Posttranslational membrane attachment and dynamic fatty acylation of a neuronal growth cone protein, GAP-43

Growth cones, the motile apparatus at the ends of elongating axons, are sites of extensive and dynamic membrane-cytoskeletal interaction and insertion of new membrane into the growing axon. One of the most abundant proteins in growth cone membranes is a protein designated GAP-43, whose synthesis increases dramatically in most neurons during periods of axon development or regeneration. We have begun to explore the role of GAP-43 in growth cone membrane functions by asking how the protein interacts with those membranes. Membrane-washing experiments indicate that mature GAP-43 is tightly bound to growth cone membranes, and partitioning of Triton X-114-solubilized GAP-43 between detergent-enriched and detergent-depleted phases indicates considerable hydrophobicity. The hydrophobic behavior of the protein is modulated by divalent cations, particularly zinc and calcium. In vivo labeling of GAP-43 in neonatal rat brain with [35S]methionine shows that GAP-43 is initially synthesized as a soluble protein that becomes attached to membranes posttranslationally. In tissue culture, both rat cerebral cortex cells and neuron-like PC12 cells actively incorporate [3H]palmitic acid into GAP-43. Isolated growth cones detached from their cell bodies also incorporate labeled fatty acid into GAP-43, suggesting active turnover of the fatty acid moieties on the mature protein. Hydrolysis of ester-like bonds with neutral hydroxylamine removes the bound fatty acid and exposes new thiol groups on GAP-43, suggesting that fatty acid is attached to the protein's only two cysteine residues, located in a short hydrophobic domain at the amino terminus. Modulation of the protein's hydrophobic behavior by divalent cations suggests that other domains, containing large numbers of negatively charged residues, might also contribute to GAP-43-membrane interactions. Our observations suggest a dynamic and reversible interaction of GAP-43 with growth cone membranes.     

Role of GAP-43 in sequestering phosphatidylinositol 4,5-bisphosphate to Raft bilayers

The lipid phosphatidylinositol 4,5-bisphosphate (PIP₂) is critical for a number of physiological functions, and its presence in membrane microdomains (rafts) appears to be important for several of these spatially localized events. However, lipids like PIP₂ that contain polyunsaturated hydrocarbon chains are usually excluded from rafts, which are enriched in phospholipids (such as sphingomyelin) containing saturated or monounsaturated chains. Here we tested a mechanism by which multivalent PIP₂ molecules could be transferred into rafts through electrostatic interactions with polybasic cytoplasmic proteins, such as GAP-43, which bind to rafts via their acylated N-termini. We analyzed the interactions between lipid membranes containing raft microdomains and a peptide (GAP-43P) containing the linked N-terminus and the basic effector domain of GAP-43. In the absence or presence of nonacylated GAP-43P, PIP₂ was found primarily in detergent-soluble membranes thought to correspond to nonraft microdomains. However, when GAP-43P was acylated by palmitoyl coenzyme A, both the peptide and PIP₂ were greatly enriched in detergent-resistant membranes that correspond to rafts; acylation of GAP-43P changed the free energy of transfer of PIP₂ from detergent-soluble membranes to detergent-resistant membranes by −1.3 kcal/mol. Confocal microscopy of intact giant unilamellar vesicles verified that in the absence of GAP-43P PIP₂ was in nonraft microdomains, whereas acylated GAP-43P laterally sequestered PIP₂ into rafts. These data indicate that sequestration of PIP₂ to raft microdomains could involve interactions with acylated basic proteins such as GAP-43.     

Mutagenesis of ser41 to ala inhibits the association of GAP-43 with the membrane skeleton of GAP-43-deficient PC12B cells: Effects on cell adhesion and the composition of neurite cytoskeleton and membrane

