Government & Nonprofit

A FOXO-Pak1 transcriptional pathway controls neuronal polarity

Description
A FOXO-Pak1 transcriptional pathway controls neuronal polarity
Published
of 16
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
   10.1101/gad.1880510Access the most recent version at doi:  2010 24: 799-813 Genes Dev.  Luis de la Torre-Ubieta, Brice Gaudillière, Yue Yang, et al.  Pak1 transcriptional pathway controls neuronal polarity − A FOXO   MaterialSupplemental   http://genesdev.cshlp.org/content/suppl/2010/04/12/24.8.799.DC1.html References   http://genesdev.cshlp.org/content/24/8/799.full.html#ref-list-1 This article cites 55 articles, 12 of which can be accessed free at: serviceEmail alerting   click here top right corner of the article orReceive free email alerts when new articles cite this article - sign up in the box at the  http://genesdev.cshlp.org/subscriptions  go to: Genes & Development  To subscribe to Copyright © 2010 by Cold Spring Harbor Laboratory Press  Cold Spring Harbor Laboratory Presson April 19, 2010 - Published by genesdev.cshlp.orgDownloaded from  A FOXO–Pak1 transcriptional pathwaycontrols neuronal polarity Luis de la Torre-Ubieta, 1,2 Brice Gaudillie`re, 1 Yue Yang, 1,2,5 Yoshiho Ikeuchi, 1,5 Tomoko Yamada, 1,5 Sara DiBacco, 1 Judith Stegmu ¨ ller, 1,6 Ulrich Schu ¨ ller, 3,7 Dervis A. Salih, 4 David Rowitch, 3,8 Anne Brunet, 4 and Azad Bonni 1,2,9 1 Department of Pathology, Harvard Medical School, Boston, Massachusetts 02115, USA;  2 Program in Neuroscience, HarvardMedical School, Boston, Massachusetts 02115, USA;  3 Department of Pediatric Oncology, Dana-Farber Cancer Institute,Harvard Medical School, Boston, Massachusetts 02115, USA;  4 Department of Genetics, Stanford University, Stanford,California 94305, USA Neuronal polarity is essential for normal brain development and function. However, cell-intrinsic mechanismsthat govern the establishment of neuronal polarity remain to be identified. Here, we report that knockdown ofendogenous FOXO proteins in hippocampal and cerebellar granule neurons, including in the rat cerebellar cortexin vivo, reveals a requirement for the FOXO transcription factors in the establishment of neuronal polarity. TheFOXO transcription factors, including the brain-enriched protein FOXO6, play a critical role in axo–dendriticpolarization of undifferentiated neurites, and hence in a switch from unpolarized to polarized neuronalmorphology. We also identify the gene encoding the protein kinase Pak1, which acts locally in neuronal processesto induce polarity, as a critical direct target gene of the FOXO transcription factors. Knockdown of endogenousPak1 phenocopies the effect of FOXO knockdown on neuronal polarity. Importantly, exogenous expression ofPak1 in the background of FOXO knockdown in both primary neurons and postnatal rat pups in vivo restoresthe polarized morphology of neurons. These findings define the FOXO proteins and Pak1 as components of acell-intrinsic transcriptional pathway that orchestrates neuronal polarity, thus identifying a novel function forthe FOXO transcription factors in a unique aspect of neural development. [ Keywords : FOXO; neuronal polarity; Pak1; transcription; axons; dendrites]Supplemental material is available at http://www.genesdev.org. Received October 30, 2009; revised version accepted February 26, 2010. Axo–dendritic polarity is a fundamental property of neu-rons that is essential for the establishment of proper neu-ronal connectivity, and provides the basis for directionalflow of information in the nervous system (Ramo´ n yCajal 1995; Kandel et al. 2000). Neuronal polarity arisesfrom the specification of undifferentiated neurites intoaxons and dendrites followed by their coordinate growth,leadingtoaneuronalshapetypicallywithalongaxonandseveral shorter