95;:qu FINES: 2:: per day per :‘w; §_ETU;‘.NLN3 LIBRH-ZY «NEEDLE: Place in bcuk return to N23949: charge from Graham recurc‘s BIOCHEMICAL AND IMMUNOLOGICAL STUDIES ON FAMILIAL ERYTHROPHAGOCYTIC LYMPHOHISTIOCYTOSIS By Clifford Gregory Wong A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Biochemistry 1980 ABSTRACT BIOCHEMICAL AND IMMUNOLOGICAL STUDIES ON FAMILIAL ERYTHROPHAGOCYTIC LYMPHOHISTIOCYTOSIS By Clifford Gregory Wong Children afflicted with familial Erythrophagocytic Lymphohistiocytosis (F EL), an inherited Childhood disorder, were recently discovered to have defects in both humoral and cellular immunity, and the occurrence of a plasma inhibitor of i_n 21E? lymphocyte blastogenesis. An autopsy examination of the liver of one FEL patient revealed an abnormally high amount of uncharacterized lipid material accumulating locally in the hepatocytes and in the infiltrating macrophages. Lipid analysis of the FEL liver was performed by isolation and separation of the total lipids by Folch solvent partitioning, silicic acid column chromatography, and analytical thin-layer Chromatography (TLC). The levels of neutral lipids were found to be increased 2-fold over normal levels, as determined by gravimetric, colorimetric, and gas—liquid chromatographic methods. This increase was primarily in triglyceride content with a modest increase in total cholesterol lipids. The cholesterol ester content in total cholesterol lipids was markedly reduced from 2996 (in normal liver) to 296. No alteration in the individual triglyceride species was found. Analysis of the neutral glycolipids and phospholipids by thin-layer Chromatography revealed no abnormalities in their TLC patterns. The major lipid abnormality in F EL was discovered in the water-soluble (ganglioside) lipid fraction where lipid-bound sialic acid, as determined by the Clifford Gregory Wong resorcinol colorimetric assay for sialic acid, was over 11-fold higher than normal. The thin-layer chromatographic analysis of the total ganglioside fraction (Folch upper phase) revealed a general increase in all ganglioside species with an apparent lOO-fold increase in lipid-bound sialic acid material with TLC mobility similar to GMZ‘ Lysosomal glycosylhydrolase assays were Conducted in liver, leukocytes, and fibroblasts of normal and FEL patients in order to determine whether the apparent ganglioside accumulation in FEL liver was due to a lysosomal enzyme deficiency. No enzyme deficiencies were found in FEL leukocytes or fibroblasts, although B-galactosidase activity in the one FEL liver examined was significantly reduced (25% of normal levels). The nature of the B-galactosidase activity decrease was investigated by means of mixing experiments with normal and FEL liver homogenates, cellulose acetate electrophoresis, and heat inactivation studies. These studies revealed the absence of a soluble inhibitor, no alteration in electrophoretic mobility of the FEL B-galactosidase isoenzymes, and the absence of a residual heat-stable isoenzyme that might have been responsible for the lowered activity in FEL liver. It was concluded that the reduction in B- galactosidase activity was not due to an enzyme defect, was only localized in liver, was not a generalized enzyme deficiency, and thus, was not the primary metabolic defect. FEL and normal liver gangliosides were preparatively isolated by chloroform:methanol solvent extraction, DEAE—Sephadex anion exchange column chromatography, Iatrobead silicic acid column chromatography, and thin-layer chromatography for immunological testing. Two putative ganglioside fractions from normal liver and one from FEL liver were found to inhibit (at 2 pg/ml) i_n 1132 antigen-stimulated lymphocyte mitogenesis, but not lectin-stimulated (Con A) mitogenesis. Initial characterization of these gangliosides by combined gas-liquid Clifford Gregory \Vong chromatography-mass spectrometry revealed a probable lactotetraose or lactoneotetraose core sequence for the oligosaccharide moiety of the glycolipid. Initial studies with the FEL plasma inhibitor did not reveal the presence of an accumulating ganglioside, although immunosuppressive activity was found in a fraction isolated by preparative TLC with similar mobility to the FEL liver ganglioside inhibitor. These studies have demonstrated for the first time highly specific and potent immunosuppressive lipid-bound sialic acid compounds in FEL and normal liver with chromatographic and chemical properties similar to gangliosides that may be involved in the pathogenesis of an inherited immunological disorder. Dedicated to my parents: to my mother, for her faith and love; and to my father, in memory of ii ACKNOWLEDGEMENTS I wish to sincerely thank all of my friends, without whose aid this work could never have been completed: Drs. Bruce Macher, Bader Siddiqui, and John Klock, for their physical and moral support; my colleagues in the lab, for their invaluable advice and assistance; Charles Caldwell, for his much appreciated artistry; and especially, Drs. Charles C. Sweeley and Stephan Ladisch, for their long-suffering patience, incredible faith, and guidance. TABLE OF CONTENTS PAGE LIST OF TABLES .................................... . ........... vi LIST OF FIGURES ............................................... viii LIST OF ABBREVIATIONS ......................................... ix INTRODUCTION .................................................. 1 LITERATURE REVIEW ............................................. 3 I. The Gangliosides ....................................... 3 A. General Background ........................... . ..... 3 B. Nomenclature ....................................... 5 C. Occurrence ......................................... 7 D. Methodology-Isolation Techniques ................... 10 E. Methodology-Characterization Techniques ............ 15 F. Glycosphingolipid and Ganglioside Anabolism ........ 20 G. Glycosphingolipid and Ganglioside Catabolism ....... 22 II. Immunoregulation by General Lipids and Lipoproteins.... 30 III. Immunological Roles for Gangliosides ............... .... 32 A. Receptors and Antigenic Cell Surface Markers ....... 32 B. Immunoregulation by Gangliosides ................... 33 MATERIALS AND METHODS.. ....................................... 36 Materials ................................................. 36 Methods ................................................... 40 I. Lysosomal Enzyme Studies ............................... 40 A. Preparation of Human Tissues and Cells ........ ..... 40 B. Lysosomal Glycosylhydrolase Assays ................. 41 II. Lipid Isolation and Characterization ................... 44 A. Lipid Extractions and Chromatography ............... 44 B. Quantitation and Characterization of Neutral Liver Lipids ............................................. 45 C. Isolation and Characterization of Gangliosides ..... 48 D. Oligosaccharide Analysis ........................... 52 E. Ganglioside Immunosuppression of Lymphocyte Blastogenesis ...................................... 53 iv PAGE RESULTS ....................................................... 55 I. Lysosomal Enzyme Studies ............................... 55 A. Liver Lysosomal Glycosylhydrolases. ............... 55 B. Leukocyte Lysosomal Glycosylhydrolases ............. 55 C. Fibroblast Lysosomal Glycosylhydrolases ..... . ...... 50 D. Studies on the Nature of the Liver Lysosomal B-Galactosidase Deficiency ................ . ........ 50 II. Liver Lipid Studies .................................... 75 A. Lipid Composition of Folch Extracts ................ 75 B. Ganglioside Studies ................................ 85 III. Liver and Spleen Oligosaccharides ...................... 107 DISCUSSION .................................................... 110 1. Identification of Liver Storage Product ................ 110 A. Lipid Analysis of FEL Liver and Plasma ............. 110 B. Oligosaccharide Analysis of FEL Liver and Spleen... 114 II. The Search for an Enzymatic Basis in FEL as a Storage Disease ................................................ 115 A. Lysosomal Enzymes in Liver, Leukocytes, and Fibroblasts ........................................ 115 B. Anabolic Defects in FEL ............................ 119 III. Immunosuppressive Activity and Structure of Liver Gangliosides ........................................... 120 IV. Identification of the FEL Plasma Immunosuppressive Factor .................................. . ........... ... 125 SUMMARY .................................................... ... 129 APPENDICES A. Case Presentation of the FEL Patient ................... 132 B. Analysis and Structural Characterization of Amino Sugars by Gas-Liquid Chromatography and Mass Spectrometry ........................................... 133 BIBLIOGRAPHY .................................................. 144 TABLE 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. LIST OF TABLES . Bacterial Toxins and their Ganglioside Receptors ......... . Oligosaccharide Structural Families in Glycosphingolipids . Structure and Occurrence of Human Gangliosides ........... . The Glycosphingolipidoses ................................ . The Mucopolysaccharidoses and Mucolipidoses .............. . Liver Lysosomal Glycosylhydrolases-I ..................... . Liver Lysosomal Glycosylhydrolases-II .................... . Leukocyte Lysosomal Glycosylhydrolases of the FEL Patient's Relatives ...................................... . Leukocyte Lysosomal B-Galactosidases of the FEL Patient's Relatives ................................................ Leukocyte Lysosomal Glycosylhydrolases From Other FEL Patients ................................................. Fibroblast Lysosomal Glycosylhydrolases .................. Total Liver Lipid Composition ............................ Neutral Lipid Fraction Composition ....................... Distribution of Triglycerides in Liver ................... Fatty Acid Composition of Liver Triglycerides ............ Total NeuNAc Content in Human Liver ...................... Quantitation of Normal Liver Ganglioside Fractions ....... Quantitation of FEL Liver Ganglioside Fractions .......... Carbohydrate Composition of Liver Immunosuppressive Fractions ................................................ Major Fatty Acid Composition of Liver Immunosuppressive Fractions ................................................ vi PAGE 25 26 56 57 58 59 59 61 76 76 83 84 84 9o 91 100 100 PAGE TABLE 21. Gangliosides of Normal and FEL Plasmas .................. 106 22. Quantitation of Liver Fatty Acids ....................... 112 23. Ion Masses Used for Mass Chromatographic Analysis ....... 122 vii LIST OF FIGURES FIGURE PAGE 1. General Pathways for Ganglioside Biosynthesis ........... 21 2. Enzyme Defects in Ganglioside Metabolism .............. .. 24 3. Liver B—Galactosidase Activity vs. pH ................... 63 4. Normal and FEL Liver B-Galactosidase Mixing Experiments. 65 5. Normal and FEL Liver B-Galactosidase Mixing Experiments with Increasing Homogenate Concentrations ............... 68 6. Heat Lability of Liver B-Galactosidases ................. 7O 7. Cellulose Acetate Electrophoresis of Liver B-Galactosidases ........................................ 72 8. Effect of Exogenous NrAcetylneuraminic Acid on Liver s—Galactosidase Activity ................................ 74 9. Thin-Layer Chromatography of Liver Neutral Lipids ....... 78 10. Thin-Layer Chromatography of Liver Neutral Glycosphingolipids ...................................... 80 11. Thin-Layer Chromatography of Liver Phospholipids ........ 82 12. Ganglioside Fractions Isolated from Normal Liver ........ 87 13. Ganglioside Fractions Isolated from FEL Liver ........... 89 14. Thin-Layer Chromatography of Total Liver Gangliosides... 94 15. Immunosuppressive Activities of Isolated Normal Liver Ganglioside Fractions ................................... 96 16. Immunosuppressive Activities of Isolated FEL Liver Ganglioside Fractions ................................... 98 17. Human Plasma Gangliosides-I ............................. 103 18. Human Plasma Gangliosides-II ............................ 105 19. Oligosaccharides from Human Liver and Spleen ....... ..... 109 viii GLC-MS GLC TLC Con A pNP— 4-MU 4-MU- FEL PHA PWM SKSD PBL NeuNAc NeuNGl Gal Glc Man GlcUA GlcNAc Cer TMS PAS GL—I GL-Z GL-3 GL-ll GL-5 LIST OF ABBREVIATIONS Miscellaneous gas-liquid chromatography-mass spectrometry gas-liquid chromatography mass spectrometry thin-layer chromatography concanavalin A p-nitrophenyl- red blood cell ll—methylumbelliferone 4—methylumbelliferyl- Familial Ertherphagocytic Histiocytosis phytohemagglutinin pokeweed mitogen Streptokinase-streptodornase peripheral blood lymphocytes N—acetylneuraminic acid, a sialic acid N—glycolylneuraminic acid galactose glucose mannose fucose glucuronic acid N-acetylglucosamine ceramide trimethylsilyl periodic acid-Schiff Reagent Glycosphingolipids glucosylcera mide lactosylceramide Gal (1 1+4GalB 1+4Gch 1+ l'Cer GalNACB1+3Galal+4GalBl+4GlCB1+1'Cer (globoside) GalNAc a1+3GalNACBI+3Galo1+4GalBl+4Gch1+l'Cer (Forssman Hapten) ix INTRODUCTION Familial Erythrophagocytic Lymphohistiocytosis (FEL), first described by Farquhar and Claireaux (I), is an inherited, fatal childhood disorder characterized by anorexia, hepatosplenomegaly, jaundice, thrombocytopenia, liver dysfunction, and hyperlipidemia (2-4). A striking autopsy feature is a widespread histiocytic infiltration of liver, spleen, lymph nodes, bone marrow, lungs, and of the gastrointestinal, genitourinary, and central nervous systems, accompanied by erythrophagocytosis at the infiltration sites. The age at onset ranges from 2 weeks to 7 years, and the average survival time is generally 6 weeks from the onset of the illness with the children succumbing either to bleeding, sepsis, or lymphocytic meningitis. The disorder is believed to be inherited as an autosomal recessive trait (5). In a recent study of 4 FEL children, Ladisch gt a_l. (6) found an immunological deficiency syndrome which included defects in both humoral and cell-mediated immunity and a plasma inhibitor of 13 gig}; lymphocyte blastogenesis. The humoral immunity defects included low antibody titers after previous immunizations and an impaired ability to respond to primary immunizations. However, immunoglobulin levels (IgA, IgM, IgG) were normal in these patients. Although a normal proportion of T and B lymphocytes were present in these patients, defects in cellular immunity were shown by anergy and by the inability of the patients' lymphocytes to proliferate in response to specific antigens; yet, responses to lectin mitogens and allogeneic cells (mixed leukocyte cultures) were normal. The level of inhibitory activity of FEL plasmas appeared to 1 2 be proportional to the degree of hyperlipidemia (measured in triglyceride levels) present. Localized fatty changes were observed in the liver autopsy examination of one FEL patient along with an accumulation of uncharacterized lipid in the infiltrating histiocytes. At present, the etiology of this disorder remains an enigma. The question on whether the immunodeficiency is the primary defect in FEL or merely secondary to another primary pathogenic mechanism remains unanswered and is the focal point of this study. From the discoveries of an immunodeficiency syndrome, the inheritance of the disorder as an autosomal recessive trait, the localized fatty changes in liver along with the accumulation of lipid in the infiltrating macrophages, and a Circulating inhibitor of. in vitro lymphocyte blastogenesis whose potency was apparently related to the degree of hyperlipidemia, a rationale was established for the investigation of lipids in FEL liver and plasma to determine the possible presence of a lipid or glycolipid storage disease, and the effects of such a possible storage product on the observed immunodeficiency. If a storage problem was to be discovered, a search for the metabolic basis of the accumulation, particularly in the likelihood of a catabolic defect, would be conducted along with the investigation of the liver lysosomal enzymes. LITERATURE REVIEW I. The Gangliosides A. General Background Glycosphingolipids, first discovered by Thudichum (8) in 1874, are amphipathic molecules composed of three basic components: a long-chain base, a fatty acid moiety, and a carbohydrate moiety which may vary in length from a single monosaccharide to a complex, branched oligosaccharide of 30—60 glycose units (9-10). The principal long-chain base of most mammalian glycosphingolipids is 4-sphingenine (commonly called sphingosine), to which saturated or unsaturated fatty acids and their a-hydroxy derivatives, varying in length from 14 to 26 carbon units, are covalently joined via amide linkages. The long-chain base and fatty acid components together constitute the hydrophobic portion (ceramide) of the glycosphingolipid that is anchored in bilayer membranes. Carbohydrate units constitute the hydrophilic portion of the glycosphingolipids and are covalently joined to the C-1 hydroxyl group of the ceramide molecule by glycosidic linkages. The carbohydrate moiety is the major determinant that expresses the wide structural and functional diversity of the glycosphingolipids. Glycosphingolipids can be divided into three main groups as determined by the chemical nature of their carbohydrate moiety: neutral (i.e. non-polar, non- acidic) glycosphingolipids, sulfatoglycosphingolipids (commonly called sulfatides), and the gangliosides. The neutral and sulfated glycosphingolipids have been reviewed extensively elsewhere (ll-15), and will not be discussed further in depth here. 4 Gangliosides are terrestrial and marine animal glycosphingolipids containing the unique acidic carbohydrate, sialic acid. They are principally located in the outer surface of plasma membranes of mammalian cells and, together with sialylated glycoproteins, are the main source of the negative charge observed on cell surfaces. Originally identified by Klenk (16) in 1935 in his investigations of brain lipids in patients with Tay-Sachs Disease and Nieman-Pick Disease, the sialic acid- containing glycosphingolipids were soon discovered in normal brain (17) and were consequently named gangliosides by Klenk (18) because they were believed to be localized specifically in ganglion cells. Yamakawa and Suzuki (19) later reported their occurrence in extraneural tissues and fluids, after which, extensive research into ganglioside structure and function has evolved. The concentration of gangliosides in all animal species is highest in brain, where they were first discovered. Since they are located at the cell surface and also in the synaptic membranes of the central nervous system, their function is of particular interest to neurochemists, especially in relation to the elucidation of synaptic transmission mechanisms at the molecular level. Metabolic disorders of gangliosides, such as Tay-Sachs Disease and GM 1 gangliosidosis, are of special interest in regard to their relationship to developmental neurobiology and mental retardation. In extraneural tissues, gangliosides are believed to occur primarily on the cell surface where they are assumed to be important in intercellular recognition. They may also be involved in a variety of cell surface functions such as the regulation of metabolism, growth, differentiation, behavior, malignant transformation, viral infection (20-22); and as surface receptors for protein hormones (thyroid stimulating hormone, luteinizing hormone, human Chorionic gonadotrOpin) (23-25), Sendai virus (26-27), interferon (28-29), serotonin (30), and 5 bacterial toxins, e.g. staphylococcal a toxin (31), botulinum toxin (32), tetanus