The role of cell adhesion molecules in neurite outgrowth on Müller cells

The roles of neural cell adhesion molecule (NCAM), L1, N-cadherin, and integrin in neurite outgrowth on various substrates were studied. Antibodies against these cell surface molecules were added to explants of chick retina and the neurites from retinal ganglion cells were examined for effects of the antibodies on neurite length and fasciculation. On laminin, an anti-integrin antibody completely inhibited neurite outgrowth. The same antibody did not inhibit neurite outgrowth on polylysine or Müller cells. Antibodies to NCAM, L1, and N-cadherin did not significantly inhibit neurite outgrowth on laminin but produced significant inhibition on Müller cells. The inhibition of neurite outgrowth on glia by anti-L1 antibodies supports the hypothesis that L1 is capable of acting in a heterophilic binding mechanism. On laminin, both anti-N-cadherin and anti-L1 caused defasciculation of neurites from retinal ganglion cells, while anti-NCAM did not. None of these antibodies produced defasciculation on Müller cells. The results indicate that these three cell adhesion molecules may be very important in interactions with glia as axons grow from the retina to the tectum and may be less important in axon-axon interactions along this pathway. No evidence was found supporting the role of integrins in axon growth on Müller cells.     

A glial cell line promotes the outgrowth of neurites from embryonic Xenopus retina

A glial cell line (XR1 cell line) derived from Xenopus retinal neuroepithelium was examined for neurite outgrowth promoting activity. A monolayer of the XR1 cells serves as an excellent substrate upon which embryonic retinal explants attach and freely elaborate neurites. The XR1 neurite outgrowth promoting activity is not secreted into the medium, but is laid down directly on the substrate where it remains active after lysing the cells by hypoosmotic shock. A polyclonal antiserum raised against membranes of the XR1 cells was effective in blocking neurite outgrowth on XR1 conditioned collagen. It is proposed that the neurite outgrowth promoting factors produced by the XR1 cells are associated with the extracellular matrix and possibly glial specific.     

Factors affecting neurite outgrowth of occipital cortical explants

The effects of various substrata including laminin, collagen gel, collagen I, and human amniotic basement membrane on neurite outgrowth of occipital cortical and diencephalic explants were studied. The results showed that the extent and pattern of growing neurites of cortical explants varied considerably depending on the substrata used. While an elaborated network of growing neurites was observed when cortical explants were plated on laminin, the most extensive neurite outgrowth was observed when collagen gel was used as the substratum. In contrast, diencephalic explants did not grow on most of the substrata. The significance of the findings are discussed.     

Differential responses of temporal and nasal retinal neurites to regional-specific cues in the mouse retinofugal pathway

Retinal explants of mouse embryos were cultured together with explants of different regions in the retinofugal pathway in order to investigate whether ventral temporal (VT) and dorsal nasal (DN) retinal neurites showed differential responses to regional-specific cues in the pathway. In the presence of the chiasm, biased outgrowth of retinal neurites was found in explants of both retinal regions, which was accompanied by a reduction in total neurite growth in the VT but not the DN retina. Such differential responses to the diffusible negative influence were also observed when explants of two retinal origins were cocultured with the ventral diencephalon, but were not found with the dorsal diencephalon that contains targets of the optic axons. Indeed, extensive neurite invasion was found in the dorsal diencephalic explants and this ingrowth was more prominent for VT than DN neurites, showing a difference in axons from a distinct position in the retina to contact-mediated stimulatory activity within the target nuclei. We conclude that neurites from different regions of the retina show differential responses to the regional-specific cues in the diencephalon. These cues exist in both diffusible and contact-mediated forms that may shape the characteristic course and organization of retinal axons in decision regions of the optic pathway and the visual targets.     

Enhancement of Neurite Outgrowth and Aspartate-Glutamate Uptake Systems in Retinal Explants Cultured with Chick Gizzard Extract

