A. Preti

Occurrence of Sialyltransferase Activity in the Synaptosomal Membranes Prepared from Calf Brain Cortex

The possible occurrence of sialyltransferase activity in the plasma membranes surrounding nerve endings (synaptosomal membranes) was studied, using calf brain cortex. The synaptosomal membranes were prepared by an improved procedure which provided: (a) a „nerve ending fraction” consisting of at least 85% well‐preserved nerve endings and containing only small quantities of membranes of intracellular origin; (b) a „synaptosomal membrane fraction” carrying high amounts of authentic plasma membrane markers (Na+‐K+ ATPase, 5′‐nucleotidase, sialidase, gangliosides) with values of specific activity four to fivefold higher than those in the „nerve ending fraction” and very small amounts of cerebroside sulphotransferase, marker of the Golgi apparatus, and of other markers of intracellular membranes (rotenone‐insensitive NADH and NADPH: cytochrome c reductases), the specific activities of which were, respectively, 0.5‐ and 0.7‐fold that in the „nerve ending fraction”. Thus the preparation of synaptosomal membranes used had the characteristics of plasma membranes and carried a negligible contamination of membranes of intracellular origin. The distribution of sialyltransferase activity in the main brain subcellular fractions (microsomes; P2 fraction; nerve ending fraction; mitochondria) resembled most closely that of thiamine pyrophosphatase, the enzyme known to be linked to the Golgi apparatus and the plasma membranes and of acetylcholine esterase, the enzyme known to be linked to either intracellular or plasma membranes. The enrichment of sialyltransferase activity in the „synaptosomal membrane fraction”, referred to the „nerve ending fraction”, was practically the same as that exhibited by authentic plasma membrane markers. All this is consistent with the hypothesis that in calf brain cortex sialyltransferase has two different subcellular locations: one at the level of intracellular structures, most likely the Golgi apparatus (as described by other authors), the other in the synaptosomal plasma membranes. The basic properties (pH optimum, V/S, V/t and V/protein relationships) and detergent requirements of the synaptosomal membrane‐bound sialyltransferase were established. The highest enzyme activities were recorded on exogenous acceptors, lactosylceramide and ds‐fetuin. The Km values for CMP‐NeuNAc were different using lactosylceramide and ds‐fetuin as acceptor substrates (0.57 and 0.135 mm, respectively); the thermal stability of the enzyme acting on glycolipid acceptor was higher than that on the glycoprotein acceptor; the effect of detergents was different when using glycoprotein from glycolipid acceptors; no competition was observed between lactosylceramide and ds‐fetuin. Thus the synaptosomal membranes carry at least two different sialyltransferase activities: one acting on lactosylceramide (and glycolipid acceptors), the other working on ds‐fetuin (and glycoprotein acceptors). Ganglioside GM3 was recognized as the product of synaptosomal membrane‐bound sialyltransferase activity working on lactosylceramide as acceptor substrate.

Parallelism of subcellular location of major particulate neuraminidase and gangliosides in rabbit brain cortex

Abstract The subcellular distribution of major particulate neuraminidase and gangliosides in the rabbit cortex was determined. The neuraminidase and gangliosides were found to be present in: (a) the nerve endings; (b) the light membranes of microsomal origin ; (c) the myelin-rich preparation obtained from the crude mitochondrial fraction. Their distribution patterns were very similar also from the quantitative point of view; in fact 40 % of the neuraminidase activity and 43.5 % of the gangliosides were recovered in the nerve endings; 50% of the enzyme and 45% of the gangliosides in the light microsomal membranes; 8 % of the neuraminidase and 11 % of the gangliosides in the myelin-rich preparation. Conversely, the preparations enriched in nuclei, mitochondria and lysosomes were practically devoid of both neuraminidase and gangliosides. The ganglioside patterns of the different subcellular fractions were similar, except for the myelin-rich subfraction which contained higher amounts of monosialoganglioside GM1. The microsomal light membranes had a slightly lower content of trisialoganglioside GTIb and tetrasialoganglioside GQ1 than the nerve endings. These results may be considered consistent with the hypothesis that neuraminidase and gangliosides are fundamental components of the neuronal plasma membrane, thus following the distribution of the different neuronal fragments after homogenization and fractionation.

MEMBRANE-BOUND NEURAMINIDASE IN THE BRAIN OF DIFFERENT ANIMALS: BEHAVIOUR OF THE ENZYME ON ENDOGENOUS SIALO DERIVATIVES AND RATIONALE FOR ITS ASSAY

—The activity of brain membrane‐bound neuraminidase on endogenous and exogenous substrates was comparatively studied in various animals (rat, chicken, rabbit, pig, calf and human). The maximum rate of hydrolysis of endogenous substrates by membrane‐bound neuraminidase (using a crude preparation of the enzyme) was different in the various animals (from 0·05 to 0·73 units, referred to 1 mg protein) and was obtained under similar but not identical optimum conditions (pH from 4·1 to 5·1; requirement or not of Triton X‐100). The maximum degree of hydrolysis of endogenous substates was also different (from 15 to 27 nmol released NeuNAc/mg protein) and was obtained within different incubation periods (from 2 to 18 h). It corresponded (in rabbit, calf, human brain only), or not, to the actual exhaustion of the endogenous substrates.

