UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Functional roles of phospholipids in exocytosis Nayar, Rajiv 1983

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-UBC_1984_A1 N39.pdf [ 8.54MB ]
Metadata
JSON: 831-1.0096591.json
JSON-LD: 831-1.0096591-ld.json
RDF/XML (Pretty): 831-1.0096591-rdf.xml
RDF/JSON: 831-1.0096591-rdf.json
Turtle: 831-1.0096591-turtle.txt
N-Triples: 831-1.0096591-rdf-ntriples.txt
Original Record: 831-1.0096591-source.json
Full Text
831-1.0096591-fulltext.txt
Citation
831-1.0096591.ris

Full Text

F U N C T I O N A L R O L E S O F P H O S P H O L I P I D S I N E X O C Y T O S I S by R A J I V N A Y A R B.Sc, The Univers i ty O f Br i t i sh Co lumb ia , 1979 A T H E S I S S U B M I T T E D I N P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F D O C T O R O F P H I L O S O P H Y in T H E F A C U L T Y O F G R A D U A T E S T U D I E S Biochemistry Depar tment W e accept this thesis as conforming to the requi red standard T H E - U N I V E R S I T Y O F B R I T I S H C O L U M B I A November 1983 © Ra j i v Nayar , 1983 In p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t o f the requ i rements f o r an advanced degree a t the U n i v e r s i t y o f B r i t i s h Co lumb ia , I agree t h a t the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and s tudy . I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e copy ing o f t h i s t h e s i s f o r s c h o l a r l y purposes may be g ran ted by the head o f my department o r by h i s o r her r e p r e s e n t a t i v e s . I t i s unders tood t h a t copy ing o r p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l ga i n s h a l l not be a l l owed w i thou t my w r i t t e n p e r m i s s i o n . Department o f %\0 CVsvHUT^S The U n i v e r s i t y o f B r i t i s h Columbia 1956 Main Mall Vancouver , Canada V6T 1Y3 Date Uo* 11. H * 3 •E-6 (3/81) 11 Abstract Cell secretion by exocytosis involves fusion between secretory granules and the surrounding plasma membrane, which results in the extracellular release of the granule contents. This event is triggered by the influx of Ca3* on stimulation, which is accompanied by an increased conversion of the lipid phosphatidylinositol (PI) to diacylglycerol, to phosphatide acid (PA) and then back to PI. This work is concerned with the roles of phospholipids in exocytosis, both with regard to the Ca2"-stimulated fusion event itself and the roles of PI and intermediates in the PI cycle. Three topics are considered in detail, namely the physical properties of PI in pure and mixed lipid systems, the possible ability of PA to act as a Ca2* ionophore and the influence of the plasma membrane inner monolayer lipid composition on the fusion event vital to the completion of exocytosis. It is shown by, employing 3 ] P - N M R and freeze-fracture techniques, that PI adopts the bilayer organization in isolation and is particularly effective for stabilizing this organization when mixed with "non-bilayer" lipids both in the absence and presence of Ca 2*. Alternatively, PA in large unilamellar vesicles exhibits properties consistent with Ca 2* ionophore capabilities. Finally, vesicles composed of phosphatidylethanolamine and phosphatidylserine (which possibly mimics the plasma membrane inner monolayer composition) can act as adjuncts to Ca2*-stimulated release of chromaffin granule contents. Theses results are consistent with the possibility that intermediates of the PI-response (e.g. PA) can enhance Ca2* influx, and indicate that the plasma membrane inner monolayer lipid composition of the secretory cell may play a vital role in the secretory event. i i i Tab le o f Contents Abstract i i L is t o f Tables v i L i s t o f F igures v i i Abbrev iat ions ix Acknowledgements x I N T R O D U C T I O N 1.1 De f in i t i on O f Exocytosis . .2 1.2 Def in i t i on O f St imulus-Secret ion Coup l i ng 2 1.3 Structure O f Membranes 4 1.4 Phys ica l Propert ies O f L ip ids 5 1.4.1 G e l - L i q u i d Crysta l l ine Phase Trans i t ion 5 1.4.2 F lu id i ty O f Membranes 6 1.4.3 Po lymorph ic Phase Transit ions 7 1.5 Asymmetr i c Nature O f B io log ica l Membranes 8 1.6 Use O f 3 1 P - N M R In Determin ing Membrane Structure . . .9 1.7 F reeze-F rac tu re E lectron Mic roscopy 12 1.8 Ca l c i um Transport A n d Regulat ion O f Secretion 15 1.8.1 Hypothes is F o r Ca l c i um Mob i l i z a t i on 16 1.9 The Phosphat idy l inos i to l -E f fec t 19 1.10 Phospha t ide A c i d ( PA ) A n d Ca l c i um Mob i l i z a t i on . . .22 1.11 De f in i t i on O f Membrane Fus ion 23 1.11.1 Lyso lec i th in A n d Membrane Fus ion 24 1.11.2 Ro l e O f Ca l c i um 25 1.11.3 Crysta l l ine M o d e l O f Fus i on 25 1.11.4 Po lymorph i c M o d e l O f Fus i on 26 1.12 The Ch roma f f i n Granu le : Structure A n d Func t ion . . .27 1.12.1 Histor ica l Perspective 28 1.12.2 Compos i t ion O f The Ch roma f f i n Granu le 28 1.12.3 Ro les O f The Ch roma f f i n G ranu le 31 1.12.4 Synex in A n d Exocytosis 34 1.13 Out l ine O f Th i s Thesis 34 II. S T R U C T U R A L P R E F E R E N C E S O F PI A N D P I - P E M O D E L M E M B R A N E S 36 2.1 I N T R O D U C T I O N 36 2.2 M A T E R I A L S A N D M E T H O D S 37 2.2.1 Pur i f icat ion O f Soyabean Phosphat idy l inos i to l 37 2.2.2 Convers ion T o The Sod ium Salt O f Phosphat idy l inos i to l 37 2.2.3 Isolation A n d Pur i f icat ion O f Phosphat idy lchol ine F r o m Egg Y o l k A n d Soyabeans 38 2.2.4 Preparat ion O f Phosphol ipase D 39 2.2.5 Preparat ion O f Phosphat idylser ine By Base-Exchange React ion 40 2.2.6 Preparat ion O f Soya Phosphat idy lethanolamine 41 2.2.7 Pur i f icat ion O f Egg A n d Soyabean Phosphat idy lethanolamine 41 2.2.8 Th i n Layer Chromatography 42 2.2.9 M i c r o - s l i d e T L C 43 2.2.10 Two -D imen s i o na l T h i n - L a y e r Chromatography ( 2 D - T L C ) iv 43 2.2.11 Phosphorus Spray Reagents F o r Identif icat ion O f Phosphol ip ids 44 2.2.12 Phosphorus Assay 44 2.2.13 F a t t y - A c i d Analys is 45 2.2.14 Preparat ion O f 3 1 P - N M R Samples A n d Exper imenta l Setup 46 2.2.15 B i layer T o Hexagona l H u Phase Trans i t ion Temperature 46 2.2.16 D iva len t Cat ion B ind ing Determinat ions 49 2.2.17 F reeze-F rac tu re Studies 50 2.3 R E S U L T S 50 2.3.1 Phase Preferences O f Phosphat idy l inos i to l . . .50 2.3.2 B i layer Stabi l izat ion Effect O f Phosphat idy l inos i to l 53 2.3.3 Character izat ion O f The "Isotropic" Components O f P I - P E M ix tu res 54 2.3.4 Influence O f Ca l c i um 59 2.3.5 Influence O f Magnes ium 62 2.4 D I S C U S S I O N 67 III. R O L E O F P H O S P H A T I D E A C I D ( P A ) A S A C A L C I U M I O N O P H O R E I N L A R G E U N I L A M E L L A R V E S I C L E S ( L U V S ) 70 3.1 I N T R O D U C T I O N 70 3.2 M A T E R I A L S A N D M E T H O D S 72 3.2.1 Synthesis O f D io l eoy l Phosphat idy lcho l ine ( D O P C ) 72 3.2.2 Synthesis O f D io l eoy l Phosphat id ic A c i d ( D O P A ) 75 3.2.3 Preparat ion O f Large Un i l ame l l a r Vesic les ( L U V s ) 75 3.2.4 Deterrmnat ipn O f Trapped Vo lumes 77 3.2.5 Determinat ion O f Leakage 78 3.2.6 F reeze-F rac tu re Studies 79 3.2.7 Ca l c i um Uptake Exper iments 82 3.3 R E S U L T S 85 3.3.1 Ionophoret ic Capabi l i t ies O f P A In L U V s 85 3.3.2 Levels O f P A Requ i red Fo r C a 2 + - u p t a k e 88 3.3.3 Effectiveness O f The Ca 2 * - s i n k 93 3.3.4 Effect O f P H O n C a 2 + - u p t a k e 98 3.3.5 C a 2 + - u p t a k e In D O P C A n d 20% PS L U V E T s + A23187 99 3.3.6 Influence O f A Membrane Potent ia l 104 3.3.7 Influence O f Proton Ionophore ( C C C P ) O n C a 2 + - u p t a k e 107 3.4 D I S C U S S I O N 110 IV. P H O S P H O L I P I D S A S A D J U N C T S F O R C A 1 * - S T I M U L A T E D R E L E A S E O F C H R O M A F F I N G R A N U L E C O N T E N T S 113 4.1 I N T R O D U C T I O N 113 4.2 M A T E R I A L S A N D M E T H O D S 114 4.2.1 Isolation O f Ch roma f f i n Granu les 114 4.2.2 Isolation O f The Ch roma f f i n G ranu l e Membranes 118 V 4.2.3 Determinat ion O f M i tochondr ia l Contaminat ion In The Large Granu le Frac t ion 118 4.2.4 Lowry Prote in Assay 119 4.2.5 L i p i d Extract ion F r o m Ch romaf f i n G ranu le Membranes 119 4.2.6 Isolation O f Erythrocyte Membrane Phosphol ip ids 120 4.2.7 Isolation A n d Pur i f i cat ion O f L ip ids 120 4.2.8 Preparat ion O f Mode l L i p i d Systems 121 4.2.9 Spectrophotometr ic Release Assay 121 4.2.10 Assay ing F o r The Release Products O f The Ch roma f f i n Granu les ". 122 4.2.11 F reeze-F rac tu re O f Ch roma f f i n Granu les A n d L i p i d Vesicles 125 4.3 R E S U L T S 125 4.3.1 Influence O f Exogenous L i p i d O n Ch roma f f i n G ranu le Release 125 4.3.2 Effect O f Ca l c i um Concentrat ion 136 4.3.3 F reeze -F rac tu re Studies 136 4.4 D I S C U S S I O N 144 V . D I S C U S S I O N 147 5.1 Speculated Mechan i sm Fo r C a 2 + Transport 149 5.2 Speculated Mechan i sm F o r Exocytosis 149 B I B L I O G R A P H Y 154 vi List of—Tables I. Compos i t ion o f the Bov ine Adrena l Ch roma f f i n Granu le 29 II. Phospho l ip id Compos i t ion o f the Ch roma f f i n Granu le Membrane 30 v i i L is t o f F igures 1. 3 1 P - N M R Spectra of Phospho l ip id Phases 10 2. F reeze -F rac tu re Nomenc lature 13 3. F reeze -F rac tu re Characterist ics o f Phosphol ip ids i n Var ious Phases 14 4. Recep to r -Con t ro l l ed Mob i l i z a t i on o f Ca l c i um 21 5. Membrane Fus ion 23 6. Subcel lu lar Dynamics o f the Ch roma f f i n C e l l 33 7. B i l ayer to Hexagonal H u Trans i t ion o f Soya P E 47 8. B i layer to Hexagonal H n Trans i t ion o f Egg P E 48 9. 3 1 P - N M R Spectra o f Soya PI 52 10. 3 1 P - N M R Spectra o f P I - P E mixtures 56 11. F reeze -F rac tu re micrographs o f P I - P E mixtures 58 12. Inf luence o f Ca 2 * on P I - P E mixtures 61 13. Inf luence o f M g 2 * on P I - P E mixtures 64 14. D iva len t cation b inding curves o f PI and PS vesicles . . 66 15. React ion scheme for synthesis o f D O P C 74 16. F r ee ze - fracture micrographs o f 20 mo l% D O P A L U V E T s . . . . 81 17. Ca 2 * -up take in 20 mol% D O P A L U V E T s 84 18. Ca 2 * -up take in 20 mol% D O P A and 20 mo l% PS L U V E T s 87 19. Ca 2 * -up take in 20 mol% D O P A containing varying amounts o f trapped phosphate 90 20. Ca 2 * -up take in L U V E T s contain ing different concentrations o f D O P A 92 21. Ca 2 * -up take i n 20 mol% D O P A L U V E T s with trapped oxalate 95 22. Ca 2 * -up take i n 20 mol% D O P A L U V E T s wi th trapped E G T A . . 97 23. Ef fect o f p H on Ca 2 *-up take 101 24. Ca 2 * -up t ake in D O P C ± A23187 L U V E T s 103 25. Ef fect o f Va l i nomyc in and C C C P on Ca 2 * -up take in 20 mo l% PS ± A23187 and in 20 mo l% D O P A L U V E T s 106 26. Ef fect o f Va l i nomyc in and C C C P on Ca 2 * -up take i n D O P C + A23187 L U V E T s 109 27. Summary o f Isolation Procedure for Ch roma f f i n Granu les o f Bov ine Adrena l Medu l l a 117 28. Cor re la t ion o f the amount o f release o f chromaf f in granule contents i n the presence o f exogenous P E - P S (3:1) vesicles 124 29. Release o f chromaff in granule contents after incubat ion in the presence o f Ca 2 * and increasing amounts o f P E - P S vesicles 127 30. Ca 2 *-s t imu la ted release o f chromaf f in granule contents after incubat ion wi th vesicles o f various composit ions 131 31. Release o f chromaf f in granule contents in the presence o f soya and egg P E -PS (3:1) vesicles 133 32. Release o f chromaf f in granule contents after incubat ion wi th vesicles composed o f erythrocyte l ip ids 135 33. Ef fect o f Ca 2 * concentration on release o f chromaf f in granule contents 138 34. F reeze- f rac ture micrographs o f chromaf f in granules and P E - P S (3:1) vesicles in the presence and absence o f 2 m M and 5 m M Ca 2 * . . . . 141 35. Freeze- f rac ture micrographs o f chromaff in granules incubated with P E - P S (3:1) vesicles in 2 m M and 5 m M C a 2 + 143 36. Proposed mechanism for faci l itated transport o f Ca 2 * .150 37. Proposed mechanism o f exocytotic release o f chromaf f in granule contents j n v ivo ix ABBREVIATIONS A23187 Calcium ionophore ATP Adenosine 5'-triphosphate ADP Adenosine 5'-diphosphate A1 20 3 Aluminum oxide BSA Bovine serum albumin CSA Chemical shift anisotropy CCCP Carbonyl cyanide- m-chlorophenyl hydrazine CHES 2-(N-Cyclohexylamino)ethanesulfonic acid CMP Cyudine-5'-monophosphate D G Diacylglycerol DMSO Dimethylsulfoxide EDTA Ethylenediaminoethantetraacetic acid E G T A Ethyleneglycol-bis-(p-aminoethyl ether)-N,N,N',N'-tetraacetic acid ^°c(f Effective chemical shift anisotropy g Centrifugal force HCI Hydrochloric acid HEPES 4-(2-Hydroxyethyl)l-piperazine ethanesulfonic acid 5-HT 5-Hydroxytryptamine Kr-phosphate Potassium phosphate K4-oxalate Potassium oxalate LUVs Large unilamellar vesicles M D H Malate dehydrogenase MES (2-[N-Morpholino] ethanesulfonic acid) MLVs Multilamellar vesicles X Mo0 3 Molybdic anhydride NAD 4 Nicotinamide adenine dinucleotide, oxidized NADH Nicotinamide adenine dinucleotide, reduced NaCl Sodium chloride NaOH Sodium hydroxide n P - N M R Phosphorus nuclear magnetic resonance Phospholipids: CL Cardiolipin DOPA Dioleoylphosphatidic acid DOPC Dioleoylphosphaudylcholine GPC L- a -Glyceryl phosphorylcholine PA Phosphatide acid PE Phosphatidylethanolamine PG Phosphatidylglycerol PS Phosphatidylserine PI Phosphatidylinositol Sph Sphingomyelin Pt/C Platinum-carbon Pi Phosphate SDS Sodium dodecyl sulfate SUVs Small unilamellar vesicles T c Gel-liquid transition temperature THF Tetrahydrofuran TLC Thin layer chromatography Aknowledgement I am indebted to Dr. P.R.Cullis for his continual support and enthusiam in this project, and also for critically editing this thesis. I would also like to acknowledge the contributions that my colleagues, particularly Dr. M.J. Hope and Dr. L. Mayer, have made in the experiments described in this thesis. Without their helpful discussions and helpful discussions and assistance, this work would not have been possible. The liberal use of Drs. Ross MacGillivray and Caroline Astell's computer terminals was greatly appreciated. Finally, I would like to acknowledge the University for providing finacial assistance during this period. x i i To my parents 1 C H A P T E R I I N T R O D U C T I O N Considerable progress in our understanding o f b iomembrane structure and funct ion has been made in the last fifty years. The importance o f membranes to bio logical processes i n an imal and plant cells and in microorganisms is indicated by the enormous diversity o f membrane-med ia ted functions, inc luding intracel lu lar compartrnental izat ion by various organelles, faci l i tated transport, cel l fusion, endo -and exocytosis, and ce l l - ce l l interactions among many others. The purpose o f this thesis is to investigate the molecular mechanisms involved in exocytosis. In particular, this thesis addresses the possibi l i ty that l ip ids might p lay funct ional roles dur ing the exocytotic process. These include direct roles in i on transport and membrane fus ion i n addit ion to a structural role o f maintain ing a permeabi l i ty barr ier between external and internal environments. Th i s chapter br ief ly reviews exocytosis wh ich is an "umbre l l a " term for distinct but related membrane-med ia ted events that eventual ly result i n secretion o f cel lu lar products into the extracellular med ium. The contents o f this chapter are d iv ided into the fo l lowing order. First, exocytosis and its synonym, "st imulus-secret ion coup l ing" are def ined. Second, some general characteristics o f membranes and the two techniques used i n this thesis to study membrane structure ( " P - N M R and freeze-fracture) are br ief ly reviewed. Th i rd , the specif ic events invo lved in exocytosis are discussed. These are (1) regulat ion o f ca lc ium transport, (2) the PI- response, (3) P A and ca lc ium mobi l i zat ion and (4) membrane fusion and proposed models for fusion. Four th , the biological fusion system employed to study the fus ion event between the secretory granule and the plasma membrane is reviewed. Th i s is the 2 chromaf f in granule system isolated f rom the adrenal medul la. F ina l l y , the approach taken to elucidate the functional roles o f phosphol ip ids in exocytosis is summar ized. 1.1 De f in i t i on O f Exocytosis Exocytosis involves fusion between the secretory vesicles o f a cel l and the cel l p lasma membrane. Th is fusion process results in the release o f the vesicle contents into the extracel lular space. Exocytosis is basic to processes o f cel l secretion and is invo lved i n the release o f a wide variety o f enzymes, hormones, and neurotransmitters f rom cells such as the newly fert i l ized egg, b lood platelets, leukocytes, mast cells, nerve cells, cells part ic ipat ing in the format ion o f k in ins, angiotensin, and erythropoiet in, and ho rmone-p roduc ing cells in the adrenal medul la , neurohypophysis, anterior pituitary, thyroid, and pancreas (reviews; Ceccare l l i et a l , 1974; Douglas, 1975; Cara fo l i et a l . , 1975). The local izat ion o f the secretory product i n membrane -bound vesicles and their export f rom the cel l by exocytosis offers several benefits. The products are protected against degradation f rom the cytoplasmic enzymes and can therefore be transported over fa i r ly long distances, as i n the case o f nerve axons. Fur thermore, the products can be released i n quantal amounts in response to a physio logica l st imulus 1.2 S t imu lus-Secre t ion Coup l i ng Doug las and R u b i n (1961) co ined the phrase "st imulus-secret ion coup l ing" wh ich includes a l l the events subsequent to st imulat ion that eventually lead to the release o f the secretory product into the extracel lular env i ronment Fo r nerves, the st imulus is an electrical depolar izat ion o f the presynaptic terminal (Katz , 1969) whereas, for hormone secretion, the st imulus is usual ly a compound which induces 3 membrane depolarizat ion. The currently accepted v iew is that the st imulus causes a redistr ibut ion or "mob i l i za t ion" o f a second messenger, calc ium, f rom various locations to regions where it can initiate secretion. The subsequent secretion either appears to have considerable structural order or be a rather amorphous process (Si l insky, 1982). The ordered type o f secretion is referred to as exocytosis. St imulus o f the cel l membrane (depolarizat ion) is suggested to open specif ic membrane channels for ca lc ium, a l lowing d i f fus ion o f ca lc ium into the cel l . Once in the cytoplasm, ca lc ium induces fus ion between the secretory granule and the plasma membrane result ing i n the discharge o f the secretory p roduc t The second "amorphous" type o f secretory process occurs when st imulus o f the plasma membrane redistributes intracel lular ca lc ium and the result ing increase i n cytoplasmic ca lc ium levels results in synthesis and secretion o f the secretory p roduc t The entry o f extracel lular ca lc ium does not appear to be necessary, nor does the secretory product need to be packaged i n a vesicle. F o r example, hormona l st imulat ion can mobi l i ze membrane ca lc ium and increase the synthesis o f cycl ic nucleotides and prostaglandins. These modulators i n turn initiate secretion either directly or indirect ly by l iberat ing ca lc ium f rom intracel lular organelles (e.g. mitochondr ia) into the cytoplasm (Carafo l i and Crampton , 1978). Examples o f this amorphous secretory process include st imulat ion o f adrenocort icotropin for cort icosteroid secretion (Rub i n and Laychock, 1978) and st imulat ion by t hy ro i d -st imulat ing hormone for thyro id hormone secretion (Wi l l i ams, 1972). Th is thesis is concerned wi th the ordered exocytotic process. Th is is clearly a membrane-med ia ted event and it is therefore appropriate to br ief ly review our understanding o f b iomembrane systems with part icular emphasis on the l i p id component 1.3 Structure O f Membranes Membranes are composed predominant ly o f l ip ids and proteins. The l ip ids are thought to prov ide the fundamental bu i ld ing blocks o f the membranes by forming a bi layer with po lar groups at the l i p i d -wa te r interface and hydrophob ic chains sequestered in the center o f the bi layer. Th is bi layer structure was first proposed by Go r te r and G rende l (1925) and by Dan ie l l i and Davson (1935). Since then, the bi layer structure o f biological membranes has received extensive support f rom numerous independent techniques (e.g. X - r a y diffract ion [Engleman, 1970], electron microscopy [Robertson, 1964], f reeze-f racture [Branton and Park, 1967], and nuclear magnetic resonance [Cul l i s , 1976]). B i layer l i p id structure was incorporated by Singer and N i co l son (1972) into a " f lu id mosaic" mode l o f b io logica l membranes. The f luidity is attr ibuted to the rotational and translational mot ion possible for phosphol ip ids in a ( l i q u i d -crystal l ine) bi layer matrix, wh i le the mosaic nature reflects integral proteins that penetrate into or through the membrane, and per iphera l proteins that are associated wi th the membrane surface. The integral proteins often contain a h igher percentage o f non -po l a r amino acids wh ich are presumably involved in a hydrophobic sequence penetrating the membrane. The per iphera l proteins are usually more polar and can 1 removed by changes i n ionic strength as they are largely attached by weak electrostatic bonds. The fluid mosaic mode l o f membranes has been widely accepted most membrane properties can be accomodated i n terms of this general hypothesis. 5 1.4 Phys ica l Propert ies o f L ip ids A typical b io logical membrane contains over a hundred different species o f l ip ids. Eukaryot ic membranes, for example, may contain phosphat idy lethanolamine, phosphatidylser ine, phosphat idy lchol ine, sphingomyel in , phospha t ide acid, phosphat idy l inos i to l , and cholesterol or other sterols. Each o f the phospho l ip id species have associated with them a range o f fatty acids o f d i f fer ing chain length and degree o f saturation. In order to determine functional roles o f ind iv idua l phosphol ip ids, mode l systems using synthetic phosphol ip ids o f wel l def ined composi t ion have been developed, and a variety o f techniques employed to moni tor the physical propert ies o f component l ipids. These inc lude electron sp in resonance, nuclear magnetic resonance, and calor imetr ic techniques to investigate the ge l - l i qu i d crystal phase characteristics, membrane f luidity, and l ip id po lymorph i sm. 1.4.1 G e l - L i q u i d Crysta l l ine Phase Behav iour The thermotropic behaviour o f a large number o f pure l ip ids have been investigated using calor imetr ic techniques (see, for example, Chapman et. ah, 1974). Pure phosphol ip ids i n aqueous dispersions undergo a phase transit ion f rom a rigid gel-state to a f lu id l iqu id-c rys ta l l ine state as the temperature is increased. Th i s transit ion temperature ( T e ) is dependent upon the l ip id class, a cy l - cha i n length, degree o f a cy l - cha in saturation, hydrat ion, and ionic environment (Chapman, 1975). In wel l def ined l i p id systems, the transit ion is h igh ly cooperative. However since, a bio logical membrane consists o f a large variety o f phospho l ip id species, the gel to l i qu id -c rys ta l l i ne transit ion is much broader. The relevance o f this type o f transit ion to bio logical membrane funct ion is not clear, except that most bio logical systems appear to require a f lu id membrane for function. 6 1.4.2 F lu id i t y O f Membranes The f lu id i ty o f a bio logical membrane reflects the degree o f disorder in the phospho l ip id acy l -cha ins , and can be moni tored by spectroscopic techniques such as elect ion spin resonance (W i r and McConne l l , 1975), nuclear magnetic resonance (Chapman and Penkett, 1966; Seelig, .1978), or fluorescent polar izat ion (Fuchs et al., 1975), and differential scanning calor imetry techniques (Chapman, 1975). The f lu id i ty is related to the ge l - l i qu i d crystall ine phase transit ion o f the component ind iv idua l l ipids, that is, membranes containing a large percentage o f unsaturated l ip ids wi th a low T c w i l l be more f lu id at a given temperature than those containing a large percentage o f saturated l ipids. Cholestero l can decrease membrane f luidity. Cholestero l is the predominant neutral l i p id found in mammal i an membranes and has dramatic effects on membrane f luidity. In a l iqu id-c rys ta l l ine med ium, cholesterol exerts a "condensation effect", increasing the degree o f order i n the acyl chains and thus decreasing f luidity. Al ternat ively, for gel-state l ip ids, cholesterol has a " l iqu i fy ing effect", preventing crystal l ization (De G i e r et al . , 1969). The presence o f cholesterol in a bio logical membrane has therefore been suggested to play a role i n modulat ing the f lu id i ty o f the bi layer. The biological s ignif icance o f membrane f lu id i ty is as yet unclear. It has been suggested to be related to the permeabi l i ty o f membranes or the control o f membrane enzyme activity. Ev idence for the former possibi l i ty is suggested by permeabi l i ty studies correlat ing the effect o f fat ty-acy l cha in composit ion o f phosphat idy lchol ine membranes to the permeabi l i ty to glycerol or erythritol (De G i e r et al . , 1968). Al ternat ively, observations o f discontinuit ies in Ar rhen ius plots o f membrane bound enzyme activities have been attributed to the "mel t ing" o f an annular l i p id shel l around the enzyme (see, for example Hesketh et al . , 1976). 7 1.4.3 Po l ymorph i c Phase Trans i t ion Apar t f rom the ge l - l i qu i d crystal l ine phase transition, phosphol ip ids can also undergo po lymorph i c phase transitions f rom the l iqu id-c rys ta l l ine bi layer structure to a hexagonal H u conf igurat ion or other non -b i l aye r structures (Cu l l i s and DeK ru i j f f , 1978a). Th is po lymorph ic phase transit ion occurs approximately 10°C above the ge l - l i qu i d crystal l ine transit ion temperature for unsaturated phosphat idylethanolamines and is dependent on hydrat ion, l i p id composit ion, p H , and divalent cation content (Cu l l i s and DeK ru i j f f , 1979). Studies on pure phospho l ip id species reveal that a signif icant proport ion o f the phosphol ip ids in a bio logical membrane preferential ly adopt ( in isolat ion) the hexagonal H H conf igurat ion at phys io logica l temperatures under condit ions o f fu l l hydrat ion. Important examples o f such l ip ids inc lude unsaturated phosphat idylethanolamines (Cu l l i s and DeK ru i j f f , 1978a and 1978b, Re i s s -Husson , 1967), monoglucosyld ig lycer ide (Wies lander et a l . , 1978), as wel l as phosphat ide acid and card io l ip in i n the presence o f ca lc ium (Papahadjopoulos et al . , 1976a; Rand et al . , 1971). In addit ion, l ip ids such as cholesterol (Cu l l i s and DeK ru i j f f , 1978a and 1978b; Cu l l i s et a l . , 1978a; Ba l ly et_ al. , 1983) and unsaturated fatty acid (Cu l l i s and Hope , 1978) induce format ion o f the hexagonal H u phase. L o w p H values can also engender the hexagonal H n phase as has been demonstrated for phosphat idylser ine (Hope and Cu l l i s , 1980) and phosphat ide acid (Far ren et al ., 1983) at p H values below 4. The po lymorph ic behaviour o f phosphol ip ids i n relat ion to funct ional abi l i t ies o f bio logical membranes has been reviewed by Cu l l i s and D e K r u i j f f (1979). They have suggested non -b i l aye r structures such as inverted cyl inders (characteristic o f the hexagonal H u phase) and inverted micel lar structures could occur as intermediates i n membrane-med ia ted processes such as membrane fusion and transbilayer transport o f l ip ids ( f l i p - f l op ) and divalent cations. 