UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Liposomal immunosuppressants for the management of organ transplantation Choice, Edward G. 1998

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

Item Metadata


831-ubc_1998-271196.pdf [ 18.21MB ]
JSON: 831-1.0088744.json
JSON-LD: 831-1.0088744-ld.json
RDF/XML (Pretty): 831-1.0088744-rdf.xml
RDF/JSON: 831-1.0088744-rdf.json
Turtle: 831-1.0088744-turtle.txt
N-Triples: 831-1.0088744-rdf-ntriples.txt
Original Record: 831-1.0088744-source.json
Full Text

Full Text

LIPOSOMAL IMMUNOSUPPRESSANTS FOR THE MANAGEMENT OF ORGAN TRANSPLANTATION By EDWARD G. CHOICE B.Sc, The University of British Columbia, 1992 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Department of Pharmacology & Therapeutics, Faculty of Medicine) We accept this as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA April 1998 © Edward G. Choice, 1998 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of The University of British Columbia Vancouver, Canada Date A p / t l <39 ; DE-6 (2/88) A B S T R A C T This thesis examines liposomal drug carriers for the delivery of immunosuppressive agents. A finite amount of cyclosporine was found to incorporate into liposomes. The amount of cyclosporine incorporated was inversely related to the amount of cholesterol present in the liposomal bilayer. Cyclosporine incorporated into liposomes remained associated with this carrier when subject to pore filtration, gel exclusion chromatography or density gradient ultracentrifugation. When donor liposomes carrying cyclosporine are mixed with acceptor liposomes devoid of cyclosporine however, a rapid redistribution of cyclosporine occurred between two liposome populations. This redistribution, proportional to the relative size of the lipid sink (acceptor) compared to the initial liposomal cyclosporine formulation, observed in vitro would predict a similar rapid but even larger loss of the cyclosporine in vivo. In mice, liposomal cyclosporine plasma elimination and tissue biodistribution of the drug confirmed the hypothesis that the cyclosporine does not stay with the liposome as predicted by the in vitro studies. Furthermore, the tissue biodistribution of cyclosporine differed from that of its liposome carrier. A comparison between liposomal versus cremophor cyclosporine resulted in insignificant differences in the plasma elimination rates and the extent of accumulation in tissues of the drug. When liposomal cyclosporine was tested in a rat heterotopic heart transplant model, the liposomal carrier accumulated in statistically significant higher levels in the cardiac grafts compared to that of hosts' hearts. Cyclosporine delivery to the cardiac grafts was, however, not improved. Therefore, liposomes are good candidates to deliver ii drugs to cardiac graft sites but cyclosporine is not readily deliverable in liposomes. Fast protein liquid chromatography (FPLC) was studied to address the challenge of characterizing drug loss. FPLC was determined to be an extremely powerful method for the separation of liposomes from plasma components in order to monitor component loss from liposomal drug formulations. This method for isolating liposomes from plasma components is regarded to have broad applicability to the study of liposomal component exchange from various liposomal drug carrier formulations. The approach taken to address the problem of rapid drug exchange was to utilize antibodies that could be covalently attached to the liposome exterior. Certain antibodies directed against lymphocytes have been identified to have immunosuppressive properties. For the purposes of establishing the feasibility of delivering immunosuppressive antibodies coupled to liposomes, however, a readily available antibody, human IgG, was utilized to study whether the antibody remains with its liposomal carrier in vivo. When the antibody-coupled liposomes are injected into severely combined immunodeficient (SCfD) mice, similar plasma elimination characteristics of human IgG and liposomes provided strong evidence for the stability of the covalent attachment between human IgG and the liposome. The FPLC separation of liposomes from plasma components developed in this thesis confirmed that the antibody, human IgG, remained attached to liposome in plasma. The biodistribution of the antibody-coupled liposome formulation varied in each type of tissue. Histochemical analysis revealed co-localization of the human IgG and the liposomes in the spleen and, to a lesser extent, in the liver. These results provide evidence for the feasibility of attaching other antibodies to liposomes in order to take advantage of the carrier's ability to accumulate at sites of ongoing rejection. ui TABLE OF CONTENTS ABSTRACT TABLE OF CONTENTS LIST OF FIGURES LIST OF TABLES ABBREVIATIONS ACKNOWLEDGMENTS DEDICATION CHAPTER 1: INTRODUCTION 1.1 Cyclosporins 1.1.1 Structure 1.1.2 Pharmacokinetics 1.1.3 Pharmacodynamics 1.1.4 Mechanism of Action 1.1.5 Toxicity 1.2 Monoclonal antibody OKT3 1.2.1 Structure of OKT3 1.2.2 Pharmacokinetic Properties of OKT3 1.2.3 Pharmacodynamic Properties of OKT3 1.2.4 Mechanism of Action of OKT3 1.2.5 Toxicity and Adverse Effects associated with OKT3 therapy 1.3 Liposomes 1.3.1 Lipid building blocks 1.3.2 Preparation and Classification of Liposomes 1.3.3 Stability of liposomes iv 1.3.4 Role of cholesterol in model membrane Systems 1.3.5 Passive targeting of liposomes 1.4 Liposomal Drug Delivery 1.4.1 Drug association with liposomes 1.4.2 Rationale for cyclosporine delivery in liposomes 1.4.3 Rationale for antibody delivery in liposomes 1.5 Immunosuppression in Organ Transplantation 1.5.1 Organ Transplantation Immunology 1.5.2 Aims of Immunosuppression in Organ Transplantation 1.6 Summary of Research Obj ectives 1.7 Research Objectives CHAPTER 2: CHARACTERIZATION OF CYCLOSPORINE DRUG INCORPORATION 57 2.1 Introduction 57 2.2 Materials and Methods 60 2.2.1 Lipids and Chemicals 60 2.2.2 Liposomes 60 2.2.3 Drug incorporation procedure 63 2.2.4 Liquid scintillation radiolabel quantitation 64 2.2.5 Effect of cholesterol, anionic lipid POPG, 65 storage temperature CsA incorporation 2.2.6 Gel exclusion chromatography 65 2.2.7 Isopycnic density gradient 66 2.3 Results 67 2.3.1 Effect of ethanol on drug incorporation 67 2.3.2 Effect of varying cyclosporin A mole% and/ 69 or lipid composition on drug incorporation 29 29 32 32 33 34 35 35 53 56 56 v 2.3.3 Effect of increasing cholesterol content in liposomes on incorporation 72 2.3.4 Stability of drug incorporation 72 2.4 Discussion 82 CHAPTER 3: MOLECULE EXCHANGE CHARACTERISTICS OF CYCLOSPORINE FROM LIPOSOMAL BILAYERS: IN VITRO STUDIES 84 3.1 Introduction 84 3.2 Materials and Methods 86 3.2.1 Lipids and Chemicals 86 3.2.2 Liposome Preparation 86 3.2.3 Incorporation of cyclosporine into vesicles 87 3.2.4 Anion Exchange Chromatography 87 3.2.5 Determination of vesicle lamellarity using 3 1P-NMR 88 3.3 Results 89 3.3.1 Interbilayer exchange and the effect of cholesterol content 91 3.3.2 Effect of multilamellarity on interbilayer exchange in vitro 93 3.4 Discussion 97 CHAPTER 4: COMPARISON OF CYCLOSPORINE AND LIPID CARRIER PHARMACOKINETICS AND BIODISTRTBUTION 98 4.1 Introduction 99 4.2 Materials and Methods 101 4.2.1 Lipids and Chemicals 101 4.2.2 Drug and lipid formulation 101 4.2.3 Mouse tail vein injections 102 4.2.4 Blood and Tissue collection 102 4.2.5 Blood and tissue analysis 102 vi 4.2.6 In vitro exchange of liposomal cyclosporine in whole blood 4.3 Results 4.3.1 Drug and Lipid Carrier Plasma Elimination Studies in a Mouse Model 4.3.2 Drug Distribution in Mouse Whole Blood 4.3.3 Drug and Liposome Tissue Biodistribution 4.4 Discussion CHAPTER 5: LIPOSOMAL CYCLOSPORINE TISSUE DISTRIBUTION IN HEART TRANSPLANTATION 5.1 Introduction 5.2 Materials and Methods 5.2.1 Lipids and Chemicals 5.2.2 Drug and Lipid Formulation 5.2.3 Rat Pancreas and Heart Transplantation 5.2.4 Tail vein inj ections 5.2.5 Blood analysis 5.2.6 Tissue analysis 5.3 Results 5.3.1 Drug and Liposome Tissue Distribution 5.3.2 Drug and Liposome Distribution in Blood versus Tissues 5.4 Discussion CHAPTER 6: SEPARATION OF LIPOSOMES FROM PLASMA LIPOPROTEINS USING FAST PROTEIN LIQUID CHROMATOGRAPHY (FPLC) 6.1 Introduction 6.2 Materials and Methods vii 6.3 Results 6.2.1 Lipids and Chemicals 147 6.2.2 Blood Samples 147 6.2.3 Isolation of lipoprotein fractions by Density Gradient Ultracentrifugation 148 6.2.4 Lipoprotein Gel Electrophoresis 148 6.2.5 FPLC Instrumentation 149 6.2.6 Preparation of Liposomes 149 6.2.7 FPLC sample preparation 150 6.2.8 Removal of very low density lipoproteins 150 6.2.9 Tritiated cholesterol exchange into plasma lipoproteins 151 6.2.10 Detection of tracer molecules 152 153 6.4 Discussion 6.3.1 Identification and definition of lipoprotein components in SCJJ3 mouse plasma using Density Gradient Ultracentrifugation and lipoprotein gel analysis 153 6.3.2 Determination of FPLC elution profiles 155 6.3.3 Removal of VLDL from liposomes and plasma mixtures prior to FPLC analysis 158 6.3.4 Characterization of liposome separation from SCUD mouse lipoproteins on the FPLC 163 6.3.5 Evaluating the general application of FPLC and modified VLDL precipitation procedure on human plasma and CD1 mouse serum samples 165 6.3.6 Comparison of radioactive cholesteryl hexadecyl ether versus non-radioactive fluorescent Dil liposome markers on the FPLC 170 173 CHAPTER 7: ANTIBODY-COUPLED LIPOSOME STABILITY AND BIODISTRTBUTION IN SCID MICE 178 7.1 Introduction 179 7.2 Materials and Methods 182 7.2.1 Lipids and Chemicals 182 7.2.2 Preparation of large unilamellar vesicles 182 7.2.3 Iodination of Human IgG 183 viii 7.2.4 Antibody Coupling to Large Unilamellar Vesicles 7.2.5 Plasma Elimination and Biodistribution of Antibody-Coupled Liposomes 7.2.6 FPLC Instrumentation and Analysis of Fractions 7.2.7 Analysis of FPLC fractions using Qualitative Dot Blotting 7.2.8 Western Gel and Transfer Blot Analysis 7.2.9 Radiochemical and Fluorimetric analysis 7.2.10 Histochemistry 7.3 Results 7.3.1 Plasma Elimination Studies 7.3.2 FPLC elution profiles, Dot Blot and Western Transfer Blot Analysis 7.3.3 Tissue Biodistribution of Antibody-Coupled Liposomes 7.3.4 Histochemistry analysis of fluorescent markers 7.4 Discussion CHAPTER 8: SUMMARY REFERENCES ix LIST OF FIGURES Page Figure 1.1 The chemical structure of cyclosporin A 3 Figure 1.2 The mechanism of action of cyclosporin A 6 Figure 1.3 The basic four peptide chain structure of immunoglobulin antibodies 12 Figure 1.4 The mechanism of action of OKT3 monoclonal antibody as an immunosuppressant 17 Figure 1.5 Lipid structure and nomenclature 22 Figure 1.6 The preparation and freeze fracture electron micrographs of liposomes 25 Figure 1.7 The cells of the reticuloendothelial system 31 Figure 1.8 The complement pathways 37 Figure 1.9 The immune response 40 Figure 1.10 The ternary MHC-antigen-TCR-CD3 complex 42 Figure 1.11 Surface molecules aiding in MHC-antigen-TCR recognition 46 Figure 1.12 Immunosuppressive drugs 51 x Figure 2.1 Isolation and purification of cyclosporin A from Oral Formulation 61 Figure 2.2 Influence of ethanol on cyclosporine incorporation 70 Figure 2.3 Influence of cholesterol on cyclosporine incorporation into POPC and POPC :POPG vesicles 73 Figure 2.4 Sephadex G-50 gel permaeation chromatography of cyclosporine-loaded POPC vesicles 75 Figure 2.5 Isopycnic density gradient centrifugation of cyclosporine-loaded POPC vesicles 76 Figure 2.6 Influence of vesicle storage on cyclosporine incorporation 80 Figure 3.1 Separation of POPC and POPCPOPG vesicles on DEAE Sephacel 90 Figure 3.2 Cyclosporine exchange from donor to acceptor vesicles 92 Figure 3.3 Influence of manganese on the 3 1 P-NMR spectra of 400nm POPC vesicles 95 Figure 3.4 Influence of vesicle lamellarity on cyclosporine exchange 96 Figure 4.1 Plasma elimination of liposomal cyclosporine 107 Figure 4.2 Redistribution of liposomal cyclosporine in plasma 110 Figure 4.3 Tissue distribution of liposomal lipid as percentage of injected dose 113 xi Figure 4.4 Concentration of liposomal lipid in major organs 114 Figure 4.5 Tissue distribution of cyclosporine as a percentage of injected dose 115 Figure 4.6 Concentration of cyclosporine in major organs 116 Figure 5.1 Rat heart transplant model 126 Figure 5.2 Rat pancreas transplantation allograft survival with Cyclosporin A post-operative management 131 Figure 5.3 Biodistribution of liposomes to tissues in a rat heart transplant model as a percentage of initial dose 133 Figure 5.4 Biodistribution of cyclosporin A to tissues in a rat heart transplant model as a percentage of initial dose 134 Figure 5.5 Biodistribution of liposomes to tissues in a rat heart transplant model on a per gram weight basis 136 Figure 5.6 Biodistribution of cyclosporin A to tissues in a rat heart transplant model on a per gram weight basis 137 Figure 5.7 Blood and tissue distribution of liposomal cyclosporin A in Lewis rats 139 Figure 6.1 1% Agarose gel electrophoresis of SCID mouse plasma lipoproteins separated by density gradient ultracentrifiigation 154 Figure 6.2 The DGUC fractions run on the FPLC and recorded as a chromatogram of absorbance measurements at 280nm 156 xii Figure 6.3 Determination of lipoprotein content in fractions separated by FPLC 159 Figure 6.4 FPLC elution profiles for human IgG, liposomes and SCID mouse plasma 161 Figure 6.5 FPLC separation of liposomes and components in human plasma 166 Figure 6.6 FPLC separation of liposomes and components in CD1 mouse serum 168 Figure 6.7 Elution profiles for radioactive and non-radioactive liposome tracers 171 Figure 7.1 Antibody coupled liposomes 191 Figure 7.2 The plasma elimination of antibody-coupled liposomes 192 Figure 7.3 FPLC elution profile of antibody-coupled liposomes 194 Figure 7.4 Dot blot assay for antibody-coupled liposomes eluted by FPLC 198 Figure 7.5 Western gel electrophoresis and transfer blot analysis for antibody-coupled liposomes in plasma 199 Figure 7.6 The biodistribution of the antibody in SCID mouse tissues after injection of antibody-coupled liposomes 201 Figure 7.7 The biodistribution of liposomes in SCLD mouse tissues after injection of antibody-coupled liposomes 202 xiii Figure 7.8 Fluorescent histochemical analysis of antibody-coupled liposomes in SCfD mouse tissues xiv LIST OF TABLES Table 2.1 Incorporation of CsA into POPC liposomes containing varying proportions of ethanol Table 2.2 Cyclosporine incorporation as a function of initial drug: lipid ratio and vesicle lipid composition Table 4.1 Distribution of liposomal cyclosporine in whole blood Table 6.1 The selective precipitation of SCID mouse VLDL from liposomes in plasma xv A B B R E V I A T I O N S 3 H - tritiated aa amino acid ACE angiotensin converting enzyme ADCC antibody-dependent cell-mediated cytotoxicity apo B apolipoprotein B B bursus; (pertaining to cell type) BCA bichininoic acid C constant (pertaining to antibody region) CcAcB calmodulin-dependent calcineurin-A calcineurin-B (complex) CD cluster designation CDR complementarity determining regions C H constant region of the antibody heavy chain CHE cholesteryl hexadecyl ether CHOL cholesterol CMV cytomegalovirus CsA cyclosporin A CTL cytotoxic T lymphocytes Da dalton(s) DAG diacylglycerol DGUC density gradient ultracentrifugation Dil 1,1' -dioctadecyl-3,3,3', 3' -tetramethyllindocarbocyanine perchlorate DPPC 1,2-Dipalmitoyl sn-glycero-3-phosphatidylcholine DSPC l,2-Distearoyl-sn-glycero-3-phosphatidylcholine DSPE-MPB 4-(4-N-Maleimidophenyl)-butyryl-1,2-Distearoyl-s«-glycero-3-phosphatidyl-ethanolamine sodium salt xvi DSPE-PEG20oo l,2-Distearoyl-5«-glycero-3-phosphatidylethanol-amine-N-Poly[ethyleneglycol-2000] EDTA ethylenediamine-tetraacetic acid ELISA enzyme-linked immunosorbent assay Fab antibody binding fragment FATMLVs frozen and thawed multilamellar vesicles F c crystallizable fragment FITC fluorescein isothiocyanate FPLC fast protein liquid chromatography g force of gravity GFR glomerular filtration rate G M I monosialoganglioside GVH graft versus host LLAMA human anti-mouse antibody HBS HEPES buffered saline HDL high density lipoprotein HEPES [4-(2-hydroxyethyl]-piperazine ethanesulfonic acid HEV high-walled endothelium of the post-capillary venules HLA human lymphocyte antigen (allele) HSA heat stable antigen hlgG human immunoglobulin G LDL intermediate density lipoprotein i.p. intraperitoneal iv. intravascular JENy interferon-gamma LL interleukin LUV large unilamellar vesicle MAC membrane attack complex MHC major histocompatibility complex MLV multilamellar vesicle xvii MWCO molecular weight cut-off NF-ATc nuclear factor activating T cell cytoplasmic component NF-ATn nuclear factor activating T cell nuclear component NMR nuclear magnetic resonance OKT3 Muromonab-CD3; Orthoclone™ T3; anti-CD3 PDPH 3-(2-pyridyldithio)-propionyl hydrazide PEG polyethylene glycol PKC protein kinase C POPC 1 -Phosphatidyl-2-oleoyl-s«-glycero-phosphatidylcholine POPG 1 -Phosphatidyl-2-oleoyl-sft-glycero-phosphatidylglycerol RES reticuloendothelial system RIA radioimmunosorbant assay SA stearylamine SAS sodium acetate buffered saline SCID severely combined immunodeficient SUV small unilamellar vesicle T thymus (pertaining to cell type) TCR T cell receptor TMP-SMZ trimethoprim-sulphamethoxazole TNFa tumor necrosis factor-alpha UV ultraviolet (radiation) V variable (pertaining to antibody region) VLDL very low density lipoprotein x times (multiplication) xviii ACKNOWLEDGMENTS This thesis would not be possible without the help of my colleagues and friends. I would like to acknowledge my committee members and research fellows: Tom Madden, Marcel Bally, Mark Meloche and Cathy Pang for their assistance and input into my project. Tom Madden, my research supervisor, has been a vital part of my scientific and personal development. Tom Madden has put a lot of his time into writing superior grants and proof reading my written work. I have fond memories of lab members, Jeff, Miranda, Cliff, Mac and Gitanjali along with Tom and Linda outside of the laboratory. Our adventures, providing much needed stress relief from the many long hours and late nights in the lab, have taken us skiing, climbing, dining and involved in team sports like volleyball and softball. I think it is suffice to say that the lab builds character and balances hard work with challenging activities outside of the lab. Of course, former lab members like Cindy, Xue Min, Liu Ping, Diane and John have moved on but were integral my early scientific research and I wish them well in their endeavors. I want to thank other members of the Liposomal Research Unit for sharing their scientific expertise, time and experiences with me as well. Namely, the PRC lab (Pieter Cullis, Kim, Dave, Steven, Nancy, Barb, Myrna, Mike, Austin, Sean, Troy, Conrad, Ken, Lome, Dora, Angel, Alison, Ismail, Hanneke, Norbert, Elisabeth, Benny and Lenore), the MJH lab (Mick Hope, Wendy, Sandy, Wilson and Sam in the Skin Barrier Lab), the B C C A Advanced Therapeutics group (Marcel Bally, Lawrence Mayer, Hafiza, Dana, Natashia, Dody, Pierrot, Ellen, Howie, Rajesh, Francis and the rest of the staff). I would also like to acknowledge the Atherosclerosis Specialty Laboratory at St. Paul's Hospital. xix My collaborative research and interest in lipid biochemistry gave me an opportunity to work with this clinically orientated group of scientists. This diverse and knowledgeable group (Haydn Pritchard, Jiri Frohlich, Amir, Andy, Lida, Jonny, Ming, Scott, Sandra, Max, Ying Ying, Mohammed, Michelle and John) have been very receptive into making me a part of the lab. My thanks goes out to Lori and Susan for their technical assistance in the organ transplantation studies and those with whom I have had the pleasure of publishing scientific works. This research project was supported by the Medical Research Council of Canada, the National Cancer Institute of Canada and by research fellowships from the University of British Columbia and the Heart and Stroke Foundation of Canada. Finally, I want to thank my best friend, Selina, for helping me organize the references in this thesis. My academic and scientific dealings with the aforementioned individuals are cherished. Thank you all for your contributions! xx DEDICATION TO M Y FAMILY FOR THEIR SUPPORT, MOTIVATION AND UNDERSTANDING xxi CHAPTER 1 INTRODUCTION The purpose of this introduction is to give a brief review of the drugs and vehicle being investigated and to present the rationale for merging immunosuppressants and liposomes into a single entity in order to improve drug potency and ultimately improve immunosuppressive therapy in organ transplantation. 1.1 CYCLOSPORIN A Cyclosporin A, purified from Tolypocladium inflation Gams and Cyclindrocapon lucidum Booth fungal culture broths by Borel and his associates in 1970, is used today as an immunosuppressant in the management of postoperative organ transplantation. Cyclosporine inhibits the lymphocyte production of cytokines such as interleukin-2. Cytokines released by lymphocytes are responsible for signaling neighboring cells and the immune system to react against foreign molecules. The immune response to grafted tissue is the major cause of organ transplant rejection. By inhibiting the production of cytokines then, cyclosporine can reduce the incidence of organ rejection. This use of this drug is limited by side effects such as nephrotoxicity and myelosuppression. Incorporating cyclosporine into a liposome drug delivery vehicle may decrease adverse side effects. The liposome, being of larger size than a molecule of cyclosporine and composed of naturally occurring lipids, has the ability to not only escape glomerular filtration but can also passively target to cells of the reticuloendothelial system including sites of inflammation and disease. Hence, any cyclosporine incorporated into these lipid vehicles can accumulate in tissues targeted by liposomes away from the kidney. The 1 structure, pharmacokinetics, pharmacodynamics and toxicity of cyclosporine are described in the following sections. Beveridge (1992) describes the clinical development of cyclosporine. 1.1.1 Structure Cyclosporin A is a cyclic endecapeptide (MW=1203) composed of seven N-methylated amino acids, a D-configuration amino acid residue #8 and a special nine carbon amino acid residue #1 unique to, and essential for, the immunosuppressive activity of cyclosporines (Figure 1.1, Borel (1994)). Intramolecular hydrogen bonding between amino acids within this ring structure makes the molecule very hydrophobic in nature. Other immunosuppressants such as FK506 and rapamycin have similar hydrophobic properties and collectively these three immunosuppressants mediate their immunosuppressive effects by binding endogenous intracellular receptors called immunophilins (Schreiber and Crabtree, 1992). 1.1.2 Pharmacokinetics Cyclosporine can be administered orally or intravenously at dosages of 5-25 mg/kg/day. It is slowly and incompletely (20-50%, mean oral bioavailability 30%) absorbed when given orally (Fiocchi et al., 1994) but due to its hydrophobic nature is bound by plasma proteins and attains a high volume of distribution within the human body (85±15L). The therapeutically desired plasma trough cyclosporine levels range from 150-300 ng/ml for immunosuppression in the first 90 days and decrease to 50-200 ng/ml for longer periods of cyclosporine therapy. The elimination half-life of cyclosporine is 24 hours and it is metabolized in the liver by the P-450 cytochromes yielding demethylated and hydroxylated metabolites to be excreted mostly in the bile and 2 C H 3 N H c II C H 3 \ C , C H 3 H / C S C H 2 H | C H \ P H 3 HO.. . C £ " / C H 3 C H 2 C H 3 C H L J U^'P u C H 3 C H 2 C H 3 C H 3 — N — C — C O — N — C — C — N — C — C O — N — C — C N — C H 2 I aa10 aa11 Jl I aa1 I aa2 II aa3 l C H 3 H C O 9 C H 3 ! | | C - C H 2 — - C a a 9 I i I C H / I ! I I C H 3 — N H O H N — C H 3 | aa8 aa7 | aa6 || aa5 | aa4 \ O C — C . — N — C O — C — N — C — C — N — C — C — N — C O — C H k "H I 1 "H I  T'" H I i "H i C H 3 ? C H 3 ? | H H 2 C H 3 C H 3 ^ H X ? H 2 L J I CH , - 6 2 I 1 l l l i ^ l l , - ' l 2 C H 3 C H 3 10 11 1 2 3 MeLeu MeVal MeBmt—Abu—MeGly 9 MeLeu D Ala Ala MeLeu—Val—MeLeu 8 7 6 5 4 Figure 1.1: The chemical structure of cyclosporin A. The cyclic molecule is composed of 11 amino acids (aa) that can display intramolecular hydrogen bonding (—) and is relatively hydrophobic in nature. A simpler schematic diagram is shown as well where the abbreviated names of the amino acids are given. Nine of the eleven amino acids in cyclosporin A are methylated (Me) and the amino acid at position 1, N-methyl-4-butenyl-4-memyl-threonine (MeBmt), is important for the drug's immunosuppressive properties. 1-a-amino butyric acid (Abu); methyl-glycine(MeGly); methyl-leucine (MeLeu); valine (Val); alanine (Ala); mefhyl-valine (MeVal). 3 to a lesser degree (<10%) in the urine. In a 70 kg person, the clearance of this immunosuppressant is 410±70 ml/min. In the blood, 40-60% of the cyclosporine is bound to lipoproteins and cyclosporine plasma half-life is 5.6±2 hours (Katzung, 1989 andBorel, 1994). 1.1.3 Pharmacodynamics Cyclosporine selectively inhibits T-helper cell activation while slightly enhancing T-suppressor cell activity leading to greater tolerance of transplanted organs. This peptide antibiotic acts by blocking an early stage in T cell differentiation. Cyclosporine can inhibit the production of factors that stimulate T lymphocyte growth such as interleukin-2 (IL-2) production (Fletcher and Goldstein, 1987). The concomitant decrease in IL-2 receptor expression leads to a functional unresponsiveness of precursor cytotoxic T lymphocytes (CTLs) to IL-2. Consequently, fewer cells of the immune system are activated into clonal proliferation needed to mount a host immune response to foreign antigens presented on graft tissues. Cyclosporine is not cytotoxic to lymphocytes at therapeutic doses and its immunosuppressive effects are reversible upon removal; therefore requiring life-long administration of this drug to organ transplant patients. The outcome of cyclosporine drug intervention in the management of organ transplants is reviewed by Haverich (1992) and Riesbeck et al, (1992) for heart transplantation, McMaster and Mirza (1994) and Schultz and Meriney (1996) for liver transplantation, Paul et al., (1994) and Wauters et al., (1994) for kidney transplantation and Sutherland, Moudry-Muns and Gruessner (1994) for pancreas transplantation. 4 1.1.4 Mechanism of Action At the molecular level, cyclosporine binds to cyclophilin, which is in the T cell cytosol (Kay, 1989, Schreiber and Crabtree, 1992 and Niebylski and Petty, 1993). Cyclophilin serves as a peptidyl-propyl cis-trans isomerase involved in the folding of proteins (Fruman, Burakoff and Bierer, 1994) but the mechanism of cyclosporine acting through cytoplasmic cyclophilin does not require this catalytic reaction (Wingard, 1991 and Dale, Foreman and Fan, 1994). Rather, the cyclophilin-cyclosporine complex (Theriault et al, 1993) has the ability to bind to a calcineurin A- calcineurin B-calmodulin complex which leads to a block in its phosphatase activity necessary for translocation into the nucleus and subsequent activation of the cytoplasmic component of the nuclear factor partially responsible for activating T cells (NF-ATc) (Figure 1.2) (Hess, Tutschka, and Santos, 1981a,b and Heitman et al, 1994). Activated NF-ATc and its nuclear component, NF-ATn, play an integral role in the initiation of transcription of the IL-2 gene (Citterio and Kahan, 1991). From Figure 1.2, it is important to recognize that the levels of cyclosporine must be maintained to keep up with cellular component turn-over and to prevent dephosphorylation of NF-ATc by the calcineurin A- calcineurin B- calmodulin complex. An intricate intracellular signal transduction pathway exists for both cyclosporin A and for a similar macrolide immunosuppressant, FK506 (Liu, 1993). The dissociation of cyclosporin A from the complex has been proposed to be an energy and temperature dependent process involving multidrug resistance pump, P-glycoprotein, which exports cyclosporin A out of the cell (Batiuk, Pazderka and Halloran, 1994 and del Moral et al, 1995). Nussenblatt and Palestine (1986) and Thomas and Gordon (1986) present comprehensive reviews on the uses of cyclosporine. 5 Antigen Presenting Cell (APC) extracellular space R R R R R R R R R l T R R R R R f f ummmumu C s A ^ A cyclophilin-CsA T cell cytosol calmodulin IRfffiflRfffifl tytr^ PLC o (+) (+) IP DAG NF-ATc calmodulin calcineurin A-calcineurin B complex nucleus NF-ATc NF-ATn complex O PKC O RAS v 1 A 0 NF-ATn NF-AT NF-AT Interleukin-2 enhancer Interleukin-2 gene Figure 1.2: The mechanism of action of cyclosporin A. Cyclosporin A (black triangles) bound to cyclophilin (aqua symbol) inhibits the calmodulin-calcineurin A-calcineurin B (CcAcB) complex phosphatase activity on nuclear factor-acyltransferase cytoplasmic component (NF-ATc). Stimulation of the T cell is brought about by antigen (red molecule) presented on the major histocompatibility complex (MHCII) of the antigen presenting cell (APC) which becomes bound by the T cell receptor (TCR). The TCR is noncovalently attached to the cell designation-3 (CD3) complex. This CD3 complex is composed of 8,y,s and a C, dimer in most cases and is responsible for transmitting the activation signal to downstream kinases. One tyrosine kinase, p59fyn, is responsible for activating phospholipase C (PLC) resulting in the generation of diacylglycerol (DAG) and inositol trisphosphate (IP3). DAG, in turn, activates protein kinase C (PKC) which transmits the activation signal to a 189 amino acid (21 kDa) cellular protein (RAS) otherwise known as p21. RAS activates the nuclear factor-acyltransferase nuclear component (NF-ATn) in the nucleus to bind to the NF-ATc and together they bind to the IL-2 enhancer region on chromosome 3 in mice or chromosome 4 in humans. Meanwhile, IP3 initiates a release of calcium (Ca2+) from intracellular stores leading to an increase in calmodulin available to bind and activate the CcAcB molecule. When cyclosporin A is present however, NF-ATc is unable to traverse into the nucleus because CcAcB cannot dephosphorylate it. Hence, there is a lack of immunological response to the foreign antigen. Cyclosporin A is continuously being metabolized and sufficient levels must be present in order for it to maintain its immunosuppressive effects on the T cell. In other words, the inhibition of CcAcB is reversible and the state of immunological anergy (unresponsiveness) can be switched to immunological rejection if cyclosporin A is not present at sufficient levels. 7 1.1.5 Toxicity Cyclosporine has a number of untoward effects that have been documented with its usage. Notably, dose-dependent nephrotoxicity (Luke et al., 1987) resulting in a decrease in glomerular filtration rate (GFR) is seen with acute and chronic treatment (Andres et al., 1994 and Furlanut, 1994). In the acute case, a rapid increase in serum creatinine levels and vasoconstriction in and around the glomerulus accompanies the reduction in GFR. Renal function may decrease by one-third as fibrosis and deterioration results from chronic exposure of this drug (Bennet et al., 1994). ). The nephrotoxicity seen with cyclosporine usage may be due to lipid peroxidation (Suleymanlar et al., 1994a), preferential constriction of the afferent renal arteriole (Graham, 1994) or the formation of reactive oxygen intermediates (Wolf, et al, 1994). Decreasing the cyclosporine dose is indicated in these cases accompanied by the use of vasodilating agents, alpha-blocking agents, ACE inhibitors and calcium channel blockers (Suleymanlar et al, 1994b and Ar'Rajab et al, 1994). The antibiotic, rifampin, increases the metabolism of cyclosporine, thereby decreasing its immunosuppressive effect in the body. An increase in incidence of viral infections, some increase in hair growth and transient liver dysfunction have been documented for patients receiving cyclosporine. Finally, malignancy developing from excessive immunosuppressive regimens disturbing immune surveillance may occur especially when the immunosuppressive activity of cyclosporine is potentiated with the use of other immunosuppressants. The incidence of malignancy is roughly 5% occurring at a mean of 16 months after transplantation (range 3-45 months, n=334) for patients on cyclosporine triple (Ritters, Garbensee and Heering, 1994) or quadruple therapy 8 (Melosky et al, 1992). Graham (1994) and Thiel et al., (1994) have assessed the long-term benefits and risks of cyclosporine therapy. 1.2 MONOCLONAL ANTIBODY OKT3 The following is a presentation on a rapidly emerging class of immunosuppressants, the antibodies. There are a number of antibodies that cross-react with T cell surface molecules and have immunosuppressive properties. OKT3 is but one of many antibodies with immunosuppressive properties and although not used to perform any research in this thesis, is described here as an example of an antibody that could benefit from liposomal delivery. The monoclonal antibody OKT3 is used as an immunosuppressive agent in the treatment of acute renal, cardiac or hepatic allograft rejection and in cases where the graft has become refractory to conventional therapy or when conventional therapy is contraindicated. This monoclonal antibody is specific for, and defines, a distinctive human T cell surface antigen. Kung, Goldstein, Reinherz and Schlossman produced this anti-CD3 antibody in 1979 using Kohler and Milstein (1975) hybridoma technology to fuse mouse spleen cells with an immortalized cell line. Orthoclone OKT3 (Muromonab-CD3) marketed by Ortho Biotech was the first monoclonal antibody to be approved by the Food and Drug Administration (FDA) for clinical use in 1980 for the acute treatment of graft versus host (GVH) disease (Martin et al, 1988). It is one of many emerging monoclonal antibodies used in transplantation (Waldmann, 1988, Chatenoud and Bach, 1993, Chavin et al, 1993, Waldmann and Cobbold, 1993, Cosimi, 1995, Dantal and Soulillou, 1995 and DeFazio, Masli and Gozzo, 1996). Other monoclonal antibodies being developed include anti-VCAM-1 and 9 anti-VLA-4 antibodies (Isobe et al, 1994 and Groczynski et al, 1995), anti-TCKa/Bantibodies (Eto et al, 1994 and Heidecke, et al, 1995, 1996), anti-TCR antibody with super antigen (Gonzalo et al, 1994), anti-CD45RB antibody (Lazarovits et al, 1996), anti-IL2R antibody (Francois et al, 1996). OKT3 was compared to high dose steroid therapy in reversal of graft rejection and was found to be significantly more effective in extending kidney graft survival (94% OKT3 vs. 75% high dose steroid therapy initially (P=0.009), and 1 year later 62% vs. 45% (P=0.029), respectively, Orfho Multicenter Transplant Study Group, 1985). OKT3 is currently being used in the management of acute transplant organ rejection in sequential immunosuppressive regimens along with azathioprine, methylprednisolone/prednisone and delayed cyclosporine. Sequential immunosuppressive therapy was assessed to be more effective than conventional standard triple therapy (cyclosporine, azathioprine and methylprednisone) for decreasing the severity and incidence of rejection of hepatic and renal grafts (Wilde and Goa, 1996). 1.2.1 Structure of OKT3 A basic four-peptide unit that is apparent upon reduction of the disulfide bonds covalently joining them characterizes each antibody. Two heavy chains (y) of 50,000 daltons each and two light chains (K or X) of 25,000 daltons form the immunoglobulin G (IgG) molecule of 150,000 daltons. The structure of the most abundant type of antibody, IgG (8-16mg/ml in normal human serum, Roitt, 1994) is diagrammed in Figure 1.3. The therapeutic monoclonal antibody OKT3 is an immunoglobulin produced by mice that has specificity for the human T cell receptor-CD3 complex. OKT3 falls into the IgG2a 10 subclass of the IgG immunoglobulin family that exhibits differences in its ability to fix complement and react with microbial proteins when compared to other IgGs. In figure 1.3, the variable (V) region domains play a role in antigen recognition while the secondary biological functions brought about by antigen-antibody binding are mediated through the constant (C) region domains. Whereas breakage of interchain disulfide bonds yields the four peptide unit of the IgG, papain cleaves the same molecule into two antigen binding fragments (Fab) and one crystallizable fragment (Fc) which lacks the ability to bind antigen. The recognition sites on the antibody or papatopes for antigen binding sites or epitopes are located at the distal variable domains and the diversity of binding is manifested in the amino acid sequences in the hypervariable regions. The N-termini of the light and heavy peptide chains are associated with this end of the molecule while the C-termini ends are in the hinge and distal constant regions, respectively. The antibody molecule is not strictly composed of amino acids but also contains carbohydrate content ranging from 3% in IgG to 13% in IgD immunoglobulin classes. 1.2.2 Pharmacokinetic Properties of OKT3 The monoclonal antibody is administered as a single intravenous bolus of 5ml of lmg OKT3/ml to dose the patient at 5mg OKT3 a day for 10 to 14 days. This dosage can be followed thirty minutes later with a hydrocortisone sodium succinate solution to decrease the incidence of cytokine release and antipyretics and antihistamines are given as needed. As induction therapy in a sequential immunosuppressive regimen, OKT3 administration is followed by azathioprine, methylprednisolone/prednisone corticosteroids and delayed cyclosporine. The amount of OKT3 in the plasma or serum 11 12 Figure 1.3: The basic four-peptide chain structure of immunoglobulin antibodies. The peptide chains, two heavy and two light, of an antibody are held together by interchain disulfide linkages. The chains and regions of the immunoglobulin G class type 2 (IgG2) antibody are identified in the figure. Various two and three-dimensional models of the antibody are shown in the accompanying figure (Adapted from Essential Immunology, Roitt, ed.). Antibodies are recognized by components of the classical complement pathway. Antibodies are also bound by cell surface F c receptors which play an integral role in mediating killing of IgG antibody coated target cells through antibody-dependent cell cytotoxicity (ADCC). Metabolism of the antibody itself is mediated through Clq of complement pathway which binds to the second constant heavy gamma (y) chain. The four peptide chains of an antibody are shown as being interwoven (2 gray chains representing the light chains, a light purple chain and a darker purple chain representing the heavy (y) chains) and held together by interchain disulfide bonds (dark purple balls) in the upper right hand panel. Finally, the antibody binding fragment composed of constant and variable light chains are illustrated as pairs of anti-parallel beta (B) pleated sheets (A-G). The variable domain contains complementary determining regions (CDR) specific to antibody idiotypes where the numbers given indicate highly variable amino acid sites that change the specificity of the antibody. 13 can be measured using radioimmunosorbant assay (RIA), competitive or sandwich ELISAs and the goal is to maintain it at 0.9-lug OKT3/ml plasma over the treatment period. OKT3 serum levels were found to vary among individuals based on age, sex, transplanted organ, treatment regimen, and anti-OKT3 antibody status (Schroeder et al, 1994). While antibodies have a plasma or serum clearance half-lives in the order of days (Russell et al., 1992 and Routledge et al., 1995), initial dosing with OKT3 results in its clearance from circulation within minutes due to its metabolism once bound to TCR-CD3 on T cell surfaces. Subsequent doses of OKT3 result in longer half-lives that could be directly attributed to lower numbers of CD3 + T cells present. The metabolism of OKT3 has been followed using peptide mapping (Kroon, Baldwin-Ferro and Lalan, 1992) and the chemical mechanisms and the functional effects of degradation identified (Rao and Kroon, 1993). OKT3, having the CD3E molecular target, was found to have a plasma half-life of approximately 18 hours over the treatment period for rejection and up to 36 hours when given prophylactically. Thus, OKT3 plasma concentrations were found to be dependent upon the number of CD3 molecules available for reactivity on circulating T cells. The OKT3 plasma concentrations do vary also according to the type of transplanted organ and the age of the patient. Unless challenged by anti-idiotypic OKT3 antibodies, OKT3 half-life was found to be indifferent between days 5 to 9 thereby indicating that the rate of immune elimination by CD3 producing T cells remains constant (Norman and Leone, 1991). Cases in which human anti-mouse antibodies (HAMA) are detected against OKT3 occur in 75% of treated patients and therefore to an large extent negate the effectiveness of the immunosuppressant. Methods to get around the induction 14 of the HAJVIA response include engineering of humanized monoclonals which contain human crystallizable fragments fused to mouse anti-human TCR-CD3e variable regions and administering OKT3 under an umbrella of anti-CD4 antibodies which allows for tolerance of murine lg epitopes. Co-administration of azathioprine and/or cyclophosphamide can block anti-OKT3 antibody production in 20%-60% of patients (Norman, Shield III and Barry, 1987 and Goldstein, Fuccello and Norman, 1986). The development of anti-OKT3 antibodies in the patient does not entirely preclude the use of OKT3 because patients with low anti-idiotypic antibody titres (<1/1000) can have OKT3 successfully re-administered. 1.2.3 Pharmacodynamic Properties of OKT3 OKT3 reacts with human peripheral and tissue T cells but not other cells in the body. It binds specifically to the 20kDa glycoprotein CD3 at its epsilon (s) chain. The CD3s chain is responsible for recognition of antigen and signal transduction in the T cell. By binding this component, OKT3 blocks binding of antigen, signal transduction and downstream processes of T cell activation leading from this integral initial step in the generation and function of effector cells. The effector cells are responsible for the production of lymphokines and cytokines that signal neighboring cells. By inhibiting the initial and early processes in the immune response, OKT3 replaces cyclosporine or delays its use in the early management of organ transplantation (Norman et al., 1988a and Norman, 1993). A decrease in circulating CD3+, CD4+ and CD8+ cells is observed within minutes of OKT3 administration. This decrease is, in part, due to an internalization of CD3 surface receptors as will be discussed in Section 1.2.4. In part, the lack of the CD3 glycoprotein on the T cell surface makes recognition of foreign antigens presented to the 15 T cell receptor (TCR) by the major histocompatibility complex (MHC) of antigen presenting cells (APCs) very difficult. The close association of the TCR heterodimer with the CD3 complex is necessary for antigen recognition and downstream signaling within the T cell. OKT3 blocks the initial and early steps in the immune response to foreign antigens. By doing so, OKT3 blocks all T cell function and is appropriately classified as a pan-T cell suppressive monoclonal antibody. While CD4+ and CD8+ cell levels do recover within 2-9 days after initial administration, CD3 + cells do not reappear until 7 days after cessation of OKT3 therapy. OKT3 was found to be efficacious in heart transplantation (Renlund et al., 1989, Mahon, 1991, Alson-Pulpon et al, 1995, Haverty, Sanders and Sheahan, 1993 and Stapleton et al., 1993), liver transplantation (Gordon et al., 1988 and Wall et al, 1995), kidney transplantation (Norman et al., 1988b and Opelz, 1995) and pancreas transplantation (Sollinger and Stratta, 1988). The treatment of refractory cardiac rejection with OKT3 occurs when other immunosuppressive agents have not worked, as outlined by Gilbert et al., (1987a,b). 