THE PURIFICATION AND PARTIAL CHARACTERIZATION OF a-GALACTOSIDASES A AND B FROM HUMAN LIVER By Kenneth James Dean A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Biochemistry 1978 a-Galactosida respectively, frCZ eluded ammonium su Se;‘1adex G—lSO ch: 1er displacement a-Galactcsida dieliifen'l-a-g-g .. .6. The noni. 6,22th for the hy: ice-ever, the anion: 3513 of this subst: $1.4 1 7: ' at eiysis of 28133 izthe presence of Tzese findings SUSE between I 8 sodium tau} csmse A also ca: nactopyranoside a acetylgalactosazini l“.;« a tieopyranoside 1~1~ga1 _ .actOpyranosi ucose. These flecificity for t 2:31. 5 he ABSTRACT THE PURIFICATION AND PARTIAL CHARACTERIZATION OF a-GALACTOSIDASES A AND B FROM HUMAN LIVER BY Kenneth James Dean a-Galactosidases A and B were purified 67,000-fold and 68,000-fold, respectively, from normal human liver. The purification procedure in- cluded ammonium sulfate precipitation, DEAR-cellulose chromatography, Sephadex 6-150 chromatography, hydroxylapatite chromatography and ampho- lyte displacement chromatography on DEAR-cellulose. o-Galactosidase A (EC 3.2.1.22) catalyzed the hydrolysis of 4dmethyl- umbelliferyl-cqgrgalactopyranoside, galabiaose and globotriaose at pH 4.5-4.6. The nonionic detergent Triton Xr100 had no effect on the pH optimum for the hydrolysis of 4-methylumbelliferyl-o—Qrgalactopyranoside; however, the anionic detergent sodium.taurocholate inhibited the hydrol- ysis of this substrate below pH 4.8. This enzyme also catalyzed the hydrolysis of galabiosylceramide and globotriglycosylceramide at pH 4.1 1n.the presence of 5.7 mM and 9.3 mM sodium taurocholate, respectively. These findings suggest that there is a strong electrostatic interaction between sodium taurocholate and o-galactosidase.A below pH 4.8. a-Galac- tosidase A also catalyzed the hydrolysis of p-nitrophenyl-Z-deoxY’a‘Dr galactopyranoside at pH 4.6, but did not hydrolyze o-nitrophenyl-ofigy acetylgalactosaminide, p-nitraphenyl-ajprglucopyranoside or o-nitrophenyl- Ofigrfucopyranoside. Furthermore, the hydrolysis of 4~methylumbelliferyl- “‘2'881actopyranoside was not inhibited by 20 mM g-galactal or 20 mM 2-- (+)-fucose. These findings suggest that o-galactoaidase A lacks absolute 8Pecificity for the hydroxyl group at C-2; however, replacement of this BrouP‘with an acetamido group appears to prevent binding with the enzyme, PCssibly due to steric hindrance. a-Galactoaidase A appears to have absolute specificity for the hydroxyl groups at C-4 and C-6. a-Galactosidase B catalyzed the hydrolysis of 4~methylumbellifery1- “fiErBalactopyranoside, o-nitrOphenyl-ofigracetylgalactosaminide, globo- triBlycosylceramide, globopentaglycosylceramide, globotriaose and glabepentaose. T1 teminal a-fi-acet greater than its tcse residues, ba” Eartha-more, the : z-fi-acetylgalactos tmtai'iing ternina that this enzyme f tether than an 3-5, thehydralysis of iafi'O-Q-galacwp or 4-nethunhellif e results is that :- sclute specificiqy. astute specificitv Kenneth James Dean globopentaose. The affinity of the enzyme toward substrates containing terminal ofigracetylgalactosamine residues was approximately 5 times greater than its affinity toward substrates containing terminal o-galac- tose residues, based on the'Michaelis constants with these substrates. Fhrthermore, the maximal velocities with substrates containing terminal aggfacetylgalactosamine were 3 times greater than those with substrates containing terminal a-galactose residues. Therefore, it appears likely that this enzyme functions as an afgracetylgalactosaminidase (EC 3.2.1.49) rather than an o—galactosidase in vivo. o—Galactosidase B also catalyzed the hydrolysis of o-nitrOphenyl-ofigrfucOpyranoside and p-nitrophenyl-Z- deoxy-ojgrgalactopyranoside, but not p-nitrophenyl-ofigrglucopyranoside or 4amethumbelliferyl-afyfacetylglucosaminide. One explanation for these results is that o—galactosidase B (afiflyacetylgalactosaminidase) lacks ab- solute specificity for the substituents at C-2 and C-6, but may have ab- solute specificity for the axial hydroxyl group at C-4. I0 my dear To my dear wife, Becky, who supported and encouraged me throughout my graduate studies. ii Iwish to exp advisor and friend ievelopnent. Ive Barker, Dr. Claret. diszussions, and t St'eeley's and Dr. . Etiutageuent . ACKNOWLEDCMENTS I wish to express my gratitude to Dr. Charles C. Sweeley, my advisor and friend, for his guidance and concern for my professional development. I would also like to express my appreciation to Dr. Robert Barker, Dr. Clarence H. Suelter and Dr. John F. Holland for many helpful discussions, and to the graduate students and postdoctorates in Dr. Sweeley's and Dr. Barker's research groups for their friendship and encouragement . iii LIST OF I I”I. LIST OF FIGL‘ ASBRE‘JIATION NERDDL'CIION PETER OF TE Current The Che:- Cont. Pu: St] The The Che: C011 Str TABLE OF CONTENTS LIST OF TABLES .................................................. LIST OF FIGURES ................................................. ABBREVIATIONS ................................................... INTRODUCTION .................................................... REVIEW OF THE LITERATURE ........................................ Current Concepts of Glycoconjugate Metabolism .............. The Chemistry and Metabolism of Glycoconjugates Containing Terminal o-Galactose Residues ................ Purification and Characterization of Glycoconjugates OOOOOOOOOOOOOOOOCOOOOOOOOOOOOOOOOOOO Structure and Function of Glycoconjugates Containing Terminal a-Galactose Residues ........... Galabiosylceramide ............................... Globotriglycosylceramide ......................... Ivh-o-Galactosyllactoneotetraglycosylceramide .... Blood Group Bchtive Glycoconjugates ............. The Biosynthesis of Glycoconjugates Containing Terminal o-Galactose Residues ...................... The Catabolism of Glycoconjugates Containing Terminal o-Galactose Residues ...................... The Chemistry and Metabolism of Glycoconjugates Containing Terminal afiNfAcetylgalactosamine Residues .. Structure and Function of Glycoconjugates Containing Terminal aggrAcetylgalactosamine REBidues OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO0.00.0.0... GlobOpentaglycosylceramide ......................... Blood Group AnActive Glycoconjugates ............... The Biosynthesis of Glycoconjugates Containing Terminal ofEfAcetylgalactosamine Residues .......... The Catabolism of Glycoconjugates Containing Terminal ofyrAcetylgalactosamine Residues .......... MATERIALS AND METHODS mterials 0.0...OOOOOOOOOOOOOO0.00.0000...OOOOOOOOOOOOOOOOOOOO. methada 000......OOOOOOOOOOOOOOO...0.00.00.00.00...OOOOOOOOOOOO Purification of a—Galactosidases A.and B from Hm Liver OOOOOOOOOOOOOOOOOOOOO00.00.0000....0.00.00.... iv Page vii viii xii 16 16 17 17 17 20 20 22 24 26 27 27 27 30 31 33 36 36 Ampho Preparatio Preps Assays $88.33, ExtraCtion .00....OOOOOOCOOOOOOOIOOOOOOOOOOOOOOOOOOOOOC Ammonium Sulfate Precipitation ........................ DEAE-Cellulose Chromatography ......................... Sephadex G-150 Chromatography ......................... Hydroxylapatite Chromatography ........................ Ampholyte Displacement Chromatography ................. Preparation of Substrates .................................. Preparation of Glycolipid Substrates .................. Globotriglycosylceramide ......................... [3H]Globotriglycosylceramide ..................... Galabiosylceramide ............................... GlobOpentaglycosylceramide ....................... Oligosaccharide Substrates ............................ Artificial Substrate Analogs .......................... p-Nitropheny1-2-Deoxy-o-2rGalactopyranoside ...... o-Nitropheny1-6-Deoxy-o-2rGa1actopyranoside (o-Nitrophenyl-o-QrFucopyranoside) ............... Assays ..................................................... Assays with Glycosphingolipid Substrates .............. Assays with Globotriglycosylceramide ............. Assays with Galabiosylceramide ................... Assays with Globopentaglycosylceramide ........... Assays with Oligosaccharide Substrates ................ Assays with Galabiaose ........................... Assays with Globotriaose ......................... Assays with Globopentaose ........................ Assays with Artificial Substrates ..................... Assays with 4-Methylumbelliferyl-o-Qr GalactOPYrmOSide 0......OOOIOOOOOOOOOOOOOOOOOO Assays with 0- or p-Nitrophenyl-ojgf Acetylgalactosaminide ......................... Assays with Other Artificial Substrates .......... Protein Assays.cocoa...coco-o00000000000010.0000... Physical Characterization of a-Galactosidases A and B ...... RESULTS OOOOOOOOOOOOOOOOOOOOOfOOOOOOOOOOOOOOCOOOOOOOOOOOOOOOO0.0. Purification of a—Galactosidases A and B ................... Page 36 36 36 36 37 37 37 37 37 39 41 41 42 42 43 44 45 45 4s 47 47 48 ' 48 48 49 49 50 50 50 51 51 52 52 Enzymatic Pr q-Galac Ar 01: Cl: a-Galac: Art Oi: Gly Characterize: HSGESION ....... Characterize: c-Galactos u-Galact a-Galac: The Cart. u-Gale Characteriz EELKBRAPHY SD“ 7 r. x ‘Ua & Enzymatic Properties of o-Galactosidases A and B ............ o—Galactosidase A ...................................... Artificial Substrates ............................. Oligosaccharide Substrates ........................ Glycosphingolipid Substrates ...................... o-Galactosidase B ...................................... Artificial Substrates ............................. Oligosaccharide Substrates ........................ Glycolipid Substrates ............................. Characterization of a-Galactosidases A and B ................ DISCUSSION 0..0......00......OOOOCOOOOOOOOOOOOO...0.00.00.00.00... Characterization of the Enzymatic PrOperties of a-GalaCtoaidasesAandB O00.0.0.0...0.000.000.0000..00... a-GalaCtosidase A ooooooooooooooooococoon-000.000.000.00 G‘salaCtOSidase B 0.0000000000000000...0.000000000000000 The Carbohydrate-Binding Specificities of a-GalaCtOSidases A and B OOOOOOOOOOOOOOOOOOOOOOOOOOOO Characterization of o-Galactosidases A and B ............. BIBLIOBWHY 00......0.0.0.0...OOOOOOOOOOOOOCOOOIOOOICOOOOOOOOOIOO APPENDIX OCOOOOOOOOOOOOO0.00......I.0.00000000000000000000COOOOOOC vi Page 66 66 66 83 83 110 110 135 160 160 182 183 183 186 187 190 196 217 hue L Purifica: from t L Summary c a-Gale Table 1. LIST OF TABLES Page Purification of o-Galactosidases A and B from Human Liver 0.0.0...OOOOOOOOOOOOOOOOOOOOOOO00..O. 53 Summary of Enzymatic Properties of a-GalaCtoaidaaes A and B OOOOOOOOOOOOOOOOOOOOOOO00.00. 177 vii Exazpl Iypic% The E; The B: L: An Exafi Struc: Te: Struc: TE? DEAE-C Gal SEphad- Gale} HydrOXj Gale Ampholj Gala Amphol} CalC NatiVQ Of a The Lin Gala EnZZF The pH Gala The Line by a The Hi1 M81 The Effe Hydré The Effe § Hit- Figure l. 9. 10. 11. 12. 13. 14. ' 15. 16. 17. 18. 19. LIST OF FIGURES Examples of Oligosaccharide Structures of Typical Glycoconjugates so...00000000000000.0000... Typical Pathway of Glycosphingolipid Biosynthesis .... The Biosynthesis of Glchpyrophosphoryldolichol ...... The Biosynthesis of a Glyprotein with Asparagine-Linked Carbohydrate .................... An Example of Glycoconjugate Catabolism .............. Structures of Glycoconjugates Containing Terminal a-Galactose Residues ..................... Structures of Glycoconjugates Containing Terminal ofgyAcetylgalactosamine Residues ......... DEAE-Cellulose Chromatography of o- Galactosidases A and B ............................ Sephadex G-150 Chromatography of o- Galactosidases A and B ............................ Hydroxylapatite Chromatography of o- GalaCtOSidases A and B oooaoooooooooaoooooooooooooo Ampholyte Displacement Chromatography of o- GalaCtOSidase A on DEAE-Ce11u108e 0.000000000000000 Ampholyte Displacement Chromatography of o- Galactosidase B ................................... Native Polyacrylamide Disc Gel Electrophoresis of o-Galactosidases A and B ....................... The Linearity of 44MU-o—Ga1 Hydrolysis by a- Galactosidase A with Respect to Time and” Enzyme concentration .0.0COOOOOOOOOOOOO0.0.0.000... The pH Optimum of 4AMU—o-Gal Hydrolysis with o- GalaCt°81dase A OOOCOOOOIOOOOOO...OOOOOOOCCOOOOOOOO The Lineweaver-Burk Plot of 4—MU-a-Gal Hydrolysis by a-Galactosidase A .............................. The Hi11.Plot of 4-MU-o—Gal Hydrolysis by a-GaIaCtOSidase A OOOOOOOOOCOOCOOOOOOOOOOOOOOOOOOO. The Effect of pH of the Km and Vmax of 4-MU-o-Ga1 Hydrolysis by o—Galactosidase A.................... The Effect of pH on Lineweaver-Burk Plots of 44MU-a-Gal Hydrolysis by o-Galactosidase A ........ viii Page 12 14 19 29 55 57 59 61 63 65 68 7O 72 74 76 78 Figure 20. Fouri Lines H} The I b) Thin- Thifl' The F Linen Ca The H Ga Line: as The H. th: LinEwE M The Hi Cal Linewe 3-0 The Hi; Gale The Lin Gal. Emly- Figure Page 20. The Effect of pH on Lineweaver-Burk Plots of 44MU-o-Ga1 Hydrolysis in the Presence of sadim TaurOChOIate 0.0.0000.........OOOOOOOOOOOO. 80 21. Fourier-Transform 13C-NMR Spectra of Artificial Substrate Analogs ................................ 82 22. Lineweaver-Burk Plot of p-NP-Z-deoxy-a-Gal Hydrolysis by a-Galactosidase A .................. 85 23. The Hill Plot of p-NP-Z-deoxy-a—Gal Hydrolysis by ““6818Ct081d883 A so...oooooooooooooooooeoooooo 87 24. Thin—Layer Chromatogram of [3H]Ga1abiaose ........... 89 25. Thin-Layer Chromatogram of [3H]Globotriaose ......... 91 26. The pH Optimum for the Hydrolysis of Galabiaose and Globotriaose by o-Galactosidase A ............ 93 27. Lineweaver—Burk Plot of GaOsez Hydrolysis by a- GalaCtosidaseAO00.0.0.........OOOOOOOOOOOOOOOOOO 95 28. The Hill Plot of GaOsez Hydrolysis by a- GalaCt081daseA00.0.00.........OOOOOOOOOOOOOOO... 97 29. Lineweaver-Burk Plot of GbOse3 Hydrolysis by s-Galactosidase A ................................ 99 30. The Hill Plot of the Hydrolysis of GbOse3 by s-Galactosidase A ................................ 101 31. Thin-Layer Chromatogram of [3H]Ga1abiosy1ceramide ... 103 32. Thin-Layer Chromatogram of [3H]G10botrig1ycosyl- cermj-de OOOOOOOOOOOOOOOOO......OOOOOOOOOOOOOOOOO. 105 33. The Linearity of GbOse3Cer Hydrolysis by o-Galactosidase A.with Respect to Time and Enzyme Concentration ............................. 107 34. The Optimum Taurocholate Concentration for the Hydrolysis of Galabiosylceramide and Globo— triglycosylcermide ......OOOCOCOC00.000.000.00... 109 35. The pH Optimum for the Hydrolysis of GaOseZCer and Gb08e3cer .....COOOOOOO0.0000000000000000000...... 112 36. Lineweaver-Burk Plot of GaOse3Cer Hydrolysis by a-GalaCtosidaseA......OOOOOOOOOOOOO0.0.0.0000... 114 37. The Hill Plot of GaOseZCer Hydrolysis by s- GalaCt081daseA00.0.00.........OOOOOOOOOOCOOO...O 11-6 38. Lineweaver-Burk Plot of GbOse3Cer Hydrolysis by a-GalaCt081dase A so...ooooooooooooooooooooooooso. 118 39. The Hill Plot of GbOse3Cer Hydrolysis by a- GalaCt081daeeA ...0.0.0.0...00.000.000.000....... 120 40. The Linearity of 4-MU-a-Ga1 Hydrolysis by a- Galactosidase B with Respect to Time and Enzme concentration ......OOOOOOOOOOOOOOOO......O 122 ix 52, 53. 54. 35. 36, 57, as, 59 so. 61. The Figure 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. The pH Optima for the Hydrolysis of 44MU-a-Gal and o-NP-o-GalNAc by a-Galac- tosidaae B 0.000.000.0000.......OOOOOOOCOOOOOO.... Lineweaver-Burk Plot of 4-MU-o-Ga1 Hydrolysis by o-Galactosidase B ................................ The Hill Plot of 4-MU-o-Gal Hydrolysis by a- Galactosidase B ............IOO......COOOOOOOOO... Thermal Inactivation Curves for o-Galactosidase A and B, and ofingcetylgalactosaminidase ActiVitj-es ......OIOOIOO...0.0.000...0.00.00.00.00 Isoelectric Focusing of a Partially Purified Preparation of o—Galactosidase B ................. The Competitive Inhibition of 4-MU-o-Ga1 Hydrol— ysis by o-NP-o-GalNAc with o-Galactosidase B ..... Lineweaver-Burk Plot of o-NP-a-GalNAc Hydrolysis by a-GalaCtOSidase B OOOOIIOOOOOOOOOOOOO0.0.0.0... The Hill Plot of o-NP-o-GalNAc Hydrolysis by a-GalaCt081dase B ....OOOOOOOOOOOOOOOOOOOO......O. Lineweaver-Burk Plot of the Hydrolysis of p-NP-2-Deoxy—o-Ga1 by o-Galactosidase B .......... The Hill Plot of p-NP-Z-Deoxy—a-Gal Hydrolysis by s-Galactosidase B ................................ Lineweaver-Burk Plot for the Hydrolysis of O‘NP-afiQrFuc by a-GalaCtOBidase B 0000000000000... The Hill Plot of o-NP-ofinguc Hydrolysis by a-GalaCtOSidase B ......OOOOOOOOOOOOIO0.0.00.00... Thin-Layer Chromatogram of [3H]Globopentaose ........ The pH Optima for the Hydrolysis of Globotriaose Globopentaose by o—Galactosidase B ............... Lineweaver-Burk Plot for the Hydrolysis of GbOBeS by a-GalaCtosidase B oooooooooooooooooooooo The Hill Plot of GbOse5 Hydrolysis by o-Galactosidase B ................................ Lineweaver-Burk Plot of GbOse3 Hydrolysis by a-GalaCtosi-dase B 00...........OOOOOOOOOOOOOOOOOOO The Hill Plot of GbOse3 Hydrolysis by o-Galactosidase B ................................ Thin-Layer Chromatogram.of [3H]Globopentag1ycosy1ceramide ................... Thin-Layer Chromatogram.of GbOse5Cer Hydrolysis by a-Galactosidase B .................. The pH Optima for the Hydrolysis of GbOse3Cer and GbOseSCer by o—Galactosidase B ............... X Page 124 126 128 130 132 134 137 139 141 143 145 147 149 151 153 155 157 159 162 164 166 65. 66. 6k 6& 6% 7L The Lin Est Isoe Thre Thre Thre Figure 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. The Sodium Taurocholate Optima for the Hydrolysis of GbOse3Cer and GbOse5Cer by a-GalaCtosidaseB ......OOOOCOOOOCOIOOOO.......... Lineweaver-Burk Plot of the Hydrolysis of GbOse5Cer by a-Galactosidase B ................... The Hill Plot of GbOseSCer Hydrolysis by c—Galactosidase B ................................ Lineweaver-Burk Plot of GbOse3Cer Hydrolysis by G‘GEIECtosidase B ooooooooooooooooooooooooooooo The Hill Plot of GbOse3Cer Hydrolysis by a-‘GalaCtOBidaseB ......OOOOOCOOOOOCOOOOO000...... Estimation of the Molecular weights of a-Galactosidases A and B by Gel Filtration ....... Isoelectric Focusing of Purified . a.GalaCt081dase A 00.000.00.000.........OOOOOOOOOO Three-Dimensional Projections of Carbohydrate Residues Hydroyzed by the BfiErAcetylhexos- minidases ......OOOOOOOOO.........OOOO0.0.0.0.... Three-Dimensional Projection of the Carbohydrate Residue Hydrolyzed by c-Galactosidase A .......... Three-Dimensional Projection of the Carbohydrate Residues Hydrolyzed by a-Galactosidase B ......... xi Page 168 170 172 174 176 179 181 189 192 194 Uridine d1; 3.;l-galactopyraf DP-GlcNAc; Uri: csine diphOSIr‘I'lat wnopyranoside , DP-NeuAc; glucc glycosylceramide peztaglycosylcer; nectetraglycosylc Galabiosylceramic M's-Gal; o-nit phenyl-Z—deoxy-a- Q-fuccpyrano side, P‘TPfl-Glc; 4-met d1=etizvmzzt>ellifa :8.lifery1-B-R.ga “nyanoside, 4- 6%5-u-k-Ara; gala ABBREVIATIONS Uridine diphosphate-o-Qrglucopyranoside, UDP-Glc; uridine diphosphate- c-Qrgalactopyranoside, UDP-Gal; uridine diphosphate-afigracetylglucosaminide, UDP-GlcNAc; uridine.diphosphate—cfigyacetylgalactosaminide, UDP-GalNAc; guan- osine diphosphate-B—erucopyranoside, GDP-Fuc; guanosine diphosphate-a-Qr' mannopyranoside, GDPéMan; cytosine monophosphate-afiyfacetylneuraminic acid, CMP-NeuAc; glucosylceramide, GlcCer; lactosylceramide, LacCer; globotri- glycosylceramide, GbOse3Cer; globotetraglycosylceramide, GbOse4Cer; globo- pentaglycosylceramide (Forssman antigen), GbOseSCer; IV”-c-galactosyllacto- neotetraglycosylceramide, IV-c-Gal-LcnOseACer; galactosylceramide, GalCer; Galabiosylceramide, GaOseZCer; 4~methylumbelliferyl-c-Qrgalactopyranoside, 4AMU-a-Gal; o-nitrOphenyl-cfgfacetylgalactosaminide, o-NP-a-GalNAc; p-nitro- phenyl-Z-deoxy-ofigrgalactopyranoside, p-NP-Z-deoxy-c-Gal; o-nitrophenyl-c- prfucOpyranoside, o-NP-c-eruc; p-nitrophenyl-c-Qrglucopyranoside, p-NP-a-Glc; 4-methylumbelliferyl-ngfacetylgalactosaminide, 4-MU-B—GalNAc; 4-methy1umbelliferyl-afyfacetylglucosaminide, 46MU-c-GlcNAc; 4dmethylumr belliferyl-B-Qrgalactopyranoside, 46MU-B-Gal; 4-methylumbelliferyl-B-Q; xylopyranoside, 4AMU-B-Xyl; 4-methy1umbelliferyl-afigrarabinopyranoside, 4AMU-c-LrAra; galactose, Gal; lactose, Lac; raffinose, Raf. xii absence which re pdysacc reviews diagno s f. INTRODUCTION The enzymes involved in the metabolism of glycoconjugates have been an area of much interest due to their involvement in human glycolipid, glycoprotein and mucopolysaccharide storage diseases. These hereditary disorders of complex carbohydrate metabolism are characterized by the absence or deficiency of specific exoglycosidases or glycosyltransferases, which results in the accumulation of the glycolipid, glycoprotein or muco- polysaccharide substrate(s) of the missing enzyme(s). Several excellent reviews on the pathology and biochemistry of these disorders (1-3) and diagnostic procedures (4) have been published. Research on these enzymes has intensified in recent years with the realization that enzyme replace- ment therapy may be efficacious in treating some of these disorders. Many of the enzymes have now been purified and their physical and kinetic prOperties partially characterized. This thesis describes the purification and partial characterization of a—galactosidases from normal human liver. One of these enzymes, c-galactosidase A (EC 3.2.1.22), is absent or deficient in Fabry's disease, an X-linked glycosphingolipidosis in which glycolipids containing terminal c-galactose residues (galabiosylceramide, globotriglycosylceramide and blood group B-active glycolipids) accumulate. The other c-galactosidase present in human liver (form B) is not absent in Fabry's disease, but in the course of these investigations was also found to be an ufigfacetylgalac- toéaminidase. .This enzyme.is.probab1yxmore active as an cfigyacetylgalac- tosaminidase (EC 3.2.1.49) rather than an c-galactosidase in ziyg, hydro- lyzing globopentaglycosylceramide (Forssman antigen), and blood group Ar active glycolipids and glycoproteins, all of which contain terminal afigf acetylgalactosamine residues. A complete review of the chemistry and metabolism of glycoconjugates is certainly beyond the scope of this thesis. A general discussion of current concepts of the biosynthesis and catabolism of glycoconjugates will be presented, followed by a discussion of the chemistry and metab- olism of glycoconjugates containing terminal o-galactose and cfigfacetyl- galactosamine residues found in man. ‘1 '1'. consist the co:- uents. classes the gre. discour. these 51 have the nature a bc‘hydrat tare .35: and 511d faunes subsutu 553115.83 hiifiroxyl: lharides The by here 1 :9“ EEtaE is ref-Er: verse St: detail th REVIEW OF THE LITERATURE The term "glycoconjugates" encompasses a large class of compounds, consisting of glycolipids, glycoproteins and proteoglycans, which have the common feature of bearing covalently attached carbohydrate substit- uents. Examples of Oligosaccharide structures typical of each of these classes of glycoconjugates are shown in Figure 1. At first approach, the great diversity of Oligosaccharide structures in glycoconjugates may discourage the reader from further study; however, close examination of these structures reveals that they are not entirely unique. Many of them have the same Oligosaccharide backbone or core, and differ only in the nature and position of substituents on the core structure. Complex car- bohydrates that contain the same Oligosaccharide backbone or core struc- ture may be classified as members of the same family. Recently,'Swee1ey and Siddiqui (5) described the classification of glycolipids into seven families with different tetrasaccharide core structures. The carbohydrate substituents on glycoproteins and proteoglycans can be classified into families on the basis of their amino acid linkage (to serine, threonine, hydroxylysine or asparagine) and the structures of their core Oligosac- charides (6-9). The compounds within each family of complex carbohydrates are related by more than their common structural features, since they also have com? mon metabolic pathways and may also have similar functions. The reader is referred to several excellent reviews (5-15), which discuss the di- verse structures, functions and metabolism of glycoconjugates in greater detail than will be presented here. 1. Current Concepts of Glycoconjugate Metabolism The biosynthetic pathways for the carbohydrate portions of glyco- lipids and glycoproteins are thought to proceed by somewhat different mechanisms. The biosynthesis of glycolipids appears to occur primarily in the Golgi apparatus (11-16). Glycosphingolipid biosynthesis is ini- tiated by the transfer of galactose or glucose from UDP—Gal or UDP-Glc directly to the hydroxyl group at C-1 of ceramide. 