8 1.5 Asymmet r i c Nature O f B io log ica l Membranes Most , i f not a l l , biological membranes exhibit transbilayer asymmetry wi th respect to their l ip id , carbohydrate, and protein distr ibutions. Tne best documented system is the erythrocyte cel l membrane. Prote in label l ing studies and lect in b ind ing experiments (Ro thman and Lenard , 1977) indicate that the outer leaflet contains oligosaccharides, the active sites o f two major proteins (acetylcholinesterase and 5'-nucleot idase), and has a l ip id composi t ion wh ich is predominant ly phosphat idy lchol ine and sphingomyel in . The inner (cytoplasmic) leaflet is predominant ly composed o f the aminophosphol ip ids, phosphat idy lethanolamine and phosphat idylser ine (Zwaal et a l . , 1973). The presence o f l i p id and prote in transbilayer asymmetry is related to different funct ional requirements on either side o f the membrane. F o r example, there is now good evidence for specific roles o f carbohydrates in cel l adhesion and recognit ion (Hakomor i et al . , 1981), and the activities o f the proteins on the inner leaflet has been associated with various intracel lular reactions. However , the reasons why phospho l ip id asymmetry is required and how it is mainta ined remain unknown. W i t h regard to exocytosis, the possibi l i ty o f asymmetr ic transbilayer l i p i d distr ibutions is o f interest as the fusion event in i t ia l ly involves a direct interaction between the outer monolayer o f the secretory vesicle and the inner monolayer o f the plasma membrane. Fur thermore, f reeze-fracture studies o f the exocytotic event in chromaff in cells (Anu i s et al . , 1979) and mast cells ( Ch i et_ al. , 1976) suggest that the membrane proteins segregate away f rom the fus ion site and therefore probably do not p lay a direct role in the actual fusion event It is most l ike ly that proteins set the stage for fusion and br ing about close apposit ion o f the two membranes. The fusion event itself must involve some sort o f structural 9 transit ion o f the phosphol ip ids. A major thrust o f this thesis is that tendency o f inner monolayer l ip ids to assume non-b i l aye r conf igurat ion could be an important factor in the fusion event dur ing exocytosis. Two techniques are employed in this work to determine the structural characteristics o f l ip ids in mode l and biological membrane systems. These are 3 1 P nuclear magnetic resonance ( 3 1 P - N M R ) and freeze-fracture electron microscopy. The nature o f these techniques and the type o f in format ion that can be obtained f rom each o f these procedures is indicated below. 1.6 Use o f 3 1 P - N M R in determining membrane structure 3 1 P - N M R is an attractive non-per tu rb ing diagnostic technique to investigate the po lymorph ic phase behaviour o f membrane phosphol ip ids. The detection o f phospho l ip id po lymorph i sm by 3 1 P - N M R has been discussed by Cu l l i s and K r u i j f f (1979). Three important points are rev iewed below. First, a large chemical anisotropy is exhib i ted by the phospho l ip id phosphorus nuclei , wh ich for large l iqu id-crys ta l l ine bi layer systems (200 nm) is on ly part ia l ly averaged by the restricted modes o f mot ion avai lable. These consist p r imar i l y o f rap id rotat ion o f the molecules about their long axes. By employ ing proton decoupl ing, it is possible to remove the dipolar broadening aris ing f rom the nearest neighbour methylene groups and this results in a characteristic "b i layer" spectrum with a low f ie ld shoulder and a h igh f ie ld peak separated by approximately 40 p p m for lamel lar phosphol ip ids. A typical b i layer spectrum is i l lustrated in F igure l a . Second, a l l g lycero l -based phosphol ip ids (PC , PE , PS, P G , PI) and sphingomyel in exhib i t a s imi lar l ineshape when in the ( l iqu id-crysta l l ine) bi layer organizat ion. 10 Phospho l ip id phases C o r r e s p o n d i n g 3 ' ? N M R spectra B ila yer s\ r% r\ ff\ - ~n sssssj/p, H exogonal (H t | ) Phases where isotropic mot ion occurs 1. Vesicles 2. Inverted micellar 3. M icellar 4. Cub ic 5. R hombic -40 ppm- H F igure 1. 3 1 P - N M R Spectra o f Phospho l ip id Phases. 11 Therefore, in a m ixed l ip id bi layer system, al l the phosphol ip ids contribute to a composite "b i layer" l ineshape. However , it should be noted that i n a smal l b i layer system (such as sonicated vesicles) vesicle tumbl ing and the lateral d i f fus ion o f the phosphol ip ids can inf luence the 3 I P - N M R spectrum obtained. In large (diameter 200 n m ) bi layer structures such as l iposomes and most bio logical membrane preparations, the process o f lateral d i f fus ion is not fast enough on the N M R timescale (10" 5 seconds) to produce an effective mot ional averaging mechanism. In smaller systems, the abi l i ty o f phosphol ip ids to diffuse lateral ly around the vesicle and vesicle tumbl ing results i n narrow resonances. These " isotropic" mot ional averaging effects are also observed for l ip ids in inverted micel lu lar conf igurat ions and in other structures, such as the cubic or rhombic phases (Luzzatt i et al., 1968; Luzzat t i and Tard ien, 1974). Thus a narrow 3 1 P - N M R signal does not prov ide an unambiguous indicat ion o f l i p id structure. A th i rd po int is that the major non -b i l a ye r structure adopted by certain membrane phosphol ip ids in isolat ion is the hexagonal Hn phase. A s shown in F igure l b , the hexagonal H n conf igurat ion consists o f long cyl inders o f phosphol ip ids whose polar headgroups are or iented towards smal l (2 n m diameter) aqueous channels. Such structures experience addit ional mot ional averaging as compared to the large bi layer structures due to the abi l i ty o f the l i p id to diffuse around the aqueous channels. A s a result, the hexagonal H u phase exhibits a characteristic 3 1 P - N M R l ineshape wh ich has reverse asymmetry compared to the bi layer l ineshape and is narrower by a factor o f two. A summary o f a l l three l ineshapes are i l lustrated in F igure 1. It shou ld be emphasized that although crit ic isms have been made (Thayer and Koh le r , 1981; Noog le et al., 1982) the close correlat ion between the po lymorph ic phase behaviour o f phosphol ip ids as detected by 3 1 P - N M R , freeze-fracture and X - r a y studies (Cu l l i s et al ., 1983; T i lcock et al ., 1983) conf i rms the val id i ty o f 3 1 P -12 N M R techniques for determining membrane l i p id structure. 1.7 Freeze- f racture Electron Mic roscopy Freeze- f racture electron microscopy is p lay ing an increasingly important role in membrane research. The technique involves four steps: (i) rap id freezing or quenching o f the sample, ( i i) fracturing, ( i i i ) shadowing and repl icat ion o f the fracture face, and (iv) visual izat ion o f the repl ica by employ ing an electron microscope. Quench ing requires the specimen to be very rapid ly f rozen in l i qu id freon or propane so as to "capture" the h igh temperature organizat ion. Ideally, product ion o f ice-crystals should also be avoided. These quenching artifacts can be avoided by employ ing recently developed fast- f reez ing techniques (Bachmann and S chm i t t - Fum ian , 1973; Heuser and Salpeter, 1979) or by using cryoprotectants such as glycerol, d imethylsul foxide, ethylene glycol, etc. However , the latter may also inf luence the substructure and produce other artifacts. Subsequently, the sample is fractured under h igh vacuum condit ions (10" 6 Tor r ) and at - 110°C . The fracture-face is "repl icated" by shadowing wi th p la t inum/carbon (P t /C ) at angles o f about 45° fo l lowed by carbon shadowing to improve the mechanical stabil ity o f the repl ica. The repl ica is washed in cleaning solutions i n order to remove any adher ing specimen and visual ized under an electron microscope. It is general ly accepted that the fracture plane proceeds through the hydrophobic interface between the two monolayers o f l i p id compr is ing the membrane bi layer (Branton, 1966; P into da S i lva and Branton, 1970). Accord ing to common nomenclature (Branton et al ., 1975), the fracture face o f the outer monolayer is termed the exoplasmic fracture face ( E F ) and the fracture face o f the hydrophobic side o f the inner monolayer is termed the protoplasmic fracture face (PF ) . Th is is i l lustrated in F igure 2a. It should be noted that in the case o f intracel lular organelles, the outer cytoplasmic monolayer would correspond to the PF and the inner luminal fracture face would correspond to EF as shown in Figure 2b. PF EF (a) (b) Plasma membrane Intracellular granules Figure 2. Freeze-Fracture Nomenclature Freeze-fracture techniques can also be used to visualize phospholipid organizations such as the bilayer, hexagonal H n , and "lipidic particles" (inter or intra-bilayer inverted micellar structures) as illustrated in Figure 3. The advantage of freeze-fracture electron microscopy is that it provides information on local structural features of membranes and membrane lipids. This is in contrast to spectroscopic, X-ray diffraction, and calorimetric techniques which provide averaged information over the whole sample. In addition, lipid vesicles can also be sized employing freeze-fracture techniques (van Venetie et al. , 1980). 14 Phospho l ip id phases Cor respond ing F racture - faces Figure 3. Freeze-Fracture C h a r a c t e r i s t i c s of Phospholipids i n Various Phases. 15 As mentioned earlier, Ca2" is the second messenger in the exocytotic process and it appears that cytosolic Ca :* regulates the exocytotic event The following sections briefly review regulation of cytosolic Ca 3 4 and the hypotheses for Ca2' mobilization. 1.8 Calcium Transport and Regulation of Secretion The concentration of free Ca 2 4 inside the cytoplasm is maintained at approximately 10~7M, compared with about 10" 3 M outside the cell. The important role of cytosolic calcium levels in various biochemical processes (including secretion) necessitates strict control of Ca 2 4 levels. Ca 2 4 influx and Ca 2 4 efflux are governed by separate processes (Akerman, K.E.0 and Nicholls, D.G., 1983). Active transport mechanisms are involved in maintaining the low basal levels of Ca 2 ' in the cytoplasm. The identification of the (Ca 2 4 + Mg2+)ATPase as a Ca 2 4 transporter in numerous cell types suggests that it is the major protein involved in pumping Ca 2 4 out of a cell (Barritt, 1981). Other active transport mechanisms which also remove cytosolic calcium exist in intracellular organelles such as the mitochondria and endoplasmic reticulum. In the case of mitochondria, Ca34-uptake is the result of a balance between a membrane potential driven Ca2'-uptake mechanism and an electroneutial efflux of Na4 ions (Carafoli and Crompton, 1978). ATP-linked Ca 2 4 transport in microsomal fractions from synaptososmes has also been described (Blaustein et al. , 1980). Calcium influx into cells occurs down a chemical gradient (Ca 2 4 0/Ca 2 4^) of about 103—10" and ah electrical gradient of about 75 mV (negative inside). An increase in cytosolic calcium in response to external stimuli could be a result of an increase in calcium influx across the plasma membrane or a release of calcium from internal stores. Since exocytosis has an apparent requirement for extracellular 16 calcium (Douglas, 1968), the former appears more likely. Upon stimulation, "Ca2*-channels" are opened in the plasma membrane during depolarization. These so-called "late Ca2*-channels" or "voltage-sensitive Ca2* inflow channels" are blocked by various bi and trivalent cations including Co 2 ', Mn2*, and La3* as well as by some organic compounds such as verapamil, D-600 and nifedipine (Baker, 1972). Sodium channel blockers such as tetrodoxin do not affect the function of Ca channels (Katz and Miledi, 1969). The mechanisms by which certain agonists (e.g. acetylcholine, adrenalin,etc.) increase the influx of calcium across the plasma membrane are not clear. It is not known whether the process involves an increase in the activity of the Ca2* transporter which catalyzes calcium influx in the resting state, or results from the opening of a different type of Ca2* channel. There have also been reports which indicate that agonists may inhibit Ca2* efflux thereby causing an increase in cytosolic calcium levels. For example, Pershadsingh and McDonald (1980) showed that insulin can inhibit the activity of the (Ca2* + Mg2")ATPase enzyme in plasma membrane vesicles derived from adipocytes. Recently, five hypotheses for calcium mobilization have been made. 1.8.1 Hypotheses for Calcium Mobilization A first hypothesis is that changes in phospholipid composition of the plasma membrane induce a conformational change in a calcium transport protein, thereby increasing its activity (Michell et al. , 1977). The idea presumes the existence of a calcium transport protein, which has not yet been shown to occur in cells. A second hypothesis suggests that diacylglycerol produced during the PI-response (section 1.9) stimulates a protein kinase which in turn phosphorylates a membrane 17 proiein responsible for altering cytosolic calcium concentration. A calcium-dependent protein kinase that is activated by diacylglycerol, phosphatidylserine and calcium was recently discovered in platelets (Takai et al. , 1979). This "protein kinase C" is suggested to either inhibit a calcium pump or induce a conformational change in the calcium gate protein, thereby causing an increase in intracellular calcium.. In this regard, a 40 K dalton polypeptide appears to be phosphorylated by protein kinase C during platelet activation (Kawahara et al. , 1980). A third hypothesis by Hirata and Axelrod (1983) proposes that the biochemical mechanism of signal transduction across biomembranes involves stimulation of the methyltransferases involved in the methylation of phosphatidylethanolamine (PE). This is suggested to affect the viscosity of the membrane, leading to increased Ca2* permeability, increased levels of cyclic nucleotides, and increased membrane phosphorylation. Thus, the biochemical changes in the cell following stimulation of receptors in response to external stimuli is coupled to phospholipid turnover. However, in conditions where marked changes in microviscosity in the erythrocyte membrane were observed, only extremely small amounts of PE were methylated (Hirata and Axelrod, 1978). This and the extremely low methylation activity observed in tissues has raised doubts that methylation of PE could account for the many physiological responses attributed to this activity (Vance and De Kiuijff, 1980). The fourth hypothesis postulates that phosphatide acid (PA) which accumulates following the phosphatidylinositol-response (section 1.9) acts as a calcium ionophore. PA can translocate calcium across both organic solvent layers and liposomal membranes (Putney, 1981). Addition of PA to smooth muscle cells (Salmon and Honeyman, 1980) and parotid slices (Putney, 1980) mimics these tissues' responses to the appropriate calcium mobilizing hormone. However, Michell has not found an 18 increase in calcium permeability on increasing erythrocyte membrane PA content (Michell et al. , 1977). Also in neutrophils, agonist-stimulated PA formation may be too slow to play a role in calcium-dependent release of lysosomal enzymes (Cockcroft et al. , 1980). These difficulties aside, transport of Ca3* by PA remains an attractive possibility which is further investigated in this thesis. The last hypothesis is that calcium is released from the interior of the plasma membrane following the breakdown of polyphosphoinositides which have a higher Ca3* binding capacity than less-phosphorylated inositol- lipids (Gil et al. , 1983). Although the idea is quite attractive, there is little supporting evidence. Also, the polyphosphoinositides have only a slight preference for calcium over magnesium. In summary, the regulation of the exocytotic event in secretion is dependent on the presence of calcium. On depolarization of the plasma membrane induced by passage of an action potential or by specific agonists, calcium ions flow into the cell or are "mobilized" via some unknown mechanism. The consequent increase in cytosolic calcium concentration triggers release. Five hypotheses have been proposed implying either the existence of a calcium transporter protein or that products of phospholipid metabolism regulate calcium mobilization. To date no "calcium channel" has been isolated, suggesting a calcium ionophore may be activated or synthesized when needed. One such candidate is phosphatide acid which is produced in the phosphatidylinositol response to binding of external ligands. Alternatively, phosphatidylinositol and/or one of its other derivatives could be directly involved in calcium transport. In this context, it is useful to briefly review what is known concerning phosphatidylinositol metabolism upon stimulation. 19 1.9 The Phosphalidylinositol Effect Thirty years ago, Hokin and Hokin (1953) first described the "phosphatidylinositol-effect" as an enhanced incorporation of 3 3P into phosphatidic acid and phosphatidylinositol following stimulation of the pancreas by acetylcholine. Further studies suggested that the PI-effect accompanying receptor mediated stimulation reflected a direct lipid involvement in cellular responses to extracellular agonists such as norepinephrine, histamine, serotonin, vasopressin, angiotensin, and thrombin. In 1975, Michell proposed that the PI-effect was closely coupled to receptor occupation and occurs prior to calcium influx and was associated with a rise in cytosolic calcium levels (Michell, 1975). The evidence cited to support this hypothesis was the close relationship between the dose-response curves for PI turnover and receptor occupation, the rapid onset of PI turnover following a stimulus and the lack of sensitivity of this process of depletion of cellular calcium. The PI-response occurs in many tissues (Michell, 1979) and is involved in the rapid responses of secretory cells or the longer-term stimulation of cell proliferation in lymphocytes, fibroblasts, denervated skeletal muscle or sympathetic ganglia. The PI-response is restricted to receptors which result in calcium mobilization on binding agonists (e.g. muscarinic, a-adrenergic and substance P) whereas p- adrenergic and glucagon receptors do not exhibit a PI-response on binding. The PI-response may therefore have a role in mobilization of calcium. Biochemically, the PI-response is a cyclic process. On receptor activation PI is degraded to 1,2-diacylglycerol which is then phosphorylated to phosphatidic acid and subsequently resynthesized to phosphatidylinositol. This is closely coupled with an increase in cytosolic calcium concentrations. The relationship between PI metabolism and changes in calcium permeability is unclear but most of the evidence 20 suggests that the PI-response precedes calcium mobilization (Michell, 1979). The three pieces of (indirect) evidence are, first, that PI-breakdown is independent of calcium and can occur at the low calcium levels found in a resting cell and furthermore is not affected by removing calcium from the bathing medium, although cell activation is drastically reduced. Second, La3* which inhibits secretion and completely suppresses calcium transport has no effect on the PI-response (Berridge, 1981). Finally, A23187, a calcium ionophore, fails to induce the PI-response in the parotid gland (Jones and Michell, 1975). The same was found in insect salivary-gland, synaptosomes, and in the pancreas. Therefore, agonist-dependent hydrolysis of PI is independent of any fluctuations in calcium concentrations. The best direct evidence linking PI-metabolism to calcium mobilization is from blowfly salivary glands, where the effect of removing PI from the membrane was investigated (Berridge, 1980). In addition to stimulating PI hydrolysis, 5-hydroxytryptamine (5-HT) was also found to inhibit the synthesis of PI. It was found that stimulation of isolated blowfly salivary glands with a high dose of 5-hydroxytryptamine resulted in almost complete inacu'vation of salivary glands. These glands could be completely resensitized if they were allowed to resynthesize PI. Therefore, the observation that 5-HT dependent "gating" of calcium correlates with alterations in the levels of PI provides strong evidence that the hydrolysis of PI plays an essential role in the opening of the so called calcium channels. Unfortunately, it is very difficult to obtain direct evidence concerning the receptor mechanism responsible for calcium mobilization in other tissues because of difficulties in measuring calcium permeability. A scheme illustrating the PI-response as a coupling event in receptor controlled mobilization of Ca2* is shown in Figure 4. 21 AGONIST + RECEPTOR (A) (R) AR 1 PI-RESPONSE Pf DG CMP ^inositol CMP-PA I^ATP ^ADP PA MOBILIZATION OF CALCIUM CELLULAR RESPONSES n e.g., Exocytosis Contraction Phosphoryl kinase activ Guanylate cyclase activ.n Increase perm, to K+ & CI Figure 4. Receptor-controlled mobilization of Calcium. 22 1.10 Phosphatidic Acid (PA) and Calcium Mobilization As indicated previously, PA may act as a calcium ionophore. Three results support this hypothesis. First, using a Pressman cell, phosphatidic acid has been shown to have characteristics consistent with a calcium ionophore as indicated by an ability to transport Ca2* through an organic solvent (Tyson et al. , 1976). Similarily, Cullis et al. (1980) have shown that PA can form a lipid-soluble complex with Ca2* in chloroform. By employing liposomes containing a calcium sensitive dye, Serhan et al. (1981) suggest that PA can translocate Ca2* across lipid membranes. Second, Salmon and Honeyman (1980) showed that synthesis and accumulation of PA is accelerated in smooth muscle cells stimulated by carbamylcholine with a similar time course to that of contraction. Alteration in PA metabolism did not seem to be a consequence of an increase in intracellular Ca2* or depolarization of the cell -membrane. Finally, introduction of submicromolar concentrations of PA rapidly produced contractions of isolated smooth muscle cells (Salmon and Honeyman, 1980). These results imply that cholinergic-induced changes in membrane Ca2* permeability in smooth muscle cells can be mediated by PA accumulation which follows the PI-response. Similar effects of PA has also been shown in the pancreas (Hokin, 1974) and platelets (Lapetina and Cuatrecasas, 1979). Finally, Putney and coworkers (1980) reported evidence that PA could act as a calcium ionophore under neurohumoral control in parotid cells. This was based on observations that receptor activation increased the PA content of the parotid and that exogenous PA stimulated a Ca2*-dependent response in the parotid. In summary, hormones, neurotransmitters and other cell stimuli appear to alter the Ca2* permeability characteristics of the plasma membrane. Electrophysiological 23 evidence indicates that the st imul i increase the number o f vo l tage-dependent " C a 2 * -channels" (Reuter, 1979). The mechanisms by wh ich these agonists induce an increase in C a 2 + in f low across the p lasma membrane remain unclear. A n attractive hypothesis is that the PI- response results i n increased P A content wh ich acts as the biological ionophore. Unfortunate ly, there is no known inhib i tor o f phosphol ipase C to test this hypothesis under in v ivo condit ions. However , the abi l i ty o f P A to both sequester C a 2 + into an organic phase and also to transport it across membranes supports such a role for phosphat id ic acid. Th i s transport abi l i ty is examined in detai l in Chapter III. 1.11 De f in i t i on o f Membrane Fus ion Membrane fusion is def ined as the process by wh ich two separate membrane bound systems unite to fo rm a single membrane. Th is is i l lustrated i n F igure 6, where two membrane bound vesicles fuse to fo rm a single vesicle. It is obvious that pr ior to F igure 5 - Membrane Fus ion the fusion event, close apposit ion o f the two membranes is necessary in order to a l low fusion to proceed. C lose apposit ion is l ike ly to be prote in mediated. Ev idence for this comes f rom the recent characterization o f the cytoplasmic ca lc ium b ind ing protein, synexin (Cruetz et al., 1978). Synexin has been suggested to sel f-24 associate into rods in the presence of calcium and cause granule membranes to aggregate into pentalamellar complexes (Cruetz et al., 1982). Binding sites for "activated" synexin have also been detected not only on the chromaffin granule membranes but also on the cytoplasmic face the chromaffin cell plasma membrane (Pollard et_aL, 1981). The actual fusion event between the two apposing membranes probably involves direct participation of the phospholipids. As illustrated in Figure 5, it is difficult, if not impossible, to imagine such a fusion event occurring while the lipid bilayer structure is maintained. At some point, the two apposing bilayers will have to adopt an alternative structure in order to allow fusion to proceed. In this regard, four molecular models of fusion have been proposed over the past decade. These are briefly reviewed below. 1.11.1 Lysolecithin and Membrane Fusion Rubin (1967), Guttler and Clausen (1969) suggested the formation of lysophospholipids in promoting membrane fusion and experimental support was presented a few years later by Lucy and coworkers when they fused hen erythrocytes in vitro with lysolecithin (Pool et al., 1970). They also observed that the cells were very, unstable and lysed within a period of 30 seconds. Studies with artificial lipid membranes have indicated that insertion of a wedge-shaped molecule such as lysolecithin could promote substantial perturbation in the packing of the lipid molecules (Haydon and Taylor, 1968). Also, due to the highly lytic properties of lysolecithin, its production during membrane fusion would need to be confined to highly localized sites so that the integrity of the rest of the membrane would be maintained. Cells have been found to possess enzymes responsible for the conversion of lysolecithin to non-lytic derivatives (Mulder and van Deenen, 1965) but the levels of enzymes responsible to generate lysolecithin and its removal appear to vary significantly between different membranes of the same cell. It has also been proposed that the high concentrations of lysolecithin found in the chromaffin granules (Blaschko et al. , 1967) may be involved in the exocytotic process. Although it is possible that membrane fusion occurring in vivo may involve bilayer to micellar transitions, no experimental support for this concept has been presented. Also, the fact that the turnover of lysolecithin in the adrenal medulla is not altered during secretion (Trifaro, 1969) argues against the participation of lysolecithin. 