Fasciclin 2


Functional coupling of L1 and Ankyrin

The L1 CAM family of cell adhesion molecules and the ankyrin family of spectrin-binding proteins are candidate collaborators in transcellular complexes used in diverse contexts in nervous systems of vertebrates and invertebrates. This report presents evidence for functional coupling between L1 and 440-kD ankyrinB in premyelinated axons in the mouse nervous system. L1 and 440-kD ankyrinB are colocalized in premyelinated axon tracts in the developing nervous system and are both down-regulated after myelination. AnkyrinB (-/-) mice exhibit a phenotype similar to, but more severe, than L1 (-/-) mice and share features of human patients with L1 mutations. AnkyrinB (-/-) mice exhibit hypoplasia of the corpus callosum and pyramidal tracts, dilated ventricles, and extensive degeneration of the optic nerve, and they die by postnatal day 21. AnkyrinB (-/-) mice have reduced L1 in premyelinated axons of long fiber tracts, including the corpus callosum, fimbria, and internal capsule in the brain, and pyramidal tracts and lateral columns of the spinal cord. L1 is evident in the optic nerve at postnatal day 1 but disappears by postnatal day 7 in mutant mice, while NCAM is unchanged. Optic nerve axons of ankyrinB (-/-) mice become dilated with diameters up to eightfold greater than normal, and they degenerate by day 20. These findings provide the first evidence of a role for ankyrinB in the nervous system and support an interaction between 440-kD ankyrinB and L1 that is essential for maintenance of premyelinated axons in vivo (Scotland, 1998).

Neural cell adhesion molecules (CAMs) of the immunoglobulin (Ig) superfamily mediate not only cell aggregation but also growth cone guidance and neurite outgrowth. Two neural CAMs, L1-CAM and TAG-1, induce the homophilic aggregation of Drosophila S2 cells but are unable to interact with each other when expressed on different cells (trans-interaction). However, immunoprecipitations from cells co-expressing L1-CAM and TAG-1 show a strong cis-interaction between the two molecules in the plane of the plasma membrane. TAG-1 is linked to the membrane by a glycosylphosphatidylinositol (GPI) anchor and therefore is unable to directly interact with cytoplasmic proteins. In contrast, L1-CAM-mediated homophilic cell adhesion induces the selective recruitment of the membrane skeleton protein ankyrin to areas of cell contact. Immunolabeling experiments in which S2 cells expressing TAG-1 were mixed with cells co-expressing L1-CAM and TAG-1 have demonstrated that the homophilic interaction between TAG-1 molecules results in the cis-activation of L1-CAM to bind ankyrin. This TAG-1-dependent recruitment of the membrane skeleton provides an example of how GPI-anchored CAMs are able to transduce signals to the cytoplasm. Furthermore, such interactions might ultimately result in the recruitment and the activation of other signaling molecules at sites of cell contacts (Malhotra, 1998).

Numb mediated endocytosis of L1

Axon growth during neural development is highly dependent on both cytoskeletal re-organization and polarized membrane trafficking. Collapsin response mediator protein-2 (CRMP-2) is critical for specifying axon/dendrite fate and axon growth in cultured hippocampal neurons, possibly by interacting with tubulin heterodimers and promoting microtubule assembly. Numb is identified as a CRMP-2-interacting protein. Numb has been shown to interact with alpha-adaptin and to be involved in endocytosis. Numb was associated with L1, a neuronal cell adhesion molecule that is endocytosed and recycled at the growth cone, where CRMP-2 and Numb colocalize. Furthermore, expression of dominant-negative CRMP-2 mutants or knockdown of CRMP-2 message with small-interfering (si) RNA inhibits endocytosis of L1 at axonal growth cones and suppresses axon growth. These results suggest that in addition to regulating microtubule assembly, CRMP-2 is involved in polarized Numb-mediated endocytosis of proteins such as L1 (Nishimura, 2003).

Transcriptional regulation of L1 and N-CAM

A binding site for the paired domain of Pax proteins (designated PBS) has been identified in the mouse N-CAM promoter. A transcription factor known to be important for development of the central nervous system, Pax-6 (Drosophila homolog: Eyeless), binds to the N-CAM PBS; the PBS can also influence N-CAM expression in vivo. Pax-6, produced in COS-1 cells, binds to the PBS through two half-sites, PBS-1 and PBS-2; mutations in both of these sites completely disrupted binding. Moreover, nuclear extracts from embryonic day (E) 11.5 mouse embryos bind to the PBS, and this binding is inhibited by antibodies to Pax-6. To determine the role of the PBS in vivo, transgenic mice were generated with N-CAM promoter/lacZ gene constructs containing either a wild-type or a mutated PBS. Mutations in PBS-1 and PBS-2 decrease the extent of beta-galactosidase expression in the mantle layer of the spinal cord limiting it to ventral regions at E11.5. At E14.5, these mutations eliminated most of the expression that was seen in the wild-type spinal cord. Taken together with previous observations that the PBS binds multiple Pax proteins, the data indicate that such binding contributes to the regulation of N-CAM gene expression during neural development (Holst, 1997).

