Local translation in neurons: visualization and function Christine E. Holt1*, Kelsey C. Martin2*, Erin M. Schuman3* 1: Department of Physiology Development and Neuroscience, Anatomy Building, Downing Street, University of Cambridge, Cambridge CB2 3DY, UK. 2. Department of Biological Chemistry, Department of Psychiatry and Biobehavioral Sciences, David Geffen School of Medicine, University of California, Los Angeles, 885 Tiverton Dr., Los Angeles, CA, 90095, USA 3: Max Planck Institute for Brain Research, Max von Laue Strasse 4, 60438 Frankfurt am Main, Germany * to whom correspondence should be addressed: ceh33@cam.ac.uk, KCMartin@mednet.ucla.edu, or erin.schuman@brain.mpg.de Abstract. Neurons are amongst the most compartmentalized and interactive of all cell types. Like all cells, neurons use proteins as the main sensors and effectors. The modification of the proteome in axons and dendrites is used to guide the formation of synaptic connections and to store information. In this review, we discuss the data indicating that an important source of protein for dendrites, axons and their associated elements arises from the local synthesis of proteins. We review the data demonstrating the presence of the machinery required for protein synthesis, the direct visualization/demonstration of protein synthesis and the established functional roles for local translation for many different neuronal functions. Finally, we consider the open questions and future directions in this field. The function of a neuron is to receive, process and potentially store information from other neurons, and then transmit an integrated signal to other cells in its network. A typical excitatory neuron receives information from 1-10,000 other neurons and transmits information to 50-100,000 neurons; information transfer occurs at specialized junctions called synapses. This complex morphology requires unique solutions to maintain and modify the proteins that are required for the correct wiring of the nervous system during development and for its function and plasticity during adulthood. Neurons and numbers. When compared to other cell types in the body, neurons possess a unique morphology. Each neuron has a cell body from which emanates a single axon and multiple dendrites. While the cell body contains the nucleus, the axons and dendrites of a neuron can comprise up to 99% of the cytoplasm. Axons can be extremely long, extending over hundreds of centimeters in length in vertebrates. Dendrites are highly branched and also quite long; for example, the average dendritic length of hippocampal pyramidal neuron is 13.5 mm1 (Fig. 1). The contacts between neurons occur at specialized junctions called synapses. This unique cellular junction usually forms between a specialized ending (axon bouton or axon terminal) of a neuron transmitting the information (the presynaptic neuron) and special compartments within dendrites (often protrusions called “spines”) of a receiving neuron (the postsynaptic neuron). During development, neurons send out long axons that navigate over hundreds or even thousands of cell body lengths to reach their postsynaptic targets. This navigational feat is accomplished by “growth cones”, the specialized tips of growing axons. These growth cones detect and react rapidly to local guidance cues that steers their movement along the route and when they reach their destination, axons branch and form. A single axon that has a length of ~ 200 cm makes about ~50,000 synapses (CA3 region of the hippocampus2). Axonal boutons contain synaptic vesicles filled with neurotransmitter as well as scaffolding and regulatory molecules and organelles that organize, promote and regulate the release of neurotransmitter. A typical pyramidal neuron hosts about 54,000 synapses its dendritic tree (~4.0 synapses per m2 in the CA1 region of the hippocampus3). The synaptic compartments within dendrites, for example the dendritic spines that form excitatory synapses, contain a special complement of proteins that include neurotransmitter receptors, as well as scaffolding, regulatory and specialized adhesion molecules. The union of proteomic studies of biochemical preparations that enrich for synapses (e.g. “synaptosomes”) has identified over 2500 proteins that are present at synapses4. This does not mean that all 2500 proteins are present at a single synapse- the proteome of a single synapse has not been measured. Within an individual pre- or postsynaptic compartment, however, the copy numbers of some proteins have been determined. For example, most excitatory pre-synaptic proteins (e.g. scaffolding molecules, synaptic vesicle proteins and membrane-associated proteins) have been reported to exhibit between 100-10,000 copies per bouton5,6. We have much less quantitative information about the postsynaptic compartment. Although the relative abundances of ~30 postsynaptic proteins have been determined7, we have the rough copy number estimates per synapse for a relatively smaller number of proteins including CamKIIa (5000), the glutamate receptors (30-80 copies) and PSD-95 (300 copies); most other measured postsynaptic proteins exhibit copy numbers between 10- 10008 (Fig.1). In addition, like all proteins, neuronal proteins have a limited half-life. Neuronal proteins measured in vitro exhibit half-lives of ~ 5.5 days9,10 and in vivo ~10 days11. Interestingly, as a group, proteins in synapses and axonal growth cones exhibit a half-life that is shorter than the average population of cellular proteins9-12. The number, size and strength of synapses in the brain changes with experience through a process known as synaptic plasticity. Experience-dependent changes in connectivity provide a means of forming and storing memories. Long lasting forms of synaptic plasticity, like long-term memory, require new RNA and protein synthesis13,14. Stimulation of individual synapses or synapse clusters has demonstrated that the stimulated synapses can undergo long-lasting changes in synaptic strength, while neighboring unstimulated synapses remain unchanged15,16. Since long-term plasticity can occur at some but not all synapses made by a given neuron, mechanisms must also exist to compartmentalize the changes in gene expression to those synapses undergoing plasticity. Taken together, the morphological complexity of neurons, the volumetric dominance of the neuronal processes and their capacity for plasticity indicate that there are special challenges to set-up, maintain and modify the proteome in axonal and dendritic synaptic compartments. The fastest protein transport times (1µm/sec) measured (in axons) indicate that it would take 11.6 days to traffic a protein from the cell body to a distal (1 meter away) axon terminal17. The processing of local information and the ensuing response (e.g. turning in response to a navigational cue during development or synaptic plasticity in the adult) more often than not require responses on the time scale of minutes, rather than hours. As discussed below, the localization of mRNA and the on-site synthesis of proteins is a conserved mechanism that appears to have evolved to fulfill local demand for new proteins at short time scales and to provide subcellular functions for development, synaptogenesis, experience-dependent plasticity, and survival in dendrites and axons. As a framework for the review, we consider which criteria must be met in order to establish that protein synthesis occurs in a particular subcellular compartment to supply proteins and to allow for proteome remodeling: i) the localization of the protein template (mRNA), the machinery (ribosomes) and regulatory elements required for protein synthesis ii) the detection of newly synthesized proteins iii) the observation that blocking local synthesis reduces protein levels or affects plasticity. We conclude the review with a discussion of some of the open questions and issues that are particularly interesting and timely. The localization of protein synthesis machinery. mRNAs. The initial identification of mRNAs in distal dendrites was a fortuitous finding made in the course of in situ hybridization studies of a handful of transcripts, including MAP218 and CamKIIa19, and beta-actin mRNA in axons20. While the limited number of mRNAs detected in this manner at first suggested that only a select number of transcripts localized to synapses, the development of more sensitive fluorescent in situ hybridization (FISH) approaches and of modern genomic technologies subsequently allowed for more complete and unbiased detection of dendritically, axonally and synaptically localized transcripts (Fig. 2). Microarray analysis and/or qPCR of mRNAs present