UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Targeted liposomes Loughrey, Helen 1989

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

Item Metadata

Download

Media
831-UBC_1989_A1 L68.pdf [ 7.3MB ]
Metadata
JSON: 831-1.0098285.json
JSON-LD: 831-1.0098285-ld.json
RDF/XML (Pretty): 831-1.0098285-rdf.xml
RDF/JSON: 831-1.0098285-rdf.json
Turtle: 831-1.0098285-turtle.txt
N-Triples: 831-1.0098285-rdf-ntriples.txt
Original Record: 831-1.0098285-source.json
Full Text
831-1.0098285-fulltext.txt
Citation
831-1.0098285.ris

Full Text

T A R G E T E D L I P O S O M E S by H E L E N C . L O U G H R E Y 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 D E P A R T M E N T O F B I O C H E M I S T R Y We accept this thesis as conforming to the required 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 S E P T E M B E R 1989 ®HELEN C . L O U G H R E Y , 1989 I n 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 t h e r e q u i r e m e n t s f o r a n a d v a n c e d d e g r e e a t 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 , I a g r e e t h a t t h e L i b r a r y s h a l l m a k e 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 a n d s t u d y . I f u r t h e r a g r e e 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 c o p y i n g o f t h i s t h e s i s f o r s c h o l a r l y p u r p o s e s m a y b e g r a n t e d b y t h e h e a d o f m y d e p a r t m e n t o r b y h i s o r h e r r e p r e s e n t a t i v e s . I t i s u n d e r s t o o d t h a t c o p y i n g 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 g a i n s h a l l n o t b e a l l o w e d w i t h o u t m y w r i t t e n p e r m i s s i o n . D e p a r t m e n t o f BlOCAWfil 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 V a n c o u v e r , C a n a d a D a t e 0ch)Wc5^ \m D E - 6 ( 2 / 8 8 ) A B S T R A C T T h i s thes is presents an o p t i m i z e d a n d gene ra l p r o c e d u r e f o r c o u p l i n g p r o t e i n s to l i p o s o m e s a n d inves t iga tes c e r t a i n aspects o f the i n t e r a c t i o n o f l i p o s o m e s w i t h c o m p o n e n t s o f the c i r c u l a t i o n . T h e o b j e c t o f these s tud ies was to d e v e l o p s t r a i g h t f o r w a r d m e t h o d s f o r the p r e p a r a t i o n o f w e l l c h a r a c t e r i z e d p r o t e i n - l i p o s o m e c o n j u g a t e s w h i c h e x h i b i t e x t e n d e d c i r c u l a t i o n h a l f - l i v e s i n the b l o o d . T h e s e f a v o r a b l e p r o p e r t i e s s h o u l d po ten t ia te the use o f p r o t e i n c o u p l e d ves i c l es i n in vivo a p p l i c a t i o n s s u c h as t a rge t i ng o r d i a g n o s t i c p r o t o c o l s . A g e n e r a l a p p r o a c h f o r the p r e p a r a t i o n o f p r o t e i n - l i p o s o m e c o n j u g a t e s was d e v e l o p e d w h i c h e m p l o y s the h i g h a f f i n i t y b i n d i n g o f s t r e p t a v i d i n f o r b i o t i n a t e d p r o t e i n s . S t r e p t a v i d i n was i n i t i a l l y a t t ached i n a n o n - c o v a l e n t m a n n e r ( v i a b i o t i n p h o s p h a t i d y l e t h a n o l a m i n e ) o r c o v a l e n t l y ( v i a m a l e i m i d o p h e n y l - b u t y r y l p h o s p h a t i d y l -e t h a n o l a m i n e , M P B - P E o r p y r i d y l d i t h i o - p r o p i o n y l p h o s p h a t i d y l e t h a n o l a m i n e , P D P - P E ) to p r e - f o r m e d l i p o s o m e s c o n t a i n i n g the v a r i o u s l i p i d d e r i v a t i v e s . It was s h o w n that the p r o c e d u r e b a s e d o n the m a l e i m i d e d e r i v a t i v e M P B - P E , was the mos t e f f i c i e n t c o u p l i n g m e t h o d . S t a n d a r d p r o c e d u r e s f o r the p r e p a r a t i o n o f M P B - P E h o w e v e r , w e r e f o u n d to resu l t i n a n i m p u r e p r o d u c t . R e c e n t l y a n e w m e t h o d f o r t he syn thes i s o f a p u r e S M P B d e r i v a t i v e o f p h o s p h a t i d y l e t h a n o l a m i n e was d e v e l o p e d ( L e w i s C h o i , u n p u b l i s h e d ) . E f f i c i e n t c o u p l i n g o f p ro te ins to l i p o s o m e s c o n t a i n i n g l o w a m o u n t s o f p u r e M P B - D P P E w a s a c h i e v e d . S u b s e q u e n t l y i t was s h o w n that gen t le i n c u b a t i o n w i t h b i o t i n a t e d p ro te i ns resu l ts i n the r a p i d a n d e f f i c i e n t g e n e r a t i o n o f p r o t e i n c o u p l e d ves i c l es . T h e s e re ta i n t h e i r a b i l i t y to i n t e r a c t w i t h d e f i n e d target cel te. A g g r e g a t i o n o f l i posomes d u r i n g the c o u p l i n g r e a c t i o n is a c o m m o n c o n s e q u e n c e o f the e f f i c i e n t c o u p l i n g o f p r o t e i n to l i posomes . A m e t h o d was t h e r e f o r e d e v e l o p e d f o r the p r e p a r a t i o n o f s m a l l h o m o g e n e o u s l y s i z e d p r o t e i n - l i p o s o m e c o n j u g a t e s b y an e x t r u s i o n p rocess w h i c h does not dena tu re the a t t a c h e d p r o t e i n . T h e s e e x t r u d e d samp les e x h i b i t e d e x t e n d e d b l o o d c i r c u l a t i o n t imes and we re s tab le f o r s i g n i f i c a n t p e r i o d s in vivo. i i The second part of this thesis investigated the in vitro interaction of liposomes of various lipid compositions with platelets. It was demonstrated that large liposomes (> 200 nm in diameter) containing negatively charged lipids (such as EPG) or thiol reactive lipid derivatives (such as MPB-PE) can induce aggregation of platelets in vitro. This interaction was mediated by complement. It is suggested that the formation of platelet-liposome microaggregates in vivo on intravenous administration of negatively charged liposomes, resulted in the transient thrombocytopenia observed in rats. This adhesion event may have also contributed to the rapid removal of aggregated protein-liposome conjugates (containing MPB-PE) from the circulation. i i i TABLE OF CONTENTS ABSTRACT ii TABLE OF CONTENTS iv LIST OF TABLES vii LIST OF FIGURES ix ABBREVIATIONS xi ACKNOWLEDGEMENTS xiii 1. INTRODUCTION 1.1 Model Membranes as Carrier Systems 1 1.2 Liposomes 2 1.3 The Preparation of Liposome Systems 3 1.3.1 Multilamellar Vesicles (MLVs) 3 1.3.2 Unilamellar Vesicles (SUVs and LUVs) 5 1.4 Development of Protein-Liposome Carrier Systems for in vitro and in vivo applications , 7 1.4.1 In vitro characterization of Protein-Liposome Conjugates 10 1.4.2 In vitro and in vivo applications of Protein-Liposome Conjugates 13 1.5 Interactions of Liposomes in vivo : 14 1.5.1 The Mononuclear Phagocytic System 14 1.5.2 Serum Protein-Liposome Interactions 17 1.5.3 Interactions of Liposomes with MPS in vitro 20 1.5.4 Passive Targeting of Liposomes 21 1.6 Summary 22 iv DEVELOPMENT OF A GENERAL METHOD FOR COUPLING PROTEINS TO LIPOSOMES 2.1 Introduction 23 2.2 Materials and Methods 2.2.1 Materials 30 2.2.2 Synthesis of N-(4-(p-Maleimidophenyl) butyryl) egg phosphatidyl-ethanolamine and N-(3-(-pyridyldithio) propionyl) egg phospha-tidylethanolamine 30 2.2.3 Preparation of liposomes 31 2.2.4 Preparation of proteins for coupling .' 32 2.2.5 Coupling of proteins to liposomes 35 2.2.6 Binding of targeted liposomes to rat erythrocytes 36 2.3 Results 2.3.1 Non-covalent method of coupling proteins to liposomes 37 2.3.2 Covalent coupling of proteins to liposomes 43 2.4 Discussion 48 OPTIMIZED PROCEDURES FOR COUPLING PROTEINS TO LIPOSOMES 3.1 Introduction : 52 3.2 Materials and Methods 3.2.1 Materials 52 3.2.2 Synthesis of pure N-(4-(p-Maleimidophenyl) butyryl) dipalmitoyl-phosphatidylethanolamine 53 3.2.3 Preparation of liposomes 54 3.2.4 Assay for maleimide reactivity 54 3.2.5 Preparation of proteins for coupling 55 3.2.6 Coupling of proteins to liposomes 55 3.2.7 In vitro targeting of streptavidin-liposomes 56 3.2.8 Flow cytometry 57 3.3 Results 3.3.1 Characterization of MPB-DPPE 57 3.3.2 Optimized coupling conditions 59 3.3.3 Targeting of streptavidin-liposomes to lymphocyte subpopulations in vitro '. 63 3.4 Discussion '. 66 4. PROTEIN-LIPOSOME CONJUGATES WITH DEFINED SIZE DISTRIBUTIONS 4.1 Introduction 69 4.2 Materials and Methods 4.2.1 Materials 70 4.2.2 Preparation of liposomes 70 4.2.3 Preparation of proteins for coupling 70 4.2.4 Coupling of proteins to liposomes 71 4.2.5 Preparation and characterization of extruded protein-liposomes : 72 4.2.6 In vitro studies of streptavidin-liposome conjugates 72 4.2.7 In vivo studies of liposomes and streptavidin-liposome conjugates 73 vi 4.3 Results 4.3.1 Effect of coupling proteins to liposomes on vesicle size 74 4.3.2 Extrusion of protein-liposome conjugates 77 4.3.4 In vitro characterization of extruded streptavidin-liposome conjugates 81 4.3.5 In vivo properties of extruded streptavidin-liposome conjugates 84 4.4 Discussion 87 5. THE BINDING OF PHOSPHATIDYLGLYCEROL LIPOSOMES TO RAT PLATELETS IS MEDIATED BY COMPLEMENT 5.1 Introduction 93 5.2 Materials and Methods 5.2.1 Materials 94 5.2.2 Preparation of platelets and plasma 95 5.2.3 Preparation of liposomes 96 5.2.4 Assay of platelet-liposome interaction 96 5.2.5 Gel electrophoresis and Western blots 97 5.3 Results 5.3.1 Lipid dependence 98 5.3.2 Requirement of complement 102 5.4 Discussion 109 6. SUMMARIZING DISCUSSION 114 BIBLIOGRAPHY 117 vii LIST OF TABLES 1.1 Examples of the approaches to the covalent coupling of proteins to liposomes 8 1.2 Cells of the Mononuclear Phagocytic System 16 2.1 Binding of targeted biotin PE liposomes to rat erythrocytes 42 2.2 Effect of lipid composition on coupling of streptavidin to MPB-EPE liposomes ... 46 3.1 Thiol dependence of streptavidin coupling to MPB-DPPE liposomes 62 3.2 Binding of biotinated proteins to streptavidin-liposome conjugates 64 4.1 Factors affecting the aggregation of extruded streptavidin-liposomes 83 4.2 Targeting of aggregated and extruded streptavidin-liposomes to erythrocytes ...... 86 4.3 Stability of streptavidin-liposome conjugates in vivo 89 5.1 Effect of EPG liposome concentration on platelet count 101 viii LIST OF FIGURES 1.1 Diagrams of freeze-fracture electron micrographs of multilamellar vesicles (MLVs) and unilamellar vesicles 4 1.2 Mechanism of interaction of target liposomes with cells 12 2.1 Modification of proteins with amine reactive N-hydroxysuccinimidyl derivatives 25 2.2 Phosphatidylethanolamine derivatives of SPDP, SMPB and NHS-biotin 26 2.3 Non-covalent (a) and covalent (b) methods of coupling streptavidin to liposomes 28 2.4 Coupling of streptavidin to liposomes containing biotin PE 38 2.5 Influence of cholesterol on coupling of streptavidin to biotin PE liposomes 40 2.6 Effect on coupling of levels of biotin per antibody and incubation ratios of anti-body to streptavidin-liposomes 41 2.7 Covalent coupling of anti-rat erythrocyte IgG to MPB-EPE and PDP-EPE liposomes 44 2.8 Effect of vesicle size on the extent of streptavidin coupled to MPB-EPE liposomes 47 2.9 Coupling of biotinated anti-rat erythrocyte IgG to streptavidin-liposomes 49 3.1 Structure of two MPB-PE derivatives 58 3.2 pH profile of reactivity of MPB-DPPE liposomes: Extent of streptavidin coupling and maleimide reactivity 60 3.3 Optimal coupling conditions for the preparation of streptavidin-liposomes 61 3.4 Targeting of streptavidin-liposomes to human peripheral lymphocytes 65 4.1 Effect of coupling streptavidin to MPB-DPPE liposomes on vesicle size 75 4.2 Freeze fracture of streptavidin-liposome conjugates 76 4.3 Extrusion of streptavidin-liposome conjugates 78 i x 4.4 Extrusion of antibody-Iiposome conjugates 79 4.5 Freeze fracture of streptavidin-liposomes before and after extrusion 80 4.6 Examination of the stability of extruded streptavidin-liposomes by QELS 82 4.7 Binding of biotinated antibodies to extruded streptavidin-liposomes 85 4.8 In vivo clearance behavior of liposome preparations 88 5.1 EPG dependent formation of platelet liposome microaggregates 99 5.2 Effect of plasma concentration on the interaction of EPG liposomes with platelets 100 5.3 Binding of EPG liposomes to platelets: requirement of plasma 103 5.4 Effect of MPB-DPPE liposomes on platelet count 104 5.5 Determination of the plasma component(s) involved in EPG liposome-platelet inter-actions .". 105 5.6 Effect of removal of C3 from plasma 107 5.7 Western blot of protein bound to EPC and EPG liposomes using anti-rat C3 IgG 108 5.8 The classical and alternative pathways of complement 110 5.9 Proteolytic cleavage fragments of the third component of complement I l l x ABBREVIATIONS BSA C3 CHOL DMF DTNB DTT EDTA EGTA EPPS EPPS pH 8.0 HBS HDL HEPES Liposomes FATMLVs LUVs MLVs SUVs lg MES MPS NEM NHS-biotin bovine serum albumin complement component 3 cholesterol dimethylformamide dithiobis-2-nitro-benzoic acid dithiothreitol ethylene diamine tetra-acetic acid ethyleneglycol-bis-(b-aminoethyl ether) N , N -tetraacetic acid N-(2-hydroxyethyl) piperazine-N'-3-propanesulphonic acid 25 mM EPPS, 150 mM NaCl, pH 8.0 25 mM HEPES, 150 mM NaCl, pH 7.5. high density lipoprotein N-(2-Hydroxyethyl) piperazine-N'-2-ethanesulphonic acid frozen and thawed multilamellar vesicles large unilamellar vesicles multilamellar vesicles small unilamellar vesicles immunoglobulin 2-(N-Morpholino) ethanesulphonic acid mononuclear phagocyte system N-ethylmaleimide N-hydroxysuccinimidyl biotin x i NMR nuclear magnetic resonance PBS phosphate buffered saline Phospholipids EPC egg phosphatidylcholine EPE egg phosphatidylethanolamine EPG egg phosphatidylglycerol DOPC dioleoyl phosphatidylcholine DPPC dipalmitoyl phosphatidylcholine DPPE dipalmitoyl phosphatidylethanolamine PDP-EPE N-[3-(2-Pyridyldithio)-propionyl] egg phosphatidylethanolamine MPB-DPPE N-[4-(p-Maleimidophenyl)-butyryl] dipalmitoyl phosphatidylethanolamine MPB-EPE N-[4-(p-Maleimidophenyl)-butyryl] egg phosphatidylethanolamine Biotin EPE biotin egg phosphatidylethanolamine BPS bovine brain phosphatidylserine PPP platelet poor plasma PRP platelet rich plasma QELS quasi-elastic light scattering SATA N-succinimidyl S-acetylthioacetate SD standard deviation SDS sodium dodecyl sulfate SMPB N-succinimidyl 4-(p-maleimidophenyl) butyrate SPDP N-succinimidyl 3-(2-pyridyldithio) propionate TLC thin layer chromatography on silica TRIS 2-amino-2-hydroxy-methylpropane-1,3-diol xii ACKNOWLEDGEMENTS I would like to acknowledge the members of Pieter Cullis's lab who created a fun atmosphere and an enthusiastic environment to work in. In particular I would like to thank Marcel Bally, Louis Choi, Kim Wong, Larry Reinish and Mick Hope for their support and contribution to various aspects of this work. I thank Tom Redelmeier for the availability of his computer and Richard Harrigan for his unlimited knowledge of computer technology; both have significantly helped in making the completion of this thesis less of a task. I would also like to thank Dr. D. Devine and Prof. R. McMaster for helpful comments throughout my Ph.D. Foremost, I want to thank Pieter for encouraging me to join the lab in my first year at UBC. I appreciate the excitement, energy and enthusiasm he has shown throughout my time in the lab. Finally, I thank my friends who have made my stay in Canada an enjoyable and educational experience that will be hard to surpass. This work (and play) was funded by a University Graduate Student Fellowship, a grant to Pieter Cullis from the National Cancer Institute of Canada and Lipex, Canada. x i i i To Mum and Dad x i v 1. INTRODUCTION 1.1 Model membranes as carrier systems In the early sixties, dispersions of phospholipids in aqueous media were visualized by electron microscopy as " minute sacs " termed "lipid somes" (Bangham et al., 1965). These lipid systems which form spontaneously in the presence of water, exhibit a closed bilayer structure similar to that observed for phospholipid in biological membranes. Such "model membrane" systems have played an important role in the understanding of the structural and physical properties of lipids in biological membranes. The observation that these structures may encapsulate ions and small, water-soluble molecules in their aqueous interior, thus sequestering them from the external environment also suggested an obvious potential for liposomes as drug carrier systems. In this regard, the ability to specifically target drugs to intracellular sites in vivo has long been a major aim of medical research. At present, chemotherapeutic regimes are often limited by the toxic side effects that arise due to systemic delivery of the drug to sensitive tissue not associated with the disease site. The potential application of liposomes as targeted drug carriers in a manner envisioned by Ehrlich (1906) has therefore generated considerable interest. The attachment of various targeting ligands has additional applications ranging from targeted liposome systems (Heath, 1987; Machy and Leserman, 1987) to liposomally based immunological and diagnostic agents (Ishimori et al., 1984; Ho et al., 1985), vaccines (Allison et al., 1974) as well as a basic tool in biochemistry (Machy and Leserman, 1987). The diversity of applications for protein-liposome conjugates has lead to the development of numerous methods for the attachment of proteins to liposomes (reviewed by Heath., 1987; Machy and Leserman, 1987). However, a general, optimized and flexible method for the preparation of protein coupled vesicles has not yet been achieved. A major part of this thesis is directed towards developing such systems (Chapters 2 and 3). 1 This introductory chapter is divided into two parts. First, various types of liposome systems and methods for the preparation of protein-liposome conjugates are summarized. The in vitro characterization of these conjugate systems with regard to retention of the functional activity of the conjugated protein and possible mechanisms for the targeted delivery of liposomally encapsulated materials to cells, are also discussed. This serves as background for the studies on targeted liposomes presented in this thesis. Second, the interaction of liposomes with components of the blood are outlined. It is well established that physical characteristics of liposomes such as size and charge can dramatically affect their in vivo behavior, presumably due to the different interactions they experience in vivo. These interactions also influence the application of protein-liposome conjugates in vivo. Studies on the interaction of liposomes with certain components of the circulation therefore constitutes a second part of this thesis. 1.2 Liposomes Potential drug carrier systems range from macromolecules (antibodies) and blood cells (erythrocytes and platelets) to synthetic carrier systems (various types of microspheres, nanocapsules and liposomes; reviewed by Machy and Leserman, 1987). Liposomes can be suggested to be the most attractive system. This is in part due to the range of lipids that can be employed which allows parameters such as release rates of encapsulated material and surface charge to be readily regulated, as well as the ease of preparing liposomes of variable size (see below; also reviewed by Hope et al., 1986) and the biodegradable nature of this delivery system. Furthermore, the amphipathic nature of liposomes facilitates the entrapment of a variety of lipophilic agents (Hashimoto, 1985; Perez-Solar et al., 1988), as well as aqueous soluble molecules such as cytotoxic drugs (Lopez-Berestein et al., 1984; Mayer et al., 1985 ), fluorescent probes (Weinstein et al., 2 1977; Allen, 1984) and larger biologically active macromolecules such as proteins (Panus and Freeman, 1988) and DNA (Wong et al., 1980). There are two major types of liposomes as defined by the lamellarity of the preparation (Figure 1.1). The first are multilamellar vesicles which are "onion skin" structures with alternating concentric rings of phospholipid bilayers and water. The second are unilamellar vesicles which exist as single bilayers surrounding an aqueous compartment. These systems are classified as small (SUVs) or large (LUVs) according to their size (SUVs, < 50 nm in diameter; LUVs, 50-200 nm in diameter). 1.3 The preparation of liposome systems A number of techniques for the preparation of liposomes exist (reviewed by Szoka and Papahadjopoulos, 1980; Hope et al., 1986). The method of choice is dependent on the application. In this section standard protocols for the preparation of liposomes most commonly employed as carrier systems are described, with particular emphasis given to three structural properties: the size of the vesicles, the number of lamellae and the trapped volume. 1.3.1 Multilamellar Vesicles (MLVs) MLVs are the first model membrane system that was developed (Bangham et al. 1965). These are formed by mechanical dispersion of a lipid film in an aqueous solution. The resulting vesicles are multilamellar, typically have diameters in excess of 1000 nm, and are heterogeneous in size (0.05-10 nm) and lamellarity. Preparations consisting of net neutral lipids such as EPC have relatively low trap volumes (-0.5 ul / /rniole lipid) due to close packing of the lamellae. The internal trap volume can be significantly 3 Figure 1.1 Diagrams and freeze-fracture electron micrographs of multilamellar vesicles (MLVs) and unilamellar vesicles. MULTILAMELLAR UNILAMELLAR VESICLES VESICLES fe. S o£^S ° * ^ b 4- TO 0*55 az. V* c*i^ o?2 ®2 * * °L**> dx** ag K w <r i i s i fa Diameter : 0-2 —10 ixm 002 — 02 jxm 4 increased by the incorporation of charged lipids. This results in swelling of the vesicles due to electrostatic repulsion between bilayers. Successive cycles of freezing and thawing also yield MLVs with increased trapped volumes (termed FATMLVs, Mayer et al., 1985a). Such liposomes are also heterogeneous in size and often exhibit unique intravesicular structures trapped between lamellae. Several procedures have been developed for the preparation of MLVs by "reverse phase" techniques. This approach involves the hydration of lipids directly from an organic solvent. Examples include" vesicles prepared from an ether-lipid solution containing buffer by bath sonication (stable plurilamellar vesicles, SPLVs) or by removal of ether under reduced pressure (REVs). Such liposome preparations can have large trapped volumes (up to 10 /d//imole lipid). A further approach which'yields liposomes exhibiting more efficient encapsulation of aqueous solutes involves the dehydration of lipids from an aqueous solution by either freeze drying or direct drying under vacuum followed by controlled rehydration (Kirby and Gregoriadis, 1984). In summary, a variety of methods are available for the preparation of MLVs which have large entrapped volumes. These systems have the capacity to encapsulate from 20 to 70 % of an aqueous phase which may contain small molecules, proteins or DNA. However, the MLV system suffers from the major disadvantage that the vesicles are large and heterogeneous in size and lamellarity which limits their general utility. For this reason, procedures for generating unilamellar liposomes of defined size have been developed. 1.3.2 Unilamellar vesicles (SUVs and LUVs) SUVs with size distributions of approximately 25-50 nm in diameter are generally prepared from MLVs by sonication (Huang, 1969) or French press techniques (Barenholz, 1979). These systems offer certain advantages for drug delivery applications. 