STUDIES OF PEPTIDE MIMICRY OF THE GROUP B STREPTOCOCCUS TYPE III

CAPSULAR POLYSACCHARIDE ANTIGEN

by

Rattanaruji Pomwised

A dissertation submitted in partial fulfillment of the requirements for the degree

of

Doctor of Philosophy

in

Microbiology

MONTANA STATE UNIVERSITY Bozeman, Montana

August 2007

© COPYRIGHT

by

Rattanaruji Pomwised

2007

All Rights Reserve

ii

APPROVAL

of a dissertation submitted by

Rattanaruji Pomwised

This dissertation has been read by each member of the dissertation committee and has been found to be satisfactory regarding content, English usage, format, citations, bibliographic style, and consistency, and is ready for submission to the Division of Graduate Education.

Dr. Mark Young

Approved for the Department of Microbiology

Dr. Tim Ford

Approved for the Division of Graduate Education

Dr. Carl A. Fox

iii

STATEMENT OF PERMISSION TO USE

In presenting this dissertation in partial fulfillment of the requirements for a doctoral degree at Montana State University, I agree that the library shall make it available to borrowers under rules of the Library. I further agree that copying of this dissertation is allowable only for scholarly purposes, consistent with “fair use” as prescribed in the U.S. Copyright Law. Requests for extensive copying or reproduction of this dissertation should be referred to ProQuest Information and Learning, 300 North

Zeeb Road, Ann Arbor, Michigan 48106, to whom I have granted “the exclusive right to reproduce and distribute my dissertation in and from microform along with the non- exclusive right to reproduce and distribute my abstract in any format in whole or in part.”

Rattanaruji Pomwised

August, 2007

iv

ACKNOWLEDGEMENTS

I would first like to thank my parents, Damnern and Samroum, and my brother,

Natthakon for all the unconditional love and support during my graduate studies. Their

love and care sustained me through this endeavor. Secondly, I would like to thank the

Royal Thai government for providing me scholarships since my first step in the U.S.

Next, I would like to thank my mentor, Dr. Seth Pincus, who introduced me to the world

of microbiology and immunology, for all his support, patience and kindness. I am

extraordinarily fortunate to have him as my advisor. This work would not have been

possible without his help and support. I would like to express my gratitude to Dr. Mark

Young for his lab collaboration and providing valuable suggestions to the project. I also

would like to thank the Young lab members especially Debbie Willits and Sue Brumfield

for being such great helps. I greatly appreciate Dr. Martin Teintze for his lab

collaboration and allowing me use his lab as my lab base during the move from Montana

to New Orleans. I would also like to thank the many members of the Pincus laboratory

both in Montana and Louisiana – their friendship and assistance has been invaluable. My

thanks also go to the Microbiology Department staff, especially Kari Cargill, who have done all of the foot-work in Montana necessary to allow me to complete my graduate work in Louisiana and in Thailand. I would like to thank my friends Emily and Barrett for packing me in their backpacks during hurricane seasons. I also would like to thank my best fiends, Waranya Thibodee, Ittiphong Leevongwat and Chirawan Thana, for all the help and support. Their help and care upheld me through all difficulties.

v

TABLE OF CONTENTS

1. INTRODUCTION ...... 1

Streptococcus ...... 1 Growth and Classification ...... 2 Group A Streptococcus...... 5 Pneumococcus (Streptococcus pneumoniae)...... 9 Other Streptococci ...... 11 Group B Streptococcus (GBS)...... 12 Identification and Morphology ...... 12 Lab Test ...... 12 Capsule and Group Antigen...... 13 Phase Variation ...... 14 GBS Infections...... 15 Disease Caused by GBS...... 15 Epidemiology and Transmission...... 16 Pathophysiology of Infection...... 17 Adherence ...... 18 Invasion...... 20 Host Injury ...... 23 Avoidance of Immune System...... 24 Induction of Sepsis Syndrome ...... 27 Use of Antibiotics to Prevent and Treat GBS Infection ...... 29 Protective Immunity to GBS...... 31 Development of a GBS Vaccine...... 33 Mimotopes ...... 36 Antigen...... 36 Epitope...... 37 Antigen-Antibody Interaction...... 38 Anti-Idiotype Antibody (Anti-Id) ...... 39 Polysaccharide Antigen ...... 40 Epitope Mapping...... 42 Mimotope...... 43 Phage Display Technology ...... 45 Peptide Mimicry of Carbohydrate Structure...... 48 Peptide Mimics of the Capsular Polysaccharide of Group B Streptococcus ...... 52 Conjugate Vaccines and Carriers...... 54 Vaccines to Microbial Polysaccharides ...... 54 Virus-Based Carrier Systems...... 57 Plant Virus-Based Vaccine ...... 58 Cowpea Mosaic Virus (CPMV)...... 60 vi

TABLE OF CONTENTS-CONTINUED

Cowpea Chlorotic Mottle Virus (CCMV) ...... 61 Research Goals...... 64

2. CCMV EXPRESSING PEPTIDES THAT MIMIC THE GROUP B STREPTOCOCCAL EPITOPE...... 66

Introduction...... 66 Materials and Methods...... 67 Engineering CCMV-Peptide S9 Chimera (Plant Expression) ...... 67 Verifying Positive Clones...... 68 In Vitro Transcription of CCMV cDNAs for Plant Inoculation ...... 69 Plant Inoculation...... 70 ELISA ...... 70 In Vitro Transcription and Translation ...... 71 Engineering CCMV-S9 Chimeras in Pichia pastoris...... 71 Protein Expression and Small Scale Time Course Experiment...... 73 Medium Scale Expression of CCMV-S9 Chimeras in P. Pastoris...... 74 Large Scale Protein Expression ...... 74 Chimeric Virus Purification...... 76 SDS PAGE, Western Blot, and Dot Blot Analysis...... 78 CCMV Wild-Type Plant Virus Purification ...... 79 Chemical Linking between CCMV Virus and S9 Peptide ...... 79 Sucrose Gradient Analysis...... 83 Results...... 84 Plant Expression System...... 84 Yeast Expression System...... 89 Chemical Linkage...... 102 Discussion...... 120

3. MURINE IMMUNE RESPONSE TO CCMV-CONJUGATED PEPTIDE MIMICS OF GBS CAPSULAR POLYSACCHARIDE...... 126

Introduction...... 126 Materials and Methods...... 128 Preparation of Antigens ...... 128 Immunization...... 129 Cell Purification and Culture ...... 131 Collection and Preparation of Blood ...... 133 Enzyme-Linked Immunosorbent Assays (ELISA) of Peptide-Specific Antibodies, GBS-Specific Antibodies, and CCMV-Specific Antibody ...... 133 ELISA of S9-Specific Antibodies Subclasses in Serum...... 134 vii

TABLE OF CONTENT-CONTINUED

ELISA of S9-Specific and GBS-Specific IgM Antibodies in Serum ...... 134 Assessment of Cytokines in Supernatants from Cultured Splenocytes and CD4 T Cells ...... 135 Results...... 136 Antigen Titration...... 136 Immune Response after Immunization...... 138 The Th1/Th2 Response Is Influenced by Antigen Conjugate Carriers...... 144 Immune Response after Injection with Heat-Killed Group B Streptococcus...... 146 Cytokine Production in Primed Splenocytes after Antigen Stimulation ...... 149 Discussion...... 151

4. PHAGE DISPLAY LIBRARY...... 156

Introduction...... 156 Materials and Methods...... 157 Peptide Library Construction...... 157 Plasmid Extraction...... 159 PCR...... 159 Single Stranded DNA (ssDNA) Extraction (373)...... 160 Phage Amplification ...... 161 Phage Purification...... 161 Phage Titration...... 162 Affinity Selection of Phage-Displayed Peptides ...... 163 ELISA and Antibodies...... 164 Conjugation of Peptides to Keyhole Limpet Hemocyanin (KLH) ...... 166 Results...... 167 Library Constructions ...... 167 Random and Affinity Selection of Phage ...... 169 Specificity and Avidity of Phage-Displayed Peptides...... 171 Specificity and Avidity of Peptides ...... 178 Specificity of KLH-Peptides Conjugates...... 184 Sequence Analysis of Peptides ...... 186 Discussion...... 189

5. CONCLUSION...... 196

REFERENCES CITED...... 201

viii

LIST OF TABLES

Table Page

1. 1 Characteristics of Medically Important Streptococci Causing Important Disease ...... 1

1. 2 Characteristics of Medically Important Streptococci Infrequently Associated with Disease ...... 2

1. 3 Summary of Peptides that Mimic Carbohydrate Structure...... 50

2. 1 Elisa Result of Leaf Crude Extract ...... 87

3. 1 Amino Acid Sequences of Peptides...... 133

4. 1 Mabs Used in This Study...... 166

4. 2 Summarized DNA Sequencing Analysis...... 170

4. 3 Summary of Binding Activity and Sequence of Peptides Which Have One Amino Acid Different from the Original S9 Peptide...... 187

4. 4 Summary of Binding Activity and Sequence of Peptides Which Have More than One Amino Acid Different from the Original S9 Peptide ...... 188

ix

LIST OF FIGURES

Figure Page

2. 1 Chemical Reaction between SPDP, CCMV and S9 ...... 81

2. 2 Chemical Reaction between Sulfo-SMCC, CCMV and S9...... 82

2. 3 PCR Analysis Recombinant Constructs...... 85

2. 4 RNA Products from In Vitro Transcription of pCC1, pCC2, pCC3-wt and Recombinant pCC3-63-S9, pCC3-102-S9, pCC3-114-S9, pCC3-129-S9, and pCC3-161-S9 ...... 86

2. 5 Introduction T7 RNA Polymerase to pCC3 Recombinant DNA...... 88

2. 6 In vitro Translation of RNA Construction...... 89

2. 7 Introduction EcoR1 and Sal1 Restriction Enzyme Sites to the Ends of pCC3-wt, pCC3-63-S9, pCC3-102-S9, pCC3-114-S9, pCC3-129-S9 and pCC3-161-S9 by PCR...... 90

2. 8 PCR Products from Positive Recombinant E. coli Clones ...... 91

2. 9 PCR Products from Recombinant Yeast Pichia pastoris Chromosomal DNA ...... 92

2. 10 In vitro Translation from pPicZa-pCC3-S9 Constructs...... 93

2. 11 Small Scale Time Course Protein Expression by Dot Blot Analysis...... 94

2. 12 Silver Staining and Western Blot Analysis of Small Scale Expression Experiment in CCMV-S9 Transformed Yeasts ...... 95

2. 13 Medium Scale CCMV-S9 Coat Protein Expression Experiment ...... 97

2. 14 Large Scale CCMV-S9 Chimeric Construct Expression Experiment ...... 99

2. 15 Virus Purification with or without the Presence of Urea...... 100

2. 16 CCMV-wt Viral Particle...... 102

2. 17 Cross-Linking between Cys of S9 and Cys of Mutant A163C...... 103

x

LIST OF FIGURES-CONTINUED

Figure Page

2. 18 Product of ssCCMV and S9 Peptide Cross-Linked via the spdp Linker...... 104

2. 19 The Conjugation Product of ssCCMv-S9 via spdp Linkers at Different pH Conditions and Different spdp Concentrations...... 105

2. 20 ssCCMV-S9 Conjugation Products via spdp Linkers at Different Peptide Concentrations...... 106

2. 21 Characterization of CCMV-S9 by Sucrose Gradient Sedimentation...... 107

2. 22 Conjugation between SubE and S9 via spdp Linkers ...... 109

2. 23 Conjugation Product SubE-spdp-S9 in Sucrose Gradient Velocity...... 110

2. 24 The Conjugation Products Between ssCCMV and S9 Peptide via Sulfo-smcc Linker with and without Iodoacetamide Treatment (IA treatment) ...... 112

2. 25 Time Course Experiment and Temperature Effect on Coupling Rreaction ...... 114

2. 26 Effect of Different Concentration of Sufo-smcc: 25, 50, 100 and 200 Molar Excess to the ssCCMV Virus in Cross-Linking Reaction ...... 116

2. 27 Effect of Peptide Concentration on Cross-Linking...... 117

2. 28 Conjugation Product ssCCMV-sulfo-smcc-S9 in Sucrose Gradient ...... 119

3. 1 Comparison of CCMV Virus Antigen in Different Antigen Preparations...... 136

3. 2 Comparison of S9 Peptide in CCMV-S9, SubE-S9, CPMV-S9, CPMV and KLH-S9 Antigens ...... 137

3. 3 Antibody Response of Individual Mice to S9 Peptide, CCMV and GBS...... 139

xi

LIST OF FIGURES-CONTINUED

Figure Page

3. 4 Antibody Response of Individual Mice to S9 Peptide CCMV and GBS...... 141

3. 5 Antibody Response of Individual Mice to S7 Peptide...... 142

3. 6 Antibody Response of Individual Mice to S9-2 Peptide ...... 143

3. 7 Antibody Response of Individual Mice to S11 Peptide...... 143

3. 8 IgG Subclass Identification from the First Immunization Experiment...... 144

3. 9 IgG Subclass Identification from the Second Immunization...... 145

3. 10 GBS-Specific IgG Antibodies after Injection of GBS...... 146

3. 11 S9-Specific IgG Antibody after Injection of GBS...... 147

3. 12 GBS-Specific IgM and S9-Specific IgM Antibodies after Injection of GBS ...... 148

3. 13 Cytokine production in primed splenocytes 3, 5 and 7 days after antigen stimulation ...... 150

4. 1 Construction of Randomly Modified S9 DNA ds Oligomer ...... 167

4. 2 Modified S9 DNA Fragment Library ...... 168

4. 3 PCR Analysis of Transfected Clones...... 169

4. 4 Direct Binding of Phage by S9 and Anti-M13 Antibodies (Set I)...... 172

4. 5 Direct Binding of Phage by S9 and Anti-M13 Antibodies (Set II)...... 173

4. 6 Direct Binding of Phage by S9 and Anti-M13 Antibodies...... 174

4. 7 Capture ELISA...... 175

4. 8 Titration of mAb binding to GBS ...... 177

4. 9 Inhibition Assay by Phages of S9 mAb Binding to GBS ...... 178

xii

LIST OF FIGURES-CONTINUED

Figure Page

4. 10 Binding Activities of Peptides to S9 mAb...... 180

4. 11 Inhibition Assay by Peptides of S9 mAb Binding to GBS ...... 181

4. 12 Inhibition Assay by Peptides of SIIIV18.C2 mAb Binding to GBS...... 182

4. 13 Inhibition Assay by Peptides of SIIIS8.C3 mAb Binding to GBS ...... 183

4. 14 Binding Activity of KLH-Peptide Conjugates by a Panel of Anti-GBS mAb ...... 184

4. 15 Inhibition Assay by KLH-Conjugates of S9 mAb Binding to GBS ...... 185

4. 16 Views of S9 mAb Bound Structure of the Peptide FDTGAFDPDWPA (Peptide S9) (277)...... 192

xiii

ABSTRACT

Capsular polysaccharide (CPS) of Streptococcus group B (GBS) is a poor immunogen and functions as T cell independent antigen, eliciting low IgG antibody with deficient immunologic memory. We previously identified a peptide, S9, which mimics CPS of type III GBS. Here we have taken steps to develop the mimetic peptide as a vaccine against GBS group III. We enhanced the immunogenicity of the peptide by presenting it on the coat protein of Cowpea Chlorotic Mottle Virus (CCMV). And we searched for better mimetic peptides by constructing a secondary phage display library. To accomplish the first goal, DNA encoding S9 was cloned into five constructions CCMV coat protein loops using recombinant DNA techniques. The results indicated that inserting the S9 peptide sequence into CCMV coat protein loops disrupted virus and virus-like particle assembly. Therefore the S9 peptide was conjugated to CCMV coat protein using chemical linkers. The CCMV-S9 conjugation products remained intact as monomer virions. Mice were immunized with the CCMV-S9, SubE-S9 (a mutant that does not assemble virions), CPMV-S9 (S9 conjugated to Cowpea Mosaic Virus) and S9 conjugated to the carrier protein KLH, with and without Freund’s adjuvant. The CCMV- S9, CPMV-S9 and KLH-S9 induce anti-S9 antibody even without the adjuvant whereas SubE-S9 induced an anti-S9 response only with adjuvant. The CCMV-S9 and CPMV-S9 predominantely induced a Th1 response with antigen-specific IgG2a and IFN-γ production, whereas the KLH-S9 predominantly induced a Th2 response with antigen- specific IgG1 and IL4/IL10 production. To accomplish the second goal, a DNA sublibrary was designed to have approximately one mutation in each displayed peptide sequence. Peptides with higher affinity to the S9 mAb were identified by affinity selection. ELISA analysis from randomly selected phage clones indicated that amino acid residues 3-5 and 7-10 of S9 peptide are important in specific binding activity to S9 monoclonal antibody. These studies identified peptides with greater affinity for the selecting antibody, i.e. enhanced antigenicity.

1

INTRODUCTION

Streptococcus

The streptococci are gram-positive spherical bacteria that characteristically form pairs or chains during growth. They are a heterogeneous group of organisms that include commensal bacteria as well as virulent pathogens. These organisms are of agricultural and medical importance. Major human pathogens include group A streptococci (S. pyogenes), group B streptococci (S. agalactiae), and the pneumococcus (S. pneumoniae), but other species are capable of causing disease. Table 1.1 and 1.2 list the major characteristics of the principal streptococcal species (2).

Table 1. 1 Characteristics of Medically Important Streptococci Causing Important Disease Group-specific Biochemical substance Reactions, Common and Hemolysis Habitat Name (Lancefield Important Important Diseases classification) Laboratory Criteria Streptococcus A Beta Throat, skin PYR test positive, Pharyngitis, pyogenes inhibited by impetigo, rheumatic bacitracin fever, glomerulonephritis Streptococcus B Alpha and Female Hippurate Neonatal sepsis and agalactiae Beta genital tract, hydrolysis, CAMP- meningitis colon positive (Christie, Atkins, Munch- Peterson test) Streptococcus None Alpha Throat Susceptible to Pneumonia, pneumoniae optochin. Colonies meningitis, soluble in bile, endocarditis Quellung reaction- positive Viridans Usually not Alpha, Mouth, Optochin-resistant. Dental caries (S streptococci typed or none throat, Colonies not mutans), untypable colon, soluble in bile. endocarditis, female Carbohydrate abscesses genital tract fermentation patterns 2

Table 1. 2 Characteristics of Medically Important Streptococci Infrequently Associated with Disease Group-specific Biochemical substance Associated Hemolysis Habitat Reactions, Important Name (Lancefield Diseases Laboratory Criteria classification) Enterococcus D None, Colon Growth in presence of Abdominal faecalis (and other alpha bile, hydrolyze abscess, urinary enterococcus) (Now esculin, growth in tract infection, is grouped into 6.5% NaCl, PYR- endocarditis genus Enterococcus positive (hydrolysis (3)) of L-pyrrolidonyl-2- naphthlamide) Streptococcus bovis D None Colon Growth in presence of Endocarditis, (nonenterococcus) bile, hydrolyze common blood esculin, no growth in isolate in colon 6.5% NaCl, degrades cancer starch Streptococcus F (ACG) and Beta Throat, Small colony variants Pyogenic anginosus untypable colon, of beta-hemolytic infections, female species. Group F are including brain genital bacitracin-resistant abscesses tract and PYR-negative. Peptostreptococcus None None, Mouth, Obligate anaerobes Abscesses (with (many species) alpha colon, multiple other female bacterial genital species) tract

Growth and Classification

Streptococci characteristically form chains of two or more organisms. The lengths of the chains vary and depend upon environmental factors. They lose their gram-positive appearance and appear to be gram-negative as the culture ages. They are facultative anaerobes and grow best at 37°C. Most streptococci produce a capsule resulting in mucoid colonies. Variants of the same strains may show different colony forms. For example, in group A streptococci, strains which produce much M protein appear as matte colonies, and strains which produce little M protein form glossy colonies. Phase variation in colony morphology has been described for streptococcal species (4-8). 3

No one system suffices to classify all streptococci. Therefore, their classification is based on different methods: hemolysis pattern, biochemical, physiological and

morphological characteristics, and immunological grouping according to Lancefield’s

method. This has led to a somewhat confusing nomenclature that includes species names, immunological grouping (e.g. group A streptococci), and/or hemolysis pattern. The

hemolysis pattern is the effect that the bacteria have on sheep red blood cells when grown

on blood agar. Many streptococci hemolyze red blood cells in vitro in varying degrees.

Complete disruption of erythrocytes is called β-hemolysis, resulting in a clear zone

around a colony. Incomplete lysis of erythrocytes with a green pigment zone around a

colony is called α-hemolysis. The formation of a greenish zone around colonies on blood

agar is due to discoloration and loss of potassium from the red cells and peroxide

production by bacteria (9). Hemolysis is the result of the secretion of soluble hemolysins;

examples of streptococcal hemolysins are streptolysin O and S of group A streptococci.

Streptolysin O is an antigenic protein (MW 60,000) that is hemolytically active in the

reduced state but inactive in the presence of oxygen (10). An antibody response to streptolysin O suggests either recent infection with streptococci or persistently high

antibody levels due to an exaggerated immune response in a hypersensitive person (11).

Streptolysin S is not antigenic. It may be inhibited by a nonspecific inhibitor present in

human sera (9, 12).

Speciation of streptococci is based on biochemical reactions. Biochemical tests

include sugar fermentation reactions, tests for the presence of enzymes, and tests for

susceptibility or resistance to certain chemical agents. PYR is the test for hydrolysis of L- 4

pyrrolidonyl-2-naphathylamide. Enzymatic hydrolysis of L-pyroglutamic acid β-

naphthylamide releases free β-naphthylamine, which is detected and shown by a color

change after adding hydrochloric acid (13). This test distinguishes group A Streptococcus

from other β-hemolytic streptococci. The physiological tests include CAMP test

(Christie, Atkins, Munch-Peterson test). It was designed to aid in the identification of S.

agalactiae. This test relies on the fact that most strains produce a diffusible, extracellular

compound that will, in conjunction with a specific β -hemolysin of Staphylococcus

aureus, cause complete lysis of sheep red blood cells in an agar medium (14, 15). The

morphological tests include the Quellung reaction tests for the presence of capsule. The

test principle is based upon the fact that capsule can prevent India ink from penetrating

into bacteria. Encapsulated cells appear to have halo around them. The reaction between

bacteria and antiserum raised against the bacterial capsule causes capsular swelling and

make it easily visualized in India ink suspension (16). Streptococcal species may also be

distinguished on the basis of starch hemolysis. Bacteria are cultivated on a plate

containing starch. Gram’s iodine is then flooded onto the plate. Clear zones around

colonies that hydrolyze starch will be observed on a blue background. Group D degrades

starch while group A, B and F don’t (17). The carbohydrate fermentation test is used to

differentiate bacteria based on their ability to oxidize or ferment specific sugars. For

example S. anginosus (viridans streptococcus) is capable of a 2,3-butanediol fermentation

as a continuation of mixed-acid fermentation (18). Group E streptococci can produce acid

from lactose but are unable to ferment trehalose and sorbitol (19). 5

The Lancefield classification is based upon immunologic differences among the group carbohydrate structures of streptococci. Classically this is performed using precipitin reactions between hot acid or enzyme extracts of the bacteria and specific antisera (20, 21). Using this method, Lancefield characterized streptococci into groups A

through T. The serologic specificity of the group-specific carbohydrate is determined by

an amino sugar. For group A streptococci, this is rhamnose-N-acetylglucosamine; for

group B, rhamnose-glucosamine polysaccharide, for group C, rhamnose-N-

acetylgalactosamine; for group D, glycerol teichoic acid containing D-alanine and

glucose; for group F, glucopyranosyl-N-acetlygalactosamine.

Group A Streptococcus

Most streptococci that contain the group A antigen are S. pyogenes. These

organisms produce a number of factors that may serve to enhance their virulence and

cause disease. They can be further classified into serotypes based on antigenicity of M

protein (22, 23).

M protein is a major virulence factor of group A streptococci. It appears as hair-

like projections of the streptococcal cell wall (24). Group A streptococci which lack M

protein are not virulent (25). Antibody to M protein is type specific. M protein binds to

fibrinogen, fibrin and their degradation products to coat the bacterial surface in order to

block complement deposition on their surface (26, 27). The secreted M protein can bind

to fibrinogen. The aggregates bind to integrin receptors on the surface of

polymorphonuclear leukocytes (PMN) leading to the activation of the host defense

mechanism of these cells, including the secretion of toxic oxygen metabolites, a variety 6

of proteolytic and glycolytic enzymes, and proinflammation cytokines, such as tumor

necrosis factor and IL-6 (27, 28). This in turn causes damage to host tissues leading to bacteria invasion into host cells. Systemic activation of this proinflammatory pathway

may lead to toxic shock syndrome.

The capsule, which is composed of hyaluronic acid and containing repeating units

of glucuronic acid and N-acetylglucosamine is considered to be an antiphagocytic

structure (29). The capsule is also an important adherence factor because it binds to

CD44 on epithelial cells and interferes with the attachment of phagocytes (29-32).

Group A streptococci also produce plasminogen-binding proteins: glyceraldehydes-3-phosphate dehydrogenase, enolase, and streptokinase (33-38). These

proteins bind to extracellular matrix metalloproteases or collagenases to facilitate tissue

invasion. This strategy enhances bacterial invasion through normal tissue barriers.

Streptococcus group A produces C5a peptidase, which cleaves six amino acids

from the carboxy terminus of C5a, which is the chemotactic factor generated by the

complement cascade (39). This event inhibits the recruitment of phagocytic cells to the

site of infection.

Group A streptococci also posses streptococcal proteinase (streptococcal exotoxin

B) which is an extracellular cysteine proteinase. It can convert IL-1β precursor to active

IL-1β, an inflammatory cytokine (40-42). The inflammatory cytokines in turn injure host

cells leading to increasing bacterial invasion (43). Group A streptococci produce a variety

of toxins such as streptolysins S and O and pyrogenic exotoxins A, B and C and also

mitogens such as streptococcal superantigen (SSA), streptococcal mitogenic exotoxin Z 7

(SMEZ). Streptolysins S and O are hemolysins, which lyse red blood cells and possibly other cell types. The three streptococcal pyrogenic exotoxins (SPEs) A, B and C are T cell superantigens, which enhance delayed-type hypersensitivity and suppress the antibody response (9). The superantigens are capable of binding to the beta chain (Vβ- chain) of a characteristic set of T-cell receptors and also to the MHC class II molecule expressed on resting B cells, monocytes and dendritic cells. Binding to the T-cell receptor and MHC class II molecule causes T cells to proliferate and release inflammatory cytokines involved in toxic shock syndrome(44-46).

Group A streptococci live on human skin and mucous membranes and can spread from person to person. The first step in the establishment of a streptococcal infection is the adherence of the bacteria to epithelial cells. Lipoteichoic acid (LTA) is an important adhesin of group A streptococci (47). LTA can interact specifically with fibronectin, a host protein that coats the epithelial cells of the oropharynx (48). The LTA molecules act as an adhesin by bridging molecules on the streptococcal surface, such as the M protein, with cell surface proteins (49). LTA interacts with the streptococcal proteins through its negatively charged polyglycerolphosphate backbone, which is attracted to positively charged residues of the surface proteins. The lipid moiety of LTA projects outward and interacts with hydrophobic fatty acid-binding sites on fibronectin and epithelial cells (49,

50). The spread of organisms depends on how the infection is acquired. Infection of the skin or mucous membranes may be localized and associated with necrosis and abscess formation. However the bacteria are also capable of rapid spread through tissues, earning it the lurid nickname “flesh eating bacteria”. The spread of bacteria occurs due to the 8

secretion of digestive enzymes including proteases, hyaluronidase, DNAase and

streptokinaes (51). The streptokinase binds to host plasminogen degrading it to plasmin

and leading to fibrin degradation (34, 35). Streptococcus Group A evokes an intense

inflammatory response in tissues, resulting from the production of proinflammatory

cytokines from PMNs and monocytes.

Group A Streptococci cause disease in several ways. They may cause damage by

direct infection: “Strep” throat, pyoderma (skin infections), necrotizing fasciitis and

septicemia. The elaboration of streptococcal toxins causes scarlet fever and toxic shock

syndrome (23, 44, 52). There may also be nonsuppurative sequelae of streptococcal

infections such as acute poststreptococcal glomerulonephritis and acute rheumatic fever.

The toxin-mediated and post-infectious syndromes result from inappropriate stimulation

of the immune system. Streptococci possess a number of antigens that cross-react with human tissues; heart muscle and valvular connective tissue (23, 45, 53, 54). Some of these epitopes are an integral part of M protein molecule. Antibodies to certain purified

M proteins react with myosin, phosphorylase and several other unidentified proteins in heart tissue (55-57). Streptococcal antigens may also evoke cross-reactive T cells. These antigens stimulate the production of cytotoxic T cells active against cultured cardiac myocytes. These cardiac reactive antibodies and T cells may be involved in the pathogenesis of rheumatic fever. The toxin SPEs are potent mitogens; they stimulate T- cell proliferation. The SPEs are superantigens interacting with conserved regions of the V domain of the T-cell receptor rather than with hypervariable regions (46, 58, 59). 9

Therefore, polyclonal activation results in an inappropriate immune response that may be

the major cause of the toxic shock syndrome.

Pneumococcus (Streptococcus pneumoniae)

Pneumococcus is placed in the viridans group. Lancefield grouping is not useful

for discriminating between the viridans streptococci therefore subdivision of the viridans

is based on biochemical and physiological characteristics. The optochin susceptibility test

is used to differentiate the pneumococci from other viridans.

Pneumococci are the major cause of bacterial pneumonia. They are surrounded by

polysaccharide capsules. The polysaccharide capsule is a potent antiphagocytic factor.

Antibodies against capsular polysaccharide have differentiated pneumococci into 90 serotypes (60). Antibodies against capsular polysaccharide are not protective (61).

Pneumococci also contain C-substance, which contains teichoic acid, a part of the cell

wall. C-substance reacts with C-reactive protein, a nonantibody β globulin from serum,

which acts as an acute phase reactant. The complex of C-substance and C-reactive protein

induces the complement cascade, leading to the release of inflammatory mediators and

opsonizing the organisms to enhance phagocytosis (62-64). Pneumococci also contain

pneumococcal surface protein A (PspA), which is located on the cell wall. PspA is a

surface protein with variable molecular size ranging from 67 to 99 kDa. It appears to

protect against the host complement system (65). PspA is highly polar, which results in

capsular charge stabilization through the electropositive end of the molecule and

prevention of complement activation through the predominant electronegative part of

PspA (65). Pneumococcus possesses hyaluronate lyase, which is part of a group of 10

enzymes called hyaluronidases (66, 67). The hyaluronidase enzyme facilitates invasion

by breaking down the extracellular matrix. The bacteria also contain pneumolysin, which

is a cytoplasmic enzyme that is released due to the action of surface pneumococcal

autolysin (68). The enzyme is cytotoxic to ciliated bronchial epithelial cells. It disrupts

tight junctions and the integrity of the bronchial epithelial monolayer. Due to the

enzyme’s function, the ability of ciliated bronchial cells to clear mucus from the lower

respiratory tract is reduced. The cytotoxic activity of pneumolysin can directly inhibit phagocyte and immune cell function, which leads to suppression of the host inflammatory and immune responses (69).

Humans are reservoirs of pneumococci. Transmission occurs from direct contact

and also airbone. Colonization by bacteria occurs in the nasopharynx (70). The outcome

of colonization is determined by the bacterial virulence and the efficiency of host defense

mechanisms. Pneumococcal adherence to host cells has been proposed to be a two-step

process. The first step is targeting an anatomic niche of the host, such as the nasopharynx,

and binding to the host cell-surface glycoconjugates on respiratory epithelial and

endothelial cells (68). The second step involves cytokine production by the host cell. This

cytokine activation results in expression of novel glycans on the surface of activated cells

and increased pneumococcal adherence (71-73).

Pneumococci cause life-threatening diseases such as pneumonia, bacteremia and

meningitis. Disease rates are high in young children, the elderly and patients with

predisposing conditions such as asplenia, chronic medical conditions or 11

immunosuppressive illnesses, particularly AIDS. Vancomycin and penicillin are the

drugs of choice for treatment.

Other Streptococci

Group C and G streptococci occur sometimes in the nasopharynx and may cause

sinusitis, bacteremia or endocarditis (74-76). Group D streptococci are the divided into

two groups, enterococci and nonenterococci. Current classification groups enterococci into a new genus Enterococcus based on 16 rRNA gene sequence analyses (3). The enterococci are normal flora of the intestinal tract and genitourinary tract. They can grow in high salt concentrations and in detergent including 60% bile (77). Most infections are urinary tract infections. Antibiotic synergism (combination use of different antibiotics) is recommended for Enterococcal infection (78). The nonenterococcal species do not grow in high salt concentrations and are sensitive to penicillin (79). Group F streptococci such as S. anginosus and S. milleri are part of the normal flora. Viridans streptococci include

S. mutans, S. salivarius and S. mitis (80). Typically they are α-hemolytic but they may be nonhemolytic. The viridans streptococci are the most prevalent members of the normal flora of the upper respiratory tract and are important for the healthy state of the mucous membranes there. They may reach the bloodstream as a result of trauma and are a principal cause of endocarditis on abnormal heart valves. Some viridans streptococci such as S. mutans synthesize large polysaccharides such as dextrans or levans from sucrose and contribute importantly to the genesis of dental caries.

12

Group B Streptococcus (GBS)

Group B streptococci cause serious disease in newborn infants and in

immunocompromised adults. Despite the fact that the bacteria are exquisitely sensitive to

antibiotics, the aggressive nature of infection leads to high morbidity and mortality. The

development of a vaccine to prevent infection in targeted groups is considered a public

health priority (81).

Identification and Morphology

Lab Tests Group B streptococci (Streptococcus agalactiae) are spherical gram- positive organisms arranged in chains. GBS can be either α or β hemolytic, if β

hemolytic, then weakly so. GBS can be differentiated from other β-hemolytic

streptococci on the basis of the hippurate test, CAMP test and serological tests. GBS are

able to produce the enzyme hippuricase, which hydrolyses sodium hippurate to glycine

and sodium benzoate. Ninhydrin reagent forms a purple color when exposed to glycine and serves as the indicator for the assay. GBS produce a diffusible extracellular compound called CAMP factor, that will, in conjunction with a specific β hemolysin of

Staphylococcus aureus act synergistically to cause complete lysis of red blood cells (14,

15). GBS contain a group specific carbohydrate antigen, that serves to distinguish it from

other groups of streptococci, and a polysaccharide capsule, which is the basis of serotype

specificity. A variety of different immunological assays have been used to define both the

group and type of GBS, latex agglutination being the most commonly used (82).

Polymerase chain reaction (PCR) can also be used for GBS identification. Primers for the 13

cfb gene, encoding CAMP factor, were used in the PCR reaction. Real-time PCR assay

for the rapid detection of GBS was developed using the same set of primers to shorten time and increase sensitivity (83, 84).

Capsule and Group Antigen The cell wall of GBS is composed of protein antigens such as R, T and Rib antigen, peptidoglycan and carbohydrate (group antigen). The capsule contains the serotyping antigen that subdivides GBS into nine subgroups and may be a major virulence factor (85).

The group B carbohydrate is a common antigen, which is found in all strains and

serotypes of GBS. The group B carbohydrate is chemically composed of rhamnose,

galactose, N-acetylglucosamine, and glucitol. These sugars are linked by phosphodiester

bond to form four different oligosaccharide units resulting in a complex and branched

subunit structure (86). The terminal position of rhamnose is considered to be part of the immunodominant epitope (87). Antibody to the group B carbohydrate is unable to protect mice from lethal challenge with viable GBS (88). All strains of GBS isolated from humans are encapsulated and can be classified on the basis of serology and capsular polysaccharide (CPS) structure. Nine GBS serotypes have been so far identified: Ia, Ib,

II, III, IV, V, VI, VII and VIII (85). The repeating unit of CPS contains combinations of

these saccharides: glucose, galactose, N-acetylglucosamine, rhamnose, and sialic acid.

There is a different arrangement of the saccharides in the repeating unit of each CPS. The

CPSs of GBS type Ia and Ib are structurally similar, differing only in the linkage of the side chain galactose to N-acetyglucosamine. Type Ia CPS has a β-(1→4) linkage and Ib

CPS has a β-(1→3) linkage (89). Type II CPS is composed of glucose, galactose, N- 14 acetylglucosamine, and sialic acid with ratio 2:3:1:1 (90). Type III CPS has a trisaccharide backbone of glucose, N-acetylglucosamine and galactose and a disaccharide side chain of sialic acid and galactose linked β-(1→4) to backbone N-acetlyglucosamine

(91). Type IV CPS is composed of glucose, galactose, N-acetylglucosamine, and sialic acid with ratio: 2:2:1:1 (92). Type V CPS is composed of trisaccharide backbone of glucose, galactose and glucose and two side chains; one side chain is composed of glucose linked β-(1→3) to backbone galactose and the other side chain linked β-(1→6) to a backbone glucose in a trisaccharide of N-acetylglucosamine, galactose and sialic acid

(93). Type VI CPS contains galactose, glucose and sialic acid in a 2:2:1 molar ratio (94,

95). Type VII CPS contains a trisaccharide backbone and trisaccharide sidechain (96).

Type VIII CPS lacks N-acetylglucosamine. The subunit contains glucose, galactose, rhamnose and sialic acid in 1:1:1:1 molar ratio (97).

Phase Variation Group B streptococci demonstrate high-frequency phase variation. Colony opacity variants were detected for GBS type III (8). Occasional transparent colonies with one or more opaque regions were detected on lucid solid medium. Opaque-transparent switches were observed in both directions. Stable opaque and transparent variants have been isolated and compared to each other. Transparent variants are more virulent compared to opaque variants (98). Opaque variants grow poorly in Todd-Hewitt medium, the standard medium used to grow GBS. Opaque variants also have differentiation among their groups. One opaque variant produces unique or overexpressed proteins of 46 and 75 kDa. One variant has capsule alteration with decreased type III antigen, increasing buoyant density. Another variant has a thinner 15

extracellular matrix and lacks the group B antigenic structure. Differences between

transparent and opaque variants, and the genetic mechanism responsible for phenotypic

switching, remain to be elucidated. Phase variation may play a role in the pathogenesis of GBS infection.

GBS Infections

Disease Caused by GBS GBS is a normal member of the human vaginal and

lower intestinal flora. About 10-30% of pregnant women are colonized with GBS (81, 99,

100). GBS are important causes of serious bacterial disease in neonates, pregnant women,

and in immunocompromised adults. In newborns, GBS diseases are characterized as

either early onset or late onset. Early onset GBS disease occurs in infants less than 7 days

of age, typically 1 to 2 days old. In early-onset disease, the neonate is infected by

exposure to GBS before or during birth. The early-onset disease is probably caused by

spread of bacteria from the maternal genital tract through ruptured membranes into the

amniotic fluid. Bacteria amplify and colonize the respiratory tract of the newborn. In

some cases the spread of bacteria occurs even without premature rupture of the

membrane. Late onset, which is less common than the early-onset disease, occurs in

infants older than 7 days of age. GBS disease in newborn babies includes sepsis,

meningitis, pneumonia, cellulitis, osteomyelitis and septic arthritis. In late-onset disease,

the two most common clinical manifestations are meningitis and bacteremia. The most

common syndrome in adults are skin infections (101, 102). Bacteremia can also occur.

GBS can also lead to pneumonia, which is a severe and has a high mortality rate (103).

Endocarditis and epiglottitis also have been reported in older children (104). A 16

substantial proportion of newborns who survive GBS infection suffer from sequelae.

Neurological sequelae occur in up to 50% of the survivors and include mental retardation,

cortical blindness, deafness, uncontrolled seizures, hydrocephalus, hearing loss and

speech and language delay. In pregnant women GBS cause urinary tract infection,

amnionitis, endometritis and wound infection (105). In nonpregnant adults infection is

most commonly seen in patients with diabetes mellitus, cancer, and AIDS. In these

people GBS can cause skin or soft tissue infection, bacteremia, genitourinary tract

infection, and pneumonia. The fatality rate for GBS disease can be as high as 4 % in

newborns (106). Fatality rates are higher in adults older than 65 than in younger adults.

Epidemiology and Transmission GBS is a commensal of the genitourinary tract in

15-40% of adult women (107). It is the leading cause of serious bacterial infection in

newborn babies. Infected newborns are exposed to the microorganism carried by their mothers either just prior to parturition or during the birth process. The risk of neonatal colonization correlates with the intensity of maternal colonization. GBS colonization rates in pregnant women have remained constant over the past decade (25-30%) (108,

109). When mothers were GBS culture positive at delivery, the vertical transmission rate

(of GBS colonization of the newborn) is between 29-72% with a mean rate of 51% (100).

If the maternal GBS culture is negative during labor, only 3% of infants become asymptomatically colonized (110). Neonates are exposed to GBS either via GBS

ascension through ruptured amniotic membranes, aspiration of contaminated amniotic

fluid, or through exposure to colonized tissues while passing through the birth canal.

Bacteria crossing intact amniotic membranes have also been reported to cause disease. 17

Organism acquisition during passage through birth canal is thought to cause late onset

disease. Horizontal transmission from the environment occurs but is rare (110). GBS type

Ia, III, and V are most prevalent in early-onset disease. Serotype III GBS are important because they account for 36% of early onset and 71% of late onset GBS infection (111).

The genetic epidemiology of disease-causing GBS has been studied by using

Hind III restriction endonuclease digest patterns (RDPs) of chromosomal DNA. Bacteria

have been classified into RDP type 1, 2, 3 for GBS type III and 1, 2, 3 and 4 for GBS

type I (112). GBS strains causing early onset septicemia belong to RDP types Ia-3 and

III-3. GBS strains of RDP type Ia-1 and III-1 have low capsular sialic acid levels and are

susceptible to tetracycline. Half of the strains in the study causing meningitis belong to

type III-2 or III-3. According to RDP typing, GBS type III shares the same clonal origins

with some strains of type II and V (112).

Pathophysiology of Infection

GBS infection in newborns can lead to death and disability. In pregnant women,

GBS causes clinical illness ranging from mild urinary tract infection to life-threatening

sepsis and meningitis. Most invasive maternal infections are bloodstream infections,

osteomyelitis, endocarditis and meningitis. Noninvasive syndromes in pregnancy include

amnionitis, endometritis, wound infections fasciitis and cellulitis. In nonpregnant adults

invasive GBS disease starts with bacteremia. The skin is the next most common site of

infection.

Newborns are easily infected by GBS because they are quantitatively and

qualitatively deficient in host defenses. The interaction between bacteria and host leading 18

to the disease can be divided into steps, including: adherence to epithelial surfaces, invasion of epithelial and endothelial cells, direct injury to host tissues, avoidance of host immunological defenses, and induction of the sepsis syndrome.

Adherence Adherence to epithelial surfaces is considered the first step in

pathogenesis. Adherence of GBS to epithelial cells plays an important role in

colonization of the rectum and vagina, as well as in initiating invasive infections of

newborns. GBS can adhere to multiple cell surface components through both protein and

non-protein adhesins (ligands) on the GBS surface. Interactions through opsonin-

mediated processes are also possible. An example of a nonprotein ligand in GBS

adherence is lipoteichoic acid (LTA). LTA is an amphiphilic glycolipid polymer

extending from bacterial cell wall. Enzymatic treatment of fetal epithelial cells with

trypsin or periodate destroys LTA binding to cells, indicating the presence of a

glycoprotein receptors on the fetal cells (113). A later study shows that the LTA may

have cytotoxic functions as well (114). Pretreatment of GBS with protease can decrease

their ability to attach to epithelial cells, implying that bacterial surface proteins are also

important in bacterial-host adherence.

Adhesins can also bind the extracellular matrix (ECM), which is a

macromolecular structure underlying epithelial and endothelial cells. ECM is composed

of a variety glycoproteins: collagen, laminin, fibronectin and fibrinogen. Several

interactions between GBS surface proteins and receptors in ECM have been reported to play roles in the adherence process. Laminin binding protein (Lmb) has been identified in

GBS. Lmb is a putative lipoprotein, which has significant homology to group A 19 streptococcal Lra I protein family, which mediates bacterial-host cell attachment. Lmb protein is expressed by most GBS strains. Mutation in the lmb gene has been reported to reduce adherence to immobilized human laminin (115).

Two fibrinogen binding proteins, FbsA and FbsB, have been identified. The FbsA has a signal peptide, a repeat region and a C-terminal wall-anchoring region with an

LPXTG motif. The repeat region is responsible for the fibrinogen binding. Deleting the fbsA gene causes a reduction in the binding activity to soluble fibrinogen, as well as adherence to and invasion of epithelial cells. Adherence, but not invasion, is partly restored by reintroducing the fbsA gene (116, 117).

GBS is reported to bind to cytokeratin 8, which is a constituent of plasma membranes (118). The bacterial protein ligands responsible for binding to cytokeratin 8 are not yet identified. Hydrophobicity of surface proteins also plays a role in mediating adherence. Protein surface antigens are highly hydrophobic whereas polysaccharide antigens containing sialic acid are highly hydrophilic. Hydrophobic strains adhere to epithelial cell better than hydrophilic strains. Treatment with neuraminidase enhances the surface hydrophobicity of hydrophilic GBS strain whereas the hydrophobic strain is unaffected by this treatment (119).

GBS expresses a cell-surface-associated protein A (CspA) encoded by cspA gene.

Csp A contains a putative signal peptide and a cell-wall anchor motif (120). Sequence analysis indicates that it shares functional and structural domains of the cell envelope- associated protease (CEP) family. The 153 kDa CspA protein is a subtilisin-like extracellular serine protease homologous to streptococcal C5a peptidase and caseinases 20

of lactic acid bacteria, but it does not cleave C5a or casein. CspA cleaves fibrinogen. It

degrades the lower α band of fibrinogen (120). The function of fibrinogen cleavage by

CspA is not clear. It has been postulated that GBS may be exploiting the adhesive

properties of a fibrin-like substance. Proteolysis of the N termini of the α and β chains of

fibrinogen lead to polymerization of cleaved fibrinogen, forming fibrin, which is highly adhesive (120). Therefore the cleavage of fibrinogen by CspA may lead to fibrin-like substance formation. This may promote the aggregation of GBS or the coating of GBS with fibrin. It has been shown that cspA mutants are significantly more sensitive than wild-type strains to opsonophagocytic killing by human neutrophils in vitro (120).

Invasion The second step in pathogenesis is invasion of epithelial and endothelial

cells. Organisms access the blood stream by direct invasion of the epithelial cell barrier.

GBS can penetrate and survive within various tissues, including placental membranes,

the alveoli and the brain microvascular endothelial cells (BMEC), the single-cell layer

which constitutes the blood brain barrier (121). GBS invade more efficiently into

pulmonary microvascular endothelial cells than endothelial cells of the pulmonary artery suggesting a tropism for particular endothelial surfaces (122, 123) and also implying that

lung endothelium may be the original site of bacterial invasion in neonates. GBS has

been shown to invade chorion and amnion epithelial cells in vitro (124). The chorion-

amnion is a multilayered structure comprised of a monolayer of chorion epithelial cells

that line the maternal side of the placenta. Amniotic fluid infections most often occur

through a ruptured chorioamniotic membrane. It is also possible that bacteria go through

intact placental membranes and then replicate in amniotic fluid. Exposure to high 21

bacterial concentrations probably leads to colonization of the lung and airways, and

subsequent pneumonia, sepsis and meningitis. Even though GBS is considered an

extracellular bacterium, it can survive in the host cell in the cytoplasm or in intracellular

vacuoles (125). However, evidence of intracellular replication has yet to be reported.

Active bacterial DNA, RNA, and protein synthesis appears to be necessary for invasion

(126). The bacterial invasion mechanism also requires host microtubular and cytoskeletal

elements. Invasion is inhibited in a dose-dependent manner by decreasing extracellular

Ca2+ level and by substances known to interfere with eukaryotic calcium regulatory systems (125). The result implies that GBS may invade cells by triggering calcium- dependent phagocytosis-like internalization mechanisms.

GBS surface proteins play significant roles in the invasion process. Surface

proteins interact with host protein receptors in the ECM and on cells. Recently, it has

been reported that the alpha (α) C protein (ACP) on the surface of GBS mediates GBS entry into cervical epithelial cells and GBS translocation across layers of these cells

(127). The ACP is characterized as one of the two proteins that constitute the C antigen.

C antigen has been found in many strains except clinical type III isolates of GBS. Based on trypsin sensitivity, C antigen is composed of two proteins, a trypsin-resistant alpha- protein and a trypsin-sensitive betaprotein (the α-C protein, ACP, and β-C protein). The

ACP is unrelated to the β- C protein. ACP is grouped into the alpha like protein family

(Alp). The members of the Alp family are ACP, Rib, R28 and Alp2 protein. Rib is identified from GBS type III (128), R28 is found on the surface of type V (129) and alp2 is identified in type V (130). ACP is composed of an N-terminal signal sequence 22

followed by a nonrepeated N-terminal region, a repeat region, a wall-anchored region

with an LPXTG motif, a short hydrophobic region that may span the cellular membrane,

and a charged C-terminal tail. It has been predicted that the repeats in Alp have structures relate to immunoglobulin (Ig) implying the possible function of Alp in recognition

(ligand-host receptor recognition). ACP interacts with host cell glycosaminoglycan

(GAG). The data indicate that ACP interacts with GAG and enters the cell cytosol by a mechanism that involves Rho GTPase-dependent actin rearrangement (127). The predicted 53 kDa Spb1 protein (surface protein of group B streptococcus-1) may be involved in the invasion process as well. Mutation in gene spb1 leads to less invasion in the HeLa cervical epithelial cell, and restoration of spb1 in trans corrected the defect

(131). The other Fbs protein, FbsB (FbsA was discussed above in adherence), also promotes invasion into epithelial cells. Deletion of the fbsB gene severely impairs the invasion of mutant GBS into lung epithelial cells (132). C5a peptidase is a surface- associated serine protease that specifically cleaves the complement component C5a, a chemotaxin for polymorphonuclear leukocytes (PMN). C5a peptidase may play roles in both the invasion process and avoidance of the host immune system. All group A streptococcal (GAS) serotypes, GBS, as well as groups C and G streptococci of human origin produce the streptococcal C5a peptidase (SCP) (133). The protein is highly conserved across species. The SCP from GAS (ScpA) is 95 to 98 % (residue) identical to

SCP from GBS (ScpB) (134). The ScpB promotes invasion by interacting with fibronectin. Mutations cause a reduction in epithelial cell invasion that can be fully restored by trans complementation with the cloned wild-type gene (135). 23

GBS capsular polysaccharide (CPS) may have a negative effect on the invasion process. The main function of CPS is generally considered to be prevention of phagocytosis by host immune cell. Wild-type encapsulated clinical isolates, an asialo and unencapsulated isogenic transposon mutants invade epithelial cells to different degrees

(136). This observation indicates that capsular polysaccharide is not essential for invasion of respiratory epithelial cells. The isogenic strains of GBS, which lack capsular sialic acid or type III capsular polysaccharide, invade and injure human umbilical vein endothelial cells three-to five fold more than the wild-type strains (126). The capsule-deficient strains also invade lung endothelial cells more than wild-type strains (137). The results indicate that capsular polysaccharide attenuates invasion. The attachment and invasion processes are regulated by oxygen and bacterial growth phase (138), as is the production of CPS. It is possible that in the beginning of the process, bacteria produce less CPS to facilitate attachment and invasion, and then later switch to high production of CPS to mask the surface proteins preventing them from being recognized by antibodyand to inhibit opsonophagocytosis.

