UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Spin labeling and analysis of erythrocyte surfaces Snoek, Robert 1985

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1985_A1 S66.pdf [ 11.52MB ]
Metadata
JSON: 831-1.0060492.json
JSON-LD: 831-1.0060492-ld.json
RDF/XML (Pretty): 831-1.0060492-rdf.xml
RDF/JSON: 831-1.0060492-rdf.json
Turtle: 831-1.0060492-turtle.txt
N-Triples: 831-1.0060492-rdf-ntriples.txt
Original Record: 831-1.0060492-source.json
Full Text
831-1.0060492-fulltext.txt
Citation
831-1.0060492.ris

Full Text

SPIN LABELING AND ANALYSIS OF ERYTHROCYTE SURFACES by ROBERT SNOEK B. SC., U n i v e r s i t y of B r i t i s h Columbia, 1976 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE 'OF DOCTOR OF PHILOSOPHY i n THE FACULTY OF GRADUATE STUDIES (Department of Chemistry) We accept t h i s t h e s i s as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA February, 1985 © Robert Snoek, 1985 3? In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t f r e e l y available for reference and study. I further agree that permission for extensive copying of t h i s thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. I t i s understood that copying or publication of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. Department of The University of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date DE-6 (3/81) - i i -ABSTRACT Spin labeling the oligosaccharides of the red c e l l membrane was achieved via selective oxidation of gal/galNAc (with galactose oxidase) or s i a l i c acid residues (with mild periodic acid) followed by reductive amination of the oxidized sugars with NaBH^CN and TEMPAMINE. Spin labeling the galactose residues resulted i n low yields and s p e c i f i c i t y , hindering analysis of the spin labeled c e l l s (SL-RBC). Higher s p e c i f i c i t y and yields were obtained by labeling s i a l i c acids. A protocol was devised which gave maximum yields with no Heisenberg exchange or membrane alterations (as detected by gel electrophoresis). Detailed analysis of the product showed the majority of the spins to be on the PAS positive membrane proteins (glycophorin A, B and C), only 8% being associated with the l i p i d s . Isolation of glycophorin A, the major sialoglycoprotein of the red c e l l membrane, revealed two modified s i a l i c acids per molecule. Successful ESR interpretations could only be done by lysing the SL-RBC (producing SL-ghosts), eliminating spins which had become internalized (rather than covalently attached to the surface) during the reductive amination step. Assuming a random d i s t r i b u t i o n of b i r a d i c a l s (since there were two spins per glycophorin), an average separation of 16 _ 2 angstroms was calculated between the nitroxides. The spin labeled s i a l i c acids exhibited r e l a t i v e l y mobile spectra with T = 9 x lO-"*"0 s. Upon addition of wheat germ agglutinin (WGA), a - i i i -l e c t i n known to bind to glycophorin, the mobility of the spin l a b e l decreased. Even though WGA binding to SL-ghosts showed complex behaviour as detected by Scatchard plots (which required compensation for WGA impurities and non-specific binding), the ESR was only sensitive to the s p e c i f i c binding, the spin mobility decreasing with increasing WGA. The fact that the spin probe was monitoring s i a l i c acids interactions was confirmed by addition of other l e c t i n s . Only l e c t i n s which interact with glycophorin altered the ESR s i g n a l . - i v -TABLE OF CONTENTS INTRODUCTION 1 CHAPTER 1 BACKGROUND 1.1 The Red Blood C e l l 5 1.1(a) The Red C e l l i n vivo 5 1.1(b) History of the C e l l Membrane Studies ... 8 1.1(c) The Red Blood C e l l Membrane 11 1.1(c)! Lipids 12 l . l ( c ) i i Proteins 14 The cytoskeleton 17 Catalytic functions, Band 3 21 Band 4.5 23 Contact or receptor proteins 24 Minor components of the membrane 27 1.2 Lectins 34 1.3 Chemical Modification 36 1.3(a) NANA Oxidation 36 1.3(b) Galactose Oxidation 37 1.3(c) Reductive Amination 38 1.4 Electron Spin Resonance 40 1.4(a) Nitroxides 40 1.4(b) Electron Spin Resonance:General 41 1.4(c) Nitroxide ESR 42 1.4(d) A r t i f a c t s 51 1.4(e) Advantages 53 CHAPTER 2. SPIN LABELING THE GALACTOSE RESIDUES OF THE RED CELL SURFACE 2.1 Introduction 54 2.2 Materials and Methods 55 2.2(a) Collection of Red Blood C e l l s 55 2.2(b) Galactose Oxidase Oxidation of Red Cells 55 2.2(c) Spin Labeling 56 2.2(d) Ghost Preparation 56 2.2(e) Microelectrophoresis 57 2.2(f) Viscometry 57 2.2(g) Electron Spin Resonance 58 2.2 (h) Quantitation ( i ) Red C e l l or Ghost Count 58 ( i i ) ESR 59 ( i i i ) Protein 59 2.2(i) Miscellaneous 60 - V -2.3 Results 2.3(a) Quantitation ( i ) Ghost Counts 61 ( i i ) Protein Quantitation of Ghosts 63 ( i i i ) ESR Integration 64 2.3(b) Galactose Spin Labeling of Red Cells ... 65 2.4 Discussion ( i ) Red C e l l s (a) Microelectrophoesis 77 (b) Spin Labeling 78 ( i i ) Tc ° f labeled ghost 83 2.5 Conclusion 85 CHAPTER 3 SPIN LABELING THE SIALIC ACID RESIDUES OF THE RED CELL SURFACE 3.1 Introduction 86 3.2 Materials and Methods 3.2(a) Periodate Oxidation of Red Cell s and Ghosts 89 3.2(b) Periodate Oxidation of Glycophorin and Fetuin 89 3.2(c) Spin Labeling of Red C e l l s 90 3.2(d) Spin Labeling Ghosts by the Method of F e l i x and B u t t e r f i e l d (1980) 91 3.2(e) Formaldehyde Assay for Red Cells 91 3.2(f) S i a l i c Acid Assays 92 3.2(g) Rouleaux 92 3.2(h) ESR 92 3.2(i) Measurements of the Distance Between Nitroxide Labels 93 3.2(j) R e v e r s i b i l i t y 93 3.2(k) Hydrolysis of the Spin Labeled Ghosts .. 94 3.2(1) Isolations 94 3.2(m) Tritium Analysis 95 3.3 Results 3.3(a) Rouleaux v_ Periodate Modification 96 3.3(b) Formaldehyde Formation 99 3.3(c) T r i t i a t e d Glycophorin 102 3.3(d) SDS PAGE Analysis 104 3.3(e) Spin Labeling Red Cell s 104 3.3(f) Background Signal 110 3.3(g) Visual Inspection of Oxidized and Spin Labeled Cel l s 110 3.3(h) Spin Labeled Ghosts 112 3.3(1) F-B SL-ghosts 112 3.3(j) Distance Measurements 115 3.3(k) R e v e r s i b i l i t y 115 - v i -3.3(1) Neuraminidase or Acid Hydrolysis of SL-ghosts 116 3.3(m) L i p i d Extraction 118 3.3(n) Selective S o l u b i l i z a t i o n 119 3.3(o) Isolation of Glycophorin 121 3.3(p) Tritium Distribution of SDS PAGE gels... 122 3.4 Discussion 125 3.4(a) Periodate Oxidation 125 3.4(b) Spin Labeling 128 3.4(c) Quantitation 129 3.4(d) Location 130 3.4(e) ESR Interpetation 133 3.4(f) Distance between Nitroxides 135 3.4(g) F-B SL-ghosts 137 3.5 Conclusion 139 CHAPTER 4 WHEAT GERM AGGLUTININ PURIFICATION AND BINDING 4.1 Introduction 141 4.2 Materials and Methods 4.2(a) Wheat Germ Agglutinin 145 4.2(b) Protein Assays 145 4.2(c) Agglutination Assay 145 4.2(d) Iodination 146 4.2(e) SDS PAGE -. 148 4.2(f) Spin Labeled Ghosts and Extractions 148 4.2(g) Binding Assay 149 4.2(h) Other Lectins 150 4.3 Results 4.3(a) WGA Addition to Extractions 151 4.3(b) WGA Plus Various SL-Ghosts 153 4.3(c) Binding Assay 156 4.3(d) Other Lectins 164 4.4 Discussion 4.4(a) WGA broadening of the ESR signal 166 4.4(b) Mechanism of Broadening 169 4.4(c) WGA Binding Isotherm 170 4.4(d) WGA Receptor on Red Cells 173 4.4(e) Correlation between SP and WGA Binding . 176 4.4(f) Binding, of Other Lectins 178 4.5 Conclusion 179 4.6 Summarizing Discussion 180 APPENDIX A SDS PAGE 181 APPENDIX B ESR Integration 184 - v i i -APPENDIX C PNA:fetuin 189 C.l(a) Fetuin Modification 189 C. l(b) Peanut Agglutinin A f f i n i t y Column 190 C.2 Results PNA:Fetuin interaction 190 C. 3 Discussion 191 APPENDIX D WGA P u r i f i c a t i o n D. l ( a ) Fast Protein Liquid Chromatography 194 D.l(b) CM-sepharose 194 D.l(c) C h i t i n Column 195 D.l(d) Ovomucoid Column 195 D. 2 Results D.2(a) SDS PAGE 196 0.2(b) A f f i n i t y Column 199 D.2(c) Cation Exchange Chromatography 201 D.3 Discussion 204 REFERENCES Introduction 208 Chapter 1 209 Chapter 2 219 Chapter 3 222 Chapter 4 227 Appendix A 231 Appendix C 231 - v i i i -LIST OF TABLES TABLE CHAPTER 1 1.1 Composition of the red c e l l membrane 11 1.2 L i p i d composition of the red c e l l membrane 12 1.3 Properties, associations and function of the red c e l l membrane fractions 20 1.4 Enzymes found i n the red c e l l membrane 28 CHAPTER 2 2.1 Ghost counting method and grams/ghost obtained 63 2.2 C e l l electrophoresis of GO or NANase treated and spin labeled red c e l l s 66 2.3 SP for NAGO SL-RBC with the addition of gal binding l e c t i n s 67 2.4 SP at different hematocrits of RBC plus 1 mM TEMPAMINE . 69 2.5 Sample manipulation of SL-RBC and resultant SP 70 2.6 Spin labeled samples made into ghosts and th e i r SP 70 CHAPTER 3 3.1 Formaldehyde formation as a function of periodate concentration and exposure time to red c e l l s 99 3.2 Formaldehyde formation as a function of periodate concentration and exposure to ghosts 100 3.3 Formaldehyde formation as a function of periodate concentration and exposure to glycophorin and fetuin ... 102 3.4 I n i t i a l TEMPAMINE concentration and the resultant spins per SL-ghost 106 3.5 Other attempts to improve spins/SL-ghost 107 3.6 Periodate concentration along with spins/SL-ghost and SP in comparison to optimal protocol 108 3.7 Spins/SL-ghost and resultant w0 112 3.8 dj/d of SL-ghosts and 1:4 and 1:8 SLAN ghosts 115 3.9 Percentage release of t r i t i u m and spins from SL-ghosts with days incubated at 4°C 116 3.10 Percentage release of t r i t i u m , spins and s i a l i c acid compared to neuraminidase treatment or acid hydrolysis of SL-ghosts or F-B SL-ghosts 117 3.11 Percentage extraction of spins and t r i t i u m by chloroform:methanol of SL-ghosts and F-B SL-ghosts 118 3.12 Percentage extraction of spins and t r i t i u m along with thei r SP of. SL-ghosts by NaOH, SDS and Triton X-100 119 3.14 Distribution of t r i t i u m on SDS PAGE of various samples . 124 - i x -CHAPTER A A.l SP of various selective extractions of SL-ghosts after the addition of WGA 1A1 A.2 SP after l e c t i n added to SL-ghosts 16A APPENDIX D D.l Percentage binding of iodinated impure Sigma WGA to c h i t i n and ovomucoid columns 199 D.2 CM-Sepharose chromatography of commercial WGA 202 D.3 FPLC chromatography of Vector and Sigma WGA 202 D.A Amino acid composition of WGA from various l i t e r a t u r e sources 206 - X -LIST OF FIGURES FIGURE CHAPTER 1 1.1 Chronological order of proposed membrane models 10 1.2 SDS PAGE of red c e l l membranes separated by Fairbanks et a l 1971 and Laemmli, 1970 16 1.3 Nomenclature for the PAS staining bands of red c e l l membranes 16 1.4 Schematic drawing of the red c e l l membrane skeleton unit c e l l 19 1.5 Proposed organization of band 3 i n the c e l l membrane ... 22 1.6 Amino acid sequence of glycophorin A and i t s arrangement i n the red c e l l membrane 26 1.7 Diagrammatic representation of the PAS proteins i n the red c e l l membrane 26 1.8 Structures of the oligosaccharides found on band 3 and glycophorin A 32 1.9 Cartoon of the outer surface of the red c e l l membrane .. 33 1.10 Oxidation of s i a l i c acid by periodate 37 1.11 Oxidation of galactose termini by galactose oxidase 38 1.12 The structure of TEMP AMINE 39 1.13 Reductive amination of an aldehyde with TEMPAMINE 39 1.14 Effect of the magnetic f i e l d upon the unpaired electron and the absorption of microwaves and the resultant ESR spectrum 42 1.15 The effect of the nuclear spin state of N on the unpaired electron 43 1.16 (A) The p r i n c i p a l axis for the nitroxide (B) Direction dependence of the Zeeman and the hyperfine interactions of the nitroxide 45 1.17 Effect of vi s c o s i t y on the ESR spectra of ni'troxides ... 47 1.18 Powder spectrum showing the heights d_ and d used i n calculating average distances between nitroxides 51 CHAPTER 2 2.1 Percentage of the red c e l l or ghost population and the trigger levels of the p a r t i c l e counter 62 2.2 ESR spectra of control RBC and NAGO SL-RBC plus SBA 68 2.3 ESR spectrum of GO SL-ghosts 72 2.4 SDS PAGE gels of normal and GO SL-ghosts 73 2.5 ESR spectra of NAGO SL-ghosts before and after addition of PNA and SBA 74 2.6 C e l l viscometry of normal and NAGO SL-RBC with the addition of PNA 76 - x i -CHAPTER 3 3.1 Pictures of red c e l l s or periodate treated red c e l l s i n 80% plasma 97 3.2 Pictures of red c e l l s or periodate treated red c e l l s i n 3% dextran 98 3.3 SDS PAGE gels of periodate oxidized ghosts 101 3.4 SDS PAGE gels of glycophorin and t r i t i u m labeled glycophorin 103 3.5 SDS PAGE gels of various red c e l l samples, oxidized and spin labeled. Stained with PAS or Basic Fuchsin ... 105 3.6 SDS PAGE gels of normal, SL-ghosts and double the periodate spin labeled ghosts 109 3.7 ESR spectra of SL-RBC and controls before and after l y s i s I l l 3.8 ESR spectrum of SL-ghosts showing parameters measured to determine SP .. 113 3.9 SDS PAGE gels of SL-ghosts and F-B SL-ghosts 114 3.10 SDS PAGE gels of the NaOH p e l l e t and the Triton X-100 pe l l e t and extract of SL-ghosts 120 3.11 SDS PAGE gels of isolated SL-glycophorin and SL-ghost .. 123 CHAPTER 4 4.1 Schematic i l l u s t r a t i o n of the disposition of the primary and secondary binding locations on the WGA dimer 142 4.2 ESR spectra of the Triton X-100 extract of SL-ghosts before and after the addition of WGA 152 4.3 ESR spectra of SL-glycophorin isolated from SL-ghosts before and after the addition of WGA 154 4.4 Graph of SP of SL-ghosts after addition of WGA (from various commercial sources) compared with the i n i t i a l WGA added and SP for the binding experiment with impure Sigma WGA of Figs. 4.12 and 4.15a 155 4.5 ESR spectra of SL-ghosts and WGA and the resultant supernatant and p e l l e t after centrifugation 157 4.6 ESR spectra of F-B SL-ghosts before and after addition of WGA 158 4.7 Binding isotherm of 1 2 5 I WGA to SL-ghosts for the t o t a l and non-specific binding of WGA 159 4.8 Binding isotherm of 1 2 5 I WGA to SL-ghosts for s p e c i f i c binding 160 4.9 Scatchard plot of s p e c i f i c binding of Fig. 4.13 161 4.10 SP and binding of 1 2 5 I WGA/SL.ghost for impure Sigma and pure Vector 1 2 5 I WGA 162 4.11 ESR spectra of SL-ghosts (A) plus WGA (C) and 2 [0] SL-ghosts (B) plus WGA (D) along with t h e i r calculated SP's 163 - x i i -4.12 ESR spectra of SL-ghosts (A) plus WGA (C). (A) relysed (B) and WGA (D) along with t h e i r calculated SP's 165 APPENDIX B B.l ESR spectra before and after being d i g i t i z e d along with the resultant f i r s t integration (also rezeroed) and the second integration of the rezeroed f i r s t integration .. 185 B.2 Placement of ESR spectrum on the pl o t t e r bed and the location of P i and P2 187 APPENDIX D D.l SDS PAGE gels of reduced impure 1 2 5 1 Sigma WGA and pure 1 2 5 I Vector WGA 197 D.2 SDS PAGE gels of impure 1 2 5 I Sigma WGA before and after exposure to red c e l l s 198 D.3 SDS PAGE gels of reduced 1 2 5 I Sigma WGA fractions run on ovomucoid and c h i t i n columns 200 D.4 SDS PAGE gels of reduced Vector WGA run on FPLC and CM-sepharose columns 203 - x i i i -ABBREVIATIONS o AChe ATP ATPase B BSA CHO c e l l s CM-sepharose Con A cpm d,d 1 DTT dpm E.C. (followed by a number) EDTA EPR ESR F-B SL-ghosts FPLC G i s o t r o p i c hyperfine constant acetylchoinesterase adenosine triphosphate adenosine triphosphatase Bohr magneton bovine serum albumin Chinese Hamster Ovary c e l l s carboxymethyl-sepharose Concanavalin A counts per minute peak-to-peak i n t e n s i t i e s i n powder spectra as defined i n Figure 1.18 d i t h i o t h r e i t o l d i s t i g r a t i o n s per minute Enzyme Commission, followed by numbers indicating c l a s s i f i c a t i o n of the enzyme ethylenediamine tetraacetate electron paramagnetic resonance electron spin resonance SL-ghosts made by the method of Fel i x & Bu t t e r f i e l d fast protein l i q u i d chromatography gauss - xiv -g x x ' 9 y y ' g z z p r i n c i p a l values of the g-tensor Q|j- ,g^ p a r a l l e l and perpendicular components of an a x i a l l y symmetric g-tensor gal galactose galNAc N-acetylgalactosamine G-3-PD glyceraldehyde-3-phosphate dehydrogenase GlcNAc N-acetyl glucosamine GlcNH 2 glucosamine GO galactose oxidase GO SL-RBC galactose oxidized and spin labeled red c e l l s GO SL-ghosts GO SL-RBC made into ghosts H externally applied magnetic f i e l d h Peak-to-peak height of a f i r s t derivative ESR signal h^g + J J peak-to-peak height for the three nitroxide resonances corresponding to = 0,+1 Hb hemoglobin Hz . Hertz IS integrator s e n s i t i v i t y K degrees Kelvin LIS lithium d i i o d o s a l i c y l a t e mono S mono sulfate M molar nij magnetic quantum number associated with the component of the nuclear spin - XV -rrig t magnetic quantum number associated with the component of the electron spin mW mi l l i w a t t s MW molecular weight n number of experiments NADH nicotinamide adenine dinucleotide hydride NAGO SL-fetuin neuraminidase and galactose oxidase treated and spin labeled fetuin NAGO SL-RBC neuraminidase, then galactose oxidase treated and spin labeled red c e l l s NAGO SL-ghosts NAGO SL-RBC made into ghosts NANA N-acetyl neuraminic acid NANA., _ N-acetyl neraminic acid with the C Q (NANA_) or / or o y o the C 9 a n c ) 8 carbons (NANA-,) removed NANase neuraminidase NH2-SL TEMPAMINE OD o p t i c a l density [ 0 ] RBC periodate oxidized red c e l l s 2 [ 0 ] SL-ghosts SL-ghosts made at twice the normal periodate concentration OSM ovum submaxillary mucin PAGE polyacrylamide gel electrophoesis PAS periodic acid Schiff base PBS phosphate buffered saline - xvi -PBS/azide phosphate buffered saline containing azide PC phosthatidylcholine PE phosphatidylethanolamine p i i s o e l e c t i c point PNA peanut agglutinin PPO 2,5 diphenyloxazole PS phosphatidylserine r mean nearest-neighbour distance between nitroxides RBC red blood c e l l RCA Ricin communis agglutinin RNA ribonucleic acid SBA soybean agglutinin l e c t i n SD standard deviation SDS sodium dodecyl sulfate SL spin l a b e l SLAN spin l a b e l analogue SL-ghosts periodate oxidized and spin labeled red c e l l s made into ghosts SL-RBC periodate oxidized and spin labeled red c e l l s T temperature T ,T ,T p r i n c i p a l values of the hyperfine tensor xx yy zz T correlation time lc , T j ^ motion about the p r i n c i p a l and molecular axis of a nitroxide - x v i i -t r i c h l o r o a c e t i c acid tracking dye N,N,N',N 1-tetramethylethylenediamine units u l t r a v i o l e t microwave frequency neuraminidase Vibrio cholerae volume per volume linewidth of the center l i n e of an nitroxide ESR signal wheat germ agglutinin weight per volume weight per weight - x v i i i -ACKNOWLEDGEMENT This project was greatly aided by the assistance of numerous people, among these are Foon Yip, who aided i n the ESR and l i p i d work, Mandy Hoskins for doing the c e l l viscometry work and John Cavanagh, i n the i s o l a t i o n of glycophorin. I would also l i k e to thank John Cavanagh and Charlie Ramey for t h e i r general help. Stimulating discussions are acknowledged with Dr. Geoffrey Herring, who also provided access to his ESR machinery and commented l i b e r a l l y on the manuscript as did Dr. Manssur Yalpani. My thanks are also due to Drs. P h i l Reid, Evan Evans, John Waterton and Mike Bernstein for advice and assistance. A great debt i s owed to my wife Sandra Sturgeon, for emotional support and c r i t i c a l reading of t h i s thesis among other things. Thanks are also due to people such as Kim Sharp, Johan Janzen, Tim Webber, Jim Van Alstine and B a s i l Chui for t h e i r discussions and comments. My greatest debt of thanks i s to Drs. Don Brooks and Laurie H a l l , who agreed to undertake t h i s combined research and whose help and encouragement made the research described i n t h i s thesis worthwhile. - 1 -INTRODUCTION Chemistry i s a science which deals with isolated systems i n a controlled, well defined manner. Molecules can be assigned three-dimensional structures, reactions are known i n great d e t a i l . Analysis at the atomic l e v e l can be eas i l y achieved, giving the chemist an extremely detailed picture of the system of int e r e s t . Biology, on the other hand, i s much less c l e a r l y defined at the molecular and atomic l e v e l . Many interactions of varying complexity occur at the same time. Some of these interactions are unknown and i t i s hard to is o l a t e these to study i n d i v i d u a l components. The b i o l o g i s t usually can't determine the cause of an event, but can only i n f e r . The question asked i n t h i s thesis i s can techniques, which have been used to analyse complex, isolated systems such as carbohydrates, peptides, oligomers etc., be extended and applied to less well defined b i o l o g i c a l systems such as c e l l membrane l e c t i n interactions and s t i l l be useful? Can chemical analysis be applied and give answers which are meaningful to biologists? Of particular interest i s the chemical and biochemical function of the plasma membrane. C e l l surfaces play a c r i t i c a l role i n the response of c e l l s to the external environment, which may contain drugs, hormones, antigens, toxins or infectious agents and may result i n agglutination, adhesion or m o t i l i t y (Hughes, 1976; Monsigny, 1979; Rauvala, 1983). The carbohydrate-containing substances, g l y c o l i p i d s and glycoproteins, are good candidates for p a r t i c i p a t i n g i n i n t e r c e l l u l a r recognition and i n the binding of regulatory molecules (Hughes, 1975). Studies concerning membrane carbohydrates have shown a l l the - 2 -carbohydrate to be anchored to the external face of the c e l l (Wlnzler, 1972; Gahmberg, 1976). This carbohydrate-rich f o r e s t - l i k e structure on the outer surface i s known as the glycocalyx (Geyen & Makovitzky, 1980) and t y p i c a l l y contains four to eight kinds of sugars (glucose, galactose, mannose, N-acetylglucosamine, N-acetylgalactosamine, fucose and s i a l i c acid (Hughes, 1975; Schrevel e t a l , 1982)). It i s the glycocalyx that i s i n contact with the external environment, be i t plasma, another c e l l surface or some i n v i t r o solution or surface. The possible form of linkages for the hexoses i s enormous. Three amino acids can produce only s i x different tripeptides while 1056 trimers are possible with three d i f f e r e n t hexoses. Biosynthesis of these oligosaccharides i s less exact than for proteins and nucleic acids (being under enzymatic control rather than that of a direct genetic template) and leads to v a r i a b i l i t y of structure known as microheterogeneity. This biosynthesis i s sensitive to e x t r a c e l l u l a r s t i m u l i and the microheterogeneity may r e f l e c t some pathophysiological state of the organism (Hatton e t a l , 1983). A considerable amount of information, environmentally s e n s i t i v e , can be "carried" by the glycocalyx, therefore. The red blood c e l l i s well suited for use i n membrane studies. It i s easy to obtain i n large numbers v i r t u a l l y free of contaminants, a c r i t e r i o n not easily met i n biology. Model recognition systems which react with s p e c i f i c elements of the glycocalyx are also available i n the form of l e c t i n s , proteins which are e a s i l y obtained i n large quantities that bind certain sugar sequences, offering a means of manipulating the system and mimicking recognition behavior i n biology. In the present work therefore, a l e c t i n / r e d c e l l system was selected for - 3 -study by quantitive chemical means, to discover the l i m i t a t i o n s and advantages of such an approach. When modifying a b i o l o g i c a l entity such as a red c e l l , one i s constrained by the media compatible with these systems. Conditions near physiological (pH 7, 280 milliosmole, aqueous) must be employed throughout the procedure i f information i s to be obtained. The next major constraint i s s e l e c t i v i t y and d e t e c t a b i l i t y of the modification. B i o l o g i c a l systems for a l l t h e i r complexity are a l l made up of r e l a t i v e l y few building blocks combined i n a variety of ways. S e l e c t i v i t y i n modification would render interpretation easier but i t i s made d i f f i c u l t by the chemical s i m i l a r i t y of completely different compounds. Also, the more selective one becomes, the lower the degree of incorporation. For example, to se l e c t i v e l y modify the most abundant protein on the red c e l l membrane (Band 3 at 1 x 10 6 copies per c e l l ) at a one to _5 one basis, results i n a concentration of 1.8 x 10 M for packed c e l l s . The reporter group introduced therefore has to be eas i l y detectable as well as report useful information about i t s environment. Spin probes (nitroxides are used i n t h i s study) are the best suited for analysis of membranes. They offer several very important features which overcome the above r e s t r i c t i o n s . They can be detected at low concentrations (down to 10~ 6 M). Natural occurrence of paramagnetism i n b i o l o g i c a l systems i s r e l a t i v e l y low, thus there i s no ambiguity regarding the source of the si g n a l . They are sensitive to molecular motion and environment, making them a good reporter group. The only s i g n i f i c a n t price paid i n using spin probes i s the possible perturbation of the system by introduction of the spin l a b e l . - A -Galactose (and N-Acetyl galactosamine) and s i a l i c acid (sugars commonly contained i n the glycocalyx) can be s e l e c t i v e l y oxidized and a spin probe attached to these oxidized sugars v i a reductive amination, a technique successfully used previously for a variety of macromolecules (Aplin, 1979; Yalpani, 1980; Bernstein, 1983). Thus, i n p r i n c i p l e , s e l e c t i v i t y and s e n s i t i v i t y are obtainable using the spin l a b e l method. A careful study of spin labeling of the red c e l l glycocalyx was thus undertaken to examine the li m i t a t i o n s and usefulness of t h i s technique as an i n s i t u membrane probe. In what follows, to provide the background necessary for interpretation, Chapter one reviews each of the above aspects of t h i s study (chemical modification, ESR, the red c e l l and l e c t i n s ) . Chapter two deals with quantitation and the effect of spin labeling the galactose residues. Chapter three discusses the effects of periodate modification of red c e l l s and compares i t with fetuin (a serum sialoglycoprotein) and glycophorin (the major sialoglycoprotein of the red c e l l membrane) as well as discussing the results of spin labeling the oxidized s i a l i c acids on red c e l l s . Chapter three also describes attempts at i s o l a t i o n of the spin labeled components of the red c e l l membrane. To study a model recognition event, the l e c t i n wheat germ agglutinin (WGA) i s used. Chapter four deals with the binding of WGA to spin labeled ghosts. - 5 -CHAPTER ONE This chapter deals with the approach used to modify the glycocalyx of the red c e l l and gives the background material necessary to understand the experimental results i n the following chapters. Section 1.1 outlines what i s known about red c e l l s with emphasis on the membrane structure and Section 1.2 explains the use of l e c t i n s i n t h i s study. Section 1.3 deals with the chemical modification necessary to insert spin probes covalently into the glycocalyx and Section 1.4 the theory of electron spin resonance needed to understand and interpret spectra. 1.1 THE RED BLOOD CELL An enormous amount of information i s available on the red blood c e l l (RBC) especially since the advent of blood transfusions. A better understanding of the red c e l l results from knowledge of i t s function and production. The f i r s t part of t h i s review b r i e f l y covers what i s known about red c e l l production and function followed by a more extensive overview of i t s membrane structure. 1.1(a) The red c e l l i n vivo Anton van Leeuwenhoek was the f i r s t to see red blood c e l l s , i n 1674, describing them as "small globules driven through a c r i s t a l i n e humidity of water" (Grimes, 1980). In 1685, Bidloo described the red c e l l as "a ballon-like-body i n which f l u i d solutions of hemoglobin were enclosed - 6 -by an external envelope" (Lee et a l , 1974). This description i s surprisingly accurate. Many reviews have been written about the red c e l l (Berlin & Berk, 1975; Brewer, 1975; Tanner, 1978; Hillman & Finch, 1974; Grimes, 1980; Williams e t a l , 1983) so only a b r i e f description i s offered here. The red blood c e l l s (or erythrocytes) are produced from stem c e l l s located i n the bone marrow of adults. Each stem c e l l undergoes a series of mitotic divisions producing sixteen smaller c e l l s c a l l e d orthochromatic normoblasts. Hemoglobin production has already started and the nucleus i s compact. Incapable of mitosis, these normoblasts produce more hemoglobin, eliminate the nucleus, organelles and enzymatic pathways not needed for red c e l l s u r v i v al yielding a c e l l c a l l e d a reticulocyte. The reticulocyte enters the c i r c u l a t i o n and with the loss of residual RNA becomes a red blood c e l l . This whole process of sp e c i a l i z a t i o n takes about seven days. As the c e l l d i f f e r e n t i a t e s and becomes more specialized, organelles and metabolic pathways are eliminated. Protein and l i p i d synthesis stops and what remains i s a discoid shaped c e l l f i l l e d with hemoglobin surrounded by the c e l l membrane. Inside t h i s c e l l are four enzymatic pathways. The major pathway ( u t i l i z i n g 90% of the glucose) i s the Embdem-Meyerhof pathway, producing ATP and NADH (which are used i n the maintenance and f l e x i b i l i t y of the c e l l ) and lactat e . The other three pathways are used to maintain the function of hemoglobin and, due to the high levels of 0 2, to maintain oxidative, reductive homeostasis. Hemoglobin i s regulated by a variety of organic phosphates, pH and C0 2 l e v e l s . Carbonic anhydrase, the second most abundant protein, f a c i l i t a t e s C0 2 transport. In the membrane are found transport proteins for glucose, lactate and HCOZ with Na +/K + - 7 -and Ca +/Mg + ATPase pumps to maintain concentration gradients. These metabolic pathways ensure red c e l l s u r v i v al and e f f i c i e n t 0^ transfer. There are other enzymes i n the c e l l but at much lower concentrations and some with no known function. The red c e l l travels about 175 miles during i t s l i f e t i m e of 120 days. I t passes through blood vessels of varying sizes and i s exposed to a wide range of shear stresses, spending most of i t s l i f e i n the c a p i l l a r y channels of the microcirculation, sometimes squeezing through openings 1/20^ i t s diameter. As the c e l l ages i t s p l i a b i l i t y decreases, membrane i s l o s t , the mean hemoglobin concentration increases and certain g l y c o l y t i c enzyme a c t i v i t i e s decrease. The reticuloendothelial tissue ( l i v e r , spleen and bone marrow) are phagocytic towards these old c e l l s . The spleen i s the most sensitive of these organs. I t s network slows down the blood flow, forcing red c e l l s to flow through 3-5 p diameter o r i f i c e s where c e l l membranes and abnormal p a r t i c l e s are removed. The red c e l l i n an adult male has an average volume (mean corpuscular volume (MCV)) of 90.1 jum"5. The mean corpuscular hemoglobin concentration 2 i s 33.9 g/dl RBC. The average surface area i s 145 jum with a diameter from 7.5 to 8.3 pm and a thickness of 2 pm. Hemoglobin constitutes 97.5% of the protein i n red c e l l s with 0.8% being enzymes and 1.7% being membrane proteins. When studying red blood c e l l s , one must be aware that no sample contains only normal human erythrocytes. Three major sources of heterogeneity are genetic v a r i a b i l i t y , especially of many enzymes and hemoglobin (there are at least 500 genetically d i s t i n c t hemoglobins catalogued i n the International Hemoglobin Information Centre), a population v a r i a b i l i t y due to the fact - 8 -that red c e l l s age and a l l stages are contained i n a red c e l l sample, and variations between donors of different age, especially the very young and old. I t might be asked i f the red c e l l i s too specialized to be useful as a representative model but a l l mammalian c e l l s are highly d i f f e r e n t i a t e d . The technical advantages of being able to obtain large, r e l a t i v e l y homogeneous quantities makes i t useful. Due to i t s specialized function, i t has a s t a t i c membrane, so one can study i t without the additional confusion of membrane turnover. The red c e l l i s also f a i r l y stable to a variety of i n v i t r o manipulations, making i t suitable for membrane studies. 1.1(b) History of c e l l membrane studies Early observations by Overton (1895) indicated the c e l l s were surrounded by a semipermeable membrane which was more permeable to nonpolar than polar solutes. Gorter & Grendel (1925), measured the surface area of a monolayer of l i p i d extracted from red c e l l s and found that there was enough l i p i d i n an erythrocyte membrane to cover twice the surface area of the c e l l . They thus proposed a bilayer structure for the l i p i d . Luckily, t h e i r underestimation of the surface area compensated for incomplete extraction of the l i p i d s i n t h e i r experiments. D a n i e l l i & Davson (1935), suggested that the l i p i d bilayer was covered by layer(s) of protein to account for the low surface tension observed. Electron micrographs of fixed membranes resulted i n the Davson-Danielli-Robertson model (Robertson, 1959). C r i t i c i s m of t h i s model arose because one would expect to see beta-sheets of protein instead of the - 9 -alpha helix found. Extracted l i p i d s t i l l gave the same electron micrographs as intact membranes and certain ions which i n theory shouldn't be permeable i n fact were. A huge variety of models was postulated (Fig. 1.1, taken from Finean & M i t c h e l l , 1981). The best accepted one, proposed by Singer & Nicholson i n 1972 i s the " f l u i d mosaic" model. This model agrees best with available data. I t includes alpha helices i n membrane proteins, f l u i d l i p i d s , l i p i d s not d i r e c t l y interacting with membrane proteins and allows for apparent mobility of macromolecules i n the plane of the membrane (Juliano, 1973). The membrane consists of a two dimensional f l u i d matrix where proteins are intercalated into the f l u i d l i p i d b i l a y e r , some a l l the way through, some only p a r t i a l l y , while others rest on the l i p i d surface. Although refinements to the f l u i d mosaic model are s t i l l being published (e.g. I s r a e l a c h v i l i , 1977) most models currently being presented are detailed studies of in d i v i d u a l membranes rather than general structures (e.g. the l a s t model proposed i n Fig. 1.1). 1975 odopted from Henderion t Unwin Fig. 1.1 Diagram i l l u s t r a t i n g the chronological order i n which the most i n f l u e n t i a l models have been proposed. Each of the structures indicated are described i n more d e t a i l i n the source publication (Finean & M i t c h e l l , 1981). 1.1(c) The red blood c e l l membrane "The red c e l l membrane i s giving up some of i t s secrets, but only slowly and, seemingly, grudgingly" (Brewer et a l , 1982). One of the major breakthroughs i n red c e l l membrane analysis came when Dodge et a l (1963) were able to lyse red c e l l s free of cytoplasmic contaminants. These membrane preparations are call e d ghosts because they are d i f f i c u l t to see unless looked at with a phase contrast microscope. Unlike previous methods, that of Dodge et a l (1963) gave reproducible ghosts with the major contaminant, hemoglobin, reduced to about 0.04% of the t o t a l weight. These ghosts were extensively analysed; Table 1.1 (taken from Juliano, 1973) l i s t s a t y p i c a l analysis of Dodge ghosts. TABLE 1.1 (taken from Juliano, 1973) COMPOSITION OF THE RED CELL MEMBRANE OF DODGE TYPE GHOSTS Protein g/ghost 6.6 x l O - - ^ % of whole c e l l 1.9 Hemoglobin g/ghost 0.51 x l O - - ^ % of whole c e l l 0.15 Cholesterol g/ghost 1.54 x 10" 1 3 mg/mg protein 0.49 Phospholipids g/ghost 3.2 x lO--*-3 mg/mg protein 0.49 Hexose g/ghost 3.5 x l O - - ^ jjg/mg protein 41 Hexosamine g/ghost 2.7 x 10~^ jjg/mg protein 41 S i a l i c Acid g/ghost 2.2 x 10" 1 4 jug/mg protein 33 Glyc o l i p i d g/ghost 2.9 x 10-14 jug/mg protein 43 - 12 -The membrane i s only 2% of the c e l l dry weight (Grimes, 1980) consisting of approximately 50% protein, 40% l i p i d and 10% carbohydrate by weight (Zwaal et a l , 1973). Recent reviews on the red c e l l membrane include Juliano, 1973, Zwaal et a l , 1973, Marchesi et a l , 1976, Tanner, 1978, Marchesi, 1978 & 1979a, Rothstein et a l , 1979, Grimes, 1980 and Shohet & Beutler, 1983. Analysis of membranes can be broken down into the studies of l i p i d s and proteins and each w i l l be dealt with i n d i v i d u a l l y . l . l ( c ) i Lipids There are approximately 5 mg of l i p i d / 1 0 1 0 c e l l s (Rendi e t a l , 1976) and a l l are located i n the membrane (see Zwaal et a l , 1973 and van Deenen & de Grier, 1974 for reviews). Table 1.2 (taken from Grimes, 1980) shows t y p i c a l values for red c e l l s . Variation i n l i p i d content i s found and i s due mainly to di f f e r e n t i s o l a t i o n techniques (Nelson, 1972). TABLE 1.2 TYPICAL VALUES FOR NORMAL HUMAN RED CELL LIPIDS, EXPRESSED AS % DRY WEIGHT (taken from Grimes, 1980) NEUTRAL GLYCO LIPIDS SPHINGO LIPIDS PHOSPHOLIPJDS lO 2 0 30 4 0 50 60 70 BO 9 0 KX> Cholesterol i Neutral Neutrol Phospholipid* Acidtc Phospholipids ^ r o c +* o i o t S io 3 u= >. a - 13 -The l i p i d i s largely i n a bilayer configuration and serves as a bar r i e r or boundary between the outside and the inside of the red c e l l . The phospholipids are asymmetric i n t h e i r d i s t r i b u t i o n i n the bilayer: 80% of the sphingomyelin, 75% of the phosphatidylcholine (PC) and a l l the g l y c o l i p i d s are i n the outer layer while 100% of the phosphatidylserine (PS) (the only major phospholipid to be charged (negatively) at physiological pH), 80% of the phosphatidylethanolamine (PE) and the majority of the phosphatidylinositol are at the inner or cytoplasmic side of the membrane (Zwaal, 1978; van Deenen, 1981; Eton, 1981). Asymmetry i n the head groups i s accompanied by asymmetry i n the hydrocarbon t a i l composition and thus i n the f l u i d i t y of the bilayer (Cogan & Schachters, 1981; Seigneuret et a l , 1984). The g l y c o l i p i d s of red c e l l s contain A, B, H, I i blood group determinants (Hakomori, 1981a) as well as the P antigens (Marcus et a l , 1981). Hakomori (1981b) raises the p o s s i b i l i t y that there may be glycolipid-associated proteins. These antigens could regulate, or be regulated by, g l y c o l i p i d s . The red c e l l l i p i d s are known to exchange with serum l i p i d s . Cholesterol i s i n rapid equilibrium with unesterified plasma cholesterol (produced by the plasma enzyme le c i t h i n - c h o l e s t e r o l acyltransferase), PC and sphingomyelin also undergo exchange while i t appears that PS and PE do not (Weed & Reed, 1966). Wallach & Verma (1975) found membrane associated carotenoids (at 7 x 10~\ of l i p i d ) which Lippert et a l (1975) also found i n variable amounts from donor to donor, possibly washed from ghosts i n some experiments. Lewis - 14 -antigens are l i p i d s which are transferred from the serum into the red c e l l membrane (Marcus & Cass, 1969). I t has been shown that shape changes can be induced i n the red c e l l by amphipathic molecules affecting the l i p i d bilayer (Lange et a l , 1982; Landman, 1984). Lipids are also involved i n c e l l fusion processes. l . l ( c ) i i Proteins The membrane proteins give the red c e l l i t s shape, s t a b i l i t y and supply the majority of surface charge. Analysis of red c e l l membrane proteins was greatly impeded by t h e i r i n s o l u b i l i t y i n ordinary buffers and u n t i l the advent of sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS PAGE), they were not very well understood. SDS disrupts noncovalent interactions and binds to polypeptide chains. The amount of SDS i s usually proportional to the mass of the polypeptide and thus proteins migrate on PAGE as a function of molecular size (Reynolds & Tanford, 1970). Kale et a l (1978) and Makino (1979) have shown that SDS does bind to membrane proteins, but certain proteins such as glycoproteins appear to migrate i n an anomalous fashion. Other problems are the choice of standards used to determine molecular weights (log(molecular size) vs mobility i s not always l i n e a r ) , and that several protein species may migrate at the same place on the gel since b i o l o g i c a l systems t y p i c a l l y contain a huge variety of proteins. Hjelmeland & Chrambach (1981) review SDS PAGE i n general and Thompson & Maddy (1982) discuss technical aspects for red c e l l s . Fairbanks et a l (1971) used a one dimensional SDS PAGE system which c l e a r l y showed at least - 15 -7 major protein bands. The numerical labeling of these bands (1 for the highest molecular weight, 2 for the next highest and so on) provided a reference point for discussion of par t i c u l a r red c e l l membrane proteins. I t was l a t e r found that the system of Laemmli (1970) gave better resolution and more bands (Thompson & Maddy, 1982). Fig. 1.2 (adapted from Thompson & Maddy, 1982) shows the protein gel p r o f i l e for the Fairbanks and the Laemmli systems along with the numbering system. As the sialoglycoproteins s t a i n poorly with the common protein s t a i n coomassie blue, they are visualized v i a periodic a c i d - s c h i f f stain (PAS). These glycoproteins, being heavily glycosylated, didn't run " i d e a l l y " on SDS PAGE (having an apparent molecular weight greater than actual (Grefrath & Reynolds, 1974)) and were discovered to combine not only with themselves but with other PAS proteins. Due to the fact that under different gel and buffer conditions they run d i f f e r e n t l y r e l a t i v e to the coomassie blue stained proteins, the PAS proteins are treated separately. There are now believed to be four i n d i v i d u a l PAS stainable components (and possible minor, uncharacterized bands) which interchange with each other. Fig. 1.3 shows the gel p r o f i l e s of the PAS proteins (adapted from Anstee, 1981) along with the most common nomenclatures and t h e i r sources. The red c e l l membrane proteins can be distinguished on the basis of th e i r ease of s o l u b i l i z a t i o n and/or function. Singer & Nicolson (1972) separated proteins into two categories by th e i r ease of s o l u b i l i z a t i o n . E x t r i n s i c (or peripheral) membrane proteins were ea s i l y extracted by manipulation of the i o n i c strength or pH. Bands 1, 2, 4, 5, and 6 are included i n t h i s category and comprise from 30 to 35% of the membrane protein. The remaining proteins were extremely d i f f i c u l t to remove, - 16 -A B C D 6 • • • 6 PAS 3 ^ mmIPAS3 8 K b — • tmm Hi>'ID F i g . 