TINS-1111; No. of Pages 9

Review

Macromolecular transport in synapse to nucleus communication Nicolas Panayotis1, Anna Karpova2, Michael R. Kreutz2, and Mike Fainzilber1 1 2

Department of Biological Chemistry, Weizmann Institute of Science, Rehovot 76100, Israel Neuroplasticity Group, Leibniz Institute for Neurobiology, Brenneckestraße 6, Magdeburg 39118, Germany

Local signaling events at synapses or axon terminals must be communicated to the nucleus to elicit transcriptional responses. The lengths of neuronal processes pose a significant challenge for such intracellular communication. This challenge is met by mechanisms ranging from rapid signals encoded in calcium waves to slower macromolecular signaling complexes carried by molecular motors. Here we summarize recent findings on macromolecular signaling from the synapse to the nucleus, in comparison to those employed in injury signaling along axons. A number of common themes emerge, including combinatorial signal encoding by post-translational mechanisms such as differential phosphorylation and proteolysis, and conserved roles for importins in coordinating signaling complexes. Neurons may integrate ionic flux with motor-transported signals as a temporal code for synaptic plasticity signaling. Communication between synapse and soma Neurons communicate at synaptic contacts by signaling events, yet their long-term responses to such stimuli require changes in gene expression at the nucleus. In contrast to many other cell types, the intracellular distances that separate synapse from soma in neurons are well beyond the effective range of diffusion-dependent mechanisms [1–4]. How then are distant stimuli sensed and decoded by the nucleus? Conversely, how are changes in transcription or translation in the soma conveyed to specific synapses or terminals? A series of events, including membrane depolarization and calcium influx, was shown to contribute to information transfer in neurons via electrophysiological encoding, while the physical transport of macromolecules was thought to serve secondary roles [5–7]. In contrast, long-distance transport of protein messengers in axons is well established as being essential for neuronal maintenance, survival, and regeneration [8– 10]. Can localized activation of macromolecules combined with intracellular trafficking also enable communication between the synapse and the soma [11]? Here we review recent advances in the understanding of macromolecular communication between the synapse and the nucleus, and Corresponding author: Fainzilber, M. ([email protected]). Keywords: synaptic plasticity; dendrite; axonal transport; nucleocytoplasmic transport; dynein; kinesin; myosin; importin; karyopherin. 0166-2236/ ß 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.tins.2014.12.001

discuss how these advances can be integrated with the more established fast coupling mechanisms underlying synaptic plasticity. Nuclear integration of synaptic events by fast coupling mechanisms The prevailing viewpoint in the literature is that calcium signals are the major route for communication of synaptic activity to the nucleus [7,12]. According to this model, the activation of receptor-gated calcium channels in the distal dendrites is propagated by action potentials or calcium waves to the soma, thereby enabling the local activation of calcium-dependent pathways connected to gene transcription. The fast genomic response to sustained rises in nuclear calcium is generally believed to be important for the induction of immediate early gene expression (IEG) within a few minutes of the stimulus [5,7,13,14]. It has been suggested that IEG transcription can be subdivided into rapid versus delayed transcription [15]. A distinctive feature is the near-instantaneous transcription by the stalling of RNA polymerase II directly downstream of the transcription start site and it is likely that very rapid transcription events require fast calcium signaling [15,16]. In the electrochemical coupling model, the activation of calcium-permeable ion channels in distal dendrites generates back-propagating action potentials or regenerative calcium waves that are conveyed along the endoplasmic reticulum (ER) to the soma [12]. Their arrival at the nucleus enables the local activation of calcium-dependent pathways connected to gene transcription. This mode of synapse to nucleus communication has been intensively studied for excitatory synapses and calcium influx generated by activation of synaptic NMDA receptors, L-type calcium channels, or intracellular calcium release [17– 20]. The resulting action potentials generated at the initial axon segment may propagate retrogradely in the soma, and spread into dendrites [21]. Propagation of such signals is influenced by the geometry of the dendritic arbor as well as spatiotemporal integration of excitatory and inhibitory inputs in the system and back-propagating action potentials were suggested to complement and enhance synaptic calcium influx [22,23]. Many forms of long-term potentiation (LTP) require NMDA receptor (NMDAR) activation and postsynaptic calcium entry [24]. Interestingly, the calcium transients generated upon synaptic activation of NMDARs are spatially restricted in spines, with low diffusion into the Trends in Neurosciences xx (2014) 1–9

1

TINS-1111; No. of Pages 9

Review dendritic shaft [25,26]. However, although calcium signals are dampened by uptake and clearance systems, regenerative wave mechanisms may allow them to cover distances of hundreds of micrometers in a number of milliseconds, bridging the gap between the synapse and the nucleus. Thus, synaptic activation can mobilize calcium from internal stores like the ER [27–29], which extends from the nucleus into the distal processes in both axons and dendrites [30,31]. Regenerative calcium waves may be initiated by calcium signals arising from voltage-gated ion channels, NMDARs, or the activation of metabotropic glutamate receptors (mGluRs) [32], and propagated by calcium-induced calcium release mediated by the activation of inositol-1,4,5-triphosphate (IP3) and ryanodine receptors (RyRs) located on the ER [33]. Although waves represent a plausible mechanism to convey calcium signals to the nucleus, the source of nuclear calcium and the mechanism of its elevation by synaptic activity are still a matter of debate [7,34]. Of note, the nucleus can generate autonomous calcium transients and a very recent study suggests that BK-type calcium-sensitive potassium channels reside on the nuclear envelope and regulate the coupling between synaptic activity and gene expression [35]. BK channels are typically plasma membrane proteins that contain a large conserved C-terminal extension, which is involved in channel gating by intracellular ions or second-messenger systems [36]. Li et al. [35] demonstrated the presence of BK channels on the nuclear envelope of rodent hippocampal neurons. Pharmacological blockade of nuclear BK channels in both intact neurons and isolated nuclei induced calcium release from the nuclear envelope and cAMP response element-binding protein (CREB)-dependent transcription [35]. These results suggest that local signaling at the nuclear envelope could link synaptic activity to nuclear calcium transients. The rapid mechanisms described above have been a primary focus for most of the field in recent years. A considerably slower process that depends on the nuclear import of proteins released from synapses may be of particular importance in coupling local synaptic activity to specific gene expression programs and potentially more sustained and delayed gene transcription [11,37]. What is the evidence for such macromolecular signaling in synapse to nucleus communication? ERK signaling from synapse to nucleus requires macromolecular trafficking Although diverse signaling molecules have been proposed to physically translocate from synapse to soma [11], the functional significance of such transport events has been contentious. NMDARs play a pivotal role in regulation of activity-dependent gene expression and the multi-protein NMDAR complex is a potentially rich source for longdistance protein messengers [38–40]. The mitogen-activated protein kinase (MAPK) cascade plays an important role in transducing synaptic signals to the nucleus, and the MAPK extracellular signal-regulated kinase (ERK) regulates gene transcription in learning and memory [41]. ERK activation occurs both at the vicinity of the stimulated synapse [4,42] and at the nucleus [6]. Strikingly, two very recent studies have now combined advanced imaging with 2