To identify proteins that bind to a regulatory element common to the genes for two neural CAMs, Ng-CAM and L1, a mouse cDNA expression library was screened with a concatamer of the sequence CCATTAGPyGA. A new homeobox gene, called Barx2, was found. The homeodomain encoded by Barx2 is 87% identical to that of Barx1, and both genes are related to genes at the Bar locus of Drosophila. In vitro, Barx2 stimulates activity of an L1 promoter construct containing the CCATTAGPyGA motif but represses activity when this sequence is deleted. Localization studies show that expression of Barx1 and Barx2 overlap in the nervous system, particularly in the telencephalon, spinal cord, and dorsal root ganglia. Barx2 is also prominently expressed in the floor plate and in Rathke's pouch. The localization data, combined with Barx2's dual function as activator and repressor, suggest that Barx2 may differentially control the expression of L1 and other target genes during embryonic development (F. Jones, 1997).

The cell adhesion molecule L1 mediates neurite outgrowth and fasciculation during embryogenesis; mutations in this gene have been linked to a number of human congenital syndromes. To identify DNA sequences that restrict expression of L1 to the nervous system, a previously unidentified segment of the mouse L1 gene containing the promoter, the first exon, and the first intron were isolated and its activity was examined in vitro and in vivo. A neural restrictive silencer element (NRSE) within the second intron prevents expression of L1 gene constructs in nonneural cells. For optimal silencing of L1 gene expression by the NRSE-binding factor RE-1-silencing transcription factor (REST)/NRSF, both the NRSE and sequences in the first intron are required. In transgenic mice, an L1lacZ gene construct with the NRSE generates a neurally restricted expression pattern consistent with the known pattern of L1 expression in postmitotic neurons and peripheral glia. In contrast, a similar construct lacking the NRSE produces precocious expression in the peripheral nervous system and ectopic expression in mesenchymal derivatives of the neural crest and in mesodermal and ectodermal cells. These experiments show that the NRSE and REST/NRSF are important components in restricting L1 expression to the embryonic nervous system (Kallunki, 1997).

The cell adhesion molecule L1 mediates axonal guidance during neural development. Mutations in its gene result in severe neurological defects. In previous studies, the promoter for the L1 gene was identified and a neural restrictive silencer element (NRSE) was shown to be critical for preventing ectopic expression of L1 during early embryonic development. The role of the NRSE has been invstigated in the regulation of L1 expression during postnatal development. In gel mobility shift experiments, the NRSE forms DNA-protein complexes with nuclear extracts prepared from the brains of postnatal mice. To examine the influence of the NRSE on postnatal patterns of L1 expression in vivo, the expression of two lacZ transgene constructs was compared, one containing the native L1 gene regulatory sequences (L1lacZ) and another (L1lacZN) lacking the NRSE. Newborn mice carrying the L1lacZN show enhanced beta-galactosidase expression, relative to L1lacZ in the brain and ectopic expression in nonneural tissues. However, and in contrast to L1lacZ mice, during postnatal development and in the adult, L1lacZN mice show an unexpected loss of beta-galactosidase expression in several neural structures, including the neural retina, cerebellum, cortex, striatum, and hippocampus. These data support the conclusion that the NRSE not only plays a role in the silencing of L1 expression in nonneural tissues during early development but also can function as a silencer and an enhancer of L1 expression in the nervous system of postnatal and adult animals (Kallunki, 1998).

To study promoter regulation in vivo for the neural cell adhesion molecule, N-CAM, homologous recombination was used to insert the bacterial lacZ gene between the transcription and translation initiation sites of the N-CAM gene. This insertion disrupts the gene and places the expression of beta-galactosidase under the control of the N-CAM promoter. Animals homozygous for the disrupted allele do not express N-CAM mRNA or protein, but the pattern of beta-galactosidase expression in heterozygous and homozygous embryos is similar to that of N-CAM mRNA in wild-type animals. The homozygotes exhibit many of the morphological abnormalities observed in previously reported N-CAM knockout mice, with the exception that hippocampal long-term potentiation in the Schaffer collaterals is identical in homozygous, heterozygous, and wild-type animals. Homozygous knockout mice have a bifurcation of the CA3 region of the hippocampus not present in wild-type animals but similar to that reported in other N-CAM knockout mice. Heterozygous animals showed an intermediate morphological phenotype with a less pronounced bifurcation. In spite of these defects, the projections from the dentate gyrus to the CA3 region were similar in mutant and wild-type animals. Homozygous animals have an increase in the number of cells in the subventricular zone on the route of migration to the olfactory bulb. They also show a reduction in the size of the olfactory bulb. Baseline synaptic physiology and LTP in hippocampal slices from heterozygous and homozygous N-CAM knockout mice appears normal (Holst, 1998).

Heterozygous mice were used to examine the regulation of the N-CAM promoter in response to enhanced synaptic transmission. Treatment of the mice with an ampakine, an allosteric modulator of alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors that enhances normal glutamate-mediated synaptic transmission, increases the expression of beta-galactosidase in vivo as well as in tissue slices in vitro. Similar treatments also increase the expression of N-CAM mRNA in the heterozygotes. The effects of ampakine in slices are strongly reduced in the presence of 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX), an AMPA receptor antagonist. Taken together, these results indicate that facilitation of AMPA receptor-mediated transmission leads to activation of the N-CAM promoter and provide support for the hypothesis that N-CAM synthesis is regulated in part by synaptic activity (Holst, 1998).