6A.second,‘m£nor 2 Figure 1. Examples of Oligosaccharide Structures of Typical Glycoconju- gates The oligosaccharides of (I) a glycolipid [II3-afflfacetylneuramin- osyl-gangliotriglycosylceramide (GMZ ganglioside that accumulates in Tay- Sachs and Sandhoff's diseases) (3,5)]; (II) a circulatory glycoprotein [this glycoprotein may also bear a Fuc(al+6) substituent on the aspara- gine-linked gracetylglucosamine residue, or a GlcNAc(Bl+4) substituent on the mannose residue at the branching point (6)] and; (III) a mucopolysac- charide [chondroitin-4-sulfate that accumulates in Morquio's Syndrome. The Oligosaccharide shown is the core structure for this mucopolysac- charide, which may contain repeating ClcUA(Bl+3)GalNAc-4-SO4 units (3)]. Mucopolysaccharides, or proteoglycans, contain much more carbohydrate than the glycoproteins do, and are sometimes referred to as glycopep- tides a II HI . CH n. ”also/\ZMH \ | OH ‘O- H~C0Ct “3°C" c00H “5°" I 0 H0 ONION HOCHZOH Hoc o-CHz-cH-CH-CH:CH-(CH2).2 -CH3 0 0H NH- co-(cn-tz):2 24 -CH3 0H HHCOCH3 0H CH3c0HN ° coon a ._c..2. CH‘OH CH20H CHZOH CH20H o A!" 11 Ho .0 OH o-c 0 0 0H HNcocn-u3 0 0H HNCOCH3 000” CHZOH 0.20.,” HNCOCH3 “may Hucocn3 CHEOH HOCH2 °~ {be III OOH HOc c00H CHZOH HNCOCH ”035 0 OH H COCH3 H HNCOCH3 Figure 2. Typical Pathway of Glycosphingolipid Biosynthesis The biosynthesis of II3-afififacetylneuraminosyl-gangliotriglycosyl- ceramide (GMZ ganglioside) is shown. C A! heUAc( LOP rc\ w Games Sphi ngosi ne Sphi ngosi ne UDP-Glc Acyl CoA \ Glucosyltransferase 0 6 Ceramide Glc(pl-l') Sph UDP—Glc \ Glucosyltransferase Acyl CM 6 Glc(pl-1')Cer : UDP-Gal \ 6 oal(pl-4)clc(pl-1')Cer \ 6 NeuAc(a2-3) Gal(p1—4)Glc(pl—l') Cer Galactosyltransferase CMP-NeuAc Sialyltransferase d : u-Acetylgalactosaminyltransferase 6M3 Gangliosidosis _l 6 GalNAc(pl-4)GaI(P1-4)Glc(Pl-l')Cer Ior2-3 NeuAc 7 pathway may exist for the biosynthesis of cerebrosides, involving the transfer of galactose or glucose to sphingosine to form psychosine (galactosylsphingosine) or glucosylsphingosine, which is then acylated by acyl CoA to form the corresponding cerebroside (17-19). The carbo- hydrate chain of the glycolipid may then be extended by the sequential transfer of monosaccharides from the appropriate sugar nucleotide (UDP- Gal, UDP-GalNAc, GDP-Fuc or CMP-NeuAc) to the nonreducing end of the oligosaccharide moiety of the acceptor glycolipid (10,11). An example of glycolipid biosynthesis is shown in Figure 2. In contrast to the direct manner in which glycolipids are synthesized, with the sequential transfer of monosaccharides from sugar nucleotide donors directly to the nascent glycosphingolipid, an intermediate acceptor is involved in the biosynthesis of the core oligosaccharide of some glyco- proteins, as shown in Figure 3. The biosynthesis of glycoproteins with asparagine-linked carbohydrate has been described by Schacter g£_gl. (8, 9), and will be presented as a general model for glycoprotein biosynthesis. However, this is not intended to preclude the existence of other pathways for glycoprotein biosynthesis; the biosynthesis of glycoproteins with serine or threonine-linked carbohydrate is probably carried out by a dif- ferent pathway (22). The biosynthesis of the core oligosaccharide of asparagine-linked complex carbohydrates is thought to take place in the rough endoplasmic reticulum (8,9). The synthesis of the core oligosaccharide is initiated by the transfer of Efacetylglucosamine from UDP-GlcNAc to a polyiSOprenoid phospholipid acceptor, dolichol phosphate (7), to formE-acetylglucosd aminylpyrophosphoryldolichol. The carbohydrate chain is then extended by the sequential transfer of gracetylglucosamine and mannose from UDP-GlcNAc and GDP-Man to form Man(Bl+4)GlcNAc(81+4)G1cNAc—P-P-Dol. The carbohydrate chain is further extended, with branching, by the sequential transfer of mannose residues from mannosylphosphoryldolichol (generated by the trans- fer of mannose from GDP-Man to dolichol phosphate) to the nonreducing end of the nascent glyc0pyr0phosphoryldolichol (7,23). The transfer of mannose residues appears to be followed by the transfer of three glucose residues to the nascent core oligosaccharide (9,25,26). The entire oligosaccharide is then transferred from dolichol pyrophosphate to specific asparagine residues on the protein (24), a process which explicitly requires the ter- minal glucose residues for effective transfer (25,26). This is followed Figure 3. The Biosynthesis of Glycopyrophosphoryldolichol The biosynthesis of the pyrophosphoryldolichol intermediate involved in the biosynthesis of asparagine-linked complex carbohydrates is shown (7). Clc. Dolichol Phosphate UD P-GlcNAc GIcNAc-®-®-Dol UD P-GlcNAc GlcNAc(Fl-4)GlcNAc-®-®-Dol GD P-Man Man(Bl-4)GlcNAc(pl-4)GlcNAc-®-®-Dol Man-®-Dol\ Man-®-Dol\ Man 0(1-6 6 /Man(Pl-4)G|cNAc(I31-4)GlcNAc-®-®-Dol Man (XI-3 6 Man-®-Dol\ 3 UD P'GIC\ Man 4 Yl-b 6 /Man(|31-4)GlcNAc(|31-4)GlcNAc-®-®-Dol Man a Man 1-3 Glc3 Man 2 by a series residues a: twist gly< questered t Glycoprote: in the Col; sequential {DP-6135A: cligcsacc..a Tue g1 C0-“‘3'-‘~18<‘=tes the Sugar 1". and anoneri are found 1 as has been ethers (29- 10 by a series of processing reactions to remove excess glucose and mannose residues and reveal the core ologosaccharide (27,28). Intracellular ves- icular glycoproteins, such as the lysosomal acid hydrolases, may be se- questered before processing to the core oligosaccharide is complete. Glycoproteins bearing only the core oligosaccharide subsequently appear in the Golgi apparatus, where the carbohydrate chains are extended by the sequential transfer of monosaccharides directly from sugar nucleotides (UDP-GlcNAc, UDP-Gal, UDP-GalNAc, GDP-Fuc, GDP-Man and CMP-NeuAc) to the oligosaccharide core, as shown in Figure 4. The glycosyl transfer reactions involved in the biosynthesis of glyco- conjugates are catalyzed by glycosyltransferases, which are specific for the sugar nucleotide donor, the oligosaccharide acceptor, and the linkage and anomerity of the glycosidic bond being formed (9-11). These enzymes are found in the cisternae of the rough endoplasmic reticulum.and Golgi, as has been mentioned, and on the plasma membrane. Roseman (10), and others (29-31), have suggested that the glycosyltransferases may be associated in multienzyme complexes, which carry out the entire synthesis of specific oligosaccharides. This concept stems from the observation that the product of one glycosyl transfer reaction becomes the substrate for the next glycosyltransferase in the sequence. Roseman (10) has also suggested that the glycosyltransferases located on the plasma membrane could play a role in intercellular adhesion. This hypothesis has been em- broiled in controversy (32,33) since its publication and has not been con- cluSively resolved. The catabolism of glycoconjugates is carried out by exoglycosidases (glycosylhydrolases), as shown in Figure 5. These enzymes have been stud- ied more extensively than the glycosyltransferases, partly because they are easily solubilized in aqueous solutions, and partly because of the involvement of many of these enzymes in human glycoconjugate storage dis- eases (1-4), whereas only one glycosyltransferase (GM3: Bfgracetylgalac- tosaminyltransferase) has been shown to be associated with a disorder of glycoconjugate metabolism (20,21). The glycosidases were thought to be localized primarily in the lysosome (34-36), however sialylated forms of these enzymes are also present in plasma (37-39), where they may play a significant role in modulating the turnover of circulating glycoproteins, and neutral glycosidases are thought to be present in the endoplasmic re- ticulum, where glycoprotein processing reactions are carried out. Indeed, 11 Figure 4. The Biosynthesis of a Glycoprotein with Asparagine—Linked Carbohydrate. The biosynthesis of the carbohydrate portion of a glycoprotein with asparagine-linked oligosaccharides is shown. This proceeds with trans- fer of the oligosaccharide from dolichol pyrophosphate to asparagine residues on the glycoprotein, followed by processing of the oligosac- charide to reveal the core structure. Carbohydrate residues are added by specific glycosyltransferases in the Golgi apparatus to complete the structure of the glycoprotein (8,9). NEUAC (a 2 HeuAckxz- 12 Man [Man 0‘1-6 Man(pl-4)GlcNAc(Pl-4)GIcNAc-®-®-Dol _ _ al-3 6ch Man2 Man / nascent protein Man 4-Man “1'6 6’ I Man(Pl-4)GlcNAc(pl-4)GIcNAc-A'sn Glcj-Manz-Man “1'3 1‘ Ser (Thr) \3 Ole \6 Man Man al-b 6 l Man(Pl-4)GlcNAc(Bl-4)GlcNAc-Alsn Man (XI-3 1‘ Ser (Thr) 2 UDP'GICNAC\ 2 UD P-Gal\ 2 CMPNeuAc\ NeuAcexz-o)Gal(pl-4)cIcNAc(pl-2)Man 6 l Man(pl-4)GlcNAc(pl-4)GlcNAc—Asn X NeuActh-é)Gal(Pl-4)GlcNAc(pl-2)Man “1'3 I Ser (Th r) l3 Calf] Figure 5. An Example of Glycoconjugate Catabolism The catabolism of the glycosphingolipids of the globo family is shown. l4 GaINAc(cl-3)GaINAcial-3)cal(a1-4)cal(pl-4)clc(pl-1')Cer a-fl-Acetylgalactosaminidase (d-Galactosidase B) 6 GalNAc(pl-3) Gal(0tl-4)Gal(P1—4)Glc(i31—l ')Cer Sandhoff's Disease I p-lfl-Acetylhexosaminidases A and B Gal(al-4)Gal(a1-4)Glc(Bl-l')Cer Fabry's Disease a—Galactosidase A Gal(fll-4)Glc'(P1-l') Cer i 6 p-Galactosidase Glc(pl-l ') Cer Gaucher's Disease p-Glucosidase Ceramide a glucosi 'hich has charides Many tvc units oligosaccl culation; rapid clea liver (LO- t'aat upta}: tor on par galactose minal E-ac. characteri; ificities ‘ lectins be: feugd in pl 611 int thlation 15 a glucosidase has been solubilized from a microsomal fraction of thyroid which has optimal activity toward dolichol pyrophasphate-linked oligosac- charides containing glucose at pH 6.5-7.5 (27). 'Many circulating glycoproteins bear carbohydrate moieties with one or two units of the trisaccharide NeuAc(c2+6)Ga1(Bl+4)G1cNAc- linked to their oligosaccharide (6). These glycoproteins have extended lifetimes in cir- culation; however, removal of the terminal sialic acid residue results in rapid clearance from circulation, with uptake by parenchymal cells in the liver (40-43). Studies by Ashwell and coworkers (43-46) have revealed that uptake of asialoglycoproteins is mediated by a membrane-bound recep- tor on parenchymal cells, which binds glycoproteins containing terminal 8- galactose residues. A receptor that binds glycoproteins containing ter- minal Efacetylgl cosamine residues has also been purified and partially characterized (47-49), and receptors with other carbohydrate-binding spec- ificities will surely be found. These receptors are often referred to as lectins because of their similarity to the carbohydrate-binding proteins found in plants (50). An interesting variation on this theme is the discovery that phos- phorylation of mannose residues on B-glucuronidase is required for rapid uptake by skin fibroblasts (51,52). This marker for rapid uptake has also been identified on other glchproteins (53-55). Circulating glycosidases may regulate the turnover of circulating glycoconjugates by hydrolyzing their terminal carbohydrate substituent and exposing their internal carbohydrate residues to hepatic receptors. Glycoconjugates associated with the plasma membrane, either as components of the membrane or bound to receptors on the cell surface, presumable en- ter the cell by endocytosis and subsequently appear in the lysosomes. Mo- nosacCharide residues are sequentially hydrolyzed from the nonreducing end of the oligosaccharides by the lysosomal glycosidases. Unlike the glycosyltransferases, which appear to have absolute speci- ficity for the sugar being transferred, the oligosaccharide acceptor, and the linkage and anomerity of the glycosidic bond being formed, several of the glycosidases lack absolute specificity for the monosaccharide being hydrolyzed (56-60), the glycoconjugate substrate (glycolipid, glycoprotein or proteoglycan) (61—63), and the linkage (but not the anomerity) of the glycosidic bond being hydrolyzed. Aside from these differences in the specifi thesis para 118 II. "5‘1 a. PCT. Sm Sitcosflce glycc'lipic 16 specificities of the glycosyltransferases and glycosidases, the biosynv thesis and catabolism of glycoconjugates appear to be carried out in parallel, but converse, pathways. II. The Chemistry and Metabolism of Glycoconjugates ContainingiTerminal c-Galactose Residues A. Purification and Characterization of Glycoconjugates Sources and methods for purifying glycolipids (5,12,64-66), glycopro- teins (6,13-15,66,67) and proteoglycans (68,69), and methods for deter- mining the composition (5,11,66-68) and structure (11,15,66) of the car- bohydrate portions of glycoconjugates have been extensively reviewed, and only recent developments in glycoconjugate purification and characteriza- tion will be discussed. .Crude glycolipid preparations have been fractionated into their individual components by silicic acid column chromatography (65,66) and thin-layer chromatography on glass plates coated with silica gel G (70). Improved resolution of glycolipids has been obtained using a new porous silica bead, Iatrobeads (Iatron Laboratories, Inc., Tokyo, Japan) (71,72). Other improvements in silicic acid chromatography include the use of -. borate-coated silicic acid, which enabled Korniat and Hof (73) to resolve glucosylceramide and galactosylceramide, and the use of DEAE-silica gel to isolate neutral glycolipids and gangliosides (74). High resolution of glycolipids has been obtained by thin-layer chromatography using plates coated with silica gel of fixed pore size (Silica Gel 60, E. Merck, Darmstadt, West Germany) in solvents containing CaClz (75,76). Recent advances in glycoprotein purification include the use of Sepharose-linked lectins to isolate glycoproteins containing specific carbohydrates (77,78), affinity chromatography (79-81), ampholyte dis- placement chromatography (82), and hydrophobic chromatography on octoyl- Sepharose (Pharmacia Fine Chemicals, Piscataway, NJ) (39). Recently developed methods for structural analysis of complex carbo- hydrates include high pressure liquid chromatography (83-88), natural abundance Fourier-transform 13C NMR spectroscopy (11,89), the use of puri- fied bacterial endoglycosidases (90-94) to hydrolyze specific internal glycosidic bonds of oligosaccharides, and the use of anhydrous hydrogen fluoride to hydrolyze Qrglycosidic bonds of oligosaccharides, but not the 17 amide linkages of proteins or Efglycosidic bonds (95). B. Structure and Function of Glycoconjugates Containing Terminal a-Galactose Residues Glycoconjugates of human origin that contain terminal a-galactose residues are quite varied in structure, ranging from the relatively simple diglycosyl glycolipid, galabiosylceramide, to the complex blood group B- active megalosaccharides (Figure 6). The occurrence and function of each of these glycoconjugates will be discussed. Galabiosylceramide. The biological function of GaOseZCer is not known at the present time, although it may serve as a precursor for galabiosylcer- amide-II3-sulfate, which probably occurs in kidney, and may play a role in sodium ion transport (95-99). The occurrence of GaOseZCer is relatively restricted in man: it is .not found in normal or Fabry plasma or erythrocytes, but has been found in normal (100,101) and Fabry (102-106) kidney, Fabry urine sediment (100), heart (108), lung (108), pancreas (105) and Tay-Sachs brain (109). This lipid is also found in kidneys of Balb/c, C57/BL and A strain mice (103). The concentration of GaOse2Cer was higher in kidneys from male mice (111) and it was found that testosterone stimulated the appearance of this glyco- lipid in kidneys of female mice (112). Globotriglycosylceramide. Globotriglycosylceramide has been identified by Naiki and Marcus (113) as the blood group Pk antigen. Individuals with this blood type are rare, as the Pk phenotype, unlike other blood group phenotypes, is inherited as a recessive trait (114-116). The PR antigen can be detected immunologically on erythrocytes (116-120), fibroblasts and lymphocytes (121-123) from individuals with this phenotype, but not on normal erythrocytes, which also contain GbOse3Cer (but in much lower lev- els). Marcus et al. (124) found that type Pk erythrocytes contain almost 5 times more GbOse3Cer than normal erythrocytes, but are devoid of globo- side (the blood group P antigen), which is the major glycosphingolipid present in normal erythrocytes (125). It was recently demonstrated that fibroblasts from individuals with this phenotype lack GbOse3Cer: Bfifir acetylgalactosaminyltransferase activity (126,127). Globotriglycosylceramide was first isolated by Klenk and Lauenstein (128) in 1953, from human erythrocytes. Sweeley and Klionsky (102) re- ported abnormal quantities of this glycolipid accumulating in the kidney 18 Figure 6. Structures of Glycoconjugates Containing Terminal a-Galactose Residues. The oligosaccharide portions of glycoconjugates containing terminal a—galactose residues are shown. They are: (I) galabiosylceramide; (II) globotriglycosylceramide[Pk glycolipid]; (III) Iva-a-galactosyllactoneo- tetraglycosylceramide [P1 glycolipid]; (IV) and (V) blood group B-active glycolipids [these oligosaccharides differ only in the linkage of the pentultimate galactose residue]; and (VI) megalosaccharide structure of blood group B glycoproteins. VI in “I (7) C) G 0!th I~ u \ (396633: \ 19 I GoI(aI—4)Go|(eI—I')Cer II GoI(°‘|—4)Gol($l-4)Glc(Pl-|')Cer III GoI(dl-4)Gol(Bl-4)GIcNAc(PI-3)Gol(Bl-4)GIciiBI-I')Cer GoHal-3)Go|(i3l-3)GlcNAc(i5|-3)GOI(I3|-4)Glc(i3|-l')Cer IV ‘ a|—2 Fuc Gol(°(I—3)GoI(BI-4)G|cNAc(BI-3)Gol(i3I—4)Glc(i3l-I')Cer V oil-2 Fuc GoHal—3)GoI(IBI-4)GICNAc(fil-3)GoI(fil-4)GIcNAc(PI-3)GoRBI-4)GIc(BI-I')Cer VI 637 protein fracticr “350' Plasma we 53‘1“! (althoug- ‘3-‘35 higher the was) from thESE isolated Ii"-3_< :35 not bear Pm *1 (141) is: group B activig if the terminal. Yes a t”minal ( and in the hu: OI glycoconjuga' “fibOhy drate 20 from a patient with Fabry's disease in 1963, and it has now been found to occur in most extraneural tissues and fluids (5). The absence of GbOse3Cer in normal brain suggests that it is not synthesized by nervous tissue. This corresponds with the absence of a-galactosidase A from normal brain (129), which would not be required, as the substrate is not present, and the lack of central nervous system involvement in Fabry's disease (130- 133). The distribution of GbOse3Cer among lipoprotein fractions in normal (134-136) and Fabry (136) plasma has been examined; the majority of this glycolipid (>602) was found to be associated with the low density lipo- protein fraction. The distribution of GbOse3Cer among lipoproteins in Fabry plasma was found to be in proportions similar to the normal distri- bution (although the concentration of GbOse3Cer in each fraction was 2-3 times higher than the control values). Ivu-a-Galactosyllactoneotetraglycosylceramide. This glycolipid has been identified by Naiki et a1. (137,138) as the human blood group P1 antigen. Individuals with the P1 phenotype are quite common (comprising approx- imately 75% of the population) (115,124), and this antigen has been de- tected immunologically on erythrocytes (117), fibroblasts and lymphocytes (122) from these individuals. Marcus (139) and Naiki et al. (138) have isolated IV“-c-Gal-LcnOse4Cer from human erythrocytes, but this glycolipid has not been purified from other sources. Eto et al. (140) and Stellner et al. (141) isolated a glycolipid from rabbit erythrocytes with blood group B activity, differing from the human P1 antigen only in the linkage of the terminal o-galactose residue. The rabbit erythrocyte glycolipid has a terminal Gal(al+3) linkage, rather than a terminal Gal(al+4) linkage found in the human P1 antigen. It appears that the blood group activity of glycoconjugates is determined not only by the identity of the terminal carbohydrate residue and its anomerity, but also by its linkage to the pentultimate residue of the oligosaccharide. The accumulation of this glycolipid in patients with Fabry's disease has not been reported, al- though it would be expected to accumulate in patients with this pheno- type. Blood Group B-Active Glycoconjugates. The blood group B antigens are glyco- lipids and glycoproteins that have terminal Gal(ol+3)Gal(Bl+3(4))GlcNAc trisaccharide groups in their carbohydrate moieties. Three different blood group B-active glycolipids have been purified 21 from human B erythrocytes (142-147) and Fabry pancreas (148), and their structures determined. Two of these glycolipids are hexaglycosylceramides differing only in the linkage of the pentultimate galactose residue, as shown in Figure 4. The third blood group B-active glycolipids found to date contain fucose; however, Eto et al. (140) and Stellner et al. (141) have purified a blood group B-active pentaglycosylceramide from rabbit erythrocytes that does not contain fucose, but does contain the terminal trisaccharide mentioned above. Gardas and Koscielak (149,150) recently reported the presence of a megaloglycolipid with blood group B activity in human erythrocyte membranes. This water—soluble substance may contain more than 20 carbohydrate residues linked to ceramide. Although the blood group B-active megaloglycolipid has not been purified or structurally characterized, a blood group H-active megaloglycolipid with 22 sugars has been purified from human erythrocyte membranes and the structure examined (151). This megaloglycolipid is presumably the precursor (the core struc- ture) for blood group A- and B-active megaloglycolipids. Blood group B-active glycoproteins have been isolated from human urine, saliva, meconium, gastric juice and ovarian cyst fluids (152-155). The structure of the carbohydrate portion of these glycoproteins is shown in Figure 6. This complex carbohydrate contains both the lacto and lactoneo structures found in blood group B-active glycosphingolipids (5,148), and the structure of the core oligosaccharide (with the terminal c-galactose residue removed) resembles the structure of the blood group H-active megaloglycolipid, suggesting similar synthetic and catabolic pathways. Blood group B-active oligosaccharides have been isolated from urine of individuals with this blood type (155). The concentration of these oligosaccharides in urine varies with diet and the secretor status of the individual, with higher concentrations of oligosaccharides found in urine from unstarved secretors. These fucose-containing oligosaccharides vary in size from 3-7 sugars and are presumably derived from partially degraded blood group B-active glycoproteins. The blood group B-active glycolipids have been reported to accumu- late in Fabry patients with this blood type (148). Tondeur and Resibois (156) reported the accumulation of mucopolysaccharides in patients with Fabry's disease, but this finding has not been supported in studies by other investigators (157-160). Unfortunately, these studies were carried out prior to the discovery of blood group B-active glycolipid accumulation 22 in Fabry pancreas (148), and the presence or absence of stored mucopoly- saccharides was not correlated with blood type. Blood group B-active glycoproteins and oligosaccharides derived from these glchproteins would be expected to accumulate in tissues of Fabry patients with this blood type, and to be present in high concentration in the urine of blood group B secretors with Fabry's disease. C. The Biosynthesis of Glycoconjugates Containing Terminal a-Galactose Residues None of the enzymes involved in the synthesis of glycoconjugates con- taining terminal u-galactose residues has been purified, except for the c-galactosyltransferase involved in the synthesis of blood group B-active glycoconjugates. The synthesis of GaOseZCer, catalyzed by crude mouse kidney homogen- ates, has been examined by Gray and coworkers (161,162). These investi- gators found that testosterone stimulated the biosynthesis of GaOseZCer in kidney of female mice (112), who had much lower levels of this glyco- lipid than male mice (111). However, this increase was not due to an alteration in c-galactosyltransferase activity, but rather appeared to be the result of an increase in galactosylceramide biosynthesis (112). From these studies, it appears that GaOseZCer levels may be regulated by the availability of galactosylceramide. The biosynthesis of GbOse3Cer has been examined by Stoffyn et al. (163) in rat kidney microsomes, and by Robbins et a1. (164-169), Hakomori et a1. (170-175), Macpherson et a1. (176-178) and others (179,180) in normal and transformed hamster cells and newborn rat kidney cells in cul- ture. These investigators found that when contact-inhibited cells became confluent, the levels of GbOse3Cer appeared to increase, but when trans- formed cells were used, the concentration of GbOse3Cer was not altered by cell population density. The biosynthesis of GbOse3Cer has been studied as a function of cell cycle in NIL cells (168) and human KB cells (181- 184). Globotriglycosylceramide, and glycolipids in general, were synthe- sized during the Cl phase of the cell cycle. Vance et al. (185) examined the biosynthesis and catabolism of neutral glycolipids in plasma of normal humans and a patient with Fabry's disease by labelling the newly synthesized glycolipids with [6,6-2Hzlglucose. These workers found that the incorportaion [6,6-2H2]g1ucose into plasma 1? her a “23 ga Rtich 23 GbOse3Cer was'reduced in the patient with Fabry's disease. It was sug- gested that the [2H]Gb0se3Cer may have been diluted by unlabeled GbOse3Cer in the patient's plasma, or that the rate of GbOse3Cer biosynthesis may be reduted in patients with Fabry's disease due to the high levels of this glycolipid that accumulate in tissues of these patients. Recently, Kijimoto-Ochiai et al. (126) examined the biosynthesis of GbOse3Cer and GbOse4Cer in fibroblasts from individuals with blood group Pk and p phenotypes. They found that individuals with the Pk phenotype lack the Bfiflracetylgalactosaminyltransferase for GbOse4Cer biosynthesis. Similarly, individuals with the p phenotype are missing the a-galactosyl- transferase involved in the biosynthesis of GbOse3Cer. These glycosyl- transferase deficiencies do not result in pathological disorders of glyco- lipid metabolism; individuals with these phenotypes appear to be physical- ly normal and to have normal life spans (126). This is in contrast to GM3 gangliosidosis (GM3: Bfifiyacetylgalactosaminyltransferase deficiency), which is characterized by hepatosplenomegaly, poor physical and motor development and central nervous system dysfunction (20,21). Nevertheless, the blood group p, PR (and H) phenotypes may be classified as hereditary glycosyltransferase deficiencies. The biosynthesis of IV“-n-Gal-LcnOse4Cer has not been investigated at the time of this writing. The synthesis of GaOseZCer, GbOse3Cer and Ivu-c-Gal-LcnOse4Cer are carried out by a-galactosyltransferases which catalyze the formation of terminal Ga1(al+4)Gal glycosidic bonds. Ko§cielak et al. (164) have sug- gested that GbOse3Cer (Pk antigen) and Ivu-c-Gal-LcnOse4Cer (P1 antigen) may be synthesized by the same c-galactosyltransferase. Galabiosylcer- amide may also be synthesized by this transferase. This hypothesis is supported by the absence of both GbOse3Cer and IV”-a-Lcn09e4Cer in eryth- rocytes of the p phenotype (115); however, this hypothesis has not been investigated in further detail. The a-galactosyltransferase involved in the biosynthesis of blood group B-active glycoconjugates catalyzes the formation of terminal Gal(o1+3)Gal(2+luFuc) glycosidic bonds. This glycosyltransferase has been purified from plasma of type B individuals by Carne et al. (187) and Nagai et al. (188). These investigators found that the enzyme was a dimer with subunits of 40,000 daltons. It required Mn+2 for maximal activity and the pH optimum was at pH 7.0-7.5 (188). The enzyme transferred 24 galactose from UDP-Gal to H—active glycoproteins, H—active megaloglycolipid and also Fuc(ol+2)Ga1, 2'-fucosyllactose and lactoffirfucopentaoBe I (187). Alterations in the biosynthesis of blood group B-active glycosphingo- lipids have been reported in human carcinomas. Blood group B antigens were missing from urinary epithelial, gastrointestinal, uterine, lung, pane creatic, and adenocarcinomas (68,188,189,257,258). Stellner and Hakomori (189) reported that the disappearance of blood group B-active glycolipids was due to the loss of the c-galactosyltransferase involved in the bior synthesis of these glycolipids. The c-galactosyltransferase involved in the biosynthesis of blood group B-active glycoconjugates is clearly distinct from the enzyme(s) in- volved in the synthesis of GaOseZCer, GbOse3Cer and Ivh-c-Gal-LcnOseACer. The blood group B-active glycocnjugates contain terminal Gal(al+3)Gal linkages, while GaOseZCer, GbOse3Cer and IV“-a-Ga1-LcnOse4Cer contain terminal Gal(al+4)Gal linkages. Replacement of a terminal Gal(al+3) linkage with a terminal Gal(cl+4) linkage, as in IV3-c-Gal-LcnOse4Cer from rabbit erythrocytes, compared with human erythrocyte Ivu-n-Gal-LcnOse4Cer, results in a change in blood group specificity. The similarities between the structures of the blood group H—active megaloglycolipid (151) and the blood group B megalosaccharide (6) suggest that these complex carbohydrates may be synthesized by a common pathway. The synthesis of these blood group-active glycoconjugates may proceed with the transfer of sugars from sugar nucleotides directly to the nascent glycolipid or glycoprotein. D. The Catabolism of Glycoconjugates Containing Terminal c-Galactose Residues The occurence of GbOse3Cer hydrolase was reported in 1967 by Brady et al. (190), who detected this hydrolase activity in particulate fractions of rat brain, liver, kidney, spleen and small intestine. These investi- gators subsequently found this enzyme activity in normal human intestine; it was absent in hemizygous male Fabry patients and reduced in'heterozyr gous female patients (191). This finding suggested that the accumulation of glycolipids in Fabry's disease was due to the absence of GbOse3Cer hydro- lase. The identity of GbOse3Cer hydrolase as an c-galactosidase was sug- gested by Kint (192) in 1970. He reported that leukocytes from a patient with subs unbei disea pcrti (193- '. ARC!" LU..