in isolated neuronal processes from Aplysia sensory neurons21, rat DRG neurons22, Xenopus or mouse retinal ganglion neurons23,24 or hippocampal neurons25-27 revealed the presence of up to hundreds of distally localized transcripts. In depth RNA sequencing combined with fluorescent in situ hybridization (FISH) and Nanostring of the stratum radiatum of rat hippocampus increased the resolution of detection to identify ~2500 mRNAs that localized to dendrites and axons of hippocampal pyramidal neurons28 (Fig. 2a). More recently, purification of excitatory synaptosomes (from vGlut1-GFP mouse) identified over 450 transcripts that are enriched in excitatory presynaptic nerve terminals, including many components of the active zone29 (Fig. 2e). mRNAs are trafficked to distal compartments in a microtubule-dependent manner20 and evidence indicates that the 3’ and 5’ untranslated (UTR) regions of mRNA play key roles in transport and localization selectivity30-34. However, despite the identification of multiple localization-required motifs35, no single motif has emerged as a canonical sequence for targeting mRNAs specifically to axons, or parts of axon, or dendrites. Small changes in UTR sequences can change the secondary structure of mRNAs that, in turn, alter the binding of RNA-binding proteins 36. Indeed, there is great diversity in the 3’UTRs of neuronal mRNAs: more than half of the mRNA population can express alternative 3’UTRs that give rise to differences in localization, half-life and presumably translational regulation, owing to the presence or absence of miRNA seed sequences33. Interestingly, plasticity changes 3’UTR choice34 and there is also data suggesting that 3’UTRs of mRNAs can be locally remodeled following plasticity33,37. Ribosomes. Ribosomes were first identified in the dendrites of primate spinal cord motoneurons38 and then observed in the dendrites of hippocampal neurons39. Ribosomes were detected in axons as early as 197040 in embryonic rabbit spinal cord and in vitro 41 but were not detected in adult CNS axons until recently29,42. Polysomes (clusters of mRNA and 2 or more ribosomes) have been detected in dendritic spines and throughout the length of the dendrite43. Recently ribosomes were observed in axon terminals using EM detection of a tagged knock-in ribosomal protein in adult mouse retinal ganglion cell axons42. One of the reasons why ribosomes eluded discovery in adult CNS axons for so long is that they are likely sparsely scattered, and are only rarely seen grouped into polysomes, for example, in the nodes of Ranvier44. Even in growing axons and in growth cones where their presence is less controversial, polysomes are not numerous and instead ribosomes appear to be commonly scattered within the cytoplasm as monosomes (single ribosomes engaged with an mRNA), sometimes lined up along the plasma membrane (Figure 2f). In yeast, translation of short proteins or proteins subject to non-sense mediated decay can occur on monosomes45. In small cytoplasmic compartments, like dendritic spines, axonal growth cones or boutons, where the supply of only a few new molecules of a protein could make a large difference functionally (see protein copy numbers above), it is possible that monosomal translation could be sufficient. Indeed, translation on monosomes could allow for synthesis for a greater number of different protein species with relatively limited ribosomal resources. Regulatory elements: co- and posttranslational processing machinery. Ribosomes, and co-translational chaperones are sufficient to synthesize, fold and traffic cytoplasmic proteins but additional machinery is required for integral membrane and secreted proteins. Membrane and secreted proteins are processed through multiple membrane-bound organelles (including the endoplasmic reticulum, the ER-Golgi intermediate compartment (ERGIC), the Golgi apparatus, the trans-Golgi network), where they are folded, assembled, and glycosylated before their delivery to the plasma membrane. While ER and several of its intrinsic proteins have has been detected in dendrites and axons, e.g. 46-49, a conventional Golgi apparatus is largely lacking50 and only sparse small Golgi outposts have been detected. Interestingly, many neuronal membrane proteins appear to exhibit glycosylation patterns consistent with Golgi bypass51. Given the abundance of membrane proteins in the dendritic and axonal transcriptomes, this suggests that membrane proteins could indeed be synthesized locally and might have a glycosylation signature that would indicate their site of synthesis. In line with this hypothesis, a dendritic ER-ERGIC-Golgi satellite (GS)-retromer microsecretory system has been described recently.52 Detection of local protein synthesis. Evidence for local protein synthesis in distal neuronal processes began with the demonstration of the metabolic incorporation of amino acids in axons in the 1960s53,54. Similar observations were made in a variety of biochemical fractions that were enriched for synaptic components (a.k.a. “synaptosomes” or “synaptoneurosomes”)55-57. These early experiments were often met with skepticism that pointed out a potential contribution of contaminating proteins synthesized in the cell body. Given these concerns, there are 4 basic approaches that have been taken to establish a local source of protein synthesis: i) physical isolation of the compartment of interest from the cell body, ii) use of photoconvertible or bleachable fluorescent reporters iii) high-resolution purification of the fraction or subcellular compartment, and iv) metabolic labelling with very short labeling periods, to eliminate the contributions of proteins made in the soma. The physical isolation of the compartment is perhaps the most solid proof for local translation, coupled with a demonstration that the observed protein synthesis signal can be blocked by a protein synthesis inhibitor. In some brain slices (e.g. from the hippocampus), for example, where the synaptic neuropil can be physically isolated from the neuronal cell bodies, local protein-synthesis dependent plasticity was first observed58,59. In Aplysia neurons, physically isolated neurites were treated with 25S- methionine and serotonin-induced protein synthesis was detected60. Physical isolation is easily accomplished in axons which can be surgically separated from their somas in vitro or in vivo61-63. For example, in isolated retinal ganglion growth cones, 3H-leucine incorporation was observed that was stimulated by different guidance molecules62. More recently, 1000-2000 mRNAs undergoing active translation in isolated embryonic and adult axons in vivo have been identified using an axon-specific Translating Ribosome Affinity Purification (TRAP) approach42,64. In addition, a highly sensitive proteomic approach combining Stable Isotope Labeling by Amino acids in Cell culture (pSILAC) with Single Pot Solid Phase-enhanced Sample Preparation (SP3) has revealed that hundreds of nascent proteins are rapidly synthesized in response to guidance molecules65,66. Fluorescent reporters which can be targeted to subcellular compartments and switched on or off or converted to different colors to define epochs of protein synthesis have been extremely useful. For example, expression of GFP-reporter in which the mRNA was targeted to the dendrites was followed by the cutting of a dendrite and stimulation of protein synthesis by a neurotrophin was observed67 (Fig. 3a). Stimulus-induced protein synthesis has also been visualized in the growth cones of retinal axons and at synapses of Aplysia neurites, isolated from their cell bodies, using translational reporters that employ photoconvertible fluorophores68,69. At cultured Aplysia sensory-motoneuron synapses the guidance cue, netrin, stimulated protein synthesis in sensory neuron neurites isolated from their cell bodies70 (Fig. 3). In addition, fluorescence recovery after photobleaching (FRAP) experiments have demonstrated the synthesis of beta-actin in branch tips in vivo71 (Fig. 3b). Continuous FRAP/image capture with a fast-folding fluorescent reporter enabled single molecule imaging of newly synthesized proteins in dendrites of hippocampal neurons and retinal growth cones72-74 and single molecule imaging of nascent peptides performed in conjunction with single molecule FISH to detect local mRNAs revealed ‘bursting’ translation in distal dendrites of hippocampal neurons75. An alternative approach makes use of new metabolic labeling approaches that enable brief labeling and/or the ability to mark or tag nascent proteins. For example, non- canonical amino acids (e.g. methionine analogs like azidohomoalanine; AHA) can