Fasciclin 2

EVOLUTIONARY HOMOLOGS (part 3/3)

CAMs and axon guidance

A mouse complementary DNAs has been cloned encoding a novel protein, Rb-8 neural cell adhesion molecule (RNCAM), with a predicted extracellular region of five immunoglobulin C2-type domains followed by two fibronectin type III domains. Alternative splicing is likely to generate two RNCAM isoforms, which are differently attached to the cell membrane. These structural features and overall sequence identity identify this protein as a novel member of a cell adhesion molecule subgroup that includes vertebrate neural cell adhesion molecule, Aplysia cell adhesion molecule, and Drosophila fasciclin II. In insects, fasciclin II is present on a restricted subset of embryonic central nervous system axons where it controls selective axon fasciculation. Intriguingly, RNCAM likewise is expressed in subsets of olfactory and vomeronasal neurons with topographically defined axonal projections. The spatial expression of RNCAM corresponds precisely to that of certain odorant receptor expression zones of the olfactory epithelium. These expression patterns thus render RNCAM the first described cell adhesion molecule with a potential regulatory role in formation of selective axonal projections important for olfactory sensory information coding (Alenius, 1997).

Axonal interactions, which are mediated by cell adhesion molecules (CAMs) as well as other types of membrane proteins, are important for sensory axon pathfinding in the developing chick hindlimb. Injection of antibodies that block the function of either L1 or N-cadherin into the limb, starting when the first sensory axons reach the plexus, alters the segmental pattern of projections along cutaneous nerves. Specific removal of polysialic acid from NCAM using the enzyme endoneuraminidase N (Endo N) also results in significant changes in cutaneous projection patterns, while injection of antibodies against NCAM itself have no obvious effect. To help understand the cellular basis for these findings, a tissue culture system was developed in which the axons from dorsal root ganglion explants grow within defined laminin lanes and an examination was carried out to see whether the same treatments increase or decrease a growth cone's tendency to be closely associated with neighboring axons. After 2 days in culture, images of the cultures were recorded, antibodies or Endo N was added, and images of the same fields were recaptured an hour later. To quantify the results, growth cones located in defined regions of the laminin lanes were classified, before and after the perturbation, as 'free' (i.e., growing primarily on the laminin substratum), 'fasciculated' (i.e., growing tightly along other neurites), or 'intermediate' (i.e., growing both on the laminin substratum and in contact with other neurites). Anti-L1 and anti-N-cadherin, but not anti-NCAM, cause an increase in defasciculated growth cones, whereas Endo N results in an increase in fasciculated growth cones. These changes in fasciculation are consistent with the changes in cutaneous projections seen in ovo perturbations. The results from these tissue culture experiments thus provide strong support for the idea that one mechanism by which CAMs affect sensory axon pathfinding in vivo is by regulating the affinity of sensory growth cones for neighboring axons, which in turn can modulate the growth cone's ability to navigate through the surrounding environment (Honig, 1998).

The neural cell adhesion molecule L1, which is present on axons and growth cones, plays a crucial role in the formation of major axonal tracts, such as the corticospinal tract and corpus callosum. L1 is preferentially transported to axons and inserted in the growth cone membrane. However, the process by which L1 is sorted to axons remains unclear. Tyr1176 in the L1 cytoplasmic domain is adjacent to a neuron-specific alternatively spliced sequence, RSLE (Arg-Ser-Leu-Glu). The resulting sequence of YRSLE conforms to a tyrosine-based consensus motif (YxxL) for sorting of integral membrane proteins into specific cellular compartments. To study a possible role of the YRSLE sequence in L1 sorting, chick DRG neurons were transfected with either human L1 cDNA that codes for full-length L1 (L1FL), a non-neuronal form of L1 that lacks the RSLE sequence (L1DeltaRSLE), mutant L1 with a Y1176A substitution (L1Y1176A), or L1 truncated immediately after the RSLE sequence (L1DeltaC77). L1FL and L1DeltaC77, both of which possess the YRSLE sequence, are expressed in the axonal growth cone and to a lesser degree in the cell body. In contrast, expression of both L1DeltaRSLE and L1Y1176A is restricted to the cell body and proximal axonal shaft. L1DeltaRSLE and L1Y1176A are integrated into the plasma membrane in the cell body after missorting. These data demonstrate that the neuronal form of L1 carries the tyrosine-based sorting signal YRSLE, which is critical for sorting L1 to the axonal growth cone (Kamiguchi, 1998a).