1.11.2 Role of Calcium It is now clearly established that calcium is essential for fusion in a wide variety of fusion events in both natural and model systems (Poste and Allison, 1972). In natural fusion systems Ca2* has been demonstrated to induce processes such as exocytosis in the chromaffin cells (Douglas, 1968), mast cells (Foreman et al, 1973), and in other neurosecretory cells (Norman, 1976), by either introducing calcium directly into the cell or by calcium ionophores A23187 and X537A. Although the exact mechanism by which calcium promotes fusion is unknown, th strong interaction between calcium and acidic phospholipids has led to two recent models of membrane fusion. 1.11.3 Crystallization Model of Fusion The interaction of calcium with phospholipids, particularly acidic phospholipids has been extensively investigated by Papahadjopoulos and coworkers (Papahadjopoulos et al., 1978; 1979). Calcium has been shown to affect the permeability properties, 26 form iwo to one (PS/Ca2*) complexes, cause crystallization of the acyl chains of acidic phospholipids and thereby raising their gel-liquid crystalline transition temperature, and to cause phase separations in mixtures of acidic and neutral phospholipid model systems. On the basis of such observations, a dual role of Ca2* in mediating fusion has been proposed. First, the presence of Ca2* is suggested to promote close apposition of adjacent membranes by enchancing electrostatic interactions between them (Newton et al. , 1978) and by forming a specific intermembrane complex (Portis et al. , 1979). Second, Ca2* induces destabilization of the apposed membranes by formation of crystalline domains of acidic phospholipids which would represent sites at which fusion would occur (Newton et al. , 1978). Support of such a mechanism of membrane fusion has been shown in the case of phosphatidylserine containing model membrane systems but as yet no evidence has been presented for fusion preceding via Ca2* induced segregation of PS in vivo. 1.11.4 Polymorphic Model of Fusion The observations that specific phospholipids in a biological membrane can adopt a non-bilayer structure in isolation (Cullis and de Kruijff, 1978a) and that Ca2* can trigger such structures in mixed model membranes (Tilcock and Cullis, 1981) has led to the proposal of a polymorphic model of membrane fusion. The main precept of such a model is that at some stage in the fusion event, whether it is mediated by protein or lipid, a portion of the lipids must experience a departure from the bilayer structure. Studies with erythrocyte ghost membranes (Hope and Cullis ,1981) illustrated that incorporation of fusogens such as oleic acid or glycerol mono-oleate at concentrations sufficient to induce cell fusion also induced a bilayer to hexagonal 27 H n phase transition in some portion of the membrane phospholipids. Also, Verkleij and coworkers (1979a; 1979b) have presented freeze-fracture evidence from model membrane systems showing the occurrence of intramembranous lipidic particles. It has been suggested that these IMP (presumably inverted lipidic micelles) were possible intermediatories in the bilayer to hexagonal H n transitions (Verkleij et al. , 1980). On the basis of these observations, a molecular model of membrane fusion has been proposed where the intermediate structure consist of inverted micellar structures (Cullis and Hope, 1978; Cullis and de Kruijff, 1979; Cullis et al. , 1980). The suggestion that inverted micelles occur in the fusion process has also been presented by other workers. For example, Lau and Chan (1975), studying alamethicin induced fusion by proton NMR, Pinto da Silva and Nogueira (1977), studying freeze-fracture results of the fusion of peripheral vesicles with plasmalemma of zoospores of the fungus, Phtophthora palmivora , and Gingell and Ginsberg (1978), from a theoretical standpoint, have all proposed similar intermediate structures. Although the polymorphic fusion model has been supported by results obtained from model membrane systems, no in vivo evidence for this mechanism is yet available. 1.12 The Chromaffin Granule: Structure and Function Once Ca 2 i is inside a secretory cell, it is directly involved in triggering the fusion event involved in exocytotic release. The final chapter of this thesis investigates the possible role of phospholipids in this Ca 2 + dependent fusion evenL The system studied is the chromaffin granule which fuses with the chromaffin cell plasma membrane during exocytosis. The ease of isolation, low cost and extensive literature on chromaffin granules make it a conven ient system. 28 1.12.1 Historical Perspective The first indication that the hormones of the adrenal medulla, noradrenalin and adrenaline, were stored in a subcellular organelle was obtained by Blashko and Welch (1953) and by Hillarp and Hokfelt (1953). They showed that the bulk of the catecholamines resided in the fraction obtained by high speed centrifugau'on after removing the cell debris by low speed centrifugation step. Electron micrographs of the adrenal medulla by Lever (1955) revealed membrane bound vesicles (200 nm diameter) which were smaller than the mitochondria and he suggested that these stored the hormones of the adrenal medulla. Sjostrand and Wetzstein (1956) obtained similar micrographs and introduced the term chromaffin granules. A morphological and biochemical characterization by Hager and Barrnett (1960) led to the conclusion that the chromaffin granules were distinct organelles involved in storage of the catecholamines. 1.12.2 Composition Of The Chromaffin Granule The two components of the chromaffin granule, water-soluble (i.e. granule contents) and water-insoluble (i.e. granule membrane) fractions can easily be separated by lysis of the chromaffin granules in a hypo-osmotic buffer solution. The composition of the contents and of the membrane are listed in Table I (Winkler and Smith, 1975). 29 Table I: Compos i t i on o f the Bov ine Adrena l Ch roma f f i n G ranu l e Const i tuent Amount , % o f total dry wt_ soluble content catecholamines adenine nucleotides prote in ca lc ium magnes ium 20.5 15 27 0.1 0.02 insoluble (membrane) content phospho l ip id cholesterol prote in ca lc ium magnes ium 17 5 8 0.06 0.02 The soluble contents are r ich in catecholamines (0.71 M ) , adenine nucleotides (0.13 M A T P ) , and proteins (210 mg/ml ) . The soluble proteins are termed chromagranins. There are at least twelve different chromagranins and their functions are sti l l unknown. The on ly prote in wi th a we l l def ined funct ion is dopam ine - 0-hydroxylase, wh ich is p redominandy found i n the membrane. The water - inso lub le proteins (presumably membrane proteins) are termed chromamembr ins and inc lude dopamine - fi- hydroxylase, Mg 2 + - a c t i v a t e d ATPase , N A D H oxidoreductase, phosphat idy l inos i to l kinase, and cytochrome 559. Spectroscopic analysis also indicates the presence o f a f lavoprotein. A characterization o f the topography o f the glycoproteins by W ink l e r and coworkers (Huber et a l ., 1979) showed that the carbohydrate port ions o f the glycoproteins faced the lumenal side o f the granule. Recent ly, Abbs and Ph i l l ips (1980) investigated the organizat ion o f the chromamembr ins and found that most o f the proteins were accessible on the cytoplasmic side o f the granules wi th at least two proteins spanning the membrane. The chromaf f in granules are relatively rich in l ip ids with a prote in to l ip id 30 ratio of 0.45 (wt/wt) in the membrane. The lipids are chiefly cholesterol and phospholipids with a high concentration of lysolecithin (about 15 %). Table II indicates the phospholipid composition of the chromaffin granule membrane. Table II: Phospholipid Composition Of The Chromaffin Granule Membrane. Phospholipid Percentage Composition Phosphatidylcholine Phosphatidylethanolamine Phosphatidylserine ' 27 34 9 Phosphatidylinositol _ Sphingomyelin Lysophosphatidylcholine Phosphatidic Acid 13 15 1 Average of all the literature values including the author's (De Oliveira-Filgueiras et al.. 1979). Efforts to elucidate possible asymmetric transbilayer distributions of phospholipids in the chromaffin granule membrane are inconclusive because about 50% of the total outer monolayer phospholipids in unsealed ghosts have been found to be inaccessible to phospholipases (De Oliveira-Filgueiras et al. , 1979; Buckland et_ al. , 1978). Although the nature of the protection is unknown, the fact that all the phospholipids can be degraded after heat-treatment of the granules is consistent with protection by membrane proteins. The only reasonably reliable results from these studies suggest mainly that the majority of phosphatidylethanolamine is localized on the outer monolayer of the granule membrane. This is in contrast to results from the red blood cell plasma membrane where most of the PE is on the inner leaflet (Verkleij et al. , 1973; Op den Kamp, 1979). Studies on the localization of lysolecithin in the chromaffin granule suggest a preferential localization in the 31 inner leaflet o f the granule membrane (Buck land et al . ,1978). 1.12.3 Roles o f the Ch romaf f i n Granu le The chromaf f in granule does not funct ion on ly as an inert reserve o f catecholamines in the adrenal medul la . It also stores several other secretory products and is also directly invo lved i n the biosynthesis and secretion o f catecholamines. F igure 6 il lustrates some o f these aspects. Apa r t f rom high levels o f catecholamines (0.71 M ) , the contents also include large quantit ies o f adenine nucleotides (0.13 M A T P ) and proteins (210 mg/m l ) (Ph i l l ips and A l l i son , 1977; W ink l e r and Westhead, 1980). The total amount o f stored substances account for approximately 63 % o f the dry weight o f the chromaf f in granule. The h igh concentration o f A T P has been suggested to be invo lved i n mainta in ing h igh concentrations o f catecholamine by format ion o f h igh molecular weight complexes (Pletscher et al . , 1974). The chromaf f in granule contains the enzyme dopamine - fi- hydroxylase wh ich plays a vital role i n the biosynthesis o f catecholamines. Since the cytosol o f the chromaf f in cel l contains potent inhib i tors o f this enzyme, the conversion o f dopamine to noradrenal in occurs inside the chromaf f in granule. Th is has been supported by 3 H -tyrosine pu l se- labe l l i ng experiments and also by the finding that reserpine, wh ich inhibits active uptake o f dopamine into the chromaf f in granule, inhibi ts synthesis o f noradrenal in i n slices o f adrenal medu l la (We iner and Rut iedge, 1966). The ult imate role o f the chromaf f in granule is in the release o f the secretory products into the blood, thus fu l f i l l ing the physio logical funct ion o f the adrenal medul la . Th i s release is accompl ished by exocytosis. Jacobj (1892) demonstrated 32 F igure 6. Subcel lu lar Dynamics O f The Ch roma f f i n Ce l l : Release o f chromaff in granule ( C G ) contents occurs by exocytosis. The empty membrane may p inch of f to give rise to several smal l coated vesicles ( C V ) . The C V lose their coating and are either degraded by lysosomal enzymes in the mult ivesicular bodies ( M V B ) or may regenerate back into a chromaf f in granule. C P , coated pit; M i t o , mitochondr ia. 34 that electrical st imulat ion o f the splanchnic nerve or direct st imulat ion o f the adrenal gland resulted in secretion o f b io logical ly active substances. Fe ldberg and coworkers (1934) later showed that acetylchol ine was the physio logical st imulus for secretion. F ina l l y , extensive studies by Douglas and coworkers (Douglas, 1968) found calc ium to be a necessary secondary messenger i n this st imulus-secret ion coupl ing event 1.12.4 Synex in and Exocytosis Synex in is a ca l c i um-b ind ing prote in or ig inal ly isolated f rom the adrenal medul la (Cruetz et al ., 1978). It causes a Ca 2 *-dependent aggregation o f isolated chromaf f in granules and is thought to be one o f the regulatory factors in the process o f exocytosis by control l ing the association o f the vesicles with the plasma membrane or to one another in the case o f compound exocytosis (Cruetz et al .. 1982). Synex in may act by reducing the threshold ca lc ium concentration required for the fus ion event in exocytosis. 1.13 Out l ine o f this thesis A s previously indicated, the purpose o f the work presented in this thesis was to investigate possible roles o f l ip ids i n exocytosis, wi th part icular emphasis on the roles o f "non -b i l aye r " l ip ids or structures. Three aspects are considered in detail. First, given P i ' s central role in exocytosis, the structural preferences o f PI are characterized in pure and mixed systems. Second, the suggestions that P A may be able to act as a Ca 2 * ionophore clearly demands examinat ion o f such abil it ies in wel l def ined uni lamel lar model systems. F ina l ly , the fusion event invo lved i n exocytosis in i t ia l ly involves an interaction between the secretory granule and the inner 35 monolayer o f the p lasma membrane. A s this inner monolayer may have a distinctive l ip id composit ion, it is appropriate to characterize possible effects o f this l i p id composit ion on the fusion process. The thesis chapters are therefore d iv ided according to these three topics as indicated below. In chapter II, the structural preferences o f PI and P I - P E mode l membranes are investigated by 3 1 P - N M R and freeze-fracture techniques. The inf luence o f divalent cations on these structural preferences is also determined. The results are discussed i n terms o f funct ional roles o f phosphat idy l inos i to l and mechanisms by wh ich Ca 2 * induces structural reorganizat ion in m ixed systems containing acidic phosphol ip ids and phosphat idy lethanolamine. Chapter III presents an investigation on the ionophoret ic properties o f phosphat id ic acid in mode l membranes, namely large uni lamel lar vesicles ( L U V s ) . Effects o f a membrane potential on Ca 2 * transport are studied. The results are compared wi th studies ut i l i z ing a wel l known ca lc ium ionophore, A23187. F ina l l y , the studies presented i n Chapter IV concern the inf luence o f sonicated vesicle systems, with l i p id composit ions wh ich may approximate the inner monolayer o f the chromaf f in granule p lasma membrane, on Ca 2 *-s t imu la ted release o f chromaf f in granule contents. C H A P T E R II Structural Preferences O f Phosphat idy l inos i to l A n d Phosphat idy l inos i to l -Phosphat idy lethanolamine Mode l Membranes 2.1 I N T R O D U C T I O N The mechanisms by wh ich some agonists can induce an in f lux o f ca lc ium across plasma membrane are not clear. F o r example, it is not known whether the process involves an increase in the rate at wh ich a ca lc ium transporter can facil itate ca lc ium inf low, or the opening o f a different type o f ca lc ium channel . A s indicated in Chapter I, i t has been suggested that the turnover o f phosphat idy l inos i to l (PI) is invo lved i n the process by wh ich agonists increase ca lc ium in f lux (M i che l l , 1979). The funct ional signif icance o f this "P l - r e sponse " is not wel l understood. T w o possibi l i t ies are apparent, (1) that phosphat idy l inos i to l and/or one or more derivatives are directly involved i n ca lc ium permeat ion, or (2) that the breakdown and resynthesis o f phosphat idyl inosi to l provides feedback regulat ion o f a receptor associated ca lc ium translocating molecule. In any event it is clear that an understanding at the molecular level o f the funct ional roles o f phosphat idyl inosi to l and its derivatives requires a detai led understanding o f relevant physical propert ies o f the various components. Th i s chapter examines the structural preferences o f hydrated phosphat idy l inos i to l in pure and mixed l i p id systems and investigated the sensitivity o f these structural preferences to divalent cations. 37 2.2 MATERIALS AND METHODS 2.2.1 Purification of Soyabean Phosphatidylinositol Phosphatidylinositol was purified from crude soya PI (Sigma, SL Louis, MO) employing first preparative high pressure liquid chromatography (Waters Prep 500) using chloroform/methanol/water/25% ammonium hydroxide (60:40:2:1, vol/vol) as the mobile phase. Typically, 5 g of crude PI was dissolved in 10 ml of chloroform and applied onto the column. Using a flow rate of 100 ml/min, fractions of 150 ml each were collected and PI was detected by thin layer chromatography (TLC). The partially pure PI (>90%) was then further purified employing carboxymethyl-cellulose column chromatography (Comfurius and Zwaal, 1971) on a 40 x 2 cm dia. column. By washing with 10 column volumes of chloroform/methanol (4:1, vol/vol) most of the contaminants were removed. PI was subsequendy eluted with chloroform/methanol (2:1, vol/vol). The final product gave one spot on two-dimensional TLC (application of 1 mg material) and was more than 97% pure with respect to phosphorus. 2.2.2 Conversion to the Sodium Salt of Phosphatidylinositol The sodium salt of PI was obtained by dissolving the dry lipid in an acidic Bligh and Dyer monophase (chloroform/methanol/0.4 M HCI) ratios which was titrated to pH 7.5 with a basic Bligh and Dyer monophase where the aqueous component was 0.5 M NaCl and 1.0 N NaOH. Addition of 0.4 volumes of water and chloroform to the total titrated volume resulted in a two-phase system. The chloroform phase containing the sodium salt of PI was dried down under nitrogen. 38 2.2.3 Isolation and Purification of PC from Egg yolk and Soyabeans Crude soya PC was purchased from Sigma (St Louis, MO). Crude PC was isolated from eggs employing 30 egg yolks which were stirred intensively in 1.25 1 of acetone in order to precipitate the lipids. The suspension was subsequently filtered through a coarse G3 glass filter. The precipitate was washed with 2 1 of acetone and the lipids were extracted three times with 500, 400, and 300 ml of chloroform/methanol (1:1) by stirring for 5 min and subsequent filtration. The combined filtrates were evaporated under vacuum and residual lipids were dissolved in about 100 ml of chloroform. To obtain a partially purified preparation of PC, the crude lipid (about 100 g) was purified by aluminum oxide chromatography. One kg of aluminum oxide (A1203) (BDH Chemicals) was suspended in chloroform/methanol (1:1) and filtered over a coarse glass filter to remove most of the fines. The washed A1203 was then suspended in chloroform and packed into a 200 X 5 cm glass column. The column was washed with 1.5 1 of chloroform and the crude PC solution was loaded onto the column. Neutral lipids (triglycerides and cholesterol) were eluted with 1.0 1 of 95% chloroform/methanol at a flow rate of 6 ml/min. The partially purified PC fractions were quickly collected by washing the column with 50% chloroform/methanol at maximum flow rate in order to prevent PC degradation to lysoPC on the column. All fractions containing PC and no lysoPC were pooled and dried down by rotary evaporation. The partially purified PC fraction was then purified by low pressure liquid chromatography on the Waters Prep LC-500 system as described by Patel and Sparrow (1978). Briefly, the compressed silica gel column was washed with 2.2 1 of chloroform/methanol/water (60:40:10) and re-equilibrated with chloroform/methanol/water (60:30:4). About twenty grams of partially purified PC dissolved in chloroform (1 gm/ml) was applied to the column, eluted at a flow-rate of 39 100 m l / rn in and 100 m l fractions were col lected. The phospho l ip id e lut ion prof i le was fo l lowed by T L C . Fract ions containing pure P C were combined and dr ied down by rotary evaporation arr iv ing at white compounds wh ich were >99% pure P C as indicated by 2 D -T L C . The pure soya P C was used to obtain P E and PS by employ ing the base-exchange capacity o f phosphol ipase D (Comfur ius and Zwaa l , 1977). Egg PS was also synthesized s imi lar i ly whi le egg P E was pur i f ied f rom total egg yolk l ip ids by H P L C . 2.2.4 Preparat ion o f Phosphol ipase D Phosphol ipase D can catalyze headgroup exchange and is very useful for the synthesis o f less abundant phosphol ip ids f rom P C . A study o f phosphol ipase D activity in var ious plant tissues by Dav idson and Long (1958) found that Savoy cabbage was the richest source o f the enzyme. Hence, a part ia l ly pur i f i ed preparat ion o f phosphol ipase D f rom fresh savoy cabbages was ut i l ized to synthesize PS and P E . Phosphol ipase D pur i f icat ion was per fo rmed as fol lows. The inner l i gh t -g reen leaves o f Savoy cabbages (4 kg) were homogenized i n a War i ng blender at max imum speed in 3 1 o f i c e - co l d water for 3 m i n intervals. The homogenate was freed f rom fiber by squeezing through four layers o f cheesecloth and then centr i fuged at 13,000 g for 30 m in . The p H o f the supernatant was adjusted to 5.5 with 4 N H C I and 250 m l fractions were qu ick ly heated to 55°C in a bo i l ing water -bath and then immediate ly cooled to 0°C. The heat- t reated filtrate was spun again at 13,000 g for 30 m i n and added to 2 volumes o f i c e - co l d acetone wi th continual st irr ing in order to precipitate the proteins. The acetone mixture was left for 2 h at 4°C. Subsequentiy, most o f the yel lowish supernatant was aspirated o f f and the precipitate was transferred into 250 m l metal centrifuge bottles and spun at 1,000 g for 5 min . The white pel let was 40 lyophilized in order to remove all the residual acetone in the preparation. The dry powder was stored at -80°C until use. For synthesis, aliquots of 500 mg were suspended in 20 ml of 0.2 M sodium acetate buffer pH 5.6 containing 0.04 M calcium chloride and continually stirred for 30 min in an ice-cold water bath. The solution was then transferred to an SS-34 centrifuge tube and spun at 17,000 g for 10 min. The brownish supernatant, corresponding to the partially purified phospholipase D fraction was used for the phospholipase D base-exchange reactions. 2.2.5 Preparation of Phosphatidylserine by Base-Exchange Reaction Both egg and soya PS were synthesized from their respective PC by employing phospholipase D (Comfurius and Zwaal, 1971). The transphosphatidylation reaction catalyzed by phospholipase D was carried out in the presence of excess L-serine either with egg PC or soya PC . The PC was dissolved in anhydrous ethyl ether at a concentration of 20 mg/ml. L-serine was first lyophilized to remove traces of methanol and subsequently dissolved at 45°C up to saturation (46 % wt/wt) in 100 ml acetate buffer (pH 5.6) containing 100 mM calcium chloride. The partially purified phospholipase D solution was added to the serine solution and an equal volume of the PC solution in ether was also added. The incubation flask was immediately closed and shaken continuously in a water-bath at 35°C. Incubation was stopped after 35 min and pressure in the nalgene incubation flask was relieved by cooling it under a stream of cold water. The mixture was transferred to a glass centrifuge bottle and spun at 1,000 g for 10 min. The top ether phase was carefully transferred into a round bottom flask and the phospholipids were extracted once more from the incubation mixture with about 50 ml of ether. The ether was removed by evaporation from combined ether fractions and the phospholipid was washed as follows in order to remove any water soluble components. 41 The phospho l ip id was dissolved in 25 m l o f ch loro form/methano l (2:1), fo l lowed by addit ion o f 6 m l o f water. The two phases were mixed with a Pasteur pipette and then centr ifuged at 1,000 g for 5 min . The top aqueous layer, containing the water-soluble components was aspirated o f f and the ch loro form l ip id phase was dr ied down under nitrogen. The y ie ld o f PS ranged f r om 20-50%. Phosphat idylser ine was pur i f i ed f rom the reaction mixture by ca rboxymethy l -cellulose ( CM- ce l l u l o s e ) chromatography as described by Com i fu r i u s and Zwaa l (1971), except that the elut ion o f the phosphol ip ids was done by a continous ch l o r o f o rm-methanol gradient rather than by a s tep -g rad ien t The pure PS fractions were pooled and converted to the sod ium salt as per section 2.2.2. The PS was shown to be > 9 9 % pure by two-d imens iona l th in layer chromatography. 2.2.6 Preparat ion o f Soya Phosphat idy lethanolamine Soya P E was also prepared by the base-exchange capacity o f phosphol ipase D. The procedure was identical to the one described for the preparat ion o f PS except that the reaction mixture contained 15 % (wt/wt) o f ethanolamine instead o f L - se r ine . The ethanolamine was neutral ized pr io r to incubat ion wi th the addi t ion o f concentrated H C I . The transphosphatidylat ion reaction in this case was much more complete with approximately 80 % yie ld. The pur i f i cat ion o f the P E was done employ ing the Waters Prep 500 as described below. 2.2.7 Pur i f i cat ion o f Egg and Soyabean P E Both egg and soya P E were pur i f ied as described by Patel and Sparrow (1978). Egg P E was pur i f ied f rom a part ia l ly pure fract ion obtained dur ing the pur i f icat ion o f egg P C , whi le the soya P E was pur i f i ed f rom the products o f the 42 transphosphatidylat ion reaction catalyzed by phosphol ipase D as described in section 2.2.6. The si l ica gel co lumn o f the Prep 500 was washed with 2.2 1 o f ch loroform/methano l /water (60:40:10) and 1.0 1 o f ch loro form/methano l /water/25 % ammon ium hydrox ide (60:30:1:1). The co lumn was then equi l ibrated wi th 2.0 1 o f the latter. Between 5 -10 g o f part ia l ly pur i f ied P E dissolved in ch lo ro form at a concentration o f 1 g /m l was appl ied to the co lumn and the phosphol ip ids were eluted at a f low rate o f 100 m l / m i n in 150 m l fractions. The pure P E fractions were pooled and rotary evaporated arr iv ing at white compounds which were > 99 % pure. Apar t f rom the T L C analysis o f PE , the bi layer to hexagonal H n transit ion temperature is very sensitive to the pur i ty o f P E . F o r example, the transit ion temperature o f soya P E is 15°C and that o f egg P E is 30°C, and sl ight impuri t ies can shift this transit ion temperature o f P E by 10°C or more. Therefore, determinat ion o f the b i l a y e r - H i L transit ion temperature o f P E was also done to access the pur i ty o f egg and soya P E . The procedure is described i n section 2.2.15. 2.2.8 Th i n Layer Chromatography ( T L C ) S i l ic ic acid was used as the adsorbent on a glass plate or on a glass slide. F o r fo l lowing elut ion profi les, microscope slides were ut i l ized whi le for quantitat ively determining the pur i ty o f phosphol ip ids, h igher resolution two -dimensional T L C was used. 43 2.2.9 M i c r o - s l i d e T L C T L C microsl ides were prepared by immers ing p la in microsl ides (2.5 X 7.5 cm) into a slurry o f si l ica gel G (50g/100 m l ch loroform) and a l lowing the th in film o f si l ica to air dry. P r io r to appl icat ion o f l ip id , the si l ica was heat activated for approximately 30 s over a hot plate and the l i p id sample was appl ied with a capi l lary tube. The T L C plate was run either i n an acid or a base solvent. The acid solvent consisted o f ch loroform/methano l /g lac ia l acetic ac id/water mixture (25:15:4:2) and the base system consisted o f a ch lo ro fo rm/methano l /ammon ia/water mixture (90:54:5.7:5.4). A f ter each run, the solvent was evaporated by gently heating the slide over a hot plate and subsequently sprayed wi th the appropriate stain for detecting specif ic phosphol ip ids, and then heated on a hot plate to develop. 