1.2.4 Mechanism of Action of OKT3 The mechanisms by which OKT3 inhibits the immune response to foreign grafted tissue are best illustrated in a diagram (Figure 1.4). OKT3 can coat and promote opsonization of circulating T cells by liver and spleen reticuloendothelial cells. Antigenic modulation of the TCR-CD3 through binding of the CD3s chain causes internalization of the CD3 glycoprotein on the surface of T lymphocytes rendering the T cell nonfunctional (Chatenoud et al., 1982 and Estabrook et al, 1983). Of particular importance to allograft rejection is a third mechanism in which OKT3 blocks functional noncirculating sessile T cells responsible for killing cells within the graft. 16 O K T 3 coats T cell 1 O K T 3 inhibits sessile T leading to T cell opsonization | ; cells from destroying graft O K T 3 down regulates C D 3 expression on T cell surface o p Antigen Presenting Cell (APC) G y l i f f i ^ MHCL TCR anti-CD3 antibody, OKT3 T cell cytosol CD3 $ tyrosine • kinase a OKT3 blocks initial and early steps in the activation of T cells v Of -] CD3 down \ l I modulation 17 Figure 1.4: The mechanism of action of O K T 3 monoclonal antibody as an immunosuppressant. OKT3 has been shown to bring about a state of immunosuppression by a combination of the three processes shown at the top of this figure. OKT3 (mouse anti-human) or 145-2C11 (hamster anti-mouse) monoclonal antibodies bind specifically to the epsilon (s) 20kDa subunit (orange) of CD3 on lymphocytes. The T cell receptor-CD3 complex is composed of two alpha (a) subunits (green), two beta (B) subunits in the TCR and delta (5, red oval), gamma (y, purple), epsilon (s, orange) and a pair of zeta (Q subunits (not shown) in the CD3. The red bar indicates points at which anti-CD3 antibodies have been shown to inhibit T cell activation. In the antigen presenting cell (APC) membrane is a major histocompatibility complex (MHC) with associated immunogen (red): 18 Other mechanisms have been proposed: the immunomodulation of graft-infiltrating lymphocytes, the elimination of activated CD3+ cells by induction of apoptosis, the modulation of CD3 complex density by shedding CD3 antigens or the whole CD3 complex, increasing lymphocyte adhesion molecule expression on peripheral blood lymphocytes resulting in increased adhesion of lymphocytes to vascular endothelium and the induction of cell-mediated cytolysis (reviewed by Wilde and Goa, 1996). 1.2.5 Toxicity and Adverse Effects associated with OKT3 therapy The first doses of OKT3 produce cytokine release syndrome (Duffy and Nestor, 1992, Doutrelepont, 1993, First, Schroederand and Hariharan, 1993 and Jeyarajah and Thistlewaite Jr., 1993) resulting in fever, chills and gastrointestinal disturbances. In an effort to minimize these side effects, varying doses of OKT3 have been administered (Norman, Kimball and Barry, 1993). The symptoms are manifest within an hour of dosing and can persist for up to two days. The symptoms of acute cytokine release syndrome experienced with OKT3 dosing are caused by massive tumor necrosis factor-alpha (TNFct), interferon-gamma (IFNy), and interleukin (IL) -2, -3 and -6 release into circulation as T cells are cleared by the reticuloendothelial system (RES). Radhakrishnan and Cohen (1993) discuss how to properly prepare the patient about to receive OKT3 therapy. The majority of the cytokine release symptoms are managed by pre-medication with hydrocortisone, acetaminophen, diphenhydramine and the use of a cooling blanket (Duffy and Nestor, 1992). Patients are also at risk of fatal pulmonary edema, especially those with predisposed left ventricular dysfunction, and precautionary measures include chest X-ray and diuresis if necessary prior to OKT3 (muromonab-CD3) administration (Costanzo-Nordin, 1993). In addition, because transplant patients are 19 immunocompromised, they are susceptible to bacterial and viral (cytomegalovirus and herpes simplex) infections. Lymphoproliferative disorders may add to the causes of morbidity and mortality associated with use of OKT3 in organ transplantation (van Wauwe, de Mey and Goossens, 1980). These adverse effects increase with high OKT3 doses (>75mg), long courses of OKT3 administration, multiple courses and early (prophylactic) treatment. Trimethoprim sulfa-methoxazole (TMP-SMZ) clears bacterial infections, acyclovir reduces the incidence of herpes simplex infections and immune serum globulin reduces the incidence and severity of cytomegalovirus (CMV) and Epstein-Barr infections associated with the OKT3 immunosuppressive regimen. Therapeutic regimens have also been developed in order to decrease adverse side effects, improve transplant success rates and patient recovery. The timing of switchover from OKT3 to the safer, and more convenient cyclosporine immunosuppressive regimen in sequential therapy is therefore also very vital to minimizing adverse reactions to the drug(s). Having given a brief review of the two types of immunosuppressants that were modeled in the thesis research, this introductory review shifts its focus. In the next section, the liposome carrier is introduced along with how it can be used to selectively deliver cyclosporine or antibodies in the management of organ transplant rejection. 20 1.3 L I P O S O M E S 1.3.1 Lipid building blocks Liposomes are aqueous filled spheres defined by a shell composed of a bilayer of lipids. The bilayer of lipids, much like a biomembrane to which proteins can be added later, can be composed of cholesterol, phospholipids, sphingolipids or other lipids (Lasic, 1997). The structures of some lipids such as cholesterol, phosphatidyl-oleoyl- and disteroyl- phosphatidylcholine (POPC and DSPC), phosphatidyl-oleoyl-phosphatidylglycerol (POPG), disteroyl-phophatidylethanolamine-polyethylene glycol-[2000] and -maleimidyl-phenyl-butaric acid (DSPE-PEG 2 0 0 0 and DSPE-MPB) are shown in Figure 1.5. 1.3.2 Preparation and Classification of liposomes A dry lipid film can be hydrated in water to spontaneously form multilamellar vesicles (MLVs) (Bangham, 1968). These vesicles are heterogeneous in size varying from 1 to 10 um in diameter and are composed of many concentric shells of phospholipid bilayers (Figure 1.6). The heterogeneity in size, lamellarity and relative large size of MLVs are undesirable because of the variability within a population of these vesicles and the fact that larger vesicles are cleared quickly from circulation (Juliano and Stamp, 1975). In order to prepare the multilamellar vesicles (MLV) for extrusion, the lipid suspension can be subjected to 5 cycles of freezing in liquid nitrogen (boiling point -196°C) and thawing in a water bath (temperature dependent on characteristic lipid transition temperature). This process disrupts and breaks the concentric bilayer shells apart resulting in vesicles having higher trapped volumes (Mayer et al., 1985). Frozen 21 R-R-O PLAl -C—O—CH 2 -C—O—CH PLD O O 1 C H 2 — O—P—Cr-r PLA2 PLC phosphatidyl--CH2CH2N+(CH3)3 -CH2CH2NH3 + OH acid choline ethanolamine glycerol 0 MPB 0 -CH2CH2NH—C—(OCH2CH2)4 5 -OCH 3 P E G 2 0 0 0 fatty acid glycerol polar head group(s) tails backbone CHOLESTEROL 22 Figure 1.5: Lipid structures and nomenclature. Lipids and phospholipids are major components in liposomes. Their structures, names and properties are illustrated in this figure (pages 22 and 23). The true phospholipid, sphingomyelin is shown and the amphipathic nature of phospholipids is illustrated. In the accompanying table are common names of fatty acids, their structural formula and the number double bonds in the fatty acyl chain. On the following page are examples of lipids: cholesterol, l-palmitoyl-2-oleoyl phosphatidylcholine (POPC) and l-palmitoyl-2-oleoyl phosphatidylglyderol (POPG). The latter two are glycerophospholipids that have an overall neutral and negative charge respectively at physiological pH. Phospholipids can be broken down at specific bonds by phospholipases (PLA-D). The presence of double bonds in the fatty acid acyl chains result in a kink (gauche isomerization) in the hydrocarbon tail resulting in less dense packing of phospholipids in liposomal bilayers. Coupling of molecules to the surface of the preformed liposome can be achieved through reaction with a maleimidophenyl-butryl (MPB) derivatized phospholipid. Longer chain length groups such as polyethylene glycol (PEG) (on average 45 repeating ethylene units for PEG 2 0 0 0) serve as a steric barrier to bilayer insertion by extrinsic molecules when present in lipid bilayers. When hydrated, lipids and phospholipids can form various structures termed liposomes in which lipids, organized in a bilayer fashion, form a sphere. 23 and thawed multilamellar vesicles (FATMLVs) generally smaller in diameter, have decreased lamellarity than MLVs (Figure 1.6) and can be further downsized by 10 cycles of high pressure extrusion (Hope et al., 1985) generating vesicles of a defined size. The most common types of vesicles generated from this method are lOOnm diameter large unilamellar vesicles (LUVs)(Figure 1.6). Extrusion produces vesicles of the size range between 80nm - 800nm with larger vesicles possessing multilamellae. Small unilamellar vesicles (SUVs) of 25nm - 80nm diameter can be produced by sonication of MLVs. The extrusion technique is a rapid procedure for preparing liposomes of a defined size. Alternatively, liposomes can be prepared by reverse phase or solvent evaporation (Szoka et al., 1980), ethanol injection (Batzri and Korn, 1973), ether infusion (Deamer and Bangham, 1976) and detergent dialysis (Mimms et al, 1981) methods of vesicle preparation. The distribution of vesicle sizes within a lipid sample or population of vesicles can be analyzed using a particle sizer based on dynamic light scattering (Frokjaer, Hjorth, and Worts, 1984). In this method, also known as photon correlation spectroscopy, a focused 5mW helium-neon (HeNe) or 75mW argon-ion laser beam is directed through a dilute lipid vesicle sample where the suspended particles randomly diffuse in Brownian motion. As the laser beam encounters the suspended particles a scattering of light occurs. These scattered light waves mutually combine or interfere with one another before arriving at a pinhole which provides a window to a photomultiplier tube detector that measures the net scattering intensity of the phase fluctuations. The phase fluctuations are proportional to particle diffusivity (D), which is related to particle size according to the Stokes-Einstein equation: 24 Figure 1.6: The preparation and freeze fracture micrographs of liposomes. One method of preparing liposomes of a defined size is by extruding them at high pressures through polycarbonate filters of a defined pore size. Lipids are hydrated to form multilamellar vesicles (MLVs) that can be further processed by freezing and thawing them to generate frozen and thawed MLVs (FATMLVs). These liposomes can be downsized to form liposomes of a defined size range. Liposomes with diameters greater than 200nm are usually multilamellar but smaller diameter liposomes are usually unilamellar. Large unilamellar vesicles (LUVs) are generated when the liposome sample is extruded through polycarbonate filters with pore sizes less than 200nm. Freeze fracture micrographs courtesy of Kim Wong, Liposome Research Unit, U.B.C.). 26 D = kl7(67T,r)Rh) where k = Boltzmann's constant T = Temperature in degrees Kelvin r| = viscosity of the solvent (r| = 1.33 for water) Rh= hydrodynamic radius of the spherical particles The Nicomp particle sizer is equipped with software to compute a Gaussian distribution of particle sizes within the sample (Nicoli, McKenzie and Wu, 1991). 1.3.3 Stability of liposomes The physical stability of liposomes can in part be derived from the stability of its chemical components. Liposomal phospholipids can hydrolyze to free fatty acids and lysophospholipids (Kensil and Dennis, 1981 and Grit et al., 1993) which tend to form micellar structures rather than organized bilayer configurations (Reiss-Husson, 1967). The rate of hydrolysis of liposome dispersions at 60°C was analyzed by Grit and Crommelin (1991 and 1992) who characterized the hydrolysis kinetics of phospholipids as a function of pH, temperature, buffer concentration and ionic strength. The effect of changing the phospholipid head group, the charge of the head group (positive, neutral or negative) and the amount of cholesterol in the liposomal bilayer were also found to effect the rate of hydrolysis and therefore liposome stability. The physical properties and functional roles of lipids in liposomal bilayers suggest the stability of liposomes is influenced by lipid packing, geometric constraints of liposome size and phase transitions between gel (solid) and liquid crystalline (fluid) bilayer states with increasing temperature (Cullis, Fenske and Hope, 1996). Acid/base catalyzed hydrolysis of the ester linkages in the phospholipid molecule and oxidation of unsaturated fatty acid tails are 27 also mechanisms in the chemical degradation of liposome preparations. These disruptive processes influencing liposome integrity are minimized by storage at a neutral pH and at refrigeration temperature. In particular regard to size, one advantage of using LUVs over MLVs and SUVs is that LUVs have a longer circulation half-life (Senior et al, 1987). The reasoning behind these observations is that larger size vesicles are more susceptible to recognition by components of the immune system clearance by the spleen and liver. Oppositely, smaller size vesicles, <80nm in diameter, have large gaps between outer monolayer phospholipids where plasma proteins, apoproteins and opsonic factor a2-macroglobulin can insert. These molecules bound to the liposome target the vesicle for metabolism in the liver (Palatini, 1992). An intermediate-sized liposome, LUVs, therefore, is optimal in minimizing the effects of immune detection and protein insertion upon introduction into blood circulation. To further improve the physical stability of the bilayer, cholesterol is included in the lipid formulation. Cholesterol was found to preserve the vesicular integrity of liposomes (Guo et al, 1980). Cholesterol molecules allow for tighter packing of the lipid bilayer making it more rigid and compact and therefore less leaky. The presence of the primary plasma lipoprotein responsible for the exchange of lipidic material between liposomes and lipoproteins, high density lipoprotein (HDL), increases the clearance rate of liposomes in vivo (Scherphof et al, 1978 and Senior, 1987). The rigidity of liposomal membranes can also be increased by addition of sphingomyelin or increasing the percentage of saturated phosphatidylcholines allowing them to evade liver and spleen-specific opsonins. More recently, stealth liposomes incorporating monosialoganglioside (GM 1) (Allen, 1987) and polyethyleneglycol (PEG)-28 containing liposomes (Woodle and Lasic, 1992) have been used to escape complement-dependent phagocytosis by preventing insertion of proteins which target the liposome for elimination. Hence G M 1 and PEG enhance circulation lifetimes of liposomes primarily by sterically hindering the direct interaction of the bilayer with opsonins. 1.3.4 Role of cholesterol in model membrane systems The role of cholesterol in model membrane systems is to stabilize the bilayer and preserve the vesicular structure of liposomes. Cell membranes possess cholesterol and of the total lipid in most biological membranes, 45mole% is cholesterol, e.g. red blood cell. Because of the molecule's overall membrane stabilizing ability, it would be advantageous to have similar proportions in liposome delivery vehicles. The amount of cholesterol in the bilayer, however, can limit the incorporation of lipophilic molecules in the bilayer. At the molecular level, cholesterol inhibits gauche isomerization (rotation about a carbon-carbon bond) of acyl chains. The resulting decrease in flexibility of the acyl chain(s), has a condensing effect when placed in a monolayer of phospholipids. For example, on a Langmuir trough, for phospholipids mixed with cholesterol, a decrease in the surface area taken up per lipid molecule is observed. Cholesterol also acts like an impurity in the bilayer by inhibiting the transition from liquid crystalline to gel state which may explain its ability to decrease the permeability of the liposomal bilayer to entrapped solutes (Inuoe, 1974). 1.3.5 Passive targeting of liposomes When liposomes are introduced into circulation, they are taken up by the immune system. Specifically, liposomes are taken up by the reticuloendothelial system (RES) which is composed of mononuclear phagocytic cells such as circulating and fixed tissue 29 macrophages (Stuart, 1970). Rapid uptake of liposomes into RES tissues (Figure 1.7, Aschoff, 1924) such as the liver and spleen has been documented by Poste et al., 1983 and the addition of cholesterol into liposomal membranes reduces the rate of RES cell uptake (Patel et al., 1983). Because cells of the RES also migrate to sites of inflammation and disease, passive targeting to these sites using liposome drug delivery systems can be of potential therapeutic advantage. Accumulation of liposome-associated molecules has been shown at sites of inflammation (Williams et al., 1986), infection (Morgan et al., 1985) and in solid tumors (Profitt et al., 1983, Ogihara et al., 1986, Jones and Hudson, 1993 and Janknegt, 1996). Egress of liposomes from the circulation is dependent on the presence of pores or fenestral gaps between endothelial cells lining the blood vessel. In the spleen and the liver a discontinuous capillary (100-1,000 nm gaps) allows the extravasation of liposomes into these tissues (Palatini, 1992). These mechanisms also explain how liposomes are taken up into sites of inflammation and disease, which have leaky blood vessels due to the underlying condition. 30 Tissue C e l l Type B r a i n M i c r o g l i a l cells L y m p h Node Macrophages L u n g A v e o l a r Macrophages L i v e r Kupffer cells Spleen Spleen Macrophages Figure 1.7: The cells of the reticuloendothelial system (RES). The mononuclear phagocytic (reticuloendothelial) system is composed of fixed macro-phages (histiocytes) and wandering macrophages. In addition to the tissues listed above, fixed macrophages are found in the bronchial tissue, bone marrow and peritoneal cavity. 31 1.4 LIPOSOMAL DRUG DELIVERY 1.4.1 Drug association with liposomes Association of a drug with the liposome can be achieved several ways. The liposome has an aqueous core where a water soluble drug can be encapsulated passively or by remote loading procedures using a pH gradient (Boman, Mayer and Cullis, 1993). Alternatively, drugs can be chemically cross-linked to one of the membrane bilayer component lipids which serves as the anchor (Chua, Fan and Karush, 1985 and Hansen et al, 1994). Finally, a hydrophobic association between the drug and the liposome bilayer resulting from intercalation of the drug among fatty acid acyl chains of phospholipids may serve as a reservoir for the drug in liposomal drug delivery. The last mentioned case is the most straightforward in the delivery of cyclosporine (Fahr, Nimmerfall and Wenger, 1994). Delivery of cyclosporin A in liposomes has been attempted by Gibson et al, (1985), Vadiei et al, (1985), Wiedmann et al, (1990), Friese et al, (1991), Gorecki et al, (1991), Stuhne-Sekalec and Stanacev, (1991) and Fahr, Holz and Fricker (1995). These reports compare liposomal cyclosporine formulations and the currently used cremophor formulation. The effects of liposomal cyclosporine on the kidney were assessed in terms of glomerular filtration rate and nephrotoxicity. It appears that while certain liposomes may decrease cyclosporine glomerular filtration rate, the majority of cases report no benefit in delivering this drug in liposomes. Similarly, when the drug pharmacokinetics and biodistribution are compared for the cremophor versus liposomal formulations, there was no significant difference. Other hydrophobic drugs, such as taxol and hydrocortisone, have also been formulated for liposomal delivery (Kaledin et al, 1981 and Riondel et al, 1992) and improved delivery was not observed. The question 32 then is, why does this happen? In developing a liposomal cyclosporine formulation, this research will address this question by testing whether the drug remains associated with the lipid carrier: an assumption that was possibly overlooked by many of the previous reports in developing liposomal cyclosporine for clinical use. In previous studies, antibodies were attached to the liposomal bilayer via the head group of a phospholipid. Chemical attachment is possible through the cross-linking of antibodies to liposomes using coupling agents. These agents include streptavidin-avidin-modified linkers or covalent linkers such as SPDP-MPB or PDPH-MPB (Loughrey et al, 1993, Debs, Heath and Paphadjopoulos, 1987 and Ansell, Tardi and Buchkowsky, 1996). These studies have investigated the coupling of antibodies to liposomes for targeting purposes. The current research studies a novel coupling procedure that covalently attaches antibody to the liposome surface and whether the antibody remains associated with the lipid carrier in circulation. In this approach, the feasibility of attaching therapeutic antibodies to liposomes can be established. 1.4.2 Rationale for cyclosporine delivery in liposomes The high incidence of nephrotoxicity seen with cyclosporin A (CsA) administration in post-operative management of organ transplantation calls for more effective means of drug delivery. Liposomes can offer a safer method of drug delivery because drug bound to liposomes will escape glomerular filtration unlike molecules of MW < 20,000, which pass through. Although cyclosporine, MW = 1203, can conceivably pass through, its tight binding (>90%) to plasma albumin MW = 68,000 prevents most of its clearance via the kidneys (Rang and Dale, 1991). However, it is probably the unbound amounts of CsA that the kidneys are chronically exposed to in 33 management of organ transplantation that cause vasoconstriction of kidney capillaries, leading to a decrease in glomerular filtration rate (GFR) and overall kidney function. It would be advantageous then, to decrease the nephrotoxic effects of CsA by using a liposome carrier. The liposome carrier will not only prevent drug from entering through the glomerulus, but also provide accumulation of CsA at the graft site by extravasation and by reticuloendothelial cell uptake. This research will test whether liposomes do accumulate at sites of ongoing rejection and whether CsA can benefit from this accumulation. Animal models of organ transplant rejection can be prepared using microsurgical techniques described by Lee (1982). 1.4.3 Rationale for antibody delivery on liposomes The covalent attachment of antibodies to the liposome surface is a novel drug delivery method. In the past, antibodies have been attached to the liposome surface for the purpose of targeting liposomes to particular cell markers. The idea of chemically attaching a therapeutic antibody to the liposome surface stems from the problem of controlling the release of the hydrophobic immunosuppressant, cyclosporine, from the liposome. The characterization of the stability of the covalent attachment of an antibody will allow for the identification of a liposomal drug delivery system applicable to therapeutic antibodies. A conventional antibody, human IgG, was used in the thesis research as a model protein antibody to determine the feasibility of delivering antibodies on liposomes. A conventional antibody was chosen in order to test whether the fundamental approach in linking a molecule, such as an antibody, covalently to the liposome was viable in the in vivo delivery of this molecule with the liposome. 34 1.5 IMMUNOSUPPRESSION IN ORGAN TRANSPLANTATION 1.5.1 Organ Transplantation Immunology Surgical transplantation is the transfer of a portion of tissue from its original site to another part of the same individual or to another individual. Definitions of terms used to describe transplants include: graft - a piece of viable tissue transplanted to another part of the body or to the body of another individual; autograft - tissue grafted back on to the original donor; isograft - graft between syngeneic (of identical genetic constitution) individuals for example, identical twins or rats or mice of the same pure line strain; allograft (homograft) - graft between allogeneic (of the same species but different genetic constitution) individuals for example, fraternal (developing from different ova) twins, man to man, one rat strain to another or one mouse strain to another; and xenograft (heterograft) - graft between xenogeneic (of different species) individuals. One of the first milestones in immunology was in the investigation of specialized cells to mediate defense against microbial organisms at the cellular level by Russian zoologist, Elie Metchnikoff in 1882. His observation of cellular immunity was of the motile transparent starfish larvae that surrounded a rose thorn introduced amongst them. His observations progressed to the microscopic study of fungal spores being sequestered by metozoan Daphnia blood cells to mammalian leukocytes engaging in, what he termed, phagocytosis of microorganisms. He defined the existence of and named two circulating phagocytes, the microphages (polymorphonuclear leukocytes) and the macrophages. In 1894, Paul Ehrlich hypothesized that surface receptors (antibody) existed for complementary antigens that were bound by a lock and key fit. Stimulation of B cells causes the production of vast numbers of the antibody to be shed into circulation. The 35 idea was not fully accepted until template and selection theories were applied to clonal selection. This occurs on a cellular basis: an antigen triggering activation and clonal proliferation of a lymphocyte to produce an expanded population of antibody forming cells that produce antibody complementary to the original antigen. The antibodies produced give the host acquired immunity against the antigen. Innate immunity, where the immunity conferred is not intrinsically affected by prior exposure to the antigen, is not mediated by lymphocytes. In the proliferation of antibody-forming cells, a certain set of cells called memory cells are set aside in anticipation of a secondary exposure and response to subsequent challenge by the same antigen. As a result, the secondary response will be stronger and faster than the initial response to the antigen. Once antigen is bound by antibody, the crystallizable fragment (Fc) of the antibody elicits the classical complement pathway to help destroy the antigen. Specifically, at least two Fcy regions bind Clq to activate the classical complement pathway (Figure 1.8, adapted from Roitt, ed., 1994). The 4 peptide structure of antibodies (Figure 1.3) are divided into 3 categories: isotypes which differ in the F c heavy chain amino acid sequence specific to a species; allotypes which are allelic forms specific to an individual and idiotypes which differ in the variable and hypervariable regions (Fv) specific to each lg molecule. These antibodies aid in binding microbes and increase the macrophage phagocytosis of the foreign antigen. T (thymus-derived) cells can either be activated against foreign particles assisted by antibodies or bind directly to the foreign antigen. In the direct activation of T cells, the antigen presenting cell (APC) is the foreign cell (e.g. the grafted cell). 36 • Figure 1.8: The complement pathways. The components of the classical and alternative pathways responsible for opsonization (sequestration and destruction) of foreign particles from the body are illustrated. The classical pathway is dependent on antibody for activation where the antibody bound to the foreign molecule (e.g. microbe) stimulates (dash-dot line) components of the classical pathway giving rise to (solid line) molecular units with protease activity ( or • ) . The result is the amplification of the signal to split C3 using C4b2b generating C3b, an opsonin. The post-C3 terminal pathway (not illustrated here) in which complement molecules C5-C8 are activated leads to formation of a membrane attack complex (MAC) that causes phagocytosis or lysis of the foreign molecule. In the alternate or properdin pathway, factor B is responsible for binding C3b, factor D for splitting C3bB to form convertase C3bBb and properdin for stabilization of C3bBb. C3bBb converts C3 to C3a anaphylatoxin, which works in conjunction with C5a to bring about inflammatory processes such as, increased blood vessel permeability and chemotactic attraction of phagocytes. The molecules of the complement system are not specific for any one antigen and do not appear in greater amounts after immunization. 38 In the indirect activation of T cells the APC is the host's own cell (self) that presents a piece of the foreign cell, as antigen, in the context of the self major histocompatibility complex (MHC). T cells are divided into T cytotoxic (Tc) and T helper (Th) cells. The role of the T cell in the immune response is illustrated in Figure 1.9. T cell receptors (composed of ap or y5(early thymic) subunits) directly linked to the CD3 complex of 5 subunits (y6s+ 2^, C,r\ or r|2) (Figure 1.10) are responsible for binding antigen presented by major histocompatibility complexes (MHCs) on antigen presenting cells (APCs). The T cell receptor interacts with the APC's major histocompatibility complex (MHC). M H C s are divided according to the genes that encode for the complex. Class I MHC (MHC,) is a 44kDa transmembrane peptide associated with B 2 microglobulin. Class II MHC (MHC,,) is composed of transmembrane heterodimers and Class III MHC (MHC,,,) is composed of a heterogeneous group of complexes that are responsible for fixing complement to C3 convertases, activating heat shock proteins and triggering TNF production.. The MHC gene cluster is a polymorphic haplotype and is inherited en bloc according to Mendelian genetics. Class I MHCs are present on all cells and are responsible for signaling cytotoxic T cells while class II MHCs are associated with B cells and macrophages to attract and signal T helper cells (Figure 1.9). The rules governing the rejection of grafts were established by genetic experiments in mice that attributed rejection reactions to polymorphic molecules encoded by genes in the histocompatibility complex (MHC in humans and H-2 in mice)(Snell, Dausset and Nathenson, 1976). Cells or organs transplanted between individuals of the same inbred strain of mice (isografts) are never rejected. Cells or organs transplanted between individuals of different inbred strains of mice (allografts) are almost always rejected. 39 foreign antigens 1. phagocytosis 3. immunogen presentation CD4+ eg. B cell, dendritic cell or macrophage 2. antigen processing 5. IL-2 & cytokines 6. inflammation, TNF, IFNy monocyte and NK activation IL-2R 4. recognition by & activation of TH CD4+ ^ c e ^ activation IL-2,-4 & -5 immunogen 5.,/ra.wr 6. proliferation i IL-4 & -5 IFNy 6. differentiation 6. recognition & cytotoxin release immunogen MHCI .g. nucleated cells, lymphoid cell 7. antibody production 7. cell death 8. antibody i release O-Z r~L 9. antigen-antibody * complex formation —\Ss£s 10. inflammation 40 Figure 1.9: The immune response. Lymphocytes and the chemicals they produce are important in the immune response to foreign antigen. An antigen presenting cell (APC) that presents part of the antigen as a processed immunogen (in yellow) on its major histocompatibility complex (MHC) takes up the foreign antigen. The MHC associated immunogen is bound by the T cell receptor (TCR) of the T helper cell (TH) with the assistance of cell determinant (designation) molecule #4 (CD4, green cross). Interleukin-1 attracts TH cells to the APC. The T H cells are activated into proliferation and form cell clones that are specific for acting against the immunogen in the form of memory T helper cells (mTH). Concurrently, the T H cells, activated by lymphokines, can cause localized inflammation, activation of B cells that generate an antibody response to the immunogen and activation of cytotoxic T cells (Tc) via interleukin-2 release (IL-2). Release of tumor necrosis factor (TNF), interferon gamma (INPy) and activation of monocytes and natural killer (NK) cells accompany the inflammation. Cytotoxic destruction of immunogen infected cells via interaction of T c TCR and CD8 (pink hexagon) with MHC class I molecules expressing the immunogen. Memory T c (mTc) arising from this series of events allows for quicker reaction to secondary responses. B cells can proliferate into memory B cells for secondary responses or differentiate and produce large amounts of antibody that can bind the immunogen. 41 B a i l l M H C I I Antigen Presenting Cell in = 20-24 amino acids (hydrophobic and charged) 42 Figure 1.10: The ternary MHC-antigen-TCR-CD3 complex. The antigen (*) is being presented by the major histocompatibility complex (MHC) to the T cell which binds the antigen through its receptor (TCR). Many other molecules and receptors play an integral part in binding and recognition to allow for the downstream activation of the T cell against this antigen to occur (see Figure 1.11). Shown in this diagram are the components of the MHC-antigen-TCR-CD3 complex. 43 The offspring of a mating between two different inbred strains will never reject grafts from either parent but either parent will almost always reject a graft derived from the offspring. The latter occurs because of co-dominant expression of both parents' histocompatibility complex proteins in the offspring's graft. Antigen-antibody binding is based on spatial complementarity and is not covalent. The forces that allow for this binding include electrostatic, hydrogen bonding, hydrophobic and Van der Waals forces. The antigen epitope consists of amino acids that make contact with the paratope amino acid sequences of the antibody. These amino acid sequences are likely to be flexible and protruding discontinuous sequences (i.e., not linear) dependent on protein folding. The affinity or strength of binding to a single antibody combining site and the avidity or functional affinity plus the enhancing effect of multivalency are terms used to describe binding. The affinity and avidity of antibody binding and B cell mediated responses are pertinent to humoral immunity while T cell-mediated responses to antigen is pertinent to cellular immunity. As part of the general immune surveillance system in the body, the lymph nodes filter and screen lymph from body tissues and the spleen filters blood. Common to both lymph and blood is an intricate network of cells of reticular origin such as macrophages and lymphocytes (bone marrow stem cells, T cells and B cells). These cells provide communication among tissues as they circulate between blood and lymph via the thoracic duct and vice versa through the high-walled endothelium of post-capillary venules (HEVs). When they come into contact with antigen either through antigen presentation to the T cell receptor-CD3 complex (T cells) or via surface lg receptors (B cells), a cascade of cellular activation occurs. 44 In cellular immunity, the T cell receptor-CD3 complex link to the MHC of APCs is relatively weak and is aided by other molecules on the surfaces of the respective cells (Figure 1.11). The predominant molecules on T cells that increase the binding affinity include CD4 on T helper cells (TH) and CD8 on T cytotoxic cells (Tc) that bind onto parts of the MHC n and MHC„ respectively. The CD (cluster designation) molecules are also known to be specific markers of their respective cells as originally assigned by monoclonal antibodies. In order for T cells to become activated, TCR/MHC-peptide binding and ligation of CD28 to B7 or HSA to its complementary receptor are essential. Blockade of the CD28:B7-mediated co-stimulatory signal results in immunosuppression (Judge, Tang and Turka, 1996) and evidence to date supports selective immunomodulation by targeting the CD29 and very late antigen (VLA) on T cells (Morimoto, Sato and Tachibana, 1996). During activation of T cells, tyrosine kinase phosphorylation is initiated by CD45 phosphatase cleavage of phosphorylated tyrosine kinase to active dephosphorylated tyrosine kinase. CD3<^  and phospholipase C (PLC) is phosphorylated by tyrosine kinase (p59fyn) leading to release of Ca 2 + from intracellular stores and protein kinase C activation. Protein kinase C (PKC) stimulates transcription upregulating IL-2 expression, synthesis of soluble cytokines and their receptors. The other branch of the lymphocyte immune response and activation, humoral immunity, involves B cells. Their activation can be T cell independent where either binding of antigen by surface lg receptors leads to polyclonal activation or antigen binding leads to cross-linking of two or more surface lg receptors. A third T cell dependent route of B cell activation requires a T helper cell to stimulate antibody production in two steps: 45 46 47 Figure 1.11: Surface molecules aiding in MHC-antigen-TCR recognition. The key cell surface molecules assisting in MHC-antigen(Ag)-TCR-CD3 ternary complex formation are CD4 on helper T cells and CD8 on cytotoxic T cells which bind the antigen presenting cell's MHC n or MHC,, respectively. Other molecules play a role in enhancing the binding and concomitant activation. These molecules include intracellular adhesion molecules (ICAM), integrins (Int), lymphocyte function associated molecules (LFA), very late integrin antigens (VLA), vascular cell adhesion molecule (VCAM), heat stable antigens (HSA), cytotoxic T lymphocyte antigens (CTLA), potent co-stimulator B7 and other cell designation/determinant (CD) molecules. 48 T helper cell receptor binding to the MHCn-peptide presented on the B cell and ligation of CD40 (on the TH) to p39 (on the B cell). T helper cells can be further divided into TH 1 and T H 2 each with contrasting roles in terms of the effectors they produce. TH 1 plays a role in inflammation, macrophage activation, delayed sensitivity, and production of interleukin, IL-2,3 and interferon, IFNy. T H 2 helps B cells produce antibody and produces IL-3,4,5,6 and 10. Both TH species produce tumor necrosis factor, TNF (TH1 TNFi?and T H 2 TNFa) and granulocyte-monocyte colony stimulating factor (GM-CSF). In humoral immunity, antibody initiates a cascade of events known as the classical complement pathway resulting in the opsonization and breakdown of the antigen. In transplantation, the grafted cells provide a source of foreign antigen. The components of the immune response: lymphocytes, macrophages and antibody and complement play an integral role in the destruction of the graft. The roles lymphocytes, macrophages and, to a lesser extent, antibodies, play in the rejection process have been discussed above and the role antibody and complement plays is reviewed by Sanfilippo and Baldwin (1997). Briefly, a biological response is triggered when the Fc portions of two or more IgG antibodies become bound by Fc receptors (FcR) on effector cells. The Fc receptors are: FcyRI, FcyRII, FcyRIII-A and B and FcsRI. FcyRI (present on monocytes, macrophages and neutrophils) and FcsRI (present on mast cells) have the highest affinity for IgG and IgE immunoglobulin Fc, respectively. Once the receptors are cross-linked by antibodies, in a process called antibody-dependent cell cytotoxicity mediates the extracellular killing of target cells coated with the immunoglobulin. 49 Several immunosuppressive agents (Figure 1.12, adapted from Roitt, 1994) are usually used in combination to prevent organ rejection. This may, however, lead to adverse drug interactions (Mignat, 1997) that may cause patient morbidity or even rejection of the graft. These immunosuppressive agents act on two types of diseases associated with the organ rejection process. In solid organ transplantation, host-versus-graft disease is characterized by host leukocytes attacking the cells of the graft. Oppositely, graft-versws-host disease is characterized by T lymphocytes present in a graft (passenger leukocytes) recognizing and attacking host cells (e.g. bone marrow trans-plant). Gill and Wolf (1995) and Faustman (1995) review the immunobiology of cellular, tissue and organ transplantation. Briefly, these reviews focus on the major strategies in achieving long-term graft survival using immunosuppressive agents and newer approaches utilizing genetically engineered, immunologically modified donor organs and xenografts. Immunosuppressants, preferably acting by several mechanisms, should target T lymphocytes in and around the donor tissue that cause immunologic rejection. An understanding for why the immunologic rejection response causes transplant rejection beings at the molecular level. A single foreign MHC molecule can precipitate a strong and rapid host T cell response irrespective of whether it presents any bound peptide. The grafted cell expressing the foreign MHC is therefore an antigen present cell (APC). Alternatively, indirect presentation of alloantigen occurs when the grafted cell is phagocytosed and through the endosomal vesicular pathway, host MHCII expresses a piece of foreign peptide. Weaker and slower responses to polymorphic alloantigens other than foreign MHC molecules are classified as rejection responses to minor histocompatibility antigens. On a cellular level, this alloantigen recognition is character-50 'o on o £ o o o is H Figure 1.12: Immunosuppressive drugs. Drugs used in immunosuppressive therapy inhibit at various points in the cell cycle. Resting cells (G 0 phase), early growth phase (Gj), D N A synthesis (S), late growth phase (G 2), mitosis (M) and return to Go/i phases. Drugs can be used in combination as in conventional or sequential immunosuppressive therapy in the management of post-organ transplantation to optimize immunosuppression. Interleukin-2 (A, IL-2) and interleukin-2 receptor IL-2R). 52 ized using the mixed leukocyte reaction. This reaction relies on the expansion of CD4+ helper T cells and CD8+ cytotoxic T cells (CTLs) brought about by their detection of foreign (donor) B cells. Similarly, in grafted tissues, alloreactive CTLs directly lyse graft endothelial and parenchymal cells while helper T cells recruit and activate macrophages that injure the graft via delayed type hypersensitivity. Thereafter, alloantibodies produced by B cells bind the endothelium, activate the complement system and injure graft blood vessels. 