Neural cell adhesion molecules of the immunoglobulin superfamily (IgCAMs) have been implicated in both the fasciculation and guidance of axons, but direct genetic evidence of a role for neural IgCAMs in vertebrate axon guidance is lacking. The L1 subfamily of vertebrate neural IgCAMs function as both homophilic and heterophilic receptors for a variety of cell-surface and extracellular ligands, and may signal through intracellular kinases or by recruitment of the fibroblast growth factor receptor. L1 itself has been implicated in many neural processes and is expressed widely in the embryonic and adult nervous systems. In humans, mutations in the L1 gene are linked with a spectrum of brain disorders, including loss of the corticospinal tract, but the mechanistic basis for these disorders is unknown. Mice that do not express L1 have defects in the guidance of axons of the corticospinal tract, a major motor control pathway projecting from the cortex to the spinal cord. Although the pathway to the caudal medulla appears normal, a substantial proportion of axons fail to cross the midline to the opposite dorsal column, as it would be normal to do. In adults, this results in a reduced decussation and in large numbers of axons projecting ipsilaterally. There is also a varying, but reduced, number of corticospinal axons in the dorsal columns of the spinal cord. These do not project beyond cervical levels. These are defects in axon guidance, because they arise during the early stages of the development of the decussation. The presence of a ligand for L1, known as CD24, specifically at the point of decussation, suggests a mechanism in which L1 functions to guide corticospinal axons across the midline. It is concluded that L1 function is necessary for the guidance of corticospinal axons across the pyramidal decussation in mice. Some of the defects in the corticospinal tract of humans with mutations in L1 could be due to errors in axon guidance at the pyramidal decussation (Cohen, 1998).

A scaffold of axons consisting of a pair of longitudinal tracts and several commissures is established during early development of the vertebrate brain. NOC-2, a cell surface carbohydrate, is selectively expressed by a subpopulation of growing axons in this scaffold in Xenopus. NOC-2 is present on two glycoproteins, one of which is a novel glycoform of the neural cell adhesion molecule N-CAM. When the function of NOC-2 is perturbed using either soluble carbohydrates or anti-NOC-2 antibodies, axons expressing NOC-2 exhibit aberrant growth at specific points in their pathway. NOC-2 is the first-identified axon guidance molecule essential for development of the axon scaffold in the embryonic vertebrate brain (Anderson, 1999).

In humans, defects of the corticospinal tract have been attributed to mutations in the gene encoding L1 CAM, a phenotype that is reproduced in L1-deficient mice. Using coculture assays, Sema3A secreted from the ventral spinal cord is reported to repel cortical axons from wild-type but not from L1-deficient mice. L1 and neuropilin-1 (NP-1) form a stable complex, and their extracellular domains can directly associate. Thus, L1 is a component of the Sema3A receptor complex, and L1 mutations may disrupt Sema3A signaling in the growth cone, leading to guidance errors. Addition of soluble L1Fc chimeric molecules does not restore Sema3A responsiveness of L1-deficient axons; instead, it converts the repulsion of wild-type axons into an attraction, further supporting a function for L1 in the Sema3A transducing pathways within the growth cone (Castellani, 2000).

To date, L1 activity in the guidance of neuronal projections has been associated with axonal fasciculation and neurite extension. Accordingly, L1 interacts with a number of cell adhesion and extracellular matrix molecules, including members of the Ig superfamily, integrins, and chondroitin sulfate proteoglycans. The finding that L1 is required for Sema3A signaling expands its range of actions and assigns it as a player in mechanisms of guidance at a distance. This raises the possibility that other fiber tracts depending upon Sema3A for their guidance might also be impaired in L1-deficient mice (Castellani, 2000).

NP-1 and NP-2 are the first proteins identified as receptors for the semaphorin family of guidance cues. Only NP-1 is necessary for Sema3A, and probably Sema3E, whereas binding sites for Sema3B and Sema3C are formed by either NP-2 homodimers or NP-1-NP-2 heterodimers. Both the coculture and collapse assays reported here demonstrate that L1 is required for the growth cone to respond to Sema3A-induced chemorepulsion but is not necessary for Sema3B, -3C, or -3E biological activity. Thus, the functional coupling of L1 to Sema3A may confer an additional level of specificity to the functions of semaphorins in the guidance of neuronal projections (Castellani, 2000).

The coimmunoprecipitation and binding experiments have showen that L1 and NP-1, but not NP-2, extracellular domains are able to interact and that L1 participates in the formation of a Sema3A receptor multimolecular complex. There are precedents for the participation of NP-1 in receptor heterocomplexes. For example, NP-1 binds to vascular endothelial growth factor receptor (VEGFR), thereby enhancing the affinity of the VEGF ligand for its receptor. Furthermore, NP-1 forms a stable complex with the transmembrane protein plexin-A1. Within this complex, plexin-A1 serves as a signal tranducer for Sema3A-induced growth cone collapse of sensory neurons. The coculture and collapse assays indicate that L1 is also necessary for neurons to respond to Sema3A. Application of L1Fc could not restore the response of L1-deficient axons to Sema3A, indicating that the intracellular domain of L1 is essential for this response. Furthermore, L1Fc does not simply block Sema3A chemorepulsion of wild-type neurons but switches it into attraction. It is therefore likely that L1-L1 homophilic interactions contribute to the outcome of Sema3A intracellular signaling in the growth cone (Castellani, 2000).