Optimal conditions for the assay of fibroblast neuraminidase with different natural substrates

A method for the assay of neuraminidase in human cultured fibroblasts has been worked out. The substrates, all naturally occurring, were: sialyloligosaccharides (alpha(2 lead to 3)sialyllactose, alpha(2 leads to 6)sialyllactose, disialyllactose), sialylglycoplipids (disialogangliosides GD1a and GD1b), sialylglycoproteins and sialylglycopeptides (ovine submaxillary glycoprotein and its pronase-glycopeptides). The method was based on the determination of the enzymically liberated N-acetylneuraminic acid (NeuAc) by a chromatographic-colorimetric microprocedure. The enzyme acted on sialyloligosaccharides and, in the presence of Triton X-100, on gangliosides, while it did not appreciably affect sialylglycoproteins and sialylglycopeptides. The optimum pH was 4.0 for all tested substrates; the Km values were higher for sialyloligosaccharides (about 10(-3) M), lower for gangliosides (about 10(-4) M); the apparent maximum velocity was higher with alpha(2 leads to 3)sialyllactose (400 mU/mg protein); the reaction rate was linear with time for up to 2 h, and with up to 0.6 mg of enzymic protein. The assay method proved to be sufficiently sensitive (3-4 nmol liberated NeuAc), simple, and reproducible (mean activity on pooled fibroblasts with alpha(2 leads to 3)sialyllactose: 400 mU +/- 6 S.E.).

Gangliosides, neuraminidase and sialyltransferase at the nerve endings.

Gangliosides are characteristic glycolipid components of the plasma membranes of mammalian cells. They are particularly abundant in the nervous tissue, specially the grey matter, where their concentration is about one tenth that of total phospholipids. The high content of gangliosides in the neuronal membranes, the great variety in the composition of their oligosaccharide chains, and their peculiar location in the outer membrane surface are enough evidence to stimulate research and speculation on the possible involvement of gangliosides in brain specific functions. As a matter of fact, gangliosides are just located- the synaptic junctions-where a specialized physiological event takes place, and definitely synaptic membranes would be and would behave differently without gangliosides. However, in order to provide a plausible working hypothesis for any specific roles of gangliosides in brain function, a more precise knowledge on the contribution given by gangliosides to the local environment of the membrane, in terms of capability and quality of interactions with both the lipid and protein components of the membrane itself, is required.

Cytosolic sialidase in developing rat forebrain

The nature and developmental profile of the soluble sialidase of rat forebrain were studied from birth to 150 days. Forebrain was extracted by two procedures, one (mild) preserving, the other (drastic) destroying nerve endings. The soluble extracts obtained by the mild procedure contained 64-78% of the total tissue cytosol, assayed as lactate-dehydrogenase; those obtained by the drastic procedure 87-94%. These latter extracts were considered as the soluble fraction containing all tissue cytosol. The cytosolic origin of the sialidase contained in the soluble extracts at all examined ages was suggested by the following evidence: (a) during extraction sialidase behaved as lactate-dehydrogenase and quite differently from ?-hexosaminidase and ?-galactosidase, enzymes of lysosomal nature present in the same extracts, (b) the sialidase content of the extract was not influenced by the presence or absence of EDTA in the medium, (c) the sialidase content in the extracts did not diminish even after prolonged centrifugation (2 h) at high speed (150,000 g). The content of cytosolic sialidase referred to g fresh tissue increased from birth to 20 days, and slowly decreased thereafter. Till 20 days the content and the developmental trend of the cytosolic enzyme were similar to that of the better known membrane bound sialidase. This latter enzyme, however, reached its maximum at about 60 days of age. The specific activity of the cytosolic sialidase was lower till 10 days of age, higher from 10 to 30 days, and equalled that of the membrane bound enzyme during adult life. Therefore rat forebrain cytosolic and membrane bound sialidases, also from the developmental point of view, behave as different enzymes.

Retinal Gangliosides: Composition, Evolution with Age. Biosynthetic and Metabolic Approaches

Gangliosides are a complex group of glycosphingolipids which contain one or more molecules of sialic acid. They are localized in a rather high amount and with a great diversity in cytoplasmic membranes of neurones especially in synaptosomes and membranes of microsomes (1–3). Gangliosides contain a sphingosine base, a fatty acid and at least one mole of hexose and sialic acid. The functional role of gangliosides is poorly understood, despite their early discovery by Klenk in metabolic disorders causing neurological diseases, named gangliosidoses (4). The accumulation of gangliosides in the metabolic disorders have been related to the disturbance in the enzymes involved in their metabolism (5).