2.2.10 Two -D imen s i o na l T h i n Layer Chromatography ( 2 D - T L C ) In order to determine the purity o f the phosphol ip ids quantitatively, use was made o f 2 D - T L C according to the method o f Broekhuyse (1969). P re -coa ted T L C plates (Si l ica G e l 60, 0.25 m m thickness, 20 x 20 cm plates) were used. They were activated at 120°C in the oven for one hour before use. A f ter sample appl icat ion, the plate was run i n the first direct ion in the base solvent system and subsequently run in the second d imens ion using the ac id solvent system. The composit ion o f the solvent systems are described in section 2.2.9. A f ter the runs, the plate was dr ied and exposed to iodine vapour for v isual izat ion o f the phospol ip ids. Pur i ty o f the phosphol ip ids was determined by scraping of f the si l ica corresponding to the specif ic phospho l ip id and determining the amount o f phosphorus by the method o f F iske and Subbarow (1925). 44 2.2.11 Phosphorus Spray Reagents for Identification Phosphorus reagent The specific reagent detection of phospholipids is the molybdenum blue reagent-Reagent I was made up by adding 40.11 g of Mo0 3 to 1 1 of 25 N H2SC\,, and boiling gently until the molybdic anhydride was dissolved. Reagent II was prepared by adding 1.78 g of powdered molybdenum to 500 ml of Reagent I and boiling gently for 15 min. The solution was cooled and decanted. The phospholipid reagent was prepared by mixing equal volumes of Reagent I and Reagent II and diluting the solution with two volumes of water. This molybdenum blue reagent is stable for months. NinhydTin Reagent The ninhydrin reagent was used to detect phospholipids containing free amino groups (e.g. PE, PS, and their lyso-derivatives). The reagent consisted of 0.2 % ninhydrin in butanol saturated with water. When sprayed with ninhydrin, lipids with a free amino group show up as red-violet spots upon heating. 2.2.12 Phosphorus Assay The amount of phosphorus was determined according to the Fiske and Subbarow method (1925). Reagent I was 70% perchloric acid, Reagent II contained 0.22% ammonium molybdate made up in 2% (vol/vol) concentrated H 2S0 4, and Reagent III was made up by dissolving 30 g of sodium bisulfite, 1 g of sodium sulfite, and 0.5 g of 1-amino-2-napthol-4-sulfonic acid at 40°C and was filtered after storing it overnight in the dark to remove crystals. The procedure for detecting the amount of phosphorus was as follows: Between 0.1 and 0.5 mol of phospholipid sample was hydrolyzed by addition of 0.5 ml of 45 perchloric acid and one hour digestion at 196°C. In order to prevent evaporation, the tubes were covered with glass marbles. When the hydrolysis was complete (i.e. solution was clear), the tubes were cooled and 14 ml of ammonium molydate reagent was added followed by 0.6 ml of the Fiske-Subarrow reagent and immediate mixing. Colour was allowed to develop by heating the tubes in a boiling water-bath for 15 min. The tubes were cooled and absorbance at 830 nm was recorded. A potassium phosphate solution (1 mM) was used to construct a standard curve. Fatty-Acid Analysis Fatty acid analyses were performed on methyl esters obtained by incubating the lipid samples in borane trifluoride (Christie, 1973). Briefly, approximately 10 mg of phospholipid was dried down in a screw-top tube and 4 ml of methanol and 1 ml of borane trifluoride was added. The tube was sealed under nitrogen and heated at 80°C for 15 min. The tube was cooled and the methyl esters extracted with 3 ml of pentane. The pentane extract was subsequently washed three times with water and then dried with a few grains of anhydrous magnesium sulfate. Fatty acid analyses were performed using a Hewlett Packard 7610A gas chromatograph fitted with a column containing 10% diethylene glycol succinate (DEGS) PS immobilized on 80/100 Supelcoport (Supelco, Inc.). It was operated in the program mode employing a temperature gradient from 150°C to 180°C. Rat liver fatty acids were used as standards. 46 2.2.14 Preparat ion o f 3 1 P - N M R samples and Exper imenta l Setup Samples for 3 1 P - N M R studies were prepared f rom appropriate mixtures o f phospho l ip id (50 /wmol total l ip id) in ch loroform. The ch lo ro form was evaporated under a stream o f nitrogen and the sample was then stored under vacuum for 2 h to remove any residual so lvent The l ip id was hydrated in 0.9 m l o f a buffer (10 % 2 H 2 0 ) containing 100 m M N a C l , 10 m M Hepes ( p H = 7.4) by vortex mix ing. Add i t ions o f C a 2 + and M g 2 * (chlor ide salts) were per formed by introducing appropriate al iquots f rom 100 m M stock solutions. 3 1 P - N M R spectra were obtained employ ing a Bruker W P - 2 0 0 Four ie r transform N M R spectrometer operat ing at 81.0 M H z . F ree induct ion decays were accumulated f rom up to 1000 transients employ ing a radiofrequency (rf) pulse width o f 11 ps,- a sweepwidth o f 20 or 50 kHz , a 0.8 s interpulse t ime and gated h igh power broad band proton decoupl ing. A n exponential filter corresponding to 50 H z l i ne -b roaden ing was appl ied pr io r to Four ie r transformation. 2.2.15 B i layer to Hexagona l H u Trans i t ion Temperature In order to determine the bi layer to hexagonal transit ion temperature of phosphat idylethanolamines, 3 1 P - N M R spectra were accumulated over the appropriate temperature range by employ ing the V A R T (variable temperature) program. F igure 7 and F igure 8 show the spectra obtained for the bi layer to hexagonal H n transit ion for soya and egg P E respectively. The i r transit ion temperatures were 15° and 30°C respectively. ~^ 20 0 20 ppm Figure 7. Bilayer to Hexagonal H M Transition of Soya Phosphatidylethanolamine. F igure 8. B i layer to Hexagona l H „ Trans i t ion o f Egg Phosphat idy lethanolamine. 49 2.2.16 D iva len t Cat ion B ind ing Determinat ions The divalent cation b ind ing determinations were per formed at 20°C employ ing equ i l ib r ium dialysis according to established protocols (Port is et al . , 1979). The extent o f Ca 2 * b ind ing was determined employ ing 4 i C a whereas the extent o f M g 2 * b ind ing was measured employ ing a Var ian Techtron A A - 5 atomic absorpt ion spectrophotometer. Fo r Ca 2 * b inding, 2 m l o f sonicated l ip id vesicles (1 ^ m o l / m l ) were dialysed (using Spectrapor 3. 18 mm) in 100 m l o f buffer (100 m M NaC l .10 m M H E P E S , p H = 7.0) ranging i n final ca lc ium concentrations f rom 0.1 to 10 m M . The samples also had the ca lc ium ionophore A23187 ( E L i l l y Co . Ltd.) at an ionophore to phospho l ip id molar rat io o f 1:100 and were also f reeze- thawed three t imes dur ing the 6 h dialysis, in order to ensure an equ i l ib r ium distr ibut ion o f the cation. Phospho l ip id associated Ca 2 * is expressed as molar ratio o f the cation to phospho l ip id . F o r the determinat ion o f M g 2 * b inding, a s imi lar protocol as above was fo l lowed wi th the fo l lowing modif icat ions. M o r e l i p id (10 jumo l /ml ) was used for dialysis. The amount o f bound M g 2 * was determined by centr i fuging the phospho l ip id vesicles at 12,000 g for 15 m i n in an Eppendor f f centrifuge and f reeze-dry ing the pe l l e t F r o m the wet and dry weights o f the pellet, the amount o f water and unbound M g 2 * was calculated. The dry pellet was resuspended by sonication i n 1 m l o f water and the amount o f phospho l ip id and magnesium was determined by the phosphorus assay and atomic absorpt ion respectively. The results were expressed as the molar rat io o f the bound cation to phospho l ip id . 50 2.2.17 F reeze-F rac tu re Studies Freeze- f rac ture studies were per formed employ ing a Balzers B A F 301 apparatus. Samples in the presence o f 25 % glycerol (by vol) were quenched f rom 20°C in l i qu id F reon . They were subsequently fractured at - 1 1 0 ° C and 2 x 10" 6 T o n and the fracture faces were repl icated by shadowing with p la t inum/carbon at a 45° angle fo l lowed by carbon shadowing (20 A " thick) to improve mechanical stabil ity. The replicas were cleaned i n Javex bleach and v iewed under a Ph i l ips 400 electron microscope. 2.3 R E S U L T S 2.3.1 Phase Preferences O f Phosphat idy l inos i to l The po lymorph ic phase preferences o f PI were investigated employ ing 3 1 P - N M R and f reeze-f racture techniques. Identif ications o f b i layer and hexagonal ( H u ) phase structures v ia 3 1 P - N M R rel ies on equivalent values o f the " r ig id latt ice" (no mot ion) chemica l shift anisotropy ( C S A ) tensors o f phospho l ip id diesters o f var ious phosphol ip ids. The 3 1 P - N M R spectrum o f the anhydrous sod ium salt o f PI is shown in F igure 9a and is effectively identical to the rigid lattice 3 1 P - N M R spectra o f P C (G r i f f i n , 1976), PS (Hope and Cu l l i s , 1980), phosphat idy lg lycerol (Farren and Cu l l i s , 1980) and sph ingomyel in (Cu l l i s and Hope, 1980) indicat ing a s imi lar conformat ion in the phosphate region for these different phosphol ip ids. O n hydrat ion (F igure 9c) a broad asymmetr ic n P - N M R spectrum with a l ow - f i e l d shoulder and h i g h - f i e l d peak is observed, which is characteristic o f l iqu id-c rys ta l l ine phosphol ip ids in a bi layer organizat ion (Cu l l i s and DeK ru i j f f , 1979). The absolute value o f the effective chemica l shift anisotropy (measured as the separation between the l ow - f i e l d shoulder and h i gh - f i e l d peak) is approx. 60 ppm which is approx. 15 - 20 p p m more 51 F igure 9. " P - N M R Spectra o f Soya PI: 81.0 M H z n P - N M R spectra o f soya PI at 30°C (a) in the anhydrous (sodium salt) fo rm and (c) fu l ly hydrated i n the presence o f excess aqueous buffer, (b) in the presence o f excess buffer and C a J + to obtain a C a 2 V P I rat io o f 5 mo l /mo l , and (d) i n the presence o f a 5 - f o l d molar excess o f Mg 2 * . The spectrum o f the anhydrous sod ium salt was obtained f rom 200 ^ m o l phospho l ip id employ ing a 50 k H z sweep width, a 20 s interpulse t ime and h igh power gated proton decoupl ing. The remain ing spectra were obtained employ ing a 50 >onol phospho l ip id , a 20 k H z sweep width, a 0.8 s interpulse time and gated proton decoupl ing. 53 than is observed for other phospho l ip id species. Th i s cou ld reflect a sl ightly different conformat ion and/or reduced mot ion in the phosphate region o f the PI polar headgroup. The addi t ion o f a five-fold molar excess o f C a 2 + or M g 2 * to the PI dispersion resulted i n precipitat ion o f the l ip id , but the 3 1 P - N M R spectra mainta ined the l ineshape consistent with lamel lar structure (F igure 9b,d). However , the spectra were somewhat broader and exhib i ted a more pronounced l ow - f i e l d shoulder (with Ca 2 * produc ing the largest effect), suggesting reduced local mot ion in the phosphate region. Such spectra are s imi lar to those observed for phosphol ip ids i n the gel phase (Cu l l i s and Hope , 1980). The effects o f Ca 2 * and M g 2 * on PI dispersions are comparable to the effects o f Ca 2 * on (unsaturated) phosphat idy lg lycero l systems (Far ren and Cu l l i s , 1980) and contrast with the inf luence o f C a 2 * on card io l ip in (Cu l l i s et a l . , 1978) and (unsaturated) phosphat id ic ac id (Verk le i j et al. , 1982; Fa r ren and Cu l l i s , 1983) for wh ich hexagonal H n phase structure can be induced. Fo r PS , on the other hand, Ca 2 * induces crystal l ine "cochleate" structure (Papahadjopoulos et al.. 1975). A n inabi l i ty o f Ca 2 * to induce such a crystal l ine structure for PI may arise i n part f rom the relatively unsaturated nature o f the acyl chains (16:0, 28% by w t ; 18:0, 11%; 18:1, 13%;18:2. 43%; 20:0, 5%). Al ternat ively, the inabi l i ty o f divalent cations to trigger H n phase format ion may be attr ibuted to the large size o f the inositol headgroup. 2.3.2 B i layer Stabi l izat ion Ef fect O f Phosphat idy l inos i to l Previous studies have shown that phosphol ip ids preferr ing the bi layer phase in isolat ion can stabil ize a net bi layer organizat ion in mixtures wi th unsaturated ( H n phase) P E (Cu l l i s and Hope , 1980; Cu l l i s et a l „ 1983). Soya PI exhibits a s imi lar abi l i ty at 30°C i n the presence o f soya phosphat idylethanolamine (which adopts the 54 H n phase above 15°C, see F igure 7) as indicated in F igure 10. The presence o f 10 mol% or 15 mol% PI results in e l iminat ion o f the hexagonal phase component as indicated by the absence o f the peak at 7 p p m which wou ld correspond to the low f ie ld peak aris ing f rom H u phase phosphol ip ids. Such PI concentrations result in a bi layer l ineshape on wh ich a narrow symmetr ic component indicat ive o f phosphol ip ids in structures a l lowing isotropic mot ional averaging is super imposed. F ina l l y , the presence o f 20 or 50 mo l% PI results i n b i layer 3 1 P - N M R l ineshapes where two distinct h i gh - f i e l d peaks are observed due to the the different values o f & ^ f o r p i ( A ^ = 6 0 ppm) and phosphat idy lethanolamine (A^jp=42 ppm). The peak at the higher f ie ld may be assigned to PI . 2.3.3 Character izat ion o f the "Isotropic" components o f P I - P E mixtures The or ig in o f the " isotropic" components observed at low PI concentrations cannot be ascertained by 3 I P - N M R . Possibi l i t ies include l ip id ic particles (Verk le i j et al., 1979b; D e K r u i j f f et al., 1978), otherwise characterized as intermembrane attachment sites (M i l l e r , 1980; H i r o and Stewart, 1981) wh ich appear to occur as intermediary structures between the bi layer and H n phases (Verk le i j et al., 1980). Al ternat ively, smal l (diameter < 200 nm) lamel lar systems can give rise to such spectra. F reeze- f rac ture visual izations can give less ambiguous indications o f the structures present, however, as indicated i n F igure 11 for systems containing 7.5 mol% PI (F igures l l a , b ) and 20 mo l% PI (F igures 11c). The 7.5 mo l% mixture reveals large l i p id structures with many l ip id ic particles distr ibuted across the fracture face. Al ternat ively, at 20 mo l% PI some smal l bi layer vesicles are apparent and l ip id ic particles are not observed. The isotropic spectral component may be attr ibuted to l ip id ic particles and associated macroscopic structures such as the "honeycomb" arrangement (Cu l l i s et al., 1980) at low (<10%) PI content, and to smal l 55 F igure 10. n P - N M R Spectra o f P I - P E : 81.0 M H z 3 1 P - N M R spectra obtained at 30°C f rom aqueous dispersions o f mixtures o f soya PI and soya phospha t idy l -ethanolamine, where PI represents 0, 5, 15, 20 and 50 mo l% o f the l ip id mixtures. Spectra were col lected as described i n the legend to F igu re 9. and methods. 40 0 - 4 0 p p m 57 F igure 11. F reeze- f rac ture micrographs o f P I - P E mixtures: F reeze- f rac ture micrographs o f aqueous dispersions o f ( A ) soya P E containing 7.5 mo l% PI, (B) the same fracture face at higher magnif icat ion, and (C) soya P E containing 20 mo l% PI. Samples were quenched f rom room temperature in the presence o f 25% glycerol. 58 59 vesicular structures at higher concentrations o f the bi layer preferr ing species. It may be noted that the ampl i tude o f the " isotropic" 3 1 P - N M R component in the systems containing 15 and 20 mol% PI was somewhat variable (compare spectra of F igures 10 and 12 ). Th is arises due to a variable generation o f smaller vesicular structures wh ich appear part icular ly sensitive to the extact PI concentration in this range. 2.3.4 Inf luence o f Ca l c i um It has been shown elsewhere that mixtures o f acidic phosphol ip ids such as PS (Verk le i j et al., 1980), card io l ip in ( D e K r u i j f f and Cul l i s , 1980) and phosphat idy lg lycerol (Farren and Cu l l i s , 1980) sensitive to the presence o f C a 2 * (and in some cases Mg 2 * ) which can trigger bi layer to hexagonal H u phase transit ions i n these systems. A s indicated i n F igure 12, Ca 2 * can also induce H n phase structure in (soya) P E systems where bi layer structure had been stabi l ized by the presence o f 15 or 20 mo l% PI . There are, however, some rather unique features. F irst, the presence o f low ca lc ium levels (Ca 2 */P I molar ratios R=0 .25 ) results i n the format ion o f structures a l lowing isotropic mot iona l averaging. These structures are not intermediates between bi layer and H n structures, however, as h igher levels o f ca lc ium (R=0 . 5 ) results in precipi tat ion o f the phospho l ip id dispersion and a 3 l P - N M R spectrum characteristic o f bi layer structure. Freeze- f racture studies were therefore per formed on the R = 0.25 samples, reveal ing the presence o f smal l apparently bi layer vesicles with an average diameter o f less than 200 nm. Thus the " isotropic" 3 1 P - N M R component may be attr ibuted to vesicular systems generated on addi t ion o f ca lc ium. The second feature o f interest concerns the h igh levels o f ca lc ium requi red to induce H n phase structure as revealed by 3 1 P - N M R . The addi t ion o f ca lc ium to attain R = 5.0 (equivalent to a ca lc ium concentration o f 30 m M ) results i n predominandy H n phase structure for the 15 mo l% PI sample. However , the presence o f such ca lc ium 60 F igure 12. Inf luence o f C a 2 + on P I - P E mixtures: 81.0 M H z " P - N M R spectra at 30°C obtained f rom aqueous dispersions o f soya P E containing 15 and 20 mo l% PI . The ratios R refer to the molar rat io of C a 1 + to PI where the divalent cation was added to the hydrated l i p i d systems as al iquots f rom a 100 m M stock solution. Other condit ions as for F igure 9. 61 15 mole % PI 2 0 mole % PI Molar ratio C a 2 " / PI 4 0 0 - 4 0 4 0 0 - 4 0 ppm ppm 62 levels i n the 20 mol% PI sample only results in a minor i ty (approx. 20%) H n phase component, the large majority (80%) o f the phospho l ip id remain ing in the bi layer organizat ion. A t higher PI concentrations ca lc ium d id not induce any H u organization. These results contrast strongly wi th the behaviour o f other acidic phospho l i p id - soya P E mixtures, where such ca lc ium concentrations can induce apparendy complete H n organization in systems containing 30 mo l% (soya) PS (Ti lcock and Cul l i s , 1981), 30 mol% (egg) phosphat idy lg lycerol (Farren and Cu l l i s , 1980), or 30 mo l% (beef heart) card io l ip in ( D e K r u i j f f and Cu l l i s , 1980). Th i s wou ld suggest that PI is a part icular ly effective bi layer stabi l iz ing agent i n the presence o f ca lc ium. 2.3.5 Inf luence O f Magnes ium The inf luence o f magnesium on m ixed PI - P E systems is i l lustrated i n F igure 13. Aga in , at low M g 2 * concentrations (R = 0.25) a large symmetr ic resonance indicat ing structures a l lowing isotropic mot ional averaging is observed, wh ich, by analogy wi th the Ca 2 * induced behaviour may be attr ibuted to a populat ion o f smal l lamel lar structures. A t h igher M g 2 * levels, however, M g 2 * does not induce H n phase components. Such behaviour is s imi lar to that observed i n (soya) P E - P S systems (Ti lcock and Cu l l i s , 1981) and indicates some specif icity in the C a 2 * - PI interaction. The different effects o f M g 2 * and Ca 2 * do not appear to be due to different aff init ies o f M g 2 * and Ca 2 * for PI, or, indeed, PS as indicated by the results o f equ i l i b r ium dialysis experiments presented in F igure 14. The b ind ing curves for M g 2 * and Ca 2 * are clearly s imi lar and do not explain the di f fer ing abil it ies o f both cations to induce H n phase structure at concentrations o f R = 5 . 0 , corresponding to cation concentrations o f more than 30 m M . Such concentrations are wel l above the concentrations at which max ima l b ind ing is observed ( 2 - 3 m M ) . It 63 F igure 13. Influence o f M g 2 * on P I - P E mixtures: 81.0 M H z 3 1 P - N M R spectra at 30°C obtained f rom aqueous dispersions o f P E containing 15 and 20 mo l% PI. The ratios R refer to the molar ratios o f M g 2 * to PI where the divalent cation was added to the hydrated l ip id systems as al iquots f rom a 100 m M stock solution. Other condit ions as for F igure 9. 15 mole°/ 0 PI 20 mole % PI Molar ratio M g 2 +/?\ ppm Ppm 65 F igure 14. D iva len t cation b ind ing curves o f PI and PS vesicles. The ratio o f bound cat ion/phospho l ip id fo l lowing equ i l ib r ium dialysis against various concentrations o f C a 2 * or Mg 2 * . ( • ), Ca 2 * and phosphat idylser ine; ( D ), Ca 2 * and phosphat idy l - inos i to l ; ( # ), M g 2 * and phosphat idylser ine; ( O ). M g 2 * a n d PI. Exper imenta l condit ions are descr ibed i n Mater ia l s and Methods . 1 2 3 4 5 6 7 8 9 10 C a 2 + concentrat ion (mM) would appear that the nature o f the C a 2 + - PI and M g 2 + - PI complexes formed are d i f ferent One possibi l i ty may be that the ca lc ium complexes o f PI tend to segregate more readily, thus effectively reducing the distr ibuted concentration o f the bi layer stabi l iz ing PI species. 2.4 D I S C U S S I O N 2.4.1 Roles O f Phosphat idy l inos i to l The object o f this study was to characterize the structural preferences o f PI i n pure and m ixed mode l systems. It has been shown that PI prefers a lamel lar organizat ion on hydrat ion as indicated by 3 1 P - N M R . A lso, whereas both ca lc ium and magnes ium precipitate these l ip id dispersions and induce effects consistent with reduced local mot ion i n the phosphate region, the 3 1 P - N M R results indicate that bi layer structure is maintained. Further , in m ixed systems wi th soya PE , PI is a part icular ly effective agent for stabi l iz ing bi layer structure at 15 mo l% and higher concentrations, inducing " intermediary" structures such as l ip id ic particles at lower levels. F ina l ly , in systems containing 20 mol% or less PI ca lc ium (but not magnesium) can trigger H n phase format ion. These observations can be discussed i n terms o f roles o f PI as related to the P I -e f fec t as wel l as the mechanism whereby ca lc ium induces structural alterations i n these systems. The predi lect ion o f PI for the bi layer organizat ion both in the presence and absence o f ca lc ium argues against a dynamic role o f PI per se i n ca lc ium transport. Th is is in contrast to the phosphat id ic acid generated dur ing the "PI - response" , wh ich can undergo structural reorganization i n the presence o f divalent cations (Papahadjopoulos et al. , 1976b; Verk le i j et al . , 1982; Far ren et al . , 1983) and has been impl icated as a ca lc ium ionophore (Tyson et a l . , 1976) as further explored 68 i n the next chapter. Th i s wou l d be consistent w i th a ro le o f P I i n v i vo wh i c h is p r imar i l y structural (serving to ma in ta in an intact permeab i l i t y barr ier) but w i th the added feature that enzymat ica l ly generated der ivat ives can p lay dynamic roles i n transbi layer t ranspor t The mechan i sm whereby ca l c ium induces (part ia l) b i layer to hexagona l H u transit ions i n (soya) P E systems where b i layer organizat ion is stabi l i zed by P I is o f interest as it appears to proceed v ia a d i f ferent mechan i sm than ca l c ium induced b i l a ye r -hexagona l transit ions i n P E systems stabi l ized by other ac id ic phospho l ip ids . In systems stab i l i zed by card io l ip in ( D e K r u i j f f and Cu l l i s , 1980) and phosphat id i c ac id ( Fa r r en et a l . , 1983) for example, ca l c i um triggers H n phase fo rmat ion by convert ing the b i layer stabi l i z ing species to a species pre fe r r ing the H u organizat ion. Th i s contrasts w i th systems stabi l i zed by PS , where ca l c i um segregates the P S into (anhydrous) crystal l ine domains, a l l ow ing the P E to revert to the H n phase it prefers i n iso lat ion (T i l cock and Cu l l i s , 1981). A l ternat ive ly , i n P E systems stab i l i zed by up to 30 mo l% (unsaturated) phosphat idy lg lycero l , ca l c i um appears to redu'ce the b i layer stabi l i z ing capacity o f this ac id ic phospho l i p i d result ing i n a d irect incorporat ion o f P G in to the H n phase matr i x ( Fa r r en and Cu l l i s , 1980). Howeve r , the behav iour o f systems stabi l i zed by P I suggest that none o f these mechan i sms apply. F o r example, ca l c i um does not induce crysta l l ine complexes i n pure P I systems, and therefore does not behave i n a comparab le manner to P S . A l ternat ive ly , there is no real ev idence that P I actual ly enters the Hn phase matr ix , as the system conta in ing 15 mo l% PI i n the presence o f excess ca l c i um (F igu re 12) st i l l appears to have a res idual "b i l ayer " 3 1 P - N M R componen t Th i s suggests that to a l im i ted extent ca l c ium is ab le to segregate P I i n these m i x ed systems, where the P I - C a J + aggregates r ema in i n a hydrated lame l la r structure. The results presented here are also consistent w i th observat ions regard ing the 69 fusogenic capabilities of PI in pure and mixed systems (Sundler and Papahadjopoulos, 1981; Sundler et al. , 1981; Sundler and Wijklander, 1983). Pure PI vesicles were resistant to fusion in the presence of calcium or magnesium even at very high levels (50 mM) in spite of aggregation induced by both cations. In mixed systems containing a fusogenic lipid such as PS, PI inhibits the Ca2*-induced fusion of the vesicles despite similar aggregation rates. The same inhibitory effect on fusion was also demonstrated for PI-PE systems where increasing the PI content progressively inhibited fusion. 70 C H A P T E R III Ro le o f Phospha t i de A c i d ( PA ) as a Ca l c i um Ionophore In Large Un i l ame l l a r Vesicles ( L U V ' s ) . 