Chicken gizzard extract promoted a long and radially directed neurite outgrowth from retinal explants of 8-day-old chick embryo in cultures of 2–3 days. The neurite outgrowth from retinal explants cultured in the absence of gizzard extract was short and restricted to the explant perimeter. The neurite outgrowth promoted by gizzard extract depended strictly on several factors. (a) Fetal calf serum and polycationic substratum were required in this culture system, (b) Pretreatment of the polyornithine-coated substratum with gizzard extract allowed the retinal explants to extend neurites even in the absence of gizzard extract in the medium. (c) Maximal neurite outgrowth was observed in retinal explants dissected from 8-day embryos, but thereafter the explants’response to gizzard extract rapidly declined and was almost lost at the 12th day. As a biochemical parameter of differentiation of cultured neuroretina, uptake systems for neurotransmitter candidates were examined in homogenates of retinal explants cultured in the absence or presence of gizzard extract. After 3 days in culture with gizzard extract, the uptake increased for aspartate and glutamate 1.6- to 1.8-fold and for γ-aminobutyric acid to a lesser degree when examined at a concentration for high-affinity uptake (10^-6M). In contrast, the uptake capacity for glycine, choline, and dopamine was not altered in explants cultured with or without gizzard extract. Kinetic analysis showed that the enhanced capacity to accumulate aspartate was not due to an alteration of K^m, but to an increase of V^max. The results suggest that one or several factors in chick gizzard muscle promote not only neurite outgrowth but also the aspartate-glutamate uptake systems in the developing neuroretina, probably related to ganglion cells.     

In vitro growth properties of Xenopus retinal neurons undergo developmental modulation

To determine whether Xenopus retinal neurons undergo intrinsic developmental changes in growth properties, retinal explants from embryos and tadpoles of different stages were grown on laminin, fibronectin, and collagen I in serum-free media. Growth was assayed in terms of a neurite growth index (NGI) and the appearance of clockwise bundles, or a clockwise growth index (CGI). The first neurites from stage 25 optic vesicles are pioneers and display a unique growth phenotype; they emerge rapidly, survive for a short time, show little substrate preferences for growth (they grow almost as well on BSA as they do on laminin and fibronectin), and form no clockwise bundles under any conditions. Neurites from progressively older retinas (stages 32-37) share with stage 25 neurites the rapid outgrowth pattern, but begin to show substrate preferences and clockwise growth. From stage 40 to 50, the mature growth pattern is expressed; a lag in initial outgrowth, long-term survival, distinct substrate preferences (they grow 10 times better on laminin and fibronectin than on BSA) and display robust clockwise growth patterns on laminin and fibronectin. The acquisition of clockwise growth is independent of optic fiber contact with the tectum or exposure to diffusible factors from mature brain tissues. The results suggest that retinal neurons undergo developmental modulation of surface adhesive properties and/or cytoskeletal organization.     

Growth cone interactions with a glial cell line from embryonic Xenopus retina

We have isolated a nonneuronal cell line from Xenopus retinal neuroepithelium (XR1 cell line). On the basis of immunocytochemical characterization using monoclonal antibodies generated in our laboratory as well as several other glial-specific antibodies, we have established that the XR1 cells are derived from embryonic astroglia. A monolayer of XR1 cells serves as an excellent substrate upon which embryonic retinal explants attach and elaborate neurites. This neurite outgrowth promoting activity appears not to be secreted into the medium, as medium conditioned by XR1 cells is ineffective in promoting outgrowth. Cell-free substrates were prepared to examine whether outgrowth promoting activity is also associated with the XR1 extracellular matrix (ECM). Substrates derived from XR1 cells grown on collagen are still capable of promoting outgrowth following osmotic shock and chemical extraction. This activity does not appear to be associated with laminin or fibronectin. Scanning electron microscopy was used to examine growth cones of retinal axons on XR1 cells and other substrates that supported neurite outgrowth. Growth cones and neurites growing on a monolayer of XR1 cells, or on collagen conditioned by XR1 cells, closely resemble the growth cones of retinal ganglion cells in vivo. A polyclonal antiserum (NOB1) generated against XR1 cells effectively and specifically inhibits neurite outgrowth on XR1-conditioned collagen. We therefore propose that neurite outgrowth promoting factors produced by these cells are associated with the extracellular matrix and may be glial specific.     

Neurotrophin-4/5 (NT-4/5) increases adult rat retinal ganglion cell survival and neurite outgrowth in vitro

Retinal ganglion cell (RGC) survival and neurite outgrowth were investigated in retinal explants from adult rats. Neutrotrophin-4/5 (NT-4/5) caused dose-dependent increases in neurite outgrowth with one-half maximal effects at approximately 0.5 ng/ml and maximal effects at 5 ng/ml. In explants treated for 7 days, the actions of NT-4/5 were similar to those of brain-derived neurotrophic factor (BDNF); with either neurotrophin, nearly twice as many RGCs survived and there was a two- to threefold increase in the number of neurites formed by RGCs. Combinations of saturating concentrations of NT-4/5 and BDNF did not enhance these in vitro effects, implying that both neurotrophins share a common signaling pathway. In contrast, nerve growth factor (NGF), neurotrophin-3 (NT-3), or ciliary nuerotrophic factor (CNTF) appeared to exert minimal influences on RGC survival or neurite outgrowth. 1994 John Wiley & Sons, Inc.     