toxin (33-34), and especially cholera toxin (35-38). A list of the bacterial toxins and their putative ganglioside receptors can be found in Table l (21). Table 1. Bacterial Toxins and their Ganglioside Receptors Bacterial Toxin Receptor cholera toxin GM 1 tetanus tox1n GD Ib’Gle botulinum toxin trisialoganglioside staphylococcal a toxin sialylparagloboside sialylparagloboside: NeuN ACBZ" BGalB 1+ 4G1CNACBI+ 363181" 401C81+ l'Cer Sialic acid, is a generic term that encompasses a variety of derivatives of neuraminic acid (5-amino—3,5-dideoxy-Q-glygrg-D-gafigt3—2-nonulopyranoso-nic acid), which usually occur naturally as either Ill-acetylneuraminic acid or N- glycolylneuraminic acid. The Q-acetyl derivatives of the sialic acids have also been found in bacteria and animals. Only the _N_-acetylneuraminic acid form of sialic acid has been found in human tissues, although _N-acetyl-ll-Q- acetylneuraminic acid has been reported in human urine (39) and bile (40). The sialic acid residues in gangliosides are joined to the oligosaccharide moiety by a- ketosidic linkages (CZ->3 or u2+6) to galactose residues, and to each other via o2+8 linkages (15). B. Nomenclature Adoption of an appropriate nomenclature descriptive of the carbohydrate moiety of glycosphingolipids has been a confusing and complex issue, as complex as the structural nature of the oligosaccharide moieties themselves. Most early systems of nomenclature were developed to designate individual components of 6 human brain gangliosides which were observed in thin-layer or paper chromatographic systems. The most commonly used system for gangliosides was that of Svennerholm (41) which was fairly easy to remember and useful for description of the major brain gangliosides. However, due to the discoveries of more complex gangliosides and the extraneural gangliosides whose carbohydrate sequences differed in isomeric glycosidic linkages as well as in their amino sugar components, Svennerholm's nomenclature system became less appropriate and extremely limited in its description of the more structurally complex gangliosides. Wiegandt (42) developed an alternative system which employed the use of trivial names to denote the different families of the oligosaccharides that have been found to occur in glycosphingolipids. Borrowing partly from Wiegandt, the IUPAC- IUB Commission on Biochemical Nomenclature (43) and the Commission on Nomenclature of Organic Compounds have considered a systematic nomenclature for the various oligosaccharide families which are summarized here in Table 2. A more detailed description of the nomenclature and the use of the prefixes and structure abbreviations is given in a review by Sweeley and Siddiqui (15). Table 2. Oligosaccharide Structural Families in Glycosphingolipids Name (prefix) Structure lacto GalB 1+3GlcNACB 1+3Ga181+4Glc+ lactoneo GalB 1+4GlcNAc8 1+3GalB 1+4Glc+ ganglio GaIB 1+3GalNAcB 1+4Ga18 1+4Glc+ globo GalNAcB 1+3Gal a 1+4GalB 1+4Glc+ globoiso GalNAcB 1+3Gal a 1+3GalB 1+4Glc+ muco GalBI-t3GalB 1+4GalB 1+4Glc+ gala GalNAc?l+3Gal?l+4Galo1+4Gal+ The prefixes given in Table 2 imply the entire structure of the root oligosaccharide up to the tetrasaccharide level, including the carbohydrate 7 sequence as well as the positions and anomeric configuration of the glycosidic linkages. So far, sialic acid sugars have only been discovered on the ganglio, gala, and lactoneo oligosaccharide families or series in human tissue. C. Occurrence Due to the large number of gangliosides (over 40) that have been discovered, only the extraneural gangliosides and the major brain gangliosides of human origin will be reviewed here. A more comprehensive review of all human neural and extraneural gangliosides and glycosphingolipids can be found elsewhere (15,44). The major distinguishing feature of extraneural gangliosides is the presence of _N_-acetylglucosamine as the amino sugar component in the oligosaccharide moiety, whereas neural gangliosides only contain N—acetylgalactosamine. The possible occurrence of _N_-acetylglucosaminc-containing gangliosides was first noted by Klenk and Lauenstein (45) in 1952 in a glycolipid hydrolysate obtained from bovine erythrocytes. Yamakawa _e_t -a_l. (46) subsequently reported the presence of GlcNAc in human, sheep, goat, rabbit, and bovine erythrocyte glycosphingolipids, and found glucosamine-containing gangliosides in their partially-purified glycosphingolipid fractions (47). The first isolation of a homogenous glucosamine- containing ganglioside was reported by Kuhn and Wiegandt (48) in 1964. With this final proof of the existence of glucosamine-containing gangliosides, a new possibility for ganglioside structures different from those of the major neural gangliosides was established. In Table 3, the structure and occurrence of selected, known human brain gangliosides, in increasing order of complexity up to the trisialogangliosides, and extraneural gangliosides are presented. The fucogangliosides are of particular interest because of their serological and possible immunological properties. Glycolipid blood group antigenic activities 88 5an SEE. $8 5an 58:: A383 Eoifiéficn «:33 as 5...... :3 sea :3 52.. am. C 33 883%.:an R m; 3 883.565 a3 :85 2 wnwnv mcoo£>.:_mcn ARAB Scoumafibuflifimun 234$ mouauoefiacoabofificmmfi 8n.m$ caoxEEfiE venom van—I300 65-50 25 n—HU £00 200 :20 Nov ~20 moo n20 s50 .50: +3201. ~2u< 2:025 N+wu 8.50m ma; muNeuNAc-GaI-Glc-Cer Su lfatide ‘//t GM3 aGaI-Gal-Glc-Cer GlcNAc-Gal-Glc-Cer l i Globo Series Ga nglio Series Neutral Ga ngliosides Glycosphingolipids Gal31~3G|cNAc-Gal-Glc-Cer GalBIMGICNAc-Gal-Glc-Cer Lacto Series Lactoneo Series Complex Complex Gangliosides Glycos hingolipids . Glycosphin olipids fiuco) ' (fucoi Fuco-gangliosides Figure 1. General Pathways for Ganglioside Biosynthesis 22 anabolism has been done by others (130,144-145). Lactosylceramide is the critical branching point for the biosynthesis of a number of glycolipid families. The addition of either a sulfate group, u-galactosyl residue, B-N-acetylglucosaminyl residue, or sialyl residue will determine the metabolic pathway and fate for the individual lactosylceramide molecule. The glycosylation of lactosylceramide appears to be primarily tissue-specific rather than substrate-specific. G. Glycosphingolipid and Ganglioside Catabolism Glycosphingolipids are sequentially catabolized by exoglycosylhydrolases, and there is no direct evidence for the existence of endoglycosylhydrolase catabolism in man analogous to glycoprotein or glycosaminoglycan degradation (146), although bacterial endoglycosylhydrolases have been used in structural studies of human glycosphingolipids. Evidence for the sequential exoglycosylhydrolase mode for glycolipid catabolism is shown in the various glycosphingolipidoses by the storage of specific glycolipids in lysosomes in which a specific exoglycosylhydrolase is defective. Since glycolipid glycosylhydrolases have acidic pH optima and the glycosphingolipidoses are lysosomal storage disorders, it is assumed that glycolipids, as well as glycoproteins and glycosaminoglycans, are normally catabolized within lysosomes. Glycoprotein and glycolipid catabolism has been demonstrated with intact lysosomal preparations by various workers (147). Many theories on the mode of action of the lysosomal glycosylhydrolases have been proposed. On the basis of studies on mucolipidosis II ("l-cell" disease), it had been formerly proposed by Neufeld gt gt. (148) that the lysosomal enzymes are normally secreted and taken 14) by adjacent cells by specific receptor-mediated pinocytosis. 23 Recent findings by Natowicz and Sly (149) have shown that 6-phosphomannose residues on lysosomal hydrolases were required for intercellular pinocytosis. They proposed a different hypothesis to explain the biogenesis of lysosomal enzymes and the findings in "I-cell" disease in which the 6-phosphomannose recognition marker in lysosomal hydrolases serves as an intracellular traffic signal that directs high- uptake forms of lysosomal enzymes to the lysosome and prevents their secretion (150). In "I-cell" disease, this marker is presumably missing. Other theories for modes of action are protease activation of latent hydrolases (147), chondroitin sulfate regulation within lysosomes (151), association of the hydrolase in membrane-bound multienzyme complexes (123), and the involvement of glycoprotein co-factors in the specific hydrolysis of glycolipids (152-154). The defects in glycolipid metabolism presented in the glycosphingolipidoses are practically all of lysosomal glycosylhydrolase origin. Excellent reviews on the metabolic defects of the glycosphingolipidoses, the mucopolysaccharidoses, and the glycoprotein storage diseases have been presented elsewhere (147,155-160). The metabolic defects associated with ganglioside metabolism are illustrated in Figure 2. The glucosamine-containing gangliosides have not been implicated in any glycosphingolipidoses nor have they been found to accumulate in other glycosphingolipidoses. The known glycosphingolipidoses associated with defects in either glycosylhydrolases or glycosyltransferases are presented in Table 4 (161). The mucopolysaccharidoses and mucolipidoses are listed in Table 5 (159,162-164). 1) Role of Neuraminidase in Glycosphingolipid Catabolism Most studies of the enzymatic release of sialic acid from gangliosides have involved the use of microbial sialidases. Most human neuraminidase preparations GM GM GM GL-2 Figure 2. 24 Gal 1e30alNAc-Gfl-Glc-Cer NeuNAc B-Galactosidase I deficient in GM1 Gangliosidosis GalNAc 1+4Gtal -Glc-Cer NeuNAc B-Hexosaminidase deficient in GM2 Gangliosidosis NeuNAc -Gal-G|c-Cer UD P-GalNAC:GM 3 GalNAc Transferase deficiency GalNAc-GgI-Glc-Cer 0M2 t NeuNAc Gal-Glc -Cer Enzyme Defects in Ganglioside Metabolism 25 Table 4. The Glycosphingolipidoses Disorder Globoid Cell Leukodystrophy Metachromatic Leukodystrophy Multiple Sulfatase Deficiency Gaucher's Disease Fabry's Disease GMl Gangliosidosis Tay-Sachs Disease Sandhoff's Disease GM3 Gangliosidosis Stored Material Gal-Cer, Gal-sphingosine SOB-Gal-Cer, SO -Ga1-Glc-Cer (squat ides) SOB-G al-Cer, SO -Gal-Glc-Cer choalesterol-sulfate, sulfated glycosamino- glycans Glc-Cer Glc-sphingosine Gal (1 1+4Gal-Glc-Cer Gala. 1+4Ga1-Cer GM , galactose-rich olig1>saccarides, keratan sulfate GM2 GMZ’ asialo-GMZ, GalNAcB 1+4Gal-Gal-Glc- Cer (globoside) GM3, GD3 Enzymic Defect galactosylceramide B-galactosidase arylsulfatase A arylsulfatase A,B,C glucosylcera mide B-glucosidase a-galactosidase A GM ganglioside B-gAlactosidase B-hexosaminidase A B-hexosaminidase A and B, asialo—GM2 UDP-GalNAc:GM GalNac transferaae Table 5. The Mucopolysaccharidoses and Mucolipidoses Disorder Mucopolysaccharidosis 1H (H urler) Mucopolysaccharidosis IS (Scheie) Mucopolysaccharidosis 11 (Hunter) Mucopolysaccharidosis IIIA (Sanfilippo A) Mucopolysaccharidosis IIIB (Sanfilippo B) Mucopolysaccharidosis IIIC (Sanfilippo C) Mucopolysaccharidosis IV (Morquio) Mucopolysaccharidosis VI (Maroteaux-Lamy) Mucopolysaccharidosis VII Mucopolysaccharidosis F Mucolipidosis I Mucolipidosis II (I-cell Disease) Mucolipidosis III 26 Stored Material heparan sulfate, der matan sulfate heparan sulfate, der matan sulfate heparan sulfate, dermatan sulfate heparan sulfate heparan sulfate heparan sulfate keratan sulfate, chondroitin sulfate der matan sulfate dermatan sulfate, heparan sulfate, chondroitin sulfate fucosyl-glycolipid, -glycosaminoglycan, -oligosaccharide uncharacterized glycolipid, mucopolysaccharide, oligosaccharide, sialy-oligosaccha- riduria. same as Mucolipid- osis I same as Mucolipid- osis I Enzymic Defect o-iduronidase o-iduronidase idu ronosulf ate sulfatase sulfamidase a-fl-acetylglucosaminidase a-glucosaminidase? N-acetylhexosaminide 76-503 sulfatase N-acetylhexosaminide Til-SO sulfatase, aryls fatase B B-glucu ronidase a-fucosidase a-neu ra minidase? several lysosomal enzymes low, but high in media, serum same as in Muco- lipidosis II 27 can liberate the external sialic acids from GT la,GD Ia’ and GM3, but not the sialic acid residue of GMl or GMZ’ thus raising the possibility of two different neuraminidases with separate specificities. The issue on the presence of two distinct neuraminidases in human tissues has not been settled due to the lack of homogenous neuraminidase preparations. A purified (3500-fold) neuraminidase preparation (164) has been demonstrated to have activity towards GM2 as well as other monosialogangliosides, polysialogangliosides, and fetuin. However, this enzyme preparation failed to catabolize the stored GM2 in Tay Sachs brain. This failure has been attributed to the fact that the mutant inactive hexosaminidase A binds to the substrate and thus prevents hydrolysis by neuraminidase (165). The activity of the other neuraminidases is unaffected in Tay-Sachs brain and current opinion seems to favor the existence of two distinct neuraminidases. However, the possibility of the requirement of a missing lipid component or glycoprotein activator similar in nature to that found for the catabolism of GMl or GMZ ganglioside by B- galactosidase or hexosaminidase A (152-153) cannot be discounted. This missing component could account for the inability of a neuraminidase to hydrolyze the sialyl residues on the gangliosides. Although neuraminidase will hyrolyze gangliosides in either the dispersed or micellar form, their activities are probably optimal at the plasma membrane level, where they appear to be membrane bound. The membrane-bound preparations of neuraminidases also appear to act preferentially on ganglioside substrates rather than on glycoproteins. Touster gt _a_l. (169) and Schengrund _e_t a_l. (170) had located the neuraminidases and their ganglioside substrates surprisingly on plasma membranes rather than in lysosomes as expected. Schengrund gt gt. (170) also found brain neuraminidase to be localized in synaptic membranes along with gangliosides and 28 their sialyltransferases. Their findings raised the possibility that neurotransmitters could be bound and released by gangliosides following the alternate actions of sialylation and de-sialylation of gangliosides at the synaptic junctions. Presently, no human disease has been attributed to a specific deficiency of glycolipid neuraminidase activity. Neuraminidase deficiency in fibroblasts and leukocytes has been reported in the mucolipidoses with heavy excretions of sialyloligosaccharides in urine (164,166). However, examinations of autopsy solid tissues revealed normal levels of neuraminidase activity (167) and normal levels of gangliosides and protein-bound sialic acid (168). Neuraminidase, therefore, does not appear to-be the key metabolic defect in the mucolipidoses. It must be noted however, that all neuraminidase assays in these studies were conducted with synthetic substrates, neuraminyl-lactose, or fetuin. Ganglioside substrates were never investigated. 2) Role of B-Galactosidase in Glycosphingolipid Catabolism GMI ganglioside, its asialo form, and various oligosaccharides and mucopolysaccharides can be hydrolyzed by a B-galactosidase that can be detected with synthetic substrates. A complete deficiency of B-galactosidase activity for both synthetic and natural substrates is found in GM 1 gangliosidosis (171). There appear to be two other distinct lysosomal B-galactosidases, one specific for only galactosylceramide and galactosyldiglyceride, as demonstrated in Krabbe's disease (172), and one specific for lactosylceramide (173). The activity of GM 1 and synthetic substrate B-galactosidase has been found to be less than 2% of normal in patients with Type 1 (infantile) GM 1 gangliosidosis, whereas in the Type 2 (juvenile) form, the B-galactosidase activity can vary from 5% to 30% of normal (171). These results, along with demonstrations that glycopeptides, oligosaccharides, and keratan sulfate are stored 29 in this disease, suggest that one common B-galactosidase can hydrolyze them all. From starch gel electrophoretic analysis of the B-galactosidase isoenzymes, all 3 isoenzymes (A,B,C) were absent in Type 1 GM 1 gangliosidosis, but only B and C were absent in Type 2. The A isoenzyme, which has the most negative charge, was the major isoenzyme in all tissue studied with more than 70% of the total B- galactosidase activity. B-Galactosidase levels, as determined with respect to synthetic substrates, are abnormal in other diseases. Ockerman _et gt.(174) and MacBrinn gt gt. (175) reported a B-galactosidase deficiency in Hurler's disease, but Kint _e_t a_1. (176) later demonstrated that the apparent B-galactosidase deficiency was due to the inhibition of synthetic B-galactosidase activity by chondroitin sulfate, which binds to the enzyme and changes its electrophoretic pattern. In the mucolipidoses, B-galactosidase levels were reported deficient (177- 179) in leukocytes and fibroblasts. However, subsequent studies have shown that the apparent B-galactosidase deficiency was not the major inherited, metabolic defect in the mucolipidoses (180-181), since the parents of the patients had normal B-galactosidase activity. Also, a number of glycosylhydrolase activities, along with that of B-galactosidase, were reported elevated in the serum of the patients, pointing to a possible lysosomal uptake defect reported for "I-cell" disease (182). Interestingly, GMl B-galactosidase levels were reported deficient in "I-cell" disease liver and leukocytes, but the enzyme activities were found normal in the plasma and leukocytes of parents of the patients (180). As a result, it was concluded that the cause of the observed GMl B-galactosidase deficiency was still uncertain, possibly deriving from inhibition by the uncharacterized storage material. 30 11. Im munoregulation by General Lipids and Lip0proteins There has been increasing interest in the possible im munoregulatory role of lipids. Free fatty acids and their esters were the first classes of lipids to be extensively evaluated. These lipids have been shown to inhibit both gt fltrg and i_n_ 2M). immune responses, including lymphocyte blastogenesis, lymphocyte cytotoxic responses, primary and secondary humoral immune responses, phagocytosis, and allograft rejection (183). Prostaglandins, especially those of the E series, inhibited a wide range of immunological responses (184). A role for tg ELIE modulation of the immune response by endogenous prostaglandins has been demonstrated. In Hodgkin's disease, prostaglandin synthesis by a suppressor cell was shown to suppress the blastogenic response of lymphocytes to phytohemagglutinin mitogen. The inhibition of prostaglandin synthesis by indomethacin abrogated this suppression (185). Studies in man defining an immunological defect due to a general alteration in lipid metabolism and levels of circulating lipids were first reported by Waddell _e_t _a_l. (186). They found that patients with Frederickson's Type IV and V hyperlipidemia, characterized by elevated levels of triglyceride and very low- density lipoprotein (Type IV), and chylomicrons (Type V) had a plasma suppressor of lymphocyte blastogenesis. Separation of lipids and lipoproteins from the plasma also re moved the suppression. Hyperlipidemia has also been found to be a constant characteristic of an inherited form of histiocytosis, familial erythrophagocytic lymphohistiocytosis (6). The immunological findings have been discussed in the introductory section of this thesis. The study of the immunoregulatory properties of lipoproteins has been the effort of Edgington's group (187). Their first observations were the identification 31 of a low-density lipoprotein, associated with viral infection, which was capable of modulating the binding of T-cells to sheep erythrocytes (E-rosettes). They found that lymphocytes bear high affinity receptors for this lipOprotein species, and that occupation of these receptors inhibited E-rosette formation of lymphocytes. Subsequently, Curtiss and Edgington (188) defined a sub-fraction of low-density lipoprotein (called LDL-In for its inhibitory activity) which was capable of modulating the lymphocyte blastogenesis response to mitogens and allogeneic cells. This lipOprotein caused irreversible suppression of the blastogenic response when preincubated with responding lymphocytes for 24 hr prior to the addition of a stimulus of blastogenesis. These investigators demonstrated that the inhibitory effect of LDL-In was a direct, primary action on the unstimulated lymphocyte, mediated by a distinct lymphocyte receptor for LDL-In not identical with the known LDL receptor found on many types of cells, including lymphocytes (189). 