be taken up by the cell, charged by the cell’s own tRNA synthetase, and incorporated in protein76. Using click-chemistry and fluorescent label, newly synthesized proteins have been detected in situ (fluorescence noncanonical amino acid tagging, FUNCAT; 77). Many demonstrations of local translation have made use of the antibiotic puromycin, which gets covalently associated with the nascent chain during protein synthesis78. Following very brief (~5-10 min) treatment with low concentrations of puromycin, protein synthesis can be visualized in situ by recognition with antibodies70,79,80. In brain tissue subjected to expansion microscopy to increase the spatial resolution, positive protein synthesis signal was detected in ~60% of dendritic spines and ~40% of excitatory and inhibitory nerve terminals following just 5 min of metabolic labeling with puromycin29. The combination of puromycin labeling with the proximity ligation assay (puro-PLA) has allowed the direct visualization of newly synthesized proteins of interest81. To date, several locally newly synthesized have been detected at synapses and in growing axons using puro-PLA, such as CamKIIa81,82, Bassoon29,81, SNAP2583, RapGEF29, Tau protein84, LB249 and VDAC249, Limk185, and several others65 (Table 1). How specific is localized translation? There is evidence for tight control over mRNA translation specificity in neuronal compartments, especially in signal-regulated translation. In growing axons, there appears to be a high degree of specificity to translation, with specific extrinsic cues triggering the synthesis of distinct proteins65. Hints of this specificity were first seen using a candidate approach where different guidance cues that exerted translation-dependent responses in turning growth cones, were found to induce the synthesis of different proteins. Netrin-1, an attractant, triggered the local synthesis of proteins that build-up the cytoskeleton (e.g. beta-actin; 69) whereas repellents Sema3A and Slit-2 induced synthesis of cytoskeletal disassembly factors86,87. Unbiased proteomic analyses has extended this differential translation model revealing that a single cue typically triggers the rapid and specific regulation of ~100 proteins, and that different cues induce distinct proteomic signatures65. Another likely important factor in the specificity of cue-induced translation is receptor activation. Different guidance cues or different neurotransmitters bind to and activate specific receptors, some of which are directly coupled to ribosomes88. In Aplysia neurons, local translation of the neuropeptide sensorin was induced by stimuli that produce translation-dependent long-term facilitation but not by stimuli that induce translation-dependent long-term depression of sensory- motor synapses55. In rodent neurons, a large-scale proteomic analysis of two opposite forms of homeostatic scaling revealed both commonly and differentially regulated newly synthesized proteins89,90. Another level of specificity may be the individual synaptic compartment in which protein synthesis is regulated by plasticity. For example, a recent study demonstrated that 3 different forms of plasticity (e.g. induced by brain-derived neurotrophic factor (BDNF), mGluR activation, or endocannabinoids) result in different signatures of protein synthesis in post-synaptic, excitatory or inhibitory presynaptic compartments29. Several mechanisms might underlie this specificity, such as microRNA regulation91, mRNA modification92, modulation of the phosphorylation of eukaryotic initiation factors93, and/or RNA-binding protein (RBP) phosphorylation94,95. This important topic here has been recently reviewed in more detail: 96,97. Established roles for local translation in protein supply and plasticity. The detection of mRNAs and translational machinery in axonal, dendritic and synaptic compartments and the visualization of local protein synthesis at synapses indicate that translation can occur at synapses, which in turn leads to questions about the physiological function of local translation in neural circuit development, survival and plasticity. An experimental challenge in demonstrating functions for local translation is the need to differentiate local translation from somatic translation with protein transport to synapses. For example, synaptic potentiation elicited by the neurotrophin BDNF requires local translation and this was demonstrated by recording from hippocampal slices in which the synaptic neuropil was physically isolated from the cell body layer58 (Fig. 3a). A similar approach demonstrated a requirement for local translation in mGluR-dependent long- term depression of synaptic transmission59 and late-phase long-term potentiation (LTP)98. In electron microscopy analyses, induction of long-term potentiation is associated with an increased number of polyribosomes detected in dendritic spines43. Restricted extracellular application of protein synthesis inhibitors to different synaptic layers has also been used to demonstrate that local protein synthesis is required for late-phase long-term potentiation99. Another form of plasticity, local homeostatic scaling, is mediated via regulation local protein synthesis. Acute blockade of synaptic activity at the synapse results in a stimulation of local translation100,101. At the Calyx of Held synapse, located ~3 mm from the cell body in mouse, a block of protein synthesis leads to an upregulation of spontaneous release events102. In addition, long-term depression of inhibitory transmission induced by endocannabinoids requires protein synthesis in the inhibitory presynaptic nerve terminal103. In the above experiments, protein synthesis inhibitors were injected, bath applied or perfused in a restricted manner. Clearly, a new generation of protein synthesis inhibitors which can be expressed in individual cells or enabled with light (e.g. 104,105) would enable better spatial and temporal control of protein synthesis inhibition and hence allow us to ascertain more readily roles for local translation. As noted above, CamKIIa is one of the most prominent dendritically localized mRNAs. Within 5 minutes of LTP induction, newly synthesized CamKIIa was detected in the dendrites106. To probe the function of local translation of dendritically localized CamKIIa mRNA in vivo, Miller et al.107 generated a mouse in which the CamKIIa 3’UTR that targeted the mRNA to dendrites was deleted, such that the protein was expressed but the mRNA was not dendritically localized. The phenotype included impairments in hippocampal-dependent forms of memory, reduced late-phase LTP, and a striking reduction in the amount of CamKIIa protein present in post-synaptic densities. Using a similar approach, Kuklin et al observed that loss of the CamKIIa 3’UTR in Drosophila showed reduced spontaneous neurotransmission and capacity for plasticity108. Together, these findings indicate that local translation of one dendritically-localized mRNA is required for functional and structural synaptic plasticity in vivo. Cultured Aplysia sensory-motor synapses provide a simple reduced preparation for investigating the function of local translation in synapse formation and plasticity. Reduction of a localized mRNA—without altering the protein levels—was found to block synapse formation109, indicating that synaptic translation of that single localized mRNA was required for synapse formation. Local perfusion of translational inhibitors to distal sensory neurites was found to inhibit serotonin-induced plasticity at the translationally- inhibited synapse but not at other stimulated synapses formed by that neuron, and sensory neurons severed from their cell bodies were capable of undergoing serotonin- induced, translation-dependent long-term facilitation that lasted 24 hours60. Together, these findings demonstrate a role for translation of localized mRNAs in both synapse formation and plasticity in this reduced, culture preparation. Local translation plays an important role during axonal development. When retinal ganglion cell axons grow towards their targets, local translation is required for them to respond appropriately (e.g. turn) to secreted guidance molecules62. Indeed, guidance molecule-driven translation of beta actin is asymmetrically required on the near-side of the growth cones (the side closest to the attractive guidance cue) to help growth cones turn in that direction69, as is the interaction with a specific RNA-binding protein that interacts with elements in the 3’UTR of the beta-actin mRNA in its transport to the near side110. In developing retinal axons in vivo, acute inhibition of protein synthesis or electroporation of