study of yellow fever vaccine viruses by Terence Neil ...epubs.surrey.ac.uk/855790/1/27606615.pdf · ANTIGENIC ANALYSIS OF EPITOPES ON YELLOW FEVER VIRUSES AND ITS VACCINES BY HAI

study of yellow fever vaccine viruses with monoclonal antibodies

by

Terence Neil Ledger

Submitted for the degree of Doctor of Philosophy

Department of Microbiology University of Surrey

Guildford Surrey GU2 5XH

U.K.September 1990

ProQuest Number: 27606615

All rights reserved

INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted.

In the unlikely event that the author did not send a com p le te manuscript and there are missing pages, these will be noted. Also, if material had to be removed,

a note will indicate the deletion.

uestProQuest 27606615

Published by ProQuest LLO (2019). Copyright of the Dissertation is held by the Author.

All rights reserved.This work is protected against unauthorized copying under Title 17, United States C ode

Microform Edition © ProQuest LLO.

ProQuest LLO.789 East Eisenhower Parkway

P.Q. Box 1346 Ann Arbor, Ml 48106- 1346

S U M M A R Y

This thesis describes the antigenic and biological analysis of 31 yellow fever (YF) vaccine viruses from three different vaccine strains (17D-204, 17DD and the French Neurotropic Vaccine [FNV]) using panels of monoclonal antibodies (McAbs).

The antigenic specificity and biological function of the epitopes recognised by the McAbs were identified. Two YF-type specific and one flavivirus group reactive McAbs were identified while the remainder were partial flavivirus group reactive.

Four panels of McAbs were used together with polyclonal anti-sera to antigenically analyse the envelope (E) protein epitopes on YF viruses using haemagglutination inhibition assays (HAI) and plaque reduction neutralisation tests (PRNT’s). Heterogeneity in the biological function of the E-protein epitopes was demonstrated. Using the technique of cluster analysis a dendrogram was produced which divided all YF viruses into three groups.

McAb neutralisation escape variants were produced to examine the function of particular epitopes in detail. One 17D-204 sub-strain specific epitope was found to affect the neurovirulence of YF 17D-204 vaccine

virus for mice.The new 17D YF vaccine virus manufactured in the

Union of Soviet and Socialist Republics (USSR) was examined in detail. The vaccine was indistinguishable from other 17D vaccines with the majority of McAbs. However, pathogenicity studies showed that the vaccine contained virions which killed mice with a reduced average survival time compared to other 17D vaccines.

Three FNV vaccine viruses were studied in HAI assays, PRNT's and pathogenicity tests. In HAI assays they were indistinguishable, while PRNT’s and pathogenicity studies were able to distinguish between them.

The 17DD sub-strain vaccine viruses and their wild-type parent strain Asibi were found to be temperature sensitive when grown in the mosquito C6-36 cell line and not the human SW13 cell line. No evidence was found to associate this phenotype with epitopes on the E-protein.

Ill

TABLE OF CONTENTSPAGE

TitleSummaryTable of Contents List of Tables List of Figures List of Abbreviations Acknowledgements

1iiiv

xviiX XV

xxviixxix

CHAPTER 1.

LITERATURE REVIEW

1. Yellow Fever virus

I. Historical considerations of the virusA) The VectorB) The Agent

II. Geographical distributionA) AmericaB) Africa

IV

III. Clinical description and Pathogenesisof the disease YF 10

A) Humans 10i) Asymptomatic I infections 11

ii) Classic severe cases 11

B ) Monkeys 12C) Mice 15D) Guinea Pigs 16E) Rabbits 16F) Hedgehogs 16G) Mosquitoes 17

IV. Yellow fever vaccines 18A) Vaccine development 18

i) French neurotropic virus 18ii) 17D vaccine virus 21

B) Research on yellow fever vaccines 27i) Mosquito Transmission 27

ii) Passage of YF virus in Different 28tissue types

iii) Virulence of YF vaccines 30iv) Molecular analysis of YF vaccine

viruses 32

2. Characteristics of the virus 37I. Taxonomic classification 37

A) Arboviruses 37B) Togaviridae 38C) Flaviviridae 39

II. Flavivirus morphology and structure 41A) Virion 41B) RNA 41

III. Flavivirus replication 43A) Uptake of virus 43B) Translation 44C) Replication 44D) Virion assembly and release 45

3. The production and use of MonoclonalAntibodies in virology 46A) History 46B) Haemagglutination and haemagglutination

inhibition assays 47C) Neutralisation tests 48

4. Antigenic and molecular analysis of theflavivirus envelope protein 49

I. Antigenic Analysis 50A) Analysis of YF virus E-protein 50B) Analysis of TEE complex E-protein 52C) Analysis of the JE subgroup

E-protein 59D) Analysis of the dengue virus complex

E-protein 69II. Molecular analysis of flaviviruses 72

III. Sequence analysis 785. Aims of Thesis 80

iVl

CHAPTER 2.

MATERIALS AND METHODS 81

1. CELL CULTURES AND VIRUSES 82

I. Cell Cultures 82A) Propagation and maintenance of cell

lines 82B) Storage of cells 84

II. Viruses 84A) Wild-type YF viruses 84B) YF Vaccine viruses 86C) Other Flaviviruses 88D) Virus plaque assays 88E) Virus TCIDgQ assay 90

III. Haemaglutinin Titrations andhaemagglutination inhibition assays 91A) Haemagglutination Titrations 91B) Haemagglutination inhibition Assays 93

2. MONOCLONAL ANTIBODIES 93

I. Production of ascitic fluid fromhybridomas 93

II. Properties of McAbs obtained fromother Laboratories 95

III. Properties of McAbs produced inthe laboratory 95

VII

IV. Determination of protein contentof ascitic fluids 98

V. Isotyping of McAbs 98VI. Preparation of multispot

immunofluorescence slides 99VII. Indirect Immunofluorescence Tests 100

VIII. Plaque reduction neutralisation tests(PRNT’s) 101A) Serum neutralisation indices 101B) The 100 PFU Test 101

3. PRODUCTION OF MONOCLONAL ANTIBODYNEUTRALISATION ESCAPE VARIANTS 102

I. Plaque purification of singleneutralisation escape variant virions from plaque reduction neutralisation tests 102A) Preparation of neutralisation

escape variants (McAb^) 102B) Confirmation of McAb neutralisation

escape variant 103II. Producing neutralisation escape variants

by a immunoprécipitation technique 104III. Producing neutralisation escape variants

in tissue culture 105IV. Producing variant virus using an Elisa

plate method 105

4. VIRAL PATHOGENESIS STUDIES 105

I. Mouse Pathogenesis Studies 106A) Adult Mice 106B) Newborn mice 107

II. Monkey Pathogenesis Tests 107

VIII

5. TEMPERATURE SENSITIVE STUDIES WITH YF VIRUSES 108I. Growth of the virus 108

II. The virus assay 108

6. RADIO-IMMUNO-PRECIPITATION OF VIRUSPOLYPEPTIDES 109

RESULTS

CHAPTER 3

DERIVATION OF MCABS 111I. Introduction 112II. Results 113

A) Cloning and secretion of ’G ’ hybridoma colonies secreting McAb 113

B) Production of McAb containing ascitic fluids from the A, B and G series of hybridoma cell lines 113

C) Antigenic reactivities of Ascites fluid in Indirect Immunofluorescence Tests 114i) The A, B and G series of ascites fluid 114

D) Serum Neutralisation Indices (SNI) of McAbs from the A, B, and G series againsttheir respective virus strains 116

E) HAI activity of McAbs from the A, B and G series against their respective virusstrains 116

F) Isotype characterisation of the A, Band G series 118

G) RIP of Virus Polypeptides 118

IX

III. Discussion 120

CHAPTER 4

ANTIGENIC ANALYSIS OF EPITOPES ON YELLOWFEVER VIRUSES AND ITS VACCINES BY HAI ASSAYSAND PRNTs 126

I. Introduction 127II. Results 127

A) Study of the Nigerian and USSR vaccines byIIF, HAI and PRNT 127

B) Analysis of virus by HAI assays usingthe A, B, G and H series of McAbs 128i) 17D-204 vaccine strains 128

ii) 17DD vaccine viruses 134iii) FNV vaccine viruses 136iv) Wild-type YF viruses 136

C) HAI assays with polyclonal anti-sera 139

D) Cluster analysis of the HAI assays withviruses against twenty seven of the McAbsused in the study 139i) All viruses and plaque purified

preparations 141ii) Viruses without plaque purified

preparations 143iii) 17D-204 vaccine viruses compared to the

Asibi wild-type strain 143E) Analysis of viruses using the A, B, F,

G and H Mcab series in PRNT's 146i) Three YF17D-204 vaccine strains and

their plaque purified virus isolates 146

ii) Analysis of two 17DD vaccines 147

F) Examination of YF viruses with Mcabs thatelicit neutralisation activity 150i) 17D-204 vaccine viruses 150

ii) 17DD vaccine viruses 152iii) FNV vaccine viruses 152iv) Wild-type YF viruses 152

G) Examination of YF viruses with polyclonal anti-sera 152

III. Discussion 153

CHAPTER 5

PRODUCTION AND ANALYSIS OF YELLOWFEVER ANTIGENIC VARIANTS 159

I. Introduction 160

II. Results 161A) Producing antigenic variants by the

immuno-precipitation technique 161B) Producing neutralisation escape

variants in tissue culture 162C) Production of variants using the

Elisa technique 164D) Production of McAb neutralisation resistant

McAb* variants from PRNT’s 166

i) Selection of the McAb* variants 166E) Indirect Immunofluorescence tests

on McAb^ variants 171i) YF17D-204 864* variants 171

XI

ii) YF17D-204 B39* variants 172iii) YF17D-204 B26* variants 172

F ) Comparison of McAb* variants with their parent virus strains in HAIassays 175i) YF17D-204 864* variants 175

ii) YF17D-204 B39* variants 178iii) YF17D-204 B26* variants 180

G) Comparison of McAb* variants with theirparent virus strains in PRNT’s with McAbs 180i) YF17D-204 864* variants 180

ii) YF17D-204 B39* variants 182H) Comparison of mouse virulence of the parent

virus strains with the variants 184i) YF17D-204 864* variants 184

a) adult mice 184ii) YF17D-204 B39* variants 186

a) adult mice 186iii) The YF17D-204 B26* variants 186

a) adult mice 186b) newborn suckling mice 186

III. Discussion 190

CHAPTER 6

EXAMINATION OF THE YF VACCINEMANUFACTURED IN THE USSR 197

I. Introduction 198II. Results 200

XII

A) Antigenic reactivity of the USSR vaccine compared to othervaccines from each strain 200i) Indirect immunofluorescence Tests 200

ii) HAI assays 202iii) PRNTs 207

B) Comparison of USSR vaccine to other vaccines strains in pathogenicitystudies 207i) In adult mice 207

ii) In newborn mice 209iii) Intranasal inoculation of mice

with the USSR vaccine 212iv) Intracerebral inoculation of a cynomolgus

monkey with the USSR vaccine 214

C) Plaque purification of the USSR vaccine 214D) Comparison of plaque purified USSR

viruses to that of the parent USSRvaccine 216i) IIF tests 216

ii) HAI assays 218iii) Pathogenicity study 218

a) In adult mice 218b) In newborn mice 221

E) Comparison of original USSR vaccine passed through mouse brain and the original vaccine passed intissue culture 223i) IIF tests 223

ii) HAI assays 225iii) Pathogenicity study 225

xiil

a) In adult mice 225b) In newborn mice 228

III. Discussion 228

CHAPTER 7

EXAMINATION OF THE FNV VACCINES 234

I. Introduction 235

II. Results 235

A) Comparison of reactivities of each FNV vaccine with other virus strainsin HAI assays with McAbs 235i) The A, B, G and H McAb series 235

ii) The F McAb series 236B) Cluster analysis of the HAI results

using the F McAbs series 239C) Compaison of reactivities of the FNV

vaccine viruses with other YF viruses using YF type, flavivirus sub-groupand group specific McAbs 241

D) Comparison of viruses in pathogenicitystudies 243i) In newborn mice 243

ii) Comparison of FNV vaccines when inoculated intracerebrally into a cynomolgus monkey 245

III. Discussion 245

XIV

CHAPTER 8

STUDY OF THE EFFICIENCY OF PLAQUING (E.O.P) OF YF VIRUSES GROWN IN THE C6-36 MOSQUITO CELL LINE AND PLAQUED AT A RESTRICTIVE TEMPERATURE OF 39.5-C 250

I. Introduction 251

II. Results 252A) Temperature sensitivity of viruses

grown in the C6-36 A. Albopictusmosquito cell line 252i) 17D-204 vaccines 252

ii) 17DD vaccines 252iii) FNV vaccines 255iv) Wild-type viruses 255

B) Analysis of the E.O.P of the 17DD Brazil plaque purified preparation when passaged from SW13 to C6-36 andthen back to SW13 cells 255

C) Antigenic analysis with McAbs of the 17DD Brazil plaque purified virus strain whengrown in the C6-36 and SW13 cell lines 256i) HAI assays 256

ii) PRNT's 256

III. Discussion 259

XV

CHAPTER 9

REACTIVITIES OF SOME McAbs in PRNTs WITH VIRUSES GROWN IN DIFFERENT CELL LINES 264

I. Introduction 265

II. Results 266A) The YF 17D-204 sub-strain specific McAb 266B) Flavivirus subgroup and group specific

McAbs 268

III. Discussion 271

CHAPTER 10

GENERAL DISCUSSION 276

REFERENCES 285

APPENDIX 312

XV!

LIST OF TABLESPAGE

1.1 Functions of YF virus proteins 341.2 Nucleotide and amino acid changes between

wild-type Asibi and the 17D-204 USAvaccine virus strain 36

1.3 Members of the Flaviviridae assigned toto antigenic complexes 42

1.4 Flaviviruses sequenced 792.1 Cell lines used in the studies 832.2 Wild-type YF viruses used in this study 852.3 YF vaccine viruses used in this study 872.4 Other flaviviruses used in these studies 892.5 Properties of envelope protein reactive

McAbs obtained from other Laboratories 962.6 Properties of the F and H McAb series 973.1 Results of indirect immunofluorescence

studies with the A, B and G series ofascites fluid 115

3.2 Serum neutralisation indices (SNI) ofMcAbs from the A, B and G series 117

3.3 HAI activity of McAbs from the A, Band G series 119

3.4 Summary of properties of the A, B and G McAb series using only the 17D-204 WHO and 17DD Brazil vaccine viruses tostudy the McAbs 123

XYll

4.1 Reactivity of the USSR and Nigerian vaccines in IIF tests, HAI assays andPRNT's 129

4.2 Analysis of YF17D-204 and the two new vaccine (Nigeria and USSR) viruses by HAI assays using the McAb A, B, G andH series 130

4.3 Analysis of the YF17DD vaccine viruses by HAI assays using the A, B, G and HMcAb series 135

4.4 Analysis of YF FNV viruses by HAI assaysusing the A, B, G, and H McAb series 137

4.5 Analysis of wild-type viruses by HAI assays using the A, B, G and H McAbseries 138

4.6 Reaction of polyclonal anti-sera in HAIassays with YF viruses 140

4.7 Survey of A, B, F, G and H series of McAbs with three YF17D-204 vaccine strains and their plaque purifiedvirus isolates in PRNT's 148

4.8 Survey of A, B, F, G and H series of McAbs with two YF17DD vaccine strainsin PRNT's 149

XVIll

4.9 Examination of YF viruses with McAbs that elicit neutralising activity inPRNT's 151

4.10 Examination of YF viruses with polyclonal anti-sera in PRNT's 154

5.1 Results of the immunoprécipitationtechnique 163

5.2 Results of the ELISA technique to generate antigenic variants toYF17D-204 WHO 165

5.3 SNI's of four YF17D-204 vaccine virus strains in reactions with neutralisingMcAbs 168

5.4 Titre of McAb* variants plaquepurified from PRNT's 169

5.5 Log Q infectivities of derived McAb* variants in PRNT's with the McAbswhich selected them 170

5.6 Comparison of indirect immunofluorescence tests between the YF17D-204 UK parent viruses and the derived YF17D 864*variant strains 173

5.7 Comparison of indirect immunofluorescence tests with the YF17D-204 WHO parent virus strain and the YF17D B39*variants 174

XIX

5.8 Results of indirect immunofluorescence tests with the YF17D B26* variantsand their parent viruses 176

5.9 Reaction of McAbs in Haemagglutination inhibition assays (HAI) with parent vaccine strains and derived 864*variants 177

5.10 Reaction of McAbs in Haemagglutination inhibition assays (HAI) with parent vaccine strains and derived B39*variants 179

5.11 Reaction of McAbs in PRNT’s with the YF17D-204 UK virus strains and thederived McAb 864* variants 181

5.12 Reaction of McAbs in PRNT's with the YF17D-204 WHO virus strain and thederived McAb B39* variants 183

5.13 Comparison of the pathogenicity of the YF17D-204 parent strains with the864* variant strains in adult mice 185

5.14 Comparison of the pathogenicity of the YF17D-204 WHO strain with the B39*variants in adult mice 187

XX

5.15 Comparison of the pathogenicity of the YF17D-204 UKpp- parent virus strain B with the B26* *E’ variants in adultmice 188

5.16 Comparison of the pathogenicity of the YF17D-204 parent strains with theB26* variants in newborn mice 189

6.1 Comparison of the USSR vaccine in IIFtests with other YF vaccine strains 201

6.2 Comparison of the USSR vaccine in HAI assays withother YF virus strains using the A, B, and H series ofMcAbs and polyclonal anti-sera 205

6.3 Comparison of the USSR vaccine in HAI assays with other YF virus strainsusing the 'F ' series of McAbs 206

6.4 Comparison of the USSR vaccine in PRNT'swith other YF virus strains usingMcAbs which elicit neutralisingactivity 208

6.5 Comparison of the USSR vaccine to other YF virus strains in pathogenicitystudies using adult mice 210

6.6 Comparison of the USSR vaccine to otherYF virus strains in pathogenicitystudies using newborn mice 213

XXI

5.7 Infectivity titres of plaque purified preparations of the USSR vaccinevirus 215

6.8 Comparison of the plaque purified USSR viruses to that of the parentvaccine strain by IIF tests 217

6.9 Comparison of the plaque purified USSR viruses to that of the parentvaccine strain by HAI assays 219

6.10 Comparison of the plaque purified USSR viruses to that of the USSR vaccine in pathogenicity studiesusing adult mice 220

6.11 Comparison of the plaque purified USSR viruses to that of the USSR vaccine in pathogenicity studiesusing newborn mice 222

6.12 Comparison of USSR vaccine passaged in the mouse brain (Mb) with that passaged first in SW13 cells byIIF tests 224

6.13 Comparison of USSR vaccine passaged in the mouse brain (Mb) with that passaged first in SW13 cells byHAI assays 226

XXI!

6.14 Comparison of USSR vaccine passagedin mouse brain (Mb) with that passaged first in SW13 cells in a pathogenicity study using adult mice 227

6.15 Comparison of USSR vaccine passaged in the mouse brain (Mb) with that passaged first in SW13 cells in a pathogenicitystudy using newborn mice 229

7.1 Examination of the FNV vaccines in HAI assays with their wild-type parent YF strain FVV using the A, B, G, and Hseries of McAbs 237

7.2 Comparison of FNV vaccines using 'F ’McAbs in HAI assays with otherstrains of YF virus 238

7.3 Comparison of reactivities of the FNV vaccines in PRNT's with other YF viruses using YF type specific, flavivirussubgroup and group specific McAbs 242

7.4 Comparison of the FNV vaccines with their wild-type parent strain (FVV) in a pathogenicity study usingnewborn mice 244

8.1 Temperature sensitivity of YF virusesgrown in the C6-36 cell line 253

XXIII

8.2 E.O.P of viruses grown in SW13 cells which were temperature sensitive whengrown in C6-36 cells 254

8.3 Antigenic analysis of the 17DD Brazilpp vaccine virus when grown in SW13 andC6-36 cells by HAI assays 258

8.4 Antigenic analysis of the 17DD Brazilpp vaccine virus when grown in SW13 and06-36 cells by PRNT's 260

9.1 PRNT's of the McAb 864 with YF virusesgrown in SW13 and C6-36 cells 267

9.2 Reactivity of flavivirus subgroup and group McAbs in PRNT's with virus grownin SW13 cells 269

9.3 Reactivity of flavivirus subgroup and group McAbs in PRNT's with virus grownin C6-36 cells 270

XMV

LIST OF FIGURESPAGE

1.1 Derivation of the FNV vaccine strains 201.2 Derivation of yellow fever 17D vaccines 231.3 YF genome organization and proteins encoded 331.4 Structure of the TBE virus E-Protein 751.5 The structure of the YF virus genome

and proteolytic processing of theYF polyprotein 77

3.1 Autoradiograph of PAGE gel of separatedvirus polpeptides 121

4.1 Cluster analysis of all 34 viruses with twenty seven McAbs which reacted withall viruses in the study 142

4.2 Cluster analysis of viruses without plaque purified viruses with the twenty seven McAbs which reacted with allviruses in the study 144

4.3 Cluster analysis of 17D-204 and wild-type Asibi viruses with the twenty seven Mcabs which reacted with all viruses inthe study 145

5.1 Passage history of USSR vaccine 1996.2 IIF tests on vaccine virus infected

Vero cells 203

XXV

6.3 Pathogenicity of the USSR vaccine when an innoculum of 200 pfu is inoculated intracerebrally 211

7.1 Cluster analysis of the HAI resultsusing the ’F ’ McAbs series 240

8.1 Passage history of the 17DD Brazilplaque purified preparation 257

XXVI

LIST OF ABBREVIATIONSADEASTCBAscm^CPEcysE.G.PE-proteinERHAHAIHAUIIFMcAbMcAb^

mg ml mM MO I NnmN°.NSPAGEPBSELISA

antibody dependent enhancement average survival time competitive binding assays centimeter squared cytopathic effect cysteineefficieincy of plating envelope protein endoplasmic reticulum haemagglutination haemagglutination inhibition haemagglutination unit indirect immunoflourescence monoclonal antibodymonoclonalresistant

antibody neutralisation

milligrammillilitremillimolarmultiplicity of infectionneutralisationnanometersnumbernonstructuralpolyacrylamide gel electrophoresis phosphate buffered salineenzyme-linked immunosorbent assay

xxvii

LIST OF ABBREVIATIONS (CONTINUED)

pfu plaque forming unitpp plaque purifiedPRNT plaque reduction neutralisatiom testRNA ribonucleic acidmRNA messenger RNARPM revolutions per minuteS/N supernatantSNI serum neutralisation indexTCID tisuue culture infectious dosets temperature sensitiveul microlitreWHO World Health OrganisationWT wild-typew/v weight per volume

Viruses

DEN DengueJE Japanese encephalitisSLE St. Louis encephalitisTBE Tickborne encephalitisWN West NileYF yellow fever

XXYlll

ACKNOWLEDGEMENTS

I would like to thank my supervisor. Dr. AlanBarrett for his untiring help and guidance to obtain one of the goals in my life. Also for help with the RIPs.

I would like to thank my mother and father whoprovided the opportunity of a good education whichenabled me to undertake this study.

Also my thanks go to all my collègues in the YFU for providing the social diversion from work. Also I must thank especially: Bijon for providing the 'H 'McAbs, Lee for undertaking the begining of the temperature study and collaborating in the mouse experiments. Last but not least Mark for his endless knowledge of computers and the Clustan programme.

I also thank David Woods for undertaking themonkey experiments. Also my thanks go to the staff of the experimental biology unit of the University of Surrey.

I am grateful to the yellow fever vaccine manufacturers for donating samples of their vaccines for use in this thesis.

I would like to thank Drs. M.K. Gentry, E.A. Gould, Roehrig, J.T. and Schlesinger, J.J. for donating their McAbs for use in this thesis.

XXIX

My thanks also go to the Science and Engineering Research Council (SERC) for providing the grant for my studies.

I would also like to thank Rita Sharman and Antoinette Williams for typing some of the figures and tables in this thesis.

Finally, as 'no man is an island' I would like to thank the staff and students of the Microbiology department who have helped me in one way or another.

XXX

CHAPTER 1

LITERATURE REVIEW

1. YELLOW FEVER VIRUS

I. Historical Considerations of the Virus

In all of the ancient medical literature there is no single word written about the disease yellow fever(YF), by that or any other name. Since epidemic YF is adramatic disease due to the quickness that acutelyinfected people die and the symptoms of the disease(e.g "black vomit"), it seems unlikely that it could have escaped the attention of even very early medical writers.

It was not until the 17th century, that YF wasdescribed as a disease entity and the term 'yellowfever' was first employed by Hughes in 1750 (Warren,1951).

A) The Vector

Until the 19th century, the disease was widely thought to be caused by airborne miasmas. Although Nott, in 1848 (Warren, 1951) suggested the mosquito as apossible agent for the dissemination of both YF andmalaria and Beauperthuy in Venezuela made a similar suggestion in 1854. However, the first serious proponent of mosquito transmission was Dr. Carlos J. Finlay (Warren, 1951; Amster, 1987) who presented evidence that

YF was spread by a specific mosquito, Stegomyia fasciata now known as Aedes aegypti. Finlay’s theory was accepted by Lazear who carried out transmission experiments with infected mosquitoes on human volunteers. These were successful and further experiments were then repeated in 1902 by Reed and Carroll (Warren, 1951) of the Yellow Fever Commission who demonstrated transmission of YF to volounteers by mosquitoes (A. aegypti) which had been previously fed on clinically ill patients proving vector transmission. They also demonstrated by direct human- human passage that the aetiologic agent was present in the blood of infected patients, and defined the period of extrinsic incubation in the mosquito, showing that a filterable agent was responsible for the disease. Although Reed’s observations were confirmed by many other workers, the filterable agent could not be isolated, propagated, established or characterized.

Until the 1930s YF was considered essentially an urban disease transmitted only by A. aegypti. This concept was strengthened by two observations. First, Marchoux and Simond (1906 a, b) failed to transmit YF to five different mosquitoes; A. scapularis, A. taeniorhynchus, Culex quinguefasciatas, Psorophora ciliata, and P. posticata and second, the dramatic disappearance of YF from all known centres of infection on the American continent due to the application of

anti-mosquito measures against A. aegypti. However, an early report by Franco et (1911) suggested YF wasacquired in the forest by mosquitoes other than A. aegypti. Bauer (1928), and Philip (1929, 1930) in West Africa and Davis and Shannon (1929, 1931) in SouthAmerica showed that other species of mosquitoes could transmit YF virus from animal to animal. The concept of jungle YF was only clearly defined by an outbreak of YF in Brazil in 1932, in the absence of A. aegypti, (Soper et al., 1933). Shannon et (1938) reported theisolation of YF virus from the Haemagogus species and the role of this genus in virus transmission to humans working in the forest was subsquently established (Bugher et al., 1944). Haddow et al. (1948) in East Africa showed that A. africanus was responsible for virus circulation between monkeys in the forest canopy and A. simpsoni for transmission from monkeys to humans. Cornet et (1979) documented the role of A.luteocephalus and A. (Diceromyia) furcifer-taylori group as principal vectors in the savannah belt of Africa and isolated YF virus from male mosquitoes. This showed that transovarial transmission as a possible mechanism of virus survival. Germain et al. (1979) reported the isolation of YF from ticks (amblyomma variegatum) collected from cattle in the Central African Republic and raises the interesting possibility that ticks have a

role in dry season maintenance or dispersal of the virus (Monath, 1984).

B ) The Agent

A number of workers at various times claimed to have discovered the agent of YF, although the only theory to receive attention in the 19th century was that of Sanarelli, a distinguished bacteriologist, who claimed that a bacillus (Bacillus icteroides) was the causative agent. Although his claim was taken seriously, it was disproven by Reed's yellow fever commission (Warren, 1951).

History again repeated itself when another eminent bacteriologist claimed the agent to be a Leptospira. Noguchi (1919) investigating a YF epidemic in Ecuador showed that leptospira icteroides produced lesions suggestive of YF in guinea pigs (Noguchi, 1925). He had prepared a vaccine against this organism, stating this gave protection against YF for six months.

However, in 1925, workers at the West African Yellow Fever Commission were unable to isolate the leptospira from YF. Then in 1927, Stokes ^ (1928),succeded in isolating the agent by inoculation of blood from patients into rhesus monkeys. Then isolating the agent and passing to another monkey. Thus, the agent now

known as YF virus, was isolated from the blood of a Ghanaian man named Asibi and later served as the parent to the 17D vaccine (Theiler and Smith, 1937a,b) . In the same year Mathis et (1928) isolated a virus in Dakarand established the French strain (FW) which became the parent to the French neurotropic virus (FNV).

II. Geographical Distribution

A) America

YF was first recognized in the Americas, occurring in Yucatan in 1648 (Blake, 1968). Although the exact place of origin of YF is uncertain, there is agreement using immunological, entomological and historical data, indicating that the virus was transported to the 'New World' from Africa (Strode, 1951; Blake, 1968) around the time of Christopher Columbus by sailing ships (Portefield, 1989).

During the 18th (Blake, 1968) and 19th (Duffy, 1968) centuries YF was one of the most important epidemic diseases in America. The disease was periodically imported by ships into ports of the Carribean, Central America and North America causing epidemics from 1693 to 1905. For example, in New Orleans

in 1853, a tenth of the population died of YF (Duffy, 1966) and again in 1905 443 persons died of YF(Bendiner, 1988). The virus even spread to Europe where during the 1800’s Madrid had 60,000 deaths (Theiler,1952) and Swansea had 27 deaths (Smith and Gibson, 1986; Meers, 1986; Porterfield, 1989).

With the introduction of anti-mosquito campaigns, for example, Gorgas from 1901 to 1908 (Patterson, 1989) and the 17D vaccine in the 20th century the incidence of YF epidemics was reduced and in the 1940's resulted in the loss of Urban YF outbreaks.

Since the 1940's outbreaks have occurred in South America exclusively of Jungle or Sylvan YF, the tree living Haemagogus mosquito being the responsible vector. There are four major countries in South America where outbreaks of Jungle YF occur regularly and account for 95% of the human cases; Peru, Bolivia, Colombia and Brazil, (Monath, 1989). Other countries where outbreaks have occured are Paraguay where there were 9 cases reported in 1974, Argentina in 1966, Eucador where there were 29 cases between 1965 and 1984, 49 cases werereported in Venezuela between 1965 and 1984, and lastly Trinidad and the Guianas. Sometimes the varying distribution of human cases and the periodicity of outbreaks reflect the actual movement of YF virus. The

most studied of these epizootic waves began in 1947, in Eastern Panama (Boshell-Manrique, 1959) and crossed the Panama Canal in 1950 reaching Mexico in 1958 but was not reported by that Government (Trapido and Galindo, 1956). Also, YF virus was isolated from migrant workers (Mendez et al., 1984) in the YF outbreak in the Apurimac Rivervalley of south central Peru in 1977, for the first time.

Recently, there have been a number of sporadic endemic cases notified in South America in 1986 and 1987, especially in the countries of Peru, Bolivia and Brazil (Weekly Epidemiology Record, 1989a,b). Also, in 1989, the Brazilian Ministry of Health advised travellers to be vaccinated against YF due to the occurence of jungle YF in 1988. Lewis, ^ aJ. (1989) and Rawlins et (1990), showed there had beensylvatic yellow fever in Trinidad between 1988 and 1989, though no human infection was reported.

B ) Africa

The first documented epidemic of yellow fever in Africa was in 1778 (Freestone, 1988), much later than those reported in the Carribean or the Americas. Serological surveys carried out in practically all parts of Africa since 1932 have delineated the boundaries of

the area in which the disease occurs. YF is generallyaccepted to be endemic or epidemic in 29 countriesbetween latitudes 15°N and 10°S (WHO, 1971).

Before the 1940s and mass immunization campaigns YF occurred in Africa as either sporadic cases of jungle YF mainly in forested areas or urban outbreaks mainly in the savanna areas. Typical urban outbreaks occurred in Nigeria, 1925 - 1926, Ghana, 1926 - 1927, and also in 1937, the Gambia, 1934 - 1935 (WHO, 1986). The first recognized epidemic in East Africa ocurred in Sudan in 1940 and was very severe. There being 15,641 cases and 1627 deaths reported among a population of 230,000.Subsequent serological surveys estimated that there hadbeen approximately 40,000 infected people with a death rate of 10% (Kirk, 1941).

From 1940 until the 1960s there was a mass vaccination programme with the French neurotropic vaccine in the french speaking countries, whereby 25 million people were immunized every 4 years. This eradicated YF from these areas (Durieux, 1956). With the loss of interest in YF in the 1960s and immunization neglected epidemics reappeared. The first in Ethiopia in 1960 - 1962 when 3000 deaths were notified. However, it was estimated that as many as 100,000 cases and 30,000 deaths occured in this area where the population

numbered one million people (Serie, 1968). Outbreaks have also occured in Senegal, 1965, Ethiopia, 1966,Ghana and Mali in 1969, and Nigeria, 1969 to 1970numbering an estimated 100,786 cases (Carey et al., 1972; Monath et ^ . , 1973), Burkino Faso, 1969 and 1983, Angola, 1971, Sierra Leone in 1975, Ghana in 1977 - 1979 and 1983, and the Gambia from 1978 - 1979 where therewas 8400 cases and 1600 deaths estimated (Monath et ,1980).

More recent epidemics have occurred in Mali in 1987 whereby 304 cases and 144 deaths were notified (Meunier, 1988; Weekly Epidemiological Record, 1989). Also, in Nigeria between 1986 and 1987 there was 5012 cases and 1367 deaths reported in the first urbanoutbreak of YF for 40 years (DeCock et ^ . , 1988; DeCock and Monath, 1988; Weekly Epidemiological Record, 1989a,b). However, an estimated total of 7500 deaths are thought to have occurred, with over 20 million inhabitants at risk (Porterfield, 1989; Nasidi et al., 1989).

Ill. Clinical description and pathogenesis of the disease YF

A) HumansIn humans, YF virus produces a viscerotropic

1 0

disease. YF shares many clinical features with other viral hemorrhagic fevers (dengue hemorrhagic fever, Crimean-Congo hemorrhagic fever and lassa fever) but it is characterised by more severe hepatic involvement than the other viruses (Monath, 1987). YF varies in severity according to the immune status of the individual, from a asymptomatic mild or moderate case to a classic severe case. Classical (Kerr, 1951) and recent reviews (Monath, 1987, Freestone, 1988) provide detailed descriptions of the clinical and pathological aspects of YF.

i ) Asymptomatic Infections

Symptoms are limited and begin with a headache, fever, nausea and vomiting, though conjunctival infection and albuminuria can also occur. However, patients usually recover in a few days.

ii) Classic Severe Cases

Clinical descriptions of severe YF, emphasize three stages: infection, remission, and intoxication.The period of infection lasts about three days in which the patient suffers abruptly from fever, severe headache, generalised malaise and weakness, lumbosacral pain, nausea and vomiting. Despite a persistent or rising severe fever^the pulse rate may decrease (Faget's

11

sign). During this phase virus is present to a high titre in the blood and the patient is infectious to the bite of mosquitoes. Then a period of remission (a period of mitigation of symptoms) follows which may last from a few hours up to 24 hours. Then the fever and symptoms reappear announcing the period of intoxication. This is shown by the return of more frequent vomiting, epigastric pain, prostration, dehydration, and the appearance of jaundice. The bleeding diathesis appears as coffee-grounds haematemesis ("black vomit"), melena, metrorrhagia and diffuse oozing from the mucous membranes. Virus is usually absent from the blood and antibodies appear during this phase. Approximately 50% of patients in this phase have a fatal outcome between the 7th and 10th day heralded by deepening jaundice, hemorrhages, tachycardia, and hypotension,encephalopathic signs and azotemia. Hypothermia, agitated delirium, intractable hiccups, stupor and coma are terminal signs.