1.5.2 Aims of Immunosuppression in Organ Transplantation Immunosuppressive therapy stresses the selective targeting of a cell subpopulation. The specific targeting of cells with cyclosporine mainly inhibits the release of lymphokines and keeps T cells in the resting phase of the cell cycle. The cellular and molecular mechanisms acting in allograft rejection include T cells that either release lymphokines and injure the endothelial cells or destroy grafted cells by cytotoxic means. The T-helper cell lymphocyte subpopulation is targeted by cyclosporine. Because of this specificity for T-helper lymphocytes, this drug is more efficacious and represents an improvement to earlier immunosuppressive drug regimens such as azathioprine, prednisone and antilymphocyte antibody treatments. Cyclosporine simply inhibits T-helper cells' functional ability to synthesize and release IL-2. As a result, the generation of cytotoxic and other effector T cells that lyse foreign cells is inhibited thereby prolonging organ graft survival and tolerance in the body and inflammatory responses and recruitment of effector cells are also inhibited. Continued administration of cyclosporine is necessary because the immunosuppressive effects of cyclosporine are reversible upon removal of this agent. 53 Immunosuppression by OKT3 relies on the physical binding of this antibody to the CD3 complex that is closely associated with the T cell receptor (TCR). The antigenic modulation caused by OKT3 binding to the CD3 complex results in a transient shut down of the T cell, leaving it susceptible to opsonization via the complement pathway. OKT3 has been shown to be more effective in decreasing the incidence of renal and hepatic graft rejection or delaying the initial incidence of rejection and equivalent to cyclosporine in the treatment of cardiac rejection. To summarize, while the molecular target for cyclosporine lies within the T cell, OKT3 targets the CD3 complex on the T cell surface and because each has a different molecular target, using both in combination would be advantageous in achieving graft survival. The alternative methods of immunosuppression in the management of autoimmune diseases and allograft rejection include the use of corticosteroids, cyclophosphamide, azathioprine, monoclonal antibodies and polyclonal antilymphocyte agents whose adverse effects are reviewed by Rossi et al., (1993). The uses of cyclosporine in autoimmune diseases provided lessons applicable to its use in transplantation (Bach, 1994). Azrin (1992) describes the use of antibodies in cardiac transplantation and clinical cardiology. Cyclosporine in combination with UV-B irradiation of bone marrow cells was also used in the prevention of GVH disease (Ohujekwe, Hardy and Oluwole, 1995). Briggs (1991) reviews current immunosuppressive therapy in renal transplantation and recommends triple therapy as well as the OKT3 regimen for steroid resistant renal allografts. In addition to anti-CD3 antibodies, anti-CD4 antibodies may have vital therapeutic roles in drug carrier systems (Olive and Mayas, 1993). The United Network for Organ Sharing (UNOS) procures 54 organs in the United States and co-ordinates transplants (Pierce et al., 1996) by matching donors to recipients in terms of HLA transplantation antigens. Their work enables not only the data recording of transplants in the United States, but also allows the efficient alert of hospital transplant teams when a donor has been identified. The most common type of grafting, allografting, (e.g. blood transfusions) was the focus of this introductory section on transplant immunology. Skin, pancreas or heart transplantation currently also fall into the allograft category where rejection as a result of immunological incompatibility can occur. Other forms of organ transplantation, xenogeneic transplantation and autogeneic transplantation where one's own tissue is grown up to produce a functional organ may become reality in the near future. The current mainstay in management of organ transplantation, however, remains to be drug intervention in the rejection process. In human organ transplantation, the major strategies to reduce graft immunogenecity are to: 1) minimize alloantigenic differences between donors and recipients by matching MHC tissue types (HLA-A to DR/Q alleles) and blood types; 2) eliminate passenger leukocytes; and 3) induce allograft tolerance using immunosuppressive agents against T cells, B cells and effector mononuclear phagocytes that produce cytokines. Vaux (1995) and Perico and Remuzzi (1997) discuss newer ways around rejection and future developments in this field. This thesis focuses on the pharmacological intervention in the immunological rejection process. 55 1.6 SUMMARY OF RESEARCH OBJECTIVES To produce a well-defined liposomal cyclosporine formulation and characterize this formulation in terms of drug incorporation, exchangeability and stability. To further characterize this liposomal cyclosporine formulation in terms of clearance and biodistribution in an animal and organ transplant model. To produce a well-defined liposomal antibody formulation and characterize its component stability in terms of whether the antibody remains associated to the liposome carrier in vivo. 1.7 RESEARCH HYPOTHESES Experiments in this research project tested the following hypotheses: Can cyclosporine be incorporated into liposomes? Can liposomes deliver incorporated cyclosporine in vivol Do liposomes preferentially accumulate at sites of ongoing tissue rejection? Do liposomes alter the biodistribution of liposomal cyclosporine? Does chemically coupled antibody remain associated with the liposome in vivol and Is it feasible to deliver immunosuppressants using liposomes? 56 CHAPTER 2 CHARACTERIZATION OF CYCLOSPORINE DRUG INCORPORATION The physical characteristics of cyclosporine and liposomes described in the introductory chapter were used to develop a method for drug incorporation. A convenient drug incorporation method utilizes ethanol to dissolve and deliver the drug into the liposomes. A description of how this was achieved for cyclosporine is described in this chapter. The effect of changing the amount of drug introduced to the liposome suspension and the effect of varying lipid composition in the liposome on overall drug incorporation are also investigated in this chapter. The main objective of these studies was to establish a well-defined set of liposomes for cyclosporine delivery. 2.1 INTRODUCTION A number of previous studies have examined the application of liposomes as carriers for the immunosuppressive agent cyclosporine (Freise et al, 1991, Venkataram et al, 1990 and Gilbert et al, 1993). These studies, however, have generated equivocal results, particularly with regard to the therapeutic properties of such systems. In the present work, the characterization of cyclosporine incorporation into well-defined liposomes, large unilamellar vesicles, have been examined along with the stability of drug association. Contrary to some earlier reports, only modest levels of cyclosporine were found to be accommodated in a liposomal membrane and the extent of drug incorporation was greatly reduced as the bilayer cholesterol content was increased. 57 While introduction of the immunosuppressive agent, cyclosporin A, has considerably improved patient survival following organ transplantation, serious adverse effects including both acute and chronic nephrotoxicity are associated with its use (Puschett et al, 1990). In addition, it has been suggested that failure to prevent rejection following lung, heart-lung or corneal transplantation may be the result of subtherapeutic drug levels in these organs following systemic administration (Dowling et al, 1990 and Tabbara, Gee and Alverez, 1989). These limitations have prompted a number of researchers to examine the application of liposomes as delivery vehicles for cyclosporine (Hseih et al, 1985, Freise et al, 1991, Venkataram et al, 1990, Gilbert et al, 1993 and Milani et al, 1993). While liposomes accumulate preferentially at sites of infection and inflammation and at tumor sites, in the absence of disease they tend not to accumulate in organs such as the heart and kidney (Ostro and Cullis, 1989, Morgan, Williams and Howard, 1985, Ogihara, Kojimi and Jay, 1986 and Gregoriadis, 1976). Administration of cyclosporine in a liposomal carrier, therefore, could potentially allow elevated drug levels to be achieved at sites of tissue rejection while reducing drug exposure to healthy organs. To date, however, studies comparing the efficacy of liposomal cyclosporine to the current cremophor formulation have been equivocal. While Hsieh and co-workers (1985) reported that liposomal cyclosporine might be less nephrotoxic than the cremophor preparation, Gibson and colleagues (1985) found no difference between these dosage forms. Similarly, while Vadiei et al, (1989b) found no reduction in glomerular filtration rate (GFR) in mice treated with cyclosporine incorporated into positively charged liposomes, a two-fold decrease in GFR was observed in animals treated with drug in negatively charged vesicles. Further, pharmacokinetics and biodistribution studies have 58 reported only very modest differences in blood clearance and organ distribution for cyclosporine incorporated into a liposomal carrier compared to the current formulation (Freise et al, 1991, Vadiei et al, 1989 and Vadiei et al, 1991b). These results are in sharp contrast to the dramatic changes in pharmacokinetics and biodistribution seen for other liposomal drugs including doxorubicin (Bally et al, 1990), amphotericin B (Lopez-Berestein, Rosenblum and Mehta, 1984) and vincristine (Mayer et al, 1993). It should be noted, however, that cyclosporine, unlike most of the other liposomal drugs studied, is hydrophobic and is believed to associate with the phospholipid hydrocarbon chains rather than being encapsulated in the aqueous core of the liposome (Wiedmann et al, 1990 and Stuhne-Sekalec and Stanacev, 1991a,b). The present study was undertaken, therefore, to characterize the incorporation of cyclosporine into well-defined liposomal systems, large unilamellar vesicles. 59 2.2 MATERIALS AND METHODS 2.2.1 Lipids and Chemicals l-Palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) and l-Palmitoyl-2-oleoyl-sn-glycero-3-[phospho-rac-(l-glycerol)], sodium salt (POPG) were obtained from Avanti Polar Lipids Inc. (Birmingham, AL, USA). Stearylamine (SA) and 5-Cholesten-3p-ol was obtained from Sigma Chemical Co. (St. Louis, MO, USA). L-cc-Dipalmitoyl-[dipalmitoyl-l-14C]-phosphatidycholine (1 4C-DPPC), L-a-Dipalmitoyl-[2-dipalmitoyl-9,10-3H(N)]-phosphatidycholine (3H-DPPC), and [mebmt-b-3H]cyclosporin A (3H-CsA) were purchased from Amersham Life Science (Little Chalfont, Buckinghamshire, England). The isolation of cyclosporin A was performed by taking the clinically used oral formulation of cyclosporin A, Sandimmune™ (Sandoz, Switzerland) and separating the components of the formulation by dry column flash chromatography (Harwood, 1985). The apparatus in Figure 2.1 fitted with a sinter of dimensions 40mm/50mm (diameter/length) containing 3ml of Sandimmune™ adsorbed on 25 grams silica gel 60, 230-400 Mesh (Merck, Germany). The components were eluted using 15ml aliquots of an ethyl acetate/hexane (Fisher Scientific, Canada) solvent mixture. Starting from a relatively non-polar mixture of ethyl acetate/hexane (0.05:14.95 v/v) and increasing the proportion of ethyl acetate (polar component) by 10% in each successive 15ml aliquot, CsA was eventually eluted with an ethyl acetate/hexane mixture of 13.85:1.15 v/v. The components in each fraction were visualized by running thin layer chromatography (TLC) on silica plates (Silica Gel 60 F-254, VWR Canada) using a running solvent of ethyl acetate/hexane (90:10 v/v). The CsA can be detected using ultraviolet light 60 1. CsA dissolved in hexane mixed with silica and rotoevaporate excess hexane off 2. CsA adsorbed to silica was transferred into glass sinter vial _ Ethyl Acetate 3. Dry column flash chromatography with mixtures of ethyl acetate and hexane run through silica gel Hexane vacuum 4. Fractions subject to TLC & UV analysis 5. CsA-rich fractions recrystallized from hexane 61 Figure 2.1: Isolation and purification of cyclosporin A from oral formulation. Cyclosporin A was isolated and purified according to the above procedures as outlined the Materials and Methods Section 2.2.1. if 62 (A,=254nm, Rf=0.35) and other components by development of the plates with ammonium molybdate solution (Fisher Scientific, Canada) (New, 1990). Fractions containing cyclosporin A are pooled and recrystallized from hexane. 2.2.2 Liposomes Preparation of Lipid Mixtures. Lipid mixtures, including radiolabeled lipids were co-solubilized in benzene/methanol (95:5 v/v), the mixture frozen in liquid nitrogen and then lyophilized (Virtis, Leybold Vacuum Products, USA) under high vacuum (60-100 mtorr) for a minimum of six hours. The lyophilized lipid mixture was hydrated with 150mM NaCl, 20mM HEPES, pH 7.4 buffer (HEPES buffered saline, HBS), followed by freezing in liquid nitrogen (5 min) and thawing (10 min at 37°C) for 5 cycles to maximize hydration. Large unilamellar vesicles were prepared by extrusion of MLVs ten times through two stacked lOOnm polycarbonate filters (Poretics Corp.) at 37°C using an Extruder (Lipex Biomembranes, Vancouver, BC, Canada) essentially as described by Hope et al, 1985. Vesicle diameters were determined by quasi-elastic light scattering using a submicron particle sizer (Nicomp Instruments model 270, Santa Barbara, CA, USA). In some samples, phosphate analysis was used to determine phospholipid concentration (Fiske and Subbarow, 1925). 2.2.3 Drug incorporation procedure Technique 1 and the effects of ethanol on drug incorporation The effect of varying the final ethanol content in the liposome suspension on efficiency of drug incorporation and vesicle size was investigated by dissolving POPC (326mg, 0.429 mmole) and CsA (13.0mg, 0.011 mmole) in 0.65ml of 95% ethanol 63 (BDH, Canada). One and one-half microliters of [mebmt-b-3H]cyclosporin A (3H-CsA) in ethanol (lmCi/ml, 10.7Ci/mmoleCsA) was added to yield a total volume of 1.0ml. Aliquots of 150ul taken from this mixture and added to 1.9ml of different ethanol/HEPES buffer solutions (see Results, Table 2.1) to generate liposomes (0.110uCi[3H-CsA]/ml, 0.805mM CsA, 31.4mM POPC). Each of the samples was extruded through 2 stacked lOOnm polycarbonate pore size filters using extrusion procedures as discussed in the preparation of large unilamellar vesicles above. Fifty microliter samples were counted before and after extrusion to determine incorporation efficiency and extruded samples are subject to Nicomp vesicle size analysis. The initial weight and mole percentages of CsA to phospholipid were 4.0 and 2.5, respectively. Technique 2 Drug incorporation into pre-formed liposomes LUVs composed of POPC (12.5 mg/ml phospholipid; 30uCi 14C-DPPC/mmole POPC) were preincubated at 37°C for 10 minutes. Cyclosporine in ethanol (lOmg/ml; 6mCi 3H-cyclosporine/mmole CsA) was then injected into the vesicle suspension while vortexing to achieve the required cyclosporineiphospholipid ratio. The mixture was then incubated at 37°C for another 5 minutes to allow drug incorporation. CsA-loaded liposomes were filtered through a 0.22um sterile filter to remove any unincorporated drug. Cyclosporin A was incorporated into liposomes in this manner for all subsequent studies described in this thesis except where noted. 2.2.4 Liquid scintillation radiolabel quantitation Cyclosporine incorporation was quantified by liquid scintillation counting of [mebmt-b-3H]cyclosporin A on a Beckman LS3801 instrument (Fullerton, CA, USA). Lipid quantitation was performed by either 1 4C(DPPC) liquid scintillation counting using 64 the Beckman instalment above or by standard phosphate assay procedures (Fiske and Subbarow, 1925). 2.2.5 Effect of cholesterol, anionic lipid POPG and storage temperature on CsA incorporation In these experiments, the effects of varying cholesterol content in the lipid bilayer, having POPG in the lipid bilayer and storing the liposomal cyclosporin A formulation on CsA incorporation were studied. The cholesterol content was varied between 0 and 45 mole% by co-lyophilizing cholesterol and POPC in benzene/methanol (95:5 v/v). One hundred microliters of CsA dissolved in ethanol (2.5LiCi[3HCsA]/ml, 25.3mg CsA/ml, 20.8mM CsA) was incorporated at 5 weight% or 3.2 mole% to lOOnm extruded vesicles of POPC/CHOL (~25mg phospholipid/ml or 32.9mM phospholipid) using technique 2 described above. The final ethanol content in all samples was 5%. The effect of including a negatively charged phospholipid, POPG at pH 7.4 on incorporation of CsA was examined similarly in combination with POPC or with both POPC and cholesterol. Again, the lipids were co-lyophilized and prepared by extrusion through lOOnm pore size filters before radiolabeled CsA was added. The stability of several liposomal cyclosporin A formulations at different temperatures was measured over 15 days by filtering the formulation through a 0.22 micron filter daily before determining the amount of radioactivity remaining in the sample. 2.2.6 Gel exclusion chromatography A radiolabeled liposomal cyclosporin A formulation (2.5 mole% CsA) was run on a 1.0 x 15 cm Sephadex G-50 gel column. The Sephadex beads made of cross-linked dextran and epichlorohydrin were dissolved (5 grams of Sephadex G-50, Sigma Chemical 65 Co., St. Louis, MO) in 100ml of HEPES buffered saline (HBS) for 2 hours at room temperature. The gel was degassed at 37°C and poured into a 1.0 x 15 cm column. Three-quarters of a milliliter of CsA-loaded liposomes (0.11 u;Ci[3H-CsA]/ml, 2.52mole% CsA, 31.4mM POPC) was placed at the top of the column and eluted with HBS in 26-1.5ml fractions. 2.2.7 Isopycnic density gradient A radiolabeled liposomal cyclosporin A prepared by injecting CsA (2.5 mole% CsA) into a pre-formed liposome formulation was run on a 0-7.5% Ficoll Type 400, (approximate molecular weight 400,000) (Sigma Chemical Co., St. Louis, MO) isopycnic density gradient. The gradient was prepared using a gradient maker (Hoefer Scientific Instruments, San Francisco, CA, model SG20L) in which 2 solutions, distilled water and Ficoll sucrose polymer dissolved in distilled water, in separate compartments, are mixed to form a continuous density gradient in 13mm x 10mm open-top Ultra-Clear™ plastic Beckman centrifuge tube. CsA-loaded liposomes (1ml) were loaded at the top of the gradient and the sample tubes centrifuged for 19 hours at 11 l,000g (30,000 rpm) at 20°C ± 1°C in a preparative ultracentrifuge (Beckman Model L2-65B, Palo Alto, CA). 66 2.3 RESULTS As noted above, cyclosporin A is a relatively hydrophobic molecule and examination of various techniques whereby it could be introduced into the hydrocarbon core of the lipid bilayer was carried out. While incorporation can be achieved by co-lyophilization of the drug:lipid mixture from organic solvents such as benzene:methanol (95:5 v/v) followed by hydration of the resulting anhydrous mixture, a more convenient technique wherein the drug was introduced into liposomes of defined size was developed. In this procedure, large unilamellar vesicles are prepared by extrusion through lOOnm pore size filters and then cyclosporine in a water-miscible solvent vector, ethanol, was injected into the vesicle suspension. Cyclosporine has a very low aqueous solubility and any drug not incorporated into the liposomal bilayer precipitates out of solution and can be efficiently removed by passing the LUVs through a 0.22 micron pore size filter. 2.3.1 Effect of ethanol on drug incorporation The effect of ethanol in destabilizing liposomal bilayers leading to leaky perturbed membranes has been well documented (Chen and Engel, 1990, Lohner (1991) and Barry and Gawrisch, 1995). Because ethanol was used in the dissolution of cyclosporin A, the effects of differing final ethanol concentrations on drug incorporation in the liposomal cyclosporin A formulation were evaluated. While the same amount of cyclosporin A was added to each liposome suspension, the volume of the injected vector was increased. The results are shown in Table 2.1 where the initial concentration of CsA was 4.0weight% or equivalent to 2.52mole% compared to that of phospholipid. 67 E e I •Si OBJ 4>| en s i £ | > - C N O 0 S ! W , >»! C i u< U ! es •ML o - 4 - -i CN oo 0 0 0 0 ooj r-~ 51 "Si SJ M CNJ as i oo «n H I CO 1 • — H r o ! cn I (N i <-*' f Bj I Of N S <=» I t i r o ; ^Di O o'j ©' ol o\l OX) E i w, C O ; l y j . 0 S ! O, C : (N O l Os I so | xrt j , - J | o i o ' l o'f - i i v> i O i v i ! <r> e (WD S 3 a o w a> S o to o a. U s-O 3 a CB o s a -g • o V o 1 © w t-H a, s o o a. XS B <D t= o o 1 ea "o .2 3 o w Jg .52 <u x< B a S « O c/j P. « O U 03 B c/i " 03 2 g o o a <U cu w §}" X H oo E i - ^ I f S i m l T t - l i o i v o ! Figure 2.2 illustrates the efficiency of cyclosporine incorporation into lOOnm vesicles composed of POPC as a function of the final ethanol concentration. Cyclosporine was initially added to the vesicle suspension at a molar ratio of 2.5% with respect to the phospholipid content and it can be seen that very good incorporation efficiencies are obtained when the ethanol content was maintained below about 10% (v/v). At higher ethanol concentrations, a progressive reduction in liposome-associated drug was observed. In all subsequent experiments, therefore, the final ethanol concentration was maintained at 5%. 2.3.2 Effect of varying cyclosporin A mole% and/or lipid composition on drug incorporation The amount of cyclosporine that could be stably incorporated into the liposomal bilayer was investigated by preparing liposomes extruded through lOOnm pore size filters. As the initial amount of cyclosporin A (mole%) added to the preparation was increased, the amount of cyclosporin A incorporated increases less than proportionately, thereby decreasing the efficiency of incorporation. The data in the first 3 entries of Table 2.2 where the amount of cyclosporin A added to the liposome was increased from 2.5 mole% to 5 mole% and the data obtained for identical lipid compositions of POPC/POPG vesicles with either 3.2 mole% or 5.0 mole% of cyclosporin A added confirm that there are decreasing returns in incorporation efficiencies when the cyclosporin A added was increased from 2.5mole% to 5.0 mole%. As a result, it was plausible to conclude that the lipid bilayer can only incorporate a finite amount of the drug and that the remaining unincorporated drug was easily separated by 0.22pm filtration. 69 Figure 2.2: Influence of ethanol on cyclosporine incorporation. Drug incorporation into POPC vesicles upon addition of 2.5 mole% cyclosporine was determined as a function of the final ethanol concentration. Incorporation was determined as described in Section 2.2.3. 70 .21 >»' « B , S I M i »i I «>! ol ! BI fli i Si "Si S f i l f oi ! O-i U : B S i s OJ E 1 B | Oj 81 si Oi I ON j oo t»-| i o x [ c x 00! O 00! Os O Oj O PH PH \ PH o l o i o PHI PH! PH CN 00 00 00 i 00 ON oil oi oi l O i O i O ! <N i O l <N O ! 2^  i o O i f?i IO I &< <OI ©I ml o o~ j <n i o i o j P<i PH! OIOI PH j PH I O O OjO j PHI OH so so ON _ VOl NO; vo <ni © j </ii cil cil ci © } © ! © O J O j O •ri! </•>! <n j © 1 o Ov u-i i © >ni © l o i o i OH ! OH i O l O S o l o ! PH( OH O i O J OH | OH a CO e es 60 S fc. "3 © a _o w e «s CQ fl *•© es I N o a e u a i «= .s © S-'l &s "Z H > 2.3.3 Effect of increasing cholesterol content in liposomes on incorporation Cholesterol reportedly has a condensing effect in liposomal bilayers (Stillwell et al, 1994 and Slotte et al, 1994). Physically, a molecule of cholesterol intercalates among phospholipids interacting with the acyl chains and positioning its B-OH moiety at the level of the phosphate moiety (lipid:medium interface) of the phospholipids. When introduced into an environment in which the lipid bilayer in its gel state, cholesterol increases the entropy and permeability of the bilayer. However, in a liquid crystalline environment, cholesterol decreases the entropy and permeability of the bilayer. Hence, the effect of cholesterol on the incorporation of cyclosporin A depends on the transition temperature of the component phospholipids. POPC and POPG have identical transition temperatures of -2°C. The overall negative charge on POPG was not expected to effect cyclosporin A incorporation which was neutral and quite hydrophobic at pH 7.4. Therefore, it can be predicted that changing the phospholipid content (ratio of POPC to POPG) would have little to no effect on drug incorporation while increasing the amount of cholesterol in the bilayer will limit the amount of cyclosporin A that can be stably incorporated. The effect on incorporation of cyclosporin A into liposomal membranes containing 0-22mole% cholesterol (POPC/POPG/CHOL) and 0-45mole% cholesterol (POPC/CHOL) agrees will the above predictions (Figure 2.3). 2.3.4 Stability of drug incorporation As discussed below, varying results have been reported regarding the amount of cyclosporine that can be incorporated into liposomes. Two techniques were therefore employed to confirm that in this system cyclosporine was stably incorporated into the lipid bilayer and was not simply in solution or present in the form of microcrystals 72 Bilayer Cholesterol Content (mole%) Figure 2.3: Influence of cholesterol on cyclosporine incorporation into POPC and POPCrPOPG vesicles. Vesicles of POPC, , and POPC:POPG (4:1), O, were prepared with varying cholesterol contents. Following addition of cyclosporine (3.2 mole %) in ethanol any unincorporated drug was removed by passage for the vesicles through a 0.22 micron filter and the eluted vesicles then assayed for cyclosporine and lipid. 73 dispersed with the vesicle suspension. First, cyclosporine-containing POPC vesicles were fractionated by gel exclusion chromatography on Sephadex G-50. Cyclosporine in the eluted fractions was quantitated using H-cyclosporine while phospholipid was assayed by phosphate analysis. As shown in Figure 2.4, phospholipid and cyclosporine co-elute in the void volume with no indication of any soluble cyclosporine eluting in the included volume. Further, drug-lipid ratios for vesicles eluted from the column are essentially identical to those of the initial sample. Additional confirmation that cyclosporine was incorporated into the liposomal bilayer was provided by isopycnic density gradient centrifugation experiments. Cyclosporine-containing POPC vesicles were layered onto continuous density gradients (0-7.5% Ficoll 400) and centrifuged at 11 l,000gav for 19 h at 20°C. This prolonged, high speed centrifugation allows all of the components in the sample to migrate down the gradient to their equilibrium density positions. Immediately following centrifugation the gradient was fractionated and phospholipid and cyclosporine quantified by liquid-scintillation counting of 1 4C-DPPC and 3H-cyclosporine, respectively. As shown in Figure 2.5, cyclosporine-containing vesicles exhibit co-migration of drug and lipid to a density intermediate between those of POPC vesicles alone and cyclosporine alone, which layers at the top of the gradient. This result confirms that for cyclosporine-loaded vesicles prepared by ethanol injection, the drug was tightly associated with the liposomal membrane. When 2.5 mole% cyclosporine was added to POPC vesicles, over 90% of the drug becomes vesicle-associated (Table 2.2). To determine whether higher membrane drug levels can be accommodated, 3.2 mole% and 5.0 mole% cyclosporine were added to 74 'o B u O 0 10 15 20 25 30 Fraction (ml) < u Figure 2.4: Sephadex G-50 gel permeation chromatography of cyclosporine-loaded POPC vesicles. Vesicles containing 2.5 mole% cyclosporine were eluted on a 1.0 x 15 cm column and fractions assayed for drug, —, and lipid, respectively. -, employing 3H-cyclosporine and 14C-DPPC, 75 0) o E i , O CL o CL 0.12 ^ E 0.08 3 0.04 o 2 4 6 8 10 12 14 Gradient Depth (ml) E jB o E O v o E O Q_ o Q_ 2 4 6 8 10 12 14 Gradient Depth (ml) 2 4 6 8 10 12 14 Gradient Depth (ml) 76 Figure 2.5: Isopycnic density gradient centrifugation of cyclosporine-loaded P O P C vesicles. Cyclosporine-loaded vesicles were layered onto a continuous density gradient (0-7.5% Ficoll 4000 and centrifuged at 111 ,000gav for 19 hours at 20°C. Gradient fractions were assayed for cyclosporine, —, and lipid, —, as described under Methods. 77 POPC vesicles and unincorporated drug subsequently removed by passage of the vesicles through a 0.22 micron fdter. Table 2.2 shows that as the initial drug level was increased the final vesicle drug content was raised but by a smaller increment and hence the efficiency of cyclosporine incorporation was reduced. This observation suggests that the lipid bilayer has a finite ability to accommodate cyclosporine and when the bilayer capacity was reached excess drug simply precipitates out of solution. Previous studies have reported that the level of cyclosporine incorporation into liposomes was strongly influenced by their lipid composition (Dowling et al, 1990 and Stuhne-Sekalec and Stanacev, 1991) and disrupts membrane architecture (Haynes et ai, 1985). This parameter also dictates to a large extent the stability and plasma clearance rates of liposomes following intravenous administration. The influence of acidic phospholipids and cholesterol on cyclosporine incorporation into large unilamellar vesicles was examined. Contrary to an earlier report by Stuhne-Sekalec and Stanacev (1991b), cyclosporine incorporation was not significantly enhanced when the acidic lipid, phosphatidylglycerol, was included in the liposomal membrane (Table 2.2). In contrast, when cholesterol was titrated into vesicles composed of either POPC or POPC:POPG (4:1), the observations here point to a rapid reduction in drug incorporation levels as a function of sterol content (Figure 2.3). For phospholipid bilayers in the liquid-crystalline state, cholesterol has a "condensing" effect resulting in an increase in lipid packing density (Ladbrooke and Chapman, 1969). Conversely, charge-charge repulsion tends to decrease lipid packing density in bilayers containing acidic phospholipids such as phosphatidylglycerol. To some extent, therefore, maximal levels of cyclosporine incorporation may simply be a reflection of lipid packing density and of the bilayer's 78 ability to incorporate drug without exposing hydrophobic domains to the aqueous medium. It should be noted that the maximal levels of liposomal cyclosporine incorporation described herein are similar to those reported by Stuhne-Sekalec and Stanacev (1991a) who employed gel permeation chromatography to separate their cyclosporine-containing liposomes from unincorporated drug. Other researchers, however, have reported studies on liposomal systems containing much higher molar ratios of cyclosporine (Hsieh et al, 1985, Freise et al, 1991, Gilbert et al, 1993 and Gorecki et al, 1991). In these studies, however, cyclosporine incorporation was either not directly measured or alternatively the liposomes were simply centrifuged and the resultant pellet assayed for drug. Any crystalline cyclosporine present, however, would likely have also pelleted resulting in an overestimation of drug incorporation. To determine whether initially incorporated cyclosporine remains vesicle-associated upon prolonged storage, drug-loaded liposomes of differing lipid composition were stored at 4°C. Any unincorporated drug was removed immediately following cyclosporine incorporation and then at regular intervals samples were refiltered to remove drug that had been subsequently lost from the bilayer. As shown in Figure 2.6, all of the vesicle compositions tested showed a small loss of cyclosporine over the first seven days but then achieved stable equilibrium levels. While part of the drug loss may reflect higher lipid packing densities at 4°C relative to the temperature at which drug incorporation was performed, it was found that similar losses occurred for samples stored at 25°C. The fact that stable drug levels are obtained on prolonged storage may indicate that "supersaturating" concentrations are initially achieved in the lipid bilayer during 79 Figure 2.6: Influence of vesicle storage on cyclosporine incorporation. Cyclosporine was loaded into POPC, • , POPC:POPG (4:1), , POPC:POPG:CHOL (4:1:1), or POPG:SA (9:1), • , vesicles and any unincorporated drug removed. The vesicles were then stored at 4°C and aliquots taken at 2, 4, 7 and 15 days. These were re-filtered to remove any cyclosporine lost from the vesicles and the sample then assayed for drug and lipid. 80 cyclosporine loading. The higher level of cyclosporine shown for POPC vesicles relative to POPC:POPG systems results from the fact that all samples were prepared at a constant POPC concentration and hence the cyclosporine to total phospholipid ratios are different. These experiments were repeated in order to determine the effect on incorporation of cyclosporin A in the situation where the cyclosporine to total phospholipid ratios are kept constant while the phospholipid type was changed. The results here demonstrate insignificant differences in cyclosporine incorporation when POPG anionic phospholipid was substituted for POPC neutral phospholipid. Overall these results demonstrate that stable cyclosporine incorporation levels can be obtained for a variety of liposomal systems. 81 2.4 DISCUSSION The present study has important implications with respect to the potential clinical application of liposomes as a delivery system for cyclosporine. These relate to both the influence of lipid composition on drug incorporation and the rate of interbilayer drug exchange. Below, these issues are discussed in turn. The location of cyclosporine in the liposomal membrane has been probed using a variety of biophysical techniques. Employing perdeuterated dipalmitoylphosphatidyl-choline, Wiedmann and colleagues examined the effect of drug incorporation on lipid order as a function of temperature (Wiedmann et al, 1990). In addition, from the depaked 2 H-NMR spectra they were able to resolve drug-induced changes in order parameter for each carbon position down the phospholipid acyl chains. At temperature below the phospholipid gel to liquid-crystalline transition temperature (Tc), cyclosporine was observed to decrease acyl chain ordering, while conversely above T c drug-containing systems showed an increased order parameter. Employing nitroxide spin labels located at different positions down the acyl chains, Stahne-Sekalec and Stanacev similarly showed a perturbation in lipid motional freedom upon introduction of cyclosporine (Stuhne-Sekalec and Stanacev, 1991b). These studies and others indicating a direct perturbation of acyl chain packing are consistent with cyclosporine being oriented in the hydrophobic core of the phospholipid bilayer. The present work clearly demonstrates, however, that liposomes have a fairly modest capacity to accommodate this relatively large hydrophobic molecule with the level of insertion being strongly influenced by lipid composition. For example, while liposomal systems composed of an equimolar POPC/POPG mixture can accommodate more than 3 mole% cyclosporine, the level of 82 drug incorporation was rapidly reduced as cholesterol was titrated into the membrane. This observation was of particular interest given the importance of cholesterol in stabilizing liposomes in blood or plasma. Several research groups have demonstrated that vesicles composed of phospholipid alone, rapidly fragment following intravenous administration with consequent loss of water-soluble drugs entrapped in their aqueous interior. The constituent lipids are then assimilated into lipoproteins or other blood components (Kirby, Clarke, and Gregoriadis, 1980 and Damen, Regts and Scherphof, 1981). This process appears to be triggered by insertion of serum proteins into the lipid bilayer and can be prevented by including cholesterol in the liposomal membrane (Kirby, Clarke, and Gregoriadis, 1980). Such cholesterol-containing systems can exhibit extended blood residency times with good retention of their aqueous contents (Mayer et al, 1989 and Mayer et al, 1993). In the case of liposomal cyclosporine, however, a compromise would need to be made between the maximum level of drug incorporation that could be achieved for a given bilayer sterol content and the optimum cholesterol ratio needed to ensure vesicle stability in serum. As indicated earlier, it was anticipated that even in the absence of any active targeting information, liposomes would tend to accumulate at sites of organ rejection due to the involvement of lymphocytes and phagocytic cells in this process (Ostro and Cullis, 1989). While the use of liposomal carriers, therefore, represents a potential mechanism whereby enhanced levels of cyclosporine could be achieved precisely where the drug was needed, this requires that the drug remains liposome-associated until the carrier reaches this site. 83 CHAPTER 3 MOLECULE EXCHANGE CHARACTERISTICS OF CYCLOSPORINE FROM LIPOSOMAL BILAYERS: IN VITRO STUDIES In developing a well-defined liposomal formulation, the rate and degree to which cyclosporine is lost from its carrier was investigated. Although cyclosporine was well incorporated into the lipid carriers in the previous chapter, the system in which it was tested did not allow for exchange into hydrophobic depots. The model systems in which liposomal cyclosporine are tested in the following research accounts for the presence of these factors. Drug release from the liposome would depend on the strength of the hydrophobic interaction between cyclosporine and lipids. The rate and extent of cyclosporine exchange from neutral liposomes to negatively charged liposomes initially devoid of cyclosporine were tested. The two liposomal populations were separated using anion exchange chromatography. Upon incubation and subsequent separation, exchange of cyclosporine was found to be rapid (within minutes) and equally distributed according to the lipid content of each liposomal population. The rate and extent of exchange was not altered with changes in liposome cholesterol content or lamellarity of the liposomes. 3.1 INTRODUCTION Having firmly established that cyclosporine can be stably associated with a defined liposomal delivery vehicle, the following in vitro experiments set out to explore the rate and degree of drug transfer between different liposomal cyclosporine formulations. The stability of drug association was evaluated with particular regard to interbilayer exchange when liposomal cyclosporine is placed in an environment where 84 other model membranes are present, specifically other liposomes. Non-cyclosporine-loaded liposomes offer a lipid sink for the hydrophobic drug to partition into in contrast to the inhospitable aqueous buffer (El Tayar et al, 1993) in which these cyclosporine-loaded liposomes are stored. 85 3.2 MATERIALS AND METHODS 3.2.1 Lipids and Chemicals l-Palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) and l-Palmitoyl-2-oleoyl-sn-glycero-3-[phospho-rac-(l-glycerol)], sodium salt (POPG) were obtained from Avanti Polar Lipids Inc. (Birmingham, AL, USA). 5-Cholesten-3P-ol was obtained from Sigma Chemical Co. (St. Louis, MO, USA). L-a-Dipalmitoyl-[dipalmitoyl-l-14C]-phosphatidycholine (14C-DPPC), L-a-Dipalmitoyl-[2-dipalmitoyl-9,10-3H(N)]-phosphatidycholine (3H-DPPC), and [mebmt-b-3H]cyclosporin A (3H-CsA) were purchased from Amersham Life Science (Little Chalfont, Buckinghamshire, England). DEAE-Sephacel (40-160um, 100-140meq per ml gel volume) was purchased from Sigma Chemical Co., St. Louis, MO, USA. The gel was washed with 5mM NaCl, 5mM HEPES, 200mM glucose, pH 7.4 to remove ethanol. 3.2.2 Liposome Preparation Preparation of Lipid Mixtures. Lipid mixtures, including radiolabeled lipids were co-solubilized in benzene/mefhanol (95:5 v/v), the mixture frozen in liquid nitrogen and then lyophilized (Virtis, Leybold Vacuum Products, USA) under high vacuum (60-100 mtorr) for a minimum of six hours. Preparation of Large Multilamellar Vesicles (MLVs) and Large Unilamellar Vesicles (LUVs). The lyophilized lipid mixture was hydrated with 150mM NaCl, 20mM HEPES, pH 7.4 buffer (HEPES buffered saline, HBS), followed by freezing in liquid nitrogen (5 min) and thawing (10 min at 37°C) for 5 cycles to maximize hydration. Large unilamellar vesicles were prepared by extrusion of MLVs ten times through two stacked lOOnm polycarbonate filters (Poretics Corp.) at 37°C using an Extruder (Lipex 86 Biomembranes, Vancouver, BC, Canada) essentially as described by Hope et al, 1985. In some experiments (see Results), larger vesicles were produced by extrusion as above through 600nm polycarbonate filters. Vesicle diameters were determined by quasi-elastic light scattering using a submicron particle sizer (Nicomp Instruments model 270, Santa Barbara, CA, USA). In some samples, phosphate analysis was used to determine phospholipid concentration (Fiske, C M . and Subbarow, Y., 1925). 3.2.3 Incorporation of Cyclosporine into Vesicles Generally, LUVs composed of POPC (12.5 mg/ml phospholipid; 30uCi 1 4 C-DPPC/mmole POPC) were preincubated at 37°C for 10 mins. Cyclosporine in ethanol (lOmg/ml; 6mCi 3H-cyclosporine/mmole CsA) was then injected into the vesicle suspension with vortexing to achieve the required cyclosporine:phospholipid ratio. The mixture was then incubated at 37°C for another 5 minutes to allow drug incorporation. CsA-loaded liposomes were filtered through a 0.22pm sterile filter to remove any unincorporated drug (see Results). 3.2.4 Anion Exchange Chromatography Interbilayer Exchange of Cyclosporine. In these experiments, drug exchange was followed from cyclosporine-loaded POPC "donor" vesicles to POPC:POPG (1:1) "acceptor vesicles" (Madden et al, 1988). Cyclosporine was loaded into the donors as described above including 3H-cyclosporine and 1 4C-DPPC to allow recovery of drug and lipid to be followed. Donors and acceptors were incubated at 25°C at final lipid concentrations of 1.25 mg/ml and 6.25 mg/ml, respectively (ratio 1:5). At 5 minutes, 30 minutes, 1 hour, 2 hours and 4 hours post-mixing, lOOpl of the sample was removed and the two vesicle populations separated by passage through 2.0ml of DEAE-Sephacel anion 87 exchange resin equilibrated with 5mM NaCl, 5mM HEPES, 200mM glucose, pH 7.4, buffer packed in 10ml Poly-Prep® chromatography columns. Donor vesicles were eluted with 3ml of 77.5mM NaCl, 55mM glucose, 5mM HEPES, pH 7.4. The acceptor vesicles were eluted with 3ml of 1.0M NaCl, 20mM HEPES, pH 7.4. Aliquots of the eluted fractions were then counted for 3H-cyclosporine and 1 4C-DPPC using a dual label program. 3.2.5 Determination of vesicle lamellarity using 3 1 P-NMR 3 ^-Nuclear magnetic resonance (NMR) can be used to quantitate the relative amounts of phosphorus or phospholipid (one phosphorus atom per phospholipid molecule) in a liposome sample with respect to the amounts within the outermost monolayer of each liposome in the liposome suspension. 3 1 P-NMR spectra were recorded with broad band proton decoupling at 81 MHz using a Bruker MSL200 NMR spectrometer equipped with a Bruker variable temperature control unit accurate to ±0.1°C. POPC/Cholesterol (19:1) multilamellar vesicles (12.5mgPOPC/ml) extruded through 600nm polycarbonate filters were analyzed at 20°C, a 50kHz sweep width, 1.5 second interpulse delay, and a pulse width of 4.7p.s. The free induction decay (FID) was recorded for 2000 transients and transformed with 50 Hz line-broadening. Ten micro-liters of 1.0M MnCl 2 was added to the sample and the free induction decay (FJD) was recorded again over 2000 transients with subsequent line broadening. Integration of the spectra supplied information on the proportion of phospholipid in the outermost monolayer of the vesicle sample. 