Position-dependent properties of retinal axons and their growth cones

The formation of the very orderly neuronal projection from the retina to the optic tectum is not yet understood, but several mechanisms are thought to be involved in a coordinated fashion. These mechanisms may include mechanical or chemical guidance in channels, guidance by spatial gradients of positional markers, gradients of temporal (maturation) markers or specific inter-axon interactions (see ref. 1 for review). The last-mentioned mechanism could explain the fibre order found in optic nerve and tract. It requires that some or all growing retinal axons can distinguish between retinal axons of various origins and grow preferentially along retinal axons originating from the same area as themselves. The in vitro experiments described here show that growth cones from the temporal half of the chick retina grow preferentially along temporal axons, whereas growth cones from nasal retina do not distinguish between nasal and temporal axons.     

Local positional cues in the neuroepithelium guide retinal axons in embryonic Xenopus brain

Growing retinal axons home to their distant target, the tectum, even when they are displaced from their normal pathway. This argues for long-range guidance mechanisms in the embryonic brain. Growth cones may orientate to diffusible attractants released from the target, as proposed in other systems, or they may use a stable distribution of positional information in the neuroepithelium. To distinguish between these possibilities, small pieces of the presumptive optic tract, through which retinal axons will normally grow, were rotated by approximately 90 degrees either clockwise or counterclockwise. When the retinal axons later encountered the rotated neuroepithelium, they also turned clockwise or counterclockwise, in correspondence with the direction of rotation. This demonstrates that long-range navigation of retinal axons in the vertebrate brain is based partly on stable, local positional factors, rather than on remote diffusible factors.     

RGM is a repulsive guidance molecule for retinal axons

Axons rely on guidance cues to reach remote targets during nervous system development. A well-studied model system for axon guidance is the retinotectal projection. The retina can be divided into halves; the nasal half, next to the nose, and the temporal half. A subset of retinal axons, those from the temporal half, is guided by repulsive cues expressed in a graded fashion in the optic tectum, part of the midbrain. Here we report the cloning and functional characterization of a membrane-associated glycoprotein, which we call RGM (repulsive guidance molecule). This molecule shares no sequence homology with known guidance cues, and its messenger RNA is distributed in a gradient with increasing concentration from the anterior to posterior pole of the embryonic tectum. Recombinant RGM at low nanomolar concentration induces collapse of temporal but not of nasal growth cones and guides temporal retinal axons in vitro, demonstrating its repulsive and axon-specific guiding activity.     

Chemotropic guidance of developing axons in the mammalian central nervous system

In the developing nervous system, axons project considerable distances along stereotyped pathways to reach their targets. Axon guidance depends partly on the recognition of cell-surface and extracellular matrix cues derived from cells along the pathways. It has also been proposed that neuronal growth cones are guided by gradients of chemoattractant molecules emanating from their intermediate or final cellular targets. Although there is evidence that the axons of some peripheral neurons in vertebrates are guided by chemotropism and the directed growth of some central axons to their targets is consistent with such a mechanism, it remains to be determined whether chemotropism operates in the central nervous system. During development of the spinal cord, commissural axons are deflected towards a specialized set of midline neural epithelial cells, termed the floor plate, which could reflect guidance by substrate cues or by diffusible chemoattractant molecules. Here we provide evidence in support of chemotropic guidance by demonstrating that the rat floor-plate cells secrete a diffusible factor(s) that influences the pattern and orientation of commissural axon growth in vitro without affecting other embryonic spinal cord axons. These findings support the hypothesis that chemotropic mechanisms guide developing axons to their intermediate targets in the vertebrate CNS.     