beta-actin translation-blocking morpholinos in isolated axons (i.e. no somal contribution) impairs axonal branching in the optic tectum71. In adult sensory axons, Perry et al. (2012)111 took advantage of the finding that the mRNA encoding the nuclear transport factor importin beta 1 had two distinct 3’UTRs, including a long 3’UTR that localized the mRNA to axons of sensory neurons and a short 3’UTR that localized the mRNA to cell bodies of sensory neurons. They targeted the long 3’UTR to generate a mouse that lacked importin beta 1 mRNA in axons, without altering the expression or function of importin beta 1 in the sensory cell body. The phenotype of this mouse included delayed axonal loss in response to axonal injury, indicative of a function for local, axonal translation of importin beta 1 in initiating the response to injury. Indeed, there are multiple examples where local axonal translation is required for the axon to mount an injury response112,113. Another important role for local translation in axons is the promotion of axon maintenance and survival. The axonal translation of LaminB2 and Bclw, for example, has been shown to be required for axon viability63,114,115. Open questions and future directions We highlight the following areas as particularly interesting owing to recent progress, new ideas, or persistent questions that still remain unanswered. Ribosome repair/remodeling. Little is known about the origin and life of ribosomes in distant neuronal compartments. It is presumed that ribosomal proteins (RPs) synthesized in the cell body are assembled into ribosomes in the nucleolus and then transported to the cytoplasm and ultimately to distal neuronal processes but recent findings point to the possibility of on-site ribosome remodeling. Transcriptome studies have repeatedly shown the abundant presence of RP-encoding mRNAs in axons and dendrites21,23,25,28 raising the possibility that RP mRNAs are locally translated. A translatome analysis has confirmed this by showing that RPs constitute the most enriched functional category of mRNAs translated in mouse retinal axons in vivo42. A proteomic study has further shown that RPs are the most enriched functional group of basally translated proteins in cultured retinal axons and that extrinsic cues, such as Netrin, trigger the rapid regulation of newly synthesised RPs65. In line with this, a recent subcellular transcriptome/proteome analysis of mouse callosal axons identified the ribosome as one of the six most enriched categories116. Several RPs are known to serve extra-ribosomal functions but the sheer number of RP mRNAs detected (>50) in dendrites and axons suggests that at least some may serve a ribosomal function. The gold standard experimental approach to answering this question directly is to perform proteomic analysis on biochemically isolated ribosomes. However, this approach presents a major challenge in neuronal compartments because of the very small amounts of purified axon/dendrite material from ribosomes can be isolated. Recent analysis of this sort has been performed on isolated axons by growing hundreds of embryonic Xenopus eyes on filters in which the cell bodies reside on one side and the axons extend to the other64. SILAC labeling combined with axonal ribosome purification and mass spec analysis suggests that locally synthesized RPs may join pre-existing ribosomes in axons, in response to cue (Netrin-1) stimulation117. If new RPs are added to ribosomes on-site in dendrites/axons, what could be their role and how would it fit with classical views on ribosome biogenesis and function? In cultured neurons, ribosomal proteins exhibit remarkably variable half-lives that range from 3-9 days10, inconsistent with the entire ribosome being synthesized and degraded together. Little is known about neuronal ribosome biogenesis and the trafficking of ribosomes to dendrites and axons. It is conceivable that a mechanism exists for maintaining/prolonging ribosomal function in distant compartments. One possibility is that basally translated RPs may be needed for the constant, on-site homeostasis of ribosomes. A second possibility follows from compelling evidence showing that mutations in individual RPs give rise to distinct, sometimes tissue-specific, phenotypes rather than a general non-thriving phenotype expected from a global loss of protein synthesis has given rise to the ‘specialized ribosome’ hypothesis118. In line with this idea is the observation that distinct guidance cues induce a specific signature of RP translational changes in axons65, raising the possibility that localized RP synthesis could ‘tune’ ribosomes in a highly localized way in axons/dendrites, perhaps determining which mRNAs are translated. RNA methylation and partitioning of transcripts at synapses. The discovery and development of tools to study reversible chemical modifications of RNAs119,120 raised interest in the possibility that epitranscriptomic modifications might play a distinct role in regulating local translation at synapses. The most abundant mRNA modification in mammalian cells involves the addition of a methyl group to the N6 position of adenine to form N6-methyladenine (m6A). Methylation of RNA has been shown to affect many aspects of RNA metabolism, including stability, transport and translation121, and the level of m6A expression is higher in the brain than in any other organ in the body120. Using m6A immunoprecipitation to sequence the m6A epitranscriptome from purified mouse forebrain synaptosomes, Merkurjev et. al (2018)122 identified 4,469 m6A peaks in 2,921 genes that were selectively enriched in synaptosomes as compared to whole forebrain extracts. Over 40% of these mRNAs were not themselves enriched at synapses, suggesting either that the methyl modification was deposited at synapses or that methylation directed the synaptic localization of the transcript. These authors also showed that several of the enzymes involved in m6A modification (“m6A writers, erasers and readers”) are present at synapses, and that shRNA-mediated reduction of m6A readers resulted in immature spine morphology and in synaptic transmission deficits. Together, these findings indicate that chemical modification of mRNAs serves to partition specific mRNAs at synapses in a manner that could alter multiple steps of mRNA metabolism (e.g. stability, translation) to control the synaptic proteome. Intercellular RNA transfer via extracellular vesicles. The capacity of synaptic compartments to autonomously regulate their proteome via local translation decentralizes the regulation of gene expression from the nucleus to the synapse123. A series of intriguing and compelling studies over the past decade have revealed a role for extracellular vesicles in transporting mRNAs and miRNAs from one cell to another. Studies conducted in Drosophila melanogaster124 and in mouse125 revealed that the neuronal Arc protein assembles into viral-like capsids that contain the Arc mRNA, and that these viral-like capsids are transferred from one neuron to a target muscle (at the Drosophila neuromuscular junction) or to a target neuron (in cultured mouse cortical and hippocampal neurons). Moreover, the transferred Arc mRNA was shown to be translated in the recipient cell. The remarkable implication of these findings is that one cell can directly regulate the proteome of a neighboring cell via the local delivery of RNA. In the context of local translation at synapses, this is a particularly provocative idea because it suggests that the regulation of gene expression during the development and plasticity of neural circuits is not just regulated in a compartmentalized fashion within neurons, but that it is regulated at the level of local spatial neighborhoods comprising multiple neurons. Liquid-liquid phase separation and RNA localization. The regulated assembly of proteins and RNAs into membraneless organelles or compartments via liquid-liquid phase separation (LLPS, 126) has emerged as an important mechanism underlying RNA trafficking and metabolism in neurons. Many RNA binding proteins associated with neurodegenerative disease, including FUS, TDP-43, tau, TIA1 and hnRNPA1, have been found to undergo LLPS in a manner that contributes to pathology127. One consequence of increased gel formation in hyperphosporylated FUS mutants is the impairment of new protein synthesis in axon terminals128, and reducing gelation of mutant FUS by overexpression of the nuclear transport protein Transportin rescues protein synthesis in these axons129. Further elucidation of the biochemistry of this phase separation process promises to provide therapeutic insights into these diseases (e.g. 130). By dynamically compartmentalizing RNAs and RNA binding proteins, LLPS is also likely to play regulatory roles in synaptic RNA localization and translation. Concluding remarks The unique anatomy of neurons enables the vast information processing they conduct. However, it also poses huge challenges for how to manage protein supply and proteome remodeling at the temporal and spatial scales needed for axon guidance, branching and viability131, the basic supply of an integral synaptic protein107, the ongoing maintenance of synaptic function and the modification of individual proteins or proteomes required for synaptic plasticity. In each of these cases, the supporting data arise from studies both in vitro and in vivo, though in vivo visualizations of protein synthesis are relatively rare owing to the extreme difficulty of the experiments (e.g. 71,132). Therefore, it is clear that the ability to assess further the necessity of local translation will require new tools with enhanced spatial and temporal resolution. Figure Legends. Figure 1. Neurons and numbers. A. Anatomy of a neuron. Neuron-specific cell parts. The sites of local protein translation and neuronal functions dependent on local translation are indicated. B. Localization, functions and per synapse copy numbers of various synaptic proteins. Protein copy numbers taken from Wilhelm et al. 2014 and Sheng and Hoogenraad, 2007. Figure 2. The localization of the protein template (mRNA), the machinery (ribosomes) and regulatory elements required for protein synthesis. a, Scheme of a postsynaptic compartment highlighting some of the transcript families for synaptically-relevant proteins identified by RNA-seq that are present in the synaptic neuropil in area CA1 of the hippocampus. Reproduced with permission from Cajigas et al., 2012. b, Fluorescence in situ hybridization (FISH) signal in cultured rat hippocampal neurons (upper panels) or adult rat hippocampal slices (lower panels) using probes designed to detect the indicated mRNAs. Dendritic processes were immunostained with an anti-MAP2 antibody (purple). Scale bars represent 15 and 25 μm, for upper and lower panels, respectively. Reproduced with permission from Tushev et al., 2018. c, FISH for sensorin mRNA (red; left panel) in Aplysia sensory-motoneuron cultures showing the localization of the mRNA at presynaptic sites, labeled by VAMP-GFP (green; right panel). An Alexa dye (blue) represents the membrane. Scale bar = 50 um. d, the stem-loop structure present in the sensorin mRNA 5-UTR that is responsible for localization (Meers et al., 2012). e, Schematic representation of a vGLUT1+ presynaptic terminal with the localization of a subset of proteins coded by mRNA detected using Next Generation RNA sequencing of fluorescently-sorted synaptosomes. Note that many presynaptic active zone-related mRNAs are enriched in vGluT1+ presynaptic terminals relative to unsorted (generic) synaptosomes whereas synaptic vesicle-related mRNAs are either significantly depleted (magenta) or not enriched by sorting. Reproduced with permission from Hafner, Donlin-Asp, et al., 2019. f, HA immuno-gold electron microscopy (EM) to label ribosomes containing the Rpl22-HA tagged protein. HA-tagged ribosomes localize to and RGC axon terminals in the superior colliculus SC from embryonic day 18.5 mice (left) or adults (right). Scale bars = 500 nm. Reproduced with permission from Shigeoka et al., 2016. Figure 3. Visualization of local protein synthesis. a, Fluorescence reporter combined with physical isolation. A cultured hippocampal neuron was transfected with a GFP reporter containing the 3’UTR from CamKIIa for dendritic targeting of the mRNA. Images of a transfected neuron before (left) and 120 min after (right) BDNF treatment; arrow points to the region of transfection. The fluorescent signal in the transected dendrite increases following BDNF treatment indicating a stimulation of local protein synthesis. Scale bar = 15 um. Reproduced with permission from Aakalu et al., 2001. b, Fluorescence reporter combined with photobleaching (FRAP). Venus-beta-actin was electroporated into Xenopus retinal ganglion cells and fluorescence was monitored in axonal branches. Shown are Venus- beta-actin hotspots forming at different sub-compartments of an axonal branch. The Kymograph displays the FRAP after 300s, along the magenta arrow. At least 4 distinct translational hotspots can be identified in this single branch.Reproduced with permission from Wong et al., 2017. c and d, Fluorescence reporter combined with physical isolation. A reporter consisting of sensorin mRNA fused to the photoconvertible fluorescent protein dendra2 was expressed in cultured Aplysia sensory neurons (SN). c, The SN soma was removed (dotted circle), and 12–18 hr later photoconverted the dendra2 signal from green to red (right panel). d, Higher magnification panels show the green (top) and the the red channel (volume control, bottom). Increased green signal (yellow arrowheads) represents newly translated reporter after spaced application of 5-hydroxytryptamine. Reproduced with permission from Wang et al., 2009. e, A newly synthesized protein-of-interest demonstrating the utility of Puro-PLA, a metabolic labeling strategy that couples puromycin as a metabolic tag (recognized by an anti-puromycin antibody) together with an antibody against the protein-of-interest, followed by the proximity-ligation-assay (see tom Dieck et al., 2015). Shown is a cultured hippocampal neuron labelled for 15 min with puromycin; newly synthesized CamKIIa is detected. Scale bar = 25 um. Table 1. Locally synthesized individual proteins detected with puro-PLA. Table 2. Established roles for local translation in neural circuit development, survival and plasticity. References. 1. Ishizuka, N., Cowan, W.M. & Amaral, D.G. A quantitative analysis of the dendritic organization of pyramidal cells in the rat hippocampus. J Comp Neurol 362, 17-45 (1995). 2. Ishizuka, N., Weber, J. & Amaral, D.G. Organization of intrahippocampal projections originating from CA3 pyramidal cells in the rat. J Comp Neurol 295, 580-623 (1990). 3. Bourne, J.N. & Harris, K.M. Coordination of size and number of excitatory and inhibitory synapses results in a balanced structural plasticity along mature hippocampal CA1 dendrites during LTP. Hippocampus 21, 354-73 (2011). 4. Pielot, R. et al. SynProt: A Database for Proteins of Detergent-Resistant Synaptic Protein Preparations. Front Synaptic Neurosci 4, 1 (2012). 5. Richter, K.N. et al. Comparative synaptosome imaging: a semi-quantitative method to obtain copy numbers for synaptic and neuronal proteins. Sci Rep 8, 14838 (2018). 6. Wilhelm, B.G. et al. Composition of isolated synaptic boutons reveals the amounts of vesicle trafficking proteins. Science 344, 1023-8 (2014). 7. Cheng, D. et al. Relative and absolute quantification of postsynaptic density proteome isolated from rat forebrain and cerebellum. Mol Cell Proteomics 5, 1158-70 (2006). 8. Sheng, M. & Hoogenraad, C.C. The postsynaptic architecture of excitatory synapses: a more quantitative view. Annu Rev Biochem 76, 823-47 (2007). 9. Cohen, L.D. et al. Metabolic turnover of synaptic proteins: kinetics, interdependencies and implications for synaptic maintenance. PLoS One 8, e63191 (2013). 10. Dorrbaum, A.R., Kochen, L., Langer, J.D. & Schuman, E.M. Local and global influences on protein turnover in neurons and glia. Elife 7(2018). 11. Fornasiero, E.F. et al. Precisely measured protein lifetimes in the mouse brain reveal differences across tissues and subcellular fractions. Nat Commun 9, 4230 (2018). 12. Deglincerti, A. et al. Coupled local translation and degradation regulate growth cone collapse. Nat Commun 6, 6888 (2015). 13. Davis, H.P. & Squire, L.R. Protein synthesis and memory: a review. Psychol Bull 96, 518-59 (1984). 14. Sutton, M.A. & Schuman, E.M. Dendritic protein synthesis, synaptic plasticity, and memory. Cell 127, 49-58 (2006). 15. Andersen, P., Sundberg, S.H., Sveen, O. & Wigstrom, H. Specific long-lasting potentiation of synaptic transmission in hippocampal slices. Nature 266, 736-7 (1977). 16. Govindarajan, A., Israely, I., Huang, S.Y. & Tonegawa, S. The dendritic branch is the preferred integrative unit for protein synthesis-dependent LTP. Neuron 69, 132-46 (2011). 17. Maday, S., Twelvetrees, A.E., Moughamian, A.J. & Holzbaur, E.L. Axonal transport: cargo-specific mechanisms of motility and regulation. Neuron 84, 292-309 (2014). 18. Garner, C.C., Tucker, R.P. & Matus, A. Selective localization of messenger RNA for cytoskeletal protein MAP2 in dendrites. Nature 336, 674-7 (1988). 