Promotion of retinal neurite outgrowth by substratum-bound fibronectin

We have examined conditions under which aggregates of embryonic chick neural retina will extend neurities in vitro. Trypsin-dispersed cells from 7-day embryonic chick neural retina were aggregated in rotation culture for 8 hr and maintained in serum-free medium on a variety of standard culture substrate. Aggregates extend few neurites on untreated plastic, glass, or collagen substrata. However, pretreatment of these substrata with human plasma fibronectin enhances their capacity to support retinal neurite outgrowth. Aggregates cultured on fibronectin-treated substrata extend long, radially oriented neurites within 36 hr in vitro. The morphology of these neurites is distinct from that seen when aggregates are cultured on polylysine-treated substrata. In the latter case, neurites are highly branched and grow concentrically around the aggregate perimeter. Addition of fibronectin to polylysine-treated substrata stimulates radial neurite outgrowth. Promotion of neurite outgrowth is dependent on the amount of fibronectin bound to the culture substratum and on the pH at which binding occurs. The requirements for fibronectin-mediated neurite outgrowth are more stringent than those previously reported for fibroblast attachment and spreading.     

Differential neurite outgrowth is required for axon specification by cultured hippocampal neurons

Formation of an axon is the first morphological evidence of neuronal polarization, visible as a profound outgrowth of the axon compared with sibling neurites. One unsolved question on the mechanism of axon formation is the role of axon outgrowth in axon specification. This question was difficult to assess, because neurons freely extend their neurites in a conventional culture. Here, we leveraged surface nano/micro‐modification techniques to fabricate a template substrate for constraining neurite lengths of cultured neurons. Using the template, we asked (i) Do neurons polarize even if all neurites cannot grow sufficiently long? (ii) Would the neurite be fated to become an axon if only one was allowed to grow long? A pattern with symmetrical short paths (20 μm) was used to address the former question, and an asymmetrical pattern with one path extended to 100 μm for the latter. Axon formation was evaluated by tau‐1/MAP2 immunostaining and live‐cell imaging of constitutively‐active kinesin‐1. We found that (1) neurons cannot polarize when extension of all neurites is restricted and that (2) when only a single neurite is permitted to grow long, neurons polarize and the longest neurite becomes the axon. These results provide clear evidence that axon outgrowth is required for its specification.     

Selective inhibition of ventral temporal but not dorsal nasal neurites from mouse retinal explants during contact with chondroitin sulphate

We have determined whether chondroitin sulphate (CS) glycosaminoglycans are sufficient to direct a selective inhibition of neurite growth from ventral temporal (VT) but not from dorsal nasal (DN) retina in mouse embryos; this may underlie the formation of axon divergence in the optic chiasm. Explants from the retinal region of embryonic day-14 mouse were grown on a laminin–polylysine substrate near to a circular spot coated with CS. In control cultures, in which no CS was added to the spot, both VT and DN retinal neurites grew extensively into the coated territory. When presented with spots coated with 10 mg/ml CS, neurite growth from the VT retina into the CS territory was dramatically reduced but that from the DN retina was not significantly affected. The selective inhibition to VT neurites was completely abolished by treatment with chondroitinase ABC, indicating a specific contribution of CS glycosaminoglycan in this regionally specific behaviour. This differential behaviour was not observed in explants presented with a lower or higher concentration of CS or in explants grown on substrate coated with a different laminin concentration. Thus, a critical ratio of CS to laminin seems to be essential to induce this differential behaviour in retinal neurites towards contact with CS. Furthermore, this behavior was not observed in explants cultured directly on a CS-rich substrate, suggesting that contact with growth-promoting molecules is necessary for the selective responses of retinal neurites during subsequent contact with CS. We concluded that CS glycosaminoglycan is sufficient to drive selective inhibition of VT but not DN neurites and that, together with a critical combination of growth-promoting factors, it may control the axon divergence process at the mouse optic chiasm.This project is partially supported by a grant from the Research Grant Council of the Hong Kong Special Administrative Region (project no. CUHK 4417/03M) and a Direct Grant from the Chinese University of HK (project no. 2004.1.051).     