B ) Monkeys

Humans or nonhuman primates infected by peripheral routes show a vicerotropic disease, though if inoculated by the intracerebral route show both encephalitic and viscerotropic disease. Stokes et al. (1928) were the

12

first workers to investigate the susceptibility of monkeys to YF, They showed that YF virus can be successfully transmitted to the indian Macacus rhesus monkey and that the clinical course of the disease and the lesions were similar to those found in human YF cases. Subsequently workers have studied other monkey species. Davis, (1930a,b) was able to demonstrate a certain degree of susceptibility of seven (the Cebus monkey, the squirrel monkey, the spider monkey, the woolly monkey, red howling monkey, true marmosets, and the black tamarin) South American monkeys to viscerotropic YF virus, while Lloyd and Penna (1933) studied the susceptibility of the same South American monkeys to the neurotropic mouse-brain-adapted YF virus. Bugher (1951), reviews the work carried out by many workers on the susceptibility of different monkey species from those of Africa and South America. Several workers showed that West African monkeys (except for the North African macacus and chimpanzee) were refractory for YF even though the virus replicated in the monkeys, suggesting signs of susceptibility to the virus. None of the West African monkeys showed any outward signs of illness. However there was a range of susceptibility of the Galago species to YF. The G. demidovii demidovii where no outward reaction was seen to inoculation of YF virus, to that of G. crassicaudatus lasiotis were they

13

died following YF inoculation. All species of cercopithecid and colobid monkeys have been shown to be viremic hosts circulating YF virus for many days. While infection of African monkeys infrequently leads to death, many of the South American monkeys died when inoculated with YF.

Monath et aJ. (1981) studied the course ofYF virus following inoculation of the rhesus monkey and showed that symptoms of fever and lethargy appeared 14 to 25 hours before death. In the last six hours acidosis, hyper-kalemia, hypotension, hyperthermia, and coma developed. Viraemia was first detected on the 2nd day post-infection and progressively increased in titre until the fifth day when there was a slight fall. No anti-YF virus antibodies developed and post-mortem samples showed there was total destruction of the hepatic structure due to acute necrosis of hepatocytes and Kupffer cells, and Councilman bodies were present. Post-mortem samples of the kidney showed varying degrees of degeneration of tubular epithelia cells and samples from the spleen and lymph nodes showed acute lymphoid necrosis.

Histopathological changes in the myocardium have been found in experimentally infected monkeys (Lloyd, 1931) adversely affecting the myocardial performance. This probably contributes to the rare phenomenon of late

14

death in humans at the end of convalescence or even weeks after complete recovery from the acute illness.

C) Mice

In 1930, Theiler demonstrated that white adult mice are susceptible to YF virus when inoculated intracerebrally and, in most cases, are resistant to intraperitoneal inoculation. These findings were soon confirmed by other workers and led to these animals being extensively used for YF studies. Theiler (1930), also showed that the passage of YF virus through mice leads to a loss of virulence for monkeys, although a progressive shortening of the average survival time (AST) is seen in mice (Theiler, 1951). On the other hand, newborn mice up to the age of two weeks old were shown to be susceptible to intraperitoneal as well as intracerebral inoculation of wild-type YF viruses and the symptoms were the same as those seen in adult mice (Theiler, 1930).

Recently, workers have demonstrated that when adult mice are inoculated intracerebrally with the 17D vaccine they have a similar average time as the parent wild-type Asibi strain and the viruses cause a neurotropic disease in mice (Fitzgeorge and Bradish, 1980a,b; Barrett and Gould, 1986a; Gibson et al., 1990).

15

D ) Guinea Pigs

Sellards (1930) and Sawyer and Frobisher (1932) (Theiler, 1951) showed that after intraperitoneal inoculation of YF virus, Guinea pigs develop a viraemia and antibodies to YF virus. Guinea pigs show no signs of illness, unless they are inoculated intracerebrally with the wild-type Asibi strain of YF virus in which case encephalitis develops (Theiler, 1933) and the animals died after an incubation period of 5 to 14 days.

E ) Rabbits

Whitman (1935) and other workers (Theiler, 1951) tested the susceptibility of the rabbit to a number of wild-type strains of YF virus. Following inoculation by either the intraperitoneal or intracerebral route it was found that the virus was at no time present in the spleen, liver, kidney, or bone marrow. Whitman, (1935) showed the animals readily developed antibodies in response to an immunizing inoculum of virus. Moreover, after a rest period of 6 weeks, they responded to a second injection by producing high titres of serum antibody.

F) HedgehogsFindlay and Clarke, (1934) showed that both the

European hedgehog (Erinaceus europaeus) and the Sudanese

16

hedgehog (E. pruneri) are both extremely susceptible to YF virus. When E. europaeus was inoculated with the Asibi strain it died between 4 to 7 days post infection. Post-mortem examination showed it had macroscopic and microscopic lesions similar to those seen in the rhesus monkey. Thus, other than humans and monkeys, hedgehogs represent the only other host in which YF virus causes viscerotropic disease. Following the initial studies of Findlay and Clarke little work has been done on YF infection of hedgehogs, presumably because of the difficulty of both obtaining and working with this animal.

G) Mosquitoes

In South America the haemagogus tree-hole living mosquitoes are of primary importance as vectors in the sylvatic cycle of YF virus spreading the virus among the monkey population. The haemagogus breed in tree-holes and during midday they feed in the forest canopy, though they are known to bite and infect humans with the YF virus in forest clearings and even inside houses near the forest (known as Jungle YF). However, in Africa there are two principal species involved in sylvatic infection (mosquito to monkey to mosquito transmission). These are the Aedes africanus which is responsible for year-round transmission of YF virus in the equatorial

17

African forests and Aedes simpsonii, now known as A. bromeliae, which links the forest cycle with humans. The principal mosquito vector in urban YF outbreaks in Africa is the 'domesticated’ Aedes Aegypti. However, in Africa significant outbreaks have occurred in the absence of A. Aegypti and have involved A. africanus and A. simpsonii

V ) Yellow fever vaccines

A) Vaccine development

Highly effective vaccines against YF have been available for over 50 years and along with smallpox and rabies vaccines, are among the very first vaccines to be developed against human disease. Two live attenuated vaccines were independently developed and are named the French neurotropic vaccine (FNV) and the 17D vaccines.

i ) French Neurotropic Vaccine virus

Attempts to produce a live vaccine strain of YF virus suitable for human vaccination were made at the Institut Pasteur, in Dakar, Senegal. The wild-type parent virus (termed the French viscerotropic virus [ F W ] ) to the vaccine was isolated from a Syrian patient named Françoise Mayali in 1927 and was attenuated by intracerebral passage in adult mice (Mathis et al..

18

1928). This attenuated virus was called the French neurotropic vaccine (FNV). The virus (FNV) had lost the ability to cause visceral YF in monkeys after extraneural inoculaton but had increased neurotropic virulence for both monkeys and mice. The first attenuated variant was identified at passage 128 and the first vaccine was used at passage 237. Later vaccines were used at the 260^^ passage. The derivation of the FNV vaccines is shown in FIGURE 1.1. The FNV vaccine was used extensively in french speaking countries of Africa between 1939 and 1980 where more than 80 million people were successfully immunized by scarification. The last vaccination took place in Malawi in 1980. Although this vaccine was more stable to heat than the 17D vaccine, reactions have been reported by Peltier (1947) and Macnamara, 1953, who reported the cases which evolved from the mass vaccination in Nigeria in the epidemic of 1951 - 1952 where 42,000 persons were vaccinated. There were 83 cases of encephalitis of which 32 died. The incidence in children was between 0.3 to 0.4% having a fatality rate of abut 40%. This led to the vaccine being discontinued in Nigeria and led to its prohibition of use in children below the age of 14 years old by the W.H.O. As a result the vaccine lost popularity and was discontinued in 1980.

19

FIGURE 1.1 DERIVATION OF THE FNV VACCINE STRAINS(ADAPTED FROM BARRETT, 1987)

F W (MAYALI) (Mathis et al., 1927)

ORIGINAL FNV (pl28)

FIRST VACCINE (p237)

LATER VACCINES (p260)

FNV-FC F N V - I P IF N V - N T

N.B. ALL FNV VACCINES WERE DISCONTINUED IN 1980 DUE TO REPORTED CASES OF ENCEPHALITIS

2 0

ii) 17D vaccine virus

The wild-type strain, from which the the 17D vaccine was derived, is called Asibi, after a patient in Ghana from whose blood the virus was isolated. The isolation was achieved by monkey to monkey passage. The attenuation process was undertaken by Theiler ^ al. (1937 a,b) who noticed that passage of this virus in mice reduced its virulence for monkeys following peripheral inoculation. These workers passaged the 18 mouse embryonic tissue culture virus, into whole chick embryo for fifty times. Thereafter, this medium was modified to whole chick embryo tissues from which head and spinal cord had been removed. By the 176^^ passage, these workers noticed that this strain, designated 17D had lost its ability of producing fatal encephalitis when injected intracerebrally into rhesus monkeys. It was also considerably less neurotropic for mice than the 114^^ passage. It is noteworthy that other isolates at the same passage (17E etc.) were not sufficiently attenuated nor could Theiler ever repeat the attenuation process (Theiler and Smith, 1937 a, b; Theiler, 1951).

The virus was shown to produce a safe and effective vaccine after the 204^^ passage. The latest 17D vaccines are derived from two distinct 17D substrains: 17D-204 and 17DD which are both independentlymaintained passage series from the original 17D. The

21

17D-204 sub-strain is derived from the 204^^ passage of Asibi virus and is now at the 233 * - 239^^ passage. The 17DD sub-strains are derived from 17D at passage 195 and subsequent passages were performed in chick embryo tissue or embryonated chick eggs to produce the 17DD vaccines which are now at passage 286 - 288. Thederivation of 17D vaccines are shown in FIGURE 1.2.

Preliminary tests of 17D vaccine were performed in YF-immune subjects and then subsequently in non-immune volunteers (Theiler and Smith, 1937a). From 1937 to 1938 the 17D vaccine produced in chick embryos was first used in vaccine trials in Brazil where some 59,000 people where vaccinated (Barrett, 1987). This vaccine was used increasingly in Latin America and then throughout the world. This lead to multiple passages of the 17D vaccine virus being maintained in different laboratories and vaccine lots were produced as needed from the current passage level of the virus available at the particular laboratory.

By 1943, however, significant problems with the 17D vaccine were encountered. The first, was that certain high level passsaged cultures of 17DD vaccine (300 - 350 passages from Asibi) failed to immuninize.However, returning to an earlier passage of virus for vaccine manufacture overcame this problem (Theiler, 1951). Another vaccine batch in Brazil showed increased

2 2

FIGURE 1.2 DERIVATION OF YELLOW FEVER 17D VACCINES'

ASIBI(THEILER ET AL., 1927, GHANA)rORIGINAL 17D (pl76)

— 17DD ' LOW (pl95)17D-204

RFSfs RF312 COLOMBIA 88

COLOMBIA SOUTH UNITED INDIA AFRICA KINGDOM

UNITED STATES OF AMERICA

FRANCE-SENEGAL

AUSTRALIAHOLLAND(p233 - p237)

WESTGERMANY

WHO ALV-free

AUSTRALIA' WEST GERMANYUSSR

(p238 - p239)(p243>

17DDHIGH

BRAZIL

SENEGALCOLOMBIA (p286 - p288)

ADAPTED FROM MONATH ^ , 1983; BARRETT, 1987

23

neurovirulence for children, where 0.36% of vaccinees showed symptoms of central nervous system disturbances. The degree of encephalitis depended on the vaccine substrains examined (Fox et ^ . , 1942). The thirdproblem was the occurrence of serum hepatitis in vaccinees in England and Brazil and American military personnel (Strode, 1951). This was shown to be due to human serum being incorporated into the vaccine to aid its stability. The problem was overcome by omitting the serum and increasing the concentration of egg protein. The development from 1945 onwards of a seed lot/batch system has been adopted to control the passage level and production of vaccine (Barrett, 1987). Since then 1000 million doses of vaccine have been distributed. While no case of viscerotropic disease has been observed, there have been fifteen cases of encephalitis reported worldwide and occurred in infants seven months or younger in age (Anonymous, 1966). However one case of lethal encephalitis has been repoted in a child 39 months old who recieved the same 17D-204 USA vaccine batch 6145, as her parents. At the autopsy after her death, a virus was isolated from the brain and shown to be serologically YF virus (Anonymous, 1966).

All vaccines manufactured today are YF 17D vaccines and both 17D-204 and 17DD sub-strains are used. The vaccine is currently manufactured in eleven

24

countries with the approval of the WHO Two countries have recently discontinued production (Holland and Colombia) while two other countries have just started production (Nigeria and USSR) (WHO, 1981; WHO, 1988). The WHO controls the production of YF vaccine, setting the preparation techniques and control tests for neurotropism and storage stability, as well as heat stability tests. The vaccine is still produced in embryonated chick eggs using the the same basicprocedures laid down in 1945. However, contamination of the YF vaccine with the avian leukosis virus (ALV) was shown (Waters, 1972) and has led the WHO to develop an ALV-free vaccine seed strain in West Germany. This was from a 17D-204 sub-strain of the virus at passage 236 and now this vaccine is presently at passage 239. This ALV-free vaccine strain is only produced by a few manufacturers (Australia, West Germany, USSR). The manufacturers have been reluctant to use this new seed as it causes more histological lesions in the brains of monkeys following the WHO safety test than the established vaccines (A.D.T. Barrett, pesonalcommunication). Currently, the 17DD substrain is only manufactured in Brazil. It is of interest that Brazil has the highest annual output of vaccine (12 million out of a total of 20 million doses produced annually,Barrett, 1987) but is still not ALV-free.

25

The monkey today is used by vaccine manufacturers to test the neurovirulence of new vaccine virus batches as recommended by the WHO (WHO Technical Report Series,1981). The monkeys should be Macca Mulatta (Macacus rhesus) species or be of equal sensitivity. The monkey is inoculated with not less than 5000 mouse LD^q and not more than 50,000 mouse in a 250ul volume. The virusinjected in the frontal lobe under anaesthetic and the monkeys are observed for 30 days by personnel familiar with encephalitis in primates. The vaccine virus must not cause severe signs of encephalitis (paralysis, inability to stand, or death) in not more than 10% of the monkeys.

Thermostability of the vaccine is a problem due to exposure at high ambient temperatures in Africa and South America. Research has been undertaken on stabilising the vaccine to heat. It has been reported that lactose, sorbitol, histidine and alanine considerably improve heat stability of lyophilised preparations (Barme & Bronnert, 1984) and it was shown in field conditions to be very stable (Georges et al., 1985) with no loss of efficacy after five months storage at +10°C.

Live attenuated vaccines like YF vaccine virus should not be administered to subjects with an impaired immunological system (i.e. lymphoma, leukemia and AIDS)

26

or patients receiving immune suppressant drugs. Vaccination in pregnant women should be delayed till after birth due to the theoretical possibility of foetal damage. Infants under 6 months of age appear more susceptible to adverse reactions and should not be vaccinated. As YF vaccine is produced in chick embryos and contains chick protein vaccination is contraindicated in a patient with a history of allergy to chick or egg protein.

Once vaccine has been administered, IgM antibodies appear, and decline after 17 days. IgG antibodies appear between days 10 and 17 and remain high. More than 99% of vaccinees seroconvert and immunity seems to last for a long-time (from 30 - 35 years, [Poland ^ , 1981]).Even though immunity appears long term, WHO vaccination certificates are only valid for ten years when revaccination is required.

Only two countries routinely vaccinate all children, these are Venezuela and Trinidad (A.D.T. Barrett, personal communication).

B ) Research on yellow fever vaccines

i ) Mosquito Transmission

17D virus has been found in the blood of vaccinees and this led to worries in the late 1930’s that the 17D

27

vaccine could be repeatedly passaged in mosquitoes and be converted back to its original virulent form. Whitman (1939) studied this and showed that A. aeqypti mosquitoes which fed on vaccinees did not become infected. Since, other workers (Duebel et ^ . , 1981;Miller et ^ . , 1986) have shown both the 17D and FNVvaccines are unable to disseminate to the head tissues of the mosquito, while wild-type virus disseminates and is transmitted (Duebel et ^ . , 1981; Miller and Adkins,1988). Although the reason for non-dissemination is unknown (Barrett, 1987) it is important that it does not occur as there were early fears that mosquito passage may result in the reversion or selection of virulent viruses.

ii) Passage of YF virus in different tissue types

Since the early studies of Theiler and Smith (1937 b) who showed the passage of YF virus in different tissues could modify the virulence of the virus, other workers have tried to modify the 17D vaccine. ’ - : Foxand Laemmert (1947) studied the infection of developing chick embryos with 17D virus and Tannock et (1980)studied the 17D vaccine prepared from decapitated embryos showing that the virus had less thermal stablility than vaccines prepared from whole embryos. Meers (1957, 1959) studied the affects of passaging 17D

28

through mouse brain and showed it became more neurovirulent on each consecutive passage.

More recently workers have passaged the 17D virus in the human tumour KB cell line and reported the virus had lost its neurovirulent character in that it would not kill adult mice when challenged intracerebrally and also protected monkeys on challenge with wild-type Asibi virus (Hallauer, 1959; Schindler & Hallauer 1963). At the same time other workers have shown the passage of Asibi virus in Hela cells results in the loss of virulence for monkeys following intracerebral or intraperitoneal inoculation (Hardy, 1963; Hearn et al., 1968). Workers have now produced the 17D vaccine in tissue culture for vaccination of humans (A.D.T. Barrett, personal communication).

Two yellow fever vaccines 17D-204 UK and 17D-204 USA were reported to be able to grow in human monocytes and the macrophage like cell line U937 (Liprandi & Walder, 1983; Schlesinger & Brandriss, 1981 a,b).Schlesinger and Brandriss in 1983 showed that thirteen IgG McAbs mediated antibody-dependent enhancement (ADE) of YF virus in the mouse P388D^ cell line, and ' showed that the Fc receptor on these cells was important for ADE to occur, ADE has been shown to occur vivo by Barrett and Gould, (1986 b) who showed that McAbs would cause A FNV and Asibi to be enhanced vivo. Recently

29

Gould and co-workers (Gould et , 1987; Gould andBuckley, 1989) have examined the vivo antibodyenhancement of neurovirulence in detail in wild-type and vaccine strains of YF virus. They reported enhancement of neurovirulence even when the McAb was given 3 days post-infection and stated that enhancement was due to the interaction in the brain between virus and antibody without the increase in virus titres.

iii) Virulence of YF vaccines

Comparison of the pathogenicity of the FNV with the 17D vaccines by Theiler and Smith (1937a,b) showed that although both FNV and 17D have lost the ability to cause viscerotropic disease, FNV was more neurotropic than 17D. Following intraspinally or intracerebrally inoculation of FNV, monkeys died of encephalitis, while the 17D vaccine caused a mild encephalitis which was only occasionally fatal. However, following extraneural inoculation the FNV vaccine caused a febrile reaction in 50% of monkeys and 30% had a lethal encephalitis whereasthe 17D vaccine is not fatal.

Due to the expense and scarcity of monkeys, most workers now use the mouse model to investigate thebiology of YF virus, Fitzgeorge & Bradish (1980a) andGibson et (1990) have shown that the FNV virus killsadult mice much quicker than the 17D vaccines as well as

30

its parent F W wild-type virus when inoculated intracerebrally. In comparison, mice infected with the 17D vaccine and its wild-type parent virus Asibi have similar average survival times (AST’s) (Fitzgeorge & Bradish, 1980a; Barrett and Gould, 1986 a; Gibson et al., 1990). It has been shown that intraperitonealinoculation of YF virus into young mice will either be fatal or not depending on the age and virus used. Wild- type F W virus will kill 8-10 day old mice and Asibi will kill a similar age group (10-12 day old), whereas 17D will only kill mice upto 5 days old but the FNV vaccine kills mice upto 21 days old (Fitzgeorge & Bradish, 1980a). Deubel ^ (1986) reported similarresults following inoculation of mice which were eight days old; while mice infected with the 17D vaccine and F W had long AST’s (8-10 days), Asibi and the FNV vaccine had shorter AST’s (5 - 6 days). Two recentstudies (Barrett & Gould, 1986a; Gibson et ^ . , 1990)have distinguished the FNV, two out of three 17DD vaccines and the wild-type Asibi from the 17D-204 vaccines by intranasal inoculation of mice, showing that the latter were not virulent. Not surprisingly, passage of the FNV in mice has resulted in the virus becoming highly neurotropic in both monkeys and mice compared to its parent and other vaccine strains.

The discovery of 17D-204 variants differing in

31

virulence by Liprandi, (1981), was an important biological finding, and has been followed up by Gould et al., (1989) who isolated variants of the 17D-204 vaccine virus having the wild-type epitope present recognised by McAb 117 and showed it to be more virulent than other plaque variants without the epitope recognised by McAb117 in weanling mice.

iV. Molecular analysis of YF vaccine viruses

Rice et (1985) published the entire sequencefor the YF17D-204 USA vaccine from the American Type Culture Collection. The genome was found to consist of 10,852 nucleotides and to be a single long open reading frame of 10,233 nucleotides with non-coding regions of118 and 511 nucleotides at the 5’ and 3' regions respectively. The structural proteins, C (capsid), M (membarane), and E (envelope) were discovered to be coded by the 5’ end of the genome, while the last three quarters of the genome coded for the non-structual proteins. By partial N-terminal amino acid sequencing a total of ten polypeptides have now been shown to be coded for by the YF virus genome in the order as shown by FIGURE 1.3. The functions and molecular weights of these ten polypeptides is listed in TABLE 1.1.

32

FIGURE 1.3 vYF GENOME ORGANISATION AND PROTEINS ENCODED(ADAPTED FROM RICE ET AL., 1985)

YF17D GENOME (10,862nt)

118nt 10,233nt 511ntCAP STRUCTURAL NONSTRUCTURAL

ns2afC| prM NSI ns2b NS3 ns4a ns4b NS5

33

TABLE 1.1 FUNCTIONS OF YF VIRUS PROTEINS

APPROXIMATEPROTEIN MOLECULAR WEIGHT GLYCOSYLATED FUNCTIONSTRUCTURAL

prM

M

13.000 - 16,000

19.000 - 23,000

8,000 - 8,500

no Nucleocapsidprotein

yes precursor toM

no virion envelope protein

51,000 - 60,000 both forms major virionenvelope

NONSTRUCTURALNSl 44,000 - 49,000 yes

ns2a 16,000 - 21,000 nons2b 12,000 - 15,000 noNS3 67,000 - 76,000 no

ns4a 24,000 - 32,000 nons4b 10,000 - 11,000 noNS5 91,000 - 98,000 no

(ADAPTED FROM RICE ET AL., 1986)

protein

soluble complement-

fixing antigenunknown

protease&helicaseunknown

replicase component ?

34

At the same time as Rice et (1985) published theentire YF vaccine gene sequence, French workers (Grange et al., 1985) published 565 nucleotides of the 3 ’-end of the genome of the YF17D-204 vaccine manufactured in France and later the 5'-end of the French vaccine was published (Despres et al., 1987). The French workers(Dupuy et ^ . , 1989) compared the two sequences andshowed very few changes in sequence between the two vaccine strains. There was a total of nine nucleotide changes, giving three amino acid changes over the whole genome. Therefore, there is a high degree of homology between the two vaccines.

The wild-type parent to 17D, strain Asibi was sequenced by Hahn et (1987) and compared to theYF17D-204 USA vaccine. Although the two YF viruses differ by more than 240 passages in different cell substrates the two RNAs only differed at 68 nucleotide positions. These differences resulted in 32 amino acid differences across the whole genome. The overall

nucleotide divergence was 0.63%, corresponding to a0.94% amino acid divergence. The envelope protein had a relatively high rate of amino acid change (2.4%) of 12 amino acid substitutions (TABLE 1.2). As the amino acid substitutions in the E-protein between the the two viruses are not conservative, Hahn ^ (1987) statedthat some of the changes could represent the differences

35

table 1.2 NUCLEOTIDE AND AMINO ACID CHANGES BETWEENTHE ENVELOPE PROTEIN OF THE WILD-TYPE ASIBI AND THE YF17D-204 USA VACCINE VIRUS STRAINS

NUCLEOTIDENUMBER

112711401482149115721750181918701887

194619652112214222192356

NUCLEOTIDE CHANGES

ASIBI 171

GCCCA

CCGC

CA

CCGC

AMINO ACID CHANGES

-204 ASIBI 17D-204

A GLYCINE ARGININEU ALANINE VALINEU ALANINE VALINEU THREONINE ISOLEUCINEC LYSINE THREONINEU NO CODING CHANGESU 1 NO CODING CHANGESA METHIONINE ISOLEUCINEU SERINE PHENYLALANINE

U PROLINE SERINEG LYSINE ARGININEG THREONINE ARGININEA PROLINE HISTIDINEA ALANINE THREONINEU

( A D A P T E D FROM H A H N E T , 1987)

36

in virulence between both viruses. To date these have not been determined. However, recently Rice et al. (1989) have developed an "infectious clone"of the 17D- 204 USA virus. Analysis of this clone should help in the understanding of the mechanism of attenuation of the 17D virus.

2. CHARACTERISTICS OF THE VIRUS

I. Taxonomic classification

A) Arboviruses

The term arbovirus (arthropod-borne) may be defined as those animal viruses which multiply in blood- feeding arthropods and are transmitted by bite to vertebrate hosts in which it also multiplies, producing a sufficient viraemia to infect another arthropod feeding on that vertebrate host. The arthropod known as the vector is usually a mosquito, tick, sandfly or midge and the principal hosts known as the reservoir are mammals and birds. Arboviruses can escape from the maintained biological cycle by vertical transmission of the virus from the arthropod to its offspring.

Several of the most important species were identified in the 1930’s including those of eastern and

western equine, tick borne and St. Louis encephalitis, though YF

37

and dengue were shown to be transmitted much earlier (Downs, 1989). The Rockefeller Foundation in the 1950’s initiated a worldwide program directed towards determining the number of Arboviruses. Casals in 1957 listed 47, then again in 1958 (Casals, 1958) the list was enlarged to 72. By 1962 the list had already doubled to 161 (Berge, 1975) and as of 1973 the list had reached 273 (Theiler and Downs, 1973). Though the 1978 edition of The International Catalogue of Arboviruses lists 388 viruses, the 1985 edition lists 504 registered viruses of which 32 are designated as probably not arthropod- borne viruses. (Karabatsos, 1985).

Casals (1957) grouped arboviruses together with the use of serological methods principally, complement fixation and haemagglutination inhibition assays. The first newly recognised viruses were sorted into serological groups A, B, and C. These groups were later designated as the Family Togaviridae containing the alphaviruses (group A) and flaviviruses (group B) (Porterfield et ^ . , 1978; Porterfield, 1980) and theFamily Bunyavirdae containing group C (Porterfield et

1975/76).

B ) Togaviridae

As the term ’arbovirus’ was only valid in a biological sense and had no revelance to the

38

replication/structure/chemistry of viruses, the Arbovirus Study group was appointed in 1967 to advise the International Committee on Taxonomy of Viruses (ICTV) on the choice of names. Following the use of 'togavirus’ as a 'jargon name' (from the Latin: Toga- a Roman mantle or cloak) for the group A and B arboviruses, the family name Togaviridae was approved (Porterfield, 1978). The family Togaviridae was first comprised of only two genera: the Alphavirus (for groupA) and the flavivirus (for group B). Later two more genera,' Rubivirus and Pestivirus, were approved (Fenner, 1975).

C) Flaviviridae

Until 1984, the flaviviruses were always regarded as a genus within the Togaviridae (White and Fenner, 1986). But in 1984, the ICTV approved the recommendation of the ICTV Togavirus Study Group to transfer the flaviviruses from the Togaviridae into a distinct family, the Flaviviridae (Westaway et ^ . , 1985). Theprefix of the family name is derived from flavus (yellow) and refers to to the type species, YF virus. The recommendation resulted from several fundamental differences with regard to flavivirus structure, replication strategy, and gene sequence. The flaviviruses have two envelope proteins while alphaviruses have

39

C The flaviviruses differ in size, the genesfor the structural proteins are located at the 5' end in a very long open reading frame and the viruses do not possess a poly (A) tail at the 3' end (Rice et al., 1985). In contrast the genes for the structural proteins for the alphaviruses are located near the 3' end of the genome and the RNA genome is polyadenylated at the 3' terminus and have a subgenomic RNA.

On the basis of cross-neutralisation tests using polyclonal, hyperimmune anti-sera De Madrid and Porterfield (1974) classified 42 group B arboviruses (flaviviruses). They were able to confirm the existence of 7 antigenic complexes or antigenic subgroups, with six of the viruses being antigenically distinct. In 1989, Calisher et analysed sixty-six flaviviruses in cross-neutralisation tests in cell culture and grouped the viruses into eight complexes with 17 other viruses still remaining antigenically distinct. This included the type virus YF. The viruses were also grouped by the principal vector responsible for transmitting the disease. The antigenic groups were:

TICK: Tick-borne encephalitisTyuleniy

MOSQUITO: Japanese encephalitisNtaya Uganda S Dengue

NO VECTOR: Rio BravoModoc

40

The complete list of members of the Flaviviridae is shown in TABLE 1.3

II. Flavivirus Morphology and Structure

A) Virion

Morphologically, in negatively stainedpreparations, virions of the flaviviridae appear as a uniformly icosahedral particle of 45 nm in diameter. This consists of an isometric, densely stained ribonucleoprotein core, between 30 to 35 nm in diameter and which is surrounded by a lipid envelope approximately 7-10 nm thick containing spikes or clublike projections (Murphy, 1980).

The virion consists of three proteins, E and M associated with the envelope and C associated with the core particle (Westaway, 1987). The chemical composition of one of the flaviviruses was 6% RNA, 66% protein, 9% carbohydrate and 17% lipid (Trent and Naeve, 1980).

B) RNA

RNA extracted from purified virions is infectious, single stranded, and of positive polarity that can immediately serve as messenger RNA (mRNA) upon uncoating in the host-cell. The genomic RNA extracted from infected cells is a 38-45S RNA and collectively known as

41

TABLE 1.3. MEMBERS OF THE FLAVIVIRIDAEASSIGNED TO ANTIGENIC COMPLEXES

(Modified from Calisher et al., 1989)

ANTIGENIC COMPLEX VIRUSES

Tickborneencephalitis

Rio Bravo

JapaneseencephalitisTyuleniyNtaya

Uganda SDengueModoc

(RSSE, Central European encephalitis), Omsk haemorrhagic fever, LI, Kyasanur Forest disease, (Langat, Phnom Penh bat, Carey Island), Negishi, Powassan, Karshi, Royal Farm.Rio Bravo, Entebbe bat, Dakar bat, Bukalasa bat, Saboya, ApoiJE, MVE, Kokobera, Alfuy, Stratford, SLE, Usutu, WN, Kunjin, KoutangoTyuleniy, Saumarez Reef, MeabanNtaya, (Tembusu, Yokose), (Israel turkey meningoencephalitis, Bagaza)Uganda S, Banzi, Bouboui, Edge HillDEN-1, DEN-2, DEN-3, DEN-4Modoc, Cowbone Ridge, Jutiapa, Sal Vieja, San Perlita

Viruses that could not be assigned to an antigenic complex are listed below:

Area, Bussuquara, Cacipacore, Gadgets Gully, Ilheus, Jugra, Kadam, Montana Myotis leukoencephalitis, Tamana bat, Naranjal, Rocio, Sepik, Wesselsbron, Sokuluk, Spondweni, yellow fever, and Zika*: Viruses in parentheses are more closely related to

each other than to other members of the antigenic complex

42

44S RNA (Westaway, 1980) has a type 1 cap structure: m^GpppAmpNp at the 5'-end and no poly (A) at the 3 '-end (Trent and Naeve, 1980; Rice et al., 1985). The base value of YF virus RNA is 27.3% A, 23.0% U, 28.4% G and 21.3% C (Rice et al., 1985).

Ill. Flavivirus Replication

A) Uptake of Virus

To initiate the entry process, the flavivirus exhibiting selective ligands on their surface must attach to corresponding receptors on target cells. These specific cell surface receptors for flavivirus bindingto host cells have not been characterized at themolecular level. Gollins and Portefield (1985) studying the mode of entry of West Nile virus into themacrophage-like cell line P388D1 concluded the virusentered by receptor-mediated endocytosis. Other workers have confirmed this using other flaviviruses: Ishak et al., (1988) using YF virus grown in Vero cells and Ng and Lau (1988), who used Vero cells infected withKunjin. However, Hase et , (1989) demonstratedthe SA strain of Japanese encephalitis virus andthe PR159 strain of Dengue-2 virus entered 06-36 cellsby direct penetration.

Due to the diverse cell tropism and pathogenicity

43

shown among the flaviviruses, it is likely that individual flaviviruses have evolved diverse entry modes. The entry mode probably depending on their characteristic life cycle under different conditions of growth and therefore further research in this area is required.

B ) Translation

Westaway (1980) has presented evidence that the translation strategy of flaviviruses involves the use of multiple iniation sites for the translocation of proteins from the virion length mRNA. However, the nucleotide sequence data for many flaviviruses indicates that the viral RNA is translated as a single long open reading frame. This produces a polyprotein structure which under goes extensive proteolytic processing (Chambers ^ a3 ., 1990).

C) Replication

The replication of flaviviruses is thought to take place in the cytoplasm of cells with the general agreement amomg investigators (Hase et , 1989) thatreplication takes place in the perinuclear region. The replication scheme for flaviviruses has been proposed by Westaway (1987) on the basis of accumulated data. Briefly, RNA replication involves the initiation and

44

completion of complementary minus-strands which are then used as templates for production of genome length plus stranded molecules. The plus stranded molecules can be used for translation of additional structural and nonstructural polypeptides, for minus strand synthesis or for encapsidation (Chambers and Rice, 1987, Hase et al.,1989).

D ) Virion Assembly and release

Electron microscopy studies usually show the flaviviruses in cytoplasmic vacuoles of infected cells around the membranes of the endoplasmic reticulum (ER) and golgi apparatus. Hase et (1989) have reportednucleocapsids of Dengue-2 forming in the cytosol and moving to the cell membrane for viral envelope acquisition, while JE virus was shown to form in the rough endoplasmic reticulum as whole virions. The membranes of the ER undergo a characteristic proliferation forming dense microtubular membranous aggregates (Chambers and Rice, 1987). The virions exit from the cells by many suggested mechanisms. These include release of many virions from vesicles either by cell lysis or reverse phagocytosis. Also by budding internally from sites of membrane junction with plasma membranes (Westaway, 1980).

45

3. THE PRODUCTION AND USE OF MONOCLONAL ANTIBODIESIN VIROLOGY

A) History

Prior to the discovery of Kohler and Milstein (1975) of the production and isolation of single cells which secrete antibody having a discrete specificity for a single epitope, polyclonal antibodies were used for serological studies. Polyclonal antibodies are still used to the present day and have played an important biological role in defining important serological groupings in virology using the techniques of haemagglutination inhibition, complement fixation and neutralisation. However, polyclonal sera reacts to more than one specific epitope and each preparation may contain antibodies with different specificities.

Kohler and Milstein (1975) used the hypothesis of Burnet (1957) who stated each mammalian B lymphocyte had the capacity to secrete a single specific antibody reacting to a single epitope. Though B lymphocytes have a short life span when cultivated vitro, they can be fused with myeloma cells to produce a hybridoma cell line. These hybridoma cells have the immortality of the myeloma cell and the antibody secretion property of the B lymphocyte. The hybridoma cells can then be cloned out in a selective medium to produce hybridomas producing a

46

monoclonal antibody (McAb) of a single specificity. In virology two of the more important biological functions defined using McAbs are those of neutralisation and Haemagglutination inhibition.

