InteractiveFly: GeneBrief

highwire: Biological Overview | Regulation | Developmental Biology | Effects of Mutation | Evolutionary Homologs | References

Gene name - highwire

Synonyms -

Cytological map position - 13A5--11

Function - signaling

Keywords - neuromuscular junction, CNS, PNS, E3 ubiquitin ligases

Symbol - hiw

FlyBase ID: FBgn0030600

Genetic map position - 1-49.8

Classification - B-box zinc finger superfamily

Cellular location - cytoplasmic

NCBI link: Entrez Gene
Recent literature
Mushtaq, Z., Choudhury, S. D., Gangwar, S. K., Orso, G. and Kumar, V. (2016). Human Senataxin modulates structural plasticity of the neuromuscular junction in Drosophila through a neuronally conserved TGFbeta signalling pathway. Neurodegener Dis 16: 324-336. PubMed ID: 27197982
Mutations in the human Senataxin (hSETX) gene have been shown to cause two forms of neurodegenerative disorders - a dominant form called amyotrophic lateral sclerosis type 4 (ALS4) and a recessive form called ataxia with oculomotor apraxia type 2 (AOA2). SETX is a putative DNA/RNA helicase involved in RNA metabolism. Although several dominant mutations linked with ALS4 have been identified in SETX, their contribution towards ALS4 pathophysiology is still elusive. In order to model ALS4 in Drosophila and to elucidate the morphological, physiological and signalling consequences, the wild-type and pathological forms of hSETX were overexpressed in Drosophila. The pan-neuronal expression of wild-type or mutant forms of hSETX induced morphological plasticity at neuromuscular junction (NMJ) synapses. Surprisingly, it was found that while the NMJ synapses were increased in number, the neuronal function was normal. Analysis of signalling pathways revealed that hSETX modulates the Highwire (Hiw; a conserved neuronal E3 ubiquitin ligase)-dependent bone morphogenetic protein/TGFbeta pathway. Thus, this study could pave the way for a better understanding of ALS4 progression by SETX through the regulation of neuronal E3 ubiquitin pathways.
Borgen, M., Rowland, K., Boerner, J., Lloyd, B., Khan, A. and Murphey, R. (2017). Axon termination, pruning, and synaptogenesis in the giant fiber system of Drosophila melanogaster is promoted by Highwire. Genetics [Epub ahead of print]. PubMed ID: 28100586
The ubiquitin ligase Highwire has a conserved role in synapse formation. This study shows that Highwire coordinates several facets of central synapse formation in the Drosophila melanogaster giant fiber system, including axon termination, axon pruning, and synaptic function. Despite the similarities to the fly neuromuscular junction, the role of Highwire and the underlying signaling pathways are distinct in the fly's giant fiber system. During development, branching of the giant fiber presynaptic terminal occurs and, normally, the transient branches are pruned away. However, in highwire mutants these ectopic branches persist, indicating that Highwire promotes axon pruning. highwire mutants also exhibit defects in synaptic function. Highwire promotes axon pruning and synaptic function cell-autonomously by attenuating a mitogen-activated protein kinase pathway including Wallenda, c-Jun N-terminal kinase/Basket, and the transcription factor Jun. A novel role for Highwire is shown in non-cell autonomous promotion of synaptic function from the midline glia. Highwire also regulates axon termination in the giant fibers, as highwire mutant axons exhibit severe overgrowth beyond the pruning defect. This excessive axon growth is increased by manipulating Fos expression in the cells surrounding the giant fiber terminal, suggesting Fos regulates a trans-synaptic signal that promotes giant fiber axon growth.

highwire (hiw) encodes a large protein (5233 amino acids) that controls synaptic growth at the larval neuromuscular junction (NMJ). The increase in synaptic structure in hiw mutants is dramatic: NMJ growth is exuberant. Hiw protein is localized at periactive zones which surround active zones at presynaptic terminals, the same region where Fasciclin II is localized. Hiw appears to function as a negative regulator of synaptic growth (Wan, 2000).

The most accessible synapse in Drosophila for a genetic approach to uncover mutants with changes in synaptic rate and extent of growth is the NMJ. In contrast to the fairly static NMJ in mammals, the Drosophila NMJ is a dynamic structure, growing new boutons and sprouting new branches during larval development. The Drosophila NMJ shares several important features with central excitatory synapses in the vertebrate brain. The Drosophila NMJ is a glutamatergic synapse and has ionotropic glutamate receptors homologous to those of humans. This synapse does not have a conspicuous basal lamina separating the two sides, as does the vertebrate NMJ, but rather appears by ultrastructural analysis to consist of close membrane apposition of pre- and post-synaptic cells. This synapse is organized into a series of boutons and branches that can be added or eliminated during development and plastic changes (Wan, 2000 and references therein).

As the Drosophila larva develops from first to third instar over a period of several days, the surface area of the postsynaptic muscle increases as much as 100-fold. During this developmental period, there is a concomitant growth of the presynaptic nerve terminal, resulting in up to a 10-fold increase in the number of boutons and a 10-fold increase in the number of active zones per bouton. A number of activity-dependent and activity-independent mechanisms appear to control the dramatic larval growth of this synapse. Long-term activity-dependent synaptic plasticity was first demonstrated at this synapse by genetic manipulations that alter neuronal activity and cAMP concentrations. K+ channel mutants, such as ether a go-go (eag) and Shaker (Sh), increase neuronal activity, and synaptic structure and branching. In addition, dunce (dnc) mutants, deficient in cAMP phosphodiesterase, increase synaptic structure. This activity-dependent synaptic growth is controlled in part by the regulation of the levels of the synaptic cell adhesion molecule (CAM) Fasciclin II (Fas II). Genetic analysis has shown that this downregulation of Fas II is both necessary and sufficient for structural plasticity at this synapse: FasII mutants that decrease Fas II levels by ~50% lead to sprouting, similar to eag and dnc, while transgenes that maintain synaptic Fas II levels suppress sprouting in eag, Sh and dnc (Wan, 2000).

A role for CAMs in the control of synaptic growth is not unique to Drosophila. In fact, CAMs of both the immunoglobulin and cadherin families are thought to play important roles in the regulation of synaptic growth in organisms as diverse as snails and mammals. Although CAMs have been implicated in the control of synaptic growth, little is known about the molecular machinery that carries out this growth process. This lack of knowledge led to a genetic approach to identify genes and genetic pathways controlling synaptic growth. Advantage was taken of a behavioral screen of the X chromosome conducted to detect walking mutants. A secondary anatomical screen was undertaken of the larval NMJ of this preselected collection of walking mutants. This screen led to the identification of the highwire gene. In hiw mutants, the specificity of motor axon pathfinding and synapse formation appears normal. However, NMJ synapses grow exuberantly and are greatly expanded in both the number of boutons and the extent and length of branches. These synapses appear normal ultrastructurally but have reduced quantal content physiologically. hiw encodes a large protein found at presynaptic terminals. Within presynaptic terminals, Hiw is localized to the periactive zone surrounding active zones; Fasciclin II (Fas II), which also controls synaptic growth, is found at the same location (Wan, 2000).

The hiw transcript encodes a protein that contains a RING-H2 finger, a domain identified in a large family of E3 ubiquitin ligases. Hiw, a potential synaptic E3 ligase, has been identified as a lethal enhancer of neuronal overexpression of fat facets, coding for a deubiquitinating protease. Overexpression of Fat facets leads to a profound disruption of synaptic growth control; there is a large increase in the number of synaptic boutons, an elaboration of the synaptic branching pattern, and a disruption of synaptic function. Genetic interactions between fat facets and highwire, a negative regulator of synaptic growth that has structural homology to a family of ubiquitin ligases, suggest that synaptic development may be controlled by the balance between positive and negative regulators of ubiquitination (Diantonio, 2001).

Phenotypic analysis of hiw has focused on the glutamatergic NMJ, but since all neurons appear to express both HIW mRNA and Hiw protein, it is reasonable to suggest that hiw functions at most, if not all, synapses. The original ethyl methanesulfonate (EMS) allele was isolated in a behavioral screen that selected for viable mutations. Subsequently 12 additional EMS alleles were isolated. All 13 EMS alleles and 3 P element alleles are viable, either alone or in trans over one another, or over either of two deficiencies that remove the gene. Several of these alleles are likely to be true nulls based on their phenotype over deficiencies and the presence of premature stop codons. Thus, the hiw mutant phenotype -- viable flies with dramatic synaptic overgrowth -- is likely to represent the true null phenotype (Wan, 2000).

In hiw mutants, motor axon pathfinding and target recognition appear normal. Synapses initially form at their normal locations on the appropriate muscles and induce appropriate postsynaptic glutamate receptor clustering. These synapses normally grow and sprout during larval development. But in hiw mutants, they are greatly expanded in both the number of boutons, and the extent and length of branches. The expanded synapses in hiw mutants have smaller-than-normal boutons but appear normal ultrastructurally in terms of active zones, T bars, clusters of synaptic vesicles, and folds of the subsynaptic reticulum. Occasionally, the overgrowth is so extensive that motor axons wander over to adjacent muscles which they would not innervate normally. This aspect of the hiw phenotype is similar to what is seen when the CAM Fas II is overexpressed on target muscles (Wan, 2000).

Physiologically, hiw synapses are functional, as is evident from the viable nature of the null mutants and the fact that they are capable of many motor behaviors. However, although greatly expanded, these synapses are not stronger, as one might have expected, but rather have a reduced quantal content. One possibility is that the reduction in quantal content is itself the primary defect and that this physiological defect leads to the increased growth in an attempt to compensate and return the strength back to normal. But this is unlikely, since mutant conditions that lead to an even greater reduction in quantal content (e.g., mutations in synaptic vesicle protein synaptotagmin) do not lead to presynaptic sprouting at this synapse. Thus, the overgrowth of the synapse appears to be a primary defect. The data suggest that the physiological defect is either secondary to the overgrowth or an independent primary defect (Wan, 2000).

Within presynaptic terminals, Hiw is localized to areas surrounding active zones: the periactive zones. Hiw is conspicuously absent from active zones. At the immuno-EM level, Hiw is associated with the plasma membrane and with vesicles near the membrane but outside the active zone. Fas II, a CAM that also controls synaptic growth, is found at the same location as Hiw at the periactive zones (Wan, 2000).

These results suggest that Hiw functions largely on the presynaptic side of the synapse. Certainly, at the NMJ, high levels of HIW mRNA are observed in motor neurons and are undetectable levels in muscles. Transgenic rescue experiments of this gene could not be performed due to the enormous size of the ORF and the complexity of the genomic region containing the hiw transcript. Nevertheless, a partial hiw cDNA transgene was used to generate a dominant-negative phenotype in which the NMJ overgrows, just as in hiw loss-of-function mutants. This dominant-negative phenotype is only observed when the Hiw fragment is expressed on the presynaptic rather than postsynaptic side of this synapse, further suggesting that Hiw protein normally functions on the presynaptic side (Wan, 2000).

rpm-1, the homologous gene in the nematode C. elegans, also seems to function cell autonomously in presynaptic neurons. The expression of RPM-1 on the presynaptic side of the synapse completely rescues function, and rpm-1 is required only in presynaptic neurons in genetic mosaic analysis (Schaefer, 2000; Zhen, 2000). Thus, taken together with the nematode data, it has been concluded that Hiw regulates synaptic growth by functioning on the presynaptic side of the synapse (Wan, 2000).

Phenotypic analysis was focused on the NMJ, where the branching pattern of individual presynaptic terminal arbors could be observed, the number of boutons could be counted, confocal microscopy could be used to image subcellular localization of components, and conventional and immuno-EM could be used to analyze identified synapses. These techniques are much more difficult to apply to identified terminal arbors and synapses in the CNS. Single cell probes that would allow this kind of quantitative analysis in the fruit fly CNS are not available (Wan, 2000).

Nevertheless, it is speculated that Hiw might function at many synapses, and not just the NMJ. (1) HIW mRNA is expressed by all CNS and PNS neurons. Interestingly, toward the end of embryogenesis, HIW mRNA is detected in the synaptic neuropil in the CNS, suggesting that HIW mRNA may be actively transported to near CNS synapses. HIW mRNA is not observed in motor or sensory axons in the periphery. HIW mRNA is not observed in the nerve roots in the CNS. Thus, it is concluded that HIW mRNA is localized in the areas where the synaptic neuropil forms. Whether HIW mRNA is expressed specifically at synapses awaits future investigation (Wan, 2000).

(2) Hiw protein, as detected with anti-Hie antibodies, is localized in the periphery at synaptic terminals of the NMJ, and in the CNS throughout the synaptic neuropil. Since no axonal staining is observed with the antibody, it is thought the staining is consistent with Hiw being localized at synapses (Wan, 2000).

(3) Although the phenotypic characterization of hiw mutants in Drosophila focuses on the glutamatergic NMJ, mutations in the homologous gene in the nematode give rise to altered synaptic structure at GABAergic synapses (Zhen, 2000) and glutamatergic and cholinergic synapses (Schaefer, 2000). Thus, Hiw protein functions at a variety of synapses in the nematode and at the glutamatergic NMJ in Drosophila, and it is expressed throughout the synaptic neuropil in Drosophila (Wan, 2000).

Why do hiw mutants have expanded synapses in Drosophila? One possibility is that synapses never properly form in hiw mutants and that the Hiw protein controls synapse formation. In this respect, the exuberant terminal motor axon branching across muscles observed in hiw mutants resembles the expanded nerve branching across muscles seen in agrin or MuSK knockout mice. But in these mutant mice, synapses do not properly form, and postsynaptic receptors do not properly cluster. These NMJs do not function, and thus these mouse mutants are lethal. In contrast, in hiw mutants, synapses form, active zones appear normal ultrastructurally, and postsynaptic receptors cluster. hiw mutant flies are viable. Thus, Hiw is not necessary for synapse formation per se (Wan, 2000).

A second and more likely possibility is that Hiw functions in a mechanism that regulates either synaptic stabilization, synaptic growth, or both. The major phenotype is synaptic overgrowth. It is difficult to distinguish between a model in which the primary function of Hiw is for synaptic stabilization versus one in which the primary function is for synaptic growth. Moreover, these two processes are likely to be intimately linked (Wan, 2000).

At the Drosophila NMJ synapse, the CAM Fas II controls synaptic stabilization. In its complete absence, the synapse retracts. But Fas II levels control synaptic growth, and these levels of Fas II are regulated by neural activity. A 50% decrease in Fas II leads to an increase in presynaptic sprouting and growth. These data suggest that while adhesion is required for synaptic stabilization, the modulation of the levels of adhesion regulate synaptic growth. Too much adhesion stabilizes the synapse and inhibits growth; a reduction of adhesion destabilizes the synapse and permits growth. It is at this interface between synaptic stabilization and the regulation of synaptic growth that Hiw appears to function (Wan, 2000 and references therein).

Mutations in other genes have been identified that lead to an increase in synaptic sprouting and growth at the Drosophila NMJ, including K+ channel mutants, such as eag and Shaker, and mutants in the cAMP phosphodiesterase dunce. Although these genetic conditions lead to synaptic sprouting, none of them are as dramatic as is hiw. No other known mutation has such a tremendous overgrowth phenotype at this synapse as does hiw. Taken together, all of these considerations have lead to the conclusion that Hiw normally functions to control synaptic stabilization and growth (Wan, 2000).

In certain respects, Hiw protein is reminiscent of two other non-Drosophila large presynaptic proteins, Bassoon and Piccolo. Piccolo is identical to a previously described protein, Aczonin. Both Bassoon and Piccolo are expressed presynaptically, both are cytoplasmic, both are enormous (>420 kDa), both contain zinc fingers (of the kind associated with protein-protein interactions), and both contain coiled-coil domains. But these two proteins also have striking differences from Hiw. Whereas Bassoon and Piccolo are expressed at presynaptic active zones around the clustered synaptic vesicles and appear to function as part of a presynaptic cytomatrix involved in the movement and clustering of synaptic vesicles, Hiw is expressed at periactive zones. In addition, there do not appear to be orthologs of Bassoon and Piccolo in the fly genome. Thus, Hiw appears to define a new function and, perhaps, a new kind of presynaptic cytomatrix that is localized to the periactive zone and functions in the control of synaptic stabilization and growth (Wan, 2000).

On the presynaptic side of a synapse, most studies have focused on the active zone and its components. The active zone contains the vesicle docking and release machinery, the clustered synaptic vesicles, the machinery to move and restrain vesicles, and the modulators and regulators of transmitter release. Synaptic function is associated with the active zone. But the active zone does not contain all functions associated with the synapse (Wan, 2000).

Evidence is presented supporting the idea that the zone outside of the active zone -- the periactive zone -- is an important area that contains the regulatory machinery for synaptic stabilization and growth. The CAM Fas II and the guanine nucleotide exchange factor (GEF) SIF are both localized in the periactive zone. Hiw protein is also localized to the periactive zone and the localization of Fas II has been confirmed. Double labeling experiments with Hiw and Fas II could not be performed because both antibodies were generated in mice. Nevertheless, it is believe that these two proteins are colocalized to the periactive zone based on their similar localization patterns, each complementary to that of Pak. Both Fas II and Hiw regulate synaptic growth. It is proposed that this region that surrounds active zones contains the machinery for a distinct set of functions, including the long-term regulation of synaptic growth and sprouting (Wan, 2000).

It is interesting to note that the structural domains predicted by the primary sequence of Hiw have been shown to mediate protein-protein interactions as well as to regulate signaling pathways. In RCC1, the seven tandem repeats form a seven-bladed propeller structure, as determined by X-ray crystallography, and are important for RCC1's function as a GEF for the nuclear small GTP binding protein Ran. A similar RCC1-like domain in a human protein p532 (HERC1) has also been demonstrated to stimulate guanine nucleotide exchange on other GTPases, such as ARF1, Rab3A, and Rab5 (see Drosophila Rab5). A second RCC1-like domain in p532 binds to clathrin and forms a cytosolic ternary complex with clathrin and Hsp70. The RCC1-like domain in Hiw might act as a GEF for an unidentified GTPase in a signaling pathway; alternatively, it might serve as a protein-protein interaction domain (Wan, 2000).

The region of the last 430 amino acids at the C terminus of Hiw is cysteine rich and is predicted to contain multiple zinc fingers, two of which can be categorized as B-box and RING finger motifs. Both the RING and the B-box motifs are zinc binding domains shown to participate in the formation of protein complexes. Recent studies also suggest a new function for the RING finger motif as an E2-dependent ubiquitin ligase. These structural motifs in Hiw may mediate protein-protein interactions and/or regulate levels of proteins via the ubiquitin pathway at periactive zones to control synaptic growth (Wan, 2000).

A region in Hiw has been found that behaves in a dominant-negative manner when expressed panneurally. It is interesting that this region does not contain either the RCC1-like domain or the zinc finger cluster. Instead, it contains the coiled-coil domain and the putative Myc binding domain. It is possible that this region in Hiw is involved in interactions between Hiw and other proteins and that the overexpression of this region interferes with such interaction and thus leads to a dominant-negative phenotype (Wan, 2000).

The hiw gene is highly conserved across phyla. There appears to be one gene in the fruit fly Drosophila (hiw), one complete gene in the nematode C. elegans (rpm-1; there is a gene duplication in C. elegans, producing a second minigene next to rpm-1 that expresses only the first 2000 amino acids of RPM-1), and one known gene in human (PAM). The function of Hiw at the synapse is also conserved between the fruit fly and the worm (Wan, 2000).

hiw mutations in the fly lead to synaptic sprouting and overgrowth, increasing the number of boutons and the complexity of presynaptic branches. No defect is seen in axon pathfinding, target selection, or the initial stages of synapse formation. rpm-1 mutations in the nematode (Zhen, 2000) lead to variable phenotypes in different types of NMJ synapses and, most noticeably, a larger presynaptic terminal with multiple active zones. EM analysis shows that the majority of the mutant GABAergic NMJs contain more than two individual presynaptic active zones within the same presynaptic terminal, while a smaller portion have fewer synaptic vesicles and instead are filled with electron-dense debris-like material. Mutant cholinergic NMJs appear to have a single larger active zone with increased number and density of synaptic vesicles. The overall axon morphology of the DD, VD, and other ventral cord motor neurons appears normal, and cholinergic synaptic transmission also appears unaffected. However, certain pre- and post-synaptic components (synaptotagmin SNT-1 and GABA receptor UNC-49) appear to be mislocalized in the mutant. In contrast, rpm-1 mutant mechanosensory neurons fail to accumulate synaptic vesicles, retract certain synaptic branches, and extend ectopic processes in the neuropil (Schaefer, 2000; Wan, 2000).

Thus, some aspects of the hiw mutant phenotype are consistent with the rpm-1 mutant phenotypes, while other aspects of the phenotypes are different. Synapses in both the nematode and the fruit fly contain presynaptic active zones with similar components. But the anatomy of synaptic contacts in the two organisms is quite distinct. In the nematode, synapses tend to be en passant contacts of one axon with another, with synaptic contacts being made along the axon shaft. In contrast, in the fruit fly, axons arborize into presynaptic terminals, and synaptic contacts are organized into boutons along the branches of the terminal arbor. At the NMJ, these terminal arbors and synaptic boutons undergo a dramatic growth process as boutons expand, new boutons are added, and new branches sprout and grow. In many respects, this synaptic organization in the fruit fly is more reminiscent of synapses in vertebrates. A dramatic overgrowth of the terminal arbor and increase in the number of boutons is observed in the Drosophila hiw mutant, while the rpm-1 mutant reveals defects in the spacing and organization of active zones (and more occasionally, retraction or expansion of synaptic branches). These differences may reflect the increased complexity of synaptic structure in the fruit fly compared with the nematode (Wan, 2000).