dendrites. A major goal in neurobiology isto elucidate the mechanisms that govern the establish-ment of neuronal polarity. Biochemical events that actlocally within neuronal processes leading to neuronalpolarity have been characterized (Craig and Banker 1994;Jan and Jan 2003; Shi et al. 2003; Schwamborn andPuschel 2004; de Anda et al. 2005; Jiang et al. 2005; Kishiet al. 2005; Yoshimura et al. 2005; Barnes et al. 2007;Shelly et al. 2007). Mounting evidence suggests thattranscriptional programs control distinct aspects of thedevelopment of axons or dendrites, including their growthand branching (Jan and Jan 2003; Goldberg 2004; Polleuxet al. 2007). These studies raise the question of whethercell-intrinsic transcriptional mechanisms might also trig-ger the initial specification of neuronal processes intoaxons and dendrites, and the establishment of the uniquepolarized morphology of neurons.Within the mammalian brain, granule neurons of thedeveloping cerebellum provide a robust system for thestudy of axon and dendrite development (Ramo´ n y Cajal1995; Powell et al. 1997). Soon after granule neurons exitmitosis in the external granule layer (EGL) of the de-veloping cerebellum, they begin to extend axons thateventually form the parallel fibers of the cerebellar cortex(Altman and Bayer 1997). Axon growth continues asgranule neurons migrate through the molecular andPurkinje cell layers to reach the internal granule layer(IGL). Once granule neurons take up residence in the IGL, 5 These authors contributed equally to this work.Present addresses:  6 Max-Planck-Institute of Experimental Medicine,Hermann-Rein-Str. 3, 37075 Go ¨ ttingen, Germany;  7 Center for Neuropa-thology and Prion Research, Ludwig-Maximilians-Universita¨t, Feodor-Lynen-St 23, 81377 Munich, Germany;  8 Department of Pediatrics, De-partment of Neurological Surgery, and Howard Hughes Medical Institute,University of California at San Francisco, San Francisco, CA 94143, USA. 9 Corresponding author.E-MAIL azad_bonni@hms.harvard.edu; FAX (617) 432-4101. Supplemental material is available at http://www.genesdev.org.Article is online at http://www.genesdev.org/cgi/doi/10.1101/gad.1880510. GENES  &  DEVELOPMENT 24:799–813    2010 by Cold Spring Harbor Laboratory Press ISSN 0890-9369/10; www.genesdev.org 799  Cold Spring Harbor Laboratory Presson April 19, 2010 - Published by genesdev.cshlp.orgDownloaded from  they elaborate dendrites. Granule neuron axonextension,migration, and dendrite development peak around thesecond to third week postnatally in the rat cerebellarcortex (Altman and Bayer 1997). Differentiated granuleneurons, like other neurons in the brain, are highlypolarized, with long axons and much shorter dendrites(Ramo´ n y Cajal 1995).Studies of neuronal morphogenesis in the cerebellarcortex suggest that distinct transcriptional mechanismsinfluence specific aspects of the development of axons anddendrites (Stegmu¨ ller and Bonni 2005; Kim and Bonni2007). The transcriptional modulator SnoN is requiredfor parallel fiber axon growth (Stegmu¨ ller et al. 2006).Similarly, the development of different phases of granuleneuron dendrite development comes under the purview of specifictranscriptionfactors.NeuroDpromotesthegrowthand maintenance of dendrites, Sp4 promotes dendriticpruning, and MEF2A stimulates synaptic dendritic differ-entiation(Gaudillie`reetal.2004;Shalizietal.2006;Ramosetal.2007).Thesestudiessuggestthatadditionalundefinedtranscriptional mechanisms might regulate other aspectsof neuronal morphogenesis in thecerebellar cortex, includ-ing establishment of the polarized neuronal shape of gran-ule neurons.Besides granule neurons of the cerebellar cortex, hip-pocampal pyramidal neurons have been employed inthe study of neuronal morphogenesis, especially in axo–dendritic polarization. Primary