1.2 The t o t a l p olypeptide content by coomassie blue s t a i n i n g (A and C) and g l y c o p r o t e i n s by PAS s t a i n i n g (B and D) separated by Fairbanks system (A and B) or separated by the Laemmli system (C and D). Polypeptides are numbered according t o Fairbanks et a l (1971) and the g l y c o p r o t e i n s by the nomenclature of Anstee (1981). Hb = hemoglobin, TD = t r a c k i n g dye. (Adapted from Thompson & Maddy, 1982) TurttvmjT glycopooriB A g l y c o p h o r i n B g l y c o p h o r i n * V r I n C g j j cophorui B o r i g i n - * , * n _ a » « « t ml '2 c(6 8 fkS-! ( P t s-JM I c II I F i g . 1.3 Nomenclature f o r the PAS s t a i n i n g bands. SDS PAGE of membranes from normal human e r y t h r o c y t e s separated by the Laemmli system. The g e l was st a i n e d w i t h PAS. (Taken from Anstee, 1981) - 17 -requiring detergents, and were believed to be intimately connected with the membrane. They were cal l e d i n t e g r a l (or i n t r i n s i c ) proteins. These include bands 3, and 7, PAS 1, 2, 3, and 4 as well as minor proteins l i k e the ATPases, acetylcholinesterase (AChe) and uncharacterized glycoproteins. The e x t r i n s i c proteins are believed to be on the inner layer of the bilayer while the int e g r a l proteins are intercalated into the bilayer and i n some cases span i t (e.g. band 3 and PAS 1) although band 7 i s believed to be on the inside. The other categorization, proposed by Marchesi (1979a), separates the membrane proteins into three functional groups. The f i r s t group has a contact or receptor function. These are proteins exposed to the outer surface which interact with the environment. The PAS proteins are examples of these. The second functional group contains the c a t a l y t i c units. The red c e l l membrane undergoes vigorous transactions with the surrounding blood plasma, exhibits HCO~ and C l ~ exchange, glucose uptake and maintains Na +/K + and C a + 2 gradients. Transporter proteins such as band 3 would be i n t h i s c a t a l y t i c category. The t h i r d group contain the supporting elements, the proteins which combine to produce a supporting sk e l e t a l structure. The cytoskeleton Several recent reviews have been written about spectrin and the cytoskeleton (Marchesi, 1979b, 1983; Palek, 1983) so only a b r i e f synopsis w i l l be presented here. The c e l l owes i t s shape, i n t e g r i t y and e l a s t i c i t y to the cytoskeleton (Gratzer, 1981). Band 1 (240k) and band 2 (220k), c o l l e c t i v e l y known as spectrin, comprise 25% of the membrane protein and form a tetramer composed of two heterodimers (band 1+2) joined head to head. - 18 -Near the t a i l end, actin (band 5) binds as does band 4.1; 4.1 i s actually two d i s t i n c t proteins c a l l e d 4.1a (78k daltons) and 4.1b (80k daltons) (Lelo & Marchesi, 1984). Ankyrin (band 2.1) binds to spectrin at a s i t e 200 angstroms d i s t a l to the end of spectrin and also to the cytoplasmic end of band 3 (Weaver & Marchesi, 1984; Weaver et a l , 1984). I t appears that bands 4.2, 4.9 and 7 are also involved but l i t t l e - i s known about them (Lux, 1979). Mueller & Morrison (1981) found that PAS 2, which they c a l l "glycoconnectin", interacts with the cytoskeletal structure, most l i k e l y through band 4.1a,b. The resul t i s a meshwork of spectrin underlying the membrane connected to band 3 via ankyrin and to PAS 2 via band 4.1a,b. I f one assumes the spectrin tetramer i s the basic unit, there are equal numbers of 4.1 and 2.1 per tetramer (1 x 1 0 5 / c e l l ) . There are about 5 times as many actins (Pinder & Gratzer, 1983) but only 3.5 x 10 4 PAS 2/ c e l l (Furthmayr, 1978) or 1/3 as many. Fig. 1.4 (taken from Cohen, 1983) shows a proposed structure for the cytoskeleton. I t has been estimated that the skeleton covers from 50 to 70% of the inner surface (Haest, 1982). Nakajima (1979) and Anderson & Lovrien (1981) postulate an association between glycophorin A (PAS 1) and spectrin although no other evidence has supported t h i s hypothesis. The spectrin complex may also bind to the inner bilayer of the phospholipids and aid i n maintaining the phospholipid asymmetry (Kuypers et a l , 1984; Haest et a l , 1980). Haest (1982) has written an excellent review concerning the interactions between the skeleton and the red c e l l membrane which i s summarized i n Table 1.3. Recently Fowler & Bennett (1984) discovered a membrane protein (a dimer consisting of a 29k and a 27k dalton monomer and about 1% of the t o t a l - 19 -membrane p r o t e i n ) they i d e n t i f y as tropomyosin. This molecule i s known t o bind to muscle F - a c t i n and they po i n t out that s i n c e there i s 1 tropomyosin dimer to 7-8 a c t i n molecules, the p o s s i b i l i t y of actomyosin c o n t r a c t i l e mechanisms should be reconsidered. F i g . 1.4 (A) Schematic drawing i l l u s t r a t i n g some of the molecular a s s o c i a t i o n s present i n a h y p o t h e t i c a l red c e l l membrane s k e l e t o n " u n i t c e l l . " The number of a c t i n molecules shown i n the s h o r t f i l a m e n t and the number of attached s p e c t r i n molecules may vary from u n i t c e l l to u n i t c e l l . (B) Two-dimensional membrane s k e l e t a l network which r e s u l t s from the head-to-head a s s o c i a t i o n of the s p e c t r i n dimers i n the u n i t c e l l s shown i n A. (Taken from Cohen, 1983) - 20 -TABLE 1.3 PROPERTIES, ASSOCIATIONS AND FUNCTION OF MEMBRANE PROTEIN FRACTIONS (Taken from HAEST, 1982) Peptide fraction Copies per cell ( X l O - 5 ) Associated state Function Association with Band 1 (spectrin) Band 2 (spectrin) 240000 220000 2.2 Dimers of band 1 and 2 form leiramers and oligomers Membrane skeleton Band 3 (to hand 1) ankyrin (to band 2) baud 4.1 Band 2.1 (ankyrin) Band 2.2 (ankyrin) Band 2.3 (ankyrin) 210000 183000 165000 1.1 Monomer Connects membrane skeleton with intrinsic domain Band 2. band 3 Band 3 95000 12 Tetramer in equilibri-um with dimers Inorganic anion transport system Skeleton attachment site Band 1. ankyrin Band 4.2. band 6 glycophorin A Band 4 1 a Band 4 1 b 80000 78000 2.3 Dimers'' Connects membrane skeleton with intrinsic domain. Specirin-aclin interaction Spectrin. glycophorin C Band 4.2 72000 2.3 Tetramer Unknown Band 3 Band 4.5/1 2 3 4 5 6 59000 -52000 1.3 1.4 0.7 1.0 1.3 1.3 Possibly monosaccharide, •--lactate and nucleoside transfer svstcm (Refs. 209. 211 and 212 resp.) Band 4 9 48000 1 Membrane skeleton (degralion product?) Band 2 Band 5 (actin) 43000 5.1 Oligomers of 10 monomers Membrane skeleton Spectrin, band 4.1 Band 6 35000 4.1 Tetramer Cilvceraldehvde-3-/>-dchvdrogenase Band 3. spectrin-aclin Band 7 29000 Glycophorin A Glycophorin B Glycophorin C 29000 - 2 5 0 0 0 8 2 0.7 0.35 Di mer Skeleton attachment site Band 3 Band 4.1 - 21 -Catalytic functions Band 3 The most abundant i n t r i n s i c protein of the red c e l l membrane i s band 3. There are 1.2 x 10^ copies/cell (Fairbanks et a l , 1971) and i t comprises 24% of the t o t a l membrane protein (Jones & Nickson, 1980). Several review a r t i c l e s on the subj'ect have been written (Jenkins & Tanner, 1977; Steck, 1978; Macara & Cantley, 1983). On SDS PAGE band 3 always appears as a broad band centered at about 100,000 daltons implying a heterogenous population (Matsumoto & Osawa, 1980) . This i s p a r t i a l l y due to the heterogeneity i n the glycosylated portion of band 3 (Mueller & Morrison, 1979; Fukuda et a l , 1979; Tsuj'i et at, 1981) and to other minor proteins comigrating with band 3. These minor proteins include acetylcholinesterase (Ott et a l , 1983), Na +/K + ATPase (Erdmann, 1982), the Ca 2 +-Mg 2 + ATPase (Gietzen & Koland, 1982; Zurini et a l , 1984), Mg 2 + ATPase (White & Ralston, 1980) and the i n s u l i n receptor (Grigorescu et a l , 1983). Blood groups A, B (Hakomori, 1981a) and I (Childs et a l , 1978) are located on band 3 i n i t s carbohydrate structure. This has been named erythroglycan, a large branched oligosaccharide of repeating gal and glcNAc units (Jarnefelt et a l , 1978; S u j i et a l , 1980). The senescent c e l l antigen has been postulated to reside on band 3 (Kay e t ' a l , 1983). Band 3 i s known to be the anion transporter (H^PO^, C l ~ , HCOj etc.) (Cabantchik et a l , 1978) and can also transport a small fraction of the neutral amino acids (Saleh & Wheeler, 1982; Young et a l , 1981) , phosphoenol pyruvate (Hamasaki et a l , 1978), water (Brown et a l , 1975) and possibly C a 2 + (Waisman et a l , 1982). - 22 -The cytoplasmic part of Band 3 i s the binding s i t e for ankyrin and cytoplasmic components such as hemoglobin (Cassoly & Salhany, 1983) aldolase, phosphofructokinase, glyceraldehyde 3-phosphate dehydrogenase (G3PD) ( G i l l e s , 1982) and possibly others. Fig. 1.5 (taken from Macara & Cantely, 1983) i s the proposed model for band 3. Although the exact structure i s n ' t known, the molecule must traverse the membrane several times to explain a l l the proteolysis and chemical modification data accumulated about i t . Band 3 may be a dimer or tetramer i n the membrane (Klingenberg, 1981) and may interact with glycophorin A (Nigg et a l , 1980) but doubts were raised by Schweizer et a l (1982) who were unable to crosslink band 3 to i t s e l f or other components using instantaneous crosslinking conditions. o o i o C Y T O P L A S M I C D O M A I N M E M B R A N E D O M A I N (40-43K) Fig. 1.5 Organization of the Band 3 polypeptide i n the red c e l l membrane, showing i t s two-domain structure. (Taken from Macara & Cantely, 1983) - 23 -Band A. 5 One function that band 3 was thought to have had was that of glucose transport. Mullins & Langdon (1980), using a potent glucose i n h i b i t o r 5 a f f i n i t y l a b e l , labeled exclusively band 3 getting 3 x 10 transporters/cell. Band A.5 labeled i f band 3 was enzymatically degraded by endogenous proteases i n Triton X-100. However, others, using cytochalasin B binding, labeled band A.5 (Shanahan, 1983). The amount varied from 3.3 x 10 A (Jones & Nickson, 1980) to 2.5 x 10 3 s i t e s / c e l l (Lin & Snyder, 1977). Carruthers & Melchior (198A) claim that band A.5 i s the transporter but band 3 also appears to mediate cytochalasin B and transport a c t i v i t y . Whitfield et a l (1983) state that bands 3 and A.5 are immunologically and s t r u c t u r a l l y related. Most investigators now believe band A.5 to be the glucose transporter (Wagner et a l , 198A; Wu et a l , 1983b; K l i p et a l , 198A). This glucose transporter comprises 3.5% of the membrane protein and i s a glycoprotein, 15% carbohydrate by weight. (Lienhard et a l , 1983). Band A.5 i s actually a region of overlapping polypeptides and 6 bands can be seen on Laemmli gels (Jones & Nickson, 1980). They range i n size from A5000 to 55000 daltons and comprise a t o t a l of 10% of the membrane 5 protein (each band numbering a l i t t l e over 1 x 10 c o p i e s / c e l l ) . Like band 3 they are also heterogenously glycosylated (Gorga et a l , 1979) and carry blood groups A, B, H, and I i (Hakomori, 1981a) Catalase, a cytosol enzyme of 60000 daltons also appears i n t h i s area (Deas et a l , 1978). Band A.5 i s also known to be the nucleoside c a r r i e r (Wu et a l , 1983a), although band 7 has been implicated along with band A.5 (Wu et a l , 1983b). - 24 -I t appears to be a multimer i n the membrane, having a radiation target size of 122000 daltons (Jarvis et a l , 1984). The nucleoside transporter comprises 0.1% of the membrane protein, numbering about 1.38 x 10 4 copies/cell (Jarvis & Young, 1982). Contact or receptor proteins The only contact proteins considered i n t h i s section are the PAS proteins, and as can be seen from Fig. 1.3, there are at least four. Reviews on these molecules have been written by Furthmayr (1981) and Anstee (1981). The most predominant and best known i s glycophorin A (PAS 1 or alpha) comprising 70% of a l l PAS protein. I t numbers about 5 x KrVcell, although values as high as 2 x 10 6 have been quoted (Anstee, 1981). I t comprises about 1.6% of the membrane protein and carries 60% of the s i a l i c acid. The amino acid sequence has been worked out, i t has 131 amino acid residues with residues 71 to 90 being hydrophobic i n nature and just large enough to span the l i p i d b i l a y e r . The carboxyl end of the molecule resides on the cytoplasmic side and the amino terminus on the outer surface i s heavily glycosylated. Sixty percent of the molecule i s carbohydrate, consisting of 15 O-glycosidically linked and 1 N-glycosidically linked oligosaccharides. Fig. 1.6 depicts the structure of the peptide chain and the location of the oligosaccharides. From the amino acid and carbohydrate composition, a molecular weight of 31000 daltons was calculated for glycophorin A. This molecule possesses M, N, and Pr blood groups as well as malarial parasite and virus receptors. Glycophorin B (PAS 3 or delta) comprises about 15% of the PAS protein and numbers about 7 x l o V c e l l (Furthmayr, 1978). I t s molecular weight i s - 25 -27000 daltons (Dohnal et a l , 1980). The amino end of t h i s molecule i s i d e n t i c a l to that of glycophorin A, blood group N, up to residue 26, which i s found for glycophorin B not to be N-glycosylated. I t i s believed that i t doesn't carry a N-glycosidically linked complex oligosaccharide. Thus, glycophorin B carries the N blood determinant as well as Ss and U ( a l l three being determined by amino acids) and the Pr group. Glycophorin C (3.5 x loVcell) according to Anstee i s actually two PAS proteins, gamma and beta. Beta i s also known as PAS 2 or glycoconnectin. Together they comprise about 10% of the PAS proteins, the remainder presumably being minor uncharacterized fractions (even band 3 stains sli g h t y with PAS due to i t s glycosylation). Absence of peptides 3000 to 4000 daltons i n size following t r y p t i c digest of these PAS molecules (except for glycophorin A) suggested no s i g n i f i c a n t cytoplasmic segment for gamma or beta although glycoconnectin has been shown to bind to band 4.1 on the cytoplasmic side. Fig. 1.7 represents a proposed model for the structure of these glycophorins (except gamma for which there i s i n s u f f i c i e n t data) (Anstee, 1981). A l l are believed to extend through the membrane with the amino terminus and the glycosylated part on the outside, and the carboxyl end at the cytoplasmic side. The conformation of these oligosaccharides adjacent to the membrane i s not known. Are they j u t t i n g straight out, or do they extend along the surface of the l i p i d layer? Due to th e i r behavior on SDS PAGE, i t has been postulated that these molecules could be dimers or oligomers i n the membrane although direct evidence i s lacking. - 26 -Fig. 1.6 The amino acids of glycophorin A, arranged to simulate, i n a very general way, the positions they might have i f the molecule was perpendicular to the l i p i d bilayer of the membrane. The s o l i d v e r t i c a l l i n e approximates the location of the inner half of the phospholipid b i l a y e r . The outer edge of the bilayer can only be approximated and i s defined by the dashed v e r t i c a l l i n e s . The 15 diamond shaped structures are the O-glycosidically linked and the other i s the N-glycosidically linked oligosaccharide. (Taken from Marchesi, 1979) M ;N F i g . 1.7 Diagrammatic representation of the location of antigens on normal erythrocyte sialoglycoproteins. The small c i r c l e s are the O-glycosidically linked oligosaccharides and the big shaded c i r c l e s the N-glycosidically linked oligosaccharides. (Taken from Anstee, 1981) - 27 -The function of these PAS proteins i s unknown. Healthy in d i v i d u a l s e x i s t who don't contain glycophorin A (Tanner & Anstee, 1976) and two people are known who lack gamma and delta (although there i s altered glycosylation of the other membrane proteins (Anstee, 1981)). The PAS proteins do supply the major source of s i a l i c acid and thus carry the majority of the surface charge. The age related antigen has been proposed to reside on glycophorin (Alderman et a l , 1981). The glycophorins interact with the cytoskeleton d i r e c t l y (via glycoconnectin) and possibly i n d i r e c t l y (Lovrien & Anderson, 1980; Nakajima, 1979). Bowles & Hanke (1977) have even postulated that glycophorin has l e c t i n a c t i v i t y , binding s p e c i f i c a l l y to galNAc residues which may be involved i n some recognition process. Even though they contain receptors for virus and malarial parasites, i t i s probably better to think of these proteins as contact proteins, t h e i r function being one of keeping the red c e l l s apart through e l e c t r o s t a t i c repulsion and s t e r i c s t a b i l i z a t i o n . Minor components of the membrane There are a number of minor proteins associated with or i n the membrane which can't be assigned roles or structures. Gahmberg (1976) labelled the red c e l l surface with galactose oxidase and ["*H] and found over 20 stainable glycoproteins. Using two dimensional gel electrophoresis (separating proteins on the basis of molecular weight and i s o - e l e c t r i c point) 100-200 reproducibly detectable proteins can be seen i n red c e l l membrane preparations (Rubin & Minkowski, 1978; Rosenblum, 1981). The majority of these proteins appear to be enzymes. Schrier (1978) reviews the membrane enzymes and l i s t s over AO he believes are d e f i n i t e l y i n the membrane (Table 1.4), - 28 -TABLE 1.4 ENZYMES WHOSE ACTIVITY IS FOUND IN THE MEMBRANE OF RED BLOOD CELLS (Taken from Schrier, 1978) I. Enzymes of nucleotide metabolism (ATPoses considered separately) 2 ,3' cAMP cyclic nucleotide 3 '-phosphohydrolase Adenyl cyclase UTPose * ' 5'AMP phosphatase* II. Enzymes of carbohydrate metabolism Neuraminidase CMP-N-ocetylneuraminic acid: glycoprotein sialyltransferase N-acetyl-tf-glucosaminidase* of-D-glucosidose* jfJ-D-glucosidase* ai-D-golactosidase" 5-D-golacTosidase* oc-L-fucosidase* tf-L-fucosidase #-D-xylosidase* <ar,-D-mannosidase* N-acetyl-/?-D-goloctosaminidase* jJ-Glucurontdase N-acetyl-goloctosaminyl transferase III. Phosphatases Vitamin B 6 phosphate phosphatase p-nitrophenyl phosphatase |not K "* dependent, for K * -dependent mtrophenyl phosphatase see ATPoses) M g * * dependent phosphoprotein phosphatase IV. Proteinases 3 proteinases pH optima 7.4, 7.4, 3.2 2 proteinases pH optima 3.4, 7.4 V. ATPoses ATPoses nol linked to fibrous membrane proteins Mg + t -ATPase No*. K *-ATPase, ouabain inhibited (K ^-dependent p-nttrophenyl phosphatase, ouabain inhibited; Co 4 " 2 . Mg* 2 -ATPase Possibly two affinities for Co , the low-affinity enzyme may be the Ca transporting en?yme Monovalent cation stimulated, not related to active Co extrusion ATPoses associated with fibrillar membrone proteins (spectrin) Ca ATPase inhibited by Mg * ' (approximately 5% os active as the Co + 1 , Mg "^-ATPase) Mg**-ATPase, low activity, stimulated by a<tin, thyf resembling octomyosin VI. Protein kinases cAMP independent, stimulated by mono and divalent cations to phosphorylate band 2, spectrin cAMP stimulated, Ca"* inhibited, monovalent cation inhibited VII. Miscellaneous NAD* glycohydrolose (DPNase) NADP + glycohydrolose (TPNase) NADH: acceotor oxidoreductase Acetylcholinesterase* - 29 -Juliano (1973) puts the enzymes into three classes. The f i r s t contain cytoplasmic enzymes, which may also be associated with the membrane and whose release usually p a r a l l e l s that of hemoglobin. The next class contains those which require more stringent conditions for extraction, s i m i l a r to those of e x t r i n s i c proteins l i k e spectrin. The l a s t class (shown i n Table 1.4) consists of the i n t r i n s i c enzymes such as NADH oxidoreductase, protein kinase, AChe, Na +K + ATPase, etc. The proteases found i n the membrane are known to cleave band 3, the PAS proteins and spectrin (Nickson & Jones, 1980; Siegel et a l , 1980; Pontremoli et a l , 1980). Yatzio et a l (1978) and Kahane et a l (1980) stress the fact that when analysing red c e l l s for membrane enzymes, one must be sure no other blood c e l l s (such as pla t e l e t s and white c e l l s ) contaminate the preparation. Sample preparation may also cause a r t i f a c t s . Hahanan (1973) states osmotic stress may lead to a r t i f a c t u a l association of enzymes with the inner surface. Aldolase, pyruvate kinase, fructose-bisphosphate (Tillmann et a l , 1975) as well as hemoglobin and 2% of the calmodulin bind to the membrane (Klinger et a l , 1984). The emerging view i s that there i s no abrupt interface between the cytosol and the inner layer of the membrane (Salhany, 1983). Methemoglobin reductase i s 35% membrane bound i n a 45000 dalton form and i s cleaved to the 29000 dalton form which elutes to the cytosol (Choury ejt al^, 1983). This association of cytosol enzymes with the membrane may be part of a complex regulatory system (Haest, 1982). Red c e l l metabolites also react with and a l t e r membrane components. Protein kinases are known to phosphorylate proteins 2, 2.1, 2.3, 3, 4.1, 4.5, 4.8, and 5 (Rsienbenlist & Taketa, 1983). Methylases a l t e r the charge of membrane proteins by methylating the ac i d i c amino acids reversibly on - 30 -bands 2.1, 3, 4.1, 4.5, and 6 (Ro et a l , 1984; Green et a l , 1983). Glutathione i s known to react with the membrane protein sulfhydryl groups of spectrin bands 3, 4.1, 4.2, 4.5, and 5 (Haest et a l , 1979). The outer surface of the membrane i s also exposed to a variety of proteins as well as hormones, drugs, etc. For instance, N-acetyl-D-galactosaminyltransferase i s found i n serum and on red c e l l s (Kim et a l , 1971). Antibodies are known to bind to red c e l l s (Muller & Lutz, 1983; Alderman et a l , 1981; Bartosz & Bryszewska, 1983). Davis & Weiss (1980) extracted by gentle agitation (glc).j-cys-glu-gly-(gly-ser)-ala, the removal of which seemed to decrease glucose transport. Serum glucose has been shown to react with membrane proteins nonenzymatically ( M i l l e r e_t a l , 1980) providing an average of 7 x 10 5 glucoses/cell (Schleicher et a l , 1982). The red c e l l interacts with many b i o l o g i c a l substances i n the nanomolar range, such as i n s u l i n , growth hormone, acetylcholine, adrenergic agents and prostaglandins, some at only a few molecules/cell (Gratzer, 1981). Other minor components on the membrane are antigens of which there are over 300 types known (Rosse, 1984) and transport proteins for compounds (such as lactate (Dubinsky & Racker, 1978), glutathione (Kondo et a l , 1981) and amino acids (Saleh & Wheeler, 1982)) for which no protein has yet been isolated. No mention has been made of the changes which occur to a red c e l l as i t ages i n vivo, about alterations which can occur i n a disease state, or to the fact that there i s considerable genetic v a r i a b i l i t y . Each of these could be chapters i n themselves. Suffice to say they add to the problem and complexity of red c e l l membrane analysis. - 31 -A major problem at present concerns the conformation of the membrane proteins i n s i t u . How they interact with each other and with the rest of t h e i r environment i s not known. Of particular interest i n t h i s study are the carbohydrates located on the PAS proteins and bands 3 and 4.5 which are depicted i n Fig. 1.8. At 15 O-glycosidic oligosaccharides per glycophorin A, there must be approximately 6 5 8 x 10 of these per red c e l l and about 5 x 10 N-glycosidically linked oligosaccharides per red c e l l . From band 3, about 1 x 10 6 oligosaccharides must reside on the surface. They appear to constitute the majority of the glycocalyx. Fig. 1.8 shows the proposed structures for these oligosaccharides. Fig. 1.8a i s the proposed structure for the O-glycosidically linked oligosaccharide of glycophorin A (Furthmayr, 1981; Thomas & Winzler, 1969; Sadler et a l , 1979; Gahmberg & Andersson, 1981; Prohaska et a l , 1981). Fig. 1.8b i s the N-glycosidically linked oligosaccharide of glycophorin A (Irimura et a l , 1981; Yoshima et a l , 1980) and Fig. 1.8c the structure of the band 3 oligosaccharide (N-glycosidically linked) as proposed by Fukuda et a l (1984). Fig. 1.9 depicts t h i s outer surface of the red c e l l (taken from Sharp, 1985). - 32 -NeuNAca2-3Gal /3 l -3GalNAcal -0 -Ser (or Thr) 6 A I N e u N A c a 2 R-Gal1-^4GlcNAclH.2Man Fuc 1 1 1 G l c N A c l ^ M a n l - V i G l c N A c l ^ / t G l c N A c * * 3 « 0 0 1 RjG a 1 1 G t c N A c 1 2 M a n l t 1Z R-»GalBl-»4(GlcNAc61-»3Ga161-»4) nGlcNAcBH2Manal Fuc R-»Gal81+4(GlcNAc81-»3GalBl-»4) GlcNAcBl Man81-»4GlcNAc81-»4GlcNAc81-»Asn \ 3 2 / R, 4 i 2 / R»»GalBl-»4(GlcNAcBl+3Gal81-»4) GlcNAcBl F i g . 1.8 The proposed structures for the oligosaccharides of glycophorin and band 3. (A) the O-glycosidically linked oligosaccharides of glycophorin, taken.from Furthmayr, 1981; (B) the N-glycosidically linked oligosaccharide of glycophorin taken from Irimura et a l , 1981 (Ri = H or NANA 2-6 (Yoshima et a l , 1980)); and (C) the oligosaccharide attached to band 3. R = H, Fuc 1-2, NANA 2-3, or NANA 2-6; R 2 = R-Gal 1-4 Glc 1-6, n = 4-5. (Taken from Fukuda et a l , 1984) - 33 -Fig. 1.9 A scale drawing attempting to show the outer surface of the red c e l l membrane. Major i n t r i n s i c proteins: band 3 (B), glycophorin A (G). Major e x t r i n s i c proteins: spectrin (S), a c t i n (Ac), band 2.1 (ankyrin) (A). L i p i d s : phospholipid bilayer (PL), g l y c o l i p i d s (GL). Carbohydrate i s shown i n s o l i d black. No attempt has been made to show the true conformations, shapes and d i s t r i b u t i o n of these components. However the r e l a t i v e amounts, average volumes and separation between band 3, glycophorin A, the g l y c o l i p i d s and the phospholipids are shown to scale, based on current estimates of t h e i r molecular weights, % carbohydrate and amounts per c e l l . Modification of Peters & Grant (1983). Calculated and drawn by K.A. Sharp (1985). - 3 4 -A useful tool i n the study of the glycocalyx would be one that interacted with the glycocalyx, mimicking recognition behavior. Lectins, known to bind to carbohydrate sequences, appear well suited for use i n model studies therefore. 1.2 LECTINS "A macromolecule contains information i n i t s sequence of subunits that determines the three-dimensional contours of i t s surface. These contours i n turn govern the recognition between one molecule and another, or between different parts of the same molecule, by means of weak noncovalent bonds." (Alberts et al_, 1983). This statement describes a wide variety of bi o l o g i c a l interactions including enzyme-substrate, antibody-antigen and c e l l - c e l l recognition. Lectins are proteins (or glycoproteins) which bind noncovalently and reversibly to carbohydrate groups without modifying them. Excluded from t h i s group are proteins with known functions such as antibodies, transport proteins, and enzymes. Lectins were discovered by Stillmark i n 1888 and were o r i g i n a l l y c a l l e d phytohemagglutinins because they could agglutinate red c e l l s and were isolated mainly from plants. In 1952, Watkins & Morgan found that s p e c i f i c monosaccharides could i n h i b i t such hemagglutination and were thus the f i r s t to show the presence of sugars on the outer surface of membranes. Indeed developments i n l e c t i n studies p a r a l l e l advances i n our knowledge of surface carbohydrates. In 1954, Boyd & Shapleigh chose the more general term, l e c t i n , from the l a t i n Lego, to pick or choose, because l e c t i n s could be isolated from animal sources also. - 35 -Lectins are i n many ways analogous to antibodies, both recognizing certain molecular "contours." Kabat (1978) has even found s i m i l a r i t i e s i n the binding s i t e s of l e c t i n s and antibodies, both showing large differences i n s i t e sizes and s p e c i f i c i t i e s . Although antibodies are more s p e c i f i c and bind to proteins as well as sugars, they are hard to obtain pure i n large amounts. Lectins can be isol a t e d r e l a t i v e l y cheaply, from a variety of sources. Several excellent reviews on l e c t i n s are Brown & Hunt (1978), Goldstein & Haynes (1978), L i s & Sharon (1973) and Nicolson (1974). Due to t h e i r a b i l i t y to bind saccharides, l e c t i n s have become useful for i s o l a t i o n and p u r i f i c a t i o n of glycoproteins and c e l l s . They are used as surface probes (e.g. blood group typing, monitoring variations i n malignant c e l l s ) as well as for s t r u c t u r a l studies and for model studies of carbohydrate binding. Lectin binding to c e l l s may e l i c i t a number of responses, depending on the l e c t i n and the c e l l receptor. Agglutination, mitogenesis, t o x i c i t y , altered transport and mimicking of an active agent i n vivo (such as i n s u l i n ) have been seen. Lectins thus enable a researcher to manipulate c e l l s i n v i t r o to study these phenomena. Lectin binding i s affected by the nature of the l e c t i n (number of binding s i t e s , binding constants, net charge and size) and the experimental conditions such as temperature, time, pH, ioni c strength, nonspecific interactions, cofactors, t e r t i a r y structure of the oligosaccharides, number of receptors per c e l l and properties of the c e l l surface. Binding can occur without agglutination (Sharon & L i s , 1975). The fact that l e c t i n s bind strongly to c e l l s , mimicking a variety of natural agents, makes them well suited for studying the complex phenomenon of protein binding. - 36 -1.3 CHEMICAL MODIFICATION Of a l l the carbohydrates on the c e l l surface, s i a l i c acid (also known as N-acetyl neuraminic acid or NANA) and galactose are the easiest to modify s p e c i f i c a l l y , s i a l i c acid because of i t s exocyclic t r i o l (not known to exist on other sugars) and galactose because of the s e l e c t i v i t y of enzymes. As reviewed e a r l i e r , the red c e l l surface i s known to contain both these residues. In t h i s work they were s e l e c t i v e l y modified by a procedure s i m i l a r to that of Aplin et a l (1979), (see also Aplin, 1979 and Bernstein, 1983). This involves selective oxidation of the NANA or galactose residues followed by reductive amination v i a NaBH^CN and a compound with a reactive amine group. The f i r s t step i s the oxidation of the terminal NANA or galactose, generating a reactive aldehyde. 1.3(a) NANA oxidation I t has been known that mild oxidation by periodate results i n selective oxidation of the exocyclic t r i o l of NANA (Van Lenten & Ashwell, 1971) (Fig. 1.10). Periodate oxidation i s commonly used i n Smith degradation to determine intersugar linkages i n oligosaccharides, with i t s r e a c t i v i t y towards v i c i n a l d i o l s i n the order e x t r a c y c l i c greater than i n t r a c y c l i c c i s greater than i n t r a c y c l i c trans (Wagh & Bahl, 1981). I f the periodate concentration i s kept low enough and the times short enough, only the most reactive d i o l s w i l l be oxidized, those on the exocyclic t r i o l of NANA, leaving the i n t e g r i t y of the carbohydrate chain intact and ensuring s p e c i f i c i t y i n the reaction. - 37 -Fig. 1.10 Oxidation of s i a l i c acid by periodate. R = red c e l l 1.3(b) Galactose oxidation The oxidation of galactose i s best performed with the enzyme galactose oxidase (G.O.) (E.C. No 1.1.3.9). This enzyme oxidizes the 6-CH20H of galactose or N-acetyl galactosamine (galNAc) producing an aldehyde and H 20 2 (see Fig. 1.11). The H 20 2 formed i n h i b i t s G.O. by a feed-back control mechanism and i s usually removed by addition of the enzyme catalase. The addition of catalase to red c e l l s being oxidized by G.O. i s not necessary, however, due to the natural presence of catalase i n these c e l l s (Liao et a l , 1973; Deas et a l , 1978). Galactose i s usually found with NANA linked to i t so to la b e l these subterminal residues also one must f i r s t remove the NANA. This can be done by either mild acid hydrolysis (usually 0.1 N F^SO^ at 80°C for one hour) or with the enzyme neuraminidase (E.C. No. 3.2.1.18). - 38 -HO OR GALACTOSE OXIDASE HO OR (NH.Ac) 0 2 H 2 0 2 OH ^ (NH.Ac) Fig. 1.11 Oxidation of galactose termini by galactose oxidase. R = red c e l l . 1.3(c) Reductive amination The aldehydes generated on NANA or galactose can be modified with the spin la b e l 2,2,6,6-tetramethyl-4-amino-piperidine-l-oxyl (often c a l l e d TEMPAMINE and indicated i n chemical pathways as SL-NH2) (Fig. 1.12). The amine group i s reductively aminated onto the aldehyde using NaBH^CN as depicted i n Fig. 1.13. As can be seen from t h i s scheme, addition of NaBt^Hj^CN results i n a t r i t i u m being placed on the covalently attached probe, res u l t i n g i n a second tag which can be followed during the reactions. NaBH,CN i s used to reduce the imine ion instead of NaBH. due 3 4 to i t s resistance to aqueous hydrolysis and i t s s p e c i f i c i t y for the intermediate S c h i f f base over the aldehyde, NaBH^ reducing the aldehyde as well (Borch et a l , 1971; Lane, 1975). One also has the option of using other compounds or probes, the only c r i t e r i a being that they contain a reactive amine, are soluble i n aqueous media and normally don't react with the b i o l o g i c a l system unless previously oxidized. - 39 -NH2 I SL Fig. 1.12 The structure of TEMPAMINE, the spin l a b e l used i n t h i s study. NH-I ' SL L H + J No B R 3 C N + H20 -CHR-NH-SL R = H or *H Fig. 1.13 Reductive amination of an aldehyde with TEMPAMINE, using NaBH3CN with or without NaB[3H]3CN. - AO -1.4 ELECTRON SPIN RESONANCE Electron spin resonance (ESR), also known as electron paramagnetic resonance (EPR), i s an important spectroscopic tool for studying b i o l o g i c a l systems. This i s usually achieved by incorporation of a spin l a b e l into the system of in t e r e s t . Information regarding membranes (especially the l i p i d portion), enzymes and other b i o l o g i c a l e n t i t i e s has been expanded by the use of ESR. A number of excellent reviews and monographs covering many aspects of theory, application and p r a c t i c a l problems are c i t e d below (Hudson & Luckhurst, 1969; Smith, 1972; Keith et a l , 1973; Likhtenstein, 1976; Berliner, 1976, 1978, 1979; Schreier et a l , 1978; Jost & G r i f f i t h , 1978; Marsh, 1981; Poole J r . , 1983). l.A(a) Nitroxides ESR i s performed on paramagnetic substances which i n t h i s study are introduced into the system by means of spin labeling (see Section 1.3). The term "spin l a b e l " was f i r s t coined by McConnell and coworkers (Stone et a l , 1965) and refers to stable free r a d i c a l s , usually a nitroxide-containing compound which contains an unpaired electron l o c a l i z e d on the nitrogen and oxygen atoms (Jost & G r i f f i t h s , 1978). Nitroxides are amongst the most stable of free radicals and were f i r s t synthesized about 1960 (Rozantsev, 1970). The nitroxide i t s e l f i s usually not involved i n chemical reactions, the reactive group being located away from the nitroxide (as i s seen for TEMPAMINE i n Fig. 1.13). Fig. 1.12 shows the structure of TEMPAMINE, the spin probe used i n t h i s study. The methyl - Al -substituents at the 2 alpha carbons prevent the nitroxide from undergoing d i s p r o p o r t i o n a t e . At pH 6-8 the reduction potential for the nitroxide i s about -150 mv, which means i t can be reduced by such agents as ascorbate, d i t h i o t h r e i t o l , and mercaptoethanol but not NaBH^, LiAlH^ or NaCNBH^. The oxidation by the nitroxide of a number of reducing agents has been used to investigate the a c c e s s i b i l i t y of t h i s group. l.A(b) Electron Spin Resonance: General Electron spin resonance occurs with a paramagnetic molecule. This molecule contains at least one unpaired electron which has a spin of + or -1/2. With the application of a magnetic f i e l d , H, energy levels between the two spin states occur as depicted i n Fig. l . l A a . With the application of a microwave frequency, V, perpendicular to the magnetic f i e l d , t r ansitions occur between these two levels (known as the Zeeman levels) at hV = gBH [1] where h = Planck's constant, B = the Bohr magneton (eTV2m) where m i s the mass and e the charge of the electron and g i s a dimensionless parameter related to the e f f e c t i v e magnetic moment of the electron, equaling 2.00232 for a free electron. The frequency of resonance i s dependent upon the applied f i e l d and most experiments, including the ones here, are conducted at the X-band (about 9.5 GHz, equivalent to a 3 cm wavelength) which corresponds to an external f i e l d of about 3.A kG. - 42 -Fig. 1.14 (A) Effect of external magnetic f i e l d strength on the energy l e v e l of the o r b i t a l s of unpaired electrons having spin states of ms = + and - 1/2. (B) Absorption of microwaves of frequency V as a function of f i e l d strength. (C) The resultant ESR spectrum, the f i r s t - d e r i v a t i v e spectrum taken from (B). (Taken from Benga, 1983) Net absorption of microwave energy from occurs at resonance due to more spins being at the lower energy state (Fig. 1.14b). These spins remain close to the equilibrium energy d i s t r i b u t i o n due to rapid relaxation and a weak H-^ . Absorption i s monitored with the aid of phase-sensitive detection using f i e l d modulation at 10 5 Hz and recorded as the f i r s t derivative of the absorption signal (Fig. 1.14c). 