Making synaptic plasticity and memory last: mechanisms of

REVIEW

Making synaptic plasticity and memorylast: mechanisms of translationalregulation

Joel D. Richter1,3 and Eric Klann2,4

1Program in Molecular Medicine University of Massachusetts Medical School Worcester, Massachusetts 01605, USA;2Center for Neural Science New York University New York, New York 10003, USA

Synaptic transmission in neurons is a measure of commu-nication at synapses, the points of contact between axonsand dendrites. The magnitude of synaptic transmission isa reflection of the strength of these synaptic connections,which in turn can be altered by the frequency with whichthe synapses are stimulated, the arrival of stimuli fromother neurons in the appropriate temporal window, and byneurotrophic factors and neuromodulators. The ability ofsynapses to undergo lasting biochemical and morpholog-ical changes in response to these types of stimuli andneuromodulators is known as synaptic plasticity, whichlikely forms the cellular basis for learning and memory,although the relationship between any one form synapticplasticity and a particular type of memory is unclear. RNAmetabolism, particularly translational control at or nearthe synapse, is one process that controls long-lastingsynaptic plasticity and, by extension, several types ofmemory formation and consolidation. Here, we reviewrecent studies that reflect the importance and challengesof investigating the role of mRNA translation in synapticplasticity and memory formation.

Changes in gene expression are required to convert short-term memory (STM), lasting less than ;1 h, to long-termmemory (LTM) in both invertebrates and vertebrates(Kandel 2001). At the cellular level, long-lasting changesin synaptic strength, typically called synaptic plasticity,refers to the ability of neurons to alter communicationwith each other via synaptic connections in response tospecific patterns of electrical stimulation and/or neuro-trophic factors, and is generally considered to underlieLTM (Malenka and Nicoll 1999). The most studied formsof long-lasting synaptic plasticity in mammals, particu-larly rodents, are long-term potentiation (LTP) and long-term depression (LTD), which refer to long-lastingincreases or decreases, respectively, in synaptic strength(Malenka and Bear 2004). Most of the work on LTP and LTDhas been conducted in the hippocampus, a structure re-

quired for memory consolidation. Similar to memory, LTPcan be defined temporally with respect to the requirementfor new gene expression: Early-phase LTP (E-LTP), likeSTM, does not require new gene expression, whereas late-phase LTP (L-LTP) does. Throughout the 1990s, molecularstudies of the regulation of gene expression in the contextof LTM formation and L-LTP focused almost exclusively ontranscription, especially the transcription factor CREB(Silva et al. 1998). However, in the last five years, therehave been several studies delineating the mechanisms oftranslational control underlying both LTM and L-LTP.

Pharmacological inhibitors of transcription and trans-lation block L-LTP induced by electrical stimulation(Klann and Dever 2004; Klann et al. 2004). Ribosomes,translation factors, and mRNA are present not only in theneuronal soma, but also in dendrites and dendritic spines(Steward and Schuman 2001), suggesting that local (synapto-dendritic) protein synthesis could trigger long-lasting syn-aptic plasticity without engaging transcription in theneuronal soma. Indeed, brain-derived neurotrophic factor(BDNF)-induced LTP, metabotropic glutamate receptor-dependent LTD (mGluR-LTD) and LTP induced by de-livery of E-LTP-inducing stimulation in the presence of anagonist of b-adrenergic receptors (bAR-LTP) are all long-lasting forms of plasticity that are blocked by proteinsynthesis inhibitors even when the neurons are physicallysevered from their cell bodies (Kang and Schuman 1996;Huber et al. 2000; Gelinas and Nguyen 2005). Moreover,the inhibition of translation initiation results in the abro-gation of L-LTP earlier than when transcription is inhibited(Kelleher et al. 2004; Banko et al. 2005) and L-LTP can beimpaired with the direct application of a protein synthesisinhibitor to dendrites (Bradshaw et al. 2003). Thus, localprotein synthesis is a critical component of several forms oflong-lasting hippocampal synaptic plasticity.

Mammalian target of rapamycin (mTOR), eIF4E-bindingprotein (4E-BP), and p70 S6 kinase (S6K) are involvedin long-lasting hippocampal synaptic plasticityand memory

mTOR is a protein kinase whose activation serves as oneof the primary triggers for the initiation of cap-dependent

[Keywords: Translation; neuron; synaptic plasticity; memoryCorrespondence.3E-MAIL [email protected]; FAX (508) 856-4289.4E-MAIL [email protected]; FAX. (713) 798-3475.Article is online at http://www.genesdev.org/cgi/doi/10.1101/gad.1735809.

GENES & DEVELOPMENT 23:1–11 � 2009 by Cold Spring Harbor Laboratory Press ISSN 0890-9369/09; www.genesdev.org 1

Cold Spring Harbor Laboratory Press on November 17, 2021 - Published by genesdev.cshlp.orgDownloaded from

http://genesdev.cshlp.org/

http://www.cshlpress.com

translation via phosphorylation of 4E-BPs and S6Ks(Richter and Sonenberg 2005). mTOR interacts the adap-tor protein Raptor, which binds both 4E-BP and S6K (Choiet al. 2003; Schalm et al. 2003); mTOR/Raptor is referredto as mTOR complex 1 (mTORC1). Rapamycin, a drugthat binds the protein FKBP12 and prevents mTOR frombinding Raptor, disrupts mTORC1 and inhibits mTOR-catalyzed 4E-BP and S6K phosphorylation (Kim et al.2002). The binding and inhibition of eIF4E by 4E-BP isregulated by mTOR-dependent phosphorylation (Gingraset al. 2001). Unphosphorylated 4E-BP binds tightly toeIF4E, whereas 4E-BP phosphorylated by mTOR does not,thereby permitting eIF4F to form and initiation to pro-ceed (Gingras et al. 2001). mTOR also impacts translationby phosphorylating S6K, which then phosphorylatesdownstream targets such as ribosomal protein S6 andeIF4B (Raught et al. 2004). Figure 1A describes some ofthe signaling events noted above.

In most systems, the phosphatidylinositol-3 kinase(PI3K) signaling pathway is upstream of mTOR. PI3Kphosphorylates the membrane phospholipid phosphati-dylinositol-4,5-bisphosphate (PIP2) converting it to PIP3,which then recruits Akt to the membrane where it isphosphorylated and activated by PDK1 (Brazil and Hem-mings 2001), as well as mTORC2, a complex of mTORbound to a second adaptor protein, Rictor (Sabatini 2006).Akt activates mTOR by inhibiting the tuberous sclerosiscomplex (TSC), a heterodimer of TSC1 (hamartin) andTSC2 (tuberin). TSC2 contains a GAP (GTPase-activatingprotein) domain for the small G-protein Rheb, and also is

a substrate for Akt. When TSC2 is phosphorylated, itsGAP activity decreases, resulting in Rheb and subsequentmTOR activation (Garami et al. 2003; Inoki et al. 2005). Adiagram depicting this signaling cascade is shown inFigure 1B. GTP-Rheb activates mTOR by antagonizingFKBP38, which binds mTOR and serves as an endogenousinhibitor of mTORC1 (Bai et al. 2007). FKBP38 is struc-turally similar to FKBP12, however, the expression ofFKBP38 is minimal in the adult rodent brain, whereasFKBP12 is very abundant (Hoeffer et al. 2008). Althoughthere is no direct evidence yet that FKBP12 binds eithermTOR or Rheb, genetic deletion of FKBP12 increases thelevel of mTORC1 (Hoeffer et al. 2008), suggesting thatFKBP12, like FKBP38, is an inhibitor of mTORC1 (Fig. 2).

Protein synthesis-dependent LTP and mGluR-LTD trig-ger activation of signaling pathways that enhance trans-lation factor phosphorylation in hippocampal slices. LTP-inducing stimulation also triggers activation of mTOR,resulting in enhanced phosphorylation of 4E-BP2, thepredominant 4E-BP isoform in the mouse brain, andenhanced eIF4F complex formation (Kelleher et al.2004; Banko et al. 2005). Moreover, LTP associated withincreased S6K phosphorylation requires both mTOR- andextracellular signal-regulated kinase (ERK) (Tsokas et al.2005, 2007; Hoeffer et al. 2008), and is correlated withincreased phosphorylation of ribosomal protein S6, a sub-strate of S6K (Kelleher et al. 2004; Tsokas et al. 2007;Antion et al. 2008b). Importantly, many of the LTP-induced changes in translation factor phosphorylationoccur in dendrites (Tsokas et al. 2005). Similar increases

Figure 1. Signaling required for translational control during long-lasting synaptic plasticity and memory. (A) Signaling that couplesgroup I mGluRs to translation initiation during mGluR-LTD. Activation of mGluR1 and mGluR5 triggers the activation of a PI3K/Akt/mTOR/4E-BP signaling cascade that is required for the formation of eIF4F and the expression of mGluR-LTD (Banko et al. 2006).mGluR-LTD also triggers the activation of a MEK/ERK/Mnk1 signaling cascades that are parallel to the mTOR cascade, with thesignaling pathways converging on eIF4E (Gallagher et al. 2004; Banko et al. 2006). (B) In most systems, phosphorylation of 4E-BP andS6K1 by mTORC1 requires Akt-dependent phosphorylation of TSC2, which inhibits its GAP activity and permits the activation ofRheb and mTOR that is bound to Raptor (mTORC1). mTORC1 is sensitive to inhibition by rapamycin bound to FKBP12. mTOR alsocan be bound to Rictor (mTORC2), which phosphorylates and activates Akt. Although there is substantial evidence consistent for a roleof mTORC1 in long-lasting plasticity and memory, very little is known about the role of mTORC2 in these processes.

