bind-ing requirement for the intact UAG motif (red)and enhanced RNA binding of recombinant Un-kempt (rUnk) in the presence of the UA-richmotif (blue) Mutations within either bindingmotif are highlighted in green and underlinedNanomolar concentrations of rUnk used in allassays are indicated (see also Supplemental FigS6C) (F) ITC binding curves of complex forma-tion between the indicated domains of Unkemptand the 18-mer HSPA8-binding site (Kd) Disso-ciation constant

Neuronal morphology program controlled by Unkempt

GENES amp DEVELOPMENT 507

Cold Spring Harbor Laboratory Press on March 26 2015 - Published by genesdevcshlporgDownloaded from

In order to determine whether Unkempt directly im-pacts protein synthesis we carried out ribosome profiling(Ingolia et al 2009) to obtain a genome-wide view ofribosome occupancy on Unkempt target messages (Sup-plemental Fig S8B) Comparative profiling of ribosome-protected fragments (RPFs) from HeLa cells with or

without ectopic expression of Unkempt revealed severaltraits typical of translation including 3-nucleotide (nt)periodicity pausing of ribosomes in the proximity ofstart and stop codons and retention of ribosomes in the59 untranslated region (UTR) segment (SupplementalFig S8C)

Figure 6 Unkempt represses translation of its target messages (AB) RNA-dependent association of Unkempt with polyribosomes (A)Polysome profiling of SH-SY5Y cells harvested in the presence of cycloheximide (CHX) or EDTA and immunoblot analysis of individualfractions for the indicated proteins See also Supplemental Figure S8A (B) Digestion of polyribosomal fractions with RNase I or MNasefollowed by immunoblot analysis of the indicated proteins for their release into the supernatant (S) from the pelleted (P) fraction (IN)Input (C) Ribosome profiling data showing total numbers of genes with significantly decreased (Down) or increased (Up) RPF count inDox-treated GFP and Unkempt-inducible versus GFP-only-inducible HeLa cells (false discovery rate [FDR] lt 5) (D) Fractions of geneswith altered RPF counts as shown in C that are bound by Unkempt considering all binding sites (All BS) or binding sites harboring theUAG motif (BS wUAG) (EndashG) Cumulative distributions of changes in ribosome occupancy for all transcripts containing Unkempt-binding sites (Bound) transcripts with one (One BS) two (Two BS) or three or more (Three+ BS) binding sites transcripts without theUAG motif (No UAG) with one UAG motif (One UAG) or with two or more UAG motifs (Two+ UAG) within the binding sites andcontrol transcripts lacking Unkempt-binding sites (Not bound) The number of genes in each category is indicated in parenthesesComparison of either set of Unkempt target transcripts with the nonbound controls showed a significant difference (P lt 00001) (HndashJ)Translational repression by Unkempt requires the UAG motif (H) Extension of the luciferase gene with native (NAT) or point-mutated(MUT) sequence corresponding to the Unkempt-binding site within the human S100A11 transcript The mutation in the critical UAGmotif (underlined) preserves the amino acid (Leu) encoded by the affected codon (vertical dashed lines) (IJ) Dual-luciferase assay usingthe native or point-mutated hybrid luciferase in wild-type (WT) and Unkempt knockdown (KD) SH-SY5Y cells (see also SupplementalFig S9C) Relative levels of the hybrid luciferase transcripts (I) and relative luminescence units (RLU) (J) are shown Error bars indicateSD (n = 3) () P lt 005 Studentrsquos t-test

Murn et al

508 GENES amp DEVELOPMENT

Cold Spring Harbor Laboratory Press on March 26 2015 - Published by genesdevcshlporgDownloaded from

Changes in the rate of translation have been shown tocorrelate with changes in ribosome occupancy of mRNAs(Ingolia et al 2009) Our global analysis of ribosomeprofiling data indicated a greater number of genes witha significantly reduced RPF count compared with geneswith an increased RPF count upon induction of Unkemptin HeLa cells (Fig 6C) Markedly the bias became sub-stantially more apparent when we considered the RNA-binding information Unkempt was found to largely targetgenes with a reduced RPF count and this trend was mostprominent for the subset of target genes harboring theUAGmotif within Unkempt-binding sites (Fig 6D) Theseobservations hinted at a repressive effect of Unkempt ontranslation of its bound transcriptsTo examine the global impact of UnkemptndashRNA in-