To investigate the molecular basis for GAP-43 function in axon outgrowth, we produced a mutant, GAP-43 (Ala₄₁), whose interaction with calmodulin in vitro was unaffected by increasing Ca²⁺ concentrations, and stably transfected it into GAP-43-deficient PC12B cells. Several lines that expressed wild-type or mutant protein at levels that resembled endogenous GAP-43 expression in PC12 controls were subcloned and characterized. GAP-43 (Ala₄₁) was significantly more extractable with Nonidet P-40 and less tightly associated with the membrane skeleton than the wild-type protein. Furthermore, GAP-43 (Ala₄₁) expression by PC12B cells profoundly affected their phenotype: First, observation of living cells using video-enhanced microscopy revealed irregular plasma membranes with numerous blebs and protrusions and neurites that appeared thin and varicose. Second, both the cells' ability to remain attached to laminin substrates and the amount of α1β1 integrin expressed on the cell surface was significantly decreased. Finally, peripherin transport, which is abnormal in PC12B cells, could be rescued by transfection of wild-type GAP-43 but not the GAP-43 (Ala₄₁) mutant. The phenotypic abnormalities resemble other cell types in which membrane skeleton/plasma membrane interactions have been functionally decoupled, and our results are consistent with the notion that these interactions may be abnormal in GAP-43 (Ala₄₁)-expressing PC12B cells, either as a direct consequence of the mutation or arising secondarily to the altered availability of calmodulin in the growing neurite. © 1996 John Wiley & Sons, Inc.     

An intracellular guanine nucleotide release protein for G0. GAP-43 stimulates isolated alpha subunits by a novel mechanism.

G protein-coupled membrane receptors activate G proteins by enhancing guanine nucleotide exchange. G0 is a major component of the growing regions (growth cones) of neurons. GAP-43 is a neuronal protein associated with the cytosolic face of the growth cone plasma membrane and stimulates binding of guanosine 5'-3-O-(thio)triphosphate (GTP gamma S) to Go (Strittmatter, S. M., Valenzuela, D., Kennedy, T. E., Neer, E. J., and Fishman, M. C. (1990) Nature 344, 836-841). Here we have examined the mechanism by which GAP-43 affects G0. Like G protein-coupled receptors, GAP-43 enhances GDP release from G0, increases the initial rate of GTP gamma S binding, and increases the GTPase activity of Go, all without altering the intrinsic kappa cat for the GTPase. Unlike the case for receptors, however, the GAP-43 effect is not blocked by pertussis toxin, nor affected by the presence or absence of beta gamma or of phospholipids. There is specificity to the interaction, in that GAP-43 increases GTP gamma S binding to recombinant alpha o and alpha i1, but not to recombinant alpha s. Thus, GAP-43 is a guanine nucleotide release protein with a novel mechanism of action, potentially controlling membrane-associated G proteins from within the cell.     

Phosphorylation-site mutagenesis of the growth-associated protein GAP- 43 modulates its effects on cell spreading and morphology

The 43-kD growth-associated protein (GAP-43) is a major protein kinase C (PKC) substrate of axonal growth cones, developing nerve terminals, regenerating axons, and adult central nervous system areas associated with plasticity. It is a cytosolic protein associated with the cortical cytoskeleton and the plasmalemma. Membrane association of GAP-43 is mediated by palmitoylation at Cys3Cys4. In vitro and in vivo, phosphorylation by PKC exclusively involves Ser41 of mammalian GAP-43 (corresponding to Ser42 in the chick protein). To identify aspects of GAP-43 function, we analyzed the actions of wild-type, membrane- association, and phosphorylation-site mutants of GAP-43 in nonneuronal cell lines. The GAP-43 constructs were introduced in L6 and COS-7 cells by transient transfection. Like the endogenous protein in neurons and their growth cones, GAP-43 in nonneuronal cells associated with the cell periphery. GAP-43 accumulated in the pseudopods of spreading cells and appeared to interact with cortical actin-containing filaments. Spreading L6 cells expressing high levels of recombinant protein displayed a characteristic F-actin labeling pattern consisting of prominent radial arrays of peripheral actin filaments. GAP-43 had dramatic effects on local surface morphology. Characteristic features of GAP-43-expressing cells were irregular cell outlines with prominent and numerous filopodia. The effects of GAP-43 on cell morphology required association with the cell membrane, since GAP-43(Ala3Ala4), a mutant that failed to associate with the cell cortex, had no morphogenetic activity. Two GAP-43 phosphorylation mutants (Ser42 to Ala42 preventing and Ser42 to Asp42 mimicking phosphorylation by PKC) modulated the effects of GAP-43 in opposite ways. Cells expressing GAP- 43(Asp42) spread extensively and displayed large and irregular membranous extensions with little filopodia, whereas GAP-43(Ala42) produced small, poorly spreading cells with numerous short filopodia. Therefore, GAP-43 influences cell surface behavior and phosphorylation modulates its activity. The presence of GAP-43 in growing axons and developing nerve termini may affect the behavior of their actin- containing cortical cytoskeleton in a regulatable manner.     