88 3.3 R E S U L T S The efficiency with which liposomes can selectively deliver cyclosporine to potential sites of tissue rejection will depend on the extent to which the drug remains vesicle-associated in the period between intravenous administration of the liposomes and their deposition at any potential sites of rejection in the transplanted organ. In turn, this will depend on the rate of exchange of cyclosporine between the vesicle bilayer and other membranes. The kinetics of such interbilayer drug exchange were therefore determined in experiments in which cyclosporine-containing vesicles composed of the zwitterionic lipid, POPC ("donor vesicles") were incubated with initially drug-free vesicles composed of POPC and the acidic phospholipid, POPG ("acceptor vesicles"). As described under Materials and Methods, the presence of POPG in the acceptor vesicles imparts a net negative surface charge that should allow mixtures of the two vesicle populations to be separated by ion exchange chromatography. That complete resolution of POPC vesicles and POPC:POPG vesicles could be achieved by DEAE-Sephacel chromatography was first confirmed in control experiments. Uncharged POPC vesicles were labeled with the non-exchangeable lipid 3H-DPPC while POPC:POPG systems were labeled with , 4 C-DPPC. The elution profile of a mixture of these two vesicle populations is shown in Figure 3.1. In the presence of a relatively low salt medium, only POPC vesicles are eluted from the column. POPC:POPG vesicles could then be recovered by washing the column in high salt. Additional experiments confirmed that separation of these two vesicle populations was not compromised when they contained cyclosporine or cholesterol (results not shown). 89 0 2 4 6 8 10 12 14 16 18 20 22 Elution Fraction Figure 3.1: Separation of POPC, — , and POPC:POPG (1:1), —, vesicles on DEAE-Sephacel. As described in Section 3.2.4, control experiments were performed to confirm that these two vesicle populations could be completely resolved on the basis of their surface charge. Fractions eluted from the column were assayed for 1 4C-DPPC (donor vesicles) and 3 H-DPPC (acceptor vesicles) by liquid scintillation counting. 90 3.3.1 Interbilayer exchange and the effect of cholesterol content on interbilayer exchange in vitro The kinetics of interbilayer drug flux were then examined by incubating cyclosporine-loaded POPC vesicles with a five-fold excess of POPC:POPG acceptors. At various times, aliquots of this mixture were separated by DEAE-chromatography and drug levels in the two liposome populations determined using 3H-cyclosporine. Controls were also run in which cyclosporine-loaded donor vesicles were incubated in the absence of acceptors and then passed down DEAE columns. Figure 3.2 shows the results of such an exchange study for vesicle systems containing different cholesterol levels (molar ratio, phospholipidxholesterol of 19:1 and 4:1). In all experiments, the cholesterol content of donor and acceptor systems was the same. Donor vesicles comprised of POPC:CHOL 19:1 were incubated in the absence of acceptor systems show a small reduction in cyclosporine content over the four hour incubation period consistent with the stability data shown in Figure 2.6. At higher cholesterol content (4:1), cyclosporine loss from donor vesicles in buffer alone was much more significant (approximately 50%) and, as suggested above, may reflect an initial "supersaturation" of the vesicle membrane during drug loading. Regardless of their cholesterol content, however, when donor vesicles are incubated with a five-fold excess of acceptors, a dramatic reduction in cyclosporine content was observed. As shown in Figure 3.2, even at the earliest time point (5 min) cyclosporine levels in the donor systems are much less than 20% of initial values and show little further change over the remaining time-course. If we assume that cyclosporine equilibrates equally between both vesicle populations, and allow for drug 91 Figure 3.2: Cyclosporine exchange from donor to acceptor vesicles. Cyclosporine loss from POPC:CHOL (19:1) vesicles incubated in the absence, O, or presence, • , of POPCPOPG acceptor vesicles and from POPC:CHOL (4:1) vesicles incubated in the absence, O , or presence, • , of acceptors are shown. 92 losses seen in control samples, it would be expected that donor drug levels would decrease to between about 8 and 13% of initial values. This was in good agreement with the data shown in Figure 3.2. The cyclosporine exchange data presented above have two important implications. First, intermembrane exchange was clearly very rapid, being essentially complete within less than five minutes. Second, all of the cyclosporine initially present in the donor vesicles was able to participate in the exchange process i.e. even drug associated with the inner monolayer was able to migrate to the outer leaflet and then transfer to acceptor membranes. This behavior can be contrasted to that of another hydrophobic compound, cholesterol. While equilibration of this sterol between two vesicle populations can occur via exchange, the half-time for this process was several hours (Nakagawa, Inoue and Nojima, 1979 and Backer and Dawidowicz, 1979). In addition, at least for some systems, flip-flop may represent a rate limiting step (Rottem, Shinar and Bittmann, 1981). 3.3.2 Effect of multilamellarity on interbilayer exchange in vitro To determine whether cyclosporine efflux was slower when the donor vesicles contain more than one bilayer, liposomes of POPC:CHOL (19:1) were prepared by extrusion through 600nm pore size filters. This resulted in formation of vesicles with a mean diameter of about 370nm. It has previously been reported that as the size of extruded liposomes was increased, the proportion of multilamellar systems present also increases (Mayer, Hope and Cullis, 1985). Vesicle lamellarity was determined by 3 , P-NMR spectroscopy employing the broadening agent, manganese. This paramagnetic ion broadens beyond detection the signal from any phospholipid headgroup to which it has access; it is not membrane permeable however, and therefore, only phospholipids facing 93 the external medium are affected. In the case of unilamellar vesicles, therefore, manganese should reduce signal intensity by 50%. Figure 3.3 shows the 3 1P-NMR spectra obtained for 370nm vesicles before and after addition of ImM MnCl 2 . Integration of these spectra showed a loss of about 28% signal intensity upon manganese addition, indicating that, on average, these liposomes comprise two concentric lamellae. Efflux of cyclosporine from these systems upon incubation with a five-fold excess of acceptor vesicles is shown in Figure 3.4. Again rapid drug loss was observed, with equilibration between the two vesicle populations being achieved within the earliest time point (5 min). Furthermore, as with the unilamellar vesicles examined earlier, the level of cyclosporine remaining in the donor vesicles was essentially that predicted based on an equal drug distribution between both vesicle populations. 94 Figure 3.3: Influence of manganese on the 3 1 P - N M R spectra of 400nm POPC vesicles. The spectra shown are from POPC vesicles before (left) and after (right) addition of the line broadening agent, MnCl 2 . This ion has access only to phospholipid head groups in the outermost monolayer and from the integrated signal intensities for the two spectra an estimation of average vesicle lamellarity can be obtained. 95 o 0 I 1 1 1 L 0 1 2 3 4 Time (hrs) Figure 3.4: Influence of vesicle lamellarity on cyclosporine exchange. Cyclosporine loss from 400nm POPC:CHOL (19:1) vesicles incubated in the absence, or presence, • , of acceptor vesicles was shown. 96 3.4 DISCUSSION The research presented in this chapter indicates that cyclosporine was able to exchange rapidly between different vesicles. Cyclosporine "flip-flop" between the inner and outer leaflet of donor vesicles was not rate limiting within the time frame of our experimental system and equilibration of the entire drug pool appears to occur within five minutes. Even when the donor population contains multilamellar vesicles, necessitating that cyclosporine transfer from one bilayer to another before being available for exchange into acceptor liposomes, there was no measurable delay in achieving complete drug equilibration. Conformational changes in cyclosporin A have been studied for the molecule in polar and apolar media (El Tayar et a l , 1993). These changes may be operative during the rapid redistribution of the drug between the liposome populations investigated in this chapter. If such exchange occurred between the liposomal carrier and cellular membranes or lipoproteins following intravenous administration, then enhanced drug delivery to tissue rejection sites might not be possible. This problem was examined in the following chapter, which directly compares the pharmacokinetics and tissue distribution of cyclosporine and its liposomal carrier. To summarize, then, in studying the drug exchange kinetics of cyclosporine from liposomes, it has been demonstrated in this chapter that cyclosporine, despite its hydrophobic character, can rapidly exchange between vesicles. This raises the possibility that, following intravenous administration, drug migration into blood components might negate the potential benefits arising from liposomal delivery. 97 C H A P T E R 4 C O M P A R I S O N O F C Y C L O S P O R I N E A N D LIPID C A R R I E R P H A R M A C O K I N E T I C S AND BIODISTRIBUTION In the preceding chapters, cyclosporine incorporation into well-defined liposomes, was characterized. This study demonstrated that only modest drug levels could be accommodated within the membrane, particularly for cholesterol-containing liposomes, and that rapid drug exchange could occur between vesicles. This raised the possibility that following intravenous administration drug migration to other blood components could negate the potential benefits arising from liposomal delivery. Therefore, examination of the pharmacokinetics and biodistribution of both cyclosporine and its liposomal carrier was necessary. It was shown that while liposomes are slowly cleared from the blood, redistribution of cyclosporine occurs much more rapidly. This liposomal loss of cyclosporine when injected into blood resulted from drug migration to the lipoproteins and, to a lesser extent, the erythrocytes. As a result, while liposomes preferentially accumulate in organs of the reticuloendothelial system following intravenous administration, tissue cyclosporine levels, in general, do not reflect the distribution profile obtained for the liposomal carrier. 98 4.1 INTRODUCTION This chapter examined the fate of liposomal cyclosporine in mice. Based on the earlier experiments in this thesis, it was predicted that the drug would rapidly dissociate from the carrier. Of particular interest in this mouse model, then, was where the drug would compartmentalize or accumulate once injected into circulation. A rapid drug flux from the liposome would probably be represented in lipoprotein sequestration of the drug. The study of the redistribution of liposomal cyclosporine into plasma lipoproteins would provide valuable information into the initial exchange of the drug upon entry into circulation. The liposomal cyclosporine formulation developed in this thesis was compared to a cremophor formulation of cyclosporine in terms of plasma elimination characteristics and tissue biodistribution. Transplantation of the organs of the reticuloendothelial system could potentially benefit from localized cyclosporine delivery with liposomes. Due to the action of macrophages and other components of the immune system, sites of inflammation and disease would also benefit from localized drug delivery with liposomes. In particular, a graft normally targeted by the immune system may passively attract liposomes. A second advantage of associating cyclosporine with liposomes was that cyclosporine would be less likely to gain access to the kidney where it has been shown to be nephrotoxic. A liposomal formulation of cyclosporine may be less toxic than the cremophor formulation which when studied gave equivocal results for cremophor vehicle nephrotoxicity. Coupled with the already adverse effects of cyclosporine on the kidney, a change to a safer liposomal vehicle that achieves equivalent or better cyclosporine potency would be 99 welcomed. The research presented in this chapter addresses the behaviour of the liposomal cyclosporine formulation developed in this thesis in a live model in which the drug and the drug carrier are followed using separate markers (radiolabels). 100 4.2 MATERIALS AND METHODS 4.2.1 Lipids and Chemicals l-Palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) was obtained from Avanti Polar Lipids Inc. (Birmingham, AL, USA). 5-Cholesten-3(3-ol was obtained from Sigma Chemical Co. (St. Louis, MO, USA). Cholesteryl hexadecyl ether [Cholesteryl-4-1 4C] ( I 4C-CHE) (NEN Research Products, Boston, Massachusetts, USA) and [mebmt-b-3H]cyclosporine A were purchased from Amersham Life Science (Little Chalfont, Buckinghamshire, England). 4.2.2 Drug and Lipid formulation Lyophilization of Lipid Mixtures. Mixtures of POPC, cholesterol and the non-exchangeable, non-metabolisable lipid marker, 14C-cholesterol hexadecyl ether ( 1 4C-CHE) (Dersen, Morselt and Scherphof, 1987), were prepared by colyophilization from benzene:methanol (95:5 v/v) (Virtis, Leybold Vacuum Products, USA) under high vacuum (60-100 mtorr) for a minimum of six hours. Preparation of Liposomes. Liposomes are prepared by hydrating the lyophilized lipid mixture with 150mM NaCl, 20mM HEPES, pH 7.4 (HEPES buffered saline, HBS) and then freezing in liquid nitrogen (5 min) and thawing (10 min at 37°C) for 5 cycles to maximize lipid hydration. Large unilamellar vesicles (LUVs) were prepared essentially as described by Hope et al, 1985. The M L V suspension was extruded ten times at 37°C through two stacked lOOnm polycarbonate filters (Poretics Corp., Livermore, CA) using an Extruder (Lipex Biomembranes, Vancouver, BC, Canada). Incorporation of Cyclosporine. POPC:CHOL (5:1) vesicles (12.5mg/ml; 170mCi 14C-CHE/mmole POPC) were incubated at 37°C. Cyclosporine dissolved in ethanol 101 (7.8mg/ml; 96mCi/mmole CsA) was then added to the vesicle suspension to give 2.5mole% cyclosporine. After incubation at 37°C for 5 minutes any unincorporated drug was removed by passage of the vesicles through a 0.22 micron filter. Cyclosporine in Cremophor. Twenty-eight microliters of [3H-CsA] (lmCi/ml) were added to 6.7ml of IV. CsA (5mg/ml, Sandimmune®, Sandoz Inc., Dorval, Quebec, Canada) and diluted with 3ml of saline to generate a O.llmgCsA/ml, 9.3mCi/ml sample of [3H-CsA]. 4.2.3 Mouse tail vein injections Plasma Clearance and Biodistribution Studies. Two hundred microliter tail vein injections were performed on 22 gram female CD1 mice (B.C. Cancer Agency, Vancouver, B.C., Canada) using either liposomal cyclosporine, "free" cyclosporine (cremophor formulation) or "empty" liposomes. 4.2.4 Blood and tissue collection Three time points were taken and three mice were sacrificed at each time point. At fifteen minutes, one hour, and four hours mice were sacrificed in carbon dioxide. Cardiac puncture to collect blood (350-500ul) was performed and the blood was subsequently spun down in ethylenediamine-tetraacetic acid (EDTA)-containing microtainers to obtain serum. Following cervical dislocation, the organs (liver, spleen, one kidney, both lungs and heart) were collected washed in ice-cold saline, blotted to remove excess fluid and then weighed. 4.2.5 Blood and tissue analysis Cyclosporine and liposome quantitation in plasma. Plasma levels of drug and/or 102 liposomes were determined by taking a 50ul volume from each plasma sample, adding 5ml of scintillation cocktail (Ultima Gold, Packard Instrument B.V., Groningen, The Netherlands) and determining the amount of 3 H and/or 1 4 C by liquid scintillation counting using a dual label program (Beckman, Fullerton, CA, USA). Cyclosporine and liposome quantitation in tissues. Tissue levels of drug and liposomal carrier were determined by homogenization of the organ (20% liver homogenate, 10% others) using a Polytron tissue homogenizer (Brinkmann Instruments, Rexdale, Ontario, Canada). Four hundred microliters of each tissue homogenate sample were digested in 500ul of 0.5M Solvable (NEN Research Products, Boston, Massachusetts, USA) at 62°C for 2 hours. The samples were allowed to cool (4°C) and lOOpl of 30% H2O2 (BDH Canada) were added to each sample to decolorize prior to addition of 5ml of Ultima Gold scintillation cocktail. Quantitation was performed on a Beckman LS3801 liquid scintillation counter using a dual label program. 4.2.6 In vitro exchange of liposomal cyclosporine in whole blood Liposomal cyclosporine was prepared as described above for the pharmacokinetic and biodistribution studies containing 3H-CsA and 1 4 C-CHE. Drug-loaded liposomes were then incubated in whole mouse blood (lmg POPC/ml) at 25°C for 4 hours. At 15 minutes, 1 hour and 4 hours aliquots (500ul) were removed and placed in EDTA microtainers (Becton Dickinson, U.S.A.). After centrifugation at 1500 rpm for 10 minutes the plasma fraction was removed and the packed cells retained for subsequent analysis of their cyclosporine content. For the plasma fraction, separation of liposomes from lipoproteins was achieved by gel exclusion chromatography on Biogel A-15m "minicolumns" essentially as described by Chonn et al. (1991). Plasma aliquots (50pl) 103 were eluted with 150 mM NaCl, 20 mM HEPES, pH 7.4 and fractions (lOOul) collected for analysis of liposomal lipid, cyclosporine and protein. Liposomal lipid and cyclosporine were quantified by dual label liquid scintillation counting, as described above, while protein was quantified using a Micro BCA protein assay (Pierce Chemical Company, Rockford, IL.). 104 4.3 RESULTS As discussed previously, selection of an appropriate lipid composition for liposomal cyclosporine is complicated by conflicting requirements with respect to drug incorporation and vesicle stability in vivo. Liposomes composed of phospholipids alone rapidly fragment following intravenous administration with their constituent lipids then assimilated into lipoproteins or other blood components (Damen, Regts and Scherphof, 1981). This process, which is believed to be triggered by insertion of serum proteins into the lipid bilayer, can be prevented by including cholesterol in the liposomal membrane (Kirby, Clarke and Gregoriadis, 1980). Such cholesterol-containing systems are stable in plasma, exhibiting extended blood residency times and good retention of their aqueous contents (Bally al, 1993 and Mayer et al, 1993). Incorporation of cyclosporine into liposomal membranes, however, is highly sensitive to their lipid composition. Increased fluidity of the lipid bilayer appears to favor drug insertion and consequently as the membrane cholesterol content is increased maximum levels of cyclosporine incorporation are greatly reduced. In the present study, therefore, a lipid composition of POPC:CHOL (molar ratio 5:1) was selected to ensure stability of the liposomal carrier in plasma while allowing moderate levels of cyclosporine to be incorporated. Subsequent pharmacokinetic and biodistribution studies examined clearance of both cyclosporine and its liposomal carrier following intravenous administration and compared the behavior of each to "free" cyclosporine (cremophor preparation) and "empty" liposomes, respectively. 4.3.1 Drug and Lipid Carrier Plasma Elimination Studies in a Mouse Model In Figure 4.1 the levels of cyclosporine and liposomal lipid remaining in the 105 plasma at 15 minutes, 1 hour and 4 hours post-injection are illustrated. Most strikingly, while the liposomal carriers are only slowly cleared from the circulation such that even after 4 hours more than 50% of the initial dose was still present, cyclosporine whether administered within liposomes or as the cremophor formulation was rapidly cleared with less than 2% of the initial dose remaining even at 15 minutes. There was no significant difference in the rate of clearance of liposomal or "free" cyclosporine and similarly blood levels of cyclosporine-loaded and "empty" liposomes are essentially the same at all time points. This result clearly establishes that cyclosporine was lost fairly rapidly from its liposomal carrier and then cleared from the blood with similar kinetics to those displayed by the free drug. 4.3.2 Drug Distribution in Mouse Whole Blood To determine whether exchange of cyclosporine occurs predominantly to erythrocytes or to the lipoproteins, an in vitro study was undertaken in which drug-loaded liposomes were incubated with mouse whole blood. At 15 minutes, 1 hour and 4 hours, aliquots were taken and centrifuged to pellet erythrocytes and white blood cells. The plasma was then fractionated by gel exclusion chromatography on Biogel A-15m to separate the liposomal carriers from lipoproteins. As shown in Table 4.1, when liposomal cyclosporine was incubated with whole blood, most of the drug remains associated with the plasma fraction with little incorporation into the erythrocytes. This result was expected, given the high cholesterol content of the erythrocyte plasma membrane and observations suggesting cyclosporine insertion was highly sensitive to membrane fluidity. When the plasma sample was fractionated by gel exclusion 106 Time Post Injection (hrs) Figure 4.1: Plasma elimination of liposomal cyclosporine. Both lipid, and cyclosporine, • , levels are shown at 30 minutes, 1 and 4 hours post injection for drug-loaded liposomes. In addition lipid levels for "empty" liposomes, , and cyclosporine levels for cremophor formulation, • , are shown. 107 Incubation Time (Minutes) Cyclosporine Distribution (%) Erythrocytes1 Plasma 15 minutes 60 minutes 12% | 88% 21% j 79% i Table 4.1: Distribution of liposomal cyclosporine in whole blood. 108 chromatography, separation of vesicles from lipoproteins can be achieved (Figure 4.2) and it can be seen that essentially all of the cyclosporine was associated with the lipoproteins with little remaining in the liposome carriers even at the earliest time point. Interestingly, when cyclosporine-loaded vesicles are incubated in buffer alone, under similar conditions to those used in the experiment above, and subsequently fractionated on Biogel A-15m, cyclosporine elutes with both the vesicles and in the included volume of the column. This suggests an equilibration between membrane-bound and soluble cyclosporine. It should be noted that this latter drug fraction was not observed in elution profiles when plasma was present, suggesting strong drug binding to the lipoproteins. 4.3.3 Drug and Liposome Tissue Biodistribution To determine whether the biodistribution of cyclosporine was altered by administration in a liposomal carrier, drug level in various organs were determined in the same animals used for the pharmacokinetic study. As for plasma clearance, both cyclosporine and liposomal lipid were followed and compared to the tissue distributions of free drug and empty liposomes. Given that cyclosporine was rapidly lost from its carrier, any change in tissue distribution would presumably reflect events occurring immediately following injection and prior to drug redistribution being complete. In Figures 4.3 and 4.4 are shown the distribution of liposomal lipid in the liver, spleen, kidney, lungs and heart for cyclosporine-loaded and empty liposomes. The radiolabeled lipid marker used in the present study, 14C-cholesterol hexadecyl ether, provides a reliable indication of liposome distribution because, unlike cyclosporine, it was not exchangeable and it was also not readily metabolized (Derksen, Morselt and Scherphof, 1987). A l l results have been corrected for tissue blood volume and are presented as 109 Elution Volume (ml) Figure 4.2: Redistribution of liposomal cyclosporine in plasma. Elution profiles are shown for plasma samples following incubation of liposomal cyclosporine in whole blood at 37°C for 15 minutes (A) or 1 hour (B). As described under Section 4.2.6, fractions eluted from Biogel A-15m "minicolumns" were assayed for liposomal lipid, • , cyclosporine, O, and protein, o . 110 Elution Volume (ml) Figure 4.2 cont'd.: Redistribution of liposomal cyclosporine in plasma. Elution profiles are shown for plasma samples following incubation of liposomal cyclosporine in whole blood at 37°C for 15 minutes (A) or 1 hour (B). As described under Section 4.2.6, fractions eluted from Biogel A-15m "minicolumns" were assayed for liposomal lipid, • , cyclosporine, O , and protein, <j . I l l percentage of total lipid injected (Figure 4.3) and pg POPC/gram tissue (Figure 4.4). As would be expected, considerable accumulation of liposomal lipid was seen in the liver even at 15 minutes with a further increase at the latter time points as vesicles are cleared from the circulation. A similar progressive increase in lipid levels was also seen in the spleen. No significant difference in biodistribution was observed between cyclosporine-loaded and empty liposomes. The biodistribution of cyclosporine administered in cremophor or in liposomes are shown in Figures 4.5 and 4.6. The data represent total tritium counts recovered from the tissues and will include both cyclosporine and any radiolabeled drug metabolites. Again, all data are presented either as percentage of total drug injected (Figure 4.5) or as mg cyclosporine/gm tissue (Figure 4.6). Drug levels in all tissues are highest at 15 minutes and gradually decline at 1 and 4 hours. There was no difference in tissue cyclosporine levels between the two drug formulations with the possible exception of the liver. At all three time points, the liposomal preparation yielded slightly higher cyclosporine levels (on a per gram weight basis) in the liver compared to the cremophor formulation. 112 CO CO o CD -92 8 CD =3 CO .CO cu o CO o c gj cp CD cn CO 2 0 . 2 5 h r j u CD . > CD -co s2 1hr CD C= CD CD CD CO =» - > CD «= CD co .2 4 h r Figure 4.3: Tissue distribution of liposomal lipid as percentage of injected dose. Liposome deposition in the lungs, liver, spleen, kidney and heart is shown for "empty" liposomes (open bars) and cyclosporine-loaded liposomes (yellow bars) at 15 minutes, 1 hour and 4 hours post i.v. administration. 113 Figure 4.4: Concentration of liposomal lipid in major organs. Relative liposome concentrations (pig POPC/gram tissue) in the lung, liver, spleen, kidney and heart are shown for "empty" liposomes (open bars) and cyclosporine-loaded liposomes (yellow bars) at 15 minutes, 1 hour and 4 hours post i.v. administration. 114 CD co O "O % CD cz 8 CD CO CO . o o i .1 $ S S CO £ 0.25hr CZ CD CD CD CO =5 . > CD <= CD CO £ 1hr cm cr >»-t: CO £ 4hr Figure 4 .5: Tissue distribution of cyclosporine as percentage of injected dose. Cyclosporine deposition in the lungs, liver, spleen, kidney, and heart are shown for "free" cyclosporine (open bars) and liposomal cyclosporine (yellow bars) at 15 minutes, 1 hour and 4 hours post i.v. administration. 115 E J 3 > O 8 CD CO CO o O L co _ o & O £= JD CD CD CO 3 > CO C o - « o - S i r CO ^ 0 . 2 5 h r £= CD CD CD CO J > II) C ( B — 1 -> CO 1*£ 1 h r — CD CD CD CO — CD L 5 > 8 -CO £ 4 h r Figure 4.6: Concentration of Cyclosporine in major organs. Relative cyclosporine concentrations (|Ltg CsA /gram tissue) for lungs, liver, spleen, kidneys and heart are shown for "free" cyclosporine (open bars) and liposomal cyclosporine (yellow bars) at 15 minutes, lhour and 4 hours post i.v. administration. 116 4.4 DISCUSSION While previous research has compared the pharmacokinetics of cyclosporine administered in liposomes or cremophor, this study was novel in that plasma clearance and biodistribution of both drug and lipid carrier were followed. In addition, the liposomal formulation used was fully characterized, particularly with respect to drug incorporation. Liposomes can accommodate only relatively modest amounts of cyclosporine with maximal levels of drug incorporation being highly sensitive to lipid composition. In some earlier studies of liposomal cyclosporine it was possible that some of drug was inserted into the lipid bilayer and some present as microcrystals dispersed with the vesicle suspension. By simultaneously monitoring the plasma clearance of liposomes and cyclosporine following intravenous injection, rapid drug loss from the carrier was demonstrated. This process occurs through an initial migration of cyclosporine to the lipoproteins and was expected given the earlier observation that this immunosuppressive agent can undergo rapid interbilayer exchange (Chapter 3). Cyclosporine bound at hydrophobic sites on the lipoproteins was then cleared from the circulation in the same manner as for "free" drug administered in cremophor. Given the similar pharmacokinetics of liposomal and free cyclosporine, it was again not surprising that their biodistributions do not differ significantly. Only in the liver were somewhat higher drug levels seen for the liposomal formulation. Interestingly, Freise et al. (1994) have recently reported increased efficacy for liposomal cyclosporine in a rat liver transplant model. They also indicated that higher drug levels may be present in this organ following administration of a liposomal formulation; however, the results 117 presented showed tissue-to-blood ratios thus preventing any direct comparison with the present data. In view of the fact that cyclosporine was rapidly lost from its liposomal carrier, it was unclear why elevated levels should be seen in this organ. While it might be suggested that this is due to liposome deposition prior to complete drug loss, similar elevated levels should then be seen in the spleen, which also rapidly sequesters the liposomal carrier. In this organ, however, similar drug levels are seen for the two formulations. Similarly the hypothesis that lipid deposition in the liver provides additional hydrophobic binding sites for the drug would also apply to the spleen which does not exhibit significantly higher cyclosporine concentrations. As indicated in the Introduction, conflicting reports have been published concerning the therapeutic properties of liposomal cyclosporine. Further, contrary to expectations for a liposomal drug, the reported pharmacokinetics and biodistribution of liposomal cyclosporine are only modestly different from those for the cremophor formulation. It was shown in the present work that these discrepancies result from rapid cyclosporine efflux from its liposomal carrier upon systemic administration. This study, therefore, re-emphasizes a critical consideration in the design of liposomal delivery systems for hydrophobic drugs such as cyclosporine: drug exchange out of the carrier must be controlled if liposomologists want to take advantage of the altered pharmacokinetics and biodistribution normally afforded by liposomal drug delivery systems (Gabizon, Shiota and Papahadjopoulos, 1989). Additional studies will need to be undertaken to identify the factors influencing drug exchange rates and to develop systems in which these rates can be controlled. Nonetheless, the biodistribution of the drug initially associated with the liposome carrier was not only equivalent to that of the 118 current i.v. formulation in the tissues studied but offers an alternative and safer vector for the delivery of cyclosporine versus the cremophor EL oil (Jiang and Acosta, 1993) that is currently used to solubilize the cyclosporine for clinical use. 119 CHAPTER 5 LIPOSOMAL CYCLOSPORINE TISSUE DISTRIBUTION IN RAT HEART TRANSPLANTATION In earlier chapters, characterization of a physically well-defined liposomal immunosuppressant formulation was described. In the formulation, the drug, cyclosporin A, was found to be associated with the liposome when subject to gel chromatography and density gradient separation techniques. Moreover, a finite amount of cyclosporin A was incorporated per mole of liposomal lipid and unincorporated drug was removed by filtration. The incorporated drug, however, readily exchanges from the liposome as evidenced in vitro and in mouse pharmacokinetic and biodistribution studies. The benefit of administering liposomal cyclosporin A was evaluated in a rat heart transplant model described in this chapter. The presence of liposomal cyclosporin A in significant levels in the heart graft was examined in a immunologically incompatible rat heart transplant model. The results taken four hours after liposomal cyclosporin A injection indicated that while cyclosporin A did not preferentially accumulate in grafted hearts, the liposome levels in the grafted heart significantly exceeded the levels in the host heart. Rats transplanted with immunologically incompatible hearts treated with free cyclosporin A and rats transplanted with immunologically compatible hearts and did not express this difference in liposome levels between host hearts and transplanted hearts. The results add confirmation to the results discussed in earlier chapters attested to the rapid loss of the hydrophobic immunosuppressant molecule, cyclosporin A, from the liposome. The results from the present study, however, took the study of delivering drugs such as immunosuppressants a step further, by unequivocally demonstrating the natural targeting 120 ability of liposomes for transplanted cardiac allografts. Although it has yet to be widely demonstrated, the generalization of liposome accumulation to other transplanted organs undergoing rejection cannot go unrecognized as a plan in the delivery of therapeutic agents. 5.1 INTRODUCTION Cyclosporin A has been used in the management of post-operative organ transplantation. However, the nephrotoxic effects of the drug can limit further administration in organ transplant patients, especially those who have associated kidney problems giving rise to hypertension. It was demonstrated that this drug can be stably incorporated into a liposome carrier and that a liposomal formulation of cyclosporin A has similar pharmacokinetics and biodistribution in CD1 mice as the current intravenous formulation of cyclosporin A used clinically. A liposomal formulation, however, may be a safer drug formulation to administer than the current drug formulation which employs a rather cytotoxic solvent emulsifier (Jiang and Acosta, 1993), cremophor EL, in which the drug is solubilized for administration. Having characterized the liposomal cyclosporin A formulation and establishing the necessary groundwork in terms of drug clearance and biodistribution in an animal model, one of the other objectives was to characterize the potential of a liposomal drug delivery system that will target a transplanted organ, in particular a transplanted heart. The inhibitory effects of cyclosporin A on T-lymphocyte dependent immune responses result in transplantation tolerance. The allograft, which has crossed major histocompatibility (MHC) barriers, is not destroyed by the recipient's immune system because CsA can inhibit the induction of cytotoxic effector cells (Hess, 121 Tutschka and Santos, 1981a). In addition, CsA indirectly favors the suppressor regulatory arm of the immune response by activating suppressor lymphocytes (Hess, Tutschka and Santos, 1981b). Cyclosporin A therapy has been associated with a number of side effects causing transplant patient morbidity. Incorporation of cyclosporin A in liposomes can reduce drug toxicity by altering the biodistribution of the drug in the body. In a second animal model, the rat, the distribution of liposomal cyclosporin A was examined. Rats have been chosen because they are generally larger than mice and hence offer a larger anatomy to work with, which was especially important when performing microsurgery to transplant a heart from one rat to another. The biodistribution of immunosuppressants in various liposome formulations in rats has been analyzed by Binder et al. (1994) and Ko et al. (1994). The majority of these liposomes end up in organs of the reticuloendothelial system, namely the liver and spleen. Organ transplants, such as heart, small bowel and liver in rats have been performed successfully by Tanabe et al. (1994), Oberhuber et al. (1994) and Sriwatanawongsa, Davies and Calne (1994), respectively. The levels of liposomal cyclosporin A (L-CsA) were measured in rats with transplanted hearts. A comparison was made for drug and liposome levels in the hosts' hearts and the transplanted hearts. In this study, the animals were divided into three groups, those with heart transplants from the same strain of rats (homeografts), those with heart transplants from a different strain but same species of rats (allografts), and allografts pretreated with cyclosporin A. 122 5.2 MATERIALS AND METHODS 5.2.1 Lipids and Chemicals l-Palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) was obtained from Avanti Polar Lipids Inc. (Birmingham, AL, USA). 5-Cholesten-3(3-ol was obtained from Sigma Chemical Co. (St. Louis, MO, USA). Cholesteryl hexadecyl ether [Cholesteryl-4-1 4C] (1 4C-CHE) (NEN Research Products, Boston, Massachusetts, USA) and [mebmt-b-3H]cyclosporin A were purchased from Amersham Life Science (Little Chalfont, Buckinghamshire, England). Cyclosporin A (CsA) was purified from the Sandimmune Oral formulation of cyclosporin A as described earlier (Section 2.2.1). 5.2.2 Drug and Lipid Formulation Cyclosporin A. The intravenous formulation of CsA (50mg/ml) was obtained from Sandimmune and diluted with normal saline (to 17mg CsA/ml) used in daily lOmgCsA/kg i.p. injections. Lyophilization of Lipid Mixtures. Mixtures of POPC, cholesterol and the non-exchangeable, non-metabolisable lipid marker, 14C-cholesterol hexadecyl ether (14C-CHE) (Derkson et al., 1987), were prepared by colyophilization from benzene:methanol (95:5 v/v) (Virtis, Leybold Vacuum Products, USA) under high vacuum (60-100 mtorr) for a minimum of six hours. Preparation of Liposomes. Liposomes are prepared by hydrating the lyophilized lipid mixture with 150mM NaCl, 20mM HEPES, pH 7.4 (HEPES buffered saline, HBS) and then freezing in liquid nitrogen (5 min) and thawing (10 min at 37°C) for 5 cycles to maximize lipid hydration. Large unilamellar vesicles (LUVs) were prepared essentially as described by Hope et al, 1985. The M L V suspension was extruded ten times at 37°C 123 through two stacked lOOnm polycarbonate filters (Poretics Corp., Livermore, CA) using an Extruder (Lipex Biomembranes, Vancouver, BC, Canada). Incorporation of Cyclosporine. POPC:CHOL (5:1) vesicles (60mg/ml; 170uCi 14C-CHE/mmole POPC) were incubated at 37°C. Cyclosporine dissolved in ethanol (30mg/ml; 37mCi/mmole CsA) was then added to the vesicle suspension to give 1.6mole% cyclosporine. After incubation at 37°C for 5 minutes, any unincorporated drug was removed by passage for the vesicles through a 0.22 micron filter. 5.2.3 Rat Pancreas and Heart Transplantation Diabetic rat pancreas transplantation was carried out as described by Latifpour et al. (1992). Wistar rats were made diabetic with a single intravenous (i.v.) 65mg/kg injection of streptozotocin (Sigma Chemical Co., St. Louis, MO., U.S.A.) dissolved in citrate buffer (pH 4.5) and blood glucose measurements were taken with Glucometer II (Ames, Elkhart, IN). After pancreas transplantation, rats were either treated with cyclosporine (lOmg/kg/day, i.p., n=6) or nothing at all (n=6). Daily blood glucose was monitored in all Wistar rats until the concentration exceeded 400mg%, at which time the pancreas transplant was considered rejected. The animals were then sacrificed with carbon dioxide gas. Male Wistar rats weighing approximately 250g were obtained from the University of British Columbia Animal Care Center, Vancouver, B.C. and inbred male Lewis rats weighing approximately 350g were obtained from Charles River, Montreal, Quebec. Dr. Mark Meloche performed the surgical transplantation of rat cardiac grafts. Lewis rats served as the recipients in all cases. The hearts were anastomosed to the abdominal great vessels which supply and return blood to the lower extremities (namely hind legs of the 124 rat) using standard microvascular techniques (Lee, S., 1987). The blood circulating through the transplanted heart causes it to beat thereby allowing the status of the transplanted heart to be assessed daily by palpation. The heart transplants consisted of a Lewis heart transplant to Lewis rat recipient or to a Wistar rat recipient. The host heart served as functional internal control while the transplanted heart does not play a functional role in pumping blood through the recipient's body. In the first set of experiments, the rats were either given daily injections of lOmg CsA/kg/day or no CsA until the graft ceased to beat. In the second study, where the biodistribution of liposomal CsA was measured, rats were divided into three treatment groups and were injected with liposomal cyclosporin A on the 7th day post-transplant (Figure 5.1). 5.2.4 Tail vein injections The transplanted hearts were monitored for 7 days and either treated with lOmg CsA/kg rat/day using a 200ul i.p. injection made from an intravenous cyclosporin A formulation (Sandimmune) diluted with saline (7% NaCl) or no treatment was given (see Table 5.1). The rats were injected with liposomal cyclosporin A and sacrificed 4 hours later on the seventh day post-transplant with an overdose of ketamine/xylazine anesthetic (150mg ketamine/kg and 18mg xylazine/kg). A time frame of 7 days post-transplant was chosen to allow for adjustment of the host to the transplant and for an adequate immune response to the graft if one was to occur. 125 Figure 5.1: Rat Heart Transplant Model. Donor hearts were transplanted from rats of the same strain (Group 1, isogeneic) or from a different rat strain (Groups 2 and 3, allogeneic) and anastomosed to the major blood vessels, descending aorta and ascending inferior vena cava, within the abdomen of recipient rats. Group 3 animals were given cyclosporin A at lOmg/kg/day. The cardiac transplants were allowed to settle in for 7 days and graft survival was monitored dairy by heart palpatioa On the seventh day post-transplant, a liposomal cyclosporine formulation was injected via the tail vein. The rats were sacrificed 4 hours thereafter and the biodistribution of the liposomal formulation determined. 127 5.2.5 Blood analysis Blood samples were taken by heart puncture immediately after the animal was sacrificed. The blood samples were placed in microtainer tubes containing EDTA (Becton Dickinson, Rutherford, New Jersey) and half the samples were centrifuged at 1500 rpm for 10 minutes (Heraeus Instruments GmbH Labortechnik, Hanau, Germany). The amounts of drug and lipid in plasma and in whole blood were determined by subjecting 50ul samples to 500jal of 0.5M Solvable digestion (NEN Research Products, Boston, Massachusetts, USA) for 2 hours in a >60°C water bath, cooled to room temperature and decolorized with 200ul of 30% hydrogen peroxide (BDH Canada) for at least 2 hours. Finally, 5ml of scintillation cocktail (Ultima Gold, Packard Instrument B.V., Groningen, The Netherlands) was added to each sample and the amount of 3 H and/or 1 4 C was determined by liquid scintillation counting using a dual label program (Beckman, Fullerton, CA, USA). 5.2.6 Tissue analysis Tissue levels of drug and liposomal carrier were determined by homogenization of the organ ( 2 0 % liver homogenate diluted in HEPES buffered saline (HBS), 10% homogenate of lung, kidney, spleen, heart and transplanted heart) using a Polytron tissue homogenizer (Brinkmann Instruments, Rexdale, Ontario, Canada). Four hundred microliters of each tissue homogenate sample were digested in 500ul of 0.5M Solvable (NEN Research Products, Boston, Massachusetts, USA) at >60°C for 2 hours. The samples were allowed to cool to room temperature and 200ul of 30% H 2 0 2 (BDH Canada) were added to each sample to decolorize prior to addition of 5ml of Ultima Gold 128 scintillation cocktail. Quantitation was performed on a Beckman LS3801 liquid scintillation counter using a dual label program. 