Richter and Klann

2 GENES & DEVELOPMENT




in mTOR phosphorylation and its downstream effectorsoccur with mGluR-LTD (Hou and Klann 2004; Bankoet al. 2006; Antion et al. 2008a; Ronesi and Huber 2008)and bAR-LTP (Gelinas et al. 2007). A diagram depictingsignaling events known to couple group I mGluRs tomTOR and its substrate 4E-BP is shown in Figure 1A.Both pharmacological and genetic manipulations oftranslation factors in mice have demonstrated that propertranslational control via mTOR is required for normalexpression of LTP (Banko et al. 2005; Tsokas et al. 2005;Antion et al. 2008b; Hoeffer et al. 2008), mGluR-LTD(Hou and Klann 2004; Banko et al. 2006; Antion et al.2008a; Ronesi and Huber, 2008), and bAR-LTP (Gelinaset al. 2007). Thus, dendritic up-regulation of translationinitiation via mTORC1 is a critical component of long-lasting, translation-dependent synaptic plasticity.

Several pharmacological studies have shown that rapa-mycin inhibition of mTORC1 signaling also blocks LTMformation in mammals (Tischmeyer et al. 2003; Parsonset al. 2006; Blundell et al. 2008; Schicknick et al. 2008). Inaddition, the training for several memory tasks coincideswith increase in the phosphorylation of the downstreammTORC1 effectors 4E-BP and S6K in the hippocampusand amygdala, a brain structure involved in fear memory(Kelleher et al. 2004; Parsons et al. 2006; Hoeffer et al.2008). Finally, behavioral studies with mice carryingmutations in 4E-BP2, S6K1, and S6K2, indicate thatmTOR signaling is required for normal LTM. 4E-BP2knockout mice display impaired hippocampus-dependentand -independent forms of memory (Banko et al. 2005,2007). Although S6K1- and S6K2-deficient mice expressL-LTP, S6K1-deficient exhibit impaired E-LTP and aredeficient in multiple forms of learning and memory(Antion et al. 2008b). S6K2-deficient mice also displayseveral memory phenotypes (Antion et al. 2008b). Thus,mice with genetic deletions of translational controlmolecules downstream from mTORC1 display a numberof learning and memory impairments.

Genetic deletion of upstream molecules that result inup-regulation of mTORC1 signaling also have an impacton memory. Mice with a conditional, postnatal deletion

of FKBP12, which results in increased levels of mTORC1and increased phosphorylation of S6K1, display enhancedcontextual fear memory as well as several examples ofperservation and repetitive behaviors that are consistentwith autistic and obsessive-compulsive behavior (Hoefferet al. 2008). In addition, training for contextual fearmemory triggers increased levels of mTORC1 and in-creased phosphorylation of S6K1 in wild-type mice (Fig.2). Both TSC1 and TSC2 heterozygous knockout miceexhibit impaired hippocampus-dependent memory(Goorden et al. 2007; Ehninger et al. 2008) and thememory impairments displayed by the TSC2 mutantmice are rescued by rapamycin treatment (Ehningeret al. 2008). These results are intriguing as mutations ineither TSC1 or TSC2 cause TSC, which often results inmental retardation and autism (DiMario 2004; Wiznitzer2004). Taken together these findings indicate that propermTOR regulation is required for normal LTM.

GCN2 and eIF2a control long-lasting hippocampalsynaptic plasticity and memory

Another translation factor involved in long-lasting syn-aptic plasticity and LTM is the eukaryotic initiationfactor eIF2, which consists of a, b, and g subunits. eIF2binds initiator Met–tRNAi

Met and GTP to form a ternarycomplex, which then interacts with the small ribosomalsubunit in its GTP-bound form. eIF2 is released from theribosome in its GDP-bound state upon GTP hydrolysis.eIF2B catalyzes the exchange of GDP for GTP, which isrequired to reconstitute a functional ternary complex fora new round of initiation (Hinnebusch 2000). Phosphor-ylation of eIF2a S51 blocks the GDP/GTP exchange,thereby slowing the dissociation of eIF2 from eIF2B andcausing a decrease in general initiation. It is important torealize that although eIF2a phosphorylation causes a gen-eral inhibition of translation, it also up-regulates theexpression of some mRNAs with upstream ORFs (uORFs)(Sonenberg and Dever 2003), one of which encodes thetranscriptional modulator ATF4 (Harding et al. 2000; seebelow).

Figure 2. Proposed mechanism formTORC1 activation following stimula-tion that induces L-LTP and trainingthat triggers LTM. FKBP12 serves as anintracellular scaffold for either an mTORinhibitory factor(s) or directly competeswith Raptor for the mTOR FRB site. Ineither scenario, FKBP12 represses mTORactivity by blocking mTORC1 (mTOR–Raptor interactions). Stimulation thatinduces translation initiation (either elec-trical stimulation that induces L-LTP ortraining that induces LTM), signals either(1) the displacement of the inhibitory fac-

tor(s) interacting with FKBP12, or (2) the sequestration of FKBP12 from mTOR, thereby permitting Raptor access to the FRB.Translation-inducing signaling promotes mTOR access to 4E-BP2 and S6K, possibly through activation of additional mTORC1-associated scaffolds (i.e., PRAS40), allowing translational initiation. Ras-homolog enriched in the brain (Rheb), proline-rich Akt/PKBsubstrate-40 kd (PRAS-40), FKBP12-binding domain (FRB), KIN (mTOR kinase catalytic domain), NRD (domain, site of serine 2448phosphorylation).

Translation and synaptic plasticity

GENES & DEVELOPMENT 3




Evidence that regulation of eIF2a phosphorylationplays an important role in long-lasting synaptic plasticityand LTM was provided by studies with GCN2 mutantmice (GCN2 is one of four kinases that can phosphorylateeIF2a), which have a decreased threshold for the induc-tion of L-LTP (Costa-Mattioli et al. 2005). This LTPphenotype also has been observed in mice with a geneticreduction of other translational repressors such as 4E-BP2(Banko et al. 2005) and TSC2 (Ehninger et al. 2008).Heterozygous eIF2a knock-in mice that harbor S51A alsohave a decreased threshold for the induction of L-LTP(Costa-Mattioli et al. 2007). Moreover, L-LTP is associ-ated with decreased eIF2a phosphorylation, and preven-tion of this dephosphorylation with Sal003, an inhibitorof eIF2a phosphatases, results in the blockade of L-LTP(Costa-Mattioli et al. 2007). ATF4, which represses syn-aptic plasticity and memory formation (Abel et al. 1998;Chen et al. 2003), is required for this impairment as L-LTPin hippocampal slices from ATF4 knockout mice cannotbe blocked by Sal0003 (Costa-Mattioli et al. 2007). Takentogether, these findings indicate that the phosphorylationof eIF2a is required for the expression of L-LTP.

The phosphorylation of eIF2a also plays an importantrole in LTM. Animal training regimens that induce LTMis associated with decreased phosphorylation of eIF2a inthe hippocampus and conversely, preventing training-induced dephosphorylation of eIF2a with Sal003 preventsLTM formation (Costa-Mattioli et al. 2007). Behavioralstudies with both GCN2 mutant mice and heterozygouseIF2a S51A knock-in mice indicate that reduction ofeIF2a phosphorylation lowers the threshold for LTMformation (Costa-Mattioli et al. 2005, 2007). Interest-ingly, with more arduous training regimens, the GCN2mutant mice have impaired LTM (Costa-Mattioli et al.2007), which also has been observed in 4E-BP2 (Bankoet al. 2005) and TSC2 (Ehninger et al. 2008) mutant mice.Thus, although reduction of eIF2a phosphorylation,which would enhance general translation, reduces thethreshold for LTM, deletion of GCN2 and other muta-tions that derepress translation initiation can be delete-rious to LTM formation.