The N-CAM gene is one among a wide range of genes that are transcribed or translated in response to heightened neural activity and after certain learning paradigms. The mechanism of activation of many of these genes has been linked to phosphorylation of the transcription factor CREB that binds to a consensus DNA element called a cAMP response element (CRE). A CRE-regulated promoter-reporter construct has been shown to be activated in slice cultures by LTP-inducing stimulation. Although no CRE has been found in the proximal 1 kb of the N-CAM promoter, the response of the entire promoter to cAMP-elevating agents has not yet been explored. It should be possible to use transfection of promoter constructs into slice cultures to identify the cis sequences within the promoter that respond to membrane depolarization via activity-regulated transcription factors. Combined with pharmacological approaches to identify signaling intermediates, the slice culture paradigm should allow an identification of potential cis- and trans-acting factors that are important for the regulation of N-CAM expression by alterations in synaptic transmission (Holst, 1998).

The cell adhesion molecule L1 regulates axonal guidance and fasciculation during development. The regulatory region of the L1 gene has been identified and shown to be sufficient for establishing the neural pattern of L1 expression in transgenic mice. A DNA element identified within this region, called the HPD, contains binding motifs for both homeodomain and Pax proteins and responds to signals from bone morphogenetic proteins (BMPs). An ATTA sequence within the core of the HPD was required for binding to the homeodomain protein Barx2 while a separate paired domain recognition motif is necessary for binding to Pax-6. In cellular transfection experiments, L1-luciferase reporter constructs containing the HPD are activated an average of 4-fold by Pax-6 in N2A cells and 5-fold by BMP-2 and BMP-4 in Ng108 cells. Both of these responses are eliminated on deletion of the HPD from L1 constructs. In transgenic mice, deletion of the HPD from an L1-lacZ reporter results in a loss of beta-galactosidase expression in the telencephalon and mesencephalon. Collectively, these experiments indicate that the HPD regulates L1 expression in neural tissues via homeodomain and Pax proteins and is likely to be a target of BMP signaling during development (Meech, 1999).

Barx1 and Barx2 are homeodomain proteins originally identified using regulatory elements of genes encoding certain cell adhesion molecules (CAMs). In the present study, regions of Barx2 were characterized that bind to regulatory elements of genes encoding three CAMs, L1, neuron-glia CAM (Ng-CAM), and neural CAM (N-CAM); domains of Barx2 were identified that regulate N-CAM transcription. The homeodomain of Barx2 is sufficient for binding to homeodomain binding sites (HBS) from all three CAM genes. The presence of a 17-amino acid Barx basic region resulted in a 2-fold decrease in binding to HBS sequences from the Ng-CAM and L1 genes, whereas it led to a 6.5-fold increase in binding to the HBS from the N-CAM promoter. Thus, the Barx basic region influences the strength and specificity of Barx2 binding to DNA. In co-transfection experiments, Barx2 repressed N-CAM promoter activity. A 24-residue N-terminal region of Barx2 is essential for repression. When this region is absent, Barx2 activates the N-CAM promoter. A 63-residue C-terminal domain is required for this activation. In GST pull-down experiments, Barx2 binds to proteins of the CREB family, CREB1 and ATF2. Overall, these findings provide a framework for understanding developmental and physiological contexts that influence repressor or activator functions of Barx2 (Edelman, 2000).

Effects of N-CAM and L1 mutation

Cell adhesion molecules (CAMs) are known to be involved in a variety of developmental processes that play key roles in the establishment of synaptic connectivity during embryonic development, but recent evidence implicates the same molecules in synaptic plasticity of the adult. In the present study, neural CAM (NCAM)-deficient mice, which have learning and behavioral deficits, were used to evaluate NCAM function in the hippocampal mossy fiber system. Morphological studies demonstrate that fasciculation and laminar growth of mossy fibers are strongly affected, leading to innervation of CA3 pyramidal cells at ectopic sites, whereas individual mossy fiber boutons appear normal. Electrophysiological recordings performed in hippocampal slice preparations reveal that both basal synaptic transmission and two forms of short-term plasticity, i.e., paired-pulse facilitation and frequency facilitation, are normal in mice lacking all forms of NCAM. However, long-term potentiation of glutamatergic excitatory synapses after brief trains of repetitive stimulation is abolished. Taken together, these results strongly suggest that in the hippocampal mossy fiber system, NCAM is essential both for correct axonal growth and synaptogenesis and for long-term changes in synaptic strength (Cremer, 1998).

Mutations in the gene for neural cell adhesion molecule L1 (L1CAM) result in a debilitating X-linked congenital disorder of brain development. In man, mutations in the L1 gene have been found to be responsible for a clinically variable X-linked recessive disorder described as either X-linked hydrocephalus, MASA syndrome or spastic paraplegia type I (SPG1). The cardinal features of these disorders are varying degrees of mental retardation and spasticity, frequently accompanied by congenital hydrocephalus and flexion deformities of the thumbs. A related phenotype that is partially dependent on genetic background is seen in the mouse L1 knockout lines. These observations not only confirm the importance of L1 for neural development but also highlight areas where L1 may have a pivotal role. For example, malformation of the corticospinal tract in both mouse and man may explain the spasticity seen in both species. Furthermore, abnormal guidance of these axons across the midline confirms a role for L1 not only in growth, but also in correct guidance of developing axons of subsets of neurons. Moreover, malformations such as underdevelopment of the anterior vermis of the cerebellum and fused thalami may be the result of abnormal migration of a subset of cells. At the neuronal cell surface L1 may interact with a variety of different molecules including itself and two other CAMs of the immunoglobulin superfamily, axonin-1 and F11. However, whether all of these interactions are relevant to normal or abnormal development has not been determined. Over one-third of patient mutations are single amino acid changes distributed across 10 extracellular L1 domains. The effects of 12 missense mutations on binding to L1, axonin-1 and F11 have been studied and for the first time it has been shown that whereas many mutations affect all three interactions, others affect homophilic or heterophilic binding alone. Perhaps the most striking observation is that L1 mutations affect binding to axonin-1 and F11 in a similar fashion and thus these two proteins may interact with L1 in a very similar, although not identical, manner. The implication from this observation is that axonin-1 and F11 are unlikely to interact with L1 at the same time. Patient pathology is therefore due to different types of L1 malfunction. The nature and functional consequence of mutation is also reflected in the severity of the resultant phenotype with structural mutations likely to affect more than one binding activity and result in early mortality. Moreover, the data indicate that several extracellular domains of L1 are required for homophilic and heterophilic interactions (De Angelis, 1999).

