TDP-43 has been found in inclusion bodies of multiple neurological disorders, including amyotrophic lateral sclerosis, frontotemporal dementia, Parkinson's disease and Alzheimer's disease. Mutations in the TDP-43 encoding gene, TARDBP, have been subsequently reported in sporadic and familial ALS patients. In order to investigate the pathogenic nature of these mutants, the effects of three consistently reported TARDBP mutations (A315T, G348C and A382T) were tested in cell lines, primary cultured motor neurons and living zebrafish embryos. Each of the three mutants and wild-type (WT) human TDP-43 localized to nuclei when expressed in COS1 and Neuro2A cells by transient transfection. However, when expressed in motor neurons from dissociated spinal cord cultures these mutant TARDBP alleles, but less so for WT TARDBP, were neurotoxic, concomitant with perinuclear localization and aggregation of TDP-43. Finally, overexpression of mutant, but less so of WT, human TARDBP caused a motor phenotype in zebrafish (Danio rerio) embryos consisting of shorter motor neuronal axons, premature and excessive branching as well as swimming deficits. Interestingly, knock-down of zebrafisfh tardbp led to a similar phenotype, which was rescued by co-expressing WT but not mutant human TARDBP. Together these approaches showed that TARDBP mutations cause motor neuron defects and toxicity, suggesting that both a toxic gain of function as well as a novel loss of function may be involved in the molecular mechanism by which mutant TDP-43 contributes to disease pathogenesis.

Mutations in the SOD1 and TARDBP genes have been commonly identified in Amyotrophic Lateral Sclerosis (ALS). Recently, mutations in the Fused in sarcoma gene (FUS) were identified in familial (FALS) ALS cases and sporadic (SALS) patients. Similarly to TDP-43 (coded by TARDBP gene), FUS is an RNA binding protein. Using the zebrafish (Danio rerio), we examined the consequences of expressing human wild-type (WT) FUS and three ALS–related mutations, as well as their interactions with TARDBP and SOD1. Knockdown of zebrafish Fus yielded a motor phenotype that could be rescued upon co-expression of wild-type human FUS. In contrast, the two most frequent ALS–related FUS mutations, R521H and R521C, unlike S57Δ, failed to rescue the knockdown phenotype, indicating loss of function. The R521H mutation caused a toxic gain of function when expressed alone, similar to the phenotype observed upon knockdown of zebrafish Fus. This phenotype was not aggravated by co-expression of both mutant human TARDBP (G348C) and FUS (R521H) or by knockdown of both zebrafish Tardbp and Fus, consistent with a common pathogenic mechanism. We also observed that WT FUS rescued the Tardbp knockdown phenotype, but not vice versa, suggesting that TARDBP acts upstream of FUS in this pathway. In addition we observed that WT SOD1 failed to rescue the phenotype observed upon overexpression of mutant TARDBP or FUS or upon knockdown of Tardbp or Fus; similarly, WT TARDBP or FUS also failed to rescue the phenotype induced by mutant SOD1 (G93A). Finally, overexpression of mutant SOD1 exacerbated the motor phenotype caused by overexpression of mutant FUS. Together our results indicate that TARDBP and FUS act in a pathogenic pathway that is independent of SOD1. Mutations in the SOD1, TARDBP, and FUS genes have been commonly identified in Amyotrophic Lateral Sclerosis (ALS). However, possible interactions between these ALS–causative genetic mutations have not been examined. Here we expressed each of three human FUS mutations (R521H, R521C, and S57Δ) in zebrafish embryos, with or without knocking down the zebrafish homolog Fus, and observed a motor phenotype consisting of significant behavioral (touch-evoked escape response) and cellular (shortened axonal projections from motor neurons) deficits due to loss of function for the R521H and R521C mutations and/or toxic gain of function solely for the R521H mutation. Wild-type FUS could rescue the Tardbp knockdown phenotype, but not vice versa, suggesting that TARDBP is upstream of FUS in this pathway responsible for motor neuron disorder. Furthermore, neither TARDBP nor FUS were able to modify and/or rescue the motor phenotype caused by mutant SOD1, and likewise SOD1 failed to rescue the phenotype of zebrafish expressing mutant TARDBP or FUS. Our results indicate that TARDBP acts upstream of FUS in a pathogenic pathway that is distinct from that of SOD1.