FMRP and long-lasting hippocampal synaptic plasticity

As mentioned earlier, mGluR-LTD is a dendritic proteinsynthesis-dependent form of synaptic plasticity that canbe induced by the selective group I mGluR agonist (RS)-3,5-dihydroxyphenylglycine (DHPG) (Huber et al. 2000;Hou and Klann 2004). The product of the Fragile X gene,FMRP, almost certainly regulates translation duringmGluR-LTD. For example, it has been reported thatmGluR-LTD is augmented in Fmr1 knockout mice(Huber et al. 2002; Hou et al. 2006). In addition, severalstudies indicate that FMRP is translated in response tostimulation of group I mGluRs in synaptosomes (an invitro preparation of synapses), cultured cortical andhippocampal neurons, hippocampal slices, and in thebrain in vivo (Todd et al. 2003; Antar et al. 2004; Houet al. 2006). Based on these results, Bear et al. (2004)

suggested the ‘‘mGluR theory of Fragile X mental re-tardation,’’ which posited that excessive mGluR-depen-dent protein synthesis leads to multiple phenotypes inFragile X syndrome (FXS). Subsequent studies showedthat DHPG-induced mGluR-LTD induced in hippocam-pal slices resulted in the rapid translation of FMRP, whichwas dependent on the group I mGluR subtype mGluR5.Surprisingly, the rapid increase in FMRP levels associatedwith mGluR-LTD was followed by the ubiquitination andrapid destruction of FMRP; conversely, inhibition of theubiquitin-proteasome pathway abrogated mGluR-LTD,as did the overexpression of FMRP (Hou et al. 2006).mGluR-LTD in wild-type mice is associated with rapidincreases in proteins whose mRNAs are bound by FMRP;such increases are abolished in Fmr1 knockout mice. Incontrast to mGluR-LTD in wild-type mice, both proteinsynthesis (Hou et al. 2006) and proteasome inhibitors(Hou et al. 2006) have no effect on mGluR-LTD in Fmr1knockout mice. These findings suggest that there isexcessive translation of normally FMRP-bound mRNAsin Fmr1 knockout mice and that these mRNAs are trans-lated during mGluR-LTD in wild-type mice. Moreover,these results also indicate that rather than an additionallevel of excessive translation, mGluR-dependent trans-lational control is absent in Fmr1 knockout mice.

How would excessive mGluR-dependent translationoccur in Fmr1 knockout mice? One possibility is thatexcessive activation of mGluRs occurs in the Fmr1knockout mice, which triggers exaggerated activation oftranslational control pathways. In an elegant series ofstudies, Dolen et al. (2007) demonstrated that Fmr1knockout mice with a 50% reduction in mGluR5 levelsdo not exhibit several FXS phenotypes. In addition,treatment of Fmr1 knockout mice with the mGluR5antagonist MPEP also reverses FXS phenotypes (Yanet al. 2005); a similar rescue of phenotypes with mGluRantagonists has been observed in a Drosophila model ofFXS (McBride et al. 2005). Moreover, deletion of FMRP inDrosophila results in memory impairments that can berescued by protein synthesis inhibitors, consistent withthe notion that excessive translation impacts cognition inFXS (Bolduc et al. 2008). Although an extensive charac-terization of translational control pathways has yet to beaccomplished, phosphorylation of PDK-1, mTOR, andS6K1 can no longer be stimulated by DHPG in Fmr1knockout mice (Ronesi and Huber 2008). Similar resultshave been observed for extracellular signal-regulatedkinase (Hou et al. 2006), which also is required for trans-lational control in long-lasting hippocampal synaptic plas-ticity and memory (Gallagher et al. 2004; Kelleher et al.2004; Banko et al. 2006). Thus, excessive basal translationand a lack of mGluR-dependent translational control arefeatures that likely contribute to plasticity and behavioralphenotypes displayed by Fmr1 knockout mice.

FMRP binds many mRNAs, and increased expressionof several of the encoded proteins has been observed ineither the brains or neurons from Fmr1 knockout mice,including Arc/Arg3.1, aCaMKII, PSD-95, SAPAP3, andMAP1B (Zalfa et al. 2003; Todd et al. 2003; Hou et al.2006; Muddashetty et al. 2007; Narayanan et al. 2008).

Richter and Klann





Consistent with studies under conditions where FMRP isreduced (Park et al. 2008), mGluR-LTD in hippocampalslices is associated with protein synthesis-dependentincreases in the levels of FMRP, MAP1B, aCaMKII, andArc/Arg3.1 (Hou et al. 2006; Waung et al. 2008). Thefunctional consequences of increased aCaMKII inmGluR-LTD are not clear, whereas MAP1B and Arc/Arg3.1 synthesis is required for mGluR-dependent endo-cytosis of AMPA receptors (Davidkova and Carroll 2007;Waung et al. 2008; Park et al. 2008). Dynamic translationof Arc/Arg3.1 also is required for the expression ofmGluR-LTD (Waung et al. 2008). Taken together, thesestudies indicate that translation of FMRP-bound mRNAscontributes to mGluR-LTD and suggest that excessivebasal translation of these mRNAs might contribute to theplasticity and behavioral phenotypes observed in FXS.

Translational control by FMRP

The molecular mechanism by which FMRP modulatestranslation has been intensively studied but remainscontroversial. There is a general consensus that FMRPinhibits translation, although there is evidence consistentwith it being a translational activator. For example,several investigators find that a substantial amount ofFMRP sediments with polysomes (Stefani et al. 2004;Darnell et al. 2005), which would be expected of anactivator of translation. On the other hand, a protein thatslows ribosome transit along an mRNA would also beexpected to sediment with polysomes, and indeed FMRPhas been suggested to do just that (Ceman et al. 2003). Inaddition, metabolic labeling of protein in the hippocam-pus of Fmr1 knockout mice exceeds that of wild-typemice (Dolen et al. 2007), again suggesting that FMRP isan inhibitor of translation. However, lessons from devel-oping systems suggest that firm conclusions from eventhese seemingly straightforward data must be viewedwith caution. For example, the completion of the finalstages of oocyte meiosis requires a translational regula-tory cascade where the very early translational activationof one (or a few) mRNA(s) induces downstream trans-lational activation of some mRNAs but translationalrepression of other mRNAs (Richter 1996). It is thereforepossible that FMRP activates the translation of somemRNAs at early times of, say, development or followingsynaptic stimulation, which causes subsequent mRNA-specific translational activation and repression events.That FMRP might repress and activate mRNAs is sug-gested by the findings of Brown et al. (2001), who observedthat of several mRNAs that were coimmunoprecipitatedwith FMRP, some sedimented to heavy polysomes, whileothers shifted to light polysomes in cells lacking FMRP.Thus, FMRP could either be bifunctional—i.e., repressingsome mRNAs while activating others—or affect allmRNAs the same way (say, repression), which is followedby a cascade of translational control that is both repres-sing and activating (see also Bagni and Greenough 2005).

A recent intriguing model has been proposed for howFMRP regulates translation that has broad implicationsfor translational control in general. Napoli et al. (2008)

recently reported that a substantial portion of FMRPsediments in fractions lighter than polysomes; suchfractions also contain CYFIP1 (cytoplasmic FMRP inter-acting protein), a factor that binds FMRP, as well as eIF4E.Surprisingly, CYFIP1 and FMRP are both retained onm7GTP (cap)-Sepharose columns, indicating that theydirectly or indirectly bind the cap. Because both proteinsare competed off the column by excess 4E-BP, Napoliet al. (2008) surmised that FMRP and CYFIP bind the capthrough an interaction with eIF4E; they further showedthat it is CYFIP1 that directly binds eIF4E. CYFIP1 con-tains a region with some similarity to 4E-BPs, but surpris-ingly, does not conform to the YXXXXLF (where F is anyhydrophobic amino acid, often a leucine) sequence that iscommon among such proteins (Richter and Sonenberg2005). Instead, CYFIP1 has a ‘‘noncanonical’’ sequencethat is predicted to form two a helices that are nearlyidentical in structure to those formed by the consensuseIF4E-binding peptide (Marcotrigiano et al. 1999; Napoliet al. 2008). The CYFIP1 a helices are stabilized bypredicted internal salt bridges and indeed the residuesthat are thought to form these bridges are necessary forCYFIP1’s interaction with eIF4E (Napoli et al. 2008).eIF4E–CYFIP1–FMRP complexes can be detected in syn-aptoneurosome preparations and, upon synaptic stimula-tion, the CYFIP1–eIF4E interaction is destroyed andFMRP-bound mRNAs undergo enhanced translation.

These results of Napoli et al. (2008) indicate that atleast one mode of FMRP-inhibited translation is analo-gous to that of CPEB. That is, an RNA-binding protein(FMRP or CPEB) is bound to an eIF4E-associated factor(CYFIP1 or Maskin) to preclude the recruitment of eIF4G,and indirectly the 40S ribosomal subunit, to the 59 end ofthe mRNA (Richter 2007). One may also infer thatmolecules with ‘‘Maskin-like’’ activities—i.e., mRNA-specific 4E-BPs—may be more widespread than thoughtpreviously. For example, Drosophila Cup (Nakamuraet al. 2004), mammalian 4E-T (Rong et al. 2008), andmammalian neuroguidin (Jung et al. 2006) all contain theYXXXXLF motif noted above and thus resemble Maskin(although Maskin has a threonine in place of the tyro-sine); CYFIP1, however, may the first among other soon-to-be-discovered molecules with ‘‘noncanonical’’ eIF4E-binding regions that could regulate translation of manysets of mRNAs by associating with different RNA-binding proteins (Fig. 3).

