THE ACTION OF CALF BRAIN SIALIDASE ON GANGLIOSIDES, SIALOGLYCOPROTEINS AND SIALOGLYCOPEPTIDES

Abstract— In agreement with other investigators it has been shown that endogenous as well as added gangliosides are a substrate for brain sialidase. The release of sialic acid was enhanced in the presence of Triton X-100; this might be due to the action of the detergent on the ganglioside micelles. The sialic acid release from endogenous gangliosides was observed over 48 h and compared with the effect of the sialidase on the endogenous glycoproteins. Though the hydrolysis of sialic acid from gangliosides is much faster in the first hours, after 48 h 40 per cent of the total bound sialic was released from both substrates at pH 4.0 and 37°C.
Sialoglycopeptides obtained from brain glycoproteins are also metabolized by the sialidase. No effect of Triton X-100 on this substrate has been observed. From sialoglycopeptides, fractions can be obtained by DEAE-Sephadex A-50 column chromatography with a sialic acid content from 8 to 26 per cent. The fractions with a high sialic acid content were about equally active towards brain sialidase as gangliosides. The results agree with the similar turnover rate observed for the carbohydrate chains from gangliosides and glycoproteins, but are in contrast to the observations of other investigators who have stated that glycoproteins are a poor substrate for brain sialidase. In our experiments bovine and ovine submaxillary mucins and sialyl-lactoses showed only slight activity compared to gangliosides and selected brain sialoglycopeptides.     

THE REGIONAL DISTRIBUTION OF CYTIDINE 5''-MONOPHOSPHO-N-ACETYL-NEURAMINIC ACID SYNTHETASE IN CALF BRAIN

—The enzyme cytidine 5′-monophospho-N-acetylneuraminic acid synthetase was studied in different parts of the calf brain. Characterization of partial purified enzyme preparations from cortical grey matter and corpus callosum by means of pH optima and K_m values, showed the enzyme of grey and white brain areas to be identical. Unexpectedly the regional differences of the enzyme activities per g wet tissue and per mg protein were very slight. From the presence of the enzyme in pure white brain areas, which are known to be poor in neuronal perikarya, and the fact that the enzyme is localized in the cell nucleus, we concluded that cytidine 5′-monophospho-N-acetylneuraminic acid is produced in glia cell nuclei and that it is very likely that biosynthesis of sialo-glycoproteins and/or ganglio-sides occurs within glia cells. The enzyme activity per μmol DNA-P is somewhat higher in grey than in white regions, indicating a slightly higher activity per neuronal than per glial nucleus. The regional differences of lipid and protein-bound sialic acid and RNA show a striking similarity and contrast to those of the enzyme. These differences are interpreted in terms of a differential content in neurons and glia cells.     

The organization of gangliosides and other lipid components in synaptosomal plasma membranes and modifying effects of calcium ion

Synaptosomes were prepared from bovine brain by zonal rotor sucrose density centrifugation. While a major fraction of lipid-bound sialic acid is included uniformly within the synaptosomal distribution profile, the sialoglycoproteins and some gangliosides do not follow this pattern, Exposure to extrasynaptosomal calcium results in alterations in the surface labeling properties of some gangliosides and membrane plasmalogens, suggesting that extrasynaptic Ca²⁺ may influence the conformation of complex lipids in synaptic plasma membranes. The level of intrinsic membrane-associated sialidase activity that liberates sialic acid from these sialoglycoconjugates parallels the synaptosomal buoyant density distribution profile, supporting a view that this enzyme resides in synaptosomal membranes in close association with a sialolipid substrate.     

Developmental Changes in the Biosynthesis of Sialoglycoprotein Substrates for Intrinsic Synaptic Sialidase

Abstract: N′-Acetyl-d -[6-³H]mannosamine was administered to 13- and 28-day-old rats by intraventricular injection. At various time intervals following the injection, synaptic membranes were prepared and the incorporation of radiolabel into sialic acid residues released from endogenous glycoproteins and gangliosides by intrinsic sialidase determined. Radiolabel was incorporated into synaptic membrane gangliosides and glycoproteins, and at all times tested, >90% of the label was associated with sialic acid. Sialic acid released from endogenous glycoproteins by intrinsic sialidase present in 28-day membranes incorporated only 20–25% as much radiolabel per nmole as sialic acid released by mild acid hydrolysis or by exogenous neuraminidase. In contrast, sialic acid released from glycoproteins present in 13-day-old membranes by intrinsic sialidase, mild acid hydrolysis, or exogenous neuraminidase all were similarly labelled. At both ages the specific radioactivity (cpm/nmol) of sialic acid released from gangliosides by the intrinsic enzyme was similar to the total ganglioside sialic acid released by mild acid hydrolysis. The results identify glycoprotein substrates for intrinsic synaptic membrane sialidase as a distinct metabolic class in the mature brain and suggest the occurrence of a developmentally related change in the metabolism of these glycoproteins.     