B ) Haemagglutination and HaemagglutinationInhibition assays

The phenomenon of haemagglutination (HA) was first described by Hirst (1941) when removing previously Influenza A infected allantoic fluid from chick embyos. He noticed the red blood cells (RBCs) had agglutinated. He then demonstrated that he could inhibit this RBC agglutination (haemagglutination) by using serum from a proven case of Influenza A. This RBC agglutination was shown to be due to the virus and was termed haemagglutination. The inhibition of RBC agglutination is called Haemagglutination Inhibition (HAI). Since its discovery HAI has been used in virology for a number of purposes including serological techniques to systematically group all arboviruses together. In 1947, Hallauer reported that he had obtained a haemagglutinating antigen from YF virus. From 1951 to 1953 Sabin and colleagues demonstrated haemagglutination in a number of flaviviruses except for YF virus. In 1953 Casals and Brown showed haemagglutination of both viscerotropic and neurotropic strains of YF virus and

47

that it was in group B of HAI reactions. Their source of antigen was the brains of newborn mice infected with the virus and the serum used was from a proven YF virus infected monkey. In 1957, Porterfield described a technique of HA and HAI using goose RBCs and reported these cells were more sensitive than chick RBCs to low titre virus.

C) Neutralisation Tests

It has been known for a number of years that antibodies induced as a result of viral infection can neutralise virions. However, the mechanism of neutralisation is still not fully understood and is the subject of debate. A number of workers have reviewed the mechanisms of neutralisation (Lafferty, 1963 a,b,c; Mandel, 1976; Della-Porta and Westaway, 1977 b; Dimmock, 1984; lorio, 1988)

An antibody which neutralises viral infectivity may affect the virion at a number of points in the replication cycle and studies using different viruses have shown a number of different mechanisms of neutralisation. Also, the number of molecules of antibody required to neutralise a virus has been debated.

Some McAbs bind to the virus and their neutralisation activity prevents the virus binding to

48

the cell (Massey, 1981). The belief that McAb bound to the virus always prevents the virus from binding to the cells has had to be altered for some viruses. In particular it has been shown that some McAbs which neutralise virus do not prevent it from binding to the cell surface. Taylor and Dimmock (1985), described IgM McAb which neutralised 99% of virus infectivity but only prevented 50% of virus attaching to BHK cells. In 1985 Gollins and Porterfield showed a new mechanism for the neutralisation of the flavivirus West Nile virus. They reported antibody could neutralise by inhibiting the intra-endosomal acid catalysed fusion step.

4. ANTIGENIC AND MOLECULAR ANALYSIS OF THE FLAVIVIRUSENVELOPE PROTEIN

The E-protein is approximately 500 amino acids in length, depending on the flavivirus studied. For most viruses it is glycosylated (asparagine-linked N- glycosylation site) with the notable exception of strains of Kunjin (Wright, 1982) and West Nile (Wengler, 1985). Biologically the E-protein encodes the functions of neutralisation, haemagglutination, haemolysis, passive protection, and antibody dependentenhancement. To date no one has reported an antibody

with complement fixation activity with this protein. Antigenically a

49

number of epitopes have been defined: sub-strain,strain, type, sub-group (or complex), super complex and flavivirus group common. At present relatively few of the 68 flaviviruses have been examined in detail. The properties of the E-protein of the viruses studied to date are discussed below with particular reference to those studies using McAbs.

I. Antigenic Analysis

A) Analysis of the YF virus E-protein

YF virus, is the prototype species of the family Flaviviridae. Five groups have prepared McAbs against this virus. In 1983, Schlesinger et prepared McAbsusing the YF17D-204 USA vaccine strain asthe immunogen. In 1985 Gould et prepared McAbsagainst the YF17D-204 UK vaccine strain, while Barrett et al. (1989) used the YF17D-204 WHO ALV-free secondaryseed and 17DD Brazil vaccine strains as the immunizing viruses to produce two sets of McAbs. Geske et al. (1983) only stated McAbs were prepared against"pathogenic" and "attenuated" strains of YF virus. A Russian group, Novokhatski et (1987) produced a McAb(Yel-2) using the Dakar strain of YF virus as theimmunizing virus. In indirect immunofluorescence tests it did not react with dengue-2, dengue-4 and Russian

50

spring summer encephalitis (RSSE) viruses. Two of these groups have reported McAbs that react with the E-protein and these have been used in competitive antibody binding assays (CBAs) to map the spatial positions of epitopes on the surface of the envelope glycoprotein. Schlesinger et al. (1984 b) used sixteen McAbs that reacted with YF17D-204 USA to map and define eight domains by the techniques of CBAs, Neutralisation tests (N), haemagglutination inhibition assays (HAI), and four sites by different serological group-reactions. They were able to show three YF type specific McAbs which constituted one antigenic site, another neutralization site spatially different from the first, was recognized by the McAb 8A3 which is YF17D-204 strain specific. The nine McAbs which neutralized Asibi virus and not YF17D- 204 USA defined six antigenic sites, of which one was associated with neutralization of YF17D-204 viruses.

On the other hand, Cammack and Gould (1986) who analysed topographically the epitope relationships on the E-protein using eleven McAbs identified five topographically distinct antigenic domains on the YF17D- 204 U.K. virus. Three were defined by one McAb each and were a YF17D-204 strain specific McAb (864), a YF vaccine specific McAb (411), and a YF-type specific McAb (825). One domain was found to consist of at least five overlapping epitopes and is thought to represent a

51

flavivirus group common antigenic site. They also compared epitope maps between the YF17D-204 UK vaccine and wild-type Asibi viruses. This revealed considerable functional differences but only one structural difference which was the epitope recognized by McAb 427. This was located in different structural configurations on each virus.

Thus, not only were domains found on YF17D-204 U.K. and absent on Asibi (i.e. strain and vaccine-specific epitopes) but also a difference in the structural configuration of an epitope was detected.

B ) Analysis of the TBE complex E-protein

The so-called TBE complex of the Flaviviridae family consists of TBE, LANGAT, NEGISHI, OMSKHAEMORRHAGIC FEVER, KYASANUR FOREST DISEASE, POWASSAN, RSSE and LOURING ILL (LI) viruses (De Madrid & Portefield, 1974; Calisher, 1989) all of which are closely related and fairly difficult to differentiate serologically. TBE could be further divided in to Western (Central European) subtype and Far Easternsubtype based on antibody absorption experiments andconfirmed by competitive RIA and peptide mapping using limited proteolysis of the structural glycoprotein (Heinz & Kunz, 1981 b). Heinz and Kunz (1981 b) stated the observed serological heterogeneity reflects the

52

potential for sequence or conformational changes within immunogenic regions of the viral glycoprotein. Changes which would not impair primary protein functions in cell infection.

Using McAbs produced against Central European TBE virus, Heinz et ^ . , (1983 b) established a model forthe antigenic structure of the TBE E-protein and showed the distribution of conserved and variable epitopes. Their clever strategy of selecting McAbs specific for different antigenic sites on the structural E-protein was achieved by isolating a panel of eight McAbs against the E-protein of the Neudorfl strain of TBE and subsequently applying the various tests of HAI, N, CBA, passive mouse protection and variant analysis. These tests analysed the spatial relationships between the different McAbs and showed they consisted of eight distinguishable epitopes on the native conformation of the E-protein . Of the two McAbs that defined a group- reactive antigenic site, McAb 6E2 defines a determinant which is altered on two viruses from another sub-groupi.e. West Nile and Murray Valley Encephalitis viruses. The type-specific determinant was defined by McAb 1G2 and the subtype-specific determinants by McAbs 1C4 (Cl) and 4D9 (A3) which allowed the discrimination between Western and Far Eastern subtypes of TBE virus. These two McAbs react with an epitope that is altered on the Far

53

Eastern subtype but which is present identically on LI virus. McAbs 1B3, 5D6, and 2E7 however reactedexclusively with TBE and LI viruses, thus defining a TBE complex specific determinant involved in HAI.

When the epitopes were topologically mapped, five independent non-overlapping antigenic sites were revealed that clustered into two complex antigenic domains. These domains were heterogeneous with respect to serological specificity as well as biological function. The effect of McAb mediated N or HAI depended on the exact location of each epitope in the 3-D structure of the E-protein. This was due to HAI reactive McAbs defining group, complex, and subtype specific epitopes. This agreed with previous studies by Clarke (1960) who used polyclonal antibody to demonstrate broad cross-reactive HI reactions with members of the TBE complex. In later papers Heinz (1986) changed the names of his McAbs to correspond with the domains which were named domains A, B and C.

Also, in 1983, a Russian group produced McAbs against another virus in the TBE complex (Novak et al.). This was the Skalica strain of TBE virus which was isolated from the bank vole. Three IgM McAbs S-8, S-15and S-16 elicited HAI activity exclusively with the Skalica strain.

In 1984 Stephenson et , prepared a panel of 16

54

McAbs using the Austrian isolate of Central European TBE virus (Neudorfl) as the immugen. Of these 16 McAbs, 9 reacted with the 58K polypeptide which was thought to be the major envelope protein while the other 7 McAbs reacted with a 51K polypeptide (which was thought to be NSl). They reported that none of their McAbs would neutralise the parent virus. However, clones Til, T13, T18, and T33/2 would neutralise the parent virus ifneutralisation was "assisted" with rabbit antibody raised against total mouse immunoglobulins. There was no correlation between HAI and "assisted" neutralisation as clones T6, T9, T18, T33/2 and T35/1 had HAI activity suggesting these two activities reside on different parts of the E-protein. However, McAbs T18 and T33/2 recognised epitopes having both biological activities and corresponded to results obtained by Heinz et al. (1983).

The McAbs were then used in an ELISA assay tocompare the antigenic relationships of (1) between TBE viruses, (2) between isolates of wider geographical origin and (3) to compare them against a Scotish isolate of LI virus.

When they compared three Austrian TBE virusisolates antigenically in the elisa assay a wide range of titres were observed and four broad groups ofepitopes could be distinguished:

55

Group 1, (T12, T16, T17) represented epitopes which were conserved in all three isolates and were present on the 51K polypeptide; Group 2 (T18, T33/2, T33/3) represented epitopes which were conserved on Neudorfl (CTEN) and St. Polten (CTES) but have changed on Hochosterwitz (CTEH) virus; Group 3 (T6, T7, T15, T33/1, T35/1) were epitopes which were conserved on CTEN and CTEH viruses, but had altered on CTES virus. The McAbs in Group 4 (T4, T9,Til, T13, T35/3) represented epitopes that had changedantigenically the most and mainly resided on the 58K polypeptide.

When TBE virus isolates from Czechoslovakia (Hypr), Russia (Sophin), and Turkey (TTE) were compared with CTEN virus there was a considerable amount of antigenic variation between these isolates. However the McAbs T16, T17 and possibly T33/1 reacted with all fourviruses. Conversely, the epitope represented by clone T12, while remaining constant in the Austrian isolates appeared to vary considerably in this second group of viruses with the notable exception of Hypr.

The epitope represented by McAb T12 was also conserved on LI virus, thus grouping LI virus among the other western subtype isolates. McAb T17, on the other hand, represented an epitope which was heavily conserved in all isolates.

The epitopes recognized by McAbs in groups 2 and 3

56

were moderately conserved in the Austrian isolates (CTEN, CTES, CTEH) and were present on either the 58K or 51K polypeptide.

Phillpotts et (1985) reported antibody-dependent enhancement with the above McAbs. Thirteen of the fourteen McAbs used only enhanced the replication of TBE virus (CTEN) and not a strain of West Nile virus (Egypt 101). These McAbs were T6, T9, Til, T12, T13,T15, T16, T17, T18, T33/1, T33/2, T33/3, T33/5. Later,in 1987, he used these McAbs in passive protection tests and reported that only one McAb, T9 which was able to neutralise the virus protected mice against viral challenge (Phillpotts et s^., 1987).

Another virus in the TBE complex that has been studied in detail is Negishi virus (Furata et al., 1984). The Japanese group has analysed antigenically Negishi virus by producing 12 McAbs, using sucrose acetone extracted Negishi virus as the immunizing antigen.

The McAbs were characterized by HAI and N tests using five flaviviruses isolated in the Pan-Pacific region (Langat, 3-Arch, JE, Apoi and Negishi viruses). The 12 McAbs were classified into eight groups after the many N and HAI tests had been combined:Group 1 McAbs (B9 and H6) reacted only with Negishi

virus in HAI and N tests.

57

Group 2 (F3) and Group 3 (DIO and F4) McAbs reacted with TBE virus complex being positive for Negishi and Langat viruses in HAI tests. F3 also reacted with Negishi virus in N tests.

Group 4 (A5) McAb was positive in HAI tests with Negishi and Langat viruses, but only neutralised JE virus and none of the TBE viruses.

Group 5 (A2) was positive in HAI tests withNegishi, Langat, JE, and Apoi viruses while it had positive N activity with JE virus and no activity with homologous Negishi virus.

Group 6 (C5) had high HAI activity with all fourflaviviruses and positive N activity with JE virus.

Group 7 (All, B1 and Fll) had high HAI activity with all four flaviviruses and N activity with JE and Apoi viruses.

Group 8 (B2) had no HAI activity but showed N activity only with the homologous Negishi virus.These workers suggested that the reaction pattern

of the McAbs showed that Negishi virus was most closely related to Langat virus followed by 3-Arch, JE, then Apoi viruses respectively.

Later, the same group of workers (Motohide et al., 1988) used ten of these McAbs to study Negishi virus in the biological reactions of HAI and several types of

58

neutralisation tests. The McAbs were classified into three groups and seven types according to the reaction pattern obtained. They also reported a mixture of 4 McAbs enhanced neutralisation in a kinetic assay.

C ) Analysis of the JE subgroup E-protein

On the basis of serological tests, especially N tests, flaviviruses were subdivided into subgroups or complexes (Chamberlain, 1980; Porterfield, 1980). One of these subgoups was known as the West Nile (WN) group and contained viruses such as Murray Valley Encephalitis (MVE), St. Louis Encephalitis (SLE), Japanese Encephalitis (JE), and WN. On further examination ofthis group of viruses by Calisher et (1989) thegroup was renamed the Japanese encephalitis antigenic complex. The E-protein epitopes of many of these viruses have been analysed in detail; JE virus (Kimura- Kuroda & Yasui, 1983, 1986; Cecilia et ^ . , 1988;Kobayashi ^ ^ . , 1984, 1985), SLE virus (Roehrig etal., 1983; Mathews & Roehrig, 1984), WN virus (Peiris et al., 1982; Besselaar & Blackburn, 1988) and MVE virus(Hawkes et ^ . , 1988; Hall ^ ^ . , 1988). These studieshave shown that the majority of E-protein epitopes canbe divided mainly into virus specific, subgroup-reactive, and flavivirus cross reactive sites.

Early studies with WN virus suggested the

59

existence of two main antigenic groups (Hamman et al.,1965). These are strains from Africa, Europe, the USSR, and the middle East which formed the Paleartic antigenic group which are serologically distinct from strains isolated in India (Gaidamovich, 1973; Hamman et al.,1966)

Peiris et (1982) were the first to produceMcAbs against WN virus. They used the Egypt 101 strain of WN virus as the immunizing strain and produced three McAbs all directed against determinants on the E- protein. The McAb F7/3 bound to the group-specific determinant and had good HAI activity but had poor N activity. Another McAb, F/lOl, which recognised a subgroup specific determinant, also had poor N activity and no HAI actvity. The last McAb, F6/16A had good N activity which was limited to the homologous virus as it only recognized a strain specific determinant. It is interesting that the Smithburn strain of WN virus was not recognised, as both strains (Smithburn and ElOl) are indistinguishable by classical polyclonal tests (Hammam et al., 1965 ).

In 1988 Besselar & Blackburn produced eleven anti-E protein reactive McAbs using a WN virus isolate from South Africa as the immunogen could be divided into four groups on the basis of cross-reactivity with various flaviviruses by indirect immunofluorescence ;

60

Group 1, WN-type specific; Group 2, WN complex-reactive; Group 3, Partially flavivirus group-reactive; Group 4, Broadly flavivirus group-reactive. One of the two McAbs in group 4 (6E8) defined a flavivirus cross-reactive HAI site present on YF, Dengue 2, Kunjin, and Wesselsbron viruses. The other flavivirus cross-reactive epitope was recognized by McAb 2A4 which unlike McAb 6E8, did not exhibit any biological activity in N and HAI tests. While they did not carry out competitive binding studies, they grouped the WN anti E-McAbs on the basis of functional activity and immunological crossreactivity to tentatively define a minimum of six epitopes.

Seven groups of workers have produced McAbs to analyse the antigenic structure of JE virus (Boquan ^ al., 1983; Gould et ^ . , 1983; Kedarnath et al.,1986; Kimura-Kuroda & Yasui, 1983, 1986; Kobayashi etal., 1984, 1985; Cecilia et , 1988; Mason et al.,1987; Srivastava, et ^ . , 1988).

The morphological, chemical, and biological characteristics of JE virus were first investigated by Kitano et (1974) who showed that the surfaceglycoprotein (E-protein) projecting from the surface induced neutralizing antibodies and also hadhaemagglutination activity. In 1983 a group of Chinese workers (Boquan et ) produced three McAbs using the

61

SA-14 strain as the antigen. McAb "32" showed HAI activity against JE, Kunjin and WN viruses. The second McAb "51" had N activity and in indirect immunofluorescence tests (IIP) only reacted with JE virus. It neutralised the avirulent 2-8 strain, 2 to 4 fold more than its parent SA-14 strain. The last McAb produced "43", showed N and IIP activity only against the JE SA-14 and 2-8 strains, but not against other JE strains (e.g. the JE JAK-2 strain).

In 1983, Gould ^ ^ . , produced a single IgM McAb using the Nakayama strain of JE virus. They reported that the McAb reacted with the NS3 (NV4) protein on western blots and it reacted in IIP tests with a number of flaviviruses.

Also in 1983, Kimura-Kuroda and Yasui produced ten McAbs against the E-protein of JE virus strain JaGAr-01. They characterized the McAbs by HAI, N, and ELISA and divided them into four types.Group 1 (McAbs 110, 111, 301, 23) was flavivirus-crossreactive with HAI reactivity but no N activity. Group 2 (McAb 109) was subgroup specific having HAI but no N activity. Group 3 (McAb 112) was JE virus strain specific having N but no HAI activity. Group 4 (McAbs 203, 302, 303) and Group 5 (McAb 204) had no HAIactivity but neutralised JE and WN viruses but not dengue-2 virus. They then also performed CBAs which

62

demonstrated the existence of atleast five distinct antigenic sites on the E-protein. Their results also separated non-HAI and N McAbs into groups 4 and 5 respectively and also showed that the group 1 site was distinct from any other site , but the sites of groups 2 and 3 seemed to be located together.

In 1986 the same workers produced other McAbsagainst the JE virus strain Nakayama-yoken. With alarger panel of 17 McAbs they were able to divide theMcAbs into eight groups using CBAs and analysed these groups for their cross-reactivity with several other flaviviruses. The results indicated that JE virusbelonged to the WN subgroup and was antigenically closely related to viruses in this group. It was moreclosely related to MVE virus, than it was to WN virusand was least closely related to SLE virus. The results of the N tests indicated McAb 503 from group 8 recognised determinants on a critical N site that was JE virus specific and had the highest N activity among the N McAbs. All the McAbs from group 7 showed high Nagainst JE and MVE viruses and one (McAb 402) also cross-neutralised WN virus. From these results they proposed that there is a variable region (recognised by McAbs in groups 3, 7, 8) on the E-protein which isimportant for infection of JE virus.

63

Using the JE virus strain Nakayama-RFVL as the immunizing strain, Kobayashi et (1984) produced 26McAbs and tested them for HAI activity with six flaviviruses: JE, MVE, Egypt 101 strain of WN, SLE, RSSE, and dengue-1. They showed that the McAbs fell into four groups : 14 JE species specific , 6 reactive withJE and MVE viruses, 3 specific to the JE-MVE-WN-SLE subgroup and 3 McAbs were flavivirus group common in reactivity. They examined and compared twenty seven strains of JE virus using McAbs 1 to 14. McAbs 13 and 14 were typical species specific while McAb 13 showed almost the same HAI titre against 26 JE virus strains. Also, McAb 5 and 6 were Nakayama strain specific McAbs. From these studies they were able to show atleast three antigenically distinct strains of JE virus existing in Japan over a period of more than 40 years. In 1985, Kobayashi ^ , used the same panel of McAbs incompetitive binding assays. They suggested the existence of at least five HAI sites on JE virus and that JE species-specific and flavivirus group reactive HAI sites were topologically distinct.

In 1986 Kedarnath et ^ . , produced a panel of sixteen McAbs against an Indian strain (733913) of JE virus that exhibited either antigenic specificity or cross-reactivity in some or all of the tests e.g. HAI

64

and N, The same group (Cecilia et a^., 1988) then usedthese McAbs to map the antigenic sites on the E-protein of JE virus and assigned biological functions to the mapped domains. Firstly two separate clusters of epitopes were revealed. Secondly, on the basis of reactivity in HAI, N, CBAs, and passive protection assays with McAbs , five functional domains (A, B, C, D, and E) were delineated. The HAI reactive flavivirus cross-reactive domain A was distinct and comprised five epitopes represented by McAbs Hx-1 to Hx-5. The HAI reactive JE specific domain B consisted of four epitopes Hs-1 to Hs-4 and was in continuum with the other domains. The non HAI reactive JE virus specific domain C contained two overlapping epitopes reactive with NHs-1 and NHs-2 McAbs. The non HAI reactive flavivirus crossreactive domain D was represented by three nonoverlapping McAbs NHx-1 to NHx-3. The last domain E contained two non-HAI cross-reactive and auto-reactive epitopes represented by McAbs NHA-1 to NHA-2 which were overlapping and were also juxtaposed to the Nhx-1 McAb reactive site.

Cecilia et (1988) concluded from theirresults:(a) Two distinct antigenic domains were associated

with HAI.

65

(b) HAI and N vivo and dji vitro were dissociated functions.

(c) Antibody-Dependent Plaque Enhancement (ADPE) activity was associated only with domain A.

(d) All McAbs reacting with epitopes in B domain had HAI,N and protective activity but failed to show ADPE.They suggested that domain B would be the most

suitable for development of synthetic or genetically engineered vaccines due to its biological function.

Srivastava et (1988) examined 42 monoclonalantibodies prepared against JE virus (strain JaOArS982) grown in C6-36 cells and classified 17 McAbs into five antigenic groups: strain specific, JE virus specific,sub-group specific reacting with MVE, WN, SLE, and JE viruses. The other two groups were subgroup and dengue specific as well as flavivirus specific.

Mason et (1987) produced ten McAbs against theNakayama strain of JE virus. Nine of these McAbs were specific for the envelope glycoprotein. McAbs 1, 2, 3,4, 6, 8 and 9 were type specific with McAbs 1, 4, 6, and 8 having high N activity ; while the other two McAbs were cross-reactive (5 and 7).

Roehrig et (1983) produced 21 McAbs specificfor the E-protein of SLE virus strain MSl-7. They

66

determined serological reactivities using indirect immunofluorescence assays and were one of the first groups to demonstrate that the E-protein of a flavivirus is divided into atleast four determinants, the type (E-1), subcomplex (E-2), supercomplex (E-3), and flavivirus group (E-4) reactive determinants. Using HAI and N assays they divided the groups into eight epitopes:El® (McAb 3B4C-7) type specific and no biological

activity.El^ (McAb 1B2C-5) type specific with only HAI

activity.El^ (McAb 6B5A-2) type specific with N and HAI

reactivity.El^ (McAb 4A4C-4) type specific with N activity only.E2 (McAb 1B5D-1) subcomplex with only HAI activity.E3 (McAb 2B5B-3) supercomplex with only HAI activity.E4® (McAb 2B6B-2) group epitope with no HAI or N

activity.E4^ (McAb 6B5C-1) group epitope with high HAI and N

activiy.Using these McAbs in a CBA assay showed that the

E-protein was divided into atleast six binding groups and that the epitopes are arranged as an overlapping continum.

Roehrig’s group (Hawkes et ^ . , 1988) have also

67

produced 10 McAbs against the Original strain of MVE virus. The McAbs were tested in an ELISA and demonstrated specificities ranging from MVE type specific to flavivirus group reactive. Eight epitopes were identified in neutralization tests and HAI assays. Some of the epitopes were identical to those found with their SLE virus McAbs and were assigned identical nomenclature. Three of the epitopes E-1® (McAb 4B6A-2), E-l^ (McAb 4B6C-2), and E-1^ (McAb 4B5A-2) were type specific for MVE virus. One epitope E-5 (McAb 4B3B-6) was shared by MVE and JE viruses. Another epitope E-6 (McAb 4B5B-10) was shared by MVE, JE and Alfuy viruses. Two epitopes E-4® (McAb 4A5D-7) and E-4^ (McAb 4A1B-9) were flavivirus group reactive. The remaining epitope was identified by the SLE virus McAb 2B5B-3. Of these epitopes E-l°, E-3, E-4^ showed both N and HAI activity. E-1^ and E-5 elicited MVE N antibodies, while E-6 showed HAI reactivity.

Their results collaborate the close antigenic relationship between viruses of the JE/SLE/MVE/WN serocomplex. Only the equivalent of the E-1^ epitope of SLE virus was not identified. Presumably, this was due to either absence of the epitope or their failure to isolate the appropiate McAb.

Also in 1988, Hall et ^ . , produced two type- specific McAbs against the French (3/51) strain of MVE

68

virus. One McAb (M-Bll) reacted with the PrM protein and the other (M-E6) reacted with both the E and NSl proteins in immunoblots. The McAbs failed to react in HAI assays with all viral antigens tested and did not neutralize the French 3/51 strain of MVE virus.

D) Analysis of the dengue virus complex E-protein

The dengue (DEN) viruses form a complex of four closely related antigenic types (types 1 to 4). Each of the four virus types can be transmitted by infected mosquito vectors. Three groups of workers have raisedMcAbs against one or more of the four DEN groups in the last ten years.

One of the first to produce McAbs against DENvirus was Dittmar et (1980) who produced 46 McAbhybridomas against DEN-3 virus. These 46 McAbs showed flourescence when added to DEN-3 infected cells in IIF tests. From this number of hybridomas they picked fourto study: numbers 31, 36, 20 and 27. They showed McAb 31was DEN-3 virus specific when assayed by HAI assays and McAb 20 was highly cross-reactive. From their data it was not possible to accurately determine which viral protein the McAbs were directed against. But as all were involved in HAI, they postulated clone 31 was directed against a type-specific determinant on the E-protein.

69

Then Gentry et al. (1982) produced 22 McAbs against the New Guinea C strain of DEN-2, which they characterised by HAI, N, IIP tests and complement fixation tests (CF). They showed five different McAbs to be type-specific for DEN-2. One of them (McAb 3H5) recognized a type-specific determinant that exhibited high N, but had low HAI activity. While the other four McAbs (1C7, 1C12, 2H3,3H1) were type-specific with high HAI but having no N activity with the homologous virus. The other seventeen McAbs were all broadly cross-reactive McAbs. In the same year, the same group (Henchal et ^ . , 1982) producedMcAbs against the four DEN virus serotypes andidentified four categories of reactions by immunofluorescence.

(i) Flavivirus group reactive(ii) Dengue complex specific(iii) Dengue subcomplex specific (e.g. DEN-1 & DEN-3)(iv) Dengue serotype specific.

This was the first time that McAbs had beenproduced for all of these unique antigenic determinants.The majority of their McAbs were flavivirus group reactive e.g. 1B8, IBIO, ICIO, 4G2, 2C4 (DEN-2 NewGuinea C strain), 1B6, (DEN-3 strain H87), 4B10, 1D7,(DEN-4 H241 strain) which generally reacted with all flaviviruses tested using IIP tests. Several other McAbs reacted solely with DEN virus serotypes. One group was DEN subcomplex specific only recognizing DEN-1 and 3

70

viruses (McAb 13E7 [DEN-1 strain Hawaian] and McAb 2F3 [DEN-3 strain H87]). A second group of McAbs were DEN complex specific, these were McAb 9D12 (DEN-1 Hawaian strain) and McAb 2H2 (DEN-4 H241 strain). The last nine McAbs they tested were type-specific by IIF tests, McAbs 15F3, 5C11 (DEN-1), 5D4 (DEN-3) also by N activity McAb 3H5 (DEN-2) or by HAI activity 3H1, 2H3, 1C7, 1C12 (DEN-2), IHIO (DEN-4).

They proposed that the second epitope (subcomplex specific) and common to DEN-1 and DEN-3 must also exist within the variable region of the E-Protein, as well as the complex specific epitope because McAbs 9D12 and 2H2 were totally unreactive with other flaviviruses.

Henchal et (1985) then used some of theseMcAbs in CBAs and analysed the epitopes of the antigenicdeterminants on the surface of the DEN-2 virus E-protein. They showed that four groups of antigenic determinants 3H5, 4E5, 2H3 were type, 9D12, 13B7 were subcomplex, 1B7 flavivirus subgroup and 4G2 wasflavivirus group specific, and identified atleast seven epitopes which could be grouped into atleast three different linkage groups. The presence of epitopes with different serological reactivities in the same linkage group was similar to the findings of Roehrig et ,(1983) and supports their conclusion that epitopes on the E-Protein of DEN virions maybe arranged as a

71

continuum of reactivities in overlapping domains.Morens et al. (1987) produced nineteen McAbs

against dengue type 4 (DEN-4) virus strain 4328-S. These McAbs were first characterized serologically in ELISA, N, and HAI assays. Twelve McAbs were flavivirus group- reactive (UH-2B2, UH-2B4, UH-2C3, UH-3D1, UH-3D3, UH-5B3, UH-6B3-4, UH-5C1, UH-7D6, UH-8A1, UH-8C2, and UH-8D5-1) and seven were DEN-4 serotype reactive (UH-3C5- Bl, UH-5C3, UH-6C3, UH-6D3, UH-2A1, UH-8A4, and UH-8A6). The last three McAbs reacted only in ELISA tests and were assumed to be directed against non-envelope determinants. The McAbs produced were then studied in antibody-dependent enhancement (ADE) experiments. They showed that McAbs raised against DEN-4 could inhance infection of DEN-2 viruses. The McAbs were serotype- reactive in HAI and N activity. Therefore they postulated some DEN-2 virus strains must have DEN-4 virus like epitopes.

II. Molecular analysis of flaviviruses

The use of McAbs to define serological relationships and to ascribe functional activities to the individual epitopes has been extensively used for the analysis of flaviviruses. In addition, McAbs are excellent tools to probe the structural properties of individual antigenic determinants. Accordingly, they

72

have increased the knowledge of the antigenic structure of proteins in general. However, to date, this potential has not been exploited to a great extent, and detailed data relating to the structural properties of epitopes in relation to their RNA sequence and peptide fragments is still fairly limited though still being extended.

A detailed study of the TBE virus E-protein has been made by Heinz et (1983). They showed thatdomain A, which contains the flavivirus group-reactive epitope Al, was sensitive to conformational change at different pH levels. This also corresponds to similar findings of Henchal et al. (1983) who showed that a flavivirus group-reactive epitope on DEN-2 virus was also sensitive to pH changes. Also Kimura & Ohyama (1988) studied the association between pH-dependent conformational changes when the virus fused to the cell membrane.

In contrast, Heinz et (1983) showed domain Bis resistant to acid pH and even guanidine-HCl urea and also SDS treatment, though it is destroyed by carboxymethylation. This suggests a role for disulphide bridges in the structure of the E-protein.

Winkler et (1984) reported evidence for astrongly stabilised core and suggested evidence of the involvement of disulphide bridges. Subsequently the nucleotide sequence of the first flavivirus genome, YF

73

virus, was published in 1985 (Rice et a^., 1985).Wengler and co-workers (Wengler et ^ . , 1985) analysedthe sequence of the E-protein of WN virus and then in1987 proposed a model for the E-protein (Nowak & Wengler1987). They showed the E-Protein of the YF and WNviruses contain twelve cysteine residues which are in conserved positions among both viruses and form six disulphide bridges; cysl-cys2, cys3-cys8, cys4-cys6, cys5-cys7, cys9-cyslO, cysll-cyslO.

Investigating proteolytic cleavage, Heinz et al. (1983) have shown that different antigenic domainscorrespond to different structural entities. Mandl et al. (1989 a) continued this work and havesequenced the proteolytic fragments and have correlated the sequences to the defined epitopes and antigenic domains recognized by their McAbs. They proposed a model for the TBE virus E-protein (FIGURE 1.4) which reveals features of folding of the polypeptide chain including the generation of discontinuous protein domains. Recently Heinz and co-workers have pepared a new panel of McAbs against TBE virus (strain Neudoerfl, Guirakhoo et ^ . , 1989) and extended the limited epitope map of Heinz et (1983) by 11 distinct epitopes.Roehrig et (1989) using computer analysis modifiedthe E-protein structure proposed by Nowak and Wengler

74

FIGURE 1.4 ..STRUCTURE OF THE TBE VIRUS E-PROTEIN (TAKEN FROM HEINZ ET , 1990)

A

'-200

OENNH2

3ocrl50-(30

TBEMVEsueJEOEM

COOH

B

1) A, B, C: correspond to different antigenic proteindomains recognised by McAbs

2) Disulphide bridges are marked by connecting solid lines (from Nowak and Wengler, 1987)

3) Thin lines indicate variable sequences4) Thick lines indicate hyoervariable sequences5) Diamonds indicate potential N-glycosylation sites for

TBE and other flaviviruses

75

(1987) and prepared synthetic peptides from the deduced amino acid sequence which were recognised by polyclonal antiviral (MVE virus) antibodies. Using antipeptide antisera in CBA's they showed spatial overlap of the R1 and R2 domains.

Other workers using other flaviviruses have studied the proteolytic cleavage events in flavivirus maturation. Wengler and co-workers (Wengler et al., 1987; Wengler & Wengler, 1989, Nowak, et al., 1989) have used WN virus to study these events, while Biedrzycka et al. (1987) have used DEN-2 virus. New Guinea C strainand have determined the amino acid sequences surrounding the likely proteolytic cleavage sites used in the formation of the E-protein and three non-structural proteins. Markoff (1989), produced mutants of DEN-4 virus by site-directed mutagenesis, and have observed that the cleavage at the capsid-prM and prM-E sites is effected entirely by signal peptidases.

Two groups of workers have studied the maturation of YF virus. A French group (Ruiz-Linares et 1989)and an American group (Chambers et ^ . , 1989, 1990) have both analysed the role and site of action of cellular proteases to produce both the structural and nonstructural proteins (FIGURE 1.5).

Mason et (1987) used McAbs (previouslydescribed above) to examine the expression of the E-

76

FIGURE 1.5 THE STUCTURE OF THE YF VIRUS GENOME AND PROTEOLYTIC PROCESSING OF THE YF POLYPROTEIN

(ADAPTED FROM CHAMBERS ET AL., 1990)

YF17D GENOME (10,862nt)

llSnt 10,233nt 512nt5'CAP- STRUCTURAL NONSTRUCTURAL

virionassembly

prManchC ' 0 NS2B

NS4A

NS2ANSl

NS4B

NS3NS1-2A NS4A-4B NS5

□ 0Ml virionrelease ?