The nine mutations that disrupt homophilic binding reside in several domains (Ig1, Ig2, Ig3, Ig4, Ig5, Ig6 and FNIII-like domain 2), indicating that the integrity of all of these domains is required for wild-type binding, although the contribution of each may vary. For those mutations that effect key structural residues their effect may not be local but may spread to conformationally linked domains. However, structural mutations are possibly limited in their ability to transmit conformational change beyond a single domain and its boundaries and, therefore, these results still implicate a large portion of the molecule in homophilic interaction. In contrast, two mutations, H210Q in Ig2 and A426D in Ig5, severely reduce homophilic binding although they are highly unlikely to disrupt domain structure. These two mutations may highlight important contact sites in L1-L1 interactions or in the formation of interdomain tertiary structure. The D598N mutation in Ig6, that has a modest but significant effect on homophilic binding, also affects a surface site. Additional evidence for the involvement of several domains in homophilic binding comes from consideration of the insect Ig superfamily adhesion molecule hemolin and domain deletion studies on chick NgCAM. X-ray crystallography of the four Ig-domain hemolin proteins indicates that a horseshoe structure can be adopted that is stabilized by interactions of the first domain with the fourth, and the second domain with the third (Su, 1998). This structure is possible due to the small seven-residue spacer region between Ig domains 2 and 3, which is also found in mammalian L1 and chick NgCAM. Moreover, many of the key residues required for hemolin folding are conserved in L1 subgroup members. There is a high degree of conservation at the putative intramolecular contact sites for L1 and hemolin and therefore it seems possible that the first four domains of L1 can also adopt a horseshoe structure. Interestingly, 50% of the known pathological human missense mutations in the first four domains of L1 lie within the regions defined as intramolecular binding sites for hemolin, even though these binding faces comprise only 13% of the residues. Those mutations that lie within these regions include the G121S, R184Q and C264Y; these are all changes that drastically affect homophilic binding. A model is therefore proposed in which the first four domains of L1 adopt an intramolecular hemolin-like fold and the extracellular domains overlap in an antiparallel fashion. The exact degree of overlap cannot be determined on the basis of these data and will require further analysis of mutated constructs. The effect of the V768F mutation suggests that antiparallel overlap may involve this domain although it is also possible that this mutation has an indirect effect through affecting the presentation of Ig domains (De Angelis, 1999 and references).

Genetic evidence indicates that cell adhesion molecules of the immunoglobulin superfamily (IgCAMs) are critical for activity-dependent synapse formation at the neuromuscular junction in Drosophila and have also been implicated in synaptic remodelling during learning in Aplysia. In mammals, a widely adopted model for the process of learning at the cellular level is long-term potentiation (LTP) in the hippocampal formation. Studies in vitro have shown that antibodies to the IgCAMs L1 and NCAM reduce LTP in CA1 neurons of rat hippocampus, suggesting a role for these molecules in the modulation of synaptic efficacy, perhaps by regulating synaptic remodelling. A role for NCAM in LTP has been confirmed in mice lacking NCAM, but similar studies have not been reported for L1. LTP was examined in the hippocampus of mice lacking L1, using different experimental protocols in three different laboratories. In tests of LTP in vitro and in vivo no significant differences were found between mutant animals and controls. Thus, contrary to expectation, the data suggest that L1 function is not necessary for the establishment or maintenance of LTP in the hippocampus. Impaired performance in spatial learning exhibited by L1 mutants may therefore not be due to hippocampal dysfunction (Bliss, 2000).

The circadian cycle is a simple, universal molecular mechanism for imposing cyclical control on cellular processes. The regulation of one of the key circadian genes, Clock, was examined in early Xenopus development. Xclk expression is found initially in the organizer region and overlying ectoderm, coincident with the neural inducer noggin. Further, noggin can induce Xclk expression in ectodermal explants along with markers of neural plate, including the cell adhesion molecule NCAM. Interestingly, NCAM is required for photic resetting of the circadian clock in mice; a mutation in NCAM that blocks its conjugation to polysialic acid results in the gradual running down of the circadian cycle in the SCN. The expression of Clock is dependent on developmental stage, not on time per se, and is mostly restricted to the anterior neural plate. It's expression can be induced by the secreted polypeptide noggin, and subsequently upregulated by Otx2, a transcription factor required for the determination of anterior fate (Green, 2001).

