Eduardo De Robertis

The neurotubular system of the axon and the origin of granulated and non-granulated vesicles in regenerating nerves

SummaryElectronmicroscope observations have been made on compressed sciatic nerves and preganglionic afferents to the superior cervical ganglia of rats. After 6 hours, the proximal regenerating stumps of both myelinated and unmyelinated axons become filled with enlarged neurotubules and vesicles. Granulated vesicles of 750–900 Å, having a dense core become abundant in all types of regenerating axons and they increase in number after 24 hours. The vesicular material is formed by dilatation and pinching off from neurotubules. The existence of a neurotubular system within the axon, connected with the Golgi complex at the perikaryon and involved in the formation of vesicles, is postulated. The presence of granulated vesicles in all types of regenerating axons and nerve terminals is discussed in relation with their possible functional significance. The distal stumps of compressed sciatic nerves show, after 6 hours, a considerable increase in membranous material within the axoplasm mainly represented by multivesicular and lamellar bodies. This reaction, which is interpreted as being autolytic, is compared with the regenerative reaction of the proximal stump.

Molecular Biology of Synaptic Receptors

A special proteolipid (a hydrophobic protein) has been extracted and purified from nerve-ending membranes and total particulate matter of gray areas of the central nervous system. Such a proteolipid shows a high affinity for binding d-tubocurarine, serotonin, and atropine and has been called receptor proteolipid. The interaction of this proteolipid with atropine sulfate was studied with light scattering and polarization of fluorescence. The changes observed, which follow a cooperative type of curve, were attributed to the aggregation of the proteolipid macromolecules. Such a phenomenon was then observed under the electron microscope. A receptor proteolipid having a high affinity for binding acetylcholine, hexamethonium, and other cholinergic drugs was isolated and purified from electric tissue of fishes and from electroplax membranes. Such a proteolipid was also extracted from membranes from which acetylcholinesterase had been removed, and it was concluded that this enzyme and the receptor proteolipid are two different macromolecules. A high affinity binding site with a dissociation constant of K1 equal to 10-7 and about ten sites with K2 equal to 10-5 were recognized in the receptor proteolipid. Under the electron microscope the receptor proteolipid of brain appears as a rod-shaped macromolecule which may assume paracrystalline arrays with 10-8 molar atropine sulfate. Similarly the receptor proteolipid from electric tissue and from skeletal muscle may form paracrystalline arrays under the action of acetylcholine and hexamethonium. A model of the cholinergic receptor based on the properties of the proteolipid is presented. Preliminary work indicates the possibility of obtaining a biophysical response to acetylcholine when the receptor proteolipid is embedded in artificial bilayered lipid membrance.

Ultrastructure and cholinergic binding capacity of junctional complexes isolated from rat brain

Abstract The results and techniques developed in the previous paper13 were used to isolate the junctional complexes from the rat brain cortex and to study their binding capacity for cholinergic blocking agents. The non-ionic detergent Triton X-100, acting on the isolated nerve ending membranes, produces disintegration of most of the limiting membranes leaving intact the junctional complexes. These complexes are composed of the two synaptic membranes joined by the intersynaptic filaments and the subsynaptic web. Some curvature and increase in thickness and electron density of the web is observed in the Triton-treated specimens. The uptakes of dimethyl-[14C]- d -tubocurarine and methyl-[14C]hexamethonium were studied, in vitro, in controls and after Triton X-100 treatment. Practically no loss of binding capacity was observed after the detergent and in some cases the uptake in counts/min per mg protein showed some increase. Comparison with the effect on acetylcholinesterase and Na+-K+-adenosinetriphosphatase was very striking because of the considerable loss of these membrane-bound enzymes. It is concluded that while these enzymes have a wider distribution, which includes the limiting membrane of the ending, the cholinergic receptor is localized in the junctional complex, probably at the subsynaptic membrane. These electron-microscopic and biochemical studies demonstrate that the junctional complexes from brain synapses may be isolated and their receptor properties studied.

Action of triton X-100 on ultrastructure and membrane-bound- enzymes of isolated nerve endings from rat brain.

Abstract The action of the non-ionic detergent Triton X-100 on several subcellular fractions from rat brain was studied with a special emphasis on the solubilization effect on protein and membrane-bound enzymes. These parameters were studied in the pellet (P) and in the supernatant (S). On the crude mitochondrial fraction, with 0.1–0.5% Triton X-100, the effect is very marked on acetylcholinesterase (23.4−13.7% in P), less marked on protein (62.1−51.3% in P) and little on monoamine oxidase (79.1−62.0% in P). On the submitochondrial fraction B after Ficoll gradient, which is mainly composed of nerve endings, the effect was very similar for acetylcholinesterase, monoamine oxidase and protein. No solubilizing action of the detergent was observed on choline acetylase, an enzyme which in the rat is related to the synaptic vesicles. On the same fraction and on the submitochondrial M1 fraction after osmotic shock, the Na+-K+-adenosinetriphosphatase showed not only solubilization but considerable activation. Extraction, without activation, was observed on the Mg2+-dependent, ouabain-insensitive adenosinetriphosphatase. On the nerve ending and synaptic membranes, isolated in the sucrose gradient of M1, considerable loss of acetylcholinesterase, Na+-K+-adenosinetriphosphatase and p-nitrophenylphosphatase occurred in P. Study with the electron microscope of the various Triton-treated fractions showed that the effect is mainly on the limiting membrane of the nerve endings causing disintegration of these isolated structures. At low concentrations the effect on mitochondria, myelin and synaptic vesicles is not very marked, a fact that is also corroborated by several biochemical parameters. The junctional complexes which comprise the synaptic membranes and the subsynaptic web appear to be more resistant to the action of the detergent.