hippocampal neurons be-come polarized in well-defined steps, beginning withthe extension of several undifferentiated neurites thatexpress markers of both axons and dendrites, followingwhich the longest process expresses axon-specific mark-ers, and the remaining neurites differentiate into den-drites (Craig and Banker 1994). Studies of neuronal po-larization in hippocampal neurons have focused on localevents in the neuronal processes (Shi et al. 2003;Schwamborn and Puschel 2004; de Anda et al. 2005; Jianget al. 2005; Kishi et al. 2005; Yoshimura et al. 2005;Jacobson et al. 2006; Arimura and Kaibuchi 2007). How-ever, a role for cell-intrinsic transcriptional mechanismsin axo–dendritic polarization has not been explored.The FOXO transcription factors are widely expressedin the developing mammalian brain (Brunet et al. 1999;Hoekman et al. 2006). While biological functions of theFOXO proteins have been characterized outside the ner-vous system (Burgering and Kops 2002; Tran et al. 2003;Accili and Arden 2004; Arden 2004; Coffer and Burgering2004; Van Der Heideet al. 2004; Barthel et al. 2005; Carterand Brunet 2007), the function of these factors in uniqueaspects of neural development have remained to be iden-tified. Interestingly, expression of the FOXO family mem-ber FOXO6 is enriched in the brain, including the cerebralcortex and hippocampus, but its function has remainedunknown (Jacobs et al. 2003; van der Heide et al. 2005;Hoekman et al. 2006).In this study, we identify a novel role for the FOXOtranscription factors, including the brain-enriched pro-tein FOXO6, in the establishment of neuronal polarityin the mammalian brain. We also identify the polarity-associated protein kinase Pak1 as a critical direct targetgene of the FOXO proteins in neuronal polarity. Collec-tively, our data define the FOXO–Pak1 pathway as a cell-intrinsic transcriptional mechanism that establishes neu-ronal polarity. Results FOXO transcription factors are required for establishment of granule neuron polarity  In situ hybridization analyses have revealed that thetranscription factors FOXO1, FOXO3, and FOXO6 are ex-pressed in the mammalian brain at a time when neuronsundergo a number of developmental events, includingneuronal polarization (Supplemental Fig. 1A,B; Hoekmanet al. 2006). In addition, we found by immunoblotting thatFOXO1, FOXO3, and FOXO6 are expressed in primary ratcerebellar granule neurons (Supplemental Fig. 1C). Theexpression of the FOXO proteins increased with matura-tion in primary granule neurons (Supplemental Fig. 1C).Together, these observations indicate that the FOXO pro-teins are expressed in developing mammalian brain neu-rons, and their temporal pattern of expression suggests apossible role in neuronal morphogenesis.To investigate if the FOXO proteins might contribute toneuronal morphogenesis, we employed a DNA template-based method of RNAi to express shRNAs targeting theFOXO proteins FOXO1, FOXO3, and FOXO6 (Gaudillie`reet al. 2002; Lehtinen et al. 2006; Yuan et al. 2008). Weconfirmed that expression of FOXO shRNAs led to theknockdown of endogenous FOXO1, FOXO3, and FOXO6in neurons (Fig. 1A). The levels of endogenous FOXO1,FOXO3, and FOXO6 mRNAwere reduced within 24–48 hin neurons after transfection with the FOXO RNAi plas-mid (Supplemental Fig. 2).To determine the effect of FOXO knockdown onneuronal morphogenesis, we transfected primary cerebel-lar granule neurons prepared from postnatal day 6 (P6) ratpups with the FOXO RNAi plasmid (U6/foxo) or controlU6 plasmid, together with a GFP expression plasmid tolabel transfected neurons. FOXO RNAi triggered a strik-ing phenotype in primary granule neurons. A significantproportionofFOXOknockdownneuronsdisplayedanon-polarized morphology (Fig. 1B,C; Supplemental Fig. 3).The control U6-transfected neurons had a