3.1 I N T R O D U C T I O N F r o m studies concerning the phase preferences o f PI (chapter two), i t is evident that PI is an effective bi layer stabi l iz ing agent wh ich maintains a bi layer organizat ion in the presence" o f divalent cations. "THence, PI in v ivo appears to be invo lved i n mainta in ing an intact permeabil ity" barr ier for the membrane. However , the products o f PI generated dur ing the PI - response offer possibi l i t ies for ca lc ium translocation. Phosphat id ic "acid ( PA ) is a key intermediary l ip id i n the P I - cyc l e wh ich is generated by phosphory lat ion o f diglyceride. Th is chapter investigates the role o f P A as a ca lc ium ionophore in large uni lamel lar vesicles ( L U V s ) . A n understanding o f the possible propert ies o f phosphat idic acid in m ixed l i p id systems clear ly requires an understanding o f the physical propert ies o f P A in isolat ion and the inf luence o f factors such as divalent cations and p H . These have been recently investigated by 3 1 P - N M R and freeze-fracture techniques (Farren et al . , 1983). It is useful to br ief ly review the results obtained. D i o l eoy l phosphat idic acid ( D O P A ) was found to adopt the hexagonal H u phase as the p H is reduced below 5.3. In the presence o f low levels o f magnes ium and calc ium, the p H at wh ich the H n organizat ion is observed is increased to 6.0. H ighe r levels o f ca lc ium and magnesium resulted i n a disorganized structure, wh ich was possibly an intermediate between bi layer and hexagonal H n phases. These observations were in agreement wi th freeze-fracture studies o f Verk le i j and coworkers (1982). The abi l i ty 71 o f divalent cations to inf luence the structural preferences o f phosphat id ic acid is o f interest part icular ly i f l ip ids present in " inverted" structures such as the hexagonal H u phase or inverted micel lar structures can reside wi th in the hydrophob ic domain o f an otherwise bi layer membrane. Such structures have been speculated to be o f funct ional signif icance in processes such as transbi layer transport o f ions (Cu l l i s et a l . , 1983). Furthermore, phosphat id ic acid has been shown to effectively sequester Ca 2 * into the organic phase (Tyson et al. , 1976; Cu l l i s et_ a l . , 1980). Th i s later study il lustrates a remarkable correlat ion between the abi l i ty o f divalent cations to induce the inverted hexagonal H n phase for aqueous l i p id dispersions and the abi l i ty o f such l ip ids to facil itate uptake o f Ca 2 * into an organic phase. These observations agree with the propos i t ion that any compound exhib i t ing ionophoret ic abi l i t ies in a bio logical membrane must also have the capabi l i ty to fo rm a l i p i d - so lub l e complex wi th the, agent to be transported. Fur thermore, the ionophore should be able to translocate the ; agent across an intact bi layer and release it into a separate aqueous compartment , ; .This abi l i ty has been hinted at by studies wi th mul t i lamel lar vesicles ( l iposomes). Sehran et al . (1981) demonstrated that P A -containing l iposomes containing the Ca 2 * sensitive metal lochromic dye, Arsenazo III, permit ted translocation o f C a 2 * across the bi layer. The purpose o f the studies presented i n this chapter was to investigate the ionophoret ic properties o f phosphat id ic acid in a more precisely def ined mode l membrane system which more accurately reflects the biological situation, that is, a large un i lamel lar vesicle. A lso , f rom electrophysiological studies, it is clear that ion gradients and membrane potential are important factors in transport processes. Therefore, their effects on ca lc ium mobi l i zat ion were studied in order to ascertain their ro le in the molecular mechanism of ion transport The results were compared to 72 studies ut i l iz ing a wel l known ca lc ium ionophore (A23187) i n L U V systems wh ich d id not contain PA . 3.2 M A T E R I A L S A N D M E T H O D S 3.2.1 Synthesis o f D io l eoy l Phosphat idy lchol ine ( D O P C ) D io l eoy l phosphat idy lchol ine was synthesized according to the procedures o f Warner and Benson (1977). The method a l lowed synthesis on a m i l l imo la r scale under m i l d condit ions and d id not require large excess o f the fatty acid acylat ing reagent The procedure invo lved prepar ing o f L - c - glyceryl phosphory lcho l ine f rom egg phosphat idy lchol ine and subsequently reacylating with oleic acid imidazol ide. Preparat ion o f L - a-glyceryl phosphory lcho l ine ( G P C ) L-a - g l y c e r y l phosphory lcho l ine was obtained i n a one-s tep procedure by alkal ine deacylat ion o f pur i f ied egg phosphat idy lchol ine (Brockerhof f and Hu rkowsk i , 1965). Egg phosphat idy lchol ine (10 g, 8 mmols) was dissolved in 100 m l o f ethyl ether i n a 200 m l glass centrifuge bottle. Unde r cont inuous st irr ing, 10 m l o f tetrabuty lammonium hydrox ide (25% in methanol , Eastman Kodak a r t 7774, Rochester, N .Y . ) was added dropwise. The reaction mixture becomes turbid and yel lowish. St i rr ing was cont inued for 1 h at room temperature. The mixture was centr i fuged at 1000 g for 10 min . to obtain a glassy precipitate o f G P C wh ich is the only insoluble p roduc t It was dissolved in approximately 5 m l o f methanol with warming and r e - -precipitated with 100 m l o f ether. Th i s was repeated three times and resulted i n pure G P C . The product was dissolved i n methanol and stored at - 2 0 °C . The y ie ld o f the reaction was approximately 90% on the basis o f the total phosphorus. 73 Preparat ion o f D io l eoy l Phosphat idy lchol ine ( D O P C ) G P C (1.3 g, 5 mmols) was dr ied down in a 500 m l round bottom flask and stored under vacuum overnight to ensure that there is no residual solvent l e f t Sod ium metal (2 g) was dissolved in 150 m l o f dry d imethy l sulfoxide ( D M S O ) overnight in a second 500 m l round bottom flask equipped wi th a drying tube. The dry G P C was dissolved i n 120 m l o f D M S O by sonicat ion in warm water. Fatty acid imidazol ide was prepared by quick ly adding carbony ld i imidazo le (A ld r i ch Chemica l Co. , M i lwaukee , W l ) (3.57 g, 22 mmols) to 20 m l o f dry tetrahydrofuran ( T H F ) containing oleic acid (S igma, SL Lou is , M O ) (5.65 g, 20 mmols) . The flask was f lushed wi th nitrogen and st irred for 45 m in over a drying tube. U p o n complet ion, the solvent ( T H F ) was removed by f lash-evaporat ion and the result ing fatty acid imidazol ide dissolved i n 20 m l o f D M S O . F o r the acylat ion reaction, the carbony ld i imidazo le oleic acid was added to the G P C - D M S O solut ion and the flask f lushed wi th nitrogen and equipped wi th a dry ing tube. Sod ium D M S O (72 ml ) was added wi th a 10 m l pipette under v igorous stirr ing. The flask was again f lushed wi th nitrogen and the reaction was a l lowed to proceed for 10 min. The reaction was terminated by cool ing the reactants i n an i ce -water bath and adding 200 m l o f co ld water and approx imate ly 10 m l o f 4 N H C I to adjust the p H to 5. The ac id i f ied reaction mixture was qu ick ly extracted three times with 200 m l o f ch loro form/methano l (2:1) and the poo led extract was subsequentiy washed four times with methanol/water (1:1). The solvent o f the washed extract was dr ied down to a reddish residue f rom which D O P C was pur i f i ed ut i l iz ing the H P L C procedure (section 2.2.3). A n y lyso P C was easily removed by carboxymethy l - cel lulose chromatography. The y ie ld o f this acylat ion reaction was typical ly around 40%. The reaction scheme is i l lustrated i n F igure 15. 74 R A Q H + f j A ^ y J » U R JV> + c o . Oleic acid Carbonyldiimidazole ^ „ . . . ., ... Fattvacid immidazolide O O II r O H + - A 2 R N N O + - ) - 2 N a C H C H \ — / LO-P-O-CH 2CH 2N (CH 3) 3 , OH L-a-glyceryl phosphoryl choline O DMSO 9 r O A R +" • R"ND-J O , -L - 2 N a N N L 0 - P - 0 - C H C H r N ( C H ) N ' OH J O D O P C « + 2 C H 3 - ^ C H 3 Figure 15. Reaction scheme for synthesis of Dioleoyl Phosphatidylcholine. 75 3.2.2 Synthesis o f D io l eoy l Phosphat id ic A c i d ( D O P A ) D io l eoy l phosphat idic acid ( D O P A ) was prepared by using phosphol ipase D to specif ical ly hydrolyze the chol ine moiety o f D O P C . The procedure was exactly as described i n section 2.2.5 with the modi f i cat ion that the aqueous phase o f the reaction contained 0.2 M sod ium acetate p H 5.6 and 100 m M ca lc ium chlor ide. The convers ion to D O P A was almost quantitative and completed wi th in 30 m in . The D O P A f rom the reaction was predominant ly in the ca lc ium salt form. D u e to solubi l i ty problems in ch l o ro fo rm-methano l -wa te r mixtures, it was impract ica l to convert it to the sod ium salt fo rm via the B l i g h - D y e r procedure. A n alternate method o f Treve lyan (1966) was found to be excellent for the sod ium salt conversion. D r y D O P A was dissolved in ch lo ro fo rm-methano l (2:1) and the solut ion was washed wi th 0.25 vo lume o f 0.1 M citrate buffer (2 moles N a salt : 1 mole acid, p H 5.2) to chelate the ca lc ium and form the sod ium salt o f phosphat id ic a c i d Citrate was subsequentiy washed out o f the lower cho lo ro form phase by addit ion o f the theoretical upper phase containing ch lo ro fo rm-methano l -wa te r (3:48:47) (Fo l ch et al., 1957). The wash was repeated three times to ensure that a l l the citrate was removed. The N a * - D O P A was dr ied down and pur i f ied by ca rboxymethy l - cel lulose chromatography as described by Comfu r i u s and Zwaa l (1971) ut i l iz ing a continuous ch lo ro fo rm-methano l gradient ( 0 -50%) rather than by a s tep -g rad ien t The D O P A was pur i f i ed to >99% as indicated by phosphorus on two-d imens iona l T L C . 3.2.3 Preparat ion o f Large Un i l ame l l a r Vesic les ( L U V s ) Recent ly, much attention has been devoted to the technology o f constructing large uni lamel lar vesicles ( L U V s ) . There are basical ly three stringent cr iter ia to be satisfied. That is, that the mode l membrane be closed, uni lamel lar , and 76 reasonably large enough to enclose an appreciable trapped vo lume and avo id any h igh ly curved areas wh ich may lead to packing problems and disorder in the hydrocarbon region (Sheetz and Chan , 1972). Two popular L U V systems are the reverse-phase evaporation L U V s (called R E V s ) (Szoka and Papahadjopoulos, 1978) and the o c t y l -glucoside L U V s prepared by detergent dialysis ( M i m m s et al.,1981). Both involve the use o f l i p id solubi l iz ing agents (organic solvents or detergents) and have various drawbacks. F o r example, lengthy dialysis procedures and ge l - f i l t ra t ion techniques are often required which never completely remove the solvent or detergent employed. Further, a variety o f protocol modif icat ions are usual ly required depending on the l ip id species employed because o f their l im i ted solubi l ty in certain organic solvents (e.g. cholesterol, phosphat idy lethanolamine i n ether or ethanol) and inadequate solubi l izat ion o f charged phosphol ip ids in non - i on i c detergents such as o c t y l -glucoside. Therefore, a technique very recendy developed by Hope and coworkers (1983) for the rap id product ion o f L U V s by repeated extrusion under moderate pressures ( < 500 psi) o f mul t i lamel lar vesicles through polycarbonate filters (100 n m pore size) was ut i l ized. Th i s procedure results i n un i lamel lar vesicles with diameters in the range 80-120 nm and with trapped vo lume in the region o f 1-2 ^1/yumol phospho l ip id . Th i s procedure has many advantages, inc lud ing the absence o f organic solvents or detergents, l ittle restriction regarding the l ip id composi t ion and concentrat ion (up to 200 j umo l /m l can be employed), and h igh trapping efficiencies (approx. 30%) that cou ld be achieved. F ina l ly , the procedure is extremely rapid ( < 15 m i n preparat ion t ime). The procedure for generating these " L U V E T s " (large un i lamel lar vesicles by extrusion techniques) is basical ly an extension o f the work o f O l son et al . (1979). They showed that the trapped volume o f large mult i lamel lar vesicles cou ld be 77 increased by extrusion under relatively low pressures (< 80 psi) through polycarbonate filters of 200 nm pore size. These vesicles were oligolamellar with a reasonably homogenous size distribution centered about the pore size of the polycarbonate filter. The LUVET preparation is as follows. Large multilamellar vesicles (LMVs or liposomes) were prepared by vortexing a dry film of lipid in a test-tube in the presence of a desired aqueous buffer. The resulting LMV dispersion was transferred into a device which extruded the liposomes through two stacked 25 nm diameter polycarbonate filters with 100 nm pore size (Nucleopore Corp., Pleasanton, CA). The extrusion device is marketed by Sciema Technical Services Ltd. (Unit 2, 2900 Simpson Rd., Richmond, Canada V6X 2P9). The LMVs (25 ,umol phospholipid/ml) were extruded by applying nitrogen pressure (100-300 psi) via a standard gas cylinder fitted with a high pressure (0-4,000 psi) regulator. The vesicles were collected and re-extruded a total of ten times to ensure reasonable homogeneity in size and unilamellarity. Occasionally, a freeze-thaw step was included in order to fuse the vesicles and form larger systems according to Pick (1981). The 100 nm LUVETs were freeze-thawed and passed through the extrusion device 10 times either employing 200 nm or 100 nm pore-size filters. This resulted in larger vesicles as indicated by freeze-fracture and trap-volume measurements. 3.2.4 Determination of Trapped Volumes To determine trapped volumes, 2 y.Ci/m\ of "Na or 1 4C-inulin (NEN, Canada) was added to the buffer in which the multilamellar vesicles were prepared. An aliquot (100 of the LUVETs was then loaded onto a 1 ml disposable syringe column packed with AcA34 Ultrogel (LKB Chemicals, Canada). The vesicles were eluted by washing the column with 350 /*1 of buffer. This was sufficient to remove all the external "Na or "C-inulin and aliquots of the eluted vesicles were assayed for lipid phosphorus 78 (Bottcher et al .. 1961, section 2.2.12); trapped " N a was determined employ ing a Beckman 800 gammacounter and trapped 1 4 C - i n u l i n was estimated using a Ph i l ips P W - 4 7 0 0 l iqu id scint i l lat ion counter. Trapped volumes were expressed as ^1 o f trapped vo lume per ^ m o l o f phospho l ip id . 3.2.5 Determinat ion o f Leakage In order to determine whether the vesicle preparations were leaky, retention o f trapped 2 2 N a inside the L U V E T s was monitored. L U V E T s with trapped 2 2 N a were prepared as described in the preceding section and the external 2 2 N a was removed by passing the vesicles on a 1.5 X 15 cm Sephadex G - 5 0 co lumn. A t various t ime intervals, al iquots (100 ju\) o f the L U V E T s were eluted by ut i l iz ing the 1 m l syringe columns packed wi th G - 5 0 by washing the co lumn wi th 300 o f buffer. The trapped 2 2 N a and l i p i d phosphorus was determined and the specif ic activity ( c p m / ^ m o l phosphol ip id) calculated. The results were expressed as the percentage o f 2 2 N a trapped inside the L U V E T s as a funct ion o f time. These leakage experiments were per formed on vesicles prepared at p H 9.0 (wi th 20 m M C H E S ) and at p H 7.4 (with 20 m M H E P E S ) containing 100 m M K + - o x a l a t e and 50 m M KC1 and a trace o f 2 2 N a . The vesicles were separated on the m in i - c o l umns at various time points and the amount o f trapped 2 2 N a was determined. The results for the 100 nm 20% P A L U V E T s imp l ied that over the time course o f the experiments (4 hours), they were non- l eaky ( < 1 0 %) to 2 2 N a at both p H 7.4 and 9.0 i n the presence or absence o f 2 m M calc ium. Th i s conf i rms the t ightiy sealed nature o f the vesicles employed in the C a 2 + - uptake experiments. 79 3.2.6 F reeze -F rac tu re Studies F reeze- f rac ture techniques were employed for s iz ing the vesicle preparations and for determining any ol igolamel lar vesicles in the L U V E T preparations. G l yce ro l (25% by vo lume) was mixed with the L U V E T s and the preparat ion quenched f rom room temperature into a freon slush. Samples were fractured and repl icated in a Balzers B A P 400D apparatus, and micrographs o f the replicas obtained by using a Zeiss 200 electron microscope. Vesic le size distr ibutions were determined by measuring the diameter o f fractured vesicles that were 50% shadowed according to the procedure o f van Venet ie and coworkers (1980). The occurrence o f o l igolamel lar vesicles in the preparations was detected by expressing the percentage o f cross-fractures through more than one bi layer in the vesicles. Accord ing to M i l l e r (1980), only a low percentage o f cross-fractures (5-10%) is detected in a mult i lamel lar preparat ion. Thus detection o f a low percentage o f cross-fractures (5%) cou ld imp ly the presence o f a signif icant number o f o l igo lamel lar vesicles in the preparat ion. Therefore, correlat ion between the size measurements and the trap vo lume determinations and detection o f < 1 % o f cross-fractures were taken as the surest indicat ion o f uni lamel lar i ty . F igure 16 shows freeze-fracture micrographs o f 100 nm sized L U V E T s prepared at pH ' s 6.0, 7.4 and 9.0. The uni lamel lar i ty o f the vesicles employed was determined by calculating the percentage o f cross-fractures in the L U V E T preparations i l lustrated in F igure 16. Less than 1% cross-fractures were found in al l three preparations at various pH ' s imp ly ing that most o f the L U V E T s were uni lamel lar . The close correlat ion between the size o f the vesicles and their trap vo lume as determined with l 4 C - I n u l i n also conf i rmed their uni lamel lar i ty. 80 F igure 16. Freeze- f rac ture micrographs o f 20 mo l% D O P A L U V E T s . 20 % D O P A L U V E T s (DOPC/DOPA/Cho l e s t e r o l ; 80 :20 :10 ) were prepared having 100 m M K*-oxalate/50 m M KC1/20 m M Buffer as the internal med ium and 175 m M KC1/20 m M Buf fer as the external med ium. L U V E T s at p H 6.0 has M E S buffer, those at p H 7.4 had H E P E S buffer and p H 9.0 L U V E T s had C H E S buffer. The L U V E T s were prepared by extrusion through the 0.1 j um fi lter as described i n the materials and methods section. The white bar corresponds to 200 nm. 82 3.2.7 Calcium Uptake Experiments For direct measurements of calcium uptake, 4 5Ca (NEN, Canada) was utilized. Typically, the LUVETs employed in the Ca2"-uptake experiments were prepared by extrusion through the 100 nm pore-size filters. After testing LUVETs of various lipid compositions, the DOPC/DOPA/Cholesterol (80:20:10) were found to form the most stable vesicles and give the most consistent results. Two types of control samples were employed. These either contained no acidic phospholipids or 20 mol% egg PS. LUVETs were formed in 125 mM potassium chloride and 20 mM potassium phosphate, at pH 7.4 and the external phosphate buffer was removed either by dialysis against 125 mM sodium chloride or potassium chloride solution buffered with 20 mM HEPES, pH 7.4, or was exchanged with a desired external buffer by passage over a 1.5 X 15 cm Sephadex G-50 desalting column. For the uptake experiments, LUVETs at a concentration of approx. 5 ^ umol phospholipid/ml were used. Additions of ionophores such as valinomycin and carbonyl cyanide-m-chlorophenylhydrazone (CCCP) at 1 ^ g/^mol phospholipid and 20 final concentration, respectively, were performed by addition from a concentrated ethanol solution. The total amount of ethanol added to the vesicle suspension was less than 1 % by volume. Control experiments did not reveal any influence of ethanol at these levels. Valinomycin (Sigma, SL Louis, MO) was used as a K* ionophore and CCCP was utilized as a H* ionophore. Whenever the calcium ionophore A23187 was utilized it was either preincorporated into the vesicles by addition during the lipid sample preparation in chloroform, or added externally from a concentrated ethanol solution. Both procedures yielded similar results. Calcium uptake was monitored by incubating the LUVETs (1 ml) in a buffer containing 2 mM final concentration of calcium chloride containing a trace of < 5Ca (2 p Ci). Aliquots (100 p\) were taken over a four hour time period and the external 83 F igure 17. C a 2 + - u p t a k e in 20 mol% D O P A L U V E T s moni tored by phosphorus assay and 3 H - D P P C as the l ip id marker. D O P C / D O P A / C h o l e s t e r o l (80:20:10) L U V E T s were sized through the 0.1 yum filters and prepared in 125 m M K.C1/20 m M potassium phosphate, p H 7.4. the external med i um was exchanged for 125 m M KC1/20 m M H E P E S , p H 7.4 on a Sephadex C—50 desalt ing co lumn. Ca l c i um chlor ide was added to a final concentration o f 2 m M f rom a stock solut ion o f 200 m M containing a trace o f 4 5 C a . The L U V E T s were separated f rom the incubat ion med i um as described i n the Mater ia l s and M e t h o d section. Amoun t o f phospho l ip id was determined either by the phosphorus assay ( O • A ) or by incorporat ion o f 3 H - D P P C into the L U V E T s ( • , A ) and employ ing the 3 H / 4 5 C a dua l - l abe l program on the l iqu id scint i l lat ion counter. C a 2 + -uptake was per formed in the absence ( O . • ) and m the presence o f va l inomyc in ( A . A ) a t a concentration o f 1 yug//i>mol phosphol ip id . 85 Ca 2 * was removed on a smal l 1 m l co lumn containing Sephadex G - 5 0 and equi l ibrated with the external buffer. The vesicles were eluted either by spinning the co lumn for 5 m in at 500 g on a bench- top centrifuge or by adding 350 JA\ o f co ld buffer. The vesicles were analyzed for " C a on a Ph i l ips P W - 4 7 0 0 l iqu id scint i l lat ion counter and for l i p id phosphorus (Bottcher et al ., 1961, section 2.1.12). Dua l - l a b e l experiments were also used to monitor ca lc ium uptake by ut i l iz ing 3 H - d i p a l m i t o y l phosphat idy lchol ine ( 3 H - D P P C ) ( f rom N E N , Canada). A trace o f 3 H - D P P C was added to the l ip id sample dur ing its preparat ion and both the amount o f phospho l ip id and ca lc ium were monitored simultaneously on the Ph i l i ps P W - 4 7 0 0 l i qu id scint i l lat ion counter employ ing a 3 H / , 5 C a dual label channel. Results o f Ca 2 *uptake experiments were expressed as nmols o f ca lc ium accumulated per ^ m o l o f phospho l ip id . F igure 17 illustrates that identical results for ca lc ium uptake were obtained by employ ing either o f the above procedures. 3.3 R E S U L T S 3.3.1 Ionophoret ic capabil it ies o f P A in L U V s The first question addressed was to ascertain whether P A cou ld act as a Ca 2 * ionophore i n a membrane. The prob lem was approached by moni tor ing the in f lux o f Ca 2 * into large un i lamel lar vesicle ( L U V E T ) systems in the presence and absence o f P A . In control L U V E T s , 20 mo l% PS was ut i l ized instead o f the P A so that both L U V E T s wou ld have the same mo l% o f acidic phosphol ip ids. The results are i l lustrated i n F igure 18. The L U V E T s employed had 125 m M KC1/20 m M K*-phosphate , p H 7.4 trapped inside the vesicle and 125 m M KC1/20 m M H E P E S , p H 7.4 as the exterior med ium. Ca 2 *-up take was only observed in L U V E T s containing PA . Th is suggests that P A has ionophoret ic capabil it ies. In the control 20% PS L U V E T s , no 86 F igure 18. C a 2 + - u p t a k e in 20 mol% D O P A and 20 mo l% PS L U V E T s . The 20% D O P A L U V E T s (DOPC /DOPA /Cho l e s t e r o l ; 80 : 20 : 10 ; A ) and the 20% PS L U V E T s (DOPC/PS/Cho les te ro l ;80 :20 :10; • ) were prepared as described in the legend to F igure 17. The internal med ium was 125 m M KC1/20 m M K*-phosphate, pH7.4 and the external med i um was 125 m M KC1/20 m M H E P E S , p H 7.4. 88 signif icant uptake was observed even though PS binds C a 2 + as avid ly as P A (Papahadjopoulos, 1968). It shou ld be noted however, that the rat io o f the amount o f Ca 2 *-up take to the avai lable amount o f phosphate (Pi) inside the vesicle is low (Ca 2 */P i = 0.3 ). A lso , the ratio o f Ca 2 * to P A is low ( C a 2 V P A = 0.05 ). Th i s raises two quest ions:( l) Is the phosphate trap an eff icient "Ca 2 * - s i n k " ? and (2) Is Ca 2 * s imply b ind ing to the P A molecules ? These questions were investigated by varying the amount o f phosphate inside the vesicles and also by employ ing other Ca 2 * chelators as Ca 2 * - s i nk s (see section 3.3.3). F igure 19 shows the levels o f Ca 2 * -up take attained inside the 20% P A L U V E T s having no phosphate, 20 m M phosphate and 40 m M phosphate. Ca 2 *-up take var ied according to the amount o f phosphate trapped inside the vesicles. Th is suggests that Ca 2 * is not just b ind ing to the P A molecules. A l so as noted in the next section, the observation that the amount o f Ca 2 *-up take increases when the L U V E T s are separated on a co lumn equi l ibrated wi th co ld Ca 2 * impl ies that Ca 2 * is not s imply b ind ing to the P A molecules. 3.3.2 Levels o f P A required for Ca 2 * -up take The next question addressed was what levels o f P A are required for t ransport A s shown in F igure 20, incorporat ion o f increasing amounts o f P A into the L U V E T s results in increasing amounts o f Ca 2 *-up take . A l l the vesicles had the same amount o f phosphate trapped inside. The levels o f Ca 2 * -up take are not l inear with the amount o f P A incorporated in the vesicles. Instead, there appears to be a threshold level o f P A that is required for opt imal uptake in the vesicle systems employed. These results are not yet understood, but cou ld suggest that more than one P A molecule participates in the transport process or that h igher P A levels lead to a 89 F igure 19. Ca 2 *-up take in 20 mo l% D O P A L U V E T s with varying amounts o f trapped phosphate. The 20 mo l% D O P A L U V E T s ( D O P C / D O P A / C h o l e s t e r o l ; 80:20:10) were prepared as described i n the legend to F igure 17. ( A ), L U V E T s without trapped phosphate (125 m M KC1/20 m M H E P E S , p H 7.4); ( O X L U V E T s with 125 m M KC1/20 m M potassium phosphate, p H 7.4 trapped and 125 KC1/20 m M H E P E S , p H 7.4 untrapped; ( • ), L U V E T s with 120 KC1/40 m M potass ium phosphate, p H 7.4 trapped and 120 KC1/40 m M H E P E S , p H 7.4 untrapped. 91 F igure 20. C a 2 + - u p t a k e in L U V E T s containing different concentrations o f D O P A . The L U V E T s were prepared as described in the legend to F igure 17. In the 5 and 15 mol% D O P A L U V E T s 15 and 5 mo l% PS was added respectively so that a l l the samples had 20 mo l% acidic phosphol ip ids. The internal med ium was 125 m M K.C1/20 m M K~-phosphate, p H 7.4 and the external med ium was 125 m M N a C l / 2 0 m M H E P E S , p H 7.4. The results were s imi lar when K C 1 / H E P E S was used as the external med ium. 93 more effective trap by formation of a Pi-Ca 2*-PA complex, for example. 3.3.3 Effectiveness of the Ca2*-sink The effectiveness of the Ca2*-sink employed in the uptake experiments was determined by employing two additional Ca2* chelators, namely EGTA and oxalate. Studies were performed on vesicles containing 20 mol% PA which were sized through the 100 nm filters. The trap volume was 1-1.2 yul/^mol phospholipid. Vesicles were made in 50 mM KC1/20 mM HEPES, pH 7.4 in the presence and absence of either 100 mM IC-oxalate or 100 mM EGTA. The external medium was exchanged over a Sephadex G-50 column. Attempts at making vesicles in higher amounts of phosphate buffer were unsuccessful. This is possibly due to an ill-defined interaction between the phospholipid and phosphate such as has been reported by other workers (Boskey and Posner, 1976; Fraley et al.. 1980). Figure 21 illustrates Ca2*-uptake when oxalate was employed as the Ca2* sink. In the absence of oxalate, very little Ca2*-uptake was observed as compared to when 100 mM K*-oxalate was present The ratio of Ca2*/oxalate = 0.6 at 4 h and approx. 