L1 could achieve this function via different interactions. (1) Its intracellular domain associates with tyrosine kinases and two serine/threonine kinases, including casein kinase II, which has been implicated in neurite fasciculation. (2) L1 stimulation is assumed to activate second-messenger systems that are common to the pathways used by fibroblast growth factor receptor (FGFR) tyrosine kinases. (3) The intracytoplasmic region of L1 interacts with the ankyrin family of spectrin binding proteins. This coupling may provide a molecular linkage between Sema3A and actin cytoskeletal dynamics within the growth cone. Alternatively, L1 could regulate the Sema3A signaling by influencing the clustering of NP-1 within the heterocomplex. As first shown for ephrins, another family of repulsive signals, and their receptors, the Eph tyrosine kinases, the oligomeric status of receptors and ligands determines their biological activity. Recent work has expanded this idea to the semaphorin family with the demonstration that monomeric, dimeric, and oligomeric forms of Sema1A confer distinct functional properties during axonal guidance in Drosophila embryos (Castellani, 2000 and references therein).

The molecular mechanisms by which L1Fc switches the Sema3A-induced response are not clear. However, two results suggest that it is achieved through activation of L1 signaling in the growth cone: (1) L1Fc alone induces a strong increase in neurite outgrowth, a process assumed to result from L1-L1 homophilic interaction; (2) the response of L1-deficient axons to Sema3A is not modulated by L1Fc, indicating that even if L1Fc can interact with NP1, transmembrane L1 in the receptor heterocomplex is crucial for the growth cone to undergo Sema3A-induced chemoattraction. L1Fc may modify the composition of the Sema3A receptor, for example, by diverting L1 into another functionally distinct transduction unit. Such changes in the interactions contracted by the different components in the complex could initiate an attractive response to Sema3A. Likewise, recent work has demonstrated that the cytoplasmic domain of axon guidance receptors and the formation of heterocomplexes are crucial in determining the type of response of the growth cone to chemotropic cues (Castellani, 2000).

It has also been shown that cyclic nucleotide levels are pivotal for the decision of the growth cone to be attracted or repelled by secreted guidance signals. Sema3A chemorepulsion can be converted into attraction by an increase in cGMP levels in the growth cone, driven by nitric oxide-dependent intracellular pathways. Accordingly, it is possible that L1Fc induces the switch by activating the synthesis of cGMP. Consistent with this, preliminary data suggest that blockade of soluble guanylate cyclase prevents the L1Fc-induced switch in the Sema3A response (Castellani, 2000).

In the present experiments, L1Fc may mimic L1-L1 trans-interaction of the growth cone with another cell or axonal surface. This implies that during the development of neuronal projections, changes in L1-dependent interactions may be able to switch growth cone responses to Sema3A from repulsion to attraction. Finally, the present study raises the more general view that specific cross-talk between cell contact and chemotropic guidance cues could be a potent mechanism to coordinate the events required for guiding axon projections toward their appropriate targets (Castellani, 2000).

Motility of the nerve growth cone is highly dependent on its dynamic interactions with the microenvironment mediated by cell adhesion molecules (CAMs). These adhesive interactions can be spatially regulated by changing the density and avidity of CAMs on the growth cone. L1, a member of the immunoglobulin superfamily of CAMs, is endocytosed at the central domain of the growth cone followed by centrifugal vesicular transport and reinsertion into the plasma membrane of the leading edge. This paper focuses on the functional significance of endocytic L1 trafficking in dorsal root ganglia neurons in vitro. The rate of L1-based neurite growth has a positive correlation with the amount of endocytosed L1 in the growth cone, whereas stimulation of neurite growth via an N-cadherin-dependent mechanism does not increase L1 endocytosis. A growth cone that migrates on an L1 substrate exhibits a steep gradient of L1-mediated adhesion (strong adhesion at the growth cone's leading edge and weak adhesion at the central domain). This gradient of L1 adhesion is attenuated after inhibition of L1 endocytosis in the growth cone by intracellular loading of a function-blocking antibody against alpha-adaptin, a subunit of the clathrin-associated AP-2 adaptor. Inhibition of L1 endocytosis by this antibody also decreases the rate of L1-dependent growth cone migration. These results indicate that the growth cone actively translocates CAMs to create spatial asymmetry in adhesive interactions with its environment and that this spatial asymmetry is important for growth cone migration (Kamiguchi, 2002).