~ 4 cf ta mated 25 with Fabry's disease were unable to hydrolyze the water-soluble artificial substrates, p-nitrophenyl-a-Qrgalactopyranoside (p-NP-c-Gal) or 4~methyl- unbelliferyl-ufigrgalactopyranoside (4-MU-a-Gal), suggesting that Fabry's disease is due to an c-galactosidase deficiency. This finding was sup- ported by similar studies using cultured skin fibroblasts and leukocytes (193-196). Structural studies on GbOse3Cer (194-201) and GaOseZCer (202) confirmed that the terminal galactose residue was linked in an u-config- uration, while the pentultimate galactose residue was linked in the 8 configuration. In 1971, Beutler and Kuhl (203) and Kint (204) reported the presence of two forms of d-galactosidase activity in normal human tissues, desig- nated c-galactosidase A and B. These forms could be distinguished by their different thermostabilities, Michaelis constants with artificial substrates, and electrophoretic mobilities at pH 7.0. Beutler and Kuhl found that only the A form was missing in patients with Fabry's disease, while a-galactosidase B persisted. Furthermore, the B form isolated from Fabry tissues, leukocytes and fibroblasts (203,206) was indistinguishable from normal o—galactosidase B activity. These a-galactosidase activities have now been examined in normal human intestinal mucosal scrapings (206), liver (206-215), heart (206), brain (209), spleen (206,216), kidney (206, 210,217,218), skeletal muscle (206), placenta (77,220-223), fibroblasts (206,214,224-228), leukocytes (206,227,228), endothelial cells (229), plasma (39,218,228,230—232), urine (218,228,233,234) and tears (235). c-Galactosidase A is the major form in these tissues and fluids, with the exception of brain, while the B form represents 20% or less of the total a-galactosidase activity. In contrast, d-galactosidase B is the major form in brain, with only traces of the A form detectable (209). The presence of two forms of c-galactosidase activity posed a dilemma to investigators in this area. It was found that treatment with C. perfringgng_neuraminidase reduced the electrophoretic mobility of c-galactosidase A to approximately that of the B form (204,206,209,219, 220). This observation led Kint (204,209) to suggest that neuraminidase treatment converted the A form to the B form. The failure of neuramini- dase treatment to alter the electrophoretic mobility of the B form (206) appeared to support this hypothesis. However, closer examination of the desialylated A form by Beutler and Kuhl (219,220) using antibodies pre- pared against purified c-galactosidases A and B, and by Romeo et al. (211) ll using isoe iiffers fr apparent t it to the of the car he B for: scrate spe is also an o-nitraphe It as; carries ou' galactosa: DI CrOSS-r Early 0f th Slycomnju absent in . pathologic. {minal a. ionj Ugates 26 using isoelectric focusing, revealed that desialylated a-galactosidase A differs from the B form in antigenicity and isoelectric point. It is now apparent that neuraminidase treatment of o-galactosidase does not convert it to the B form, although it appears that sialic acid may be a component of the carbohydrate moiety of d-galactosidase A and may not be present in the B form. Subsequent studies by Dean et a1. (60,236,237) on the sub, strate specificities of a-galactosidase A and B revealed that the B form is also an ofoacetylgalactosaminidase, with much greater activity toward o-nitrophenyl-ofigfacetylgalactosaminide (o—NP-a—GalNAc) and GbOseSCer (Forssman antigen) than towards substrates containing terminal a-galac- tose residues. It was concluded from these studies that a-galactosidase B probably functions as an afflfacetylgalactosaminidase rather than an c-galactosidase in vivo. This finding was confirmed by Schram et al. (238) using antibodies to the purified enzymes. It appears that a single enzyme, a-galactosidase A (EC 3.2.1.22), carries out the hydrolysis of terminal o-galactose residues, regardless of their linkage to the oligosaccharide. The absence of this enzyme in patients with Fabry's disease results in the accumulation of glycolipids, and probably glycoproteins, glycopeptides and oligosaccharides, containing terminal a-galactose residues. III. The Chemistry and Metabolism of Glycoconjugates Containing Terminal a-N-Acetylgglactosamine Residues The glycoconjugates of human origin that contain terminal affiracetyl- galactosamine residues share the property of having blood group A activity or cross-reactivity. In contrast to the common occurrence in humans of many of the glycoconjugates containing terminal a-galactose residues, glycoconjugates containing terminal degracetylgalactosamine residues are absent in a majority of the population. Furthermore, there are no known pathological disorders of metabolism involving glycoconjugates containing terminal ufiyfacetylgalactosamine residues. Nevertheless, these glyco— conjugates are of interest because of their blood group activity (115) and their involvement in human carcinomas (68,188,189,239,240,257—260). 27 A. Structure and Function of Glycoconjugates Containing Terminal o-N-Acetylgalactosamine Residues The Structures of the carbohydrate portions of the glycoconjugates found in man containing terminal affiracetylgalactosamine residues are shown in Figure 7. The occurrence and function of each of these glyco- conjugates will be discussed. Globgpentaglycosylceramide. In 1911, Forssman (241) reported that the injection of extracts from guinea pig kidney into rabbits resulted in the formation of hemolysins for sheep red blood cells. The antigen that evoked the formation of these hemolysins, the Forssman antigen, was sub- sequently found to be a heterogenetic antigen, occurring in many species of animals and bacteria (242). Brunius (242) found that the antigen was a glycolipid containing galactosamine, and Papirmesiter and Mallette (243) reported that it contained hexose, hexosamine, fatty acid and base. The structure of the Forssman antigen isolated from horse spleen was de- termined by Siddiqui and Hakomori (244) in 1971. The accuracy of the structure proposed by Siddiqui and Hakomori has been confirmed using Forssman antigen from dog intestine (245), goat erythrocytes (246) and guinea pig tissues (247). The Forssman antigen (GbOseSCer) has not been isolated from human erythrocytes, and was thought not to occur in humans until Hakomori et al. (239) reported its presence in gastrointestinal mucosa of some normal individuals. The presence of GbOseSCer in gastro- intestinal mucosa was not dependent on the blood type of the individual. These investigators also suggested that GbOse5Cer may be present in low levels in erythrocytes from some individuals. Globopentag1ycosy1ceramide is also expressed in human hepatic biliary adenocarcinoma tissue (240) and in gastric and colonic tumors (239). Globopentag1ycosylceramidecross-reacts extensively with blood group A antigens (248,249). This specificity probably resides in the terminal GalNAc(ul+3) moiety, which is common to GbOse5Cer and the blood group A antigens. Blood Group AeActive Glycoconjugates. The blood group A antigens are glycolipids and glycoproteins that have a common terminal tetrasaccharide, shown below. Hakomori et al. (250) have isolated and partially GalNAc(al+3)Gal(Bl+4)G1cNAc 01+2 Fuc 28 Figure 7. Structures of Glycoconjugates Containing Terminal afngcetylgalactosamine Residues ElVigp The oligosaccharide portions of glycoconjugates containing terminal afigfacetylgalactosamine residues are shown. They are: (I) globopentagly- cosylceramide (Forssman antigen); (II), (III) and (IV) blood group A-ac- tive glycolipids, a fourth A-active glycolipid has been described by Hakomori et al. (250) that resembles IV, but with excessive GlcNAc and an additional branching structure and; (V) the megalosaccharide struc- ture of blood group A-active glycoproteins. E r , (T) GOHMUU!‘ 29 I GoINAc(otl-3)Gol(|3l—3)Gol(aI-4)Go|($l-4)GIC(PI—I')Cer GoINAc(Oil-3)Gol(fil-4)GICNAC(i3l—3)(Go|)n(P|-4)Glc(i3l-l')Cer Isl—2 Fuc II GoINAc(«I-3)Gol(i3I--4)GICNAC(BI-3)GoI(BI—4)GICNAC(BI- 3)(Gol)n(t3|-4)Glc(DI-I')Cer III ldl-Z Fuc Fuc dI-Z GolNAc(80% of its original activity after 4 hours at 50°C, pH 4.3. The Michaelis constant with p-NP-u-GalNAc was 3.1 mM and the hydrolysis of this substance was competitively inhibited by N—acetylgalactosamine and galactose. Tallman et al. (263) observed that one form of placental c-galactosidase also exhibited a-hexosaminidase activity, however this activity was not described in further detail. In 1977, Dean et a1. (60, 236,237) reported that a-galactosidase B from normal human liver was, in fact, an o-N-acetylgalactosaminidase, hydrolyzing GbOseSCer and o-NP-c- GalNAc at much greater rates than substrates containing terminal c—galac- tose residues. This finding suggests that a-N—acetylgalactosaminidase lacks absolute specificity for the substituent at C-2 of galactose. Schram et 31. (215,238) confirmed the identification of a-galactosidase B as an a-N—acetylgalactosaminidase using antibodies prepared against the purified enzyme. I From the studies of Yamamoto (262) and Dean et a1. (60,236,237) 11'- appears that a single o-N—acetylgalactosaminidase catalyzes the hydro- 1Ysis of globopentaglycosylceramide and blood group A antigens. This enzyme may be constitutive, since it was detected in tissues from per- 3033 With various blood types; however, the presence of Forssman antigen in the gastrointestinal mucosa or in erythrocyte membranes was not exam- ined and the presence of this antigen may induce the appearance 0f “'31..- acetylgalactosaminidase . lfl-‘F' -‘ "F .pna—.\..-~ q VHF 9* \n .\, .'.‘..I ...). Gene Dry 1 MATERIALS AND METHODS MATERIALS SOLVENTS General Dry Methanol D ry Pyridine Dry Acetonitrile CHEMICUXLS I_)_-Galactal _I_)_- (+) -Fucose S o dium Borohydride Sodium Borohydride- [3H] Fluram DETERGENTS \ SOdium Tauro cho late Triton x-100 \ 4rMethylumbellifery1- o"“JQI-Galactopyranoside P‘Nitrophenyl-a-N- cetylgalactosam'i'nide P‘Nitrophenyl-o-Q- a.lactopyranoside Solvents were redistilled by constant flow-rotary evaporation. Methanol (500 ml) was refluxed over magnesium turnings (2.5 g) and iodine (0.1 g) for 1 hour and then distilled and stored over 3A Davison molecular sieves. Pyridine was refluxed over barium oxide for 1 hour and then distilled and stored over potassium hydroxide pellets. Acetonitrile was redistilled and stored over type 3A Davison molecular sieves. Research Products International Corp. Elk Grove Village, IL Aldrich Chemical Company, Inc. Milwaukee, WI Sigma Chemical Co. St. Louis, MD New England Nuclear Boston, MA Pierce Chemical Co. Rockford, IL Calbiochem La Jolla, CA Research Products International Corp. Elk Grove Village, IL Research Products International Corp. Elk Grove Village, IL Research Products International Corp. Elk Grove Village, IL Sigma Chemical Co. St. Louis, MO 33 i7} 646‘ Mt . n A 9', ;‘ (3:4)?» 3. ——-— USA.) (30 44Methylumbellifery1-B- N-Acetylgalactosaminide CHROMATOGRAPHY SUPPLIES ENZ DEAR-Cellulose (DE-52) Sephadex G-150 Sephadex G-25 (Fine) Hypatite C (Hydroxylapatite) Unisil (100-200 mesh) (Silicic Acid) Iatrobeads Silica Gel G Thin-Layer Chromato- graphy Plates Silica Gel 60 Thin-Layer Chromato- graphy Plates 3% OV-225 on Supelcoport (80-100 mesh) 5% SE-30 on Gas Chrom Q (80-100 mesh) 37!. SP-2340 on Supelcoport (100-120 mesh) Dowex 50W-X8 (SO-100 mesh) AG 501-X8 {20-50 mesh) AG 1-xa (20-50 mesh) Dowex l-X8 (SO-100 mesh) YMEHS \ Ga lactose Oxidase P1'-'<>nase H01- se Radish Peroxidase 34 Pierce Chemical Co. St. Louis, MO Whatman, Inc. Clifton, NJ Pharmacia Fine Chemicals, Inc. Piscataway, NJ Sigma Chemical Co. St. Louis, NJ Clarkson Chemical Company, Inc. Williamsport, PA Clarkson Chemical Company, Inc. Williamsport, PA Iatron Laboratories, Inc. Tokyo, Japan Analtech, Inc. Newark, DE E 0 Meer Darmstadt, Germany Supelco, Inc. Bellefonte, PA Applied Science Laboratories, Inc. State College, PA Supelco, Inc. Bellefonte, PA Sigma Chemical Co. St. Louis, MO Bio-Rad Laboratories Richmond, CA Bio-Rad Laboratories Richmond, CA Sigma Chemical Co. St. Louis, MO Worthington Biochemical Corp. Freehold, NJ Calbiochem La Jolla, CA Worthington Biochemical Corp. Freehold, NJ mrr ARV'YY I '1: F‘LLLL'L . . .. Bio-E Pro Sta For fl messafil] 5t. Lav: Hedrcal lensing 35 MISCELLANEOUS REAGENTS Bio-Solv BBS-3 Beckman Instruments, Inc. Fullerton, CA PPO Research Products International Corp. (2,5-Diphenyloxazole) Elk Grove Village, IL Dimethyl-POPOP Research Products International Corp. (1,4-bis-2-[4-methy1-5- Elk Grove Village, IL phenyloxazolyl] -benzene Carrier Ampholytes LKB Produkter AB (pH 3-5) Bromma, Sweden Phenylmethylsulfonyl- Sigma Chemical Co. fluoride St. Louis, MO Protein Molecular Weight Boehringer Mannheim Corp. Standards New York, NY Porcine intestines were obtained from Peets Packing Company, Chessaning, MI. Normal human liver was kindly provided by Dr. Leo Walker, St. Lawrence Hospital, Lansing, MI; Dr. Wanderly De Mendonza, Ingham Medical Center, Lansing, Mi; Dr. Lawrence Simpson, E.W. Sparrow Hospital, Lansing, MI and Dr. Allen Yates, Dept. of Pathology, Ohio State University, Columbus, OH. Tissues were stored frozen at -80°C prior to use. All other reagents and materials used in these studies were of reagent grade . METHODS I. Purification of o-Galactosidases A and B from Human Liver Extraction. Approximately 1 kg of normal human liver was homogenized with two volumes (w/v) of 1 mM phenylmethylsulfonylfluoride (PMSF) in a Waring blender at 4°C. It was necessary to stir the PMSF in distilled water for several hours, with warming, to dissolve it. The PMSF solution was not buffered, as the pH of the liver homogenate in distilled water approxi- mated the pH of maximum stability for hepatic a-galactosidases at pH 6.5 (212). The homogenate was centrifuged at 100,000 x g for 1 hour to obtain a clear supernatant. Ammonium Sulfate Precipitation. The 100,000 x g supernatant was adjusted to 302 of saturation with ammonium.sulfate, with gentle stirring, at 4°C. The solution was centrifuged to remove precipitated material. The supere natant was adjusted to 60% of saturation with ammonium.sulfate, and the precipitate was collected by centrifugation. The pellet was suspended in approximately 200 m1 of distilled water and dialyzed for 24 hours ver- sus 10 liters of 10 mM sodium phosphate buffer, pH 6.5. The buffer was changed once during dialysis. DEAR-Cellulose Chromatography. The dialyzed solution was centrifuged to remove precipitated material and was applied to a DEAE-cellulose column (4.3 x 25 cm) that had been prepared according to the manufacturer's in- structions and equilibrated with 10 mM sodium phosphate buffer, pH 6.5.-8t 4°C. The enzymes were eluted with a linear gradient of NaCl {0-0.3 M NaCl) in sodium phosphate buffer. The gradient was monitored with a Radiometer- Copenhagen Model CDM3 conductivity meter. Fractions containing a-galacto- sidase activity were pooled and concentrated approximately lO-fold in an Amicon Model 52 ultrafiltration apparatus equipped with an 800 ml reser- voir and a PM 10 membrane. Sephadex G-150 Chromatography. The crude enzyme solution (<50 ml) was applied to a Sephadex G-150 column (3.0 x 110 cm) that had been prepared according to the manufacturer's instructions and equilibrated with 10 mM sodium phosphate buffer, pH 6.5, containing 0.02% sodium azide, at 4°C. The enzymes were eluted with the same buffer. The fractions containing a-galactosidase activity were pooled and concentrated approximately 10- fold in the Amicon Model 52 ultrafiltration cell equipped with a PM 10 36 a linea l v ‘) U1 n LC I 1 am a can: i: i-fi*~" uh—‘d'y x 37 membrane. The concentrated solution was dialyzed versus two changes of 1 mM sodium phosphate buffer, pH 6.5, for 24 hours. Hydroxylapatite Chromatography. The dialyzed enzyme solution was applied to a hydroxylapatite column (3.0 x 20 cm) that had been equilibrated with 1 mM sodium phosphate buffer, pH 6.5, at 4°C. The enzymes were eluted with a linear gradient of sodium phosphate buffer, pH 6.5 (1-50 mM), and then with 200 mM sodium phosphate buffer, pH 6.5. The gradient was monitored with a Radiometer-Copenhagen Model CDM3 conductivity meter. The fractions containing a-galactosidases A and B were pooled separately. Ampholyte Displacement Chromatography. a-Galactosidases A and B, that had been separated by hydroxylapatite chromatography, were applied individually to DEAE-cellulose columns (1 x 5 cm) that had previously been equilibrated with 10 mM sodium phosphate buffer, pH 5.0. The column was then washed with 2 volumes of distilled water to remove the buffer. The enzymes were eluted with 15 m1 of a carrier ampholyte solution consisting of 1.0 ml of carrier ampholytes (pH 3-5) diluted 1:15 with distilled water. Fol- lowing application of the ampholyte solution, the column was washed with distilled water. The fractions (0.5 ml) containing a-galactosidase ac- tivity were pooled and the ampholytes were removed by Sephadex G-150 chromatography (2 x 90 cm column) as previously described. II. Preparation of Substrates A. Preparation of Glycolipid Substrates Globotriglycosylceramide was purified from porcine small intestines as described below. Galabiosylceramide was purified from Fabry kidney (264) and globopentaglycosylceramide was purified from canine intestines (265) . Globotriglycosylceramide. The procedure of Suzuki et al. (266) was modi- fied for the purification of GbOse3Cer. Approximately 23 kg of porcine small intestines were trimmed of mesenteric fat and connective tissue and washed with water. The intestines were cut into approximately 5 cm segments and homogenized in 5 volumes of acetone in a Haring blender. The homogenate was stored overnight at 4°C. The acetone was removed by vacuum filtration on a Buchner funnel with Whatman No. 4 (fast) filter paper. The residue was air-dried at room temperature. The acetone powder was divided into 1 kg lots and extracted with 3 volumes of 38 chloroform-methanol, 2:1 (v/v) in a waring blender at high speed. The extracts were collected by vacuum filtration on a Buchner funnel with Whatman No. 4 (fast) filter paper. The filter residues were re-extracted with 3 volumes of chloroform-methanol, 1:1 (v/v) and then 3 volumes of chloroformdmethanol, 1:1 (v/v). The extracts were collected, pooled with the previous extracts and dried by rotary evaporation. The dry residue was dissolved in approximately 1.1iter of chloroform-methanol, 2:1 (v/v) and vacuumrfiltered on a Buchner funnel to remove undissolved material. The solution, which had appeared pale yellow as a suspension, was deep red following filtration. The filter residue was resuspended in chloro- formamethanol, 2:1 (v/v) and re-filtered. The filtrate was concentrated by rotary evaporation to a final volume of approximately 300 ml, and 3 liters of acetone were added slowly with stirring. The mixture was cooled to 4°C in the cold room. The suspension was then centrifuged at 7,000 x g for 15 minutes at 4°C in stainless steel centrifuge bottles to col- lect the precipitated glycolipids. The pellets were removed from the centrifuge bottles and the supernatant was discarded. Approximately 65 g of crude glycolipids were obtained. Saponification of contaminating neutral lipids and phospholipids was carried out by dissolving the crude glycolipids in 300 m1 of chloroform and adding an equal volume of 1.0 N aqueous potassium hydroxide. The solution was incubated at 37°C for 12 hours, with stirring. After 12 hours the solution was neutralized to approximately pH 7 with concentrated HCl, transferred to dialysis tubing and dialyzed versus several changes of distilled water for 5 days at 4°C. The lipids retained by the dialysis tubing were concentrated by rotary evaporation using absolute ethanol to azeotrope the water. The glyco- lipids were then lyophilyzed to dryness. Approximately 60 g of glyco- lipids were recovered. Approximately 50 g of the glycolipid preparation were dissolved in 150 m1 of chloroform and applied to a Unisil column (500 g of Unisil; 4.4 x 63 cm) that had been washed with several volumes of chloroform. The column was eluted by a linear gradient of increasing methanol con- centration in chloroform, prepared by mixing 2 liters of chloroform with 2 liters of methanol in a gradient maker. Fractions (10 ml) were col- lected and periodically examined by thin-layer chromatography on silica gel G thin-layer plates (250 u). Thin-layer plates of neutral glycolipids were developed in chloroform-methanol-water, 65:25:4 (v/v/v), and visualized by iediue and heati Tail-lay actuanol' vapcrs, the spra of the r resorcin with 80 tilled v be prote 801' which we‘ stand at :5 the o 1.7: VOIL 39 by iodine vapors or by spraying with 0.5% orcinol in 4 N sulfuric acid and heating the plate at 100°C until the dark purple color developed. Thin-layer chromatograms of gangliosides were developed in chloroform- methanol—7Z ammonium hydroxide, 55:40:10 (v/v/v), and visualized by iodine vapors, orcinol spray, or by spraying with resorcinol reagent, covering the sprayed area with a clean glass cover plate (to prevent evaporation of the reagent) and heating at 100°C until the blue color developed. The resorcinol reagent was prepared by mixing 10 ml of 2% aqueous resorcinol with 80 ml of concentrated-HC1, 0.25 ml of 0.1 M aqueous CuSO4 and dis- tilled water to a final volume of 100 ml. The resorcinol reagent should be protected from light and refrigerated when not in use (267). Some of the glycolipids were still contaminated by phospholipids, which were removed by mild alkali-catalyzed methanolysis as described by Vance and Sweeley (125). The lipids were dissolved in 10 m1 of 0.6 N methanolic NaOH and 10 ml of chloroform. The solution was allowed to stand at room temperature for 1 hour, after which 1.2 volumes (referring to the original volume of methanolic NaOH used) of 0.5 N methanolic HCl, 1.7 volumes of water and 3.4 volumes of chloroform were added and the bi- phasic system was mixed. The mixture was then centrifuged at low speed to recover the glycolipids in the lower, chloroform phase. Next, the chloroform phase was washed twice with theoretical upper phase (268) [chloroform-methanoldwater, 3:48:47 (v/v/v)] and then dried under a stream of nitrogen at 50°C. An alternative to gradient elution of the Unisil column is stepwise elution with solutions containing increasing concentrations of methanol in chloroform, as described by Suzuki et al. (266). The final yield was approximately 600 mg of pure globotriglycosyl- ceramide. 1?H]Globotriglycosylceramide. Globotriglycosylceramide was labeled with tritium at C-6 of the terminal galactose residue by the method of Suzuki and Suzuki (269). A portion (50 mg) of the purified globotriglycosyl—‘ ceramide (above) was placed in a large screw-cap test tube with 4 ml of freshly distilled tetrahydrofuran and 4 ml of 0.1 M potassium phosphate buffer, pH 7.0. Galactose oxidase (427 units) was dissolved in 0.5 ml of 0.1 M potassium phosphate buffer, pH 7.0, and was added to the reaction mixture.. The solution was incubated at room temperature for 4 hours with gentle shaking. An additional 427 units of galactose oxidase, dissolved ~ in 0.5 a wmnu tion um WV): ‘ washed : 3:548:47 hydrofu' 0.1 N X. tempera additict was the: mthano] loner p‘: phase, c of nitrc sodium t room ter Eezh OXAZQ (a. 3 527.8% 40 in 0.5 ml of potassium phosphate buffer was added, and the incubation was continued overnight. The tetrahydrofuran was then removed from the solu- tion under a stream of nitrogen and 5 volumes of chloroformdmethanol, 2:1 (v/v), were added. The upper phase was removed and the lower phase was washed once with theoretical upper phase (268) [chloroformdmethanoldwater, 3:48:47 (v/v/v)]. The lower phase was dried in vacuo and 5 ml of tetra- . hydrofuran plus 0.4 m1 of [3H]-sodium borohydride solution (10 mCi/ml in 0.1 N NaOH) were added. The sample was incubated, with shaking, at room temperature overnight. Excess sodium borotritiide was destroyed by the addition of 0.7 ml of 10 N acetic acid in the hood. The tetrahydrofuran was then removed under a stream of nitrogen and 5 volumes of chloroform- methanol, 2:1 (v/v), were added. The upper phase was removed and the lower phase was washed once with theoretical upper phase. The lower phase, containing [3H]-1abeled glycolipid, was then dried under a stream of nitrogen and 5 m1 of tetrahydrofuran and 10 mg of solid, unlabeled sodium borohydride were added. The solution was incubated overnight at room temperature, with shaking, after which the excess borohydride was destroyed by the addition of 0.7 ml of 10 N acetic acid in the hood. Tetrahydrofuran was removed under a stream of nitrogen and 5 volumes of chloroform-methanol, 2:1 (v/v), were added. The upper phase was removed and the lower phase was washed 10—15 times with theoretical upper phase. This exhaustive washing reduced the background radioactivity of the assay to a practical level. The [3H]-1abeled GbOse3Cer was purified by prepara- tive thin-layer chromatography on 500 u silica gel G-coated plates de- veloped in chloroformdmethanoldwater, 65:25:4 (v/v/v). The [3Hl-labeled glycolipid was located by brief exposure to iodine vapors, or by scanning the thin-layer plate on a Varian Aerograph Berthold Radio Scanner. The glycolipid was eluted form the silica gel scrappings by washing the gel thoroughly with chloroform-methanoldwater, 100:50:10 (v/v/v). Radioactivity was determined with a Beckman LS-150 liquid scintilp lation counter. Scintillation solvent was prepared according to the method of Suzuki and Suzuki (269) and contained 7.0 g PPO (2,5-diphenyl- oxazole) and 0.6 g of dimethyl POPOP (1,4-bis-2-[methyl-S-phenyloxazolyl]- benzene) and 100 ml of Bio-Solv BBS-3 in 1000 m1 of scintillation toluene. Prior to the addition of scintillation solvent, the sample was dried under a stream of nitrogen and redissolved in 0.5 m1 of water. The sample was counted in 10 ml of scintillation solvent. i Galabiae fornali: ddnal g describ M A‘Lfine a..... purifie the ter sylcera 3 :1 of phcspha percxid buffer . fwdh titaal « continut added, cf gala: 05 nitrc with the (351). 