5 However, the small size of these vesicles results in a small radius of curvature which can lead to vesicle instability (as evidenced by a tendency to fuse to form larger systems) and furthermore limits the aqueous entrapped volume (< 0.2 ul / /rniole lipid). The disadvantages of SUVs are avoided by the preparation of larger unilamellar vesicles (LUVs). A straightforward method for the preparation of stable LUVs involves the sequential extrusion of MLVs through polycarbonate filters of various sizes (Hope et al., 1985). This is a rapid procedure for the generation of homogeneously sized populations of vesicles with sizes of 50-200 nm in diameter. Furthermore, extrusion can be carried out at high lipid concentrations, resulting in high aqueous trapping efficiencies and has the particular advantage over other methods in that exposure to reagents such as organic solvents or detergents is avoided, as are harsh" procedures such as sonication. Other useful methods for the preparation of LUVs involve the hydration of lipids from organic phases. A widely accepted method for L U V production employs the previously noted "reverse phase" evaporation technique. This procedure involves the formation of an emulsion of lipid in solvents such as diethylether with aqueous buffer followed by careful removal of organic solvent under reduced pressure (for a detailed description see Szoka and Papahadjopoulos, 1978). The resulting dispersion is a heterogeneous mixture of multi- and unilamellar vesicles which requires extrusion through polycarbonate filters to achieve relatively homogeneous preparations of unilamellar vesicles. The general application of this technique is limited to the encapsulation of materials which are insensitive to the possible denaturing effects of solvents and is further restricted by the solubility of various lipids in the solvent employed. Advantages include high trapping efficiencies and the ability to produce vesicles at high lipid concentrations. Detergent methods, which were first devised for reconstitution of integral membrane proteins into lipid bilayers have also been employed in the preparation of 6 LUVs (Kagawa and Packer, 1971; Mimms et al., 1981). Such procedures are dependent on the type of lipid used, result in low trapping efficiencies and are very time consuming due to the methods required for the removal of detergent from the bilayer. Furthermore, elimination of detergent is a major problem which may affect the permeability properties of the preparation. For these reasons, this method has generally been confined to reconstitution of transmembrane proteins in LUVs. In summary, LUVs prepared by extrusion techniques and in combination with reverse phase evaporation procedures represent versatile methods for the generation of L U V systems and the encapsulation of bioactive materials. These homogeneously sized liposome preparations can also be prepared in the optimal size range for drug delivery to cells (50-100 nm in diameter, see Section 1.4.1). 1.4 Development of protein-liposome carrier systems for in vitro and in vivo applications Protein-liposome conjugates include carrier systems which have the potential to bind specifically to soluble or membrane associated target ligands. Numerous techniques have been developed for the preparation of these systems (reviewed by Heath, 1987; Machy and Leserman, 1987). The most widely used procedures have employed chemical coupling regimes, which involve either the incorporation of a hydrophobic derivative of the targeting molecule into liposomes or the direct covalent attachment of the molecule of interest to pre-formed vesicles (see Table 1.1; also reviewed in Chapter 2, Section 2.1.). Presently, major limitations of coupling methods concern variability with respect to coupling efficiency, exposure to harsh conditions during chemical coupling or preparation of liposomes as well as a lack of generality and flexibility. The use of protein-liposome conjugates in targeting requires the retention of the binding activity of liposomally bound protein and also in some cases the cell specific delivery of liposomally encapsulated contents. In this section, in vitro 1 Table 1.1. Examples of the approaches to the covalent coupling of proteins to liposomes. Proteins coupled to lipids before the formation of liposomes Derivatized form of Ligand reactive group Reference lipid or fatty acid and/or type of ligand N-E-(5-dimethylamino- sulphydryl Sinha and Karush, 1979 naphthaline-1 -sulf onyl groups -L-lysine coupled to PE Hydroxysuccinimide amino groups Huang et al., 1980 ester of palmitic acid Carbodiimide derivative amino groups of Janson and Mallet, 1981 of PE citraconylated F(ab')2 Amino containing lipids binds non-covalently Bayer et al., 1984 reacted with N-hydroxy to avidin labelled Urdal and Hakomori, 1980 -ester of biotin components 8 Table 1.1 (continued) Proteins coupled after the formation of liposomes Derivatized form of Ligand reactive group Reference lipid or fatty acid and/or type of ligand Carbodiimide carboxyl groups of: derivatized PE IgG Dunnick et al., 1975 streptavidin Rosenberg et al., 1987 PE modified with amino groups Torchilin et al., 1978 glutaraldehyde or dimethylsuberimidate Periodate oxidized- amino groups of: reduction of IgG Heath et al., 1981 gangliosides avidin Urdal and Hakomori, 1980 PE reacted with sulphydryl SPDP groups of: PDP-protein A Leserman et al., 1980 Fab Martin et al., 1981 PE reacted with sulphydryl SMPB groups of Fab Martin et al., 1982 PDP-IgG Bragman et al., 1984 SATA-IgG Derksen and Scherphof, 1985 9 characterizations of protein-liposome conjugates are reviewed in relation to these points. A brief description of the applications of such conjugates in vitro and in vivo follows. 1.4.1 In vitro characterization of protein-liposome conjugates The specificity of the interaction of antibody bearing liposomes with cells has been clearly shown in vitro by the absolute requirement for the appropriate targeting molecule for binding to occur. This has been demonstrated by the selective association of liposomally encapsulated fluorescent markers with cells to levels which reflect the density of the target antigen (Leserman et al., 1981). For most applications, the intracellular delivery of liposomal contents is essential. The delivery of liposomally entrapped material such as cytotoxic drugs to defined cell populations (determined by the inhibition of DNA synthesis or cell death) has been achieved under mixed cell culture conditions (Leserman et al., 1981; Heath et al., 1983). Thus, encapsulated biological agents which are normally non-specific in nature or which are inactive due to an inability to cross the cell membrane, may be delivered to cells in a specific manner by prior attachment of a relevant targeting molecule to liposomes. In this regard, it has been shown that the nature of the antigenic determinant as well as the cell type play a critical role. For example, the presence of an antigen on a defined cell population and subsequent binding of targeted liposomes to such cells, does not necessarily result in the expression of the cytotoxic effect of the liposomally encapsulated drug (Weinstein et al., 1978; Machy et al., 1982). This indicates that the association of an antigen with a cell population does not necessarily result in internalization of protein-liposome conjugates. However, while these limitations may restrict the capacity of this system to selectively influence certain cell populations in vitro, they may not be so critical in the in vivo scenario. For example, localization of liposomes to defined sites in vivo by targeting molecules, followed by leakage of 10 liposomally encapsulated drug in the vicinity of the target site should result in a greater therapeutic index than achieved by systemic administration of unencapsulated material. Targeted liposomes can interact with cells in several different ways (Figure 1.2). It is generally accepted that the cytoplasmic delivery of liposomally encapsulated agents to cells proceeds by an endocytic mechanism and not a fusion process (Machy and Leserman, 1987). This is suggested by the inhibition of the cytotoxic effect of encapsulated drugs on treatment of cells with chloroquine or ammonium chloride (both increase the intra-lysosomal pH), by incubations at 4°C, by the inclusion of cytochalasins (which act on microfilaments), cholchicine (which acts on microtubules) or finally by the addition of metabolic inhibitors such as 2-deoxy-D-glucose and NaN3 (Leserman et al., 1981; Heath et al., 1983; Huang et al., 1983). Furthermore, with the exception of phagocytic cell types, it has been clearly demonstrated that large liposomes are.: internalized less efficiently than small liposomes, with an upper size restriction of approximately 100 nm (Machy et al., 1983; Matthay et al., 1984; Heath et al., 1985). The cellular uptake of liposomes by endocytosis results in exposure of liposomally encapsulated materials to acidic conditions, proteolytic enzymes and as indicated above, limits the size of the vesicle carrier system employed. A number of applications perceived for protein-liposome conjugates are restricted by these factors (examples are use of pH sensitive drugs such as cytosine arabinoside and the delivery of protease susceptible molecules such as proteins). For these reasons, alternative routes for the delivery of liposomal reagents have been proposed. In particular, targeted delivery of liposomally encapsulated materials such as DNA has been achieved by combining the use of immunoliposomes with electroporation techniques (Machy et al., 1988) and by fusion mediated by polyethylene glycol (Godfrey et al., 1983) or by liposomally bound fusogenic proteins from Sendai virus (Gitman and Loyter, 1984). The potential of pH sensitive immunoliposomes (Connor and Huang, 1986) in mediating delivery of liposomally entrapped material has as yet to be investigated. 11 Figure 1.2 Mechanisms of interaction of targeted liposomes with cells. A. Non-specific binding Sgg o= so B. Specific binding C. Endocytosis 12 In summary, target proteins (such as antibodies) when conjugated to liposomes retain their biological function. This technology can result in the targeting of liposomes to cells. Subsequent uptake and release of liposome contents to the cell interior limits the type of liposome employed and the nature of the encapsulated contents. However, in the case of cytotoxic drugs, it is likely that localization and subsequent slow release of liposome contents to target tissues in vivo will have significant benefit, even though the liposome itself may not reach the cell interior. 1.4.2 In vitro and in vivo applications of protein-liposome conjugates The targeted delivery of liposomally encapsulated material to defined cell populations in vitro has many potential applications. Such carrier systems can selectively ... ; • destroy contaminating cells in vitro such as residual tumor cells (Heath et al., 1983) or mature T cells (Machy and Leserman, 1984) which are capable of provoking graft versus host disease (Good et al., 1983). Protein-liposome conjugates have also been exploited in liposome immunoassays which are based on agglutination (Kung et al., 1985) or complement mediated lysis (Ishimori et al., 1984) of hapten-bearing liposomes. Furthermore, diagnostic assay procedures have been described which exploit particular biophysical properties of protein coupled vesicles such as liposome instability which is induced by the binding of antigen to the protein portion of the conjugate (Ho and Huang, 1985). This technology also represents an interesting tool for the study of biological phenomena such as the mechanism of endocytosis of membrane determinants (Machy and Leserman, 1987) and has been further exploited in the delivery of macrornolecules such as DNA to specific cells in vitro (Machy et al., 1988). At present, the in vivo use of protein-liposome conjugates is restricted to applications which require rapid removal of molecules, such as toxic chemicals, from the blood (Campbell et al., 1980) as protein coupled vesicles are rapidly cleared from the 13 circulation (Papahadjopoulos and Gabizon, 1987). A systematic study which evaluates the contribution of various factors to this in vivo clearance behavior has as yet to be initiated. Some reports indicate that the chemical attachment of protein to liposomes can result in an increase in the polydispersity of the original liposome preparation (Bredehorst, 1986). One study suggests that the efficient coupling of proteins to liposomes results in aggregation of vesicles (Heath et al., 1980). It is plausible that the enhanced blood clearance behavior of protein coupled vesicles arises in part due to the size of aggregated protein-liposomes which would be removed from the circulation in a size dependent manner (see Section 1.5.1). In this regard, studies on the effect of attaching proteins to liposomes on the aggregation of liposomes as well as on their in vivo clearance behavior have been performed in this thesis (Chapter 4). 1.5 Interactions of liposomes in vivo Many in vivo applications envisioned for protein-liposome conjugates require extended half-lives in the circulation and accessibility to various anatomical sites in the body. In this regard, the effect of different physical characteristics of liposomes on the in vivo behavior of liposomes are discussed in this section, with particular emphasis given to methods of manipulating the in vivo biodistribution and behavior of liposomes. As the usual mode of administration of liposomes is generally intravenously (i.v.), the effect of exposure of liposomes to blood components will also be addressed. 1.5.1 The Mononuclear Phagocytic System Following intravenous (i.v.) injection, liposomes are rapidly cleared from the blood via uptake by organs and cells of the mononuclear phagocytic system (MPS; Poste et al., 1982, 1983). This system consists of various tissue cell types as well as circulating 14 cells such as monocytes (Table 1.2, Bradfield, 1984). In vivo studies have shown that the bulk of the injected vesicles accumulate in the liver and spleen and to a lesser extent in the lung, lymph nodes and bone marrow (Poste, 1983). It is generally accepted that the active phagocytic capacity of cells in these organs results in the internalization of liposomes. However, as this process may occur slowly, it is possible that the initial localization of liposomes to a specific organ is more a function of the constraints imposed by the vasculature. For example, liposomes can enter spaces outside the blood in organs where vessels are lined with sinusoidal capillaries as found in the liver, spleen and bone marrow (Weiss and Greep, 1977). This type of passive filtration may result in the concentration of a liposomally encapsulated drug at a desired location in vivo, as similar incomplete capillary networks are found at many sites of infection and inflammation as well as at several types of solid tumors (Peterson,,L979; EQlkman^ J98JD-. In general, i.v. use of liposomes as targeted carrier systems is restricted by the inaccessibility of liposomes to various extravascular sites. The biodistribution of liposomes can be significantly altered by administration of liposomes intraperitoneally (i.p.), intramuscularly (i.m.) or subcutaneously (s.c; reviewed by Poste, 1983; Machy, 1987). Liposomes injected at these sites can persist for considerable times and as illustrated by i.p. injections, liposomes may also enter alternative compartments such as the lymphatics (Ellens et al., 1981). Passive (i.e. the natural fate of non-targeted liposomes in the body) or targeted localization of liposomes to various anatomical sites in vivo may be enhanced by extending the circulation times of liposomes' in the blood. To achieve this aim, an understanding of factors which influence the in vivo behavior and biodistribution of liposomes is required. At least four major factors have been implicated to play significant roles. First, vesicles composed solely of phospholipids are rendered leaky in serum or plasma (see Section 1.5.2). This effect is remedied to a large extent by the incorporation of cholesterol (Gregoriadis and Davis, 1979) and by using phospholipids 15 Table 1.2 Cells of the Mononuclear Phagocytic System (MPS). Cell type Tissue Kupffer cells liver Histiocytes connective tissue Macrophages lung spleen bone marrow peritoneal cavity Monocyte blood Osteoclast bone Microglia nervous system 16 with high gel to liquid crystalline transition temperatures (Blok et al., 1975; Senior et al., 1982). Second, small liposomes such as SUVs composed of long chain, saturated phospholipids are cleared less readily than larger liposomes of similar composition (Senior, 1987). For example, the circulation half-life of a liposomal preparation is increased from minutes to hours when SUVs are used in place of MLVs. Third, high lipid doses extend the circulation residence times of liposomes in the blood which has been attributed to the saturation of MPS mediated uptake. This observation has been exploited in the development of long lived liposomes by pre-dosing animals with liposomes or other inert particles which localize in MPS cells (Abra et al., 1980). Such an approach, termed MPS blocking (Abra et al., 1980) is not particularly advantageous as the MPS system plays a critical role in host defense function (Altura,' 1980) and such methods result in MPS dysfunction (Poste, 1983).; Finally,.negatively^charged, liposomal systems are cleared more rapidly than neutral vesicles, an effect that may be related to the greater affinity of serum proteins for charged liposomes (see Section 1.5.2). Thus, small liposomes (< 100 nm in diameter) composed of cholesterol and long chain, saturated phospholipids at high lipid doses yield stable liposome preparations which exhibit long circulation half-lives in the blood. More recently, it has been shown that the presence of gangliosides such as G M i can further increase the circulation life-times of liposomes (Gabizon and Papahadjopoulos, 1988). These long lived liposome preparations offer potential advantages for "passive" targeting to tumors and other tissues as significant levels of these liposomes were detected in body compartments such as implanted tumors, the carcass and skin (Gabizon and Papahadjopoulos, 1988). 1.5.2 Serum protein-liposome interactions Plasma components play an important role in the behavior of liposomes in vivo as initially indicated by the instability of certain liposome preparations in the presence of 17 plasma (reviewed by Senior, 1987). Furthermore, the assumption of an overall negative liposome surface potential on incubation of liposomes with plasma (Black and Gregoriadis, 1976) and the association of plasma proteins with liposomes of various compositions in a differential manner (reviewed by Bonte and Juliano, 1986), indicates that the interaction of specific plasma proteins with different liposome preparations may be a major factor which determines their fate in vivo. In this part of the introduction, a brief description of the types of proteins which interact with liposomes will be given. This includes lipoproteins, complement factors and other serum proteins as well as certain circulating cells. The contribution of plasma lipoprotein components to the instability of liposome preparations in the circulation has been extensively studied in vitro in order to develop non-leaky liposomes for in vivo applications (Allen and Cleland, 1980). The key role of a specific plasma lipoprotein population in liposome destabilization has been implied by the transfer of liposomal phospholipid to HDL lipoproteins with concomitant release of an entrapped aqueous marker (Chobanian et al., 1979). This loss of liposomal integrity is characteristic of liposome preparations composed of pure phosphatidylcholines and is more pronounced for liquid crystalline systems. The major apolipoprotein of HDL, ApoA-1 has been implicated in this destabilization effect (Klausner et al., 1985), however the physiological significance of these observations is not yet well defined. Finkelstein and Weissmann (1979) were the first investigators to suggest that complement components, which are involved in diverse processes (see review by Whaley, 1985) may induce liposome instability. More recently, specific complement components have been shown to associate with negatively charged vesicles (Thielens and Colomb, 1986) ., This interaction with anionic liposomes is further indicated by the consumption of complement (as measured by haemolysis of antibody coated sheep erythrocytes) on incubation of negatively charged liposomes with serum. This includes phosphatidylserine-liposomes (Comis and Easterbrook-Smith, 1986) and 18 phosphatidylglycerol-liposomes (Chonn et al., 1989). The association of complement factors in a specific manner with anionic liposomes may also contribute to the rapid removal of these liposomes from the circulation. This is suggested by the complement dependent adhesion of negatively charged liposomes to platelets to form platelet-liposome microaggregates. Such aggregates of significant size would be expected to be cleared rapidly from the circulation by MPS (for further details see below and in Chapter 5). Various other serum proteins have been shown to bind to liposomes (reviewed by Bonte and Juliano, 1986). For example, anionic phospholipids bind vitamin K dependent serum proteins via C a + + ions and enhance the conversion of the zymogen forms of the clotting proteins to active proteases (reviewed by Jackson, 1980). This' association may exert an important influence on the removal- of negatively: charged hposomeS- ;from i ;the ; ; circulation. Other proteins such as the high molecular weight glycoprotein fibronectin, which is involved in many cell surface phenomena (cell-cell adhesion, phagocytosis and haemostasis), binds to vesicles containing phosphatidic acid, phosphatidylglycerol or phosphatidylethanolamine. Human C reactive protein which is present during acute inflammation processes and which can activate complement, has also been shown to associate with phosphatidylcholine liposomes. Proteins such as albumin, a2 macroglobulin and immunoglobulin also bind to liposomes. With the exception of the monocytic cells of the MPS (see Section 1.5.3), the interactions of liposomes of various lipid compositions with blood cell types has received little attention. When liposomes are introduced via the intravenous route in vivo, red cells, leukocytes and platelets are exposed to liposomes before they reach the MPS system. Few studies have indicated that these components may affect the in vivo biodistribution of liposomes. Recently, it has been shown that platelets interact with negatively charged liposomes. For example, intravenous administration of phosphatidylglycerol (PG) 19 c o n t a i n i n g l i p o s o m e s to rats has b e e n s h o w n to resu l t i n a r a p i d t r ans ien t t h r o m b o c y t o p e n i a ( R e i n i s h et a l . , 1989). T h i s t rans ien t r e d u c t i o n i n p la te le t c o u n t is assoc ia ted w i t h seques t ra t i on o f p la te le ts a n d l i p o s o m e s i n the l i v e r a n d l u n g ( D o e r s c h u k et a l . , 1989) . T h e p o t e n t i a l r o le o f p la te le ts i n the r e m o v a l o f l i p o s o m e s f r o m the c i r c u l a t i o n was sugges ted b y a t rans ien t l i p o s o m e - p l a t e l e t i n t e r a c t i o n o b s e r v e d in vitro. T h e na tu re o f th i s i n t e r a c t i o n a n d the i n v o l v e m e n t o f v a r i o u s s e r u m p r o t e i n s i n th is p rocess is i n v e s t i g a t e d i n C h a p t e r 5 o f th is thes is . 1.5.3 Interactions of liposomes with MPS in vitro T h e m e c h a n i s m w h e r e b y l i p o s o m e s are r e c o g n i z e d a n d r e m o v e d f r o m the c i r c u l a t i o n b y the m o n o n u c l e a r p h a g o c y t i c s y s t e m is l a r g e l y u n e x p l o r e d . T h e in vivo d i f f e r e n t i a l ra te o f c l e a r a n c e o b s e r v e d f o r n e u t r a l , n e g a t i v e a n d p o s i t i v e l i p o s o m e p r e p a r a t i o n s e n c o u r a g e d a n u m b e r o f g roups to i n v e s t i g a t e the e f f e c t o f s u r f a c e c h a r g e o n l i p o s o m e i n t e r a c t i o n s w i t h v a r i o u s ce l l s o f the M P S in vitro. H o w e v e r , as i .v a d m i n i s t r a t i o n o f l i p o s o m e s resu l ts i n p r i o r e x p o s u r e o f l i p o s o m e s to p l a s m a as w e l l as c i r c u l a t i n g c e l l s , a n d l i p o s o m e s i r r e s p e c t i v e o f c h a r g e assume a n o v e r a l l n e g a t i v e s u r f a c e p o t e n t i a l o n i n c u b a t i o n w i t h p l a s m a ( B l a c k a n d G r e g o r i a d i s , 1976) , the r e l e v a n c e o f these in vitro s tud ies to the in vivo m e c h a n i s m o f l i p o s o m e c l e a r a n c e is q u e s t i o n a b l e . It has b e e n w e l l e s t a b l i s h e d that p l a s m a cons t i t uen ts c a n coa t p a r t i c l e s i n t r o d u c e d i n t o the c i r c u l a t i o n thus e n h a n c i n g t h e i r u p t a k e b y p h a g o c y t i c ce l l s (a p rocess t e r m e d o p s o n i z a t i o n ; S a b a , 1970) . T h i s o b s e r v a t i o n -suggests tha t l i p o s o m a l l y b o u n d s e r u m p r o t e i n s a re i m p o r t a n t i n the i n t e r a c t i o n o f l i p o s o m e s w i t h M P S ce l l s . In th i s r e g a r d , i t has b e e n c l e a r l y s h o w n that the l e v e l o f l i p o s o m e s assoc ia ted w i t h m o n o c y t e s is s i g n i f i c a n t l y e n h a n c e d i n the p resence o f s e r u m o r b y p r i o r c o a t i n g w i t h a g g r e g a t e d I g G ( F i n k e l s t e i n et a l . , 1981) . 20 1.5.4 Passive targeting of liposomes The natural tendency of i.v. administered liposomes to accumulate within mononuclear phagocytes has been extensively exploited in a number of therapeutic regimes. For example, liposome encapsulated drugs such as antimonial drugs (Alving et al., 1978; New et al., 1978) and antibiotics (such as amphotericin, Lopez-Berestein and Juliano, 1987) have successfully been employed to eliminate intracellular parasites that reside within the phagocytic cells of the MPS. Delivery of immunomodulators by liposomes to target cells such as macrophages has resulted in enhanced killing of tumor cells in vitro and in vivo (Fidler et al., 1980). Other applications include the administration of liposomally encapsulated proteins (for example enzynles or hormones) or specific genes, in an effort to alleviate diseases arising-due to genetic deficiencies (Gregoriadis et al., 1971; Wong et al., 1980; Nicolau et al., 1984). Several researchers have shown that liposome encapsulated molecules can preferentially accumulate at sites of inflammation (Williams et al., 1986), infection (Morgan et al., 1981) and in some solid tumors (Ogihara et al., 1986). This passive targeting of liposomes to disease sites is clearly an effective method for drug delivery as slow release of drugs at defined sites will result in minimal systemic toxic side effects, thus contributing to a more efficacious approach to chemotherapy. In summary, the passive targeting of liposomes to MPS and blood monocytes and the capacity of liposome to infiltrate inflammation or diseased sites in the body, offers significant potential for site-specific delivery of biologically active agents in the treatment of a number of diseases. With an increased understanding of factors that influence these interactions and their functional consequences and by exploiting liposomes which exhibit long circulation half-lives, it may also be plausible to develop protocols for specific targeting of liposomes to other tissue sites in vivo for diagnostic as well as therapeutic applications. 21 1.6 S u m m a r y The preceding review clearly shows that liposomes represent a flexible carrier system for aqueous and lipophilic molecules. Furthermore, the natural fate of liposomes in vivo can be exploited in passive targeting protocols. The work presented in this thesis is aimed at developing the use of liposomes in applications requiring protein-liposome conjugates. Such systems have very broad potential applications. The initial chapters of this thesis focus on characterizing methods for the efficient attachment of proteins to liposomes in which the type of liposomally attached protein can easily be varied, thus facilitating the rapid preparation of a spectrum of protein-liposome conjugates for diverse applications (Chapter 2 and 3). These studies also show that the attachment of protein to liposomes can alter the monodisperse nature of the original vesicle preparation. For this reason, a method was developed for the generation of protein-liposome conjugates with a better defined size distribution (Chapter 4). This technique facilitated an investigation into vesicle aggregation and it is shown that such factors can contribute to the rapid clearance of protein coupled vesicles from the circulation. Finally, in Chapter 5, it is shown that negatively charged liposomes interact with platelets, resulting in microaggregates. These could play a role in the transient thrombocytopenia observed when such liposomes are injected intravenously. 22 2. DEVELOPMENT OF A GENERAL METHOD FOR COUPLING PROTEINS TO LIPOSOMES. 2.1 Introduction More than 80 years ago Paul Ehrlich envisioned the use of "bodies which possess a particular affinity for a certain organ . . . , as a carrier by which to bring therapeutically active agents to the organ in question (Ehrlich, 1906)." During the last decade, there has been considerable progress in the application of this concept to the elimination of cells that are reactive with antibodies coupled to toxic agents (Vitteta et al., 1983). As liposomes represent an efficient system for encapsulating cytotoxic materials, the technology developed in the preparation of such drug or toxin-antibody conjugates has been applied to prepare protein-liposome carrier systems. Initial attempts to generate targeted vesicles relied on the non-specific association of IgG with liposomes (Weissmann et al., 1974; Gregoriadis and Neerunjun, 1975; Huang and Kennel, 1979). These procedures however result in denaturation of protein^ are limited by variations in stability and the extent of liposomal coupled protein observed for different immunoglobulins, and favour non-specific interactions with cells via Fc receptors (Machy and Leserman, 1987). For these reasons, chemical methods for the attachment of targeting molecules to liposomes have been developed (summarized in Chapter 1, Table 1.1). The introduction of a hydrophobic group such as a fatty acid or lipid molecule into proteins, has been reported by a number of workers (Sinha and Karush, 1979; Huang et al., 1980; Janson et al., 1980). The incorporation of such hydrophobic protein derivatives into liposomes has been achieved by various methods including detergent dialysis or sonication of lipid mixtures containing the protein derivative. A major advantage of this approach is that the lipophilicity of the derivative results in efficient incorporation of protein into liposomes (85-90%, Huang and Kennel, 1980). However, 23 the requirement of harsh methods such as sonication for the preparation of such protein-liposome conjugates results in variable levels of inactivation of the protein (Huang et al., 1979), leaky vesicles due to the presence of detergent (Machy and Leserman, 1987) and limits the methods available for the encapsulation of materials in liposomes (Heath et al., 1986). The attachment of proteins to pre-formed liposomes has several advantages over the incorporation of hydrophobic protein derivatives into liposomes. In particular, methods for the production of sized liposome preparations (Hope et al., 1985) and procedures developed for the efficient encapsulation of solutes (Mayer et al., 1985 )^ can be utilized. Efficient coupling protocols based on the reactive N-hydroxysuccinimide esters of carboxyl acids have been developed such that the reactive portion of the lipid is exclusively reactive with a particular group on the protein molecule (Figure 2.1 and Figure 2.2; Carlsson et al., 1978; Leserman et al., 1980; Martin et al., 1981; Martin et al., 1982; Duncan et al., 1983). Carboxyl groups activated in this manner react very efficiently with amino groups at neutral pH to produce an amide bond. This type of coupling approach (which involves the modification of phosphatidylethanolamine (PE) with prosthetic groups as originally described by Uemura and Kinsky (1972)) results in high protein liposome coupling ratios while minimizing inefficient side reactions such as homopolymerization of protein or liposomes. An optimized coupling protocol is a procedure that facilitates the rapid and efficient preparation of a diversity of protein-liposome conjugates such that the coupled protein remains associated with liposomes in a stable and functional manner. Studies in this direction have lead to the development of protein coupled vesicles which can subsequently bind various targeting proteins of interest. Presently this approach is limited to the attachment of immunoglobulins to liposomes (Leserman et al., 1980; Matthay et al., 1986). The potential use of sandwich techniques based on the high affinity of biotinated proteins for avidin conjugated to vesicles (Urdal et al., 1980; 24 Figure 2.1 Modification of proteins with amine reactive N-hydroxysuccinimidyl ester derivatives. Reduced with DTT Biotinated Protein PDP-Protein 25 F i g u r e 2.2 Phosphatidylethanolamine derivatives of SPDP, SMPB and NHS-biotin. O l i R , _ C - 0 - C H 2 R - C _ 0 - C H O O II • II II / / O H C - 0 - p - 0 - C H 2 - C H 2 - N - C - ( C H 2 ) - S - S ( ' H ' H 2 2 O N Pyridyl-dithio-propionyl PE (PDP-PE) R , - C - 0 - C H 2 O R 2 - C - 0 - C H O H £ H C - 0 - P - 0 - C H 2 - C H 2 - N - C - ( C H 2 ) 3 H 1 - " Maleimidophenyl-butyryl PE (MPB-PE) O R , - C - 0 - C H 2  R 2 - C - 0 - C H O O H C - 0 - P - 0 - C H 2 - C H 2 - N - C - ( C H 2 ) H 6~ H Biotin PE ? f r y C - ( C H 2 ) 4 - 1 ± N H 26 Godfry et al., 1981) has recently been described. Significant background binding (Finn et al.,1980) which has been attributed to the basic nature of avidin, has however precluded the general use of this system in the generation of targeted liposome systems. For this reason streptavidin, a protein produced and exported by the organism Streptomyces avidinii (Chaiet et al., 1963), which has a similar high affinity for biotin but is neutral at physiological pH (pi of 6.5), has been suggested to represent a superior protein (Finn et al., 1981). In this chapter, methods which utilize the streptavidin-biotin sandwich system are examined with regard to developing a general, rapid and efficient procedure for ^conjugation of proteins to pre-formed liposomes. Two approaches are investigated. First, a non-covalent protocol which involves the cross-linking of biotinated proteins to biotin EPE containing liposomes via streptavidin (Figure 2.3(a)) and. .second, .binding..Of: biotinated proteins to streptavidin covalently coupled to liposomes containing thiol reactive lipid derivatives (Figure 2.3(b)). It is shown that the two procedures result in significant levels of streptavidin coupled to liposomes. The covalent coupling methods based on the thiol reactive lipid derivatives PDP-EPE (Leserman et al., 1980, Martin et al., 1981) and MPB-EPE (Martin et al., 1982) are shown to be more efficient methods of attaching protein to liposomes. Moreover, the covalent conjugation procedure based on the maleimide lipid MPB-EPE, is shown to represent a more flexible and efficient v coupling method than the procedure based on PDP-EPE. The usefulness of both non-covalently and covalently liposomally conjugated forms of streptavidin in the preparation of protein conjugated vesicles is demonstrated by the rapid association of biotinated proteins with streptavidin conjugated liposomes. Such antibody conjugated systems are shown # to bind to defined cell types as determined by the liposomally associated antibody. 27 Figure 2.3 Non-covalent (a) and covalent methods (b) of attaching streptavidin to liposomes. (a) + \ Liposome containing Biotin PE \ (STREPTAVIDIN) Streptavidin-Biotin PE Liposome Conjugate - C - Biotin 28 Figure 2.3 (continued) (b) ,itl<"*'fl. o * i N O ' Liposome containing MPB-PE S|-N-C-(CH2)-S-S V ;?? H 2 2 N Liposome containing PDP-PE N H 2 2HN{STREPTAVIDIN)0 ( C H 2 ) - S - H Reduced PDP-Streptavidin „ l l ) » J U , , V s . „ tr-. o II ||-N-C-(CH2)3 <NH2 2HN-(STREPTAVIDIN)0 (CH2) I o s i II I ;i-N-C-(CH2)-S s H 2 Streptavidin-MPB-PE Liposome Conjugate Streptavidin-PDP-PE Liposome Conjugate 29 2.2 Materials and methods 2.2.1 Materials Egg Phosphatidylcholine (EPC) and egg-phosphatidylethanolamine (EPE) were obtained from Avanti Polar Lipids. Biotin-phosphatidylethanolamine (biotin EPE), N-succinimidyl 3-(2-pyridyldithio) propionate (SPDP), N-succinimidyl 4-(p-maleimidophenyl) butyrate (SMPB), were obtained from Molecular Probes. Cholesterol, inulin, tyramine, biotin, streptavidin, Sephadex G-50, FITC-cellite, Triton-X-100, pepsin, N-(2-hydroxyethyl) piperazine-N'-3-propanesulphonic acid (EPPS), N-2-Hydroxyethyl-piperazine-N'-2-ethanesulphonic acid (HEPES) and 2-(N-Morpholino) ethanesulphonic acid (MES) were obtained from Sigma. Carrier free Na 1 2 5I in 0.1 M NaOH, 1 4 C cholesterol, 3 H N -hydroxysuccinimide-biotin (NHS-biotin) and S H biotin were obtained from Amersham. 3 H dipalmitoyl-phosphatidylcholine ( 3H DPPC) was obtained from Nuclear England Nuclear. Anti-human and anti-rat erythrocyte IgG were obtained from Cappel, Iodobeads and N-hydroxysuccinimide-biotin (NHS-biotin) were obtained from Pierce and Sepharose CL-4B was obtained from Pharmacia. 1 2 5 I tyramine inulin was a generous gift from Dr.E.Sommerman (Sommerman et al., 1984). All other reagents were of standard chemical grade. 2.2.2 Synthesis of N-(4-(p-Maleimidophenyl)butyryl) egg phosphatidylethanolamine (MPB-EPE) and N-(3-(2-pyridyl-dithio)propionyl) egg phosphatidylethanolamine (PDP-EPE) « The synthesis of MPB-EPE was carried out according to the procedure of Martin et al. (1982; see structure in Figure 2.2). EPE (50 pmole) was dissolved in 5 ml of freshly distilled anhydrous methanol containing 50 /jmoles of freshly distilled 30 triethylamine and 100 mg SMPB. The reaction was carried out at room temperature under nitrogen and its progress was monitored using thin layer chromatography (TLC, running solvent: chloroform/ methanol/ water, 65: 25: 4). Following an 18 hour incubation, 95% of the EPE was converted to a faster running product. Methanol was removed under reduced pressure and the sample was redissolved in chloroform. The mixture was extensively washed with 1% NaCl to remove unreacted SMPB and residual triethylamine. Analysis of the product by two dimensional TLC (1. Base: chloroform/ methanol/ 25% N H 3 / H 2 0 , 90: 54: 5.7: 5.3; 2. Acid: chloroform/ methanol/ acetic acid/ H2O, 60: 30: 8: 2.85) indicated the presence of two components (Rf values in acid dimension: 0.93 and 0.783) and confirmed by *H NMR analysis (L.Choi, unpublished observations). Both products were shown to be ninhydrin negative, phosphate positive and ran as a single component in the solvent system used in monitoringthe-progress-of — the reaction (running solvent chloroform/ methanol/ water, 65: 25: 4). The synthesis of PDP-EPE was performed as described by Leserman et al. (1984; see structure in Figure 2.2). Briefly, EPE (50 /tmole) was dissolved in 3.5 ml of chloroform/ methanol (9:1) and added to 1.5 ml methanol containing SPDP (60 /*mole) and triethylamine (100 /xmole). After a 4 hour incubation at room temperature, TLC (solvent chloroform/ methanol/ water, 65: 25: 4) analysis indicated 99% conversion of EPE to a faster running product. This reaction mixture was washed with 10 ml of phosphate buffer saline prior to removal of the organic phase under reduced pressure. Further analysis by two dimensional TLC (acid and base solvent systems described above) and *H NMR indicated a single product which was greater than 98% pure. 2.2.3 Preparation of liposomes Large unilamellar liposomes (LUVs) were prepared as described by Hope et al. (1985). Appropriate aliquots of lipid mixtures in chloroform were deposited in a tube 31 and dried to a lipid film under a stream of nitrogen followed by high vacuum for two hours. The lipid was then hydrated in various buffers and extruded through two stacked filters of various pore size. Phospholipid was estimated by the colorimetric method of Fiske and Subbarrow (1925). Briefly, samples containing between 0.05 and 0.2 /xmole phospholipid were digested in 0.6 ml of 70% HCIO4 at 190°C for 1 to 2 hours. Subsequently, 7 ml of ammonium molybdate reagent (0.22%, w/v, ammonium molybdate in 2% H2SO4, v/v) and 0.6 ml of Fiske-Subbarrow reagent (30 g NaHS0 4, 1 g Na 2S0 4 and 0.5 g l-amino-2-napthol-4-sulphonic acid in 200 ml water heated to 40°C). The reaction mixture was incubated at 100°C for 15 minutes, cooled to room temperature and the absorbance at 815 nm was determined. In some experiments, trace amounts of 1 4 C cholesterol were incorporated in the lipid mixture and samples were assayed by scintillation counting in a Packard Tri Carb liquid scintillation analyzer. For targeting experiments, 1 2 5I labelled-inulin was included in the hydration buffer. After extrusion of the lipid sample, unencapsulated inulin was removed by column chromatography on Sepharose CL-4B equilibrated with HEPES buffered saline (25 mM HEPES, 150 mM NaCl pH 7.5; HBS). 1 2 5I was measured using a Packard Auto-Gamma 5650 gamma counter. 2.2.4 Preparation of proteins for coupling (i) SPDP modification of proteins: Streptavidin (5-10 mg/ml in HBS) was modified with the amine reactive reagent, SPDP according to published procedures (Carlsson et al., 1978). Briefly, SPDP (25 mM, methanol stock) was incubated at a 10 molar ratio to streptavidin at room temperature for 30 minutes in HBS. To estimate the extent of modification, a portion of the reaction mixture was passed down Sephadex G-50 columns equilibrated with HBS to remove 32 unreacted SPDP. The extent of modification of streptavidin was determined by estimating the protein concentration at 280 nm (molar extinction coefficient, E28O : 1.8 x 105) prior to the addition of dithiothreitol (DTT) and the 2-thiopyridone concentration at 343 nm (E343 : 7550) 10 minutes after the addition of DTT (25 mM). The remainder of the reaction mixture was reduced with DTT (25 mM, 10 minutes) and the thiolated product was isolated by gel exclusion on Sephadex G-50 equilibrated with various buffers. The product was immediately used in coupling experiments. In the case of IgG (5-10 mg/ml in HBS), following the modification of the protein with SPDP, the protein was fluorescently labelled with FITC-cellite as described below. Prior to the treatment of the protein with DTT, the sample was separated from unreacted reagents on Sephadex G-50 equilibrated with an acetate buffer (100 mM NaCl, 100 mM Na acetate, pH 5.0), to protect against the reduction of the intrinsic disulphides of the protein. The extent of modification of streptavidin was 5-6 SPDP molecules per protein while the modification of the antibody preparation resulted in 2-3 molecules of SPDP per protein. (ii) Biotination of antibodies: Antibodies were derivatized with biotin according to the procedure of Bayer et al. (1976). NHS-biotin in 50 pi of dimethylformamide (DMF) was incubated at a 2-10 molar ratio to antibody (10 mg in 0.95 ml), in 1 M NaCl, 0.1 M NaHCC«3, pH 8.0, for 4 hours at room temperature. The sample was extensively dialysed against HBS, overnight at 4°C. The extent of biotination of antibody was estimated from a standard curve employing trace amounts of S H NHS-biotin in the above reaction and assuming an extinction coefficient, E28O of 2.02 x 105 for antibody. Biotinated F(ab')2 fragments were prepared by digestion of biotinated antibody with pepsin in 0.1 M Na acetate, pH 4.5, at 37°C overnight (Nisoff et al., 1960). The products were fractionated on a Sephadex G -33 150 column and fractions containing F(ab'>2 fragments, as determined by a 10 % SDS polyacrylamide gel (Laemmli, 1970), were pooled. (iii) Fluorescent labelling of proteins: Proteins (5-10 mg/ml in HBS) were fluorescently labelled with FITC-cellite (Rinderknecht, 1962). Briefly, FITC-cellite (50% weight of protein in 1 ml of 150 mM NaCl, 0.2 M NaHCC>3, pH 8.8) was incubated with protein for 20 minutes at room temperature. The fluorescein labelled product was isolated after centrifugation of the mixture in a microcentrifuge by gel exclusion on Sephadex G-50 equilibrated with HBS. The concentration of protein (mg/ml), was determined by correcting for the absorbance contributed by the fluorescein label at 280 nm using the following equation: (OD280 -0.35xOD49s)/OD280 for 1 mg/ml solution of protein. FITC was assayed by monitoring the fluorescence at 520 nm using a SLM-aminco SPF-500C spectrofluorometer with an excitation wavelength of 495 nm. (iv) Iodination of streptavidin: Streptavidin was iodinated using iodobeads (Markwell, 1982). Briefly, iodobeads were washed twice with phosphate buffered saline (PBS) and blotted dry on filter paper prior to iodination reactions. Streptavidin (100 ug in 0.5 ml PBS) was added to carrier free Na 1 2 5 I (1 mCi) in a microcentrifuge tube and the iodination reaction was initiated by the addition of one iodobead. After 15 minutes incubation at room temperature, the iodinated streptavidin was removed from the tube and unreacted radio-iodide was removed from the product by gel filtration on Sephadex G-50 equilibrated with PBS. Greater than 90% of the gel excluded material was precipitable by the addition of cold trichloroacetic acid (final concentration of 20%, w/v). 3 4 2.2.5 Coupling of proteins to liposomes (i) Non-covalent procedure Liposomes (various molar ratios of EPC: cholesterol, 1 pmole total lipid, lOOnm vesicles) containing biotin EPE (0.15-0.05 mole %) were normally incubated with streptavidin (1 pCi 1 2 6I /mg, 4 mg/ml in 20 mM EPPS, 150 mM NaCl pH 8.0; EPPS pH 8.0) at room temperature at a 10 fold excess of streptavidin to biotin EPE. Samples were fractionated on Sepharose CL-4B columns to separate liposomally bound streptavidin from free streptavidin. The ratio of streptavidin bound per pmole lipid was determined by counting 1 2 5 I and a standard phosphate assay for phospholipid (Section 2.2.3). (ii) Covalent protocol Liposomes of various lipid compositions containing 5 mole % of the thiol reactive lipids MPB-EPE or PDP-EPE, were incubated with reduced PDP-proteins (100 pg protein / pmole total lipid, 1 mM final lipid concentration) in EPPS pH 8.0, overnight. Uncoupled protein was removed by gel chromatography on Sepharose CL-4B equilibrated with HBS. The extent of coupling of streptavidin to liposomes was measured by the binding of 3 H biotin to liposome associated streptavidin. Briefly, streptavidin-liposomes (0.25 /xmoles lipid in 0.5 ml) were incubated with 3 H biotin (3.85 nmoles in 25 pi, 15.4 nmoles//iCi) for 10 minutes and unbound 3 H biotin was removed by gel filtration on Sepharose CL-4B equilibrated with HBS. The extent of binding of 3 H biotin to a sjreptavidin standard (100 pg; OD280 for 1 mg/ml solution of streptavidin : 2.77) after gel exclusion on Sephadex G-50, was used as a reference for the calculation of coupling ratios. For the determination of the extent of antibody coupled to liposomes, 35 samples (200 pi) were dissolved in ethanol (1.8 ml) and the liposome associated fluorescence was correlated to a known quantity of fluorescein labelled antibody. (iii) Binding of biotinated antibody or FCab')^  fragments to streptavidin-liposomes Antibody (or F(ab')2) streptavidin-liposomes were prepared by incubating biotinated, fluorescein labelled antibody (or F(ab')2; 1 mg/ml, 1-4 biotin molecules / antibody) with streptavidin-liposomes (1 /imole/ml) for 30 minutes at room temperature, at a 2-8 fold mole ratio of antibody (or F(ab')2) to streptavidin. Uncoupled antibody or F(ab')2 was removed by gel exclusion on Sepharose CL-4B equilibrated with HBS. The extent of coupling was determined by measuring the amount of liposomally associated fluorescence as described in Section 2.2.4 (iii). 2.2.6 Binding of targeted liposomes to rat erythrocytes 1 2 6 l-inulin liposomes (25 A»mole EPC, 25 nmole biotin EPE; 0.027 pCi 1 2 5I-inulin / pmole lipid) were prepared as described in Section 2.2.3. For erythrocyte cell binding studies, rat or human erythrocytes were washed with EPPS pH 8.0, three times. Streptavidin-liposomes and antibody-liposomes prepared as in Section 2.2.5(i) and (iii) respectively were incubated with 109 erythrocytes (in some cases after prior labelling of with 50 pg of antibody) for 1 hour at 4°C. Cells were subsequently washed three times with EPPS pH 8.0, and counted for 1 2 5I to determine levels of erythrocyte-associated liposomes. 36 2.3 Results 2.3.1 Non-covalent method of coupling proteins to liposomes The first series of experiments explored the use of streptavidin in cross-linking biotinated antibodies to liposomes containing biotinated phosphatidylethanolamine (biotin EPE, Figure 2.2 and 2.3(a)). Initially, the binding of streptavidin to liposomes containing biotin EPE was characterized. This binding is a crucial step, as the four biotin binding sites on streptavidin can give rise to inter-vesicle coupling, with associated aggregation and precipitation. The major aims of these experiments were therefore first to show that streptavidin couples to liposomes containing biotin EPE and second to determine the maximum liposomal concentration of biotin EPE for which streptavidin induced aggregation can be avoided. Biotin EPE was therefore incorporated into EPC liposomes at a molar ratio of 0.1% (with respect to EPC) and incubated in the presence of streptavidin (10 fold molar excess with respect to biotin EPE). As shown in Figure 2.4, binding proceeded rapidly to achieve streptavidin-biotin EPE ratios of 1:12 (mole/mole). This association was found to be independent of pH over the pH range 5-9 (data not shown), and all subsequent coupling of streptavidin to liposomes was performed for 30 minutes at pH 8.0. In the preceding experiment, a maximum of 5.8 pg of streptavidin was bound per pmole lipid, which corresponded to about 7-8 copies of streptavidin per vesicle for a vesicle diameter of 0.1 pm. In order to increase the amount of streptavidin bound per liposome (and thus increase the number of sites available for biotinated antibody to couple, to liposomes), the amount of biotin EPE in the vesicles was increased over the range 0-0.5 mole % of total lipid, maintaining a constant ratio of streptavidin to total lipid. As shown in Figure 2.5 when the amount of biotin EPE was increased, a linear increase in the amount of streptavidin bound per pmole lipid was observed, both in the 37 Figure 2.4 Coupling of streptavidin to liposomes containing biotin phosphatidylethanolamine. Liposomes (99.9 mole % EPC, 0.1 mole % biotin EPE) were prepared as described in Section 2.2.3. Streptavidin (1 /iCi 1 2 5I/mg, 4 mg/ml) was incubated at room temperature with vesicles at a 10-fold excess to biotin EPE in EPPS pH 8.0. At various time points, aliquots were fractionated on Sepharose CL-4B columns (5 ml) to separate liposomally bound streptavidin from free streptavidin. The ratio of streptavidin (ug) bound per umole lipid was determined as in Section 2.2.5(i). Non-detectable levels of streptavidin were associated with control liposomes (100 mole % EPC, data not shown). S o I •i—i -t-> 4) U CQ 3. Time (mins.) 38 presence and absence of cholesterol. In the presence of 50 mole % cholesterol, a 50% increase in the streptavidin bound per /xmole total lipid was observed compared to EPC liposomes. If the amount of biotin EPE was increased further than 0.35% (in the presence of cholesterol) and 0.5% (in the absence of cholesterol) with respect to total lipid, complete aggregation and precipitation of liposomes was observed on addition of streptavidin. Thus the maximum number of copies of streptavidin per 100 nm vesicle that can be achieved without aggregation is approximately 40, both in the presence and absence of cholesterol. The next step in the coupling procedure involved the attachment of biotinated antibody to streptavidin coated liposomes. A preliminary study was performed to determine whether an antibody (anti-rat erythrocyte IgG) which had four biotin molecules covalently bound, would associated with pre-formed streptavidin-liposomes. Approximately equimolar amounts of antibody were bound per liposomal streptavidin irrespective of the amount of antibody available (1-8 antibody molecules per streptavidin; Figure 2.6(a)). Furthermore, equimolar amounts of antibody bound to streptavidin-liposomes irrespective of the amount of antibody biotination (1-6 biotins per antibody; Figure 2.6(b)). In the absence of liposomally bound streptavidin, no detectable levels of antibody bound to liposomes (data not shown). Thus, since streptavidin-liposomes can be prepared with 40 copies of streptavidin bound per vesicle, the maximum number of copies of antibody bound per vesicle is 40. In order to demonstrate that these targeted liposomes associated specifically with target cells, biotinated anti-rat erythrocyte IgG or derived biotinated F(ab')2 fragments were coupled to vesicles containing 125I-labelled tyramine inulin. Approximately 7 copies of IgG( or F(ab')2 were coupled per 100 nm vesicle. As shown in Table 2.1, little non-specific binding of biotinated liposomes to rat or human erythrocytes was observed. Anti-rat erythrocyte IgG or F(ab')2 liposome complexes bound specifically to rat erythrocytes. Such specific binding was effectively blocked by the presence of 1000 39 Figure 2.5 Influence of cholesterol on streptavidin coupling to liposomes containing variable amounts of biotinated phosphatidylethanolamine. Vesicles (25 /zmole EPC, 25-125 nmole biotin EPE (O), or 12.5 /xmole cholesterol, 12.5 umole EPC, 25-125 nmole biotin EPE (9) were prepared as in Figure 2.4. Streptavidin (1 /xCi,125I/mg, 0.68 mg) was incubated with liposomes (1 umole lipid) for 30 minutes at room temperature in EPPS pH 8.0. The ratio of streptavidin bound per /xmole lipid was determined as in Section 2.2.5(i). 35 0.00 0.10 0.20 0.30 0.40 0.50 % biotin EPE 40 Figure 2.6 Effect on coupling of (a) incubation ratio of biotinated antibody to streptavidin-liposomes and (b) biotination levels of antibody. Streptavidin-liposomes (6.3 /xg streptavidin / /zmole lipid) were prepared as outlined in Figure 2.4. Fluorescein labelled, biotinated antibody preparations were incubated with streptavidin-liposomes for 10 minutes and the extent of coupling was determined after column chromatography on Sepharose CL-4B by measuring liposomally associated fluorescence. Negligible binding of biotinated antibody to EPC liposomes or native antibody to EPC liposomes containing 0.1 mole % biotin PE was detected. 2 3 4 5 8 7 Molar ratio of Antibody to Streptavidin No. of Blotlna per Antibody 41 Table 2.1 Binding of targeted biotin EPE liposomes to rat erythrocytes. Sample No. of Liposomes bound per erythrocyte Rat Human Liposomes 18 N.D. Streptavidin-liposomes 20 N.D. (a) Pre-incubation with IgG + streptavidin-liposomes 542 N.D. IgG streptavidin-liposomes 416 11 IgG streptavidin-liposomes + biotin 73 N.D. F(ab')2 streptavidin-liposomes 302 11 F(ab')2 streptavidin-liposomes + biotin 50 N.D. 1 2 5I-inulin liposomes (25 /xmole EPC, 25 nmole biotin EPE, 0.027 /iCI 125I-inulin//imole lipid) were prepared as described in Section 2.2.3. Streptavidin-liposomes were prepared as described in Figure 2.4. Anti-rat erythrocyte antibody (or F(ab')2) streptavidin-liposomes were prepared as in Section 2.2.5(iii). For binding studies, lipid (0.62 /imole/ml) was incubated with 109 erythrocytes in each experiment with the exception of (a), where the lipid concentration was 1.76 /imole/ml. Cells were washed three times with EPPS pH 8.0 and were counted to determine levels of erythrocyteTassociated liposomes. (N.D., not done) 42 molar excess of biotin to streptavidin. Pre-incubation of rat erythrocytes with biotinated antibody, followed by the addition of streptavidin-liposomes also resulted in the binding of liposomes to rat erythrocytes. For these targeted systems, a maximum of 20% of total lipid were bound to rat erythrocytes, which is comparable to binding efficiencies of other targeted vesicle systems (Machy et al., 1982; Bragman et al., 1983; Heath et al., 1983). 2.3.2 Covalent coupling of proteins to liposomes Efficient methods for the covalent attachment of proteins to liposomes have exploited the reactivity of the succinimidyl derivatives SMPB (Martin et al., 1982) and SPDP (Leserman et al., 1980; Martin et al., 1981),.with.amino groups (Figure.2.1, Figure. 2.2 and Figure 2.3(b)). The derivatization of egg-phosphatidylethanolamine (EPE) with both reagents was readily achieved under alkali conditions in organic solvents (Figure 2.2). Both lipid derivatives when incorporated into liposomes at 5 mole % (based on total modified lipid) bind thiolated proteins with an optimum pH of 8.0 (PDP-EPE derivative, Martin et al., 1982; MPB-EPE derivative, Dr.M.B.Bally, unpublished observations). In this Section, to determine which coupling lipid represented the more versatile coupling approach, a comparison of factors affecting the extent of thiolated protein binding to liposomes, was examined. In Figure 2.7, the effect of cholesterol on the reactivity of derivatized lipids (5 mole % of total lipid) incorporated into liposomes with thiolated IgG (5 SPDP molecules per IgG)» is shown. The results indicated that significant levels of IgG coupled to liposomes containing PDP-EPE did not occur unless greater than 20 mole % cholesterol was incorporated into EPC liposomes. In contrast, levels of 12 pg IgG/^mole lipid were obtained for vesicles containing MPB-EPE, even in the absence of cholesterol. Similar coupling efficiencies have been obtained for smaller proteins such as protein A (MW: 41,000; Dr.M.B.Bally, unpublished observations), 43 Figure 2.7 Coupling of anti-rat erythrocyte IgG to EPC liposomes containing thiol reactive lipids (5 mole %), as a function of cholesterol content (0-45 mole % of total lipid). Thiolated IgG (4 SPDP groups/IgG, 100 /jg) was fluorescently labelled and incubated with (O) PDP-EPE or (#) MPB-EPE liposomes (1 /miole lipid, 1 mM final lipid concentration, labelled with 3 H DPPC) overnight, in EPPS pH 8.0 buffer. After removal of uncoupled IgG by gel chromatography on Sepharose CL-4B, the amount of IgG bound per /jmole lipid was determined as described in Section 2.2.5(H). 35 0 0 5 10 15 20 25 30 35 40 45 Cholesterol ( mol. %) 44 indicating that this differential reactivity of the two lipid derivatives is independent of the thiolated protein employed. The lower levels of antibody coupled to vesicles containing PDP-EPE versus MPB-EPE, under the examined conditions, indicate that the coupling lipid MPB-EPE was the more efficient cross linker. For this reason, all subsequent experiments were carried out with this lipid product. In an attempt to develop a general coupling procedure for attaching various proteins of interest to liposomes, the coupling of thiolated streptavidin to MPB-EPE liposomes of various lipid compositions, was examined. As shown in Table 2.2, efficient coupling of streptavidin to liposomes was observed for a variety of lipid compositions (30-40% of initial protein bound to liposomes). Notably, the incorporation of 45 mole % cholesterol in EPC liposomes did not result in increased levels of liposomally bound streptavidin, as observed for IgG (see above). Also, the incorporation of a "negatively charged lipid such as phosphatidylserine or unsaturated forms of phosphatidylcholine, did not affect the levels of streptavidin conjugated to vesicles. The effect of vesicle size on the coupling reaction was also examined. As shown in Figure 2.8, similar levels of streptavidin were bound to liposomes extruded through filters of 50-200 nm in pore size. A significant reduction in coupling levels was observed for liposomes prepared by extrusion through filters of 400 nm or larger in pore size, which is likely due to the multilamellar character of these liposomes. This would result in a decrease in the amount of coupling lipid available to couple with thiolated streptavidin. In Figure 2.9, the retention of the affinity of streptavidin covalently coupled to liposomes for biotinated proteins is examined. Incubation of biotinated antibody (4 biotin molecules per antibody) with streptavidin-liposome conjugates resulted in the binding of 40 /ig IgG bound per nmole lipid. This corresponds to the association of about one biotinated antibody per two streptavidin molecules and a coupling efficiency of 40%. The extent of binding was observed to be relatively independent of IgG biotination over 45 Table 2.2 Effect of lipid composition on coupling. Lipid Composition Hg Streptavidin / umole Lipid DOPC/CHOL 38 DOPC 36 EPC/CHOL 42 EPC 36 EPC/CHOL/BPS 38 EPC/BPS 32 Streptavidin was incubated with MPB-EPE liposomes (EPC or dioleoyl phosphatidylcholine (DOPC), CHOL, bovine brain phosphatidylserine (BPS) and MPB-EPE at molar ratios of 55, 45, 5 and 5 % respectively; liposomes contained trace amounts of 1 4 C cholesterol) as described for antibody in Figure 2.7. The extent of coupling of streptavidin to liposomes was determined by binding of 3 H biotin to conjugated vesicles as described in Section 2.2.5(ii). 46 Figure 2.8 Effect of vesicle size on the extent of streptavidin coupled to liposomes. Liposomes (5 mole % MPB-EPE, 45 mole % cholesterol, 50 mole % EPC) of various sizes were prepared by extrusion through filters of different pore size (50-600 nm). Coupling of streptavidin and the extent of streptavidin conjugated to liposomes was performed as described in Table 2.2. Vesicle diameter as defined by filter pore size (nm) 47 the range of 2-4 biotin molecules/IgG, as was observed previously (data not shown). Levels of biotinated antibody bound to liposomes in the absence of conjugated streptavidin were not detectable (data not shown). 2.4 Discussion In this chapter, covalent and non-covalent methods of attaching proteins to liposomes are examined with particular emphasis placed on the development of a general, flexible and efficient coupling procedure. Methods based on a sandwich system which exploits the high affinity interaction of streptavidin with biotin are described, which should facilitate the attachment of a variety of proteins of interest to a single type of protein conjugated liposome system. Here it is shown that streptavidin can associate in a non-covalent manner with biotin EPE liposomes or covalently in a modified form with liposomes containing thiol reactive lipid derivatives. Both forms of conjugated streptavidin vesicles readily bind biotinated protein molecules. It is further demonstrated that streptavidin-liposome conjugates bind to cells in a specific manner either by prior labelling of cells with defined biotinated antibodies or by attachment of biotinated antibodies to streptavidin conjugated liposomes before cellular binding studies. Initially, a non-covalent method of attaching streptavidin to liposomes containing biotin EPE was investigated. Streptavidin-liposome conjugates with up to 35 copies of protein bound per liposome (assuming a vesicle size of 0.1 um), can readily be achieved in this manner. However, in order to avoid aggregation of biotin EPE liposomes by the tetrameric biotin binding protein streptavidin, high incubation levels of streptavidin to biotin EPE were necessary, which results in inefficient protein to lipid coupling ratios (3% of initial protein). Although this represents a disadvantage with regard to direct targeting protocols, the rapid interaction of streptavidin with biotinated components may be exploited in indirect targeting applications. Furthermore, the aggregation properties of 48 Figure 2.9 Coupling of biotinated anti-rat erythrocyte IgG to streptavidin-liposome conjugates. Streptavidin-liposomes were prepared as described in Table 2.2 with approximately 40 ng streptavidin bound per /rniole lipid. Biotinated antibody (4 biotins/IgG) was fluorescently labelled and incubated for 30 minutes with streptavidin-liposomes at a two fold molar ratio of antibody to streptavidin. The sample was fractionated on Sepharose CL-4B and the extent of antibody (•) coupled to liposomes ( O ) was determined on the excluded fractions as described in Section 2.2.5 (iii). Fraction No. 49 this system offer a sensitive method for the development of the diagnostic side of streptavidin-liposome con j ugates. In an attempt to develop efficient methods of attaching streptavidin to liposomes which avoid excessive aggregation and precipitation effects, covalent coupling methods based on the two thiol reactive lipid derivatives MPB-EPE and PDP-EPE were addressed. When incorporated into liposomal systems these lipids have been employed to couple proteins which contain endogenous thiol groups (Martin et al., 1982) or have had thiol groups introduced by modification with the amine reactive hetero-bifunctional reagent SPDP (Barbet et al., 1981; Bragman et al., 1983, Leserman et al., 1984; Bragman et al. 1984). Here it is demonstrated that SMPB derivatized PE functions as a more efficient and versatile coupling reagent than PDP-EPE, which may be due to the longer spacer arm associated with this lipid derivative. Coupling of thiolated IgG to liposomes containing either lipid derivative is shown to be more efficient in the presence of cholesterol. It is feasible that this effect occurs due to steric constraints associated specifically with larger immunoglobulins at the liposomal surface as coupling of small protein molecules such as streptavidin to MPB-EPE liposomes is not influenced by addition of cholesterol. Similarly, with regard to the non-covalent method of preparing streptavidin-liposome conjugates, the addition of cholesterol to liposomes containing biotin EPE may alleviate the steric hindrance associated with a short spacer arm. In comparison to other coupling procedures, the covalent attachment of streptavidin to MPB-EPE liposomes represents an efficient coupling procedure (coupling efficiencies of 30%). As indicated here, the coupling of biotinated IgGs to streptavidin-liposomes can be up to 20-40% more efficient when compared to covalent coupling procedures. It is thus feasible that streptavidin conjugated vesicles may be utilized in an efficient manner for the preparation of various protein conjugated liposomes by prior biotination of the protein molecule of interest. 50 Recently streptavidin conjugated liposomes have been prepared by utilizing carbodiimide derivatives (Rosenberg et al. 1987). The exclusive reactivity of thiolated streptavidin with MPB-EPE liposomes as described here has a particular advantage over the above covalent coupling approach as protein-protein crosslinking associated with the use of carbodiimide reagents cannot occur. In summary, it is shown that streptavidin-liposome conjugates can readily be prepared by non-covalent and covalent coupling procedures. Furthermore, it is demonstrated that such conjugates rapidly associate with isolated biotinated antibodies or cells pre-labelled with biotinated antibodies, resulting in targeted vesicle systems. This technology should be readily extended to a large variety of antibodies and other ligands of interest in direct as well as indirect sandwich targeting protocols and thus satisfies many of the requirements of a general protocol of generating targeted liposomal systems. With regard to the methodology employing the SMPB derivatized EPE as a coupling reagent, although this derivative was readily synthesized, two reaction products were generated during the derivatization. This was not noted in previous work employing this lipid. Two dimensional TLC and 1 H NMR studies indicated that pure MPB-EPE comprised 60% of the product. As the presence of two components is not ideal, attempts were made to identify the contaminant and develop a better method for the synthesis of pure MPB-EPE. The results of these studies are discussed in Chapter 3. 51 3. OPTIMIZED PROCEDURES FOR COUPLING PROTEINS TO LIPOSOMES 3.1 Introduction In the previous chapter, general methods for coupling biotinated proteins to vesicles via conjugation with streptavidin were described. This approach utilizes the tight association of biotin to streptavidin and has a number of potential advantages. For example, it is known that biotination of IgGs does not significantly influence binding to antigens (Bayer et al., 1979; Heitzman and Richards, 1974). In addition, coupling of biotinated proteins to streptavidin requires only gentle incubation procedures as opposed to exposure to relatively harsh reducing conditions during covalent coupling protocols. As indicated in Section 2.4, covalent coupling of proteins such as streptavidin to liposomes via the maleimide lipid derivative MPB-PE, is a more efficient method than that based on PDP-PE or a non-covalent approach which employs biotin PE. Further characterization of this lipid derivative however, indicated that previous protocols for synthesizing MPB-PE are subject to contamination by a ring-opened by-product, and a new procedure for the synthesis of pure MPB-PE was thus developed (Choi,L., to be published). The studies presented in this chapter are developed in two stages. First, optimized procedures for the covalent coupling of streptavidin to liposomes via a pure form of MPB-DPPE are determined. Subsequently, it is shown that these streptavidin coated liposomes rapidly and efficiently bind biotinated proteins which leads to liposome conjugates that exhibit specific targeting properties in vitro. 3.2 Materials and Methods 3.2.1 Materials Dipalmitoyl phosphatidylethanolamine (DPPE) was obtained from Avanti Polar Lipids. 52 /?-mercaptoethanol, ethylene diamine tetra-acetate (EDTA), dithiobis-2-nitro-benzoic acid (DTNB), N-ethylmaleimide (NEM), bovine serum albumin (BSA), carboxyfluorescein (further purified as described by Ralston et al., 1981), biotinated -protein A, biotinated-alkaline phosphatase and biotinated-succinylated concanavalin A were obtained from Sigma. Biotinated anti-B 1 (PAN-B, IgG 2a) and biotinated anti-T 11 (E-rosette, IgG 1) were obtained from Coulter Electronics. 2-amino-2-hydroxy-methyl propane-1,3-diol (TRIS) was obtained from Biorad and Ficoll Paque was obtained from Pharmacia. 3.2.2 Synthesis of N-[4-(p-Maleimidophenyl)-butyrylJ-dipalmitoylphosphatidyl-ethanolamine (MPB-DPPE) Pure MPB-DPPE used through out this thesis was kindly prepared and analyzed by Dr. Louis S. Choi. Briefly, MPB-DPPE was synthesized by reacting DPPE (69 mg) with SMPB (65 mg) in chloroform (5 ml) containing triethylamine (10 mg) at 40°C. After two hours, TLC on silica showed conversion of DPPE to a faster running product (solvent system: chloroform/ methanol/ acetonitrile/ water, 75: 16: 5: 4; Rf : 0.6). The solution was diluted with chloroform (10 ml) and washed several times with NaCl (0.9 %) to remove by-products of the reaction. The solution was further concentrated in vacuo and the solid residue was triturated and recrystalized from diethylether to remove unreacted SMPB. Further recrystalization from diethylether/ acetonitrile yielded a pure product as indicated by *H NMR analysis (Bruker W40, 400 MHz) and low resolution mass spectroscopy (KRATOS MS 50). Mass spectra were obtained with AEI MS9 instrumentation at the B.C. Regional Mass Spectroscopy Center. 53 3.2.3 Preparation of liposomes Large unilamellar vesicles (LUVs) were prepared as described by Hope et al. (1985). Briefly, appropriate amounts of lipid mixtures dissolved in chloroform were deposited in a tube and dried to a lipid film under a stream of nitrogen followed by high vacuum for two hours. Normally lipid samples (50-54 mole % EPC, 45 mole % cholesterol, 1-5 mole % thiol reactive lipid) were hydrated in 150 mM NaCl, 25 mM HEPES, 25 mM MES, pH 6.5 and extruded 10 times through 2 stacked 100 nm filters. Just prior to coupling experiments, samples were titrated to the appropriate pH with NaOH. For studies on the thiol dependence of the coupling procedure, liposomes containing 1 mole % (for coupling) or 5 mole % (for maleimide reactivity) pure MPB-DPPE were prepared at pH 6.5 as described above, titrated to pH 7.5 with NaOH and an aliquot was incubated with /9-mercaptoethanol for 5 minutes at a molar ratio of 10 moles ^-mercaptoethanol / mole of maleimide containing lipid. Liposomes were separated from free j8-mercaptoethanol on Sephadex G-50 equilibrated with 25 mM HEPES, 25 mM MES, 150 mM NaCl, pH 7.5. The coupling efficiency and the reactivity of the maleimide group of quenched liposomes was compared to that of unquenched samples. Lipid was estimated either by the colorimetric method of Fiske and Subbarrow (1925) or by trace amounts of 1 4 C cholesterol present in the lipid mixture. This was performed by scintillation counting in a Packard Tri Carb liquid.scintillation analyzer. 3.2.4 Assay for maleimide reactivity Reactivity of the maleimide group of MPB-DPPE lipid was estimated by the thiol binding of /?-mercaptoethanol to the lipid derivative and back titration of unbound /3-mercaptoethanol with Ellman's reagent, dithiobis-2-nitro-benzoic acid (DTNB) as described by Sedlack et al. (1968). Liposomes (5 mole % MPB-DPPE, 50 mole % EPC, 54 45 mole % CHOL, 1 u mole lipid in 200 ul) were incubated with /?-mercaptoethanol (100 j i l of 1 mM) at pH 8.2 (0.2 M Tris CI, 20 mM EDTA, 1 % Triton-X-100 (w/v), pH 8.2; 1.6 ml) for 30 minutes at room temperature. DTNB (100 ul, 20 mM in methanol) was added and the absorbance was measured at 412 nm after 30 minutes. 3.2.5 Preparation of proteins for coupling Streptavidin (5 mg/ml in 25 mM HEPES, 150 mM NaCl, pH 7.5; HBS), was modified with the amine reactive reagent, SPDP according to published procedures (Carlsson et al., 1978; see Section 2.2.4(i)). Biotinated anti-erythrocyte IgG (4 biotins bound per antibody molecule) was prepared according to the method of Bayer et al. (1976; see Section 2.2.4(H)). 3.2.6 Coupling of proteins to liposomes (i) Coupling of streptavidin to liposomes The coupling of streptavidin to liposomes was performed by incubating reduced PDP-modified streptavidin with liposomes containing pure MPB-DPPE at a ratio of 100 ng protein / pmole lipid (1 mM final lipid concentration) at various pH values. Unassociated protein was removed by gel filtration on Sepharose CL-4B equilibrated with HBS. The extent of coupling of streptavidin to liposomes was assayed by monitoring the binding of 3 H biotin to streptavidin conjugated liposomes containing trace amounts of 1 4 C cholesterol (see Section 2.2.5(H)). 55 ( i i ) B i n d i n g o f b i o t i n a t e d p ro te i ns to s t r e p t a v i d i n - l i p s o m e s All biotinated proteins were fluorescently labelled with FITC-cellite as described for IgG (see Section 2.2.4(iii)). Proteins were incubated at a two fold molar ratio to streptavidin coupled to liposomes for 10 minutes. Unassociated protein was removed by gel chromatography on Sepharose CL-4B pre-equilibrated with HBS. The extent of liposome associated protein was determined as described for fluorescently labelled IgG (described in Section 2.2.4(iii)). Background binding of all biotinated proteins was shown to be negligible (data not shown). 3.2.7 In vitro targeting of streptavidin-liposome conjugates Liposomes with entrapped carboxyfluorescein (15 mM) were coupled to thiolated streptavidin as described above at pH 7.5 and a final lipid concentration of 2.5 mM. The coupling reaction was quenched with N-ethylmaleimide (500 molar ratio to streptavidin) after 4 hours, streptavidin-liposome conjugates were isolated by gel exclusion on Sepharose CL-4B and levels of liposomally associated streptavidin were determined as described previously. For targeting experiments, human blood was collected in EDTA (25 mM in PBS). Human peripheral blood leukocytes were isolated by standard protocols using Ficoll Paque (Boyum, 1968) and suspended in PBS containing 2% BSA (w/v) and 0.01% Na azide (w/v) at 4°C prior to binding studies. Cells (106) were aliquoted into round bottom microtitre wells, washed and incubated with antibody (Til and B l , 5 and 10 pg respectively in 100 pi PBS) or alone in PBS for 1 hour at 4°C. After washing twice with PBS, cells were incubated with streptavidin-liposome conjugates (0.2 /zmoles in 200 pi PBS) for a further hour at 4°C. The cells were then washed three times with PBS and analyzed by flow cytometry. 56 3.2.8 Flow cytometry Cell associated fluorescence was measured with an EPICS Profile Analyzer (Coulter Electronics, Inc.). Cells were illuminated with the 488 nm line of an argon ion laser. Fluorescence was measured behind a 515 to 530 nm band-pass filter. Fluorescence signals were gated on the basis of a right angle versus forward light scatter cytogram to restrict analysis to signals from single cells. Amplifiers were set in the log area mode. For statistical analysis of histograms, region 1 was arbitrarily set with the lower channel at the base of the right shoulder of the histogram of the control sample (min.: 2.705, max.: 1023). 3.3 Results 3.3.1 Characterization of MPB-DPPE The derivatization of phosphatidylethanolamine (EPE or DPPE) with SMPB is a straightforward reaction that readily generates a lipid product which runs as a single component under certain solvent conditions as analyzed by thin layer chromatography on silica (Martin et al., 1982). However, further characterization of the reaction products by two dimensional thin layer chromatography (see Section 2.2.2) and by proton NMR indicated that this product contained a significant impurity which could account for up to 40 % of the material. This contaminant was identified as a ring open form of MPB-DPPE which is generated by methanolic cleavage of the maleimide moiety (L. Choi, unpublished observations; Figure 3.1). A new method for the preparation of a pure MPB-DPPE lipid derivative was thus developed which avoids the requirement of methanol during the synthesis and isolation of MPB-DPPE (L.Choi, to be published). The purity of this lipid product was shown by *H NMR analysis. For example, the J H 57 Figure 3.1 Structure of two MPB-PE lipid derivatives fl, , // w D P P E - N H - C - ( C H 2 ) 3 - N Structure of pure MPB-DPPE fl • . // \ N D P P E - N H - C - ( C H 2 ) 3 o 3^TT O Structure of ring open form of MPB-DPPE 58 NMR spectra of the final product exhibited low field resonances attributed to the aromatic protons of the phenyl groups (chemical schift 6: 7.3) and vinyl protons (6: 6.86) of SMPB with loss of the signal attributed to N-hydroxysuccinimide group (5: 2.86). Mass spectroscopy studies confirmed the purity of the lipid derivative by the presence of a molecular ion at 955 which corresponds to a molecular formula of CsiH^OnPNa for the Na salt of MPB-DPPE. 3.3.2 Optimized coupling conditions Optimal conditions for coupling thiolated streptavidin to liposomes containing pure MPB-DPPE were investigated. The pH dependence of the binding of thiolated-streptavidin to MPB-DPPE liposomes and the stability of the maleimide function were initially established. As shown in Figure 3.2, the amount of liposomally conjugated protein increased rapidly at pH values greater than 7.0. However, incubation of liposomes containing MPB-DPPE at pH values of 7.0 and above resulted in a corresponding increase in the rate of degradation of the maleimide group of the derivatized lipid. At pH 7.5 after 18 hours incubation, significant levels of streptavidin were coupled to liposomes (45%) with acceptable loss of maleimide reactivity (65% remaining). For this reason, a pH of 7.5 was chosen for further optimization of the coupling reaction. In Figure 3.3. a time course relating streptavidin binding to liposomes and reactivity of the maleimide lipid is presented. The results indicate that optimal levels of streptavidin conjugated to liposomes (approx. 37 pg / pmole lipid) were obtained with minimal degradation of the maleimide group after an incubation period of 8 hours at pH 7.5 and at room temperature. The requirement for protein associated thiol groups in the coupling procedure is illustrated in Table 3.1. Prior exposure of MPB-DPPE liposomes to 0-mercaptoethanol 59 Figure 3.2 Effect of pH on the levels of streptavidin conjugated to liposomes containing pure MPB-DPPE and on the stability of the maleimide moiety. Liposomes (1 mole % or 5 mole % MPB-DPPE, 54 or 50 mole % EPC, 45 mole % cholesterol, trace amounts of ^ C cholesterol) were incubated with thiolated streptavidin (1 mole % MPB-DPPE liposomes) or alone (5 mole % MPB-DPPE liposomes for maleimide assay) at various pHs (pH: 5.5-9.0) overnight. The level of streptavidin conjugated to liposomes (# ) and the stability of the maleimide group ( O) were determined as detailed in Section 3.2.4 and 3.2.6. 60 Figure 3.3 Optimal coupling conditions for the conjugation of streptavidin to liposomes containing pure MPB-DPPE. Liposomes (1 or 5 mole % MPB-DPPE, 54 or 50 mole % EPC, 45 mole % cholesterol) were incubated with thiolated streptavidin (1% MPB-DPPE liposome) or alone (5% MPB-DPPE liposomes for maleimide assay) at pH 7.5. At various times streptavidin-liposome conjugates were separated from free streptavidin by gel chromatography on Sepharose CL-4B and the levels of protein conjugated to liposomes ( # ) and the stability of the maleimide group ( O ) were determined as detailed in Section 3.2.4 and 3.2.6. 9 I Xi i—t O 4 Xi (X (0 u -•-> CO tao a. 40--30 20-10-x>o—o 6 Time (hrs.) 8 10 10  --40 o I o >> -t-> -f-> o ei <a FH Oi --20 ^ S 6^  12 61 Table 3.1. Thiol dependence of the coupling of streptavidin to liposomes containing pure MPB-DPPE. Sample % Maleimide Reactivity ug Streptavidin / pmole Lipid 0 hrs. 8 hrs. 8 hrs. MPB-DPPE liposomes 100 73 36.0 /?-mercaptoethanol treated MPB-DPPE liposomes 11 0 2.5 MPB-DPPE liposomes + unthiolated streptavidin 100 77 0 Liposomes (1 or 5 mole % MPB-DPPE, 54-50 mole % EPC, 45 mole % cholesterol) were quenched with /7-mercaptoethanol (10 molar excess to MPB-DPPE) for 5 minutes at pH 7.5, exchanged on Sephadex G-50 equilibrated with HBS and incubated with streptavidin or alone as described in Figure 3.3. After 8 hours incubation, the extent of streptavidin conjugated to liposomes and the reactivity of the maleimide group was determined for control (unquenched MPB-DPPE liposomes or unthiolated streptavidin) and quenched samples (see Section 3.2.3, 3.2.4 and 3.2.5 for Methods). 62 resulted in a decrease in the extent of liposomally conjugated streptavidin when quenched samples were compared to control MPB-DPPE liposomes. This was paralleled by a decrease in the detectable reactivity of the maleimide group of the lipid derivative. Furthermore, native streptayidin did not associate with liposomes containing the maleimide lipid. 3.3.3 Targeting to lymphocyte subpopulations in vitro The object of this investigation was to establish a general, efficient method of attaching various ligands to liposomes. The results to this point demonstrate that streptavidin can be efficiently coupled to liposomes containing a pure form of the maleimide lipid derivative MPB-DPPE. To show the applicability of this system in attaching various types of targeting molecules of interest to liposomes, the binding of a variety of biotinated proteins to streptavidin-liposomes was examined. As shown in Table 3.2, on incubation of various biotinated proteins with streptavidin conjugated liposomes, approximately 2 protein molecules binds for every 3 molecules of streptavidin. The extent of binding of biotinated proteins to streptavidin coupled vesicles is independent of the size of the biotinated protein (MW: 42,000-150,000 D). Liposomes coated with streptavidin can obviously be used directly as targeted systems by attachment of a biotinated antibody (see Section 2.3.1). Alternatively, an indirect procedure is possible whereby biotinated antibodies are first associated with target cells and the streptavidin coated liposomes are introduced subsequently. In this chapter, targeting of streptavidin-liposomes employing this later approach is demonstrated utilizing flow cytometry. As shown in Figure 3.4, incubation of liposome-streptavidin conjugates (containing encapsulated carboxyfluorescein) with cells pre-labelled with a biotinated monoclonal antibody specific for peripheral B cells (BI), resulted in the fluorescent labelling of approximately 20 % of the total lymphocyte 63 Table 3.2 Binding of biotinated proteins to streptavidin-liposomes. Protein ug/umole Lipid nmole//imole Lipid Molar Ratio Protein : Streptavidin Anti-human Erythrocyte IgG (mw: 150 kD) 62.6 0.417 1 : 1.68 Alkaline Phosphatase (mw: 140 kD) 77.7 0.555 1 : 1.25 Protein A (mw: 43 kD) 20.3 0.482 1 : 1.46 Succinylated Concanavalin A (mw: 55 kD) 26.4 0.480 1 : 1.46 Streptavidin-liposomes with 45.2 /jg protein bound / /rniole lipid were prepared as described in Figure 3.3. Fluorescein labelled biotinated proteins were incubated with conjugated liposomes at a 2 fold molar excess to streptavidin for 10 minutes at pH 7.5. The extent of coupling of biotinated proteins to streptavidin-liposomes was determined after gel chromatography of samples on Sepharose CL-4B by measuring the levels of fluorescence associated with liposomes for protein and scintillation counting for lipid. 64 Figure 3.4 Targeting of streptavidin-liposome conjugates via biotinated monoclonal antibodies to human peripheral lymphocytes. Streptavidin-liposome conjugates (38.8 fig streptavidin per /xmole lipid) with entrapped carboxyfluorescein (15 mM) were prepared as described in Section 3.2.7. Lymphocytes (106) were incubated with biotinated antibodies (10 /xg, Bl ( B ); 5 ug, T l 1 ( C )) or in PBS (A ) for 1 hour at 4°C. After two washes, streptavidin-liposome conjugates (0.2 /xmoles lipid) were added, incubated for an hour at 4°C and cells were washed thrice with PBS. Samples were subsequently examined for cell associated fluorescence by flow cytometry (Section 3.2.8; LFL1, log of fluorescence). H1H H » COUNT LFL1 PERCENT MEAN SO I HFCV 1 2.705 1823. 78 1.4 4.159 2.7BB 45.8 KIN HAI COUNT LFL1 fERCENT MEAN SO I KPCV 1 2.785 1823. 1815 28.3 22.11 3.B9 2.57 I HFCV 1 2.785 1823. 4585 91.7 35.46 1.93 11.7 65 population (Fig. 3.4B). In comparison, similar studies with a biotinated anti T cell antibody (Til) resulted in the labelling of approximately 90% of lymphocytes (Fig. 3.4C). These results are consistent with the expected cell distribution of the antigens defined by BI (Stashenko et al., 1980) and by T i l (Howard et al., 1981). The specificity of these conjugates is indicated by the negligible background binding of streptavidin-liposome conjugates to lymphocytes in the absence of biotinated antibodies (Fig. 3.4A). 3.4 Discussion Covalent attachment of liposomes to proteins such as antibodies which are directed against cell surface antigens associated with transformed cells, may have therapeutic potential. However, at the present time, such targeted liposomal systems have mainly been used for in vitro applications such as diagnostic assays (Kung et al.,1985). In order to exploit the full potential of targeted carrier systems a versatile and reliable methodology for coupling is required. In this chapter, an approach to rapidly couple IgGs and biotinated proteins to streptavidin liposomal systems, prepared by an improved coupling procedure, is described. This technology should be applicable to a variety of other bioactive peptides. Previous literature methods for the synthesis of the maleimide lipid derivative MPB-PE lead to the generation of a contaminant, ring open form of the expected lipid product. A new protocol for the synthesis of a pure SMPB derivative of phosphatidylethanolamine was subsequently developed (L.Choi, to be published). In this Chapter, coupling conditions for the conjugation of protein to liposomes were optimized such that the integrity of the maleimide function of pure MPB-DPPE was retained in order to generate well characterized protein-liposome conjugates. Significant amounts of protein were associated with MPB-DPPE liposomes (equivalent to 30 ug streptavidin per pmole lipid) at a pH of 7.5 after 8 hours with minimal loss of maleimide function 66 (approximately 85-90 % reactivity remaining). This amount of liposomally associated streptavidin corresponds to about 75 copies of streptavidin bound per 100 nm vesicle. Coupling efficiencies of up to 50 % are readily achieved under the optimized conditions outlined in this study. Similar efficiencies have been attained only on incorporation of higher levels of MPB-EPE in liposomes (5 mole %; Bragman et al., 1984). The increased efficiency observed here likely reflects the pH of the conjugation reaction rather than the impurity of the derivatized lipid. For example, conjugation of thiolated protein to MPB-DPPE liposomes at pH 6.7 (conditions used by Bragman et al., 1984) results in a 4.5 fold reduction in the levels of liposomally conjugated protein when compared to levels obtained at the optimum pH of 7.5 (Figure 3.2). This efficient coupling of protein to liposomes containing low levels of MPB-PE' is of particular importance as higher concentrations of this anchor molecule (>2.5 mole %) dramatically affect liposome stability (Bredehorst et al., 1986). Finally, with regard to other covalent methods for the preparation of protein-liposome conjugates, equivalent coupling efficiencies have been reported in a limited number of cases (Barbet et al., 1981, Rosenberg et al., 1987). The covalent method of conjugating proteins to liposomes under optimized coupling conditions requires the presence of protein thiol groups. As illustrated here, in the absence of endogenous free sulphydryl groups, exogenous thiol groups are introduced by the modification of protein amino functions with the hetero-bifunctional reagent SPDP. The deprotection of the SPDP thiol moiety which requires exposure of the derivatized protein to a strong reducing agent such as DTT may result in the loss of the biological function of the molecule (Heath et al., 1983). In order to circumvent this problem, a sulphydryl-introducing reagent that does not require the use of such harsh reagents has been synthesized (Duncan et al., 1983). The alternative approach exploited here utilizes the high affinity of streptavidin-liposome conjugates for biotinated proteins. This sandwich method of preparing protein-liposome conjugates requires the 67 prior modification of the protein of interest with the amine reactive derivative N -hydroxysuccinimide biotin, a procedure that does not appear to significantly influence the function of proteins (Heitzman et al., 1974; Bayer et al., 1979). This modification does not need to be extensive, as little as a single biotin coupled to IgG is sufficient for the efficient coupling to streptavidin coated vesicles (refer to Section 2.3.1). As indicated here, the coupling of biotinated proteins to streptavidin conjugated liposomes is rapid (< 10 minutes) and can be 20 to 40 % more efficient than covalent coupling procedures (see Section 2.3.1 and 2.3.2). In summary, a method for the preparation of liposomes coated with streptavidin covalently bound to MPB-DPPE via an improved coupling procedure, is described. It is demonstrated that such protein-liposome conjugates rapidly and efficiently conjugate with biotinated proteins resulting in targeted vesicle systems. The potential application of these conjugates in targeting and diagnostic regimes is illustrated by the specific binding of such conjugates to lymphocyte subpopulations via defined biotinated monoclonal antibodies. 68 4. PROTEIN-LIPOSOME CONJUGATES WITH DEFINED SIZE DISTRIBUTIONS. 4.1 Introduction A method based on the coupling of liposomes containing a pure MPB-DPPE lipid derivative to thiolated streptavidin is described in Chapter 3 which is efficient and subsequently allows the straightforward conjugation of a wide variety of biotinated proteins to liposome systems. Previous reports indicate that the attachment of proteins to liposomes can affect the integrity of liposome preparations (Heath et al., 1980; Bredehorst et al., 1986). Work described in this Chapter focuses on the physical properties of protein-liposome conjugates as various applications envisioned for such conjugates in vitro and in vivo require physically well characterized systems. It is demonstrated, by two physical techniques, that the attachment of protein to liposomes by various coupling protocols results in vesicle aggregation. The amount of liposomally attached protein was shown to determine the extent of vesicle cross-linking. A novel method for the generation of protein-liposome conjugates with defined size distributions, obtained by the extrusion of protein-coupled vesicles through filters of defined pore size is described. This procedure provides a relatively gentle method of producing protein-liposome conjugates of stable size. No significant denaturation of the attached protein is observed as indicated by the binding of free biotin or biotinated antibodies to streptavidin-liposome conjugates. The ability of immuno-streptavidin-liposome conjugates to bind to target cells is shown. With regard to developing protein coupled liposomes for in vivo applications, the influence of the size of protein-liposome conjugates on blood clearance behavior was investigated in mice. The results indicate that the size of the conjugate is a major factor that determines the circulation behavior of liposome conjugates. 69 4.2 Materials and methods 4.2.1 Materials Anti-human erythrocyte IgG was obtained from Cappel. 1 4 C cholesterol and S H cholesterol-hexadecyl-ether were obtained from NEN. S H and 1 4 C biotin were obtained from Amersham. Mice (average weight of 21g) were obtained from Jackson laboratories. For other reagents, refer to Sections 2.2.1 and 3.2.1. 4.2.2 Preparation of liposomes Large unilamellar liposomes were prepared as described in Section 3.2.3. (Hope et al., 1985). Briefly, a dry lipid film was hydrated with 25 mM MES, 25 mM HEPES, 150 mM NaCl pH 6.5 and extruded through two stacked 100 nm'or 50 nm filters 10 times. Prior to coupling experiments, samples were titrated to pH 7.5 with NaOH. Lipid was estimated either by the colorimetric method of Fiske and Subbarrow (1925, Section 2.2.3) or by incorporating trace amounts of 1 4 C cholesterol or S H cholesterol-hexadecyl-ether in the lipid mixture. 4.2.3 Preparation of proteins for coupling Streptavidin and anti-human IgG (10 mg/ml and 20 mg/ml respectively in 25 mM HEPES, 150 mM NaCl, pH 7.5, HBS) were modified with the amine reactive reagent, SPDP according to published procedures (Carlsson et al., 1978; see Section 2.2.4(i) for details). IgG was fluorescently labelled with FITC-cellite as described in Section 2.2.4 (iii), after a 30 minutes incubation with SPDP. The extent of modification of streptavidin was 5-6 SPDP molecules per protein while the modification of the 70 antibody preparation resulted in 2-3 molecules of SPDP per protein as estimated by the release of 2-thiopyridone (see Section 2.2.4(i)). 4.2.4 Coupling of proteins to liposomes The coupling of proteins to liposomes was performed by incubating the reduced PDP-modified protein with liposomes (54 mole % EPC, 45 mole % cholesterol, 1 mole % MPB-DPPE (synthesized as described in Section 3.2.2)), sized through filters of 50 or 100 nm pore size, at a ratio of 100 pg protein / /rniole lipid (5 mM-30 mM final lipid concentration) at pH 7.5 as described in Section 2.2.5 (ii). The reaction was quenched at various times by the addition of N-ethylmaleimide (500 molar ratio to protein, methanol stock). For in vivo experiments, samples were further quenched with /3-mercaptoethanol (10 molar ratio with respect to N-ethylmaleimide) after a 2 hour incubation of the reaction mixture with N-ethylmaleimide. Uncoupled protein was removed by gel filtration on Sepharose CL-4B equilibrated with HBS. The extent of coupling of streptavidin or antibody to liposomes was measured by the binding of S H or 1 4 C biotin to streptavidin or the level of liposomal associated fluorescence for antibody. Lipid was estimated by monitoring 1 4 C cholesterol or 3 H cholesterol hexadecyl ether levels by scintillation counting (see Section 2.2.5(ii)). Prior to the non-covalent attachment of streptavidin to liposomes, streptavidin was fluorescently labelled with FITC-cellite as described above for IgG (Section 2.2.4 (iii)). Streptavidin (4.1 mg) was incubated for 40 minutes with liposomes (54.75 mole % EPC, 45 mole % cholesterol, 0.25 mole % biotin EPE) at a 10 molar excess to biotin EPE as described previously (Chapter 2, Section 2.2.5 (i)). The extent of coupled streptavidin was determined after gel filtration on Sepharose CL-4B as described above for IgG. 71 4.2.5 Preparation and characterization of extruded protein-liposome samples Protein-liposome conjugates (5 mM or 20 mM final lipid concentration) were extruded 10 times through two stacked millipore filters (50 or 100 nm). Lipid recovery was estimated by scintillation counting of an aliquot of the extruded sample. The size of the protein-coupled-vesicles before and after extrusion was estimated by freeze fracture techniques and by quasi-elastic light scattering (QELS). For freeze fracture, liposome preparations were mixed with glycerol (25% by volume) and frozen in liquid nitrogen. Samples were fractured and replicated employing a Balzer BAF 400 D apparatus and micrographs of replicas were obtained using a Phillips 400 electron microscope. QELS measurements were obtained using a Nicomp Model 270 submicfo particle sizer operating at 632.8 nm and 5 mW. All freeze fracture studies were kindly performed by Dr. K. Wong. 4.2.6 In vitro studies of streptavidin-liposome conjugates Anti-human IgG was biotinated (Bayer et al., 1976) and fluorescein labelled as described in Section 2.2.4 (ii) and (iii) respectively. Binding of biotinated IgG to streptavidin-liposome conjugates extruded through two stacked 100 nm filters was examined by incubating antibody in the presence or absence of biotin (5000 molar ratio to antibody) at a two fold molar ratio to streptavidin attached to liposomes (0.5 /mioles lipid, 45.1 p% streptavidin / pmole lipid, 179 nm in diameter) for 10 minutes. Samples (0.5 ml) were chromatographed on Sepharose CL-4B and fractions (0.5 ml) were assayed for fluorescein labelled antibody and S H cholesterol hexadecyl ether for lipid as described above (see Section 4.2.4). In vitro targeting of streptavidin-liposomes (aggregated or extruded preparations) to human red blood cells via biotinated anti-human IgG was examined after prior 72 labelling of erythrocytes with IgG. Antibody (1 nmole) was incubated with 108 erythrocytes for 1 hour at 4°C and cells were washed twice with HBS. Aggregated or extruded streptavidin-liposome samples (0.25 /rnioles lipid; 3 H cholesterol hexadecyl ether as lipid marker; 42.9 and 39.6 pg streptavidin bound / /rniole lipid respectively; 530 nm and 132 nm in diameter) were incubated with cells in the presence or absence of biotin (5000 molar ratio to antibody) for 1 hour at 4°C. Cells were washed three times with HBS and 100 u\ aliquots were quenched with peroxide (100 p\ in 800 pi methanol) and assayed for cell associated radioactivity in scintillation fluid (5 ml) in a Beckman model LS 3801 scintillation counter. 4.2.7 In vivo studies of liposome preparations For in vivo studies streptavidin-liposome conjugates were prepared at a final lipid concentration of 30 mM and an incubation period of 15 minutes, essentially as described in Section 2.2.5 (ii). Liposomal lipid was quantified employing the non-metabolizable, non-exchangeable lipid marker 3 H cholesterol-hexadecyl-ether (Huang, L., 1983; Stein et al., 1980, specific activity: 0.23 pCi / mg total lipid). For scintillation counting, 50-100 pA plasma was added to 5 ml Pico-Fluor 40 scintillation cocktail and samples were counted in a Beckman model LS 3801 scintillation counter. Unbound streptavidin was removed by gel chromatography on Sepharose CL-4B. A portion of the sample was extruded 10 times through two stacked 50 or 100 nm filters immediately prior to injection. As controls, liposomes containing MPB-DPPE (54 mole % EPC, 45 mole % cholesterol, 1 mole % MPB-DPPE) were prepared at pH 6.5 as described in Section 4.2.2. An aliquot of the lipid sample was titrated to pH 7.5 with NaOH, quenched with /J-mercaptoethanol (10 molar excess to MPB-DPPE) and free 0-mercaptoethanol was removed by gel filtration on Sephadex G-50 equilibrated with HBS. Unquenched MPB-DPPE liposomes were exchanged on Sephadex G-50 equilibrated with 73 HBS prior to in vivo experiments. Liposomes (55 mole % EPC, 45 mole % cholesterol) were prepared in HBS. For in vivo plasma lipid level determinations, mice (4-8 / time point) were injected with samples via the tail vein at a dose of 100 mg total lipid / kg. Blood was collected in EDTA treated microcontainers and plasma was prepared by centrifuging (200 x g) whole blood for 10 minutes in a clinical centrifuge. Total plasma volume per animal was taken to be 4.55% of mean body weight. Control blood samples containing known amounts of liposomes showed that only a minor fraction of the liposomal lipid was associated with the pelleted blood cells. The recovery of liposomes was similar if determined from whole blood or from plasma. The levels of streptavidin associated with liposomes in vivo was determined by the binding of 1 4 C biotin to a plasma sample isolated 1 and 4 hours post injection. 4.3 Results 4.3.1 Effect of coupling proteins to liposomes on vesicle size Initial efforts were centered on characterizing the influence of protein conjugation on vesicle size. As shown in Figure 4.1, an increase in the amount of protein bound to liposomes results in a significant increase in vesicle size as recorded by QELS (Figure 4.1(B)). The initial rapid coupling of streptavidin to vesicles correlates with a rapid increase in the size distribution of the preparation. In order to confirm this observation employing a different technique and to examine the morphology of the larger systems, aliquots of the same coupling system were examined by freeze fracture. The results presented in Figure 4.2, clearly show that the increase in size as measured by QELS is related to vesicle aggregation. However, after extended periods of incubation, a 74 Figure 4.1 Effect of coupling streptavidin to liposomes on vesicle size. Liposomes (54 mole % EPC, 45 mole % CHOL, 1 mole % MPB-DPPE, 5 mM final lipid concentration, 100 nm) were incubated with streptavidin (100 ng protein / /imole lipid) overtime at pH 7.5. At various time points, the reaction was quenched by addition of N-ethylmaleimide (500 molar ratio to protein) and free streptavidin was removed by gel filtration on Sepharose CL-4B. Extent of coupled streptavidin was determined by 3 H biotin binding (A,©) and vesicle size was estimated by QELS (B,Q). 0 2 4 6 8 10 12 Time (hr) 75 Figure 4.2 Freeze fracture of streptavidin-liposome preparations. Streptavidin-liposome samples quenched at 0.5 (A), 2 (B), 4 (C) and 18 (D) hours were prepared as described in Figure 4.1 and examined by freeze fracture. 76 significant number of large vesicles (> 200 nm) are observed, which arise presumably due to fusion events following aggregation. 4.3.2 Extrusion of protein-liposome conjugates An optimized coupling protocol should not affect the overall size distribution of the conjugated system. In an attempt to achieve small, homogeneously sized protein-liposome conjugates, the effects of extruding aggregated, conjugated vesicles through filters with 100 nm pore size were examined for liposomes with attached streptavidin (Figure 4.3) or antibody (Figure 4.4). The coupling reaction mixtures were quenched with N-ethylmaleimide at various times and the size of the coupled samples prior to and after extrusion was estimated by QELS (Figure 4.3(B)-and Figure 4.4(B)). The extent of coupled protein was determined after extrusion of conjugated samples (Figure 4.3(A) and Figure 4.4(A)). Irrespective of the amount of protein coupled to the liposomes, both types of coupled vesicles were readily extruded and the resulting preparations fell within a narrow size range. For example, extrusion of liposomes with attached streptavidin (25-60 pg I /umole lipid) resulted in vesicle sizes of 120-140 nm in diameter as compared to initial size distributions of 150 to more than 500 nm (Figure 4.3.(B)). Similarly, extrusion of antibody-liposome conjugates (15-35 pg protein / /unole lipid) resulted in smaller vesicles of narrow size distribution (90-110 nm) compared to a size range of 130-230 nm prior to extrusion (Figure 4.4(B)). Importantly, the loss of lipid for both protein coupled vesicles during the extrusion process was minimal (85-90% lipid recovery). These results demonstrate that highly aggregated preparations of vesicles with high levels of conjugated protein can be extruded efficiently and the resulting preparations are of a similar size. The observation that vesicle aggregation occurs on conjugation of protein to liposomes is not unique to the covalent procedure. Aggregation also occurs during the 77 Figure 4.3 Extrusion of streptavidin-liposome conjugates. Streptavidin was coupled to liposomes (100 nm) containing 1 mole % MPB-DPPE at a final lipid concentration of 20 mM. At various time points aliquots of the reaction mixtures were quenched with N -ethylmaleimide and diluted to 5 mM lipid concentration before extrusion through 100 nm filters. The extent of coupled streptavidin (A,#) was estimated by 3 H biotin binding to streptavidin-liposomes after gel chromatography of lipid samples on Sepharose CL-4B. The size of the liposome streptavidin conjugates was estimated by QELS before and after extrusion (B, O) as described in Section 4.2.5. 70 "3 50 + B x; 40 el | 30 Cti $ 204! u tic 0< 0 g 200 o 1B0 100'} 80 B ,o o o o 1E0 850 380 460 660 Size o! aggregate before extrusion (nm) 3 4 5 Time (hr) 8 78 Figure 4.4 Extrusion of antibody-liposome conjugates. Fluorescein labelled antibody was coupled to liposomes (100 nm) containing 1 mole % MPB-DPPE at a final lipid concentration of 20 mM. At various time points aliquots of the reaction mixtures were quenched with N-ethylmaleimide and diluted to 5 mM lipid concentration before extrusion through 100 nm filters. The extent of coupled antibody (A, #) was determined by estimating the levels of liposomally associated fluorescence after chromatography of lipid samples on Sepharose CL-4B. The size of the antibody-liposome conjugates was estimated by QELS before and after extrusion (B.O) as described in Section 4.2.5. 40 ft 3 30 + r H o S j. 30+/ O rO • i H 3. 0*-0 «o too 160 100 o M 60 B o o o g 0 i n 140 160 160 800 Size of aggregate before extrusion (nm) 3 4 5 Time (hr) 8 79 F i g u r e 4.5 Freeze fracture of streptavidin-liposomes before and after extrusion. Streptavidin was coupled to liposomes at a f ina l l i p i d concentration of 20 m M for 8 hours as described in Figure 4.1. The sample was di luted to 5 /xmoles / m l pr ior to extrusion. Non-covalent attachment of streptavidin to liposomes containing biot in E P E (0.25 mole %) was performed as described in Section 4.2.4 at a f i n a l l i p i d concentration of 5 m M . Samples were examined by freeze fracture before and after extrusion through 100 n m fi l ters. S t r e p t a v i d i n - M P B - D P P E liposomes before ( A ) and after (B) extrusion; s t reptavidin-biot in E P E liposomes before (C) and after (D) extrusion. 80 non-covalent attachment of streptavidin to liposomes containing biotin EPE (Chapter 2, Section 2.3.1). To demonstrate the general application of the extrusion process as a means of generating sized populations of protein-liposome conjugates, the effect of extrusion of streptavidin coupled covalently to liposomes containing MPB-DPPE or non-covalently bound to liposomes containing biotin EPE, was examined by freeze fracture (Figure 4.5). Both types of streptavidin-liposome conjugates were observed to be highly aggregated prior to extrusion. After extrusion the coupled vesicles existed as monomers or dimers with the maximum aggregate observed to be a conglomerate of 4 vesicles. In the case of the non-covalent coupling procedure (Figure 4.5(C) and 4.5(D)), significant loss of lipid occurred (50%) during the extrusion of coupled vesicles. 4.3.4 In vitro characterization of extruded streptavidin-liposome conjugates The stability of extruded samples containing covalently bound streptavidin with respect to size is addressed in Figure 4.6. QELS measurements indicate an initial small (30 nm) rapid increase in the size of the preparation after extrusion. This was reflected by increased aggregation of the extruded vesicles as indicated by freeze fracture (results not shown). As shown in Table 4.1, the level of reaggregation observed 3 hours after extrusion of various streptavidin-liposome conjugates was minimal when compared to the aggregated state of the samples prior to extrusion. Such reaggregation of liposomes was not observed when MPB-DPPE liposomes were extruded with thiolated-streptavidin which had been quenched by prior incubation with 0-mercaptoethanol (Table 4.1). This indicates that reaggregation was not due to non-specific association of protein with liposomes. It was found that the incorporation of negatively charged lipids such as phosphatidylserine, or the presence of low or high ionic strength buffers did not prevent reaggregation (data not shown). Reduction of the amount of streptavidin coupled to vesicles (Table 4.1) resulted in a corresponding decrease in the extent of reaggregation 81 Figure 4.6 Examination of the stability of extruded streptavidin-liposome conjugates by QELS. Streptavidin-liposomes with approximately 51 ug protein / /xmole lipid were prepared by incubating thiolated streptavidin with liposomes containing 1 mole % MPB-DPPE for 8 hours at a final lipid concentration of 20 mM. After removal of unbound streptavidin by gel filtration on Sepharose CL-4B, the sample was diluted to 5 mM lipid and extruded 10 times through two stacked 100 nm filters. At various time points the size of the extruded preparation was determined by QELS ( O )• Size of streptavidin-liposome conjugates prepared as in Figure 4.1 are shown for comparison (#). 4 5 0 -O 50 t 0 1 2 3 4 Time ( hrs. ) 82 Table 4.1 Factors affecting the aggregation of extruded streptavidin-liposomes. Sample QELS Size Estimates of Streptavidin-Characteristics Liposome Conjugates (nm) /ig Streptavidin Lipid Before After /nmol. Lipid Conc.(mM) Extrusion Extrusion 0 hrs 8 hrs 0C 2.5 110 104 104 17.1a 2.5 177 109 119 31.6a 2.5 232 119 140 45.3a 2.5 286 123 154 45.1* 5.0 403 174 197 45.1 . 15.0 403 174 197 45.lJ'd 5.0 403 174 182 45.1b'e 5.0 403 174 188 a Liposome samples (54 mole % EPC, 45 mole % CHOL, 1 mole % MPB-DPPE) were prepared with different levels of coupled streptavidin by quenching the coupling mixture (20 mM final lipid concentration) with N-ethylmaleimide at various time points, k Streptavidin-liposomes were prepared at a final lipid concentration of 30 mM and an incubation period of 15 minutes. c Streptavidin (50 ng) quenched with N-ethylmaleimide was extruded with liposomes (1 nmole, 2.5 mM final lipid concentration) containing 1 mole % MPB-DPPE. ^ Extruded samples were kept on ice for 3 hours prior to QELS measurements. e Extruded samples were frozen immediately after extrusion and thawed just prior to QELS measurements. 83 observed 3 hours after extrusion. Varying the lipid concentration of the extruded sample did not significantly affect reaggregation. Streptavidin coupled to liposomes which were frozen immediately after extrusion, maintained their original size distribution on thawing. Finally, storage of the extruded samples at 4°C resulted in increased stability of liposome size. The extrusion of protein-liposome aggregates represents a gentle method of preparing sized protein conjugated vesicles. This was illustrated by the retention of the ability of streptavidin-liposome conjugates to bind biotin (Figure 4.3) and biotinated proteins (Figure 4.7) after extrusion. The specificity of this interaction was shown by the absence of liposomally associated antibody in the presence of an excess of biotin (5000 molar ratio to antibody, Figure 4.7(b)). It is further shown that extruded conjugates when compared to aggregated streptavidin-liposomes bind as efficiently in a biotin dependent manner to target erythrocytes which were pre-labelled with biotinated anti-human erythrocyte IgG (Table 4.2). These observations indicate that the extrusion process minimally affects the potential use of streptavidin-liposome conjugates in targeting protocols. 4.3.5 In vivo properties of extruded streptavidin-liposome conjugates Large liposomes are rapidly removed from the blood circulation when compared to small preparations (Hunt, 1982; Sota et al., 1986). It is therefore possible that the rapid clearance observed for targeted systems in vivo (Wolff et al., 1984; Papahadjopoulos et al., 1988) could be partly due to aggregation of liposomes. The time required for clearance from the blood of certain control liposome preparations (Figure 4.8(A)) as well as aggregated and extruded streptavidin-liposome conjugates (Figure 4.8(B)) in mice were therefore examined. Aggregated streptavidin-liposomes (530 nm in diameter as indicated by QELS) were cleared rapidly from the circulation; only 3% of the initial 84 Figure 4.7 Binding of biotinated anti-human erythrocyte IgG to extruded streptavidin-liposome conjugates. Extruded streptavidin-liposome conjugates with 45.1 /xg streptavidin bound / /xmole lipid were prepared as described in Figure 4.3. Fluorescein labelled biotinated antibody was incubated at a 2 fold molar ratio to streptavidin in the absence (A) or presence (B) of biotin (5000 fold to antibody) for 10 minutes. Samples were subsequently chromatographed on Sepharose CL-4B equilibrated with HBS and fractions were assayed for protein by fluorescence measurements (#) and for lipid ( 0 , 3 H cholesterol acyl ether marker) by scintillation counting. B 0.12 A 0 2 4 6 B 10 12 14 18 IB 20 22 24 26 Fraction No. B -TO.12 0# # % 9 li) 0J CD O #-#=W0.00 0 2 4 6 8 10 12 14 16 IB 20 22 24 26 Fraction No. 85 Table 4.2 Targeting of aggregated and extruded streptavidin-liposomes to human red blood cells in vitro. Sample nmoles Lipid / 108 Erythrocytes Aggregated streptavidin-liposomes: Anti-human IgG + Streptavidin-liposomes 28.29 Anti-human IgG, Biotin + Streptavidin-liposomes 2.27 Extruded streptavidin-liposomes: Anti- human IgG + Streptavidin-liposomes 23.85 • Anti-human IgG, Biotin + Streptavidin-liposomes 2.01 Aggregated and extruded streptavidin-liposomes were prepared as in Figure 4.3. The extent of coupled streptavidin and the size of the conjugated systems were 42.9 pg streptavidin / nmole lipid and 530 nm in diameter for aggregated conjugates, and 39.6 pg streptavidin / umole lipid and 132 nm in diameter for extruded conjugates. Biotinated anti-human erythrocyte antibody (1 nmole) was incubated with human erythrocytes (108) for 1 hour at 4°C. After washing twice with HBS, streptavidin-liposomes (0.25 pmole lipid) were added to cells in the presence or absence of biotin (5000 molar excess to antibody), incubated for 1 hour at 4°C and cells were washed three .times with HBS. Cell associated radioactivity was determined by scintillation counting of samples (100 A»0 after quenching with peroxide (100 pi in 800 pi methanol) in 5 ml scintillation fluid. 86 lipid dose remained in the circulation 4 hours after injection. Extrusion of these protein-vesicle conjugates through 50 or 100 nm polycarbonate filters resulted in preparations with size distributions of 139 and 187 nm respectively. Both of these preparations showed extended blood circulation times in vivo, with 48 and 32 % of the initial dose remaining in circulation after 4 hours. When compared to EPC/ CHOL vesicles (125 nm), the presence of covalently bound protein on liposomes of similar size (139 nm), increased the rate of removal of liposomes from the circulation (80 and 48 % respectively remaining in circulation at 4 hours). No significant difference in the circulation of MPB-DPPE liposomes (normal or quenched with /9-mercaptoethanol, 170 nm in diameter) was observed when compared to EPC / cholesterol preparations of 197 nm in diameter. The stability of covalently conjugated streptavidin-liposomes in vivo was demonstrated by the binding of biotin to liposome samples isolated from plasma at 1 and 4 hours post injection (Table 4.3). A slight loss of biotin binding capacity of streptavidin coupled liposomes was observed for samples isolated from plasma, which could arise due to the absorption of serum components to the vesicles, the inactivation of streptavidin by proteolysis or the binding of endogenous biotin to the preparation. 4.4 Discussion Protein-liposome conjugates have many potential applications, ranging from systems of diagnostic ability (Ho et al., 1985) to systems specifically targeted to disease \ sites in vivo (Gregoriadis, 1982). A major focus of this thesis has been directed towards developing general procedures for the generation of protein-liposome conjugates which allow efficient and straightforward coupling of a variety of proteins to liposomes and which result in conjugated systems of defined size distributions and protein-lipid ratios. As indicated elsewhere (Chapter 2 and 3) the coupling of streptavidin to liposomes 87 Figure 4.8 In vivo clearance rates of liposome preparations. Streptavidin was coupled to liposomes (50 and 100 nm,) at a final lipid concentration of 30 mM and incubation period of 15 minutes, quenched with N-ethylmaleimide for 2 hours followed by an overnight incubation with /J-mercaptoethanol. Control liposomes containing MPB-DPPE were titrated to pH 7.5 and exchanged on Sephadex G-50 equilibrated with HBS. EPC / cholesterol liposomes were made up in HBS. Mice were injected with lipid at a dose of 100 mg / kg. Plasma was prepared from EDTA whole blood at specific time points and aliquots were analyzed by scintillation counting (Section 4.2.7). Size of extruded samples were determined by QELS. (A): EPC / cholesterol, 125 nm (#); EPC / cholesterol, 197 nm (•); MPB-DPPE liposomes, 170 nm ( quenched A , unquenched, T ) ; (B): aggregated 100 nm streptavidin-liposomes, 530 nm ( • ); streptavidin-liposomes extruded through 100 nm, 187 nm (A); streptavidin-liposomes extruded through 50 nm, 139 nm (O)-Time (hr) 88 Table 4.3 Stability of streptavidin-liposome conjugates in vivo. Streptavidin-Liposome conjugate (QELS size estimates) ng Streptavidin /umol. Lipid Prior to After administration Administration 1 hr. 4 hr. Aggregated 42.9+/-0.1 43.1+/-0.8 29.8+/-0.8 (>530 nm) Extruded 41.1+/-2.8 35.4+/-0.2 32.9+/-0.3 (187 nm) Extruded 47.1+/-0.5 44.5+/-1.4 39.1+/-0.6 (139 nm) • The amount of streptavidin attached to liposomes was determined by the binding of 1 4 C biotin to lipid samples or pooled plasma samples from three mice, 1 and 4 hours post injection (see Section 4.2.7). 