Michihiro Igarashi

Localization and targeting of SCG10 to the trans-Golgi apparatus and growth cone vesicles

SCG10 is a membrane‐associated, microtubule‐destabilizing protein of neuronal growth cones. Using immunoelectron microscopy, we show that in the developing cortex of mice, SCG10 is specifically localized to the trans face Golgi complex and apparently associated with vesicular structures in putative growth cones. Consistent with this, subcellular fractionation of rat forebrain extracts demonstrates that the protein is enriched in the fractions containing the Golgi apparatus and growth cone particles. In isolated growth cone particles, SCG10 was found to be particularly concentrated in the growth cone vesicle fraction. To evaluate the molecular determinants of the specific targeting of SCG10 to growth cones, we have transfected PC12 cells and primary neurons in culture with mutant and fusion cDNA constructs. Deletion of the amino‐terminal domain or mutations within this domain that prevented palmitoylation at cysteines 22 and 24 abolished Golgi localization as well as growth cone targeting, suggesting that palmitoylation of the amino‐terminal domain is a necessary signal for Golgi sorting and possibly transport of SCG10 to growth cones. Fusion proteins consisting of the amino‐terminal domain of SCG10 and the cytosolic proteins stathmin or glutathione‐S‐transferase colocalized with a Golgi marker, α‐mannosidase II, and accumulated in growth cones of both axons and dendrites. These results reveal a novel axonal/dendritic growth cone targeting sequence that involves palmitoylation.

Effects of inhalational anesthetics on biochemical events in growing neuronal tips.

BackgroundThe Influence of general anesthetics on developing organs has been a source of concern for many years. The central nervous system, which is developing rapidly at the time of birth, is of special Interest in this regard. In this study, the biochemical characteristics of developing neural tips (growth cones) were examined after exposure to anesthetics to elucidate the molecular mechanism by which long-lasting alterations in the nervous system, including neuroter-atogenicity, as previously described, were evoked. MethodsNeonatal rats were exposed to an atmosphere containing i inhalational anesthetics (1% halothane or 75% nitrous oxide) or a control atmosphere (25% O2 and 75% N2) for 6 h at postnatal day 1. After this exposure, growth cone particles were isolated from the forebrain using a recently devised cell fractionation method at postnatal days 2, 3, 4, and 5. Protein composition, phosphoprotein patterns, and protein kinase C (PKC) activities of the isolated growth cones were compared between each group exposed to anesthetics and the control group. The dose-response relationship of the action of anesthetics on PKC activity was also examined (at 0.5 and 0.75% halothane and 25 and 50% N2O). ResultsThe increase in body weight and brain wet weight were not significantly affected by exposure to either anesthetic. No apparent influence on protein composition was observed by sodiumdodecylsulfate polyacrylamide gel electrophoresis (SDS-PAGE). However, calcium-dependent protein phosphorylation of the 46 kDa protein and of the 80 kDa protein, which is reported to be mediated by PKC, were significantly reduced after exposure to the anesthetics. A direct assay of PKC activity in growth cone particles indicated that PKC activity in the growth cone was 70.6 ± 9.6% of the control value at 24 h after exposure to 1% halothane, and 63.2 ± 4.9% after exposure to 75% nitrous oxide. Exposure to 0.75% halothane or 50% nitrous oxide had a similar, but lesser, effect on this parameter. In contrast, exposure to 0.5% halothane or 25% nitrous oxide evoked no apparent effect. Thus, the PKC activity in growth cone particles, which is thought to play an important role in signal transduction in the developing brain, was shown to be affected by exposure to Inhalational anesthetics over a range of concentrations. ConclusionsConsidering the crucial role of growth cones in the establishment of the neuronal network, the interruption of signal transduction in the growth cone at a time that is critical in directing the neurite extension may evoke a long-lasting alteration in the neural network. Therefore, the effect of inhalational anesthetics on the growth cone enzyme observed in this study may have a major role in the mechanism that Induces morphologic or behavioral neuroabnormalities in later life.

Stimulation of L‐type Ca2+ channel in growth cones activates two independent signaling pathways