Identification of intrinsic sialidase and sialoglycoprotein substrates in rat brain synaptic junctions

Sialidase activity associated with rat brain synaptic junctions (SJ) and synaptic membranes (SM) was determined. Both fractions released sialic acid from exogenous glycopeptides and gangliosides. SJ accounted for 5-10% of the total sialidase activity recovered from SM following extraction with Triton X-100, and the specific activity of SJ sialidase was 60% of that of the parent SM fraction. Intrinsic SJ sialidase hydrolysed 12-15% of the sialic acid associated with endogenous SJ glycoproteins. Sialic acid residues associated with SJ glycoproteins were labelled with sodium borotritide and SJ proteins fractionated by affinity chromatography on concanavalin A-agarose. SJ glycoproteins that reacted with concanavalin A (con A+ glycoproteins) accounted for 25% of the total SJ [3H]sialic acid. Intrinsic SJ sialidase hydrolysed 20% of the [3H]sialic acid associated with these glycoproteins. Each molecular weight class of con A+ glycoprotein previously shown to be a specific component of the postsynaptic apparatus contained sialic acid and was acted on by intrinsic SJ sialidase.     

Lipid composition of human neural tumors.

Gangliosides, cholesterol, and phospholipids were quantitated in the tissues of 11 human neural tumors and the cells of two gliomas cultured in vitro. All tumor tissues contained higher water concentrations but lower total lipid concentrations than either human grey or white matter. In general they contained less cholesterol, sphingomyelin, and serine glycerophospholipid but more choline glycerophospholipid than white matter. Concentrations of total ganglioside sialic acid were intermediate between grey and white matter. Compared with normal brain, all tumors had greater proportions of the structurally less complex gangliosides and smaller proportions of the more complex gangliosides. This was most marked in the rapidly growing tumors while the better differentiated astrocytomas contained the greatest proportions of complex gangliosides. The cells of the cultured tumors contained amounts of total lipid and total phospholipid similar to their parent tissues. However, the cultures had less cholesterol, sphingomyelin, and total ganglioside than their parent tissues. There were significant amounts of choline and ethanolamine plasmalogens in both cultures and parent tissues. The ganglioside patterns of both cultures were complex but they contained a greater proportion of structurally simpler gangliosides than their parent tissues.-Yates, A. J., D. K. Thompson, C. P. Boesel, C. Albrightson, and R. W. Hart. Lipid composition of human neural tumors.     

Substrate specificity and inhibitor studies of a membrane-bound ganglioside sialidase isolated from human brain tissue

A ganglioside-specific sialidase that controls cellular functions such as growth, differentiation, and adhesion has been observed in a variety of cells, but its characterization proved difficult due to firm membrane attachment and lability of the purified enzyme. Here we report on the specificity toward gangliosides and susceptibility to certain inhibitors of a ganglioside sialidase solubilized and purified 5100-fold from human brain. The sialidase removed terminal sialic acids from gangliosides GM3, GM4, GD3, GD2, GD1 a, GD1 b, GT1 b and GQ1 b, but was inactive toward gangliosides with sialic acid in a branching position (as in GM1 and GM2). Lyso-GM3 and -GD1a were good substrates, too, whereas O-acetylation of the sialic acid as in 9-O-acetyl-GD3 caused strongly reduced cleavage. The new influenza virus drug 4-guanidino-2-deoxy-2,3-dehydro-N-acetylneuraminic acid (Zanamivir) exhibited an IC50 value of about 7 x 10(-5) M that was in the range of the 'classical' sialidase inhibitor 2-deoxy-2,3-dehydro-N-acetylneuraminic acid; the bacterial sialidase inhibitor 4-nitrophenyloxamic acid, however, was ineffective. The glycosaminoglycans heparan sulfate, heparin, chondroitin sulfates A and B, as well as dextran sulfate and suramin, were all strongly inhibitory, suggesting that glycosaminoglycans present on the cell surface or in the extracellular matrix may influence the ability of the sialidase to alter the ganglioside composition of the membrane.     