89 results in a flexible basic system which subsequently allows the straightforward conjugation of a wide variety of biotinated proteins. This chapter focuses on the physical properties of protein-liposome conjugates. It is shown that such conjugates are susceptible to aggregation during the coupling procedure, which is particularly notable at high protein to lipid ratios. A major finding is that aggregated systems can be re-extruded through filters of defined pore size, resulting in smaller, homogeneously sized systems with enhanced size stability and corresponding decreased blood clearance rates in vivo. The observation that liposome protein conjugates tend to aggregate during the conjugation process (particularly at high protein to lipid ratios) is consistent with previous observations. For example, Bredehorst et al. (1986) have fourid that increased amounts of protein (Fab fragments) conjugated to liposomes resulted in an increase in the polydispersity of vesicle populations. It has also been observed that conditions which increase the coupling efficiency of protein to liposomes, such as lipid concentration and the ratio of protein to lipid in the coupling incubation step, affects the extent of vesicle-aggregation observed by negative staining (Heath et al., 1980). The increased degree of aggregation obtained for highly coupled liposome systems can be attributed to inter-vesicle cross-linking via the protein molecule. The presence of large liposome systems (> 200 nm in diameter) as observed by freeze fracture after extended coupling periods, also indicates that liposome fusion can occur subsequent to the aggregation step associated with the coupling of protein to liposomes. The demonstration that highly aggregated liposome-conjugates with various amounts of bound protein, can be readily extruded through filters of 50 or 100 nm in pore size yielding a population of conjugates with narrow size distribution is a primary finding of this investigation. The size of the re-extruded preparations as determined by QELS, were slightly larger than the initial liposome preparation, which is consistent with the presence of vesicle dimers as observed by freeze fracture. For the chemically coupled 90 liposome-protein conjugates, greater than 85% of lipid was recovered after extrusion of conjugates. The retention of the binding capacity of streptavidin-liposomes for biotin and biotinated antibody indicates that extrusion does not result in significant protein denaturation and loss of binding activity. This is further indicated by the biotin dependent binding of extruded streptavidin-liposome conjugates to human erythrocytes via biotinated anti-human erythrocyte IgG (Table 4.2). After extrusion of highly aggregated liposome preparations, reaggregation of vesicles occurs (Figure 4.6). As indicated by the QELS and freeze-fracture studies, this reaggregation is relatively minor (size increases of 20-30 nm over 3 hours) as compared to the aggregation before extrusion. This reassociation is insensitive to a variety of protocols. For example, stable liposome-coupled samples with different levels of attached protein (17-50 /xg / /xmole lipid) were prepared at lipid concentrations ranging from 5 mM to 30 mM (Table 4.1) with little influence on reaggregation. Minimal reaggregation was observed when extruded samples were stored at 4°C or frozen immediately after the extrusion process. It is possible that during the coupling process some vesicle aggregation corresponds to cross-linking of protein to the lipid MPB-DPPE on two or more vesicles. Extrusion of such an aggregate may result in smaller systems, with exposed protein associated MPB-DPPE removed from previously adjacent vesicles. Exposure of such hydrophobic groups would be expected to lead to aggregation. The rate of clearance of liposomes from the circulation is dependent on the size of the preparation; the larger the liposome the'faster it is removed (Hunt, 1982; Sota et al., 1986). In this regard, limited success has been achieved for the targeting of protein-liposoipe conjugates to specific sites in vivo, due in part to their rapid sequestration from the circulation by the reticuloendothelial system (Papahadjopoulos et al., 1988). As shown here, the extent of aggregation of the coupled-liposomes significantly alters the blood clearance behavior of the conjugated preparations. For example, aggregated 91 streptavidin-liposomes (> 530 nm in diameter) were rapidly removed from the circulation (< 3% remaining after 4 hours). In comparison, extended circulation times were obtained for extruded conjugates i.e. 32 and 48% of the initial lipid dose remained in circulation 4 hours post-injection for samples of 187 nm and 139 nm in diameter respectively. The enhanced circulation times observed for smaller protein-liposome conjugates indicates that aggregation of the preparation can be a major factor that determines the lifetimes of conjugates in vivo. It should be noted however, that the blood clearance of protein-liposome conjugates was always greater than for control samples of similar size, indicating that the presence of protein on liposomes contributes to some extent to an enhanced clearance of liposomes from the circulation. The presence of the thiol reactive coupling lipid MPB-DPPE in liposomes does not significantly affect the in vivo clearance when compared to EPC/cholesterol liposomes, suggesting that the binding of thiol containing serum proteins does not affect the in vivo properties of liposomes. In summary, in this chapter a technique for the generation of sized protein-liposome conjugates by extrusion through filters of defined pore size is presented. The procedure allows easy manipulation of the physical size of protein coupled liposomes without affecting the binding activity of the protein. Stable protein-vesicle conjugates of defined size distribution can readily be prepared with various amounts of protein attached to liposomes by this technique. The enhanced blood circulation times of extruded conjugates and the retention of the biotin binding capacity of extruded streptavidin-liposomes 4 hours post injection in mice, indicate that extruded preparations of streptavidin coupled liposomes will be capable of binding to biotinated molecules in vivo. 92 5. THE BINDING OF PHOSPHATIDYLGLYCEROL LIPOSOMES TO RAT PLATELETS IS MEDIATED BY COMPLEMENT. 5.1 Introduction The potential of targeting liposomes within the vasculature is a major application envisioned for protein-liposome conjugates in vivo (Poste, 1983). Various factors however, have been shown to influence the behavior of liposomes in the circulation. For example, as illustrated in Chapter 4 and in other studies, liposome (or protein-liposome conjugate) size affects the residence times of liposomes in the blood (Abra et al., 1981, Hunt, 1982; Sommerman et al., 1986; Sota et al., 1986). Lipid composition also significantly affects the clearance rate of liposomes from the blood (reviewed by Machy and Leserman, 1987; Senior, 1987). With regard to surface charge, the incorporation of negatively charged lipids such as phosphatidylglycerol in liposomes results in decreased circulation times in the blood and an enhanced propensity of the liposomes for mononuclear phagocytic cell types (Juliano and Stamp, 1975). It has been well documented that liposomes bind a variety of serum components (Bonte and Juliano, 1986).. Specifically, negatively charged EPG containing liposomes bind significantly more serum proteins than neutral liposomes. As indicated in Section 1.5.2, these liposomally associated serum components may play a role in the rapid clearance of EPG liposomes from the blood stream. Recent studies have shown that intravenous injection of negatively charged liposomes induces a transient thrombocytopenia in rodents (Reinish et al, 1988; Doerschuk et al, 1988). This effect is most striking for liposome preparations containing the negatively charged lipid phosphatidylglycerol (EPG). The transient reduction in platelet counts is associated with sequestration of platelets and liposomes in the liver and lung (Doerschuk et al, 1988). This indicates a potential role of platelets in the removal of liposomes from the circulation following intravenous injection. 93 Preliminary investigations suggest that the transient thrombocytopenia results from a transient liposome-platelet interaction (Reinish et al, 1988). The nature of this interaction is clearly of interest. It is plausible that serum components associated with EPG liposomes mediate the binding of liposomes to platelets. For this reason, the in vitro interaction of liposomes with platelets was investigated. In this chapter it is shown that EPG containing liposomes associate with rat platelets in platelet rich plasma. This interaction, which results in the formation of platelet microaggregates, is shown to be dependent oh the presence of plasma. The sensitivity of this adhesion process to treatment of plasma with heat, purified cobra venom factor and removal of the complement factor C3, are consistent with a requirement for complement in the binding of EPG liposomes to platelets. The inclusion of reactive lipid derivatives such as the thiol binding lipid MPB-DPPE in liposomes is also shown to promote the adhesion of liposomes to platelets. 5.2 Materials and Methods 5.2.1 Materials Apyrase (grade 1), Na heparin (porcine intestinal mucosa, grade 1), 4-chloro-naphthol and cholesterol were obtained from Sigma. Egg phosphatidylglycerol (EPG) were purchased from Avanti Polar Lipids. Purified cobra venom factor (Naja naja kaouthia, obtained from Diamedix) and peroxidase labelled anti-goat IgG (H and L, purchased from Jackson Immunoresearch Laboratories Inc.) were generous gifts from Dr. D. Devine. Goat anti-rat C3 was obtained from Organon Teknika. Cyanogen bromide activated Sepharose CL-4B was obtained from Pharmacia. All other chemicals were of standard analytical grade. Female rats of albino Wistar strain were obtained from the 94 University of British Columbia Animal Care Center or from Charles River Laboratories. The animals weighed between 225-250g at the time of use. 5.2.2 Preparation of platelets and plasma Rats were injected with heparin (10 units/kg) prior to collection of blood by heart puncture. Blood (7 ml) was collected in tubes containing 100 units of heparin. Platelet rich plasma (PRP) and platelet poor plasma (PPP) were prepared by standard haematological techniques (Packham, 1977). Briefly, whole blood was centrifuged at 400g for 20 minutes at 22°C. The resulting supernatant was spun at 800g for 30 minutes at 22°C to generate PPP. Plasma was heat inactivated by heating plasma' at 54°C for 30 minutes. Plasma was treated with cobra venom factor (160 /xg/ml; Muller-Eberbard, 1966) for 30 minutes at 37°C. The complement dependent hemolysis of antibody-sensitized erythrocytes (Mayer, 1961) in the presence of C a + + and M g + + was negligible for the plasma preparations either after heat treatment or cobra venom treatment. EDTA plasma (45%) containing 20 mM EDTA was titrated to pH 7.5 with NaOH. For generation of plasma deficient in complement component 3 (C3), goat anti-rat C3 was coupled to cyanogen bromide activated Sepharose CL-4B by standard procedures (1.5 mg antibody bound/ml of packed sepharose, Axen et al., 1967). Plasma (1 ml) was incubated with goat anti-rat C3 sepharose (1 ml packed sepharose in 1 ml 25 mM HEPES, 150 mM NaCl, HBS) for 1 hour at 4°C. The resulting preparation showed no detectable C3 as determined by Western blot analysis. Serum was obtained from coagulated blood by centrifuging at 800g for 20 minutes at room temperature. Washed platelets were isolated from PRP (9 volumes diluted with 1 volume of acid citrate dextrose) by centrifuging at 800g for 20 minutes at room temperature and resuspended in Tyrodes-HEPES buffer (Mustard et al., 1972) containing apyrase (0.5 mg/ml) and heparin (50 units/ml). The 95 concentration of platelets was adjusted to 400,000/ml and the final concentrations of HEPES, C a + + and M g + + were 20 mM, 2 mM and 1 mM respectively. 5.2.3 Preparation of liposomes Frozen and thawed multilamellar vesicles (FATMLVs, Mayer et al., 1985) were prepared by first dissolving the lipids in chloroform and then evaporating the solvent under nitrogen, followed by drying under high vacuum for 1 hour. The dried lipid film was hydrated in HBS, and then freeze-thawed 5 times employing liquid nitrogen for the freezing cycles. 5.2.4 Assay of platelet-liposome interaction Initially, the interaction of liposomes with platelets was visualized by light microscopy. Briefly, liposomes (20 pi of 5 mM total lipid in HBS) were incubated with PRP (400,000 platelets/ml, 80% plasma) at room temperature. Phase contrast photomicrographs of samples were taken after a 15 minute incubation period with a 35 mm camera mounted on a Leitz Laborlux-D light microscope (final magnification 480 x). The association of liposomes with platelets was also indicated by a reduction of platelet count as determined using an automated blood counter (Coulter Counter Model T660) after gentle mixing of liposomes with platelets. The basis of the assay is that EPG liposome induced aggregation of platelets results in the exclusion of the aggregate from the platelet window, resulting in a decline in the platelet count recorded by the counter. Since FATMLVs can be of a similar size to platelets, lipid samples (EPC/ cholesterol and EPG/ EPC/ cholesterol; 50 pM) contributed to the platelet count (~ 1 and 2 % respectively, see Section 5.3.1). In this study, the platelet count at time t is expressed as 96 a percentage of the platelet count at time zero (which also includes the small contribution from liposomes). 5.2.5 Gel electrophoresis and Western blots Liposomes with associated plasma proteins were prepared by incubating liposomes (5 /imoles total lipid in 0.1 ml) with normal or EDTA (20 mM final concentration) treated plasma (4.5 ml) diluted with 100 mM HEPES, 150 mM NaCl pH 7.5 (total volume 6 ml) for 15 minutes at room temperature and were washed extensively at 4°C with HBS in the presence of C a + + (2 mM) and M g + + (1 mM) for normal plasma liposomes or in the absence of divalent cations for EDTA plasma liposome samples. Liposomes with associated plasma proteins (1 pmole lipid equivalent) were prepared for gel electrophoresis by delipidation according to the method of Wessel and Flugge (1984), and solubilized in reducing sample buffer (100 p\) for 1 hour at 37°C. Plasma (diluted 50x with reducing sample buffer) was prepared for gel electrophoresis as described for the liposome samples. Aliquots (1 pi) were examined by discontinuous SDS-polyacrylamide gel electrophoresis (SDS-PAGE) on a 10-15% gradient resolving gel using the PhastGel system (Pharmacia). Gels were washed in transfer buffer (20 mM Tris-acetate, 2 mM EDTA and 0.01% SDS (w/v) at pH 7.4) for 20 minutes and electrophoretically transferred for 45 minutes onto nitrocellulose (Nitroplus 2000, Micron Separations) in a mini electrophoretic transfer apparatus (Pharmacia) at 50 mA at 4°C. After gel transfer, nitrocellulose was quenched' with 5% (w/v) powdered milk, 2% (w/v) BSA in phosphate buffered saline (PBS) overnight, washed and incubated with goat anti-rat C34 (75 pg/ml in PBS) for 3 hours. After extensive washing, rabbit anti-goat antibody (lOOOx diluted in 2% (w/v) BSA, 5% (w/v) skim powder milk, PBS) was incubated for 90 minutes, washed and labelled proteins were visualized with the peroxide immuno-stain using the chromagen 4-chloro-naphthol. 97 5.3 Results 5.3.1 Lipid dependence of platelet-liposome interactions In previous studies, negatively charged liposome preparations were shown to induce a transient thrombocytopenia in rodents (Reinish et al, 1988; Doerschuk et al, 1988). This in vitro study was initiated in order to determine what factors are involved in this event. Initially the interactions of liposomes with platelets were examined by light microscopy. It is shown that incubation of EPG liposomes with platelet rich plasma (PRP) for 15 minutes resulted in the formation of platelet microaggregates, in which the negatively charged liposomes form an integral part (Figure 5.1 A). The requirement of EPG in this process was shown by the absence of aggregates on incubation of EPC/ cholesterol liposomes with platelet rich plasma (Figure 5.IB). This association of EPG liposomes with platelets was also shown to result in a time dependent reduction in platelet counts as recorded by a Coulter counter (Figure 5.2 and Table 5.1). The effect of various concentrations of plasma and EPG liposomes on the EPG induced decline in platelet counts was examined in order to maximize this response. As shown in Figure 5.2, a rapid decline in platelet count was observed over 15 minutes on addition of EPG liposomes to PRP. The extent of the decline in platelet count induced by EPG liposomes (50 ttm lipid) was minimally affected by the levels of plasma (25-65%). Thus, in all subsequent plasma reconstitution experiments, the final concentration of plasma was 45%. In Table 5.1*. a significant decrease in platelet counts was observed at 15 minutes for a range of EPG liposome concentrations (50-500 pM lipid). .As liposomes at higher lipid concentrations interfere with platelet count estimates (Table 5.1), lipid concentrations of 50 pM lipid were employed subsequently. 98 Figure 5.1 Formation of platelet-liposome microaggregates: requirement for EPG Liposomes (100 nmole total lipid of (A) 10 mole % EPG, 45 mole % EPC, 45 mole % CHOL or (B) 55 mole % EPC, 45 mole % CHOL) were incubated with PRP (400,000 platelets/ml, 80% plasma) for 15 minutes at room temperature and subsequently examined by light microscopy. 99 Figure 5.2 Effect of plasma concentration on the interaction of EPG liposomes with platelets. EPG liposomes (10 pi of 5 mM) were added to washed platelets reconstituted with plasma (25%-65%) and platelet number was determined over time using a Coulter counter as described in Section 5.2.4. (O) 25% plasma, (#) 45% plasma, and (A) 65% plasma. 0-1 1 H 1 1 — 1 0 5 10 15 20 25 30 Time ( mins. ) 100 Table 5.1 Effect of EPG liposome concentration on platelet count.3 Lipid concentration ( / i M ) Lipid contribution to platelet count" (%) Platelet count at 15 min. c (%) 0 0 95 50 2. 64 100 4 58 250 10 48 500 20 45 (a) EPG liposomes (10 /il of 5-50 mM lipid) were added to washed platelets reconstituted with plasma (45 %) and platelet count was determined as described in Section 5.2.4. (k) This is expressed as the percentage of the total count in the platelet window (at zero time) which arises from the liposomes. (c) This is expressed as the counts in the platelet window at 15 min. as a percentage of the counts at time zero. 101 5.3.2 Requirement of complement In Figure 5.3 the requirement for plasma is illustrated by the absence of a reduction in the platelet count for washed platelet preparations in the presence of EPG liposomes. Addition of plasma resulted in a rapid decline in platelet count which was equivalent to that obtained for PRP. The lipid dependence of this interaction indicated by light microscopy studies, is also illustrated by the slight, much slower platelet count decline observed on incubation of EPC/ cholesterol liposomes with PRP. It is further illustrated that synthetic lipid derivatives such as the thiol reactive lipid MPB-DPPE, can induce a decline in platelet counts, indicating an association of such liposomes with platelets (Figure 5.4). To identify the plasma factor(s) required for the liposome dependent reduction in platelet count, EPG liposomes were incubated with washed platelets to which serum or plasma derivatives had been added (Figures 5.5 and 5.6). The substitution of serum by plasma did not affect the rate or extent of EPG liposome binding to platelets, indicating that blood coagulation activity was not involved in the binding process. However, the addition of heat inactivated plasma (54°C for 30 minutes) did not result in any decline in platelet counts induced by EPG liposomes, demonstrating that the plasma factor is heat labile. Treatment of plasma with purified cobra venom factor also resulted in loss of the capacity of EPG liposomes to induce aggregation as indicated by a decline in platelet count. Finally, a requirement for divalent cations in the process was shown by the lack of any EPG liposome induced decline in platelet count in the presence of EDTA treated plasma. These results suggest that the EPG liposome induced reduction in platelet counts is dependent on a complement factor. To conclusively demonstrate the requirement of complement for platelet-liposome interactions, plasma deficient in C3 was prepared by immunoaffinity chromatography on 102 Figure 5 . 3 Binding of EPG liposomes to platelets: requirement of plasma. Liposomes (50 nmoles of 10% EPG, 45 mole % CHOL, 45 mole % EPC or 55 mole % EPC, 45 mole % CHOL) were added to PRP or washed platelets in the presence or absence of plasma, gently mixed and sampled at various times for platelet count as outlined in Section 5.2.4. ( A ) PRP and EPG liposomes, (#) washed platelets, plasma and EPG liposomes, (A) washed platelets and EPG liposomes and (O) washed platelets, plasma and EPC liposomes. 103 Figure 5.4 Effect of MPB-DPPE liposomes on platelet count. Liposomes (50 nmoles of 1 mole % MPB-PE, 45 mole % CHOL, 54 mole % EPC (•) and 45 mole % CHOL, 45 mole % EPC, 10 mole % EPG (O)) were added to PRP, gently mixed and sampled at various times for platelet count as outlined in Section 5.2.4. 100 80 -o a o wo & 60 V -o- — C ) » 4 0 -20 05 i P-. o EH 0 10 15 20 Time ( mins. ) 25 30 104 Figure 5.5 Determination of the plasma component(s) involved in the binding of EPG liposomes to platelets: EPG Liposomes (50 nmoles) were added to washed platelet preparations reconstituted with various types of plasma or serum (45%) and assayed for platelet count. Characterization of plasma component: (A) plasma; (•) serum; (A) EDTA plasma; ( O ) heat inactivated plasma and (#) cobra venom treated plasma. C 100 3 o o s 8 0 -o m 1 ti g, 60 - - 9! q> 40-S 20-o 5 10 15 20 25 30 Time ( mins. ) 105 goat anti-rat-sepharose. Removal of C3, as confirmed by Western blot analysis (data not shown), totally inhibited the reduction in platelet counts (Figure 5.6). To further demonstrate the possible involvement of C3 and / or C3 cleavage products in the binding process, a Western blot of protein associated with EPG liposomes after incubation with plasma was probed with goat anti-rat C3 IgG (Figure 5.7). The third component of complement C3, is composed of two disulphide linked polypeptide chains of molecular weight (MW) 123,000 (a chain) and 76,000 (0 chain, Daha et al., 1979). As shown in Figure 5.7, two major bands were detected by blot analysis of plasma (lanes 1 and 2) with goat anti-rat C3 IgG on a 10-15% reducing SDS-PAGE gel. Some minor labelling of the band of MW 65-70 kD with control non-immune goat serum was observed with all samples (data not shown). To confirm the identity of the higher MW band as the a chain of C3, plasma was treated with purified cobra venom factor (lane 3). This results in the disappearance of the high molecular weight band with the concomitant appearance of a low molecular weight band (MW: approximately 40 kD), a pattern which is consistent with the formation of the rat equivalent of the C3 cleavage fragment, iC3b. Analysis of protein associated with EPG liposomes after 15 minutes incubation with plasma (lane 4), indicates a similar protein profile to that obtained for cobra venom treated plasma, with the additional appearance of an unidentified high molecular protein band (approximate 140-150 kD). The interaction of liposomes with platelets requires the presence of negatively charged lipids, cations and intact complement factors. To demonstrate that levels of liposomally associated C3 (and C3 cleavage fragments) reflect these requirements, protein associated with EPC liposomes incubated with plasma (lane 5), and EPG liposomes incubated with EDTA plasma (lane 6) or cobra venom treated plasma (lane 7) were screened with anti-rat C3 IgG. The results indicate that negligible levels of the complement factor C3 or its cleavage products associate with liposomes under these conditions. 106 Figure 5.6 Effect of removal of C3 from plasma on binding of EPG liposomes to platelets: EPG liposomes (50 nmoles) were added to washed platelets reconstituted with normal plasma (45%, O) or plasma that had been treated with goat anti-rat C3 coupled sepharose (#). 100 o o o m o O 8 0 -g. 60 -O. - C ) v 4 0 -2 0 -0 'to PH •—I a! +-> o 10 20 30 40 50 60 Time ( mins. ) 107 Figure 5.7 Western blot of protein bound to EPC and EPG liposomes using goat anti-rat C3 IgG. Samples were prepared as in Section 5.2.5 and run on a 10-15% reducing SDS-PAGE gel. C3 and iC3b cleavage products are identified. Lane 1, plasma treated as described for liposome samples; lane 2, plasma; lane 3, cobra venom treated plasma; lane 4, C3 and C3 cleavage fragments associated with EPG liposomes after 15 minutes incubation with plasma; lane 5, EPC liposomes incubated with plasma; lane 6, EPG liposomes incubated with EDTA plasma; lane 7, EPG liposomes incubated with cobra venom treated plasma. L A N E 1 2 3 4 5 6 7 — 200 ac3 C¥iC3b 1 16 96 55 43 36 29 18 12 108 5 . 5 Discussion In this chapter, it is shown that platelet-liposome microaggregates readily form in vitro on incubation of negatively charged EPG liposomes with platelets in the presence of plasma. This study was initiated in order to determine the nature of the factors involved in this adhesion event. The extent of liposome-platelet binding was followed by platelet count measurements using a Coulter counter assay. The participation of complement in platelet-liposome interactions was initially suggested by the heat labile nature of the mediator. This was further confirmed by experiments in which treatment of plasma with cobra venom factor, which specifically degrades the complement factor C3 (Muller-Eberbard et al., 1966), or removal of C3 by treatment of plasma with goat anti-rat C3 coupled to sepharose, abolished the EPG liposome-platelet interaction Previous work has shown that the rapid clearance of EPG liposomes from the blood is associated with a transient drop in circulating platelets (Reinish et al, 1988; Doerschuk et al, 1988). Biodistribution studies also show that EPG liposomes and platelets are sequestered together within the lung as well as the liver (Doerschuk et al, 1988). These observations indicate that platelets are involved in the clearance behavior of EPG liposomes. As it is well established that particular matter is cleared in vivo by cells of the mononuclear phagocytic system (MPS) in a size dependent manner (Altura, 1980; Hunt, 1982), the observation that EPG liposome-platelet microaggregates of > 20 fim in diameter readily form in vivo would suggest that the size of the platelet-liposome aggregate leads to an enhanced clearance rate. of EPG liposomes from the circulation. Furthermore, the in vivo transient reduction in platelet count may arise from trapping of these microaggregates within the micro-vasculature of the lung and the liver. Various serum factors which bind to liposomes (Bonte and Juliano, 1986) have a specific role in phagocytosis of particles by various MPS cell types (Saba, 1970). In this regard, studies in this Chapter show that components of the complement cascade, which 109 Fig ure 5.8 Classical and Alternative pathways of complement activation. (Taken from Kundel et al., 1985) Figure 5.9 A schematic representation of the proteolytic cleavage fragments of the third component of complement. S-C-0 _1_J C3 s^ s sT i i C3bBb /C4b2a C3a| H S O C - O R s-s s-s C3b Factor I IDI . 