Host Injury After invasion, bacteria proliferate. The proliferation is associated with injury to host tissues. However, proliferation is not essential for injury. Direct cell injury facilitates penetration and spread of the organisms. GBS products, including cytotoxins and enzymes, can directly mediate injury. GBS β-hemolysin has been shown to be directly cytotoxic to lung epithelial cells (139). It acts as pore-forming cytolysin

(137). Nonhemolytic variants didn’t induce cell injury in this model, while hyperhemolytic variants derived by chemical or transposon mutagenesis are extremely 24 toxic; resulting in disruption of cytoplasmic and nuclear membranes, loss of microvillus architecture, and marked swelling of cytoplasm and organelles (139). GBS also produce several proteinases, which can degrade tissue. Cell-associated collagenase activity has been postulated. It has been shown that antibodies against collagenase from Clostriduium histolyticum cross-react with cell-associated proteins produced by GBS, and inhibit GBS hydrolysis of a synthetic peptide collagen analog (140). GBS also contain hyaluronan lyase. This enzyme degrades extracellular hyaluronan, which is an important component of ECM and may promote persistent colonization of the vagina (141, 142). Neutrophils and macrophages have hyraluronan receptors, CD44. Thus this enzyme may inhibit their ability to interact with infected host cells. A GBS hyaluronate lyase has been identified, which shares 50.7% amino acid identity with hyaluronidases from S. pneumoniae but not from GAS and Clostridium perfringens (141).

Avoidance of Immune System To firmly establish infection, GBS must avoid immune clearance by both innate and acquired immune mechanisms. After penetration into lung tissue or the blood stream, an immunological response is recruited to clear the organism. The central elements of the response are neutrophils, although macrophages are also involved in bacterial clearance. Effective phagocytosis of GBS by neutrophils requires opsonization, requiring serotype-specific anti-GBS immunoglobulin, and both classical and alternative complement pathways. In order to avoid opsonophagocytosis

GBS possess several factors that interfere with the phagocytosis mechanism. The main factor is the type-specific polysaccharide capsule. Sialic acid rich CPS inhibits complement alternative pathway activation, as seen in adult sera deficient in type-specific 25

anticapsular antibodies (143). Resistance of serotype III GBS to opsonophagocytosis is

proportional to the sialic acid content of the capsular polysaccharide. Removal of sialic

acid by treatment with neuraminidase or by transposon-insertional mutagenesis increases

deposition of opsonic C3 (144). A GBS type III mutant strain, which is capsule-deficient,

binds to C3b, the active form of C3, in greater amounts than the non-mutant parent strain.

The parent strain binds predominantly to C3bi, the inactive form leading to reduction of

C3b deposition on bacterial cell surface. The sialic acid rich capsule prevents C3b deposition, presumably through acquisition of Factor H (FH) from host plasma (144).

Capsular sialic acid limits C5a production (145). As noted earlier, C5a is an important chemoattractant that recruits neutrophils and macrophages. The mechanism by which the capsule inhibits C5a production is yet to be defined. The ability of GBS to inhibit C5a production was observed in type III GBS, which have high capsular sialic acid. The C5a- ase might not be necessary for type III strains to cause invasive disease because the high

sialic acid content inhibits C5a production.

Several GBS surface proteins are also involved in escaping host immunity. C5a

peptidase functions both in invasion, as described above, and in immune avoidance.

ScpA and ScpB cleave the human C5a molecule between residues His-67 and Lys-68, causing the release of a short C-terminal C5a fragment (135). An ScpB-negative mutant shows increased sensitivity to phagocytosis by immune cells in the presence of complement. GBS surface protein Alp (alpha-like protein, described earlier under bacterial invasion) also functions in avoiding the host immune system. The repeat regions of Alp (described earlier) affect immune escape by reducing the immune response to the 26

whole protein and in particular to the N-terminal region. The β component of the C

protein can bind to IgA (146). This binding can interfere sterically with deposition of

opsonically active complement protein C3 on the GBS surface. The β component of the

C protein also binds to the soluble complement inhibitor factor H (fH) (147). The fH-

binding sites on the β component consisted of discontinuous and partially homologous

sequences. Therefore, GBS bind multiple sites on fH, making the association more stable.

Binding of the GBS β component does not block active sites on fH. Thus active sites on

fH are free to inhibit C3b deposition and opsonophagocytosis (147).

The lipoteichoic acid (LTA) has been shown to play a role in avoidance of the

immune system. LTA mutant strains display an increased susceptibility to human

defensins. Moreover, the mutant strains are less virulent than the wild-type (148). Killing

mechanisms of phagocytic cells involve production of reactive oxygenmetabolites

(oxidative burst). The production of reactive oxygen intermediates including superoxide anion, hydrogen peroxide and hydroxyl radicals are toxic to GBS and to host cells, damaging DNA, RNA, protein and lipid. GBS has mechanisms to evade the oxidative burst by using enzyme Mn-cofactored superoxide dismutase (SodA) (149). The enzyme converts the superoxide anion to molecular oxygen and hydrogen peroxide, which can be metabolized by bacterial catalases and/or peroxidases. SodA mutants display an increase in susceptibility to killing by macrophages.

GBS has been shown to induce apoptosis in macrophages (150-152). GBS does

not need to be in the macrophage cytoplasm to promote apoptosis. The apoptotic pathway

initiated by GBS is dependent on host protein synthesis (but independent of caspase-1 27

and –3). Apoptosis of macrophages can decrease antigen presentation and inhibit the

generation of immune responses.

Induction of Sepsis Syndrome The last stage in disease pathogenesis is induction

of the sepsis syndrome. The sepsis syndrome arises from the body’s extreme attempts to eliminate the GBS. While attempting to clear the pathogen, overreaction of the immune system may cause damage to the host.

If the epithelial barrier and immunological clearance fail, GBS establish

bacteremia in newborns. These infants may develop septic shock as consequence. Severe

early-onset GBS disease is clinically different from septic shock syndrome resulting from

infection with gram-negative bacteria. GBS sepsis syndrome includes systemic

hypotension, persistent pulmonary hypertension, tissue hypoxemia and acidosis,

temperature instability, neutropenia and multiple-organ system failure. GBS induce many

cytokines. Predominant among these are Th1 and proinflammatory cytokines, and less

Th2 cytokines (153, 154). GBS infection has been reported to stimulate TNF-α,

Interleukin-1 (IL-1), IL-2, gamma-interferon, IL-6, IL-8, and IL-12 production. TNF is a

key mediator, but other cytokines also play a role in the unique GBS sepsis syndrome

(155).

GBS also induces TNF-α production from macrophages and monocytes. TNF-α

release is relevant to the pathophysiology of neonatal sepsis. TNF-α is believed to play

an important role in morbidity and mortality. It is a proinflammatory cytokine that

produces fever, metabolic acidosis, capillary leak and cardiovascular collapse leading to

systemic hypotension. TNF-α production is inhibited by monoclonal antibodies against 28

CR3 and CR4, which are receptors for C3 and C4 complement. Therefore the ability of

plasma and serum to amplify TNF-α production in response to GBS is dependent on the

activity of the complement pathway (156). It has been shown that group B polysaccharide

and GBS peptidoglycan, which are cell wall components, induce greater production of

TNF-α than CPS or LTA (156, 157). GBS induce TNF-α independently from CD14

(158). Studies have shown that heat-killed GBS type III induces host cell signal

transduction that leads to activation of transcription factors NF-kB and AP-1 that can lead

to TNF production. The MAPK (p38 mitogen-activated protein kinase) pathway is also

involved in TNF-α production by GBS (155). Mutant strains lacking CPS can induce

TNF-α production to the same degree as wide type strains (155).

GBS also induce production of other cytokines. The production of many of these

cytokines follows TNF-α release (153). IL-1 may induce septic shock. Treatment with an

IL-1 receptor antagonist can reduce the symptoms. GBS also induces nitric oxide

production in respiratory epithelial cell leading to lung injury (159). The nitric oxide may

also play an important role in meningitis because GBS induce a respiratory burst in the

CNS. The resulting production of hydrogen peroxide and superoxide leads to CNS

damage. GBS also induce human monocytes to express cyclooxygenase 2 (COX 2),

which is involved in many inflammatory states (160).

IL-12 also may play a role in the induction of neonatal sepsis and clearance of organisms. Pretreatment with anti IL-12 increases lethality in a mouse model (161).

Administration of recombinant IL-12 improves survival time and decreases GBS in blood. 29

GBS may induce membrane injury indirectly by alteration of host cell processes.

The presence of GBS in the lower uterine cavity promotes oxygen radical-induced

damage to adjacent fetal membranes.

Use of Antibiotics to Prevent and Treat GBS Infection

The Centers for Disease Control and Prevention (CDC) recommends screening

pregnant women for Group B strep two to four weeks before labor by using the standard

culture method. Women who have a positive GBS result will be given four hours of

antibiotic treatment during labor (162).

GBS vertical transmission can be inhibited by ampicillin chemoprophylaxis

administered at least 4 hours before delivery to achieve bactericidal effect. Giving

penicillin G 4 hours before delivery (intrapartum antibiotics) is also effective (163).

Penicillin administration to infants shortly after birth (postpartum prophylaxis) reduces

early-onset GBS. Antibiotics used preferentially are ampicillin, clindamycin, gentamycin

and penicillin. The combination of penicillin and gentamycin reduces the rate of GBS

infection (164). The CDC recommends intravenous antibiotics during labor for women

who have maternal risk factors (162). These risk factors include premature delivery, maternal fever, and prolonged rupture of the amniotic membranes. The use of vaginal disinfectants such as Chlorhexidine was investigated in Sweden. The result shows that vaginal disinfectants are less effective than antibiotic prophylaxis. However, the use of vaginal disinfectants leads to less resistance when compared to the use of antibiotics.

Penicillin is the drug of choice for known GBS infections. However, because

symptoms and clinical suspicion often precede the definitive bacteriologic diagnosis, 30

empirical therapy for neonatal meningitis is established, often with a combination of

ampicillin, a third generation cephalosporin that penetrates the CNS, and/or an

aminoglycoside. General antibiotics are administered for the first 48-72 hours until

species are identified, susceptibility testing is completed, and a clinical response is

observed. Penicillin is then used for a total of 10 to 14 days (164). In neonates, the

recovery rate is about 75-80 % without sequelae and 5-10% with sequelae (165).

Neonatal fatality rate after treatment is 5-7% (164). In adult infection with GBS and in

late-onset neonatal disease, penicillin is also the drug of choice. The recovery rate (50-

60%) is lower compared to the early onset disease (166).

Even though penicillin is the drug of choice for prophylaxis and treatment of GBS

disease and resistance to penicillin has not been reported, other antibiotics such as

macrolides have been recommended as secondary agents for the patients with penicillin

allergy. It has been reported that clinical GBS isolates exhibit degrees of resistance to

antibiotics such as erythromycin (18%), clindamycin (8%) and tetracycline (>80%) (167).

Resistance mechanisms include ribosomal modification by methylase encoded by an erm

gene and drug effux by membrane-bound protein encoded by a mef gene.

Hypothermia has been proposed as an adjunctive treatment for severe bacterial

meningitis in a rabbit model of severe GBS meningitis (164). The application of

hypothermia initiated shortly after antibiotic therapy improves short-term physiologic

measures associated with brain injury.

31

Protective Immunity to GBS

The absence of maternal antibody to GBS capsular antigens has been identified as an

important factor in the pathogenesis of GBS infections in newborns (168). Presumably,

transplacental transfer of maternal anti-GBS antibody results in sufficient immunity in the

neonate to protect against the development of infection. The absence of maternal antibody in

the face of vaginal colonization with GBS places the infant at highest risk. Why some women

who carry GBS fail to mount an antibody response is still a matter of conjecture. Bacterial

factors may play a role. Although host factors play an important role, it has been shown that

women whose infants have had GBS infection are able to produce anti-GBS antibodies in

response to immunization with GBS carbohydrate (169).

Lancefield originally demonstrated the protective efficacy of anti-GBS antibodies

using rabbit antisera (170). There are eight serotypes of GBS. Lancefield established that

protective efficacy is type specific (85). Antisera raised against type II or III GBS do not

protect against infection with type I bacteria (171). These experiments also show that

antibodies to group B determinants (i.e. expressed on all GBS) are not protective. Using

monoclonal antibodies Pincus, et al. have confirmed these findings, and demonstrated that the

protective epitope on the type III capsular polysaccharide (CPSIII) contains sialic acid (172-

174). Monoclonal antibodies directed against non-sialic acid determinants on CPSIII are not protective (175). Although there have been reports that group specific antibodies can protect

(176), most studies confirm Lancefield’s original observation. Studies of Pincus, and those of others, indicate that one reason for the failure of group specific antibodies to protect is that 32

there is limited expression of these epitopes on the surface of intact GBS (172). Protein

antigens of GBS can also elicit protective antibodies (177, 178).

The role of CPSIII in bacterial virulence has been studied, in part because the

protective antibody response to the capsular antigen is so well characterized. Variants lacking

the type III capsule have markedly reduced virulence (136, 144, 179). Transposon mutagenesis

has demonstrated that sialylation of CPSIII is critical for pathogenicity (179). The type III capsule is anti-complementary, inhibiting the activation of the alternate pathway of complement activation. Phase variants and transposon mutants lacking the capsule spontaneously activate complement. Sialic acid plays an important role in the ability of CPSIII to inhibit complement deposition (145, 180). It has been postulated that the reason antibodies against sialic acid containing epitopes of CPSIII are protective is that they may block this

complement-inhibiting activity of CPSIII, thus facilitating C3 deposition, perhaps by both

classical and alternate pathways, as well as interacting with Fc receptors on neutrophils (181).

Protective antibodies to CPSIII appear to be directed at a single, repeating epitope

containing sialic acid (181). This structure also appears to function as a virulence factor, in part

because it is able to prevent activation and deposition of complement by the bacterial surface.

Antibodies that bind to this target block the anti-complementary activity, as well as serve to

opsonize GBS. The capsular polysaccharides of other GBS serotypes express similar critical

epitopes against which protective antibody is directed.

GBS surface proteins play important roles during different stages of an infection.

Several GBS surface proteins have been purified and characterized. Some of them have been shown to induce protective antibody against GBS. The first surface protein antigen 33

identified in GBS was the C protein, which we have described earlier. Lancefield et al. have shown that antibodies to the C antigen are protective against GBS in a mouse model

(182, 183). All of the alpha-like proteins (Alps) confer protective immunity against GBS

in animal models (184, 185). With the exception of ACP, biological functions of these

proteins still are yet to be defined. It has been shown that immunization with ScpB

induces protective antibody in mice (186). The 45 kDa Sip protein has a wall-anchoring

motif and induces protective immunity against lethal infection with GBS (187). It is

interesting that Sip from all nine GBS serotypes is highly conserved; implying Sip could

be a good candidate for vaccine development. The function of Sip protein is not known.

Another surface protein, which has been shown to elicit a protective antibody, is R

antigen. Like C antigen, the R antigen is immunogenic on intact GBS. The biological

function of R antigen is still not known. Antibodies raised against a R antigen protecte

mice against infection with R-antigen positive strain of GBS, but not from R negative

strains (188, 189).

Development of a GBS Vaccine

Protective immunity of GBS can be achieved by maternal antibodies to the CPS

of GBS, which is transferred through placenta. Like other bacterial CPS, GBS CPS is a

poor immunogen and acts as a T-independent antigen resulting in IgM production instead

of IgG and lack of immunological memory. An initial vaccine consisting of CPSIII

underwent both animal and then human testing (190). Although the vaccine elicited

antibodies in only 60% of pregnant women immunized, the passive transfer of anti-GBS

antibodies to the neonate was shown, thus demonstrating the feasibility of maternal 34

immunization (191). Due to the low immunogenicity of the purified CPS, alternative

approaches are being examined.

One strategy to improve immunogenicity of the CPS is to couple it to carrier

proteins, to obtain a CPS-protein conjugate vaccine. Different coupling strategies and

protein carriers have been proposed. Tetanus toxoid (TT), cholera toxin B subunit, and

GBS proteins have been used as protein carriers. Tetanus toxoid has been most

thoroughly evaluated. Coupling procedures include binding cyanogen bromide-activated

CPS to TT via the spacer adipic acid dihydrazide (192, 193). Another strategy is to

conjugate CPS to TT without linkers (194, 195). GBS type III CPS is treated with

periodate oxidation method to form free aldehyde groups on the sialic acid side chain.

Then the treated polysaccharide is coupled with TT via the aldehyde groups.

TT-CPS conjugates have been tested in animals and in humans. The ability of

CPSIII-TT, administered to the mother, to protect GBS-infected neonates was

demonstrated in animals (193, 196). Subsequent clinical trials of this vaccine have demonstrated an opsonic antibody response in approximately 90% of recipients (195). At

the same time, experimental studies conjugating the CPS of other GBS serotypes to tetanus toxoid have been performed. The protective efficacy of a tetravalent (types Ia, Ib,

II, and III, which account for >90% of clinical infections) polysaccharide-tetanus toxoid

conjugate vaccine has been demonstrated in animals (197, 198). The conjugate vaccine

produces predominantly IgG1, in a mouse model, implying a Th2 bias of the vaccine.

Prior exposure to GBS and TT can affect the antibody response (195). 35

Since colonization of genital and lower intestinal tracts is critical in GBS

transmission, effective mucosal immunity may play an important role in vaccine-

mediated protection by limiting GBS colonization. IgA is the main protective antibody in

mucosal immunity. CPSIII was coupled to cholera toxin B subunit (CTB) by reductive

amination (199). The conjugated vaccine was administered to mice. Immunization elicits

both mucosal and systemic immune responses with high IgA, and IgG antibody against

CPS and GBS in sera and at mucosal sites. The immunization route affects the immune

response obtained: intranasal immunization elicits the highest level of IgA and IgG in the

lungs, peroral immunization leads to highest antibody levels in the intestine, vaginal vaccination generates highest antibody levels in vagina, and rectal administration produces the highest response in the rectum (200, 201).

A second approach to generate a protective anti-GBS antibody response is by

using GBS antigenic surface proteins, which are expressed on most or all GBS serotypes,

as vaccine candidates. GBS surface antigenic proteins were identified and tested for

ability to elicit protective antibody against GBS, as discussed above. Surface antigenic

proteins, ACP protein, β-C protein, Sip and Rib have been tested in animals as potential

candidate antigens for a vaccine (128, 187, 202, 203). All of them confer protective

antibody against GBS, and the sera from vaccinated animals protecta mice from GBS

lethal challenge. In an effort to elicit the greatest degree of protection by using multiple

GBS antigens, vaccines have been designed in which the CPS is conjugated to a GBS

protein. The alpha-C protein is conjugated to CPSIII using the reductive amination

method (202). Low titers of antibody to type III GBS are elicited. Adjuvant is required 36

for immunization. Compared to GBS-TT vaccine, these proteins provide less impressive

protection. Moreover, some proteins are not expressed in all clinical isolates. For

example, beta-C protein is expressed in type Ia and Ib, but not in III. Even though Rib

and Sib are more abundantly expressed in clinically significant strains, not all of them

contain these proteins. Generally GBS, which do not contain Rib, do express alpha-C

protein. The ScpB protein, found in all clinical GBS isolates is also believed to be good

candidate for vaccine development. Immunization with purified ScpB increases the

bacterial clearance in lung (204).

Mimotopes

Antigen

An antigen is a substance that can be bound specifically by an antibody, T-cell receptor, or other specific immune recognition molecule. In this thesis, we will confine our discussion of antigenicity to recognition by antibody. Some antigens can induce

antibody production; some do not. An antigen, which is able to induce an immune

response, is called an immunogen. Thus not all antigens are immunogenic, but all

immunogens are antigenic. For example, a hapten is a small molecular weight molecule

that when bound to a carrier protein can elicit antibodies that bind to the hapten. Thus the hapten alone is antigenic, that is it can be bound by antibody, but not immunogenic.

When coupled to a carrier it is both. Immunogenicity results from four properties of the immunogen: foreignness, molecular size, chemical composition, and susceptibility to antigen processing and presentation. An immunogen has to be recognized as nonself by 37

the immune system. The best immunogens generally have molecular weights greater than

25 kDa, molecules less than 5 kDa have poor immunogenicity (205). Antigens with high

complexity in chemical composition generally have high immunogenicity. Proteins are

good immunogens, simple carbohydrates with repeated chains are poorly immunogenic,

as are lipids and DNA. Molecules that are not degraded or processed by antigen presenting cells are poorly or not at all immunogenic, for example plastics.

Epitope

Antibody binds to specific sites on the antigen, which are referred to as antigenic determinants or epitopes. Epitopes can be bound directly by antibodies, which are also the antigen receptors on B lymphocytes. B cells and T cells often recognize different epitopes on the same molecule. The T-cell receptor does not recognize native antigen as do B cells, but recognizes antigen that has been processed into antigenic peptides and presented in association with MHC molecules. Therefore destruction of the conformation of a protein does not affect its T cell epitopes. Antigens recognized by T cells must possess two different interaction sites: the epitope, which interact with the T cell receptors, and the agretope, which interacts with an MHC molecule.

Properties of epitopes for antibody/B cells include accessibility of the epitope,

hydrophilicity, and structural mobility. Antigenic sites should be accessible on the

surface of the molecule to be recognized by antibodies binding to the native protein. In

general the native conformation of proteins in aqueous environments is to fold with the

hydrophobic residues buried in the interior and hydrophilic residues exposed on the 38

surface (206). Segmental mobility of the polypeptide backbone also contributes to

antigenicity.

Antigen-Antibody Interaction

Detailed structural analyses of the interactions between epitopes and antibodies

have been reported. Here we will discuss the interaction of Fab antibody fragments to

protein antigens: hen egg white lysozyme (HEL) (207-210), influenza virus neuraminidase (211, 212) and angiotensin II (213). Interactions of several of these antibodies with anti-idiotypic antibodies have also been studied and compared to the interaction with antigen (214, 215). B cell epitopes can contain sequential or nonsequential amino acids. In the latter case, the native protein is folded to bring the amino acids into close juxtaposition.

The interaction between antigen and antibody is influenced by the structure of the

antigen, as well as the sequence of the portions of the antibody that make contact with the

antigen, the so-called complementarity determining regions (CDRs). Angiotensin II is a

small antigen, folded into a compact structure that interacts with the antibody within a

deep and narrow cleft (216). Epitopes of globular proteins, such as HEL interact with

antibody on the surface of the protein and in a very shallow binding cleft (209). The

binding interaction between antibody and antigen requires good complementarity

between proteins and reasonable juxtapositions of polar residues to permit hydrogen bond

formation. Studies of antigen/antibody binding in HEL reveal water molecules in cavities

between the two that provide a bridge for hydrogen bonds (207, 209). Two monoclonal

antibodies (mAb) specific for the influenza neuraminidase, NC41 and NC10, have been 39

studied (211, 212). The neuraminidase epitope for NC10 overlaps that for NC41. These

two mAbs bind differently to antigen. Five of the six CDRs of NC41 make contact with

antigen while four CDRs of NC10 make contact.. Thus antibody/antigen interactions

reflect both structures (212).

Anti-Idiotype Antibody (Anti-Id)

The amino acid variability of the antibody CDRs allows them to function as

antigenic determinants that can in turn lead to the production of new antibodies, termed

anti-idiotypic antibodies (anti-Id). The interactions between the anti-lysozyme antibody

D1.3 (Ab1) and its anti-Id (Ab2) and its antigen HEL were studied (215). The anti-Id

antibody binds to 13 amino acids on D1.3 mainly from the CDRs; whereas seven amino

acids make contact with the nominal antigen, HEL. The main chain conformation of the

D1.3 CDRs in the anti-Id Ab complex shows significant differences from the complex with HEL, especially the side chains of the CDRs. Thus the conformation of the antibody binding site is influenced by the antigen it binds. As discussed below, anti-idiotypic

antibodies have been used as mimics of antigen, since both bind the same antibody. In

this case it has been said that the anti-idiotype carries the internal image of the antigen,

although as the HEL data show, this internal image is not perfect (215, 217). An example

of internal image mimicry is seen in the angiotensin system. The anti-anti-Id (Ab3) binds

strongly to the antigen angiotensin (218, 219). Crystallographic studies show that when

complexed with Ab3, the angiotensin adopts a conformation that resembles the CDR3 of

a light chain (220). This suggests a possible mechanism for the interactions; in which a 40

CDR of Ab2, presumably CDR3 of the light chain, would carry a similar conformation to

that of the angiotensin

Polysaccharide Antigen

The repeating polymeric structures of polysaccharide and DNA are poor

immunogens. They function as T-independent antigens eliciting primarily low affinity

IgM antibodies and failing to induce immunologic memory. Cross reactions among these

structures demonstrate how similar epitopes, having different chemical structures on very

different molecules can mimic each other. Capsular polysaccharide from group B

meningococcus and Escherichia coli K1 are poorly immunogenic (221-224). The

α(2→8)-linked N-acetyl neuraminic acid (Neu Nac) polymer, which is found in the E.

coli K1, Klebsiella, and group B meningococcus CPS component can be bound specifically by monoclonal IgM antibody (NOV) (223, 224). The antibody is protective against E. coli in newborn rats. The mAb also cross-reacts with polynucleotides and denatured DNA. The cross-reactions might result from a similar distribution of negative charges between the carboxyl group of α(2→8)Neu Nac and the phosphates of the denatured DNA or polynucleotides (224). This finding suggests that the distribution of negative charges on certain antigens is the basis of cross reactivity in this system. More generally, it suggests that antibody cross reactivity is based on structural similarities

NOV between antigens. The amino acid sequence of IgM VH is 106 residues and belongs to

subgroup VB III, and VL is 109 residues and belongs to subgroup II (222).

Antigenic cross-reactions could lead to disease. A human anti-DNA monoclonal

autoantibody, 16/6, is found to share an idiotypic determinant with antibody from patients 41

with active systemic lupus erythematosus (SLE), and with other anti-DNA mAb from

unrelated patients (225). Amino acid analysis reveals that first 40 light chain amino

terminal residues are identical among the antibodies sharing this idiotype. Six mAb from

patients with Waldenstrom’s macroglobulinemia that react with Klebsiella pneumoniae

polysaccharides were tested for their ability to bind to nucleic acid antigens (225). The

study shows that one of these IgMs bound to polynucleotides and to single stranded

DNA. This antibody is also the only one to share this idiotype. The binding between this

IgM and the polynucleotides is inhibited by Klebsiella polysaccharide 30 (K30). The 16/6

mAb is found to react weakly with the K30. The results indicat that anti-polysaccharide

antibodies share an idiotypic determinant with anti-nucleic acid antibodies and

immunoglobulin from SLE and Waldenstrom‘s macroglobulinemia, a cancer of IgM-

secreting plasma cells.

Dextran is a polysaccharide consisting of repeating d-glucose residues. It has also

been shown to be a T-independent antigen. In classic experiments that defined the size of the antigen-combining site of immunoglobulin, the combining sites of antibodies to

α(1→6) dextran are shown to have an upper size limit of six to seven oligosaccharide glucose residues (226). The lower limit, based on the ability of the antibody to cross-react in precipitin tests with isomaltose oligosaccharide coupled to BSA, is between one and two glucoses (226, 227). Synthetic glycolipids prepared by coupling isomaltose oligosaccharide to steary-lamine are T-independent antigens in mice as well (228).

42

Epitope Mapping

The nature of peptide and protein antigenic determinants has also been studied and has led to useful approaches to making specific antisera, to designing vaccines, and for mapping the antibody response to protein antigens. A key concept underlying much of this work is that a short peptide can serve as a surrogate for the epitope on the intact protein. Using epitope mapping techniques, both continuous and discontinuous peptide epitopes have been defined (229, 230). In continuous epitopes, the amino acids encoding the antigenic determinant are adjacent to each other in the primary protein structure.

Discontinuous epitopes have amino acids that are brought together by the folding of the native protein. Denaturation may destroy discontinuous epitopes, and monoclonal antibodies recognizing such epitopes will not generally function in SDS-PAGE immunoblotting techniques (Western blots) that require protein denaturation.

In the epitope mapping procedure for continuous epitopes, overlapping peptides

covering the length of a protein are designed and tested for antibody binding. For example, a total of 208 overlapping hexapeptides covering the total 231 amino acid sequence of the immunologically important coat protein VP1 of foot and mouth disease virus (FMDV), type O1were synthesized on a solid phase (230) The attached peptides

were tested for antigenicity by an ELISA using a variety of antisera. The result indicates

that an immunodominant epitope is located between amino acids 146-152 of that protein.

Then 120 hexapeptides representing complete single point amino acid replacement sets

were synthesized and tested for retention of antigenicity. The result indicates that leucine 43

residues at position 148 and 151 were essential for binding of antibodies in antisera raised

against FMDV (230).

An example of a discontinuous epitope was identified with a mAb raised against

FMDV type A10 (229). The peptides mimicking the epitope were identified. No sequence

homology was found in capsid protein VP1 of FMDV type A10. The result suggests that protein folding, in which amino acid residues from three regions distant from one another in the primary sequence, are brought into close proximity at this epitope.

Mimotope

Identifying peptides that represent epitopes suggested the mimotope concept, in

which molecules that mimic the structure of antigenic determinants are constructed and

tested for antigenicity and for the ability to elicit cross-reactive antibodies to the original

antigen. Just as many pharmacologically active drugs are considered to be mimics of the

natural ligands for protein receptors, mimotopes are mimics of antigenic determinants for

protein antibodies. Because the universe of peptide mimotopes could not readily be

constructed from the primary amino acid sequence of a protein antigen, the attempt to

identify mimotopes requires large libraries of peptides, particularly of discontinuous epitopes. This has been accomplished through combinatorial chemistry and through genetic techniques such as phage display.

An alternate approach to designing mimotopes for antibodies is to use the

resemblance of anti-idiotype (anti-Id) to antigen. The idiotype (Id) of an antibody is that

part which is antigenically unique as determined by a second set of antibodies made

against the first one. Thus idiotypes are in the variable region in, at, or near the 44

combining site. Thus the combining site defines both antigen specificity and the idiotype

of an antibody. In 1964, Jerne proposed that the immune system is regulated by networks

of anti-idiotypic antibodies acting as a self-contained homeostatic network (231).

According to this hypothesis, the Id-anti-Id interactions regulate the immune response of

a host to a given antigen. The Id regulatory mechanism influences the prevalence of

different combining sites and antibody specificity. Based upon the structural studies we

have previously described, the anti-idiotype may resemble the antigen. According to the

network concept, immunization with a given antigen will generate the production of Ab1

against the antigen. When sufficient Ab1 is produced, it then induces an Ab2 (anti-Id)

against its idiotype. Some of these Ab2 can mimic the structures of external antigens.

Thus by manipulating the idiotypic network it may be possible to elicit antibodies that

cross react with the original immunogen. In this case the anti-Id is said to contain the

internal image of the antigen. Immunization with anti-Id Ab2 generates anti-anti-Id Ab3,

which recognizes the original antigen. However, there may be differences in the structure

of Ab3 and Ab1. Based upon the ability to mimic the internal image of antigen, anti-

idiotypic immunization has been studied extensively for vaccines against cancer (232)

and infectious agents. Anti-Ids elicit antibodies to microbial antigens, including proteins,

LPS, and capsular polysaccharides (233-235). More about the use of anti-idiotypes to

mimic microbial polysaccharides is detailed below.

Combinatorial protein-display libraries permit the examination of an extremely

large number of possible sequences (libraries), having as many as 109 individual members. This may be done biologically using genetic approaches or chemically through 45

directed synthesis. Biological approaches include displaying peptide sequences on phages, microbes, and even eukaryotic cells. Combinatorial design provides a complementary approach for understanding sequence and structure compatibility and discovering novel sequences that have a specific structure. In addition to creating a library with a high degree of diversity, the combinatorial approach provides materials which can be assayed for structure (e.g. sequence if protein) or function. Synthetic combinatorial libraries are not restricted to proteins that can be made by living organisms, but rather may be made with D-amino acids, carbohydrates, novel molecules (e.g. dendrimers), and so forth. Thus synthetic libraries cover a larger portion of chemical space (the range of chemical compounds that can have biological effects) than do biological expression systems. Moreover they may be used to select for materials that function in non-natural environments that would kill the biological vector. However, in our studies we have used phage display technology because of the ready availability of peptide-display libraries (e.g. Burritt (236), Smith (237)).

Phage Display Technology

Based on the mimotope concept, phage-displayed technology was introduced by

Smith (238). The construction of random peptide or gene fragment display libraries

allows researchers to map contact points in protein-protein interactions. The technology

depends upon an inserted random mutation at an appropriate location and the expression of the mutated sequence on the viral coat protein of viable phages.

The filamentous bacteriophage M13 and its derivatives have often been used as

library vehicles. The phage consists of a stretched-out loop of single-stranded DNA 46 sheathed in a tube composed of several thousand copies of the major coat protein pVIII.

At the tips of the virion, pIII and pVI minor coat proteins are involved in host-cell binding and in termination of the assembly process. Genes II and X encode replication proteins. Gene V encodes a single-stranded DNA binding protein. Gene IX encodes amino coat protein responsible for assembling pIII in the viral coat. Genes I and IV encode morphogenetic proteins. The three structural proteins, pIII, pVI and pVIII have been used for insertion of foreign sequences in the peptide libraries (237, 239). PIII is most commonly used, followed by pVIII, and very rarely pVI. Proteins pVIII are expressed at very high levels (2700 copies/phage), whereas protein pIII is expressed in only three to five copies/phage. Because pVIII represents the majority of the virion, marked constraints are placed on the insertions that may be placed in this protein. Either the length of the insert or the number of copies of the pVIII expressing the mutant protein must be limited. Insertion of greater than nine amino acid residues abolishes the viral assembly process (239). The incorporation of peptides in the pVIII protein without significant reduction in viral assembly requires either co-infection of the host E. coli with a helper phage, which encodes a wild type pVIII protein or the expression of wild-type pVIII protein on a plasmid in the infected cell (240).

The pIII protein is less critical in viral assembly and still binds to the phage receptor even with large inserts. The pIII can bear fusion proteins containing several hundreds residues at or near the amino terminus without significant loss of function. The expression of foreign peptides on pIII proteins has a minimal effect on the infectivity or assembly of virions (237, 239). 47

The assembly process of phages requires interaction between their structural

proteins and two extra non-structural proteins (pI and pIV) and also host (E. coli)

proteins (241). Displayed proteins fused to the amino-terminus of pIII and pVIII fold in

the periplasm before incorporating into the phage coat. The displayed protein, which can

contain cysteine residues, may form disulfide bonds preventing incorporation of coat

protein into the viral particle. Study of “random” insertions into pVIII indicates a bias for

certain peptides, presumably those with a suitable structural conformation (242, 243).

M13 phages infect E. coli through fertility pili (F factor). They attach to pili prior

to infection. The model for initiation of infection has been studied (244). The extreme

amino-terminal end of pIII (g3p-D1) is pulled off the next upstream domain (g3p-D2) by

interaction with the bacterial cell surface protein TolA. The interaction allows for viral

disassembly and pilus retraction. Amino-terminal inserts on pIII do not greatly affect this process.

Construction of a phage-display library requires manipulation of the double- stranded replicative form (RF) of the vector. The RF can be harvested in the same manner as plasmids. The DNA library is inserted into the vector and then complete RFs with the

DNA insertions are electroporated into competent E. coli cells. Peptide sequences to be displayed are encoded by the insertions into the genes encoding the phage coat proteins.

These may be random sequences, DNA obtained from sources that one is attempting to mimic (e.g. epitope libraries constructed from genomes of pathogenic microbes), or mixtures of constrained and random sequences. Diversity in the library can be estimated from the number of transfection events following the introduction of the recombinant 48

vector DNA into host cells. The estimation is based on the assumption that few

displayed-peptides are represented more than once.

Antibody has been used to select display libraries in attempts to identify antigenic

determinants recognized by these antibodies. Initially this was restricted to identifying

peptides bound by anti-protein antibodies. These approaches are able to define not just linear epitopes, but also conformational epitopes. The results demonstrat immunodominant regions of proteins, and also identify potential vaccine antigens.

Interestingly, phage display has also been used to identify peptides that mimic

carbohydrate and other non-protein structures.

Peptide Mimicry of Carbohydrate Structure

Although it may seem odd to seek peptides that mimic carbohydrates, naturally

occurring carbohydrate-mimetic peptides can be found. For example, the tendamistat and

other proteins that function as inhibitors of α–amylase have WRY tripeptide sequences

that bind in the enzyme’s active site (245, 246), whose “natural” substrate is the

carbohydrate amylose. Phage display technology has been use to identify peptide ligands

for the carbohydrate-binding protein concanavalin A (Con A) (247). Peptides with a

consensus sequence, YPY, bind Con A with affinity comparable to the carbohydrate

ligand. The mimetic peptide has been shown to competitively inhibit carbohydrate-

specific binding.

Initial studies that demonstrated peptide mimicry of microbial polysaccharides

were based upon the anti-idiotypic approach. Westerink et al. developed a monoclonal

anti-Id that contains the internal image on its combining site, which mimics 49

meningococcal group C capsular polysaccharide (248, 249). The anti-Id elicits a T-

dependent antibody response in mice directed toward CPS of meningococcal group C.

The sequences of the variable regions of heavy and light chains of the anti-Id reveal that

the CDR 3 is unique, in that the sequence tract YRY is exposed on the surface.

Immunizing mice with the mimetic peptide results in significant anti-meningococcal

group C antibody response. Immunized mice are protected against infection with a lethal

dose of bacteria. On the basis of these studies, other groups, including our own, have used a variety of methods to elicit antibodies to microbial carbohydrate and LPS antigens.

These are summarized in table 1.3.

Epitope mapping techniques and sequence analysis from anti-Id and phage-

displayed libraries demonstrate that antibodies bind to epitopes of three to five amino

acids within a sequence. The fine specificity of some carbohydrate-binding proteins maps

to peptides containing central W/YXY residues. Peptides with the aromatic motif

W/YXY are defined to mimic several carbohydrate subunits. These include YPY as a

mimic of mannose (ConA (247)), WRY found to mimic α(1→4) glucose, WLY (amylase

inhibitor (245) which mimics Lewis Y (250), and YRY as a mimic of the major C

polysaccharide α(2-9) sialic acid of Neisseria meningitidis(251). A protective antibody

against Cryptococcus neoformans glucuronoxylomanna was selected that displays a binding motif of TPXW(M/L)(M/L) (252). The peptide doesn’t interact with non- protective anti carbohydrate antibodies to Cryptococcus which indicates both the

specificity of selected peptides and the potential for peptide structures to differentiate

between protective and nonprotective antibodies (253). 50

Table 1. 3 Summary of Peptides that Mimic Carbohydrate Structure Carbohydrate Mimetic Comment Reference α-D-Mannopyranoside Peptide from phage- Similar consensus sequence obtained (242) displayed peptide independently by two groups library

Meningococcal group C Anti-idiotype and Elicits protective antibody when used (248, 249) polysaccharide phage-displayed anti-Id and peptide derived from anti-Id peptide library mAb as an immunogen

Siayl Lewis X antigen Peptide from phage- Peptide is a specific inhibitor of (254) displayed peptide neutrophil adhesion library

Group A streptococcal Anti-idiotype Elicits polysaccharide cross-reactive (234, 255) polysaccharide antibody when used as an immunogen

Group B streptococcal ScFv antibody (from Elicits protective antibody when used (256) polysaccharide type III anti-idiotype) as an immunogen

Group B streptococcal Peptide from phage- Elicits polysaccharide cross-reactive (1) polysaccharide type III displayed peptide antibody when used as an immunogen library

Cryptococcal Peptide from phage- Peptides are good antigenic mimetics (252, 257- glucuronoxylomannan displayed peptide but poor immunogenic mimetics 259) library

Pseudomonas Anti-idiotype Elicits protective antibody when used (260) aeruginosa as an immunogen lipooligosaccharide

Neisseria meningitidis Peptide from phage- Elicits protective antibody when used (261) group A displayed peptide as an immunogen and it is preliminary lipooligosaccharide library evidence of immunogenic mimicry

Neisseria gonorrhoeae Anti-idiotype Elicits anti-lipooligosaccharide cross- (262) lipooligosaccharide reactive antibodies

Adenocarcinoma- Peptide from phage- Elicits antibody responses that cross- (263, 264) associated carbohydrate displayed peptide react with human tumor-associated antigens (Lewis Y) library carbohydrate antigen.

HIV carbohydrate Peptide from phage- Elicits carbohydrate cross-reactive (265, 266) antigens displayed peptide antibodies library

Streptococcus Anti-idiotype Elicits protective antibody when used (235) pneumoniae as an immunogen

Streptococcus Peptide from phage- Elicits protective antibody when used (267, 268) pneumoniae capsular displayed peptide as an immunogen polysaccharide library 51

Table 1.3-Continued Summary of Peptides that Mimic Carbohydrate Structure Carbohydrate Mimetic Comment Reference Haemophilus influenza Peptide from phage- Elicits protective antibody when used (269) lipooligsaccharide displayed peptide as an immunogen library

Brucella Peptide from phage- Elicits weak immune response (270) lipopolysaccharide displayed peptide library

P-selectin Peptide from phage- Peptide inhibits the binding of glycosid (271) displayed peptide ligand for P-selectin. library

Gal-α1,3-Gal Peptide from phage- Melibiose was able to inhibit the (272) displayed peptide binding of the human natural anti-αGal library antibody to the peptide competitively

Glycosphingolipids Peptide from phage- Peptide inhibits β–galactosidase (273) displayed peptide activity towards lactotetraosylceramide. library

Lipopolysaccharide Peptide from phage- Biopanning on LPS-conjugated epoxy (274) displayed peptide beads was used to identify peptides library

Salmonella enterica Peptide from phage- Biding activity between mAb to the (275) serovar Typhi displayed peptide mimotope was blocked by CPS Vi library antigen of the bacteria.

Shigella flexnerii LPS Peptide from phage- Peptide can block the binding activity (276) displayed peptide between the bacterial LPS and mAb library

In conclusion, phage-displayed technology has proven to be a valuable multipurpose tool with which to probe protein-ligand interactions. Phage libraries are a rich source of unexpected functional diversity. Because of the very large library sizes that can be achieved, it is tempting to maximize the number of varied positions and then carry out many rounds of selection in order to increase the chance of identifying a rare, highly functional clone. It also provides an economical alternative way to identify mimic peptides. The technology requires only basic understanding of genetics to carry out 52

experiments. These reasons make the phage technology grow explosively. It has become

an indispensable tool for protein engineering. There will be further improvements in

library construction such as vector design and selection methodology that will further

enhance phage display technology.

Peptide Mimics of the Capsular Polysaccharide of Group B Streptococcus

In the Pincus lab, we have used a highly protective mAb designated S9 to select a

phage-displayed peptide library for peptides that mimic the GBS type III CPS (1). The S9

mAb is highly protective against experimental infection. With S9 treatment, animals survive an otherwise lethal bacterial infection. In isolating phage that bind S9, two

peptides have been identified, depending on the desorption conditions. With acidic

desorption, peptide WENWMMGNA was identified, and with the basic desorption,

peptide, FDTGAFDPDWPA. ELISA results demonstrate that these peptides bind to S9

and no other antibodies tested. Phage blocks the binding of S9 to type III GBS, but

doesnot block the binding of a group specific mab to GBS. Phage displaying peptide

FDTGAFDPDWPA shows a greater inhibition of GBS binding than WENWMMGNA.

Antibody S9, other type-specific mAbs, and polyclonal anti-GBS type III antisera bind

the synthetic peptide FDTGAFDPDWPA. The FDTGAFDPDWPA is named peptide S9

after the antibody S9, which is used for its selection. The binding of S9 to GBS is

inhibited by free S9 peptide with an IC50 of 30 ug/ml. The CPS and intact GBS could

block the binding activity between S9 peptide and polyclonal anti-GBS antibodies. The

result indicates that the S9 peptide bears antigenic properties of GBS type III capsular 53

polysaccharide. A conjugate S9 peptide with carrier proteins (KLH, BSA or OVA) induces significant antibody responses against both GBS and purified CPS after a single immunization. The results indicate that this peptide mimetic of the GBS capsular polysaccharide type III is both antigenic and immunogenic.

To investigate the origins of peptide-carbohydrate mimicry, the conformational

preferences of peptides that mimic the GBS type III CPS have been investigated by NMR

spectroscopy (277). The NMR spectra of the peptide FDTGAFDPDWPA reveal that the

peptide exists in two isoforms due to cis-trans isomerization. The ratio of trans to cis

isomer is estimated to be 7:3. For the trans-isomer, a type I β–turn is deleted from

residues Asp-7 through Trp-10. The type I β-turn is not represented in the population of

structures of the cis isoform. The S9 mAb also binds at - Asp-7 through Trp-10. Two new

peptides were synthesized, the peptide 1A, FDTGAFDPDWPYD, and peptide 1B,

FDTGAFDPDWPY. The Ala-12 was replaced by a tyrosine residue in both analogues.

An extra Asp residue was incorporated at the C-terminus of the 1A sequence to test the

possibility that the negative charge imparted by the Asp residue might mimic the negative

charge of the sialic acid component of the original peptide. The peptide analogues were

tested for biological activity and investigated by NMR. Peptide 1A is shown to be able to

compete with the CPS approximately twice as effectively as either the original peptide or

peptide 1B. Thus, the addition of an Asp residue at the C-terminus appears to increase the

relative avidity of the antibody for the peptide. NMR study reveals that peptide 1A, like

the original peptide, has the propensity to form type I β-turns in solution. 54

In this thesis I further investigate the structural basis of the peptide mimicry of the

GBS type III CPS carbohydrate epitope. We designed a sublibrary, expressing a mutated

version of the S9 peptide with 3% random nucleic acids at each position of the original

S9 nucleotide sequence, creating approximately one mutation in each inserted sequence.

Peptides with higher affinity to the S9 mAb were identified by affinity selection. From

this sublibrary, we also explored effects of the amino acid changes in the antigenic S9

peptide. The relationship of antigenicity and amino acid in each position is discussed in

the thesis.

Conjugate Vaccines and Carriers

Vaccines to Microbial Polysaccharides

Capsular polysaccharide (CPS) is both a microbial virulence factor and the target

antigen of many protective antibodies. As a virulence factor, CPS functions by inhibiting

phagocytosis by host immune cells (278). CPS is often a T-independent antigen that

predominantly elicits low avidity IgM antibodies and fails to induce immunological

memory. The lack of immunogenicity is most apparent in infants less than 2 years of age.

Although the polysaccharide is recognized by the B cell receptor (i.e. antibody), it usually

cannot be presented to T cells in conjunction with major histocompatibility complex

(MHC) class II molecules. The lack of specific T cell interactions in the immune

response limits the immunogenicity of the polysaccharide, the development of memory B

cells, and class-switched antibody. On subsequent antigen exposure, maturation of the immune response does not occur. However, anti-CPS antibodies are frequently protective

and the induction of anti-CPS antibody to certain bacteria is believed to be the hallmark 55 of prevention (279-281). Thus, polysaccharide capsule is attractive as a vaccine antigen.

Because the immune response to purified CPS vaccines is generally not satisfactory, further approaches to this problem must be taken.

Hemophilus influenzae type b (Hib) is one of the leading causes of invasive bacterial infection in young children (282). It possesses a polyribosyl ribitol phosphate

(PRP) capsule (283). Hib is a significant cause of bacterial infection including meningitis, septicemia, epiglottitis, pneumonia and septic arthritis, especially in children under 5 years of age. Studies have been shown that antipolysaccharide antibody is protective against invasive Hib disease (284). Thus PRP is attractive as vaccine antigen. However, purified PRP induces low titers of serum antibodies that are usually insufficient to protect against invasive disease (285-287), particularly in the target age groups. Hib conjugate vaccines, which contain PRP conjugated to peptide carriers (Diphtheria toxoid, CRM 197 mutant C. diphtheriae toxin protein, N. meningitidis protein outer membrane complex and tetanus toxoid) are shown to be more immunogenic and to show boosted responses characteristic of T-dependent memory (288, 289). Estimates of efficacy for prevention of invasive disease range from 90-100% for up to 1 year following vaccination with Hib conjugates.

The success of Hib conjugate vaccine has led to the improvement of vaccines to bacteria such as Streptococcus pneumoniae and Neisseria meningitidis. Recent results of efficacy trails of heptavalent pneumococcal conjugate vaccine are promising.

Streptococcus pneumoniae is a common bacterial agent causing a variety of infections including mucosal infections, pneumonia, arthritis, pericarditis, peritonitis and severe 56 invasive infections such as meningitis and septicemia (290). Pneumococcal polysaccharide vaccines are not effective in children under 2 years old because they are T cell-independent immunogens. The seven-valent pneumococcal conjugate vaccine includes seven purified CPS of S. pneumoniae, each coupled to a CRM 197 (291). The seven-valent pneumococcal conjugate vaccine demonstrat remarkable efficacy against invasive pneumococcal disease (292). Neisseria meningitidis is a major cause of bacterial meningitis and sepsis. Effective conjugate vaccines for the prevention of menigococcal disease cause by group A, C and Y strains have been licensed (293). Immunogenicity of these meningococcal polysaccharides is markedly improved by conjugation to carrier proteins.

The protein carriers in conventional vaccine development induce T cell responses.

T cell memory produced by conjugate vaccines is specific for peptide sequences derived from the carrier proteins of these vaccines. However, carrier protein specific T cells are unable to contribute to a secondary response to an invasive disease because they are not specific for polysaccharide antigen. This phenomenon may explain some cases of the existence of immunological memory without protective immunity.

In an effort to induce immune responses of greater magnitude, or of different quality, different forms of carriers are being investigated. For example, our own studies use a novel carrier, an RNA plant virus. Such carriers present antigen in repeated forms, and in association with virus structures.

We will now describe the development of virus carriers for presenting antigens.

Carrier structures may be used to present antigens other than bacterial polysaccharides. 57

The immune response to any material considered a hapten, because of its molecular size

or other properties, may be enhanced by conjugation to a carrier. Examples include

peptides and low molecular weight organic compounds. There are a variety of different

carrier approaches. The most popular method is chemical coupling to large proteins such

as keyhole limpet hemocyanin (KLH) or in the vaccine conjugates described in the

section above. An alternative method exists for haptenic peptide antigens, fusing the

DNA encoding the desired peptide to the gene encoding the carrier protein and producing

the protein in a genetic expression system or using it directly as a DNA vaccine.

Virus-Based Carrier Systems

Conventional carrier proteins when administered systemically may induce a

strong antibody response, but often they can be inefficient in inducing T helper (Th) or

cytotoxic T lymphocyte (CTL) responses. They are also inefficient in the induction of mucosal responses to the immunogens. The expression of antigen on viral coat proteins

by genetically fusing to viral DNA or chemically fusing to viral capsid protein, results in

a particulate structure providing a more effective stimulus for induction of local and

systemic immune responses. Virus particles are examples of highly organized and

complex structures (294). The degree of complexity of antigens is crucial to the

efficiency and quality of B cell response. Moreover, repetitiveness of antigen is used as a

self/nonself discrimination by B cells as well as by innate defenses such as Toll receptors

(295).

There are two strategies in using viruses as delivery systems or vaccine vectors.

The first strategy is making use of viruses that replicate in the host for induction of 58

immune responses, while the second strategy immunizes with preassembled virions or pseudovirions expressed in heterologous expression systems such as bacteria (E. coli),

yeast, mammalian cells, or the natural host (e.g. plant and insect viruses). Examples of

viral vectors suggested for the first approach include modified poliovirus (296) and

human rhinovirus (297). In these cases, live viruses are proposed for use as vaccine

vectors. Alternatively in the second approach, whole viruses or virus-like particles

(VLPs) self-assembled in heterologous systems, may be used. Intact viruses that have

been proposed for vaccines include plant viruses, such as plum pox virus (298), cowpea

mosaic virus (CPMV) (299), and others. The first VLP used was hepatitis virus surface

antigen containing particles expressed in yeast (300). Examples of VLPs obtained using

this approach include rotavirus (301), picornavirus (302), orbivirus (303), calicivirus

(304), herpesvirus (305) and parvovirus (306).

The strategy to create efficient delivery systems from virus or VLP consists of

three steps. The first step is identification of permissive sites in the virus particle that do

not interfere with virus structure and assembly and maintain the full immunogenicity of

the foreign epitope. The second step is purification and characterization of chimeric

viruses or chimeric VLPs. The third step is investigation of the immunogenic properties

of the chimeras.

Plant Virus-Based Vaccine

Recent advances in the molecular biology of plant viruses have enabled the

development of expression systems that use viral vectors to produce large amounts of protein in a short period of time. The general approach is to insert the foreign gene into 59 the viral genome under the control of a strong subgenomic promoter. The resulting recombinant viruses can then be introduced into the appropriate host plant by mechanical inoculation, where the virus can spread systemically throughout the plant. If the construct allows viable virus particles to be produced, the recombinant virus has the potential to spread from plant to plant. Expression of recombinant protein using viral vectors is a fast and relatively simple method for examining proteins. The infected plants can produce high amounts of protein within 1-4 weeks of inoculation. These systems can be divided according to the architecture of the virus capsid into helical (rod-shaped) viruses or icosohedral viruses. Helical viruses consist of a single strand of RNA around which is wound the capsid protein. Icosahedral viruses encapsulate their genome and associated proteins in a 20-sided particle of a defined size. An example of a helical virus system is tobacco mosaic virus (TMV). TMV was used as a virus vector to express epitope from

Hepatitis C virus (307). Another example is plum pox potyvirus (PPV). It has been developed into antigen presentation system (298). One or two copies of the 15 amino acid epitope from canine parvovirus (CPV) are inserted into the N-terminus of the capsid coat protein. Neutralizing antibodies are detected against CPV. An example of icosahedral viruses is tomato bushy stunt virus (TBSV) (308). A 13 amino acid peptide from HIV-1 p

120 with 3 amino acid linker sequence is expressed at the C terminus of the coat protein from TBSV in tobacco plants. The resulting virions resembled wild-type particles. The

HIV epitope is recognized by monoclonal antibodies and the immunization of mice results in high antibody response to both the HIV peptide and the virus vector. Cowpea 60

mosaic virus (CPMV) and cowpea chlorotic mottle virus (CCMV) are other examples of

icosahedral viruses and are the viruses used in our studies.

Cowpea Mosaic Virus (CPMV) CPMV is a positive-strand RNA virus of the family comoviridae. It has a narrow host range. Most experimental work has been carried

out in its natural host, the legume cowpea. It is a nonenveloped virus of approximately

300 A° diameter that can be isolated from infected plants in yields of 1-2 grams per

kilogram of leaves. The genome consists of two RNA molecules, RNA-1 and RNA-2.

RNA-2 encodes the capsid proteins. A systemic infection can be achieved by mechanical

abrasion of the leaf and application of both viral RNAs. An icosahedral capsid consists of

60 copies each of the large (L) and small (S) subunits. Two capsid proteins form three

distinct β-barrel domains which are arranged in the capsid in a pseudo T=3 lattice. The

crystal structure of the virus particle has been solved (309). One of the loops on the S protein was chosen for insertion of an epitope from the VP1 protein of foot and mouth

disease (FMDV) (310). The result indicates that the insertion of this epitope interfers with

CPMV infectivity and spread.

Johnson et al. have used a surface exposed reactive lysine on CPMV as the site for chemical modification (311-313). CPMV possess a single uniquely reactive lysine residue per asymmetric unit (60 per virus particle). CPMV reacts with fluorescein N- hydroxysuccinimide (NHS) ester and isothocyanate, which selectively react with available lysine residues. At neutral pH, wild-type CPMV is found to carry up to 60 dye

molecules per virion, in a dose-dependent manner (with a maximal yield at dye:viral

protein = 200:1 ). At pH 8.3, 1.5 dye molecules per asymmetric unit (90 per virion) are 61

bound at a 200:1 ratio, whereas when the ratio was 4000:1 a maximum loading of 240

dye molecules per virion is observed. In conclusion, under maximal conditions, up to 4

lysine residues per asymmetric unit could be labeled.

CPMV has a stable structure. The half-life of particles in 80% DMSO is

approximately 30 minutes. This allows chemistry to be performed on the viruses even

with relatively hydrophobic organic compounds. The study further reveals that CPMV can be conjugated to biotin. In this application, in this thesis, lysine residues on CPMV

were conjugated to a peptide that mimics the type III GBS epitope. We studied the ability

of this chimera to induce an immune response in mice.

Cowpea Chlorotic Mottle Virus (CCMV) CCMV is a member of the Bromovirus

genus and in the Bromoviridae family. The bromovirus group has five members: CCMV,

brome mosaic virus, broad bean mottle virus, cassia yellow blotch virus and spring

beauty latent virus. These viruses have a 28 nm icosahedral capsid. The genome consists

of four positive-sense single stranded RNA molecules encapsulated in three virions, each

with similar or identical capsid structures. RNA 1 and 2 are monocistronic and encode

nonstructural proteins involved in RNA-dependent RNA replication. The dicistronic

RNA 3 contains open reading frames for movement protein and the coat protein. The coat

protein is translated from RNA 4, a subgenomic RNA transcribed from minus-strand

RNA 3. CCMV induces an extensive systemic chlorosis in cowpea. Continuous

propagation of CCMV in cowpea results in an attenuated variant, CCMV, which induces

mild green mottle symptoms (314). Movement protein plays a crucial role in determining

host specificity (315) and also has been shown to modulate symptom expression (316). 62

The structure of CCMV was determined to 3.2 A° resolution by X-ray

crystallography (317). The capsid is composed of 180 chemically identical protein

subunits. The quaternary structure of CCMV displays a T = 3, quasi-symmetry with 32

prominent capsomers. The capsomers occur as 12 pentamers and 20 hexamers assembled

from the coat protein. Each protein subunit is comprised of an eight-stranded,

antiparallel, β-barrel core (from amino acids 52-176). The capsomers are formed from these β barrels. The C-terminal (amino acids 176-190) and N-terminal (amino acids 1-51) of the coat protein extend in opposite directions away from the β barrel core. The

pentameric and hexameric capsomers of the virion are linked through C-terminal

extensions originating from two coat proteins. This linkage stabilizes the noncovalent

dimer. The C-terminal extension from one coat protein subunit extends across two-fold or

quasi two-fold axes to interact with subunits in adjacent capsomers. The N-terminus from

a nearby coat protein subunit clamps the invading C-terminal arm. The N-terminal

extensions of the threefold adjacent coat proteins converge at the quasi-sixfold vertices of

the virion and form a hexameric tubular structure (β-hexamer). The β-hexamer is

proposed to stabilize virion structure. An eight-stranded, antiparallel, β-barrel core (from

amino acids 52-176) results in the outward projection of five loops. In the Mark Young

lab, the DNA encoding each loop was engineered to have BamH I restriction enzyme

sites to allow for the insertion of DNA encoding epitopes. In our study, we inserted DNA

encoding mimetic peptide of GBS type III capsular polysaccharide in the BamH I site

located on each loop of CCMV coat protein. Analysis of these chimeras was studied in

this project. 63

Wild-type CCMV is stable around pH 5.0 and sediments at 88 S. At pH 7.0 and

low ionic strength, the particle undergoes a transition state to a swollen form, where the average size of particle increases by roughly 10%. Swelling is the result of a radial expansion of the virion capsid around the three-fold axes, where there are divalent calcium ion binding sites (318). Young et al. analyzed a salt stable mutant of CCMV

(ssCCMV) (319). The ssCCMV resists disassembly in 1.0M NaCl, pH 7.5, whereas the wild-type virions completely disassemble into RNA and capsid protein under these conditions. Sequence analysis of the ssCCMV revealed a single A to G nucleotide change

at position 1484 of RNA 3 (position 134 of RNA 4). The nucleotide change results in a

lysine (K) to arginine (R) change at position 42 of the coat protein. Expression of the

K42R mutant coat protein in E. coli and then in vitro assembly produces virions with the

salt stable phenotype. The ssCCMV was used in our study to couple with a mimetic

peptide of type III GBS epitope. Another mutant, designated SubE mutant (discussed

later), was also conjugated to the peptide. We studied the ability of these chimeras to

induce an immune response in mice.

To study CCMV as a virus-based protein carrier, Young et al. developed methods

to produce CCMV in high yield in a Pichia pastoris heterologous expression system.

Expression of the coat protein in this system results in the assembly of particles as virus- like particles (VLP). The study of VLP from this expression system indicates that the

VLP are visibly indistinguishable from virus particles produced in the natural host. This system has the advantage of providing a method for expressing a range of viral protein mutants using large-scale fermentation technology. They also have investigated the 64 ability of a wide variety of CCMV coat protein mutants to be expressed and assembled into VLPs using this system. The SubE mutant, in which eight basic amino acid residues

(K and R) of the N-terminus are replaced with acidic residue glutamic acids, was modified to express in the yeast expression system. The wild-type virus and the Sub E mutant results in equivalent level of protein accumulation, as determined by immunoblot assays, of about 0.05 to 0.5 mg (from a gram of wet yeast) (320). Transmission electron microscopic analysis of CCMV VLP demonstrates intact protein cages for the wild-type and the SubE mutant. The same system was also applied in our study to produce VLP expressing the mimetic peptide epitope of GBS type III.

Both CCMV and CPMV were used in our study to present the peptide mimic epitope of GBS type III CPS. CCMV were either genetically engineered or chemically conjugated with the peptide to produce the chimera. Mice were immunized with all chimeras and the immune response against the chimera was studied in this thesis work.

Research Goals

GBS is a major causative agent of neonatal sepsis and meningitis. Our laboratory previously identified a peptide, S9, which is an antigenic mimic of the epitope that is the target of protective antibody. The goals of this study were to enhance the S9 peptide immunogenicity, for the purpose of GBS vaccine development, and to study the structural basis of antigenic mimicry by the S9 peptide. In chapter two, we conjugated the S9 peptide to CCMV as a platform to enhance the immune response. Subsequently, in chapter 3, the CCMV-S9 constructs were injected in to mice and the immunogenicity of 65

the constructs was characterized. Finally in chapter four, we created mutants of the S9 sequence to study the structural basis of antibody binding to the S9 peptide and to search for better mimetic peptides.

66

CCMV EXPRESSING PEPTIDES THAT MIMIC THE GROUP B STREPTOCOCCAL EPITOPE

Introduction

We chose to express the GBS mimetic peptide in the plant virus Cowpea

Chlorotic Mottle Virus (CCMV). The virus serves as a carrier for the mimetic peptide.

CCMV was chosen for this purpose because of its repetitive structure allowing multiple epitopes to be expressed, because the Young lab had invested time and energy in developing virus mutants that could be used for epitope expression, and because large amounts of virus could be obtained in both natural and recombinant expression systems.

CCMV is a member of the Bromoviridae family of viruses. CCMV has a 28 nm

icosahedral capsid. The genome consists of four positive-sense single stranded RNA molecules encapsulated in three virions, each with similar or identical capsid structures.

RNA 1 and 2 are monocistronic and encode nonstructural proteins involved in RNA- dependent RNA replication. The dicistronic RNA 3 contains open reading frames for movement protein and the coat protein. The coat protein is translated from RNA 4, a subgenomic RNA transcribed from minus-strand RNA 3. CCMV induces an extensive systemic chlorosis in cowpea. The structure of CCMV determined by X-ray crystallography demonstrates that the virion is comprised of 180 copies of the coat protein subunit arranged on a T = 3 icosahedral structure with 12 pentamer and 20 hexamer capsomeres (317). The Young laboratory has studied expression and assembly of virions and virus-like particles (VLPs) based on CCMV. These were utilized in these studies. 67

Initially, genetic approaches were taken to express the S9 mimetic peptide

FDTGAFDPDWPA, first in infectious virus, then as VLPs made in a yeast expression system. These viruses failed to assemble and function, and so the final constructs, used later in this thesis research were made by chemical conjugation.

Materials and Methods

Engineering CCMV-Peptide S9 Chimera (Plant Expression)

Five plasmids each encoded loop-specific BamH1 cloning sites where peptides

could be expressed on the virus coat, at amino acid positions 63, 102, 114, 129 or 161 in

the wild-type coat protein, were kindly provided by the Young lab. We inserted DNA to

encode the S9-peptide (FDTGAFDPDWPA) within the loop sequence. Plasmid vectors

for each specific loop are referred to as pCC3-63, pCC3-102, pCC3-114, pCC3-129 and

pCC3-161. Plasmids were digested with BamH1 for 2 h at 37°C. Then the digested

plasmids were dephosphorylated with a calf intestinal alkaline phosphatase (CIP). The

sense encoding the S9 peptide, 5’G ATC CCA [TTC GAT ACG GGC GCG TTC GAT

CCA GAT TGG CCA GCG] GG 3’ (IP[FDTGAFDPDWPA]) with phosphate at 5’ end

and the complementary strand (antisense strand), were purchased from Integrated DNA

Technologies. The red highlights indicate the BamHI overhang, the letters in the

parentheses indicate the S9 coding region, and blue letters indicate S9 peptide. Each

strand was heated at 95°C for 3 min and rapidly cooled down for 5 min. Sense and

antisense strands 25 ul of 50 ng/ul were mixed together, giving 1:1 molar ratio reaction.

The mixture was incubated 15 min at 50°C and then allowed to cool down to room

temperature. The S9 peptide genes were ligated into the compatible ends of restriction 68

digested BamH1 sites. The ligation reaction was performed with a ratio of 1:3, vectors: inserts with ligation reaction composed of 1 ul T4 DNA ligase (Promega, Madison, WI) and 1 ul of 10x ligation buffer (containing 100mM mgCl2 and 10 mM ATP, Promega) in

10 ul total volume. The reaction was incubated at 4°C overnight to facilitate ligation. The digested plasmids alone were used as a control. The ligation reactions were then transformed into XL2-Blue ultracompetent E. coli cells (Stratagene, Cedar Creek, TX) by mixing 2 ul of the reaction with 100 ul of the competent cells. The transformation reactions were incubated in an ice tub for 30 min and then briefly in 67°C for 42 sec. LB broth was added to the transformation reactions and the mixtures were incubated at 37°C for 1 h. The transformation reactions were spread on LB media agar with 100 ug/ul ampicillin supplement and incubated at 37°C overnight. Single colonies were picked and subcultured in LB broth. Clones were tested for the presence of the S9 insert by PCR and further, by sequencing.

Verifying Positive Clones

Clones containing inserts were identified by two steps: PCR and then sequencing

analysis. Selected clones were cultured in 3 ml LB broth medium. Plasmids were

extracted from cells by alkaline lysis method (321). 3 mls of overnight culture in LB

broth with complement of ampicillin 100 ug/ml were spun down to collect a cell pellet.

Cells were resuspended in 100 ul of 20 mM Tris buffer pH 8 and then incubated at room

temperature for 5 min. 200 ul of lysis buffer were added and the mixture was mixed by

inversion. The mixture was incubated for 10 min and then 150 ul of KAC/AC (5 M

potassium acetate, 10 M (57%) acetic acid) were added. The mixture was mixed 69 vigorously and placed in ice for 30 min. Cell debris was removed by centrifugation at 12 krpm 10 min twice. Supernatant was collected and was then mixed with an equal volume of phenol/chloroform/isopropanol. The upper phase was collected and transferred to another eppendorf tube. DNA was precipitated by mixing the mixture well with 2X volume of 95% ethanol and 1/10 volume of 3 M sodium acetate pH 5. The mixture was incubated at –20°C overnight. DNA was collected by centrifuging12 krpm 10 min. The

DNA pellet was washed with 70% ethanol and resuspended in10 ul water. 2 ul of DNA suspension was loaded in 0.8% agarose gel. Purified plasmids were amplified by PCR using primers 20 and 21, kindly provided by the Young lab. The PCR products were run in 1% agarose electrophoresis gels. Positives clones were sent for DNA sequencing analysis (Davis Sequencing, Inc, Davis, CA).

In Vitro Transcription of CCMV cDNAs for Plant Inoculation

Plasmids pCC1, pCC2, pCC3-wt and pCC3-S9 constructs were prepared for in vitro transcription by linearization with Xba1 followed by proteinase K (Sigma, St. Louis,

MO) digestion for 30 min at 65°C, phenol-chloroform extraction and ethanol DNA precipitation. Chimeric CCMV-S9 RNAs were synthesized by in vitro transcription from approximately 1.0 ug of linearized DNA with T7 Megascript (Ambion, Austin, TX) in a

20 ul reaction, containing RNA polymerase enzyme and its buffer and ribonucleotides, supplied by the company. RNA quantitation and integrity were determined by UV spectrophotometry (1OD260 = 44 ug/ml) and visualization on denatured agarose gel electrophoresis.

70

Plant Inoculation

The 20 ul mixtures containing 1 ug of each RNA 1, 2 and wt RNA 3 or chimeric

RNA 3-S9 were spotted on carbon dusted leaf of cowpea plants (Vigna unguiculata (L.)

var California Blackeye) and tobacco plants (Nicotiana benthamiana) at the primary

stage (two leaves). The inoculation was performed on 2-4 leaves. Four weeks after

inoculation, leaves were collected and kept at –80°C. Virus from collected samples was

extracted. Collected leaves were immersed in liquid nitrogen and crushed in

homogenization buffer (0.2 M sodium acetate, 0.2 M acetic acid, 0.01 M disodium

EDTA, pH4.8). The crude extract was tested for the presence of virus by ELISA and

observation under electron microscope.

ELISA

ELISA was performed as follows. Briefly, 100 ul of 1:5000 dilution of 1 mg/ml

purified rabbit anti-CCMV polyclonal antibody with phosphate buffer was absorbed on

the surface of ELISA plates at 4°C overnight. Plates then were blocked with blocking

buffer (containing BSA) 200 ul at 4°C at least 6 h. Plates were washed with ELISA

washing buffer, 10 mM phosphate buffer pH 7.4, 150 mM NaCl, 0.05% Tween 20. Plant

crude extracts were loaded in prepared plates and then incubated at 4°C overnight. Plates

then were washed again with washing buffer. A 1:500 dilution of alkaline phosphatase

(AP)-conjugated rabbit anti-CCMV polyclonal antibody was loaded in the plates and incubated at room temperature for 4 h. After washing with ELISA washing buffer, 100 ul 71

of substrate, p-nitrophenyl phosphate, were added (Sigma, St. Louis, MO). The resulting color reaction was measured with ELISA plate reader at absorbance 405 nm.

In Vitro Transcription and Translation

The CCMV-S9 plasmid constructs, pCC3-63, pCC3-102, pCC3-114, pCC3-129

and pCC3-161, encoding coat protein gene were PCR amplified using upstream primer,

which contained a T7 RNA polymerase promoter element and a start codon, ATG. The

PCR reaction resulted in a PCR product, which contained T7 RNA polymerase promoter

element and a start codon at the 5’ and stop codon at 3’ end. DNA quantitation and

integrity was determined by UV spectrophotometry and visualized by agarose gel

electrophoresis. Chimeric CCMV-S9 RNA was in vitro transcribed from PCR product by

using RNA transcription kit (mMessage mMachine@T7 kit, Ambion, Austin, TX). RNA

quantitation and integrity were determined by UV spectrophotometry and visualized by

denatured agarose gel electrophoresis. Resulting 5’ capped mRNAs were then in vitro

translated using cell free wheat germ extracts translation kit (Promega, Madison, WI).

Radioactive (S35) methionine was incorporated into the reaction. The proteins then were

visualized by using gel electrophoresis and autoradiography.

Engineering CCMV-S9 Chimeras in Pichia pastoris

CCMV wt and CCMV-S9 chimeric genes were subcloned from their original

plasmid vector pCC3-wt and pCC3-S9s into the P. pastoris expression plasmid pPicZa

(Invitrogen) to construct pPicZa 63-S9, pPicZa 102-S9, pPicZa 114-S9, pPicZa 129-S9

and pPicZa 161-S9. PCR was used to amplify the DNA cassette using primers: 5’ GAT 72

AGT AAG AAT TCA TGT CTA CAG TCT T 3’ as upstream primer, and 5’ GTA ACG

GTC GAC AGC GGG C 3’ as downstream primer. The upstream primer was complementary to a region encoding the coat protein start codon ATG (underlined sequence), and was created to introduce an EcoR I endonuclease restriction site (red block sequence) into the 5’ end of the amplified fragment. The downstream primer was complementary to the region encoding the coat protein stop codon, TAG and was also designed to contain SalI endonuclease restriction enzyme site. Thus the PCR products contained both EcoRI and Sal1 sites. Agarose gel electrophoresis was used to follow the

PCR products. The products were purified from gels using QIAquick DNA/gel purification kits (Qiagen, Valencia, CA).

The pPicZa plasmid vectors were digested with EcoRI and XhoI. The XhoI digestion created SalI compatible ends. Then plasmids were dephosphorylated by CIP as described above. The 5 digested PCR products from pCC3-S9s were ligated into digested pPicZa plasmids. The ligation reactions were then transformed to XL-2 Blue ultracompetent E. coli cells as described. Transformed E. coli were spread on low salt LB agar plates containing 25 ug/ml Zeocin. Single colonies from each ligation reaction were randomly selected and cultured overnight in low salt LB media under Zeocin selection

(25 ug/ml). Recombinant plasmids were extracted from clones by the alkaline lysis method (321). Recombinant clones were verified by PCR and DNA sequencing.

Recombinant plasmids containing pCC3-S9 were extracted by the alkaline lysis method (321) and then linearized with SacI. The linearized plasmids were then electroporated into competent P. pastoris X-33 mut+ yeast (Invitrogen) using a T800L 73

Electro Cell Manipulator (BTX, San Diego, CA) electroporation unit according to the

manufacturers’ protocol. The pPicZa does not contain a yeast origin of replication.

Transformants can be obtained only if the recombination occurs between plasmid and

Pichia genome. Zeocin resistant transformants were selected after 2-3 days growth on

YPDS agar plates (1% w/v yeast extract, 2% w/v dextrose, 1M sorbitol, 2% w/v agar)

containing Zeocin (100 ug/ml). Single colonies were subcultured into 3 ml of YPDS

broth containing Zeocin (100 ug/ml) and incubated at 30 °C overnight. Direct PCR-based

screening of clones was performed (322). Briefly, 10 ul of P. pastoris culture (1:10

diluted with water) was disrupted using 25 U lytiase (Sigma). Primers for CCMV coat

protein were used to amplify DNA. The PCR products were observed by gel

electrophoresis and sent for DNA sequencing analysis (Devis Sequencing, Inc., Davis,

CA).

Protein Expression and Small Scale Time Course Experiment

Positive yeast clones were cultured in 25 ml MGY (1.34% w/v yeast nitrogen base without amino acids (YNB), 1% v/v glycerol, 0.00004% w/v biotin) 30°C for 18 h

250 rpm. For induction of the coat protein expression, the cells were harvested by centrifugation, washed once with MM medium (1.34% w/v YNB, 0.00004% biotin, 0.5% methanol) and resuspended in 150 ml MM medium. During the incubation, methanol was

added to the culture at 24 h intervals, each time to give a final concentration of 0.5%.

Samples of 1 ml were collected every day for 7 days. Cells were stored at –80°C. 74

Medium Scale Expression of CCMV-S9 Chimeras in P. Pastoris

Five positive recombinant Pichia clones (pCC3-63, pCC3-102, pCC3-114, pCC3-

129 and pCC3-161) were cultured in 50 ml MGY for 24-48 h at 30 °C on a 250 rpm incubation shaker. Cells were harvested by centrifugation and washed with MM medium.

The washed harvested cells were then inoculated into 500 ml of MM medium. 100 % methanol 2.5 ml (0.5% final concentration) was added to the culture at 24 h intervals. The culture was incubated for five days at 30°C in 250 rpm incubation shaker. Cells were harvested by centrifugation and kept in -80°C for further analysis.

Large Scale Protein Expression

Improved overall recombinant protein yields in Pichia have been achieved by large scale Pichia fermentation process. The fermentation method followed the

Invitrogen Pichia fermentation process guidelines. A Bioflo 3000 bench-top fermenter

(New Brunswick Scientific) with a 5L water-jacketed glass fermentation vessel was used in the project. The fermentation procedure was separated into three phases, glycerol batch phase, glycerol fed-batch phase and methanol fed-batch phase. Dissolved oxygen (DO) measurement was used as the tool to observe and manipulate the metabolic rate of the culture during all 3 phases of fermentation process.

The positive yeast clone producing the largest amount of protein was selected based on time course data. This recombinant clone was cultured in 300 ml MGY medium

(10% of initial fermentation volume) at 30°C, 250-300 pm, 16-24 h as an inoculum seed for the fermentation. The glycerol batch phase began with adding the inoculum seed into 75

the fermentor vessel containing 3L of fermentation basal salts medium (85% v/v

phosphoric acid, 0.93 g/L CaSO4, 18.2 g/L K2 SO4, 14.9 g/L Mg SO47H2O, 4.13 g/L

KOH and 4% v/v glycerol). The batch culture was allowed to grow until the glycerol was completely consumed (18-24 h). The DO was set to 100% before the culture started to grow. As the culture grew, oxygen was consumed resulting in a decrease of DO. DO was maintained at 20% by adding oxygen and providing agitation. When glycerol in the medium was completely utilized by culture (18-24 h), yeast stopped growing. The oxygen in medium was not consumed. Thus the DO was close to 100%. Therefore, we measured the DO when the agitation tool and the oxygen pump stopped. If the DO measurement was close to 100%, the next phase was started. Next, the glycerol-fed batch was started by pumping in a culture medium containing 50% w/v glycerol at a feed rate of 18.15 ml/h/L initial fermentation volume. This phase improved the cellular yield from

180 to 220 g/L wet cells. Glycerol feeding was carried out for 4 h and stopped 30 min before starting the next fermentation phase to assure that cells used up all glycerol in the medium. The DO measurement was again used as an indicator of when to start the next phase. DO close to 100% indicated the complete consumption of glycerol. The methanol-fed batch phase was initiated by gradually feeding 100% methanol into the vessel to allow cells to adapt to growth on methanol. The DO measurement was used as the indicator to tell how cells adapted to the methanol feeding. A reduction of the DO indicated that the culture was able to adapt to methanol. However if the DO was lower than 20%, it indicated that the yeasts had died and the methanol feeding needed to be stopped. After full adaptation, methanol was fed with the rate of approximately 7.3-10.9 76

ml/h/lL feed rate. The entire methanol fed-batch phase lasted approximately 70-100 h

with a total of approximately 750-1000 ml methanol fed per liter of initial volume.

Temperature during all fermentation phases was maintained at 30°C. The DO was kept above 20% through out the fermentation process by agitating and adding oxygen. Foam was controlled by manually adding an antifoam agent, Silwel L-7602 (Witco

Corporation, Michigan). The cellular yield could reach a final level of 350-450 g/L after the full fermentation cycle. Cells were harvested by centrifugation and wet cells were preserved in -80 °C for further analysis.

Chimeric Virus Purification

Frozen cells from medium scale culture were thawed and lysed using acid-washed

glass beads in an equal volume of virus buffer (0.1 M sodium acetate/acetic acid, 1 mM

sodium azide, 1 mM disodium EDTA, pH 4.8, 0.22 um filtered). The mixture was

vortexed vigorously 4 times for 5 min, with 2 min on ice in between. Cell debris and

glass beads were separated by centrifugation at low speed. Chimeric viruses in

supernatant were precipitated by adding 10% (w/v) polyethylene glycol (PEG) MW

8000. The mixture was then slowly stirred overnight at 4°C. The mixtures were

centrifuged and the pellets were resuspended in 1/10 initial supernatant volume virus

buffer. The mixture was then centrifuged at low speed. Supernatants containing virus

were kept in -80°C for further analysis.

For the fermentation experiment, 100 g of collected cells were lysed by grinding

with 235 g acid-washed glass beads in a bead-beater (BioSpec Products-BSP, model

1107900) in the presence of 200 ml homogenization buffer. The grinding cycle process 77

was to grind for 5 min and rest on ice for 3 min. To completely break cells, 5 cycles of

grinding were performed. Beads and cell debris were separated by centrifugation and approximately 0.2 g of cell debris was kept at -80°C for further analysis. Supernatant was

added with the same buffer to a final volume of 500 ml and impurities eliminated by

using filter paper (Water/bushman) 0.6 um. 200 ul of filtrate was kept at -80°C for further

analysis and the remainder was centrifuged using a Beckman Type 30 Rotor at 25 krpm

for 2 h. Supernatant was kept at -80°C for further analysis and the pellet was resuspended

in 1 ml virus buffer. The virus mixture was centrifuged at low speed to pellet out debris and aggregates.

To improve viral purification from the fermentation culture, diethylamino ethanol

(DEAE) was applied in the process. 100 g of collected cells were lysed by grinding in the

presence of 200 ml 50mMTris-HCl buffer pH 6.5. In one experiment the grinding process

was performed in the presence of 2 M urea to increase recombinant viral protein

solubility and disrupt noncovalent bonds between the viral coat protein and yeast cell

debris. Beads and cell debris were separated by centrifugation and approximately 0.2 g of cell debris was kept at -80°C for further analysis. Supernatant was added to 50 mM Tris-

HCl buffer pH 6.5 to have final volume of 500 ml. DEAE 2 g was added to the supernatant. The mixture was stirred slowly at 4°C for 2 h. The mixture was then filtered

(Buchner funnel and filter paper (Water/bushman 90 mm) and washed with 50 ml of the same buffer twice. The filter cake was resuspended in 50 ml virus buffer supplement with

0.5 M NaCl. The suspension was filtered and the filtrate was collected. To elute all chimeric constructs from the filter cake, the filter cake was resuspended one more time 78

with 50 ml of the same virus buffer supplement with 0.5 M NaCl. The suspension was filtered. The filtrate was collected and pooled with the filtrate from the previous step and then centrifuged using a Beckman Type 30 Rotor at 25 krpm for 2 h. The virus- containing pellet was resuspended in 1 ml virus buffer. The virus suspension was briefly centrifuged at low speed to pellet out debris and aggregates.

SDS PAGE, Western Blot, and Dot Blot Analysis

For [S35]-labeled in vitro protein translation products, samples were boiled with

gel loading buffer (0.5 M Tris buffer pH 6.8: 10% SDS: Glycerol (1:1:1), bromophenol

blue, 0.036% betamercaptoethanol) and then loaded onto pre-cast 12.5% SDS

polyacrylamide gel according to the manufacturer guidelines (PhastSystem, Pharmacia

Biotech, Uppsala, Sweden). For other purposes, samples were boiled in the gel loading

buffer and then loaded onto 12.5% SDS polyacrylamide gels. Gels were visualized by

silver staining or proteins in gel were transferred to polyvinylidene fluoride (PDVF)

membranes (Sigma), by using western blot apparatus (Bio-Rad, Laboratories, Hercules,

CA). For dot blot analysis, prepared boiled samples were directly dotted into membranes

and air-dried for 10 min before proceeding. Membranes from both western and dot blot

were then blocked of any nonspecific binding overnight at 4°C with blocking buffer (5% non fat dry milk, 0.2% tween 20 in PBS buffer pH7.4). Membranes were washed briefly with ELISA washing buffer 3 times for 1 min each. Membranes were incubated overnight at 4°C with blocking buffer containing primary antibody: either 5 ug/ml S9 monoclonal antibody or 1/5000 anti-ccmv antibody. After 3 washes with the ELISA washing buffer, membranes were incubated with blocking buffer containing secondary antibody either 79

1/1000 alkaline phospatase-conjugated anti-mouse IgM antibody or 1/1000 Horseradish

peroxidase-conjugated anti-rabbit antibody. The colorimetric reaction was then

developed (Bio-Rad, Laboratories, Hercules, CA).

CCMV Wild-Type Plant Virus Purification

Frozen CCMV infected leaves 100 g were blended for approximately 3 min in

100 ml homogenization buffer. The homogenate was filtered through 2 layers of

cheesecloth. The filtrate was allowed to sit on ice for 1 h before low speed centrifugation

at 20,000 g for 20 min. The supernatant was precipitated by adding 10% (w/v) PEG MW

8000 followed by overnight stirring at 4ºC. The pellet containing CCMV virus was

collected by centrifugation and resuspended in virus buffer. A second round of viral

precipitation was carried out with 15% (w/v) PEG MW 8000 and resuspended in virus buffer. Next the suspension containing CCMV virus was purified by centrifuging in 38%

(w/v) CsCl at 38000 rpm for 20 h (Beckman, SW41). The virus band was collected and dialyzed against virus buffer. Viral concentration and impurity were determined by measuring UV absorbance ratio Abs260/280 and observing the presence of virus under

transmission electron microscope (TEM). The Abs260/280 =1.5-1.7 is considered good. The

viral concentration was calculated based on absorbance at 260 nm (1Abs260 = 0.17

ug/ml).

Chemical Linking between CCMV Virus and S9 Peptide

Two approaches were used to chemically conjugate S9 peptide to CCMV. The

first was to conjugate through a thiol group by forming a disulfide bond between S9 and 80

CCMV. S9 peptide with an added cysteine at the amino terminal was synthesized for this

purpose. The CCMV mutant, CCMV A163C with cysteine substituted at position 163,

was used in the experiment. The reaction was performed in the presence of 10 mM

CuSO4, 1 mg of virus, and 1 mg of S9 peptide at room temperature for 2.5 h. For analysis, 10 ul of the coupling reaction were loaded on SDS nonreducing PAGE gel. Gels were silver stained to visualize the conjugation product.

The second approach used the chemical linkers N-succinimidyl 3-(2-

pyridyldithio) propionate (spdp) and Sulfosuccinimidyl 4-(N-maleimidomethy)

cyclohexane-1-carboxylate (sulfo-smcc), which were purchased from Pierce (Rockford,

IL).

The spdp is a heterobifunctional cross-linker. It is composed of two functional

groups, NHS ester and 2-pyridyldithio group. The NHS ester reacts with the primary

amine on the viral coat protein. The 2-pyridyldithio group reacts with sulfhydryl at the

amino terminal of S9 peptide, forming disulfide bond between the linker and S9 peptide

and releasing pyridine 2-thione (Figure 2.1).

To prepare for the cross-linking reaction, virus ssCCMV, originally kept in the

virus buffer pH 4, was dialyzed against 20 mM NaH2PO4 pH 7.5. After dialysis, the pH of virus was measured again and adjusted with 1M NaH2PO4 pH 9 to pH 7.5. The freshly

prepared spdp linker in dimethyl sulfoxide (dmso) was used in 5-fold molar excess to

CCMV capsid protein (MW of CCMV capsid protein = 22,000). The pH of the reaction

was adjusted to pH 7.5 and the reaction was incubated at room temperature for 1 h in a

shaker. The reaction was stopped by incubating 1 h at room temperature with 50 ul of 81 hydroxylamine 1.5 M pH 8.5. The pH of the reaction was adjusted to 8.5 with 5 M sodium acetate pH 4.5. The S9 peptide in a 50-fold molar excess to CCMV was added.

The pH of the reaction was adjusted to pH 6.5 with 5 M sodium acetate buffer pH 4.5.

The reaction was performed at room temperature for 1.5 h. Then 5 ul of reaction sample was mixed with 5 ul of SDS loading buffer without β-mercapto ethanol and boiled for 10 min. The samples were loaded onto a 12% SDS-PAGE gel and visualized by silver staining or transfer to a PVDF membrane for western blot analysis. The conjugation product was also characterized by sucrose gradient velocity analysis.

O

N O C C C S S H H O 2 2 N O

SS

pH 7.5

SS NH2

NH C C C S S H H O 2 2 N

HS Disulfide exchange Peptide pH 6.5

SS SS1 SS2

NH CC C S S NH CC C S S C CC NH O H2 H2 O H2 H2 H2 H2 O

+ S + S S N N N

Figure 2. 1 Chemical Reaction between SPDP, CCMV and S9

Sulfo-smcc consists of an NHS-ester and a maleimide group connected with a spacer arm. NHS-esters react with primary amines on CCMV, and maleimides react with the thiols on the cysteine-containing S9 peptide. Primary amines are the principal targets for the NHS-ester. Thus the e-amine of lysine available on the capsid protein will react 82

with the NHS-ester. An amide bond is formed when the NHS ester conjugation reagent

reacts with primary amines releasing N-hydroxysuccinimide (Figure 1. 2).

Figure 2. 2 Chemical Reaction between Sulfo-SMCC, CCMV and S9

To begin the coupling process, ssCCMV in the virus buffer pH 4.8 was dialyzed

against 20 mM NaH2PO4 pH 7.5. The pH of the virus was measured again before the

coupling reaction, and adjusted to pH 7.5 with 1M pH 9 NaH2PO4. Tris (2-carboxyethyl)

phosphine (TCEP) 0.5 M was added to the virus, to a final concentration of 50 mM, to reduce protein disulfide bonds in viral coat protein. The reaction was allowed to continue

at room temperature for 1 h. An alkylating sulfhydryl reagent 2-iodoacetamide (IA) 0.5

M was added to the virus to 50 mM final concentration and incubated at room

temperature for 30 min. The virus solution was purified to get rid of excess TCEP and the

IA by using microcon 100 filters (Millipore, Billerica, MA). The virus was washed

extensively 3 times with 20 mM NaH2PO4 pH 7.5 and eluted with the same buffer.

Freshly prepared sulfo-smcc linker was used in 25, 50, 100 or 200-fold molar excess to 83

the virus. After adding the linker, the pH of the reaction was remeasured and adjusted to

7 with 1 M NaH2PO4 pH 9. The reaction continued for 12 h at 4°C. To stop the amide

reaction between virus and the linker, lysine was added to the reaction to a final

concentration of 1 mMand kept for 15 min at room temperature. The excess linker and lysine were removed by passing through a size-exclusion filter microcon 100, washed extensively 3 times and eluted with the 20 mM NaH2PO4 pH 6. The S9 peptide was

pretreated with 50 mM TCEP for 1 h at room temperature prior to coupling with the

derivatized virus. The pretreated S9 peptide was used in 12.5, 25, 50, 100 and 200 molar

excess to the ssCCMV. The pH of the reaction was adjusted to pH 6 with 1 M NaH2PO4 pH 9. The reaction of maleimides with thiols was incubated for 6 h at 4°C. 5 ul of reaction sample was mixed with 5 ul of SDS loading buffer containing β-mercapto ethanol and boiled for 10 min. The samples were loaded into 12% SDS-PAGE gel and visualized by silver staining technique or transfer to a nitrocellulose membrane for western blot analysis. The conjugation product was also characterized by sucrose gradient velocity analysis.

Sucrose Gradient Analysis

Samples were separated on 10-40% or 5-25% sucrose (w/v virus buffer)

sedimentation velocity gradients. Gradients were poured in Beckman SW41 centrifuge

tubes (10 ml/tube) using a Gradient Master (BioComp Instruments, New Brunswick,

Canada). 500 ul samples were overlaid and then centrifuged at 37,000 rpm for 3 h. The

gradients were manually carefully fractionated obtaining 1 ml/fraction. Abs280 of

gradient fractions was measured for presence of virus. 10 ul of each fraction were mixed 84

with loading dye to load into SDS-PAGE gel analysis. Gels were silver stained or/and blotted to PVDF membranes for western blot analysis.

Results

Plant Expression System

Plasmids containing cDNA of RNA3 were kindly provided by the Young lab.

Plasmids of RNA3 were designated pCC3-63, pCC3-102, pCC3-114, pCC3-129 and pCC3-161 according to the position of loop-specific BamH1 cloning sites. A synthetic

oligonucleotide encoding the S9 peptide sequence (FDTGAFDPDWPA) flanked with

BamHI restriction sites was inserted into the loop sequence BamH1 site of each plasmid.

The positive recombinant clones were verified by PCR and DNA sequencing analysis.

PCR products from positive clones are 677 bp compared to vectors, which all are 632 bps

resulting in the slight mobility shift for 45 bp differences (Figure 2.3).

All positive clones were sent for DNA sequencing analysis to confirm the result.

Only one positive recombinant clone from each construct was kept for further study and

were named pCC3-63-S9, pCC3-102-S9, pCC3-114-S9, pCC3-129-S9 and pCC3-161-

S9. Each construct needed to be transcribed in vitro to RNA. For in vitro transcription,

recombinant clones and cDNA of CCMV-RNA1 and CCMV-RNA2 (kindly provided

from Young lab) were linearized with XbaI, followed by proteinase K treatment,

phenol/chloroform extraction and ethanol precipitation. RNA1, RNA2, RNA3-wt and

chimeric CCMV-S9 RNAs were synthesized by in vitro transcription from linearized

DNA with T7 Megascipt (Ambion, Austin, TX) as described in the Materials and 85

Methods section. RNA quantitation and integrity were determined by UV spectrophotometry and visualization on denaturing agarose gel (Figure 2.4). We obtained approximately 100 ug of RNA1 and RNA2 and approximately 12 ug of wt RNA3 and all chimeric RNA3-S9s from in vitro transcription process.

PCC3-63 pCC3-102

pCC3-114 pCC3-129

pCC3-161

Figure 2. 3 PCR Analysis Recombinant Constructs S9 DNA sequences, 45 bp, was inserted into pCC3-63, pCC3-102, pCC3-114, pCC3-129, and pCC3-161 plasmid. PCR analysis using primer 20 and 21 was applied to identify positive recombinant clones. The first lane in each panel is 1 Kb standard DNA marker. The second lane is the pCC3-63,pCC3-102, pCC3-114, pCC3-129 or pCC3-161 plasmid vector. The remainders are recombinant clones. The mobility shifts indicated the positive recombinant clones.

86

From left to right are RNA products from in vitro transcription pCC1 (cDNA of RNA1), pCC2 (cDNA of RNA2), pCC3-wt (cDNA of wt-RNA3), the recombinant clone pCC3-63 and the recombinant clone pCC3-102 respectively.

From left to right are RNA products from in vitro transcription pCC1 (cDNA of RNA1), pCC2 (cDNA of RNA2), pCC3-wt (cDNA of wt-RNA3), the recombinant clone pCC3-114 and the recombinant clone pCC3-129 respectively.

From left to right are RNA products from in vitro transcription pCC1 (cDNA of RNA1), pCC2 (cDNA of RNA2), pCC3-wt (cDNA of wt-RNA3) and the recombinant clone pCC3-161 respectively.

Figure 2. 4 RNA Products from In Vitro Transcription of pCC1, pCC2, pCC3-wt and Recombinant pCC3-63-S9, pCC3-102-S9, pCC3-114-S9, pCC3-129-S9, and pCC3-161- S9 Positive recombinant pCC3-S9 clones, pCC1 encoding coat protein1 and pCC2 encoding protein 2 were in vitro transcribed to obtain RNAs. The RNA products were visualized by gel electrophoresis.

For plant inoculation, 1 ug of RNA1, RNA2, and RNA3-wt or chimeric RNA3-S9 from in vitro transcription reaction were mixed together and 20 ul/leaf administered.

Leaves of cow pea plant (Vigna unguiculata (L.) var. California Blackeye) and tobacco 87 plants (Nicotiana benthamiana) were carbonrundum dusted at the primary stage (two leaves) to create abrasion. The RNA cocktails were manually inoculated. Three weeks following inoculation, infected leaves were collected. The leaf crude extract was subjected to ELISA to detect the presence of wt CCMV and the chimeric viral particles

(Table 2.1).

Table 2. 1 Elisa Result of Leaf Crude Extract Construct Tobacco Plant Cowpea Plant

Repetition1 Repetition2 Repetition1 Repetition2

Mean+ SD Mean+ SD Mean+ SD Mean+ SD

Wild-type 0.672+0.002* 0.571+0.009 1.205+0.006 1.206+0.008

63-S9 chimera 0.05+0.003 0.071+0.010 0.051+0.007 0.087+0.008

102-S9 chimera 0.055+0.005 0.056+0.005 0.021+0.005 0.065+0.005

114-S9 chimera 0.061+0.007 0.091+0.004 0.029+0.007 0.098+0.005

129-S9 chimera 0.089+0.006 0.051+0.007 0.009+0.007 0.188+0.004

161-S9 chimera 0.095+0.007 0.082+0.008 0.018+0.005 0.095+0.008

PBS 0.087+0.008 0.080+0.004 0.042+0.010 0.098+0.009

*OD405 detected in ELISA well, mean of triplicate samples + SD Inoculated leaves were collected after 3 weeks of infection. Leaves were frozen by immersing in liquid nitrogen. Then, 20 g of frozen leaves were crushed by placing in a blender with 20 ml of homogenization buffer. Supernatants were collected after centrifugation. Each crude extract was diluted 1:3 with PBS buffer and 100 ul of the dilution were loaded onto anti-CCMV coated wells. Elisa was performed as described in the materials and methods section

88

The ELISA result indicated the absence of chimeric viral particles in the crude

extract when compared with wild type inoculation. The crude extracts were then studied

by transmission electron microscope for the presence of viral particles. Virus particles

were seen following inoculation with constructs encoding wild-type virus, but not with the chimeric constructs (not shown). The RNA from each construct was in vitro

translated to assure that each construct was capable of expressing a translated protein.

The CCMV-S9 plasmid constructs, pCC3-63-S9, pCC3-102-S9, pCC3-114-S9, pCC3-

129-S9 and pCC3-161-S9, encoding recombinant S9-coat protein gene were PCR amplified using primer, which contained T7 RNA polymerase promoter element. The

PCR amplified fragments were quantified and visualized by gel electrophoresis (Figure

2.5).

Figure 2. 5 Introduction T7 RNA Polymerase to pCC3 Recombinant DNA The CCMV-S9 plasmid constructs, pCC3-63, pCC3-102, pCC3-114, pCC3-129, and pCC3-161 were PCR amplified to introduce T7 RNA polymerase promoter to the 5’ end of amplified products. From left to right are 1kb DNA marker, pCC3-wt, pCC3-63, pCC3-102, pCC3-114, pCC3-129, pCC3-161 and negative control (water).

Chimeric CCMV-S9 RNA was in vitro transcribed from PCR products by using

T7 mMessage mMachine (Ambion, Austin, TX) resulting in 5’ capped mRNAs. The chimeric CCMV-S9 RNAs were then in vitro translated using cell free wheat germ 89 extract translation kit (Promega, Madison, WI). Radioactive (S35)methionine was incorporated into the reaction. The proteins then were visualized by using gel electrophoresis and autoradiography (Figure 2.6).

pCC3-wt 61-S9 102-S9 114-S9 !29-S9 161-S9

Figure 2. 6 In vitro Translation of RNA Construction.

The RNA from each construct was able to be in vitro translated resulting in

CCMV coat protein-S9 product. This data assured that each construct is capable of expressing a translated protein. All these data indicated that the insertion of the S9 peptide into extended loops of the CCMV viral capsid protein may inhibit viral production in its natural host, the cow pea plant.

Yeast Expression System

Due to our failure to express infectious CCMV-S9 chimeras in plants, which are the natural host, we chose a yeast expression system as an alternative method to create chimeric CCMV-S9 capsids. The Pichia expression system has been used widely to express high levels of heterologous proteins and intact virions (320, 323-325). Pichia 90

pastoris is a methylotrophic yeast cable of using methanol as a sole carbon source. It

contains the AOX1 gene responsible for alcohol oxigenase activity in the cell. We subcloned cDNA of RNA3-wt and RNA3-S9 from each loop construct to the Pichia

expression vector, pPicZA. The insertion of recombinant and wt cassettes was

accomplished in two steps. First, the genes for the RNA3s encoding wt and recombinant

constructs were PCR amplified from the pCC3 recombinant clones, and EcoR1 and Sal1

restriction enzyme sites added to the ends of the sequences during the PCR. The PCR

products were verified by gel electrophoresis (Figure 2.7) and DNA sequencing analysis.

The slight band shifts indicated the different size between wt and CCMV-S9 construct.

Then the PCR products were digested with enzymes EcoR1 and Sal1.

1 kb wt 63-S9 102-S9 114-S9 129-S9 161-S9

Figure 2. 7 Introduction EcoR1 and Sal1 Restriction Enzyme Sites to the Ends of pCC3- wt, pCC3-63-S9, pCC3-102-S9, pCC3-114-S9, pCC3-129-S9 and pCC3-161-S9 by PCR From left to right are 1 kb DNA marker, pCC3-wt, pCC3-63-S9, pCC3-102-S9, pCC3- 114-S9, pCC3-129-S9 and pCC3-161-S9

91

Second, the digested products were ligated into prepared pPicZA vector, which

was cut with EcoR1 and Xho1 and dephosphorylated with CIP. The constructs were linearized by using Kpn1 restriction endonuclease to reduce negative clones. The Kpn1

cut only the plasmid vector not recombinant plasmids. The recombinant plasmids were

transformed into E. coli competent cells. Positive recombinant clones were verified by

PCR (Figure 2.8). Clones, which have same size PCR products as positive control pCC3-

161-S9 were selected and sent for DNA sequencing analysis. The positive recombinant

clones were named pPicZa 63-S9, pPicZa 102-S9, pPicZa 114-S9, pPicZa 129-S9, and

pPicZa 161-S9.

Figure 2. 8 PCR Products from Positive Recombinant E. coli Clones From left to right are 1 kb DNA marker, positive control from pCC3-161-S9, pPicZa 63- S9, pPicZa 102-S9, pPicZa 114-S9, pPicZa 129-S9, pPicZa 161-S9 and negative control (water)

Next, the positive recombinant plasmids were extracted and linerized with Sac1

and then electroporated into Pichia pastoris competent cells. The Zeor transformants were 92

selected after 2-3 days growth on YPDS agar plates containing Zeocin 0.1 mg/ml. Direct

PCR-based screening of clones was performed as described in the Materials and Methods

section. PCR products were visualized by agarose gel electrophoresis (Figure 2.9) and

then sent for DNA sequencing analysis to confirm positive clones.

Figure 2. 9 PCR Products from Recombinant Yeast Pichia pastoris Chromosomal DNA Transformants from YPDS plates containing Zeocin were selected and PCR amplified by using primers 20 and 21. From left to right are DNA marker 1kb, PCR product from pPicZa wt transformant (kindly provided by Young lab), recombinant pPicZa 63-S9, pPicZa 102-S9, pPicZa 114-S9, pPicZa 129-S9, pPicZa 161-S9 transformants and pPicZa wt transformant.

To show that cloning of CCMV-S9 DNA inserts into pPicZa plasmid was in

frame and able to express CCMV-S9 coat protein, in vitro translation was performed. The yeast positive clones containing pPicZa 63-S9, pPicZa 102-S9, pPicZa 114-S9, pPicZa

129-S9, and pPicZa 161-S9 respectively, were PCR amplified using upstream primer, which contained a bacterial phage T7 RNA polymerase promoter element and a start codon. Chimeric CCMV-S9 RNA was in vitro transcribed from PCR products by using

T7 mMessage mMachine (Ambion, Austin, TX). Resulting 5’ capped mRNAs were then 93 in vitro translated using cell free wheat germ extracts translation kit (Promega, Madison,

WI). The proteins then were visualized by using gel electrophoresis and autoradiography

(Figure 2.10). The result showed that the cloning of CCMV-wt and recombinant CCMV-

S9 into Pichia plasmids were in frame and capable of being translated in vitro.

Positive pCC3-wt 61-S9 102-S9 114-S9 129-S9 161-S9

Figure 2. 10 In vitro Translation from pPicZa-pCC3-S9 Constructs

The ability of the positive yeast clones to express CCMV coat protein was investigated. The expression of CCMV-S9 coat proteins in transformed yeast was analyzed by adding methanol as inducer. Methanol induces the alcohol oxidase (AOX) promoter, which controls expression of integrated coat protein construct. Small scale time course experiments were created to observe the ability of recombinant yeasts to express chimeric CCMV-S9 coat protein. As described in the Materials and Methods section, the positive recombinant yeast clones were cultured in 25 ml MGY medium at 30°C for 18 h

250 rpm as inoculums for small scale time course experiments. Then, the cells were harvested by centrifugation and washed with MM medium and resuspended in 150 ml

MM medium. The inducer of gene expression, 100% methanol, was added to the culture 94

at 24 h intervals. Samples were collected every day for 7 days. Dot blot analysis was applied to evaluate protein expression (Figure 2.11).

Day 1 2 3 4 5 6 7 CCMV

Pichia wt

63-S9

102-S9

114-S9

129-S9

161-S9

Figure 2. 11 Small Scale Time Course Protein Expression by Dot Blot Analysis Recombinant yeast expressing chimeric CCMV coat protein 63-S9, 102-S9, 114-S9, 129- S9 and 161-S9 were cultured for 7 days in methanol inducing medium. Culture samples were collected each day. Samples were centrifuged and resuspended in 100 ul SDS gel loading buffer. The cell suspensions were boiled for 10 min and cooled down in an ice tub. Samples were centrifuged. 3 ul of supernatants were dotted onto the nitrocellulose membrane. The samples from first horizontal lane to last horizontal lane are nontransformed Pichia wt, transformed yeast with CCMV 63-S9 102-S9, 114-S9, 129-S9 and 161-S9 respectively. From left to right, they are transformed yeasts or yeast-wt on day 1 to day 7 respectively and the last lane is the positive control 0.5 ug/ml of CCMV- wt purified from yeast, kindly provided from Young lab. The membrane was blocked with blocking buffer at 4°C over night. After washing with washing buffer, membrane was incubated with anti-CCMV antibody 1/5000 dilution and finally 1/1000 Horseradish peroxidase-conjugated anti-rabbit antibody. Colorimetric detection was developed.

The dot blot result indicated that there was no CCMV protein detected from any

CCMV-S9 construct, but we easily detected product from the CCMV wild-type 95

construct. The undetectable product in the dot blot may result from insufficient CCMV

protein in samples. Western blot analysis with anti-CCMV antibody was performed using

the same samples from day 7 (Figure 2.12).

1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9

Figure 2. 12 Silver Staining and Western Blot Analysis of Small Scale Expression Experiment in CCMV-S9 Transformed Yeasts Recombinant yeasts expressing chimeric CCMV-S9 coat proteins, 63-S9, 102-S9, 114- S9, 129-S9 and 161-S9 and CCMV-wt coat protein were cultured in methanol- inducing media for 7 days as described in the Materials and Methods section. Samples were collected at day 7. Cell pellets from 1.5 ml cultures were resuspended in SDS loading buffer and boiled for 10 min and placed on ice. Samples were centrifuged and 5 ul of CCMV-wt and 20 ul of CCMV-S9 were loaded into 2 of 10% SDS PAGE gels. From left to right, samples are CCMV-wt day 6, CCMV-wt day 7, high molecular weight marker, 63-S9 day 7, 102-S9 day 7, 114-S9 day7, 129-S9 day 7, 161-S9 day 7 and 161-S9 day 6. One gel was visualized by silver staining and the other was analyzed by western blot analysis using anti-CCMV antibody 1/5000 as primary antibody and mouse anti-rabbit IgG-HRP as secondary antibody. Colorimetric detection was developed.

The western blot result indicated that recombinant CCMV proteins were

expressed in each construct. However the expression yield of the S9 constructs might

have been too low to be detected by dot blot analysis. CCMV wt cassette was expressed

in higher yields, compared to the chimeric constructs. To prepare sufficient CCMV-S9 96

chimeric protein, a medium scale culture of 500 ml for 7 days was performed. Five positive recombinant Pichia clones (pCC3-63, pCC3-102, pCC3-114, pCC3-129 and pCC3-161) were cultured in non-methanol media, MGY for 24-48hs at 30 °C with 250 rpm incubation shaker, and then transferred to methanol-induced medium for 5 days.

Yeast cells were harvested by centrifugation and CCMV-S9 coat proteins were purified

from yeast cells. Samples were collected from each step of purification process for dot

blot analysis (Figure 2.13). Dot blot analysis showed that most of the CCMV-S9

construct was associated with the cell pellet. However, with the yeast transformed with

CCMV 161-S9, a detectable amount of CCMV-S9 can be visualized in the final

purification step. Samples from each purification step were observed for the presence of

viral particles by transmission electron microscopy. We could not detect any chimeric

virus particles at any step of the purification but we could detect CCMV wt viral particle

from the final purification step. The dot blot and EM data suggest two explanations. The

first is that the purification protocol was not optimized resulting in less sensitive

detection of proteins. The second is that the low production of protein may be the result

of inappropriate conditions in cell growth or protein production resulting in less biomass

or desired protein at the start of the purification process. A high density fermentation

culture was introduced to circumvent this obstruction. 97

CCMV 63-S9 CCMV 102-S9 1 2 3 4 5 1 2 3 4 5

++ CCMV 114-S9 CCMV 129-S9 1 2 3 4 5 1 2 3 4 5

+ +

CCMV 161-S9 1 2 3 4 5

+

Figure 2. 13 Medium Scale CCMV-S9 Coat Protein Expression Experiment Recombinant yeasts expressing CCMV-S9 coat protein were cultured in non-methanol medium for 2 days and transferred to methanol-inducing medium for 5 days. Cells were harvested and kept in -80 ºC for viral purification. Samples of each purification step were collected for dot blot analysis. Frozen cells were lysed by adding acid-washed glass beads in equal volume virus buffer. The mixture was vortexed vigorously. Cell debris and glass beads were separated by centrifugation at low speed. Cell debris (sample1) was kept in - 80 ºC. The supernatant (sample2) was mixed with PEG 8000. The mixtures were stirred slowly overnight at 4 °C, centrifuged and the supernatant (sample 3) stored at -80 ºC. The pellets were resuspended in virus buffer. Mixtures were briefly centrifuged with low speed. The pellets (sample 4) were stored at -80 ºC and the supernatants (sample 5) containing virus were kept in -80°C for further analysis. Collected samples (number 1 to 5) were mixed with SDS loading buffer and boiled for 10 min and immediately put in an ice tub. Then 10 ul of each sample were loaded onto nitrocellulose membranes. The membranes were blocked with blocking buffer. Then the sample dots were determined by anti-CCMV antibody dilution 1/5000 as a primary antibody and anti-rabbit IgG-HRP dilution 1/1000 as a secondary antibody. Colorimetric detection was developed. The symbol + represents a CCMV-wt positive control purified from yeast cells.

98

Yeasts transformed with pcc3-161-S9 were picked to culture in the fermenter

because they showed the highest protein expression in the small scale time course

experiment and also the medium scale experiment. The fermentation method followed the

Invitrogen Pichia fermentation process guidelines and the details are described in the

Materials and Methods section.

Cells from the fermenter were harvested by centrifugation and divided into 100 g

aliquots and kept at -80 ºC. Samples from each viral purification step were loaded onto

nitrocelluose membranes for dot blot analysis with anti-CCMV antibody (Figure 2.14).

Dot blot analysis indicated that 161-S9 chimeric protein was mostly maintained in the

yeast cell debris (Figure 2.14, sample 1). The chimeric construct can be detected in later

purification steps. However, it was still associated with pellet as seen in the dot blot

analysis. The construct was found to associate with the DEAE particles in the filter cake,

(Figure 2.14: sample 3 and sample 8). Even at the final step of viral purification, the dot

blot analysis indicated that construct was associated with the pellet, (Figure 2.14, sample

11). This data shows that the association between the chimeric CCMV-S9 construct and

the pellet or yeast cell debris was strong. A smaller amount of the chimeric construct was

found in later purification step.

In order to dissociate chimeric CCMV-S9 construct from cell debris and pellet,

urea was applied at the first step of virus purification. The presence of 2 M urea during

the breaking of yeast cell was expected to increase recombinant viral protein solubility

and disrupt noncovalent association between the viral coat protein and yeast cell debris.

As shown in figure 2.15, most of the chimeric protein was still in the cell pellet. 99

1 2 3 4 5

6 7 8 9 10

11 + + -

Figure 2. 14 Large Scale CCMV-S9 Chimeric Construct Expression Experiment The recombinant yeast clone, CCMV 161-S9, was cultured in the fermenter. Cells were harvested by centrifugation and lysed by grinding vigorously with acid-washed glass beads in a bead-beater in the presence 50mMTris-HCl buffer pH 6. Beads and cell debris (sample 1) were separated by centrifugation. Supernatant (sample 2) was used for further steps. DEAE was added to the supernatant. The mixture was stirred slowly at 4°C for 2 h. The mixture was then filtered and the filtrate (sample 4) was kept at -80°C for the dot blot analysis. The filter cake was washed twice with the Tris buffer. The filtrate after the first wash and the second wash were mixed together (sample 5) and kept at -80°C for the dot blot analysis. The filter cake (sample 3) was resuspended in virus buffer supplement with 0.5 M NaCl. The suspension was filtered and the filtrate (sample 6) was collected. The filter cake was resuspended again with the virus buffer. The suspension was again filtered. The filtrate (sample 7) was collected and pooled with the filtrate from earlier step (sample 6). The filter cake (sample 8) was kept at -80°C for the dot blot analysis. The pooled filtrate was centrifuged at 25 krpm for 2 h. The supernatant (sample 9) was kept at -80°C for the dot blot analysis and the pellet was resuspended in the virus buffer. The mixture was centrifuged briefly at low speed to remove impurities. The supernatant (sample 10), which was supposed to contain virus and the pellet (sample 11) was kept at - 80°C for the dot blot analysis. Collected samples (number 1 to 11) were mixed with SDS loading buffer and boiled for 10 min and immediately put in an ice tub. Then 10 ul of each sample were loaded onto a nitrocellulose membrane. The membranes were blocked with blocking buffer. Then the sample dots were detected by anti-CCMV antibody and anti-Rabbit IgG-HRP. Colorimetric detection was developed. The symbol + represents a CCMV-wt positive control purified from yeast cells. The symbol – represents a non transformed yeast wt as negative control.

100

Anti-CCMV antibody S9 antibody Without Urea Without Urea 1 2 3 4 5 6 1 2 3 4 5 6

7 8 9 10 11 12 7 8 9 10 11 12

2M Urea 2M Urea 1 2 3 4 5 6 1 2 3 4 5 6

7 8 9 10 11 7 8 9 10 11

CCMV yeast S9 S9 CCMV yeast S9 S9 + - - - + +

Figure 2. 15 Virus Purification with or without the Presence of Urea Cells were lysed by grinding vigorously with acid-washed glass beads in a bead-beater in the presence 50mMTris-HCl buffer pH 6 without 2 M Urea (upper panel) or with 2 M urea (lower panel). Beads and cell debris (sample 1) were separated by centrifugation. Supernatant (sample 2) was used in further steps. DEAE was added to the supernatant. The mixture was stirred slowly at 4°C for 2 h. The mixture was then filtered and the filtrate is sample 4. The filter was washed with the Tris buffer twice. The filtrate after the first wash and the second wash were mixed together (sample 5). The filter cake (sample 3) was resuspended in virus buffer supplemented with 0.5 M NaCl. The suspension was filtered and the filtrate (sample 6) was collected. The filter cake was resuspended again with the virus buffer. The suspension was again filtered. The filtrate (sample 7) was collected and pooled with the filtrate from earlier step (sample 6). The filter cake was sample 8. The pooled filtrate was centrifuged at 25 krpm for 2 h. The supernatant was saved (sample 9) and the pellet was resuspended in the virus buffer. The mixture was centrifuged briefly at low speed to remove impurities. The supernatant (sample 10), which was supposed to contain virus and the pellet (sample 11) was kept at -80°C for the dot blot analysis. Collected samples (number 1 to 11) were mixed with SDS loading buffer and boiled for 10 min and immediately put in an ice tub. Then 10 ul of each sample were loaded onto a nitrocellulose. The membranes were blocked with blocking buffer. Then the sample dots were determined with anti-CCMV antibody and anti-rabbit IgG-HRP or 5 ug/ml S9 antibody and anti-mouse IgM-AP (right panel). Colorimetric detection was developed. The symbol + represents a CCMV-wt purified from yeast cells using the same protocol with the absence of urea. The CCMV-wt sample at final step was collected for the dot blot analysis with anti-CCMV antibody. The symbol – represents crude extract from a non transformed yeast wt as a negative control. S9 peptide concentration 0.2 ug/ml and 0.02 ug/ml were loaded onto the membrane as positive controls for the dot blot analysis with S9 antibody.

101

The presence of 2 M urea during the breaking of yeast cells was expected to increase recombinant viral protein solubility and disrupt noncovalent association between

the viral coat protein and yeast cell debris. As shown in figure 15, most of the chimeric

protein was still in the cell pellet. The presence of urea did not affect the interaction

between the chimeric construct and cell debris. CCMV-S9 viral particles could not be

obtained from the purification with or without urea application. Dot blot analysis with S9

mAb indicated that CCMV-S9 construct can be expressed but was found in the pellet with cellular debris. On the other hand, the CCMV wt construct could be obtained easily

with the same protocol even without urea. These data indicated that the chimeric CCMV-

S9 construct can be expressed but cannot be separated from yeast cell debris or pellet particles, probably because it failed to form intact virus-like particles, but rather aggregates and precipitates.

Samples from each purification step were observed for the presence of viral

particles by transmission electron microscopy. We could not detect any chimeric virus

particles at any purification steps, whereas wt CCMV viral particle was observed in final

step purification (figure 2.16)

These results demonstrated that chimeric coat proteins fail to assemble virus-like

particles, just as they failed to assemble infectious virions during plant expression.

Because we are able to produce both infectious virions, and virus-like particles with wild-

type constructs, we believe that these results clearly demonstrate the inability of the

CCMV carrying the S9 epitope to assemble a functional capsid. 102

To present the S9 peptide on a viral particle, chemical linkage of S9 peptides to

CCMV viral particles was pursued in our next experiment.

Figure 2. 16 CCMV-wt Viral Particle Yeast cells transformed with CCMV-wt construct were lysed by the method described in the Materiasl and Methods section. Next the suspension containing CCMV virus was purified with 38% (w/v) isopycnic CsCl2 density gradient by centrifuging at 38000 rpm for 20 h (Beckman, SW41). The virus band was collected and dialyzed against virus buffer. Viral concentration and purity were determined by measuring UV absorbance ratio Abs260/280 (Abs260/280 =1.5-1.7 considered good, 1Abs260 = 5.87) and observed under transmission electron microscope (TEM).

Chemical Linkage

To present peptide S9 on the surface of the CCMV virion, we first applied a

chemical method to conjugate S9 peptide to CCMV through disulfide formation. Thus we

introduced an additional C-terminal cysteine residue to the S9 peptide, creating a thiol

group for disulfide linkage. Mutant CCMV, A163C, which has Cys instead of original

Ala at position 163, was chosen for use. The coupling reaction was performed as

described in the Materials and Methods section. The coupling products were loaded onto

SDS non-reducing gels and visualized by silver staining (Figure 2.17). CCMV A163C mutant strain virus can conjugate with S9 peptide through disulfide bond formation. The 103

oxidizing agent CuSO4 can cause a minor improvement of the chemical reaction. The

problem experienced with this cross-linking method is that the conjugation product was

aggregated. The aggregation occurred when the product was left at 4°C in the virus

buffer. The aggregation reaction may be the result of continuous covalent disulfide bonds

forming between viruses. The insoluble conjugation product could not be utilized in the

vaccination, which is our purpose. Heterobifunctional linkers, sulfo-smcc and spdp were

then introduced in the experiment to crosslink CCMV to the S9 peptide and circumvent

the aggregation problem.

CCMV CCMV-S9 CCMV CCMV-S9 + CuSO4

Crosslinking product {{

Figure 2. 17 Cross-Linking between Cys of S9 and Cys of Mutant A163C CCMV mutant, A163C, 0.5 mg/ml, 2 ml was mixed with S9 peptide 0.5 mg/ml, 2 ml in the presence or absence of 40 ul 1 M CuSO4. The mixture was incubated at room temperature for 2 h. 10 ul of the mixture were mixed with 10 ul SDS PAGE gel buffer without mercaptoethanol and boiled for 10 min and then placed in the ice tub immediately. 10 ul of the sample were loaded into 12% nonreducing SDS PAGE gel. From left to right are mutant A163C, the cross-linked A163C-S9 without CuSO4, the cross-linkedA163C without S9 peptide, and the A163C-S9 with CuSO4.

104

The spdp linker is composed of a N-hydroxysuccinimide (NHS-ester) group and

2-pyridyldithil group (Figure 2.1). The NHS-ester will react with lysine associated

amines of the CCMV coat protein leading to a conjugation product via amide bond

formation. The reaction occurs at pH 7-8. Thus CCMV wt is not suitable for conjugation

because it is only stable at a low pH (pH4) (317). The salt stable CCMV mutant

(ssCCMV), which can tolerate a high pH and high ionic strength (319) was used in the

experiment instead of CCMV wt.

The ssCCMV virus was first conjugated to S9 by using the spdp linker. ssCCMV

was mixed with 3-fold molar excess to CCMV at pH 7.5 and the S9 peptide with 50-fold

molar excess of spdp. The pH of the reaction was adjusted to be pH6.5 and was

performed at room temperature for 1.5 h. The conjugation product was visualized by

silver staining and characterized by western blot analysis (Figure 2.18).

1 2 3 1 2

1 = ssCCMV 2 = ssCCMV-spdp-S9 2 ul 3 = ssCCMV-spdp-S9 5 ul

Figure 2. 18 Product of ssCCMV and S9 Peptide Cross-Linked via the spdp Linker. ssCCMV and S9 were conjugated via the linker spdp. The conjugation products were loaded onto 12% SDS-PAGE and visualized by silver staining or transferred to nitrocellulose membranes for western blot analysis with CCMV antibody

105

The silver staining and western blot result indicated that the ssCCMV was conjugated to S9 peptides via the spdp linker. The extra bands seen in the western blot

analysis reflects the conjugation products.

Since primary amine reacts with NHS-ester at pH 7-9, to get the best coupling

reaction, we performed an experiment with two different pH conditions, 7 and 7.5. We also varied the concentration of spdp linkers to be 1, and 2-fold molar excess to CCMV

virus. The conjugation products were loaded into SDS-PAGE gel and visualized by silver

staining (Figure 2.19). A greater degree of coupling was observed at pH 7.5and more

cross-linked products were obtained when the spdp linker was equimolar to the virus.

1 2 3 4

1 = ssCCMV 1000 ng 2 = ssCCMV-2x spdp –S9 pH 7.5 3 = ssCCMV-2x spdp –S9 pH 7 4 = ssCCMV-1x spdp –S9 pH 7.5

Figure 2. 19 The Conjugation Product of ssCCMv-S9 via spdp Linkers at Different pH Conditions and Different spdp Concentrations. The spdp linkers were used in 1 or 2-fold molar excess to CCMV. The reaction between the linkers and viruses was performed at pH 7 or 7.5 (pH of reactions was adjusted by adding NaH2PO4 1 M pH 9 for 1 hr at room temperature in the shaker. The conjugation products were loaded onto 12% SDS-PAGE and visualized by silver staining.

We next performed coupling experiments with different S9 peptide concentrations

to analyze if concentration of peptide affects the coupling reaction resulting in more, or less, CCMV-S9 conjugation product (Figure 2.20). Different concentrations of S9 106

peptide, 0.5,1, 10, 20, 30, 40 and 50-fold molar excess to CCMV were separately added

to the spdp-CCMV derivatized and the pH of reaction was controlled at 6.5. Adding

peptides at less than 10-fold molar excess to CCMV reduced the cross linking reaction.

Using a 10-fold molar excess of peptide to CCMV seemed optimal since a 50-fold molar

excess of peptide to CCMV did not provide a better yield of cross-linked material.

1 2 3 4 5 6 7 8 9 1 = ssCCMv 2 = ssCCMV-spdp 3 = ssCCMV- spdp - 0.5x S9 4 = ssCCMV- spdp - 1x S9 5 = ssCCMV- spdp - 10x S9 6 = ssCCMV- spdp - 20x S9 7 = ssCCMV- spdp - 30x S9 8 = ssCCMV- spdp - 40x S9 9 = ssCCMV- spdp - 50x S9

Figure 2. 20 ssCCMV-S9 Conjugation Products via spdp Linkers at Different Peptide Concentrations The ssCCMV was incubated with the fresh prepared spdp linker at 1:1 molar ratio. The reaction was allowed to continue at room temperature pH 7.5. After 1 h, the reaction was stop by incubating with 1.5 M hydroxylamine pH 8.5 for 1 h. Different concentrations of S9 peptide were used in 0.5, 1, 10, 20, 30, 40 and 50-fold molar excess to CCMV. The reactions were allowed to continue at room temperature for 1.5 h. 5 ul of reaction sample was mixed with 5 ul of SDS loading buffer without β-mercapto ethanol and boiled for 10 min and placed on ice. The samples were loaded onto 12% SDS-PAGE and visualized by silver staining.

The CCMV-S9 conjugates were next characterized by sucrose gradient

sedimentation. The ssCCMV-spdp-S9 conjugates were overlaid on the top of a 5-25%

sucrose density gradient and sedimented. The gradients were manually fractionated and

test samples loaded onto SDS-PAGE (Figure 2.21). 107

A B M ss 1 3 5 7 9 11 13 15 M ss-S9 1 3 5 7 9 11 13 15

C M ss 17 19 21 ss-S9 17 19 21

Figure 2. 21 Characterization of CCMV-S9 by Sucrose Gradient Sedimentation ssCCMV was incubated with freshly prepared spdp linker at a 1:1 molar ratio. The reaction was allowed to continue at room temperature pH 7.5. After 1 h, the reaction was stopped by incubating with 1.5 M hydroxylamine pH8.5 for 1 h. S9 peptides were used in 10-fold molar excess to CCMV. The reactions were allowed to continue at room temperature for 1.5 h. 500 ul of the conjugation product, and unconjugated ssCCMV virions were loaded 5-25% sucrose gradients and centrifuged at 37,000 rpm for 3 h. The gradients were fractionated 1 ml/fraction (21 fractions/tube). 10 ul of odd numbered fractions were mixed with SDS loading buffer without β-mercapto ethanol and boiled for 10 min, loaded onto 12% SDS-PAGE, and visualized by silver staining. The odd number fractions, 1-15, from ssCCMV were loaded onto gel A. The odd numbered fractions, 1- 15, from ssCCMV-spdp-S9 conjugate were loaded onto gel B. The remaining odd number fractions, 17-21, from ssCCMV were loaded into the first half of gel C and the remaining odd number fractions, 17-21, from ssCCMV-spdp-S9 were loaded into the second half of gel C (C figure). The ss indicates ssCCMV mutant and ss-S9 indicates ssCCMV-spdp-S9 conjugates. M indicates protein molecular weight marker.

108

The ssCCMV nonconjugate virion fell in the middle of the gradient (fraction 9-

11), whereas the CCMV-S9 conjugate fell in the bottom of the gradient (fraction 21). The

result indicated that the conjugation product exhibited higher sedimentation velocities

compared to the unconjugated ssCCMV particles. These higher sedimentation velocities of the conjugated particles probably indicate that the conjugated particles aggregated together. This may result from unexpected disulfide bond exchanging between

derivatized CCMV (CCMV-spdp) (Figure 2.1), thus causing aggregation that results in different sedimentation velocities.

To attain coupling of S9 to CCMV without aggregation, the mutant CCMV, subE,

was used. SubE is a CCMV mutant stain, in which eight basic amino acid residues (K and

R) of the N-terminus were replaced with the acidic residue glutamic acid (E) (320).

Conjugation to SubE virions leads to efficient viral disassembly (unpublished data). The

SubE-S9 construct serves as a control for the effect of fully assemble virus, since it

represents the peptide conjugated to a monomer form of virus coat protein. Thus we

conjugated SubE to S9 via spdp linkers at 1:1 molar ratio of spdp:CCMV. The reaction

was allowed to continue at room temperature pH 7.5, 1 h, and stopped by incubating with

1.5 M hydroxylamine pH8.5 for 1 h. The S9 peptides were used in 10-fold molar excess

to SubE. The reactions were allowed to continue at room temperature for 1.5 h. The

subE-S9 conjugates were visualized by SDS-PAGE analysis and analyzed by western

blot analysis (Figure 2.22). The conjugated SubE-S9 products were able to be detected

by western blot analysis against anti-CCMV antibody. 109

SubE SubE-S9 SubE SubE-S9

Figure 2. 22 Conjugation between SubE and S9 via spdp Linkers SubE CCMV and S9 were conjugated via the linker spdp. The conjugation products were loaded onto the 12% SDS-PAGE and visualized by silver staining or transferred to the nitrocellulose membrane for western blot analysis. Anti CCMV antibody and enzyme conjugated mouse anti rabbit IgG antibody were used in the western blot.

SubE-S9 conjugates were next characterized by sucrose gradient sedimentation.

The SubE-spdp-S9 conjugates were overlaid on the top of 5-25% sucrose gradients andsedimented. The gradients were manually fractionated and loaded onto SDS-PAGE

(Figure 2.23). The unconjugated SubE virions fell in the middle of the gradient (fraction

5-7). The SubE-S9 conjugate fell in the first fraction of the gradient (fraction 1). This result indicated that cross linking SubE with S9 peptide via the spdp linker caused the virus to fall apart. Thus the conjugation product from SubE is capsid protein subunit conjugated to the S9 peptide. The result was confirmed by observation of virus under

EM. We could not see any particles from SubE-S9 conjugate. 110

A B M SubE 1 3 5 7 9 11 13 M SubE-S9 1 3 5 7 9 11 13

Figure 2. 23 Conjugation Product SubE-spdp-S9 in Sucrose Gradient Velocity The SubE was incubated with the fresh prepared spdp linker at a1:1 molar ratio. The reaction was allowed to continue at room temperature pH 7.5. After 1 h, the reaction was stopped by incubating with 1.5 M hydroxylamine pH8.5 for 1 h. The S9 peptides were used in 10-fold molar excess to CCMV. The reactions were allowed to continue at room temperature for 1.5 h. 500 ul of the conjugation product and or unconjugated SubE were loaded on 5-25% sucrose gradients and then centrifuged at 37,000 rpm for 3 h. The gradients were fractionated 1 ml/fraction (21fractions/tube). 10 ul of odd numbered fraction were mixed with SDS loading buffer without β-mercapto ethanol and boiled for 10 min, loaded onto 12% SDS-PAGE and visualized by silver staining. The odd number fractions, 1-13, from SubE were loaded into gel A from the third lane (one lane apart from the protein molecular weight marker, to prevent contamination from reducing buffer β-mercapto ethanol contained in the markers). The odd numbered fractions, 1-13, from SubE-spdp-S9 conjugate were loaded into gel B from the third lane (one lane apart from the protein molecular weight marker.

Another linker sulfo-smcc was introduced to replace the spdp linker to create non-

aggregated ssCCMV-S9 particles. The sufo-smcc linker is composed of an NHS ester 111

group and maleimide group. As with the spdp linker, the NHS ester group reacts with a

primary amine in the CCMV coat protein, likely lysine. The amide bond is formed, releasing N-hydroxysuccinimide (Figure 2.2). The maleimide group reacts with thiol group of the peptide when the pH of the reaction mixture is maintained between 6.5 and

7.5. To preclude the aggregation between virus particles through the disulfide linkage that

might occur during the amide bonding, the ssCCMV was treated with tris (2-

carboxyethyl) phosphine (TCEP) to a final concentration of 50 mM. The TCEP reduces

the disulfide formation without affecting the other protein functional groups in the virus,

especially free amines. To assure that all disulfide bonds were reduced the ssCCMV was treated with 2-iodoacetamide (IA), an alkylating sulfhydryl reagent. The IA binds

covalently with cysteine on the virus. Thus there was no available cys from the virus to

react with the maleimide group. Cross-linking between ssCCMV and S9 was performed

with or without the IA treatment (Figure 2.24).

Figure 2.24 shows that treatment with 50 mM IA did improve the conjugation

result. With or without IA treatment, the virus can cross-link to the peptide via the sulfo- smcc linker, resulting in the conjugated product. With IA treatment, a dimer aggregation

band did not occur. The result showed that pretreating ssCCMV with 2-iodoacetamide

reduced disulfide formation between the viruses resulting in less dimer product

formation.

112

Without iodoacetamide With iodoacetamide M ss ss ss ss ss ss M ss ss ss ss ss ss nIA nIA nIA nIA nIA IA IA IA IA IA m m m m m m m m sc sc sc sc sc sc sc m m m m S9 S9

Dimer aggregation band Conjugation band Conjugation band

Figure 2. 24 The Conjugation Products Between ssCCMV and S9 Peptide via Sulfo-smcc Linker with and without Iodoacetamide Treatment (IA treatment) ssCCMV (ss) in 20 mM NaH2PO4 pH 7.5 was treated with TCEP to a final concentration of 50 mM at room temperature for 1 h. The 2-iodoacetamide (IA) 0.5 M was added to the virus to 50 mM final concentration, and incubated at room temperature for 30 min. The virus solution was purified to eliminate excess TCEP and the IA by using the microcon 100 (m). Virus was washed extensively with 20 mM NaH2PO4 pH 7.5 and eluted with the same buffer. Fresh prepared sulfo-smcc (sc) linker was used in 50-fold molar excess to the virus. After adding the linker, the pH of the reaction was remeasured and adjusted to be 7.5 with 1 M NaH2PO4 pH 9. The reaction was for 1 h in room temperature. Lysine was added to the reaction to have 1 mM final concentration and kept for 15 min at room temperature to stop the reaction. The excess linker and lysine were removed by using m100 and washed and eluted with the 20 mM NaH2PO4 pH 7.5. The S9 peptide was pretreated with TCEP to a final concentration of 50 mM for 1 h at room temperature prior to coupling with the derivatized virus. The pretreated S9 peptide was used in 50 molar excess to the ssCCMV. The pH was controlled at pH 7 with 1 M NaH2PO4 pH 9. The reaction of maleimides with thiols was for 1.5 h in room temperature. 5 ul of each step in the cross-linking process was mixed with 5 ul of SDS loading buffer containing β- mercapto ethanol and boiled for 10 min, loaded into 12% SDS-PAGE, and visualized by silver staining.

113

We next asked if the incubating temperature and time period of the reaction

between the ssCCMV virus and the linker affects the coupling reaction. To answer the

question, the ssCCMV was treated with TCEP and IA before adding the linker. The

sulfo-smcc linker was used in 50-fold molar excess to the ssCCMV. The virus was

incubated with the linker in 2 conditions, at room temperature or at 4°C. The length of

incubation was 2, 6 or 24 h for the room temperature condition and 6, 12, 18 or 24 for the

4°C temperature condition. After removing excess linkers out of the solution by using m100, pretreated S9 peptide with TCEP was added in 50-fold molar excess to the ssCCMV. The conjugation product was analyzed by SDS-PAGE gel (Figure 2.25). The silver staining data indicated that conjugation can occur in both condition. NHS reactions with primary amine can occur in 6 h and reached the maximum in 12 h. Incubating the virus with the linker for 24 h at room temperature pushed the NHS reaction, resulting in multimer conjugation between the viruses. The reaction at 4°C slows the amide forming reaction compared to the reaction at room temperature. The slower reaction rate allowed the virus to conjugate as amonomer, since multimer coupling products cannot be detected when the reaction was performed at 4°C. Performing the reaction at 4°C allowed more selective amide formation than did performing the reaction at room temperature, and thus prevented the formation of multimers. 114

Incubation at room temperature M ss ss ft ss ss ft ss ss ss M ft ss ss ss ft ss ss IA IA IA IA IA IA IA IA IA IA IA m m m m m m m m m m sc2 sc2 sc2 sc6 sc6 sc6 sc24 sc24 sc24 m m m m m m S9 S9 S9

Multimer coupling product

Incubation at 4°C M ss ss ft ss ss ft ss ss ss M ft ss ss ss ss ss ss ss ss IA IA IA IA IA IA IA IA IA IA IA IA IA IA m m m m m m m m m m m m m sc6 sc6 sc6 sc12 sc12 sc12 sc18 sc18 sc18 sc24 sc24 sc24 m m m m m m m m S9 S9 S9 S9

Coupling product

Figure 2. 25 Time Course Experiment and Temperature Effect on Coupling Rreaction The ssCCMV (ss) was treated with TCEP and IA before adding the linker. The sulfo- smcc linker was used in 50-fold molar excess to the ssCCMV. The virus was incubated with the linker in 2 conditions, at room temperature (upper panel) or at 4°C (lower panel). The length of incubation was 2, 6 or 24 h (sc2, sc6 and sc24) for the room temperature condition and 6, 12, 18 or 24 h (sc6, sc12, sc18 and sc24) for the 4°C temperature condition and the reaction was stopped by adding lysine to final concentration 1mM. After removing excess linkers out of the solution by using m100 (m), TCEP-treated S9 peptide was added at 50X molar excess to the ssCCMV. 5 ul of samples from each step of cross-linking process were mixed with SDS loading gel buffer containing β-mercapto ethanol and analyzed by SDS-PAGE gel. M is the protein molecular weight marker, ss stands for ssCCMV, ft is the flow through after passing reaction solution through microcon 100. The sc represents the linker sulfo-smcc and the number after sc represents the incubation time length. For example, sc6 means incubating with the sulfo-smcc linker for 6 h.

115

To ask if the linker concentration affected the cross-linking reactivity, different

concentrations of sulfo-smcc, (25, 50, 100 or 200-fold molar excess to the ssCCMV

virus), was added in the cross-linking reaction. The conjugation products were analyzed

by SDS-PAGE gel analysis (Figure 2. 26).

The primary lysines on the ssCCMV are shown to react with NHS-ester from linker in a dose dependent manner causing different amounts of ssCCMV-sufo smcc-S9 conjugation products. Using the linker in 100-fold molar excess to the virus pushed the reaction to the plateau. There was not much difference between using the linker in 100 and 200-fold molar excess to the virus. The derivatized ssCCMVs with a higher concentration of sulfo-smcc linker (200-fold molar excess) were found in the flow through in the step of passing the derivatized virus through the size-exclusion filter microcon 100. However, with lower concentrations of linker, 25, 50 and 100-fold molar excess, the derivatized ssCCMV-linker cannot be detected in the flow through. Based on this result, the next experiment was performed by using the linker sulfo-smcc in 100-fold molar excess to the ssCCMV virus.

To investigate if the S9 peptide concentration affected the cross-linking, different

concentrations of S9 peptide: 12.5, 25, 50, 100 or 200-fold molar excess to the ssCCMV

virus, were tested in the cross-linking reaction. The conjugation products were analyzed

by SDS-PAGE gel analysis (Figure 2.27).

116

M ss ss ft ss ss ft ss ss M ss ft ss ss ss ss ft ss ss IA IA IA IA IA IA IA IA IA IA IA m m m m m m m m m m 25sc 25sc 25sc 50sc 50sc 50sc 100sc 100sc 100sc m m m m m m S9 S9 S9

M ss ss ft ss ss IA IA IA m m m 200sc 200sc 200sc m m S9

Figure 2. 26 Effect of Different Concentration of Sufo-smcc: 25, 50, 100 and 200 Molar Excess to the ssCCMV Virus in Cross-Linking Reaction The ssCCMV (ss) was treated with TCEP and IA before adding the linker. The sulfo- smcc linker was used in 25, 50, 100 and 200-fold molar excess to the ssCCMV. The pH of reaction is 7. The virus was incubated with the linker at 4°C for 12 h and the reaction was stopped by adding lysine to final concentration 1 mM. After removing excess linkers out of the solution by using m100 (m), TCEP-treated S9 peptide was added at 50 molar excess to the ssCCMV. 5 ul of samples from each step of cross-linking process were mixed with SDS loading gel buffer containing β-mercapto ethanol and analyzed by SDS- PAGE gel. M is the protein molecular weight marker. The ss stands for ssCCMV. The ft is the flow through after passing reaction solution through microcon 100. The sc represents the linker sulfo-smcc and the number before sc represents the concentration of the linker in term of molar ratio excess to the ssCCMV. For example, 25sc means that the linker was used in 25 molar excess to the ssCCMV virus.

117

Figure 2.27 indicated that using peptide in 12.5-fold molar excess to the virus

provided the least conjugation product. Using peptide in 50-fold molar excess to the virus increased the reaction between the maleimide group of the linker and thiol group of S9 leading to high conjugation productivity. Adding more peptide than 50-fold molar excess to the virus did not improve the conjugation.

ss 200 100 50 25 12.5 M

Figure 2. 27 Effect of Peptide Concentration on Cross-Linking The ssCCMV (ss) was treated with TCEP and IA before adding the linker. The sulfo- smcc linker was used in 100-fold molar excess to the ssCCMV. The pH of the reaction was 7. The virus was incubated with the linker at 4°C for 12h and the reaction was stopped by adding lysine to final concentration 1 mM. After removing excess linkers out of the solution by using microcon100, TCEP-treated S9 peptide was used in 12.5, 25, 50, 100 or 200-fold molar excess to the ssCCMV. The reaction was performed at 4°C for 6 h. 5 ul of samples from each step of cross-linking process were mixed with SDS loading gel buffer containing β-mercapto ethanol and analyzed by SDS-PAGE gel. From left to right are ssCCMV (ss), coupling product with 200, 100, 50, 25, 12.5-fold molar excess to the ssCCMV and the last lane is protein molecular weight marker (M).

118

The ssCCMV-sulfo smcc-S9 particle was characterized by sucrose gradient

sedimentation velocity. The ssCCMV was treated with TCEP and IA before adding the

linker. The sulfo-smcc linker was used in 100-fold molar excess to the ssCCMV, pH 7.

The virus was incubated with the linker at 4°C for 12h and the reaction was stopped by

adding lysine to final concentration 1 mM. After removing excess linkers out of the

solution by using microcon100, TCEP-treated S9 peptide was used in 50-fold molar

excess to the ssCCMV. The reaction was performed at 4°C for 6 h. The ssCCMV-sulfo

smcc-S9 conjugation products were loaded on to 5-25% sucrose gradient and sedimented.

The gradients were manually fractionated and loaded into SDS-PAGE gel (Figure 2.28).

The sucrose gradient data showed that ssCCMV-sulf smcc-S9 particles remained intact and stayed as monomer virions, since they exhibited the same sedimentation velocities as underivatized ssCCMV particles. The ssCCMV-sulf smcc-S9 particles were observed under TEM, and particles were found.

Both conjugation products, SubE-spdp-S9 (Figure 2.22) and ssCCMV-sufo smcc-

S9 (Figure 2.27), were purified by passing through size exclusion microcon 100 filters.

The conjugation products were washed several times with sterilized 20 mM NaH2PO4 pH

6. The SubE-S9 coat protein-peptide conjugation product and ssCCMV-S9 particle were ready for the immunization experiment, which will be discussed in later chapters

119

A B ssCCMV ssCCMV-sufo smcc-S9 conjugate

M ss 1 3 5 7 9 11 13 15 M ss-S9 1 3 5 7 9 11 13 15

C ssCCMV ssCCMV-sufo smcc-S9 conjugate

M ss 17 19 21 ssS9 17 19 21

Figure 2. 28 Conjugation Product ssCCMV-sulfo-smcc-S9 in Sucrose Gradient 500 ul of the conjugation product and unconjugated ssCCMV virions were loaded on the top of 5-25% sucrose gradients and then centrifuged at 37,000 rpm for 3 h. The gradients were fractionated 1 ml/fraction (total 21 fractions/tube). 10 ul of odd numbered fractions were mixed with SDS loading buffer without β-mercapto ethanol and boiled for 10 min, loaded onto 12% SDS-PAGE and visualized by silver staining. The odd number fractions, 1-15, from ssCCMV (ss) were loaded into gel A. The odd numbered fractions, 1-15, from ssCCMV-sulfo smcc-S9 conjugate (ss-S9) were loaded into gel B. The remaining odd number fractions, 17-21, from ssCCMV were loaded into the first half of gel C and the remaining odd number fractions, 17-21, from ssCCMV-sulfo smcc-S9 were loaded into the second half of gel C. The ss indicates ssCCMV mutant, and the ss-S9 indicates ssCCMV-sufo smcc-S9 conjugates. M indicates protein molecular weight marker.

120

Discussion

The use of plant viruses for recombinant protein expression has been studied extensively, including for vaccine development. In our study, we used CCMV as the model for the design of a new vector for protein expression. We describe in detail the techniques that we used to prepare chimeric CCMV and also CCMV virus-like particles

(VLP). The X-ray crystal structure of CCMV reveals that viral coat protein has a core structure consisting of an eight stranded, antiparallel β-barrel motif, with exposed loops

(326-328). The N-terminus and the C-terminus of the coat protein subunit extends in opposite directions from the β-barrel motif and plays an important role in virion assembly and stability (329-331). The CCMV coat protein can be expressed in E. coli and can be assembled in vitro into either empty virions or in combination with viral RNA (329).

Based on three-dimension EM and X-ray crystal analyses, the Young lab identified permissive sites on exposed loops of coat protein subunit for foreign DNA insertion.

Loop-specific BamH1 cloning sites, where peptides could be expressed on the virus coat protein subunit, were created.

We initially inserted DNA to encode the S9-peptide (FDTGAFDPDWPA) within the loop sequence. The S9 peptide is a mimotope of type III Streptococcus group B capsular polysaccharide, which was shown to be both antigenic and immunogenic (1).

Recombinant RNA3 encoding the S9 peptide sequence was in vitro transcribed for plant inoculation. Chimeric recombinant virus particles were expected after inoculating on cowpea and tobacco plants. The ELISA results revealed that no intact chimeric particles were obtained, whereas the wild-type CCMV particles did produce virus following plant 121

inoculation. In vitro translation demonstrated that inserts were in frame and could be

translated. Results of inserting other peptides: peptide 44, HIV mimetic peptide

(unpublished data) and peptide 11, cell-targeting peptide (332, 333), into CCMV exposed loops were consistent with our results (unpublished data). However, insertion of peptide

11 in loop position 129 did yield recombinant virus, but with loss of infectivity. The productive yield of CCMV-peptide 11 was low, and the plant infection was localized not

systemic. It has been shown that movement protein of CCMV depends on the coat

protein to mediate the cell-to-cell movement (334, 335). All data indicates that insertion

of a peptide sequence into CCMV exposed coat protein loops leads to disruption of the

coat protein structure and its ability to assemble. Insertion of peptides into viral coat

proteins to create antigenic chimeras has been demonstrated in icosahedral virus such as

poliovirus (296), human rhinovirus (297) and also cowpea mosaic virus (CPMV) (299).

With these viruses the difficulty arises with a limitation of the size of peptide that can be

inserted and the loss of insert after serial passaging; whereas for CCMV, viral assembly

appears to be blocked

The second attempt to present the S9 peptide incorporated into CCMV coat

protein was by creating self-assembling recombinant S9-viral coat protein in

heterologous systems resulting in procapsids called virus-like particles (VLP). We chose

P. pastoris as our heterologous expression system. Yeast requires methanol for protein production and has been used to produce other VLPs such as bacteriophage Qbeta (324)

Norovirus (336), Hepatitis B virus (337) and Hepatitis C virus (338). For this, we

subcloned the DNA cassette encoding S9-CCMV recombinant coat protein into a Pichia 122 pastoris, expecting intracellular recombinant VLP production. DNA sequencing analysis and in vitro translation results indicated that the cassettes were in frame and able to be translated. The dot blot results indicated that recombinant VLP were obtained after methanol induction. However, the recombinant protein yield was low, even when we used the fermentor batch to produce the recombinant yeasts. In the fermentor batches, we produced recombinant proteins but could not purify the recombinant chimeric VLP from the yeast cell debris, whereas wt CCMV VLPs can be obtained easily from the same purification. Urea was applied to disrupt noncovalent associations between the viral coat protein and yeast cell debris, but we were still unable to gain CCMV-S9 VLP. This data indicated that the insertion of S9 into the CCMV virus coat protein also inhibits the formation of intact chimeric VLP, just as it does infectious virions. However, other chimeric plant viruses have been successfully made as VLPs (299, 307, 308). It has been shown that success in chimeric VLP production is dependent upon the insertion sites into the coat protein. In the development of epitope expression in CPMV, the insertion of foot-and-mouth disease virus (FMDV) epitope into the S protein results in genetic instability resulting from homologous recombination (299, 310). Repositioning of the epitope insertion site results in the successful assembly of viral particles (310, 339, 340).

However, some chimeric CPMV constructs are incapable of systemic infection, probably because the coat protein modifications interfers with viral movement (299). In our study, it is likely that coat protein modification in CCMV interfered with viral assembly.

Our third attempt to produce S9-CCMV particle was by using chemical methods to conjugate the CCMV to the S9 peptide. We first conjugated dS9 and CCMV through 123

disulfide bond formation between thiol groups on the virus coat protein and the C-

terminal of peptide. The CCMV mutant, CCMV A163C with cysteine substituted at

position 163, was used in the experiment because position 163 is on the exposed loops

and considered to be suitable as B-cell epitope. The result indicated that conjugation

products could be obtained, but they were aggregated resulting in precipitated particles at the bottom of the tube when the product was stored at 4°C. The aggregation reaction may have resulted from continuous covalent disulfide bonding between viruses.

We next used the chemical linkers, spdp and sulfo-smcc to conjugate the virus

and peptide. The NHS ester group on spdp or sulfo-smcc reacts with primary amines on

the viral coat protein. The 2-pyridyldithio group of the spdp linker reacts with sulfhydryl

at the amino terminal of S9 peptide. A maleimide group of sufo-smcc reacts with the

thiols on the S9 peptide. The coupling reaction requires pH 6-7.5. Wild-type CCMV is

stable at pH 5 and low ionic strength. Therefore, we used the salt stable mutant of CCMV

(ssCCMV), which resists disassembly in 1.0M NaCl, pH 7.5 (319, 341). The result

indicated that the ssCCMV can conjugate to the S9 via spdp linker but the conjugation

product again aggregated. According to the sucrose gradient analysis, the conjugation

product was found in the bottom of the gradient whereas the unconjugated ssCCMV was

found in the middle of gradient. The data indicated aggregation of the conjugation

product, which may result from unexpected disulfide bond exchanging between

derivatized CCMV (CCMV-spdp). Then we changed the linker to sulfo-smcc to create

nonaggregated S9-CCMV particles. We could obtain the conjugation product with the

help of 2-iodoacetamide (IA), an alkylating sulfhydryl reagent. The IA binds covalently 124

with cysteine on the virus, resulting in no available cys from virus to react with the

maleimide group. The sucrose gradient analysis revealed that the ssCCMV-S9 particles

remained intact and stayed as monomers, since they exhibited the same sedimentation

velocities as underivatized ssCCMV particles. To create the coat protein subunit-S9

conjugate, the SubE mutant in which eight basic amino acid residues (K and R) of the N-

terminus were replaced with acidic residue glutamic acid was used (320). The sucrose

gradient analysis indicated that conjugation of SubE virions to S9 results in efficient viral disassembly, since the coupling product was found at the top of the gradient. The western blot analyses showed that the yield of conjugation products was low and less virus seemed to be conjugated. One possible explanation is that there might be few accessible lysines on the coat protein, resulting in less conjugation product. A second explanation is that the coupling reaction might not be suitable, disrupting natural coat protein folding of viral particle. Anther virus, Cowpea Mosaic Virus (CPMV), was used to conjugate to fluorescein N-hydroxysuccinimide (NHS) and biotin (311, 312). At neutral pH, wild-type

CPMV was found to carry up to 60 dye molecules per virion, in a dose-dependent manner. At pH 8.3, 1.5 dye molecules per asymmetric unit (90 per virion). Under maximal conditions, up to 4 lysine residues per asymmetric unit could be labeled (311,

312). The conjugation between CPMV and S9 was kindly performed by the Johnson Lab,

(Scripps Institute, La Jolla CA). The conjugation product yield was higher than CCMV-

S9 conjugation (discussed more in next chapter). In conclusion, the CCMV-S9 particles

and CCMV coat protein subunit-S9 conjugates (made with the SubE mutant, which fails 125 to assemble) were obtained from the experiment and were ready to use in immunization experiments, which will be discussed in the next chapter.

126

MURINE IMMUNE RESPONSE TO CCMV-CONJUGATED PEPTIDE MIMICS OF GBS CAPSULAR POLYSACCHARIDE

Introduction

Group B streptococci (GBS) (Streptococcus agalactiae) is a commensal of the genitourinary tract in 15-40% of healthy adult women (342). It is reported as a major cause of neonatal sepsis and meningitis in developed countries (99), with type III GBS predominate (100, 343). Neonatal GBS infection is acquired during delivery by vertical transmission. Neonates are exposed to GBS either via GBS ascending through ruptured amniotic membranes, aspiration of contaminated amniotic fluid, or through exposure to colonized tissues while passing through the birth canal (344). Immunity to GBS infection in humans (168), mice (345) and rats (346) is due to type-specific capsular polysaccharide antibody. GBS capsular polysaccharide (CPS) is both a virulence factor and the target of a protective humoral immune response (143, 168, 191, 345-348). GBS vaccine development has been based on immunization with CP or CP conjugated to a protein carrier, leading to protective levels of maternal IgG type specific anti-GBS antibodies, which could be transplacentally passed to the newborn. Vaccines containing

CPS of group B streptococcus induced protective IgG antibodies in only 60% of vacinees (191, 346). Conjugation of CPS to protein carriers, including tetanus toxoid

(TT) (192, 346) and choleratoxin B subunit (199) induce much better protective IgG antibodies. GBS CPS is a poor immunogen and acts as a T-independent antigen resulting in IgM production instead of IgG, and a lack of immunological memory. 127

An alternate approach is to design antigenic mimics, and relevant to our approach is using peptide-display phage libraries to search for peptides that mimic carbohydrate epitopes. The technique was developed by G. P. Smith (238) and is a powerful tool to identify mimotopes for vaccine development. Several carbohydrate mimetic epitopes have been reported by using this technique (1, 273, 276, 349, 350). In our lab, previously, we identified peptide mimics of a protective epitope of GBS by the means of a phage display library (1). The S9 mAb demonstrating in vivo protection against GBS type III

(172, 174) was used to select displayed peptides. The S9 peptide, FDTGAFDPDWPA was isolated. The S9 peptide is shown to be both antigenic and immunogenic.

Immunization of mice with peptide-carriers results in the production of GBS antibodies

(1).

Several systems have been developed to improve vaccines for small peptides

epitopes. Conjugation to carrier protein such as TT or keyhole limpet hemocyanin (KLH) has been extensively used in experimental studies to test candidates for vaccine development. An alternative scheme to enhance the immunogenicity of the peptides is to fuse the peptide sequences to the genes coding for virus capsid proteins, which have

ability in self-assembly to be virus-like particles (VLP), or chemically conjugate peptides

to virion or to VLP. Several plant viruses have been successfully used as expression

vectors to present epitopes for vaccine development (308, 310, 340, 351, 352). Our

studies have focused on the use of the plant virus Cowpea Chlorotic Mottle Virus

(CCMV) to present the S9 peptide on viral coat proteins. The structure of CCMV,

determined by X-ray crystallography, demonstrates that the virion is comprised of 180 128

copies of the coat protein subunit (317). The attempt to genetically modify CCMV coat protein to express the S9 peptide is discussed in the previous chapter. We failed to genetically express S9 fused to the viral coat protein, due to failure of virus assembly. We

then chemically conjugated the S9 peptide to the CCMV virus and to the VLP of CCMV.

In this study, we investigated antibody responses of mice immunized with the virus-S9

conjugate compared with KLH-S9 conjugate and evaluated their immunogenicity.

Materials and Methods

Preparation of Antigens

The S9 peptide, which mimics the type III GBS CPS, was chemically conjugated

to CCMV ss and SubE mutants, as described in the previous chapter. CCMV wild type

was obtained from the Young lab (Montana State University). Cowpea Mosaic Virus

(CPMV) wild type and conjugated to S9 peptide were kindly provided by Dr. John E.

Johnson (Scripps Institute, La Jolla CA). Synthetic peptides (Sigma Laboratories (USA))

containing Cys at the COOH peptide terminals were conjugated to maleimide derivatized

KLH (Pierce Chemical Co., Rockford, IL). Maleimides react with sulfhydryls at pH 6.5-

7.5 to form stable thioether bonds. Peptides (300 ug) were mixed with KLH (300 ug) (1:1

ratio) in phosphate buffer pH 7 in a reaction volume of 300 ul. The reaction was allowed

to continue at room temperature for 3 h. The unconjugated peptides were removed by

using microcon 100 (Millipore, Billerica, MA) filtration. The conjugated KLH-peptides

were quantified by using BCA protein assay reagent kit. 129

We used ELISA to determine the amount of S9 peptide and virus in each

preparation. Protein concentrations were measured using BCA Protein Assay Reagent Kit

from Pierce (Rockford, IL). The first titration was performed to compare CCMV

concentration in each preparation. 96 well microtiter plates were coated with the 0.2, 0.1,

0.05, 0.025 and 0.0125 ug/ml (based upon BCA concentration) of CCMV, CCMV-S9,

SubE-S9 or KLH-S9. Then the coated plates were blocked with 1% BSA in phosphate-

buffered saline (PBS). Primary antibody, rabbit anti-CCMV (1:3000 dilution) was

incubated in microtiter wells at 4°C for 18 h. Plates were washed with ELISA washing

buffer, 10 mM phosphate buffer pH 7.4, 150 mM NaCl, 0.05% Tween 20 and incubated

with alkaline phosphatase-conjugated anti-rabbit Ig (Zymed Laboratories, South San

Francisco CA) at a 1/10,000 dilution for 6 h, followed by washing and the addition of

colorimetric substrate p-nitrophenyl phosphate (Sigma Chemical) 1 mg/mL in 10%

diethanolamine buffer, pH 9.6, with 0.5mM MgCl2. Absorbance at 405 nm was determined using a microplate reader (EL-320, Bio-Tek Instruments, Winooski, VT). A

second assay was performed to measure the amount of S9 peptide in each antigen

preparation. Plates were coated with the 0.4, 0.2, 0.1, 0.05, 0.025 and 0.0125 ug/ml of

ccmv-S9, subE-S9, cpmv-S9, cpmv or KLH-S9. S9 mAb at 5ug/ml in PBS buffer was

used as the primary antibody and alkaline phosphatase-conjugated anti-mouse Igs (H+L

chain specific, Zymed) at 1/10,000 in PBS was used as the secondary antibody.

Immunization

Two sets of immunizations were performed. Female BALB/c mice 8-10 weeks

were used. We compared the antibody responses to immunization with different antigen 130 preparations, with and without adjuvant, and the ability to prime for secondary immune responses to GBS.

In the first experiment, mice were separated into 2 main groups (20 mice in each group): adjuvant and no adjuvant. In the adjuvant group, mice were immunized with antigens, which were emulsified in a 1:1 ratio with complete Freund’s adjuvant (CFA,

Difco, Detroit, MI) for the primary immunization, and with incomplete Freund’s adjuvant

(IFA) for booster injections. The non-adjuvant groups were immunized with antigens which were diluted with PBS for primary and booster immunizations. Each group was separated into 5 subgroups (4 mice in each subgroup) each receiving a different antigen:

CCMV, CCMV-S9, SubE-S9, CPMV-S9 and KLH-S9. Mice were immunized with following amount of antigens: CCMV 2 ug, CCMV-S9 20 ug, SubE-S9 20 ug, CPMV-S9

10 ug, KLH-S9 10 ug. The concentration of antigen was based upon ELISA results characterizing the amount of S9 peptide coupled to each antigen, as determined by

ELISA. Mice were bled to collect sera before immunization (Bleed 1). Primary immunization was given subcutaneously (sc) and boosters intraperitonealy (ip). Booster immunization followed the primary immunization by four weeks and one week after boosting, all mice were bled to collect sera (Bleed 2) to measure antibody. Sera were diluted to appropriate concentrations in PBS to measure different antibody responses:

1/200 to test the antibody response to GBS; 1/1000 to test the antibody response to other mimetic peptides: S9-2, S7 and S11; and 1/5000 to test the antibody response to CCMV and S9 peptide. ELISA was used to measure total Ig and specific subclasses of antibodies. 131

Six subgroups of mice from this experiment were chosen for boosting with GBS:

CCMV, CCMV-S9 and KLH-S9 from both the adjuvant and non-adjuvant groups. Five

weeks after boosting with peptide, these mice were injected with heat-killed GBS.

Overnight cultures of GBS type III strain 125 in Todd Hewitt broth was spun down and

washed three times with sterilized PBS and then resuspended into sterilized PBS.

Absorbance at 600 nm of suspended GBS was adjusted to OD 0.9. The cells were killed

by heating at 60 °C for 30 min. Mice were injected subcutaneously with 0.1 ml of heat-

inactivated culture. The heat-inactivated culture was checked for viability by plating on

Todd Hewitt agar plates. Sera were collected at 3 and 7 days after challenge (Bleed 3 and

Bleed 4). Antibody to GBS, S9 peptide and CCMV was measured.

In the second experiment, 24 mice were separated into 4 groups (6 mice in each group). Each group received a different antigen: CCMV, CCMV-S9, CPMV-S9 and

KLH-S9. Groups of mice were immunized subcutaneously, without adjuvant, with

following amount of antigens; CCMV 2 ug, CCMV-S9 20 ug, CPMV-S9 10 ug, and

KLH-S9 10 ug. Mice were bled to collect sera before immunization (Bleed 1). Four

weeks later, mice were reimmunized ip with the same amount of antigen. One week after

boosting, all mice were bled to collect sera (Bleed 2) and antibody to GBS, S9 and

CCMV was measured. Mice with high anti-S9 and anti-GBS antibody titers were chosen

to be investigated for cytokine production.

Cell Purification and Culture

Mice from the second experiment with high anti-S9 and anti-GBS antibody titers

were bled (Bleed 3) and immunized ip with the same antigen and dose as in the prime 132

and boost. Immunized and age-matched naive mice were bled and sacrificed 4 days after

the second booster immunization. Spleens and mesenteric lymph nodes were removed

aseptically. The spleens and mesenteric nodes were pooled, single-cell suspensions prepared by aspiration into syringes, washed in culture medium (RPMI 1640), and incubated on ice for 30 min in RPMI 1640. After one wash in RPMI 1640 medium, the cells were resuspended in 10 ml of the same medium and layered onto 10 ml of

Lymphocyte M separation medium (Cedarlane Laboratories limited, Ontario, Canada).

The gradient was spun at 1750 rpm for 30 min at room temperature. Cells at the interface

of the gradient were collected, washed twice, and resuspended in cold culture media

(RPMI 1640 containing 2% fetal calf serum (FCS)), 100 U/ml penicillin, 100 g/ml

streptomycin, 50 mM 2-mercaptoethanol and 2mM l-glutamine (Gibco BRL,

Gaithersburg, MD). Cells were counted and used as splenocytes in some experiments, in

others the cells were passed through a column (Mouse T cell Recovery Column Kit from

Cedarlane Laboratories Limited, Ontario, Canada) to prepare CD4 T cells. Cells were

plated 100 ul per well (2×105 cell per well) into sterile 96-well microplates (Costar,

Corning Inc., NY) and stimulated either with medium, CCMV, CCMV-S9, S9, KLH-S9,

KLH, all at 20 ug/ml, or PHA-M (Sigma Chemical) at 10 ug/ml. Cultures were

incubated for 7 days at 37 ◦C in 5% CO2. The supernatants were collected at day 3, 5 and

7 and frozen at −20 ◦C until analyzed for cytokine production.

133

Collection and Preparation of Blood

Blood samples were taken from the lateral saphenous vein. Serum was prepared

by centrifugation and stored at −20 ◦C. Sera were diluted to appropriate concentration with PBS/1% BSA/0.01% Na Azide before ELISA.

Enzyme-Linked Immunosorbent Assays (ELISA) of Peptide-Specific Antibodies, GBS- Specific Antibodies, and CCMV-Specific Antibody

CCMV or the peptide Ags: S9, S9-2, S7 and S11 were coated onto microtiter

wells (Immulon, Costar) at 5 µg/ml for at least 18 h. Amino acid sequences of each

peptide are indicated in Table 3.1.

Table 3. 1 Amino Acid Sequences of Peptides Peptide Amino acid Description References sequence S9 FDTGAFDPDWPA Mimic epitope of type III, GBS (1) group B S9-2 WENWMMGNAGC Mimic epitope of type III, GBS UO group B S7 WGNWQDRLKC Mimic epitope of GBS group B, UO C, G S11 EEYPHGPEPC Mimic epitope of type I, GBS UO group B Note; UO : unpublished observation by Dr. Seth Pincus

GBS were coated onto microtiter wells using poly-L-lysine and glutaraldehyde, as

described elsewhere (353). Plates were blocked with 1% BSA and used within 1 week.

Plates were washed with ELISA washing buffer (10 mM phosphate buffer pH 7.4, 150 mM NaCl, 0.05% Tween 20) prior to use. Test sera at appropriate dilutions were added to antigen-coated wells in triplicate and incubated at 4°C for 18 h. The plates were washed 134

and incubated with alkaline phosphatase-conjugated anti mouse-Ig (heavy and light chain

specific, Zymed laboratories, South San Francisco, CA) for 4 h at room temperature.

Following incubation with the secondary antibody, plates were washed and the colorimetric substrate p-nitrophenyl phosphate (Sigma Chemical) was added. Absorbance at 405nm was determined using a microplate reader (EL-320, Bio-Tek Instruments,

Winooski, VT).

ELISA of S9-Specific Antibodies Subclasses in Serum

S9-peptide was coated onto microtiter wells (Immulon, Costar) at 5 µg/ml. Mouse

sera at a dilution of 1/2000 were added in triplicate to the plates and incubated at 4°C for

18 h. The plates were washed with ELISA washing buffer and incubated with rabbit

antisera specific for -mouse IgG1, IgG2a, IgG2b or IgG3 (Zymed laboratories, South San

Francisco, CA) diluted 1/8000 at 4°C for 18 h. Following a washing step, goat anti-rabbit

IgG antibody conjugated with alkaline phosphatase (Zymed Laboratories) was incubated in the wells for 6 h, followed by washing and addition of colorimetric substrate and A405 was determined.

ELISA of S9-Specific and GBS-Specific IgM Antibodies in Serum

S9-peptide and GBS plates were prepared as described in section 4. Sera diluted

1/1000 for S9 plates and 1/200 for GBS plates were added in triplicate to the plates and

incubated at 4°C for 18 h. The plates were washed and incubated with rabbit anti-mouse

IgM μ chain specific antibody (Zymed laboratories, South San Francisco, CA diluted

1/1000). Goat anti-rabbit IgG antibody conjugated with alkaline phosphatase (Zymed 135

Laboratories), diluted 1/10,000 was incubated in wells for 6 h, followed by washing and

addition of colorimetric substrate and A405 was determined.

Assessment of Cytokines in Supernatants from Cultured Splenocytes and CD4 T Cells

A sandwich ELISA was used to determine the amount of gamma interferon (IFN-

γ), interleukin 4 (IL -4) and IL-10 produced by the splenocytes and CD4 T cells during

antigen activation. Antibody pairs, recombinant standards and ELISA reagent for the

assay were purchased from eBioscience, Inc. Corning Costar 9018 96-well ELISA plates

were coated in100 µl/well of capture antibody in Coating Buffer. Plates were sealed and

incubated overnight at 4°C. Wells were blocked with 200 µl/well of assay diluent

(provided with reagents) and incubated at room temperature for 1 hour. The standards

(IFN-γ, IL-4 or IL-10) were diluted by 2-fold serial dilutions to create appropriate

standard curves and were added in duplicate to the plates. The supernatants from cell media were added in duplicate to the plates in a series of two-fold dilutions. We then added 100 ul/ well of the diluted biotin-conjugated detection antibody. Plates were sealed and incubated at room temperature for 1 hour. After 5 washes, we added 100 ul/well avidin-peroxidase and the plates were incubated at room temperature for 30 min. After 7 washes, 100ul/ well substrate, 2,2-azinobis-3-ethylbenzthiazoline-sulfonic acid with 0.3%

H2O2 were added and incubated at room temperature for 15 min. The reaction was

stopped with a stop solution (0.7M SDS in DMF and purified water, 1:1). The absorbance

was measured at 450 nm. In each sample, the final concentration of IFN-γ, IL-4 or IL-10 136

was determined from the standard curves by calculating the mean concentration from two

dilutions.

Results

Antigen Titration

To immunize with the same amount of the relevant antigen (S9 peptide) in

preparations having different carrier proteins, methods of coupling, and coupling

efficiencies; CCMV-S9, SubE-S9, CCMV, KLH-S9and cpmv-S9 were titrated by

ELISA. Antigens were diluted in PBS. The graphs in figure 3.1 and 3.2 display the results.

3.00

2.50 CCMV-S9 2.00 SubE-S9 1.50 KLH-S9 SubE 1.00 CCMV 0.50 Optical Density 405 nm 0.00 0 0.05 0.1 0.15 0.2 0.25 Protein Concentration (ug/ml)

Figure 3. 1 Comparison of CCMV Virus Antigen in Different Antigen Preparations 96 well microtiter plates were coated with the 0.2, 0.1, 0.05, 0.025 and 0.0125 ug/ml of CCMV, CCMV-S9, SubE, SubE-S9 or KLH-S9. The coated plates were then blocked with 1% BSA in PBS. Anti-CCMV serum (1:3000 dilution) was incubated in the microtiter wells at 4°C for 18 h. Plates were washed and incubated with alkaline phosphatase-conjugated anti-rabbit Ig for 4 h, followed by washing and addition of colorimetric substrate p-nitrophenyl phosphate. Absorbance at 405 nm was determined. Results are means and standard errors.

137

2.50

2.00

1.50 CCMV-S9 SubE-S9 1.00 KLH-S9

0.50 Optical Density nm405 0.00 0 0.1 0.2 0.3 0.4 0.5 Protein Concentration (ug/ml)

2.50

2.00

1.50 CPMV-S9 CPMV 1.00 KLH-S9

0.50 Optical Density 405 nm 0.00 0.00 0.10 0.20 0.30 0.40 0.50 Protein Concentration (ug/ml)

Figure 3. 2 Comparison of S9 Peptide in CCMV-S9, SubE-S9, CPMV-S9, CPMV and KLH-S9 Antigens Plates were coated with 0.4, 0.2, 0.1, 0.05, 0.025 and 0.0125 ug/ml of CCMV-S9, SubE- S9, CPMV-S9, CPMV and KLH-S9. Plates were blocked with 1% BSA and washed with ELISA washing buffer. S9 mAb 5ug/ml in PBS buffer was added and incubated at 4°C for 18 h. After washing with ELISA washing buffer, alkaline phosphatase-conjugated anti-mouse Igs 1/10,000 in PBS were added into the wells and incubated at room temperature for 4 hours, followed by washing and addition of colorimetric substrate p- nitrophenyl phosphate. Absorbance at 405 nm was determined.

138

At the same nominal protein concentrations, the amount of CCMV antigen in

SubE-S9 and CCMV-S9 was 10 fold less than the amount of CCMV antigen in

unconjugated CCMV (Figure 3.1). This may have resulted from a degree of denaturation

of the S9-conjugated viruses, and loss of antigenicity of the virus coat protein.

The amount of S9 peptide in the KLH-S9 is approximately equal to the amount of

S9 antigen in the CPMV-S9 (Figure 3.2). The amount of S9 in the KLH-S9 is two-fold higher than the amount of S9 protein antigen in the CCMV-S9 and SubE-S9 (Figure 3.2).

According to these titration results, we decided to choose concentrations of CCMV 2 ug,

CCMV-S9 20 ug, SubE-S9 20 ug, CPMV-S9 10 ug, KLH-S9 10 ug in immunization experiments.

Immune Response after Immunization

In the first experiment, we immunized 2 groups of mice, with and without

Freund’s adjuvant. Mice in the adjuvant group were immunized with antigens which

were emulsified in complete Freund’s adjuvant and then 4 weeks later were reimmunized

with the same amount of each antigen emulsified in incomplete Freund’s. Mice in the

non-adjuvant group were immunized with antigens diluted in PBS. Each group was

separated into 5 subgroups (4 mice in each subgroup), each receiving a different antigen:

CCMV, CCMV-S9, SubE-S9, CPMV-S9 and KLH-S9. Groups of mice were immunized

with the amount of antigen determined above. Mice received a primary, and 4 weeks

later, a booster immunization. One week after the booster, sera were collected and tested

for antibody response to S9 peptide, CCMV and GBS (Figure 3.3).

139

S9

3

2.5

2 Bleed1 1.5 Bleed2 1

Optical Density 405nm Density Optical 0.5

0 M1 M2 M3 M4 M5 M6 M7 M8 M9 M10 M11 M12 M13 M14 M15 M16 M17 M18 M19 M20 M21 M22 M23 M24 M25 M26 M27 M28 M29 M30 M31 M32 M33 M34 M35 M36 M37 M38 M39 M40 CCMV CCMV-S9 SubE-S9 CPMV-S9 KLH-S9 CCMV CCMV-S9 SubE-S9 CPMV-S9 KLH-S9

CFA PBS Mouse

CCMV

3.00 2.50 2.00 Bleed1 1.50 Bleed2 1.00 0.50

Optical Density 405 nm Density Optical 0.00 M1 M2 M3 M4 M5 M6 M7 M8 M9 M10 M11 M12 M13 M14 M15 M16 M17 M18 M19 M20 M21 M22 M23 M24 M25 M26 M27 M28 M29 M30 M31 M32 M33 M34 M35 M36 M37 M38 M39 M40 CCMV CCMV-S9 SubE-S9 CPMV-S9 KLH-S9 CCMV CCMV-S9 SubE-S9 CPMV-S9 KLH-S9

CFA PBS Mouse

GBS

1.8 1.6 1.4 1.2 1 Bleed1 0.8 Bleed2 0.6 0.4 Optical Density 405 nm Density Optical 0.2 0 M1 M2 M3 M4 M5 M6 M7 M8 M9 M10 M11 M12 M13 M14 M15 M16 M17 M18 M19 M20 M21 M22 M23 M24 M25 M26 M27 M28 M29 M30 M31 M32 M33 M34 M35 M36 M37 M38 M39 M40 CCMV CCMV-S9 SubE-S9 CPMV-S9 KLH-S9 CCMV CCMV-S9 SubE-S9 CPMV-S9 KLH-S9

CFA PBS Mouse

Figure 3. 3 Antibody Response of Individual Mice to S9 Peptide, CCMV and GBS Plates were coated with S9 peptide, CCMV or GBS. Prebleed (Bleed 1) and post-booster sera, (Bleed 2) at dilutions of 1/5000 for S9, 1/1000 for CCMV and 1/200 for GBS, were added into triplicate wells. Plates then were incubated at 4°C for 18 h. After washing with ELISA washing buffer, the anti-mouse IgG dilution 1/1000 was added into wells and incubated for 4 hours at room temperature. Colorimetric substrate p-nitrophenyl phosphate was added. Absorbance at 405 nm was measured. Results are means and standard errors. 140

ELISA results showed that sera from mice immunized with CCMV-S9, CPMV-

S9 and KLH-S9, with or without adjuvant, made antibody response to the S9 peptide and that immunizing with Freund’s adjuvant increased the antibody response to the S9 peptide. Mice immunized with SubE-S9 antigen, with or without adjuvant were able to induce anti-CCMV response but not an anti-S9 response. Mice immunized with unconjugated CCMV, with or without adjuvant, made antibody responses to CCMV but not the S9 peptide. Adjuvant had little effect on the anti-CCMV response in mice immunized with CCMV, which was different than the effect of adjuvant on the anti-S9 response. It is difficult to judge the development of anti-GBS antibody response because prebleeds had high levels of “natural” anti-GBS antibody, and in controls not immunized with S9, the anti-GBS antibody increased with time. To confirm the results of the first experiment, we repeated the immunizations with a subset of these antigens (Figure 3.4).

We immunized 4 groups of mice (6 mice per group) with the following antigens using no adjuvant: CCMV, CCMV-S9, CPMV-S9 and KLH-S9. All mice were bled to collect sera before immunization (Bleed 1). Mice received the primary immunization sc. Four weeks later, mice were boosted with the same antigen, given ip. Sera were collected 1 week after the booster immunization. Sera collected from CCMV-S9, KLH-S9 and CPMV-S9 bound to plates coated with S9 antigen. CPMV-S9 induced higher anti-S9 antibody levels than CCMV-S9 and KLH-S9. Control mice, immunized with unconjugated CCMV, did not produce anti-S9. Similar problems as described above (Figure 3.3) made analysis of the anti-GBS difficult, but it appears that a small number of mice in CCMV-S9 and

CPMV-S9 groups produced anti-GBS antibody. 141

S9

4.50 4.00 3.50 3.00 2.50 Bleed1 2.00 Bleed2 1.50 1.00 0.50 Optical Density nm405 0.00 M1 M2 M3 M4 M5 M6 M1 M2 M3 M4 M5 M6 M1 M2 M3 M4 M5 M6 M1 M2 M3 M4 M5 M6 CCMV CCMV-S9 CPMV-S9 KLH-S9 Mouse

CCMV 1.20 1.00 0.80 Bleed1 0.60 Bleed2 0.40 0.20

OpticalDensity 405 nm 0.00 M1 M2 M3 M4 M5 M6 M1 M2 M3 M4 M5 M6 M1 M2 M3 M4 M5 M6 M1 M2 M3 M4 M5 M6 CCMV CCMV-S9 CPMV-S9 KLH-S9 Mouse

GBS 1.00 0.90 0.80 0.70 0.60 Bleed1 0.50 0.40 Bleed2 0.30

Optical Density Optical 0.20 0.10 0.00 M1 M2 M3 M4 M5 M6 M1 M2 M3 M4 M5 M6 M1 M2 M3 M4 M5 M6 M1 M2 M3 M4 M5 M6 CCMV CCMV-S9 CPMV-S9 KLH-S9 Mouse

Figure 3. 4 Antibody Response of Individual Mice to S9 Peptide CCMV and GBS Plates were coated with S9 peptide, CCMV or GBS. Prebleed (Bleed 1) and post-booster sera, (Bleed 2) at dilutions of 1/5000 for S9, 1/1000 for CCMV and 1/200 for GBS, were added into triplicate wells. Plates then were incubated at 4°C for 18 h. After washing with ELISA washing buffer, the anti-mouse IgG dilution 1/1000 was added into wells and incubated for 4 hours at room temperature. Colorimetric substrate p-nitrophenyl phosphate was added. Absorbance at 405 nm was measured. Results are means and standard errors. 142

To test whether CCMV-S9, KLH-S9 and CCMV were able to induce cross

reactivity to other phage selected peptides, collected sera from these mice were added to

ELISA plates coated with S7, S9-2 and S11. These peptides were selected with the

indicated monoclonal antibody (S7, S9, or S11). The S9-2 peptide has a completely

different sequence than the S9 peptide used here. The ELISA results showed no cross

reactivity between anti-S9 antibody and any of these other peptides representing group B streptococcal carbohydrate antigens (Figure 3.5, Figure 3.6 and Figure 3.7).

0.90 0.80 0.70 0.60 0.50 0.40 0.30 0.20 Optocal Density 405 nm 405 Density Optocal 0.10 0.00 M1 M2 M3 M4 M5 M6 M7 M8 M17 M18 M19 M20 M21 M22 M23 M24 M25 M26 M27 M28 M37 M38 M39 M40 positive negative CCMV CCMV-S9 KLH-S9 CCMV CCMV-S9 KLH-S9

CFA PBS Mouse

Figure 3. 5 Antibody Response of Individual Mice to S7 Peptide S7 plates were prepared as described in Materials and Methods. Bleed 2 from the first sequence of immunizations was tested. S7 mAb was used as a positive control and PBS was used as a negative control. Plates were incubated at 4°C for 18 h. After washing with ELISA washing buffer, enzyme conjugated anti-mouse IgG was added and incubated for 4 hours at room temperature. Colorimetric substrate p-nitrophenyl phosphate was added. Absorbance at 405 nm was measured and shown as mean and SEM. 143

0.80 0.70 0.60 0.50 0.40 0.30 0.20 0.10 0.00 Optical Density 405 nm Density Optical M1 M2 M3 M4 M5 M6 M7 M8 M17 M18 M19 M20 M21 M22 M23 M24 M25 M26 M27 M28 M37 M38 M39 M40 positive negative CCMV CCMV-S9 KLH-S9 CCMV CCMV-S9 KLH-S9

CFA PBS Mouse

Figure 3. 6 Antibody Response of Individual Mice to S9-2 Peptide S9-2 plates were prepared as described in Materials and Methods. Bleed 2 from the first sequence of immunizations was tested. S9-2 mAb was used as a positive control and PBS was used as a negative control. Plates were incubated at 4°C for 18 h. After washing with ELISA washing buffer, enzyme conjugated anti-mouse IgG was added and incubated for 4 hours at room temperature. Colorimetric substrate p-nitrophenyl phosphate was added. Absorbance at 405 nm was measured and shown as mean and SEM.

0.60 0.50 0.40 0.30 0.20 0.10 0.00 Optical Density nm 405 Density Optical M1 M2 M3 M4 M5 M6 M7 M8 M17 M18 M19 M20 M21 M22 M23 M24 M25 M26 M27 M28 M37 M38 M39 M40 positive negative CCMV CCMV-S9 KLH-S9 CCMV CCMV-S9 KLH-S9

CFA PBS Mouse

Figure 3. 7 Antibody Response of Individual Mice to S11 Peptide S11 plates were prepared as described in Materials and Methods. Bleed 2 from the first sequence of immunizations was tested. S11 mAb was used as a positive control and PBS was used as a negative control. Plates were incubated at 4°C for 18 h. After washing with ELISA washing buffer, enzyme conjugated anti-mouse IgG was added and incubated for 4 hours at room temperature. Colorimetric substrate p-nitrophenyl phosphate was added. Absorbance at 405 nm was measured and shown as mean and SEM. 144

The Th1/Th2 Response Is Influenced by Antigen Conjugate Carriers

To determine the Th1/Th2 profile of the antibody response, the subclass profile of

the humoral IgG response, antigen-specific IgGl, IgG2a and IgG2b, and IgG3, was analyzed using ELISA as described in Materials and Methods (Figure 3.8 and Figure

3.9).

2.00

1.80

1.60

1.40

1.20 IgG1 IgG2a 1.00 IgG2b 0.80 IgG3 0.60 Optical Density 405 nm 0.40

0.20

0.00 M5 M6 M7 M8 M34 M35 M36 M37 M38 M39 M40 M13 M14 M15 M16 M17 M18 M19 M20 M25 M26 M27 M28 M33 CCMV - S9 CPMV - S9 KLH- S9 CCMV - S9 CPMV - S9 KLH- S9

PBS CFA Mouse

Figure 3. 8 IgG Subclass Identification from the First Immunization Experiment Bleed 2 sera from the first immunization were added to ELISA plates coated with S9 peptide. Following this, rabbit anti mouse IgG1, IgG2a, IgG2b or IgG3 antisera were added and incubated at 4°C for 18 h. Goat anti rabbit IgG antibody conjugated with alkaline phosphatase was then added to wells for 6 h, followed by washing and addition of colorimetric substrate. A405 was determined using a microplate reader. Results are means and SEM of triplicate samples.

Antibody subclass was determined on Bleed 2 sera from mice in the first set of

immunizations (Figure 3.8), which had been immunized with KLH-S9, CPMV-S9, and

CCMV-S9, with or without adjuvant. In the non-adjuvant group, CCMV-S9 and CPMV- 145

S9 predominantly induced antigen-specific IgG2a whereas KLH-S9 predominantly

induced antigen-specific IgG1 (Figure 3.8). When adjuvant was used, these differences

disappeared, all subgroups showed a mixed Th1/Th2 response as reflected by equivalent levels of antigen specific IgG1, IgG2a and IgG2b (Figure 3.8).

The result was confirmed using sera from the second immunization experiment.

Immunization with the virus carrier conjugate induced predominantly antigen-specific

IgG2a for a Th1 type of response, whereas KLH-S9 predominantly induced antigen-

specific IgG1, a Th2 pattern (Figure 3.9).

2.50

2.00 IgG1 1.50 IgG2a 1.00 IgG2b IgG3 0.50

Optical Density 405 nm 0.00 M1 M2 M3 M4 M5 M6 M1 M2 M3 M4 M5 M6 M1 M2 M3 M4 M5 M6

CCMV-S9 CPMV-S9 KLH-S9 Mouse

Figure 3. 9 IgG Subclass Identification from the Second Immunization Bleed 2 sera from the second immunization were added to ELISA plates coated with S9 peptide. Following this, rabbit anti mouse IgG1, IgG2a, IgG2b or IgG3 antisera were added and incubated at 4°C for 18 h. Goat anti rabbit IgG antibody conjugated with alkaline phosphatase was then added to wells for 6 h, followed by washing and addition of colorimetric substrate. A405 was determined using a microplate reader. Results are means and SEM of triplicate samples.

146

Immune Response after Injection with Heat-Killed Group B Streptococcus

We next asked if mice primed with the S9 peptide were able to produce rapid

GBS-specific antibody response after exposure to intact GBS. From the first

immunization, six groups of mice (CCMV, CCMV-S9, KLH-S9), with and without

adjuvant, were injected with heat-killed GBS. Sera were collected 3 and 7 days after

injection (Bleed 3 and Bleed 4) to test for the induction of S9 and GBS-specific antibodies. All groups of mice produced small amounts of GBS-specific antibodies

(Figure 3.10) and S9-specific IgG antibodies (Figure 3.11) at day 7, but not day 3 after

immunization.

2.50 2.00 Bleed2 1.50 Bleed3 1.00 Bleed4 0.50

Optical Density 405 nm 405 Density Optical 0.00 M1 M2 M3 M4 M5 M6 M7 M8 M17 M18 M19 M20 M21 M22 M23 M24 M25 M26 M27 M28 M37 M38 M39 M40 CCMV CCMV-S9 KLH-S9 CCMV CCMV-S9 KLH-S9

CFA PBS Mouse

Figure 3. 10 GBS-Specific IgG Antibodies after Injection of GBS Mice were injected with heat-killed GBS. Sera were collected 3 and 7 days after injection (Bleed 3 and Bleed 4) to test for GBS-specific IgG antibodies nd compared to Bleed 2 sera (obtained prior to exposure to GBS). Sera were diluted to 1/200 and added to GBS coated plates in triplicate and incubated. The plates were washed and incubated with alkaline phosphatase-conjugated anti mouse-Ig, followed by washing and addition of colorimetric substrate. A405 was determined using a microplate reader. Results are mean and SEM.

147

3.00 2.50 2.00 Bleed2 1.50 Bleed3 1.00 Bleed4 0.50

Optical Density 405 nm Density Optical 0.00 M1 M2 M3 M4 M5 M6 M7 M8 M17 M18 M19 M20 M21 M22 M23 M24 M25 M26 M27 M28 M37 M38 M39 M40 CCMV CCMV-S9 KLH-S9 CCMV CCMV-S9 KLH-S9

CFA PBS Mouse

Figure 3. 11 S9-Specific IgG Antibody after Injection of GBS Mice were injected with heat-killed GBS. Sera were collected 3 and 7 days after injection (Bleed 3 and Bleed 4) to test for S9-specific IgG antibodies, and compared to Bleed 2 sera (obtained prior to exposure to GBS). Sera were diluted to 1/5000 and added to S9 coated plates in triplicate and incubated. The plates were washed and incubated with alkaline phosphatase-conjugated anti mouse-Ig, followed by washing and addition of colorimetric substrate. A405 was determined using a microplate reader. Results are mean and SEM.

All groups of mice produced more GBS-specific and S9-specific IgM antibodies

at day 7, but not day 3, after immunization (Figure 3.12). The results confirmed that GBS

immunization results in the production of anti-S9 antibody. But the results did not

provide any evidence that immunizing with the S9 peptide resulted in priming for a

challenge with GBS.

148

GBS

2.00 1.80 1.60 1.40 Bleed2 1.20 1.00 Bleed3 0.80 0.60 Bleed4 0.40 0.20 0.00 Optical Density 405 nm M1 M2 M3 M4 M5 M6 M7 M8 M17 M18 M19 M20 M21 M22 M23 M24 M25 M26 M27 M28 M37 M38 M39 M40 CCMV CCMV-S9 KLH-S9 CCMV CCMV-S9 KLH-S9

CFA PBS Mouse

S9

3.00 2.50 2.00 Bleed2 1.50 Bleed3 1.00 Bleed4 0.50 0.00 Optical Density 405 nm M1 M2 M3 M4 M5 M6 M7 M8 M17 M18 M19 M20 M21 M22 M23 M24 M25 M26 M27 M28 M37 M38 M39 M40 CCMV CCMV-S9 KLH-S9 CCMV CCMV-S9 KLH-S9

CFA PBS Mouse

Figure 3. 12 GBS-Specific IgM and S9-Specific IgM Antibodies after Injection of GBS Mice were injected with heat-killed GBS. Sera were collected at 3 and 7 days after injection (Bleed 3 and Bleed 4) to test for GBS-specific IgM antibodies, and compare to Bleed 2 sera (obtained prior to exposure to GBS). Sera were diluted to 1/200 for GBS- specific IgG antibody response or diluted to 1/5000 for S9 -specific IgG antibody response. The diluted sera were added to GBS or S9 coated plates coated plates in triplicate and incubated. The plates were washed and incubated with rabbit anti-mouse IgM, and then alkaline phosphatase-conjugated anti rabbit-IgG. A405 was determined using a microplate reader. Results are mean and SEM.

149

Cytokine Production in Primed Splenocytes after Antigen Stimulation

Because the subclass analysis of the antibody response suggested that immunizing

with antigen in association with virus induces a Th1 type of response, whereas antigen

conjugated to KLH induced a Th2 response, we examined cytokine expression in helper

T cells following antigen stimulation. Selected groups of mice from the second immunization were boosted a second time ipwith the same antigen. Immunized and naive mice were sacrificed 4 days after the last booster. The spleens and mesenteric lymph nodes were removed aseptically. Splenocytes were purified and were plated 100 ul per well (2×105 cell per well) into sterile 96-well microplates (Costar, Corning Inc., NY) and

stimulated either with CCMV, CCMV-S9, S9 peptide alone, KLH-S9, or KLH, at

concentration 20 ug/ml, PHA at concentration 10 ug/ml (positive control), or with sterile

RPMI-medium supplement with 10% FCS (negative control). Cytokine production on

days 3, 5 and 7 was measured and is represented in figures 3.13. The data indicated that

splenocytes primed with CCMV or CCMV-S9 and then stimulated with CCMV or

CCMV-S9 produced large amounts of IFN-γ, with variable amounts of IL10 and no IL4.

On the other hand, splenocytes primed with KLH-S9 and then stimulated with KLH-S9

or KLH produced IL4 and IL10, and little IFN-γ. There appeared to be a greater overall cytokine response when the immunogen is associated with virions. These data showed that immunizing with virus particles expressing S9 provide a Th1-biased response whereas immunization with protein, KLH-S9 induces a Th2-biased response. These results are consistent with the Th1 vs Th2 bias indicated by the subclass of antibody produced. 150

9000.00

8000.00

7000.00 3 day

6000.00

5000.00 CCMV 4000.00 CCMV-S9 KLH-S9 3000.00

2000.00

1000.00 Cytokine concentration (pg/ml) concentration Cytokine

0.00 S9 S9 S9 KLH PHA KLH PHA -1000.00 KLH PHA CCMV CCMV CCMV KLH-S9 KLH-S9 KLH-S9 medium medium medium CCMV-S9 CCMV-S9 CCMV-S9 IL-4 pg/ml IL-10 pg/ml IFN-g pg/ml

6000.00 5 day 5000.00

4000.00

3000.00 CCM V CCMV-S9 2000.00 KLH-S9

1000.00 Cytokine Concentration (pg/ml) Concentration Cytokine

0.00 S9 S9 S9 KLH PHA KLH PHA KLH PHA CCMV CCMV CCMV KLH-S9 KLH-S9 -1000.00 KLH-S9 medium medium medium CCMV-S9 CCMV-S9 CCMV-S9 IL-4 pg/ml IL-10 pg/ml IFN-g pg/ml

9000.00 7 day 8000.00

7000.00

6000.00

5000.00 CCMV 4000.00 CCMV-S9 KLH-S9 3000.00

2000.00

1000.00 Cytokine Concentration (pg/ml)

0.00 S9 S9 S9

-1000.00 KLH PHA KLH PHA KLH PHA CCMV CCMV CCMV KLH-S9 KLH-S9 KLH-S9 medium medium medium CCMV-S9 CCMV-S9 CCMV-S9 IL-4 pg/ml IL-10 pg/ml IFN-g pg/ml

Figure 3. 13 Cytokine production in primed splenocytes 3, 5 and 7 days after antigen stimulation Mice immunized with CCMV, CCMV-S9 or KLH-S9 were sacrificed. Splenocytes were purified, cultured in vitro, and stimulated either with antigen, PHA, or medium. The supernatants were collected on day 3, 5 and 7 and analyzed for cytokine production. Results are from individual mice immunized with the antigen indicated in the legend.

151

Discussion

Immunization with capsular polysaccharide (CPS) of GBS has been known to

induce protective antibodies (85). Vaccines using CPS (191) or CPS conjugated to

protein carriers (193, 346) have been developed. In our lab, we used an alternative

approach to GBS vaccine development. A CPS mimetic peptide, S9, was previously

identified by the means of phage display (1). The S9 peptide consists of 12 amino acids

and is immunogenic when coupled to large proteins. Conjugating peptide epitope to large proteins such as KLH is a common method to enhance immunogenicity of the peptide.

An alternative way, using viruses to present epitopes by forming chimeric viruses and virus-like particles has been developed (310, 340, 352, 354). CPMV, an icosahedral virus comprised of 60 capsomers, has been used as a viral vector to express an epitope on the virus coat protein (299, 313).

In the present study, we aimed to improve the immunogenicity of the GBS

mimotope, S9 peptide, by chemically coupling it to the plant virus particle, CCMV. We

choose the CCMV as virus vector to deliver the S9 peptide to the immune system

because the X-ray crystal structure of CCMV has been resolved (326-328) and it has been

developed as a new expression vector in the Young lab( Montana State University).

CCMV is a member of Bromoviridae virus family. It is an icosahedral virus 28 nm in

diameter. The structure of CCMV has been determined to 3.2 A° resolution (317).The

virus particle is composed of 180 copies of the coat protein. The S9 peptide was

chemically conjugated to four different carriers: CCMV, CPMV, SubE (CCMV mutant)

and KLH. The CPMV-S9 conjugate is a gift from Dr. John E. Johnson (Scripps Institute, 152

La Jolla CA). CPMV is a picorna-like icosahedral virus with size of 30 nm with 60

copies of coat protein subunit (355). The crystal structure of CPMV has been refined

(356). CPMV has been used as a virus vector to present many epitopes in chimeric form

and as chimeric VLPs (299, 310, 356-358). The amount of S9 peptide coupled to

CPMV-S9 was two-fold higher than the amount of S9 protein antigen in the CCMV-S9

and SubE-S9. The reactive lysine side chains on the virus coat protein were used in the

conjugation procedure for both CCMV-S9 and CPMV-S9. This may indicate that lysines

are less accessible on the CCMV coat protein than on CPMV.

Mice immunized with CCMV-S9 can produce anti-S9 antibody even without

adjuvant. With the help of adjuvant, the antibody response to S9 peptide increased. The

anti-S9 antibody response from mice immunized with CPMV-S9 and KLH-S9 conjugate

seemed slightly higher than the response from mice immunized with CCMV-S9. The degree of organization of antigens has been shown to be important and to have effects on the B cell response (295, 352, 359). The S9 antigen in the CCMV-S9 conjugate might not be optimal for antibody production. Even though the amount of S9 peptide in each antigen was equal in each sample, the amount of S9 per virus particle was less in the

CCMV-S9 conjugate. Thus, more repetition of antigen in CPMV-S9 enhanced antibody response. This explanation was supported by the result of SubE-S9 conjugation immunization. SubE-S9 conjugate is a monomeric form of mutant CCMV-S9 conjugate.

The SubE is the CCMV mutant in which eight basic amino acid residues (K and R) of the

N-terminus were replaced with acidic residue glutamic acid(320). Conjugating S9 to the

SubE led to a virus that falls apart (monomeric form). Mice immunized with SubE-S9 153 conjugate without adjuvant resulted in less anti-S9 antibody production. When adjuvant was used, the response increased but the level was still lower than response from other conjugates. Experiments using papaya mosaic virus-like particles fused to a hepatitis C virus epitope also demonstrate that the immunogenicity is highly dependent upon antigen organization. Immunization with the chimeric VLP induces long-lasting higher antibody response compared to immunization with the monomeric form (352).

In the second immunization experiment, the result indicated a better anti-S9 response in mice immunized with CPMV-S9 when compared with the KLH-S9. However this observation cannot be obviously seen in the first immunization experiment, but is consistent with our other observations that virion-associated antigen may be more immunogenic than antigen attached to soluble proteins (352).

Sera from all mice immunized with CCMV-S9, CPMV-S9, and KLH-S9 were tested for cross reactivity to other peptides (peptide S9-2, S7 and S11) representing group

B streptococcal carbohydrate antigens. S7 and S11 were detected using Mabs specific for other GBS antigens, whereas S9-2 was from the same set of experiments that produced the S9 peptide, but eluted at acid, rather than alkaline pH (1). The result indicated no cross reactivity between antibody elicited by immunization with any of the S9-antigen preparations, and any peptides. This shows that the S9 peptide elicits antibodies specifically to this peptide, and there is little if any antigenic mimicry among the peptides.

The anti-GBS antibody response was tested in sera from all mice. The result is difficult to interpret. Control mice showed an increase in anti-GBS antibody with time, 154 suggesting possible cross-reactions with the normal flora of the mice (mice were tested and negative for GBS). Also results were weak and inconsistent. In the first immunization experiment, an anti-GBS response seemed to be induced in mice immunized with or without adjuvant with CCMV-S9, CPMV-S9, SubE-S9 and KLH-S9.

However, when the sera from the second immunization experiment were tested for anti-

GBS antibody, an anti-GBS antibody response was only seen in CPMV-S9, when compared to the prebleed sera.

Immunized mice were injected with heat-killed GBS to test the ability of the S9 conjugates to prime for a secondary response to the microbial antigen expressed on the killed GBS. The results indicate that immunization with the S9 peptide was unable to do so. Interestingly, the injection of the GBS elicits a strong anti-peptide IgM response at day 7.

The goal of vaccination is to provide protection from infection. Effective stimulation of the immune system depends upon stimulation of strong T-cell response. In general, Th1 cells stimulate cell-mediated immunity (CMI), while Th2 induce an antibody-mediated response. In vaccine design, the type of immune response is important because different immune responses have different effects on different pathogens. In general, Th1 cells are required for the elimination of intracellular pathogens, whereas the

Th2 cells are required for inhibition of attachment or entry of microbes during extracellular stages of the infectious cycle. The route of immunization and the form of the antigen and delivery method can have effects on the immune response. The antibody subclass profile (IgG1, IgG2a, IgG2b and IgG3 in mice), can suggest a Th1 or Th2 155

response. In our study, mice immunized without adjuvant with the CCMV-S9 and

CPMV-S9 conjugate predominantly produce anti-S9 specific IgG2a, whereas KLH-S9

predominantly induced anti-S9 specific IgG1. The antibody profile result suggested that

S9 conjugated to virus particle induced Th1 response, whereas conjugation S9 peptide to

soluble protein (KLH) induced a Th2 response. Notably, the Th1/Th2 bias was obscured

when adjuvant was used in the immunization. To substantiate that the antigen, in the form of virus particles affects the Th1/Th2 response, we examined cytokine expression in helper T cells following antigen stimulation. The data indicated that splenocytes primed with CCMV or CCMV-S9 and then stimulated with CCMV or CCMV-S9 produced the

Th1 cytokine, IFN-γ. On the other hand, splenocytes primed with KLH-S9 and then stimulated with KLH-S9 or KLH produced the Th2 cytokines, IL4 and IL10. The ability of multimeric VLP vaccines to induce a Th1 biased response has been shown before

(360-363). This may result from the multimeric form of the antigen, or it may be the effect of RNA within the VLP or virion contributing to adjuvanticity. CCMV virion and pseudovirion assembly has been shown to be dependent upon the inclusion of RNA within the capsid (364-368). Whatever the mechanism, our study clearly indicated that immunizing with virus particles expressing S9 induces a Th1 response, whereas immunization with protein, KLH-S9, induced a Th2 response.

156

PHAGE DISPLAY LIBRARY

Introduction

Group B streptococci (GBS) can cause neonatal sepsis and meningitis (369).

Neonates are exposed to GBS either via GBS ascension through ruptured amniotic

membranes, aspiration of contaminated amniotic fluid, or through exposure to colonized

tissues while passing through the birth canal (344). GBS possess a polysaccharide

capsule which allows the GBS to avoid opsonophagocytosis (145). Capsular

polysaccharides (CPS) are poorly immunogenic and induce a T-independent response

without immunologic memory. GBS vaccines using conjugates of the carbohydrate to a

protein carrier have been developed and are able to generate a T-dependent response

(370).

In our lab, we have followed an alternative strategy, using peptides that mimic the

structure of microbial polysaccharides. Phage display technique was used to seek

peptides that mimic carbohydrate epitopes. The S9 mAb, which demonstrates in vivo

protection against GBS type III (172, 174) was used to select displayed peptides. The S9

peptide, FDTGAFDPDWPA was isolated. The S9 peptide was shown to be both

antigenic and immunogenic (1), specifically blocking the binding of antibody to type III

CPS and eliciting antibody that cross-reacts with GBS.

In this thesis we further investigated the structural basis of the peptide mimicry of

the GBS type III CPS carbohydrate epitope. We designed a sublibrary, expressing a mutated version of the S9 peptide with 3% random nucleic acids at each position of the 157

original S9 nucleotide sequence, creating approximately one mutation in each inserted

sequence. Peptides with higher affinity to the S9 mAb were identified by affinity

selection. From this sublibrary, we also explored effects of the amino acid changes in the

antigenic S9 peptide. Phages displaying modified S9 peptides therefore were randomly

selected. Peptides from both affinity selection and random selection were characterized.

The results allow us to map the antigenic portion of the S9 peptide, and suggest ways to

generate better mimotopes of GBS.

Materials and Methods

Peptide Library Construction

To create a library expressing mutated versions of the S9 peptide, we inserted oligonucleotide sequences between the BglII sites of the filamentous phage vector

fUSE2 (237). The inserted sequence is expressed at the amino terminus of the pIII coat

protein, giving the final amino acid sequence H2N-A EDPG [FDTGAFDPDWPA] GGP

TD. The modified S9 oligonucleotide sense strands (New England Biolabs) were

designed so that the nucleotide sequence, indicated within the brackets, consisted of 97%

of the appropriate nucleotide with the other 3% consisting of the other three nucleotides

at each position, thus creating approximately one mutation in each inserted sequence:

5’CAAA G GATCC AGGA [TTC GAT ACG GGC GCG TTC GAT CCA GAT TGG

CCA GCG] GGT GGG CCC ACG G 3’. The antisense strands, which were used to

prime the fill-in reaction, were 5’GAT GGA TCC GTG GGC CCA CC 3’. The red highlight indicates BamHI sites. Each strand was heated at 95°C for 3 min and rapidly 158

cooled down for 5 min. Sense S9 library strands 25 ul of 40 ng/ul were mixed with 2 ul of

40 ng/ul antisense strands, giving 1:1 molar ratio reaction. The mixture was incubated 15

min at 50 °C and 15 min at 37 °C and then allowed to cool down to room temperature.

Klenow Fragment (Takara), its buffer, and dNTP mix were added to the mixture to have

total volume of 250 ul. The reaction was incubated at 37°C 3 h and at 65°C for 5 min.

The oligonucleotide library was then digested with BamH1 (Promega). The double

stranded mutated S9 oligonucleotide DNA library was then purified by phenol-

chloroform extraction, followed by ethanol precipitation. The library was ligated into Bgl

II digested, and dephosphorylated fUSE2. The ligation reaction was performed at a vector to insert ratio of 1:2 with T4 DNA ligase (Promega, Madison, WI) and 1 ul of 10x ligation buffer (containing 100mM mgCl2 and 10 mM ATP, Promega) in 10 ul total volume. The reactions were incubated at 4°C overnight to facilitate ligation. Digested and dephosphorylated fUSE2 plasmids were used as a control. The ligation reactions were precipitated by mixing the mixture well with two volumes of 95% ethanol and 1/10 volume of 3 M sodium acetate pH 5. The mixture was incubated at –20°C overnight.

DNA was collected by centrifuging12 krpm 30 min. The DNA pellet was washed with

70% ethanol and air dried. DNA was resuspended in10 ul water. Two transfections were performed by electroporating 4 µl DNA into MC1061 E. coli (371, 372). SOC medium

with 0.2 ug/ml tetracycline was added to the transformation reactions and the mixtures

were incubated in an incubated shaker at 37°C for 1 h. The transfection reactions were

spread on LB media agar plates supplemented with 40 ug/ml tetracycline and incubated

at 37°C overnight, or added to 200 ml LB broth media supplemented with 40 ug/ml 159

tetracycline and incubated with shaking at 37°C overnight. Phages were purified from the

culture by using large scale phage purification methods described below.

Plasmid Extraction

A modified alkaline lysis method was used to purify plasmid DNA (321). 3 ml of overnight culture in LB broth with tetracycline 20 ug/ml were spun to collect the cell pellet. Cells were resuspended in 100 ul of 20 mM Tris buffer pH 8 and then incubated at room temperature for 5 min. 200 ul of lysis buffer (fresh prepared 0.2 N NaOH, 1% SDS) were added and the mixture was mixed. The mixture was incubated at RT for 10 min and then 150 ul of KAC/AC (5 M potassium acetate, 10 M (57%) acetic acid) were added.

The mixture was mixed vigorously and sat in ice for 30 min. Cell debris was removed by centrifugation at 12 krpm 10 min twice. Supernatant was collected and was mixed with equal volume of phenol/chloroform/isopropanol. The upper phase was collected. DNA was precipitated with 95% ethanol and sodium acetate, and washed with 70% ethanol.

DNA was resuspended in10 ul water. DNA was characterized by running on 0.8-1.5% agarose gels.

PCR

Double stranded plasmid DNA (RF form of phage) was studied to assure the

presence of DNA inserts. DNA from individual transfectant colonies was mixed with

PCR buffer with MgCl, dNTP mix, primers (fUSE2 primer1 (TCAAGATCCGTGGGC-

C CACC) and fUSE2 primer2 (GCTTCTCTTATGATTGACCGTCTGCG)), Taq 160

polymerase, and the DNA template. PCR cycles were 94°C for 30 s, 60°C for 30 s, and

72°C for 30 s. 5 ul of PCR product were loaded onto 1% agarose gels.

Single Stranded DNA (ssDNA) Extraction (373)

Phage clones containing inserts, as verified by PCR, were cultured in 1.5 ml LB containing 20 ug/ml tetracycline and incubated overnight at 37°C with shaking. Cultures were centrifuged and 1 ml of supernatant was mixed with 150 ul PEG/NaCl solution

(16.7%/3.3M). The tube was inverted several times and kept at 4°C overnight. The tube was centrifuged at 12 krpm for 10 min and all supernatant was removed. The pellet, containing precipitated phage virions, was dissolved in 200 ul TE buffer (10 mM

Tris·HCL 1mM EDTA pH8). An equal volume of phenol/chloroform solution

(phenol:chloroform = 1:1) was added to the phage solution and mixed well. After centrifugation, 200 ul of the aqueous phase was transferred to another tube. The ssDNA was precipitated by adding 20 ul of 3M sodium acetate pH6 and 440 ul 95% ethanol and kept at 4°C overnight. DNA was obtained by high speed centrifugation at 12 krpm for 30 min. The pellet was washed with 500 ul 70% ethanol and allowed to air-dry. The pellet containing ssDNA was resuspended in 20 ul of sterilized deionized water. The ssDNA was kept in -20°C until sent for sequencing analysis (Davis Sequencing, Inc, Davis, CA), using fUSE2 primer, CCCTCATAGCGTAACG. Peptide sequences were inferred from the DNA sequences using MacVector 6.5.3 and EditView 1.0.1.

161

Phage Amplification

E. coli stain K91Kan was cultured in a 10 ml LB broth containing 100 ug/ml

kanamycin in an incubation shaker 37°C until an OD 600 of a 1/10 dilution reached 0.2.

Purified phages or the eluate from the phage affinity selection were added to the culture

and incubated for 15 min at 37°C without shaking. A fresh 10 ml of LB media containing

0.2 ug/ml tetracycline was added and incubated at 37°C with high speed shaking (250

rpm) for 30 min. Tetracycline was added to the culture to a final concentration of 20

ug/ml. The total culture was spread on an LB agar plate containing 20 ug/ml tetracycline and 100 ug/ml kanamycin in a 9x24 inch glass baking dish. The plate was incubated at

37°C overnight. On the next day, 25 ml of TBS (50mM Tris-Cl, 150 mM NaCl pH7.5) with 0.5% Tween 20 was flooded over the culture plate. An alcohol-flamed sterilized glass spreader was used to gently remove colonies from the plate. Phage particles were purified and quantified by phage titration method as described below. The phages were further reamplified using 100 ml of E. coli stain K91Kan. Purified phage particles from the previous amplification were added to the culture, incubated, and purified as described above.

Phage Purification

Phage were purified as follows (374). The overnight culture of phage-infected

K91Kan was centrifuged at 8 krpm for 10 min. The supernatant was transferred and

mixed with 0.15 volume of PEG/NaCl. The phage particles were allowed to precipitate

overnight at 4°C. The pellet containing phages was collected by centrifugation at 10 162

krpm for 30 min. The pellet was resuspended with 15 ml TBS buffer. The phages were

reprecipitated 3 times with PEG/NaCl. After the final round of precipitation, the pellet

was resuspended in 1 ml TBS phage buffer. Phage were quantified by titration, preserved

by adding 50% glycerol, and stored at 4°C. Phage protein concentration was measured by

using BCA protein assay reagent kit (Pierce, Rockford, IL).

Phage Titration

Phage were titrated as follows (375) (376). Starved E. coli K91Kan were used as host cells for phage titration. K91Kan cells were cultured in 20 ml LB media supplemented with 100 ug/ml kanamycin at 37°C with shaking until reaching OD600

0.4-0.5. Cells were centrifuged at 2.5 krpm for 10 min and resuspended in 20 ml 80 mM

NaCl. The cell suspension was incubated in the 37°C shaker incubator for 45 min and

centrifuged 2.5 krpm for 10 min. Cells were resuspended in 1 ml cold NAP buffer (80

mM NaCl and 50 mM NH4H2PO4, adjusted pH to 7 with NH4OH) and kept at 4°C until

use. Purified phages were 10-fold serially diluted with TBS phage buffer with 100 ul

final volume. Phages were allowed to infect the cells at room temperature for 10 min and

the mixture was added with 1 ml LB supplemented with 0.2 ug/ml tetracycline. The infection was allowed to continue in the 37°C shaking incubator for 40 min. 100 ul/plate of infected cells were spread on LB agar supplemented with 40 ug/ml tetracycline and

100 ug/ml kanamycin. Culture plates were incubated overnight in the 37°C incubator.

Infected colonies able to grow on the tetracycline/kanamycin plates were counted and calculated as transducing units (TU titer).

163

Affinity Selection of Phage-Displayed Peptides

S9 monoclonal antibody (mAb) was diluted with coupling buffer (0.1 M

NaHCO3, 0.5 M NaCl, pH8.3-8.5) to 1.5 mg/ml. Cyanogen bromide-activated Sepharose

4B (Sigma Chemical Co., St. Louis, MO) was washed and swollen in 200 ml cold 1mM

HCl per gram of dry gel for 30 min at 4°C . The resin was washed with 200 ml cold

distilled water 5 times. The resin was then mixed with S9 mAb at 3 mg of Ab/ml of beads

in the coupling buffer overnight at 4°C rotating end-over-end. The S9 mAb-resin was spun down at low speed for 1 min and washed 3X with the coupling buffer. Supernatants were collected for protein assay by using BCA protein assay reagent kit (Pierce,

Rockford, IL). The quantity of S9 mAb conjugated to the matrices was determined by

subtracting total S9 mAb protein released from each coupling step from the input S9

mAb protein. Unreacted groups on resin were blocked with 0.2 M glycine pH 8 overnight at 4°C. The mAb-conjugated resin was separated into three aliquots and kept at 4°C in

0.02% azide until ready to use. The S9 mAb-conjugated resins were mixed with 1x1012-13

phage (1 ml) from the phage library at 4°C overnight. The mixture was loaded into a 5 ml

syringe plastic barrel (Becton, Dickinson and company, Franklin Lakes, NJ). The

unbound phages were removed by washing 10 times with phage buffer. Bound phages

were eluted with 0.5 M NH4OH pH 11 and the pH of the eluate was neutralized

immediately with 1 M acetic acid. The titer of phage in each wash and in each eluate was

determined by the phage titration method as described above. The eluted phage were then

amplified in E. coli strain K91Kan, as described above, to a TU titer of 1012 and reapplied

to the column for a second round of affinity selection. The same incubation, washing, and 164 elution procedures were used. The yield was determined from the second round. The eluted phages were reamplified in K91Kan for the third round of affinity selection. After the last round of selection, individual phages were rescued by infecting starved E. coli strain K91Kan and plated on the LB agar media supplemented with 40 ug/ml tetracycline and 100 ug/ml kanamycin. The yield was determined. Phage ssDNA were purified and sent for sequencing analysis. Phages were purified and characterized for binding to selecting, and other, Abs using ELISA.

ELISA and Antibodies

Direct binding ELISA was used to measure the binding of phages, peptides, and

KLH-peptide conjugates to select Abs. 100 ul of the appropriate concentration of antigen were loaded onto microtiter wells (Immulon 2, Dynatech, McLean, VA). Plates were blocked with ELISA blocking buffer (1% BSA in phosphate-buffered saline (PBS)) 4°C overnight. 100 ul of appropriate concentrations of the following antibodies: S9, S7, S11, or anti-M13 were incubated in microtiter wells at 4°C for 18 h. Unbound antibody was removed by washing with ELISA washing buffer (10 mM phosphate buffer pH 7.4, 150 mM NaCl, 0.05% Tween 20). After several washings, 100 ul of 1/1000 alkaline phosphatase-conjugated anti-Ig was added and incubated at room temperature for 6 h.

After washing, 100 ul of colorimetric substrate p-nitrophenyl phosphate (Sigma

Chemical) (1 mg/mL in 10% diethanolamine buffer, pH 9.6, with 0.5mM MgCl2) was added. A405 was determined using a microplate reader (EL-320, Bio-Tek Instruments,

Winooski, VT). 165

Antigen capture ELISA was also used. Plates were coated with 3 ug/ml of mAbs

S9, S11, or B6.1. After blocking with ELISA blocking buffer, 100 ul of each phage dilution was added. After incubating at 4°C overnight, anti-M13 antisera, at a 1/10,000 dilution was added and incubated at 4°C overnight. Plates were washed and incubated with alkaline phosphatase-conjugated anti-rabbit IgG, followed by washing and the addition of substrate, p-nitrophenyl phosphate. Absorbance at 405 was determined.

ELISA assays for phage or peptide inhibition of binding activity of mAb to Group

B streptococcus (GBS) were performed by coating microtiter wells with GBS (353).

Plates were blocked with ELISA blocking buffer overnight at 4°C. The inhibition assay was performed by preincubation of the mAb with different concentrations of phages or peptides. Small scale purified phage clones were diluted with phage buffer to have concentration of 3 ug/ml. Phages were two-fold serially diluted ranged from 3 ug/ml to

0.006 ug/ml. Peptides were diluted to be 2.5, 0.5, 0.1, 0.2 and 0.004 ug/ml. The KLH- peptide conjugates were two-fold serially diluted, and ranged from 0.5 ug/ml to 0.001 ug/ml. The concentration of mAb used for the inhibition assay was in the middle of the linear portion of the binding curve between mAb and GBS (S9 mAb ascites was diluted to 1/25,000, SIIIV18.C2 was diluted to 1/250,000, SIIIS8.C3 was diluted to 1/3,000). The mixture was allowed to incubate at room temperature for 1 h before placing 100 ul of the mixture in wells and incubating 4°C overnight. The plates were washed with ELISA washing buffer, and incubated with alkaline phosphatase-conjugated anti-mouse Ig, followed by washing and addition of colorimetric substrate p-nitrophenyl phosphate

(Sigma Chemical). Absorbance at 405 nm was measured. 166

ELISA assays for binding activity of sera from immunized mice or mabs to GBS

were performed by coating microtiter wells with GBS (353). Information regarding the

mAbs used in this experiment are listed in table 4.1.

Table 4. 1 Mabs Used in This Study Name Specificity S9 Type III capsular polysaccharide S11 Beta-C protein of type I and II GBS SIIIS8.C3 Complete type III S3.1A6 Type III capsular polysaccharide S3.2A6 Type III capsular polysaccharide S3.1B1 Type III capsular polysaccharide SIIIV18.C2 Complete type III B6.1 Candida albicans

Plates were blocked with ELISA blocking buffer overnight at 4°C. Diluted

antibody was loaded onto the GBS coated plates and incubated at 4°C overnight. After washing, alkaline phosphatase-conjugated anti-mouse Ig was added and incubated. After another wash, colorimetric substrate p-nitrophenyl phosphate was added. A405 was determined using a microplate reader.

Conjugation of Peptides to Keyhole Limpet Hemocyanin (KLH)

Synthetic peptides (Sigma Laboratories (USA)) containing Cys at the COOH

peptide terminals were conjugated to maleimide derivatized KLH (Pierce Chemical Co.,

Rockford, IL). Maleimides react with sulfhydryls at pH 6.5-7.5 to form stable thioether

bonds. Peptides (300 ug) were mixed with KLH (300 ug) in phosphate buffer pH 7 in a

reaction volume of 300 ul. The reaction was allowed to continue at room temperature for

3 h. The unconjugated peptides were removed by using microcon 100 (Millipore,

Billerica, MA) filtration. The conjugated KLH-peptides were quantified by using the 167

BCA protein assay reagent kit. The specificity of conjugated peptides binding to mAb and ability to inhibit mAb to GBS were characterized by ELISA.

Results

Library Constructions

To identify peptides that would engage the S9 mAb more effectively and to map the binding motif within the S9 peptide, we created a mutated phage library based on the

S9 peptide sequence having approximately one mutation per phage. The library was constructed by inserting a randomly modified version of the S9 DNA sequence between the BglII sites of the filamentous phage vector fUSE2. To make the oligomers that would form the library, the sense strands were designed to have 97% correct nucleotide and 3% consisting of the other three nucleotides in each position of the S9 coding sequence.

Klenow fragment DNA polymerase was used to extend the antisense strands complementary to the sense strands resulting in the oligomer of modified S9 DNA fragments (Figure 4.1)

Figure 4. 1 Construction of Randomly Modified S9 DNA ds Oligomer Red, white and blue areas indicate BamHI site, the modified S9 library and bases encoding unmodified AA’s, respectively.

168

The modified S9 DNA oligomers were digested with BamH1 restriction enzyme.

The digested products were purified and loaded onto the 12% polyacrylamide gels

(Figure 4.2).

Annealing S9

Annealing S9 + klenow fragment

Annealing S9 + klenow fragment + BamHI+purification

50bp DNA ladder

Figure 4. 2 Modified S9 DNA Fragment Library Sense S9 library strands 25 ul of 40 ng/ul were mixed with 2 ul of 40 ng/ul antisense strands. Klenow fragment (Takara), its bufferand dNTP mix were added to the mixture to have total volume of 25 ul. The reaction was incubated at 37 °C at least 3 h and at 65 °C for 5 min. The complete set of double stranded S9 DNA oligomers was then digested with BamH1 and purified by using phenol-chloroform extraction followed by ethanol precipitation.

The library was ligated into the ends of restriction digested Bgl II sites of fUSE2 vector located at the amino terminal part of the pIII coat protein giving the final amino acid sequence H2N-A EDPG [FDTGAFDPDWPA] GGPT D. The blue letters indicate the S9 peptide sequence that was modified. Two batches of electroporation were performed, which gave a transfection efficiency of 4.7 x 104 and 5 x 104 colonies, respectively. Only 36 colonies derived from transfection with control dephosphorylated fUSE2 plasmid.

169

Random and Affinity Selection of Phage

Phages were selected using two different screening protocols, random selection

and affinity selection. In random selection, the entire first electroporated reaction was

spread on the LB supplemented with 40 ug/ml tetracycline with appropriate dilution to

provide isolated colonies on plates. 1,224 isolated colonies were randomly selected for

PCR analysis to identify positive clones containing inserts in the correct orientation. A

642 bp PCR product was expected from the positive clones (Figure 4.3). From 1224

random-selecting clones, 493 clones (40.3%) contained foreign DNA inserts in the

correct orientation.

Figure 4. 3 PCR Analysis of Transfected Clones Recombinant plasmids (RF form) were isolated from each clone and then PCR amplified. 5 ul of PCR products were analyzed by 2% agarose gel electrophoresis. The first lane was 1 kb DNA marker. The last lane is negative

Single stranded phage DNAs from the total 493 positive clones were analyzed by

DNA sequencing. Table 4.2 displays a summary of the DNA sequencing analysis.

170

Table 4. 2 Summarized DNA Sequencing Analysis Number of clones Affinity selection Description Random selection Second Third round round Base change, resulting in amino acid change 120 5 4 Base change without amino acid change 17 1 2 Original S9 identical 356 9 18 Subtotal 493 15 24 Total 532 Single stranded DNA from 532 recombinant clones from random selection and affinity selection were sent for sequencing analysis (Davis Sequencing, Inc, Davis, CA).

The sequencing result indicated that 72.2% of positive phage clones encoded the

S9 identical sequence, FDTGAFDPDWPA, without any DNA base alteration. Only

27.8% of the positive phage clones contained modified S9 DNA sublibrary inserts. From

120 phage clones contained the modified S9, 41 clones encoded redundant amino acid

sequences. In conclusion only 79 clones encoded modified S9 sequences. All 79 clones

were cultured and the phages were purified for further characterization.

In the affinity selection, the second electroporation batch was added to 200 ml LB broth media supplemented with 40 ug/ml tetracycline and cultured overnight in a shaking

incubator at 37 °C overnight. Phages were purified from the culture by using the large scale phage purification method described above. After amplification, 1.2x1013 phages

were obtained for phage displayed peptide selection with S9 monoclonal antibody. S9

mAb was conjugated to cyanogen bromide-activated sepharose, containing 0.7 mg of S9 mAb conjugated to the resin. The S9-conjugated matrices were separated into three aliquots, resulting in approximately 0.2 mg of S9 mAb for each affinity selection round.

After the first affinity selection round, the yield of affinity selected phage was 5x105. The 171

output phages were amplified to have 8x1012 TU for the second round of affinity selection. The yield in the second round was 1x108. The output phages were amplified to

4x1012 TU for the last round of affinity selection, yielding 4.8x109 phage. 15 and 24

clones from the second and the third round were sequenced, respectively. Single stranded

phage DNAs were purified and sent for sequencing analysis. A summary of the sequencing data is displayed in the table 3.2. Four clones isolated from the second round and three clones isolated from the final round displayed a modified sequence. Three (2 from the second round, one from the final) displayed FDTGAFDPDWRA (designated

RP1.8).

Specificity and Avidity of Phage-Displayed Peptides

In order to determine the binding between the selected phages and mAbs, ELISAs

using different antibodies were performed. A total of 86 phages, from both random and

affinity selection, were tested. Phage binding to S9 mAb and anti-M13 antibody is shown

in Figure 4.4 and 4.5. With the same phage concentration, binding by anti-M13 was more

or less equivalent for all phage preparations. In contrast, binding of the mutant phages by

the S9 mab was different, with some having increased binding, some equivalent, some

less, and some lacking binding altogether; when compared to phage displaying the

original S9 peptide. Six out of seven phages selected by affinity selection showed

increased S9 binding activity. A phage displaying the wrong orientation of S9

surprisingly showed some degree of binding by S9 mAb.

172

g activity was measured. was measured. sequences. sequences. ation of the S9 Individual Phage Individual Individual Phages Individual 50 ug/mlfor S9ug/ml,1 and and antiM13 binding activity respectively. For S9 bindingactivity (A) 100

3.00 2.50 2.00 1.50 1.00 0.50 0.00 2.00 1.80 1.60 1.40 1.20 1.00 0.80 0.60 0.40 0.20 0.00

Optical Density 405 nm 405 Density Optical Optical Density 405 nm 405 Density Optical

Figure 4. 4 Direct binding of phage by S9 and anti-M13 antibodies (set I) I) (set antibodies anti-M13 and S9 by phage of binding Direct 4 4. Figure with ELISA plates were coated phages at bindin antiM13 For conjugate. IgG-AP anti-mouse adding by detected was activity binding the and was added ug/ml 15 mAb S9 of ul (B), 100 ul of antiM13 1/10,000 was added and the binding activity was detected by adding anti-rabbit IgG. Absorbance at 405 nm 405 at Absorbance IgG. anti-rabbit adding by detected was activity binding the and added was 1/10,000 antiM13 of ul 100 (B), orient wrong the displaying phage a represents bar pink The peptide. S9 original the displaying phage a represents bar blue The coding sequence. The green bar represents a fUSE2 original phage vector. The purple bars represent phage displaying modified S9 modified displaying phage represent bars purple The vector. phage original a fUSE2 bar represents green The sequence. coding A B

173

g activity was measured. was measured. sequences. sequences. ation of the S9 Individual Phages Individual Phages 50 ug/mlfor S9ug/ml,1 and and antiM13 binding activity respectively. For S9 bindingactivity (A) 100 1.60 1.40 1.20 1.00 0.80 0.60 0.40 0.20 0.00

3.00 2.50 2.00 1.50 1.00 0.50 0.00

m n 405 Density Optical Optical Density 405 nm 405 Density Optical

coding sequence. The green bar represents a fUSE2 original phage vector. The purple bars represent phage displaying modified S9 modified displaying phage represent bars purple The vector. phage original a fUSE2 bar represents green The sequence. coding ul of S9 mAb 15 ug/ml was added and the binding activity was detected by adding anti-mouse IgG-AP conjugate. For antiM13 bindin antiM13 For conjugate. IgG-AP anti-mouse adding by detected was activity binding the and was added ug/ml 15 mAb S9 of ul Figure 4. 5 Direct binding of phage by S9 and anti-M13 antibodies (set II) II) (set antibodies anti-M13 and S9 by phage of binding Direct 5 4. Figure with ELISA plates were coated phages at A B nm 405 at Absorbance IgG. anti-rabbit adding by detected was activity binding the and added was 1/10,000 antiM13 of ul 100 (B), orient wrong the displaying phage a represents bar pink The peptide. S9 original the displaying phage a represents bar blue The

174

To confirm these results, phages with increases in binding activity to the S9 mAb and some phages with reduction in binding activity were recultured and retested (Figure

4.6). The data were consistent with the previous data.

A

3.000

2.500

2.000

1.500

1.000

Optical Density 405 nm Density Optical 0.500

0.000

n 2 3 2 2 8 2 9 2 6 1 0 4 .2 .8 2 6 S9 S9 o p- 10 8 1 6 7 1 3 8 1 2 5 1 1 1 1 ti r rp-5 1 150 161 1 1 192 208 2 2 261 278 2 3 318 414 5 5 826 1. 2.10 2.13 2. vector a

orient g n ro w Individual Phages

B

2.500

2.000

1.500

1.000

0.500 Optical Density 405 nm Density Optical

0.000

r n 2 5 3 0 2 1 8 1 1 8 4 6 .2 2 6 S9 S9 to p- 1 8 12 6 68 0 19 6 78 1 1 14 5 2 1 1.8 1 .1 c tio rp- r 1 150 1 1 172 192 2 2 232 2 2 286 3 3 4 520 5 8 1. 2.10 2.13 2 ve ta n rie o g n ro w Individual Phages

Figure 4. 6 Direct Binding of Phage by S9 and Anti-M13 Antibodies To measure binding by S9 and M13 antibodies ELISA plates were coated with phages at 50 ug/ml and 1 ug/ml, respectively. For S9 binding activity (A and B) 100 ul of S9 mAb 15 ug/ml was added and the binding activity was detected by adding anti-mouse IgG-AP conjugate. For antiM13 binding activity (C and D), 100 ul of antiM13 1/10,000 was added and the binding activity was detected by adding anti-rabbit IgG. Absorbance at 405 nm was measured. The blue bar represents a phage displaying the original S9 peptide. The pink bar represents a phage displaying the wrong orientation of the S9 coding sequence. The green bar represents a fuse2 original phage vector. The purple bars represent phage displaying modified S9 sequences. 175

In order to determine the specificity and avidity of binding of phage-displayed

peptides by antibody, a capture ELISA was designed, in which purified phages were

captured by S9 mAb and detected with anti-M13 antibody. ELISA plates were coated

with S9 mAb or irrelevant antibodies, B6.1 (antibody against Candia albicans) and S11

(mAb IgM against Beta-C protein of type I and II GBS). The immobilized antibodies

were incubated with different concentrations of the phage clones. Phages were detected

with antiM13 antibody and the binding activity was measured (Figure 4.7).

4.500

4.000

3.500

3.000

172 2.500 1.2 2.16 S9 2.000 Fuse 2 vector wrong oreintation Optical Density 405 nm Density Optical 1.500

1.000

0.500

0.000 0.000 0.004 0.008 0.016 0.031 0.063 0.125 0.250 B6.1 S11 Individual Phage Concentration (ug/ml)

Figure 4. 7 Capture ELISA Plates were coated with S9, S11 or B6.1mabs. Phage clones 172, 1.2, 2.16, parental S9, original fuse2, and phages with wrong orientation insert at different concentrations were added. After incubating at 4 °C overnight, anti-M13 antibody was added and incubated. The reactions were detected with alkaline phosphatase conjugated anti-rabbit IgG and addition of colorimetric substrate. Absorbance at 405 nm was observed.

176

The ELISA results show that phage-displayed peptides bound to S9 mAb in a

concentration-dependent manner. Phage clone 1.2 from the second round affinity

selection and 2.16 from the third round affinity selection gave stronger reactions than the parental S9 phage, suggesting higher binding avidity by the S9 mAb. fUSE2 vector phage was not recognized by S9 mAb, whereas the phage-displayed wrong orientation S9 insert showed some degree of binding activity. No phage clones were recognized by the irrelevant antibodies B6.1 and S11.

To determine whether peptides displayed as phage fusion protein are mimetic of a

group B streptococcus (GBS) determinant, we tested the ability of phage-displayed

peptides to inhibit the binding of GBS by S9 mAb. We chose those phages with the

greatest direct binding activity for testing. To perform the inhibition experiment, different

concentrations of phages were premixed with mAb and allowed to incubate for 1 h at

room temperature and then loaded onto the plates coated with GBS. The concentration of

S9 mAb used for the inhibition assay was in the middle of the linear portion of the

binding curve (Figure 4.8), which was 1/25,000. The inhibition assay (Figure 4.9)

indicated that phage clone 10 showed highest activity in blocking the binding activity of

S9 mAb to GBS. Phage clone 1.2 was also a good inhibitor. However, inhibition obtained

from both clones was not much different from that obtained from phage displaying the

original S9 peptide. The original fUSE2 phage did not show any inhibition. 177

3.000

2.500

2.000 S3.2A6 1.500 SIII S8.C3 S9 Ascites 1.000

0.500 Optical Density 405 nm

0.000 1/1000 1/5000 1/25000 1/75000 1/125000 Monoclonal Antibody Dilution

3.000 2.500

2.000 S3.1A6 1.500 SIIIV 18.C2 1.000 S3.1B1 0.500

Optical Density 405 nm 0.000

0 0 00 0 00 50 50 500 2 250 2 / /125000 /6 12 1 1 1 /3 1 /156 1 Monoclonal Antibody Dilution

Figure 4. 8 Titration of mAb binding to GBS ELISA assays for mAb interaction with Group B streptococcus (GBS) were performed by coating microtiter wells with GBS. MAbs were two-fold serially diluted and binding was detected with alkaline phosphatase-conjugated anti-Ig and colorimetric substrate p- nitrophenyl phosphate. A405 was determined. 178

0.600

0.500

S9 0.400 rp5 10 0.300 172 219 1.2 0.200 Fuse2 vector Optical density 405 nm density Optical

0.100

0.000 0.001 0.01 0.1 1 10 Phage Concentration (ug/ml)

Figure 4. 9 Inhibition Assay by Phages of S9 mAb Binding to GBS ELISA assays for phage inhibition of mAb interaction with Group B streptococcus (GBS) were performed by coating microtiter wells with GBS. The inhibition assay was performed by preincubation of S9 with different phage concentrations and adding to GBS coated wells. Binding was detected with alkaline phosphatase-conjugated anti-mouse Ig followed by washing and addition of colorimetric substrate p-nitrophenyl phosphate (Sigma Chemical). Absorbance at 405 nm was determined.

Specificity and Avidity of Peptides

In order to ensure that the interaction between the selected phages and S9 mAb is caused by insert sequences, peptides representing sequences from phage clones, which

showed higher binding activity to S9 mAb and the ability to compete with GBS for S9

binding, were synthesized. Binding of S9 mAb to the synthetic peptides was demonstrated by ELISA (Figure 4.10). 179

10 ug/ml 3 ug/ml 1 ug/ml

PBS

WO

1

M

-

2 .

1

rp

2

M -

172

1

M -

72

1

6

1 .

2

rp

2

1

.

1

-

rp

8

.

1

-

rp

2

.

1

-

rp

6

2

8

4

41

1 31 of original S9 DNA sequence.

or with PBS as negative control. S9 mAb was added and or with PBS as negative control.

6

8

2

8 Peptide

27

9

1

2

2

9

1

2

17

0

5

1 ti-mouse IgG-AP conjugate. Absorbance at 405 nm was observed. Note that 82

peptides to S9 mAb peptides

10 tide from the wrong orientation

3

5

-

p r

2

-

rp

2

T -

9

s

1

T

- Binding activities of 9

s

9 S

0.800 0.700 0.600 0.500 0.400 0.300 0.200 0.100 0.000 Optical Density 405 nm 405 Density Optical

Figure 4. 4. 10 Figure 10 ug/ml with peptides at 1, 3 and were coated ELISA plates the binding was detected by adding an WO peptide represents the pep

180

While there is a concentration dependent effect of the S9 peptide binding by S9

mAb, several of the peptides show very good binding even at the lowest concentration

tested, indicating a higher avidity interaction with S9 mab. The peptide representing the

wrong orientation of S9 DNA did not show binding activity to S9 mAb. Other peptides,

which showed strong binding activity when they were displayed on phage particles

(Figure 4.5, 4.6 and 4.7) showed low binding values.

To determine whether peptides are mimetic of the group B streptococcus (GBS)

determinant, we tested the ability of peptides to compete with S9 mAb binding to GBS.

Different concentrations of peptides were premixed with S9, SIIIS8.C3 or IIIIV18.C2

mAbs and allowed to incubate for 1 h at room temperature and then loaded onto plates

coated with GBS. The concentration of mAb used for the inhibition assay was in the

middle of the linear portion of the binding curve (Figure 4.8), which was 1/25,000 for S9,

1/250,000 for SIIIV18.C2 and 1/3,000 for SIIIS8.C3. The inhibition assay (Figure 4.11)

indicated that peptides which had stronger binding activity with S9 mAb when compared

to the original S9 peptide also were better inhibitors of binding of S9 mAb to GBS. There

was no inhibition when peptides, which had lower binding activity with S9 mAb than the

parental S9 peptide, were used as inhibitors. All peptides were better inhibitors of the

binding reaction between GBS and SIIIV18.C2 mAb IgM when compared with the

original S9 peptide (Figure 4.12). None of the peptides showed inhibition between GBS

and SIIIS8.C3 mAb IgG2a (Figure 4.13). The inhibition assay indicated that certain modified S9 peptides can be better mimetic peptides than the original S9 peptide.

181

2.000

1.800

1.600

1.400

1.200

1.000

0.800 OpticalDensity 405 nm

0.600

0.400

0.200

0.000 0.004 0.02 0.1 0.5 2.5 Peptide Concentration (ug/ml)

Figure 4. 11 Inhibition Assay by Peptides of S9 mAb Binding to GBS ELISA assays for peptide inhibition of mAb interaction with GBS were performed by coating microtiter wells with GBS. The inhibition assay was performed by preincubation of the S9 mAb with different peptide concentrations. S9 mAb ascites was diluted to 1/25,000. The mixture was allowed to incubate at room temperature for 1 h before placing in wells. Red lines represent inhibition curves when peptides were used. The blue line represents inhibition curve when S9 original peptide was used. The green line represents inhibition curve when peptide from the wrong orientation S9 DNA sequence was used.

182

1.400

1.200

1.000

0.800

0.600 Optical Density 405 nm 405 Density Optical

0.400

0.200

0.000

0.004 0.02 0.1 0.5 2.5

Peptide Concentration (ug/ml)

Figure 4. 12 Inhibition Assay by Peptides of SIIIV18.C2 mAb Binding to GBS ELISA assays for peptide inhibition of mAb interaction with GBS were performed by coating microtiter wells with GBS. The inhibition assay was performed by preincubation of the mAb with different peptide concentrations. SIIIV18.C2 mAb was diluted to 1/250,000. The mixture was allowed to incubate at room temperature for 1 h before placing in wells. Red lines represent inhibition curves when peptides were used. The blue line represents inhibition curve when S9 original peptide was used. The green line represents inhibition curve when peptide from the wrong orientation S9 DNA sequence was used. 183

0.600

0.500

0.400 405 nm y 0.300 tical Densit tical p O 0.200

0.100

0.000 0.004 0.02 0.1 0.5 2.5 Peptide Concentration (ug/ml)

Figure 4. 13 Inhibition Assay by Peptides of SIIIS8.C3 mAb Binding to GBS ELISA assays for peptide inhibition of mAb interaction with GBS were performed by coating microtiter wells with GBS. The inhibition assay was performed by preincubation of the mAb with different peptide concentrations. SIIIS8.C3 mAb was diluted to 1/3,000. The mixture was allowed to incubate at room temperature for 1 h before placing in wells. Red lines represent inhibition curves when peptides were used. The blue line represents inhibition curve when S9 original peptide was used. The green line represents inhibition curve when peptide from the wrong orientation S9 DNA sequence was used.

184

Specificity of KLH-Peptides Conjugates

To determine whether multivalent peptides were more effective at inhibiting GBS

mab interactions, peptides were conjugated to the carrier protein KLH. We compared the binding of several closely related anti-GBS mAb to different KLH-peptide conjugates by

ELISA (Figure 4.14). At the same protein concentration, KLH-S9 bound to mAbs more strongly than the S9 peptide alone. Compared with KLH-S9 conjugate, KLH-219 bound more strongly to S9 mAb. None of the KLH-peptides bound to S3.1A6, S3.2A6 and

S3.1B1, which detect a conformational carbohydrate determinant.

4.000 3.500 KLH-S9 KLH-rp5 3.000 KLH-10 405 405 nm 2.500 y KLH-172 2.000 KLH-219 1.500 KLH-1.2 S9 1.000 tical Densit

p KLH

O 0.500 GBS 0.000 S3.1A6 S3.1B1 SIII SIIIV S9 S3.2A6 anti-KLH PBS S8.C3 18.C2 ascites Monoclanal Antibody

Figure 4. 14 Binding Activity of KLH-Peptide Conjugates by a Panel of Anti-GBS mAb ELISA plates were coated with KLH-peptide conjugates, S9 peptide, KLH or GBS. Anti- GBS mabs, anti-KLH, or PBS as negative control, were added and the binding activity was detected by adding anti-mouse IgG-AP conjugate. Absorbance at 405 nm was observed. 185

Competitive ELISA s between the KLH-peptides and the S9 mAb was then

performed (Figure 4.15).

1.400

1.200

1.000 KLH-S9 KLH-rp5 0.800 KLH-10 KLH-172 KLH-219 0.600 KLH-1.2

Optical Density 405 nm KLH 0.400 S9

0.200

0.000 0.000 0.001 0.010 0.100 1.000 Concentration (ug/ml)

Figure 4. 15 Inhibition Assay by KLH-Conjugates of S9 mAb Binding to GBS GBS coated plates were blocked with ELISA blocking buffer overnight at 4°C. The inhibition assay was performed by preincubation of the S9 mAb with different KLH- peptide concentrations. The mixture was allowed to incubate at room temperature for 1 h before placing 100 ul of the mixture in wells and incubating 4°C overnight. The plates were washed with ELISA washing buffer, and incubated with 100 ul of alkaline phosphatase-conjugated anti-mouse Ig, followed by washing and addition of colorimetric substrate p-nitrophenyl phosphate. Absorbance at 405 nm was read.

The competitive ELISA indicated that all KLH-peptides and S9 inhibit the

binding of S9 mAb to GBS. KLH-S9 gave greater inhibition of binding than did the unconjugated peptide at higher concentrations, indicating that the multivalent ligand had greater avidity, even though the free peptide gave greater inhibition at lower 186 concentrations, where there were more antigenic molecules and binding was not yet saturated. Other peptide-KLH conjugates behaved as expected compared to the S9-KLH conjugate.

Sequence Analysis of Peptides

Binding activity and sequence of the peptides described in this chapter are summarized in table 4.3 and table 4.4. Table 4.3 displays peptides, which have one amino acid different from the original S9 peptide. The data indicate that changes amino acid at positions 3, 4, 5, 7, 8, 9, 10 resulted in a decrease in binding activity to S9 mAb compared to the original S9 peptide. Changing amino acid at positions 1, 2, 6, 11 and 12 led to increased or unchanged binding activity compared to the original S9 peptide. Table

4.4 shows the information for peptides containing more than one amino acid different from the original S9 peptide. ELISA data analysis was separated into three groups. Group

I represents a population of peptides with changes at positions 7, 8, 9 and 10. Group II represents peptides with changes of positions 7, 8, 9 and 10 combined with changes at other positions. Group III represents peptides with changes outside positions 7, 8, 9 and

10. Changing amino acids in positions 7, 8, 9 and 10 led to a reduction in binding by S9 mAb even though other changes are at positions 1, 2, 6, 11 and or 12, which alone may have increased avidity. Group III indicates that amino acid combination changes in positions outside 7, 8, 9 and 10 reduced or slightly reduced the binding activity by the S9 mAb.

187

Table 4. 3 Summary of Binding Activity and Sequence of Peptides Which Have One Amino Acid Different from the Original S9 Peptide peptide selection Amino Acid Position Binding 1 2 3 4 5 6 7 8 9 10 11 12 Activity S9 Original F D T G A F D P D W P A 47 Random L S 445 Random C S/- 516 Random S S/- 554 Random V S 219 Random H + 315 Random G S 836 Random A S/+ Rp-1.2 Affinity E S/+ 172M1 R+Syn* Y S/- Rp2 Random P S/- 32 Random S - 310 Random A - 712 Random R - 819 Random K - 443 Random A - Rp12 Random V - 39 Random T - 189 Random G S/- 312 Random S - Rp5 Random I + 150 Random C + Rp1.12 Affinity L S/+ 42 Random E - 82 Random H - 112 Random V - 306 Random N - 372 Random Y - 439 Random A - 533 Random G - 22 Random A - 227 Random R - 259 Random L - 379 Random Q - 172M2 R+Syn* T - 28 Random E - 693 Random V - 719 Random A - Rp1.15 Affinity G - 144 Random G - 168 Random R - 311 Random S - 192 Random T S/+ 414 Random S + 520 Random A S/+ Rp1.8 Affinity R S/+ Rp2.10 Affinity Q + Rp2.16 Affinity L +

188

Table 4.3-Continued Summary of Binding Activity and Sequence of Peptides Which Have One Amino Acid Different from the Original S9 Peptide peptide selection Amino Acid Position Binding 1 2 3 4 5 6 7 8 9 10 11 12 Activity S9 Original F D T G A F D P D W P A 10 Random G S 278 Random S S/+ 318 Random P S 820 Random R S 172 M1 and M2 are peptides modified from peptide 172. S, S/+, +, S/- and - represent same, slightly increased, increased, slightly decreased, and decreased binding by S9 mAb compared to the parental peptide.

Table 4. 4 Summary of Binding Activity and Sequence of Peptides Which Have More than One Amino Acid Different from the Original S9 Peptide peptide selection Amino Acid Position Binding 1 2 3 4 5 6 7 8 9 10 11 12 activity S9 Origin F D T G A F D P D W P A Group I 74 Random A L T - 146 Random G A - 545 Random E Q S - 261 Random R R - 850 Random V E L - Group II 113 Random C G - 475 Random G K E - 51 Random G L - 61 Random C K T S - 208 Random I G R S 172 Random Y T - 238 Random G A - 319 Random H T - 390 Random S A - 457 Random H L - 74 Random A L T - 90 Random A C - 156 Random S G - 237 Random A Y - 378 Random M A - 608 Random L N L - 850 Random V E L - 214 Random P G - Rp2.13 Affinity G R S

189

Table 4.4-Continued Summary of Binding Activity and Sequence of Peptides Which Have More than One Amino Acid Different from the Original S9 Peptide peptide selection Amino Acid Position Binding 1 2 3 4 5 6 7 8 9 10 11 12 activity S9 Origin F D T G A F D P D W P A Group III 3 Random A P - 121 Random G L S 164 Random V V - 173 Random D L - 190 Random N S Q - 204 Random A S A - 286 Random L S - 291 Random C V T - 309 Random E E S 332 Random V G S/- 436 Random Z A - 612 Random L P - 624 Random L Q S/- 650 Random C K -

Discussion

The peptide-display phage library technique has served as a powerful system to

identify mimetic peptides. In our lab, this technique has been used successfully to define

a mimotope of type III Streptococcus group B capsular polysaccharide epitope identified by the protective antibody S9. This is designated the S9 peptide (1). Here in this study, the technique was used to clarify the nature of the portion of the peptide bound by antibody and to identify better antigen mimetic peptides. Phage fUSE2 vector, which has been used in our study, originally was developed by Smith lab (237, 373, 375, 377-379) and proved to be successful in identifying epitopes identified by antibodies to African horsesickness virus (380), Hepatitis C virus (381), HIV (382), Bluetongue virus (383), and the rickettsia, Cowdria ruminantium (384). The fUSE2 vector is a type 3 vector, which accepts DNA inserts and encodes fusion protein of the pIII molecule. The foreign 190

peptides are displayed on all five pIII molecules on a virion (377). We decided to choose

the type 3 vectors because the type III vector can display long foreign peptides, whereas

the type 8 vector, which allows insertion of foreign peptides in pVIII molecules, can only

display short foreign peptides (242, 385). To achieve our aims, we created the modified

S9 sublibrary to have approximately one mutation in each inserted original S9 sequence.

Transforming the recombinant DNA into competent cells gave a low efficiency (4.85 x

104) compared to the Puc19 vector control (1011). This resulted in less diversity in the library. However, the need for library diversity was not a priority in this study since the

library need only cover mutations in a 12-mer peptide. The diversity of our library was

further diminished according to the sequencing analysis. Most clones with inserts had an

unmutated version of the peptide encoding gene, indicating that our mutagenesis was less efficient than hoped. Cloning the library into a single cloning site of the fUSE2 vector resulted in two possible orientations giving 40.3% of transformant with foreign DNA inserts in the correct orientation. The library in this study was small in size, but it was large enough to analyze effects of amino acid changes in the 12 amino acid S9 peptide and search for better mimetic peptides based on the original S9 sequence.

Affinity selection with the S9 mAb was used to seek out peptides with higher binding activity. It has been shown that the yield in the first round of selection is very

important (375, 376). If the yield is not high enough in the first round (<1% of the input

phage), clones have a good chance of being lost (373). Even though, the yield in the first

round of the selection was low (5x105), the increasing yield in the second round of our

experiment indicated the success of affinity selection. In the third round of affinity 191 selection, the yield increased by 1 log suggesting that the phage population was saturated with phages displaying high binding activity to S9 mAb. Sequencing analysis of clones from the second and the third round of affinity selection revealed that the S9 original sequence was found most commonly among the phage clones. Only six unique peptide sequences were identified either in the second or the third round of affinity selection. The peptides are RP1.2, RP1.12 and RP1.15 from the first round of selection and RP2.10,

RP2.13 and RP2.16 from the second round of selection. One unique deduced peptide,

RP1.8, was detected from both rounds of the selection (from 2 clones in the second round and one clone in the final). Six out of seven phages selected by affinity selection showed increased or equivalent binding by the S9 mab. One clone showed reduced S9 binding. It has been shown that there is always a background of non-specifically bound phages, especially when low stringency phage elution conditions are used, as we did (373, 386).

Even so, the affinity selection protocol appeared to select for phage with enhanced binding by the mab, i.e. enhanced immunogenicity.

The relationship between peptide sequence and antibody binding is summarized in tables 3.3 and 3.4. The data show that amino acid positions 7, 8, 9, 10 are most important for binding activity by the S9 mAb. Changes in these positions resulted in reductions in binding by the S9 mAb. In position 7 and 9, the original S9 residue is D, which is an acidic amino acid. Even changing the residue from D to other acidic amino acids, E and H, still led to reduced binding activity by the S9 mAb. In position 10, the original residue is W, an aromatic amino acid. Changing this residue also led to reduced binding. We do not know if a change to another aromatic amino acid would have 192 preserved binding. These results are consistent with the data reported by Johnson et al.

(277). These NMR studies of the S9 peptide, as bound by the S9 mAb, indicated that the mAb bound to peptide residues 7 to 10, DPDW (Figure 4.16).

Figure 4. 16 Views of S9 mAb Bound Structure of the Peptide FDTGAFDPDWPA (Peptide S9) (277)

Epitope mapping techniques and sequence analysis from anti-Id and phage- displayed libraries have demonstrated that antibodies bind to an epitope of three to five amino acids within a sequence. The fine specificity of some carbohydrate-binding proteins maps to peptides containing central W/YXY residues (387-392). A protective antibody against Cryptococcus neoformans glucuronoxylomannan selected phage displaying a binding motif of TPXW(M/L)(M/L) (393). Peptide mimicry of the carbohydrate epitope of Mycobacterium leprae is W(T/R)LGPY(V/M) (349). The peptide motif, DPDW from the S9 is different but there are some similarities. There are aromatic

(W) and hydrophobic (W) residues. The acidic residue, D, may also imitate properties of the GBS capsular saccharide epitope, in which sialic acid plays a key role. 193

Our study also found that AA positions 3 to 5, TGA, play an important role in

peptide binding. The mutation in this motif led to a reduction in binding activity by the

S9 mAb. In position 3, the original amino acid residue is T, a non-aromatic amino acid

with a hydroxyl group. Changing to S, which is also a non-aromatic amino acid with

hydroxyl groups, still led to a reduction in binding avidity. In position 4, the original

amino acid residue is G, which is an aliphatic amino acid. Changing to A still led to the

reduction in peptide avidity. In position 5, the original amino acid residue is A, another aliphatic amino acid. Changing to V (peptide Rp12) or to G (peptide 189), also aliphatic amino acids, led to reductions in peptide avidity, although less so with G than V. The hydrophobicity of each amino acid residue may also play an important role: V is very highly hydrophobic whereas the A and G possess less hydrophobicity. Thus the TGA motif may possess hydrophobic properties that are important in antibody binding. This result indicated that the peptide may imitate the hydrophobic properties of the native bacterial capsule.

All data indicated that amino acid residues at positions 1, 11, and 12 are less important in binding to the S9 mAb. But altering positions 2 and 6 can lead to increased antibody binding. No single amino acid alteration in position 1 (F), 11 (P) and 12 (A) had an effect on peptide binding by the S9 mAb. A combination of amino acid alterations in these positions (peptide 332 and 624) led to slightly decreased peptide binding. Amino acid alterations at position 2 led to unchanged or increased binding to the S9 mAb, which depended upon the amino acid properties. In peptide 219, changing the amino acid from

D to H, which is more hydrophobic, resulted in increased binding activity to the S9 mAb. 194

Alteration of position 6 (F), which is aromatic and highly hydrophobic, to non-aromatic

but still highly hydrophobic amino acids (I,C,L) led to an increase in the binding activity

to the S9 mAb. It would be interesting to investigate the result of the amino acid

alteration from a hydrophobic to a non-hydrophobic amino acid in this position.

The results of sequence analysis of peptides obtained randomly, and by avidity

binding to the S9 mAb, were consistent. Five of seven phages from the affinity selection,

have amino acid alterations at positions 2, 6, 11 and/or 12 which resulted in increased

binding to the S9 mAb. Several phages with increased binding to the S9 Ab randomly

selected were not picked by the affinity selection. Several clones were lost during the

affinity selection process, and we did not fully explore the effects of different elution methods

Data from our study indicated that peptides with stronger binding by S9 mAb also

were better inhibitors of the binding of S9 mAb to GBS. When peptides which had lower

binding than the parental peptide were used as inhibitors, no inhibition of the S9 mab-

GBS interaction was seen. The result implied that peptides with higher avidity for the S9

mAb are also better inhibitors of GBS binding to the S9 mAb. Peptides did not bind to

irrelevant antibodies, B6 (antibody against Candida albicans) and S11 (antibody against

Beta-C protein of type I and II GBS), showing that the peptides we isolated were specific

for the S9 mAb.

We next asked whether the peptides were bound by other mabs that react with

GBS type III polysaccharide. Antibody SIIIV18.C2 mAb (specific for the complete type

III CPS) bound a similar array of peptides as S9, whereas an antibody with similar 195 specificity, SIIIS8.C3, binds to none. Antibodies to the conformational type III CPS epitope do not bind the peptides either. .

The results presented here indicated that phage peptide sublibraries can be powerful tools for isolating high affinity peptides and mapping the original peptide mimotope.Immunization with phages displaying these peptides (E. Moran, S.H. Pincus, unpublished) did not lead to significantly higher anti-GBS titers than immunization with the original S9 phage. At best, only a small fraction of the anti-peptide antibody is anti-

GBS. These results have implications for the design of peptide mimotope vaccines to carbohydrate antigens.

196

CONCLUSION

Streptococcus group B, (GBS) is a major cause of neonatal sepsis and meningitis.

Even though the pathogen can be successfully treated with antibiotics, deaths due to infection still occur (100, 342). The main virulence factor of GBS is capsular polysaccharide (CPS), which is anti-phagocytic (94, 394). The antibody against CPS confers protection against GBS infection (170, 194, 346). GBS vaccines using CPS alone were of limited effectiveness. This is likely because CPS of GBS, as a carbohydrate antigen, is a T-independent antigen resulting in low IgG induction and lack of immunologic memory. Conjugation with protein carriers has improved its immunogenicity and conferred protection against GBS infection (192, 195, 199).

To develop a vaccine against GBS, an alternative approach using a phage displayed library was applied to search for a peptide that mimics the CPS of GBS. The S9 peptide, FDTGAFDPDWPA, was identified and shown to be both antigenic and immunogenic. S9-KLH conjugate conferred antibody against S9 peptide and GBS (1).

In this study we continued the focus on GBS vaccine design. The goal of this thesis research is first to enhance immunogenicity of the S9 peptide, second to investigate the structural basis of the peptide mimicry of the GBS type III CPS carbohydrate epitope, and third to identify better mimic peptides.

The standard method to improve immunogenicity of small peptides is by conjugating them to protein carriers such as KLH and tetanus toxoid. In our study, we used an alternative method by presenting the peptide on plant virus coat protein. Plant viruses have been used as vaccine vectors to deliver peptides or small proteins to the 197

immune system (reviewed in reference (395)). There are two approaches to accomplish

these using recombinant techniques. The first strategy is to create replicative chimeric

virus expressing peptide by fusing the foreign protein to the virus coat protein. The

second approach is based on the property of virus coat proteins to self-assemble on

expression in bacteria, such as Escherichia coli or in yeast, such as Saccharomyces

cerevisiae or Pichia pastoris. In our study, Cowpea Chlorotic Mottle Virus (CCMV) was

chosen to carry the S9 peptide. The capsid of CCMV is composed of 180 copies of coat

protein subunits. The CCMV structure has been revealed by X-ray crystallography(317).

Based on the structure and X-ray crystallography, the Young lab has modified coat

protein gene at permissive sites for peptide insertion. Each site was located on the

exposed loop of CCMV coat protein. Nucleic acids encoding the peptide S9 were cloned

into each site in a separate construct. The RNA of the recombinant constructs was in vitro

transcribed and manually inoculated on the natural host, cowpea plant and tobacco. No

infection with chimeric virus was obtained. The result suggested that inserting peptide

into coat protein disrupted virus assembly process. The coat protein of CCMV can self-

assemble in P. pastoris, a heterologous expression system, resulting in the formation of

virus-like particles (VLP). The cassettes of chimeric S9-coat protein genes for each

construct were subcloned into the P. pastoris expression plasmid pPicZa. The S9-CCMV coat protein from each construct can be expressed inside the yeast cells. However, intact recombinant VLP could not be purified from yeast cell debris, while the wild-type VLP can be obtained easily. This data indicates that the insertion of the S9 into the CCMV coat protein might cause virus to fall apart or disrupt viral assembly. We therefore 198

presented the S9 peptide on the CCMV coat protein via chemically conjugating CCMV

and S9 through chemical linkers. Two CCMV mutants were used in the coupling

reaction, salt stable mutant (ssCCMV) and SubE mutant. Because the ssCCMV (K42R) can resist disassembly in 1.0M NaCl, pH 7.5 (319, 341) whereas the wild type CCMV is

only stable at pH4 and lower ionic strength, and the coupling reaction is done at the

higher pH and ionic strength, the salt stable mutant is the appropriate virion for

conjugation. The ssCCMV-S9 conjugation resulted in monomeric intact virions. SubE is

a mutant that cannot assemble virions, and thus serves as a control with the S9 peptide

conjugated to the monomeric coat protein. We next tested if the CCMV-S9 conjugate can

enhance S9 peptide immunogenicity. Mice were immunized, with or without adjuvant,

with different antigens, CCMV, CCMV-S9, SubE-S9, CPMV-S9 (kindly provided from

Dr. John E. Johnson), and KLH-S9. Mice immunized with CPMV-S9, CCMV-S9 and

KLH-S9 produced an antibody response to S9 and GBS without the help of adjuvant.

When adjuvant was used in the immunization, the antibody response was augmented.

SubE-S9 conjugate only induced an antibody response with the help of adjuvant. The

data suggested that organization of antigen is important in immunogenicity. In fact the

anti-S9 antibody response from mice immunized with CPMV-S9 conjugate was higher

than the response from mice immunized with KLH-S9 conjugate. The data suggested that

conjugation of peptide to virus coat protein created repetitive structures expressing the

peptide, leading to enhanced in immunogenicity.

The antigen-specific IgG subclass profile was characterized. Mice immunized

with both virus-S9 conjugates expressed predominant IgG2a, whereas mice immunized 199

with KLH-S9 conjugate produced predominant IgG1, suggesting Th1/Th2 bias with the

different carriers. The cytokine profile analysis confirmed this, showing that splenocytes

primed with CCMV or CCMV-S9 and then stimulated with CCMV or CCMV-S9

produced predominantly IFN-γ, whereas splenocytes primed with KLH-S9 and then

stimulated with KLH-S9 or KLH produced predominantly IL4 and IL10. These data

suggested that immunizing with virus particle expressing S9 provides a Th1 biased

response whereas immunization with protein, KLH-S9 induces a Th2 biased response.

Even though, the data indicates that CCMV is not suitable for use as vaccine

vector because insertion of peptides into the virus coat protein disrupts virus assembly,

this study elucidated immunological issues. The study indicated that organization and the

composition of the carrier for a peptide antigen is important to its immunogenicity.

Virions are particulate antigens, with repetition of the basic epitope, and may also have

accompanying nucleic acids. These differences appear to make them more immunogenic

than carrier proteins, and may also influence the Th1/Th2 orientation of the resulting

immune response. This knowledge could benefit vaccine design.

The second goal of the study was to characterize the structural basis of S9

mimetic antigen. To accomplish this goal, we designed a DNA sublibrary, expressing a

mutated version of the S9 peptide with 3% random nucleic acids at each position of the original S9 nucleotide sequence, creating approximately one mutation in each inserted sequence. By affinity selection, peptides with better binding activity to the S9 mAb were

obtained. By random selection, it was shown that phage clones with amino acid

alterations in positions 3-5 and 7-10 had reduction in S9 antibody binding activity. This 200 data indicates that these residues are involved in antibody binding of the peptide. These findings are consistent with those of an NMR study of antibody binding to the S9 peptide

(277). The characterization of peptides with higher affinity for antibody showed that changing certain amino acids to more hydrophobic amino acid residues increased binding. This suggests that the S9 peptide mimics the hydrophobicity of the native capsular polysaccharide.

In summary, this thesis examined an alternative approach toward GBS vaccine development. Phage display libraries have been used in our study and proven to be a powerful tool to search for peptides that mimic carbohydrate epitopes. The technique also allows us to explore the structural basis of the mimetic peptide providing a better understanding of the structural basis for mimicry epitope. By using CCMV plant virus as a vaccine vector we also show that using particulate antigens influences the immunogenicity of peptide antigens.

Our original goals in this study were to enhance the immunogencity of S9 as a

GBS mimetic and to study the structural basis of the peptide. Our results indicated that we were not able to improve upon the original S9 in terms of eliciting an anti GBS response. However, our studies have provided answers to the structural questions of S9 memetic structure.

201

REFERENCES CITED

1. Pincus, S. H., M. J. Smith, H. J. Jennings, J. B. Burritt, and P. M. Glee. 1998. Peptides that mimic the group B streptococcal type III capsular polysaccharide antigen. Journal of Immunology 160:293-298.

2. Schaechter, M., L. C. Madoff, and B. I. Eisenstein. 1993. Streptococci. In Mechanisms of microbial disease, 2 ed. T. S. Satterfield, ed. Williams& Wilkins. 198-212.

3. Hardie, J. M., and R. A. Whiley. 1997. Classification and overview of the genera Streptococcus and Enterococcus. Society for Applied Bacteriology Symposium Series 26:1S-11S.

4. Jones, G. W., D. B. Clewell, L. G. Charles, and M. M. Vickerman. 1996. Multiple phase variation in haemolytic, adhesive and antigenic properties of Streptococcus gordonii. Microbiology 142 ( Pt 1):181-189.

5. Kim, J. O., and J. N. Weiser. 1998. Association of intrastrain phase variation in quantity of capsular polysaccharide and teichoic acid with the virulence of Streptococcus pneumoniae. Journal of Infectious Diseases 177:368-377.

6. Weiser, J. N. 1998. Phase variation in colony opacity by Streptococcus pneumoniae. Microb Drug Resist 4:129-135.

7. Weiser, J. N., Z. Markiewicz, E. I. Tuomanen, and J. H. Wani. 1996. Relationship between phase variation in colony morphology, intrastrain variation in cell wall physiology, and nasopharyngeal colonization by Streptococcus pneumoniae. Infection and Immunity 64:2240-2245.

8. Pincus, S. H., R. L. Cole, M. R. Wessels, M. D. Corwin, E. Kamanga-Sollo, S. F. Hayes, W. Cieplak, Jr., and J. Swanson. 1992. Group B streptococcal opacity variants. Journal of Bacteriology 174:3739-3749.

9. Wannamaker, L. W. 1983. Streptococcal toxins. Reviews of Infectious Diseases 5 Suppl 4:S723-732.

10. Palmer, M., I. Vulicevic, P. Saweljew, A. Valeva, M. Kehoe, and S. Bhakdi. 1998. Streptolysin O: a proposed model of allosteric interaction between a pore- forming protein and its target lipid bilayer. Biochemistry 37:2378-2383.

202

11. Kawakita, S., T. Takeuchi, J. Inoue, T. Onishi, and Y. Uemura. 1981. Infection of group A streptococcus and antibody response to extracellular antigens. Japanese Circulation Journal 45:1384-1390.

12. Ginsburg, I. 1999. Is streptolysin S of group A streptococci a virulence factor? APMIS 107:1051-1059.

13. Wellstood, S. A. 1987. Rapid, cost-effective identification of group A streptococci and enterococci by pyrrolidonyl-beta-naphthylamide hydrolysis. Journal of Clinical Microbiology 25:1805-1806.

14. Phillips, E. A., J. W. Tapsall, and D. D. Smith. 1980. Rapid tube CAMP test for identification of Streptococcus agalactiae (Lancefield group B). Journal of Clinical Microbiology 12:135-137.

15. Darling, C. L. 1975. Standardization and evaluation of the CAMP reaction for the prompt, presumptive identification of Streptococcus agalactiae (Lancefield group B) in clinical material. Journal of Clinical Microbiology 1:171-174.

16. Mundy, L. S., E. N. Janoff, K. E. Schwebke, C. J. Shanholtzer, and K. E. Willard. 1998. Ambiguity in the identification of Streptococcus pneumoniae. Optochin, bile solubility, quellung, and the AccuProbe DNA probe tests. American Journal of Clinical Pathology 109:55-61.

17. Fertally, S. S., and R. Facklam. 1987. Comparison of physiologic tests used to identify non-beta-hemolytic aerococci, enterococci, and streptococci. Journal of Clinical Microbiology 25:1845-1850.

18. Tao, L., I. C. Sutcliffe, R. R. Russell, and J. J. Ferretti. 1993. Cloning and expression of the multiple sugar metabolism (msm) operon of Streptococcus mutans in heterologous streptococcal hosts. Infection and Immunity 61:1121- 1125.

19. Thomas, T. D., and K. W. Turner. 1981. Carbohydrate Fermentation by Streptococcus cremoris and Streptococcus lactis Growing in Agar Gels. Applied and Environmental Microbiology 41:1289-1294.

20. Efstratiou, A., and W. R. Maxted. 1979. Serological grouping of streptococci by slide agglutination. Journal of Clinical Pathology 32:1228-1233.

21. Herbst, H., D. Lavanchy, and D. G. Braun. 1983. Grouping of haemolytic streptococci by monoclonal antibodies: determinant specificity, cross-reactivity and affinity. Ann Immunol 134D:349-371.

203

22. Lancefield, R. C. 1962. Current knowledge of type-specific M antigens of group A streptococci. Journal of Immunology 89:307-313.

23. Martin, J. M., and M. Green. 2006. Group A streptococcus. Semin Pediatr Infect Dis 17:140-148.

24. Mora, M., G. Bensi, S. Capo, F. Falugi, C. Zingaretti, A. G. Manetti, T. Maggi, A. R. Taddei, G. Grandi, and J. L. Telford. 2005. Group A Streptococcus produce pilus-like structures containing protective antigens and Lancefield T antigens. Proceedings of the National Academy of Sciences of the United States of America 102:15641-15646.

25. Moses, A. E., M. R. Wessels, K. Zalcman, S. Alberti, S. Natanson-Yaron, T. Menes, and E. Hanski. 1997. Relative contributions of hyaluronic acid capsule and M protein to virulence in a mucoid strain of the group A Streptococcus. Infection and Immunity 65:64-71.

26. Hanski, E., and M. Caparon. 1992. Protein F, a fibronectin-binding protein, is an adhesin of the group A streptococcus Streptococcus pyogenes. Proceedings of the National Academy of Sciences of the United States of America 89:6172-6176.

27. Ellen, R. P., and R. J. Gibbons. 1972. M protein-associated adherence of Streptococcus pyogenes to epithelial surfaces: prerequisite for virulence. Infection and Immunity 5:826-830.

28. Courtney, H. S., D. L. Hasty, Y. Li, H. C. Chiang, J. L. Thacker, and J. B. Dale. 1999. Serum opacity factor is a major fibronectin-binding protein and a virulence determinant of M type 2 Streptococcus pyogenes. Molecular Microbiology 32:89- 98.

29. Wessels, M. R., and M. S. Bronze. 1994. Critical role of the group A streptococcal capsule in pharyngeal colonization and infection in mice. Proceedings of the National Academy of Sciences of the United States of America 91:12238-12242.

30. Schrager, H. M., J. G. Rheinwald, and M. R. Wessels. 1996. Hyaluronic acid capsule and the role of streptococcal entry into keratinocytes in invasive skin infection. Journal of Clinical Investigation 98:1954-1958.

31. Dale, J. B., R. G. Washburn, M. B. Marques, and M. R. Wessels. 1996. Hyaluronate capsule and surface M protein in resistance to opsonization of group A streptococci. Infection and Immunity 64:1495-1501.

204

32. Schrager, H. M., S. Alberti, C. Cywes, G. J. Dougherty, and M. R. Wessels. 1998. Hyaluronic acid capsule modulates M protein-mediated adherence and acts as a ligand for attachment of group A Streptococcus to CD44 on human keratinocytes. Journal of Clinical Investigation 101:1708-1716.

33. McKay, F. C., J. D. McArthur, M. L. Sanderson-Smith, S. Gardam, B. J. Currie, K. S. Sriprakash, P. K. Fagan, R. J. Towers, M. R. Batzloff, G. S. Chhatwal, M. Ranson, and M. J. Walker. 2004. Plasminogen binding by group A streptococcal isolates from a region of hyperendemicity for streptococcal skin infection and a high incidence of invasive infection. Infection and Immunity 72:364-370.

34. Wang, H., R. Lottenberg, and M. D. Boyle. 1995. Analysis of the interaction of group A streptococci with fibrinogen, streptokinase and plasminogen. Microbial Pathogenesis 18:153-166.

35. Wang, H., R. Lottenberg, and M. D. Boyle. 1995. A role for fibrinogen in the streptokinase-dependent acquisition of plasmin(ogen) by group A streptococci. Journal of Infectious Diseases 171:85-92.

36. Pancholi, V., and V. A. Fischetti. 1993. Glyceraldehyde-3-phosphate dehydrogenase on the surface of group A streptococci is also an ADP-ribosylating enzyme. Proceedings of the National Academy of Sciences of the United States of America 90:8154-8158.

37. Pancholi, V., and V. A. Fischetti. 1998. alpha-enolase, a novel strong plasmin(ogen) binding protein on the surface of pathogenic streptococci. Journal of Biological Chemistry 273:14503-14515.

38. Pancholi, V., and V. A. Fischetti. 1997. A novel plasminogen/plasmin binding protein on the surface of group A streptococci. Advances in Experimental Medicine and Biology 418:597-599.

39. Stockbauer, K. E., D. Grigsby, X. Pan, Y. X. Fu, L. M. Mejia, A. Cravioto, and J. M. Musser. 1998. Hypervariability generated by natural selection in an extracellular complement-inhibiting protein of serotype M1 strains of group A Streptococcus. Proceedings of the National Academy of Sciences of the United States of America 95:3128-3133.

40. Chiang-Ni, C., C. H. Wang, P. J. Tsai, W. J. Chuang, Y. S. Lin, M. T. Lin, C. C. Liu, and J. J. Wu. 2006. Streptococcal pyrogenic exotoxin B causes mitochondria damage to polymorphonuclear cells preventing phagocytosis of group A streptococcus. Med Microbiol Immunol 195:55-63.

205

41. Ohara-Nemoto, Y., M. Sasaki, M. Kaneko, T. Nemoto, and M. Ota. 1994. Cysteine protease activity of streptococcal pyrogenic exotoxin B. Canadian Journal of Microbiology 40:930-936.

42. Kuo, C. F., J. J. Wu, K. Y. Lin, P. J. Tsai, S. C. Lee, Y. T. Jin, H. Y. Lei, and Y. S. Lin. 1998. Role of streptococcal pyrogenic exotoxin B in the mouse model of group A streptococcal infection. Infection and Immunity 66:3931-3935.

43. Hidalgo-Grass, C., M. Dan-Goor, A. Maly, Y. Eran, L. A. Kwinn, V. Nizet, M. Ravins, J. Jaffe, A. Peyser, A. E. Moses, and E. Hanski. 2004. Effect of a bacterial pheromone peptide on host chemokine degradation in group A streptococcal necrotising soft-tissue infections. Lancet 363:696-703.

44. Matsubara, K., and T. Fukaya. 2007. The role of superantigens of group A Streptococcus and Staphylococcus aureus in Kawasaki disease. Curr Opin Infect Dis 20:298-303.

45. Weiss, K. A., and M. Laverdiere. 1997. Group A Streptococcus invasive infections: a review. Canadian Journal of Surgery 40:18-25.

46. Sriskandan, S., L. Faulkner, and P. Hopkins. 2007. Streptococcus pyogenes: Insight into the function of the streptococcal superantigens. International Journal of Biochemistry and Cell Biology 39:12-19.

47. Beachey, E. H., and W. A. Simpson. 1982. The adherence of group A streptococci to oropharyngeal cells: the lipoteichoic acid adhesin and fibronectin receptor. Infection 10:107-111.

48. Beachey, E. H., and H. S. Courtney. 1987. Bacterial adherence: the attachment of group A streptococci to mucosal surfaces. Reviews of Infectious Diseases 9 Suppl 5:S475-481.

49. Courtney, H. S., J. B. Dale, and D. L. Hasty. 1997. Host cell specific adhesins of group A streptococci. Advances in Experimental Medicine and Biology 418:605- 606.

50. Beachey, E. H., and H. S. Courtney. 1989. Bacterial adherence of group A streptococci to mucosal surfaces. Respiration 55 Suppl 1:33-40.

51. Collin, M., and A. Olsen. 2003. Extracellular enzymes with immunomodulating activities: variations on a theme in Streptococcus pyogenes. Infection and Immunity 71:2983-2992.

206

52. Stevens, D. L. 1997. The toxins of group A streptococcus, the flesh eating bacteria. Immunological Investigations 26:129-150.

53. Kaplan, M. H. 1979. Rheumatic fever, rheumatic heart disease, and the streptococcal connection: the role of streptococcal antigens cross-reactive with heart tissue. Reviews of Infectious Diseases 1:988-986.

54. Stollerman, G. H. 1967. Prospects for a vaccine against group A streptococci: the problem of the immunology of M proteins. Arthritis and Rheumatism 10:245-255.

55. Vashishtha, A., and V. A. Fischetti. 1993. Surface-exposed conserved region of the streptococcal M protein induces antibodies cross-reactive with denatured forms of myosin. Journal of Immunology 150:4693-4701.

56. El-Demellawy, M., R. El-Ridi, N. I. Guirguis, M. Abdel Alim, A. Kotby, and M. Kotb. 1997. Preferential recognition of human myocardial antigens by T lymphocytes from rheumatic heart disease patients. Infection and Immunity 65:2197-2205.

57. Lymbury, R. S., C. Olive, K. A. Powell, M. F. Good, R. G. Hirst, J. T. LaBrooy, and N. Ketheesan. 2003. Induction of autoimmune valvulitis in Lewis rats following immunization with peptides from the conserved region of group A streptococcal M protein. Journal of Autoimmunity 20:211-217.

58. Mollick, J. A., G. G. Miller, J. M. Musser, R. G. Cook, D. Grossman, and R. R. Rich. 1993. A novel superantigen isolated from pathogenic strains of Streptococcus pyogenes with aminoterminal homology to staphylococcal enterotoxins B and C. Journal of Clinical Investigation 92:710-719.

59. Proft, T., P. D. Webb, V. Handley, and J. D. Fraser. 2003. Two novel superantigens found in both group A and group C Streptococcus. Infection and Immunity 71:1361-1369.

60. Henrichsen, J. 1999. Typing of Streptococcus pneumoniae: past, present, and future. American Journal of Medicine 107:50S-54S.

61. Nielsen, S. V., U. B. Sørensen, and J. Henrichsen. 1993. Antibodies against pneumococcal C-polysaccharide are not protective. Microbial Pathogenesis 14:299-305.

62. Winkelstein, J. A. 1981. The role of complement in the host's defense against Streptococcus pneumoniae. Reviews of Infectious Diseases 3:289-298.

207

63. Jarva, H., T. S. Jokiranta, R. Würzner, and S. Meri. 2003. Complement resistance mechanisms of streptococci. Molecular Immunology 40:95-107.

64. Brown, E. J., S. W. Hosea, and M. M. Frank. 1983. The role of antibody and complement in the reticuloendothelial clearance of pneumococci from the bloodstream. Reviews of Infectious Diseases 5 Suppl 4:S797-805.

65. Ren, B., M. A. McCrory, C. Pass, D. C. Bullard, C. M. Ballantyne, Y. Xu, D. E. Briles, and A. J. Szalai. 2004. The virulence function of Streptococcus pneumoniae surface protein A involves inhibition of complement activation and impairment of complement receptor-mediated protection. Journal of Immunology 173:7506-7512.

66. Paton, J. C., P. W. Andrew, G. J. Boulnois, and T. J. Mitchell. 1993. Molecular analysis of the pathogenicity of Streptococcus pneumoniae: the role of pneumococcal proteins. Annual Review of Microbiology 47:89-115.

67. Kostyukova, N. N., M. O. Volkova, V. V. Ivanova, and A. S. Kvetnaya. 1995. A study of pathogenic factors of Streptococcus pneumoniae strains causing meningitis. FEMS Immunology and Medical Microbiology 10:133-137.

68. Jedrzejas, M. J. 2001. Pneumococcal virulence factors: structure and function. Microbiology and Molecular Biology Reviews 65:187-207 first page, table of contents. 69. Rubins, J. B., and E. N. Janoff. 1998. Pneumolysin: a multifunctional pneumococcal virulence factor. Journal of Laboratory and Clinical Medicine 131:21-27.

70. Feldman, C., R. Read, A. Rutman, P. K. Jeffery, A. Brain, V. Lund, T. J. Mitchell, P. W. Andrew, G. J. Boulnois, H. C. Todd, and et al. 1992. The interaction of Streptococcus pneumoniae with intact human respiratory mucosa in vitro. European Respiratory Journal 5:576-583.

71. Cundell, D. R., J. N. Weiser, J. Shen, A. Young, and E. I. Tuomanen. 1995. Relationship between colonial morphology and adherence of Streptococcus pneumoniae. Infection and Immunity 63:757-761.

72. Barocchi, M. A., J. Ries, X. Zogaj, C. Hemsley, B. Albiger, A. Kanth, S. Dahlberg, J. Fernebro, M. Moschioni, V. Masignani, K. Hultenby, A. R. Taddei, K. Beiter, F. Wartha, A. von Euler, A. Covacci, D. W. Holden, S. Normark, R. Rappuoli, and B. Henriques-Normark. 2006. A pneumococcal pilus influences virulence and host inflammatory responses. Proceedings of the National Academy of Sciences of the United States of America 103:2857-2862. 208

73. Tong, H. H., L. M. Fisher, G. M. Kosunick, and T. F. Demaria. 1999. Effect of tumor necrosis factor alpha and interleukin 1-alpha on the adherence of Streptococcus pneumoniae to chinchilla tracheal epithelium. Acta Otolaryngol 119:78-82.

74. Shah, M., R. M. Centor, and M. Jennings. 2007. Severe acute pharyngitis caused by group C streptococcus. Journal of General Internal Medicine 22:272-274.

75. Hashikawa, S., Y. Iinuma, M. Furushita, T. Ohkura, T. Nada, K. Torii, T. Hasegawa, and M. Ohta. 2004. Characterization of group C and G streptococcal strains that cause streptococcal toxic shock syndrome. Journal of Clinical Microbiology 42:186-192.

76. Jankowski, S., K. MacLachlan, J. Stephenson, and P. Thomas. 2002. Necrotising infections and group G streptococcus. Lancet 359:2277.

77. Flahaut, S., A. Hartke, J. C. Giard, A. Benachour, P. Boutibonnes, and Y. Auffray. 1996. Relationship between stress response toward bile salts, acid and heat treatment in Enterococcus faecalis. FEMS Microbiology Letters 138:49-54.

78. Davis, D. R., J. B. McAlpine, C. J. Pazoles, M. K. Talbot, E. A. Alder, C. White, B. M. Jonas, B. E. Murray, G. M. Weinstock, and B. L. Rogers. 2001. Enterococcus faecalis multi-drug resistance transporters: application for antibiotic discovery. Journal of Molecular Microbiology and Biotechnology 3:179-184.

79. Bosley, G. S., R. R. Facklam, and D. Grossman. 1983. Rapid identification of enterococci. Journal of Clinical Microbiology 18:1275-1277.

80. Koranyi, K. I., R. Ray, V. D. Scott, and C. G. Wahoff. 1981. Meningitis caused by a streptococcus of the viridans group. Clinical Pediatrics 20:526.

81. 1996. Prevention of perinatal group B streptococcal disease: a public health perspective. In In Morbid. C. f. D. C. a. Prevention, ed. Morbid. Mortal. Wk. Rep. 1.

82. Slotved, H. C., J. Elliott, T. Thompson, and H. B. Konradsen. 2003. Latex assay for serotyping of group B Streptococcus isolates. Journal of Clinical Microbiology 41:4445-4447.

83. Ke, D., C. Menard, F. J. Picard, M. Boissinot, M. Ouellette, P. H. Roy, and M. G. Bergeron. 2000. Development of conventional and real-time PCR assays for the rapid detection of group B streptococci. Clinical Chemistry 46:324-331.

209

84. Reglier-Poupet, H., G. Quesne, E. Le Theo, M. Dommergues, P. Berche, P. Trieu- Cuot, and C. Poyart. 2005. Prospective evaluation of a real-time PCR assay for detection of group B streptococci in vaginal swabs from pregnant women. European Journal of Clinical Microbiology and Infectious Diseases 24:355-357.

85. Lancefield, R. C., and E. H. Freimer. 1966. Type-specific polysaccharide antigens of group B streptococci. Journal of Hygiene 64:191-203.

86. Michon, F., E. Katzenellenbogen, D. L. Kasper, and H. J. Jennings. 1987. Structure of the complex group-specific polysaccharide of group B Streptococcus. Biochemistry 26:476-486.

87. Michon, F., R. Chalifour, R. Feldman, M. Wessels, D. L. Kasper, A. Gamian, V. Pozsgay, and H. J. Jennings. 1991. The alpha-L-(1----2)-trirhamnopyranoside epitope on the group-specific polysaccharide of group B streptococci. Infection and Immunity 59:1690-1696.

88. Anthony, B. F., N. F. Concepcion, and K. F. Concepcion. 1985. Human antibody to the group-specific polysaccharide of group B Streptococcus. Journal of Infectious Diseases 151:221-226.

89. Jennings, H. J., E. Katzenellenbogen, C. Lugowski, and D. L. Kasper. 1983. Structure of native polysaccharide antigens of type Ia and type Ib group B Streptococcus. Biochemistry 22:1258-1264.

90. Jennings, H. J., K. G. Rosell, E. Katzenellenbogen, and D. L. Kasper. 1983. Structural determination of the capsular polysaccharide antigen of type II group B Streptococcus. Journal of Biological Chemistry 258:1793-1798.

91. Wessels, M. R., V. Pozsgay, D. L. Kasper, and H. J. Jennings. 1987. Structure and immunochemistry of an oligosaccharide repeating unit of the capsular polysaccharide of type III group B Streptococcus. A revised structure for the type III group B streptococcal polysaccharide antigen. Journal of Biological Chemistry 262:8262-8267.

92. Wessels, M. R., W. J. Benedi, H. J. Jennings, F. Michon, J. L. DiFabio, and D. L. Kasper. 1989. Isolation and characterization of type IV group B Streptococcus capsular polysaccharide. Infection and Immunity 57:1089-1094.

93. Wessels, M. R., J. L. DiFabio, V. J. Benedi, D. L. Kasper, F. Michon, J. R. Brisson, J. Jelinkova, and H. J. Jennings. 1991. Structural determination and immunochemical characterization of the type V group B Streptococcus capsular polysaccharide. Journal of Biological Chemistry 266:6714-6719. 210

94. von Hunolstein, C., S. D'Ascenzi, B. Wagner, J. Jelinkova, G. Alfarone, S. Recchia, M. Wagner, and G. Orefici. 1993. Immunochemistry of capsular type polysaccharide and virulence properties of type VI Streptococcus agalactiae (group B streptococci). Infection and Immunity 61:1272-1280.

95. Kogan, G., D. Uhrin, J. R. Brisson, L. C. Paoletti, D. L. Kasper, C. von Hunolstein, G. Orefici, and H. J. Jennings. 1994. Structure of the type VI group B Streptococcus capsular polysaccharide determined by high resolution NMR. Journal of carbohydrate chemistry 13:1071-1078.

96. Kogan, G., J. R. Brisson, D. L. Kasper, C. von Hunolstein, G. Orefici, and H. J. Jennings. 1995. Structural elucidation of the novel type VII group B Streptococcus capsular polysaccharide by high resolution NMR spectroscopy. Carbohydrate Research 277:1-9.

97. Kogan, G., D. Uhrin, J. R. Brisson, L. C. Paoletti, A. E. Blodgett, D. L. Kasper, and H. J. Jennings. 1996. Structural and immunochemical characterization of the type VIII group B Streptococcus capsular polysaccharide. Journal of Biological Chemistry 271:8786-8790.

98. Pincus, S. H., R. L. Cole, E. Kamanga-Sollo, and S. H. Fischer. 1993. Interaction of group B streptococcal opacity variants with the host defense system. Infection and Immunity 61:3761-3768.

99. Gibbs, R. S., S. Schrag, and A. Schuchat. 2004. Perinatal infections due to group B streptococci. Obstetrics and Gynecology 104:1062-1076.

100. Schuchat, A. 1998. Epidemiology of group B streptococcal disease in the United States: shifting paradigms. Clinical Microbiology Reviews 11:497-513.

101. Edwards, M. S., and C. J. Baker. 2005. Group B streptococcal infections in elderly adults. Clinical Infectious Diseases 41:839-847.

102. Farley, M. M. 2001. Group B streptococcal disease in nonpregnant adults. Clinical Infectious Diseases 33:556-561.

103. Munoz, P., A. Llancaqueo, M. Rodriguez-Creixems, T. Pelaez, L. Martin, and E. Bouza. 1997. Group B streptococcus bacteremia in nonpregnant adults. Archives of Internal Medicine 157:213-216.

104. Kotiw, M., G. W. Zhang, G. Daggard, E. Reiss-Levy, J. W. Tapsall, and A. Numa. 2003. Late-onset and recurrent neonatal Group B streptococcal disease associated with breast-milk transmission. Pediatric and Developmental Pathology 6:251-256. 211

105. Clay, L. S. 1996. Group B streptococcus in the perinatal period. A review. Journal of Nurse-Midwifery 41:355-363.

106. Platt, J. S., and W. F. O'Brien. 2003. Group B streptococcus: prevention of early- onset neonatal sepsis. Obstetrical and Gynecological Survey 58:191-196.

107. Munoz, P., T. Coque, M. Rodriguez Creixems, J. C. Bernaldo de Quiros, S. Moreno, and E. Bouza. 1992. Group B Streptococcus: a cause of urinary tract infection in nonpregnant adults. Clinical Infectious Diseases 14:492-496.

108. Ferrieri, P., P. P. Cleary, and A. E. Seeds. 1977. Epidemiology of group-B streptococcal carriage in pregnant women and newborn infants. Journal of Medical Microbiology 10:103-114.

109. Anthony, B. F., D. M. Okada, and C. J. Hobel. 1978. Epidemiology of group B Streptococcus: longitudinal observations during pregnancy. Journal of Infectious Diseases 137:524-530.

110. Anthony, B. F., D. M. Okada, and C. J. Hobel. 1979. Epidemiology of the group B streptococcus: maternal and nosocomial sources for infant acquisitions. Journal of Pediatrics 95:431-436.

111. Gray, B. M., D. G. Pritchard, and H. C. Dillon, Jr. 1989. Seroepidemiology of group B streptococcus type III colonization at delivery. Journal of Infectious Diseases 159:1139-1142.

112. Nagano, Y., N. Nagano, S. Takahashi, K. Murono, K. Fujita, F. Taguchi, and Y. Okuwaki. 1991. Restriction endonuclease digest patterns of chromosomal DNA from group B beta-haemolytic streptococci. Journal of Medical Microbiology 35:297-303.

113. Henneke, P., S. Morath, S. Uematsu, S. Weichert, M. Pfitzenmaier, O. Takeuchi, A. Muller, C. Poyart, S. Akira, R. Berner, G. Teti, A. Geyer, T. Hartung, P. Trieu- Cuot, D. L. Kasper, and D. T. Golenbock. 2005. Role of lipoteichoic acid in the phagocyte response to group B streptococcus. Journal of Immunology 174:6449- 6455.

114. Goldschmidt, J. C., Jr., and C. Panos. 1984. Teichoic acids of Streptococcus agalactiae: chemistry, cytotoxicity, and effect on bacterial adherence to human cells in tissue culture. Infection and Immunity 43:670-677.

212

115. Spellerberg, B., E. Rozdzinski, S. Martin, J. Weber-Heynemann, N. Schnitzler, R. Lutticken, and A. Podbielski. 1999. Lmb, a protein with similarities to the LraI adhesin family, mediates attachment of Streptococcus agalactiae to human laminin. Infection and Immunity 67:871-878.

116. Schubert, A., K. Zakikhany, G. Pietrocola, A. Meinke, P. Speziale, B. J. Eikmanns, and D. J. Reinscheid. 2004. The fibrinogen receptor FbsA promotes adherence of Streptococcus agalactiae to human epithelial cells. Infection and Immunity 72:6197-6205.

117. Schubert, A., K. Zakikhany, M. Schreiner, R. Frank, B. Spellerberg, B. J. Eikmanns, and D. J. Reinscheid. 2002. A fibrinogen receptor from group B Streptococcus interacts with fibrinogen by repetitive units with novel ligand binding sites. Molecular Microbiology 46:557-569.

118. Tamura, G. S., and A. Nittayajarn. 2000. Group B streptococci and other gram- positive cocci bind to cytokeratin 8. Infection and Immunity 68:2129-2134.

119. Wibawan, I. T., C. Lammler, and F. H. Pasaribu. 1992. Role of hydrophobic surface proteins in mediating adherence of group B streptococci to epithelial cells. Journal of General Microbiology 138:1237-1242.

120. Harris, T. O., D. W. Shelver, J. F. Bohnsack, and C. E. Rubens. 2003. A novel streptococcal surface protease promotes virulence, resistance to opsonophagocytosis, and cleavage of human fibrinogen. Journal of Clinical Investigation 111:61-70.

121. Nizet, V., K. S. Kim, M. Stins, M. Jonas, E. Y. Chi, D. Nguyen, and C. E. Rubens. 1997. Invasion of brain microvascular endothelial cells by group B streptococci. Infection and Immunity 65:5074-5081.

122. Rubens, C. E., S. Smith, M. Hulse, E. Y. Chi, and G. van Belle. 1992. Respiratory epithelial cell invasion by group B streptococci. Infection and Immunity 60:5157- 5163.

123. La Penta, D., P. Framson, V. Nizet, and C. Rubens. 1997. Epithelial cell invasion by group B streptococci is important to virulence. Advances in Experimental Medicine and Biology 418:631-634.

124. Winram, S. B., M. Jonas, E. Chi, and C. E. Rubens. 1998. Characterization of group B streptococcal invasion of human chorion and amnion epithelial cells In vitro. Infection and Immunity 66:4932-4941. 213

125. Valentin-Weigand, P., H. Jungnitz, A. Zock, M. Rohde, and G. S. Chhatwal. 1997. Characterization of group B streptococcal invasion in HEp-2 epithelial cells. FEMS Microbiology Letters 147:69-74.

126. Gibson, R. L., M. K. Lee, C. Soderland, E. Y. Chi, and C. E. Rubens. 1993. Group B streptococci invade endothelial cells: type III capsular polysaccharide attenuates invasion. Infection and Immunity 61:478-485.

127. Baron, M. J., G. R. Bolduc, M. B. Goldberg, T. C. Auperin, and L. C. Madoff. 2004. Alpha C protein of group B Streptococcus binds host cell surface glycosaminoglycan and enters cells by an actin-dependent mechanism. Journal of Biological Chemistry 279:24714-24723.

128. Stalhammar-Carlemalm, M., L. Stenberg, and G. Lindahl. 1993. Protein rib: a novel group B streptococcal cell surface protein that confers protective immunity and is expressed by most strains causing invasive infections. Journal of Experimental Medicine 177:1593-1603.

129. Areschoug, T., M. Stalhammar-Carlemalm, C. Larsson, and G. Lindahl. 1999. Group B streptococcal surface proteins as targets for protective antibodies: identification of two novel proteins in strains of serotype V. Infection and Immunity 67:6350-6357.

130. Lachenauer, C. S., and L. C. Madoff. 1996. A protective surface protein from type V group B streptococci shares N-terminal sequence homology with the alpha C protein. Infection and Immunity 64:4255-4260.

131. Adderson, E. E., S. Takahashi, Y. Wang, J. Armstrong, D. V. Miller, and J. F. Bohnsack. 2003. Subtractive hybridization identifies a novel predicted protein mediating epithelial cell invasion by virulent serotype III group B Streptococcus agalactiae. Infection and Immunity 71:6857-6863.

132. Gutekunst, H., B. J. Eikmanns, and D. J. Reinscheid. 2004. The novel fibrinogen- binding protein FbsB promotes Streptococcus agalactiae invasion into epithelial cells. Infection and Immunity 72:3495-3504.

133. Herbert, M. A., C. J. Beveridge, and N. J. Saunders. 2004. Bacterial virulence factors in neonatal sepsis: group B streptococcus. Curr Opin Infect Dis 17:225- 229.

134. Cunningham, M. W. 2000. Pathogenesis of group A streptococcal infections. Clinical Microbiology Reviews 13:470-511. 214

135. Jiang, S. M., M. J. Cieslewicz, D. L. Kasper, and M. R. Wessels. 2005. Regulation of virulence by a two-component system in group B streptococcus. Journal of Bacteriology 187:1105-1113.

136. Hulse, M. L., S. Smith, E. Y. Chi, A. Pham, and C. E. Rubens. 1993. Effect of type III group B streptococcal capsular polysaccharide on invasion of respiratory epithelial cells. Infection and Immunity 61:4835-4841.

137. Gibson, R. L., C. Soderland, W. R. Henderson, Jr., E. Y. Chi, and C. E. Rubens. 1995. Group B streptococci (GBS) injure lung endothelium in vitro: GBS invasion and GBS-induced eicosanoid production is greater with microvascular than with pulmonary artery cells. Infection and Immunity 63:271-279.

138. Johri, A. K., J. Padilla, G. Malin, and L. C. Paoletti. 2003. Oxygen regulates invasiveness and virulence of group B streptococcus. Infection and Immunity 71:6707-6711.

139. Nizet, V., R. L. Gibson, E. Y. Chi, P. E. Framson, M. Hulse, and C. E. Rubens. 1996. Group B streptococcal beta-hemolysin expression is associated with injury of lung epithelial cells. Infection and Immunity 64:3818-3826.

140. Jackson, R. J., M. L. Dao, and D. V. Lim. 1994. Cell-associated collagenolytic activity by group B streptococci. Infection and Immunity 62:5647-5651.

141. Pritchard, D. G., J. O. Trent, X. Li, P. Zhang, M. L. Egan, and J. R. Baker. 2000. Characterization of the active site of group B streptococcal hyaluronan lyase. Proteins 40:126-134.

142. Baker, J. R., and D. G. Pritchard. 2000. Action pattern and substrate specificity of the hyaluronan lyase from group B streptococci. Biochemical Journal 348 Pt 2:465-471.

143. Edwards, M. S., D. L. Kasper, H. J. Jennings, C. J. Baker, and A. Nicholson- Weller. 1982. Capsular sialic acid prevents activation of the alternative complement pathway by type III, group B streptococci. Journal of Immunology 128:1278-1283.

144. Lewis, A. L., V. Nizet, and A. Varki. 2004. Discovery and characterization of sialic acid O-acetylation in group B Streptococcus. Proceedings of the National Academy of Sciences of the United States of America 101:11123-11128.

145. Takahashi, S., Y. Aoyagi, E. E. Adderson, Y. Okuwaki, and J. F. Bohnsack. 1999. Capsular sialic acid limits C5a production on type III group B streptococci. Infection and Immunity 67:1866-1870. 215

146. Jerlstrom, P. G., S. R. Talay, P. Valentin-Weigand, K. N. Timmis, and G. S. Chhatwal. 1996. Identification of an immunoglobulin A binding motif located in the beta-antigen of the c protein complex of group B streptococci. Infection and Immunity 64:2787-2793.

147. Jarva, H., J. Hellwage, T. S. Jokiranta, M. J. Lehtinen, P. F. Zipfel, and S. Meri. 2004. The group B streptococcal beta and pneumococcal Hic proteins are structurally related immune evasion molecules that bind the complement inhibitor factor H in an analogous fashion. Journal of Immunology 172:3111-3118.

148. Poyart, C., E. Pellegrini, M. Marceau, M. Baptista, F. Jaubert, M. C. Lamy, and P. Trieu-Cuot. 2003. Attenuated virulence of Streptococcus agalactiae deficient in D-alanyl-lipoteichoic acid is due to an increased susceptibility to defensins and phagocytic cells. Molecular Microbiology 49:1615-1625.

149. Poyart, C., E. Pellegrini, O. Gaillot, C. Boumaila, M. Baptista, and P. Trieu-Cuot. 2001. Contribution of Mn-cofactored superoxide dismutase (SodA) to the virulence of Streptococcus agalactiae. Infection and Immunity 69:5098-5106.

150. Fettucciari, K., E. Rosati, L. Scaringi, P. Cornacchione, G. Migliorati, R. Sabatini, I. Fetriconi, R. Rossi, and P. Marconi. 2000. Group B Streptococcus induces apoptosis in macrophages. Journal of Immunology 165:3923-3933.

151. Fettucciari, K., I. Fetriconi, R. Mannucci, I. Nicoletti, A. Bartoli, S. Coaccioli, and P. Marconi. 2006. Group B Streptococcus induces macrophage apoptosis by calpain activation. Journal of Immunology 176:7542-7556.

152. Ulett, G. C., and E. E. Adderson. 2005. Nitric oxide is a key determinant of group B streptococcus-induced murine macrophage apoptosis. Journal of Infectious Diseases 191:1761-1770.

153. Kwak, D. J., N. H. Augustine, W. G. Borges, J. L. Joyner, W. F. Green, and H. R. Hill. 2000. Intracellular and extracellular cytokine production by human mixed mononuclear cells in response to group B streptococci. Infection and Immunity 68:320-327.

154. Rosati, E., K. Fettucciari, L. Scaringi, P. Cornacchione, R. Sabatini, L. Mezzasoma, R. Rossi, and P. Marconi. 1998. Cytokine response to group B streptococcus infection in mice. Scandinavian Journal of Immunology 47:314- 323.

216

155. Vallejo, J. G., P. Knuefermann, D. L. Mann, and N. Sivasubramanian. 2000. Group B Streptococcus induces TNF-alpha gene expression and activation of the transcription factors NF-kappa B and activator protein-1 in human cord blood monocytes. Journal of Immunology 165:419-425.

156. Levy, O., R. M. Jean-Jacques, C. Cywes, R. B. Sisson, K. A. Zarember, P. J. Godowski, J. L. Christianson, H. K. Guttormsen, M. C. Carroll, A. Nicholson- Weller, and M. R. Wessels. 2003. Critical role of the complement system in group B streptococcus-induced tumor necrosis factor alpha release. Infection and Immunity 71:6344-6353.

157. Vallejo, J. G., C. J. Baker, and M. S. Edwards. 1996. Roles of the bacterial cell wall and capsule in induction of tumor necrosis factor alpha by type III group B streptococci. Infection and Immunity 64:5042-5046.

158. Medvedev, A. E., T. Flo, R. R. Ingalls, D. T. Golenbock, G. Teti, S. N. Vogel, and T. Espevik. 1998. Involvement of CD14 and complement receptors CR3 and CR4 in nuclear factor-kappaB activation and TNF production induced by lipopolysaccharide and group B streptococcal cell walls. Journal of Immunology 160:4535-4542.

159. Goodrum, K. J., L. L. McCormick, and B. Schneider. 1994. Group B streptococcus-induced nitric oxide production in murine macrophages is CR3 (CD11b/CD18) dependent. Infection and Immunity 62:3102-3107.

160. Maloney, C. G., S. D. Thompson, H. R. Hill, J. F. Bohnsack, T. M. McIntyre, and G. A. Zimmerman. 2000. Induction of cyclooxygenase-2 by human monocytes exposed to group B streptococci. Journal of Leukocyte Biology 67:615-621.

161. Mancuso, G., V. Cusumano, F. Genovese, M. Gambuzza, C. Beninati, and G. Teti. 1997. Role of interleukin 12 in experimental neonatal sepsis caused by group B streptococci. Infection and Immunity 65:3731-3735.

162. Schrag, S. J., R. Gorwitz, K. Fultz-Butts, and A. Schuchat. 2002. Prevention of perinatal group B streptococcal disease: revised guidelines. C. f. D. C. a. Prevention, ed. Morbid. Mortal. Wk. Rep. 1.

163. Logsdon, B. A., and D. T. Casto. 1997. Prevention of group B Streptococcus infection in neonates. Annals of Pharmacotherapy 31:897-906.

164. Alkalay, A. L., P. A. Brunell, J. S. Greenspon, and J. J. Pomerance. 1996. Management of neonates born to mothers with group B streptococcus colonization. Journal of Perinatology 16:470-477. 217

165. Schrag, S. J., S. Zywicki, M. M. Farley, A. L. Reingold, L. H. Harrison, L. B. Lefkowitz, J. L. Hadler, R. Danila, P. R. Cieslak, and A. Schuchat. 2000. Group B streptococcal disease in the era of intrapartum antibiotic prophylaxis. New England Journal of Medicine 342:15-20.

166. Yang, Y. J., C. C. Liu, and S. M. Wang. 1998. Group B streptococcal infections in children: the changing spectrum of infections in infants. Journal of Microbiology, Immunology, and Infection 31:107-112.

167. Heelan, J. S., M. E. Hasenbein, and A. J. McAdam. 2004. Resistance of group B streptococcus to selected antibiotics, including erythromycin and clindamycin. Journal of Clinical Microbiology 42:1263-1264.

168. Baker, C. J., and D. L. Kasper. 1976. Correlation of maternal antibody deficiency with susceptibility to neonatal group B streptococcal infection. New England Journal of Medicine 294:753-756.

169. Baker, C. J., M. A. Rench, and D. L. Kasper. 1990. Response to type III polysaccharide in women whose infants have had invasive group B streptococcal infection. New England Journal of Medicine 322:1857-1860.

170. Lancefield, R. C., M. McCarty, and W. N. Everly. 1975. Multiple mouse- protective antibodies directed against group B streptococci. Special reference to antibodies effective against protein antigens. Journal of Experimental Medicine 142:165-179.

171. Lancefield, R. C. 1938. Two serological types of group B hemolytic streptococci with relatd, but not idntical type specific substances. Journal of Experimental Medicine 67:25.

172. Pincus, S. H., A. O. Shigeoka, A. A. Moe, L. P. Ewing, and H. R. Hill. 1988. Protective efficacy of IgM monoclonal antibodies in experimental group B streptococcal infection is a function of antibody avidity. Journal of Immunology 140:2779-2785.

173. Shigeoka, A. O., S. H. Pincus, N. S. Rote, D. G. Pritchard, J. I. Santos, and H. R. Hill. 1985. Monoclonal antibody preparations for immunotherapy of experimental GBS infection. Antibiotics and Chemotherapy 35:254-266.

174. Shigeoka, A. O., S. H. Pincus, N. S. Rote, and H. R. Hill. 1984. Protective efficacy of hybridoma type-specific antibody against experimental infection with group-B Streptococcus. Journal of Infectious Diseases 149:363-372. 218

175. Egan, M. L., D. G. Pritchard, H. C. Dillon, Jr., and B. M. Gray. 1983. Protection of mice from experimental infection with type III group B Streptococcus using monoclonal antibodies. Journal of Experimental Medicine 158:1006-1011.

176. Raff, H. V., P. J. Siscoe, E. A. Wolff, G. Maloney, and W. Shuford. 1988. Human monoclonal antibodies to group B streptococcus. Reactivity and in vivo protection against multiple serotypes. Journal of Experimental Medicine 168:905-917.

177. Madoff, L. C., J. L. Michel, and D. L. Kasper. 1991. A monoclonal antibody identifies a protective C-protein alpha-antigen epitope in group B streptococci. Infection and Immunity 59:204-210.

178. Christensen, K. K., P. Christensen, G. Duc, W. H. Hitzig, V. Linden, B. Muller, and R. A. Seger. 1984. Human IgG antibodies to carbohydrate and protein antigens in mouse protection tests with group B streptococci. Pediatric Research 18:478-482.

179. Rubens, C. E., M. R. Wessels, L. M. Heggen, and D. L. Kasper. 1987. Transposon mutagenesis of type III group B Streptococcus: correlation of capsule expression with virulence. Proceedings of the National Academy of Sciences of the United States of America 84:7208-7212.

180. Marques, M. B., D. L. Kasper, M. K. Pangburn, and M. R. Wessels. 1992. Prevention of C3 deposition by capsular polysaccharide is a virulence mechanism of type III group B streptococci. Infection and Immunity 60:3986-3993.

181. Campbell, J. R., C. J. Baker, T. G. Metzger, and M. S. Edwards. 1988. Functional activity of class-specific antibodies to type III, group B streptococcus. Pediatric Research 23:31-34.

182. Edwards, M. S., C. J. Baker, and D. L. Kasper. 1979. Opsonic specificity of human antibody to the type III polysaccharide of group B Streptococcus. Journal of Infectious Diseases 140:1004-1008.

183. Edwards, M. S., A. Nicholson-Weller, C. J. Baker, and D. L. Kasper. 1980. The role of specific antibody in alternative complement pathway-mediated opsonophagocytosis of type III, group B Streptococcus. Journal of Experimental Medicine 151:1275-1287.

184. Gravekamp, C., D. L. Kasper, J. L. Michel, D. E. Kling, V. Carey, and L. C. Madoff. 1997. Immunogenicity and protective efficacy of the alpha C protein of group B streptococci are inversely related to the number of repeats. Infection and Immunity 65:5216-5221. 219

185. Gravekamp, C., D. L. Kasper, L. C. Paoletti, and L. C. Madoff. 1999. Alpha C protein as a carrier for type III capsular polysaccharide and as a protective protein in group B streptococcal vaccines. Infection and Immunity 67:2491-2496.

186. Cheng, Q., B. Carlson, S. Pillai, R. Eby, L. Edwards, S. B. Olmsted, and P. Cleary. 2001. Antibody against surface-bound C5a peptidase is opsonic and initiates macrophage killing of group B streptococci. Infection and Immunity 69:2302-2308.

187. Brodeur, B. R., M. Boyer, I. Charlebois, J. Hamel, F. Couture, C. R. Rioux, and D. Martin. 2000. Identification of group B streptococcal Sip protein, which elicits cross-protective immunity. Infection and Immunity 68:5610-5618.

188. Moyo, S. R., J. A. Maeland, and R. V. Lyng. 2001. The putative R1 protein of Streptococcus agalactiae as serotype marker and target of protective antibodies. APMIS 109:842-848.

189. Erdogan, S., P. K. Fagan, S. R. Talay, M. Rohde, P. Ferrieri, A. E. Flores, C. A. Guzman, M. J. Walker, and G. S. Chhatwal. 2002. Molecular analysis of group B protective surface protein, a new cell surface protective antigen of group B streptococci. Infection and Immunity 70:803-811.

190. Baker, C. J., M. S. Edwards, and D. L. Kasper. 1978. Immunogenicity of polysaccharides from type III, group B Streptococcus. Journal of Clinical Investigation 61:1107-1110.

191. Baker, C. J., M. A. Rench, M. S. Edwards, R. J. Carpenter, B. M. Hays, and D. L. Kasper. 1988. Immunization of pregnant women with a polysaccharide vaccine of group B streptococcus. New England Journal of Medicine 319:1180-1185.