Chemotropic effect of specific target epithelium in the developing mammalian nervous system

Developing nerve fibres are guided to their targets by specific directional cues which are thought to be expressed in the tissues along the route and may involve the extracellular matrix. Another possibility, that directional cues emanate from the target itself, is consistent with the recent demonstration of homing behaviour by ectopic retinal ganglion axons and our previous demonstration that early trigeminal neurites grow directly to their virgin peripheral target in vitro. Here we show that this chemotropic effect is precisely limited to the trigeminal system; trigeminal ganglion neurites grow directly to their own target field but not to the adjoining field, normally innervated by the geniculate ganglion; furthermore, the trigeminal field does not influence the growth of geniculate neurites. Also, when trigeminal ganglia are co-cultured with isolated tissue layers of their target, neurites grow only towards the epithelial and not the mesenchymal component. These findings suggest that trigeminal epithelium is specified to attract correct innervation and that pathway mesenchyme, in which preformed guidance cues have been postulated, may provide favourable conditions for nerve fibre growth but not govern its direction.     

Retinotopic order in the absence of axon competition

The retinotectal projection has long been studied experimentally and theoretically, as a model for the formation of topographic brain maps. Neighbouring retinal ganglion cells (RGCs) project their axons to neighbouring positions in the optic tectum, thus re-establishing a continuous neural representation of visual space. Mapping along this axis requires chemorepellent signalling from tectal cells, expressing ephrin-A ligands, to retinal growth cones, expressing EphA receptors. High concentrations of ephrin A, increasing from anterior to posterior, prevent temporal axons from invading the posterior tectum. However, the force that drives nasal axons to extend past the anterior tectum and terminate in posterior regions remains to be identified. We tested whether axon-axon interactions, such as competition, are required for posterior tectum innervation. By transplanting blastomeres from a wild-type (WT) zebrafish into a lakritz (lak) mutant, which lacks all RGCs, we created chimaeras with eyes that contained single RGCs. These solitary RGCs often extended axons into the tectum, where they branched to form a terminal arbor. Here we show that the distal tips of these arbors were positioned at retinotopically appropriate positions, ruling out an essential role for competition in innervation of the ephrin-A-rich posterior tectum. However, solitary arbors were larger and more complex than under normal, crowded conditions, owing to a lack of pruning of proximal branches during refinement of the retinotectal projection. We conclude that dense innervation is not required for targeting of retinal axons within the zebrafish tectum but serves to restrict arbor size and shape.     

Formation of target-specific neuronal projections in organotypic slice cultures from rat visual cortex.

A characteristic feature of the mammalian cortex is that projection neurons located in distinct cortical layers send their axons to different targets. In visual cortex, cells in layers 2 and 3 project to other cortical areas, whereas cells in layers 5 and 6 project to subcortical targets such as the lateral geniculate nucleus. The proper development of these projections is crucial for correct functioning of the visual system. Here we show that specific connections are established in an organotypic culture system in which rat visual cortex slices are co-cultured with another slice of the visual cortex or with a thalamic slice. The laminar origin and cellular morphology in vitro of cortical projections to other cortical regions or to subcortical targets are remarkably similar to those seen in vivo. In addition, axons of projecting cells are not restricted to particular pathways, but appear instead to grow directly towards their appropriate target. These observations raise the possibility that chemotropic attraction from the target areas may play an important part in the development of the cortical projection pattern.     

Wnt-Ryk signalling mediates medial-lateral retinotectal topographic mapping

Computational modelling has suggested that at least two counteracting forces are required for establishing topographic maps. Ephrin-family proteins are required for both anterior-posterior and medial-lateral topographic mapping, but the opposing forces have not been well characterized. Wnt-family proteins are recently discovered axon guidance cues. We find that Wnt3 is expressed in a medial-lateral decreasing gradient in chick optic tectum and mouse superior colliculus. Retinal ganglion cell (RGC) axons from different dorsal-ventral positions showed graded and biphasic response to Wnt3 in a concentration-dependent manner. Wnt3 repulsion is mediated by Ryk, expressed in a ventral-to-dorsal decreasing gradient, whereas attraction of dorsal axons at lower Wnt3 concentrations is mediated by Frizzled(s). Overexpression of Wnt3 in the lateral tectum repelled the termination zones of dorsal RGC axons in vivo. Expression of a dominant-negative Ryk in dorsal RGC axons caused a medial shift of the termination zones, promoting medially directed interstitial branches and eliminating laterally directed branches. Therefore, a classical morphogen, Wnt3, acting as an axon guidance molecule, plays a role in retinotectal mapping along the medial-lateral axis, counterbalancing the medial-directed EphrinB1-EphB activity.     