Neuronal polarity is, at least in part, mediated by the differential sorting of membrane proteins to distinct domains, such as axons and somata/dendrites. This study investigated the pathways underlying the subcellular targeting of NgCAM, a cell adhesion molecule residing on the axonal plasma membrane. Following transport of NgCAM kinetically, surprisingly a transient appearance of NgCAM was observed on the somatodendritic plasma membrane. Down-regulation of endocytosis resulted in loss of axonal accumulation of NgCAM, indicating that the axonal localization of NgCAM was dependent on endocytosis. The data suggest the existence of a dendrite-to-axon transcytotic pathway to achieve axonal accumulation. NgCAM mutants with a point mutation in a crucial cytoplasmic tail motif (YRSL) are unable to access the transcytotic route. Instead, they were found to travel to the axon on a direct route. Therefore, the results suggest that multiple distinct pathways operate in hippocampal neurons to achieve axonal accumulation of membrane proteins (Wisco, 2003).

In the developing chick hindlimb, sensory axons, which grow together in bundles as they extend distally, and the motoneuron axons they encounter, express the cell adhesion molecule L1. Following injection of function-blocking anti-L1 antibodies into the limb at stage 25, some sensory axons choose inappropriate peripheral nerves even though motoneuron pathfinding is unaffected. To further elucidate L1's role, the effects of this perturbation were assessed using pathway tracing, immune labeling, confocal microscopy, and electron microscopy. After L1 blockade, sensory axons were still bundled and closely apposed. However, clear signs of decreased adhesion were detectable ultrastructurally. Further, sensory axons grow into the limb more slowly than normal, wandering more widely, branching more frequently, and sometimes extending along inappropriate peripheral nerves. Sensory axons that ultimately projected along different cutaneous nerves show increased intermixing in the spinal nerves, due to errors in pathfinding and also to a decreased ability to segregate into nerve-specific fascicles. These results suggest that, in the highly complex in vivo environment, as in tissue culture, L1 stimulates axon growth and enhances fasciculation, and that these processes contribute to the orderly, timely, and specific growth of sensory axons into the limb (Honig, 2002).

Dynamic microtubules explore the peripheral (P) growth cone domain using F actin bundles as polymerization guides. Microtubule dynamics are necessary for growth cone guidance; however, mechanisms of microtubule reorganization during growth cone turning are not well understood. These issues are addressed by analyzing growth cone steering events in vitro, evoked by beads derivatized with the Ig superfamily cell adhesion protein apCAM. Pharmacological inhibition of microtubule assembly with low doses of taxol or vinblastine result in rapid clearance of microtubules from the P domain with little effect on central (C) axonal microtubules or actin-based motility. Early during target interactions, F actin assembly and activated Src, but few microtubules, are detected at apCAM bead binding sites. The majority of microtubules extend toward bead targets after F actin flow attenuation occurs. Microtubule extension during growth cone steering responses is strongly suppressed by dampening microtubule dynamics with low doses of taxol or vinblastine. These treatments also inhibit growth cone turning responses, as well as focal actin assembly and accumulation of active Src at bead binding sites. These results suggest that dynamic microtubules carry signals involved in regulating Src-dependent apCAM adhesion complexes involved in growth cone steering (Suter, 2004).

The role of the cell adhesion molecule NrCAM for axonal growth and pathfinding was investigated in the developing retina. Analysis of the distribution pattern of NrCAM in chick embryo retina sections and flat-mounts shows its presence during extension of retinal ganglion cell (RGC) axons; NrCAM is selectively present on RGC axons and is absent from the soma. Single cell cultures show an enrichment of NrCAM in the distal axon and growth cone. When offered as a substrate in addition to Laminin, NrCAM promotes RGC axon extension and the formation of growth cone protrusions. In substrate stripe assays, mimicking the NrCAM-displaying optic fibre layer and the Laminin-rich basal lamina, RGC axons preferentially grow on NrCAM lanes. The three-dimensional analysis of RGC growth cones in retina flat-mounts reveals that they are enlarged and form more protrusions extending away from the correct pathway under conditions of NrCAM-inhibition. Time-lapse analyses show that these growth cones pause longer to explore their environment, proceed for shorter time spans, and retract more often than under control conditions; in addition, they often deviate from the correct pathway towards the optic fissure. Inhibition of NrCAM in organ-cultured intact eyes causes RGC axons to misroute at the optic fissure; instead of diving into the optic nerve head, these axons cross onto the opposite side of the retina. These results demonstrate a crucial role for NrCAM in the navigation of RGC axons in the developing retina towards the optic fissure, and also for pathfinding into the optic nerve (Zelina, 2005).