NERVE ENDINGS IN METHIONINE SULPHOXIMINE CONVULSANT RATS, A NEUROCHEMICAL AND ULTRASTRUCTURAL STUDY*

THE FINDING that dogs underwent seizures after ingestion of flour from NC13-treated wheat (MELLANBY, 1946) led to the isolation of methionine sulphoximine (MSO) as the toxic convulsant factor (BEKTLEY, MCDEXMOTT, MORAN, PACE and WHITEHEAD, 1950; MISANI and REINER, 1950; REINER, MISANI and WEISS, 1950). Because of the similarity of the convulsions to those of human epilepsy this agent has been the object of numerous physiological and neurochemical studies (see WOLFE and ELLIOT, 1962). Several investigations pointed toward a possible causal relationship between MSO and a disequilibrium in glutamine-glutamic acid metabolism. PETERS and TOWER (1959) found a marked inhibition of glutamine synthesis in brain slices of MSO-treated cats, and a relationship between amidation of glutamic acid in brain and seizures was also postulated in rats after MSO or acoustic epileptogenic stimulation by KOLOUSEK, HORAK and JIRACEK (1959). SELLINGER and WEILER (1963) reported that in vitro MSO competitively inhibits glutamine synthetase (GS), and more recently LAMAR and SELLINGER (1965) found that in vivo the drug inhibits GS irreversibly. In this last paper some preliminary observations on the subcellular distribution of labelled MSO were made and evidence was reported that it is preferentially bound to the microsomal fraction isolated from an homogenate in water, which is particularly rich in GS (DE BALBIAN VERSTER, SELLINGER and HARKIN, 1965). Recently SALGANICOFF and DE ROBERTIS (1965) analysed the subcellular distribution of GS and other enzymes related to the metabolism of glutamate, glutamine and GABA. The fractionation procedure, carried out in sucrose solutions under standard conditions and with electron microscopic control, enabled them to follow the localization of certain enzymes in well defined subcellular fractions of the CNS such as mitochondria, myelin, and particularly nerve endings, synaptic vesicles and membranes. In agreement with the results of WAELSCH (1959) and SELLINGER and DE BALBIAN VERSTER (1962), GS was found to be concentrated in the microsomal fraction.

Glutamic acid decarboxylase inhibition and ultrastructural changes by the convulsant drug allylglycine

Abstract The effects in vivo and vitro of the convulsant drug allylglycine on the activity of glutamic acid decarboxylase (GAD), aminobutyrate aminotransferase, glutamine synthetase and aspartate- and alanine-aminotransferases of the rat cerebral cortex were studied. The most significant result was the finding of an inhibition of GAD, during the period of convulsion, which was even greater by addition in vitro of the drug. This inhibition is not by way of the cofactor, pyridoxal phosphate. Preincubation of the homogenate with allylglycine enhanced the inhibition, while preincubation in buffer-substrate produced a protective effect. The inhibition of GAD was correlated with a decrease of 40 per cent in the concentration of γ-aminobutyric acid in the cerebral cortex. In the convulsant rat, ultra-structural alterations of some nerve endings of the cerebral cortex were observed. After cell fractionation, such altered nerve endings were preferentially found in the GAD-rich (nonaminergic) fraction of isolated nerve endings. The possible mechanism of the convulsions induced by allyglycine is discussed and a specific effect on GAD-rich inhibitory nerve endings is postulated.

The secretory cycle of supraoptic neurons in the rat

SummaryAfter perfusion with formaldehyde and glutaraldehyde the supraoptic nucleus and infundibular process of the neurohypophysis of the rat were dissected and prepared for electronmicroscope observation. This study was carried out in specimens under normal water balance, in others fed on dry food and in rats submitted to forced hydration.Two extreme types of neurons with intermediary stages were recognized in the normal supraoptic nucleus. The main difference between them is in the content of ribosomes, development and dilatation of the vacuolar system and in the number of elementary neurosecretory granules. In both types lysosome-like particles are observed. The volume of the elementary granules increases 1.7 times along the hypothalamic-hypophyseal tract while the increase in the dense core of the granule is of the order of 4.3 times.After forty-eight hours on dry food there is a general depletion of secretory granules from the perikaryon and nearby axons, the ribosomes are numerous and the endoplasmic reticulum is dilated in all cells and contains a macromolecular filamentous material. With more prolonged dehydration the neurosecretory granules reappear in relation to the Golgi complex and the vacuolar system becomes progressively flattened. With forced hydration the number of granules in the perikaryon increases considerably.These observations are interpreted as indicative that the early stages of synthesis take place at the level of the ribosomes. The product, in a dilute macromolecular form, is transferred into the cisternae of the endoplasmic reticulum and then condensed into granules within the Golgi complex. The increase in size of the granules along the axon is discussed in relation to the progressive increase in hormone content.