Studies on brain membrane-bound neuraminidase. I. General properties of the enzyme prepared from calf brain

Abstract 1. 1. A crude preparation of membrane-bound neuraminidase (mucopolysaccharide N -acetylneuraminylhydrolase, EC 3.2.1.18), obtained from calf brain, was used. This preparation was free from the soluble and lysosomal neuraminidases and depleted of endogenous substrates. 2. 2. The enzyme showed: (a) optimum pH 4.0 for gangliosides GD1a, GD1b, GT1b and GQ1; 3.8 for ovine submaxillary mucin, ovine submaxillary mucin-sialoglycopeptides and brain sialoglycopeptides; 3.1 for sialyllactose (C-3 isomer); (b) apparent higher affinity for gangliosides ( K m of the order 10 −5 M) than for sialyllactose ( K m 0.68·10 −3 M), ovine submaxillary mucin and sialoglycopeptides; (c) higher V for gangliosides (1.26–2.12 units/mg protein), lower for sialyllactose (1.18 units/mg protein) and sialoglycoprotein substrates (0.27–0.41 units/mg protein); (d) inhibition by excess substrate (over 0.15-0.2 mM) only in the case of gangliosides; (e) maximum rate of hydrolysis of gangliosides at 70 °C (24 units/mg protein in the case of ganglioside GD1a); (f) considerable stability. 3. 3. Na + and Li + did not influence the enzyme activity; K + activated below 0.1 M; NH 4 + started inhibiting at 0.01 M. All bivalent cations tested inhibited the enzyme: Hg 2+ from 10 −6 M, Cu 2+ from 10 −5 M, Ca 2+ from 10 −3 M. Anions had no appreciable influence on the enzyme activity, at concentrations up to 5·10 −2 M.

Studies on brain cytosol neuraminidase. II. Extractability, solubility and intraneuronal distribution of the enzyme in pig brain

The origin and properties of cytosolic neuraminidase (acylneuraminyl hydrolase, EC 3.2.1.18) from pig brain were studied. 1. The brain extracts containing the cytosol derived from neuronal bodies and glial cells carry 0.69 munits neuraminidase/g fresh tissue. The behaviour of neuraminidase during extraction closely paralleled that of authentic cytosolic enzyme, lactate dehydrogenase; whereas, it differed from that of the lysosomal enzymes, beta-hexosaminidase and beta-galactosidase, also found in the extracts. 2. Nerve endings from either crude or purified preparations, when treated by hypoosmotic shock, released neuraminidase activity up to a maximum of 1.25 munits/g fresh tissue. The behaviour of releasable neuraminidase was always identical to that of lactate dehydrogenase and very similar to that of ATPase and acetylcholinesterase. Typical lysosomal enzymes, however, such as beta-galactosidase and beta-hexosaminidase, behaved differently under the same conditions. This neuraminidase activity is thought to be derived from the cytosol of nerve endings. 3. The specific activity of neuraminidase in nerve-ending cytosol is 15--20 times that in neuronal body and glial cell cytosol. Some properties (pH, Km value, V/t relationship) of the cytosolic enzymes of different origin are similar; others (stability on standing at 4 degrees C; resistance to freezing and thawing) are different. Hypoionic solutions caused both cytosolic neuraminidases to slowly precipitate and to assume a stable insoluble form which was still active.

Assay of brain particulate neuraminidase III. Preparation of the enzyme devoid of endogenous substrates

Abstract 1. 1.|The crude preparation of brain particulate neuraminidase (mucopolysaccharide N-acetylneuraminylhydrolase, EC 3.2.1.18) (0–105 000 × g pellet, prepared from rabbit brain homogenate in isotonic sucrose) was depleted from endogenous (intrinsic) substrates by autolysis performed under the following conditions: 0.18 M sodium acetate buffer, 0.3% Triton X-100, final pH 4.2, 25°, 5 h incubation. 2. 2.|After this treatment only 54% protein, 51% sialoglycoproteins, 89% gangliosides, 75% total and lipid-bound phosphorus remained particulate. Neuraminidase remained firmly particulate and fully active (97% recovery). Thus, the 105 000 × g sediment obtained after the above treatment is the enzyme preparation devoid of endogenous substrates. 3. 3.|The basic properties of the enzyme present in the preparation devoid of endogenous substrates were the same observed in the starting crude preparation. The Km for disialoganglioside GD1a was 4.8·10−6 M; the vmax 1.6 nmoles released N-acetylneuraminic acid per min. 4. 4.|The examined properties of particulate neuraminidase depleted from endogenous substrates by our procedure were very similar to those of the enzyme prepared according to Z. Leibovitz and S. Gatt (Biochim. Biophys. Acta, 152 (1968) 136). However, the final enzyme recovery was 5 times higher with our procedure.