The function of p120 catenin in filopodial growth and synaptic vesicle clustering in neurons

At the developing neuromuscular junction (NMJ), physical contact between motor axons and muscle cells initiates presynaptic and postsynaptic differentiation. Using Xenopus nerve-muscle cocultures, we previously showed that innervating axons induced muscle filopodia (myopodia), which facilitated interactions between the synaptic partners and promoted NMJ formation. The myopodia were generated by nerve-released signals through muscle p120 catenin (p120ctn), a protein of the cadherin complex that modulates the activity of Rho GTPases. Because axons also extend filopodia that mediate early nerve-muscle interactions, here we test p120ctn's function in the assembly of these presynaptic processes. Overexpression of wild-type p120ctn in Xenopus spinal neurons leads to an increase in filopodial growth and synaptic vesicle (SV) clustering along axons, whereas the development of these specializations is inhibited following the expression of a p120ctn mutant lacking sequences important for regulating Rho GTPases. The p120ctn mutant also inhibits the induction of axonal filopodia and SV clusters by basic fibroblast growth factor, a muscle-derived molecule that triggers presynaptic differentiation. Of importance, introduction of the p120ctn mutant into neurons hinders NMJ formation, which is observed as a reduction in the accumulation of acetylcholine receptors at innervation sites in muscle. Our results suggest that p120ctn signaling in motor neurons promotes nerve-muscle interaction and NMJ assembly.     

Reciprocal Regulation of Axonal Filopodia and Outgrowth during Neuromuscular Junction Development

Background

The assembly of the vertebrate neuromuscular junction (NMJ) is initiated when nerve and muscle first contact each other by filopodial processes which are thought to enable close interactions between the synaptic partners and facilitate synaptogenesis. We recently reported that embryonic Xenopus spinal neurons preferentially extended filopodia towards cocultured muscle cells and that basic fibroblast growth factor (bFGF) produced by muscle activated neuronal FGF receptor 1 (FGFR1) to induce filopodia and favor synaptogenesis. Intriguingly, in an earlier study we found that neurotrophins (NTs), a different set of target-derived factors that act through Trk receptor tyrosine kinases, promoted neuronal growth but hindered presynaptic differentiation and NMJ formation. Thus, here we investigated how bFGF- and NT-signals in neurons jointly elicit presynaptic changes during the earliest stages of NMJ development.

Methodology/Principal Findings

Whereas forced expression of wild-type TrkB in neurons reduced filopodial extension and triggered axonal outgrowth, expression of a mutant TrkB lacking the intracellular kinase domain enhanced filopodial growth and slowed axonal advance. Neurons overexpressing wild-type FGFR1 also displayed more filopodia than control neurons, in accord with our previous findings, and, notably, this elevation in filopodial density was suppressed when neurons were chronically treated from the beginning of the culture period with BDNF, the NT that specifically activates TrkB. Conversely, inhibition by BDNF of NMJ formation in nerve-muscle cocultures was partly reversed by the overexpression of bFGF in muscle.

Conclusions

Our results suggest that the balance between neuronal FGFR1- and TrkB-dependent filopodial assembly and axonal outgrowth regulates the establishment of incipient NMJs.     

Fibroblast growth factor receptor signaling in Xenopus retinal axon extension

Fibroblast growth factor receptors (FGFRs) and N-cadherin both regulate axon extension in developing Xenopus retinal ganglion cells (RGCs). Cultured cerebellar neurons have been shown to require FGFR activity for N-cadherin–stimulated neurite outgrowth, raising the possibility that N-cadherin is a FGFR ligand. To investigate this possibility in the developing visual system, retinal neurons were transfected with a dominant-negative FGFR (XFD) and plated on purified N-cadherin substrates. XFD-expressing neurons extended markedly shorter processes than control GFP-expressing neurons, implicating a role for FGFRs in N-cadherin–stimulated neurite outgrowth. To examine whether N-cadherin and FGFRs share the same pathway or use distinct second messenger pathways, specific inhibitors of implicated signaling molecules were added to neurons stimulated by N-cadherin, basic fibroblast growth factor (bFGF), or brain-derived nerve factor (BDNF) (which stimulates RGC outgrowth by a FGFR-independent mechanism). Diacylglycerol (DAG) lipase and Ca²⁺/calmodulin kinase II inhibitors both significantly reduced outgrowth stimulated by N-cadherin or bFGF but not by BDNF. Furthermore, we show that inhibiting DAG lipase activity in RGC axons extending in vivo toward the optic tectum reversibly slows axon extension without collapsing their growth cones. Thus, a common second-messenger signaling pathway mediating both N-cadherin– and bFGF-stimulated neurite extension is consistent with a model in which N-cadherin directly modulates the FGFR or a model whereby both FGFR and N-cadherin regulate the same second-messenger system. © 1998 John Wiley & Sons, Inc. J Neurobiol 37: 633–641, 1998     