9'; Ell ( 'u’Ere adc OVEInigi 9.7 ml ‘ reamed '1 4:1 (V/t 41 Galabiosylceramide. Galabiosylceramide, which had been purified from a formalin-fixed Fabry kidney (264), was [3H]-labeled at C-6 of the ter- minal galactose residue by the galactose oxidase-[3H]borohydride method described above. Globopentaglycosylceramide. Globopentaglycosylceramide, which had been purified from canine intestines (265), was tritium labeled at C-6 of the terminal Nfacetylgalactosamine residue as follows. Globopentaglyco- sylceramide (20 mg) was placed in a large screwbcapped test tube with 4 m1 of freshly distilled tetrahydrofuran and 4 ml of 0.1 M potassium phosphate buffer, pH 7.0. Galactose oxidase (427 units) and horse radish peroxidase (1000 units) were dissolved in 0.5 ml of potassium phosphate buffer and added to the reaction mixture. The solution was incubated for 4 hours at room temperature with gently shaking, and then an addi- tional 427 units of galactose oxidase were added. The incubation was continued overnight and then a final 427 units of galactose oxidase were added. The incubation was terminated 24 hours after the final addition of galactose oxidase by the removal of the tetrahydrofuran under a stream of nitrogen and the addition of 5 volumes of chloroformdmethanol, 2:1 (v/v). The upper phase was removed and the lower phase was washed once with theoretical upper phase (chloroform-methanol-water, 3:48:47 (v/v/v)) (261). The lower phase dried in vacuo and 5 ml of tetrahydrofuran and 0.4 m1 of [3H]-sodium borohydride solution (10 mCi/ml in 0.1 N NaOH) were added. The sample was incubated with shaking at room temperature overnight. Excess sodium borotritiide was destroyed by the addition of 0.7 ml of 10 N acetic acid in the hood. The tetrahydrofuran was then removed under a stream of nitrogen and 5 volumes of chloro ormrmethanol, 2:1 (v/v), were added. The upper phase was removed and the lower phase was washed once with theoretical upper phase. The lower phase was washed once with theoretical upper phase. The lower phase, containing the [3H]-labe1ed glyolipid, was then dried under a stream of nitrogen and 5 ml of tetrahydrofuran and 10 mg of solid, unlabeled sodium borohydride were added. The solution was incubated overnight at room temperature with shaking, after which the excess borohydride was destroyed by the addition of 0.7 ml of 10 N acetic acid in the hood. Tetrahydrofuran was removed under a stream of nitrogen and 5 volumes of chloroformdmethanol, 2:1 (v/v), were added. The upper phase was removed and the lower phase was washed 10—15 times with theoretical upper phase. This exhaustive dashin level. tograp rather. by bri on the Rf ide from t , ICI‘E‘S 42 washing reduced the background radioactivity of the assay to a practical level. The [3H]GbOse5Cer was purified by preparative thin-layer chroma- tography on a 500 u silica gel G-coated plate developed in chloroform- methanol-water, 65:25:4 (v/v/v). The [3H]—1abeled glycolipid was located by brief exposure to iodine vapors, or by scanning the thin-layer plate on the Varian Aerograph Berthold Radio Scanner. The [3H]GbOse5Cer had an Rf identical to that of authentic GbOseSCer. The glycolipid was eluted from the silica gel scrapings by washing the gel thoroughly with chloro- formdmethanol-water, 100:50:10 (v/v/v). B. Oliggsaccharide Substrates Oligosaccharides were prepared from [3H]-labeled and unlabeled GaOseZCer, GbOse3Cer and GbOseSCer by ozonolysis followed by treatment with mild base (264,270, 271). The procedure described below was used without modification for the preparation of GbOse3, GaOsez and GbOseS. Ozone was generated by passing oxygen through a high voltage spark chamber equipped with a Supelco high voltage generator. The generation of ozone was confirmed by bubbling the gas through a 2% K1 solution. The glyco- lipid was dissolved in dry methanol to make a 0.4% solution and ozone bubbled through the solution for 4-5 hours. Following ozonolysis,~the solvent was removed and the residue was dissolved in 0.2 M sodium car- bonate to make a 1% solution. The reaction mixture was incubated at room temperature overnight. The reaction was terminated by neutralizing the A solution on a column (1 x 8 cm) of Dowex SOWAXB (H+) ion exchange resin (50-100 mesh). Following application of the reaction mixture to the column, it was eluted with water. The aqueous solution was extracted once with 5 volumes of chloroformdmethanol, 2:1 (v/v), and 3 times with hexane to remove the degraded products of ceramide. The aqueous phase was lyo- philyzed to obtain the oligosaccharide. The oligosaccharides were examined by thin-layer chromatography on silica gel G thin¥layer plates developed in n-propanol-acetic acid—water, 85:12:3 (v/v/v), and by gas-liquid chromatography as trimethylsilyl methyl- glycosides (125). h C. Artificial Substrate Analogs The synthetic substrate analogs, 45MU-c-Gal and p-NP-c-Gal, were ob- tained from commercial sources. The artificial substrate, p-NP-a-GalNAc. became un inrestiga p—hP-a-Ga analogs, z-D-fucor I prepared a-Tiitroe': hug—nub Sweeley tied of 1 atetylat: heated a 0 reel «Peking tone. ature. Y.‘ ' ‘13”; q 43 became unavailable from commercial sources during the course of these investigations, and it was necessary to synthesize o-NP-a-GalNAc and p—NP-o-GalNAc, as previously described (133). The synthetic substrate analogs, p-nitrophenyl-Z-deoxy-c-_D_-galactopyranoside and o-nitrophenyl- a—R—fucopyranoside (o-nitrophenyl-6-deoxy-o-I_)_—galactopyranoside), were prepared by the following methods. p-Ni trophenyl-Z-Deoxy—c-D-Ga lac topyranoside . Sweeley (133) and Woods and Kramer (272) were modified for the prepara- Q—Galactal was The methods of Dean and tion of p-nitrophenyl-Z-deoxy-a—QB—galactopyranoside. acetylated by the method of Wallenfels and Lehman (273), as follows. Anhydrous sodium acetate (1.8 g) and acetic anhydride (60 ml) were heat- D-Galac- ed in a 100 m1 round-bottom flask at 100-120°C for 30 minutes. tal (3.0 g) was added to the acetylation mixture and the solution was heated at loo—120°C for 30 minutes. The acetylation mixture was allowed ‘30 c001 to room temperature and was then poured slowly into 100 m1 of cold 5‘7. aqueous sodium carbonate, with stirring. The mixture was exe- tracted twice with chloroform and the combined extracts were washed once With water. The chloroform was removed by rotary evaporation at 460°C to Yield g-galactal-tri-Q-acetate (5.5 g, 98% yield) as a yellow oil. Phenyl-tri-Q—acetyl-Z-deoxy-o—g-galactopyranoside was prepared by Shaking Q—galactal-tri—Q—acetate (5.5 g) and phenol (2.0 g) with 4 drops Of c011 centrated HCl on a New Brunswick shaker for 4 hours at room temper- ature - The reaction was terminated by neutralizing the reaction mixture With 4 drops of 12.5 N NaOH. The reaction mixture was dissolved in ethyl- ene dichloride and washed once with water, twice with 0.25 N NaOH and then one “30 re time with water. The lower, ethylene dichloride phase, .was dried With Q«alciurn chloride and the solvent removed by rotary evaporation at I"I6()"‘(: to yield an oil. Phenyl-tri—Q-acety1-2-deoxy-a-Q-ga1actopyranoside Was recrystallized from 95% ethanol as flat white plates (yield 1.9 g, ““P- 14o-142°c). Phenyl—tri-Q—acetyl-2-deoxy—o-D—galactopyranoside (1.8 g), dissolved in 7’ 5 ml of acetic acid, was nitrated with 3 m1 of a nitration mixture consisting of 3 volumes of 90% (fuming) nitric acid and 10 volumes of acetic anhydride. The solution was incubated at 37°C for 2 hours and then 18 m1 of cold 2 M potassium acetate was added and the solution was allowed to Stand at room temperature for 3 hours. The reaction mixture was then e"“7-‘1‘acted 3 times with chloroform and the pooled chloroform extracts were A_— washed water . vent re 952 et‘ I‘ chloro vent h 44 washed once with cold 2 M aqueous sodium carbonate and once with cold The chloroform phase was dried with calcium chloride and the sol- water . The product crystallized readily from vent removed by rotary evaporation. 95% ethanol as long white needles (yield 0.7 g, m.p. ISO-152°C). The product was Q—deacetylated by dissolving the crystals (0.7 g) in chloroform (10 m1) and heating the solution to boil until 420% of the sol- Anhydrous methanol (8 ml) was added and the solution vent had escaped. Sodium methoxide (0.4 ml), prepared pre- was heated to boiling again. viously by mixing anhydrous methanol with sodium metal (3 g) in an ice bath in the hood until the evolution of hydrogen gas had ceased, was added to the boiling reaction mixture. The solution was removed from the heat, stoppered, and incubated at room temperature for 4 hours. the end of the incubation, 1 drop of acetic acid was added to adjust the The solvent was removed on the rotary evaporator At solution to NpH 6.5. and p-nitrophenyl-Z-deoxy-a-R—galactopyranoside was recrystallized from 95% ethanol (yield 0.5 g, m.p. 178-181°C). Fourier-transform 13C-NMR SPGCtra were obtained on a 15.08 mHz Bruker WP-60 spectrometer. O‘Nitro hen l-6-Deo -a-D-Galacto ranoside o-Nitro hen l-a-D-Fuco- W. 2-(+)-Fucose (3.0 g) was acetylated by the method of walletxfels and Lehman (273) as previously described for the synthesis of Tetra-Q—acetyl-B-g-fuCOpyranoside was obtained as a p‘NP~2-deoxy-a-Gal. Phenyl-trifigracetyl-c-Drfucopyranoside Yellow 011 (6.0 g, 99% yield). was prepared by the method of Conchie et al. (274), modified as described below- Tetra-gracetyl-B-g-fucopyranoside was placed in a 100 m1 round- bottom flask with 6.0 g of freshly distilled phenol. Anhydrous zinc chloride (2.0 3) (prepared by fusing zinc chloride in a crucible and allowing it to cool in a desiccator), disablved'in 5 m1 of acetic..a¢id- acetic anhydride, 95:5 (v/v), was added. The reaction mixture was im- merged in an oil bath at 60-80°C and the pressure was reduced to 50-70 tor-r With a water aspirator. These reaction conditions were maintained £01.. 2 hours with stirring, and then the heat was removed and the pressure returned to normal. The reaction mixture was allowed to cool overnight. The resulting red syrup was dissolved in ethylene dichloride and washed sucqessively with water, 0.2 N NaOH and twice more with water, to remove Zinc chloride and unreacted phenol. The lower ethylene dichloride phase, Q0“taining phenyl-tri-Q—acetyl-c—D—fucopyranoside, was dried with calcium ch'-‘~Or'ide and the solvent removed by rotary evaporation. The product was active 39.-i: 13% St was w. pheny form, 45 obtained as a yellow oil (6.3 g, 94% yield). Phenyl-tri-ancetyl-a—QrfucOpyranoside was nitrated as previously described for the preparation of p-nitrOphenyl-Z-deoxy-c-Qrgalactopyrano- side. The mixture of o- and p-nitrophenyl-trifigracetyled-Qrfucopyranosides, obtained as a red oil (3.8 g, 51% yield), was defigfacetylated as previously described. A mixture of o- and p-nitrophenyl-o-QrfucOpyranosides was ob- tained as a red oil (1.9 g, 36% yield). The o-nitrophenyl isomer was purified by Iatrobead chromatography. A column of Iatrobeads (1.5 x 30 cm) was prepared using Iatrobeads (not heat- activated) suspended in chloroform. The mixture of o— and p-nitrophenyl- a—Qrfucopyranosides, dissolved in approximately 2 m1 of chloroform contain- ing several drops of methanol, was applied.to the column and the column= was washed with approximately 10 ml of chloroform. The o- and p-nitro- phenyl-a-Drfucopyranosides were eluted with a methanol gradient in chloro- form, prepared by mixing 100 ml of chloroform with 100 ml of chloroform- methanol, 7:3 (v/v), in a gradient maker. The ortho isomer eluted first, followed by the para isomer contaminated by the ortho isomer. Fractions containing pure o-nitrophenyl-a-nyucopyranoside were pooled and recrystal- lized from absolute ethanol as white needles (0.8‘g, 15% yield,.m.p.'l90- 13 193°C). Fourier-transform C-NMR spectra were obtained on a 15.08 mHz Bruker WP-60 spectrometer.- III. Assays The ability of a—galactosidase A and c-galactosidase B (digracetyl- galactosaminidase) to hydrolyze glycosphingolipids, oligosaccharides and artificial substrates containing terminal a-galactose or ojgracetylgalac- tosamine residues was investigated. The reaction conditions for assays with these substrates are described below. A. Assays with Glycosphingglipid Substrates The glycosphingolipids used in these studies are not soluble in water, and a detergent (sodium taurocholate) was therefore employed to solubilize them. The effect of sodium taurocholate on the hydrolysis of glycosphingo- lipids by a—galactosidases A and B will be discussed in a later section. Assays with Globotriglycosylceramide. The reaction conditions for the hydrolysis of GbOse3Cer were different for the reactions catalyzed by C! 3‘- P9! 46 c-galactosidase A and c-galactosidase B (afigfacetylgalactosaminidase). The assay with o—galactosidase A was carried out as follows. An aliquot (100 pl) of 1 mM [3H]GbOse3Cer (1000 cpm/nmole) in chloroformdmethanol, 2:1 (v/v), was mixed in a test tube with 100 pl of 0.5% sodium tauro- cholate (w/v) in theoretical upper phase (chloroform-methanol-water,3:48: 47 (v/v/v). The solvent was removed under a stream of nitrogen and the residue was redissolved in 50 ul of Gomori citrate-phosphate buffer, pH 4.1 (275), with sonication. An aliquot (up to 50 ul) of the enzyme solution to be assayed was then added. The concentration of protein in the reaction mixture should not exceed 40 ug/lOO ul, as it has been re- ported (207,217,218) that higher protein concentrations inhibit the hy- drolysis of GbOse3Cer, presumably by binding to the negatively charged detergent-glycolipid mixed micelles. Following the addition of the en- zyme solution, the reaction mixture was taken up to a final volume of 100 pl with glass distilled water. The final concentrations.of GbOse3Cer and sodium taurocholate were 1.0 mM and 9.3 mM, respectively. The re- action mixture was incubated at 37°C for 1 hour and then the reaction was terminated by the addition of 100 pl of water, 400 pl of methanol and 200 pl of chloroform.. The homogeneous solution was mixed on a Vortex mixer and then an additional 200 pl of chloroform was added, as described by Bligh and Dyer (276). The biphasic system was mixed and then centrifuged to resolve the phases. The upper phase was removed with a Pasteur pipet and the lower phase was washed once with 0.5 ml of theoretical upper phase (chloroformdmethanol-water, 3:48:47 (v/v/v)). 'The pooled upper phases were washed once with 1 m1 of theoretical lower phase (chloroformrmethanol- water, 86:14:1 (v/v/v)). The upper phase was then transferred to a scin- tillation vial and dried under a stream of nitrogen. The residue was dissolved in 0.5 m1 of water and 10 m1 of scintillation solvent, consisting of 7.0 g PPO, 0.6 g of dimethyl-POPOP and 100 m1 of Biosolv BBS-3 dissolved in 1,000 m1 of toluene (269), was added. The solution was mixed and counted id a Beckman LS-150 liquid scintillation counter. The assay for GbOse3Cer hydrolysis by a-galactosidase B (ajgfacetyl- galactosaminidase) differed slightly from the assay with aegalactosidase A. An aliquot (100 pl) of 1 mM [3H] GbOse3Cer (1000 cpm/nmole) in chloro- form—methanol, 2:1 (v/v), was mixed in a test tube with 80 ul of 0.5% sodium taurocholate (w/v) in theoretical upper phase (chloroformdmethanol- water, 3:48:47 (v/v/v)). The solvent was removed under a stream of nitroger 0.1 h s< enzyme aixture water. 1.0 a! 7°C f phase moved sanic An a1 added 47 nitrogen and the residue was redissolved, with sonication, in 50 ul of 0.1 M sodium citrate buffer, pH 4.3. An aliquot (up to 50 ul) of the enzyme solution to be assayed was then added, and the reaction mixture was taken up to a final volume of 100 pl with glass-distilled water. The final concentrations of GbOse3Cer and sodium taurocholate were 1.0 mM and 7.4 mM, respectively. The reaction mixture was incubated at 37°C for 6 hours, the reaction was terminated‘and the liberated [3H]ga- lactose was quantitated as previously described for the assay with a-ga- lactosidase A. Assays with Galabiosylceramide. The hydrolysis of GaOseZCer was examined with a-galactosidase A only. An aliquot (100 pl) of 1 mM [3H]GaOse2Cer (1000 cpm/nmole) in chloroform-methanol, 2:1 (v/v), was mixed in a test tube with 60 ul of 0.5% sodium taurocholate (w/v) in theoretical upper phase (chloroformdmethanol-water, 3:48:47 (v/v/v)). The solvent was re- imoved under a stream of nitrogen and the residue was redissolved, with sonication, in 50 ul of Gomori citrate-phosphate buffer, pH 4.1 (275). An aliquot (up to 50 ul) of the enzyme solution to be assayed was then added and the reaction mixture was taken up to a final volume of 100 pl with glass;distilled water. The final concentrations of GaOseZCer and sodium taurocholate were 1.0 mM and 5.6 mM, respectively. The reaction mixture was incubated for 1 hour at 37°C, the reaction was terminated and the liberated [3H]ga1actose was quantitated as previously described for the assay with GbOseBCer. Assays with Globopentaglycosylceramide. An aliquot (100 pl) of 1 mM [3H]GbOse5Cer (500 cpm/nmole) in chloroformdmethanol, 2:1 (v/v), was mixed in a test tube with 80 ul of 0.5% sodium taurocholate (w/v) in theoretical upper phase (chloroform-methanol-water, 3:48:47 (v/v/v)). The solvent was removed under a stream of nitrogen and the residue was redissolved, with sonication, in 50 ul of 0.1 M sodium citrate buffer, pH 3.9. An aliquot (up to 50 ul) of the enzyme solution to be assayed was then added and the reaction mixture was taken up to a final volume of 100 pl with glass-dis- tilled water. The final concentrations of GbOseSCer and sodium taurocholate wer 1.0 mM and 7.4 mM, respectively. The reaction mixture was incubated at 37°C for 30 minutes, the reaction was terminated by the addition of 4 m1 of chloroformdmethanol, 2:1 (v/v), water (0.85 ml) was added, and the biphasic system was thoroughly mixed on a Vortex mixer and then centri- fuged to resolve the phases. The upper phase was washed once with 2.5 m1 ”."-_-..- _ of theor« dried in describ and sod Reactic I-galaz descrifi £58229 .....__ lactos- Cpl-’12: stream rate-g 232333 to a f tratic for 1 48 of theoretical lower phase (chloroform-methanoldwater, 86:14:1 (v/v/v), dried in a scintillation vial and counted as previously described. B. Assays with Oligosaccharide Substrates The oligosaccharides used in these studies were derived from glyco— sphingolipids by ozonolysis and treatment with mild base, as previously described. These complex carbohydrates were soluble in aqueous solutions, and sodium taurocholate was therefore omitted from the reaction mixtures. Reaction conditions for the hydrolysis of GaOseZ, GbOse3 and GbOses by a-galactosidase A or a-galactosidase B (oijacetylgalactosaminidase) are described below. Assays with Galabiaose. The hydrolysis of GaOsez was examined with a-ga— lactosidase A only. An aliquot (250 ul) of aqueous [3H]GaOse2 (500 cpm/nmole) was placed in a test tube and the solvent was removed under a stream of nitrogen. The residue was redissolved in 25 pl of Gomori cit- rate-phosphate buffer, pH 4.5 (275), and an aliquot (up to 25 ul) of the enzyme solution to be assayed was added. The reaction mixture was adjusted to a final volume of 50 ul with glass-distilled water. The final concen- tration of GaOsez was 5.0 mM. The reaction mixture was incubated at 37°C for 1 hour and then the reaction was terminated by the addition of 100 pl of methanol. The total reaction mixture was applied to a 500 u silica gel G thin-layer plate, along with galactose and lactose standards, which were spotted on either side of the reaction mixture. The chromatograpm was developed in n-propanol—acetic aciddwater, 85:12:3 (v/v/v), in a paper- lined tank that had been allowed to equilibrate for several hours prior to use. The galactose and lactose standards were visualized by covering the lanes of the reaction mixtures with a glass plate and spraying the exposed lanes with 0.5% orcinol in 4 N sulfuric acid. The plate was then placed in an oven at @100°C until a purple color developed. Regions of the unsprayed lanes that corresponded in Rf to the galactose standard were scrapped and the [3H]galactose was eluted in a small column with 30 ml of 50% ethanol. The solvent was removed under a stream of nitrogen and the [3H]galactose was quantitated as previously described. Assays with Globotriaose. The hydrolysis of GbOse3, catalyzed by both a-galactosidase A and o-galactosidase B (afflyacetylgalactosaminidase), was examined. The assay for GbOse3 hydrolysis by a-galactosidase A was car- ried out as follows. An aliquot (250 pl) of aqueous 1 mM [3H]GbOse3 (500 cps/moi stream 0 rate-ph enzyme a mafh tration for l h! isolate 49 cpm/nmole) was placed in a test tube and the solvent was removed under a stream of nitrogen. The residue was redissolved in 25 ul of Gomori cit- rate-phosphate buffer, pH 4.5 (275) and an aliquot (up to 25 ul) of the enzyme solution to be assayed was added. The reaction mixture was adjusted to a final volume of 50 ul with glass-distilled water. The final concen- tration of GbOse3 was 5.0 mM. The reaction mixture was incubated at 37°C for 1 hour and then the reaction was terminated. The [3H]ga1actose was isolated and quantitated as previously described for the assay with GaOsez. Assays for GbOse3 hydrolysis by a-galactosidase B were carried out as follows. An aliquot (250 pl) of aqueous 1 mM [3H]GbOse3 was placed in a test tube and the solvent was removed under a stream of nitrogen. The residue was dissolved in 25 ul of Gomori citrate-phosphate buffer, pH 4.8 (275) and an aliquot (up to 25 ul) of the enzyme solution to be assayed was added. The reaction mixture was adjusted to a final volume of 50 ul with glass-distilled water. The final concentration of GbOse3 was 5.0 mM. The reaction mixture was incubated at 37°C for 6 hours and then the reac- tion was terminated. The [3H]ga1actose was isolated and quantitated as previously described for the assay with GaOsez. Assays with Globopentaose. The hydrolysis of GbOse5 was examined with a—galactosidase B (cgflfacetylgalactosaminidase) only. An aliquot (250 pl) of aqueous 1 mM [3H]GbOse5 (500 cpm/nmole) was placed in a test tube and the solvent was removed under a stream of nitrogen. The residue was re- dissolved in 25 ul of Gomori citrate-phosphate buffer, pH 4.4, and an ali- quot of the enzyme solution to be assayed (10 ul) was added. The reac- tion mixture was adjusted to a final volume of 50 ul with glass-distilled , water. The final concentration of GbOseS was 5.0 mM. The reaction mix- ture was incubated at 37°C for 45 minutes and then the reaction was ter- minated and the liberated [BHINracetylgalactosamine was isolated and quan- titated as previously described for the assay with GaOsez. C. Assays with Artificial Substrates Artificial substrates were employed to monitor the purification of u-galactosidases A and B because of their availability and the ease of the assay compared with assays employing glycolipid or oligosaccharide sub- strates. In addition, artificial substrates were employed to probe the substrate specificities of these enzymes. ate. be ESCGQCE a Turne' 50 Assays with 44Methylumbelliferyl-a-D-Galactopyranoside. The procedure described below is a modification of the method of Desnick et al. (228). A stock solution of 5.0 mM 44MU-a-Ga1 was prepared in Gomori citrate-phos- phate buffer, pH 4.6 (275). An aliquot (300 p1) of this solution was placed in a test tube and an aliquot (up to 50 p1) of the enzyme solution to be assayed was added and the reaction mixture was adjusted to a final volume of 350 pl with glass-distilled water. The final concentration of 44MU—o-Gal in the reaction mixture was 4.3 mM. The reaction mixture was incubated at 37°C for 10-30 minutes, after which the reaction was termin- ated by the addition of 4.65 ml of 0.1 M ethylenediamine, pH 11. Fluor- escence was read on an Aminco J4-7439 Fluoro-Colorimeter, equipped with a Turner 110-811 (Corning 7-60) excitation filter and Turner 110-831 and 110-816 (Wratten 48 and 2A) emission filters. The fluorimeter was first calibrated with standard solutions of 4-methyumbelliferone that were pre- pared and stored in water, and diluted up to volume with 0.1 M ethylene- diamine just before use. Assays with 0- or p-Nitrophenyl-o-N-Acetylgalactosaminide. The procedure described below is a modification of the method of Sung and Sweeley (265). An aliquot (50 p1) of 10 mM o- or p-NP-a-GalNAc in 0.1 M sodium citrate buffer, pH 4.3, was placed in a test tube. An aliquot (up to 50 pl) of the enzyme solution to be assayed was added and the reaction mixture was adjusted to a final volume of 100 p1. The final concentration of o-NP-a-GalNAc was 5.0 mM. The reaction mixture was incubated at 37°C for 15-30 minutes, after which the reaction was terminated by the addition of 3.0 m1 of saturated sodium borate, pH 9.7, and the optical density of the solution was read at 410 nm in a Gilford 2400 spectrophotometer. Assays with Other Artificial Substrates. Assays with 4~methylumbellifery1- Bgfifacetylgalactosaminide, to check for contamination by Bfflracetylhexos- aminidase, were carried out as follows. An aliquot (50 p1) of 2.0 mM 4-MU-B-GalNAc in Gomori citrate phosphate buffer, pH 4.4 (275), was placed in a test tube. An aliquot (up to 50 p1) of the enzyme solution to be assayed was added and the reaction mixture was adjusted to a final volume of 100 pl. The final concentration of 4-MU-B-GalNAc was 1.0 mM. The reaction mixture was incubated at 37°C for 10 minutes, after which the reaction was terminated by the addition of 4.9 m1 of 0.1 M ethylenediamine and fluorescence was read on an Aminco J4-7439 Fluoro-Colorbmeter, as pre- viously described for assays with 44MU-a—Gal. Assa phenyl-a- assays vi velocitie describe: Lem et rem 51 Assays with p-nitropheny1-2-deoxy-a-Qrgalactopyranoside and o-nitro- phenyl-cfiprfucopyranoside were carried out as previously described for assays with o- or p-NP-a-GalNAc. Michaelis constants (Km) and maximal velocities (vmax) were calculated by weighted double reciprocal plots, as described by Wilkinson (278), on a CDC 6500 computer. Protein Assays. Protein concentrations were determined by the method of Lowry et al. (277) using bovine serum albumin (0-200 pg/ml). IV. Physical Characterization of a-Galactosidases A and B Isoelectric focusing of a-galactosidases A and B was carried out in an LKB 8101 isoelectric focusing column in pH 3-5 carrier ampholytes, in a sucrose density gradient, as described by Vesterberg (279). Native poly- acrylamide disc gel electrophoresis was carried out as described by Wrigley (280). The gels were stained for enzyme activity as described by Gabriel (281), or for protein as described by Malik and Berrie (282). The molecular weights of native a-galactosidases A and B were deter— mined by gel filtration on Sephadex G-150 (3.0 x 110 cm column) with 0.1 M sodium phosphate buffer, pH 6.5, using horse heart cytochrome c, bovine pancreas chymotrypsinogen A, Bovine serum albumin, rabbit muscle aldolase and beef liver catalase as molceular weight standards. .4” heads; in the A aisc as shc areh Iaphy ifib-fal {1:29.45 RESULTS I. Purification of u-Galactosidases A and B The purification of a—galactosidases A and B from normal human liver is summarized in Table l. o-Galactosidase A was purified about 67,000- fold, with approximately 16% recovery, and c-galactosidase B was purified 68,000-fold, with approximately 9% recovery. Following homogenization, ammonium sulfate precipitation and dialysis, the crude enzyme preparation was chromatographed on DEAF-cellulose, as shown in Figure 8. BfgyAcetylhexosaminidase B eluted at a very low salt concentration, well ahead of the a-galactosidases. However, Bfifiyacetyl- hexosaminidase A eluted between a-galactosidases A and B, and was contained in the fractions pooled for further purification. BfiflrAcetylhexosaminidase A also chromatographed with the a-galactosidase activity on Sephadex G-150, as shown in Figure 9. The most effective steps in the purification scheme were hydroxylapatite chromatography and ampholyte displacement chromatog- raphy on DEAR-cellulose. Hydroxylapatite chromatography gave approximately 400—fold purification of the d-galactosidases and completely resolved the A and B forms, with the B form eluting at 10 mM phosphate and the A form eluting at 30 mM phosphate, as shown in Figure 10. BfiflrAcetylhexos- aminidase A (not shown) eluted at 40 mM phosphate, contaminating d-galacto- sidase A. Ampholyte displacement chromatography on DEAE-cellulose is shown in Figures 11 and 12. Bfingcetylhexosaminidase A eluted at pH 5.0, ahead of d-galactosidase A at pH 4.7, and a-galactosidase B, at pH 4.5. The purified a-galactosidases were examined by native polyacrylamide disc gel electrophoresis, as shown in Figure 13. o-Galactosidase A.migrat- ed ahead of the B form, and appeared to have some minor contaminants, when relatively large amounts of protein were used. ‘o-Galactosidase B was appar- ently homogeneous, even when relatively large amounts of protein were used. 52 “uentmmnvennv Hmnyiflvl-ziv \HN>H.