teractions on ribosome occupancy of the targeted tran-scripts we binned the transcripts according to theirbinding to Unkempt number of binding sites and pres-ence or absence of the UAG motif Markedly Unkempttarget genes exhibited a significant drop in ribosomeoccupancy compared with nontargets an effect thatfurther increased in the presence of multiple bindingsites (Fig 6EndashG) These data demonstrate that Unkemptreduces translational efficiency of its target mRNAs bylowering the ribosome occupancy without concurrentchanges in transcript abundance It should be noted thatsince only a fraction of all cells analyzed by ribosomeprofiling successfully induce Unkempt upon treatmentwith Dox (31) (Supplemental Fig S9A) the observedimpact on translational efficiency is likely an underesti-mate of the repressive activity of UnkemptTo validate the inhibitory effect of Unkempt on trans-

lation in neuronal cells we carried out immunoblottingfor high-confidence target genes that showed no signifi-cant changes at the transcript level upon knockdown inSH-SY5Y cells Proteins CCT5 HNRNPK and DDX5 allof which are encoded by mRNAs that rank among the top10 of all Unkempt targets (Supplemental Table S2A)were expressed at notably higher levels in Unkempt-deficient compared with wild-type SH-SY5Y cells (Supple-mental Fig S9B) In contrast nontarget controlsmdashGAPDHand histone H3mdashwere expressed at comparable levels inboth conditions

The RNA recognition element is requiredfor translational control by Unkempt

To determine whether the translational repression ofUnkempt target genes in vivo requires a direct interac-tion between Unkempt and the identified RNA recogni-tion element we used a modified dual-luciferase reporterassayWe took awell-definedRNA-binding site of a strongUnkempt target gene S100A11 and inserted it proximalto the C terminus of the firefly luciferase gene so as tomimic the respective position in the endogenous genewhile preserving the reading frame (Fig 6H) We alsoprepared a mutant construct in which we converted thecritical UAGmotif into UCG which retained the encodedamino acid (leucine) but abolished the binding by Un-kempt as informed by the EMSA assay (Fig 5E) Upon

transfection of either of these two constructs along withthe Renilla luciferase vector into either control or Un-kempt-depleted SH-SY5Y cells we detected comparablelevels of each modified luciferase mRNA by RT-qPCRanalysis (Fig 6I) However measurements of lumines-cence in the control cells revealed about a threefold lowerabundance of the modified luciferase produced from theconstruct containing the native binding site sequence(NAT) compared with the construct with the pointmutation (MUT) (Fig 6J) In contrast Unkempt-depletedcells showed amuch smaller albeit significant differenceof 25 in firefly luciferase activity We repeated theexperiment with a binding site of a different top-scoringmRNA target of UnkemptDpy30 and observed a similarif not more profound repressive effect of UnkemptndashmRNA interaction on translation (about fivefold) thatwas essentially eliminated upon depletion of Unkempt(Fig 6J Supplemental Fig S9C) Taken together thesedata indicate that the translational control by Unkemptcritically depends on recognition of its binding elementwithin target mRNAsSimilar to Unkempt a few other RBPs have been

proposed to largely act by regulating translation of targettranscripts although the mechanisms of their transla-tional control are not well understood (Polesskaya et al2007 Abdelmohsen et al 2011 Darnell et al 2011 Penget al 2011 Cho et al 2012 Kwan et al 2012Wilbert et al2012 Brummer et al 2013) FMRP which like Unkemptprimarily targets coding regions of mRNAs was shown torepress protein synthesis by stalling the translocatingribosomes (Darnell et al 2011) However a runoff exper-iment with puromycin revealed that Unkempt unlikeFMRP shifted to lighter fractions of a sucrose gradientsuggesting its association with actively translocating butnot stalled ribosomes (Supplemental Fig S9D) Moreovera global correlation of RNA-binding positions of Un-kempt with ribosome profiling data unveiled enrichedbinding of Unkempt to sites on mRNAs just upstream ofendogenously paused ribosomes inconsistent with theribosome stalling model in which an RBP would beexpected to block the translocation of ribosomes andlocate downstream from their clusters (Supplemental FigS9E) We hypothesize that translational repression byUnkempt entails a different process such as interferencewith translation initiation which is a generally rate-limiting and commonly regulated step (SupplementalFig S9F Besse and Ephrussi 2008)