Involvement of p120 catenin in myopodial assembly and nerve-muscle synapse formation

At developing neuromuscular junctions (NMJs), muscles initially contact motor axons by microprocesses, or myopodia, which are induced by nerves and nerve-secreted agrin, but it is unclear how myopodia are assembled and how they influence synaptic differentiation at the NMJ. Here, we report that treatment of cultured muscle cells with agrin transiently depleted p120 catenin (p120ctn) from cadherin junctions in situ, and increased the tyrosine phosphorylation and decreased the cadherin-association of p120ctn in cell extracts. Whereas ectopic expression of wild-type p120ctn in muscle generated myopodia in the absence of agrin, expression of a specific dominant-negative mutant form of p120ctn, which blocks filopodial assembly in nonmuscle cells, suppressed nerve- and agrin-induction of myopodia. Significantly, approaching neurites triggered reduced acetylcholine receptor (AChR) clustering along the edges of muscle cells expressing mutant p120ctn than of control cells, although the ability of the mutant cells to cluster AChRs was itself normal. Our results indicate a novel role of p120ctn in agrin-induced myopodial assembly and suggest that myopodia increase muscle-nerve contacts and muscle's access to neural agrin to promote NMJ formation.     

Bilaterally Symmetric Populations of Chicken dI1 (Commissural) Axons Cross the Floor Plate Independently of Each Other

Axons use temporal and directional guidance cues at intermediate targets to set the rate and direction of growth towards their synaptic targets. Our recent studies have shown that disrupting the temporal guidance process, by unilaterally accelerating the rate at which spinal dI1 (commissural) axons grow, resulted in turning errors both in the ventral spinal cord and after crossing the floor plate. Here we investigate a mechanistic explanation for these defects: the accelerated dI1 axons arrive in the ventral spinal cord before necessary fasciculation cues from incoming dI1 axons from the opposite side of the spinal cord. The identification of such an interaction would support a model of selective fasciculation whereby the pioneering dI1 axons serve as guides for the processes of the bilaterally symmetrical population of dI1 neurons. To test this model, we first developed the ability to “double” in ovo electroporate the embryonic chicken spinal cord to independently manipulate the rate of growth of the two bilateral populations of dI1 axons. Second, we examined the requirement for a putative bilateral interaction by unilaterally ablating the dI1 population in cultured explants of chicken embryonic spinal cord. Surprisingly, we find no evidence for a bilateral dI1 axon interaction, rather dI1 axons appear to project independently of each other.     

Wnt Signalling Promotes Actin Dynamics during Axon Remodelling through the Actin-Binding Protein Eps8

Upon arrival at their synaptic targets, axons slow down their growth and extensively remodel before the assembly of presynaptic boutons. Wnt proteins are target-derived secreted factors that promote axonal remodelling and synaptic assembly. In the developing spinal cord, Wnts secreted by motor neurons promote axonal remodelling of NT-3 responsive dorsal root ganglia neurons. Axon remodelling induced by Wnts is characterised by growth cone pausing and enlargement, processes that depend on the re-organisation of microtubules. However, the contribution of the actin cytoskeleton has remained unexplored. Here, we demonstrate that Wnt3a regulates the actin cytoskeleton by rapidly inducing F-actin accumulation in growth cones from rodent DRG neurons through the scaffold protein Dishevelled-1 (Dvl1) and the serine-threonine kinase Gsk3β. Importantly, these changes in actin cytoskeleton occurs before enlargement of the growth cones is evident. Time-lapse imaging shows that Wnt3a increases lamellar protrusion and filopodia velocity. In addition, pharmacological inhibition of actin assembly demonstrates that Wnt3a increases actin dynamics. Through a yeast-two hybrid screen, we identified the actin-binding protein Eps8 as a direct interactor of Dvl1, a scaffold protein crucial for the Wnt signalling pathway. Gain of function of Eps8 mimics Wnt-mediated axon remodelling, whereas Eps8 silencing blocks the axon remodelling activity of Wnt3a. Importantly, blockade of the Dvl1-Eps8 interaction completely abolishes Wnt3a-mediated axonal remodelling. These findings demonstrate a novel role for Wnt-Dvl1 signalling through Eps8 in the regulation of axonal remodeling.     