Enhanced level of site-specific proteolysis of GAP-43 protein during early stages of brain development

GAP-43 protein of nerve terminals (B-50, F1, F57, pp46, neuromodulin) is thought to be one of key proteins involved in the control of outgrowth of neurites, release of neuromediators, synapse plasticity, etc. GAP-43 is usually considered as a whole protein. Along with the intact protein, nerve cells also contain two large native fragments of GAP-43 deprived of four or of about forty N-terminal amino acid residues (GAP-43-2 and GAP-43-3, respectively). The full-length GAP-43 is predominant in the mature brain. However, the ratio of the full-length protein and its fragments can vary under different physiological conditions. Changes in the GAP-43 proteins (the full-length protein and its fragments) were studied during embryonal and postnatal development of rat brain. The GAP-43 proteins were found to be expressed not later than on the 12-13th day of embryogenesis. Then their contents increased, and, until the 10th day after birth, GAP-43-3 dominated rather than the full-length protein. It is suggested that during this period the activity of a specific protease, which cleaves the N-terminal peptide of about 40 residues from the full-length GAP-43 molecule, is increased. The cleavage occurs in the region responsible for the interaction of GAP-43 with calmodulin. In the full-length molecule, this region is responsible also for the recognition of Ser41 residue by protein kinase C during phosphorylation. Another functionally important region that determines, in particular, the attachment of GAP-43 to the plasma membrane is cleaved from the main part of the molecule together with the N-terminal peptide. Thus, the specific fragmentation of GAP-43 that depends on developmental stage should be considered as a controlled structural rearrangement fundamentally affecting the functions of this protein.     

Arachidonic Acid, but Not Sodium Nitroprusside, Stimulates Presynaptic Protein Kinase C and Phosphorylation of GAP-43 in Rat Hippocampal Slices and Synaptosomes

Abstract: Activation of protein kinase C (PKC) and phosphorylation of its presynaptic substrate, the 43-kDa growth-associated protein GAP-43, may contribute to the maintenance of hippocampal long-term potentiation (LTP) by enhancing the probability of neurotransmitter release and/or modifying synaptic morphology. Induction of LTP in rat hippocampal slices by high-frequency stimulation of Schaffer collateral-CA1 synapses significantly increased the PKC-dependent phosphorylation of GAP-43, as assessed by quantitative immunoblotting with a monoclonal antibody that recognizes an epitope that is specifically phosphorylated by PKC. The stimulatory effect of high-frequency stimulation on levels of immunoreactive phosphorylated GAP-43 was not observed when 4-amino-5-phosphonovalerate (50 µM), an N-methyl-d -aspartate (NMDA) receptor antagonist, was bath-applied during the high-frequency stimulus. This observation supports the hypothesis that a retrograde messenger is produced postsynaptically following NMDA receptor activation and diffuses to the presynaptic terminal to activate PKC. Two retrograde messenger candidates—arachidonic acid and nitric oxide (sodium nitroprusside was used to generate nitric oxide)—were examined for their effects in hippocampal slices on PKC redistribution from cytosol to membrane as an indirect measure of enzyme activation and PKC-specific GAP-43 phosphorylation. Bath application of arachidonic acid, but not sodium nitroprusside, at concentrations that produce synaptic potentiation (100 µM and 1 mM, respectively) significantly increased translocation of PKC immunoreactivity from cytosol to membrane as well as levels of immunoreactive, phosphorylated GAP-43. The stimulatory effect of arachidonic acid on GAP-43 phosphorylation was also observed in hippocampal synaptosomes. These results indicate that arachidonic acid may contribute to LTP maintenance by activation of presynaptic PKC and phosphorylation of GAP-43 substrate. The data also suggest that nitric oxide does not activate this signal transduction system and, by inference, activates a distinct biochemical pathway.     