The L1 adhesion molecule regulates axon growth and is mutated in the X-linked mental retardation syndrome CRASH (acronym for corpus callosum agenesis, retardation, aphasia, spastic paraplegia, hydrocephalus). A novel role for L1 as a potentiator of neuronal cell migration to extracellular matrix proteins through ß1 integrins and intracellular signaling to mitogen-activated protein (MAP) kinase has been identified. L1 potentiates haptotactic migration of B35 neuroblastoma cells toward fibronectin, vitronectin, and laminin through the signaling intermediates c-Src, phosphatidylinositol-3 kinase, and MAP kinase. L1 potentiated migration toward fibronectin through alpha5ß1 integrin in human embryonic kidney 293 cells and depends on determinants of L1 endocytosis: dynamin I, c-Src, and the AP2/clathrin binding site (Arg-Ser-Leu-Glu) in the neuronal splice form of L1. L1 clustering on the cell surface enhances the internalization of activated ß1 integrins and L1 into distinct endocytic vesicles. L1-potentiated migration, enhancement of ß1 integrin endocytosis, and activation of MAP kinase are coordinately inhibited by mutation of an RGD sequence in the sixth immunoglobulin-like domain of L1. Moreover, three CRASH mutations in the L1 cytoplasmic domain (1194L, S1224L, Y1229H), two of which interfere with ankyrin association, inhibits L1-potentiated migration and MAP kinase activation. Function-blocking antibodies to L1 and ß1 integrin retards the migration of 5-bromo-2'-deoxyuridine-labeled mouse cerebellar granule cells in slice cultures, underscoring the potential physiological relevance of these findings. These studies suggest that L1 functionally interacts with ß1 integrins to potentiate neuronal migration toward extracellular matrix proteins through endocytosis and MAP kinase signaling, and that impairment of this function by L1 cytoplasmic domain mutations may contribute to neurological deficits in CRASH (Thelen, 2002).

A new mouse line has been produced in which the sixth Ig domain of the L1 cell adhesion molecule has been deleted. Despite the rather large deletion, L1 expression is preserved at normal levels. In vitro experiments showed that L1-L1 homophilic binding is lost, along with L1-alpha5beta1 integrin binding. However, L1-neurocan and L1-neuropilin binding are preserved and sema3a responses are intact. Surprisingly, many of the axon guidance defects present in the L1 knockout mice, such as abnormal corticospinal tract and corpus callosum, were not observed. Nonetheless, when backcrossed on the C57BL/6 strain, a severe hydrocephalus was observed and after several generations, became an embryonic lethal. These results imply that L1 binding to L1, TAG-1, or F3, and L1-alpha5beta1 integrin binding are not essential for normal development of a variety of axon pathways, and suggest that L1-L1 homophilic binding is important in the production of X-linked hydrocephalus (Itoh, 2004).

The neural cell recognition molecule Close Homolog of L1 (CHL1) is required for neuronal positioning and dendritic growth of pyramidal neurons in the posterior region of the developing mouse neocortex. CHL1 was expressed in pyramidal neurons in a high-caudal to low-rostral gradient within the developing cortex. Deep layer pyramidal neurons of CHL1-minus mice are shifted to lower laminar positions in the visual and somatosensory cortex and develop misoriented, often inverted apical dendrites. Impaired migration of CHL1-minus cortical neurons is suggested by strikingly slower rates of radial migration in cortical slices, failure to potentiate integrin-dependent haptotactic cell migration in vitro, and accumulation of migratory cells in the intermediate and ventricular/subventricular zones in vivo. The restriction of CHL1 expression and effects of its deletion in posterior neocortical areas suggests that CHL1 may regulate area-specific neuronal connectivity and, by extension, function in the visual and somatosensory cortex (Demyanenko, 2004).

CHL1 is the most recently identified member of the L1 family. Its amino acid sequence is ~60% identical to L1 in the extracellular region and ~40% identical in the cytoplasmic domain. CHL1 is a strong promoter of neurite outgrowth in vitro and may bind homophilically as well as heterophilically. CHL1 knockout mice display aberrant guidance of olfactory axons and hippocampal mossy fibers, and these mice are defective in cognitive processing of spatial information and attention. In humans, mutations in the CHL1 gene ortholog, CALL, are associated with the human 3p syndrome, which is characterized by mental retardation or low IQ, and delayed speech and motor development. L1 mutations induce a mental retardation syndrome that is more complex than the 3p syndrome and may cause spasticity, corpus callosum agenesis, and optic atrophy. Many of these features are phenocopied in L1 knockout mice. Polymorphisms in both the CHL1/CALLand L1 genes have been linked to increased risk for schizophrenia. These findings suggest a role for CHL1 in human brain development and function (Demyanenko, 2004).

Migration of cortical neurons to their final location in the cerebral cortex and the ensuing development of dendrites and axons are fundamental processes in the formation of the cerebral cortex. The neocortical region of the brain is comprised of distinct areas with a common organization but with different functions and patterns of neuronal connectivity. An important goal is to elucidate the molecules and mechanisms that govern cortical area-specific neuronal differentiation. Transcription factors with graded expression in the embryonic cortex are required for normal patterning, but less is known about area-specific guidance molecules for migrating neurons, whose expression may be governed by such factors. CHL1 emerged in a screen for genes differentially expressed in the caudal versus rostral neocortex of the rat. CHL1 mRNA is localized in a spatially and temporally graded pattern in migrating neuronal precursors and in differentiating, postmitotic pyramidal neurons enriched in layer V. The graded pattern of CHL1 expression suggests a function in cortical area-specific development (Demyanenko, 2004).