High-performance thin-layer chromatography and densitometric determination of brain ganglioside compositions of several species.

Improved resolution of complex brain ganglioside mixtures was achieved by high-performance thin-layer chromatography. The percentage distribution of individual gangliosides was then determined by direct densitometric seanning, employing a transmittance mode, of the resorcinol-positive spots on the plate. As little as 90 pmol (29 ng) of lipid-bound sialic acid could be detected with a good signal-to-noise ratio. A linear detector response was observed up to 3.0 μg of lipid-bound sialic acid. The brain white matter ganglioside patterns of eight animal species, including human, chimpanzee, monkey, chicken, bovine, sheep, and pig, were examined in detail. In addition, human brain gray matter, rat cerebral, rat brain gray matter, and rat cerebellar ganglioside patterns were also studied. Ganglioside G_M4 (G₇) was found to be one of the major components in primate and chicken brain white matter, but it represented only a minor ganglioside in other species. Other major gangliosides in all brain samples studied were G_M1, G_D1a, G_D1b, and G_T1b. G_M1 was more abundant in white matter than in gray matter. G_T1a, a recently discovered ganglioside species, was found in all species examined, but was most abundant in the rat cerebellum. The latter source also contained high proportions of G_T1b and G_Q1b.     

Distribution of acid and alkaline deoxyribonucleases in white and grey matter regions of growing and aging chick brain

The activities of acid and alkaline deoxyribonucleases in the white and grey matter areas of growing and old chick cerebrum were measured. Two marker enzymes for glial cells, butyrylcholinesterase and carbonic anhydrase were also measured in these regions. Higher specific activities of both butyrylcholinesterase and carbonic anhydrase were found in the white matter region at all the stages studied. Acid and alkaline deoxyribonuclease activities were observed in both white and grey matter. The decrease in the specific activity of acid deoxyribonuclease with advancement of age was more pronounced as compared to the alkaline deoxyribonuclease Marked reduction in total acid deoxyribonuclease activity in white matter, beyond the age of 130 days, was observed. On the other hand, total alkaline deoxyribonuclease activity in both white and grey matter continued to increase with age Further, the activity per mg of DNA also increased in white matter of the old brain. These results indirectly suggest a continued role for alkaline deoxyribonuclease in glial cells formed at a later age.     

Specificity of the N1 and N2 sialidase subtypes of human influenza A virus for natural and synthetic gangliosides

Sialyl-linkage specificity of sialidases of the human influenza A virus strains, A/Aichi/2/68 (H3N2) and A/PR/8/34 (H1N1) were studied using natural and synthetic gangliosides. The sialidase of the A/Aichi/2/68 strain hydrolyzed the terminal Neu5Acalpha2-3Gal sequence but not the Neu5Acalpha2-3 linkage on the inner Gal of GM1a, which is a ganglioside that has the gangliotetraose chain (Galbeta1-3GalNAcbeta1-4- (Neu5Acalpha2-3)Galbeta1++ +-4Glcbeta1-Cer). The sialidase hydrolyzed the Neu5Ac on the inner Gal of GM2, which had a shorter gangliotriose chain. GM4, which had the shortest chain (Neu5Acalpha2-3Galbeta1-Cer) of the gangliosides, had a lower substrate specificity. The N1 and N2 sialidase subtypes of the human influenza A virus had no significant variation in their substrate specificity for the gangliosides. Analysis of 11 synthetic gangliosides, which contained various ceramide or sialic acid moieties, demonstrated that A/Aichi/2/68 (H3N2) sialidase recognized the ceramide and sialic acid moiety and the length and structure of the sialyl sugar chain.      

Purification and characterization of a membrane-bound sialidase from pig liver

A membrane-bound sialidase in pig liver microsomes was solubilized with a nonionic detergent, IGEPAL CA630, and purified to homogeneity by sequential chromatographies on SP-Toyopearl, Butyl-Toyopearl (1st), SuperQ-Toyopearl, Hydroxyapatite, Butyl-Toyopearl (2nd), GM1-Cellulofine affinity, and sialic acid-Cellulofine affinity columns. The molecular weight of the purified enzyme was estimated to be 57 kDa on SDS-PAGE. The pH optimum was 4.8 for the activity measured using 4-methylumbelliferyl-alpha-N-acetylneuraminic acid (4MU-Neu5Ac) as the substrate. The enzyme activity was inhibited by 2-deoxy-2,3-dehydro-N-acetylneuraminic acid, iodoacetamide and p-chloromercuribenzoic acid. While the enzyme could effectively hydrolyze 4MU-Neu5Ac, it failed to significantly cleave a sialic acid residue(s) from sialyllactose, glycoproteins or gangliosides at pH 4.8. These results suggest that the purified enzyme is a novel sialidase with a substrate specificity distinct from those of known membrane-bound sialidases in mammalian tissues.     