Finally, how can the difference between what is reported for Hiw in Drosophila be reconciled with the initial observation of the expression and potential function of Pam, the Hiw homolog in human? Pam was discovered as a putative Myc binding protein that is expressed in the nucleus of tissue culture cells (Guo, 1998). No evidence has been found for nuclear localization of Hiw in Drosophila. Moreover, Hiw lacks part of the region that has been identified as the Myc binding domain. No evidence was found for genetic interactions in Drosophila between hiw and dmyc; moreover, dmyc mutants do not show a synaptic phenotype in Drosophila. At present, nothing is known about Pam function in mammals. It is also possible that additional Hiw homologs may exist in mammals. Thus, whether Hiw homologs in vertebrates function only as presynaptic regulators of synaptic stabilization and growth or also have taken on additional functions remains to be determined (Wan, 2000).

Highwire regulates guidance of sister axons in the Drosophila mushroom body

Axons often form synaptic contacts with multiple targets by extending branches along different paths. PHR (Pam/Highwire/RPM-1) family ubiquitin ligases are important regulators of axon development, with roles in axon outgrowth, target selection, and synapse formation. This study reports the function of Highwire, the Drosophila member of the PHR family, in promoting the segregation of sister axons during mushroom body (MB) formation. Loss of highwire results in abnormal development of the axonal lobes in the MB, leading to thinned and shortened lobes. The highwire defect is attributable to guidance errors after axon branching, in which sister axons that should target different lobes instead extend together into the same lobe. The highwire mutant MB displays elevation in the level of the MAPKKK Wallenda/DLK (dual leucine zipper kinase), a previously identified substrate of Highwire, and genetic suppression studies show that Wallenda/DLK is required for the highwire MB phenotype. The highwire lobe defect is limited to α/β lobe axons, but transgenic expression of highwire in the pioneering α'/β' neurons rescues the phenotype. Mosaic analysis further shows that α/β axons of highwire mutant clones develop normally, demonstrating a non-cell-autonomous role of Highwire for axon guidance. Genetic interaction studies suggest that Highwire and Plexin A signals may interact to regulate normal morphogenesis of α/β axons (Shin, 2011).

In Drosophila, highwire is best studied for its role in restraining synaptic terminal growth at the NMJ. Studies in the fly neuromuscular system did not, however, find a role for highwire in motoneuron axon guidance. The current study demonstrates that highwire does regulate axon guidance of MB neurons in Drosophila. The gross morphological defects, such as the short α lobe and thinning of either the α or β lobe, present in the highwire MB lobes of the adult could be attributable to defects in either the development or maintenance of axons. However, similar defects were observed in both the developing and adult MB, so the phenotype is not attributable to degeneration of previously formed axons. The defect is also inconsistent with a gross alteration in axon outgrowth or guidance. The α/β axons form, path-find appropriately through the peduncle, and branch at the appropriate location. Instead, the data suggest a selective deficit in responding to guidance cues at this choice point. After bifurcation of the axon, sister branches do not segregate into distinct lobes as in WT but rather travel together into the same lobe. This phenotype is consistent with loss of homotypic repulsion of sister branches and/or the inability to respond to selective guidance cues targeting the axons to particular lobes (Shin, 2011).

In both fly and worm, PHR proteins sculpt synaptic terminals by restraining Wnd/DLK MAPKKK activity. In Drosophila, highwire acts as an ubiquitin ligase to limit the abundance of Wnd/DLK. Excess Wnd/DLK protein overactivates a MAP kinase signaling pathway that promotes synaptic terminal overgrowth. This study demonstrates that highwire-dependent downregulation of Wnd/DLK is also required for segregation of sister branches of α/β axons and, hence, proper MB development. In the absence of highwire, levels of Wnd/DLK are elevated in the axons of the developing MB. Furthermore, genetic deletion of Wnd/DLK suppresses the highwire-dependent phenotypes, demonstrating that Wnd/DLK is required for the aberrant behavior of α/β axons in the highwire mutant. Attempted were made to test whether overexpression of Wnd/DLK phenocopies highwire mutant MB by driving Wnd/DLK transgenic expression with OK107-Gal4 or MB subset Gal4 lines, including α'/β'-specific NP2748-Gal4. However, the strong overexpression of Wnd/DLK resulted in either lethality or massive cell death in the MB, probably because of the excess activation of downstream JNK MAPK signaling (Shin, 2011).

Although the relationship between the PHR ubiquitin ligase and DLK kinase is clear in flies and worms, studies in vertebrate systems paint a murkier picture. Analysis of PHR mutants in mice and zebrafish consistently demonstrate an important role in various aspects of axon development. However, the molecular mechanism of PHR action and the potential involvement of DLK in vertebrate axons is less clear. In cultured sensory axons from the Phr1 mutant magellan, axon morphology is disrupted and DLK protein is mislocalized. In addition, pharmacological inhibition of p38, a MAP kinase that can be downstream of DLK, reduced the size of the abnormally large growth cones present in these mutant axons. These findings are consistent with a role for DLK activity in generating the Phr1-dependent phenotypes. However, in an independently generated Phr1 mutant, no gross change was observed in DLK levels and it was found that genetic deletion of DLK failed to suppress either corticothalamic axon guidance defects or motoneuron sprouting defects. In zebrafish, mutations in the PHR ortholog esrom disrupt axon guidance and lead to an increase in JNK activation, consistent with a role for DLK, but inhibition of neither JNK nor p38 can suppress the esrom phenotypes, arguing against a functional role for DLK (Hendricks, 2009). The finding that Wnd/DLK is essential for axonal phenotypes in Drosophila while it is dispensable for at least some axonal phenotypes in mice and fish suggests that there is no simple relationship between PHR targets and the cellular function of PHR proteins. PHR proteins do interact with a number of other proteins besides DLK, and so the mechanism of PHR-dependent axonal phenotypes is likely context dependent (Shin, 2011).

During MB development, the α/β neurons are the last to be born and the last to extend their axons into the MB lobes. These α/β axons follow the path established by the earlier-born γ and α'/β' axons. In the highwire mutant, the α/β axons form short, thin, or absent α/β lobes, whereas the γ and α'/β' axons form morphologically normal lobes. Although such results would be consistent with a unique requirement for highwire in α/β neurons, a series of findings instead indicate that highwire is required in α'/β' neurons and indirectly affects the development of α/β axons via a non-cell-autonomous mechanism. First, in the highwire mutant, Wnd/DLK levels are elevated in γ, α'/β', and α/β axons, demonstrating that the Highwire ligase is likely targeting Wnd/DLK in all three cell types. Second, in the highwire mutant, the sister branches of the α/β axons fail to segregate but instead travel into the same lobe. However, within a brain hemisphere, most of the sister branches choose the same lobe, resulting in either a thickened α or β lobe. Hence, the decision as to which lobe to enter is apparently not determined independently by each axon. Third, expression of highwire in the earlier-born α'/β' neurons is sufficient to rescue the defects in the α/β lobes. Fourth, in single-cell highwire α/β clones, sister axons segregate normally in an otherwise heterozygous background. Together, these data demonstrate a non-cell-autonomous requirement for highwire (Shin, 2011).

How might α'/β' axons affect the guidance decision of α/β axons? Misexpression of the cell adhesion molecule Fas II in the α'/β' neurons leads to the loss of either α or β projections, demonstrating that inter-axonal interactions can affect α/β axon development and suggesting that α'/β' axons act as “pioneering axons” for the later-arriving α/β axons. Because the α'/β' axons form morphologically normal lobes in the highwire mutant, the defect is likely at the molecular level, potentially involving a change in either membrane-associated or secreted guidance cues. In the vertebrate CNS, the highwire ortholog Phr1 is also required for a non-cell-autonomous mechanism that guides cortical axons. In the absence of Phr1, cortical axons stall at the corticostriatal border and do not contribute to the internal capsule. In contrast, after conditional excision of Phr1 exclusively in cortical neurons, these same cortical axons can now cross this choice point and path-find to the thalamus. Hence, the requirement for PHR proteins for non-cell-autonomous axon guidance mechanisms is evolutionarily conserved, although there is no evidence that the molecular mechanism is conserved (Shin, 2011).

To investigate the molecular mechanism of the non-cell-autonomous requirement for highwire, genetic interactions between highwire and candidate guidance molecules were tested. The data suggest that Highwire promotes a Plexin A signaling mechanism that is required for proper α/β lobe development. Loss of a single copy of the plexin A gene has no effect on MB development in an otherwise WT background but enhances the phenotype of a weak highwire allele. Furthermore, RNAi-mediated knockdown of plexin A in the MB has a very similar phenotype to loss of highwire, with abnormal thickness of α/β lobes and shortened α lobes. Hence, Plexin A is required for normal MB development. Plexins are receptors for semaphorins, and both Sema-1a and Sema-5c are required for normal MB development. The genetic studies did not uncover a genetic interaction between either of these semaphorins and highwire, but the absence of such an interaction does not rule out the involvement of these or other semaphorins. Two potential models are consistent with these genetic interaction studies. First, Plexin A may function to downregulate Wnd/DLK, potentially via inhibition of Rac GTPase signaling. In the absence of either plexin A or highwire, Wnd/DLK activity would be upregulated, disrupting axonal interactions between α'/β' axons and α/β axons via unknown mechanisms. Alternatively, excess Wnd/DLK activity in the highwire mutant could disrupt the Plexin A signaling pathway that is necessary for α/β lobe development. The mechanisms by which Highwire and Plexin A signaling converge will be the subject of future studies (Shin, 2011).

Highwire regulates presynaptic BMP signaling essential for synaptic growth; Highwire interacts with Medea

Highwire, a putative RING finger E3 ubiquitin ligase, negatively regulates synaptic growth at the neuromuscular junction (NMJ) in Drosophila. hiw mutants have dramatically larger synaptic size and increased numbers of synaptic boutons. Hiw binds to the Smad protein Medea (Med). Med is part of a presynaptic bone morphogenetic protein (BMP) signaling cascade consisting of three receptor subunits, Wit, Tkv, and Sax, in addition to the Smad transcription factor Mad. When compared to wild-type, mutants of BMP signaling components have smaller NMJ size, reduced neurotransmitter release, and aberrant synaptic ultrastructure. BMP signaling mutants suppress the excessive synaptic growth in hiw mutants. Activation of BMP signaling, which in wild-type does not cause additional growth, in hiw mutants does lead to further synaptic expansion. These results reveal a balance between positive BMP signaling and negative regulation by Highwire, governing the growth of neuromuscular synapses (McCabe, 2004).

To search for proteins that interact with Hiw, a yeast two-hybrid screen of a Drosophila cDNA library was carried out using two regions of the Hiw protein as bait (HWB1: aa 2063–3461; and HWB2: aa 4082–5233). From a screen of 4 × 107 transformants, three candidate proteins were isolated that interacted positively with HWB1 and one candidate protein that showed positive interaction with HWB2. One of the three candidate proteins that interacted with HWB1 was identified as the C-terminal region of the second mad homology domain (MH2) of Med (aa 616-745). The Med clone did not self-activate, and when positive interaction was tested by cotransforming the Med clone together with either HWB1 or HWB2, a strong interaction was found with HWB1 but no interaction was seen with HWB2. Furthermore, GST-fused Med protein is able to bind to in vitro translated preparations of Hiw. The binding region for Med was between amino acids 2063 and 3461 of Hiw (included in HWB1), while binding was not observed with the region of Hiw found in HWB2. Interestingly, the Med binding domain of Hiw includes the region used to generate the partial Hiw transgene, Hiw-DN, that produces a dominant-negative hiw phenotype when expressed in the nervous system. The results of yeast two-hybrid experiments together with in vitro binding results show that Med and Hiw proteins can interact and that the MH2 region of Med is sufficient for this interaction (McCabe, 2004).

This study demonstrates that Hiw negatively regulates the BMP signaling cascade, required for the normal growth and function of neuromuscular synapses in Drosophila. hiw mutants have a dramatic overexpansion of synaptic structures, with many more synaptic boutons than wild-type. In contrast to hiw mutants, mutants of Med have smaller synapses and many fewer synaptic boutons compared to wild-type. Med is part of a presynaptic signaling cascade that includes three cell surface receptors (Wit, Tkv, and Sax) and two intracellular transcription factors (Med and Mad) that transmits a BMP signal from the NMJ to the motoneuron nucleus. Genetic removal of either Med or Wit can completely suppress the synaptic overgrowth in hiw mutants, while activation of BMP signaling in hiw mutants produces even more synaptic growth. These findings provide evidence for a functional link between the action of Hiw and BMP signaling to control synaptic growth at the Drosophila NMJ (McCabe, 2004).

The type II BMP receptor Wit and the BMP ligand Gbb are required for synaptic growth at the neuromuscular junction. While Wit is required in presynaptic neurons for normal NMJ growth, the requirement for Gbb in postsynaptic muscles is consistent with a retrograde BMP signal. Yet it was not clear from these studies how Wit carried out this function, whether by a local signaling mechanism within the synapse or via a signaling cascade. Evidence is provided that strongly suggests BMP signaling through Wit regulates synaptic growth primarily by altering transcription. Two lines of evidence support this conclusion. (1) Mutants of both of the Smad transcription factors Med and Mad have similar defects in synaptic growth, presynaptic ultrastructure, and function to mutants of the cell surface receptors wit, tkv, and sax. (2) Phosphorylated Mad is absent from the nucleus of motoneurons in wit, sax, and tkv mutants. The absence of P-Mad in the nuclei of motoneurons in BMP receptor mutants combined with the similarity of synaptic phenotypes between mutants in receptors and mutants of intracellular Smads argues that BMPs exert their influence on synapses primarily by signaling to the nucleus rather than having a local synaptic activity (McCabe, 2004).

The specificity of synaptic BMP signaling seems to be maintained by Wit, which is expressed only in the nervous system, while Med, Mad, Tkv, and Sax are found in many tissues. This conclusion is supported by the finding of developmental defects in many tissues such as the fat body in Med, Mad, tkv, and sax mutants that are not found in wit mutants. Recently, an independent study found synaptic defects in sax and Mad mutants (Rawson, 2003). These studies have been extended, demonstrating that Med and Tkv, in addition to Mad and Sax, are required in presynaptic neurons for normal NMJ growth, and mutants of all these molecules have similar characteristic defects in presynaptic active zone ultrastructure. This study further shows, by rescuing wit mutants with a pair of Tkv::Wit chimeric receptors, that Tkv and Wit function together in vivo, and that Tkv is localized at the NMJ. All these data therefore suggest that Wit, Tkv, and Sax receive a retrograde BMP signal from muscles by Gbb (and possibly other BMP ligands) at the NMJ and transmit this signal to the nucleus via Mad and Med to induce transcriptional change. This neurotrophic signal is essential for the coordination of presynaptic NMJ expansion with postsynaptic muscle growth (McCabe, 2004).

While BMP signaling plays an essential role in neuromuscular junction expansion, BMP signaling mutants also have dramatic reductions in the levels of neurotransmitter release and aberrant presynaptic ultrastructure at active zones. Several pieces of data suggest the role of BMP signaling in the regulation of neurotransmitter release may be separable from its role in synaptic growth. Restoration of Gbb in the nervous system of gbb mutants can rescue neurotransmitter release to wild-type levels while not restoring normal synaptic size. This result is reminiscent of Fasciclin II mutations, which also have reduced synaptic size but normal neurotransmitter release. Furthermore, Wit is necessary for the homeostatic regulation of neurotransmitter release. It may be that the involvement of BMP signaling in this process is independent of, but complementary to, its role in regulating synaptic structural growth (McCabe, 2004).

Despite the central requirement for BMP signaling in synaptic growth, when attempts were made to increase BMP signaling in motoneurons, no synaptic overgrowth beyond wild-type levels was seen. These observations are explained by proposing the presence of a negative regulatory process that tightly controls the levels of synaptic BMP signaling. Hiw is a key and necessary component of this regulatory process (McCabe, 2004).

Hiw is an extremely large protein, making in vitro confirmation of its ubiquitination activity difficult. Despite the absence of direct biochemical data, several lines of evidence suggest that Hiw does function as an E3 ubiquitin ligase. Hiw has a signature RING-H2 finger, a domain that has a general function in ubiquitin-mediated protein degradation. RING fingers can function as modules that interact with E2 ubiquitin-conjugating enzymes to catalyze ubiquitination. Futhermore, hiw mutants have a strong genetic interaction with the deubiquitinating enzyme Fat Facets. Overexpression of either Fat Facets or the yeast deubiquitinating protease UBP2 in presynaptic neurons produces a synaptic overgrowth phenotype very similar to the hiw mutant phenotype. Given this evidence, a model is proposed whereby Hiw negatively regulates BMP signaling at the NMJ by a ubiquitination-dependant mechanism, antagonizing BMP signaling and controlling synaptic growth (McCabe, 2004).

In support of this model, Hiw has been shown to specifically binds Med protein in both yeast two-hybrid and in vitro binding assays. This is consistent with a function for Hiw as an E3 ubiquitin ligase, since these proteins specifically bind to their substrates before targeting them for proteolysis. Unfortunately, several antibodies against mammalian Smad4 failed to detect Med, precluding assays of Med's ubiquitination status. Interestingly, however, the region in Hiw that interacts with Med is included in the sequence of a partial Hiw transgene, Hiw-DN, that causes synaptic overgrowth when expressed in the nervous system. Since this transgene does not include the RING finger domain, its dominant-negative effect could be mediated by its ability to inhibit the binding of Med by endogenous Hiw. In addition to physical interaction between Hiw and Med, genetic removal of Med has been demonstrated to suppress the increase in the number of synaptic boutons in hiw mutants to Med mutant levels. This implies that the dramatic increase in the number of synaptic boutons in hiw mutants is completely dependant upon the presence of Med. This increase is also suppressed by wit mutants, showing that it is Med's role as part of a BMP signaling cascade that mediates its suppression of hiw. Furthermore, it was shown that synaptic overgrowth due to the overexpression of the deubiquitinating enzymes Faf or UBP2 is also suppressed by disrupting BMP signaling. These results together support the model whereby Hiw regulates BMP signaling via a ubiquitin-dependent mechanism (McCabe, 2004).

Overexpression of a constitutively active Tkv type I receptor transgene in neurons does not cause any overgrowth at the NMJ. Similarly, loss-of-function mutants of the inhibitory Smad, Dad, does not show any synaptic overgrowth. Consistent with the model that Hiw regulates BMP signaling, in hiw mutants, activation of BMP signaling can now lead to further synaptic overgrowth. This suggests that while hiw mutants may have elevated levels of BMP signaling, further activation of the BMP pathway can induce yet more synaptic growth. By activating BMP signaling using transgenic constitutively active type I receptors, other factors that could conceivably limit the signal can presumably be bypassed, such as the availability of Wit or Gbb (McCabe, 2004).

In contrast to the current findings, dad mutants have previously been reported to produce large numbers of extra synaptic boutons. While the current study examined only one homoallelic mutant combination of dad, several homo and heteroallelic mutant combinations were examined, to eliminate the possibility of second site mutations, and this prior result was not confirmed. Thus the discrepancy remains unresolved. Consistent with the current data, it has been shown that overexpression of Wit cannot induce synapse overgrowth, despite the ability of overexpressed type II receptors to activate signaling in the absence of ligand. The current results are also supported by findings of Rawson (2003) (McCabe, 2004).

While the current data indicate that synaptic structural growth can be controlled by Hiw regulating BMP signaling, neurotransmitter release at the NMJ does not seem to be governed by an identical mechanism. BMP mutants have decreased neurotransmitter release, in addition to reduced numbers of synaptic boutons, when compared to wild-type. In contrast, hiw mutants have many more synaptic boutons than wild-type, but despite this, neurotransmitter release in hiw mutants is also reduced to levels similar to that of BMP mutants. Interestingly, double mutants of hiw;wit or hiw;Med have levels of neurotransmitter release similar to those of wit or Med mutants alone. This result indicates that the role of Hiw in controlling neurotransmitter release is distinct from its role as a negative regulator of synaptic structural growth. Previous results support this idea; the unknown retrograde signal that controls the homeostasis of neurotransmission at the NMJ is disrupted in wit mutants but remains functional in hiw mutants. Other aspects of the hiw mutant phenotype also appear to be independent of BMP signaling. Individual synaptic bouton size is reduced in hiw mutants, a phenotype not observed in BMP mutants and not suppressed by the inhibition of BMP signaling in hiw mutants. Also, the excessive degree of synaptic branching and arborization observed in hiw mutants is only partially suppressed by the disruption of BMP signaling. It is likely therefore that Hiw regulates other molecules responsible for these aspects of synaptic development (McCabe, 2004).