1.4(c) Nitroxide ESR In the nitroxide spin l a b e l , the electron also interacts with the nuclear spin of the nitrogen nucleus, a phenomenon known as hyperfine - A3 -inter a c t i o n . The spin Hamiltonian for a nitroxide electron i s given by Ji = J{£ Zeeman) + X ( n y P e r r " i n e ) + K ( d i p o l a r ) + H exchange) [2] The nuclear Zeeman term, being n e g l i g i b l e , has been omitted. The hyperfine interaction occurring between the electron and the nitrogen nucleus (I = 1), results i n the c h a r a c t e r i s t i c 3-line spectrum (m = 1, 0, -1, Fig. 1.15). Small hyperfine interactions occur with the beta-hydrogens but are not resolved. Hyperfine interactions also occur with 13 15 the alpha C nuclei (1.1% natural abundance) and N (O.A% natural abundance) and are seen as s a t e l l i t e peaks at high signal-to-noise. Magnetic Field Fig. 1.15 The eff e c t of the spin state of nitrogen nuclei (I = +1, 0, -1) on the energy l e v e l of the unpaired electron. The magnitude of the magnetic f i e l d f e l t by the electron w i l l now depend on the magnetic f i e l d from the nearby nucleus as well as the applied magnetic f i e l d . Also shown are some parameters used to analyse the spectrum, h + i , hg and h_i are the height of the l o w - f i e l d , mid-field and the h i g h - f i e l d l i n e s respectively and WQ i s the linewidth of the mid-field l i n e (see equations 9 and 10). (Adapted from Benga, 1983) - 44 -Fig. 1.16a shows the p r i n c i p a l axes for the nitroxide. The direction-dependence of the Zeeman and hyperfine interactions i s seen i n Fig. 1.16b. Here a spin l a b e l i s i n a diamagnetic host c r y s t a l and the spectrum run at various orientations to the magnetic f i e l d . The g- and T-tensors are then measured at the different orientations. For most nitroxides, g = 2.009, g = 2.006 and g = 2.002 with T ' Mxx ' ayy a z z xx T = 6G and T = 32G (T = 37G for TEMPAMINE). These values d i f f e r yy £.£. z z from compound to compound and the p o l a r i t y of the environment (Berliner, 1978). Both terms can be approximated as a x i a l l y symmetric with g xx = g, yy [3] and T. xx = T yy [4] with [ 5 ] and T z z s T l [6] - 45 -z i Fig. 1.16 (A) Schematic representation of the nitroxide group showing the unpaired electron i n the nitrogen p z o r b i t a l (taken from Smith, 1972). (B) T- and g- anisotropics i n d i - t - b u t y l nitroxide oriented i n a host c r y s t a l at room temperature. The c r y s t a l was rotated i n the molecular yz plane. At 0° and 90° the external f i e l d lay along the z and y axes respectively. (Taken from Jost & G r i f f i t h s , 1978). - 46 -The electron d i s t r i b u t i o n i n the nitroxide i s influenced by the p o l a r i t y of the environment. This i s eas i l y seen when one pictures the nitroxide bond i n i t s resonance form N-0* «-» N+-0" Polar solvents s t a b i l i s e the io n i c form resulting i n increased spin density at the nitrogen and hence increased magnitude of the is o t r o p i c s p l i t t i n g constant a„, defined below. The value of a of TEMPAMINE varies from o o 16.99 G i n water to 15.22 G i n n-hexane (Knauer & Napier, 1976). In d i l u t e solutions at room temperature with fast i s o t r o p i c motion occurring, the nitroxide spectrum i s usually three sharp l i n e s (Fig. 1.16). The g and T anisotropies are averaged out and are characterized by the is o t r o p i c s p l i t t i n g constant a Q. a„ = 1/3(T +T +T ) [7] o xx yy zz L J The g values also average out g = 1/3(g +g +g ) [8] M uxx yyy M z z L J Fig. 1.17 shows the effect of a l t e r i n g the spin probe motion v i a temperature. Decreasing the temperature increases solvent v i s c o s i t y and thus decreases the motion of the probe. At low temperatures, a pol y c r y s t a l l i n e array of nitroxides i s obtained and the l i m i t i n g lineshape spectrum i s obtained (often c a l l e d the "powder," " r i g i d " or "glass" spectrum). Very l i t t l e molecular motion occurs. A l l possible orientations Fig. 1.17 The ef f e c t of vis c o s i t y on ESR spectra. The structures of the spin labels used are given at the head of the corresponding column of spectra. The molecular motion was controlled by a l t e r i n g the solvent v i s c o s i t y . Most mobile, top spectrum, least mobile (characterized by the s p l i t t i n g 2T), bottom spectrum. (Taken from Jost & G r i f f i t h s , 1978) - 48 -of the nitroxide contribute to the spectrum, which i s simply the sum of resonances due to the orientations i n Fig. 1.16 together with a l l others. The outer extremes are due to the nitroxides orientated about the z-axis, p a r a l l e l to the external f i e l d , with s p l i t t i n g = 2Tz_,. The correlation time for rotati o n a l reorientation i s denoted T . *c Between the non-viscous solution for which T = l C T ^ s and the r i g i d state, characterized by T c = 10~ 7s, p a r t i a l averaging of the anisotropic quantities occurs. The spectrum i s sensitive to motional effects between these two extremes, a time scale relevant to the motions of a variety of b i o l o g i c a l systems. As the motion of the probe slows down, asymmetric broadening of the spectrum occurs (see Fig. 1.17). The direc t i o n of the g anisotropy i s such that the h i g h - f i e l d t r a n s i t i o n , which corresponds to m^  = -1, begins to broaden before the low- f i e l d l i n e , which i n turn broadens before the centre l i n e . From ESR spectra one can calculate an apparent correlation time by measuring linewidths and peak-to-peak heights (h). Two often used formulae are: T c = 6 x l 0 - 1 0 w n [ ( h 0 / h _ 1 ) 1 / 2 + ( h 0 / h 1 ) 1 / 2 - 2 ] s [9] or T c = 6 x l 0 " 1 0 w 0 [ ( h 0 / h _ 1 ) 1 / 2 - l ] s [10] (see Nordio, 1976, Fe l i x & B u t t e r f i e l d , 1980 and Keith et a l , 1970) where h^ are the peak heights of the 1, 0 or -1 l i n e s and w^  i s the linewidth of the centre l i n e (see Fig. 1.15). - 49 -This treatment assumes that the molecular motion i s i s o t r o p i c , that tumbling motion i s s u f f i c i e n t l y slow to influence the linewidth and that the three l i n e s do not overlap. At the X-band, the equations are applicable -11 -9 from 5 x 10 to 5 x 10 s. Very slow tumbling may be characterized by the s p l i t t i n g (2T, Fig. 1.17) between the outermost extrema of the spectra. For accurate calculations, spectral simulation, i s necessary to minimize effects of unresolved hyperfine s p l i t t i n g and conformational interconversions of the probe, which also contribute to spectral broadening (Schreier et a l , 1978). Small probes i n i s o t r o p i c solutions can s t i l l undergo anisotropic motion (Schreier et a l , 1978) and i f attached to other compounds may also exhibit anisotropy. I f anisotropic motion i s present, at least two T values must be calculated, Tn about the p r i n c i p a l axis (the z axis i n Fig. 1.16) and Tj_ about the molecular axis the x axis i n Fig. 1.16) (see Luoma et a l , 1982 and Schreier et a l , 1978). When the rotational anisotropy i s small and motion rapid, Tc i s often calculated using equations [9] and [10]. Sometimes Tc i s dispensed with completely and the h Q/h ^ or h 0/h^ +h 0/h ^ Quoted as empirical indices of mobility as was done by Sharom & Grant (1977) and Davoust et a l (1981). In the present work the exact nature of the tumbling motion i s considered less important than changes i n tumbling rates due to external perturbations or the physical state of the system. Two aspects of the Hamiltonian yet to be discussed are the electron-electron exchange and the electron-electron dipolar interaction. In electron-electron, or Heisenberg exchange, the wavefunctions of two different unpaired electrons overlap and e l e c t r o s t a t i c interactions tend to - 50 -couple the spins. The exchange energy i s i s o t r o p i c ; i n solution exchange occurs when the radicals diffuse together and c o l l i d e . The electrons exchange spin states rapidly, shortening the l i f e t i m e of the states without affecting the t o t a l energy of the system. In d i l u t e solutions such as those used i n t h i s thesis, such events occur s u f f i c i e n t l y rarely that they don't contribute s i g n i f i c a n t l y to the linewidth (Chapter 3, Table 3.6). Exchange broadening by the paramagnetic metal ion complex Fe(CN)^" occurs, resulting i n a broadened nitroxide signal without frequency s h i f t or exchange narrowing (Morse, 1977; Aplin, 1979). A l l nitroxides accessible to t h i s agent broaden completely at high enough Fe concentration, leaving only protected nitroxides to give a si g n a l . This i s one way of assaying the location of nitroxides i n complex systems. The dipolar interaction i s direction-dependent and i s averaged to zero by rapid tumbling, so that i t has an opposite viscosity/T dependence to that of strong exchange (which depends on c o l l i s i o n frequency), increasing as motional averaging becomes less complete at high v i s c o s i t y . I t reaches a 2 3 l i m i t i n g value which depends on (l-3cos e)/r (where 6 i s the angle between the vector r_ and the external f i e l d H and r i s the interelectronic distance between the two nit r o x i d e s ) . From empirical c a l i b r a t i o n methods ( o r i g i n a l l y done by Kokorin et a l , 1972, and s p e c i f i c a l l y for TEMPAMINE by Waterton & H a l l , 1979; Aplin, 1979 and Yalpani, 1980), one i s able to estimate the mean nearest-neighbour distance i n a pol y c r y s t a l l i n e array between nitroxides i f kept below 2 mM (Heisenberg exchange also affects lineshape at high r a d i c a l concentrations). At these low concentrations, p a r t i a l saturation i s -A d i f f i c u l t to avoid, even at the lowest microwave power of 1.6 x 10 W. - 51 -Fig. 1.18 shows the spectral parameters used to calculate the distances. From studies at 77°K, equation [11] was phenomenologically derived (see Waterton & H a l l , 1979; Aplin, 1979 and Yalpani, 1980) c y d = (dj/d) d i l + 0.58r - 3 [11] where (d-^/d) d i l are the spectral parameters defined i n F i g . 1.18 at i n f i n i t e d i l u t i o n of the spin probe (where no dipolar interactions occur). The above equation i s v a l i d i n the range 1.0 to 2. A nm. Fi g . 1.18 Powder spectrum showing the heights d^ and d used i n cal c u l a t i n g average distances between nitroxides i n equation 11. l.A(d) A r t i f a c t s Assuming the instrumentation i s acceptable (see Jost & G r i f f i t h s , 1978 - 52 -and Marsh, 1981) and the sample contains no other paramagnetic compounds to broaden spectra, the only r e a l problem i n deriving information from ESR experiments i s spectral misinterpretation. One must have an experienced eye when analysing spectra to avoid such misinterpretation (Berliner, 1978). For instance, l o c a l i z e d increased concentrations of the spin l a b e l may re s u l t and/or the spin probe may be insoluble and precipitate. This results i n spectra ranging from a single broad l i n e (due to Heisenberg exchange) to a powder-type spectrum. Both would be superimposed over the t y p i c a l three l i n e s giving the resultant spectrum a broadened appearance. Conversely, the probe may hydrolyse o f f or be attached to a proteolytic fragment, resulting i n a more mobile spectrum. More mobile signals are easier to detect ( i t takes only 1/40 as much free signal to give the same peak height as an immobilized signal (Jost & G r i f f i t h , 1978)) and thus even very minor hydrolysis w i l l y i e l d a resultant spectrum which looks more mobile than that which characterizes the bulk of the sample. One must remember also that a reporter group i s being introduced which could s i g n i f i c a n t l y perturb the system of in t e r e s t . There i s also ambiguity as to the type of motion the probe r e f l e c t s . Is i t attached to a segment which i s more mobile or more r i g i d than the rest of the system? Is a change induced i n the spectrum just a change i n the immediate v i c i n i t y of the nitroxide or i s the rest of the system also changing i n a s i m i l a r manner? I t i s d i f f i c u l t to approximate the anisotropic tumbling model (Berliner, 1978), so computer simulations used to f i t experimental results often provide a number of values which equally well describe the data (Schreier et a l , 1978). - 53 -1.4(e) Advantages One advantage of the spin probe method over other spectroscopic ones i s the s e n s i t i v i t y of ESR: very l i t t l e sample i s required, a necessary condition for working on many b i o l o g i c a l systems. Distances between spin labels and other paramagnetic materials can be determined. The p o l a r i t y of the nitroxide environment i s e a s i l y obtainable as i s i t s molecular tumbling rate. The time scale of 10~ 7s to lO'^^s i s useful and (although not used i n t h i s thesis) with the advent of saturation transfer ESR (see Hyde & Dalton, 1972 and Devaux et a l , 1981), correlation times from 10~^s to lO'^s are accessible. Modern spectrometers are equipped with computers which can d i g i t i z e spectra and u t i l i z e sophisticated programs to manipulate the data, resulting i n more accurate interpretations. One can quantitate the spin l a b e l yields by double integration of spectra, and, with the chemistry now available, l a b e l almost any material. - 54 -CHAPTER TWO 2.1 INTRODUCTION The introduction of spin labels into red c e l l s i s not a new technique. Numerous studies have dealt -with spin labeled l i p i d s being inserted into the bilayer (e.g. Shiga et a l , 1977; Suda et a l , 1980; B u t t e r f i e l d , 1982). The peptide parts of membrane proteins have been modified with spin labels (e.g. Fung & Simpson, 1979; Yamaguchi et a l , 1982) and i n some cases the spin probe of interest wasn't attached to any component but was allowed to permeate into the red c e l l (e.g. Ross & McConnell, 1975; Schnell et a l , 1983). Information on properties such as membrane f l u i d i t y (Shiga & Maeda, 1980), protein conformation changes (Lammel & Maier, 1980; Schneider & Smith, 1970), p r o t e i n - l i p i d i n teraction ( B i e r i et a l , 1975; Wallach et a l , 1974), transport mechanisms (Ross & McConnell, 1975; Zimmer et a l , 1981) and vis c o s i t y and e l a s t i c i t y of the c e l l membrane (Noji et a l , 1981) have been elucidated by the above techniques. Only one other laboratory (that of Butterfield) aside from those assoc-iated with the present work has attached the spin probe to the glycocalyx of the red c e l l , an important part of the membrane (see Introduction). The findings of t h i s laboratory w i l l be discussed i n the following chapters. This chapter deals with quantitation of the spin labeled red c e l l system and covers the modification of galactose i n the glycocalyx. These studies show the necessity for eliminating the background spin probe (by lysing the c e l l ) and the need for high y i e l d s and s p e c i f i c i t y i n the modification of the red c e l l membrane for spectral interpetation. - 55 -2.2 MATERIALS AND METHODS 2.2(a) Collection of red blood c e l l s Blood was drawn from healthy human volunteers by venipuncture into the anticoagulant sodium c i t r a t e (0.38%) or EDTA (10.5 mg/7 ml blood). Occasionally, blood was obtained from the Red Cross (in citrate-phosphate-dextrose (Masouredis, 1972)). Blood was used the same day i t was collected unless otherwise indicated. P l a s t i c ware was used i n a l l handling of blood. Outdated blood was rejuvenated to restore depleted ATP le v e l s by the method of V a l e r i & Zaroulis (1972). Red c e l l s were separated from the plasma by table top centrifugation (1000 x g for f i v e min) and then washed three times i n phosphate buffered saline (16.7mM Na 2P0 4, 3.3 mM NaH2P04, 130.4 mM NaCl pH 7.4 (PBS) i n 0.025% NaN^ (PBS/azide)) removing the buffy coat (containing white c e l l s and p l a t e l e t s ) each time. A wash r a t i o of one volume red c e l l s to 25 volumes PBS/azide was used unless otherwise indicated. A l l blood samples were kept at 4°C or on ice unless otherwise indicated. 2.2(b) Galactose oxidase oxidation of red c e l l s Red c e l l s were f i r s t treated with the enzyme neuraminidase from Vibrio  cholerae (Calbiochem-Behring Corp., La J o l l a , CA) at one volume of red c e l l s to f i v e volumes of neuraminidase (0.04 U per 10 ml of 37 mM t r i s HC1 pH 6.9, 114 mM NaCl, 4 mM CaCl 9) and incubated for one hour at 37°C. These - 56 -c e l l s were then washed three times i n PBS and incubated with one volume of galactose oxidase (Dactylium dendroides from ICN Pharmaceuticals Inc., Cleveland, Ohio) at 10 units/ml PBS for one hour at room temperature. In some cases the neuraminidase step was l e f t out. In one instance the neuraminidase and galactose oxidase were added together i n PBS and incubated for an hour at 37°C and then washed three times i n PBS/azide. In a l l procedures controls were also run and treated i d e n t i c a l l y except no oxidizing agent was present. 2.2(c) Spin labeling To one volume of washed oxidized c e l l s or controls was added one volume of 2,2,6,6-tetramethyl-4-aminopiperdine-l-oxyl (TEMPAMINE) from Aldrich Chemical Co., Milwaukee, Wisconsin. (Originally TEMPAMINE from Molecular probes Inc. (Piano, Texas) was used but was found to be impure (M. Yalpani, personal communication)). The concentration of TEMPAMINE was varied from 0.6 to 4 mg/ml PBS/azide. One volume of NaBH^CN (Aldrich Chemical Co.) at fiv e times the molar concentration of TEMPAMINE i n PBS/azide was added at the same time and the c e l l s incubated for two hours at room temperature. The reaction was stopped by d i l u t i n g the solution by a factor of three with PBS/azide and washing three times i n PBS/azide. 2.2(d) Ghost preparation Lysis of red c e l l s , oxidized c e l l s , labeled c e l l s and controls was done by the method of Dodge et a l (1963) i n 20 id e a l milliosmoles phosphate - 57 -buffered solution at pH 8.0 i n 0.025% azide. One volume of c e l l s was lysed with AO volumes of the l y s i s buffer. Ghosts were centrifuged at 20,000 x g's at A°C for 20 minutes. Care was taken to remove any dense small white p e l l e t at the bottom of the tubes which could contain proteolytic enzymes (Fairbanks et a l , 1971). The procedure was repeated three more times producing (usually) pearly white membrane preparations (ghosts). 2.2(e) Microelectrophoresis The electrophoretic m o b i l i t i e s of red c e l l s or modified red c e l l s were measured i n a c y l i n d r i c a l chamber es s e n t i a l l y as described by Seaman & Heard (1961) (see also Seaman, 1975). The chamber was immersed i n a water bath at 25°C and measurements were made at an e l e c t r i c f i e l d strength of A.O volts/cm. Electrodes were of s i l v e r / s i l v e r chloride. 2.2(f) Viscometry A viscometric assay was performed i n a Couette viscometer (Contraves LS 20) as described by Greig & Brooks (1979). This involves measurement of the shear stress resulting from the application of a constant shear rate to a red c e l l suspension. I f upon exposure to a l e c t i n an increase i n the shear stress occurs, agglutination has taken place. The f r a c t i o n a l increase provides a quantitative index of the degree of aggregation present. - 58 -2.2(g) Electron Spin Resonance Spectra were recorded at the X-band i n the derivative absorption mode on a Varian E-3. Spectrometer settings (modulation amplitude, f i l t e r time constant and scan rate) were chosen to avoid spectral d i s t o r t i o n and power levels were non-saturating and the f i e l d always increased from l e f t to r i g h t . The settings chosen were 5.7 mW power and a 1 gauss (G) modulation. A l l aqueous samples were placed i n a f l a t c e l l of 200 jul capacity and recorded at ambient temperature. Samples were also run at constant temperature on a homodyne spectrometer with a Varian 12 inch magnet and d i g i t i z e d . This was kindly provided by Dr. F.G. Herring. 2.2(h) Quantitation ( i ) Red c e l l or ghost count Red c e l l counts were determined by t h e i r hematocrit (volume fraction) i n c a p i l l a r y tubes spun down for f i v e min on an International micro-capillary centrifuge (International Equipment Corp., Needham Hts., MA), assuming 1.1 x 1 0 ^ packed cel l s / m l . At low concentrations, red c e l l s were counted i n a hemocytometer (a chamber which holds a known volume of red c e l l s over a grid pattern; c e l l s are then counted v i s u a l l y with a microscope) or a Model 112 CL TH/RWP p a r t i c l e counter ( P a r t i c l e Data Inc., Elmhurst, 111). Ghosts were o r i g i n a l l y counted microscopically i n an improved Neubauer hemocytometer, but due to t h e i r poor v i s i b i l i t y and low density were l a t e r determined on the p a r t i c l e counter also. For ghosts, the counter trigger levels were altered from the red c e l l settings (10 lower to 50 upper) to 5 lower and 50 upper, (other settings were current 1/4, gain 68, function switch at delta, - 59 -li n e a r control and a 76 u o r i f i c e ) . Dilutions (from 200-400) for counting were done i n PBS/azide by weight to ensure accuracy of d i l u t i o n . ( i i ) ESR Double integrations were done using free spin l a b e l of known concentration i n PBS/azide as standards. These were o r i g i n a l l y performed on the Varian E-3 with a P a c i f i c Precision Co. MP-1012A integrator. The second integration was obtained by cutting out and weighing the peaks. Integrator s e n s i t i v i t y (IS) was defined as (mass/gain) x # spins/ml of standard. The spins/ml of the unknown was determined by i t s (mass/gain) x IS. The double integrations could also be performed on the spectra with a Hewlett Packard 9815A calculator interfaced with a 9872A plot t e r using a program written by K.A. Sharp (see Appendix B). Double integrations were also performed via computer on d i g i t i z e d spectra run at constant temperature on the homodyne spectrometer kindly provided by Dr. F.G. Herring. Sample size i n t h i s case was limited to 24 ;ul. ( i i i ) Protein Protein concentrations of ghost suspensions were determined by the Lowry procedure (Lowry et a l , 1951) as modified by Markwell et a l (1978) using human serum albumin (Sigma Chemical Co., St. Louis, Mo) as the standard. Absorbance at 280 nm was determined by the method of V i c t o r i a et a l (1981) except i n 0.5% SDS instead of 0.2% SDS. - 60 -2.2(i) Miscellaneous K 3Fe(CN) 6 (80mM) (BDH Chemicals Canada Ltd, Vancouver) was added to spin labeled red c e l l s and red c e l l s plus spin l a b e l according to Morse (1977). Lectins from Glycine max (soybean agglutinin (SBA)) type VI, Bandeiraie  s i m p l i c i f o l i a , Dolichis b i f l o r u s (horse gram), Abrus communis ( j e q u i r i t y bean agglutinin, and Ricinus communis (Castor bean type 11 (RCA 11)) a l l bind galactose derivatives and were purchased from Sigma. Concanavalin A (Con A), a mannose binding l e c t i n , was from Calbiochem-Behring. Bound s i a l i c acid was assayed as i n Reid et a l (1977). - 61 -2.3 RESULTS  2.3(a) Quantitation ( i ) Ghost counts I t was found using the hemocytometer that only 8 of the 80 squares could be counted s u f f i c i e n t l y accurately. This was due to the fact that red c e l l ghosts didn't s e t t l e onto the grid pattern so one had to constantly raise and lower the microscope's f i e l d of v i s i o n to ensure that one had counted a l l the ghosts. Thus 8 random squares were counted and the number multiplied by 10 to give the t o t a l i n that g r i d . Using the p a r t i c l e counter i t was found that the average apparent volume of the ghosts was lower than that of red c e l l s (Fig. 2.1) so the window settings had to be lowered to ensure a l l the ghosts were counted. I t was observed that the % of the t o t a l population that occupied the lowest volume i n t e r v a l (5-10 s e t t i n g ) , although consistent for each ghost preparation, varied considerably from sample to sample. The only correlation of sample treatment to the % i n t h i s i n t e r v a l was that the ghosts incubated at 37°C were larger i n size and thus occupied a smaller % of t h i s window (Fig. 2.1). The in t e r n a l consistency was much better for the p a r t i c l e counter than for the hemocytometer. For the p a r t i c l e counter, each sample was consistent (within 9 +_ 7%, n = 30) regardless of d i l u t i o n . When samples of known ghost counts were accurately diluted and counted again the results were within 7 _+ 6% of that calculated. The % i n the 5-10 window was also very consistent (_+ 5%) for a given sample, but varied tremendously from sample to sample, giving r i s e to inconsistencies i n the 10-50 window r e s u l t s . - 62 -Fig . 2.1 Graph of the d i s t r i b u t i o n of red c e l l populations and the window settings of the p a r t i c l e counter. Abscissa, the % distribution/window ordinate, the window se t t i n g . * * red c e l l s ; O ® ghosts o -O ghosts a f t e r incubation at 37°C for AO hours. - 63 -The advantages of the p a r t i c l e counter are that i t i s faster, more consistent, and more c e l l s can be counted per sample (thousands compared to about 100 for the hemocytometer). Using the p a r t i c l e counter one can count ghost populations to within 10% accuracy. ( i i ) Protein quantitation of ghosts The modified Lowry gave a good standard curve (coef f i c i e n t of determination = 0.992). Table 2.1 tabulates the results and also l i s t s values quoted i n the l i t e r a t u r e . TABLE 2.1 COUNTING METHOD AND GRAMS PROTEIN/GHOST OBTAINED GRAMS PROTEIN/GHOST METHOD X l O 1 3 (+ SD) RANGE (n) Hemocytometer 3.67 + 0.38 3.A-A.0 (6) C e l l counter 6.15 + 0.6A 5.7-6.6 (6) Fairbanks et a l (1971) 5.7 Dodge et a l (1963) 6.6 Khodadad & Weinstein (1982) 5.5 The OD at 280 nm of ghosts (1 mg/ml) i n 0.5% SDS (used to s o l u b i l i z e the protein and prevent l i g h t scatter) was 1.2 _+ 0.1 (range 1.07-1.32, n = 6), i d e n t i c a l to that of V i c t o r i a et a l (1981) i n 0.2% SDS. The results of the l a s t two sections support the use of the p a r t i c l e counter as the method of choice for determining ghosts/ml. - 64 -( i i i ) ESR integration Or i g i n a l l y the integrations were performed on the Varian E-3 using a P a c i f i c Precision integrator. This was found to be tedious and inaccurate. The trace had to be cut out and weighed to obtain the second integration and there was always baseline d r i f t on the integrator at the gains used (5-10 x 10 5). Errors came i n the placement of the sample, the rep r o d u c i b i l i t y of power settings and the microwave coupling (see Goldberg, 1978). When a sample was taken out, put back i n , the machine retuned and another integration run, the r e l i a b i l i t y turned out to be low (a 24% err o r ) . Calculating spins/ml would have approximately doubled t h i s error due to error i n IS (calculated from the spin standard) plus the error i n the integration of the unknown. A program written by K.A. Sharp for a Hewlett Packard calculator and plotte r u t i l i z e d manual d i g i t i z a t i o n of the ESR spectrum (150 points, see Appendix B). Over an extended period of time (3 years), the IS using _9fi Sharp's program was 2.56+0.82 x 10 (n = 55, range 1.36-5.79), an uncertainty of 33%. The advantage of t h i s program i s ease. One can integrate the spectrum when time i s available and can reintegrate i f needs be (reintegrating the same spectrum has a 4 + 1 ^ accuracy, n = 6, range 1.7-5.5). Integrating the same sample but a dif f e r e n t spectrum was fraught with the problems c i t e d above (gain accuracy and sample repr o d u c i b i l i t y ) and was found to produce differences from 3 to 23%, averaging 15 _+ 10%. The problem with t h i s program i s that the spectrum can't be too broad or have too much baseline d r i f t . Considering how the f i r s t integration i s zeroed (see Appendix B), one would expect the SL-ghost integrations to be slig h t y underestimated, due to s l i g h t broadening, compared to the free spin l a b e l . - 6 5 -Quantitation v i a the temperature controlled, computer d i g i t i z e d spectrometer was di f f e r e n t . Sample size was smaller due to the temperature controlled cavity ( 2 4 jul compared with 2 0 0 J J I for the Varian E - 3 ) . One put i n a standard sample containing a known number of spins (as opposed to a known concentration); unknown samples had accurately measured volumes (measured by weight). The spectra were d i g i t i z e d , stored on tape and integrated via a computer program devised by P.S. P h i l l i p s and F.G. Herring (Herring & P h i l l i p s , 1 9 8 5 ) . The baseline of t h i s integration was then zeroed and the second integration performed. The answers obtained for spins/ml were very close to those calculated v i a the Hewlett-Packard (H-P) program on a i d e n t i c a l sample run on the Varian E - 3 . The H-P program yielded 2 . 5 x 1 0 1 6 spins/ml and the computer d i g i t i z e d spectrum yielded 2 . 3 5 x 1 0 1 6 (another sample gave 7 . 3 x 1 0 1 6 for the H-P and 9 . 9 x 1 0 1 6 for the computer d i g i t i z e d ) . A l l values quoted as spins/ml or spins/ghost are derived from the H-P program. The end result of t h i s quantitation i s that when quoting spins/ghost an uncertainty of about 2 0 - 4 0 % i s inherent i n the c a l c u l a t i o n . 2 . 3(b) Galactose spin labeling of red c e l l s I t was found from c e l l microelectrophoresis data (Table 2 . 2 ) that the simultaneous combination of neuraminidase and galactose oxidase (GO) wasn't as e f f i c i c e n t i n removing s i a l i c acid as was the sequence of neuraminidase digest followed by GO. This i s shown by the fact that the electrophoretic mobility of the c e l l s treated simultaneously didn't decrease as much as i n - 66 -the sequential experiment (Table 2.2). In these experiments (collaborating with M.A. Bernstein and R. Greig) a t o t a l of 5 x 10 4 s p i n s / c e l l was found for the sequential digest and spin labeling of the red c e l l . TABLE 2.2 CELL ELECTROPHORESIS OF ENZYME TREATED AND SPIN LABELED RED CELLS Sample Mobility (jum sec-J-V - 1 cm) After spin labeling Control -1.08 + 0.03 -1.08 + 0.01 NANase treated red c e l l s -0.65 + 0.03 -0.66 + 0.01 NANase then GO treated red c e l l s -0.66 + 0.05 -0.65 + 0.05 NANase with GO treated red c e l l s -0.90 + 0.02 -0.90 + 0.07 Further experiments were conducted with either neuramindase followed by GO (called NAGO labeled c e l l s ) or with GO by i t s e l f (called GO labeled c e l l s ) , followed by reductive amination. Analysis of these spin labeled c e l l s ( S L - c e l l s ) , labeled at 0.6 mg/ml TEMPAMINE, showed a freely mobile spin probe population whose T (correlation time, Chapter 1, section A) didn't change upon addition of a variety of galactose binding l e c t i n s , even though i n a l l cases the c e l l s were obviously agglutinated (Fig. 2.2 and Table 2.3). A l l ESR spectral data are .expressed i n terms of t h e i r spectral parameter, SP, defined below (the spectral parameter i s a r e f l e c t i o n of the spins' mobility, the higher the parameter the slower the mobility): - 67 -SP = ( r v/h^ )^ + (h Q /h + 1 ) x / z -2 TABLE 2.3 SP FOR NAGO SL-RBC PLUS LECTINS SAMPLE RELATIVE GAIN1 SP NAGO SL-RBC 1 0.10 + 0.01 Con t r o l 2 40 0.11 NAGO SL-RBC + 250 ug l e c t i n SBA 1.24 0.07 RCA 11 1.24 0.11 PNA 2.5 0.09 Bandeiraie s i m p l i c i f o l i a 2.5 0.10 ^Gain of samples on the Varian E-3 spectrometer r e l a t i v e to that set for SL-RBC 2Red c e l l s treated as i n section 2.2b and c but without enzymes NANase and GO It became apparent that the signal observed was that of free spin label trapped inside the red c e l l ( f i r s t observed by Morse, 1977). Washing red c e l l s exposed to spin l a b e l (at 4 mg/ml) u n t i l no signal was detected i n the supernatants s t i l l resulted i n an eas i l y detectable signal from the red c e l l p e l l e t s (SP = 0.19). Lysing these c e l l s at a 1:1 r a t i o with ghost buffer produced a spin l a b e l population i n the supernatant (SP = 0.03). In another experiment, 80mM K^Fe(CN)6 (isosmomolar to red c e l l s ) was added to these c e l l s and a signal was s t i l l detected i n the red c e l l p e l l e t s (with an increased SP = 0.19). Addition of 80 mM K,Fe(CN) to - 68 -Fig. 2.2 ESR spectra of (A) control c e l l s (treated as i n (B) but without neuraminidase or GO added) and (B) NAGO SL-RBCs made as described i n protocol, section 2.2(b) and (c). (C) i s sample (B) with the addition of 250 jug of SBA l e c t i n . See Table 2.3 for d e t a i l s . - 69 -one mM TEMPAMINE i n PBS resulted i n a completely broadened spectrum barely detectable at a gain 48 times higher. Table 2.4 l i s t s the spectral parameter vs the hematocrit of red c e l l s exposed to TEMPAMINE with or without 80mM K 3Fe(CN) 6. TABLE 2.4 SP AT DIFFERENT HEMATOCRITS OF RBC PLUS TEMPAMINE Hematocrit of red c e l l s SP i n the presence plus 1 mM TEMPAMINE SP of 80mMK3Fe(CN)6 41% 0.04 0.19 76% 0.13 0.23 84% 0.16 0.19 Another confirmation of spin l a b e l noncovalently associating with the red c e l l came from the observation that the signal decreased with increased washing. Perhaps the most sensitive indicator was the ESR signal i t s e l f . Samples containing free spin l a b e l always exhibited a free population spectrum which overrode that of the covalently bound species. Table 2.5 l i s t s the SP observed following different sample manipulations. In some cases extensive washing resulted i n spectral changes. A NAGO SL-red c e l l after 10 washings produced a spectrum (SP = 0.33) which was noticeably changed from that of the controls (SP = 0.15), but addition of the j e q u i r i t y bean l e c t i n (250 yjg) s t i l l resulted i n no further spectral changes (SP = 0.34) even though obvious agglutination had occurred. - 70 -TABLE 2.5 SAMPLE MANIPULATION OF SL-RBC AND RESULTANT SP SAMPLE NAGO SL-RBC GO SL-RBC SAMPLE MANIPULATION SPINS/CELL SP SPINS/CELL SP Washed f i v e times 2.5 x 10 6 0.24 1.7 x 10 6 0.16 Washed seven times 1.9 x 10 5 0.30 1.1 x 10 5 0.22 Made into ghosts 1.9 x 10 5 0.62 1.1 x 10 5 0.55 Better analysis could be done u t i l i z i n g ghosts made from spin labeled red c e l l s . NAGO and GO SL-ghosts had SP varying from 0.41 to 0.62 averaging 0.54 +_ 0.09 (n = 4) (Table 2.6) with the GO labeled ghosts being more consistent. TABLE 2.6 SPIN LABELED SAMPLES MADE INTO GHOSTS AND THEIR SP SAMPLE SPINS/GHOST SP NAGO SL-ghosts 2.0 X 10 5 0.62 0.8 X 10 5 0.42 GO SL-ghosts 5 X 10 5 0.58 1 X 10 5 0.55 - 71 -Fluctuations are to be expected due to low yiel d s (the maximum y i e l d 5 5 was 5 x 10 spins/ghost with the resultant gain being 10 x 10 ). Fig. 2.3 shows a t y p i c a l spectrum of GO SL-ghosts. The resultant SDS PAGE (Fig. 2.4) shows that these labeled ghosts are comparable to the controls (red c e l l s treated as stated i n the methods for spin labeling red c e l l s but no galactose oxidase present). Addition of 250 jjg (50 J J I) of PNA or SBA to 1 ml of NAGO SL-ghosts resulted i n v i s i b l e agglutination and a s l i g h t increase i n the spectral parameters (Fig. 2.5). -12 -Fig. 2.3 An ESR spectrum of GO SL-ghosts made as described i n the text. I t was run on a Varian E-3 with the gain set to 10 X 10 5, modulation 1 G, and power at 5.7 mW. SP = 0.58 - 73 -A B I / v——•*• ' t, FRONT FRONT Fig. 2.4 SDS PAGE of (A) GO SL-ghosts made as i n protocol and (B) untreated ghosts. The top scans are densitometric tracings of coomassie blue stained gels loaded with 45 ug of membrane protein. The bottom scans are of PAS stained gels loaded with 180 ug of membrane protein. The numbering system i s that of Fairbanks et a l , 1971. - 74 -Fig. 2.5 ESR spectrum of (A) NAGO SL-ghosts made as described i n the text along with the spectra of these ghosts with the addition of 250 jul (50 p i ) l e c t i n s , (B) PNA and (C) SBA. The spectral parameters, SP, are also indicated. - 75 -To further study the effects of PNA upon red c e l l s , neuraminidase treated c e l l s and NAGO SL-red c e l l s were exposed to PNA i n a viscometer to assay the agglutination induced by PNA with these treated c e l l s . Unmodified red c e l l s sheared i n the presence of PNA showed no noticeable agglutination i n the viscometer. The neuraminidase treated and the NAGO SL-ce l l s , however, exhibited behaviour c h a r a c t e r i s t i c of strong agglutination. Fifteen micrograms of PNA added to a 47% solution of NAGO SL-cells showed the t y p i c a l shear-enhanced agglutination response when plotted as shear stress against time (Fig. 2.6) (Greig & Brooks (1979) analysed the system with respect to Con A agglutination). This system, however, was not i n h i b i t e d or reversed by galactose, the monosaccharide i n h i b i t o r of PNA. Because t h i s measurement works only on red c e l l s , the free spin l a b e l couldn't be eliminated and the resultant spectrum from the highly agglutinated c e l l s was no diff e r e n t from the c e l l s not exposed to PNA or to the viscometer, a l l giving the t y p i c a l free spin l a b e l spectra. - 76 -o D_ E C\J I S I T I X CO CO LU CU I— CO en UJ ZLZ CO 10 8 6 0 B • PNA i i i i i 0 10 15 20 25 30 TIME (MIN) F i g . 