~ :HHV-huj—vn abhAUhhtu an udfiudh < moanmwuvflvuflrvtundd~i ‘nbllnv .VAU such .10.- - . . envdwwknasu -t .. q MVN ahUNYHe 53 om.a New.~ Nm.H H.mca oq.a hm.o No.0 ma.o Awa.owa\moaoanv Ao«a\moaoaov huw>auuo oawaoomo uuomomono NH oaH.m om mmo.m eNH.m mme.~ msa.ao ome.oe moan: Houou H0>H1H 55m BORN m Uflm 4 mmmdUflQOUUQHQUIB MO GOHUQUflMfln—"Sm oee.ae amo.ae maa.ma was.m o.mq n.0H H.o wHomI we e.H mem.e m.oH m we H.H ome.ao s.ao a knamuwoumaouno unuaoomanmwp ouhaongam ma o.mm omm.H m.om m we m.mm saw m.as < zonmuwoumaouao pneumooamxouvho m m.e a.mo H.NOH annaumoumaoueu omauo xmeanaom w m.m 0.0 «.mm hammuwoumaouso omoHoHHmolmo woumumoom puma amassed onH .cesaoo ouwumomakxoupmn Bo om x o.m o co poaaaam whoa moemquDmBouoo cedaoo omHIu xovwnmom an eocfimuno mummwfimouomamwla vofiwauoa mHHoHuume one m use < mommpfimouomamold mo housewoumaouso mufiumomamxouvam .OH ouowfim 59 + (w up) [at oquoqd] O O O O O ([3 to an V m N 9 1 l l n l l 1 O 'O N LO 5" o O ‘9 75 IOO |25 50 25 I I I j ..- ugw/luu/pazKImpAq ovmog-xa-dN-d sa|ouuu number fraction 60 041 a .wouooaaoo mama HE A >Hmumaaxoumem mo mdoeuomuh .m.q no on wousHm < ommvumouomamolo .mlm mo .mouhaozaew uofiuumo mo coauoaom m noes moaned was omoasaaoolmfiumz m vow < mommwfimouumamulo mo mfimouoneouuooam How omen opafimahuowhaom m>Huwz .mH ouowfim 65 4 a... § 4 I I ., r «were .. 446* .s ”......le .... ‘. ‘ 3 than.» u l.._ 66 II. Enzymatic Properties of c-Galactosidases A and B A. c-Galactosidase A Artificial Substrates. An artificial substrate (4-MU-u-Gal) was used to monitor the purification of c-galactosidase A. The linearity of this assay with respect to time and enzyme concentration is shown in Figure 14. The pH optimum for 44MU-d-Gal hydrolysis in the absence of detergents was at pH 4.6, as shown in Figure 15. The presence of the nonionic detergent Triton X9100 had no effect on the pH optimum of 4-MU-c-Gal hydrolysis, however the anionic detergent, sodium taurocholate, used to solubilize glycosphingolipids in aqueous reaction mixtures, inhibited the hydrolysis of 4-MU-a-Gal below pH 4.8. The Michaelis constant (Km) for 4-MU-c-Gal hydrolysis was 2.9 mM at pH 4.6, with a Hill slope of 1.0, as shown in Figures 16 and 17. The maximal velocity was 28.7 pmoles/min-mg. The Km was not dependent on pH, as shown in Figure 18A, and Lineweaver-Burk plots of 44MU-o-Gal hydrolysis at pH 3.7,4.0,4.5 and 4.8 were linear, as shown in Figure 19. The Vmax was dependent on pH, as shown in Figure 188. The presence of 9.3 mM sodium taurocholate in the reaction mixture had a pro- found effect on the kinetics of 4-MU-c-Gal hydrolysis at pH 3.7,4.0,4.5, and 4.8, as shown in Figure 20. The Lineweaver-Burk plot of 4-MU-c—Ga1 hydrolysis at pH 3.7 was hyperbolic, however the hyperbolic character of the kinetics diminished as the pH was raised. The artificial substrates p-nitrophenyl—Z-deoxy-a-27ga1actopyrano- side and o-nitropheny1-6-deoxy-c-Qrgalactopyranoside (o-nitrophenyl-cfiyr fucopyranoside) were synthesized to examine the substrate specficity of o-galactosidase.Am- Fourier-transform 13 C-NMR spectra of these artificial substrates are shown in Figure 21. Figure 21A is the Fourier-transform 13C-NMR spectrum of p-NP-c-Gal, which was used as a standard. The anomeric carbon of p-NP—B-Gal (spectrum not shown) appeared at 101.3 ppm. The 130- NMR spectrum of p-nitrophenyl-Z-deoxy—trifgracetylec-Qrgalactopyranoside is shown in Figure 21B. Authentic p-NP-Z-deoxy-c-Gal was not sufficiently soluble in water to obtain a spectrum, so the 13C-NMR spectrum of acetylated p-NP-Z-deoxy-a-Gal is shown. The anomeric carbon appeared at 96.4 ppm and the C-2 methylene appeared at 30.0 ppm. Signals at 170 ppm and 21 ppm are from the acetyl groups. The 13C-NMR spectrum of o-NP-a—DrFuc is shown in Figure 210. The anomeric carbon appeared at 99.1 ppm and the signal at 16.6 ppm was due to the C-6 methyl group. 67 Figure 14. The Linearity of 4-MU-a-Ga1 Hydrolysis by a-Galactosidase A with Respect to Time and Enzyme Concentration The hydrolysis of 4-MU-a-Ga1 by a-galactosidase A was linear with respect to time for up to 1 hour (top) and with respect to enzyme con— centration up to 12 pg/350 pl (bottom). 68 200 ISO- e339 a»: a O 5 IOO‘ .notouazuv 3.06: (min.) Time _ O 5 200 ISO-e IOOd 6:5 32.06»... 30:95sz 3.9:: (pg) Enzyme 69 .Ammmv acumen oumoomozoloumuufio fiuoEou ow uoo powuumo mama mammmm use .w.q me Scamp Susana: no gases? 2: @3325 3320893 538.... :5 ma H:5 .87x 833 2a 3. fit, .8 .835 uaowuuump on zufia o.q me on mm3 4 ommvfimouomamwlo oufia Hmolslpzsq mo mamhaoupan use How aoafiuoo ma o£H .< ommwamouomamwlo sues mammaouwam Hewlslpzlq mo EDEHuoo we may .mH ouowfim 70 0.0 0.m 04v 00_Ix :3...» 5:: m0 .0: 322.852 Eatom .28 mm aw. .5933 o .? 100.. ma-o-nw-v sa|ouru °ugLu.|w/psz.(|oap,(q 71 xuwooam> onu ca uouum any .wE.:HE\mmHoEn n.m~ mmB yams .ofia\moaoao o.~« aaoumafixouaom mm3 mucmawuommma > 2: saw as ad £3 Toasters fit. so. 2a . < mmmwfimODUmHmwls an mfimkHouwzm Hourslbzaq mo uoam xuomluw>mw3mcHA one .oH muowam 72 0.N m. ..-22. 0._ m\_ m0 :5 m.~ "Ev. 3019.357? 1N0.0 100.0 A/l (l_a|ou1u . ugw) 73 .o.H mo macaw m :ufia .ummofifi mmB mammflouvho Hoolslszle mo uoao Hafiz moa < mmmeemouomomouc so mawaooueam Hmuuaupzus mo Boom Hoax any .aH magmas 74 On =26. oeozmgm 0. on o._ 00 ..o -.__r__ _ _-_b?_ p . _ _.O r r .00 Ho._ 0 T- . H rofi Ho. .oouonsinv u on 75 Figure 18. The Effect of pH on the KM and Vmax of 4—MU-a-Ga1 Hydrolysis by a-Galactosidase A. The effect of pH on the Michaelis constant (A), and Vmax (B) is shown. The standard deviation of the Michaelis constants was $0.1 mM and for the maximal velocities was i0.8 pmoles/min-mg. 76 Vmax 4.0- 36. E 3.0“ v. 2.0 30.0- 20.04 38.553205... I0.0- XOE 5.0 4.5 4.0 3.5 pH 77 .naa\mmaoan o.NH kamuwafixouomm mos muomamuommma muaooao> onu ca pound any .w.¢ was m.q .o.e .n.m mo um voowamxo mums < ommvemouumamwlo up mwmzaoupmo Hmolslszlq mo mofiumaflx may < omowfimouumamwla an mfimzaouvzm Hmwlslbzsq mo muoam xuomlum>mm3oowa no mo mo uoommm 0:9 .ma madman 78 ..-—2E. m: ,o.~ m. 0._ 00 o x \, \\\.0 0 I A O 4 A . r..0 . m. m6 .9 rmd M . w a 2. ... m 0..» .¢. .. 5| En .4. o I I n _ 0 .0 35— V I0 1M.O 79 Figure 20. The Effect of pH on Lineweaver-Burk Plots of 4-MU-a—Ga1 Hydrol-: ysis in the Presence of Sodium Taurocholate The kinetics of 44MU—a-Gal hydrolysis by a-galactosidase A were ex- amined in the presence of 9.3 mM sodium taurocholate at pH 3.7, 4.0, 4.5 and 4.8. The error in the velocity measurements was approximately 12.0 nmoles/min. 80 pH 4-MU-a-Gal ‘t- 3.7 with 9.3 mM Taurocholate 0’57 + 4.0 -0- 4.5 -0- 4.8 0.4 . TU) 0 B E C ,5 0.3- E A 0.2< > 2 OJ T I l I I 1 I l O 0.5 LO l.5 2.0 US (mM") 81 Figure 21. Fourier-Transform 13C-NMR Spectra of Artificial Substrate Analogs. Fourier-transform proton-decoupled 13C-NMR spectra of p-nitrophenyl- q-Qrgalactopyranoside, (A); p-nitrophenyl-triggfacetyl-2-deoxy-a-2;galac- t0pyranoside, (B); and o-nitrophenyl-a-Q;fucopyranoside are shown. Spectra were obtained on a 15.08 mHz Bruker WP-60 NMR spectrometer. 82 A , 1i! 4» v- w “New ..‘M MM” we? mowi w WW“ WWW 200 :50 :00 5'0 *1 PPM 83 d-Galactosidase A catalyzed the hydrolysis of p—nitrophenyl-Z-deoxy— cfigrgalactopyranoside (p-NP-Z-deoxy-c-Gal). The hydrolysis of this sub- strate was optimal at pH 4.6. The Km for the reaction was 8.2 mM, as shown in Figure 22, and the Vmax was 37.4'pmo1es/min°mg. The Hill slope was 1.0, as shown in Figure 23. c-Galactosidase A did not catalyze the hydrolysis of o-NP-dfinguc, o-NP-c-GalNAc or p-NP-c-Glc. In addition, the hydrolysis of 4-MU-c—Gal was not inhibited by 10 mM o-NP-a-GalNAc, 20 mM grgalactal or 20 mM 27(+)-fucose. Oligosaccharide Substrates. Thin-layer chromatograms of the oligosac- charides derived from [3H]GaOse2Cer and [3H]GbOse3Cer are shown in Figures 24 and 25, respectively. Examination of these oligosaccharides by gas- 1iquid chromatography, as described by Vance and Sweeley (125), showed that [3H]Ga0se2 contained only galactose and [3H]GbOse3 contained galactose and glucose in the ratio of 2:1. The specific activities of [3H]GaOse2 and [3H]GbOse3 were 25,733 cpm/nmole and 86,169 cpm/nmole, respectively. The assays with these oligosaccharides were linear with respect to time (up to 3 hours) and the concentration of enzyme (up to 130 pg). The hydrolysis of GaOse2 and GbOse3 was optimal at pH 4.5 (Figure 26). The Km with GaOsez was 5.5 mM, as shown in Figure 27, and the Vmax was 15.8 pmoles/min-mg. The Hill slope was 1.0, as shown in Figure 28. The Km with GbOse3 was 4.1 mM (Figure 29) and the Vmax was 13.8 pmoles/min‘mg. The Hill slope for the hydrolysis of GbOse3 was 1.1 (Figure 30). Glycosphingplipid Substrates. Thin-layer chromatograms of purified [3H]GaOse2Cer and [3H]GbOse3Cer are shown in Figures 31 and 32, respectively. Examination of these glycolipids by gas-liquid chromatography showed that GaOseZCer contained only galactose and GbOse3Cer contained galactose and glucose in the ratio of 2:1. The specific activities of [3H]GaOse2Cer and [3H]GbOse3Cer were 25,858 cpm/nmole and 26,768 cpm/nmole, respectively. Assays with GaOseZCer and GbOse3Cer were linear with respect to time and enzyme concentration, as shown in Figure 33. These glycolipids are not soluble in water, and the anionic detergent sodium taurocholate was used to solubilize these substrates in the aqueous reaction mixture. Optimal hydrolysis of GaOseZCer and GbOse3Cer was observed with 5.7 mM and 9.3 mM sodium taurocholate, respectively (Figure 34). No activity could be de- tected when the nonionic detergent Triton X-100 was substituted for sodium taurocholate (concentrations of 0.3-1.2 mM Triton X-100 and 1.0-8.0 mM GbOse3Cer were examined). The pH optimum for the hydrolysis of GaOseZCer .4 no .ofia\moaoao 0.0HH hamumafixoumom mmB mandamuommma hufiuoao> sou sq Hound ooH .we.cfie\moHoE: «.nm mmB ama> moo new :8 ~.w mm3 HowlolaxoowINIAZIo oufis EM onH < owmcfimouomamuuo an mammaouwmm HmolslzxoowlmlmZIo mo uoam xusmluo>mm3ooHA .NN madman 85 0._ 0.0 ...:s. 0.0 m \_ v.0 . N0 .25 we "Ev. .ooIsLSmoumuaza 10.. I 0. N I 0. N) /\/l (l_sa|0uui -U!UJ) 86 .o.H mo oooam m zuHB .umocfifi mmB s ommvamouomamwls an mammaouwhs HmulolmxoowINImZIo mo uoaa Hafiz use 4 mmmpfimouomHmOIo an mammaoumhm HmUIslmxoovINImZIa mo uoam Hafiz use .mN musmam 87 .903 :25. 22.3.6 “wow. ago. new 0. LPL - F b----- p b p _.o a Inge o h no. T .05“ _oo-s->xouo-~-az-a n (500 .Hoofiouo Sofia wouwamoma> mums 8 no mououvhnooumu uoH .mHoeo\5oo mmn.mw mo xua>fiuoo owwaooem m we: was .uoaomom oavmm vaonuuom nomuwouo< smfium> a no woodman was mnfiumooommowaao m>Huom0Hpmu one .vuwvomum a mo mom: mama mmooamwmu was umouoma .wmou Iomaoo .oumHo adamalcfiou 0 How moHHHm s so nonfimuoo was omOfianmemm. mo Emuwoumsouno umamalofisu 05H mmomfiomamummm. mo Emuwoumaouzu pummAIoaoH .qN muowfim ,I 523 89 No.08 2m 3&703 $33-23 2.0210535-.. .Honwouo nufis wouaamnmw> muo3 m” mounuwzconnmo one .oHoan\ano moH.ow mo >ua>wuom cameoonm m was vnm nonnmom oapmm wHosuumm namuwouo< nmfium> n no wonnmom mm3 pneumooommowfiao o>auom0Hcmu moH .mpuownmum mm mom: mums omonammou wnm omouoma .omouomH Inc .ouman nommHInfinu 0 How mafiafim n no wonfimupo mn3 uwomfiuuonoawmmm. mo amuwoumfiouno nonmalnwnu may mmomfiuuonoHummm. mo emuwoumaouno HoamqlnfinH .mm madman 91 988 ,g- .05 I 52.0 3&7an 3.03-204 23041000020-.. 92 .amamv assumes a umoomooalmumuu as u . a oEo0 nuaa m 0 ma um Hmafiuao mmz mmmonu 0nd N omomu mo mfimhaon can was < ommvfimOuomHmo 0 m I A omomfiuuo Loam 0nd mmomfinmamu mo mfimhaouvmm ecu How E: a ans 0 m0 use .0 N muowfim 93 0. ”no I N In - 0 0 a) (0 <2. 2 " o o '3' I fl I I I 1 <2 0. o In N -' 0 Ill /paz£|01pKH arousqns salowu 94 .n«a\moaoan n.0H kaoumafixouonm mm3 munoaounmmoa muaooao> onu na Hound any .w5.naa\moHoE: w.mH mmB x08> onu 0nd :5 m.m mm3 < ommvfimouooamwlo ho Nomomo mo mannaouvhn sou now EM 039 < ommwamouomamolo mo mfimzaouwhm Nomomo mo uoam xunmluw>mo3onHA .nu ounwfim 95 ...:e. m: 0.~ m... 0.. ....0 0 L p p . b p p p p p p \ p - .00 I ..l. A O .....0 . M m. -00 m m. a . ...... 5.5 0.0 ...Ev. . .00 3300 96 Figure 28. The Hill Plot of GaOse2 Hydrolysis by a-Galactosidase A The Hill plot of GaOse2 hydrolysis by o-galactosidase A was linear, with a slope of 1.0. 97 I 600502 5.0. 0 I03 1 05. DJ : 0.053 i 000' I I IVIUIII I I IIIIUII I h 0.I 0.: LG 5.0 no.0 50.0 Substrate (mm (log) .n«a\m0Han m.o hauumafixounnm mmz munmaouommma hufiooao> mnu nfi Hound 059 m" .mn.nfia\moaonn w.mH mwa x05> moo 0nd :8 H.0 was 4 ommvfimouomamwro no momoow mo mfimzaouwmz 050 you EM one 4 ommwfimouomHmUIo an mfimxaounkm momonu mo uoam xenmluo>mmsmnHA .mN ounwfim 99 0. O "N I- .‘Q a .2 .“2 0 r0 - .. m as 1’ ~ .0 E s 0 x f I I I f O a? “2 t “2 . O 0 0 O (' sa|0wu.ugw) A n (mM-') l/S 100 .H.H mo macaw m cufia ummnfia mm3 4 mmmmfimouomamwle so mmwoou mo mammaonpnn mnu mo uoan Hafiz mnH a ammsnmouumoaoru as masons mo mamaoousam was mo scam Hafiz use .om unswnn 101 as. .25. 20:003.. 0.0m 0.0. Cd 0.. 0.0 . ..0 pp P - PPb-bpp p h bb-bbbb - b .o. ”no. m 1.0 1 ”n0 "0.. 1 .0.» 3.03 m 0.0. hxow,‘ (ban 102 .Honwouo nufiz emsfiamomfi> mum3 meanwaoohaw may .mHoEn\Eeo www.mm mo aufi>fiuom cameomom m cm; 0cm .umnnmom 040mm waonuumm oomuwoum4 nmaum> m no wnfinnmom no pmumooa mm3 nfinfiaoomaw m>Huum0H0mu mcH .mnmnnmum 0 mm 0mm: mama meaofiaoohaw Hmuunmn Hmnwummunfi mnfinmu .mumHn HmamHInfisu 0 Hmw mofiHHm m no 0mnfimuoo mmz mnaamnmoaxmowomamwmmm. mo Bmuwoum50uno umnmalnanu mny mmfiemumoahmOHanmoflmm. mo smuwoumaouno ummmqlnfiofi .Hm munmfim 103 . . cmomomo as 20 resonanceo ... 300308 , utoomamoao Icmooco .. 0020 30 N00. 3.2-0 104 .Honauuo :uH3 mosaHman> mumB unamaaoozaw may .mHoan\Eoo 005.0m wo >0H>auum cameomam m 0m: 0nm .umnnmom 040mm naonuumm nomuwoum4 anum> m no mnannmom an 0mumo IOH ems vfiafiaoonaw m>HuomOH0mu moH .mnmvnmum m mm 0mm: muma meaaHHOUMHw ammunmn Hmnaummuna mnfinmu .mumHm um>mHInanu o Hmm mafiafim m no 0mnfimuoo 003 mwflamumuammoo>H0fiuuonoawHmm. mo amuwoumeouno ummmalnwou mLH mvfiamnmoaamooxamfiuuonoaw”mm. mo Smuwoumsounu nmxmqlnwne .mm madman 105 I Soto -.aomamOQo -..0aam030 .-.oomamoDo Icmoooo ....020 .0axwmo. 2¢2-0 T. .003.08 0.0 106 Figure 33. The Linearity of GbOse3Cer Hydrolysis by c-Galactosidase A with Respect to Time and Enzyme Concentration The hydrolysis of GbOse3Cer was linear with respect to time for up to 4 hours (top) and with respect to enzyme concentration, up to 40 pg/100 p1 (bottom). 80.0 107 60.01 40.00 20.00 nmoles GbOseSCer Hydrolyzed l.0 60.0 V 2.0 30 410 (hrs) 50 6.0 Time 40.00 20.0; nmoles GbOse3Cer / hr O l0.0 T 20.0 40.0 50.0 (pg) 30.0 Enzyme 108 .Amnuv H.0 ma .nmwmnn mumnomonmImumuuao Huoaoo na uno wmauumo mums mammm4 .mHm>Huomnmmu .mumaonoounmu anfiv I00 :5 m.m 0nm 28 n.m nufl3 Hmaauao mm3 4 mmmvwmouumamole mo umummmono 0nm umommmomo mo mannaouwhn mnH m w mcfiamumoazmoo I H Huuoooao wnm mvfiamumoamefianmu mo mfimzaonwzm man now nOHumuunmonou mumaonoouan anEHumo mnH .0m madman 109 2535. 20.2.0039? 0.N. 0.0 0.0 0.0 0 P p p p . . p p p - F 0 ..0m emu~00000I\\ .00 O 500m0m080\ o Ink. 0 r00. IOU-”XOW °/o $stle 110 and GbOse3Cer was at pH 4.1 (Figure 35), in contrast to the pH optimum for the hydrolysis of 44MU-o-Gal in the presence of 9.3 mM sodium taurocholate, at pH 4.8, and for 4AMU-a-Gal, p-NP-2-deoxy—a-Gal, GaOsez and GbOse3 in the absence of sodium taurocholate, at pH 4.5-4.6. The Lineweaver-Burk plot of GaOseZCer hydrolysis by u-Galactosidase A was sigmoidal, with inhibition at high substrate concentrations (Figure 36). The apparent Km with GaOseZCer was 0.28 mM and the apparent vmax was 1.09 pmoles/minomg. The Hill plot of GaOsezCer hydrolysis was not linear, but approximated a line with a slape of 1.1, as shown in Figure 37. The kinetics of GbOse3Cer hydrolysis catalyzed by o-galactosidase A resembled those with GaOseZCer. The Lineweaver—Burk plot of GbOse3Cer was sigmoidal, with inhibition at high substrate concentrations. The apparent Km with GbOse3Cer was 0.18 mM and the apparent Vhax was 1.16 pmoles/min-mg (Figure 38). The Hill plot of GbOse3Cer hydrolysis was not linear, but approximated a line with a slepe of 1.0, as shown in Figure 39. B. a-Galactosidase B Artificial Substrates. The artificial substrate 44MU-c-Gal was used to monitor the purification of o-galactosidase B. The hydrolysis of this sub- strate was linear with respect to time and enzyme concentration (Figure 40). The pH optimum for the hydrolysis of 4-MU-o-Ga1 was at pH 4.7, as shown in Figure 41. The Km was 6.8 mM and the Vmax was 18.73 pmoles/min°mg, as shown in Figure 42. The Hill slope was 1.1, as shown in Figure 43. deg: Acetylgalactosaminidase activity co-chromatographed with c-galactosidase B throughout the purification procedure (Figures 8-10 and 12). To determine whether a single enzyme catalyzed the hydrolysis of both u—galactose and cfigracetylgalactosamine residues, or whether these activities only coincided during purification, the physical and kinetic properties of these activities were investigated. Both c-galactosidase B and afflracetylgalactosaminidase activities were thermostable compared with a-galactosidase A, retaining approximately 60% of their original activity after 30 minutes at 55°C (Fig- tire 44). Furthermore, the thermal inactivation curves for u-galactosidase 13 and cfiflyacetylgalactosaminidase activities coincided. Isoelectric focus. ibng of a partially purified preparation of c-galactosidase B revealed that l>oth c-galactosidase B and ufigyacetylgalactosaminidase had isoelectric EDoints at pI - 4.5 (Figure 45). BjflrAcetylhexosaminidase A, contaminating tihis preparation, had pI - 5.0. To determine whether a single enzyme 111 .Amnmv mumwwno mumonmonoImumuoHo Huoaou nH uno 0mauumo mama mummmm may .%Hm>fiuomnmmu .muma Ionoounmu Enawom 28 m.m 0nm 28 n.m nufia H.0 mm on Hmaaueo mm3 umommmonu wnm umommmomo mo mammaouwhn may umummmoou 0nm umommmomu mo mammaouvmm mnu pom snaflueo me mnH .mm mnanm 112 -o- GoOseZCer -A- GbOseSCer I I l l l T I 0. 0. 0. 0. 0. o, 0 0 V to N — Iu/pazfllmpflq aiousqns 500qu O 5.0 4.0 3.0 113 .u£\mmaoan N08 m.ow hamumafixouaom mms munmamuommma auauoam> mnu nH uouum mnH .wa.nHa\mmaoan mo.H mo > unmumnmm nm 0nm EB mm.o mo BM unmumoam nm nuHB .Hmwaofiwfim 003 umommmomw mo mammaouwwn mnu now uoam xnnmlum>mm3mnHA may 4 mmmwfimouomHmUIs an mammaounmm umommmomu mo uoam xunmlum>mm3mnHA .0m madman 114 0.0N ..-:e. 0: 0.0. 0%: 0pm . O o 0.0 H I A o I0._ M r m. m w -0._ m. I0.N .mommmo0o 115 .o.H mo macaw 0 £003 mnflH m 030 .HmmnHH uon mm3 mammaouwhn umommmomo mo uoaa Hafiz mnH .4 mmmvfimouomamols ho mfimhaouvzm umonmomu 0o uoam Hafiz mew .mm munmfim 116 :25. 0.0. 0.0 phpb- - p .00.. 0._ 0.0 ppph-p P p — 0.030000 _.0 00. p-pF- - p b p .0. .00N00000 .0. I00. IIIII "ox: (600 117 .HB.Hn\mmHoan o.0H >Hmumaflxounnm mm3 munmamunmmma muwooam> mnu n0 uouum msa .we.nHE\0mHoE: 0H.H mo xm8> unmumanm nm wnm 25 mm.o 00 EM unmumnam nm £003 .Hmwfioawfim mm3 mamzaonwzn umommmonw mo moan xuoqum>mm3mnaq m£H 4 mmmwamouumHmOIo >0 mammaouwxm umommmooo mo uoam xunmlum>mm3mnHA .wm munwfim 118 r_ m é V é V ('V 0. o. o. 0. O O O O (pazKIOJpMI 180985009 anwU/Iw/Ju) A/l 20 l5 l0 (mM'I) I/S 119 .o.H 0c macam 0 £003 mnHH 0 0mumawxcuaa0 unn .u0mnHH ucn 003 000%H0u005 um0mm0000 0c moan HHH: mSH 0 ammenmouumomous an mamaaousam 000000000 no 0000 Hoax 0:0 .00 000000 120 3.03 co: 22.5 m 00.6 00... 00.0 9.0 nmd .0.0 b? b h D D? b b I D b - Co 1 n. 0.0 m o._ n. 0.0 rod. 0 mod... 3283... 30$...an w (A - "“‘M/A (owns 60.) 121 Figure 40. The Linearity of 4—MU—a-Gal Hydrolysis by a-Galactosidase B with Respect to Time and Enzyme Concentration The Hydrolysis of 4-MU-a-Gal by a—galactosidase B was linear with respect to time for up to 1 hour (top), and with respect to enzyme con- centration, up to 100 ug/lOO pl (bottom). Hydrolyzed nmoles 4-MU-d-Gol nmoles 4-MU-d-Gol Hydrolyzed/ min. 122 -« /O ”50.00 o IOO.04 50.04 0 I I I I I I I 1 O 30 60 90 I20 Time (min) 150.0 1 Ioo.o~ O 50.0« o * I I I 1 T 1 o 25 50 75 IOO |25 |50 Enzme (pg) 123 .m.q mm on Hmsfiumo mm3 mahuco mafia kn u2 oaouw>m m:u How msfiumo mm mze .Hq ouswfim 124 042.000-0210 a. 30-0-2570 ¢ 00.000 ..OdON [0.00m °ugw/pazA|01pKH amusqns salowu 125 .cHa\00Hoad o.~fi zaoumaaxouaam 003 muamBmuSmmma mufiooam> mnu 0H uouuo 039 .wa.aaa\moHan mn.wa was xma> ozu wow 28 w.o mma m mmmpamouomamwld ma Hmoldlpzwq mo mwmxaouwxn mnu How EM 0:8 0 ommwfimouomamwlc zn mamxaouwmm HmUIdIDZIq mo uoam xusmluo>mm3mcHA .Nq madman 126 0._ $2.5 0._ p w\_ :5 0.0 "Ex 30-05.20 .96 [9.0 /\/| (|_sa|owu-ugw) 127 .H.H mo macaw m zuHB .ummcHH mmB m mmmvfimouumamwla up mammaouvka amalalszlq mo uoam Hafiz 039 m ommwflmouomHmUIU mp mamxaouwzm Hmoudlszlq mo uoam Hafiz one .mq ouswwm 30: :25 28.3.6 .000. 0.0. on o._ 0.0 _.o 128 hbb b P npnhbb- b n b-nhbhb - b —O. BOIUIDSTv n 10.0. A_XDwA (50!) 129 .m.0 ma 00 comm 00 0:0 00.2.30 003 035930005" H0§0SH .ofiuo05n H0593. .l 00.3.2503. 0000H0w8000000H0wH>000usu coau0>wuo0dH H0Eu0£H .qq 0u=wfim 130 IO [fig/woo 20 30 4O 50 IO (minutes) time 131 emf—V " HQ 00: m 0000H0ouo0a00|0 .Ammmv wu00u0000> zn 000H00000 00 .uc0fiv0um hufi0cmw 000uon0 0 0H .mlm ma .00uka Ionaa0 HOHuu00 mag nua3 nasaoo wafimsoom 0HH000H000« Hoaw qu :0 0H 008uomu0m 003 wcwmaoom 0Hu000H000H m 0000H00000H00I0 mo coau0u0m0um 00HMHusm hHH0Huu0m 0 mo wcwmdoom ofiuuo0a0o0H .mq 0uawfim 132 2600836> 00. 0m 00 O? O 0.? . 20 no. . . 8 . 0 d . H + 0.? 0._L + .. ix Om: E 0.0% If 30260052-? .o. 60-6-3240 .0: 04.260-00-020 r0._ .O.N fi fi 00.0 . .00 m 0 . w 6.0. s 1m.O m. 6.0. m. u D. W .0.. 6.0.0. m p + / . w. .u .3 133 .GHB\00Hoac o.HH 3H0u0afixouna0 003 000080030002 %uaooa0> 0:0 Ga 00000 0£H .za 0.~ . 00 0003 .00z000uuumzuo 00 000000000 000>000000000 003 Hmoucupzuq 00 000000000: 0:0 .000000aaoo 0:0 .00 000000 134 ON 0._ 172,5 0._ m\_ 0 ¢_._ mNN find. d “2:; o 0:0 Ga 00000 0£H .wa.cfia\00aoan mm.mm 003 x0a> 0:0 0:0 :8 m.H 003 m 000vfimouo0a0wlc an 0Houwzn 030 you EM 058 m 0000H00000H0015 mp mammaouvmm u0030afiq .mq 0udwfim 137 $205 m\_ O.N n... 0._ 0.0 H 1.0.0 A 0.. .36 u. U 0 0. rvod 8 .25 0.. ...Ex s0 0 ( 042.005-02-00 rmod .H.H mo 000:0 0 :003 .000CHH 003 m 0000H00000H0wld %: 0H0>Houw%: o mnu CH uouum msH .wa.cfi8\mmaoan n.mam was me> mnu cam 25 m.qa mm3 ameldlxxomvlmlmZIm mo mfimhaouwms mnu How EM wnH m mmmwfimOuumHmotd An HmolclzxomolmlmZIQ mo mamxaouwmm mLu mo uoam xusmlnm>mm3mcHA .me muawfim 141 fgé m: ON w... 0._ 0.0 o .1 \ 10.. n A T m. nod M d w o my: . E u a 57:. NV. n X SF .nouUonmoumuazh rod 3. 142 .o.H mo macaw m :ufiB .ummCaH mmB mHm%Houvzn Hmwldlmxomvlmlmzlm mo uoaa Haw: m:& m mmmwflmouumamold x3 mflm%aouwmm HamldlzxomQINImZIa mo uoam Hafiz mLH .om muswfim 143 . 32. 22.5 28.33 0.09 0.3 0.0. 0.... 0.. ..o pr-IPP b D P DPPPP P P b P DDDP- b b .000 We 2.0 t no.0 "0.. O .nouuuaxooo-miz-q . o.» A-‘DwA (60:) 144 .c«8\mmaoas o.HH mamumafixouamm mm3 muamaqummma zuaooam> mnu ca uouum one .wa.cfia\mmaoan oa.o mma xma> mzu van 28 w.m mmB usmhmlalmZIo mo mwmhaouwhn mnu uom BM may m mmmwflmouumamuna %n USWMMIGIAZIO mo mHmzaouwzm mzu pom uoam xusmlum>mmamafiq .Hm muswfim 145 uvd ...22. m: . m... 0._ nd 0 \ 1.0 O I. / .. A v~.o ) M. U . u. m m. a .3 (L :2 mduev. . 3...-m-a-az-o 146 .o.H mo macaw m nufi3 .ummcaa mm3 mammaouvxc osmhmldlmZIo mo uoaa HHfim mfiH m mmmwfimouumHmUId kn mfimhaouw%m UDWMMIaImZIo mo uoam Hafiz mLH .Nm muswflm 147 32. :25 28.3.5 .8 od. 0.» 0.. no 3 hr hp hhhhbhF - h bh-bhbb b P _°o mac. n ".6 T mg "0._ uau-m-u-az-o H 0..» 148 .Hocfiuuo nufiz wmuaamsmfi> oum3 moumuvhzonumo 038 .mHoEa\Bmo Hmo.m mo zufi>fiuum afiwfiomam m was van .umccmum Ofivmm vaonuumm namuwouw< cmaum> a co @wccmom mms wwwumsuommowaao m>Huom0Humu may .mumaa umxmalcficu o How moHHHm m :0 wmcfimuno mmB mmomuammonoawmmma mo amumoumaounu ummmalafisu may mmomuGQQOLOHommmH mo amuwoumaouno H¢>MAIGH£H .mm muswfim 149 I ”mane ¢ r. 3.028% ‘ n 3080.. .