How is synaptic growth maintained? A model is proposed whereby a positive BMP signaling cascade is negatively regulated by the ubiquitin-protein ligase action of Hiw on the Smad Med. This model, however, leaves an important question unanswered: how is the balance between these two opposing forces maintained? This question cannot be answered with current knowledge, but some scenarios can be suggested. One possibility is that the level of phosphorylated Mad competes for binding of Med with Hiw. Once phosphorylated by type I receptors, Mad forms a complex with Med, and the formation of this complex is required for efficient signaling to the nucleus. It is conceivable that an equilibrium exists between the binding of phosphorylated Mad to Med and the binding of Med to Hiw. This equilibrium could potentially act to set a consistent level of BMP signaling and thus normal synaptic growth at the NMJ. Another alternative is that the ability of Hiw to block BMP signaling could be regulated by a third protein that is itself under the transcriptional control of BMP signaling. In this scenario, activation of BMP signaling leads to increased levels of this third protein that in turn activate Hiw's ability to target Med for ubiquitination, completing a negative feedback loop. Future studies will allow these and other possibilities to be tested to further dissect the opposing molecular forces that govern synaptic growth and function (McCabe, 2004).

Pam and its ortholog Highwire interact with and may negatively regulate the TSC1.TSC2 complex

Tuberous Sclerosis Complex (TSC) is an autosomal dominant disorder associated with mutations in TSC1, which codes for hamartin, or TSC2, which codes for tuberin. The brain is one of the most severely affected organs, and CNS lesions include cortical tubers and subependymal giant cell astrocytomas, resulting in mental retardation and seizures. Tuberin and hamartin function together as a complex in mammals and Drosophila. This paper reports the association of Pam, a protein identified as an interactor of Myc, with the tuberin-hamartin complex in the brain. The C terminus of Pam containing the RING zinc finger motif binds to tuberin. Pam is expressed in embryonic and adult brain as well as in cultured neurons. Pam has two forms in the rat CNS, an approximately 450-kDa form expressed in early embryonic stages and an approximately 350-kDa form observed in the postnatal period. In cortical neurons, Pam co-localizes with tuberin and hamartin in neurites and growth cones. Although Pam function(s) are yet to be defined, the highly conserved Pam homologs, HIW (Drosophila) and RPM-1 (Caenorhabditis elegans), are neuron-specific proteins that regulate synaptic growth. This study shows that HIW can genetically interact with the Tsc1.Tsc2 complex in Drosophila and can negatively regulate Tsc1.Tsc2 activity. Based on genetic studies, HIW has been implicated in ubiquitination, possibly functioning as an E3 ubiquitin ligase through the RING zinc finger domain. Therefore, it is hypothesized that Pam, through its interaction with tuberin, could regulate the ubiquitination and proteasomal degradation of the tuberin-hamartin complex particularly in the CNS (Murthy, 2004).

Highwire restrains synaptic growth by attenuating a MAP kinase signal

Highwire is an extremely large, evolutionarily conserved E3 ubiquitin ligase that negatively regulates synaptic growth at the Drosophila NMJ. Highwire has been proposed to restrain synaptic growth by downregulating a synaptogenic signal. This study identifies such a downstream signaling pathway. A screen for suppressors of the highwire synaptic overgrowth phenotype yielded mutations in wallenda, a MAP kinase kinase kinase (MAPKKK) homologous to vertebrate DLK and LZK. wallenda is both necessary for highwire synaptic overgrowth and sufficient to promote synaptic overgrowth, and synaptic levels of Wallenda protein are controlled by Highwire and ubiquitin hydrolases. highwire synaptic overgrowth requires the MAP kinase JNK and the transcription factor Fos. These results suggest that Highwire controls structural plasticity of the synapse by regulating gene expression through a MAP kinase signaling pathway. In addition to controlling synaptic growth, Highwire promotes synaptic function through a separate pathway that does not require Wallenda (Collins, 2006).

JNK signaling affects many cellular processes, often by regulating transcription factor activity that leads to changes in gene expression. A common downstream effector of JNK-mediated changes in gene expression is the AP-1 complex of Fos and Jun transcription factors, which can regulate synaptic growth at the Drosophila NMJ. To investigate whether Drosophila Fos or Jun (known as D-fos and D-jun, respectively) are required for highwire-dependent synaptic overgrowth, each was inhibited by expressing dominant-negative transgenes that contain the DNA binding and dimerization domains of Fos and Jun but lack the transcriptional activation domains. Expression of these dominant-negative transgenes in postmitotic neurons allowed circumvention of early embryonic requirements for D-fos and D-jun (Collins, 2006).

When FosDN and JunDN are neuronally expressed in a wild-type background, there is a modest trend toward inhibition of synaptic growth. When expressed in a highwire mutant background, the FosDN transgene confers dramatic suppression of the highwire synaptic phenotype, reducing bouton number and branching (42%) and increasing the intensity of staining for synaptic vesicle markers at the synapse. The reduction in highwire-dependent synaptic overgrowth is much greater than the reduction of growth in a wild-type background. In contrast, JunDN does not suppress the highwire phenotype. This suggests the existence of a pathway that is separate from AP-1, consistent with results in Drosophila demonstrating that D-Fos can act independently of D-Jun. The requirement for D-Fos in highwire synaptic overgrowth suggests that the highwire phenotype involves changes in gene expression rather than exclusively local changes to the synapse (Collins, 2006).

If FosDN acts downstream of Wallenda to inhibit synaptic overgrowth, it should also suppress the synaptic overgrowth caused by overexpressing wallenda. Indeed, when FosDN was coexpressed with UAS-wnd in neurons, FosDN could suppress the wallenda gain-of-function phenotype, leading to a 38% reduction in synaptic bouton number, a 52% reduction in synaptic branching, a 54% increase in bouton size, and a 3.8-fold increase in the intensity of staining of synaptic vesicle markers. This is consistent with D-Fos acting downstream of Wallenda to promote synaptic growth. Therefore, the synaptic overgrowth phenotypes caused by loss of highwire and by overexpression of wallenda are similar in their requirements for the transcription factor D-Fos (Collins, 2006).

Current models suggest that Highwire functions as an E3 ubiquitin ligase to downregulate a signaling pathway that promotes synaptic growth. This study identified a MAPKKK, Wallenda, whose protein levels are controlled by Highwire and the activity of ubiquitin hydrolases. Wallenda is both necessary for highwire-dependent synaptic overgrowth and sufficient to promote synaptic growth. Downstream of Wallenda, the MAP kinase JNK and transcription factor Fos are required for highwire-dependent synaptic overgrowth. It is proposed that Highwire restrains synaptic growth by downregulating the MAPKKK Wallenda, thereby inhibiting signaling through the JNK MAP kinase and the Fos transcription factor. In the absence of highwire, this signaling pathway is overactive, leading to changes in gene expression that result in excessive synaptic growth (Collins, 2006).

The regulation of the MAPKKK Wallenda is conserved in Drosophila and C. elegans (Nakata, 2005). In both organisms, the synaptic phenotype of highwire/rpm-1 requires the Wallenda/DLK-1 MAPKKK and downstream MAPK signaling. However, the downstream MAPK pathways diverge: in C. elegans, the rpm-1 phenotype requires a p38 MAP kinase (Nakata, 2005), while the highwire phenotype requires JNK signaling. This suggests that regulation of the specific MAPKKK Wallenda/DLK-1, rather than a particular downstream MAP kinase pathway, is a fundamental activity of Highwire and its orthologs (Collins, 2006).

Since Highwire functions as an E3 ubiquitin ligase to restrain synaptic growth, Wallenda is a compelling candidate target for the following reasons: (1) wallenda functions downstream of highwire and is essential for the synaptic overgrowth in highwire mutants; (2) increasing the levels of Wallenda by overexpression is sufficient to confer synaptic overgrowth; (3) Highwire regulates Wallenda protein levels through a posttranscriptional and most likely posttranslational mechanism. Each of the points above is conserved in C. elegans (Nakata, 2005 ). (4) Wallenda protein levels are regulated by ubiquitination in vivo, since inhibiting ubiquitination by overexpressing ubiquitin hydrolases increases the levels of Wallenda protein. (5) The RING domain of the C. elegans homolog rpm-1 can interact with the Wallenda homolog DLK-1 (Nakata, 2005) and stimulate its ubiquitination when both are overexpressed in 293T cells (Collins, 2006).

Targeting a MAPKKK, which sits at the top of a MAP kinase signaling pathway, is an attractive mechanism for spatially and temporally controlling a synaptogenic signal without affecting downstream components shared by multiple MAPK signaling cascades. Restraining MAP kinase signaling is essential for controlling diverse cellular processes, including cell proliferation, differentiation, and apoptosis. The targeting of MAPKKKs by specific ubiquitin ligases may be a powerful and general mechanism for regulating MAP kinase signals (Collins, 2006).

While Wallenda is an essential mediator of the highwire mutant phenotypes in both Drosophila and C. elegans, an endogenous synaptic function for Wallenda has not yet been identified in either organism: the wallenda mutants have surprisingly normal synapse morphology and function. This may be due to another pathway that compensates for the loss of wallenda function. Such redundancy would obscure the role of wallenda. A second possibility is that wallenda functions in an aspect of synaptic growth that is not detected or required under laboratory culture conditions. For instance, wallenda could promote synaptic growth as part of a structural plasticity program that responds to unknown experience-dependent stimuli. A third possibility is that Wallenda does not normally function at synapses, but its upregulation in highwire mutants causes a neomorphic phenotype. In this scenario, the regulation of Wallenda by Highwire is required for normal synaptic development, but endogenous Wallenda would not itself regulate the synapse. The neuropil and synaptic localization of Wallenda and the vertebrate homolog DLK (Hirai, 2005) is, however, consistent with a synaptic function (Collins, 2006).

As an activator of MAP kinase signaling, Wallenda and its homologs might also control other processes beyond the synapse. Functional studies in vertebrates suggest that DLK and JNK signaling regulate neuronal migration and axon outgrowth in the developing cortex (Hirai, 2002). Outside of the nervous system, DLK influences keratinocyte differentiation, and LZK is highly expressed in the pancreas, liver, and placenta. In Drosophila, wallenda mutants are female sterile. It is predicted that the regulation of DLK and LZK is conserved from worms and flies to vertebrates. Therefore, the vertebrate homologs of Highwire might regulate some of these neuronal and/or extraneuronal developmental processes (Collins, 2006 and references therein).

Highwire is a large, multidomain protein that, in addition to acting as an E3 ubiquitin ligase, has been shown to inhibit adenylate cyclase, influence TSC signaling and pteridine biosynthesis, and interact with the myc oncogene and the co-SMAD Medea. It is remarkable that throughout millions of years of evolution, members of the Highwire family have retained an exceptionally large size and complex domain structure. An attractive explanation for this conservation is that this molecule could serve as an intersection point for multiple signaling pathways, integrating MAP kinase and other signals during neural development (Collins, 2006).

The ubiquitin ligase activity alone could be responsible for regulating more than one downstream target. Interactions with components of TSC (tuberin/hamartin) and TGF-β signaling pathways suggest that Highwire might target either or both of these pathways. The model that Highwire regulates TGF-β signaling through interaction with the co-SMAD Medea has received considerable attention. Since the TGF-β pathway regulates synaptic growth at the NMJ, it has been proposed that synaptic overgrowth of highwire mutants is caused by overactivity of this pathway. Null alleles of wit, which completely disrupt TGF-β signaling at the NMJ, can partially suppress the highwire phenotypes: they partially suppress the increase in bouton number, but show little or no suppression of the reduced bouton size and the reduced intensity for synaptic vesicle markers. This partial suppression of highwire by wit is consistent with the model that overactive TGF-β signaling contributes to the highwire phenotype. However, the data are also consistent with the alternate model that TGF-β signaling and Highwire act in parallel pathways. An assay for the activity of TGF-β signaling is to stain for phosphorylated-MAD (phospho-MAD), the major transducer of BMP signals in Drosophila, in motoneuron nuclei. No change was detected in the levels of phospho-MAD staining in highwire mutants compared to wild-type. This assay is sensitive to changes in pathway activity—neuronal expression of the constitutively active type I receptor thick veins leads to a 40% increase in phospho-MAD staining. Interestingly, this increase in TGF-β signaling does not lead to excess synaptic growth. Combining a highwire mutant with expression of constitutively active thick veins does cause excess growth, but it does not lead to any further increase in phospho-MAD staining. These data are consistent with highwire and TGF-β signaling acting in parallel pathways (Collins, 2006).

Whether or not Highwire regulates TGF-β signaling, it is likely to target an additional pathway. Highwire not only restrains synaptic growth, but also promotes synaptic function. Synaptic function requires the ubiquitin ligase activity of Highwire and is sensitive to the levels of the ubiquitin hydrolase fat facets. This study demonstrates that this regulation of neurotransmitter release does not require Wallenda. Therefore, Highwire must regulate at least two distinct molecular pathways. If Wallenda is a substrate whose downregulation is essential for restraining synaptic growth, there is likely another substrate for Highwire whose downregulation promotes neurotransmitter release (Collins, 2006).

Downstream of Wallenda, the JNK MAP kinase and Fos transcription factor are required for the highwire synaptic morphology phenotype. Therefore, Highwire attenuates a JNK signaling pathway that presumably controls gene expression to regulate synaptic growth. Previous studies have implicated JNK-dependent transcriptional control in activity-dependent growth of the Drosophila NMJ. However, this previously described pathway is probably distinct from the JNK signal that is controlled by Highwire and activated by Wallenda. The previously described role for JNK requires AP-1, a heterodimer of Fos and Jun transcription factors; inhibiting either D-Fos or D-Jun disrupts this pathway. In contrast, highwire-induced overgrowth requires D-Fos, but not D-Jun. The Wallenda pathway could therefore involve a homodimer of D-Fos or another transcription factor that interacts with Fos. Such D-Jun-independent functions of D-Fos have been described previously in Drosophila. The differential requirement for transcription factors suggests that the output of Wallenda signaling cannot simply be activation of JNK, but instead activation of JNK in a particular spatial or temporal context, such as in the presence of cofactors that influence downstream signaling (Collins, 2006).

In addition to transcription factors, substrates for activated JNK include components of the cytoskeleton. Because the NMJ is distant from the motoneuron nucleus, and because vertebrate DLK colocalizes with tubulin in axonal regions of the brain, it was initially expected that the Highwire/Wallenda/JNK pathway would influence synaptic morphology through local action upon the synaptic cytoskeleton. Instead, a requirement was identified for a transcription factor and presumably changes in gene expression. However, this does not exclude an interaction with the cytoskeleton or local changes at the synapse. It is possible that Highwire regulates the Wallenda signal in the cell body. However, the observation that Wallenda accumulates in the synapse-rich neuropil and at the NMJ when Highwire is absent suggests that Wallenda could become activated at the synapse. This would imply the need for a mechanism to transport the activated JNK signal back to the nucleus. In addition, cell-wide changes in gene expression must then be translated into localized growth at the synapse. Activated Wallenda at the synapse is an attractive candidate to integrate changes in gene expression with regulation of the synaptic cytoskeleton to control synaptic growth (Collins, 2006).

DFsn collaborates with Highwire to down-regulate the Wallenda/DLK kinase and restrain synaptic terminal growth

The growth of new synapses shapes the initial formation and subsequent rearrangement of neural circuitry. Genetic studies have demonstrated that the ubiquitin ligase Highwire restrains synaptic terminal growth by down-regulating the MAP kinase kinase kinase Wallenda/dual leucine zipper kinase (DLK). To investigate the mechanism of Highwire action, DFsn has been identified as a binding partner of Highwire and characterized the roles of DFsn (CG4643) in synapse development, synaptic transmission, and the regulation of Wallenda/DLK kinase abundance. This study identified DFsn as an F-box protein that binds to the RING-domain ubiquitin ligase Highwire and that can localize to the Drosophila neuromuscular junction. Loss-of-function mutants for DFsn have a phenotype that is very similar to highwire mutants -- there is a dramatic overgrowth of synaptic termini, with a large increase in the number of synaptic boutons and branches. In addition, synaptic transmission is impaired in DFsn mutants. Genetic interactions between DFsn and highwire mutants indicate that DFsn and Highwire collaborate to restrain synaptic terminal growth. Finally, DFsn regulates the levels of the Wallenda/DLK kinase, and wallenda is necessary for DFsn-dependent synaptic terminal overgrowth. In conclusion, the F-box protein DFsn binds the ubiquitin ligase Highwire and is required to down-regulate the levels of the Wallenda/DLK kinase and restrain synaptic terminal growth. It is proposed that DFsn and Highwire participate in an evolutionarily conserved ubiquitin ligase complex whose substrates regulate the structure and function of synapses (Wu, 2007; full text of article).

Highwire and the C. elegans homolog, RPM-1, act as ubiquitin ligases to regulate synaptic development. Liao (2004) proposed that, in C. elegans, RPM-1 participates in an atypical SCF ubiquitin ligase complex with the F-box protein FSN-1. Consistent with this hypothesis, RPM-1 binds to FSN-1 as well as to Skp-1 and Cullin-1, core components of SCF complexes. In addition, FSN-1 null mutants have very similar phenotypes to rpm-1 mutants at GABAergic synapses, but weaker phenotypes in DD motoneurons and sensory neurons. The difference in phenotypes suggests that RPM-1 interacts with other F-box proteins in addition to FSN-1, acts as a ubiquitin ligase without an F-box partner, or has ubiquitin-independent functions. The target of the RPM-1/FSN-1 complex in C. elegans is not clear. Biochemical and genetic data indicate that the receptor tyrosine kinase ALK is the functionally relevant target for FSN-1, while the MAPKKK DLK is the functionally relevant target for RPM-1. The data in Drosophila support the model from worms that RPM-1 and FSN-1 form a functional ubiquitin ligase complex, but simplify the model by demonstrating that in Drosophila both components target the same substrate (Wu, 2007).

This study demonstrates that Highwire binds the Drosophila homolog of FSN-1, DFsn. Therefore, the physical association of Highwire/RPM-1 and DFsn/FSN-1 is evolutionarily conserved. While eukaryotic genomes can encode hundreds of F-box proteins like DFsn, in worms, flies, mice, and humans there is only a single F-box protein that also contains an SPRY domain, and each is more closely related by sequence to each other than to other F-box proteins. Since the binding of DFsn/FSN-1 to Highwire/RPM-1 is conserved, it is speculated that the mouse and human homologs of DFsn, F-box protein 45 and hCG1734196, will bind to and function with Phr and PAM, the mouse and human homologs of Highwire, respectively. Indeed, expression analysis demonstrates that both the F-box protein 45 and Phr are expressed in a very similar pattern in the mouse brain (Wu, 2007).

These results suggest that the interaction of Highwire with DFsn is required for Highwire activity. Loss-of-function mutants for highwire and DFsn have qualitatively and quantitatively similar phenotypes - both are required to restrain synaptic terminal growth and promote synaptic release. Both Highwire and DFsn are necessary to down-regulate the levels of the MAPKKK Wallenda/DLK, and wallenda mutants suppress the morphological but not physiological phenotypes of both highwire and DFsn. Finally, genetic data support the model that Highwire and DFsn function together during synaptic development - DFsn mutants enhance the phenotype of a highwire hypomorph but not of a highwire null. All of these data are consistent with the model that Highwire and DFsn act together to form a functional ubiquitin ligase complex. In this model, the ligase complex targets Wallenda/DLK to restrain synaptic terminal growth, and an unknown substrate to promote synaptic function. It is speculated that the targeting of the Wallenda/DLK MAPKKK by the Highwire/DFsn complex will be conserved from worms to mammals. While Highwire and DFsn collaborate for synaptic development, the male sterility of DFsn but not highwire mutants suggests that DFsn has Highwire-independent functions in other developmental processes (Wu, 2007).

Protein turnover of the Wallenda/DLK kinase regulates a retrograde response to axonal injury

Regenerative responses to axonal injury involve changes in gene expression; however, little is known about how such changes can be induced from a distant site of injury. This study describes a nerve crush assay in Drosophila to study injury signaling and regeneration mechanisms. Wallenda (Wnd), a conserved mitogen-activated protein kinase (MAPK) kinase kinase homologous to dual leucine zipper kinase, was found to function as an upstream mediator of a cell-autonomous injury signaling cascade that involves the c-Jun NH(2)-terminal kinase MAPK and Fos transcription factor. Wnd is physically transported in axons, and axonal transport is required for the injury signaling mechanism. Wnd is regulated by a conserved E3 ubiquitin ligase, named Highwire (Hiw) in Drosophila. Injury induces a rapid increase in Wnd protein concomitantly with a decrease in Hiw protein. In hiw mutants, injury signaling is constitutively active, and neurons initiate a faster regenerative response. These data suggest that the regulation of Wnd protein turnover by Hiw can function as a damage surveillance mechanism for responding to axonal injury (Xiong, 2010).