The mutations P56S and T46I in the gene encoding vesicle-associated membrane protein-associated protein B/C (VAPB) cause ALS8, a familial form of amyotrophic lateral sclerosis (ALS). Overexpression of mutant forms of VAPB leads to cytosolic aggregates, suggesting a gain of function of the mutant protein. However, recent work suggested that the loss of VAPB function could be the major mechanism leading to ALS8. Here, we used multiple genetic and experimental approaches to study whether VAPB loss of function might be sufficient to trigger motor neuron degeneration. In order to identify additional ALS-associated VAPB mutations, we screened the entire VAPB gene in a cohort of ALS patients and detected two mutations (A145V and S160Δ). To directly address the contribution of VAPB loss of function in ALS, we generated zebrafish and mouse models with either a decreased or a complete loss of Vapb expression. Vapb knockdown in zebrafish led to swimming deficits. Mice knocked-out for Vapb showed mild motor deficits after 18 months of age yet had innervated neuromuscular junctions (NMJs). Importantly, overexpression of VAPB mutations were unable to rescue the motor deficit caused by Vapb knockdown in zebrafish and failed to cause a toxic gain-of-function defect on their own. Thus, Vapb loss of function weakens the motor system of vertebrate animal models but is on its own unable to lead to a complete ALS phenotype. Our findings are consistent with the notion that VAPB mutations constitute a risk factor for motor neuron disease through a loss of VAPB function.

Development and function of highly polarized cells such as neurons depend on microtubule-associated intracellular transport, but little is known about contributions of specific molecular motors to the establishment of synaptic connections. In this study, we investigated the function of the Kinesin I heavy chain Kif5aa during retinotectal circuit formation in zebrafish. Targeted disruption of Kif5aa does not affect retinal ganglion cell differentiation, and retinal axons reach their topographically correct targets in the tectum, albeit with a delay. In vivo dynamic imaging showed that anterograde transport of mitochondria is impaired, as is synaptic transmission. Strikingly, disruption of presynaptic activity elicits upregulation of Neurotrophin-3 (Ntf3) in postsynaptic tectal cells. This in turn promotes exuberant branching of retinal axons by signaling through the TrkC receptor (Ntrk3). Thus, our study has uncovered an activity-dependent, retrograde signaling pathway that homeostatically controls axonal branching.

Dynactin subunit 1 is the largest subunit of the dynactin complex, an activator of the molecular motor protein complex dynein. Reduced levels of DCTN1 mRNA and protein have been found in sporadic amyotrophic lateral sclerosis (ALS) patients, and mutations have been associated with disease, but the role of this protein in disease pathogenesis is still unknown. We characterized a Dynactin1a depletion model in the zebrafish embryo and combined in vivo molecular analysis of primary motor neuron development with live in vivo axonal transport assays in single cells to investigate ALS-related defects. To probe neuromuscular junction (NMJ) function and organization we performed paired motor neuron-muscle electrophysiological recordings and GCaMP calcium imaging in live, intact larvae, and the synapse structure was investigated by electron microscopy. Here we show that Dynactin1a depletion is sufficient to induce defects in the development of spinal cord motor neurons and in the function of the NMJ. We observe synapse instability, impaired growth of primary motor neurons, and higher failure rates of action potentials at the NMJ. In addition, the embryos display locomotion defects consistent with NMJ dysfunction. Rescue of the observed phenotype by overexpression of wild-type human DCTN1-GFP indicates a cell-autonomous mechanism. Synaptic accumulation of DCTN1-GFP, as well as ultrastructural analysis of NMJ synapses exhibiting wider synaptic clefts, support a local role for Dynactin1a in synaptic function. Furthermore, live in vivo analysis of axonal transport and cytoskeleton dynamics in primary motor neurons show that the phenotype reported here is independent of modulation of these processes. Our study reveals a novel role for Dynactin1 in ALS pathogenesis, where it acts cell-autonomously to promote motor neuron synapse stability independently of dynein-mediated axonal transport.