Based on these results, an investigation was undertaken to examine a potential role for CHL1 in development of cortical pyramidal neurons in the mouse neocortex. It was anticipated that CHL1 might modulate cortical neuron migration, because it has been shown that expression of CHL1 in transfected HEK293 cells strongly potentiates haptotactic migration to extracellular matrix proteins in vitro. Moreover, the stimulation of migration by CHL1 in these cells is dependent on integrins, which have been implicated in cortical neuron adhesion and migration. This study identifies a cortical area-specific role for CHL1 in modulating radial migration and apical dendrite development of pyramidal cells (Demyanenko, 2004).

In the mammalian brain, ongoing neurogenesis via the rostral migratory stream (RMS) maintains neuronal replacement in the olfactory bulb throughout life. Mechanisms that regulate the final number of new neurons in this system include proliferation, migration and apoptosis. The polysialylated isoforms of the neural cell adhesion molecule (PSA-NCAM) act as a pro-survival molecule in immature newborn neurons. Confocal microscopic analysis revealed a threefold increase in TUNEL-positive cells in the subventricular zone (SVZ) and the RMS of transgenic animals lacking the gene encoding NCAM (NCAM-/-), as compared with wild types. The enhanced apoptotic cell death occurred specifically in the population of mCD24-positive newborn neurons, but not in GFAP-positive astrocytes. Using in vitro cultures of purified SVZ-derived neurons, it has been demonstrated that the loss or inactivation of PSA on NCAM, as well as the deletion of NCAM, lead to reduced survival in response to neurotrophins including BDNF and NGF. These changes in cell survival are accompanied by an upregulation of p75 neurotrophin receptor expression in vitro as well as in vivo. Furthermore, the negative effects of PSA-NCAM inactivation on cell survival can be prevented by the pharmacological blockade of the p75 receptor-signaling pathway. It is proposed that PSA-NCAM may promote survival by controlling the expression of the p75 receptor in developing neurons (Gascon, 2007).

Trafficking of NgCAM to the axonal surface

The trafficking of two endogenous axonal membrane proteins, VAMP2 (Drosophila homolog: Synaptobrevin-62A) and NgCAM, has been examined in order to elucidate the cellular events that underlie their polarization. VAMP2 is delivered to the surface of both axons and dendrites, but preferentially endocytosed from the dendritic membrane. A mutation in the cytoplasmic domain of VAMP2 that inhibits endocytosis abolished its axonal polarization. In contrast, the targeting of NgCAM depends on sequences in its ectodomain, which mediate its sorting into carriers that preferentially deliver their cargo proteins to the axonal membrane. These observations show that neurons use two distinct mechanisms to polarize proteins to the axonal domain: selective retention in the case of VAMP2; selective delivery in the case of NgCAM (Sampo, 2003).

Nearly all neurons are polarized into two structurally and functionally distinct domains, the axon and the dendrites. Consistent with the different physiological properties of axons and dendrites and their different roles in cell signaling, many cell surface proteins are preferentially distributed either to the axonal or somatodendritic domain. In a general sense, the trafficking pathways involved in the biosynthesis of integral membrane proteins are well understood. These proteins are synthesized in the rough endoplasmic reticulum, pass through the Golgi complex, and are packaged into carrier vesicles, which are transported into the axons and dendrites where they deliver their contents to the plasma membrane by exocytic fusion. With respect to the trafficking of proteins destined for different destinations within the cell, however, many fundamental questions remain unanswered. Where along these pathways does the trafficking of axonal and dendritic proteins diverge? What underlying mechanisms lead to the selective localization of such proteins on the cell surface? With regard to the second question, there are two general mechanisms that could account for the selective localization of polarized proteins on the cell surface. Selectivity along the trafficking pathways en route to the plasma membrane could ensure that proteins destined for different domains are segregated from one another into different carriers and only delivered to the plasma membrane of the appropriate domain. Alternatively, proteins could be delivered equally to the plasma membrane of both domains but retained on the cell surface only in the appropriate domain (Sampo, 2003).

Both mechanisms, selective delivery and selective retention, contribute to the maintenance of polarity in epithelial cells. Many apical proteins and basolateral proteins are sorted into distinct transport carriers, either as they exit the Golgi complex or within an endosomal compartment in the cell periphery, and these carriers deliver their cargoes exclusively to the apical or basolateral surface. Other proteins, such as Na,K-ATPase and ß1 integrin, are delivered in equal amounts to both domains. The proteins that reach the inappropriate domain are rapidly removed by endocytosis, whereas those that reach the appropriate domain interact with submembranous cytoskeletal proteins, which stabilize them in the membrane and prevent their endocytosis (Sampo, 2003 and references therein).