polarized mor-phology with long Tau1-positive, MAP2-negative axons,and short Tau1-negative, MAP2-positive dendrites (Fig.1D,E). In contrast, the nonpolarized FOXO knockdowngranule neurons had multiple morphologically similarprocesses that were positive for both the axonal markerTau1 and the dendrite marker MAP2 (Fig. 1D,E). Toquantify the loss of polarization in granule neurons uponFOXO knockdown, we measured the ratio of Tau1 orMAP2 signal in the longest process compared with thesecond-longest process, which respectively represent theaxon and a dendrite in control neurons (Kishi et al. 2005).Enrichment of Tau1 was significantly reduced and en-richment of MAP2 signal was significantly increasedin the longest process in granule neurons upon FOXOknockdown (Fig. 1F). We subjected control and FOXO de la Torre-Ubieta et al.800 GENES  &  DEVELOPMENT  Cold Spring Harbor Laboratory Presson April 19, 2010 - Published by genesdev.cshlp.orgDownloaded from  knockdown granule neurons to morphometric analyses.In control granule neurons, as in most neurons in thebrain, the longest process is the axon while the other,shorter processes develop into dendrites. FOXO knock-down neurons exhibited significantly longer secondaryprocesses (dendrites in control), while the longest process(axon in control) was significantly shorter as comparedwith control U6-transfected neurons (Fig. 1G). In controlexperiments, FOXO knockdown did not affect theimmunoreactivity of markers of post-mitotic granule Figure 1.  FOXO transcription factors estab-lish neuronal polarity in cerebellar granuleneurons. (  A ) Granule neurons were electro-porated before plating using the Amaxanucleofection kit with the control U6 orU6/foxo RNAi plasmid. Four days aftertransfection, lysates were subjected to im-munoblotting with a FOXO1, FOXO3, orFOXO6 antibody. FOXO RNAi substantiallyreduced levels of endogenous FOXO1,FOXO3, and FOXO6 in neurons. The aster-isk indicates nonspecific band. ( B ) Cerebellargranule neurons transfected with the controlU6 or U6/foxo RNAi plasmid and a GFPexpression plasmid were subjected 4 d aftertransfection to immunocytochemistry withan antibody to GFP (see Supplemental Fig. 3for additional lower-magnification panels).Arrows, arrowheads, and asterisks indicatedendrites, axons, and cell body, respectively.Bar, 50  m m. ( C ) Granule neurons transfectedand analyzed asin  B  werescoredas polarizedor nonpolarized. FOXO knockdown signifi-cantly increased the number of neurons thatfail to acquire a polarized morphology ( P  < 0.01;  t -test,  n  =  3). ( D–F  ) Granule neuronswere transfected with the Amaxa electro-poration device with the control U6 or U6/foxo RNAi plasmid and the GFP expressionplasmid and grown at low density. Five daysafter transfection, neurons were subjectedtoimmunocytochemistrywiththeGFPanti-bodyandanantibodytothedendriticmarkerMAP2 ( D ) or the axonal marker Tau1 ( E ). En-richment of Tau1 and MAP2 was quanti-fied in  F  . Tau1 and MAP2 enrichment aredefined as the intensity of Tau1 or MAP2immunostaining in the longest neurite di-vided by the intensity in the second-longestneurite. FOXO knockdown neurons dis-played significantly increased MAP2 en-richment ( P  <  0.001;  t -test,  n  =  3) andsignificantly reduced Tau1 enrichment ( P  < 0.01;  t -test,  n  =  3) when compared withcontrol U6-transfected neurons. Arrowheadsand arrows point to the longest process andother processes, respectively. Asterisks in-dicate cell bodies. ( G ) Morphometric analysis of granule neurons transfected as in  B  revealed that FOXO RNAi significantly reduced thelength of the longest process (axon in control), and concomitantly increased the length of secondary processes (dendrites in control) ( P  < 0.001;  t -test, 213 neurons measured). ( H  ) Lysates of 293Tcells transfected