The Highwire ubiquitin ligase promotes axonal degeneration by tuning levels of Nmnat protein

Axonal degeneration is a hallmark of many neuropathies, neurodegenerative diseases, and injuries. Using a Drosophila injury model this study has identified a highly conserved E3 ubiquitin ligase, Highwire (Hiw), as an important regulator of axonal and synaptic degeneration. Mutations in hiw strongly inhibit Wallerian degeneration in multiple neuron types and developmental stages. This new phenotype is mediated by a new downstream target of Hiw, the NAD+ biosynthetic enzyme nicotinamide mononucleotide adenyltransferase (Nmnat), which acts in parallel to a previously known target of Hiw, the Wallenda dileucine zipper kinase (Wnd/DLK) MAPKKK. Hiw promotes a rapid disappearance of Nmnat protein in the distal stump after injury. An increased level of Nmnat protein in hiw mutants is both required and sufficient to inhibit degeneration. Ectopically expressed mouse Nmnat2 is also subject to regulation by Hiw in distal axons and synapses. These findings implicate an important role for endogenous Nmnat and its regulation, via a conserved mechanism, in the initiation of axonal degeneration. Through independent regulation of Wnd/DLK, whose function is required for proximal axons to regenerate, Hiw plays a central role in coordinating both regenerative and degenerative responses to axonal injury (Xiong, 2012).

Since the discovery of the dramatic inhibition of degeneration by the WldS mutation, many studies have focused upon the action of the NAD+ biosynthetic enzyme isoforms, Nmnat1, Nmnat2, and Nmnat3, which in some circumstances can confer protection against axonal degeneration (reviewed in Coleman, 2010). Most of these studies involve gain-of-function overexpression experiments; it has been difficult to address the role of endogenous Nmnat enzymes in this process. Recent observations indicate that endogenous Nmnat activity plays an essential role in neuronal survival, and its depletion leads to neurodegeneration. In addition, recent studies in vertebrate neurons suggest that the cytoplasmic isoform, Nmnat2, has a short half-life in neurons. An attractive model proposes that Nmnat2 is rapidly turned over in axons, and that its loss in the distal stump of an axon, which has become disconnected from its cell body, leads to the initiation of Wallerian degeneration (Xiong, 2012).

Some aspects of this model are supported by current in vivo characterization in Drosophila. This study identfies Hiw, a highly conserved protein with features of an E3 ubiquitin ligase, as an important regulator of Wallerian degeneration. Hiw's role in this process involves the Nmnat protein, whose levels in axons and synapses are regulated post-transcriptionally by Hiw function. In hiw mutants, Wallerian degeneration is strongly inhibited, and the increased level of Nmnat protein in hiw mutants is both required and sufficient to inhibit degeneration (Xiong, 2012).

While the localization of endogenous Hiw in Drosophila is not known, homologues in mice and Caenorhabditis elegans have been detected in axons and at synapses, so it is in the appropriate location to target the destruction of Nmnat in distal axons. However, it remains to be determined whether the down-regulation of Nmnat in the distal stump per se is the trigger for Wallerian degeneration. When HA-Nmnat was overexpressed, axons were protected from degeneration long after the rapid disappearance of detectable protein in the distal stump. It is possible that even very low levels of Nmnat protein are sufficient to protect from degeneration. It is also formally possible that the basal levels of Nmnat before injury, rather than the disappearance of Nmnat after injury, is an important determinant of degeneration. It is also acknowledged that axonal degeneration likely involves additional steps downstream or in parallel to the regulation of Nmnat by Hiw. While overexpression of Hiw can induce a reduction in HA-Nmnat levels, it was not possible to observe an enhanced rate of degeneration when Hiw was overexpressed (Xiong, 2012).

Studies almost a decade ago suggested a role for the ubiquitin protease system (UPS) in the initiation of Wallerian degeneration (Zhai, 2003). It is tempting to propose that this role is manifested by the regulation of Nmnat by Hiw. However the current observations caution against a simple interpretation that Hiw regulates Nmnat via the UPS, since Hiw can promote disappearance of Nmnat protein in cells in a manner unaffected by proteasome inhibitors. Moreover, in vivo, inhibition of the proteasome had only a minor effect upon Nmnat levels in a wild-type background. However in hiw mutants, Nmnat levels were very sensitive to the function of the proteasome. It is interpreted that additional ubiquitin ligases and the UPS may regulate Nmnat independently of Hiw (Xiong, 2012).

Regardless of the role of the proteasome, the current observations suggest that ubiquitin plays an important role in Nmnat regulation. Overexpression of the yeast de-ubiquitinating protease UBP2 leads to increased levels of Nmnat protein and inhibition of Wallerian degeneration, in a manner that requires endogenous Nmnat. Future studies of the mechanism by which Hiw regulates Nmnat will therefore consider potential proteasome-independent roles of ubiquitination. Of note, in yeast UBP2 has been shown to preferentially disassemble polyubiquitin chains linked at Lys63, which have been found to perform non-proteolytic functions in DNA repair pathways, kinase activation, and receptor endocytosis. The possibility should also be considered that Hiw regulates Nmnat indirectly: since thus far it has not been possible to detect any ubiquitinated Nmnat species, it is possible that an intermediate, yet unknown, regulator of Nmnat may be the actual substrate of ubiquitination. Nevertheless, co-immunoprecipitation studies from S2R+ cells indicate that Hiw and Nmnat have the capacity to interact (Xiong, 2012).

The mechanism and cellular location of Nmnat's protective action is a highly debated subject. Observations in the literature point to both NAD+-dependent and NAD+-independent models for the strong protection by the WldS mutation (Coleman, 2010). The location of its protective action may be the mitochondria, since mitochondrially localized Nmnat can protect axons from degeneration. However golgi/endosomal localized Nmnat2 can also be protective. The findings suggest that mutation of hiw leads to an increase in the pool of endogenous Nmnat that functionally impacts degeneration (Xiong, 2012).

While the site of endogenous Nmnat function during axonal degeneration remains to be identified, this study found that the levels of ectopically expressed mouse Nmnat2 were specifically increased in the hiw mutant background. In contrast, the levels of nuclearly localized mNmnat1 or mitochondrially localized mNmnat3 were unaffected by Hiw. Since Nmnat2 has a short half-life in vertebrate neurons, it is intriguing to propose that it is regulated by Hiw orthologs via an analogous mechanism (Xiong, 2012).

Since Nmnat2 does not appear to localize to mitochondria, does this favor a non-mitochondrial activity, such as function as a chaperone, for the protective action? It remains challenging to determine the exact location of protection, since the most apparent changes in Nmnat protein may not necessarily be the functionally relevant changes (Xiong, 2012).

A previously characterized target of Hiw regulation is the Wallenda (Wnd) MAP kinase kinase kinase. This axonal kinase is also capable of inhibiting Wallerian degeneration in motoneurons. The protective action of Wnd requires a downstream signaling cascade and changes in gene expression mediated by the Fos transcription factor. Loss of nmnat does not affect this signaling cascade nor does it affect the protective action of Wnd. Conversely, loss of wnd does not affect the protection caused by overexpressing nmnat. Importantly, the regulation of Nmnat by Hiw does not appear to require Wnd function, and Wnd and Nmnat can protect axons independently of each other. These findings favor the model that Wnd and Nmnat are both regulated by Hiw and influence axonal degeneration through independent mechanisms (Xiong, 2012).

The Wnd kinase plays additional roles in neurons, which can be genetically separated from Nmnat function. These include regulation of synaptic growth: a dramatic synaptic overgrowth phenotype in hiw mutants is fully suppressed by mutation of wnd, but is not at all affected by knockdown of nmnat. Wnd/DLK also promotes axonal sprouting in response to axonal injury, which is also unaffected by nmnat knockdown. It is therefore clear that by regulating both Wnd and Nmnat, Hiw regulates multiple independent pathways in neurons (Xiong, 2012).

It is intriguing that the actions of both Wnd and Nmnat promote cellular responses to axonal injury. Axonal regeneration requires an initiation of a growth program within the axon, which depends upon the function of Wnd and its homologues. Equally important is a clearance of the distal stump to make room for the regenerating axon. Since both Wnd and Nmnat are transported in axons a model is proposed in which Hiw function in the distal axon terminal could simultaneously promote destruction of Nmnat in the distal stump, and accumulation of Wnd in the proximal stump. The latter is observed after injur, and is required to promote new axonal growth. The actual location in which Hiw regulates Nmnat remains to be determined. As an upstream regulator of both sprouting in the proximal stump and degeneration of the distal stump, Hiw may play a central role in regulating the ability of a neuron to regenerate its connection after injury (Xiong, 2012).

Importantly, the protective action of Nmnat may not be limited to Wallerian degeneration. The WldS mutation can protect neurons from degeneration in a wide variety of paradigms, from models of neurodegenerative disease, diabetic neuropathy, excitotoxity, and loss of myelination. These findings suggest that action and regulation of Nmnat function is broadly important for neuronal function and maintenance. As a critical regulator of Nmnat, the Hiw ubiquitin ligase and its vertebrate homologues deserve further scrutiny for potential roles in human health and disease (Xiong, 2012).

Drosophila Syncrip modulates the expression of mRNAs encoding key synaptic proteins required for morphology at the neuromuscular junction

Localized mRNA translation is thought to play a key role in synaptic plasticity, but the identity of the transcripts and the molecular mechanism underlying their function are still poorly understood. This study shows that Syncrip, a regulator of localized translation in the Drosophila oocyte and a component of mammalian neuronal mRNA granules, is also expressed in the Drosophila larval neuromuscular junction, where it regulates synaptic growth. RNA-immunoprecipitation followed by high-throughput sequencing and qRT-PCR were used to show that Syncrip associates with a number of mRNAs encoding proteins with key synaptic functions, including msp-300, syd-1 (RhoGAP100F), neurexin-1, futsch, highwire, discs large, and alpha-spectrin. The protein levels of MSP-300, Discs large, and a number of others are significantly affected in syncrip null mutants. Furthermore, syncrip mutants show a reduction in MSP-300 protein levels and defects in muscle nuclear distribution characteristic of msp-300 mutants. These results highlight a number of potential new players in localized translation during synaptic plasticity in the neuromuscular junction. It is proposed that Syncrip acts as a modulator of synaptic plasticity by regulating the translation of these key mRNAs encoding synaptic scaffolding proteins and other important components involved in synaptic growth and function (McDermott, 2014).

Localized translation is a widespread and evolutionarily ancient strategy used to temporally and spatially restrict specific proteins to their site of function and has been extensively studied during early development and in polarized cells in a variety of model systems. It is thought to be of particular importance in the regulation of neuronal development and in the plastic changes at neuronal synapses that underlie memory and learning, allowing rapid local changes in gene expression to occur independently of new transcriptional programs. The Drosophila neuromuscular junction (NMJ) is an excellent model system for studying the general molecular principles of the regulation of synaptic development and plasticity. Genetic or activity-based manipulations of synaptic translation at the NMJ has previously been shown to affect the morphological and electrophysiological plasticity of NMJ synapses. However, neither the mRNA targets nor the molecular mechanism by which such translational regulation occurs are fully understood (McDermott, 2014).

Previously work identified CG17838, the fly homolog of the mammalian RNA binding protein SYNCRIP/hnRNPQ, which was named Syncrip (Syp). Mammalian SYNCRIP/hnRNPQ is a component of neuronal RNA transport granules that contain CamKIIα, Arc, and IP3R1 mRNAs and is thought to regulate translation via an interaction with the noncoding RNA BC200/BC1, itself a translational repressor. Moreover, SYNCRIP/hnRNPQ competes with poly(A) binding proteins to inhibit translation in vitro and regulates dendritic morphology (Chen, 2012) via association with, and localization of, mRNAs encoding components of the Cdc-42/N-WASP/Arp2/3 actin nucleation-promoting complex. Drosophila Syp has a domain structure similar to its mammalian homolog, containing RRM RNA binding domains and nuclear localization signal(s), as well as encoding a number of protein isoforms. It was previously shown that Syp binds specifically to the gurken (grk) mRNA localization signal together with a number of factors previously shown to be required for grk mRNA localization and translational regulation (McDermott, 2012). Furthermore, syp loss-of-function alleles lead to patterning defects indicating that syp is required for grk and oskar (osk) mRNA localization and translational regulation in the Drosophila oocyte (McDermott, 2014).

This study shows that Syp is detected in the Drosophila third instar larval muscle nuclei and also postsynaptically at the NMJ. Syp is required for proper synaptic morphology at the NMJ, as syp loss-of-function mutants show a synaptic overgrowth phenotype, while overexpression of Syp in the muscle can suppress NMJ growth. Syp protein associates with a number of mRNAs encoding proteins with key roles in synaptic growth and function including, msp-300, syd-1, neurexin-1 (nrx-1), futsch, highwire (hiw), discs large 1 (dlg1), and α-spectrin (α-spec). The protein levels of a number of these mRNA targets, including msp-300 and dlg1, are significantly affected in syp null mutants. Furthermore, in addition to regulating MSP-300 protein levels, Syp is required for correct MSP-300 protein localization, and syp null mutants have defects in myonuclear distribution and morphology that resemble those observed in msp-300 mutants. It is proposed that Syp coordinates the protein levels from a number of transcripts with key roles in synaptic growth and is a mediator of synaptic morphology and growth at the Drosophila NMJ (McDermott, 2014).

The results demonstrate that Syp is required for the appropriate branching of the motoneurons and the number of synapses they make at the muscle. These observations are potentially explained by the finding that Syp is also required for the correct level of expression of msp-300, dlg1 and other mRNA targets. Given that it was previously shown that Syp regulates mRNA localization and localized translation in the Drosophila oocyte, and studies by others have shown that mammalian SYNCRIP/hnRNPQ inhibits translation initiation by competitively binding poly(A) sequences (Svitkin, 2013), these functions of Syp as occurring at the level of translational regulation of the mRNAs to which Syp binds. Our data are also consistent with other work in mammals showing that SYNCRIP/hnRNPQ is a component of neuronal RNA transport granulesthat can regulate dendritic morphology via the localized expression of mRNAs encoding components of the Cdc-42/N-WASP/Arp2/3 actin nucleation-promoting complex (McDermott, 2014 and references therein).

Translation at the Drosophila NMJ is thought to provide a mechanism for the rapid assembly of synaptic components and synaptic growth during larval development, in response to rapid increases in the surface area of body wall muscles or in response to changes in larval locomotion. The phenotypes observed in this study resemble, and are comparable to, those seen when subsynaptic translation is altered genetically or by increased locomotor activity. In syp null mutants, NMJ synaptic terminals are overgrown, containing more branches and synaptic boutons. Similarly, bouton numbers are increased by knocking down Syp in the muscle using RNAi. In contrast, overexpression of Syp in the muscle has the opposite phenotype, resulting in an inhibition of synaptic growth and branching. Furthermore, expressing RNAi against syp in motoneurons alone does not result in a change in NMJ morphology, indicating that Syp acts postsynaptically in muscle, but not presynaptically at the NMJ to regulate morphology. Interestingly, pan-neuronal syp knockdown or overexpression using Elav-GAL4 also results in NMJ growth defects, revealing that some of the defects observed in the syp null mutant may be attributed to Syp function in neuronal cell types other than the motoneurons, such as glial cells, which are known to influence NMJ morphology. Finally, while Syp is not required in the motoneuron to regulate synapse growth and is not detected in the motoneuron, the possibility cannot be excluded that Syp is present at low levels in the presynapse and regulates processes independent of synapse morphology. A further detailed characterization of the cell types and developmental stages in which Syp is expressed and functions is required to better understand the complex phenotypes that were observe (McDermott, 2014).

RNA binding proteins have emerged as critical regulators of both neuronal morphology and synaptic transmision, suggesting that protein production modulates synapse efficacy. Consistent with this, it has been shown in a parallel study that Syp is required for proper synaptic transmission and vesicle release and regulates the presynapse through expression of retrograde Bone Morphogenesis Protein (BMP) signals in the postsynapse. In this role, Syp may coordinate postsynaptic translation with presynaptic neurotransmitter release. These observations provide a good explanation for how Syp influences the presynapse despite being only detectable in the postsynapse. This study has shown that Syp associates with a large number of mRNAs within third instar larvae, many of which encode proteins with key roles in synaptic growth and function. Syp mRNA targets include msp-300, syd-1, nrx-1, futsch, hiw, dlg1, and α-spec. Syp negatively regulates Syd-1, Hiw, and DLG protein levels in the larval body wall but positively regulates MSP-300 and Nrx-1 protein levels. Dysregulation of these multiple mRNA targets likely accounts for the phenotypes that were observed. Postsynaptically expressed targets with key synaptic roles that could explain the synaptic phenotypes that were observed in syp alleles include MSP-300, α-Spec, and DLG. For example, mutants in dlg1 and mutants where postsynaptic DLG is destabilized or delocalized have NMJ morphology phenotypes similar to those observed upon overexpression of Syp in the muscle. Presynaptically expressed targets include syd-1, nrx-1, and hiw. However, this study has shown that syp knockdown in presynaptic motoneurons does not result in any defects in NMJ morphology. The RIP-Seq experiments were carried out using whole larvae and will, therefore, identify Syp targets in a range of different tissues and cells, the regulation of which may or may not contribute to the phenotype that were observed in syp mutants. It is, therefore, possible that Syp associates with these presynaptic targets in other neuronal cell types such as the DA neurons of the larval peripheral nervous system. It is also possible that Nrx-1 or Hiw are expressed and required postsynaptically in the muscle, but this has not been definitively determined. syp alleles may provide useful tools to examine where key synaptic genes are expressed and how they are regulated (McDermott, 2014).

The identity of localized mRNAs and the mechanism of localized translation at the NMJ are major outstanding questions in the field. To date, studies have shown that GluRIIA mRNA aggregates are distributed throughout the muscle. The Syp targets identified in this study, such as msp-300, hiw, nrx-1, α-spec, and dlg1, are now excellent candidates for localized expression at the NMJ. Ultimately, conclusive demonstration of localized translation will involve the visualization of new protein synthesis of targets during activity-dependent synaptic plasticity. Biochemical experiments will also be required to establish the precise mode of binding of Syp to its downstream mRNA targets, the basis for interaction specificity, and the molecular mechanism by which Syp differentially regulates the protein levels of its mRNA targets at the Drosophila NMJ. Despite the fact that mammalian SYNCRIP is known to associate with poly(A) tails, this study and other published work have revealed that Syp can associate with specific transcripts. How Syp associates with specific mRNAs is unknown, and future studies are needed to uncover whether the interaction of Syp with specific transcripts is dictated by direct binding of the three Syp RRM RNA binding domains or by binding to other specific mRNA binding proteins. It is also possible that specific mRNA stem–loops, similar to the gurken localization signal, are required for Syp to bind to its mRNA targets (McDermott, 2014).

This study shows that msp-300 is the most significant mRNA target of Syp. MSP-300 is the Drosophila ortholog of human Nesprin proteins. These proteins have been genetically implicated in various human myopathies. For example, Nesprin/Syne-1 or Nesprin/Syne-2 is associated with Emery-Dreifuss muscular dystrophy (EDMD) as well as severe cardiomyopathies. Moreover, Syp itself is increasingly linked with factors and targets that can cause human neurodegenerative disorders. Recent work has revealed that SYNCRIP/hnRNPQ and Fragile X mental retardation protein (FMRP) are present in the same mRNP granule, and loss of expression of FMRP or the ability of FMRP to interact with mRNA and polysomes can cause cases of Fragile X syndrome. Separate studies have also shown that SYNCRIP interacts with wild-type survival of motor neuron (SMN) protein but not the truncated or mutant forms found to cause spinal muscular atrophy, and Syp genetically interacts with Smn mutations in vivo. Understanding Syp function in the regulation of such diverse and complex targets may, therefore, provide new avenues for understanding the molecular basis of complex disease phenotypes and potentially lead to future therapeutic approaches (McDermott, 2014).



In a first attempt to understand Highwire function, both its mRNA and protein distribution were examined. For protein localization, a monoclonal antibody was generated against Hiw. Analysis by in situ hybridization shows that HIW mRNA is largely, if not exclusively, neural specific. Antisense probes label the ventral neurogenic region as early as stage 5. The staining persists in the neurogenic region at later stages and reaches its peak around stage 13, when both CNS and PNS neurons clearly express this transcript. After embryonic stage 14, the strength of CNS and PNS cell body expression starts to decrease. By stages 16 and 17, weak staining of neuronal cell bodies is observed throughout the CNS. However, by this stage, darker staining was also observed in a pattern within the CNS that coincides with the longitudinal axon tracts found where the synaptic neuropil forms; the same pattern was observed with four independent RNA in situ probes). This expression of mRNA is not associated with specific cell bodies, and thus it is concluded that it represents mRNA in axonal processes or synaptic regions (Wan, 2000).

Hiw protein is detectable at early embryonic stages in the CNS. In contrast to the mRNA expression, a peak of protein expression around stage 13 is not seen. Rather, antibody staining reveals that Hiw protein expression in the CNS is, by and large, constant from stage 13 through the end of embryogenesis. Within the CNS, most of the protein is expressed in the region of the longitudinal axon tracts around which the synaptic neuropil forms. No protein is seen in the CNS nerve roots, none in the peripheral motor and sensory nerves, and very little in the commissural axon pathways. Thus, Hiw expression correlates with where the synaptic neuropil forms in the CNS. Outside the CNS, the predominant location of Hiw protein expression is at NMJ synapses in the embyro. Protein expression is first observed during stage 16, just as these synapses are forming (Wan, 2000).