Although growth cones respond to various modulators of neurite outgrowth, such as neurotrophins, neurotransmitters, and cell adhesion molecules, the signal‐transducing mechanisms for these modulators in growth cones are unclear. Since recent studies have suggested that the signals of these modulators are mediated by Ca2+ influx through L‐type voltage‐sensitive Ca2+ channels (VSCCs) in the growth cone, we examined L‐type VSCC‐dependent signaling pathways, using isolated growth cones (IGCs) from developing rat forebrains. Binding assays revealed that L‐type VSCC is enriched in growth cone membrane and gradually decreased in amount developmentally, while N‐type VSCC has the opposite tendency. In intact IGCs, Bay K 8644 (BK, an L‐type agonist) induced much more rapid elevation of [Ca2+]i than that in adult synaptosomes. Ca2+‐dependent phosphorylation of GAP‐43 and MARCKS protein by protein kinase C (PKC) was enhanced in the IGC by BK, resulting in the release of these proteins from the membrane, which is consistent with our recent report. In addition, the Ca2+‐dependent degradation of brain spectrin (fodrin) by calpain was also enhanced by BK or GABA, consequently inducing the release of α‐actinin from the membrane skeleton of the growth cones. The activities of PKC and calpain were not inhibited by inhibitors of the other, indicating that these reactions occur independently. Our results suggest that Ca2+ influx through L‐type VSCCs activates two distinct signaling branches, probably in the different domains of the growth cone, i.e., Ca2+‐dependent phosphorylation of GAP‐43 and MARCKS protein, and Ca2+‐dependent degradation of brain spectrin and the release of α‐actinin by calpain. J. Neurosci. Res. 51:682–696, 1998.

Muscarinic acetylcholine receptors are expressed and enriched in growth cone membranes isolated from fetal and neonatal rat forebrain: Pharmacological demonstration and characterization

Nerve growth cones, the motile tips of growing neurites, are closely related to the exact pathway finding, and their roles for synaptogenesis have been proposed to be modified by some neurotransmitters. In the present study, to clarify the expression and the ontogeny of muscarinic acetylcholine receptors in growth cones, growth cone membranes from fetal and neonatal rat forebrain were isolated, and muscarinic receptors in growth cone membrane were pharmacologically characterized, by using the [3H]quinuclidinyl benzilate as a labeled ligand. The specific binding sites for [3H]quinuclidinyl benzilate had already been detected in growth cone membrane on embryonic day (E)17 (Bmax = 557 fmol/mg protein: KD = 19.7 pM) and gradually increased in amount without significant changes in the KD values from E17 to postnatal day (P)5. [3H]Quinuclidinyl benzilate binding sites in growth cone membrane were several times higher than that in the P2-fraction-derived membranes, and in perinuclear membranes. Competitive inhibition studies showed that the proportion of high-affinity sites for pirenzepine (M1-subtype) to total [3H]quinuclidinyl benzilate binding sites in growth cone membrane was significantly lower than that in adult synaptic plasma membranes. In contrast, the proportion of high-affinity sites for AF-DX 116 (M2-subtype) was significantly higher than that in adult synaptic plasma membranes (E17 growth cone membrane: M1, 29.5%; M2, 56.9%; adult synaptic plasma membrane: M1, 63.6%, M2, 5.9%). Electron micrographic examination revealed that there were no significant morphological differences among growth cone particle fractions at the developmental stages which we examined, and that mature synaptic elements did not contaminate the growth cone particle fractions. Biochemical examination by electrophoresis and the phosphorylation study of the growth cone particle fractions showed that the protein composition and the phosphoprotein pattern did not change markedly during these stages. Our results suggest that muscarinic receptors were expressed and more concentrated in growth cone membrane than in other membrane portions from perinatal rat forebrain, and that they may play some role in the axonal guidance in growth cone via receptor subtype-specific signal transduction mechanisms.

Subtypes of protein kinase C in isolated nerve growth cones: Only type II is associated with the membrane skeleton from growth cones

The growth cone particle (GCP) fraction was isolated from fetal and neonatal rat brains and the distribution of protein kinase C subtypes in the fraction was examined by using subtype-specific antibodies. The main subtype in the GCP fraction from fetal forebrain was type II, and type III was also present, but not type I. The pattern was not altered from embryonic day 17 to postnatal day 5. The membrane skeleton subfraction from the GCP fraction contained type II, but far less amount of type III. Our results suggest that type II and type III may be closely related to the functions of growth cones but that they appear to be associated with distinct signal transduction processes.

Minimal residues in linker domain of syntaxin 1A required for binding affinity to Ca2+/calmodulin-dependent protein kinase II.

The linker domain is important for the conformational change syntaxin 1A, which enables it to act as a SNARE for exocytosis. We found that when applied exogenously, the linker domain is a potent inhibitor of exocytosis through inhibiting interaction between autophosphorylated CaMKII and endogenous syntaxin 1A (Ohyama et al. [ 2002 ] J. Neurosci. 22:3342–3351). To identify the simplest and the most potent inhibitor for exocytosis, we further characterized the linker domain and determined the minimal number of residues required for CaMKII binding. The minimal length of the CaMKII‐binding site was 145–172 residues and a loss of G172 considerably weakened affinity for CaMKII. The basic amino acid clusters, R151 and K146, were indispensable for binding, whereas R148 was not. A comparison of the CaMKII‐binding in several syntaxin isoforms revealed that the substitution of S162 attenuated CaMKII‐binding activity. These results suggest that S162 is an important residue as well as the basic amino acid cluster within region 145–172 of the linker domain.

Characteristics of gangliosides including O-acetylated species in growth cone membranes at several developmental stages in rat forebrain.