12 _v_iy_9_ studies with LDL-In have defined an inhibition of murine primary immune responses, as measured by the generation of plaque-forming (antibody- producing) splenic cells following immunization with sheep erythrocytes (190). Investigation of i_n_ ligrg antibody synthesis induced by pokeweed mitogen stimulation of peripheral blood lymphocytes have also demonstrated the inhibitory effect of LDL-In. Their studies indicated that the target for LDL-In was the lymphocyte, and that both T and B lymphocyte functions were modulated by LDL- In (191). Studies of very low-density lipoproteins (VLDL) by Chisari (192) have demonstrated similar im munosuppressive activity for this lipOprotein. Bieber gt gt. (193) have recently isolated a neutral glycosphingolipid derived from LDL in serum of Hodgkin's disease patients which inhibited E-rosette forming cells. Inhibition by LDL from normal serum was not found. The inhibitory activity was distinct from that of C-reactive protein and complement Clq. TLC analysis of the glycolipid identified the inhibitor as either a mono- or 32 diglycosylceramide. Interestingly, commercial preparations of galactosylceramide selectively inhibited Hodgkin's disease E-rosette forming cells. Lipids have also been implicated in modulation of the host immune response to tumors. Using macrophage killing of tumor cells as an assay system, Chapman and Hibbs (194) have shown that a low-density lipoprotein present in human, murine, and fetal bovine serum was capable of inhibiting the normal tumoricidal response of activated macrophages. Schultz _e_t .a_l. (195) have shown that prostaglandins can also modulate the tumoricidal function of macrophages. Raz gt gt. (196) demonstrated inhibition of _ig mtg macrophage-tumor cell interaction by membrane vesicles which had been shed by the tumor cells, suggesting that the shedding may produce self-protection of the tumor cells from host immune destruction. Analogous inhibition of _ig _v_i_t£9_ B-cell antibody synthesis was reported by Freimuth gt a_l. (197) from the shed, high molecular weight (over 2 6 X 10 daltons) membrane vesicles containing gangliosides with theta antigen activity from murine lymphoblastoid cell lines. 111. Immunological Roles for Gangliosides A. Receptors and Antigenic Cell Surface Markers The potential of glycolipids to act as cell surface antigens has been long recognized. Some of the well-studied glycolipid antigens include the Forssman antigen and the human blood group antigens (198). The ganglioside nature of certain lymphoid cell surface markers has been reported. Esselman and Miller (199) demonstrated the gangliosidic nature of theta (Thy-I) antigen on mouse thymocytes, and recently, differentiated Thy 1.1 and Thy 1.2 antigenic activity from two distinct ganglioside fractions separated by TLC and differentially susceptible to neuraminidase degradation (200). The identity and structure of these gangliosides are unknown at present. 33 It has been reported that the macrophage receptor for the lymphokine, migration inhibition factor (MIF), is a ganglioside which is derived from the acidic, water-soluble glycolipid fraction of guinea pig macrophage total lipid extracts. Pretreatment of macrOphages with macrophage derived glycolipids markedly enhanced their response to MIF, presumably by incorporation into macrophage plasma membrane. Because pretreatment of the glycolipids or the target macrophages with o-fucosidase or neuraminidase destroyed the macrophage responsiveness to MIF (201-202), it was thought that the MIF receptor was a fucoganglioside with TLC mobility between GM and GD 1 a (203). 1 Other workers (204) have found that bovine brain mixed gangliosides contained a minor unidentified component that inactivated guinea pig and rat lymphocyte MIF and macrophage activating factor (MAF). The identity of the ganglioside was not established, but absorption studies with commercial preparations of purified GMI’ GMZ’ GM3, GD Ia’ and GT 1 were negative. Miura _e_t gt. (205) have recently reported the specific inhibition of MIF by a fucosylated glycolipid (RM) isolated from rat peritoneal macrophages which had the structure: GalB 1+3Gal(2+l aFuc)B 1+3GalNAcB 1+3GalB 1+4Gch 1+1'Cer. GM3 and a blood group B-active glycolipid with identical structure with glycolipid RM except for a GlcNAc substitution were not inhibitory. B. Im munoregulation by Gangliosides Early studies (206) showed that the addition of mixed brain gangliosides to bacterial antigens prior to injection into rabbits resulted in a reduction in magnitude and duration of the primary response (IgM production) and complete inhibition of the secondary response (IgG production). 34 Miller and Esselman (207) first documented modulation of the i_n ytttg immune response by gangliosides. The addition of GMl ganglioside to splenic cultures incubated with sheep erythrocytes depressed the anti-sheep hemolytic plaque response. They suggested that this effect was due to a direct effect on B lymphocyte terminal differentiation into antibody-producing plasma cells. Studies of helper and suppressor T-cells in the mouse (208) demonstrated that the conditioned media of the mouse antigen-induced T-suppressor cells, which inhibited the primary i_n _vi_tr_g immune response to a heterologous antigen, contained a glycolipid released by the T-cells. Treatment of the T-suppressor cell media with either anti-GMl or anti-Thy 1.2 would remove the inhibitory activity. Isolation of the inhibitory glycolipid by TLC revealed a glycolipid with a mobility similar to brain GM 1' Other glycolipids isolated from the media were not inhibitory. The inhibitory activity of this ganglioside on the B-cell response was found to be antigen non-specific, suggesting that the ganglioside might have important general immunoregulatory properties. Lengle gt gt. (209) have demonstrated that lymphocyte blastogenic response to conconavalin A was inhibited by bovine brain gangliosides. These gangliosides caused reversible inhibition of RNA and DNA synthesis. Some structure-function data was obtained in that purified preparations of GTl gave the highest inhibition, followed by, in decreasing order of effectiveness, GMZ’ GDla’ and finally GMI’ which had very low inhibitory activity. Ryan and Shinitsky (210) have also shown an inhibitory effect of bovine brain gangliosides on lymphocyte blastogenesis, although they claimed that the inhibition was exclusively towards B-cell response. The B-cell specificity was demonstrated by the fact that only E. ggtt lipopolysaccharide mitogen stimulation was inhibited, whereas phytohemagglutinin stimulation (T-cell specific) was unaffected at identical ganglioside concentrations. 35 Stewart gt gt. (211) recently reported that human brain gangliosides also have inhibitory activity toward lymphocyte blastogenesis as measured by con A- stimulation and the mixed leukocyte culture assay (M LC). In summary, gangliosides have been found to have a variety of immunological roles: I. They are antigenic in that specific antibodies can be directed against them. 2. They function as cell surface markers on murine thymocytes (theta antigen). 3. They are macrophage cell surface receptors for lymphokines, such as macrophage activating factor and migration inhibiton factor. 4. They are potent inhibitors of T and B lymphocyte blastogenesis and antibody synthesis, and as such, may be the biochemical mediator by which suppressor T-cells and tumor cells suppress immune responses. MATERIALS AND METHODS Materials Chemicals Pharmgcia (Piscatawaj, N. J.) DEA E-Sephadex A-25 Dextran T 500 Pfanstiehl (Waukegan, Ill.) N-acetylgalactosamine N-acetylglucosamine N-acetylmannosamine galactose glucose fucose mannose Calbiochem (La Jolla, Ca.) sodium taurocholate Maltxh (Ngwark, Delaware) Silica Gel G TLC plates Silica Gel H TLC plates E. Merck (Cincinnati, Ohio) Silica Gel 60 TLC plates Silica Gel 60 High Performance TLC plates Sigma (St. Louis, Mo.) p-nitrophenyl-Bfl-acetylgalactosaminide p-nitrophenyl- Bfl-acetylglucosaminide p-nitrophenyl-B-galactoside tyrosine bovine serum albumin sodium borohydride galactose glucose stachyose phenyl- a-N-acetylglucosaminide _N_-acetylneuraminic acid hexamethyldisilazane trimethylchlorosilane mannitol 36 37 dichlorodim ethylsilane glycine tris-(hydroxymethyl)-aminomethane Boehringer-Mannheim (Indianapolis, Ind.) 2(3'-methoxy)phenyl- o-fl-acetylneuraminide Eastman Kodak (Rochesteg N.Y.) m-methoxyphenol p-nitrophenol Koch-Light (Colnbrook, Buckinghamshire, England) 4-methylumbelliferyl- u-galactoside 4-methylumbelliferone 4-methylumbelliferyl- B-glucoside 4-methylumbelliferyl- B-galactoside 4-methylumbelliferyl- B-N-acetylglucosaminide 4-methylumbelliferyl- B-N-acetylgalactosamide 4-methylumbelliferyl- a-fucoside 4-methylumbellif eryl- a-mannoside 4-methylumbelliferyl-sulfate A. H. Thomas (Philadelphia, Pa.) dialysis tubing Teflon screw-cap liners Iatron Labs (TokyoLJyan) Iatrobeads 6RS 8060 Fisher (FairlawnLNJJ orcinol resorcinol iodine Redi-Plates (Silica Gel G TLC plates) sulfuric acid hydrochloric acid acetic acid Mallinckrodt (St. LouisLMo.) silver carbonate sodium thiosulfate citric acid sodium Citrate sodium phosphate sodium acetate Matheson Gas (Joliet, Ill.) hydrogen chloride Harleco (PhiladelphigL Pa.) Folin-Ciocalteau reagent 38 Clarkson (Williamsport, Pa.) UnisilIlOO—ZOO mesh) Airco Industrial Gases (MontvaleLNJJ hydrogen, 90.5% pure compressed air nitrogen, 99% pure helium, 99% pure Melco (Bellefonte, Pa.) GM ganglioside bov ne brain mixed gangliosides cholesterol cholesterol oleate oleic acid phosphatidylcholine phosphatidylethanolamine phosphatidylserine l-monopalmitin 1,2-dipalmitin triolein fatty acid methyl esters NIH standard mix 3% 55-30 on Supelcoport (80-100 mesh) arachidic acid (C20:0) Solvents All solvents listed here are of reagent grade and re-distilled: Mallinckrodt (St. Louis, Mo.) methanol chloroform pyridine (stored over KOH pellets) toluene acetic anhydride l-propanol 2-propanol l-butanol acetone hexanes petroleum ether(b.p. 300-750) diethyl ether Commercial Solvents (Terre Haute, Ind.) absolute ethanol 39 Tissues Fibroblast cell lines, leukocytes, and plasma samples from patients were donated by Dr. Stephan Ladisch, UCLA Med. Ctr. Other normal and outdated plasma samples were donated by the American National Red Cross, Lansing, Mi. Human livers, spleens, and brains were donated by Dr. Stephan Ladisch and Dr. Allan Yates, Ohio State Univ. 40 Methods 1. Lysosomal Enzyme Studies A. Pryaration of Human Tissues and Cells 1) Frozen human liver or spleen samples were weighed and homogenized as described by Suzuki (212) in distilled water with a Polytron homogenizer for a pulse interval of l min repeated three times. The concentration of tissue in the final homogenate was 10% (w/v). The tissue cells were further disrupted by freeze- thawing three times and the mixture was centrifuged at 1,000 X g for 10 min to remove the insoluble cellular debris. Aliquots of the supernatant fraction were used for subsequent assays. 2) One to two m1 of packed skin fibroblasts were subjected to hypotonic lysis by suspension in an equal volume of double-distilled water as described by Suzuki (212). The fibroblasts were further disrupted by freeze-thawing three times, centrifuged at 1100 X g for 5 min to precipitate the cellular debris, and aliquots taken from the supernatant fraction for enzyme assays. 3) Leukocytes were prepared as described by Snyder and Brady (213). Two ml of a 5% Dextran in isotonic saline (0.9% NaCl) solution was added to 10 ml of heparinized blood and the resulting mixture was allowed to settle at room temperature for 45 min. The leukocyte layer was carefully removed by Pasteur pipet and centrifuged for 10 min at 600 X g. The resulting pellet was resuspended and washed twice with 2 ml isotonic saline solution and centrifuged to obtain the pellet. The pellet was suspended in 3 ml of distilled water for 90 sec to lyse any remaining contaminant RBCs after which 1 ml of 3.6% NaCl solution was added to return the suspension to isotonicity. The leukocyte suspension was centrifuged at 41 600 X g for 10 min to obtain a RBC-free leukocyte pellet. This leukocyte preparation could either be stored frozen at -20° with 1 ml isotonic saline or lysed immediately by suspension in 1 ml distilled water, freeze-thawing three times, amd centrifugation at 1120 X g. The supernatant fraction was used for subsequent enzyme assays. B. Lysosomal Glycosjlhydrolase Assays l) Assays with p—nitrophenylglycoside substrates A typical assay mixture for the determination of lysosomal glycosylhydrolase activity with p-nitrophenylglycoside substrates contained: 50 pl of 0.6 M sodium citrate or sodium citrate-phosphate buffer, 200 pl of substrate(l-10 mM concentration) in a 13 mm X 100 mm test tube, 0-50 pl of enzyme, and distilled water to bring the total volume of the assay to 300 pl. The assay mixture was incubated at 370 for a time interval that ranged from 5 min to 3 hr. The reaction was stopped by the addition of 3 ml of 0.6 M potassium borate buffer (made by adjusting 0.6 M boric acid to pH 10.4 with a 5 M KOH solution). Controls were incubated at 37° without enzyme for appropriate time intervals, followed by the simultaneous addition of enzyme and stopping solution. Absorbance was read at 420 nm on a Gilford Model 250 Spectrophotometer. Due to large amounts of protein and insoluble matter in some liver homogenates, 1 ml of I-pentanol/chloroform (1:5) was added to the final solution to precipitate proteinaceous matter at the interface. A standard curve of p-nitrophenol was linear from 0-300 nmoles. Duplicate assays were done at two different time points and at two different enzyme concentrations to insure linearity of the assay with respect to time and protein concentration. 42 2) Assays with 4-metmlumbelliferylglycoside substrates A typical assay mixture for the determination of lysosomal glycosylhydrolase activity with 4-methylumbelliferylglycoside substrates contained: 300 pl of a 1-10 mM substrate solution in 0.1 M sodium citrate, citrate- phosphate, or acetate buffer, 0-50 pl of enzyme in a 10 mm X 75 mm disposable test tube, and distilled water to bring the total volume of the assay mixture to 350 pl.After an incubation period at 37° for 0-6 hr, the reaction was stopped with 2 ml of a 0.2 M glycine-NaOH pH 10.8 buffer. Controls were incubated without enzyme and stopped with the simultaneous addition of enzyme and stapping solution. The liberated 4-methylumbelliferone was measured in an Aminco fluorometer- colorimeter Model 347439 with an excitation wavelength of 365 nm and an emission wavelength of 448 nm using the assay test tube directly as a cuvette. A standard curve of 4-methylumbelliferone gave a linear fluorescence response from 0-30 nmoles. Duplicate assays were done at two different time points and at two different enzyme concentrations to insure linearity of the assay with respect to time and protein concentration. 3) Measurement of neuraminidase activity A typical assay mixture for the determination of neuraminidase activity (214) contained: 30 pl of a 10 mM 2(3'-methoxyphenyl)- a-N-acetylneuraminide solution in 0.1 M sodium acetate pH 5.0 buffer, 10-50 pl of enzyme, and additional acetate buffer mixed in a 10 mm X 75 mm disposable test tube to a final volume of 200 pl. The reaction mixture was incubated at 370 for intervals ranging from 10 min to 16 hr and stopped by the addition of 1.5 ml of 10% sodium carbonate. Upon addition of 0.2 ml of 2 N Folin-Ciocalteau phenol reagent, the tube contents were vortexed, allowed to stand for 20 min, and the absorbance read in a Gilford Model 250 spectrophotometer at a wavelength of 750 nm. 43 In the event of excessive interference from protein in the phenol reagent reaction, a modification of the technique was performed as follows (215): 0.2 ml of toluene was added instead of sodium carbonate as the stopping solution, the reaction mixture was vortexed and centrifuged in a Sorvall desk top centrifuge at 1000 X g for 5 min, and the toluene upper layer was removed by Pasteur pipet. This toluene extraction was repeated twice more at which point the toluene extract could either be stored at -200 or treated with 0.5 ml 10% sodium carbonate in order to extract the released methoxyphenol into the lower alkaline, aqueous phase. After carefully removing the toluene upper phase by Pasteur pipet, 0.25 ml of Folin-Ciocalteau phenol reagent (diluted 1:2 with water) was added, and the entire solution was vortexed and allowed to stand for 20 min. Using either tyrosine or m-methoxyphenol as a standard, linearity of absorbance at 750 nm was obtained from 0-300 nmoles. Duplicate assays were done at two different time points and at two different enzyme concentrations to insure linearity of the assay with respect to time and protein concentration. 4) Protein assays All protein assays were done precisely as described by Lowry _e_t gt.(216). Using bovine serum albumin as standard, the assay gave a linear absorbance response at 750 nm from 0-80 pg of protein. 5) Cellulose acetate electrophoresis studies on liver B-galactosidases Aliquots (30 pl) from the liver homogenates were applied by a special Gelman sample applicator onto 1 X 6 inch polyacetate strips (Gelman Sephraphore III) previously wetted in Gelman High Resolution tris-barbiturate pH 8.8 buffer and mounted on a Gelman electrophoresis apparatus model 51170 with dual chambers 44 containing the tris-barbiturate buffer. The strips were electrophoresed at 4° with a 250-volt electrical field across the strips (2.5 milliamperes per strip) for 50 min. To stain for enzymatic activity, the strips were carefully removed by tweezers from the apparatus and placed between a pair of 18.5 cm Whatman no. 1 filter papers wetted with 1 mM 4-methylumbelliferyl-B-galactoside in 0.1 M NaCl, Na citrate pH 4.5 buffer. The "sandwich" was placed in a large, covered Petri dish and incubated for 3 hr at 37°. Afterwards, the cellulose strips were transferred into another filter paper "sandwich" wetted with 0.2 M glycine-NaOH pH 10.8 buffer for 5 min. The strips were viewed under a portable long wavelength U.V. lamp for location of the B-galactosidases. II. Lipid Isolation and Characterization A. Lipid Extractions and Chromatography 1) Folch Extraction of Human Liver Lipid extractions as described by Folch e_t gt.