In third instar larvae, protein expression is observed at the NMJ and at the CNS neuropil. Type I synaptic boutons are clearly stained at all NMJs on all muscles. Weaker staining of type II boutons is detected. The synaptic staining is largely absent at the NMJ in hiw mutant alleles ND8, -9, and -69, and is greatly reduced in hiw mutant alleles hiwEMS, ND16, and ND51. In addition, a transgenic line of flies (pUAS-HIWC1-LD) was generated that ectopically expresses a fragment of Hiw protein containing the epitope used for generating the antibody. The Hiw mAb detects this Hiw protein fragment ectopically expressed in muscles or in tracheal branches, further confirming its specificity. When the horseradish peroxidase (HRP) product is overdeveloped, the antibody also faintly stains the surface of muscles and each of their attachment sites. However, it is believed that this staining is due to the recognition of an unrelated protein(s) since this staining is still present at about the same level in all of the hiw mutants (Wan, 2000).

To study the localization of Hiw at individual boutons, confocal microscopy was used. The staining is not uniform in boutons but rather is confined to patches. Double labeling experiments were performed with Hiw and other synaptic markers. Antibodies against Syt and Dlg label synaptic terminals in a rather uniform fashion. Double labeling with anti-Hiw and anti-Syt, as well as anti-Dlg, confirms that Hiw is not uniform within synaptic boutons (Wan, 2000).

To further identify the position and significance of these Hiw patches, antibodies that label active zones versus the areas around active zones, called periactive zones, were sought. The postsynaptic membrane just opposite the presynaptic active zones can be labeled with antibodies against an epitope tag on transgenic glutamate receptor subunits (MHC-myc-DGluRIIB). For double labeling with anti-Hiw, however, another active zone marker that was not a mouse monoclonal antibody was needed. The Drosophila homolog of the Pak cytoplasmic kinase is such a probe. The glutamate receptor subunit and Pak are precisely colocalized. By double labeling Hiw and Pak, it was found that most Hiw patches are nonoverlapping with Pak staining spots and form a pattern complementary to that of Pak. These data indicate that Hiw is localized to the periactive zones which surround the active zones (Wan, 2000).

A similar pattern of expression is seen for the CAM Fas II. Fas II staining is seen in nonoverlapping areas around Pak staining. The staining with both Fas II and Hiw fills the space outside the Pak staining areas, suggesting that both are localized around but not in active zones (Wan, 2000).

One additional antibody was used to confirm the distribution of Hiw protein outside active zones. Dap160 is a presynaptic membrane-associated protein that physically interacts with dynamin and is organized into an 'endocytic honeycomb' that surrounds active zones. Double labeling between Hiw and Dap160 reveals that the two proteins have largely overlapping distributions, further supporting the idea that Hiw protein is localized to subboutonic regions outside the active zones (Wan, 2000).

To examine the periactive zone localization of Hiw protein in greater detail, immuno-EM analysis was performed. Hiw protein is found to be associated with both the plasma membrane of boutons in all periactive zone areas and with clusters of vesicles outside active zones. However, active zones, T bars, and vesicles clustered at active zones are devoid of staining. Some Hiw staining is also observed associated with postsynaptic membranes opposite periactive zones, even though in situ hybridization analysis does not reveal muscle expression of hiw mRNA. In immuno-EM serial section reconstruction of boutons containing 20 active zones, the nonstaining active zones (which contained T bars and clusters of clear synaptic vesicles) measured 570 ± 50 nm, while the areas between them (the periactive zones) were all stained by the anti-HIW antibody and measured 1389 ± 144 nm. Given the variability of these measurements, the Hiw exclusion zone width of 570 ± 50 nm in the immuno-EM is in the same range as, but is a bit wider than, an independent assessment of active zone width in conventional EM of 469 ± 24 nm. It is not known whether this ~50 nm (~10%) difference on each side represents a real difference or whether the width of the active zone was simply underestimated by classic ultrastructural criteria (Wan, 2000).

Highwire function at the Drosophila neuromuscular junction: spatial, structural, and temporal requirements

Highwire is a large, evolutionarily conserved protein that is required to restrain synaptic growth and promote synaptic transmission at the Drosophila neuromuscular junction. Current models of highwire function suggest that it may act as a ubiquitin ligase to regulate synaptic development. However, it is not known in which cells highwire functions, whether its putative ligase domain is required for function, or whether highwire regulates the synapse during development or alternatively sets cell fate in the embryo. A series of transgenic rescue experiments were performed to test the spatial, structural, and temporal requirements for highwire function. Presynaptic activity of highwire is both necessary and sufficient to regulate both synapse morphology and physiology. The Highwire RING domain, which is postulated to function as an E3 ubiquitin ligase, is required for highwire function. In addition, highwire acts throughout larval development to regulate synaptic morphology and function. Finally, it is shown that the morphological and physiological phenotypes of highwire mutants have different dosage and temporal requirements for highwire, demonstrating that highwire may independently regulate the molecular pathways controlling synaptic growth and function (Wu, 2005).

Hiw is reported to localize to periactive zones, the regions surrounding but not including the active zone, in presynaptic terminals. It has been proposed that the periactive zone is a specialized domain that regulates synaptic growth because at least three proteins important for regulating synaptic growth localize there: Hiw, FasII, and Still life. However, the evidence that Hiw localizes to periactive zones comes from staining with the monoclonal antibody 6H4. 6H4 does not recognize Hiw. Synaptic staining with 6H4 persists in all hiw alleles, including excision mutants that delete the region of the protein against which the antibody was raised. In addition, 6H4 is unable to recognize misexpressed transgenic Hiw. To determine where Hiw is localized, sera were generated to 10 independent peptide and fusion protein antigens. Although a single serum was found that recognizes Hiw on an immunoblot, sera that work for immunocytochemistry were not generated. As such, the endogenous localization pattern of Hiw cannot be determined. GFP-tagged UAS-GFP-Hiw transgenes were generated that are fully capable of rescuing hiw mutants. The GFP-tagged Hiw does localize to synapses, although it does not extensively colocalize with synaptic vesicles, periactive zones, or active zones. The transgenic protein does appear to be enriched in synapses, because it is barely detectable in the preterminal axon. However, this overexpressed protein is visible in cell bodies and axon bundles. Similar results are found with GFP-tagged RPM-1, which is detected in synapses, axons, and cell bodies of C. elegans. It is concluded that Hiw is likely present at synapses (Wu, 2005).

A large number of hiw alleles have been generated, including many with stop codons. All of these nonsense alleles express truncated protein and indeed also express some full-length protein attributable to readthrough of the stop codon. This finding raises the possibility that no hiw alleles are true nulls. In fact, both screens that identified hiw alleles selected for viable mutants, which may have biased the screens against true null alleles. Two new excision mutants of hiw have been generated: one that deletes at least the N-terminal 2784 amino acids (hiwDeltaN), and another that appears to delete the C-terminal 2449 amino acids (hiwDeltaC). Although either allele could potentially express a truncated but partially functional protein, these are the first alleles that do not express any full-length protein. Like the previously identified alleles, these mutants are viable and have synaptic phenotypes that are essentially identical to a previously characterized genetic null allele, hiwND8. Therefore, the previously described hiw phenotype is likely the null phenotype (Wu, 2005).

RNA in situ analysis has demonstrated that hiw is abundantly expressed in the CNS, but that cannot exclude the possibility of lower-level expression in other tissues, such as muscle. In fact, in vertebrates, PAM has been detected in skeletal muscles. This study has addressed the problem of where hiw functions. Presynaptic hiw was found to be both necessary and sufficient for generating a morphologically and physiologically normal synapse. Postsynaptic expression has no detectable effect on synaptic development. These results are consistent with findings in worms and zebrafish. The conserved site of action for hiw supports the view that the mechanism of action of hiw may also be conserved (Wu, 2005).

It is likely that Hiw and its homologs function as E3 ubiquitin ligases. It contains a RING motif that can function as E3 ubiquitin-protein ligase in vitro, and this domain of the Hiw protein is the most highly conserved across species. In flies, hiw genetically interacts with the deubiquitinating protease fat facets, and, in worms, RPM-1 physically interacts with proteins that form an E3 ubiquitin ligase complex. In addition, mutations have been identified in flies and worms that are predicted to disrupt the RING domain; however, these mutant proteins may not be expressed. Therefore, it is unknown whether Hiw ligase activity is necessary for none, some, or all of the functions of hiw at the synapse. An hiw transgene encoding a mutant RING domain has been expressed. The mutations disrupt two conserved cysteine residues that are required for activity of RING domain ubiquitin ligases. This mutant protein is abundantly expressed, but cannot rescue any of the hiw mutant phenotype. Moreover, this RING mutant acts as a potent dominant negative. This suggests that the mutant protein is not grossly misfolded but instead likely interacts with Hiw-binding proteins. This supports the assertion that the RING domain, and ubiquitin ligase activity, is essential for hiw-dependent regulation of both synaptic morphology and transmission (Wu, 2005).

Because Hiw restrains synaptic growth, it was predicted that overexpression of Hiw could reduce synaptic complexity, generating small NMJs with few boutons. No such gain-of-function was observed, indicating that endogenous Hiw levels are not limiting for synaptic growth control. No gain-of-function phenotypes have been reported for rpm-1 either. These results are consistent with the finding that RPM-1 binds to a multiprotein ubiquitin ligase complex. Overexpressing a single member of that complex may not lead to increased function. In fact, the very weak dominant-negative phenotype caused by overexpression of wild-type Hiw suggests that excess Hiw may disrupt the stoichiometry of such a functional complex (Wu, 2005).

Attempts were made to distinguish between two models for the temporal requirement for hiw function. In hiw mutants, synaptic growth is excessive from the time of initial synapse formation. Hence, hiw may act exclusively during embryogenesis to set cell fate, programming the motoneuron for normal or excessive growth. Alternatively, hiw may function throughout development to regulate the synaptic growth rate of the motoneuron. This second model would allow for modification of synaptic growth during development and would be consistent with a function for hiw in activity-dependent synaptic plasticity. To investigate the temporal requirements for hiw, advantage was taken of the GeneSwitch Gal4 system for tissue-specific and temporally specific control of gene expression. It was found that expression of hiw after embryogenesis leads to a decrease in synaptic bouton number and a complete rescue of synaptic function. Hence, hiw not only acts in the embryo but can function throughout the larval growth period to control synaptic morphogenesis and transmission. It is noted that addition of hiw during larval development does not reverse the excess growth that occurred in the embryo and early larval stage. This may explain why similar studies in C. elegans found a requirement for rpm-1 early in synaptic development. hiw is required early in development, but the quantitative analysis of synaptic growth and function that is feasible at the Drosophila NMJ reveals that hiw also functions later in development (Wu, 2005).

Through manipulation of the timing of hiw expression, differential rescue of the morphological and physiological phenotypes of hiw was found. When hiw is expressed late, synaptic transmission is fully rescued, although there is still a twofold to threefold increase in bouton number. Hence, the physiological phenotype is not secondary to developmental defects at the synapse. The morphological and physiological phenotypes of hiw are genetically separable, because an hiw, fat facets double mutant shows a partial rescue of physiology with no rescue of morphology. However, the complete rescue of physiology by the late expression of hiw does more than show that morphology and physiology are independent: it raises the intriguing possibility that hiw functions as a short-term regulator of synaptic function. If hiw has an ongoing role in setting synaptic strength in addition to a developmental role controlling synaptic growth, it is in an ideal position to integrate and coordinate signals that regulate morphological and physiological synaptic plasticity (Wu, 2005).


Each abdominal hemisegment in the Drosophila embryo and larva has a stereotyped pattern of 30 muscles, each identifiable by its size, shape, body wall insertion position, and expression of molecular markers. About 45 motor neurons extend axons to innervate specific muscles in each hemisegment (Wan, 2000).

At the larval NMJ, there are two major types of synaptic terminals. Type I synapses have larger boutons and use glutamate as their neurotransmitter, while type II synapses have smaller boutons containing a variety of vesicles, including dense core vesicles filled with other transmitters, such as neuropeptides and biogenic amines. Type I boutons are surrounded postsynaptically by a conspicuous subsynaptic reticulum (SSR) consisting of multiple folds and invaginations of the muscle cell membrane. Type I synapses can be further divided into type Ib (big) and type Is (small) based on the sizes of their boutons. These synapses undergo dynamic growth during larval development. As the muscle increases its volume, there is a parallel increase in the number of synaptic boutons and the number and complexity of branches (Wan, 2000).

In a pilot effort to look for synaptic structural mutants, a collection of viable mutants on the X chromosome was screened, looking for structural defects at the NMJ of third instar larvae. These 230 mutant lines have various degrees of walking defects in adults. Antibodies against synaptotagmin (Syt) and Fas II were used as markers to label NMJ synaptic terminals of third instar larvae. One mutant line, hiw, has a dramatic synaptic structure phenotype. This initial hiwEMS mutant allele and all 12 subsequent hiwND mutant alleles, as well as 3 P element alleles, behave as viable recessive alleles. They are all viable as transheterozygotes with each other or with deficiencies that remove the gene (Wan, 2000).

In hiw mutants, the presynaptic boutons visualized by Syt staining are more numerous at all NMJs. Compared with wild-type synapses, hiw synapses have a greatly expanded branching pattern. The presynaptic terminal arbors contain many more Fas II-positive branches, the branches are longer, and the branch orders are increased. The synaptic area, the muscle surface area occupied by presynaptic structures, is also increase in hiw mutants. This synaptic overgrowth phenotype is 100% penetrant because expanded presynaptic arborizations and an increased number of boutons are observed at every NMJ in every segment of every mutant third instar animal (Wan, 2000).

Several aspects of the mutant phenotype were quantified. The number of type I boutons on muscles 6/7 and on muscle 4 were counted. hiw mutants have a 2-fold or greater increase in the number of boutons. After correcting for muscle area, hiw synapses also have a 2-fold increase in synaptic size compared with wild type. The combination of hiwEMS chromosome in trans over deficiencies show essentially identical phenotypes. Developmental curves of both hiw and wild-type larvae can be fitted with similar types of power functions, but hiw synapses have a larger synaptic size at first instar and grow at a higher rate during further larval stages. On average, there is a decrease of bouton size in hiw mutants. Given the dramatic increase in the number of boutons, this decrease in bouton size might reflect some limitation of one or more structural components as a result of oversprouting. Type Ib and type Is boutons still can be distinguished from one another in hiw mutants, although both are smaller than their wild-type counterparts. Type II boutons and branches seem more numerous in hiw mutants. However, this phenotype is less dramatic and more difficult to score, and thus was not quantitated (Wan, 2000).

To compare branching, three aspects of the branching pattern on muscle 12 were measured. Muscle 12 was chosen for this analysis because its innervation is relatively planar and accessible, and the branching pattern of that innervation is relatively simple in wild-type larvae, making it easier to quantitate branch number and complexity. First, the total length of this synapse (of only those branches containing type I boutons) was examined as a function of the length of the muscle. From the first larval instar stage onward, the synaptic length in hiw mutants is greater than in wild type. This relationship is maintained throughout larval development (Wan, 2000).

Second, the length of the two longest branches was measured. These branches are more than 2-fold longer in hiw mutants compared with wild type. Third, the branching pattern of this synapse was examined. The branches containing only type I boutons were divided into different groups based on their branch order and the number of branches were counted. There is a dramatic increase in the complexity of branches at hiw synapses; mutant synapses have many more branches and higher order branches. The average number of branches is 16.5 at hiw synapses on muscle 12, compared with 7.7 at the same synapse in third instar wild-type larvae. hiw synapses on muscle 12 often have several fifth order branches, something that is rarely observed in wild type (Wan, 2000).

Although the morphology of NMJ synapses is dramatically expanded in hiw mutants, the specificity of connections is not altered. The initial motor axon pathfinding and target selection appear normal. Antibodies that label both CNS and peripheral axon pathways (anti–Fas II and BP102 mAbs) reveal a normal pattern of axon pathways and a normal pattern of muscle innervation. Longer and larger presynaptic terminal branches were observed on the surfaces of muscles in hiw mutants toward the end of embryogenesis, but the specificity of contacts was normal. It appears that hiw mutants display the synaptic overgrowth phenotype from the onset of synaptogenesis during embryogenesis. This is consistent with the observation that hiw first instar larvae, even 30 min after hatching, already display a dramatic increase in bouton number (34 versus 17 in wild type at muscles 6/7) and increase in synaptic span (56% of muscle length in hiw versus 27% in wild type at muscle 12) (Wan, 2000).

To what extent are synaptic components localized normally in hiw mutant synapses? Postsynaptic glutamate receptors appear to be correctly localized just opposite active zones as determined by expression of the MHC-myc-DGluRIIA and MHC-myc-DGluRIIB transgenes in hiw mutant larvae. The presynaptic protein Syt and the synaptic CAM Fas II are expressed normally at hiw mutant synapses. Several other synaptic proteins (Discs-large [DLG], Pak, and Dap160) are also localized correctly in hiw mutants. Thus, many of the molecular events associated with synapse formation, including clustering of postsynaptic receptors and organization of presynaptic vesicles, appear normal in hiw mutants. This is consistent with the observation that hiw mutants are viable and are not severely impaired in motor behaviors (Wan, 2000).

To further examine the relationship of FasII and hiw, a test was performed to see whether overexpression of Fas II either pre- or post-synaptically can rescue the hiw mutant phenotype. Fas II overexpression cannot rescue the hiw mutant phenotype. This suggests that FasII is not downstream of hiw in the genetic control of synaptic growth (Wan, 2000).

Given the dramatic structural phenotype observed at hiw NMJ synapses, it was of interest to see whether the ultrastructure of these mutant synapses is affected. Serial sections of both hiw and wild-type boutons at muscle 6 (segment A3) of third instar larvae were examined in the electron microscope (EM). It was found that the hiw mutant synapses have active zones with T bars, clusters of synaptic vesicles, and other features that appear wild-type. The postsynaptic SSR also appears normal in size and shape. Some of these characteristics were quantitated from serial sections of 9 hiw boutons and they were compared with similar data from 18 wild-type boutons. Although hiw boutons are, on average, smaller than wild-type boutons, they nevertheless have the same relative bouton surface area per active zone. In addition, the average maximum width of individual active zones (as measured by membrane apposition, location of T bars, and clusters of synaptic vesicles) was 441 in hiw compared with 469 in wild type. Finally, the number of T bars per active zone is statistically similar. It is concluded from these data that the fundamental organization, size, and spacing of active zones in hiw mutants is relatively normal (Wan, 2000).

Given that hiw synapses, although greatly expanded, appear normal ultrastructurally, it was of interest to see whether synaptic transmission is altered at these mutant synapses. One might imagine that the expanded structure would lead to a stronger synapse. Surprisingly, it was found that these expanded synapses are actually weaker physiologically. Intracellular recordings were performed from muscle 6 (segment A3) of third instar larvae. This muscle is innervated by type I glutamatergic boutons from motor neurons RP3 and 6/7b. The postsynaptic responses to both spontaneous and evoked transmitter release were measured. The mean amplitude of spontaneous miniature junctional potentials (mEJPs), also called quantal size, is typically considered a measure of postsynaptic sensitivity to transmitter (in principle, it might also reflect the amount of neurotransmitter packed into each synaptic vesicle, although this is less likely). In contrast, the mean amplitude of the excitatory junctional potential (EJP) depends on both the postsynaptic receptor function and the amount of transmitter released from the presynaptic terminal (Wan, 2000).

The average mEJP amplitude is only slightly reduced in hiw mutants (0.65 ± 0.03 mV versus 0.91 ± 0.04 mV in wild type). However, the evoked release is greatly reduced; the mutant shows a 66% decrease compared with wild type (7.75 ± 0.86 mV versus 22.58 ± 1.73 mV, respectively). Thus, the quantal content (number of vesicles released) of the mutant synapses (13.18 ± 1.41, which is estimated by dividing the mean EJP amplitude by the mean mEJP amplitude and adjusting for nonlinear summation), is about 34% that of wild type (39.22 ± 4.64). Quantal content is an indication of presynaptic function. Since an obvious reduction in synaptic vesicle distribution was not observed in hiw mutants, the reduction in quantal content suggests more subtle defects in the release mechanism (Wan, 2000).

As a first step to gaining insight into the mechanisms of Hiw function, transgenic flies carrying different Hiw fragments were generated. One such transgene (pUAS-HIWC1-LD) contains a Hiw fragment between amino acids 2418 and 3461, fused to enhanced green fluorescent protein (EGFP). When this transgene was expressed panneurally using elav-Gal4 in an otherwise wild-type background, expanded synaptic structures were observed at the NMJ of third instar larvae. This phenotype is qualitatively similar to that of hiw loss-of-function mutants, although quantitatively it is not as severe. When the same transgene was expressed in muscles using the panmuscle 24B-Gal4 enhancer trap line, no phenotype at the NMJ was observed. This result further suggests that Hiw functions on the presynaptic side to control synaptic growth (Wan, 2000).