In addition to the mechanism by which FMRP affectstranslation, the sequence(s) to which it binds is alsocomplex owing to the fact that the protein contains 2KH (RNP K homology) domains and a RGG box. FMRPhas been reported to bind an unusual intramolecularduplex structure known as a G-quartet through theRGG box (Darnell et al. 2001), a small noncoding dsRNA(BC1) via a previously undescribed RNA-binding motif(Zalfa et al. 2003, 2005), and a loop–loop pseudoknot‘‘kissing complex’’ via KH domain 2 (Darnell et al. 2005).However, the only portion of FMRP linked to the FXS isKH domain 2; one individual with an I304N mutationwithin this region displays several characteristics of thesyndrome. Interestingly, the kissing complex, when added






in trans, induces a large shift in the sedimentation profileof FMRP such that it is almost exclusively in the mRNPfraction. This result draws a clear connection betweenthe FXS, FMRP KH domain 2, and the kissing complex. Ofcourse, ‘‘the absence of evidence is not evidence ofabsence,’’ and the other domains of FMRP and the RNAsto which they bind could also contribute to the FXS.

The exon junction complex (EJC) and the regulation ofsynaptic strength

The EJC consists of four core proteins that are probablydeposited on most, if not all, exon–exon junctions fol-lowing intron removal from nuclear pre-mRNA. The fourproteins, eIF4AIII, Y14, Mago, and MLN51 travel with themRNA as it is exported to the cytoplasm where they helpdictate the fate of the transcript. Although it is generallythought that the first, or pioneer round of translationcauses the dissociation of the EJC from mRNA, prior tothis event, this complex can regulate mRNA translation,localization, and destruction in conjunction with otherancillary proteins (Chang et al. 2007; Giorgi and Moore2007; Le Hir and Seraphin 2008; Ma et al. 2008). Theseprocesses are often interconnected; for example, the EJCcan stimulate translation before it dissociates from themRNA during the first ‘‘pioneer round’’ of translationthat is important for RNA quality control. Together withUpf1 and other factors, the EJC can influence nonsense-mediated mRNA decay (NMD), a surveillance mecha-nism to ensure that mRNAs with aberrant stop codonsare destroyed and do not make improper proteins thatcould be deleterious to cells.

As noted earlier, translation at synapses is regulated byseveral factors including mTOR and its effectors andFMRP. Certainly additional translational control mecha-nisms/factors operate at synapses, and Giorgi et al. (2007)have proposed an intriguing new one. They noted thateIF4AIII, the EJC component, displays a dendritic as wellas cell body localization in cultured hippocampal and

cortical neurons (in tissue culture cells such as Hela, thepreponderance of eIF4AIII is nuclear), and interacts withsome dendritic mRNAs such as that encoding arc/arg3.1.Arc/arg3.1 is an immediate early gene whose transcrip-tion is induced by a variety of agents and behaviors in thehippocampus (Waung et al. 2008), all of which probablylead to the activation of N-methyl-D-aspartate receptors(Steward and Worley 2001a). The arc/arg3.1 39 untrans-lated region (UTR) is formed from three exons, and thuswould be expected to have two EJC complexes; hence, theobserved coimmunoprecipitation of this mRNA witheIF4AIII. From this observation, Giorgi et al. (2007)surmised that arc/arg3.1 mRNA could be transported tothe synapto-dendritic compartment in a translationallydormant form accompanied by the EJC, and that uponsynaptic stimulation, a transient burst of arc/arg3.1 pro-tein synthesis would occur, followed soon thereafter bydestruction of the mRNA. Because arc/arg3.1 mRNAEJCs would be located in the 39 UTR, one or a few pioneerrounds of translation presumably would not inducedissociation of the EJC from the mRNA yet the mRNAmay still be subject to NMD-like destruction. In thisscenario, protein synthesis at synapses would be highlyregulated, since after very little translation, the RNAwould be destroyed. The evidence that this is the caserests primarily on fact that a knockdown of eIF4AIIIA incultured neurons leads to increased levels of dendriticarc/arg3.1 protein and RNA levels. The eIF4AIII knock-down also induces increased excitatory synaptic strength,most likely via the addition of glutamate receptors atsynapses (Giorgi et al. 2007).

While attractive, the model of Giorgi et al. (2007) wouldseem to be inconsistent with other observations of arc/arg3.1 mRNA and protein distribution in vivo. Forexample, certain behaviors in rats lead to substantialarc/arg3.1 protein levels in hippocampal cell bodies,indicating that the mRNA is not repressed in thatlocation (Ramırez-Amaya et al. 2005). Similar observa-tions are made when the rat hippocampus is subjected toelectrical stimulation that induces LTP (Steward and

Figure 3. Translational control by 4E-BPs. 7mG (cap)-dependent translation depends on ordered interactions among eIF4E, eIF4G, thelarge multisubunit eIF3, and the 40S ribosomal subunit. The eIF4E–eIF4G interaction can be disrupted by any one of three 4E-BPs,which inhibits the translation of many mRNAs by sequestering eIF4E (Richter and Sonenberg 2005). The eIF4E–eIF4G interaction canalso be disrupted by Maskin, Neuroguidin (Ngd), Bruno, or CYFIP1. Because these proteins are tethered to specific mRNAs via CPEB(and its binding sequence, the CPE), Bruno (and its binding sequence, the BRE), or FMRP (in some cases acting through the smallnoncoding BC1 RNA), translation is inhibited on only a subset of mRNAs.

Richter and Klann





Worley 2001b; Steward et al. 2007). It thus remainsunclear whether, or to what extent, arc/arg3.1 mRNAmay be transported in dendrites in an inactive form.

CPEB-regulated molecular circuitry

CPEB is a sequence-specific RNA-binding protein thatstimulates translation by inducing cytoplasmic poly(A)elongation (Richter 2007). In neurons, CPEB is found atpost-synaptic sites (as well as the cell body) where inresponse to synaptic activity, it induces polyadenylationand translation of several mRNAs (Wu et al. 1998; Huanget al. 2002; Shin et al. 2004; Du and Richter 2005). Theimportance of this protein for translation in the brain wasdemonstrated in a CPEB knockout mouse where thetaburst-induced LTP was reduced in hippocampal ShafferCA-1 neurons (Alarcon et al. 2004). In addition, CPEBknockout mice have a deficit in extinction, a type ofmemory where behavioral responses diminish and even-tually become extinct when there is no reinforcement ofthe memory (Berger-Sweeney et al. 2006). Although ex-tinction requires the formation of new memories, theunderlying mechanisms by which it occurs are probablydistinct from those of memory acquisition and consoli-dation (Able and Lattal 2001).

The key to understanding how CPEB might influencethese complex phenotypes surely lies in the identificationof target mRNAs. To this end, Zearfoss et al. (2008) haveidentified grown hormone (GH) as one protein whoselevel is reduced ;10-fold in the CPEB knockout hippo-campus. GH mRNA contains no 39 UTR cytoplasmicpolyadenylation elements (CPEs), the binding sites forCPEB, and both GH mRNA and pre-mRNA are reduced inthe knockout versus wild-type hippocampus. This resultsuggested that an mRNA encoding a transcription factorthat regulates GH gene expression might be under thedirect control of CPEB. Indeed, c-jun is just such a factor;

it is reduced in the hippocampus of CPEB knockout mice,its 39 UTR contains CPEs, and it coimmunoprecipitatesthe promoter of the GH gene in wild type, but not CPEBknockout mice. Surprisingly, GH itself induces LTP inhippocampal slices that, like electrical stimulation, isreduced in the CPEB knockout mouse. Moreover, the LTPinduced by GH and theta burst stimulation is reduced ifslices are incubated with cordycepin, a drug that inhibitspolyadenylation. These and other results suggest that GHacts in both autocrine and paracrine fashion to regulationplasticity through CPEB control of c-jun mRNA trans-lation (Fig. 4).