Discussion

Coregulation of physiologically related genes by RBPs haspreviously been linked with different aspects of cellularmorphogenesis For instance in mammalian brainsa multitude of RBPs implicated in neurological disordersincluding FUS FMRP NOVA SAM68 and Staufen pro-teins regulate mRNAs with synaptic functions and affectthemorphology of neuronal dendritic spines (Fujii et al 2005Ule et al 2005 Goetze et al 2006 Dictenberg et al 2008Vessey et al 2008 Ruggiu et al 2009 Darnell et al 2011Iijima et al 2011 Wagnon et al 2012 Klein et al 2013) In

Neuronal morphology program controlled by Unkempt

GENES amp DEVELOPMENT 509

Cold Spring Harbor Laboratory Press on March 26 2015 - Published by genesdevcshlporgDownloaded from

contrast to these RBPs our data suggest that Unkemptaffects neuronal morphology by regulating a molecularprogram that is intrinsic to cells of diverse lineages Suchtarget selectivity may explain at least in part the capacityof ectopic Unkempt to establish a neuronal-like mor-phology in nonneuronal cells further suggesting thattissues other than brain could use Unkempt for shapingof their cells Consistent with this possibility the initialreport onUnkempt noted its widespread expression duringthe earliest stages of fruit fly development (Mohler et al1992) and expression analysis in mice indicates the pres-ence of moderate levels of Unkempt in numerous tissues(Supplemental Fig S1C) Future studies are warranted toinvestigate the effects ofUnkempt on cellmorphogenesis indifferent tissues Of note since acute gene silencing usingshRNA delivery by in utero electroporation can lead to off-target effects and associated artifacts (Baek et al 2014) invivo analyses with additional less error-prone approacheswill be essentialThe mode of Unkempt binding to target mRNAs is

unexpected and differs from that of most other RBPsstudied to date Indeed the extent of sequence specificityof several other RBPs seems comparable with that ofUnkempt yet Unkempt exhibits a higher definition ofbinding Thus there likely exist additional determinants toguide Unkempt to its unique binding sites on cognatetranscripts This assumption is supported by the observa-tion that the consensus Unkempt recognition elementUAG_gap_WWW relatively poorly predicts the actualbinding sites By varying the length of the spacer (gap) be-tween both motifs and the definition of the UA-rich motifwe found that the sequences matching UAGNNUUUconsensus predicted the actual binding sites with thegreatest specificity still out of all sites within the CDSmatching this sequence only 23 were bound by Un-kempt representing just 11 of all Unkempt-bindingsites in the CDS We speculate that one auxiliary land-mark could be provided by the proximity or cobinding ofribosomes which could also reduce the need for sequencescanning by Unkempt along the targeted mRNA How-ever the exact reason and the requirement if any for thecommonly singular contact sites of Unkempt on mRNAsremain to be investigatedThe association of Unkempt with large polyribosomes

along with its impact on ribosome occupancy of targetmRNAs posit translation as a key post-transcriptionalprocess regulated by Unkempt The fact that Unkemptcontrols translation of proteins that themselves regulatetranslation as well as the cytoskeleton and traffickingsuggests a highly hierarchical structure of the RNAoperon in which Unkempt plays the role of the lsquolsquoregulatorof the regulatorsrsquorsquo This idea is further supported by thefinding of several RBPs (each regulating its own molecu-lar program) among Unkempt targets including PARK7RBFOX2 Staufen proteins ELAVL4 and several hetero-geneous nuclear ribonucleoproteins (Supplemental TableS2) One implication of such hierarchical activity is thatUnkempt does not need to act at the actual sites ofstructural rearrangements during morphogenesis but caninstate through translational control a global molecular

program that in turn remodels cellular morphology Itshould be noted that Unkempt might regulate cellmorphogenesis by additional means that are unrelatedto its RNA binding in the cytoplasm including forexample a putative nuclear activity and interactionswith other proteinsUnkemptrsquos sequence specificity mode of mRNA bind-

ing and potent impact on protein translation togetherform an RNA operon that can be extremely sensitive tomutations within the binding sites of Unkempt As seenwith point mutations of the hybrid luciferase transcriptsa single nucleotide exchange can have a profound effecton protein levels without affecting the transcript abundanceor the encoded amino acid sequence Broadly speaking thissuggests that synonymous geneticmutations also known aslsquolsquosilentrsquorsquo mutations could have a substantial impact on geneexpression through the action of RBPs with narrowlydefined position-dependent regulatory capacities compa-rable with that of Unkempt With hundreds of unstudiedRBPs encoded by the human genome such sensitivity topoint mutations could provide a novel explanation for thefrequent but often neglected association of synonymousmutations with human disease Given the severity of thephenotypes observed upon depletion of Unkempt in fliesand mice one could envision that such mutations orcompromised activity of Unkempt protein itself wouldmanifest in embryonic lethality or give rise to debilitat-ing neurological disorders in humans