Genetic interaction between highwire and fat facets

The covalent attachment of ubiquitin to cellular proteins is a powerful mechanism for controlling protein activity and localization. Ubiquitination is a reversible modification promoted by ubiquitin ligases and antagonized by deubiquitinating proteases. Ubiquitin-dependent mechanisms regulate many important processes including cell-cycle progression, apoptosis and transcriptional regulation. Ubiquitin-dependent mechanisms regulate synaptic development at the Drosophila neuromuscular junction (NMJ). Neuronal overexpression of the deubiquitinating protease Fat facets leads to a profound disruption of synaptic growth control; there is a large increase in the number of synaptic boutons, an elaboration of the synaptic branching pattern, and a disruption of synaptic function. Antagonizing the ubiquitination pathway in neurons by expression of the yeast deubiquitinating protease UBP2 also produces synaptic overgrowth and dysfunction. Genetic interactions between fat facets and highwire, a negative regulator of synaptic growth that has structural homology to a family of ubiquitin ligases, suggest that synaptic development may be controlled by the balance between positive and negative regulators of ubiquitination (Diantonio, 2001).

Synaptic morphology is dynamic; once formed, synapses expand, retract, and remodel throughout life. This plasticity underlies the refinement of neuronal circuits during development and may be critical for plasticity in the adult brain. To identify molecular mechanisms regulating the morphological growth of synapses, a genetic screen was performed for molecules whose neuronal overexpression disrupts synaptic growth control at the Drosophila NMJ. A collection of flies capable of the targeted overexpression of endogenous Drosophila genes was screened and two lines, EP(3)381 and EP(3)3520, were identified whose overexpression in the nervous system leads to synaptic overgrowth. Both EP(3)381 and EP(3)3520 overexpress fat facets (faf), a deubiquitinating protease. Endogenous faf transcript is strongly and widely expressed in the developing central nervous system (CNS), demonstrating that neuronal expression of faf from the EP lines produces overexpression, not misexpression, of the transcript (Diantonio, 2001).

Anatomical analysis at the NMJ reveals that neuronal expression from both EP(3)381 and EP(3)3520 leads to an increase both in the number of synaptic boutons and in the synaptic span (the extent of the muscle covered by the synapse). This increase is not seen in flies that do not overexpress faf v or that overexpress a non-functional faf gene (elav-Gal4 crossed to EP(3)381faf-). Neuronal overexpression of faf also causes an increase in the number of synaptic branches as quantified by the number of branch points. Postsynaptic expression of faf does not affect synaptic morphology (Diantonio, 2001).

To assess the physiological consequence of neuronal faf overexpression, both spontaneous and evoked neurotransmitter release were analysed at muscle 6 of third instar larvae. Despite the greatly expanded size of the NMJ with faf overexpression, the amplitude of evoked excitatory junctional potentials (EJPs) is markedly reduced. Given that the amplitude of miniature EJPs (mEJPs) shows only a small, albeit significant, reduction, a large decrease was measured in quantal content (the number of vesicles released by the nerve) as measured by dividing the EJP amplitude by the mEJP amplitude. Neuronal overexpression of faf also leads to a reduction in the frequency of spontaneous mEJPs. The reduction in both quantal content and mEJP frequency indicates a presynaptic defect in neurotransmitter release. Other presynaptic mutants with even greater reductions in quantal content do not show a structural overgrowth, thus the anatomical phenotype described above is probably a direct consequence of faf overexpression and not a secondary consequence of this physiological phenotype (Diantonio, 2001).

faf antagonizes ubiquitin-dependent mechanisms by deubiquitinating target proteins. Alterations in synaptic structure and function owing to overexpression of faf suggest that ubiquitin-dependent mechanisms normally act to regulate the developing synapse. However, faf, a characterized deubiquitinating protease, might have other functions. To investigate the role of deubiquitination in the regulation of synaptic development, transgenic flies were generated capable of the targeted overexpression of the yeast deubiquitinating protease UBP2. This enzyme antagonizes ubiquitin-dependent mechanisms in yeast and has overlapping substrate specificity with FAF+9. Overexpression of yeast UBP2 in the nervous system of Drosophila leads to marked synaptic overgrowth and a severe reduction in presynaptic transmitter release. This phenotype is very similar to that seen with faf overexpression. Hence, antagonizing ubiquitin-dependent mechanisms by overexpression of deubiquitinating proteases markedly affects synaptic development (Diantonio, 2001).

To identify molecular pathways regulated by faf, a genetic interaction screen was performed to identify genes that enhance the faf overexpression phenotype. The X chromosome was screened for viable mutations that are lethal in combination with neuronal overexpession of faf. 7,000 chromosomes were screened and 15 lethal enhancers, 12 of which form one complementation group, were identifed. These 12 mutants are alleles of the highwire (hiw) gene and share the synaptic overgrowth phenotype described for loss-of-function hiw mutants. The hiw loss-of-function phenotypes are very similar to the faf gain-of-function phenotypes described here, with a large increase in the number of synaptic boutons, branches, and synaptic span, a small decrease in quantal size, and a large decrease in quantal content. The hiw transcript encodes a greater than 5,000 amino-acid protein that is localized to synapses and that contains a RING-H2 finger, a domain recently identified in a large family of E3 ubiquitin ligases. Hence, a potential synaptic E3 ligase was identified as a lethal enhancer of neuronal overexpression of faf. This genetic interaction provides further evidence that ubiquitination may have a central role in regulating synaptic growth and function (Diantonio, 2001).

To further investigate the genetic relationship between hiw and faf, double mutants were generated between loss-of-function alleles of hiw and faf. faf loss-of-function mutants have phenotypes in the developing eye and female germ line. In faf mutants, no defects were found in either synaptic morphology or function, possibly due to genetic redundancy between faf and one of the 17 other putative deubiquitinating proteases in the Drosophila genome. Although no phenotype was found for faf mutants in otherwise wild-type flies, in the sensitized background of a hiw mutant a requirement for faf is found. Two different loss-of-function alleles of faf both suppress the physiological phenotype of hiw, leading to a more than doubling of both quantal content and mEJP frequency. Hence endogenous faf activity acts to inhibit neurotransmitter release in a hiw background, much as increased faf activity inhibits neurotransmitter release in an otherwise wild-type background. Mutants of faf do not suppress the synaptic overgrowth seen in hiw, indicating that the physiological and morphological phenotypes in hiw are genetically separable. This suggests that either these phenotypes are mediated by different hiw substrates, or the anatomical phenotype is more sensitive to disruption of ubiquitin-dependent mechanisms. Finally, the inability to suppress the small quantal size defect suggests that this phenotype, seen in both hiw mutants and with overexpression of faf, may be a secondary consequence of synaptic overgrowth (Diantonio, 2001).

Decreases in postsynaptic activity induce a compensatory increase in presynaptic transmitter release, demonstrating that a homeostatic mechanism regulates synaptic strength during development. To assess the relationship between homeostatic and ubiquitin-dependent regulation, glutamate receptor function was disrupted in a hiw mutant. Postsynaptic expression of a dominant negative glutamate receptor (DgluRIIA-M/R) in a hiw mutant leads to an 18% decrease in quantal size. Since homeostatic compensation still occurs in a hiw mutant, it is suggested that homeostatic and ubiquitin-dependent regulation are mechanistically distinct (Diantonio, 2001).

The data presented here indicate that ubiquitin-dependent mechanisms regulate synaptic development at the Drosophila NMJ and suggest that a balance between positive and negative regulators of ubiquitination controls the structure and function of the synapse. Antagonizing ubiquitination by the neuronal overexpression of the deubiquitinating proteases faf or yeast UBP2 leads to synaptic overgrowth and defects in neurotransmitter release. This phenotype is very similar to the loss-of-function phenotype of hiw, a putative synaptic E3 ubiquitin ligase. Gain-of-function mutants of faf enhance hiw and loss-of-function alleles of faf suppress hiw. It is proposed that hiw-dependent ubiquitination controls the level or activity of critical regulatory molecules at the synapse, and that these molecules can be deubiquitinated by faf and other deubiquitinating proteases. Ubiquitinated proteins have been identified at mammalian synapses, and ubiquitin-processing enzymes can regulate long-term potentiation and facilitation, therefore control of ubiquitination by molecules such as HIW and FAF could be a widely used mechanism for regulating synaptic growth and function (Diantonio, 2001).

A mechanism distinct from highwire for the Drosophila ubiquitin conjugase Bendless in synaptic growth and maturation

The signaling mechanisms that allow the conversion of a growth cone into a mature and stable synapse are yet to be completely understood. Ubiquitination plays key regulatory roles in synaptic development and may be involved in this process. Previous studies identified the Drosophila ubiquitin conjugase bendless (ben) as important for central synapse formation, but the precise role it plays has not been elucidated. These studies indicate that Ben plays a pivotal role in synaptic growth and maturation. An incipient synapse is present with a high penetrance in ben mutants, suggesting that Ben is required for a developmental step after target recognition. Cell-autonomous rescue experiments were used to show that Ben has a presynaptic role in synapse growth. The TARGET system was harnessed to transiently express UAS-ben in a ben mutant background and a well defined critical period for Ben function was identified in establishing a full-grown, mature synaptic terminal. The protein must be present at a time point before but not during the actual growth process. Phenotypic evidence is provided demonstrating that Ben is not a part of the signal transduction pathway involving the well characterized ubiquitin ligase Highwire. It is concluded that Bendless functions as a novel developmental switch that permits the transition from axonal growth and incipient synapse formation to synaptic growth and maturation in the CNS (Uthaman, 2008).

The results from this study have given new insights into how ubiquitin system components establish functional synaptic connections. The temporal analysis of Bendless has been critical in illustrating its role as a developmental 'switch' in converting a growth cone into a mature synapse. As mentioned previously, the ben mutation is the result of a single amino acid change in the conserved catalytic core of the conjugase domain of the protein. This highlights the fact that the conjugase activity of the protein is necessary for the observed synaptic phenotype. Analysis of synaptic growth in Drosophila has primarily been done at the peripheral synapse of the NMJ. Components of the ubiquitin system, such as the ubiquitin ligase hiw, the deubiquitinating protease faf, and the synapse-associated E3 ligase PDZRN3, are known to play important roles in the growth and function of the fly NMJ. Significant studies have been performed with particular regard to the conserved family of hiw ubiquitin ligases. In Drosophila, hiw functions as a negative regulator of synapse development as mutants exhibit dramatic synaptic overgrowth at the larval NMJ. In Caenorhabditis elegans, loss of function of the hiw homolog rpm-1 results in multiple phenotypes at the NMJ as well as in the CNS. At the NMJ, some NMJs exhibit enlarged presynaptic terminals containing multiple active zones, whereas others contain underdeveloped or absent presynaptic terminals. In the worm mechanosensory circuit, the sensory neurons were found to retract synaptic branches, extend ectopic axons, and fail to accumulate synaptic vesicles, whereas some of the motor neurons exhibited phenotypes such as altered synaptic organization, branching, and overgrowth. Downstream signaling components have been isolated for both hiw and rpm-1 in Drosophila and C. elegans, respectively, and a number of conserved elements have been identified. Mutations in homologs of hiw in zebrafish and mice are also known to cause a variety of synaptic disruptions (Uthaman, 2008).

Ben and Hiw play distinct roles in synapse growth. This study has analyzed the novel roles played by these ubiquitin system components at the giant fiber system (GFS) central synapse. ben and hiw loss of function result in very different phenotypes, with ben specimens exhibiting synaptic undergrowth and hiw specimens exhibiting synaptic overgrowth. Ben function does not involve JNK, a well characterized downstream signaling partner identified for Hiw in Drosophila. It is also interesting to compare and contrast the role Hiw plays at a peripheral synapse with the role Ben plays at a central synapse. hiw mutants exhibit a presynaptic overgrowth phenotype at the NMJ, whereas ben mutants exhibit a reduction in presynaptic growth in the CNS. Also, Hiw does not localize to the nucleus and was found to regulate synaptic growth throughout development, whereas Ben has nuclear as well as cytosolic localization and only functions in a critical time period. Finally, Hiw activity is associated with the bone morphogenetic protein (BMP) retrograde signaling pathway that is known to be dependent on the retrograde motor. No evidence was found that Bendless function is dependent on the retrograde motor. All these data underline the fact that there are distinct targets for the ubiquitination cascades involving Ben and Hiw (Uthaman, 2008).

Functional neuronal circuits are established through a series of events: neurite outgrowth, axon guidance, target recognition, synapse formation, and synaptic growth and maturation. When the bendless mutant was originally characterized, Ben was thought to play an important role in either axon guidance or target recognition (Muralidhar, 1993; Oh, 1994). Analysis of the ben mutant clearly shows that Ben has an important role in synaptic growth. A number of specimens exhibit dye coupling between the GF and the motorneuron dendrite demonstrating that an incipient synapse is still formed and that the mutant phenotype arises from a failure of this immature connection to grow into a mature synapse. In addition, both gap junctional and chemical components are present at ben mutant terminals with synaptic vesicle marker localization as well as ultrastructural analyses (Uthaman, 2008).

The current view of synapse formation is that a nascent synapse can be rapidly assembled from material present in a growth cone in prepackaged vesicles and packets. After this primary rapid assembly of a nascent synapse, a secondary slower growth and maturation process takes place to result in a stable mature synapse. An insightful study on the Drosophila kinesin immaculate connections (imac) has shown it to be a permissive regulator of presynaptic maturation at the larval NMJ. Imac was found to be involved in the anterograde transport of synaptic vesicle precursors to the tip of the growth cone, an initial stage of synaptogenesis. In ben specimens, synaptic vesicles are still transported all the way down to the tip of the truncated terminal as evidenced by the localization of GFP-tagged synaptotagmin and synaptobrevin. The data strongly suggest that the ben mutant phenotype is resultant at a point after synaptic vesicular transport. Hence, it is concluded that the bendless terminal is an incipient synapse that fails to grow and mature and that Ben is a permissive regulator whose function is required for the initiation of a secondary process in presynaptic growth (Uthaman, 2008).

It is counterintuitive that, although Bendless is required for synaptic growth and maturation, the data show that it is not required during the growth process. This highlights the important role Ben plays as a developmental switch. Transient expression of UAS-ben before the growth of the presynaptic terminal was sufficient to rescue the ben phenotype anatomically and physiologically, but expression during the growth period had no effect. This suggests that Bendless is not involved in the actual growth process but rather has to be present in advance to alter signaling and initiate changes that allow growth to take place. Here it is essential to differentiate between axonal and synaptic growth, because it has been determined previously that axonal growth is unaffected in ben mutants (Muralidhar, 1993). Hence, Ben function is required to permit axonal growth to switch to synaptic growth (Uthaman, 2008).

The molecules in the signaling pathway of this novel mechanism remain to be further investigated. In conclusion, tight spatial and temporal control of synaptic connectivity in the nervous system is undoubtedly crucial to normal function. Determining how exactly Bendless regulates the formation of a mature synapse will give future novel insights into this phenomenon (Uthaman, 2008).

The Drosophila epsin 1 is required for ubiquitin-dependent synaptic growth and function but not for synaptic vesicle recycling

The ubiquitin-proteasome system plays an important role in synaptic development and function. However, many components of this system, and how they act to affect synapses, are still not well understood. This study used the Drosophila neuromuscular junction to study the in vivo function of Liquid facets (Lqf), a homolog of mammalian epsin 1. The data show that Lqf plays a novel role in synapse development and function. Contrary to prior models, Lqf is not required for clathrin-mediated endocytosis of synaptic vesicles. Lqf is required to maintain bouton size and shape and to sustain synapse growth by acting as a specific substrate of the deubiquitinating enzyme Fat facets. However, Lqf is not a substrate of the Highwire (Hiw) E3 ubiquitin ligase; neither is it required for synapse overgrowth in hiw mutants. Interestingly, Lqf converges on the Hiw pathway by negatively regulating transmitter release in the hiw mutant. These observations demonstrate that Lqf plays distinct roles in two ubiquitin pathways to regulate structural and functional plasticity of the synapse (Bao, 2008).

One important finding from this study is that Lqf does not play a detectable role in SV endocytosis. Multiple lines of evidence obtained from electrophysiological, ultrastructural and optical imaging studies support this conclusion. This is the first in vivo study of Lqf or epsin 1 on SV recycling. The finding is also clearly surprising given that epsin 1 has been highly implicated to play a key role in the initiation of clathrin-coated vesicle formation and endocytosis. Does the observation reflect the special property of the fly NMJ? Lqf lacking either the ENTH domain or the clathrin-interacting C-terminus has been shown to rescue the mutant phenotype in the developing eye. These rescue results are intriguing, but they do not readily support a specific role for Lqf in CME. In particular, there are no clear mechanisms on how these truncated fragments could fulfill Lqf's clathrin-dependent functions. Interestingly, RNA interference and small interfering RNA-induced knockdown of epsin 1 fails to block the internalization of EGF receptors in HeLa cells. There is also evidence that epsin 1 functions only in clathrin-independent endocytosis. Furthermore, Lqf has been shown to be required for endocytosis of select receptors but not of all receptors. More importantly, Lqf itself is not required for receptor-mediated endocytosis. Rather, Lqf appears to signal select ligands (such as Delta/Serrate/Lag2) for internalization or recycling. Hence, these studies lend strong support to observations that Lqf does not play a significant role in CME of SVs (Bao, 2008).

It should be noted that recent studies reveal that the epsin 1-interacting protein Eps15 is required for SV recycling in both C. elegans and Drosophila. However, Eps15 is required to maintain the level of endocytotic proteins in nerve terminals. Strikingly, key endocytotic proteins such as Dynamin and Dap160 are reduced in synaptic boutons by ∼90% and ∼80%, respectively, in eps15 mutants. These observations make it difficult, if not impossible, to assign a direct role for Eps15 in CME (Bao, 2008).

Synapse development is a highly regulated process involving a large number of molecules. The first suggestion that Lqf could have a potential role in synapse development came from studies of its deubiquitinating enzyme Faf. This notion was further supported by a direct biochemical demonstration that Lqf is a specific substrate of Faf. The current studies provide the first experimental test of this hypothesis by showing that Lqf acts downstream of Faf in promoting synaptic overgrowth. This effect on NMJ growth appears to be Faf dependent as lqf mutations alone do not dramatically affect bouton numbers. It is interesting to note that neuronal overexpression of Lqf promotes bouton budding but does not mimic the exuberant synaptic overgrowth induced by overexpression of Faf. Hence, it is suggested that Lqf is necessary but insufficient for synaptic overgrowth, raising the possibility that Lqf is not the only substrate of Faf in motoneurons (Bao, 2008).

Another important finding emerging from this study is that two distinct UPS pathways may be employed at the Drosophila larval NMJ to regulate synapse growth. The Hiw/RPM-1/Phr1 proteins have a conserved role in inhibiting presynaptic development in Drosophila, C. elegans and mammals. In C. elegans and Drosophila, the substrates of RPM-1/Hiw have been shown to be MAP kinases and MAPKKK. The current study indicates that Lqf is unlikely a substrate of Hiw in conditioning synaptic growth. In contrast, this study shows that the Faf pathway is a positive regulator of synaptic growth at the NMJ in which Lqf is an essential substrate. Hence, it is suggested that Hiw and Faf/Lqf are two distinct UPS pathways that regulate synapse development in Drosophila (Bao, 2008).

However, the relationship between the Faf and Hiw pathways in synapse development is rather complex. Intriguingly, the MAPKKK Wnd is required for synaptic overgrowth mediated by both Hiw and Faf pathways. One possibility is that Wnd acts downstream of Lqf to fulfill the function of both the Hiw and the Faf pathways. However, this idea is inconsistent with the observation that unlike lqf mutants, the wnd null mutant itself has no morphological or electrophysiological defect. More importantly, wnd mutations do not suppress the transmitter release defect seen in the hiw mutant, whereas the lqf mutant does. Alternatively, it is suggested that Hiw and Faf act through two parallel pathways and that the suppression of Faf-induced overgrowth by the wnd mutation may be mediated by Fos/Jun kinase signaling. Based on the observation that overexpression of Ubp2A increases neuronal Wnd levels, it is possible that Faf may also use Wnd as a substrate for synaptic overgrowth. However, this has yet to be tested experimentally (Bao, 2008).

Recent genetic studies have revealed an interesting feature of synapse growth and function that closely depends on protein turnover by specific UPS pathways. In Drosophila, faf or lqf mutations are capable of partially suppressing the defect in transmitter release in hiw mutants. This partial suppression is specific and should not be viewed simply as a reduction of transmitter release in faf or lqf mutant backgrounds by hiw mutations. If there were no partial suppression by faf or lqf mutations, the amplitude of EJPs would be similar to that in hiw single mutants. Because faf null mutations reduce Lqf levels, it is reasonable to suggest that Lqf acts downstream of Faf to inhibit synaptic transmission in hiw mutants. Unlike the functional interactions with hiw, however, faf or lqf mutations do not affect synaptic overgrowth in hiw mutants. Differing from lqf and faf mutations, wnd mutations fully suppress synaptic overgrowth but do not affect synaptic physiology in the hiw mutant. Hence, different ubiquitin pathways can specifically dissociate synapse growth from function (Bao, 2008).

The physiological stimuli involved in such selective modulation of synapse growth and function remain to be identified. Given the conserved role of the ubiquitin-proteasome system in synaptic plasticity across animal species, the findings reported in this study may have general neurobiological implications. In particular, it is noted that the Faf homolog in mouse, Usp9x or Fam is differentially expressed in different regions of the brain. Such spatial distribution patterns may provide a means for Usp9x to locally regulate synaptic function. Importantly, Usp9x is localized at synapses, where calcium influx rapidly regulates its enzymatic activity and deubiquitination of epsin 1. Hence, Faf and Lqf/epsin 1 are good candidate mediators of activity-dependent synaptic plasticity (Bao, 2008).