Regulation of axonal growth and neuromuscular junction formation by neuronal phosphatase and tensin homologue signaling

During the development of the vertebrate neuromuscular junction (NMJ), motor axon tips stop growing after contacting muscle and transform into presynaptic terminals that secrete the neurotransmitter acetylcholine and activate postsynaptic ACh receptors (AChRs) to trigger muscle contraction. The neuron-intrinsic signaling that retards axonal growth to facilitate stable nerve–muscle interaction and synaptogenesis is poorly understood. In this paper, we report a novel function of presynaptic signaling by phosphatase and tensin homologue (PTEN) in mediating a growth-to-synaptogenesis transition in neurons. In Xenopus nerve–muscle cocultures, axonal growth speed was halved after contact with muscle, when compared with before contact, but when cultures were exposed to the PTEN blocker bisperoxo (1,10-phenanthroline) oxovanadate, axons touching muscle grew ∼50% faster than their counterparts in control cultures. Suppression of neuronal PTEN expression using morpholinos or the forced expression of catalytically inactive PTEN in neurons also resulted in faster than normal axonal advance after contact with muscle cells. Significantly, interference with PTEN by each of these methods also led to reduced AChR clustering at innervation sites in muscle, indicating that disruption of neuronal PTEN signaling inhibited NMJ assembly. We thus propose that PTEN-dependent slowing of axonal growth enables the establishment of stable nerve–muscle contacts that develop into NMJs.     

Multiple Neurotrophic Factors from Skeletal Muscle: Demonstration of Effects of Basic Fibroblast Growth Factor and Comparisons with the 22-Kilodalton Choline Acetyltransferase Development Factor

Extracts of skeletal muscle contain chromatographically distinct molecules that enhance the cholinergic development of cultured embryonic rat spinal cord neurons. We have recently purified a 20-22 kilodalton anionic polypeptide choline acetyltransferase (ChAT) development factor (CDF) from rat skeletal muscle extracts that stimulates the development of ChAT activity in rat spinal cord cultures. The maximum increase in the level of ChAT activity achieved by this factor, however, is less than that achieved by the addition of the crude extract. We now show that muscle extract also contains mitogenic activity that is immunologically related to basic fibroblast growth factor (bFGF) and also that recombinant bFGF stimulates ChAT development in rat spinal cord cultures. bFGF, however, differs from CDF in its physiochemical, chromatographic, and immunological properties and by its action on nonneuronal cells. Individually, CDF and bFGF each enhance the level of ChAT activity in rat spinal cord cultures two- to threefold after 2 days of treatment. However, their combined actions result in a five- to sixfold enhancement of ChAT activity, suggesting that they are affecting cholinergic development through different means. The demonstration that extracts of rat skeletal muscle contain two biochemically and immunologically distinct polypeptides, with additive effects on cultured embryonic spinal cord neurons, provides additional evidence for the involvement of multiple target-derived neurotrophic factors in the regulation of cholinergic development.     

Muscular dystrophy associated mutations in caveolin-1 induce neurotransmission and locomotion defects in Caenorhabditis elegans

Mutations in human caveolin-3 are known to underlie a range of myopathies. The cav-1 gene of Caenorhabditis elegans is a homologue of human caveolin-3 and is expressed in both neurons and body wall muscles. Within the body wall muscle CAV-1 localises adjacent to neurons, most likely at the neuromuscular junction (NMJ). Using fluorescently tagged CAV-1 and pre- and post-synaptic markers we demonstrate that CAV-1 co-localises with UNC-63, a post-synaptic marker, but not with several pre-synaptic markers. To establish a model for human muscular dystrophies caused by dominant-negative mutations in caveolin-3 we created transgenic animals carrying versions of cav-1 with homologous mutations. These animals had increased sensitivity to levamisole, suggesting a role for cav-1 at the NMJ. Animals carrying a deletion in cav-1 show a similar sensitivity. Sensitivity to levamisole and locomotion were also perturbed in animals carrying a dominant-negative cav-1 and a mutation in dynamin, which is a protein known to interact with caveolins. Thus, indicating an interaction between CAV-1 and dynamin at the NMJ and/or in neurons.     