(87) were done on human livers as follows: A weighed, frozen sample of liver (10 g) was minced and homogenized with 20 volumes of chloroform:methanol (2:1) in sand by mortar and pestle. The homogenate was filtered through Whatman no.1 filter paper, the insoluble residue re-extracted again with the same volume of chloroform-methanol (2:1), refiltered, and the filtrates combined. A 0.05 M NaCl solution amounting to 20% of the pooled volume was added to the lipid extract in a large separatory funnel. After separation of the lipid extract into 2 phases upon standing overnight at 4°, the upper aqueous phase (40% of the total volume) was washed twice with theoretical lower phase (chloroform:methanol:water (86:14:1)), and the lower chloroform phase (60% of total volume) was washed twice with theoretical upper phase (chloroform:methanol:0.58% NaCl (3:48:47)) as described by Folch gt gt.(87). 45 The pooled upper phase and lower phase extracts were then evaporated i_n_ vacuo on a rotary evaporating apparatus . 2) Unisil Silicic Acid Column Chromatography Unisil silicic acid column chromatography was used to fractionate the Folch lower phase lipids into 3 major classes. Unisil (20 g/g lipid) was activated at 80° overnight, slurried with chloroform, and quickly poured into a column with 3 bed volume washes of chloroform. The lower phase lipids were chromatographed and batch-eluted with 5 bed volumes of chloroform (neutral lipid fraction), 10 bed volumes of acetone:methanol (9:1) (neutral glycolipid fraction), and 8 bed volumes of methanol (phospholipid and other acidic lipids fraction), as described by Esselman gt g_l. (217) and by Vance and Sweeley (111). Each of the 3 fractions was evaporated i_n_ vacuo and redissolved in 5 m1 chloroform:methanol (2:1). B. Quantitation and Characterization of Neutral Liver Lipids l) Gravimetric quantitation of lipids Total lipid extracts from the Folch lower phase and the 3 major lipid fractions derived from silicic acid column chromatography of the total liver lipids were weighed on a Mettler analytical balance. Solvents and water were thoroughly removed from the samples to be weighed by rotary evaporation in a 35° water bath, followed by further evaporation under a stream of nitrogen with 3 additions of absolute ethanol to remove all traces of water. The samples were further dried overnight in a desiccator gt vacuo. 46 2) Qualitative Analysis of Liver Lipids The 3 major lipid fractions were each taken up in 5 ml of chloroform:methanol (2:1) and aliquots spotted on TLC plates (activated for 1 hr at 110°) for further analysis. Silica Gel G, 250 micron thick plates were used for the analysis of both the neutral lipid and glycolipid fractions. Silica Gel H (no CaSOn binder), 250 micron thick plates were used for the analysis of the phospholipid fraction. The neutral lipids were developed in petroleum ether:diethyl ether:acetic acid (90:10:l), and the glycolipids and phospholipids were developed in chloroform:methanol:water (100:42:6). All lipids were visualized by iodine vapor for preparative TLC. For qualitative analysis (218), neutral lipids were visualized by H250 charring and by H250 :acetic acid (1:1) spray reagent specific for 4 4 cholesterol residues, glycolipids were visualized by orcinol-H250“ spray reagent, and phospholipids by molybdenum blue spray reagent. 3) Direct Probe Low Resolution Mas Spectrometry of Triglycerides The triglycerides from the neutral lipid fraction (400 pl aliquot from 5 ml fraction) of both normal and patient liver were isolated by preparative TLC on silica gel G, 500 micron thick plates, developed in petroleum ether:diethyl ether:acetic acid (90:10:1) and visualized by iodine vapor. The triglyceride spots were scraped off by razor blade and poured into a 1.5 cm i.d. X 15 cm column plugged with a glass wool bed support, and the silica gel was washed with 5 column volumes of chloroform:methanol (2:1). The eluants were evaporated, redissolved in 1-2 ml chloroform:methanol (2:1), and aliquots taken for direct probe mass spectrometry and methanolysis. Mass spectra were obtained with a PDP-8/e computer-assisted Varian Model CH5 mass spectrometer using a direct probe, electron impact mode. Results were recorded on an U.V. oscillograph where the 47 peaks were mass assigned and integrated manually. Mass spectrometer conditions were as follows: probe temperature program 240°-300°, ionization energy 70 eV, accelerating voltage 3 kV, and ion source temperature 260°. 4) Gas Chromatographic Analysis of Fatty Acid Methyl Esters from Liver Triglycerides Aliquots from the isolated triglycerides were subjected to methanolysis in 3 ml 0.75N HCl in methanol (methanolic HCI prepared as described by Chambers and Clamp (112)) at 80° for 24 hr. The hydrolysates were partitioned with 2 ml hexane to remove the fatty acid methyl esters, the hexane upper phase removed, and the hexane wash repeated twice more. The hexane washes were pooled, washed twice with 2 ml water, dried under a stream of nitrogen gas, and the residue redissolved in 100-200 pl hexane. Methyl arachidate (C20:0) was added as an internal standard for quantitation. Gas-liquid chromatography was performed on a F and M Hewlett-Packard Model 402 with dual flame ionization detectors on a 6 ft X 2 mm glass column containing 3% SE-30 on Supelcoport 80/100 mesh, operated at 170° isothermal. To calculate triglyceride molar content, the total fatty acid molar amount was divided by 3. 5) Quantitation of Cholesterol Lipids in Liver Determination of cholesterol in liver was performed colorimetrically as described by Abel gt a_l.(219). Total cholesterol was quantitated by assaying 200 pl aliquots from the whole neutral lipid fraction. Free cholesterol was quantitated following the isolation of the cholesterol by preparative TLC developed in petroleum ether:diethy1 ether:acetic acid (90:10:l). Cholesterol ester amount was calculated as the difference between the total cholesterol and free cholesterol amount per equivalent weight of liver. 48 C. Isolation and Characterization of Gangliosides 1) Folch Upper Phase Lipids- TLC and Quantitation The Folch upper phase lipid extract was lyophilized, redissolved in water, and dialyzed for 24 hr against 10 volumes of distilled water at 4° with 4 changes of water, lyophilized, and redissolved in 1 ml chloroform:methanol (2:1). Insoluble material was removed by filtration through glass wool-stoppered Pasteur pipet columns. TLC analysis of the gangliosides was done on silica gel G 250 micron thick plates developed twice in chloroform:methanol:7%NHuOH (55:40:10) in paper-lined tanks saturated with developing solvent. The plates were visualized by both iodine vapor and by resorcinol spray reagent (14). Aliquots were taken from the 1 ml lipid solution for sialic acid colorimetric quantitation by resorcinol as described by Svennerholm (220) and modified by Spiro (221). 2) Ganglioside Isolation by DEAE—Sghadex Column Chromatography The extraction of gangliosides from large amounts of human liver tissue was done by a slight modification of the procedure described by Ledeen gt gt.(50) and Ueno gt gt.(222). DEAE-Sephadex A-25 was prepared for chromatography as follows: 2.2 g beads/g tissue were converted from the chloride to the acetate form by mixing into a slurry with chloroform:methanol:0.8 M sodium acetate (30:60:8). The slurry was decanted and the washing repeated 4 more times. The slurry was left to settle and stand overnight, after which the solution was again decanted and replaced by chloroform:methanol:water (30:60:8). Decantation and resuspension of the DEAE- Sephadex was repeated 3 times. The slurry was poured into a 3 cm i.d. X 35 cm glass column and washed with 4 column volumes of chloroform:methanol:water (30:60:8). 49 A 100-500 g sample of liver (wet weight) was homogenized with a Polytron homogenizer in 20 volumes of chloroform-methanol (1:1). After stirring overnight at 4°, the homogenate was filtered to remove the insoluble residue. The insoluble residue was re-extracted with 10 volumes of chloroform:methanol (1:1) and refiltered. The filtrates of both washes were combined and evaporated _ig ygcgg. The lipid extract was redissolved in chloroform:methanol:water (30:60:8) and applied to a 3 cm i.d. X 35 cm column of pre-equilibrated DEAE—Sephadex A-25 (acetate form). Ten column volumes of the solvent followed by 2 column volumes of methanol were passed through the column to elute all non-acidic lipids. The acidic lipids (sulfatides, phospholipids, and gangliosides) were eluted with 9 column volumes of 0.2 M sodium acetate in methanol and then dried i_n w. In order to remove phospholipids, alkaline methanolysis was performed by redissolving the extracted lipids in 0.5 N NaOH in methanol and incubating the mixture for 4 hr at 37°. The solution was neutralized with 5 N acetic acid in methanol and dried gt w. The residue was redissolved in water and dialyzed against distilled water at 4° with 4 changes over a period of 24 hr. The dialysate was lyophilized. 3) Iatrobead Column Chromatography The lyophilized lipids were fractionated by Iatrobead chromatography as described by Momoi gt gt. (99) utilizing a linear gradient of chloroform:methanol:water from 60:40:2 to 30:70:4 on a 2 cm i.d. X 24 cm glass column of Iatrobead silicic acid pre-equilibrated with the starting solvent. Small aliquots of each 10 ml fraction collected were qualitatively analyzed by development on silica gel 60 TLC plates in a chloroform:methanol: 0.2% CaCl2 (60:40:9) solvent and appropriate fractions pooled. Some ganglioside fractions were further purified either by preparative TLC in the same developing solvent or by re- chromatography on the Iatrobead column. The final ganglioside fractions were 50 quantitated colorimetrically by the resorcinol assay (220-221) and shown to be free of contaminating substances by TLC. 4) Extraction of Plasma Gangliosides The extraction of lipids from plasma (1-10 ml) was performed in a similar manner as that described above for liver ganglioside extraction with 20 volumes of chloroform:methanol (1:1). After the chloroform:methanol filtrates were pooled and dried i_n_ ya_c_1tg, the dried lipids were redissolved in the original volume (1-10 ml) with chloroform:methanol (1:1) solvent and applied on to silica gel 60 TLC plates. The plates were developed in three different solvent systems as described by Harth gt a_l. (104) in order to separate the gangliosides from all other lipids. The gangliosides were identified by comparison with standard gangliosides and visualized by either iodine vapor or resorcinol spray reagent. They were scraped off the plate and eluted from the silica gel with successive washes of chloroform:methanol (1:1), chloroform:methanol:water (50:50:15), and methanol. The recovered gangliosides were quantitated by the resorcinol method as described earlier (220-221). 5) Carbohydrate and Fatty Acid Composition of Liver Immunosuppressive Substance Aliquots of the ganglioside fractions that were immunosuppressive were further purified by preparative TLC on 200 micron thick silica gel 60 plates in the triple solvent system described by Harth e_t gt.(104). The bands were visualized by iodine, scraped off, and eluted as described for the plasma gangliosides. The samples were dried in 1.3 cm X 10 cm Teflon-lined screw-capped test tubes and methanolyzed as described by Esselman gt gt. (217). Three ml of anhydrous l N HCl-methanol was added to the sample, the samples capped tightly, and the tubes 51 allowed to stand at 80° for 18 hr. The samples were cooled to room temperature and 2 ml hexane was added to extract the fatty acid methyl esters. The upper layer of hexane was removed by Pasteur pipet and the 2 ml hexane wash repeated twice more. The hexane washes were pooled and dried under a stream of nitrogen in separate 1 dram vials with Teflon-lined screw caps, whereas the methanol layer was neutralized by 10-50 mg AgZCO3 powder added gradually into the methanol solution with frequent vortexing. When the methanol solution was pH 6 by litmus paper test, 300 pl of acetic anhydride was added and the mixture allowed to stand at room temperature for 8-12 hr to re—N-acetylate all amino sugars. The solution was centrifuged for 4 min at 400 X g and the methanol carefully transferred by Pasteur pipet into a 1 dram vial with Teflon-lined screw cap and dried under a stream of nitrogen. The AgCl precipitate was washed twice more with 1 ml methanol, centrifuged, and the washes transferred into the l dram vial. The methanol washes were finally dried under nitrogen and were ready for trimethylsilylation and GLC analysis. For sugar derivatization (223), 50-100 pl of TMS reagent (pyridine:hexamethyldisilazane:trimethylchlorosilane 10:4:2) was added to the hydrolyzed glycolipid samples, and the mixture allowed to stand at room temperature for 30 min before injection of 2-3 pl aliquots into a Hewlett- Packard Model 5840A gas liquid chromatograph for carbohydrate analysis. The pooled hexane washes were dried under nitrogen in their vials, redissolved in 50-100 pl hexane, and 2-3 pl aliquots analyzed by GLC. Conditions for the GLC analyses were as follows: nitrogen carrier gas flow-20 m1/min; temperature program- 140° to 240° at 3°/min (carbohydrate analysis), 140° to 240° at 5°/min (fatty acid analysis); glass column- 6 ft x 2 mm i.d. with 3% SE-30 on 80-100 mesh Supelcoport. Identifications of the trimethylsilylated carbohydrates and fatty acid methyl esters detected by gas chromatography were confirmed by combined GLC- 52 M5 on a Hewlett Packard Model 5895 using the same GLC conditions as described above with ionization energy- 70 eV, ion source temperature- 200°, and electron multiplier- 2200 volts. 6) Total Sialic Acid Determination in Liver bLGas-Liquid Chromatography To determine the total sialic acid content in liver, 1-2 ml aliquots of 10% liver homogenates were lyophilized in Teflon-lined screw-capped test tubes. The dried samples were hydrolyzed in 4 ml 0.05 N HCl in methanol for 2 hr at 80° as described by Ledeen and Yu (224). After 3 washes with 6 ml hexane to remove fatty acid methyl esters, the methanol solution was centrifuged at 1000 X g for 10 min to pellet the insoluble matter, and the methanol was carefully removed by Pasteur pipet into a 1 dram vial with Teflon-lined screw cap. The insoluble matter was washed twice with 1 ml methanol, centrifuged, and the washes combined and dried in the vial. Then 50-100 nmoles of phenyl-o-N-acetylglucosaminide was added to the vial as an internal standard, and the entire content dried under nitrogen. The samples were trimethysilylated with 100-200 p1 of the TMS reagent described earlier by Laine gt gt. (223). Finally, 2-3 p1 aliquots of the derivatized sample were analyzed with a Hewlett-Packard Model 5840A gas chromatograph containing a 6 ft X 2 mm glass column of 3% SE-30 on 80-100 mesh Supelcoport operated at 215° isothermal. D. Oligosaccharide Analysis Oligosaccharides from liver and spleen were analyzed by a modification of the method described by Humbel and Collart (225). Five grams of tissue were homogenized in 10 ml distilled water for l min on a Polytron homogenizer to yield a 33% homogenate. The homogenates were centrifuged at 50,000 X g for 45 min at 4° to give a clear supernatant which was spotted in 10-20 pl aliquots onto silica 53 gel G 250 micron thick TLC plates (Fisher Redi-Plates). The plates were developed in butanolzacetic acid:water (3:3:2) and visualized by orcinol spray (14) for neutral sugar content and by resorcinol spray (14) for sialic acid content. E. Ganglioside lmmunosuppression of Lymphocyte Blastogenesis Immunosuppressive activities of all gangliosides, presented in the form of liposomes, were tested by Dr. Stephan Ladisch using the procedure described by Muchmore and Blaese (226) by assessing the effect of the glycolipids on human lymphocyte i_n Mg proliferative responses. The glycolipid-containing liposomes were prepared in the following manner: glycolipids to be tested were dissolved in a small amount of chloroform:methanol (1:1) in 4 ml glass vials. Chromatographically pure cholesterol and phosphatidylcholine in the same solvent were combined with the glycolipids in the respective ratio 5:5:1 by weight at room temperature. The solvent was evaporated under a stream of nitrogen and all final traces of solvent removed by further evaporation i_n yggtng. The lipid mixture thus prepared was resuspended with sterile 0.9% NaCl solution by sonication at 4° for 5-20 min duration with pulse intervals of l min in a bath sonicator. The liposome-incorporated glycolipids were tested in the lymphocyte blastogenesis assay in final calculated ganglioside concentrations of 2- 20 pg/ml, and added in 0.1 ml volumes/ 1.0 ml peripheral blood lymphocyte suspension. The lymphocyte blastogenesis assay was carried out in the following manner: normal human peripheral blood was anticoagulated with preservative-free heparin, and subjected to ficoll-hypaque density gradient sedimentation by the method of B2) yum (227). The peripheral blood lymphocytes (PBL) collected at the interface were washed three times and resuspended in medium (RPMI 1640 containing penicillin, streptomycin, and glutamine) containing 10% autologous 54 plasma before the addition of the test lipids for the assay. Controls for each experiment with test lipids were PBLs incubated in the carrier liposomes minus the test lipid, and PBLs in complete medium alone. The final concentration of the cell preparation was 1.35 X 106 lml, with 0.15 ml being plated per well. The appropriate antigen or tissue culture medium alone (unstimulated control wells) was added (0.05 ml) after an 18 hr preincubation of the cells with the liposomes being tested for immunoregulatory activity. All stimulants were used at concentrations giving optimal blastogenic responses. Stimulants used in the experiments were phytohemagglutinin (PHA), concanavalin A (Con A), pokeweed mitogen (PWM), Candida albicans antigen, streptokinase-streptodornase (SKSD), tetanus toxoid, and diphtheria toxoid. All incubations were performed at 37° in a humidified 5% COZ-95% air atmosphere. Cultures were incubated for the previously determined optimum duration of culture: 3 days for mitogen responses and 6 days for antigen responses. At the end of the culture period, 0.5pCi tritiated thymidine was added, the incubation continued for 4.5 hr, and the cells harvested onto glass fiber filter paper using a Brandel automated harvester. Tritiated thymidine incorporation was measured as counts per minute on a Beckman 8- counter. All tests were performed in triplicate and the mean net stimulation (compared to control wells containing medium instead of a stimulant) determined. Control wells (without a stimulant) were also counted to determine direct stimulatory or inhibitory activity of the added gangliosides. Other control wells were stained with trypan blue at the end of the culture period, and the viable cell count determined to eliminate non-specific cytotoxicity as a cause of apparent inhibition of the blastogenic response. A 50% inhibition of a blastogenic response in a test was considered indicative of the presence of suppressive activity. Furthermore, the presence of a dose-related inhibition by an added lipid was considered indicative of a suppressive effect. RESULTS 1. Lysosomal Enzyme Studies A. Liver Lysosomal (Ecosylhydrolases Studies with liver lysosomal enzymes indicated a lower B-galactosidase activity in the patient's liver when compared to that of a normal liver, although his other enzyme levels were within the range of the normal liver levels of activity (Table 6). His o-mannosidase activity was also somewhat lower than normal , although not to the extent of the decreased levels of B-galactosidase. B- Galactosidase activity of the patient was 34% of normal levels on a per mg protein basis, but even lower (25% of normal) on the basis of an equivalent amount of liver. All other enzyme specific activities studied were also reduced when based on equivalent amounts of liver; however, their activities were not as reduced as that of B-galactosidase (Table 7). Classical lysosomal storage diseases generally present a deficiency of a given lysosomal enzyme with activities around 0.1-10% of normal levels. Such an obvious deficiency of any particular lysosomal glycosylhydrolase was not apparent here. B. Leukocyte Lysosomal Glycosylhydrolases Table 8 lists the activities of selected leukocyte lysosomal enzymes from relatives of the FEL patient under study. The activities of these lysosomal enzymes of his relatives were all within the normal range and no clear deficiency of any enzyme activity could be demonstrated. Table 9 lists the results of a separate leukocyte preparation for B-galactosidase analysis in the patient's 55 56 Table 6. Liver Lysosomal Glycosylhydrolases- I * Enzymes Specific Activities FEL (% of normal) Normal (n=1) B-Galactosidase 890 (34) 2640 B-N-Acetylglucosaminidase 3293 (1 16) 2835 a-Galactosidase 95.5 (1 15) 83.0 a-tNt-Acetylgalactosaminidase 14.7 (68) 21.7 a-Neuraminidase 2.33 (9 4) 2.48 a-F ucosidase 393 (7 2) 546 B-Glucuronidase 944 (7 4) 1280 a-Mannosidase 42.8 (47) 91.3 *nmoles substrate hydrolyzed/hr] mg protein. The standard error for all values was less than 10%. Substrates used: 1 mM 4-MU-B-Gal in 0.1 M NaCl, Na citrate pH 4.5 1 mM 4-MU- o-Fuc in 0.1 M Na citrate-0.2 M Na phosphate pH 6.0 1 mM 4-MU-B-GlcNAc in 0.1 M Na citrate-0.2 M Na phosphate pH 4.4 10 mM pNP- a-GalNAc in 0.1 M Na citrate pH 5 1 mM 4-MU-a-Gal in 0.1 M Na citrate-0.2 M Na phosphate pH 4.4 10 mM 4-MU-B-GlcUA in 0.1 M Na acetate pH 4.8 5 mM 4-MU-a-Man in 0.1 M Na citrate-0.2 M Na phosphate pH 4.4 57 Table 7. Liver Lysosomal Glycosylhydrolases- II Enzymes FEL B-Galactosidase 35,600 B-N-Acetylglucosaminidase 247,000 a-Galactosidase 7,160 a-N- Acetylgalac tosaminidase 1,102 o-Neuraminidase 261 u-Fucosidase 22,000 B-Glucuronidase 34,000 a-Mannosidase 2,400 * Specific Activities (% of normal) Normal (n=l) (25) 145,200 (77) 319,000 (77) 9,340 (45) 2,440 (68) 385 (57) 38,666 (40) 84,500 (37) 6,450 *nmoles substrate hydrolyzed/hr/ g equivalent weight liver. The standard error of all values was less than 10%. Substrates used: C.f. Table 6. 