with the control U6 or U6/foxo RNAi plasmid together with anexpression vector encoding GFP-tagged FOXO6 (FOXO6-WT) or the RNAi-resistant mutant FOXO6 (FOXO6-Res) were subjected toimmunoblotting with the GFP antibody ( top  panel) or an antibody to ERK1/2 ( bottom  panel). ( I–K  ) Granule neurons transfected with thecontrol U6 or U6/foxo RNAi plasmid, together with the FOXO6-Res expression plasmid or its control vector and an expression plasmidencoding DsRed, were subjected 4 d after transfection to immunocytochemistry with an antibody to DsRed. FOXO6-Res significantlyreduced the percentage of nonpolarized neurons in the background of FOXO RNAi ( P  <  0.01; ANOVA,  n  =  3). The length of the longestprocess (axon in control) was significantly reduced and the length of secondary processes (dendrites in control) was significantly increasedupon FOXO RNAi ( P  <  0.001; ANOVA, 200 neurons measured), but not in FOXO6-Res-expressing neurons in the background of FOXOknockdown, when compared with control U6-transfected neurons. Arrows, arrowheads, and asterisks indicate dendrites, axons, and cellbody, respectively. Bar, 50  m m. FOXO–Pak1 transcriptional control polarityGENES  &  DEVELOPMENT 801  Cold Spring Harbor Laboratory Presson April 19, 2010 - Published by genesdev.cshlp.orgDownloaded from  neurons, including the neuron-specific class III  b -tubulin(Tuj1) and the transcription factor MEF2A (SupplementalFig. 4). In addition, FOXO knockdown did not haveadverse effects on cell survival, and thus the impairmentof neuronal polarity in FOXO knockdown neurons wasnot associated with reduced cell survival (SupplementalFig. 5). Together, these results suggest that knockdown of the FOXO proteins impairs axo–dendritic polarization ingranule neurons.Todeterminethe specificityof theFOXORNAi-inducedneuronal polarity phenotype, we performed a rescue exper-iment. We generated expression plasmids encoding rescueforms of FOXO1, FOXO3, and FOXO6 by introducingsilentmutationsinthecDNAencodingtheFOXOproteinsdesigned to render them resistant to FOXO RNAi (FOXO-Res). We confirmed that expression of FOXO shRNAsfailed to effectively induce knockdown of FOXO1-Res(Yuan et al. 2008), FOXO3-Res (Lehtinen et al. 2006), andFOXO6-Res (Fig. 1H). We next tested if expression of therescue forms of FOXO proteins suppresses the FOXORNAi-induced phenotype in granule neurons. Expressionof FOXO1-Res or FOXO3-Res significantly, albeit partially,reversed the FOXO RNAi-induced phenotype in granuleneurons (Supplemental Fig. 6A,B). Expression of FOXO6-Res restored the polarized morphology of granule neuronsin the background of FOXO RNAi (Fig. 1I,J). Expression of the FOXO rescue proteins on their own had little or noeffect on polarity in granule neurons (Supplemental Fig.6C). FOXO6-Res also reversed the dual effect of FOXORNAionthegrowthofthelongestandsecondaryprocessesin granule neurons (Fig. 1K). Together, these results in-dicate that the FOXO RNAi-induced phenotype is theresult of specific knockdown of FOXO proteins, ratherthan off-target effects of RNAi or nonspecific activation of the RNAi machinery. Our results also suggest that, amongthe FOXO proteins, FOXO6 is the prominent though notexclusive member that promotes neuronal polarity.To further characterize the relative roles of the FOXOproteins in the establishment of neuronal polarity, wegenerated U6/foxo1, U6/foxo3, and U6/foxo6 RNAi plas-mids encoding shRNAs targeting each of the three FOXOproteins specifically (Supplemental Fig. 7A,B). In contrastto FOXO RNAi inducing the knockdown of FOXO1,FOXO3, and FOXO6 (see Fig. 1), knockdown of each of the three FOXO proteins alone failed to impair polarity ingranule neurons (Supplemental Fig. 7C), suggesting thatFOXO1, FOXO3, and FOXO6 have redundant functionsin the establishment of neuronal polarity. Accordingly,the combined expression of FOXO1, FOXO3, and FOXO6shRNAs impaired polarity in granule neurons, thus phe-nocopying the effect of FOXO shRNAs (SupplementalFig. 