-m g: 3.23.00 Anumrmmv .o.oiuu_o< utoodLocoaoiu: r..- 2m 150 .wmm: muma Ammmv mummwsn mumnamosaumumuufiu «uoaou .mam>fiuumammu .q.q mm cam m.q 39 um Hmsfiuao mm3 m mmmwfimouomammla mp mmmono cam mowonu mo mfimhaouuzn mnH m mmmvfimouumamold mp mmomuammopoao cam mmomfiuuonoao mo mfimxaouwmm mnu How meHuao ma use .qm madman 151 0,0 0.0 0d 04m - p — p b p o . .00 o d u 10.. . rm; @808 Av M IO.N $08 + + .06. 1OAK .Odn Jq/pazKIOJpKH amusqns sa|owu 152 mzu cw uouum 03H .u£\mmaoac m.ou %Houmaaxouaam mm3 muamamusmmma muaooaw> .we.cHE\mmHoan «N.m mmB xma> mnu van :8 n.m mma mmmono mo mwmzaouvmc mnu you BM use m mmmwfimouomamUId mp mmmono mo mfimzaouwmm mnu wow uoam Musmlum>mw3maHA .mm Guzman 153 O.m 1-2:: m) @808 /\/| (I_sa|ouJu.Jq) 154 .o.H mo macaw m SuwB .ummcfia mm3 mfimhaouwxz momOpo mo uoaa HHHm mnH m wmmvfimOuomamuld mp mammaouvkm mmmonu mo uoam HHHm mna .om muswfim 155 I0.0 (60:) A - P p p- b y- y— b h p— p- : b p 1111!! I 1 IHHII I "”111 l I Q 0 “2 -: I0 - ID - o O Q Q A_XDLUA 50.0 I0.0 5.0 I.O 0.5 O.l mm vs (log) 156 .un\mmaoa: m.ow mawumafixouuam mmB mucmamuammma muauoam> ms» ca Houuo 05H .ma.cHB\moHoE: no.¢ was me> mnu cam 28 H.m mm3 m mmmvwmouomawwld ha mmmono mo mwmzaouvkc map How BM mnH m mmmwfimouomamunc >9 mwmhaouwxm mmwono mo uoam xuamlum>mm3mcHa .nm munwwm 157 A/l (l_sa|owu.1q) 158 .o.H mo macaw m zufl3 .ummcHH mmz m mmmvwmouumamwld >3 mfimxaouvms mmmoaw mo uoaa Hafiz mnH m ammufimouomfimouu xn mflmmaouwxm mmmOQU mo uoam Hafiz may .wm muswflm 159 0. 9 A I L- m I m (D F O - .0 L9 .. unit] I IIWITII I Q Q In ' -: ID — o’ 0 A_xow (500 A 50.0 5.0 10.0 (mM) LO 0.5 O.I Substrate (log) 160 Vmax was 4.07 umoles/min-mg. The Hill plot of GbOse3 hydrolysis was lin- _ ear, with a slape of 1.0 (Figure 59). Glycolipid Substrates. The thin-layer chromatogram of [3H]GbOseSCer is shown in Figure 59. The specific activity of the [3H]-labe1ed glycolipid was 12,177 cpm/nmole. The hydrolysis of this glycolipid was linear with respect to time (up to 4 hours) and enzyme concentration (up to 100 ug with 20 minute incubations). A thin-layer chromatogram.of GbOseSCer hydrolysis by u-galactosidase B is shown in Figure 60. The hydrolysis of this glycolipid was optimal at pH 3.9 (Figure 61) with 7.4 mM sodium tauro- cholate (Figure 62). The Lineweaver-Burk plot of GbOse5Cer hydrolysis was sigmoidal, resembling the plot of GbOse3Cer hydrolysis by o-galactosidase A. The Km with GbOseSCer was 0.59 mM (Figure 63) and the Vmax was 0.27 umoles/min-mg. The Hill plot of this activity also deviated from linear- ity (Figure 64), but approximated a line with a slope of 1.0. a-Galactosidase B also catalyzed the hydrolysis of GbOse3Cer, however prolonged incubation periods were required to detect activity with this sub- strate. The hydrolysis of GbOse3Cer was linear with respect to time (up to 10 hours) and enzyme concentration (up to 180 ug). The hydrolysis of this substrate was optimal at pH 4.3 (Figure 61) with 9.3 mM sodium tauro- cholate (Figure 62). Unlike the Lineweaver-Burk plots of the hydrolysis of other glycolipids by o-galactosidases A and B, the Lineweaver-Burk plot of GbOse3Cer hydrolysis was linear (Figure 65). The Km with GbOse3Cer was 0.35 mM and the Vmax was 0.18 umoles/min-mg. The Hill plot was linear, with a slope of 1.1 (Figure 66). The enzymatic properties of o-galactosidases A and B are summarized in Table 2. III. Characterization of a-Galactosidases A and B The molecular weights of a-galactosidases A and B were estimated by gel filtration on Sephadex G-150, as shown in Figure 67. The molecular weight of o-galactosidase A.was approximately 104,000 daltons and o-galac- tosidase B (ofiflyacetylgalactosaminidase) was 90,000 daltons. The isoelec- tric point of o-galactosidase A.was 4.7 (Figure 68), and a-galactosidase B was 4.5, as previously shown (Figure 45). 161 .Hocfiouo mugs vafiHmamH> mum3 mafiafiaoomaw may .mHosa\aau nma.~a mo %ufi>fiuom owwaommm m mm: mam .umccmum oavmm vaonuumm namumoumd cmwum> m zuaa waficamom an wmumu IOH mm3 vfimfiaouxam m>Huomoavmu mnH .mummcmum m mm mom: mums mvamfiaoo%aw Hmuusma HmcfiummuCfi mafiamo .mumHa umhmalaanu 0 How moHHHm m do mocamuno mm3 mwfismumoammoomammucmaonoawHmmg mo Smumoumaouso umxmalcfinu 05H meEmumua%moo%HwmuammonoauHmmH mo SmuwoumEouno umhmAICana .am muawfim 162 I_:oto -..om.moQo -..om..0ne Icoooom l m o _ 0.9 icene no. 3-2-0 .8388. (am 163 Figure 60. Thin-Layer Chromatogram of GbOseSCer Hydrolysis by a-Galacto- sidase B The hydrolysis of GbOseSCer by a-galactosidase B was examined by thin- layer chromatography. Canine intestinal neutral glycolipids were used as a standard. The standard contained glucosylceramide (GL-l), lactosylcer- amide (CL-2), globotriglycosylceramide (CL-3), globotetraglycosylceramide (CL-4) and globopentaglycosylceramide (GL—S). 164 det. blank assay assay std. C=M=W 65=45=|O taurocholate- 165 .mummmsa oumuuao aafiwom 28 on mam mumaonoouamu Edwvom SE «.5 wmsfimucou umommmono zuH3 mmusuxfia cOHuummu may .Ammmv mummwan mumnmmosalmumuufio «90800 mam mumaosuouamu sawcom ZS m.m wmcfimucoo umommmono nuws mmuauxfia coauommu use .mam>auomammu .a.m ma mam m.q mm mama “mommmonu was umommmono mo mfimhaouv>£ mnu pom mafiuao mm 0&8 m ammuamouomamuus an “mommmono new umommmosu mo mamaaoueam was “om assume as may .Ho magmas 166 0d 06 06 0d _ . _ . . o ‘ O .06. /. 1CON 0 emu mwOQO IO: ..O.Om chMmmOnw n1. 167 .m.m mm .mumuufio Bafiwom :8 on wmafimuaoo umUmmmono sufis mauauxfia dofiuommu any .Amnav m.q ma .ummwan mumnamocalmumuuao Huoaoo vmcfimucoo “mommmoao nuaB mmuauxaa coauommu any .%Hm>wuommmmu m .mumaofiuouamu aafivom 28 «.5 was 28 m.m Lufi3 Hmaaumo mums umu mwono mam umommmono mo mfimzaoummz mna m mmmvwmouomamuld %n umommmono cam umommmonu mo mHm%H0pw%: man now mafiuao mumaoaooude aawvom age .No muawam 168 $595 3205030... 0.N_ 0.0. 0 m 0.0 0.? ON 0 L .8388 -0.0N 10.0? w000 10.00 [0.00. sgsAIOJpAH |ouugxow 0/0 169 .u£\mmaofia m.ow mamumaw Ixoumnm was muamsmHSmmma muwooam> mnu ca uouum any .wa.afia\mmaoaa nm.o m0 me> mamumaam cm mam Ea mm.o 00 EM uamummmm am nufi3 .Hmmfioawwm mm3 m mmmwflmouomamwlu kn mfimhaouwmn “mommmonu mo uoam xuamlum>mmBmSHA 05H 0r m mmmwfimouomHmUId kn umummmono mo mfimzaouvhm mam mo uoam xuamlum>mm3mcHA .mo muswwm 170 0.0. 1.2.5 3. 0.N_ 0.0 06 0 ...0 | / A .Nd ) N m. ,3 w 0 m. S. .22 and"; .....o (I .mommmofi 171 .o.H mo macaw m Sufi3 mafia m vmumaflxoumam man .ummcHH uoa mmB mfimzaouwmc umUmmmono mo moan Hafiz mnH m mmmwwmouumHmOIo >3 mammaouvmm “mommmoau mo uoam Hafiz mLH .qo madman 172 30: 22:: maozmnam 0.0. 0.0 0._ m0 _.0 n0. .0. app» p . app-p.- _ a ppppappp _ _o. T .00. n ...A n_.o m A . I A .n.0 U n 0 . .m. n0._ W .mommmog ”on 0.0. 173 3.0 9. 3X“ .u:\mmaoaa n.0H hamumafixoummm mma muamsmuammma mufiooam> msu ca uouum any. .wa.aH8\mmHoE: a> mnu man 28 mm.o mm3 m ammwfimouomammls mp umommmonu wo mamwaouvmn mnu wow SM one m mmmmamouomamolo >9 mflmmaouwmm umommmono mo uoam xuamlum>mm3mafia .mo madman 174 1.2.5 m: 90. ow om 0.... . o..~ . o .2. n A M 0 r0._ “mu 0 m ‘I s. .2... and "Ex 6: .8988 n r .— 175 .H.H wo macaw m :uHB .ummafla mm3 mwmzaouvhs umommmono mo uoam HHflm m£H m mmmwfimouomamond kn mmeaoumxm umommmono mo moam HHHm mna .oo muawfim .Ii .176 loo; 22.5 Entansm 06. ed 0._ no 3. no. 5. pun—hp - - ppppp—P _ L b-nppr b r —.o O .oommmoQo 0.0m 177 Table 2. Summary of the Enzymatic Properties of o-Galactosidases A and B a-Galactosidase A a-Galactosidase B PH Km pH substrate optimum. (mM) vgax optimum (mM) vgax GaOseZCer 4.1 0.28 1.09 ND1 ND ND GbOse3Cer 4.1 0.18 1.16 4.3 0.35 0.18 GbOseSCer NH? NH2 NH2 3.9 0.59 0.27 GaOse2 4.5 5.5 15.76 ND ND ND GbOse3 4.5 4.1 13.76 4.8 9.1 4.07 GbOseS NH3 NH3 NH3 4.4 3.7 9.24 4-MU-a-Gal 4.6 2.9 28.68 4.7 6.8 18.73 o-NP-a-GalNAc NH“ NH” NH“ 4.3 1.3 59.95 p-NP-Z-deoxy-a-Gal 4.6 8.2 37.40 4.7 14.7 215.70 o-NP-a-QrFuc NH5 NH5 NH5 4.7 8.8 0.16 * umoles/min-mg 1 not determined with 2.0 mM GbOseSCer after 3 hours at 37°C with 5.0 mM GbOse5 after 2 hours at 37°C with 10.0 mM o-NP-a-GalNAc after 1 hour at 37°C 2 hydrolysis not detected 3 hydrolysis not detected ” hydrolysis not detected detected with 10.0 mM o-NP-a-QrFuc after 1 hour at 37°C 5 hydrolysis not 178 .Aaadaoo Eu OHH x o.mV omalo wamsmmm do coaumuuafiw Haw hp vmumEHumm mums m mam < mommwfimouomamwlo mo munwfim3 umaaomaoa msH coaumuuafim Hmo %n m was < mammvfimouomHmUIa mo munwfimz umaaomaoz mnu mo cowumeflumm .mo muswfim 179 2062, 836222 mo... 0.0 0.0 0.0 06 0.v F . _ r _ _ p 0 32200 1.0 32024 mo one I < smog . _ o .0 .N0 m amou_mo.oo_oou.o .00 1V0 4 5005338620 .m.0 0 2.020010 0 .00 ADM 180 ONOQ u Hg was < ammvamouumamoud .Amfimv wumnumumm> >3 vmnfiuommm mm .mlm mm .mmumaoxmam umfiuumo mag spas neaaou wcfimaoom oauuumamoma Hoam mMA cm ca mmauomumm mm3 < mmmvflmouumamwla mmfimwuam mo wsfimaoow uauuomamomH a mmmwamouomamols wowwfiuam mo wcwmaoom oauuomamomH .wo muawwm 181 ..mDEDZ 00:00..“— Om. ON. 0: OO. 00 cm ON 00 . p a p P p r O a d O 4 4 4 w . 1 a IO.ON o m C d O 4. IOI 4 a o S A . .09. ..v o q o w . , ... ORV: o 10.00 m . 9 m. 1 o #000 H (A m. . mu. 0 m L r000. M a p / 1 T0.0N_ w w. 0.0 -<>- DISCUSSION The salient results of the investigations described in this thesis are: (i) the purification of a-galactosidases A and B to higher specific activities than previously reported, (ii) the characterization of the en- zymatic properties of a-galactosidase A with several artificial and natural substrates, and the effect of nonionic and anionic detergents on the hy- drolysis of these substrates, (iii) the identification of a-galactosidase B as an afigfacetylgalactosaminidase, and the characterization of the en- zymatic properties of this enzyme with several artificial and natural sub- strates and (iv) the preparation of two novel artificial substrate analogs, p-nitrcphenyl-Z-deoxy-o-gfgalactopyranoside and o-nitrophenyl-6-deoxy—ae27 galactopyranoside (o-nitrophenyl-a-Q;fucopyranoside), and the use of these substrate analogs to investigate the carbohydrate-binding specificities of a-galactosidases A and B. These findings are discussed in greater detail below. 1. Purification of a-Galactosidases A and B The purification of aegalactosidases A and B described here achieved higher specific activities with 4-MU-a-Gal than values previously reported by Romeo et al. (211) (2,867 nmoles/min-mg and 246 nmoles/min-mg for a-ga- lactosidases A and B, respectively), Hayes and Beutler (222) (8,500 nmoles/min'mg for a-galactosidase A) and Kusiak et al. (223) (4,683 nmoles/min-mg and 4,517 nmoles/min-mg for a-galactosidases A and B, respec- tively). The success of this purification is primarily due to the use of ampholyte displacement chromatography. This procedure was first described by Leaback and Robinson (82) for the purification of isozymes of Bfigracetyl- hexosaminidase B. These workers reported that ampholyte displacement chrom- atography gave greater resolution of Bfifiracetylhexosaminidase B isozymes than did isoelectric focusing, and could be carried out in only a few hours, in contrast with the periods of several days often required for isoelectric focusing. Hayes and Beutler (222) and Kusiak et al. (223) employed Concan- avalin A Sepharose chromatography in their purifications of a—galactosidases. This step was not included in the purification scheme described here, 182 183 because of the poor recoveries of glycosidases from Con-A sepharose that were reported when this support first became available (77). However, the recoveries of glycosidases from.Con-A sepharose chromatography that were reported by Mayes and Beutler (222) and Kusiak et al. (223) are acceptable, and this may be a useful step to employ in future purifications of these enzymes. Octyl-Sepharose chromatography (39) may also be a useful step to employ in future purifications of these enzymes. II. Characterization of the Enzymatic Properties of a-Galactosidases A and B A. a-Galactosidase A u-Galactosidase A catalyzed the hydrolysis of several artificial and natural substrates, including 4-MU-a-Gal, GaOseZCer, CbOse3Cer, GaOsez, GbOse3 and p-NP—Z-deoxy-a-Gal. The glycolipids were not freely soluble in the aqueous reactions mixtures, and the anionic detergent sodium taurocholate was used to solubilize them as glycolipid-taurocholate mixed micelles (283). The hydrolysis of the water-soluble artificial substrate 44MU-a-Ga1 was optimal at pH 4.6, however the pH optimum for the hydrolysis of the glyco- lipid substrates was lower, at pH 4.1. Furthermore, the Michaelis constants and maximal velocities with these substrates were approximately one order of magnitude smaller than those observed with 44MU-o—Gal (Table 2). The differences in the pH optima, Michaelis constants and maximal vel- ocities with 44MU-a-Gal and the glycolipid substrates were investigated. The hydrolysis of the water-soluble oligosaccharides derived from GaOseZCer and GbOse3Cer by ozonolysis and treatment with base was examined, and the hydrolysis of 4éMU-a-Gal was examined in the presence of sodium taurocholate and Triton X-100. The hydrolysis of GaOse2 and GbOse3 was optimal at pH 4.5, corresponding with the pH optimum for the hydrolysis of 4+MU-a-Ga1. Sim- ilarly, the Michaelis constants with these oligosaccharides approximated the Michaelis constant with 4-MU-a-Gal, however the maximal velocities with GaOsez and GbOse3 were approximately 50% of that observed with 44MU-a-Gal. This difference probably reflects the relative lability of the glycosidic bond of the artificial substrate due to the aromatic group it is linked to, rather than a difference in the catalytic ability of the enzyme with these different substrates. The differences in pH optima and kinetics with water- soluble substrates and glycolipid substrates, therefore, did not appear to 184 be related to any differences in the interaction of the enzyme with the carbohydrate portion of these molecules. Because the differences in the pH optima and kinetics of the hydrolysis of water-soluble and glycolipid substrates did not appear to be related to the carbohydrate portion of the molecules, the effect of detergents on the hydrolysis of 44MU-a—Gal was investigated. The nonionic detergent Triton X-100 had no effect on the pH optimum for the hydrolysis of 4-MU-a-Ga1. However, sodium taurocholate, at the same concentration used to solubilize the glycolipid substrates, had a profound effect on both the pH optimum and kinetics of 4AMU-a-Gal hydrolysis. Sodium taurocholate inhibited the hydrolysis of 4AMU-a-Gal below pH 4.8 and produced hyperbolic Lineweaver-Burk plots for the hydrolysis of this substrate below pH 4.8. The observation that the hyperbolic character of the Lineweaver-Burk plots diminished as the pH increased, suggests that the effect of sodium tauro- cholate on the hydrolysis of 4-MU-a-Ga1 may be due to an electrostatic in- teraction between sodium taurocholate and a-galactosidase A. This hypoth- esis is supported by the finding that o—galactosidase A has pI - 4.7. Be- low pH 4.7 the enzyme would hear a positive charge, taurocholate would be negatively charged (pK - 1.4 (284)), and there would be a strong electro- static interaction. Above pH 4.7 the enzyme and the detergent would both bear negative charges and their interaction would be minimal. In view of the fact that 4éMU-a-Gal is water-soluble, its hydrolysis by a-galactosidase A is not dependent on the interaction of the enzyme with taurocholate-substrate mixed micelles. Indeed, the hydrolysis of 44MU-a-Ga1 is inhibited by sodium taurocholate at low pH. In contrast, the hydrolysis of glycolipids should be dependent on the interaction of o-galactosidase A with glycolipid-taurocholate mixed micelles. This hypothesis is supported by the lower pH optimum observed with the glyco- lipid substrates, and the lower apparent Km, which probably reflects the strong electrostatic interaction between the taurocholate micelles and the enzyme, rather than the affinity of a—galactosidase A for the substrate. The lower maximal velocity observed with the glycolipid substrates may be due to allosteric interactions with the detergent, or the low pH of the reaction mixture. The Lineweaver—Burk plots with the water-soluble substrates (44MU-a-Ga1, GaOse2 and GbOse3) were linear, conforming to typical Michaelis-Menton kinetics. The Hill plots with these substrates were also 185 linear, with slopes of ~1.0, suggesting that there is no cooperativity in the binding of the substrate with the enzyme. In contrast, the Lineweaver- Burk plots with GaOseZCer and GbOse3Cer were sigmoidal, as described by Gatt et al. (283) for enzymes that utilize micellar lipid substrates. The Hill plots with these glycolipids also deviated from linearity, suggesting that the detergent had an allosteric effect on the interaction of the en- zyme with the substrate (285). Fung and Sweeley (286) have suggested that monomeric, rather than micellar taurocholate may be involved in these a1- 1osteric interactions. Although it may be argued that the use of detergents to solubilize glycolipids in aqueous solutions is quite an unnatural environment for the enzyme to encounter its substrate, it should be pointed out that, in vivo, glycolipids are components of membranes which may be negatively charged, and electrostatic interactions between the enzymes of glycolipid metab- olism and membranes may play an important role in bringing the enzyme and substrate together. Li and coworkers (212-214) have isolated a glycoprotein which they describe as an 'activator' for glycolipid hydrolases. These investigators use this 'activator' protein to replace detergents in their reaction mix- tures containing glycolipids. However, this protein is nonspecific, it stimulates the hydrolysis of both neutral glycolipids and gangliosides by their respective hydrolases, and appears to act as a detergent. Fur- thermore, these investigators have not demonstrated the presence of this 'activator' protein in the lysosome, the predominant site of glycolipid catabolism. An interesting hypothesis is that this 'activator' protein may not be involved in glycolipid catabolism at all, but may serve as a carrier protein for glycolipids during glycolipid biosynthesis. As dis- cussed in the first section of this thesis, the biosynthesis of some gly- colipids and glycoproteins, such as the blood group-active glycolipids and glycoproteins, may be carried out by a common pathway. This 'acti- vator' protein may solubilize the glycolipids to enable then to interact with the same glycosyltransferases that carry out the biosynthesis of gly- coproteins. This hypothesis has not been investigated, but poses an in- teresting alternative to the proposed function of the 'activator' protein. The hydrolysis of IVu-a-Gal-Lcn03e4Cer and blood group B-active gly- colipids and glycoproteins was not examined, but a—galactosidase A is pro- bably responsible for the hydrolysis of these glycoconjugates in vivo. 186 The blood group B-active glycolipids accumulate in Fabry patients with this phenotype and blood group B-active glycopeptides and oligosaccha- rides would also be expected to accumulate in these patients. The ac- cumulation of IVu-a-Gal-LcnOse4Cer in Fabry patients with the P1 pheno- type has not been reported, but this glycolipid would also be expected to accumulate in these patients. B. a-Galactosidase B a—Galactosidase B and ajyfacetylgalactosaminidase activities co-chro- matographed throughout the purification procedure. Furthermore, these hydrolase activities had coincident thermal inactivation curves and iso- electric points, and the hydrolysis of 4+MU-o-Ga1 was competitievly in- hibited by o-NP-a-GalNAc. The Km for the hydrolysis of the latter was approximately 5 times lower than the Km with 44MU-u-Gal and a Vmax that was approximately 3 times greater than that with 4+MU-a-Gal. These findings suggest that a-galactosidase B is, in fact, an avid afififacetyl- galactosaminidase. This enzyme also catalyzed the hydrolysis of GbOse3Cer, GbOseSCer, GbOse3 and GbOseS. The Michaelis constants and maximal velocities with the glycolipids were quite similar, however this was undoubtedly due to the presence of sodium taurocholate in the reaction mixtures. The Michaelis constants with the oligosaccharides derived from CbOse3Cer and GbOseSCer approximated those observed with 4éMU-a-Ga1 and o-NP-a-GalNAc, respectively. The maximal velocities with these oligosaccharides were approximately 5 times lower than those observed with the artificial sub- strates. However, this difference probably reflects the relative lability of the glycosidic bond of the artificial substrates due to the aromatic groups that they are linked to. The predominant physiological activity of o-galactosidase B is some- what speculative, but glycolipids containing terminal a-galactose residues accumulate in Fabry's disease in spite of normal or even somewhat elevated levels of a-galactosidase B in these patients. It appears, therefore, that a-galactosidase B is more likely to function as an ejyfacetylgalactosamini- dase rather than an o-galactosidase in viva. It would therefore be more accurate to refer to it as an afigyacetylgalactosaminidase (EC 3.2.1.49). Globopentag1ycosylceramide (Forssman antigen) and blood group A-active glycolipids and glchproteins contain terminal ofiflfacetylgalactosamine 187 residues and are likely substrates for this enzyme. These findings have been confirmed by Schram et al. (238), who found that antibodies prepared against a-galactosidase B precipitated both a-ga- lactosidase B and ujgracetylgalactosaminidase activities from a crude u—galactosidase B preparation from normal human liver. In addition, this antibody precipitated a-galactosidase activity from a crude a-galactosidase B preparation from Fabry liver (215). Purified porcine liver ojgfacetyl- galactosaminidase also catalyzed the hydrolysis of 4AMU-a-Ga1 with a Km that was 5 times greater than that with p-NP-a-GalNAc and a Vmax that was more than 4 times lower than that with p-NP-u-GalNAc (60). C. The Carbohydrate Binding Specificities of o-Galactosidases A and B Several of the lysosomal glycosidases lack absolute specificity for the carbohydrate residues that they hydrolyze. Chester et al. (287) re- ported that human liver B-glucosidase also catalyzed the hydrolysis of 4-MU-B-Ga1, p-NP-B-QrFuc, 44MU-B-Xyl and 4-MU-a-grAra. This suggests that B-glucosidase lacks absolute specificity for the substituent at C-5, and the conformation of the hydroxyl group (axial or equatorial) at C-4. However, all of these substrates had equatorial hydroxyl groups at C-2 and C-3, and these groups may be important in the binding of the substrate at the active site, or the mechanism of the enzymatic hydrolysis of the glycosidic bond. Wallenfels and Neil (288) have suggested that the equatorial hydroxyl group at C-2 may play a role in the mechanism of the hydrolysis catalyzed by B-galactosidase. These investigators suggested that the developing carbonium ion at C-1 may be stabilized by the forma- tion of a 1,2-epoxide, which is subsequently hydrolyzed to yield free galactose as the product. The Bfigfacetylhexosaminidases hydrolyze the 4~methylumbelliferyl derivatives of both Bfiflfacetylgalactosamine and nyfacetylglucosamine (56,57), demonstrating a lack of absolute specificity for the conformation of the hydroxyl group at C-4, as shown in Figure 69. a-Galactosidase A catalyzed the hydrolysis of 44MU-a—Gal and o-NP-Z-deoxy-a-Gal, but not p-NP-a-Glc, o—NP-a-GalNAc or o-NP-a-QrFuc. Furthermore, the hydrolysis of 4+MU-c-Ga1 was not competitively inhibited by o-NP-a-GalNAc, p-NP-a-Glc, 27(+)-fucose or Qrgalactal. These findings suggest that the hydroxyl group at C-2 does not play a significant role in the binding of the carbohydrate at the active site of the enzyme, or 188 Figure 69. Three—Dimensional Projections of Carbohydrate Residues Hydrol- yzed by the 87§7Acetylhexosaminidases The Bfiflfacetylhexosaminidases catalyze the hydrolysis of both 3- ‘Efacetylglucosamine (I) and Bjfifacetylgalactosamine (II) residues. This suggests a lack of absolute specificity for the configuration of the hydroxyl group at C-4 (shaded). 189 190 in the mechanism of the enzymatic hydrolysis of these substrates (Figure 70). However, replacement of the hydroxyl group at C-2 with an acetamido group results in a loss of activity. This may be due to steric hindrance in binding the 2-acetamido-2-deoxy-substrate at the active site of the enzyme. The failure of e-galactosidase A to hydrolyze p-NP-a-Glc and o-NP-a-QrFuc, and the failure of 44MU-a-Gal hydrolysis to be competitively inhibited by p-NP-aGlc and 27(+)-fucose, suggest that the axial hydroxyl group at C-4 and the hydroxyl group at C-6 may be important in binding the carbohydrate at the active site of the enzyme. These findings, together with the failure of grgalactal to inhibit the hydrolysis of 4-MU-a—Gal, suggest that the mechanism of the enzymatic hydrolysis of glycosides by o-galactosidase A.may proceed through a different mechanism than that pro- posed by Wallenfels and Weil (288) for B-galactosidase. a-Galactosidase B (aggracetylgalactosaminidase) catalyzed the hydrol- ysis of 44MU-a-Gal, o-NP-a-GalNAc, p-NP-Z-deoxy-a-Gal and o-NP-u-QrFuc. These findings suggest that his enzyme lacks absolute specificity for the substituents at C-2 and C-6, as shown in Figure 71. Unlike a-galactosidase A, the B form was also an a~§7acetylgalactosaminidase, although the lower pH optimum for the hydrolysis of o-NP-o-GalNAc suggests that a change in the conformation of the enzyme may be required to accommodate the bulky acetamido substituent. In addition, it has been reported that dfigfacetyl- galactosaminidase also catalyzes the hydrolysis of afflfacetyltalosamine residues (60), as shown in Figure 71. The preparation of artificial substrate analogs that are missing specific substituents can provide insight into the carbohydrate-binding specificities and possibly the mechanism of the reactions catalyzed by the enzymes of glycoconjugate metabolism. The use of deoxy or halogenated carbohydrates may also provide insight into the carbohydrate-binding speci- ficities of glycosyltransferases, where they could serve as dead-end carbo- hydrate acceptors. Therefore, these compounds might serve as inhibitors of specific steps in glycoconjugate biosynthesis or catabolism. III. Characterization of a-Galactosidases A and B The molecular weight of o-galactosidase A from.human liver was approximately 104,000 daltons, in agreement with molecular weight esti- mates for a—galactosidase A by Mapes et al. (231) and Kusiak et al. (223). 