CPEB is also found in invertebrates, and in Aplysiasensory neurons where CPEB RNA has been ablated by anantisense oligonucleotide, long-term facilitation (LTF),a form of plasticity, is not properly maintained (Si et al.2003a). However, the isoform of CPEB in Aplysia neuronsdiffers from the CPEB described above in mammals inthat it contains a long stretch of glutamine residues.Polyglutamine is sometimes found in proteins that havecharacteristics of a prion, an infectious agent consistingentirely of protein that is self-reproducing. This observa-tion, plus the fact that CPEB RNA is detected in Aplysianeurons suggested to Si et al. (2003a,b) that this CPEBisoform might assume a prion-like structure followingsynaptic stimulation, thereby forming a protease-resis-tant tag at synapses. If so, then perhaps CPEB itself, asopposed to proteins derived from CPEB-stimulated trans-lation, might comprise the tag that is thought to distin-guish stimulated from naive synapses. Si et al. (2003b)indeed showed that Aplysia CPEB had some features ofa prion in vitro, such as resistance to protease and fastsedimentation rate in sucrose gradients. The most com-pelling evidence, however, comes from experiments inyeast, where Alpysia CPEB was shown to assume twoforms: one that is aggregated (i.e., prion-like) and one thatis not (Si et al. 2003b). Surprisingly, not only was the

Figure 4. Model for CPEB-regulated mo-lecular circuitry in neurons. (1) CPEB inthe synapto-dendritic compartment is acti-vated via N-methyl-D-aspartate receptors(NMDAR) that are stimulated by calcium.(2) Active CPEB binds the CPE in the c-junmRNA and stimulates translation. (3)Newly synthesized c-jun protein is trans-ported in retrograde fashion to the nucleuswhere it stimulates GH transcription. (4)GH is synthesized and secreted; it theninteracts with receptors on the same cell(autocrine) or nearby cells (paracrine). TheGH receptors signal through phospho-JAK2and phospho-STAT3, which enters the nu-cleus to stimulate gene transcription (5),where it modifies plasticity and other sig-naling events (4). (6) Newly formed NMDAand AMPA receptors maintain LTP.






aggregated form of CPEB the only one to bind RNA invitro, the aggregated form converted the nonaggregatedform into an aggregated form. Such epigenetic inheri-tance is a fundamental hallmark of prion formation. Siet al. (2003b) hypothesized that synaptic stimulationmight cause the neuronal Aplysia CPEB isoform toassume a prion-like state, which could stimulate thetranslation of some RNAs, cause it to alter its substratespecificity, or release some mRNAs from an inhibitedstate. Si et al. (2003b) further suggest that once in a prionform, CPEB would need no further stimulation (e.g., bykinases) to maintain its activity.

If polyglutamine-containing CPEB forms a prion ininvertebrate neurons, then what about the polyglut-amine-lacking CPEB in vertebrate neurons? Vertebratescontain three additional CPEB-like genes, all of which areexpressed in the brain (Theis et al. 2003). Two of theseother CPEB-like proteins do contain some polyglutamine,although they are not nearly as long as that in the AplysiaCPEB. Moreover, these other CPEB-like proteins do nothave a strong affinity for the CPE and do not supportcytoplasmic polyadenylation (Huang et al. 2006). Thus,the relationship between vertebrate CPEB proteins andprions, if any, remains to be determined. Nonetheless, itis noteworthy that the Drosophila CPEB isoform calledOrb2 is found in a head structure (the mushroom bodies)that is important for LTM and contains polyglutamine;when the polyglutamine is deleted, LTM, but not STM isimpaired (Keleman et al. 2007). While these data do notindicate whether prion formation, or even mRNA trans-lation was involved, they do point to the importance ofthe glutamine stretch for CPEB isoform function inmemory formation.

Conclusions

Overwhelming evidence indicates that synapto-dendriticmRNA translation is a necessary component of long-lasting synaptic plasticity, but many of the most intrigu-ing questions remain unanswered. Synaptic connectionsare strengthened by LTP and weakened by LTD, yet bothappear to require synapto-dendritic mRNA translationthat is stimulated by many of same upstream kinasesignaling events. Do LTP and LTD lead different spectra ofproteins that are synthesized, and if so, how does thisoccur? Conversely, if the same proteins are synthesized,how do synapses ‘‘know’’ whether to strengthen orweaken? In either case, synaptic tagging, the deposition ofa tag or mark at synapse, once it is stimulated to distinguishbetween those that are naıve from those that are experi-enced (Frey and Morris 1997) presumably is involved.Kelleher et al. (2004) has suggested that specific tags forLTP and LTD permit the ‘‘capture’’ of specific proteins atthe synapse, which in turn produce either LTP or LTD. Ofcourse, this begs the question of the nature of the tag, andwhat distinguishes an LTP tag from an LTD tag. Perhapsrelatively rare mRNAs specific for either LTP or LTD aretranslated at synapses, and that the encoded proteinscapture subsequent newly made proteins that produce thebiochemical and morphological changes that we recognize

as strengthened or weakened synapses. However, proteinsynthesis inhibitors do not block synaptic tagging in thehippocampus (Frey and Morris 1997), which argues againstnewly synthesized proteins being a synaptic tag for eitherLTP or LTD. Clearly, the identification of the full panoply oftranslation factor modifications, mRNA-binding proteins,and mRNAs that are translated at synapses is both essentialand challenging.

It is also worth noting that some of the same trans-lational control mechanisms that influence LTP and LTMunder normal conditions are impaired under certainpathologies of the brain. For example, during ischemiaand reperfusion, which often is associated with deficientLTM, there are long-lasting decreases in protein synthesisand increased phosphorylation of eIF2a (DeGracia et al.2002). In addition, increased eIF2a phosphoryation in thebrain is associated with epilepsy (Carnevalli et al. 2006)and neuronal culture models of Parkinson’s disease (PD)(Holtz and O’Malley 2003). Moreover, eIF2a phosphory-lation increases in the hippocampus of mouse models ofAlzheimer’s disease (Page et al. 2006; Kim et al. 2007), aswell as brains of AD patients (Kim et al. 2007). Mutationsin phosphatase and tensin homolog on chromosome ten(PTEN), a tumor suppressor gene involved in upstreamregulation of Akt and mTOR signaling, have beenreported in autistic individuals with macrocephaly (But-ler et al. 2005). Mice with a neuron-specific deletion ofPTEN exhibit excessive Akt/mTOR signaling and behav-iors consistent with autism (Kwon et al. 2006). Asmentioned earlier, mutations in either TSC1 or TSC2,both of which result in up-regulated mTOR signaling,cause TSC that often results in mental retardation andautism (DiMario 2004; Wiznitzer 2004). Finally, misre-gulation of mTOR signaling has been implicated in theexpression of a cellular marker of neuronal loss in a mousemodel of PD (Malagelada et al. 2006). Thus, multiple linesof evidence implicate aberrant translational control,especially via altered eIF2a phosphorylation and mTORsignaling, in several neurological disorders.

References

Abel, T., and Lattal, K.M. 2001. Molecular mechanisms ofmemory acquisition, consolidation and retrieval. Curr. Opin.

Neurobiol. 11: 180–187.Abel, T., Martin, K.C., Bartsch, D., and Kandel, E.R. 1998.

Memory suppressor genes: Inhibitory constraints on thestorage of long-term memory. Science 279: 338–341.

Alarcon, J.M., Hodgman, R., Theis, M., Huang, Y.S., Kandel,E.R., and Richter, J.D. 2004. Selective modulation of someforms of schaffer collateral-CA1 synaptic plasticity in mice witha disruption of the CPEB-1 gene. Learn. Mem. 11: 318–327.

Antar, L.N., Afroz, R., Dictenberg, J.B., Carroll, R.C., andBassell, G.J. 2004. Metabotropic glutamate receptor activa-tion regulates fragile x mental retardation protein and FMR1mRNA localization differentially in dendrites and at synap-ses. J. Neurosci. 24: 2648–2655.

Antion, M.D., Hou, L., Wong, H., Hoeffer, C.A., and Klann, E.2008a. mGluR-dependent long-term depression is associatedwith increased phosphorylation of S6 and synthesis of EF1A,but remains expressed in S6K-deficient mice. Mol. Cell. Biol.

28: 2996–3007.

Richter and Klann





Antion, M.D., Merhav, M., Hoeffer, C.A., Reis, G., Kozma, S.C.,Thomas, G., Schuman, E.M., Rosenblum, K., and Klann, E.2008b. Deletion of S6K1 and S6K2 leads to divergent alter-ations in learning, memory, and synaptic plasticity. Learn.

Mem. 15: 29–38.Bagni, C. and Greenough, W.T. 2005. From mRNP trafficking to

spine dysmorphogenesis: The roots of fragile X syndrome.Nat. Neurosci. 6: 376–387.

Bai, X., Ma, D., Liu, A., Shen, X., Wang, Q.J., Liu, Y., and Jiang, Y.2007. Rheb activates mTOR by antagonizing its endogenousinhibitor, FKBP38. Science 318: 977–980.

Banko, J.L., Poulin, F., Hou, L., DeMaria, C.T., Sonenberg, N.,and Klann, E. 2005. The translation repressor 4E-BP2 iscritical for eIF4F complex formation, synaptic plasticity,and memory in the hippocampus. J. Neurosci. 25: 9581–9590.

Banko, J.L., Hou, L., Poulin, F., Sonenberg, N., and Klann, E. 2006.Regulation of eukaryotic initiation factor 4E by convergingsignaling pathways during metabotropic glutamate receptor-dependent long-term depression. J. Neurosci. 26: 2167–2173.

Banko, J.L., Merhav, M., Sonenberg, N., Rosenblum, K., and Klann,E. 2007. Behavioral alterations in mice that lack the translationrepressor 4E-BP2. Neurobiol. Learn. Mem. 87: 248–256.