7D,E). Collectively, our data suggest that FOXO6collaborates with FOXO1 and FOXO3 to induce neuronalpolarity and promote the dual morphogenesis of axonsand dendrites.We next characterized the temporal dynamics of theFOXO RNAi-induced polarity phenotype. In analysesof cohorts of granule neurons, we found that the majorityof control P6 neurons acquire a polarized morphologybetween the first and second day after plating (Fig. 2A).Remarkably, FOXO RNAi-transfected granule neuronsdid not convert to a polarized morphology, and remainedin a nonpolarized state throughout the course of theanalysis.To determine whether FOXO proteins control thetransitionfromanonpolarizedtoapolarizedmorphology,we performed time-lapse analyses of individual controland FOXO knockdown granule neurons. Neurons wereclassified into five distinct stages (Powell et al. 1997).Stages 1–2 and stages 3–5 represent nonpolarized andpolarized neurons, respectively. At the time of initialobservation, both control and FOXO knockdown granuleneurons were found in both polarized and nonpolarizedmorphologies (Fig. 2B,C). During the ensuing 86 h of observation, control granule neurons that were initiallypolarized remained polarized, and neurons that wereinitially nonpolarized converted to a polarized morphol-ogy (Fig. 2C). In contrast, FOXO knockdown neurons thatwere initially nonpolarized did not convert to a polarizedmorphology throughout the 86 h of observation (Fig.2B,C). Interestingly, FOXO knockdown neurons thatwere initially polarized remained polarized (Fig. 2C).Quantification of these analyses revealed that FOXOknockdown blocked polarization in nearly 70% of neu-rons that were initially nonpolarized, while only 15% of control granule neurons that were initially nonpolarizedremained nonpolarized at the last time point of observa-tion (Fig. 2D). However, none of the FOXO knockdown orcontrol granule neurons that were initially polarizedbecamenonpolarized atthelast time point ofobservation(Fig. 2C). Together, these results suggest that the FOXOproteins trigger a switch from nonpolarized to polarizedmorphology in neurons. FOXO transcription factors orchestrate axo–dendritic polarization in hippocampal neurons We next asked if the function of the FOXO transcriptionfactors in the establishment of neuronal polarity isspecific to cerebellar granule neurons, or if the FOXOproteins play a generalized role in neuronal polarity inmammalian neurons. We therefore characterized the roleof FOXO transcription factors in primary hippocampalneurons, an established system in the study of neuronalpolarization (Craig and Banker 1994). Induction of FOXORNAiinhippocampal neuronssignificantly increasedthenumber of nonpolarized neurons, leading to a threefoldincrease in the percentage of nonpolarized neurons ascompared with control U6-transfected neurons (Fig. 3A).ThelargemajorityofcontrolU6-transfectedhippocampalneurons had a polarized morphology. These neuronsdisplayed a long Tau1-positive, MAP2-negative axon,and multiple short Tau1-negative, MAP2-positive den-drites (Fig. 3B,C). The nonpolarized FOXO knockdownhippocampal neurons had multiple morphologically sim-ilar processes that were both Tau1- and MAP2-positive(Fig. 3B,C). Quantification of the ratio of Tau1 or MAP2signal in the longest process compared with the second-longest process—which represent the axon and the den-drite in control neurons, respectively (Kishi et al. de la Torre-Ubieta et al.802 GENES  &  DEVELOPMENT  Cold Spring Harbor Laboratory Presson April 19, 2010 - Published by genesdev.cshlp.orgDownloaded from
Search
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks