Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Development and characterization of a liposomal subunit vaccine against Neisseria Gonorrhoeae Parmar, Manjeet M. 1999

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

Item Metadata

Download

Media
831-ubc_1999-464059.pdf [ 8.98MB ]
Metadata
JSON: 831-1.0099437.json
JSON-LD: 831-1.0099437-ld.json
RDF/XML (Pretty): 831-1.0099437-rdf.xml
RDF/JSON: 831-1.0099437-rdf.json
Turtle: 831-1.0099437-turtle.txt
N-Triples: 831-1.0099437-rdf-ntriples.txt
Original Record: 831-1.0099437-source.json
Full Text
831-1.0099437-fulltext.txt
Citation
831-1.0099437.ris

Full Text

DEVELOPMENT AND CHARACTERIZATION OF A LIPOSOMAL SUBUNIT VACCINE AGAINST NEISSERIA GONORRHOEAE By MANJEET M PARMAR B.Sc, The University of British Columbia, 1994 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 September 1999 © Manjeet M. Parmar, 1999 In p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l l m e n t of the requirements f o r an advanced degree at the U n i v e r s i t y of B r i t i s h Columbia, I agree t h a t the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and study. I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e copying o f t h i s t h e s i s f o r s c h o l a r l y purposes may be gran t e d by the head of my department or by h i s or her r e p r e s e n t a t i v e s . I t i s understood t h a t copying or p u b l i c a t i o n of t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be allowed without my w r i t t e n p e r m i s s i o n . Department of P^,4AM/J^C^C>6^ dM>) TtftifflfcajICS The U n i v e r s i t y o f B r i t i s h Columbia Vancouver, Canada Date Serr^ZeZ } 11 A B S T R A C T This thesis is concerned with the development and characterization of a subunit vaccine against Neisseria gonorrhoeae. The major gonococcal outer membrane protein, Protein I (Por), was selected as the subunit component because it is antigenically conserved between gonococcal strains. Isolated proteins, however, often do not elicit a protective immune response, either because they are not efficiently taken up by antigen presenting cells (APC) or because antibodies generated against the denatured protein do not recognize the native conformation. In the present research, therefore, Protein I was reconstituted into liposomes. These lipid bilayer structures should be capable of maintaining Por in its native conformation and may be efficiently accumulated by APC. The protein reconstitution process was characterized with regard to the efficiency of protein insertion and the ease of detergent removal. Subsequently, the Por proteoliposomes were characterized as to their size, lamellarity, Por orientation in the bilayer and antibody binding efficiency. The detergents octyl glucopyranoside (OGP) and sodium cholate were compared with respect to efficiency of detergent removal and protein incorporation. The rate of OGP removal was greater than for cholate during dialysis. During OGP-mediated reconstitution, essentially complete protein incorporation was achieved at a protein-lipid ratio of 0.01:1. It was observed, however, that the degree of protein incorporation was dependent on the initial protein-lipid ratios. Increasing the concentration of Por protein relative to phospholipid in the reconstitution mixture resulted in inefficient protein incorporation at ratios of 0.02:1 or higher. Reconstitution studies using cholate indicated that protein insertion into liposomes was less efficient than during OGP-mediated reconstitution at the same initial protein-lipid ratios. Subsequent experiments examined Por reconstitution into liposomes consisting of a Ill solely bilayer-forming lipid, 1-palmitoyl, 2-oleoyl phosphatidylcholine (POPC) or mixtures of POPC with a non-bilayer-forming lipid, 1-palmitoyl, 2-oleoyl phosphatidylethanolamine (POPE). These studies showed no significant differences in incorporation as a function of POPC POPE ratio. Examination of Por orientation in these proteoliposomes suggested that over 80% of the protein was oriented facing outwards in the same "hairpin loop" fashion found in the native bacterial membrane. The conclusion from these studies is that OGP is a preferred detergent for protein reconstitution, providing efficient insertion into the liposomal bilayer in an orientation comparable to the native conformation. Por protein was also reconstituted into liposomes containing positively or negatively charged lipids. Protein reconstitution into systems composed of negatively charged lipids, 1-palmitoyl, 2-oleoyl phosphatidylserine (POPS) or 1-palmitoyl, 2-oleoyl phosphatidylglycerol (POPG), exhibited similar protein incorporation efficiencies as neutral POPC systems when the acidic lipid was present at 5% (by wt). However, increasing the amount of anionic lipid up to 25% resulted in a decrease in protein incorporation efficiency. Essentially complete protein incorporation was achieved when Por was reconstituted into positively charged, dioleyl dimethylammonium chloride- or DODAC-containing liposomes at 5, 10 or 25% DODAC. Interestingly, increasing the concentration of cationic lipid resulted in a shift of the protein/lipid peak towards the top of the isopycnic density gradient, indicating a decrease in density of the reconstituted systems. These results indicate that efficient protein incorporation can be achieved for liposomal systems of differing lipid composition. Reconstituted Por proteoliposomes were characterized by quasi-elastic light scattering size analysis (QELS) and cryo-electron microscopy (CTEM) to determine proteoliposome size and morphology. Such systems exhibited a mean vesicle diameter of greater than 0.3 iv microns and were observed as heterogeneous structures with regard to size and lamellarity using CTEM. A potential subunit vaccine would have to be sterilized before it could be safe for use in humans. Conventional techniques such as heat or steam sterilization would be inappropriate due to potential denaturation of the protein subunit. Terminal filtration through 0.2 micron filters would be adequate for vaccine sterilization; however, reconstituted systems with a mean diameter of 0.3 microns would be too large and unsuitable for direct sterile filtration. Therefore, these proteoliposomes would have to be size-reduced prior to the filtration/sterilization step. In this thesis, it is shown that reconstituted systems can be size-reduced to 100 nm unilamellar vesicles by extrusion, without significant loss of protein or lipid. These extruded systems were then suitable for sterilization by terminal filtration. In a comparative study, the reconstitution of meningococcal outer membrane protein (MOMP) was characterized with regard to degree of protein insertion and detergent removal. The rate of octyl glucoside removal during MOMP reconstitution followed the same kinetics as seen for gonococcal Por reconstitutions. However, the efficiency of protein incorporation was lower than that of Por incorporation at the same initial protein and lipid concentrations. In addition, MOMP reconstitution was examined in the presence of a zwitterionic detergent, Empigen BB, which has been shown to restore the antigenicity of purified meningococcal proteins. Isopycnic density gradient centrifugation studies showed that liposomes were not formed, and hence no protein incorporation occurred during dialysis from an Empigen BB-containing reconstitution mixture. The results of this comparative study would suggest that protein reconstitution occurs more efficiently in the case of a single polypeptide, such as Por, than for mixtures of different membrane proteins, such as MOMP. ELISA assays were performed to determine the antibody binding activities of various Por liposome formulations using both anti-Por monoclonal antibodies and immunized rabbit sera. Consistently higher levels of antibody binding were obtained for Por liposomes prepared as described herein compared to reconstituted systems prepared as described in earlier publications from a different research group. Moreover, neutral proteoliposomes had a higher antibody binding activity compared to negatively or positively charged liposomes. Following the in vitro antigenicity analysis, the in vivo immunogenic properties of Por proteoliposomes and free Por were characterized and compared in a murine model. Mice were immunized by intraperitoneal or intradermal injections of either free Por protein or proteoliposomes containing neutral, cationic or anionic lipids. In addition to the overall antibody titers, the immune sera were characterized with regard to the ratio of the particular serotypes to determine whether the immune response elicited was humoral, indicated by elevated IgGl, or cell-mediated, marked by a predominant IgG2a response. Analysis of mouse immune sera showed that neutral and positively charged proteoliposomes, as well as free Por, induced similar antibody titers and these titers were greater than titers elicited by anionic proteoliposomes. These differences in immune response were seen following administration by either the intraperitoneal or intradermal routes of immunization. Cationic proteoliposomes, however, induced irritation and inflammation at the site of intradermal injection. Examination of the antibody serotypes in the immune sera indicated that immunization via the intraperitoneal route induced predominant IgGl responses whereas intradermal inoculation elicited greater IgG2a antibody titers. These results suggest that intradermal immunization might be more effective than the intraperitoneal route for vi generating a cell-mediated immune response, which is required for fighting intracellular infections. vii T A B L E O F C O N T E N T S Page ABSTRACT ii T A B L E OF CONTENTS vii LIST OF FIGURES xi LIST OF TABLES xiv ABBREVIATIONS xv A C K N O W L E D G E M E N T S xviii DEDICATION xix CHAPTER 1: INTRODUCTION 1 1.1 Vaccines and Immunology 1 1.1.1 Vaccine strategies 2 1.1.2 Humoral immunity 4 1.1.3 Cell-mediated immunity 10 1.2 Neisseria gonorrhoeae 16 1.2.1 Characteristics of the gonococcal organism 16 1.2.2 Characteristics of gonococcal infection 19 1.2.3 Treatment of gonococcal infection 20 1.2.4 Potential vaccine target antigens 23 1.3 Gonococcal Protein I 26 1.3.1 Characteristics of gonococcal protein I 26 1.3.2 Structure of protein I 30 1.3.3 Function of protein I 33 1.3.4 Antigenic properties of gonococcal protein I 34 1.4 Immunological Adjuvants 36 1.4.1 Liposomes: Model biomembranes 37 1.4.2 Liposome preparation and characterization 41 1.4.3 Liposomes as drug delivery systems 47 1.4.4 Liposomes as immunological adjuvants 48 viii 1.5 Research Hypotheses 52 1.6 Specific Research Objectives 52 CHAPTER 2: MATERIALS A N D METHODS 54 2.1 Lipids, chemicals, and reagents 54 2.2 Reconstitution of gonococcal protein I (Por) into liposomes 55 2.2.1 Reconstitution of soluble Por 55 2.2.2 Reconstitution of lyophilized Por 56 2.2.3 Reconstitution of Por into anionic and cationic proteoliposomes 57 2.3 Reconstitution of meningococcal proteins (MOMP) 58 2.4 Analytical Procedures 59 2.4.1 Isopycnic density gradient centrifugation 59 2.4.2 Protein and phospholipid quantitation 59 2.5 Protease digestion 60 2.6 SDS-Polyacrylamide gel electrophoresis 60 2.7 Gel scanning densitometry 61 2.8 Size reduction of reconstituted proteoliposomes by extrusion 61 2.9 Quasi-elastic light scattering (QELS) 62 2.10 Cryo-transmission electron microscopy 62 2.11 Antibody binding evaluation 62 2.11.1 Sample preparation for antigenicity tests 62 2.11.2 Porin monoclonal antibodies 63 2.11.3 ELISA and inhibition ELISA assays 63 2.12 Antibody binding activity of anionic and cationic Por proteoliposomes 65 2.12.1 Biotinylation of goat anti-mouse IgG 65 2.12.2 ELISA antibody binding assays 65 2.13 In vivo antigenicity of Por proteoliposomes 67 2.13.1 Immunization 67 2.13.2 ELISA antibody assays 67 2.13.3 ELISA antibody isotyping assays 68 IX 2.14 Tissue histology 69 2.15 Statistical methods 69 CHAPTER 3: FACTORS INFLUENCING PROTEIN 70 INCORPORATION INTO LIPOSOMES 3.1 INTRODUCTION 70 3.2 RESULTS 73 3.2.1 Detergent removal during reconstitution of Por protein 73 3.2.2 Influence of Por protein/lipid ratio on reconstitution efficiency 76 3.2.3 Proteoliposome size 83 3.2.4 Reconstitution of meningococcal outer membrane protein (MOMP): 85 Residual detergent levels 3.2.5 Characterization of M O M P incorporation during reconstitution 87 3.2.6 Empigen BB-mediated M O M P reconstitution into liposomes 90 3.3 DISCUSSION 92 CHAPTER 4: INFLUENCE OF LIPID COMPOSITION ON POR 98 RECONSTITUTION A N D CHARACTERIZATION OF THE RESULTING PROTEOLIPOSOMES 4.1 INTRODUCTION 98 4.2 RESULTS 101 4.2.1 Por Protein reconstitution determined by isopycnic density 101 gradient centrifugation 4.2.2 Por orientation in reconstituted proteoliposomes 105 4.2.3 Size reduction of reconstituted Por proteoliposomes 109 4.2.4 Reconstitution of Por into liposomes composed of charged lipids 113 4.2.5 Vesicle morphologies 120 4.3 DISCUSSION 127 CHAPTER 5: THE ANTIGENIC CHARACTERIZATION OF POR 130 PROTEOLIPOSOMES 5.1 INTRODUCTION 13 0 5.2 RESULTS 132 5.2.1 Antibody binding to Por proteoliposomes determined using 132 an ELISA assay 5.2.2 In vitro antibody binding activity of charged Por proteoliposomes 137 5.2.3 In vivo immune responses to proteoliposomes 137 5.2.4 Antibody isotyping of immune sera 142 5.2.5 Histology 145 5.3 DISCUSSION 147 CHAPTER 6: SUMMARY 152 REFERENCES 157 X I L I S T O F F I G U R E S Page Figure 1.1 6 The humoral immune response Figure 1.2 9 The pathways of complement activation Figure 1.3 12 The cell-mediated immune response Figure 1.4 18 The surface components of the gonococcal outer membrane Figure 1.5 28 Models of proteins IA and IB orientations in the gonococcal outer membrane Figure 1.6 31 D N A sequence of Protein IA of gonococcal strain FA19 Figure 1.7 32 D N A sequence of Protein IB of gonococcal strain RIO Figure 1.8 38 The eukaryotic plasma membrane Figure 1.9 40 General phospholipid structure Figure 1.10 42 Structure of lipid model membranes Figure 111 45 Protein reconstitution into liposomes Figure 3.1 74 N-Octyl-(3-D-glucopyranoside levels during dialysis Figure 3.2 75 Sodium cholate levels during dialysis Figure 3.3 78 OGP-mediated Por reconstitution into POPC liposomes (P/L=0.01) X l l Figure 3.4 80 OGP-mediated Por reconstitution into POPC liposomes (P/L=0.02) Figure 3.5 82 Cholate-mediated Por reconstitution into POPC liposomes (P/L=0.01) Figure 3.6 86 OGP levels during MOMP reconstitution Figure 3.7 89 OGP-mediated MOMP reconstitution into POPC liposomes Figure 3.8 91 Empigen BB/cholate-mediated MOMP reconstitution Figure 4.1 102 Isopycnic density gradient centrifugation profile for Por reconstituted into POPC liposomes Figure 4.2 103 Isopycnic density gradient centrifugation profile for Por reconstituted in POPC POPE (1:1) liposomes Figure 4.3 104 Isopycnic density gradient centrifugation profile for insoluble Por reconstituted in POPC POPE (1:1) liposomes Figure 4.4 106 Trypsin and a-chymotrypsin cleavage of detergent-solubilized and reconstituted Por protein. Figure 4.5 112 Recovery of protein and phospholipid following extrusion of Por proteoliposomes. Figure 4.6 117 Por protein incorporation into POPC/POPG liposomes Figure 4.7 119 Por protein incorporation into POPC/POPS liposomes Figure 4.8 122 Por protein incorporation into POPC/DODAC liposomes Figure 4.9 Cryo-electron micrographs of reconstituted Por proteoliposome formulations viewed under different magnifications Figure 5.1 In vitro Por antibody binding activity Figure 5.2 In vitro Por antibody binding activity with rabbit anti-sera Figure 5.3 Effect of charged liposomes on antibody binding activity of proteoliposome formulations Figure 5.4 Anti-Por IgG titers of mice immunized with Por protein preparations Figure 5.5 Effect of immunization route on IgG isotypes of anti-Por antibodies Figure 5.6 Photograph of skin cross-section of the intradermal injection site xiv L I S T O F T A B L E S Page Table 1.1 Outline of the regimen and mechanism of action of antibiotics against Neisseria gonorrhoeae 21 Table 1.2 Amino acid compositions of gonococcal proteins IA and EB isolated from bacterial strains FA19 and RIO, respectively 29 Table 3.1 QELS Size analysis of Por proteoliposomes reconstituted from OGP and cholate 84 Table 4.1 Trypsin and a-chymotrypsin cleavage of reconstituted Por-proteoliposomes: Densitometric analysis of SDS-PAGE gel 108 Table 4.2 Size reduction of reconstituted Por proteoliposomes by extrusion 110 Table 4.3: QELS size analysis and protein incorporation efficiency of Por proteoliposomes reconstituted with varying lipid composition 114 A B B R E V I A T I O N S Ab antibody ADCC antibody-dependent cytotoxic cells Ag antigen APCs antigen presenting cells BCA bicinchoninic acid BCR B cell receptor CD cluster designation CFA complete Freund's adjuvant CMC critical micelle concentration CMI cell-mediated immunity CRs complement receptors CS circumsporozoite CTL cytotoxic T lymphocyte Da dalton DGI disseminated gonococcal infection ELISA enzyme-linked immunosorbent assay FA Freund's adjuvant FATMLV freeze and thawed multilamellar vesicles Fc crystallizable fragment FCR crystallizable fragment receptor g force of gravity xvi GMP general manufacturing procedure HBS HEPES-buffered saline H B V hepatitis B virus HEPES N-[2-hydroxyethyl]piperazine-N'-[2-ethanesulfonic acid] HI humoral immunity HIV human immunodeficiency virus id. intradermal i.p. intraperitoneal IFNy interferon-gamma IgG immunoglobulin G IL interleukin kDa kilodalton LOS lipooligosaccharide LPS lipopolysaccharide L U V large unilamellar vesicle MAb monoclonal antibody M A C membrane attack complex mBc memory B cell M H C major histocompatibility complex M L V multilamellar vesicle M 0 macrophage MOMP meningococcal outer membrane protein mTc memory cytotoxic T cell xvii M W molecular weight OGP octyl-P-D-glucopyranoside P G peptidoglycan PI gonococcal protein I PID pelvic inflammatory disease PII gonococcal protein II PHI gonococcal protein III POPC l-Palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine POPE 1 -Palmitoyl, 2-oleoyl-sn-glycero -3-phosphoethanolamine POPG 1 -Palmitoyl, 2-oleoyl-sn-glycero -3 -phosphoglycerol POPS 1 -Palmitoyl, 2-oleoyl-sn-glycero -3-phosphoserine QELS quasi-elastic light scattering RER rough endoplasmic reticulum RES reticuloendothelial system SDS sodium dodecyl sulfate SUV small unilamellar vesicle TCR T cell receptor T H T helper cell TNF tumor necrosis factor ACKNOWLEDGEMENTS xvm Several colleagues and friends were instrumental in the preparation, development and eventual completion of this doctoral thesis. I would like to acknowledge my committee members Drs. Tom Madden, David Godin, Morley Sutter, and Lawrence Mayer for their contributions and assistance to my research project. The research supervisor is essential for providing guidance and support in a graduate student's research and education. I would like to thank Tom Madden for giving me the opportunity to work with him and providing invaluable supervision and input into my thesis and other written work. Laboratory work is a crucial part of science that often becomes stressful. As a supervisor, Tom has provided much needed reprieves from the long days in the lab as we have been rock climbing, skiing, and, on a number of occasions, dinner outings. I would also like to acknowledge my lab colleagues present and past: Miranda, Cliff, Jeff, Gitanjali, Ed, Cindy, Xue Min, and John for their contributions in and out of the laboratory. My thanks go to Dana Masin for her assistance in the animal immunology studies. I would also like to acknowledge Drs. Milan Blake and Katarina Edwards for their assistance in this research. This research was supported by grants from National Institute of Health and scholarships from the Science Council of British Columbia. D E D I C A T I O N TO MY WIFE AND FAMILY FOR THEIR CONSTANT SUPPORT AND MOTIVATION 1 CHAPTER 1 INTRODUCTION The following chapter provides an introduction to vaccines and the host immune response. In addition, the characteristics of the gonococcus, mechanisms by which it avoids host defenses to cause infection and disease, current therapy of infection, as well as the rationale for using liposomes as a carrier for a gonococcal subunit vaccine are discussed. 1.1 Vaccines and Immunology The primary aim of vaccination is to generate an adaptive immune response that is highly specific for a particular pathogen and provides a memory of the pathogen such that subsequent challenge by the organism will be met with an enhanced immune response leading to inactivation and elimination of the pathogen. Vaccines have proven to be effective weapons for disease prevention. Vaccination has led to substantially lower incidences of disease, particularly diseases such as measles, mumps, rubella, pertussis, tetanus and poliomyelitis. The benefits of immunization can be illustrated by the virtual eradication of smallpox. However, there are a number of diseases for which vaccines are not readily available. Diseases such as malaria, cholera, human immunodeficiency virus (HIV) and toxin-producing E. coli kill millions of people annually. Obstacles that have become evident and have slowed the development of vaccines against these diseases are antigenic variation, hypersensitivity to the antigen, adverse reactions due to contamination with endotoxins and reversion of the attenuated organism to the wild type or pathogenic form. 2 1.1.1 Vaccine strategies Immunization has involved the administration of live, attenuated organisms (polio, measles and mumps) or killed whole organism vaccines (pertussis, influenza and typhoid). Microorganisms can be attenuated by mutation and selection of an avirulent strain which is then grown in culture. Mutants are constantly monitored for absence of virulence and antigen retention. The problems with attenuated vaccines are antigen instability and the danger of reversion to virulent forms, which can then cause disease (Ross, 1998). Researchers have indicated that genetic mutations contribute to the reversion to neurovirulence of attenuated vaccines, as observed in cases of vaccine-associated paralytic poliomyelitis (Nkowane et al, 1987 and L i et al, 1996). In contrast, whole, killed-organism vaccines are prepared by chemical inactivation or irradiation with y-rays. Advantages of using inactivated vaccines are that they are more stable and will not revert to virulent forms as with attenuated preparations. However, killed vaccine formulations have been associated with safety problems. For instance, endotoxin contamination in pertussis vaccination has been associated with severe febrile convulsions and encephalitis in infants leading to, in severe cases, brain damage and even death (Miller et al, 1981 and Cherry, 1992). Alternative vaccine strategies using specific antigens from pathogens have been examined. Use of purified antigens can decrease the risk of adverse reactions associated with attenuated or killed preparations. Capsular polysaccharides, exotoxins, recombinant antigens, D N A and synthetic peptides have been investigated as potentiafantigen targets for vaccines (Kuby, 1997). Polysaccharide vaccines against Neisseria meningitidis groups A and C, but not B, have proven effective for eliciting humoral immunity through thymus-3 independent B cell activation (Buchanan et al. 1998). However, polysaccharides do not bind to major histocompatibility complex (MHC) molecules and thus are unable to activate T helper cells or trigger the development of memory cells. Conjugation of polysaccharide to a protein carrier permits binding to the MHC molecules leading to T H cell activation and induction of memory B cells in response to the polysaccharide antigen. However, this type of vaccine lacks the ability to induce memory T cells against the pathogen. Other vaccine preparations, such as diphtheria or tetanus toxoid vaccines, are effective in inducing anti-toxoid antibodies that bind and neutralize bacterial toxin (Kuby, 1997). One of the concerns of using exotoxins in a vaccine relates to the need to ensure that there is complete detoxification without excessive modification of the epitope. Recombinant protein or synthetic peptide vaccines have been developed, in particular, against the hepatitis B virus. Although recombinant vaccines are capable of inducing humoral immunity (HI), they are taken up by endocytosis and processed as exogenous antigens and are therefore unable to activate major histocompatibility complex (MHC) class I-restricted T cells that are required for induction of cell-mediated immunity (CMI). Synthetic peptides are often poorly immunogenic or they tend to preferentially elicit HI over CMI. In order to improve the immune response to weak antigens, peptides and proteins are often combined with adjuvants, as described in a later section. A host's ability to fight an infection is influenced by the type of immune response, HI or CMI. The humoral response plays an important role in combating extracellular pathogens, whereas the cell-mediated immunity is involved in fighting intracellular infections. The development and characterization of a vaccine should take into account the type of immune response that is being generated and whether or not this response will be able to confer 4 protection against the specific pathogen. The components and mechanisms of humoral and cell-mediated immunity are described below. 1.1.2 Humoral Immunity Humoral immunity is that arm of the immune system that is mediated by immunoglobulins which are a group of glycoproteins present in serum and tissue fluids. Immunoglobulins are also commonly known as antibodies and can either be free in blood and lymph or can be surface-attached on B cells acting as receptors for antigens. Antigen recognition by B cells stimulates their maturation, proliferation and differentiation into antibody-forming cells (AFCs) or plasma cells which secrete large amounts of antibody against the particular antigen (Figure 1.1, adapted from Roitt et al, 1996). In addition, a certain fraction of the B cell population develops into longer-lived memory B cells that retain the antigen-binding specificity. Antibody molecules consist of four polypeptide chains, two identical light chains and two heavy chains linked together by disulphide bonds. The amino terminal ends of the heavy chains constitute the ends of the Fab portion of the antibody molecule and contain the site for antigen binding. The C-terminal ends or Fc portion of the heavy chains function as the ligand for binding to surface receptors (FCR) on phagocytic cells, B cells, and antibody-dependent cytotoxic (K) cells (ADCC). Antibody Fab portion recognition and binding of surface-exposed antigen on bacteria leads to opsonization in which Fc receptors on the surface of macrophages (M0) or ADDCs bind to Fc portion of the antibody attached to bacterial surface antigen resulting in bacterial cell phagocytosis and lysis. Macrophages can act as antigen-presenting cells (APCs) by ingesting a pathogen or foreign particles and processing and presenting antigen in association with cell surface 5 Figure 1.1: The humoral immune response. Antigen binds to B cells that have the appropriate surface B cell receptor (BCR). Cells are stimulated leading to clonal selection, cell proliferation and maturation into antibody-forming cells (AFCs) and memory B cells (mB). Antibody secreted by the AFCs and the memory cells have the same antigen-binding specificity. Secreted antibodies then bind to antigen resulting in opsonization of antigen or microbes expressing antigen on the cell surface. Antibody-antigen complex binds to Fc receptor (FCR) on the surface of the macrophage (Mphg) and triggers cell-mediated immunity, as well as phagocytosis of the antigen. Antigen-microbe complexes bind to antibody-dependent cytotoxic cells (ADCC) via the FCR, resulting in microbe lysis. 6 V • • Antigen f^BCR f Bcellj Clonal Selection Proliferation +Maturation AFc) ( A F C ) (AFf?) Y y y + V • • Antigen Opsonization V Y Mfcr (Mphg) Cell-mediated Immunity Phagocytosis 7 molecules called major histocompatibility complexes (MHCs). Antigen presentation on M H C molecules can trigger recognition by B cells leading to maturation, proliferation and differentiation into AFCs and memory B cells. In addition to phagocytosis and lysis, antibody-antigen complexes can trigger complement activation and serum killing. The complement system is a network of proteins or zymogens that require proteolytic cleavage in order to become active. The role of complement is to provide defense against bacteria by mediating lymphocyte activation, as well as triggering opsonization and lysis of target cells. Complexes formed by antibody binding to bacterial surface antigens trigger activation of this system through the classical or alternative pathways in which serum C4 is cleaved to C4b which becomes surface bound by binding to protein or carbohydrates on the bacterium. Surface-bound C4b then binds C2 which activates a cascade of reactions leading to cleavage of C3 to C3b which deposits on the bacterial cell surface (Figure 1.2, adapted from Roitt et al, 1996). Covalently attached complement proteins C3b and C4b act as opsonins enhancing phagocytosis by functioning as ligands for complement receptors on phagocytic cells. Furthermore, fixed C3b can bind serum C5 which is converted to a convertase to generate C5b which subsequently binds C6, C7, C8, C9 to form C5b-9, the membrane attack complex. Complement-mediated killing occurs in several ways. Complement C3b-coated target cell can bind to complement receptors (CRs) on phagocytic cells triggering endocytosis or phagocytosis and cell death. Alternatively, binding of effector cells to complement fragments on the target can trigger activation and chemotaxis of leukocytes capable of producing bacterial cell death. In addition, the membrane attack complex (MAC), C5b-9, forms a hydrophobic "plug" or pore-forming molecule. This plug inserts into the target cell membrane causing osmotic disruption and cell lysis. 8 Figure 1.2: The pathways of complement activation. Activation of the complement system, which is a part of the innate immune system functioning in opsonization and removal of foreign pathogens from the body, is illustrated. Classical pathway (Panel A) links the adaptive immune system, antibody, to the innate immune system, complement. Antibody-microbe complexes bind Clq and trigger a series of reactions to generate the enzyme C4b2a that cleaves C3 to C3b and deposits on the cell surface and acts as an opsonin. The alternative pathway (Panel B) involves the surface-attached C3b, which binds factor B. Factor D then cleaves factor B to generate the enzyme complex C3bBb that combines with properdin (P) to enhance cleavage C3 to C3b The enzymes C4b2a and C3bBb cleave C5 to C5b leading to the deposition and binding of C6, C7, C8 and C9 to form the membrane attack complex (MAC), which is a pore-forming molecule that causes osmotic disruptions in the cell membrane resulting in cell death. 10 1.1.3 Cell-mediated immunity The second component of the immune system is cell-mediated immunity (CMI), which is mediated primarily by lymphocytes and phagocytes and where antibodies play a secondary role acting as links in some cell-mediated reactions. T cell-independent interactions of an organism and phagocytic cells can lead to uptake and killing of the organism. As mentioned above, deposition of complement on the organism enables binding of macrophages via cell-surface receptors and subsequent pathogen lysis. Alternatively, macrophages and other cells can release cytokines upon recognition of the organism, which leads to stimulation and activation of other leukocytes. T cell-dependent cell-mediated immune responses represent a major host defense mechanism. The uptake, processing and presentation of antigen in association with M H C molecules on the APC surface triggers T helper (T H) cell activation, in particular CD4 + - T H cells (Figure 1.3, adapted from Roitt et al, 1996). T helper cells modulate the various types of cellular cooperation by releasing specific combinations of cytokines. There are two major subsets of T H cells, TH1 and TH2, each having a specific cytokine profile (Mosmann and Coffman, 1989). The differentiation and selection of the effector mechanism is determined by the nature of the antigen presentation on the surface of the APC. TH1 type cells are generally stimulated by M H C class I-restricted antigen presentation. These CD4 + -T cells produce cytokines IL-2 and IFNy, which are effective stimulators of B cell activation and production of IgG2a, but not much IgGl . In addition, EL-2 stimulates cytotoxic CD8 + lymphocytes via the IL-2 receptor to recognize and kill infected host cells. In contrast, antigen recognition in association with M H C class II molecules stimulates CD4 + -T cells to produce cytokines IL-4 and IL-6 which are efficient helper cells for B cell activation and 11 Figure 1.3: The cell-mediated immune response. Antigen is recognized, taken up and processed by antigen-presenting cells (APCs), such as macrophages and dendritic cells. The antigen is then presented on the surface of the cell in association with major histocompatibility complex class I (MHCl) or class II (MHCII). C D 4 + - T helper (TH) recognize antigen via the T cell receptor (TCR) and stimulate the activation of T H cells. When the antigen is presented on M H C l molecules, TH1 cells are activated releasing cytokines, such as tumor necrosis factor (TNF), interferon gamma (fFN) and interleukins 2 (IL-2) and 12 (EL-12). These cytokines stimulate macrophages and natural killer (NK) cells, which mediate phagocytosis and lysis of the pathogen. Cytokines also stimulate C D 8 + cytotoxic T (Tc) cells, which recognize and kill infected host cells. Activation of Tc also generates memory cytotoxic T (mTc) cells that retain long-lived memory of the specific antigen. Antigen presentation on MHCII molecules triggers the activation of TH2 cells and release of cytokines, such as DL-2 and EL-4. These cytokines trigger B cell activation and the release of EL-4, EL-6 and IL-10, which stimulate antibody-forming cells (AFC) to secrete antibodies. 12 • • Antigen Antigen uptake & processing \y* • l Y A P C + TffrH) CD4+ MHciV_yMHcn ^ v _ y Antigen Recognition & Activation of T H cells M H C l Recogniton MHCII Recognition IL-2R mTc IL-2R Tc CD8+ T N F . I F N Y IL -2 , IL-12 (Mphg) ( N K ^ ) Recognition of Phagocytosis of Infected Host Cells Pathogen CD8+ IL -2 , I L - 4 I L - 1 , I L - 6 . I L 10 V ( A F C ) ( A F C 13 proliferation for antibody production, especial ly I g G l and IgE . C h o w and coworkers (1998) demonstrated that co - immuniza t ion o f hepatitis B vi rus and T H 1 type cytokines, EL-12 or I F N y , stimulated T H 1 cel l development w i t h a concomitant increase i n I g G 2 a product ion. In contrast, antigen and EL-4 co-inject ion induced elevated I g G l levels w i t h enhanced T H 2 cel l development, but suppressed T H 1 differentiation and I g G 2 a product ion. E leva t ion o f T H 1 cells and cytokines have been shown to increase the C D 8 + T cel l cy to ly t ic act ivi ty towards M. tuberculosis-infected macrophage target cells (Skinner et al., 1997). Exogenous antigen is taken up into endosomes or lysosomes and degraded by acid hydrolys is to peptides. M H C class II molecules are heterodimers o f heavy (a) and light (P) glycoprote in chains that are synthesized i n the rough endoplasmic re t iculum ( R E R ) complexed to a polypeptide chain ca l led the invariant chain (Ii) and then transported through the G o l g i complex to the acidic endosomal or lysosomal compartment (Jensen, 1997). The invariant chain is bound to the M H C class II b ind ing groove w h i c h is comprised o f a P sheet supporting two a hel ical domains. Here , the I i becomes dissociated and the M H C class II molecule binds peptide. Crys ta l lography studies have shown that the ends o f the b ind ing groove are open and opt imal ly b ind peptides o f 12-20 amino ac id residues w i t h at least three variable anchoring residues (Stern and W i l e y , 1994 and M a l c h e r e k et al, 1995). Peptides w i t h fewer than 13 amino acid residues generally have less than m a x i m a l b ind ing activities. The endosome is then transported to the cel l surface where the M H C class II- peptide complex is exposed to the external environment. T helper cells recognize and b ind to the p e p t i d e - M H C class II complex v i a the T cel l receptor ( T C R ) and trigger a T H 2 - t y p e response. T H 2 cells release cytokines (EL-4, I L - 6 , EL-10) w h i c h stimulate B cel l proliferation, maturation and differentiation into A F C s and memory cells . 14 Endogenous or cytoplasmic antigen, in A P C , is processed by proteases and then is transported by a transmembrane transporter to the rough endoplasmic re t iculum ( R E R ) . In the R E R , proteosome complexes are processed and bound to M H C class I molecules. In contrast to the class II b ind ing groove, the ends o f the M H C class I b ind ing cleft are closed (Rammensee et al, 1993 and Stern and W i l e y , 1994). Moreove r , endogenously bound peptides are restricted i n length to eight or nine amino acids and are buried deep i n the groove by anchoring residues at the amino and ca rboxy l te rmini o f the peptide through hydrogen bonding and V a n der Waa l s forces (Matsumara et al, 1992, Y o u n g et al, 1995 and Fremont et al., 1995). The ternary complex is transported to the cel l surface v i a the G o l g i complex and exposed for T cel l recogni t ion. M H C molecules that do not have peptide bound to their clefts are unstable and prompt ly dissociate and are degraded intracel lularly. A n t i g e n presentation on M H C class I molecules triggers T H 1 type ce l l act ivation. T helper cells b ind v i a the T C R and release cytokines that stimulate T cells, B cells and monocytes. In particular, T H 1 cytokines, I L - 2 and I F N - y activate cytotoxic T cells (Tc) , as w e l l as natural k i l l e r ( N K ) cells and macrophages. Induct ion o f T c cells, N K cel ls and macrophages directs cel l-mediated cytotoxic i ty and hence the cel l -mediated immune response. Cel l -media ted cytotoxic i ty produces target ce l l lys is by several different mechanisms. Cy to tox ic cel ls can degranulate and release a monomer ic pore-forming protein cal led perforin that binds to the pathogen ce l l membrane i n the presence o f c a l c ium ions. M o n o m e r s po lymer ize to form transmembrane channels that cause the target ce l l to become leaky. Subsequently, the T c releases degradative enzymes that penetrate through the polyperfor in channels to produce cel l death. In addit ion, the release o f tumor necrosis factor ( T N F a or P) and I F N y can b ind to surface receptors on the target cel l and cause ce l l death. 15 The exact mechanism is not known , but the l iga t ion o f the agents at the ce l l surface may induce alterations i n the internal metabol ic pathways w i t h i n the organism to cause death o f the cel l . 16 1.2 NEISSERIA GONORRHOEAE Gonorrhea is a sexually transmitted disease caused by a bacterium, Neisseria gonorrhoeae. In 1995, nearly 400,000 cases o f gonorrhea i n the U n i t e d States were reported to the Centers for Disease C o n t r o l ( N I A T D , 1998), whereas approximately 5,500 cases were reported i n Canada ( L C D C , 1998). Howeve r , it is estimated that 800,000 cases o f gonorrhea occur annually i n the Un i t ed States, w i th an annual cost estimated at close to $1.1 b i l l i o n for the treatment o f gonorrhea and its compl ica t ions . D u r i n g 1995, it is estimated that there were over 60 m i l l i o n new cases o f gonorrhea among adults w o r l d w i d e ( W H O , 1995). The prevalence o f this disease, medica l costs and the dramatic rise o f antibiotic-resistant strains o f gonococcus underscore the need for a means o f preventing and cont ro l l ing gonorrhea. Therefore, the development o f an effective gonococca l vaccine is an important research objective for many research groups around the wor ld . The characteristics o f the gonococca l organism, current methods o f treatment and potential vaccine targets are described below. 1.2.1 Characteristics o f the gonococca l organism Neisseria gonorrhoeae belongs to the fami ly o f organisms ca l led Neisseriaceae w h i c h are non-flagellated, spherical or rod-shaped organisms c o m m o n l y occur r ing in pairs or as short-chained groups. G o n o c o c c i are gram negative bacteria that colonize mucosa l sites. The structural characteristics o f the gonococcus have been described by several researchers (Maeland , 1977, M o r s e and B r o o k s , 1985 B r o o k s , 1985a). A s shown i n F igure 1.4, the gonococca l ce l l envelope consists o f three major layers. The cytoplasmic membrane is the innermost layer o f p r imar i ly l ip ids and proteins surrounding the cytoplasm and the cel lular machinery. The middle layer consists o f the pept idoglycan, w h i c h is constructed by a 17 Figure 1.4: The surface components of the gonococcal outer membrane. The outer membrane structures that are key factors i n gonococca l pathogenesis are shown. The abbreviations are. P I , protein I; PI I , protein II; P H I , protein III; L P S , l ipopolysacchar ide; l i p id , fo rming the phosphol ip id bi layer o f the outer membrane; pi lus , po lymer o f p i l i n proteins forming hair - l ike extensions that are considerably longer than illustrated; peptidoglycan, a structure o f sugars and amino acids. A r r o w s indicate the proposed connections between inner and outer membrane proteins to the pept idoglycan (Adapted f rom B r i t i g a n e f a/., 1985 and B r o o k s , 1985a). 18 19 l inkage o f a series o f sugars and amino acids. The pept idoglycan ( P G ) forms a r ig id structure and maintains the structural integrity o f the ce l l . The outermost layer o f the ce l l envelope is a complex outer membrane. The structures w i t h i n this outer membrane are considered to be the key virulence factors in the pathogenesis o f disease. The interactions o f these complexes w i t h host mucosal surfaces and immune system mediate the development o f disease. The internal side o f the outer membrane is thought to be connected to the pept idoglycan v i a membrane proteins. The midd le layer is composed o f phosphol ip ids and hydrophobic components such as the hydrophobic membrane-bound domains o f proteins that assist in the attachment to the pept idoglycan layer. The external surface o f the gonococcus presents the hydrophi l ic regions o f the membrane-bound protein and other structures such as p i l i and l ipopolysaccharide. These surface-exposed structures interact w i t h the external environment and function in faci l i tat ing and mediat ing gonococca l infection. 1.2.2 Characteristics o f gonococca l infect ion H u m a n s are the natural hosts for Neisseria gonorrhoeae where gonococca l infection results i n a w ide spectrum o f sequelae ( B r o o k s and Donegan , 1985 and B r i t i g a n et al., 1985). A n i m a l models for gonococca l infect ion have been found to be difficult to maintain, as animal hosts tend to have natural resistance to infect ion (Brooks , 1985b). In addit ion, infect ion in animals does not m i m i c the infect ion in humans. Urethri t is is a c o m m o n symptom o f gonococca l infect ion in men and is characterized by m i l d dysuria and urethral discharge after a two to seven day incubat ion per iod. In some men, the disease can develop to epididymit is , painful in f lammat ion o f the ep id idymis . Infections in females often lead to more serious condit ions. Ini t ial ly, gonococca l infect ion is pr imar i ly , a l though not 20 exclus ively , manifested in the endocervix and is associated w i t h unusual or increased vagina l discharge. Some indiv iduals experience dysuria , increased frequency o f urinat ion, pe lv ic pain and menstrual abnormalit ies. In some instances, w o m e n can develop salpingitis , inf lammat ion o f the fa l lopian tubes, w h i c h is characterized by rapid onset o f pe lv ic , lower abdominal and, in some cases, back and leg pain (Brooks , 1985c). Compl ica t ed salpingitis can lead to ectopic pregnancy and infert i l i ty. In addit ion, spread o f infection f rom the fa l lopian tubes or by lymphat ic drainage to the l iver can cause l iver dysfunct ion or perihepatitis. Other mucosal infections that often occur are rectal and pharyngeal infection, p r imar i ly due to rectal-genital and oral-genital contact, respectively. Compl ica t ed gonococca l infections are referred to as disseminated gonococca l infections ( D G I ) and are characterized by fever and sk in lesions in the early stages (Brooks , 1985d). Les ions usual ly occur on distal parts o f the arms and legs such as joints o f toes and fingers, inc lud ing the general area o f the hands and feet. A s the condi t ion progresses, septic arthritis develops, mark ing the onset o f the secondary stage o f D G I . In severe disease situations, D G I is manifested as endocarditis or meningit is . 1.2.3 Treatment o f gonococca l infect ion Ant ib io t i c s are the current therapy for combat ing gonococca l infections. P e n i c i l l i n was w i d e l y prescribed for the treatment o f gonorrhea and remains the standard drug for treatment i n many developing nations. H o w e v e r , due to the emergence o f penic i l l inase-producing strains o f Neisseria gonorrhoeae and the development o f newer more potent antibiotics, change in the therapy regimen has become inevitable. The current first choice o f ant imicrobia l therapy is a combina t ion o f a m o x i c i l l i n and c lavulan ic ac id (Table 1.1, adapted 21 Regimen Antibiotic Class Antimicrobial Agents Mechanism of Action Firs t choice Pen ic i l l i n s + (3 lactamase inhib i tor A m o x i c i l l i n + C l a v u l a n i c A c i d Inhibi t ion o f pept idoglycan synthesis Cephalosporins Ceftr iaxone, cefuroxime, cefotaxime Inhib i t ion o f pept idoglycan synthesis Second choice F luoroquinolones C ip ro f loxac in , enoxacin, perf loxacin , nor f loxac in Inhibi t ion o f D N A synthesis Table 1.1: Out l ine o f the regimen and mechanism o f act ion o f antibiotics against Neisseria gonorrhoeae. 22 f rom R a n g et al, 1995). A m o x i c i l l i n , a P-lactam antibiotic, s imi lar to pen ic i l l in , interferes w i t h the formation o f the pept idoglycan layer o f the gonococca l ce l l envelope. It acts by inh ib i t ing the transpeptidation enzyme responsible for c ross - l ink ing o f peptide chains dur ing pept idoglycan synthesis (Arms t rong et al, 1987). C l a v u l a n i c ac id is a P-lactamase inhibi tor that protects the antibiotic from enzymat ic degradation. F o r the treatment o f P-lactamase-producing bacteria, spec t inomycin was a c o m m o n replacement for pen ic i l l in ; however, there is a propensity to develop resistance to spec t inomycin (Br i t igan et al, 1985). Ano the r group o f first-line ant imicrobia l agents are the cephalosporins, ceftriaxone being the prototypical agent. A s w i t h the penic i l l ins , cephalosporins interfere w i t h pept idoglycan synthesis. Increased resistance to these agents has developed due to plasmid-mediated and chromosomal p-lactamases that have greater hydro ly t i c act ivi ty towards cephalosporin. D u e to mul t i -drug resistance, newer agents have been introduced for the treatment o f gonococca l infection. C ip ro f loxac in is a broad-spectrum antibiot ic be long ing to a group o f synthetic compounds cal led the f luoroquinolones. These ant imicrobials exhibi t excellent act ivi ty against many organisms resistant to pen ic i l l in , cephalosporin and aminoglycosides . The therapeutic act ion o f these agents occurs during D N A replicat ion. In order for D N A synthesis to proceed, D N A gyrase (topoisomerase II) must in i t i a l ly u n w i n d the posi t ive supercoil o f D N A and generate a negative supercoil . The introduct ion o f a negative supercoil in D N A permits transcription or replicat ion. D r u g s such as c ip ro f loxac in inhibi t D N A synthesis by b ind ing and b l o c k i n g D N A gyrase function. A l t h o u g h ant imicrobia l therapy is an effective method o f eradicating gonococca l infections, the emergence o f greater numbers o f antibiotic-resistant strains o f Neisseria gonorrhoeae requires the continued development o f newer, more expensive agents to 23 ci rcumvent the resistance mechanisms. Therefore, the concept o f prevention v i a vaccina t ion is rece iv ing considerable attention as an addi t ional method o f combat ing gonococca l infection, especial ly in deve lop ing countries where antibiotic-resistant bacteria are prevalent. 1.2.4 Potential vaccine target antigens F o r gonococcal infect ion to occur, attachment to host mucosal cells is essential. The role o f var ious gonococca l surface structures in the attachment process has been reviewed by several researchers (Br i t igan et al, 1985 and B l a k e and Got sch l i ch , 1987). P i l i are ha i r - l ike extensions radiating f rom the surface o f the gonococca l ce l l . E a c h p i lus is composed o f a series o f protein subunits, p i l i n proteins, that range i n molecular weights o f 18,000-20,000 daltons. The size and antigenic characteristics vary depending o n the particular strain. Adherence o f the gonococcus is mediated p r imar i ly by the p i l i , as pi l ia ted organisms exhibi t greater adherence to human eukaryotic cel ls than non-pi l ia ted gonococc i . The pi lus can also mediate the spread o f infection f rom the endocervix to the fa l lopian tubes by faci l i ta t ing attachment o f bacterial cel ls to sperm and enabl ing the penetration o f the cervical mucous barrier and migrat ion to the fa l lopian tissue. In addit ion, p i l i can inhibi t neutrophil phagocytosis o f bacterial cells . The importance o f the pi lus i n the pathogenesis o f infect ion has made p i l i n peptides targets for potential vacc ine candidates. H o w e v e r , gonococca l p i l i have been observed to have a h igh degree o f antigenic heterogeneity (Buchanan, 1975). Th is has been demonstrated by marked p i l i n var iab i l i ty amongst strains, as w e l l as w i t h i n an ind iv idua l gonococca l strain (Swanson et al., 1987). P i l i n var iabi l i ty , therefore, has lead to the targeting o f other surface structures as potential vacc ine antigens. 24 Prote in II is an outer membrane protein w i t h a molecular weight ranging between 24,000-30,000 daltons, depending on the condit ions used i n pur i f ica t ion and solubi l iza t ion . S imi la r to p i l i , protein II is thought to play an important role i n the adherence to certain mucosal , as w e l l as phagocyt ic cells such as neutrophils. In particular, protein II is thought to be invo lved in the c lumping o f organisms such that the adherence o f a single gonococcus to the mucosal surface may a l l o w the attachment o f many more gonococc i . An t igen i c var ia t ion in protein II is quite prevalent and enables gonococc i to survive at various different sites o f infection in male and female hosts. Studies have shown that a single gonococca l strain can express either no, one or several types o f protein II, each having a different molecular weight and be ing ant igenical ly distinct ( B l a c k et al, 1984). The antigenic diversity o f protein II species has been demonstrated by the reaction o f different protein II species f rom the same strain w i t h a specific protein II serum ( D i a z and Hecke l s , 1982). The results were indicat ive o f very little cross-reactivity between different protein II molecules and a h igh degree o f antigenic var iab i l i ty i n the surface-exposed protein structure (Heckels , 1981). A s indicated, phase variations enable the organism to escape host immune responses and adapt efficiently, enabl ing them to co lonize particular microenvironments . Therefore, protein II has proven to be less than ideal as an antigen for a gonococca l vaccine. Another outer membrane constituent that has been considered as a vaccine antigen is protein III. Th is protein ranges in molecula r weight f rom 30,000-31,000 daltons, depending on the reducing condit ions and it is a h igh ly ant igenical ly conserved protein c o m m o n to al l strains o f gonococc i (B lake and Go t sch l i ch , 1983 and B l a k e et al, 1988). It is located in close p rox imi ty to the major gonococca l outer membrane protein, protein I ( M c D a d e and Johnston, 1980). Protein III is i nvo lved in media t ing serum resistance by p rov id ing a b ind ing 25 site for b l o c k i n g antibodies. Studies have shown that antibodies to protein III can b lock the bactericidal act ivi ty o f human sera (R ice et al, 1994, R i c e et al, 1985, R i c e et al, 1986 and V i r j i and Hecke l s , 1988). The effect may be exerted by compet i t ion between b l o c k i n g antibody and bactericidal antibody for complement act ivat ion and deposi t ion o n the ce l l surface (Joiner et al, 1985a). Joiner and coworkers (1985b) showed that complement C 3 deposit ion was enhanced six- to nine-fold by b l o c k i n g antibody and that b l o c k i n g ant ibody inhibi ted k i l l i n g i n a dose-dependent manner. These observations suggest that complement depletion may be a mechanism o f serum resistance. Further studies showed that enhanced complement activation and deposi t ion by the b l o c k i n g antibody lead to the formation o f a non-bactericidal C 5 b - 9 complex that was i n a different molecular configurat ion than the bactericidal C 5 b - 9 required for serum lysis and k i l l i n g o f bacteria (Joiner, 1983, Joiner, 1985c). The depletion o f complement factors may reduce the number o f effective membrane attack complexes and hence impai r the host 's abi l i ty to fight infection. D u e to the antigenic var ia t ion o f p i l i , protein II and serum resistance mediated by b l o c k i n g antibodies, the gonococca l organism has developed several mechanisms to evade and thwart the immune system and ensure its survival and facilitate the pathogenesis o f infection. Research has therefore focussed on protein I, a major gonococca l membrane component as a potential antigen target. 26 1.3 G O N O C O C C A L P R O T E I N I 1.3.1 Characteristics o f gonococca l protein I The characteristics o f gonococca l protein I have been rev iewed by G o t s c h l i c h and coworkers (1988). Prote in I is the major gonococca l membrane protein (60%), having a molecular weight in the range o f 32-40 k D a (Johnston and Gotsch l i ch , 1974). It is an antigenically conserved membrane constituent that serves as the basis for classif icat ion o f a large number o f gonococca l strains (Johnston et al, 1976, Buchanan and Hi ldebrandt , 1981 and B l a k e and Gotsch l i ch , 1983). Studies have shown that protein I is a channel fo rming outer membrane protein termed a por in (Douglas et al, 1981). Isolation, pur i f icat ion and protease digestion studies have indicated the existence o f t w o types o f protein I that adopt different conformations w i t h i n the gonococca l outer membrane. Prote in I A has been observed to be inserted completely into the membrane w i t h on ly a smal l fragment exposed on the surface (Barrera and Swanson, 1984). Prote in I B is oriented wi th both ends inserted into the bi layer w i t h the loop port ion extending out from the per ip lasmic surface (B lake et al, 1981, B l a k e and Gotsch l i ch , 1982 and B l a k e and Got sch l i ch , 1983). A mode l o f the proposed protein I structures is i l lustrated i n F igure 1.5 (adapted from B l a k e et al, 1981 and Barrera and Swanson, 1984). The general amino acid compos i t ion determined for proteins I A and I B f rom two gonococcal strains is shown i n Table 1.2 (adapted from B l a k e and Gotsch l ich , 1982). 27 Figure 1.5: Models of proteins IA and IB orientations in the gonococcal outer membrane. Protein I A is almost completely bur ied w i t h i n the membrane w i t h only a smal l terminal , exposed portion. Treatment o f membrane-bound P I A w i t h chymot ryps in or t rypsin does not produce any cleavage fragments. Howeve r , treatment w i th proteinase K ( P K ) yields t w o fragments, a smal l fragment (1.5 k D a ) and a membrane-associated fragment (33.5 kDa) . In contrast, protein I B is situated w i t h its terminal portions bur ied i n the outer membrane w i t h an external loop region exposed on the gonococca l surface. C h y m o t r y p s i n treatment generates two membrane-associated fragments o f 22 ( C l ) and 14 k D a (C2) , whereas t rypsin, in i t ia l ly , cleaves P I B into 28 (T2) and 10 k D a ( T l ) fragments. P ro longed trypsin treatment cleaves the 28 k D a fragment to a membrane-bound 21 k D a fragment and a smaller, soluble fragment. In contrast to P I A , P I B is in i t i a l ly c leaved by proteinase K into fragments resembling those generated by chymotryps in . P ro longed digest ion cleaves P K 2 fragment into a smaller soluble fragment P K 3 . (Adapted from B l a k e et al., 1981 and Bar re ra and Swanson, 1984). 29 A m i n o A c i d Content (moi o ' r e s idue /mol o f protein) A m i n o A c i d Strain F A 1 9 - P r o t e i n I A Strain R I O - P r o t e i n I B Neutra l 198 (0.64)* 223 (0.68) A l i p h a t i c 155 (0.50) 179 (0.54) G l y c i n e 37 43 A l a n i n e 27 31 V a l i n e 28 29 Leuc ine 16 19 Isoleucine 6 8 Serine 23 31 Threonine 18 18 A r o m a t i c 38 (0.12) 3 7 ( 0 . 1 1 ) Phenylalanine 15 15 Tyros ine 18 18 Tryptophan 5 4 Sulphur-containing 0 2 (0.01) Meth ion ine 0 2 Cyste ine 0 0 Imino acids/proline 5 (0.02) 5 (0.02) D i c a r b o x y l i c amino acids 61 (0.20) 67 (0.20) Aspartic/asparagine 33 34 Glutamic /g lu tamine 28 33 B a s i c amino acids 48 (0.16) 3 9 ( 0 . 1 2 ) His t id ine 12 8 A r g i n i n e 16 11 L y s i n e 20 20 Tota l 307 329 Table 1.2: A m i n o ac id composi t ions o f gonococca l proteins I A and EB isolated f rom bacterial strains F A 1 9 and R 1 0 , respectively. The amino acid content was determined from the specific D N A sequence o f each protein I. *Numbers in parentheses represent the fraction o f the total number o f residues represented i n each class. 30 1.3.2 Structure o f protein I A s mentioned, gonococca l protein I is d iv ided into t w o immunochemica l classes, P I A and P I B , based on the reactions w i t h specific monoc lona l antibodies (Knapp et al, 1984). Protein I A is usual ly associated w i t h strains capable o f invading the human b l o o d stream and causing systemic gonococca l infections or disseminated disease. Strains bearing protein I B are typ ica l ly responsible for urogenital disease or pe lv ic inf lammatory disease (P ID) . Prote in I B has a molecular weight o f 34-38 k D a and w h e n membrane-bound is susceptible to t rypsin and chymotryps in cleavage, whereas protein I A has a lower molecular weight and is generally resistant to cleavage by these specif ic proteases ( B l a k e et al., 1981). Carbonett i and Spar l ing (1987) c loned the P I A gene and determined the D N A sequence encoding protein I o f one gonococca l strain. The predicted N- t e rmina l amino ac id sequence o f protein I A o f strain F A 1 9 (Figure 1.6) matched that o f a k n o w n protein I A (strain 120176-2) and was very s imi lar to that o f the protein I B o f strain RIO (F igure 1.7) (B lake and Go t sch l i ch , 1982, B l a k e , 1985 and Go t sch l i ch et al, 1987). The predicted amino acid sequence ( 3 0 7 A A ) o f protein I A F A 1 9 predicts a molecular weight o f 33,786 daltons, w h i c h c losely approximates the apparent molecular weight o f 34,000 (Carbonett i and Sparl ing, 1987). S imi la r ly , the calculated molecular weight based on the amino ac id sequence ( 3 2 9 A A ) o f protein I B o f strain RIO agrees w i t h that o f the k n o w n value o f 36 k D a as determined by S D S - P A G E (Gotsch l ich et al, 1987). Compar i sons o f the gonococca l proteins I A and I B pr imary structures to that o f the major porins o f Escherichia coli ( O m p F / O m p C ) showed a h igh degree o f sequence homology, indicat ing the poss ib i l i ty o f a c o m m o n ancestral o r ig in (Carbonett i and Sparl ing, 1987 and G o t s c h l i c h et al, 1987). Moreove r , the D N A sequences o f both P I A and PUB genes predicted the same typ ica l 19 amino ac id signal peptide as shown 31 - o -a < < CJ o < < o o < < < < < o o p CJ E-S-H < < CJ .2 < se o - E-< o < l a o S3 P «>8 •5 o - H E-H < < 3 < o o < > o CJ o ° CJ « cj < o 5,2 J < o ^5 y E- < CJ o o p 5 'CJ o o o o < < H CJ O < < < CJ o < E— o o CJ o o CJ E-CJ CJ CJ CJ o < - o o != CJ E— t— O o H CJ o o < o o <i < o E-I O I " a* Q -J CJ CJ •3 y .2 o < o . CJ o o 3 CJ u H J o ^ y < o < o o _ CJ £• o o o J . P OH f -o o o CJ JJ E-'— < » < J < v. O o CJ . C O O O o E— 2 < CJ o CJ < 5 <-> < o u CJ £.2 > b < o < o £ CJ < O r- O a s JS CJ < o . < - C o O o H cfl X CJ o < CJ o < o » 2 J= p . o JC'O O O I 8 Jp Cu E-3 < — < O o o CJ - 8 t . E— 5 CJ < O y E- E-oo CJ i> C5 < o _ < to E-1 > O . CJ o CJ _ H >S - o £>8 < CJ 5 o |S a. Q < O CJ « ^ H = 5 < < CJ u O 3 < S H J E-O Q 3 O . CJ o VO o ^> 0 00 _ o > o = 2' ^ CJ J= CJ H < u O # E- 5 8 cu E— cfl < < O ~3> CJ ^ CJ o o _ E-ra E— > O M < ^ y < o ^3 O O „ o 12 _ CJ £ • < U 0 -J < ^ O O o c < 55 tTT < t§ c_ H 0» CJ M H c y < < {-"ra CJ > O _ CJ < < -e H cu E— CD H J p X 0 ^8 18 u. O ca < £ c j < O a. ^  3U (J _ H ra H > O < 0 t CJ E-T< H E-^ O 55 c CJ 03 < < < i- H Cfl o -1" CJ 00CJ < CJ O o » 3 >, < -J < u. CJ fe o <M < •a 9 ° 5 cfl CJ 3 < H < J p - E— Cfl < < < ia O cn < J < cfl < E ^ < ra O ^8 o o <- ^ c CJ < O u. o c < 5< g I? ^ < ^ 2 J2 < < o cfl < - O c ° cfl < < E ^ ^ 0 CU CJ cu CJ sp ra " ^8 O > o - < . CJ >.< H .E-« CJ 38 >• R (/} C-H ra H •5 O o ra E-> o „ < % < O CJ H E-0 cs ON 0 0 u E-J= CJ >,CJ 58 12 -J < ^ O >> O O O 12 < < < < O 55 ^8 = 2 J3 P <. 0 „ 0 B-O E- H c_ E-H H ^ H - C O O O H ra CJ £ 0 <! 0 s8 wa f - i u O 0 0 H > O - < = 2 <u E— J E— u CJ « CJ ^8 5 ° u H u CJ M H X Q ra CJ 4 CJ < 0 s>8 < CJ .-^ E- H cfl < ^2 u t -u CJ V- CJ " CJ u_ ra u_ > b eu O E-Arg CGC Cfl < < 0 0 O - 8 H E-C CJ _ < C3 0 _ < > O « O <8 n 0 < < • £ > b 0 0 ^ CJ 0 5 _ 0 ra f — > O u 2 J5 E-cu H — H ra > b OD CJ •58 _ O J ? E-> O E & " E- H c < — < 0 3 4 O <c 0 < O O ra 0 ^ O H E-CJ o < < < CJ G E— O E-, < < E-^ H cfl < - 2 >^ CJ •I < X Q 00 CJ < c j 3 2 J CJ >,E-58 _ CJ ra H > o 5 8 5^ _ CJ ra H > o t - < ra C5 5 CJ _ < ra H > O <D CJ E P * 2 _ < o o ^ y ON "es « w u o w o e o DC <•> O *3 o u s 3 cu z Q vo 3 M l <L> > 3 a. <L> -o c -o c CD - O c 3 4) 12 a. CD a, IS c .2P^-c. cn t~-O ^ O GO c e a O N C/3 ' " O S c CD CD O 3 -e C T S3 (D Q a QJ > t o w - O -13 cd CD C cd ^ .a O T3 e cn U S o 3 7 ^ -a cd 1 g ^ C CD 'S 0 E c c« CD • ° § -^ 1 H Q. 32 o t -O CS ON cs m f i o H H E-O < O E~ O u y o 3 O - 1 o u. o s» < - 3 I S < < O O < U < < o o < < < < < < o u % E-H O E— E— < < H E-i O O E-O O o o u y V < o o f— H O O o y o < < y o < < < y o E~ 53 3 < < O < K y _ < > o - h 03 f-, °" - y E- E-O E- < o 3 < rt E-> O O O o «8 a? - J < U — E-s < < o H E-3 O u E_ •J O _ o rt > O < o < o •s o rt EJ < y » u ^ o ^ o o t-* U 3 E-< o u • a y rt E_ o rt ° .2 y < o » o •J < _ o $Z . y JC-O o o 1^  <: < :>y 3 8 3 2 at H -J o & y < o 3 < O o >-y 5 8 GO < . a) O - 8 O O y - < 3 < ° O 5 § ^ 8 E- < s < 5 2 a S3 O _ < > o O 2< < u o = < ^ 8 E- < E-o o o 2 H o - 9 o o o o -J <: = 2 J f-58 > o s= cu —^< u O V Q - < 5S J < < < O o p . O H H i - E->.o 58 ^ 8 _ o > o s 8 M E-a 9-5 ° ° O = < ^ § 3 O Q E~ •s S U O r , 0 B-O t- E-s < ^ o o o o Jj o o O o = 2 J3 S ^ E— < < -1 < . O ^ o o o i - o o o B-O E- H - h < ^ a < < „ o > O „ o < < 3 < o ^ o a o ^ 8 c ° < < to * ^ >> < J < 3 9 E 8 cu (J o s q 3 < < < 3 " 5 E-J O u. o o o c—. o o o 5 O u O u. < a> O cu o < o H E— 5 S < > O fe 8 . o o o o fe o co < h Cu E-. O ^ O O O | 8 cu H o O s o « <; < O ^ 8 > a U Q 3 O E— H | 8 O o » o < o — < o S ^ o « o ^ o P < * 8 5 8 H E-O O o O 3 2 >—] P o o o < o _ o < 8 Sj: o O J= E-cu p o o u O c O = < b ° H E— = y 3 2 - i t o § > o « u 2 < u E— i - H - 2 o o ^ 8 5 5 = y CO < , < < < 53 2 wi E— S 2 < o H o o . o %o o o _ < > o 3 2 ^5 o o *« > a 3 2 J o u, E~ S o ° H < u E-S o 5 5 < O < O u, O O o o < < o a o o o o o < E— O O O U o H H O o a u, o J= o „ < •§ < O O fe o &o E— 3 y < < •- < u O S o u-3 < < E— o o fe o o *s < £ O < O 5 8 E— H 3 2 n < St < a (Si <£, < o ^3 o 3 a o <! o > o s o E-1 — H rt t— > 5 E— E-_ o 13 E— > O e < ° - y < O E- E-t. E— < < ^ y < O X o ii < J2 O < O fe O ai < °-fe < o — E-rt f-< > o i . H 7= O 5- < . O o o D o J= H CU UH _ O 5 8 s tj « o < o u. O H E— ._ E— S O oo [-. _ H rt E^ > O oo y < o °^ y Cu 0 _ o P^ > o E-S < < < . y uC o O O o O E-cu p oo E-^ 8 « y 4 y < o s 8 oo u_ < "rt H > O O o E-1 1 8 y O O < j < ^ y O O o o o 3 3 2 &8 . y o o « y < O u, E~ s o E—< — H rt E-< > O 3 2 J3t 3 " < o u, E-u O ?8 o < J < u- y u o M H OJ O & y < o . y H H 3 < — < rt o <" y ^ H O o _ y rt H > o 8 b i Ct © "3 es u u o u o e o o pa H H _c "S +J o u u & <U 3 er CD V ) z Q CU 3 00 > 3 CL, <D e <3 0> C 3 *-*—< o, OJ a, 0 0 o ctf • — 00 g CT\ cci 1 CU o o ;3 CU O 3 « cr -g Q S o C ^ CU cu . ^ ^ H o w CTS CU c 13 * .2 O -T3 X ! C oo oo <D CU «» S o I CU 3 ^ -a cd I O c oo _ cu "5 ° c c 03 CU CU 3 cr <u H Cu 33 in Figures 1.6 and 1.7. Const ruct ion o f a P I gene clone and the expression o f P I i n Escherichia coli was obtainable. Prote in produced by the clone was detectable w i t h ant i-PI monoc lona l antibodies and was observed to be identical i n size to native protein I. Howeve r , prolonged expression o f protein I dur ing overnight growth was lethal to the E. coli cells, possibly due to the act ivi ty o f the gonococca l por in or alterations in the structural integrity o f the bacterial outer membrane (Carbonett i and Spar l ing, 1987). 1.3.3 Func t ion o f protein I Studies o f outer membrane porins o f Escherichia coli have shown that these monomers adopt a pr imar i ly j3-sheet secondary structure (Grav i to and Rosenbusch, 1980 and Grav i to et al, 1983). In the formation o f the channel, these monomers combine to adopt a quaternary structure to form trimers. Prote in I o f Neisseria gonorrhoeae is also a por in and forms trimers (B lake and Go t sch l i ch , 1982 and Doug la s et al, 1981). Studies examin ing the incorporat ion o f protein I into l i p i d vesicles showed that the protein formed channels in the bi layer, increasing the permeabil i ty to ions and var ious macromolecules , such as sugars (Greco et al, 1980, Douglas et al, 1981). Y o u n g and coworkers (1983) demonstrated that protein I functions as an anion-selective, hydroph i l i c channel w h e n incorporated into art if icial planar bilayers. Incorporated protein was shown to self-associate into a t r imer ic structure forming a voltage-dependent aqueous pore o f at least 11A i n diameter. Studies have also demonstrated that gonococca l protein I can be spontaneously transferred as functional channels from the bacterial membrane into art if icial l i p i d bilayers, as w e l l as red b lood cells (B lake and Go t sch l i ch , 1983 and L y n c h et al, 1984). These results suggest that the " s p i k i n g " o f host cells by protein I w i th its associated ionophor ic properties may enable 34 the bacterium to invade non-phagocyt ic host epithelial cel ls as shown by W a r d et al. (1974) and M c G e e et al. (1978). Infection o f surface epi thel ial cells facilitates spread o f infect ion to under ly ing cells and hence further contributes to the pathogenesis o f gonococca l disease. 1.3.4 An t igen i c properties o f gonococca l protein I The gonococca l surface structures play a cr i t ical role i n the pr imary interaction w i t h the host and facilitate the development o f infect ion (Swanson, 1981). E a r l y studies showed that crude outer membrane preparations cou ld p rov ide some degree o f protection against infection in a guinea p ig mode l (Buchanan and A r k o , 1977). These results suggest that there are antigens on the ce l l surface capable o f induc ing i m m u n i t y to gonococca l infect ion and thus potentially o f great interest in the research and development o f a gonorrhea vaccine (Heckels , 1978 and Gotsch l i ch , 1984). Ant ibod ies against these outer membrane components have been shown to be bacter icidal and opsonic. Therefore, antibodies may provide protection to host cells by inh ib i t ing bacterial ce l l attachment at the mucosal surface or by promot ing phagocytosis and complement-mediated serum k i l l i n g ( W a r d et al., 1978 and V i r j i , 1981). Buchanan and coworkers (1980) showed that indiv iduals that had experienced an episode o f gonococca l pe lv ic inf lammatory disease generated antibodies against the organism that provided some protect ion against recurrent salpingitis. These studies provide further evidence that protective immun i ty may be possible by u t i l i z i n g an outer membrane constituent i n a vaccine formulat ion. Howeve r , the large degree o f antigenic var ia t ion in many o f these surface antigens has prompted a search for ant igenical ly conserved cel l components (Zak et al., 1984). 35 Protein I is an ant igenical ly stable protein that has been shown to elicit the product ion o f bactericidal and opsonic antibodies ( R i c e et al, 1980, Sarafian, 1983 and G u l a t i et al, 1991). In addition, the central role o f protein I i n the pathogenesis o f gonococca l infection by " s p i k i n g " or transferring into the host epithelial ce l l membrane to trigger endocytosis o f the bacterial ce l l makes protein I an attractive target i n the development o f a gonococca l vaccine In vitro studies have shown that monoc lona l antibodies against outer membrane protein I A exhibi ted bactericidal and opsonic act ivi ty and were effective i n protecting epithelial cells f rom the cytotoxic effects o f gonococc i ( V i r j i et al, 1987). V i r j i and coworkers also demonstrated that anti-PEB antibodies were also bactericidal and opsonic w i t h the epitope for antibody b ind ing located at, or close to, the chymotryps in cleavage site w i t h i n the surface-exposed loop region (Fletcher et al, 1986). An t ibod ies directed towards PEB were observed to be more protective against infect ion o f epi thel ial cel ls than were a n t i - P I A antibodies ( V i r j i et al, 1986 and V i r j i et al, 1987). These observations may reflect the greater transferring abi l i ty o f P I A into host epithelial ce l l membranes during infection (Blake , 1985). These protein I characteristics support the proposi t ion that protein I w o u l d be an important antigenic component o f a potential gonococca l subunit vaccine. 36 1.4 I M M U N O L O G I C A L A D J U V A N T S A s mentioned earlier, due to the potential tox ic i ty o f convent ional vaccines, it w o u l d be preferable to identify and purify a single, conserved antigen or generate recombinant subunit or synthetic peptides. Hence , the target and specif ici ty o f the immune response w o u l d be control led and opt imized. Unfortunately, peptide or protein antigens are often non-immunogen ic or only weak ly immunogen ic when administered alone (Richards et al, 1988). In some cases, the purif ied antigen does elici t an antibody response; however, the immune response generated may be directed towards certain determinants that may not provide protective immuni ty against infection. This may be a result o f inadequate surface presentation o f the epitope required for antibody b ind ing and opsonizat ion o f the who le organism (Wetz ler et al., 1988). The immune response to weak immunogens may be ampli f ied when administered w i t h an adjuvant. A number o f adjuvants have been investigated for immune potentiating act ivi ty for use i n vaccines (Ede lman , 1980). M i n e r a l oils , such as Freund ' s adjuvant ( F A ) , have been shown to have ant ibody-st imulat ing properties; however, this is associated w i t h severe side effects, and thus is not acceptable for human subjects. A l u m i n u m hydroxide (or alum) is the on ly adjuvant that has been approved by the F D A for use i n humans. It is less than ideal , as it induces granulomas and painful inf lammat ion at the site o f injection and therefore research has continued to develop safer and more effective adjuvants. L i p o s o m e s have been indicated to have potential as immuno log ica l adjuvants and possible alternatives to alum and F A ( A l v i n g , 1987, A l v i n g , 1991 and Gregoriadis , 1990). The f o l l o w i n g sections describe the structure and funct ion o f l ip ids , methods o f l iposome preparation and role o f l iposomes as i m m u n o l o g i c a l adjuvants. 37 1.4.1 L iposomes : M o d e l b iomembranes B i o l o g i c a l membranes are composed o f a w ide range o f molecules , o f w h i c h l ip ids are major bu i ld ing blocks . The p lasma membrane o f the eukaryotic ce l l acts as a permeabil i ty barrier between the external environment and the internal cel lular machinery as w e l l as p rov id ing a matrix w i t h w h i c h membrane proteins can be associated. Proteins are often inserted into and through this l i p i d b i layer (Figure 1.8, adapted from C u l l i s and H o p e , 1991) The abi l i ty o f l ip ids to adopt a b i layer conformat ion is dictated by the amphipathic characteristics o f these membrane l ipids . L i p i d structure, as shown i n F igure 1.9, consists o f a polar head group that is hydroph i l i c in nature and a non-polar or hydrophobic tail . In the bi layer structure, polar heads are oriented towards the aqueous environment, whereas the hydrophobic acyl chain region is sequestered f rom water. The f luidi ty o f the result ing membrane is dependent on the nature o f the acyl chain compos i t ion and other factors such as sterol content. A n important aspect o f membrane function is the role that integral membrane proteins play in the b io log ica l act ivi ty o f the ce l l . In order to examine the characteristics o f particular proteins, they can be isolated, purif ied and inserted into wel l -def ined, l i p i d model membranes. A variety o f membrane proteins have been reconstituted into such model membranes w h i c h are termed l iposomes (Madden , 1986, M a d d e n , 1988). L iposomes consist o f a l ip id b i layer surrounding an aqueous core. The l i p i d b i layer can be constructed wi th different l i p i d species depending o n the desired l iposome characteristics. 38 Cytoplasm F i g u r e 1.8: T h e e u k a r y o t i c p l a s m a m e m b r a n e . L i p i d , protein and carbohydrates are c losely associated i n the eukaryotic p lasma membrane as described by the f lu id mosaic mode l (Taken f rom C u l l i s and H o p e , 1991). 39 F i g u r e 1.9: G e n e r a l p h o s p h o l i p i d s t ruc tu re . Phosphol ip ids are major b u i l d i n g b locks o f b io log i ca l membranes. The structure o f a b i layer - forming phosphol ip id , l - p a l m i t o y l - 2 - o l e o y l phosphat idylchol ine ( P O P C ) , is illustrated. A s a result o f their amphipathic nature, l ip ids i n an aqueous environment orient w i t h their acyl chains ( R 2 and R 3 ) shielded from the water, whereas the polar head group is directed towards the aqueous medium. The polar head group consists o f the phosphate group, g lycero l backbone and chol ine, the R l substituent. L i p o s o m e s o f defined characteristics can be prepared w i t h var ious substitutions at groups R l , R 2 and R 3 . F o r example, shown are polar head groups serine, ethanolamine, and g lyce ro l that can be substituted to obtain phosphol ip ids w i t h different charges and headgroup sizes. 40 41 1.4.2 L i p o s o m e preparation and characterization A variety o f different techniques have been used to prepare l iposomes. These methods generate systems differing i n size and lamellari ty. The classical method o f preparing l iposomes was first described over 30 years ago (Bangham et al, 1965). The procedure involves the hydrat ion and dispersion o f a dr ied l i p i d f i l m i n water resulting i n spontaneous vesiculat ion. The l iposome popula t ion generated by this technique is usual ly heterogeneous in size (1 um-20 um) (Figure 1.10) and consists o f vesicles w i t h several concentric lamellae. These systems are cal led large mul t i lamel lar vesicles ( M L V ) . Other methods o f l iposome preparation have been rev iewed by S z o k a and Papahadjopoulos (1980). Some o f these procedures invo lve d i s so lv ing l i p i d i n organic solvent and then s l owly injecting or infusing the sample into aqueous buffer. Reverse phase evaporation has also been used and involves m i x i n g l i p i d d isso lved i n organic solvent w i th aqueous buffer fo l lowed by removal o f the solvent under partial v a c u u m to fo rm a th ick ge l o f hydrated l i p i d w h i c h is then diluted to generate large uni lamel la r vesicles ( L U V ) (0.1 um-1 um). Smaller uni lamel lar vesicles o f 20-50 n m can be prepared by sonicat ion o f M L V (Huang, 1969). A l t h o u g h the above-mentioned procedures are suitable for generating l iposomes, the vesicles are often heterogeneous in terms o f size, lamel lar i ty and internal trapped volumes w i t h i n the sample population. V e s i c l e characteristics, part icularly size, lamellar i ty and l i p i d compos i t ion play an important role i n the function o f l iposomes as drug de l ivery systems (Ostro and C u l l i s , 1989 and C u l l i s et ai, 1989) and immunoadjuvants ( A l v i n g , 1987), as w i l l be discussed later. L iposomes w i t h h igh trapped volumes can be prepared from hydrated M L V by subjecting these vesicles to f ive cyc les o f freezing and thawing, w h i c h repeatedly breaks and reforms l i p i d bi layers y i e ld ing vesicles w i t h homogeneous solute distributions 42 F i g u r e 1 . 1 0 : S t r u c t u r e o f l i p i d m o d e l m e m b r a n e s . The structure o f mul t i lamel la r vesicles ( M L V ) , composed o f two or more concentric lamel lae i n an " o n i o n - l i k e " structure, is i l lustrated schematical ly in ( A ) and in a freeze-fracture electron micrograph ( B ) . A schematic representation and a freeze-fracture electron micrograph depict ing the single b i layer structure o f large un i lamel la r vesic les ( L U V ) are shown i n (C) and (D) , respectively. F o r the electron micrographs, the bar shown represents 200 n m and ar row indicates the direct ion o f shadowing (Taken from M a d d e n , 1997). 43 and higher trapped volumes and efficiencies ( M a y e r et al, 1985). H i g h trapping efficiencies are required, for example, for opt imal drug loading in l iposomal drug del ivery systems (Cu l l i s et al, 1989). F rozen and thawed vesicles ( F A T M L V ) can then be size-reduced by a rapid extrusion procedure in w h i c h the F A T M L V are passed ten times through two stacked polycarbonate filters o f defined pore size ( H o p e et al, 1985). Ex t ru s ion o f vesic les through a 100 n m pore size filter has been shown to generate a homogeneous populat ion o f large uni lamel lar ves ic le o f defined size (about 90 nm) ( M a y e r et al, 1986). F o r protein incorporat ion into l iposomes, the freeze-thaw method w o u l d be inappropriate, as the temperature extremes may result i n denaturation o f the protein and alterations in the functional properties o f the reconstituted protein ( M a d d e n et al, 1983 and L y n c h et al, 1984). S imi la r ly , the use o f organic solvents i n the preparation o f proteoliposomes w o u l d l i ke ly result in protein precipi tat ion and denaturation. In addit ion, due to the variable lamellar i ty o f M L V , m u c h o f the protein w o u l d l i ke ly be sequestered wi th in the inner lamellae and therefore these vesicles w o u l d not be suitable for examin ing the function o f intr insic or membrane-spanning proteins ( N i c h o l l s et al, 1980). Furthermore, Shek and colleagues (1983) showed that protein entrapped in L U V el ic i ted a greater antibody response than that generated w i t h M L V , indicat ing that there was a better surface presentation o f protein i n L U V . Therefore, a non-denaturing procedure is required for protein reconstitution into uni lamel lar l iposomes. Detergent dialysis is a c o m m o n technique employed in the incorporat ion o f protein into l iposomes ( M i m m s et al, 1981) i n a functionally active state (Madden et al, 1983). T y p i c a l l y , l i p i d and protein are co -solubi l ized in detergent, w h i c h is then s l o w l y removed by passive dia lys is against aqueous buffer (Figure 1.11). A s the detergent concentration is decreased be low its cr i t ica l mice l l a r 44 Figure 1.11: Protein reconstitution into liposomes. The process o f detergent dialysis dur ing protein reconsti tution into l iposomes is illustrated. Protein so lubi l ized i n detergent is m i x e d w i t h l i p i d i n detergent and then the detergent is s lowly removed by dialysis . A s the detergent concentration is lowered be low the cr i t ica l mice l la r concentration ( C M C ) , l i p i d molecules rearrange to form bilayers w i t h protein incorporated i n the l i p i d membrane. 46 concentration, the l i p id spontaneously vesiculates w i t h l iposomal protein either inserted into the bi layer, trapped wi th in the aqueous core or attached to the l iposome surface. Dens i ty gradient centrifugation experiments have shown free protein to migrate in a peak separate from l i p id , l i ke ly as denatured protein (Stahn et al, 1992). It has been shown that vesicle size can influence the c i rcula t ion l ifetimes o f l iposomes (Juliano and Stamp, 1975 and Senior et al, 1985). M L V and S U V , depending on their l i p i d composi t ion, have short half- l ives i n c i rcula t ion, whereas L U V generally have longer residence times ( A l l e n et al, 1989). The enhanced clearance rates o f M L V may be due to the greater degree o f interaction between vesicles and plasma proteins or l ipoproteins (Scherphof et al, 1978 and F inke l s te in and Wei s sman , 1979). F o r S U V , the bi layer has a very smal l radius o f curvature and the l ip ids experience a greater degree o f stress, producing packing defects in the l i p i d b i layer ( C u l l i s and H o p e , 1991). The b i layer defects make the vesicles unstable and leaky and thus susceptible to the insertion or b ind ing o f l ipoproteins, and apoproteins, facil i tat ing their rapid clearance from circulat ion. Enhanced c i rcu la t ion times w o u l d be o f particular importance for the del ivery o f chemotherapeutic agents; however, for a l iposomal subunit vaccine , clearance o f the vesic les to the ret iculoendothelial system ( R E S ) w o u l d enhance antigen de l ivery and uptake by A P C , such as macrophages, B cells and dendrit ic cells. Target ing and de l ivery o f l iposomes to the R E S w o u l d therefore lead to an enhanced immune response. The l i p i d compos i t ion o f l iposomes can influence the characteristics o f the l iposomal system. Studies have shown that addit ion o f cholesterol results in a less permeable and more r ig id membrane. Th is results i n reduced leakage from l iposomes and decreased b ind ing o f high-density l ipoprote in ( A l l e n , 1981), w h i c h translates into extended c i rcula t ion t imes 47 ( A l l e n , 1989). These aspects have part icular ly importance in the encapsulation, retention and del ivery o f chemotherapeutic agents. Incorporat ion o f g lyco l ip ids into the bi layer l i p i d mixture results in increased c i rcula t ion half- l ives , possibly by p rov id ing a steric hindrance preventing apolipoprotein from gain ing access to and inserting into the l i p i d b i layer ( A l l e n , 1989). The particular characteristics o f l i p i d bi layers p lay an important role i n the fate o f l iposomes in vivo, especial ly w i t h regard to the pharmacokinet ics o f encapsulated chemotherapeutic agents. S imi la r ly , the charges o f the l ip ids compos ing the b i layer can influence the fate o f l iposomes. Studies have shown that cationic l iposomes are taken up by macrophages more efficiently than neutral or anionic l iposomes (Nakanish i et al, 1997). Therefore, the l i p i d compos i t ion o f a l i posomal subunit vaccine should be characterized to determine the effects on protein insertion, orientation wi th in the l i p i d b i layer and the influence on antigenicity. 1.4.3 L iposomes as drug del ivery systems L iposomes have been w i d e l y used as carrier systems for var ious chemotherapeutic agents, part icularly antifungal and antineoplastic drugs (Meh ta and Lopez-Beres te in , 1989 and Northfel t et al, 1998). Encapsula t ion o f drugs i n l iposomes has been shown to be w e l l tolerated wi th increased efficacy and decreased tox ic i ty compared to free drug. ( M a y e r et al, 1989, B o m a n et al, 1994 and Chang et al, 1997). An t i - t umor agents are often associated wi th cardiac toxic i ty ; however , encapsulation in l iposomes results i n reduced cardiac uptake and cardiac toxic i ty (Rahman et al, 1980 and G a b i z o n et al, 1982). The increased drug efficacy is probably due to decreased clearance by cel ls o f the R E S , prolonged c i rcu la t ion l ife-t ime and accumulat ion and release at the tumor site. L i p o s o m a l encapsulation o f 48 therapeutic agents appears to be an effective approach to therapy and may have advantages compared to conventional methods o f chemotherapy. These studies also demonstrate that l iposomes are effective carrier systems that are safe as carriers i n a potential subunit vaccine. 1.4.4 L i p o s o m e s as i m m u n o l o g i c a l adjuvants Adjuvants are thought to function by t w o possible mechanisms ( A l l i s o n and Bya r s , 1986). One mechanism is that the adjuvant creates a depot at the injection site and retards antigen clearance (Ant imis ia r i s et al, 1993). Reduced loca l clearance thus prolongs the duration o f release and interaction w i t h antigen presenting cells ( A P C ) . The loca l in f lammat ion generated by formulat ions containing a lum or F A may result in the migra t ion and infi l tration o f A P C to the site o f inoculat ion, hence faci l i tat ing an immune response. A second mechanism o f act ion is the act ivat ion o f macrophages, a subset o f the A P C . F o r example, the presence o f l ipopolysacchar ide ( L P S ) or muramyl dipeptide ( M D P ) in the adjuvant preparation can stimulate macrophages to release inter leukin 1 (EL-1) (Gregoriadis , 1990). The combinat ion o f antigen and EL-1 stimulates T cells to produce EL-2 and other cytokine factors, w h i c h trigger cel l -mediated ( C M I ) or humora l immun i ty (HI) . V a c c i n a t i o n by de l iver ing antigens in l iposomes may be advantageous over other preparations due to the targeting and uptake o f l iposomes by macrophages, dendrit ic cel ls and B cells w h i c h are major sites i n the pathways o f i m m u n o l o g i c a l antigen processing (Unanue, 1984). Investigators have also shown that macrophages are a major subset o f antigen-presenting cel ls capable o f T H 1 ce l l ac t ivat ion (Brewer et al, 1994) Dep le t ion o f macrophages resulted in the deplet ion o f T H 1 cell-associated cytokines as w e l l as suppression o f I g G 2 a antibodies. A s mentioned earlier, the locat ion o f antigen processing 49 dictates the type antigen presentation on the A P C surface B r i e f l y , endogenous peptides are presented on M H C class I molecules, whereas exogenous antigens are associated w i t h M H C class II surface molecules. Immuniza t ion w i t h soluble antigens alone results i n exogenous antigen processing through the endosomal compartment and therefore insufficient presentation on M H C class I molecules and poor induct ion o f C T L (Hard ing et al, 1991). Howeve r , antigen del ivery in l iposomes to A P C leads to phagocytosis o f l iposomal antigen and processing to the t rans-Golgi where antigen is bound to M H C class I molecules resulting i n a C T L response (Reddy et al, 1992, N a i r et al, 1992 and R a o et al, 1999). The results o f these studies indicate that l iposomes may be efficient adjuvants for e l ic i t ing a cel l -mediated immune response for protection against intracellular pathogens. M o r g a n and coworkers (1984) showed that an Ep te in -Bar r antigen i n complete Freund ' s adjuvant ( C F A ) induced on ly a weak and delayed response, whereas l i posomal ly encapsulated antigen el ici ted higher and prolonged antibody titers. Studies w i t h a cloned, synthetic, malar ia c i rcumsporozoi te ( C S ) peptide has been shown to be non- immunogenic on its o w n , but could be rendered h igh ly antigenic by incorporat ion into l iposomes ( A l v i n g et al., 1986). Furthermore, the anti-peptide antibodies generated by immuniza t ion w i t h l i p o s o m a l - C S peptide reacted w e l l w i t h intact ( l ive and k i l l ed ) sporozoite organisms. M o r e importantly, antibody titers in i m m u n i z e d rabbits remained elevated over a prolonged duration f o l l o w i n g immuniza t ion ( A l v i n g et al, 1986 and A l v i n g , 1987). In addit ion to e l ic i t ing protective humoral immuni ty , a l i p o s o m a l - C S peptide vacc ine has also been shown to induce cyto ly t ic T lymphocyte ( C T L ) responses w h i c h are considered to be important in induc ing protective immuni ty (Whi te et al, 1993). 50 Pre l iminary research evaluated and compared the antigenic properties o f gonococca l protein I either g iven by i t se l f or combined w i t h various adjuvants (Wetzler et al, 1988, Wetz le r et al, 1989a, Wetz le r et al, 1991 and Wetz l e r et al, 1992). These studies showed that protein I was capable o f e l ic i t ing antibodies i n a rabbit immuniza t ion mode l either g iven alone, in a formulat ion w i t h Freund ' s adjuvant, adsorbed to a lum or incorporated into l iposomes. A l t h o u g h antibodies were generated against the antigen by each o f the formulations, observations indicated that Freund ' s adjuvant and l iposome preparations el ici ted the highest level o f surface reactive, bactericidal antibodies. In addit ion, antisera from pro teol iposome- immunized rabbits were far superior to protein I -a lum sera i n the agglutination and opsonophagocytosis o f who le gonococca l organisms (Wetz le r et al, 1988). The results o f these studies and those o f W h i t e and coworkers (1993) suggest that l iposomes are effective adjuvants in e l ic i t ing both humoral and cel l -mediated immuni ty and are w e l l tolerated and have few adverse effects. Aspects such as these make l iposomes attractive alternatives to a lum for a potential gonococca l subunit vaccine . The rise in the number o f antibiotic-resistant strains and the enormous cost associated w i t h the treatment o f gonorrhea and its compl ica t ions have prompted research into methods o f disease prevention, namely the development o f a vaccine against the gonococca l organism. Pre l iminary research by Wetz l e r and coworkers (1988) showed that a l iposomal gonococca l subunit vaccine exhibi ted immunogen ic act ivi ty i n the rabbit immuniza t ion model . Howeve r , the l iposomal formulat ion needs to be further characterized w i t h regard to the protein incorporat ion and orientation, l i p i d compos i t ion and detergent removal dur ing the reconstitution process. In addit ion, the act ivi ty o f the proteol iposome formulat ion i n e l i c i t ing humoral and cel l -mediated immuni ty needs to be evaluated. The research presented i n this 51 thesis deals w i t h the development and characterization o f a subunit vacc ine against Neisseria gonorrhoeae. 52 1.5 RESEARCH HYPOTHESES The historical basis o f vacc ina t ion has been to use k i l l ed or attenuated whole organisms, w h i c h are therefore non-pathogenic, to elicit protective immune responses. Howeve r , these preparations contain a large number o f antigens that may g ive rise to adverse reactions to the vaccine itself. In order to avoid these compl ica t ions , it w o u l d be preferable to develop a subunit vaccine based o n a single, conserved protein rather than the who le organism. Unfortunately, separation o f proteins from membranes renders them non-immunogen ic or only w e a k l y immunogen ic and unable, therefore, to elicit a protective immune response. A d d i n g an immunoadjuvant to the formulat ion, however , can restore the antigenicity o f these isolated proteins. It is proposed that incorporat ing a purif ied bacterial antigen into l iposomes could restore its native protein orientation and, therefore, restore and enhance the antigenicity o f the membrane protein to elicit an effective immune response. 1.6 SPECIFIC RESEARCH OBJECTIVES The objectives o f the research were to: 1. Characterize gonococca l membrane protein incorporat ion into l iposomes and determine the factors inf luencing protein reconstitution. F o r a comparison, evaluate the efficiency o f meningococca l outer membrane proteins ( M O M P ) incorporat ion into l iposomes. 2. Determine the b iophys ica l characteristics o f l iposomal gonococca l P o r vaccine i n terms o f protein orientation, vesicle size and ves ic le morphology . 53 3. Characterize in vitro antibody b ind ing act iv i ty to reconstituted P o r proteol iposome as a function o f l i p i d composi t ion . 4. Determine antigenicity o f P o r proteoliposomes in vivo us ing a murine model . Compare antibody titers for proteoliposomes and free protein. 5. Evaluate the nature o f the antibody response i n a murine model , humoral or cel l-mediated, based o n immunog lobu l in serotypes. 54 C H A P T E R 2 M A T E R I A L S A N D M E T H O D S 2.1. L i p i d s , chemicals and reagents l -Pa lmi toy l -2-o leoy l - sn-g lycero-3-phosphochol ine ( P O P C ) ( M W 760.1) was purchased from Nor thern L i p i d s Inc., Vancouver , B . C . 1-Palmitoyl , 2 -o leoyl -sn-g lycero-3-phosphoethanolamine ( P O P E ) ( M W 718), 1-Palmitoyl , 2-oleoyl-sn-glycero-3-phosphoserine ( P O P S ) ( M W 784) and 1-Palmitoyl , 2 -o leoyl -sn-g lycero-3-phosphoglycero l ( P O P G ) ( M W 771) were purchased from A v a n t i Po la r L i p i d s , Alabaster , A L . N , N - d i o l e y l - N , N -d ime thy lammonium chlor ide ( D O D A C ) ( M W 582.5) was obtained from I N E X Pharmaceuticals, Vancouver , B . C . N-Octy l -P -D-glucopyranos ide ( O G P ) , H E P E S , (N- [2 -hydroxyethyl]piperazine-N'-[2-ethanesulfonic acid]) ( M W 238.3), chol ic ac id ( 3 a , 7 a , 1 2 a -Tr ihydroxy-5p-cho lan-24-o ic acid) ( M W 430.6) (sodium salt), bovine serum a lbumin ( B S A ) and o-phenylenediamine (1,2-benzenediamine) d ihydrochlor ide ( M W 181.1) were obtained f rom S igma C h e m i c a l C o . , St. L o u i s , M O . [ 1 4 C]-N-Octy l -P -D-g lucopyranos ide ( 1 4 C - O G P ) was purchased f rom A m e r i c a n Rad io labe led Chemica l s , Inc., St. L o u i s , M O . L - a -d ipa lmi toy l - [2 -pa lmi toy l -9 ,10- 3 H(N)] -phospha t idy lcho l ine ( 3 H - D P P C ) and [ 3 H ] - c h o l i c ac id were obtained from Dupont , Bos ton , M A . The enzymes used, a -chymot ryps in ( E C . 3.4.21.1) and trypsin ( E C . 3.4.21.4), were purchased from C a l b i o c h e m , L a Jol la , C A . F i c o l l 400 was obtained f rom Pharmacia , Uppsa la , Sweden. The reagents for the protein assay were purchased from Pierce, R o c k f o r d , I L . The reagents used i n the sodium dodecyl sulfate po lyacry lamide gel electrophoresis and S M - 2 Biobeads were obtained from B i o r a d , Hercules , C A . The reagents for the B C A protein assay and N H S - L C - B i o t i n were purchased 55 from Pierce, R o c k f o r d , DLL. The peroxidase-conjugated streptavidin, goat anti-mouse I g G , rabbit anti-mouse I g G and mouse anti-rabbit I g G were obtained from R o c k l a n d , Gi lber t sv i l l e , P A . The gonococca l membrane protein I B (Por) preparations used i n this study were i n purif ied form (free o f R m p and L O S ) (Wetz le r et al, 1989b) and were supplied by D r . M i l a n S. B l a k e , Rockefe l le r Un ive r s i ty , N e w Y o r k , N Y through D y n C o r p P R I . T w o P o r samples were used i n the course o f these studies: A solut ion o f P o r (4 mg/ml) i n 0.1 M Tr i s -H C l , 0.2 M N a C l , 10 m M E D T A , 0 .05% Z 3 , 1 4 , 0 .02% A z i d e at p H 8.0 and a l y o p h i l i z e d sample prepared made under G o o d Manufac tu r ing Pract ices ( G M P ) . The meningococca l outer membrane proteins containing detoxif ied l ipool igosacchar ide (2:3 wt . /vol . ) ( M O M P ) and E m p i g e n B B ( N - d o d e c y l - N , N - d i m e t h y l g l y c i n e ) ( M W 272) were supplied by D r . W e n d e l l Zo l l inge r , Wal te r R e e d A r m y Institute, Wash ing ton , D C . 2.2. Reconst i tut ion o f gonococca l protein I (Por) into l iposomes 2.2.1 Reconst i tu t ion o f soluble P o r P o r protein was reconstituted at different protein- to- l ipid ratios f rom O G P or cholate to determine the effect o f protein concentrat ion on protein incorporat ion efficiency. In addition, the kinetics o f detergent removal was moni tored during the reconstitution process. P o r protein was di luted to 0.5 m g / m l or 1.0 m g / m l protein i n either 400 m M O G P , 20 m M H E P E S p H 7.4 or 200 m M sodium cholate, 20 m M H E P E S p H 7.4. These solutions (3 ml) were then added to 150 m g o f P O P C and the phosphol ip id was d issolved by gentle vor tex ing at 2 5 ° C . The radiolabel [ 3 H ] - D P P C (0.03 p C i / m g phosphol ip id) was included as a l i p i d marker. A l i q u o t s (1 ml) o f the O G P solutions were set aside as the 0 hour dialysis t ime point or were transferred to dialysis tubing and d ia lyzed for either 20 hours or 125 hours at 56 4 ° C against 500 volumes o f 150 m M N a C l , 20 m M H E P E S , p H 7.4. Exte rna l buffer changes were made at 20 and 50 hours. In some experiments, removal o f O G P was fo l lowed us ing [ 1 4 C ] - O G P (2 p C i / m l ) . D u r i n g dialysis , samples were taken from the dialysis tubes after 0, 1, 2, 4, 8, 16, 24 and 36 hours o f dialysis . S imi l a r ly , al iquots (1 ml) o f the cholate-containing sample were set aside as the 0 hour dialysis t ime point and were transferred to dialysis tubing and d ia lyzed for either 60 hours or 125 hours at 4 ° C against 500 volumes o f 150 m M N a C l , 20 m M H E P E S , p H 7.4. Exte rna l buffer was changed at 20, 50 and 100 hours. In some cholate-mediated reconstitution experiments, the radiolabel [ 3 H ] - c h o l i c ac id (3.3 p C i / m l ) was inc luded in the reconstitution mixture. D u r i n g dialysis , al iquots were taken from the dialysis tubes at 0, 2.5, 5, 10, 20, 30, 50, 70,130 and 150 hour t ime points. Res idua l O G P and cholate concentrations were determined based o n scint i l la t ion count ing o f [ I 4 C ] - O G P and [ 3 H ] - c h o l i c acid, respectively, in a B e c k m a n L S 3801 instrument (Ful ler ton, C A ) . 2.2.2 Reconst i tut ion o f l y o p h i l i z e d P o r L y o p h i l i z e d P o r protein was so lub i l i zed i n 400 m M O G P , 20 m M H E P E S , p H 7.4 by gentle homogenizat ion us ing a glass/teflon homogenizer . Some batches o f purif ied P o r contained smal l quantities o f insoluble material w h i c h was removed by first centrifuging the sample at 2000 rpm on a Bax te r Megafuge 1.0 (Heraeus Instruments) for 3 minutes and then passing the supernatant through a 0.2 m i c r o n cel lulose acetate filter ( M i c r o Fi l t ra t ion Systems) to y i e ld a clear solution. The protein concentrat ion was adjusted to 1 m g / m l and aliquots (1 ml) o f this solut ion were then added to either 50 m g P O P C or 50 m g P O P C P O P E (1:1 weight ratio). The phosphol ipids were d issolved by gentle vor tex ing at 2 5 ° C and the samples were then placed in SpectraPor II dialysis tubing (6.4 m m diameter) and d ia lyzed at 57 4°C against 500 volumes o f 150 m M N a C l , 20 m M H E P E S , p H 7.4 for 36 hours w i th one change o f external buffer at 16 hours. The radiolabel [ 3 H ] - D P P C (0.03 u C i / m g phosphol ipid) was inc luded as a phosphol ip id marker. 2.2.3 Reconst i tu t ion o f P o r into anionic and cat ionic proteoliposomes T o assess the effect o f l i p i d species and charge on protein incorporat ion efficiency, P o r protein was reconstituted into l iposomes o f va ry ing l i p i d content. B r i e f l y , dry phosphol ip id , P O P C , was weighed w i t h 5, 10 or 2 5 % (by wt.) o f P O P G or P O P S i n a total o f 50 m g l i p id . M i x t u r e s o f P O P C and 5, 10 or 2 5 % (by wt.) (50 m g total l ip id) D O D A C (23.28 mg/ml) in benzene:methanol (95:5 v /v ) were prepared by co - lyoph i l i za t ion f rom benzene:methanol (95:5 v/v) (Vi r t i s , Gardiner , N Y ) under h igh v a c u u m (60-100 mtorr) for a m i n i m u m o f 5 hours. A solut ion o f 400 m M O G P , 20 m M H E P E S at p H 7.4 was added to the dry l i p i d mixtures. The radiolabel [ 3 H ] - D P P C (0.03 u C i / m g phosphol ip id) was inc luded as a l i p id marker. L i p i d s were so lub i l i zed by gentle vor tex ing and sonicat ion at 2 5 ° C . P o r protein from the stock solution (4 mg/ml) was added to the l i p i d mixture to give protein and l i p i d concentrations o f 1.0 m g / m l and 50 m g / m l , respectively, i n 400 m M O G P , 20 m M H E P E S p H 7.4. F o r control vesicles, H B S , p H 7.4 was added to the detergent/lipid mixture instead o f P o r protein. E a c h sample was transferred to SpectraPor dialysis tubing (6.4 m m diameter) and d ia lyzed for 125 hours at 4 ° C against 500 vo lumes o f 150 m M N a C l , 20 m M H E P E S , p H 7.4 w i th external buffer changes at 20 and 50 hours. F o l l o w i n g dialysis , reconstituted samples were analyzed by i sopycn ic density gradient centrifugation, Q E L S size analysis, protease digestion and cryo-electron microscopy, as described below. 58 2.3 Reconst i tut ion o f meningococca l proteins ( M O M P ) M e n i n g o c o c c a l proteins were reconstituted f rom O G P and E m p i g e n B B to determine the rate o f detergent removal and eff iciency o f protein incorporat ion. Howeve r , it was observed that E m p i g e n B B alone was insufficient to dissolve phosphol ip id pr ior to reconstitution. Therefore, proteoliposomes were reconstituted from a mixture o f sod ium cholate and E m p i g e n B B , as described be low. M O M P (0.5 mg/ml) i n 400 m M O G P , 20 m M H E P E S , p H 7.4 was prepared and added (2 ml) to 50 m g P O P C . The radiolabel [ 1 4 C ] - O G P (2 p C i / m l ) was inc luded as a detergent marker. The phosphol ip id was d isso lved by gentle vor tex ing at 2 5 ° C and samples were then placed in SpectraPor II dialysis tubing (6.4 m m diameter) and d ia lyzed for 140 hours at 4 °C against 500 volumes o f 150 m M N a C l , 20 m M H E P E S , p H 7.4 i n the presence or absence o f S M - 2 Biobeads (2 g). Exte rna l buffer changes were made at 20, 50 and 116 hours. D u r i n g dialysis , al iquots were taken from the dia lys is tubes at 0, 2.5, 5, 10, 20, 30, 50, 116 and 140 hour t ime points. M O M P (0.5 mg/ml) i n 200 m M sod ium cholate, 20 m M H E P E S , p H 7.4 w i t h 5.6% E m p i g e n B B was prepared and added to dry P O P C (50 mg/ml) . The phosphol ip id was dissolved by gentle vor tex ing at 2 5 ° C and the samples were then placed i n dialysis tubing and d ia lyzed for 12 days at 4 °C against 500 volumes o f 150 m M N a C l , 20 m M H E P E S , p H 7.4 w i th external buffer changes at 36, 82, 142, 214 and 250 hours. F o l l o w i n g dialysis , reconstituted samples were analyzed by i sopycn ic density gradient centrifugation and Q E L S size analysis. 59 2.4 A n a l y t i c a l procedures 2.4.1 Isopycnic density gradient centrifugation A continuous F i c o l l gradient was prepared (0 -10% F i c o l l ) in 150 m M N a C l , 20 m M , p H 7.4 using a Gradient M a k e r (Hoefer Scient i f ic Instruments, San Franc isco , C A ) . The reconstituted proteoliposomes (500 ul) were loaded o n the gradient w h i c h was then centrifuged in a B e c k m a n S W 41 T i swing ing bucket rotor on a B e c k m a n L 2 - 6 5 B ultracentrifuge at 110 ,000g a v for 24 hours at 4 °C . The gradients were then fractionated into 500 u l fractions (see F igure Legends for details) and analyzed for protein and l i p i d content. The density gradient profiles were representative o f at least t w o separate experiments. 2.4.2 Protein and phosphol ip id quantitation Phospho l ip id concentrations and specific activit ies were determined by assaying l i p i d phosphorus content (F iske and Subbarow, 1925) and by l i q u i d scint i l la t ion count ing o f the samples on a B e c k m a n L S 3801 l i qu id sc in t i l la t ion counter Protein concentrations were determined by a modi f ied Pierce B C A protein assay for microti ter plates (Pierce, R o c k f o r d , I L ) . A 50 u l al iquot o f each standard, blank or di luted u n k n o w n sample was pipetted into the appropriate microti ter plate wel ls . Then, 50 u l o f 0 .5% S D S was added, fo l lowed by the addi t ion o f 100 ul o f w o r k i n g reagent to each w e l l . The microti ter plates were incubated at 3 7 ° C for 2 hours. Absorbance was measured at 540 n m on a B i o t e k 96 we l l microt i ter plate reader ( B i o T e k Instruments, W i n o o s k i , V T ) . Prote in concentrations in the samples were determined f rom a bovine serum a lbumin ( B S A ) standard curve. 60 2.5 Protease digestion The susceptibil i ty o f reconstituted P o r protein to proteolytic cleavage by either t rypsin or a -chymot ryps in was determined for the proteol iposome fractions obtained fo l l owing i sopycnic density gradient centrifugation. In the case o f reconstituted systems prepared wi th either P O P C or P O P C P O P E (1:1), aliquots (25 p i containing 17.5 p g protein, P O P C systems or 30 p g protein, P O P C : P O P E systems) were treated w i t h either a -chymotryps in (0.5 pg) or t rypsin (0.2 pg) at 37 ° C for 15 minutes. F o l l o w i n g enzyme digestion, the proteoliposomes were subjected to a del ip idat ion procedure (see be low) pr ior to SDS-po lyac ry l amide gel electrophoresis. 2.6 SDS-po lyac ry l amide gel electrophoresis ( S D S - P A G E ) P r io r to S D S - P A G E , reconstituted proteoliposomes were first delipidated. T o each sample, 400 p i methanol, 200 p i ch lo roform and 300 p i d is t i l led water were added. Samples were vortexed and centrifuged at 2000 rpm for 10 minutes. The top layer above the protein interface was carefully removed and discarded. A n addit ional 300 p i o f methanol was added to each sample fo l lowed by vor tex ing and centrifugation at 2500 rpm for 10 minutes. The supernatant was removed and discarded. The protein pellets were then dried under nitrogen gas and resuspended in 25 p i o f 1% S D S . SDS-po lyac ry l amide gel electrophoresis was undertaken us ing the discontinuous buffer system ( L a e m m l i , 1970) employ ing a M i n i Protean I f D u a l Slab C e l l apparatus (Biorad) wi th a 19.5%(w/v) separating gel and a 4 .5%(w/v) stacking gel . Samples were dissociated i n 0.062 M T r i s / H C l buffer, p H 6.8 containing 2%(w/v) S D S , 10%(w/v) sucrose, 5%(v/v) 2-f3-mercaptoethanol and 0 .001%(w/v) bromophenol blue at 9 5 ° C for 4 minutes. 61 Samples were loaded on the stacking gel a long w i t h molecular weight standards and electrophoresis was conducted at a constant voltage o f 80 V through the stacking gel and 130 V through the resolving gel un t i l the b romophenol blue t racking dye was approximately 5 m m from the gel bottom. Proteins were detected by 0 . 1 % Coomass ie B l u e stain i n f ixat ive (40% methanol, 10% acetic acid) and destained w i t h 4 0 % methanol, 10% acetic acid. Ge l s were then si lver stained us ing the B i o r a d s i lver staining kit without modif icat ion. 2.7 G e l scanning densitometry T o quantitate proteolytic digestion, S D S gels were scanned us ing the L K B - B r o m m a Ul t rascan X L Laser Densitometer. The relative amounts o f each peptide band were determined by w e i g h i n g peaks f rom the densitometer scan profi le and calcula t ing percentage relative to the total peak weights for that lane. 2.8 Size reduction o f reconstituted proteoliposomes by extrusion F o l l o w i n g reconstitution, proteoliposomes were size-reduced us ing an extrusion procedure (Hope et al, 1985 and M a y e r et al, 1986a). B r i e f l y , reconstituted systems were placed in an Ext ruder ( L i p e x Biomembranes , Vancouver , B . C . , Canada) and extruded 10 times through two (stacked) polycarbonate filters (Costar, Cambr idge , M A ) o f defined pore size (100-600 nm) under nitrogen pressures o f 100-400 psi at 3 7 ° C . F o l l o w i n g extrusion, ves ic le size distributions were determined us ing quasi-elastic l ight scattering ( Q E L S ) . In addition, samples were analyzed for protein and l i p i d concentration. 62 2.9 Quasi-elastic l ight scattering ( Q E L S ) Reconst i tuted systems were analyzed to determine ves ic le size distributions by quasi-elastic light scattering ( Q E L S ) analysis us ing a N i c o m p M o d e l 270 Submic ron Par t ic le Sizer as described previously (Ko lchens et al, 1993). 2.10 Cryo- t ransmiss ion electron mic roscopy ( c r y o - T E M ) Reconst i tuted vesicles were analyzed us ing the technique o f cryo-transmission electron microscopy. Br i e f l y , sample f i lms were prepared i n a custom-buil t environmental chamber under control led temperature ( 2 5 ° C ) and humid i ty condit ions. The films were then vi t r i f ied by rapid freezing i n l i q u i d ethane and transferred to a Zeiss E M 902 transmission electron microscope for analysis. The specimens were kept b e l o w 108 °K during the transfer and v i e w i n g procedures to prevent sample perturbation and ice formation. The microscope was operated in zero-loss, br ight-f ield mode and at an accelerating voltage o f 80 k V (Bel lare etal, 1988 and Dubochet etal, 1988). 2.11 A n t i b o d y b ind ing evaluat ion 2.11.1 Sample preparation for antigenicity tests The antigenic properties o f five reconstituted P o r proteol iposome formulations were compared us ing an antibody b ind ing assay. In this experiment, two protein batches were examined ( G M P and M S l l j e A r m p ) and these were reconstituted w i t h either P O P C alone or P O P C P O P E (1:1) (as described above). F o l l o w i n g reconstitution, these samples were s ize-reduced by extrusion as described above. A fifth sample contained P O P C P O P E (1:4) and was prepared w i t h P o r MSIIJBA/TM/J according to a procedure described previously (Wetz le r 63 et al, 1988, Wetz le r et al, 1992). B r i e f l y , the dry l i p i d was so lub i l i zed in a solut ion containing MS11JBA/TM£> (2 mg/ml) in 400 m M O G P , 20 m M H E P E S , p H 7.4 and then d ia lyzed against 500 volumes o f 150 m M N a C l , 20 m M H E P E S , p H 7.4 at 4 ° C for 36 hours w i t h changes o f external buffer after 10 and 20 hours. F o l l o w i n g dialysis , the sample was sonicated in a water bath at 3 7 ° C for 20 minutes to produce small uni lamel lar vesicles. 2.11.2 P o r i n monoclona l antibodies The monoclonal antibodies ( M A b ) u t i l i zed i n these studies were produced and characterized us ing purif ied porins (Wetz le r et al, 1988) f rom various strains. The hybr idoma cells producing these M A b s were c loned by l i m i t i n g d i lu t ion and acc l imat ized for growth in serumless H y b r i m a x media ( B R I , Ba l t imore , M D ) . The M A b s were purif ied us ing a c o l u m n packed w i t h H A - U l t r a m a x ( S i g m a C h e m i c a l C o . , St. L o u i s , M O . ) as previously described (Stanker et al, 1985). The antibody fractions were pooled and the antibodies precipitated by the addi t ion o f an equal v o l u m e o f 3 0 % (wt. /vol .) polyethylene g l y c o l 8000 ( J T . Baker , Ph i l l ip sburg , N J ) . The precipitates were col lected by centrifugation at 30,000g for 20 minutes, the supernatant discarded and the antibodies resuspended i n P B S . The purif ied M A b s were stored at 4 ° C unti l used. 2.11.3 E L I S A and inh ib i t ion E L I S A assays Mic ro t i t e r plates (Nunc- Immuno Plate I IF, V a n g a r d International, Neptune, N . J . ) were sensitized by adding 0.1 m l per w e l l o f the purif ied por in from the strain M S I l jeAr/wp, 2.0 pg /ml i n 0.1 sodium bicarbonate buffer, p H 9.6. The plates were incubated overnight at r oom temperature. The plates were washed five t imes w i t h 0 .9% N a C l , 0 .05% T w e e n 20, 10 64 m M sodium acetate, p H 7.0, 0 .02% sodium azide. The Neisse r ia l por in monoc lona l antibodies were di luted i n P B S and added to the plate and incubated for 2 hours at r oom temperature. The plates were again washed as before and the secondary antibody, a lkal ine phosphatase-conjugated goat anti-mouse I g G and I g M (Tago Inc., Bur l ingame , C A ) , was di luted in P B S / 0 . 5 % T w e e n 20, added to the plates and incubated for 1 hour at r oom temperature. The plates were washed as before and /?-ni trophenyl phosphate (S igma Phosphatase Substrate 104) (1 mg/ml ) in 0.1 M diethanolamine, 1 m M M g C ^ , 0.1 m M ZnCl2, 0 .02% sodium azide, p H 9.8 was added. The plates were incubated at r oom temperature for 1 hour and the absorbance at 405 nm was determined us ing an E L 3 1 1 s x Automated M i c r o p l a t e Reader ( B i o - T e k Instruments, Inc., W i n o o s k i , V T ) . Con t ro l we l l s lacked either the pr imary and/or secondary antibody. Th i s was done to obtain a titre for each monoclona l antibody w h i c h w o u l d g ive a h a l f m a x i m a l reading i n the E L I S A assay. Th i s titer for each por in monoc lona l antibody was then used in an inh ib i t ion E L I S A . The microti ter plate was sensitized and washed as before. L iposomes containing the indicated amount o f porins (measured i n | i g protein), as w e l l as w h o l e bacteria, were added to a separate V-bot tomed microti ter plate (Nunc - Immuno Plate 9 6 V , Vanga rd International, Neptune, N . J . ) and di luted i n P B S A n equal v o l u m e o f each o f the por in monoc lona l antibodies, di luted i n P B S , was added to the wel l s such that the final concentration o f the antibody represented their specific " h a l f m a x i m a l " titers. The total vo lume i n each w e l l was 150 u.1. The plates were incubated for 2 hours on a horizontal plat form shaker. The microti ter plates were then centrifuged at 3000 rpm for 5 minutes i n a R C 5 Superspeed refrigerated centrifuge (Dupont Instruments, W i l m i n g t o n , D E ) us ing a S H - 3 0 0 0 rotor w i t h microplate carriers. A n aliquot (100 ul) f rom each w e l l o f the V - b o t t o m microti ter plate was 65 transferred to the pre-washed porin-sensi t ized microti ter plate. Th is plate was incubated for 2 hours, washed and the conjugated second antibody added as stated. The plate was then processed and read as described. The inh ib i t ion was calculated as fo l lows : 1 - ( E L I S A value after absorption) ( E L I S A value o f unabsorbed ant ibody) E a c h o f the assays was done in tr ipl icate o n different days. The inh ib i t ion values represent the mean o f these assays. 2.12 A n t i b o d y b ind ing act ivi ty o f anionic and cat ionic P o r proteoliposomes 2.12.1 B io t i ny l a t i on o f goat anti-mouse I g G In order to measure antibody b inding , b io t in was conjugated to the detecting antibody as described below. Goat anti-mouse I g G (5 mg) was resuspended i n 100 u l P B S . The antibody was then passed d o w n 1 m l Sephadex G - 5 0 spin columns hydrated in P B S that had been pre-spun at 1000 rpm for 3 minutes in a Labofuge 400 table top centrifuge (Heraeus, Germany) . F o l l o w i n g centrifugation, 1 u l o f freshly prepared N F f S - L C - b i o t i n (156 m M i n dimethyl sulfoxide) was added to the antibody mixture and incubated for 30 minutes at ambient temperature. The sample was then centrifuged d o w n pre-spun 1ml Sephadex G - 5 0 spin columns hydrated in P B S . The final sample was di luted 3-fold and stored at 4 ° C . 2.12.2 E L I S A antibody b ind ing assays Mic ro t i t e r plates (Nunc - Immuno Plate, Denmark) were sensitized by adding 0.1 m l / w e l l o f the purif ied P o r protein f rom the strain M S l l j B A r m p , 2 ug /ml in 0.1 M sodium bicarbonate buffer, p H 9.6. The plates were incubated overnight at 4 ° C . M e a n w h i l e , 66 reconstituted samples were di luted in H B S / B S A and aliquots between 0-150 u l pipetted into E p p e n d o r f tubes and the total vo lume was made up to 150 u l w i t h H B S / B S A . Neisser ia l por in monoclona l antibody was di luted i n H B S / B S A and 0.1 m l was added to each tube. The samples were gently vortexed and incubated overnight at 4 ° C . Af ter 24 hours, sensitized plates were washed once w i t h P B S and then were b locked w i t h b l o c k i n g buffer ( P B S / B S A ) for 30 minutes at room temperature. The b l o c k i n g buffer was removed and 0.1 m l o f each p r o t e o l i p o s o m e / M A b incubat ion mixture was plated ( in duplicate). A 100 u l al iquot o f H B S / B S A was added to each w e l l to make the total vo lume to 200 ul . Plates were incubated overnight at 4 ° C . F o l l o w i n g the proteol iposome plating, plates were washed w i t h P B S and then were b locked w i t h P B S / B S A fo l l owed by the addi t ion o f 200 u l / w e l l o f a 1/1000 di lu t ion o f biot inylated goat anti-mouse I g G and incubated for 1 hour at ambient temperature. F o l l o w i n g reaction w i t h biotin-conjugated immunog lobu l in , plates were washed and b locked and then 200 u l /we l l o f a 1/10000 d i lu t ion o f streptavidin horse radish peroxidase were added to the plates and incubated for 60 minutes at ambient temperature. Plates were washed three t imes w i t h P B S and b locked w i t h b l o c k i n g buffer and 0.2 m l / w e l l o f o-phenylenediamine in 100 m M N a C l , 50 m M citrate, p H 5.0 was added. C o l o r was developed and reaction was terminated w i t h the addi t ion o f 50 u l / w e l l o f 2 M H C 1 . The absorbance was measured at 490 n m on a 96 w e l l microt i ter plate reader ( B i o T e k Instruments, W i n o o s k i , V T ) . 67 2.13 7/7 vivo antigenicity of Por proteoliposomes 2.13.1 Immunization B A L B / c mice (females; 8 weeks old) were immunized intraperitoneally (200 pi) and intradermally (30 ul) with neutral, anionic and cationic proteoliposome formulations at an antigen dose of 1 pg Por. Injections were made at 0, 3 and 7 weeks and blood was collected via the tail vein at 2, 5 and 9 weeks. Samples were centrifiiged and sera were collected and analyzed by ELISA assays (see below). 2.13.2 ELISA antibody titer assays Microtiter plates (Nunc-Immuno Plate, Denmark) were sensitized by adding 0.1 ml/well of the purified Por protein from the strain MSI IJBAT/W/?, 2 pg/ml in 0.1 M sodium bicarbonate buffer, pH 9.6. The plates were incubated overnight at 4 °C. After 24 hours, sensitized plates were washed once with PBS and then were blocked with blocking buffer (PBS/BSA) for 30 minutes at room temperature. After the plates were blocked, sera from immunized mice were serially diluted in HBS/BSA and added to Por-sensitized plates and incubated overnight at 4°C. After the incubation, plates were washed with PBS and then were blocked with PBS/BSA followed by the addition of 200 pl/well of a 1/1000 dilution of biotinylated goat anti-mouse IgG and incubated for 1 hour at ambient temperature. Following reaction with biotin-conjugated antibody, plates were washed and blocked and then 200 pl/well of a 1/10000 dilution of streptavidin horse radish peroxidase were added to the plates and incubated for 60 minutes at ambient temperature. Plates were washed three times with PBS and blocked with blocking buffer and 0.2 ml/well of o-phenylenediamine in 100 mM NaCl, 50 mM citrate, pH 5.0 was added. Color was developed and the reaction was 68 terminated wi th the addi t ion o f 50 u l / w e l l o f 2 M H C 1 . The absorbance was measured at 490 n m on a 96 w e l l microti ter plate reader. 2.13.3 E L I S A antibody isotyping assays Mic ro t i t e r plates (Nunc - Immuno Plate, Denmark) were sensitized by adding 0.1 m l / w e l l o f the purif ied P o r protein f rom the strain MS11JBA /7W£>, 2 ug /ml in 0.1 M sodium bicarbonate buffer, p H 9.6. The plates were incubated overnight at 4 ° C . Af ter 24 hours, sensitized plates were washed once w i t h P B S and then were b locked w i t h b l o c k i n g buffer ( P B S / B S A ) for 30 minutes at r oom temperature. After the plates were b locked , sera were diluted and plated on Por-sensi t ized plates and incubated overnight at 4 ° C . Plates were washed wi th P B S and were then b locked w i t h P B S / B S A f o l l o w e d by 1 hour incubat ion w i t h 200 u l /we l l o f rabbit anti-mouse I g G (specific for isotypes G I and G2a) . Th is reaction was fo l lowed by a 1 hour incubat ion w i t h 200 u l / w e l l o f 1/1000 d i lu t ion o f b iot inyla ted mouse anti-rabbit I g G . Plate incubations were performed at r oom temperature. F o l l o w i n g reaction w i t h biotin-conjugated immunog lobu l in , a l l plates were washed and b locked as before and then 200 u l / w e l l o f a 1/10000 di lu t ion o f streptavidin horse radish peroxidase were added to the plates and incubated for 60 minutes at ambient temperature. Plates were washed three times w i t h P B S and b locked w i t h b l o c k i n g buffer and 0.2 m l / w e l l o f o-phenylenediamine i n 100 m M N a C l , 50 m M citrate, p H 5.0 was added. C o l o r was developed and reaction was terminated wi th the addi t ion o f 50 u l / w e l l o f 2 M H C 1 . The absorbance was measured at 490 n m on a 96 w e l l microti ter plate reader. 69 2.14 Tissue his tology M i c e showing side effects due to the P o r proteoliposomes exhibi ted inf lammat ion at the site o f injection. Therefore, the tissue at the injection site was excised and f ixed w i t h formalin . W h e n the tissue had been fixed, the fixative was washed out and the tissue was then dehydrated and embedded in paraffin. The hardened paraffin was mounted o n a micro tome and cut into five m i c r o n slices. The thin sections were then placed on slides and paraffin was removed f rom the tissue w i t h xylene . W h e n the paraffin was d issolved out, the tissue was re-hydrated, stained w i t h hematoxy l in and eosin ( H & E stain) and again dehydrated. The sample was permanently fixed in place w i t h a mount ing med ium and covered w i t h a coversl ip. The tissue samples were then v i e w e d in normal bright field mode w i t h a Zeiss A x i o p h o t microscope (West Germany) . 2.15 Statistical methods The statistical s ignif icance o f the antibody titer data was determined by a single factor A N O V A fo l lowed by a post hoc Scheffe's test for mul t ip le comparisons. The antibody isotype data were analyzed w i t h the Student Mest. A P value o f less than 0.05 was considered to be statistically significant. 70 CHAPTER 3 FACTORS INFLUENCING PROTEIN INCORPORATION INTO LIPOSOMES Factors inf luencing protein incorporat ion into l iposomes are addressed in this chapter. Compar i son was made o f protein reconsti tution eff iciency dur ing the preparation o f proteoliposomes f rom a mixture o f membrane proteins ( M O M P ) or f rom a single protein (Por). Add i t iona l ly , the rate o f detergent removal and protein incorporat ion eff iciency were examined for different detergents. F ina l ly , the effects o f protein- to- l ipid ratio, reconstitution t ime and protein species on incorporat ion eff iciency are outl ined. Character izat ion o f the proteoliposomes is covered in Chapter 4. 3.1 INTRODUCTION The technique o f protein reconsti tution has been useful i n s tudying var ious membrane proteins. Studies have investigated the functions o f membrane receptor proteins such as benzodiazepine and insu l in receptors by so lub i l i za t ion and subsequent incorporat ion into art if icial l i p i d bi layers (Anho l t et al, 1986 and Tranum-Jensen et al, 1994). In addit ion, various membrane-bound enzymes have been reconstituted into l iposomes w i t h g o o d recovery o f enzyme act ivi ty (Anhol t , 1988 and Dr iessen and W i c k n e r , 1990). B y isola t ion o f ind iv idua l proteins into planar bi layers, the structure, function and orientation o f specific proteins in their native, membrane-bound state can be determined. Isolation and reconstitution o f i nd iv idua l membrane proteins is useful i n the reconstitution o f antigens in potential vacc ine preparations. Convent iona l vaccines use attentuated or k i l l ed who le organisms that may cause m i l d to severe reactions (Lauter ia et al, 71 1974, B a r k i n and P ich ichero , 1979 and Ro i t t et al, 1996). Th i s p roblem has prompted the development o f subunit and peptide vaccines that are based on single, conserved antigens. Unfortunately, these vaccines are often poor immunogens for p rov id ing protective i m m u n i t y i n the absence o f an adjuvant (Richards et al, 1996 and W e t z l e r et al, 1992). The use o f l iposomes as potential immunoadjuvants and vaccine carriers has been extensively reviewed ( A l v i n g , 1987, Gregoriadis , 1990 and A l v i n g , 1991). The incorporat ion o f purif ied antigens into l iposomes has been shown to restore the immunogenic i ty o f the antigen. L o c a l i z a t i o n o f the antigen into the l iposomal membrane a l lows for surface presentation o f antigen to B cells for induct ion o f humoral immuni ty and to antigen presenting cells for cel l -mediated immuni ty . L i p o s o m e s have the abi l i ty to elici t both a cel lular-mediated immune response ( G a r c o n and S ix , 1991) and a humoral immune response (Ther ien et al, 1990). Studies have shown l iposomes to be effective immunopotentiators in hepatitis A (Just et al, 1992 and G l i i c k et al, 1992) and inf luenza vaccines (G l i i ck , 1992 and Stahn et al, 1992). In addition, l iposomes have adjuvant act ivi ty in vaccines against protozoan (Whi te et al, 1993) and bacterial organisms (Mut t i l a inen et al, 1995). In addi t ion to Neisseria gonorrhoeae, another Neisse r ia l organism for w h i c h a vaccine is currently being sought is Neisseria meningitidis. M e n i n g i t i s (meningococcal) disease arises from various serogroups and varies f rom country to country. A large proport ion o f this disease is caused by strains A , B , C , Y and W 1 3 5 . H i g h l y effective capsular polysaccharide vaccines against strains A , C , Y and W 1 3 5 have been developed. Unfortunately, group B capsular polysacchar ide vaccines are found to be essentially nonimmunogenic . The fact that serogroup B , the predominant cause o f meningococca l 72 disease in many temperate countries, lacks polysacchar ide immunogen ic i ty has prompted the research and development o f alternative vaccines based on membrane proteins. Studies have shown adjuvant-protein vaccines to have markedly improved antibody responses ( W a n g and Frasch, 1984 and Frasch et al, 1987). Furthermore, proteosome-based vaccines have been shown to be immunogenic (Ruegg et al., 1990) and outer membrane vaccines can increase immuni ty to group B meningococc i (Zo l l inger et al, 1987, Rosenqvis t et al, 1988 and Zo l l i nge r and M o r a n , 1991). Based on these studies, the reconstitution o f a mixture o f meningococca l membrane proteins into l iposomes was characterized. The development o f a potential l iposome vaccine requires that the condit ions for opt imal protein incorporat ion be determined. Therefore, the research in the present chapter deals w i t h the characterization o f the process o f detergent-mediated protein reconsti tution and addresses the factors inf luencing protein incorporat ion into l iposomes. In particular, the kinetics o f detergent removal and residual detergent levels i n the reconstituted proteol iposome sample must be evaluated. The detergent concentration in the sample has relevance w i t h regard to sample stability and vaccine safety. In addit ion, the characterization o f protein incorporat ion during dialysis and the effect o f protein- to- l ipid ratio are essential for op t imiz ing the condit ions for efficient protein incorporat ion. The protein incorporat ion efficiencies during the preparation o f proteol iposomes f rom a mixture o f membrane proteins ( M O M P ) and a single protein (Por) were examined. 73 3.2 R E S U L T S 3.2.1 Detergent removal dur ing reconsti tution o f P o r protein The non- ionic surfactant oc ty l g lucopyranos ide ( O G P ) was selected for these studies because it is reported to be a relat ively m i l d , non-denaturing detergent and has a relat ively h igh cr i t ical mice l la r concentration ( C M C ) o f 21 m M ( G o u l d et al, 1981). The rate o f removal o f O G P (ini t ial concentration 400 m M ) was moni tored by assaying for [ I 4 C ] - O G P during dialysis . A s shown in F igure 3.1, detergent removal is rapid, part icularly dur ing the first 16 hours, w h i c h l i ke ly represents the removal o f monomer ic or mice l la r detergent. W h e n P o r proteoliposome formation has occurred, it is l i ke ly that subsequent O G P removal f rom the vesicles w i l l be constrained by the rate o f O G P " f l i p - f l o p " f rom the inner monolayer to the outer monolayer. After 36 hours o f dialysis , however , O G P remaining i n the P O P C and P O P C P O P E systems was 1.5% and 3 .3% o f the starting concentration, respectively. This corresponds to 6.0 m M and 13.3 m M O G P , or 9% and 2 0 % relative to the l i p i d concentration, respectively. These concentrations are w e l l be low the C M C o f 21 m M . R e m o v a l o f the ionic surfactant, cholate, during P o r protein reconsti tution was also determined. Interestingly, cholate removal was m u c h s lower than the removal o f O G P (Figure 3.2). O n l y after about 50 hours o f d ia lys is is the residual cholate concentration at or be low the C M C , w h i c h is reported to be 13-16 m M (Chattopadhyay and L o n d o n , 1984 and Z h a n g et al, 1996). After 140 hours o f dialysis , residual cholate in the reconsti tution mixture was 1.3% o f ini t ia l , w h i c h corresponds to 2.6 m M , w e l l be low its C M C . In contrast, O G P levels are w e l l be low the C M C after only 20 hours o f dialysis . 74 0 . 4 5 Time (hours) Figure 3.1: N-Octyl- (3-D-glucopyranoside levels dur ing dialysis . Res idua l O G P was moni tored by assaying for 1 4 C - O G P during the dia lys is procedure. S h o w n are detergent levels for P O P C proteoliposomes ( • ) and P O P C : P O P E pro teo l iposomes(T) . 75 Figure 3.2: Sod ium cholate levels dur ing dia lys is . The levels o f sod ium cholate were measured by assaying for [ 3 H]-cho l i c ac id dur ing the dialysis procedure. 76 3.2.2 Influence o f P o r protein/ l ipid ratio o n reconstitution eff iciency The influence o f prote in / l ip id ratio on eff iciency o f protein incorporat ion into l iposomes dur ing reconstitution was examined. Pro teol iposome samples containing gonococca l protein I (Por) and P O P C were prepared from O G P (See section 2.2) at in i t ia l protein/ l ipid ratios (wt/wt) o f either 0.02 or 0.01. Samples taken pr ior to dialysis , at both protein/ l ipid ratios, show a broad band in the top ha l f o f the gradient (Figures 3 . 3 A and 3 .4A) . A large proport ion o f the protein is seen to migrate to the bot tom o f the gradient. F o l l o w i n g dialysis for 20 hours, a prote in/ l ip id band extending over a relat ively narrow density range (gradient depth 3-5 ml) is observed. H o w e v e r , there is st i l l a significant proport ion o f free P o r protein, w h i c h migrates to the bot tom o f the gradient. Th i s represents 3 0 % and 5 0 % o f total protein for P / L ratios o f 0.01:1 and 0.02:1, respectively. Th i s observation indicates protein incorporat ion is incomplete after 20 hours. Af ter 125 hours o f dialysis , a further narrowing o f the phosphol ip id band is seen w i t h essentially a l l o f the P o r protein associated w i t h this l i p i d band at the lower protein- to- l ipid ratio (0.01:1) (Figure 3 .3C) . It is important to note, however, that at the higher protein- to- l ipid ratio (0.02:1), approximately 17% o f P o r protein remains un-associated w i t h l i p i d and migrates to the bottom o f the gradient (Figure 3 .4C) . Th i s suggests that protein incorporat ion is saturable and l imi ted by the phosphol ip id concentration. Prote in reconstitution was also characterized us ing the ion ic surfactant, cholate. Isopycnic density gradient profiles for P o r / P O P C (P/L=0.01) mixtures reconstituted for either 0, 60 or 125 hours from cholate are shown in F igure 3.5. The in i t ia l sample (time zero) shows the l i p i d band dispersed over a broad density range w i t h i n the top ha l f o f the gradient, 77 Figure 3.3: OGP-med ia t ed P o r reconstitution into P O P C l iposomes (P/L=0.01) . I sopycnic density gradient centrifugation profiles for gonococca l Pro te in I l iposomes reconstituted from O G P : A ) 0 hours, B ) 20 hours and C ) 125 hours. Samples were centrifuged at 110 ,000g a v at 4 ° C for 24 hours on a continuous F i c o l l gradient (0-10%). Pro te in (f i l led circles) . Phospho l ip id (open circles) . 0 1 2 3 4 5 6 7 8 9 10 11 12 13 Gradient Depth (ml) 79 F i g u r e 3.4: OGP-med ia t ed P o r reconstitution into P O P C l iposomes (P/L=0.02) . Isopycnic density gradient centrifugation profiles for gonococca l Pro te in I l iposomes reconstituted from O G P : A ) 0 hours, B ) 20 hours and C ) 125 hours. Samples were centrifi iged at 110 ,000g a v at 4 ° C for 24 hours on a continuous F i c o l l gradient (0-10%). Prote in ( f i l led circles) . Phospho l ip id (open circles). 81 Figure 3.5: Cholate-mediated P o r reconsti tution into P O P C l iposomes (P/L=0.01) . Isopycnic density gradient centrifugation profiles for gonococca l Prote in I l iposomes reconstituted from cholate: A ) 0 hours, B ) 60 hours and C ) 125 hours. Samples were centrifuged at 110 ,000g a v at 4 ° C for 24 hours on a continuous F i c o l l gradient (0-10%). Protein ( f i l led circles) . Phospho l ip id (open circles) . 83 whereas P o r protein is observed to migrate near the bot tom o f the gradient. Af ter 60 and 125 hours o f dialysis , there are increases, 7 0 % and 77%, respectively, in protein incorporat ion as the l i p i d and protein are observed to co-migrate d o w n the gradient. Howeve r , for this protein/ l ipid ratio there is a significant fraction (approximately 23%) o f unincorporated Por , in contrast to proteoliposomes prepared f rom O G P at the same prote in / l ip id ratio (Figure 3.3). These results indicate that P o r protein incorporat ion is less efficient dur ing reconstitution f rom cholate. 3.2.3 Proteol iposome size Reconst i tuted systems were examined us ing quasi-elastic l ight scattering analysis ( Q E L S ) to determine ves ic le size distributions. Reconst i tu t ion f rom O G P resulted in proteoliposomes exhib i t ing a relat ively broad size dis tr ibut ion w i t h mean ves ic le diameter about 317 n m and standard deviat ion o f 180 n m (Table 3.1). It has been reported that vesicles reconstituted f rom cholate have mean diameters o f about 50 nm, depending on l i p i d composi t ion ( M a d d e n et al. 1983 and M a d d e n , 1986). Ves i c l e s o f this size w o u l d be ideal for sterile fil tration o f a potential vacc ine candidate pr ior to use in a c l i n i ca l setting. It was observed, however, that reconstitution o f P o r proteoliposomes f rom cholate resulted in systems w i t h mean ves ic le diameter greater than 500 n m exh ib i t ing a C h i 2 o f 5 and thus w o u l d have to be size-reduced pr ior to steri l izat ion by terminal filtration through 0.2 m i c r o n filters. This C h i 2 value is an indicator o f the goodness-of-fit between the actual size distr ibution and a calculated Gauss ian distr ibution. A C h i 2 va lue o f less than 2 w o u l d indicate good agreement o f the raw data to a Gauss ian fit. The observed C h i 2 o f 5, therefore, suggests that the distr ibution profile may be skewed towards larger or smaller ves ic le sizes or 84 Reconst i tu t ion M e a n V e s i c l e Diamete r (nm) Standard D e v i a t i o n Detergent (nm) O G P 317 180 Chola te 543 277 O G P / E x t r u d e d Ves ic les 96 21 Table 3.1: Q E L S Size analysis o f P o r proteoliposomes reconstituted from O G P and cholate. 85 a b imoda l dis tr ibut ion may exist. P o r proteoliposomes are further characterized by C T E M in the next chapter to examine morpho logy and size distributions. 3.2.4 Reconst i tut ion o f meningococca l outer membrane protein ( M O M P ) : Res idua l detergent levels. Reconst i tu t ion experiments were conducted us ing two different detergent preparations to determine the effect o f detergent properties on proteol iposome formation. The surfactant solutions under invest igat ion were octy l g lucos ide and a combina t ion o f sod ium cholate and E m p i g e n B B . Differences in ion ic character and cr i t ical mice l la r concentrations o f these detergents may influence membrane protein incorporat ion into l iposomes. The effects o f these detergents on protein incorporat ion efficiency and residual detergent levels dur ing the dialysis were examined. In previous reconstitution experiments, O G P removal dur ing gonococca l protein reconstitution was quite rapid and residual detergent levels were found to be w e l l be low the cr i t ical mice l le concentration. R e m o v a l o f O G P dur ing meningococca l protein reconstitution i n the presence and absence o f S M - 2 Biobeads (B io rad) was moni tored by assaying for [ 1 4 C ] -O G P during dialysis . A s shown i n F igure 3.6, detergent removal f rom the reconstitution mixture is rapid, part icularly dur ing the first 15 hours. It was anticipated that Biobeads , polystyrene polymers , when added to the external buffer, w o u l d b ind to detergent molecules that have already been removed f rom the reconsti tution mixture. Detergent molecules b ind ing to the Biobeads in the external buffer w o u l d further increase the concentration gradient and lead to an increase in the rate o f O G P removal f rom the reconstitution mixture. However , the rate o f removal was s imi la r w i t h and wi thout the addit ion o f Biobeads . Th is 86 0 20 40 60 80 100 120 140 Time (hours) Figure 3.6: O G P levels dur ing M O M P reconstitution. The levels o f N-oc ty l - f3 -D-g lucopyranos ide (OGP) were measured by assaying for [ 1 4 C ] - O G P during the dia lys is process. O G P levels in absence o f S M - 2 Biobeads (open circles) and in presence o f S M - 2 Biobeads (open triangles). 87 suggests that either the rate l i m i t i n g step i n O G P removal was not the concentration gradient across the dialysis membrane or that the Biobeads d id not b ind O G P to any great extent. W h e n proteol iposome formation has occurred, it is l i k e l y that addit ional O G P removal from the vesicles w i l l be determined by the rate o f O G P exchange from the inner monolayer to the outer monolayer. After 140 hours o f dialysis , O G P remain ing i n both samples was 0 .25% o f the ini t ia l concentration. This corresponds to 1 m M O G P w h i c h is w e l l be low the C M C o f approximately 21 m M ( G o u l d et al, 1981). The kinetics o f O G P removal dur ing meningococca l protein reconstitution fo l lowed very c lose ly the rate observed dur ing gonococca l protein reconstitution. 3.2.5 Character izat ion o f M O M P incorporat ion dur ing reconsti tution H a v i n g determined the rates o f removal o f O G P , the reconsti tution process was characterized by moni to r ing protein incorporat ion into l iposomes as the detergent was d ia lyzed away. M e n i n g o c o c c a l outer membrane proteins ( M O M P ) and P O P C were dissolved i n O G P and d ia lyzed for 20 and 125 hours (as described under Methods) . A n aliquot o f the or ig inal sample was also retained. I sopycnic density gradient centrifugation was then used to monitor vesicle format ion and protein incorporat ion dur ing the reconstitution process. The in i t ia l sample (time zero) shows a broad distr ibution o f phosphol ip id i n the top ha l f o f the gradient (Figure 3.7). Some M O M P is also found associated w i t h the l i p i d band, l i ke ly due to spontaneous ves icula t ion w h e n the sample is di luted on the gradient. F o r this in i t ia l sample, the remaining protein is seen to migrate as a peak near the bot tom o f the gradient. Af ter 20 or 125 hours o f dialysis , however, it can be seen that co-migrat ion o f phosphol ip id and protein occurs to a posi t ion approximately a third 88 Figure 3 . 7 : OGP-med ia t ed M O M P reconsti tution into P O P C l iposomes. Isopycnic density gradient centrifugation profiles for meningococca l O M P l iposomes reconstituted f rom O G P : A ) 0 hours, B ) 20 hours and C ) 125 hours. Samples were centrifuged at 110 ,000g a v at 4 ° C for 24 hours on a continuous F i c o l l gradient (0-10%). Protein ( f i l led circles) . Phospho l ip id (open circles). 0 1 2 3 4 5 6 7 8 9 10 11 12 13 Gradient Depth (ml) 90 o f the w a y d o w n the gradient. Furthermore, the prote in / l ip id band extends over a relat ively narrow density range. A p p r o x i m a t e l y 8 0 % o f the protein associated w i t h the l i p i d , w i t h 2 0 % free protein at the bot tom o f the gradient. Th is profi le indicates a h igh M O M P incorporat ion efficiency dur ing the reconstitution procedure, but incomplete compared to O G P - m e d i a t e d gonococca l protein incorporat ion at the same prote in / l ip id ratio. 3.2.6 E m p i g e n B B - m e d i a t e d M O M P reconsti tution into l iposomes Prote in isola t ion studies o f meningococca l membrane proteins have indicated that inc lus ion o f a zwit ter ionic detergent during protein so lub i l i za t ion and pur i f icat ion can restore the antibody b ind ing capacity o f membrane proteins ( M a n d r e l l and Zo l l inge r , 1984). Therefore, E m p i g e n B B , a c o m m o n l y used zwi t te r ion ic detergent, was used i n meningococca l protein reconstitution experiments and the results were then compared to reconstitution w i t h oc ty l glucoside. F o l l o w i n g d ia lys is for 12 days, the prote in- l ip id mixtures were observed to be transparent, suggesting that either any vesicles present were very smal l , or that no vesicula t ion had occurred. Dens i ty gradient centrifugation showed little or no protein associated w i t h the phosphol ip id band as the phosphol ip id component was observed to migrate in the top ha l f o f the gradient. M o s t o f the protein was seen i n a peak fraction at the bottom o f the gradient (Figure 3.8). The separation o f the t w o peaks indicates that ves ic le formation d id not occur during dia lys is ; therefore, protein incorporat ion cou ld not occur. 91 Figure 3 . 8 : E m p i g e n BB/cho la te -media ted M O M P reconstitution. Isopycnic density gradient centrifugation profile for meningococca l O M P l iposomes reconstituted from E m p i g e n B B / c h o l a t e . Sample as centrifuged at 110 ,000g a v at 4 ° C for 24 hours on a continuous F i c o l l gradient (0-10%). Protein ( f i l led circles) . Phospho l ip id (open circles) . 92 3.3 D I S C U S S I O N The subject o f phosphol ip id and protein so lub i l i za t ion and subsequent proteol iposome reconsti tution has been rev iewed extensively (Helenius and Simons, 1975, A l l e n et al, 1980, Lich tenberg et al, 1983, Hje lmeland , 1990 and M a d d e n , 1988). Studies o f detergents and l iposomes have suggested a three-stage mode l o f detergent-liposome interaction (Paternostre et al, 1988 and A n g r a n d et al, 1997). D u r i n g detergent-induced solubl izat ion, the first stage involves the par t i t ioning o f non-mice l la r detergent into the l iposomal b i layer resulting i n increased membrane permeabi l i ty wi thout solubi l iza t ion. Stage two corresponds to the gradual disrupt ion o f the b i layer and the emergence o f detergent-lipid m i x e d micel les and detergent saturation o f the l iposomes. The final stage constitutes the complete so lubi l iza t ion and convers ion o f l iposomes into detergent-lipid micel les . It is assumed that detergent-mediated reconstitution o f l iposomes or proteoliposomes fo l lows the reverse o f the so lub i l i za t ion process (R igaud et al, 1988). Ini t ia l ly , lipid/detergent and/or lipid/protein/detergent are present as m i x e d micel les . A s detergent is removed, m i x e d mice l les become unstable leading to the format ion o f structures composed o f l ip ids and proteins that eventually fo rm vesicles. In the final phase, residual intrabilayer detergent is removed. The incorporat ion o f protein into the b i layer is thought to occur v i a two possible mechanisms. Ei ther l i p i d vesic les are formed and the protein then inserts into these preformed vesicles. Al te rna t ive ly , protein may become incorporated into the bi layer during the ves icula t ion process. Reconst i tu t ion studies have indicated that the presence o f m i x e d mice l les is required for protein incorporat ion to occur (Rigaud et al, 1988). In addition, these studies have suggested that for in i t ia l detergent concentrations above the cr i t ical mice l la r concentration, complete protein incorporat ion cou ld be obtained. 93 Researchers have suggested that the compos i t ion o f result ing proteol iposomes w i l l be determined by the relative in i t ia l concentrations o f l i p i d , protein and detergent ( M i m m s et al, 1981). These experiments indicated that in i t ia l detergent concentrations need to be 15-fold higher than the C M C for complete so lubi l iza t ion o f the phosphol ip id . F o l l o w i n g detergent dialysis , complete ves icula t ion o f the phosphol ip id was observed. W h e n the detergent/l ipid molar ratio prior to dialysis was 5:1, the m i n i m u m concentration required for obtaining a clear solut ion, only about 5 0 % o f the l i p i d was found to be in vesicular form, whereas the remainder was i n non-vesicular fo rm ( M i m m s et al., 1981). Ini t ial detergent-to-lipid and in i t ia l protein- to- l ipid ratio may affect the protein incorporat ion efficiency. The present study compares protein reconsti tution into l iposomes generated f rom a non- ionic or ionic detergent. The detergents used, oc ty l g lucopyranoside ( O G P ) and sodium cholate, have been employed i n either protein i so la t ion and/or reconstitution studies (Paternostre et al, 1988, R i g a u d et al, 1988, and A n g r a n d et al, 1997). O G P is a non- ionic detergent that has previously been used in the reconsti tut ion o f cy tochrome-P450 membrane proteins (Schwarz et al., 1984). One advantage to us ing this detergent is that it has a relat ively h igh cr i t ical mice l la r concentration (21 m M ) and can be rapidly removed during the dialysis procedure ( G o u l d et al, 1981). A s the present study illustrates, complete P o r protein incorporat ion into l iposomes can be achieved by O G P - m e d i a t e d reconstitution, as has been reported for other proteins (R igaud et al, 1988 and A n g r a n d et al, 1997). Ana lyses o f reconstituted proteoliposomes have shown residual detergent levels to be w e l l be low the C M C , levels that are non-membrane ly t ic (Angrand etal, 1997). S o d i u m cholate, an ionic detergent, has also been used in the reconsti tution o f membrane proteins into l iposomes ( M a d d e n et al, 1983 and R i g a u d et al, 1988). Studies 94 report that proteoliposomes generated by cholate-mediated reconstitution tend to be smaller i n diameter than vesic les generated from O G P (Schwarz et al, 1984). Howeve r , it was observed that Por-conta in ing vesicles generated from cholate were as large or larger than vesicles prepared from O G P . In addition, the rate o f removal o f cholate was much s lower than that o f O G P . Furthermore, cholate-mediated reconstitution resulted i n lower protein incorporat ion efficiencies than for O G P - m e d i a t e d reconstitution. The differences between these two detergents may be attributed to two factors. Firs t , the anionic charge on the cholate may be invo lved in an electrostatic interaction w i t h protein residues. Th i s interaction cou ld have resulted i n some protein denaturation, thus preventing complete incorporat ion o f the gonococca l membrane protein. Second, the steroid-l ike structure, l ower C M C and negative charge on cholate may retard or substantially reduce the rate o f dissociat ion o f detergent molecules f rom the l ip id/prote in complexes , result ing in s lower detergent removal . Studies suggest that the presence o f polar h y d r o x y l groups o f cholate i n the hydrophobic core o f l iposomes and its lower C M C may play a role in the t ime required to generate proteoliposomes (R igaud et al, 1988 and Lich tenberg , 1985). S l o w i n g the reconsti tution process may favor protein denaturation and/or aggregation and hence reduce the eff ic iency o f protein incorporation. A n a l y s i s o f the detergent-mediated reconstitution experiments presented in this paper indicates octy l g lucos ide to be the detergent o f choice for proteol iposome reconstitution, i n particular, for the incorporat ion o f bacterial membrane proteins into a l iposomal matr ix to be used to produce a subunit vaccine. A s mentioned earlier, it has been suggested that the composi t ion and characteristics o f the reconstituted system w i l l be determined by the relative in i t ia l concentrations o f l i p i d , protein and detergent ( M i m m s et al, 1981). W i t h an ini t ia l detergent concentration b e l o w its 95 C M C , only partial protein incorporat ion was attainable (R igaud et al, 1988). Comple te protein incorporat ion into l iposomes was achieved w h e n in i t ia l detergent levels were above the C M C Studies i n v o l v i n g the incorporat ion o f the cy tochrome-P450 enzyme system into l iposomes us ing O G P have reported the efficiency o f protein incorporat ion to be dependent on the ini t ia l prote in- to- l ipid ratio (Schwarz et al, 1984). Th is is consistent w i t h the present results w h i c h indicate that protein incorporat ion is saturable for a g iven l i p i d concentration. A s observed, ini t ia l protein- to- l ipid ratios above this saturation l imi t result i n incomplete protein incorporat ion w i t h consequent denaturation and aggregation o f non-incorporated protein. F o r a subunit vaccine preparation to be most effective, it w o u l d be advantageous to have m a x i m u m presentation o f the antigenic determinants on the surface o f the l iposome. In addition, complete antigen incorporat ion into l iposomes may reduce tox ic i ty and increase immunogenic i ty o f the antigen, subsequently y i e l d i n g a vaccine w i t h increased potency and efficacy. O G P - m e d i a t e d reconstitution o f men ingococca l outer membrane proteins was observed to be efficient w i th a large propor t ion o f protein associated wi th the phosphol ip id as determined by i sopycnic density gradient centrifugation. H o w e v e r , M O M P incorporat ion was incomplete at the same P / L ratio o f 0.01 where gonococca l P o r protein was completely incorporated into l iposomes. Th i s may have been due to heterogeneity o f the M O M P sample, as it consisted o f a mixture o f proteins ranging in molecular weight f rom 25-100 k D a (data not shown). Aggrega t ion and denaturation o f certain protein molecules may have prevented incorporat ion o f the approximately 2 0 % protein observed at the bot tom o f the gradient Therefore, a single protein w o u l d be preferred i n a potential subunit vacc ine formulat ion. 96 In addit ion to O G P , another detergent mixture that was employed in M O M P proteol iposome reconstitution was E m p i g e n B B w i t h sod ium cholate. E m p i g e n B B is a relat ively m i l d , zwi t ter ionic detergent that has been c o m m o n l y used in protein so lubi l iza t ion and purif icat ion studies (Lowther t et al, 1995 and M u k h l i s et al, 1986). M a n y detergents have been used in protein isolat ion experiments; however, protein pur i f icat ion often results i n loss o f the antigenic determinants and thus the antigenicity o f the protein (Chris t ie et al, 1988). E m p i g e n BB-ex t r ac t ed antigen preparations exhibi t an abi l i ty to elici t greater antibody responses than preparations extracted w i t h various other detergents (Jennings et al, 1988 and Jennings and Er turk , 1990). These studies indicate E m p i g e n B B to be mi lder and less denaturing on isolated proteins a l l o w i n g for the retention o f antigenic activi ty. Based on these properties, meningococca l proteol iposomes were reconstituted f rom a mixture o f sodium cholate and E m p i g e n B B to compare these vesicles to proteoliposomes formed f rom O G P . The present study demonstrates, however, that no ves icula t ion or protein incorporat ion occurred during reconstitution f rom a chola te /Empigen B B mixture. Despi te prolonged dialysis , effective removal was . not achieved and hence no protein reconstitution was possible. The C M C for E m p i g e n B B (1.2 m M ) (de la M a z a et al, 1998) is lower than those o f O G P and sodium cholate but this factor alone does riot appear to be sufficient to explain w h y detergent removal was not achieved. One poss ib i l i ty is that, w i th in m i x e d micel les containing phosphol ip id , Por , cholate and E m p i g e n B B , the effective C M C o f each detergent compound is s ignif icantly lower than for the i nd iv idua l pure surfactant. In this regard, it should be noted that the zwit ter ionic character o f E m p i g e n B B is s imi lar to some membrane phosphol ipids ( A l l e n and Humphr ies , 1975). 97 In summary, protein incorporat ion was complete w h e n reconsti tuting proteoliposomes from O G P at a prote in / l ip id ratio o f 0.01. Howeve r , incorporat ion was incomplete at higher P / L ratios above 0.02, indica t ing that protein incorporat ion is saturable and l imi ted for a g iven l i p i d concentration. In contrast, it was observed that protein incorporat ion was inefficient and incomplete during sod ium cholate-mediated reconsti tution at a P / L o f 0.01. Interestingly, reconsti tution studies us ing E m p i g e n B B / c h o l a t e detergent mixture showed that a mixture o f a zwi t te r ionic and ionic detergent cou ld not be removed by dialysis and thus no vesicula t ion or protein incorporat ion occurred during the reconsti tution process. These results indicated that O G P w o u l d be the ideal detergent to use i n subsequent reconstitution experiments. In addit ion, reconsti tution o f a single outer membrane protein, Por , resulted i n greater incorporat ion eff iciency compared to the reconsti tution o f a mixture o f outer membrane proteins, M O M P . 98 CHAPTER 4 INFLUENCE OF LIPID COMPOSITION ON POR RECONSTITUTION AND CHARACTERIZATION OF THE RESULTING PROTEOLIPOSOMES In the previous chapter, factors inf luencing protein incorporat ion were investigated. These factors ranged from detergent properties, reconsti tution t ime, protein- to- l ipid ratios and the nature o f the membrane protein sample, single protein or a mixture o f membrane proteins. F r o m the observations and other considerations (see Sect ion 1.1.1), it is concluded that a single protein w o u l d be preferred in a potential vaccine formulat ion. Therefore, having characterized incorporat ion o f both P o r and M O M P , subsequent studies focussed on P o r proteoliposomes. In this chapter, b iophys ica l and antigenic properties o f a l iposomal gonococca l P o r vaccine are characterized. The objective o f these experiments was to determine the effect o f l i p i d compos i t ion on P o r incorporat ion into l iposomes and the orientation o f P o r i n the l i p i d bi layers . 4.1 INTRODUCTION A s mentioned earlier, Neisseria gonorrhoeae is a mucosa l pathogen that colonizes by compet ing w i t h loca l microf lora for adherence to the mucosal epithelial cells and the organism has adapted several mechanisms for faci l i ta t ing infect ion and avo id ing the host immune response (Br i t igan et al., 1985). Individuals w i t h gonococca l infections produce bactericidal , opsonic antibodies that may poss ib ly protect host cells by inh ib i t ing bacterial attachment at mucosal surfaces or by promot ing phagocytosis and complement-mediated serum k i l l i n g ( W a r d et al, 1978 and V i r j i , 1981). These observations have prompted 99 research o f gonococca l surface structures as possible vacc ine target antigens and have revealed protein I as a potential candidate. Prote in I (Por) is a major, channel- forming protein that orients in a "ha i rp in" fashion w i t h both ends inserted i n the p lasma membrane, w i t h the loop port ion extending out f rom the per ip lasmic surface (Greco et al, 1980 and B l a k e et al., 1981). Studies have shown that protective, bacter icidal antibodies are directed towards, and b ind an epitope located wi th in , the surface-exposed loop region ( V i r j i et al., 1986 and Fletcher et al, 1986). Ear l ie r studies reported incorporat ion o f P o r into l i p i d bi layers and compared this l iposomal formulat ion to previously used vaccines containing the P o r protein (Wetz ler et al, 1988 and Wetz le r et al, 1992) These studies demonstrated that combin ing gonococca l protein I w i t h a luminum phosphate or Freund ' s adjuvant increased the overa l l level o f por in -reactive antibodies, as compared to P o r alone; however, there was a reduced titer o f bactericidal antibodies that were reactive w i t h the gonococca l ce l l surface. The reduct ion i n bactericidal antibody titers was l i ke ly due to inadequate presentation o f the protein loop region as a result o f protein denaturation and aggregation produced by the adjuvants. In contrast, protein I incorporated into l iposomes e l ic i ted the highest titer o f surface-reactive, bactericidal antibodies in the rabbit model . These observations suggest that l iposomes w o u l d be preferred adjuvants in a gonococca l subunit vaccine. It has been shown that l iposomes are removed from ci rcula t ion and accumulate i n organs o f the reticuloendothelial system, such as the lungs, l iver , spleen and bone mar row ( A l l e n and Chonn , 1987), where they are taken up by antigen presenting cells ( A P C ) . Research on macrophages, w h i c h are a major subset A P C , has suggested that the ce l l surface charge may influence particulate phagocytosis (Mutsaers and Papadimi t r iou , 1988). Studies 100 conducted by N a k a n i s h i et al. (1997) showed that posi t ively charged l iposomes were taken up more efficiently by macrophages than neutral or negatively charged carriers. In addit ion, researchers have suggested that antigenic ac t iv i ty may be dependent o n the phys ica l state o f the phosphol ip id moiety (Gomez-Gut ie r rez et al., 1994 and Gomez-Gut i e r r ez et al, 1995). These studies indicate that the polar head group, the electrostatic interactions between the antigen and phosphol ip id , as w e l l as the fatty ac id compos i t ion o f the phosphol ip id may influence the recovery o f the antigenic act ivi ty o f pur i f ied antigens. The early studies conducted by Wetz l e r and coworkers (1988, 1992) examined l i p i d mixtures o f P O P C : P O P E ; therefore, studies were conducted to determine the influence o f l i p i d composi t ion, P O P C alone or P O P C : P O P E , o n incorporat ion and protein orientation. In addition, P o r protein was reconstituted into l iposomes composed o f anionic or cat ionic l ip ids to determine the effect o f l i p i d charge on incorporat ion efficiency. P o r Proteol iposomes were also characterized in terms o f ves ic le size and morphology. Furthermore, the issue o f steri l izat ion o f a l iposomal gonococca l subunit vaccine formulat ion was also addressed. 101 4.2 RESULTS 4.2.1 P o r Prote in reconstitution determined by i sopycn ic density gradient centrifugation Previous studies by W e t z l e r and coworkers (1988, 1992) examined P o r incorporated in l iposomes consis t ing o f mixtures o f P O P C : P O P E , whereas reconsti tution experiments described in the previous chapter were conducted w i t h P O P C alone. Therefore, P o r protein was reconstituted into l iposomes w i t h different l i p i d compos i t ion , P O P C alone or a mixture o f P O P C P O P E , to determine the effect o f l i p i d compos i t ion o n incorporat ion efficiency. F o l l o w i n g reconstitution, proteoliposomes were analyzed by i sopycn ic density gradient centrifugation to separate components w i t h i n a mixture on the basis o f their specific densities. In Figures 4.1 and 4.2 are shown the density gradient profiles obtained for proteoliposomes reconstituted w i t h either P O P C alone or P O P C : P O P E . It can be seen that co-migrat ion o f phosphol ip id and protein occurs to a posi t ion approximate ly m i d w a y d o w n the gradient. Protein incorporat ion was observed to be 90%, w h i c h was associated wi th the l i p i d band, indicat ing efficient incorporat ion dur ing reconstitution. Furthermore, in the case o f systems prepared from P O P C alone, the prote in / l ip id band extends over a relat ively narrow density range indicat ing that the vesicles are h igh ly homogenous w i t h respect to protein- to- l ipid ratio. In the case o f reconstituted systems prepared w i t h P O P C P O P E , however, two narrow bands o f proteol iposomes are observed (Figure 4.2). These t w o bands appear to arise f rom vesicle populat ions o f differ ing protein- to- l ipid ratio. Inspection o f the data presented in F igure 4.2 shows that the larger, denser ves ic le fraction exhibits a protein-to - l ip id ratio ( P / L ) o f 0.026, whereas the l ighter fraction exhibits a P / L o f 0.012. These in i t ia l experiments also established the importance o f r emov ing any insoluble or aggregated 102 0 1 2 3 4 5 6 7 8 9 10 11 12 13 Gradient Depth (ml) Figure 4.1: I sopycnic density gradient centrifugation prof i le for P o r reconstituted in P O P C l iposomes. The proteoliposome sample was centrifuged at 110 ,000g a v at 4°C for 20 hours on a continuous F i c o l l gradient (0-10%). S h o w n are the protein ( • ) and phosphol ip id ( O ) concentrations for each fraction. 103 0 1 2 3 4 5 6 7 8 9 10 11 12 13 Gradient Depth (ml) Figure 4.2: I sopycnic density gradient centrifugation profi le for P o r reconstituted i n P O P C P O P E (1:1) l iposomes. The proteol iposome sample was centrifuged at 110 ,000g a v at 4 ° C for 20 hours on a continuous F i c o l l gradient (0-10%). S h o w n are the protein ( • ) and phosphol ip id ( O ) concentrations for each fraction. 104 Figure 4 .3: Isopycnic density gradient centrifugation profi le for insoluble P o r reconstituted in P O P C P O P E (1:1) l iposomes. In this sample, the P o r protein was not filtered pr ior to reconstitution. The proteol iposome sample was centrifuged at 110 ,000g a v at 4 ° C for 20 hours on a continuous F i c o l l gradient (0-10%). S h o w n are the protein ( • ) and phosphol ip id ( O ) concentrations for each fraction. 105 P o r protein pr ior to reconstitution. A s described i n the M e t h o d s section, some batches o f P o r protein contained smal l amounts o f insoluble material . Th i s denatured or aggregated protein was not incorporated into l iposomes dur ing reconsti tution and could be separated from the proteoliposomes by i sopycn ic density gradient centrifugation. Th is is i l lustrated in F igure 4.3 w h i c h shows the density gradient profi le for reconstituted systems prepared w i t h P O P C : P O P E using a P o r protein batch that had not been subjected to fil tration. A g a i n protein bands are seen associated w i t h the l iposomes, as i n F igure 4.1 and 4.2, but i n addition, protein is found near the bot tom o f the gradient w i t h no associated l i p id . In a l l subsequent studies, therefore, any insoluble material was removed from the O G P - s o l u b i l i z e d P o r (as described under Methods ) pr ior to reconstitution. 4.2.2 P o r orientation in reconstituted proteoliposomes A s indicated earlier, in the bacterial p lasma membrane, P o r is folded i n a "ha i rp in" loop configuration w i t h the loop port ion exposed on the per ip lasmic surface. Ideally, P o r should retain this orientation in the reconstituted proteoliposomes, thereby ensuring that the major antigenic site on the loop domain was exposed. T o evaluate protein orientation, reconstituted systems were incubated w i t h t rypsin or a -chymot ryps in . Cleavage sites for these proteases are located i n the loop domain as discussed i n Sect ion 1.3.1. F o l l o w i n g protease digestion, samples were analyzed by S D S - P A G E . A s shown in F igure 4.4, when O G P - s o l u b i l i z e d P o r is incubated w i t h a - chymot ryps in (lane C ) , the 3 9 K protein is c leaved to produce 2 3 K , 1 7 K and 14 .5K fragments. In contrast, t rypsin cleavage (lane D ) produced 2 9 K , 2 3 K , 2 1 K and 14 .5K fragments. These results correlate w e l l w i t h the protease digestion studies o f purif ied outer membranes reported previously ( B l a k e et al, 1981). In 106 F i g u r e 4.4: Tryps in and a -chymot ryps in cleavage o f detergent-solubil ized and reconstituted P o r protein. In lane A are shown molecular weight standards, phosphorylase b ( 9 7 . 4 K ) , serum a lbumin (66 .2K) , ova lbumin ( 4 5 K ) , carbonic anhydrase ( 3 I K ) , soy bean t ryps in inhib i tor (21 .5K) and ly sozyme (14 .5K) . The remaining lanes represent the f o l l o w i n g : P o r so lubi l ized in O G P alone (lane B ) , P o r in O G P treated wi th a - c h y m o t r y p s i n (lane C ) , P o r in O G P treated wi th t rypsin (lane D ) , Por reconstituted in P O P C l iposomes treated w i t h a -chymotryps in (lane E ) , P o r in P O P C l iposomes treated w i t h t rypsin (lane F ) , P o r reconstituted in P O P C P O P E l iposomes treated wi th a - chymot ryps in (lane G ) , P o r in P O P C P O P E l iposomes treated w i t h t rypsin (lane H ) , a - chymot ryps in alone (lane I), and trypsin alone (lane J). 107 this earlier work , a -chymot ryps in was found to cleave P o r protein I ( 3 4 K ) into 2 3 K and 1 4 K fragments wh i l e t rypsin severed at t w o sites to produce 3 fragments. Ini t ia l ly , t ryps in cleaves the protein into 2 8 K and 1 0 K fragments. The 2 8 K fragment is then further digested to a 2 I K fragment and a smaller undetected fragment. In the present study, however, incubat ion w i t h t rypsin also produced 2 3 K and 14 .5K fragments w h i c h appear to result f rom residual ct-chymotryps in activity. A lpha -chymot ryps in can cleave the 2 9 K fragment generated by trypsin into a 2 3 K fragment and a smaller undetected fragment. W h e n Por , reconstituted into either P O P C or P O P C P O P E vesicles, is incubated w i t h t rypsin or a - chymot ryps in (Figure 4.4 lanes E - H ) cleavage patterns s imi lar to the O G P - s o l u b i l i z e d protein are seen. H o w e v e r , for the reconstituted systems, a port ion o f uncleaved protein can also be seen. This protease-resistant populat ion l i ke ly represents reconstituted P o r that is inward ly oriented, w i t h the loop domain facing the vesicle interior, and hence unavai lable to external protease. T o determine the relative proportions o f outwardly and inward ly oriented P o r in reconstituted proteoliposomes, gels were scanned us ing a laser densitometer to determine the relative amounts o f each peptide band. F r o m the densitometer scan, fragments produced by protease digestion accounted for an average o f 8 3 . 5 % o f the total protein in each o f the protein digests (Table 4.1). These results suggest that over 8 0 % o f reconstituted P o r was outwardly oriented wi th in the l i p i d bi layer w i t h the loop domain exposed on the external surface o f the l iposome as has been described in studies on the bacterial ce l l membrane (B lake et al, 1981). Studies have shown that l iposomes are impermeable to enzymes (Oberholzer et al. 1999). Therefore, the 2 0 % o f the P o r protein that was not degraded by the enzymes may represent a proport ion o f P o r protein that was either inward ly oriented or sequestered to the inner lamellae o f mul t i lamel lar vesicles and thus not susceptible to protease cleavage. 108 Peak Areas (°/ o) B a n d L a n e B L a n e E L a n e F L a n e G L a n e H M o i . W t . 39000 100 15.2 13 19.1 18.6 29000 - 19.5 - 13 23000 - 41.4 25.3 43 25.7 21000 - - 12.9 - 14.9 17000 - 18.8 - 18.3 -14500 - 24.5 29.4 19.6 27.8 Tota l Fragments 0 84.7 87.1 80.9 81.4 % Table 4.1: T ryps in and a -chymot ryps in cleavage o f reconstituted Por-proteol iposomes: Densi tometr ic analysis o f S D S - P A G E gel . S h o w n are the areas under each band as a percentage o f the total for the corresponding lane. 109 4.2.3 Size reduction o f reconstituted P o r proteoliposomes Pharmaceut ical product ion o f a l i posomal P o r vaccine w o u l d be greatly facili tated i f the reconstituted proteoliposomes cou ld be steri l ized by terminal fi l tration. Reconst i tu t ion f rom O G P , however, generates systems o f mean diameter greater than about 500 n m w h i c h therefore cannot be steri l ized by passage through a 0.2 m i c r o n filter. Therefore, experiments were conducted to examine whether smaller vesicles cou ld be generated by extrusion o f the proteoliposomes through polycarbonate filters o f defined size. Proteol iposomes were sequentially extruded through 600, 400, 200 and 100 n m polycarbonate filters and after each extrusion step, vesicle size distr ibution was examined by Q E L S and protein and phosphol ip id recovery also determined. A s shown in F igure 4.5, little or no protein or phosphol ip id loss occurs on extrusion o f proteoliposomes reconstituted from either P O P C or P O P C P O P E . A s expected, however, the mean diameter o f the extruded systems is reduced as filter pore size is decreased (Table 4.2). Proteol iposomes reconstituted from P O P C generally exhibi t s l ight ly smaller mean diameters f o l l o w i n g extrusion compared to systems prepared from P O P C : P O P E . F o l l o w i n g extrusion through 200 or 100 n m pore size filters, for example, P O P C proteoliposomes have mean diameters o f about 134 and 88 nm, respectively, whereas P O P C P O P E proteoliposomes g ive mean diameters o f about 196 and 104. In addition, vesicles produced by extrusion through 200 n m and 100 n m filters exhibi t relat ively nar row size distributions as indicated by the calculated standard deviations (Table 4.2). It should be anticipated that P o r proteoliposomes prepared from P O P C and extruded through either 100 n m or 200 n m filters were then suitable for s ter i l izat ion by terminal filtration. In the case o f proteoliposomes prepared from P O P C P O P E , however, on ly systems extruded through 100 nm filters were smal l enough to be steri l ized through a 0.2 m i c r o n filter. The ves ic le 110 L i p i d F i l t e r Pore Size M e a n Diamete r Standard D e v i a t i o n C o m p o s i t i o n (nm) (nm) (nm) P O P C Ini t ial 494.3 286.0 600 251.1 118.1 400 210.1 70.4 200 133.8 40.0 100 88.5 25.7 P O P E P O P C Ini t ial 975.7 655.3 600 592.9 357.7 400 347.1 161.8 200 195.6 64.0 100 104.0 33.8 Table 4 .2: Size reduction o f reconstituted P o r proteoliposomes by extrusion. I l l Figure 4.5: Recove ry o f protein and phospho l ip id f o l l o w i n g extrusion o f P o r proteoliposomes. Reconst i tuted P O P C ( A ) or P O P C P O P E ( B ) proteoliposomes were extruded through sequentially smaller pore size polycarbonate filters as described under Mater ia ls and Methods . 113 morphologies o f reconstituted and extruded proteoliposomes are discussed later in the next section. 4.2.4 Reconst i tut ion o f P o r into l iposomes composed o f charged l ip ids P o r protein was reconstituted into l iposomes containing charged l ip ids , anionic or cationic, to determine the effect o f l i p i d properties on proteol iposome formation and protein incorporation. The charged l ip ids u t i l i zed were the negatively and pos i t ive ly charged l ip ids P O P S (or P O P G ) and D O D A C , respectively. Differences in ion ic character o f these l i p i d species may influence the P o r protein incorporat ion eff iciency into l iposomes. G o n o c o c c a l protein I (Por) was reconstituted into l iposomes consis t ing o f P O P C and 5, 10 and 2 5 % (by wt.) P O P S , P O P G or D O D A C from O G P (as described under Methods) . A s discussed in Sect ion 4.2.1, reconstituted neutral proteoliposomes were found to have a mean vesicle diameter o f about 500 n m (Table 4.2). A s shown previous ly in F igure 4 .1 , reconstitution o f P O P C l iposomes resulted in the majority (85%) o f the P o r protein be ing associated w i t h the phosphol ip id band and co-migra t ing in the top ha l f o f the gradient. A peak fraction o f free protein (15 %) was observed at the bot tom o f the gradient, indicat ing that there is incomplete protein incorporat ion into P O P C vesicles at this protein- to- l ipid ratio (P/L=0.02) . In studies employ ing the anionic l ipids , P O P S and P O P G , reconstituted systems were found to have diameters o f about 500 n m and standard deviations greater than 150 nm, characteristics w h i c h were s imi lar to that o f vesicles composed solely o f P O P C (Table 4.3). The reconstituted proteoliposomes, containing P O P C alone or a mixture o f P O P C w i t h anionic l ip id , exhibi ted large C h i 2 values (greater than five), indicat ing the ves ic le popula t ion had a mul t i -modal distr ibution. In addit ion, anionic proteoliposomes exhibi ted s imi lar 114 Sample Charged Lipid (%) Mean Vesicle Diameter (nm) Standard Deviation (nm) POPC 0 494.3 286 POPC/POPG 5 459 195 10 473 193 25 434 149 POPC/POPS 5 519 259 10 419 200 25 412 209 POPC/DODAC 5 702 301 10 720 354 25 1180 675 Table 4.3: Q E L S size analysis and protein incorporat ion efficiency o f P o r proteoliposomes reconstituted wi th va ry ing l i p i d composi t ion . 115 gradient profiles to neutral l iposomes; protein and l i p i d were observed to co-migrate in peak fractions spanning a narrow density range w i t h free, unincorporated protein at the gradient bot tom (Figures 4.6 and 4.7). F o r a mixture containing 5% P O P G , reconstitution generated vesicles w i t h lower protein incorporat ion efficiency (75%) compared to that observed for neutral l iposomes. A n increase i n the proport ion o f P O P G (10 or 25%) resulted in a decrease i n the amount o f P o r protein (65%) associated w i t h the l i p i d fraction (Figures 4 . 6 B and 4 .6C) . Reconst i tu t ion o f 5% and 10% P O P S samples y ie lded proteol iposomes w i t h approximately 9 0 % and 8 0 % protein incorporat ion (Figure 4 . 7 A and 4 . 7 B ) , w h i c h was s imi lar to that seen for P O P C proteoliposomes w h e n taking into account a 5% var iab i l i ty between reconstitution preparations. A s shown in F igure 4 . 7 C , an increase i n the P O P S content to 2 5 % resulted i n a decrease (75%) i n the relative amount o f protein incorporated. These profiles indicate that the electrostatic interaction o f the charged l i p i d and charged domains o f the protein may be inf luencing protein structure, subsequently affecting the efficiency o f protein insertion into the bi layer . P o r reconstitution in vesicles conta ining pos i t ive ly charged l i p id , D O D A C , resulted i n signif icantly larger proteoliposomes and these systems displayed very different i sopycn ic density gradient profiles (Table 4.3 and F igure 4.8) f rom those seen w i t h neutral or anionic l i p i d mixtures. It was observed that increasing the relative amount o f D O D A C from 5-25% resulted in an increase in ves ic le diameter f rom approximate ly 700 n m to about l u M w i t h standard deviations ranging from about 300 to 700 nm. The ves ic le populations were also observed to exhibit a mul t i -modal size dis tr ibut ion as suggested by a C h i 2 greater than five. Reconst i tuted proteoliposomes containing 5% D O D A C exhibi ted complete protein incorporat ion as indicated by co-migra t ion o f the protein and l i p i d fractions and the absence 116 Figure 4 .6: P o r protein incorporat ion into P O P C / P O P G l iposomes. Isopycnic density gradient centrifugation profile for gonococca l protein I incorporated into P O P C l iposomes containing va ry ing amounts o f P O P G : A ) 5%, B ) 10 % and C ) 2 5 % . Samples were centrifuged at 110 ,000g a v at 4 ° C for 24 hours on a continuous F i c o l l gradient (0-10%). Prote in (circles). Phospho l ip id (triangles). 118 Figure 4.7: P o r protein incorporat ion into P O P C / P O P S l iposomes. Isopycnic density gradient centrifugation profile for gonococca l protein I incorporated into P O P C l iposomes containing va ry ing amounts o f P O P S : A ) 5%, B ) 10 % and C ) 2 5 % . Samples were centrifuged at 110 ,000g a v at 4 ° C for 24 hours on a continuous F i c o l l gradient (0-10%). Prote in (circles). Phospho l ip id (triangles). 120 o f free protein at the bot tom o f the gradient (Figure 4.8 A ) . Interestingly, as shown i n figure 4 . 8 B , an increase i n the D O D A C content to 10% resulted in a shift o f the prote in/ l ip id peak distr ibution towards the top o f the gradient. A complete shift o f the prote in / l ip id fraction to the top o f the gradient was observed after reconstitution w i t h 2 5 % D O D A C , indicat ing that proteol iposome density was less than the lowest F i c o l l concentrat ion (Figure 4 .8C) . Reconst i tu t ion w i t h cat ionic l i p i d resulted in complete protein incorporat ion; however, the properties o f the reconstituted systems were altered relative to neutral or anionic vesicles, as demonstrated by the shift o f the prote in / l ip id band towards a lower density. These changes were associated w i t h reduced antibody b ind ing act ivi ty compared to neutral and anionic proteoliposomes, as discussed in Chapter 5. 4.2.5 V e s i c l e morphologies Proteol iposome morphologies were also examined us ing cryo-electron microscopy ( C T E M ) . P o r proteoliposomes reconstituted f rom O G P detergent are shown i n Figures 4 . 9 A and B . Consistent w i t h Q E L S results, C T E M analysis shows that vesicles are heterogeneous w i t h regard to lamellar i ty and morphology. U n d e r l o w magnif ica t ion (Figure 4 . 9 A ) , vesicles are observed as aggregated structures o f va ry ing morphology w i t h large vesicles enc los ing many smaller vesicles, however , these vesicles may have been super-imposed on one another and thus they are perceived to be mul t i lamel lar . These smaller vesicles become discernable under higher magnif icat ion; however, P o r protein, due to its smal l size, is not v i s ib le w i t h i n the l i p i d bilayers. S imi l a r vesicles conta ining a pos i t ive ly charged l i p id , D O D A C , or a negatively charged l i p id , P O P S , display s imi la r morphologies (Figures 4 . 9 C - D and 4 . 9 E - F , respectively). Reconst i tuted systems were also size-reduced by extrusion through two, 121 Figure 4.8: P o r protein incorporat ion into P O P C / D O D A C l iposomes. Isopycnic density gradient centrifugation profile for gonococca l protein I incorporated into P O P C l iposomes containing vary ing amounts o f D O D A C : A ) 5%, B ) 10% and C ) 2 5 % . Samples were centrifuged at 110 ,000g a v at 4 ° C for 24 hours on a continuous F i c o l l gradient (0-10%). Protein (circles). Phospho l ip id (triangles). Gradient Depth (ml) 123 Figure 4.9: Cryo-e lec t ron micrographs o f reconstituted P o r proteol iposome formulations v i ewed under different magnifications. A ) and B ) P O P C / P o r , C ) and D ) P O P C / D O D A C / P o r , E ) and F) P O P C / P O P S / P o r , G ) and H ) Ex t ruded P O P C / P o r . 124 126 stacked polycarbonate filters o f 100 n m pore size (see Me thods ) in an attempt to generate uni lamel lar vesicles o f a un i form size distr ibution. A s shown in Figures 4 . 9 G and H , extruded proteoliposomes are approximately 7 0 % uni lamel la r and spherical w i t h an average diameter o f about 100 nm. Furthermore, the extrusion procedure resulted i n m i n i m a l loss o f protein and l i p id , as discussed in Sect ion 4.2.3. 127 4.3 D I S C U S S I O N P o r is the major constituent o f the gonococca l outer membrane, compr i s ing up to 6 0 % o f the protein content (Johnston and Gotsch l i ch , 1974 and Johnston and Gotsch l i ch , 1976). Furthermore, anti-Por antibodies have been shown to confer protection against subsequent challenge by gonococca l organisms ( V i r j i et al, 1987 and P l u m m e r et al, 1989). V a c c i n e trials employ ing gonococca l membrane blebs, however, have fai led to demonstrate a protective effect ( A r m i n j o n et al, 1987, Tramont, 1989 and Gula t i et al, 1991). T w o factors may have contributed to this lack o f success. Firs t , contaminat ion o f the blebs by other membrane constituents, notably Rmp (protein III), may have triggered the product ion o f b l o c k i n g antibodies. Recent studies, for example, have shown that the presence o f anti-Rmp antibodies results in an increased susceptibi l i ty to gonococca l infect ion (R ice et al, 1986 and P lummer et al, 1993). Second, P o r proteins may not have been adequately presented i n their native tertiary structure, resulting in on ly l o w levels o f ant i -Por antibody being produced. A s described below, the l iposomal P o r vaccine described herein avoids both o f these problems. T o prevent contaminat ion by Rmp, P o r protein was isolated from gonococca l strains from w h i c h the rmp gene had been deleted (Wetzler et al, 1989a and Wetz l e r et al, 1989b). Por ins isolated from these strains have been shown to be funct ional ly and ant igenical ly identical to those isolated from isogenic w i l d type parent strains. Pur i f ied P o r was then reintroduced into a bi layer environment by reconsti tution w i t h phosphol ip ids to generate wel l -def ined proteoliposomes. Eff ic ient and essentially complete P o r insertion cou ld be demonstrated when ful ly so lub i l i zed protein was employed. It was noted, however, that any insoluble protein in i t i a l ly present during reconsti tution was not incorporated into the resulting proteoliposomes. Pro te in reconsti tution was compared for systems consis t ing o f a 128 mixture o f phosphol ip id species ( P O P C : P O P E ) or a single l i p i d ( P O P C ) . The l i p i d mixture selected for this study consisted o f a b i layer fo rming species ( P O P C ) together w i t h a l i p i d ( P O P E ) w h i c h i n isola t ion can adopt the non-bilayer, hexagonal H n phase (Epand, 1985). It has previously been suggested that the hydrophobic , b i layer-spanning domain o f intr insic membrane proteins may more readily be accomodated in bi layers composed o f m i x e d l i p i d species due to the abi l i ty o f l ip ids possessing different dynamic molecula r shapes to pack at the l ip id-prote in interface wi thout creating b i layer defects (Navarro et al, 1984). In the present study, however, no differences in the eff iciency o f protein reconsti tution were seen between P O P C vesicles and P O P C P O P E systems. In these reconstituted systems, P o r was predominantly inserted in the same outwardly oriented configurat ion found i n the native membrane. Furthermore, the external ha i rp in loop domain was readily c leaved by t rypsin or a -chymot ryps in to y i e l d characteristic peptide fragments. A g a i n , this is consistent w i t h the protein exist ing i n its native tertiary structure. The research presented i n this chapter characterized the effect o f l iposome charge on protein incorporat ion efficiency. It was observed that P o r incorporat ion into cationic l iposomes was more efficient than for neutral or negatively charged systems. In addit ion, the incorporat ion o f larger proportions o f the cat ionic l i p i d , D O D A C , resulted i n changes in density o f the proteoliposomes as seen by the shift i n the prote in / l ip id fractions toward the top o f the i sopycnic density gradient (Figure 4.8). Th is density change may have been due to the increase in mean l i p i d cross-sectional area introduced by the carbon double bonds o f D O D A C . In addition, the absence o f phosphate groups and the lower molecular weight o f D O D A C compared to P O P S or P O P G may have also produced the shift o f the proteoliposomes toward a lower density. 129 Ear l ie r studies have shown that proteoliposomes generated by reconstitution from O G P detergent were largely un i lamel la r w i t h a mean diameter o f about 200 n m ( M i m m s et al, 1981). In contrast, the cryo-electron micrographs presented i n this chapter indicate that reconstituted systems are heterogeneous in their size, lamel lar i ty and morphology. Reconst i tuted proteoliposomes composed o f neutral or charged l ip ids were observed to be o f variable lamellar i ty w i t h mean diameters greater than 500 nm. These systems cou ld be size-reduced to vesicles o f about 100 n m mean diameter us ing an extrusion procedure (Hope et al., 1985 and M a y e r et al, 1986). Smal ler vesicles w o u l d be favorable w i t h regard to steri l izat ion and safety for human use. Conven t iona l heat steri l izat ion or i rradiat ion o f the vaccine formulat ion w o u l d not be suitable steri l izat ion methods because they w o u l d l i ke ly cause denaturation o f the protein and loss o f the antigenic determinants. In order to retain the immunogenic characteristics o f the vaccine, the preparation w o u l d have to be sterilized under non-denaturing condit ions. One method that w o u l d be suitable is terminal fi l tration. Th is procedure involves passing the reconstituted sample through a 0.2 m i c r o n filter that effectively removes any mic rob ia l organisms present i n the sample. C lea r ly this requires that the proteoliposomes be less than 200 n m i n diameter. Therefore, the extrusion procedure w i l l a l l ow vaccine preparation us ing G o o d Manufac tu r ing Pract ices ( G M P ) under clean condit ions that do not require ful l aseptic processing, w h i c h w i l l greatly reduce the complexi ty and cost o f vacc ine manufacture. In addit ion, the size-reduction o f proteoliposomes to L U V could lead to greater P o r protein exposed on the external l iposome surface than for M L V , w h i c h may have some P o r protein sequestered i n internal lamellae. Therefore, L U V w o u l d l i ke ly have more efficient antigen presentation and thus lead to an improved immune response. 130 CHAPTER 5 THE ANTIGENIC CHARACTERIZATON OF POR PROTEOLIPOSOMES The previous chapter characterized P o r protein orientation i n the l i p i d b i layer and the effect o f l i p i d compos i t ion on protein incorporat ion efficiency, ves ic le size and morphology. In the present chapter, P o r proteoliposomes, composed o f neutral l i p i d alone or i n mixtures w i t h anionic or cationic l i p id , are characterized to determine the effects o f l i p i d compos i t ion on in vitro antibody b ind ing and in vivo immunogenic i ty . 5.1 INTRODUCTION Liposomes , administered intravenously, are removed f rom c i rcula t ion and accumulate in organs o f the ret iculoendothelial system ( R E S ) , such as the lungs, l iver , spleen and bone mar row ( A l l e n and C h o n n , 1987). Preferential clearance o f l iposomes to tissues o f the R E S may be an advantageous aspect o f us ing l iposomes as carriers o f vaccines, because this may enable targeting and uptake o f l iposomes by macrophages, w h i c h are major participants i n antigen processing (Unanue, 1984). Research w i t h macrophages has suggested that the ce l l surface charge may influence particulate phagocytosis (Mutsaers and Papadimi t r iou , 1988). Studies conducted by N a k a n i s h i et al. (1997) showed that pos i t ive ly charged l iposomes were taken up more efficiently by macrophages and were more potent inducers o f humoral and cel l-mediated immuni ty than neutral or negatively charged carriers. Research conducted by Wetz l e r et al. (1988 and 1992) has suggested that a l iposomal P o r formulat ion w o u l d be effective at induc ing humora l immuni ty . The i r results showed that P o r proteoliposomes exhibi t h igh in vitro ant ibody b ind ing act ivi ty and thus g o o d surface 131 presentation o f the epitope for antibody recogni t ion compared to a lum or Freund ' s adjuvant preparations containing P o r protein. The results presented i n Chapter 4 demonstrated that gonococca l protein EB (Por) can be efficiently incorporated into l iposomes i n an orientation s imilar to that seen i n the native bacterial membrane. H o w e v e r , i n developing effective vaccines for particular diseases, several factors must be taken into account. Init ial protection to infection appears to be conferred through a humora l immune response mediated by antibodies against the pathogen ( A h m e d and Gray , 1996). Some intracel lular pathogens, such as Mycobacterium tuberculosis, appear to induce cel l -mediated immuni ty (Skinner et al, 1997). Vacc ines , therefore, should be designed to induce a broad immune response i n v o l v i n g a humoral immun i ty against the in i t ia l exposure and cel lular responses for protection against intracellular infection. The studies outl ined i n the present chapter examine the influence o f phosphol ip id compos i t ion on the in vitro and in vivo antigenic properties o f l iposomal gonococca l subunit vaccine formulations. P o r proteoliposomes described in the previous chapter are characterized w i t h regard to their antibody b ind ing act iv i ty and immunogen ic i ty i n a murine immuniza t ion model . In addit ion, the mouse immune response is characterized to determine the effect o f the route o f inocula t ion on immunogenic i ty . Furthermore, the antibody titers el ici ted are analyzed for i m m u n o g l o b u l i n serotypes to determine the type o f immune response induced, humoral or cel l-mediated. 132 5.2 RESULTS 5.2.1 A n t i b o d y b ind ing to P o r proteoliposomes determined us ing an E L I S A assay The b ind ing affinity o f proteoliposomes for anti-Por antibody was examined for reconstituted systems prepared w i t h either P O P C or P O P C P O P E (1:1) using two different batches o f puri f ied P o r protein ( G M P and M S 1 1 ) . In an earlier study, P o r had been reconstituted into proteoliposomes composed o f P O P C P O P E (1:4) using a different procedure to that described here. F o r comparison, therefore, s imi lar proteoliposomes were prepared and their antibody b ind ing act ivi ty determined. It is important to note that neither who le cells f rom gonococca l strain M S l l j e A r / w p nor any o f the five l iposomal preparations demonstrated any b ind ing act ivi ty w i t h antibodies directed at epitopes that appear to be buried wi th in the environment o f the gonococca l outer membranes ( M . S . B lake , unpubl ished results). U s i n g an inh ib i t ion E L I S A assay, b ind ing o f three surface reactive, monoc lona l antibodies ( M A b ) and a po lyc lona l rabbit serum against each o f the proteol iposome formulations was examined. A s shown i n F igure 5.1, each o f the P o r proteol iposome preparations showed inhibi tory act ivi ty i n the E L I S A assay us ing the M A b s , 1F11, 3H1 and 6F9 . These results suggest that P o r protein is oriented in the l iposomal membrane i n a confirmation that is s imi lar to that seen in the native bacterial membrane. In F igure 5.1 A , P o r proteoliposomes prepared from P O P C w i t h M S I 1 P o r and P O P C P O P E (1:1) w i t h G M P P o r both exhibited 5 0 % inhib i t ion at approximately 32 p g and 35 p g Por , respectively, whereas proteoliposomes prepared w i t h P O P C and G M P P o r and P O P C P O P E (1:1) w i t h M S I 1 P o r demonstrated 5 0 % inhib i t ion at 40 p g and 52 pg, respectively. Ves i c l e s prepared from P O P C : P O P E (1:4), according to Wetz le r and coworkers (1992), showed the lowest 133 F i g u r e 5.1: In vitro P o r antibody b ind ing act ivi ty. Inhib i t ion E L I S A assay showing antibody b ind ing activities o f f ive Por - l iposome preparations w i t h monoc lona l antibodies ( M A b s ) generated against various gonococca l strains: 1F11 M A b s . ( A ) , 3H1 M A b s (B) and 6F9 M a b s (C) . The proteoliposome formulations consisted o f proteoliposomes prepared from G M P P o r protein w i t h P O P C ( • ) and P O P C P O P E (1:1) ( A ) ; proteoliposomes prepared from M S 11 P o r reconstituted wi th P O P C ( • ) and P O P C P O P E (1:1) ( T ) and proteoliposomes prepared from M S I 1 P o r wi th P O P C P O P E (1:4) ( O ) . 135 potency w i t h 5 0 % inh ib i t ion at 55 ug P o r protein. In figure 5. I B , P O P C l iposomes w i t h M S 11 P o r and P O P C wi th G M P P o r demonstrated s imi lar potencies w i t h 5 0 % inh ib i t ion at 35 ug P o r protein, w h i c h were greater than the potency o f P O P C P O P E (1:1) systems w i t h G M P P o r w h i c h exhibi ted 5 0 % inh ib i t ion w i t h 41 ug Por . Proteol iposomes composed o f P O P C P O P E at a ratio o f 1:1 or 1:4 w i t h M S 11 P o r required 52 ug o f P o r to produce 5 0 % inhibi t ion. S imi la r b inding activit ies were observed w i t h monoc lona l antibody 6 F 9 (Figure 5 .1C) . Overa l l , the highest inh ib i t ion was seen for P O P C vesicles containing M S 11 Por . Th i s compar ison also held w h e n the assay was conducted using a po lyc lona l rabbit serum (Figure 5.2). A s shown in F igure 5.2, P O P C / M S 1 1 P o r proteoliposomes exhibi t 5 0 % inhib i t ion w i t h approximately 12 ng Por , whereas greater than 25 ug P o r is required for 5 0 % inhib i t ion w i t h each o f the other four formulations. A n a l y s i s o f the data indicates that P O P C / M S 1 1 P o r proteoliposomes have 3-fold greater b ind ing act ivi ty than P O P C P O P E (1:4) w i th M S 11 Por . In who le ce l l b i n d i n g assays, it was seen that approximately 10 9 gonococc i o f the strain M S I IJBAATW/? exhibi ted 3 0 % b ind ing , whereas 18 ug o f P O P C / M S 1 1 gave the same 3 0 % b ind ing act ivi ty ( M . S . B l a k e , unpubl ished results). Therefore, us ing the P O P C / M S 1 1 P o r formulat ion as the standard, a theoretical estimate o f the number o f P o r proteins per bacterium can be derived. U s i n g the equation: P o r proteins/Bacter ium = ( D / M ) x A + C D = A m o u n t o f P o r protein i n m g M = M o l e c u l a r weight o f P o r A = A v a g a d r o ' s N u m b e r 6.02 xlO 2 3 C = N u m b e r o f gonococca l cells 136 10 100 Porin-Liposome (pg protein) Figure 5.2: In vitro P o r antibody b ind ing act ivi ty w i t h rabbit anti-sera. Inhibi t ion E L I S A assay showing antibody b ind ing activit ies o f f ive l iposome preparations w i t h a po lyc lona l rabbit serum, antiserum 2-859. The proteol iposome formulat ions consisted o f proteoliposomes prepared from G M P P o r protein w i t h P O P C ( • ) and P O P C P O P E (1:1) (A); proteoliposomes prepared f rom M S 11 P o r reconstituted w i t h P O P C ( • ) and P O P C P O P E (1:1) ( T ) and proteoliposomes prepared from M S 11 P o r wi th P O P C P O P E (1:4) ( O ) . 137 Based on a protein molecular weight o f 3 6 K , the calcula t ion w o u l d suggest that each gonococcus contains 3 x 10 5 molecules o f P o r protein or 10 5 porins per bacterium, w h i c h is good agreement w i t h other estimates ( N i k a i d o , 1993). 5.2.2 In vitro antibody b ind ing act ivi ty o f charged P o r proteoliposomes Proteol iposomes composed o f P O P C alone or a mixture o f P O P C and 10% charged l i p i d , P O P S or D O D A C , were prepared and analyzed for their b ind ing affinity against an anti-Por monoclona l ant ibody us ing an inh ib i t ion E L I S A assay. Subsequent to the b ind ing studies described above, it was observed that improved assay design a l l owed much greater sensit ivity and detection o f nanogram quantities o f P o r protein. In F igure 5.3, P O P C P o r proteoliposomes exhibi ted 3 8 % inhib i t ion w i t h 2.4 ng P o r protein, whereas proteoliposomes containing P O P S or D O D A C demonstrated 3 5 % and 17% inhibi tory activi ty, respectively. A t 4.5 ng P o r protein, 5 0 % inhib i t ion was observed for P O P C proteoliposomes. P O P S -containing l iposomes showed 4 2 % inh ib i t ion w i t h 4.5 ng P o r protein, whereas on ly 2 4 % inhib i t ion was exhibi ted by D O D A C - c o n t a i n i n g l iposomes. Con t ro l l iposomes were found to have negl ig ible antibody b ind ing activi ty. The results suggest that the membrane changes introduced by D O D A C may prevent P o r protein from adopting the correct bi layer orientation required for epitope presentation and ant ibody b ind ing . 5.2.3 In vivo immune responses to proteoliposomes T o determine the antigenic properties o f P o r proteoliposomes, eight mice were inoculated w i t h either P o r incorporated into l iposomes, composed o f P O P C alone or in mixtures o f P O P C w i t h 10% P O P S or D O D A C , or free (non-reconstituted) P o r protein. 138 Figure 5.3: Effect o f charged l iposomes on antibody b ind ing activity. Inhibi t ion E L I S A showing b ind ing activities o f l iposome preparations w i t h ant i -Por monoc lona l antibody. The l iposomes formulations were: control P O P C ( O ) , control P O P C / P O P S (A) , control P O P C / D O D A C ( V ) , P O P C / P o r ( • ) , P O P C / P O P S / P o r ( A ) and P O P C / D O D A C / P o r ( T ) 139 Adminis t ra t ions were at 1 ug P o r protein g i v e n either intraperitoneally (i.p.) or intradermally ( i d . ) at week ly intervals (3 injections). The result ing immune response was analyzed by measuring the antibody titer. It was anticipated that the first inocula t ion w o u l d induce a pr imary antibody response and the formation o f memory cells . U p o n booster injections, the antigen w o u l d trigger a faster and more intense secondary ant ibody response. In F igure 5.4, it can be seen that antibody titers el ici ted by the first P o r proteol iposome inocula t ion were not significantly higher than titers observed for control l iposomes. The first intraperitoneal booster injection at 3 weeks induced an average 50-fold (1.7 l og unit) increase i n ant ibody titer, whereas a 100-fold (2 l og unit) increase was observed for the proteoliposomes after the first i d . booster inoculat ion. The marked increase i n titers signal the strong secondary response that occurs after the first inoculat ion. The second booster injection o f P o r proteoliposomes intraperitoneally or intradermally at 7 weeks produced average increases in antibody titers o f 18-fold and 14-fold, respectively (Figure 5.4). F o r the i.p. route, the free P o r preparation e l ic i ted an antibody titer o f 4.62 l og units after 9 weeks, w h i c h was sl ight ly higher than the titer (4.24 l og units) induced by D O D A C -containing P o r proteoliposomes, but was not statistically significant (P>0.05) (Figure 5 .4A) . However , the titer el ici ted by free P o r control was s ignif icant ly higher (P<0.05) than the titers, 3.83 and 3.90 l og units, induced by P O P C and P O P S - c o n t a i n i n g P o r proteoliposomes. There was no significant difference between the antibody titers generated by the neutral and anionic proteoliposomes. F o r the i d . route o f inoculat ion, free P o r protein, cat ionic and neutral proteol iposome formulations e l ic i ted titers o f 4 .81, 4.66 and 4.58 l o g units, respectively, but were not s ignif icantly different (P>0.05) (Figure 5 .4B) . In contrast, anionic proteoliposomes induced a s ignif icant ly lower antibody titer o f 4.13 l og units (P<0.05). 140 Figure 5.4: A n t i - P o r I g G titers o f mice i m m u n i z e d w i t h P o r protein preparations. A ) Intraperitoneal and B ) Intradermal immuniza t ion . H B S control ( O ) , control P O P C (A), control P O P C / P O P S ( • ) , control P O P C / D O D A C ( V ) , Free P o r control ( • ) , P O P C / P o r ( • ) , P O P C / P O P S / P o r ( • ) , P O P C / D O D A C / P o r ( • ) . A n t i b o d y titer was determined by taking the log o f the reciprocal d i lu t ion that gave an absorbance o f 0.2. E a c h data point represents the mean titer o f eight animals per t ime point. 142 Con t ro l l iposomes d id not generate an appreciable ant ibody titer by either route o f inoculat ion. The effect o f the route o f administrat ion was examined for the four P o r preparations. A t 9 weeks, neutral and cat ionic proteoliposomes induced signif icant ly higher antibody titers when administered intradermally (P<0.05). The route o f inocula t ion d id not produce any significant difference i n titers induced by free P o r protein (P>0.05). S imi la r ly , there was no difference between antibody responses for anionic proteoliposomes el ic i ted by either i.p. or i d . injection. 5.2.4 A n t i b o d y i so typing o f immune sera In addi t ion to determining the antibody titers produced by the proteoliposomes, the immune serum was characterized for antibody subclasses present; more specif ical ly, the relative amounts o f I g G l or I g G 2 a were determined. In the case o f mice inoculated intraperitoneally w i t h free or cationic P o r proteoliposomes, sera obtained at nine weeks d id not exhibit a predominance o f I g G l or I g G 2 a (Figure 5 .5A) . H o w e v e r , sera from mice inoculated w i t h neutral ( P O P C ) proteoliposomes contained a s ignif icant ly higher fraction o f I g G l antibodies than I g G 2 a (P<0.05). Converse ly , anionic P o r proteoliposomes induced an approximately two- fo ld greater amount o f I g G 2 a antibodies compared to I g G l (P<0.05). U p o n i d . inoculat ion, examinat ion o f the immune sera indicates a shift towards higher I g G 2 a / I g G l ratios from 1.7:1 to 2.2:1 (Figure 5 .5B) . It can be seen that sera induced by free P o r and cationic P o r proteoliposomes consisted o f a two- fo ld higher level o f I g G 2 a compared to I g G l ; however, the relative difference was not statistically significant (P>0.05). S imi l a r ly , neutral and anionic proteol iposome immune sera contained a s ignif icant ly higher (P<0.05) 143 Figure 5.5: Effect o f immuniza t ion route on I g G isotypes o f ant i -Por antibodies. A ) Intraperitoneal and B ) Intradermal immuniza t ion . Absorbance at 490 gives a measure o f the relative concentrations o f the I g G l and I g G 2 a antibody subclasses. D a t a represent the mean ± S E M o f five animals. 145 proport ion o f I g G 2 a antibodies compared to I g G l . The ratio o f the t w o antibody subclasses may be indicat ive o f the type o f immune response that is be ing induced. A humoral response is associated w i t h elevated levels o f I g G l , whereas a predominant I g G 2 a response is indicat ive o f a cel l -mediated immune response ( M o s m a n n and Coffman, 1989). 5.2.5 H i s to logy It was observed that animals rece iv ing the cat ionic proteoliposomes exhibi ted inf lammat ion at the site o f intradermal injection that was absent for the other proteol iposome and empty l iposome formulations. H i s t o l o g y o f the sk in showed that empty, control l iposomes d id not induce loca l in f lammat ion or the inf i l t rat ion o f leukocytes to the injection site and that the his tologic appearance o f the sk in was s imi lar to that o f normal , naive tissue (Beyaert etal., 1991). In contrast, however, cationic proteol iposome immuniza t ion produced an intense loca l immune response, characterized by the large infi l t rat ion o f neutrophils, mast cells and, to a lesser degree, eosinophils i n response to the D O D A C , i n association w i t h P o r protein antigen (Figure 5.6). Neut rophi l s were observed as cells w i t h mul t i - lobed dark, violet-stained nuclei surrounded by a lighter, pink-stained cytoplasm. M a s t cells were also seen at the site o f injection, w h i c h were characterized by deep, viole t staining o f the cytoplasm and the nucleus, such that the nucleus cou ld not be dis t inguished from the cytoplasm. Other subsets o f lymphocytes that were observed were monocytes (histocytes) and eosinophils . Eos inoph i l s were identif ied by the presence o f a b i - lobed nucleus, whereas monocytes were characterized by an indented, horseshoe-shaped nucleus. These results suggest that cationic l i p id , D O D A C , may be tox ic to the cells and thus causing the loca l inf lammat ion seen at the injection site. 146 Figure 5.6: Photograph o f sk in cross-section o f the intradermal injection site v i e w e d at 40x magnif ica t ion A ) P O P C l iposome and B ) P C / D O D A C / P o r l iposome immuniza t ion . 147 5.3 D I S C U S S I O N Liposomal vaccine preparations have been shown to elicit humoral (Therien et al, 1990) and cellular immune responses (Garcon and Six, 1991). Research using chemical markers has demonstrated that mammalian cell membranes exhibit a net negative charge (Danon et al., 1972). Work on murine peritoneal macrophages demonstrated that cationic ferritin was internalized whereas native or anionic ferritin was not, suggesting that the negative charge on the cell surface influences particulate phagocytosis (Mutsaers and Papadimitriou, 1988). Extrapolation of these results suggests that the surface charge on liposomes may effect phagocytosis by macrophages, and thus influence the type of immune response. In addition, experiments with rat peritoneal macrophages have shown that liposome-cell association can be enhanced with an increase in positive surface charge (Schwendener et al., 1984). Studies conducted by Nakanishi et al. (1997) showed that positively charged liposomes were taken up more efficiently by macrophages and were more potent inducers of humoral and cell-mediated immunity than neutral or negatively charged carriers. The antibody-binding activities of various Por liposome formulations were determined using both anti-Por monoclonal antibodies with known reactivity against gonococcal organisms and an immunized rabbit serum. The results showed that proteoliposomes prepared from P O P C with MS11 Por exhibited high antibody binding activity with several monoclonal anti-Por antibodies, as well as with rabbit immune serum. A l l the Por proteoliposome formulations prepared, using the methodology described herein, exhibited consistently higher levels of antibody binding compared to systems prepared from P O P C P O P E (1:4), as described previously (Wetzler et al, 1992) (Figure 5.1 and 5.2). 148 E L I S A assays were used to determine the effect o f l iposome charge o n in vitro antibody b inding . Despi te the higher protein incorporat ion eff iciency that was observed for D O D A C - c o n t a i n i n g proteoliposomes described in Chapter 4, E L I S A assays indicated that neutral and negatively charged systems had s imi la r antibody b ind ing activit ies and profiles. Neut ra l ( P O P C ) P o r proteoliposomes had an overa l l 5 0 % greater antibody b ind ing act ivi ty than cat ionic P o r proteoliposomes (Figure 5.3). These results suggest that membrane changes produced by D O D A C and/or charge interaction w i t h negatively charged residues in the P o r protein reduced the surface presentation o f the epitope for antibody recognit ion. The antibody b ind ing data indicate that neutral P o r proteoliposomes have the greatest epitope presentation and thus a greater proport ion o f P o r protein i n its native conformation. In vivo immuniza t ion studies i n a mouse mode l indicated that free protein and l iposomal P o r formulations a l l induced s imi la r antibody titers that were greater than the control empty l iposome formulations. Immuniza t ion v i a the intraperitoneal route resulted i n free P o r protein e l ic i t ing the highest antibody titer fo l lowed by cat ionic l iposomes. The neutral and negatively charged proteol iposomes were the next efficacious o f the preparations tested (Figure 5.4). Free Por , cationic and neutral proteoliposomes e l ic i ted s imi lar ant ibody profiles when administered v i a the intradermal route. A l t h o u g h purif ied P o r protein has been shown to be immunogenic i n our study as w e l l as i n previous studies, a humoral response alone is not sufficient in p rov id ing protection against the gonococca l organism. W e t z l e r et al. (1988) showed Por inserted into l iposomes i n its native orientation el ici ted antibodies that agglutinated intact organisms and had a higher bacter ic idal and opsonic act ivi ty than a lum-generated sera. In addition, who le ce l l absorption studies and synthetic peptide E L I S A assays indicated that proteoliposome antisera contained a higher percentage o f antibodies 149 against the surface-exposed epitopes o f P o r protein required for opsonizat ion and phagocytosis. Wetz le r et al. (1992) further demonstrated proteol iposome-induced ant i -Por I g G had a greater functional react ivi ty than free P o r or alum-generated sera against surface exposed epitopes o f protein I i n an intact gonococca l organism. A l t h o u g h pos i t ive ly charged proteol iposomes were effective in e l i c i t ing a humoral immune response, it was observed that animals rece iv ing this formulat ion exhibi ted inf lammat ion at the site o f intradermal inject ion that was absent for the other proteol iposome and empty l iposome formulations. H i s t o l o g y o f the sk in revealed that there was an infi l t rat ion o f neutrophils, eosinophils and monocytes indicat ing that D O D A C , i n association w i t h antigen, causes an intense loca l immune response. Studies have shown that cationic l iposomes are h igh ly tox ic toward phagocytic cells , such as macrophages, but not toward non-phagocytic T lymphocytes ( F i l i o n and P h i l l i p s , 1997). M i c e rece iv ing empty D O D A C proteoliposomes d id not exhibit any symptoms o f tox ic i ty nor was there any inf lammat ion at the site o f injection. These results might suggest that empty D O D A C l iposomes do not trigger an immune response; however, the presence o f a foreign antigen, Por , may initiate phagocytosis and a subsequent cytokine response result ing i n the marked infi l t rat ion o f macrophages at the injection site. U p o n contact or uptake o f the cationic l iposomes, macrophages may process and present the antigen on the M H C molecule to elici t an immune response; however, the presence o f cat ionic l i p i d may cause some ce l l toxic i ty . Macrophage tox ic i ty may also trigger the induct ion Of cyto toxic T lymphocytes and thus further enhancing the immune response. A s i n the case o f a lum, inf lammat ion at the site o f inocula t ion suggests that cationic proteol iposomes w o u l d be undesirable as a candidate vaccine. 150 A l t h o u g h a humoral immun i ty (FA) may afford protection against an extracellular pathogen, cel l-mediated immuni ty ( C M I ) w o u l d be required to combat an intracellular challenge. In order for a C M I response to occur, the antigen has to be endogenously processed into the cytosol and presented by the major h is tocompat ib i l i ty complex ( M H C ) class I molecules on antigen presenting cells, such as macrophages (Brac ia le et al, 1987 and D a l M o n t e and Szoka , 1989). Prev ious studies have suggested that l iposomal antigen is taken up and degraded i n lysosomes and then recycled to endosomes and subsequently presented to T-cel ls i n association w i t h M H C - c l a s s II molecules (Hard ing et al, 1991). The i r research suggests that l iposomal immuniza t ion favors the induct ion o f H I over C M I . Howeve r , other research has shown that l iposomal antigen does enter the cytoplasm and is processed in the t rans-Golgi o f murine macrophages, whereas free soluble antigen is l i ke ly degraded in endosomes and is unable to reach the t rans-Golgi (Rao et al, 1997). These results may expla in w h y l iposomal antigens have been able to induce cytotoxic T lymphocytes ( C T L ) , whereas free soluble antigen d id not (Reddy et al, 1992). Moreove r , the type o f immune response has been associated w i t h a particular antibody subclass. T y p i c a l l y , T h l type responses are characterized by C T L s and high levels o f I g G 2 a induct ion, whereas higher I g G l levels associated w i t h very li t t le or no induct ion o f C T L s are indicat ive o f Th2 type responses ( M o s m a n n and Cof fman , 1989). Studies o f D N A vaccines have demonstrated the balance between the T h l - and Th2-type response ( C h o w et al, 1998). M i c e i m m u n i z e d w i t h a hepatitis B virus ( H B V ) D N A vaccine and a T h l cytokine gene were shown to have enhanced T h l cel ls and C T L s w i t h concomitant increase in I g G 2 a antibodies. A marked reduct ion o f T h 2 cells and I g G l antibodies accompanied these observations. Converse ly , co-inject ion o f H B V D N A vaccine and a Th2 cytokine gene 151 produced an increase in Th2 cel ls and I g G l levels w i t h a suppression o f T h l ce l l differentiation and reduction in I g G 2 a product ion. Others researchers have also demonstrated the balance between T h l and T h 2 type responses (Kostense et al, 1998). Some investigators have also suggested that the route o f administrat ion o f a vacc ine may influence the immunog lobu l in subclass and cy tokine response (Pertmer et al, 1996 and L e e and Sung, 1998). The sk in has a populat ion o f h igh ly efficient A P C , such as dermal dendrit ic cells and Langerhans ' cel ls (Anjeure et al, 1999). Dendr i t i c cel ls ( D C ) have essential function in the development o f the immune response against mic rob ia l pathogens, as w e l l as tumors. An t igen s t imulat ion o f D C leads to the act ivat ion and mob i l i za t ion o f D C , w h i c h can migrate and carry antigen from the sk in to the T h cells located in the l y m p h nodes. D N A vaccinat ion studies have indicated that cutaneous D N A immuniza t ion can activate and mob i l i z e D C , st imulating them to produce I L - 1 2 , a T h l - t y p e cytokine, and thus t r iggering a Th l -p redominan t immune response (Jakob etal, 1998 and Jakob etal, 1999). The results o f the immune serotyping showed that upon intraperitoneal injection, there was either no predominant antibody subclass or a tendency toward higher I g G l antibody levels, as seen for the neutral proteol iposomes (Figure 5.5). Af te r intradermal immuniza t ion , a shift toward higher I g G 2 a / I g G l ratios was observed for each o f the proteol iposome formulations, indica t ing that a T h l type response may be prevalent. These results might suggest that the intradermal route o f vaccina t ion may be effective for targeting D C to induce a more predominant T h l type response and hence the induct ion o f C T L response for protection against intracellular infection. H o w e v e r , detailed studies o f cy tokine responses and C T L assays are required to further characterize the cel l -mediated immune response. 152 CHAPTER 6 SUMMARY This thesis describes the development and characterization o f l iposomal gonococca l subunit vaccine formulations. Exper iments examined the incorporat ion o f the major gonococca l outer membrane protein (Por) into l iposomes and characterized the factors inf luencing protein reconstitution. In particular, the effects o f prote in- l ip id ratio, l i p i d species, l iposome charge, duration o f dia lys is and detergent properties were examined. F o r comparison, a meningococca l protein preparation, representing a mixture o f different proteins, was reconstituted into l iposomes. The b iophys ica l properties o f reconstituted P o r proteoliposomes were characterized i n terms o f protein orientation w i t h i n the l i p i d bi layer , as w e l l as vesicle size and morphology. Furthermore, studies characterized the in vitro and in vivo antigenicity o f P o r proteoliposomes. G o n o c o c c a l membrane protein, Por , was reconstituted into l iposomes us ing detergent dialysis . The rate o f detergent removal was observed to be more rapid for non-ionic , oc ty l glucopyranoside ( O G P ) than for the ion ic detergent sodium cholate. Res idua l levels o f detergents after extended dialysis were b e l o w the cr i t ica l mice l l e concentration ( C M C ) , at concentrations that are un l ike ly to affect proteol iposome stability. Cholate-mediated reconstitution resulted in incomplete P o r protein incorporat ion into l iposomes, whereas a l l o f the protein was reconstituted into the l i p i d b i layer during O G P - m e d i a t e d reconstitution at prote in- l ip id ratios o f 0.01:1. Th is difference may result f rom the s lower rate o f removal o f sodium cholate during dia lys is and hence the greater opportunity for protein aggregation and/or denaturation pr ior to bi layer formation. The results o f this research demonstrate that 153 detergent properties influence protein incorporat ion and thus the choice o f detergent for product ion o f proteoliposomes must be taken into account when characterizing the reconstitution process. Studies were performed to examine the effects o f prote in- l ip id ratios o n protein incorporat ion efficiency. It was observed that for a g iven l i p i d concentration there is a f ixed amount o f protein that can be incorporated into the l i p i d bi layer . Reconst i tu t ion experiments w i t h protein concentrations above this level resulted i n increased amounts o f unincorporated protein as observed f o l l o w i n g i sopycn ic density gradient centrifugation. T o m a x i m i z e protein insertion into the bi layer, the development o f an effective subunit vacc ine requires determination o f the factors for opt imal incorporat ion and surface presentation o f the protein antigen. Eff ic ient use o f purif ied protein antigens w o u l d reduce the amount o f antigen required and hence decrease vacc ine costs. Another aspect o f protein reconsti tution examined i n this research related to ves ic le size and morphology. U s i n g both quasi-elastic light scattering ( Q E L S ) and cryo-electron microscopy ( C T E M ) , reconstituted proteoliposomes were found to be heterogeneous structures vary ing i n lamel lar i ty and shape w i t h mean diameters i n i n excess o f 300 nm. A potential vaccine for human use w o u l d have to be steri l ized during preparation. Convent iona l steri l izat ion methods, such as steam steri l izat ion or i on i z ing radiation, w o u l d l i k e l y be unsuitable due to protein and l i p i d degradation. Therefore, terminal f i l trat ion o f proteoliposomes through 0.2 u m filters was examined as a potential s ter i l izat ion procedure. However , reconstituted samples were in i t i a l ly too large to be steri l ized by filtration and therefore it was necessary to size-reduce the proteoliposomes pr ior to terminal f i l trat ion. A g a i n , the techniques o f Q E L S and C T E M were used to conf i rm that proteoliposomes cou ld 154 be size-reduced to about 100 n m i n diameter us ing an extrusion procedure w i t h m i n i m a l loss o f protein and phosphol ip id . These size-reduced systems w o u l d then be suitable for terminal filtration through 0.2 p m filters. Furthermore, C T E M revealed that approximately 7 0 % o f the size-reduced proteoliposomes were uni lamel lar . The research presented in this thesis examined the effects o f l i p i d species and charge on protein incorporat ion and P o r antigenicity. Reconst i tu t ion studies w i t h b i layer- ( P O P C ) and non-bi layer- forming ( P O P E ) l ip ids showed that there were no differences i n protein incorporat ion efficiencies, or the orientation o f P o r w i th in the l i p i d bi layers, for different l i p i d formulations. In addition, in vitro antibody b ind ing experiments indicated that proteoliposomes consis t ing o f neutral b i layer fo rming l i p i d ( P O P C ) composi t ions exhibi ted greater antibody b ind ing act ivi ty and cross-reactivity w i t h var ious strain specific anti-Por antibodies than reconstituted systems containing P O P C P O P E mixtures. B o t h P O P C and P O P C P O P E P o r proteoliposomes exhibi ted greater antibody b ind ing act ivi ty than a reconstituted system prepared as described in earlier publ icat ions from a different research group. Moreover , the P o r formulations developed i n the present research exhibi ted h igh reactivity w i t h rabbit immune serum, further indica t ing that there is a h igh degree o f surface exposure o f the antibody recogni t ion site. P o r protein was also reconstituted into l iposomes containing anionic or cationic l ip ids to characterize the effect o f charge on protein insert ion eff ic iency and the antigenic properties o f the resulting proteoliposomes. F o r P o r proteoliposomes containing 5 we igh t% anionic l i p i d ( P O P S or P O P G ) , comparable protein incorporat ion efficiencies were seen compared to neutral proteoliposomes. Increasing the we igh t% composi t ion o f anionic l i p i d resulted in a decrease in incorporat ion eff iciency. In contrast, there was essentially complete 155 protein incorporat ion when P o r was reconstituted into cationic l iposomes containing up to 25 we igh t% D O D A C . Subsequent in vitro ant ibody b ind ing studies revealed that neutral proteoliposomes exhibi ted greater b inding act ivi ty than negatively or pos i t ive ly charged systems. This may indicate that neutral proteol iposomes have a greater surface exposure o f the epitope for antibody recogni t ion and b ind ing or that antibody b ind ing is inhibi ted by electrostatic effects. Studies were then conducted in mice to evaluate the immune response to P o r proteol iposome formulations. These studies compared antibody titers over three inoculat ions by either intraperitoneal or intradermal inject ion o f proteoliposomes o f differing l i p i d composi t ion and free P o r protein. The results o f in vivo immuniza t ion o f mice show that free protein, neutral and cat ionic proteoliposomes el ic i ted s imi la r antibody titers that were greater than titers induced by anionic proteoliposomes by either route o f administrat ion. A l t h o u g h these results indicate that each o f the proteol iposome formulat ions and free P o r protein are effective at induc ing an antibody titer, a h igh titer alone may not be sufficient for protection against infection. Corre la t ion o f the in vitro and in vivo results indicate that efficient epitope presentation and h igh antibody titers w o u l d l i k e l y induce protective immuni ty . The fact that free protein induces a h igh antibody titer may not be sufficient for e l ic i t ing protective immuni ty , because antibodies generated against denatured protein may not recognize the native P o r protein conformation. Therefore, an effective antibody titer must contain a h igh concentration o f antibodies directed against the correct surface epitope in order to trigger opsonizat ion and phagocytosis o f the gonococca l pathogen. The antibody b ind ing assays indicate P O P C proteoliposomes to be the most effective formulat ion i n terms o f epitope presentation and induct ion o f effective ant ibody titers. 156 Immune sera were characterized w i t h regard to the specific ant ibody serotypes generated by immuniza t ion w i t h P o r proteoliposomes o f different l i p i d compos i t ion and free P o r protein. The ratios o f I g G l and I g G 2 a serotypes may indicate whether a humoral or ce l l -mediated immune response is predominant. Intradermal inocula t ion w i t h proteoliposomes el ici ted higher I g G 2 a titers and an increase i n the ratio o f I g G 2 a / I g G l , whereas intraperitoneal immuniza t ion induced greater I g G l titers than IgG2a . These f indings suggest that proteoliposome immuniza t ion v i a the intradermal route enables antigen del ivery to the cytoplasm and subsequent antigen processing and presentation to the major his tocompat ibi l i ty complex class I molecule . Therefore, proteoliposomes may be potential inducers o f cel l-mediated immuni ty and associated cytotoxic T lymphocytes ( C T L s ) that are required for combating intracellular infections. The research presented in this thesis outlines the development and characterization o f a l iposomal gonococca l subunit vaccine. The experiments described herein demonstrate the conditions for effective and opt imal incorporat ion o f a membrane protein antigen. A n t i g e n i c characterization o f these proteoliposomes suggests that this system may have potential as a l iposomal subunit vaccine against Neisseria gonorrhoeae. The p romis ing results reported in this thesis w i l l hopefully lead to further characterization o f this preparation in terms o f protection against gonococcal challenge, induct ion o f serum lysis and k i l l i n g o f intact organisms, as w e l l as induct ion o f C T L s for conferring protection against intracellular infection, especial ly at mucosal sites. Ul t ima te ly , it is hoped that such a candidate subunit vaccine w o u l d be used c l i n i ca l l y i n the prevent ion or control o f gonococca l infection. 157 REFERENCES A h m e d , R . and Gray , D . (1996) Immuno log ica l memory and protective immuni ty : understanding their relation. Science 272, 54-60. A l l e n , J . C . and Humphr ies , C . (1975) The use o f zwi t te r ionic surfactants i n the agarose chromatography o f b io log ica l membranes. F E B S Letters 57, 158-162. A l l e n , T M . (1981) A study o f phosphol ip id interactions between high-density l ipoproteins and smal l un i lamel la r vesicles. B i o c h i m i c a et B i o p h y s i c a A c t a 640, 385-397. A l l e n , T M . and Chonn , A . (1987) La rge uni lamel la r l iposomes w i t h l o w uptake into the ret iculoendothelial system. F E B S Letters 223, 42-46. A l l e n , T . M . , Hansen, C . and Rutledge, J. (1989) L iposomes w i t h prolonged c i rcula t ion times: factors affecting uptake by ret iculoendothelial and other tissues. B i o c h i m i c a et B i o p h y s i c a A c t a 981, 27-35. A l l e n , T . M . , Romans , A . Y . , Kercret , H . and Segrest, J .P. (1980) Detergent removal dur ing membrane reconstitution. B i o c h i m i c a et B i o p h y s i c a A c t a 601, 328-342. A l l i s o n , A C . and Byars , N . E . (1986) A n adjuvant formulat ion that selectively el ici ts the formation o f antibodies o f protective isotypes and o f cel l -mediated immuni ty . Journal o f Immunolog ica l M e t h o d s 95, 157-168. A l v i n g , C . R . (1987) L iposomes as carriers for vaccines. In: L iposomes , from B i o p h y s i c s to Therapeutics. Ostro, M . J . , editor. M a r c e l Dekke r , Inc., N e w Y o r k . 195-218. A l v i n g , C . R . (1991) L iposomes as carriers o f antigens and adjuvants. Journal o f Immunolog ica l M e t h o d s 140, 1-13. A l v i n g , C . R . , Richards , R . L . , M o s s , J . , A l v i n g , L L , Clements , J .D . , Shiba, T , K o t a n i , S., W i r t z , R . A . and Hockmeye r , W T . (1986) Effectiveness o f l iposomes as potential carriers o f vaccines: applications to cholera t ox in and human malar ia sporozoite antigen. V a c c i n e 4, 166-172. Angrand , M . , B r i o l a y , A . , R o n z o n , F . and R o u x , B . (1997) Detergent-mediated reconstitution o f a g lycosyl -phosphat idyl inos i to l -prote in into l iposomes. European Journal o f B i o c h e m i s t r y 250, 168-176. A n h o l t , R . R . H . (1988) Funct iona l reconsti tution o f the olfactory membrane: incorporat ion o f the olfactory adenylate cyclase in l iposomes. B iochemis t ry 27, 6464-6468. Anho l t , R . R . H . , A e b i , U . , Pedersen, P . L . and Snyder, S H . (1986) So lub i l i za t ion and reassembly o f the mi tochondr ia l benzodiazepine receptor. B iochemis t ry 25, 2120-2125. 158 Anjeure, F , M a r t i n , P . Ferrero, I., Fraga, M L . , del H o y o , G . M . , Wr igh t , N . and A r d a v i n , C . (1999) Def in i t ion o f dendri t ic ce l l sub-populations present i n the spleen, Peyer ' s patches, l y m p h nodes and sk in o f the mouse. B l o o d 93, 590-598. An t imis i a r i s , S . G . , Jayasekera, P . and Gregor iadis , G . (1993) L iposomes as vacc ine carriers. Incorporation o f soluble and particulate antigens in giant vesicles. Journal o f Immuno log ica l Me thods 166, 271-280. A r m i n j o n , P . , Cadoz , M . , M o r s e , S .A . , R o c k , J .P. and Sarafian, S . K . (1987) Bacter ia l and Opson ic Ac t i v i t i e s o f Sera f rom Individuals Immunize w i t h a G o n o c o c c a l Prote in I V a c c i n e [Abstract E -92 ] . In: Proceedings o f the A n n u a l M e e t i n g o f the A m e r i c a n Society for M i c r o b i o l o g y . Wash ing ton , D C : A S M . Arms t rong , J A . , Wee , S H . , G o o d w i n , C S . and W i l s o n D H . (1987) Response o f Campylobacter pyloridis to antibiotics, b i smuth and an acid-reducing agent in vitro, an ultrastructural study. Journal o f M e d i c a l M i c r o b i o l o g y . 24(4), 343-350. Bangham, A D . , Standish, M M . and Watk ins , J . C . (1965) Di f fus ion o f univalent ions across the lamellae o f swol len phosphol ipids . Journal o f M o l e c u l a r B i o l o g y 13, 238-252. B a r k i n , R . M . and P ich ichero , M E . (1979) Diphtheria-pertussis-tetanus vaccine: reactogenicity o f commerc ia l products. Pediatr ics 63, 256-260. Barrera , O. and Swanson, J. (1984) Pro te in I A and I B exhibi t different exposures and orientations i n the outer membranes o f Neisseria gonorrhoeae. Infection and Immuni ty 44, 565-568. Bel la re , J R . , D a v i s , H . T . , Scr iven, L E . and Ta lmon , Y . (1988) Cont ro l led environment v i t r i f ica t ion system: an improved sample preparation technique. Journal o f E l ec t ron M i c r o s c o p y Techniques 10, 87-111. Beyaert , R , D e Potter, C , Vanhaesebroeck, B . , V a n R o y , F . and Fiers , W . (1991) Induct ion o f inf lammatory ce l l infi l t rat ion and necrosis in normal mouse sk in by the combined treatment o f tumor necrosis factor and l i t h ium chlor ide. A m e r i c a n Journal o f Pa tho logy 138, 727-739. B l a c k , W . J . , Schwalbe, R . S . , N a c h a m k i n , I. and Cannon , J . G . (1984) Character izat ion o f Neisseria gonorrhoeae protein II phase var ia t ion by use o f monoc lona l antibodies. Infection and Immuni ty 45, 453-457. B l a k e , M . S . (1985) Implicat ions o f the active role o f gonococca l porins i n disease. In. The Pathogenic Neisseriae. Schoolnik , G . K . , editor. A m e r i c a n Society for M i c r o b i o l o g y , Wash ing ton , D C . 251-258. 159 B l a k e , M . S . and Gotsch l i ch , E C . (1982) Pur i f i ca t ion and partial characterization o f the major outer membrane protein o f Neisseria gonorrhoeae. Infection and Immuni ty 36, 277-283. B l a k e , M . S . and Gotsch l i ch , E C . (1983) G o n o c o c c a l membrane proteins: speculation o n their role in pathogenesis. In: Progress i n A l l e r g y . H o s t Parasite Relat ionships i n G r a m -Negat ive Infections. Hanson , L A . , K a l l o s , P . and Westphal , O . (Eds.) Base l : Karger , 33, 298-313. B l a k e , M . S . , Gotsch l ich , E C . and Swanson, J. (1981) Effects o f proteolyt ic enzymes o n the outer membrane proteins o f Neisseria gonorrhoeae. Infection and Immuni ty 33,212-222. B l a k e , M . S . , Ly t ton , E . J . , Seiff, M E . and Got sch l i ch , E C . (1988) Studies on gonococca l protein III. In. Bacter ia -Host C e l l Interaction. H o r w i t z , M . A . , editor. A l a n R . L i s s , Inc., N e w Y o r k , 85-97. B o m a n , N . L . , M a s i n , D , M a y e r , L . D . , C u l l i s , P R . and B a l l y , M B . (1994) L i p o s o m a l vincr is t ine w h i c h exhibits increased drug retention and increased c i rcula t ion longevi ty cures mice bearing P388 tumors. Cancer Research 54, 2830-2833. Brac ia le , T . J . , M o r r i s o n , L . A . , Sweetser, M X , Sambrook, I , Gething, M . - J . and Brac ia le , V L . (1987) A n t i g e n presentation pathways to class I and class II M H C - r e s t r i c t e d T lymphocytes . Immunolog ica l R e v i e w s 98, 95-114. Brewer , J . M . , R i c h m o n d , J. and Alexander , J . (1994) The demonstration o f an essential role for macrophages i n the in vivo generation o f I g G 2 a antibodies. C l i n i c a l and Exper imenta l Immunology 97(1), 164-171. Br i t igan , B E , Cohen , M . S . and Spar l ing, P F . (1985) G o n o c o c c a l infection: a mode l o f molecular pathogenesis. N e w E n g l a n d Journal o f M e d i c i n e 312, 1683-1694. B r o o k s , G . F . (1985a) Structures and characteristics o f the gonococca l ce l l envelope. In. G o n o c o c c a l Infection. B r o o k s , G . F . and Donegan , E A . (Eds.) E d w a r d A r n o l d , L o n d o n , pp. 9-22. B r o o k s , G . F . (1985b) Pathogenesis and i m m u n o l o g y o f gonococca l infection. In: G o n o c o c c a l Infection. B r o o k s , G . F . and Donegan , E A . (Eds.) E d w a r d A r n o l d , L o n d o n , pp. 51-82. B r o o k s , G . F . (1985c) Salpingit is . In: G o n o c o c c a l Infection. B r o o k s , G . F . and Donegan , E A . (Eds.) E d w a r d A r n o l d , L o n d o n , pp. 105-120. B r o o k s , G . F . (1985d) Disseminated gonococca l infection. In: G o n o c o c c a l Infection B r o o k s , G . F . and Donegan , E A . (Eds.) E d w a r d A r n o l d , L o n d o n , pp. 121-131. B r o o k s , G . F . and Donegan, E A . (1985) U n c o m p l i c a t e d gonococca l infect ion. In: G o n o c o c c a l Infection. B r o o k s , G . F . and Donegan , E A . (Eds.) E d w a r d A r n o l d , L o n d o n , pp. 85-104. 160 Buchanan , R . M . , Aru lanandam, B P . and Me tzge r , D W . (1998) DL-12 enhances antibody responses to T-independent polysaccharide vaccines i n the absence o f T and N K cells . Journal o f Immuno logy 161, 5525-5533. Buchanan , T M . (1975) An t igen ic heterogeneity o f gonococca l p i l i . Journal o f Expe r imen ta l M e d i c i n e 141, 1470-1475. Buchanan , T M . and A r k o , R . J . (1977) Immuni ty to gonococca l infect ion induced by vaccina t ion w i t h isolated outer membranes o f Neisseria gonorrhoeae i n guinea pigs. Journal o f Infectious Diseases 135, 879-887. Buchanan , T M . and Hildebrandt , J F . (1981) Ant igen-spec i f ic serotyping o f Neisseria gonorrhoeae, characterization based upon pr inc ipa l outer membrane protein. Infection and Immuni ty 32, 985-994. Buchanan , T . M . , Eschenbach, D A , K n a p p , J S . and H o l m e s , K K . (1980) G o n o c o c c a l salpingitis is less l i ke ly to recur w i t h Neisseria gonorrhoeae o f the same pr inc ipa l outer membrane protein antigenic type. A m e r i c a n Journal o f Obstetrics and G y n e c o l o g y 138, 978-980. Carbonett i , N H . and Sparl ing, P F . (1987) M o l e c u l a r c l o n i n g and characterization o f the structural gene for protein I, the major outer membrane protein o f Neisseria gonorrhoeae. Proceedings o f the N a t i o n a l A c a d e m y o f Sciences o f the U n i t e d States o f A m e r i c a 84, 9084-9088. Chang , C . W . , Barber, L . , Ouyang , C , M a s i n , D . , B a l l y , M B . and M a d d e n , T D . (1997) P lasma clearance, b iodis t r ibut ion and therapeutic properties o f mitoxantrone encapsulated i n conventional and sterically s tabi l ized l iposomes after intravenous administrat ion i n B D F 1 mice. B r i t i s h Journal o f Cancer 75, 169-177. Chattopadhyay, A . and L o n d o n , E . (1984) F luoromet r ic determination of. cr i t ical mice l l e concentration avo id ing interference from detergent charge. A n a l y t i c a l B i o c h e m i s t r y 139, 408-412. Cherry , J . D . (1992) Pertussis: the trials and tr ibulations o f o l d and new pertussis vaccines. V a c c i n e 10, 1033-1038. C h o w , Y . - H . , Ch iang , B . - L . , L e e , Y . - L . , C h i , W . - K , L i n , W . - C , Chen , Y . - T . , and Tao , M -H . (1998) Development o f T h l and Th2 populat ions and the nature o f immune responses to hepatitis B virus D N A vaccines can be modulated by co-de l ivery o f var ious cytokine genes. Journal o f Immuno logy 160, 1320-1329. 161 Chris t ie , K . E . , Flaegstad, T . and Traavik , T. (1988) Character izat ion o f B K virus-specif ic antibodies i n human sera by western immunoblo t t ing : use o f a zwi t te r ionic detergent for restoring the ant ibody-binding capacity o f electroblotted proteins. Journal M e d i c a l V i r o l o g y 24, 183-190. C u l l i s , P R . and H o p e , M . J . (1991) P h y s i c a l properties and functional roles o f l ip ids i n membranes. In: B iochemis t ry o f L i p i d s . L ipopro te ins and Membranes . Vance , D . E . and Vance , J. (Eds.) E l sev ie r Science Publishers . N e w Y o r k . 1-41. C u l l i s , P R . , M a y e r , L . D . , B a l l y , M B , M a d d e n , T . D . and Hope , M . J . (1989) Generat ing and loading o f l iposomal systems for drug-del ivery applications. A d v a n c e d D r u g D e l i v e r y R e v i e w s 3, 267-282. D a l M o n t e , P . and Szoka , F C . Jr. (1989) Effect o f l iposome encapsulat ion on antigen presentation in vitro, compar ison o f presentation by peritoneal macrophages and B cel l tumors. Journal o f Immuno logy 142, 1437-1443. Danon , D , Golds te in , L , M a r i k o v s k y , Y . , and Skutelsky, E . (1972) U s e o f cat ionized ferri t in as label o f negative charges on ce l l surfaces. Journal Ultrastruct . Res . 38:500. de la M a z a , A . , L o p e z , O , Coderch , L . and Parra, J L . (1998) So lub i l i za t ion o f phosphat idylchol ine l iposomes by the amphoteric surfactant dodecyl betaine. Chemis t ry and Phys ics o f L i p i d s 94, 71-79. D i a z , J . - L . and Hecke l s , J . E . (1982) An t igen i c var ia t ion o f outer membrane protein II i n co lon ia l variants o f Neisseria gonorrhoeae P 9 . Journal o f Genera l M i c r o b i o l o g y 128, 585-591. Douglas , J .T. , L e e , M . D . and N i k a i d o , H . (1981) Pro te in I o f Neisseria gonorrhoeae outer membrane is a porin . F E M S M i c r o b i o l o g y Letters 12, 305-309. Driessen, A . J . M . and W i c k n e r , W . (1990) So lub i l i za t ion and functional reconsti tution o f the protein-translocation enzymes o f Escherichia coli. Proceedings o f the Na t iona l A c a d e m y o f Sciences o f the U n i t e d States o f A m e r i c a 87, 3107-3 111. Dubochet , J . , A d r i a n , M . , Chang , J.J . , H o m o , J . C , Lepaul t , J . , M c D o w a l l , A W . and Schultz , P . (1988) Cryo-e lec t ron microscopy o f v i t r i f ied specimens. Quarter ly R e v i e w i n B i o p h y s i c s 21 , 129-228. Ede lman , R . (1980) V a c c i n e adjuvants. R e v i e w s o f Infectious Diseases 2, 370-383. Epand , R . M . (1985) H i g h sensit ivity differential scanning calor imetry o f the bi layer to hexagonal phase transitions o f diacylphosphat idylethanolamines. Chemis t ry and Phys ics o f L i p i d s 36, 387-393. 162 F i l i o n , M C . and Ph i l l i p s , N C . (1997) T o x i c i t y and immunomodula to ry act ivi ty o f l iposomal vectors formulated w i t h cat ionic l ip ids toward immune effector cells. B i o c h i m i c a et B i o p h y s i c a A c t a 1329, 345-356. F inkels te in , M C . and Weissmann , G . (1979) E n z y m e replacement v i a l iposomes. Var ia t ions in l i p i d compos i t ion determine l iposomal integrity i n b io log ica l f luids. B i o c h i m i c a et B i o p h y s i c a A c t a 587, 202-216. F i ske , C M . and Subbarow, Y . (1925) The colour imetr ic determination o f phosphorus. Journal o f B i o l o g i c a l Chemis t ry 66, 375-400. Fletcher, J . N . , Zak , K . , V i r j i , M . and Hecke l s , J E . (1986) M o n o c l o n a l antibodies to gonococca l outer membrane protein I: locat ion o f a conserved epitope on protein I B . Journal o f Genera l M i c r o b i o l o g y 132, 1611-1620. Frasch, C . E . , M o c c a , L . F . and Karpas , A B . (1987) Appearance o f new strains associated w i t h group B meningococca l disease and their use for rapid vacc ine development. An ton ie van Leeuwenhoek. 53, 395-402. Freemont, D . H . , Stura, E . A . , Matsumura , M , Peterson, P . A . and W i l s o n , I A . (1995) Crys ta l structure o f an H - 2 K b - o v a l b u m i n peptide complex reveals the interplay o f pr imary and secondary anchor positions in the major h is tocompat ib i l i ty complex b ind ing groove. Proceeding o f the Na t iona l A c a d e m y o f Sciences o f the U n i t e d States o f A m e r i c a 92, 2479-2483. Gab izon , A . , Dagan, A . , Goren , D . , Barenholz , Y . and Fuks , Z . (1982) L i p o s o m e s as in vivo carriers o f adr iamycin: reduce cardiac uptake and preserved anti tumor act ivi ty i n mice. Cancer Research 42, 4734-4739. Garcon , N . M . J , and S ix , H R . (1991) U n i v e r s a l vacc ine carrier: l iposomes that provide T -dependent help to weak antigens. Journal o f I m m u n o l o g y 146, 3697-3702. G l i i c k , R . (1992) Immunopotentiat ing reconstituted inf luenza v i rosomes ( I R I V s ) and other adjuvants for improved presentation o f smal l antigens. V a c c i n e 10, 915-919. G l i i c k , R . , M i s c h l e r , R , Brantschen, S., Just, M . , Al thaus , B . and C r y z , Jr., S.J. (1992) Immunopotentiat ing reconstituted inf luenza virus v i rosome vaccine de l ivery system for immuniza t ion against hepatitis A . Journal o f C l i n i c a l Investigation 90, 2491-2495. Gomez-Gut ie r rez , J . , Rodr iguez-Crespo , I , Peterson, D L and Gavi lanes , F . (1994) Reconst i tu t ion o f hepatitis B surface antigen proteins into phosphol ip id vesicles. B i o c h i m i c a et B i o p h y s i c a A c t a 1192:45. Gomez-Gut ie r rez , J . , Rodr iguez-Crespo , I , Peterson, D L and Gavi lanes , F . (1995) Ant igen ic i ty o f hepatitis B surface antigen proteins reconstituted w i t h phosphol ipids . B i o c h i m i c a et B i o p h y s i c a A c t a 1233, 205-212. 163 Gotsch l ich , E C (1984) Deve lopment o f a gonorrhoea vaccine: prospects, strategies and tactics. B u l l e t i n o f the W o r l d Hea l th Organiza t ion 62, 671-680. Gotsch l i ch , E C , Seiff, M E . and B l a k e , M . S . (1988) Studies on gonococca l protein I. In: Bacter ia -Host C e l l Interaction. H o r w i t z , M A . , editor. A l a n R . L i s s , Inc., N e w Y o r k . 63-73. Gotsch l i ch , E C , Seiff, M . E . , B l a k e , M . S . and K o o m e y , M . (1987) P o r i n protein of Neisseria gonorrhoeae: c l on ing and gene structure. Proceedings o f the Na t iona l A c a d e m y o f Sciences o f the U n i t e d States o f A m e r i c a 84, 8135-8139. G o u l d , R . J . , Ginsberg , B . H . , and Spector, A . A . (1981) Effects o f oc ty l P-glucoside on insu l in b ind ing to solubi l ized membrane receptors. B iochemis t ry 20, 6776-6791. Grav i to , R M . and Rosenbusch, J .P. (1980) Three-dimensional crystals o f an integral membrane protein: an ini t ia l x-ray analysis. Journal o f C e l l B i o l o g y 86, 327-329. Grav i to , R . M . , Jenkins, J . , Jansonius, J . N . , Ka r l s son , R K . and Rosenbusch , J .P. (1983) X - r a y diffraction analysis o f matr ix por in , an integral membrane protein from Escherichia coli outer membranes. Journal o f M o l e c u l a r B i o l o g y 164, 313-327. Greco , F . , B l a k e , M . S . , Gotsch l i ch , E C and M a u r o , A . (1980) M a j o r outer membrane protein o f Neisseria gonorrhoeae forms channels in l i p i d b i layer membranes. Federat ion Proceedings 39, 1813-1817. Gregoriadis , G . (1990) Immunolog ica l adjuvants: a role for l iposomes. Immuno logy Today 11, 89-97. Gula t i , S., R i c e , P A , B l a k e , M . S . , Sarafian, S . K . , M o r s e , S . A . , Q u e n t i n - M i l l e r , M . J . and A r m i n j o n , F . (1991) A n t i b o d y Responses i n S i x Volunteers Immunized w i t h a G o n o c o c c a l Protein I Vacc ine . In: Neisser iae 1990-Proceedings o f the Seventh International Pathogenic Neisse r ia Conference (Ber l in) . Ach tman , M . , K o h l , P . , M a r c h a l , C , M o r e l l i , G . , Seiler, A . , Thiesen, B . (Eds.) B e r l i n , Wal t e r de Gruyter , 229-34. Hard ing , C . V . , C o l l i n s , D . S . , Slot, J . W . , Geuze , H . J . and Unanue, E R . (1991) L i p o s o m e -encapsulated antigens are processed i n lysosomes, recycled, and presented to T cells. C e l l 64, 393-401. Hecke l s , J .E . (1978) The surface properties o f Neisseria gonorrhoeae: topographical distributions o f the outer membrane protein antigens. Journal o f Genera l M i c r o b i o l o g y 108, 213-219. Hecke ls , J . E . (1981) Structural compar i son o f Neisseria gonorrhoeae outer membrane proteins. Journal o f Bac te r io logy 145, 736-742. 164 Helenius , A . and Simons, K . (1975) So lub i l i za t ion o f membranes by detergents. B i o c h i m i c a et B i o p h y s i c a A c t a 415, 29-79. Hje lmeland , L M . (1990) R e m o v a l o f detergents from membrane proteins. M e t h o d s i n E n z y m o l o g y 182, 277-282. H o p e , M . J . , B a l l y , M B , Webb , G . and C u l l i s , P R . (1985) Produc t ion o f large un i lamel la r vesicles by a rapid extrusion procedure. Character izat ion o f size distr ibution, trapped vo lume and abi l i ty to maintain a membrane potential. B i o c h i m i c a et B i o p h y s i c a A c t a 812, 55-65. Huang , C . - H . (1969) Studies on phosphat idylchol ine vesicles. Fo rmat ion and phys ica l characteristics. B iochemis t ry 8, 344-351. Jakob, T., Walke r , P S , K r i e g , A . M . , U d e y , M C . and V o g e l , J . C . (1998) A c t i v a t i o n o f cutaneous dendritic cells by C p G - c o n t a i n i n g ol igodeoxynucleot ides : a role for dendrit ic cel ls i n the augmentation o f T h l responses by immunos t imula tory D N A . Journal o f I m m u n o l o g y 161, 3042-3049. Jakob, T , Walker , P .S . , K r i e g , A . M . , v o n Stebut, E . , U d e y , M C . and V o g e l , J .C . (1999) Bacter ia l D N A and C p G - c o n t a i n i n g o l igodeoxynucleot ides activate cutaneous dendrit ic cel ls and induce H - 1 2 production: impl ica t ions for the augmentation o f T h l responses. International A r c h i v e s o f A l l e r g y & I m m u n o l o g y 118, 457-461. Jennings, R . and Erturk, M . (1990) Compara t ive studies o f H S V - 1 antigens solubi l i sed f rom infected cells by us ing non- ionic or zwi t ter ionic detergents. Journal M e d i c a l V i r o l o g y 31, 98-108. Jennings, R , Quas im, T. , Sharrard, R . M . , H o c k l e y , D . and Potter, C . W . (1988) Zwi t t e r ion ic detergent solubi l isa t ion o f H S V - 1 surface antigens. A r c h i v e s o f V i r o l o g y 98, 137-153. Jensen, P . E . (1997) Peptide b ind ing and antigen presentation by class II h is tocompat ib i l i ty glycoproteins. B i o p o l y m e r s 43, 303-322. Johnston, K . H . and Gotsch l i ch , E C . (1974) Isolation and characterization o f the outer membrane o f Neisseria gonorrhoeae. Journal o f Bac t e r io logy 119, 250-257. Johnston, K . H . , Ho lmes , K K . and Go t sch l i ch , E C . (1976) The serological c lassif icat ion o f Neisseria gonorrhoeae. I. Isolation o f the outer membrane complex responsible for serotypic specificity. Journal o f Exper imenta l M e d i c i n e 143, 741-758. Joiner, K . A . , Scales, R . , Warren , K . A . , Frank, M M . and R i c e , P . A . (1985b) M e c h a n i s m o f action o f b l o c k i n g i m m u n o g l o b u l i n G for Neisseria gonorrhoeae. Journal o f C l i n i c a l Investigation 78, 1765-1772. 165 Joiner, K . A . , Warren, K . A . , B r o w n , E . J , Swanson, J. and Frank, M M . (1983) Studies on the mechanism o f bacterial resistance to complement-mediated k i l l i n g . C 5 b - 9 forms h igh molecular weight complexes w i t h bacterial outer membrane constituents on serum-resistant but not on serum- sensitive Neisseria gonorrhoeae. Journal o f I m m u n o l o g y 131, 1443-1451. Joiner, K . A . , Warren , K . A . , Frank, M M . and R i c e , P . A . (1985a) B l o c k i n g i m m u n o g l o b u l i n G enhances complement consumpt ion and deposi t ion on Neisseria gonorrhoeae. In: The Pathogenic Neisseriae. Schooln ik , G . K . , editor. A m e r i c a n Socie ty for M i c r o b i o l o g y , Wash ing ton , D C . 431-434. Joiner, K . A . , Warren , K . A . , Hammer , C . and Frank, M M . (1985c) Bac te r ic ida l but not nonbactericidal C 5 b - 9 is associated w i t h dist inct ive outer membrane proteins i n Neisseria gonorrhoeae. Journal o f Immuno logy 134, 1920-1925. Jul iano, R . L . and Stamp, D . (1975) The effect o f particle size and charge on the clearance rates o f l iposomes and l iposome encapsulated drugs. B i o c h e m i c a l and B i o p h y s i c a l Research Communica t ions 63, 651-658. Just, M . , Berger, R . , Drechsler , H . , Brantschen, S. and G l u c k , R . (1992) A single vaccina t ion w i t h an inactivated hepatitis A l iposome vacc ine induces protective antibodies after only t w o weeks. V a c c i n e 10, 737-739. K n a p p , J .S. , Tam, M R , N o w i n s k i , R . C . , H o l m e s , K K . and Sandstrom, E . G . (1984) Serologica l c lassif icat ion o f Neisseria gonorrhoeae w i t h use o f monoc lona l antibodies to gonococca l outer membrane protein I. Journal o f Infectious Diseases 150, 44-48. Ko lchens , S., R a m a s w a m i , V . , Birgenheier , J . , Nett , L . and O ' B r i e n , D F . (1993) Quas i -elastic l ight scattering determination o f the size dis tr ibut ion o f extruded vesicles. Chemis t ry and Phys ics o f L i p i d s 65, 1-10. Kostense, S., Sun, W . H . , Cottey, R , Tay lor , S.F. , Ha rme l ing , S., Zander, D , Smal l , P . A . Jr. and Bender , B . S . (1998) Interleukin 12 adminis t ra t ion enhances T h l act ivi ty but delays recovery from influenza A virus infect ion in mice . A n t i v i r a l Research 38, 117-130. K u b y , J. (1997) Immunology . 3 r d edit ion. W . H . Freeman and Company , N e w Y o r k . Labora tory Centre for Disease Con t ro l . (1998) Sexual ly transmitted disease survei l lance i n Canada: 1995 annual report. C C D R 2 4 S 1 , 1-32. L a e m m l i , U . K . (1970) Cleavage o f structural proteins dur ing the assembly o f the head o f bacteriophage T 4 . Nature 227, 680-685. Lauter ia , S.F. , Kantz le r , G . B . , H i g h , P C , L e e , J D . and W a l d m a n , R H . (1974) A n attenuated influenza virus vaccine: reactogenicity, t ransmissibi l i ty , immunogenic i ty , and protective efficacy. Journal o f Infectious Diseases 130, 380-383. 166 Lee , S W . and Sung, Y . C . (1998) Immuno-s t imulatory effects o f bacterial-derived p lasmids depend on the nature o f the antigen i n intramuscular D N A inoculat ions. Immunology 94, 285-289. Lichtenberg , D . (1985) Character izat ion o f the so lubi l iza t ion o f l i p i d bi layers by surfactants. B i o c h i m i c a et B i o p h y s i c a A c t a 821, 470-478. Lichtenberg , D , Robson , R . J . and Dennis , E A . (1983) So lub i l i za t ion o f phosphol ip ids by detergents: structural and kinet ic aspects. B i o c h i m i c a et B i o p h y s i c a A c t a 737, 285-304. L i , J . , Zhang , L B , Y o n e y a m a , T , Y o s h i d a , H , Sh imizu , H , Y o s h i i , K , Hara , M , N o m u r a , T , Yosh iku ra , H , M i y a m u r a , T . and Hag iwara , A . (1996) Genet ic basis o f the neurovirulence o f type 1 pol ioviruses isolated from vaccine-associated paralytic patients. A r c h i v e s o f V i r o l o g y 141, 1047-1054. Lowther t , L A , K u , N O . , L i a o , I , Cou lombe , P . A . and Omary , M B . (1995) E m p i g e n B B : a useful detergent for so lubi l iza t ion and b iochemica l analysis o f keratins. B i o c h e m i c a l and B i o p h y s i c a l Research Communica t ions 206, 370-379. L y n c h , E C , B lake , M S , Gotsch l i ch , E C . and M a u r o , A (1984) Studies o f porins: Spontaneously transferred from who le cells and reconstituted f rom purif ied proteins o f Neisseria gonorrhoeae and Neisseria meningitidis. B i o p h y s i c a l Journal 45, 104-107. M a d d e n , T D . (1986) Current concepts in membrane protein reconstitution. Chemis t ry and Phys ics o f L i p i d s 40, 207-226. M a d d e n , T D . (1988) Protein reconstitution: methodologies and applications. International Journal o f B iochemis t ry 20, 889-895. M a d d e n , T D , Hope , M . J . and C u l l i s , P R . (1983) L i p i d requirements for coupled cytochrome oxidase vesicles. B iochemis t ry 22, 1970-1974. M a d d e n , T D . (1997) M o d e l membrane systems. In: Pr inc ip les o f M e d i c a l B i o l o g y . V o l u m e 7 A . Membranes and C e l l S igna l ing . Bi t tar , E E . and Bit tar , N . (Eds.) J A I Press Inc. pp. Greenwich , C o n n . 1-17. Mae land , J A . (1977) Immunochemis t ry o f the gonococcus . In: Gonorrhea: E p i d e m i o l o g y and Pathogenesis. Skinner, F A , W a l k e r , P D . and Smith , H . (Eds.) A c a d e m i c Press, N e w Y o r k . pp. 67-81. Malche rek , G , Gnau , V , Jung, G , Rammensee , H . - G . and M e l m s , A . (1995) Supermotifs enable natural invariant chain-derived peptides to interact w i t h many major his tocompat ibi l i ty complex-class II molecules. Journal o f Exper imen ta l M e d i c i n e 181, 527-536. 167 M a n d r e l l , R . E . and Zo l l inge r , W D . (1984) U s e o f a zwi t te r ionic detergent for the restoration o f the ant ibody-binding capacity o f electroblotted meningococca l outer membrane proteins. J. Immunol . Me thods 67, 1-11. Matsumura , M . , Freemont, D . H . , Peterson, P . A . and W i l s o n , L A . (1992) E m e r g i n g pr inciples for the recognit ion o f peptide antigens by M H C class I molecules . Science 257, 927-934. M a y e r , L . D . , Hope , M . J . and C u l l i s , P R . (1986) Ves i c l e s o f variable sizes produced by a rapid extrusion procedure. B i o c h i m i c a et B i o p h y s i c a A c t a 858, 161-168. M a y e r , L . D . , Hope , M . J , C u l l i s , P R . and Janoff, A . S . (1985) Solute distributions and trapping efficiencies observed in freeze-thawed mul t i lamel la r vesicles. B i o c h i m i c a et B i o p h y s i c a A c t a 817, 193-196. M a y e r , L . D . , Ta i , L . C . L . , K o , D S C , M a s i n , D , Ginsberg , R . S . , C u l l i s , P R . and B a l l y , M B . (1989) Influence o f ves ic le size, l i p i d compos i t ion , and drug- to- l ip id ratio on the b io log ica l activity o f l iposomal doxorub ic in in mice. Cancer Research 49, 5922-5930. Mcdade , R . L . Jr. and Johnston, K . H . (1980) Character izat ion o f serologica l ly dominant outer membrane proteins o f Neisseria gonorrhoeae. Journal o f Bac te r io logy 141, 1183-1191. M c G e e , Z . A . , M e l l y , M . A . , Gregg , C R , H o r n , R G , Tay lo r -Rob inson , D , Johnson, A P . and M c C u t c h a n , J . A . (1978) V i r u l e n c factors o f gonococc i : studies us ing human fa l lopian tube organ cultures. In: Immunobio logy o f Neisseria gonorrhoeae. B r o o k s , G . F . , Gotsch l i ch , E C , Ho lmes , K . K . , Sawyer, W D . and Y o u n g , F E . (Eds.) A m e r i c a n Socie ty for M i c r o b i o l o g y , Wash ing ton , D C . 258-262. Meh ta , R . T . and Lopez-Beres te in , G . (1989) Effect o f l iposome encapsulation on toxic i ty and antifungal act ivi ty o f polyene antibiotics. In: L iposomes i n the Therapy o f Infectious Diseases and Cancer. Berestein, G . and F id le r , I.J. (Eds . ) A l a n R . L i s s , Inc., N e w Y o r k . 263-273. M i l l e r , D . L . , Ross , E . M . , Alders lade , R , B e l l m a n , M H . and R a w s o n , N . S . B . (1981) Pertussis immunisa t ion and serious acute neurological i l lness i n chi ldren. B r i t i s h M e d i c a l Journal 282, 1595-1599. M i m m s , L . T . , Z a m p i g h i , G , N o z a k i , Y , Tanford , C . and Reynolds , J . A . (1981) Phospho l ip id vesicle formation and transmembrane protein incorporat ion us ing octy l glucoside. B iochemis t ry 20, 833-840. M o r g a n , A . J . , Epste in , M . A . and N o r t h , J .R. (1984) Compara t ive immunogen ic i ty studies on Eps te in-Bar r vi rus membrane antigen ( M A ) gp340 wi th novel adjuvants i n mice, rabbits, and cotton-top tamarins. Journal o f M e d i c a l V i r o l o g y 13, 281-292. 168 M o r s e , S .A . and B r o o k s , G . F . (1985) Neisseria gonorrhoeae taxonomy, co lony phenotypes, and disease. In: G o n o c o c c a l Infection. B r o o k s , G . F . and Donegan , E A . (Eds.) E d w a r d A r n o l d , L o n d o n , pp. 1-8. M o s m a n n , T R . and Coffman, R . L . (1989) T h l and T h 2 cells: different patterns o f lymphok ine secretion lead to different functional properties. A n n u a l R e v i e w o f I m m u n o l o g y 7, 145-173. M u k h l i s , F A , Jennings, R , Stephenson, T . J , Quas im, T . and Potter, C . W . (1986) Character izat ion and immunogenic i ty o f H S V - 1 antigens obtained f o l l o w i n g zwi t te r ionic detergent treatment. V a c c i n e 4, 191-196. Mutsaers , S .E . and Papadimi t r iou , J M . (1988) Surface charge o f macrophages and their interaction w i t h charged particles. Journal L e u k o c y t e B i o l o g y 44, 17-26. Mut t i l a inen , S , I d a n p a a n - H e i k k i l a , I , W a h l s t r o m , E , N u r m i n e n , M , M a k e l a , P H . and Sarvas, M . (1995) The Neisseria meningitidis outer membrane protein P I produced i n Bacillus subtilis and reconstituted into phosphol ip id vesicles el ici ts antibodies to native P I epitopes. M i c r o b i a l Pathogenesis 18 ,423-436 . N a i r , S. Z h o u , F , Reddy , R , Huang , L . and Rouse , B . T (1992) Soluble proteins del ivered to dendrit ic cells v i a pH-sensi t ive l iposomes induce pr imary cy to toxic T lymphocyte responses in vitro. Journal o f Exper imenta l M e d i c i n e 175, 609-612. Nakan i sh i , T , K u n i s a w a , J , Hayash i , A , Tsutsumi, Y , K u b o , K , Nakagawa , S , Fuj iwara , H , Hamaoka , T. and M a y u m i , T . (1997) Pos i t ive ly charged l iposome functions as an efficient immunoadjuvant in induc ing immune responses to soluble proteins. B i o c h e m i c a et B i o p h y s i c a Research Communica t ions 240, 793-797. Navar ro , J , T o i v i o - K i n n u c a n , M . and Racker , E . (1984) Effect o f l i p i d compos i t ion o f the calcium/adenosine 5'-triphosphate coup l ing ratio o f the C a 2 + - A T P a s e o f sarcoplasmic ret iculum. B iochemis t ry 23, 130-135. N I A I D - N a t i o n a l Institute o f A l l e r g y and Infectious Diseases. (1998) Fact Sheet: Gonorrhea. Bethesda, M D . N i c h o l l s , P . Hi ldebrandt , V . and W r i g g l e s w o r t h , J M . (1980) Orientat ion and react ivi ty o f cytochrome aa3 heme groups i n proteoliposomes. A r c h i v e s o f B i o c h e m i s t r y and B i o p h y s i c s 204, 533-543. N i k a i d o , H . (1993) Transport across the bacterial outer membrane. Journal o f Bioenerget ics and Biomembranes 25, 581-589. N k o w a n e , B M , Wass i l ak , S . G . F , Orenstein, W . A , Bar t , K . J , Schonberger, L B , H i n m a n , A R . and K e w , O . M . (1987) Vaccine-associa ted paralytic po l iomyel i t i s . U n i t e d States: 1973 through 1984. Journal o f the A m e r i c a n M e d i c a l A s s o c i a t i o n 257, 1335-1340. 169 Northfelt , D . W , Dezube, B . J . , Thommes , J A , M i l l e r , B . J , F i s c h l , M . A , F r i e d m a n - K i e n , A . , K a p l a n , L D , D u M o n d , C , M a m e l o k , R D . and Henry , D H . (1998) Pegyla ted- l iposomal doxorubic in , b l eomyc in , and vincr is t ine in the treatment o f A J D S - r e l a t e d K a p o s i ' s sarcoma: results o f a randomized phase III c l in ica l t r ial . Journal o f C l i n i c a l O n c o l o g y 16, 2445-2451. Oberholzer , T , M e y e r , E , A m a t o , I., Lus t ig , A . and M o n n a r d , P . - A . (1999) E n z y m a t i c reactions in l iposomes us ing the detergent-induced l iposome loading method. B i o c h i m i c a B i o p h y s i c a A c t a 1416, 57-58. Ostro, M . J . and C u l l i s , P R . (1989) U s e o f l iposomes as injectable-drug del ivery systems. A m e r i c a n Journal o f Hosp i t a l Pharmacy 46, 1576-1587. Paternostre, M T , R o u x , M . and R i g a u d J L . (1988) M e c h a n i s m s o f membrane protein insertion into l iposomes dur ing reconsti tution procedures i n v o l v i n g the use o f detergents. 1. So lub i l i za t ion o f large uni lamel lar l iposomes (prepared by reverse-phase evaporation) by tri ton X - 1 0 0 , oc ty l glucoside, and sod ium cholate. B iochemis t ry 27, 2668-2677. Pertmer, T M , Roberts, T R . and Haynes , J .R. (1996) Influenza v i rus nucleoprotein-specif ic immunog lobu l in G subclass and cytokine responses el ic i ted by D N A vaccinat ion are dependent on the route o f vector D N A delivery. Journal o f V i r o l o g y 70, 6119-6125. P lummer , F A , Chubb , H , Simonsen, J . N , Bos i re , M , Slaney, L , M a c l e a n , I, N d i n y a -A c h o l a , J O , W a i y a k i , P . and Brunham, R . C . (1993) A n t i b o d y to Rmp (outer membrane protein 3) increases susceptibi l i ty to gonococca l infection. Journal o f C l i n i c a l Investigation 91, 339-343. P lummer , F A , Chubb, H , Simonsen, J . N , K i m a t a , J . , Bos i re , M , N d i n y a - A c h o l a , J O , and N g u g i , E . N . (1989) E p i d e m i o l o g i c evidence for the development o f serovar-specific immuni ty after gonococca l infection. Journal o f C l i n i c a l Invest igat ion 83, 1472-1476. Rahman , A . , Kess ler , A . , M o r e , N , S ik i c , B , R o w d e n , G , W o o l l e y , P . and Schein, P S , (1980) L i p o s o m a l protection o f adr iamycin- induced card io toxic i ty i n mice. Cancer Research 40, 1532-1537. Rammensee, H . - G , Fa lk , K . and Rotzschke , O . (1993) Peptides naturally presented by M H C class I molecules. A n n u a l R e v i e w o f Immuno logy 11, 213-244. Rang , H P , Dale , M M , Ritter, J M . and Gardner, P . (1995) Pharmacology. C h u r c h h i l l L iv ings tone I n c , N e w Y o r k . pp. 718-743. Rao , M , Ro thwe l l , S W , Wassef, N . M , K o o l w a l , A . B , and A l v i n g , C . R . (1999) Tra f f i ck ing o f l iposomal antigen to the t rans-Golg i o f murine macrophages requires both l iposomal l i p i d and l iposomal protein. Exper imenta l C e l l Research 246, 203-211. 170 R a o , M , R o t h w e l l , S W , Wassef, N M , Pagano, R E , and A l v i n g , C R . (1997) V i s u a l i z a t i o n o f peptides derived from l iposome-encapsulated proteins i n the t rans-Golg i area o f macrophages. Immuno logy Letters 59, 99-105. Reddy , R , Z h o u , F , Na i r , S , Huang , L , and Rouse , B . T . (1992) In vivo cytotoxic T lymphocyte induct ion w i t h soluble proteins administered i n l iposomes. Journal o f Immuno logy 148, 1585-1589. R i c e , P . A , M c C o r m a c k , W . M . and Kasper , D L . (1980) Na tura l serum bactericidal act ivi ty against Neisseria gonorrhoeae isolates from disseminated, l oca l ly invasive, and uncompl ica ted disease. Journal o f I m m u n o l o g y 124, 2105-2109. R i c e , P . A , M c Q u i l l e n , D P , Gula t i , S , Jani, D . B , Wetz ler , L M , B l a k e , M . S . and Gotsch l ich , E C . (1994) Serum resistance o f Neisseria gonorrhoeae. D o e s it thwart the inf lammatory response and facilitate the t ransmission o f infection? A n n a l s N e w Y o r k A c a d e m y o f Sciences 730, 7-14. R i c e , P . A , M i l t o n , M R . and B l a k e , M . S . (1985) Immunog lobu l in G antibodies in normal human serum directed against protein III b lock k i l l i n g o f serum resistant Neisseria gonorrhoeae by immune human sera. In: The Pathogenic Neisseriae. Schooln ik , G . K , editor. A m e r i c a n Society for M i c r o b i o l o g y , Wash ing ton , D C . 427-430. R i c e , P . A , V a y o , H . E , M i l t o n , M R . and B l a k e , M . S . (1986) Immunog lobu l in G antibodies directed against protein III b lock k i l l i n g o f serum resistant Neisseria gonorrhoeae by immune serum. Journal o f Exper imen ta l M e d i c i n e 164, 1735-1748. Richards , R . L , A l v i n g , C R . and Wassef, N M . (1996) L i p o s o m a l subunit vaccines: effects o f l i p i d A and a luminum hydroxide on immunogenic i ty . Journal o f Pharmaceut ical Sciences 85, 1286-1289. Richards , R . L , Hayre , M D , H o c k m e y e r , W . T . and A l v i n g , C R . (1988) L iposomes , l i p i d A , and a luminum hydroxide enhance the immune response to a synthetic malar ia sporozoite antigen. Infection and Immuni ty 56, 682-686. R igaud , J L , Paternostre, M T . and Bluza t , A . (1988) Mechan i sms o f membrane protein insertion into l iposomes dur ing reconsti tution procedures i n v o l v i n g the use o f detergents. 2. Incorporation o f the l ight-dr iven proton pump bacteriorhodopsin. B iochemis t ry 27, 2677-2688. Roi t t , I , Brostoff, J . and M a l e , D . (Eds.) (1996) Immunology , 4 t h edit ion. M o s b y , Barce lona . Rosenqvist , E , Har thug, S , F r a h o l m , L O , H 0 i b y , E A , B e v r e , K . and Zo l l inge r , W . D . (1988) A n t i b o d y responses to serogroup B meningococca l outer membrane antigens after vaccinat ion and infection. Journal o f C l i n i c a l M i c r o b i o l o g y 26, 1543-1548. 171 Ruegg , C . L , Jaffe, R . I , Ros te r , B , Sadoff, J . C , Zo l l inge r , W D , L o w e l l , G . H . and Strand, M . (1990) Preparation o f proteosome-based vaccines. Corre la t ion o f immunogen ic i ty w i t h physical characteristics. Journal o f Immuno log i ca l M e t h o d s 135, 101-109. Sarafian, S . K , Tarn, M R . and M o r s e , S A . (1983) G o n o c o c c a l protein I-specific opsonic I g G in normal human serum. Journal o f Infectious Diseases 148, 1025-1032. Scherphof, G , Roerdink , F , Wai te , M . and Parks , J. (1978) Dis integrat ion o f phosphat idylchol ine l iposomes in plasma as a result o f interaction w i t h high-density l ipoproteins. B i o c h i m i c a et B i o p h y s i c a A c t a 542, 296-307. Schwarz, D , Gast, K , M e y e r , H . W , L a c h m a n n , U , C o o n , M . J . and Ruckpau l , K . (1984) Incorporation o f the cytochrome P-450 monoxygenase system into large uni lamel la r l iposomes us ing octylglucoside , especial ly for measurements o f protein diffusion in membranes. B i o c h e m i c a l and B i o p h y s i c a l Research Communica t ions 121, 118-125. Schwendener, R . A , L a g o c k i , P . A . and Rahman , Y E . (1984) The effects o f charge and size on the interaction o f uni lamel lar l iposomes w i t h macrophages. B i o c h i m i c a et B i o p h y s i c a A c t a 772, 93-101. Senior, J , Crawley , J . C . W . and Gregor iadis , G (1985) Tissue distr ibution o f l iposomes exhibi t ing long half-l ives in the c i rcula t ion after intravenous injection. B i o c h i m i c a et B i o p h y s i c a A c t a 839, 1-8. Shek, P . N , Y u n g i , B . Y . K . and Stanacev, N . Z . (1983) C o m p a r i s o n between mul t i lamel lar and uni lamel lar l iposomes in enhancing ant ibody formation. Immuno logy 49, 37-44. Skinner, M . A , Y a u n , S , Prestidge, R , Chuk , D , Watson , J D . and Tan , P . L . J . (1997) Immuniza t ion w i t h heat-ki l led Mycobacterium vaccae stimulates C D 8 + cytotoxic T ce l l specific for macrophages infected wi th Mycobacterium tuberculosis. Infection and Immuni ty 65 ,4525-4530 . Stahn, R , Schafer, H , K u n z e , M , M a l u r , J , Ladhoff , A . and Lachmann , U . (1992) Quantitative reconsti tution o f isolated inf luenza haemagglut inin into l iposomes by the detergent method and the immunogenic i ty o f haemagglut inin l iposomes. A c t a V i r o l o g i c a 36, 129-144. Stanker, L H , Vander laan, M and Juarez-Salinas, H (1985) One-step purif icat ion o f mouse monoc lona l antibodies f rom ascites f lu id by hydroxylapt i te chromatography. Journal o f Immunolog ica l M e t h o d s 76, 157-169. Stern, L . J . and W i l e y , D . C (1994) An t igen i c peptide b ind ing by class I and class II his tocompat ibi l i ty proteins. Structure 2, 245-251. Swanson, J. (1981) Surface-exposed protein antigens o f the gonococca l outer membrane. Infection and Immuni ty 34, 804-816. 172 Swanson, J , Robbins , K , Barrera , O , C o r w i n , D , Bos l ego , J , C iak , J , B l a k e , M . and K o o m e y , J M . (1987) G o n o c o c c a l p i l i n variants i n experimental gonorrhea. Journal o f Exper imenta l M e d i c i n e 165, 1344-1357. Szoka , F . Jr. and Papahadjopoulos, D . (1980) Compara t ive properties and methods o f preparation o f l i p i d vesicles ( l iposomes). A n n u a l R e v i e w o f B i o p h y s i c s and Bioeng inee r ing 9, 467-508. The r i en , H . - M , L a i r , D . and Shahum, E . (1990) L i p o s o m a l vaccine: influence o f antigen association on the kinetics o f the humora l response. V a c c i n e 8, 558-562. Tramont, E C . (1989) G o n o c o c c a l Vacc ines . C l i n i c a l M i c r o b i o l o g y R e v i e w 2, S74-S77. Tranum-Jensen, J , Christ iansen, K , Car lsen, J , B renze l , G . and V in t en , J. (1994) M e m b r a n e topology o f insu l in receptors reconstituted into l i p i d vesicles. Journal o f M e m b r a n e B i o l o g y 140, 215-223. Unanue, E R . (1984) Antigen-present ing function o f the macrophage. A n n u a l R e v i e w o f Immunology 2, 395-428. V i r j i , M . (1981) The effect o f antibody directed against outer-membrane antigens on the virulence o f a pilated variant o f Neisseria gonorrhoeae. F E M S M i c r o b i o l o g y Letters 12, 355-358. V i r j i , M . and Hecke l s , J E . (1988) Nonbac te r ic ida l antibodies against Neisseria gonorrhoeae: evaluation o f their b l o c k i n g effect on bacter icidal antibodies directed against outer membrane antigens. Journal o f General M i c r o b i o l o g y 134, 2703-2711. V i r j i , M , Fletcher, J . N , Zak , K . and Hecke l s , I E . (1987) The potential protective effect o f monoc lona l antibodies to gonococca l outer membrane protein I A . Journal o f Genera l M i c r o b i o l o g y 133, 2639-2646. V i r j i , M , Zak , K . and Hecke l s , I E . (1986) M o n o c l o n a l antibodies to gonococcal outer membrane protein I B : use i n invest igat ion o f the potential protective effect o f antibodies directed against conserved and type-specif ic epitopes. l ou rna l o f Genera l M i c r o b i o l o g y 132, 1621-1629. W a n g , L Y . and Frasch, C E . (1984) Deve lopment o f a Neisseria meningitidis group B serotype 2b protein vaccine and evaluat ion in a mouse model . Infection and Immuni ty 46, 408-414. W a r d , M E , Lambden , P R , Hecke ls , I E . and Watt , P . J . (1978) The surface properties o f Neisseria gonorrhoeae: determinants o f susceptibi l i ty to antibody complement k i l l i n g . Journal o f Genera l M i c r o b i o l o g y 108, 205-212. 173 W a r d , M E , Watt , P . J . and Robertson, J . N . (1974) The human fa l lopian tube: a laboratory mode l for gonococca l infection. Journal o f Infectious Diseases 129, 650-659. Wetz le r , L M , B l a k e , M . S . , and Gotsch l i ch , E C . (1988) Character izat ion and specif ici ty o f antibodies to protein I o f Neisseria gonorrhoeae produced by injection w i t h various protein I-adjuvant preparations. Journal o f Exper imen ta l M e d i c i n e 168, 1883-1897. Wetz ler , L M , B l a k e , M . S . , and Gotsch l i ch , E C . (1989a) Prote in I (Por) o f Neisseria gonorrhoeae as and immunogen: l iposomes, proteosomes, and the lack o f b l o c k i n g antibodies. In: Transactions or the A m e r i c a n A s s o c i a t i o n o f Phys ic ians . W a v e r l y Press, Ba l t imore . 78-90. Wetz le r , L M , B l a k e , M . S . , and Gotsch l i ch , E C . (1991) G o n o c o c c a l por in vaccine. In: Neisser iae 1990-Proceedings o f the Seventh International Pathogenic Ne i s se r i a Conference (Ber l in) . A c h t m a n , M , K o h l , P , M a r c h a l , C , M o r e l l i , G , Seiler, A , Thiesen, B . (Eds.) B e r l i n , Wal te r de Gruyter , 313-319. Wetz le r , L M , B l a k e , M . S . , Bar ry , K . and Go t sch l i ch , E C . (1992) G o n o c o c c a l por in vaccine evaluation: compar ison o f por proteosomes, l iposomes, and blebs isolated f rom Rmp deletion mutants. Journal o f Infectious Diseases 166, 551-555. Wetz le r , L M , B l a k e , M . S . , Gotsch l i ch , E C , B l a k e , M . S . and K o o m e y , J M . (1989b) The construction and characterization o f Neisseria gonorrhoeae l ack ing protein III in its outer membrane. Journal o f Exper imenta l M e d i c i n e 169, 2199-2209. Whi te , K , K r z y c h , U , Gordon , D . M , Porter, T G , Richards , R L , A l v i n g , C . R , Dea l , C D , Ho l l ingda le , M , Si lverman, C , Sylvester, D . R , B a l l o u , W . R . and Gross , M . (1993) Induct ion o f cy to ly t ic and antibody responses us ing Plasmodium falciparum repeatless c i rcumsporozoi te protein encapsulated in l iposomes. V a c c i n e 11, 1341-1346. W H O - W o r l d Hea l th Organiza t ion (1995) W H O ini t iat ive on H I V / A I D S and sexual ly transmitted infections ( H I S ) A n ove rv i ew o f selected curable sexually transmitted diseases. Geneva, Switzer land. Y o u n g , A C M , Nathenson, S . G . and Sacchettini , J .C . (1995) Structural studies o f class I major his tocompat ibi l i ty complex proteins, insight into antigen presentation. F A S E B Journal 9, 26-36. Y o u n g , D . - E , B l a k e , M , M a u r o , A . and Cohen , Z A . (1983) Properties o f the major outer membrane protein from Neisseria gonorrhoeae incorporated into mode l l i p i d membranes. Proceedings o f the Na t iona l A c a d e m y o f Sciences o f the U n i t e d States o f A m e r i c a 80, 3831-3835. Zak , K , D i a z , J . - L , Jackson, D . and Hecke l s , J . E . (1984) An t igen ic var ia t ion dur ing infect ion w i t h Neisseria gonorrhoeae, detection o f antibodies to surface proteins i n sera o f patients w i t h gonorrhea. Journal o f Infectious Diseases 149, 166-174. 174 Zhang , X , Jackson, J K . and Bur t , H M . (1996) Determinat ion o f surfactant cr i t ical mice l le concentration by a novel fluorescence depolar izat ion technique. Journal o f B i o c h e m i c a l and B i o p h y s i c a l Me thods 31, 145-150. Zo l l inge r , W . D . and M o r a n , E . (1991) M e n i n g o c o c c a l vaccines-present and future. Transactions o f the R o y a l Society o f T rop i ca l M e d i c i n e H y g i e n e 85, 37-43. Zo l l inge r , W . D , Bos lego , J , F r a h o l m , L O , Ray , J S , M o r a n , E E . and Brandt , B L . (1987) H u m a n bactericidal antibody response to meningococca l outer membrane protein vaccines. An ton ie van Leeuwenhoek. 53, 403-411. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0099437/manifest

Comment

Related Items