191 Figure 70. Three-Dimensional Projection of the Carbohydrate Residue Hydrolyzed by a-Galactosidase A a-Galactosidase A catalyzes the hydrolysis of o-galactose and 2-deoxy-a-galactose residues, suggesting a lack of absolute speci- ficity for the hydroxyl group at C-2 (shaded). 193 Figure 71. Three Dimensional Projections of Carbohydrates Hydrol- yzed by a-Galactosidase B a-Galactosidase B catalyzes the hydrolysis of offifacetylgalactos- amine residues (I), ofEfacetyltalosamine residues (II), a-galactose residues (III) and u-Qrfucose residues (not shown), suggesting a lack of absolute specificity for the substituents at C-2 and C-6 (shaded). 194 “......u—I 195 Kusiak et al. (223) found that a-galactosidase A was a dimer, with sub- units of 57,700 daltons. The molecular weight of the B form from human liver was 90,000 daltons, in agreement with the molecular weight deter- mined by Kusiak et al. (223). These workers found that the B form was a dimer with subunits of 47,700 daltons. The isoelectric points of a-galactosidases A and B were at pH 4.7 and pH 4.5, respectively, in agreement with values reported by Beutler and Kuhl (219). Bishop and Sweeley (39) reported that a-galactosidase A from normal human plasma had pI - 3.7. Treatment of this enzyme with C. perfringens neuraminidase shifted the isoelectric point to 4.3. These findings suggest that a-galactosidase A from plasma may be sialylated, with oligosaccharide substituents similar to those shown in Figure 1. However the tissue a-galactosidases, which were not susceptible to neur- aminidase treatment, may a different type of oligosaccharide substituent. Sung (289) found that porcine liver affifacetylgalactosaminidase did not contain sialic acid, but appeared to have an oligomannose type.of oligo- saccharide. This finding suggests that the metabolism of tissue forms of the glycosidases may be quite different from the metabolism of secreted forms of these enzymes, such as those found in plasma. This hypothesis is supported by studies by Desnick et al. (290), who found that plasma a-galactosidase A had a much longer survival in circulation than splenic u-galactosidase A, in Fabry patients receiving these enzymes. As dis- cussed in the first section of this thesis, the presence of terminal sialic acid residues on the carbohydrate moiety of glycoproteins prolongs their survival in circulation. However, removal of the sialic acid res- idues greatly reduces the survival of these glycoproteins in circulation. Henderson et al. (291) have reported the presence of a mannose—binding protein in livers of rabbits and rats. This receptor could be responsible for the rapid clearance of tissue forms of a—galactosidase A from circu- lation. BIBLIOGRAPHY 10. ll. 12. 13. 14. BIBLIOGRAPHY Lipids and Lipidoses, (G Schettler, Ed), Springer-Verlag (New York, 1967). Lysosomes and Storage Diseases, (HG Hers and F van Hoof, Eds), Academic Press (New York, 1973). The Metabolic Basis of Inherited Disease,_4th Ed., (JB Stanbury, JB Wyngaarden and DS Fredrickson, Eds), McGraw-Hill (New York, 1978). Practical Enzymglggy of the Sphingolipidoses, (RH Glew and SP Peters, Eds), Alan R. Liss (New York, 1977). Sweeley, CC and Siddiqui, B: Chemistry of Mammalian Glycolipids. In The Glycoconjugates: Volume I, Mammalian Glycoproteins and Glyco- lipids, (MI Horowitz and W Pigman, Eds), Academic Press (New York, 1977), p 459. Kornfeld, R and Kornfeld, S: Comparative Aspects of Glycoprotein structure. Ann. Rev. Biochem. 42, 217 (1976). Waechter, CJ and Lennarz, WJ: The Role of Polyprenol-Linked Sugars in Glycoprotein Synthesis. Ann. Rev. Biochem. 42, 95 (1976). Schacter, H, Marasimham, S and Wilson, JE: Biosynthesis and Catabolism of Glycoproteins. In GlycoProteins and Glycolipids in Disease Pro- cesses, ACS Symposium Series (Washington, 1978), (in press). Schacter, H and Roseman, S: Mammalian Glycosyltransferases: Their Role in the Synthesis and Function of Glycoproteins. In Biochemistny of Proteoglycans and Glycoproteins, (WJ Lennarz, Ed), (in press). Roseman, S: The Synthesis of Complex Carbohydrates by Multiglycosyl- transferase Systems and Their Potential Function in Intercellular Adhesion. Chem. Phys. Lipids 2, 270 (1970). Sweeley, CC, Fung, Y-K, Macher, BA, Moskal, JR and Nunez, HA: Structure and Metabolism of Glycolipids. In Glycoproteins and Glyco- lipids in Disease Processes, ACS Symposium Series (Washington, 1978), (in press). Macher, BA and Sweeley, CC: A Survey of Glycosphingolipids; Struc- ture, Biological Source and Properties. In Methods in Enzymology, Vol. L (Part C): Complex Carbohydrates, (V Ginsberg, Ed), Academic Press (New York, 1978) p 236. Glycoproteins: Their Composition, Structure and Function, Part A, Zed Ed., (A Gottschalk, Ed), Elsevier (New York, 1972). Glycoproteins: Their Composition, Structure and Function, Part B, 196 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 197 Zed Ed., (A Gottschalk, Ed), Elsevier (New York, 1972). The Glycoconjugates: Volume I,,Mammalian Glycoproteins and Glyco- lipids, (MI Horowitz and W Pigman, Eds), Academic Press (New York, 1977). Pigman, W: General Aspects. In The Glycoconjugates: Volume I, Mammalian Glycoproteins and Glycolipids, (MI Horowitz and W Pigman, Eds), Academic Press (New York, 1977) p l. Cleland, WW and Kennedy, EP: The Enzymatic Synthesis of Psychosine. J. Biol. Chem. 235, 45 (1960). Hildebrand, J, Stoffyn, P and Hauser, G: Biosynthesis of Lactosyl- ceramide by Rat Brain Preparations and Comparisons with the Formation of Ganglioside GMl and Psychosine During Developement. J. Neurochem. 11, 403 (1970). Curtino, JA and Caputto, R: Enzymatic Synthesis of Glucosylsphingosine by Rat Brain Microsomes. Lipids Z, 525 (1972). Max, SR, Maclaren, NK, Brady, R0, Bradley, RM, Rennels, MB, Tanaka, J, Garcia, JH and Cornblath, M: GMB (Hematoside) Sphingolipodys- trophy. New Engl. J. Med. 291, 929 (1974). Maclaren, NK, Max, SR, Cornblath, M, Brady, R0, Ozand, PT, Campbell, J, Rennels, M, Mergner, WJ and Garcia, JH: GMB Gangliosidosis: A Novel Human Sphingolipodystrophy. Pediatrics 22, 106 (1976). Young, JD, Tsuchiya, D, Sandlin, DE and Holroyde, M: Synthetic Sub- strate for UDPfiflfAcetylgalactosamine: Mucin Transferase. Fed. Proc. 32, 1684 (1978). Kang, MS, Spencer, J and Elbein, AD: Glycoprotein Synthesis in the Aorta. Fed. Proc. 31, 1685 (1978). Hart, GW, Grant, GA and Bradshaw, RA: Primary Structural Requirements for the Glycosylation of Proteins. Fed. Proc. 21, 1684 (1978). Spiro, MJ, Spiro, RG and Bhoyroo, VD: Utilization of Oligosaccharide- Lipids in Glycoprotein Biosynthesis by Thyroid Enzyme. Fed. Proc. 31, 1684 (1978). Chen, WW: The Role of Glucose-Containing Oligosaccharide-Lipid in Glycosylation of Protein Acceptors. Fed. Proc. 32, 1685 (1978). Spiro, RG, Spiro, MJ and Bhoyroo, VD: Processing of Carbohydrate Units of Glycoproteins: Action of a Thyroid Glucosidase. Fed. Proc. 31, 1684 (1978). Hubbard, SC, Turco, SJ, Liu, T, Sultzman, L and Robbins, PW: Process- ing of Asparagine-Linked Oligosaccharides In VivO and In Vitro. Fed. Proc. 31, 1684 (1978). 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 198 Gatt, S: Enzymatic Aspects of Sphingolipid Degradition. Chem. Phys. Lipids 2, 235 (1970). Arce, A, Maccioni, HJ and Caputto, R: The Biosynthesis of Ganglio- sides: The Incorporation of Galactose, NfAcetylgalactosamine and .NfAcetylneuraminic Acid into Endogenous Acceptors of Subcellular Particles from Rat Brain In Vitro. Biochem. J. 121, 483 (1971). Maccioni, HJ, Arce, A, Zanda, C, and Capputo, R: Rat Brain Micro- somal Gangliosides: Accessibility to a Neuraminidase Preparation and the Possible Existence of Different Pools in Relation to Their Biosynthesis. Biochem. J. 138, 291 (1975). Shur, BD and Roth, S: Cell Surface Glycosyltransferases. Biochim. Biophys. Acta 415, 473 (1975). Keenan, TW and Morré, DJ: Glycosyltransferases: Do They Exist on the Surface Membrane of Mammalian Cells? FEBS Letters 55, 7 (1975). Weinreb, NJ, Brady, R0 and Tappel, AL: The Lysosomal Localization of Sphingolipid Hydrolases. Biochim. Biophys. Acta 159, 141 (1968). Lusis, AJ and Paigen, K: PrOperties of Mouse o-Galactosidase. Biochim. Biophys. Acta 437, 487 (1976). Callahan, JW, Lassila, EL, Den Tandt, W and Philippart, M: Alpha- ‘EfAcetylgalactosaminidase: Isolation, Proterties and Distribution of the Human Enzyme. Biochem. Med. 1, 424 (1973). Swallow, DM, Stokes, DC, Carney, C and Harris, H: Differences Between the NfAcetyl-Hexosaminidase Isozymes in Serum and Tissues. Ann. Hum. Genet. (London) 21, 287 (1974). Willcox, P and Renwick, AGG: The Effect of Neuraminidase on the Chromatographic Behaviour of Eleven Acid Hydrolases from Human Liver and Plasma. Eur. J. Biochem. 13, 579 (1977). Bishop, DF and Sweeley, CC: Plasma a-Galactosidase A: Properties and Comparisons with Tissue a-Galactosidases. Biochim. Biophys. Acta (1978), (in press). Morell, AG, Irvine, RA, Sternlieb, I, Scheinberg, TH and Ashwell, G: Physical and Chemical Studies on Ceruloplasmin: V. Metabolic Studies on Sialic Acid-Free Ceruloplasmin In Vivo. J. Biol. Chem. 243, 155 (1968). Morell, AG, Gregoriadis, G, Scheinberg, IH, Hickman, J and Ashwell, G: The Role of Sialic Acid in Determining the Survival of Glycopro- teins in the Circulation. J. Biol. Chem. 246, 1461 (1971). Rogers, JC and Kornfeld, S: Hepatic Uptake of Proteins Coupled to Fetuin Glycopeptides. Biochem. Biophys. Res. Comm. 4;, 622 (1971). Pricer, WE, Jr, and Ashwell, G: The Binding of Desialylated 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 199 Glycoproteins by Plasma Membranes of Rat Liver. J. Biol. Chem. 246, 4825 (1971). Hudgin, RL, Pricer, WE, Jr, Ashwell, G, Stockert, RJ and Marell, AG: The Isolation and Properties of a Rabbit Liver Binding Protein Spec- ific for Asialoglycoproteins. J. Biol. Chem. 249, 5536 (1974). Kawasaki, T and Ashwell, G: Chemical and Physical Properties of an Hepatic Membrane Protein that Specifically Binds Asialoglycoproteins. J. Biol. Chem. 251, 1296 (1976). Kawasaki, T and Ashwell, G: Carbohydrate Structure of Glycopeptides Isolated from an Hepatic Membrane-Binding Protein Specific for Asialo- glycoproteins. J. Biol. Chem. 251, 5292 (1976). Stockert, RJ, Morell, AG and Scheinberg, IH: The Existence of a Second Route for the Transfer of Certain Glyc0proteins from the Circulation into the Liver. Biochem. Biophys. Res. Comm. 68, 988 (1976). Stahl, P, Schleisinger, PH, Rodman, JS and Doebber, T: Recognition of Lysosomal Glycosidases In Vivo Inhibited by Modified Glycoproteins. Nature 264, 86 (1976). Kawasaki, T and Ashwell, G: Isolation and Characterization of an Avain Hepatic Binding Protein Specific for NfAcetylglucosamine- Terminated Glycoproteins. J. Biol. Chem. 252, 6536 (1977). Simpson, DL, Thorne, DR and Loh, HH: Lectins: Endogenous Carbohydrate- Binding Proteins from Vertebrate Tissues: Functional Role in Recog- nition Processes? Life Sciences 22, 727 (1978). Kaplan, A, Achord, DT and Sly, WS: Phosphohexyl Components of a Lyso- somal Enzyme Are Recognized by Pinocytosis Receptors on Human Fibro- blasts. Proc. Natl. Acad. Sci. USA 14, 2026 (1977). Kaplan, A, Fischer, D and Sly, WS: Correlation of Structural Features of PhOSphomannans with Their Ability to Inhibit Pinocytosis of Human B-Glucuronidase by Human Fibroblasts. J. Biol. Chem. 253, 647 (1978). Kaplan, A, Fischer, D, Acord, DT, and Sly, WS: Phosphohexosyl Recog- nition is a General Characteristic of Pinocytosis of Lysosomal Glyco- sidases by Human Fibroblasts. J. Clin. Invest. 99, 1088 (1977). Sando, GN and Neufeld, EF: Recognition and ReceptoréMediated Uptake of a Lysosomal Enzyme, a-LrIduronidase, by Cultured Human Fibroblasts. Cell 12, 619 (1977). Ullrich, K, Mersmann, G, Weber, E and Van Figura, K: Evidence for Lysosomal Recognition by Human Fibroblasts via a Phosphorylated Carbohydrate Meiety. Biochem. J. 170, 643 (1978). WOollen, JW, Heyworth, R and Walker, PG: Studies on Glucosaminidase: 3. Testicular NfAcetyl-B-Galactosaminidase. Biochem. J. 18, 111 (1961). 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 200 Robinson, D and Stirling, JL: §7Acety1-B-Glucosaminidase in Human Spleen. Biochem. J. 107, 321 (1966). Sandhoff, K: The Hydrolysis of Tay—Sachs Ganglioside (TSG) by Human .NrAcetyl—B-QrHexosaminidase A. FEBS Letters 11, 342 (1970). Wenger, DA, Okada, S and O'Brien, JS: Studies on the Substrate Spec- ificities of Hexosaminidase A and B from Liver. Arch. Biochem. Biophys. 153, 116 (1972). Dean, KJ, Sung, S-SJ and Sweeley, CC: Purification and Partial Char- acterization of Human Liver o—Galactosidases: Is a-Galactosidase B an afingcetylgalactosaminidase? In Enzymes of Lipid Metabolism, F (P Mandel, L Freysz and S Gatt, Eds), Plenum PublishingUCorp. (New York, 1978), p 515. Wolfe, LS, Senior, RG and Ng Ying Kin, NMK: The Structures of Oligo- saccharides Accumulating in the Liver of GMl-Gangliosidosis, Type I. J. Biol. Chem. 249, 1828 (1974). V Ng Ying Kin, NMK and Wolfe, LS: Oligosaccharides Accumulating in the Liver from a Patient with GMz-Gangliosidosis, Varient O (Sandhoff- Jatzkewitz Disease). Biochem. Biophys. Res. Comm. 22, 837 (1974). Ng Ying Kin, NMK and Wolfe, LS: Characterization of Oligosaccharides and Glycopeptides Excreted in the Urine of GMl-Gangliosidosis Patients. Biochem. Biophys. Res. Comm. fig, 123 (1975). Hakomori, S-I and Ishizuka, I: The Table of Glycolipids. In Handbook of Biochemistry and-Molecular Biology: Lipids, Carbohydrates, Steroids, 3rd, Ed., (GD Fasman, Ed) CRC Press (Cleveland, 1975) p 416. Methods in Enzymology,Vol. XIV,_Lipids, (JM Lowenstein, Ed), Academic Press (New York, 1969). Methods in Enzymology, Vol. XXVIII (Part B): Complex Carbohydrates, (V Ginsberg, Ed), Academic Press (New York, 1972). Methods in Carbohydrate Chemistry, Vol. VII: General Methods, Gly- cosaminoglycans and Glycoproteins, (RL Whistler and JN BeMiller, Eds), Academic Press (New York, 1976). Yang, H-J and Hakomori, S-I: A Sphingolipid Having a Novel Type of Ceramide and Lacto-N-Fucopentaose III. J. Biol. Chem. 246, 1192 (1971). Chambers, RE and Clamp, JR: An Assessment of Methanolysis and Other Factors Used in the Analysis of Carbohydrate—Containing Material. Biochem. J. 125, 1009 (1971). Laine, RA, Stellner, K and Hakomori, S-I: Isolation and Characteri- zation of Membrane Glycosphingolipids. Methods in Membrane Biology 3, 205 (1974). 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 201 Susumu, A, Isobe, M and Nagai, Y: High Performance Preparative Column Chromatography of Lipids Using a New Porous Silica, Iatrobeads. Biochim. BiOphys. Acta 424, 98 (1976). Momoi, T, Ando, S and Nagai, Y: High Resolution Preparative Column Chromatographic System for Gangliosides Using DEAF-Sephadex and a New Porous Silica, Iatrobeads. Biochim. Biophys. Acta 441, 448 (1976). Korniat, BK and Hof, LB: Isolation and Characterization of Gluco- cerebrosides from Adult Rat Brain. J. Neurochem. 29, 557 (1978). Kundu, SK and Ray, SK: A Rapid and Quantitative Method for the Iso- lation of Gangliosides and Neutral Glycosphingolipids by DEAR-Silica Gel Chromatography. J. Lipid Res. 12, 390 (1978). Ando, S and Yu, R: Isolation and Characterization of a Novel Tri— Sialoganglioside, GTla’ from Human Brain. J. Biol. Chem. 252, 6247 (1977). Iwamori, M and Yoshitaka, N: A New Chromatographic Approach to the Resolution of Individual Gangliosides: Ganglioside Mapping. Biochim. Biophys. Acta 528, 257 (1978). Beutler, E, Guinto, E and Kuhl, W: Placental Acid Hydrolase Purifi- cation on Concanavalin ArSepharose. J. Lab. Clin. Med. 22, 672 (1975). Ross, TT, Hayes, CE and Goldstein, IJ: Carbohydrate-Binding Properties of an Immobilized o-Qrcalactopyranosyl-Binding Protein (Lectin) from the Seeds of Bandeirea simplificifolia. Carb. Res. 41, 91 (1976). Barker, R, Olsen, KW, Shaper, JH and Hill, RL: Agarose Derivatives of Uridine Diphosphate and NfAcetylglucosamine for the Purification of a Galactosyltransferase. J. Biol. Chem. 247, 7135 (1972). Paulson, JC, Beranek, WE and Hill, RL: Purification of a Sialyltrans- ferase from Bovine Colostrum by Affinity Chromatography on CDP-Agarose. J. Biol. Chem. 252, 2356 (1977). Dale, GL and Beutler, E: Enzyme Replacement Therapy in Gaucher's Disease: A Rapid High-Yield Method for Purification of Glucocerebro- sidase. Proc. Natl. Acad. Sci. USA 12, 4672 (1976). Leaback, DH and Robinson, HK: Ampholyte Displacement Chromatography- A New Technique for the Separation of Proteins Illustrated by the Resolution of B-N-Acetyl-D-Hexosaminidase Isoenzymes Unresolvable by Isoelectric Focusing or Conventional Ion-Exchange Chromatography. Biochem. Biophys. Res. Comm. 21, 248 (1975). Evans, JE and McCleur, RH: High Pressure Liquid Chromatography of Neutral Glycosphingolipids. Biochim. Biophys. Acta 270, 565 (1972). McCleur, RH and Evans, JE: Preparation and Analysis of Benzoylated Cerebrosides. J. Lipid Res 14, 611 (1973). Bresle, JM and Lassalas, B: Separation des Oses des Di- et 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 202 Triholosides par Chromatographie D'Bchange D'Ions. Ann. Biol. Anim. Biochim. Biophys. 11, 545 (1974). Iwamori, M, Moser, HW, McCleur, RH and Kishimoto, Y: 3-Ketosphingo- lipids: Application to the Determination of Sphingolipids which Con- tain 4-Sphingenine. Biochim. Biophys. Acta 380, 308 (1975). McCleur, RH and Evans, JE: Quantitative Analysis of Brain Galactosyl- ceramide by High Performance Liquid Chromatography of Their Perben— zoylated Derivatives. J. Lipid Res. 11, 412 (1976). Suzuki, A, Hands, S and Yamakawa, T: Separation of Molecular Species of Glucosylceramide by High Performance Liquid Chromatography of Their Benzoyl Derivatives. J. Biochem. (Tokyo) 29, 1181 (1976). Sillerud, LO, Prestegard, JH, Yu, RK, Schafer, DE and Konigsberg, WH: Assignment of the 13C Nuclear Magnetic Resonance Spectrum of Aqueous Ganglioside GM1 Micelles. Biochemistry (1978), (in press). Arakawa, M and Muramatsu, M: Endo-BfigrAcetylglycosaminidase Acting on the Carbohydrate Moieties of Glycoproteins. J. Biochem. (Tokyo) 12, 307 (1974). Tarentino, AL and Maley, F: Purification and Properties of an Endo- BfififAcetylglucosaminidase from Streptomyces griseus. J. Biol. Chem. 249, 811 (1974). Koide, N and Muramatsu, T: Endo-BfNTAcetylglucosaminidase Acting on Carbohydrate Moieties of Glyc0proteins. J. Biol. Chem. 249, 4897 (1974). Ito, S, Muramatsu, T and Kobata, A: Endo-BjfifAcetylglucosaminidases Acting on Carbohydrate Moieties of Glycoproteins: Purification and Properties of the Two Enzymes with Different Specificities from Clostridium perfringens. Arch. Biochem. Biophys. 111, 78 (1975). Tai, T, Yamashita, K and Kobata, A: The Substrate Specificity of Endo-Bengcetylglucosaminidase CII and H. Biochem. Biophys. Res. Comm. 12, 434 (1977). Mort, AJ and Lamport, DTA: Anhydrous Hydrogen Fluoride Deglycosylates Glycoproteins. Analyt. Biochem. 22, 289 (1977). Karlsson, K—A, Samuelsson, BE and Steen, G0: The Sphingolipid Compo- sition of Bovine Kidney Cortex, Medulla and Papilla. Biochim. Biophys. Acta 316, 317 (1973). 'Karlsson, K—A, Samuelsson, BE and Steen, G0: Lipid Pattern and Na+- K+-Dependent Adenosine Triphosphatease Activity in the Salt Gland of Duck Before and After Adaptation to Hypertonic Saline. J. Membr. Biol. 2, 169 (1971). Karlsson, KeA, Samuelsson, BE and SLeen, G0: The Lipid Composition and Na+-K+¥Dependent Adenosine Triphosphatase Activity of the Salt 99. 100. 101. 102. 103. 104. 105. 106. 107. 108. 109. 110. 111. 112. 203 (Nasal) Gland of Eider Duck and Herring Gull: A Role for Sulfatides in Sodium-Ion Transport. Eur. J. Biochem. 12, 243 (1974). Ureda, T, Eqawa, K and Nagai, Y: Enhancement of Sulfatide Metabo- lism in Hypertrophied Kidney of C3H/He Mouse with Reference to [Na+,K+]—Dependent ATPase. Jap. J. Exptl. Med. 42, 87 (1976). Makita, A: Biochemistry of Organ Glycolipids, II. Isolation of Human Kidney Glycolipids. J. Biochem. (Tokyo) 22, 269 (1964). Mfirtensson, E: Sulfatides of Human Kidney: Isolation, Identification, and Fatty Acid Composition. Biochim. Biophys. Acta 116, 521 (1966). Sweeley, CC and Klionsky, B: Fabry's Disease: Classification as a Sphingolipidosis and Partial Characterization of a Novel Glyco- “ lipid. J. Biol. Chem. 238, PC 3148 (1963). Sweeley, CC and Klionsky, B: Glycolipid Lipidosis: Fabry's Disease. - In The Metabolic Bases of Inherited Disease, (JB Stanbury, JB ' ‘5 Wyngaarden and DS Fredrickson, Eds) McGrawbHill (New York, 1966) p 618. Christensen Lou, HO: A Biochemical Investigation of Angiokeratoma Corporis Diffusum. Acta Pathol. Microbiol. Scand. 22, 332 (1966). Schibanoff, JM, Kamoshita, S and O'Brien, JS: Tissue Distribution of Glycosphingolipids in a Case of Fabry's Disease. J. Lipid. Res. 19, 515 (1969). Makita, T: A Study on Glycolipid in Fabry's Disease. Jap. J. Exp. Med. 22, 35 (1969). Desnick, RJ, Sweeley, CC and Krivit, W: A Method for the Quantitative Determination of Neutral Glycosphingolipids in Urine Sediment. J. Lipid Res. 11, 31 (1971). Desnick, RJ, Bleiden, LD, Sharp, HL and Moller, JH: Cardiac Valvular Anomalies in Fabry's Disease. Circulation 21, 818 (1976). Gatt, S and Berman, ER: Studies on Brain Lipids in Tay-Sachs Disease- I. Isolation of Two Sialic Acid-Free Glycolipids. J. Neurochem.'1g, 43 (1963). Adams, EP and Gray, GM: The Carbohydrate Structures of the Neutral Glycolipids in Kidneys of Different Mouse Strains with Special Refernce to the Ceramide Dihexosides. Chem. Phys. Lipids 2, 147 (1968). ‘ Coles, L, Hay, JB and Gray, GM: Factors Affecting the Glycosphingo- lipid Composition of Murine Tissues. J. Lipid Res. 11, 158 (1970). Hay, JB and Gray, GM: The Effect of Testosterone on the Glycosphingo- lipid Composition of Mouse Kidney. Biochim. Biophys. Acta 202, 566 (1970). 4p. -- 11): Iutbllfiwl. :01... ...! ...n o, i a 113. 114. 115. 116. 117. 118. 119. 120. 121. 122. 123. 124. 125. 126. 204 Naiki, M and Marcus, DM: Human Erythrocyte P and PR Blood Group Antigens: Identification as Glycolipids. Biochim. BiOphys. Acta 22, 1105 (1974). Kortekangos, AE, Kaarsalo, E, Malartin, L, Tippett, P, Gavin, J, Noades, J, Sanger, R and Race, RR: The Red Cell Antigen Pk and Its Relationship to the P System: The Evidence of Three More Pk Families. Vox Sang. 12, 385 (1965). Race, RR and Sanger, R: Blood Groups in Man, 6th Ed., Blackwell Scientific Publications (Oxford, 1975). Matson, GA, Swanson, J, Noades, J, Sanger, R and Race, RR: A 'New' Antigen and Antibody Belonging to the P Blood Group System. Amer. J. Hum. Genetics 11, 26 (1959). Landsteiner, K and Levine, P: Further Observations on Individual Differences in Human Blood. Proc. Soc. Exp. Biol. Med. 22, 941 (1927). Levine, P, Bobbitt, OB, Waller, RK and Kulmichel, A: Isoimmunization of a New Blood Factor in Tumor Cells. Proc. Soc. Exp. Biol. Med. 11, 403 (1951). Sanger, R: An Association Between the P and Jay Systems of Blood Groups. Nature 176, 1163 (1955). Naiki, M and Marcus, DM: An Immunochemical Studey of the Human Blood Group P1, P and PR Glycosphingolipid Antigens. Biochemistry 12, 4837 (1975). Gurner, BW and Coombs, RRA: Examination of Human Leucocytes for ABO, MN, Rh, T a, Lutheran and Lewis Systems of Antigens by Means of Mixed Ery€hrocyte-Leucocyte Agglutination. Vox Sang. 2, 13 (1958). Fellous, M, Couillin, P, Neauport-Sautes, C, Frezal, J, Bilardon, C and Dausset, J: Studies of Human Alloantigens on Man-Mouse Hybrids: Possible Syntheny Between HL-A and P-Systems. Eur. J. Immunol..2, 543 (1973). Fellous, M, Gerbal, A, Tessier, C, Frezal, J, Dausset, J and Salmon, C: Studies on the Biosynthetic Pathway of Human P Erythrocyte anti- gens Using Somatic Cells in Culture Vox Sang. 22, 518 (1974). Marcus, DM, Naiki, M and Kundu, SK: Abnormalities in the Glycosphingo- lipid Content of Human PR and p Erythrocytes. Proc. Nat. Acad. Sci. USA 12, 3263 (1976). Vance, DE and Sweeley, CC: Quantitative Determination of the Neutral Glycosyl Ceramides in Human Blood. J. Lipid Res. 2, 621 (1967). Kijimoto-Ochai, S, Naiki, M and Makita, A: Defects of Glycosyltrans- ferase Activities in Human Fibroblasts of Pk and p Blood Group Pheno- types. Proc. Natl. Acad. Sci. USA 12, 5407 (1977). 127. 128. 129. 130. 131. 132. 133. 134. 135. 136. 137. 138. 139. 140. 205 Fellous, M, Gerbal, A, Nobillot, G and weils, J: Studies on the Biosynthetic Pathway of Human Erythrocyte Antigen Using Genetic Complementation Tests Between Fibroblasts from Rare p and Pk Pheno- type Donors. Vox Sang. 21, 262 (1977). Klenk, E and Lauenstein, K: Uber die Glycolipoide und Sphingomyeline des Stromas der Pferdeerythrocyten. Happe-Seyler's Z. Physiol. Chem. 295, 164 (1953). Kint, JA and Huys, A: Effect of Bacterial Neuraminidase on the Iso- enzymes of Acid Hydrolases of Human Brain and Liver. In Glycolipids, Glycoproteins and Muc0polysaccharides of the Nervous System, (V Zambotti, G Tettamanti and M Arrigoni, Eds) Adv. Exptl. Med. Biol. 22, 273 (1972). Kint, JA and Carton, D: Fabry's Disease. In Lysosomes and Stora e Dis- eases, (HG Hers and F van Hoof, Eds), Academic Press (New York, 1973) p 357. Desnick, RJ, Klionsky, B and Sweeley, CC: Fabry's Disease (a-Galacto- sidase A Defiency). In The Metabolic Basis of Inherited Disease, 4th 222, (JB Stanbury, JB Wyngaarden and DS Fredrickson, Eds), McGraw-Hill (New York, 1978) p 810. Kahlke, W: Angiokeratoma Corporis Diffusum (Fabry's Disease). In Lipids and Lipidoses, (G Schettler, Ed), Springer-Verlag (New York, 1967) p 332. Dean, KJ and Sweeley, CC: Fabry Disease. In Practical Enzymology of the §phingglipidoses, (RH Glew and SP Peters, Eds), Alan R. Liss (New York, 1977) p 173. Van Den Bergh, FAJTM and Tager, JM: Localization of Neutral GIYCOP sphingolipids in Human Plasma. Biochim. Biophys. Acta 441, 391 (1976). Dawson, G, Kruski, AW and Scanu, AM: Distribution of Glycosphingo- lipids in the Serum Lipoproteins of Normal Human Subjects and Patients with Hypo- and Hyperlipidemias. J. Lipid Res. 11, 125 (1976). Clarke, JTR, Stoltz, JM and Mulcahey, MR: Neutral Glycosphingolipids of Serum LipOproteins in Fabry's Disease. Biochim. Biophys. Acta 431, 317 (1976). Marcus, DM, Naiki, M, Fang, J and Ledeen, R: The Structure of a Gly- cosphingolipid with Blood Group P1 Activity. Fed. Proc. 22, 646 (1975). Naiki, M, Fang, Dedeen, R and Marcus, DM: Structure of the Human Erythrocyte Blood Group Pl Glycosphingolipid. Biochemistry 12, 4831 (1975). Marcus, DM: Isolation of a Substance with Blood Group P1 Activity from Human Erythrocyte Stroma. Transfusion.11, 16 (1971). Eta, T, Ichikawa, Y, Nishimura, K, Ando, S and Yamakawa, T: Chemistry (”...—J 141. 142. 143. 144. 145. 146. 147. 148. 149. 150. 151. 152. 153. 206 of the Posthemolytic Residue or Stroma of Erythrocytes XVL. Occur- ance of Ceramide Pentasaccharide in the Membrane of Erythrocytes and Reticulocytes of Rabbit. J. Biochem. (Tokyo) 22, 205 (1968). Stellner, K, Saito, H and Hakomori, S-I: Determination of Aminasugar Linkages of Ceramide Pentasaccharides of Rabbit Erythrocytes.and of Forssman Antigens. Arch. Biochem. Biophys. 