CLEAVAGE EVENTS:♦ signalase (rapid, in lumen of intracellular membranes)I Dibasic (rapid, in cytoplasm)* Dibasic (delayed in cytoplasm)4 Dibasic (delayed, in lumen of intracellular membranes)V Novel (delayed, in lumen of intracellular membranes)

77

protein of JE virus in E. coli. The clone J7-ls, which only contained 135 of the 500 amino acids of the E-Protein, still retained immunoreactivity with each ofthe nine McAbs. The workers detected two immunogenic regions in the E-protein which they presumed are exposed on the surface of the virion. Only one domain isrecognized by all nine anti-E McAbs, between amino acids 280 to 414 of the E-protein. This segment ends 83residues from the C-terminus of the protein.

Ill Sequence Analysis

A number of groups have sequenced the entire or parts of the genome of a limited number of flaviviruses with many strains also being sequenced. TABLE 1.4 lists the flaviviruses which the entire genomes have been sequenced. Two groups of workers have sequenced the E-protein of the YF17D-204 vaccine strain (Rice ^ al., 1985; Despres et ^ . , 1987) while Hahn et (1987)have sequenced the Asibi virus. Lobigs et al., (1987)have localized and sequenced a neutralisation YF type- specific determinant on the E-Protein of YF virus at amino acid positions 71 or 72.

78

TABLE 1.4 FLAVIVIRUSES SEQUENCED

Virus/strain Author

DEN-2 Jamaica DEN-2 SIDEN-2 New Guinea C DEN-4 Caribbean814669 JE JaOArS982 JE Beijing—1 JE SA-14 JE SA-14-14-2 Kunj in MRM61C TBE Neudoerfl TBE Sofjin WN Nigeria YF 17D-204 USA YF 17D-204 France YF Asibi

Deubel et al., 1988 Hahn et al., 1988 Irie §t al., 1989 Mackow et al., 1987 Sumiyoshi et , 1987 Hashimoto et , 1988 Nitayaphan et , 1990

Coia et al., 1988 Mandl et , 1989 b Pletnev ^ , 1990Castle ^ , 1986Rice et al., 1985 Dupuy et al., 1989 Hahn et , 1987

DEN-3 H87 Osatomi and Sumiyoshi, 1990

79

5. AIMS OF THESIS

Previous studies (Liprandi, 1981; Buckley and Gould, 1985 and Barrett and Gould, 1986a) have shown that YF vaccine viruses from different manufacturers are antigenically and biologically distinct. The aim of this thesis was to examine and compare in detail all the YF vaccine viruses produced by manufacturers in different countries using different conditions of manufacture and different seed viruses. In particular, the studies would concentrate on the E-protein of the viruses. Therefore, panels of McAbs were generated and the epitopes of the E-protein of different strains and sub-strains of YF vaccine viruses examined using these McAbs in the biological assays of HAI and PRNT.

Also, the pathogenicity of the YF vaccine viruses will be compared using the mouse and monkey as models to test the neurovirulence of different strains of YF virus.

Finally, a comparison will be undertaken of the properties of YF vaccine viruses when passaged in vertebrate and invertebrate cell lines. Each cell passaged virus will be tested for temperature sensitivity as well as differences in biological function in PRNT's and HAI assays.

80

CHAPTER 2

MATERIALS AND METHODS

81

1: CELL CULTURES AND VIRUSES

I. Cell Cultures

A) Propagation and maintenance of cell lines

All cell lines (TABLE 2.1) were maintained on Eagle’s minimal essential medium [EMEM] (Eagle, 1959), with the addition of L-glutamine (2mM), penicillin G (100 units/ ml) (Sigma chemical company Limited, Poole, Dorset), streptomycin sulphate (0.1 mg/ml), sodium bicarbonate (0.075% final concentration) (Sigma chemical company Limited). The media contained 2% (v/v)(maintenance media) or 10% (v/v) (growth media) newborn calf serum (NBCS, mycoplasma and virus free. Flow Laboratories, Irvine, Scotland) for SW13, LLCMK-2, CV-1 cell lines. VERO, BHK-21, C6-36 and CEF cells weremaintained on 2% (v/v) (maintenance media) or 10% (v/v) (growth media) foetal calf serum (FCS, mycoplasma and virus tested. Flow Laboratories). Cell culture media were obtained from Flow or Gibco Laboratories.

All the cell lines ( except for C6-36 cells) were trypsinized when the cell monolayer was confluent, using 5 mis of a versene (0.05% [w/v] EDTA, lOOmM Sodium

82

o O CO ro oM cy» s: K <Hj 1 H « 1 Ow W 1 H 10» toH ?to

< >0 cr >tr wO 0 W<H- CO0 «rt H*H-O3

ti H) O M > 0 w X" Cd X > g X > QH H- U 0 0 0 c p. p) H- H) p. y H- Ml MH- Ct H- H Qi 1—' 3 & tr Pi H Pi 0 Pi H MS H 0 < 0 pj pj 5 3 H- 3 0 3 H-0 O X 0 0 5 0 0 0 0 C 0 0H tr 0 H y 0 0 ►C 0 OM 0 0 p. p) 0 5 3 0 H0 S 0 H 3 Qi 0 3 0 0 3 0 H* g0 w & 0 tr 0 0 0 0 0 kP 0 0 0 iQ M W0 ft H M 0 b M ft M H M 3 1—' M COH W M'0 0 H 0 M 0 H X M 0 MH 0 H- Q 0 0 0 H- d5 H*n 0 M H 3 M 3 3 COl_i p. H*ft H H- H* H- H* 0 EdH- 0 5 c 0 5 3 3 3 3 3 D3 0 0 H- 0 0 o 0 0 00 3 3 3 H0 X X" g3 0 00 •C

< < < < < î?> < 0 <!H- H- H- H- H- H- H H-b H H H H O 0 Ii3 3 3 3 3 M 3 rtkP 30 0 0 0 0 0 0 3 0A 0 0*3 tî ts *Ü 3 % M dH H 0 H H- 0 Ii 00 0 0 O O 0 Pi 0 o 0tJ 0 T3 0 0 00 0 0 0 0 0 0 0 0 0kP kP kP kP kp 0 kP k< v£|0 0 0 0 0 0 0 0 0ft ft ft ft rt •C ft rtH- H- H- H- H- 0 H- H*0 0 0 O 0 0 o3 3 3 3 3 3 3

IHMCO

D h3 C (D b rt U1 M H) tu O tl H O Di C" H ro

rtCQ d h OQ 5 Kj (Q

gMft

g g

cCO>

o00hœHioMoH-w00m0

8m

§

;

MCOH-4OHi

CO H-

? !H- W

CO S BS5-H CQ

t) H 0 3 W CO rt ft 0 H-r4Hi

&oh0ftoHH-0W

§

œ00 _ H Mct*0y w 0 btH&oH0ftoHH-0m

83

hydroxide, 0.0025% [w/v] Phenol red/PBS solution) /trypsin (1:250, 2.5% in Hanks Balanced salt solution[Hanks, 1949], Flow Laboratories) PBS solution (Oxoid Limited, Basingstoke). The C6-36 cell line was subcultured by manually tapping the tissue culture flask and splitting the contents in a 1:5 dilution.

B ) Storage of cells

Cells were frozen in cell medium containing 20% FBS (Flow Laboratories) and 7.5% Dimethyl Sulphoxide (Sigma Chemical Company). The cryotubes (Nunc) were wrapped in cotton wool in a foam box to ensure the temperature decreased slowly and then placed at -70°C for 24 hours before being transferred to liquid nitrogen for long-term storage.

II. Viruses

A) Wild-Type YF Viruses

The wild-type YF viruses used in this study are shown in TABLE 2.2. All viruses were obtained as freeze dried preparations which were rehydrated with 1ml of distilled water and aliquoted into lOOul amounts and

84

K H > < rh < 4(Ji 00 0 p. h{ p. h{ >kO H- M 0 0 (D W

kO or C P O 3 <J\ H- 0 H- 0 0 H- HCO 0 H 3* nH 0

1 30MN)

03 I<H-0ftH-O3 n

5CO CO t3 o CO o 10 0 0 3* 0 o hi3 3 P 0 3 3 K0kP0H

0kP0M

3 30

0kP0M

3Sk

S

<

§CO > H H) hj CO 3 H O' 3 HO 0 H H- 0 0 H- 0 H H rt 0 0 0 0 CD3 3 3' kO rt H O H 3 O 0 0 3 0 ^ 3 0 M0 0 3 00 N 3 H H P i 3 kP X 0 H is: 0 H UirtkP g O k P 3 0 0 0 3 0 P i 0 0 P i 03* 0 0 ^ 0 • ft H N 0 0 0 ft ft ft 0 CÎ

H 3 0 0 > 0 3' 4 0 0 COP- J?>|0 tt^P 10 0 P i p- Ii pi 0 H

•cl- 0 H- kP H l(+ 3 3 Irt H- 0 0 0 Ofp 0 0 kO H) > 3 P O' H) 3 Hi P

10 H- 0IH Qj H (Jl M P i 10 0 H- 10 p- Ii 10 O P 0 Hor* Lo0 3 H) IP' 0 Ift O O X• 0 0 B p . . p. • 3 p* 3 X^ 3 rt 03 3 H- 0 " H 10 0 p-

H- 0 P 3* 0 0 P i kO 0 1m 0 0 0 w' 0 0 3 s, O > to ft H

H H) P i B H M <; H 'O'P " 1^*0 o rakO H- H H- 0 ^ kO 0 P- kO -w» pj 0 000 3 0 H 3

HkOcy>(Jl

03'

3 H CO rt P kO 1. 0 3 00 • 0 00 •

CO • ft 00 P- ^ 0

3 rt

ftH 0 P- kO pj 0 CO P- 3 '-J ft

< CO

1K

n 4 O hr) o0 0 D 0 0 O 0

3 P Ii 3 Ii 3 Kjft ft • O ft ft rt ft 0

0 P* 0 0 P»o hj 13 ft o Ii O 0o O 0 ffi • 0 0 o 0 0 o 0 COH 0 M P" H O H 0 ^ d oH 3 Ml CO p| 3 H 3 H) H 3 H) d 3 3P- rt 0 P- 0 P- rt 0 P- ft 0 CO P- Ii3 P Ii O rt pi 3 Ii i 3 Ii > < O0 0 N 0 o 0 o w 0 0

H D S kP H) H D p* O Iip- W 0 M P- X p. 0

a 0 0 0 d 0 d 0 p*CO 0 Ii 3 CO 0 CO 0 ft> 0 kP > 0 > 0

0 0 0 00 0 0

85

stored at -70°C in cryotubes (Nunc/Gibco, Uxbridge) as virus seed stocks.

Wild-type YF viruses were grown by infecting a confluent 25cmf tissue culture flask (Nunc/Gibco) of SW13 cells with approximately 0.01 PFU/cell. After 72 hours or when CPE was observed in 90% of the cells, the tissue culture media was clarified by centrifugation in plastic universals (Sterilin, Feltham, Middx.) for 5 minutes at 2000rpm in a MSE bench-top centrifuge. The supernatant was then dispensed into 1ml aliquots and frozen at -70°C in cryotubes.

Wild-type YF viruses were also propagated in a confluent tissue culture flask of C636 cells using a similar M.O.I as before. After five days the media was clarified by centrifugation and the supernatant stored at -70°C.

B ) YF Vaccine Viruses

All YF vaccine viruses were obtained from individual manufacturers ( TABLE 2.3) as freeze dried preparations which were stored at -20°C until required. YF vaccine viruses were propagated by taking a freeze- dried ampoule of commercial vaccine thawed out at 37°c and then rehydrated with 1ml of sterilised distilled water.

85

3O•H+J3T333 X0 Pk O>HP0rH 0H 0(M0XÜ0X

I0 mCO 0 •HrH 4J05 05 0 UII0 "u

rH3 P P3 3 3O 0 0•H +J ■PÜ *3 0 03 3 3 P 3 0X rH P4 3 P-i -H3 XO CO +J 3 +J 3+J 3 Û 3 PU3 0 +J+J "C ■H -H•H ■P +J4-> 0 00 3 33 H HH

0+J5•H-PWcH

s-pp0AOX

0+>•H■PWs05 Ü •H6uEH>1S

0iH(D§00CdPU

tIoÜ

cy» 0^ If) CM CO If)CO CO CO CO CO COCO CO CO CO CO CO

Ô!zi CO

rHf: COo >4+»3m

s

00cHCOS

CO5

I

COCO

If) OrH If)

rH (y>CO 1 1X4 so CO CO 4P cyio CO H 00 00

o 34 C'-rH

H O

3 3"r4 •H rH 0rH A 3 03 e tr> 3P z O 0 3+J rH 3 P0 0 0 IP3 u CO

3O•H+J3 00P CO CO3 CO COft0PftCOT5 CO0 O•H If)4H _•H ^P cy>3 00ft 1 Tf

rH 00 000 O \ rH3 rH VOtr g \ m3 0 CT> <fH P 00ft 4H CO

&0Ü4J0

T5C05iHrHOw

CO 0. 3CO P•HMPIm<E4

o o ftCO1

CO1 gCO CO PI ft ,ftÛ Q ÎD o P P ft

r-' rH r> rH < < u Q ft ft ftH (Û ^ rH 0 Ty TP3 \ 3 \ 'vl' TTCO >1-H 'Vj* 00 >1t H ^ O o O o O o OrH rH Ü \ H H 0 \ CO CM CO CO CO CO COCO p Ü rH CO P U rH Q Û Q Û Q Q Qft 3 3 CM ft 3 3 CO r c- C' r-'ft ft > — ft ft > rH rH rH rH rH H rH

g

o(MQc—I

87

0 04-> Ü3 •H P >1 0 >14J > O 0 0 3•H P (M O ft •H 3■P 0 rH 0 P ft0 CO 0 o ft 0 g 75u 3 4-> 3 0 •H ft O 0 00 H >4 3 3! 0 0 ft 3 U > pu P ■P >1 Ü •H ft 0 P •H P3 f: 0 •rH 0 P O r— 1 1— 1 0 tr P •HÜ -p H-» O -P O ft 0 ft 3 0 >Ü p 3 0 rH 0 ft •H cyi ftcr 3 p 75 753 3 P 3 O •H 3 ftco >4 CO PI Q(H 0 O H P PQ P 0 cn ft cr ft in 03 0 A -H O rH iH ft 1—1 3 03 0 3 H > 0 ft 0 v— 73 ' ft 3 3 rH 03 ft PI 3 g 3 ft rH tr 3 ft VO

IS: 3 O PI >1 - •H *• 3 0 PrH 1—1 0 Ü ft 3: • 3 •H 0 • P3 3 -H rH rH rH| 3 ft >0 3p P ■P 1—1 '3 3l 0 31 3 3 0 3i 75+J 0 3 0 rH ft 0 % P _ 33 73 ft 15 'H ft! ft ft! u ft O0 0 3: 01 01 0 Ü Üu ft ft 3 ft 03 73rH5 0•H CD3

3 73 O rH5 §•H Ü3

•H 3 Ü ft Ü ^ 3 C > CO

000Pft1—1 ft0 3 3 < 3 < 3 H' 0> 0 0 • 0 • 0 o p0 •fH •H ft •H M rH CM P1—1 •P ft L-, ft L-J P 1 •Hn 3 in 3 3 rH Q >0 r> ft CO CO P CO P P M « « •H r-x

& CM ft CM CM 3 CM 3 rH 3 rH % % % > H 03 ft ft rH ft rH O ^ •H0 0 0 0 p g 00 P P ft P ft 3 0 O3 ft ftC ftC 0 P ftft Ü Ü % ft 3730•rHMH•rHP CM

•H :>ift ft •rHP 73 •HP 75

0 1—1 133•H. 3 o 3 0 3 0 >O ft CO ft 0 ft 0 C% CM o c •H •H CO inVD CD C O 0 rH CM 0 3 0 3 in 00 in VO 3Si C- CM CO 3 \ 3 tr 3 tr rH o Oo 1 CD cy> O cr g rH cr o cr o w 00 rH VO •H+> W CO rH 3 O \ 3 Ü 3 Ü CM CD cH ft3 rH P ft rH 0 iH 0 ft 3m ft ft >4 ft p ft P 0

3Ü•fH3Ü•H

§%

•HëS'o5

>i 3 P P 3 fHu 3 -H 4H (M •H < rd+> •rH P < < ft 03 TJ 0 CO ft3 3 &> 3! 73 5 75O H •H +J •p 0O ft 3OCO §CO

ft•Hë

fHp§

0 +ft 0 ft ft cQ ft CO ft ft c < c g

0 ft c < c ft ft ft CO CO CO CO3 H CO CO CO ft >i p p P p P PP Ü P•H H H* H' 3 H* H*> O ft- o O o o g o O o O O O O

CM «CM CM CM CM •H CM CM CM CM CM CM CMQ P Q P P P P p P P P P P Pr- M r-' ft r-' r-srH rH rH rH rH H ^ H rH rH rH rH rH rH

87a

0•H 0 ft 0 N•H 75 3rH tH P0 ft UP g rH O0 0 3 75p •H ft rH3 H ft 15 3ft 0 0 O 15O ft Ü ü 03 3 0 O(H ft H O3 0 ft O3 1—1 33 0 3 Üft ft P 33 -H 75ft > 3•H 3ft 75 k0 3 3 3 H

?

M

co 0P 0 < P

0 3 3 co 3 «75 0 0 p 0 pft 0 V-- ft pd1—1 0 •H r4 0 §3 3 P p 0 0 3 0 33 ft 3 p 3 ft -H U0 ft p ft•H P o-H ft 3 o 3 rH ft 3 co pÜ 3 û ft O H 3 ft p3 ft U 0 ft ffi 3ft •H 0 U •H ft 0ft P ft ftO 0 0 ft 0 P3 3 ft p 3 0ft H 3 0 H ft-H 0 kft U03H

0 3 3> 0 00 •H -H1—1 ft ft3 C 30 k k P 00 P CO0 p k 3 CM 3 CM3 A ft0 0 00 P P3 A ftft 75 750 0•H •Hft ft•H •H

P P. 3 3O ft ft% < 0 rH 0 rHM CM 3 W 3 •Ü in 00 cr M cr rHft rH 3 3 003 rH rHm ft ft

M

M

§

I

33 1—1 •H 1—1

p -H •H 3ft 0 N Cn3 0 3 E o z 03 3 P rH 3O ft m 0 0U u CO

cop

0H3Pk

g

fttsi N ftft < < P P kCO ft ft O O <CO m m u u Pp p p p p Pp p p p p Pr r-. C"H rH rH rH rH rH

uk

k

ftH

k k

87b

Then lOOul of virus was used to infect a confluent 25cm^ tissue culture flask of SW13 cells. This virus preparation was then used as the seed stock to infect other confluent cell lines when that particular cell line derived virus was required in an experiment.

C ) Other Flaviviruses

Other flaviviruses listed in TABLE 2.4 were used to make spot slides for the use in indirect immunofluorescence studies. They were propagated and frozen as described above for YF virus.

D ) Virus Plaque Assay

All virus titrations were performed in LLCMK-2 or Vero cell monolayers prepared in six welled tissue culture plates (Nunc or Gibco). Ten-fold dilutions of virus were made in PBS and lOOul of each dilution was inoculated onto a monolayer of an appropiate well which was previously washed once with PBS. Following incubation for 30 minutes at room temperature (RT), the wells were overlaid with 6mls of a mixture of 2% (w/v)agar (Sigma chemical company Limited) and double strength Eagle's minimal essential medium containing 4%

88

0 0 X—X0 < 0 <3 CO 3 CO0 p 0 p0 0•rH •rH0 p rH 0 P rH 0

O 0 3 0 3P p P •H P P •H3 O ft rH O ft rHO ft 3 rH ft 3 rHCO 0 O 0 00 U U 0 U U

P P0 ft 0 ftft P ft P3 o 3 00 k 0 kU U

wIg I—I• 3ft ft CO -H ft 01 “I

00 <3 CO0 p0•rHP rH 00 3P P •H0 ft rHft 3 rH0 00 U UP0 ftft p3 00 kU

COwHICOMCOIkHPwCOPCOMCOI§kI

3-H-Huoft0yft1Ü

303SI

3O-H oft o3 p-H> CM0 1P kA MS P

33■H6

HP

3•H0> 43r4ê

Eh5

OJ•H?H

3 UVO

-H 3 r> rH VO3 0 1 CM CMP 0 3 in k CMftCO ' 1

EH Ü

3-H1H

CMI0s-H>

u30CM 3 •H 0 3 3 Ü

3 0 0 P P

0•Hft•H0 rH 0 3 0 ft 3 ft 0a s sP w >

0•Hft0 -r|F dO ftS 0Ü Ü •H 3 EH M

0s•H>ft3g3P

0rH•HPft 3 0 P 0 -H >

3-HH

89

foetal bovine serum (Flow Laboratories) and 4% DEAE- Dextran (Sigma chemical company Limited). After 6 days incubation at 37°C in a CO^ ( 5% in atmosphere ) incubator, 2mls of a second 'overlay ' containing 0.008% neutral red (BDH Chemicals, Eastleigh, Hampshire) was added to each well. After 24 hours incubation at 37°C in the dark, plaques were visualised via the use of a light box (Genetic Research Instruments).

E ) Virus TCIDg Assay

All the virus assays were performed on SW13 cellmonolayers prepared in 24 well tissue culture plates(Nunc or Gibco). The monolayers were washed once with PBS and then ten-fold dilutions of the relevant virus were made in PBS and lOOul of each dilution was inoculated onto the monalyers of an appropiate wellacross the row of five wells, as one remains as acontrol plate. This procedure was repeated with the same virus dilutions across the plate with the other three rows of six wells. Following 30 minutes incubation at RT the cell monolayers were overlaid with 2mls of virus maintenance media and placed in a CO^ incubator for up to four days. After four days incubation the virus media was poured off and the cell monolayer was stained with 0.2% crystal violet. The tissue cell

90

destruction was noted and used to work out the TCID^^ for the particular virus using the equation below:

TCIDgQ = M - A (S - 0.5)

M = Dilution containing the highestconcentration of virus

A= LOG q dilution factorS = Sum of proportion of positive

cultures0.5 = Constant

III Haemagglutinin Titrations and Haemagglutination Inhibition Assays

A) Haemagglutination Titrations

Haemagglutination assays were carried out according to that of Clarke and Casals (1958). and asmodified for use in microtitre plates by Sever et al.(1964).

The diluent Borate Saline Bovine Albumin (BABS),(50mM boric acid, 120mM sodium chloride, pH 9.0 withsodium hydroxide, containing 0.4% [w/v] bovine serum albumin) was first added (50ul/well, except for the first row) to the plates (96 conical well plates, [NUNC or GIBCO]).

The viral haemaglutinin was produced by

91

propagating the appropiate virus for up to 5 days in tissue culture flasks containing confluent monolayers of C636 cells. The clarified virus supernatent was then assayed for HA activity, whereby 50ul of virus was added to the first well of each row of the plate and diluted two-fold across the row of the microtitre plate.

Gander erythrocytes (Tissue Culture Services, Slough) suspended in dextrose-gelatin-veronal buffer (0.058% [w/v] veronal (barbitone), 0.038% [w/v] sodiumbarbitone, 0.06% gelatin, 0.14mM calcium chloride,0.49mM magnesium chloride, 1.45 M sodium chloride, 55.6 mM D-glucose, pH 7.4), were standardised so that a 1:40 dilution of the cell suspension had an optical density of 0.45 at 490nm. The cells were then diluted 1:240 with the 'red cell' pH diluent, a buffer made up of monobasic (150 mM sodium chloride, 200 mM monobasic sodium phosphate) and dibasic (150mM sodium chloride, 200mM dibasic sodium phosphate) sodium phosphate solutions mixed in predetermined proportions to produce the desired final pH after 50ul was added to the plates containing an equal volume of BABS, pH 9.0.

The plate was incubated at room temperature for upto 90 minutes and then read to give the optimum pH for the highest HA titre of the virus showing complete agglutination.

92

B) Haemagglutination Inhibition Assays

These assays were carried out after the HA titre was noted on day five using the same sort of 96 welled plate as in (A) above. BABS diluent was added (25ul) to each well (except the first row) and the monoclonal antibody (1:10 dilution) was serially diluted in two fold dilutions (begining with 50ul in the first well and then carrying over 25ul each time). Eight HA units of the YF virus in 25ul of BABS was added to each well and then the plate was incubated for two hours at room temperature. Then a 50ul aliquot of gander erythrocytes (diluted 1:240 in the appropiate ’red cell’ pH diluent) was added to each well. After 60 minutes incubation the HI antibody titre was read as the highest dilution to show complete inhibition of haemagglutintion.

The results of HAI assays can be converted into Log^Q values (Appendix) and used to produce a dendrogram by using a computer programme called Clustan Analysis as described by Fitch and Margoliash (1967) and Rico-Hesse et al. (1987).

2: MONOCLONAL ANTIBODIES

I. Production of ascitic fluid from hybridomasAn ampoule containing the desired hybridoma was

obtained from liquid nitrogen, thawed out rapidly at

93

37°C and centrifuged at 2000 RPM for five minutes. The supernatant containing dimethyl sulphoxide was poured off and the cells were resuspended in Leibovitz L15 medium (FLOW) supplemented with 10% FCS and 0.2% hypoxanthine thymidine (HT)(SIGMA) in a 25cmf tissue culture flask.

When the flask was confluent, the hybridoma cells were transferred to a 80cm^ and then to a 175cmf by simple aggitation of the flask. The contents of a confluent 175cm? was then centrifuged at 2000 RPM for 5 minutes and the supernatant was poured off. The cell pellet was then washed in PBS, centrifuged as before and then resuspended in 300ul of PBS. The cell suspension was then inoculated intraperitoneally into a previously (4-7 days earlier) ’pristaned' (intraperioneal injection of 0.5ml of 2, 6, 10, 14-tetra methyl pentedecane[Sigma]) female Balb C mouse (six weeks old, OLAC, UK ). This procedure was repeated twice and the mouse examined daily for the swelling of the peritoneal cavity due to the build up of ascitc fluid. Two to four weeks after the third innoculation the fluid was 'drained’ by the insertion of a wide bore hypodermic needle (19G, Luer, Reading, Berkshire) into the peritoneal cavity. This allowed the ascitic fluid to run off into a sterile 30ml plastic universal (NUNC) which was centrifuged at 2000 RPM for 5 minutes, then the ascitic fluid was pippetted

94

off and stored at -20°C. This procedure was repeated on consecutive occasions whereby ascitic fluid from each ’draining' was pooled until no more ascitic fluid was produced. Then the ascetic fluid for that particular hybridoma was frozen at -70°c in 1ml aliquots until required.

II. Properties of McABs Obtained from other laboratories

McAbs were kindly provided as freeze dried ascitic fluids by E.A. Gould, J.T. Roehrig, M.K. Gentry and J.J. Schlesinger and stored at -20°C until required. McAbs were prepared for use by thawing out at 37oC and then rehydrating with the required amount of distilled water. The properties of the McAbs obtained and used are listed in TABLE 2.5.

III. Properties Of McAbS Produced in the Laboratory McAbs already produced against YF17DD BRAZIL

vaccine virus (H McAbs) by Sil et ^ . , (manuscript in preparation) and YF FNV IP vaccine virus (F McAbs) prepared by A. Jennings were used in this study and their properties are shown in TABLE 2.6.

95

>1 >H QA p;EH P•Ü Is O0) M Ü•Ü ü =-H

> • <o « •u Hp4 s

ÜH

H§

EHb

HWiCQh)i-g

«a P0) m CN •H0 1 1 AO H H •H oc P P M (Ng Ü Ü < 1E P P DE M H rH Q Q >H H

mdk'H>HW

Cw

0CNJ1QrHk>H

+ + + + + + + +

H + + + + + + + + + + + +

5*-HO-H-HO&CQ

0u1CQ

&OUo

IITSrH•H15Ipq>H

CDC-Hüü>Ik>4

a a ad b d0 0 o

CD k o u a aa

CDO a Ü CM ü d d a a a>1 I 0 o d d d >i

H S EH iH P rH k o 0 0 EHCO 1 rd td CP k y k 1

à-H h -H rH •H A A ü CD CDA >H A k A b d >4U >4 k CQ CQfd td tda a a

mOJ

0)go(0H

(ÜCN]ütpH

mH

td td td td < td td td td td td td tdCM CM co CN CN CN CN CN CN CN CN CN CN CNCD « CD CD CD CD tP CD CD CD CD CD e o üCP P tP tP CP tp CD tp tP CP tP CP tp tp tpH H H H H H H H H H H H H H

(NÜ

en CN rH en1 1 1 1 1 o

rH CN CO in n 00 m p U m u rHrH rH rH H CN TT KO 10 IT) H CD en en HrH CO 00 00 00 00 00 m m m m < CN

CN vo CD CD CD

96

TABLE 2,6 PROPERTIES OF THE F AND H McAbs SERIES

McAb Specificity HAI PRNTYF

Immunogen

FIO Partial Group + +

F14 Group + +

F16 I I NT +

F17 Partial Group + +

F19 I I - +

F20 I I + +

F21 I I NT +

F23 I I + -

F25 Group + -

F27 I I NT -

F28 YF & Wesselsbron + -

F33 Partial Group + —

H3 YF-Type + +

H5 YF-Vaccine + +

H6 YF17D Vaccine + +

H8 Partial Group + -

HIO 17DD sub-strain + +

Hll Partial Group + -

H14 YF-Vaccine + Asibi + —

H16 Group + -

H17 Partial Group + —

H20 I I + —

H21 I I + -

H22 I I + -

H23 I I + -

H24 I I + NTH25 Group + +

H27 Partial Group + —

H29 I I + NTH30 Group + —

H31 I I + +

H32 I I + -

H33 YF-Type + +

FNV-IP

17DD BRAZIL

NT: Not Tested

97

IV. Determination of the Protein Content of Ascitic Fluids.

McAb ascitic fluid was diluted 1:1000 mixed with the protein sensitive reagent (diluted 1:5, BioRad protein assay mixture) and after 15 minutes incubation at room temperature vortexed using a whirlimixer. The absorbance of this mixture was then read with the use of a spectrophotometer at a fixed wavelenght of 595 nm and the result compared graphically to the curve of the standard protein provided.

V. Isotyping of McAbs

The McAbs were isotyped with the use of a mouse monoclonal antibody isotyping kit (Amersham). The kit provides the reagents for determining the class, subclass, and light chain type of mouse monoclonal antibodies.

A 1:500 dilution of each ascites fluid under test is prepared in PBS and 5 mis of this is added to a container containing a sample stick, then incubated for 15 mins at RT. These typing sticks are nitrocellulose membranes having segments each containing peroxidase labelled goat antimouse antibodies specific for each Isotype class. A final segment of the stick is a

98

positive control which contains mouse antibody.After incubation the fluid was decanted off and

the stick was washed twice with 5mls of a 20mM Tris- buffered Saline (137mM sodium chloride)-Tween (0.1% V/V Tween 20) (TBS-T). After decanting off the last wash, 5mls of a 0.2% (diluted in TBS-T) solution of the supplied peroxidase labelled antimouse antibody was added and incubated for 15 mins. Then the sticks where washed twice with TBS-T and then 5mls of a substrate (4- chloro-l-naphthol [one tablet] in lOmls methanol addedto 30% v/v solution of hydrogen peroxide in TBS-T) andincubated with agitation for fifteen minutes. Thesubstrate was decanted and after the sticks were washed with three changes of distilled water H20 the results were read.

VI. Preparation of multispot immunofluorescence slides

25 cm^ tissue culture flask(s) (NUNC) containing a monolayer of Vero cells was inocculated with the chosen virus(es) as descibed in Part 1 (II). The tissue culture flask was incubated at 37°C in a CO^ incubator for 48hours. Before any CPE was observed, the Vero cells were trypsinised, centrifuged at 1000 RPM and resuspended in 3 mis of virus growth media. Then 25ul aliquots of the cell suspension containing approximately 10^ cells were

99

pippetted on to each spot of the twelve spot Poly Tetra-Flouro-Ethylene (PTFE) coated slide(s) (Hendley Essex) which were then incubated at 37°C for 4-12 hours. The spot slide(s) were then washed once in PBS, then once in ice-cold acetone and then finally fixed by placing in fresh ice cold acetone forlS minutes. The spot slides were then stored at -20°C prior to use.

VII. Indirect Immunofluorescence Tests

For immunofluorescence the spot slides were thawed out at room temperature and then incubated for 30 minutes at 37oC with 25ul of each McAb (diluted 1:50) required to be tested. The positive control was a

(Htà)flavivirus group reactive McAbând the negative control was PBS.

The spot slides were washed twice for 15 minutes with PBS, air-dried and 25ul of 1:40 dilution of goat anti-mouse polyvalent immunoglobulin conjugated to fluorescein isothiocyanate (SIGMA) was added to each spot and the slides were incubated at 37°C for 30 minutes. The spot slides were then again washed twice with PBS for 15 minutes and then once in distilled water. After finally being air-dried, mounting media (10% saline in glycerol) (Sigma chemical company limited) was added and the slides were examined under

100

ultraviolet light with the aid of a Leitz dialux 20 epifluorescent microscope for fluorescence. Results of immunofluorescence tests were confirmed by two individuals reading slides independently.

VIII. Plaque Reduction Neutralisation Tests (PRNTs)

A) Serum neutralisation Indices

These were performed in 6 well tissue culture plates (NUNC), using the plaque assay technique as described in 1. II D (above) except that the virus dilutions were incubated overnight at 4°C with a 1:20 dilution of McAb prior to infecting a confluent cell monolayer of Vero cells, as described by Buckley and Gould (1985) and modified by Barrett et (1989) whoused 1% noble agar and 0.015% DEAE in the overlay medium. The assays were stained on the sixth day with neutral red and the results were read on the seventh day as neutralization titres expressed as the LOG^g reduction in virus titre.

B) The 100 PFU Test

This was also performed in six well dishes as described above except that the virus was diluted to a concentration of 100 pfu in lOOul of PBS and incubated

101

with lOOul of McAb, diluted 1:20 in PBS for 24 hours at 4°C. This mixture was then added to a confluent Vero cell monolayer, incubated at room temperature for 30 minutes and then overlaid with overlay medium. The results were recorded after seven days and expressed as either full neutralisation (90% or more) where the McAb neutralised all the virus or negative neutralisation where the virus was either partially or not neutralised by the McAb. If a McAb fully neutralized 100 pfu of a particular virus, this assay was repeated using the PRNT technique to gain the serum neutralization index (SNI), stated as the log^g reduction in virus titre.

3: PRODUCTION OF MONOCLONAL ANTIBODYNEUTRALISATION ESCAPE VARIANTS

I. Plaque Purification Of Single NeutralisationEscape Variant Virions from Plaque Reduction Neutralisation Tests.

A) Preparation of Neutralisation Escape Variants(McAb^)

If a particular McAb neutralised one of the virus strains approximately 4000 fold or more in a SNI PRNT, individual plaque isolates that had escaped

102

neutralisation were cloned by plaque purification as described by Wiktor and Kowprowski (1980). This method involved the use of a small glass pasteur pippette to suck up an individual plaque isolate in a plug of agar, which was then frozen in 1ml of virus growth medium at -70°C. The antibody escape variant was thenthawed at 37°C and lOOul was used to inoculate confluent monolayers of SW13 cells. After 30 minutes incubaton at room temperature virus maintenance media was added. The virus infected cells were incubated at 37°C for up to seven days or until 90% CPE occured to prepare a seed stock. The virus was then clarified by centrifugation and frozen in 1ml aliquots at -70°C until required.

B ) Confirmation of McAb Neutralisation Escape Variant

The putative McAb neutralisation escapevariant (McAb*) was first re-identified by repeating the antibody treatment protocol to observe that this individual plaque isolate still escaped neutralisation for the second time. Secondly, spot slides of virus infected Vero cells of this seed stock were used inindirect immunofluorescence tests. This was to check whether the epitope for that particular McAb used toproduce the neutralisation escape variant was present or absent on that virus. If the variant re-escaped

103

neutralization and the epitope for that particular McAb was absent, the virus was deemed to be a true neutralization escape variant.