Trends in Neurosciences xxx xxxx, Vol. xxx, No. x

biochemical approaches to provide convincing evidence of the need for transport of ERK signaling complexes in specific paradigms of synapse to nucleus communication [43,44]. Zhai et al. [43] used a fluorescent ERK activity reporter to examine spatial parameters of signaling under different parameters of LTP. They evaluated the effect of inducing LTP in single versus multiple dendritic spines of rat CA1 hippocampal pyramidal neurons on the propagation of ERK signaling to the nucleus. Stimulation of three or more spines generated a signal that propagated to the nucleus, if the stimulated spines were located in at least two distinct dendritic branches (Figure 1A). Moreover, Zhai et al. further showed that this mechanism allows both spatial integration of signals from multiple branches and temporal summation of spatially distinct signals separated by intervals of 30 min. The onset of the nuclear response was delayed by tens of minutes, in correlation with the distance of the stimulation, suggesting that propagation was via a relatively slow mechanism [43]. The effective diffusion range of ERK has been estimated to be approximately 30 mm in hippocampal dendrites [2] and, moreover, the signaling range might be additionally constrained by phosphatase activity in the cytoplasm [3]. How then might a phosphorylated signaling molecule such as ERK traverse distances of a few hundred microns in a phosphatase-rich environment from synapse to nucleus while maintaining its signaling capacity over periods of tens of minutes or more? Karpova et al. [44] characterized such a mechanism in the course of a study aimed at understanding how the nucleus can distinguish between signaling of synaptic versus extrasynaptic NMDARs. Synaptic and extrasynaptic NMDARs have distinct roles in synaptic plasticity, transcription, and cell death [45,46]. The transcriptional response to synaptic NMDAR activation is biased toward cell survival and plasticity genes, whereas extrasynaptic NMDAR signaling primarily induces expression of transcripts involved in cell death pathways [44,47]. ERK activity was found to be required for nuclear translocation of the protein Jacob after specific activation of synaptic NMDARs. Although activation of extrasynaptic NMDARs also induced accumulation of Jacob in the nucleus, ERK activity was not required in the latter case. A series of biochemical, proteomic, and cell biological analyses then showed that Jacob is an ERK substrate, and that synaptic, but not extrasynaptic, NMDAR activation leads to the phosphorylation of Serine180 in Jacob by ERK. Phosphorylated Jacob was protected from dephosphorylation en route to the nucleus by association with proteolytic cleavage fragments of the intermediate filament a-internexin, together with ERK (Figure 1B). The phosphorylation state of Jacob upon arrival in the nucleus then determines the transcriptional response and physiological outcome (Figure 1C), and it is tempting to speculate that Jacob operates as a mobile hub that docks NMDAR-derived signalosomes to CREB and potentially other nuclear target sites [44]. This transport mechanism is strikingly analogous to that previously described for protected transport of phosphorylated ERK from injured peripheral axons to the soma in association with proteolytic fragments of the intermediate filament

TINS-1111; No. of Pages 9

Review

Trends in Neurosciences xxx xxxx, Vol. xxx, No. x

(A) (i)

Single spine

(ii)

3–7 spines

(iii)

One vs mulple branches

Ca2+

Ras CaMKII

Raf Ca2+

MEK1/2

PKC p p

ERK1/2 Elevated nuclear ERK acvity

Elevated nuclear ERK acvity

Spaotemporal signal integraon over mulple dendric branches (B)

(C)

α-i p p n te rn ER Jacob p exin K1

Ca2+

To the nucleus

Synapc NMDAR

/2

(i) Mi

β cro

To

tub

the

α

Ca2+ ule

nu

Dynein

s

cle

Ras

Raf MEK1/2

us

ERK1/2

Extrasynapc NMDAR

CNS p en

(ii)

ER

n

β Mi

cro

To

PNS

tub

the

ule

cle

α-internexin p

2

S180

us

p p

ERK1/2

Jacob

ERK1/2

α

S180

α Dynein

s

nu

K1 /

Cell survival

p Vim

S180

Jacob α

p

Jacob α

CREB Nucleus Cell death TRENDS in Neurosciences

Figure 1. Spatial signaling by ERK and Jacob. (A) Nuclear ERK is activated by spatially distributed inputs. Stimulation of dendritic spines induces both ERK activation and propagation toward the nucleus, as shown by Zhai et al. [43]. Spatial characteristics of the input pattern influence nuclear ERK activation, since (i) stimulating a single synapse fails to generate the pERK signal, while the stimulation of (ii) three to seven spines distributed on two to seven branches of the same neuron induces a significant increase of pERK in the nucleus. Moreover, (iii) the stimulation of several spines on the same branch is prone to signal saturation, whereas two sets of stimulation applied to different branches separated by >30 mm induce increased nuclear response. (B) Preservation of phosphorylation signals during long-distance transport. Schematics describing the signaling complexes of (i) pJacob and pERK with the retrograde transport machinery in CNS neurons [44] and (ii) pERK with the retrograde transport machinery in peripheral sensory axons after injury [48]. In both models, intermediate filaments (a-internexin for CNS neurons and vimentin in PNS neurons after injury) were shown to conserve the phosphorylation status of ERK along its travels in a phosphatase-rich environment during synapse-to-nucleus and axon-to-nucleus transport. (C) The phosphorylation state of Jacob reports the source of its activation to the nucleus. As shown in [44], synaptic NMDAR activation by glutamate leads to the phosphorylation of ERK and its subsequent binding and activation of Jacob. Transport of pJacob to the nucleus induces plasticity-related gene transcription. By contrast, activation of extrasynaptic NMDARs recruits the CREB shut-off pathway, wherein a non-phosphorylated form of Jacob enters the nucleus to compete with the phosphorylation of CREB and thus promote cell death. Abbreviations: CAMKII, calcium/calmodulin-dependent protein kinase type II; CNS, central nervous system; CREB, cAMP response element-binding protein; ERK, extracellular signal-regulated kinase; MEK, mitogen-activated protein kinase (MAPK)/ERK kinase; NMDAR, NMDA receptor; PKC, protein kinase C; PNS, peripheral nervous system. a, importin a; b, importin b1; p, phosphorylation.