1.0 at 10 h (not shown) while the ratio of Ca2*/PA = 0.3 and 0.5 at 4 and 10 h respectively. When EGTA, a very effective calcium chelator, was employed as the Ca2*-sink (100 mM EGTA, pH 7.4 [KOH] ), results similar to the oxalate experiments were observed. This is shown in Figure 22. The EGTA Ca2*-sink is clearly much more effective than the oxalate trap. This is illustrated by the rapid rate of Ca2*-uptake where the Ca 2*/EGTA ratio of 1.0 is attained within 1 h. Another point alluded to previously which implies Ca2* uptake rather than binding is also illustrated in Figure 22, and concerns the fact that when the vesicles were separated on the mini-columns from the incubation mixture, 94 Figure 21. Ca2*-uptake in 20 mol% DOPA LUVETs with trapped oxalate. The LUVETs (DOPC/DOPA/Cholesterol;80:20:10) were made in 100 mM potassium oxalate/50 mM KC1/20 mM HEPES, pH 7.4 and the external medium was exchanged with 175 mM K.C1/20 mM HEPES, pH 7.4. Ca i+-uptake was monitored in the absence ( • ) and in the presence ( A ) of valinomycin. Valinomycin was added at (1 ^mol phospholipid. Ca2+-uptake was also monitored in the control 20 mol% DOPA LUVETs made in 175 mM KC1/20 mM HEPES, pH 7.4 ( • ). UPTAKE (nmol Ca/umol PL) K ) ^ f > O O O 96 F igure 22. C a 2 + - uptake i n 20 mo\% D O P A L U V E T s with trapped E G T A . The L U V E T s (DOPC/DOPA/Cho l e s t e r o l ; 80 :20 :10 ) were made up i n 100 m M E G T A / 5 0 m M K.C1/20 m M H E P E S , p H 7.4 and the external med ium was exchanged for 275 m M KC1 /20 m M H E P E S , p H 7.4. Du r i ng the Ca 2 *-up take experiment, the L U V E T s were separated on columns equi l ibrated with 2 m M calc ium chlor ide ( • ) and on columns without any ca lc ium chlor ide( • ). The control 20 mo l% D O P A L U V E T s had 175 m M KC1/20 m M H E P E S , p H 7.4 as the internal and external mediums( 0 ). UPTAKE (nmol Ca/umol PL) 98 approximately 25% more Ca2* was found to be associated with the vesicles when they were separated on columns equilibrated with 2 or 10 mM (cold) calcium chloride in the buffer rather than on columns containing no calcium. This argues against Ca2* binding to the outside of the vesicles influencing the results because one would expect the excess of cold calcium on the column to exchange with any "Ca bound to the exterior of the vesicles while they were progressing down the column. This would consequently result in lower amounts of " C a associate with the vesicles. The results obtained were just the opposite. Although EGTA is a more effective Ca2*-sink, oxalate was utilized for subsequent experiments investigating the effect of pH due to the pH dependence of the EGTA-metal ion complex. These complexes are stable at high pH, however, at lower pH values (below pH 7) EGTA is not as effective a Ca2* chelator due to the competition of protons for the carboxyl and amine groups (Thach and Newburger, 1972). In summary, the results from varying the amounts of phosphate trap and traps employing other calcium chelators such as oxalate and EGTA in 20% PA LUVETs indicate that Ca2*-uptake into the vesicles is observed which cannot be simply attributed to Ca2*-binding to PA. Also, the increased levels of uptake observed when the columns were pre-equilibrated with (cold) Ca2* support Ca2*-uptake as opposed to binding to PA. 3.3.4 Effect of pH on Ca2*-uptake PA has two pK 's, the First one at 3.5 and the second one at 8.0 (Papahadjopoulos, 1968). Most of the uptake experiments were done at pH 7.4 where PA has an average charge slightly less than -1. Therefore, it was of interest to investigate the pH dependence on Ca2*-uptake. LUVETs (20% PA) containing 100 mM K*-oxalate/50 mM KC1 were prepared with either 99 20 mM MES buffer (pH 6.0), 20 mM HEPES buffer (pH 7.4) or 20 mM CHES buffer (pH 9.0). The results are illustrated jn Figure 23. No calcium was found associated with the vesicles at pH 6.0 while at pH 9.0, significantly greater amounts of calcium were found associated widi the vesicles than those observed at pH 7.4. This is consistent with the observation that PA binds more strongly with bivalent metals at pH 7.4 than at pH 6.0 (Papahadjopoulos, 1968). At pH 9.0, PA has two negative charges and its affinity for calcium is presumably enhanced. It did not appear that the Ca2*-sink employed is an effective chelator of Ca3* at this pH. This is because in contrast to similar Ca2*-uptake experiments done at pH 7.4, the presence of the Ca2*-sink did not change the rate or amount of Ca2*-uptake (results not shown). This suggests that at pH 9.0, the observed trapping is primarily due to binding of Ca2* to PA. 3.3.5 Ca2*-uptake in DOPC and 20% PS LUVETs + A23187 In order to ascertain whether Ca2*-uptake was specific to PA-containing vesicles, similar uptake experiments were done in 20% PS and pure DOPC LUVETs. As shown in Figure 24, pure DOPC LUVETs failed to take up any calcium. The same was observed for 20% PS LUVETs (see Figure 18). This implies that Ca2*-uptake is specific for PA containing systems. The ionophoretic properties of PA were also compared with those observed for the fungal calcium ionophore, A23187, which was incorporated into DOPC LUVETs at a molar ratio of 1:500 phospholipid. Ca2*-uptake by these DOPC+A23187 LUVETs is illustrated in Figure 24. The internal medium was 100 mM oxalate/50 mM KC1/20 mM HEPES, pH 7.4 and the external medium was 175 mM KC1/20 mM HEPES, pH 7.4. LUVETs with no oxalate were also employed. As illustrated in Figure 24, the amount of Ca2*-uptake 100 F igure 23. Effect o f p H on Ca 2 *-uptake . 20 mol% D O P A L U V E T s (DOPC/DOPA/Cho l e s t e r o l ; 80 :20 :10 ) having 100 m M potassium oxalate/50 m M KC1/20 m M buffer as the internal med ium and 175 m M KC1/20 m M buffer as the external med ium were employed. ( • ) L U V E T s had 20 m M M E S p H 6.0 buffer; ( # ) L U V E T s had 20 m M H E P E S , p H 7.4 buffer; ( • ) L U V E T s had 20 m M C H E S , p H 9.0 buffer. 102 F igure 24. Ca 2 * -up t a ke i n D O P C + A23187 L U V E T s . The D O P C L U V E T s were made up in 100 m M potass ium oxalate/50 m M KC1 /20 m M H E P E S , p H 7.4 and the external med ium was exchanged for 175 m M KC1 /20 m M H E P E S , p H 7.4 on a Sephadex G - 5 0 . co l umn as descr ibed in the Mater ia l s and Methods section. (A ). D O P C + A23187 (1:500 mo l /mo l phospho l ip id ) L U V E T s without any oxalate i n the internal med i um (i.e. 175 K.C1/20 m M H E P E S , p H 7.4 as the internal and external med iums); ' ( • ), D O P C + A23187 (1:500 mo l /mo l phospho l ip id) L U V E T s wi th oxalate i n the internal med ium; ( • ), D O P C L U V E T s wi th oxalate i n the internal med ium but wi thout A23187. 1 2 3 4 T I M E (h) 104 is s ignif icantly greater when oxalate is present as a Ca 2 * - s i n k inside the vesicles. These results are very s imi lar to those observed i n PA - con t a i n i ng systems (see F igure 21). S imi lar results were also observed i n 20% PS + A23187 L U V E T s in wh ich 40 m M K*-phosphate was employed as the Ca 2 * sink. Th i s is i l lustrated i n F igure 25. 3.3.6 Influence o f a membrane potent ia l The next quest ion investigated was to determine why a s low rate o f Ca 2 * -up take is observed and also why the amount o f C a 2 + taken up d id not saturate the calculated amount o f the C a 2 + - s i n k s employed. In part icular, it was o f interest to know whether a K* d i f fus ion potential (K* inside) cou ld enhance Ca 2 * uptake by a K .* -Ca 2 + exchange process faci l i tated by the presence o f a K* ionophore. The reasoning beh ind these experiments fo l lows f rom the results o f Ba l l y et a l . (1983a) who show that l ipoph i l i c cations such as safranine can be accumulated to h igh levels inside (egg P C ) L U V E T s i n the presence o f a va l inomyc in induced K* d i f fus ion potential. Th is uptake proceeds as a sa f ran ine-K* transbi layer exchange process. In the case o f the Ca 2 * -up take investigated here, the i onopho re t i c -Ca 2 * transport complex may have characteristics o f a l ipoph i l i c cation, a l lowing net Ca 2 * -up take correlated with K* eff lux. Therefore, the inf luence o f va l inomyc in , a K* ionophore, was investigated i n 20% PA , 20% PS + A23187 and D O P C + A 2 3 1 8 7 L U V E T systems exhib i t ing transbi layer N a V K * gradients (K* inside). Concentrat ions o f va l inomyc in employed were 1 ^ g / yumol phospho l ip id . The presence o f va l inomyc in in 20% P A and 20% PS + A23187 signif icantiy increases both the rate and amount o f Ca 2 * -up t ake as indicated i n F igure 25. The vesicles had 125 m M KC1/40 m M K*-phosphate, p H 7.4 as the internal med ium and 135 m M N a C l / 2 0 m M H E P E S , p H 7.4 as the external med ium. The trap vo lume of the vesicles was 105 F igure 25. Effect o f Va l i nomyc in and C C C P on Ca 2 *-up take in 20 mol% PS + A23187 and i n 20 mo l% D O P A L U V E T s . 20 mo l% PS L U V E T s ( D O P C / P S / Cholesterol;80:20:10) and 20 mo l% D O P A L U V E T s (DOPC/DOPA/Cho l e s t e r o l ; 80 :20 :10 ) were prepared i n 125 m M KC1/40 m M potassium phosphate, p H 7.4 and the external med ium was exchanged for 135 m M N a C l / 4 0 m M H E P E S , p H 7.4 on a Sephadex G - 5 0 co lumn as described in the Mater ia ls and Methods section. Va l i nomyc in was added at a concentration o f l/ug/>imol phospho l ip id and C C C P was added to a final concentrat ion o f 20 juM f r om concentrated stock solutions, (a) 20 mo l% PS L U V E T s were employed: ( • ) 20 mo l% PS L U V E T s in the presence and absence o f both va l inomyc in and/o r C C C P ; ( • ) 20 mol% PS + A23187 (1:200 mo l /mo l phospho l ip id) L U V E T s ; ( A ), 20 mo l% P S + A 2 3 1 8 7 L U V E T s with va l inomyc in; ( O ). 20 mo l% PS + A23187 L U V E T s with va l inomyc in and C C C P . (b) 20 mo l% D O P A L U V E T s were employed: ( # ), 20 mo l% D O P A L U V E T s ; ( A ), 20 mo l% D O P A L U V E T s with va l inomyc in; ( O )• 2 0 mo\% D O P A L U V E T s with va l inomyc in and C C C P . Results with C C C P alone were s imi lar to the ones wi th va l inomyc in and C C C P . UPTAKE (nmol Ca/umol PL) 107 approximately 1-1.2 ^ 1 / ^ m o l phospho l ip id . Va l i nomyc in enhanced the rate o f C a 2 * -uptake and the amount o f Ca 2 * taken up approached the calculated levels expected to saturate the C a 2 + - s i n k (40-45 nmols P i /^umol phosphol ip id) . S imi lar results were observed in the case o f D O P C + A 2 3 1 8 7 L U V E T s as i l lustrated in F igure 26. In this case, 100 m M K*-oxa la te/50 m M KC1/20 m M H E P E S , p H 7.4 was the internal med ium and 175 m M N a C l / 2 0 m M H E P E S , p H 7.4 was the external med ium. W h e n 20% P A L U V E T s were employed with the above conditions, s imi lar effects were observed. The control 20% PS and D O P C L U V E T s showed no signif icant uptake in the presence or absence o f the ionophores. Th i s impl ies that va l inomyc in itself does not affect the transport o f Ca 2 * in the L U V E T s . The va l inomyc in response was abol ished when KC1 was present inside and outside the vesicles suggesting that Ca 2 * -up take was coupled to K* eff lux down its concentrat ion grad ient 3.3.7 Inf luence o f proton ionophore ( C C C P ) on Ca 2 *-up take The inf luence o f the proton ionophore, carbonyl c yan ide -m-ch lo rohyd razone ( C C C P ) , on ca lc ium uptake was also investigated. There were two reasons for this. First, i f P A is able to transport Ca 2 * into the L U V E T s in a cycl ic fashion, then it must " f l i p " back to the vesicle exterior. It may be suggested that the protonated fo rm wou ld move across the bi layer more easily, result ing in a net ef f lux o f protons. Al ternat ively, i f the accumulated Ca 2 * is chelated by the phosphate sink, an increase i n inter ior proton concentration may be expected. A s a result, this might slow the rate o f Ca 2 *-up take . Therefore, C C C P at a final concentration o f 20 , u M was employed i n the presence or absence o f va l inomyc in. A s is i l lustrated in F igures 25 and 26, C C C P enhances both the rate and amount o f Ca 2 * taken up to the calculated levels o f Ca 2 * expected to saturate the employed sinks. Th is is observed in al l three L U V E T systems; 20% PA , 20% PS + A23187, and 108 F igure 26. Effect o f Va l i nomyc in and C C C P on Ca 2 *-up take in D O P C ± A23187 L U V E T s . The D O P C L U V E T s were made up i n 100 m M potassium oxalate/50 m M KC1/20 m M H E P E S , p H 7.4 and the external med ium was exchanged for 175 m M N a C l / 2 0 m M H E P E S , p H 7.4 on a Sephadex G - 5 0 co lumn as described in the Mater ia ls and Methods section. ( • ), D O P C + A23187 (1:500 mo l /mo l phosphol ip id); ( O ), D O P C + A 2 3 1 8 7 L U V E T s + va l inomyc in; ( A ), D O P C + A 2 3 1 8 7 L U V E T s + C C C P ; ( • ), D O P C + A 2 3 1 8 7 L U V E T s + va l inomyc in + C C C P ; ( A ) D O P C L U V E T s without A23187 in the presence o f va l inomyc in and C C C P . 110 D O P C + A 2 3 1 8 7 . In the controls, C C C P i tse l f d id not a l low any Ca 2 * to be taken up. In summary, both C C C P and va l inomyc in have s imi lar effects on Ca 2 * -up take . W h e n used separately or in conjunct ion in 20% P A , 20% PS + A23187, or D O P C + A23187 L U V E T systems, both ionophores increase the rate and amount o f Ca 2 * -up t ake . These effects suggest that i on f luxes o f both ions might p lay important roles in the Ca 2 * -up t a ke process. The increase in the rate o f Ca 2 * -up t a ke dur ing the presence o f these ion f luxes imp l ies that they cou ld be acting to dr ive Ca 2 * -up t a ke in membranes contain ing a ca l c ium ionophore. 3.4 D I S C U S S I O N The results o f this chapter prov ide strong indicat ions that D O P A can act as a Ca 2 * ionophore in L U V systems. First, C a 2 * can be accumulated in to L U V systems (wi th a phosphate "s ink" inside) when D O P A is present, whereas no such uptake is observed when no D O P A is present, o r when (egg) PS is substituted for D O P A . Second, the amount o f C a 2 * accumulated is dependent on the amount and type o f trap employed. C lear l y , in order for C a 2 * to be sensitive to the trap on the ins ide o f the vesicle, it must be transported to the vesicle inter ior. Th i r d , the vesicle associated Ca 2 * does not appear to be mere ly bound to D O P A on the vesicle exterior, as passage o f the vesicles through a gel filtration co l umn p re -equ i l i b r a t ed wi th " co l d " C a 2 * does not result in exchange wi th 4 5 C a associated w i th the vesicle system. Rather , an increase i n the vesicle associated 4 5 C a is observed, suggesting that some o f the entrapped Ca 2 * leaks out o f the vesicles when passed over co lumns wh ich do not contain Ca 2 * . F ina l l y , the t ime course o f Ca 2 * uptake and amount o f accumulated C a 2 * obta ined i n vesicles conta in ing 20% D O P A is rough ly comparab le to that observed for pure D O P C vesicles conta in ing the same trap, in the presence o f the Ca 2 * ionophore, A23187 I l l (compare F igu re 21 and F igu re 24). There are, however, st i l l some prob lems. F i rst , wh i le the amount o f C a 2 ' taken up is sensitive to the amount o f P i trap employed (F igure 19) it is also sensitive to the amount o f D O P A contained in the vesicle membrane (F igure 20), wi th increased uptake at h igher P A concentrations. It may be expected that i f P A was on ly act ing as an ionophore, then faster uptake wou ld be observed at h igher P A contents but that the net amount o f C a 2 * accumulated at extended t imes wou ld be the same. Th i s does not appear to be the case, suggesting that wh i le the accumulated Ca 2 * is trap dependent, some b ind ing also occurs wi th the D O P A . Th i s effect c lear ly warrants further investigation, poss ib ly emp loy ing a more eff ic ient C a 2 4 trap such as E G T A or enhancing the rate o f uptake (by incubat ing at h igher temperature, or faci l i tat ing H* and K* f luxes - see below) to ensure that equ i l i b r i um has been reached. The second area that remains to be we l l character ized concerns the t ransmembrane fluxes o f monova lent cations (K* and H*) accompany ing C a 2 * uptake. It is clear f r om the results presented i n F igure 26 that the rates o f C a 2 * uptake into D O P C vesicles in the presence o f A23187 are marked ly increased by the presence o f the K 4 ionophore , va l inomyc in , and/o r the proton ionophore, C C C P . S im i l a r effects are observed for D O P A conta in ing systems (F igure 25 (b)). The effects o f C C C P in the A23187 dependent uptake are consistent wi th the electroneutral nature o f this carr ier, wh i ch exchanges Ca 2 * for 2 H* (N icho l l s , 1982) and it may be that uptake med ia ted by D O P A proceeds by a s imi lar mechan ism. It is certainly un l i ke ly that the P A is able to move f r om one side to the other o f the membrane i n a charged fo rm. It is c lear however that that these f luxes must be separated f r om the possib le inf luences o f a membrane potent ia l set up by a K* d i f fus ion gradient i n the systems invest igated i n F igures 25 and 26 due to the N a * / K 4 ion gradients imposed. D u e to t ime l imi tat ions on the preparat ion o f this thesis these exper iments cou ld not be per fo rmed. 112 A final area that remains to be resolved concerns the strong inf luence o f p H on D O P A mediated uptake o f C a 2 * (F igure 23). In part icular, it is not presentiy understood why an abrupt c u t - o f f in C a 2 " transport shou ld occur at p H 6.0 and lower as this p H valve does not correspond to the lower pK. o f the P A headgroup ( p K a = 3 . 5 ) . These results are in disagreement w i th a recent report that egg P A does not a l low transbi layer transport i n mu l t i l ame l la r vesicles (Ho lmes and Yoss, 1983). Th i s observat ion has been used to quest ion the ionophoret ic abi l i ty o f P A . However , under the condi t ions employed in their exper iments, the results are not total ly unambiguous. F i rst , mu l t i l ame l la r vesicles were used wh ich usual ly have < 1 0 % o f the total internal vo lume accessible to the external bi layer. Therefore, one wou ld expect to observe reduced levels o f Ca 2 * -up t ake . Second, the h igh ion ic strength condi t ions used in the exper iments (145 m M K C 1 , 100 m M T r i s - H C l , p H 7.45) cou ld affect the aff in i ty o f P A for Ca 2 * . Th i r d , a re lat ively saturated species o f P A (egg P A ) was used, and it may be that the acyl cha in saturat ion o f P A inf luences it ionophoret ic capabi l i t ies. In part icular, the P A generated f r om PI i n the b io log ica l s ituat ion is h igh ly unsaturated and cou ld be a more effective Ca 2 * ionophore . Fou r t h , the M L V s contained 10 -20 mo l% dicety lphosphate wh i ch w i l l compete for Ca 2 * . F ina l l y , the exper iments mon i to red Ca 2 * - up t a k e by emp loy ing the Ca 2 *-sens i t i ve dye, arsenazo III. Th i s dye can interact w i th membranes in an as yet i l l - d e f i n ed manner (C.P.S. T i l cock , personal commun ica t ion) wh i ch cou ld also affect the ionophoret ic abi l i ty o f P A . In summary, the results presented in this chapter are consistent w i th an abi l i ty o f P A to act as a Ca 2 * ionophore, but the detai ls invo lved are c lear ly compl icated. Howeve r the results presented do prov ide a sound basis and mot iva t ion for further invest igations o f the ionophoret ic capacit ies o f P A and the mechanisms invo lved, but are i n no way the last word on the subject 113 C H A P T E R I V Phospho l ip ids as Adjuncts to Ca 2 * -S t imu l a t ed Release O f Ch r oma f f i n G r anu l e Contents 4.1 I N T R O D U C T I O N In the prev ious two chapters, the propert ies o f two l ip id species invo lved i n the PI - response have been examined. Th i s response is invo lved in the process whereby Ca 2 * is "mob i l i z ed " inside the target cel l . It has been shown that PI has propert ies consistent w i th mainta in ing a stable b i layer structure even in the presence o f d ivalent cations, whereas P A exhib i ts propert ies wh i ch appear to be consistent w i th an abi l i ty to translocate Ca 2 * across the bi layer. In this chapter the consequences o f increased cystosol levels o f C a 2 * i n secretory cel ls is examined, namely, the fus ion between the secretory granule and the sur round ing p lasma membrane wh ich results in extracel lular release o f granule contents. The approach that is taken concerns the roles o f l ip ids i n the Ca 2 * - s t imu l a t ed fus ion event o f exocytosis. In part icular, it is probab le that the inner mono layer o f the secretory cel l p lasma membrane has a l i p i d compos i t ion s imi la r to that observed for the inner monolayer o f the erythrocyte membrane wh ich is composed p r imar i l y o f P E and PS (Zwaa l et al., 1977). B i layers composed o f P E and PS i n the appropr iate ratios are sensitive to the presence o f Ca 2 * , wh i ch can tr igger b i layer to hexagonal ( H n ) transit ions in mu l t i l ame l la r systems (T i l cock and Cu l l i s , 1981) and can st imulate fus ion between un i lamel lar vesicles wi th this l i p i d compos i t ion (Hope et al., 1983). It may therefore be suggested that when intracel lu lar levels o f C a 2 * are raised the inner mono layer o f the p lasma membrane may be susceptible to fus ion wi th the secretory granules. In this chapter the abi l i ty o f P E - P S vesicles to undergo C a 2 * -114 st imulated fus ion wi th ch romaf f i n granules is examined as a first approx imat ion to the b io log ica l s i tuat ion. 4.2 M A T E R I A L S A N D M E T H O D S 4.2.1 Isolat ion o f Ch r oma f f i n G ranu l e s A mod i f i ca t ion o f the methods o f Sm i th and W i n k l e r (1967) and o f He l l e et a l . (1971) was used to isolate the ch romaf f i n granules f rom bovine adrenal medu l l a . Between twenty and thirty fresh bov ine adrenal glands were obta ined f r om a local abattoir, where they were p laced on ice unt i l ready for iso lat ion. A l l subsequent steps were carr ied out at 4° C . Preparat ion o f the Large G r anu l e F rac t ion The adrenal glands were defatted and dissected free o f cort ical tissue. The adrenal medu l lae were p laced in i c e - c o l d 0.3 M sucrose solut ion conta in ing 10 m M H E P E S and 1 m M E D T A at p H 7.0 (this so lut ion is referred to as "buf fered sucrose"). Homogen i za t i on o f the adrenal medu l lae was more convenient ly carr ied out emp loy ing a B r i n kman Po ly t ron (5 s) and three subsequent passes in a Potter homogen izer w i th a mo t o r - d r i v e n tef lon pestie. The homogenate was centr i fuged at 800 g for 10 m i n and the pe l let (cel l debris) was discarded. The l ow speed supernatant was then centr i fuged at 17,000 g for 10 m i n to pe l le t the denser ch romaf f i n granules. T h e upper f luf fy b rown layer on the pe l let (p r imar i l y mi tochondr ia and lysosomes) was carefu l ly decanted and the p ink pel let was washed wi th buf fered sucrose. The h igh speed centr i fuge step was repeated twice after resuspending the p ink pe l let conta in ing the ch romaf f i n granules i n 40 m l o f buf fered sucrose. The f ina l pel lets 115 obta ined after the h igh speed centr i fugat ions were poo led and resuspended in approx imate ly 10 m l o f buf fered sucrose to give a prote in concentrat ion o f around 30 . m g / m l . The fract ion contained part ia l ly pur i f i ed chromaf f in granules and was referred to as the "La rge G ranu l e F rac t i on " ( L G F ) . The large granule fract ion was stable in an i so -osmot i c med i um and had a smal l amount o f m i tochondr ia l contaminat ion (see pur i ty sect ion 4.2.3). In the major i ty o f the release experiments, the large granule fract ion rather than the h igh ly pur i f i ed granules was emp loyed because o f the marked osmot ic fragi l i ty o f the purer preparat ions. Preparat ion o f the h igh ly pur i f i ed ch romaf f i n granules D u e to the h igh density o f the ch romaf f i n granules relative to that o f the mi tochondr ia and lysosomes, the chromaf f in granules can be h igh ly pu r i f i ed by centr i fugat ion through a hyperosmot ic sucrose med i um . Abou t 2 m l o f the large granule fract ion was layered onto 30 m l o f 1.6 M sucrose and centr i fuged at 80,000 g for 1 h. Several d i f ferent layers resulted after the s tep-grad ient centr i fugat ion. These are i l lustrated and numbered in F igu re 27. Ana lys i s o f the fractions by Smi th and W i n k l e r (1967) have shown that the interface layers 2 and 2' conta in most o f the mitochondr ia] and lysosomal activity. The p ink sediment, 5, corresponded to the h igh ly pu r i f i ed chromaf f in granules and were used for l i p id extract ion and determinat ion o f the phospho l i p i d compos i t ion o f the ch romaf f i n granule membrane. The scheme o f iso lat ion is i l lustrated i n F i gu re 27. 116 Figure 27. Summary of the isolation procedure for chromaffin granules of bovine adrenal medulla. Step 4 was repeated twice. Bovine Adrenal Medullae (i) Homogenized in buffered sucrose Hooogenate (2) Pellet e l l debris, nuclei) 750 g, 10 min. Low Speed Supernatant (3) 17,000 g 10 min. High Speed Pellet resuspended in buffered sucrose Supernatant (microsomes, c e l l sap) (4) Fluffy layer 4 -17,000 g, 10 min. Large Granule Fraction (resuspended in buffered sucrose) (5) Supernatant (microsomes, c e l l sap) 2 ml. layered on 30 ml. 1.6 M sucrose r 2 ml. of large granule fraction 1.6 H sucrose (6) 80,790 g, 60 min. Purified chromaffin granules 0.3 M Sucrose MM 1.6 M Sucrose —|— 1 (clear solution) —2 (opaque, dark brown) 2' (dark brown) 3 (opaque, light brown) A (light brown) 5 (pink purified chromaffin granules) 118 4.2.2 Isolation of the Chromaffin Granule Membranes The chromaffin granule membranes were prepared by lysis of the chromaffin granules in a hypo-osmotic medium (10 mM HEPES, pH 7.0), followed by a freeze-thaw cycle, and subsequent centrifugation at 27,000 g for 30 min. This was repeated four or five times until the absorbance of the supernatant at 265 nm. after centrifugation was less than 5% of the initial value. The brown pellet obtained after the centrifugation step corresponded to the chromaffin granule membrane fraction. 4.2.3 Determination of Mitochondrial contamination in the Large Granule Fraction Since the major contaminant of the chromaffin granule preparation is mitochondria, it was of interest to determine the extent of impurity in the large granule fraction. The enzyme, malate dehydrogenase (MDH), was used as a mitochondrial marker enzyme. MDH catalyzes the following reaction in the mitochondrial matrix: Oxaloacetate + NADH + H' > L-Malate + NADH* NADH absorbs at 340 nm, therefore the decrease in absorbance over time was taken as corresponding to MDH activity. The A(340nm)/min/mg protein was taken as the specific activity of MDH, which was proportional to the amount of mitochondria in the sample. Fraction 2 in the ultra-centrifugation step of the preparation was assumed to correspond to the pure mitochondrial fraction. The stock solutions needed for the assay were 0.1 M phosphate buffer, pH 7.5, 15 mM oxaloacetate solution (2mg/ml in phosphate buffer), and 12 mM NADH (10 mg/ml). The reaction was performed in, a 5 ml cuvette by adding the ingredients in the following order: 2.83 ml phosphate buffer, 0.10 ml oxaloacetate solution, 0.05 ml NADH 119 solut ion, g iv ing a final concentrat ion o f 95 m M , 0.5 m M , and 0.2 m M respectively. The react ion was started by add i t ion o f 0.02 m l o f the sample. The decrease in absorbance was recorded for about a minute at 340 nm . The reference cel l had al l the "cockta i l " except N A D H . The specif ic activity o f M D H in the large granule fract ion and the h igh ly pur i f i ed chromaf f in granules was approx imate ly 15% and 5% o f the pure mi tochondr ia l fract ion respectively. 