Bear, M.F., Huber, K.M., and Warren, S.T. 2004. The mGluRtheory of fragile X mental retardation. Trends Neurosci. 27:370–377.

Berger-Sweeney, J., Zearfoss, N.R., and Richter, J.D. 2006. Re-duced extinction of hippocampal-dependent memories inCPEB knockout mice. Learn. Mem. 13: 4–7.

Blundell, J., Kouser, M., and Powell, C.M. 2008. Systemic inhi-bition of mammalian target of rapamycin inhibits fear mem-ory reconsolidation. Neurobiol. Learn. Mem. 90: 28–35.

Bolduc, F.V., Bell, K., Cox, H., Broadie, K.S., and Tully, T. 2008.Excess protein synthesis in Drosophila fragile X mutantsimpairs long-term memory. Nat. Neurosci. 11: 1143–1145.

Bradshaw, K.D., Emptage, N.J., and Bliss, T.V. 2003. A role fordendritic protein synthesis in hippocampal late LTP. Eur. J.

Neurosci. 18: 3150–3152.Brazil, D.P., and Hemmings, B.A. 2001. Ten years of protein

kinase B signalling: A hard Akt to follow. Trends Biochem.Sci. 26: 657–664.

Brown, V., Jin, P., Ceman, S., Darnell, J.C., O’Donnell, W.T.,Tenenbaum, S.A., Jin, X., Feng, Y., Wilkinson, K.D., Keene,J.D., et al. 2001. Microarray identification of FMRP-associ-ated brain mRNAs and altered mRNA translational profilesin fragile X syndrome. Cell 107: 477–487.

Butler, M.G., Dasouki, M.J., Zhou, X.P., Talebizadeh, Z., Brown,M., Takahashi, T.N., Miles, J.H., Wang, C.H., Stratton, R.,and Pilarski, R., et al. 2005. Subset of individuals with autismspectrum disorders and extreme macrocephaly associatedwith germline PTEN tumour supressor gene mutations. J.Med. Genet. 42: 318–321.

Carnevalli, L.S., Pereira, C.M., Jaqueta, C.B., Alves, V.S., Paiva,V.N., Vattem, K.M., Wek, R.C., Mello, L.E., and Castilho,B.A. 2006. Phosphorylation of the a subunit of translationinitiation factor-2 by PKR mediates protein synthesis in-hibition in the mouse brain during status epilepticus. Bio-

chem. J. 397: 187–194.Ceman, S., O’Donnell, W.T., Reed, M., Patton, S., Pohl, J., and

Warren, S.T. 2003. Phosphorylation influences the transla-tion state of FMRP-associated polyribosomes. Hum. Mol.

Genet. 12: 3295–3305.Chang, Y.-F., Imam, J.S., and Wilkinson, M.F. 2007. The non-

sense-mediated decay RNA surveillance pathway. Annu.

Rev. Biochem. 76: 51–74.Chen, A., Muzzio, I.A., Malleret, G., Bartsch, D., Verbitsky, M.,

Pavlidis, P., Yonan, A.L., Vronskaya, S., Grody, M.B., and

Cepeda, I., et al. 2003. Inducible enhancement of memorystorage and synaptic plasticity in transgenic mice expressingan inhibitor of ATF4 (CREB-2) and C/EBP proteins. Neuron

39: 655–669.Choi, K.M., McMahon, L.P., and Lawrence Jr., J.C. 2003. Two

motifs in the translational repressor PHAS-I required for effi-cient phosphorylation by mammalian target of rapamycin andfor recognition by raptor. J. Biol. Chem. 278: 19667–19673.

Costa-Mattioli, M., Gobert, D., Harding, H., Herdy, B., Azzi, M.,Bruno, M., Bidinosti, M., BenMamou, C., Marcinkiewicz, E.,Yoshida, M., et al. 2005. Translational control of hippocam-pal synaptic plasticity and memory by the eIF2a kinaseGCN2. Nature 436: 1166–1173.

Costa-Mattioli, M., Gobert, D., Stern, E., Gamache, K., Colina,R., Cuello, C., Sossin, W., Kaufman, R., Pelletier, J., andRosenblum, K., et al. 2007. eIF2a phosphorylation bidirec-tionally regulates the switch from short- to long-term syn-aptic plasticity and memory. Cell 129: 195–206.

Darnell, J.C., Jensen, K.B., Jin, P., Brown, V., Warren, S.T., andDarnell, R.B. 2001. Fragile X mental retardation proteintargets G quartet mRNAs important for neuronal function.Cell 107: 489–499.

Darnell, J.C., Fraser, C.E., Mostovetsky, O., Stefani, G., Jones,T.A., Eddy, S.R., and Darnell, R.B. 2005. Kissing complexRNAs mediate interaction between the Fragile-X mentalretardation protein KH2 domain and brain polyribosomes.Genes & Dev. 19: 903–918.

Davidkova, G. and Carroll, R.C. 2007. Characterization of therole of microtubule-associated protein 1B in metabotropicglutamate receptor-mediated endocytosis of AMPA receptorsin hippocampus. J. Neurosci. 27: 13273–13278.

DeGracia, D.J., Kumar, R., Owen, C.R., Krause, G.S., and White,B.C. 2002. Molecular pathways of protein synthesis inhibi-tion during brain reperfusion: Implications for neuronalsurvival or death. J. Cereb. Blood Flow Metab. 22: 127–141.

DiMario Jr., F.J. 2004. Brain abnormalities in tuberous sclerosiscomplex. J. Child Neurol. 19: 650–657.

Dolen, D., Osterweil, E., Rao, B.S., Smith, G.B., Auerbach, B.D.,Chattarji, S., and Bear, M.F. 2007. Correction of Fragile Xsyndrome in mice. Neuron 56: 955–962.

Du, L., and Richter, J.D. 2005. Activity-dependent polyadenyla-tion in neurons. RNA 11: 1340–1347.

Ehninger, D., Han, S., Shilyansky, C., Zhou, Y., Li, W., Kwiat-kowski, D.J., Ramesh, V., and Silva, A.J. 2008. Reversal oflearning of deficits in a Tsc2+/� mouse model of tuberoussclerosis. Nat. Med. 14: 843–848.

Frey, U., and Morris, R.G. 1997. Synaptic tagging and long-termpotentiation. Nature 385: 533–536.

Gallagher, S.M., Daly, C.A., Bear, M.F., and Huber, K.M. 2004.Extracellular signal-regulated protein kinase activation isrequired for metabotropic glutamate receptor-dependentlong-term depression in hippocampal area CA1. J. Neurosci.

24: 4859–4864.Garami, A., Zwartkruis, F.J., Nobukuni, T., Joaquin, M., Roccio,

M., Stocker, H., Kozma, S.C., Hafen, E., Bos, J.L., andThomas, G. 2003. Insulin activation of Rheb, a mediator ofmTOR/S6K/4E-BP signaling, is inhibited by TSC1 and 2.Mol. Cell 11: 1457–1466.

Gelinas, J.N., and Nguyen, P.V. 2005. b-Adrenergic receptor acti-vation facilitates induction of a protein synthesis-dependent lastphase of long-term potentiation. J. Neurosci. 25: 3294–3303.

Gelinas, J.N., Banko, J.L., Hou, L., Sonenberg, N., Weeber, E.J.,Klann, E., and Nguyen, P.V. 2007. ERK and mTOR signalingcouple b-adrenergic receptors to translation initiation to gatethe induction of protein synthesis-dependent LTP. J. Biol.Chem. 282: 27527–27535.






Gingras, A.C., Raught, B., and Sonenberg, N. 2001. Regulation oftranslation initiation by FRAP/mTOR. Genes & Dev. 15: 807–826.

Giorgi, C. and Moore, M.J. 2007. The nuclear nurture andcytoplasmic nature of localized mRNPs. Semin. Cell Dev.Biol. 18: 186–193.

Giorgi, C., Yeo, G.C., Stone, M.E., Katz, D.B., Burge, C.,Turrigiano, G., and Moore, M.J. 2007. The EJC factor eIF4AIIImodulates synaptic strength and neuronal protein expres-sion. Cell 130: 179–191.

Goorden, S.M., van Woerden, G.M., van der Weerd, L., Cheadle,J.P., and Elgersma, Y. 2007. Cognitive deficits in Tsc1+/�micein the absence of cerebral lesions and seizures. Ann. Neurol.62: 648–655.

Harding, H.P., Novoa, I., Zhang, Y., Zeng, H., Wek, R., Schapira,M., and Ron, D. 2000. Regulated translation initiationcontrols stress-induced gene expression in mammalian cells.Mol. Cell 6: 1099–1108.

Hinnebusch, A.G., 2000. Mechanism and regulation of initiatormethionyl-tRNA binding to ribosomes. In Translationalcontrol of gene expression (eds. N. Sonenberg et al.), pp.185–244. Cold Spring Harbor Laboratory Press, Cold SpringHarbor, NY.