Mitochondrial clustering at the vertebrate neuromuscular junction during presynaptic differentiation

During vertebrate neuromuscular junction (NMJ) development, presynaptic motor axons differentiate into nerve termini enriched in synaptic vesicles (SVs). At the nerve terminal, mitochondria are also concentrated, but how mitochondria become localized at these specialized domains is poorly understood. This process was studied in cultured Xenopus spinal neurons with mitochondrion-specific probe MitoTracker and SV markers. In nerve-muscle cocultures, mitochondria were concentrated stably at sites where neurites and muscle cells formed NMJs, and mitochondria coclustered with SVs where neurites were focally stimulated by beads coated with growth factors. Labeling with a mitochondrial membrane potential-dependent probe JC-1 revealed that these synaptic mitochondria were with higher membrane potential than the extrasynaptic ones. At early stages of bead-stimulation, actin-based protrusions and microtubule fragmentation were observed in neurites at bead contact sites, suggesting the involvement of cytoskeletal dynamics and rearrangement during presynaptic differentiation. Treating the cultures with an actin polymerization blocker, latrunculin A (Ltn A), almost completely abolished the formation of actin-based protrusions and partially inhibited bead-induced mitochondrial and SV clustering, whereas the microtubule disrupting agent nocodazole was ineffective in inhibiting the clustering of mitochondria and SVs. Lastly, in contrast to Ltn A, which blocked bead-induced clustering of both mitochondria and SVs, the ser/thr phosphatase inhibitor okadaic acid inhibited SV clustering but not mitochondrial clustering. These results suggest that at developing NMJs, synaptogenic stimuli induce the clustering of mitochondria together with SVs at presynaptic terminals in an actin cytoskeleton-dependent manner and involving different intracellular signaling molecules.     

Distinct roles of muscle and motoneuron LRP4 in neuromuscular junction formation

Neuromuscular junction (NMJ) formation requires precise interaction between motoneurons and muscle fibers. LRP4 is a receptor of agrin that is thought to act in cis to stimulate MuSK in muscle fibers for postsynaptic differentiation. Here we dissected the roles of LRP4 in muscle fibers and motoneurons in NMJ formation by cell-specific mutation. Studies of muscle-specific mutants suggest that LRP4 is involved in deciding where to form AChR clusters in muscle fibers, postsynaptic differentiation, and axon terminal development. LRP4 in HEK293 cells increased synapsin or SV2 puncta in contacting axons of cocultured neurons, suggesting a synaptogenic function. Analysis of LRP4 muscle and motoneuron double mutants and mechanistic studies suggest that NMJ formation may also be regulated by LRP4 in motoneurons, which could serve as agrin's receptor in trans to induce AChR clusters. These observations uncovered distinct roles of LRP4 in motoneurons and muscles in NMJ development.     

β-Catenin gain of function in muscles impairs neuromuscular junction formation

Neuromuscular junction (NMJ) formation requires proper interaction between motoneurons and muscle cells. β-Catenin is required in muscle cells for NMJ formation. To understand underlying mechanisms, we investigated the effect of β-catenin gain of function (GOF) on NMJ development. In HSA-β-cat(flox(ex3)/+) mice, which express stable β-catenin specifically in muscles, motor nerve terminals became extensively defasciculated and arborized. Ectopic muscles were observed in the diaphragm and were innervated by ectopic phrenic nerve branches. Moreover, extensive outgrowth and branching of spinal axons were evident in the GOF mice. These results indicate that increased β-catenin in muscles alters presynaptic differentiation. Postsynaptically, AChR clusters in HSA-β-cat(flox(ex3)/+) diaphragms were distributed in a wider region, suggesting that muscle β-catenin GOF disrupted the signal that restricts AChR clustering to the middle region of muscle fibers. Expression of stable β-catenin in motoneurons, however, had no effect on NMJ formation. These observations provide additional genetic evidence that pre- and postsynaptic development of the NMJ requires an intricate balance of β-catenin activity in muscles.     