Segmental independence and age dependence of neurite outgrowth from embryonic chick sensory neurons

Targets in limb regions of the chick embryo are further removed from the dorsal root ganglia that innervate them compared with thoracic ganglion-to-target distances. It has been inferred that axons grow into the limb regions two to three times faster than into nonlimb regions. We tested whether the differences were due to intrinsic properties of the neurons located at different segmental levels. Dorsal root ganglia (DRG) were isolated from the forelimb, trunk, and hind limb regions of stage 25–30 embryos. Neurite outgrowth was measured in dissociated cell culture and in cultures of DRG explants. Although there was considerable variability in the amount of neurite outgrowth, there were no substantive differences in the amount or the rate of outgrowth comparing brachial, thoracic, or lumbosacral neurons. The amount of neurite outgrowth in dissociated cell cultures increased with the stage of development. Overall, our data suggest that DRG neurons express a basal amount of outgrowth, which is initially independent of target-derived neurotrophic influences; the magnitude of this intrinsic growth potential increases with stage of development; and the neurons of the DRG are not intrinsically specified to grow neurites at rates that are matched to the distance they are required to grow to make contact with their peripheral targets in vivo. We present a speculative model based on Poisson statistics, which attempts to account for the variability in the amount of neurite outgrowth from dissociated neurons. © 1995 John Wiley & Sons, Inc.     

Preferential outgrowth of central nervous system neurites on astrocytes and Schwann cells as compared with nonglial cells in vitro

I have compared central nervous system (CNS) neurite outgrowth on glial and nonglial cells. Monolayers of glial cells (astrocytes and Schwann cells) or nonglial cells (e.g., fibroblasts) were prepared and were shown to be greater than 95% pure as judged by cell type-specific markers. These monolayers were then tested for their ability to support neurite outgrowth from various CNS explants. While CNS neurites grew vigorously on the glial cells, most showed little growth on nonglial cell monolayers. Neurites grew singly or in fine fascicles on the glial cells at rates greater than 0.5 mm/d. The neurite outgrowth on astrocytes was investigated in detail. Scanning and transmission electron microscopy showed that the neurites were closely apposed to the astrocyte surface and that the growth cones were well spread with long filopodia. There was no evidence of significant numbers of explant- derived cells migrating onto the monolayers. Two types of experiments indicated that factors associated with the astrocyte surface were primarily responsible for the vigorous neurite outgrowth seen on these cells: (a) Conditioned media from either astrocytes or fibroblasts had no effect on the pattern of outgrowth on fibroblasts and astrocytes, and conditioned media factors from either cell type did not promote neurite outgrowth when bound to polylysine-coated dishes. (b) When growing CNS neurites encountered a boundary between astrocytes and fibroblasts, they stayed on the astrocytes and did not encroach onto the fibroblasts. These experiments strongly suggest that molecules specific to the surfaces of astrocytes make these cells particularly attractive substrates for CNS neurite outgrowth, and they raise the possibility that similar molecules on embryonic glial cells may play a role in guiding axonal growth during normal CNS development.     

The effect of cyclic AMP on neuritic outgrowth in explant cultures of developing chick olfactory epithelium

The effect of cyclic AMP (cAMP) analogs and phosphodiesterase (PDE) inhibitors on neurite outgrowth was studied in explant cultures of olfactory neurons. Nasal pits from 5- or 6-day-old chick embryos were minced, explanted into culture dishes, and grown in a serum-free medium. One of the cyclic AMP analogs, dibutyryl cyclic AMP (dbcAMP) or 8-bromo-cyclic AMP (8-Br-cAMP), or one of the PDE inhibitors, theophylline or isobutylmethylxanthine (IBMX), was added to the culture medium. The explants were examined for neurite outgrowth after 2 days in vitro. Db-cAMP increased the number of explants expressing neurites by 25-35% over control cultures, whereas 8-Br-cAMP had essentially no effect at the same concentrations. Addition of dibutyryl cyclic GMP (dbcGMP) gave no increase in neurite outgrowth, thus indicating that the effect of enhancing neuritic growth is specific to cAMP and not cyclic nucleotides in general. The resulting increase in neurite outgrowth is due to the cyclic nucleotide component of dbcAMP, since both IBMX and theophylline, which elevate intracellular cAMP, also increased neurite outgrowth significantly. When forskolin was added to the culture medium, there was a trend to increased neurite outgrowth; this was significantly enhanced when a subthreshold concentration of theophylline was added in addition to the forskolin.     

Tenascin is accumulated along developing peripheral nerves and allows neurite outgrowth in vitro.