Growth cones, the motile tips of extending neuronal processes, are involved in accurate synaptogenesis. To study the developmental changes in ganglioside composition including O-acetylated gangliosides in growth cones, we analyzed the gangliosides in growth cone membranes (GCM) prepared from rat forebrains at different developmental stages. At several stages, GCM contained significantly larger amounts of gangliosides than the other membrane subfractions. The ganglioside content of GCM increased in amount with development. Moreover, in GCM, the relative amount of GD3 gradually decreased, and that of GD1a dramatically increased. There were significant differences in the composition of ganglioside species between GCM and the perinuclear plasma membrane subfraction (NM); most importantly, GCM had a higher ratio of GD1a to GM3 plus GD3 than NM. There were three different O-acetylated gangliosides in GCM: O-acetyl-GD3, O-acetyl-GT1b, and O-acetyl-GQ1b. The molar ratio of O-acetyl-GD3 decreased in GCM at later stages (5% of the total gangliosides at embryonic day 17, to 1% at postnatal day 5). However, those of the other two O-acetylated gangliosides were almost constant (1-2% of the total). Our results show that there are significant differences in ganglioside content and composition between the membrane subfraction of growth cones and the perinuclear portion. This suggests that several species of gangliosides, including O-acetyl-GD3, play a role in growth cone function.

Biochemical characterization of nerve growth cones isolated from both fetal and neonatal rat forebrains: the growth cone particle fraction mainly consists of axonal growth cones in both stages

Nerve growth cones are responsible for the exact pathway finding, and for the establishment of neurocytoarchitecture. To elucidate the developmental changes of biochemical characteristics of nerve growth cones, growth cone particle (GCP) fractions were isolated biochemically from embryonal day 17 (E17) rat forebrain and from postnatal day 5 (P5). There were no significant differences in protein phosphorylation pattern in a Ca(2+)-dependent manner between E17-GCP fraction and that of P5. As for the membrane lipid composition, molar ratios of cholesterol to total phospholipids were well conserved during these ages. The immunoreactivity to anti-synaptophysin monoclonal antibody as a marker of mature synaptic elements could not be detected either in E17-GCP or P5-GCP fractions. To exclude the possibility of the contamination of dendritic elements, RNA contents and immunoreactivity to anti-high molecular weight microtubule-associated protein 2 (MAP2) monoclonal antibody were examined. RNA contents of the GCP fractions were extremely low compared to those of other subcellular fractions both in E17 and P5. No immunoreactivities to anti-MAP2 antibody were observed in either GCP fraction. Our results suggest that the GCP fractions, isolated from forebrains of E17 to P5 rat, are free from the contamination of the synaptic elements, and that the GCP fractions are mainly composed of axonal growth cones.

The mechanism of the neurotransmitter release in growth cones

The growth cone is considered the precursor of the presynaptic terminal. To elucidate the minimal molecular machinery required for exocytosis, we examined the characteristics of α‐latrotoxin‐induced exocytosis in growth cones. In isolated growth cones (IGC), neurotransmitters were released in a SNARE‐dependent manner, but rab3A cycling was blocked. By supplying rabphilin, a rab3A acceptor found in low levels in IGC, the IGC obtained as high an exocytotic efficiency as adult synaptosomes, and the complete GDP‐GTP conversion of rab3A occurred on growth cone vesicles (GCV). GCVs bound SNAREs but not NSF or α‐SNAP; whereas in the rabphilin‐supplied IGC, GCVs recruited both NSF and α‐SNAP, to form the SNARE‐NSF‐SNAP complex. These results suggest that rab3A cycling is dependent upon the accumulation of rabphilin and is completed later than the SNARE mechanism, and that rabphilin is involved in determining the efficiency of exocytosis by modifying the SNARE mechanism. J. Neurosci. Res. 60:743–753, 2000.

Actin-binding proteins in the growth cone particles (GCP) from fetal rat brain: a 44 kDa actin-binding protein is enriched in the fetal GCP fraction

Neuronal growth cones, the motile tips of growing neurites, are thought to play a significant role in nerve growth. To study the role of actin in their motility, we examined actin-binding proteins in growth cone particles (GCP) isolated from fetal rat brain, using a blot-overlay method with biotinylated actin. Among the more than ten species of actin-binding proteins in the GCP, a 44 kDa protein was found specifically in growth cones and was enriched in the cytoskeletal and the membrane skeletal subfractions from the GCP. This protein binds to actin in a Ca(2+)- and Mg(2+)-dependent manner, and ATP enhances its binding to actin. The protein was predominantly present in the fetal GCP, but it is expressed at a much lower level in the neonatal GCP and not detected in adult synaptosomes. The protein also bound to a deoxyribonuclease I column and was eluted by EGTA-containing buffer. The 44 kDa protein appears to be a novel actin-binding protein, since none of the known actin-binding proteins exhibit this combination of properties. Our results suggest that the protein may be involved with the early stages of neurite extension.