Aplysia apCAM, cell adhesion and synaptic plasticity

Dynamic cytoskeletal rearrangements are involved in neuronal growth cone motility and guidance. To investigate how cell surface receptors translate guidance cue recognition into these cytoskeletal changes, a novel in vitro assay has been developed where beads, coated with antibodies to the immunoglobulin superfamily cell adhesion molecule apCAM or with purified native apCAM, replace cellular substrates. These beads associate with retrograde F-actin flow, but in contrast to previous studies, are then physically restrained with a microneedle to simulate interactions with noncompliant cellular substrates. After a latency period of approximately 10 min, an abrupt increase in bead-restraining tension is observed, accompanied by direct extension of the microtubule-rich central domain toward sites of apCAM bead binding. Most important, retrograde F-actin flow is attenuated only after restraining tension has increased and only in the bead interaction axis where preferential microtubule extension occurs. These cytoskeletal and structural changes are very similar to those reported for growth cone interactions with physiological targets. Immunolocalization using an antibody against the cytoplasmic domain of apCAM reveals accumulation of the transmembrane isoform of apCAM around bead-binding sites. These results provide direct evidence for a mechanical continuum from apCAM bead substrates through the peripheral domain to the central cytoplasmic domain. By modulating functional linkage to the underlying actin cytoskeleton, cell surface receptors such as apCAM appear to enable the application of tensioning forces to extracellular substrates, providing a mechanism for transducing retrograde flow into guided growth cone movement (Suter, 1998).

Long-term facilitation of the connections between the sensory and motor neurons of the gill-withdrawal reflex in Aplysia requires five repeated pulses of serotonin (5-HT). The repeated pulses of 5-HT initiate a cascade of gene activation that leads ultimately to the growth of new synaptic connections. Several genes in this process have been identified, including the transcriptional regulators apCREB-1, apCREB-2, apC/EBP, and the cell adhesion molecule apCAM, which is thought to be involved in the formation of new synaptic connections. The transcriptional regulators apCREB-2 and apC/EBP, as well as a peptide derived from the cytoplasmic domain of apCAM, are phosphorylated in vitro by Aplysia mitogen-activated protein kinase (apMAPK). The cDNA encoding apMAPK (see Drosophila MAPK) has been cloned and apMAPK activity is increased in sensory neurons treated with repeated pulses of 5-HT and by the cAMP pathway. These results suggest that apMAPK may participate with cAMP-dependent protein kinase during long-term facilitation in sensory cells by modifying some of the key elements involved in the consolidation of short- to long-lasting changes in synaptic strength. How might the PKA and MAPK pathways converge in Aplysia sensory neurons? In the simplest case, the MAPK pathway would be downstream from the PKA pathway. In many cells, PKA negatively regulates the MAPK pathway by phosphorylating Raf-1. However, in B-Raf-containing cells, PKA activates the MAPK pathway by signaling through Rap1, a member of the Ras family of small G proteins. PKA may thus activate MAPK by phosphorylating the Aplysia homolog of Rap1, therby activating B-Raf, MEK, and MAPK (Martin, 1998).

Cell adhesion molecules play important roles in axon guidance and synapse formation. Recent studies suggest that the expression of some of these molecules can be regulated either by electrical activity or by specific neurotransmitters. The expression of neural cell adhesion molecule (NCAM)-like molecules in Aplysia, designated apCAM, is downregulated from the surface of sensory neurons by 5-HT, a transmitter known to evoke long-term changes in the structure and function of these neurons. Whether the distribution of apCAM on the surface of other neurons can be regulated by treatments with other neurotransmitters known to evoke long-term functional and structural changes in Aplysia neurons was tested, as well as the consequences of treatments with the neurotransmitters on the pattern of growth cone-neurite interactions. Applications of the neuropeptide Phe-Met-Arg-Phe-amide (FMRFamide) that evoke long-term synaptic depression also reduce apCAM expression on the surface of motor cell L7 via a mechanism that appears to be similar to the mechanism mediating the 5-HT-induced change in the sensory cells. Specific treatments that affect apCAM distribution on the surface of their respective cells (5-HT on sensory cells and FMRFamide on motor cell L7) mimic treatment with monoclonal antibodies against apCAM by evoking a significant reduction in the fasciculation of growth cones with other neurites extending from homologous cells (Peter, 1994).

FMRFamide (see Drosophila FMRFamide) evokes long-term inhibition of the sensorimotor connection of Aplysia, including structural alterations in the presynaptic sensory cell. FMRFamide also evokes a down-regulation of the adhesion molecule apCAM from the surface of the postsynaptic motor cell L7. The second messenger pathways mediating the long-term actions of FMRFamide on both the pre- and postsynaptic cells were examined to determine whether the activation of each pathway is required for the expression of long-term functional and structural plasticity. Inhibition of the lipoxygenase pathway of arachidonic acid metabolism, but not the cyclooxygenase pathway, blocks the long-term changes in the presynaptic sensory cell evoked by FMRFamide. The down-regulation of apCAM in L7 appears to be mediated by cAMP-dependent activation of protein kinase A. Blocking the cAMP-dependent changes also blocks FMRFamide-induced long-term functional and structural changes. These results suggest that the expression of long-term heterosynaptic inhibition in Aplysia may require concomitant presynaptic and postsynaptic changes, each transduced by specific second messenger systems (Wu, 1994).