The extracellular matrix protein, tenascin, appears in a restricted pattern during organ morphogenesis. Here we studied the expression of tenascin along developing peripheral nerves in chick embryos and tested its activity as a substrate for cultured neurons. Motor axons grow out through the tenascin-rich, anterior part of the sclerotome. Shortly after, tenascin surrounds axon fascicles of ventral roots. At the limb levels, outgrowing axons accumulate in the tenascin-containing girdle region forming a plexus. In the limb, tenascin first appears in bracket-like structures that surround the precartilage cell condensations of the femur and humerus, respectively. These regions coincide with the channels along which axons first grow in from the girdle plexus to form the limb nerves. Later, the major tenascin staining is associated with the cartilage and tendon primordia, and not with the limb nerves. We used tenascin as a substrate for cultured neural explants and single cells in order to test for its function in neurite outgrowth. Dissociated embryonic neurons of various types attached to mixed polylysine/tenascin substrates and sprouted rapidly after a lag of several hours. Outgrowth was inhibited and neurites were detached by anti-tenascin antibodies. On substrates coated with tenascin alone, neurite outgrowth was achieved from 3 day spinal cord explants. Whereas growth cones were well spread and rapidly moving, the neurites were poorly attached, straight and rarely branched. We speculate that in vivo tenascin allows axonal outgrowth, but inhibits branching and supports fasciculation of newly formed axons.     

Antiserum Against Neurite Outgrowth Factor in Chick Gizzard Extract and Its Inhibitory Effect on Neuritic Response in Cultured Ciliary Neurons

Abstract: Antiserum against a neurite outgrowth factor (NOF) of gizzard extract that promotes neurite outgrowth from dissociated ciliary ganglionic neurons (CG neurons) of 8-day-old chick embryo was prepared to determine whether or not the antiserum inhibits neurite outgrowth from cultured neurons or explants of chick and murine tissues. When CG neurons were cultured on a polyornithine-coated well exposed to NOF (NOF-bound POR well), marked neurite outgrowth was observed. When NOF-bound POR wells were exposed to antiserum, neurite outgrowth from CG neurons was gradually inhibited with increasing amounts of antiserum, while exposure to preimmune serum did not prevent neurite outgrowth. Antiserum had no effect on neuronal survival during a 48-h incubation. The diluted antiserum, which produced nearly 100% inhibition of the NOF activity, was almost equally active in suppressing the activity of NOFs in conditioned media (CM) of various chick embryo tissues, but showed much less inhibitory effects on NOFs in CM of murine tissues. The appearance of neurites from explants of spinal cord, dorsal root ganglion, or retina of chick embryo was also inhibited by the antiserum. These results indicate that antiserum against NOF from gizzard extract suppressed the activity of NOFs from various sources, and that there are species differences in NOFs, at least between chick and murine.     

Oriented axon outgrowth from avian embryonic retinae in culture

The neural retina of avian embryos was spread on a membrane filter and cut in any desired orientation. Strips cut across the retina of 4- to 7-day chick or 3- to 6-day quail embryos were explanted onto collagen gels. Vigorous neurite outgrowth was seen for about 3 days, by which time many neurites were 3 mm long. Horseradish peroxidase (HRP) labeling showed that the cells producing the neurites were large and formed a layer near the inner limiting membrane, indicating that the neurites in vitro were axons of retinal ganglion cells. The size of the neurite population and the regions from which neurites emerged vaired with the donor age, while most neurites sprouted from the side of the explant formerly closest to the optic fissure. This pattern closely resembled that of axon growth in the normal retina, as revealed by SEM, silver staining, and HRP labeling. Mitotic inhibitors (Ara-C and FUdR) did not alter the neurite outgrowth. Pretreatment of retinae with trypsin or collagenase did not disorganize axons at the time of explantation, but tended to equalize neurite emergence on each side of the retinal strips. We suggest that microenvironmental factors, especially the enzyme-labile inner limiting membrane, are important for axon guidance in the retina.     