Mass spectrometric analysis of GAP-43/neuromodulin reveals the presence of a variety of fatty acylated species

GAP-43 (neuromodulin) is a protein kinase C substrate that is abundant in developing and regenerating neurons. Thioester-linked palmitoylation at two cysteines near the GAP-43 N terminus has been implicated in directing membrane binding. Here, we use mass spectrometry to examine the stoichiometry of palmitoylation and the molecular identity of the fatty acid(s) attached to GAP-43 in vivo. GAP-43 expressed in either PC12 or COS-1 cells was acetylated at the N-terminal methionine. Approximately 35% of the N-terminal GAP-43 peptides were also modified by palmitate and/or stearate on Cys residues. Interestingly, a variety of acylated species was detected, in which one of the Cys residues was acylated by either palmitate or stearate, or both Cys residues were acylated by palmitates or stearates or a combination of palmitate and stearate. Depalmitoylation of membrane-bound GAP-43 did not release the protein from the membrane, implying that additional forces function to maintain membrane binding. Indeed, mutation of four basic residues within the N-terminal domain of GAP-43 dramatically reduced membrane localization of GAP-43 without affecting palmitoylation. These data reveal the heterogeneous nature of S-acylation in vivo and illustrate the power of mass spectrometry for identification of key regulatory protein modifications.     

Palmitoylation alters protein activity: blockade of G(o) stimulation by GAP-43.

The addition of palmitate to cysteine residues enhances the hydrophobicity of proteins, and consequently their membrane association. Here we have investigated whether this type of fatty acylation also regulates protein-protein interactions. GAP-43 is a neuronal protein that increases guanine nucleotide exchange by heterotrimeric G proteins. Two cysteine residues near the N-terminus of GAP-43 are subject to palmitoylation, and are necessary for membrane binding as well as for G(o) activation. N-terminal peptides, which include these cysteines, stimulate G(o). Monopalmitoylation reduces, and dipalmitoylation abolishes the activity of the peptides. The activity of GAP-43 protein purified from brain also is reversibly blocked by palmitoylation. This suggests that palmitoylation controls a cycle of GAP-43 between an acylated, membrane-bound reservoir of inactive GAP-43, and a depalmitoylated, active pool of protein.     

Protective effect of the membrane skeleton on the immunologic reactivity of the human red cell Rho(D) antigen

It has recently been shown that the 30,000 m.w. Rho(D) protein is associated with the membrane skeleton of the human red cell. We have studied the effects of the membrane skeleton on the immunoreactivity of the Rho(D) antigen present in Rho(D)+ membranes. Solubilization of the membranes with the Triton X-100 detergent and centrifugation of the extracts showed that more than 90% of the immunoreactive Rho(D) antigen sedimented with the membrane skeleton structures. The skeleton-bound Rho(D) antigen could be solubilized by disruption of the skeleton in low ionic strength medium. The removal of the membrane skeleton structure before the solubilization of the membranes with detergent resulted in the inactivation of the majority of the Rho(D) antigen. The effect of the membrane skeleton on the stability of the Rho(D) antigen was additionally studied in detergent extracts prepared from native and skeleton-free membranes. The assay of the Rho(D) antigen activity in the extracts showed that the Rho(D) antigen was 100 times more sensitive to the detergent inactivation in skeleton-free membranes than in native membranes. These results indicate that the membrane skeleton is important for stabilizing the immunoreactive form of the Rho(D) protein on the red cell membrane.     

GAP-43 Augmentation of G Protein-Mediated Signal Transduction Is Regulated by Both Phosphorylation and Palmitoylation

Abstract: The neuronal protein GAP-43 is concentrated at the growth cone membrane, where it is thought to amplify the signal transduction process. As a model for its neuronal effects, GAP-43 protein injection into Xenopus laevis oocytes strongly augments the calcium-sensitive chloride current evoked by the G protein-coupled receptor stimulation. We have now examined a series of GAP-43 mutants in this system and determined those regions of GAP-43 required for this increase in current flux. As expected, palmitoylation inhibits signal amplification in oocytes by blocking G protein activation. Unexpectedly, a second domain of GAP-43 (residues 35–50) containing a protein kinase C phosphorylation site at residue 41 is also necessary for augmentation of G protein-coupled signals in oocytes. This region is not required for activation of isolated G_o but is necessary for GAP-43 binding to isolated calmodulin and to isolated protein kinase C. Substitution of Asp for Ser⁴¹ inactivates GAP-43 as a signal facilitator in oocytes. This mutation blocks GAP-43 binding to both protein kinase C and calmodulin. Thus, GAP-43 regulates an oocyte signaling cascade via coordinated, simultaneous G protein activation and interaction with either calmodulin or protein kinase C.     

GAP-43 as a plasticity protein in neuronal form and repair.