129 5.3 R E S U L T S In order to test the effectiveness of cyclosporin A in preventing organ rejection, diabetic Wistar rats had microsurgery performed on them to replace their unresponsive pancreas with healthy Lewis pancreases. The survival of the pancreatic graft could easily be monitored by measuring the blood glucose level in the rat. This model allowed us to observe the effectiveness of cyclosporin A in allogeneic rat organ transplantation. In the Wistar rats (n=6) not treated with cyclosporin A post-pancreas transplant, the blood glucose levels increased from basal levels of ~120mg% after 8 to 18 days post-transplant with a mean rejection time of 12 days post-transplant. Figure 5.2 illustrates the blood glucose levels of untreated diabetic rats before and after pancreas transplant. Pancreas transplant rejection was determined to be complete when blood glucose levels reached 400mg% glucose. When 10 mg/kg/day of cyclosporin A was administered intraperitoneally, however, all six Wistar pancreases were tolerated and survived after cyclosporin A therapy was removed 30 days post-transplant. As Figure 5.2 indicates, these cyclosporin A-treated rats (n=6) had no increase in blood glucose levels 34 days post-transplant. We performed allogeneic heart transplants where Wistar hearts were transplanted into Lewis rats. Rats (n=6) were treated with lOmg CsA/kg/day i.p. while others received no treatment (n=6). Effective doses of this immunosuppressant have been found to range from 5 to 25 mg/kg/day (Green, 1981) and cardiac allografts have been reported to be maintained with a dose of 15mg/kg/day in rats (White et al., 1980). Kawahara et al, (1980) have reported toxic pulmonary infection resulting in 40% mortality in rats with 130 700 0 I I 1 I I I I I I I I I I 0 5 10 15 20 25 30 35 40 45 50 55 Day(s) Post Pancreas Transplant Figure 5.2: Rat pancreas transplantation allograft survival with cyclosporin A post-operative management. Six pancreas transplants were performed in which pancreases isolated from Lewis rats are transplanted (on day zero) into chemically-induced diabetic Wistar rats with elevated blood glucose levels. Rats were given either no drug (filled symbols) or lOmg CsA/kg/day i.p. (open symbols) and blood glucose levels measured. Blood glucose levels were used to assess the functional ability and state of the transplanted pancreas and rejection of the allograft was determined to occur when blood glucose levels returned to the elevated levels seen prior to transplantation (i.e. exceeding 400mg%). Cyclosporin A-treated rats had CsA injections discontinued on day 20 and all rats were terminated when blood glucose levels were over 400mg%. Pancreas allograft survival was maintained for at least 34 days in CsA-treated animals without complications. The mean rejection times of transplants with and without cyclosporin A intervention are shown in the figure. 131 transplanted hearts at lOmg CsA/kg/day i.p. The rats receiving cyclosporin A (lOmg CsA/kg/day) survived for over 30 days while those not receiving cyclosporin A had their heart grafts fail at day 13±1 as assessed by heart palpation. This result served as the basis for designing liposomal cyclosporin A biodistribution experiments (See Methods). 5.3.1 Drug and Liposome Tissue Distribution A dramatic difference in the levels of POPC/CHOL (5:1) liposomes found in allogeneic transplanted hearts when compared to levels was found in the hosts' hearts. This result seen in group 2 rats, Figure 5.3, was not observed in isogeneic transplant group 1 rats. Liposome accumulation in allogeneic transplanted hearts of group 3 rats was less pronounced and essentially ameliorated with 8 days of cyclosporin A treatment (lOmg CsA/kg/day) prior to radiolabeled liposomal cyclosporin A injection. In all groups tested, the liposome accumulation was greatest in the liver followed by the spleen, kidney, lung and heart as a percentage of recovered dose. More liposomes were present in the transplanted hearts of group 2 rats, which were undergoing rejection. The increase in liposomes in the transplanted hearts of group 2 rats has a noticeable effect on the amounts accumulating in the liver tissue of these group 2 rats when compared to the amounts in the liver tissue of groups 1 and 3 (Figure 5.3). This observation could be related to the fact that circulating macrophages, arising from the liver have the ability to engulf liposomes before infiltrating into the foreign heart graft as a part of the host's natural immune response. The biodistribution of cyclosporin A (Figure 5.4) is different from that of the liposomes even when these are administered together in a formulation. 132 20 CD CO O TJ "ra 16 c CO c o u cu 3 Ui OT 0) E o (A O Q. 12 Group 1 Group 2 Group 3 Figure 5.3: Biodistribution of liposomes to tissues in a rat heart transplant model. Tissue distribution of liposomal lipid as a percentage of initial dose (corrected for plasma content) in Lewis rats. Liposome deposition in the lungs, liver, spleen, kidney, heart (open columns) and transplanted heart (hatched columns) is illustrated for CsA-loaded liposomes four hours post i.v. tail vein injection. 133 0) W o 75 c 0) O O o 3 (A (A O Q. (A O U >» o 30 25 20 L 15 U 10 5 U Mf^M M^^M M/JLM w ^ V # "H^4fW Group 1 Group 2 Group 3 Figure 5.4: Biodistribution of cyclosporin A to tissues in a rat heart transplant model. Tissue distribution of CsA as a percentage of initial dose in Lewis rats. Deposition in the lungs, liver, spleen, kidney, heart (open columns) and transplanted heart (hatched columns) is illustrated for liposomal CsA four hours post i.v. tail vein injection. 134 Although the results indicated that both liposomes and CsA have a high propensity for uptake by the liver, the similarities end there (Figures 5.3 and 5.4). The liposome content in spleens was higher than in the kidneys while the reverse holds true for CsA when the tissue content was expressed as a percentage of recovered dose (Figures 5.5 and 5.6, respectively). The amount of CsA in the lungs was similar to amounts accumulating in the spleens, while amounts in the hearts and transplanted hearts of group 1 and 3 rats indicated no statistical difference. The statistically higher amounts of CsA in the transplanted hearts of group 2 rats compared to the amounts in the hosts' hearts are not apparent when these results are corrected for the weights of the organs (Figures 5.3 and 5.4) for liposomes and CsA biodistribution, respectively. The liposome content in the tissues collected was greatest in the spleen and liver when corrected for the weights of these organs. The level of liposomes in group 2 rat transplanted hearts where it was undergoing rejection as a result of crossing major histocompatibility (MHC) barriers (Wistar to Lewis) was as great as the levels found in the spleens and livers of these rats. More importantly, the levels in the transplanted hearts were dramatically higher than in the hosts' hearts of group 2 rats. The results also indicate that this was not the case for isogeneic heart transplants (group 1) or when the allogeneic heart transplant recipients were treated with lOmg CsA/kg/d prior to liposomal CsA injections (group 3). The CsA content in the tissues tested did not differ among the groups when corrected for the weights of the organs except for a pronounced decrease in the levels measured for the livers of group 3 rats. The lower levels in these livers was due to the fact that the daily 135 CD 3 OT OT 'S E co w O) •a "5. OT o o •E c 0 c o o CD 3 OT OT 'S CD E o OT O a 1600 1200 h-800 \-400 h ' 1 1 1 1 1 1 1 1 ' 1 1 1 1 • IV/J VSSA Group 1 Group 2 Group 3 Figure 5.5: Biodistribution of liposomes to tissues in a rat heart transplant model. Concentration of liposomal lipid in major organs (open columns) and transplanted hearts (hatched columns) of Lewis rats. The liposomal concentrations (nmoles lipid/gram tissue, corrected for plasma content) in the lung, liver, spleen, kidney, heart and transplanted heart for CsA-loaded liposomes four hours post i.v. tail vein injection. 136 CD 3 (A .12 E CO < to o (0 o E c c CO c o o CD 3 (0 (A CD _C 'E O a (A O O >» o jgjr.Mf Mf.di^ Jtf.A¥ V % W ^ v s ^ 4 / ^ v s ^ 4 > ^ Group 1 Group 2 Group 3 Figure 5.6: Biodistribution of cyclosporin A to tissues in a rat heart transplant model. Concentration of CsA in major organs (open columns) and transplanted hearts (hatched columns) of Lewis rats. Relative CsA concentrations (nmoles CsA/gram tissue) in the lungs, liver, spleen, kidney, heart, transplanted heart are shown for liposomal CsA four post i.v. tail vein injection. 137 treatments with CsA prior to injection of radiolabeled liposomal CsA prompted the efficient metabolism and elimination of cyclosporin A by the P450 metabolizing enzymes in the liver. Hence, less of the radiolabeled CsA, injected as liposomal CsA, accumulated in the liver. 5.3.2 Drug and Liposome Distribution in Blood versus Tissues Analysis of blood indicated that 10-22% of the recovered cyclosporin A dose remained 4 hours post-injection in all three groups tested (Figure 5.7). Most of this cyclosporin A was found to be sequestered in plasma proteins and not in liposomes. Conversely, between 75 and 83% of the liposome-recovered dose remained in the blood 4 hours post-injection. Hence, whereas most of the liposomes remained in circulation after 4 hours, the drug had already been distributed rapidly into the tissues. These results are in agreement with previous in vitro exchange and mouse model studies where the drug was found to quickly transfer out of the liposomal bilayer when a lipid sink, such as other liposomes or biomembranes, becomes available. Hence, association of the drug with POPC/CHOL (5:1) liposomes does not significantly alter drug distribution in the body. 138 Percentage distribution (of recovered dose) Figure 5.7: Blood and tissue distribution of liposomal cyclosporin A in Lewis rats. The relative amounts (percentage of recovered dose) of liposome carrier (top 3 bars) and cyclosporin A (bottom 3 bars) in the blood (open bar portion) or in the tissues (hatched bar portion) is pictured for liposomal cyclosporin A four hours post i.v. injection. Numbers 1, 2 and 3 correspond to the 3 treatment groups. 139 5.4 DISCUSSION In this study, the immunosuppressant, cyclosporin A, was incorporated into lipid vesicles known as liposomes. This drug and lipid combination was potentially useful in post-operative organ transplantation in the selective delivery of cyclosporine to sites of ongoing rejection. The immunosuppressant allows for tolerance of the graft by inhibiting the proliferation of the recipient's lymphocytes responsible for destroying foreign particles such as a transplanted organ. Because liposome carriers accumulate at sites of disease and inflammation, they may deliver associated drug to these areas of interest. When applied in the rat heart transplant model, the ideal pharmacokinetics of liposomal cyclosporin A would be uptake into the grafted tissue where circulating macrophages and monocytes recognize and engulf the liposomes carrying cyclosporin A. The present study sought to measure the levels of drug and liposome carrier in 5 rat tissues plus the transplanted heart following i.v. injection of liposomal cyclosporin A seven days after heart transplantation to determine if appreciable amounts of liposomal cyclosporin A accumulated in particular tissues. Two preliminary studies were carried out to investigate the effectiveness of cyclosporin A in a rat organ transplant model and the survival of rat heart graft transplants, respectively. These experiments served as controls for the major study in which the biodistribution of liposomal cyclosporin A (L-CsA) was analyzed for three rat treatment groups. The control studies resulted in allogeneic organ transplants that survived 12 and 13 days on average for the pancreas and heart, respectively, when no cyclosporin A post-operative management was provided. In contrast, grafts were tolerated and survived well past 34 and 30 days for pancreas and heart transplants, 140 respectively, when treated with lOmg/kg/day i.p. cyclosporin A. These results provided us with information necessary about how well cyclosporin A works in the rat model and an idea of the time scale before complete rejection occurs. The major study consisted of 3 groups of rats with heart transplants. In the first group, considered as control animals (n=3), donor hearts were taken from Lewis rats and transplanted into similar inbred Lewis rats (syngeneic transplant). In this situation, the graft from an inbred rat species, is genetically identical to the host transplantation antigens and therefore the host immune system will not recognize it as foreign. One would not expect stimulation of lymphocytes to infiltrate and destroy this type of graft. The second group (n=4) composed of donor hearts from Wistar rats (genetically incompatible) were transplanted into Lewis recipients (allogeneic transplant). The cardiac graft from a different strain of rat in this case would cause allorecognition and a subsequent immune rejection. Based on earlier experiments characterizing the duration of cardiac graft survival between these two rat strains, an appropriate assessment of liposomal drug accumulation to the graft undergoing rejection can be made seven days post-transplant. The third group (n=4) was similar to the second group with the only exception being that each recipient rat received daily i.p. injections of cyclosporin A (lOmg/kg/day) starting one day prior to heart transplantation and finishing on the day of injection of liposomal cyclosporin A. The heart was transplanted into the recipient's abdominal cavity where the aorta and pulmonary artery are anastomosed to the descending abdominal aorta and ascending inferior vena cava, respectively. Seven days post-transplant the rats were injected with radiolabeled liposomal cyclosporin A via the tail vein and sacrificed 4 hours post-injection. 141 In Figure 5.6, comparing cyclosporin A tissue content as a percentage of recovered dose, it was apparent that over 70% of the recovered tissue cyclosporin A dose accumulates in the liver followed by approximately 10% of the recovered tissue cyclosporin A dose in the kidneys. There are two statistically significant differences among the three groups tested. The first difference was in the amount of cyclosporin A in the spleens of group 1 rats in which there are lower levels of the drug present in comparison to spleens of rats in groups 2 and 3. The second difference was the amount of cyclosporin A in the transplanted heart tissue of group 2 rats and the amount in the hosts' heart tissue. An explanation of the former observation stems from the fact that group 1 rats are of isologous transplant and the immune system of these recipients has not been prompted to mount an immune response to the isograft. Hence, the proliferation of lymphocytes that would be arising from the spleen (the organ being an integral part of the RES) in response to foreign antigens such as in groups 2 and 3 homologous transplants does not occur. As a result, the spleen does not take up as much cyclosporin A, leaving the drug for hepatic metabolism. The larger amount of the drug in transplanted heart tissue than in the control heart of group 2 animals was explained by the larger size of the transplanted heart which could weigh up to 5 times more than the hosts' hearts. The clinical accumulation of lymphocytes in acute allograft rejection has been similarly observed (Arai et al., 1992) and may explain our observations of enlarged allogeneic hearts. The results for the delivery of POPC/CHOL (5:1) liposomes are encouraging. The preferential accumulation of liposomes in the allogeneic transplanted hearts of cyclosporin A non-treated rats (group 2) has established that there was passive targeting 142 of these vesicles by cells of the immune system. The dramatic difference lies in the accumulation of liposomes in the transplanted hearts of group 2 rats in which there was a 44-fold more liposomes in the transplanted heart compared to the hosts' heart (Figure 5.5). Friese et al. (1994) have shown that a liposomal CsA formulation results in increased liver graft survival and that there was preferential accumulation of the liposomal formulation in the liver when compared to a formulation of CsA in saline following bolus injection. The results for group 2 liposomal biodistribution are different from those of group 1 in which the heart transplants were of the isogeneic type and no rejection was expected to occur. In group 3, in which allogeneic heart transplantation was followed up with lOmg/kg/day of cyclosporin A i.p. treatment, different results were again obtained due to the suppression of the immune response. In groups 1 and 3, then, statistically different levels of liposomes between the hosts' heart and the transplanted heart were not achieved. However, the higher levels of liposomes in the transplanted hearts of group 2 rats were not accompanied by a proportional increase of cyclosporin A at this graft site. This finding for cyclosporin A levels confirms earlier results (Ouyang et al. (1995) and Choice et al. (1995)) that can be summarized as a rapid transfer of the drug out of POPC/CHOL liposomes when lipid sinks such as other liposomes and/or biomembranes are present. One explanation of the former observation was that only a portion of the CsA drug molecule adsorbs weakly to the liposome surface. When the liposome comes into contact with disruptive protein binding (albumin, ct-globulin), hydrophobic lipid interactions or lipid extraction by lipoproteins (HDL) the drug is released from the liposome. Hence, the feasibility of a liposomal cyclosporin A formulation to target transplanted organs such as 143 the heart and to prevent the nephrotoxicity associated with the use of this drug remains questionable. The liposome carrier, however, will serve as an excellent tool for delivering other drugs to transplanted organs based on the results here for cardiac allografts. 144 C H A P T E R 6 S E P A R A T I O N O F LIPOSOMES F R O M P L A S M A LIPO-PROTEINS USING F A S T PROTEIN LIQUID C H R O M A T O G R A P H Y (FPLC) In the past, the amount of drug remaining with the liposome was either measured in plasma samples directly or the liposomes were separated and then measured for drug content. The experiments performed for liposomal cyclosporine have presented the problem of rapid drug loss from the liposome. One central problem in the accurate measurement of agents associated with the liposome was the proper separation of the liposomes from all other confounding factors to the accurate measurement of agents remaining associated with the liposome. A rapid and efficient method for separating liposomes from plasma lipoproteins was established and presented in this chapter. Previous methods of separation have been laborious and time consuming, requiring either repetitive BioGel centrifugation of spin columns or conventional column chromatography employing elution times of several days. A new technique was sought to replace these existing techniques and improve the accuracy of measuring the amount of agents remaining with liposomes injected into plasma. The possibility of separating liposomes and plasma lipoproteins was investigated using fast protein liquid chromatography (FPLC). The peak fractions were identified for very low density lipoprotein (VLDL), low density lipoprotein (LDL) and high density lipoprotein (HDL) in mouse plasma samples. Similar peak fraction identities were made for human plasma lipoprotein samples. Liposomes in the 140-160nm size range eluted early on the FPLC when run alone and co-eluted with the earliest FPLC eluting lipoprotein fraction, VLDL, in liposome and plasma sample mixture. The liposomes were dual labeled with tritium and 145 a fluorescent probe for identification purposes. The remaining plasma components were eluted off the FPLC separate of liposomes and VLDL. A simple precipitation method was employed to exclude VLDL from plasma prior to separation on the FPLC to obtain only liposomes in the earliest eluting peak fractions. Taken together, our results have shown that FPLC can be used to separate liposomes from plasma lipoproteins. 6.1 INTRODUCTION The study of liposomal formulation stability in the blood has been precluded in the past by the lack of an effective and efficient method of separating liposomes from plasma macromolecules. In some cases, the drug delivered with the liposome becomes lost or sequestered in blood components. Fast protein liquid chromatography (FPLC) was investigated as a technique that could be used for the separation of liposomes from individual plasma proteins and lipoproteins. In the case of liposomal drug delivery, such a separation would then allow quantitation of drug associated with the liposomal carrier, plasma proteins or lipoprotein fractions. Alternatively, interaction of blood components with the liposomal carrier could be determined. Previous attempts to separate liposomes from plasma lipoproteins have been laborious and time consuming, requiring either crude repetitive centrifugation of mini-columns (Chonn, Semple, and Cullis, 1991) or conventional column chromatography employing elution times of several days (Rodrigueza, Pritchard and Hope, 1993). Therefore, an efficient method for the separation of liposomes was required. In this chapter, we describe the use FPLC in the separation of liposomes from plasma lipoproteins. 146 6.2 MATERIALS AND METHODS 6.2.1 Lipids and Chemicals l,2-Distearoyl-sn-glycero-3-phosphotidylcholine (DSPC) and 1,2-Distearoyl-phosphotidylethanolamine-N-(poly[ethyleneglycol)-2000) (DSPE-PEG2 0 0 0) were obtained from Avanti Polar Lipids Inc. (Alabaster, AL, USA). Cholesterol (CHOL) was obtained from Sigma Chemical Co. (St. Louis, MO, USA) and N-[4-(p-maleimidophenyl)butaroyl]-l,2-disteroyl-5«-glycero-3-phosphotidylethanolamine sodium salt (DSPE-MPB) from Northern Lipids Inc. (Vancouver, B.C., Canada). Cholesteryl hexadecyl ether (3H-CHE) (NEN Research Products, Boston, Massachusetts, USA) and l,l'-dioctadecyl-3,3,3',3'-tetramethylindocarbocyanine perchlorate (Dil, MW=933.88, Molecular Probes, Eugene, Oregon, USA) served as lipid tracers. 6.2.2 Blood samples Blood was obtained from either human normolipidemic females (Atherosclerosis Specialty Laboratory, Healthy Heart Program, St. Paul's Hospital, Vancouver, B.C.), CD1 mice (Jackson Laboratories) or SCID Rag 2 (severely combined immunodeficient, recombinase activation gene-2 deficient) mice (B.C. Cancer Agency and Research Center Joint Animal Facility, Vancouver, B.C.) (Shinkai et al, 1992). Blood samples of the above were spun down in 500 pi plasma or serum Microtainers (Becton Dickinson, USA) at 1500 x g on a tabletop centrifuge (Centronics S-103 NAR, Japan) at room temperature for 10 minutes. The serum and plasma samples were frozen down to -20°C and thawed on the day they were used. 6.2.3 Isolation of lipoprotein fractions by Density Gradient Ultracentrifugation (DGUC) 147 This process was carried out according to earlier published methods (Mills, Lane and Weech, 1984). Briefly, the lipoproteins in a pooled SCID plasma sample (2.88ml) were sequentially fractionated in solutions of different densities. The densities of the final solutions, d,=1.006g/ml, d2=T.019g/ml, d3=1.063g/ml were checked using an Anton Parr precision density meter. The samples were placed in 5.8ml Quick-Seal polyallomer tubes, heat sealed and ultracentrifuged (Beckman Model 50.4Ti rotor) at 40,000 rpm for 18 hours at 15°C. Four fractions corresponding to VLDL, IDL, LDL and HDL and the plasma proteins were isolated in solutions of d<1.006g/ml, d=1.006-1.019g/ml, d=1.019-1.063g/ml and d>1.063, respectively. Lipoprotein fractions (1ml each) were dialyzed (Spectra/Por Membranes, MWCO=12,000-14,000, The Spectrum Companies, Gardena, CA) for 48 hours in 150mM NaCl, 25mM N-[2-hydroxyethyl]piperazine-N'-[2-ethanesulfonic acid] (HEPES) pH 7.4 to remove the high salt content in the fractions. 6.2.4 Lipoprotein Gel Electrophoresis Qualitative gel electrophoresis was performed using a Ciba Corning Lipoprotein Electrophoresis System (Ciba Corning Diagnostics Corporation, Alameda, CA). Briefly, a 1% agarose universal gel/8 was loaded with 1-2)0.1 of a lipoprotein fraction and placed in the electrophoresis cell to run for 35 minutes in universal buffer pH 8.6. The gel was blotted dry and further heated dry using a hot stream of air. The gel was allowed to cool for 5 minutes before staining with Fat Red 7B. Destaining was performed using methanol/dH20 (30:70v/v). The gel was scanned into a computer (ScanJet 4C, Hewlett Packard). 6.2.5 FPLC Instrumentation 148 The FPLC instrument consisted of a LKJ3 brommer pump (Germany) linked to a BioRad Econo System Controller running a Pharmacia Superose 6 H.R. (10mm diameter X 30cm length) column and a Pharmacia Superose 12 prep grade (16mm diameter X 50cm length) column in series. The flow rate was set at 0.5ml/min at 3 bar pressure run on a pump for a total run time of 240 minutes. (Conversions: 1 bar = 0.069 p.s.i. = 1 atmosphere = O.lMPa = 760 mm Hg). The BioRad Econo System Controller (BioRad, Canada) was set up to collect 60 fractions of lmL each. The eluting fractions were also monitored by a spectrophotometer attached in series to measure the absorbance of the fractions at 280nm. The rurming buffer (mobile phase) was 150mM NaCl, lOmM Tris, 0.03% NaN 3 at pH = 7.4. Fraction collection was set to commence at 70 minutes and end at 190 minutes (that is, elution volume collection between 36 and 95ml). 6.2.6 Preparation of Liposomes DSPC, cholesterol, DSPE-PEG 2 0 0 0, DSPE-MPB at a molar ratio of 52:45:2:1, respectively, fifty microcuries of 3H-CHE and 0.75umoles of Dil in the lipid mixture (150umoles total lipid) were dissolved in benzene/methanol (95:5 v/v), frozen in liquid nitrogen and dried down in a lyophilizer (Virtis, Leybold Vacuum Products, USA) under high vacuum (60-100 mtorr) for five hours. The dry lipid film was hydrated with 2ml of HEPES buffer saline (HBS, 150mM NaCl, 20mM HEPES, pH 7.4), transferred into a cryovial and frozen in liquid nitrogen and thawed in a 56°C water bath for 5 cycles. The liposomes were subsequently size-reduced through two stacked lOOnm pore size filters using an extrusion procedure (Hope et al, 1985). The resultant liposomes were analyzed using a Nicomp particle sizer and determined to have a mean diameter of 140-160nm. 149 The specific radioactivity of the liposome formulation was 0.33mCi [3H]/mmole total lipid. 6.2.7 FPLC sample preparation The samples for injection onto the FPLC were made up of 250ul of either DGUC fractions or antibody in buffer or serum or plasma and/or 1 umole of liposomal lipid. The volume of the sample was adjusted to 350ul with the addition of FPLC running buffer (150mM NaCl, lOmM Tris, 0.03% NaN3 at pH = 7.4). The samples were lightly vortexed and loaded onto the FPLC within 10 minutes and run at room temperature as described in the FPLC section above. Some samples were treated using the precipitation procedure outlined in S ection 6.2.8. 6.2^ .8 Removal of very low density lipoprotein (VLDL) The FPLC column setup does not allow for separation of VLDL from liposomes. A modified (dodeca)-tungstophosphoric acid (PTA, BDH, England), magnesium chloride (MgCL, BDH Inc., Toronto, Canada) precipitation procedure (Burstein, Scholnick and Morfin, 1970, Mills, Lane and Weech, 1984 and Schriewer, Kohnert and Assman, 1984) was used to precipitate apo-B containing lipoproteins, VLDL and LDL, from mouse plasma. One hundred microliters of a lOmM total lipid liposome formulation was added to 250ul of SCID plasma and the precipitating reagents added. VLDL and LDL were precipitated out using a solution consisting of 4g (dodeca)-tungstophosphoric acid (BDH, Canada) dissolved in 16ml of a 1M sodium hydroxide (Fisher Scientific, Canada) and 84ml of deionized water (4% PTA) and a 2M magnesium chloride solution (2M MgCl2). The ratio of 4% PTA to 2M MgCl 2 was kept constant at 4:1 (v/v). The contents, in an Eppendorf tube, are lightly vortexed and allowed to cool at 4°C for 15 minutes before the 150 precipitate was pelleted by microcentrifugation (Eppendorf 5417C, Germany) at 6800 x g for thirty minutes at 4°C. The resultant supernatant was removed and run on the FPLC and the fractions were analyzed. The appropriate volumes of 4%PTA and 2M MgCl 2 were varied to achieve differing degrees of VLDL and LDL precipitation depending on the plasma lipoprotein characteristics of each pooled serum or plasma sample. 6.2.9 Tritiated cholesterol exchange into plasma lipoproteins The relative lipoprotein content of fractions obtained from FPLC can be determined using a readily exchangeable marker such as tritiated cholesterol which equilibrates into lipoproteins. Subsequent selective precipitation of apo B-containing lipoproteins, VLDL and LDL, using the method, can effectively allow for estimation of VLDL and LDL contribution to the absorbance measurement at 280nm. Plasma lipoproteins were labeled with 3H-cholesterol using the following procedure. Filter paper discs (5 mm diameter, Whatman International, Maidstone, Kent, U.K.) were blotted with 5ul 3H-cholesterol (Amersham Life Sciences, Little Chalfont, Bucks, England, lmCi/ml, 75Ci/mmole) and allowed to dry for 15 minutes at room temperature. This disc was placed in a 8mm x 75mm glass test tube with 350 ul of SCID mouse plasma, vortexed and incubated for 24 hours at 4°C. After the 24 hour incubation, 250 pi was combined with 100 pi of FPLC running buffer and loaded onto the FPLC. The elution profile was followed at 280nm and fractions assayed for 3H-cholesterol by liquid scintillation counting. It was determined from precipitation studies (Section 6.2.8) that 5ul of 2M MgCl 2 and 1.25ul of 4%PTA effectively removed apo B-containing lipoproteins, VLDL and LDL, from 250ul of plasma. An identical SCID mouse plasma sample was therefore incubated with tritiated cholesterol (5 pi) for 24 hours at 4°C then treated with the above 151 volumes of precipitation reagents and made up to 350ul This mixture was cooled to 4°C and centrifuged at 8000 rpm for 30 minutes. The supernatent from this sample was loaded onto the FPLC apparatus. 6.2.10 Detection of tracer molecules The liposomes can be detected with a non-metabolisable, non-exchangeable 3 H CHE radiolabel that was quantitated using in a liquid scintillation cocktail (Pico-Fluor™ 40, Packard Instruments) on a scintillation counter (Tri-carb 1900TR, Packard Canberra, Meriden, CT). The fluorescent, non-radioactive liposome tracer molecule, Dil , was detected using fluorimetric measurements (Perkin Elmer LS50, USA) of the probe in liposome membrane with the excitation wavelength maximum determined to be at 518nm using a slit width of 2.5nm and emission wavelength measured at 570nm over a 5nm slit width. 152 6.3 R E S U L T S 6.3.1 Identification and definition of lipoprotein components in SCID mouse plasma using DGUC and lipoprotein gel analysis The agarose gel electrophoresis stained for the SCID mouse plasma lipoproteins in different density fractions is shown in Figure 6.1. The lipoproteins are separated on the basis of their electrophoretic mobility with the cathodic dietary triglyceride transporters, chylomicrons, at or nearest to the point of application (origin) followed by cholesterol transporting beta lipoproteins, pre-beta lipoproteins and highly mobile, anodic alpha lipoproteins. Lipoprotein gels have been used to detect the presence of abnormal electrophoretic banding patterns in plasma samples from patients with cardiovascular disease (Boisvert, Spangenberg and Curtiss, 1995). The purpose of this study was to detect components of SCID mouse plasma by identifying the banding patterns from density gradient fractions. The lipoproteins run on lane 2 (d>1.063g/ml) are from the HDL and albumin density gradient fraction. In lane 3 (d=1.019-1.063g/ml), loaded with the IDL fraction there are no lipoproteins present. This might be explained by the fact that the mouse species is known to have very low levels of IDL lipoproteins, if any. More apparent, however, was the LDL lipoprotein fraction (d=l.006-1.019g/ml) in lane 4 of the gel electrophoresis. In lane 5 (d<1.006), lipoproteins of the beta electrophoretic mobility type or chylomicrons loaded from the VLDL density gradient fraction are present. Al l comparisons were performed against known standards diagrammed in the instruction booklet accompanying the lipoprotein gel kit from Ciba Corning. The results of this gel separation are confirmed by FPLC 153 Figure 6.1: 1% Agarose gel electrophoresis of SCID mouse plasma lipoproteins separated by density gradient ultracentrifugation. The original plasma (lane 1) shows heavy staining for alpha migrating lipoproteins. While in the HDL/albumin fraction (lane 2) there was obvious alpha migrating lipoproteins, an absence of any IDL (lane 3) lipoproteins and trace amounts of LDL (lane 4) lipoproteins were observed. The VLDL lipoprotein fraction (lane 5) contained beta migrating lipoproteins. Each lane was compared to the migration of known standards illustrated in the lipoprotein gel kit instruction booklet (Ciba Corning). 154 analysis of the respective fractions as evidenced by absorbance units (note corresponding scales) of the peaks as measured by A 2 8 0 in Figure 6.2. 6.3.2 Determination of FPLC elution profiles The elution profiles for the DGUC separated plasma fractions run on the FPLC column apparatus are shown in Figure 6.2. The FPLC column separates plasma components on the basis of their size. This figure illustrates that the VLDL lipoprotein component was eluted between 44-50ml. This population was large enough so that it was excluded from the highly cross-linked agarose gel matrix, i.e. it eluted at the void volume. Based on the exclusion limits for the first column, a Superose 6 (4 X 107 MW protein) and the second column connected in series, a Superose 12 (2 X 106 MW protein), VLDL (around lOOnm in diameter) eluted in the void volumes of both columns. The IDL was undetectable in SCID mouse plasma according to our absorbance measurements at 280nm, lipoprotein gel analysis and FPLC of the density gradient fraction. Similarly, the density gradient procedure did not allow for modest quantities of LDL to be resolved. The lipoprotein gel and FPLC analysis of the fraction isolated by DGUC would indicate the presence of a lipoprotein in this fraction. Based on the migration of this lipoprotein fraction on the gel and its elution on the FPLC, the presence of a preceding shoulder in the elution profile was most likely due to cross-contamination from the VLDL (beta-lipoprotein) fraction. Note how the band in lane four in Figure 6.1 runs similarly to the one in lane 5 but its intensity is much less than the latter. Similarly, on the FPLC, the LDL-sized particles were found to be in the HDL/albumin density gradient sample. The lipoprotein particles, classified on the basis of their relative densities, do exist in a distribution of sizes, 155 *UIUQ$Z (g) 3 D U B q j o s q y UIUQS^ © 3 D u e q j o s q y Figure 6.2: The D G U C fractions run on the F P L C and recorded as a chromatogram of absorbance measurements at 280nm. The points of elution on the FPLC apparatus for DGUC fractions from SCID mouse plasma were determined here. Two Absorbance (@ 280nm) scales are shown where an asterisks (*) denotes the scale being used for the particular elution profile is the one on the right. The VLDL fraction* (purple line) elutes at 44-52ml, while the lipoproteins in the IDL fraction* (green line) (minor contaminants from VLDL and LDL) was barely-visible. The LDL fraction (red line) contains fractions that correspond to a early eluting shoulder (43-48ml) and a peak (51-56ml). The lipoproteins in the HDL/albumin fraction with 3 major peaks at 50-57ml, 62-68ml and 70-80ml are assigned LDL/plasma proteins, HDL and albumin/plasma proteins, respectively. 157 however. The FPLC analysis of each DGUC fraction may therefore represent lipoprotein subspecies (e.g. HDL l 5 HDL 2 and HDL 3). Otherwise, it could be that there was a problem with DGUC and the LDL contaminated the HDL/albumin fraction. In the separation of the HDL/albumin fraction, the first peak (50-56ml) was determined to contain some LDL. The latter two distinct peaks (62-69ml and 71-82ml) resolved in the HDL/albumin fraction correspond to HDL and albumin along with other plasma protein components found in SCID mice, respectively. They are by far the largest and contain abundant amounts of protein (in terms of A 2 8 0 measuring the tyrosine content) found in SCID plasma. The cholesterol exchange results (Figure 6.3) indicate the presence of LDL (minor component) and plasma proteins (major component) in the second eluting peak in SCID plasma samples. The precipitation procedure discussed below was able to remove the apo B-containing lipoproteins, VLDL and LDL, from the sample which resulted in the loss of tritiated cholesterol eluting fractions, corresponding to the first two peaks for these lipoproteins in the FPLC elution profile. 6.3.3 Removal of VLDL from liposome and plasma mixtures prior to FPLC analysis The VLDL fraction co-elutes with liposomes in the void volume based on the FPLC elution profiles (Figure 6.4). In some cases, it may be desirable to eliminate the VLDL from the sample to be injected onto the FPLC as the VLDL might confound the results observed in the liposome FPLC fractions. For example, if preferential exchange of liposomal formulation components to VLDL occurs (or was suspected) then interpretation of results of the fractions corresponding to the first peak on the elution profile would be inconclusive. The modified method to selectively remove most of the 158 o oo o X 50 60 70 80 Elution Volume (ml) PH Q 2 o X U T3 o 00 o X < 30 40 50 60 70 80 Elution Volume (ml) PH Q s U -*-» to "3 x U T3 D 159 Figure 6.3: Determination of lipoprotein content in fractions separated by F P L C . Plasma lipoproteins were labeled with 3H-cholesterol for 24 hours and 250pl was combined with lOOpl of FPLC running buffer and loaded onto the FPLC (upper elution profile). Similarly, an identical sample was incubated for 24 hours, 250ul withdrawn, and subject to selective precipitation of apoB-containing lipoproteins using 5pl of 2M MgCl 2 and 1.25pl of 4%PTA for precipitation and run on the FPLC (lower elution profile). The elution profiles were followed at 280nm and fractions assayed for 3 H -cholesterol by liquid scintillation counting. The open circles represent the plasma lipoprotein elution profile as determined by absorbance measurements (left axis) and the filled circles represent relative cholesterol content in the fractions as determined by tritium counting of cholesterol (right axis). 160 *uiu()8£ <D 3 3 u e q j o s q y U1UQ8Z (g) 9Dueqjosqy Figure 6.4: FPLC Elution profiles. The elution profiles were obtained for the following samples: Liposomes (lumole total lipid, red line); Human IgG (30ug total protein, black line*); Liposomes (lumole total lipid) mixed with Human IgG (30ug total protein)(green line*); SCID mouse plasma (250u,l, blue line); SCID mouse plasma precipitated with 5pl 4%PTA and 1.25ul 2M MgCL, (pink line); and SCID mouse plasma and liposomes (lumole total lipid) precipitated with 5pl 4%PTA and 1.25(0,1 2M MgCl 2 (gray line). * Denotes: Absorbance is read off the right axis. 162 VLDL component out of the liposome and plasma sample mixture prior to FPLC with minimal loss to the liposome population relies on selective precipitation of apo B-containing lipoproteins, VLDL and LDL, by PTA and MgCl 2 . The nature of the VLDL lipoprotein fraction renders itself more susceptible to this precipitation method than other lipoprotein fractions and, as demonstrated in this paper, more susceptible than liposomes. The PTA/MgCl 2 precipitation procedure was tested on liposomes in plasma with the results as shown in Table 6.1. Optimal precipitation of the VLDL fraction was achieved with a 20% loss of liposomes by using 5ul of 4%PTA and 1.25ul of 2M MgCl 2 in 250pl of SCID mouse plasma and 1 pmole of lipid (liposomes) made up to a total volume of 350pl with HBS (Table 6.1). This precipitation protocol allowed for almost (>90%) complete removal of the VLDL fraction from the plasma sample when liposomes are excluded from the 35Out sample and would be valuable for studies excluding liposome component exchange into the VLDL. It was not necessary, however, to precipitate out the VLDL component to study liposome component exchange especially if radiolabel or fluorescent tracer molecules are employed. 6.3.4 Characterization of liposome separation from SCID mouse lipoproteins on the FPLC A full characterization of the separation of liposomes, SCID plasma, an example of a foreign molecule that is within the columns' agarose gel fractionation range, human IgG, or combinations thereof are illustrated in Figure 6.4. Liposomes were eluted between 45 and 51ml whereas the major human IgG peak was found to elute between 66 163 1/5 o o a S o Q s o a U co c a > u <u 4) H ,3 E and 72ml although the presence of human IgG was found over a wider range of 50 to 78ml. A mixture of liposomes and human IgG was also analyzed on the FPLC. They were effectively resolved on the FPLC. The PTA/MgCl 2 precipitation procedure was repeated for SCID mouse plasma and was found to be satisfactory in removing the VLDL component from the sample and therefore can used to remove the majority of the VLDL component in plasma containing liposomes prior to FPLC analysis as summarized in Table 6.1. 6.3.5 Evaluating the general application of FPLC and modified VLDL precipitation procedure on human plasma and CD1 serum samples The separation of many serum and plasma components was possible by FPLC. The elution profiles for human plasma (Figure 6.5) and CD1 mouse serum (Figure 6.6) illustrate not only liposome separation from plasma or serum components but also the difference between species plasma samples. The most profound difference was in the elution of HDL in human versus mouse samples. The resolution between the HDL and the albumin/protein peak was poor for human samples but well-resolved in mouse samples. The first peak, corresponding to VLDL, between 44-49ml was effectively precipitated out by PTA/MgCl 2 in both human plasma and CD1 mouse serum. A second elution profile results from mixing liposomes and serum ex vivo for 10 minutes before performing FPLC. The fractions were assayed for tritium counts and all recovered counts were found to correspond to the first peak in the elution profile. This result was consistent for all injected samples that contained radiolabeled liposomes. The liposomes were therefore found to be intact and stable in serum over the entire handling time (10 165 (uiurjK)33irec4JOsqv Figure 6.5: F P L C separation of liposomes and components in human plasma. The FPLC was used to separate liposomes mixed with human plasma before (dark pink*) and after (brown*) VLDL precipitation with reference to liposome recovery on the left y-axis. The elution profiles based on absorbance at 280nm were also obtained for liposomes mixed with human plasma (red line); human plasma (green line); and human plasma in which VLDL had been precipitated (blue line). 