vimentin [48,49]. Moreover, a recent study showing that proline-rich synapse-associated protein 2 (ProSAP2)/SH3 and multiple ankyrin repeat domains 3 (Shank3) translocates to the nucleus in an activity-dependent manner [50] provides further support for the idea that mobile hubs coordinate larger signaling complexes in synapse to

nucleus transport. ProSAP2/Shank3 is a synaptic scaffolding protein that directly interacts with proteins such as Lapser1 [51] and Abelson interacting protein 1 (Abi-1) [52], which have also been shown to transit to the nucleus from synaptic sites and have roles in regulation of gene transcription. 3

TINS-1111; No. of Pages 9

Review The findings outlined above reveal how transport of ERK-containing signaling complexes allows the neuron to distinguish between localized versus distributed synaptic stimulation [43], or between synaptic versus extrasynaptic activation of the same receptor [44]. In addition to ERK and Jacob, a burgeoning list of signaling molecules has been implicated in synapse to nucleus communication [11,37,53]. Transcriptional targets have been identified for some of these macromolecular messengers, for example Arc/Arg3.1 and Bdnf for Jacob [44,54], c-fos, Arc, and Zif268 for CREB-regulated transcription coactivator 1 (CRTC1) [55], and Nap1, JunD, and HLA-DOB for nuclear factor-kB (NF-kB) [56]. Many of these candidate protein messengers, including axin interactor, dorsalization associated-1D (AIDA1D), CREB2, NF-kB, and Jacob itself, contain nuclear localization signals (NLS) that can be recognized by classical nuclear import factors from the importin a family [57]. Moreover, ERK itself can be imported to the nucleus via phosphorylation of a short three-amino acid motif that interacts with importin 7 [58]. These findings have motivated closer examination of possible roles of nuclear import factors in the cytoplasmic transport of protein messengers. Nuclear targeting of synaptic signals Early suggestions that nuclear import mechanisms might be implicated in long-distance retrograde signaling were based on the hypothesis of shared regulatory mechanisms for both neuronal plasticity and neuronal injury responses [59]. Direct evidence for long-distance transport of importin a with the dynein retrograde motor was first obtained in injured sensory axons [60] and then also shown in models of long-term synaptic plasticity [61]. It should be noted that at least six importin a genes are expressed in any given mammal, and multiple importin a gene products and isoforms are expressed in different cells [57,62,63]. Hence, for the purpose of clarity in this review, we will use the generic term importin a throughout, without specifying which individual isoforms are implicated in each case. Different modes of regulation of importin complex formation have been described in axonal injury versus synaptic signaling. In the axon injury model, a series of studies have demonstrated that calcium-dependent local translation of importin b1 is a critical trigger for retrograde complex formation (Figure 2A) [60,64,65]. In contrast, studies of the postsynaptic compartment demonstrated an association of importin a with the cytoplasmic NLS of the NR1-1a subunit of the NMDAR [66]. Activation of NMDARs in cultured hippocampal neurons or triggering of late-phase CA3–CA1 LTP in hippocampal slices releases importin a from NMDARs. Interestingly, this release is not observed for stimuli that trigger early-phase LTP [66]. Thus, activity regulation and phosphorylation of NMDARs can free importin a for cargo binding and transport in specific paradigms of synaptic plasticity (Figure 2B). In addition to the importins, NLS bearing cargo molecules are also activated and regulated by translational and post-translational mechanisms. Transcription factors must access the nucleus to fulfill their primary roles in 4

Trends in Neurosciences xxx xxxx, Vol. xxx, No. x

cells, and therefore have importin-binding capacity. For example, the RelA/p65 subunit of NF-kB is recruited to importins upon phosphorylation-mediated release of its inhibitor IkBa, following synaptic activation of NMDARs (Figure 2B) [67]. Intracellular transport of NF-kB can occur by either dynein-dependent transport or facilitated diffusion [68,69], and studies in different cell types have revealed that dynein-dependent transport is critical mainly for the longer distances of transport required in neurons [70,71]. Moreover, the NLS of RelA/p65 was shown to be essential for NF-kB association with the dynein motor complex after synaptic stimulation of hippocampal neurons [71]. Other transcription factors have been reported to translocate over distance in neuronal processes under diverse physiological paradigms. Early observations on the phosphorylation dynamics of signal transducer and activator of transcription 3 (STAT3) in axon versus soma led to the suggestion that it might act as a retrograde signaling factor [72]. A recent proteomic array approach revealed a battery of transcription factors associated with dynein in injured axons, one of which was STAT3 [73]. Activated STAT3 was then shown to arise from local translation and phosphorylation in the injured axon, and its transport with dynein and importins provides a retrograde injury signal to the soma in both sensory and hippocampal neurons [73–75]. Interestingly, STAT3 signaling was recently implicated in long-term depression (LTD), although nuclear translocation was not required for this function [76]. Indeed, STAT3 may exert local cytoplasmic effects on the cytoskeleton that are distinct from its transcriptional activity [77], thus it may have multiple roles in synapse to nucleus signaling at different subcellular junctions. Finally, the transcription factor CREB is one of the most studied components in neuronal plasticity pathways, and a wealth of studies have described how LTP-inducing stimuli trigger CREB phosphorylation and transcriptional activation [6,78]. Electrochemical mechanisms are thought to be the primary route for conveying synaptic events to CREBmediated transcription [7], and although CREB itself has been reported in both embryonic dendrites [79] and mature peripheral axons [80], evidence of a function for dendritically transported CREB in neuronal plasticity is unavailable to date. However, recent studies have revealed functionally important trafficking of CREB modulators and effectors from the synapse. The transcription repressor CREB2 was reported to translocate from synapse to nucleus under experimental paradigms of LTD, but not LTP [81]. An importin a is required for CREB2 transport from distal dendrites to the soma in both rodent hippocampal and Aplysia sensorimotor synapses [81]. Interestingly, testing of Jacob under similar NMDAR stimulation paradigms in hippocampal neurons revealed importin-dependent nuclear accumulation under LTP, but not LTD [82]. CRTC1 (also known as TORC) is another CREB pathway modulator recently reported to translocate from synapse to nucleus [55]. CRTC1 binds CREB to promote transcription, and synaptic activity regulates its transfer from synapse to nucleus in hippocampal neurons. Activity-dependent CRTC1 nuclear translocation is required for the complete transcriptional response of specific CRE-containing genes,

TINS-1111; No. of Pages 9

Review

Trends in Neurosciences xxx xxxx, Vol. xxx, No. x

Axon to nucleus

(A)

Nucleus (i)

Imp-β1 AAAAAA

RanBP1 AAAAAA

Dynein

(ii)

