Phosphate availability regulates root system architecture in Arabidopsis.

Plant root systems are highly plastic in their development and can adapt their architecture in response to the prevailing environmental conditions. One important parameter is the availability of phosphate, which is highly immobile in soil such that the arrangement of roots within the soil will profoundly affect the ability of the plant to acquire this essential nutrient. Consistent with this, the availability of phosphate was found to have a marked effect on the root system architecture of Arabidopsis. Low phosphate availability favored lateral root growth over primary root growth, through increased lateral root density and length, and reduced primary root growth mediated by reduced cell elongation. The ability of the root system to respond to phosphate availability was found to be independent of sucrose supply and auxin signaling. In contrast, shoot phosphate status was found to influence the root system architecture response to phosphate availability.

Botulinum neurotoxin A blocks synaptic vesicle exocytosis but not endocytosis at the nerve terminal.

The supply of synaptic vesicles in the nerve terminal is maintained by a temporally linked balance of exo- and endocytosis. Tetanus and botulinum neurotoxins block neurotransmitter release by the enzymatic cleavage of proteins identified as critical for synaptic vesicle exocytosis. We show here that botulinum neurotoxin A is unique in that the toxin-induced block in exocytosis does not arrest vesicle membrane endocytosis. In the murine spinal cord, cell cultures exposed to botulinum neurotoxin A, neither K(+)-evoked neurotransmitter release nor synaptic currents can be detected, twice the ordinary number of synaptic vesicles are docked at the synaptic active zone, and its protein substrate is cleaved, which is similar to observations with tetanus and other botulinal neurotoxins. In marked contrast, K(+) depolarization, in the presence of Ca(2+), triggers the endocytosis of the vesicle membrane in botulinum neurotoxin A-blocked cultures as evidenced by FM1-43 staining of synaptic terminals and uptake of HRP into synaptic vesicles. These experiments are the first demonstration that botulinum neurotoxin A uncouples vesicle exo- from endocytosis, and provide evidence that Ca(2+) is required for synaptic vesicle membrane retrieval.

Neuronal sensitivity to tetanus toxin requires gangliosides.

Tetanus toxin produces spastic paralysis in situ by blocking inhibitory neurotransmitter release in the spinal cord. Although di- and trisialogangliosides bind tetanus toxin, their role as productive toxin receptors remains unclear. We examined toxin binding and action in spinal cord cell cultures grown in the presence of fumonisin B(1), an inhibitor of ganglioside synthesis. Mouse spinal cord neurons grown for 3 weeks in culture in 20 microM fumonisin B(1) develop dendrites, axons, and synaptic terminals similar to untreated neurons, even though thin layer chromatography shows a greater than 90% inhibition of ganglioside synthesis. Absence of tetanus and cholera toxin binding by toxin-horseradish peroxidase conjugates or immunofluorescence further indicates loss of mono- and polysialogangliosides. In contrast to control cultures, tetanus toxin added to fumonisin B(1)-treated cultures does not block potassium-stimulated glycine release, inhibit activity-dependent uptake of FM1-43, or abolish immunoreactivity for vesicle-associated membrane protein, the toxin substrate. Supplementing fumonisin B(1)-treated cultures with mixed brain gangliosides completely restores the ability of tetanus toxin to bind to the neuronal surface and to block neurotransmitter release. These data demonstrate that fumonisin B(1) protects against toxin-induced synaptic blockade and that gangliosides are a necessary component of the receptor mechanism for tetanus toxin.

Syntaxin and 25-kDa synaptosomal-associated protein: differential effects of botulinum neurotoxins C1 and A on neuronal survival.

The Clostridium botulinum neurotoxins (BoNTs) A and C1 cleave specific proteins required for neuroexocytosis. We demonstrated that, in intact neurons, BoNT A cleaves 25-kDa synaptosomal-associated protein (SNAP-25), and BoNT C1 cleaves both syntaxin and SNAP-25 (Williamson et al.: Mol Biol Cell 6:61a, 1995; J Biol Chem 271:7694-7699, 1996). Here, we compare the actions of BoNT A and BoNT C1 on mature and developing mouse spinal cord neurons in cell culture and demonstrate that BoNT C1 is severely neurotoxic. In mature cultures, synaptic terminals become enlarged shortly after BoNT C1 exposure, and, subsequently, axons, dendrites, and cell bodies degenerate. Electron microscopy confirms that early degenerative changes occur in synaptic terminals when the somatic cytoplasm appears normal. In newly plated cultures, few neurons survive exposure to BoNT C1. Whereas both BoNT A and BoNT C1 cleave SNAP-25, BoNT A has no adverse effect on neurite outgrowth, synaptogenesis, or neuron survival. This cytotoxicity is unique to BoNT C1, is specific to neurons, and is initiated at the synaptic terminal, suggesting either a novel role for syntaxin or additional actions of BoNT C1. The neurodegeneration induced by BoNT C1 may be significant in terms of its efficacy for the clinical treatment of dystonia and spasticity.