167 (uiu()8£) a a u B q j o s q y ( U O J P B J J / j e i l l U I J O %) A\l9A0Da*[ 9UIOSOdn % Figure 6.6: F P L C separation of liposomes and components in C D 1 mouse serum. The FPLC was used to separate liposomes mixed with mouse serum before (dark pink*) and after (brown*) V L D L precipitation with reference to liposome recovery on the left y-axis. The elution profiles based on absorbance at 280nm were also obtained for liposomes mixed with mouse serum (red line); mouse serum (green line); and mouse serum in which V L D L had been precipitated (blue line). 169 minutes or up to 1 hour for samples subject to PTA/MgCl 2 precipitation) and during the 4 hour FPLC run. The radiolabel marker indicated that the liposomes are eluted within 2 hours after the sample was loaded onto the column. Hence, the versatility of this separation technique applied in the separation of various biological fluids was demonstrated in the separation of SCID plasma lipoproteins and liposomes. The characterization of liposome separation from SCID mouse plasma lipoproteins could be generalized to liposome separation from many types of plasma, serum or other biological fluids. 6.3.6 Comparison of radioactive cholesteryl hexadecyl ether versus non-radioactive fluorescent Dil liposome markers on the FPLC A hypothesis stating that the lipophilic fluorescent marker, Dil , would be a suitable marker for liposomes was tested by FPLC separation technology. Although the Dil probe can be incorporated into pre-formed liposomes (Claassen, 1992), it was co-lyophilized along with our other lipids. The fluorescent probe present at 0.5mole% quantities relative to total lipid in the liposome sample was compared to a non-exchangeable non-metabolisable radiolabel marker, 3H-CHE (0.33mCi/mmole of lipid) (Derksen, Morselt, and Scherphof, 1987). It was determined from the elution profile of these fluorescently labeled liposomes that the Dil lipophilic probe does not readily exchange out of our liposome formulation and was stable in SCID plasma with negligible quantities in the separated lipoprotein fractions (Figure 6.7). As a tracer of liposomes in terms of amount of liposomal lipid, the amount of Dil probe as assayed for each fraction slightly overestimates by approximately 25% the amount of liposomal lipid in 170 Elution Volume (ml) Figure 6.7: Elution profiles for radioactive and non-radioactive liposome tracers. Dil (squares) and H-CHE (circles) liposomes separated from SCID plasma lipoproteins (diamonds) by FPLC. 171 comparison to the radiolabel marker in the peak fractions. The margin of error may be due to concentration, volume or the environment in which the probe was assayed relative to the standard samples gauged for specific fluorescence. The effect of these variables in combination could be non-linear. Indeed, quantification based on the fluorescence readings at excitation maximum 518nm and emission maximum 570nm in samples containing plasma resulted in some error (unpublished data). 172 6.4 DISCUSSION Liposomes are microscopic spheres composed of one or more concentric bilayers of lipid surrounding an aqueous core (Bangham, 1968). The vast amount of research on liposomes spans from investigations studying the transfer and exchange of phospholipids from the bilayer membrane into lipoproteins (Damen, Regts and Scherphof, 1981) to devising ways to increase the liposome circulation lifetime for purposes of drug delivery (Allen, 1994). Their application as drug delivery vehicles is of increasing clinical importance with liposomal formulations of amphotericin B (Ringden et al., 1991) and doxorubicin (Working et al., 1994). Drugs are encapsulated into liposomes because liposomes are found to accumulate at sites of infection and disease. The localization of any drugs associated with the liposome may therefore be advantageous in the treatment of the infected or diseased site. The separation of liposomes from red blood cells continues to be efficiently carried out by centrifugation. In this chapter, a description of the use of fast protein liquid chromatography (FPLC) in the separation of liposomes from plasma lipoproteins was introduced. FPLC is a method used to separate macromolecules by passing them through a column of support media capable of filtering molecules based on their size. Whereas the common technique of high resolution liquid chromatography (HPLC) relies on modified silica as a physically strong media support, the advent of highly cross-linked agarose has made FPLC possible (Andersson et al., 1985). Without adequate physical cross-linking of agarose polymers, the gel would be too soft be compatible with a high pressure system such as the apparatus commonly used in HPLC. Two types of highly cross-linked agarose resins, Superose 6 and Superose 12 in FPLC that were shown to 173 offer superior separation of proteins, were used in this experiment (Eggesbo et al., 1996). The advantages of FPLC over conventional column chromatography are: shorter run times, higher resolution, smaller sample volumes necessary and less packing material required. Previous studies have investigated the use of FPLC in lipoprotein analysis (Vedie et al., 1991, Lehmann, Bhargava and Gunzel, 1993 and Hjerten and Kunquan, 1981). More recently, FPLC was used to study the differences in the relative amounts of each of the lipoprotein components in relationship to dietary changes and genetic differences of subjects (Averna et al, and Lau et al., 1995). Fast high performance liquid chromatography was found to successfully separate liposomes from plasma or serum components. This technique utilized highly cross-linked agarose in filtration chromatography. Optimal separation of the samples was achieved by using a combination of gel matrix types, Superose 6 and Superose 12, in different size columns. In contrast to existing methods such as gravity filtration methods of separation which are both time and material intensive, the procedure here required less than 4 hours to resolve the liposome and plasma/serum components in our samples and less than 130ml of Superose gel matrix. In addition, two-column FPLC increased resolution of the lipoprotein components especially between the HDL and lipoprotein remnants, albumin and a,-glycoprotein (Marz and Gross, 1986). The uncertainty generated in the gel (Figure 6.1) of lipoprotein fractions from density ultracentrifugation upon comparison to FPLC elution profiles of the same samples can be explained by exploring the cholesterol content in mouse plasma fractions as analyzed by FPLC. Radiolabeled cholesterol incubated with SCID plasma for 24 hours yielded the results shown in Figure 6.3. The elution profile for SCID plasma and 174 corresponding cholesterol content associated with each fraction is shown in Figure 6.3a. VLDL and LDL particles both contain cholesterol which was precipitated out as shown by the lack of cholesterol in the elution profile corresponding to these fractions (Figure 6.3b). The second peak major peak in the SCID mouse elution profile, therefore, represents LDL and other plasma proteins. In addition, the SCID mouse lipoprotein studies indicate that the earliest eluting fraction on the FPLC was VLDL and the other major SCID mouse lipoprotein was HDL eluting separate from and later than VLDL on the FPLC. The resolution of the various lipoprotein fractions in addition to the separation of these two major SCID mouse lipoprotein components would be advantageous for studying the effects a drug formulation has on the characteristics of each lipoprotein fraction especially in blood, serum or plasma samples in other species of animals. Notwithstanding, the goal of separating liposomes from lipoproteins was achieved using FPLC as co-eluting VLDL can be separated prior to FPLC using established PTA/MgCl 2 precipitation procedures. In the current research, liposome-plasma/serum samples were efficiently separated by FPLC. The FPLC separation technique was also used to determine if liposomes could be separated from free human IgG antibody. The antibody, having being smaller in size and molecular weight compared to a liposome, would be partially retained by the two columns (see fractionation range in Materials & Methods) and elute later than the liposomes. The results generated (Figure 6.4) demonstrated that the FPLC technique worked for the separation of this liposome-antibody mixture. Secondly, the method developed in this chapter will have a direct impact in the study of drug exchange, 175 antibody stability and component transfer from liposomes when they are introduced into a biological fluid such as plasma or serum. In one application of the technique is the introduction of a non-radioactive fluorescent probe, l,l'-dioctadecyl-3,3,3',3'-tetramethyllindocarbocyanine perchlorate (Dil), as a possible candidate for a liposome marker. The FPLC can then quantitatively compare its ability to remain associated with a liposome by the lack of exchange of this marker into plasma components. Dil was compared to a well-established non-exchangeable, non-metabolisable radiolabeled liposome marker, tritiated cholesteryl-hexadecyl ether [3H-CHE]. The positively charged, lipophilic fluorescent probe, Dil , was incorporated into the liposome bilayer at 0.5 mole% of total liposome lipid in the presence of SCID mouse plasma. The fluorescent probe was assayed for in the fractions eluted from the FPLC columns (Figure 6.7). The fluorescence was restricted to fractions in which the liposome eluted and corresponded to the liposome lipid content in these fractions as determined by tritiated cholesteryl hexadecyl ether. Therefore, for the purposes of demonstrating where the liposome elution peak arises in our study, the Dil marker was found to be a suitable marker. The quantitation of lipid as assayed by fluorescence measurements, however, requires further study in order to correct for sample differences. For example, the ratio of tritium counts to fluorescence units prior to mixing and separation of liposomes and SCID mouse plasma was found to be 84.8 while the average ratio for peak liposome fractions was 67.6. This represents a 20.3% drop from the original tritium count to fluorescence unit ratio. This result could either mean a paradoxical loss of the non-metabolisable non-exchangeable tritium marker CHE or a flaw in the fluorescence readings for the fractions eluted off the FPLC. The effects leading to this discrepancy should therefore be further investigated. Nonetheless, 176 Dil should be viewed as a very good qualitative marker of liposomes and its quantitative aspects should hopefully be resolved in the near future. In conclusion, the use of the FPLC technology as a tool for separating liposomes from lipoproteins sets an important stage from which many questions regarding their interactions can be answered. For liposomology, the transfer of a liposome component into one or more of the plasma lipoproteins (or lack thereof) may lead to important insights on molecule exchangeability and rates. For lipoprotein research, the changes in levels of particular lipoprotein components as a result of liposome drug therapy may be of interest. FPLC provides a tool to study a wealth of questions regarding the components associated with liposomes that have been injected into the circulation. FPLC was a useful technique for the separation of liposomes from lipoproteins. 177 C H A P T E R 7 A N T I B O D Y - C O U P L E D L I P O S O M E STABILITY AND BIODISTRIBUTION IN SCID M I C E The delivery of an antibody using liposomes was characterized in a model system utilizing a conventional antibody, human IgG. This antibody was chosen instead of an antibody with immunosuppressive properties because the question being posed was whether the antibody remains attached to its lipid carrier. To manufacture and isolate an antibody with immunosuppressive properties would be an exhaust of time and finances if an antibody chemically attached to the liposome surface did not remain with the liposome in vivo. Hence, the feasibility of delivering therapeutic antibodies can be established by initially demonstrating whether the antibody-coupled liposome attachment is stable in circulation. Ultimately, the methods could be generalized for the attachment of other proteins, antibodies or therapeutic agents such as one of the immunosuppressive antibodies introduced earlier in this thesis. In order to evaluate the stability of the attachment between the human immunoglobulin G (hlgG) antibody and the liposome, hlgG antibodies were covalently attached to liposomes using a novel carbohydrate coupling procedure to yield a 20ug of hlgG/pmole of lipid liposome formulation which was then injected into SCID (severely combined immunodeficient) mice. The elimination pharmacokinetics of hlgG antibody and liposomes from plasma was identical over a 4 hour-time course. Fast protein liquid chromatography (FPLC) was used to separate liposomes from plasma lipoproteins to ascertain that the hlgG antibody remained with the liposome. This analysis of the plasma resulted in co-elution of the peaks for hlgG antibody and liposomes. Taken together with 178 qualitative dot-blot and Western transfer blot analysis for the hlgG antibody in plasma, the results have shown that the hlgG antibody covalently attached to liposomes was not lost from the liposome while in circulation. In the biodistribution studies, both hlgG and liposomes preferentially accumulated in the spleen and liver. Over the four hour-time course, however, hlgG levels decreased in these tissues while liposome levels to these tissues increased. Therefore, the fate of hlgG does not parallel that of liposomes after they exit the circulation. Upon further histological analysis, evidence of co-localization of hlgG antibody and liposomes in the spleen and liver was revealed. In conclusion, while the hlgG antibody-coupled to the liposomal carrier remained stable in circulation over a four hour-time course, the biodistribution into different tissues demonstrated different degrees of antibody and liposome accumulation not consistent with initial hlgG antibody to lipid ratios. 7.1 INTRODUCTION In the delivery of drugs using liposomes, the question of how much of the drug, when injected into circulation, remains with the liposome has been posed. Similarly, because of extensive research into lipid packing and interactions with neighboring lipids in the liposome membrane bilayer, lipids have been known to also exchange within and out of the bilayers of liposomes. Advances in targeted liposomes and stealth liposomes have added to the breadth of this question. In targeted liposomes, receptors for cell surface molecules or antibodies have been attached in novel ways to the liposome surface. The degree and stability of this attachment predict how effective the targeted delivery works. If the antibody is lost due to chemical bond breakage or exchange from 179 the liposomal carrier, the chance of attaining its target as an entity attached to a liposome decreases. In the same respect, monosialoganglioside (GM 1) and polyethylene glycol (PEG) molecules in stealth liposomal formulations have a tendency to exchange out of the liposome over time at rates dependent on the length of the acyl chains anchoring the molecule to the bilayer (Maruyama et al, 1992 and Bedu-Addo et al, 1996). Effective localization of the stealth liposomes relies on increased circulation lifetime of the liposome which in turn is dependent in part on the retention of the protecting molecule. The question of how to measure the amount of any liposomal formulation component remaining with the liposome in circulation not bound by plasma lipoproteins has posed a technical challenge for scientists studying liposomes. In the preceding chapter, fast protein liquid chromatography (FPLC) was found to separate liposomes from plasma lipoproteins. The FPLC separation technique developed in Chapter 6 was used in this study to demonstrate the stability of antibody-coupled liposomes in vivo. A method to covalently attach antibodies to the liposome surface in an orientation-specific manner that theoretically results in improved epitope binding to the receptor was employed (Ansell et al, 1996). The stability of a novel carbohydrate coupling linkage joining a conventional antibody, human IgG, to fluorescent Dil labeled DSPC/CHOL/MPB-DSPE/DSPE-PEG 2 0 0 0 (52:45:2:1) liposomes was examined. The characterization of stability of the antibody-coupled liposome formulation was followed in plasma elimination and biodistribution studies carried out in SCID mice. In addition, the F c specific carbohydrate chemistry used to link the antibody to the liposomal surface might inhibit F c recognition and binding by F c receptors on T cells. OKT3, a humanized monoclonal antibody used in the management of organ transplantation (Boyd and James, 180 1989 and Dantal and Soulillou, 1991), when administered causes cytokine release syndrome which is thought to be mediated through antibody F c binding. Specifically, the early toxic effects of antibody administration are mediated through complement activation requiring interaction with at least three antibody components: a disulfide bond in the N-terminal end of CH2, carbohydrate moieties and paired CH3 domains (Brekke, Michaelsen and Sandlie, 1995). Therefore, the physical and chemical modification of carbohydrate moieties might reduce the relative toxicity of the administered antibody. In this chapter, characterization of the in vivo stability of a liposome formulation containing DSPE-MPB reactive to an activated human IgG is described. In demonstrating the formation of stable antibody-liposome formulations for use in vivo, an extension of this technology to the general coupling of antibodies to liposomes will hopefully be realized. In particular, the practical application of this research in liposomal delivery of therapeutically immunosuppressive monoclonal antibodies like OKT3 should be considered. The effect of immunosuppressive anti-CD3 antibodies has been extensively investigated. In particular, a hamster antimouse anti-CD3, 145-2C11 has been tested in murine allograft rejection studies (Abbs et al., 1994, Vossen et al., 1994a and 1995a,b). 181 7.2 MATERIALS AND METHODS 7.2.1 Lipids and Chemicals 1,2-Distearoyl-sn-Glycero-3 -Phosphatidylcholine (DSPC) and 1,2-Distearoyl-sn-Glycero-3-Phosphatidylethanolamine-N-(Poly[ethyleneglycol-2000] (DSPE-PEG2 0 0 0) were obtained from Avanti Polar Lipids Inc. (Alabaster, AL, USA). Cholesterol (CHOL) was obtained from Sigma Chemical Co. (St. Louis, MO, USA) and 4-(4-/V-Maleimido-phenyl)-butyryl-l ,2-Distearoyl-sn-glycero-3-Phosphatidyl-ethanolamine sodium salt (DSPE-MPB) from Northern Lipids Inc. (Vancouver, B.C., Canada). Cholesteryl hexadecyl ether [cholesteryl-1,2-3H(N)] (3H-CHE) (NEN Research Products, Boston, Massachusetts, USA) and l,r-dioctadecyl-3,3,3',3'-tetramethyllindocarbocyanine perchlorate (Dil, Molecular Probes, Eugene, Oregon, USA) served as tracer molecules in the liposome. 7.2.2 Preparation of large unilamellar vesicles Fifty microcuries of 3H-CHE and 0.5mole% of Dil lipophilic fluorescent probe were added to a DSPC/CHOL./DSPE-PEG2 0 0 0/DSPE-MPB (52:45:2:1 m/m) lipid mixture. Lipids and probes were dissolved in benzene/methanol (95:5 v/v) and dried down on a lyophilizer (Virtis, Leybold Vacuum Products, USA) under high vacuum (60-100 mtorr) for five hours. The dry lipid film was hydrated with 2ml of HEPES buffered saline (HBS, 150mM NaCl, 20mM HEPES, pH 7.4), transferred into a cryovial and frozen in liquid nitrogen and thawed in a 56°C water bath for 5 cycles. The MLVs were subsequently size-reduced through two stacked lOOnm pore size filters using an extrusion procedure as published by Hope et al. (1985). The resultant lipid particles were analyzed 182 by a Nicomp particle sizer and determined to have a mean diameter of 140-160nm and a specific radioactivity of 0.33pCi [3H]/pmole total lipid. 7.2.3 Iodination of Human IgG Three iodobeads (Pierce, Rockford, Illinois) made of N-chloro-benzenesulfonamide derivatized onto non-porous polystyrene beads were washed with HEPES buffer saline (HBS, 150mM NaCl, 20mM HEPES, pH 7.4) and added to 125ul of 40mg human IgG/ml (5mg human IgG total) and 430pl of HBS. One millicurie of carrier-free sodium 1 2 5I (Amersham, Mississauga, Ontario) was carefully added and the iodination was halted after 6 minutes by running the sample on a 15ml Sephadex G-50 gel exclusion column equilibrated and run with HBS collecting fractions of 0.8ml each. Fractions 13-16 corresponding the Human IgG were pooled and the protein concentration was determined according to A 2 8 0 measurements (A 2 8 0 / 1.35 = protein concentration (mg/ml)). The specific radioactivity of the pooled 125I-human IgG was determined using gamma counting (Section 7.3.8). 7.2.4 Antibody Coupling to Large Unilamellar Vesicles Human IgG (Miles Laboratories, Gibco) was covalently attached to the surface of liposomes at room temperature using the procedure described by Ansell et al. (1996) modified for the purposes of this study. Briefly, one milliliter of 1 2 5I human IgG (2494 CPM/ug protein, specific radioactivity) made up to 22mg/ml with human IgG was oxidized with sodium metaperiodate (2mg/ml of antibody), Sigma Chemical Co., St. Louis, MO, USA) for 90 minutes in the dark while stirring and then excess sodium metaperiodate was removed using a 15ml Sephadex G-50 (Sigma Chemical Co., St. Louis, MO, USA) column in sodium acetate buffer (SAS, lOOmM Sodium acetate 183 (trihydrate) 13.61 g/1, 50mM NaCl 2.92g/l, pH-4.5) and 0.7ml fractions were collected on a Micro-fraction collector (Gilson, Middleton, WI). The fractions containing high protein content (A 2 8 0 > 1) were subsequently pooled and reacted with 3-(2-pyridyldithio)-propionyl hydrazide (PDPH, Pierce, Rockford, Illinois) cross-linker at 0.9mg per ml of pooled human IgG for 5 hours in the dark while stirring. The reaction was terminated by passing the reaction mixture on a 30ml Sephadex G-50 column using SAS and 0.7ml fractions were collected. The crosslinker attached to the human IgG was then reduced using dithiothreitol (DTT, Sigma Chemical Co., St. Louis, MO, USA) at 3.86mg DTT per ml of pooled fractions with A 2 8 0 > 1 for 15 minutes. This reaction was stopped by passing the reaction mixture down a 50ml Sephadex G-50 column equilibrated in HEPES buffered saline (HBS, 150mM NaCl, 20mM HEPES, pH 7.4). The amount of protein recovered was quantitated using the following equation: Amount of protein (mg/ml) = A280/1.35. The antibody was coupled to MPB-DSPE on the liposomes at 75u,g of protein per umole of total lipid in the liposome formulation for 16 hours with stirring. The resultant mixture was loaded onto a 60ml Sepharose CL4B column and HBS was used to elute the antibody-attached liposomes. The amount of protein was either determined by a modified Micro BCA assay (Pierce, Rockford, Illinois, USA) or by gamma counting based on specific radioactivity of the 125I-human IgG (Section 7.3.8). 7.2.5 Plasma Elimination and Biodistribution of Antibody-Coupled Liposomes SCID Rag 2 (severely combined immunodeficient) mice (B.C. Cancer Agency and Research Center Joint Animal Facility, Vancouver, B.C.) were used in this study (Bosma, Custer and Bosma, 1983). These animals were chosen based on their inability to produce whole antibodies which were found to interfere with secondary antibody 184 detection of antibody-coupled liposomes (personal communication). The animals weighed 20 grams on average and were supplied with food and water ad libitum. The mice were randomly divided into 4 groups of 4 in the experimental design which called for plasma and tissue samples to be collected 0.5, 1, 2 and 4 hours after a tail vein injection of 250pl of an antibody-coupled liposome formulation (lOmM lipid). The mice were sacrificed with carbon dioxide. Blood was taken by heart puncture employing a 1ml insulin syringe and cervicals dislocated before tissue collection. The blood samples of the above were spun down in 500ul plasma Microtainers (Becton Dickinson, USA) at 2000 rpm on a tabletop centrifuge (Centronics S-103 NAR, Japan) at room temperature for 10 minutes. The plasma was analyzed to determine plasma elimination characteristics of the liposome formulation and was also used for FPLC analysis as described below. Tissues were collected for quantitation of liposomal lipid and antibody (lung, heart, liver, kidney and spleen) and histochemical analysis (liver, kidney and spleen). Mouse tissues collected for quantitation of liposomal lipid (3H-CHE) and human IgG were processed as follows. Tissue samples were weighed and gamma counted (AutoGamma 85650, Packard Canada) for human IgG antibody content as described in more detail in the Radiochemical and Fluorimetric Analysis section below. SCID mouse tissue levels of liposomal carrier were determined by homogenization (20% liver homogenate, 10% others, diluted in HBS) using a Polytron tissue homogenizer (Brinkmann Instruments, Rexdale, Ontario, Canada) and four hundred microliters of each tissue homogenate then digested in 500ul of 0.5M Solvable (NEN Research Products, Boston, Massachusetts, USA) at 60°C for 2 hours. The samples were allowed to cool to room temperature. Fifty microliters of EDTA, lOpl of 2M HC1 and 200ul of 30% H 2 0 2 were added to decolorize 185 before addition of 5ml of Ultima Gold scintillation cocktail (Canberra Packard, Mississauga, Ont.). The samples were then subjected to liquid scintillation counting. 7.2.6 FPLC Instrumentation and Analysis of Fractions The FPLC set up consisted of a pump (LKB Brommer 2150, Germany) hooked up to a BioRad Econo System Controller running a Pharmacia Superose 6 H.R. (10mm diameter X 30cm length = 23.56mL) column and a Pharmacia Superose 12 H.R. (16mm diameter X 50cm length = 100.5mL) column in series. The flow rate was set at 0.5ml/min at 3 bar pressure run by the pump for a total run time of 240 minutes. (Conversions: 1 bar = 0.069 p.s.i. = 1 atmosphere = O.lMPa = 760mmHg). The BioRad Econo System Controller (BioRad, Canada) was set up to collect 60 fractions of lmL each. The running buffer (mobile phase) was 150mM NaCl, lOmM Tris and 0.03% NaN 3. Plasma samples (25Oui each) were made up to 35Oui with running buffer and injected into the FPLC loading coil (capacity of 500ul). The eluting fractions were monitored by a spectrophotometer attached in series to measure the absorbance of the fractions at 280nm. Fraction collection was set to begin at 70 minutes and end at 190 minutes (i.e., elution volume collection between 36 and 95 ml). The fractions were characterized using radiochemical counting, fluorimetric analysis and dot blotting as described below. 7.2.7 Analysis of FPLC fractions using Qualitative Dot Blotting Dot blots of the FPLC fractions for the presence of human IgG were performed according to Immuno-Blot® assay kit instructions using a 96 well sandwich chamber (Bio-Dot apparatus, BioRad Laboratories) outfitted to an aspirator that draws 50ul samples through a 0.45um thick polyvinylidene chloride (PVDF) membrane (Immobilon 186 P, Millipore Corp., Medford, MA). Briefly, the membrane was incubated with goat anti-human IgG secondary antibody linked to alkaline phosphatase. After washing the PVDF membrane of excess unbound secondary antibody, the presence of human IgG was visualized using 5-bromo-4-chloro-3-indolylphosphate p-toluidine salt (BCIP), an alkaline phosphatase substrate that was added to be cleaved by the enzyme to yield a blue chromogenic product. In turn, this product reduces nitroblue tetrazolium chloride (NBT) to form an insoluble purple precipitate (Pierce, Rockford, Illinois, USA) that qualitatively identified human IgG as present (dark dot) or absent (faint dot) in the FPLC fractions tested. 7.2.8 Western Gel and Transfer Blot Analysis Protein gel electrophoresis was performed to identify human IgG in SCID mouse plasma. Samples were run on a 10% BIS/acrylamide resolving gel over a 5% BIS/acrylamide stacking gel (Mini-Protean Gel Apparatus Kit, BioRad Laboratories, Hercules, CA). Samples consisting of human IgG, Dil labeled liposomes, human IgG-Dil liposomes and SCID mouse plasma were heated to 95°C in reducing buffer for 4 minutes, prior to loading (10-20ul each) on the gel. High molecular weight markers (Amersham) were also run. Samples were run on duplicate gels for 95 minutes at 150 volts. One gel was stained with 0.1% Commassie blue (BioRad, diluted in 10% acetic acid, 40% methanol and distilled water) while the duplicate was subjected to transfer blot analysis using polyvinylidene chloride (PVDF) membrane (Immobilon P, Millipore Corp., Medford, MA). The transfer gel was blotted for 1 hour at 100 volts according to Mini-Protean Transfer Blot instructions (BioRad Laboratories) in a cell apparatus, employing a 25mM Tris, 192mM glycine and 20% methanol (v/v) running buffer. 187 Human IgG antibody was visualized with a goat anti-human IgG-alkaline phosphatase secondary antibody in a similar fashion for visualization of dot blots described above. 7.2.9 Radiochemical and Fluorimetric analysis The presence of radiolabeled human IgG was determined by 1 2 5I gamma counting (AutoGamma 85650, Packard Canada). The rate of 1 2 5I radioactive decay to the nearest day was factored into the specific radioactivity of the iodinated antibody when measurements were made. Al l samples were placed into 16 x 100mm glass test tubes for gamma counting. Dil as a fluorescent, non-radioactive liposome tracer molecule, evaluated in Chapter 6 was used to track the liposome. Fluorimetric measurements (Perkin Elmer LS50, USA) of the probe in the liposome membrane were taken at an excitation wavelength maximum of 518nm (2.5nm slit width) and an emission wavelength maximum measured at 570nm over a 5nm slit width for 1ml samples. The liposomes were also detected using a non-metabolisable, non-exchangeable 3H-CHE radiolabel that was quantitated by liquid (Pico-Fluor™40, Packard Instruments) scintillation counting (Tri-carb 1900TR, Packard Canberra, Meriden, CT). Each sample was placed in a 7ml glass scintillation vial, immersed in 5ml of scintillation cocktail, capped and vortexed. The samples were placed in the liquid scintillation counter and counted for 2 minutes each. The spillover of gamma counts into the tritium channel was minimized by having excess tritium counts in the sample and was further corrected for by using a standard curve of gamma counts that were run under liquid scintillation conditions not having tritium label. Therefore, twenty-five microliters of each SCID mouse plasma was analyzed on the gamma counter, made up to one milliliter with HBS 188 for fluorimetric measurement and this whole sample analyzed for tritium counts with 5ml of scintillation cocktail. 7.2.10 Histochemistry Tissue samples of liver, kidney and spleen were harvested from animals and preserved in 10% formalin. The tissues were mounted on cork with water soluble tissue embedding compound (Tissue-Tek, Miles Laboratories) and frozen in a dewar of 2-methylbutane cooled in liquid nitrogen for 1 minute before 12 micrometer thick sections were cut at -20°C on a cryotome (Micron, Heidelberg). Tissue sections were mounted on chrome alum-coated slides. Slides were washed with PBS, blotted dry and treated with a goat anti-Human IgG-FLTC (heavy and light chains) secondary antibody (Caltag Laboratories, Burlingame, CA) diluted with PBS and 0.1% Triton (l:100v/v). The fluorescence probes, l,r-dioctadecyl-3,3,3',3'-tetramethyllindocarbocyanine perchlorate (Dil) and fluorescein isothiocyanate (FJTC) were visualized on a Zeiss epifluorescent microscope (Axiophot, Switzerland) using rhodamine and FLTC optics, respectively for dual label visualization. The Dil (red channel) and FLTC (green channel) fluorescence was recorded as prints on Snappy software (version 2.0, Play Incorporated, www.play.com). 189 7.3 RESULTS 7.3.1 Plasma Elimination Studies A liposome formulation was designed and tested for its ability to stably carry an antibody while in circulation. The liposome components were selected based on several criteria. These criteria took into account the need for a stable liposome, relatively longer circulation lifetimes and stability of the modified lipid anchor to the liposome surface. The lipid components in their relative molar ratios, DSPC/CHOL/DSPE-PEG 2 0 0 0/DSPE-MPB (52:45:2:1) were determined to yield liposomes of 140-160nm in mean diameter. The antibody-coupled liposome formulation (Figure 7.1), containing several probes, was injected into SCID mice. The probes were 3H-CHE and Dil liposome markers and 1 2 5I-human IgG for the antibody. The specific radioactivities of the 1 2 5I and tritium radiolabels were determined to be 2494 CPM/ug human IgG antibody and 779,860 DPM/mmole lipid, respectively. A quantitative measurement for the Dil probe resulted in a specific fluorescence 460 relative fluorescence units for every 50 nmoles of lipid. Liposomes have been previously fluorescently tagged using either fluorescein-DPPE (Singh et al, 1996) or Dil (Claassen, 1992 and Litzinger et al, 1993). The results of radiolabel and Micro BCA protein assay of the coupled formulation yielded a protein to lipid ratio of 22ug human IgG/umole of lipid. This represents a 29% efficiency in coupling of the antibody to the liposome surface. Although coupling efficiencies of as high as 40%) can be achieved with other antibodies using this method, the level attained in this experiment for this specific antibody was modest (Ansell et al, 1996). The SCID mice were given a 250)0.1 injection of a lOmM total lipid liposome formulation (22ug human IgG/umole lipid) by the tail vein and the clearance characteristics were followed 190 O i T3 cu CO 1 a 9 1 N 0 u X o ; C JS } 'S 0 = 0 1 c < • X z 1 CU z L . 3 II m 11 I 0 c O o s P c 3 cu 80 C _ X B "C S jg c cu m ' E 3 * CU C J 3 o s o. o x 3 S 8 £ 1 o g O a) .9- E o a— <u Si X o 0 E « 3 3 DO -a cU -g s c -° .3. 1-U ° X a u o 5 cu <u c > 0 •a fi cu 3 EG o ' 3 <-bO 53 «< — E £ <u a cu -J3 T3 •£ J2 = 2 E o y X5 T3 O D. CU * -£ ~ CU 2 3 <S a 3 .s W CU CU C/3 g Q o ° ~ o 2 11 cu ^ .22 o c « = xV CU 2 x "O X o d d Circulation Time (hours) Figure 7.2: The plasma elimination of Antibody coupled liposomes. The plasma clearance characteristics of human IgG antibody coupled to D S P Q C H O L / D S P r > P E C ^ o o o ^ P E - M P B (52:45:2:1) liposomes according to radiolabel and fluorescent tracer molecules in the formulation. The white squares represent the liposome content in plasma as determined by fluorescent probe D i l assay while the purple circles represent the same liposome as determined by tritium radiolabel analysis of plasma samples. The blue diamonds represent antibody clearance according to gamma counting of I human IgG antibody. 192 over a four hour-time course (Figure 7.2). The samples were first analyzed for gamma counts followed by fluorescence intensity measurements before being further processed for liquid scintillation counting for tritium counts. At the very first time point, 30 minutes post-injection, roughly 50% of the liposome formulation (tritium counts and Dil fluorescence) was cleared from plasma and begins to level off to 32% of the injected dose still in plasma circulation at 4 hours. As shown in Figure 7.2, the clearance of the antibody (125I gamma counts) mirrors that of the liposome. The ratio of protein to liposome lipid was statistically unchanged from time zero, 22.25 ± 1.90 ug/umole to time four hours, 20.98 ± 1.93 ug/umole. 7.3.2 FPLC elution profiles, Dot Blot and Western Transfer Blot Analysis The plasma isolated from these mice at various times within the 4 hour study was subjected to FPLC analysis. The FPLC can readily separate liposomes from the majority of the lipoproteins found in plasma (Chapter 6). In addition, non-covalently attached human IgG antibody was found to resolve in a separate peak away from the liposome elution peak when run on Superose 6 and Superose 12 FPLC agarose gel columns. Whether the coupling procedure used can withstand the interactions occurring in vivo in plasma was investigated. If the liposome formulation were unstable, the probes, radiolabel and fluorescent markers for antibody and lipid would neither co-migrate nor elute in similar fractions on the FPLC apparatus. The results of FPLC analysis of plasma taken from mice at 0.5, 1, 2, and 4 hours post-injection are illustrated as elution profiles in Figure 7.3. The lipid was tracked in terms of radiolabel counts as well as fluorescence intensity for 1ml fractions that were collected while the human IgG antibody was tracked quantitatively using radiolabeled 125I-human IgG and qualitatively using dot-blot assay 193 200 0.00 35 40 45 50 55 60 65 70 75 80 85 90 95 Elution Volume (ml) 200 175 150 125 c o U 7 5 -a '§. 50 25 \ I ^ I VI i v A 35 40 45 50 55 60 65 70 75 Elution Volume (ml) 2.00 1.75 1.50 1.25 1.00 0.75 0.50 0.25 0.00 85 90 95 194 195 Figure 7.3: F P L C elution profile of Antibody-coupled Liposomes. The FPLC plasma elution profiles for SCID mouse plasma lipoproteins (orange triangles), liposomes as determined by fluorescent Dil assay (blue squares) or tritium radiolabel (pink circles) and human IgG by 1 2 5I gamma counting (green diamonds). Plasma was isolated from SCID mice (n=16) at 30 minutes (a), 1 hour (b), 2 hours (c) and 4 hours (d) after injection of antibody-coupled liposomes. 196 employing an anti-human IgG (heavy and light chains) alkaline phosphatase enzyme secondary antibody (Figure 7.4). Although there was a slight quantitative discrepancy in the amount of liposome lipid in the peak fractions, the stability of the antibody-coupled liposome formulation was revealed by FPLC analysis. The antibody co-elutes with the liposome and the radiolabel ratio of antibody to liposome lipid remains similar to the initial formulation even after four hours. The relative amount of liposomal formulation loaded onto the FPLC in a 250ul plasma injection steadily decreased, reflecting a loss of the formulation from plasma in agreement with the clearance results (Figure 7.2). Just as significantly, any peaks for 125I-gamma counts of the antibody in fractions further down the elution profile where any free uncoupled antibody would have eluted were not observed. Qualitatively, dot blot results (Figure 7.4) shown support the evidence given for the antibody remaining attached to the liposome. The presence of the human IgG in the plasma was also confirmed using Western gel analysis and transfer blot analysis of the plasma sample (Figure 7.5). The controls and a 30 minute plasma sample were run on duplicate polyacrylamide gels and one was stained with 0.1% Commassie blue. The other was transferred, blotted and hybridized with a goat anti-human alkaline phosphatase secondary antibody to confirm the presence of the human IgG primary present in plasma. The bands identifying the antibody either before it was injected as a liposomal formulation or 30 minutes after it was placed in circulation and removed subsequently are similar in migration pattern on the transfer blot analysis. Because the formulation was diluted by blood and was partially cleared at 30 minutes, the resulting quantity that could be detected in the lanes representing the formulation in SCID mouse plasma was limited. 197 Fraction # Elution Volume (ml) 1 2 3 4 5 6 7 8 9 10 11 12 36 37 38 39 40 41 42 43 44 45 46 47 Fraction # 13 14 15 16 17 18 19 20 21 22 23 24 Elution Volume (ml) 48 49 50 57 52 53 54 55 56 57 58 59 * Fraction # 25 26 27 28 29 30 31 32 33 34 35 36 Elution Volume (ml) 60 61 62 63 64 65 66 67 68 69 70 77 Fraction # 37 38 39 40 41 42 43 44 45 46 47 48 Elution Volume (ml) 72 73 74 75 76 77 78 79 SO 81 82 83 Fraction # 49 50 51 52 53 54 55 56 57 58 59 60 Elution Volume (ml) 84 85 86 67 88 89 90 91 92 93 94 95 Figure 7.4: Dot blot assay for Antibody coupled Liposomes eluted off FPLC Dot blot assay for the human IgG antibody in the fractions eluted off the FPLC. The dot blot results support the presence of human IgG antibody in FPLC fractions corresponding to liposome elution. Note the qualitative dark dots in fractions 11-15 corresponding to elution volumes 46-50ml on the FPLC elution profile. 198 o o o <0 O <-H Tfr ^ m <N — Cffl cd o CN CN 7.3.3 Tissue Biodistribution of Antibody-coupled Liposomes The biodistribution of the antibody-coupled liposome formulation into SCID mouse tissues was evaluated for the levels of antibody and liposome lipid (Figures 7.6 and 7.7). The levels of the antibody are greatest in the liver and spleen followed by lung, kidney and heart in descending order. The levels of the antibody decrease over the four hour tissue sampling time-course except for the levels in the kidney, which remain the same at the four time points chosen (n=4 per group). The levels of liposomes on a per gram tissue basis increase, over the four hour time-course for tissues of the reticuloendothelial system (RES), namely liver and spleen. The high levels of liposomes entrapped in the spleen probably contribute to the significant amounts of human IgG antibody present in this tissue. 7.3.4 Histochemistry analysis of fluorescent markers The results thus far provide evidence that the antibody remains stably attached to the liposome surface in plasma. What happens to the formulation when it exits the circulation? Some answers to this question were provided by isolating tissue samples for histochemical analysis. Investigation of the liver, spleen and kidney tissues were carried out. The results seen under fluorescence microscopy of cryosectioned tissues employing the original Dil label for liposomes and a mouse anti-human-FITC secondary antibody for the antibody are shown in Figure 7.8. The spleen was abundant in Dil liposomes and FITC label green fluorescence identifying human IgG. The co-localization of the two probes was evidenced in the cropped photos. In the liver, modest amounts of Dil (red channel) liposome labeling and minute quantities of FITC antibody labeling (green channel) were observed. Although not readily evident in this photo, there appears to be 200 0.5 hr 1 hr 2 hr 4 hr Figure 7.6: The biodistribution of Antilxidy in SOD mouse tissues. The biodistribution of the human IgG antibody in the lung, heart, liver, kidney and spleen tissues of SCID mice after injection of human IgG antibody coupled liposomes. On a per gram basis (error bars represent standard deviation), the antibody was present in high levels in the liver and spleen. Over the entire four hour time course, the decrease in antibody was most pronounced in the lung and liver while loss was also recorded for the spleen. 201 Figure 7.7: The biodistribution cf Uposomes iii SQD niiuse tissues. The biodistribution of the liposome in the lung, heart, liver, kidney and spleen tissues of SCID nice after injection of human IgG antibody coupled liposomes over a four hour time course. Significant levels of the lipid carrier were found in the spleen at all four time points on a per gram tissue basis. The specific accunwlation of the liposome to the liver was pronounced relative to lung, heart and kidney tissues but levels remained constant over the four hour time course. 