Signaling cargo α NLS

CAS α Imp- α

Lesion

(iii)

+

Imp-β1 Ca2+ β

AAAAAA

RanBP1 AAAAAA

Ran

+ –

+

α

RanBP1

Microtubule +

Ca2+

(B)

Synapse to nucleus

p p

α

Ca2+

CAMKII p50

PKC

Imporns +

α

α

ERK α

Jacob

p65

p

IκBα

H HAP1

CREB2

Dynein F-acn Microtubule

TRENDS in Neurosciences

Figure 2. Importin-dependent transport in neuronal processes. (A) Importin-dependent transport of signals from axon to nucleus after peripheral nerve injury. Under normal conditions (i) an importin a is found in axons in a complex with CAS and the retrograde molecular motor dynein, while importin b1 and RanBP1 are present predominantly as mRNA. Following lesion (ii), both RanBP1 and importin b1 are translated locally in a calcium-dependent manner. RanBP1 and other cofactors (not shown) free importin a for association with importin b1, thus generating high affinity binding sites for NLS-bearing signaling proteins. (iii) The complex is then retrogradely transported toward the nucleus by virtue of the interaction of importin a with dynein [10]. (B) Importins are implicated in activity-dependent synapse to nucleus transport in CNS neurons. Synaptic activity and NMDAR activation trigger calcium entry and the subsequent activation of protein kinases such as PKC and CAMKII. Jeffrey et al. [66] demonstrated that an importin a can be anchored at the synapse under basal activity via an interaction with certain splice isoforms of the NR1 subunit of the NMDAR. Phosphorylation of the latter by PKC elicits the release of importin a, which is then available to bind Jacob (itself released from an interaction with caldendrin upon synaptic calcium elevation) for retrograde transport to the soma [44]. Synaptic activation of NMDARs also leads to phosphorylation of the NF-kB inhibitor IkBa by CAMKII, thereby unmasking the NF-kB p50/RelA(p65) NLS for importin a binding and subsequent transport [67]. Abbreviations: CAMKII, calcium/calmodulin-dependent protein kinase type II; CAS, Crk-associated substrate; CREB2, cAMP response element-binding protein 2; ERK, extracellular signal-regulated kinase; HAP1, huntingtin-associated protein 1; Htt, huntingtin; NF-kB, nuclear factor-kB; NLS, nuclear localization signal; NMDAR, NMDA receptor; PKC, protein kinase C; RanBP1, Ran-specific binding protein 1. a, importin a; b, importin b1; IkBa, NF-kB inhibitor alpha.

and localized glutamate stimulation reveals that CRTC1 exit is specific to the stimulated dendrite [55]. These findings suggest that physical translocation of CREB repressors or modulators may provide spatial or quantitative information to the nucleus on the stimulated synapses. Multiple regulatory and transport pathways from synapse to nucleus The NMDAR complex, as outlined above, fulfills central roles in synaptic signaling and neuronal plasticity. A diverse array of molecules with dual synaptic and nuclear

functions have been shown to associate with the complex (Figure 3), however, the mechanisms of activation or release of these long-distance modulators are, as yet, poorly understood. It will be critical to determine the complete range of mechanisms that enable local NMDAR signaling to induce divergent nuclear responses. One interesting option of localized encoding for long-distance signaling is the utilization of local protein synthesis, as has been described for axonal injury responses [10]. Indeed, translational control of dendritic mRNAs is required for various forms of synaptic plasticity [83,84], and synaptic protein 5

TINS-1111; No. of Pages 9

Review

Trends in Neurosciences xxx xxxx, Vol. xxx, No. x

NMDAR acvaon

LTP

Ca2+

AIDA1 2

α

LTD

Ca2+

Ca

4

3

Abi-1

RelA/p65

2+

Calcineurin

ERK Jacob

CRTC1

CREB2 6 Abi-1

+

α

Myosin 5

Dynein

Nucleus

F-acn

+

1 +

H

Synapc cargo

α

7 Synapc cargo

β -

α

+

Microtubules

TRENDS in Neurosciences

Figure 3. Long-distance transport from synapse to nucleus. (1) Dynamic microtubules can transiently invade a subset of spine synapses following NMDAR activation in a strictly Ca2+-dependent manner. (2) Neuronal importins can associate with several synapto-nuclear messengers at synaptic sites but it is currently unclear how this interaction might be regulated. (3,4) Signaling downstream of NMDAR that eventually elicits nuclear transport might also encode that type of NMDAR activation. (3) Ca2+-dependent (Jacob and CRTC1) and independent (AIDA1, blue) dissociation of protein messengers from NMDAR has been reported and local phosphorylation via ERK in the case of Jacob or dephosphorylation via calcineurin in the case of CRTC1 is required to exit the synapse. (4) Interestingly, NMDAR-dependent LTP but not LTD induces nuclear translocation of Jacob, while the reverse occurs for CREB2. (5) For many proteins binding to a dynein motor, likely via association with an importin a, is a prerequisite for long-distance transport. Whether this transport requires microtubule invasion of dendritic spines is presently unclear. (6) Alternatively, proteins like Abi-1 might leave the synapse utilizing a myosin motor. (7) At present it is also unknown whether the macromolecular transport complex is assembled in spines or in the dendritic shaft below the synapse. Another interesting question is whether Htt might substitute for importin b1 in some cases. It has been reported that an association with Htt is essential for the dissociation of RelA/p65 from synaptic sites. Abbreviations: Abi-1, Abelson interacting protein 1; AIDA1, axin interactor, dorsalization associated 1; CREB2, cAMP response element-binding protein 2; CRTC1, CREB-regulated transcription coactivator 1; ERK, extracellular signal-regulated kinase; Htt, huntingtin; LTD, long-term depression; LTP, long-term potentiation; NMDAR, NMDA receptor.