19. Burgin, K.E. et al. In situ hybridization histochemistry of Ca2+/calmodulin- dependent protein kinase in developing rat brain. J Neurosci 10, 1788-98 (1990). 20. Bassell, G.J. et al. Sorting of beta-actin mRNA and protein to neurites and growth cones in culture. J. Neurosci. 18, 251-265 (1998). 21. Moccia, R. et al. An unbiased cDNA library prepared from isolated Aplysia sensory neuron processes is enriched for cytoskeletal and translational mRNAs. J Neurosci 23, 9409-17 (2003). 22. Zheng, J.Q. et al. A functional role for intra-axonal protein synthesis during axonal regeneration from adult sensory neurons. J Neurosci 21, 9291-303 (2001). 23. Zivraj, K.H. et al. Subcellular profiling reveals distinct and developmentally regulated repertoire of growth cone mRNAs. J Neurosci 30, 15464-78 (2010). 24. Gumy, L.F. et al. Transcriptome analysis of embryonic and adult sensory axons reveals changes in mRNA repertoire localization. RNA 17, 85-98 (2011). 25. Poon, M.M., Choi, S.H., Jamieson, C.A., Geschwind, D.H. & Martin, K.C. Identification of process-localized mRNAs from cultured rodent hippocampal neurons. J Neurosci 26, 13390-9 (2006). 26. Zhong, J., Zhang, T. & Bloch, L.M. Dendritic mRNAs encode diversified functionalities in hippocampal pyramidal neurons. BMC Neurosci 7, 17 (2006). 27. Taylor, A.M. et al. Axonal mRNA in Uninjured and Regenerating Cortical Mammalian Axons. The Journal of Neuroscience 29, 4697-4707 (2009). 28. Cajigas, I.J. et al. The local transcriptome in the synaptic neuropil revealed by deep sequencing and high-resolution imaging. Neuron 74, 453-66 (2012). 29. Hafner, A.S., Donlin-Asp, P.G., Leitch, B., Herzog, E. & Schuman, E.M. Local protein synthesis is a ubiquitous feature of neuronal pre- and postsynaptic compartments. Science 364(2019). 30. Eom, T., Antar, L.N., Singer, R.H. & Bassell, G.J. Localization of a beta-actin messenger ribonucleoprotein complex with zipcode-binding protein modulates the density of dendritic filopodia and filopodial synapses. J Neurosci 23, 10433-44 (2003). 31. Andreassi, C. et al. An NGF-responsive element targets myo-inositol monophosphatase-1 mRNA to sympathetic neuron axons. Nat Neurosci 13, 291-301 (2010). 32. Taliaferro, J.M. et al. Distal Alternative Last Exons Localize mRNAs to Neural Projections. Mol Cell 61, 821-33 (2016). 33. Tushev, G. et al. Alternative 3' UTRs Modify the Localization, Regulatory Potential, Stability, and Plasticity of mRNAs in Neuronal Compartments. Neuron 98, 495-511 e6 (2018). 34. Fontes, M.M. et al. Activity-Dependent Regulation of Alternative Cleavage and Polyadenylation During Hippocampal Long-Term Potentiation. Sci Rep 7, 17377 (2017). 35. Kislauskis, E.H., Zhu, X. & Singer, R.H. Sequences responsible for intracellular localization of beta-actin messenger RNA also affect cell phenotype. J Cell Biol 127, 441-51 (1994). 36. Turner-Bridger, B. et al. Single-molecule analysis of endogenous beta-actin mRNA trafficking reveals a mechanism for compartmentalized mRNA localization in axons. Proc Natl Acad Sci U S A 115, E9697-E9706 (2018). 37. Andreassi, C. et al. 3′UTR cleavage of transcripts localized in axons of sympathetic neurons. bioRxiv, 170100 (2019). 38. Bodian, D. A Suggestive Relationship of Nerve Cell Rna with Specific Synaptic Sites. Proc Natl Acad Sci U S A 53, 418-25 (1965). 39. Steward, O. & Levy, W.B. Preferential localization of polyribosomes under the base of dendritic spines in granule cells of the dentate gyrus. J Neurosci 2, 284-91 (1982). 40. Tennyson, V.M. The fine structure of the axon and growth cone of the dorsal root neuroblast of the rabbit embryo. J Cell Biol 44, 62-79 (1970). 41. Bunge, M.B. Fine structure of nerve fibers and growth cones of isolated sympathetic neurons in culture. J Cell Biol 56, 713-35 (1973). 42. Shigeoka, T. et al. Dynamic Axonal Translation in Developing and Mature Visual Circuits. Cell 166, 181-92 (2016). 43. Ostroff, L.E., Fiala, J.C., Allwardt, B. & Harris, K.M. Polyribosomes redistribute from dendritic shafts into spines with enlarged synapses during LTP in developing rat hippocampal slices. Neuron 35, 535-45 (2002). 44. Walker, B.A. et al. Reprogramming axonal behavior by axon-specific viral transduction. Gene Ther 19, 947-55 (2012). 45. Heyer, E.E. & Moore, M.J. Redefining the Translational Status of 80S Monosomes. Cell 164, 757-69 (2016). 46. Gardiol, A., Racca, C. & Triller, A. Dendritic and postsynaptic protein synthetic machinery. J Neurosci 19, 168-79 (1999). 47. Horton, A.C. & Ehlers, M.D. Dual modes of endoplasmic reticulum-to-Golgi transport in dendrites revealed by live-cell imaging. J Neurosci 23, 6188-99 (2003). 48. Merianda, T.T. et al. A functional equivalent of endoplasmic reticulum and Golgi in axons for secretion of locally synthesized proteins. Mol Cell Neurosci 40, 128-42 (2009). 49. Cioni, J.M. et al. Late Endosomes Act as mRNA Translation Platforms and Sustain Mitochondria in Axons. Cell 176, 56-72 e15 (2019). 50. Hanus, C. & Ehlers, M.D. Secretory outposts for the local processing of membrane cargo in neuronal dendrites. Traffic 9, 1437-45 (2008). 51. Hanus, C. et al. Unconventional secretory processing diversifies neuronal ion channel properties. Elife 5(2016). 52. Mikhaylova, M., Bera, S., Kobler, O., Frischknecht, R. & Kreutz, M.R. A Dendritic Golgi Satellite between ERGIC and Retromer. Cell Rep 14, 189-99 (2016). 53. Koenig, E. Synthetic Mechanisms in the Axon. I. Local Axonal Synthesis of Acetylcholinesterase. J Neurochem 12, 343-55 (1965). 54. Giuditta, A., Dettbarn, W.D. & Brzin, M. Protein synthesis in the isolated giant axon of the squid. Proc Natl Acad Sci U S A 59, 1284-7 (1968). 55. Rao, A. & Steward, O. Evidence that protein constituents of postsynaptic membrane are locally synthesized: analysis of proteins synthesized within synaptosomes. Journal of Neuroscience 11, 2881-2895 (1991). 56. Weiler, I.J. & Greenough, W.T. Potassium ion stimulation triggers protein translation in synaptoneurosomal polyribosomes. Mol. Cell. Neurosci. 2, 305-314 (1991). 57. Torre, E.R. & Steward, O. Demonstration of local protein synthesis within dendrites using a new cell culture system that permits the isolation of living axons and dendrites from their cell bodies. J. Neurosci. 12, 762-772 (1992). 58. Kang, H. & Schuman, E.M. A requirement for local protein synthesis in neurotrophin-induced hippocampal synaptic plasticity. Science 273, 1402-6 (1996). 59. Huber, K.M., Kayser, M.S. & Bear, M.F. Role for rapid dendritic protein synthesis in hippocampal mGluR-dependent long-term depression. Science 288, 1254-1256 (2000). 60. Martin, K.C. et al. Synapse-specific, long-term facilitation of Aplysia sensory to motor synapses: a function for local protein synthesis in memory storage. Cell 91, 927- 938 (1997). 61. Eng, H., Lund, K. & Campenot, R.B. Synthesis of beta-tubulin, actin, and other proteins in axons of sympathetic neurons in compartmented cultures. J Neurosci 19, 1-9 (1999). 62. Campbell, D.S. & Holt, C.E. Chemotropic responses of retinal growth cones mediated by rapid local protein synthesis and degradation. Neuron 32, 1013-26 (2001). 63. Yoon, B.C. et al. Local translation of extranuclear lamin B promotes axon maintenance. Cell 148, 752-64 (2012). 64. Shigeoka, T., Jung, J., Holt, C.E. & Jung, H. Axon-TRAP-RiboTag: Affinity Purification of Translated mRNAs from Neuronal Axons in Mouse In Vivo. Methods Mol Biol 1649, 85-94 (2018). 65. Cagnetta, R., Frese, C.K., Shigeoka, T., Krijgsveld, J. & Holt, C.E. Rapid Cue- Specific Remodeling of the Nascent Axonal Proteome. Neuron 99, 29-46 e4 (2018). 66. Hughes, C.S., Sorensen, P.H. & Morin, G.B. A Standardized and Reproducible Proteomics Protocol for Bottom-Up Quantitative Analysis of Protein Samples Using SP3 and Mass Spectrometry. Methods Mol Biol 1959, 65-87 (2019). 67. Aakalu, G., Smith, W.B., Nguyen, N., Jiang, C. & Schuman, E.M. Dynamic visualization of local protein synthesis in hippocampal neurons. Neuron 30, 489-502 (2001). 68. Wang, D.O. et al. Synapse- and stimulus-specific local translation during long- term neuronal plasticity. Science 324, 1536-40 (2009). 69. Leung, K.M. et al. Asymmetrical beta-actin mRNA translation in growth cones mediates attractive turning to netrin-1. Nat Neurosci 9, 1247-56 (2006). 70. Kim, S. & Martin, K.C. Neuron-wide RNA transport combines with netrin- mediated local translation to spatially regulate the synaptic proteome. Elife 4(2015). 71. Wong, H.H. et al. RNA Docking and Local Translation Regulate Site-Specific Axon Remodeling In Vivo. Neuron 95, 852-868 e8 (2017). 72. Tatavarty, V. et al. Single-molecule imaging of translational output from individual RNA granules in neurons. Mol Biol Cell 23, 918-29 (2012). 