Neurons exhibit a remarkable plasticity of form, both during neural development and during the subsequent remodelling of synaptic connectivity. Here we review work on GAP-43 and G0, and focus upon the thesis that their interaction may endow neurons with such plasticity. We also present new data on the role of G proteins in neurite growth, and on the interaction of GAP-43 and actin. GAP-43 is a protein induced during periods of axonal extension and highly enriched on the inner surface of the growth cone membrane. Its membrane localization is primarily due to a short amino terminal sequence which is subject to palmitoylation. Binding to actin filaments may also assist in restricting the protein to specific cellular domains. Consistent with its role as a "plasticity protein," there is evidence that GAP-43 can directly alter cell shape and neurite extension, and several theses have been advanced for how it might do so. Two other prominent components of the growth cone membrane are the alpha and beta subunits of G0. GAP-43 regulates their guanine nucleotide exchange, which is an unusual role for an intracellular protein. We speculate that GAP-43 may adjust the "set point" of responsiveness for G0 stimulation by receptors, thereby altering the neuronal propensity to growth, without actually causing growth. To begin to address how G protein activity affects axon growth, we have developed a means to introduce guanine nucleotide analogs into sympathetic neurons. Stimulation of G proteins with GTP-gamma-S retards axon growth, whereas GDP-beta-S enhances it. This is compatible with G protein registration of inhibitory signals.     

Identification of two proteins that bind to a pyrimidine-rich sequence in the 3'-untranslated region of GAP-43 mRNA.

GAP-43 is a membrane phosphoprotein that is important for the development and plasticity of neural connections. In undifferentiated PC12 pheochromocytoma cells, GAP-43 mRNA degrades rapidly ( t = 5 h), but becomes stable when cells are treated with nerve growth factor. To identify trans- acting factors that may influence mRNA stability, we combined column chromatography and gel mobility shift assays to isolate GAP-43 mRNA binding proteins from neonatal bovine brain tissue. This resulted in the isolation of two proteins that bind specifically and competitively to a pyrimidine-rich sequence in the 3'-untranslated region of GAP-43 mRNA. Partial amino acid sequencing revealed that one of the RNA binding proteins coincides with FBP (far upstream element binding protein), previously characterized as a protein that resembles hnRNP K and which binds to a single-stranded, pyrimidine-rich DNA sequence upstream of the c -myc gene to activate its expression. The other binding protein shares sequence homology with PTB, a polypyrimidine tract binding protein implicated in RNA splicing and regulation of translation initiation. The two proteins bind to a 26 nt pyrimidine-rich sequence lying 300 nt downstream of the end of the coding region, in an area shown by others to confer instability on a reporter mRNA in transient transfection assays. We therefore propose that FBP and the PTB-like protein may compete for binding at the same site to influence the stability of GAP-43 mRNA.     

GAP-43 as a plasticity protein in neuronal form and repair

Neurons exhibit a remarkable plasticity of form, both during neural development and during the subsequent remodelling of synaptic connectivity. Here we review work on GAP-43 and G₀, and focus upon the thesis that their interaction may endow neurons with such plasticity. We also present new data on the role of G proteins in neurite growth, and on the interaction of GAP-43 and actin. GAP-43 is a protein induced during periods of axonal extension and highly enriched on the inner surface of the growth cone membrane. Its membrane localization is primarily due to a short amino terminal sequence which is subject to palmitoylation. Binding to actin filaments may also assist in restricting the protein to specific cellular domains. Consistent with its role as a ?plasticity protein,”? there is evidence that GAP-43 can directly alter cell shape and neurite extension, and several theses have been advanced for how it might do so. Two other prominent components of the growth cone membrane are the α and β subunits of G₀. GAP-43 regulates their guanine nucleotide exchange, which is an unusual role for an intracellular protein. We speculate that GAP-43 may adjust the ?set point”? of responsiveness for G₀ stimulation by receptors, thereby altering the neuronal propensity to growth, without actually causing growth. To begin to address how G protein activity affects axon growth, we have developed a means to introduce guanine nucleotide analogs into sympathetic neurons. Stimulation of G proteins with GTP-γ-S retards axon growth, whereas GDP-β-S enhances it. This is compatible with G protein registration of inhibitory signals. © 1992 John Wiley & Sons, Inc.     