Guidance of optic nerve fibres by N-cadherin adhesion molecules

The dendritic branches (neurites) of developing neurons migrate along specific pathways to reach their targets. It has been suggested that this migration is guided by factors present on the surface of other neurons or glial cells. The molecular nature of such factors, however, remains to be elucidated. N-cadherin is a cell-surface glycoprotein which belongs to the cadherin family of cell-cell adhesion molecules. This adhesion molecule is expressed in various neuronal cells as well as in glial cells of the central and peripheral nervous systems in vertebrate embryos and recent immunological studies suggested that N-cadherin may play a role in guiding the migration of neurites on myotubes or astrocytes. To further examine this possibility, we used a molecular-genetic approach; that is, we examined the outgrowth of chicken embryonic optic axons on monolayer cultures of Neuro 2a or L cells transfected with the complementary DNA encoding chicken N-cadherin. The data indicate that N-cadherin is used as a guide molecule for the migration of optic axons on cell surfaces.     

Substance P-immunoreactive retinal ganglion cells and their central axon terminals in the rabbit

Retinal ganglion cells are the projection neurons that link the retina to the brain. Peptide immunoreactive cells in the ganglion cell layer (GCL) of the mammalian retina have been noted but their identity has not been determined. We now report that, in the rabbit, 25-35% of all retinal ganglion cells contain substance P-like (SP) immunoreactivity. They were identified by either retrograde transport of fluorescent tracers injected into the superior colliculus, or by retrograde degeneration after optic nerve section. SP immunoreactive cells are present in all parts of the retina and have medium to large cell bodies with dendrites that ramify extensively in the proximal inner plexiform layer. Their axons terminate in the dorsal lateral geniculate nucleus, superior colliculus and accessory optic nuclei, and these terminals disappear completely after contralateral optic nerve section and/or eye enucleation. In the dorsal lateral geniculate nucleus large, beaded, immunoreactive axons and varicosities make up a narrow plexus just below the optic tract, where they define a new geniculate lamina. The varicosities make multiple synaptic contacts with dendrites of dorsal lateral geniculate nucleus projection neurons and presumptive interneurons in complex glomerular neuropil. This is direct evidence that some mammalian retinal ganglion cells contain substance P-like peptides and strongly suggests that, in the rabbit, substance P (or related tachykinins) may be a transmitter or modulator in a specific population or populations of retinal ganglion cells.     

Pathway selection by growth cones of identified motoneurones in live zebra fish embryos

How is the adult pattern of connections between motoneurones and the muscles that they innervate established during vertebrate development? Populations of motoneurones are thought to follow one of two patterns of development: (1) motor axons initially follow stereotyped pathways and project to appropriate regions of the developing muscle or (2) motor axons initially project to some regions that are incorrect, the inappropriate projections being eliminated subsequently. Here we observed individually identified motoneurones in live zebra fish embryos as they formed growth cones and as their growth cones navigated towards their targets. We report that from axogenesis, each motor axon followed a stereotyped pathway and projected only to the specific region of the muscle appropriate for its adult function. In addition, the peripheral arbor established by each motoneurone was restricted to a stereotyped region of its own segment and did not overlap with the peripheral arbor of the other motoneurones in that segment. We conclude that the highly stereotyped pattern of innervation seen in the adult is due to initial selection of the appropriate pathway, rather than elimination of incorrect projections.     

Growth-cone attraction to netrin-1 is converted to repulsion by laminin-1.