4.2.4 L ow r y Pro te in Assay Prote in was measured accord ing to the procedure o f Lowry et a l . (1951), using crystal l ine bov ine serum a lbum in as a standard. Stock solut ions requ i red for the assay were 2% potassium tartate, 1% copper sulfate, and 2% sod ium carbonate dissolved in 0.1 N sod ium hydrox ide. So lut ion I was freshly prepared by m i x i ng 98 m l o f sod ium carbonate wi th 1.0 m l o f copper sulfate. So lut ion II was prepared by d i lu t ing 2 N Fo l i n ' s Reagent to 1 N . The assay was carr ied out by add ing 5.0 m l o f so lut ion I to 1.0 m l o f sample and water. A f te r 10 m in , 0.5 m l o f so lut ion II was added fo l lowed by immed ia te vortex ing. Absorbance at 550 n m was recorded after 30 m in . L I P I D I S O L A T I O N A N D P U R I F I C A T I O N 4.2.5 L i p i d Ext ract ion f rom Ch r oma f f i n G ranu l e Membranes L i p i d s were extracted by the procedure o f B l i gh and Dye r (1959). Br ie f ly , a sample o f ch romaf f i n granule membranes was d i lu ted to 5 m l w i th water. Ano the r 2.1 vo lumes o f methano l and 1.0 vo lume o f ch l o ro fo rm was added and the one -phase solut ion 120 was st irred for 15 m in . T o extract the l ip ids, a two phase system was produced by the addi t ion o f 1.0 vo lume o f ch lo ro fo rm and 1.0 vo lume o f water. Th i s c loudy solut ion was centr i fuged on a bench - top centr i fuge at approx imate ly 3,000 r pm for 5 m in . The top wate r - so lub le fract ion (conta in ing proteins, ions, etc.) was aspirated o f f and the rema in ing ch lo ro fo rm phase conta in ing the l ip ids was ro ta ry -evapora ted in a round bottom flask under vacuum. The dry l i p i d was redissolved in some ch lo ro fo rm and stored at - 2 0 ° C under nitrogen. 4.2.6 Isolat ion o f Erythrocyte Memb rane Phospho l ip ids The erythrocyte membrane phospho l ip ids were a gift f r om D r . M . J . Hope . They had been isolated f r om human erythrocytes and pur i f i ed us ing low pressure l i qu i d chromatography on si l ic ic ac id and carboxymethy l cel lulose co lumns. The l ip ids were eluted by mixtures o f ch lo ro fo rm and methano l , and were > 9 9 % pure wi th respect to phosphorus (Hope and Cu l l i s , 1979). C h l o r o f o r m mixtures o f the outer erythrocyte monolayer consisted o f 44 mo l% P C , 44 mo l% Sph., and 12 mo l% PE , wi th an equ imo lar amount o f cholesterol . T he inner mono layer mix ture consisted o f 47 mo l% P E , 28 mo l% PS, 15 mo l% P C , and 10 mo l% Sph., with an equ imo la r amount o f cholesterol (Zwaa l et al., 1978). 4.2.7 Isolat ion and Pur i f i ca t ion o f L i p i d s The procedures for the iso lat ion and pur i f i ca t ion o f phosphat idy lcho l ine , phosphat idy le thano lamine, and phosphat idy lser ine are out l ined in sections 2.2.3 to 2.2.7. Cho les te ro l and bovine heart ca rd io l i p in (sod ium salt) were purchased f r om S igma ( S i Lou i s , M O ) . A l l the l ip ids were checked for pur i ty by 2 D - T L C and their po l ymorph i c phase preferences were character ized by 3 1 P - N M R techniques. Br ie f l y , the 121 soya P E and egg P E had a b i layer to hexagonal transit ion temperature o f 15°C and 30°C respectively. Ca rd i o l i p i n in the presence o f ca lc ium adopted a hexagonal H „ conf igurat ion. 4.2.8 Preparat ion o f M o d e l L i p i d Systems The phospho l ip id mode l systems used were sonicated vesicles. They were obta ined by m ix ing appropr iate quantit ies o f l i p i d i n ch lo ro fo rm wh ich was then evaporated under a stream o f nitrogen and subsequent ly stored under vacuum for at least 2 h. The l i p i d was hydrated in the buf fered sucrose solut ion and sonicated intermittent ly (30 s sonicat ion fo l lowed by 30 s intervals i n an i ce -wate r bath) by emp loy ing a t ip sonicator. Sonicat ion was cont inued unt i l the dispers ion became opt ica l ly clear (approx. 5 min) . 4.2.9 Spectrophotometr ic Release Assay Typ i ca l l y , chromaf f in granules (0.8-1.2 mg prote in , 4 0 - 5 0 uJ o f the large granule fract ion) in buf fered sucrose were incubated i n the presence o f sonicated vesicles (F igure 32), C a C l 2 , o r N a C l at 25°C for 15 m i n in a total vo lume o f 1.0 m l Whe re C a C l 2 was not added, a corresponding mi l l i o smo ls o f N a C l was added in order to m in im i ze release due to osmot ic differences. The incubat ion was stopped by add i t ion o f 3.0 m l i c e - c o l d buf fered sucrose and the ch romaf f i n granules were pel leted at 17,000 g for 10 m in . The supernatant was assayed for the release o f the ch romaf f i n granule contents, namely prote in, A T P , and catecholamines. A reproduc ib le level o f about 25% background release was observed i n a l l controls i n agreement w i th pub l i shed observat ions (H i l l a rp , 1959; M o r r i s et al., 1977). Release o f contents was hence expressed as a percentage after subtract ing the background. A measure o f the total 122 release o f the ch romaf f i n granule contents was taken as t he ' amoun t release after lys ing the chromaf f in granules in 5 m l o f 10 m M H E P E S , p H 7.0, fo l lowed by a f reeze-thaw cycle. 4.2.10 Assay ing for the Release Products o f the Ch r oma f f i n G ranu les Pro te in The amount o f prote in released in the supernatant was measured by the procedure o f L ow r y et a l . (1951). The determinat ion was made in 1.0 m l o f the supernatant and the blank contained 1.0 m l o f buf fered sucrose. Catecho lamines The amount o f catecholamines was measured accord ing to the procedure o f von Eu l e r and Hambe rg (1949). B r ie f l y , to 1.0 m l o f the supernatant, 1.0 m l o f 1 M acetate buf fer ( p H 6.0), 50 u l o f 10% S D S , and 0.2 m l o f 1 N iod ine solut ion was added. A f t e r exactiy 10 m in , 0.2 m l o f 0.5 M sod ium thiosulf i te was added to b leach the co lor o f the ox id i zed catecholamines and absorbance at 530 n m was recorded immediate ly . Tota l Contents Since the contents o f the ch romaf f i n granules absorb quite strongly at 265 nm, a very convenient assay for mon i to r ing release was s imp ly to determine the absorbance o f the supernatant at 265 n m (Edwards et al. . 1974; M o r r i s et a l „ 1977). It was found that the results obta ined by this assay correlated very favorably to those obta ined by the prote in and catecholamine determinat ions (see F igu re 28). Therefore, i n the major i ty o f the release exper iments, the A (265nm) assay was emp loyed . 123 F igure 28. Cor re la t i on o f the amount o f release o f the chromaf f in granule contents in the presence o f exogenous P E - P S (3:1) vesicles w i th subsequent add i t ion o f ca lc ium ch lor ide as determined by mon i to r ing the absorbance at 265 n m ( A ), the amount o f prote in released ( O ). and the amount o f catecholamine released ( • ). F o r details, see Mate r ia l s and Methods . R is def ined as the mo la r rat io o f exogenous (vesicular) l i p i d to endogenous (chromaf f in granule) phospho l i p id . 7. Release XI to 125 4.2.11 F reeze -F rac tu re o f Ch r oma f f i n G ranu l e s and L i p i d Vesic les F reeze - f rac tu re studies were per fo rmed on the und i lu ted 1.0 m l samples descr ibed i n section 4.2.9. G l y ce ro l was added as a cyroprotectant to a final concentrat ion o f 25%, and the samples were immediate ly frozen in l i qu id slush o f F reon . The f rozen samples were stored in l i qu id nitrogen unt i l the fracture procedure. F reeze - f rac tu re was done accord ing to standard procedures (see section 1.9) by employ ing a Balzer 's B A F 400D apparatus, and repl icas were v iewed by employ ing a Ph i l i p s 400 electron microscope. 4.3 R E S U L T S 4.3.1 Inf luence o f Exogenous L i p i d on Ch roma f f i n G ranu les Release The abi l i ty o f exogenous l i p i d to act as adjuncts for Ca 2 * - s t imu la ted release o f granule contents was per fo rmed using spectrophotometr ic assays as indicated in the Mater ia l s and M e t h o d section. The results obta ined are summar i zed in F igure 29, wh i ch shows that the presence o f P E - P S (3:1) vesicles at a concentrat ion corresponding to a mode l membrane phospho l i p id to chromaf f in granule membrane phospho l i p id rat io (R ) o f 5 (mo l /mo l ) results in complete release i n the presence o f ca lc ium. Th i s contrasts strongly w i th the results obta ined when either P E - P S (3:1) vesicles or C a 2 * are emp loyed separately. N o release for C a 2 * concentrat ions as h igh as 10 m M was observed. Th i s c lear ly establishes a requ i rement for both P E - P S (3:1) vesicles and Ca 2 * for release o f ch romaf f in granule contents. Fur ther , the amount o f release is sensitive to the order in wh ich these agents are added to the ch romaf f i n granule preparat ions. Less release is observed i f the granules are incubated wi th C a 2 + p r io r to incubat ion wi th P E - P S (3:1) vesicles. Th i s behav iour is interpreted as ar is ing 126 F igure 29. Release o f ch romaf f in granule contents after incubat ion in the presence o f C a 2 " and increasing amount o f exogenous P E - P S phospho l i p id as assayed by spectrophotometr ic techniques (see Mater ia l s and Me thods ) : ( # ) incubat ion i n the presence o f P E - P S (3:1) vesicles (15 m in ) where 5 m M C a C l j was introduced after 10 m i n ; ( • ) incubat ion in the presence o f P E - P S (3:1) vesicles (15 min) where 5 m M C a C l 2 was present in the granule suspension pr io r to int roduct ion o f the vesicles. Bu f fe red sucrose was employed th roughout R is def ined as the mo la r rat io o f exogenous (vesicular) l i p i d to endogenous (chromaf f in granule) phospho l i p id . 100 80 CO S 60 a-CD rr 40 20 128 f rom Ca 2 * - i n du c ed prec ip i tat ion o f the sonicated vesicles to f o rm the hexagonal H u phase before the vesicles can interact w i th the granule membranes. The mechan ism whereby P E - P S (3:1) vesicles act as adjuncts for Ca 2 * - s t imu l a t ed release o f ch romaf f i n granule contents is o f part icu lar interest It may be suggested that the ch romaf f i n granule release results f r om Ca 2 * - i n du c ed fus ion o f the vesicles w i th the granule membrane. Lys i s may then occur as a result o f the fus ion event itself, or the presence o f " non -b i l a ye r " l i p id in the granule membrane wh ich can then no longer support b i layer structure. A s indicated elsewhere (T i l cock and Cu l l i s , 1981), the add i t ion o f C a 2 * to P E - P S systems results in a structural segregation o f the PS component in to crystal l ine (presumably cochleate) regions (Papahadjopou los et ah, 1975), a l low ing the P E to revert to the hexagonal H n phase it prefers i n isolat ion. Quest ions then arise as to whether it is the abi l i ty o f C a 2 * to induce crystal l ine cochleate structures, or hexagonal H n phase organizat ion (or both), wh i ch is re lated to the lyt ic e f fec t These questions were approached by testing the abi l i ty o f vesicles composed o f pure PS and pure card io l ip in ( C L ) to act as adjuncts for Ca 2 * - s t imu l a t ed release. Whereas C a 2 * induces format ion o f a crystal l ine, apparently anhydrous cochleate structure for PS dispersions (Papahadjopou los et al. . 1975; H o p e and Cu l l i s , 1979), the add i t ion o f C a 2 * to C L mode l systems triggers format ion o f the hexagonal H n phase (Rand and Sengupta, 1972; Cu l l i s et al . . 1978). A s shown in F igu re 30, both PS and C L vesicles are effective adjuncts for Ca 2 * - s t imu l a t ed release o f ch romaf f i n granule contents. A l so , incubat ion with egg P E - P S vesicles instead o f soya P E - P S vesicles was tested since the b i layer to hexagonal temperature o f egg P E was between 25°C and 30°C. Therefore, pure egg P E wou ld adopt a b i layer conformat ion at 20°C and a hexagonal H n conformat ion at 35°C. F r o m F igu re 31, it is ev ident that s igni f icant release o f ch romaf f i n granule contents on ly occurred when the egg P E - P S vesicles were 129 incubated at 35C. F o r compar ison, incubat ion wi th soya P E - P S (3:1) vesicles resulted i n s ignif icant release at both 20°C and 35°C. Th i s exper iment strongly suggests part ic ipat ion o f non -b i l a ye r hexagonal H n phases in Ca 2 * - s t imu la ted release o f the chromaf f in granules as one o f the parameters related to the lyt ic even t A s a contro l for the above experiments, incubat ion wi th P C - P S (3:1) and P C - P S (1:1) vesicles resulted i n no s ignif icant release (F igure 30). These vesicles do not undergo any structural transformat ions in the presence o f Ca 2 * . It was o f interest to extend these observat ions to mode l systems wh i ch may more closely approx imate the compos i t ion o f the inner monolayer o f the adrenal cel l p lasma membrane. O n the assumpt ion that the transbi layer d istr ibut ion o f phospho l ip ids i n this membrane is s imi lar to that observed for the erythrocyte (the on ly p lasma membrane wel l character ized in this regard), the abi l i ty o f vesicles composed o f inner mono layer erythrocyte phospho l ip ids to induce release was examined. Th i s protocol also assumes that the outer mono layer o f the vesicle systems composed o f inner mono layer phospho l ip ids reflects the or ig ina l l i p i d compos i t ion. Th i s may be s l ight ly d i f ferent due to asymmetr ic l i p id d is t r ibut ion in sonicated systems (Berden et al., 1975). It has been shown elsewhere that the inner mono layer mode l system (part ia l ly) adopts the hexagonal H n phase i n the presence o f Ca 2 * , whereas outer monolayer systems remain i n a b i layer organizat ion (Hope and Cu l l i s , 1979). A s indicated in F igu re 32, the inner mono layer system can act as an adjunct for C a 2 * -st imulated release o f granule contents, whereas the outer mono layer system does not. It may be noted that somewhat larger ratios o f exogenous (vesicular) inner mono layer phospho l i p id to endogenous (granule) phospho l i p i d are requ i red than for the other adjunct systems i l lustrated here, a s ituat ion wh i ch is tentatively attr ibuted to the instabi l i ty o f the sonicated inner mono layer vesicles, as ind icated by an increasingly c loudy dispers ion obta ined as a funct ion o f t ime after sonicat ion. 150 F igu re 30. Ca 2 *-s t imu la ted release o f ch romaf f i n granule contents after incubat ion wi th vesicles o f var ious composit ions: ( O ) incubat ion in the presence o f pure soya PS vesicles (15 min) where 5 m M C a C l 2 was int roduced after 10 m in ; ( • ) incubat ion in the presence o f pure ca rd io l i p in vesicles (15 min) where 5 m M C a C l 2 was introduced after 10 m i n ; ( • ) incubat ion in the presence o f P C - P S (3:1) vesicles (15 m in ) where 5 m M C a C l 2 was int roduced after 10 m in . R is def ined as the mo la r rat io o f exogenous (vesicular) l i p id to endogenous (chromaf f in granule) phospho l i p id . 100 8 0 cn 03 0) <D GC 6 0 4 0 2 0 132 F igure 31. Release o f ch romaf f in granule contents in the presence o f soya and egg P E - P S (3:1) vesicles (15 m in ) at 20°C and 35°C: ( • ) incubat ion i n the presence o f soya P E - P S vesicles at 20°C and 35°C w i th subsequent addi t ion o f 5 m M C a C l 2 ; incubat ion i n the presence o f egg P E - P S vesicles and 5 m M C a C l 2 at 20°C ( # ) and at 35°C ( A ). R is def ined as the molar rat io o f exogenous (vesicular) l i p i d to endogenous (chromaf f in granule) phospho l i p id . 133 134 F igure 32. Release o f ch romaf f i n granule contents after incubat ion wi th vesicles composed o f erythrocyte l ip ids: ( • ) incubat ion in the presence o f outer mono layer l ip ids (15 min) where 5 m M C a C l 2 was int roduced after 10 m in ; ( • ) incubat ion in the presence o f inner mono layer l ip ids (15 m in ) where 5 m M C a C l 2 was int roduced after 10 m in . R is def ined as the mo la r rat io o f exogenous (vesicular) l i p id to endogenous (chromaf f in granule) phospho l ip id . 135 136 4.3.2 Ef fect o f C a l c i um Concent ra t ion In al l the above experiments, excess C a 1 ' (5 m M ) was used so that the l im i t i ng factor was the amount o f phospho l i p i d vesicles. W h e n the amount o f Ca 2 * was made l im i t ing , it was found that approx imate ly 2 m M Ca 2 * was requ i red to induce effect ive release o f the chromaf f in granule contents (F igure 38) when P E - P S (3:1) vesicles were employed as adjuncts. Th i s value corresponds wi th the concentrat ion o f C a 2 * requ i red to induce format ion o f the hexagonal H n phase in analogous P S - P E l iposomal systems (T i l cock and Cu l l i s , 1981). 4.3.3 F r eeze -F ra c tu re Studies A s ind icated above, the abi l i ty o f P E - P S and other vesicle systems to act as adjuncts for Ca 2 * - s t imu l a t ed release o f granule contents was assumed to arise f r om fusion o f the mode l systems wi th the granules. In order to p lace this hypothesis on a f i rmer foundat ion, the P E - P S (3:1) ves i c l e - ch romaf f i n granule systems was studied by employ ing f reeze- f rac ture techniques. F igu re 34 shows the inf luence o f 2 and 5 m M Ca 2 * on the chromaf f in granule (F igure 34a - c ) and P E - P S (3:1) vesicle systems (F igure 34d- f ) - The presence o f 2 m M Ca 2 * (F igure 34b) does not result in s igni f icant changes in granule d is t r ibut ion or size whereas 5 m M C a 2 * produces aggregation but not fus ion (fus ion is def ined by the observat ion o f larger fu l ly round systems). Th i s is consistent w i th the results o f other workers (Edwards et al., 1974; D a h l et a l , 1979) who observe aggregation (wi thout m i x i ng o f internal compartments) at C a 2 * concentrat ions less than 10 m M and some fus ion (appearance o f larger rounded structures) at h igher Ca 2 * levels, a l though Eke rd t et al . (1981) do report fus ion at lower C a 2 * levels. In the P E - P S (3:1) vesicle systems, 2 m M levels o f C a 2 * result in larger systems conta in ing l i p id i c part ic les, some regions o f hexagonal H n structure 137 F igure 33. Effect o f ca lc ium concentrat ion on the release o f ch romaf f i n granule contents incubated in the presence o f P E - P S (3:1) vesicles to obtain a rat io R o f exogenous (vesicular) phospho l i p id to endogenous (chromaf f in granule) phospho l i p i d o f 4.0 for 15 m i n where the C a C l 2 was added after 10 m in . Calcium Cone. (mM) 139 (not shown), and also some spira l structures s imi lar to cochleate P S - C a 2 * domains (Papahadjopoulos et al., 1975). These may be attr ibuted to the abi l i ty o f Ca 2 * to segregate PS into crystal l ine domains in these m ixed systems (T i l cock and Cu l l i s , 1981). The presence o f 5 m M Ca 2 * (F igure 34f) results in large regions o f hexagonal H u phase, in agreement wi th 3 1 P - N M R results (not shown here), as we l l as regions characterist ic o f cochleate structure. F igure 35 i l lustrates the inf luence o f Ca 2 * on the chromaf f in granules incubated with P E - P S (3:1) sonicated vesicles. F igure 35b depicts the s i tuat ion after incubat ion wi th 2 m M Ca 2 * . The largeT structures observed ( 5 - 1 0 t imes larger than the isolated granules) exhib i t in t ramembrane particles on the concave ( P F ) face wh ich are o f s imi lar size as the P F part icles o f the intact granules, whereas on the convex ( E F ) face, the part icles are much less distinct, again corresponding to the situat ion for the intact granules (F igure 35a). These observations, together w i th the fact that fus ion o f the P E - P S (3:1) vesicle systems by 2 m M Ca 2 * produces much di f ferent structures (F igure 34e), establ ish that the large structures arise at least i n part, f rom the granules. The density o f in t ramembranous part icles on the P F face o f the large fused system o f F igure 35b is approx imate ly 360 /um 2 , or about 40% o f the P F part ic le density o f the parent granules. Th i s part ic le d i lu t ion may be attr ibuted to the presence o f exogenous l i p id der ived f rom the P E - P S (3:1) vesicles. In some cases (results not shown), the P F part ic le density o f the large fused granules was not s igni f icant iy d i f ferent f rom that o f no rma l granules, ind icat ing some var iab i l i ty i n the number o f P E - P S vesicles accompany ing g ranu le -g ranu le fus ion. H i ghe r levels o f Ca 2 * (5 m M ) led to appreciable perturbat ion o f the large fused systems as shown in F igure 35c. Patches o f apparent ly H n phase or cochleate l i p i d structure are observed to be int imate ly associated wi th the membrane o f the fused granule system. 140 F igure 34. F reeze- f rac tu re micrographs o f chromaf f in granules and P E - P S (3:1) vesicles in the presence and absence o f 2 m M and 5 m M Ca 2 * . F r e e z e -fracture micrographs o f the fo l l ow ing: (a) ch romaf f i n granules in the absence o f Ca 2 * , showing both convex ( E F ) and concave ( P F ) fracture faces; (b) ch romaf f in granules in the presence o f 2 m M Ca 2 * ; (c) ch romaf f i n granules in the presence o f 5 m M Ca 2 * ; (d) sonicated P E - P S (3:1) vesicles in the absence o f C a 2 * ; (e) P E - P S (3:1) vesicles in the presence o f 2 m M Ca 2 * (the arrow indicates rows o f l i p id i c part ic les); (f) P E - P S (3:1) vesicles i n the presence o f 5 m M Ca 2 * . The white bars represent 200 nm. and the direct ion o f shadowing is ind icated by the ar rowhead in each micrograph. 141 142 F igu re 35. F reeze- f rac tu re micrographs o f ch romaf f in granules incubated w i th P E - P S (3:1) vesicles in 2 m M and 5 m M Ca 2 * . F reeze- f rac tu re micrographs o f the fo l lowing: (a) ch romaf f i n granules in the presence o f sonicated P E - P S (3:1) vesicles where the rat io o f ch romaf f in granule l i p i d to exogenous phospho l ip id is 1 to 4 and where the sample was prepared i n the absence o f C a 2 * ; (b) the same preparat ion as (a) but incubated i n the presence o f 2 m M Ca 2 * ; (c) the same preparat ion as (a) but incubated i n the presence o f 5 m M Ca 2 * ; (d) a mic rograph at h igher magni f i cat ion depict ing the interact ion between ch romaf f i n granules and exogenous P E - P S (3:1) vesicular l i p id after incubat ion i n the presence o f 2 m M Ca 2 * (upper por t ion) ; (e) same as (d) but i n the presence o f 5 m M Ca 2 * . A r r ows indicate part icles wh ich exh ib i t characterist ics o f l i p id i c part ic les. The white bars represent 400 nm, and the direct ion o f shadow is indicated by the ar rowhead in each mic rograph. 143 144 The nature o f the part ic les and other features observed in the presence o f 2 and 5 m M C a 2 * are o f interest and are indicated at h igher magni f i cat ion in F i gu re 35d,e. In part icular, the 2 m M Ca 2 * micrograph o f F igu re 35d wou ld appear to correspond to a large P E - P S (3:1) system fus ing wi th a chromaf f in granule system. Th i s is suggested by the di f ferent nature o f the part ic les observed in the fracture face. In the l i p i d system, the " l i p id i c part ic les" tend to l ine up i n rows (see F i gu re 34e), whereas the in t ramembrane part ic les o f the granule do not. 4.4 D I S C U S S I O N The results o f this chapter concern the observat ion that l i p i d vesicles wh i ch undergo structural transformations i n the presence o f C a 2 * can act as adjuncts for C a 2 ' - s t i m u l a t e d release o f ch romaf f i n granule contents. Th i s abi l i ty is associated wi th fus ion o f the vesicles wi th the granule, and lysis may result due either to the fus ion event i tsel f or as a result o f the presence o f non -b i l a ye r l i p i d i n the granule membrane . The c ommon property o f the phospho l i p id vesicles exh ib i t ing this ab i l i ty is that Ca 2 * induces aggregation and format ion o f larger structures for the vesicle system in isolat ion. A s both card io l ip in and phosphat idy lser ine systems are able to act as C a 2 * adjuncts, it wou ld appear that the detai led nature o f this larger structure (i.e. hexagonal H n or cochleate) is not a determin ing factor. Th i s suggests that the instabi l i ty o f the adjunct vesicles i n the presence o f C a 2 * is re l ieved on fus ion either w i th each other or w i th the ch romaf f i n granule membranes accord ing to their p rox im i ty and that i n the process o f embedd ing themselves in the granule membrane the integrity o f that membrane is d isrupted. It may be noted that the abi l i ty o f the sonicated vesicles to induce fus ion between ch romaf f i n granules corresponds closely w i th the abi l i ty o f (negatively charged) phospha t i dy lg l y ce ro l -145 and phosphat idy l ser ine-conta in ing vesicles to induce fus ion between cul tured cells in the presence o f Ca 2 * (Papahadjopoulos et al., 1973). The detai led mechan ism invo lved in this process remain a matter for speculat ion and is discussed in the next chapter. A l t hough the results in this chapter i l lustrate that phospho l ip id membranes that approx imate the inner p lasma membrane in l i p id compos i t ion can act as adjuncts for for Ca 2 * - s t imu l a t ed release o f ch romaf f i n granules, the mode l systems employed do not total ly ref lect the phys io log ica l s i tuat ion. T w o ma in prob lems remain . F i rs t , the level o f ca lc ium emp loyed in the exper iments is h igher than that found inside a cel l . It wou ld be more appropr iate i f the release o f the chromaf f in granule contents or fusion between the chromaf f in granules and the l i p i d vesicles can be shown to occur at lower ca lc ium concentrations. Recent ly, H o n g et al . (1982a; 1982b) have fused l i p id vesicles at m ic romo la r ca lc ium concentrat ions i n the presence o f synex in. S im i l a r exper iments between the chromaf f in granules and the l i p i d vesicles need to be done to ascertain whether the ca lc ium threshold i n this system can also be lowered. However , it can also be argued that the local ca l c ium concentrat ion at the fus ion site on the membrane can be s ignif icant iy h igher than the cytosol ic concentrat ion. A l so , the ca l c ium concentrat ion at wh ich fus ion was observed corresponds to the concentrat ion requ i red to induce format ion o f the hexagonal H n phase i n analogous P E - P S l iposomal systems (T i l cock and Cu l l i s , 1981). Second, the exper iments ut i l i zed smal l sonicated un i lamel lar vesicles (approx. 25 n m in diameter). These vesicles have a h igh degree o f curvature and do not accurately reflect the phys io log ica l s i tuat ion where the chromaf f in granules fuse wi th a p lanar p lasma membrane. Therefore, it wou ld be o f interest to study the inf luence o f large un i lamel lar vesicles o f s imi la r l i p i d compos i t ion ( P E - P S ) o n C a 2 * -st imulated release o f ch romaf f i n granule contents. P re l im inary exper iments us ing homogenous P E - P S L U V preparat ions have shown fus ion between the L U V s and the 146 chromaffin granules (results not shown). The fusion event appears to be significantly less leaky in this case. 147 C H A P T E R V D I S C U S S I O N The results presented in this thesis prov ide new in format ion regard ing the roles o f l ip ids in var ious events associated wi th exocytosis. The studies presented i n Chapte r II and III are mot ivated by the observat ion that a rap id cyc l ing o f P I ( through d iacy lg lycero l and P A ) occurs on st imulat ion o f a secretory ce l l , wh i ch is accompanied by in f lux o f Ca 2 * . Th i s leads to the possibi l i ty that P I or a der ivat ive plays an active role i n Ca 2 * t ranspor t The results presented i n Chap te r II suggest that PI remains i n a stable lamel lar organizat ion in the presence o f Ca 2 * , a property wh i ch wou ld not be expected to lead to enhanced Ca 2 * inf lux. However , the studies presented i n Chap te r III prov ide some support for a role o f P A as a C a 2 * ionophore, a l though this capacity and the mechan isms invo lved rema in to be more precisely def ined. The invest igations o f Chap te r I V concern the event subsequent to the in f lux o f C a 2 " , namely the fus ion o f the secretory vesicle wi th the p lasma membrane. The part icular system studied concerns Ca 2 * - s t imu l a t ed interactions between ch romaf f i n granules and P E - P S vesicles, wh i ch prov ide a crude mode l for the interactions between the granules and the inner mono layer o f the p lasma membrane in the intact adrenal ce l l . It is shown that such vesicles fuse readi ly wi th the granules in the presence o f Ca 2 * , lead ing to the poss ib i l i ty that the inner mono layer l i p i d compos i t ion is important to the exocytot ic event i n v ivo. Before progressing to var ious speculat ions wh i ch may be made on the basis o f these studies, it is impor tant to br ie f ly summar i ze the di f f icul t ies invo lved in extrapolat ing these results to processes i n v ivo. 148 W i t h regard to the possible role o f P A as a C a 2 * ionophore i n v ivo, there are at least three prob lems. First, 20 mo l% P A was requ i red in order to observe s ignif icant C a 2 * uptake. Th i s is much larger than the level o f P A normal ly found in the p lasma membrane (approx. 