II. Producing Neutralisation Escape Variants by aImmunoprécipitation Technique

The virus required in the experiment was diluted to give 1x10^ PFU in 250ul of PBS and was mixed with1.25ul, 12.5ul, or 25ul of pre-spLun (2 minutes at13000 RPM in a bench-top microfuge) ascites fluid of relevant McAb. After this mixture was incubated for 45 minutes at RT, 25ul of Goat anti-mouse immunoglobulin G (SIGMA) was added, then the mixture was incubated at 37°C for 30 minutes and then 25ul of 10% (w/v) Protein A (SIGMA) was added. After 45 minutes incubation at RT the mixture was microfuged at 13000 RPM for 5 minutes and the pellet was discarded. Then 20 ul of the supernatant (S/N) was diluted in a serial ten fold dilution and plated out on six well plates (Nunc) for each McAb/virus reaction at each dilution as described before for virus plaque assays. While the rest of the reaction mixture was frozen at -70°C until required in further experiments. Production of escape variants were assumed if the virus/McAb plaque assays gave a 4000 fold reduction in virus titre.

104

III. Producing Neutralisation Escape Variants in TissueCulture.

The appropiate McAb (6ul) was added to 120ul of the desired virus and incubated for 45 minutes at RT before being inoculated onto a confluent layer of SW13 cells. This mixture was then incubated on the cell sheet for 30 minutes and after virus growth medium was added they were placed in a CO^ incubator and checked every day for signs of CPE. The virus was harvested on the production of 90% CPE or after six days growth whichever the earlier. This procedure was repeated twice more and then each virus passage was tested for its ability to escape neutralisation with the relevant McAb. If the virus population escaped neutralisation and the epitope for that particular McAb was absent in IIP studies it was de&med to be a neutralisation escape resistant variant.

IV. Producing variant virus using an Elisa Plate Method

The specific monoclonal antibody ascitic fluid(s) is first clarified by centrifugation at 1300 RPM for 2 minutes and then diluted 1:20 in coating buffer pH 9.6 (15mM Na^COg /35mM NaHCO^) (Wreghitt and Morgan-Capner, 1990). Then lOOul of the diluted McAb is added to a well

105

of a flat bottomed ELISA plate and incubated 0/N at 4°C. The remaining solution in the well of the plate is then pippetted off and lOOul of virus containing 1x10* PFU is added to each well. After the plate is incubated for 2 hours at RT, most of the virus is pipetted into cryotubes and frozen at -70°C. The remaining 20ul of virus is serially diluted in PBS and plagued on six well tissue culture plates. If there is a 4000 or more foldreduction in virus infectivity the virus samples areassumed to be antigenic variants and individual plaquepurifications were made. These were then grown up andstudied in indirect immunofluorescence tests to see if the epitope for that particular McAb is present or absent. If absent the virus samples are assumed to be true variants.

4: VIRAL PATHOGENESIS STUDIES

I. Mouse Pathogenesis Studies

A) Adult MiceIn pathogenesis studies, female mice 4-5 weeks old

and weighing 20-25 grms (Strain TO, OLAC, U.K.), were used. Light ether anaethetised mice were inoculated from either a freshly thawed, whirlimixed ampoule of virus or

106

a diluted, whilimixed preparation of virus. Mice were inoculated with either 20ul of virus by the intracerebral route or 20ul of virus by the intranasal* route. The mice were then monitored daily for up to 37 days and mouse illness or death per day was recorded. The average survival time (AST) plus or minus the standard error of the mean (SEM) was calculated for each experiment where by :

AST = No. of deaths / day + SEM = <Tn-l Total No. of mice JrT"

B) Newborn miceNewborn mice no-more than 48 hours old were

inoculated by the intraperitoneal route with a 50ul volume of virus containing 200 pfu. Mice were monitored for up to 15 days and mouse death was recorded daily. The average survival time (AST) plus or minus the standard error of the mean (SEM) was calculated for each experiment as above.

II. Monkey Pathogenesis Tests

Cynmologus monkeys were inoculated with 1x10^ pfu in a volume of 500ul by the intracerebral route. Monkeys were examined for signs of encephalitis for up to 30 days.

1 0 7

5: TEMPERATURE SENSITIVE STUDIES WITH YF VIRUSES

I. Growth of The VirusVirus was propagated in either SW13 or C636 cells

for three or five days respectively, then clarified by centrifugation and frozen in 1ml aliquots at -70°C until required for use.

II. The Virus Assay

Plaque assays were performed as described previously except that LLCMK-2 or Vero cells were used and the assays were incubated at 37°C or 39.5°C. The efficiency of plating of the virus which compares the ability of the virus to plaque at both the two temperatures was caculated by using the equation below:

E.O.P. = pfu/ml at 39.5 C pfu/ml at 37.0°C

108

6 : Radio-immuno-precipitation of virus polypeptides

One 5cm petri dish containing a confluentmonolayer of Vero cells was infected with the YF17D- 204 UK vaccine virus using an MCI of 5, another which lOOul of PBS had been added was used as a the mock infected control dish. The two petri dishes wereincubated for 30 minutes at room temperature and then

FC5 Mmwiaaaaogprior to incubating in a CO^ incubator medium was added to the petri dishes. At 48 hours post infection the virus medium was removed and 1ml of PBS was added to the plate which was incubated at 37°C for two hours to starve the cells of methionine. Then after 50 hours post infection the PBS was removed and SOOul of PBScontaining 25uCi ^ S was added and incubated at 37°C for 4 hours. Then the label was removed from the petridishes and the cells were then removed from the petridishes in a volume of 500ul to 1.5ml microfuge tubes. The cells were pelleted by centrifugation at 13,000 rpm for 3 seconds. The S/N was removed and a lOOul of detergent was added (PBS containing 1% NP40, 1%deoxycholate and 2% Triton X-100), prior to to whirly- mixing. After incubation at 37°C for 30 minutes thesamples were frozen at -20°C.

On thawing the sample, 20ul of the radiolabelledS/N was added to 2ul of clarified McAb with the addition

109

of 80ul of detergent, mixed and incubated at 37°C for one hour. Then lOOul of excess protein A was added and the sample was again incubated at 37°C for 1 hour. Then the complex was pelleted by centrifugation at 13,000 rpm for 2 minutes, the S/N was removed and the pellet was washed with 200ul detergent. After centrifugation for 2 minutes, the S/N was discarded and the pellet wasresuspended in 50ul of detergent and after addition of20ul of cracking buffer (10% SDS, 2*5% 2-mercaptoethanol and 25% glycerol in lOmM Tris pH 7.2. Coloured with small amount of bromophenol blue)and boiled for 3 minutes.

The samples were then seperated on a 10 30%linear gradient polyacrylamide gel. After staining with coomassie blue (Ig in lOOmls methanol, lOOmls distilled water, 36 mis glacial acetic acid) anddestaining (7% glacial acetic acid, 50% methanol, and 43% distilled water) the gel was dried and put up for autoradiography.

110

RESULTS

CHAPTER 3

DERIVATION OF McAbs

111

I. Introduction

The flavivirus E-protein has been shown to beimportant for the induction of protective immunity, including neutralization antibodies. The E-protein also encodes the biological properties of haemagglutination (Della-Porta & Westaway, 1977; Heinz et al., 1981; Kitano et , 1974; Quereshi & Trent,1973) and haemolysis (Cammack & Gould, 1985).

To examine the E-protein of YF virus in detail panels of McAbs prepared against3ifferent wild-typeand vaccine strains of YF virus and recognising different epitopes are required.

Therefore, this chapter describes the derivation and properties of three series of McAbs which are used in later studies in this thesis. Series A was prepared by Ms. A. Pryde, series B by A.R. Medlen and series G by Ms. G. Lewis in collaboration with the author. Basically, hybridoma cell lines already established andsecreting McAb were cultivated to produce ascites inmice as described in the Methods section. Two series of hybridoma cell lines (series B and G) involved 17DD Brazil as the immunogen and the other (series A) used YF17D-204 WHO as the immunogen. These three panels of McAb containing ascitic fluids were used to examine the 17D-204 WHO and 17DD Brazil vaccine virus strains in IIF

112

tests, PRNTs and HAI assays to investigate the antigenic and biological properties of the epitopes recognised by the McAbs.

II. Results

A) Cloning and selection of ’G ’ hybridoma coloniessecreting McAb

Hybridoma colonies showing vigorous cell growth were selected for cloning by serial two-fold dilution of the cells and clones obtained at the end point were screened by IIF for the presence of antibodies to YF17DD Brazil. The positive single clones were expanded and recloned and those still positive by IIF tests were grown up for inoculation into pristane treated mice.

B ) Production of McAb containing ascitic fluids from the A, B and G series of hybridoma cell lines

Hybridoma cells from the A, B and G series were inoculated into pristane treated mice as described in the Methods section. Ascitic fluids were produced from each hybridoma cell line, clarified by centrifugation, aliquoted and stored at -70°C until

113

required. In total three A (A3, A4, A6), six B (B19, B26, B37, B39, B44 and B45) and seven G (G6, G14, G16, G23, G46, G49 and G65) McAb containing ascitic fluids were collected for use in future experiments.

C) Antigenic reactivities of ascites fluid in Indirect Immunofluorescence tests

i) The A, B and G series of ascites fluid

A selection of viruses from different flavivirus sero-groups as described by Calisher et (1989) wereused in indirect immunofluorscenee tests (IIF).

The majority of the McAbs (A3, A4, A6, B19, B26,B37, B44, B45, G6, G14, G16, G49, G65) were found to bepartially flavivirus group reactive, only McAbs B39 and G23 were YF type specific and McAb G46 reacted with all flaviviruses tested (TABLE 3.1).

McAbs G6, G14 and G16 did not react with allthe wild-type strains of YF virus studied. McAb G6 did not react with Asibi, McAb G14 did not react with Dak 1279, and McAb G16 did not react with Asibi and B4.1.

114

TABLE 3.1 RESULTS OF INDIRECT IMMUNOFLUORESCENCE

Viruses^

STUDIES WITH THE A, B AND G SERIES OF ASCITES FLUID

McAbsA A A B B B B B B G G G G G G G

1 2 3 3 4 4 1 1 2 4 4 63 4 6 9 6 7 9 4 5 6 4 6 3 6 9 5

17D-204 + + + + + + + + + + + + + + + +17DD + + + + + + + + + + + + + + + +FNV IP + + + + + + + + + + + + + + + +ASIBI + + + + + + + + + - + - + + + +B4.1 + + + + + + + + + + + - + + + +FVV + + + + + + + + + + + + + + + +Y5 + + + + + + + + + + + N + + + +DAK 1279 + + + + + + + + + + - N + + + +

LGT TP21 JE 5-3 WN-INDIA DEN-2 NGC

+ + + - + -

+ + — — — +

— + — — — +

+ + +

+ - N + — N + + N

+ + +

+ - +

+ + -

+ - -

+ :

N:$.

POSITIVE REACTION NEGATIVE REACTION NOT TESTEDABBREVIATIONS FOR FLAVIVIRUSES ARE THOSE DESCRIBED IN THE METHODS CHAPTER

115

D ) Serum Neutralisation Indices (SNI) of McAbs from the A, B, and G series against their respective virus strains

Plaque reduction neutralisation tests (PRNTs) were performed using the technique described in the Methods section to test the ability of each of the McAbs to neutralise the YF17D-204 WHO and the YF17DD Brazil vaccine viruses. Using virus grown in the SW13 cell line, four (McAbs A6, B37, B39, B44) of the sixteenMcAbs neutralized the YF17D-204 WHO vaccine strain whil only two (McAbs G15 and G46) neutralised the YF17DD Brazil vaccine strain (TABLE 3.2). It was noted that none of the B series of McAbs neutralised 17DD Brazil (the immunogen for this series of McAbs) and only three of them (B37, B39, B44) neutralised YF17D-204 WHO, withMcAb B39 neutralising this vaccine virus to a high titre ( 4 LOG^g units ).

E ) HAI activity of McAbs from the A, B and G series against their respective virus strains

McAbs from the A, B and G series were used in HAI assays as described in the Method section to determine if they elicited HAI activity against the two YF vaccine strains 17D-204 WHO and 17DD Brazil. Interestingly,

116

TABLE 3.2 SERUM NEUTRALISATION INDICES (SNI) OF McAbs FROM THE A, B and G SERIES

McAb YF Vaccine Viruses17D-204 WHO 17DD Brazil

A3 — —A4 — —A6 2.0*B19 -B26 — —B37 1.5B39 4.4B44 1.0B45G6 “ —G14 — —G16 - 1.0G23G46 - 1.0G49 — —G55

SNI’s ARE EXPRESSED AS THE LOG^q REDUCTION IN VIRUS TITRESNI <0.5

117

McAbs A3, A4, A6 did not elicit HAI activity with YF17D- 204 WHO (the immunogen for these McAbs), similarly McAbs B19, B39, 06, 014, 023, 046, 049 did not elicit HAIagainst their immunogen; YF17DD Brazil. However, the majority of these McAbs did elicit HAI from the other virus, for example McAb 06 had a HAI titre of 320 against the 17D-204 WHO vaccine virus and McAb A3 had a titre of 80 against the YF17DD Brazil vaccine virus (TABLE 3.3). McAbs A6, B37 and B45, reacted to low titre with only YF17DD Brazil while McAbs B26, B44, and 065elicited HAI against both YF vaccine virus strains tested. In contrast, McAbs A4, B19 and 049 elicted noHAI activity against either vaccine virus studied.

F) Isotype Characterisation of the A, B and O series

Using an isotyping system supplied by Amersham pic, the A, B and O series of McAbs were isotyped using the methodology described in chapter 2. While all the A and B series of McAbs were of an IgOl subtype, the 0 series of McAbs are nearly all of the Ig02a isotype except for McAb 065 which had an Ig02b isotype.

O) RIP of virus Polypeptides

The A and B series of McAbs have already been usedin an radio-immuno-precipitation protocol and run on

118

TABLE 3.3 HAI ACTIVITY OF McAbs FROM THE A, B and GSERIES

McAb YF Vaccine Viruses17D-204 WHO 17DD Brazil

A3 - 80A4 -A6 20B19B26 40 40B37 - 20B39 40B44 20 80B45 - 20G6 320G14 40016 NT NT023 80046 40 —049065 40 160

: HAI TITRE EXPRESSED AS THE RECIPROCAL OF THE DILUTION INHIBITING HAEMAOOLUTINATIONHAI TITRE LESS THAN 20

NT; NOT TESTED

119

polyacrylamide gels. They were all shown to recognise the E-protein of YF virus (A.D.T. Barrett, personal communication) and McAb A6 recognised a conformational epitope (M. Bouloy, personal communication). One of the G series of McAbs, G23 was used in a radio-immuno- precipitation reaction as described in Chapter 2. The final reaction was run on a PAGE gel, stained, destained and set up for autoradiograhy. FIGURE 3.1 shows that McAb G23 precipitates the E (p54) protein of YF virus from infected cells.

Ill. Discssion

This Chapter descibes the characterisation of 16 McAbs produced in the Author's laboratory using two different YF vaccine virus strains as immunogens (17D- 204 W.H.O. and 17DD Brazil). The properties of the McAbs are summarised in TABLE 3.4. The McAbs were tested for reactivity in IIF tests with three of the eight representative antigenic subgroups of the genus Flavivirus. Those subgroups were the JE/WN, TBE, and dengue complex groups and also the three vaccine strains of YF virus (17D-204, 17DD, and FNV IP) and five wild-type strains from Africa and South America were used. Preliminary analysis in IIF tests using McAbs from the A and B series (Barrett et al., 1989) showed that most

120

FIGURE 3.1 AUTORADIOGRAPH OF PAGE GEL OF SEPERATEDVIRUS POLYPEPTIDES

Track A: non immunoprecipitated virus infected cell lysatesTrack B: E-protein from virus infected cell lysates

immunoprecipitated with McAb G23

NS5 (P98)_

NS3 (P77)_

E (p54)_ NS1 (p4fit

actin_

NS4a (p26)_

PreM (p23).

NS2a (pi91NS2b (p17l

A B

121

McAbs were YF type specific except for McAbs A6, B19,B26, and B44 which were partial group reactive. However, using different viruses from the flavivirus serogroups showed that the majority of McAbs were partiallyflavivirus group reactive except for McAb B39 whichremained YF-type specific.

In the IIF tests with the three series of McAbs, A, B and G, three antigenic specificities were identified. These were; flavivirus group common,flavivirus partial group reactive and YF-type specific.Previously, Schlesinger et (1983) identifiedantibodies showing YF17D-204 sub-strain, YF-type andflavivirus group specificities. Gould et (1985)identified five antigenic specificities on the E-protein of a YF vaccine virus. These were; 17D-204 sub-strain specific, vaccine (17D-204, 17DD, and FNV) specific, YF- type specific, partial group specific and flavivirus group reactive. Thus none of the McAbs had the YF substrain or vaccine specificity described by the previous workers.

The 16 McAbs were also used in PRNTs and HAIassays with both virus immunogens and showed a rangeof patterns of reactivities. Similar reactivities have been reported by Schlesinger ^ al. (1983) and Gould et al. (1985). It was noted that most of the McAbs would

122

TABLE 3.4 SUMMARY OF PROPERTIES OF THE A, B AND G

McAbA3A4A5B19B26B37B39B44B45G5G14G16G23G46G49G65

McAb SERIES USING ONLY THE 17D-204 WHOAND 17DD BRAZIL VACCINE VIRUSES TO

IsotypeIgGl

NTIgG2a

IgG2bIgG2a

STUDY THE McAbs

Specificity Partial Group

YF-Type Partial Group

YF-Type Group

Partial Group

YFHAI PRNT Immunogen

+++

++NT++

17D-204 WHO

17DD BRAZIL

NT

POSITIVE REACTION NEGATIVE REACTION NOT TESTED

123

not neutralise the virus immunogens (TABLE 3.2) and most of the Mcabs had little or no HAI activity with the homologous virus (TABLE 3.3). All McAbs produced are of the IgG isotype either IgGl, IgG2a or IgG2b similar to those McAbs produced by Schlesinger et (1983) andisotyped by Monath et (1984).

Although the McAbs descibed in this Chapter were only studied with the two parent strains in biological assays (HAI and PRNT's) it still gives an indication of differences n the biological function of the E-protein epitopes that they recognise. Thisindicates that care should be taken in theinterpretation of the biological function of the same epitope on different viruses. The protein content of the McAb containing ascitic fluids were determined as described in Chapter 2 and found to be similar (3- 5mg/ml) (data not shown). Thus, the difference in biological function could not be attributable to the variation in concentration of McAbs. The heterogeneous response of viruses being either neutralised or fully resistant to neutralisation has been reported previously by Buckley & Gould (1985). They stated that a 'critical area' on the virus is required. This 'critical area wasfirst referred to by Della-Porta and Westaway (1977 b).In the next Chapter the response of McAbs in biological assays (PRNT's and HAI assays) against a variety of

124

strains of YF vaccine viruses is examined,

125

CHAPTER 4

ANTIGENIC ANALYSIS OF EPITOPES ON YELLOW FEVER VIRUSES BY HAI ASSAYS AND PRNT's

126

I. Introduction

In the previous Chapter differences in the abilities of epitopes of the E-protein of YF vaccine viruses to elicit HAI and neutralisation activities with McAbs were observed. In this chapter, a large panel of McAbs (A, B and G series described in the last chapter and two other McAb series prepared in the authors laboratory [the F series of McAbs prepared by A. Jennings and the H series of McAbs prepared by B.K. Sil]) were used to analyse 31 YF vaccine viruses in HAI assays and PRNTs and compared with three wild-type YF viruses.

In addition, polyclonal anti-sera were also used in both HAI assays and PRNT's to compare their properties of eliciting HAI and neutralising activity with the McAbs used in this Chapter.

II. Results

A) Study of the Nigerian and USSR vaccine by IIF, HAIand PRNT

The two new YF 17D vaccines were studied using YF-type (McAb H3), YF vaccine specific (McAbs H5, 411), 17D strain (McAb H6), 17DD (McAb HIO) and 17D-204 (McAb 864) sub-strain specific McAbs in IIF, HAI and PRNT's.

127

As seen in TABLE 4.1, both vaccine strains have epitopes recognised by all McAbs with the exception of McAb HIO in IIF studies. There is a similar reactivity in HAI studies except McAb H3 does not elicit HAI with theNigerian vaccine and H6 does not elicit HAI with theUSSR vaccine strain. In PRNT's, neither of the vaccines were neutralised by any of the McAbs used.

From this data, one is able to assume that bothviruses are derived from the 17D-204 sub-strain of YF vaccine viruses.

B ) Analysis of viruses by HAI assays using the A, B,G and H series of McAbs

i) 17D-204 Vaccine strains

As seen in TABLE 4.2, 32 McAbs were used to screen 17, 17D-204 vaccine viruses and also 4 plaque purifiedpreparations from the vaccines manufactured in France, South Africa and the UK. The vaccines came from twelve different manufacturers as described in the Methods Section. Overall, the vaccines display a heterogeneous HAI response dependent upon the virus and McAbcombination being analysed. Examining particular viruses in detail, for example comparison of two Australianvaccine batches in HAI assays, one having the avian

128

TABLE 4.1

A)

REACTIVITY OF THE USSR AND NIGERIAN VACCINES IN IIF TESTS, HAI ASSAYS AND PRNT's

IIF TESTS

VIRUSES McAbsH3 H5 H6 HIO 411 864

USSR + + + + +NIGERIA + + + + +

B) HAI ASSAYS

VIRUSES McAbsH3 H5 H6 HIO 411 864

USSR + + - + NTNIGERIA + + NT NT

C) PRNT's VIRUSES

USSRNIGERIA

McAbsH3 H5 H6 HIO 411 864- - - — NT NT— — — — NT NT

+: POSITIVE REACTION NEGATIVE REACTION

NT: NOT TESTED

129

TABLE 4.2 ANALYSIS OF YF17D-204 AND THE TWO NEW VACCINE (NIGERIA AND USSR) VIRUSES BY HAI ASSAYS USING THE McAb A, B, G AND H SERIES

: HAI TITRE EXPRESSED AS THE RECIPROCAL OF THEDILUTION INHIBITING HAEMAGGLUTINATIONHAI TITRE LESS THAN 20

NT: NOT TESTEDVIRUS ABBREVIATIONS ARE THOSE USED IN THEMETHODS CHAPTEREXCEPT FOR USA VACCINES WHERE A: P-16065,B: 6145, C: 2848, D: 4E158

130

TABLE 4.2

VIRUSES'McAbs

AUS033.

AUS 1 008

COL915

DAK78LF

FR FRpp FRG HOL IND RF318

SA894A

A3 20 80 — 320 80 160 160 80 20 40 320A4 40 - 20 80 80 - - 20 40 40 20A5 40 160 160 80 — 40 40 40 -

B19 _ 20 20 20B26 160 80 20 40 20 160 80 — 80 20 320B37 40 40 160 160 80 40 80 80 160 20 160B39 80 80 80 - 160 20 160 80 160 — 640B44 160 80 160 320 160 80 80 160 80 80 640B45 80 40 40 320 40 160 40 40 40 40 320

G6 20 40 20G14 NT 640 NT NT NT 40 80 NT NT NT 640G23 160 40 640 80 160 640 80 320 80 640G46 - 40 - - 20 — 40G49 - - - - - - 20 20G65 640 320 160 320 160 160 160 640 - 80

H5 80 20 320 20 20 40 20 80H6 40 320 - 160 - 10 80 - 40 80 40H8 80 640 20 20 80 40 160 40 80 40 40HIO - - - - 20 - . — 20 40Hll 20 - 40 320 20 40 - 40 80 40 160H14 - - - - 20 80 — 40 40H20 160 - 80 320 - . - - 320 40H21 NT NT NT NT 320 NT NT 160 NT NT NTH22 80 NT NT 160 NT 40 NT NT 20 NTH23 - 40 - 320 - - 160 20 160H24 - 640 NT NT 20 40 — 20 i. NTH25 160 20 - - 20 160 80 — 160 — 40H27 320 - NT NT NT 160 20 80 640 NTH30 80 160 - 320 - 20 20 20 80 640H31 40 40 20 160 20 20 160 40 40 320H32 40 320 80 - 160 - 160 160 40 20H33 80 160 - 40 20 40 80 40 80 160 20

131

TABLE 4.2

McAbs VIRUSES*

SA SApp UK UKpp UKpp 10802 +wt -wt A

USA* B C

WHO17D

NIG USSR

A3 80 160 40 80 80 40 80 20 160 160A4 20 - . - 40 40 - - 80 - - 80 40A6 80 40 80 80 20 20 160 — — 320 160

B19 20 _B26 40 80 - - 20 - - 80 320 40 160 —

B37 160 80 40 80 80 - 20 80 40 — 160 160B39 160 640 160 80 80 - 20 160 160 40 80 320B44 160 320 320 640 320 40 40 80 640 20 160 160B45 160 80 20 160 80 20 20 80 160 — 80 40

G6 40 40 320 20G14 NT 640 NT NT NT NT NT NT 640 40 NT NTG23 160 640 40 320 160 20 40 160 320 80 160 80G46 - - 40 20 20 - - — — 40 ' — —

G49 20 20 20 - - 20 - — — 20 —

G65 80 640 160 160 160 ' — — 80 320 40 640 320

H5 40 40 80 40 _ 40 80 20 40H6 20 320 - - - - - 40 160 80 40 —

H8 - 640 80 80 80 - - 20 640 320 40 40HIO - 320 20 20 40 10 - - — 20 _ 20Hll 20 160 80 640 80 320 - 160 — 80 160 40H14 80 - 40 40 40 160 20 80 - 80 — 40H20 - - 160 160 320 160 - 40 - 320 40 320H21 NT NT 80 160 160 NT NT NT NT 160 NT 640H22 80 NT NT NT NT 160 40 40 NT NT NT NTH23 - 640 - - 20 — - - 640 80 _ _

H24 NT - 160 80 40 NT NT - 640 80 — 40H25 40 40 - 80 160 320 - 160 40 - 640 —

H27 - - 80 160 320 NT NT 320 - 80 640 NTH30 160 - 40 20 40 - 40 - — 160 — 20H31 40 20 80 80 40 - 20 - - 80 _ 40H32 40 20 320 160 80 - - - - 320 — 160H33 - 20 40 40 40 - - 80 - 80 - 20

132

leukosis virus (ALV) as a contaminant (008) and theother prepared using an ALV-free seed (033-1), showed that these viruses show more similarity in reaction with the B and G McAbs than to the A and H McAbs. McAbs B26,B44, G55 react to a high HAI activity with both of thevaccines while McAbs H6, H8, H20, H24 and H27differentiate the vaccines by reacting with either one or the other vaccine viruses to a different extent (four fold or more difference in HAI titres). The plaque purified preparations, two from the UK vaccine, one with (+wt) and one without (-wt) a YF wild-type specific epitope (recognized by McAb 117 [Gould, et ^ . , 1989]), one from the French and one from the South African vaccine all showed similar HAI titres as their parent strains. Although there were differences in HAI titres between the McAbs with different viruses, there was considerable homogeneity in the response of particular McAbs (HAI or no HAI) with certain viruses. For example, McAbs B26, B37, B39, B44, B45, G23, G65, H8, H32inhibited haemagglutination to a high dilution with 70% of the viruses studied. In comparison McAbs B19, G46,G49, showed very little or no HAI activity with any of the viruses. Other McAbs (G5, H20, H23) had intermediate HAI activity with the viruses studied, having high HAI activity with some, and low HAI activity with others. It

133

was also noted that McAb HIO, though being 17DD vaccinespecific elicited HAI activity with the 17D-204 SAplaque purified preparation. Two vaccines from new manufacturers (17D Nigeria and 17D USSR) were studied with 34 McAbs. McAbs A6, B37, B39, B44, G23, G65,reacted with high HAI activity with both of the 17Dvaccines, while McAbs A3, B26, Hll, H20, H25 and H32showed considerable differences in HAI activity between both of the vaccine strains. Other Mcabs used eitherreacted with low or no HAI titres with both viruses, or reacted with one and not , the other with low HAIactivity.

ii) 17DD vaccine viruses

The 17DD vaccines and two plaque purifications (one from the 17DD Brazil vaccine, and the other from17DD Colombia) were studied with 34 McAbs. As seen inTABLE 4.3, McAbs from the B and H series had high HAI activity with the 17DD vaccine viruses but to a lesser extent with the two plaque purified preparations. There seemed to be more differences between these plaque purified preparations and their parent strains than the two parent strains to each other. Although McAb HIO is 17DD specific, it only reacts with the 17DD vaccineswith a low HAI activity.

134

TABLE 4.3 ANALYSIS OF THE YF17DD VACCINE VIRUSES BY HAI

McAbASSAYS USING THE A, B, G AND H McAb SERIES

Viruses*BRAZ BRAZpp COL COLpp DAK

A3 80* — _ 160 80A4 - - 20 - 20A6 20 40 40 40 40B1 _ _ _ 20B2 40 40 160 320 40B3 20 160 40 160 20B3 - 160 40 80 -

B44 80 160 40 160 80B45 20 80 80 320 -

G5 — 20 40 20G23 - 20 320 160 -

G45 - - 20 - 20G49 - - - - -

G65 - 40 160 80 -

H3 160 20 320 40 160H5 80 160 80 40 160H6 40 40 40 - 80H8 160 20 160 40 320HIO 40 - 40 40 40Hll 40 160 40 - 80H14 40 - 40 80 80H16 160 NT 160 NT 160H17 160 NT 160 NT 160H20 320 80 320 160 320H21 160 NT 160 NT 160H23 80 - 80 - 80H24 80 NT 80 - 80H25 - 20 - 160 -

H27 160 NT 160 160 160H29 160 NT 160 NT 160H30 320 - 320 - 160H31 80 40 80 - 80H32 640 320 640 160 1280H33 160 - 160 40 80; HAI TITRE EXPRESSED AS THE RECIPROCAL OF THE

DILUTION INHIBITING HAEMAGGLUTINATIONHAI TITRE LESS THAN 20

NT: NOT TESTED*: VIRUS ABBREVIATIONS ARE THOSE USED IN THE

METHODS CHAPTER

135

iii) FNV vaccine virusesThree different preparations of the French

neurotropic vaccine virus (FNV) were examined with 30 McAbs. As shown in TABLE 4.4, there are different groups of reactions. The first, where McAbs elicit very low or no HAI activity with all three viruses (McAbs B19, G46,G49, H6 [17D specific], and HIO [17DD specific]). The second where McAbs (e.g. B26, B37, B44, B45, Hll, H20and H31) elicit HAI activity with all three vaccine viruses (FNV FC, FNV IP, FNV NT), another where McAbs elicit HAI activity with two out of the three FNV’s (e.g. McAbs A3, A6, B39, G23, G65, H5, H8, H14, H22,H25, H30 and H33). The last was McAb H23 which elicitedHAI with only one of the vaccines (FNV IP).

iv) Wild-type YF virusesThree wild-type viruses were examined using 35

McAbs (TABLE 4.5). Two of the viruses were from Africa (Asibi and FVV) and one was from South America (B4.1). Nearly 50% of the McAbs used had high HAI activity with all three of the viruses. McAbs H5, H6, and HIO had noreactivity as they are vaccine, 17D and 17DD specific respectively. Other McAbs which had low or no HAI activity were B19, G46, G49, H23 and also H25 which isflavivirus group reactive. McAbs B26 and G65 had similar HAI titres with the Asibi and FVV wild-type

136

TABLE 4.4 ANALYSIS OF YF FNV VIRUSES BY HAI ASSAYSUSING THE A, B, G AND H McAb SERIES

McAb

A3

YFFNV FC

Viruses* FNV IP

40*FNV NT

80A4 80 - 20A6 20 80 80B19 _ 20B26 80 40 80B37 80 40 40B39 160 20 -

B44 320 160 160B45 40 40 40G6 _G23 160 80 -

G46 - - -

G49 20 40 -

G65 640 80 -

H5 20 160 80H6 - 20 -

H8 40 - 160HIO - - -

Hll 160 40 80H14 - 640 40H20 640 160 320H22 - 160 320H23 - 160 -

H24 - NT 80H25 640 320 -

H27 160 NT 40H30 - 160 320H31 40 80 40H32 - 20 640H33 80 — 160

HAI TITRE EXPRESSED AS THE RECIPROCALDILUTION INHIBITING HAEMAGGLUTINATIONHAI TITRE LESS THAN 20


METHODS CHAPTER

137

TABLE 4.5 ANALYSIS OF WILD TYPE VIRUSES BY HAI ASSAYS USING THE A, B, G AND H McAb SERIES

McAb YF Viruses*ASIBI FW* B4.1

A3 - 160 20A4 40A6 20 -B19 - - -B26 40 80B37 40B39 40 80 40B44 40 160 80B45 80 20 20G6 - 80 80G14 NT 40 80G23 160G46 — 40 —G49 20G65 80 40H3 160 160 80H5 — — —H6 NTH8 160 160 160HlO — — —Hll 160 80 160H14 80H16 320 160 320H17 320 160 160H20 160 320 320H21 160 - 160H23H24 80 80 80H25 - - -H27 80 40 80H29 80 160 160H30 80 160 80H31 80 80 160H32 320 320 320H33 160 80 160

*: HAI TITRE EXPRESSED AS THE RECIPROCAL OF THEDILUTION INHIBITING HAEMAGGLUTINATIONHAI TITRE LESS THAN 20


METHODS CHAPTER

138

strains from Africa but no HAI activity with the wild-type strain B4.1 from South America. McAb G6 had HAI activity with only the F W and B4.1 strains of wild-type YF virus. McAb G23 only elicited HAI activity with Asibi virus, while McAb A3 elicited a high HAI titre with FW.

C) HAI assays with polyclonal anti-sera

A number of YF viruses were examined in HAI assays using polyclonal mouse anti-sera made against 17D-204 USA, FNV FC, and Asibi viruses. As seen in TABLE 4.6, the polyclonal anti-sera elicit high HAI activity with most of the viruses studied. However, there is still a heterogeneous HAI response seen with some of the viruses (e.g. 17D-204 Colombia, the FNV strains and the 17DDstrains) as seen with the McAbs used previously.