transcripts traffic from nucleus to synapse in translationally silenced ribonucleoprotein granules [85,86]. This modality of nucleus-to-synapse communication sets the stage for local translation of mRNAs that are critical for synaptic plasticity, but it is not yet clear if such local translation events also generate synapse-to-nucleus messengers such as Jacob [87]. Regardless, local translation can induce longlasting changes in synapse activity [88], and ribosomal complexes are enriched at postsynaptic sites [89]. Local protein synthesis can be stimulated downstream of NMDARs, mGluR, and the neurotrophin receptor tyrosine kinase TrkB, supporting the involvement of dendritic translation of trafficked mRNAs in synaptic plasticity [90–92]. It will be interesting to examine how such mechanisms might be integrated at the NMDAR with Ca2+ 6

signals and kinase and phosphatase pathways that might also be activated by other membrane receptors (Figure 3). It is also still unknown whether the transport complex for nuclear trafficking is assembled at synaptic sites or in dendrites (Figure 3). As already outlined above, importin a can directly associate with certain NMDARs [66]. A direct transit from receptor association to transport complex formation might be possible when dynamic microtubules transiently invade a subset of spine synapses [93]. In contrast, certain protein messengers such as Abi-1 need actin-based motors for nuclear transport and their trafficking from synapse to nucleus is not necessarily microtubulebased [52,94]. Interestingly, microtubule entry to spines requires sustained NMDAR-dependent Ca2+-transients [95], and a connection between microtubule transport and NMDAR complexes might be facilitated by their independent interactions with neural cell adhesion molecule (NCAM) [96,97]. It will be interesting to test whether importin a already associates with a dynein motor for active retrograde transport along microtubules in dendritic spines, as is the case in axons (Figure 2). Finally, it will be interesting to find out whether signaling endosomes might also play a role in synapse to nucleus signaling, analogous to their well-established role in survival and maintenance signaling in the axonal compartment [98,99]. A recent study used microfluidics chambers to investigate dendrite to nucleus signaling in embryonic cortical neurons, demonstrating that dendritic application of BDNF elicited a transcriptional response in the soma [100]. Intriguingly the response was dependent on activity of the BDNF receptor TrkB in the soma, consistent with involvement of a signaling endosome [100]. BDNF-induced retrograde transport of TrkB in the dendrites of striatal neurons was reported to require Huntingtin (Htt), the protein mutated in Huntington’s disease [101]. Interestingly, Htt can also function as a molecular adaptor for nonvesicular cargoes [102], and a role for Htt in synapse to nucleus transport has also been proposed (Figure 3). It is not clear at this stage whether this might reflect a specific binding adaptor role of Htt in signaling complexes [101,102], or a more general role in energy supply to transport complexes [103]. From all of the above evidence, it is increasingly apparent that multiple mechanisms are implicated in synapse to nucleus trafficking events. Concluding remarks: How do neurons integrate different modalities of synapse to soma signaling? The recent progress described above raises the pressing question of how and to what end would a neuron integrate electrophysiological and motor-driven modalities of synapse to soma signaling? One intriguing possibility is that the motor-dependent signals provide a temporal component for information transfer from synapse to soma. The average velocities of motor-driven transport [104] are orders of magnitude lower than those of ionic waves or electrophysiological signals [105,106], thus a synaptic event transduced by both mechanisms will generate biphasic fast and slow signals that arrive in the soma at different times. There is precedent for such temporal encoding in transcription factor regulation, where fast [calcium/calmodulin-dependent protein kinase type IV

TINS-1111; No. of Pages 9

Review Box 1. Outstanding questions  How is the synaptic origin of protein messengers encoded? Is it by translational or post-translational mechanisms at the synapse itself, or by temporal dynamics of signal trafficking?  How are protein messengers released from the synapse?  What are the main transporters of protein messengers from the synapse to the nucleus? Are these importin complexes, kinase scaffolding molecules, or others?  Does dendritic trafficking of macromolecular complexes influence transport of axonal signals and vice versa? How do neurons integrate pre- and post-synapse communication with the nucleus?  How are ‘fast’ calcium signals integrated with ‘slow’ motordependent signals in synapse to nucleus communication and neuronal plasticity?  What are the functionally important transcriptional targets of macromolecular synapse-to-nucleus messengers? Are these the same transcriptional targets as those activated by calcium-based mechanisms?  Which aspects of neuronal plasticity definitively require macromolecular communication between the synapse and the nucleus? Does it serve an auxiliary role for other aspects of plasticity?

(CaMKIV)-dependent] and slow (Ras- or MAPK-dependent) pathways were shown to converge on CREB, with different downstream consequences [107,108]. Even if the ultimate molecular messenger in the soma is shared by both signaling mechanisms, as may be the case for ERK, the pulsatile nature of temporal encoding may relay information to the nucleus. The specificity of such signaling systems is dependent on noise levels in the intracellular milieu, and this might be a significant challenge for decoding pulses of ubiquitous signaling molecules such as ERK [109]. Indeed, computational modeling of the fidelity of information transfer in a system comprising a fast ionic wave signal and slower molecular motor trafficked signals suggested that multiple slow signals would be required to ensure robustness of information transfer [110]. Alternatively, frequency encoding is likely to be more robust and to convey more information than amplitude encoding for intracellular signaling, since information can be encoded both in the amplitude of the oscillations and in their frequency, as well as the potential for encoding multiple frequencies in a complex oscillating signal [111]. Both theoretical [112] and experimental analyses in non-neuronal cells [113–116] have shown that an oscillating signal can lead to a more consistent output than a non-oscillating signal, and frequency encoded signaling was recently proposed to function in axon growth and length control in neurons [117,118]. Regardless of whether the macromolecular signals in synapse to nucleus communication are amplitude or frequency encoded, it will be intriguing to determine which aspects of plasticity require long-distance protein messengers. New genetic models and methods to interrogate systems at the subcellular level will be key to answering these questions (Box 1). Acknowledgments Our research on these topics was supported by the European Research Council (Grant Agreement # 339495) and the Chaya Professorial Chair in Molecular Neuroscience (M.F.); the French Ministry of Foreign Affairs International Volunteers Program (N.P.); the Center for Behavioral Brain Sciences Magdeburg (A.K.); and the German Research Foundation, the Leibniz Foundation, the European Union Marie Curie International Training Network NPlast, and a Deutsch-Israelische Projektfo¨rderung grant (M.R.K.).