58 n... :a 88:0 62 s. 3 5 0.....-8 32.. .25 8 n... :a 38:0 ...z .2 1o 5 8.2.6 .82.. .25 S a... :a 8282.. 62 2 No .850 62 2 3 5 30-8-32-.. .25 3 ed :a 88:0 62 .662 2 3 5 308-224 2:. _ e... :a 82868 62 2 N5 .3650 62 .2 3 5 22208.32... 2.: a "com: woumbmnsm A82 can“ 80— mm? 83:; 2m .3 Coho Eon-Bum 05. .5303 wE\E\voN>_o.6.E 30:33 8.9:: * 0.:— 9m: Now 3.3 ”Kw ”om «A: 9:» 9.0.0:...15 _ R wwn wwm N R a a am» «on N—N 03.5.2-5 Rees awn seen was weNM _aN ANRN NAN wamN _aa LRNe me~ e~n~ e- wek_ Re_ 86.22068 86.30... 83339.. 0:00am * KN— wa Nwm Nam cam saw— nmoa one me.—m01m :32... FEB: :32... 3E3: 205.. 0.0:: 20:: 3503.: C053 .5508 pound». «um—ham mozuflom 952qu am". on» no m30_BU>c_>30>_U 389.0qu 8.6950.— .w 033. 59 Table 9. Leukocyte Lysosomal B-Galactosidase of the F EL Patient's Relatives Subject Specific Activities" sister l+22 mother 358 father 347 uncle #24 uncle 396 uncle 439 grandmother 45‘} normal adult 372 *nmoles substrate hydrolyzed/hr] mg protein. The standard error for all values was less than 1096. Substrates used: 1 mM lt-MU-B-Gal in 0.1 M NaCl, Na citrate pH 5 Table 10. Leukocyte Lysosomal Glycosylhydrolases From Other FEL Patients 0 o O O * Specific Activmes Subject B-Galactosidase B-N-Acetylglucosaminidase J. D. (FEL) 894 5778 Ju. C. (FEL) 323 2298 Je. C. (FEL) 331 210i; Normal 672 3562 Normal M3 2511 It nmoles substrate hydrolyzed] hr/ mg protein. The standard error for all values was less than 1096. Substrates used: 1 mM ll-MU-B-Gal in 0.1 M NaCl, Na citrate pH 4.5 1 mM ll-MU-B-GICNAC in 0.1 M Na citrate pH (4.5 60 relatives. The enzyme activities of all the relatives were close in value to the one normal adult tested. The lysosomal glycosylhydrolase levels of other FEL patients (Table 10) were examined to study the variation of enzyme activity in other patients similarly afflicted. Both B-galactosidase and B-hexosaminidase activities of the other FEL patients showed no real differences from the normal controls. The apparent increase of lysosomal enzyme activity of JD. could be explained by the lower amounts of protein used in each assay (approximately 3396 of the other patients' and normals' protein levels tested). C. Fibroblast Lysosomal Glycosylhydrolases The lysosomal enzyme activities from the fibroblast cell lines of several FEL patients, their parents, 2 GMl gangliosidosis lines, an agammaglobulinemia, and normal controls were compared. The results are listed in Table 11. No apparent lysosomal enzyme deficiencies were found although a-fucosidase and B- glucuronidase activities were elevated above normal ranges in most FEL patients. D. Studies on the Nature of the Liver Lysosomal B-Galactosidase Deficiency B-Galactosidase activity was measured over a wide pH range to determine whether the lowered enzyme activity in the FEL patient was due to a pH optimum shift of activity. As can be seen in Figure 3, results of this experiment revealed identical pH optimum curves for both normal and FEL liver B-galactosidase. Thus, the apparent low enzyme activity was not due to a shift in pH optimum for B- galactosidase activity. To determine if the B-galactosidase deficiency was due to the presence of an enzyme inhibitor in the FEL liver, mixing experiments were conducted with equal amounts of normal and FEL liver homogenates. As can be seen in Figure 4, the enzyme activity of the mixed homogenates was median to the individual 61 .3 Ia 32.7.9... m2 2 N5 .825 m2 .2 do 5 no-5 -224 2.: 2 n... :a 88:6 m2 .62 _2 ..o 5 Basis: sé _ :33 220 w: on cam 32050.53 mz w: con 853:8 xmmmm comov m6 In 329.0%. m2 2 N6 .3330 m2 .2 #0 5 20.9224 SE n 3.: In 329.6% oz .2 N6 .3956 m2 _2 1o 5 ”2220-93.21. 5:: fl as :a 82.8% «2 .2 N5 .meU «z .2 1o 5 8.2.5.32; SE n 3 :a 838... oz .2 3 E $06-93;; .25 2 ed Ia oumcamoca oz .2 N5 .3838 m2 2 10 E van—-5557: SE _ "ENQNDV com: 33335 .3316 E mowmmmma 2.: L3? @3332 3o? 8:: zoo =< .82 can“ 82 mm? m02¢> omega 3: no ..oto vcmvcmum .5395 wE\E\ooN>_o.uE 38333 no mfloEcm 3: E 3 ca 3 an son 2.5 uczsnoEmEEmwm SQ on n2 «2 :2 3 o 3828: Eu 3%» so 33 R :2 m3 5 Na 3 emogmoficmu .25 so on: Na 3 an N” on 62 7.5.5: 3.8 R 3 SN .5 on NR 35.8 S: o... 2 NE 2 2 o? 2225 .u.m .58 a: 3 :N 8 Nu w? 1282 .u.v_ we; 8 M3 Rm NE .3 as 3m": .0 $9 cm 2. 8m 3 3 ms 3m": .0 3mm 5: 9.. SN 2 5 Rs 3m”: .03 8? 2m 3 «R as R an 3m": .05.” 0mm.Uflo< 2:8an oENncm 3836.3.»8030 _mEomOm.3 «mm—205E .2 2an 62 Figure 3. Liver B-Galactosidase Activity vs. pH Aliqugts (30 pl) from normal liver (H) and FEL liver (0—0) were assayed at 37 for 10 and 15 min incubations in chplicate and the average activities were reported. Substrate concentrations used in all assays were 1 mM 1+- methylumbelliferyl-B-galactoside in 0.1 M NaCl, Na citrate, 0.2 M Na phosphate buffer. Units of activity were defined as nmoles of substrate hydrolyzed per hr. 200 UNITS 100 63 Figure 3. 64 Figure 4. Normal and FEL Liver B-Galactosidase Mixing Experiments Enzyme activities were measure at different time intervals for normal liver (0), FBI. liver (0), and mixed (5096 of each) normal and FEL liver (+) homogenates. Aliquots (20 pl) were incubated at 37, and the averages of duplicate asays were reported. Substrate concentration in all assays was identical to that described in Figure 3. 65 60 30 TIME (MIN) O 0 4 2 DmN>..O~_o>: wh<~=mn3m mmaOSZ Figure 4. 66 activity levels of the normal and FEL liver homogenates, indicating the absence of an inhibitory substance. The effect of mixing increasing amounts of normal and FEL liver homogenates together (Figure 5) resulted also in a median level of B- galactosidase activity compared to the normal and FEL homogenates tested alone. The indication here was, again, the absence of an inhibitory substance at various concentrations of homogenates, and that the absence of inhibitory activity was not dependent on its concentration in the assay. Heat inactivation studies (Figure 6) were done to determine whether the decreased FEL liver B-galactosidase activity was due to the presence of a residual, heat-stable isoenzyme. In both normal and FEL liver enzymes, the decrease in enzyme activity due to pre-incubation at 42° was identical (73% reduction of activity), demonstrating that the FEL activity was not due solely to the presence of a heat-stable isoenzyme. . Cellulose acetate electrOphoresis of the liver B-galactosidases (Figure 7) revealed 2 major isoenzyme forms (A and B), as visualized by activity staining with the fluorescent substrate, lt-methylumbelliferyl-B-galactoside. There was no alteration in electrophoretic mobility of the FEL liver B-galactosidase isoenzymes. A decrease in the level of B-galactosidase A activity could easily be discerned, but due to the relatively low level of the B isoenzyme activity, a decrease in 8- galactosidase B could not be determined with certainty. The effect of exogenous sialic acid addition on B-galactosidase activity was investigated (Figure 8) at l X, 10 X, and 100 X the total sialic acid concentration found in the FEL liver lipid extract. Inhibition of enzyme activity was not found at any level of added sialic acid, although a slight decrease was seen in FEL and normal liver at 0.01 mM NeuNAc. 67 Figure 5. Normal and FEL Liver B-Galactosida'se Mixing Experiments with Increasing Homogenate Concentrations Enzyme activities were measured for normal liver (0), FEL liver (0), and mixed (composed 5096 of each) normal and FEL liver homogenates (+). Increasing aliquots of liver homogenates were added to the standard B-galactosidase assay described in Figure 3. The assays were incubated for 5 and 10 min, and the averages of duplicate assays at the two time points were reported. Units were defined as nmoles substrate hydrolyzed per-hr. 150 100 UNITS 50 68 l I I l 10 20 30 40 pl HOMOGENATE ADDED / ASSAY Figure 5. 50 69 Figure 6. Heat Lability for Liver B-Galactosidases Normal liver (H) and FEL liver (O—O) homogenates were preincubated at 42° for the time durations described followed by the standard B-galactosidase assay described in Figure 3. The numbers in parentheses were the percentages of total enzyme activity remaining after preincubation. All values reported were averages of duplicate assays run at 10 and 20 min. 70 7o - UNITS 0 IO 20 3O 40 MIN PREINCUBATION AT 42° Figure 6. 71 Figure 7. Cellulose Acetate Electrophoresis of Liver B-Galactosidases Normal and FEL liver lysosomal B-galactosidases were visualized by fluorescent activity staining of the polyacetate strips as described in Methods. 72 NORMAL Figure 7. 73 Figure 8. Effect of Exogenous N-Acetylneuraminic Acid on Liver B-Galactosidase Activity Equal aliquots from normal liver ( C ) and FEL liver ( O ) homogenates were incubated for 10 and 20 min with exogenous sialic acid. Activities were reported as averages of duplicate assays at the 2 time points. Concentrations of sialic acid reported were the final concentrations in each assay. Assay conditions were identical to those in Figure 3. Units of activity were defined as nmoles of substrate hydrolyzed per hr. 5U - 'I m "..'.'. Z 3 30 " ' ‘V 40 III " " (l l .1 .1 II 0.01 0.1 1.0 74 mM NeuNAc Figure 8. 75 II. Liver Lipid Studies A. Lipid Composition of Folch Extracts The Folch lower phase liver lipids were separated into 3 major fractions consisting of a neutral lipid fraction, a glycolipid fraction, and a phospholipid fraction by Unisil silicic acid chromatography as described in Methods. The quantities of each fraction in normal and FEL liver were determined gravimetrically (Table 12). Total FEL liver lipids were elevated 142% over the normal liver levels, this increase deriving completely from the neutral lipid fraction (167% of normal liver neutral lipids levels). Analysing the neutral lipids further, this elevation appeared to originate from the large increases in triglycerides and cholesterol lipids (187% and 158%, respectively, of normal liver levels) as seen in Table 13. Interestingly, free cholesterol in FEL liver was elevated to such an extent (218% of normal liver levels) that it accounted for almost all the cholesterol lipids present. Free cholesterol comprised 71% of normal liver total cholesterol lipids and an abnormally high 98% of FEL liver total cholesterol lipids. TLC analysis (Figures 9-11) of the 3 lipid fractions of the FEL and normal liver isolated by Unisil silicic acid chromatography revealed no apparent alterations in lipid profiles except in the FEL patient's neutral lipid fraction, where a large increase of triglycerides was seen. An increase in free cholesterol was also discernable on the TLC plate. In Figure 10, the orcinol- positive bandcorresponding to GL-5 was due to GM3 ganglioside carried over into the Folch lower phase partition. The - symbol marked bands that were phospholipid contaminants in the acetone-methanol wash. Mass spectral analysis (Table 14) and fatty acid analysis (Table 15) were performed on the isolated triglycerides to confirm the identity of the triglycerides and to determine whether any alterations in the fatty acid constituents had occurred. The results demonstrated that there were no major changes in the relative distribution of individual triglyceride species 76 Table 12. Total Liver Lipid Composition g/ g wet weight liver Lipid Fraction Norma1* FEL Total Lipids . 0.074 1 0.003 0.105 1 0.014 Neutral Lipids 0.048 1 0.001 0.080 1 0.002 Glycolipids 0.004 1 0.001 0.005 1 0.001 Phospholipids 0.021 1 0.001 0.020 1 0.002 i*All values are averages of duplicate samples and are reported with their standard error. Table 13. Neutral Lipid Fraction Composition pmoles/ g wet weight liverl Normal FEL Free Cholesterol 8.7 1 0.5 19.0 1 0.7 Total Cholesterol 12.3 1 0.7 19.4 _+_ 0.8 Total Triglycerides 23.9 1 3.0 44.6 1 5.1 Total Triglycerides 2 20.3 1 2.1 37.9 1 4.0 (mg/ g wet weight liver) 1All values are averages of duplicate samples and are reported with their standard error. 2 Average molecular weight from M.S. and GLC data was 850. 77 Figure 9. Thin-Layer Chromatography of Liver Neutral Lipids Normal liver (lane 1) and FEL liver (lane 2) neutral lipids from the Folch extracts were chromatographed along with standard neutral lipids. In lane 3 (top to bottom), the standard lipids were triolein, oleic acid, and 1,2-dipalmitin; in lane 4, cholesterol, and in lane 5,1-monopalmitin. The + denotes spots that gave a positive reaction for cholesterol lipids with H SO a:cetic acid ospray reagent. The TLC plate was also visualized by H2504 gha rlIIing at 130° for general lipid visualization. 78 was$+ '. .-.-1.318 . > iobieéxfié—Ofigin .. mean” I 2 Figure 9. 79 Figure 10. Thin-Layer Chromatography of Liver Neutral Glycosphingolipids Equivalent amounts of FEL (lane 1) and normal (lane 2) liver neutral glycolipids were chromatographed along with a mixed glycolipid standard derived from horse kidney. The symbol (-) denoted bands which did not give a positive reaction for the presence of carbohydrate with orcinol spray reagent. 80 GL'G .L'4 GL'B STD Figure 10. 81 Figure 11. Thin-Layer Chromatography of Liver Phospholipids Normal liver (lane 1) and FEL liver (lane 2) phospholipids from the Folch lower phase extracts were developed and visualized as described in Methods. The standard phospholipids run were sphingomyelin (lane 3), phospatidylcholine (lane 4), phosphatidylethanolamine (lane 4), and phosphatidylserine (lane 5). 82 0" , ., 1 . ... ‘- . I I '- u . ‘I -\ -‘ T. o o . . .‘ D c . -. ' .‘I I ’. - o 1. ‘ . - ‘ ~ . ‘ -- u n , o c . . I ‘. - Q. ~ ‘ f.‘ .. a 2. ' ’ 4... ~ a ‘ ' ‘ . ,M I I 5:530}: - I! ‘1002‘ 4.‘ 1" * ‘ 134 .3; ., ~- .:...,“1 fl. 1 ”8 ($133: 1'13; 3:6") ‘ Q h! o 3., I ‘ i. ‘ 0J5 fl. , » . v o .' ‘ . . \ 4““? o . as ' .r.“ ,9 ‘, 0 .g . b_;-‘ j o" ’0' ' :. ~ .1 L ‘ a, . o «r -‘." ' A .V. ' a ,. “ - “ . . ' V ) I - - - . 1 2 34 5 Figure 11. 83 Table 14. Distribution of Triglycerides in Liver Molecular Probablefl Weight Species Normal F EL 800 16:1/16:1/16:1 1.7 3.6 804 16:0/16:0/16:1 2.8 2.6 828 16:1/16:1/18:1 10.8 7.7 830 16: 1/16: 1/18:0 or 16:0/18:1/16:1 7.9 9.0 832 16:0/16:0/18:l or 18:0/16:0/ 16:1 4.5 7.5 834 16:0/16:0/18:0 2.0 1.5 856 18: 1/18:1/16:1 21.8 17.0 858 18:1/18:1/16:0 or 18:0/18:l/16:1 22.5 19.2 860 18:0/18:0/16:1 or 18:0/16:0/18:l 6.2 7.3 884 18:1/18:1/18:1 11.3 15.5 886 18: l/18:1/l8:0 5.7 6.4 888 18:0/18:0/18:1 2.8 1.8 *From+calculated ion intensity measurements of the triglycerides mass ions M+ and (M-18) ions (-H 0). Fatty acids composition of the triglycerides are designated in terms of chain lgngthmumber of double bonds. 84 Table 15. Fatty Acid Composition of Liver Triglycerides Per Cent Distribution Fatty Acid Normal 13E]; 14:0 2.3 2.0 16:0 31.6 32.0 16: l 6.8 7.6 18:0 6.7 7.6 18:1 52.6 50.7 Table 16. Total NeuNAc Content in Human Liver Nanomolesper gLiver Normal FEL Total liverl 2 1720 1 173 2230 1 260 Folch upper phase extract 8.9 1 4.4 101.3 1 15.0 All values are averages of duplicate assays on duplicate samples and are reported with their standard error. éCalculated by GLC analyses as described in Methods. Calculated by resorcinol colorimetric assay. 85 type nor in the distribution of component fatty acids of the liver triglycerides in FEL liver compared to normal liver. Calculations on Table 14 to determine the relative composition of the individual fatty acids in the liver triglycerides corroborate quite well the values determined by GLC in Table 15, especially if it is assumed that those triglyceride species of a given molecular weight with multiple species possibilities are predominantly composed of the palmitic acid-containing species. Data for the quantitation of triglycerides in Table 13 and Table 15 were derived and quantitated from gas-liquid chromatographic data of the methanolyzed products of the isolated triglycerides using methyl arachidate as an internal standard . No endogenous methyl arachidate was detected in the fatty acid methyl ester products obtained from the methanolysis of the triglycerides. B. Ganglioside Studies 1) Isolation and Quantitation of Liver Gangliosides Human liver gangliosides were isolated by DEAE—Sephadex column chromatography and fractionated by combined Iatrobead silicic acid column chromatography-preparative TLC into fractions based on their column elution times and TLC mobilities (Figure 12 and Figure 13). Total gangliosides and other sialic acid-containing components in liver were quantitated on the basis of sialic acid by GLC analysis of the mild methanolysis (0.05 N HCl-methanol hydrolysis at 80°) products of the human liver homogenates. Total gangliosides in the Folch upper phase lipids and those gangliosides fractionated by DEAE—Sephadex-latrobead column chromatography were quantitated by the colorimetric resorcinol assay for sialic acid. These results were summarized in Tables 16-18. Although the sialic acid content of the FEL liver was slightly elevated over the normal liver (1.3-fold), the major increase was seen in the Folch upper phase (11.4-fold) where the major sources of sialic acid were undoubtedly gangliosides. Thus, the increase of sialic 86 .ucowmoc .33... 350.58.. 