Growing axons are guided by both diffusible and substrate-bound factors. Growth cones of retinal neurons exhibit chemoattractive turning towards the diffusible factor netrin-1 in vitro and are guided into the optic nerve head (ONH) by localized netrin-1. Here we report that, in Xenopus, laminin-1 from the extracellular matrix (ECM), converts netrin-mediated attraction into repulsion. A soluble peptide fragment of laminin-1 (YIGSR) mimics this laminin-induced conversion. Low levels of cyclic AMP in growth cones also lead to the conversion of netrin-induced attraction into repulsion, and we show that the amount of cAMP decreases in the presence of laminin-1 or YIGSR, suggesting a possible mechanism for laminin's effect. At the netrin-1-rich ONH, where axons turn sharply to leave the eye, laminin-1 is confined to the retinal surface. Repulsion from the region in which laminin and netrin are coexpressed may help to drive axons into the region where only netrin is present, providing a mechanism for their escape from the retinal surface. Consistent with this idea, YIGSR peptides applied to the developing retina cause axons to be misdirected at the ONH. These findings indicate that ECM molecules not only promote axon outgrowth, but also modify the behaviour of growth cones in response to diffusible guidance cues.     

Nerve fibre topography in the retinal projection to the tectum.

The orderly layout of visual fibres in the optic nerves of cichlid fishes suggests that they are guided most of the way to the brain by contact with their neighbours. Before they reach the visual map in the tectum, however, their arrangement in the cross-section of the nerve is reorganised in a way which requires longer-range guidance.     

Retinofugal fibres change conduction velocity and diameter between the optic nerve and tract in ferrets

In earlier studies of central nervous fibre tracts, it was tacitly assumed that individual axons are relatively uniform along their length. In the retinofugal pathway in particular, axon diameter, myelin thickness and correlated conduction properties have been treated as constant throughout the optic nerve, chiasm and tract. We report here that the conduction velocities of fibres contributing to the early components of the compound action potential are significantly greater in the optic tract than in the optic nerve of ferrets, and also that the diameters of the largest retinofugal fibres increase from nerve to tract. This observation raises significant questions about the developmental mechanisms in the central nervous system that relate the axons, their diameters, and the glia with which they are myelinated. In addition, it indicates that studies that have relied on the constancy of conduction velocity along the retinofugal course may require reappraisal.     

Segregation of functionally distinct axons in the monkey's optic tract

The classical neuro-ophthalmologic literature describes the organization of the primate's optic tract as containing a single topographic representation of the complete contralateral visual hemifield. In contrast, cats have separate visual field representations for the optic axons of the functionally distinct retinal ganglion cell classes. As the line of decussation for each ganglion cell class in the cat occupies a different location on the retinal surface, whereas in primates they are all superimposed, such a species difference might be expected. We report that implants of horseradish peroxidase placed in either the deep or superficial extremes of the monkey's optic tract produce retrograde labelling of distinct retinal ganglion cell classes, and produce anterograde labelling confined to distinct laminae of the lateral geniculate nucleus. Hence, the optic tract of the primate cannot contain a single representation of the contralateral visual hemifield; rather, independent visual field representations for the functionally distinct optic axons must exist. Their anatomical segregation may account for the clinical observation of selective impairments of distinct visual abilities following partial interruption of the optic tract in man.     

Disoriented pathfinding by pioneer neurone growth cones deprived of filopodia by cytochalasin treatment

A major question in developmental neurobiology is how developing nerve cells accurately extend processes to establish connections with their target cells. This problem involves both the nature of cues for growth cone guidance and also the question of how growth cones survey their environment for cues and respond by altering their direction of migration. The filopodia which normally extend from neuronal growth cones have been shown to affect growth cone steering in vitro and it has been proposed that they function in vivo in the detection of and response to guidance cues. This hypothesis could be tested in vivo if growth cones which normally have filopodia could be induced to migrate in their absence. The pair of Ti1 neurones are the first neurones to extend axons through the limb buds of embryonic grasshoppers. We report here an examination of the migration of Ti1 pioneer growth cones deprived of filopodia by culture in agents which disrupt actin microfilaments. Under these conditions, axons continue to extend but a large percentage of growth cones are highly disoriented. Our results indicate that Ti1 filopodia are not necessary for axonal elongation in vivo but that they are important for correctly oriented growth cone steering.     