Trends in Neurosciences xxx xxxx, Vol. xxx, No. x

References 1 Neves, S.R. et al. (2008) Cell shape and negative links in regulatory motifs together control spatial information flow in signaling networks. Cell 133, 666–680 2 Wiegert, J.S. et al. (2007) Diffusion and not active transport underlies and limits ERK1/2 synapse-to-nucleus signaling in hippocampal neurons. J. Biol. Chem. 282, 29621–29633 3 Kholodenko, B.N. et al. (2010) Signalling ballet in space and time. Nat. Rev. Mol. Cell Biol. 11, 414–426 4 Harvey, C.D. et al. (2008) The spread of Ras activity triggered by activation of a single dendritic spine. Science 321, 136–140 5 Greer, P.L. and Greenberg, M.E. (2008) From synapse to nucleus: calcium-dependent gene transcription in the control of synapse development and function. Neuron 59, 846–860 6 West, A.E. and Greenberg, M.E. (2011) Neuronal activity-regulated gene transcription in synapse development and cognitive function. Cold Spring Harb. Perspect. Biol. 3, a005744 7 Bading, H. (2013) Nuclear calcium signalling in the regulation of brain function. Nat. Rev. Neurosci. 14, 593–608 8 Howe, C.L. and Mobley, W.C. (2005) Long-distance retrograde neurotrophic signaling. Curr. Opin. Neurobiol. 15, 40–48 9 Ibanez, C.F. (2007) Message in a bottle: long-range retrograde signaling in the nervous system. Trends Cell Biol. 17, 519–528 10 Rishal, I. and Fainzilber, M. (2014) Axon-soma communication in neuronal injury. Nat. Rev. Neurosci. 15, 32–42 11 Karpova, A. et al. (2012) Long-distance signaling from synapse to nucleus via protein messengers. Adv. Exp. Med. Biol. 970, 355–376 12 Adams, J.P. and Dudek, S.M. (2005) Late-phase long-term potentiation: getting to the nucleus. Nat. Rev. Neurosci. 6, 737–743 13 Bading, H. et al. (1993) Regulation of gene expression in hippocampal neurons by distinct calcium signaling pathways. Science 260, 181–186 14 Cohen, S. and Greenberg, M.E. (2008) Communication between the synapse and the nucleus in neuronal development, plasticity, and disease. Annu. Rev. Cell Dev. Biol. 24, 183–209 15 Saha, R.N. and Dudek, S.M. (2013) Splitting hares and tortoises: a classification of neuronal immediate early gene transcription based on poised RNA polymerase II. Neuroscience 247, 175–181 16 Saha, R.N. et al. (2011) Rapid activity-induced transcription of Arc and other IEGs relies on poised RNA polymerase II. Nat. Neurosci. 14, 848–856 17 Adams, J.P. et al. (2009) NMDA receptor-independent control of transcription factors and gene expression. Neuroreport 20, 1429–1433 18 Zhao, M. et al. (2005) Pattern-dependent role of NMDA receptors in action potential generation: consequences on extracellular signalregulated kinase activation. J. Neurosci. 25, 7032–7039 19 Dolmetsch, R.E. et al. (2001) Signaling to the nucleus by an L-type calcium channel-calmodulin complex through the MAP kinase pathway. Science 294, 333–339 20 Thiagarajan, T.C. et al. (2005) Adaptation to synaptic inactivity in hippocampal neurons. Neuron 47, 725–737 21 Waters, J. et al. (2005) Backpropagating action potentials in neurones: measurement, mechanisms and potential functions. Prog. Biophys. Mol. Biol. 87, 145–170 22 Koester, H.J. and Sakmann, B. (1998) Calcium dynamics in single spines during coincident pre- and postsynaptic activity depend on relative timing of back-propagating action potentials and subthreshold excitatory postsynaptic potentials. Proc. Natl. Acad. Sci. U.S.A. 95, 9596–9601 23 Nakamura, T. et al. (1999) Synergistic release of Ca2+ from IP3sensitive stores evoked by synaptic activation of mGluRs paired with backpropagating action potentials. Neuron 24, 727–737 24 Bliss, T.V. and Collingridge, G.L. (1993) A synaptic model of memory: long-term potentiation in the hippocampus. Nature 361, 31–39 25 Yuste, R. et al. (2000) From form to function: calcium compartmentalization in dendritic spines. Nat. Neurosci. 3, 653–659 26 Hagenston, A.M. and Bading, H. (2011) Calcium signaling in synapseto-nucleus communication. Cold Spring Harb. Perspect. Biol. 3, a004564 27 Finch, E.A. and Augustine, G.J. (1998) Local calcium signalling by inositol-1,4,5-trisphosphate in Purkinje cell dendrites. Nature 396, 753–756 28 Takechi, H. et al. (1998) A new class of synaptic response involving calcium release in dendritic spines. Nature 396, 757–760 7

TINS-1111; No. of Pages 9

Review 29 Emptage, N. et al. (1999) Single synaptic events evoke NMDA receptor-mediated release of calcium from internal stores in hippocampal dendritic spines. Neuron 22, 115–124 30 Merianda, T.T. et al. (2009) A functional equivalent of endoplasmic reticulum and Golgi in axons for secretion of locally synthesized proteins. Mol. Cell Neurosci. 40, 128–142 31 Berridge, M.J. (1998) Neuronal calcium signaling. Neuron 21, 13–26 32 Berridge, M.J. (2009) Inositol trisphosphate and calcium signalling mechanisms. Biochim. Biophys. Acta 1793, 933–940 33 Jaffe, D.B. and Brown, T.H. (1994) Metabotropic glutamate receptor activation induces calcium waves within hippocampal dendrites. J. Neurophysiol. 72, 471–474 34 Bootman, M.D. et al. (2009) An update on nuclear calcium signalling. J. Cell Sci. 122, 2337–2350 35 Li, B. et al. (2014) Nuclear BK channels regulate gene expression via the control of nuclear calcium signaling. Nat. Neurosci. 17, 1055–1063 36 Salkoff, L. et al. (2006) High-conductance potassium channels of the SLO family. Nat. Rev. Neurosci. 7, 921–931 37 Jordan, B.A. and Kreutz, M.R. (2009) Nucleocytoplasmic protein shuttling: the direct route in synapse-to-nucleus signaling. Trends Neurosci. 32, 392–401 38 Emes, R.D. et al. (2008) Evolutionary expansion and anatomical specialization of synapse proteome complexity. Nat. Neurosci. 11, 799–806 39 Husi, H. et al. (2000) Proteomic analysis of NMDA receptor-adhesion protein signaling complexes. Nat. Neurosci. 3, 661–669 40 Trinidad, J.C. et al. (2008) Quantitative analysis of synaptic phosphorylation and protein expression. Mol. Cell. Proteomics 7, 684–696 41 Carasatorre, M. and Ramirez-Amaya, V. (2013) Network, cellular, and molecular mechanisms underlying long-term memory formation. Curr. Top. Behav. Neurosci. 15, 73–115 42 Hardingham, G.E. et al. (2001) A calcium microdomain near NMDA receptors: on switch for ERK-dependent synapse-to-nucleus communication. Nat. Neurosci. 4, 565–566 43 Zhai, S. et al. (2013) Long-distance integration of nuclear ERK signaling triggered by activation of a few dendritic spines. Science 342, 1107–1111 44 Karpova, A. et al. (2013) Encoding and transducing the synaptic or extrasynaptic origin of NMDA receptor signals to the nucleus. Cell 152, 1119–1133 45 Hardingham, G.E. et al. (2002) Extrasynaptic NMDARs oppose synaptic NMDARs by triggering CREB shut-off and cell death pathways. Nat. Neurosci. 5, 405–414 46 Ivanov, A. et al. (2006) Opposing role of synaptic and extrasynaptic NMDA receptors in regulation of the extracellular signal-regulated kinases (ERK) activity in cultured rat hippocampal neurons. J. Physiol. 572, 789–798 47 Hardingham, G.E. and Bading, H. (2010) Synaptic versus extrasynaptic NMDA receptor signalling: implications for neurodegenerative disorders. Nat. Rev. Neurosci. 11, 682–696 48 Perlson, E. et al. (2005) Vimentin-dependent spatial translocation of an activated MAP kinase in injured nerve. Neuron 45, 715–726 49 Perlson, E. et al. (2006) Vimentin binding to phosphorylated Erk sterically hinders enzymatic dephosphorylation of the kinase. J. Mol. Biol. 364, 938–944 50 Grabrucker, S. et al. (2014) The PSD protein ProSAP2/Shank3 displays synapto-nuclear shuttling which is deregulated in a schizophrenia-associated mutation. Exp. Neurol. 253, 126–137 51 Schmeisser, M.J. et al. (2009) Synaptic cross-talk between N-methylD-aspartate receptors and LAPSER1-beta-catenin at excitatory synapses. J. Biol. Chem. 284, 29146–29157 52 Proepper, C. et al. (2007) Abelson interacting protein 1 (Abi-1) is essential for dendrite morphogenesis and synapse formation. EMBO J. 26, 1397–1409 53 Mosca, T.J. and Schwarz, T.L. (2010) The nuclear import of Frizzled2C by Importins-beta11 and alpha2 promotes postsynaptic development. Nat. Neurosci. 13, 935–943 54 Dieterich, D.C. et al. (2008) Caldendrin-Jacob: a protein liaison that couples NMDA receptor signalling to the nucleus. PLoS Biol. 6, e34 55 Ch’ng, T.H. et al. (2012) Activity-dependent transport of the transcriptional coactivator CRTC1 from synapse to nucleus. Cell 150, 207–221 8