202 KIDNEY D i l Liposomes FITC Antibodies Superimposed LIVER Dil Liposomes FITC Antibodies Superimposed SPLEEN Dil Liposomes FLTC Antibodies Superimposed 205 Figure 7.8: Fluorescent histochemical analysis of Antibody-coupled Liposomes in SCID mouse tissues. Histochemical analysis of 12um thick sections of SCID mouse kidney (p.203), liver (p.204) and spleen (p.205). The liposomes tracked with Dil (red channel; rhodamine optics) and human IgG is detected by goat anti-human IgG-FITC conjugate (green channel; FITC optics). Both images are superimposed together to illustrate the co-localization of the two species (if present) in the respective tissues. The scale bar is 20um in length in each photo. 206 localized liver-specific metabolism of the antibody-coupled liposomes resulting in a relatively lower FITC green fluorescence for a given amount of Dil red fluorescence upon comparison of spleen and liver photos. The calculations for the ratio of antibody to liposome lipid ratio at 2 hours indicate that the levels found in the spleen are about half compared to the ratio found in the plasma or the original injected sample. Finally, in the kidney while there are liposomes present, the lack of a green FITC antibody signal indicates that the human IgG antibody initially coupled to the liposome became dissociated from its carrier in the kidney. These qualitative photographic results are consistent with the biodistribution results obtained with the radiolabeled tracers described above. 207 7.4 DISCUSSION Of the most popular immunosuppressants used in the management of organ transplantation in current clinical use, many of them are relatively hydrophobic drugs such as cyclosporin A and FK506. In the last decade, however, a new generation of immunosuppressants has emerged as viable drugs in the quest to achieve immunological tolerance of grafted tissues. Of particular interest is the monoclonal antibody OKT3 (Kung et al., 1979 and Rao and Kroon, 1993) which has demonstrated excellent results in prolonging graft survival and in rescue therapy (Martin et al., 1988 and Norman and Leone, 1991). The antibody modulates TCR/CD3 expression and activity resulting in a depletion of T cells that play a key role in graft rejection. The antibody specifically binds the CD3s subunit of the TCR/CD3 complex on thymocytes (Bluestone et al., 1987). Even with genetic engineering of this anti-TCR CD3e antibody, the early cytokine release syndrome associated with the use of OKT3 has yet to be ameliorated. The highly immunogenic F c portion becomes bound in the immune response and accounts for the early toxic effects observed (Sawchuk, Gates and Hirsch, 1995). In many cases, human anti-mouse antibodies (HAMA) are synthesized against OKT3, thereby limiting its therapeutic use in management of acute graft rejection. Strategies to deliver anti-CD3 (Fa b)2 antibody fragments (Hirsch, Archibald and Gress, 1991) have resulted in superior immunosuppressive properties in allogeneic bone marrow transplant (Blazar et al., 1993 and Blazar, Taylor and Vallera, 1994), in a murine skin transplant model (Vossen et al, 1995) and in thymectomized mice (Sawchuk et al., 1990 and Vossen et al, 1994). Chimeric mouse/human anti-CD3 antibodies are also being constructed in efforts to decrease the immunogencity of the antibody (Arakawa et al., 1996). 208 The equivalent of the mouse anti-human antibody OKT3 is the hamster anti-mouse 145-2C11 which also has been extensively studied in mouse models of immunosuppressive therapy in organ transplantation (Leo et al., 1987 and Hendrickson et ah, 1995). The cell line that produces this antibody was obtained from the American Type Tissue Culture (ATCC). Unfortunately, the amount of antibody that had to be isolated from the cell supernatent or ascites fluid for an experiment was not feasible for our purposes. Other allotypic and isotypic immunosuppressive antibody molecules have similarly been developed and used in animal experimentation (Tomonari, 1988, Hirsch et al, 1988, 1989, Miescher, Schreyer and MacDonald, 1989, Coulie et al, 1991 and Tsuchida et al, 1994). A formidable alternative for the delivery of antibody-type drugs, such as OKT3, in the form of an antibody-coupled liposome was presented in this chapter. Specifically, a carbohydrate-based coupling procedure (Ansell et al, 1996) used to link a readily available and characterizable antibody, human IgG, through the F c portion onto the liposome surface. The antibody and lipid components of this formulation were tracked to determine whether the antibody remains with the liposome. Immobilization of biosynthetic lipid-tagged antibodies reconstituted into liposomes (Laukkanen, Alfthan and Keinanen, 1995) and other chemical coupling methods have been attempted using biotin-streptavidin (Loughrey et al, 1990), SPDP-maleimide (Hansen et al, 1995) linkers and (Fa b)2 antibody fragments via either thioether (Martin and Papahadjopoulos, 1982), disulfide (Martin, Hubell and Papahadjopoulos, 1981) or novel (Shahinian and Silvius, 1995) coupling procedures have been used to produce antibody-liposome conjugates. Smaller molecules such as conjugated and unconjugated peptides capable of inducing am 209 antibody and cytotoxic T lymphocyte response have been delivered in liposomal carriers (Alving et al, 1995). Recently, antibodies attached to polyethylene glycol (PEG) chains on liposomes were also investigated (Zalipsky et al. 1993, 1997 and Maruyama et al., 1994). In the present investigation, the stability of a novel antibody-coupled liposome formulation was tested in vivo. The coupling method of choice required an activated antibody to be conjugated through a propionyl hydrazide to a maleimide moiety anchored by a phospholipid to the liposome surface. In comparison to liposome formulations of similar lipid composition devoid of Dil , this formulation has a relatively short (thirty minute) half-life. The faster clearance might be attributed to the presence of human IgG antibody present on the exterior surface of the liposome formulation or Dil present in the bilayer membrane. Liposomes with galactose-terminated PEG molecules on the surface demonstrated similar increased elimination rates from plasma (t,/2 = 0.3hours) which have been attributed in large part to uptake by liver Kupffer cells (Shimada et al, 1997). Liposomes have been previously fluorescently tagged using either fluorescein-DPPE (Claassen, 1992) or Dil (Litzinger, Buiting and van Rooijen, 1992 and Russell, Graveley and Coxon, 1992). The effect of fluorescent labeling on liposome clearance has not been fully documented but qualitative illustrations of these liposomes entrapped in the liver and spleen within 30 minutes (Claassen, 1991) and 2 hours after injection are reported by Buiting et al., 1992. In comparison to the many foreign antibodies that have been injected into circulation that have demonstrated in vivo half-lives in the order of days (Russell, 1992 and Routledge, 1995), initial injection of OKT3 into humans revealed rapid clearance within hours (Henell and Norman, 1993). The actual plasma elimination 210 rates do vary widely among individuals and the degree to which a monoclonal is humanized is important in the rate of clearance. The primary sites of degradation of antibodies such as OKT3 have been shown by peptide mapping (Rao and Kroon, 1994) to be deamidation (asparagine residues) and oxidation (cysteine and methionine residues). Monoclonal antibody conjugates as drugs have been designed with plant toxins and radionucleotides to target tumors (Ren, 1991). In this study, human IgG, and a liposome were coupled and the antibody's circulation half-life in a SCID mouse model was effectively decreased. This increase in plasma elimination of the antibody was attributed to the larger presentation surface of liposomes compared to free antibody monomers. Liposomologists have exploited this property of liposomes in the preparation of immunoadjuvants especially in charged liposome formulations without PEG present as a steric barrier to opsonization. What was more important in this study, however, was that by devising a method to covalently link an antibody to the liposome surface, the immediate loss or exchange of the molecule from the liposome was limited. The plasma clearance results indicate that both antibody and liposome remain together and intact over the four hour-time course. The stability of such a formulation in vivo supports a model of antibody delivery that would be advantageous in the delivery of therapeutic antibodies such as OKT3. By chemically coupling this therapeutic antibody, the benefits of passive liposomal targeting or uptake into sites of disease, inflammation or of particular interest, allogeneic tissue graft sites are possible (Chapter 5). In the development of liposomal immunosuppressants, prior attempts to keep an immunosuppressant such as cyclosporin A with the liposome over extended periods in vivo were unsuccessful (Choice et al., 1995). Using the same liposome formulation, 211 however, it was discovered that the liposome does preferentially accumulate in rat heart grafts (Chapter 5). It has long been touted that increasing liposome circulation lifetimes will enable better liposomal targeting of drugs associated with liposomes. This is true especially when extravasation plays an integral role in liposomal drug delivery as in the case of tumor-targeted delivery of anticancer agents (Papahadjopoulos, 1991). Because the vasculature in sites of disease, inflammation and tissue grafts yields itself highly permeable, due to wide interendothelial junctions, the chance of liposomal drug deposition increases in these areas. The SDS-PAGE gel and transfer blot identified the presence of the antibody in plasma. The immunoassay was highly sensitive in being able to detect nanogram quantities of human IgG. These results confirmed radiochemical probe clearance data. Furthermore, the FPLC separation of plasma components and analysis of fractions clearly demonstrated that the components of the liposomal formulation were intact and stable and any free antibody in plasma was undetectable. Similar FPLC results were observed for these antibody coated liposomes incubated in plasma at 37°C and at 65°C over 24 hour-time courses ex vivo. The liver represents a major site of metabolism of the antibody. High levels of the human IgG antibody were found in this tissue. One reason for the constant levels found in the first four hours in the kidney was that the simultaneous collection and excretion of the free metabolized antibody occurs through the kidneys. The quantity of the initial antibody-coupled liposome formulation cleared from plasma at the first time point could account for any free antibody metabolized and cleared by the liver and kidney at the early time points. The high levels of the antibody observed in the spleens of SCID 212 mice may be a direct effect of liposome sequestration in this organ of the reticuloendothelial system (Figure 7.6). Hence, the antibody found in the liver and spleen would reflect the ability of the reticuloendothelial organs to naturally target foreign particles that have been marked for clearance. In fact, it would then be very convenient to target a therapeutic antibody such as OKT3 to the organs of the RES (eg. liver transplant) using liposomes. The liposome deposition in the tissues investigated results in a steady increase in liposome content in the spleen, levels remaining constant in the heart, liver and kidney and a decrease in lung liposome levels over the four hour time course (Figure 7.7). In Chapter 5, similar levels of liposomes in the spleen and liver were found which can be compared to the equivalent levels on a per gram tissue basis to allogeneically grafted rat hearts. The uptake of liposomes in RES tissues is due to tissue macrophages in the liver (Kupffer cells) and macrophages lining the splenic cords and venous sinusoids of the spleen. Additionally, the spleen is very effective in filtering and removing particulate blood borne matter and effete red and white blood cells. In this respect, the large uptake of liposomes and attached antibody by the spleen was probably also due to a positively charged Dil fluorescent probe and the antibody in the liposomal formulation. The evidence presented here, in conjunction with liposome deposition in rat heart transplantation results, would support the notion that delivery of a therapeutic immunosuppressive monoclonal antibody using liposomes to graft sites is possible. To summarize the histochemical results, the antibody was co-localized with the liposome in the spleen and liver and the majority of the liposome formulation in the kidney was devoid of human IgG antibody (Figure 7.8). Dual labeling studies of Dil and FITC in spleen sections have been reported by Buiting et al. (1993) and liposomal 213 extravasation studies have been previously studied by Unezaki et al. (1996). Based on the groundwork studies performed in this thesis, the co-localization of antibody and liposomes in tissues occurred using histochemical fluorescent probes can be similarly determined for other antibody-coupled liposome formulations. The effect of varying the lipid composition on clearance and biodistribution kinetics is briefly described here to provide ways to test and improve the current antibody-coupled liposome formulation to achieve better drug delivery. Although other lipids and coupling techniques can be used, the discussion is limited to the components of the current liposome formulation. The primary phospholipid, DSPC, is one of the longer acyl chain phospholipids used by liposomologists. It provides for a relatively stronger and thicker bilayer and the addition of 45 mole% cholesterol stabilizes and strengthens it for in vivo applications (Klibanov et al., 1990). The use of DSPC/CHOL/PEG liposomes has also been investigated by Marayama et al., 1992. The presence of DSPE-PEG20oo affords a steric and hydrophilic barrier to opsonins or proteins (Senior et al., 1990) that would otherwise readily insert into the naked liposome bilayer membrane. Allen et al. (1987) reported a corresponding decrease in RES uptake of liposomes incorporating PEG when compared to liposomes without PEG. Increasing the molar content of this lipid would prevent access to the lipid bilayer (by sterically hindering plasma protein interaction) thereby increase the circulation lifetime of such a liposome formulation (Lasic et al, 1991, Papahadjopoulos et al, 1991 and Torchilin and Papisov, 1994). It has been previously shown that 2 mole% DSPE-PEG2000 was an appropriate PEG content to use in antibody-coupled liposome formulations (unpublished data). A balance between the protective effect of this lipid against opsonins and the 214 degree of antibody coupling to be achieved with DSPE-MPB (1 mole%) had to be struck (Klibanov et al, 1991). Clearly, too much PEG in the bilayer would prevent both coupling chemistry and target binding. The latter was demonstrated for liposomes incorporating PEG5 0oo (Mori et al, 1991). Carbohydrate chemical coupling of the activated antibody was performed on pre-formed liposomes. If greater amounts of DSPE-PEG2000 were used compared to the same amount of DSPE-MPB, coupling was found to be less efficient or absent (unpublished results). While the lipid components may be suitable for coupling with human IgG, both the steps in the work-up of activated antibody and the ratio of the various lipids in the formulation should be adjusted. Further investigation on the carbohydrate coupling procedure needs to be performed especially when different antibodies or different liposome compositions are to be used. The implications on liposomal drug delivery of varying PEG and cholesterol content in the liposomal bilayer are discussed further by Bedo-Addo et al. (1996). Maximal steric stabilization (surface protection) was achieved with low concentrations (<10%) of a short chain PEG (e.g. PEG 1000) phosphatidylethanolamine (PE) and high concentrations (>30%) of cholesterol. The implications of varying lipid dose are discussed by Oja et al. (1996). The lipid dose in these studies was 83mg total lipid per kg of mouse. As suggested in the Introduction, the interference brought about by carbohydrate chemical coupling on the structure and chemical bonding of an antibody could alter the immune response to a foreign antibody. When antibody was coupled using this carbohydrate technique to liposomes, a protein A column did not retain the resultant formulation (personal communication). This provides evidence for the lack of F c binding when antibody is coupled via it's carbohydrate moiety to liposomes. The F c plays a role 215 in stimulating the cytokine release syndrome (CRS) caused by the administration of therapeutic antibodies, namely OKT3. If masked or modified in such a way that protein G can no longer bind it, a parallel lack of recognition could prevent CRS. Thus, patients receiving therapeutic antibodies such as OKT3 could experience a benefit in reduced toxicity with the advent of OKT3-coupled liposomes. The immunogenicity triggering classical pathway complement activation and humoral responses leading to increased rates of antibody clearance was found to be mediated by the F c domain (Anderson et al., 1990 and Burton and Woof, 1992). The toxicity of antibody administration was also shown to be mediated via the F c region (Sawchuk, Gates and Hirsch, 1995) and glycosylated Fc sites (Routledge et al, 1995). In vitro evidence for decreased crystallizable fragment (Fc) receptor binding upon single amino acid modification in the F c region was given by Alegre et al. (1992). By decreasing F c receptor binding, a subsequent decrease in the antibody immunogenecity should occur. It follows then that chemical carbohydrate modification in the F c linking antibody to liposome will decrease the relative toxicity of the antibody. The immunogeneicity of antibody-coupled liposomes attached via a F c carbohydrate moiety should also be addressed in an immunocompetent mouse (SCUD RAG2 mice used in this study lack active T and B cells that have receptors for F c and do not produce immunoglobulins). Whereas Harding et al. (1997) report liposomal PEG-grafted antibodies to be more immunogenic than the free antibody, the advantage of masking the F c (based on the lack of binding by protein A column of carbohydrate coupled antibody formulation (person communication)) should be evaluated. 216 Liposomes have been recently studied for their potential effect on effector cells and in transplantation. Positively charged liposomes were evaluated for their toxicity towards murine macrophages and T lymphocytes (Filion and Phillips, 1997) and liposomes were used to mediate transfection of HLA-DR alpha chain gene in order to induce graft tolerance (Aleksic et al, 1997). In the former study, cationic liposomes were highly toxic to phagocytic cells but 10mole% of DPPE-PEG2000 was found to completely abolish the toxic effects of synthetic cationic liposome formulation. In the latter study, a gene/liposome mixture was delivered by intracoronary infusion into pigs and yielded the expression of the gene product on the pig endothelium for 48 hours. This approach could help overcome barriers to xenotransplantation. If the transfer of an organ recipient's HLA genes into animals is possible, reactions to xenogeneic transplants may be decreased. In conclusion, antibody-coupled liposomes using a carbohydrate coupling technique were developed and a number of techniques to address the question of whether the antibody-coupled liposomes are stable in vivo were demonstrated. The evidence in this chapter indicates that the human IgG antibody-coupled through a carbohydrate-PDP-MPB-linker to the liposome remains stable in SCID mouse plasma over a 4 hour-time course. Moreover, the levels of antibody and liposome in plasma and other tissues were quantitated and coexistence of both using fluorescent probes were qualitated. The methodical use of techniques and the development of assays in this research will aid in defining and providing future evidence for the development of antibody-coupled liposomal drug formulations. 217 C H A P T E R 8 S U M M A R Y The characterization of liposomal immunosuppressants for management of organ transplantation was investigated in the research presented in this thesis. The most widely used immunosuppressant drug, cyclosporine, was characterized in terms of incorporation into liposomal vesicles, exchange between vesicle populations, stability in vitro and in vivo, elimination characteristics in circulation and biodistribution into animal tissues. Liposome delivery of cyclosporine was also evaluated in rat heart transplantation. These studies were performed as part of the research objectives outlined in the introductory chapter of this thesis. A well-defined liposome formulation of cyclosporine was characterized in terms of lipid composition, the effect of charged lipids and cholesterol content. Their preparation by extrusion yielded vesicles of around 120nm in diameter. A finite amount of cyclosporin A remains associated with the liposomal carrier when subjected to filtration through a 0.2um filter, gel exclusion chromatography or density gradient ultracentrifugation. While the rationale of incorporating cyclosporine into a liposome was theoretically sound, the research has demonstrated that the drug does not remain stably associated with the liposomal carrier when other depots such as liposomes devoid of cyclosporine are present. Hence, drug exchange or release from the liposome was rapid and complete within minutes of mixing with other liposomes or blood components. When liposomal cyclosporine was injected into mice, a predictable rapid release of cyclosporine was again evident as plasma elimination and tissue biodistribution characteristics of the drug and the liposome carrier did not coincide with each other. 218 While liposomes preferentially accumulated in the spleen (on a per gram weight basis), cyclosporine deposition was largely concentrated in the liver. In the rat heart transplant model then, it was not surprising that levels of cyclosporine did not correspond to the high levels of liposomes achieved in the allogeneic cardiac graft. Any beneficial effects that would be afforded by liposomal delivery of cyclosporine do not result from increased potency or decreased nephrotoxicity of the drug due to localized delivery of the drug to graft sites because the results in this thesis provide evidence that this does not occur. The rat heart transplant model was successful in demonstrating a localized and a statistically significant accumulation of liposomes to an organ undergoing rejection. The results in this thesis have shown that liposomal cyclosporine pharmacokinetics do not differ from the clinical intravenous cyclosporine formulation in cremophor oil. A benefit of dispensing cyclosporine in liposomal form may, however, be realized by replacing the cremophor oil, which was found to be relatively more toxic than liposomes, with liposomes as a solubilizing agent for the drug. At this point in the thesis project, an emphasis was placed on keeping the drug with the liposome carrier. The chemical structure and hydrophobic nature of cyclosporine precluded its use in derivatizing it for liposomal entrapment or covalent attachment. The approach to this problem was to test the feasibility of delivering antibodies using liposomes. Many antibodies with immunosuppressive properties are currently being investigated. The therapeutic monoclonal antibody, OKT3, for example, has been used in combination with other immunosuppressants and in rescue therapy in organ rejection. This thesis describes the use of human IgG (hlgG), that was chemically coupled to form antibody-coupled liposomes using a carbohydrate coupling method. A reliable technique to track the 219 antibody-coupled liposome drug formulation in the presence of plasma was required. In response, fast protein liquid chromatography (FPLC) analysis was developed for the separation of liposomes from plasma lipoproteins. With direct application to antibody-coupled liposomes, the FPLC was extensively characterized for the elution profiles of liposomes, lipoproteins from human and mouse serum and plasma and other small molecules such as antibodies. The carbohydrate coupling procedure was used for human IgG antibody coupling to sterically stabilized liposomes. This formulation was studied in terms of antibody coupling efficiency, plasma elimination characteristics of the antibody and the liposome and biodistribution in vivo. The quantitative (radiolabel and fluorescence) and qualitative (fluorescence and immunochemistry) results provide evidence for stable association of the covalently coupled antibody to the liposome over a four hour time-course. The knowledge and training obtained in the characterization and development of liposomal immunosuppressants utilizes recent advances in scientific drug delivery research. FPLC separation of liposomes from the majority of plasma proteins has many applications in the field of liposomology. For example, the stability or exchange of any liposomal component can be determined by isolating liposomes from plasma and assaying for the amount remaining in the liposome fraction(s). Alternatively, if the components readily exchange into plasma lipoproteins or albumin, the extent to which they are sequestered into these fractions can similarly be determined. Hence, the time-course of drug release from liposomes can be determined. The FPLC technique is limited in the speed (4 hours) required to separate liposomes and plasma components and the size of the sample that can be loaded onto the columns. The future of the antibody-coupled 220 liposome formulation may involve target binding studies in which a suitable model would be that of an anti-CD3 antibody to either CD3 immobilized on a gel matrix or cells possessing the CD3 antigen in cell culture or in vivo. The model system for attachment of antibodies to liposome carriers and the evidence for the stability of the attachment of the antibody in vivo, establish part of the groundwork for the delivery of therapeutic antibodies using liposomes. The impact of this work, in conjunction with the demonstration of liposomal accumulation at organ rejection sites, will hopefully lead to the selective delivery of therapeutic antibody using liposomes in clinical transplantation. 221 REFERENCES Abbs, L C , Clark, M. , Waldmann, H., Chatenoud, L., Koffman, C G . and Sacks, S.H. (1994) Sparing of first dose effect on monovalent anti-CD3 antibody used in allograft rejection is associated with diminished release of proinflammatory cytokines. Therapeutic Immunology 1:325-331. Alegre, M.-L., Collins, A.M., Pulito, V.L., Brosius, R.A., Olson, W.C., Zivin, R.A., Knowles, R., Thistlewaite, J.R., Jolliffe, L.K. and Bluestone, J.A. (1992) Effect of a single amino acid mutation on the activating and immunosuppressive properties of a "humanized" OKT3 monoclonal antibody. The Journal of Immunology 148(11):3461-3468. Aleksic, I., Ren, M. , Popov, A., Freimark, D., Blanche, C , Czer, L., Trento, A. and Barath, P. (1997) In vivo liposome-mediated transfection of HLA-DR alpha-chain gene into pig hearts. European Journal of Cardio-Thoracic Surgery 12(5):792-797. Allen, T.M. (1981) A study of phospholipid interactions between high-density lipoproteins and small unilamellar vesicles. Biocimica et Biophysica Acta 640(2):385-397. Allen, T.M. and Chonn, A. (1987) Large unilamellar liposomes with low uptake into the reticuloendothelial system. Federation of European Biochemical Societies 223(1 ):42-46. Allen, T.M., Hansen, C. and Rutledge, J. (1989) Liposomes with prolonged circulation times: factors affecting uptake by reticuloendothelial and other tissues. Biochimica et Biophysica Acta 981:27-35. Allen, T.M., Hansen, C , Martin, F., Redemeann, C. and Yau-Young, A. (1991) Liposomes containing synthetic lipid derivatives of poly(ethylene glycol) show prolonged circulation half-lives in vivo. Biochimica et Biophysica Acta 1066:29-36. Allen, T.M. (1994) Long-circulating (sterically stabilized) liposomes for targeted drug delivery. Trends in Pharmacological Sciences 15:215-220. Alonso-Pulpon, L., Serrano-Fiz, S., Rubio, J.A., Cavero, M.A., Silva, L., Sanchez-Vegazo, I., Burgos, R., Montero, C.G., Maicas, C , Kreisler, M. , Crespo, M.L., Tellez, G. and Ugarte, J. (1995) Efficacy of low-dose OKT3 as cytolytic induction therapy in heart transplantation. Journal of Heart and Lung Transplantation 14(1): 136-142. Alving, C.R., Koulchin, V., Glenn, G.M. and Rao, M . (1995) Liposomes as carriers of peptide antigens: Induction of antibodies and cytotoxic T lymphocytes to conjugated and unconjugated peptides. Immunological Reviews 145:5-31. 222 Anderson, C.L., Shen, L., Eicher, D.M., Wewers M.D. and Gill, J.K. (1990) Phagocytosis mediated by three distinct Fc gamma receptor classes on human leukocytes. Journal of Experimental Medicine 171(4): 1333-45. Andersson, T., Carlsson, M. , Hagel, L., Pernemalm, P. and Janson, J.-C.(1985) Agarose-based media for high resolution gel filtration of biopolymers. Journal of Chromatography 326:33-44. Andres, A., Mazuecos, A., Morales, J.M., Praga, M. , Martinez, M.A., Usera, G., Araque, A. and Rodicio, J.L. (1994) Renal allograft infarction is a cause of early kidney transplant loss in the cyclosporine A era. Transplantation Proceedings 26(5):2544-2545. Ansell, S.M., Tardi, P.G. and Buchkowsky, S.S. (1996) 3-(2-Pyridyldithio)-propionic acid hydrazide as a cross-linker in the formation of liposome-antibody conjugates. Bioconjugate Chemistry 7:490-496. Ar'Rajab, A., Dawidson, I.J.A., Sentementes, J.T. and Harris, R.B. (1994) Deleterious effect of cyclosporines on the ischemic kidney in the rat and protection by the calcium antagonist verapamil. Transplantation Proceedings 26(5):2894-2895. Arai, H., Yuda,T., Goodgold, H.M., Martin, T.W., Swartz, M.T. and Pennington, D.G. (1992) Asymmetric intramyocardial lymphocytic accumulation during acute rejection in rat heart and heart-lung transplantation. The Journal of Heart and Lung Transplantation ll(4):763-772. Arakawa, F , Kuroki, M. , Kuwahara, M. , Senba, M. , Ozaki, H., Matsuoka, Y., Misumi, Y., Kanda, H. and Watanabe, T. (1996) Cloning and sequencing of the V H and V ^ genes of an anti-CD3 monoclonal antibody, and construction of a mouse/human chimeric antibody. Journal of Biochemistry 120:657-662. Aschoff, L. (1924) Das reticulo-endotheliale system. Ergebn. Inn. Med. KinderheUk 26:1-118. Averna, M. , Marcovina, S.M., Noto, D., Cole, T.G., Krul, E.S. and Schonfeld, G. (1995) Familial Hypobetalipoproteinemia is not associated with low levels of lipoprotein(a). Arteriosclerosis, Thrombosis and Vascular Biology 15(12):2165-2175. Azrin, M.A. (1992) The use of antibodies in clinical cardiology. American Heart Journal 124(3):753-768. Bach, J.-F. (1994) Lessons for transplant immunosuppression from the usage of cyclosporine in autoimmune diseases. Transplantation Proceedings 26(5):3077-3081. Backer, J.M. and Dawidowicz, E.A. (1979) The rapid transmembrane movement of cholesterol in small unilamellar vesicles. Biochim. Biophys. Acta 551:260-270. 223 Baldwin 3™ W.M ., Pruitt, S.K., Brauer, R.B., Daha, M.R. and Sanfilippo, F. (1995) Complement in organ transplantation. Contributions to inflammation, injury, and rejection. Transplantation 59(6):797-808. Bally, M.B., Nayar, R., Masin, D., Hope, M.J., Cullis, P.R. and Mayer, L.D. (1990) Liposomes with entrapped doxorubicin exhibit extended blood residence times. Biochimica et Biophysica Acta 1023(1): 133-139. Bally, M.B., Nayar, R., Masin, D., Cullis, P.R. and Mayer, L.D. (1990) Studies on the myelosuppressive activity of doxorubicin entrapped in liposomes. Cancer Chemotherapy & Pharmacology 27(1): 13-19. Bangham, A.D. (1968) Membrane models with phospholipids. In: Progress in Biophysics and Molecular Biology Vol. 18, Butler, J .AV. and Noble, D., eds., Pergamon Press, Oxford, pp.29-95. Barry, J.A. and Gawrisch, K. (1995) Effects of ethanol on lipid bilayers containing cholesterol, gangliosides, and sphingomyelin. Biochemistry 34(27): 8852-8860. Batiuk, T.D., Pazderka, F. and Halloran, P.F. (1994) How do cells recover from inhibition by cyclosporine? Transplantation Proceedings 26(5):2831-2832. Batzri, S. and Korn, E.D. (1973) Single bilayer liposomes prepared without sonication. Biochimica et Biophysica Acta 298(4):1015-1019. Bedu-Addo, F.K., Tang, P., Xu, Y.and Huang, L. (1996) Effects of polyethylene glycol chain length and phospholipid acyl chain composition on the interaction of polyethylene-phospholipid conjugates with phospholipid: Implications in liposomal drug delivery. Pharmaceutical Research 13(5):710-717. Bennett, W.M., Burdmann, E., Andoh, T., Elzinga, L. and Francescnini, N. (1994) Nephrotoxicity of immunosuppressive drugs. Minerals, Electrolytes and Metabolism 20:214-220. Beveridge, T. (1992) Clinical development of cyclosporine. Transplantation Proceedings 24(4, Suppl. 2):64-66. Beveridge, T., Gratwohl, A., Michot, F., Niederberger, W., Nuesch, E., Nussbaumer, K., Schaub, P. and Speck, B. (1981) Cyclosporin A: pharmacokinetics of a single dose in man and serum levels after multiple dosing in recipients of allogeneic bone-marrow grafts. Current Therapeutic Research 30:5-18. Binder, J., Mishina, E.V., Jusko, W.J. and Kupiec-Weglinski, J.W. (1994) Prolongation of cardiac allograft survival in rats by liposome-encapsulated methylprednisolone. Transplantation 58(5):633-634. 224 Blazar, B.R., Hirsch, R., Gress, R.E., Carroll, S.F. and Vallera, D.A. (1991) In vivo administration of anti-CD3 monoclonal antibodies or immunotoxins in murine recipients of allogeneic cell-depleted marrow for the promotion of engraftment. The Journal of Immunology 147(5): 1492-1503. Blazar, B.R., Taylor, P.A., Snover, D.C, Bluestone, J.A. and Vallera, D A . (1993) Nonmitogenic anti-CD3F(ab')2 fragments inhibit lethal murine graft-versus-host disease induced across the major histocompatibility barrier. The Journal of Immunology 150(l):265-277. Blazer, B.R., Taylor, P.A. and Vallera, D.A. (1994) In vivo or in vitro anti-CD3s chain monoclonal antibody therapy for the prevention of lethal murine graft-versus-host disease across the major histocompatibility barrier in mice. The Journal of Immunology 152:3665-3674. Bluestone, J.A., Pardoll, D., Sharrow, S.O. and Fowlkes, B.J. (1987) Characterization of murine thymocytes with CD-3associated T-cell receptor structures. Nature 326(5):82-84. Boisvert, W.A., Spangenberg, J., and Curtiss, L.K. (1995) Treatment of severe hypercholesterolemia in apolipoprotein E-deficient mice by bone marrow transplantation. The Journal of Clinical Investigation 96:1118-1124. Boman, N.L., Mayer, L.D. and Cullis, P.R. (1993) Optimization of the retention properties of vincristine in liposomal systems. Biochimica et biophysica Acta, 1152:253-258. Borel, J.F. (1994) Cyclosporine. In: Textbook of Immunopharmacology, 3rd edition, Dale, M.M., Foreman, J.C., and Fan, T.D., eds., Blackwell Scientific Publications, Oxford. pp.320-329. Bosma, G.C., Custer, R.P. and Bosma, M.J. (1983) A severe combined immunodeficiency mutation in the mouse. Nature 301(10):527-530. Boyd, J.E. and James, K. (1989) Human monoclonal antibodies: their potential, problems, and prospects. Advances in Biotechnological Processes 11:1-43. Brekke, O.H., Michaelsen, T.E. and Sandlie, I. (1995) The structural requirements for complement activation by IgG: does it hinge on the hinge? Immunology Today 16(2):85-90. Briggs, J.D. (1991) A critical review of immunosuppressive therapy. Immunology Letters 29:89-94. Bristow, M.R., Gilbert, E.M., Renlund, D.G., DeWitt, C.W., Burton, N.A. and O'Connell, J.B. (1988) Use of OKT3 monoclonal antibody in heart transplantation: Review of the initial experience. The Journal of Heart Transplantation 7(1): 1-11. 225 Buiting, A.M.J., De Rover, Z., Claassen, E. and Rooijen, N. (1993) In vivo distribution of particulate antigens and liposomes in murine spleen. A possible role in the humoral immune response. Immunobiology 188:13-22. Burstein, M. , Scholnick, H.R. and Morfin, R (1970) Rapid method for the isolation of lipoproteins from human serum by precipitation with polyanions. Journal of Lipid Research ll(6):583-95. Chatenoud, L. and Bach, J.-F. (1993) Therapeutic monoclonal antibodies in transplantation. Transplantation Proceedings 25(l):473-474. Chatenoud, L., Baudrihaye, M.F., Kreis, H., Goldstein, G., Schhindler, J. and Bach, J.-F. (1982) Human in vivo antigenic modulation induced by the anti-T cell OKT3 monoclonal antibody. European Journal of Immunology 12:979-982. Chavin, K.D., Qin, L., Lin, J., Kaplan, A.J. and Bromberg, J.S. (1993) Anti-CD2 and anti-CD3 monoclonal antibodies synergize to prolong allograft survival with decreased side effects. Transplantation 55(4):901-908. Chen, C.-H. and Engel, S.G. (1990) The susceptibility of membrane vesicles to interaction with ethanol. Chemistry and Physics of Lipids 52(3-4): 179-187. Choice, E., Ayyobi, A.F., Pritchard, P.H. and Madden, T.D. (1998) Separation of liposomes from plasma lipoproteins using FPLC technology. Analytical Biochemistry, submitted. Choice, E, Meloche, R.M., Madden, T.D. (1998) Liposomal Cyclosporin A: Clearance and Biodistribution in a Rat Heart Transplant Model. Transplantation, submitted. Choice, E., Masin, D., Bally, M. , Meloche, M . and Madden T.D. (1995) Liposomal cyclosporine: Comparison of drug and lipid carrier pharmacokinetics and biodistribution. Transplantation 60(1):1006-1011. Chonn, A., Semple, S.C, Cullis, P.R. (1991) Separation of large unilamellar liposomes from blood components by a spin column procedure: Towards identifying plasma proteins which mediate liposome clearance in vivo. Biochim. Biophys. Acta 1070:215-222. Chu, C.T., Oury, T.D., Enghild, J.J. and Pizzo, S.V. (1994) Adjuvant-free in vivo targeting: Antigen delivery by a2-macroglobulin enhances antibody formation. The Journal of Immunology 152:1538-1545. Chua, M.M., Fan, S.T. and Karush, F. (1985) Attachment of immunoglobulin to liposomal membrane via protein carbohydrate. Biochimica et Biophysica Acta 800:291-300. 226 Citterio, F. and Kahan, B.D. (1991) Cyclosporine's dual mechanism of action in inhibiting generation of the cytoplasmic activation signal and the nuclear response thereto. Transplantation 51(6): 1248-1257. Claassen, E. (1992) Post-formation fluorescent labelling of liposomal membranes. In vivo detection, localisation and kinetics. Journal of Immunological Methods 147:231-240. Colvin, R.B. (1990) Cellular and molecular mechanisms of allograft rejection. Annual Reviews in Medicine 41:361-375. Compendium of Pharmaceutical Specialties (Canada) (1997) 32 n d Edition. Canadian Pharmaceutical Assoiciation, Ottawa, Canada. Cosimi, A.B. (1995) Future of monoclonal antibodies in solid organ transplantation. Digestive Diseases and Sciences 40(l):65-72. Costanzo-Nordin, M.R. (1993) Cardiopulmonary effects of OKT3: Determinants of hypotension, pulmonary edema, and cardiac dysfunction. Transplantation Proceedings 25(2 Suppl. l):21-24. Coulie, P.G., Uyttenhove, C , Wauters, P., Manolios, N. , Klausner, R.D., Samelson, L.E. and Van Snick, J. (1991) Identification of a murine monoclonal antibody specific for an allotypic determinant on mouse CD3. European Journal of Immunology 21:1703-1709. Cullis, P.R., Fenske, D.B. and Hope, M J . (1996) Physical properties and functional roles of lipids in membranes. In: Biochemistry of Lipids, Lipoproteins and Membranes, Vance, D.E. and Vance, J.E., (eds.), Elsevier Science Publishers, Amsterdam, pp. 1-41. Dale, M.M., Foreman, J.C. and Fan, TP., Eds. (1994) Textbook of immunopharmacology, 3 r d edition, Oxford, Boston, Blackwell Scientific Publications. p.352. Damen, J., Regts, J., Scherphof, G. (1981) Transfer and exchange of phospholipids between small unilamellar liposomes and rat plasma high density lipoproteins. Dependence on cholesterol content and phospholipid composition. Biochim. Biophys. Acta 665:538-545. Dantal, J. and Soulillou, J.P. (1991) Use of monoclonal antibodies in human transplantation. Current Opinion in Immunology 3(5):740-747. De Fazio, S.R., Masli, S. and Gozzo, J.J. (1996) Effect of monoclonal anti-CD4 and anti-CD8 on skin allograft survival in mice treated with donor bone marrow cells. Transplantation 61(1): 104-110. 227 Deamer, D. and Bangham, A.D. (1976) Large volume liposomes by an ether vaporization method. Biochimica et Biophysica Acta 443(3):629-634. Debs, R.J., Heath, T.D. and Paphadjopoulos, D. (1987) Targeting of anti-Thy 1.1 monoclonal antibody conjugated liposomes in Thy 1.1 mice after intravenous administration. Biochimica et Biophysica Acta 901:183-190. del Moral, R.G., O'valle, F., Andujar, M. , Lucena, M.A., Lopez-Hidalgo, J., Ramirez, C , Medina-Cano, M.T., Aguilar, D. and Gomez-Morales, M. (1995) Relationship between P-Glycoprotein expression and cyclosporin A in kidney. American Journal of Pathology 146(2):398-408. Derksen, J.T., Morselt, H.W., and Scherphof, G.L. (1987) Processing of different liposome markers after in vitro uptake of immunoglobulin-coated liposomes by rat liver macrophages. Biochimica et Biophysica Acta 93 l(l):33-40. Donatsch, P., Abisch, E., Homberger, M. , Traber, R., Trapp, M. and Voges, R. (1981) A radioimmunoassay to measure cyclosporin A in plasma and serum samples. Journal of Immunoassay 2(1): 19-32. Doutrelepont, J.M., Abramowicz, D., Borre, B., Lemoine, L., De Pauw, P., Vereerstraeten, P., Vanherweghem, J.L. and Goldman, M . (1993) Prophylactic OKT3: Practical considerations for the prevention of first-dose reactions. Transplantation Proceedings 25(2 Suppl. l):45-46. Dowling, R.D., Zenati, M. , Burckart, G.J. et al. (1990) Aerosolized cyclosporine as single-agent immunotherapy in canine lung allografts. Surgery 108:198-205. Duffy, M.M. and Nestor, A. (1992) Nursing Guidelines for Muromonab-CD3 (OKT3). American Nephrology Nurses' Association Journal 19(5):493-495. Eggesbo, J.B., Lyberg, T., Aspelin, T., Hjermann, I., and Kierulf, P. (1996) Different binding of 125I-LPS to plasma proteins from persons with high or low HDL. Scandinavian Journal of Clinical & Laboratory Investigation 56(6): 533-543. El Tayar, N. , Mark, A.E., Vallat, P., Brunne, R.M., Testa, B. and van Gunsteren, W.F. (1993) Solvent-dependent conformation and hydrogen-bonding capacity of cyclosporin A: Evidence from partition coefficients and molecular dynamics simulations. Journal of Medicinal Chemistry 36(24):3757-3764. Estabrook, A., Berger, C.L., Mittler, R., LoGerfo, P., Hhardy, M. and Edelson, R.L. (1983) Antigenic modulation of human T-lymphocytes by monoclonal antibodies. Transplantation Proceedings 15( 1 ):651 -656. 228 Eto, M. , Yoshikai, Y., Nishimura, Y., Hiromatsu, K., Maeda, K., Nomoto, K., Kong, Y.-Y., Kubo, R.T., Kumazawa, J. and Nomoto, K. (1994) Inhibition of allograft rejection by anti-T-cell receptor -aB monoclonal antibodies preserving resistance to bacterial infection. Immunology 81:198-204. Fahr, A., Nimmerfall, F. and Wenger, R. (1994) Interactions of cyclosporine and some derivatives with model membranes: Binding and ion permeability changes. Transplantation Proceedings 26(5):2837-2841. Fahr, A., Holz, M. and Fricker, G. (1995) Liposomal formulations of cyclosporin A: influence of lipid type and dose on pharmacokinetics. Pharmaceutical Research 12(8): 1189-1198. Fiocchi, R., Mamprin, F., Gamba, A., Glauber, M. , Catania, S., Cattaneo, G., Torre, L., Bertocchi, C , Ruhmann, R., Colombo, D., Binetti, G. and Ferrazzi, P. (1994) Pharmacokinetic profile of cyclosporine in long-term heart transplanted patients treated with a new oral formulation. Transplantation Proceedings 26(5):2994-2995. Faustman, D. (1995) Strategies for circumventing transplant rejection: modification of cells, tissues and organs. Trends in Biotechnology 13(3): 100-105. Filion, M.C. and Phillips, N.C. (1997) Toxicity and immunomodulatory activity of liposomal vectors formulated with cationic lipids toward immune effector cells. Biochimica et Biophysica Acta - Biomembranes 1329(2):345-356. First, M.R., Schroeder, T.J. and Hariharan, S. (1993) OKT3-Induced cytokine-release syndrome: Renal effects (Cytokine Nephropathy). Transplantation Proceedings 25(2 Suppl. l):25-26. Fiske, C. and Subbarow, Y. (1925) The calorimetric determination of phosphorus. Biol. Chem. 66:375-400. Fletcher, M. and Goldstein, A.L. (1987) Recent Advances in the understanding of the biochemistry and clinical pharmacology of interleukin-2. Lymphokine Research 6(1):45-57. Francois, C , Dantal, J., Sorel, M. , Cherel, M. , Miyasaka, M. , Soulillou, J.