1%). It may be that more unsaturated varieties o f P A are effective at lower concentrations. Th i s remains to be investigated. Second, the uptake process observed in the L U V system is quite slow, a l though the presence o f K* and H~ ionophores can s ign i f icandy enhance the rate. F ina l l y , there remains the quest ion whether P A is generated fast enough in v i vo to fu l f i l a role as a ca lc ium ionophore. The re lat ion between the chromaf f in granu le-ves ic le results and the fus ion events invo lved i n exocytosis is also not unambiguous. First, the h igh concentrat ions o f ca lc ium (1 m M ) requ i red for the fus ion event between the ch romaf f i n granules and the P E - P S vesicles are much h igher than may be expected i n the cytosol. However , i t can be argued that the local C a 2 * concentrat ion at the fus ion site m ight be s igni f icant ly h igher than the cytosol ic Ca 2 * concentrat ion or that proteins, such as synexin, result in h igher effective C a 2 * accumulat ion. Second, it remains to be determined whether the inner mono layer o f the chromaf f in cel l p lasma membrane contains predominant ly P E and PS. F ina l l y , fus ion was observed between the h igh ly curved sonicated vesicle system and the chromaf f in granule, and the re lat ion o f this to fus ion o f the granule wi th the p lasma membrane may be quest ioned. In this regard, p re l im inary exper iments (not inc luded here) suggest that P E - P S (3:1) L U V systems also fuse wi th the chromaf f in granules. Reso lut ions o f these points w i l l require addi t iona l work. However , the results presented here do permi t to a role o f P A as a Ca 2 * ionophore, and do support the poss ib i l i ty that p lasma membrane inner mono layer compos i t ion o f P E and PS cou ld be important to exocytosis. It is o f interest to speculate on the possible mechanisms 149 that cou ld be invo lved. 5.1 Speculat ive M o d e l for C a 2 * Transport Chapte r III indicates that phosphat id ic ac id in large un i lamel lar vesicles can act as a ca l c ium ionophore to transport C a 2 * across a phospho l i p id bi layer. A l t hough the exact molecu lar mechan ism by wh ich phosphat id ic acid is able to translocate Ca 2 * ions is not apparent, the mechan ism must obv ious ly invo lve some sort o f f l i p - f l o p process o f a Ca 2 * -phospha t i d i c ac id complex. The abi l i ty o f P A to adopt non -b i l a ye r organizat ion in the presence o f Ca 2 * gives interest ing possibi l i t ies. L i p i d s present in " inver ted" structures cou ld reside w i th in a hydrophob i c doma in such as the inter ior o f a l i p i d membrane. These structures wou ld have a hydrophob i c exter ior and a hydroph i l i c inter ior, characteristics norma l l y ascr ibed to ionophores such as va l inomyc in . A possib le mechan ism whereby the dynamic format ion o f inverted micel les in a b i layer can act as a permeabi l i ty pathway or "channe l " for both l ip ids and po lar molecules such as C a 2 * / i n F igure 36. Such transient events cou ld be tr iggered by changes in the local concentrat ions o f non -b i l a y e r species (e.g. P A ) wh i ch , in the presence o f C a 2 * give rise to the inverted mice l l a r transport intermediate. 5.2 Speculat ive M o d e l for Exocytos is The re lat ionship between the results presented in Chapte r I V and exocytotic release o f ch romaf f i n granule contents i n v i vo is not immediate ly obvious. However , H o p e and Cu l l i s (1979) have demonstrated that mode l systems composed o f erythrocyte l ip ids i n the propor t ions found in the inner leaf let o f the erythrocyte membrane adopt the hexagonal H n conf igurat ion i n the presence o f Ca 2 * . Thus, i f the inner 150 (a) 888888 8888886 (b) m^mm wsrnm (c) (d ) F igu re 36. Proposed mechan i sm for ' fac i l i ta ted transport o f C a J 151 mono layer o f the adrenal cel l p lasma membrane has a s imi lar l i p i d compos i t ion as the erythrocyte, the presence o f C a 2 * w i l l destabi l ize it in the sense that the preferred conf igurat ion for a large fract ion o f the l i p id w i l l be in the hexagonal H n phase. A s has been indicated elsewhere (Cu l l i s et a l . , 1980), format ion o f hexagonal H n structure f r om prev ious ly b i layer systems appears to proceed as an inter b i layer event. Thus, the strain exper ienced by the inner monolayer o f the p lasma membrane in the presence o f C a 2 * cou ld be re l ieved by interact ing wi th closely apposed granule membranes to f o rm short " inver ted" cy l inders ( H n structure) or inverted mice l les ( l ip id ic part ic les; Verk le i j et a l . , 1979) wh i ch wou ld also prov ide intermediar ies in the exocytot ic event A mode l depict ing such a mechan ism for exocytot ic release is i l lustrated i n F igure 37. C r i t i c i sms o f this mode l and s imi lar models suggesting inverted mice l la r intermediates in fusion ( Lau and Chan , 1975; P in to da S i lva and Nague i ra , 1977; G i nge l and G insbe rg , 1978; Cu l l i s and Hope , 1978) inc lude the fact that " l ip id i c part ic les" wh i ch may be interpreted as ar is ing f rom inverted micel les, only appear to occur wel l after the actual fus ion event in mode l systems (see, for example, Bearer et al., 1982). However , the recent the rmodynamic analysis o f S iegel (1983) suggest that inverted mice l la r intermediates f o rm ing between closely apposed bi layers are l i ke ly to exh ib i t short ha l f - l i v e s ( < 10~4 s). A s a result, Siegel suggests that in a study such as that o f Bearer et a l . (1982) over 5 X 10 } fus ion cross- f ractures wou ld have to be examined to observe a single " l ip id i c part ic le" . 152 Figure 37. Proposed mechanism of exocytotic release of chromaffin granule contents in vivo . PM refers to the chromaffin cell plasma membrane whereas C G denotes the chromaffin granule. 153 154 B I B L I O G R A P H Y 1. Abbs , M.T. and Ph i l l ips , J . H . (1980) B i o ch im . B iophys . Acta . 558: 7 3 -84. 2. A ke rman , K .E .O . and N i cho l l s , D . G . (1983) Trends B iochem. Sci . 8: 63 -64 . 3. Anu i s , D., Hesketh , J .E . and Dev i l l i e rs , G . (1979) C e l l T issue Res. 197: 433 -441 . 4. Bachmann, L. and S c hm i t t - F um i an , W . W . (1973) in F r ee ze -E t ch i ng Techn iques and App l i ca t ions (Benedett i , E.L. and Fava rd , P., Eds.) 63 -72 , Paris. 5. Baker, P.F. (1972) Prog. B iophys . M o l . B i o l . 24: 177-223 . 6. Ba l ly , M.B., Hope , M.J. , van Echte ld , C .J .A . and Cu l l i s , P.R. (1983a) B iochemis t ry (submitted). 7. Ba l ly , M.B., T i lcock, C.P.S., Hope , M . J . and Cu l l i s , P.R. (1983) C a n . J . B i o chem. 61: 346-352 . 8. Ba r r i o , G . J . (1981) Trends B iochem. Sci . 6: 322 -325 : 9. Bearer, E.L., Duzgunes, N. , F r i end , D.S. and Papahadjopoulos, D . (1982) B i o ch im . B iophys . Ac ta . 693: 93 -98 . 10. Berden, J.A., Barker , R .W. and Radda , G.K.. (1975) B i o ch im . B iophys . Ac ta . 375: 186-208. 11. Berr idge, M . J . (1980) Trends Pharmaco l . Sci . 1: 419-424 . 12. Berr idge, M . J . (1981) M o l . Ce l l . Endroc r ino l . 24: 115-140. 13. B laschko, H . and We l ch , A . D . (1953) A r c h . E x p d . Patho l . Pha rmako l . 219: 17 -22 . 14. B laschko, H. , F i r emark , H . , Sm i th , A . D . and W ink l e r , H . (1967) B i o chem. J . 104: 545-549 . 15. B lauste in, M.P., Ratz laft , R .W. and Schweitzer, E.S. (1980) F ed . Proc. 39: 2790-2795. 16. B l i gh , E .G . and Dye r , W . J . (1959) C a n . J . B i o chem. Phys io l . 37: 9 1 1 -917. 17. Boskey, A . L . and Posner, A .S . (1976) Ca lc i f . T issue Res. 19: 273. 18. Bottcher, C.J. , Radda , G . K . and Shennan, C D . (1978) B i o ch im . B iophys . Acta . 513: 321-337 . 19. Branton, D . (1966) Proc. Na t l . A cad . Sc i . ( U S A ) 55: 1048-1056. 20. Branton, D . and Park, J . (1967) J . U l t rast ruc. Res . 19: 283. 155 21. Branton, D., Bul l ivant, S., G i l u l a , N.B., Karnovsky , M.J. , M o o r , H. , Muh le tha le , K., No r t h , D .H . , Parker, L , Satir, B., Satir, P., Spath, V., Staehel in, J.A., Steere, R.L. and Weis te in , R.S. (1975) Science 190: 5 4 -56. 22. Brockerhof f , H . and Hu r kowsk i , M . (1965) Can . J . B i ochem. 43: 1777. 23. Broekhuyse, R . M . (1969) C l i n . C h i m . Acta . 23: 457 -461 . 24. Buck land , R .M . , Radda , G . K . and Shennan, C D . (1978) B i o ch im . B iophys . Acta . 513: 321-337. 25. Cara fo l i , E. and C rompton , M . (1978) Cur r . Top ics M e m b r . Transp. 10: 151-216. 26. Ca ra fo l i , E., C l emen t i . F., D rab i kowsk i , W . and Marg re th , A . (1975) i n C a l c i um Transfer in Cont rac t ion and Secret ion, N o r t h - H o l l a n d , Ams te rdam. 27. Ceccare l l i , B., Me ldo les i , J . , and C lement i , F. (1974) in Advances i n Cytopharmaco logy. V o l 2, 127-139, Raven Press, N .Y . 28. Chand le r , D.E . and Heuser , J .E. (1980) J . C e l l B i o l . 86: 666-674. 29. Chapman , D . (1975) Q u a r t Rev . B iophys . 8: 185-235. 30. Chapman , D. and Penkett, S.A. (1966) Nature ( London) 211: 1304-1305. 31. Chapman , D., U rb i na , J . and Keough , K . M . (1974) J . B i o l . C h e m . 249: 2512-2521. 32. C h i , E.Y., Lagunoff , D . and Koeh le r , J .K . (1976) Proc. N a d . A cad . Sci . ( U S A ) 73: 2823-2827. 33. Chr is t ie , W . W . (1973) L i p i d Ana lys is , 85 -102 , Pergamon Press, Ox f o r d . 34. Comfur ius , P. and Zwaa l , R .F .A . (1977) B i o ch im . B iophys . Acta . 488: 3 6 -42. 35. C r o c k c r o f t S., Bennett, J.P. and Gomper t s , B .D. (1980) Na tu re ( London) 288: 275-277 . 36. Cruetz, C.E., Pazoles, C J . and Po l l a rd , H.B. (1978) J . B i o l . C h e m . 253: 2858-2866. 37. Cruetz , C.E. , Scott, J .H . , Pazoles, C J . and Po l l a rd , H.B. (1982) J . Ce l l u l a r B iochem. 18: 87 -97 . 38. Cu l l i s , P.R. (1976) F E B S L E T T . 68: 173-176. 39. Cu l l i s , P.R. and D e K ru i j f f , B. (1978a) B i o ch im . B iophys . Ac ta . 507: 207-218 . 40. Cu l l i s , P.R. and D e K ru i j f f , B. (1978b) B i o ch im . B iophys . Ac ta . 513: 156 31 -42 . . 41. Cu l l i s , P.R. and D e Krui j fT, B. (1979) B i o ch im . B iophys . Ac ta . 559: 399-420 . 42. Cu l l i s , P.R. and Hope , M.J . (1978) Na tu re ( London) 271: 672-674. 43. Cu l l i s , P.R. and Hope , M . J . (1980) B i o ch im . B iophys . Acta . 597: 533 -542. 44. Cu l l i s , P.R. and Verk le i j . A . J . (1979) B i o ch im . B iophys . Ac ta . 552: 546 -551 . 45. Cu l l i s , P.R., D e KruijfT, B., Hope , M.J. , Nayar , R. and Schmid , S.L. (1980) Can . J . B i ochem. 58: 1091-1100. 46. Cu l l i s , P.R., D e KruijfT, B., Hope , M.J. , Ve rk le i j , A .J . , Nayar , R., Fa r ren , S.B., T i l cock , C.P.S., Madden , T . D . and Ba l ly , M .B . (1983) i n Memb r ane F l u i d i t y in B io logy (A lo i a , R.C., Ed.) V o l . I, Chapte r 2, 3 9 - 81 , A cadem i c Press, N . Y . 47. Cu l l i s , P.R., V an Di jck , P .W.M. , D e Krui j fT, B. and D e G i e r , J . (1978a) B i o ch im . B iophys . Ac ta . 513: 218-230 . 48. Cu l l i s , P.R., Verk le i j , A . J . and Ververgaert, P .H.J .Th. (1978b) B i o ch im . B iophys . Ac ta . 513: 11 -20 . 49. D a h l , G . , Ekerdt , R. and G ra t z l , M . (1979) Symp. Soc. Exp . B i o l . 33: 349-368 . 50. Dan i e l l i , J .F . and Davson , H . (1935) J . C e l l . C omp . Phys io l . 5: 4 9 5 -508. 51. Dav id son , F . M . and Long , C . (1958) B i o chem. J . 69: 458-466 . 52. D e G i e r , J., Manders loot , J .G . and V a n Deenen , L . L . M . (1969) B i o ch im . B iophys . Acta . 150: 666-675. 53. D e G i e r , J . , Manders loot , J .G . and V a n Deenen , L . L . M . (1969) B i o ch im . B iophys . Ac ta . 173: 143-145. 54. D e Krui j fT, B. and Cu l l i s , P.R. (1980) B i o ch im . B iophys . Ac ta . 602: 477-490 . 55. D e Krui j fT, B., Ve rk le i j , A .J . , V a n Echte ld , C.J.A., Ger r i t sen , W.J . , Mombe r s , C , Noo rdam , P.C. and D e G i e r , J . (1978) B i o ch im . B iophys . Ac ta . 555: 218-230 . 56. D e O l i v e i r a - F i l gue i r a s , O . M . , V a n den Besselaar, A . M . H . P . and V a n den Bosch, H . (1979) B i o ch im . B iophys . Ac ta . 558: 73 -84 . 57. Doug las , W . W . (1968) B r i L J . Pha rmaco l . 34: 451-474 . 58. Doug las , W . W . (1975) in Handbook o f Phys io logy (B laschko, H. , Sayers, G . 157 and Smi th , A . D., Eds.) Section 7, V o l . V I : The Adrena l G l a n d , 367-388, W i l l i ams and W i l k i n s , Ba l t imore . 59. Douglas , W . W . and R u b i n , R.P. (1961) J . Phys io l . ( London) 159: 40 -57 . 60. Edwards , W., Ph i l l ips , J . H . and Mo r r i s , S.J. (1974) B i o ch im . B iophys . Acta . 356: 164-173. 61. Eng leman , D . N . (1970) J . M o l . B i o l . 47: 115. 62. Fa r ren , S.B. and Cu l l i s , P.R. (1980) B i o chem. B iophys. Res . C o m m u n . 97: 182-191 . 63. Fa r ren , S.B. and Cu l l i s , P.R. (1983) B iochem. B iophys. Res. C o m m u n . 97: 182-191 . 64. Fe ldberg , W., M i n z , B. and Tsudz imura , H . (1934) J . Phys io l . ( London) 81: 286-304. 65. F i ske , C H . and Subbarow, Y . (1925) J . B i o l . C h e m . 66: 375-400 . 66. Fo l ch , J . , Lees, M . , S loane-Stan ley , G . H . (1957) J . B i o l . C h e m . 226: 497. 67. Fo reman , J . C , Momga r , J . C and Gomper t s , B .D . (1973) Na tu re ( London) 245: 249 -251 . 68. F ra ley , R., Wi l schut , J . , Duzgunes, N., Smi th , C . and Papahadjopoulos, D . (1980) B iochemis t ry 19: 6021-6029. 69. Fuchs , P., Paro la, A . . Robb ins , P.W. and Blout , E.R. (1975) Proc. N a d . A cad . Sc i . ( U S A ) 72: 3351-3354. 70. G i l , D.W., B rown , S.A. Seeholzer, S .H . and W i l dey , G . M . (1983) L i f e Sci. 32: 2043-2046. 71. G i nge l l , D . and G insbe rg , L. (1978) in Ce l l Surface Rev iews (Poste, G . and N i co l son , G .L . , Eds.) V o l . 5, 833-792, N o r t h - H o l l a n d , Ams te rdam. 72. Go i t e r , E. and G rende l , F. (1925) J . Exper . M e d . 41: 439 -443 . 73. G r i f f i n , R . G . (1976), J . A m . C h e m . Soc. 98: 851-853 . 74. Gu tde r , F. and C lausen, J . (1969) B i o chem. J . 115: 959-968 . 75. Hager , P. and Barrnett, R.J . (1960) i n Ad renerg i c Mechan i sms (Vane, R.J., Wo l s tenho lme, G . E . W . and O 'Conno r , Eds.) 83 -99 . 76. H a k o m o r i et a l . J . C e l l B i o l . 88: 127-159 , Parts I, II, III. 77. Haydon , D .A . and Tay lor , J .L . (1968) Na tu re ( London) 217: 739-740 . 78. He l l e , K.B., F la tmark , T., Se r c k -Hannssen , G . and Lonn ing , S. (1971) B i o ch im . B iophys . Ac ta . 226: 1-8 158 79. Heske th , T.R., Smi th , G .A . , Hous lay , M .D . , M c G i l l , K.A., B i rdsa l l , N . J .M. , Metca l fe , J .C. and War ren , G .B . (1976) B iochemist ry 15: 4145-4151. 80. Heuser , J . and Salpeter, R. (1979) J . C e l l B i o l . 82: 150-173 . 81. H i l l a r p , N .A . (1959) Acta . Phys io l . Scand. 47: 271-278. 82. H i l l a r p , N - A . and Hokfe l t , B. (1953) Ac ta . Phys io l . Scand. 30: 55 -68 . 83. H i ra ta , F. and Axe l rod , J . (1978) Na tu re (London) 275: 219-220. 84. H i ra ta , F. and Axe l r od , J . (1983) in Pharmaco log ica l and B iochemica l Aspects o f Neurotransmit ter Receptors (Yosh ida , H . and Yamamu ra , H.T., Eds.) 287-294, John W i l e y and Sons, N .Y . 85. H o k i n , M .R . (1974) in Secretory Mechan i sms o f Exocr ine G l a nd s (Tho rn , N . A and Peterson, O .H . , Eds.) 101-112, Munksgaa rd , Copenhagen. 86. H o k i n , M .R . and H o k i n , L .E. (1953) J . B i o l . C h e m . 203: 967-977 . 87. Ho lmes , R.P. and Yoss , N .L . (1983) Na tu re ( London) 305: 637-638 . 88. Hong , K... Duzgunes, N . and Papahadjopoulos, D. (1982a) B iophys . J . 37: 297-305 . 89. Hong , K., Duzgunes, N. , Ekerdt , R. and Papahadjopoulos, D . (1982b) Proc. N a d . A cad . Sci . ( U S A ) 79: 4642-4644. 90. Hope , M . J . and Cu l l i s , P.R. (1979) F E B S L E T T . 107: 323-326. 91. Hope , M . J . and Cu l l i s , P.R. (1980) B i o chem. B iophys. Res. C o m m u n . 92: 846-852 . 92. Hope , M . J . and Cu l l i s , P.R. (1981) B i o ch im . B iophys. Ac ta . 640: 82 -90 . 93. Hope , M.J. , Ba l ly , M.B., van Echte ld , C .J .A . and Cu l l i s , P.R. (1983) B iochemis t ry (submitted) 94. Hope , M.J., Wa lke r , D .C . and Cu l l i s , P.R. (1983) B iochem. B iophys . Res . C o m m u n . 110: 15-22 . 95. Hube r , E., Kon i g , P., Schuler, G . , Abere r , W. , P lanner , H . and W ink l e r , H . (1979) J . Neu rochem. 32: 35 -47 . 96. Jacobj, C . (1892) A r c h . Exp t i . Pa tho l . Pha rmako l . 29: 171-211 . 97. Jacobson, K. and Papahadjopoulos, D . (1975) B iochemist ry 14: 152-161 . 98. Jones L . M . and M i c h e l l , R . H . (1975) B i o chem. J . 148: 479-485 . 99. Kagayama , M . and Douglas , W . W . (1974) J . C e l l B i o l . 62: 519-526 . 100. Ka t z B. and M i l e d i , R. (1969) J . Phys i o l . ( London) 203: 407-436. 159 101. Kawahara , Y. , Taka i , Y . , M i nakuch i , R., Sano, K and N i sh i zuka , Y . (1980) B i o chem. B iophys . Res. C o m m u n . 97: 309-317 . 102. Lapent ine, E G . and Cuatrecasas, P. (1979) B i o ch im . B iophys. Acta . 573: 394-402 . 103. L au , A . L . Y . and Chan , S.I. (1975) Proc. N a d . Acad . Sci . ( U S A ) 72: 2170-2174. 104. Lever , J .D . (1955) Endocr ino logy 57: 621-635 . 105. Lowry , O .H . , Rosebrough, N.J., F a n , A . L . and Rande l l , R.J. (1951) J . B i o l . C h e m . 193: 265-275. 106. Luzzat t i , V . and Tard ien , A . (1974) A n n . Rev . Phys. C h e m . 25: 79 -94 . 107. Luzzatt i , V., G u l i k - K r z y w i c k i , T. and Tard ien , A . (1968) Nature ( London) 218: 1031-1034. 108. M i c h e l l , R .H . (1975) B i o ch im . B iophys . Ac ta . 415: 81-147 . 109. M i c h e l l , R .H . (1979) Trends B iochem. Sci . 4: 128-131. 110. M i c h e l l , R .H . , Jafferj i , S.S. and Jones, L . M . (1977) Adv . Exp . M e d . B i o l . 83 : 447-464 . 111. M i l l e r , R . G . (1980) Nature ( London) 287: 166-167. 112. M i m m s , L.T., Zamp igh i , G . , Nozak i , Y . and Tan fo rd , C . (1981) B iochemis t ry 20: 833-840 . 113. Mo r r i s , S.J., Schober, R. and Schultens, H .A . (1977) B i o ch im . B iophys . Acta . 464 :65-81 . 114. Mu l de r , E. and Van Deenen , L . L . M . (1965) B i o ch im . B iophys . Ac ta . 106: 348-359 . 115. Newton , C , Pangborn, W. , N i r , S. and Papahadjopoulos, D . (1978) B i o ch im . B iophys . Acta . 506: 281-287 . 116. N i cho l l s , D . G . (1982) i n Bioenerget ics p 30 - 33 , Academic Press, N .Y . 117. Noog le , J .H . Maracek, J.F., Manda l , S.B., van Vanet ie, R., Rogers, J . , Ja in, M . K . and Rami rez , F. (1982) B i o ch im . B iophys . Ac ta . 691: 240-248 . 118. N o r m a n , T .C . (1976) I n t Rev . C y L (Bourne, G . H . and Dan ie l l i , J.F., Eds.) V o l . 46. 1-77, Academ i c Press, N . Y . 119. O l son , R , Hun t , C.A. , Szoka, F.C., V a i l , W . J . and Papahadjopoulos, D . (1979) B i o ch im . B iophys . Ac ta . 557 9 - 2 3 . 120. O p den K a m p , J .A .E . (1979) A rm . Rev . B i o chem. 48: 4 7 - 7 1 . 160 121. Papahadjopoulos, D . (1968) B i o ch im . B iophys . Acta. 163: 240-254. 122. Papahadjopoulos, D . (1978) in Ce l l Surface Rev iews (Poste, G . and N i co l son , G .L . , Eds.) V o l . 5, 766 -791 . N o r t h - H o l l a n d , Ams te rdam. 123. Papahadjopoulos, D., H u i , S., V a i l , W . J . and Poste, G . (1976a) B i o ch im . B iophys . Ac ta . 448: 245-264. 124. Papahadjopoulos, D., Poste, G . and Schaeffer, B.E. (1973) B i o ch im . B iophys . Ac ta . 323, 23 -42 . 125. Papahadjopoulos, D., Poste, G . and Va i l , W . J . (1979) in Me thods i n Memb rane B io logy ( K o r n , E.D., Ed.) V o l . 10, 1-121, P l e num Press, N . Y . 126. Papahadjopoulos, D., V a i l , W.J . , Jacobson, K. and Poste, G . (1975) B i o ch im . B iophys. Ac ta . 394: 483 -491 . 127. Papahadjopoulos , D., V a i l , W.J . , Pangborn , W . A . and Poste, G . (1976) B i o ch im . B iophys. Acta . 448: 265-283 . 128. Patel , K . M . and Sparrow, J.T. (1978) J . C h r o m a L 150: 542-547 . 129. Pershads ingh, H .A . and M c D o n a l d , J . M . (1980) J . B i o l . C h e m . 255: 4087-4093. 130. Ph i l l i ps , J .H . and A l l i son , Y . P . (1977) Neurosc ience 2: 147-152 . 131. P ick, U . (1981) A r c h . B i ochem. B iophys . 212: 186-194. 132. P in to da Si lva, P. and Branton, D . (1970) J . C e l l B i o l . 457: 598-605 . 133. P in to da S i lva, P. and Nogue i ra , M . L . (1977) J . Ce l l . B i o l . 73: 1 6 1 -181. 134. Pletscher, A., D a Prada, M . , Berneis, K .H . , Steffen, H. , Lu to ld , B. and Weder , H . G . (1974) Adv . Cy topharmac. 2: 257-264. 135. Poole, A .R . , Howe l l , J.I. and Lucy , J . A . (1970) Na tu re ( London ) 227: 810 -813 . 136. Port is, A . , Newton , C , Pangborn , W . and Papahadjopoulos, D . (1979) B iochemis t ry 18: 780-790 . 137. Poste, G . and A l l i son , A . C . (1973) B i o ch im . B iophys . Ac ta . 300: 4 2 1 -465. 138. Putney, J .W. (1981) L i f e Sc i . 29: 1183-1194. 139. Putney, J.W., Weiss , S.J., V a n der Waa ls , C M . and Hadd j i , R .A . (1980) Na tu re ( London) 284: 345-347 . 140. R and , R.P. and Sengupta, S. (1972) B i o ch im . B iophys . Ac ta . 255: 484 -492. 161 141. Rand , R.P., T in ier . D .O. and Fast, P .G . (1971) Chem . Phvs. L i p i d s 6: 333-342 . 142. Re i s s -Hu s son , F. (1967) J . M o l . B i o l . 25: 363-382. 143. Reuter , H . (1979) A n n . Rev . Phys io l . 41: 413-424. 144. Rober t son , J .D . (1964) in Ce l l u l a r Membranes in Deve lopment (Locke , M . , Ed.) 1-81, Academic Press Inc., N .Y . 145. Ro thman , J . E and Lcnard , J . (1977) Science 195: 743-753 . 146. R u b i n , H . (1967) i n The Speci f ic i ty o f C e l l Surfaces (Davies, B .D. and War ren , L., Eds.) 181-194, Prent ice H a l l , Eng l ewoon -C l i f f s , N.J . 147. Rub i n , R.P. and Laychock, S.R. (1978) A n n . N .Y . A cad . Sc i . 307: 377-325. 148. Sa lmon, D . M . and Honeyman , T .W . (1980) Nature ( London) 284: 344-345. 149. Seel ig, J . (1978) B i o ch im . B iophys . Ac ta . 575: 105-140. 150. Serhan, C , Anderson, P., G oodma n , E , Du r ham , P. and We issman, G . (1981) J. B i o l . C h e m . 256: 2736-2741. 151. Sheetz, M .P . and Chan , S.J. (1972) B iochemis t ry 11: 4573 152. Siegel , D.P. (1983) B iophys. J . ( in press). 153. S i l insky, E M . (1982) F r ed . Proc. 41: 2169-2171. 154. Singer, S.J. and N i co l son , G . L . (1972) Science 175: 720 -731 . 155. Sjostrand, F.S. and Wetzste in, R. (1956) Exper ient ia . 12: 196-199 . 156. Smi th , A . D . ' and W ink l e r , H . (1967) B iochem. J . 103: 483 -485 . 157. Sundler , R. and Papahadjopoulos, D . (1981) B i o ch im . B iophys . Ac ta . 649: 751-758 . 158. Sundler , R. and Wi jkander , J . (1983) B i o ch im . B iophys. Ac ta . 730: 3 9 1 -394. 159. Sundler , R., Duzgunes, N . and Papahadjopoulos, D. (1981) B i o ch im . B iophys . Acta. 649: 751-758 . 160. Szoke, F. and Papahadjopoulos, D . (1978) Proc. N a d . A cad . Sc i . ( U S A ) 75: 4194-4198. 161. Taka i , Y . , K i sh imo to , A., Iwasa, Y . , Kawaha ra , Y . , M o r i , T . and N i sh i zuka , Y . (1979) J . B i o l . C h e m . 254: 3692-3695. 162. Thach , R.E. and Newburger , M .R . (1972) i n Research Techn iques i n B iochemis t ry and Mo l e cu l a r B io logy, p 12, W . A . Ben jamin , Inc., Ca l i f o rn i a . 162 163. Thayer , A . M . and Koh l e r , S.J. (1981) B iochemist ry 20: 6831-6834. 164. T i l cock, C.P.S. and Cu l l i s , P.R. (1981) B i o ch im . B iophys . Ac ta . 641: 189-201 . 165. T i lcock, C.P.S.. Bal ly , M.B., Far ren , S.B., Cu l l i s , P.R., G rune r , S . M . and K i r k , G . (1983) B iochemis t ry (submitted). 166. Treve lyan , W .E . (1966) J . L i p i d Research. 7 : 445-447 . 167. Tr i fa ro , J . M . (1969) M o l . Pharmaco l . 5: 382-393 . 168. Tyson, C .H . , Zande, H.V. and G r een , D.E . (1976) J . B i o l . C h e m . 251: 1526-1332. 169. V a n Venet ie, R., Leun i ssen-B i j ve l t , J . , Ve rk le i j , A . J . and Ververgaert, P . H J . T h . (1980) J . M i c roscopy 118 : 401-408 . 170. Vance, D.E . and de K ru i j f f , B. (1980) Nature ( London) 288: 277-279. 171. Verk le i j , A . J . V an Echte ld , C.J.A., Ger r i t sen, W.J . , Cu l l i s , P.R., V a n den Bosch, H . and V a n Deneen , L . L . M . (1980) B i o ch im . B iophys . Acta . 600: 620-624. 172. Verk le i j , A .J . , D e Maagd , R., Leun i ssen-B i j ve l t , J . and D e Kru i j f f , B. (1982) B i o ch im . B iophys . Acta . 684: 255-262 . 173. Verk le i j , A .J . , Mombe r s , C , Ger r i t sen, W.J . , Leun i ssen-B i j ve l t , J . and Cu l l i s , P.R. (1979) B i o ch im . B iophys . Acta . 555: 358-362 . 174. Ve rk l e i j , A . J . , Mombe r s , C , Leun i ssen-B i j ve l t , J . and Ververgaert , P .H.J .Th. (1979) Na tu re ( London) 279: 162-163. 175. Verk le i j , A . J . , Zwaa l , R.F.A. , Roe lo fsen, B., Comfu r i u s , P., Kaste l i jn , D., V a n Deenen , L . L . M . (1973) B i o ch im . B iophys . Ac ta . 323: 178-193 . 176. V o n Eu ler , U.S. and Hambe rg , U . (1949) Ac ta . Phys io l . Scand. 47: 2 7 1 -278. 177. Warner , H . G . and Benson, A . A . (1977) J . L i p i d Research. 18: 548-549 . 178. We iner , N . and Rudedge , C O . (1966) i n Mechan i sms o f release o f B iogen ic Am ine s ( V o n Eu ler , U.S., Ros i l l , S. and Uvnas , B., Eds.) 307-318, Pe rgamon Press, Ox f o rd . 179. Wies lander , A., U lm i u s , J . , L i n db l om , G . and Fon te l l , K. (1978) B i o ch im . B iophys . Acta . 554: 340-357 . 180. W i l l i ams , J.A. (1972) Endoc r ino logy 90: 1452-1458. 181. W ink l e r , H . and Smi th A . D . (1975) i n Handbook o f Phys io logy (B laschko, H . . Sayers, G . and Smi th , A .D . , Eds.) Sect ion 7, V o l . V I : The Ad rena l G l a n d . 321-339 , W i l l i ams and W i l k i n s , Ba l t imore . 163 182. W ink l e r , H . and Westhead, E. (1980) Neurosc ience 5: 1803-1823. 183. W i r , S .H . and M c C o n n e l , H . M . (1975) B iochemist ry 14: 847-854. 184. Zaka i , N. , K u l k a , R . G . and Loyter , A . (1977) Proc. N a d . A cad . Sci . ( U S A ) 74: 2411-2421. 185. Zwaa l , R.F.A. , Comfu r i u s , P. and V a n Deenen, L . L . M . (1977) Na tu re ( London) 268 . 358-360. 186. Zwaa l , R.F.A. , Roe lo fsen , B. and Co l l ey , C M . (1973) B i o ch im . B iophys . Ac ta . 300: 159-182. 

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}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            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:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0096591/manifest

Comment

Related Items