The synaptic growth that accompanies 5-HT-induced long-term facilitation of the sensory to motor neuron connection in Aplysia is associated with the internalization of apCAM at the surface membrane of the sensory neuron. Epitope tags were used to examine the fate of each of the two apCAM isoforms (membrane bound and GPI-linked). Only the transmembrane form is internalized. This internalization can be blocked by overexpression of transmembrane constructs with a single point mutation in the two MAPK consensus sites, as well as by injection of a specific MAPK antagonist into sensory neurons. These data suggest MAPK phosphorylation at the membrane is important for the internalization of apCAMs and, thus, may represent an early regulatory step in the growth of new synaptic connections that accompanies long-term facilitation (Bailey, 1997).

Long-term facilitation by serotonin (5-HT) of Aplysia sensorimotor synapses in culture is accompanied by two changes: an increase in the number of sensory cell branches and varicosities contacting the major axons of the target motor cell L7, and a downregulation of Aplysia cell adhesion molecules (apCAM) from the surface of the presynaptic sensory cell. A hypothesis that the two changes may be linked suggests that the 5-HT-induced decrease of apCAM levels from sensory neurites may defasciculate sensory neurites from one another and make the surface of the motor axons a more attractive substrate for new growth and synapses. Developing cultures were used to examine the relationship of neuritic branching, varicosity formation, and efficacy of the connections formed by sensory cells to levels of apCAM expression on the motor cell. A determination was made of the consequences of apCAM expression for the pattern of sensory cell growth and synapse formation of 5-HT applied during the early period of interaction between sensory and motor cells (day 1 or 2 in culture). The number of sensory cell branches and varicosities, and the ability of sensory growth cones to fasciculate with L7 axons and form chemical connections correlate with the level of apCAM expression on different regions of L7. Early exposure to 5-HT increases the number of sensory cell branches and varicosities contacting newly regenerated distal neurites of L7 to levels that would normally occur when the sensory neurites interact with the major proximal axons of L7. Treatment with 5-HT also modulates the efficacy of the developing synaptic connections. The change in synapse efficacy is accompanied by an increase in the formation of new sensory varicosities and branches with pioneering growth cones extending on the major axons of L7. The results are consistent with the hypothesis that treatment with 5-HT modulates local differences in the expression of cell adhesion molecules on the surface of the interacting cells, making motor neurites more attractive for sensory growth cones, thereby affecting new sensory neuritic growth and synapse formation (Zhu, 1994).

Both 5-HT and FMRFamide evoke long-lasting changes in the efficacy of sensorimotor (SN-L7) synapses of Aplysia, structural alterations of the presynaptic sensory cell, and cell-specific downregulation in the distribution of the adhesion molecule apCAM. The cell-specific changes in apCAM, related to vertebrate NCAM and Drosophila Fas II, contribute to the formation of new presynaptic varicosities by 5-HT and the elimination of existing presynaptic varicosities by FMRFamide. The formation of new sensory varicosities is directed by the presence of preexisting zones on the motor axon that are enriched for apCAM. Moreover, there was a further enrichment of apCAM levels at existing sensory varicosities contacting the motor axon beginning at 1 hr and lasting 24 hr after treatment with 5-HT. As was found for synapse formation during the early stages of cell-cell interaction, incubation with anti-apCAM mAb blocks the 5-HT-induced long-term changes in synaptic efficacy and the accompanying changes in sensory neuron structure. Long-term synaptic depression with FMRFamide is accompanied by an overall decline of apCAM levels. Treatment with FMRFamide evokes an even greater decline in apCAM levels at sites of sensory varicosities that precede the structural changes and persist especially at sites where sensory varicosities are eliminated. These results suggest that neurotransmitters evoke both cell- and site-specific changes in the levels of adhesion molecules that can influence either the formation or the elimination of presynaptic varicosities that accompany long-term heterosynaptic modulation of a behaviorally relevant synaptic connection (Zhu, 1995).