Trends in Neurosciences xxx xxxx, Vol. xxx, No. x

56 Kaltschmidt, B. et al. (2006) NF-kappaB regulates spatial memory formation and synaptic plasticity through protein kinase A/CREB signaling. Mol. Cell. Biol. 26, 2936–2946 57 Perry, R.B. and Fainzilber, M. (2009) Nuclear transport factors in neuronal function. Semin. Cell Dev. Biol. 20, 600–606 58 Chuderland, D. et al. (2008) Identification and characterization of a general nuclear translocation signal in signaling proteins. Mol. Cell 31, 850–861 59 Ambron, R.T. and Walters, E.T. (1996) Priming events and retrograde injury signals. A new perspective on the cellular and molecular biology of nerve regeneration. Mol. Neurobiol. 13, 61–79 60 Hanz, S. et al. (2003) Axoplasmic importins enable retrograde injury signaling in lesioned nerve. Neuron 40, 1095–1104 61 Thompson, K.R. et al. (2004) Synapse to nucleus signaling during long-term synaptic plasticity: a role for the classical active nuclear import pathway. Neuron 44, 997–1009 62 Yasuhara, N. et al. (2009) The role of the nuclear transport system in cell differentiation. Semin. Cell Dev. Biol. 20, 590–599 63 Marfori, M. et al. (2011) Molecular basis for specificity of nuclear import and prediction of nuclear localization. Biochim. Biophys. Acta 1813, 1562–1577 64 Yudin, D. et al. (2008) Localized regulation of axonal RanGTPase controls retrograde injury signaling in peripheral nerve. Neuron 59, 241–252 65 Perry, R.B. et al. (2012) Subcellular knockout of importin beta1 perturbs axonal retrograde signaling. Neuron 75, 294–305 66 Jeffrey, R.A. et al. (2009) Activity-dependent anchoring of importin alpha at the synapse involves regulated binding to the cytoplasmic tail of the NR1-1a subunit of the NMDA receptor. J. Neurosci. 29, 15613–15620 67 Meffert, M.K. et al. (2003) NF-kappa B functions in synaptic signaling and behavior. Nat. Neurosci. 6, 1072–1078 68 Shrum, C.K. et al. (2009) Stimulated nuclear translocation of NFkappaB and shuttling differentially depend on dynein and the dynactin complex. Proc. Natl. Acad. Sci. U.S.A. 106, 2647–2652 69 Wellmann, H. et al. (2001) Retrograde transport of transcription factor NF-kappa B in living neurons. J. Biol. Chem. 276, 11821–11829 70 Mikenberg, I. et al. (2006) TNF-alpha mediated transport of NFkappaB to the nucleus is independent of the cytoskeleton-based transport system in non-neuronal cells. Eur. J. Cell Biol. 85, 529–536 71 Mikenberg, I. et al. (2007) Transcription factor NF-kappaB is transported to the nucleus via cytoplasmic dynein/dynactin motor complex in hippocampal neurons. PLoS ONE 2, e589 72 Lee, N. et al. (2004) STAT3 phosphorylation in injured axons before sensory and motor neuron nuclei: potential role for STAT3 as a retrograde signaling transcription factor. J. Comp. Neurol. 474, 535–545 73 Ben-Yaakov, K. et al. (2012) Axonal transcription factors signal retrogradely in lesioned peripheral nerve. EMBO J. 31, 1350–1363 74 Ohara, R. et al. (2011) Axotomy induces axonogenesis in hippocampal neurons through STAT3. Cell Death Dis. 2, e175 75 Shin, J.E. et al. (2012) Dual leucine zipper kinase is required for retrograde injury signaling and axonal regeneration. Neuron 74, 1015–1022 76 Nicolas, C.S. et al. (2012) The Jak/STAT pathway is involved in synaptic plasticity. Neuron 73, 374–390 77 Selvaraj, B.T. et al. (2012) Local axonal function of STAT3 rescues axon degeneration in the pmn model of motoneuron disease. J. Cell Biol. 199, 437–451 78 Lonze, B.E. and Ginty, D.D. (2002) Function and regulation of CREB family transcription factors in the nervous system. Neuron 35, 605– 623 79 Crino, P. et al. (1998) Presence and phosphorylation of transcription factors in developing dendrites. Proc. Natl. Acad. Sci. U.S.A. 95, 2313– 2318 80 Cox, L.J. et al. (2008) Intra-axonal translation and retrograde trafficking of CREB promotes neuronal survival. Nat. Cell Biol. 10, 149–159 81 Lai, K.O. et al. (2008) Importin-mediated retrograde transport of CREB2 from distal processes to the nucleus in neurons. Proc. Natl. Acad. Sci. U.S.A. 105, 17175–17180 82 Behnisch, T. et al. (2011) Nuclear translocation of jacob in hippocampal neurons after stimuli inducing long-term potentiation but not long-term depression. PLoS ONE 6, e17276