73. Ifrim, M.F., Williams, K.R. & Bassell, G.J. Single-Molecule Imaging of PSD-95 mRNA Translation in Dendrites and Its Dysregulation in a Mouse Model of Fragile X Syndrome. J Neurosci 35, 7116-30 (2015). 74. Strohl, F. et al. Single Molecule Translation Imaging Visualizes the Dynamics of Local beta-Actin Synthesis in Retinal Axons. Sci Rep 7, 709 (2017). 75. Wu, B., Eliscovich, C., Yoon, Y.J. & Singer, R.H. Translation dynamics of single mRNAs in live cells and neurons. Science 352, 1430-5 (2016). 76. Dieterich, D.C., Link, A.J., Graumann, J., Tirrell, D.A. & Schuman, E.M. Selective identification of newly synthesized proteins in mammalian cells using bioorthogonal noncanonical amino acid tagging (BONCAT). Proc. Natl. Acad. Sci. 103, 9482-9487 (2006). 77. Dieterich, D.C. et al. In situ visualization and dynamics of newly synthesized proteins in rat hippocampal neurons. Nat Neurosci 13, 897-905 (2010). 78. Nathans, D. Puromycin Inhibition of Protein Synthesis: Incorporation of Puromycin into Peptide Chains. Proc Natl Acad Sci U S A 51, 585-92 (1964). 79. Smith, W.B., Starck, S.R., Roberts, R.W. & Schuman, E.M. Dopaminergic stimulation of local protein synthesis enhances surface expression of GluR1 and synaptic transmission in hippocampal neurons. Neuron 45, 765-79 (2005). 80. David, A. et al. Nuclear translation visualized by ribosome-bound nascent chain puromycylation. J Cell Biol 197, 45-57 (2012). 81. tom Dieck, S. et al. Direct visualization of newly synthesized target proteins in situ. Nat Methods 12, 411-4 (2015). 82. Sambandan, S. et al. Activity-dependent spatially localized miRNA maturation in neuronal dendrites. Science 355, 634-637 (2017). 83. Batista, A.F.R., Martinez, J.C. & Hengst, U. Intra-axonal Synthesis of SNAP25 Is Required for the Formation of Presynaptic Terminals. Cell Rep 20, 3085-3098 (2017). 84. Li, C. & Gotz, J. Somatodendritic accumulation of Tau in Alzheimer's disease is promoted by Fyn-mediated local protein translation. EMBO J 36, 3120-3138 (2017). 85. Rajgor, D., Sanderson, T.M., Amici, M., Collingridge, G.L. & Hanley, J.G. NMDAR-dependent Argonaute 2 phosphorylation regulates miRNA activity and dendritic spine plasticity. EMBO J 37(2018). 86. Wu, K.Y. et al. Local translation of RhoA regulates growth cone collapse. Nature 436, 1020-1024 (2005). 87. Piper, M. et al. Signaling mechanisms underlying Slit2-induced collapse of Xenopus retinal growth cones. Neuron 49, 215-28 (2006). 88. Tcherkezian, J., Brittis, P.A., Thomas, F., Roux, P.P. & Flanagan, J.G. Transmembrane receptor DCC associates with protein synthesis machinery and regulates translation. Cell 141, 632-44 (2010). 89. Schanzenbacher, C.T., Langer, J.D. & Schuman, E.M. Time- and polarity- dependent proteomic changes associated with homeostatic scaling at central synapses. Elife 7(2018). 90. Schanzenbacher, C.T., Sambandan, S., Langer, J.D. & Schuman, E.M. Nascent Proteome Remodeling following Homeostatic Scaling at Hippocampal Synapses. Neuron 92, 358-371 (2016). 91. Bellon, A. et al. miR-182 Regulates Slit2-Mediated Axon Guidance by Modulating the Local Translation of a Specific mRNA. Cell Rep 18, 1171-1186 (2017). 92. Yu, C.C. et al. Epigenetic Modulation on Tau Phosphorylation in Alzheimer's Disease. Neural Plast 2019, 6856327 (2019). 93. Cagnetta, R. et al. Noncanonical Modulation of the eIF2 Pathway Controls an Increase in Local Translation during Neural Wiring. Mol Cell 73, 474-489 e5 (2019). 94. Lepelletier, L. et al. Sonic Hedgehog Guides Axons via Zipcode Binding Protein 1-Mediated Local Translation. J Neurosci 37, 1685-1695 (2017). 95. Sasaki, Y. et al. Phosphorylation of zipcode binding protein 1 is required for brain-derived neurotrophic factor signaling of local beta-actin synthesis and growth cone turning. J Neurosci 30, 9349-58 (2010). 96. Richter, J.D., Bassell, G.J. & Klann, E. Dysregulation and restoration of translational homeostasis in fragile X syndrome. Nat Rev Neurosci 16, 595-605 (2015). 97. Gkogkas, C., Sonenberg, N. & Costa-Mattioli, M. Translational control mechanisms in long-lasting synaptic plasticity and memory. J Biol Chem 285, 31913-7 (2010). 98. Vickers, C.A., Dickson, K.S. & Wyllie, D.J. Induction and maintenance of late- phase long-term potentiation in isolated dendrites of rat hippocampal CA1 pyramidal neurones. J Physiol 568, 803-13 (2005). 99. Bradshaw, K.D., Emptage, N.J. & Bliss, T.V. A role for dendritic protein synthesis in hippocampal late LTP. Eur J Neurosci 18, 3150-2 (2003). 100. Sutton, M.A. et al. Miniature neurotransmission stabilizes synaptic function via tonic suppression of local dendritic protein synthesis. Cell 125, 785-799 (2006). 101. Sutton, M.A., Wall, N.R., Aakalu, G.N. & Schuman, E.M. Regulation of dendritic protein synthesis by miniature synaptic events. Science 304, 1979-83 (2004). 102. Scarnati, M.S., Kataria, R., Biswas, M. & Paradiso, K.G. Active presynaptic ribosomes in the mammalian brain, and altered transmitter release after protein synthesis inhibition. Elife 7(2018). 103. Younts, T.J. et al. Presynaptic Protein Synthesis Is Required for Long-Term Plasticity of GABA Release. Neuron 92, 479-492 (2016). 104. Goard, M. et al. Light-mediated inhibition of protein synthesis. Chem Biol 12, 685-93 (2005). 105. Elamri, I. et al. A New Photocaged Puromycin for an Efficient Labeling of Newly Translated Proteins in Living Neurons. Chembiochem 19, 2458-2464 (2018). 106. Ouyang, Y., Rosenstein, A., Kreiman, G., Schuman, E.M. & Kennedy, M.B. Tetanic stimulation leads to increased accumulation of Ca(2+)/calmodulin-dependent protein kinase II via dendritic protein synthesis in hippocampal neurons. J Neurosci 19, 7823-33 (1999). 107. Miller, S. et al. Disruption of dendritic translation of CaMKIIalpha impairs stabilization of synaptic plasticity and memory consolidation. Neuron 36, 507-19 (2002). 108. Kuklin, E.A. et al. The Long 3'UTR mRNA of CaMKII Is Essential for Translation- Dependent Plasticity of Spontaneous Release in Drosophila melanogaster. J Neurosci 37, 10554-10566 (2017). 109. Lyles, V., Zhao, Y. & Martin, K.C. Synapse formation and mRNA localization in cultured Aplysia neurons. Neuron 49, 349-56 (2006). 110. Yao, J., Sasaki, Y., Wen, Z., Bassell, G.J. & Zheng, J.Q. An essential role for beta-actin mRNA localization and translation in Ca2+-dependent growth cone guidance. Nat Neurosci 9, 1265-73 (2006). 111. Perry, R.B. et al. Subcellular knockout of importin beta1 perturbs axonal retrograde signaling. Neuron 75, 294-305 (2012). 112. Koley, S., Rozenbaum, M., Fainzilber, M. & Terenzio, M. Translating regeneration: Local protein synthesis in the neuronal injury response. Neurosci Res 139, 26-36 (2019). 113. Willis, D.E. & Twiss, J.L. The evolving roles of axonally synthesized proteins in regeneration. Curr Opin Neurobiol 16, 111-8 (2006). 114. Cosker, K.E., Fenstermacher, S.J., Pazyra-Murphy, M.F., Elliott, H.L. & Segal, R.A. The RNA-binding protein SFPQ orchestrates an RNA regulon to promote axon viability. Nat Neurosci 19, 690-696 (2016). 115. Hillefors, M., Gioio, A.E., Mameza, M.G. & Kaplan, B.B. Axon viability and mitochondrial function are dependent on local protein synthesis in sympathetic neurons. Cell Mol Neurobiol 27, 701-16 (2007). 116. Poulopoulos, A. et al. Subcellular transcriptomes and proteomes of developing axon projections in the cerebral cortex. Nature (2019). 117. Shigeoka, T. et al. On-site ribosome remodeling by locally synthesized ribosomal proteins in axons. bioRxiv, 500033 (2018). 118. Shi, Z. & Barna, M. Translating the genome in time and space: specialized ribosomes, RNA regulons, and RNA-binding proteins. Annu Rev Cell Dev Biol 31, 31-54 (2015). 119. Dominissini, D. et al. Topology of the human and mouse m6A RNA methylomes revealed by m6A-seq. Nature 485, 201-6 (2012). 120. Meyer, K.D. et al. Comprehensive analysis of mRNA methylation reveals enrichment in 3' UTRs and near stop codons. Cell 149, 1635-46 (2012). 121. Wang, Y. & Zhao, J.C. Update: Mechanisms Underlying N(6)-Methyladenosine Modification of Eukaryotic mRNA. Trends Genet 32, 763-773 (2016). 122. Merkurjev, D. et al. Synaptic N(6)-methyladenosine (m(6)A) epitranscriptome reveals functional partitioning of localized transcripts. Nat Neurosci 21, 1004-1014 (2018). 123. Holt, C.E. & Schuman, E.M. The central dogma decentralized: new perspectives on RNA function and local translation in neurons. Neuron 80, 648-57 (2013). 124. Ashley, J. et al. Retrovirus-like Gag Protein Arc1 Binds RNA and Traffics across Synaptic Boutons. Cell 172, 262-274 e11 (2018). 125. Pastuzyn, E.D. et al. The Neuronal Gene Arc Encodes a Repurposed Retrotransposon Gag Protein that Mediates Intercellular RNA Transfer. Cell 172, 275- 288 e18 (2018). 