Olfactory sensory neurons (OSNs) are individually specified to express one odorant receptor (OR) gene from ~1000 different alternatives and project with precision to the glomeruli, topographically defined convergence sites in the olfactory bulb. Although ORs partially determine the location of convergence sites, the mechanism ensuring that axons with different OR identities do not co-converge is unknown. RNCAM (OCAM, NCAM2) is assumed to regulate a broad zonal segregation of projections by virtue of being a homophilic cell adhesion molecule that is selectively expressed on axons terminating in a defined olfactory bulb region. RNCAM is related to NCAM, Drosophila Fasciclin 2 (Fas2) and Aplysia CAM (apCAM) in terms of sequence similarity and domain structure. NADPH diaphorase activity serves as an independent marker for RNCAM-negative axons. Analyses of transgenic mice that ectopically express RNCAM in NADPH diaphorase-positive OSNs show that the postulated function of RNCAM in mediating zone-specific segregation of axons is unlikely. Instead, analyses of one OR-specific OSN subpopulation (P2) reveal that elevated RNCAM levels result in an increased number of P2 axons that incorrectly co-converge with axons of other OR identities. Both Gpi-anchored and transmembrane-bound RNCAM isoforms are localized on axons in the nerve layer, while the transmembrane-bound RNCAM is the predominant isoform on axon terminals within glomeruli. Overexpressing transmembrane-bound RNCAM results in co-convergence events close to the correct target glomeruli. By contrast, overexpression of Gpi-anchored RNCAM results in axons that can bypass the correct target before co-converging on glomeruli located at a distance. The phenotype specific for Gpi-anchored RNCAM is suppressed in mice overexpressing both isoforms, which suggests that two distinct RNCAM isoform-dependent activities influence segregation of OR-defined axon subclasses (Alenius, 2003).

Thus RNCAM-negative axons and their terminals within glomeruli in the OB can be visualized by NADPHd histochemistry. This novel marker for Z1 axons has allowed the function of zonally restricted expression of RNCAM in OE to be addressed. Transgenic mice with ectopic RNCAM expression in Z1 and control mice show an indistinguishable topography of NADPHd-positive and NADPHd-negative OB regions. This finding indicates that Z1 axons project to the dorsomedial OB and avoid the ventrolateral part, irrespective of whether they express RNCAM or not. Conversely, axons of a defined RNCAM-positive OSN subpopulation (P2) are not attracted to the NADPHd-positive (Z1) region of the OB in RNCAM transgenic mice. This indicates that RNCAM does not determine formation of a broad regional division of projections from OE to OB. A candidate molecule with such functions is the semaphorin receptor neuropilin 2, which has been shown to mediate zonal segregation, but not axonal convergence, of olfactory sensory neurons located in the vomeronasal organ. Supporting the notion that other genes may regulate the formation of zone-specific projections in the main olfactory system is the finding that neuropilin 2 mRNA is expressed in a step gradient, with highest relative levels in Z4 and lowest in Z1 (Alenius, 2003).

If RNCAM does not determine formation of a broad regional division of projections, it is possible that the topographic distribution of RNCAM expression has evolved to play a role in axon convergence and/or structural plasticity for neurons in Z2-4. In fact, previous studies of OSN-expressing tagged ORs corresponding to a RNCAM-negative Z1 OR (M72) and a RNCAM-positive Z2 OR (P2) have provided evidence for heterogeneity in glomerular formation during development and sensitivity to targeted deletion of an olfactory cyclic nucleotide-gated channel subunit. To determine if RNCAM function influences axon convergence, the P2-IRES-tau-lacZ allele was used to probe for navigational errors of P2 axons. One finding is that increased expression of RNCAM in OSNs results in an increased number of glomeruli that are innervated by axons that represent more than one OR identity. This indicates that normal level of RNCAM is important for the formation of highly refined axon projections of OSNs, to which RNCAM expression is normally confined. Taken together, three important functional characteristics of RNCAM have been uncovered: (1) RNCAM does not determine the division of OSN projections into Z1 and Z2-4, respectively; (2) normal regulated levels of RNCAM are important for accurate OR-specific axon segregation of Z2-4 OSNs; (3) specific activities of RNCAM splice variants have been identified that influence formation of precise topographic axon connections in the vicinity and distantly relative to the correct target (Alenius, 2003).

Mammalian NCAM and synaptic function

Neural recognition molecules such as the neural cell adhesion molecule (NCAM) have been implicated in synaptic plasticity, including long-term potentiation (LTP), sensitization, and learning and memory. The major isoform of NCAM carrying the longest cytoplasmic domain of all NCAM isoforms (NCAM180) is predominantly localized in postsynaptic membranes and postsynaptic densities of hippocampal neurons, with only a proportion of synapses carrying detectable levels of NCAM180. To investigate whether this differential expression of NCAM180 may correlate with distinct states of synaptic activity, LTP was induced by high-frequency stimulation of the perforant path and the percentage of NCAM180 immunopositive spine synapses determined in the outer third of the dentate molecular layer of the dentate gyrus by immunoelectron microscopy. Twenty-four hours following induction of LTP by high-frequency stimulation, the percentage of spine synapses expressing NCAM180 increases from 37% (passive control) to 70%. This increase is inhibited by the noncompetitive NMDA receptor antagonist MK801. Following repeated LTP induction on 10 consecutive days with one tetanization each day, 60% of all spine synapses were NCAM180 immunoreactive. Compared to passive control animals, the percentage of NCAM180 expressing synapses in low-frequency stimulated animals decreased from 37% to 28%. Spine synapses in the inner part of the dentate molecular layer not contacted by the afferents of the perforant path does not change the percentage of NCAM180-expressing synapses. The results obtained by the postembedding immunogold staining technique confirm the difference in NCAM180 expression of spine synapses between passive control and potentiated animals. These observations suggest a role for NCAM180 in synaptic remodeling accompanying LTP (Schuster, 1998).