2.6 The shear s t r e s s recorded as a f u n c t i o n of time f o r aggregating e r y t h r o c y t e s . Red c e l l s were suspended at a f i n a l hematocrit o f 47% and 0.9 ml placed i n the cup of the viscometer. Temperature was maintained at 37.0 +_ 0.1°C. At a constant shear r a t e of approximately 49s--'- (37 rpm) a steady b a s e l i n e shear s t r e s s , due to the v i s c o s i t y of the unaggregated c e l l suspension i s obtained. PNA was added as i n d i c a t e d i n a volume of 40 u l . (A) i s the n e u r a m i n i d a s e t r e a t e d red c e l l s with 20 ;ug of PNA added and (B) i s the NAGO SL-RBC with 10 /jg PNA added. - 77 -2. A DISCUSSION OF SPIN LABELING RESULTS (i ) Red c e l l s (a) Microelectrophoresis The neuraminidase and GO treatment together was not as e f f i c i e n t as the sequential addition of these two enzymes as seen by t h e i r electrophoretic mobility measurements. Mobility i s a measure of the surface charge distributed through the glycocalyx (Levine et a l , 1983) to which s i a l i c acid i s the major contributor (Eylar et a l , 1962; Seaman et a l , 1977). Vassar et a l (1972) found a 60% decrease i n electrophoretic mobility of red c e l l s after neuraminidase treatment (however, Luner et a l (1975) found that removal of 83% of the s i a l i c acid with neuraminidase resulted i n an 80% drop i n the electrophoretic mobility). Table 2.2 shows that mobility decreased A0% for the neuraminidase and NAGO treated c e l l s , indicating that not a l l the s i a l i c acid had been removed. The red c e l l s that had been treated simultaneously with neuraminidase and GO showed only a 16% decrease i n mobility indicating that less s i a l i c acid had been removed. This decrease i n s i a l i c acid removal i s probably due to the incubation buffer being optimal for GO and not neuraminidase (requiring lower pH and C a 2 + ) , thus the neuraminidase was not as e f f i c i e n t . The important res u l t from t h i s data i s that the m o b i l i t i e s of these c e l l s were not altered by the reductive amination step. This indicates that t h i s type of modification r e s u l t s i n no further change i n the surface charge of the red c e l l as detected by microelectrophoresis. The SDS PAGE gels of GO SL-ghosts don't appear di f f e r e n t from those of normal red c e l l s (Fig. 2.A), showing that the GO labeling method i s mild. - 78 -(b) Spin labeling Due to low yields (5 x 10 5 s p i n s / c e l l maximum) the spectra are noisy (gains set to 10 x 10 ) and d r i f t can occur (times for running a spectrum are from 8-30 min). The resultant signal-to-noise r a t i o i s thus lower than desired. For these reasons spectral parameters (SP) have been quoted instead of Tn calculations, since w (the center l i n e peak width) i s hard to determine accurately (Lee et a l , 1980). These weak signals make spectral interpretation more d i f f i c u l t and uncertainties high (20% accuracy at the gains quoted). This could mask s l i g h t systematic alterations i n the samples, since changes of 10% would go unnoticed and 20% changes would be viewed as questionable. The major problem i n spin labeling b i o l o g i c a l systems i s the removal of non-reacted spin probe. This population i s usually free unless partitioned into a different environment or precipitated out. Spin probes measure the micro-environment, as opposed to the macro-environment, of a solution. The solution may appear very viscous and s t i l l give ESR signals of freely spinning nitroxides, i f the micro-environment of that nitroxide has a low visc o s i t y (see Aplin, 1979 and Morse et a l , 1979). This freely spinning population i s readily detected because i t s spectrum exhibits the t y p i c a l three sharp l i n e s which can be seen down to about 10~ 6 M. This spectrum i s superimposed over the covalently attached spectrum and can override the bound si g n a l . This free spin probe i s inside the red c e l l ( f i r s t observed by Morse, 1977, see also Morse et a l , 1979, and Bartosz & Leyko, 1980). Extensive washing of the red c e l l s never eliminated the signal from the controls ( c e l l s + TEMPAMINE) even i f washed, l e t s i t overnight and washed again. No - 79 -signal was detected i n the supernatants of these c e l l s , but upon lysing (1:1 with ghost buffer) a signal was observed i n the supernatant (SP = 0.03). For red c e l l controls (red c e l l s plus TEMPAMINE), SP was 0.135+0.015, indicating a s l i g h t increase i n Xc (Tc for free TEMPAMINE i n aqueous solutions has been reported to be 4.3 x 10~^s (Bartosz & Leyko, 1980) or 4.6 x 1 0 - 1 1 s (Morse et a l , 1979), which corresponds to a spectral parameter of about 0.04 assuming wQ= 1.5 G and K = 6.5 x 1 0 - 1 0 and using equation [9] i n Chapter 1 (p. 48) which agrees well with the SP obtained for TEMPAMINE i n PBS/azide i n t h i s work: 0.03 + 0.01 (n = 10)). As the hematocrit of controls was increased, so did SP (Table 2.4). The fact that SP increased with hematocrit indicates that the spin probe inside was spinning at a slower rate, the more mobile signal i n solution e x t r a - c e l l u l a r l y originating from slow leakage out of the c e l l . I f one extrapolates the data i n Table 2.4 to 100% hematocrit, an SP of 0.198 i s obtained (coefficient of determination = 0.997) which results i n a calculated T/c °f l* 9 x 10~^s. Interestingly, using the data i n Table 2.4, i t was calculated that at hematocrits of 36% or les s , the signal obtained from control suspensions would be equivalent to that of TEMPAMINE in PBS, the signal from inside the red c e l l s being t o t a l l y undetectable. Additional proof of the location came using the method of Morse, 1977 (addition of 80 mM Fe(CN)^, a paramagnetic ion which broadens the nitroxide signal of TEMPAMINE (Morse, 1977; Aplin, 1979)). At 80 mM K,Fe(CN),, the signal i s completely broadened out for the spin l a b e l standard i n PBS (as was found by Morse, 1977) but i t s addition to red c e l l s spin labeled or with spin l a b e l added to them s t i l l resulted i n a signal from the red c e l l p e l l e t s (Table 2.4) (Morse, 1977). K,Fe(CN), doesn't - 80 -enter red c e l l s (Morse, 1977; Mishra & Passow, 1969; Kaplan et a l , 1973), thus the observed signal must be from spins protected by the red c e l l , presumably inside i t . A T of 1 . 9 x 10~^s was calculated for the TEMPAMINE inside red c e l l s , i n good agreement with Morse (1977) (1.8A x 10" 1 0s) and Morse et a l (1979) (2.0 x 1 0 " 1 0 s ) . Spin labeling the red c e l l s via NAGO or GO resulted i n spectra such as that shown i n Fig. 2.2. As seen i n Table 2.3, addition of a variety of gal/galNAc binding l e c t i n s (and Con A, a mannose binding l e c t i n ) at concentrations which produce v i s i b l e agglutination resulted i n no change i n the spectra. The non-covalently bound spectrum predominated. Extensive washing resulted i n a decreased signal but the free signal couldn't be eliminated unless the c e l l s were lysed and made into ghosts (Table 2.5, Fig. 2.3). For the washed NAGO SL-red c e l l s (SP = 0.33), the spectra of which began to resemble those of SL-ghosts (SP = 0.5A), the addition of j e q u i r i t y bean l e c t i n s t i l l resulted i n no detectable spectral change upon agglutination (SP = 0.3A). This could be due to the free spin l a b e l masking any minor change induced by the l e c t i n , or the l e c t i n agglutination may actually produce no change. Only when the samples had been lysed were the signals v i s u a l l y d ifferent and SP increased by about 50% (Table 2.5 and 2.6). I t appears that very few, i f any, spins were removed upon lysing the c e l l s , yet the SP were dramatically d i f f e r e n t . I t might be argued that the nitroxide environment was altered upon the formation of ghosts, but the spectra of NAGO SL-cells also s t a r t to show t h i s type of increase i f enough free spin l a b e l can be removed (SP = 0.33). Only when these c e l l s were made into ghosts ( a l l the free spin l a b e l - 81 -removed) was there any detectable change upon the addition of PNA and SBA, and even then i t was only about 20% i n the spectral parameter (Fig. 2.5). The spin labeled ghosts produced are s t i l l agglutinable by l e c t i n s . Since PNA binds only to desialylated red c e l l s (Goldstein & Haynes, 1978) the NAGO SL-method was used to see i f the spectrum would be altered upon addition of PNA or SBA l e c t i n s . PNA agglutinates NAGO SL-red c e l l s as seen from the viscometry of these c e l l s , PNA producing a t y p i c a l l e c t i n p r o f i l e (Greig & Brooks, 1979, Fig. 2.6). Cel l s not treated with neuraminidase don't agglutinate while neuraminidase digested and NAGO SL-cells do. This shows that even though the c e l l s are modified with the spin l a b e l , they s t i l l bind PNA. Shear induced agglutination was not reversed by galactose, possibly due to other PNA binding s i t e s being exposed during shearing. PNA binding to the NAGO SL-cells was also indicated by the fact that v i s i b l e agglutination occurred with these c e l l s and the resultant SL-ghosts upon addition of the PNA l e c t i n . Interpretation of the effects of PNA on NAGO SL-ghosts i s complex due to the multiple membrane components being labeled, low y i e l d s , the p o s s i b i l i t y of unbound lab e l contributions and lack of knowledge of the PNA binding s i t e s . Carter & Sharon (1977) i s o l a t e d , v i a a PNA a f f i n i t y column, desialylated glycophorin and a 27 K component from red c e l l s . Jaffe et a l (1979) tentatively i d e n t i f i e d desialylated glycophorin A and a second glycoprotein (of 58-61 K) as the PNA receptors. GO has a s p e c i f i c i t y for terminal C^ and C^ OH groups of galactose and galNAc (Hamilton et a l , 1973; Schlegel et a l , 1968). Yields as high as 2-4 x 10 7 3 H / c e l l have been quoted for just the G0/NaB3H4 method (Gattegno et a l , 1981; Aminoff et a l , 1981) and are improved employing NAGO - 82 -vs GO, with two additional proteins being labeled (Gahmberg, 1976), indicating that a large number of exposed terminal gal and galNAc residues are present on the red c e l l . The GO/NaBC^H]^ labeling method shows at least 18 proteins labeled (Gahmberg, 1976) (PAS proteins, band 3, band 4.5 Mueller et a l , 1979 and high molecular weight components, Jokinen, 1981; Abraham & Low, 1980) with about 40% of the l a b e l on the l i p i d s (Gahmberg & Hakomori, 1973; Steck & Dawson, 1974). Since PNA appears to bind to only two of the 18 proteins and since 40% of the l a b e l i s on the l i p i d s , one wouldn't expect a large change i n the ESR spectrum with PNA binding, as only a small fraction of the t o t a l spin population would be perturbed. As can be seen from Fig. 2.5, there i s a s l i g h t increase i n SP with the addition of PNA and SBA. This increase could be due to the spin l a b e l slowing down s l i g h t l y but the changes are very close to experimental uncertainties. The fact that both PNA and SBA increase SP supports the assumption that these l e c t i n s are binding to the spin labeled gal/galNAc residues, however. One problem re s u l t i n g from removal of s i a l i c acid to l a b e l the galactose residues i s that a sugar unit which also carries a negative charge has been l o s t . Desialylating red c e l l s renders them agglutinable to PNA. The question a r i s e s , i s t h i s due to gal residues being exposed or i s i t , wholly or i n part, a r e s u l t of removal of s t e r i c or e l e c t r o s t a t i c hindrance by the s i a l i c acid. Removal of s i a l i c acid decreases the surface charge (Table 2.2) and t h i s fact alone appears to render them more agglutinable (Luner et a l , 1975; Greenwalt & Steene, 1974). The system has been perturbed and i t i s hard to determine i f the effect one sees i s due to t h i s perturbation alone or to secondary effects. - 83 -( i i ) T c of labeled ghosts GO or NAGO SL-red c e l l s made into ghosts had a calculated T c of 1.1 x 10" 1 0s (assuming the SP = 0.54 _+ 0.09, wQ = 2.2 (Chapter 3, Table 3.7) and K = 6.5 x l O " 1 0 ) . This value i s close to that found for the NAGO-treated fetuin (5.2 x 10~ 1 0s) and NAGO treated BSM (3.5 x 1 0 _ 1 0 s ) of Aplin et a l (1979), but i s much closer to the values obtained for spin labeled s i a l i c acid residues on ghosts by Fel i x & B u t t e r f i e l d (8.4 x 10" 1 0s) (1980), for fetuin by Aplin et a l (7.9 x 10" 1 0s) (1979) and for glycophorin by Lee & Grant (9.6 x 10~ 1 0s) (1979). A l l of the above have faster molecular motion than anticipated based on t h e i r molecular s i z e , implying that these sugar units are more free than the bulk of the molecule. The fact that S L - s i a l i c acid residues gave s i m i l a r results to the labeled gal residues implies a general freedom of motion for the oligosaccharides. This implies that the glycocalyx of the red c e l l surface (or more s p e c i f i c a l l y , the terminal gal/galNAc residues) are f a i r l y mobile. The spectrum of the GO SL-ghosts may be a composite of several different populations ( l i p i d s , PAS proteins, bands.3 and 4.5), although the yields were too low for t h i s type of analysis (labeling with NaBE^Hj^CN s t i l l resulted i n low t r i t i u m l a b e l i n g ) . The fact that PNA and SBA didn't a l t e r the signal very much implies a heterogeneous population (or possibly only minor interactions with the SL-gal/galNAc residues). I f the population i s heterogeneous, then t h i s r e s u l t i s very surprising, because about 40% of t h i s population are g l y c o l i p i d s and the appearance of f a i r l y homogeneous spectra indicates that the majority of the gal/galNAc residues are experiencing the same type of environment. The actual values for X c - 84 -quoted and i n the l i t e r a t u r e are only estimates of the true T c » however, which requires computer simulation to calculate properly (and a knowledge of how the probe i s actually r o t a t i n g ) . The advantage of using SL-sugar residues i n t h i s context, on the other hand, i s that one now has, i n p r i n c i p l e , an independent confirmation that the molecule of interest i s actually binding to the sugar unit, something which i s usually taken for granted but not shown. - 85 -2.5 CONCLUSION A spin la b e l was successfully attached to the surface of red c e l l s v i a the GO or the NAGO method followed by reductive amination with TEMPAMINE. No detectable surface charge alterations occurred due to the reductive amination step. The red c e l l samples had to be lysed to eliminate unbound spin l a b e l , to allow for proper spectral interpretation and to observe any possible spectral alterations upon the addition of l e c t i n s . From the Tc calculations, i t appears that the terminal gal/galNAc residues on the glycocalyx are a l l f a i r l y mobile, having a correlation time of 7.7 +1 x lO'^seconds. They appear to be heterogeneous i n t h e i r d i s t r i b u t i o n , being only p a r t i a l l y sensitive to PNA and SBA agglutination. Converting NAGO SL-red c e l l s into ghosts does not appear to a l t e r the mobility of these terminal sugars and thus one can use t h i s system for studying the glycocalyx. - 86 -CHAPTER 3 3.1 INTRODUCTION Due to low yields and the lack of s p e c i f i c i t y found i n the galactose oxidase reactions, the s i a l i c acids on the red c e l l surface were chosen to be modified. On the red c e l l membrane, the main s i a l i c acid containing proteins are the PAS proteins (glycophorins A, B and C and minor components). There i s no enzymatic way of oxidizing these sugars, so mild periodate treatment was used, which under proper conditions i s s p e c i f i c for s i a l i c acid. Periodate modification of red c e l l surfaces was f i r s t carried out i n 1948 by Hirst on fowl red c e l l s . This was one of the f i r s t indications that the c e l l surface contained sugars. The c e l l s ' a b i l i t y to absorb influenza virus was altered. In 1949, St. Groth made the same observation for human red c e l l s and found that t h i s effect was sensitive to the extent of periodate modification. Stewart (1949) found these periodate treated c e l l s were able to form antibodies i n rabbits and that these c e l l s could become agglutinable by the rabbit's own serum. Springer (1963), reviewing the l i t e r a t u r e , states that blood type a c t i v i t y was decreased upon periodate modification, with the M and N being the most susceptible followed by the Rh Q(D), indicating that these two blood types are most l i k e l y carbohydrate in nature. In 1964, Spiro showed that s i a l i c acid was s e l e c t i v e l y oxidized i n the periodate modification of f e t u i n , and was completely destroyed after t h i r t y - 87 -min exposure to periodate at a 7:1 r a t i o of periodate to s i a l i c acid. Van Lenten & Ashwell (1971) confirmed t h i s with the periodate oxidation of orosomucoid. Blumenfeld et a_l (1972) were the f i r s t to periodate oxidize red c e l l s and reduce the product with NaBC^H]^, thus labeling the newly generated aldehydes of the c e l l surface. They found a correlation between the p r o f i l e and the PAS stainable proteins on SDS PAGE. Neuraminidase digest of these treated ghosts (Liaos et a l , 1973) showed the major released product to be NANA., ( s i a l i c acid with the Cg and removed) with a concomitant decrease i n the PAS and H staining. Mueller e_t a l (1976), using the Laemmli gel system, labeled the ghost via the periodate/NaB[^H] 4 method and found a l l the PAS proteins (8 stainable components v i s i b l e ) labeled. The periodate oxidation method has become the method of choice for labeling the s i a l i c acids of c e l l membranes. Lymphocytes (Presant & Parker, 1976; Spiegel & Wilchek, 1983), synaptic plasma membranes (Cruz & Gurd, 1980) and p l a t e l e t s (Rotman et a l , 1980; Steiner et a l , 1983) have been labeled t h i s way. Thiol mannosyl hydrazide (Rando & Bangerton, 1979), b i o t i n hydrazide (Heitzmann & Richards, 1974; Skutelsky et a l , 1977), (via NaB[ 3H] 4) (Steiner et a l , 1983; Steck & Dawson, 1974; Kahane et a l , 1976), arylalkylamines (Schweizer et a l , 1982), eosin derivatives (Cherry et a l , 1980), thiolhydrazide (Taylor & Wo, 1980), fluoresceine amine (Abraham & Low, 1980) and spin labels (Aplin et a l , 1979, Felix & B u t t e r f i e l d , 1980) have been attached to the c e l l surface using t h i s method. - 88 -This chapter deals with the selective modification of s i a l i c acid v i a mild periodate oxidation followed by spin labeling with TEMPAMINE and NaBH^CN (reductive amination). Effects of periodate oxidation on red c e l l s and ghosts was analysed and compared to sialoglycoproteins glycophorin and fetuin. Attempts at improving y i e l d of s p i n s / c e l l with minimal perturbation of the membrane followed. Higher yie l d s (compared to GO SL-ghosts) resulted i n the opportunity to analyse SL-ghosts i n more d e t a i l . Results indicate selective labeling had occurred and information about the S L - s i a l i c acids could be obtained. - 89 -3.2 MATERIALS AND METHODS 3.2(a) Periodate oxidation of red c e l l s and ghosts To one volume of packed red c e l l s (washed as i n Chapter 2, section 2.2(a)) or ghosts (prepared as i n Chapter 2, section 2.2(d)) was added one volume of NalO^ of varying concentrations (0-10 mM) i n PBS/azide (defined i n Methods, Chapter 2, section 2.2(a)). These c e l l s were incubated for 10 min unless otherwise indicated and the reaction stopped by d i l u t i n g the solution by a factor of three with PBS/azide and then washed three times at a wash volume r a t i o of 25:1 (PBS to red c e l l s ) spun at 1,000 x g for fiv e min or 40:1 (Dodge buffer (20 idea l milliosmole phosphate buffer, pH 8.0 i n 0.025% azide) to ghost) spun down at 20,000 x g at 4°C for 20 min. 3.2(b) Periodate oxidation of glycophorin or fetuin Glycophorin was isolated from red c e l l s by the method of Marchesi & Andrews (1971). Neuraminidase digest was performed on glycophorin as follows: to 10 mg of glycophorin i n 10 ml Krebs-Ringer solution (0.121 M NaCl, 5 mM KC1, 1 mM KH 2P0 4, 1 mM MgS04, 17 mM Na2HP04) was added 0.2 U of Vibrio cholerae neuraminidase (Calbiochem). This was incubated at 37°C for 24.5 hours. The sample was dialysed against Kreb-Ringers solution u n t i l no free s i a l i c acid (determined by the method of Reid et a l , 1977) could be detected i n the sample. Lowry assays were performed as i n Lowry (1951) as modified by Peterson (1977). Fetuin (type IV) was purchased from Sigma. - 90 -To a known concentration of sialoglycoprotein (glycophorin or fetuin) was added NalO^ i n PBS. After certain time i n t e r v a l s the solution was monitored for formaldehyde (by the method of Nash, 1953, as modified by Reid et a l , 1977). 3H labeling of glycophorin Five milligrams of glycophorin i n 5 ml of 0.1 M sodium acetate pH 5.5 i n 0.025% azide was incubated with a 2 molar excess of NalO^ for 2 hours at 4°C i n the dark. This reaction was stopped by the addition of 0.3 ml of 0.1 M glucose and the sample dialysed against water and 0.1 M phosphate buffer, pH 7.1. Next, 25 mCi of NaBE^H]^ (Amersham) was added and incubated for 1 hour at room temperature followed by 2.8 mg of cold NaBH^ and incubation for another 35 minutes. The product was dialysed against 0.1 M sodium acetate pH 5.0 and then against water/0.025% azide u n t i l no more was released. The product was then freeze dried. 3.2(c) Spin labeling of red c e l l s Experimental conditions were varied for the periodate treated c e l l s to optimize labeling and are tabulated i n the re s u l t s . Occasionally, NaB[ 3H] 3CN (Amersham) at about 3 x 10 8 cpm/ml i n the NaBH^CN solution was used. Other reagents were borane dimethyl amine (Aldrich), 2,2,6,6-tetramethyl-A-amino-piperidine-l-oxyl (Sigma), the spin la b e l analogue (SLAN). Optimal conditions for labeling the red c e l l s are quoted i n the Results (section 3.3(e)) for ease of comparison. - 91 -3.2(d) Spin labeling ghosts by the method of F e l i x & B u t t e r f i e l d (1980) 9 To 50 ml of ghosts (at 5.4 x 10 ghosts/ml) was added 50 ml of 2 mM NalO^ i n 0.1 M sodium acetate pH 5. This was incubated for 10 minutes at 4°C with gentle mixing. The oxidation was stopped by addition of 450 ml of 5 mM arsenite i n 50 mM phosphate buffer pH 9.2. The samples were spun down at 20,000 x g for 20 minutes at 4°C. To the p e l l e t was added 100 ml of 1 mM TEMPAMINE and 100 ml of 1 mM NaBH^CN with an a c t i v i t y of 8 x 10 7 cpm/ml (both i n 5 mM phosphate buffer, pH 8) and incubated overnight at 4°C (a t o t a l of 19.5 hours). The suspension was spun down as before and then washed i n Dodge buffer f i v e times at a volume r a t i o of 1:25 ghosts to buffer. Controls were treated i d e n t i c a l l y except no periodate was added. 3.2(e) Formaldehyde assay for red c e l l s The formaldehyde was assayed v i a a modification of the method of Nash (1953). Due to s l i g h t l y s i s of red c e l l s by the periodate oxidation step, an additional step was included. After the addition of the periodate and i t s cessation by the addition of KI/Na^O^, the protein i n solution was precipitated via the method of Somogyi (1945). In t h i s procedure 50 u l of ZnSO^ (5% by weight) was added to the 300 pi sample, mixed, then 50 u l of BatOH^ (0.3 M) added and the sample mixed and centrifuged on an Eppendorf table top centrifuge. This method was found to successfully precipitate the i n t e r f e r i n g protein (more e f f e c t i v e l y than 10% t r i c h l o r o a c e t i c acid or (NH^JSO^). The supernatant was then assayed according to Nash (1953). Standard samples were treated i d e n t i c a l l y . - 92 -3.2(f) S i a l i c acid assays S i a l i c acid was assayed by the thiob a r b i t u r i c acid method of Aminoff (1961) as modified by Culling et a l (1977), or by the resorcinol assay of Jourdian et a l (1971). Bound s i a l i c acid was assayed as i n Reid et a l (1977). 3.2(g) Rouleaux Red c e l l s attach side-by-side to form what i s c a l l e d rouleaux i n the presence of certain macromolecules (e.g. fibrinogen and globulins found i n the plasma or dextran solutions). An assay of c e l l - c e l l surface interactions are t h e i r a b i l i t y to form rouleaux. A 5% suspension of washed periodate oxidized red c e l l s was exposed to an 80% plasma solution (in PBS) or a 3% solution of dextran 70 (Pharmacia) i n 1% albumin. The dextran concentration was determined i n a polarimeter. Red c e l l s that had been oxidized with 3.5 mM NalO^ and washed, were exposed to plasma or dextran and a varying concentration of N-Acetyl glucosamine (GlcNAc), glucosamine (glcNh^) or glucose. A l l data was obtained v i a microscopic inspection of the mixtures. 3.2(h) ESR ESR samples were run as described i n Chapter 2. Temperature controlled experiments were done on a temperature controlled homodyne spectrometer employing a Varian 12 inch magnet. The samples were put i n 20 u l c a p i l l a r y - 93 -tubes and flame sealed at one end making sure no heat damage occurred to the sample. 3.2(i) Measurements of the distance between nitroxide la b e l Various samples of red c e l l s were labeled according to the optimal protocol with the inclusion of a varying r a t i o of spin l a b e l analogue to TEMPAMINE, from 9:1, 4:1 to no spin la b e l analogue. This ensured that the reacting amine group concentration stayed the same and only the nitroxide concentration decreased by the d i l u t i o n s quoted. These labeled c e l l s were then made into ghosts as i n the protocol. Spectra were run at 77°K on a l l the samples using a Dewar inser t on the Varian E-3 containing l i q u i d nitrogen. Spectra were run at 0.16 mW microwave power, the lowest available to prevent saturation. S i l i c a tubing (3 mm o.d.) sealed at one end was used and solutions were introduced using a te f l o n syringe. 3.2(j) Reversiblity Spin labeled 3H labeled ghosts, made v i a the optimal protocol were l e f t s i t t i n g at 4°C i n PBS/azide. At selected times aliquots were removed and spun down and the supernatant and p e l l e t analysed for spin la b e l and t r i t i u m content. - 94 -3.2(k) Hydrolysis of the spin labeled ghosts Neuraminidase digest was performed on the spin labeled ghosts (SL-ghosts) with enzyme from Clostridium perfringens (Sigma type VI) or Vibrio cholerae (Calbiochem-Behring Corp., La J o l l a , CA) at 1.3 III per 6 x 9 o 10 ghosts at 37 C for 39 hours at pH 5.1. Native red blood c e l l s were incubated only for 1.5 hours under i d e n t i c a l conditions. Acid hydrolysis was performed with 0.1 N H 2S0 4 at 80°C for one Q hour at one ml ^ SO^ to one ml of SL-ghosts (6 x 10 ghosts/ml). The samples were then spun down (20,000 x g for 20 min at 4°C), the supernatant and the p e l l e t separated and neutralized with NaOH. In a l l cases, appropriate controls (SL-ghosts or normal ghosts were treated i d e n t i c a l l y , except no neuraminidase or acid was added) were included. 3.2(1) Isolations SL-ghosts were s e l e c t i v e l y s o l u b i l i z e d with 0.1 M NaOH or 0.5% Triton X-100 i n 56 mM borate buffer, pH 8 as described by Steck & Yu (1973) and Yu et a l (1973). Lipids were isolated with chloroform:methanol as described by Saito & Hakomori (1971) or via the method of Folch et a l (1957). Glycophorin was isolated from ghosts or SL-ghosts by the method of Marchesi & Andrews (1971). B r i e f l y , to every gram of lyo p h i l i z e d ghosts are added 43 ml of 0.3 M LIS (lithium d i i o d o s a l i c y l a t e ) i n 0.05 M t r i s pH 7.4. After incubating for 20 min, 80 ml of ice cold water was added and allowed - 95 -to incubate for one hour at A C. The supernatant was collected by centrifugation at 18,000 x g for one hour at A°C and 50% phenol i n water added at equal volume to the supernatant. The phases were allowed to separate and the upper phase collected and dialysed against water. This was freeze-dried and then the protein washed twice i n i c e cold absolute ethanol and resuspended i n water, dialysed and freeze dried. 3.2(m) Tritium analysis SDS PAGE was performed as described i n Appendix A. The gels were s l i c e d at 1 mm int e r v a l s and counted for by the method of Aloyo (1979). In some cases the gels were frozen at -70°C immediately after being run and s l i c e d , and i n other cases the gels were fixed and stained before being s l i c e d and counted. No differences i n the 3H p r o f i l e s were noticed but a decrease of 50% i n the cpm/slice was obtained for the stained gels. T r i t i a t e d samples (5-10 u l ) were s o l u b i l i z e d i n 10 ml Atomlite (NEN, Boston, MA) or i n 10 ml toluene (scintanalyzed (Fisher)) to which had been added 6 g of PPO (Kodak-Eastman), 10 ml NCS tissue s o l u b i l i z e r (Amersham) and 10 ml hyamine hydroxide (Packard) per l i t e r of toluene. Samples were allowed to incubate at room temperature for two days, with occasional shaking, before being counted. Absolute disintegration rates (disintegration per minute (dpm)) were determined by adding various amounts of erythrocytes mixed with a constant known l e v e l of t r i t i u m . With the use of the experimental quench curve, the number of counts per minute (cpm) was converted into dpm. The moles of NaBH^CN which reacted was calculated -12 from the s p e c i f i c a c t i v i t y (2 to 308 x 10 mmole/dpm). - 96 -3.3 RESULTS 3.3(a) Effects of periodate modification on rouleaux formation Red c e l l s were modified as described i n the Methods section 3.2(a) with concentrations varying from 0 to 10 mM periodate. Up to 3.5 mM periodate, no obvious alterations were v i s i b l e but 5 mM exposure resulted i n noticeable swelling and 10 mM periodate caused l y s i s . These modified c e l l s were exposed to blood plasma. Control c e l l s i n 80% plasma formed rouleaux (Fig. 3.1). As the periodate concentration increased, less rouleaux were formed. At 0.5 mM, the rouleaux formation had decreased to about 50% that of controls and by 3.5 mM periodate, the c e l l s were swollen and clumps had formed; at 5 mM and higher, c e l l l y s i s resulted, and large clumps formed. This pattern was also observed for red c e l l s suspended i n dextran solutions (Fig. 3.2), with 2 mM periodate modification, s i g n i f i c a n t decreases were observed i n rouleaux formation. In place of rouleaux clumps were formed which got bigger the higher the periodate concentration used to modify these c e l l s . At 2 mM periodate modification, the clumps were already being observed. A concentration of 3.5 mM periodate was chosen to modify c e l l s due to noticeable clumps being formed but no l y s i s occurring. To these systems of oxidized red c e l l s and plasma or dextran were added the sugars glcNAc, glcNFL, and glucose at A, 10 and AO mg/ml. For plasma, glcNAc at AO mg/ml completely i n h i b i t e d the formation of clumps and t h i s i n h i b i t i o n was noticeable even at A mg/ml. The controls at AO mg/ml glcNAc were crenated however. Forty mg/ml glcNhL p a r t i a l l y inhibited clumping and had no effect - 97 -F i g . 3.1 Oxidation of red c e l l s vs t h e i r a b i l i t y to form rouleaux. (A) i s a picture of normal rouleaux i n 80% serum. (B), (C), and (D) are periodate treated red c e l l s exposed to 80% serum, (B) 1 mM, (C) 2 mM, (D) 5 mM periodate. Magnification 1600 X. - 98 -F i g . 3.2 The e f f e c t of periodate oxidation of red c e l l s upon t h e i r a b i l i t y to form rouleaux i n 3% dextran. (A) normal rouleaux formation with unmodified red c e l l s and (B), (C), and (D) periodate treated c e l l s . (B) 0.5 mM, (C) 1 mM, (D) 2 mM periodate. Magnification 1600 X. - 99 -on the controls (they s t i l l formed rouleaux). No effect was produced by glucose. For the dextran system, whether the c e l l s were oxidized or not, there was only a s l i g h t decrease i n the agglutination with a l l the sugars. 3.3(b) Formaldehyde formation Table 3.1 l i s t s the amount of formaldehyde formed due to periodate oxidation of red c e l l s as performed i n the protocol with varying times and periodate concentrations. The amount of formaldehyde i s expressed as the % of s i a l i c acid modified assuming 3 x 10 7 s i a l i c a c i d s / c e l l . TABLE 3.1 FORMALDEHYDE FORMATION1 AS A FUNCTION OF PERIODATE CONCENTRATION TO RED CELLS exposure time of periodate (min) Molar excess of periodate to s i a l i c a c i d 2 Four Eight 10 20 7% 11% 12% 15% -'•Expressed as the % of t o t a l s i a l i c acid on a red c e l l 2Assuming the t o t a l s i a l i c acid on a red c e l l i s 3 X 10 7 Repeated experiments at four f o l d molar excess (2 mM) of periodate to s i a l i c acid proved to be variable, with the average being 15 +_ 9% of the s i a l i c acid being modified (n = 5, range 7-22). Table 3.2 l i s t s the amounts of formaldehyde formed upon the oxidation of ghosts and neurmaminidase treated ghosts for various times and oxidation conditions i n PBS. - 100 -TABLE 3.2 FORMALDEHYDE FORMATION1 WITH PERIODATE OXIDATION OF GHOSTS Periodate exposure Molar excess of periodate to t o t a l s i a l i c a c i d 2 Ghosts NANase treated Ghosts 3 7 15 28 15 28 10 A2.5+.5 74+8 15 55+6 66+2 5.1+2 11+2 30 72+6 76+1 11 13 •••Expressed as the % of t o t a l s i a l i c acid on a red c e l l , r esults an average of four experiments 2Assuming the t o t a l s i a l i c acid on a red c e l l i s 3 x 10 7  380 + 8% i f the s i a l i c acid removed Fig. 3.3 shows the SDS PAGE gels of the oxidized ghosts. Table 3.3 l i s t s the amounts of formaldehyde formed upon the oxidation of glycophorin, neuraminidase treated glycophorin and fetuin for various times and oxidation conditions i n PBS. Glycophorin contained 37.6% (w/w) lowry protein and 18.2 +_ 0.7% (w/w) s i a l i c acid. Fetuin contained 6.45% (w/w) s i a l i c acid. - 101 -Fig. 3.3 .Densitometric scans of SDS PAGE gels (5% pH 6.1) scans of (A) coomassie blue (45 ug membrane protein 595 nm) or (B) PAS (180 jjg membrane protein 525nm) of (1) untreated ghosts, (2) seven molar excess of periodate to s i a l i c acid or (3) 15 molar excess periodate oxidation of ghosts. Numbering according to Fairbanks et a l , 1971. TD = tracking dye. - 102 -TABLE 3.3 FORMALDEHYDE FORMATION* WITH PERIODATE OXIDATION OF GLYCOPHORIN AND FETUIN Molar excess of periodate to t o t a l s i a l i c acid Glycophorin Fetuin Periodate NANase2>3 exposure (min) 2 3 A 5 3 8 treated A 8 10 9+1 9+5 9+1 16+6 5+1 21+10 21+10 20 58+2 5A+6 100+3 82+8 9 80+5 98+3 60 93+3 111+5 121+5 122+5 O.A ^expressed as a % of the t o t a l s i a l i c acid (average of A experiments) 2average of two experiments 3 5 _+ 5% of the bound s i a l i c acid remaining 3.3(c) T r i t i a t e d glycophorin The sample of glycophorin used was found to contain 31.4% protein and 18.9 _+ 1.0% s i a l i c acid. In 0.1 M sodium acetate 1 mg/ml glycophorin has an absorbance of 0.A28 at 280 nm. After t r i t i a t i o n , i t was found that over 99% of the unreacted t r i t i u m was released during the f i r s t two hours of d i a l y s i s against acetate buffer, pH 5. The resultant a c t i v i t y was 1.A3 x 10 7 cpm/mg glycophorin. Fig. 3.A shows the gel p r o f i l e s of the t r i t i u m labeled and unmodified glycophorin. These gels were s l i c e d immediately after being run. - 103 -E C LD C\J LO Q m O 1 A \ TD FRONT 300000 0 10 20 30 SLICE NUMBER 40 50 60 Fig. 3.4 SDS PAGE of 50 ug glycophorin (reduced) run on a 10% laemmli gel. (A) PAS sta i n of isolated glycophorin. (B) Periodate and NaB[3H]4 treated glycophorin. The gel was s l i c e d i n 1 mm in t e r v a l s and counted for t r i t i u m . TD = tracking dye. - 104 -3.3(d) SDS PAGE analysis ( i ) PAS staining PAS staining as i n Appendix A, yielded the t y p i c a l PAS p r o f i l e for red c e l l membranes (Figs. 3.5, 3.11, 3.12). I f periodate oxidation of the gels was omitted i n the staining of the gels v i a PAS but everything else kept constant, only the l i p i d s stained. This staining procedure has been called Basic Fuchsin staining. Gels of periodate treated c e l l s with or without spin labeling were analysed by t h i s method and a new band corresponding to PAS 1 was found also to st a i n (at a lower i n t e n s i t y ) . Fig. 3.5 shows the results of staining variously treated red c e l l s . This additional staining technique ( i . e . omitting oxidation on the gels) was applied to various samples throughout t h i s work and i s noted where used. 3.3(e) Spin labeling red c e l l s The optimal conditions used to spin l a b e l red c e l l s were as follows: to one volume of freshly washed packed red c e l l s was added one volume of 2 mM NalO^ (a four molar excess to s i a l i c acid) i n PBS/azide. After incubation at room temperature for 10 minutes, the oxidized c e l l s ([0] RBC) were diluted three fold with ice cold PBS/azide and washed three times i n ice cold PBS/azide. One volume of 23 mM TEMPAMINE (4 mg/ml PBS/azide) and one volume of 48 mM NaBl-LCN (3 mg/ml PBS/azide) with or without 1-4 x 10 8 -12 3 cpm/ml (spe c i f i c a c t i v i t y 2.9 to 308 x 10 mmole/dpm) NaB[ H]3CN were added. After two hours incubation at room temperature, the solution was diluted three fold with ice cold PBS/azide and washed three times. F i g . 3.5 SDS PAGE of 5% g e l s , pH 6.1 of (A) PAS s t a i n i n g and (B) Basic f u s c h i n s t a i n i n g o f 180 pg of membrane p r o t e i n of SL-ghosts made as i n p r o t o c o l , s e c t i o n 3-3(e), with varying amounts of spins/SL-ghost. (1) 7.3 x 10 6, (2) 1.7 x 10 6, (3) 0.18 x 1 0 6 and (A) c e l l s periodate o x i d i z e d but not exposed t o TEMPAMINE. The i n s e r t scans i n (B) are samples at f u l l gain on the d e n s i t o m e t e r . TD = t r a c k i n g dye. - 106 -The resultant spin labeled red c e l l s (SL-RBCs) were then lysed by the method of Dodge et a l (1963) and denoted spin labeled ghosts (SL-ghosts). The following data shows how t h i s protocol was arrived at. Various attempts were made to optimize the y i e l d . Tables 3.A-3.6, show the results of alterations from the above protocol. A l l conditions were as quoted above unless stated i n the table ( i . e . the tables only quote the variations from the above protocol). Table 3.4 shows the spins/ghost vs the i n i t i a l TEMPAMINE concentration. Table 3.5 l i s t s other manipulations done to try to improve labeling yields along with the i n i t i a l TEMPAMINE concentrations used. TABLE 3.4 TEMPAMINE CONCENTRATION AND SPINS/GHOST I n i t i a l Result TEMPAMINE cone (mM) Spins/ghost x 10" 6 % of t o t a l s i a l i c t r i a l s acid modified 1 % of oxidized s i a l i c a c i d 2 3 6 8 17 23 29 35 45 0.4 + 0.3 1.6 + 1.3 2.0 + 1.7 3.5 + 2.0 4.0 + 1.7 3.1 + 1.8 3.9 + 1.3 3.1 2 10 6 10 20 4 3 1 0.1 0.5 6.5 8 36 44 77 89 69 87 69 12 13 10 13 10 ^assuming the t o t a l s i a l i c acid content i n a red c e l l i s 3 X 10 7  2assuming 15% of the s i a l i c acids are periodate oxidized - 107 -TABLE 3.5 OTHER ATTEMPTS TO IMPROVE SPIN LABELING OF RED CELLS Manipulation TEMPAMINE cone (mM) Resultant spins/ ghost X 10" 6 Rejuvenated old blood 23 1.13 0.3 M Sorbital (diluted 8:1 35 6.5 with PBS) instead of PBS 17 3.6 8 3.6 0 2 bubbled i n the PBS buffer 23 2.8 Incubation overnight at 4°C 23 2.5 8 1.2 Recrystallized NaBH3CN 23 1.2 Boranedimethyl complex 23 0.95 instead of NaBH3CN No reducing agent 23 0.34 No oxidation 23 0.35 Table 3.6 compares results of variation i n the i n i t i a l oxidation conditions with that of the f i n a l spins/ghost and also when SL-RBCs are reoxidized (when s t i l l red c e l l s ) and the whole labeling procedure repeated. The s h i f t i n the spectral parameter compared to the standard method i s also shown. Fig. 3.6 shows SDS PAGE gel scans of various products quoted i n Table 3.6. - 108 -TABLE 3.6 INCREASING PERIODATE CONCENTRATION ALONG WITH SPINS/GHOST AND SP % increase i n periodate % increase i n resultant % increase i n i n i t i a l l y usedl spins/ghost 1 resultant SP 1 A3 23 100 79 + 52 (n = 3) 6 + 1 i n comparison to the protocol used i n section 3.3(e) SL-RBC reoxidized and labeled as i n optimum protocol resulted i n a 19% increase i n the resultant spins/ghost - 109 -2 2 FRONT FRONT FRONT FRONT FRONT FRONT A B C Fig. 3.6 5% SDS PAGE (pH 6.1) gels of (1) 45 ug membrane protein coomassie blue stained and scanned, (595 nm), and (2) 180 ug membrane protein PAS stained and scanned, (525 nm), of (A) untreated red c e l l s , (B) SL-ghosts made as described i n optimum protocol section 3.3(e) and (C) SL-ghosts made as i n (B) except at twice the periodate concentration. The numbering system according to Fairbanks et a l (1971). TD = tracking dye, Hb = hemoglobin. - n o -3.3(f) Background signal As i n Chapter 2, i t was found that a large population of spin l a b e l had entered the c e l l which could be removed only by lysing the c e l l s . Fig. 3.7 shows the difference i n the spectra before and after lysing of spin labeled c e l l s and controls (red c e l l s plus TEMPAMINE). Care must be taken i n l y s i n g red c e l l s . I t was found that the samples could s i t overnight at 4°C but could not be frozen or else the membranes appeared to fragment. The control (red c e l l s plus TEMPAMINE then washed and lysed) spectrum 5 varied from nondetectable to an average of 6 + 2 x 10 spins/ghost (n = 25). Fig. 3.7b shows a t y p i c a l spectrum obtained for control ghosts. 