PHYSICO-CHEMICAL CHARACTERISTICS AND REGIONAL DISTRIBUTION STUDIES OF GP-350, A SOLUBLE SIALOGLYCOPROTEIN FROM BRAIN

The apparent molecular weight of GP-350, a sialoglycoprotein from calf and rat brain, has been determined by SDS-polyacrylamide gel electrophoresis. The electrophoretic mobility corresponds to the mobility of a polypeptide with a molecular weight of 11,600 ± 200. On this basis it can be calculated that only one sialic acid residue is present/GP-350 molecule. From isoelectric focusing experiments it appeared that the isoelectric point of GP-350 is about 2. The determination of the amide content of the polypeptide chain showed that out of 22.0 acidic amino acid residues of glutamic acid and aspartic acid, only 4.9 residues are amidated. The total amount of the basic amino acid residues lysine and arginine, is 6.5. So, per molecule GP-350 10.6 acidic amino acid residues are not counteracted by basic amino acid residues. The surplus of the acidic amino acid residues as well as sialic acid result in the pronounced acidic character of GP-350. This fact is supported by the electrophoretic experiments. The carbohydrate-polypeptide linkage type has been studied by alkaline sodium borohydride treatment. Two thirds of all the galactosamine was destroyed, whereas the amount of glucosamine remained the same. Amino acid analysis indicated a decrease in serine and threonine with a concomitant small increase in alanine. These data point to the occurrence of linkages between the carbohydrate chain and the polypeptide core of the galactosamineserine or –threonine type. Per molecule GP-350 about two residues of galactosamine are destroyed, indicating that two carbohydrate chains of this binding type are present. Only one of these chains can be terminated by a sialic acid residue. The other carbohydrate chain may be terminated by fucose. Regional distribution studies showed the presence of GP-350 in all brain areas studied; in relatively large amounts in the regions rich in ganglia such as caudate nucleus, cerebellar grey matter, pons and medulla oblongata, and in relatively small amounts in the regions poor in ganglia such as corpus callosum, cerebral white, cerebral grey and cerebellar white matter. GP-350 is also present in the pituitary gland. In the cerebrospinal fluid a glycoprotein is present with the same electrophoretic mobility as GP-350. However, this glycoprotein gave no precipitin reaction with GP-350 specific antiserum. Moreover, the amino acid composition was quite different from that of GP-350. Subcellular distribution study revealed that GP-350 is present in the soluble cell fraction and in the synaptosomal membrane fraction, whereas it is absent from the purified nuclei, mitochondria, myelin, and also from the microsomal fraction.     

Enzymatic properties of sialidase from the ovary of the starfish, Asterina pectinifera

A sialidase [EC 3.2.1 18] was isolated and highly purified from the ovary of the starfish, Asterina pectinifera, and its enzymatic properties were compared with those of human placental sialidase. The final preparation gave one broad protein band corresponding to sialidase activity on polyacrylamide gel electrophoresis. The molecular weight of the enzyme was 360 000 by HPLC on Sigma Chrome GFC-1300 and Sephadex G-150 column chromatography, and 55 000 by SDS-PAGE, suggesting the presence of a hexamer in the native protein. The optimum pH was between 3.0 and 4.0, and the enzyme liberated sialyl residues from the following compounds: α(2-3) and α(2-6) sialyllactose, colominic acid, fetuin, transferrin, gangliosides GM₃, GD_1a and GD_1b. The enzyme was strongly inhibited by 4-aminophenyl and methyl thio-glycosides of sialic acid, but not by those glycosides of 5-amino sialic acid or sialic acid methyl ester. The enzyme was also highly inhibited by sulfated glucan and glycosaminoglycans. The substrate specificity and the effects of inhibitors on starfish sialidase were very similar to those of human placental sialidase.     