Clostridial neurotoxins and substrate proteolysis in intact neurons: botulinum neurotoxin C acts on synaptosomal-associated protein of 25 kDa.

Clostridial neurotoxins are zinc endopeptidases that block neurotransmission and have been shown to cleave, in vitro, specific proteins involved in synaptic vesicle docking and/or fusion. We have used immunohistochemistry and immunoblotting to demonstrate alterations in toxin substrates in intact neurons under conditions of toxin-induced blockade of neurotransmitter release. Vesicle-associated membrane protein, which colocalizes with synaptophysin, is not detectable in tetanus toxin-blocked cultures. Syntaxin, also concentrated in synaptic sites, is cleaved by botulinum neurotoxin C. Similarly, the carboxyl terminus of the synaptosomal-associated protein of 25 kDa (SNAP-25) is not detectable in botulinum neurotoxin A-treated cultures. Unexpectedly, tetanus toxin exposure causes an increase in SNAP-25 immunofluorescence, reflecting increased accessibility of antibodies to antigenic sites rather than increased expression of the protein. Furthermore, botulinum neurotoxin C causes a marked loss of the carboxyl terminus of SNAP-25 when the toxin is added to living cultures, whereas it has no action on SNAP-25 in vitro preparations. This study is the first to demonstrate in functioning neurons that the physiologic response to these toxins is correlated with the proteolysis of their respective substrates. Furthermore, the data demonstrate that botulinum neurotoxin C, in addition to cleaving syntaxin, exerts a secondary effect on SNAP-25.

Bafilomycin A1 inhibits the action of tetanus toxin in spinal cord neurons in cell culture.

Tetanus toxin (TeNT) is one of the clostridial neurotoxins that act intracellularly to block neurotransmitter release. However, neither the route of entry nor the mechanism by which these toxins gain access to the neuronal cytoplasm has been established definitively. In murine spinal cord cell cultures, release of the neurotransmitter glycine is particularly sensitive to blockade by TeNT. To test whether TeNT enters neurons through acidic endosomes or is routed through the Golgi apparatus, toxin action on potassium-evoked glycine release was assayed in cultures pretreated with bafilomycin A1 (baf A1) or brefeldin A (BFA). baf A1, which inhibits the vacuolar-type H(+)-ATPase responsible for endosome acidification, diminishes the staining of acidic compartments and interferes with the action of TeNT in a dose-dependent manner. TeNT blockade of evoked glycine release is inhibited by 50 and 90% in cultures pretreated with 50 and 100 nM baf A1, respectively, compared with cultures treated with the inhibitor alone. The effects of baf A1 are fully reversible. In contrast, BFA, which disrupts Golgi function, has no effect on TeNT action. These findings provide evidence that TeNT enters the neuronal cytoplasm through baf A1-sensitive acidic compartments and that TeNT is not trafficked through the Golgi apparatus before its translocation into the neuronal cytosol.

Cytotoxic effects of a chimeric protein consisting of tetanus toxin light chain and anthrax toxin lethal factor in non-neuronal cells.

The light chain of tetanus toxin is a zinc endoprotease that inhibits neurotransmitter release by selective proteolysis of the synaptic vesicle-associated protein synaptobrevin/vesicle-associated membrane protein. Cellubrevin is a homologue of synaptobrevin that is found in most cell types and is also a substrate for tetanus toxin. The lack of receptors for tetanus toxin on most cell types has made studies of tetanus toxin action in non-neuronal cells difficult. To characterize tetanus toxin effects in non-neuronal cells, a fusion protein consisting of the 254 amino-terminal amino acids of lethal factor (LF) of anthrax toxin and tetanus toxin light chain (LC) was prepared. This protein (LF-LC) inhibited evoked glycine release from primary spinal cord neurons at concentrations between 1.0 and 100 ng/ml. LF-LC was cytotoxic to RAW 264.7, ANA-1 cells (mouse macrophage cell lines), and Chinese hamster ovary cells in a dose-dependent manner. These effects required the presence of protective antigen, the receptor binding component of anthrax toxin. In contrast, LF-LC was not cytotoxic to RBL-2H3, Vero, or mouse hybridoma cell lines. Mutagenesis of conserved amino acids (His237 and Glu234) in the zinc-binding motif of LC resulted in fusion proteins having no biological activity. LF-LC did not inhibit regulated secretion of serotonin in RBL-2H3 cells or constitutive secretion in any non-neuronal cell lines as measured in several different assays. We suggest that the cytotoxic effects of LF-LC result from inhibition of a specific intracellular membrane fusion event mediated by cellubrevin.