Materials and methods

iCLIP experiments

All iCLIP experiments on SH-SY5Y cells HeLa cells and mouseembryonic brains were carried out in replicates using polyclonalrabbit anti-UNK antibody from Sigma (HPA023636) by adheringto a published protocol (Konig et al 2011) See also the Supple-mental Material

RNA sequencing (RNA-seq)

RNA-seq libraries were prepared from total RNA extracted fromSH-SY5Y cells transfected with nontargeting or UNK targetingsiRNA as described in the Supplemental Material Total RNAwas polyA-selected fragmented reverse-transcribed and se-quenced according to the TruSeq protocol (Illumina)

Ribosome footprinting

HeLa cells inducibly expressing GFP and Unkempt or GFP alonewere lysed and the lysates were treated with RNase I and spun topellet the ribosomes Ribosome-protected RNAwas isolated anddeep sequencing libraries were generated and sequenced asdescribed in the Supplemental Material

Polysome profiling

Whole mouse embryonic brains or SH-SY5Y cells were treatedwith cycloheximide lysed layered on top of a sucrose gradientand centrifuged The gradient was fractionated concomitantwith recording of the absorbance and proteins from each fractionwere precipitated and analyzed by immunoblotting

Murn et al

510 GENES amp DEVELOPMENT

Cold Spring Harbor Laboratory Press on March 26 2015 - Published by genesdevcshlporgDownloaded from

Purification of recombinant Unkempt and EMSAs

Full-length mouse Unkempt was purified from Dox-inducibleHeLa S3 cells by immunoaffinity purification Synthetic oligo-ribonucleotides were radioactively labeled and incubated withthe recombinant Unkempt and RNA binding was assessed bynative gel electrophoresis

Functional classification of Unkempt target genes

The enrichment of gene ontology categories in each set of Un-kempt targets was analyzed using the online tool DAVID (httpdavidabccncifcrfgov) Pathway analysis was performed using theWeb-based Ingenuity Pathway Analysis program (Ingenuity Sys-tems http wwwingenuitycom)

Analysis of high-throughput sequencing data

Processing and genomic mapping of all high-throughput se-quencing reads including the analyses of differential transcriptabundance (RNA-seq data) identification and characterization ofUnkempt-binding sites (iCLIP data) and analyses of differentialribosome occupancy (ribosome profiling data) are described indetail in the Supplemental Material

Accession numbers

The ArrayExpress accession numbers for the iCLIP RNA-seqand ribosome profiling data are E-MTAB-2279 E-MTAB-2277and E-MTAB-2278 respectively

Acknowledgments

We thankYoichiro Sugimoto AndreaDrsquoAmbrogio TomChittendenand Angeles Fernandez-Gonzales for advice and assistancewith the experiments Timothy Mitchison Julian Keuroonig LindaVanAelstMarc KirschnerNicholas Ingolia David PellmanDavidvan Vactor John Flanagan and members of the Shi laboratory fordiscussion and comments on the manuscript Crt Gorup for hiswork on the iCount pipeline and Francois Guillemot for thepCAGGS-IRES-RFP vector This study was supported by theNancy Lurie Marks Post-doctoral Fellowship (JM) an EMBLInterdisciplinary Postdocs (EIPOD) fellowship (KZ) Stuart HQand Victoria Quan Fellowship (YJY) National Institute of GeneralMedical Sciences grant T32GM007753 (YJY) National Instituteof Neurological Disorders and Stroke grants R01 NS032457 andR01 NS35129 (CAW) the Manton Center for Orphan DiseaseResearch (CAW) and National Institutes of Health grantsGM058012 CA118487 and MH096066 (YS) CAW is an In-vestigator of the Howard Hughes Medical Institute YS is anAmerican Cancer Society Research Professor YS is also a co-founder of Constellation Pharmaceuticals Inc and a member ofits scientific advisory board