THE DISTRIBUTION OF LIPIDS IN THE HUMAN NERVOUS SYSTEM-V. GANGLIOSIDES AND ALLIED NEUTRAL GLYCOLIPIDS OF INFANT BRAIN

—Gangliosides and allied neutral glycosylceramides were isolated from human infant (2-24 months of age) cerebral cortex and white matter. The individual glycolipids were separated quantitatively by a combination of column and thin-layer chromatographic methods on silica gel, DEAE-cellulose and Sephadex G-25. In cerebral cortex G_D1a and G_M1 were the major fractions and constituted more than 70 per cent of the total gangliosides. The concentrations of neutral glycolipids, except for galactosylceramides, were very low: lactosylceramide and glucosylceramide comprised 30 and 5 nmol/g wet weight, respectively. In white matter their concentrations were 10 times higher. The ganglioside concentration was only 50 per cent of that in cerebral cortex: the difference was accounted for mainly by the much lower content of the major di- and trisialogangliosides. Stearic acid was the predominant fatty acid of all brain gangliosides. G_M3, and G_D3 had a considerable content of the very long-chain fatty acids, C₂₂-C₂₄, particularly in the white matter. Glucosylceramide and lactosylceramide had almost identical fatty acid patterns between each other in cerebral cortex and white matter. In the cerebral cortex stearic acid and in the white matter the very long-chain acids predominated. d20:1 Sphingosine comprised more than 20 per cent of total sphingosine in all the gangliosides of the G_l- and G₂-series. G_M3, and G_D3 like lactosylceramide contained significantly less of d20:1 sphingosine. The findings suggest the existence of separate compartments for the biosynthesis of the gangliosides. Glucosylceramides and lactosylceramides of white matter have the same ceramide composition as the galactosylceramides with normal fatty acids and are thus unlikely to be intermediates in the metabolism of the major brain gangliosides which have a completely different fatty acid composition.     

Substrate specificities of a bacterial sialidase and rat liver ganglioside GM3 sialyltransferase

Sialidases cleave off sialic acid residues from the oligosaccharide chain of gangliosides in their catabolic pathway while sialyltransferases transfer sialic acid to the growing oligosaccharide moiety in ganglioside biosynthesis. Ganglioside G_M3 is a common substrate for both types of enzymes, for sialidase acting on ganglioside G_M3 as well as for ganglioside G_D3 synthase. Therefore, it is possible that both enzymes recognize similar structural features of the sialic acid moiety of their common substrate, ganglioside G_M3. Based on this idea we used a variety of G_M3 derivatives as glycolipid substrates for a bacterial sialidase (Clostridium perfringens) and for G_D3 synthase (of rat liver Golgi vesicles). This study revealed that those G_M3 derivatives that were poorly degraded by sialidase also were hardly recognized by sialyltransferase (G_D3 synthase). This may indicate similarities in the substrate binding sites of these enzymes.     

Visualization of Sialidase Activity in Mammalian Tissues and Cancer Detection with a Novel Fluorescent Sialidase Substrate

Sialidase removes sialic acid from sialoglycoconjugates and plays crucial roles in many physiological and pathological processes. Various human cancers express an abnormally high level of the plasma membrane-associated sialidase isoform.Visualization of sialidase activity in living mammalian tissues would be useful not only for understanding sialidase functions but also for cancer diagnosis. However, since enzyme activity of mammalian sialidase is remarkably weak compared with that of bacterial and viral sialidases, it has been difficult to detect sialidase activity in mammalian tissues. We synthesized a novel benzothiazolylphenol-based sialic acid derivative (BTP-Neu5Ac) as a fluorescent sialidase substrate. BTP-Neu5Ac can visualize sialidase activities sensitively and selectively in acute rat brain slices. Cancer cells implanted orthotopically in mouse colons and human colon cancers (stages T3-T4) were also clearly detected with BTP-Neu5Ac. The results suggest that BTP-Neu5Ac is useful for histochemical imaging of sialidase activities.     