155, 464 (1973). Hakomori, S-I: Glycosphingolipids Having Blood Group ABH and Lewis Specificities. Chem. Phys. Lipids 2, 96 (1970). Koécielak, J, Piasek, A, G6rniak, H, Gardas, A and Gregor, A: Struc- tures of Fucose-Containing Glycolipids with H and B Blood-Group 2‘ Activity and of Sialic Acid and Glucosamine-Cantaining Glycolipid of Human-Erythrocyte Membrane. Eur. J. Biochem..21, 214 (1973). Hanfland, P and Egli, H: Quantitative Isolation and Purification of Blood Group-Active Glycosphingolipids from Human B Erythrocytes. Vox Sang. 22, 438 (1975). H Hanfland, P and Egge, H: Mass Spectrometric Analysis of Permethylated Glycosphingolipids 1. Sequence of Two Blood-Group Active Glyco- sphingolipids I. Sequence of Two Blood-Group Active Glycosphingo- lipids fram Human B Erythrocyte Membranes. Chem. Phys. Lipids 12, 243 (1975). Hanfland, P: Characterization of B and H Blood-Group Active Glyco- sphingalipids from Human B Erythrocyte Membranes. Chem. Phys. Lipids 12, 105 (1975). Wherrett, JR and Hakomori, S-I: Characterization of a Blood Group B Glycolipid Accumulating in the Pancreas of a Patient with Fabry's Disease. J. Biol. Chem. 248, 3046 (1973). Gardas, A and Koécielak, J: New Farm of Ar, B- and H-Blaod Group- Active Substances Extracted form Erythrocyte Membranes. Eur. J. Biochem.‘22, 178 (1973). Gardas, A and Koscielak, J: Megaloglycolipids-Unusually Complex Glyco- sphingolipids of Human Erythrocyte Membrane with A,B,H and I Blood Group Specificity. FEBS Letters 22, 101 (1974). Gardas, A: A Structural Study on a Macro-Glycolipid Containing 22 Sugars Isolated from Human Erythrocytes. Eur. J. Biochem.'22, 177 (1976). Watkins, WM: Blood-Group Substances. Science 152, 172 (1966). Watkins, WM: Blood Group Substances. In Glygoproteins: Their Compo- sition, Structure and Function, (A Gottschalk, Ed), Elsevier (New York, 1972) p 830. Pigman, W: Blood Group Glycoproteins. In The Glycoconjugates, Vol. I. Mammalian Glycoproteins and C1ycoli ids, (MI Horowitz and W Pigman, Eds), Academic Press (New York, 1977) p 181. 0._—J 154. 155. 156. 157. 158. 159. 160. 161. 162. 163. 164. 165. 166, 115 7. 168 o 207 Lundblad, A: Urinary Glycoproteins, Glycopeptides and Oligosac- charides. In The Glycoconjugates, Vol. I. Mammalian Glycoproteins and Glycolipids, (MI Horowitz and W Pigman, Eds), Academic Press (New York, 1977) p 441. Tondeur, M and Resibois, A: Fabry's Disease in Children: An Electron Microscopic Study. Virchaw Arch. (Zellpath.) 1, 239 (1969). Scriba, K: Zur Pathogenese des Angiokeratoma Corporis Diffusum Fabry mit Cardiavasorenalem Symptomenkomplex. Vehr. Deutsch Ges. Path..23, 221 (1950). Frost, P, Tanaka, Y and Spaeth, GL: Fabry's Disease - Glycolipid Lipidosis. Histochemical and Electron Microscopic Studies of Two Cases. Amer. J. Med. 22, 618 (1966). Ferrans, VJ, Hibbs, RB and Burda, CD: The Heart in Fabry's Disease: A Histochemical and Electron Microscopic Studey. Am. J. Cardiol. 22, 95 (1969). Lehner, T and Adams, CWM: Lipid Hisochemistry of Fabry's Disease. J. Pathol. Bact. 22, 411 (1968). Coles, L and Gray, GM: The Biosynthesis of Digalactosylceramide in the Kidney of the C57/BL Mouse. Biochem. BiOphys. Res. Commun. 22, 520 (1970). Gray, GM: The Effect of Testosterone on the Biosynthesis of the Neutral Glycsphingalipids in the C57/BL Mouse Kidney. Biochim. Biophys. Acta 239, 494 (1971). Stoffyn, A, Stoffyn, P and Hauser, G: Structures of Trihexosylcer- amide Biosynthesized In Vitro by Rat Kidney Galactosyltransferase. Biochim. BiOphys. Acta 360, 174 (1974). Robbins, PW and Macpherson, 1: Control of Glycolipid Synthesis in a Cultured Hamster Cell Line. Nature 229, 569 (1971). Sakiyama, H, Gross, SK and Robbins, PW: Glycolipid Synthesis in Nor- mal and Virus-Transformed Hamster Cell Lines. Proc. Natl. Acad. Sci. USA 22, 872 (1972). Sakiyama, H and Robbins, PW: The Effect of Dibutyryl Adenosine 3'— 5'-Cyclic Monophosphate on the Synthesis of Glycolipids by Normal and Transformed Nil Cells. Arch. Biochem. Biophys. 154, 407 (1973). Sakiyama, H and Robbins, PW: Effect of Transformation by Hamster Sarcoma Virus on the Glycolipid Composition of Secondary Hamster Embryo Cells and the Nil Cell Line. In Vitro 2, 331 (1974). Wolf, BA and Robbins, PW: Cell Mitotic Cycle Synthesis of Nil Ham- ster Glycolipids Including the Forssman Antigen. J. Cell Biol. 21, 676 (1974) Robbins, PW and Macpherson, I: Glycolipid Synthesis in Normal and 169. 170. 171. 172. 173. 174. 175. 176. 177. 178. 179. 180. 181. 208 Transformed Animal Cells. Proc. Roy. Soc. (London) (Series B) 177, 49 (1971). Hakomori, S-I and Murakami, WT: Glycolipids of Hamster Fibroblasts and Derived Malignant-Transformed Cell Lines. Proc. Natl. Acad. Sci. USA 22, 254 (1968). Hakomori, S-I: Cell Density-Dependent Changes of Glycolipid Concen- trations in Fibroblasts and Loss of This Response in Virus-Trans- formed Cells. Proc. Natl. Acad. Sci. USA 21, 1741 (1970). Kijimoto, S and Hakomori, S—I: Enhanced Glycolipid: a-Galactosyl- transferase Activity in Contact-Inhibited Hamster Cells, and Loss of this Response in Polyoma Transformants. Biochem. Biophys. Res. Comm. 22, 557 (1971). Yogeeswaren, G, Laine, RA and Hakomori, S—I: Mechanism of Cell Con- tact-Dependent Glycolipid Synthesis: Further Studies with Glycolipid Glass Complex. Biochem. Biophys. Res. Comm.'2g, 591 (1974). Gahmberg, CG and Hakomori, S-I: Organization of Glycolipids and Glycoproteins in Surface Membranes: Dependency on Cell Cycle and on Transformation. Biochem. Biophys. Res. Comm. 22, 283 (1974). Gahmberg, CG and Hakomori, S-I: Surface Carbohydrates of Hamster Fibroblasts: 1. Chemical Characterization of Surface-Labeled Glyco- sphingolipids and a Specific Ceramide Tetrasaccharide for Trans- formants. J. Biol. Chem. 229, 2438 (1975). Zavada, J and Macpherson, I: Transformation of Hamster Cell Lines In Vitro by a Hamster Virus. Nature 225, 24 (1970) Critchley, DR and Macpherson, 1: Cell Density Dependent Glycolipids in Nil 2 Hamster Cells, Derived Malignant Cells and Transformed Cell Lines. Biochim. Biophys. Acta 296, 145 (1973). Chandrabose, KA and Macpherson, IA: Glycolipid Glycosyl Transferases of a Hamster Cell Line in Culture: I. Kinetic Constants, Substrate and Donor Nucleotide Sugar Specificities. Biochim. Biophys. Acta _4_2_9_, 96 (1976). Chandrabose, KA Graham, JM and Macpherson, IA: Glycolipid Glycosyl Transferases of a Hamster Cell Line in Culture: II. Subcellular Distribution and the Effect of Culture Age and Density. Biochim. Biophys. Acta 222, 112 (1976). Brady, R0 and Fishman, PH: Biosynthesis of Glycolipids in Virus- Transformed Cells. Biochim. Biophys. Acta 355, 121 (1974). Richardson, CL, Baker, SR, Morré, DJ and Keeman, TW: Glycosphingo- lipid Synthesis and Tumarigenesis: A Role for the Golgi Apparatus in the Origin of Specific Receptor Molecules of the Mammalian Cell Surface. Biochim. Biophys. Acta.211, 175 (1975). Chatterjee, S and Sweeley, CC: Biosynthesis of Proteins, Nucleic Acids 182. 183. 184. 185. 186. 187. 188. 189. 190. 191. 192. 193. 194. 209 and Glycosphingolipids by Synchronized KB Cells. Biochem. Biophys. Res. Comm. 22, 585 (1973). Chatterjee, S, Sweeley, CC and Velicer, LF: Glycosphingolipids of Human KB Cells Grown in Monolayer, Suspension, and Synchronized Cultures. J. Biol. Chem. 250, 61 (1975). Chatterjee, S, Velicer, LF and Sweeley, CC: Glycosphingolipid Glyco- syl Hydrolases and Glycosidases of Synchronized KB Cells. J. Biol. Chem. 250, 4972 (1975). Lockney, M, Moskal, JR, Fung, Y-K and Macher, BA: The Effect of Butyrate on Glycosphingolipid Metabolism in Human Epidermoid Carci- noma (KB) Cells. Fed. Proc. 21, 1766 (1978). Vance, DE, Krivit, W and Sweeley, CC: Metabolism of Neutral Glyco- sphingolipids in Plasma of Normal Human and a Patient with Fabry's Disease. J. Biol. Chem. 250, 8119 (1975). Carne, LR and Watkins, WM: Human Blood Group B Gene-Specific a-3- Galactosyltransferase: Purification of the Enzyme in Serum by Bio- specific Adsorption onto Blood Group 0 Erythrocyte Membranes. Biochem. Biophys. Res. Comm. 11, 700 (1977). Nagai, M, Daue, V, Muencsh, H and Yoshida, A: Human Blood Group Glyco- syltransferase: II. Purification of Galactosyltransferase. J. Biol. Chem. 253, 380 (1978). Hayashi, A, Matsubara, T, Hakomori, S-I and Andrews, HD: Sphingo- lipids with Leb Activity, and the Co-Presence of Le3-, Leb-Glyco- lipids in Human Tumor Tissue. Biochim. Biophys. Acta 202, 225 (1970). Stellner, K, Hakomori, S-1 and Warner, GA: Enzymatic Conversion of "HI-Glycolipid" to A or B-Glycolipid and Deficiency of These Enzyme Activities in Adenocarcinoma. Biochem. Biophys. Res. Comm. 22, 439 (1973). Brady, R0, Gal, AE, Bradley, RM and Mfirtinsson, E: The Metabolism of Ceramide Trihexosides: I. Purification and Properties of an Enzyme that Cleaves the Terminal Galactose Molecule of Galactosylgalactosyl- glucosylceramide. J. Biol. Chem. 121, 1021 (1967). Brady, RO, Gal, AE, Bradley, RM, Martinssan, E, Warshaw, AL and Lester, L: Enzymatic Defect in Fabry's Disease: Ceramidetrihexosidase Deficiency. New Engl. J. Med. 276, 1163 (1967). Kint, JA: Fabry's Disease: Alpha-Galactosidase Deficiency. Science 167, 1268 (1970). Romeo, G and Migeon, BR: Genetic Inactivation of the a-Galactasidase Locus in Carriers of Fabry's Disease. Science 170, 180 (1970). Brady, RO, Ulendorf, BW and Jacobson, CB: Fabry's Disease: Antenatal Detection. Science 172, 174 (1971). LIV—IA 195. 196. 197. 198. 199. 200. 201. 202. 203. 204. 205. 206. 207. 208. 209 O 210 Clarke, JTR, Knack, J and Crawhall, JC: Ceramide Trihexosidosis (Fabry's Disease) Without Skin Lesions. New Engl. J. Med. 284, 233 (1971). Brady, RO: Prenatal Diagnosis of Lipid Storage Diseases. Clin. Chem. 12, 811 (1970). Handa, S, Ariga, T, Miyatake, T and Yamakawa, T: Presence of a-Anomeric Glycosidic Configuration in the Glycolipids Accumulated in Kidney with Fabry's Disease. J. Biochem. (Tokyo) 22, 625 (1971). Bensaude, I, Callahan, J and Philippart, M: Fabry's Disease is an a-Calactosidosis: Evidence for an a-Configuration in Trihexosylcer- amide. Biochem. Biophys. Res. Comm. 22, 913 (1971). Li, Y-T, Li, S-C: Anomeric Configuration of Galactose Residues in Ceramide Trihexosides. J. Biol Chem. 246, 3769 (1971). Clarke, JTR, Wolfe, LS and Perlin, AS: Evidence for a Terminal 9'2? Galactopyranosyl Residue in Galactosylgalactosylglucosylceramide from Human Kidney. J. Biol. Chem. 246, 3763 (1971). Hakomori, S-I, Siddiqui, G, Li, Y—T, Li, S-C and Hellerqvist, CF: Anomeric Structures of Globoside and Ceramide Trihexoside of Human Erythrocytes and Hamster Fibroblasts. J. Biol. Chem. 246, 2271 (1971). Li, Y-T, Li, S-C and Dawson, G: Anomeric Structure of Ceramide Di- galactoside Isolated from the Kidney of a Patient with Fabry's Di- sease. Biochim. BiOphys. Acta 260, 88 (1970). Beutler, E and Kuhl, W: Fabry's Disease: Structural or Regulatory Mutation? J. Lab. Clin. Med. 12, 987 (1971). Kint, JA: On the Existence and the Enzymatic Interconversion of the Isozymes of a-Galactosidases in Human Organs. Arch. Int. Physiol. Biochem. 12, 633 (1971). Rietra, PJGM, Van Den Bergh, FAJTM and Tager, JM: Properties of the Residual a-Galactosidase Activity in the Tissues of a Fabry Hemi- zygote. Clin. Chim. Acta 22, 401 (1975). Beutler, E, Guinto, E and Kuhl, W: Variability of o-Galactosidase A and B in Different Tissues of Man. Am. J. Hum. Genet. 22, 42 (1973). Ho, MW: Hydrolysis of Ceramide Trihexoside by a Specific a-Galacto- sidase from Human Liver. Biochem. J. 133, 1 (1973). Ho, MW, Beutler, E, Tennent, L and O'Brien, JS: Fabry's Disease: Evidence for a Physically Altered a-Galactosidase. Am. J. Hum. Genet. 22, 256 (1972). Kint, JA and Huys, A: Effect of Bacterial Neuraminidase on the Iso- enzymes of Acid Hydrolases of Human Brain and Liver. In Advances in Experimental Medicine and Biology, Vol. 25, Glycolipids, Glycoproteins and Mucopolysaccharides of the Nervous System, (V Zambotti, G ‘ " 210. 211. 212. 213. 214. 215. 216. 217. 218. 219. 220. 221. 222. 223. 211 Tettamanti and M Arrigoni, Eds), Plenum Press (New York, 1972) p 2730 Kano, I and Yamakawa, T: The Properties of a-Galactosidase Remain- ing in Kidney and Liver of Patients with Fabry's Disease. Chem. Phys. Lipids 12, 283 (1974). Romeo, G, Di Matteo, G, D'Ursa, M, Li, S-C and Li, Y-T: Character- ization of Human a-Galactosidase A and B Before and After Neuramini- dase Treatment. Biochim. Biophys. Acta 391, 349 (1975). Romeo, G, Di Matteo, G, D'Ursa, M, Wan, C-C, Li, S-C and Li, Y-T: Hydrolysis of Ceramide Trihexoside and Melibiose by a-Galactosidase A and B of Human Liver. Biochem. Expt. Biol. 11, 289 (1974/1975). Li, S-C, Wan, C-C, Mazotta, MY and Li, Y-T: Requirement of an Acti- vator for the Hydrolysis of Sphingolipids by Glycosidases of Human Liver. Carb. Res. 22, 189 (1974). Li, S-C and Li, Y-T: An Activator Stimulating the Enzymatic Hydrolysis of Sphingolipids. J. Biol. Chem. 251, 1159 (1976). Schram, AW, Hamers, MN, Brouwer-Kelder, B, Donker-Kooperman, W and Tager, JM: Enzymological Properties and Immunological Characteriza- tion of a-Galactosidase Isoenzymes from Normal and Fabry Liver. Biochim. Biophys. Acta.221, 125 (1972). Snyder, PD, Jr, Wold, F, Bernlohr, RW, Dullum, C, Desnick, RJ, Krivit, W and Candie, RM: Enzyme Therapy 11. Purification of Human a-Galacto- sidase A: Stabilization to Heat and Protease Degradation by Complex- ing with Antibody and by Chemical Modification. Biochim. BiOphys. Acta 222, 432 (1974). Kano, I and Yamakawa, T: Human Kidney a-Galactasidase: Multiplicity and Enzyme Activities for Ceramide Trihexoside and Some Aryl a-Galac- tosidases. J. Biochem. (Tokyo) 12, 347 (1974). Mapes, CA and Sweeley, CC: Preparation and Properties of an Affinity Column Adsorbent for Differentiation of Multiple Forms of a-Galacto- sidase Activity. J. Biol. Chem. 248, 2461 (1973). Beutler, E and Kuhl, W: Purification and Pr0perties of Human a-Galac- tosidases. J. Biol. Chem. 247, 7195 (1972). Beutler, E and Kuhl, W: Relationship Between Human a-Galactosidase Isozymes. Nature New Biol. 239, 207 (1972). Johnson, WC and Brady, R0: Ceramide Trihexosidase From Human Placenta. In Methods in Enzymology, Vol. XXVIII (Part B): Complex Carbohydrates, (V Ginsberg, Ed), Academic Press (New York, 1972) p 849. Mayes, JS and Beutler, E: Alpha-Galactosidase A from Human Placenta: Stability and Subunit Size. Biochim. Biophys. Acta 484, 408 (1977). Kusiak, JW, Quirk, JM, Brady, R0 and Mook, GE: Purification and 224. 225. 226. 227. 228. 229. 230. 231. 232. 233. 234. 235. 236. 237. 212 Properties of the Two Major Isozymes of a-Galactosidase from Human Placenta. J. Biol. Chem. 253, 184 (1978). Romeo, G and Migeon, BR: Genetic Inactivation of the e-Galactosidase Locus in Carriers of Fabry's Disease. Science 170, 180 (1970). Wood, S and Nadler, HL: Fabry's Disease: Absence of an a-Galacto- sidase Isoenzyme. Am. J. Hum. Genet. 12, 250 (1972). Beutler, E and Kuhl, W: Biochemical and ElectrOphoretic Studies of a-Galactosidase in Normal Man, Patients with Fabry's Disease and in Equidae. Am. J. Hum. Genet. 22, 237 (1972). Crawhall, JC and Banfalvi, M: Fabry's Disease: Differentiation Between Two Forms of a-Galactosidase by Myoinosital. Science 177, 527 (1972). Desnick, RJ, Allen, KY, Desnick, SJ, Raman, MK, Bernlohr, RW and Krivit, W: Fabry's Disease: Enzymatic Diagnosis of Hemizygotes and Heterozygotes. J. Lab. Clin. Med. 21, 157 (1973). * Johnson, DL and Desnick, RJ: Molecular Pathology of Fabry's Disease: Physical and Kinetic Properties of a-Galactosidase A in Cultured Human Endothelial Cells. Biochim. Biophys. Acta (1978), (in press). Mapes, CA, Anderson, RL and Sweeley, CC: Trihexosylceramide: Galac- tosyl Hydralase in Normal Human Serum and Plasma and Its Absence in Patients with Fabry's Disease. FEBS Letters 1, 180 (1970). Mapes, CA, Suelter, CH and Sweeley, CC: Isolation and Characteriza- tion of Ceramide Trihexosidases (Form A) from Human Plasma. J. Biol. Chem. 248, 2471 (1973). Bishop, DF, Wampler, DE, Sgouris, JT, Bonefeld, RJ, Anderson, DK, Hawley, MC and Sweeley, CC: Pilot Scale Purification of a—Galacto- sidase A from Cahn Fraction IV-l of Human Plasma. Biochim. Biophys. Acta 212, 109 (1978). Rietra, PJGM, Tager, JM and Borst, P: Detection and Properties of an Acid a-Galactosidase (Ceramide Trihexosidase) in Normal Human Urine. Biochim. Biophys. Acta 279, 436 (1972). Rietra, PJGM, Molenaar, JL, Hamers, MN, Tager, JM and Borst, P: Investigation of the a-Galactosidase Deficiency in Fabry's Disease Using Antibodies Against the Purified Enzyme. Eur. J. Biochem. 22, 89 (1974). Del Monte, MA, Johnson, DL, Cotlier, E, Krivit, W and Desnick, RJ: Diagnosis of Fabry's Disease.by Tear a-Galactosidase A. New Engl. J. Med. 290, 57 (1974). Dean, KJ, Sung, SS-J and Sweeley, CC: Purification and Partial Char- acterization of Human Liver a-Galactosidases: Is a-Galactosidase B an afifirAcetylgalactosaminidase? Fed. Proc. 22, 731 (1977). Dean, KJ, Sung, SS-J and Sweeley, CC: The Identification of 238. 239. 240. 241. 242. 243. 244. 245. 246. 247. 248. 249. 250. 251. 213 a-Galactosidase B from Human Liver as an aegyAcetylgalactosaminidase. Biochem. Biophys. Res. Comm. 11, 1411 (1977). Schram, AW,.Hamers, MN and Tager, JM: The Identity of a-Galactosidase B from Human Liver. Biochim. Biophys. Acta 482, 138 (1977). Hakomori, S, Wang, S-M and Young, WW, Jr: Isoantigenic Expression of Forssman Glycolipid in Human Gastric and Colonic Mucosa: Its Possible Identity with "A-Like Antigen" in Human Cancer. Proc. Natl. Acad. Sci. USA 12, 3023 (1977). Kawanami, J: The Appearance of Forssman Hapten in Human Tumors. J. Biochem. (Tokyo) 11, 783 (1972). Forssman, J: Die Herstellung Hochwertiger Spezifischer Schafhamo- lysine ohne Verwendung von Shafblut: Ein Beitrag zur Lehre van Heterolager Antikorperbildung. Biochem. Z. 21, 78 (1911). Brunius, FE: Chemical Studies on the True Forssman Hapten, the Car- responding Antibody and their Interaction, Aktiebologet Fahlcrantz, Boktryckeri (Stockholm, 1936). Papirmesiter, B and Mallette, MP: The Isolation and Some Properties of the Forssman Hapten form Sheep Erythrocytes. Arch. Biochem. Biophys. 21, 94 (1955). Siddiqui, B and Hakomori, S-I: A Revised Structure for the Forssman Glycolipid Hapten. J. Biol. Chem. 246, 5766 (1971). Sung, SS-J, Esselman, WJ and Sweeley, CC: Structure of a Pentahexosyl- ceramide (Forssman Hapten) from Canine Intestine and Kidney. J. Biol. Chem. 248, 6528 (1973). Taketomi, T, Hara, A, Kawamura, N and Hayashi, M: Further Investi- gations on Chemical-Structure of Forssman Globoside Obtained from Caprine Erythrocyte Stroma. J. Biochem. (Tokyo) 12, 197 (1974). Kijimoto, S, Ishibashi, T and Makita, A: Biosynthesis of Forssman Hapten form Globoside by afingcetylgalactosaminyltransferase of Guinea Pig Tissues. Biochem. Biophys. Res. Comm..22, 177 (1974). Schiff, F and Adelberger, L: und Antigene. I. Mitteilung. 335 (1925). Ueber Blutgruppenspecifische Antikorper Z. Immunitaetsforsch. Exp. Ther. 22, Buchbinder, L: Heterophile Phenoma in Immunology. Arch. Pathol. 12, 841 (1935). Hakomori, S-I, Stellner, K and watanabe, K: Four Antigenic Varients of Blood Group A Glycolipid: Examples of Highly Complex, Branched Chain Glycolipid of Animal Cell Membrane. Biochem. Biophys. Res. Comm. 22, 1061 (1972). Koécielak, J, Miller-Podraza, H, Krauze, R and Piasek, A: Isolation and Characterization of Poly(g1ycosyl)ceramides (Megaloglycolipids) 252. 253. 254. 255. 256. 257. 258. 259. 260. 261. 262. 263. 264. £265, 214 with A, H and I Blood-Group Activities. (1976). Eur. J. Biochem. 11, 9 Kijimoto, S and Hakomori, S-I: Contact-Dependent Enhancement of Net Synthesis of Forssman Glycolipid Antigen and Hematoside in NIL-Cells at the Early Stage of Cell-to-Cell Contact. FEBS Letters 22, 38 (1972). Whitehead, JS, Bella, A, Jr and Kim YS: An a-N-Acetylgalactosaminyl- transferase from Human Blood Group A Plasma I. Purification and Agar- ose Binding Properties. J. Biol. Chem. 249, 3442 (1974). Whitehead, JS, Bella, A, Jr and Kim YS: An a-N—Acetylgalactosaminyl- transferase from Human Blood Group A Plasma 11. Kinetic and Physica- chemical Properties. J. Biol. Chem. 249 3448 (1974). Nagai, M, Dave, V, Kaplan, BE and Yoshida, A: Human Blood Group Glyco- syltransferases I. Purification of EyAcetylgalactosaminyltransferase. J. Biol. Chem. 253, 377 (1978). Schwyzer, M and Hill, RL: Porcine A Blood Group-Specific ngcetyl- galactosaminyltransferase, I. Purification from Porcine Submaxillary Glands. J. Biol. Chem. 252, 2338 (1977). Schwyzer, M and Hill, RL: Porcine A Blood Group-Specific EfAcetyl- galactosaminyltransferase II. Enzymatic Properties. J. Biol. Chem. 252, 2346 (1977). Kay, HEM and Wallace, DM: A and B Antigens of Tumors Arising from Urinary Epithelium. J. US Natl. Cancer Inst. 26 1349 (1961). Davidsohn, I, Kovarik, S and Lee, CL: A,B, and O Substances in Gas- trointestinal Carcinoma. Arch. Pathol. 21, 381 (1966). Dabelsteen, E and Dindborg, JJ: Loss of Epithelial Blood-Group Sub- stance-A in Oral Carcinomas. Acta Pathol. Microbiol. Scand. (Section A) 21, 435 (1973). Davidsohn, I: Early Immunologic Diagnosis and Prognosis of Carcinoma. Amer. J. Clin. Pathol. 21, 715 (1972). Yamamoto, H: Studies on Enzymes which Decompose Blood Group Sub- stances. 30. A-Decomposing Enzymes Derived from Human and Pig Livers. Med. Biol. 12, 286 (1968). Tallman, JF, Pentchev, PG and Brady, RO: An Enzymological Approach to the Lipidoses. Enzyme 12, 136 (1974). Mapes, CA: Studies on the a-Galactosidases of Normal and Fabry Plasma. Ph.D. Thesis (Michigan State University, 1972). Sung, S-SJ and Sweeley, CC: Purification and Partial Characterization of a-N-Acetylgalactosaminidase from Porcine Liver. In Current Trends in Sphingolipidoses and Allied Disorders, (BW Volk and L Schneck, Eds), Plenum Press (New York, 1976) p 323. 266. 267. 268. 269. 270. 271. 272. 273. 274. 275. 276. 277. 278. 279. 280. 281, 215 Suzuki, C, Makita, A and Yosizawa, Z: Glycolipids Isolated from Por- cine Intestine. Arch. Biochem. Biophys. 127, 140 (1968). Svennerholm, L: Quantitative Estimation of Sialic Acids II. A Color- imetric Resorcinal-Hydrochloric Acid Method. Biochim. Biophys. Acta 22, 604 (1957). Folch, J, Lees, M and Stanley, GHS: A Simple Method for the Isolation and Purification of Total Lipids from Animal Tissues. J. Biol. Chem. 226, 497 (1957). Suzuki, Y and Suzuki, K: Specific Radioactive Labeling of Terminal EfAcetylgalactosamine of Glycosphingolipids by the Galactose Oxidase- Sodium Borohydride Method. J. Lipid Res. 12, 687 (1972). Wiegandt, H: Structure and Function of Gangliosides. Angew. Chem. Intern. Ed. 1, 87 (1968). Wiegandt, H and Baschang, G: Die Gewinnung des Zuckeranteilles der Glykosphingolipide durch Ozonolyse und Fragmentierung. Z. Naturforschg. 20b, 164 (1965). . Woods, GF and Kramer, DN: Dihydropropane Addition Products. J. Amer. Chem. Soc. 22, 2246 (1947). Wallenfels, K and Lehmann, J: Synthese Patentieller Substrate der: B-Galactasidase. Ann. 635, 166 (1960). Conchie, J, Levvy, GA and Marsh, CA: Methyl and Phenyl Glycosides of the Common Sugars. In Advances in Carbohydrate Chemistry,_Val. 12, (ML Wolfram and RS Tipson, Eds), Academic Press (New York, 1957) p 157. Gomori, G: Preparation of Buffers for Use in Enzyme Studies. In Methods in Enzymology, Vol. I, General Preparative Procedures, (SP Colowick and NO Kaplan, Eds), Academic Press (New York, 1955) p 138. Bligh, EG and Dyer, WJ: A Rapid Method of Total Lipid Extraction and Purification. Can. J. Biochem. Physiol.'21, 911 (1959). Lowry, OH, Rasebrough, NJ, Farr, AL and Randall, RJ: Protein Measure- ment with the Folin Phenol Reagent. J. Biol. Chem. 193, 265 (1951). Wilkinson, GN: Statistical Estimations in Enzyme Kinetics. Biochem. J. 22, 324 (1961). Vesterberg, 0: Isoelectric Focusing of Proteins. In Methods in En- zymology, Vol. XXII, Enzyme Purification and Related Techniques, (WB Jakoby, Ed), Academic Press (New York, 1971) p 389. Wrigley, CW: Analytical Disc Gel Electrophoresis. In Methods in En- gyma1ggy, Val. XXII, Enzymegpurification and Related Techniques, (WB Jakoby, Ed), Academic Press (New York, 1971) p 565. Gabriel, 0: Locating Enzymes on Gels. In Methods in Enzymology,,Vol. 282. 283. 284. 285. 286. 287. 288. 289. 290. 291. 292. 216 XXII, Enzyme Purification and Related Techniques, (WB Jakoby, Ed), Academic Press (New York, 1971) p 578. Malik, N and Berrie, A: New Stain Fixative for Proteins Separated by Gel Isoelectric Focusing on Coomassie Brilliant Blue. Analyt. Biochem.‘22, 173 (1972). Gatt, S, Barenholz, Y, Borkovski-Kubilu, I and Leibovitz-Ben Gershon, Z: Interaction of Enzymes with Lipid Substrates. In Sphingolipids, Sphingolip1doses and Allied Disorders, (BW Vold and SM Aronson, Eds), Plenum Press (New York, 1972) p 237. The Merck Index, 9th Ed., (M Windholz, Ed), Merck and Company, Inc. (Rahway, 1976) p 8852. Segal, 1H: Enzyme Kinetics, John Wiley and Sons (New York, 1975) p 371. Fung, Y-K and Sweeley, CC: Interaction of Sodium Taurocholate with Canine Liver ajfifAcetylgalactosaminidase. Fed. Proc. 21, 1766 (1978). Chester, MA, Hultberg, B and Ockerman, P-A: The Common Identity of Five Glycosidases in Human Liver. Biochim. Biophys. Acta 429, 517 (1976). Wallenfels, K and Weil, R: B-Galactosidase. In The Enzymes,gVol. VII, 3rd Ed., Academic Press (New York, 1972) p 617. Sung, S-SJ: Structure of Dog Intestinal Forssman Hapten and Purifica- tion and Partial Characterization of Forssman Hapten Hydrolase (deg: Acetylgalactosaminidase, EC 3.2.1.49) From Porcine Liver. Ph.D. Thesis (Michigan State University, 1977). Desnick, RJ, Dean, KJ, Grabawski, GA and Sweeley, CC: Enzyme Therapy: Differential Metabolic Effectiveness of a-Galactosidase A Isozymes in Fabry's Disease. (in preparation). Henderson, LA, Brown, TL and Thorpe, SR: Mannose-Dependent Uptake of Circulating Ribonucleas-B by Non-Parenchymal Cells in Liver. ’Fed. Proc. 21, 1501 (1978). Yamashina, I and Kawasaki, T: Isolation and Biological Properties of Mannan-Binding Proteins. Fed. Proc. 21, 1501 (1978). APPENDIX APPENDIX List of Publications Dean, KJ, Sung, S-SJ and Sweeley, CC: Purification and Partial Character- ization of Human Liver a—Galactosidases: Is a-Galactosidase B an “IE? Acetylgalactosaminidase? Fed. Proc. 22, 731 (1977). Dean, KJ, Sung, S-SJ and Sweeley, CC: The Identification of a-Galacto- sidase B from Human Liver as an ajngcetylgalactosaminidase. Biochem. Biophys. Res. Comm..11, 1411 (1977). Dean, KJ and Sweeley, CC: Fabry Disease. In Practical Enzymology of the Sphingolipidoses, (RH Glew and SP Peters, Eds), Alan R. Liss, Inc., (New York, 1977) p. 173. Dean, KJ, Sung, S-SJ and Sweeley, CC: Purification and Partial Character- ization of Human Liver a-Galactosidases: Is a-Galactosidase B an “TE? Acetylgalactosaminidase? In Enzymes of Lipid Metabolism, (P Mandel, L Freysz and S Gatt, Eds), Plenum Publishing Corp. (New York, 1978) p. 515. 217