53 ooE—mnm; cam 35:58 NUMU *NdsocmfoEchouoco—zu E pogo—05c v.53 £033 noun—a 0.: cm .00 moEm co poi—mam v.83 EnocwoumEoEu c828 pmonobfl cam xoomcaomumGlc+Cer or NeuNAc+Gal+GlcNAc+Gal(+NeuNAc)+Glc+Cer The assignment of the NeuNAc residues are based on known structures of glucosamine—containing gangliosides (Table 3). No internally-linked NeuNAc residues have yet been discovered in human extraneural lactoneo series gangliosides. It must be remembered that the ganglioside structure assignment is only tentative. The sialylparagloboside structure assigned to fraction H cannot really be possible if judged by its TLC mobility, which suggests a disialo structure or possibly a lactoneohexaglycose core sequence for the ganglioside. The ganglioside fractions may also be a heterogeneous mixture of gangliosides with similar TLC mobilities. Further purification of the ganglioside fractions may eventually be required. Upon isolation of sufficient amounts for further study, the positions of NeuNAc residues on the oligosaccharide moiety can be determined either by mass spectral analysis of the intact permethylated glycolipid and/or susceptibility of the gangliosides to L cholera or Q perfringens neuraminidases which would only hydrolyze the external (on the non-reducing terminus of the oligosaccharide moiety) NeuNAc residues. There was a possibility of a contaminant in the isolated liver fractions, albeit a minor one, following the purification procedures in this study. Free sialic acid or neutral monosaccharides migrate to about the position of standard GT lb in the TLC solvent system employed. Larger oligosaccharides or glycopeptides would not move above the origin. Although the liver immunosuppressive fractions were not homogeneous, as seen by TLC analysis (Figure 12 and Figure 13), they did appear to contain only gangliosides, as only resorcinol-positive bands were visible. Evidence for the identification of the liver immunosuppressive lipid components as gangliosides so far rests in their chromatographic behavior on silicic 125 acid column chromatography and thin-layer chromatography, and their retention on a DEAE-Sephadex anion exchange column, which indicates their acidic properties. Their solubility in chloroform:methanol (1:1), chloroform:methanol:water (30:60:8), and methanol point out their amphipathic nature as polar, water-soluble lipids, and their positive reaction with resorcinol identifies their sialic acid components. Subsequent studies with the effect of neuraminidase on the putative ganglioside fractions in regard to their TLC mobility and immunosuppressive activity should verify their tentative identification. Future work in the identification of the immunosupressive gangliosides will depend on obtaining sufficient amounts of both normal and FEL liver gangliosides. With sufficient material, methylation studies, total carbohydrate and sphingosine analysis, and anomeric studies of the carbohydrate linkages could be conducted to elucidate the structure of the immunosuppressive glycolipid. IV. Identification of the FEL Plasma Immunosuppressive Factor Since essentially all circulating gangliosides and neutral glycosphingolipids in plasma were shown to be intimately associated with lipoproteins (267-269), the hypothesis that the inhibitory activity associated with LDL could be accounted for by the ganglioside components (7) was tested by isolation of the ganglioside- enriched Folch upper phase plasma lipids and performance of the PBL mitogenic inhibition assay with the isolated lipids quantitated on the basis of sialic acid at 2, 5, and 10 pg/ml ganglioside concentration. Complete inhibition of the PBL mitogenic response to PHA, Con A, and PWM was found at 5 and 10 pg/ml, but only slight (1%) inhibition was seen at 2 pg/ml. A comparison of the work of Curtiss and Edgington (188) with whole intact LDL as inhibitor showed that if their inhibitory activity was based on the known concentration of ganglioside in LDL, the 126 effective inhibitory concentration of the gangliosides would be comparable to the studies performed with the lipoprotein-derived ganglioside fractions isolated in this study. These results with the Folch upper phase lipoprotein lipids could only be seen as preliminary due to the presence of a multitude of other lipids that also could be partitioned into the Folch upper phase; namely, phospholipids, sulfatides, neutral glycosphingolipids with three or more carbohydrate units length, and even small quantities of neutral lipids. Non-lipid materials such as nucleotides or small oligosaccharides could also be present. More definitive answers to the role of plasma gangliosides would depend on their isolation from other lipids and non- lipids, especially with improved microscale methods (105), which would be useful with pathological samples of limited quantity. Preliminary investigations (7) into the nature of the FEL plasma inhibitor were conducted by isolation of the ganglioside fractions and the other lipids by preparative TLC (Figure 17), and testing their immunosuppressive activities at their equivalent concentrations in plasma before isolation. Initial results were promising as only the strong resorcinol-positive band corresponding to GD 1 b in the TLC analysis (from the plasma lipids labeled "FEL" in Figure 18) had im munosuppressive activity. The isolated neutral lipids were highly cytotoxic. Further studies on the nature of the plasma inhibitor were attempted with permethylation analysis of the intact inhibitor for carbohydrate sequence determi- nation, and with tests on the effect of specific exo-glycosylhydrolase treatment on inhibitory activity. These studies are still in progress. Further purification attempts on the FEL plasma inhibitor revealed the putative Gle-like ganglioside to be free sialic acid, a finding confirmed by GLC analysis and the dialysability of the resorcinol-positive band. There was a small possibility that a ganglioside was present which co-migrated with free sialic acid. If this were true, the potency of the inhibitory factor would be greater than 127 originally believed. However, none of the other FEL plasmas analyzed (no. 127 and 169 in Figure 18) showed any detectable resorcinol-positive bands analogous to the one seen in the original FEL plasma (labeled "FEL" ) studied. Therefore, the identity of a ganglioside as the FEL immunosuppressive factor was still uncertain. The possibility exists that the putative plasma immunosuppressive ganglioside normally is present in nanogram or picogram quantities in plasma and is elevated to a barely detectable level in FEL plasma or liver as a result of the observed liver ganglioside accumulation. The elevated amount of the immunosuppressive ganglioside might just exceed the threshold level at which the ganglioside is im munosuppressive. The ganglioside may even be macrophage-derived and may be elevated to the threshold level due to the abnormal proliferation of histiocytes observed in this disorder. Future work will be directed at the study of macrophage gangliosides and the comparison of those gangliosides to those observed accumulating in FEL liver, particularly to the liver immunosuppressive ganglioside fractions. An important note should be made here that free sialic acid was not immunosuppressive, indicating that the inhibitory activity was authentic even though the identity of the factor was unknown. In Figure 17, a strong iodine- positive band was seen of comparative TLC mobility to the sialic acid-containing band near the origin. This band was common to all the FEL plasmas analyzed and was co-purified along with the sialic acid band in the one FEL plasma tested for inhibitory activity. One might speculate that this band might be the inhibitory factor. Comparative studies of the immunosuppressive activity of isolated gangliosides and other lipids from other FEL plasmas and normal plasmas are still in progress. PBL mitogenesis inhibition tests were also conducted with purified glycosaminoglycans. Keratin sulfate, chondroitin sulfate, and heparan sulfate had 128 no inhibitory activity at equivalent molar concentrations to sialic acid in the FEL plasma. Recent work by others has demonstrated the presence of an uncharacter- ized immunosuppressive factor or factors in normal plasma (270-278), hyperlipidemia Type IV plasma (186), and more specifically in the plasma 8- lipoproteins (186,188,279). Although some of these studies have tentatively identified the factor as an anionic, large-molecular weight protein with o- or B- globulin electrophoretic mobility, the exact nature of the inhibitory factor was still unclear due to the wide variance of biochemical properties attributed to it, as well as to the diverse methods of its isolation. As a result, the common identity or relationship of all these factors reported would be extremely difficult to ascertain. Admittedly, the connection between plasma or LDL ganglioside and the immunosuppressive factor is currently circumstantial, and the connection between the FEL plasma inhibitor and ganglioside even more tenuous. However, the results obtained thus far justify further investigations of the immunoregulatory properties of plasma gangliosides as well as the liver gangliosides, which are the probable source of plasma gangliosides (27.1). SUMMARY Biochemical studies were conducted on the autopsy liver and plasma of a child afflicted with the hereditary disorder, Familial Erythrophagocytic Histiocytosis. Lipid accumulation was found in localized regions of the liver as well as in the infiltrating macrophages. Humoral and cell-mediated immunity defects were noted, and a circulating immunosuppressive factor was found in the hyperlipidemic plasma. Studies were carried out to identify the accumulating lipid material in the liver and to ascertain its relationship to the plasma immunosuppressive factor. Lipid analysis of the F EL liver revealed a mild fatty liver condition with a 2-fold increase in triglyceride levels along with a marked decrease in the ratio of cholesterol esters to free cholesterol. No abnormalities in the neutral glycolipid or phospholipid levels or patterns as determined by TLC were observed. High lipid- bound sialic acid levels, with over an 11-fold increase over the levels of a comparative normal liver were the most striking feature found in the F EL liver. Although a generalized increase of all ganglioside species was found, a major proportion of the gangliosides consisted of the "simple" GM3 and (3M2 type. Preliminary analyses of spleen and plasma showed no concomitant increase in gangliosides even though the plasma was hyperlipidemic. Investigations into other possible sources of the accumulating material were conducted and no accumulation of oligosaccharides or free sialic acid was found in liver. Mucopolysaccharides or glycopeptides in liver were not investigated as the occurrence of a mucopolysaccharidosis or glycopeptide storage disease was unlikely in FEL as presented here. 129 130 To check the possibility of a lysosomal storage disease with a specific glycosylhydrolase deficiency, numerous liver lysosomal enzymes were assayed. Of all the lysosomal enzymes tested, only B-galactosidase activity was found to be decreased to 25% of a comparable normal liver level. Studies on the nature of the reduced B-galactosidase activity revealed the absence of a soluble enzyme inhibitor. No alterations were observed in the electrophoretic pattern of the enzyme and in the pH optimum of activity, nor was a residual heat-stable isoenzyme form responsible for the low enzyme activity level seen in the FEL liver. The evidence pointed to a decrease in activity of all liver lysosomal B- galactosidase isoenzyme forms, although the decrease was not as marked as those observed in GMl gangliosidosis, a lysosomal storage disease with a characterized B-galactosidase deficiency. The low B-galactosidase activity measured in FEL liver was probably not the primary metabolic defect in FEL as fibroblasts and leukocytes from other FEL patients did not exhibit a similar decrease in enzyme activity. No lysosomal enzyme was seen to account for the localized increase of liver gangliosides thus far. Studies on the inhibition of lymphocyte mitogenesis were conducted with isolated liver gangliosides to determine whether the accumulated gangliosides could account for the observed immune deficiency syndrome in FEL. Two ganglioside fractions from normal liver and one fraction from F EL liver were found to have immunosuppressive activity. The ganglioside immunosuppression was unique from the work of others in that the effective levels were lower than reported elsewhere, and in that only specific antigen-stimulated lymphocyte proliferative responses were inhibited by the gangliosides. Plant mitogen stimulat- ion of PBL proliferation was unaffected. Partial characterization of the gangliosides identified a common monosialo—ganglioside as the FEL and normal liver im munoSUppressive factor and a 131 unique disialoganglioside in normal liver with either a lactoneotetraose or lactotetraose sequence for the core oligosaccharide moiety of the glycosphingolipid. However, the structure assignment is only tentative as the ganglioside fractions have not been shown to be homogeneous and the amounts available for characterization were suboptimal. Characterization of the plasma immunosuppressive factor is still in progress, but inhibitory activity could be demonstrated in the plasma lipid extracts. The biochemical basis for FEL is still undetermined and appears to be rather complex in that neutral lipid metabolism, glycolipid metabolism, and immune dysfunction might all be somehow inter-related in the pathogenesis of this inherited disorder. Although the etiology of this disease has not been elucidated, important information in the identification of glycolipids in immunoregulation has been obtained. This is the first known demonstrable instance of possible human visceral ganglioside involvement in immunosuppression, as well as in the possible pathogenesis of an immune dysfunction. The results from this study join the ever- mounting evidence of others of an important biological role for glycosphingolipids in the regulation of mammalian immunological systems. APPENDICES APPENDIX A Case Presentation of the FEL Patient The patient was a two-year-old boy who presented the classic symptoms of the disorder. These symptoms were recurrent fever, anorexia, anemia, hepatosplenomegaly, abnormal liver function, hyper- lipidemia, and prominent erythrophagocytosis in the bone marrow and lymphoid tissues. Widespread tissue infiltration by morphologically normal cells of the macrophage series was also noted. Serum-glutamate- oxaloacetate-transaminase (SGOT), serum-glutamate-pyruvate-transaminase (SGPT), and serum-lactate-dehydrogenase (LDH) were elevated, and serum- haptoglobulin was low. Fibrinogen level was low and the fasting plasma triglyceride level was 420 mg/dl. Absolute leukocyte counts were normal. Total immunoglobulin levels were normal, but antibody titers after primary immunizations to various bacterial antigens and toxoids were abnormally low. Restoration of normal antibody titers to bacterial toxoids was achieved following booster immunizations. The immunological findings have been reported elsewhere (6). At the time of death, the patient reportedly had both bacterial infection (sites unknown) and evidence of liver damage (7). Summaries of the liver biopsy report, autopsy report, and clinical records, which are all available at the National Cancer Institute, were communicated to the author for the purpose of this investigation (7). 132 APPENDIX B [6] Analysis and Structural Characterization of Amino Sugars by Gas-Liquid Chromatography and Mass Spectrometry BY Cur-roan G. WONG, SUN-SANG Josam SUNG, AND CHARLES C. Swmav Department of Biochemistry, Michigan State University, East Lamina, Michigan Introduction Amino sugars are important components in the oligosaccharide structures of glycoproteins, glycosphingolipids, mucopolysaccharides, bacterial pep- tidoglycans, lipopolysaccharides, and antibiotic substances and in the free oligosaccharides of urine and milk. Within the past ten years, combined gas-liquid chromatography-mass spectrometry has become practical for the sub-microgram-scale identification and characterization of these complex carbohydrates. In this chapter are presented the retention behavior of several kinds of amino sugar derivatives on gas-liquid chromatography (glc) and the major ions produced from these substances by electron impact ionization mass spectrometry (ms). The three most common types of derivatization for carbohydrates are acetylation, methylation, and trimethylsilylation. Although the preparation of acetyl derivatives of monosaccharides is a simple technique, there are a few complications. When dealing with alditols produced by borohydride reduction of sugars, borate complexes are formed and can interfere with the acetylation reaction (1). Thus, it is important to remove borate prior to the acetylation step. Another problem may arise in the possible decomposition of sugar acetates on the column, as reported by Bishop et a1. (2), Perry (3) and Gunner et a]. (4). Stellner et a1. (5) have reported very poor recoveries of their partially methylated hexosaminitol acetates due to the inherent design of individual glc-ms models. Trimethylsilylation of sugars, as reported by Sweeley et al. (6), is a simple and rapid method for derivatization. However, it must be kept in mind that the treatment of hexosamine hydrochlorides with trimethylchlorosilane and hexamethyldisilazane in pyridine will not yield silyl derivatives of the amino 55 unions m casement! Copyright (D I980 by Academe Press. Inc. Quinn“. VOL. vm All rights of reproduction in any form reserved. ISBN 0-12 746208-2 133 . 134 56 GENERAL muons or SEPARATION AND ANALYSIS groups. However, use of N, 0-bis(trimethylsilyl)acetamide (BSA) (7, 8) or N,0-bis(trimethylsilyl)trifluoroacetamide (BSTFA) (22, 23) as the tri- methylsilylating reagent has been shown to efl'ectively trimethylsilylate all functional groups. Methylation is an important method for structure elucidation of complex polysaccharides (9). A convenient method for methylation of carbohydrates, giving very high yields of permethylated derivatives, has been developed by Hakomori (10, Vol. VI [64]). Complete methylation of all accessible func- tional groups, including the N-acetamido groups, can be accomplished in one step. The oligosaccharide products themselves have been analyzed by mass spectrometry up to molecular weights approaching 2,000. Subsequent acetylation of the acid-hydrolyzed and reduced alditols yields the partially methylated alditol acetates. Separation of these substances on an ECNSS-M GC column and MS analysis of the eluted components gives structural in- formation of a polysaccharide in regard to glycosidic linkages and carbo- hydrate composition (9). The individual partially methylated alditol acetates . are identified not only by their glc retention times, but also by their charac- teristic fragmentation patterns on the mass spectrometer. Procedures Equipment Relative retention times of the various sugars reported here were deter- mined with a Hewlett-Packard F & M Model 402 gas chromatograph equipped with a hydrogen flame ionization detector. The earrier gas was nitrogen, with flow rates between 40 and 50 ml/min. Glass columns con- taining various packings were 6 h x 2 mm (i.d.). The GC-MS runs were performed on an LKB 9000 gas chromatograph- mass spectrometer, interfaced to a PDP 8/e minicomputer (Digital Equip- ment Co., Maynard, Mass.) for data compilation and analysis. The mass spectrometer was operated at 70 eV with an accelerating voltage of 3.5 kV and an ion source temperature of 290°. The helium carrier gas flow rate was approximately 30 ml/min. Coiled glass columns were 6 ft x 2 mm (i.d.). 