References

Abdelmohsen K Tominaga K Lee EK Srikantan S Kang MJKimMM Selimyan R Martindale JL Yang X Carrier F et al2011 Enhanced translation by Nucleolin via G-rich ele-ments in coding and non-coding regions of target mRNAsNucleic Acids Res 39 8513ndash8530

Avet-Rochex A Carvajal N Christoforou CP Yeung KMaierbrugger KT Hobbs C Lalli G Cagin U Plachot CMcNeill H et al 2014 Unkempt is negatively regulated bymTOR and uncouples neuronal differentiation from growthcontrol PLoS Genet 10 e1004624

Baek ST Kerjan G Bielas SL Lee JE Fenstermaker AG NovarinoG Gleeson JG 2014 Off-target effect of doublecortin familyshRNA on neuronal migration associated with endogenousmicroRNA dysregulation Neuron 82 1255ndash1262

Besse F Ephrussi A 2008 Translational control of localizedmRNAs restricting protein synthesis in space and time NatRev Mol Cell Biol 9 971ndash980

Brummer A Kishore S Subasic D Hengartner M Zavolan M2013 Modeling the binding specificity of the RNA-bindingprotein GLD-1 suggests a function of coding region-locatedsites in translational repression RNA 19 1317ndash1326

Carballo E Lai WS Blackshear PJ 1998 Feedback inhibition ofmacrophage tumor necrosis factor-a production by triste-traprolin Science 281 1001ndash1005

Cho J Chang H Kwon SC Kim B Kim Y Choe J HaM Kim YKKim VN 2012 LIN28A is a suppressor of ER-associatedtranslation in embryonic stem cells Cell 151 765ndash777

Darnell JC Van Driesche SJ Zhang C Hung KY Mele A Fraser CEStone EF Chen C Fak JJ Chi SW et al 2011 FMRP stallsribosomal translocation on mRNAs linked to synaptic functionand autism Cell 146 247ndash261

Davis RL Weintraub H Lassar AB 1987 Expression of a singletransfected cDNA converts fibroblasts to myoblasts Cell 51987ndash1000

Dictenberg JB Swanger SA Antar LN Singer RH Bassell GJ2008 A direct role for FMRP in activity-dependent dendriticmRNA transport links filopodial-spine morphogenesis tofragile X syndrome Dev Cell 14 926ndash939

Dotti CG Sullivan CA Banker GA 1988 The establishment ofpolarity by hippocampal neurons in culture J Neurosci 81454ndash1468

Fujii R Okabe S Urushido T Inoue K Yoshimura A TachibanaT Nishikawa T Hicks GG Takumi T 2005 The RNAbinding protein TLS is translocated to dendritic spines bymGluR5 activation and regulates spine morphology Curr Biol

15 587ndash593Goetze B Tuebing F Xie Y Dorostkar MM Thomas S Pehl U

Boehm S Macchi P Kiebler MA 2006 The brain-specificdouble-stranded RNA-binding protein Staufen2 is requiredfor dendritic spine morphogenesis J Cell Biol 172 221ndash231

Hall TM 2005 Multiple modes of RNA recognition by zincfinger proteins Curr Opin Struct Biol 15 367ndash373

Iijima T Wu K Witte H Hanno-Iijima Y Glatter T Richard SScheiffele P 2011 SAM68 regulates neuronal activity-dependentalternative splicing of Neurexin-1 Cell 147 1601ndash1614

Ince-Dunn G Okano HJ Jensen KB Park WY Zhong R Ule JMele A Fak JJ Yang C Zhang C et al 2012 Neuronal Elav-like (Hu) proteins regulate RNA splicing and abundance tocontrol glutamate levels and neuronal excitabilityNeuron 751067ndash1080

Ingolia NT Ghaemmaghami S Newman JR Weissman JS 2009Genome-wide analysis in vivo of translation with nucleotideresolution using ribosome profiling Science 324 218ndash223

Keene JD 2007 RNA regulons coordination of post-transcriptionalevents Nat Rev Genet 8 533ndash543

Kent WJ Sugnet CW Furey TS Roskin KM Pringle TH ZahlerAM Haussler D 2002 The human genome browser atUCSC Genome Res 12 996ndash1006

Klein ME Younts TJ Castillo PE Jordan BA 2013 RNA-bindingprotein Sam68 controls synapse number and local b-actinmRNA metabolism in dendrites Proc Natl Acad Sci 1103125ndash3130