Differential effects of tetanus toxin on inhibitory and excitatory neurotransmitter release from mammalian spinal cord cells in culture.

The effect of tetanus toxin on depolarization-evoked and spontaneous synaptic release of inhibitory and excitatory neurotransmitters was examined in murine spinal cord cell cultures. Toxin action on the release of radiolabeled glycine and glutamate was followed over time intervals corresponding to the early phase of convulsant activity through the later phase of electrical quiescence. Tetanus toxin inhibited potassium-evoked release of [3H]glycine and [3H]glutamate in a time- and dose-dependent manner. Ninety minutes after the application of toxin (6 x 10(-10) M), the stimulated release of [3H]glycine was blocked completely, whereas stimulated release of [3H]glutamate was not blocked completely until 150-210 min after toxin application. Fragment C, the binding portion of the tetanus toxin molecule, had no effect on stimulated release of either transmitter. The spontaneous synaptic release of [3H]glycine was blocked totally within 90 min of toxin exposure. In contrast, the spontaneous release of [3H]glutamate, in toxin-exposed cultures, was elevated to nearly twice that of control cultures at this time. Thus, toxin-induced convulsant activity is characterized by a reduction in the spontaneous synaptic release of inhibitory neurotransmitter with a concomitant increase in the release of excitatory neurotransmitter, as well as the more rapid onset of blockade of depolarization-evoked release of inhibitory versus excitatory neurotransmitter.

Uptake, metabolism, and release of N-[3H]acetylaspartylglutamate by the avian retina.

N-Acetylaspartylglutamate (NAAG) is a nervous system-specific dipeptide that is released from retinal neurons on depolarization. In the present study, extracellular metabolism, uptake, and release of [3H]NAAG were examined in the chick retina. After in vitro incubation with NAAG radiolabeled in the glutamate moiety, [3H]glutamate and [3H]NAAG increased in retinal cells through time- and temperature-dependent processes, which were reduced in the absence of extracellular sodium. Coincubation of cells with [3H]NAAG and aspartylglutamate or phosphate resulted in the decreased extracellular appearance of [3H]glutamate, produced by hydrolysis of radiolabeled NAAG, and a consequent increased availability of [3H]NAAG for transport into the retinal cells. When this tissue was incubated with radiolabeled NAAG, glutamate, glutamine, or aspartate under similar conditions, only [3H]NAAG served as a significant source for the appearance of intracellular [3H]NAAG. These data support the conclusion that [3H]NAAG can be transported into retinal cells, whereas [3H]glutamate transport is the predominant process after release of this amino acid from NAAG by extracellular peptidase activities. After uptake, [3H]NAAG entered a cellular pool, from which the peptide was secreted under depolarizing conditions and in a calcium-dependent manner.

Isolation, ultrastructure, and partial characterization of collagen from the perineurium of the Florida lobster, Panulirus argus.

Highly concentrated extracellular filaments in the perineurium of the Florida spiny lobster, Panulirus argus, were isolated using ultracentrifugation and linear sucrose gradients. The pellet obtained was highly enriched for the filaments as observed by transmission electron microscopy. Fibril diameter and axial periodicity measurements were obtained from filaments positively and negatively stained with uranyl acetate. A period between 14.0 and 25.0 nm and an average fibril diameter of 15.0 nm were observed. The filaments proved resistant to solubilization by most conventional agents and by several collagenases. NaOH (0.1 M at 100 degrees C) safely dissolved the filaments for measurements of protein content by the Lowry method and carbohydrate content with anthrone reagent. These tests revealed a protein content of approximately 84% and a high carbohydrate content of approximately 15%. Polyacrylamide electrophoresis of an acid-pepsin filament extract revealed a highly concentrated band (approximately 100,000) corresponding to the alpha-1 and alpha-2 bands of vertebrate type I collagen. Wide angle X-ray diffraction yielded meridional reflections that confirmed the filaments as collagen when compared with mammalian collagen X-ray diffraction. The amino acid composition was determined with a computer-assisted Beckman amino acid analyzer, which showed a glycine content of 279 residues/1000. Hydroxylysine and hydroxyproline were present in lower concentrations than expected.

Effect of optic nerve transection on N-acetylaspartylglutamate immunoreactivity in the primary and accessory optic projection systems in the rat.