Autophagy promotes synapse development in Drosophila

Autophagy, a lysosome-dependent degradation mechanism, mediates many biological processes, including cellular stress responses and neuroprotection. This study demonstrates that autophagy positively regulates development of the Drosophila larval neuromuscular junction (NMJ). Autophagy induces an NMJ overgrowth phenotype closely resembling that of highwire (hiw), an E3 ubiquitin ligase mutant. Moreover, like hiw, autophagy-induced NMJ overgrowth is suppressed by wallenda (wnd) and by a dominant-negative c-Jun NH2-terminal kinase (bskDN). Autophagy promotes NMJ growth by reducing Hiw levels. Thus, autophagy and the ubiquitin-proteasome system converge in regulating synaptic development. Because autophagy is triggered in response to many environmental cues, these findings suggest that it is perfectly positioned to link environmental conditions with synaptic growth and plasticity (Shen, 2009).

Autophagy involves multiple steps, including induction, autophagosome formation, fusion of autophagosomes with lysosomes, and recycling of autophagy components. Disrupting any of these steps impairs autophagy. Several highly conserved ATG genes encoding core components of the autophagy machinery have been identified in yeast. Mutations in genes, including atg1, -2, -6, and -18, have also been isolated and characterized in Drosophila. To assess the role of autophagy in NMJ development, the effects were examined of mutations in atg genes, whose normal functions span the entire process: atg1 is defective in autophagy induction, atg6 is defective in autophagosome formation, and atg2 and -18 are defective in retrieval of other ATG proteins from autophagosomes. Regardless of the step impaired, all of these atg mutants exhibited significant reduction in NMJ size. These results demonstrate that a basal level of autophagy is required to promote NMJ development (Shen, 2009).

Overexpression of atg1+ is sufficient to induce high levels of autophagy in larval fat bodies and salivary glands. If autophagy is a positive regulator of NMJ development, an increase in autophagy might enhance synaptic growth. Consistent with previous studies, panneuronal overexpression of UAS-atg1+ under the control of C155-Gal4 or elav-Gal4 drivers induced high levels of autophagy in the nervous system, as indicated by increased staining with LysoTracker, an acidophilic dye which has been used to assess autophagy by labeling acidic structures, including lysosomes. Under these conditions, bouton number increased more than twofold. To further verify that this NMJ overgrowth was caused by elevated autophagy rather than to some other effect of atg1+ overexpression, whether mutations in other atg genes suppress this phenotype was examined. For this purpose, a null allele of atg18 (atg18δ) was generated. Removal of one copy of atg18+ had no affect on NMJ growth in an otherwise wild-type background but significantly suppressed NMJ overgrowth caused by atg1+ overexpression. Removal of both copies of atg18+ conferred almost complete suppression. Therefore, NMJ overgrowth caused by atg1+ overexpression is primarily caused by elevated levels of autophagy (Shen, 2009).

As a further test, NMJ morphology was examined after feeding larvae with rapamycin, which induces autophagy by inhibiting TOR (target of rapamycin), the key negative regulator of autophagy. Wild-type larvae fed rapamycin exhibited striking NMJ overgrowth similar to that caused by overexpressing atg1+. Rapamycin-induced NMJ overgrowth was completely suppressed by mutations in atg18. Collectively, these results demonstrate that autophagy is a key positive regulator of NMJ growth (Shen, 2009).

Wairkar (2009) observed NMJ undergrowth in atg1 mutants but did not see overgrowth with atg1+ overexpression. This discrepancy likely results from the use of different UAS-atg1+ transgenes. For example, Wairkar was able to obtain only partial (~50%) rescue of NMJ undergrowth in atg1 mutants by overexpression of their UAS-atg1rescue construct, whereas this study obtained complete rescue of this phenotype (Shen, 2009).

Atg1 has several functions unrelated to autophagy. It was found that axonal transport is disrupted in atg1-null mutants, which is a result also recently reported by Toda (2008) and Wairkar (2009). In addition, Atg1 suppresses translation by inhibiting the S6K kinase (Lee, 2007; Scott, 2007) and controls active zone density by inhibiting extracellular signal-regulated kinase (ERK) signaling (Wairkar, 2009). However, several lines of evidence indicate that these functions of Atg1 are not responsible for the NMJ phenotypes observed when Atg1 activity was altered. (1) atg2 or -18 mutants exhibited similar NMJ undergrowth but did not have defects in axonal transport. Thus, in agreement with Toda (2008), it is concluded that Atg1's role in axonal transport is distinct from its function in autophagy and NMJ growth. (2) Blocking or activating translation by overexpressing a dominant-negative S6K transgene or constitutively activated S6K transgenes by elav-Gal4 driver had little affect on NMJ growth. Moreover, coexpression of any of the three constitutively activated S6K transgenes failed to suppress NMJ overgrowth caused by atg1+ overexpression. Thus, the role of Atg1 in S6K-dependent translation does not contribute to the NMJ phenotypes associated with manipulations of Atg1. (3) An ERK mutation does not affect NMJ growth. Although this ERK mutation suppresses the deficit in active zone density in atg1 mutants, it does not suppress atg1's NMJ undergrowth phenotype (Wairkar, 2009), indicating that it is not mediated by the ERK pathway. Collectively, these results demonstrate that altered levels of autophagy are primarily responsible for the effects of Atg1 on NMJ development (Shen, 2009).

NMJ overgrowth induced by autophagy is distinctive and offers potential clues about pathways that may be involved. Formation of multiple long synaptic branches containing many small diameter boutons without any hyperbudding or satellite boutons most closely resembles the hiw phenotype, suggesting that autophagy and Hiw may function through the same pathway. Recent evidence indicates that Hiw inhibits NMJ growth by down-regulating Wnd, which in turn activates a Jun kinase encoded by bsk (basket). NMJ overgrowth in hiw is suppressed by mutations of wnd and by a dominant-negative mutation of bsk (bskDN; Collins, 2006). If the phenotypic similarity between hiw and increased autophagy reflects convergence on a common pathway, autophagy-induced NMJ overgrowth should also be suppressed by wnd and bskDN. Indeed, this is what was observed in this study. These results strongly support the idea that autophagy and Hiw converge on a Wnd-dependent MAPK signaling pathway to regulate NMJ development (Shen, 2009).

If autophagy and Hiw act via a common pathway, where do they converge? As a positive regulator of NMJ growth, autophagy could promote degradation of a negative regulator. An intriguing possibility is that Hiw is the negative regulator affected by autophagy. If a decrease in Hiw levels is responsible for NMJ overgrowth when autophagy is elevated, restoration of Hiw should suppress overgrowth. This possibility was tested by coexpressing wild-type Hiw with Atg1; Atg1-mediated NMJ overgrowth was found to be significantly suppressed. This suppression is not simply an indirect consequence of the dilution of GAL4 caused by addition of a second UAS element because coexpression of UAS-nwk+ did not suppress such NMJ overgrowth. This result also shows that Nwk (Nervous wreck), another negative regulator of NMJ growth, is not an apparent target of autophagy, as predicted by differences in phenotypes. Thus, autophagy appears to regulate NMJ growth through its effects on particular presynaptic proteins, and Hiw represents a key downstream effector (Shen, 2009).

To further test whether autophagy promotes NMJ growth by limiting Hiw, one copy of hiw+ was eliminated to determine whether this further decrease in Hiw levels enhanced the effects of atg1+ overexpression. In an otherwise wild-type background, loss of one copy of hiw+ had no affect, but it significantly enhanced atg1+-induced NMJ overgrowth. The phenotype of hiw homozygotes overexpressing atg1+ was no more extreme than hiw homozygote alone. The absence of any additive or synergistic effects further supports the hypothesis that autophagy promotes NMJ development by down-regulating Hiw (Shen, 2009).

Because Hiw antibodies do not work for immunohistochemistry, Hiw was visualized using a fully functional GFP-tagged construct to test directly whether abundance of Hiw is affected by autophagy. In an otherwise wild-type background, Hiw-GFP was strongly expressed in neurons throughout the ventral ganglion and brain lobes driven by C155-Gal4, as detected by anti-GFP staining. However, in larvae co-overexpressing atg1+, the GFP signal was reduced by ~60% relative to anti-HRP staining. This result was confirmed by Western blot analysis. Reduction of Hiw-GFP is not caused by the dilution of GAL4 by the presence of a second UAS element because coexpression of UAS-myr-RFP did not affect abundance of Hiw-GFP. These results further indicate that autophagy promotes NMJ growth by down-regulating Hiw (Shen, 2009).

These results indicate that NMJ overgrowth caused by elevated autophagy is primarily caused by reduction in Hiw. Is the converse also true? Is NMJ undergrowth in atg mutants caused by elevated levels of Hiw? To address these questions, Hiw-GFP was expressed in neurons using C155-Gal4 in various backgrounds. Hiw-GFP levels were significantly elevated in atg1 and -6 mutants compared with the controls, consistent with the idea that Hiw is down-regulated by autophagy. If this increase in Hiw is a primary cause of NMJ undergrowth in atg loss-of-function mutants, eliminating Hiw should prevent this undergrowth; i.e., mutations in hiw should be epistatic to atg mutations. Thus, NMJ morphology was examined in hiw; atg2 and hiw; atg18 double mutants, and it was found that hiw was completely epistatic, demonstrating the role of elevated levels of Hiw in NMJ undergrowth of atg mutants (Shen, 2009).

A more direct test is to determine whether overexpression of Hiw can reduce NMJ size. However, this experiment is complicated because overexpression of Hiw by a relatively weak neuronal driver (elav-Gal4) does not affect NMJ size, whereas overexpression of Hiw by a strong neuronal driver (Elav-GeneSwitch) has a modest dominant-negative effect. To determine whether increased levels of Hiw can limit NMJ growth, it appears necessary to overexpress Hiw at an intermediate level. Therefore, NMJs were examined in larvae overexpressing UAS-hiw+ via C155-Gal4. C155-Gal4/+; UAS-hiw+/+ female larvae exhibited very mild NMJ undergrowth. Stronger undergrowth was observed in C155-Gal4/Y; UAS-hiw+/+ male larvae. This difference is consistent with higher levels of C155-Gal4 expression in males than in females, owing to dosage compensation. No differences were observed in NMJ growth between C155-Gal4/+ female and C155-Gal4/Y male larvae, indicating that the undergrowth phenotypes are dependent on the levels of Hiw overexpression and not on differences in gender or expression of GAL4 alone. Thus, moderate increases in Hiw levels result in NMJ undergrowth. Furthermore, the modest NMJ undergrowth in C155-Gal4/+; UAS-hiw+/+ larvae was enhanced when one copy of atg1+, -2+, or -6+ was removed. Together, these results indicate that elevated levels of Hiw account for most of the NMJ undergrowth in atg mutants. However, excess Hiw cannot fully explain NMJ undergrowth in atg mutants because NMJ undergrowth caused by Hiw overexpression is less severe than that of atg1 and -18 mutants. Thus, when autophagy is impaired, additional negative regulators may accumulate to depress NMJ growth. It is also likely that elevated levels of Hiw target proteins other than Wnd to limit synaptic growth because loss-of-function mutations of wnd do not affect NMJ development (Shen, 2009).

Because autophagy is generally thought of as a nonselective bulk degradation process, the idea that autophagy regulates NMJ growth primarily through its effects on Hiw levels seems difficult to understand at first. However, recent studies demonstrate that autophagy can also operate in a substrate-selective mode in regulating specific developmental events (Rowland, 2006; Zhang, 2009). For example, in Caenorhabditis elegans, when postsynaptic cells fail to receive presynaptic contact, GABAA receptors selectively traffic to autophagosomes (Rowland, 2006). However, the detailed mechanism of such selectivity is unknown. Zhang identified SEPA-1 as a bridge that mediates the specific recognition and degradation of P granules by autophagy in C. elegans. Thus, one possibility is that Hiw is specifically targeted to autophagosomes via a mechanism that remains to be elucidated. It is also possible that many presynaptic proteins besides Hiw are degraded by autophagy, but it is the reduction in Hiw that primarily affects NMJ size. Moreover, although the idea that autophagy regulates Hiw directly is favored, the possibility cannot be ruled out that autophagy promotes degradation of Hiw through an indirect mechanism involving the proteasome or other pathway (Shen, 2009).

In principle, autophagy could be acting on either side of the NMJ to regulate its development. Because atg1+ overexpression in muscle results in lethality at the first larval instar, it was not possible to assess whether this affects NMJ growth. Although a postsynaptic role of autophagy in NMJ development cannot be ruled out, several results suggest that the effects of autophagy are primarily presynaptic: neuronal expression of UAS-atg1+ is sufficient to completely rescue the NMJ undergrowth in atg1 mutants, the Hiw-Wnd pathway functions presynaptically (Wu, 2005; Collins, 2006), and hiw is completely epistatic to autophagy for NMJ growth (Shen, 2009).

Autophagy is of particular interest as a regulator of synaptic growth because it is triggered in response to many environmental cues. These results demonstrate that decreasing or increasing autophagy from basal levels results in corresponding effects on synaptic size. Thus, autophagy is perfectly positioned to link environmental conditions with synaptic growth and plasticity. As such, it is intriguing to speculate on a role for autophagy in learning and memory (Shen, 2009).

A permissive role of mushroom body α/β core neurons in long-term memory consolidation in Drosophila

Memories are not created equally strong or persistent for different experiences. In Drosophila, induction of long-term memory (LTM) for aversive olfactory conditioning requires ten spaced repetitive training trials, whereas a single trial is sufficient for LTM generation in appetitive olfactory conditioning. Although, with the ease of genetic manipulation, many genes and brain structures have been related to LTM formation, it is still an important task to identify new components and reveal the mechanisms underlying LTM regulation. This study shows that single-trial induction of LTM can also be achieved for aversive olfactory conditioning through inhibition of highwire (hiw)-encoded E3 ubiquitin ligase activity or activation of its targeted proteins in a cluster of neurons, localized within the α/β core region of the mushroom body. Moreover, the synaptic output of these neurons is critical within a limited posttraining interval for permitting consolidation of both aversive and appetitive LTM. It is proposed that these α/β core neurons serve as a 'gate' to keep LTM from being formed, whereas any experience capable of 'opening' the gate is given permit to be consolidated into LTM (Huang, 2012).

The current study began with the finding that 24 hr memory resulting from single session was enhanced in two hiw mutant alleles. This enhanced memory component was identified as facilitated LTM, given that it was sensitive to protein synthesis inhibition. The behavioral effect of hiwδRING and the presence of a hiw-GAL4 line allowed mapping the neural circuitry to a cluster of MB α/β core neurons, within which Hiw and its downstream targets regulate LTM. Furthermore, it was shown that the MB α/β core neurons are involved in the consolidation of both aversive and appetitive LTM. In conclusion, the observations that the MB α/β core neurons are capable of both facilitating and limiting LTM suggests a working model in which the ability to form LTM is gated through these neurons. The significance of these results is further elaborated below (Huang, 2012).

Not only synthesis but also degradation of proteins plays a critical role in the remodeling of synapses, learning, and memory. Altered memory formation in ubiquitin ligase mutants has been reported in mice and Drosophila. This study reports that Hiw, an evolutionarily conserved E3 ubiquitin ligase, negatively regulated LTM formation through restraining its downstream target Wallenda (Wnd). The results indicated Hiw function as an inhibitory constraint, as the memory suppressor gene, on LTM formation. Removal of this suppressor or direct activation of its downstream signals could lead to the facilitated LTM induction without the repetitive training that is normally required. So far, the physiological consequence of Hiw or Wnd in core neurons still remains an open question. Because of the extensively shared components with hiw’s function in synaptic growth and transmission, the attenuated Hiw activity may elevate Wnd level and then lead to the excessive synaptogenesis or abnormal synaptic activity in core neurons. It will also be interesting to test whether the hiw-mediated LTM facilitation shares some common molecular components with the corkscrew-regulated spacing effect in LTM induction, in which the MB α/β lobes also play an important role (Huang, 2012).

In a recent report, Hiw was shown to regulate the axon guidance in MB (Shin, 2011). The morphological defect was also observed in MB α/β lobes in a portion of hiw mutant flies. It is striking that hiw mutants with MB defect can form LTM even more efficiently, given the observation of LTM impairment in another MB structural mutant, ala. However, comparing to the total loss of vertical lobes (including α and α') in ala mutant, most hiw mutants had the abnormal thickness of α/β lobes caused by the unequal distribution of the MB axonal projections between the α and β lobes, and about 39% of hiwDN mutant had the shortened α lobe. One of the possible explanations is that the remaining function of α/β lobe in hiw mutants is sufficient to support the LTM. Moreover, expression of Hiw dominant-negative protein or acutely increasing Wnd protein level in MB was sufficient to promote LTM but did not give rise to any observable gross morphological change in MB. Thus, it is suggested that Hiw mediates memory phenotype through a different mechanism from the one that led to the structure change in the MB (Huang, 2012).

The involvement of MB in the hiw-mediated LTM facilitation led an examination of the function of this structure in LTM regulation. It has been well documented that the MB, a bilateral brain structure that consists of approximately 2,500 neurons in each hemisphere, plays the central role in olfactory memories, both aversive and appetitive. Intrinsic MB neurons are organized into physically distinct α, β, α', β' and γ lobes. All three lobes exhibit different functions in memory processing, such that the output of α/β lobes is required for retrieval of memory, α' β' lobes are transiently required to stabilize memory or to retrieve immediate memory, and γ lobe mediate a rutabaga-dependent mechanism and dopaminergic signal to support short-term memory (STM) and LTM formation. Moreover, memory traces mapped to different lobes exhibit different temporal features (Huang, 2012).

Through gene expression patterns and enhancer trap lines, each lobe of MB can be classified into more specific subgroups such as the posterior, surface, and core regions in the α/β lobes. The current work shows that α/β core neurons play a distinct role in LTM induction. The synaptic outputs of these neurons are critical during consolidation of LTM for both aversive and appetitive conditioning, but these neurons are not involved in LTM cellular consolidation per se because a landmark of LTM cellular consolidation, CREB-mediated protein synthesis, occurs in non-MB neurons. Thus, one of the roles for this cluster of neurons can be viewed as simply providing connections to channel learning information to the downstream neurons in which LTM is formed. However, the remarkable feature of enabling single-trial induction of aversive LTM through targeted genetic manipulation of this cluster of neurons suggests that they play a unique permissive role in determining whether an experience should be consolidated (Huang, 2012).

This newly identified function for permitting an experience to be consolidated leads to proposal of a gating theory. This theory proposes that the α/β core neurons serve as a 'gate,' and activation of this gating mechanism functions as a checkpoint that keeps LTM from being formed for general experiences, whereas only specific experiences, capable of 'opening' this gate, can and are bound to trigger LTM consolidation and to form LTM ultimately. There is little survival advantage in committing never-to-be-repeated episodes to memory, particularly because the very act of LTM formation may be deleterious to the fly. In contrast, repetitively occurring experience, such as spaced repetitive aversive conditioning, and events critical for survival, such as finding food or single-trial appetitive conditioning, would be able to 'open' the gate, and therefore, LTM is formed for such experiences (Huang, 2012).


Presynaptic terminals contain highly organized subcellular structures to facilitate neurotransmitter release. In C. elegans, the typical presynaptic terminal has an electron-dense active zone surrounded by synaptic vesicles. Loss-of-function mutations in the rpm-1 gene result in abnormally structured presynaptic terminals in GABAergic neuromuscular junctions (NMJs), most often manifested as a single presynaptic terminal containing multiple active zones. The RPM-1 protein has an RCC1-like guanine nucleotide exchange factor (GEF) domain and a RING-H2 finger. RPM-1 is most similar to the Drosophila presynaptic protein Highwire (HIW) and the mammalian Myc binding protein Pam. RPM-1 is localized to the presynaptic region independent of synaptic vesicles and it functions cell autonomously. The temperature-sensitive period of rpm-1 coincides with the time of synaptogenesis. rpm-1 may regulate the spatial arrangement, or restrict the formation, of presynaptic structures (Zhen, 2000).

rpm-1 mutations were isolated based on the abnormal morphologies of presynaptic terminals visualized by the synaptic vesicle-tagged GFP markers expressed in different types of neurons, hence the name regulator of presynaptic morphology. This analysis of the GABAergic NMJs has revealed that loss of rpm-1 function causes two types of defects: overdeveloped presynaptic terminals that contain multiple presynaptic active zones and underdeveloped presynaptic terminals that have few synaptic vesicles. In the strongest mutant background, both types of abnormality are present, and the overdeveloped presynaptic terminals appear to be predominant (Zhen, 2000).

rpm-1 is widely expressed in the nervous system. Different types of neurons appear to be affected in different manners and to different extents in rpm-1 mutants. The defects in cholinergic NMJs are similar to, but less severe than, those of GABAergic NMJs, and the abnormal presynaptic terminals either contain few vesicles or have elongated presynaptic active zones. No obvious axonal morphological defects are observed in GABAergic and cholinergic motor neurons of the ventral nerve cord. By contrast, the axons of mechanosensory and SAB motor neurons have extra branches, often bypass their targets, and make few synapses. It is not known whether the synapses of these mechanosensory and SAB neurons are affected at the ultrastructural level. Such 'bypass' phenotypes could be caused by a failure in recognizing targets or could be a secondary effect of failures in the initiation and stabilization of synapse formation. Formation of NMJs between ventral cord motor neurons and body muscles does not depend critically on the correct axon pathfinding of motor neurons and appears to be initiated by motor neurons responding to signals from muscle arms. However, mechanosensory and SAB neurons may form synapses onto their targets only after they are guided to the target regions. The different mutant phenotypes in mechanosensory and motor neurons may reflect the differences in how synapses are formed in different neurons or may indicate that rpm-1 has different functions in different neurons (Zhen, 2000).