3.3(g) Visual inspection of oxidized and spin labeled c e l l s Another reason for lysing the red c e l l s came from the observation that a gradual change overcame the periodate oxidized red blood c e l l s whether or not they were spin labeled. A solution of 23 mM TEMPAMINE/NaBH^CN caused s l i g h t hemolyis as determined by the v i s u a l appearance of hemoglobin i n the solution. As the TEMPAMINE concentration increased, l y s i s was more noticeable. In a l l cases, the hemolysis was minor. Over an extended period of time (6 hours or more), these oxidized (spin labeled or not) red c e l l s became darker red i n color and more viscous compared to the controls. I f l e f t overnight at 4°C, the alterations were even more noticeable. Rouleaux formation by SL-RBCs was also i n h i b i t e d (they formed clumps l i k e the periodate treated c e l l s ) , while the controls formed rouleaux. - I l l -Fig. 3.7 ESR spectra of the control (red c e l l s plus TEMPAMINE then washed) before (A) and after (C) lysing and of SL-RBCs before (B) and after (D) ly s i n g ( c e l l s prepared as described i n the optimal protocol section 3.3(e)). A = 7 X 10.5, c = A X 10 5, B = 1.8 X 10 6 and D = l.A X 10 6 s p i n s / c e l l . - 112 -3.3(h) Spin labeled ghosts Fig. 3.8 shows a t y p i c a l spectrum for spin labeled ghosts along with the parameters measured. The spectral parameter, SP, was defined as on page 67 SP = ( h Q / h _ 1 ) 1 / 2 + ( h Q / h + 1 ) 1 / 2 - 2 and for SL-ghosts, SP = 0.64 +0.097 (n = 50). Table 3.7 shows the peak-to-peak linewidth of the center l i n e (wQ) vs the spins/ghost and shows that over the range quoted the change i n wQ i s negligible (+2%). TABLE 3.7 SPINS/GHOST AND RESULTANT w0 Spins/SL-ghost w0 (guass) 7.3 X 10 6 2.26 1.7 X 10 6 2.16 0.18 X 10 6 2.21 The SP of SL-ghosts decreased 21% as the temperature increased from 20°C to 29°C and increased 21% as the temperature decreased from 20°C to 12°C. 3.3(i) F-B SL-ghosts Ghosts spin labeled by the method of Fel i x & B u t t e r f i e l d (1980) as described i n section 3.2(d). The resultant spins/ghost was 0.14 x 10 6 (0.5% of the t o t a l s i a l i c acid modified). The SP of the F-B SL-ghosts was 0.52. Fig. 3.9 show the SDS PAGE gels of SL-ghosts and F-B SL-ghosts. - 113 -Fig. 3.8 A t y p i c a l spectrum of SL-ghosts made as described i n the text section 3.3(e). w0 i s the linewidth of the mid-field l i n e , h + i , h 0, and h i are the height of the l o w - f i e l d , mid-field and the h i g h - f i e l d lines'respectively, and are used to measure the spectral parameter SP as defined i n section 3.3(h). - 114 -PAS1 Fig. 3.9 Densitometric scans of SDS PAGE gels (12% Laemmli). On each gel was loaded 180 ;jg of reduced membrane protein of (A) SL-ghosts or (B) F-B SL-ghosts. PAS s t a i n of (1) control ghosts (no oxidation) and (2) ghosts made as i n optimal protocol section 3.3(e) for SL-ghosts and section 3.2(d) for F-B SL-ghosts. Gel (3) i s (2), s l i c e d i n 1 mm in t e r v a l s and counted for t r i t i u m . - 115 -3.3(j) Distance measurements Table 3.8 l i s t the parameters measured (as defined i n Chapter 1, p. 51) for samples run at 77°K prepared as stated i n Methods. I t was found that baseline d r i f t was substantial and had to be accounted for before meaningful measurements could be taken (due to the gain being high). The SLANrSL of 8:1 and 4:1 represent the i n f i n i t e d i l u t i o n of the spin l a b e l . Given d,/d (for SL-ghosts) = (d,/d) at i n f i n i t e d i l u t i o n + 0.58r" 3 TABLE 3.8 MEASURED PARAMETERS FOR DISTANCE CALCULATIONS Sample Spins/SL-ghost di/d SL-ghosts 5.2 x 10 6 0.539+0.001 3.0 x 10 6 0.541 1:4 SLAN ghosts 9.0 x 10 5 0.46 + 0.03 3.4 x 10 5 0.49 1:8 SLAN ghosts 2.0 x 10 5 0.49 an average distance of 23 +_ 2 angstroms (n = 5) between nitroxides of the spin labeled ghosts was calculated. 3.3(k) R e v e r s i b i l i t y SL-ghosts made as described by optimal conditions (section 3.3(e)) with 8 3 the inclusion of 1.5 x 10 cpm/ml of NaB[ Hj^CN produced SL-ghosts with 3.0 x 10 6 spins/ghost and 4.7 x 10~ 5 dpm/ghost. Table 3.9 l i s t s - 116 -the time and the quantity of spins and t r i t i u m found i n the supernatant as well as the supernatant spectral parameters (SP). TABLE 3.9 % RELEASE OF SPINS AND TRITIUM WITH TIME OF INCUBATION OF SL-GHOSTS Incubation % of t o t a l found i n the supernatant SP of spins i n (days) Tritium Spins the supernatant 4 5.8 11 0.04 16 10 14 0.03 26 13 10 0.03 The signal on the SL-ghosts was stable; i t could s t i l l be detected months l a t e r . Some samples exhibited a decrease i n t h e i r spectral parameter with time (weeks) while others increased. Re-lysing the spin labeled ghosts again the next day resulted i n only a 2% increase i n the parameter. 3.3(1) Neuraminidase or acid hydrolysis of SL-ghosts Attempts were made to hydrolyse the spin labeled s i a l i c acid o f f the red c e l l membranes of SL-ghosts or F-B SL-ghosts. The majority of e f f o r t was via neuraminidase digest because i t i s milder than acid hydroylsis. Table 3.10 l i s t s the results of the various attempts, quoting the % of the s i a l i c acid, spin l a b e l and t r i t i u m recovered i n the supernatant as well as the r e l e v a n t s p e c t r a l parameters, SP. - 117 -TABLE 3.10 HYDROLYSIS OF SL-GHOSTS OR F-B SL-GHOSTS Time Sample Method 1 % recovered i n the supernatant SP (hr) S i a l i c acid spins t r i t i u m super p e l l e t 25 SL-ghosts CP 82 5 0.14 0.45 none 0 0 0.46 control ghost CP 116 27.5 SL-ghosts VC 76+11 20 27 0.09 0.64 none 0 20 10 0.15 0.76 control ghost VC 100 10 0 none 0 8 F-B SL-ghost VC 47 0 40 0.65 none 0 0 30 0.78 39.5 SL-ghost CP 38 4 0.08 0.45 none 0 5 0.11 0.46 SL-ghost H+ 17 73 30 0.21 0.11 F-B SL-ghost H + 30 96 40 0.08 ineuraminidase from:CP - Clostridium perfringens, VC - Vibrio Cholerae, none - no enzyme present i n incubation buffer, H + - acid hydrolysis 2Ghosts prepared i d e n t i c a l l y to SL-ghosts but not exposed to periodate The absorption spectra of the s i a l i c acids assay products for the various fractions showed no differences from the standards for the thioba r b i t u r i c acid or the resorcinol assay. - 118 -3.3(m) Li p i d extraction T r i t i a t e d samples of SL-ghosts or F-B SL-ghosts were made as i n the protocol. Analysis was carried out on the spins and the t r i t i u m counts for SL-ghosts but due to the low spins/ghost of F-B SL-ghosts only the t r i t i u m could be analysed. In both cases SDS PAGE samples were run and the amount of t r i t i u m i n the sample associated with the l i p i d (the peak that runs with the tracking dye) (Fig. 3.9) was determined (Table 3.13). Table 3.11 l i s t s the results found for each type of analysis. TABLE 3.11 EXTRACTION OF SPINS AND TRITIUM FROM SL-GHOSTS BY CHLOROFORM:METHANOL % extracted by CHCI3 :methanol % t r i t i u m associated Sample Spins Tritium with l i p i d on SDS PAGE SL-ghosts 1 0 + 3 2 0 + 4 10+2.5 Control 1 13 53 63 F-B SL-ghosts 13 23 26 F-B c o n t r o l 1 54 1 c e l l s treated i d e n t i c a l l y except not periodate oxidized Only 30% of the chloroform:methanol extracted t r i t i u m associated with the Folch upper layer extraction, the layer associated with the glyco l i p i d s . - 119 -3.3(n) Selective s o l u b i l i z a t i o n Extractions of SL-ghosts and F-B SL-ghosts were done as i n Methods (section 3.3(e)) and the quantity of spins or t r i t i u m s o l u b i l i z e d by each of these methods l i s t e d i n Table 3.12. Included i n t h i s table are the SP of these fractions as well as for 1% SDS extraction of SL-ghosts (F-B SL-ghosts having too weak a spin l a b e l signal to be detected i n any of these fra c t i o n s ) . TABLE 3.12 EXTRACTION OF SPINS AND TRITIUM FROM SL-GHOSTS % extracted SP Extraction method Spins Tritium Triton X-100 t o t a l . extracted p e l l e t 82 +A 18 60 0.53 0.50 0.56 + 0.12 + 0.07 NaOH t o t a l extracted p e l l e t 11 +6 89 13 0.25 0.18 O.Al + 0.06 + 0.1 1% SDS t o t a l 100 100 0.72 Fig. 3.10 shows the accompanying SDS PAGE gel analysis of these fractions. - 120 -L I P I D FRONT TD L I P I D FRONT Fig. 3.10 PAS sta i n of SDS PAGE of extractions of SL-ghosts. (1) 0.1 M NaOH p e l l e t , (2.) Triton X-100 p e l l e t and (3) Triton supernatant. Samples made as described i n the text (section 3.2(n)) and diluted 1:1 with sample buffer, reduced and 100 u l of each run of 5% gels (pH 6.1). TD = tracking dye. - 121 -3.3(o) Isolation of glycophorin SL-ghosts made as i n the protocol resulted i n 3.0 x 10 6 spins/ghost and 4.7 x 10~ 5 dpm/ghost. Non-labeled ghosts were run i n p a r a l l e l during the i s o l a t i o n procedure. From 25 ml of SL-ghosts 5.8 mg of SL-glycophorin was recovered. Calculating that 25 ml SL-ghosts equaled 4.29 x 1 0 1 7 spins and 7.09 x 10 6 dpm t o t a l , 54% of the spins co-isolated with the glycophorin fraction and 15 _+ 2% of the t r i t i u m did (over 25% of the t r i t i u m dialysed away). At 8.3 mg/ml i n PBS, SL-glycophorin had an SP = 0.73+0.04 and at 0.6 mg/ml, an SP = 0.64. For SL-glycophorin, 106 _+ 4 jjg of bound s i a l i c acid was found per mg of SL-glycophorin and 131 +_ 4 jjg s i a l i c acid/mg^ of normal glycophorin (19% decrease for SL-glycophorin, equivalent to 2.5 s i a l i c acid). The resorcinol assay was used and a l l samples were found to have i d e n t i c a l absorption spectra to standards. SL-glycophorin had a calculated 2.1 spins/glycophorin (4.1 x lO"*"6 spins/mg or 16% of the t o t a l s i a l i c acid) and 2 x 10 5 dpm of 3H/mg glycophorin (at a s p e c i f i c a c t i v i t y of 2.18 x lO-"*"0 mmole/dpm, t h i s i s equivalent to 1.35 moles of NaBH^CN to mole glycophorin). The interpretation of the values i s i n doubt since Burness & Pardo (1981) showed that glycophorin isolated by the method used here resulted i n co-isolation of PAS 3, 4 and minor proteins (14% of the t o t a l PAS s t a i n ) . The isotope effect for NaB[ 3H] 3CN i s unknown (for NaB[ 3H] 4, i t appears to be 0.5 eq per NANA., aldehyde, not the expected 1 eq, according to Liao et a l , 1973). I f one assumes that the only s i t e for 3H i n these preparations i s the reduced NANA-SL bond, comparing the r a t i o of 3H to spins i n - 122 -SL-glycophorin implies an isotope effect of 1.35/2.1 = 0.64 (Borch & Durst, 1969, observed an isotope effect for LiBD^CN). Fig. 3.11 shows some SDS PAGE gels run of isolated SL-glycophorin i n p a r a l l e l with SL-ghosts and normal ghosts for comparison. The SP of SL-glycophorin (at 8.3 mg/ml) decreased 22% as the temperature increased 9.5°C (from 20.5°C to 30°C) and increased 13% when the temperature dropped from 20.5°C to 11°C. 3.3(p) Tritium d i s t r i b u t i o n on SDS PAGE gels Table 3.13 l i s t s the distrubution of the t r i t i u m on SDS PAGE gels of variety of samples, comparing them to isolated SL-glycophorin and 3 - 3 H-glycophorin made by the 10 /NaB[ H], method. - 123 -LIPID A B Fig. 3.11 Densitometric scans of 12% Laemmli SDS PAGE gels of reduced samples of (A) 278 ug isolated SL-glycophorin and (B) 180 ug membrane protein of SL-ghosts. (1) PAS s t a i n , (2) basic fuschin s t a i n , (3) gel (1), s l i c e d into 1 mm thicknesses and counted for t r i t i u m . TD = tracking dye. - 124 -TABLE 3.13 DISTRIBUTION OF TRITIUM IN VARIOUS SDS PAGE SAMPLES Sample above PAS 1 PAS 1 between PAS 1 and Li p i d L i p i d (A) SL-ghosts: 1. gel run and s l i c e d 2. gel run, stained and s l i c e d 3. 1.6x10^ spins/ghost 4. 4 x l 0 5 spins/ghost 5. c o n t r o l 1 23+2% 19+8% 12% 11% 52+1% 52+7% 60% 56% 22+1% 20+2 % 20% 19% 5.5+0.7% 6.6+2.5% 7.8% 8.7% 63% (B) F-B SL-ghost: 1. gel run, stained and s l i c e d 2. c o n t r o l 1 15% 33% 13% 26% 54% (C) SL-glycophorin: 1. isolated from SL-ghosts 2. I0 4/NaB[ 3H] 4 labeled 28+2% 18% 58% 61% 13+2% 16% 1.5+0.7% 4.8% 1 c e l l s treated i d e n t i c a l l y but not periodate oxidized - 125 -3.4 DISCUSSION The effects of spin labeling upon red c e l l s were studied i n two parts: periodate oxidation and reductive amination (addition of the spin label TEMPAMINE and NaBH^CN). Analysis and optimization of these two steps was done with the aim of producing maximum yields with the least perturbation to the red c e l l s . Once t h i s was accomplished, analysis of the product took place to determine where the modification occurred. 3.4(a) Periodate oxidation Exposure of red c e l l s to high periodate concentrations (10 molar excess to s i a l i c acid (equivalent to 5 mM periodate) or higher) results i n immediate swelling and hemolysis. For instance, i t was reported i n 1949 (St. Groth), that decreased osmotic resistance occurred i n red c e l l s exposed to 16 molar excess periodate. Stewart (1949) found that even a four molar excess w i l l r esult i n hemolysis and c e l l i n s t a b i l i t y i f the incubation times are long (30 min compared with the 10 min used i n t h i s work). Short incubation times and low periodate concentrations therefore are important not only for selective s i a l i c acid modification (Chapter 1, p. 36) but also to maintain red c e l l i n t e g r i t y . ( i ) Altered surface properties These periodate modified c e l l s behave d i f f e r e n t l y from normal c e l l s i n other ways as well i n that surface properties have been altered. Periodate - 126 -treatment i n h i b i t s the formation of rouleaux (Figs. 3.1, 3.2). Large, more symmetrical aggregates are produced instead, more t y p i c a l of antibody-induced agglutination than rouleaux. The theory of rouleaux formation i s not completely known, but macromolecular bridging of polymers i n solution i s believed to be the cause (Brooks, 1976; Skalak et a l , 1981; F r i t z J r , 1984). Dextran and plasma appear to induce agglutination by different mechanisms as seen by the effects of the monosaccharides glcNAc and glcNH 2. GlcNAc (at 181 mM) inh i b i t e d and glcNH 2 p a r t i a l l y i n h i b i t e d agglutination of periodate (3.5 mM) oxidized red c e l l s by plasma (also shown by Novagrodsky (1973) for glcNAc and galNAc), but had no effect on dextran agglutinated oxidized c e l l s . Periodate oxidized c e l l s have been found to be agglutinable by a l l adult sera (Stewart, 1949; Moskowitz & Treffers, 1950) due to the presence of a natural antibody (Yachmin & Gardner, 1961). This antibody i s probably p r e f e r e n t i a l l y bound to the monosaccharides while dextrans are l i k e l y non-specifically adsorbed onto a variety of structures on the red c e l l surface. ( i i ) Formaldehyde formation Oxidation of red c e l l s by periodate results i n the formation of formaldehyde (Table 3.1) due to the cleavage of the C^-Cg d i o l of s i a l i c acid, resulting i n an aldehyde on Cg (reduction of t h i s aldehyde to the alcohol results i n a product cal l e d NANAQ). Another molecule of periodate can then oxidize the Cg-aldehyde C7-0H, producing CHOOH and an aldehyde on C 7 (which, when reduced, produces NANA7). Mild oxidation of s i a l i c - 127 -acid with periodate has the potential for producing two products, the C g or the Cj aldehyde. In either case, for each formaldehyde produced, one s i a l i c acid has been oxidized. Removal of s i a l i c acid by the enzyme neuraminidase resulted i n no formaldehyde formation (Tables 3.2, 3.3). Increased periodate concentration or exposure times results i n increased formaldehyde formation (Table 3.1) and l y s i s of red c e l l s . The r e a c t i v i t y of red c e l l s to periodate appeared to be lower than ghosts (also seen by Blumenfeld et a l , 1972 and Taylor & Wu, 1980), isolated glycophorin or fetuin (Tables 3.2 and 3.3). Periodate modification didn't appear to a l t e r isolated glycophorin (Fig. 3.4) or red c e l l s (Fig. 3.5) as analysed by SDS PAGE, but oxidized ghosts showed major changes i n the gel p r o f i l e s compared to untreated ghosts (Fig. 3.2). High molecular weight components were seen i n the coomassie blue st a i n (protein) p r o f i l e s of periodate treated ghosts and the protein bands were broader compared to untreated ghosts. The PAS staining p r o f i l e also showed material at the top of the gel which was unable to penetrate the gel due to extensive denaturation and cross-linking. This s e n s i t i v i t y of ghosts to periodate has also been seen by Gahmberg et a l (1978), who found concentrations as low as 0.1 mM periodate altered SDS PAGE p r o f i l e s of periodate treated ghosts while concentrations of 2 mM had no effect on red c e l l s . Hence, although ghosts are more reactive to periodate than red c e l l s , one can't increase the degree of oxidation by using ghosts due to t h i s s e n s i t i v i t y to periodate, which apparently induces membrane protein crosslinking. - 128 -The value of 2 mM periodate (a four to one r a t i o of periodate to s i a l i c acid; exposure time of 10 min at room temperature) appears to be the upper l i m i t for red c e l l s (altered PAS p r o f i l e of 4 mM treated c e l l s also Fig. 3.6). Under these conditions, 15 _+ 9% of the s i a l i c acids are oxidized as determined by formaldehyde formation (section 3.3(b)). Massamiri et a l (1979) found at a molar r a t i o of 10 periodates to one s i a l i c acid, 28% of the s i a l i c acids were oxidized. Assuming l i n e a r i t y , 14% would have been oxidized at a fi v e to one r a t i o , i n close agreement to that found here. 3.4(b) Spin labeling Spin labeling the oxidized red c e l l s (by reducing the aldehyde via reductive amination, Chapter 1, section 1.3) didn't appear to reverse any effects induced by periodate oxidation. These spin labeled c e l l s also showed inhi b i t e d rouleaux formation (in the presence of plasma) and underwent a slow change with time (six hours or more), as did the oxidized c e l l s . The oxidized or spin labeled oxidized c e l l s became more viscous and darker red compared to normal c e l l s . No changes were noticed i n the SDS PAGE coomassie blue p r o f i l e (Fig. 3.6). These changes could have been due to some metabolic pump or in t e r n a l metabolite being altered as postulated by Brossmer & Bohm (1974) who found that the red c e l l glucose u t i l i z a t i o n decreased and K + e f f l u x increased after exposure to periodate, even at periodate l e v e l s lower than needed to cause red c e l l l y s i s . Spin labeling the red c e l l s also resulted i n a large population of spins going inside the c e l l rather than being attached to the oxidized sugars (as found i n Chapter 2, section 2.3(b)). The presence of TEMPAMINE inside the - 129 -red c e l l s and the slow i r r e v e r s i b l e alterations induced by the oxidation step necessitated the additional step of lysing the c e l l s a fter they had been spin labeled (Fig. 3.7), removing the majority of non-covalently attached spin la b e l and stopping the slow alterations i n the red c e l l s . 3.4(c) Quantitation The optimal concentration of TEMPAMINE required to maximize the spins/ghost was 23 mM (Table 3.4), producing 4 + 1.7 x 10 6 spins/ghost. This value i s si m i l a r to that found for periodate treated red c e l l s by Cherry et a l (1980) of 10 6 eosin molecules/cell, Taylor & Wo (1980) 1.5 x 10 6 t h i o l hydrazides and Schweizer et a l (1982) 7.2 x 10 6 arylalkyldiamines per c e l l . Table 3.5 shows that no manipulation of the reaction media other than a l t e r i n g the periodate concentration (Table 3.6) improved labeling. Much higher yi e l d s were not expected since only 15% of the s i a l i c acids were oxidized at 2 mM periodate. At 23 mM TEMPAMINE, 89 + 38% of these oxidized s i a l i c acids were reductively aminated (Table 3.4). Incomplete reduction of the aldehydes was anticipated due to short incubation times (two hours as opposed to the usual 12 hours or longer, Lane (1975)). Incomplete reduction of the oxidized s i a l i c acids was also detected by the basic fuschin s t a i n , staining only the membrane components already oxidized. Figs. 3.11 and 3.12 show a faint band corresponding to PAS 1 for SL-ghosts indicating that some oxidized s i a l i c acid s t i l l remained. Untreated ghosts had no basic fuschin s t a i n except i n the l i p i d region (believed to be due to auto-oxidation of l i p i d s and l a b i l e - 130 -acetal phospholipids whose products react with the S c h i f f s reagent (Pearse, 1968)). The labeling procedure labels approximately 15% of the s i a l i c acids of red c e l l membranes (13% as detected by spins/SL-ghost, Table 3.A, 16% as detected by the decrease i n detectable s i a l i c acid for isolated SL-glycophorin and between 12 to 2A% as determined by the hydrolysis attempts, Table 3.10). The values from the hydrolysis experiments are based on the assumption that the modified s i a l i c acids ( i . e . those spin labeled) are insensitive to neuraminidase. The hydrolysed s i a l i c acids showed no maximum s h i f t i n the resorcinol assay as was seen by Van Lenten & Ashwell (1971) and Liao et a l (1973) for NANA.,, implying only unmodified s i a l i c acid was released. Decreased s e n s i t i v i t y of neuraminidase to NANA7 has been seen by S u t t a j i t & Winzler (1971), Blumenfeld et a l (1972) and Liao et a l (1973) (the exocyclic hydroxyls are needed for optimal a c i t i v i t y (Ashwell & Morell, 197A)). 3.A(d) Location The spin labeled components appear to be associated mainly with the PAS proteins, predominately glycophorin A. Red c e l l s oxidized but not spin labeled showed a very heavily stained PAS 1 with basic fuschin (Fig. 3.A). As the degree of spin labeling increased (by increasing the i n i t i a l TEMPAMINE concentration up to 23 mM) the amount of basic fuschin s t a i n at PAS 1 decreased, u n t i l only a very f a i n t band was seen for SL-ghosts made as described i n the optimal protocol. - 131 -The PAS stain showed the same trend, decreasing i n intensity the higher the degree of spin labeling occurring (Fig. 3.4). In t h i s case only the non-reductively aminated s i a l i c acids were stained. Dahr (1974, 1976), Tanner (1978) and Furthmayr (1981) believe that the PAS s t a i n depends almost e n t i r e l y on the s i a l i c acid i n the membrane glycoproteins ( l i p i d s stain oxidized or not and are therefore insensitive to PAS staining for s i a l i c acid, Fig. 3.4). Glycophorin A contains 70% and glycophorin B 10% of the t o t a l s i a l i c acid (Singer & Morrison, 1974; Anstee, 1981) and 70% and 15% of the PAS s t a i n respectively (Burness & Pardue, 1981; Furthmayr, 1981). The PAS proteins (glycophorins A, B and C) comprise at least 80% of the s i a l i c acid i n the red c e l l membrane. Band 3 (Furthmayr et a l , 1976; Fukuda et a l , 1979; Koziaraz et a l , 1978), the glucose transporter (band 4.5) (Sogin & Hinkle, 1978) and acetylcholinesterase (Bjerrum & Bog-Hansen, 1976; Ravazzolo et a l , 1983) also contain s i a l i c acid, as do the l i p i d s (about 2% of the t o t a l s i a l i c acid appear on the l i p i d s , Wherratl, 1973, Marcus et a l , 1981, Beeley e t a l , 1977). Using NaB[3H]^CN i n the reductive amination step resulted i n the SDS PAGE gel 3H p r o f i l e s of SL-ghosts corresponding to the PAS p r o f i l e s (Figs. 3.9, 3.11, 3.12). Over 50% of the t r i t i u m was associated with PAS 1 (Table 3.13). Published data obtained v i a the periodate/NaBC^Hj^ method shows sim i l a r results (Blumenfeld et a l , 1972; Steck & Dawson, 1974; Mueller et a l , 1976; Gahmberg & Anderson, 1977; Gattegno et a l , 1983). Selective extractions of SL-ghosts, made i n the presence of NaBC^Hl^CN, with NaOH, Triton X-100 and chloroform:methanol showed that the t r i t i u m co-isolated with the spin population which co-isolated with the PAS proteins (Table 3.12). I t was found that 89% of the spins and 87% of - 132 -the t r i t i u m was not extracted by 0.1 M NaOH and 82% of the spins and 60% of the t r i t i u m was extracted with the Triton. These two fractions are known to contain the majority of PAS proteins (Fig. 3.10) (Yu et a l , 1973; Steck & Yu, 1973; Mueller & Morrison, 1981). Chloroform:methanol, used to extract l i p i d s , extracted only 7% of the spins and 6.5% of the t r i t i u m . This was comparable to the amount of t r i t i u m found i n the l i p i d area on SDS PAGE gels of the SL-ghosts (8.4%) (Fig. 3.9, Table 3.13). Only about 30% of the chloroform:methanol extracted t r i t i u m and spins co-isolated with the g l y c o l i p i d fraction i n a Folch extraction (the upper layer) (section 3.3 (1)), which represents about 2% of the t o t a l . This low degree of l i p i d labeling of periodate treated red c e l l s was also found by Cherry et a l (1980), Abraham & Low (1980), Steck & Dawson (1974), Mueller et a l (1976) and Gahmberg & Anderson (1977). Isolation of glycophorin from SL-ghosts labeled with NaBC^Hj^CN showed that 54% of the spins co-isolated with the glycophorin f r a c t i o n . The t r i t i u m p r o f i l e of SDS PAGE gels of the isolated SL-glycophorin was i d e n t i c a l to that of the PAS p r o f i l e (Fig. 3.11) with 58% of the t r i t i u m associated with PAS 1. Coomassie blue staining was i d e n t i c a l to the PAS stain suggesting no other proteins were isolated (data not shown). The amount of detectable bound s i a l i c acid i n the glycophorin isolated from SL-ghosts was decreased by a factor of 2.5 (19%) compared to glycophorin isolated from i d e n t i c a l c e l l s not treated. This decrease was probably due to the non-detectable s i a l i c acids being modified (spin labeled) and no longer sensitive to the bound s i a l i c acid assay (due to the exocyclic t r i o l being destroyed). There i s good agreement between the decrease i n the bound s i a l i c acid and the number of spins/glycophorin (2.1 - 133 -spins/SL-glycophorin, which equals 84% of the oxidized s i a l i c acids or 16% of the t o t a l s i a l i c acid on SL-glycophorin), v e r i f y i n g that the two spin labels per glycophorin isolated from SL-ghosts were located on the s i a l i c acids. A problem with the spin labeling procedure of red c e l l s was the . appearance of a small population of spins which appeared to be non-covalently associated with the red c e l l membranes. This population seemed to slowly leach out into solution i n suspensions of SL-ghosts and control ghosts (red c e l l s exposed to TEMPAMINE and then washed and lysed). R e v e r s i b i l i t y experiments (section 3.3(D) show a spin population which migrates into solution with time (approximately 10% of the t o t a l spins i n 10 days, Table 3.8). This spin population was also found i n the supernatants of hydrolysis experiments; spins could be recovered i n the supernatants of SL-ghosts not exposed to neuraminidase (Table 3.10). The population was characterized by a spectral parameter, SP, (0.06-0.14) smaller than that of SL-ghosts (SP = 0.64) and was seen also i n the NaOH and chloroform:methanol extracts of SL-ghosts. These fractions (the neuraminidase, NaOH, chloroform:methanol, and PBS extractions) represented only a minor component of the t o t a l spins per SL-ghost, however. 3.4(e) ESR interpretation SL-ghosts have an SP of 0.64 +_ 0.097 and exhibited negligible Heisenberg exchange (Table 3.6). This SP can be converted to a correlation time (T c) using the formula T = 6.5 x 10" 1 0 w„ SP (seconds) x o - 134 -Although accurate f c calculations need rigorous computer simulations (Chapter 1, p. 49), the formula i s v a l i d i f one compares values within a known system, assuming the same type of motion i s always occurring to the probe. For SL-ghosts, T c i s 9.2+1.4 x 10" 1 0s and 7.4 x 1 0 - 1 0 s for F-B SL-ghosts. Literature values for s i a l i c acid spin labeled red c e l l s are 9.6 x 10" 1 0s (Felix & B u t t e r f i e l d , 1980) and 10 x 10" 1 0s (Aplin et a l , 1979). Ross et a l (1983) obtained 12-13 x 10~ 1 0s spin labeling the s i a l i c acids of lymphocyte membranes. Cherry at a l (1980) found two molecular motions for an eosin probe attached to the s i a l i c acids of red c e l l s , a fast (nanosecond range) motion -5 -7 believed to be the probe i t s e l f and a slower (10 to 10 s) motion believed to be cooperative motion of the oligosaccharide chains. Unfortunately, the slower time scale was not examined i n the present ESR experiments. For isolated SL-glycophorin at 8.3 mg/ml, SP = 0.73 ( T C = 10 x 10" 1 0s) and at 0.64 mg/ml, SP = 0.64 ( f = 9.2 x 1 0 " 1 0 s ) . Higher concentrations of SL-glycophorin appear to cause the spin la b e l to slow down. Lee & Grant (1979) also observed a decrease (18%) i n the mobility of s i a l i c acid spin labeled glycophorin as the concentration of glycophorin increased (from 0.6 mg/ml to 69.1 mg/ml) and attributed i t to an increase i n the l o c a l head group density. This increase i n SP with increased glycophorin may also r e f l e c t solution behavior of glycophorin, since isolated glycophorin i s known to aggregate i n aqueous solutions (Grefrath & Reynolds, 1974). From 10 to 20 monomers per aggregate have been reported (Edgmond et a l , 1979; Springer et a l , 1966), possibly i n the form of a loose - 135 -aggregate of dimers (Silverberg et a l , 1976). The change i n SP may r e f l e c t a change i n the size of the glycophorin aggregate with increased glycophorin concentration. Lee & Grant (1979) calculated a T c of 9.6 x 10~ 1 0s for t h e i r SL-glycophorin and Aplin et al^ (1979) 7.9 x lO -"*"^ for s i a l i c acid spin labeled fetui n . Low et a l (1982) found isolated glycophorin from fluorescein amine labeled red c e l l s had an increased rota t i o n a l freedom compared to labeled red c e l l s and concluded that on the membrane, o l i g o -saccharide density may immobilize the s i a l i c acids via hydrogen bonding. A l l the T c calculations for spin labeled s i a l i c acid either on the red c e l l membrane or on isolated proteins give values (7.4 - 10 x 10~"'"us) f a i r l y s i m i l a r regardless of whether on the membrane or not. No mobility increase was found upon i s o l a t i o n of SL-glycophorin or i n the Triton X-100 and SDS extractions (Table 3.12). The difference between the observations of Low et a l (1982) and the spin l a b e l studies may r e f l e c t the type of probes used (spin l a b e l as opposed to fluorescence). 3.4(f) Distance between nitroxides —5 6 At a spin concentration of 5 x 10 M (5 x 10 spins/SL-ghost, 6 x 9 10 ghosts/ml), the average separation of nitroxides would be 174 angstroms i f a l l were randomly distributed i n three dimensions i n solution (Aplin, 1979; Yalpani, 1980). I f one assumes that the spins are evenly distributed 1 over the c e l l surface, a distance of 53 angstroms would separate the nitroxides (the red c e l l has a surface area of 145 urn 3). Clearly both assumptions are wrong because the s e n s i t i v i t y of the d-,/d measurement i s - 136 -from 10 to 24 angstroms (Aplin, 1979; Yalpani, 1980). Changes i n d^/d were noticed above 10 6 spins/SL-ghosts, however, indicating that the spins were closer than predicted i f t h e i r d i s t r i b u t i o n s were random (Table 3.8). Using equation [11] i n Chapter 1, an average nitroxide separation of 23 angstroms i s implied from the experimental d-^ /d values assuming random 3-dimensional d i s t r i b u t i o n . Assuming the majority of the spins are on glycophorin (at two spins/glycophorin, t h i s equals 1 to 4 x 10 6 s p i n s / c e l l or 25 to 100% of the t o t a l spins/SL-ghost), there are a potential 31 NANA molecules to label on glycophorin A and a s i m i l a r number on glycophorin B. Lovrien & Anderson (1980) state glycophorin has an average radius of 25 angstroms and an average distance of 220 angstroms between glycophorin molecules i n the membrane. I f there are two spins/glycophorin, the above averages suggest the distance measured i s intramolecular. In t h i s case, a more appropriate model would be to assume a random d i s t r i b u t i o n of b i r a d i c a l s . This results i n a calculated 16 angstrom average distance between nitroxides (the distance altered by a factor of 0.7, Kokorin et a l , 1972). Due to low signal-to-noise and baseline d r i f t , a more detailed analysis was not attempted. There i s l i t t l e information available regarding how glycophorin i s situated on the c e l l surface. Stibenz & Geyer (1980), based on model calculations, believe that glycophorin i s hydrogen-bonded to the headgroups of the l i p i d b i l a y e r , looping back on i t s e l f . Skutelsky et a l (1977), through electon microscopy studies, and Levine et a l (1983), through electrophoretic mobility studies, found the s i a l i c acid to be 50-70 angstroms away from the membrane. I f f u l l y extended, glycophorin could reach 168 angstroms beyond the l i p i d bilayer (Stibenz & Geyer, 1980). - 137 -Ruppel et a l (1982) found that the conformation of glycophorin i n vesicles was concentration dependent, small amounts of glycophorin spreading out onto the lipid-water interface and high concentrations protruding into the aqueous phase. The lack of a good model for glycophorin oligosaccharides makes the randomly distributed b i r a d i c a l the best approximation. The s i a l i c acids are the terminal residues of the 16 oligosaccharide chains and certainly could be more than 16 angstroms apart. The 16 angstrom distance separating the spins, which are attached to two of the possible 31 s i a l i c acids, suggests that only selective s i a l i c acids are being labeled, perhaps the most exposed and accessible residues. 3.4(g) F-B SL-ghosts In the work discussed throughout t h i s chapter, spin labeled ghosts cal l e d F-B SL-ghosts, made by the method of Fel i x & B u t t e r f i e l d (1980) have also been analysed. They were shown to contain 0.14 x 10 6 spins/F-B SL-ghost and had 25% of the t r i t i u m l a b e l associated with l i p i d (Fig. 3.9, Table 3.13) compared to 4 x 10 6 spins/SL-ghost and 8% associated with the l i p i d for SL-ghosts made vi a the present protocol (section 3.3(e)). The increase i n the amount of t r i t i u m associated with the l i p i d for F-B SL-ghosts could be due to the higher r e a c t i v i t y of ghosts to periodate, compared-with red c e l l s . Subtle rearrangement of the l i p i d s seems to occur when red c e l l s are made into ghosts (Juliano, 1973; Zwaal et a l , 1973; Marchesi et a l , 1976) which may make them more reactive. Increased labeling of the l i p i d i n ghosts by IO^/NaBC^H]^ was also seen by Liao et a l (1972) and Gahmberg & Anderson (1977) (but not Steck & Dawson, 1974). Fe l i x - 138 -& B u t t e r f i e l d (1980) also found that improved yi e l d s (by increasing the i n i t i a l TEMPAMINE concentration) resulted i n Heisenberg exchange, thus l i m i t i n g them to low yields of spins/ghost. They calculated that 40% of the s i a l i c acids had been labeled by t h e i r technique. I f t h i s were true, they would have obtained 1.2 x 1 0 7 s p i n s / c e l l , 2.7 times higher than obtained here for SL-ghosts. The figure of 40% came from determining the decrease of t r i t i u m incorporation into periodate oxidized ghosts before and after spin labeling, however. They are thus r e a l l y stating that 40% of the oxidized s i a l i c acids had been labeled. The reason for the careful analysis carried out above w i l l become clear i n Chapter 4, when the effects of the wheat germ agglutinin l e c t i n binding to these SL-ghosts w i l l be analysed to examine the possible use of t h i s spin labeling technique for monitoring changes i n s i a l i c acid. - 139 -3.5 CONCLUSION Although oxidation and labeling of red c e l l s has become routine, new information i s s t i l l obtainable. In t h i s work, careful analysis has been done on the spin labeling of red c e l l s . Extensive quantitation was carried out at a variety of levels and correlations have been found between the amount of formaldehyde formed, the amount of spins/SL-ghost and the i s o l a t i o n of SL-glycophorin and t r i t i u m from SL-ghosts. It i s quite clear that the protocol devised results i n two spins/glycophorin, showing that i n spite of a complex substrate, s p e c i f i c i t y and quantitation are both possible. Surface alterations were produced by periodate oxidation and l y s i s occurred at high periodate l e v e l s . Conditions were found where minimal perturbation of the resultant SL-ghosts occurred as detected by SDS PAGE and lack of Heisenberg exchange. The labeling procedure was limited by the periodate oxidation step. The only other published procedure for spin labeling red c e l l s i a l i c acids (Felix & B u t t e r f i e l d , 1980) was l i m i t e d by the TEMPAMINE concentration (increased concentration resulted i n Heisenberg exchange). The protocol i n t h i s work d i f f e r s from F e l i x & B u t t e r f i e l d i n that the red c e l l not the ghost i s periodate oxidized, producing SL-ghosts with more spins/ghost and without Heisenberg exchange. Detailed analysis of the SL-ghosts showed that the majority of the label was on the PAS-stainable proteins (as detected by t r i t i u m l a b e l i n g , selective s o l u b i l i z a t i o n and by a new staining procedure c a l l e d Basic Fuchsin) with only about 8% being associated with the l i p i d s . Occasionally, a small fraction of the spin l a b e l appeared to be associated with, but not - 1A0 -covalently attached to, the SL-ghosts. It appeared i n the supernatants of NaOH extractions and i n various buffer incubations. This fraction appeared very mobile i n comparison to the SL-ghosts and slowly leached out into the solution. Since there are two spins/glycophorin molecule, i t appears that certain s i a l i c acids were s e l e c t i v e l y spin labeled. Assuming a random b i r a d i c a l d i s t r i b u t i o n , the average distance between the spins was calculated to be 16 +_ 2 angstroms. - 141 -CHAPTER A 4.1 INTRODUCTION The l e c t i n wheat germ a g g l u t i n i n (WGA), was used as a model to study the s p e c i f i c b i n d i n g of a p r o t e i n to s p i n l a b e l e d ghosts. WGA was f i r s t discovered i n 1963 as an impurity i n wheat germ l i p a s e (Aub et a l , 1963). I t i s now b e l i e v e d to be a dimer o f 36000 d a l t o n s , high i n d i s u l f i d e bonds (no f r e e s u l f h y d r y l groups are dete c t a b l e (Nagata & Burger, 1974; Rice & E t z l e r , 1975)). This makes WGA very s t a b l e i n 0.1 N HC1, 0.05 M NaOH or at 60°C (Nagata & Burger, 1974). From c e n t r i f u g a t i o n s t u d i e s , i t was found that the WGA d i s s o c i a t e s i n t o i t s monomer form at low pH with a pK g = 4 f o r t h i s r e a c t i o n (Monsigny et a l , 1979). The bindi n g s i t e s on wheat germ a g g l u t i n i n (WGA) were f i r s t p o s t u l a t e d t o be s p e c i f i c f o r glcNAc by Burger & Goldberg (1967) when that monosaccharide was found to i n h i b i t WGA bindin g to c e l l s . S i a l i c a c i d appeared important a l s o s i n c e neuraminidase t r e a t e d c e l l s showed decreased b i n d i n g . I n h i b i t i o n o f WGA bi n d i n g (by mono-, d i - , t r i - and ol i g o s a c c h a r i d e s ) t o red c e l l s , g l y c o p r o t e i n s e t c . became a standard method f o r determining the s p e c i f i c i t y of WGA. A l l e n et a l (1973) were the f i r s t to propose t h a t WGA contained s u b s i t e s , due to the increased a f f i n i t y f o r ( g l c N A c ) n up to n = 3, p o s t u l a t i n g t h a t WGA could accept three glcNAc molecules i n the beta 1-4 c o n f i g u r a t i o n . This s u b s i t e theory has been the framework on which a l l other models have been based. I t i s accepted that s u b s i t e s e x i s t , but how they i n t e r a c t i s s t i l l i n question (see G r i v e t et a l , 1983; Kronis & Carver, 1984c). - 1A2 -S i a l i c acid also appears to interact with the WGA binding s i t e , as shown by fetuin and OSM binding to WGA a f f i n i t y columns (Peters et a l , 1979; Bhavanadan & K a t l i c , 1979; Monsigny et a l , 1980). A l l reported X-ray crystallography work on WGA has been done by Wright (197A, 1977, 1979, 1980, 1981 and'1984). The c r y s t a l consists of two monomers (or protomers). The binding s i t e s are located i n the contact region between the protomers and binding requires p a r t i c i p a t i o n of both protomers. Fig. A.l Schematic i l l u s t r a t i o n of the disposition of the primary and secondary binding locations on the WGA dimer. The domains of protomer 1 are labelled A i , B i , C i , Di, and those of protomer 11 A i i , B i i , C i i , D ; Q . Each unique binding location i s subdivided into subsites (small c i r c l e s ) . GlcNAc oligomers bind at subsites 1, 2, and 3 i n both binding locations, whereas NANA oligosaccharides u t i l i z e subsites i n the primary binding location only. Subsite 1 i s shaded to indicate strongest binding interactions. (Taken from Wright, 1980) - 1A3 -There are two primary s i t e s which are accessible to glcNAc, NANA and s i a l y l l a c t o s e , and two secondary s i t e s which are poorly accessible to (glcNAc)2 and not at a l l to NANA (although there i s no obvious reason for the i n a c c e s s i b l i t y , Wright, 1980). The primary s i t e can be divided into subsites 1, 2, and 3 and, although the secondary s i t e i s less well defined, can also be divided into three subsites (Fig. A . l ) . An unsubstituted acetoamido group at C-2, and equatorial OH at C-3 and C-A are required for binding. Substitutions at C - l , C-A and C-6 are allowed, and WGA can also accommodate inte r n a l sequences of glcNAc i n oligosaccharides. Analysis of binding data (for a ligand binding to a macromolecule or c e l l ) usually involves binding isotherms (a plot of the amount bound (B) vs the equlibrium concentration of the ligand (F)) or Scatchard plots (Scatchard, 19A9) which i s a plot of B/F vs F derived from the equation B/F = K(B Q - B), where B Q i s the number of potential binding s i t e s and K i s the association constant. Analysis of t h i s type assumes that the system i s i n equilibrium. Scatchard plots are useful because the number of binding s i t e s and association constants can ea s i l y be extrapolated i f these plots are l i n e a r . Deviations from l i n e a r i t y (a single straight l i n e implying only one type of binding s i t e ) can be sensitive indicators of the type of binding. Positive cooperativity results i n a concave l i n e with a well pronounced maximum. Negative cooperativity results i n a concave l i n e as does heterogeneity of the binding s i t e s . Systems which don't show a single a f f i n i t y can be treated by curve f i t t i n g the data to predictions of mathematical models (see Dahlquist, 1978). - 1AA -WGA binding has been analysed v i a Scatchard plots. Solution studies employing equlibrium d i a l y s i s (Nagata & Burger, 197A), c i r c u l a r dichroism (Thomas et a l , 1977), proton nmr (Jordan et a l , 1977; Kronis & Carver, 1982), deuterium nmr (Neurohr et a l , 1981), fluorine nmr (Midouz et a l , 1980), i n t r i n s i c fluorescence (Kronis & Carver, 198Ac), and fluorescently labeled saccharides (Clegg et a l , 1983) i n which saccharides were added to WGA i n solution and some parameter measured, a l l showed li n e a r Scatchard plots at a l l WGA concentrations. (glcNAc) n, NANA and sialylactose a l l bound to WGA. Unlike the crystallographic studies of WGA, these studies a l l showed four equivalent sites/WGA dimer (except Jordan et al_ (1981) who found two sites/dimer). In solution, therefore, no secondary s i t e s are observed. The binding of WGA to c e l l s appears to be more complex and of higher a f f i n i t y than to simple sugars. WGA binding to Chinese hamster ovary (CHO) c e l l s (Stanley & Carver, 1977), fat c e l l s (Cuatrecasas, 1973), kidney 21 CB c e l l s (Monsigny et a l , 1980), mouse thymus c e l l s (Monsigny et a l , 1979) and red c e l l s (Adair & Kornfeld, 197A) a l l gave nonlinear Scatchard plots of a type that implied multiple and heterogeneous s i t e s . This chapter deals with the binding of WGA to spin labeled ghosts and to various fractions isolated from these ghosts. Emphasis i s placed on the effect of WGA on the spin probe. Other l e c t i n s (of varying s p e c i f i c i t y ) were also used to determine the s e n s i t i v i t y of the spin probe. - 145 -4.2 MATERIALS AND METHODS  4.2(a) Wheat germ agglutinin Wheat germ agglutinin was obtained from Sigma Chemical Co, St. Louis Mo, EY laboratories Inc, San Mateo, CA, Bethesda Research Laboratories (BRL) Inc, Gaithersburg, MD, and from Vector Laboratories Inc, Burlingame, CA. Pu r i f i c a t i o n of the commercial WGA samples was done by cation exchange chromatography (Appendix D). A l l commercial preparations contained some impurity (a yellow pigment) varing from 7% for Vector WGA to 14% for EY WGA with Sigma WGA varying (depending on the l o t ) i n i t s impurity. Detailed analysis i s shown i n Appendix D. The WGA SL-ghost experiments state which WGA was used. 4.2(b) Protein assays Protein assays were o r i g i n a l l y done by the method of Lowry et a l (1951) as modified by Peterson (1977) with human serum albumin (Sigma) as the standard. O.D.2go was found to be more r e l i a b l e and consistent and was used even though i t was less sensitive. O.D.2gn was determined on a Beckman Model 25 spectrophotometer (Beckman Instruments Inc., Irvine, CA). 