VEGF mediates commissural axon chemoattraction through its receptor Flk1

Growing axons are guided to their targets by attractive and repulsive cues. In the developing spinal cord, Netrin-1 and Shh guide commissural axons toward the midline. However, the combined inhibition of their activity in commissural axon turning assays does not completely abrogate turning toward floor plate tissue, suggesting that additional guidance cues are present. Here we show that the prototypic angiogenic factor VEGF is secreted by the floor plate and is a chemoattractant for commissural axons in vitro and in vivo. Inactivation of Vegf in the floor plate or of its receptor Flk1 in commissural neurons causes axon guidance defects, whereas Flk1 blockade inhibits turning of axons to VEGF in vitro. Similar to Shh and Netrin-1, VEGF-mediated commissural axon guidance requires the activity of Src family kinases. Our results identify VEGF and Flk1 as a novel ligand/receptor pair controlling commissural axon guidance.     

The role of sensory input in motor neuron sprouting control

Primary sensory neurons project to motor neurons directly or through interneurons and affect their activity. In our previous paper we showed that intramuscular sprouting can be affected by changing the sensory synaptic input to motor neurons. In this work, motor axon sprouting within a peripheral nerve (extramuscular sprouting) was induced by nerve injury at such a distance from muscle so as not to allow nerve-muscle trophic interactions. Two different procedures were carried out: (1) sciatic nerve crush and (2) sciatic nerve crush with homosegmental ipsilateral L3-L5 dorsal rhizotomy. The number of regenerating motor axons innervating extensor digitorum longus muscle was determined by in vivo muscle tension recordings and an index of their individual conduction rate was obtained by in vitro intracellular recordings of excitatory postsynaptic end-plate potentials in muscle fibers. The main findings were: (1) there are more regenerated axons distally from the lesion than parent axons proximally to the lesion (sprouting at the lesion); (2) sprouting at the lesion was negatively affected by homosegmental ipsilateral dorsal rhizotomy; (3) the number of motor axons innervating extensor digitorum longus muscle extrafusal fibers counted proximally to the lesion increased following nerve injury and regeneration but this did not occur when sensory input was lost. A transient innervation of extrafusal fibers by &#110 motor neurons may explain the increase of motor axons counted proximally to the lesion.     

The role of sensory input in motor neuron sprouting control

Primary sensory neurons project to motor neurons directly or through interneurons and affect their activity. In our previous paper we showed that intramuscular sprouting can be affected by changing the sensory synaptic input to motor neurons. In this work, motor axon sprouting within a peripheral nerve (extramuscular sprouting) was induced by nerve injury at such a distance from muscle so as not to allow nerve-muscle trophic interactions. Two different procedures were carried out: (1) sciatic nerve crush and (2) sciatic nerve crush with homosegmental ipsilateral L3-L5 dorsal rhizotomy. The number of regenerating motor axons innervating extensor digitorum longus muscle was determined by in vivo muscle tension recordings and an index of their individual conduction rate was obtained by in vitro intracellular recordings of excitatory postsynaptic end-plate potentials in muscle fibers. The main findings were: (1) there are more regenerated axons distally from the lesion than parent axons proximally to the lesion (sprouting at the lesion); (2) sprouting at the lesion was negatively affected by homosegmental ipsilateral dorsal rhizotomy; (3) the number of motor axons innervating extensor digitorum longus muscle extrafusal fibers counted proximally to the lesion increased following nerve injury and regeneration but this did not occur when sensory input was lost. A transient innervation of extrafusal fibers by gamma motor neurons may explain the increase of motor axons counted proximally to the lesion.     

Ephrin-As play a rhombomere-specific role in trigeminal motor axon projections in the chick embryo

In this study, we investigate the possible role of ephrin-Eph signaling in trigeminal motor axon projections. We find that EphA receptors are expressed at higher levels by rhombomere 2 (r2) trigeminal motor neurons than by r3 trigeminal motor neurons in the chick embryo. Mapping of rhombomere-specific axon projections shows that r2 and r3 trigeminal motor neurons project to different muscle targets, including the mandibular adductor and the intermandibularis muscles respectively. Ephrin-A5 is expressed in these muscles, especially in some regions of the intermandibularis muscle, and can cause growth cone collapse of both r2 and r3 motor axons in vitro. We demonstrate that in vivo overexpression of ephrin-A5 in the intermandibularis muscle, or overexpression of dominant-negative EphA receptors in trigeminal motor neurons leads to a reduction in branching of r3-derived motor axons specifically. Overexpression of full-length EphA receptors impairs the formation of r3 projections to the intermandibularis muscle. These findings indicate that ephrins and their Eph receptors play a role in trigeminal motor axon topographic mapping and in rhombomere 3-derived projections in particular.