Konig J Zarnack K Rot G Curk T Kayikci M Zupan B TurnerDJ Luscombe NM Ule J 2010 iCLIP reveals the function ofhnRNP particles in splicing at individual nucleotide resolutionNat Struct Mol Biol 17 909ndash915

Neuronal morphology program controlled by Unkempt

GENES amp DEVELOPMENT 511

Cold Spring Harbor Laboratory Press on March 26 2015 - Published by genesdevcshlporgDownloaded from

Konig J Zarnack K Rot G Curk T Kayikci M Zupan B TurnerDJ Luscombe NM Ule J 2011 iCLIPmdashtranscriptome-widemapping of proteinndashRNA interactions with individual nu-cleotide resolution J Vis Exp 50 e2638

Kwan KY Lam MM Johnson MB Dube U Shim S Rasin MRSousa AM Fertuzinhos S Chen JG Arellano JI et al 2012Species-dependent posttranscriptional regulation of NOS1by FMRP in the developing cerebral cortex Cell 149 899ndash911

Leppek K Schott J Reitter S Poetz F HammondMC Stoecklin G2013 Roquin promotes constitutive mRNA decay via a con-served class of stem-loop recognition motifsCell 153 869ndash881

Li X Quon G Lipshitz HD Morris Q 2010 Predicting in vivobinding sites of RNA-binding proteins using mRNA second-ary structure RNA 16 1096ndash1107

Liang J Song W Tromp G Kolattukudy PE Fu M 2008Genome-wide survey and expression profiling of CCCH-zincfinger family reveals a functional module in macrophageactivation PLoS ONE 3 e2880

Lunde BM Moore C Varani G 2007 RNA-binding proteinsmodular design for efficient function Nat Rev Mol Cell Biol8 479ndash490

Matsushita K Takeuchi O Standley DM Kumagai Y Kawagoe TMiyake T Satoh T Kato H Tsujimura T Nakamura H et al2009 Zc3h12a is an RNase essential for controlling immuneresponses by regulating mRNA decay Nature 458 1185ndash1190

Miller JW Urbinati CR Teng-Umnuay P Stenberg MG ByrneBJ Thornton CA Swanson MS 2000 Recruitment of humanmuscleblind proteins to (CUG)(n) expansions associated withmyotonic dystrophy EMBO J 19 4439ndash4448

Mohler J Weiss N Murli S Mohammadi S Vani K Vasilakis GSong CH Epstein A Kuang T English J et al 1992 Theembryonically active gene unkempt of Drosophila encodesa Cys3His finger protein Genetics 131 377ndash388

Niehrs C Pollet N 1999 Synexpression groups in eukaryotesNature 402 483ndash487

Noctor SC Martinez-Cerdeno V Ivic L Kriegstein AR 2004Cortical neurons arise in symmetric and asymmetric divisionzones and migrate through specific phases Nat Neurosci 7136ndash144

Peng S Chen LL Lei XX Yang L Lin H Carmichael GG Huang Y2011 Genome-wide studies reveal that Lin28 enhances thetranslation of genes important for growth and survival ofhuman embryonic stem cells Stem Cells 29 496ndash504

Polesskaya A Cuvellier S Naguibneva I Duquet A Moss EGHarel-Bellan A 2007 Lin-28 binds IGF-2 mRNA and partic-ipates in skeletal myogenesis by increasing translationefficiency Genes Dev 21 1125ndash1138

Ray D Kazan H Cook KB Weirauch MT Najafabadi HS Li XGueroussov S Albu M Zheng H Yang A et al 2013 Acompendium of RNA-binding motifs for decoding generegulation Nature 499 172ndash177

Rounbehler RJ Fallahi M Yang C SteevesMA Li W Doherty JRSchaub FX Sanduja S Dixon DA Blackshear PJ et al 2012Tristetraprolin impairs myc-induced lymphoma and abolishesthe malignant state Cell 150 563ndash574

Ruggiu M Herbst R Kim N Jevsek M Fak JJ Mann MAFischbach G Burden SJ Darnell RB 2009 Rescuing Z+ agrinsplicing in Nova null mice restores synapse formation andunmasks a physiologic defect in motor neuron firing ProcNatl Acad Sci 106 3513ndash3518