4.2(c) Agglutination assay (microtitres) The assay i s a modification of that of Sever (1962). A 50 jul suspension of red blood c e l l s (4% v/v) i n phosphate buffered saline (PBS) was mixed - 146 -with 50 u l of a WGA solution of varying concentrations (in PBS). The suspensions were l e f t undisturbed i n the microtitre plate for 60 minutes. The non-agglutinated c e l l s s e t t l e and flow down the sides of the conical w e l l , forming a small knob at the bottom. The agglutinated c e l l s do not s e t t l e but coat the sides of the cone, appearing more diffuse by eye. The microtitre quoted i s the lowest f i n a l concentration of WGA which caused agglutination. Unless stated the f i n a l concentration of the red c e l l suspension was 2%. O r i g i n a l l y fresh red c e l l s (washed) were used for these microtitres but glutaraldehyde fixed red c e l l s (stable for several months at 4°C) were l a t e r used for ease and because consistent red c e l l samples could be used for a l l studies (see Turner & Liener, 1975). The c e l l s were fixed i n 0.1% glutaraldehyde (Ladd Research Industries) and then extensively washed i n PBS and stored at 4°C. Because fixed c e l l s , being r i g i d , don't deform and pack down as much as fresh c e l l s (Vassar et a l , 1972), a f i n a l concentration of 2.5% (v/v) was used i n the microtitre assay. 4.2(d) Iodination Iodination of proteins involves the i n i t i a l oxidation (with lactoperoxidase or iodo-beads, N-chloro-benzenesulfonamide derivatized polystyrene beads) of the radioiodide which then reacts with the tyrosines of proteins ( e l e c t r o p h i l i c substitution of the ortho hydrogens of the phenolic r i n g , Regoeczi, 1984). Iodination was o r i g i n a l l y done with lactoperoxidase (Sigma) i n a solution of 22 mg WGA/ml PBS containing 50 mM glcNAc (to prevent iodination of the binding s i t e s ) . A four ml aliquot of - 1A7 -th i s WGA solution was combined with 0.1 mg lactoperoxidase and one mCi of 125 c a r r i e r free Na I (Amersham, Arlington Heights, 111) at room temperature. The reaction was i n i t i a t e d by the addition of 20 p i of 0.06% H 20 2 and 15 minutes l a t e r an additional 10 p i of H 20 2 was added. The reaction was stopped with the addition of 0.1% (w/v) of KI/NaN^ after 30 minutes. The solution was then dialysed (using 8000 MW cut o f f d i a l y s i s bags) against four 1 of PBS at A°C repeatedly u n t i l "*"2 5I i n the solution was negligible. This procedure was modified i n some experiments by substituting 0.5A ml Biorad Enzymobeads (lactoperoxidase immobilized on Sepharose) for the lactoperoxidase. With the advent of Pierce Chemical Co (Rockford, 111) Iodo-beads, the procedure was modified again. To two ml of a 20 mg/ml of WGA i n PBS/azide or 0.05 M sodium acetate, 0.05 M NaCl, pH A.3/azide, both containing 50 mM 125 glcNAc, were added fiv e iodo-beads and 0.1 mCi Na I. After occasional mixing, the solution (minus the iodo-beads) was loaded onto a CM-Sepharose CL-6B column and eluted with 0.05 M NaCl acetate buffer, pH A.3 i n 0.025% 125 sodium azide. This column was washed u n t i l a l l free I was eluted. The 125 bound I WGA was then eluted with the same buffer at 0.A5 M NaCl. In a l l cases the iodination step was monitored v i a t r i c h l o r o a c e t i c acid (TCA) p r e c i p i t a t i o n . A very small volume (1 p i ) of the iodination mixture was mixed with 1 ml of 1 mg/ml bovine serum albumin, 1 ml of 20% TCA added, mixed, spun down and 1 ml of the supernatant taken o f f and both fractions 125 counted. The free I distributed evenly amongst the two fractions but 125 the bound I pelleted with the albumin and the % incorporation was calculated. - 148 -The free I was separated from the bound via d i a l y s i s or column chromatography either on a sephadex G-10 or a CM-cellulose CL-6B column. 125 Analysis of the I WGA and various fractions was also performed via instant thin layer chromatography (TLC). A drop of solution was added near the base of a 10 x 2 cm s t r i p of p o l y s i l i c i c acid gel (Gelman Instruments Co, Ann Arbor, Mi.) This was put into a PBS chamber and the PBS solvent allowed to run almost to the top. The s t r i p was a i r dried, cut into 1 cm 125 125 s t r i p s and counted for I. Free I ran with the solvent and the 125 I protein stayed at the base. 4.2(e) SDS PAGE SDS PAGE was performed as described i n Appendix A. 4.2(f) Spin labeled ghosts and extractions Spin labeled ghosts were made as i n the optimum protocol of Chapter 3 (Section 3.3(e)). The extracted fractions of Chapter 3 were tested for t h e i r a b i l i t y to react with various preparations of WGA (from different sources). The l i p i d extract was concentrated, dried, resuspended i n chloroform, degassed, dried again and resuspended i n PBS. This was kindly done by Foon Yip. 3 - 149 -4.2(g) Binding assay To varying concentrations of iodinated WGA (_+ 50 mM glcNAc, Calbiochem-Behring) were added SL-ghosts or normal ghosts. After one hour of incubation at room temperature (with mild shaking) these samples were spun down at 20000 x g at 4°C for 20 minutes. The supernatant was removed 125 and i t s I a c t i v i t y determined on a LKB Compugamma. The a c t i v i t i e s of the p e l l e t s were determined i n a si m i l a r manner. ESR spectra were run on the pe l l e t s from the binding assay as described i n Chapter 2. Binding isotherms were obtained by pl o t t i n g the amount bound to ghosts or SL-ghosts vs the concentration of iodinated WGA i n solution at equlibrium. The t o t a l amount of WGA bound/SL-ghost i s referred to as t o t a l binding. Non-specific binding i s defined as the amount of WGA bound/SL-ghost i n the presence of 50 mM glcNAc. Specific binding was defined as the t o t a l binding minus non-specific binding. To analyse the binding data cor r e c t l y , several things had to be accounted for: 125 ( i ) Free I. This was monitored v i a the quick TLC plate method as described i n the iodination section (4.2(d)). Duplicate samples were run before and after the binding experiment and the average res u l t taken as the 125 amount of free I. This varied from 3-7% of the t o t a l a c t i v i t y . ( i i ) Impurity i n the WGA. I n i t i a l experiments were done on nonpurified commercial preparations of WGA (58% pure, see Appendix D Fig. D.l)). This impurity was compensated for by assuming only the pure WGA bound to the - 150 -pe l l e t s (SDS PAGE analysis of the pe l l e t s showed that the binding WGA was between 80 and 100% pure). Later experiments were run with WGA puri f i e d on the CM CL-6B cation exchange column. The iodinated WGA preparations a l l had microtitres of 5-9 pg/ml showing that they were s t i l l active, ( i i i ) Trapped volume i n the p e l l e t . Ghosts are leaky and less dense than red c e l l s . I t was found (by taking samples of the p e l l e t s and spinning them down hard i n c a p i l l a r i e s ) that 55 +_ 8% (40-71%) of the p e l l e t volume was trapped supernatant. This was confirmed by washout experiments i n which the WGA-ghost p e l l e t was repeatedly washed i n PBS, eluting a l l the non- or weakly bound WGA. The amount of WGA binding to the sample tubes was found to be negligible (a maximum of 0.9% to the gamma counting tubes). A l l data shown are the average of duplicate points. 4.2(h) Other l e c t i n s Ricin communis 1 (a galactose binding l e c t i n ) , peanut agglutinin (a galactose binding l e c t i n ; (Chapter 2) and soybean agglutinin (a galNAc/galactose binding l e c t i n ) were a l l purchased from EY laboratories. Concanavalin A (Con A) (a mannose binding l e c t i n ) and slug l e c t i n (Limax  flavus) (a s i a l i c acid binding l e c t i n ) were purchased from Calbiochem-Behring Corp, La J o l l a , CA. Lectin solutions (0.5 ml at 10 mg/ml i n PBS except Con A at 50 mg/ml and slug at 2 mg/ml) were added to 0.25 ml of SL-ghosts and incubated for 30 minutes. The mixture was then spun down and the p e l l e t ESR run. - 151 -A. 3 RESULTS A.3(a) ESR analysis of WGA addition to extractions of SL-ghosts Table A . l l i s t s the spectral parameters for the mixtures of WGA and the SL-ghost-derived fractions described i n Chapter 3. TABLE A.l SP OF THE VARIOUS SELECTIVE EXTRACTIONS OF SL-GHOSTS PLUS WGA volume volume WGA added Fraction (pi) (ul) (mg) SP NaOH pe l l e t 200 supernatant 200 Triton X-100 pe l l e t 200 supernatant 200 Isolated SL-glycophorin (1 mg/ml) 500 Sigma 100 (2.5 mg) 300 (7.5 mg) 75 (1.9 mg) 225 (5.7 mg) 100 (2.5 mg) 7.5 mg/ml f i n a l A.5 mg/ml f i n a l Vector 500 (0.5 mg) 0.A3 0.63 0.22 0.19 0.55 0.11 (resultant supernatant) 1.36 (pellet) 0.13 (resultant supernatant) 2.2 (pellet) 0.02 (resultant supernatant) 1.23 (pellet) For Triton X-100 extractions of SL-ghosts, i t was found that a precipitate formed upon addition of Sigma WGA i f the f i n a l WGA concentration was A mg/ml or greater. The ESR spectra are shown i n Fig. A.2. Addition of 0.5 ml iodinated p u r i f i e d Vector WGA i n PBS (1 mg/ml at A.9A x 10 6 cpm/mg) to an equal volume of a SL-glycophorin solution (1 mg/ml) - 152 -Fig. 4.2 ESR spectra of the Triton X-100 extract of SL-ghosts before and after the addition of Sigma WGA (4.5 mg/ml f i n a l concentration). - 153 -resulted i n the formation of a p e l l e t containing 96% of the spins with a r a t i o of 1.5 moles glycophorin to 1 mole WGA. The supernatant contained 3.6% of the spins and an equilibrium concentration of 142 ug/ml of WGA. Fig. 4.3 shows the ESR spectrum and the resultant spectral parameters. The SP decreased by 20% with an increase of 9°C (from 20°C to 29°C) and increased by 26% with a decrease i n temperature from 20°C to 10°C. The l i p i d extract (isolated from SL-^H-ghosts with chloroform: methanol) was concentrated and dried down. This l i p i d was extracted from the equivalent of about 1.25 ml of SL-ghosts. Only 6.6% of the t r i t i u m i n the l i p i d s resuspended into the PBS solution. The resultant ESR signal was very weak. Addition of an equal volume of WGA (4 mg/ml PBS) resulted i n no change i n t h i s signal (SP = 0.02) (detailed analysis was very d i f f i c u l t , due to the extremely weak s i g n a l ) . Addition of WGA to the supernatants of the neuraminidase treated SL-ghosts of Chapter 3 resulted i n no change i n SP, remaining that of free spin l a b e l (SP = 0.03). 4.3(d) WGA plus various SL-ghosts Fig. 4.4a i s a plot of the resultant SP of p e l l e t s from various commercial WGAs added 1:1 (v/v) to SL-ghosts (of various preparations). The t o t a l WGA concentrations are l i s t e d because no binding assays were performed on these samples. Fig. 4.4a i s an accumulation of a variety of experiments and shows the v a r i a b i l i t y that arises. For comparison, Fig. 4.4b i s a plot of SP vs t o t a l WGA (the i n i t i a l amount added 1:1 with SL-ghosts) from a binding assay done with impure Sigma WGA. The binding data are shown i n SL-GLYCOPHORIN SP = 0.64 SUPERNATANT SP = 0 .12 PELLET SP = 1.23 F i g . 4.3 ESR s p e c t r a of SL-glycophorin (1 mg/ml) before and a f t e r the a d d i t i o n of an eaual volume of p u r i f i e d Vector WGA (1 mg/ml). cuuitiun or an equal - 155 -0 1 2 3 4 5 INITIAL WGA (mg/ml) 1:1 (v/v) SL-GHOST Fig. A.A Graph of SP after addition of WGA (1:1) (v/v) with SL-ghosts. Total WGA concentration added on the abscissa and the resultant SP of the SL-ghosts/WGA p e l l e t on the ordinate. (A) the res u l t of a variety of SL-ghost and WGA samples. (B) i n i t i a l conditions used i n the binding of impure Sigma WGA of Fig. A.12 to A.15a. WGA source, , * Sigma, + Vector, ^ EY - 156 -Figs. 4.7 to 4.9 and the SP vs WGA bound/SL-ghost shown i n Fig. 4.10a. Fig. 4.5 shows the ESR spectra of WGA binding to SL-ghosts and the resultant supernatant signal after the p e l l e t was incubated at 4°C for two days. Spin labeled ghosts were also made according to the protocol of Felix & Bu t t e r f i e l d (1980), Chapter 3. Fig. 4.6 shows the effect of adding an equal volume of vector WGA (10 mg/ml PBS) to these ghosts along with t h e i r spectral parameters. 4.3(e) Binding assay WGA binding to SL-ghosts was the same as that to unlabeled ghosts. Fig. 4.7 shows the resultant binding isotherms for a variety of binding experiments. Fig. 4.8 shows the resultant s p e c i f i c binding of these experiments. Fig. 4.9 i s the Scatchard plot for the s p e c i f i c binding of Fig. 4.8 (B i s defined as the bound WGA x 10 and F i s defined as the free WGA, which i s the equilibrium WGA concentration l e f t i n the binding supernatant after the WGA has bound to the SL-ghosts). Fig. 4.10 shows the graphs of the spectral parameter, SP, vs the amount of WGA bound for the t o t a l , non-specific and the s p e c i f i c binding. Fig. 4.11 compares the results of two binding experiments which vary i n the way the SL-ghosts were prepared. One was prepared by the optimal protocol and the other i d e n t i c a l l y except with twice the periodate concentration. The figure i l l u s t r a t e s the ESR spectra before and after the addition of WGA to these samples. F i g . 4.5 ESR spectra of SL-ghosts (200 p i ) p l u s 6.6 mg EY WGA (100 p i ) incubated at 4°C f o r two days followed by s e p a r a t i o n of the supernatant and p e l l e t (the supernatant contained 10% of the t o t a l s p i n s ) . Also quoted are t h e i r c a l c u l a t e d SP's. - 158 -Fig. 4.6 ESR spectra of F-B SL-ghosts before and after addition of an equal volume of p u r i f i e d Vector WGA (10 mg/ml). Also quoted are the calculated SP. - 159 -8 Fig. A. 7 The graph of WGA bound per SL-ghost vs the amount of WGA l e f t i n solution (binding isotherm). The s o l i d l i n e s are the t o t a l binding of 1 2 5 I WGA to SL-ghosts and the dotted l i n e s the nonspecific binding (WGA binding to SL-ghosts i n the presence of 50 mM GlcNAc). o Sigma WGA (58% pure) * pure Sigma WGA + pure Vector WGA (using SL-ghosts i n Fig A.13b, i . e . relysed again overnight). The bottom graph i s an expanded region of the upper graph. 60 50 40 c/> 30 o (J I co 20 10 0 0 8 EQUILIBRIUM CONC. WGA (mg/ml) F i g . 4.8 The bindin g isotherm o f 1 2 5 I WGA t o SL-ghosts as described i n F i g . 4.6. S p e c i f i c b i n d i n g i s defined as the t o t a l b i n d i n g minus the n o n - s p e c i f i c binding (binding i n the presence o f 50 mM glcNAc). Symbols as defined i n F i g . 4.7. - 161 -(\J I x E CO D 100 CO o CD I I CO < CD m 2 4 6 WGA/SL-GHOST X 10~ 5 8 10 Fig. A.9 Scatchard plot (B/F vs F) of s p e c i f i c binding of Fig. A.8. B i s defined as the amount of s p e c i f i c WGA bound per SL-ghost (x 10 5) and F i s defined as the amount of WGA l e f t i n the solution (equilibrium binding i n ug/ml) after WGA binding to SL-ghosts. Extrapolation of the graph to the X-axis (the amount of s p e c i f i c WGA bound) yields the number of binding s i t e s for the s p e c i f i c WGA binding per SL-ghost. The lower graph i s an expansion of the upper graph. Symbols as i n Fig. A.7. - 162 Q_ CO 1. 7 CL CO 50 75 100 125 WGA/SL-GHOST (X 10" 5) Fig. 4.10 Graph of SP vs the number of molecules of 1 2 5 I WGA bound/SL-ghost from the binding isotherms of Figs. 4.7 and 4.8. (A) (spectra run on the Varian E-3 at ambient temperature) for 58% pure Sigma WGA. (B) (spectra run at a constant temperature of 20°C) pure Vector WGA binding to SL-ghosts of Fig. 4.17b. ^ t o t a l WGA/SL-ghost, * non-specific (binding of WGA in the presence of 50 mM GlcNAc) and ° spe c i f i c binding (defined as the t o t a l minus the non-specific binding). - 163 -Fig. 4.11 ESR spectra of SL-ghosts (A) prepared as i n the optimum labeling protocol i n Chapter 3 section 3-3(h) and (C) prepared the same way except at twice the normal periodate concentration (2[0] SL-ghosts). (B) i s SL-ghosts plus Sigma WGA binding at 6.9 X 10 5 t o t a l and 4.2 x 10 3 s p e c i f i c WGA/SL-ghost. (D) i s 2[0] SL-ghosts plus Sigma WGA at 6.3 x 10 6 t o t a l and 4.2 x 106 s p e c i f i c WGA/ghost. The calculated SP for each spectrum i s also quoted. - 164 -Fig. 4.12 shows the same SL-ghosts prepared as i n the protocol, WGA added, relysed to eliminate what appeared to be an unbound signal (incubated overnight i n ghost buffer and washed at 40:1 volume) and WGA added again. This was done as a test to determine the r e l i a b i l i t y of the SL-ghost preparation. These SL-ghosts were then used i n a binding assay (with Vector WGA) with the results shown i n Figs. 4.7, 4.8 and 4.10. 4.3(f) Other l e c t i n s Table 4.2 l i s t s the spectral parameters of the l e c t i n p e l l e t s formed as described i n Methods. TABLE 4.2 % INCREASE IN SP AFTER LECTIN ADDED TO SL-GHOSTS1 Lectin added % increase i n SP RCA PNA ConA 29 0 6 5 30 Soybean Slug 10.5 ml l e c t i n at 10 mg/ml i n PBS (except Con A at 50 mg/ml and Slug at 2 mg/ml) added to 0.25 ml SL-ghosts Fig. A.12 ESR spectra of SL-ghosts (A) before and after addition, (C) of equal volumes of p u r i f i e d BRL WGA (37.5 mg/ml). Spectrum (B) i s the SL-ghosts of (A) relysed i n Dodge buffer and (D) i s the r e s u l t of adding equal volumes of p u r i f i e d Sigma WGA (21.5 mg/ml) to (B). Both (A) and (B) had 2.9 x 10 6 spins/SL-ghost. Also quoted are the SP for each spectrum. SL-ghosts (B) were used i n the binding isotherm of Fig. A.7 with pure Vector WGA, but the spectra on which the data i n Fig. A.10 were based were run 10 hours l a t e r than those shown here). - 166 -4.4 DISCUSSION 4.4(a) WGA broadening of the ESR signal Addition of WGA results i n an increase i n the spectral parameter of SL-ghosts (Figs. 4.4, 4.10 and 4.12) and extracted preparations (Table 4.1, Figs. 4.2 and 4.3). The only cases i n which WGA didn't increase the SP were those involving the NaOH and the neuraminidase supernatants. When WGA was added to fractions which resulted i n precipitates (SL-ghosts, SL-glycophorin and the Triton X-100 extraction) the remaining signal (about 5%) had the characteristices of a free spin l a b e l , with a low SP (Figs. 4.2 4.3, and 4.5). This implies that there i s a fa i n t unbound signal present i n SL-ghosts. I t either slowly dissociates with time, or incomplete reduction could allow slow reversal of labe l i n g . These spins didn't interact with the WGA. Addition of a l l the various WGA preparations to SL-ghosts (Fig. 4.4) caused the spectral parameters to increase. The biggest changes i n SP didn't come at the highest WGA concentrations because the SL-ghost preparations d i f f e r i n the amount of the unbound (background) ESR signal which i s insensitive to WGA. This background signal caused the variations in SP changes with WGA concentration. This insensitive f r a c t i o n , although low i n concentration, interferes with spectral interpetation of WGA binding to SL-ghosts. Addition of WGA to SL-ghosts treated with double the amount of periodate yielded a signal which was a superposition of two separate spin populations, one broadened by WGA and the other WGA insensitive (Fig. 4.11). This insensitive population was - 167 -believed to be a second spin l a b e l s i t e which was produced by the higher amount of periodate oxidation but i t could also represent non-interacting, noncovalently attached spins since no differences were found i n the SDS PAGE t r i t i u m p r o f i l e s of these ghosts compared to SL-ghosts. This type of interference was also noticed i n other binding experiments where an addition of over 10 mg/ml Vector WGA resulted i n only minor broadening of the ESR signal (Fig. 4.10b). P l o t t i n g the spectral parameter vs the amount of WGA bound/SL-ghost i n t h i s case showed l i t t l e c orrelation between the two (Fig. 4.10b). Letting the samples s i t for 4 days at 4°C resulted i n a 20% decrease i n the spectral parameters, suggesting either proteolysis or a leaching out of the non-bound spins. No experimental method was found which would r e l i a b l y eliminate t h i s s i g n a l . The SL-ghosts i n which the signal appeared had been made as described i n the protocol i n Chapter 3 (Section 3-3(e)) and tested with the addition of WGA to see i f there was homogeneous broadening (Fig. 4.12a and c ) . A fai n t free signal was observed so the ghosts were incubated at a 40:1 r a t i o with l y s i s buffer overnight at 4°C, lysed again and analysed by the addition of WGA (Fig. 4.12b and d). The spectrum appeared to be satisfactory so the binding experiment was performed and spectra recorded with the results i l l u s t r a t e d i n Fig. 4.10b. No obvious decreases or alterations i n these spectra were v i s i b l e i n the SL-ghosts which were relysed. Visual inspection of the spectra obtained for Fig. 4.10b, taken 10 hr after that shown i n Fig. 4.12d, again revealed a weak but noticable WGA insensitive population. Seigneuret et a l (1984) also noted t h i s problem with spin labeled l i p i d s inserted into the red c e l l bilayer where a small but noticable percentage became free. - 168 -The only experiments to give an increase i n SP with increasing WGA i n which detailed binding measurements were made were the i n i t i a l experiments done with the impure Sigma WGA. However, i n most cases, SP increased as a function of increasing WGA concentration (Fig. 4.4). The broadening of the spectra by WGA has been reported by Lee & Grant (1979, 1980) for SL-glycophorin, as was also found here for isolated SL-glycophorin (Fig. 4.3), for spin labeled NANA of fetuin (Kwok & Landsberg, 1982) and for spin labeled NANA of lymphocytes (Ross et a l 1983, although they don't mention which l e c t i n s they used). Binding of antibodies to nitroxides also show spectral broadening (Rey & McConnell, 1976). An unusual result was reported by Fel i x et a l (1982) i n that they found a decrease i n with the addition of WGA to t h e i r SL-ghosts (F-B SL-ghosts). As shown i n Chapter 3 (Discussion), F-B SL-ghosts are comparable to those made here except have a lower number of spins/ghost and a greater proportion associated with the l i p i d s . The signal from t h i s type of ghost i s extremely low (at 2 x 10 6 gain on a Varian E-3 with a 200 u l sample size) and not detectable from the small volumes needed for temperature controlled experiments (24 u l ) . In the present work addition of an equal volume of 10 mg WGA/ml to these ghosts resulted i n a 55% increase i n the spectral parameters (Fig. 4.6), not a decrease as Fel i x et a l (1982) observed. As discussed above, the discrepancy could be associated with a large population of non-interacting spins, or, since t h e i r l e c t i n concentration was low (1.5 mg/ml) possibly not enough binding occurred and the a l t e r a t i o n was due to experimental fluctuations. They had a considerable attachment of the spin probe to g l y c o l i p i d s (30%). At low WGA concentrations spin labeled gangliosides appear to f i r s t decrease i n their T c and as the WGA increases "Cc al s o does (Lee et a l , 1980). Felix et a l - 169 -(1982) attribute t h e i r decrease i n f c (33%) to the hypotonic buffers they used, as opposed to the higher i o n i c strength PBS used i n the other experi-ments, and to complex glycophorin-membrane protein interactions which, following WGA binding, were f e l t to leave the probe i n a less r e s t r i c t i v e environment. In t h i s work the yields of spins/ghost were low for F-B SL-ghosts so no detailed analysis could be done. Errors were high due to the low signal-to-noise r a t i o . The l i t e r a t u r e supports the finding that WGA does broaden the ESR signal of S L - s i a l i c acid.moieties. The above re s u l t s , and the shape of the broadened spectra (Figs. 4.4 and 4.11) cast considerable uncertainty on the use of SP to describe the reduction i n spin l a b e l motion associated with the experimental manipulations used i n t h i s study. Clearly, the broadened signal i s complex and most l i k e l y contains contributions from several types of spectra. This i s pa r t i c u l a r l y true when the background signal i s appreciable (Figs. 4.5 and 4.10b). In the absence of a detailed spectral analysis, however, SP i s used i n t h i s thesis as an index of changes i n spin la b e l motion. I t should not be interpreted as representing a uniform change i n correlation time for the entire spin population. 4.4(b) Mechanism of broadening Peters et al,(1979) found WGA bound better to ovine submaxillary mucin and fetuin i f the NANA was converted to NANA^, suggesting that the exocyclic chain i s a potential source of s t e r i c hindrance. NANA7 glycophorin glycopeptides are retained on WGA a f f i n i t y columns (Bahavanadan & K a t l i c , 1979) and Kahan et a l (1976) used l0r/NaB[ 3H] 4 treated - 170 -red c e l l s to monitor the i s o l a t i o n of glycophorin on a WGA a f f i n i t y column. Hence, oxidation per se does not seem to reduce WGA a f f i n i t y for NANA. However, Lee and Grant (1980) found that SL-glycophorin (labeled on the s i a l i c acids) was a weaker i n h i b i t o r of WGA agglutination than unmodified glycophorin, again suggesting that side chain constituents can interfere with WGA.binding to some degree. In t h i s work, since only about 13% of the s i a l i c acids were modified on the SL-ghost, s p e c i f i c WGA binding to SL-ghosts didn't d i f f e r from that of normal ghosts. The fact that the spin l a b e l signal broadens upon the addition of WGA implies i t i s attaching to, or very near, s i a l i c acid. The exocyclic t r i o l i s the furthest removed from the binding s i t e i n c r y s t a l studies (Wright 1980), however, and since WGA can also bind i n t e r n a l sugar units, direct binding might not be expected. Using the Stokes-Einstein relationship t = 4Tf^r"5/3kT (where r = the radius of the protein, equal to 25 angstroms for WGA (Wright, 1980), and q = the vi s c o s i t y of the solution) a value of -8 o 1.7 x 10 s at 22 C i s calculated for WGA i n solution. For the largest SP of a WGA/SL-ghost p e l l e t (1.7) a Tc of about 3 x 10" 9s i s implied using equation [9] i n Chapter 1. For the WGA/SL-glycophorin p e l l e t , T/c = 1.7 x 10" 9s (SP = 1.23, Table 4.1). I f WGA was binding to the nitroxide i t s e l f , therefore, much higher SPs would have been produced. Hence i t seems l i k e l y that the signal broadening i s due to s t e r i c hindrance as opposed to actual binding to the spin probe. 4.4(c) WGA binding isotherm WGA binding to red c e l l s i s complex. WGA appears to bind s p e c i f i c a l l y to red c e l l s i n a p o s i t i v e l y cooperative fashion ( i . e . i n i t i a l l y bound - 171 -material enchances subsequent binding) at low WGA concentrations and there appears to be two types of s p e c i f i c binding s i t e s as detected by Scatchard plots (Anderson & Lovrien, 1981; Adair & Kornfeld, 1974), as well as considerable non-specific adsorption. The cooperativity and two binding s i t e s have also been found for CHO c e l l s by Stanley & Carver (1977). Sigmodial curves (an indication of positive cooperativity) have been seen i n the isotherms of WGA binding to glycophorin-containing liposomes (Ketis et a l , 1980; Redwood et a l , 1975) and to ganglioside liposomes (Redwood & Polefka, 1976). Addition of 50 mM glcNAc usually results i n instantaneous release of WGA from c e l l s (Anderson & Lovrien, 1981; Evans & Leung, 1984). This r e v e r s i b i l i t y by glcNAc has usually been taken to i n f e r that WGA i s binding to glcNAc residues. However, glcNAc i s found to reverse and i n h i b i t binding of WGA to g l y c o l i p i d vesicles not containing any glcNAc (Boldt et a l , 1977). Wright (1980) and Cuatrecasas (1973) postulate that glcNAc binds to an unoccupied s i t e on WGA causing either conformational changes i n the l e c t i n resulting i n dissociation of the WGA from the c e l l (Cuatrecasas, 1973) or de s t a b i l i z i n g WGA:WGA aggregates on the c e l l surface (Wright, 1980) (native WGA crystals disintegrate i n the presence of glcNAc but are stable i n the presence of NANA, Wright, 1979). Whether NANA on the c e l l binds to a l l four WGA s i t e s i s unknown. In a l l cases, 50 mM glcNAc appears not to be an ef f e c t i v e i n h i b i t o r of t o t a l WGA binding to ghosts or SL-ghosts. Non-specific binding appears to be very large (Fig. 4.7). Perhaps lysing the red c e l l s into ghosts exposes other s i t e s not normally available to WGA, as suggested by Horisberg & Rosset (1977). These investigators found twice the number of receptors on isolated membranes as was found via c o l l o i d a l gold-WGA labeling of red - 172 -c e l l s . The non-specific binding was highest for SL-ghosts that had been re-lysed overnight (Fig. 4.7). These SL-ghosts also had a WGA insensitive signal (as shown by the lack of correlation of SP vs WGA bound per SL-ghost, Fig. 4.10b), showing that the d e t a i l s of sample preparation were important. Perhaps excessive lysing produced inside-out ghosts (Steck, 1974) which rendered the S L - s i a l i c acid inaccesible to WGA. I f impurities and non-specific binding are compensated for, a l l the s p e c i f i c binding data, taken over an extended period of time, are reproducible (Fig. 4.8) even though there was scatter for the t o t a l WGA/SL-ghost i n the binding isotherm (Fig. 4.7). Points from both p u r i f i e d WGA and impure WGA preparations for which the amount of binding was calculated from the WGA band on SDS PAGE (Appendix D) f a l l on the same l i n e . This indicates that non-specific binding fluctuates and causes the scatter i n the data, due apparently to variations i n ghost preparations. Cooperativity plus at least two different classes of binding s i t e s for WGA result i n a Scatchard plot far removed from the straight l i n e obtained for simple saccharide binding to WGA i n solution. There are also indications that WGA binding to c e l l s may not be i n equilibrium. Scatchard plots of WGA binding to CHO c e l l s or fat c e l l s are effected by the number of c e l l s i n the assay (Stanley & Carver, 1977; Cuatrasas, 1973). Data sent to E. Evans by Lovrien shows the same phenomenon for red c e l l s (Evans, personal communication). One possible explanation for t h i s i s the fact that WGA binding to c e l l s i s very fast (Schneble & Bachi, 1975) and the rate of release slow (Cutrecasas, 1973) possibly resulting i n the c e l l s i n i t i a l l y exposed to l e c t i n binding most of the WGA, leaving less i n solution for the remaining c e l l s . - 173 -For WGA binding to SL-ghosts, therefore, i t i s not surprising that the Scatchard plots are complex. Detailed analysis of a l l the s p e c i f i c binding data v i a Scatchard analysis indicated that binding did not f i t a simple model (Fig. A.9). Inconsistencies i n the Scatchard plots for the various experiments are also due to the s e n s i t i v i t y of t h i s plot compared to binding isotherms (compare Fig. A.8 to A.9). At low levels of binding, there i s positive cooperativitiy (as depicted by a positive slope for the Scatchard pl o t , Fig. A.9), where the WGA broadening i s strongest. The best high WGA concentration data gives saturation at about 9 x 10 6 WGA/SL-ghosts, si m i l a r to that c i t e d i n the l i t e r a t u r e (8 x 10 6, Adair & Kornfeld, 197A; 1.2 x 10 7, Anderson & Lovrien, 1981 and from 1.0 x 10 7 for old to 1.7 x 10 7 for young red c e l l s , Choy et a l , 1979). A.A(d) WGA receptor on red c e l l s Although WGA binding to red c e l l s has been studied for some time, the exact location of the receptor(s) has not been i d e n t i f i e d with certainty, as i s seen from the following. WGA a f f i n i t y columns w i l l bind glycophorin (Kahane et a l 1976; Adair & Kornfeld, 197A), acetylcholinesterase (Ravozzolo et a l , 1983), band 3 (Bjerrum et a l , 1981, although Futhmayr et a l , 1976 is o l a t e band 3 by c o l l e c t i n g the material that f a i l s to bind to a WGA a f f i n i t y column), band A.5 (the glucose transporter) (Froman et a l , 1981) and g l y c o l i p i d s (Bowles & Hanke, 1977). WGA w i l l bind to glycophorin i n solution (Fig. A.3 and Lee & Grant, 1979) and to glycophorin-containing vesicles (Ketis & Grant, 1982; Redwood et a l , 1975). WGA w i l l also bind to vesicles made of human red c e l l l i p i d s i f they are high i n s i a l o g l y c o l i p i d s (Van der Steen et a l , 1983), to ghosts so - 174 -extensively protease treated the no protein can be detected (Gordon et a l , 1977) and to vesicles containing globoside and ceramide trihexose isolated from red c e l l s (Boldt et a l , 1977), although Rendi et a l (1976) reported that WGA would not bind to vesicles made from red c e l l l i p i d s . Although WGA binds to desialylated red c e l l s (Adair & Kornfeld, 1974; Schnebli & Bache, 1975 and Anderson & Lovrien, 1980), i t i s with a weaker association constant (Adair & Kornfeld, 1974). WGA columns bind desialylated glycophorin (Kahne et a l , 1976), implying s i a l i c acid i s not the only binding s i t e . However, Bhavanandan & K a t l i c (1979) found asialoglycophorin to be a poor i n h i b i t o r of WGA hemagglutination. Burness & Pardoe (1983) found as the s i a l i c acid content of glycophorin decreased, so did i t s a b i l i t y to i n h i b i t WGA hemagglutination (a 28% decrease i n the s i a l i c acid content resulted i n a 42% decrease i n i n h i b i t i o n ) . However, Fukuda & Osawa (1973) could release 50% of the s i a l i c acid with no effect on i t s a b i l i t y to i n h i b i t WGA hemagglutination. This suggests that certain s i a l i c acids are more important than others for WGA binding. A simi l a r conclusion was drawn by Kronis & Carver (1982) based on studies of sialylactose binding to WGA, the alpha 3 structure binding better than the NANA alpha 6-lactose (the alpha 3 gal and the alpha 6 galNAc both are found on glycophorin). Adair & Kornfeld (1974) found the RBC components released from a WGA a f f i n i t y column in h i b i t e d better than glycophorin isol a t e d v i a the LIS/phenol method. Kahane et a l (1976) didn't find any PAS 3 i n the material released from a WGA column by glcNAc, implying that not a l l the PAS positive components react equally with WGA (although Bjerrum et a l , 1980 found that PAS 3 did react with WGA). - 175 -Bhavanandan & K a t l i c (1979) state that WGA s p e c i f i c i t y i s r e l a t i v e rather than absolute, depending of the density of receptors. High densities of s i a l i c acid and glcNAc lead to strong WGA binding, thus the topography of c e l l surfaces i s important i n the formation of stable bonds. The results presented i n t h i s thesis show that WGA added to various fractions isolated from ghosts reacts to the fractions r i c h i n PAS proteins (the NaOH p e l l e t , the Triton X-100 extract and to isolat e d SL-glycophorin) (Table A.l and Figs. A.2 and A.3). SDS PAGE gels of these fractions show that unless the WGA i s reduced a large f r a c t i o n of PAS stainable components won't run into the gel , implying that.the. WGA i s complexing with the PAS proteins (data not shown). The isolated l i p i d , however, didn't appear to interact with the WGA when resupended i n PBS. The experimental evidence obtained i n t h i s work strongly supports the idea that i n i t i a l binding of WGA to red c e l l s occurs at one s i t e per 5 glycophorin. Cooperativity i s seen i n Scatchard plots up to 2 to 10 x 10 WGA/cell (Fig. A.9) ( Anderson & Lovrien, 1980 reported 5 x 10 5), approximately equal to the number of glycophorin copies i n the red c e l l membrane (Anstee, 1981; Furthmayr, 1981). SL-glycophorin reacted strongly with WGA, completely p r e c i p i t a t i n g out upon the addition of WGA (except for 3.6%, believed to be non-covalently associated, spins) at a r a t i o of 1.5 moles glycophorin to 1 mole of WGA (Fig. A.7). For a precipitate to form, 2 moles of glycophorin to 1 mole WGA i s expected, but due to the oligomeric nature of glycophorin i n aqueous solutions (existing i n aggregates of 10 to 20 monomers, Chapter 3, Discussion), t h i s lower molar r a t i o of glycophorin to WGA i s expected. The resultant ESR signal i n t h i s precipitated p e l l e t was broad (SP = 1.23), l i k e that of SL-ghost with WGA. Visual inspection of - 176 -the p e l l e t spectrum reveals two superimposed spectra, one broad, which may possibly be a combination of unbound and WGA bound SL-glycophorin. SDS PAGE gels of t h i s p e l l e t again wouldn't run into the gel unless reduced f i r s t , even though the i n d i v i d u a l components ran into the gels, reduced or not (data not shown). No non-covalently-associated spins were detected i n the p e l l e t even when the sample was run at 10°C (as determined by v i s u a l inspection of the spectrum for sharp l i n e s ) . Since most of the spins are associated with the PAS proteins (Chapter 3) and as shown here, t h i s spin population i s sensitive to the addition of WGA, WGA i s almost certainly binding to glycophorin on ghosts. Even though there i s a variety of s i t e s on the red c e l l capable of binding WGA, the i n i t i a l binding i s to only glycophorin as shown by the altered ESR spectrum of SL-ghosts after the addition of the l e c t i n . 