Intrinsic differences in the growth rate of early nerve fibres related to target distance

Target field innervation in the developing vertebrate nervous system coincides with the onset of important trophic interactions. Two factors that determine the timing of this event are the distance axons have to grow to reach their targets, which are known to vary, and the rate at which they grow. There have been few studies of axonal growth rate at this stage of development and no comparative study of the relationship between growth rate and target distance. Embryonic chick cranial sensory neurons are located in discrete ganglia and the distance axons have to grow to reach their targets is different for each ganglion, ranging from several hundred to several thousand microns. Here, I show that these neurons differ in their in vivo growth rates; neurons with more distant targets growing faster. In vitro, single isolated neurons from each of these populations grow at a similar rate to that observed in vivo, indicating that growth rate is an intrinsically determined property of neurons before they reach their targets.     

Order in the initial retinotectal map in Xenopus: a new technique for labelling growing nerve fibres

Retinal nerve fibres form an orderly map of visual space in several centres in the vertebrate brain. Such topographic maps are a common feature of central nervous system organization, yet the way in which they develop is poorly understood. Early nerve projections in the fetal and neonatal mammalian brain have been found in several cases to be less restricted than those in the adult, suggesting that nerve fibres may initially form a diffuse set of connections in their target structure from which the adult map is sculpted by the elimination of terminals. Indeed, previous electrophysiological data indicate that the retinotectal map in Xenopus laevis might be initially disorganized. We report here, however, that the retinotectal projection is ordered from the beginning of tectal innervation (stage 39/40). We demonstrate this first autoradiographically by tracing groups of growing ganglion cell axons which we labelled by incubating sectors of eye rudiments, before axonal outgrowth, in 3H-proline and replacing them orthotopically. Separate labelling of dorsal and ventral parts of the initial projection showed that retinal fibres are organized topographically, as in the adult, in the tectal rudiment and throughout much of the pathway. Second, we show that visual responses are ordered in the tectum from the first stage that they can be mapped (stage 40). We conclude that the topographic ordering of retinotectal connections develops as a result of directed axonal outgrowth.     

Retinal ganglion cells lose response to laminin with maturation

The decisive role played by adhesive interactions between neuronal processes and the culture substrate in determining the form and extent of neurite outgrowth in vitro has greatly influenced ideas about the mechanisms of axonal growth and guidance in the vertebrate nervous system. These studies have also helped to identify adhesive molecules that might be involved in guiding axonal growth in vivo. One candidate molecule is laminin, a major glycoprotein of basal laminae which has been shown to induce a wide variety of embryonic neurones to extend neurites in culture. Moreover, laminin is found in large amounts in injured nerves that can successfully regenerate but is absent from nerves where regeneration fails. However, it is unclear to what extent the mechanisms that regulate axonal regeneration also operate in the embryo when axon outgrowth is initiated. Here we have examined the substrate requirements for neurite outgrowth in vitro by chick embryo retinal ganglion cells, the only cells in the retina to send axons to the brain. We show that while retinal ganglion cells from embryonic day 6 (E6) chicks extend profuse neurites on laminin, those from E11 do not, although they retain the ability to extend neurites on astrocytes via a laminin-independent mechanism. This represents the first evidence that central nervous system neurones may undergo a change in their substrate requirements for neurite outgrowth as they mature.     

Axon guidance in cultured epithelial fragments of Drosophila wing

Growing axons can be guided by a number of different cues: adhesive substrates, diffusible factors, electrical fields and even factors intrinsic to the neurone itself have all been shown to affect axon orientation and outgrowth in vitro. However, in most intact systems it has proved difficult to test directly the role played by these putative guidance cues. Here, we describe a system, the developing wing of the fruitfly, in which we have tested simultaneously two putative guidance mechanisms, physical constraints to axon growth (channels) and the position of neuronal somata (guideposts), using surgical techniques. We show that pioneer sensory axons can navigate correctly and form their normal stereotyped pattern of axon bundles in wing fragments that apparently lack both physical and neural cues. This technique allows access to the surface along which neuronal pathfinding takes place, making possible a wide range of experimental manipulations on the developing system.