To evaluate the contributions of the pre- versus post-synaptic expression of NCAM in regulation of synaptic efficacy, dissociated hippocampal cells from NCAM-deficient and wild-type mice were cultured in homo- and heterogenotypic combinations. Double recordings from synaptically coupled neurons maintained in hetero-genotypic cocultures show that synaptic strength of excitatory but not inhibitory synapses depend on expression of NCAM post- but not pre-synaptically. This correlates with higher levels of potentiation and synaptic coverage of NCAM-expressing neurons compared to NCAM-deficient neurons in heterogenotypic cocultures. Synaptic density is the same in homogenotypic cultures of NCAM-deficient and wild-type neurons as well as in heterogenotypic cocultures in which glutamate receptors are blocked. These observations indicate that the relative levels of postsynaptic NCAM expression control synaptic strength in an activity-dependent manner by regulating the number of synapses (Dityatev, 2000).

The data demonstrate that the preference of NCAM-expressing cells as postsynaptic targets is abolished when glutamate receptors are inhibited in the heterogenotypic cocultures, suggesting that activation of glutamate receptors is involved in target selection at least at certain developmental stages. These results are interesting in view of studies demonstrating that activation of AMPA receptors increases the activity of the NCAM promoter, and postsynaptic expression of the largest NCAM isoform, NCAM180, is increased following long-term potentiation of synapses in the dentate gyrus. Thus, activation of glutamate receptors can upregulate expression of NCAM and other adhesion molecules and enhance the relative difference in attractiveness of postsynaptic targets. This likely happens at a relatively slow time scale of hours and days. The study also shows that postsynaptic NCAM is important for rapid activity-dependent synaptic modifications such as LTP, since no significant potentiation of amplitude of spontaneous EPSCs is induced in response to glutamate stimulation in NCAM-deficient neurons in homo- or heterogenotypic situations. Interestingly, the levels of potentiation in NCAM +/+ neurons relative to NCAM -/- neurons are higher in heterogenotypic cocultures than in homogenotypic cultures, again highlighting the role of a differential expression of NCAM in regulation of synaptic functions. These experiments link the current work with previous studies that have made use of NCAM antibodies, NCAM-derived peptides, and knockout mice to demonstrate an involvement of NCAM in LTP. This study adds a novel point to these data, implicating postsynaptically expressed NCAM in enhancement of synaptic strength. The importance of postsynaptic NCAM for LTP is in agreement with recent data obtained in hippocampal slices, in which the largest postsynaptic major isoform of NCAM, NCAM180, appears to be an essential determinant in synaptic plasticity, since transgenic reintroduction of NCAM180 into NCAM knockout animals normalizes long-term potentiation in comparison to wild-type animals, with NCAM knockout mice being deficient in long-term potentiation. The observation that NCAM is an important molecule in enhancing synaptic strength by a postsynaptic mechanism will be instrumental in defining the molecular mechanisms that underlie the positioning of adhesion molecules via the cytoskeleton and via signal transduction mechanisms in activity-dependent processes (Dityatev, 2000).

The neural cell adhesion molecule (NCAM) regulates synapse formation and synaptic strength via mechanisms that have remained unknown. This study shows that NCAM associates with the postsynaptic spectrin-based scaffold, cross-linking NCAM with the NMDA receptor and Ca²⁺/calmodulin-dependent protein kinase II α (CaMKIIα) in a manner not firmly or directly linked to PSD95 and α-actinin. Clustering of NCAM promotes formation of detergent-insoluble complexes enriched in postsynaptic proteins and resembling postsynaptic densities. Disruption of the NCAM-spectrin complex decreases the size of postsynaptic densities and reduces synaptic targeting of NCAM-spectrin-associated postsynaptic proteins, including spectrin, NMDA receptors, and CaMKIIα. Degeneration of the spectrin scaffold in NCAM-deficient neurons results in an inability to recruit CaMKIIα to synapses after NMDA receptor activation, which is a critical process in NMDA receptor-dependent long-term potentiation. The combined observations indicate that NCAM promotes assembly of the spectrin-based postsynaptic signaling complex, which is required for activity-associated, long-lasting changes in synaptic strength. Its abnormal function may contribute to the etiology of neuropsychiatric disorders associated with mutations in or abnormal expression of NCAM (Sytayk, 2007).

Fasciclin 2 Evolutionary homologs: Return part 1/3 | part 2/3

Home page: The Interactive Fly © 1995, 1996 Thomas B. Brody, Ph.D.

The Interactive Fly resides on the
Society for Developmental Biology's Web server.