4-4(e) Correlation between SP and WGA binding The fact that the SL-ghost spectral parameters do increase with the addition of WGA suggests that the spin probe i s sensitive to WGA binding. This increase i s l i n e a r i n the amount s p e c i f i c a l l y bound up to 5 x 10 5 WGA/ghost (for impure Sigma WGA, Fig. 4.10a). For nonspecific binding there appears to be no effect upon spin motion u n t i l about 3.6 x 10 5 WGA/ghost, where again SP s t a r t s to increase. These breaks i n the plot at 3.6 and 5.4 5 x 10 WGA/SL-ghost are close to the number of glycophorins/red c e l l . This suggests that 1:1 binding of WGA to SL-glycophorin on the c e l l results i n spin label broadening. After t h i s 1:1 binding with glycophorin, the non-specific binding broadens also. - 177 -The selective broadening at low WGA levels indicates the WGA does indeed bind to glycophorin. The correlation i s only seen i f one calculates the s p e c i f i c binding, implying that the non-specific binding occurs to something other than the PAS proteins. In t h i s low WGA concentration region, the binding i s cooperative (Fig. 4.9) yet one s t i l l sees the correlation indicating labeling doesn't affect the cooperative mechanism. Scatchard analysis (Fig. 4.9) shows complex binding yet the SP vs WGA bound i s l i n e a r i n t h i s region. One can conclude that ESR provides a more s p e c i f i c picture of WGA binding than adsorption measurements i n t h i s instance since the effects of non-specific binding or of the impurities i n impure radiolabeled preparations are not seen. It has been postulated that there i s a population of s i a l i c acids on red c e l l s (40% of the t o t a l s i a l i c acids) which are more "external" than the others, as indicated by t h e i r a c c e s s i b i l i t y to neuraminidase and periodate treatment i n intact c e l l s i n comparison to ghosts (Singer & Morrison, 1974; Blumenfeld et a l , 1972). The lower a c c e s s i b i l i l t y of a fraction of the red c e l l s i a l i c acids may be due to the size of neuraminidase and periodate since i t i s known that tetramethyl ammonium ions are excluded r e l a t i v e to K + and Na + (Brooks, 1973) or possibly due to subtle rearrangements i n the glycocalyx after l y s i s . Since only 15% of the s i a l i c acids get oxidized (roughly two per glycophorin for isolated SL-glycophorin at 131 ug s i a l i c acid/mg, Chapter 3), one would assume the most accessible would react with periodate and thus be spin labeled. One would also expect these external s i a l i c acids to be the f i r s t seen by WGA. Since i t appears that the spin labels are on the average 16 angstroms apart, (Chapter 3) WGA could easily accomodate both S L - s i a l i c acids located on glycophorin (the distances - 178 -between binding s i t e s on WGA are from 22 angstroms, from a primary to a secondary s i t e to 42 angstroms for the two secondary s i t e s with the primary s i t e s being 31 angstroms apart (Wright, 1980)). As the WGA concentration increases, the binding would be to a variety of s i t e s and the spin l a b e l broadening would become less (in proportion). This would explain why the s p e c i f i c broadening st a r t s to l e v e l o f f and the non-specific to increase. 4.4(f) Binding of Other Lectins The s e l e c t i v i t y of t h i s spin l a b e l for s i a l i c acid-binding proteins i s also seen i n Table 4.2. Lectins such as PNA, ConA and SBA don't bind to glycophorin or s i a l i c acid residues and have no effect on the SP after addition to SL-ghosts. Slug- l e c t i n ( s p e c i f i c for s i a l i c acid) binding to SL-ghosts does cause an increase i n SP (30%). Of interest i s the RCA l e c t i n which also increased SP upon exposure to SL-ghosts. I t has been postulated that RCA doesn't bind to glycophorin (Adair & Kornfeld, 1974; Jakobovits et a l , 1981; Triche et a l , 1975) yet glycophorin can i n h i b i t RCA hemagglutination (Adair & Kornfeld, 1974), RCA binds a l l 3H/G0 labeled glycoproteins of red c e l l s (Tsao et a l , 1981) of which glycophorin i s one, (Chapter 2) and i t binds to band 3 and PAS 1 on SDS PAGE gels (Tanner & Anstee, 1976). The data i n Table 4.2 implies that RCA does indeed bind to glycophorin. - 179 -4.5 CONCLUSION Addition of WGA to SL-ghosts, selective extractions of SL-ghosts or isolated SL-glycophorin results i n decreased spin l a b e l m o b l i l i t y as detected by an increase i n the spectral parameter. WGA appears to bind to glycophorin as determined by i t s interaction with selective extractions of ghosts and i t s p r e c i p i t a t i o n of isolated glycophorin. WGA binding to SL-ghosts was indistinguishable from that to normal ghosts, showing that spin labeling the c e l l s didn't a l t e r them detectably with respect to t h i s reaction. Binding analysis showed high and variable non-specific binding. I f noncovalent (background) ESR signal was removed a l i n e a r increase of spectral parameter vs the amount of s p e c i f i c a l l y bound WGA/SL-ghost was observed. The binding of impure Sigma WGA was more e f f i c i e n t at broadening the ESR signal than Vector WGA which had been p u r i f i e d . For s p e c i f i c 5 binding, the change i n SP started to l e v e l o f f at about 5 x 10 WGA/ghost and was interpeted to indicate s p e c i f i c binding of one mole WGA per mole of glycophorin. More general binding occurs at higher WGA levels as determined by an increase i n plots of SP vs non-specific binding and by a reduced correlation between s p e c i f i c binding and spectral a l t e r a t i o n s . WGA also broadened the ESR signal of F-B SL-ghosts, contrary to the results reported by Felix & B u t t e r f i e l d (1980). The proposed mechanism of broadening by WGA i s that of s t e r i c hindrance by the WGA molecule of the spin probe. Only s i a l i c acid-binding proteins affected the ESR signal of SL-ghosts indicating that the spin probe s e n s i t i v i t y was limited to t h i s moiety. - 180 -4.6 SUMMARIZING DISCUSSION The o r i g i n a l question asked i n t h i s thesis was can techniques used i n chemistry on complex isolated systems be applied successfully to less well defined b i o l o g i c a l systems in_ s i t u ? A positive answer i s implied by the results presented above. ESR proved to be sensitive enough to analyse a s p e c i f i c component of the glycocalyx of the c e l l membrane. Conditions could be found where maximum labeling and minimal perturbations occurred with retention of s p e c i f i c i t y for glycophorin. Elimination of i n t e r f e r i n g spins was a prerequisite for understanding and interpreting spin labeling data, however. Spin labeling s i a l i c acid proved to be more informative than spin labeling the gal/galNAc residues, perhaps due to t h e i r a c c e s s i b i l i t y (the s i a l i c acids are the terminal residues of oligosaccharides on membrane sialoglycoproteins). This a c c e s s i b i l i t y also appeared to be responsible for the s e n s i t i v i t y of these spin labeled s i a l i c acids to WGA binding. Spin labeling reported s e l e c t i v e l y on the s p e c i f i c binding of WGA (detected as an increased SP) without necessitating compensation for non-specific binding or impurities i n WGA. In fact, impure WGA was found to be more effec t i v e i n ESR studies than p u r i f i e d commercial WGA. Out of the complex binding of WGA to red c e l l membranes, the spin label monitored s p e c i f i c binding to glycophorin s i a l i c acids. This s e l e c t i v i t y was also found with Slug and RCA l e c t i n binding. Hence, i n s i t u spin labeling would seem to be a promising way to study any receptor a c t i v i t y i n which s i a l i c acid i s thought to be involved. - 181 -APPENDIX A SDS PAGE SDS PAGE was run on 5 or 10% c y l i n d r i c a l gels as follows, following general instructions provided by Bio-rad Laboratories. A l l reagents were of electrophoresis purity from Bio-Rad Laboratories, Richmond CA. The gel was f i r s t cast into 7 x 12.5 mm glass tubes. The gel solution consisted of 5% (w/v) acrylamide, 0.13% (w/v) bisacrylamide (or 10% acrylamide and 0.26% bisacrylamide), 0.1% (w/v) SDS, 0.03% (v/v) tetramethylethylenediamine (TEMED) and 0.05% (w/v) ammonium persulfate for the 5% acrylamide and 0.01% ammonium persulfate for the 10% acrylamide, i n a buffer of 0.205 M t r i s acetate pH 6.1. The gel was layered with isobutanol to ensure a f l a t surface and allowed to polymerize at room temperature for one hr. The isobutanol was then rinsed o f f with d i s t i l l e d water. The sample buffer consisted of 12% sucrose, 1% (w/v) SDS i n 0.041 M t r i s acetate pH 6.1. I f the sample was to be reduced 0.04 M d i t h i o t h r e i t o l (DTT) or 10% mercaptoethanol was used. Samples were incubated at 100°C for three minutes and then the tracking dye, Pyronin Y was added. I f the sample was i n solution, the sample buffer was added at a volume to ensure that the sample buffer was never diluted more than 50% (ghost samples were usually mixed 50:50 with the sample buffer). I f the sample was freeze-dried protein, f i n a l protein concentrations of 1 mg/ml sample buffer were usually made. - 182 -The electrophoresis buffer was 0.205 M t r i s acetate pH 6.1 with 0.1% (w/v) SDS. A Hoefer Scientic Instrument (San Fransisco CA) gel electrophoresis chamber model DE 102 was used. Electrophoresis was run at s l i g h t l y less than 1 watt/tube at 4°C. A discontinous system as described by Laemmli (1970) was also used with a 5% stacking gel and either a 10, 12 or 15% s°eparating gel. Staining. Gels were f i r s t fixed (after being run) i n methanol: acetic acid: water (5:1:5) overnight, and i f staining for protein were also incubated with 0.02% coomassie blue. The coomassie blue stained gels were then destained i n 10% (v/v) acetic acid u n t i l the background was clear. I f staining for carbohydrate, the PAS (periodic acid Schiff base) st a i n was used. The fixed gels were incubated i n 1% periodate i n 5% acetic acid for one hour. The periodate was then eliminated by f i r s t washing i n 0.5% arsenite i n 5% acetic acid and then i n 0.1% arsenite i n 5% acetic acid. The arsenite was then washed out with 5% acetic acid and the Schiff reagent was added and the gels incubated overnight. The gels were then destained with 0.1% sodium metabisulfite i n 0.01 M HC1 repeatedly u n t i l no pink color could be seen with the addition of formaldehyde. The Schiff reagent consisted of one g of Basic Fuchsin with one g of sodium metabisulfite i n 200 ml of 0.1 N HC1. The mixture was f i l t e r e d with activated charcoal. For Basic Fuchsin s t a i n , the same procedure was followed except the periodate oxidation step was eliminated. - 183 -Scanning gels Gels were scanned on a Helena Auto Scanner model R4-077 (Helena Laboratories, Beaumont, Texas) which sets the largest peak to 100% absorbance or on a Beckman model 25 spectrophotmeter attached to a 24-25 ACC recorder. Coomassie blue stained gels were recorded at 595 nm and the PAS and Basic Fuchsin stained gels at 525 nm. Radioactive gels were s l i c e d i n 1 mm int e r v a l s with a Bio Rad gel s l i c e r model 195. Iodinated gel s l i c e s were counted with an LKB Compugamma counter and t r i t i a t e d gel s l i c e s were counted by the method of Aloyo (1979). Molecular weight estimations Molecular weight standards hemoglobin (16000 daltons), human serum albumin (66000 daltons), cytochrome c (12400 daltons) and ovalbumin (45000 daltons) were used. Their dimers and trimers also appeared on the gels. A plot of log(molecular weight) vs t h e i r R^ . (distance protein traveled/distance tracking dye traveled) was made and from the R^ . of the unknowns a molecular weight was calculated. - 184 -APPENDIX B ESR INTEGRATION ESR spectra were run on the Varian E-3 at ambient temperature using a f l a t c e l l of 200 j j l capacity. Samples were run several times, scan rates and response times such that the maximum peak-heights were ensured for each run. The spectrum, at a l a t e r time, was put onto a Hewlett Packard 9872A plott e r i n conjunction with the 9815A Hewlett Packard calculator, d i g i t i z e d (150 points) and plotted (program written by K.A. Sharp, l i s t e d below). This d i g i t i z e d spectrum was integrated and the resu l t also plotted. Four points were selected as zero points as i n f i g . B.l and a second integration was performed (on t h i s rezeroed f i r s t integration) and i t s numerical value stated. Fig. B.l shows the i n i t i a l ESR spectrum, the d i g i t i z e d spectrum and the resultant integrations. The program for the Hewlett Packard calculator i s l i s t e d on the three pages following Fig. B . l . - 185 -F i g . B . l (A) ESR spectrum of SL-ghosts, (B) d i g i t i z e d spectrum of (A), (C) the f i r s t and second i n t e g r a t i o n s o f the d i g i t i z e d spectrum (B). The dotted l i n e i s the a c t u a l f i r s t i n t e g r a t i o n and the s o l i d l i n e the rezeroed f i r s t i n t e g r a t i o n . The second i n t e g r a t i o n i s of the rezeroed f i r s t i n t e g r a t i o n along with i t s numerial value. - 186 -OPERATION 1. Set up the plott e r position the spectrum on the pl o t t e r bed as shown i n Fig. B.2 2. Load the program 3. When 'P]/ move the d i g i t i z i n g s i t e (in pen # A s t a l l ) with the cursor controls on the p l o t t e r to position P]_ as defined i n Fig. B.2 and press ENTER on plotter A. When 'P 2' move the d i g i t i z i n g s i t e with the cursor controls to position P 2 as defined i n Fig. B.2 and press ENTER on plotter 5. When 'X min, max, Y min, max' enter 55 RUN/STOP, 89 RUN/STOP, -1 RUN/STOP, 0 RUN/STOP t h i s sets the scale for the pl o t t e r 6. When 'LOAD' a. enter 1 (means YES) RUN/STOP i f p l o t t i n g a stored spectrum from tape and go to step 8 b. enter 0 (means NO) RUN/STOP i f d i g i t i z i n g an ESR spectrum 7. D i g i t i z i n g the spectrum the d i g i t i z e r pen goes to point (0,0) on the p l o t t e r . a. move the d i g i t i z e r s i t e onto the spectrum by moving the v e r t i c a l cursor controls b. press ENTER on the p l o t t e r , d i g i t i z e r pen automatically moves an increment along the x-axis c. repeat 7a and b u n t i l 150 points have been d i g i t i z e d 8. When 'NEW PAPER' replace spectrum with blank paper then press RUN/STOP 9. The d i g i t i z e d spectrum i s plotted with pen #1 10. When 'STORE' a. enter 1 RUN/STOP and then the f i l e # RUN/STOP i f the spectrum i s to be stored b. enter 0 RUN/STOP i f the spectrum i s not to be stored 11. Plotter proceeds to plot the f i r s t integration with pen # 1 12. When 'A ZERO PTS' move d i g i t i z e r s i t e over the f i r s t of the four zero points (lowest x) and press ENTER on p l o t t e r , repeat for the other three points, the four points are printed after the l a s t one has been entered 13. Press RUN/STOP to plot the rezeroed f i r s t integration with pen # 2 - 187 -14. When 'OK* a. enter 0 RUN/STOP i f the rezeroed integration unsatisfactory and repeat step 12 and 13 b. enter 1 RUN/STOP i f rezeroed integration i s good, t h i s integration i s automatically replotted 15. When 'STORE' a. enter 1 RUN/STOP and the f i l e # RUN/STOP i f want the rezeroed integration to be stored b. enter 0 RUN/STOP i f don't want the integration stored 16. after step 15 the pl o t t e r automatically plots the second integration with pen # 3 and prints i t s numerical value. 1 F i g . B.2 Placement of the ESR spectrum onto the plott e r bed and the location of Pj_ and P 2. 00 02 03 04 05 06 07 06 09 10 11 12 13 14 15 17 19 20 21 22 23 25 27 28 29 30 31 33 34 35 36 38 39 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 71 73 74 75 76 77 78 79 80 81 82 84 85 86 87 89 90 91 92 93 94 95 96 FUNCT * CLEAR ffREGS CFG CFG CFG CFG CFG 1 6 0 0REGS 4 PENSL PRNTcc. P 1 7 END* DGTZR PRNTc* P 2 ? END « DGTZR XjY ROLLt X*Y FUNCT 0 PRNTc\ X TAB M I N i M A X TAB 7 LINE Y TAB M I N t M A X TAB 7 END * STOP STOP STOP STOP SCALE PRNTfX L 0 A D ? END ot 1 STOP IF X=Y GOTO 0 ENTER t 1 MOVE 1 STO i 1 5 0 STO FOR DGTZR - 188 -L05 F A*F 98 X#Y 200 ENTER + 99 STO I A 201 0 311 PRINT 101 CLEAR 202 PRNT« 312 RCL I E 102 1 204 F 314 STO R152 103. PLOTR 205 I 316 PRINT 105 NEXT A 206 L 317 RCL I G 106 GOTO 0122 207 E 319 STO R153 108 LBL 208 TAB 321 PRINT 05 209 * 322 RCL I H 110 PRNT<X 210 END* 324 STO R154 112 F 211 STOP 326 PRINT 113 I 212 RCDATA 327 STCP 114 L 213 CLEAR 328 PENUP 115 E 214 MOVE 330 LBL 116 t 216 PLOTA 04 117 END* 218 1 332 3 118 0 219 5 333 PENSL 119 ENTERf 220 0 335 RCL 0 120 STOP 221 PLOTA 336 STO A 121 LOAD 223 PENUP 337 RCL E 122 PRNTfX 225 1 338 STO F 124 N 226 STO A 339 RCL R152 125 E 227 RCL ROOO 341 RCL R151 126 W 229 STO C 343 _ 127 TAB 230 FOR A*F 344 RCL E 128 P 231 RCL A 345 RCL D 129 A 232 1 346 _ 130 P 233 - 347 1 131 E 234 STO B 348 IF SFG 5 132 R 235 RCL C 349 CLEAR 133 LINE 236 RCL I A 350 STO I 134 P 238 STO C 351 CFG 5 135 L 239 + 352 FOR A»F 136 0 240 2 353 RCL A 137 T 241 354 RCL D 138 7 242 RCL I B 355 139 ENDCX 244 + 356 RCL I 140 STOP 245 STO I A 357 * 141 1 247 5 358 RCL R151 142 IF X=Y 248 i 360 + 143 SFG 1 249 RCL A 361 RCL I A 144 PENUP 250 PENUP 363 XJ.Y 146 1 252 PLOTA 364 147 PENSL 254 PLOTA 365 IF SFG 2 149 CLEAR 256 NEXT A 366 STO I A 150 MOVE 257 3 368 5 153 PLOTA 258 PENSL 369 • 154 1 260 LBL 370 RCL A 155 5 03 371 PLOTA 156 0 262 CLEAR 373 NEXT A 157 PLOTA 263 MOVE 374 RCL E 159 PENUP 265 PLOTA 375 1 161 IF CFG 1 267 1 376 + 162 GOTO LOl 268 5 377 STO A 164 0 269 0 378 RCL G 165 STO A 270 PLOTA 379 STO F 166 1 272 PENUP 380 RCL R153 167 5 274 4 382 RCL R152 168 0 275 PENSL 384 _ 169 STO F 277 PRNTot 385 RCL G 170 FOR A-»F 279 4 386 RCL E 171 RCL I A 280 TAB 387 -173 RCL A 281 Z 388 i 174 PLOTA 282 E 389 IF SFG 5 176 NEXT A 283 R 390 CLEAR 177 LBL 284 0 391 STO I 01 285 TAB 392 CFG 5 179 PENUP 286 P 393 FOR A»F 181 PRNT<* 287 T 394 RCL A 183 S 288 S 395 RCL E 184 T 289 7 396 -185 0 290 ENOW. 397 RCL I 186 R 291 DGTZR 398 * 187 E 293 INT 399 RCL R152 188 294 STO D 401 + 189 END* 295 DGTZR 402 RCL I A 190 1 297 INT 404 405 406 407 Y - W 191 STOP 298 STO E 192 193 IF X=Y GOTO 0197 299 301 DGTZR INT IF SFG 2 STO I A 195 GOTO 0213 302 STO G 409 410 5 197 1 303 DGTZR 198 5 305 INT 199 0 306 307 309 STO H RCL I D STO R151 411 412 414 415 416 417 418 419 420 421 423 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 442 443 445 446 447 448 450 451 452 453 455 456 458 459 461 463 464 465 466 467 468 469 RCL A PLOTA NEXT A RCL G 1 + STO A RCL H STO F RCL R154 RCL R153 RCL H RCL G IF SFG 5 CLEAR STO I CFG 5 FOR A*F RCL A RCL G RCL I # RCL R153 + RCL I A X#Y IF SFG 2 STO I A 5 RCL A PLOTA NEXT A PENUP IF SFG 2 GOTO 0480 PRNT* 0 K 7 END 1 STOP IF X=Y 470 GOTO 472 GOTO 0474 0477 474 SFG 2 475 GOTO L04 477 CFG 2 478 GOTO L03 480 PRNT «* 482 S 483 T 484 0 485 R 486 E 487 ? 488 ENO <* 489 1 490 STOP 491 IF X=Y 492 GOTO 0496 494 GOTO 496 1 497 5 498 1 499 ENTER t 500 0 501 PRNT«. F I L E # 0512 503 504 505 506 507 508 ? 509 END <* 510 STOP 511 RCDATA 512 2 .513 PENSL 515 516 517 518 519 521 522 523 524 525 526 527 528 529 530 532 533 534 535 536 538 539 541 542 543 544 545 546 547 549 550 551 552 553 554 555 556 557 558 559 560 562 564 565 566 567 568 569 570 571 572 573 575 576 578 580 581 582 584 585 587 588 590 RCL D 1 • STO A RCL I D STO C RCL STO FOR RCL 1 H F A»F A STO B RCL C RCL I A STO C RCL I B + STO I A 2 0 0 NOP RCL A PLOTA NEXT A 0 ENTER t 1 ENTER t 1 ENTER t 1 5 0 YAXIS PRNT<* 2 N D TAB I N T END* RCL I F PRINT GOTO L06 LBL A 1 STO- A RCL I A RCL A PLOTA RETURN LBL 06 ENO - 189 -APPENDIX C I l l u s t r a t e d i n t h i s and the following Appendix are two examples of l e c t i n interactions, the fetuin:PNA system and WGA a f f i n i t y p u r i f i c a t i o n . Fetuin contains s i x sialo-oligosaccharide chains, of which three are binding s i t e s for PNA when the s i a l i c acid i s removed. Fetuin i s thus an excellent model to study for the neuraminidase, galactose oxidize spin-labeling reaction. WGA was p u r i f i e d by a variety of methods and raises questions concerning the s e l e c t i v i t y of the i s o l a t i o n method. C.l(a) Fetuin modification Fetuin (Type IV from Sigma) was iodinated by the lactoperoxidase method (see Chapter 4, section 4.2(d)) which resulted i n an a c t i v i t y of 4.10 +_ 0.6 8 125 125 x 10 cpm/mg. Free I was separated from I fetuin on a Sephadex G-25 (Pharmacia Fine Chemicals, Uppsala, Sweden) column. This was done 125 before each use to ensure no free I interfered with the experiment. Fetuin was desialylated with neuramindase (Vibrio cholerae) at 0.1 U/ml i n 0.05 M sodium acetate pH 5.5, 0.154 M NaCl, 9 mM CaCl 2, 0.025% azide at 37°C for 51.5 hours. Fetuin was spin labeled by the neuraminidase, galactose'oxidase method as i n Aplin et a l (1979) producing NAGO SL-fetuin. This method was found to release 73 + 3% of the bound s i a l i c acid. - 190 -C.l(b) Peanut agglutinin a f f i n i t y column The procedure i s a that of Cuatrecasas (1970). Five grams of wet Sepharose AB (Pharmacia) were modified with 0.3 g of CNBr dissolved i n O.A ml of a c e t o n i t r i l e diluted 1:1 with water. The CNBr activated Sepharose was suspended i n 10 ml of 0.1 M sodium bicarbonate, 0.5 M NaCl, pH 8.3 which also included 0.2 M gal , and the PNA added (Miles-Yeda, Rehovot, Israel or Sigma) (1 mg/ml i n the buffer with 0.2 M ga l ) . This was s t i r r e d gently for two hours at room temperature and then packed into polystyrene columns (Isolab Inc., Akron, OH) of 6 ml capacity and washed. The amount of PNA coupled to the beads was determined as the amount added reduced by the amount recovered i n the washing solutions. The resultant column had 0.7 mg PNA/ml Sepharose. C.2 Results PNA Fetuin interactions Addition of 16 pg (6 x 10^ cpm) fetuin i n PBS/azide to the 5 ml column of PNA sepharose, which bound desialylated RBCs, resulted i n 33% of the fetuin nonspecifically bound to the column (protein could not be removed with 0.2 M g a l , 1 M NaCl or ethylene g l y c o l ) . No other binding to the PNA column was observed either for fe t u i n , neuraminidase treated fetuin (with no detectable bound s i a l i c acid) or NAGO SL-fetuin even i f allowed to equilibrate for 15 min to ensure maximum binding at A°C. - 191 -Addition of PNA l e c t i n (2 mg i n 400 p i plus 18 mg SL-fetuin (SP = 0.03) i n 1 ml, both i n t r i s buffer) produced a precipitate. This precipitate was washed 3 times i n the t r i s buffer (each wash being 2 ml), with each successive wash decreasing the amount of material i n the p e l l e t (SP = 0.13 for the p e l l e t and 0.03 for the solution). Analysis of the p e l l e t on SDS PAGE showed i t to contain PNA and fetuin (data not shown). C.3 Discussion Fetuin i s a f e t a l c a l f serum sialoglycoprotein which bears s i x oligosaccharide types. Three are N-glycosidically linked and three O-glycosidically linked (Nilsson et a l , 1979; Baenziger & Fiete, 1979). The removal of s i a l i c acid exposes the gal beta l-3galNAc residue of the O-glycosidically linked oligosaccharides, a strongly binding sugar unit for PNA (Goldstein & Haynes, 1978). This O-glycosidically linked oligosaccharide i s also believed to be present on glycophorin A (15/molecule, Anstee, 1981, Furthmayr, 1981 and Lisowska et a l , 1980), the major sialoglycoprotein of the red c e l l membrane. NAGO SL-fetuin has already been studied by Aplin et a l (1979) so the modification of the system was not analyzed here. Spin labeling fetuin by the NAGO method produced no alterations as detected by SDS PAGE (data not shown). Since not a l l (73%) of the s i a l i c acid was removed by the neuraminidase, the p o s s i b i l i t y arose of populations of fetuin heterogeneous with regard to s i a l i c acid content and thus spin labeling. One would then expect that not a l l of the NAGO SL-fetuin would bind to a PNA a f f i n i t y column. - 192 -The PNA column made (0.7 mg PNA/ml) was found to bind desialylated red c e l l s i n a manner which was i n h i b i t e d or reversed with 10 mM gal i n PBS. The same preparation was found not to bind any of the above mentioned fetuin or modified fetuin fractions, however. A high degree of nonspecific binding occurred which was d i f f i c u l t to reverse. Sepharose i s known to allow some hydrophobic interactions (Lotan et a l , 1977) which may explain t h i s r e s u l t . Lotan et a l (1977) used a high capacity, charge free, non-leaching po l y a c r y l i c hydrazide sepharose column, with 5 mg PNA/ml column, and obtained some desialylated fetuin binding. I t i s known (at least for wheat germ a f f i n i t y columns) that clusters of receptors determine whether or not binding w i l l occur (Yamamoto et a l , 1981). The l e c t i n density of the column also appears to influence binding (Bhavanadan & K a t l i c , 1979), with low density columns binding only high clusters of binding s i t e s (such as are present on red c e l l s ) and the binding of lower density clusters (such as fetuin) occurring only when there i s a high l e c t i n density on the gels. Dulaney et a l (1979) l i s t several variables which appear to be important i n the binding of glycoproteins to l e c t i n s (such as temperature, time of contact, pH and i o n i c strength). Conditions chosen here were considered optimal by Lotan et a l (1977) so the most l i k e l y explanation for the lack of binding i s that of the PNA density of the sepharose column. The desialylated red c e l l s have a high density of receptor s i t e s available and bind to the PNA column. The desialylated fetuin (spin labeled or not) may not have bound because the receptor density and l e c t i n density were too low. I t i s also possible that the nonspecific binding (perhaps due to the impurities i n the fetuin) interfered with s p e c i f i c binding, s t e r i c a l l y hindering access to these s p e c i f i c s i t e s (unfortunately, the nonspecifically - 193 -saturated gels were not tested for t h e i r a b i l i t y to bind desialylated red c e l l s ) . Solutions studies support the hypothesis that i t i s the l e c t i n density of the PNA a f f i n i t y column which i s the factor d i c t a t i n g the binding since the SL-fetuin w i l l bind to PNA i n solution even though not completely desialylated (presumably because the l o c a l density of PNA or fetuin i s higher than on the g e l ) . Of the 12 possible gal residues to be spin labeled on desialylated f e t u i n , only three are optimal for PNA. About 30% of the s i a l i c acid remained on the molecules, probably decreasing the r a t i o of SL to optimal PNA binding s i t e , yet t h i s NAGO SL-fetuin s t i l l interacted with PNA i n solution. - 19A -APPENDIX D D.l WGA P u r i f i c a t i o n D.l(a) Fast protein l i q u i d chromatography (FPLC) The FPLC system (Pharmacia) includes a gradient programmer GP-500, two P-500 pumps and a single path monitor UV-1 set at 280 nm, whose output was displayed on a Hewlett Packard 7100 B s t r i p chart recorder. The column used was a Pharmacia mono S HR5/5 cation exchange column bearing -CH2-S0~ groups. This column was equilibrated with 0.05 M sodium acetate, 0.05 M NaCl, 0.025% azide at pH A. 3 (buffer A) which was also the buffer i n which WGA was dissolved. Buffer B was i d e n t i c a l to buffer A except the NaCl was raised to 0.A5 M. The program used was 0% buffer B for the f i r s t 15 minutes followed by 100% buffer B for 3 minutes then 0% buffer B for the l a s t 3 minutes. For the f i r s t 15 minutes, the flow rate was set at 0.5 ml/min, to allow WGA to bind to the column. After buffer B was applied, the flow rate was increased to 2 ml/min and remained at t h i s rate for the rest of the run. The column was re-equilibrated with buffer A for 3 minutes of the run, making the column ready for another WGA sample. A maximum of 10 mg WGA (in 500 JJI buffer) can be loaded on the column per run. After the WGA runs were finished, the column was washed with 0.1 M NaOH to elute any remaining protein. - 195 -D.l(b) CM-Sepharose CL-6B The procedure used was that of Kronis & Carver (1982), a modification of LaCelle (1979). To a 190 ml (2 x 73 cm) column of CM-Sepharose CL-6B (Pharmacia) equilibrated i n 0.05 M NaCl, 0.05 M sodium acetate, pH 4.3 (in 0.025% NaN3) was added 5 ml of 37 mg/ml WGA i n the same buffer and the column washed with t h i s buffer. The eluent was monitored on a LKB Uvicord Type 4701A at 253.7 nm and the WGA was eluted with 0.45 M NaCl i n the pH 4.3 acetate buffer. The column was subsequently washed with 0.1 M NaOH to remove residual protein. D.l(c) Chitin column The c h i t i n column (in 0.01 M t r i s HC1, pH 8.5, 1 M NaCl, 0.025% NaN?) was prepared according to Bloch & Burger (1974), using c h i t i n (crab s h e l l s ) purchased from Sigma. D.l(d) Ovomucoid column The ovomucoid column was prepared according to Marchesi (1972) using ovomucoid type 111-0 trypsin i n h i b i t i o r from Sigma: 974 mg of ovomucoid was dissolved i n 50 ml 0.2 M sodium bicarbonate, pH 8.6, i n 0.02% NaN^ and added to 38 ml of Sepharose CL-4B (Pharmacia) which had been CNBr activated by the addition of 10 g of CNBr i n 65 ml 2 M sodium carbonate. The coupling eff i c i e n c y was 10%, providing 1.2 mg/ml ovomucoid Sepharose. The WGA was loaded onto the column (1 x 18 cm) and the column - 196 -washed i n 0.05 M phosphate buffer, pH 7, 0.25 M NaCl and 0.025% NaN^. The a f f i n i t y bound WGA was then eluted with 0.1 M acetic acid (pH = 2.9). D.2 RESULTS D.2(a) SDS PAGE A sensitive measure of WGA impurity was obtained when iodinated WGA preparations (as described i n Chapter A section A.2(d)) were run on SDS PAGE. Sigma WGA l o t # A9C-97A0 was iodinated via the Iactoperoxidase/H 20 2 method, yie l d i n g an a c t i v i t y of 8.533 x 10 6 cpm/mg WGA. A 10% Laemmli SDS PAGE (Appendix A) was run on a 25 ug sample of iodinated WGA (reduced with d i t h i o t h r e i t o l ) mixed with 100 jug of cold WGA 195 of the same l o t (Fig. D.l). Only 58% of the I lab e l was found to be associated with the WGA peak, indicating about A0% impurity. Vector WGA, iodinated by the Iodo-bead method and p u r i f i e d on the CM-Sepharose column, was 97% pure, according to the SDS PAGE p r o f i l e (Fig. D.l). Iodinated WGA (Sigma WGA (l o t 90F-9515) 6.AA x 10 7 cpm/mg) had a microtitre the same as unmodified Sigma WGA (agglutinated 2% red c e l l s down to 5 jjg/ml). On a 10% Laemmli gel about 70% of the radiolabel associated with the WGA peak (Fig. D.2). Addition of red c e l l s ( f i n a l concentration 195 of 1%) to a 3.0 /jg/ml solution (in PBS) resulted i n only 19% of the I being l o s t from the supernatant bound within red c e l l p e l l e t . SDS PAGE of the WGA solution before and after addition of red c e l l s , revealed that one major peak had disappeared. The maximum a c t i v i t y had shifted from gel s l i c e s 13-18 to 27-30 (Fig. D.2). - 197 -3000 3000 B Fig. D.l SDS PAGE (12% laemmli) of reduced WGA, s l i c e d at 1 mm in t e r v a l s and counted for 1 2 5 I . (A) impure Sigma WGA (6 pg, l o t # 49C-9740) iodinated by the lactoperoxidase method. (B) pure Vector WGA (10yug) iodinated by the Iodo-bead method, shown here for comparison. - 198 -0 10 20 30 40 50 60 0 10 20 30 4 0 50 60 S L I C E NUMBER F i g . D.2 SDS PAGE (10% Laemmli) of unreduced 125I WGA. Gels s l i c e d at 1 mm in t e r v a l s and counted for 1 2 5 I . (A) stock 1 2 5 I Sigma WGA (A pg) and (B) the supernatant 1 2 5 I WGA (0.2 pg) after expusre to red c e l l s . - 199 -D.2(b) A f f i n i t y columns P u r i f i c a t i o n of the Sigma WGA ( l o t # 90F-9515) was attempted using a c h i t i n and an ovomucoid a f f i n i t y column. Table D.l tabulates the results of the a f f i n i t y columns with WGA, stating the amount of WGA loaded and the amount eluted. Fig. D.3 shows the SDS PAGE of these various fractions. TABLE D.l % BINDING OF IODINATED SIGMA WGA TO CHITIN OR OVOMUCOID COLUMNS A f f i n i t y column used Chi t i n column^ Ovomucoid column 2 WGA loaded (mg) 0.8 1.2 % unbound 35 95 % a f f i n i t y e l u t e d 3 5.4 1.2 % non-specifically bound 4 56 3.8 -'•the 19 ml column has a capacity of 185 mg WGA (Bloch & Burger, 1974) 2 t h e 18 ml column (1.2 mg ovomucoid/ml) has a capacity of 11 mg WGA (Marchesi, 1972) 3For the c h i t i n column, a f f i n i t y WGA eluted with 0.05 M HC1 for the ovomucoid column, a f f i n i t y WGA eluted with 0.1 M acetic acid 4iodinated Sigma WGA which could not be eluted. Analysis of the various fractions by SDS PAGE showed that the a f f i n i t y p u r i f i e d WGA of the c h i t i n column didn't bind the material which bound to red c e l l s while the ovomucoid column did ( a l l the radiolabel was situated i n the top 25% of the g e l , where the high molecular weight component which bound red c e l l s was situated (Fig. D.3)). - 200 -Fig . D.3 SDS PAGE (10%, pH 6.1) of several column fractions of 1 2 5 I Sigma WGA reduced. (A) the ovomucoid column fra c t i o n s , s l i c e d at 1 mm intervals and counted for I25j (cpm/slice) (except gel 3 where four gels were s l i c e d at two mm i n t e r v a l s , the s l i c e s i n register combined and counted for 1 2 5 I which resulted i n a decrease i n resolution because the s l i c e s were thicker and perfect alignment among gels was impossible) and (B) the c h i t i n column fractions stained with coomassie blue and scanned at 595 nm. (1) 88 pg of WGA before being added to the columns, (2) the WGA fra c t i o n which didn't bind to the column and (3) the WGA which bound to the column and was eluted with (A) acetic acid or (B) HC1. - 201 -D.2(c) Cation exchange chromatography The method which proved to be best for purifying WGA from commercial sources was cation exchange chromatography. Two methods were employed to purify WGA, a CM-Sepharose CL-6B (bearing C00~ groups) and FPLC, using mono S HR 5/5 (-Cht^SO") cation exchange column. Tables D.2 and D.3 tabulate the results of both types of column chromatography and Fig. D.4 the resultant SDS PAGE gels of Vector WGA from the cation exchange column fractions. No s a l t gradients were used to separate the l e c t i n into i t s i s o l e c t i n s as i n Kronis & Carver (1980), since the ind i v i d u a l i s o l e c t i n s are less soluble than when combined and the binding of each i s o l e c t i n was indistinguishable by nmr, c i r c u l a r dichroism, fluorescence and binding to Chinese hamster ovary c e l l s (LaCelle, 1979). The fraction which eluted i n high s a l t was considered the p u r i f i e d WGA. Any fractions not associated with the high s a l t (pure) fraction were considered impure. The protein that eluted with NaOH was always yellow and also considered an impurity. After the WGA was iodinated i n the acetate pH 4.3 buffer, i t was 125 separated from free I and p u r i f i e d simultaneously on the CM column. Vector WGA iodinated and run down the column gave the same results as i n Table D.3 (93% pure). - 202 -TABLE D.2 CM-SEPHAROSE CHROMATOGRAPY OF COMMERCIAL WGA Source Vector  t o t a l t i t r e * recovered (ug/ml) Sigma 2  t o t a l t i t r e recovered (ug/ml) EY t o t a l t i t r e recovered (ug/ml) Original 7 12.5 % unbound 0 A.2 3 500 6 13A % eluted by 0.A5 M NaCl 92 7 92 18 80 8 % eluted by 0.1 M NaOH 8 1A7 3.6 100 IA % recovered 90 89 92 Iminimum agglutination concentration 2 a l l fractions were yellow ^altered absorption spectrum, maxima at 320 nm and 282 nm TABLE D.3 FPLC CHROMATOGRAPHY OF VECTOR AND SIGMA WGA Source Vector t o t a l recovered t i t r e (ug/ml) 2 Sigma-1 t o t a l recovered t i t r e (ug/ml) Original % unbound O.A % eluted by 0.A5 M NaCl 97 % eluted by 0.1 M NaOH 2.A 13 38 13 25 7.7 86 3 20 60 25 72 ^fr a c t i o n eluted with 0.A5 M NaCl on CM-Sepharose of Table A.3 2minimum agglutination concentration 3 s t i l l yellow at 2 mg/ml - 203 -Fig. D.A SDS PAGE of various Vector WGA fractions (50 pg) from cation exchange columns, reduced and stained with coomassie blue. (A) 15% laemmli gels of the FPLC fractions, (B) 10%, pH 6.1 gels of CM-Sepharose fractions. (1) the i n i t i a l Vector WGA, (2) WGA which eluted with high s a l t and (3) protein eluted with 0.1 N NaOH. The numbers above the protein bands are the calculated molecular weights (+_ 2000 daltons). - 204 -D.3 General discussion Cation exchange chromatography detected impurities i n commercial preparations, (the fraction which did not bind to the columns at a l l and the yellow pigment eluted with NaOH) (Tables D.2 and D.3). LaCelle (1979) found an impurity which did not bind to a CM-Sepharose column which lowered the binding of Sigma WGA to CHO c e l l s . This impurity was f e l t to be toxic to the CHO c e l l s . In the present work, only 0.4% of the t o t a l f a i l e d to bind to the CM or mono S columns (Vector) (Tables D.2 and D.3). I t agglutinated red c e l l s at about three times the concentration of pu r i f i e d WGA. For Sigma WGA, t h i s fraction comprised about 7-8% of the protein and had comparable binding to the p u r i f i e d fraction (Tables D.2 and D.3). The yellow pigment has also been referred to as an impurity. This may be a misnomer since i t does agglutinate c e l l s , although at a higher concentration (140 compared to 7-20 ug/ml for pure WGA). Analysis of t h i s fraction shows a complex mixture of proteins on SDS PAGE. At least four bands are present, ranging from 7000 to 28000 daltons (Fig. D.4). One major fraction had a molecular weight of 16000 +_ 2000 daltons, very close to that of WGA (18000 +_ 2000 daltons). This pigment was also seen by Bouchard et a l (1976). LeVine et a l (1972) found that combining i t with WGA greatly enhanced the s o l u b i l i t y (at neutral pH) of WGA. Thus impurities may aid i n s o l u b i l i z a t i o n of WGA and also bind to c e l l s . Bassett (1975) claims that the i s o l a t i o n method chosen selects a WGA fraction with a ch a r a c t e r i s t i c s o l u b i l i t y , the amino acid composition - 205 -r e f l e c t i n g the s o l u b i l i t y . Nagata & Burger (1974) state WGA i s s l i g h t y soluble i n water while Bouchard et a l (1976) find WGA soluble up to 20 mg/ml water. A l l stock solutions of WGA were soluble up to 20 mg/ml i n PBS. Some Sigma preparations were equally soluble i n water. Other preparations (e.g. BRL) precipitated out when dialysed against water; Vector WGA precipitated (at 20 mg/ml) i n 0.05 M acetate, 0.05 M NaCl, pH 4.3, i f l e f t for 4 hours at room temperature. Table D.4 l i s t s eight d i f f e r e n t reported amino acid compositions for WGA, a l l normalized to a molecular weight of 18000 (except for Wright et a l , 1984, who use 20600). Some amino acids are consistent while others appear to vary considerably (aspartic acid varies from 13.6 to 18/monomer, glutamic acid from 15.7 to 19/monomer). WGA i s known to contain four i s o l e c t i n s (1, 11 , 11., and 111) (Rice & E t z l e r , 1975; LaCelle, 1979), the t e t r a p l o i d s t r a i n containing no i s o l e c t i n 11 (Rice, 1976) and the hexaploid strains (the source usually used for WGA) varying i n t h e i r r e l a t i v e amounts (Rice, 1976; Lotan et a l , 1973), thus explaining the amino acid v a r i a b i l i t y . Addition of red c e l l s to iodinated WGA (Sigma WGA ( l o t 90F-9515)) 125 resulted i n only 19% of the I being l o s t from the supernatant bound within red c e l l p e l l e t . SDS PAGE of the WGA solution before and after addition of red c e l l s , revealed that one major peak had disappeared (Fig. D.2). The ovomucoid a f f i n i t y column also appeared to bind t h i s protein fraction but the c h i t i n column did not (Fig. D.3 and Table D.l). The i s o l a t i o n method chosen selects a WGA. Perhaps, l i k e many bi o l o g i c a l e n t i t i e s , co-factors are required to perform i t s function i n vivo (which are mainly unknown for l e c t i n s ) . Glucosaminidases, proteases and - 206 -TABLE D.4 AMINO ACID COMPOSITION OF WGAa r e f e r - 1 2 3 4 5 6 7 8 9 10 ance i s o -l e c t i n 1 11 1 11 1 11 11 amino-acid Asp 15.3 17 14 13.6 15 18 14.9 15.1 15 15 15 15.3 15 Thr A.6 5 4 4.6 5 A 5.0 4.0 4 3 4.2 3.7 4 Ser 13.3 15 12 11.3 12 1A 13.7 13.5 8 9 10.4 9.7 16 Glu 16.0 16 19 16.6 16 19 16.4 16.0 17 16 16.1 15.7 15 Pro 5.6 6 6 5.3 5 8 5.2 6.2 5 5 4.9 5.5 6 Gly ' 39.7 AO 38 36.2 Al A3 41.0 40.5 40 37 40.7 40.4 42 Ala 9.8 12 9 9.9 10 10 10.1 9.4 10 10 9.7 9.2 10 Cys 27.A 32 28 33.2 30 1A 37.7 36.6 nd nd 32.2 32.0 32 Val 1.1 1 1 1.5 1.5 1 0.9 1.1 1 1 0.9 0.8 1 Met 0.9 1 2 1.8 3 3 2.0 2.0 1 1 1.4 1.2 2 He 1.8 3 2 2.3 1.5 2 2.1 2.1 2 2 1.7 2.3 2 Leu A.l 6 4 4.5 5 A 4.0 4.0 4 4 3.8 4.1 4 Tyr 7.1 6 7 6.8 7 7 8.0 7.4 6 6 7.4 6.5 7 Phe 2.8 3 3 3.0 3 3 3.3 2.9 3 2 2.8 2.3 2 His 1.7 1 2 1.7 2 2 0.0 2.0 0 2 0.1 2.0 2 Lys 7.5 5 7 6.1 8 8 7.9 6 7 8 6.7 7.9 6 Arg A.O 3 4 3.8 5 5 4.0 3.4 5 3 3.8 3.9 4 Try nd 2 3 2.3 3 - 3.2 2.9 nd nd 1 t o t a l 162 174 165 16A 173 16A 177 177 nd nd 171 nd = not determined 1 and 11 are i s o l e c t i n s WGA 1 and WGA 11 f o r m a l i z e d to a molecular weight of 18000 for the monomer except reference 10 which has a molecular weight of 20600 reference 1. Shaper e_t a l , 1973 2. Bassett, 1975 3. Nagata & Burger, 1974 4. Allen et a l , 1973 5. Bouchard et a l , 1976 6. Erni et a l , 1980 7. Rice & Et z l e r , 1975 8. LaCelle, 1979 9. Wright, 1981 10. Wright et a l , 1984. - 207 -high molecular weight contaminants could bind c h i t i n columns (Bloch & Burger, 1974) and poor recoveries were found using ovomucoid columns that had been used previously (Bhavanandan & K a t l i c , 1979). Brown et a l (1976) isolat e d a mitogen from wheat germ on a c h i t i n column, Nagata & Burger (1974) co-isolated a trypsin i n h i b i t o r along with WGA and WGAs of varying s o l u b i l i t y and impurities have been shown. Perhaps, i t i s these impurities which aid WGA i n i t s function (in Chapter 4, impure Sigma binding to SL-ghosts produced the broadest spectra). Just as the PNA:fetuin system showed that the density of the l e c t i n or receptor i s important i n i t s s e l e c t i v i t y , the WGA studies show that the i s o l a t i o n procedure of the l e c t i n may also be important. -208-REFERENCES INTRODUCTION Aplin, J. Ph D. Thesis, University of B r i t i s h Columbia (1979) Bernstein, M.A. Ph D. Thesis, University of B r i t i s h Columbia (1983) Geyer, G., Makovitzky, J. J. Micro. 