D ) Cluster analysis of the HAI assays with viruses against twenty seven of the McAbs used in the study

All the HAI titres for 27 McAbs which reacted with each virus were converted into Log^^ values (Appendix). Any HAI titres below ten were given a log^^ value of 1.0, due to the zero point of the HAI assay was taken as ten. The values were then typed into the computer

139

TABLE 4.6 REACTION OF POLYCLONAL ANTI-SERAIN HAI ASSAYS WITH YF VIRUSES

VIRUSES

17D-204RF31BDAKARCOLOMBIAUKUKpp+wt UKpp-wt USA 6145

POLYCLONAL ANTI-SERA 17D-204 USA FNV FC ASIBI

3208020160640160160

16016020

320320160640

32032040

320320160NT

17D USSR 320 160 160

17DDBrazilppBrazil

20 4040

80NT

FNVFCIPNT

402040

NTNTNT

WILD-TYPEASIBIFVV

640640

640640

NTNT

NT:

HAI TITRE EXPRESSED AS THE RECIPROCAL OF THE DILUTION INHIBITING HAEMAGGLUTINATIONHAI TITRE LESS THAN 20NOT TESTED

140

programme and the cluster analysis was performed on: i) all the 34 viruses, ii) all viruses except the plaque purifications and iii) all the 17D-204 vaccine virus strains were compared to the wild-type Asibi strain.

i) All viruses and plaque purified preparations

FIGURE 4.1, shows the result of the Cluster analysis with all the viruses examined in HAI assays with 27 of the McAbs which reacted with all the viruses. As can be seen there are three main groups, group 1 consists of the wild-type viruses and some of the vaccine strains, group 2 a large group of 17D-204 vaccine strains and group 3 containing only two viruses: the revertant USA virus P-16065 and FNV IP. Wild-type strains Asibi, F W and B4.1 grouped fairly close together in the cluster analysis, although Asibi is separated from F W and B4.1. All three 17DD vaccines are in the same group. Group 1, but while the 17DD vaccines produced in Brazil and Dakar are closely related to each other and to the vaccines RF318 and FNV NT, the 17DD Colombia vaccine is more closely related to the wild-type Asibi virus. It was noted that the majority of the YF vaccines are grouped together into group 2, except for the 17D-204 WHO, 17DD Colombia, 17DD Brazil, 17DD Dakar, FNV NT and RF318 vaccine viruses which are all in group 1, with the

141

FIGURE 4.1. CLUSTER ANALYSIS OF ALL 34 VIRUSES WITH TWENTY SEVEN McAbs WHICH REACTED WITH ALL VIRUSES IN THE STUDY

o o o o • o o oœ œ -Vi O) cn w(O o fU OJ cnIV) u ^ • ■ -U1 O) . CO

-Vio

GROUP 1

ASIBI 17DD COL

17D204 WHO FW

BA. 1 17DD BRAZ 17DD DAK FNV NT RF318

GROUP 2

17D20A HOL 17D USSR 17D20A DK

UKpp+ OKpp-

17D204 FR 17D20A FRpp 17DD COLpp 17D20A DAK

17D NIG 17D20A USA 2848 17D204 AUS 033-1

17D204 IND FNV FC

17D204 SA 10802 17D204 USA 6145

17D204 COL 17DD BRAZpp17D 204 FRG

17D204 AUS 008 17D204 SApp

17D204 USA 4E158 17D204 SA 894A

GROUP 3P-16065 FNV IP

142

wild type viruses. The 17D-204 UK vaccine and its plaque purified viruses were shown previously to react similarly with most of the McAbs (TABLE 4.2) and in the clustan analysis they are also very close together (group 2).

ii) Viruses without the plaque purified preparations

The cluster analysis was recalculated to exclude the plaque purified viruses (FIGURE 4.2). Nonetheless, the viruses still group into a similar pattern as before although some of the viruses have differences in theirdetailed relationships within group 2. For example, the YF17D-204 Dakar vaccine which was grouped with two different plaque purified preparations has now moved closer to the FNV-FC vaccine. YF17D-204 Colombia has moved away and is now closer to group 3.

iii ) 17D-204 vaccine viruses compared to the Asibiwild-type strain

FIGURE 4.3, shows a cluster analysis involving the 17D vaccines and their wild-type parent virus strain Asibi with the 27 McAbs. Of the 19, 17D vaccinesexamined, the 17D-204 WHO ALV-free seed stock showedthe most similarity to Asibi virus. The other 17D- 204 vaccines group together showing a similarity of

143

FIGURE 4.2. CLUSTER ANALYSIS OF VIRUSES WITHOUT PLAQUE PURIFIED VIRUSES WITH THE TWENTY SEVEN McAbs WHICH REACTED WITH ALL VIRUSES IN THE STUDY

O ' O O O o o-Nl -s C7> cn COCD O O fUo> CO CD cn fV) CO COcn

ASIBI

17DD COL17D204 MHO

F W

17DD BRAZ17DD DAK

FNV NTRF318

17D204 HOL17D USSR17204 OK17D204 FR

17D204 SA 10802

17D204 USA 614517D NIG

17D204 USA 284817D204 AUS 033-1

17D204 INDFNV FC

17D204 DAK17D204 FRG

17D204 AUS 00817D204 SA 894A

17D204 USA 4E15817D204 COL

P-16065FNV IP

144

FIGURE 4.3. CLUSTER ANALYSIS OF 17D204 AND WILD-TYPEASIBI VIRUSES WITH THE TWENTY SEVEN McAbsWHICH REACTED WITH ALL VIRUSES IN THESTUDY

ASIBI

17D204 WHO

17D204 HOL

17D ossa17D204 UK

17D204FR

17D204 SA 10802

17D204 USA 6145

17D NIC

17D204 USA 2848

170204 AUS 033-1

170204 IND

R P 318

170204 OAK

170204 FRG

170204 AOS 008

170204 SA 894A

170204 USA 4E158

P-16065

170204 COL

145

reaction with the McAbs used. However, there is still a separate group containing vaccines P-16065 and 17D-204 Colombia which show the least reactivity with the McAbs to the other vaccines.

E ) Analysis of viruses using the A, B, F, G and HMcAbs series in PRNT's

Thirty-two YF viruses were studied in PRNT's with the A, B and G McAbs derived in chapter 3 and also the ’F ’ and 'H ’ series of McAbs whose properties are described in the Methods section (Chapter 2).

i ) Three YF17D-204 vaccine strains and their plaquepurified virus isolates

Three 17D-204 viruses ( France [Fr], UK and South Africa [SA]) were compared in PRNT’s with one plaque purified strain from both the Fr and SA vaccines and two from the UK vaccine strain, one with a wild-type (+wt) and another without a wild-type epitope (-wt) (recognised by McAb 117, [Gould et al., 1989]). Of the 36 McAbs used in PRNT's to survey these YF virus strains, only 18 would neutralise one or more of the vaccines. McAbs FIG, F19, F25, F27, F28, G6, G16, G46, G49, G65, H5, H8, HIO, Hll, H14, H23, H27 and H33 failed

146

to neutralise any of the above viruses. As can be seen in TABLE 4.7, there is a heterogeneous neutralisation response by the McAbs where each McAb only neutralises particular viruses. Only McAb G23, neutralised all the strains used, with the exception of the vaccine manufactured in France. Also, apart from the French plaque purified preparation, McAb B26 neutralises all plaque purified strains and none of the parent strains. It neutralises the 17D204 UKpp- strain to a high SNI value of 3.8. McAbs A3, H3 and H6 only neutralised the 17D-204 SA vaccine and McAb A6 only neutralised the French plaque purified preparation.

McAb G14, neutralised the vaccines from both France and the UK as well as the SApp preparation but not the SA vaccine. McAbs B44 and B45 neutralised the UKpp+wt strain while McAb H25 neutralised both this and the 17D-204 SA vaccine. The F Mcabs only neutralised the UK vaccine strain, except for McAb F14 which also neutralised the UKpp- strain.

ii) Analysis of two 17DD vaccines

Out of the 31 McAbs tested (TABLE 4.8) using thetwo YF 17DD vaccines from Brazil and Colombia, only 7McAbs (F14, F19, G16, G46, H3, HIO, H33) neutralised the

147

TABLE 4.7 SURVEY OF A, B, F, G and H SERIES OF McAbs WITH THREE YF17D-204 VACCINE STRAINS AND THEIR PLAQUE PURIFIED VIRUS ISOLATES IN PRNT's

17D-204 VIRUSESMcAb Fr Frpp SA SApp UK UKpp-wt UKpp+wtA3 — — 1 . 0 — — — —A6 — 0.7 — — — — —B19 — — — 0.7 — — —B26 - - - 1.1 - 3.8 0.9B3 7 — — — — — — —B39 — — — — — 1.7 —B44 — — — — — — 0.8B45 — — — — — — 1.3F14 — — — — 2.3 1.4 —FI 6 — — — — 0.7 — —F17 — — — — 0.6 — —F20 — — — — 0.9 — —F21 - - NT NT 1.0G14 0.7 - - 0.7 1.0G23 - 1.8 1.0 1.9 3.2 1.3 2.3H3 — — 0 . 7 — — — —H6 — — 2.0 — — — —H25 - - 1.0 - - - 0.7

*: ANTIBODY TITRES EXPRESSED AS LOG q REDUCTION INVIRUS TITRE

-: NEUTRALISATION TITRE £ 0.5NT: NOT TESTED

McAbs THAT DID NOT NEUTRALISE ANY OF THE YF17D-204VIRUSES ABOVE ARE LISTED BELOW:FIO, F19, F25, F27, F28, G6, G16, G46, G49,G65, H5, H8, HIO, Hll, H14, H23, H27, H33.

148

TABLE 4.8 SURVEY OF A, B, F, G and H SERIES OF McAbs WITH TWO YF17DD VACCINE STRAINS IN PRNT's

YF17DD VIRUSES COLOMBIA BRAZIL

McAbsF14 - 1.0*F19 - 0.8G16 1.0 1.0G46 - 1.0H3 3.0 3.0HIO - 1.0H33 - 3.0

*: ANTIBODY TITRES EXPRESSED AS LOG^^ REDUCTION INVIRUS TITRENEUTRALISATION TITRE £ 0.5

3 McAbs THAT DID NOT NEUTRALISE ANY OF THE YF17DDVIRUSES ABOVE ARE LISTED BELOW:A3, A5, B19, B26, B37, B39, B44, B45, FIO, F16,F17, F20, F21, F25, F27, F28, G6, G23, G49, G65,H5, H6, H14, H25,

149

Brazil vaccine while only two (G16 and H3) neutralised the Colombia vaccine.

F) Examination of YF viruses with McAbs that elicitneutralisation activity

i) 17D-204 vaccine viruses

As with the HAI results there was a heterogeneous response of epitopes eliciting neutralising activity (TABLE 4.9). Out of the seventeen viruses examined, only seven were neutralised by one or more of the eight McAbs used. McAb A6, neutralised the 17D-204 Colombia and 17D-204 WHO vaccines to 2 log^^s, while only neutralising the revertant USA vaccine P-16065 to nearly 1 log Q. The USA vaccine batch 6145, was only neutralised by McAb B26, while the WHO vaccine was only neutralised by McAb B39. McAb F14 neutralised the UK and WHO vaccine viruses, while McAb G23 neutralised the UK, South Africa and Holland vaccine viruses. The 17D- 204 SA vaccine virus was also neutralised by McAbs H3, H6 and H25. None of the eight McAbs which elicit neutralising activity, neutralised the vaccines produced in either Nigeria and Russia.

150

TABLE 4.9 EXAMINATION OF YF VIRUSES WITH McAbs THAT ELICIT NEUTRALISING ACTIVITY IN PRNT'S

VIRUSES17D-204AUS 033-1AUS 008COL 915DAK 78LFFRFRGHOLINDRF312RF318SAUKUSA 6145 USA P-16065 USA 4E158 USA 2848 WHO17D

17DD

FNV

NIGUSSRBRAZCOLDAKFCIPNT

WILD-TYPEASIBIFVVB4.1

A6

2.4

McAbs B26 B39 F14

3.30.7

2.0

1.01.8

G23 H3 H6 H25

2.3

1.8

1.0 0.73.2

1.9 1.0

4.3 2.5

1.0NT

1.51.91.9

1.8

3.03.03.0

0.6 2.0

ANTIBODY TITRES VIRUS TITRE

1.4

EXPRESSED AS LOG10 REDUCTION IN

NT:NEUTRALISATION TITRE <0.5 NOT TESTEDVIRUS ABBREVIATIONS ARE THOSE USED IN THE METHODSCHAPTER

151

ii) 17DD vaccine viruses

Of the eight McAbs used, only McAb H3 previously shown to neutralise the 17DD Brazil and Colombia vaccines also neutralised the 17DD Dakar vaccine (TABLE 4.9).

iii) FNV vaccine virusesAll three of the FNV vaccine viruses were

neutralised by one or more of the eight McAbs used in the study. McAb F14, neutralised all the FNV vaccines to a SNI value of 1.5 to 1.9. The FNV IP strain was only neutralised by McAb F14, while the FNV FC strain was also neutralised by McAbs A6 and G23. However, the FNV NT strain was neutralised by McAbs A6, F14, H3 and H25.Thus, the FNV vaccine strains show a heterogeneity of epitopes eliciting neutralisation, like that of the 17D- 204 and 17DD vaccine strains.

iv) Wild-type YF viruses

Of the eight McAbs used, only one McAb neutralised only one of the wild-type viruses. This was McAb F14, which neutralised FW.

G) Examination of the YF viruses with polyclonalanti-sera

Since only seven of 17 YF17D-204 vaccines were

152

neutralised by the McAbs used, polyclonal anti-sera were used in PRNT’s with one virus from each strain of virus studied. As seen in TABLE 4.10, all the three polyclonal anti-sera raised against Asibi, FNV FC, and 17D-204 USA viruses neutralised these strains in PRNT's.

III. DiscussionIn the early phase of YF 17D vaccine development

and production, a number of different sub-strains of YF vaccine were found to be unsatisfactory when studied in pathogenicity tests and human trials. The instability of YF vaccines were resolved by the establishment of the seed/lot system in 1945. However, the difference of substrain origin of the early vaccines continues to be reflected in the vaccines manufactured today, both the 17D-204 and 17DD sub-strains are at different passage levels (233 to 239 and 286 - 288 respectively). Although these vaccines have proven highly eficacious and safe they have not been fully characterised. In particular, the two new manufacturers of YF vaccines (USSR and Nigeria) refer to their vaccines as '17D'. The results in this chapter indicate that both viruses are 17D-204 sub-strain viruses, although this has only been

153

TABLE 4.10 EXAMINATION OF YF VIRUSES WITHPOLYCLONAL ANTI-SERA IN PRNT's

VIRUS POLYCLONAL ANTI-SERA17D-204 FNV-FC ASIBI

USA

ASIBI >2.1 >2.1 >2.1

YF17D204 USA (4E158)

1.1 1.6 1.4

YF17D204 UK 2.1 >3.4 >3.4

FNV IP 0.5 1.8 1.6

YF 17DD BRAZ 0.8 2.2 1.8

: ANTIBODY TITRES EXPRESSED AS LOG^g REDUCTION INVIRUS TITRE

154

confirmed for the USSR vaccine (Elbert, personalcommunication).

Previously Monath et (1983) geneticallycharacterised the 17DD vaccines by ribonucleaseoligonucleotide fingerprinting and Gould and co-workers (Gould et al., 1985; Buckley & Gould 1985; and Cammack & Gould, 1986) have examined some of the vaccines in PRNTs and HAI assays. This is the first study to compare all the YF vaccines manufactured in every centreantigenicall in HAI assays and neutralisation tests with a large panel of McAbs. Overall, the results demonstrate heterogeneity in the E-protein epitopes that elicit HAI and neutralisation. Since this work involved the use of McAbs it could be argued that the McAbs are examining only some epitopes. Thus, the heterogeneous HAI results with the McAbs were confirmed by the use of polyclonal antibodies (TABLE 4.6)

The biological activities of the McAbs can be grouped into four biological reactive groups; those that have the property of both neutralisation and HAI, those with neutralisation only, those with only HAI activity and those with neither of the biological activities. This is similar to other workers who have noted that the biological properties of HAI and neutralisation in flaviviruses were associated with distinct epitopes (Kimura & Kuroda, 1983; Cammack and Gould, 1986). Also

155

no evidence of linkage between HAI and neutralisation activity was shown. The McAbs used in this study confirm the studies of Casals and Brown (1954) who stated that HAI shows the greatest serological overlap and neutralisation tests the least. Although each McAb used may elicit HAI with many YF vaccine and wild-type viruses, it will only neutralise specific YF vaccine or wild-type viruses with the appropiate 'critical site' or 'critical area' present (Della-Porta and Westaway, 1977 b).

The results of the biological activity of single epitopes on the E-protein of YF virus in HAI assays in this Chapter show that each McAb has a heterogeneous response with each YF vaccine virus strain used in the study, both qualitatively and quantitively. The reactivity of 27 McAbs in HAI assays with all the viruses was examined in a cluster analysis. As shown in Figures 4.1, 4.2 and 4.3, the combined McAb HAIreactivities were able to differentiate three major groups of viruses. The first contained the wild-type viruses and 17DD vaccines, the second contained the majority of the vaccine strains and the third containedonly two vaccine strains (FNV IP and the revertant virusof the YF 17D-204 USA batch 6145, named P-16065).However, when the 17D-204 vaccines were studied in the cluster analysis with only the wild-type Asibi virus,

156

parent to the vaccine strains, the YF17D-204 Colombia vaccine, merges into a separate group with the P-16065. This shows that these two viruses are more distantly related to the remainder of the YF17D-204 vaccines. There is no evidence to suggest that the 17D-204 Colombia vaccine is neurotropic like P-16065. Previously Monath et (1983) have used T ribonucleaseoligonucleotide mapping to genetically compare YFvaccine viruses. They found a high degree of homology

')between the 17D vaccine viruses, with the 17DD derived vaccines being more closely related to each other than the 17D-204 vaccines. Thus, in this study, antigenic analysis of the E-protein of YF vaccines with McAbs has revealed very similar degrees of relatedness as the T mapping study. Analysis of the three batches of 17D-204 USA vaccine and two batches of 17D-204 SA vaccine revealed that these batches of vaccine are closely related, but not identical. This variation presumably reflects differences due to manufacturing the vaccine in embryonated chick eggs on different dates.

The biological activity of the epitopes recognised by the McAbs described previously in Chapters 2 and 3 were also examined in PRNT’s (TABLES 4.7, 4.8, 4.9).Overall, the majority of the McAbs did not neutralise the YF viruses. Of the McAbs that did, they only neutralised specific YF viruses and not all of the 32 YF

157

viruses studied. This heterogeneous response of MoAbs in PRNT’s has been described previously by other workers (Buckley and Gould, 1985; Barrett ^ al., 1989).

Clearly, there is heterogeneity in the biological activity of the E-protein epitopes of vaccine viruses from different manufacturers in HAI and PRNT's. Whether or not this has any practical considerations in the production and efficiency of YF vaccine is not known and requires epidemiological studies in human vaccinees.

Finally, two McAbs neutralised particular viruses to a high titre. McAb B39, which recognises a YF-type specific epitope, neutralised the YF17D-204 WHO vaccine and McAb B26, which recognises a partial flavivirus group reactive epitope, neutralised the YF17D-204 USA vaccine batch 6145. These two McAbs are used in the next Chapter to produce McAb neutralisation escape variants to examine the epitopes of the YF vaccines in more detail.

158

CHAPTER 5

PRODUCTION AND ANALYSIS OF YELLOW FEVERANTIGENIC VARIANTS

159

1. IntroductionIn the previous chapter, the results showed that

envelope protein epitopes of YF vaccine viruses react with McAbs to different extents in PRNT’s and HAI assays. Therefore, the biological function of these epitopes was examined in detail by the generation of antigenic variants.

The selection of antigenic variants depends upon the virus populations being heterogeneous, including virions lacking envelope protein epitopes. Accordingly, the specificity of McAbs for particular epitopes allows the selection of virions in the population lacking the epitope recognised by the McAb. This chapter describes four procedures to produce and analyse antigenic variants of YF virus using McAbs. The best method of producing antigenic variants was by selection of McAb neutralisation escape variants (McAb* variants). However, as shown in the previous chapter few McAbs neutralise YF virus. Thus, three other procedures to select and prepare antigenic variants, using McAbs, are described and compared to that of the neutralisation escape variants. These are immunoprécipitation by the addition of goat anti-mouse immunoglobulin to virus-McAb complexes, capture of virus using McAb attached to ELISA plates and growth of the virus in tissue culture in the

160

presence of a McAb.

II. Results

A) Producing____antigenic____variants____by_____theImmunoprécipitation technique.

The methodology for this procedure is described in detail in Chapter 2. Briefly, virus was incubated with a McAb, followed by incubation with goat anti-mouse immunoglobulin to form virus-antibody complexes. These were then precipitated by high speed centrifugation and the supernatant was examined for residual infectivity and compared to appropiate controls.

Four McAbs, recognising different epitopes on YF virus were used in different combinations with six different viruses in the imunoprecipitation technique. McAb 117 (wild-type specific) which previously showed no neutralising activity in PRNTs (Gould ^ ^ . , 1989)and McAb 6B3B-3 a flavivirus group reactive McAb which previously was shown to neutralise wild-type YF virus (Roerhig et , 1983; Barrett et al., 1990) were usedto produce antigenic variants against the YF17D-204 UKpp with the wild-type epitope (+ve WT). Also McAb 411, a non-neutralising vaccine specific McAb (Gould et al., 1985) was used against four vaccine strains used in the

161

study. McAb 864 which recognises the 17D-204 sub-strain specific epitope and neutralises virus was used as the positive control and PBS was used as the negative control (instead of any McAb). Also, a normal standard virus plaque assay was carried out to get the standard titre of the virus used.

The results, in Table 5.1, show that the control neutralising McAb, 864, reduced the virus titre of only the YF17D-204 vaccine virus strains that it recognised. None of the other McAbs examined reduced the infectivity titre of any of the viruses significantly. Thus, the procedure was not continued further.

B ) Producing Neutralising Escape Variants in TissueCulture.

McAbs 6B3B-3 and 6A3C-4 (Flavivirus group reactive) were used in this experiment. McAbs 6B3B-3 and 6A3C-4 were incubated with the wild-type virus 1899/81 B4.1 for 45 minutes at room temperature and then this mixture was inoculated onto a confluent monolayer of SW13 cells and incubated up to seven days or when 90% CPE occurred. The above procedure was repeated consequtively twice more with the same McAb. Each

162

TABLE 5.1 RESULTS OF THE IMMUNOPRECIPITATION TECHNIQUE

McAbs & or CONTROL

STANDARDPLAQUEASSAY

17DDBRAZIL

4.8*

UKpp+

5.3

17D-204UKpp-

6.3

SApp

4.3

FNVNT

4.3

WILDTYPEB4.14.0

PBS + GAM 4.9 5.5 6.0 4.0 4.5 4.2864 + GAM Nt 1.0 1.9 1.6 Nt NtONLY 864 Nt 1.3 2.0 <0.9 Nt Nt

117 + GAM Nt 5.2 Nt Nt Nt 4.3ONLY 117 Nt 6.5 Nt Nt Nt 4.4

411 + GAM 4.8 5.4 5.8 4.8 4.2 NtONLY 411 4.9 5.4 5.9 Nt Nt Nt

6B3B-3 + GAM Nt 5.3 Nt 4.0 Nt NtONLY 6B3B-3 Nt Nt Nt Nt Nt <0.9

: TITRES ARE EXPRESSED AS THE LOG^q INFECTIVITY INVIRUS TITRE

GAM: GOAT ANTI-MOUSE IMMUNOGLOBULINNt: NOT TESTED

163

passage was titrated to determine the presence of virus. The results of these virus titrations by the standard plaque assay technique described in the methods section showed that virus was not detected. Therefore, this method of producing antigenic variants was not continued further.

C) Production of variants using the ELISA technique

McAbs 411, H5 (vaccine specific) and H6 (YF17D specific) were incubated overnight at 4°C, in a 96 well flat-bottomed tissue culture plate. McAb B39 (YF-type specific) which neutralises YF17D-204 WHO was included as the positive control and McAb P6 (made against Coxsackie A9 virus) as the negative control. The virus (YF17D-204 WHO) was then added to the plate and incubated for 2 hours at room temperature. After this period, the remaining virus supernatant from each well was titrated for infectivity using the standard virus plaque assay technique. As shown in Table 5.2, there was no significant reduction in virus titre with any of the McAbs used including McAb B39 which neutralises the virus. Therefore, this procedure was not continued.

164

TABLE 5.2 RESULTS OF THE ELISA TECHNIQUE TOGENERATE ANTIGENIC VARIANTS TO YF17D-204 WHO

McAb/PBS LOG^q LOG^qINFECTIVITY REDUCTION IN

INFECTIVITY

PBS 4.3411 4.2 0.1H5 4.1 0.2H6 3.9 0.4

B39 4.0 0.3P6 4.3

165

D ) Production of McAb neutralisation resistant(McAb^) variants from PRNTs

i) Selection of McAb^ variants

The detailed procedure for the selection of McAb* variants is described in the Methods section (chapter2). Briefly, a McAb was incubated with a virus at room temperature for 45 minutes or overnight at 4°C (neutralisation titres were the same for both incubation regimes [data not shown]) and then titrated in plaque assays. The McAb* were selected by picking plaques, amplifying in cell culture and then confirming that the progeny virus were resistant to neutralisation by the same McAb. Only McAbs that neutralised virus titres 1000 fold or more were selected as the aim of this procedure was to select and generate stable antigenic variants. Thus, a 1000 fold reduction in infectivity titresindicate that each plaque pick represents on average 1 in a 1000. Since the genome of YF virus is 10,862 nucleotide bases long, a mutation frequency of 1 in a 1000 is equivalent, on average to 10.8 nucleotidechanges randomly distributed across the entire genome of the virus.

McAb* variants were plaque purified from PRNTs, if the SNI was above a value of 3. This was done by

166

incubating McAb 854 with two plaque purified strains of the YF17D-204 UK vaccine, parent A with a wild type epitope (+WT) and parent B without a wild type epitope (-WT). Since the YF-type specific McAb B39 does not neutralise the parent A strain and only neutralises the parent B strain to a SNI of 2, it was used to produce McAb^ variants from the YF17D-204 WHO secondary seed vaccine. Finally, McAb B26 was also used to produce McAb^ variants from the parent B strain of the YF17D-204 UK vaccine and the YF17D-204 USA (6145) vaccine (TABLE 5.3)

All McAb^ variants were grown up in SW13 cells and harvested on the fifth day post-infection or when 90% CPE was observed, whichever was the sooner. The McAb* variants were then titrated. As can be seen in TABLE 5.4, only McAb* variants derived from McAbs 864 and B39gave infectivity titres detectable by plaque assay.These McAb* variants were then re-identified by escaping neutralisation for the second time with the relevant McAb used to select the respective McAb* variants (TABLE 5.5). All McAb* variants showed very little (^ 0.1) orno neutralisation with its respective McAb (e.g. 864 orB39 ) and were selected. Three were derived from the parent strain A and were designated YF17D 864* Al, A2,

167

TABLE 5.3 SNI's of FOUR YF17D-204 VACCINE VIRUS STRAINS IN REACTIONS WITH NEUTRALISING McAbs

McAb SYF

VIRUSES864 B39 B26

17D-204 UK 1.0*STRAIN A 3.9 - 0.9

17D-204 UKpp+STRAIN B 3.6 1.7 2.6

17D-204 UKpp-17D-204 USA 6145 0.8 - 3.317D-204 WHO >4.3 4.4

SNI's ARE EXPRESSED AS THE LOG^q REDUCTION IN VIRUS TITREDENOTES NEUTRALISATION TITRE <0.6

168

TABLE 5.4 TITRE of McAb* VARIANTS PLAQUEPURIFIED FROM PRNTs

LOGio LOG.oMcAb* VARIANTS CPE INFECTIVITY TCID5,YF17D 864* Al + 5.0 NTYF17D 864* A2 + 4.9 NTYF17D 864* A3 + 4.9 NTYF17D 864* B1 + 5.3 NTYF17D 864* B2 + 5.2 NTYF17D B39* D1 + 3.4 NTYF17D B39* D2 + 3.6 NTYF17D204 UKpp- + 5.0 5.2YF17D B26* El + <0.9 4.0YF17D B26* E2 + <0.9 2.0YF17D B26* E3 + <0.9 4.5YF17D B26* E4 + <0.9 3.0YF17D-204 USA + 6.0 4.5YF17D B26* FI + <0.9 2.0YF17D B26* F2 + <0.9 2.5

+: POSITIVE REACTION NT: NOT TESTED

169

TABLE 5.5 LOGIO INFECTIVITIES OF DERIVED McAb* VARIANTS IN PRNT's WITH THE McAbs WHICH SELECTED THEM

LOG^q infectivity REDUCTION INLOGô INFECTIVITY

YF17D MINUS WITHMcAb* VARIANTS McAb McAb

864* Al 5.1 6.3 -

864* A2 5.3 6.3 -

864* A3 5.5 6.4 -

864* B1 5.5 6.4 -

864* B2 5.3 5.2 0.1B39* D1 3.8 3.8 -

B39* D2 4.6 4.6

170

A3, while two were selected from strain B and were designated YF17D 864* B1 and B2 and two, designated YF17D B39* D1 and D2 were selected from the YF17D-204 WHO vaccine. In further studies described below the McAb* variants were also passaged in Vero cells for IIF studies and epitopes to the relevant McAbs from which they were derived from were shown to be absent, thus indicating that the McAb* variants lacked the appropiate epitope and appeared stale on passage. The B26*variants which gave no detectable infectivity titre in plaque assays were designated B26* El, E2, E3 and E4 for those plaque purified from the UKpp -wt epitope or B26* FI and F2 from the YF17D-204 USA vaccine batch 6145. Although the latter variants contained no virusdetectable by plaque assay, virus was nonethelessdetected in TCID^q assays. For comparison, the results obtained with the parent viruses to the B26* variants are also shown in TABLE 5.4.

E) Indirect Immunofluorescence Tests on McAb* variants

i ) The YF17D-204 864* variantsA panel of 24 four McAbs recognising different

epitopes on YF vaccine viruses was used in IIF tests to examine and compare the parent strains and the McAb

171

864* variants derived from them. As can be seen in TABLE 5.6 all McAbs, except McAbs 117 and 864, recognize epitopes on all the virus variants, parent A and B strains and the YF17D-204 UK vaccine. The epitope recognised by McAb 864 was present on the YF17D-204 UK vaccine and the two plaque purified parent A and B strains but not on any of the McAb 864* variants. On the other hand, the YF wild-type specific McAb (117) only recognised the epitope on the parent A strain and the McAb 864* variants made from this parent strain.

ii) The YF17D-204 B39* variants

The same panel of 24 McAbs was used to compare the McAb B39* variants with the parent YF17D-204 WHO vaccine virus. All McAbs, excluding McAbs 117 and B39, recognised epitopes on both the McAb B39* variants and their parent YF17D-204 WHO vaccine strain (TABLE 5.7). However the epitope recognized by McAb 117 was neither present on the two McAb* B39 variants nor their parent YF17D-204 WHO virus and the epitope recognised by McAb B39 was present on YF17D-204 WHO vaccine virus was absent from both the McAb* B39 variants.

iii) The YF17D-204 B26* variants

Although both sets of the McAb* B26 variants gave no titre in virus plaque assays, they showed CPE and

172

TABLE 5.6 COMPARISON OF INDIRECT IMMUNOFLUORESCENCE STUDIES BETWEEN THE YF17D-204 U.K. PARENT VIRUSES AND THE DERIVED YF17D 864* VARIANT STRAINS

YF17D-204 VIRUSES

McAbs864117411A3A4A6B19B26B37B39B44B45G23G65H3H5H6H14H25H33612813825843868

UK

+

+

+++

++++++

++

++++++

+++++

B

+ +++ +

YF17D 864* VARIANTS Al A2 A3 B1 B2

+++++++++

++

++++++

+ + + ■ + +

+++

++++++

++

++++++

+++++

+++++++++++

++

++++++

+++++++++++++++++++

+ ++ ++ ++ ++ +

++++++++++++++++++++++++

+++++++++++++++++

+++++++++++++++++

+ ++ ++ ++ ++ +

+: POSITIVE REACTION IN INDIRECT IMMUNOFLUORESCENCE NEGATIVE REACTION IN INDIRECT IMMUNOFLUORESCENCE

173

TABLE 5.7 COMPARISON OF INDIRECT IMMUNOFLUORESCENCE STUDIES WITH THE YF17D-204 WHO PARENT VIRUS STRAIN AND THE YF17D B39* VARIANTS

YF17D-204 YF17D B39* VARIANTS McAbs WHO D1 D2

B39 + — —864 + + +117 - -411 + + +A3 + + +A4 + + +A6 + + +B19 + + +B26 + + +B37 + + +B44 + + +B45 + + +G23 + + +H3 + + +H5 + + +H6 + + 4-H14 + + +H25 + + +H33 + + +612 + + +813 + + +825 + + +846 + + +868 + + +

+: POSITIVE REACTION IN INDIRECT IMMUNOFLUORESCENCE NEGATIVE REACTION IN INDIRECT IMMUNOFLUORESCENCE

174

gave an infectivity titre in a TCID^q test. Each virus was examined in infected Vero cells by IIF tests using a flavivirus group reactive McAb (H25), a YF-type specific McAb (2E10) and McAb B26 (TABLE 5.8). The results of the IIF tests showed the presence of virus antigen in the cells and the epitopes recognized by the flavivirus group reactive McAb H25 and the YF-type specific McAb 2E10. However, the epitope recognised by McAb B26 was absent in the McAb* variant infected cells but not those infected by the parent viruses.

F ) Comparison of the McAb* variants with their parentvirus strains in HAI assays

i) The YF17D-204 864* variants

A panel of McAbs was used to compare the neutralisation escape variants with their parent strains in their capacity to elicit HAI activity (TABLE 5.9). While a number of McAbs tested (A3, A6, B37, B44, B45,G23, G65, H33) elicited HAI activity to high titre withthe parent strains A and B, they elicited either no detectable or low HAI activity with the McAb 864* variants. McAb 864, which was used to produce the McAb*

175

TABLE

McAbs

5.8 RESULTS OF INDIRECT IMMUNOFLUORESCENCESTUDIES WITH THE YF17D B26 VARIANTS ANDTHEIR PARENT VIRUSES

PARENTYF17D-204VIRUSES

B26* VARIANTS

UKpp- USA6145 El E2 E3 E4 FI F2

H25 + + + + + + + +2E10 + + + + + + + +B26 + +

+: POSITIVE REACTION IN INDIRECT IMMUNOFLUORESCENCENEGATIVE REACTION IN INDIRECT IMMUNOFLUORESCENCE

176

TABLE 5.9 REACTION OF McAbs IN HAEMAGGLUTINATIONINHIBITION ASSAYS (HAIs) WITH PARENT VACCINE STRAINS AND DERIVED 864* VARIANTS

YF17D-204VIRUSES YF17D McAb* VARIANTS

McAbsUK A B Al A2 A3 B1 B2

864 *80 640 80 _ _ _ _ _

117 20 80 20 - - - - -

411 40 40 20 - - - - -

A3 40 80 80 20 20 20 20 40A6 40 80 80 20 20 20 20 40B19 — _ _

B26 - - 20 - - - - -

B37 40 80 80 - 20 - - 20B39 160 80 80 — - - - -

B44 320 640 320 - 20 20 20 20B45 20 160 80 20 20 20 20 20G23 40 320 160 _ 20 20 20G46 40 20 20 20 20 20 20 20G65 160 160 160 - 20 20 20 20H3H33 40 80 80

*: HAI TITRE EXPRESSED AS THE RECIPROCAL OF THE DILUTIONINHIBITING HAEMAGGLUTINATION HAI TITRE LESS THAN 20

177

variants, elicited HAI activity only with the parent virus strains A and B but not the McAb* variants. However, some McAbs (B19, H3) used in the study had nodetectable HAI activity with either the parent virus strains or their McAb* variants. McAb 117 only had high HAI activity with the Parent A strain (which possesses a wild-type epitope recogised by McAb 117) and not with the McAb* variants produced from this parent A strain.


The same panel of McAbs was used to compare the ability of the 17D-204 WHO parent strain with the B39* variants to elicit HAI activity (TABLE 5.10). While a number of McAbs tested (864, 411, B26, B39, G23, G46,G65, and H33) elicited HAI activity with the parent WHOvaccine strain, they elicited either no or low HAI activity with the McAb B39* variants. McAb B39, which was used to produce the McAb* variants elicited HAI activity only with the parent WHO vaccine virus but not the McAb* variants. Some McAbs (117, A3, A6, B19, B37,and H3) used in the study had low HAI activity or no HAI activity with either the parent virus or the B39*variants. McAb 864, however had high HAI activity withthe parent WHO vaccine but low activity with the B39* variants.