Little is known of mechanisms regulating presynaptic differentiation. rpm-1 was identified in a screen for mutants with defects in patterning of a presynaptic marker at certain interneuronal synapses. The predicted RPM-1 protein contains zinc binding, RCC1, and other conserved motifs. In rpm-1 mutants, mechanosensory neurons fail to accumulate tagged vesicles, retract synaptic branches, and ectopically extend axons. Some motor neurons branch and overgrow; others show altered synaptic organization. Expression of RPM-1 in the presynaptic mechanosensory neurons is sufficient to rescue phenotypes in these cells. Certain rpm-1 phenotypes are temperature sensitive, revealing that RPM-1 function can be bypassed by maintaining mutants at the permissive temperature at stages commensurate with synapse formation in wild-type animals. These results indicate that RPM-1 functions cell autonomously during synaptogenesis to regulate neuronal morphology (Schaefer, 2000).

What mechanism in the presynaptic mechanosensory neurons is regulated in vivo by RPM-1? There are at least two possibilities. First, RPM-1 could regulate neuronal maturation. The data are consistent with a model whereby mechanosensory synapses fail to mature, generating retraction of the synaptic branch. Although fewer synaptic branches are observed than in wild type, additional transient branches could be missed in the analyses of staged populations. Such growth cone dynamics would be consistent with the 20-55 µm/hr pace of VD motor neuron migration. Moreover, if RPM-1 is a regulator of synaptic maturation, the rpm-1 mutant phenotype of ectopic axonal targeting could be secondary to failure to form stable synapses. Perhaps such failure generates a retrograde signal, which leads to ectopic growth toward the VNC. Alternatively, the ectopic growth could be a primary defect, resulting from failure of the axonal growth cone to mature into a stable and static structure in the lateral midbody region (Schaefer, 2000).

In a different scenario, RPM-1 could be involved in mechanisms of outgrowth at the appropriate subcellular locus. Normally, PLM extends a synaptic branch perpendicularly from the middle of the axon, in a defined region between PVM and the vulva. Perhaps, in rpm-1 mutants, the machinery for synaptic branch extension is misrouted to the inappropriate intracellular location. However, such machinery could normally be distributed uniformly along the axon, but local cues regulating the locus of branch extension are somehow disrupted in rpm-1 animals. Normal and ectopic targeting are not mutually exclusive: a single PLM neuron can display both a synaptic branch and ectopic growth from the end of the axon (Schaefer, 2000).

The first hypothesis, that RPM-1 is part of a mechanism to effect neuronal maturation, is favored. This explanation is more consistent with the phenotype of synaptic branch retraction. Moreover, a role in maturation is compatible with the motor neuron phenotypes. The SAB motor neurons in rpm-1 sprout additional branches, similar to sprouting in SAB neurons induced by deficits in synaptic activity. In the DNC neuropil, there were fewer presynaptic densities. Wider gaps between these labeled presynaptic specializations could simply reflect that many specializations are missing in mutant animals. Labeled presynaptic specializations in the DNC appear to aggregate in mutants. Perhaps these aggregations represent the addition of active zones within individual motor neurons. Alternatively, they may be generated by a redistribution of existing synapses (Schaefer, 2000).

Observations of these changes in presynaptic labeling and in morphology of mechanosensory and motor neurons thus may reflect a primary defect in synapse formation. Failure of synapses to form or to mature may, in turn, generate distinct cellular responses, depending on cell type (Schaefer, 2000).

If indeed RPM-1 is involved in neuronal maturation at the time of synaptogenesis, its function is not absolutely required: labeling with synaptic markers indicates that many synapses do differentiate in rpm-1 animals. Neuronal patterning is grossly normal. For the screen, a gross behavioral phenotype from loss, per se, of chemical synapses made by mechanosensory neurons was not predicted; these synapses are not required for the touch response. Nevertheless, the essentially normal behavior of mutant animals confirms that there are no general synaptic defects (Schaefer, 2000).

During synapse formation, specialized subcellular structures develop at synaptic junctions in a tightly regulated fashion. Cross-signalling initiated by ephrins, Wnts and transforming growth factor-beta family members between presynaptic and postsynaptic termini are proposed to govern synapse formation. It is not well understood how multiple signals are integrated and regulated by developing synaptic termini to control synaptic differentiation. FSN-1 is a novel F-box protein that is required in presynaptic neurons for the restriction and/or maturation of synapses in Caenorhabditis elegans. Many F-box proteins are target recognition subunits of SCF (Skp, Cullin, F-box) ubiquitin-ligase complexes. fsn-1 functions in the same pathway as rpm-1, a gene encoding a large protein with RING finger domains. FSN-1 physically associates with RPM-1 and the C. elegans homologues of SKP1 and Cullin to form a new type of SCF complex at presynaptic periactive zones. Evidence is provided that T10H9.2, which encodes the C. elegans receptor tyrosine kinase ALK (anaplastic lymphoma kinase), may be a target or a downstream effector through which FSN-1 stabilizes synapse formation. This neuron-specific, SCF-like complex therefore provides a localized signal to attenuate presynaptic differentiation (Liao, 2004).

Genetic studies using a set of overlapping deletions centered at the piebald locus on distal mouse chromosome 14 have defined a genomic region associated with respiratory distress and lethality at birth. The candidate gene Phr1 that is located within the respiratory distress critical genomic interval has been isolated and characterized. Phr1 is the ortholog of the human Protein Associated with Myc as well as Drosophila highwire and Caenorhabditis elegans regulator of presynaptic morphology 1. Phr1 is expressed in the embryonic and postnatal nervous system. In mice lacking Phr1, the phrenic nerve fails to completely innervate the diaphragm. In addition, nerve terminal morphology is severely disrupted, comparable to the synaptic defects seen in the Drosophila hiw and C. elegans rpm-1 mutants. Although intercostal muscles were completely innervated, they also showed dysmorphic nerve terminals. In addition, sensory neuron terminals in the diaphragm were abnormal. The neuromuscular junctions showed excessive sprouting of nerve terminals, consistent with inadequate presynaptic stimulation of the muscle. On the basis of the abnormal neuronal morphology seen in mice, Drosophila, and C. elegans, it is proposed that Phr1 plays a conserved role in synaptic development and is a candidate gene for respiratory distress and ventilatory disorders that arise from defective neuronal control of breathing (Burgess, 2004).

Changes in axon outgrowth patterns are often associated with synaptogenesis. Members of the conserved Pam/Highwire/RPM-1 protein family have essential functions in presynaptic differentiation. This study shows that C. elegans RPM-1 negatively regulates axon outgrowth mediated by the guidance receptors SAX-3/robo and UNC-5/UNC5. Loss-of-function rpm-1 mutations cause a failure to terminate axon outgrowth, resulting in an overextension of the longitudinal PLM axon. PLM overextension observed in rpm-1 mutants is suppressed by sax-3 and unc-5 loss-of-function mutations. PLM axon overextension is also induced by SAX-3 overexpression, and the length of extension is enhanced by loss of rpm-1 function or suppressed by loss of unc-5 function. Loss of rpm-1 function in genetic backgrounds sensitized for guidance defects disrupts ventral AVM axon guidance in a SAX-3-dependent manner and enhances dorsal guidance of DA and DB motor axons in an UNC-5-dependent manner. Loss of rpm-1 function alters expression of the green fluorescent protein (GFP)-tagged proteins, SAX-3::GFP and UNC-5::GFP. RPM-1 is known to regulate axon termination through two parallel genetic pathways; one involves the Rab GEF (guanine nucleotide exchange factor) GLO-4, which regulates vesicular trafficking, and another that involves the F-box protein FSN-1, which mediates RPM-1 ubiquitin ligase activity. glo-4 but not fsn-1 mutations affect axon guidance in a manner similar to loss of rpm-1 function. Together, the results suggest that RPM-1 regulates axon outgrowth affecting axon guidance and termination by controlling the trafficking of the UNC-5 and SAX-3 receptors to cell membranes (Li, 2008).

The PAM/Highwire/RPM-1 (PHR) proteins are signaling hubs that function as important regulators of neural development. Loss of function in C. elegans E3 ubiquitin-protein ligase rpm-1 and Drosophila Highwire results in failed axon termination, inappropriate axon targeting, and abnormal synapse formation. Very mild abnormalities in behavior have been found in animals lacking PHR protein function. Therefore, it was hypothesized that large defects in behavior might only be detected in scenarios in which evoked, prolonged circuit function is required, or in which behavioral plasticity occurs. rpm-1 loss-of-function mutants had relatively mild abnormalities in exploratory locomotion, but exhibited large defects in evoked responses to harsh touch and learning associated with tap habituation. Rescue analysis indicated that RPM-1 function in the mechanosensory neurons affects habituation. Transgenic expression of RPM-1 in adult animals failed to rescue habituation defects, consistent with developmental defects in rpm-1 mutants resulting in impaired habituation. Genetic analysis showed that other regulators of neuronal development that function in the rpm-1 pathway (including glo-4, fsn-1, and dlk-1) also affected habituation. Overall, these findings suggest that developmental defects in rpm-1 mutants manifest most prominently in behaviors that require protracted or plastic circuit function, such as learning (Giles, 2015).

Visual system development is dependent on correct interpretation of cues that direct growth cone migration and axon branching. Mutations in the zebrafish esrom gene disrupt bundling and targeting of retinal axons, and also cause ectopic arborization. By positional cloning, it was established that esrom encodes a very large protein orthologous to PAM (protein associated with Myc)/Highwire/RPM-1. Unlike motoneurons in Drosophila highwire mutants, retinal axons in esrom mutants do not arborize excessively, indicating that Esrom has different functions in the vertebrate visual system. Esrom has E3 ligase activity and modulates the amount of phosphorylated Tuberin, a tumor suppressor, in growth cones. These data identify a mediator of signal transduction in retinal growth cones that is required for topographic map formation (D'Souza, 2005).

Zebrafish esrom mutants have an unusual combination of phenotypes: in addition to a defect in the projection of retinal axons, they have reduced yellow pigmentation. The pigment phenotype was investigated and, from this, evidence is provided for an unexpected defect in retinal neurons. Esrom is not required for the differentiation of neural crest precursors into pigment cells, nor is it essential for cell migration, pigment granule biogenesis, or translocation. Instead, loss of yellow color is caused by a deficiency of sepiapterin, a yellow pteridine. The level of several other pteridines is also affected in mutants. Importantly, the cofactor tetrahydrobiopterin (BH4) is drastically reduced in esrom mutants. Mutant retinal neurons also appear deficient in this pteridine. BH4-synthesizing enzymes are active in mutants, indicating a defect in the regulation rather than production of enzymes. Esrom has recently been identified as an ortholog of PAM (protein associated with c-myc), a very large protein involved in synaptogenesis in Drosophila and C. elegans. These data thus introduce a new regulator of pteridine synthesis in a vertebrate and establish a function for the Esrom protein family outside synaptogenesis. They also raise the possibility that neuronal defects are due in part to an abnormality in pteridine synthesis (Le Guyader, 2005).

Synapses display a stereotyped ultrastructural organization, commonly containing a single electron-dense presynaptic density surrounded by a cluster of synaptic vesicles. The mechanism controlling subsynaptic proportion is not understood. Loss of function in the C. elegans rpm-1 gene, a putative RING finger/E3 ubiquitin ligase, causes disorganized presynaptic cytoarchitecture. RPM-1 is localized to the presynaptic periactive zone. RPM-1 negatively regulates a p38 MAP kinase pathway composed of the dual leucine zipper-bearing MAPKKK DLK-1, the MAPKK MKK-4, and the p38 MAP kinase PMK-3. Inactivation of this pathway suppresses rpm-1 loss of function phenotypes, whereas overexpression or constitutive activation of this pathway causes synaptic defects resembling rpm-1lf mutants. DLK-1, like RPM-1, is localized to the periactive zone. DLK-1 protein levels are elevated in rpm-1 mutants. The RPM-1 RING finger can stimulate ubiquitination of DLK-1. These data reveal a presynaptic role of a previously unknown p38 MAP kinase cascade (Nakata, 2005).


Search PubMed for articles about Drosophila highwire

Bao, H., Reist, N. E. and Zhang, B. (2008). The Drosophila epsin 1 is required for ubiquitin-dependent synaptic growth and function but not for synaptic vesicle recycling. Traffic 9(12): 2190-205. PubMed Citation: 18796008

Burgess, R. W., Peterson, K. A., Johnson, M. J., Roix, J. J., Welsh, I. C. and O'Brien, T. P. (2004). Evidence for a conserved function in synapse formation reveals Phr1 as a candidate gene for respiratory failure in newborn mice. Mol. Cell. Biol. 24: 1096-1105. 14729956

Coleman, M. P. and Freeman, M. R. (2010). Wallerian degeneration, wld(s), and nmnat. Annu Rev Neurosci 33: 245-267. PubMed ID: 20345246

Collins, C. A., Wairkar, Y. P., Johnson, S. L. and DiAntonio, A. (2006). Highwire restrains synaptic growth by attenuating a MAP kinase signal. Neuron 51(1): 57-69. 16815332

Diantonio, A., et al. (2001). Ubiquitination-dependent mechanisms regulate synaptic growth and function. Nature 412: 449-452. 11473321

D'Souza, J., Hendricks, M., Le Guyader, S., Subburaju, S., Grunewald, B., Scholich, K. and Jesuthasan, S. (2005). Formation of the retinotectal projection requires Esrom, an ortholog of PAM (protein associated with Myc). Development 132(2): 247-56. 15590740

Giles, A. C., Opperman, K. J., Rankin, C. H. and Grill, B. (2015). Developmental function of the PHR protein RPM-1 is required for learning in Caenorhabditis elegans. G3 (Bethesda) [Epub ahead of print]. PubMed ID: 26464359

Guo, Q., Xie, J., Dang, C. V., Liu, E. T. and Bishop, J. M. (1998). Identification of a large Myc-binding protein that contains RCC1-like repeats. Proc. Natl. Acad. Sci. 95: 9172-9177. 9689053

Hendricks, M., and Jesuthasan, S. (2009). PHR regulates growth cone pausing at intermediate targets through microtubule disassembly. J. Neurosci. 29: 6593-6598. PubMed Citation: 19458229

Hirai, S., et al. (2002). MAPK-upstream protein kinase (MUK) regulates the radial migration of immature neurons in telencephalon of mouse embryo. Development 129: 4483-4495. 12223406

Hirai, S., et al. (2005). Expression of MUK/DLK/ZPK, an activator of the JNK pathway, in the nervous systems of the developing mouse embryo. Gene Expr. Patterns 5: 517-523. 15749080

Huang, C., Zheng, X., Zhao, H., Li, M., Wang, P., Xie, Z., Wang, L. and Zhong, Y. (2012). A permissive role of mushroom body α/β core neurons in long-term memory consolidation in Drosophila. Curr Biol 22: 1981-1989. PubMed ID: 23063437

Le Guyader, S., Maier, J. and Jesuthasan, S. (2005). Esrom, an ortholog of PAM (protein associated with c-myc), regulates pteridine synthesis in the zebrafish. Dev. Biol. 277(2): 378-86. 15617681

Li, H., Kulkarni, G. and Wadsworth, W. G. (2008), RPM-1, a Caenorhabditis elegans protein that functions in presynaptic differentiation, negatively regulates axon outgrowth by controlling SAX-3/robo and UNC-5/UNC5 activity. J. Neurosci. 28(14): 3595-603. PubMed Citation: 18385318

Liao, E. H, Hung, W., Abrams, B. and Zhen, M. (2004) An SCF-like ubiquitin ligase complex that controls presynaptic differentiation. Nature 430: 345-350. 15208641

McCabe, B. D., et al. (2004). Highwire regulates presynaptic BMP signaling essential for synaptic growth. Neuron 41(6): 891-905. 15046722

McDermott, S. M., Yang, L., Halstead, J. M., Hamilton, R. S., Meignin, C. and Davis, I. (2014). Drosophila Syncrip modulates the expression of mRNAs encoding key synaptic proteins required for morphology at the neuromuscular junction. RNA 20(10): 1593-606. PubMed ID: 25171822

Muralidhar, M. G. and Thomas, J. B. (1993). The Drosophila bendless gene encodes a neural protein related to ubiquitin-conjugating enzymes. Neuron 11(2): 253-66. PubMed Citation: 8394720

Murthy, V., et al. (2004). Pam and its ortholog Highwire interact with and may negatively regulate the TSC1.TSC2 complex. J. Biol. Chem. 279: 1351-1358. PubMed Citation: 14559897

Nakata, K., et al. (2005). Regulation of a DLK-1 and p38 MAP kinase pathway by the ubiquitin ligase RPM-1 is required for presynaptic development. Cell 120: 407-420. 15707898

Oh, C. E., McMahon, R., Benzer, S. and Tanouye, M. A. (1994). bendless, a Drosophila gene affecting neuronal connectivity, encodes a ubiquitin-conjugating enzyme homolog. J. Neurosci. 14(5 Pt 2): 3166-79. PubMed Citation: 8182464

Rawson, J. M., Lee, M., Kennedy, E. L. and Selleck, S. B. (2003). Drosophila neuromuscular synapse assembly and function require the TGF-beta type I receptor saxophone and the transcription factor Mad. J. Neurobiol. 55: 134-150. 12672013

Schaefer, A. M., Hadwiger, G. D. and Nonet, M. L. (2000). rpm-1, a conserved neuronal gene that regulates targeting and synaptogenesis in C. elegans. Neuron 26: 345-356. 10839354

Shen, W. and Ganetzky, B. (2009). Autophagy promotes synapse development in Drosophila. J Cell Biol. 187(1): 71-9. PubMed Citation: 19786572

Shin, J. E. and DiAntonio, A. (2011). Highwire regulates guidance of sister axons in the Drosophila mushroom body. J. Neurosci. 31(48): 17689-700. PubMed Citation: 22131429

Toda, H., et al. (2008). UNC-51/ATG1 kinase regulates axonal transport by mediating motor-cargo assembly. Genes Dev. 22: 3292-3307. PubMed Citation: 19056884

Uthaman, S. B., Godenschwege, T. A. and Murphey, R. K. (2008). A mechanism distinct from highwire for the Drosophila ubiquitin conjugase bendless in synaptic growth and maturation. J. Neurosci. 28(34): 8615-23. PubMed Citation: 18716220

Wairkar, Y. P., et al. (2009). Unc-51 controls active zone density and protein composition by downregulating ERK signaling. J. Neurosci. 29: 517-528. PubMed Citation: 19144852

Wan, H. I., et al. (2000). Highwire regulates synaptic growth in Drosophila. Neuron 26: 313-329. 10839352

Wu, C., Wairkar, Y. P., Collins, C. A. and DiAntonio, A. (2005). Highwire function at the Drosophila neuromuscular junction: spatial, structural, and temporal requirements. J. Neurosci. 25(42): 9557-66. 16237161

Wu, C., Daniels, R. W. and Diantonio, A. (2007). DFsn collaborates with Highwire to down-regulate the Wallenda/DLK kinase and restrain synaptic terminal growth. Neural Develop. 2:16. PubMed citation: 17697379

Xiong, X., Wang, X., Ewanek, R., Bhat, P., Diantonio, A. and Collins, C. A. (2010). Protein turnover of the Wallenda/DLK kinase regulates a retrograde response to axonal injury. J Cell Biol 191: 211-223. PubMed ID: 20921142

Xiong, X., Hao, Y., Sun, K., Li, J., Li, X., Mishra, B., Soppina, P., Wu, C., Hume, R. I. and Collins, C. A. (2012). The Highwire ubiquitin ligase promotes axonal degeneration by tuning levels of Nmnat protein. PLoS Biol 10: e1001440. PubMed ID: 23226106

Zhai, Q., Wang, J., Kim, A., Liu, Q., Watts, R., Hoopfer, E., Mitchison, T., Luo, L. and He, Z. (2003). Involvement of the ubiquitin-proteasome system in the early stages of wallerian degeneration. Neuron 39: 217-225. PubMed ID: 12873380

Zhen, M., Huang, X., Bamber, B. and Jin, Y. (2000). Regulation of presynaptic terminal organization by C. elegans RPM-1, a putative guanine nucleotide exhanger with a RING-H2 finger domain. Neuron 26: 331-343. 10839353

Biological Overview

date revised: 20 August 2013

Home page: The Interactive Fly © 2017 Thomas Brody, Ph.D.