Studies of neuronal protein targeting have revealed motifs that are required for their appropriate localization, but it is unclear if these motifs mediate selective sorting and delivery or selective retention. Both mechanisms could plausibly contribute to the maintenance of polarity in nerve cells, and there have been few experimental tests to distinguish between them. In the case of dendritic targeting, the available evidence favors the selective sorting and delivery model. The same motifs that govern the targeting of some dendritic proteins in neurons have been shown to mediate selective delivery to the basolateral surface in epithelial cells. Moreover, live-cell imaging studies have shown that carriers labeled by expression of GFP-tagged transferrin receptor, a dendritic protein, are transported into dendrites but excluded from axons, implying that such carriers deliver their protein contents only to the somatodendritic membrane. Similar results have been observed for several other GFP-tagged dendritic proteins, including the acidic amino acid transporter EAAT3, the EGF receptor, and the metabotropic glutamate receptor mGluR1a , indicating that selective sorting and delivery is likely to contribute to the polarization of many dendritic proteins (Sampo, 2003 and references therein).

The mechanisms that underlie the polarization of axonal proteins are less well understood, but several lines of evidence raise the possibility that selective retention rather than selective delivery may play a particularly important role. (1) In contrast to the situation for dendritic proteins, transport carriers labeled following expression of the GFP-tagged axonal protein NgCAM are transported into dendrites as well as axons. It is not known if the NgCAM-containing carriers that enter dendrites fuse with and deliver their contents to the dendritic membrane or simply are returned to the cell body and eventually reach the axon. (2) Axons and dendrites differ in the molecular composition of their submembranous cytoskeleton and in the components of their endocytic machinery, which could allow for the selective retention of proteins in the axon (Sampo, 2003 and references therein).

This study analyzes the trafficking of two endogenously expressed axonal proteins, NgCAM and VAMP2, in order to determine which mechanism, selective retention or selective delivery, accounts for their polarity. L1 and its chick homolog NgCAM are members of the Ig superfamily of neural cell adhesion molecules, which are thought to play a role in axonal pathfinding and fasciculation. When NgCAM is expressed in cultured hippocampal neurons, it is highly polarized to the axonal surface. VAMP2 is a synaptic vesicle v-SNARE that is required for calcium-dependent exocytosis at presynaptic specializations. Although VAMP2 is a component of synaptic vesicles, a significant fraction of VAMP2 is also present on the axonal surface. GFP-tagged VAMP2 is highly polarized to the axonal surface when expressed in cultured hippocampal neurons, and carriers containing GFP-tagged VAMP2 are transported into both dendrites and axons, like those containing NgCAM. Thus, existing data concerning the trafficking of both NgCAM and VAMP2 are equally compatible with the selective retention and selective delivery models. In order to determine which mechanism accounts for the polarization of these proteins, this study has assessed whether these proteins are preferentially endocytosed from the dendritic surface. Regions within these proteins have been identified that are required for their polarization to the axonal surface and whether these regions are likely to mediate selective retention or selective sorting has been examined. The findings indicate that the polarization of these two axonal proteins depends on distinct mechanisms: selective retention in the case of VAMP2, selective delivery in the case of NgCAM (Sampo, 2003).

The axonal polarization of NgCAM depends on information contained in the FnIII repeats in its extracellular domain. By analogy to results on apical targeting in epithelial cells, deletions within the ectodomain of NgCAM could well disrupt its sorting, allowing it to enter an inappropriate population of carriers, which deliver cargoes to the dendritic membrane. Mutations in the ectodomain are unlikely to interfere with endocytosis. Indeed, NgCAM mutants (lacking the FnIII domains) inappropriately reach the dendritic surface, and their endocytosis from the dendritic surface is readily detectable by antibody uptake. These observations strongly imply that wild-type NgCAM is preferentially delivered to the axonal membrane. Additional methods that allow direct visualization of the fusion of NgCAM carriers with the plasma membrane, such as total internal reflection fluorescence microscopy (TIR-FM), will be required to quantify differences in rates of exocytosis of these carriers in axons and dendrites. If NgCAM is selectively delivered to the axonal plasma membrane while VAMP2 is delivered equally to both axonal and dendritic domains, these two proteins must reach the membrane via different carriers. In addition, the preferential fusion of NgCAM- but not VAMP2-containing carriers with the axonal membrane implies that the former contain unique v-SNAREs (or other proteins that govern interaction with the fusion machinery) and that these proteins interact with axon-specific t-SNAREs or other tethering proteins. If this model is correct, it remains to be determined how VAMP2 and NgCAM become segregated into different carriers and where in the cell their trafficking diverges (Sampo, 2003).

The somatodendritic endosomal regulator NEEP21 facilitates axonal targeting of L1/NgCAM

Correct targeting of proteins to axons and dendrites is crucial for neuronal function. Axonal accumulation of the cell adhesion molecule L1/neuron-glia cell adhesion molecule (NgCAM) depends on endocytosis. Two endocytosis-dependent pathways to the axon have been proposed: transcytosis and selective retrieval/retention. This study shows that axonal accumulation of L1/NgCAM occurs via nondegradative somatodendritic endosomes and subsequent anterograde axonal transport, which is consistent with transcytosis. Additionally, the neuronal-specific endosomal protein NEEP21 (neuron-enriched endosomal protein of 21 kD) was identified as a regulator of L1/NgCAM sorting in somatodendritic endosomes. Down-regulation of NEEP21 leads to missorting of L1/NgCAM to the somatodendritic surface as well as to lysosomes. Importantly, the axonal accumulation of endogenous L1 in young neurons is also sensitive to NEEP21 depletion. It is proposed that small endosomal carriers derived from somatodendritic recycling endosomes can serve to redistribute a distinct set of membrane proteins from dendrites to axons (Yap, 2008).

