The L1 cell adhesion molecule (L1CAM) participates in neuronal development. Mutations in the human L1 gene can cause the neurological disorder CRASH (corpus callosum hypoplasia, retardation, adducted thumbs, spastic paraplegia, and hydrocephalus). This study presents genetic data that shows that L1-like adhesion gene 2 (LAD-2), a Caenorhabditis elegans L1CAM, functions in axon pathfinding. In the SDQL neuron, LAD-2 mediates dorsal axon guidance via the secreted MAB-20/Sema2 and PLX-2 plexin receptor, the functions of which have largely been characterized in epidermal morphogenesis. We use targeted misexpression experiments to provide in vivo evidence that MAB-20/Sema2 acts as a repellent to SDQL. Coimmunoprecipitation assays reveal that MAB-20 weakly interacts with PLX-2; this interaction is increased in the presence of LAD-2, which can interact independently with MAB-20 and PLX-2. These results suggest that LAD-2 functions as a MAB-20 coreceptor to secure MAB-20 coupling to PLX-2. In vertebrates, L1 binds neuropilin1, the obligate receptor to the secreted Sema3A. However, invertebrates lack neuropilins. LAD-2 may thus function in the semaphorin complex by combining the roles of neuropilins and L1CAMs.

Zebrafish (Danio rerio) is a widely used model organism in genetics and developmental biology research. Genetic screens have proven useful for studying embryonic development of the nervous system in vivo, but in vitro studies utilizing zebrafish have been limited. Here, we introduce a robust zebrafish primary neuron culture system for functional nerve growth and guidance assays. Distinct classes of central nervous system neurons from the spinal cord, hindbrain, forebrain, and retina from wild type zebrafish, and fluorescent motor neurons from transgenic reporter zebrafish lines, were dissociated and plated onto various biological and synthetic substrates to optimize conditions for axon outgrowth. Time-lapse microscopy revealed dynamically moving growth cones at the tips of extending axons. The mean rate of axon extension in vitro was 21.4±1.2 µm hr−1 s.e.m. for spinal cord neurons, which corresponds to the typical ∼0.5 mm day−1 growth rate of nerves in vivo. Fluorescence labeling and confocal microscopy demonstrated that bundled microtubules project along axons to the growth cone central domain, with filamentous actin enriched in the growth cone peripheral domain. Importantly, the growth cone surface membrane expresses receptors for chemotropic factors, as detected by immunofluorescence microscopy. Live-cell functional assays of axon extension and directional guidance demonstrated mammalian brain-derived neurotrophic factor (BDNF)-dependent stimulation of outgrowth and growth cone chemoattraction, whereas mammalian myelin-associated glycoprotein inhibited outgrowth. High-resolution live-cell Ca2+-imaging revealed local elevation of cytoplasmic Ca2+ concentration in the growth cone induced by BDNF application. Moreover, BDNF-induced axon outgrowth, but not basal outgrowth, was blocked by treatments to suppress cytoplasmic Ca2+ signals. Thus, this primary neuron culture model system may be useful for studies of neuronal development, chemotropic axon guidance, and mechanisms underlying inhibition of neural regeneration in vitro, and complement observations made in vivo.

The local translation of β-actin is one mechanism proposed to regulate spatially-restricted actin polymerization crucial for nearly all aspects of neuronal development and function. However, the physiological significance of localized β-actin translation in neurons has not yet been demonstrated in vivo. To investigate the role of β-actin in the mammalian central nervous system (CNS), we characterized brain structure and function in a CNS-specific β-actin knock-out mouse (CNS-ActbKO). β-actin was rapidly ablated in the embryonic mouse brain, but total actin levels were maintained through upregulation of other actin isoforms during development. CNS-ActbKO mice exhibited partial perinatal lethality while survivors presented with surprisingly restricted histological abnormalities localized to the hippocampus and cerebellum. These tissue morphology defects correlated with profound hyperactivity as well as cognitive and maternal behavior impairments. Finally, we also identified localized defects in axonal crossing of the corpus callosum in CNS-ActbKO mice. These restricted defects occurred despite the fact that primary neurons lacking β-actin in culture were morphologically normal. Altogether, we identified novel roles for β-actin in promoting complex CNS tissue architecture while also demonstrating that distinct functions for the ubiquitously expressed β-actin are surprisingly restricted in vivo.

The proper localization of ß-actin mRNA and protein is essential for growth cone guidance and axon elongation in cultured neurons. In addition, decreased levels of ß-actin mRNA and protein have been identified in the growth cones of motor neurons cultured from a mouse model of Spinal Muscular Atrophy (SMA), suggesting that ß-actin loss-of-function at growth cones or pre-synaptic nerve terminals could contribute to the pathogenesis of this disease. However, the role of ß-actin in motor neurons in vivo and its potential relevance to disease has yet to be examined. We therefore generated motor neuron specific ß-actin knock-out mice (Actb-MNsKO) to investigate the function of ß-actin in motor neurons in vivo. Surprisingly, ß-actin was not required for motor neuron viability or neuromuscular junction maintenance. Skeletal muscle from Actb-MNsKO mice showed no histological indication of denervation and did not significantly differ from controls in several measurements of physiologic function. Finally, motor axon regeneration was unimpaired in Actb-MNsKO mice, suggesting that ß-actin is not required for motor neuron function or regeneration in vivo.