178

TABLE 5.10 REACTION OF McAbs IN HAEMAGGLUTINATIONINHIBITION ASSAYS (HAIs) WITH PARENTVACCINE STRAINS AND DERIVED B39*VARIANTS

McAbs864117411A3A6B19B26B37B39B44B45G23G46G65H3H33

YF17D-204WHO

3202040

404020

804040

40

YF17D McAb* VARIANTS D1 D2

20 20

20 20 20 20

20 - 20

20 20 20 20

HAI TITRE EXPRESSED AS RECIPROCAL OF THE DILUTION INHIBITING HAEMAGGLUTINATIONHAI TITRE LESS THAN 20

179

iii) The B26* variantsThe B26* variants were assayed for a

haemagglutination (HA) titre as described in Chapter 2. The titre for each McAb* variant was found to be very low (^8 HAU / 50ul) and so the variants were not studied in HAI assays.

G) Comparison of McAb* variants With Their Parent Virus Strains in PRNT's with McAbs

i) The YF17D-204 864* Variants

A panel of fourteen envelope (E) protein reactive McAbs, of which eight were previously shown to neutralise, were used in PRNT's to study and compare the parent strains with their McAb* variants (TABLE 5.11).

Two of the McAbs neutralised one or more of the McAb 864* variants. The first McAb, B26, neutralised all the variants and both the parent A and B strains but not the YF17D-204 UK vaccine. McAb G23 neutralised both the parent A and B strains, the YF17D-204 U.K. vaccine strain and variants 864* A3, 864* Bl, 864* B2, but notthe variants 864* Al and 864* A2. Of the other McAbs tested, McAbs B44 and B45, neutralised the parent A strain with a neutralisation index of approximately 1

180

TABLE 5.11 REACTION OF McAbs IN PRNTs WITH THE YF17D-204 U.K. VIRUS STRAINS AND THEDERIVED McAb 864 VARIANTS

McAbs

YF17D-204 VIRUSES UK A B

YF17D 864* VARIANTS Al A2 A3 Bl B2

864 1.0 3.9 3.6

117411A3A6B19B26B37B39B44B45G23G65H3H33

0.9 2.61.7

0.8 — 1.3 —

3.2 2.3 1.3

3.4 2.1 1.2 2.1 1.8

1.0 1.5 0.7

: ANTIBODY TITRES EXPRESSED AS LOG ^ REDUCTION IN VIRUS TITREDENOTES NEUTRALISATION TITRE < 0.6

181

and not any of the McAb* variants from this parent A strain. Thus, indicating alteration in the biological function of these epitopes in the variants. McAb B39 neutralised only the parent B strain while McAb 864, which previously was shown to neutralise the parent A and B strains, also neutralised the YF17D-204 UK vaccine strain but did not neutralise any of the variants produced from it. McAbs 117, 411, A3, A6, B19, B37, G65, H3 and H33 neutralised none of the variant or parent viruses.

ii) The YF17D-204 B39* Variants

Of the 15 McAbs used (TABLE 5.12), seven McAbs either neutralised the parent YF17D-204 WHO strain or the McAb* B39 variants derived from it. McAb 864 was the only McAb which neutralised both the parent YF17D-204 WHO strain and albeit lower, the two B39* variants D1 and D2. McAb A6 only neutralised the parent YF17D-204 WHO and McAbs B26 and G23 only neutralised, poorly, the B39* variants D1 and D2. McAb B39, which had been shown previously to neutralise the parent YF17D-204 WHO strain, did not neutralise the variants which were produced using this McAb. McAbs 117, 411, A3, B19, B45, G65, H3, and H33 neutralised neither the parent YF17D-204 WHO nor the B39* variant strains D1 and D2.

182

TABLE 5.12 REACTION OF McAbs IN PRNTs WITH THEYF17D-204 WHO VIRUS STRAIN AND THE DERIVED McAb B39-* VARIANTS

YF17D-204 VIRUS YF17D B39* VARIANTS

WHO D1 D2McAbsB39 4.4117 - -411 - - -864 >4.3 2.0 2.0A3 -. - -A6 2.0 - -B19 -B26 - 0.6 0.7B37 1.5 0.9 -B44 1.0 0.8B45 - - -G23 - 0.7 0.7G65 - - -H3 - - -H33 - - -

*: ANTIBODY TITRES EXPRESSED AS LOGIO REDUCTION IN VIRUS TITRE.DENOTES NEUTRALISATION TITRE < 0.5

183

H ) Comparison of mouse virulence of the Parent VirusStrains With the Variants.

i) The YF17D-204 864* variants

a) Adult mice

Female mice 4-5 weeks old and weighing 20-25g (Strain TO, Olac, U.K.), were inoculatedintracerebrally with 10,000 pfu of virus and were monitored daily for up to 37 days. The majority of mice tested with four (Bl, Al, A2, A3) out of the seven 864* variants remained healthy showing no signs of sickness or death until the end of the study, while mice inoculated with the YF17D-204 UK and the parent A strain died (TABLE 5.13). One variant, B2, killed (9/9) of the mice and therefore was more virulent than the parent strain B from it was derived, which only killed 17.7% (8/45) of the mice. Another variant Bl, derived from the same parent strain B, was more attenuated than the parent strain B and killed no mice (0/26). The variants of strain A were also attenuated; Al killed no mice, while A2 and A3 killed 11% (1/9) and 13% (1/8) of mice respectively.

184

TABLE 5.13 COMPARISON OF THE PATHOGENICITY OF THEYF 17D-204 PARENT STRAINS WITH THE 864* VARIANT STRAINS IN ADULT MICE*

YF17D-204VACCINEVIRUSES

N. DEAD %N°. INOCULATED MORTALITY

AVERAGESURVIVAL

TIME*

UK

A UKpp+

10/10

17/20

100.0

85.0

10.3 + 0.6

10.1 + 0.5

864* Al 864* A2 864* A3

0/81/91/8

<U2.511.012.5

N.A.N.A.N.A.

B UKpp- 8/45 17.7 9.5 + 0.5

864* Bl 864* B2

0/269/9

4 3.8 100.0

N.A.9.2 + 1.1

N.A.

MICE WERE INOCULATED WITH 10,000 pfu BY THE INTRACEREBRAL ROUTEAVERAGE SURVIVAL TIME IN DAYS + THE STANDARD ERROR OF THE MEAN (SEM)NOT APPLICABLE

185


a) Adult mice

Mice inoculated intracerebrally with either the parent WHO strain and the B39* variants died ( as shown in TABLE 5.14).

iii) The YF17D-204 B26* variants

a) Adult mice

Only two (El and E3) of the B26* variants from the 17D-204 UKpp- parent B strain were examined due to their low infectivity titre. 200 TCID^q of parent B strain, B26* El and B26* E3 viruses were each inoculated intracerebrally. As seen in TABLE 5.15, the UKpp- parent strain killed only 33.3% of the mice with an AST of 12.0+1.0 days, while neither of the variants B26* El and E3 killed any of the mice.

b ) Newborn mice

The two other variants (B26* E2 and B26* E4) of the 17D-204 UKpp- parent B strain and the 17D-204 USA parent and its two B26* variants (B26* FI and B26* F2)were inoculated intraperitoneally with 10 TCID^q of virus in a volume of 50ul. As shown in TABLE 5.16, only

186

TABLE 5.14 COMPARISON OF THE PATHOGENICITY OF THEYF17D-204 WHO STRAIN WITH THEB39 VARIANTS IN ADULT MICE


N. DEAD %nV inoculated mo rt a l i t y

AVERAGESURVIVALTIME*

WHOB39* D1 B39* D2

8/89/97/9

100.0100.078.0

11.0 + 0.7 9.0 + 1.010.0 + 0.9

MICE WERE INOCULATED WITH 10,000 pfu BY THE INTRACEREBRAL ROUTE.AVERAGE SURVIVAL TIME IN DAYS + THE STANDARD ERROR OF THE MEAN (SEM).

187

TABLE 5.15 COMPARISON OF THE PATHOGENICITY OF THEYF17D-204 UKpp-- PARENT VIRUS STRAIN BWITH THE B26* ' E' VARIANTS IN ADULTMICE*

YF17D-204VACCINE N°. DEAD %

AVERAGESURVIVAL

VIRUSES N°. INOCULATED MORTALITY TIME*

B UKpp- 2/5 33.3 12.0 + 1.0

B26* El 0/8* ^12.5 NAB26* E3 0/8* 41 2.5 NA

MICE WERE INOCULATED WITH 200 TCID50 UNITS IN AVOLUME OF 20ul BY THE INTRACEREBRAL ROUTE.AVERAGE SURVIVAL TIME IN DAYS + THE STANDARD ERROR OF THE MEAN (SEM).

$. MICE WERE MONITORED FOR 21 DAYS

188

TABLE 5.16 COMPARISON OF THE PATHOGENICITY OF THE YF17D-204 PARENT STRAINS WITH THE B26R* VARIANTS IN NEWBORN MICE*


N. DEAD %N^7 INOCULATED MORTALITY

AVERAGESURVIVALTIME*

B UKpp- B26* E2 B26* E4

4/60/140/13

66.6^7.1^7.7

12.0 + 0.0 NA NA

USA 6145 B26* FI B26* F2

3/161/170/11

25.05.8

^9.1

16.3 + 1.2 NA NA

MICE WERE INOCULATED WITH 10 TCID^q IN A VOLUME OF 5Oui BY THE INTRAPERITONEAL ROUTE.AVERAGE SURVIVAL TIME IN DAYS + THE STANDARD ERROR OF THE MEAN (SEM).

189

the parent strains 17D-204 USA and 17D-204 UK strain B killed 25.0% and 66% respectively of mice with AST’s of16.3 + 1.2 and 12 + 0.0 days. None of the B26* variants killed newborn mice, with the exception of B26* FI which killed one mouse (5.8%) on day twelve.

Ill Discussion

The studies described in this chapter have shown that the most successful method of producing antigenic variants by selection with McAbs was by plaque purifiction from PRNTs. It was not found possible to prepare antigenic variants with McAbs that do not neutralise virus. In the case of the "ELISA" and "Immunoprécipitation" techniques this may reflect incorrect pH buffers. However, the amount of time required for this investigation could not be justified by the Author.

Three McAbs produced variants from four different virus strains. Of the variants produced, three sets gave virus titres when titrated by plaque assay, while variants derived from McAb B26 did not. However, infectivity titres could be determined by use of TCID^q assays. The reason for the lack of plaquing is not understood but may relate to the function of the epitope

190

recognised by McAb B26 on the surface of the virus. It may also be significant that 6 variants derived from two parent vaccine viruses all share this property while the two parent viruses do not. The B26* variants were examined by indirect immunofluoresence using flavivirus group (H25) and YF type (2E10) specific McAbs, in addition to the partial group reactive McAb, B26, from which they had escaped neutralisation. As seen in TABLE 5.8, viral antigen was detected in the cells which possessed the epitopes for both McAbs H25 and 2E10, while the epitope for B26 was not present. This would indicate replication within the cells be, it complete or abortive. The B26* variants were also examined in pathogenicity studies (TABLES 5.15 and 5.16), and except for B26* FI none of the variants killed mice. Therefore, it would appear that the B26* variants are attenuated, however further studies are required to see if the B26* variants replicated in the mouse, be it complete or abortive. This could be achieved by extracting virus from the brains of mice or challenging mice pre-inoculated with the variants with a virulent YF vaccine virus to see if the mice have acquired protective immunity.

McAb* variants derived using McAbs B39 and 864 were examined in IIF, HAI and PRNTs with a panel of

191

McAbs. In IIFs all McAb* variants were found to lack the appropiate epitope. All McAb^ variants showed very little (<0.1) or no neutralisation and little or no HAl activity with their respective McAbs indicating that the variants had also lost the biological activity associated with the epitope recognised by their respective McAb. The McAb* variants did not regain the epitope upon atleast three passages in cell culture, indicating that the variants were probably stable and lacked the epitope recognised by the selecting McAb.

The differences in serum neutralisation indices (SNl) and HAl titres between the parent strains and McAb^ variants can not simply be explained on the basis of the lack of epitopes on the relevant virus strains. IIF tests showed that all McAbs, except McAbs B39, 864and 117, recognize epitopes present on all the McAb* variants, the YF17D-204 UK and WHO vaccines and the parent A and B YF17D-204 UKpp strains (TABLES 5.6 and 5.7). While biological assays (PRNT and HAl) showed the epitopes had lost these functions. Presumably the selection of the antigenic variants results in a conformational change in the envelope protein of the virus. The majority of the McAbs produced in the Author's laboratory elicit HAl activity (as seen in the last chapter). While the majority of the McAb* variants produced lack any HAl activity but have

192

retained the haemagglutinin as determined in HA assays. Thus, it is possible that the loss of HAl activity is due to the McAbs being prepared against one or few antigenic domains on the envelope protein. Clearly epitope mapping is required to answer this question.

Buckley and Gould (1985) suggested that as McAb 864 so effectively neutralised YF17D-204 vaccine virus, neutralisation of YF virus was unlikely to be merely effected by steric hindrance to cell receptors but more likely that changes in the glycoprotein conformation occur which impedes virus penetration into cells. The above PRNTs, IIF tests and HAl assays show that the epitope recognized by McAb 864 has altered, possibly causing a change in conformational structure to the E- protein.

While it is recognized that the E-protein of YF virus is responsible for the biological properties of neutralisation and haemagglutination, it has not previously been shown to be involved in the attenuation of YF virus. Two studies have reported detailed analysis of the E-protein of YF virus with McAbs. Schlesinger et al. (1984) identified nine epitopes while Cammack andGould (1986) identified five epitopes on a single major domain and an additional four other domains which had single epitopes. One of these single epitopes was identified as being 17D-204 sub-strain specific and was

193

recognised by McAb 864. This McAb was used to produce the 864* variant virus strains analysed above.

One of the variants, YF17D 864* B2, was moreneurovirulent than its parent strain B, while four out of the seven 864* variants selected were not neurovirulent for adult mice following intracerebral inoculation, even when an inoculum of 10,000 pfu was used.

It is also of interest that a wild-type epitope recognised by McAb 117 remains on the YF17D 864* variants Al, A2, and A3 that do not possess the McAb 864 epitope and have become attenuated in mice when injected intracerebrally. This contrasts with the findings of Gould ^ (1989) who found that a variant of YF17D-204 vaccine virus which possessed the epitopes recognised by the McAbs 864 and 117 was neurovirulent for mice. There has been a previous report describing YF 17D-204 UK and 17D-204 SA vaccine viruses differing in virulence for mice (Liprandi, 1981). Moreover, Gould and co-workers (Gould ^ , 1985; Cammack and Gould, 1986;Cane and Gould, 1988; Gould et , 1989) have inferred that the E-protein is involved in the determination of mouse neurovirulence.

The results above establish that alteration of neurovirulence of YF17D-204 vaccines for mice is linked to a unique epitope on the E-protein: that of the

194

17D-204 sub-strain specific epitope recognised by McAb 864 and not a YF-type specific epitope recognised by McAb B39.

However, we can not exclude the possibilty that other E-protein epitopes, including YF-type specific epitopes, may alter mouse neurovirulence of YF17D-204 vaccine viruses. To date one other McAb* variant has been shown to have altered neurovirulence. A McAb* variant of Central European Tick-borne encephalitis virus produced by Heinz et (1990) using aneutralising McAb that recognises the type specific epitope B4 was attenuated for adult mice, though only when inoculated subcutaneously and not intracerebrally.

The epitope recognized by McAb 864 is involved in a number of biological activities: neutralisation(Buckley and Gould, 1985), haemagglutination (Gould et al., 1985), haemolysis (Cammack and Gould, 1985) andpassive protection (Gould et ^ . , 1986). Clearly theepitope is also important in determining the neurovirulence of YF virus inoculated intracerebrally in mice. However, how a 17D-204 sub-strain specific epitope is involved in altering the neurovirulence is not clear. This will require a comparitive study of the role of other E-protein epitopes in the pathogenesis of YF virus. The results above support the proposal of Hahn et al. (1987) that the determinants of attenuation of YF

195

virus may reside on the E-protein. These workers identified 12 amino acid diferences between the E- protein of the wild type Asibi virus and the YF17D-204 vaccine virus. However, which of these amino acid(s) forms part of the epitope on the YF virus E-protein recognised by McAb 864, or is responsible for the alteration in virulence is not known. The determination of the nucleotide (and deduced amino acid) sequence of the E-protein genes of the 864* variants described in this chapter is required. Finally, it can not be excluded that there are nucleotide mutations within the envelope protein gene of the 864* variants but outside the epitope recognised by McAb 864 or alternatively mutations outside the envelope protein gene which are causing effects in the E-protein. However, IIF, HAl and PRNT data, together with the extensive neutralising selection pressure of the McAb 864 indicate that the mutation(s) are probably within the E-protein epitope recognised by McAb 864.

196

CHAPTER 6

EXAMINATION OF THE YF VACCINE MANUFACTURED IN THE USSR

197

I. IntroductionRecently, two new centres have been approved by

the World Health Organisation (WHO) to manufacture YF vaccines. These are the Federal Laboratory Service, Lagos, Nigeria and the Institute of Poliomyelitis and Viral Encephalitides, Moscow, USSR. Studies on the USSR vaccine indicated that it may contain viruswith different properties to the other YF virus vaccines studied. Accordingly, this chapter reports experiments carried out to analyse the vaccine manufactured in the USSR.

The Authors’ laboratory received two one doseampoules of the YF vaccine manufactured in the USSR(from Dr. J. Stephenson, Public Health Laboratory Service, Centre for Applied Microbiological Research, Porton), which were part of batch or lot 57. The ampoule label stated it was "17D" vaccine. One of the ampoules was reconstituted in 300ul of distilled water and lOOulof this sample was used to infect SW13 cells for a seedstock. The passage history of this USSR vaccine seed stock is shown in FIGURE 6.1.

The USSR vaccine virus was then compared antigenically in IIF tests, HAl assays and PRNT’s with McAbs and polyclonal anti-sera, and pathogenically with

198

FIGURE 6.1. PASSAGE HISTORY OF USSR VACCINEIN AUTHORS' LABORATORY

USSR/57300ul dHgO

RECONSTITUTEDVACCINE

HAl ASSAYS - ----C636 CELLS"

lOOul virus/SW13 __ --PRNT'sCELLS ""'-PATHOGENICITY STUDY

VERO ___IIF TESTS "CELLS

SW13 --- PATHOGENICITY STUDYCELLS

MOUSEBRAIN

HAl ASSAYS C636CELLS"

SW13 ___ PATHOGENICITY STUDYCELLS

VERO IIF TESTSCELLS

199

other YF vaccine and wild-type strains. In IIF tests, the USSR vaccine virus was studied with the 17D-204 substrain specific McAb 854. McAb, 864 recognised only 10% of the USSR vaccine infected cells.Therefore, plaque purified preparations were made of the USSR vaccine and were compared in antigenic and pathogenic studies to the parent USSR vaccine seed stock as described below.

II. Results

A) Antigenic reactivity of the USSR vaccine compared to other vaccines from each strain

i ) Indirect Immunofluorescence TestsThe USSR vaccine was compared with

representatives of the YF virus vaccine strains (YF17D-204 UK, YF FNV IP, YF 17DD Brazil) and the other newvaccine strain produced in Nigeria for the expression ofepitopes recognised by McAbs in IIF tests. The presenceof these epitopes were examined using McAbs whichrecognise specific epitopes on the E-protein of YFvirus. Of interest were the presence of the flavivirusgroup common, YF type specific, vaccine specific, 17DDand 17D-204 sub-strain specific epitopes.

The IIF tests were performed as described inChapter 2. As seen in TABLE 6.1, the epitope recognised

200

TABLE 6.1 COMPARISON OF THE USSR VACCINE IN IIFTESTS WITH OTHER YF VACCINE STRAINS

YF Vaccines McAbsH3 H5 HIO H25 H33 411 864

17-204 UK + + - + + + +17DD BRAZIL + + + + + + -

FNV IP + + - + + + -

NIGERIA + + - + + + +

USSR + + + + + + St

+: POSITIVE IMMUNOFLUORESCENCE-: NEGATIVE IMMUNOFLUORESCENCE

+ & 10% POSITIVE & 90% NEGATIVE IMMUNOFLUORESCENCEOF VIRUS INFECTED CELLS

201

by the flavivirus group common McAb (McAb H25), was present on all the viruses studied. This was also true for McAbs, H3 and H33 (YF type) and McAbs H5 and 411 (vaccine specific) which recognised epitopes present on all the vaccine virus strains. McAb HIO recognised the 17DD specific epitope only on the 17DD Brazil vaccine strain. However, McAb 864 which recognises the 17D-204 sub-strain specific epitope was present on the YF17D-204 UK vaccine and the new Nigerian vaccine virus infected cells, indicating the latter vaccine was probably a 17D- 204 sub-strain. In comparison, McAb 864 showed only partial immunofluorescence with the USSR vaccine. Using multispot slides with cells infected with the USSR vaccine, McAb H25 (flavivirus group common) causedimmunofluorescence with nearly all the cells indicating that they were infected. Incontrast, with McAb 864 only 10% of the cells showed fluorescence. This indicated that only 10% of the virus infected cells contained antigen expressing the 17D-204 sub-strain epitope. (As shown in the photographs in FIGURE 6.2)

ii) HAl assaysTo study the biological activity of E-protein

epitopes of the USSR vaccine, the USSR virus was examined and compared to seven other YF viruses in HAl

202

FIGURE 6.2 IIF TESTS ON VACCINE VIRUSINFECTED VERO CELLS

a) USSR + McAb H25 b) USSR + McAb 864

c) UK + McAb 864 d) PBS control

203

assays with the A, B and H series of McAbs, as well as polyclonal anti-sera (TABLE 6.2) and then also with the ’F ’ series of McAbs (TABLE 6.3).

The HAl assays were performed as described in Chapter 2.

The results of McAbs, A3, B39, B44, H5, H20, H25, H33 are shown in TABLE 6.2, and have been described previously in Chapter 4. To summarise, all the McAbs were used to compare eight viruses which included two wild-type viruses ( F W and Asibi). Also, included in addition to the USSR vaccine were viruses from the FNV strain, one from the 17DD and two from the 17D-204 substrains, and P-16065 revertant from the YF17D-204 USA vaccine batch 6145. The HAl activity of the USSR vaccine with the McAbs was similar to that of the other vaccine strains reacting to the same extent as the YF17D-204 UK vaccine with both YF-type (McAbs B39, H33) and vaccinespecific (McAb H5) McAbs. Similar results were also obtained with the polyclonal anti-sera made to the FNV FC and 17D-204 USA vaccines. In TABLE 6.3, the results with the 'F' McAbs are summarised. Again, the USSR vaccine E-protein epitopes react similarly with McAbs to the YF17D-204 USA vaccine batch 6145 strain. Thus, in

204

TABLE 6.2 COMPARISON OF THE USSR VACCINE IN HAl ASSAYS WITH OTHER YF VIRUS STRAINS USING THE A, B AND H SERIES of McAbs AND POLYCLONAL SERA

YF VIRUSES McAbsPolyclonal

SeraA3 B39 B44 H5 H20 H25 H33 17D-204

USAFNVFC

FVV 160* 80 160 - 320 - 80 640 640Asibi - 40 40 - 160 - 160 640 640FNV IP 40 20 160 160 160 320 - - 2017DD BRAZIL 80 - 80 80 320 - 160 - 4017D-204 UK 40 160 320 40 160 - 40 160 32017D-204 USA (6145)

80 20 40 - - - - 160 640

P-16065 40 - 40 - 160 320 - 20 4017D USSR 160 320 160 40 320 20 320 160

HAl TITRE EXPRESSED AS RECIPROCAL OF THE DILUTION INHIBITING HAEMAGGLUTINATION.HAl TITRE LESS THAN 20.

205

TABLE 6.3 COMPARISON OF THE USSR VACCINE INHAl ASSAYS WITH OTHER YF VIRUS STRAINSUSING THE ’F SERIES OF McAbs

YF VIRUSES

F WAsibiFNV IP17DD BRAZILI7D-204 UK17D-204 USA (6145)

P-1606517D USSR

McAbsFIO F17 F19 F20 F23 F25 F28 F334020

20

80

20 4040 80

- 160

80

80 40

8080 8020 80

20 — 80

4020

80 80 80 160

20408040 160 20 160

20- 160 80 20 80 320

: HAl TITRE EXPRESSED AS RECIPROCAL OF THE DILUTION INHIBITING HAEMAGGLUTINATION.

■: HAl TITRE LESS THAN 20.

206

HAl assays the USSR vaccine appears similar to the other YF 17D vaccines.

iii) PRNTsThe USSR vaccine was compared with the same group

of YF virus strains (as in the HAl assays) using McAbs previously shown to elicit neutralisation activity. The results of using McAbs A6, B26, B39, F14, G23, H3, H6and H25 are described previously in Chapter 4. A summary of the results are decribed in TABLE 6.4. While one or more of the McAbs neutralise the other virus strains (except for Asibi virus), the USSR vaccine was not neutralised by any of the McAbs. However, this is not a significant result due as other YF virus vaccines were also not neutralised by this group of McAbs (described in Chapter 4).

B ) Comparison of the USSR vaccine to other vaccinestrains in pathogenicity studies

i) In adult miceFemale mice 4 - 5 weeks old and weighing 20-25g

(Stain TO, OLAC, UK) were inoculated with various concentrations of virus in a volume of 0.02 mis by the intracerebral route. For comparison, mice were inoculated with ten fold dilutions of the 17D-204 USA

207

TABLE 6.4 COMPARISON OF THE USSR VACCINE IN

YF VIRUSES

F WAsibiFNV IP17DD BRAZIL17D-204 UK17D-204 USA (6145)

17D-204 USA (4E158)

P-1606517D USSR

PRNT's WITH OTHER YF VIRUS STRAINSUSING McAbs WHICH ELICIT NEUTRALISINGACTIVITY

McAbsA6 B26 B39 F14 G23

*- — — 1.4 —

- - - 1.9 -- - - 1.0 -

- - - 2.3 3.2— 3.3 — — —

0.7

H3 H6

3.0

H25

ANTIBODY TITRES EXPRESSED AS LOG-|o REDUCTION IN VIRUS TITREDENOTES NEUTRALISATION TITRE < 0.6

208

and P-16065 revertant viruses corresponding to 2x10^, 2x10^ and 2x10^ pfu. The mice were monitored daily for up to 21 days. The average survival time in days and the percentage mortality were determined and compared with the data obtained by Gibson ^ (1990). As seen inTABLE 6.5, although all viruses killed 100% of the mice following an inoculation of 2x10* pfu they had different average survival times (ASTs). The ASTs varied from 5.8 + 0.2 days for FNV IP to 10.3 + 0.3 days forAsibi virus. In comparison mice infected with the USSR vaccine had an average survival time of 6.0 + 0.3 days. When an inoculum of 200 pfu was used the YF17D-204 USA vaccine batch had a similar AST (10.5 + 0.7 days) to that following inoculation of 20,000 pfu while the revertant vaccine, P-16065, had an extended AST of 9.8 + 0.6 days. However, when an inoculum of 200 pfu was used for the USSR vaccine the AST of the infected mice could not be calculated due to a biphasic response with mice dying at 3 to 4 days or 11 to 12 days post infection. (FIGURE 6.3)

ii) In newborn miceNewborn mice (1 to 4 days old), strain TO, were

inoculated intraperitonealy with 50ul containing 200 pfu

209

TABLE 6.5 COMPARISON OF THE USSR VACCINE TOOTHER YF VIRUS STRAINS IN PATHOGENICITYSTUDIES USING ADULT MICE*

VIRUS pfu%

MORTALITY AST + S.E.M.*

FW* 2x10* 100.0 9.5 + 0.4Asibi* 2x10* 100.0 10.3 + 0.3FNV IP* 2x10* 100.0 5.8 + 0.217DD BRAZ* 2x10* 100.0 8.3 + 0.317D-204 UK 1x10* 100.0 10.3 + 0.617D-204 USA (6145)

2x10*2x10=2x10=

100.0100.0100.0

10.1 + 0.5 9.1 + 0.7

10.5 + 0.7P-16065 2x10*

2x1032x10=

100.0100.0100.0

7.3 + 0.2 7.1 + 0.5 9.8 + 0.6

17D USSR 2x10%2x10*2x10=

100.0100.071.4

6.2 + 0.2 6.0 + 0.3

?

* : DATA FROM[ GIBSON ET , (1990)20ul OF VIRUS WAS ROUTE INTO GROUPS

INOCULATED BY OF 8 MICE

THE INTRACEREBRAL

^: AVERAGE THE MEAN

SURVIVAL TIME IN DAYS + STANDARD ERROR OF

?: AST UNABLE TO BE CALCULTED

210

FIGURE 6.3 PATHOGENICITY OF THE USSR VACCINE WHENAN INOCULUM OF 200 pfu IS INOCULATED INTRACEREBRALLY

5 - 4 - 3 - 2 ■

1

DEAD

MORTALITY = 71%AST = ?

1 2 3 4 5 6 7 8 9N . OF DAYS

10 11 12 13

211

of virus. The mice were monitored for up to 15 days and the percentage mortality and AST + S.E.M were calculated for each virus. As shown in TABLE 6.6, all the vaccine strains used killed 100% of the mice. However, the USSR and the FNV IP vaccine strains had the shortest AST of7.9 + 0.2 and 8.5 + 0.2 days respectively. The vaccinerevertant, P-16065, and the 17D-204 USA vaccine strains had similar AST of 10.7 + 0.4 and 11.2 + 0.3respectively. These were statistically longer than that of the 17D USSR vaccine (p<0.001). Only the wild-type Asibi and F W virus strains had a shorter AST than the USSR vaccine.

iii) Intranasal Inoculation of mice with the USSRvaccine

Groups of nine female mice 4 - 5 weeks old and weighing 20-25g (Stain TO, OLAC, UK) were inoculated intranasally with different concentrations of virus in a volume of 0.02ml. Each group of mice were monitored for up to 26 days. Only one mouse receiving a dose of 2x10* pfu died on the 19 day post-inoculation (Results not shown).

212

TABLE 6.6 COMPARISON OF THE USSR VACCINE TOOTHER YF VIRUS STRAINS IN PATHOGENICITYSTUDIES USING NEW-BORN MICE

N. DEAD %VIRUS N°. INOCULATED MORTALITY AST + S.E.M.*

AsibiFVV

FNV IP17D-204 UK

17D-204 USA (6145)

P-1606517D USSR

10/108/810/105/11

10/10

10/1011/11

10010010045100

100100

4.1 + 0.43.1 + 0.1 8.5 + 0.214.6 + 1.511.2 + 0.3

10.7 + 0.47.9 + 0.2

: 50ul OF VIRUS CONTAINING 200 pfu WAS INOCULATED BYTHE INTRAPERITONEAL ROUTE

*: AVERAGE SURVIVAL TIME IN DAYS + STANDARD ERROR OF THE MEAN

213

iv) Intracerebral Inoculation of a cynomolgus monkeywith the USSR vaccine

The monkey was inoculated by the intracerebral route with 3.5x10^ pfu of the USSR vaccine in a volume of SOOul, the monkey was monitored upto 18 days post inoculation for signs of any adverse affects. The monkey remained healthy and showed no visible signs of disease.

C) Plaque purification of the USSR Vaccine

The immunofluorescence tests on the USSR vaccine suggested that it consisted of a heterogeneous population of virions, where only 10% of virions expressed the 17D-204 sub-strain epitope recognised by McAb 864. To isolate these variant virion populations, the USSR vaccine was plaqued on a six well tissue culture plate, as described in Chapter 2. After the assay had been overlaid with neutral red, plaques were observed. Ten single isolated plaques were picked with the aid of a glass pasteur pippette (Wiktor & Kowprowski, 1980). The ten plaque purified preparations labelled; I through to X, were grown up in SW13 cells as described in the Methods section. They were all then plaque assayed to assertain the presence of virus (TABLE 6.7). These plaque purified populations of virus were

214

TABLE. 6.7 INFECTIVITY TITRES OF PLAQUE PURIFIEDPREPARATIONS OF THE USSR VACCINE VIRUS

PLAQUE >URIFIE

PREPARATION (pfu/ml)PURIFIED LOG^Q INFECTIVITY

I 3.7II 5.0III 4.0IV 4.0VVI 5.5VII 5.8VIII 5.0IX 6.0X 6.0

NO PLAQUES DETECTABLE AFTER SEVEN DAYS INCUBATION,

215

then compared with the parent USSR vaccine strain in IIF, HAl and pathogenicity studies, as described below.

D ) Comparison of the plaque purified USSR viruses tothat of the parent USSR vaccine

i ) IIF TestsThe nine plaque purified (pp) viruses, which gave

an infactivity titre, were compared with the parent USSR vaccine in IIF tests with 11 McAbs from the A, B. G, and H series. As above the USSR vaccine gave 100% immunofluorescence on infected cells with all the McAbs studied with the exception of McAb 864, which only reacted with 10% of the infected cells. Examination of thee USSRpp virus infected cells showed that although most McAbs recognised epitopes on each USSRpp virus there was a heterogeneous response with McAbs 411 (vaccine specific) and 864 (17D-204 sub-strain specific) (TABLE 6.8). The cells infected with USSRpps I, II, VI and VII, gave no immunofluorescence with McAb 864, while cells infected with USSRpps VIII and X, did not fluoresce with either McAbs 864 or 411. In contrast, both of the epitopes recognised by McAbs 864 and 411 were detected by immunofluorescence of cells infected with USSRpps III, IV, and IX.

216

TABLE 6.8 COMPARISON OF THE PLAQUE PURIFIED USSR VIRUSES TO THAT OF THE PARENT VACCINE STRAIN BY IIP TESTS

USSRMcAbs I II Ill IV VI VII VIII IX X VACC]A6 + + + + + + + + + +B39 + + + + + + + + + +023 + + + + + + + + + +H3 + + + + + + + + + +H5 + + + + + + + + + +H6 + + + + + + + + + +H14 + + + + + + + + + +H25 + + + + + + + + + +H33 + + + + + + + + + +411 + + + + + + - + - +864 — — + + _ _ + + &

+ & —

POSITIVE IMMUNOFLUORESCENCE NEGATIVE IMMUNOFLUORESCENCE ONLY 10% POSITIVE IMMUNOFLUORESCENCE

217

ii) HAI assays

Six McAbs recognising different epitopes were used to examine and compare the USSR parent vaccine with seven of the USSRpps in HAI assays (TABLE 6.9). The vaccine specific McAb, H5, elicited HAI activity with the parent USSR vaccine and all of the USSRpps examined, except for USSRpp VIII. The other vaccine specific McAb, 411, elicited high HAI activity with the parent and all USSRpps with the exception of USSRpps VIII and X. Interestingly McAb G65, a partially flavivirus group reactive McAb elicited high HAI activity with the USSR vaccine, but elicited either low or no HAI activity with all of the USSRpps. McAbs H3, H33 and H6 elicited either low or no HAI activity with both the vaccine and the USSRpps.

iii) Pathogenicity studies

a) In adult mice

Adult female mice were inoculated with 2x10^ pfu in 20ul volume by the intracerebral route. The mice were monitored for 21 days and the percentage mortality and the AST were calculated. The results are shown in TABLE 6.10. Of the seven USSRpps tested in the pathogenicity

218

TABLE 6.9 COMPARISON OF THE PLAQUE PURIFIEDUSSR VIRUSES TO THAT OF THE PARENT VACCINE STRAIN BY HAI ASSAYS

USSRMcAbs II III VI VII VIII IX X VACCINE411 80* 80 160 160 20 160 20 80G65 20 — — — — 80 — 320H3 20 20 40 20 - 20H5 80 20 160 320 - 320 160 40H6 20 - - - 20 40 20H33 — — — — — — — 20

: HAI TITRE EXPRESSED AS RECIPROCAL OF THE DILUTION INHIBITING HAEMAGGLUTINATION.HAI TITRE LESS 20.