Influence of ganglion age,nonneuronal cells and substratum on neurite outgrowth in culture

This study characterizes the outgrowth patterns of superior cervical ganglia (SCG) obtained from embryonic (E15), perinatal (E20–21), and adult (P35) rats when placed in culture on various substrata. Outgrowth morphology, degree of fasciculation, and outgrowth length were examined on collagen (COL), polyornithine (PO), polylysine (PL), fibronectin (FN), and nonneuronal cells (NNCs) from the ganglion. COL and FN supported extensive neuritic outgrowth; PO and PL provided poor support. Outgrowth pattern, degree of fasciculation, neurite growth rate, and the number of NNCs in the outgrowth varied considerably depending upon the COL configuration. When undiluted COL (～5 mg/ml) was air dried, a three-dimensional loose fibrillar network was formed. Upon COL dilution or gelling undiluted COL by ammoniation, an essentially two-dimensional layer was formed. On two-dimensional COL, NNCs were able to proliferate and migrate extensively from ganglia of all ages; their presence influenced the form and extent of neurite growth. E15, E20, and P35 neurites responded differently to their endogenous NNCs. E15 neurites extended in relation to NNC surfaces and were predominantly nonfasciculated. E20 neurites became more fasciculated in the presence of NNCs that exhibited morphological and behavioral differences from those migrating from E15 ganglia. E20 neurite bundles became defasciculated when they extended into E15 outgrowth. Far fewer neurites grew from P35 explants in the presence of their NNCs. Three-dimensional COL greatly slowed NNC migration and thus allowed investigation of neurite outgrowth from ganglia of differing age in the absence of NNCs. We conclude that neuritic outgrowth patterns on varying substrata reflect not only neurite differences depending upon ganglion age but also variation in the behavior of accompanying NNCs.     

Embryonic and regenerating Xenopus retinal fibers are intrinsically different

Growth and guidance behavior of Xenopus embryonic (ER) (optic vesicle stage 25/26) and regenerating retinal fibers (stage 47/50 newly regenerating NR, and actively regenerating RR, respectively) have been studied in vitro on a variety of substrates in serum-free media. RR retinas receive a prior conditioning lesion 12-14 days before explantation while NR retinas are explanted immediately after axotomy. The substrates include plastic (UN), polylysine (PL), polyornithine (PO), laminin (LM), fibronectin (FN), and collagen type I (CO). Two kinds of experimental situations were tested, one in which substrates were derivatized to plastic as a planar surface, while the second involved the addition of a substrate as a soluble supplement to dishes derivatized with PL. A neurite growth index (NGI), based on density of neurite outgrowth and axon lengths, is determined for each fiber type on all substrates. Embryonic and regenerating fibers are phenotypically different fiber types; each displays a specific "substrate preference profile" (SPP), reflecting differential growth on each substrate. ER neurites grow equally well on all planar substrates, including plastic, but do not grow on CO (SPP, LM = FN = PL = PO = UN greater than CO). Both NR and RR neurites show distinct substrate preferences, but RR neurites grow more vigorously (SPP, LM greater than CO greater than PL = PO greater than FN). In media supplemented with LM, FN or CO, the SPPs showed little change but the neurite bundle patterns were qualitatively different. Only regenerating neurites display clockwise growth in laminin (LM) and fibronectin (FN)-supplemented media. Under no conditions do embryonic fibers exhibit this pattern which suggests that embryonic and regenerating retinal fibers also differ in cytoskeletal organization. Evidence of intrinsic growth differences in vitro suggest that embryonic and regenerating retinal fibers may not respond to identical guidance cues during in vivo development and regeneration of retinotectal connections.     

PTPmu regulates N-cadherin-dependent neurite outgrowth

Cell adhesion is critical to the establishment of proper connections in the nervous system. Some receptor-type protein tyrosine phosphatases (RPTPs) have adhesion molecule-like extracellular segments with intracellular tyrosine phosphatase domains that may transduce signals in response to adhesion. PTPmu is a RPTP that mediates cell aggregation and is expressed at high levels in the nervous system. In this study, we demonstrate that PTPmu promotes neurite outgrowth of retinal ganglion cells when used as a culture substrate. In addition, PTPmu was found in a complex with N-cadherin in retinal cells. To determine the physiological significance of the association between PTPmu and N-cadherin, the expression level and enzymatic activity of PTPmu were perturbed in retinal explant cultures. Downregulation of PTPmu expression through antisense techniques resulted in a significant decrease in neurite outgrowth on an N-cadherin substrate, whereas there was no effect on laminin or L1-dependent neurite outgrowth. The overexpression of a catalytically inactive form of PTPmu significantly decreased neurite outgrowth on N-cadherin. These data indicate that PTPmu specifically regulates signals required for neurites to extend on an N-cadherin substrate, implicating reversible tyrosine phosphorylation in the control of N-cadherin function. Together, these results suggest that PTPmu plays a dual role in the regulation of neurite outgrowth.