119:407-414 (1980) Gahmberg, C.G. J. B i o l . Chem. 251:510-515 (1976) Hatton, M.W.C., Marz, L. and Regoeczi, E. Trends i n B i o l . S c i . 8:323-325 (1983) . Hughes, R.C. i n "Essays i n Biochemistry" vol.11 1-36 (Campbell, P.N. and Aldridge, W.N. eds.) (1975) Hughes, R.C. "Membrane Glycoproteins" Butterworths, London (1976) Monsigny, M. B i o l . C e l l u l a i r e 36:209-212 (1979) Rauvala, H. Trends i n B i o l . S c i . 8:287-291 (1983) Schrevel, J . , Gros, D. and Monsigny, M. Prog. Histoc. Cytoc. 14: 1-269 (1982) Winzler, R.J. i n "Glycoproteins, Their Composition, Structure and Function" second edition 1268-1299 (Gottschalk, A. ed.) El s v i e r pub. Co. Amsterdam (1972) Yalpani, M. Ph D. Thesis, University of B r i t i s h Columbia (1980) -209-CHAPTER 1 Alberts, B., Bray, D., Lewis, J . , Raff, M., Roberts, K. and Watson, J.D. 'Molecular Biology of the C e l l ' Garland Publishing, N.Y. (1983) Alderman, E.M., Fudenberg, H.H. and Lovins, R.E. Blood 58:351-359 (1981) Anderson, R.A. and Lourien, R. i n 'Erythrocyte Membrane 2: Recent C l i n i c a l and Experimental Advances' 207-226, A. Liss Inc., N.Y. (1981) Anstee, D.J. Sem. i n Hematology 18:13-31 (1981) Aplin, J.D. Ph.D. Thesis, UBC (1979) Aplin, J.D., Bernstein, M.A., C u l l i n g , C.F.A., H a l l , L.D. and Reid, P.E. Carb. Res. 70:C9-C12 (1979) Bartosz, B. and Bryszewska, M. C e l l B i o l . Int. Report 7:1 (1983) Benga, G. i n 'Biochemical Research Techniques, p. 79-117 Wrigglesworth, J. M. (ed) Wiley and Sons N.Y. (1983) B e r l i n , N.I. and Berk, P.D. i n 'The Red C e l l , ' v.2, 2nd ed., p. 954-1016. Surgenor, D.M.N, (ed), Academic Press, N.Y. (1975) Berliner, L.J. Methods Enzymol. 49:418-480 (1978) Berliner, L.J. (ed) 'Spin Labeling, Theory and Application' Academic Press, N.Y. (1976) Berliner, L.J. (ed) 'Spin Labeling I I , Theory and Application' Academic Press, N.Y. (1979) Bernstein, M.A. Ph.D. Thesis, UBC (1983) Borch, R.F., Bernstein, M.D. and Durst, H.D. J. Amer. Chem. Soc. 93:2897-2904 (1971) Bowles, D.J. and Hanke, D.E. FEBS Lett. 82:34-38 (1977) Boyd, W.C., Shapleigh, E. J. Immun. 73:226-231 (1954) Brewer, G.J. i n 'The Red C e l l , ' v. 1, 2nd ed., p. 387-433. Surgenor, D.M. (ed), Academic Press, N.Y. (1975) Brewer, G.J., Dick, R.D., Aster, J.C. and Hebbel, R.P. Progress i n C l i n . B i o l . Res. 97:25-31 (1982) Brown, J.C. and Hunt, R.C. Int. Rev. Cytol. 52:277-349 (1978) Brown, P.A., Feinstein, M.B. and Shaafi, R. Nature 254:523-525 (1975) -210-Cabantchik, Z.I., Knauf, P.A. and Rothstein, A. Biochim. Biophys. Acta 515:239-302 (1978) Carruthers, A. and Melchior, D.L. Biochemistry 23:2712-2718 (1984) Cassoly, R. and Salhony, J.M. Biochim. Biophys. Acta 745:134-139 (1983) Childs, R.A., F e i z i , T., Fukuda, M. and Hakomori, S. Bioc. J . 173:333-336 (1978) Choury, D., Reghis, A., Pichard, A.-L. and Kaplan, J.-C. Blood 61:894-898 (1983) Cogan, U. and Schachters, D. Biochemistry 20;6396-6403 (1981) Cohen, CM. Sem. i n Hematology 20:150 (1983) D a n i e l l i , J.F. and Davson, H. J. C e l l . Comp. Physiol. 5:495-508 (1935) Davis, R.J. and Weiss, J.B. Bioc. Soc. Tran. 8:315-316 (1980) Davoust, J . , Michel, V., Spik, G., Montreuil, J. and Devaux, P.E. FEBS Lett. 125: 271-276 (1981) Deas, J.E., Lee, L.T. and Howe, C. Biochem. Biophys. Res. Commun. 82:296-304 (1978) Devaux, P.F., Davoust, J. and Rousselet, A. Bioc. Soc. Symp. 46:207-222 (1981) Dodge, J.T., M i t c h e l l , C. and Hanahan, D. Arch. Biochem. Biophys. 100:119-129 (1963) Dohnal, J.C, Potempa, CA. and Garvin, J.E. Biochim. Biophys. Acta 621:255-264 (1980) Dubinsky, W.P. and Racker, E. Memb. B i o l . 44:25-36 (1978) Erdmann, E. i n 'The Red C e l l Membrane: A Methodological Approach* p. 251-262 E l l o r y , J.C. & Ypung, J.D. (eds) Academic Press, N.Y. (1982) Eton, J.W. i n 'Erythrocyte Membranes 2: Recent C l i n i c a l and Experimental Advances,' p. 1-4 (1981) Fairbanks, G., Steck, T.L. and Wallach, D.F.H. Biochemistry 10:2606-2617 (1971) F e l i x , J.B. and B u t t e r f i e l d , D.A. FEBS l e t t e r s 115:185-188 (1980) Finean, J.B. and M i t c h e l l , R.H. i n 'Membrane Structure,' p. 1-35, Finean, J.B. & M i t c h e l l , R.H. (eds), Elsevier/North Holland (1981) -211-Fowler, V.M. and Bennett, V. J. B i o l . Chem. 259:5978-5989 (1984) Fukuda, M., Fukuda, M.N. and Hakomori, S. J. B i o l . Chem. 254:3400-3703 (1979) Fukuda, M., D e l l , A., Oates, 3.E. and Fukuda, M.N. 3. B i o l . Chem. 259:8360-8273 (1984) Furthmayr, H. 3. Supra. Mol. Struc. 9:79-95 (1978) Furthmayr, H. i n 'Biology of Carbohydrates,' v. 1, p. 123-198, Ginsberg, V. & Robbins, P. (eds). Wiley & Sons, N.Y. (1981) Gahmberg, C.G. 3. B i o l . Chem. 254:510-515 (1976) Gahmberg, C.G. and Andersson, L.C. Eur. 3. Bioc. 122:581-586 (1982) Gietzan, K. and Kolandt, 3. Bioc. 3. 207:155-159 (1982) G i l l e s , R.J. TIBS 7:1-2 (1982) Goldstein, I . J . and Haynes, C.E. Adv. Carb. Chem. B i o l . 35:128-340 (1978) Gorga, F.R., Baldwin, S.A. and Lienhard, G.E. Biochem. Biophys. Res. Commun. 91:955-961 (1979) Gorter, E. and Grendel, F. J. Expt. Med. 41:439-443 (1925) Gratzer, W.B. Bioc. J. 198:1-8 (1981) Green, G.A., Sikka, S.C., Kalro, V.K. J. B i o l . Chem. 258:12958-12966 (1983) Grefrath, S.P. and Reynold, J.A. Proc. Natl. Acad. S c i . USA 71:3913-3916 (1974) Grigorescu, F., White, M.F. and Kahn, CR. J. B i o l . Chem. 258:13708-13716 (1983) Grimes, A.J. 'Human Red C e l l Metabolism.' Blackwell S c i e n t i f i c , Oxford (1980) Haest, C.W.M. Biochim. Biophys. Acta 694:331-352 (1982) Haest, C.W.M., Kamp, D. and Deuticke, B. Biochim. Biophys. Acta 557:363-371 (1979) Haest, C.W.M., Plasa, G., Kamp, D. and Deuticke, B. i n 'Membrane Transport i n Erythrocytes,' p. 198-123, Lassen, U.V., Ussin, H.H. & Wieth, O.J. (eds) (1980) Hakamori, S. Sem. i n Hematology 18:39-62 (1981a) -212-Hakamori, S. i n 'International C e l l Biology,' p. 744-748, Schweizer, H.G. (ed). Springer-Verlag, N.Y. (1981b) Hamasaki, N., Harjoro and Minakami, S. Bioc. J. 170:39-46 (1978) Hanahan, D.J. Biochim. Biophys. Acta 300:319-340 (1973) Hillman, R.S. and Finch, CA. 'The Red C e l l Manual,' 4th ed. Davis Co., Philadelphia (1974) Hjelmeland, L.M. and Chrambach, A. Electrophoresis 2:1-11 (1981) Hudson, A. and Luckhurst, G.R. Chem. Rev. 69:191-225 (1969) Hyde, J.S. and Dalton, L.R. Chem. Phys. Lett. 16:568-572 (1972) Irimura, T., Tsuj'i, T., Tagami, S., Yamamoto, K and Osawa, T. Biochemistry 20:560-566 (1981) I s r a e l a c h v i l i , J.N. Biochim. Biophys. Acta 469:221-225 (1977) Jarnefelt, J . , Rush, J . , L i , Y. and Laine, R.A. J. B i o l . Chem. 253:8006-8009 (1978) J a r v i s , S.M., Fincham, D.A., E l l o r y , J.C, Paterson, A.R.P., and Young, J.D. Biochim. Biophys. Acta 772:227-230 (1984) Ja r v i s , S.M. and Young, J.D. i n 'Red C e l l Membranes: A Methodological Approach,' p. 263-273, E l l o r y , J.L. & Young, J.D. (eds). Academic Press, N.Y. (1982) Jenkins, R.E. and Tanner, M.J.A. Bioc. J. 161:139-147 (1974) Jones, M.N. and Nickson, J.K. FEBS Lett. 115:1-8 (1980) Jost, P. and G r i f f i t h s , O.H. Methods. Enzymol. 49:369-418 (1978) Juliano, R.L. Biochim. Biophys. Acta 300:341-378 (1973) Kabat, E.A. J. Supra. Mol. Struct. 8:79-88 (1978) Kahane, I., Ben-Chetrit, E., S h i f t e r , A., and Rachmilewitz, E.A. Biochim. Biophys. Acta 596:10-17 (1980) Kale, K., Kresheck, G.C and Van der Kooi, G. Biochim. Biophys. Acta 535: 334-341 (1978) Kay, M.M.B., Goodman, S.R., Sorensen, K., Whitfield, C.F., Wong, P., Zaki, L. and Rudloff, V. Proc. Natl. Acad. S c i . USA 80:1631-1635 (1983) Keith, A.D., B u i f i e l d , G. and Snipes, W. Bioph. J. 10:618-629 (1970) -213-Kelth, A.D., Sharnoff, M. and Cohn, G.E. Biochim. Biophys. Acta 300:379-419 (1973) Kim, Y.S., Perdome, J . , B e l l a , A. J r . and Nordberg, J. Proc. Natl. Acad. S c i . USA 68:1753-1756 (1971) Klingenberg, M. Nature 290:449-454 (1981) Klinger, R., Wetzker, R., Wenz, I., Dinjuo, U., ReiBmann, R. and Frunder, F. C e l l Calcium 5:167-175 (1984) K l i p , A., Dezial, M. and Walker, D. Biochem. Biophys. Res. Commun. 122:218-224 (1984) Knauer, B.B. and Napier, J . J . J. Amer. Chem. Soc. 98:4395-4400 (1976) Kokarin, A.I., Zamarayev, K.I., Grigoryan, G.L., Ivanov, V.P. and Rozanysev, E.G. B i o f i z i k a 17:31-42 (31-39 i n the english trans.) (1972) Kondo, T., Dale, G.'and Beutler, E. Biochim. Biophys. Acta 645:132-(1981) Kuypers, F., van Linde-Sibernius-Trip, M., Roelofsen, B., Tanner, M.J.A., Anstee, D.J. and Op Den Kamp, J.A.C. Bioc. J. 221:931-934 (1984) Laemmli, U.K. Nature 227:680-685 (1970) Landman, K.A. J. Theor. B i o l . 106:329-351 (1984) Lane, C.F. Synthesis March 135-146 (1975) Lange, Y., Gough, A. and Steck, T.L. J. Memb. B i o l . 69:113-124 (1982) Lee, A.G., B i r d s a l l , N.J.M. and Metcalfe, J.C. i n 'Methods i n Membrane Biology,' v. 2, p. 134, Korn, E.D. (ed) (1974) Lee, P.M. and Grant, C.W.M. Bioc. Biop. Res. Commun. 90:856-863 (1979) Lelo, T.L. and Marchesi, V.T. J. B i o l . Chem. 259:4603-4608 (1984) Liao, T.-H., Gallop, P.M. and Blumenfeld, 0.0. J. B i o l . Chem. 248:8247-8253 (1973) Lienhard, G.E., Baldwin, J.M., Baldwin, S.A., Gorga, F.R. i n 'Structure & Function of Membrane Proteins,' p. 325-333, Quagliaiello, E. & Palmier, F. (eds). Elsevier Publishing (1983) Likhtenstein, G.I. 'Spin Labeling .Methods i n Molecular Biology,' Wiley-Interscience (1976) Lin , S. and Snyder, C.E. J r . J. B i o l . Chem. 252:5464-5471 (1977) -214-Lippert, J.L., Gorczyca, L.E. and Meiklejohn, G. Biochim. Biophys. Acta 382:51-57 (1975) L i s , H. and Sharon, N. Ann. Rev. B i o l . 42:541-574 (1973) Lovrien, E.R. and Anderson, R.A. J. C e l l . B i o l . 85:534-548 (1980) Luoma, G.A., Herring, F.G. and Marshall, A.G. Biochemistry 24:6591-6598 (1982) Lux, S.E. Nature 28^:426-429 (1979) Macara, I.G. and Cantley, L.C. i n 'Cell Membranes: Methods and Reviews,' v. 1, p. 41-87, Elson, E., Frasier, W. & Glaser, L. (eds) (1983) Makino, S. Adv. Bioph. 12:131-184 (1979) Marchesi, V.T. B i r t h Defects Original A r t i c l e Series 14:127-138 (1978) Marchesi, V.T. Sem. i n Hematology 16:3-20 (1979a) Marchesi, V.T. J. Membrane B i o l . 51:101-131 (1979b) Marchesi, V.T. Blood 61:1-11 (1983) Marchesi, V.T., Furthmayr, H. and Tomita, M. Annu. Rev. B i o l . 45:667-698 (1976) Marcus, D. and Cass, L. Science 164:553 (1969) Marcus, D., Kunda, S.K. and Suzuki, A. Sem. Hemat. 18:63-71 (1981) Marsh, D. Mol. B i o l . Biochem. Biophys. 31:51- (1981) Matsumoto, A.M.E. and Osawa, T. J. Bioc. 87:847-854 (1980) M i l l e r , J.A., Gravallese, E. and Bunn, H.F. J. C l i n . Inv. 65:896-901 (1980) Morse 11, P.D. Biochem. Biophys. Res. Comm. 77:1486-1491 (1977) Mueller, T.J., L i , Y.T. and Morrison, M. J. B i o l . Chem. 254:8103-8106 (1979) Mueller, T.J. and Morrison, M. i n 'Erythrocyte Membranes 2: Recent C l i n i c a l and Experimental Advances,' p. 95-112. A. Liss Inc., N.Y. (1981) Muller, H. and Lutz, H.U. Biochim. Biophys. Acta 729:249-254 (1983) Mullins, R.C. and Langdon, R.G. Biochemistry 19:1205-1212 (1980) Nakajima, M., Yoskimoto, R., Irimura, T. and Osawa, T. J. Bioc. 86:583-586 (1979) -215-Nelson, G.F. i n 'Blood and Lipoprotein Quantitation, Composition and Metabolism,' p. 3, Nelson, G.J. (ed). Wiley Interscience, N.Y. (1972) Nickson, J.K. and Jones, M.D. Bioc. Soc. Trans. 8:300-309 (1980) Nicolson, G.L. Int. Rev. Cytol. 39:89-190 (1974) Nigg, E.A., Bron, C , Girardet, M. and Cherry, R.J. Biochemistry 19:1887-1893 (1980) Nordio, P.L. i n 'Spin Labeling, Theory and Applications p.29-35 (Berliner, L.J. ed.) Acad. Press NY (1976) Ott, P., Ariano, B.H., Binggeli, Y. and Brodbeck, U. Biochim. Biophys. Acta 729:193-199 (1983) Overton, E. V i e r t e l j a h r s s c h r i f t der Natarforschenden Gesellschaft Zurich 40:159 (1895) Palek, J. (ed) Sem. i n Hematology 20 (1983) Peters, L.M. and Grant, C.W.M. Advances i n Exp. Med. and B i o l . 174:119-131 (1983) Pinder, J.C. and Gratzer, W.B. J. C e l l . B i o l . 96:768-775 (1983) Pontremoli, S., Sparatore, B., Melloni, E., Salamino, F., Michetti, M., M o r e l l i , A., Benatti, U. and de Flora, A. Biochim. Biophys. Acta 630:313-322 (1980) Poole, CP. J r . 'A Comprehensive Treatise on Experimental Techniques,' 2nd ed. J. Wiley & Sons (1983) Prohaska, R., Koerner,Jr., T.A.W., Armitage, I.M. and Furthmayr, H. J. B i o l . Chem. 256:5781-5791 (1981) Rendi, R., Kuettner, CA. and Gordon, J.A. Biochem. Biophys. Res. Commun. 72:1071-1076 (1976) Reynolds, J. and Tanford, CT. J. B i o l . Chem. 245:5161-5165 (1970) Ro, J.-Y., Maria, P.D. and Kim, S. B i o l . J. 219:743-749 (1984) Robertson, J.D. Biochem. Soc. Symp. 16:3-43 (1959) Rosenblum, B.B. i n 'Erythrocyte Membranes 2: Recent C l i n i c a l and Experimental Advances,' p. 251-265 (1981) Rosse, W.F. i n 'Zinsser Microbiology,' 18th ed., p. 341-352, J o k l i k , W.K., Wi l l e t , H.P. & Amos, D.B. (ed). Appelton Century Croft-Norwalk Conn (1984) -216-Rothstein, A., Knauf, P.A. and Cabantchik, Z.I. i n 'Surface of Normal and Malignant C e l l s , ' p. 356-387, Hughes, R.C. (ed). J. Wiley & Sons, N.Y. (1979) Rozantsev, E.G. 'Free Nitroxyl Radicals.' Plenum, N.Y. (1970) Rsienbenlist, K. and Taketa, F. J. B i o l . Chem. 258:11384-11390 (1983) Rubin, R.W. and Minkowski, C. Biochim. Biophys. Acta 509:100-110 (1978) Sadler, J.E., Paulson, J.C. and H i l l , R.L. J. B i o l . Chem. 254:2112-2119 (1979) Saleh, E.A. and Wheeler, K. Biochim. Biophys. Acta 684:157-171 (1982) Salhany, J.M. J. C e l l . B i o l . 23:211-222 (1983) Schleicher, E., Scheller, L. and Weiland, O.H. Biochem. Biophys. Res. Commun. 99:1011-1019 (1981) Schreier, S., Polnaszek, C.F. and Smith, I.CP. Biochim. Biophys. Acta 515:375-436 (1978) Schrier, S.L. Blood 50:227-238 (1978) Schweizer, E., Angst, W. and Lutz, H.U. Biochemistry 21:6807-6918 (1982) Seigneuret, M., Zachocoski, A., Hermann, A. and Devaux, P.E. Biochemistry 23:4271-4275 (1984) Shanahan, M.F. Biochemistry 22:2750-2756 (1983) Sharon, F.J. and Grant, C.W.M. Biochem. Biophys. Res. Commun. 74:1039-1045 (1977) Sharon, N. and L i s , H. Methods i n Memb. B i o l . 3:147-200 (1975) Sharp, K.A. Ph D. thesis University of B. C. (1985) Shohet, S.B. and Beutler, E. i n 'Hematology,' 3rd ed., p. 345-353, Williams, W.J., Beutler, E., Ersleu, A.J. & Lichtman, M.A. (eds). McGraw-Hill (1983) Siegel, D.L., Goodman, S.R., Branton, D. Biochim. Biophys. Acta 598:517-527 (1980) Singer, S.J. and Nicholson, G.L. Science 175:720-731 (1972) Smith, I.CP. i n 'Biological Applications of Electron Spin Resonance Spectroscopy,' p. 483-539, Bolton, J.R., Borg, D. & Swartz, H. (eds). Wiley-Interscience (1972) -217-Steck, T.L. J. Supr. Mol. Struc. 8:311-324 (1978) Stillmark, H. 'Uber R i z i n , ein g i f t i g e s Fermentaus dem Samen von Ricinus  communis L und einigen anderen Euphorbiaccon, 1 Inaug. Diss. Dorpat (1888) Stone, T.J., Backman, T., Nordio, P.L. and McConnell, H.M. Proc. Natl. Acad. S c i . USA 54:1010-1017 (1965) S u j i , T.T., Irimura, T. and Osawa, T. Bioc. J . 187:687-686 (1980) Tanner, J.M. 'Current Topics i n Membranes and Transport, 1 v. 11, p. 279-325 (1978) Tanner, J.M. and Anstee, D.J. Bioc. J . 153:271-277 (1976) Thomas, D.B. and Winzler, R.J. J. B i o l . Chem. 244:5943-5946 (1969) Thompson, S. and Maddy, A.H. i n 'Red C e l l Membrane. A Methodological Approach,' p. 67-93, E l l o r y , J.C. & Young, J.D. (eds). Academic Press, N.Y. (1982) Tillman, W., Cordua, A. and Schroter, W. Bioc. Biop. Acta 382:157-171 (1975) Ts u j i , T., Irimura, T. and Osawa, T. Carb. Res. 92:328-332 (1981) Van Deenen, L.L.M. FEBS Lett. 123:3-15 (1981) Van Deenen, L.L.M. and De Grier, J . i n 'The Red Blood C e l l , ' 2nd ed., p. 147-211, Surgenor, D.M. (ed). Academic Press, N.Y. (1974) Van Lenton, L. and Ashwell, G. J. B i o l . Chem. 246:1889-1894 (1971) Wagh, P.V. and Bahl, O.P. CRC C r i t . Rev. Bioc. 10:307-377 (1981) Wagner, H., Zimmer, G. and Lacke, L. Biochim. Biophys. Acta 771:99-102 (1984) Waisman, D.M., Smallwood, J . , Lafreniere, D. and Rasmussen, H. FEBS Lett. 145:337-340 (1982) Wallach, D.F.M., Verma, S.P. Biochim. Biophys. Acta 382:542-551 (1975) Watkins, W.M., and Morgan, W.M. Nature 177:521-522 (1952) Waterton, J.C. and H a l l , L.D. J. Amer. Chem. Soc. 101:3697-3698 (1979) Weaver, D.C. and Marchesi, V.T. J. B i o l . Chem. 259:6165-6169 (1984) Weaver, D.C, Pasternach, G.R. and Marchesi, V.T. J. B i o l . Chem. 259:6170-6175 (1984) -218-Weed, R.I. and Reed, C.F. Amer. J. Med. 41:681-698 (1966) White, M.D. and Ralston, G.B. Biochim. Biophys. Acta 599:569-579 (1980) Whitfield, C , Kay, M.M.D., M i l l e r , J. and Goodman, S.R. Fed. Proc. 42:2197 (1983) Williams, W.J., Beutler, E., ERsler, A.J. and Lichtman, M.A. (eds) •Hematology,' 3rd ed. McGraw-Hill (1983) Wu, J.S.R., Kwong, F., J a r v i s , S.M. and Young, J.D. J. B i o l . Chem. 258:13745-13751 (1983a) Wu, J.S.R., J a r v i s , S.M. and Young, J.D. Bioc. J. 214:995-997 (1983b) Yalpani, M. Ph D Thesis University of B. C. (1980) Yatzio, S., Abelink, P., Rachmilewitz, E.A., C i v i d a l l i , G. and Kahane, I. Israel J. Med. S c i . 14:1124-1126 (1978) Yoshima, H., Furthmayr, H. and Kobata, A. J. B i o l . Chem. 255:9713-9718 (1980) Young, J.D., Jones, S.E.M. and E l l o r y , J.C. Biochim. Biophys. Acta 645:157-160 (1981) Z u r i n i , M., Krebs, J . , Penniston, J.T. and C a r o f o l i , E. J. B i o l . Chem. 259: 618-627 (1984) Zwaal, R.F.A. Biochim. Biophys. Acta 515:163-205 (1978) Zwaal, R.F.A., Roelofren, B. and Colley, CM. Biochim. Biophys. Acta 30: 159-182 (1973) -219-CHAPTER 2 Abraham, G. and Low, P. S. Biochim. Biophys. Acta 597:285-291 (1980) Aminoff, D., Ghalambor, M.A. and Hinrich, C.J. i n 'Erythrocyte Membranes 2: Recent C l i n i c a l and Experimental Advances' p. 269-278. Alan Liss Inc. N.Y. (1981) Aplin, J. A. Ph D Thesis, University of B r i t i s h Columbia (1979) Aplin, J. A., Berstein, M.A., Cu l l i n g , C.F.A., H a l l , L.D. and Reid, P.E. Carb. Res. 70:c9-cl2 (1979) Bartosz, G. and Leyko, W. Blut Al:131-136 (1980) B i e r i , V.G. and Wallach, D.F.H. Biochim. Biophys. Acta A06:A15-A23 (1975) B u t t e r f i e l d , D.A. i n 'Biological Magnetic Resonance' V.A p.1-78 Berliner, L.J. and Reuben, J (eds). Plenum press N.Y. (1982) Carter, W.G. and Sharon, N. Arch. Biochem. Biophys. 180:570-582 (1977) Dodge, J.T., M i t c h e l l , C. and Hanahan, D. Arch. Biochem. Biophys. 100:119-129 (1963) Eylar, E.H., Madoff, M.A., Brody, O.V. and Oncley, J.L. J. B i o l . Chem. 237:1992-2000 (1962) Fairbanks, G., Steck, T.L. and Wallach, D.F.H. Biochemistry 10:2606-2617 (1971) F e l i x , J.B. and B u t t e r f i e l d , D.A. FEBS l e t t . 115:185-188 (1980) Fung, L.W.M. and Simpson, M.J. FEBS l e t t . 108:269-273 (1979) Gahmberg, G.C. J. B i o l . Chem. 251:510-515 (1976) Gahmberg, G.C. and Hakomori, S J. B i o l . Chem. 2A8:A311-A317 (1973) Gattegno, L., Perret, G., Fabia, F., Bladsier, D. and C o r n i l l o t , P. Carb. Res. 95:283-290 (1981) Goldberg, I.B. J. Mag. Res. 32:233-242 (1978) Goldstein, I . J . and Hayes, C.E. Adv. Carb. Chem. 35:127-340 (1978) Greenwalt, T.J. and Steane, E.A. B r i t . J. Haem. 25:207-215 (1973) Greig, R.A. and Brooks, D.E. Nature 282:738-739 (1979) -220-Hamiltion, G.A., DeJersey, J. and Adolf, P.K. in'Oxidases and Related Redox systems' p.103-124. King, T.E., Manson, H.S. and Morrison, M (1973) Hatton, M.W.C. and Regoeczi, E. Biochim. Biophys. Acta 438:339-346 (1976) Herring, F.G. and P h i l l i p s . P.S. J. Mag. Res. 62 to be published (1985) Ja f f e , C.L., L i s , L. and Sharon, N. Bioc. Biophys. Res. Comm. 91:402-409 (1979) Jokinen, M. Scand. J. Haem. 26:272-280 (1981) Kaplan, J . , Canonico, P.G. and Caspary, W.J. Proc. Nat. Acad. S c i . USA 70:66-70 (1973) Khodadad, J.K. and Weinstein, R.S. J. Membr. B i o l . 72:161-171 (1982) Lammel, B. and Maier, G. Biochim. Biophys. Acta 622:245-358 (1980) Lee, P.M., Ketis, N.V., Barber, K.R. and Grant, C.W.M. Biochim. Biophys. Acta 601:302-314 (1980) Lee, P.M. and Grant, C.W.M. Biochem. Biophys. Res. Comm. 90:856-863 (1979) Levine, S, Levine, M., Sharp, K.A. and Brooks, D.E. Biophys. J. 42:127-135 (1983) ~~ Lowry, O.H., Rosenbrough, N.J., Farr, A.L. and Randall, R.J. J. B i o l . Chem. 193:265-275 (1951) Luner, S.J., Strugeon, P., Szklarek, D. and McQuistion, D.T. Vox Sang. 28:184-199 (1975) Markwell, M.A.K., Haas, S.M., Biegar, L.L. and Tolbert, N.E. Anal. Bioc. 87:206-210 (1978) Masouredis, S.P. i n 'Hematology' p. 1308-1319 Williams et a l eds. McGraw H i l l Book Co. N.Y. (1972) Mishra, R.K. and Passow, H. J. Memb. B i o l . _l:214-244 (1969) Morse 11, P.D. Biochem. Biophys. Res. Comm. 77:1486-1491 (1977) Morse 11, P.D., Lusczakoski, D.M. and Simpson, D.A. Biochemistry 18:5021-5029 (1979) Mueller, T.J., L i , Y.-T. and Morrison, M. J. B i o l . Chem. 254:8103-8106 (1979) Noji, S, Inoure, F. and Kon, H. J.Bioc. Biop. Methods 5:251-258 (1981) -221-Oshiro, Y. and Eylar, E.H. Arch. Bioc. Biop. 127:476-489 (1968) Reid, P.E., Cu l l i n g , C.F.A., Ramey, C.W.., Dunn, W.L. and Clay, M.G. Can. J. Bioc. 55:493-503 (1977) Ross, A.H. and McConnell, H.M. Biochemistry 14:2793-2798 (1975) Schlegel, R.A., Gerbeck, CM. and Montgomery, R., Carb. Res. 7:193-199 (1968) Schneider, H. and Smith, I.CP. Biochim. Biophys. Acta 219:73-80 (1970) Schnell, K.F., Elbe, W., Kasbauer, J. and Kaufmann, E. Biochim. Biophys. Acta 732:266-275 (1983) Seaman, G.V.F. and Heard, D.H. Blood 18:599-604 (1961) Seaman, G.V.F. i n The Red C e l l V. I I p.1135-1229 Surgeon, R. ed. (1975) Seaman, G.V.F., Knox, R.J., Nordt, F.J. and Regen, D.M. Blood 50:1001-1011 (1977) Shelton, K.R. and Rogers, K.S. Anal. Bioc. 44:134-142 (1971) Shiga, T., Suda, T. anmd Maeda, N. Biochim. Biophys. Acta 466:231-244 (1977) Shiga, T. and Maeda, N. Biorheology 17:485-499 (1980) Steck, T.L. and Dawson, G. J. B i o l . Chem. 249:2135-2142 (1974) Suda, T., Maeda, N. and Shiga, T. J. Bioc. 87:1703-1713 (1980) Surolia, A.K., Appukuttan, P.S., Pain, D. and Bachhawat, B.K. Anal. Bioc. 105:436-440 (1980) Wallach, D.F.H., Verma, S.P., Weidekamm, E. and B i e r i , V. Biochim. Biophys. Acta 356:53-67 (1974) Yamaguchi, T., Koga, M., Takehara, H. and Kimoto, E. FEBS l e t t . 141:53-55 (1982) V a l e r i , P.S. and Zaroulis, C.G. New England J. Med. 287:1307-1313 (1972) Vassar, P.S., Hards, J.M., Brooks, D.E., Hagenberger, B. and jSeaman, G.V.F. J. C e l l B i o l . 53:809-818 (1972) V i c t o r i a , E.J., Mahan, L.C and Masouredis, S.P. Proc. Nat. Acad. S c i . USA 78:2898-2902 (1981) Zimmer, G. Lacko, L. and Kruger, E. Bioc. Pharm. 30:2362-2364 (1981) -222-CHAPTER 3 Abraham, G. and Low, P.S. Biochim. Biophys. Acta 597:285-291 (1980) Aloyo, V.J. Anal. Bioc. 99:116-118 (1979) Aminoff, D. Bioc. J. 81:384-392 (1961) Anstee, D.J. Sem. i n Hemat. 18:13-31 (1981) Aplin, J.D. Ph.D. Thesis, University of B r i t i s h Columbia (1979) Aplin, J.D., Bernstein, M.A., Cu l l i n g , C.F.A., H a l l , L.D. and Reid, P.E. Carb. Res. 70:c9-cl2 (1979) Ashwell, G. and Morell, A.G. Adv. Enzymol. 99:19-24 (1974) Beeley, J.G., Blackie, R. and Baxter, A. Bioc. Soc. Trans. 5:1725-1726 (1977) Bjerrum, O.J. and Bog-Hansen, T.C. Biochim. Biophys. Acta 455:66-89 (1976) Blumenfeld, 0.0., Gallop, P.M. and Liao, T.H. Biochem. Biophys. Res. Comm. 48:242-251 (1972) Borch, R.F. and Durst, H.D. J. Amer. Chem. Soc. 91:3396-3397 (1969) Brooks, D.E. i n "Human Blood Groups, F i f t h Int. Convoc. Immunol. Buffalo NY 27-35, Karger, Basel (1977) Brossmer, R. and Bahn, B. FEBS l e t t e r s 42:116-118 (1974) Cherry, R.J., Nigg, E.A. and Beddard, G.S. Proc. Nat. Acad. S c i . USA 77:5899-5903 (1980) Cruz. T.F. and Gurd, J.W. Anal. Bioc. 108:139-145 (1980) Cu l l i n g , C.R.A., Reid, P.E., Ramey, C.W., Dunn, W.L. and Clay, M.G. Can. J. Bioc. 55:778-782 (1977) Dahr, W., Uhlenbruck, G. and Bird, G.W.G. Vox Sang. 27:28-42 (1974) Dahr, W., Uhlenbruck, G., Schmalisch, R. and Janssen, E. Human Genetics 32:121-135 (1976) Dodge, J.T., M i t c h e l l , C. and Hanahan, D. Arch. Bioc. Biop. 100:119-129 (1963) Egmond, M.R., Williams, R.J.P., Welsh, E.J. and Rees, D.A. Eur. J. Bioc. 97:73-83 (1979) -223-F e l i x , J.B. and B u t t e r f i e l d , D.A. FEBS Letters 115:185-188 (1980) Folch, J . Less, M. and Stanley, G.H.S. J. B i o l . Chem. 226:497-509 (1957) F r i t z J r , O.G. Biophys. J. 46:219-228 (1984) Fukuda, M., Fukuda, M.F. and Hakomori, S. J. B i o l . Chem. 254:3700-3703 (1979) Furthmayr, H., Kahane, I. and Marchesi, V.T. J. Memb. B i o l . 26:173-187 (1976) ~~ Furthmayr, H. i n "Biology of Carbohydrates" Vol. 1 123-198 (Ginsburg, V. and Robbins, P. eds.) Wiley and Sons, N.Y. (1981) Gahmberg, C.G. and Andersson, L.C. J. B i o l . Chem. 252:5888-5894 (1977) Gahmberg, C.G., Virtanen, I. and Wartiovaara, J . Bioc. J. 171:683-686 (1978) Gattegno. L., Durand, G., Feger, J . , Perret, G., Felon, M. and C o r n i l l o t , P. Carb. Res. 117:255-262 (1983) Grefrath, S.P. and Reynold, J.A. Proc. Natl. Acad, S c i . USA 71:3913-3916 (1974) — Heitzmann, H. and Richards, F.M. Proc. Nat. Acad. S c i . USA 71:3537-3541 (1974) Helenius, A. and Simon, K. Biochim. Biophys. Acta 415:29-79 (1975) H i r s t , G.K. J. Expt. Med. 87:301-314 (1948) Juliano, R.L. Biochim. Biophys. Acta 300:341-378 (1973) Jourdian, G.W., Dean, C. and Roseman, S. J. B i o l . Chem. 246:430-435 (1971) Kahane, I., Furthmayr, H. and Masrchesi, V.T. Biochim. Biophys. Acta 426:464-476 (1976) Kokarin, A.I., Zamarayev, K.I., Grigoryan, G.L., Ivanov, V.P. and Rozanysev, E.G. B i o f i z i k a 17:31-42 (31-39 i n the english trans.) (1972) Kokorin, A.I., Zamarayev, K.I., Grigoryan, G.L., Ivanov, V.P. and Rozantsev, E.G. B i o f i z i k a 17:31-42 (31-39 i n the english translation) (1972) Koziarz, J . J . , Kohler, H. and Steck, T.L. J. Supramol. Structure (suppl 2): 215 (1978) Lane, C.F. Synthesis, March:135-146 (1975) Lee, P.M. and Grant, C.W.M. Bioc. Biop. Res. Commun. 90:856-863 (1979) -224-Levine, S., Levine,M., Sharp, K.A. and Brooks, D.E. Biophy. J. 42:127-135 (1983) ~~ Liao, T.H., Gallop, P.M. and Blumenfeld, 0.0. J. B i o l . Chem. 248:8247-8253 (1973) Lovrien, E.R. and Anderson, R.A. J. C e l l B i o l . 85:534-548 (1980) Low, P.S., Cramer, W.A., Abraham, G., Bone, R. and Ferguson-Segall, M. Arch. Bioc. Biop. 214:675-680 (1982) Lowry, O.H., Rosebrouagh, N.J., Farr, A.L. and Randall, R.J. J. B i o l . Chem. 193:265-275 (1951) Marchesi, V.T. and Andrew, E.P. Science 174:1247-1248 (1971) Marchesi, V.T., Furthmayr, H. and Tomita, M Ann. Rev. B i o l . 45:667-698 (1976) ~~ Marcus, D.M., Kundo, S.K. and Suzuki, A. Sem. i n Hemat. 18:63-71 (1981) Massamiri, Y., Durand, G., Richard, A., Feger, J. and Agneray, J. Anal. Bioc. 97:346-351 (1979) Moskowitz, M. and Treffers, H.P. Science 111:717-719 (1950) Mueller, T.J., Dow, A.W. and Morrison, M. Bioc. Biop. Res. Commun. 72:94-98 (1976) — Mueller, T.J. and Morrison, M. i n "Erythrocyte Membranes 2: Recent C l i n i c a l and Experimental Advances" 95-112 Liss In. N.Y. (1981) Nash, T. Bioc. J. 55:416-421 (1953) Pearse, A.G.E. i n 'Histochemistry, Theoretical and Applied' p. 460-467 Churchill Ltd Londaon (1968) Peterson, G.L. Anal. Bioc. 83:346-356 (1977) Presant, CA. and Parker, S. J. B i o l . Chem. 251:1846-1870 (1976) Rando, R.R. and Bangerton, F.W. Biochim. Biophy. AScta 557:354-362 (1979) Ravazzolo, R., Garre, C and Ajmar, F. Scan. J. C l i n . Lab Invest. 43:123-126 (1983) Reid, P.E., C u l l i n g , C.F.A., Ramey, C.W., Dunn, W.L. and Clay, M.G. Can. J. Bioc. 55:493-503 (1977) Ross, T.E., Campbell, CD. and Sharom, F.J. Can. Fed. B i o l . Soc. 26:241 (1983) -225-Rotman, A., Linder, S. and Pribluda, V. FEBS Letters 120:85-89 (1980) Ruppel, D., Kapitza, H.G., Galla, H.J., S i x l , F. and Sackmann, E. Biochim. Biophys. Acta 69:17-21 (1982) Saito, T. and Hakomori, S. J. Lip i d Res. 12:257-259 (1971) Schweizer, E., Angst, W. and Lutz, H.U. Biochemistry 21:6807-6818 (1982) Silverberg, M., Furthmayr, H. and Marchesi, V.T. Biochemistry 15:1448-1454 (1976) Singer, J.A. and Morrison, M. Biochim. Biophy. Acta 343:598-608 (1974) Skalak, R., Zarda, P.R., Jan, K.M. and Chien, S. Biophy. J. 35:771-781 (1981) Skutelsky, K., Danon, D., Wilchek, M. and Bayer, E.A. J. U l t r a s t r c t r . Res. 61:325-335 (1977) Sogin, D.C. and Hinkle, D.C. J. Supramol. Struct. 8:447-453 (1978) Springer, G.F. Bacteriological Rev. 27:191-227 (1963) Springer, G.F. and Yang, H.J. Vox Sang 35:255-264 (1978) Springer, G.F., Nagai, Y. and Tegtmayer, H. Biochemistry 5_:3254-3271 (1966) Somogyi, M. J. B i o l . Chem. 160:69-73 (1945) Spiegel, S. and Wilchek, M. Molecular and Ce l l u l a r Bioc. 55:183-190 (1983) Spiro, R.G. J. B i o l . Chem. 239:567-573 (1964) ( de St. Groth G.F. Australian J. Exp. B i o l . Med. S c i . 27:65-81 (1949) Steck, T.L. and Yu, J. J. Supramol. Structure 1_:220-232 (1973) Steck, T.L. and Dawson, G. J. B i o l . Chem. 249:2135-2142 (1974) Steiner, B., Clemetson, K.J. and Luscher, E.F. Thrombosis Res. 29:43-52 (1983) Stewart, F.S. J. Path. Bact. 61:456-458 (1949) Stibenz, D. and Geyer, G. F o l i a Haematol. 107:787-792 (1980) S u t t a j i t , M. and Winzler, R.J. J. B i o l . Chem. 246:3398-3404 (1971) Tanner, M.J.A. Current Topics i n Membranes and Transport 11:279-325 (1978) -226-Taylor, K.E. and Wu, Y.C. Bioc. International 1_: 353-358 (1980) Van Lenten, L. and Ashwell, G. J. B i o l . Chem. 246:1880-1804 (1971) Vassar, P.S., Hards, J.M., Brooks, D.E., Hagenberger, B. and Seaman, G.V.F. J. C e l l B i o l . 53:809-818 (1972) Wherrett, J.R. Biochim. Biophys. Acta 326:63-73 (1973) Winkelhake, J.L., Kusumi, A., Mckean, L. and Mandy, W.J. J. B i o l . Chem. 259:2171-2178 (1984) Yachnin, S. and Gardner, F.H. Blood 18:349-363 (1961) Yalpani, M. Ph. D. Thesis, University of B r i t i s h Columbia (1980) Yu, J., Fischman, D.A. and Steck, T.L. J. Supromol. Structure 1:233-240 (1973) Zwaal, R.F.A., Roelofren, B. and Colley, CM. Biochim. Biophys. Acta 30:159-182 (1973) -227-Chapter 4 Adair, W.L. and Kornfeld, S. J. B i o l . Chem. 249:4696-4704 (1974) Al l e n , A.K., Neuberger, A. and Sharon, N. Bioc. J. 131:155-162 (1973) Anderson, R.A. and Lovrien, R. i n "Erythrocyte Membranes 2: Recent C l i n i c a l and Experimental Advances" 207-226 Alan Liss In. N.Y. (1981) Anstee, D.J. Sem i n Hematology 18:13-31 (1981) Aub, J.C., Tieslau, C. and Lankester, A. Proc. Nat. Acad. S c i . USA 50: 613-619 (1963) Bhavanandan, V.P. and K a t l i c , A.W. J. B i o l . Chem. 254:4000-4008 (1979) Bjerrum, O.J., Ramulau, J . , Bock, E. and Bog-Hansen, T.C. Receptors and Recognition B v o l . 11:115-156 (1981) Blumenfeld, 0.0., Gallop, P.M. and Liao, T.H. Biochem. Biophys. Res. Comm. 48:242-251 (1972) Boldt, D.H., Speckert, S.F., Richards, R.L. and Alaving, CR. Bioc. Biop. Res. Commun. 74:208-214 (1977) Bowles, D.J. and Hanke, D.E. FEBS Letters 82:34-38 (1977) Brooks, D.E. J. C o l l . S c i . 43:687-726 (1973) Burger, M.M. and Goldberg, A.R. Proc. Nat. Acad. S c i . USA 57:359-366 (1967) Burness, A.T.H. and Pardoe, I.U. J. Chromatography 259:423-432 (1983) Choy, Y.M., Wong, S.L. and Lee, CY. Bioc. Biop. Res. Commun. 91:410-415 (1979) Clegg, R.M., Loontiens, F.C, Sharon, N. and Jovin, T.M. Biochemistry 22:4797-4804 (1983) Cuatrecasas, P. Biochemistry 12:1312-1323 (1973) Dahlquist, F.W. Methods i n Enzymol. 40:270-299 (1978) Evans, E. and Leung, A. J. C e l l B i o l . 98:1201-1208 (1984) F e l i x , J.B. and B u t t e r f i e l d , D.A. FEBS Letters 115:185-188 (1980) F e l i x , J.B., Green, L.L. and B u t t e r f i e l d , D.'A. L i f e Sciences 31:1001-1009 (1982) Froman, G., Lundahl, P. and Acevedo, F. FEBS Letters 129:100-104 (1981) . -228-Fukuda, M. and Osawa, T. J. B i o l . Chem. 248:5100-5105 (1973) Furthmayr, H., Kahane, I. and Marchesi, V.T. J. Membr. B i o l 26:173-187 (1976) — Furthmayr, H. i n 'Biology of Carbohydrates,' v. 1, p. 123-198, Ginsberg, V. & Robbins, P. (eds). Wiley & Sons, N.Y. (1981) Gordon, J.A., Staehelin, L.A. and Kuettner, CA. Exp. C e l l Res. 110:439-448 (1977) Grivet, J.-Ph., Midoux, P., G a t t e l l i e r , P., Delmotte, F. and Monsigny, M. i n "Structure, Dynamics, Interactions and Evolution of B i o l o g i c a l Macromolecules" 329-349, (Helene, C. ed. ) Reidel (1983) Horisberg, M. and Rosset, J. J. Histochem. Cytochem. 25:295-305 (1977) Jokobivits, A., Eshdat, Y. and Sharon, N. Bioc. Bioph. Res. Comm. 100:1484-1490 (1981) Jordan, F., Bassett, E. and Redwook, W.R. Bioc. Biop. Res. Commun. 75:1015-1021 (1977) Jordan, F., Bahr, H., Patrick, J. and Woo, P.W.K. Arch. Bioc. Biop. 207:81-86 (1981) Kahane, I., Furthmayr, H. and Marchesi, V.T. Biochim. Biophy. Acta 426:464-476 (1976) Ketis, N.V., Girdlestone, J. and Grant, C.M.W. Proc. Nat. Acad. S c i . USA 77:3786-3790 (1980) Ketis, N.V. and Grant, C.W.M. , Biochim. Biophy. Acta 685:347-354 (1982) Kohn, J. and Wilchek, M. Bioc. Biop. Res. Commun. 84:7-14 (1978) Kronis, K.A. and Carver, J.P. Biochemistry 21:3050-3057 (1982) Kronis, K.A. and Carver, J.P. Biochemistry i n press (1984 a,b,c) Kwok, B.C.P. and Landsberger, F.R. Biophy. J. 37:144a (1982) Lee, P.M. and Grant, C.W.M. Bioc. Biop. Res. Commun. 90:856-863 (1979) Lee, P.M. and Grant, C.W.M. Can. J. Bioc. 58:1197-1205 (1980) Lee, P.M., Ketis, N.Y., Barber, K.R. and Grant, C.W.M. Biochim. Biophys. Acta 601:302-314 (1980) Lovrien, R.E. and Anderson, R.A. J. C e l l B i o l . 85:527-541 (1980) -229-Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. J. B i o l . Chem. 193:265-275 (1951) March, S.C., Parikh, I. and Cuatrecasas, P. Anal. Bioc. 60:149-152 (1974) Midoux, P., Grivet, J.-Ph. and Monsigny, M. FEBS Letters 120:29-32 (1980) Monsigny, M., Sene, C., Obrenovitch, A., Roche, A.C., Delmotte, F. and Boschetti, E. Eur. J. Bioc. 98:39-45 (1979) Monsigny, M., Roche, A., Sene, C , Maget-Dana, R. and Delmotte, F. Eur. J. Bioc. 104:147-153 (1980) Nagata, Y. and Burger, M.M. J. B i o l . Chem. 249:3116-3122 (1974) Neurohr, K.J., Lacelle, N., Mantsch, H.H. and Smith, I.CP. Biophys. J. 32:931-938 (1980) Peters, B.P., Ebisu, S., Goldstein, I . J . and Flashner, M. Biochemistry 18:5505-5511 (1979) Peterson, G.L. Anal. Bioc. 83:346-356 (1977) Ravazzola, R., Garre, C. and Ajmar, F. Scand. J. C l i n . Lab Invest. 43:123-126 (1983) Redwood, W.R., Jansons, V.K. and Patel, B.C. Biochim. Biophy. Acta 406:347-361 (1975) Redwood, W.R. and Polefka, T.G. Biochim. Biophy. Acta 455:631-643 (1976) Regoeczi. E. 'Iodine-labeled Plasma Proteins' V. 1 CRC Press Inc. Florida (1984) Rendi, R., Kuettner, CA. and Gordon, J.A. Bioc. Biop. Res. Commun. 72:1071-1076 (1976) Rey, P. and McConnell, H. Bioc. Biop. Res. Commun. 73:248-254 (1976) Rice, R.H. and Et z l e r , M.E. Biochemistry 14:4093-4099 (1975) Ross, T.E., Campbell, CD. and Sharom, F.J. Can. Fed. B i o l . Soc. 26:241 (1983) Scatchard, G. Ann. N.Y. Acad. S c i . 51:660-672 (1949) Schnebli, H.P. and Bachi, T. Exp. C e l l Res. 91:175-183 (1975) Seigneuret, M,. Zachocoski, A., Hermann, A. and Devaux, P.E. Biochemistry 23:4271-4275 (1984) -230-Sever, J.L. J. Immunol. 88:320-329 (1962) Singer, J.A. and Morrison, M. Biochim. Biophy. Acta 343:598-608 (1974) Smith, L. and Hochmuth, R.M. J. C e l l B i o l . 94:7-11 (1982) Stanley, P. and Carver, J.P. Proc. Nat. Acad. S c i . USA 74:5056-5059 (1977) Steck, T.L. Meth. Memb. B i o l . 2:245-281 (1974) Tanner, M.J.A. and Anstee, D.J. Bioc. J. 153:265-270 (1976) Thomas, M.W., Walbarg J r . , E.F. and Jirgensons, B. Arch. Bioc. Biop. 178:625-630 (1977) Triche, T.J., T i l l a c k , T.W. and Kornfeld, S. Biochim. Bioph. Acta 394:540-549 (1975) Tsao, D. and Kim, Y.S. J. B i o l . Chem. 256:4947-4950 (1981) Turner, R.H. and Liener, I.E. Anal. Bioc. 68:651-653 (1975) Van der Steen, A.T.M., Taraschi, T.F., Voorhout, W.F. and de K r u i j f f , B. Biochim. biophy. Acta 733:51-64 (1983) Vassar, P.S., Hard, J.M., Brooks, D.E., Hagengerger, B. and Seaman, G.V.F. J. C e l l B i o l . 53:809-818 (1972) Wright, C.S., Keith, C , Langride, R., Nagata, Y. and Burger, M.M. J. Mol. B i o l . 87:843-846 (1974) Wright, C.S. J. Mol. B i o l . 111:439-457 (1977) Wright, C.S. J. Mol. B i o l . 132:53-60 (1979) Wright, C.S. J. Mol. B i o l . 141:267-291 (1980) Wright, C.S. J. Mol. B i o l . 145:453-561 (1981) Wright, C.S., Gavilanes, F. and Peterson, D.L. Biochemistry 23:280-287 (1984) -231-Appendlx A Aloyo, V.J. Anal. Bioc. 99:161-178 (1979) Laemmli, U.K. Nature 227:680-685 (1970) Appendix C All e n , A.K., Neuberger, A. and Sharon, N. Bioc. J . 131:155-162 (1973) Anstee, D. Sem. i n Hematology 18:13-31 (1981) Aplin, J. A., Berstein, M.A., C u l l i n g , C.F.A., H a l l , L.D. and Reid, P.E. Carb. Res. 70:c9-cl2 (1979) Baenziger, J.U. and Fiete, D. J. B i o l . Chem. 254:789-795 (1979) Bassett, E.W. Preparative Bioc. 5:461-477 (1975) Bhavanadan, V.P. and K a t l i c , A.W. J. B i o l . Chem. 254:4000-4008 (1979) Bloch, R. and Burger, M.M. Bioc. Biop. Res. Commun. 58:13-19 (1974) Bouchard, P., Moroux, Y., T i x i e r , R., Priv a t , J.-P. and Monsigny, M. Biochimie 59:1247-1253 (1976) Brown, J.M., Leon, M.A. and Lightbody, J.J. J. Immunology 117:1976-1980 (1976) Cuatrecasas, P. J. B i o l . Chem. 245:3059-3065 (1970) Dulaney, J.T. Molecular and C e l l u l a r Bioc. 21:43-63 (1979) Erni, B., De Boeck, H., Loontiens, F.G. and Sharon, N. FEBS Letters 120:149-154 (1980) Furthmayr, H. in'Biology of carbohydrates V. 1' p.123-198, Ginsburg, V. and Robbins, P (eds). Wiley and Sons, N.Y. Kronis, K.A. and Carver, J.P. Biochemistry 21:3050-3057 (1982) LaCelle, N. Ph. D. thesis, University of Toronto (1979) LeVine, D., Kaplan, M.J. and Greenaway, P.J. Bioc. J . 129:847-856 (1972) Lisowska, E., Duk, M. and Dahr, W. Carb. Res. 79:103-113 (1980) Lotan, R., Gussin, A.E.S., L i s , H. and Sharon, N. Bioc. Biop. Res. Commun. 52:656-662 (1973) -232-Lotan, R., Beattie, G., Hubbell, W. and Nicolson, G.L. Biochemistry 16:1787-1794 (1977) Lovrien, R.E. and Anderson, R.A. J. C e l l B i o l . 85:527-541 (1980) Marchesi, V.T. Methods i n Enzymology 28B:354-356 (1972) Nagata, Y. and Burger, M.M. J. B i o l . Chem. 249:3116-3122 (1974) Nilsson, B., Norden, N.E and Svensson,'S. J. B i o l . Chem. 254:4545-4553 (1979) Rice, R.H. and Etzler, M.E. Biochemistry 14:4093-4099 (1975) Rice, R.H. Biochim. Biophy. Acta 444:175-180 (1976) ' Shaper, J.H., Barker, R. and H i l l , R.L. Anal. Bioc. 53:564-570 (1973) Wright, C.S. J. Mol. B i o l . 145:453-561 (1981) Wright, C.S., Gavilanes, F. and Peterson, D.L. Biochemistry 23:280-287 (1984) Yamamoto, K., T s u i j i , T., Matsumoto, I, and Osawa, T. Biochemistry 20:5894-5899 (1981) 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0060492/manifest

Comment

Related Items