192. Lagergard, T., J. Shiloach, J. B. Robbins, and R. Schneerson. 1990. Synthesis and immunological properties of conjugates composed of group B streptococcus type III capsular polysaccharide covalently bound to tetanus toxoid. Infection and Immunity 58:687-694.

193. Wessels, M. R., L. C. Paoletti, D. L. Kasper, J. L. DiFabio, F. Michon, K. Holme, and H. J. Jennings. 1990. Immunogenicity in animals of a polysaccharide-protein conjugate vaccine against type III group B Streptococcus. Journal of Clinical Investigation 86:1428-1433.

194. Marques, M. B., D. L. Kasper, A. Shroff, F. Michon, H. J. Jennings, and M. R. Wessels. 1994. Functional activity of antibodies to the group B polysaccharide of group B streptococci elicited by a polysaccharide-protein conjugate vaccine. Infection and Immunity 62:1593-1599. 220

195. Kasper, D. L., L. C. Paoletti, M. R. Wessels, H. K. Guttormsen, V. J. Carey, H. J. Jennings, and C. J. Baker. 1996. Immune response to type III group B streptococcal polysaccharide-tetanus toxoid conjugate vaccine. Journal of Clinical Investigation 98:2308-2314.

196. Paoletti, L. C., R. C. Kennedy, T. C. Chanh, and D. L. Kasper. 1996. Immunogenicity of group B Streptococcus type III polysaccharide-tetanus toxoid vaccine in baboons. Infection and Immunity 64:677-679.

197. Baker, C. J., M. A. Rench, M. Fernandez, L. C. Paoletti, D. L. Kasper, and M. S. Edwards. 2003. Safety and immunogenicity of a bivalent group B streptococcal conjugate vaccine for serotypes II and III. Journal of Infectious Diseases 188:66- 73.

198. Paoletti, L. C., M. R. Wessels, A. K. Rodewald, A. A. Shroff, H. J. Jennings, and D. L. Kasper. 1994. Neonatal mouse protection against infection with multiple group B streptococcal (GBS) serotypes by maternal immunization with a tetravalent GBS polysaccharide-tetanus toxoid conjugate vaccine. Infection and Immunity 62:3236-3243.

199. Shen, X., T. Lagergard, Y. Yang, M. Lindblad, M. Fredriksson, and J. Holmgren. 2000. Systemic and mucosal immune responses in mice after mucosal immunization with group B streptococcus type III capsular polysaccharide- cholera toxin B subunit conjugate vaccine. Infection and Immunity 68:5749-5755.

200. Shen, X., T. Lagergard, Y. Yang, M. Lindblad, M. Fredriksson, and J. Holmgren. 2001. Group B Streptococcus capsular polysaccharide-cholera toxin B subunit conjugate vaccines prepared by different methods for intranasal immunization. Infection and Immunity 69:297-306.

201. Shen, X., T. Lagergard, Y. Yang, M. Lindblad, M. Fredriksson, and J. Holmgren. 2000. Preparation and preclinical evaluation of experimental group B streptococcus type III polysaccharide-cholera toxin B subunit conjugate vaccine for intranasal immunization. Vaccine 19:850-861.

202. Larsson, C., M. Stalhammar-Carlemalm, and G. Lindahl. 1996. Experimental vaccination against group B streptococcus, an encapsulated bacterium, with highly purified preparations of cell surface proteins Rib and alpha. Infection and Immunity 64:3518-3523.

203. Martin, D., S. Rioux, E. Gagnon, M. Boyer, J. Hamel, N. Charland, and B. R. Brodeur. 2002. Protection from group B streptococcal infection in neonatal mice by maternal immunization with recombinant Sip protein. Infection and Immunity 70:4897-4901. 221

204. Cheng, Q., S. Debol, H. Lam, R. Eby, L. Edwards, Y. Matsuka, S. B. Olmsted, and P. P. Cleary. 2002. Immunization with C5a peptidase or peptidase-type III polysaccharide conjugate vaccines enhances clearance of group B Streptococci from lungs of infected mice. Infection and Immunity 70:6409-6415.

205. Kuby, J. 1997. Immunology. W.H. Freeman, New York.

206. Hjelm, F., F. Carlsson, A. Getahun, and B. Heyman. 2006. Antibody-mediated regulation of the immune response. Scand J Immunol 64:177-184.

207. Bhat, T. N., G. A. Bentley, G. Boulot, M. I. Greene, D. Tello, W. Dall'Acqua, H. Souchon, F. P. Schwarz, R. A. Mariuzza, and R. J. Poljak. 1994. Bound water molecules and conformational stabilization help mediate an antigen-antibody association. Proc Natl Acad Sci U S A 91:1089-1093.

208. Braden, B. C., H. Souchon, J. L. Eisele, G. A. Bentley, T. N. Bhat, J. Navaza, and R. J. Poljak. 1994. Three-dimensional structures of the free and the antigen- complexed Fab from monoclonal anti-lysozyme antibody D44.1. J Mol Biol 243:767-781.

209. Padlan, E. A., E. W. Silverton, S. Sheriff, G. H. Cohen, S. J. Smith-Gill, and D. R. Davies. 1989. Structure of an antibody-antigen complex: crystal structure of the HyHEL-10 Fab-lysozyme complex. Proc Natl Acad Sci U S A 86:5938-5942.

210. Sheriff, S., E. W. Silverton, E. A. Padlan, G. H. Cohen, S. J. Smith-Gill, B. C. Finzel, and D. R. Davies. 1987. Three-dimensional structure of an antibody- antigen complex. Proc Natl Acad Sci U S A 84:8075-8079.

211. Colman, P. M., W. G. Laver, J. N. Varghese, A. T. Baker, P. A. Tulloch, G. M. Air, and R. G. Webster. 1987. Three-dimensional structure of a complex of antibody with influenza virus neuraminidase. Nature 326:358-363.

212. Tulip, W. R., J. N. Varghese, W. G. Laver, R. G. Webster, and P. M. Colman. 1992. Refined crystal structure of the influenza virus N9 neuraminidase-NC41 Fab complex. J Mol Biol 227:122-148.

213. Worobec, R. B., J. H. Wallace, and C. G. Huggins. 1972. Angiotensin-antibody interaction. II. Thermodynamic and activation parameters. Immunochemistry 9:239-251.

214. Garcia, K. C., P. M. Ronco, P. J. Verroust, A. T. Brunger, and L. M. Amzel. 1992. Three-dimensional structure of an angiotensin II-Fab complex at 3 A: hormone recognition by an anti-idiotypic antibody. Science 257:502-507. 222

215. Bentley, G. A., G. Boulot, M. M. Riottot, and R. J. Poljak. 1990. Three- dimensional structure of an idiotope-anti-idiotope complex. Nature 348:254-257.

216. Couraud, P. O. 1986. Structural analysis of the epitopes recognized by monoclonal antibodies to angiotensin II. J Immunol 136:3365-3370.

217. Metzger, D. W., A. Miller, and E. E. Sercarz. 1980. Sharing of an idiotypic marker by monoclonal antibodies specific for distinct regions of hen lysozyme. Nature 287:540-542.

218. Garcia, K. C., S. V. Desiderio, P. M. Ronco, P. J. Verroust, and L. M. Amzel. 1992. Recognition of angiotensin II: antibodies at different levels of an idiotypic network are superimposable. Science 257:528-531.

219. Budisavljevic, M., P. M. Ronco, and P. J. Verroust. 1990. Angiotensin II (AII)- related idiotypic network. III. Comparative analysis of idiotopes and paratopes borne by monoclonal antibodies raised against AII (AB1) and its internal image (AB3). J Immunol 145:1440-1449.

220. Garcia, K. C., P. Ronco, P. J. Verroust, and L. M. Amzel. 1989. Crystallization and preliminary X-ray diffraction data of an anti-angiotensin II Fab and of the peptide-Fab complex. J Biol Chem 264:20463-20466.

221. Liao, J., K. G. Nickerson, S. Bystricky, J. B. Robbins, R. Schneerson, S. C. Szu, and E. A. Kabat. 1995. Characterization of a human monoclonal immunoglobulin M (IgM) antibody (IgMBEN) specific for Vi capsular polysaccharide of Salmonella typhi. Infect Immun 63:4429-4432.

222. Gawinowicz, M. A., G. Merlini, S. Birken, E. F. Osserman, and E. A. Kabat. 1991. Amino acid sequence of the FV region of a human monoclonal IgM (NOV) with specificity for the capsular polysaccharide of the group B meningococcus and of Escherichia coli K1, which cross-reacts with polynucleotides and with denatured DNA. J Immunol 147:915-920.

223. Kabat, E. A., J. Liao, E. F. Osserman, A. Gamian, F. Michon, and H. J. Jennings. 1988. The epitope associated with the binding of the capsular polysaccharide of the group B meningococcus and of Escherichia coli K1 to a human monoclonal macroglobulin, IgMNOV. J Exp Med 168:699-711.

223

224. Kabat, E. A., K. G. Nickerson, J. Liao, L. Grossbard, E. F. Osserman, E. Glickman, L. Chess, J. B. Robbins, R. Schneerson, and Y. H. Yang. 1986. A human monoclonal macroglobulin with specificity for alpha(2----8)-linked poly- N-acetyl neuraminic acid, the capsular polysaccharide of group B meningococci and Escherichia coli K1, which crossreacts with polynucleotides and with denatured DNA. J Exp Med 164:642-654.

225. Naparstek, Y., D. Duggan, A. Schattner, M. P. Madaio, F. Goni, B. Frangione, B. D. Stollar, E. A. Kabat, and R. S. Schwartz. 1985. Immunochemical similarities between monoclonal antibacterial Waldenstrom's macroglobulins and monoclonal anti-DNA lupus autoantibodies. J Exp Med 161:1525-1538.

226. Kabat, E. A., and A. E. Bezer. 1958. The effect of variation in molecular weight on the antigenicity of dextran in man. Arch Biochem Biophys 78:306-318.

227. Kabat, E. A. 1966. The nature of an antigenic determinant. J Immunol 97:1-11.

228. Lai, E., E. A. Kabat, and L. Mobraaten. 1985. Genetic and nongenetic control of the immune response of mice to a synthetic glycolipid, stearylisomaltotetraose. Cell Immunol 92:172-183.

229. Geysen, H. M., S. J. Rodda, and T. J. Mason. 1986. A priori delineation of a peptide which mimics a discontinuous antigenic determinant. Mol Immunol 23:709-715.

230. Geysen, H. M., R. H. Meloen, and S. J. Barteling. 1984. Use of peptide synthesis to probe viral antigens for epitopes to a resolution of a single amino acid. Proc Natl Acad Sci U S A 81:3998-4002.

231. Jerne, N. K. 1974. Towards a network theory of the immune system. Ann Immunol (Paris) 125C:373-389.

232. Herlyn, D., R. Somasundaram, W. Li, and H. Maruyama. 1996. Anti-idiotype cancer vaccines: past and future. Cancer Immunol Immunother 43:65-76.

233. Schreiber, J. R., G. B. Pier, M. Grout, K. Nixon, and M. Patawaran. 1991. Induction of opsonic antibodies to Pseudomonas aeruginosa mucoid exopolysaccharide by an anti-idiotypic monoclonal antibody. J Infect Dis 164:507-514.

234. Monafo, W. J., N. S. Greenspan, J. A. Cebra-Thomas, and J. M. Davie. 1987. Modulation of the murine immune response to streptococcal group A carbohydrate by immunization with monoclonal anti-idiotope. J Immunol 139:2702-2707. 224

235. McNamara, M. K., R. E. Ward, and H. Kohler. 1984. Monoclonal idiotope vaccine against Streptococcus pneumoniae infection. Science 226:1325-1326.

236. Burritt, J. B., C. W. Bond, K. W. Doss, and A. J. Jesaitis. 1996. Filamentous phage display of oligopeptide libraries. Analytical Biochemistry 238:1-13.

237. Smith, G. P., and V. A. Petrenko. 1997. Phage Display. Chemical Reviews 97:391-410.

238. Smith, G. P. 1985. Filamentous fusion phage: novel expression vectors that display cloned antigens on the virion surface. Science 228:1315-1317.

239. Cabilly, S. 1999. The basic structure of filamentous phage and its use in the display of combinatorial peptide libraries. Molecular Biotechnology 12:143-148.

240. Greenwood, J., A. E. Willis, and R. N. Perham. 1991. Multiple display of foreign peptides on a filamentous bacteriophage. Peptides from Plasmodium falciparum circumsporozoite protein as antigens. Journal of Molecular Biology 220:821-827.

241. Russel, M. 1993. Protein-protein interactions during filamentous phage assembly. Journal of Molecular Biology 231:689-697.

242. Kishchenko, G., H. Batliwala, and L. Makowski. 1994. Structure of a foreign peptide displayed on the surface of bacteriophage M13. Journal of Molecular Biology 241:208-213.

243. Makowski, L. 1993. Structural constraints on the display of foreign peptides on filamentous bacteriophages. Gene 128:5-11.

244. Riechmann, L., and P. Holliger. 1997. The C-terminal domain of TolA is the coreceptor for filamentous phage infection of E. coli. Cell 90:351-360.

245. Murai, H., S. Hara, T. Ikenaka, A. Goto, M. Arai, and S. Murao. 1985. Amino acid sequence of protein alpha-amylase inhibitor from Streptomyces griseosporeus YM-25. Journal of Biochemistry 97:1129-1133.

246. Vertesy, L., V. Oeding, R. Bender, K. Zepf, and G. Nesemann. 1984. Tendamistat (HOE 467), a tight-binding alpha-amylase inhibitor from Streptomyces tendae 4158. Isolation, biochemical properties. European Journal of Biochemistry 141:505-512.

225

247. Oldenburg, K. R., D. Loganathan, I. J. Goldstein, P. G. Schultz, and M. A. Gallop. 1992. Peptide ligands for a sugar-binding protein isolated from a random peptide library. Proceedings of the National Academy of Sciences of the United States of America 89:5393-5397.

248. Westerink, M. A., and P. C. Giardina. 1992. Anti-idiotype induced protection against Neisseria meningitidis serogroup C bacteremia. Microbial Pathogenesis 12:19-26.

249. Prinz, D. M., S. L. Smithson, and M. A. Westerink. 2004. Two different methods result in the selection of peptides that induce a protective antibody response to Neisseria meningitidis serogroup C. Journal of Immunological Methods 285:1-14.

250. Luo, P., G. Canziani, G. Cunto-Amesty, and T. Kieber-Emmons. 2000. A molecular basis for functional peptide mimicry of a carbohydrate antigen. Journal of Biological Chemistry 275:16146-16154.

251. Moe, G. R., S. Tan, and D. M. Granoff. 1999. Molecular mimetics of polysaccharide epitopes as vaccine candidates for prevention of Neisseria meningitidis serogroup B disease. FEMS Immunology and Medical Microbiology 26:209-226.

252. Valadon, P., G. Nussbaum, L. F. Boyd, D. H. Margulies, and M. D. Scharff. 1996. Peptide libraries define the fine specificity of anti-polysaccharide antibodies to Cryptococcus neoformans. Journal of Molecular Biology 261:11-22.

253. Beenhouwer, D. O., R. J. May, P. Valadon, and M. D. Scharff. 2002. High affinity mimotope of the polysaccharide capsule of Cryptococcus neoformans identified from an evolutionary phage peptide library. Journal of Immunology 169:6992-6999.

254. Murali, R., and T. Kieber-Emmons. 1997. Molecular recognition of a peptide mimic of the Lewis Y antigen by an anti-Lewis Y antibody. Journal of Molecular Recognition 10:269-276.

255. Harris, S. L., L. Craig, J. S. Mehroke, M. Rashed, M. B. Zwick, K. Kenar, E. J. Toone, N. Greenspan, F. I. Auzanneau, J. R. Marino-Albernas, B. M. Pinto, and J. K. Scott. 1997. Exploring the basis of peptide-carbohydrate crossreactivity: evidence for discrimination by peptides between closely related anti-carbohydrate antibodies. Proceedings of the National Academy of Sciences of the United States of America 94:2454-2459.

226

256. Magliani, W., L. Polonelli, S. Conti, A. Salati, P. F. Rocca, V. Cusumano, G. Mancuso, and G. Teti. 1998. Neonatal mouse immunity against group B streptococcal infection by maternal vaccination with recombinant anti-idiotypes. Nature Medicine 4:705-709.

257. Fleuridor, R., A. Lees, and L. Pirofski. 2001. A cryptococcal capsular polysaccharide mimotope prolongs the survival of mice with Cryptococcus neoformans infection. Journal of Immunology 166:1087-1096.

258. Valadon, P., G. Nussbaum, J. Oh, and M. D. Scharff. 1998. Aspects of antigen mimicry revealed by immunization with a peptide mimetic of Cryptococcus neoformans polysaccharide. Journal of Immunology 161:1829-1836.

259. Zhang, H., Z. Zhong, and L. A. Pirofski. 1997. Peptide epitopes recognized by a human anti-cryptococcal glucuronoxylomannan antibody. Infection and Immunity 65:1158-1164.

260. Harris, S. L., A. S. Dagtas, and B. Diamond. 2002. Regulating the isotypic and idiotypic profile of an anti-PC antibody response: lessons from peptide mimics. Molecular Immunology 39:263-272.

261. Charalambous, B. M., and I. M. Feavers. 2000. Peptide mimics elicit antibody responses against the outer-membrane lipooligosaccharide of group B neisseria meningitidis. FEMS Microbiology Letters 191:45-50.

262. Gulati, S., D. P. McQuillen, J. Sharon, and P. A. Rice. 1996. Experimental immunization with a monoclonal anti-idiotope antibody that mimics the Neisseria gonorrhoeae lipooligosaccharide epitope 2C7. Journal of Infectious Diseases 174:1238-1248.

263. Kieber-Emmons, T., P. Luo, J. Qiu, T. Y. Chang, I. O, M. Blaszczyk-Thurin, and Z. Steplewski. 1999. Vaccination with carbohydrate peptide mimotopes promotes anti-tumor responses. Nature Biotechnology 17:660-665.

264. Kieber-Emmons, T., B. Monzavi-Karbassi, B. Wang, P. Luo, and D. B. Weiner. 2000. Cutting edge: DNA immunization with minigenes of carbohydrate mimotopes induce functional anti-carbohydrate antibody response. Journal of Immunology 165:623-627.

265. Chen, X., G. Scala, I. Quinto, W. Liu, T. W. Chun, J. S. Justement, O. J. Cohen, T. C. vanCott, M. Iwanicki, M. G. Lewis, J. Greenhouse, T. Barry, D. Venzon, and A. S. Fauci. 2001. Protection of rhesus macaques against disease progression from pathogenic SHIV-89.6PD by vaccination with phage-displayed HIV-1 epitopes. Nature Medicine 7:1225-1231. 227

266. Pashov, A., G. Canziani, B. Monzavi-Karbassi, S. V. Kaveri, S. Macleod, R. Saha, M. Perry, T. C. Vancott, and T. Kieber-Emmons. 2005. Antigenic properties of peptide mimotopes of HIV-1-associated carbohydrate antigens. Journal of Biological Chemistry 280:28959-28965.

267. Harris, S. L., M. K. Park, M. H. Nahm, and B. Diamond. 2000. Peptide mimic of phosphorylcholine, a dominant epitope found on Streptococcus pneumoniae. Infection and Immunity 68:5778-5784.

268. Lesinski, G. B., S. L. Smithson, N. Srivastava, D. Chen, G. Widera, and M. A. Westerink. 2001. A DNA vaccine encoding a peptide mimic of Streptococcus pneumoniae serotype 4 capsular polysaccharide induces specific anti- carbohydrate antibodies in Balb/c mice. Vaccine 19:1717-1726.

269. Hou, Y., and X. X. Gu. 2003. Development of peptide mimotopes of lipooligosaccharide from nontypeable Haemophilus influenzae as vaccine candidates. Journal of Immunology 170:4373-4379.

270. De Bolle, X., T. Laurent, A. Tibor, F. Godfroid, V. Weynants, J. J. Letesson, and P. Mertens. 1999. Antigenic properties of peptidic mimics for epitopes of the lipopolysaccharide from Brucella. Journal of Molecular Biology 294:181-191.

271. Molenaar, T. J., C. C. Appeldoorn, S. A. de Haas, I. N. Michon, A. Bonnefoy, M. F. Hoylaerts, H. Pannekoek, T. J. van Berkel, J. Kuiper, and E. A. Biessen. 2002. Specific inhibition of P-selectin-mediated cell adhesion by phage display-derived peptide antagonists. Blood 100:3570-3577.

272. Zhan, J., Z. Xia, L. Xu, Z. Yan, and K. Wang. 2003. A peptide mimetic of Gal- alpha 1,3-Gal is able to block human natural antibodies. Biochemical and Biophysical Research Communications 308:19-22.

273. Taki, T., D. Ishikawa, H. Hamasaki, and S. Handa. 1997. Preparation of peptides which mimic glycosphingolipids by using phage peptide library and their modulation on beta-galactosidase activity. FEBS Letters 418:219-223.

274. Kim, Y. G., C. S. Lee, W. J. Chung, E. M. Kim, D. S. Shin, J. H. Rhim, Y. S. Lee, B. G. Kim, and J. Chung. 2005. Screening of LPS-specific peptides from a phage display library using epoxy beads. Biochemical and Biophysical Research Communications 329:312-317.

275. Tang, S. S., W. S. Tan, S. Devi, L. F. Wang, T. Pang, and K. L. Thong. 2003. Mimotopes of the Vi antigen of Salmonella enterica serovar typhi identified from phage display peptide library. Clinical and Diagnostic Laboratory Immunology 10:1078-1084. 228

276. Phalipon, A., A. Folgori, J. Arondel, G. Sgaramella, P. Fortugno, R. Cortese, P. J. Sansonetti, and F. Felici. 1997. Induction of anti-carbohydrate antibodies by phage library-selected peptide mimics. European Journal of Immunology 27:2620-2625.

277. Johnson, M. A., M. Jaseja, W. Zou, H. J. Jennings, V. Copie, B. M. Pinto, and S. H. Pincus. 2003. NMR studies of carbohydrates and carbohydrate-mimetic peptides recognized by an anti-group B Streptococcus antibody. Journal of Biological Chemistry 278:24740-24752.

278. Levy, N. J., A. Nicholson-Weller, C. J. Baker, and D. L. Kasper. 1984. Potentiation of virulence by group B streptococcal polysaccharides. Journal of Infectious Diseases 149:851-860.

279. Zou, W., R. Mackenzie, L. Therien, T. Hirama, Q. Yang, M. A. Gidney, and H. J. Jennings. 1999. Conformational epitope of the type III group B Streptococcus capsular polysaccharide. Journal of Immunology 163:820-825.

280. Kasper, D. L. 1995. Designer vaccines to prevent infections due to group B Streptococcus. Proceedings of the Association of American Physicians 107:369- 373.

281. Boyer, K. M., L. S. Kendall, C. K. Papierniak, M. E. Klegerman, and S. P. Gotoff. 1984. Protective levels of human immunoglobulin G antibody to group B streptococcus type Ib. Infection and Immunity 45:618-624.

282. Muñoz, A. I. 1980. Hemophilus influenzae infections: a brief review. Clin Pediatr 19:86-90.

283. Anderson, P., and D. H. Smith. 1977. Isolation of the capsular polysaccharide from culture supernatant of Haemophilus influenzae type b. Infection and Immunity 15:472-477.

284. Anderson, P., D. H. Smith, D. L. Ingram, J. Wilkins, P. F. Wehrle, and V. M. Howie. 1977. Antibody of polyribophate of Haemophilus influenzae type b in infants and children: effect of immunization with polyribophosphate. Journal of Infectious Diseases 136 Suppl:S57-62.

285. Parke, J. C., Jr., R. Schneerson, J. B. Robbins, and J. J. Schlesselman. 1977. Interim report of a controlled field trial of immunization with capsular polysaccharides of Haemophilus influenzae type b and group C Neisseria meningitidis in Mecklenburg county, North Carolina (March 1974-March 1976). Journal of Infectious Diseases 136 Suppl:S51-56. 229

286. Peltola, H., H. Kayhty, A. Sivonen, and H. Makela. 1977. Haemophilus influenzae type b capsular polysaccharide vaccine in children: a double-blind field study of 100,000 vaccinees 3 months to 5 years of age in Finland. Pediatrics 60:730-737.

287. Smith, D. H., G. Peter, D. L. Ingram, A. L. Harding, and P. Anderson. 1973. Responses of children immunized with the capsular polysaccharide of Hemophilus influenzae, type b. Pediatrics 52:637-644.

288. Anderson, P., M. Pichichero, R. Insel, P. Farsad, and M. Santosham. 1985. Capsular antigens noncovalently or covalently associated with protein as vaccines to Haemophilus influenzae type b: comparison in two-year-old children. Journal of Infectious Diseases 152:634-636.

289. Anderson, P., M. E. Pichichero, and R. A. Insel. 1985. Immunogens consisting of oligosaccharides from the capsule of Haemophilus influenzae type b coupled to diphtheria toxoid or the toxin protein CRM197. Journal of Clinical Investigation 76:52-59.

290. Musher, D. M. 1992. Infections caused by Streptococcus pneumoniae: clinical spectrum, pathogenesis, immunity, and treatment. Clinical Infectious Diseases 14:801-807.

291. O'Brien, K. L., A. J. Swift, J. A. Winkelstein, M. Santosham, B. Stover, R. Luddy, J. E. Gootenberg, J. T. Nold, A. Eskenazi, S. J. Snader, and H. M. Lederman. 2000. Safety and immunogenicity of heptavalent pneumococcal vaccine conjugated to CRM(197) among infants with sickle cell disease. Pneumococcal Conjugate Vaccine Study Group. Pediatrics 106:965-972.

292. O'Brien, K. L., L. H. Moulton, R. Reid, R. Weatherholtz, J. Oski, L. Brown, G. Kumar, A. Parkinson, D. Hu, J. Hackell, I. Chang, R. Kohberger, G. Siber, and M. Santosham. 2003. Efficacy and safety of seven-valent conjugate pneumococcal vaccine in American Indian children: group randomised trial. Lancet 362:355- 361.

293. Ramsay, M. E., N. Andrews, E. B. Kaczmarski, and E. Miller. 2001. Efficacy of meningococcal serogroup C conjugate vaccine in teenagers and toddlers in England. Lancet 357:195-196.

294. Bachmann, M. F., and R. M. Zinkernagel. 1996. The influence of virus structure on antibody responses and virus serotype formation. Immunology Today 17:553- 558.

230

295. Bachmann, M. F., H. Hengartner, and R. M. Zinkernagel. 1995. T helper cell- independent neutralizing B cell response against vesicular stomatitis virus: role of antigen patterns in B cell induction? European Journal of Immunology 25:3445- 3451.

296. Burke, K. L., G. Dunn, M. Ferguson, P. D. Minor, and J. W. Almond. 1988. Antigen chimaeras of poliovirus as potential new vaccines. Nature 332:81-82.

297. Arnold, G. F., D. A. Resnick, Y. Li, A. Zhang, A. D. Smith, S. C. Geisler, A. Jacobo-Molina, W. Lee, R. G. Webster, and E. Arnold. 1994. Design and construction of rhinovirus chimeras incorporating immunogens from polio, influenza, and human immunodeficiency viruses. Virology 198:703-708.

298. Fernandez-Fernandez, M. R., J. L. Martinez-Torrecuadrada, J. I. Casal, and J. A. Garcia. 1998. Development of an antigen presentation system based on plum pox potyvirus. FEBS Letters 427:229-235.

299. Porta, C., V. E. Spall, J. Loveland, J. E. Johnson, P. J. Barker, and G. P. Lomonossoff. 1994. Development of cowpea mosaic virus as a high-yielding system for the presentation of foreign peptides. Virology 202:949-955.

300. Valenzuela, P., A. Medina, W. J. Rutter, G. Ammerer, and B. D. Hall. 1982. Synthesis and assembly of hepatitis B virus surface antigen particles in yeast. Nature 298:347-350.

301. Sabara, M., M. Parker, P. Aha, C. Cosco, E. Gibbons, S. Parsons, and L. A. Babiuk. 1991. Assembly of double-shelled rotaviruslike particles by simultaneous expression of recombinant VP6 and VP7 proteins. Journal of Virology 65:6994- 6997.

302. Urakawa, T., M. Ferguson, P. D. Minor, J. Cooper, M. Sullivan, J. W. Almond, and D. H. Bishop. 1989. Synthesis of immunogenic, but non-infectious, poliovirus particles in insect cells by a baculovirus expression vector. Journal of General Virology 70 ( Pt 6):1453-1463.

303. French, T. J., J. J. Marshall, and P. Roy. 1990. Assembly of double-shelled, viruslike particles of bluetongue virus by the simultaneous expression of four structural proteins. Journal of Virology 64:5695-5700.

304. Plana-Duran, J., M. Bastons, M. J. Rodriguez, I. Climent, E. Cortes, C. Vela, and I. Casal. 1996. Oral immunization of rabbits with VP60 particles confers protection against rabbit hemorrhagic disease. Archives of Virology 141:1423- 1436. 231

305. Thomsen, D. R., L. L. Roof, and F. L. Homa. 1994. Assembly of herpes simplex virus (HSV) intermediate capsids in insect cells infected with recombinant baculoviruses expressing HSV capsid proteins. Journal of Virology 68:2442- 2457.

306. Brown, C. S., J. W. Van Lent, J. M. Vlak, and W. J. Spaan. 1991. Assembly of empty capsids by using baculovirus recombinants expressing human parvovirus B19 structural proteins. Journal of Virology 65:2702-2706.

307. Nemchinov, L. G., T. J. Liang, M. M. Rifaat, H. M. Mazyad, A. Hadidi, and J. M. Keith. 2000. Development of a plant-derived subunit vaccine candidate against hepatitis C virus. Archives of Virology 145:2557-2573.

308. Joelson, T., L. Akerblom, P. Oxelfelt, B. Strandberg, K. Tomenius, and T. J. Morris. 1997. Presentation of a foreign peptide on the surface of tomato bushy stunt virus. Journal of General Virology 78 ( Pt 6):1213-1217.

309. Lin, T., Z. Chen, R. Usha, C. V. Stauffacher, J. B. Dai, T. Schmidt, and J. E. Johnson. 1999. The refined crystal structure of cowpea mosaic virus at 2.8 A resolution. Virology 265:20-34.

310. Usha, R., J. B. Rohll, V. E. Spall, M. Shanks, A. J. Maule, J. E. Johnson, and G. P. Lomonossoff. 1993. Expression of an animal virus antigenic site on the surface of a plant virus particle. Virology 197:366-374.

311. Wang, Q., E. Kaltgrad, T. Lin, J. E. Johnson, and M. G. Finn. 2002. Natural supramolecular building blocks. Wild-type cowpea mosaic virus. Chemistry and Biology 9:805-811.

312. Wang, Q., T. Lin, J. E. Johnson, and M. G. Finn. 2002. Natural supramolecular building blocks. Cysteine-added mutants of cowpea mosaic virus. Chemistry and Biology 9:813-819.

313. Chatterji, A., L. L. Burns, S. S. Taylor, G. P. Lomonossoff, J. E. Johnson, T. Lin, and C. Porta. 2002. Cowpea mosaic virus: from the presentation of antigenic peptides to the display of active biomaterials. Intervirology 45:362-370.

314. de Assis Filho, F. M., O. R. Paguio, J. L. Sherwood, and C. M. Deom. 2002. Symptom induction by Cowpea chlorotic mottle virus on Vigna unguiculata is determined by amino acid residue 151 in the coat protein. Journal of General Virology 83:879-883.

232

315. Mise, K., R. F. Allison, M. Janda, and P. Ahlquist. 1993. Bromovirus movement protein genes play a crucial role in host specificity. Journal of Virology 67:2815- 2823.

316. Rao, A. L., and G. L. Grantham. 1995. A spontaneous mutation in the movement protein gene of brome mosaic virus modulates symptom phenotype in Nicotiana benthamiana. Journal of Virology 69:2689-2691.

317. Speir, J. A., S. Munshi, G. Wang, T. S. Baker, and J. E. Johnson. 1995. Structures of the native and swollen forms of cowpea chlorotic mottle virus determined by X-ray crystallography and cryo-electron microscopy. Structure 3:63-78.

318. Albert, F. G., J. M. Fox, and M. J. Young. 1997. Virion swelling is not required for cotranslational disassembly of cowpea chlorotic mottle virus in vitro. Journal of Virology 71:4296-4299.

319. Fox, J. M., X. Zhao, J. A. Speir, and M. J. Young. 1996. Analysis of a salt stable mutant of cowpea chlorotic mottle virus. Virology 222:115-122.

320. Brumfield, S., D. Willits, L. Tang, J. E. Johnson, T. Douglas, and M. Young. 2004. Heterologous expression of the modified coat protein of Cowpea chlorotic mottle bromovirus results in the assembly of protein cages with altered architectures and function. Journal of General Virology 85:1049-1053.

321. Feliciello, I., and G. Chinali. 1993. A modified alkaline lysis method for the preparation of highly purified plasmid DNA from Escherichia coli. Analytical Biochemistry 212:394-401.

322. Linder, S., M. Schliwa, and E. Kube-Granderath. 1996. Direct PCR screening of Pichia pastoris clones. Biotechniques 20:980-982.

323. Ao, J. Q., J. W. Wang, X. H. Chen, X. Z. Wang, and Q. X. Long. 2003. Expression of pseudorabies virus gE epitopes in Pichia pastoris and its utilization in an indirect PRV gE-ELISA. Journal of Virological Methods 114:145-150.

324. Freivalds, J., A. Dislers, V. Ose, D. Skrastina, I. Cielens, P. Pumpens, K. Sasnauskas, and A. Kazaks. 2006. Assembly of bacteriophage Qbeta virus-like particles in yeast Saccharomyces cerevisiae and Pichia pastoris. Journal of Biotechnology 123:297-303.

233

325. Martinez-Donato, G., N. Acosta-Rivero, J. Morales-Grillo, A. Musacchio, A. Vina, C. Alvarez, N. Figueroa, I. Guerra, J. Garcia, L. Varas, V. Muzio, and S. Duenas-Carrera. 2006. Expression and processing of hepatitis C virus structural proteins in Pichia pastoris yeast. Biochemical and Biophysical Research Communications 342:625-631.

326. Johnson, J. E., and W. Chiu. 2000. Structures of virus and virus-like particles. Current Opinion in Structural Biology 10:229-235.

327. Johnson, J. E., and J. A. Speir. 1997. Quasi-equivalent viruses: a paradigm for protein assemblies. Journal of Molecular Biology 269:665-675.

328. Rossmann, M. G., and J. E. Johnson. 1989. Icosahedral RNA virus structure. Annual Review of Biochemistry 58:533-573.

329. Zhao, X., J. M. Fox, N. H. Olson, T. S. Baker, and M. J. Young. 1995. In vitro assembly of cowpea chlorotic mottle virus from coat protein expressed in Escherichia coli and in vitro-transcribed viral cDNA. Virology 207:486-494.

330. Vriend, G., B. J. Verduin, and M. A. Hemminga. 1986. Role of the N-terminal part of the coat protein in the assembly of cowpea chlorotic mottle virus. A 500 MHz proton nuclear magnetic resonance study and structural calculations. Journal of Molecular Biology 191:453-460.

331. Johnson, J. E. 1996. Functional implications of protein-protein interactions in icosahedral viruses. Proceedings of the National Academy of Sciences of the United States of America 93:27-33.

332. Graf, J., Y. Iwamoto, M. Sasaki, G. R. Martin, H. K. Kleinman, F. A. Robey, and Y. Yamada. 1987. Identification of an amino acid sequence in laminin mediating cell attachment, chemotaxis, and receptor binding. Cell 48:989-996.

333. Kazmin, D. A., T. R. Hoyt, L. Taubner, M. Teintze, and J. R. Starkey. 2000. Phage display mapping for peptide 11 sensitive sequences binding to laminin-1. Journal of Molecular Biology 298:431-445.

334. Rao, A. L., and B. Cooper. 2006. Capsid protein gene and the type of host plant differentially modulate cell-to-cell movement of cowpea chlorotic mottle virus. Virus Genes 32:219-227.

335. Sasaki, N., M. Kaido, T. Okuno, and K. Mise. 2005. Coat protein-independent cell-to-cell movement of bromoviruses expressing brome mosaic virus movement protein with an adaptation-related amino acid change in the central region. Archives of Virology 150:1231-1240. 234

336. Xia, M., T. Farkas, and X. Jiang. 2007. Norovirus capsid protein expressed in yeast forms virus-like particles and stimulates systemic and mucosal immunity in mice following an oral administration of raw yeast extracts. Journal of Medical Virology 79:74-83.

337. Li, H. Z., H. Y. Gang, Q. M. Sun, X. Liu, Y. B. Ma, M. S. Sun, and C. B. Dai. 2004. Production in Pichia pastoris and characterization of genetic engineered chimeric HBV/HEV virus-like particles. Chinese Medical Sciences Journal 19:78-83.

338. Falcon, V., C. Garcia, M. C. de la Rosa, I. Menendez, J. Seoane, and J. M. Grillo. 1999. Ultrastructural and immunocytochemical evidences of core-particle formation in the methylotrophic Pichia pastoris yeast when expressing HCV structural proteins (core-E1). Tissue and Cell 31:117-125.

339. McLain, L., C. Porta, G. P. Lomonossoff, Z. Durrani, and N. J. Dimmock. 1995. Human immunodeficiency virus type 1-neutralizing antibodies raised to a glycoprotein 41 peptide expressed on the surface of a plant virus. AIDS Research and Human Retroviruses 11:327-334.

340. Porta, C., and G. P. Lomonossoff. 1998. Scope for using plant viruses to present epitopes from animal pathogens. Reviews in Medical Virology 8:25-41.

341. Fox, J. M., F. G. Albert, J. A. Speir, and M. J. Young. 1997. Characterization of a disassembly deficient mutant of cowpea chlorotic mottle virus. Virology 227:229- 233.

342. Schuchat, A. 1999. Group B streptococcus. Lancet 353:51-56.

343. Wenger, J. D., A. W. Hightower, R. R. Facklam, S. Gaventa, and C. V. Broome. 1990. Bacterial meningitis in the United States, 1986: report of a multistate surveillance study. The Bacterial Meningitis Study Group. Journal of Infectious Diseases 162:1316-1323.

344. Daugaard, H. O., A. C. Thomsen, U. Henriques, and A. Ostergaard. 1988. Group B streptococci in the lower urogenital tract and late abortions. American Journal of Obstetrics and Gynecology 158:28-31.

345. Baltimore, R. S., D. L. Kasper, and J. Vecchitto. 1979. Mouse protection test for group B Streptococcus type III. Journal of Infectious Diseases 140:81-88.

235

346. De Cueninck, B. J., T. K. Eisenstein, T. S. McIntosh, G. D. Shockman, and R. M. Swenson. 1983. Quantitation of in vitro opsonic activity of human antibody induced by a vaccine consisting of the type III-specific polysaccharide of group B streptococcus. Infection and Immunity 39:1155-1160.

347. Baltimore, R. S., C. J. Baker, and D. L. Kasper. 1981. Antibody to group B Streptococcus type III in human sera measured by a mouse protection test. Infection and Immunity 32:56-61.

348. Baltimore, R. S., D. L. Kasper, C. J. Baker, and D. K. Goroff. 1977. Antigenic specificity of opsonophagocytic antibodies in rabbit anti-sera to group B streptococci. Journal of Immunology 118:673-678.

349. Youn, J. H., H. J. Myung, A. Liav, D. Chatterjee, P. J. Brennan, I. H. Choi, S. N. Cho, and J. S. Shin. 2004. Production and characterization of peptide mimotopes of phenolic glycolipid-I of Mycobacterium leprae. FEMS Immunology and Medical Microbiology 41:51-57.

350. Gevorkian, G., E. Segura, G. Acero, J. P. Palma, C. Espitia, K. Manoutcharian, and L. M. Lopez-Marin. 2005. Peptide mimotopes of Mycobacterium tuberculosis carbohydrate immunodeterminants. Biochemical Journal 387:411-417.

351. Yusibov, V., D. C. Hooper, S. V. Spitsin, N. Fleysh, R. B. Kean, T. Mikheeva, D. Deka, A. Karasev, S. Cox, J. Randall, and H. Koprowski. 2002. Expression in plants and immunogenicity of plant virus-based experimental rabies vaccine. Vaccine 20:3155-3164.

352. Denis, J., N. Majeau, E. Acosta-Ramirez, C. Savard, M. C. Bedard, S. Simard, K. Lecours, M. Bolduc, C. Pare, B. Willems, N. Shoukry, P. Tessier, P. Lacasse, A. Lamarre, R. Lapointe, C. Lopez Macias, and D. Leclerc. 2007. Immunogenicity of papaya mosaic virus-like particles fused to a hepatitis C virus epitope: evidence for the critical function of multimerization. Virology 363:59-68.

353. Rote, N. S., N. L. Taylor, A. O. Shigeoka, J. R. Scott, and H. R. Hill. 1980. Enzyme-linked immunosorbent assay for group B streptococcal antibodies. Infection and Immunity 27:118-123.

354. Gleba, Y., S. Marillonnet, and V. Klimyuk. 2004. Engineering viral expression vectors for plants: the 'full virus' and the 'deconstructed virus' strategies. Current Opinion in Plant Biology 7:182-188.

355. Lomonossoff, G. P., and J. E. Johnson. 1991. The synthesis and structure of comovirus capsids. Progress in Biophysics and Molecular Biology 55:107-137. 236

356. Lin, T., C. Porta, G. Lomonossoff, and J. E. Johnson. 1996. Structure-based design of peptide presentation on a viral surface: the crystal structure of a plant/animal virus chimera at 2.8 A resolution. Folding and Design 1:179-187.

357. Liu, L., M. C. Canizares, W. Monger, Y. Perrin, E. Tsakiris, C. Porta, N. Shariat, L. Nicholson, and G. P. Lomonossoff. 2005. Cowpea mosaic virus-based systems for the production of antigens and antibodies in plants. Vaccine 23:1788-1792.

358. Xu, F., T. D. Jones, and P. B. Rodgers. 1996. Potential of chimaeric plant virus particles as novel, stable vaccines. Developments in Biological Standardization 87:201-205.

359. Jegerlehner, A., T. Storni, G. Lipowsky, M. Schmid, P. Pumpens, and M. F. Bachmann. 2002. Regulation of IgG antibody responses by epitope density and CD21-mediated costimulation. European Journal of Immunology 32:3305-3314.

360. Yang, R., F. M. Murillo, H. Cui, R. Blosser, S. Uematsu, K. Takeda, S. Akira, R. P. Viscidi, and R. B. Roden. 2004. Papillomavirus-like particles stimulate murine bone marrow-derived dendritic cells to produce alpha interferon and Th1 immune responses via MyD88. Journal of Virology 78:11152-11160.

361. Bach, P., E. Kamphuis, B. Odermatt, G. Sutter, C. J. Buchholz, and U. Kalinke. 2007. Vesicular stomatitis virus glycoprotein displaying retrovirus-like particles induce a type I IFN receptor-dependent switch to neutralizing IgG antibodies. Journal of Immunology 178:5839-5847.

362. Ball, J. M., D. Y. Graham, A. R. Opekun, M. A. Gilger, R. A. Guerrero, and M. K. Estes. 1999. Recombinant Norwalk virus-like particles given orally to volunteers: phase I study. Gastroenterology 117:40-48.

363. Ye, L., J. Lin, Y. Sun, S. Bennouna, M. Lo, Q. Wu, Z. Bu, B. Pulendran, R. W. Compans, and C. Yang. 2006. Ebola virus-like particles produced in insect cells exhibit dendritic cell stimulating activity and induce neutralizing antibodies. Virology 351:260-270.

364. Brown, D. R., J. T. Bryan, J. M. Schroeder, T. S. Robinson, K. H. Fife, C. M. Wheeler, E. Barr, P. R. Smith, L. Chiacchierini, A. DiCello, and K. U. Jansen. 2001. Neutralization of human papillomavirus type 11 (HPV-11) by serum from women vaccinated with yeast-derived HPV-11 L1 virus-like particles: correlation with competitive radioimmunoassay titer. Journal of Infectious Diseases 184:1183-1186.

237

365. Reyburn, H. T., D. J. Roy, B. A. Blacklaws, D. R. Sargan, and I. McConnell. 1992. Expression of maedi-visna virus major core protein, p25: development of a sensitive p25 antigen detection assay. Journal of Virological Methods 37:305- 320.

366. Svirshchevskaya, E. V., L. Alekseeva, A. Marchenko, N. Viskova, T. M. Andronova, S. V. Benevolenskii, and V. P. Kurup. 2002. Immune response modulation by recombinant peptides expressed in virus-like particles. Clinical and Experimental Immunology 127:199-205.

367. Tegerstedt, K., A. V. Franzen, K. Andreasson, J. Joneberg, S. Heidari, T. Ramqvist, and T. Dalianis. 2005. Murine polyomavirus virus-like particles (VLPs) as vectors for gene and immune therapy and vaccines against viral infections and cancer. Anticancer Research 25:2601-2608.

368. Veenstra, J., I. G. Williams, R. Colebunders, L. Dorrell, S. E. Tchamouroff, G. Patou, J. M. Lange, I. V. Weller, J. Goeman, S. Uthayakumar, I. R. Gow, J. N. Weber, and R. A. Coutinho. 1996. Immunization with recombinant p17/p24:Ty virus-like particles in human immunodeficiency virus-infected persons. Journal of Infectious Diseases 174:862-866.

369. Prevention., C. f. D. C. a. 1996. Prevention of perinatal group B streptococcal disease: a public health perspective. In Morbid. Mortal. Wk. Rep. 1.

370. Madoff, L. C., L. C. Paoletti, J. Y. Tai, and D. L. Kasper. 1994. Maternal immunization of mice with group B streptococcal type III polysaccharide-beta C protein conjugate elicits protective antibody to multiple serotypes. Journal of clinical investigation 94:286.

371. Dower, W. J., J. F. Miller, and C. W. Ragsdale. 1988. High efficiency transformation of E. coli by high voltage electroporation. Nucleic Acids Research 16:6127-6145.

372. Speyer, J. F. 1990. A simple and effective electroporation apparatus. Biotechniques 8:28-30.

373. Parmley, S. F., and G. P. Smith. 1988. Antibody-selectable filamentous fd phage vectors: affinity purification of target genes. Gene 73:305-318.

374. Parmley, S. F., and G. P. Smith. 1989. Filamentous fusion phage cloning vectors for the study of epitopes and design of vaccines. Adv Exp Med Biol 251:215-218.

375. Smith, G. P., and J. K. Scott. 1993. Libraries of peptides and proteins displayed on filamentous phage. Methods in Enzymology 217:228-257. 238

376. Yu, J., and G. P. Smith. 1996. Affinity maturation of phage-displayed peptide ligands. Methods in Enzymology 267:3-27.

377. Smith, G. P. 1993. Surface display and peptide libraries. Gene 128:1-2.

378. Murray, A., R. G. Smith, K. Brady, S. Williams, R. A. Badley, and M. R. Price. 2001. Generation and refinement of peptide mimetic ligands for paratope-specific purification of monoclonal antibodies. Analytical Biochemistry 296:9-17.

379. Scott, J. K., and G. P. Smith. 1990. Searching for peptide ligands with an epitope library. Science 249:386-390.

380. Bentley, L., J. Fehrsen, F. Jordaan, H. Huismans, and D. H. du Plessis. 2000. Identification of antigenic regions on VP2 of African horsesickness virus serotype 3 by using phage-displayed epitope libraries. Journal of General Virology 81:993- 1000.

381. Pereboeva, L. A., A. V. Pereboev, L. F. Wang, and G. E. Morris. 2000. Hepatitis C epitopes from phage-displayed cDNA libraries and improved diagnosis with a chimeric antigen. Journal of Medical Virology 60:144-151.

382. Lundin, K., A. Samuelsson, M. Jansson, J. Hinkula, B. Wahren, H. Wigzell, and M. A. Persson. 1996. Peptides isolated from random peptide libraries on phage elicit a neutralizing anti-HIV-1 response: analysis of immunological mimicry. Immunology 89:579-586.

383. Wang, L. F., D. H. Du Plessis, J. R. White, A. D. Hyatt, and B. T. Eaton. 1995. Use of a gene-targeted phage display random epitope library to map an antigenic determinant on the bluetongue virus outer capsid protein VP5. Journal of Immunological Methods 178:1-12.

384. Fehrsen, J., and D. H. du Plessis. 1999. Cross-reactive epitope mimics in a fragmented-genome phage display library derived from the rickettsia, Cowdria ruminantium. Immunotechnology 4:175-184.

385. Petrenko, V. A., G. P. Smith, X. Gong, and T. Quinn. 1996. A library of organic landscapes on filamentous phage. Protein Engineering 9:797-801.

386. Tsunetsugu-Yokota, Y., M. Tatsumi, V. Robert, C. Devaux, B. Spire, J. C. Chermann, and I. Hirsch. 1991. Expression of an immunogenic region of HIV by a filamentous bacteriophage vector. Gene 99:261-265.

239

387. Oldenburg, K. R., D. Loganathan, I. J. Goldstein, P. G. Schultz, and M. A. Gallop. 1992. Peptide ligands for a sugar-binding protein isolated from a random peptide library. Proceedings of the national academy of sciences of the united states of america 89:5393-5397.

388. Scott, J. K., D. Loganthan, R. B. Easley, X. Gong, and I. J. Goldstein. 1992. A family of concanavalin A-binding peptides from a hexapeptide epitope library. Proceedings of the national academy of sciences of the united states of america 89:5398-5402.

389. Murai, H., S. Hara, T. Ikenaka, A. Goto, M. Arai, and S. Murao. 1985. Amino acid sequence of protein a-amylase inhibitor from Streptomyces griseosporeus YM-25. Journal of biochemistry 97:1129-1133.

390. Kieber-Emmons, T., B. Monzavi-Karbassi, B. Wang, P. Luo, and D. B. Weiner. 2000. DNA Immunization with minigenes of carbohydrate mimotopes induce functional anticarbohydrate antibody response. Journal of immunology 165:623- 627.

391. Kieber-Emmons, T., P. Luo, J. Qiu, T. Y. Chang, I. O, M. Blaszczyk-Thurin, and Z. Steplewski. 1999. Vaccination with carbohydrate peptide mimotopes promotes anti-tumor responses. Nature biotechnology 17:660-665.

392. Moe, G. R., S. Tan, and D. M. Granoff. 1999. Molecular mimetics of polysaccharide epitopes as vaccine candidates for prevention of Neisseria meningitidis serogroup B disease. FEMS Immunology and medical microbiology 26:209-226.

393. Molenaar, T. J. M., C. C. M. Appeldoorn, S. Haas, A. M. de, I. N. Michon, A. Bonnefoy, M. F. Hoylaerts, H. Pannekoek, T. C. v. Berkel, J. Kuiper, and E. A. Biessen. 2002. Specific inhibition of P-selectin-mediated cell adhesion by phage diplay-derived peptide antagonists. Blood 100:3570-3577.

394. Martin, T. R., J. T. Ruzinski, C. E. Rubens, E. Y. Chi, and C. B. Wilson. 1992. The effect of type-specific polysaccharide capsule on the clearance of group B streptococci from the lungs of infant and adult rats. Journal of Infectious Diseases 165:306-314.

395. Awram, P., R. C. Gardner, R. L. Forster, and A. R. Bellamy. 2002. The potential of plant viral vectors and transgenic plants for subunit vaccine production. Advances in Virus Research 58:81-124.