Hoeffer, C.A., Tang, W., Wong, H., Santillan, A., Patterson, R.J.,Martinez, L.A., Tejada-Simon, M.V., Paylor, R., Hamilton,S.L., and Klann, E. 2008. Removal of FKBP12 enhancesmTOR/Raptor interactions, LTP, memory, and autistic-likeperseveration. Neuron 60: 832–845.

Holtz, W.A., and O’Malley, K.L. 2003. Parkinsonian mimeticsinduce aspects of unfolded protein response in death ofdopaminergic neurons. J. Biol. Chem. 278: 19367–19377.

Hou, L., and Klann, E. 2004. Activation of the phosphoinositide3-kinase–Akt–mammalian target of rapamycin signalingpathway is required for metabotropic glutamate receptor-dependent long-term depression. J. Neurosci. 24: 6352–6361.

Hou, L., Antion, M.D., Hu, D., Spencer, C.M., Paylor, R.E., andKlann, E. 2006. Dynamic translational and proteasomal regu-lation of fragile X mental retardation protein controls mGluR-dependent long-term depression. Neuron 51: 441–454.

Huang, Y.S., Jung, M.Y., Sarkissian, M., and Richter, J.D. 2002. N-methyl-D-aspartate receptor signaling results in Aurora ki-nase-catalyzed CPEB phosphorylation and a CaMKII mRNApolyadenylation at synapses. EMBO J. 21: 2139–2148.

Huang, Y.S., Kan, M.C., Lin, C.L., and Richter, J.D. 2006. CPEB3and CPEB4 in neurons: Analysis of RNA-binding specificityand translational control of AMPA receptor GluR2 mRNA.EMBO J. 25: 4865–4876.

Huber, K.M., Kayser, M.S., and Bear, M.F. 2000. Role for rapiddendritic protein synthesis in hippocampal mGluR-depen-dent long-term depression. Science 288: 1254–1257.

Huber, K.M., Gallagher, S.M., Warren, S.T., and Bear, M.F. 2002.Altered synaptic plasticity in a mouse model of fragile Xmental retardation. Proc. Natl. Acad. Sci. 99: 7746–7750.

Inoki, K., Corradetti, M.N., and Guan, K.L. 2005. Dysregulationof the TSC–mTOR pathway in human disease. Nat. Genet.

37: 19–24.Jung, M.Y., Lorenz, L., and Richter, J.D. 2006. Translational

control by neuroguidin, a eukaryotic initiation factor 4E andCPEB binding protein. Mol. Cell. Biol. 26: 4277–4287.

Kandel, E.R., 2001. The molecular biology of memory storage: Adialogue between genes and synapses. Science 294: 1030–1038.

Kang, H. and Schuman, E.M. 1996. A requirement for localprotein synthesis in neurotrophin-induced hippocampal syn-aptic plasticity. Science 273: 1402–1406.

Keleman, K., Kruttner, S., Alenius, M., and Dickson, B.J. 2007.Function of Drosophila CPEB protein Orb2 in long-

term memory courtship memory. Nat. Neurosci. 10: 1587–1593.

Kelleher 3rd, R.J., Govindarajan, A., Jung, H.Y., Kang, H., andTonegawa, S. 2004. Translational control by MAPK signaling inlong-term synaptic plasticity and memory. Cell 116: 467–479.

Kim, D.H., Sarbassov, D.D., Ali, S.M., King, J.E., Latek, R.R.,Erdjument-Bromage, H., Tempst, P., and Sabatini, D.M. 2002.mTOR interacts with raptor to form a nutrient-sensitivecomplex that signals to the cell growth machinery. Cell 110:163–175.

Kim, H.S., Choi, Y., Shin, K.Y., Joo, Y., Lee, Y.K., Jung, S.Y., Suh,Y.H., and Kim, J.H. 2007. Swedish amyloid precursor proteinmutation increases phosphorylation of eIF2a in vitro and invivo. J. Neurosci. Res. 85: 1528–1537.

Klann, E. and Dever, T.E. 2004. Biochemical mechanisms fortranslational regulation in synaptic plasticity. Nat. Rev.Neurosci. 5: 931–942.

Klann, E., Antion, M.D., Banko, J.L., and Hou, L. 2004. Synapticplasticity and translation initiation. Learn. Mem. 11: 365–372.

Kwon, C.H., Luikart, B.W., Powell, C.M., Zhou, J., Matheny,S.A., Zhang, W., Li, Y., Baker, S.J., and Parada, L.F. 2006. Ptenregulates neuronal arborization and social interaction inmice. Neuron 50: 377–388.

Le Hir, H., and Seraphin, B. 2008. EJCs at the heart of trans-lational control. Cell 133: 213–216.

Ma, X.M., Yoon, S.O., Richardson, C.J., Julich, K., and Blenis, J.2008. SKAR links pre-mRNA splicing to mTOR/S6K1-medi-ated enhanced translation efficiency of spliced mRNAs. Cell

133: 303–313.Malagelada, C., Ryu, E.J., Biswas, S.C., Jackson-Lewis, V., and

Greene, L.A. 2006. RTP801 is elevated in Parkinson brainsubstantia nigral neurons and mediates death in cellularmodels of Parkinson’s disease by a mechanism involvingmammalian target of rapamycin inactivation. J. Neurosci. 26:9996–10005.

Malenka, R.C. and Bear, M.F. 2004. LTP and LTD: An embar-rassment of riches. Neuron 44: 5–21.

Malenka, R.C. and Nicoll, R.A. 1999. Long-term potentiation—A decade of progress? Science 285: 1870–1874.

Marcotrigiano, J., Gingras, A.C., Sonenberg, N., and Burley, S.K.1999. Cap-dependent translation initiation in eukaryotes isregulated by a molecular mimic of eIF4G. Mol. Cell 3: 707–716.

McBride, S.M., Choi, C.H., Wang, Y., Liebelt, D., Braunstein, E.,Ferreiro, D., Sehgal, A., Siwicki, K.K., Dockendorff, T.C.,Nguyen, H.T., et al. 2005. Pharmacological rescue of synapticplasticity, courtship behavior, and mushroom body defects ina Drosophila model of fragile X syndrome. Neuron 45: 753–764.

Muddashetty, R.S., Kelic, S., Gross, C., Xu, M., and Bassell, G.J.2007. Dysregulated metabotropic glutamate receptor-dependenttranslation of AMPA receptor and postsynaptic density-95mRNAs at synapses in a mouse model of fragile X syndrome.J. Neurosci. 27: 5338–5348.

Nakamura, A., Sato, K., and Hanyu-Nakamura, K. 2004. Dro-sophila cup is an eIF4E binding protein that associates withBruno and regulates oskar mRNA translation in oogenesis.Dev. Cell 6: 69–78.

Napoli, I., Mercaldo, V., Boyl, P.P., Eleuteri, B., Zalfa, F., DeRubeis, S., Di Marino, D., Mohr, E., Massimi, M., andFalconi, M., et al. 2008. The Fragile X syndrome proteinrepresses activity-dependent translation through CYFIP1,a new 4E-BP. Cell 134: 1042–1054.

Narayanan, U., Nalavadi, V., Nakamoto, M., Thomas, G.,Ceman, S., Bassell, G.J., and Warren, S.T. 2008. S6K1 phos-phorylates and regulates FMRP with the neuronal proteinsynthesis-dependent mTOR signaling cascade. J. Biol. Chem.283: 18478–18482.

Richter and Klann





Page, G., Rioux Bilan, A., Ingrand, S., Lafay-Chebassier, C., Pain,S., Perault Pochat, M.C., Bouras, C., Bayer, T., and Hugon, J.2006. Activated double-stranded RNA-dependent proteinkinase and neuronal death in models of Alzheimer’s disease.Neuroscience 139: 1343–1354.

Park, S., Park, J.M., Kim, S., Kim, J.A., Shepherd, J.D., Smith-Hicks, C.L., Chowdhury, S., Kaufmann, W., Kuhl, D., andRyazanov, A.G. , et al. 2008. Elongation factor 2 and fragileX mental retardation protein control the dynamic translationof Arc/Arg3.1 essential for mGluR-LTD. Neuron 59: 70–83.

Parsons, R.G., Gafford, G.M., and Helmstetter, F.J. 2006. Trans-lational control via the mammalian target of rapamycin path-way is critical for the formation and stability of long-term fearmemory in amygdala neurons. J. Neurosci. 26: 12977–12983.

Ramırez-Amaya, V., Vazdarjanova, A., Mikhael, D., Rosi, S.,Worley, P.F., and Barnes, C.A. 2005. Spatial exploration-inducedarc mRNA and protein expression: Evidence for selective,network-specific reactivation. J. Neurosci. 25: 1761–1768.

Raught, B., Peiretti, F., Gingras, A.C., Livingstone, M., Shahba-zian, D., Mayeur, G.L., Polakiewicz, R.D., Sonenberg, N., andHershey, J.W. 2004. Phosphorylation of eucaryotic transla-tion initiation factor 4B Ser422 is modulated by S6 kinases.EMBO J. 23: 1761–1769.