TINS-1111; No. of Pages 9

Review 83 Holt, C.E. and Schuman, E.M. (2013) The central dogma decentralized: new perspectives on RNA function and local translation in neurons. Neuron 80, 648–657 84 Swanger, S.A. and Bassell, G.J. (2011) Making and breaking synapses through local mRNA regulation. Curr. Opin. Genet. Dev. 21, 414–421 85 Speese, S.D. et al. (2012) Nuclear envelope budding enables large ribonucleoprotein particle export during synaptic Wnt signaling. Cell 149, 832–846 86 Fritzsche, R. et al. (2013) Interactome of two diverse RNA granules links mRNA localization to translational repression in neurons. Cell Rep. 5, 1749–1762 87 Kindler, S. et al. (2009) Dendritic mRNA targeting of Jacob and Nmethyl-D-aspartate-induced nuclear translocation after calpainmediated proteolysis. J. Biol. Chem. 284, 25431–25440 88 Kindler, S. and Kreienkamp, H.J. (2012) Dendritic mRNA targeting and translation. Adv. Exp. Med. Biol. 970, 285–305 89 Zalfa, F. et al. (2006) mRNPs, polysomes or granules: FMRP in neuronal protein synthesis. Curr. Opin. Neurobiol. 16, 265–269 90 Graber, T.E. et al. (2013) Reactivation of stalled polyribosomes in synaptic plasticity. Proc. Natl. Acad. Sci. U.S.A. 110, 16205–16210 91 Klann, E. and Dever, T.E. (2004) Biochemical mechanisms for translational regulation in synaptic plasticity. Nat. Rev. Neurosci. 5, 931–942 92 Buffington, S.A. et al. (2014) Translational control in synaptic plasticity and cognitive dysfunction. Annu. Rev. Neurosci. 37, 17–38 93 Jaworski, J. et al. (2009) Dynamic microtubules regulate dendritic spine morphology and synaptic plasticity. Neuron 61, 85–100 94 Jordan, B.A. et al. (2007) Activity-dependent AIDA-1 nuclear signaling regulates nucleolar numbers and protein synthesis in neurons. Nat. Neurosci. 10, 427–435 95 Merriam, E.B. et al. (2013) Synaptic regulation of microtubule dynamics in dendritic spines by calcium, F-actin, and drebrin. J. Neurosci. 33, 16471–16482 96 Sytnyk, V. et al. (2006) NCAM promotes assembly and activitydependent remodeling of the postsynaptic signaling complex. J. Cell Biol. 174, 1071–1085 97 Perlson, E. et al. (2013) Dynein interacts with the neural cell adhesion molecule (NCAM180) to tether dynamic microtubules and maintain synaptic density in cortical neurons. J. Biol. Chem. 288, 27812–27824 98 Cosker, K.E. and Segal, R.A. (2014) Neuronal signaling through endocytosis. Cold Spring Harb. Perspect. Biol. 6, a020669 99 Bronfman, F.C. et al. (2014) Spatiotemporal intracellular dynamics of neurotrophin and its receptors. Implications for neurotrophin signaling and neuronal function. Handb. Exp. Pharmacol. 220, 33–65

Trends in Neurosciences xxx xxxx, Vol. xxx, No. x

100 Cohen, M.S. et al. (2011) Neurotrophin-mediated dendrite-to-nucleus signaling revealed by microfluidic compartmentalization of dendrites. Proc. Natl. Acad. Sci. U.S.A. 108, 11246–11251 101 Liot, G. et al. (2013) Mutant Huntingtin alters retrograde transport of TrkB receptors in striatal dendrites. J. Neurosci. 33, 6298–6309 102 Marcora, E. and Kennedy, M.B. (2010) The Huntington’s disease mutation impairs Huntingtin’s role in the transport of NF-kappaB from the synapse to the nucleus. Hum. Mol. Genet. 19, 4373–4384 103 Zala, D. et al. (2013) Vesicular glycolysis provides on-board energy for fast axonal transport. Cell 152, 479–491 104 Schiavo, G. et al. (2013) Cytoplasmic dynein heavy chain: the servant of many masters. Trends Neurosci. 36, 641–651 105 Munoz-Garcia, J. and Kholodenko, B.N. (2010) Signalling over a distance: gradient patterns and phosphorylation waves within single cells. Biochem. Soc. Trans. 38, 1235–1241 106 Cho, Y. et al. (2013) Injury-induced HDAC5 nuclear export is essential for axon regeneration. Cell 155, 894–908 107 Finkbeiner, S. et al. (1997) CREB: a major mediator of neuronal neurotrophin responses. Neuron 19, 1031–1047 108 Wu, G.Y. et al. (2001) Activity-dependent CREB phosphorylation: convergence of a fast, sensitive calmodulin kinase pathway and a slow, less sensitive mitogen-activated protein kinase pathway. Proc. Natl. Acad. Sci. U.S.A. 98, 2808–2813 109 Voliotis, M. et al. (2014) Information transfer by leaky, heterogeneous, protein kinase signaling systems. Proc. Natl. Acad. Sci. U.S.A. 111, E326–E333 110 Kam, N. et al. (2009) Can molecular motors drive distance measurements in injured neurons? PLoS Comput. Biol. 5, e1000477 111 Cheong, R. and Levchenko, A. (2010) Oscillatory signaling processes: the how, the why and the where. Curr. Opin. Genet. Dev. 20, 665–669 112 Tostevin, F. et al. (2012) Reliability of frequency and amplitude decoding in gene regulation. Phys. Rev. Lett. 108, 108104 113 Yissachar, N. et al. (2013) Dynamic response diversity of NFAT isoforms in individual living cells. Mol. Cell 49, 322–330 114 Cai, L. et al. (2008) Frequency-modulated nuclear localization bursts coordinate gene regulation. Nature 455, 485–490 115 Albeck, J.G. et al. (2013) Frequency-modulated pulses of ERK activity transmit quantitative proliferation signals. Mol. Cell 49, 249–261 116 Aoki, K. et al. (2013) Stochastic ERK activation induced by noise and cell-to-cell propagation regulates cell density-dependent proliferation. Mol. Cell 52, 529–540 117 Albus, C.A. et al. (2013) Cell length sensing for neuronal growth control. Trends Cell Biol. 23, 305–310 118 Rishal, I. et al. (2012) A motor driven mechanism for cell length sensing. Cell Rep. 1, 608–616

9