Effect of neurotoxic divalent cations on the activity of the intrinsic nerve ending membrane-associated sialidase of bovine brain

Exposure to Hg²⁺ below 10 M destroys synaptosomal membrane-associated sialidase of bovine brain in situ. Inhibition by Cu²⁺ occurs only at relatively higher concentrations, and is demonstrable after the synaptosomal nembrane preparation has been presaturated with Cu²⁺. Pb²⁺ does not inhibit enzymatic activity. Hg²⁺ does not exert a significant effect on the free energy of association of monomeric brain gangliosides into aggregates, or on the stability of the aggregate forms, as estimated by ultracentrifugal analysis of the ion-independent moment of ganglioside micelles as a function of concentration. Hg²⁺ inhibits synaptic membrane sialidase acting both in situ on the native sialocompounds in the membrane, or on exogenous ganglioside. Kinetic analyses of the exogenous activity in membranes exposed to Hg²⁺ reveal loweredV _max values but no substantial change inK _m for synaptosomal membrane gangliosides. These findings suggest that the powerful inhibitory effect exerted by Hg²⁺ on nerve ending membrane sialidase is enzyme directed, not substrate directed. It may be postulated that part of the neurotoxic effect of low levels of Hg²⁺ stems from an interference with synaptic metabolism by the destruction of membrane-associated sialidase. This enzyme can serve the purpose of modulation of synaptic negative charge density by releasing bound, strongly anionic, sialic acid from highly concentrated sialocompounds in the membrane.Author to whom correspondence should be sent.     

Molecular cloning and expression of mouse brain sialidase.

Sialidase (EC 3.2.1.18) catalyzes the release of sialic acid from sialo-oligosaccharides, gangliosides, or sialo-glycoproteins. In this investigation, we cloned a novel cDNA for mouse brain sialidase and expressed the cDNA in COS-7 cells. This 1,699 bp cDNA codes for a 41.6 kDa protein consisting of 372 deduced amino acid residues. In COS-7 cells transiently transfected with the cDNA, a 250-fold increase was observed in specific activity toward 2'-(4-methylumbelliferyl)-alpha-D-N-acetylneuraminic acid. Similarity searches of the nonredundant GenBank peptide sequence database by the PSI-BLAST program identified rat, hamster, human, and bacterial sialidases homologous to this mouse brain sialidase. Amino acid sequence identities to rat and hamster sialidases (84% and 77%, respectively) suggest that this form of sialidase is conserved in rodents. Sequence identities to human and mouse lysosomal sialidases (30% and 28%, respectively) indicate that the mouse brain sialidase is distinct from the lysosomal enzyme. Mouse brain sialidase has two amino acid sequence motifs common to bacterial sialidases: the 'F/YRIP' motif and the 'Asp-box' motif. The 'F/YRIP' motif is present near the N terminus while two 'Asp-box' motifs are present downstream.     

Regional, cellular and subcellular distribution of calcium-activated cyclic nucleotide phosphodiesterase and calcium-dependent regulator in porcine brain.

Abstract— Cyclic nucleotide phosphodiesterase activity (calcium-dependent and calcium-independent) and CDR (calcium-dependent regulator protein of phosphodiesterase) are present in all ten brain regions examined, with the specific activity of both being highest in areas predominating in grey matter and lowest in areas consisting largely of white matter. Fractionation of cerebral cortex into neuronal and glial perikarya shows that cyclic nucleotide phosphodiesterase and CDR are present at approximately equal levels in both cell types. Subcellular fractionation reveals that calcium-sensitive enzyme specific activity, as well as CDR are highly localized in the 100,000 g supernatant fraction. Regional, cellular and subcellular distribution studies indicate that the distribution of CDR closely parallels the distribution of enzyme specific activity, suggesting that the levels of cyclic nucleotide phosphodiesterase and CDR may be synchronized.     

Content and accessibility of sialic acid on the surface of rat hepatocytes during primary culture

The content and accessibility of terminal sialic acid and galactose residues of rat hepatocytes in primary culture were determined by in situ labeling using either periodate or sialidase/galactose oxidase treatment followed by sodium borotritiide reduction. Rat erythrocytes which were used for comparison showed a strongly enhanced tritium incorporation into galactose after sialidase treatment. In contrast, with freshly prepared rat hepatocytes only a small amount of galactose labeling was achieved after sialidase treatment. The amount of galactose labeled following sialidase treatment increased with time in culture up to day 6 and roughly paralleled the increase of the total sialic acid content. Major changes of sialic acid-containing glycoconjugates were restricted to the gangliosides. There was a transient drop in surface labeling of ganglioside-associated sialic acid on the first day in culture. The specific radioactivity of the in situ-tritiated ganglioside-sialic acid also fell by 50% in this period. Between day 2 and 4, there was an increase in gangliosidesialic acid labeling but the specific radioactivity of the sialic acid remained constant. This indicates that newly synthesized gangliosides but not the preexisting ones were accessible to periodate oxidation. The data allow conclusions about turnover and topology of the sialic acid-containing glycolipids.