Richter, J.D. 1996. Dynamics of poly(A) addition and removalduring development. In Translational control (eds. J.W.B.Hershey et al.), pp. 481–504. Cold Spring Harbor LaboratoryPress, Cold Spring Harbor, NY.

Richter, J.D. 2007. CPEB: A life in translation. Trends Biochem.

Sci. 32: 279–285.Richter, J.D. and Sonenberg, N. 2005. Regulation of cap-dependent

translation by eIF4E inhibitory proteins. Nature 433: 477–480.Ronesi, J.A. and Huber, K.M. 2008. Homer interactions are

necessary for metabotropic glutamate receptor-inducedlong-term depression and translational activation. J. Neuro-

sci. 28: 543–547.Rong, L., Livingstone, M., Sukarieh, R., Petroulakis, E., Gingras,

A.C., Crosby, K., Smith, B., Polakiewicz, R.D., Pelletier, J.,Ferriuolo, M.A., et al. 2008. Control of eIF4E cellular localiza-tion by eIF4E-binding proteins, 4E-BPs. RNA 14: 1318–1327.

Sabatini, D.M. 2006. mTOR and cancer: Insights into a complexrelationship. Nat. Rev. Cancer 6: 729–734.

Schalm, S.S., Fingar, D.C., Sabatini, D.M., and Blenis, J. 2003.TOS motif-mediated raptor binding regulates 4E-BP1 multi-site phosphorylation and function. Curr. Biol. 13: 797–806.

Schicknick, H., Schott, B.H., Budinger, E., Smalla, K.H., Riedel,A., Seidenbecher, C.I., Scheich, H., Gundelfinger, E.D., andTischmeyer, W. 2008. Dopaminergic modulation of auditorycortex-dependent memory consolidation through mTOR.Cereb. Cortex 18: 2646–2658.

Shin, C.Y., Kundel, M., and Wells, D.G. 2004. Rapid, activity-induced increase in tissue plasminogen activator is mediatedby metabotropic glutamate receptor-dependent mRNA trans-lation. J. Neurosci. 24: 9425–9433.

Si, K., Lindquist, S., and Kandel, E.R. 2003a. A neuronal isoform ofthe aplysia CPEB has prion-like properties. Cell 115: 879–891.

Si, K., Giustetto, M., Etkin, A., Hsu, R., Janisiewicz, A.M.,Miniaci, M.C., Kim, J.H., Zhu, H., and Kandel, E.R. 2003b.A neuronal isoform of CPEB regulates local protein synthesisand stabilizes synapse-specific long-term facilitation in aply-sia. Cell 115: 893–904.

Silva, A.J., Kogan, J.H., Frankland, P.W., and Kida, S. 1998. CREBand memory. Annu. Rev. Neurosci. 21: 127–148.

Sonenberg, N. and Dever, T.E. 2003. Eukaryotic translation initi-ation factors and regulators. Curr. Opin. Struct. Biol. 13: 56–63.

Stefani, G., Fraser, C.E., Darnell, J.C., and Darnell, R.B. 2004.Fragile X mental retardation protein is associated with

translating polyribosomes in neuronal cells. J. Neurosci. 24:7272–7276.

Steward, O. and Schuman, E.M. 2001. Protein synthesis atsynaptic sites on dendrites. Annu. Rev. Neurosci. 24: 299–325.

Steward, O. and Worley, P.F. 2001a. Selective targeting of newlysynthesized arc mRNA to active synapses requires NMDAreceptor activation. Neuron 30: 227–240.

Steward, O., and Worley, P.F. 2001b. A cellular mechanism fortargeting newly synthesized mRNAs to synaptic sites ondendrites. Proc. Natl. Acad. Sci. 98: 7062–7068.

Steward, O., Huang, F., and Guzowski, J.F. 2007. A form ofperforant path LTP can occur without ERK1/2 phosphorylationor immediate early gene expression. Learn. Mem. 14: 433–445.

Theis, M., Si, K., and Kandel, E.R. 2003. Two previouslyundescribed members of the mouse CPEB family of genesand their inducible expression in the principal cell layers ofthe hippocampus. Proc. Natl. Acad. Sci. 100: 9602–9607.

Tischmeyer, W., Schicknick, H., Kraus, M., Seidenbecher, C.I., Staak,S., Scheich, H., and Gundelfinger, E.D. 2003. Rapamycin-sensitive signalling in long-term consolidation of auditorycortex-dependent memory. Eur. J. Neurosci. 18: 942–950.

Todd, P.K., Mack, K.J., and Malter, J.S. 2003. The fragile Xmental retardation protein is required for type-I metabo-tropic glutamate receptor-dependent translation of PSD-95.Proc. Natl. Acad. Sci. 100: 14374–14378.

Tsokas, P., Grace, E.A., Chan, P., Ma, T., Sealfon, S.C., Iyengar,R., Landau, E.M., and Blitzer, R.D. 2005. Local proteinsynthesis mediates a rapid increase in dendritic elongationfactor 1A after induction of late long-term potentiation. J.

Neurosci. 25: 5833–5843.Tsokas, P., Ma, T., Iyengar, R., Landau, E.M., and Blitzer, R.D.

2007. Mitogen-activated protein kinase upregulates the den-dritic translation machinery in long-term potentiation bycontrolling the mammalian target of rapamycin pathway. J.

Neurosci. 27: 5885–5894.Waung, M.W., Pfeiffer, B.E., Nosyreva, E.D., Ronesi, J.A., and

Huber, K.M. 2008. Rapid translation of Arc/Arg3.1 selec-tively mediates mGluR-dependent LTD through persis-tent increases in AMPAR endocytosis rate. Neuron 59:84–97.

Wiznitzer, M. 2004. Autism and tuberous sclerosis. J. Child

Neurol. 19: 675–679.Wu, L., Wells, D., Tay, J., Mendis, D., Abbott, M.A., Barnitt, A.,

Quinlan, E., Heynen, A., Fallon, J.R., and Richter, J.D. 1998.CPEB-mediated cytoplasmic polyadenylation and the regula-tion of experience-dependent translation of a-CaMKII mRNAat synapses. Neuron 21: 1129–1139.

Yan, Q.J., Rammal, M., Tranfaglia, M., and Bauchwitz, R.P. 2005.Suppression of two major Fragile X Syndrome mouse modelphenotypes by the mGluR5 antagonist MPEP. Neurophar-macology 49: 1053–1066.

Zalfa, F., Giorgi, M., Primerano, B., Moro, A., Di Penta, A., Reis,S., Oostra, B., and Bagni, C. 2003. The fragile X syndromeprotein FMRP associates with BC1 RNA and regulates thetranslation of specific mRNAs at synapses. Cell 112: 317–327.

Zalfa, F., Adinolfi, S., Napoli, I., Kuhn-Holsken, E., Urlaub, H.,Achsell, T., Pastore, A., and Bagni, C. 2005. Fragile X mentalretardation protein (FMRP) binds specifically to the braincytoplasmic RNAs BC1/BC200 via a novel RNA-bindingmotif. J. Biol. Chem. 280: 33403–33410.

Zearfoss, N.R., Alarcon, J.M., Trifilieff, P., Kandel, E., andRichter, J.D. 2008. A molecular circuit composed of CPEB-1 and c-jun controls growth hormone-mediated synapticplasticity in the mouse hippocampus. J. Neurosci. 28:8502–8509.






10.1101/gad.1735809Access the most recent version at doi: 23:2009, Genes Dev.

Joel D. Richter and Eric Klann translational regulationMaking synaptic plasticity and memory last: mechanisms of

References

http://genesdev.cshlp.org/content/23/1/1.full.html#ref-list-1

This article cites 108 articles, 47 of which can be accessed free at:

License

ServiceEmail Alerting

click here.right corner of the article or

Receive free email alerts when new articles cite this article - sign up in the box at the top

Copyright © 2009 by Cold Spring Harbor Laboratory Press


http://genesdev.cshlp.org/lookup/doi/10.1101/gad.1735809

http://genesdev.cshlp.org/content/23/1/1.full.html#ref-list-1

http://genesdev.cshlp.org/cgi/alerts/ctalert?alertType=citedby&addAlert=cited_by&saveAlert=no&cited_by_criteria_resid=protocols;10.1101/gad.1735809&return_type=article&return_url=http://genesdev.cshlp.org/content/10.1101/gad.1735809.full.pdf

http://genesdev.cshlp.org/cgi/adclick/?ad=55564&adclick=true&url=https%3A%2F%2Fhorizondiscovery.com%2Fen%2Fcustom-synthesis%2Fcustom-rna%3Futm_source%3DCSHL_RNA%26utm_medium%3Dbanner%26utm_campaign%3Dcustom_synth%26utm_term%3Doligos%26utm_content%3Djan21



Documents

Making synaptic plasticity and memory last: mechanisms of