N-CAMs and Organogenesis

Classical cell dissociation/reaggregation experiments with embryonic tissue and cultured cells have established that cellular cohesiveness, mediated by cell adhesion molecules, is important in determining the organization of cells within tissue and organs. N-CAM-deficient mice were examined to determine whether N-CAM plays a functional role in the proper segregation of cells during the development of islets of Langerhans. In N-CAM-deficient mice the normal localization of glucagon-producing alpha cells in the periphery of pancreatic islets is lost, resulting in a more randomized cell distribution. In contrast to the expected reduction of cell-cell adhesion in N-CAM-deficient mice, a significant increase in the clustering of cadherins, F-actin, and cell-cell junctions is observed, suggesting enhanced cadherin-mediated adhesion in the absence of proper N-CAM function. These data together with the polarized distribution of islet cell nuclei and Na+/K+-ATPase indicate that islet cell polarity is also affected. Finally, degranulation of beta cells suggests that N-CAM is required for normal turnover of insulin-containing secretory granules. Taken together, these results confirm in vivo the hypothesis that a cell adhesion molecule, in this case N-CAM, is required for cell type segregation during organogenesis. Possible mechanisms underlying this phenomenon may include changes in cadherin-mediated adhesion and cell polarity (Esni, 1999).

A surface antigen specific for the collecting duct (CD) epithelium has been immunopurified. Microsequencing of three polypeptides has identified the antigen as the neuronal cell adhesion molecule L1. The kidney isoform shows a deletion of exon 3. L1 is expressed in the mesonephric duct and the metanephros throughout CD development. In the adult CD examined by electron microscopy, L1 is not expressed on intercalated cells but is restricted to CD principal cells and to the papilla tall cells. By contrast, L1 appears late in the distal portion of the elongating nephron in the mesenchymally derived epithelium and decreases during postnatal development. Immunoblot analysis shows that expression, proteolytic cleavage, and the glycosylation pattern of L1 protein are all regulated during renal development. L1 is not detected in epithelia of other organs developing by branching morphogenesis. Addition of anti-L1 antibody to kidney or lung organotypic cultures induces dysmorphogenesis of the ureteric bud epithelium but not of the lung. These results suggest a functional role for L1 in CD development in vitro. It is further postulated that L1 may be involved in the guidance of the developing distal tubule and in the generation and maintenance of specialized cell phenotypes in CD (Debiec, 1998).

Considerable evidence points to an involvement of neural cell adhesion molecule (NCAM) in myoblast fusion. Changes in the level of NCAM expression, isoform specificity, and localization in muscle cells and tissues correspond to key morphogenetic events during muscle differentiation and repair. Furthermore, anti-NCAM antibodies have been shown by others to reduce the rate of myoblast fusion, whereas overexpression of NCAM cDNAs increases the rate of myoblast fusion compared to controls. A novel fusion assay, based on intracistronic complementation of lacZ, in combination with fluorescent X-gal histochemistry and immunocytochemistry, was used to assess levels of NCAM expression in individual muscle cells. The results indicate that a substantial proportion of newly fused myoblasts have NCAM expression levels unchanged from the levels of the surrounding unfused population, suggesting that increased expression of NCAM is not required for wild-type myoblasts to fuse. Moreover, pure populations of primary myoblasts isolated from mice homozygous null for NCAM and therefore lacking the molecule, when placed in differentiation medium, consistently fused to form contractile myotubes with kinetics equivalent to wild-type primary myoblasts. It is concluded that the increase in expression of NCAM, although typically observed during myogenesis, is not essential to myoblast fusion to form myotubes (Charlton, 2000).

Two questions arise from these findings. (1) Why in myogenesis should there be an increase in NCAM expression which is conserved across species? (2) What is the significance of studies in which overexpression of NCAM enhances the rate of fusion in cultured myoblasts and antibodies to NCAM decrease that rate? It is thought that NCAM may be important both in the higher order organization of muscle tissue and in the formation of neuromuscular contacts leading to synaptogenesis. In development, a high level of NCAM is transiently expressed in embryonic muscle and then lost in normal adult muscle except at neuromuscular junctions. The expression of a highly polysialylated form of NCAM has been temporally and ultrastructurally correlated with the separation of chick secondary muscle fibers from the primary fibers along which they form, and in neuromuscular development, the overall pattern of NCAM expression on muscle fibers in vivo parallels their time course of innervation. Anti-NCAM antibodies have been reported to block the formation of contacts between neuron processes and myotubes in vitro and have been reported to block reinnervation of synaptic sites in frog muscle basal lamina after surgical denervation. Polysialylated NCAM has also been implicated in the branching and bundling patterns of axon outgrowth and in the innervation of developing and regenerating chick muscle. In agreement with this putative role in muscle fiber innervation, NCAM, especially polysialylated NCAM expression, is also increased on young regenerating fibers of mice. However, in NCAM-null mice, although endplates of neuromuscular junctions are reduced, no deficits in motor function or ability to reinnervate were reported, suggesting that effects upon these developmental processes must be minor or compensated for in vivo by other molecules. Thus, despite the abundance of reports to the contrary, it is concluded that an essential function for NCAM in any aspect of neuromuscular development has yet to be established (Charlton, 2000 and references therein).

CAMs and axon guidance

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

Fasciclin 2: Biological Overview | Regulation | Developmental Biology | Effects of Mutation | References

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.