Evidence has been presented in recent years that support the hypothesis that N-acetylaspartylglutamate (NAAG) may be involved in synaptic transmission in the optic tract of mammals. Using a modified fixation protocol, we have determined the detailed distribution of NAAG immunoreactivity (NAAG-IR) in retinal ganglion cells and optic projections of the rat. Following optic nerve transection, dramatic losses of NAAG-IR were observed in the neuropil of all retinal target zones including the lateral geniculate nucleus, superior colliculus, nucleus of the optic tract, the dorsal and medial terminal nuclei and suprachiasmatic nucleus. Brain regions were microdissected and NAAG levels measured by a radioimmunoassay (RIA) (IC50: NAAG = 2.5 nM, NAA = 100 microM; smallest detectable amount = 1-2 pg/assay). Large decreases (50-60%) in NAAG levels were detected in the lateral geniculate, superior colliculus and suprachiasmatic nucleus. Moderate losses (25-45%) were noted in the pretectal nucleus and the nucleus of the optic tract. Smaller changes (15-20%) were detected in the paraventricular nucleus and the pretectal area. These results are consistent with a synaptic communication role for NAAG in the visual system.

Ultrastructure studies of the extracellular filaments in the ventral nerve of Panulirus argus.

The ventral nerve cord of the spiny lobster, Panulirus argus was examined by transmission and scanning electron microscopy. Tannic acid mordant stain was used to enhance extracellular filaments. The ventral nerve cord is surrounded by an unusual perineurial sheath composed primarily of interwoven extracellular filaments. Gap junctions were found associated with the glial cells making up the perineurium. The axo-glial wrappings also contained extracellular filaments associated in bundles rather than uniformly around the axons. The extracellular filaments of the perineurium and axo-glial wrappings appeared to be morphologically identical with diameters ranging from 10-15 nm.

Calcium-dependent release of N-acetylaspartylglutamate from retinal neurons upon depolarization.

N-Acetylaspartylglutamate (NAAG) is present in high concentrations specifically in the nervous system. Its neuronal distribution, presence in synaptic vesicles and its excitatory actions support the hypothesis that this dipeptide participates in communication between neurons. Following the incorporation of [3H]glutamate by frog retinal cells in vivo, the release of radiolabeled glutamate, GABA and NAAG was studied during acute incubation of the retina in vitro. Release of the radiolabeled amino acids and dipeptide was stimulated by elevated extracellular potassium. The release required the presence of extracellular calcium. These data are the first which demonstrate the release of NAAG following biosynthesis from a radiolabeled precursor and are consistent with synaptic release of this dipeptide.

Ultrastructural localization of N-acetylaspartylglutamate in synaptic vesicles of retinal neurons.

Localization of N-acetylaspartylglutamate (NAAG) in a variety of central and peripheral neurons, as well as its receptor-mediated activation of membrane conductance have led to speculation that this peptide has a role in chemical neurotransmission. We previously identified NAAG in retinal neurons of several species, including the grass frog, and now have determined its ultrastructural distribution within the plexiform layers of this amphibian retina. NAAG immunoreactivity was localized within vesicles in synaptic endings of presumptive amacrine and bipolar neurons in the inner plexiform layer. Additionally, the peptide was present in vesicles within ribbon synapses in the outer plexiform layer, a result suggestive of release from photoreceptor cells. These data support the hypothesis that NAAG is secreted at points of synaptic contact between neurons, including retinal amacrine, bipolar and photoreceptor cells.

Regulation of cell shape in Euglena gracilis. V. Time-dependent responses to Ca2+ agonists and antagonists.

The daily changes in cellular shape observed in growth-synchronized cultures of Euglena gracilis Klebs strain Z, were altered by exposure to Ca2+ channel agonists and antagonists. The response of the cells to these pharmacological agents depended, in part, on the time in the growth cycle that the cells were exposed. The Ca2+ channel blockers verapamil and nifedipine and the intracellular Ca2+ antagonist TMB-8 all caused cell rounding when elongated cells from the middle of the light cycle were treated. These results were the same as with other methods used to deprive cells of extracellular Ca2+, such as exposure to EGTA or resuspension in Ca2+-free medium. The cell response in mid light cycle to the channel blockers was reversible by simultaneous exposure to CaCl2, and the nifedipine response was also reversed by simultaneous exposure to the structurally related Ca2+ agonist BAY-K 8644. Exposure of cells in the first hour of the light cycle to verapamil, nifedipine or TMB-8 caused an unexpected result. Instead of preventing the round cells from elongating in the first portion of the light cycle, as do LaCl3, EGTA or resuspension in Ca2+-free medium, a greater than expected percentage of elongated cells was found in the treated population. This represents the first instance in which the biological clock control over the rate and extent of cell elongation was accelerated. The calcium agonist CGP-28392 did not have an effect on cell elongation in the early portion of the light cycle but caused cell rounding in the middle of the light cycle. The calcium agonist BAY-K 8644 did not cause any shape changes alone, but was capable of reversing the effects of nifedipine in the middle of the light cycle.