2-Acetamido-2-deoxyalditob and 2-Amino-2-deoxyalditob In separate one-dram vials with Teflon-lined screw-caps, about 2.5 mg of each N-acetylhexosamine and hexosamine hydrochloride (Sigma Chem. Co., St. Louis, Mo. and Pfanstiehl Laboratories, Waukegan, 111.) are mixed with 15 mg of sodium borohydride in 1.5 ml of water. The reaction is allowed to proceed overnight (6 h) at 4° and is stopped by the dropwise addition of 135 [6] GLC-MS or AMINO scouts 57 glacial acetic acid until the pH of the solution is acid (pH 2—3) and hydrogen gas can no longer be seen bubbling from the solution. The acidified solutions are then taken to dryness under a stream of nitrogen on a 50° water bath with successive additions of methanol (total volume approximately 20 ml) and evaporation to remove borate completely as the volatile trimethylborate ester. Finally, the dried residues are dissolved in 0.5 ml Of water and used as stock solutions for derivatization. Methyl 2-Acetarnido-2-deoxyglycosides Into a Teflon-lined, screw-capped test tube (10 x 1.3 cm) containing 5 mg of the N-acetylhexosamine, 3 ml of 0.75N anhydrous methanolic HCl (V o1. IV [21], Vol. VI [69], Vol. VII [34]) is added; and the mixture is heated at 80° for 3 h. Losses of solvent from leaky caps are minimized by mo- mentarily loosening the cap after about 10 min heating to reduce the pressure. After methanolysis, powdered silver carbonate is added in small portions to neutralize the reaction mixture (pH 6 by litmus paper test). For further conversion to N-acetyl derivatives, 0.3 ml of acetic anhydride is added to the tubes, and the reaction mixtures are kept at 20°-25° for 6 h. The mixtures are centrifuged; the supernatant fraction is transferred to a one-dram, Teflon-lined, screw-capped vial; and the solvent is removed by a stream Of nitrogen. The silver chloride precipitate is washed twice with 2-ml portions of anhydrous methanol (Vol. VB [3]), and the combined super- natants are quantitatively transferred to one-dram vials and dried down under nitrogen. The methyl-2-acetamido-2-deoxyhexosides are redissolved in 1 ml of water and used as standards (5 mg/ml) for subsequent derivatization. For some biological samples, an incubation time of 18-24 h is preferred for quantitative acid-catalyzed methanolysis. Partially Methylated Alditol Acetate: Permethylation of carbohydrates is done under dry nitrogen by the method of Hakomori (10). Hexane is redistilled after refluxing with 20 g/l of barium oxide for 2 h and is stored over sodium. Dimethyl sulfoxide is dried by refluxing with 50 g/liter of barium oxide for 2 h, redistilled, and stored over molecular sieves (Vol. VI [64], Vol. VII [26]). All other solvents are redistilled. A sample (0.9 g of 57°/o Oil emulsion) of sodium hydride (Alfa Inorganics, Beverly, Mass.) is washed 7 times with lS-ml portions Of dry redistilled hexane. Dry redistilled dimethylsulfoxide (10 ml) is added and allowed to react at 65°-70° for about 90 min, until the bubbling of hydrogen ceased. The methylsulfinyl ion solution (0.5 ml) is added to a solution of 0.5 g of the sample in 0.5 ml of dimethylsulfoxide, and the mixture is allowed to react for 30 min with periodic sonication. Two ml of redistilled iodo- 136 58 GENERAL mops OF SEPARATION AND ANALYSIS methane (methyl iodide) (Pflatz and Bauer, Stamford, Conn.) is then slowly added, and the mixture is allowed to stand for 2 h at 20°—25°. The reaction mixtures are then mixed with 5 ml of chloroform and washed twice with 5 ml of water, once with 5 ml of a 20% solution of sodium thiosulfate (Na28203), and three times with water. The organic phases are evaporated to dryness under nitrogen with the aid of absolute ethanol to remove water by azeotropic distillation, and the residues are hydrolyzed in 0.5 ml of 0.5N H2SO, in 95% acetic acid for 24 h at 85°. Water (0.5 ml) is then added, and heating is continued for an additional 5 b at 85°. A small column containing 2 ml of Dowex 1X8 anion-exchange resin [acetate form] (SO-100 mesh) is used to absorb the sulfate, the permethylated carbohydrates being eluted with 2-3 ml of acetic acid. The hydrolyzate is transferred to a l-dram vial and evaporated to dryness under nitrogen. Reduction with 0.5 ml of sodium borohydride (10 mg/ml) for 2 h at 20-25° yields the partially methylated alditols. After the addition of several drops of glacial acetic acid, the solutions are dried under nitrogen. Borate is removed as its methyl ester, as described above, using 1-2 drops of acetic acid and 2 ml of methanol and heating in a 50° water bath for 5 min under a stream of nitrogen. Esterification is repeated three more times. The dried sample is acetylated in 0.5-l ml of acetic anhydride for 60-90 min at 100°. After drying under nitrogen with the aid of toluene, the sample is dissolved in 2 ml of dichloromethane (methylene chloride), washed three times with 1-2 ml of water, redried under nitrogen, and redissolved in 0.5-l ml of dichloromethane for GC and GC-MS analyses. Partially methylated glucosaminitol (2-amino—2-deoxy-D-glucitol) ace- tates may be synthesized from D-glucosamine hydrochloride by the method of Tai et al. (11). Trimethylsilyl Derivatives Reagent I.—Pyridine (redistilled, stored over KOH), 10 volumes (Vol. II [43], [53], [63], [73]; Vol. IV [73]; Vol. VII [2]). Hexamethyldisilazane (com- mercial reagent), 4 volumes. Trimethylchlorosilane (commercial reagent), 2 volumes. The reagents are added to a 7-ml, screw-capped test tube with a Teflon-lined cap, mixed, and centrifuged. If moisture is excluded, the deriva- tizing solution can be used for 1 week. Reagent II.—N,0—Bis(trimethylsilyl)trifluoroacetamide (BSTFA) contain- ing 1% trimethylchlorosilane (Pierce Chemical Co., Rockford, Illinois). Z-Amino-Z-deoxy-O-trimethylsilylhexosides.—Reagent I (100 pl) is pipeted into dry, l-dram, Teflon-lined, screw-capped vials containing 125 pg Of amino sugar. The mixture is allowed-to stand at 20°-25° for 30 min. An 137 [6] GLS—MS or AMINO SUGARS 59 appropriate aliquot (1-3 pl) is injected immediately into the gas chromato- graph, for the N-trimethylsilyl hexosamine derivatives are present in appre- ciable amounts after 2 h at room temperature. 2-Deoxy-2-trirnethylsilylamino-0—trimethylsilylhexosides.—Into l-dram Tef- lon-lined screw-capped vials containing 125 pg of amino sugar is added 50 pl of dry pyridine, followed by 50 pl of BSTFA (Reagent II). The sealed vial is heated at 80° for 30 min, and an aliquot is injected into the GC. (Note: N-acetyl derivatives do not form any N-trimethylsilyl amide under these conditions). Acetate Derivatives Acetic anhydride (100 pl) and dry pyridine (100 pl) are added to dry, l-dram, Teflon-lined, screw-capped vials containing 250 pg of amino sugar. Thesealedvialsareheatedat 100° for4h,afterwhich2mlofredistilled toluene is added and the mixture is dried by evaporation under a stream of nitrogen at 50°. This addition of toluene and subsequent evaporation are repeated once more to ensure the complete removal of acetic anhydride and pyridine. A solution of the acetylated sugar in 200 pl of dry, redistilled methylene chloride is used for GC analysis. Results Tables I-XII are a summary of the relative GC retention times and the major ions found in the mass spectra of each denoted amino sugar. Since stereoisomers and anomers of the carbohydrate derivatives give similar mass spectra, with small differences in peak intensity, the mass spectrum of only one stereoisomer is given. 138 ‘I'AaLE l Retention Times of the Peracetylated Amino Sugars Relative Other Compound retention time‘ references 2—Acetamido-1.3.4.6-tetrs-O-acetyl-2-deoxy-D-glucose’ 0.36. 2.57 2-Acetamido- l,3.4,6-tetra-O—acetyl-2-deoxy-D-ga1actose' 0.37, 0.43, 2.53, 2.77 2-Aoetamido- l,3.4,6-tetrs-O-acetyl-2-deoxy-D-mannose’ 0.46, 2.33, 2.48, 2.90 Z-Acetamido- l.3,4,5,6-pents-O-acetyl-Z-deoxy-D-glucitol‘ 1.76 (12) 2-Acetsmido-l.3.4,5.6-penta-0-acetyl-2-deoxy-D-galactitol‘ 2.34 (12) Z-Acetamido-l,3,4,5,6-penta-O-scetyl-Z-deoxy-D-mannitol‘ 2.83 (12) ‘ Retention times are relative to 1.2.3.4.5.6.7-hepta-0-acetylperseitol on a column of 3'/. Poly A-103 on Gas Chrom Q “ll/120 mesh (Applied Science Laboratories. Inc, State College, Pa). 'Isothermal at 200°; internal standard retention time was 11 min. ‘ Isothermal at 210°: internal standard retention time was 6.8 min. TAILS 11’ Retention Times of the Z-Acetamido-Z-deoxy—O-trimethylsilyl Sugar: Relative Compound‘ retention time' 2-AcetamidO-2-deoxy-1.3.4.6-tetra-O-trimethylsilylglucopyranoside 0.93, 1.69 2-Acetamido-2-deoxy-l,3,4,6-tetra-O-trimethylsilylgalactopyranoside 1.35, 1.56 2-Acetamido-2-deoxy-l,3,4,6-tetra-0-trimethylsilylmannopyrsnoside 0.95. 1.27 Methyl 2-acetamido-2-deoxy-3,4,6-tri-O-trimethylsilylglucopyrsnoside 1.31, 1.41.156 Methyl 2-acetamido-2-deoxy-3,4,6—tri-O-trimethylsilylgalactopyranoside 1.14, 1.37 Methyl 2-acetamido-2-deoxy-3,4.6-tri-O-trimethylsilylmannopyranoside 0.94. 1.60 Z-Acetamido-Z-deoxy-1,3,4,5,6-pents-0-trimethylsilylglucitol 1.69 2-AcetamidO-2-deoxy-15.4.5.6-penta-O-trimethylsilylgalactitol 1.77 2-Acetsmido-2-deoxy-1.3.4.5.6-penta-O-trimetbylsilylmannitol 1.88 ‘ Derivatized with either Reagent l or II. ' Retention times are relative to 1.2.3.4. 5.6-hexa-0-trimethy1silylmannitol (10.5 min) on 3‘/, SP 2100 on Supelcoport 801100 mesh (Supelco, Inc. Bellefonte. Pa.) at column temperature of 180° isothermal. 'I'AaLE III Retention Times of the Z-Ammo-Z-deoxy-O-trimethylsily Alditols Relative Compound retention time‘ 2-AminO-2-deoxy-1.3.4.5,6-penta-O-trimethylsilylglucitol' 1.22 2-Amino-2-deoxy-1,3,4,5.6-penta-0-trimethylsilylgalactitol’ 1.18 Z-Amino-Z-deoxy-1.3.4,5,6-penta-0-trimethylsilylmannitol‘ 1.23 2-Deoxy-2-trimethylsilylamino-1.3,4.5,6-penta-O-trimethylsilylglucitol‘ 0.92 2- Deoxy-Z-trimethylsilylamino— 1.3.4.5,6-penta-0-trimethylsilyl galactitol ‘ 0.91 2-Deoxy-2-trimethylsilylamino-1.3.4,5,6-penta-0-trimethylsilylmannitol‘ 0.92 ‘ Retention time relative to 1. 2. 3.4. 5. 6-hexa-O-trimethylsilylmannitol (10.4 min) on 37. SP 2100. Supelco- port 80/111) mesh (Supelco. Inc, Bellefonte, Pa.) at column temperature of 180° isothermal. ’ Derivatized with trimethylsilylating Reagent I. ‘Derivstized with trimethylsilylating Reagent II. 60 139 I‘m IV Retention Times of the Partially O-Methylated Z-N-methylglucosaminitol Acetates Position of O-CH, groups Relative retention time‘ 3, 4, 6 1.1!) (retention time - 8.3 min) 3, 6 1.68 3, 4 2.19 4, 6 2.51 3 2.91 4 3.64 6 4.51 ‘Isothermal at 190°, 3% OV-210 on Supelcoport 80/ 11!) mesh (Supelco. Inc, Bellefonte. Pa.). TAaLE VI Major Fragment Ions Observed in the Mass Spectrum of Tm v 2-Aeetamrdo—l.3,4,5.6-penta-0— Major Fragment I ans Observed in acetyl-Z-deoxy-o-glucitol the Mass Spectrum of (M W - 433)‘ 2-A cetamido-l, 3, 4, 6-tetra-0-acetyl- 2-deoxy-o-galactopyranose M/t 3010““ intensity (MW - 389)‘ 43 IND [C1‘I;,CO]+ m/e Relative Intensity 60 21.4 84 73.8 43 100.0 [CI-1;,CO]+ 85 20.3 72 10.9 102 22.9 84 10.4 114 8.9 97 6.9 1 15 7.9 108 3.3 126 15.1 110 0.9 139 12.5 114 48.9 144 23.0 (M’-289) 126 9.1 145 7.7 139 13.6 151 9.4 144 1.3 156 8.6 150 4.7 157 4.8 156 19.7 168 7.4 168 6.6 216 1.5 181 7.2 217 1.4 198 5.0 288 0.3 199 6.7 289 0.4 210 1.8 318 8.3 (M’-73—42) 241 14.3 360 1.1 (M’-73) 330 2.6 (M’-59) 374 0.2 (M’-59) 346 1.5 (M’-43) 390 0.1(M’-43) ‘ References 13 and 14 give detailed ‘ References 15 and [6 give detailed descriptions of fragmentation pathways descriptions of fragmentation pathways and Identifications of ions. and identifications of Ions. 61 140 Tm VII Major Fragment Ions Observed in the Mass Spectra of Partially 0- Methylated 2- N- Methylglucosaminitol Acetate: Position of CH30- groups m/e 3,4,6 3,6 3,4 4,6 3 4 6 43 + + + + + + + 45 + + + + + + + 74 + + + + + + + 87 + + + + + + + 98 + + + + + + + 1 16 + + + + + + + 124 + + 128 + 129 + + + + + + 142 + + + + + + 145 + + 158 + + + + + + + 161 + + + 170 + + + + 173 + 189 + + 202 + + + + 205 + 230 + 233 + 261 + 274 + ‘Referenees 5, 9, ll. 16. I 7, and I8 give detailed descriptions of fragmentation pathways and identifications of ions. 62 Tau; VIII Major Fragment Ions Observed in the Mass Spectrum of 2-Acetamido-2-deoxy-l,3,4,5,6— penta-O-trimethylsilylmannitol (MW - 583)‘ We Relative intensity 73 1111.0 [(C1-1,),Si]+ 103 18.5 132 23.1 147 23.8 157 16.9 174 14.9 186 29.4 205 18.2 217 25.1 247 13.2 276 12.4 319 18.4 378 7.6 (LP-205) 390 4.0(M’-90-103) 478 1.5 (M’-15—90) 480 1.9 (hf-103) 568 6.1 (M ’ ~15) ‘Referenees 8. 16. and I9 give detailed descriptions of fragmentation pathways and identifications of ions. TABLE X Major Fragment Ions Observed in the Mass Spectrum of 2-Deoxy-2-trimethylsilylamino— I ,3, 4,5 ,6-penta-O-trimethylsilyl- o-mannitol (MW - 613)“ m/e Relative intensity 73 46.9 [(CH,),Si] * 103 6.5 204 100.0 205 21.1 217 5.0 307 1.4 420 2.0 (M*-103-90) 510 1.8 (M*-103) 598 0.5 (hr-15) ‘ References 8. 16, and 19 give detailed descriptions of fragmentation pathways and identifications of ions. 141 63 Tau 1x Major Fragment Ions Observed in the Mass Spectrum of 2-Arnino—2-deoxy-I , 3. 4, 5 , 6-penta-0- trimethylsilyl-o-glucitol (M W - 541)‘ m/e Relative intensity 73 83.2 [(CH,),Si] ‘ 103 20.2 132 32.1 147 17.5 204 25.7 205 10.9 217 100.0 258 8.8 348 7.2 (M*-103-90) 438 6.6 (hf—103) 451 0.3 (LP-90) 526 5.7 (M’olS) “References 8, l6.and 19 give detailed descriptions of fragmentation pathways and identifications of ions. Tm XI Major Fragment Ions Observed in the Mass Spectrum of Methyl 2-Acetamido-2-deoxy- 3, 4 , 6- tri-O-trimethylsilyl-D - galaetOpyranoside (M W - 45 I)‘ m/e Relative intensity 73 91.6 [(C1-1,),Si]+ 75 14.4 131 26.5 147 20.9 173 100.0 204 10.9 217 7.6 218 10.9 226 0.8 247 11.2 259 3.9 314 2.8 330 0.5 (M’-31-90) 346 O.4(M*-15-90) 'Referenees 8, I6, 20, and 21 give detailed descriptions of fragmentation pathways and identifications of ions. 142 GENERAL METHODS OF SEPARATION AND ANALYSIS Tm xn Major Fragment Ions Observed in the Mass Spectrum of 2-Acetarmdo-2-deoxy-I,3,4,6- tetra-OotrimethylsilyI-o- galactopyranoside (M W - 509)‘ m/e Relative intensity 73 73.0 [(CH,),Si] ’ 103 5.7 117 5.8 131 21.7 147 13.8 173 100.0 204 10.4 217 9.5 233 3.9 305 1.5 314 3.2 (M*-15-90-90) 404 0.8 (M*-15-90) 494 1.6 (M*-15) ‘Referenw 8, I6, 20, and 21 give detailed descriptions of fragmentation pathways and identifications of ions. Acknowledgments WewishtothankMr.Jack1-IartenandDr.FrankMartinfortheirtechnicalassistancein this work. This work was supported in part by grants from the National Institute of Arthritis, Metabolism and Digestive Diseases (AM 12434) and the Biotechnology Resources Branch (RR 00480) of the National Institutes of Health. (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) References Blake and G. N. Richards, Carbohydr. Res., 14, 375 (1970). Bishop, F. P. Cooper, and R. K. Murray, Can. J. Chem., 41, 2245 (1963). 0“ T. B. Perry, Can. J. Biochem, 42, 451 (1964). W. Gunner, J. K. N. Jones, and M. B. Perry. Can. J. Chem., 39, 1892 (1961). Stellner, 1'1. Saito, and S.-I. Hakomori, Arch. Biochem. Biophys., 155, 464 (1973). . . Sweeley, R. Bentley, M. Makita, and W. W. Wells, J. Amer. Chem. Soc., 85, 2497 3). . ebe, H. Finkbeiner, and D. M. White, J. Amer. Chem. Soc., 88, 3390 (1966). J. Kirkkfinen and R. Vihko, Carbohydr. Res., 10, 113 (1969). B. Lindberg, Methods Enzymol., 28, 178 (1972). S.-I. Hakomori, J. Biochem. (Tokyo), 55, 205 (1964). 1'. Tai, K. Yamashita, and A. Kobata, J. Biochem. (Tokyo), 78, 679 (1975). Z . D “COW“ “an (12) (13) (14) (l5) (16) (17) (18) (19) (20) (21 ) (22) (23) 143 [6] GLC-MS or AMINO squns 65 W. Niedermeier and M. Tomana. Anal. Biochem., 57, 363 (1974). K. Heyns and H. Scharmann, Justus Liebigs Ann. Chem., 667, 183 (1963). R. C. Daugherty, D. Horton, K. D. Philips, and J. D. Wander, Org. Mass Spectrom., 7, 805 (1973). H. Bjérndal, C. G. Hellerqvist, B. Lindberg, and S. Svensson, Angew. Chem., biternat. Ed, 9, 610 (1970). J. LOnngren and S. Svensson, Adv. Carbohydr. Chem. Biochem., 29, 41 (1974). W. Stofi'el and P. Hanfiand. Hoppe-Seyler‘s Z. Physiol. Chem., 354, 21 (1973). G. O. H. Schwarzmann and R. W. Jeanloz, Carbohydr. Res., 34, 161 (1974). M. Dizdaroglu, D. Henneberg, and C. van Sonntag, Org. Mass Spectrom., 8, 335 (1974). D. C. Waugh. ‘1'. Redford, J. D. Hribar, S. Hanessian, M. Bieber, G. Dawson, and C. C. Sweeley, J. Amer. Chem. Soc., 91, 1728 (1969). P. I... Coduti and C. A. Bush, Anal. Biochem., 78, 21 (1977). C. W. Gehrke and K. Leimer, J. Chromatogr., 57, 219 (1971). R. E. Hurst, Carbohydr. Res., 3, 143 (1973). BIBLIOGRAPHY 7. 10. 11. 12. 13. 14. 15. 16. 17. 18. BIBLIOGRAPHY Farqmar, ILW. and A.E. Claireaux, Arch. Dis. Child. _21 (1952) 519-525. Mozziconacci, P., C. Nezelof, C. Attal, F. Girard, Pham-Huu-Trung, J. Weil, B. Desbuquois, and M. Gadot, Arch. Fr. Pediat. _2_; (1965) 385-408. Botha, J.B.C., L.B. Kahn, R.O.C. Kaschula, S. Afr. Med. 3. £2 (1975) 1305- 1308. Landrieu, P. and 3).-J. Choulot, Arch. Fr. Pediat. 3_3_ (1976) l197-503. Perry, M.C., E.G. Harrison, .‘lr., 13.0. Burgert, Jr., and (1.5. Gilchrist, Cancer 2_8_ (1976) 209-218. Ladisch, S., B. Holiman, D.G. Poplack, and RM. Blaese, Lancetl(l978) 581- 583. Ladisch, S., unpublished observations. Thudichum, J.L.W., Rep. Med. Officer (London) _3_ (1874) 113. Gardas, A. and J. Koscielak, FEBS Letters 92 (1974) 101-104. Koscielak, 3., H. Miller-Podraza, R. Krauze, and A. Piasek, Eur. J. Biochem. _7_l (1976) 9-18. Martensson, E., Prog. Chem. Fats Other Lipids _12 (1969) 365-407. Stoffel, W., Ann. Rev. Biochem. 59 (1971) 57-82. McKibbin, .‘l.M. in "The Carbohydrates", vol. ZB, W. Pigman and D. Horton, edit., second edition, Academic Press, N.Y., (1970) pp. 711-738. Laine, R.A., K. Stellner, and S. Hakomori, Meth. Membrane Biol. _2_ (1974) 205-244. Sweeley, C.C. and B. Siddiqui, in "The Glycoconjugates", vol. 1, M.1. Horowitz and W. Pigman, edit., Academic Press, N.Y, (1977) pp. l459-540. Klenk, E., Hoppe-Seyler's Z. Plysiol. Chem. 235 (1935) 24-36. Blix, G., Skand. Arch. Physiol. Q (1938) 46-51. Klenk, E., Hoppe-Seyler's Z. Physiol. Chem. 273 (1942) 76-86. 144 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 145 Yamakawa, T. and S. Suzuki, J. Biochem. (Tokyo) _3_8 (1951) 199-212. Fishman, P.H. and R.O. Brady, Science 194 (1976) 906-915. Yamakawa, T. and Y. Nagai, Trends in Biochem. Sci. _3_ (1978) 128-131. Nagai, Y. and M. Iwamori, Molec. Cell. Biochem. 22 (1980) 81-90. Mullin, B.R., P.H. Fishman, G. Lee, S.M. Aloj, F.D. Ledley, R.J. Winand, L.D. Kohn, and R.O. Brady, Proc. Natl. Acad. Sci. U.S.A. 1_3 (1976) 842-846. Lee, G., S.M. Aloj, and L.D. Kohn, Biochem. Biophys. Res. Commun. _7_7_ (1977) 434-441. Lee, G., S.M. Aloj, R.O. Brady, and L.D. Kohn, Biochem. Biophys. Res. Commun. 2 (1976) 370-377. Haywood, A.M., J. Mol. Biol. 82 (1974) 427-436. Holmgren, J., L. Svennerholm, H. Elwing, P. Fredman, and 5. Strannegard, Proc. Natl. Acad. Sci. U.S.A. 110980) 1947-1950. Besancon, F. and H. Ankel, Nature 222 (1974) 478-480. Vengris, V.E., P.H. Reynolds, Jr., MD. Hollenberg, and PM. Pitha, Virology Z; (l976)486—493. Woolley, D.W. and B.W. Gommi, Proc. Natl. Acad. Sci. U.S.A. 5_3_ (1965) 959- 963. Kato, I. and M. Naiki, Infect. Immunol. 1_3 (1976) 289-291. Simpson, L.L. and MM. Rapport, J. Neurochem. 18 (1971) 1751-1759. van Heyningen, W.E. and P. Miller, J. Gen. Microbiol. _23 (1961) 107-119. van Heyningen, W.E., J. Gen. Microbiol. fl (1963) 375-387. van Heyningen, W.E., C.C.J. Carpenter, N.F. Pierce, and W.B. Greenough 111, J. Infect. Dis. 124 (1971) 415-418. Holmgren, J ., I. Lonnroth, L. Svennerholm, Infect. Immun. 8 (1973) 208-214. King, C.A. and W.E. van Heyningen, J. Infect. Dis. 127 (1973) 639-647. Cuatrecasas, P., Biochem. 12 (1973) 3547-3558, 3558-3566, 3567-3576,3577- 3581. Carrion, A., R. Bourrillon, and J.A. Cabezas, Rev. Esp. Fisiol. 26 (1970) 171- 174. Cabezas, J .A. and M. Ramos, Carbohyd. Res. _2_4 (1972) 486-488. Svennerholm, L., J. Neurochem. fl (1963) 613-623. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 146 Wiegandt, H., Adv. Lipid Chem. 2 (1971) 249-289. IUPAC-ICB Commission on Biochemical Nomenclature, Lipids 22 (1977) 455- “68. Macher, B.A. and C.C. Sweeley, Meth. Enzymol. 