M-calpain-mediated cleavage of GAP-43 near Ser41 is negatively regulated by protein kinase C, calmodulin and calpain-inhibiting fragment GAP-43-3

Neuronal protein GAP-43 performs multiple functions in axon guidance, synaptic plasticity and regulation of neuronal death and survival. However, the molecular mechanisms of its action in these processes are poorly understood. We have shown that in axon terminals GAP-43 is a substrate for calcium-activated cysteine protease m-calpain, which participates in repulsion of axonal growth cones and induction of neuronal death. In pre-synaptic terminals in vivo, in synaptosomes, and in vitro, m-calpain cleaved GAP-43 in a small region near Ser41, on either side of this residue. In contrast, micro-calpain cleaved GAP-43 in vitro at several other sites, besides Ser41. Phosphorylation of Ser41 by protein kinase C or GAP-43 binding to calmodulin strongly suppressed GAP-43 proteolysis by m-calpain. A GAP-43 fragment, lacking about forty N-terminal residues (named GAP-43-3), was produced by m-calpain-mediated cleavage of GAP-43 and inhibited m-calpain, but not micro-calpain. This fragment prevented complete cleavage of intact GAP-43 by m-calpain as a negative feedback. GAP-43-3 also blocked m-calpain activity against casein, a model calpain substrate. This implies that GAP-43-3, which is present in axon terminals in high amount, can play important role in regulation of m-calpain activity in neurons. We suggest that GAP-43-3 and another (N-terminal) GAP-43 fragment produced by m-calpain participate in modulation of neuronal response to repulsive and apoptotic signals.     

Signaling mechanisms of daidzein-induced axonal outgrowth in hippocampal neurons

We aim to study the mechanisms underlying the neurotrophic effect of daidzein (Dz) in hippocampal neurons. Dz-enhanced axonal outgrowths manifested growth cone formation and increased immunostaining intensity of growth-associated protein 43 (GAP-43) in growth cones. Consistent with this, Dz increased GAP-43 phosphorylation and its membrane translocation without affecting total GAP-43 levels. In the presence of Dz, significant increase in the immunoreactivity for estrogen receptor (ER) β, but not ERα, was observed on the membrane of cell bodies and growing axons. Dz also induced the activation of protein kinase C α (PKCα), which was inhibited by the ICI182,780 pretreatment. Similarly, Dz-promoted axonal elongation was blocked by ICI182,780 and Gö6976. Moreover, Dz-stimulated activation of GAP-43 was specifically abolished by Gö6976, suggesting PKCα being the upstream effector of GAP-43. Taken together, our data suggest that Dz triggers an ERβ/PKCα/GAP-43 signaling cascade to promote axonal outgrowths in cultured hippocampal neurons.     

A novel 87,000-Mr protein associated with acetylcholine receptors in Torpedo electric organ and vertebrate skeletal muscle

To identify proteins associated with nicotinic postsynaptic membranes, mAbs have been prepared to proteins extracted by alkaline pH or lithium diiodosalicylate from acetylcholine receptor-rich (AChR) membranes of Torpedo electric organ. Antibodies were obtained that recognized two novel proteins of 87,000 Mr and a 210,000:220,000 doublet as well as previously described proteins of 43,000 Mr, 58,000 (51,000 in our gel system), 270,000, and 37,000 (calelectrin). The 87-kD protein copurified with acetylcholine receptors and with 43- and 51-kD proteins during equilibrium centrifugation on continuous sucrose gradients, whereas a large fraction of the 210/220-kD protein was separated from AChRs. The 87-kD protein remained associated with receptors and 43-kD protein during velocity sedimentation through shallow sucrose gradients, a procedure that separated a significant amount of 51-kD protein from AChRs. The 87- and 270-kD proteins were cleaved by Ca++- activated proteases present in crude preparations and also in highly purified postsynaptic membranes. With the exception of anti-37-kD antibodies, some of the monoclonals raised against Torpedo proteins also recognized determinants in frozen sections of chick and/or rat skeletal muscle fibers and in permeabilized chick myotubes grown in vitro. Anti-87-kD sites were concentrated at chick and rat endplates, but the antibodies also recognized determinants present at lower site density in the extrasynaptic membrane. Anti-210:220-kD labeled chick endplates, but studies of neuron-myotube cocultures showed that this antigen was located on neurites rather than the postsynaptic membrane. As reported in other species, 43-kD determinants were restricted to chick endplates and anti-51-kD and anti-270-kD labeled extrasynaptic as well as synaptic membranes. None of the cross reacting antibodies recognized determinants on intact (unpermeabilized) myotubes, so the antigens must be located on the cytoplasmic aspect of the surface membrane. The role that each intracellular determinant plays in AChR immobilization at developing and mature endplates remains to be investigated.