Sato T Katagiri K Gohbara A Inoue K Ogonuki N Ogura AKubota Y Ogawa T 2011 In vitro production of functionalsperm in cultured neonatal mouse testesNature 471 504ndash507

Srivastava M Simakov O Chapman J Fahey B Gauthier MEMitros T Richards GS Conaco C Dacre M Hellsten U et al

2010 The Amphimedon queenslandica genome and theevolution of animal complexity Nature 466 720ndash726

Stefani G Fraser CE Darnell JC Darnell RB 2004 Fragile Xmental retardation protein is associated with translatingpolyribosomes in neuronal cells J Neurosci 24 7272ndash7276

Sugimoto Y Konig J Hussain S Zupan B Curk T Frye M Ule J2012 Analysis of CLIP and iCLIP methods for nucleotide-resolution studies of proteinndashRNA interactions Genome

Biol 13 R67Uehata T Iwasaki H Vandenbon A Matsushita K Hernandez-

Cuellar E Kuniyoshi K Satoh T Mino T Suzuki Y StandleyDM et al 2013 Malt1-induced cleavage of regnase-1 in CD4+

helper Tcells regulates immune activation Cell 153 1036ndash1049Ule J Ule A Spencer J Williams A Hu JS Cline M Wang H

Clark T Fraser C Ruggiu M et al 2005 Nova regulates brain-specific splicing to shape the synapse Nat Genet 37 844ndash852

Vessey JP Macchi P Stein JM Mikl M Hawker KN Vogelsang PWieczorek K Vendra G Riefler J Tubing F et al 2008 A lossof function allele for murine Staufen1 leads to impairment ofdendritic Staufen1-RNP delivery and dendritic spine morpho-genesis Proc Natl Acad Sci 105 16374ndash16379

Vierbuchen T Ostermeier A Pang ZP Kokubu Y Sudhof TCWernig M 2010 Direct conversion of fibroblasts to func-tional neurons by defined factors Nature 463 1035ndash1041

Vinuesa CG Cook MC Angelucci C Athanasopoulos V Rui LHill KM Yu D Domaschenz H Whittle B Lambe T et al2005 A RING-type ubiquitin ligase family member requiredto repress follicular helper T cells and autoimmunity Nature435 452ndash458

Wagnon JL Briese M Sun W Mahaffey CL Curk T Rot G Ule JFrankel WN 2012 CELF4 regulates translation and localabundance of a vast set of mRNAs including genes associatedwith regulation of synaptic function PLoS Genet 8 e1003067

Wang ET Cody NA Jog S Biancolella M Wang TT Treacy DJLuo S Schroth GP Housman DE Reddy S et al 2012Transcriptome-wide regulation of pre-mRNA splicing andmRNA localization bymuscleblind proteinsCell 150 710ndash724

Wilbert ML Huelga SC Kapeli K Stark TJ Liang TY Chen SXYan BY Nathanson JL Hutt KR Lovci MT et al 2012LIN28 binds messenger RNAs at GGAGA motifs and regu-lates splicing factor abundance Mol Cell 48 195ndash206

Murn et al

512 GENES amp DEVELOPMENT

Cold Spring Harbor Laboratory Press on March 26 2015 - Published by genesdevcshlporgDownloaded from

101101gad258483115Access the most recent version at doi 2015 29 501-512 Genes Dev

Jernej Murn Kathi Zarnack Yawei J Yang et al finger protein UnkemptControl of a neuronal morphology program by an RNA-binding zinc

Material

Supplemental

httpgenesdevcshlporgcontentsuppl20150303295501DC1html

References

httpgenesdevcshlporgcontent295501fullhtmlref-list-1

This article cites 49 articles 15 of which can be accessed free at

License

Commons Creative

httpcreativecommonsorglicensesby-nc40at Creative Commons License (Attribution-NonCommercial 40 International) as described

) After six months it is available under ahttpgenesdevcshlporgsitemisctermsxhtmlsix months after the full-issue publication date (see This article is distributed exclusively by Cold Spring Harbor Laboratory Press for the first

ServiceEmail Alerting

click hereright corner of the article orReceive free email alerts when new articles cite this article - sign up in the box at the top

httpgenesdevcshlporgsubscriptionsgo to Genes amp Development To subscribe to

copy 2015 Murn et al Published by Cold Spring Harbor Laboratory Press

Cold Spring Harbor Laboratory Press on March 26 2015 - Published by genesdevcshlporgDownloaded from