126. Hyman, A.A., Weber, C.A. & Julicher, F. Liquid-liquid phase separation in biology. Annu Rev Cell Dev Biol 30, 39-58 (2014). 127. Alberti, S. & Hyman, A.A. Are aberrant phase transitions a driver of cellular aging? Bioessays 38, 959-68 (2016). 128. Murakami, T. et al. ALS/FTD Mutation-Induced Phase Transition of FUS Liquid Droplets and Reversible Hydrogels into Irreversible Hydrogels Impairs RNP Granule Function. Neuron 88, 678-90 (2015). 129. Qamar, S. et al. FUS Phase Separation Is Modulated by a Molecular Chaperone and Methylation of Arginine Cation-pi Interactions. Cell 173, 720-734 e15 (2018). 130. Seidler, P.M. et al. Structure-based inhibitors of tau aggregation. Nat Chem 10, 170-176 (2018). 131. Cioni, J.M., Koppers, M. & Holt, C.E. Molecular control of local translation in axon development and maintenance. Curr Opin Neurobiol 51, 86-94 (2018). 132. Leung, L.C. et al. Coupling of NF-protocadherin signaling to axon guidance by cue-induced translation. Nat Neurosci 16, 166-73 (2013). 133. Seo, et al., DAP5 increases axonal outgrowth of hippocampal neurons by enhancing the cap-independent translation of DSCR1.4 mRNA Cell Death and Disease 10:49 (2019). 134. Mattioli et al., Alternative 3′ UTRs direct localization of functionally diverse protein isoforms in neuronal compartments Nucleic Acid Research 47, 2560-2573 (2019). 135. Jones et al., Rapid, experience-dependent translation of neurogranin enables memory encoding PNAS, 115 (25) (2018). Table 1 Newly synthesized proteins of interest detected with Puro-PLA Protein detected Species/ model system compartment method references CamK2a rat hippocampal neuron culture dendrite endogenous tom Dieck et al 2015 Sambandan et al 2017 (Ref 81, 82) Bassoon rat hippocampal neuron culture along dendrites VGluT1 positive presynaptic terminals endogenous tom Dieck et al 2015 Hafner et al 2019 (Ref 81, 29) RapGEF rat hippocampal neuron culture VGluT1 positive presynaptic terminals endogenous Hafner et al 2019 (Ref 29) Tau mouse hippocampal neuron culture dendrite endogenous Li & Götz 2017 (Ref 84) LaminB2 Xenopus retinal ganglion cell culture axon endogenous Cioni et al 2019 (Ref 49) VDAC2 Xenopus retinal ganglion cell culture axon endogenous Cioni et al 2019 (Ref 49) LimK1 rat cortical neuron culture dendrite endogenous Rajgor et al 2018 (Ref 85) DSCR1.4 mouse hippocampal neuron culture axon 5‘3‘-UTR- containing expression reporter Seo et al., 2019 (Ref 133) DAP5 mouse hippocampal neuron culture axon endogenous Seo et al., 2019 (Ref 133) CDC42 isoform mESC-derived neuron culture neurites CDS-E7- 3’UTR expression reporter Mattioli et al., 2019 (Ref 134) Neurogranin mouse cortical neuron culture dendrite endogenous Jones et al. 2018 (Ref 135) Calmodulin mouse cortical neuron culture dendrite endogenous Jones et al. 2018 (Ref 135) SNAP25 rat hippocampal neuron culture axon nascent presynapses endogenous Batista et al 2017 (Ref 83) Table 2: Established Roles for Local Translation in Protein Supply and Plasticity Function Species/ model system Compartment Method Reference BDNF-induced synaptic potentiation rat hippocampal slices Synaptic neuropil of area CA1 Physical isolation Kang and Schuman, 1996 (ref 58) Serotonin- induced synaptic facilitation cultured Aplysia sensory and motoneurons Sensory to motoneuron synapses Physical isolation Martin et al., 1997 (ref 60) mGluR1/5- induced long- term depression rat hippocampal slices Synaptic neuropil of area CA1 Physical isolation Huber et al., 2000 (ref 59) Guidance decisions Xenopus retinal ganglion cells in culture Axonal growth cones Physical isolation and protein synthesis inhibitors and translation inhibition of -actin Campbell & Holt, 2001 (ref 62); Leung et al, 2006.(ref 69) Maintenance of synaptic CamKIIa Mouse synapses Postsynaptic density of forebrain synapses Deletion of CamKIIa 3‘UTR Miller et al., 2002 (Ref 107) Late-phase long-term potentiation rat hippocampal slices Synaptic neuropil of area CA1 Restricted perfusion of a protein synthesis inhibitor Bradshaw et al., 2003 (Ref 99) Late-phase long-term potentiation rat hippocampal slices Synaptic neuropil of area CA1 Physical isolation Vickers et al., 2005 (Ref 98) Local homeostatic scaling Rat hippocampal neuron cultures dendrites Local perfusion of a protein synthesis inhibitor Sutton et al., 2006 (Ref 100) Synapse formation cultured Aplysia sensory and motoneurons Sensory to motoneuron synapses mRNA knockdown Lyles et al., 2006 (Ref 109) Injury response Mouse sensory neuron axons axons Deletion of long 3’UTR of importin beta 1 Perry et al., 2012 (Ref 111) Maintenance of presynaptic beta-catenin Cultured hippocampal neurons in microfluidic chambers Axon terminals Fluidically isolated application of a protein synthesis inhibitor Taylor et al., 2013 (Ref 27) Endocannabinoi d-induced plasticity of inhibitory transmission Rat hippocampal slices Inhibitory axon terminals Injection of a protein synthesis inhibitor Younts et al. 2016 (ref 103) Presynaptic protein clustering and neurotransmistt er release Cultured hippocampal neurons in microfluidic chambers Axon terminals Fluidically isolated application of a protein synthesis inhibitor Batista et al., 2017 (Ref 83) Spontaneous neurotrasmitter release Mouse Calyx of Held synapse Axon terminals Bath application of protein synthesis inhibitor Scarnati et al., 2018 (Ref 102) Axon branching Xenopus retinal neurons in vivo Axon terminals Translation inhibition of -actin in somaless axons Wong et al., 2017 (Ref 71) Axon maintenance Xenopus retinal neurons in vivo Axons Translation inhibition of LB2 in somaless axons Yoon et al., 2012 (Ref 63) Axon maintenance Rat dorsal root ganglion sensory neurons in vitro Axons Compartmentalized axon cultures and shRNA to Sfpq, LaminB2 or Bclw Cosker et al., 2016 (Ref 114) Axon maintenance Rat sympathetic superior cervical ganglion neurons in vitro Axons Compartmentalized axon cultures and protein synthesis inhibitors Hillefors et al., 2007 (Ref 115) Neurotransmitter receptors: 20-80 (e.g. GluA1, GluN1) Copy numbers in the post- and presynaptic compartment Scaffolding and signaling (PSD95: 300, SynGAP: 300, Shank: 150, CaMKII: 5000) ~2500 different proteins detected at synapses Scaffolding and signaling (piccolo: 100, basson: 440, Rim1: 40, CaMKII: 5000) Ion channels and adhesion (Ca2+ channels 200-400, cadherin 170, neurexins 40-100 ) Vesicle associated (synaptotagmin: 240, synapsin 60) b a Presynaptic terminals and axons Dendritic shafts and spines Protein supply to synapses 4 synapses per micron in dendrites in vivo; up to 10,000 receiving synapses and 100,000 output synapses per neuron Input specificity Growth cones Neuron Local synthesis Cable length: up to 10s of mm for dendrites up to 1 meter for axons Dendrites t0 t1 Axon Gria1,2,3 Grin1 Grin2a,b Grm5,7 Gripap1 Dlg1,2,3,4 Dlgap2,3,4 Homer1,2 Shank1,3 Mpp2,3 Lrrc7 Sipa1l1 Cnksr2 Psd, 3 Gabbr2 Gabra1,2,3,5 Gabrb1,2,3 Gabrg2,3 Pip4k2a,b,c Pip5k1a,c Cdipt Impa1 Inpp4b,5b,5j Itpr1 Prkcg Ppp1ca,cb,cc Ppp2ca,cb Ppp1r12b,15b,16a Ppp1r9a,9b Ppp2r1a,3c Ppm1b,e,g,k Prkce Actb Arc Map2 Map1a, b Arpc2,4,5 Calb1,2 Camk2a Camkk1, 2 Hpcal4 Ptk2b Gnaq Prkcg Prkacb Ryr2 Chrm1-5 Epha4,5,6 Htr2c Htr5b Slc12a5 Slc17a5,a7 Slc30a3,4,9 Slc4a10 Voltage-gated ion channels Presynaptic terminal Corresponding transcripts are: Enriched Depleted Signaling/ metabolism DelGEF RapGEF4 Aldolase C Fatty acid synthetase 1 Glutamine synthetase Adenylate cyclase 1 PTPRD Complexin-3 RIM1 RIM2 RIM3 KIF1A KIF1B KIF5C KIF16B Myo1C Spectrin β elF-2 elF-4E3 elF-5A elF-3A HLC-7 eEF-1A1 eEF-1B eEF-2 Translation Transport Stx6 Munc-13B SNAP25 Synaptotagmin-1 V-type ATPase* CLCN3 CSP Rab3A RIM-BP1 Liprin α1 Liprin α3 Bassoon Synaptophysin-1 Synaptogyr Synapsin-1 SCAMP1 SCAMP5 Active zone Catenin-α2 Catenin-β1 Dishevelled-1 FAM31A Phosphatase 2B catalitic subunit γ DAG kinase iota PSD and scaffolds Calcium signaling and kinases Other receptors Inhibitory receptors Glutamate receptors and interactors Transporters CytoskeletonPhosphatases Translation factors Phosphatidylinositol signaling Cacna1a,b,d,e,g,h,i Cacna2d3 Cacnb2,3 Cacng2,3,5,6,7 Kcna1,2,3,4,6 Kcnab1,2 Kcnma1 Scn2a1 Scn4b Eef1a1,b2,g,2 Eif3a,b,c,d,e,f Eif4a1,a2,b,e Eif4g1,2,3 Fmr1 Eif5,5a,5b Eftud1 Sensorin GFP-VAMP Synaptic vesicle CamKIIa Calm1 a b c d e f G C G G C C U A A U U G A G A A A C G C A U U A A U UA A C G U A U G C G A C U A A A C C A A A A G 5′ 3′ K Synapse Synapse Post Post Pre Pre L AdultE18.5 a SN-MN SN MN SN soma removal Photoconversion Pre-5X5HT De nd ra De nd ra A4 A5 A6 A7 Post-5X5HT 0 30 60 90 120 150 180 0” 30” 50” 100” Branch tip Tip (T) Shaft (S) Within branch Branch point 150” 300” 210 240 270 300 T S FRAP kymograph Ve nu s- β- ac tin cu m ula tiv e FR AP Proximal Distal Tim e (s) b c d e