-P. and Jacques, Y. (1996) Antibodies directed at mouse EL-2-R a and B chains act in synergy to abolish T-cell proliferation in vitro and delayed type hypersensitivity reaction in vivo. Transplant International 9:46-50. Freise, C.E., Hong, K., Clemens, L.E. et al. (1991) Characterization of a cyclosporine-containing liposome. Transplant Proc. 23:473-474. Freise, C.E., Liu, T., Hong, K. et al. (1994) The increased efficacy and decreased nephrotoxicity of a cyclosporine liposome. Transplantation 57:928-932. 229 Frokjaer, S., Hjorth, E.L. and Worts, O. (1984) Stability testing of liposomes during storage. In: Liposome Technology, Volume I, Preparation of Liposomes. Gregoriadis, G., ed. CRC Press, Boca Raton, Florida p.241. Fruman, D.A., Burakoff, S.J. and Bierer, B.E. (1994) Immunophilins in protein folding and immunosuppression. FASEB 8:391-400. Furlanut, M. , Baraldo, M. , Pea, F., Albanese, M.C., Albertini, A., and Puricelli, C. (1994) Effect of fluctuations of blood cyclosporine concentrations on renal function. Transplantation Proceedings 26(5):2574-2575. Gabizon, A., Shiota, R. and Papahadjopoulos, D. (1989) Pharmacokinetics and tissue distribution of doxorubicin encapsulated in stable liposomes with long circulation times. Journal of the National Cancer Institute 81(19):1484-1488. Gennis, R.B. (1977) Protein-lipid interactions. Annual Review of Biophysics & Bioengineering 6:195-238. Gibson, S.M., Shimamum, T., Strauss, G., Maesaka, J.K. (1985) Effect of liposome encapsulation on the toxicity of cyclosporin A. Kidney International 27:340. Gilbert, E.M., Dewitt, C M . , Eiswirth, C.C. Renlund, D.G., Menlove, R.L., Freedman, L.A., Herrick, C M . , Gay, W.A. and Bristow, M.R. (1987a) Treatment of refractory cardiac allograft rejection with OKT3 monoclonal antibody. The American Journal of Medicine 82:202-206. Gilbert, E.M., Eisworth, C.C, Renlund, D.G., Menlove, R.L., DeWitt, C.W., Freedman, L.A., Herrick, C M . Gay, W.A. and Bristow, M.R. (1987b) Use of Orthoclone OKT3 monoclonal antibody in cardiac transplantation: Early experience with rejection prophylaxis and treatment of refractory rejection. Transplantation Proceedings 19(2):45-53. Gilbert, B.E., Wilson, S.Z., Garcon, N.M., Wyde, P.R. and Knight, V. (1993) Characterization and administration of cyclosporine liposomes as a small-particle aerosol. Transplantation 56(4):974-977. Gill, R.G. and Wolf, L. (1995) Immunobiology of cellular transplantation. Cell Transplantation 4(4):361-370. Goldstein, G., Fuccello, A.G. and Norman, D.J. (1986) OKT3 monoclonal antibody plasma levels during therapy and the subsequent development of host antibodies to OKT3. Transplantation 42:507-511. 230 Gonzalo, J.A., Baizeras, E., Gonzalez-Garcia, A., George-Chandy, A., Van Rooijen, N , Martinez, C. and Kroemer, G. (1994) Differential in vivo effects of a superantigen and an antibody targeted to the same T cell receptor. Journal of Immunology 152:1597-1608. Gordon, R.D., Tzakis, A.G., Iwatsuki, S., Todo, S., Esquivel, CO. , Marsh, J.W., Stieber, A., Makowka, L. and Starzl, T.E. (1988) Experience with Orthoclone OKT3 monoclonal antibody in liver transplantation. American Journal of Kidney Diseases 11(2): 141-144. Gorecki, D.C, Jakobisiak, M. , Kruszewski, A , Lasek, W. (1991) Evidence that liposome incorporation of cyclosporine reduces its toxicity and potentiates its ability to prolong survival of cardiac allografts in mice. Transplantation 52:766-769. Graham, R.M. (1994) Cyclosporine: mechanisms of action and toxicity. Cleveland Clinic Journal of Medicine 61(4):308-313. Green, CJ . (1981) Immunosuppression with Cyclosporin A: A review. Diagnostic Histopathology 4:157-174. Gregoriadis, G. N. (1976) The carrier potential of liposomes in biology and medicine. Engl. J. Med. 295:704-710. Grit, M. , Zuidam, N.J., Underberg, W.J. and Crommelin, D.J. (1993) Hydrolysis of partially saturated egg phosphatidylcholine in aqueous liposome dispersions and the effect of cholesterol incorporation on hydrolysis kinetics. Journal of Pharmacy & Pharmacology 45(6):490-495. Grit, M . and Crommelin, D.J. (1992) The effect of aging on the physical stability of liposome dispersions. Chemistry & Physics of Lipids. 62(2): 113-122. Grit, M . and Crommelin, D.J. (1993) Chemical stability of liposomes: Implication for their physical stability. Chemistry & Physics of Lipids 64(l-3):3-18. Groczynski, R.M., Chung, S., Fu, X. -M. , Levy, G., Sullivan, B. and Chen, Z. (1995) Manipulation of skin graft rejection in alloimmune mice by anti-VCAM-1 :VLA-4 but not anti-ICAM-l:LFA-l monoclonal antibodies. Transplant Immunology 3(1):55-61. Guo, L.S., Hamilton, R.L., Goerke, J., Weinstein, J.N. and Havel, R.J. (1980) Interaction of unilamellar liposomes with serum lipoproteins and apoliproteins. Journal of Lipid Research 21(8):993-1003. Hansen, C.B., Moase, E.H., Kao, G.Y., Zalipsky, S. and Allen, T.M. (1994) Antibody conjugation to sterically-stabilized liposomes. Third Annual Liposome Research Days Conference, Vancouver, B.C., Canada. 231 Hansen, C.B., Kao, G.Y., Moase, E.H., Zalipsky, S. and Allen, T.M. (1995) Attachment of antibodies to sterically stabilized liposomes: evaluation, comparison and optimization of coupling procedures. Biochimica et Biophysica Acta 1239:133-144. Harding, J.A., Engbers, C M . , Newman, M.S., Goldstein, N.I. and Zalipsky, S. (1997) Immunogeneicity and pharmacokinetic attributes of poly(ethylene glycol)-grafted immunoliposomes. Biochimica et Biophysica Acta - Biomembranes 1327(2):181-192. Harwood, L.M. (1985) "Dry-Column" Flash Chromatography. Aldrichimica Acta 18(1):25. Haverich, A. (1992) Cyclosporine in heart transplantation. Transplantation Proceedings 24(4):82-84. Haverty, T.P., Sanders, M . and Sheahan, M . (1993) OKT3 treatment of cardiac allograft rejection. The Journal of Heart and Lung Transplantation 12(4):591-598. Haynes, M. , Fuller, L., Haynes, D.H. and Miller, J. (1985) Cyclosporine partitions into phospholipid vesicles and disrupts membrane architecture. Immunology Letters 11:343-349. Heidecke, C.-D., Hancock, W.W., Jakobs, F., Zantl, N. , Kurple, R., Westerholt, S., Sewczik, T., Deusch, K. and Kupiec-Weglinski, J. (1995) a/B-T cell receptor-directed therapy in rat cardiac allograft recipients. Transplantation 59(l):78-84. Heidecke, C.-D., Zantl, N. , Maier, S., Varzaru, A., Hager, B., Kupiec-Weglinski, J. and Hancock, W.W. (1996) Induction of long-term rat renal allograft survival by pretransplant T cell receptor-o/5-targeted therapy. Transplantation 61(2):336-339. Heitman, J., Cardenas, M.E., Breuder, T., Hemenway, C , Muir, R.S., Lim, E. Goetz, L., Zhu, D., Lorenz, M. and Dolinski, K. (1994) Antifungal effects of cyclosporine and FK 506 are mediated via immunophilin-dependent calcineurin inhibition. Transplantation Proceedings 26(5):2833-2834. Hennell, R. and Norman D.J. (1993) Monitoring OKT3 treatment: Pharmacodynamic and pharmacokinetic measures. Transplantation Proceedings 25(2 Suppl.l):83-85. Henrickson, M . , Reid, J., Bellet, J.S., Sawchuk, S.S. and Hirsch, R. (1995) Comparison of in vivo efficacy and mechanism of action of antimurine monoclonal antibodies directed aganist TCR aB (H57-597) and CD3 (145-2C11). Transplantation 60(8):828-835. Hess, A.D., Tutschka P.J. and Santos, G.W. (1981a) Effect of cyclosporin-A on human lymphocyte responses in vitro - induction of tolerance to alloantigens. Transplantation Proceedings 13(1 Pt l):374-378. 232 Hess, A.D., Tutschka, P.J. and Santos, G.W. (1981b) The effect of Cyclosporin A on T-lymphocyte subpopulations. In: Cyclosporin A. White, D.J.G., ed. Elsevier Biomedical Press, Amsterdam, The Netherlands, pp.209-231. Hirsch, R., Eckhaus, M. , Auchincloss Jr., H., Sachs, D.H. and Bluestone, J.A. (1988) Effects of in vivo administration of anti-T3 monoclonal antibody on T cell function in mice: I. Immunosuppression of transplantation responses. The Journal of Immunology 140(11):3766-3772. Hirsch, R., Gress, R.E., Pluznik, D.H., Dckhaus, M . and Bluestone, J.A. (1989) Effects of in vivo administration of anti-CD3 monoclonal antibody on T cell function in mice: II. In vivo activation of T cells. The Journal of Immunology 142(3):737-743. Hirsch, R., Archibald, J. and Gress, R.E. (1991) Differential T cell hyporesponsiveness induced by in vivo adminstration of intact or F(ab')2 fragments of anti-CD3 monoclonal antibody. The Journal of Immunology 147(7):2088-2093. Holland, J.W., Hui, C , Cullis, P.R. and Madden, T.D. (1996) P o l y e t h y l e n e glycol)-lipid conjugates regulate the calcium-induced fusion of liposomes composed of phosphatidylethanolamine and phosphatidylserine. Biochemistry 35(8):2618-2624. Hope, M.J., Bally, M.B., Webb, G., Cullis, P.R. (1985) Production of large unilamellar vesicles by rapid extrusion procedure. Characterization of size distribution, trapped volume and ability to maintain a membrane potential. Biochim Biophys. Acta 812:55-65. Hsieh, H.H., Schreiber, M. , Stowe, N , Novick, S., Deodhar, S., Barna, B., Pippenger, C. and Shamberger, R. (1985) Preliminary Report: The use of liposome-encapsulated cyclosporine in a rat model. Transplantation Proceedings 17(1): 1397-1400. Huang, L. (1983) Liposome-Cell Interactions in vitro. In: Liposomes. Ostro, M.J., ed. Marcel Dekker, N Y . pp. 87-124. Huang, S.K., Lee, K.D., Hong, K., Friend, D.S. and Papahadjopoulos, D. (1992) Microscopic localization of sterically stabilized liposomes in colon carcinoma-bearing mice. Cancer Research 52:5135-5143. Inoue, K. (1974) Permeability properties of liposomes prepared from dipalmitoyllecithin, dimyristoyllecithin, egg lecithin, rat liver lecithin and beef brain sphingomyelin. Biochimica et Biophysica Acta 339(3):390-402. Inoue, K. and Kitagawa, T. (1974) Effect of exogenous lysolecithin on liposomal membranes. Its relation to membrane fluidity. Biochimica et Biophysica Acta 363(3):361-372. 233 Isobe, M , Suzuki, J. Yagita, H., Okumura, K., Yamazaki, S., Nagai, R., Yazaki, Y. and Sekiguchi, M. (1994) Immunosuppression to cardiac allografts and soluble antigens by anti-vascular cellular adhesionmolecule-1 and anti-very late antigen-4 monoclonal antibodies. Journal of Immunology 153(12):5810-5818. Janknegt, R. (1996) Liposomal formulations of cytotoxic drugs. Supportive Care in Cancer 4(4):298-304. Jeyarajah, D.R. and Thistlewaite Jr. J.R. (1993) General aspects of cytokine-release syndrome: Timing and incidence of symptoms. Transplantation Proceedings 25(2 Suppl. 1): 16-20. Jiang, T. and Acosta, D. (1993) An in vitro model of cyclosporine-induced nephrotoxicity. Fundamental and Applied Toxicology 20:486-495. Jones, M.N. and Hudson, M.J.H. (1993) The targeting of immunoliposomes to tumour cells (A431) and the effects of encapsulated methotrexate. Biochimica et Biophysica Acta 1152:231-242. Judge, T.A., Tang, A. and Turka, L.A. (1996) Immunosuppression through blockade of CD28:B7-mediated costimulatory signals. Immunologic Research 15(l):38-49. Juliano, R.L. and Stamp, D. (1975) The effect of particle size and charge on the clearance rates of liposomes and liposome encapsulated drugs. Biochemical and Biophysical Research Communications 63(3):651-658. Kaledin, V.I., Matienko, N.A., Nikolin, V.P., Gruntenko, Y.V., Budker, V.G. (1981) Intralymphatic administration of liposome-encapsulated drugs to mice: possibility for suppression of the growth of tumor metastases in the lymph nodes. Journal of the National Cancer Institute 66(5):881-887. Katzung, B.G., ed. (1989) Basic and Clinical Pharmacology, 4 t h edition, Appleton & Lange, Norwalk, Connecticut, p.937. Kawahara, K., Sutherland, D.E., Rynasiewicz, J. and Najarian, J.S. (1980) Prolongation of heterotopic cardiac allografts in rats by cyclosporin A. Surgery 88:594-600. Kay, J.E. (1989) Inhibitory effects of cyclosporin A on lymphocyte activation. In: Cyclosporin Mode of action and clinical applications, Thompson, A.W. (ed.), Kluwer Academic Publishers, Boston, pp. 1-23. Kensil, C.R. and Dennis, E.A. (1981) Alkaline hydrolysis of phospholipids in model membranes and the dependence on their state of aggregation. Biochemistry 20(21):6079-6085. 234 Kirby, C , Clarke, J., Gregoriadis, G. (1980) Effect of the cholesterol content of small unilamellar liposomes on their stability in vivo and in vitro. Biochem. J. 186:591. Klibanov, A.L., Maruyama, K , Torchilin, V.P. and Huang, L. (1990) Amphipathic polyethyleneglycols effectively prolong the circulation time of liposomes. Federation of European Biochemical Societies 268(l):235-237. Klibanov, A.L., Maruyama, K , Beckerleg, A.M. , Torchilin, V.P. and Huang, L. (1991) Activity of amphipathic poly(ethylene glycol) 5000 to prolong the circulation time of liposomes depends on the liposome size and is unfavorable for immunoliposome binding to target. Biochimica et Biophysica Acta 1062:142-148. Ko, S., Nakajima, Y., Kanehiro, H., Yoshimura, A. and Nakano, H. (1994) The pharmacokinetic benefits of newly developed liposome-incorporated FK506. Transplantation 58(10): 1142-1144. Kohler, G. and Milstein, C. (1975) Continuous cultures of fused cells secreting antibody of predefined specificity. Nature 256(5517):495-497. Kroon, D.J., Baldwin-Ferro, A. and Lalan, P. (1992) Identification of sites of degradation in a therapeutic monoclonal antibody by peptide mapping. Pharmaceutical Research 9(11):13861393. Kung, P.C., Goldstein, G., Reinherz, E.L., Schlossman, S.F. (1979) Monoclonal antibodies defining distinctive human T cell surface antigens. Science 206:346-349. Ladbrooke, B.D. and Chapman, D. (1969) Thermal analysis of lipids, proteins and biological membranes. A review and summary of some recent studies. Chem. Phys. Lipids 3:304-356. Land, W. (1994) Clinical immunosuppressive matrix in organ transplantation. Transplantation Proceedings 26(5):2479-2484. Lasic, D.D. (1997) Liposomes (Chapter 6) In: Liposomes for gene delivery. Lasic, D.D., ed. CRC Press, Boca Raton pp.67-112. Lasic, D.D., Martin, F.J., Gabizon, A., Huang, S.K. and Papahadjopoulos, D. (1991) Sterically stabilized liposomes: a hypothesis on the molecular origin of the extended circulation times. Biochimica et Biophysica Acta 1070:187-192. Latifpour, J., Gousse, A., Yoshida, M. and Weiss, R.M. (1992) Effect of insulin and dietary myoinositol on muscarinic receptor alterations in diabetic rat bladder. The Journal of Urology 147:760-763. Lau, P.P., Cahill, D.J., Zhu, H.J. and Chan, L. (1995) Ethanol modulates apolipoprotein B mRNA editing in the rat. Journal of Lipid Research 36:2069-2078. 235 Laukkanen, M L . , Alfthan, K. and Keinanen, K. (1994) Functional immunoliposomes harboring a biosynthetically lipid-tagged single-chain antibody. Biochemistry 33(38): 11664-11670. Lazarovits, A.L, Poppema, S., Zhang, Z., Khandaker, M. , Le Feuvre, C.E., Singhal, S.K., Garcia, B.M., Ogasa, N , Jevnikar, A.M., White, M.J., Singh, G., Stiller, CR. and Zhong, R.Z. (1996) Prevention and reversal of renal allograft rejection by antibody against CD45RB. Nature 380:717-720. Lee, R.J. and Huang, L. (1996) Folate-targeted, anionic liposome-entrapped polylysine-condensed DNA for tumor cell-specific gene transfer. Journal of Biological Chemistry 271(14):8481-8487. Lee, S. (1987) Experimental Microsurgery. Igaku-Shoin Medical Publishers, Inc., New York. pp. 3-39,151-182,266-276. Lehmann, R., Bhargava, A.S. and Gunzel, P. (1993) Serum lipoprotein pattern in rats, dogs and monkeys, including method comparison and influence of menstrual cycle in monkeys. European Journal of Clinical Chemistry & Clinical Biochemistry 31(10):633-637. Leo, O., Foo, M. , Sachs, D.H., Samelson, L.E. and Bluestone, J.A (1987) Identification of a monoclonal antibody specific for the murine T3 polypeptide. Proceedings in the National Academy of Sciences USA 84:1374-1378. Lim, H.J., Masin, D., Madden, T.D. and Bally, M.B. (1997) Influence of drug release characteristics on the therapeutic activity of liposomal mitoxantrone. Journal of Pharmacology & Experimental Therapeutics 281(l):566-573. Lin, Y , Sobis, H., Vandeputte, M. and Waer, M. (1994) Induction therapy of leflunomide and cyclosporine allows for long-term xenograft survival under cyclosporine alone. Transplantation Proceedings 26(5):3052. Litzinger, D.C, Buiting, A.M.J., van Rooijen, N. and Huang, L. (1993) Effect of liposome size on the circulation time and intraorgan distribuition of amphipathic poly(ethylene glycol)-containing liposomes. Biochimica et Biophysica Acta 1190:99-107. Liu, G., Excoffon, K.J., Benoit, P., Ginzinger, D.G., Miao, L., Ehrenborg, E., Duverger, N , Denefle, P.P., Hay den, M.R. and Lewis, M.E. (1997) Efficient adenovirus-mediated ectopic gene expression of human lipoprotein lipase in human hepatic (HepG2) cells. Human Gene Therapy 8(2):205-14. Lohner, K. (1991) Effects of small organic molecules on phospholipid phase transitions. Chemistry & Physics of Lipids 57(2-3):341-362. 236 Lopez-Berestein, G., Rosenblum, M.G. and Mehta, R. (1984) Altered tissue distribution of amphotericin B by liposomal encapsulation: comparison of normal mice to mice infected with Candida albicans. Cancer Drug Delivery 1(3): 199-205. Loughrey, H.C., Choi, L.S., Wong, K.F., Cullis, P.R. and Bally, M.B. (1993) Preparation of streptavidin-liposomes for use in ligand specific applications. In: Liposome Technology Volume III, Gregoriadis, G., ed. CRC Press, Boca Raton, pp. 163-177. Luke, D.R., Kasiske, B.L., Matzke, G.R., Awni, W.M. and Keane, W.F. (1987) Effects of cyclosporin A on the isolated perfused rat kidney. Transplantation 43(6):795-799. Madden, T.D., Tilcock, C.P.S., Wong, K. and Cullis, P.R. (1988) Spontaneous vesiculation of large multilamellar vesicles composed of saturated phosphatidylcholine and phosphatidylglycerol mixtures. Biochemistry 27(24):8724-8730. Mahon, P.M. (1991) Orthoclone OKT3 and cardiac transplantation: An overview. Critical Care Nurse ll(8):42-48. Martin, F.J., Hubbell, W.L. and Papahadjopoulos, D. (1981) Immunospecific targeting of liposomes to cells: a novel and efficient method for covalent attachment of Fab' fragments via disulfide bonds. Biochemistry 20(14):4229-4238. Martin, P.J., Hansen, J.A., Anasetti, C , Zutter, M. , Durnam, D., Storb, R. and Thomas, E.D. (1988) Treatment of acute graft-versus-host disease with anti-CD3 monoclonal antibodies. American Journal of Kidney Diseases 11(2):149-152. Maruyama K , Takizawa T., Yuda, T., Kennel, S.J., Huang, L. and Iwatsuru, M. (1995) Targetability of novel immunoliposomes modified with amphipathic poly(ethylene glycol)s conjugated at their distal terminals to monoclonal antibodies. Biochimica et Biophysica Acta 1234:74-80. Maruyama, K., Yuda, T., Okamato, A., Kojima, S., Suginaka, A. and Iwatsuru, M . (1992) Prolonged circulation time in vivo of large unilamellar liposomes composed of distearoylphosphatidylcholine and cholesterol contairiing amphipathic poly(ethyleneglycol). Biocimica et Biophysica Acta 1128:44-49. Marz, W. and Gross, W. (1986) Analysis of plasma lipoproteins by ultracentrifugation in a new fixed angle rotor: evaluation of a phosphotungstic acid/MgC12 precipitation and a quantitative lipoprotein electrophoresis assay. Clinica Chimica Acta 160(1):1-18. Mayer, L.D., Hope, M.J. and Cullis, P.R. (1986) Vesicles of variable sizes produced by a rapid extrusion procedure. Biochimica et Biophysica Acta 858:161-168. 237 Mayer, L.D., Tai, L . C , Ko, D.S., Masin, D., Ginsberg, R.S., Cullis, P.R. and Bally M.B. (1989) Influence of vesicle size, lipid composition, and drug-to-lipid ratio on the biological activity of liposomal doxorubicin in mice. Cancer Research 49: 5922-5930. Mayer, L.D., Nayar, R., Thies, R.L., Boman, N.L., Cullis, P.R. and Bally, M.B. (1993) Identification of vesicle properties that enhance the antitumour activity of liposomal vincristine against murine LI210 leukemia. Cancer Chemotherapy & Pharmacology 33:17-24. McMaster, P., Mirza, D.F. (1994) Optimal use of cyclosporin A in liver transplantation. Transplantation Proceedings 26(5):2662-2664. Miescher, G.C, Schreyer, M. and MacDonald, H.R. (1989) Production and characterization of a rat monoclonal antibody against the murine CD3 molecular complex. Immunology Letters 23:113-118. Mignat, C. (1997) Clinically significant drug interactions with new immunosupressive agents. Drug Safety 16(4):267-278. Milani, J.K., Pleyer, U., Dukes, A , Chou, H.J., Lutz, S., Ruckert, D., Schmidt, K.H. and Mondino, B.J. (1993) Prolongation of corneal allograft survival with liposome-encapsulated cyclosporine in the rat eye. Ophthalmology 100(6):890-896. Morgan, J.R., Williams, L.A., Howard, C B . (1985) Technetium-labelled liposome imaging for deep-seated infection. British Journal of Radiology 58:35-39. Mori, A., Kilibanov, A.L., Torchilin, V.P. and Huang, L. (1991) Influence of the steric barrier activity of amphipathic poly(ethyleneglycol) and ganglioside GMj on the circulation time of liposomes and on the target binding of immunoliposomes in vivo. Federation of European Biochemical Societies 284(2):263-266. Morimoto, C , Sato, T. and Tachibana, K. (1996) Selective immunomodulation: utilization of CD29/VLA molecules. Artificial Organs 20(8):828-831. Mowat, J.J., Mok, M.J., MacLeod, B.A. and Madden, T.D. (1996) Liposomal bupivacaine. Extended duration nerve blockade using large unilamellar vesicles that exhibit a proton gradient. Anesthesiology 85(3):635-643. Nicoli, D.F., McKenzie, D.C and Wu, J.-S. (1991) Application of dynamic light scattering to particle size analysis of macromolecules. American Laboratory 23(17):32-37. Niebylski, C D . and Petty, H.R. (1993) Physical characterization of cyclosporine binding sites in lymphocytes. Biophysical Journal 64:701-708. Norman, D.J., Sheild III, CF . and Barry, J.M. (1987) Therapeutic use of OKT3 monoclonal antibody for acute renal allograft rejection. Transplantation 38:695-701. 238 Norman, D.J., Shield III, C.F., Barry, J., Bennett, W.M., Henell, K., Kimball, J., Funnell, B. and Hubert, B. (1988a) Early use of OKT3 monoclonal antibody in renal transplantation to prevent rejection. American Journal of Kidney Diseases 11(2): 107-110. Norman, D.J., Barry, J.M., Bennett, W.M., Leone, M. , Henell, K , Funnell, B. and Hubert, B. (1988b) The use of OKT3 in cadaveric renal transplantation for the rejection that is unresponsive to conventional anti-rejection therapy. American Journal of Kidney Diseases ll(2):90-93. Norman, D.J. and Leone, M.R. (1991) The role of OKT3 in clinical transplantation. Pediatric Nephrology 5(1):130-136. Norman, D.J. (1993) Rationale for OKT3 monoclonal antibody treatment in transplant patients. Transplantation Proceedings 25(2 Suppl. 1): 1-3. Norman, D.J., Kimball, J.A. and Barry, J.M. (1993) Cytokine-release syndrome: Differences between high and low doses of OKT3. Transplantation Proceedings 25(2 Suppl. l):35-38. Nussenblatt, R.B. and Palestine, A.G. (1986) Cyclosporine: immunology, pharmacology and therapeutic uses. Survey of Ophthalmology 31(3): 159-169. Oberhuber, G., Schmid, T., Thaler, W., Kaltenbacher, A., Schirmer, M. , Liliemark, J., Herbst, P., Geisen, F., Margreiter, R. and Konwalinka, G. (1994) Evidence that 2-chlorodeoxyadenosine in combination with cyclosoporine prevents rejection after allogeneic small bowel transplantaton. Transplantation 58(6):743-745. Ogihara, L, Kojimi, S., Jay, M . (1986) Differential uptake of gallium-67-labeled liposomes between tumors and inflammatory lesions in rats. J. Nucl. Med. 27:1300-1307. Ohajekwe, O.A., Hardy, M.A. and Oluwole, S.F. (1995) Prevention of graft-versus-host disease and the induction of transplant tolerance by low-dose UV-B irradiation of B M cells combined with cyclosporine immunosuppression. Transplantation 60(12): 1510-1516. O'Leary, T.J., Ross, P.D., Lieber, M.R. and Levin, I.W. (1986) Effects of cyclosporine A on biomembranes. Biophysical Journal 49:795-801. Olive, D. andMawas, C. (1993) Therapeutic applications of anti-CD4 antibodies. Critical Reviews in Therapeutic Drug Carrier Systems 10(l):29-63. Opelz, G. (1995) Efficacy of rejection prophylaxis with OKT3 in renal transplantation. Transplantation 60(11): 1220-1224. 239 Ortho Multicenter Transplant Study Group (1985) A randomized clinical trial of OKT3 monoclonal antibody for acute rejection of cadaveric renal transplants. New England Journal of Medicine 313:337-341. Ostro, M.J., Cullis, P.R. (1989) Use of liposomes as injectable-drug delivery systems. Am. J. Hosp. Pharm. 46:1576-1587. Ouyang, C , Choice, E., Holland, J., Meloche, M . and Madden T.D. (1995) Liposomal cyclosporine: Characterization of drug incorporation and interbilayer exchange. Transplantation 60(1):999-1006. Palatini, P. (1992) Deposition kinetics of phospholipid liposomes In: Neurobiology of Essential Fatty Acids. Bazan, N.G. et al., ed., Plenum Press, New York, pp.375-391. Papahadjopoulos, D., Allen, T.M., Gabizon, A., Mayhew, E., Matthay, K., Huang, S.K., Lee, K.D., Woodle, M . C , Lasic, D.D., Redemann, C. and Martin, F.J. (1991) Sterically stabilized liposomes: Improvements in pharmacokinetics and antitumor therapeutic efficacy. Proceedings of the National Academy of Sciences U.S.A. 88:11460-11464. Patel, H.M., Tuzel, N.S. and Ryman, B.E. (1983) Inhibitory effect of cholesterol on the uptake of liposomes by liver and spleen. Biochimica et Biophysica Acta 761(2): 142-151. Paul, L . C , Chea, R., Davidoff, A., and Benediktsson, H. (1994) Efficacy of long-term cyclosporine treatment to prevent chronic renal allograft rejection in rat. Transplantation Proceedings 26(5):2567-2568. Palatini, P. (1992) Deposition kinetics of phospholipid liposomes. In: Neurobiology of essential fatty acids. Bazan, N.G., ed. Plenum Press, New York pp.375-391. Perico, N. and Remuzzi, G. (1997) Prevention of transplant rejection: current treatment guidelines and future developments. Drugs 54(4):533-570. Pierce, G.A., Graham, W.K., Kauffman, Jr., H.M., Wolf, J.S. and the United Network for Organ Sharing (UNOS) (1996) The united network for organ sharing: 1984 to 1994. Transplantation Proceedings 28(1): 12-15. Proffitt, R.T., Williams, L.E., Presant, C A , Tin, G.W., Uliana, J.A., Gamble, R.C. and Baldeschwieler, J.D. (1983) Liposomal blackade of the reticuloendothelial system: improved tumor imaging with small unilamellar vesicles. Science 220(4596):502-505. Puschett, J.B., Greenbery, A., Holley, J. and McCauley, J. (1990) The spectrum of ciclosporin nephrotoxicity. Am. J. Nephrol. 10:296-309. Radhakrishnan, J. and Cohen, D.J. (1993) Cytokine-release syndrome: General risk-factor modification - Preparation of high-risk patients for use of OKT3. Transplantation Proceedings 25(2 Suppl. l):60-62. 240 Rao, P.E. and Kroon, D.J. (1993) Orthoclone OKT3: Chemical mechanisms and functional effects of degradation of a therapeutic monoclonal antibody. In: Stability and characterization of protein and peptide drugs. Wang, Y.J. nad Pearlman, R., eds., Plenum Press, New York. pp. 135-158. Reiss-Husson, F. and Luzzati, V. (1967) Phase transitions in lipids in relation to the structure of membranes. Advances in Biological & Medical Physics 11:87-107. Ren, E.C. (1991) Cellular and molecular approaches to developing human monoclonal antibodies as drugs. Annals of the Academy of Medicine 20(l):61-70. Renlund, D.G., O'Connell, J.B., Gilbert, E.M., Hammond, M.E., Burton, N.A., Jones, K.W., Karwande, S.V., Doty, D.B., Menlove, R.L., Herrick, C M . , Lee, H.R., Gay Jr., W.A. and Bristow, M.R. (1989) A prospective comparison of murine monoclonal CD-3 (OKT3) antibody-based and equine antithymocyte globulin-based rejection prophylaxis in cardiac transplantation. Transplantation 47(4):599-605. Riesbeck, K., Schatz, H., Ostraat, O., Tufveson, G., Forsgren, A. and Ekberg (1995) Enhancement of the immunosuppressive effect of cyclosporin A by ciprofloxacin in a rat cardiac allograft transplantation model. Transplant International 8:96-102. Riondel, J., Jacrot, M. , Fessi, H., Puisieux, F. and Potier (1992) Effects of free and liposome-encapsulated taxol on two brain tumors xenografted into nude mice. In vivo 6(l):23-27. Ritters, B., Garbensee, B. and Heering, P. (1994) Malignancy under immunosuppressive therapy including cyclosporine. Transplantation Proceedings 26(5):2656-2657. Robinson, S.T. (1993) Administration of OKT3 in the operating room. Transplantation Proceedings 25(2 Suppl. l):41-42. Robinson, S.T., Barry, J.M. and Norman, D.J. (1993) The hemodynamic effects of intraoperative injection of muromonab CD3. Transplantation 56(2):356-358. Rodrigueza, W.V., Pritchard, P.H and Hope, M.J. (1993) The influence of size and composition on the cholesterol mobilizing properties of liposomes in vivo. Biochimica et Biophysica Acta. 1153(1):9-19. Roitt, I (ed.) (1994) Essential Immunology, 8 th edition. Blackwell Scientific Publications, London, England, pp. 1-56. Rossi, S.J., Schroeder, T.J., Hariharan, S. and First, M.R. (1993) Prevention and management of the adverse effects associated with immunosuppressive therapy. Drug Safety 9(2): 104-131. 241 Routledge, E.G., Falconer, M.E., Pope, H., Lloyd, I.S. and Waldmann H. (1995) The effect of aglycosylation on the immunogenicity of a humanized therapeutic CD3 monoclonal antibody. Transplantation 60(8):847-853. Russell, R.G.G., Graveley, R., Coxon, F., Skjodt, H., Del Pozo, E., Elford, P. and Mackenzie, A. (1992) Cyclosporin A. Mode of action and effects on bone and joint tissues. Scandinavian Journal of Rheumatology 21(Suppl. 95):9-18. Russell, S.J., Llewelyn, M.B. and Hawkins, R.E. (1992) Principles of antibody therapy. British Medical Journal 305:1424-1429. Sanfilippo, F. and Baldwin, W.M. (1997) Antibody and complement in graft rejection. Transplantation Proceedings 29(1-2):179-180. Sawchuk, S.S., Gates, R. and Hirsch, R. (1995) Contrasting in vivo effects on T helper cell functions induced by mitogenic (intact) versus nonmitogenic (F(ab')2) anti-CD3 monoclonal antibody. Transplantation 60(11):1331-1337. Scherphof, G., Roerdink, F., Waite, M. and Parks, J. (1978) Disintegration of phosphotidylcholine liposomes in plasma as a result of interaction with high-density lipoproteins. Biochimica et Biophysica Acta 542(2):296-307. Schreiber, S.L. and Crabtree, G.R. (1992) The mechanism of action of cyclosporin A and FK506. Immunology Today 13(4): 136-142. Schroeder, T.J., Hariharan, S, and First, M.R. (1993) Antibody response to OKT3 and methods for monitoring. Transplantation Proceedings 25(2 Suppl. 1):77-80. Schroeder, T.J., Michael, AT . , First, M.R., Hariharan, S., Bhat, G., Hanto, D.W., Ryckman, F.C. and Balistreri, W.F. (1994) Variations in serum OKT3 concentration based upon age, sex, transplanted organ, treatment regimen, and anti-OKT3 antibody status. Therapeutic Drug Monitoring 16:361-367. Schultz, S.L. and Meriney, D.K. (1996) Cyclosporine and tacrolimus: a comparison of immunosuppressants used for liver transplantation. Dimensions of Critical Care Nursing 15(4):187-197. Schulz, M. , Schuurman, H-J., Joergensen, J., Steiner, C , Meerloo, T., Kagi, D., Hengartner, H., Zinkernagel, R.M., Schreier, M.H., Burki, K. and Ledermann, B. (1995) Acute rejection of vascular heart allografts by perforin-deficient mice. European Journal of Immunology 25:474-480. Senior, J.H. (1987) Fate and behavior of liposomes in vivo: A review of controlling factors. Critical reviews in therapeutic drug carrier systems 3(2): 123-193. 242 Senior, J., Delgado, C , Fisher, D., Tilcock, C. and Gregoriadis, G. (1990) Influence of surface hydrophilicity of liposomes on their interaction with plasma protein and clearance from the circulation: Studies with poly(ethylene glycol)-coated vesicles. Biochimica et Biophysica Acta 1062:77-82. Shahinian, S. and Silvius, J.R. (1995) Doubly-lipid-modified protein sequence motifs exhibit long-lived anchorage to lipid bilayer membranes. Biochemistry 34(11):3813-22. Shield III, C.F. (1993) Consequences of anti-OKT3 antibody development: OKT3 reuse and long-term graft survival. Transplantation Proceedings 25(2 Suppl. l):81-82. Shinkai, Y., Rathbun, G., Lam, K.P., Oltz, E.M., Stewart, V., Mendelsohn, M. , Charron, J., Datta, M. , Young, F., Stall, A.M. and Alt, F.W. (1992) RAG-2-deficient mice lack mature lymphocytes owing to inability to initiate V(D)J rearrangement. Cell 68(5):855-867. Silvius, J.R. and Zuckermannn, M.J. (1993) Interbilayer transfer of phospholipid-anchored macromolecules via monomer diffusion. Biochemistry 32(12):3153-3161. Singh, A.K., Kilpatrick, P.K. and Carbonell, R.G. (1996) Application of antibody and fluorophore-derivatized liposomes to heterogeneous immunoassays for D-dimer. Biotechnology Progress 12(2):272-280. Slotte, J.P., Jungner, M. , Vilcheze, C. and Bittman, R (1994) Effect of sterol side-chain structure on sterol-phosphatidylcholine interactions in monolayers and small unilamellar vesicles. Biochimica et Biophysica Acta. 1190(2):435-443. Snell, G.D., Dausset, J. and Nathenson, S.G. (eds.) (1976) Histocompatibility. New York, Academic Press pp. 1-401. Sollinger, H.W. and Stratta, R.J. (1988) Experience with OKT3 in vascularized pancreas transplantation. American Journal of Kidney Diseases 11(2): 145-148. Sriwatanawongsa, V., Davies, H.F. and Calne R.Y. (1994) Successful transplantation of rat liver grafts. Transplantation 59(6):908-910. Stapleton, D.D., Ventura, H.O., Grundtner, S.E., Smart, F.W., Price, H.L., Van Meter, C. and Ochsner, J.L. (1993) Induction immunosuppression with the monoclonal antibody OKT3 after cardiac transplantation. The American Journal of the Medical Sciences 306(1):16-19. Starzl, T.E., Demtris, A.J., Rao, A.S., Thomson, A.W., Trucco, M. , Murase, N. , Zeevi, A. and Fontes, P. (1994) Spontaneous and iatrogenically augmented leukocyte chimerism in organ transplant recipients. Transplantation Proceedings 26(5):3071-3076. 243 Stillwell, W., Ehringer, W.D., Dumaual, A.C. and Wassail, S.R. (1994) Cholesterol condensation of alpha-linolenic and gamma-linolenic acid-containing phosphatidylcholine monolayers and bilayers. Biochimica et Biophysica Acta. 1214(2):131-136. Stuart, A.E. (1970) The Reticuloendothelial System. E. & S. Livingstone, Edinburgh and London, pp.349-387. Stuhne-Sekalec, L. and Stanacev, N. (1991a) Liposomes as cyclosporin A carriers: ESR study of cyclosporin A interaction with l,2-dimyristoyl-sn-3-phosphatidylglycerol liposomes. Journal of Microencapsulation 8(4):455-463. Stuhne-Sekalec, L. and Stanacev, N.Z. (1991b) Liposomes as carriers of cyclosporin A. J. Microencapsulation 8:441-446. Suleymanlar, G., Guleymanlar, I., Shapiro, J.I. and Chan, L. (1994a) Possible role of lipid peroxidation in cyclosporine nephrotoxicity in rats. Transplantation Proceedings 26(5):2888-2889. Suleymanlar, G., Guleymanlar, I., Shapiro, J.I. and Chan, L. (1994b) Efficacy of verapamil in prevention of cyclosporine nephrotoxicity in the rat. Transplantation Proceedings 26(5):2896-2898. Sutherland, D.E.R., Moudry-Munns, K.C. and Gruessner, A. (1994) Pancreas transplant outcome with or without biological anti-T-cell therapy for induction immunosuppression with use of cyclosporine. Transplantation Proceedings 26(5):2752-2755. Szoka Jr., F. and Papahadjopoulos, D. (1980) Comparative properties and methods of preparation of lipid vesicles (liposomes). Annual Review of Biophysics & Bioengineering 9:467-508. Szoka, F., Olson, F., Heath, T., Vail, W., Mayhew, E. and Papahadjopoulos, D. (1980) Preparation of unilamellar liposomes of intermediate size (0.1-0.2 mumol) by a combination of reverse phase evaporation and extrusion through polycarbonate membranes. Biochimica et Biophysica Acta 601(3):559-571. Tabbara, K.F., Gee, S.S., Alvarez, H. (1989) Ocular bioavailability of cyclosporine after oral administration. Transplantation Proceedings. 20:656-659. Tanabe, M. , Todo, S., Murase, N., Irish, W., Miyazawa, H., Fujisaki, S., and Starzl, T.E. (1994) Combined immunosuppressive therapy with low dose FK506 and antimetabolites in rat allogeneic heart transplantation. Transplantation 58(l):23-27. Theriault, Y., Logan, T.M., Meadows, R., Yu, L., Olejniczak, E.T., Holzman, T.F., Simmer, R.L. and Fesik, S.W. (1993) Solution structure of the cyclosporin A/cyclophilin complex by NMR. Nature 361:88-91. 244 Thiel, G., Bock, A., Spondlin, M. , Brunner, F.P., Mihatsch, M. , Rufli, T. and Landmann, J. (1994) Long-term benefits and risks of cyclosporin A (Sandimmun) - An analysis at 10 years. Transplantation Proceedings 26(5):2493-2498. Thomas, S.E. and Gordon, D.S. (1986) Cyclosporine. Southern Medical Journal 79(2):205-214. Tibell, A. and Norrlind (1994) Cyclosporin A in fat emulsion carriers: studies on the immunosuppressive potential, using the heterotopic heart transplant model in rats. Transplant International 7(6):43 8-441. Timmerman, L.A., Clipstone, N.A., Ho, S.N., Northrop, J.P. and Crabtree, GR. (1996) Rapid shuttling of NF-AT in discrimination of Ca 2 + signals and immunosuppression. Nature 383:837-840. Tomonari, K. (1988) A rat antibody against a structure functionally related to the mouse T-cell receptor/T3 complex. Immunogenetics 28:455-458. Torchilin, V.P. and Papisov, M.I. (1994) Why do polyethylene glycol-coated liposomes circulate so long? Journal of Liposome Research 4(l):725-739. Tsuchida, M. , Hirahara, H., Matsumoto, Y., Abo, T. and Eguchi, S. (1994) Induction of specific unresponsiveness to cardiac allografts by short-term administration of anti-T cell receptor aB antibody. Transplantation 57(2):256-262. Unezaki, S., Maruyama, Kazuo, Hosoda, J-L, Nagae, I., Koyanagi, Y., Nakata, M . , Ishida, O., Iwatsuru, M. and Tsuchiya, S. (1996) Direct measurement of the extravasation of polyethyleneglycol-coated liposomes into solid tumor tissue by in vivo fluorescence microscopy. International Journal of Pharmaceutics 144:11-17. Vadiei, K., Lopez-Berestein, G., Perez-Solar, R., Luke, D.R. (1991) Tissue distribution and in vivo immunosuppressive activity of liposomal cyclosporine. Drug Metab. Dispos. 19:1147-1151. Vadiei, K., Perez-Solar, R., Lopez-Berestein, G., Luke, D.R. (1989) Pharmacokinetic and pharmacodynamic evaluation of liposomal cyclosporine. Int. J. Pharm. 57:125-131. van Wauwe, J.P., de Mey J.R. and Goossens, J.G. (1980) OKT3: A monoclonal anti-human T lymphocyte antibody with potent mitogenic properties. The Journal of Immunology 124(6):2708-2713. Vedie, B., Myara, I., Pech, M.A., Maziere, J.C., Maziere, C , Caprani, A. and Moatti, N. (1991) Fractionation of charge-modified low density lipoproteins by fast protein liquid chromatography. Journal of Lipid Research. 32(8): 1359-69. 245 Venkataram, S., Awni, W.M., Jordan, K., Rahman, Y.E. (1990) Pharmacokinetics of two alternative dosage forms for cyclosporine: liposomes and intralipid. J. Pharm. Sci. 79:216-219. Vossen, A.C.T.M., Tibbe, G.J.M., van Oudenaren, A., Vredendaal, A.E.C.M., Benner, R. and Savelkoul, H.F.J. (1994a) A rat anti-mouse CD3 monoclonal antibody induces long-term skin allograft survival without inducing side effects. Transplantation Proceedings 26(6):3157-3158. Vossen, A.C.T.M., Knulst, A.C., Tibbe, G.J.M., van Oudenaren, A., Baert, M.R.M., Benner, R. and Savelkoul, H.F.J. (1994b) Suppression of skin allograft rejection in mice by anti-CD3 monoclonal antibodies without cytokine-related side effects. Transplantation 58(2):257-261. Vossen, A.C.T.M., Tibbe, G.J.M., Benner, R. and Savelkoul, H.F.J. (1995a) T-lymphocyte and cytokine-directed strategies for inhibiting skin allograft rejection in mice. Transplantation Proceedings 27(l):380-382. Vossen, A.C.T.M., Tibbe, G.J.M., Kroos, M.J., van de Winkel, J.G.J., Benner, R. and Savelkoul, H.F.J. (1995b) Fc receptor binding of anti-CD3 monocloanal antibodies is not essential for immunosuppression, but triggers cytokine-related side effects. European Journal of Immunology 25:1492-1496. Waldmann, H. (1988) Monoclonal antibodies for organ transplantation: Prospects for the future. American Journal of Kidney Diseases 11(2): 154-158. Waldmann, H. and Cobbold, S. (1993) The use of monoclonal antibodies to achieve immunological tolerance. Immunology Today 14(6): 247-251. Wall, W.J., Ghent, C.N., Roy, A., McAlister, V.C., Grant, D.R.and Adams, P.C. (1995) Use of OKT3 monoclonal antibody as induction therapy for control of rejection in liver transplantation. Digestive Diseases and Sciences 40(l):52-57. Wang, S., Lee, R.J., Cauchon, G., Gorenstein, D.G. and Low, P.S. (1995) Delivery of antisense oligodeoxyribonucleotides against the human epidermal growth factor receptor into cultured KB cells with liposomes conjugated to folate via polyethylene glycol. Proceedings of the National Academy of Sciences of the United States of America 92(8):3318-22. Wang, D.-C, Taraschi, T.F., Rubin, E. and Janes, N. (1993) Configurational entropy is the driving force of ethanol action on membrane architecture. Biochimica et Biophysica Acta 1145(1): 141-148. Wassef, N.M., Matyas, G.R., and Alving, C.R. (1991) Complement-dependent phagocytosis of liposomes by macrophages: Suppressive effects of "stealth" lipids. Biochemical and Biophysical Research Communications 176(2) 866-874. 246 Wauters, IP., Rosman, J.B., Descombes, E., Genton, A. and Mosimann, F. (1994) Ten-year experience with low-dose cyclosporine A in renal transplantation. Transplantation Proceedings 26(5):2528-2529. White, D.J.G., Rolles, K., Ottawa, T. and Turell, O. (1980) Cyclosporin A-induced long term survival of fully incompatible skin and heart grafts in rats. Transplantation Proceedings 12(2):261-265. Wiedmann, T.S., Trouard, T., Shekar, S.D. et al. (1990) Interaction of cyclosporin A with dipalmitoylphosphatidylcholme. Biochim. Biophys. Acta 1023:12-18. Wilde, M.I. and Goa, K.L. (1996) Muromonab CD3: A reappraisal of its pharmacology and use as prophylaxis of solid organ transplant rejection. Drugs 51(5): 865-894. Williams, G.M., Irwin, B., Burdick, J. and Pennington, L. (1986) Intravenous cyclosporine and kidney function: The John Hopkins Experience. Transplantation Proceedings 18(2, Suppl. l):66-68. Wingard Jr., L.B. (1991) Human pharmacology: molecular-to-clinical, Mosby-Year Book, St. Louis, p. 819. Wissing, M. , Marchant, A., Moser, M. , Flamand, V., Leo, O., Abramowicz, D., Urbain, J. and Goldman, M . (1991) Modulation of murine host-versus-graft disease by anti-CD3 monoclonal antibody. Clinical experimental Immunology 83:333-337. Wolf, A., Clemann, N. , Frieauff, W., Ryffel, B. and Cordier, A. (1994) Role of reactive oxygen formation in the cyclosporin-A-mediated impairment of renal functions. Transplantation Proceedings 26(5):2902-2907. Woodle, M.C. and Lasic, D.D. (1992) Sterically stabilized liposomes. Biochimica et Biophysica Acta 113:171-199. Young, J.M., Spires, D.A., Bedord, C.J., Wagner, B., Balleron, S.J. and De Young, L .M. (1984) The mouse ear inflammatory response to topical arachidonic acid. The Journal of Investigative Dermatology 82(4):367-371. Zalipsky, S. (1993) Synthesis of an end-group functionalized polyethylene glycol-lipid conjugate for preparation of polymer-grafted liposomes. Bioconjugate Chemistry 4(4):296-299. 247 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items