Proconvulsant and anxiogenic effects of n-butyl β carboline-3-carboxylate, an endogenous benzodiazepine binding inhibitor from brain

The discovery of n-butyl beta carboline-3-carboxylate (beta CCB) as an endogenous substance of brain capable of interacting with the central benzodiazepine receptor, and the fact that this beta carboline increases in the cerebral cortex of rats undergoing acute stress, led us to study the pharmacological properties of beta CCB in mice. Using 3-mercaptopropionic acid in subconvulsant doses, it was found that this beta carboline, although not being a convulsant, has a proconvulsant action, as indicated by the number of mice undergoing convulsions and the reduction in latency. This proconvulsant effect was observed both with IP or ICV injections and was blocked by the benzodiazepine receptor antagonist RO 15-1788. In an open-field test the injection of 0.3 mg/kg of diazepam increased the number of squares crossed, while beta CCB had the opposite effect, reducing the squares crossed in a dose dependent manner between 1 and 30 mg/kg. This drug also increased the time of freezing and decreased the number of rearings. These changes were partially counteracted by the injection of 3.6 mg/kg of RO 15-1788. In the plus-maze test, 10 mg/kg chlordiazepoxide increased the number of entries and the time spent in the open arms, while the beta carboline produced the opposite effect. The conclusion reached is that beta CCB has both proconvulsant and anxiogenic actions, behaving as an inverse agonist for the central benzodiazepine receptor.

Species differences in subcellular distribution of choline acetylase in the CNS. A study of choline acetylase, acetylcholinesterase, 5-hydroxytryptophan decarboxylase, and monoamine oxidase in four species.

THE isolation of ACh-rich and ACh-poor nerve endings from the CNS, and the demonstration that ACh is concentrated in a fraction rich in synaptic vesicles, made it of interest to study the subcellular localization of ChAc in those fractions. In rat brain, a direct correlation between ACh content and ChAc activity in the two nerve ending fractions was observed; this result indicates that both the transmitter and the synthesizing enzymes are topographically related within the synaptic complex. A similar close relationship was found in the fraction containing the synaptic vesicles, which were separated from osmotically disrupted nerve endings. In spite of some solubilization, ACh and ChAc were found to have 3.6and 5.6-fold concentrations, respectively, in synaptic vesicles to the total homogenate (DE ROBERTIS, PELLECRINO DE IRALDI, RODR~GUEZ DE LORES ARNAIZ and SALGANICOFF, 1962; DE ROBERTIS, RODRiCUEZ DE LORES ARNAIZ, SALGANICOFF, PELLECRINO DE IRALDI and ZIEHER, 1963). WHITTAKER, MICHAELSON and KIRKLAND (1964), using osmotic shock and a gradient technique to separate the vesicular fraction from guinea pig brain, confirmed our results for ACh but found that ChAc was not related to the synaptic vesicles. According to them, ChAc is essentially a soluble enzyme which is similar to lactate dehydrogenase and potassium in its subcellular distribution. These authors appeared to question the validity of our results without giving consideration to a possible species difference or taking into account the different technical approaches used. In view of this conflicting report, we have reinvestigated the localization of ChAc in the rat and guinea pig brain and extended this study to the rabbit and pigeon. A new and more precise technique for ChAc was used (MCCAMAN, 1963; MCCAMAN and HUNT, 1965). In this micromethod [14C]acetyl CoA and choline are used as

Subcellular distribution and chemical nature of the receptor for 5-hydroxytryptamine in the central nervous system.

Abstract— In vitro binding experiments with 5‐hydroxy[14C]tryptamine (3.3 × 10−6 M) were carried out on subcellular fractions of the cat brain. The highest specific activity was observed in some fractions of nerve‐ending membranes isolated from the hypothalamus, basal ganglia, and gray areas of the mesencephalon. The specificity of this high affinity binding was demonstrated by competition with reserpine, butanolamide of lysergic acid, and desmethylimipramine. With butanol‐water extraction the [14C]5‐HT was found in the butanol while the gangliosides were separated in the water phase. Several experiments with thin layer and column chromatography suggest that in the organic phase the [14C]5‐HT is not bound to the lipids but to a special proteolipid. This proteolipid is different from that found in myelin and has similar chromatographic properties to that previously observed in the proteolipid which binds d‐[14C]tubocurarine in nerve‐ending membranes of the cerebral cortex.