22 (1978) 236-251. Klenk, E. and K. Lauenstein, Hoppe—Seyler's Z. Physiol. Chem. 291 (1952) 249—258. Yamakawa, T., M. Matsumoto, and S. Suzuki, J. Biochem. (Tokyo) fl (1956) 63-72. Yamakawa, T., R. Irie, and M. Iwanaga, J. Biochem. (Tokyo) _4_8_ (1960) 490- 507. Kuhn, R. and H. Wiegandt, Z. Naturforsch. B. _12 (1964) 80-81. Kuhn, R. and H. Wiegandt, Z. Naturforsch. B. _12 (1964) 256-257. Ledeen, R., R.K. Yu, and L.F. Eng, J. Neurochem. 2_1(1973)829-839. Hooghwinkel, G.J.M., P.F. Borri, and G.W. Bruyn, Neurology, _1_6_ (1966) 934- 936. Ledeen, R., K. Salsman, and M. Cabrera, Biochem. 2 (1968) 2287-2295. Ledeen, R., K. Salsman, J. Gonatas, and A. Taghavy, J. Neuropathol. Exp. Neurol. 24 (1965) 341-351. Seyfried, T.N., S. Ando, and R.K. Yu, J. Lipid Res. 29 (1978) 538-543. Vanier, M.T., M. Holm, J.E. Manson, and L. Svennerholm, J. Neurochem. 22 (1973) 1375-1384. Svennerholm, L., Acta Chem. Scand. 29 (1965) 1506-1507. Svennerholm, L., Acta Chem. Scand. 20 (1956) 1048. Svennerholm, L., Biochem. Biophys. Res. Commun. 2 (1962) 436-411. Wiegandt, H., Hoppe—Seyler's Z. Physiol. Chem. 354 (1973) 1049-1056. Klenk, E. and M. Naoi, Hoppe-Seyler's Z. Physiol. Chem. 349 (1968) 288-292. Kuhn, R. and H. Wiegandt, Chem. Ber. 26 (1963) 866-880. Kuhn, R. and H. Wiegandt, Z. Naturforsch. B. 28 (1963) 541-543. Ando, S. and R.K. Yu, J. Biol. Chem. 252 (1977) 6247-6250. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 147 Macher, B.A., T. Pacuszka, B.R. Mullin, C.C. Sweeley, R.O. Brady, and P.H. Fishman, Biochim. Bigahys. Acta 588 (1979) 35-43. Ghidoni, R., S. Sonnino, G. Tettamanti, H. Wiegandt, and V. Zambotti, 2: Neurochem. 22 (1976) 511-515. Iwamori, M. and Y. Nagai, J. Biochem. (Tokyo) fl (1978) 1601-1608. Ando, S. and R.K. Yu, Abstr. Fourth Int. Symp. Glycoconjugates, Woods Hole, (1977), l. Siddiqui, B. and S. Hakomori, Biochim. Biophys. Acta 330 (1973) 147-155. Wherret, J.R., Biochim. Biophys. Acta 326 (1973) 63-73. Yu, R.K. and R.\V. Ledeen, J. Lipid Res. 22 (1972) 680-686. Li, Y.-T., J.F. Manson, M.-T. Vanier, and L. Svennerholm, J. Biol. Chem. 228 (1973) 2634-2636. Wiegandt, H., Eur. J. Biochem. 42 (1974) 367-369. Rauvala, H., T. Krusius, and J. Finne, Biochim. Biophys. Acta 531 (1978) 266-274. Siddiqui, 3., 3.5. Whitehead, and Y.S. Kim, J. Biol. Chem. 253 (1978) 2168- 2175. Watanabe, K., S.-1. Hakomori, R.A. Childs, and T. Feizi, J. Biol. Chem. 254 (1979)3221-3228. Watanabe, K., M. Powell, and S.-I. Hakomori, J. Biol. Chem. 253 (1978) 8962-8967. Feizi, T., R.A. Childs, S.-I. Hakomori, and M.E. Powell, Biochem. J. 173 (1978)245-254. Niemann, H., K. Watanabe, S. Hakomori, R.A. Childs, and T. Feizi, Biochem. Biophys. Res. Commun. _82 (1978) 1286-1293. Suzuki, A., I. Ishizuka, and T. Yamakawa, J. Biochem. (Tokyo) _72 (1975) 947- 954. Sonnino, S., R. Ghidoni, G. Galli, and G. Tettamanti, J. Neurochem. _32 (1978)947-956. Ohashi, M. and T. Yamakawa, J. Biochem. (Tokyo) fl (1977) 1675-1690. Baumann, H. and S. Hakomori, Proc. Am. Assoc. Cancer Res. _12 (1978) 227. Folch-Pi, J ., Adv. Exp. Med. Biol. 22 (1972) 63-76. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 148 Brunngraber, E.G., G. Tettamanti, and B. Berra, in "Glycolipid Methodology", L.A. Witting, edit., American Oil Chem. Soc., Champaign, Il, (1976) pp. 159-186. F olch, J., M. Lees, G.H. Sloane-Stanley, J. Biol. Chem. 226 (1957) 497-509. Suzuki, K., J. Neurochem. _l_2 (1965) 629-638. Tettamanti, G., F. Bonali, S. Marchesini, and V. Zambotti, Biochim. Biophys. Acta 292 (1973) 160-170. Trams, E.G. and Lauter, C.J., Biochim. Bioghys. Actg, 6_0_ (1962) 350-358. Rouser, G., D. Kritchevsky, D. Heller, and E. Lieber, J. Am. Oil Chem. Soc. 39 (1963) 425-454. Penick, R.J., M.H. Meisler, and R.H. McCluer, Biochim. Biophys. Acta 116 (1966) 279-287. Ghidoni, R., S. Sonnino, and G. Tettamanti, Lipids 1_3_ (1978) 820-822. Sandhoff, K., K. Harzer, and H. Jatzkewitz, Hoppe-Sejler's Z. Physiol. Chem. 2Q (1968) 283-287. Rouser, G., D. Kritchevsky, and A. Yamamoto, in "Lipid Chromatographic Analysis", vol. 1, G.V. Marinetti, edit., Marcel Dekker, New York, (1967) pp. 99-162. Iwamori, M. and Y. Nagai, Biochim. Biophys. Acta 528 (1978) 257-267. Kundu, S.K. and S.K. Roy, J. Lipid Res. _l_9_ (1978) 390-395. Fredman, P., O. Nilsson, J.-L. Tayot, and L. Svennerholm, Biochim. Biophys. Acta 212 (1980) 42-52. Momoi, T., S. Ando, and Y. Nagai, Biochim. Biophys. Acta 441 (1976) 488- 497. Ando, S. M. Isobe, and Y. Nagai, Biochim. Biophys. Acta 424 (1976) 98-105. Ripphahn, J. and H. Halpaap, J. Chromat. 112 (1975) 81-96. Ando, S. N.-C. Chang, and R.K. Yu, Anal. Biochem. 22 (1978) 437-450. Zanetta, J.-P., F. Vitiello, and J. Robert, J. Chromat. 137 (1977) 481-484. Harth, S., H. Dreyfus, P.F. Urban, and P. Mandel, Anal. Biochem. g (1978) 543-551. Irwin, C.C. and L.N. Irwin, Anal. Biochem. 24 (1979) 335-339. Evans, J.E. and R.H. McCluer, Biochim. Biophys. Acta 270 (1972) 565-569. Ullman, M.D. and R.H. McCluer, J. Lipid Res. _l_8_ (1977) 371-378. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 118. 119. 120. 121. 122. 123. 124. 125. 126. 127. 128. 129. 130. 131. 149 Suzuki, A., S. Handa, and T. Yamakawa, J. Biochem. (Tokyo) _82 (1977) 1185- 1 187. Yamazaki, T., A. Suzuki, S. Handa, and T. Yamakawa, J. Biochem. (Tolgo) 8_6 (1979) 803-809. Tjaden, U.R., J.H. Krol, R.P. Van Hoeven, E.P.M. Oomen-Meulemans, and P. Emmelot, J. Chromat. 136 (1977) 233-243. Vance, B.E. and C.C. Sweeley, J. Lipid Res. 2 (1967) 621-630. Chambers, R.E. and J.R. Clamp, Biochem. J. 125 (1971) 1009-1018. Yu, R.K. and R.W. Ledeen, J. Lipid Res. _l_l (1970) 506-516. Gaver, R.C. and C.C. Sweeley, J. Am. Oil Chem. Soc. 4_2 (1965) 294-298. Samuelsson, K. and B. Samuelsson, Chem. Phys. Lipids _5 (1970) 44-79. Lauter, C.J. and E.G. Trams, J. Lipid Res. 2 (1962) 136-138. Naoi, M., Y.C. Lee, and S. Roseman, Anal. Biochem. _5_8_ (1974) 571-577. Ackman, R.C., Meth. Enzymol. 22 (1969) 329-382. Tschb pe, G., HoppeSeyler's Z. Physiol. Chem. 354 (1973) 1291-1298. Ginsburg, V., editor, Meth. Enzymol. 22 (1972) 699-897. Kobata, A., Anal. Biochem. 100 (1979) 1-14. \Venger, D.A. and S. Wardell, J. Neurochem. 22 (1973) 607-612. Rauvala, H., FEBS Letters §_5_ (1976) 229-233. Sugano, K., M. Saito, and Y. Nagai, FEBS Letters _8_2 (1978) 321-325. Falk, K.-E., K.-A. Karlsson, and B.E. Samuelsson, Arch. Biochem. Biophys. 122 (1979) 164-176, 177-190, 191-202. Sillerud, L.O., J.H. Prestegard, R.K. Yu, D.E. Schafer, and W.H. Konigsberg, Biochem. _12 (1978) 2619-2628. Harris, P.L. and E.R. Thornton, J. Am. Chem. Soc. 100 (1978) 6738-6745. Sweeley, C.C., Y.-K. Fung, B.A. Macher, J.R. Moskal, and H.A. Nunez, in "Glycoproteins and Glycolipids in Disease Processes" , E.F. Walborg, Jr., edit., Am. Chem. Soc., Washington, D.C., (1978) pp.47-85. Laine, R.A. and O. Renkonen, J. Lipid Res. l_6_ (1975) 102-105. 132. 133. 134. 135. 136. 137. 138. 139. 140. 141. 142. 143. 144. 145. 146. 147. 148. 149. 150. 151. 152. 150 ijrndal, H., C.-G. Hellerquist, B. Lindberg, and S. Svensson, Angew. Chem., Int. Ed. Enp2. 2 (1970) 610-619. Lindberg, B. Meth. Enzymol. 22 (1972) 178-195. Rauvala, H. and J. Khrrk‘hinen, Carbohyd. Res. _5Q (1977) 1-9. Sweeley, C.C. and G. Dawson, Biochem. Biophys. Res. Commun. 22 (1969) 6- l#. Karlsson, K.-A., l. Pascher, W. Pimlott, and B.E. Samuelsson, Biomed. Mass Spectrom. 2(1974149-56. Karlsson, K.-A., Biochem. 22 (1974) 3643-3647. Karlsson, K.-A., Prog. Chem. Fats other Lipids 26 (1978) 207-230. Wiegandt, H. and G. Baschang, Z. Naturforsch. B. 2_0 (1965) 164-166. Hakomori, 5., J. Lipid Res. 2 (1966) 789-792. Ando, S., K. Kon, Y. Nagai, and T. Murata, J. Biochem. (Tokyo) 2 (1977) 1623-1631. Roseman, 5., Chem. Phys. Lipids 2 (1970) 270-297. Schachter, H., in "The Glycoconjugates", vol. 2, M.I. Horowitz and W. Pigman, edit., Academic Press, N.Y., (1977) pp. 87-181. Dawson, G., in "The Glycoconjugates", vol. 2, M.I. Horowitz and W. Pigman, edit., Academic Press, N.Y., (1977) pp. 255-284. Wherret, J.R., in "Glycolipid Methodology", L.A. Witting, edit., Am. Oil Chem. Soc., Champaign, Il, (1976) pp.215-232. Patel, V., in "The Glycoconjugates", vol. 2, M.I. Horowitz and W. Pigman, edit., Academic Press, N.Y., (1977) pp.185-229. Dawson, G., in "The Glycoconjugates", vol. 2, M.l. Horowitz and W. Pigman, edit., Academic Press, N.Y., (1977) pp. 285-336. Neufeld, E.F., G.N. Sando, A.J. Garvin, and L.H. Rome, J. Supramolec. Struc. g (1977)95-101. Sly, W.S., in "Structure and Function of the Gangliosides", L. Svennerholm, P. Mandel, H. Dreyfus, and P.-F. Urban, edit., Plenum Press, N.Y., (1980) pp. 433-451. Sly, W.S., A. Gonzalez-Noriega, M. Natowicz, H.D. Fischer, and J.P. Chambers, Fed. Proc. 28_ (1979) 467. Avila, J.L. and J. Convit, Biochem. J. 160 (1976) 129-136. Hechtman, P., Can. J. Biochem. _5_2 (1977) 315-324. 153. 154. 155. 156. 157. 158. 159. 160 161. 162. 163. 164. 165. 166. 167. 168. 169. 170. 171. 172. 151 Li, S.-C., and Li, Y.-T., J. Biol. Chem. 251 (1976) 1159-1163. Mraz, W., G. Fisher, and H. Jatzkewitz, FEBS Letters 27: (1976) 104-109. Hall, C.W., I. Liebaers, P. Di Natale, and E.F. Neufeld, Meth. Enzymol. 29 (1978) 439-456. Sandhoff, K., Angew. Chem.,lnt. Ed. Engl. 22 (1977) 273—285. Neufeld, E.F., T.W. Lim, and L.J. Shapiro, Ann. Rev. Biochem. 22 (1975) 357-376. Stanbury, J.B., J.B. Wyngaarden, and D.S. Fredrickson (editors), in "The Metabolic Basis of Inherited Disease", fourth edition, McGraw-Hill, N.Y. (1978). Dorfman, A. and R. Matalon, Proc. Natl. Acad. Sci. U.S.A. 22 (1976) 630- 637. Brady, R.O., Angew. Chemfllnt. Ed. Engl. 22 (1973) 1-11. Suzuki, K., Meth. Enzymol. 22 (1978) 456-488. Alhadeff, J.A., A.L. Miller, D.A. Wenger, and J.S. O'Brien, Clin. Chim. Acta 2 (1974) 307-313. Spranger, J.W. and H.-R. Wiedemann, Humangenetik 2 (1970) 113-139. Cantz, M., J. Gehler, and J. Spranger, Biochem. Biophys. Res. Commun. fl (1977) 732-738. Tallman, J.F. and R.C. Brady, Biochim. BIOphys. Acta, 293 (1973) 434-443. Strecker, G., J.C. Michalski, J. Montreuil, and J.P. Farriaux, Biomed. 22 (1976) 238-239. Suzuki, Y. and K. Fukuoka, Clin. Chim. Acta 22 (1979) 107-112. Berra, B., S. DiPalma, C. Lindi, and K. Sandhoff, Cell. Molec. Biol. 22 (1979) 281-284. ' Touster, O., N.N. Aronson, J.T. Dulaney, and H. Hendrickson, J. Cell Biol. 11 (1970) 604-618. Schengrund, C.-L., D.S. Jensen, and A. Rosenberg, J. Biol. Chem. 247 (1972) 2742-2746. O'Brien, J.S. in "The Metabolic Basis of Inherited Disease", J.B. Stanbury, J.B. Wyngaarden, and D.S. Fredrickson, edit., third edition, McGraw-Hill, N.Y., (1972) Pp. 639-6620 Wenger, D.A., M. Sattler, C. Clark, H. Tanaka, K. Suzuki, and G. Dawson, Science 188 (1975) 1310-1312. 173. 174. 175. 176. 177. 178. 179. 180. 181. 182. 183. 184. 185. 186. 187. 188. 189. 190. 191. 192. 193. 152 Rushton, A.R. and G. Dawson, Biochim. Biophys. Acta, 388 (1975) 92-105. Ockerman, P.-A., B. Hultberg, and O. Erickson, gin. Chim. Acta 2_5_ (1969) 97-102. MacBrinn, M., S. Okada, M. Woollacott, V. Patel, M.W. Ho, A.L. Tappel, and J.S. O'Brien, N. Engl. J. Med. 281 (1969) 338-343. Kint, J.A., G. Dacremont, D. Carton, E. Orye, and C. Hooft, Science 181 (1973) 352-354. Orii, T., R. Minami, K. Sukegawa, S. Sato, S. Tsugawa, K. Horino, R. Miura, and T. Nakao, Tohoku J. Exp. Med. 107 (1972) 303-315. Okada, S., T. Yutaka, T. Kato, C. Ikehara, H. Yabuuchi, M. Okawa, M. lnui, and H. Chiyo, Eur. J. Pediatr. 130 (1979) 239-249. Leroy, J.G., M.W. Ho, M.C. MacBrinn, K. Zielke, J. Jacob, and J.S. O'Brien, Pediatr. Res. §_ (1972) 752-757. Suzuki, Y., K. Fukuoka, J.J. Wey, and S. Handa, Clin. Chim. Acta 2 (1977) 91-970 Wenger, D.A., T.J. Tarby, and C. Wharton, Biochem. Biophys. Res. Commun. 8_2_ (1978) 589-595. Hultberg, B., A. Isaksson, S. Sjb blad, and P.A. Ockerman, Clin. Chim. Acta £12 (1980) 33—38. Meade, C.J. and J. Mertin, Adv. Lipid Res. 26_ (1978) 127-165. Pelus, L.M. and H.R. Strausser, Life Sci. 22 (1977) 903-913. Goodwin, J.S., R.P. Messner, A.D. Bankhurst, G.T. Peake, J.H. Saiki, and R.C. Williams, Jr., N. Engl. J. Med. 297 (1977) 963-968. Waddell, C.C., O.D. Taunton, and J.J. Twomey, J. Clin. Invest. fl (1976) 950-954. Chisari, F.V. and T.S. Edgington, J. Exp. Med. 142 (1975) 1092-1107. Curtiss, L.K. and T.S. Edgington, J. Immunol. 116 (1976) 1452-1458. Curtiss, L.K. and T.S. Edgington, J. Clin. Invest. fl (1978) 1298-1308. Curtiss, L.K., D.H. DeHeer, and T.S. Edgington, J. Immunol. 118 (1977) 648- 652. Curtiss, L.K. and T.S. Edgington, J. Clin. Invest. 6_3 (1979) 193-201. Bieber, M.M., D.P. King, S. Strober, and H.S. Kaplan, Clin. Res. 2_7 (1979) 81A. 194. 195. 196. 197. 198. 199. 200. 201. 202. 203. 204. 205. 206. 207. 208. 209. 210. 211. 212. 213. 214. 215. 153 Chapman, Jr., H.A., and J.B. Hibbs, Jr., Science 1_91 (1977) 282-285. Schultz, R.M., N.A. Pavlidis, W.A. Stylos, and MA. Chirigos, Science 202 (1978) 320-321. Raz, A., R. Goldman, l. Yuli, and M. Inbar, Cancer Immunol. Immunother. 2 (1978) 53-59. Freimuth, W.W., H.C. Miller, and W.J. Esselman, J. Immunol. 123 (1979) 201-208. Marcus, D.M. and G.A. Schwarting, Adv. Immunol. _22 (1976) 203—240. Esselman, W.J. and H.C. Miller, J. Exp. Med. 139 (1974) 445-450. Wang, T.J., W.W. Freimuth, H.C. Miller, and W.J. Esselman, J. Immunol. 121 (1978) 1361-1365. Liu, D.Y., K.D. Petschek, H.G. Remold, and J.R. David, J. Immunol. 124 (1980) 2042-2047. Higgins, T.J., A.P. Sabatino, H.G. Remold, and J.R. David, J. Immunol. 121 (1978) 880-886. Liu, D.Y., T.J. Higgins, K.D. Petschek, H.G. Remold, and J.R. David, Fed. Proc. 22 (1978) 1400. Poste, G., R. Kirsh, and I.J. Fidler, Cell. Immunol. 4_4_ (1979) 71-88. Miura, T., S. Handa, and T. Yamakawa, J. Biochem. (Tokyo) 22 (1979) 773- 776. Agarwal, M.K. and E. Neter, J. Immunol. 107 (1971) 1448-1456. Miller, H.C. and W.J. Esselman, J. Immunol. 115 (1975) 839-843. Esselman, W.J. and H.C. Miller, J. Immunol. 119 (1977) 1994-2000. Lengle, E.E., R. Krishnaraj, and R.C. Kemp, Cancer Res. 22 (1979) 817-822. Ryan, J.L. and M. Shinitzky, Eur. J. Immunol. 2 (1979) 171-175. Stewart, S.S., R.L. Whisler, and A.J. Yates, Abstr 13th Int Leukocyte Culture Conference (1979) 78. Suzuki, K., Meth. Enzymol. 22, (1972) 839-841. Snyder, R.A. and R.C. Brady, Clin. Chim. Acta 22, (1969) 331-338. Palese, P., D. Bucher, and E.D. Kilboume, lgplied Microbiol. 22, (1973) 195- 201. Thomas, G.H., R.E. Tipton, L.T. Chien, L.W. Reynolds, and C.S. Miller, Clin. Genetics g, (1978) 369-379. 216. 217. 218. 219. 220. 221. 222. 223. 224. 225. 226. 227. 228. 229. 230. 231. 232. 233. 234. 154 Lowry, O.H., N.J. Rosebrough, A.L. Farr, and R.J. Randall, J. Biol. Chem. 122.(1951)265-275. Esselman, W.J., R.A. Laine, and C.C. Sweeley, Methods Enzymol. 22 (1972) 140-156. Skipski, V.P. and M. Barclay, Meth. Enzymol. fl (1969) 530-598. Abell, L.L., B.B. Levy, B.B. Brodie, and F.E. Kendall, J. Biol. Chem. 195 (1952) 357-366. Svennerholm, L., Biochim. Biophys. Acta 22 (1957) 604-611. Spiro, R.G., Meth. Enzymol. 2 (1966) 3-25. Ueno, K., S. Ando, and R.K. Yu, J. Lipid Res. 22 (1978) 863-871. Laine, R.A., W.J. Esselman, and C.C. Sweeley, Meth. Enzymol. 22 (1972) 159-1670 Ledeen, R. and R.K. Yu, in "Biological Diagnosis of Brain Disorders", S. Bogoch, edit., Spectrum Publications, Inc., Flushing, N.Y., (1973) pp. 377- 379. Humbel, R. and C. Collart, Clin. Chim. Acta 6_2 (1975) 143-145. Muchmore, AN. and RM. Blaese, J. Immunol. 118 (1977) 881-886. Bb yum, A., Scand. J. Clin. Lab. Invest. (Suppl. 97) a (1968) 51-76. Wenger, D.A., M. Sattler, C. Clark, and C. Wharton, Life Sci. 22 (1976) 413- 420. Kwiterovich, Jr., P.O., H.R. Sloan, and D.S. Fredrickson, J. Lipid Res. 22 (1970)322-330. Billing, B.H., H.J. Conlon, D.E. Hein, and L. Schiff, J. Clin. Invest. 22 (1953) 214-225. Ralli, E.P., S.H. Rubin, and S. Rinzler, J. Clin. Invest. 22 (1941) 93-97. Wolfe, L.S., R.G. Senior, and N.M.K. Ng Ying Kin, J. Biol. Chem. 249 (1974) 1828-1838. Ng Ying Kin, N.M.K. and L.S. Wolfe, Biochem. Biophys. Res. Commun. 22 (1974) 837-844. Strecker, G. and J. Montreuil, Clin. Chim. Acta 22 (1971) 395-401. 235. 236. 237. 238. 239. 240. 241. 242. 243. 244. 245. 246. 247. 248. 249. 250. 251. 252. 253. 254. 155 Tsay, G.C., G. Dawson, and R. Matalon, J. Clin. Invest. 22 (1975) 711-718. Tsay, G.C., G. Dawson, and S.-S.J. Sung, J. Biol. Chem. 251 (1976) 5852- 5859. Strecker, G., M.-C. Peers, J.-C. Michalski, T. Hondi-Assah, B. Fournet, G. Spik, J. Montreuil, J.-P. Farriaux, P. Maroteaux, and P. Durand, Eur. J. Biochem. Z2 (1977) 391-403. Michalski, J.-C., G. Strecker, B. Fournet, M. Cantz, and J. Spranger, FEBS Letters 22 (1977) 101-104. Den Tandt, W.R. and M.A.H. Giesberts, Biochem. Med. _7_ (1973) 441-451. Van Hoof, F. and H.G. Hers, Eur. J. Biochem. _7_ (1968) 34-44. Dean, K.J. and C.C. Sweeley, J. Biol. Chem. 254 (1979) 10001-10005. Tanaka, H. and K. Suzuki, J. Biol. Chem. 250 (1975) 2324-2332. Norden, A.G.W. and J.S. O'Brien, Arch. Biochem. Biophys. 159 (1973) 383- 392. Ho, M.W. and J.S. O'Brien, Science 165 (1969) 611-613. Ockerman, P.A. and P. Kb h11n, Acta Paediatr. Scand. 21 (1968) 281-284. Spranger, J.W., F. Koch, V.A. McKusick, J. Natzschka, H.-R. Wiedemann, and H. Zellweger, Helv. Paediatr. Acta 22 (1970) 337-362. Bach, G., M. Cantz, C. Hall, and E. Neufeld, Biochem. Soc. Trans. 2 (1973) 231-234. O'Brien, J .S., Clin. Genetics fl (1978) 55-60. Potier, M., G. Beauregard, M. Belisle, L. Mameli, N. Hong, S.B. Melancon, and L. Dallaire, Clin. Chim. Acta 22 (1979) 97-105. Li, S.-C., T. Nakamura, A. Ogamo, and Y.-T. Li, J. Biol. Chem. 254 (1979) 10592-10595. Hechtman, P. and D. Le Blanc, Biochem. J. 167 (1977) 693-701. Fishman, P.H., S.R. Max, J.F. Tallman, R.O. Brady, N.K. Maclaren, and M. Cornblath, Science 187 (1975) 68-70. Max, S.R., N.K. Maclaren, R.C. Brady, R.M. Bradley, M.B. Rennels, J. Tanaka, J.H. Garcia, and M. Cornblath, N. Engl. J. Med. 291 (1974) 929-931. Sherlock, 5., Am. J. Med. 22 (1970) 693-706. 255. 256. 257. 258. 259. 260. 261. 262. 263. 264. 265. 266. 267. 268. 269. 270. 271. 272. 273. 274. 275. 276 O 156 Paronetto, F., Postgrad. Med. 22 (1973) 156-160. Hsu, C.C.S. and C.M. Leevy, Clin. Exp. Immunol. 2 (1971) 749-760. Wands, J.R. and J.L. Dienstag, Yale J. Biol. Med. 22 (1978) 615-623. Nakao, M., Y. Mizoguchi, T. Monna, S. Yamamoto, and S. Morisawa, Acta Hepato-gastroent. 2_5 (1978) 335-343. Chisari, F.V., J. Immunol. 121 (1978) 1279-1286. Schumacher, K., G. Maerker-Alzer, and U. Wehmer, Nature 222 (1974) 655- 656. Kh rkkhinen, J. and R. Vihko, Carbohyd. Res. fl (1969) 113-120. DeJongh, D.C., T. Radford, J.D. Hribar, S. Hanessian, M. Bieber, G. Dawson, and C.C. Sweeley, J. Am. Chem. Soc. 22 (1969) 1728-1740. Wong, C.G., S.-S.J. Sung, and C.C. Sweeley, Meth. Carbohyd. Chem. 2 (1980) 55-65. Mononen, I. and J. K8 rkk'hinen, FEBS Letters 22 (1975) 190-193. Hayashi, A. and I. Yano, in "Chemistry of Lipids", T. Yamakawa and S. Nojima, editors, Kagaku-dojiu, Tokyo (1974) pp. 128-170. Wiegandt, H., Hoppe-Seyler's Z. Pthiol. Chem. 354 (1973) 1049-1056. Dawson, G., A.W. Kruski, and A.M. Scanu, J. Lipid Res. _1_7_ (1976) 125-131. Van den Bergh, F.A.J.T.M. and J.M. Tager, Biochim. Biophys. Acta 441 (1976) 391-402. Clarke, J.J.R., J.M. Stolz, and MR. Mulcahey, Biochim. Biophys. Acta 431 (1976) 317-325. Nelken, D., H. Ovadia, and N. Hanna, Eur. J. Immunol. 2 (1979) 176-177. Sohnle, P.G. and C. Collins-Lech, Cell. Immunol. fl (1979) 428-436. Hanna, N., H. Ovadia, and D. Nelken, Immunol. 23 (1978) 1007-1015. Miller, F. and G.S. Habicht, Int. Arch. Allergy appl. Immunol. 22 (1977) 228- 238. Mannick, J.A. and K. Schmid, Transplant. 2 (1967) 1231-1238. Mowbray, J., Immunol. 2 (1963) 217-225. Phillips, M.S., C.B. Carpenter, and P. Lane, Ann. N.Y. Acad. Sci. 249 (1975) 236-247. 157 277. Glasgow, A.H., S.R. Cooperband, J.C. Occhino, K. Schmid, and J.A. Mannick, Fed. Proc. 22 (1971) 652. 278. Cooperband, S.R., R. Nimberg, K. Schmid, J.A. Mannick, Transplant. Proc. 2 (1976) 225-242. 279. Skibinski, G., Z. Wieczorek, and J. Zwolinski, Arch. Immunol. Therap. Exp. 26_ (1978) 351-356.