1 s-s s-I jC3b C3dg Factor I HS O-C-OR M ' ' : f — n s-s s-s C3c I I C3d C3e C3g H S O - C - O R ^s _L s-s s-s (Figure donated by D r . D . Devine) 1 1 1 have been implicated in the removal of bacteria from the circulation, associate with EPG liposomes. This group of sequentially interacting proteins can be activated by immunoglobulins (aggregated or complexed to antigen; termed the Classical Pathway) and also by non-immunological materials (carbohydrate polymers, many components of bacterial and yeast cell walls, and certain cell membranes; termed the Alternative Pathway; Henson and Ginsberg, 1981). The pivotal role of the third complement component C3, in this cascade is illustrated by the convergence of the two pathways at this point (Figure 5.8). Activation of complement results in the generation of a diversity of cleavage fragments of C3 (Figure 5.9) which can interact with soluble complement components (which can result in their stabilization or the inhibition of the complement cascade) as well as various membrane receptors distributed on a variety of cell types (which result in the processing of immune complexes or promotes binding and phagocytosis of coated particles by various cells; reviewed by Whaley, 1985; Lambris, 1988). Analysis of liposomally associated protein by Western blot techniques with anti-rat C3 IgG demonstrates the lipid specific association of proteolytic fragments of C3 (corresponding to the rat equivalent of human iC3b; Whaley, 1985) with EPG liposomes, under conditions which result 'in maximum aggregation of EPG liposomes and platelets. It is well established that platelets from rodents have CR1 type receptors that bind C3b and iC3b avidly (Henson and Ginsberg, 1981; Manthei et al., 1988). The above findings support the possibility that the deposition of C3b and/or iC3b on EPG liposomes results in the adhesion of EPG liposomes to rat- platelets via the CR1 receptor. This interpretation is consistent with the observations that negatively charged liposomes consume complement (Comis et al.; 1986; Chonn et al., 1989) and can bind purified C3b in vitro (Thielens and Colomb, 1986). Furthermore, human platelets which lack CR1 receptors (Manthei et al, 1988), do not interact with EPG liposomes in the presence of human plasma (unpublished observations), even though incubation of EPG liposomes 112 with human plasma results in consumption of C3 (Chonn et al., 1989). However, other components such as C4b (which associate with negatively charged liposomes; Thielens and Colomb, 1986) as well as adhesive proteins such as fibronectin (which bind to anionic liposomes and cleavage fragments of C3; Bonte and Juliano, 1986; Leivo and Engvall, 1986) may play a role in mediating this interaction. The finding that platelet microaggregates are formed on incubation with EPG liposomes in a complement dependent manner is not limited to negatively charged inert particles such as liposomes. Similar observations have implicated complement in the adhesion of rabbit platelets to negatively charged microcapsules (Muramatsu and Kondo, 1984). Here it is shown that liposomes containing the thiol reactive coupling lipid MPB-DPPE induce a decline in platelet counts when added to platelet rich plasma. This finding is consistent with the observation that other thiol reactive lipid derivatives such as PDP-PE, when incorporated into liposomes bind the third component of complement (Okada et al, 1988). In summary, the results of this chapter show that in order for EPG liposomes to bind to platelets in vitro, the complement component C3 must be present. The formation of platelet-liposome microaggregates of significant size offers a feasible explanation for the rapid removal of negatively charged liposome from the blood circulation. This adhesion phenomena is not restricted to negatively charged particles but is also shown to be characteristic of liposomes containing thiol reactive lipid derivatives. These findings stress the critical role that lipid structure plays in the in vitro interaction of platelets with liposomes which may influence their in vivo behavior. 113 6. SUMMARIZING DISCUSSION The studies presented in this thesis were initiated to further the application of protein-liposome conjugates as a general biochemical or pharmaceutical targeted-carrier system. Specifically, this thesis has addressed limitations in this area of liposome technology which concern the lack of generality and flexibility of available methods for the conjugation of proteins to liposomes. Physically well characterized protein coupled vesicles which have favorable in vivo properties are described. These conjugates have promising potential as targeted carriers in in vivo applications. Initially, a' generalized method for the preparation of protein-liposome conjugates, which employs the affinity of streptavidin for biotinated proteins, is described. This approach requires the prior coupling of streptavidin in a non-covalent manner to liposomes containing biotin EPE or more efficiently by the covalent attachment of streptavidin to liposomes containing MPB-DP^E (Chapters 2 and 3). Gentle incubation with biotinated proteins results in the rapid and efficient preparation of protein-liposome conjugates. These retain their biological function and bind in a specific manner to defined cells. The covalent method of coupling proteins to liposomes via the derivatized lipid MPB-DPPE was further investigated, primarily due to the development of a better method of preparing pure MPB-DPPE. Optimized coupling conditions are shown to result in the efficient conjugation of proteins to liposomes containing low concentrations of MPB-DPPE (1 mole %, Chapter 3). Similar coupling efficiencies have been obtained only on incorporation of higher concentrations of MPB-PE (5 mole %) in liposomes, levels which adversely affect liposomal integrity. These observations may be of importance in the use of protein-liposome. conjugates in applications which are dependent on the retention of small molecules (such as fluorescent compounds and hydrophilic drugs) within the aqueous compartment of liposomes. 114 A physical study of the effect of attaching proteins to liposomes indicated that aggregation of protein-liposome conjugates is a problem during coupling protocols (Chapter 4). This can drastically enhance the clearance rate of protein coupled vesicles from the circulation. A method was therefore developed for the preparation of small, homogeneously sized protein coupled vesicles by extrusion techniques. A significant finding of this thesis is that aggregated streptavidin-liposome conjugates, after extrusion exhibit long circulation half-lives. These small, extruded conjugates with well defined size distributions retain their ability to bind biotin after extended circulation times in the blood and thus should be capable of binding to target cells via defined biotinated ligands in vivo. The interaction of serum proteins with liposomes may also affect the in vivo properties of protein-liposome conjugates. For example, large liposomes containing negatively charged lipids (such as EPG) or thiol reactive lipid derivatives (such as MPB-PE) associate with rat platelets in vitro to form platelet microaggregates (Chapter 5). This process is mediated by complement. Previous studies have shown that intravenous administration of anionic liposomes induces a transient thrombocytopenia in rats. This is associated with an enhanced clearance rate of liposomes from the circulation. This is suggested to be due to the formation of large platelet-liposome aggregates in the blood which are subsequently removed rapidly from the circulation by the MPS. The adhesion of platelets to liposomes is restricted however to large liposome preparations (< 0.2 /un in diameter; Reinish et al., 1989) and thus minimally affects the in vivo application of small sized liposomes containing EPG and MPB-PE. In summary, the studies described in this thesis lead to a number of interesting areas of research which require further investigation. First, the generalized, straightforward and optimized method for the preparation of small homogeneously sized protein-liposome conjugates, which exhibit enhanced blood circulation times, should assist in the application of protein coupled vesicles in targeting and diagnostic protocols 115 in vivo. Specifically, the versatile coupling methodology described here is not restricted to the use of antibodies as the targeting ligand but may take advantage of the targeting potential of other proteins such as growth factors, small peptides as well as lectins. Furthermore, cocktail mixtures of biotinated proteins can readily be used which exploit more than one property of the target cell. A further advantage is that two targeting approaches are possible; biotinated ligands may be initially attached to streptavidin-liposome conjugates or alternatively, cells may be pre-labelled with biotinated targeting molecules in vivo followed by the administration of streptavidin-liposome conjugates. The latter approach may have significant advantages over direct targeting protocols, as it is anticipated that the clearance behavior of pre-formed targeted-liposomes will be dependent on the type of protein attached to vesicles (due to particular properties such as the size of the protein and carbohydrate content). With regard to the optimization of the targeting properties of protein-liposome conjugates, further work is required to determine whether factors such as protein size, the number of copies of protein per vesicle, size of protein-liposome conjugates as well as the type of linker and length of the spacer arm employed in the coupling reaction will contribute to an enhanced immune response to liposomally bound protein. These studies will also be informative in the development of liposomally based vaccines. A second aspect of interest focuses on the interaction of liposomes of various compositions with platelets. Further studies are required to identify the platelet receptor involved in this adhesion process and to determine whether serum components other than complement are involved. Furthermore, as in vivo studies indicate that the association of platelets with EPG liposomes is a transient process (Reinish et al., 1988), an investigation of factors involved in this transient event may provide useful information regarding the mechanism of clearance of liposomes from the blood. 116 BIBLIOGRAPHY Abra.R.M. and Hunt,C.A. (1981) Biochim. Biophys. Acta §66, 493-503. Abra,R.M., Bosworth.M.E. and Hunt,C.A. (1980) Res. Comm. Chem. Pathol. Pharmacol. 22, 349-360. Allen,T.M. (1984) In "Liposome Technology", vol III, Gregoriadis,G. (Ed.), CRC Press Inc., pp 177-182. Allen.T.M. and Cleland,L.G. (1980) Biochim. Biophys. Acta 527, 418-426. Allison,A.C. and Gregoriadis.G. (1974) Nature (London) 25.2, 252-254. Altura,B.M. (1980) Adv. Microcirculation 9, 252-294. Alving,C.R., Steck, E.A. and Chapman.W.L. (1978) Proc. Natl. Acad. Sci. USA 75,, 2959-2963. Axen,R., PorathJ. and Ernback,S. (1967) Nature (London) 214. 1302-1304. Bangham,A.D., Standish.M.M. and Watkins.J.C. (,1965) J.Mol.BioL 13^.238-252.-Barbet,J., Machy,P. and Leserman,L.D. (1981) / . Supramol. Struct. Cell Biochem. 16, 243-258. Barenholz.Y., Amselem„S. and Lichtenberg,D. (1979) FEBS Lett. 22, 210-214. Bayer,E.A. and Wilchek.M. (1984) In "Liposome Technology", vol I, Gregoriadis, G. (Ed), CRC Press Inc., Boca Raton, Florida, pp 127-135. Bayer.E.D., Rivnay.B. and Skutelsky,E. (1979) Biochim. Biophys. Acta 5_5_Q, 464-473. Bayer.E.A., Wilchek,M. and Skutelsky,E. (1976) FEBS Lett. 68, 240-243. Black.C.D.V. and Gregoriadis,G. (1976) Biochem. Soc. Trans. 4, 253-256. Blok.M.C, van der Neut-Kok,E.C.M., van Deenan.L.L.M. and de Grier,J. (1975) Biochim. Biophys. Acta 406,, 187-192. Bonte,F. and Juliano,R.L. (1986) Chem. Phys. Lip. 42, 359-372. Boyum.A. (1968) Scand. J. Cin. Lab. Invest. 21_, Suppl. 97, 9-13. Bradfield.J.W.B. (1984) In "Microspheres and Drug Therapy, Pharmaceutical, Immunological and Medical Aspects", Davis,S.S., Ilium, L., McVie,J.G. and Tomlinson, E. (Eds'), pp 25- 42. Bragman,K.S., Heath,T.D. and Papahadjopoulos.D. (1984) J. Natl. Cancer Inst. 73, 127-131. 117 Bragman.K.S., Heath,T.D. and Papahadjopoulos,D. (1983) Biochim. Biophys. Acta 730. 187-195. Bredehorst,R., Ligler,F.S., Kusterbeck,A.W., Chang.E.L., Gaber,B.P. and Vodel,C. (1986) Biochemistry 25, 5693-5698. Campbell,P.L, Harding,N.G.L., ryman,B.E. anmd Tyrrell,D.A. (1980) Eur. J. Biochem. 109. 87-92. Carlsson.J., Drevin,H. and Axen,R. (1978) Biochem. J. 173. 723-737. Chaiet,L., Miller.T.W., Tausig,F. and Wolf,F.J. (1963) Antimicrob. Agents Chemother. 3, 28-32. Chobanian,J.V., Tall,A.R. and Brecher,P.I. (1979) Biochemistry 1_8, 180-186. Chonn.A., Devine.D. and Cullis.P.R. (1989) (to be published). Comis,A. and Easterbrook-Smith,S.B. (1986) FEBS Lett. 122, 321-327. Daha.M.R., Stuffers-Heiman,M., Kilstra.A. and Van Es.L.A. (1979) Immunology 36, 63-70. DerksenJ.T.P. and Scherphof, G.L. (1985) Biochim. Biophys. Acta 814, 151-155. Doerschuk,C.M., Gie.R.P., Bally.M.B., Cullis,P.R. and Reinish, L.W. (1988) Thrombosis and Hemostasis 60, 524-529. Duncan.R.J.S., Weston.P.D. and Wrigglesworth,R. (1983) Anal. Biochem. JJ2, 68-73. Dunnick,J.K., McDougall.I.R., Aragon,S., Goris,M.L. and Kriss.J.P. (1975) J. Nucl. Med. i£ , 483-487. Ehrlich.P. (1906) Collected Studies on Immunology. New York, John Whiley. Vol II, pp. 442-447. Ellens.H., Morselt.H. and Scherphof.G. (1981) Biochim. Biophys. Acta 624, 10-18. FidlerJ.J., Hart.I.R., Raz,A., Fogler.W.F., Kirsh,R. and Poste,G. (1980) In "Liposomes and Immune-biology", Tom.B.H. and Six.H.R. (Eds.), New York, Amsterdam, Elsevier / North Holland, pp 109-118. Finkelstein,M.C. Kuhn.S.H., Schieren,H., Weissmann.G. and Hoffstein,S. (1981) Biochim. Biophys. Acta 6J3, 286-302. Finkelstein,M.C. and Weissmann,G. (1979) Biochim. Biophys. Acta 5_8_2, 202-209. Finn,F.*M., Iwata,N., Titus,G. and Hofman,K. (1981) Heppe-Seyler's Z. Physiol. Chem. 362. 679-684. Finn,F.M., Titus,G. and MontibellerJ.A. (1980) J. Biol. Chem. 255, 5742-5746. Fiske.C.H. and Subbarrow,Y. (1925) J. Biol. Chem. 66, 325-329. 118 FolkmanJ. (1985) Adv. Cancer Res. 43_, 175-196. Gabizon.A. and Papahadjopoulos,D. (1988) Proc. Natl. Acad. Sci. USA 8_5_, 6949-6953. Gitman,A.G. and Loyter,A. (1984) J. Biol. Chem. 252, 9813-9820. Godfrey.W., Doe,B., Wallace.E.F., Bredt,B. and Wofsy.L. (1981) Exp. Cell Res. 135, 137-145. Godfrey,W., Doe,B. and Wofsy, L. (1983) Proc. Natl. Acad. Sci. USA 8J), 2267-2271. Good,R.A., Kapoor.N. and Reisner.Y. (1983) Cell. Immunol. §2, 36-54. Gregoriadis, G. , Senior, J. and Trouet, A. (1982) Eds. In "Targeting of drugs", Nato Advances Study Institute Series. Series A, Life sciences, Vol. 47. Gregoriadis,G. and Davis,C. (1979) Biochem. Biophys. Res. Commun. 82, 1287-1293. Gregoriadis,G. and Neerunjun,E.D. (1975) Biochem. Biophys. Res. Commun. £5_, 537-544. Gregoriadis,G.,Leathwood,P.D. and Ryman.B.E. (1971) FEBS Lett. 14, 95-99. Hashimoto,K., Loader.J.E., Knight,M.S. and Kinsky,S.C.-(1985)B/oc/»'/w. Biophys. Acta... -816. 169-178. Heath,T.D. (1987) In " Methods in Enzymology", 142, part B, pp 111-118. Heath.T.D. and Martin,F.J. (1986) Chem.Phys.Lipids 4fl, 347-358. Heath,T.D., Lopez, N.G. and Papahadjopoulos.D. (1985) Biochim. Biophys. Acta 820, 74-84. Heath,T.D., Montgomery,J.A., Piper.J.R. and Papahadjopoulos.D. (1983) Proc. Natl. Acad. Sci. USA 80, 1377-1381. Heath,T.D., Frayley.R.T. and Papahadjopoulos.D. (1981) Science 210, 539-541. Heath.T.D., Robertson.D., Birbeck,M.S.C. and Davies,A.J.S. (1980) Biochim. Biophys. Acta 522, 42-62. Heitzmann,H. and Richards,F.M. (1974) Proc. Natl. Acad. Sci. USA 7J., 3537-3541. Henson.P.M and Ginsberg.M.H. (1981) Immunological reactions in platelets. In "Platelets in biology and pathology". Vol. 2,. GordonJ. (Ed.). Elsevier/North Holland Biomedical Press, Amsterdam, The Netherlands. Ho.R.J.Y. and Huang,L. (1985) J. Immunol. 134, 4035-4040. Hope,M.J., Bally,M.B., Mayer,L.D., Janoff.A.S. and Cullis,P.R. (1986) Chem. Phys. Lip. 40, 89-107. Hope,M.J., Bally,M.B., Webb,G. and Cullis,P.R. (1985) Biochim. Biophys. Acta. 8J2, 55-65. 119 Howard,F.D., Ledbetter.J.A., Wong,J., Bieber.C.P., Stinson.E.B. and Herzenberg,L.A. (1981) J. Immunol. 126, 2117-2121. Huang,A. Kennel,S. and Huang.L. (1983) J. Biol. Chem. 258., 14034-14040. Huang,A., Huang.L. and Kennel.SJ. (1980) /. Biol. Chem. 255_, 8015-8018. Huang.C.H. (1969) Biochemistry 8, 344-352. Huang.L. In " Liposomes ", (1983), Ostro.M.J. (Ed.), New York, Marcel Deker, pp 87-124. Huang,L. and Kennel,S.J. (1979) Biochemistry 18, 1702-1707. Hunt.A.C. (1982) Biochim. Biophys. Acta 7_1£, 450-463. Ishimori.Y., Yasuda,T., Tsumita,T., Notuski.M., Koyama,M. and Tadakuma,T. (1984) J. Immunol. Methods 75, 351-360. Jackson.C.M. (1980) Ann. Rev. Biochem. 49, 765-811. Janson.V.K. and Mallet.P.L. (1981) Analyt. Biochem. Ill, 54-59. Juliano,R.L. and Stamp.D. (1975) Biochem. Biophys. Res. Commun. (Q, 651-658. Kagawa,Y. and Packer,E. (1971) J. Biol. Chem. 246, 5477-5487. Kirby.C. and Gregoriadis,G. (1984) In "Liposome Technology", vol I, Gregoriadis, G. (Ed.), CRC Press, Boca Raton, Florida, pp 19-27. Kitagawa,T., Shimozono,T., Aikawa.T., Yoshida,T. and Nishimura, H. (1981) Chem. Pharm. Bull. 29_, 1130-1135. Klausner,R.D., Bluthmenthal,R., Innerarity.T. and Weinstein,J.N. (1985) J. Biol. Chem. 260. 13719-13727. Kundel,S.L., Ward,P.A., Caporale.L.H. and Vogel, K. (1985) In "Immunology III", Bellanti,J.A. (Ed.), W. B. Saunders Company, pp 106-116. Kung.V.T. Maxim,P.E., Veltri.R.W. and Martin,F.J. (1985) Biochim. Biophys. Acta 839. 105-109. Lambris.J.D. (1988) Immunol. Today 9 , 387-393. Lammmli.U.K. (1970) Nature (London) 227, 680-685. Leivo,I and Engvall,E. (1986) J. Cell. Biol. 103, 1091-1100. Leserman.L.D., Machy.P. and BarbetJ. (1984) In 'Liposome Technology", Vol. Ill, Gregoriadis,G. (Ed.), CRC Press Inc., Boca Raton, Florida, 29-40. Leserman,L.D., Machy,P. and Barbet,J. (1981) Nature (London) 293., 226-228. 120 Leserman L.D., Barbet,J.,Kourilsky,F. and Weinstein.J.N. (1980) Nature (London), 288. 602-604. Lopez-Berestein,G. and Juliano,R.L. (1987) In "Liposomes: From Biophysics to Therapeutics", Ostro,M.J. (Ed), New York, Marcel Dekker, pp 253-276. Lopez-Berestein.G, Hopfer,R.L., Mehta.R., Mehta,K., Hersh.M.E. and Juliano.R.L. (1984) J. Infect. Dis. 13Q, 278-283. Loughrey,H.,Bally,M.B. and Cullis,P.R. (1987) Biochim. Biophys. Acta. 20J., 157-160. Machy,P., Lewis,F„ McMillan,L. (1988) Proc. Natl. Acad. Sci. USA £5, 8027-8031. Machy.P. and Leserman,L. (1987) In "Liposomes in Cell Biology and Pharmacology", John Libbey and Co. LTD, London. Machy,P. and Leserman,L. (1983) Biochim. Biophys. Acta 730. 3131-320. Machy.P., Barbet,J. and Leserman.L.D. (1982) Proc. Natl. Acad. Sci. USA 79, 4148-4152. Manthei.U., Nickells.M.W., Barnes.S.H., Ballard.L.L., Cyi.W. and Atkinson,J.P. (1988) J. Immunol. 140, 1228-1235. Markwell, M.A.K. (1982) Anal. Biochem. 125, 427-432. Martin.F.J. and Papahadjopoulos.D. (1982) J. Biol. Chem. 25J7, 286-288. Martin,F.J., Hubbell.L. and Papahadjopoulos,D. (1981) Biochemistry 20., 4229-4238. Matthay.K.K., Heath,T.D., Badger,C.C, Bernstein,I.D.and Papahadjopoulos.D. (1986) Cancer Res. 4J>, 4904-4910. Matthay.K.K., Heath,T.D., and Papahadjopoulos.D. (1984) Cancer Res. 44, 1880-1886. Mayer.L.D., Hope.MJ., Cullis.P.R. and Janoff,A.S. (1985a) Biochim. Biophys. Acta 817. 193-196. Mayer,L.D., Bally.M.B., Hope.M.J. and Cullis,P.R. (1985b) Biochim. Biophys. Acta 816. 294-302. Mayer,M.M. (1961) In 'Experimental Immunochemistry", Thomas, Kabat, E.A. and Mayer, M.M. (Eds.), Springfield, pp. 133-145. Mimms,L.T., Zampighi,G., Nozaki.Y., Tanford,C. and Reynolds.J.A. (1981) Biochemistry 20_, 833-840. Morgan,J.R., Williams.K.E., Davies et al., (1981) J. Med. Microbiol. 14, 213-217. Muller-Eberbard,H.J., Nilsson,U., Dalmasso,A. and Polley.M. (1966) Arch. Pathol. §2, 205-213. Muramatsu.N. and Kondo.T. (1984) Biochim. Biophys. Acta 774, 75-80. 121 Mustard,J.F., Perry,D.W., Ardlie,N.G. and Packham,M.A. (1972) Brit. J. Haem. 22, 193-204. New,R.R.C, Chance,M.L., Thomas,S.C. and Peters,W. (1978) Nature, (London), 272, 55-59. Nicolau,C, Legrand.A. and Soriano.P. (1984) In "Cell Fusion", Pitman Books, London, Ciba Foundation Symposium 103, pp 254-267. Nisoff,A., Wissler,F.C, Lipman,L.N. and Woernly,D.L. (1960) Arch. Biochem. Biophys. 52, 230-244. OgiharaJ., Kojima,S. and Jay,M. (1986) Eur. J. Nucl. Med. U, 405-411. Okada,N., Yasuda,T. and Okada,H. (1988) Biochem. Biophys. Res. Commun. \5\, 743-748. Packham.N.S., Kinlough-Rathbone.R.L. and Mustard,J.F. (1977) In "Platelet Function Testing", Day.HJ., Holmsen.H. and Zucker,M.B. (Eds.), DHEW publications, No. (NIH) 78-1087, pp. 143-145. Panus.P.C. and Freeman,B.A. (1988) In "Liposomes as Drug Carriers", Gregoriadis,G.(Ed.), John Wiley and Sons Ltd, pp 473-484. PapahadjopouIos,D. and Gabizon, A. (1987) In "Annals of the New York Academy of Sciences", Juliano,R.L. (Ed.) 5JV7, 64-74. Perez-Soler.R., Khokhar,A.R. and Lopez-Berestein,G. (1988) In "Liposomes as Drug Carriers", Gregoriadis,G.(Ed.), John Wiley and Sons Ltd, pp 401-418. Peterson.H.I. (1979) In "Tumor Blood Circulation", Peterson,H.I (Ed), Boca Raton, Florida, CRC Press Inc., pp 152-175. Poste,G. (1983) Biol. Cell 47, 19-38. Poste.G., Bucana,C, Raz,A., Bugelski.P., Kirsh.R. and Fidler.I.J. (1982) Cancer Res. 42, 1412-1422. Ralston,E., Hjelmeland,L.M., Klausner.R.D., WeinsteinJ.N. and BlumenthaI,R. (1981) Biochim. Biophys. Acta 249, 133-137. Reinish.L.W., Bally,M.B., Loughrey, H.C. and Cullis,P.R. (1988) Thrombosis and Hemostasis 60, 518-523. Rinderknecht.H. (1962) Nature (London) 21, 167-171. Rosenberg.M.B., Breakefield,X.O. and Hawrot.E. (1987) / . Neurochem. 48, 865-875. Saba.T.M. (1970) Arch. Intern. Med. 126, 1031. Sedlak,J. and Lindsay,R.h. (1968) Anal. Biochem. 25, 192-205. 122 Senior,J.H. (1987) In "CRC Critical Reviews in Therapeutic Drug Carrier Systems", Vol. 3, pp 123-193. Senior.J. and Gregoriadis,G. (1982) Life Sci. 10, 2123-2136. Sinha.D. and Karush,F. (1979) Biochem. Biophys. Res. Commun. 9J), 554-560. Sommerman,E.F., Chonn,A.,Yeung,C, Pritchard.P.H. and Cullis,P.R. (1986) Biochim. Biophys. Acta, (submitted). Sommerman,E.F., Pritchard.P.H. and Cullis.P.R. (1984) Biochem. Biophys. Res. Commun. 122. 319-324. Sota,Y., Kiwada,H. and Kato,Y. (1986) Chem. Pharm. Bull. 34, 4244-4252. Stashenko,PM Nadler.L.M., Hardy,R. and Schlossman,S.F. (1980) J. Immunol. 125, 1678-1685. Stein.Y., Halperin.G. and Stein.O. (1980) FEBS Lett. U, 104-106. Szoka,F. and Papahadjopoulos.D. (1980) Ann. Rev. Biophys. Bioeng. 9 , 467-508. Szoka.F. and Papahadjopoulos.D. (1978) Proc. Natl. Acad: Sci. USA 79_, 4194-4198. Tack,B.F., Harrison.R.A., JanatovaJ., Thomas.M.L. and Prahl.J.W. (1980) Proc. Natl. Acad. Sci. USATL, 5764-5768. Thielens.N.M. and Colomb.M.G. (1986) Eur. J. Immunol. 16, 617-622. Torchilin,V.P., Goldmacher,V.S. and Smirniv,V.N. (1978) Biochem. Biophys. Res. Commun. 85_, 983-990. Uemura.K. and Kinsky,S.C. (1972) Biochemistry U, 4085-4094. Urdal.D.L. and Hakomori.S. (1980) J. Biol. Chem. 255., 10509-10516. Vitetta,E.S., Krolick.K.A., Miyama-Inaba.M., Cushley.W. and Uhr.J.W. (1983) Science 219. 644-650. Weinstein.J.N., Blumenthal,R., Sharrow,S.O. and Henkart.P.A. (1978) Biochim. Biophys. Acta 509, 272-288. Weiss,L. and Greep.R.O. (1977) Histology, Fourth edition, McGraw Hill, San Fransisco. Weissmann.G., Brand,A. and Franklin,E.C. (1974) J. Clin. Invest. 5 1 , 536-543. WesseLD. and Flugge.U.I. (1984) Anal. Biochem. 138, 141-143. Whaley,K. (ed). (1985) In "Methods in complement for clinical immunologists", Church Livingstone, Edinburgh, Scotland, pp. 1-21. Williams,B.D., O'Sullivan.M.M. Saggu,G.S. et al., (1986) Br. J. Rheum. 25, 98-103. 123 Wolff,B- and Gregoriadis.G. (1984) Biochim. Biophys. Acta £Q2, 259-273. Wong,T.K., Nicolau.C. and Hofschneider.P.H. (1980), Gene JJL 87-93. Yoshitake,S., Yamada,T., Ishikawa,E. and Masseyeff,R. (1979) Eur. J. Biochem. 101. 395-399. Youle.RJ. and Neville.D.M.Jr. (1980) Proc. Natl. Acad. Sci. USAJ1, 5483-5486. 124 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0098285/manifest

Comment

Related Items