219

TABLE 6.10 COMPARISON OF THE PLAQUE PURIFIED USSRVIRUSES TO THAT OF THE USSR VACCINE INPATHOGENICITY STUDIES USING ADULT MICE

VIRUS* N? DEAD%

MORTALITY AST + S.E.M.+N°. INOCULATED

USSRVACCINE 8/8 100 6.0 + 0.4

II 0/8 ^13 N/AIII 3/7 42 15.0 + 1.0VI 3/8 38 13.3 + 0.3

VII 5/7 71 13.4 + 0.9VIII 2/6 33 12.0 + 0.0*

IX 4/8 50 12.5 + 0.9X 2/6 33 11.5 + 1.5*

*: 20ul OF VIRUS CONTAINING 1x10^ pfu WAS INOCULATED BY THE INTRACEREBRAL ROUTEAVERAGE SURVIVAL TIME IN DAYS + STANDARD ERROR OF THE MEAN

$: NOT STATISTICALLY SIGNIFICANT AS FEW MICE DIED

220

study, none killed 100% of the mice as the parent strain did. The USSRpp II killed no mice, while USSRpps X, VIII, VI, III and IX killed from 33% to 50% of mice. The USSRpp VII was the most neurovirulent of the seven plaque purified preparations in that it killed 71% of the mice. The AST of all the plaque purified preparations were similar being between 11.5 and 13.4 days post-infection, with the exception of USSRpp III which had a longer AST of 15 days. However, the AST of the USSRpps was twice that of the parent USSR vaccine strain which killed 100% of mice in 6.0 + 0.4 days.

b) In newborn mice

Newborn mice (1 to 4 days old) were inoculated intraperitoneally with 200 pfu of virus in a volume of 50ul. The mice were monitored for up to 21 days and the AST and percentage mortality were calculated. As shown in TABLE 6.11, the USSR parent vaccine strain was more virulent than any of USSRpps killing 100% of mice with an AST of 7.9 + 0.2 days. While USSRpps II, IX, X killed 60%, 61% and 83% of mice respectively, USSRpps VIII,III, VI killed 0%, 25% and 33% of mice respectively.Although the mortality rate of the USSRpps varied they have a similar AST of 12.2 to 13.5 days with USSRpp III again having the longest AST.

221

TABLE 6.11 COMPARISON OF THE PLAQUE PURIFIED USSR VIRUSES TO THAT OF THE USSR VACCINE IN PATHOGENICITY STUDIES USING NEWBORN

VIRUS

MICE

N. DEAD%

MORTALITY

(6145)

AST + S.E.M.+N*. INOCULATED

USSRVACCINE 11/11 100 7.9 + 0.2

II 3/5 60 12.6 + 0.8III 2/8 25 13.5 + 2.5*VI 5/15 33 12.6 + 0.2

VIII 0/10 ^10 N/AIX 8/13 61 12.2 + 0.6X 10/12 83 12.4 + 0.3

17D-204 USA 10/10 100 11.2 + 0.3

50ul OF VIRUS CONTAINING 200 pfu WAS INOCULATED BY THE INTRAPERITONEAL ROUTE

$.

AVERAGE SURVIVAL TIME IN DAYS + STANDARD ERROR OF THE MEANNOT STATISTICALLY SIGNIFICANT AS FEW MICE DIED

222

E ) Comparison of the original USSR vaccine passedthrough mouse brain and the original vaccine passed in tissue culture

The brain of a 4 - 5 week old paralysed mouse, which had previously been inoculated with the passage one USSR vaccine grown in SW13 cells, was removed on the 5 day post inoculation. The brain was homogenised through a 19G needle and syringe, centrifuged and the supernatant aliquoted into lOOul amounts and frozen at -70%. This preparation was then used to inoculate a confluent cell sheet of SW13 cells. This virus stock (known as Mb) was then compared with the passage one USSR vaccine that had been passaged through SW13 cells as described below.

i ) IIP Tests

Vero cells, previously infected with the USSR vaccine and Mb preparation as described in Chapter 2. were examined using a panel of 9 McAbs. As shown in TABLE 6.12, all the McAbs reacted with both preparations with the exception of McAb 864, which recognised the 17D-204 sub-strain specific epitope in 100% of the Mb infected cells while this epitope was only detected in

223

TABLE 6.12 COMPARISON OF USSR VACCINE PASSAGED INTHE MOUSE BRAIN (Mb) WITH THAT PASSAGED FIRST IN SW13 CELLS BY IIF TESTS

McAb VIRUS STRAINSUSSR USSRin in

SW13 CELLS MOUSE BRAINA6 + +H3 + +H5 + +H6 + +H14 + +H25 + +H33 + +411 + +864 4- & — +

+; POSITIVE IMMUNOFLUORESCENCE -: NEGATIVE IMMUNOFLUORESCENCE

224

10% of the cells infected with the passage one SW13 grown USSR vaccine.

ii) HAI assays

The USSR YF vaccine virus which had been propogated in mouse brain was compared to that propagated in SW13 cells in HAI assays as described previously (Chapter 2) using six McAbs recognising different epitopes on YF virus (TABLE 6.13). Most of the McAbs used (McAbs 411, H3, H6, H33) reacted to a similar HAI activity with both of the viruses, but McAbs G65 and H5 distinguished between the two viruses. McAb G65 elicited high HAI activity with the SW13 propagated virus compared to that grown in mouse brain while McAb H5 elicited a high HAI activity with the mouse brain grown virus compared to the SW13 grown virus.

iii) Pathogenicity study

a) In adult mice

Adult mice were inoculated with 2x10^ pfu of virus as described above. As shown in TABLE 6.14, both the viruses killed 100% of the mice with a similar AST.

225

TABLE 6.13 COMPARISON OF USSR VACCINE PASSAGED IN THE MOUSE BRAIN (Mb) WITH THAT PASSAGED FIRST IN SW13 CELLS BY HAI ASSAYS

McAb VIRUS STRAINS

411G55H3H5H6H33

USSRin

SW13 CELLS*80

3202040

20

USSRin

MOUSE BRAIN 160 40 40

320 40

: HAI TITRE EXPRESSED AS RECIPROCAL OF THE DILUTION INHIBITING HAEMAGGLUTINATION.

■: HAI TITRE LESS THAN 20.

226

TABLE 6.14 COMPARISON OF USSR VACCINE PASSAGED INTHE MOUSE BRAIN WITH THAT PASSAGED FIRST IN SW13 CELLS IN A PATHOGENICITY STUDY USING ADULT MICE

%VIRUS N. DEAD MORTALITY AST + S.E.M.+N°. INOCULATED

USSRin 8/8 100 6.0 + 0.4

SW13 CELLS

USSRin 8/8 100 5.6 + 0.5

MOUSE BRAIN

: 20ul OF VIRUS CONTAINING 2x10^ pfu WAS INOCULATED BY THE INTRACEREBRAL ROUTE

: AVERAGE SURVIVAL TIME IN DAYS + STANDARD ERROR OFTHE MEAN

227

b ) In newborn miceMice were inoculated and monitored as described

previously above. As shown in TABLE 6.15, both the viruses killed 100% of the mice. However, the AST's of the infected mice were different. The mouse brain grown virus had an average survival time of 5.3 + 0.1 days, and the SW13 grown virus had an AST of 7.9 + 0.2 days. The difference in AST's was statistically significant(p<0.001).

Ill. Discussion

Like any other virus popuation the YF vaccine viruses have a potential to undergo changes in its properties. Early workers (Theiler, 1951; Fox et al., 1942; Fox and Penna, 1943) have demonstrated either decreases or increases in neurovirulent properties of the YF vaccines depending on its passage history. More recently, other workers have been able to differentiate YF vaccine viruses when inoculated into mice (Fitzgeorge and Bradish 1980a, b; Liprandi, 1981; Barrett and Gould, 1986), and by the use of McAbs (Schlesinger,1983; Gould et ^ . , 1985) and gentically by T ribonuclease oligonucleotide fingerprinting (Monath et al., 1983). The Authors laboratory obtained the USSR

228

TABLE 6.15 COMPARISON OF USSR VACCINE PASSAGED INTHE MOUSE BRAIN (Mb) WITH THAT PASSAGEDFIRST IN SW13 CELLS IN A PATHOGENICITYSTUDY USING NEWBORN MICE

%VIRUS* N. DEAD_____ MORTALITY AST + S.E.M.+N°. INOCULATED

USSRin 11/11 100 7.9 + 0.2

SW13 CELLS

USSRin 13/13 100 5.3 + 0.1

MOUSE BRAIN

: 50ul OF VIRUS CONTAINING 200 pfu WAS INOCULATED BYTHE INTRAPERITONEAL ROUTE

: AVERAGE SURVIVAL TIME IN DAYS + STANDARD ERROR OFTHE MEAN

229

vaccine and examined it antigenically with McAbs and neurovirulence in a monkey and mice.

The USSR is one of two new centres which have been recently approved by the WHO (WHO, 1988) to manufacture the YF vaccine. The vaccine is a 17D-204 sub-strain virus, using the 17D-204 WHO ALV-free secondary seed as the inoculum. It is produced in minced chicken embryos that have been removed from eggs and grown in culture, with 200,000 doses being prepared annually (Elbert, personal communication).

The biological properties of the USSR vaccine virus were examined in this Chapter using the methods of HAI assays, PRNTs, IIF tests and pathogenicity studies. This vaccine strain was compared to different vaccine strains produced by different manufacturers.

As shown in HAI assays (TABLES 6.2 and 6.3) using both polyclonal anti-sera and McAbs, the USSR vaccine strain was shown to react more similar to the other 17D- 204 vaccine virus strains manufactured in the United Kingdom and the United States of America than the 17DD, FNV-IP vaccine strains or the two wild-type YF viruses from Africa (Asibi and FW). However, when the USSR vaccine virus was analysed in PRNT’s it was not neutralised by any of the McAbs used in the study (TABLE 6.3). This result was not considered significant

230

as many of the other 17D-204 vaccines studied were not neutralised. When, the USSR vaccine was examined in IIF tests, only approximately 10% of the virus infected cells displayed immunofluorescence with McAb 864 which recognises the 17D-204 sub-strain specific epitope. In contrast, McAb H25, which recognises a flavivirus group common epitope, caused 100% of the virus infected cells to fluoresce. This indicated that antigen and possibly many virions in the USSR vaccine population, lack the 17D-204 sub-strain specific epitope. On subsequent plaque purification of the USSR vaccine strain and passaging of these USSRpps, nine different isolates were produced. IN IIF tests, six of these USSRpps lacked the 17D-204 vaccine epitope, while of these six, two lacked the vaccine epitope (i.e. 17D-204, 17DD and FNVspecific) recognised by McAb 411. To date, of 20 vaccines examined from different YF vaccine manufacturers, the USSR vaccine is the only one where McAb 864 has given an unusual pattern of immunofluorescence. Therefore, it appears that the USSR vaccine contains a heterogeneous population of virions. This is not unusual and has been reported previously. For example, Liprandi (1981), isolated different plaque variants from the YF17D-204 UK and YF17D-204 SA vaccine virus strains differing in plaque size and virulence.

Subsequently , the USSR parent vaccine strain was

231

compared in a pathogenicity study to other 17D-204, 17DD and FNV vaccine strains using both adult mice and newborn mice. It was found that the USSR vaccine killed both ages of mice with an average survival time that was much shorter than other 17D vaccine strains (except FNV IP). When the inoculum was reduced to 200 pfu in adult mice, a biphasic response was obtained with mice dying at 3 - 4 days or 11 - 12 days post inoculation. This indicated a mixed population of virions with respect to virulence. However, when the USSRpps were inoculated into adult and newborn mice, they were found to less virulent and took twice as long to die as the parent virus. Thus, biologically the plaque purified viruses appear more like other 17D vaccine strains. The significance of the absence of the 17D-204 sub-strain specific and vaccine epitopes recognised by McAbs 864 and 411 respectively, on the USSRpps is not clear. The author has generated McAb^ variants (Chapter 5) to the 864 epitope and has shown the loss of this epitope alters the virulence of the 17D-204 strains studied. However, in this study the parent USSR vaccine was very neurovirulent for mice and apparently only a small proportion of the virions in the population possess the 17D-204 specific epitope recognised by McAb 864. Thus, it would appear likely that none of the plaque purified viruses represent the virion responsible for virulence

232

with the short AST and that these virulent viruses represent only a small proportion of the population. The parent virus had no adverse affect when it was inoculated intracerebrally into a monkey. This would imply that the "mouse virulent" virus is not virulent in monkeys or that insufficient virions were inoculated into the monkey for it to have an effect. The latter seems unlikely as the monkey received 3.5x10® pfu.

Clearly further investigation is required to isolate the "mouse virulent" virion. However, the statistically shorter AST of the virus passaged in mouse brain may indicate that the virulent virus is being selected and enriched in mouse brain. Accordingly, selection of plaque purified virions from the USSR vaccine propagated in mouse brain may aid in the indentification of the virulent virus in the USSR vaccine. Also, another ampoule or batches of USSR vaccine should be examined, to see if this virulent population is also present in other batches of the USSR vaccine. Alternatively, the virulent virus may not be YF virus. Clearly, a careful study of the neutralisation kinetics of the USSR vaccine with YF type specific McAbs (if available) would help answer this question.

233

CHAPTER 7

EXAMINATION OF THE FNV VACCINES

234

I. Introduction

As discussed in chapter 1, the FNV vaccine was used extensively in the French speaking countries of Africa. Over 80 million people were vaccinated. Although, this vaccine was more stable to heat than 17D vaccine it was discontinued in 1980 due to cases of post-vaccinal encephalitis being reported. Three different strains of the FNV vaccine were examined below. These were FNV IP, FNV NT and FNV FC and are described in Chapter 2. The three vaccines were examined antenically in HAI assays and PRNT’s and biologically in pathogenicity studies in mice and monkeys. The results are discussed below.

II. Results

A) Comparison of reactivities of each FNV vaccine with other virus strains in HAI assays with McAbs

i) The A, B, G and H McAb Series

The FNV vaccines were compared to their wild-type parent strain (FW) using the A, B, G and H McAbs. The reactivities of these McAbs have been described previously in Chapter 4. To summarise, McAbs from these series were used to compare the three FNV vaccines in HAI assays with the wild-type YF F W strain. As shown in

235

TABLE 7.1, most McAbs (B26, B37, B44, Hll, H20, H31)elicited HAI with all four of these YF viruses. However, there were some exceptions. As expected, the vaccine specific McAb H5 did not elicit HAI with the F W wild- type virus, but elicited high HAI activity with FNV IP and FNV NT, but low HAI activity with FNV FC. Out of three YF-type specific McAbs, B39, did not elicit HAI activity with FNV NT, while McAb G23 did not elicit HAI activity with either FNV NT or FW. The last YF-type specific McAb, H33, elicited high HAI activity with all viruses with the exception of FNV IP. Finally, McAbs H8, and A3, did not elicit HAI with FNV IP and FNV FC respectively.

ii) The F McAb Series

The ’F' series of envelope protein reactive McAbs, which were prepared using FNV IP as the immunogen (By A. Jennings), were used to study and compare the F W and FNV vaccines in HAI assays, as well as another African wild-type YF virus, Asibi. The FNV vaccines were also compared to other YF vaccine viruses. As shown in TABLE 7.2, the 'F ' McAbs elicited similar HAI activity with each of the three FNV vaccines. However, the F W wild- type YF virus reacted more similar to the wild-type Asibi virus than the FNV vaccines. McAb F17 did not

236

TABLE 7.1 EXAMINATION OF THE FNV VACCINES IN HAIASSAYS WITH THEIR WILD-TYPE PARENT YF STRAIN F W USING THE A, B, G and H SERIES OF McAbs

McAbs VirusesF W FNV FC FNV IP FNV NT

A3 160* — 40 80B26 80 80 40 80B37 40 80 40 40B39 80 160 20 -

B44 160 320 160 160G23 - 160 80 -

G65 40 640 80 -

H5 - 20 160 80H8 160 40 - 160Hll 80 160 40 80H20 320 640 160 320H30 160 - 160 320H31 80 40 80 40H33 80 80

■

160

*: HAI TITRE EXPRESSED AS RECIPROCAL OF THIINHIBITING HAEMAGGLUTINATION HAI TITRE LESS THAN 20

237

TABLE 7.2 COMPARISON OF FNV VACCINES USING 'F 'McAbs IN HAI ASSAYS WITH OTHER STRAINS OF YF VIRUS

YF Viruses McAbs

FIG F17 F19 F20 F23 F25 F28 F33FNVFNV FC - 160* - 20 160 20 20 40FNV IP - 80 - 20 80FNV NT 20 160 - 20 80 20 - 20

wild-type

F W 40 - - 160 80 40 80 80ASIBI 20 - - 80 80 20 80 160

17D17D-204 UK - 80 40 - 80 4017D-204 USA 80 - - 40 160 20 160 80

(6145)P—16065 20 40 — — 20 — — —17D USSR 40 80 - 160 80 20 80 320

17DD Brazil 20 - - - 20 - - 20

HAI TITRE EXPRESSED AS RECIPROCAL OF THE DILUTION INHIBITING HAEMAGGLUTINATION HAI TITRE LESS THAN 20

238

elicit HAI activity with the F W virus, but elicited high HAI activty with all three FNV vaccines. On the other hand, McAb F28 did not elicit HAI activity with the FNV vaccines it did elicit HAI activity with the wild-type F W virus. When the FNV vaccines were compared to the 17D vaccines the profile of HAI activity elicitedby the McAbs was more similar to that of the 17D-204vaccines than the 17DD Brazil vaccine. The F Mcabs elicited very low or no HAI activity with the 17DDBrazil vaccine viruses.

B ) Clster analysis of the HAI results using the FMcAbs series

All the HAI titres for each McAb reaction against each virus was converted in LOG^^ values and entered into the Clustan computer programme. The dendrogram, shown in FIGURE 7.1, gives two groups, one containing the wild-type YF viruses together with a fewvaccine viruses (17D-204 USA 6145, 17D USSR and 17DDBrazil) and the other containing the three FNVvaccine viruses (FNV FC, FNV IP, FNV NT) and the other17D-204 vaccine viruses (17D-204 UK, P-16065) used in the study. This dendogram was different from thatconstructed using McAbs derived from 17D virus immunogens (using the A, B, G and H McAbs, FIGURE 4.1) except that it still groups the wild-type YF viruses

239

FIGURE 7.1 CLUSTER ANALYSIS OF THE HAIRESULTS USING THE ’F ’ McAbsSERIES

o o o o o O o o Ô05 05 u IV) o05- tn 05 uo> IV) . œ . . o 05 ■ s

ASIBI

17D204 USA 6145

USSR

17DD BRA2

FNV FC

FNV NT

P-16065

17D204 UK

240

together with the 17DD Brazil vaccine virus in the same group.

C ) Comparison of reactivities of the FNV vaccine viruses in PRNT * s with other YF viruses using YF type, flavivirus sub-group and group specific McAbs

Of all the F McAbs only McAb F14 neutralised the YF FNV vaccine viruses and the wild-type F W virus (A. Jennings, personal communication). Therefore, McAbs produced in other Laboratories which used other flaviviuses as immunogens were examined in neutralisation tests. These McAbs and their properties are described in Chapter 2. The YF type specific McAb 2E10 fully neutralised all the YF viruses studied (TABLE 7.3), except for the 17D-204 WHO strain. The flavivirus subgroup McAb 2B5B-3 was able to distinguish between FNV FC and the other two FNV vaccines (FNV NT and FNV IP viruses) and the wild-type F W virus, by neutralising the FNV FC virus to a much higher titre than the otherYF viruses. The flavivirus group reactive McAb 6B6C-1reacted similar by neutralising the FNV FC viruscompletely. McAb 1B7, was able to differentiate the FNVNT virus from the other two FNV vaccines by neutralising FNV IP and FNV FC viruses but not FNV NT virus.

241

TABLE 7.3 COMPARISON OF REACTIVITIES OF THE FNVVACCINES IN PRNT's WITH OTHER YF VIRUSESUSING YF TYPE SPECIFIC, FLAVIVIRUSSUBGROUP AND GROUP SPECIFIC McAbs

McAbs Viruses

17D204WHO FC

FNVIP NT

ASIBI F W B4.1

2E10 3.2* >4.0 >6.0 >4.3 >3.0 >4.0 >4.02B5B-3 NT >4.0 0.6 0.6 >3.0 1.5 >4.06B6C-1 NT >4.0 0.6 - >3.0 1.2 >4.01B7 NT 1.0 1.1 1.0

NT

ANTIBODY TITRES EXPRESSED AS LOG^^ REDUCTION IN VIRUS TITREDENOTES NEUTRALISATION TITRE <0.5 NOT TESTED

242

D) COMPARISON OF VIRUSES IN PATHOGENICITY STUDIES

Previous workers, using intracerebralinoculation of adult mice, have shown that the FNV vaccine virus kills mice with a shorter AST than other vaccine viruses (Barrett and Gould, 1986; Deubel et al., 1986). Gibson ^ (1990) studied the three differentFNV vaccine viruses in adult mice and showed that FNV IP and FNV NT viruses had short AST’s (5-7 days) while FNV FC virus had a relatively long AST (11 days). Therefore, the above results were extended by examining the FNV vaccines in newborn mice and monkeys.

i ) Newborn mice

Each virus studied was diluted to a titre of 200 pfu in a volume of 50ul and inoculated intraperitoneally as described previously. The mice were monitored for 21 days and the AST and percentage mortality were calculated. As seen in TABLE 7.4 all viruses killed 100% of the newborn mice inoculated. Mice infected with the parent YF wild-type strain F W had a shorter AST (3.1 + 0.1 days) than the Asibi wild-type strain (4.1 + 0.4days). Mice infected with FNV FC and FNV NT viruses had very similar ASTs, 12.5 + 0.3, 12.2 + 0.3 respectivelysimilar to the AST’s of the 17D-204 USA (11.2 + 0.3) and

243

TABLE 7.4 COMPARISON OF THE FNV VACCINES WITH THEIR WLID-TYPE PARENT STRAIN (FW) IN A PATHOGENICITY STUDY USING NEWBORN MICE

VIRUSN. DEAD

N°. INOCULATED MORTALITY AST+ S.E.M.*

Asibi 10/10F W 8/8FNV FC 5/5FNV IP 10/10FNV NT 11/1117D-204 USA 10/10

(6145)P-16065 10/10

100100

100100100100

100

4.1 +0.43.1 + 0.1

12.4 + 0.28.5 + 0.2

12.2 + 0.311.2 + 0.3

10.7 + 0.4

200 pfu OF VIRUS IN A VOLUME OF 50ul WAS INOCULATED BY THE INTRAPERITONEAL ROUTE

AVERAGE SURVIVAL TIME.IN DAYS + STANDARD ERROR OF THE MEAN

244

p-16065, the revertant 17D-204 USA vaccine batch 6145 (10.7 + 0.4 days). Mice infected with FNV IP virus had a much shorter AST of 8.5 + 0.2 days which issignificantly shorter (p<0.001) than that of the other two FNV vaccine viruses.

ii) Comparison of FNV vaccines when inoculatedIntracerebrally into a monkey

Two of the FNV vaccinces were examined forneurovirulence in cynmologous monkeys. Each monkey was inoculated with 1x10^ pfu of virus in a volume of 500ul by the intracerebral route and examined daily. Themonkey inoculated with the FNV FC virus died on day nine post infection, while the monkey inoculated withFNV IP virus died much earlier on day six postinfection.

Ill. Discussion

The F W strain of YF virus was attenuated by Mathis et (1928) by passage of the virus throughmouse brain to produce the FNV vaccine virus. Thevaccine was initially given alone by subcutaneous inoculation and later was given together with YF immune serum. However, severe reactions followed injection

245

(Laigret, 1937) and the introduction of the 17D vaccine (attenuated by Theiler et (1937) by passage throughchick embryo tissue) restricted the use of this vaccine. However, workers at the Institute Pasteur in Dakar, Senegal improved the FNV vaccine so it would afford protection if it was given by linear scarification and no YF immune serum was required to be administered (Peltier 1947).

The vaccine was produced from desiccated mouse brain which had previously been inoculated by the intracerebral route with the FNV vaccine virus. The virus had received 250 passages by intracerebral inoculation of adult mice. Unfortunately, this vaccine was also reported to produce post-vaccinal encephalitis in children and the WHO recommended that it should not be given to chidren under the age of 14 years (WHO, 1959).

In this Chapter three different isolates of FNV from three different Laboratories: FNV FC, FNV IP, FNVNT (properties described in Chapter 2, their exact passage histories are unknown) were examined antigenically in PRNT’s and HAI assays, and biologically in pathogenicity tests.

The overall result of the McAb studies was that the FNV vaccine viruses were similar to each other and distinct from other YF vaccine viruses. The summary of

246

the results using the A, B, G and H McAbs in TABLE 7.1 showed that the three FNV viruses react similarly with a positive HAI activity with most McAbs, although the quality of HAI activity was different between the McAbs used. Accordingly, as shown in the dendrogram of the cluster analysis of all vaccines in Chapter 4 (FIGURE4.1), the FNV viruses are differentiated from each other so that each is in a different group. On the other hand the dendrogram constructed using the F McAbs (FIGURE 7.1) does not differentiate the FNV vaccineviruses to a great extent but groups them together away from the wild-type YF virus strains. Deubel ^ al. (1986) using a different method than Monath ^ al. (1983) reported the T oligonucleotide mapping of FW, Asibi, FNV and 17D-204 viruses and that the Asibi and F W wild-type strains had 96.4% homology while comparisons of F W and FNV gave 83.6% homology and Asibi and 17D gave 79.6% homology. The same type of groupingis shown when using the F McAbs which were made againstFNV IP. The homology between the F W and 17D-204vaccines was 81.8% and this reflects why the A, B, G and G McAbs (which where made against 17D vaccines, as described in Chapter 3) do not group the FNV vaccines together in the Cluster analysis (FIGURE 4.2, Chapter 4).

247

Of the F McAbs, only one McAb (F14) neutralised the FNV vaccine viruses to above 1 LOG^g or more. Thus the three FNV vaccines were then examined in PRNT’s using McAbs produced in other Laboratories. As shown in TABLE 7.3, the YF type specific McAb 2E10 did not differentiate any of YF viruses studied. However, McAbs 2B5B-3 and 6B3B-3 differentiated the FNV FC virus from the two other FNV viruses.

Biologically, mouse virulence studiesdifferentiated the FNV vaccine viruses in two ways. First intracerebral inoculation of adult mice showed that FNV IP and FNV NT viruses killed mice with a shorter AST than FNV FC virus infected mice. Second, intraperitoneal inoculation of newborn mice grouped FNV NT and FNV FC viruses together on the basis of AST of infected mice, while FNV IP virus infected mice had a shorter AST. Thus, as expected virulence is multifactorial and there are no apparent correlations to the results obtained with antigenic studies using McAbs.

Two of the three FNV vaccines were further examined for neurovirulence by inoculating each virus into cynmologous monkeys by the intracerebral route. As described previously in the Results the monkey infected with the FNV IP virus had a much shorter AST (killing the monkey on day 6 post infection) than that infected with the FNV FC virus (killing on day 9 post

248

infection) tested. These two FNV vaccine viruses (FNV IP and FNV FC) were apparently more neurovirulent than the revertant P-16065 virus from the 17D-204 USA vaccine batch 6145. The cynmologous monkey inoculated with P- 16065 virus, became paralysed on day 12 and died on day 14 post infection (A.D.T. Barrett, personal communication). Therefore the increased virulence for monkeys of the FNV vaccine viruses requires more studies to investigate the reason for this increased virulence. Comparison of the nucleotide sequence of the genomes of the three FNV viruses to their parent wild-type virus (FW) and to the 17D-204 vaccines may give a molecular answer to the basis of neurovirulence. The study of McAb* variants of the FNV strains may also add more information as to why they are more neurovirulent.

249

CHAPTER 8

STUDY OF THE EFFICIENCY OF PLAQUING (E.O.P) OF YF VIRUSES GROWN IN THE C6-36 MOSQUITO CELL LINE AND PLAGUED AT A RESTRICTIVE TEMPERATURE OF 39.5°C

250

I. IntroductionThere are three different YF virus vaccine

strains: FNV (attenuated from the wild-type F W virus[Mathis et , 1928] ) and the other two are sub-strains derived from the original 17D vaccine (attenuated from the wild-type Asibi virus [Theiler and Smith, 1937]). The 17D-204 virus sub-strain is derived from the 204th passage of Asibi virus in chick embryos with the nerve tissue removed, while the 17DD sub-strain is derived from the 195th passage and subsequent passage in embryonated chicken eggs.

The production of temperature sensitive variants was discussed by Lwoff (1959) who suggested viruses should be passaged in cell lines incubated below 37°C. One biological property of some attenuated vaccine derivatives is temperature sensitivity (ts). For example, the experimental attenuated DEN-2 vaccine candidates are ts (Eckels et ^ . , 1976, 1980).Therefore, the three YF vaccine strains (FNV, 17DD and 17D-204) were examined for ts when grown in mosquito 06- 36 cells at 28°C. Any virus found to be ts when plaqued at 39.5°C was examined following growth of the original virus in the human SW13 cell line at 37°C.

251

II. Results

A) Temperature sensitivity of viruses grown in the C6-36 A. Albopictus mosquito cell line

i) 17D-204 vaccines

The efficiency of plaquing at 39.5°C versus 37.0°C of thirteen YF vaccine viruses and four plaque purified preparations, one each from France and South Africa, and two antigenically different plaque purified preparations from the United Kingdom (described previously in Chapter 5 of this thesis) were studied. With the exception of one of the early (1941) manufactured 17D-204 vaccines (RF318), none of the 17D-204 vaccines grown in C6-35 cells were ts. The RF318 vaccine had an efficiency of plaquing of 0.07 (TABLE 8.1).

ii) 17DD vaccines

The 17DD Brazil, Colombia and Dakar vaccines grown in C6-36 cells were shown to be ts. In addition the plaque purifications selected from 17DD Brazil and Colombia were also shown to be ts (TABLE 8.1). The efficiency of plaquing at 39.5°C versus 37.0°C was between 0.04 and 0.19.

However, when the original viruses were grown in SW13 cells none of the 17DD derived vaccines above were ts at 39.5°C (TABLE 8.2).

252

TABLE 8.1 TEMPERATURE SENSITIVITY OF YF VIRUSES GROWNIN THE C6-36 CELL LINE

LOG^q INFECTIVITYVIRUSES 39.5°C 37.0°C E.O.P."

COL 5.9 6.4 0.32DAK 5.9 5.7 1.58FR 4.1 4.0 1.25FRpp 5.8 5.6 1.58FRG 4.6 4.0 3.90HOL 3.3 3.3 1.00IND 6.3 6.2 1.25RF312 5.9 6.3 0.39RF318 4.0 5.1 0.07

17D204 SA 894A 5.9 6.2 0.50SApp 4.0 4.2 0.63UK 5.6 6.0 0.39UKpp+ 4.8 5.4 0.25UKpp- 5.0 5.0 1.00USA 4E158 5.9 6.0 0.79USA 2848 6.1 6.3 0.63WHO 7.0 7.4 0.39

17D NIG 6.7 6.5 1.58USSR 6.8 6.9 0.79

BRAZ 5.0 5.5 0.19BRAZpp 3.9 5.3 0.04

17DD COL 5.0 6.0 0.10COLpp 4.0 5.3 0.05DAK 5.3 6.4 0.08FC 4.0 4.4 0.39

FNV IP 6.6 7.0 0.39NT 5.5 5.8 0.50ASIBI 4.2 6.7 0.003

Wild-Type ASIBI LPFC 3.2 4.5 0.05F W 5.5 5.7 0.63B4.1 <0.9* 6.8 <1.3x10"

fO VISIBLE PLAQUES SEEN WHEN THE VIRUS WASPLAQUED AT 39.5°C E.O.P. = pfu/ml at 39.5 C

pfu/ml at 37.0°C

253

TABLE 8.2 E.O.P OF VIRUSES GROWN IN SW13 CELLS

VIRUSES17DD

WHICH WERE TEMPERATURE SENSITIVE WHENGROWN IN C636 CELLS

LOG^q pfu/ml of SW13 Grown

Viruses39.5°C 37.0°C

E.O.P of Viruses

SW13 C6-36+Grown Grown

BRAZBRAZppCOLCOLppDAK

5.63.76.5 3.36.6

5.73.8 6.03.46 .5

0.790.793.160.791.25

0.190.040.100.050.08

WILD-TYPEASIBI 3.0 3.3 0.50 0.003: E.O.P. = pfu/ml at 39.5 C

pfu/ml at 37.0°C: RESULTS FROM TABLE 8.1

254

iii) FNV vaccines

The three different FNV vaccines FNV IP, FNV FC, FNV NT were not ts when grown in the C6-36 cell line (TABLE 8.1).

iV) Wild-type viruses

The efficiency of plaquing of four wild-type viruses grown in C6-36 cells were examined. As seen in TABLE 8.1 the Asibi parent virus and the large plaque purified variant from it were ts (efficiency of plaquing of 0.003 and 0.05 respectively). In comparison, the F W was not ts when grown in C6-36 cells. Finally, the South American isolate B4.1 when grown in C6-36 cells was ts with no plaques visible at 39.5°C, even when neat virus was used (i.e. E.O.P of £1.3x10"^).

However when the original Asibi parent virus was grown in SW13 cells it was not ts (TABLE 8.2).

B ) Analysis of the E.O.P. of the 17DD Brazil plaque purified preparation when passaged from SW13 to C6-36 and then back to SW13 cells

The original 17DD Brazil plaque purified preparation was first passaged in SW13 cell line using an MOI of 0.01, then C6-36 cells using an MOI of 0.01, and then back into SW13 cells using an MOI of 0.1. Then

255

each passaged virus was plaqued at 37°C and 39.5°C to test for ts (FIGURE 8.1). The first SW13 passage was not ts, while the C6-36 passage was ts, no visible plaques could be seen on the second SW13 passage when plaqued at 39.5°C even when neat virus was used. However the infectivity titre of this virus was 1x10 pfu/ml when plaqued at 37°C. (i.e. E.O.P <7.9x10'^)

While the infectivity titre of the virus increased with each passage of virus the HA titre was higher inthe 06-36 cell line than in the SW13 cell line (FIGURE8.1) and this is reflected in the Log ^ pfu/HAU values of the 06-36 (pi) and SW13 (p2) passaged 17DD Brazilpp viruses.

C) Antigenic analysis with McAbs of the 17DD Brazilplaque purified virus strain when grown in the 06-36 and SW13 cell lines

i) HAI assays

The plaque purified preparation of the 17DDBrazil vaccine virus grown in SW13 cells was compared to that grown in 06-36 cells. As shown in TABLE 8.3, with the exception of McAb H31 which displayed a 8-folddifference in HAI titre, McAbs elicited similar HAI activity with both viruses grown in the two different

256

FIGURE 8.1 PASSAGE HISTORY OF THE 17DD BRAZIL PLAQUE PURIFIED PREPARATION

PASSAGE Log^Q HA Log^QHISTORY pfu/ml titre pfu/HAU

37°C 39.5°C EOP MO I17DD BRAZILpp (seed stock)

SW13 (pi) 3.8 3.7 0.79 0.01

C6-36 (pi) 5.3 3.9 0.04 0.01 64 2.2

SW13 (p2) 7.0 <0.9* <7.9x10'^ 0.1 32 4.2

: NO VISIBLE PLAQUES SEEN WHEN THE VIRUS WAS PLAQUED AT 39.5°C.

257

TABLE 8.3 ANTIGENIC ANALYSIS OF THE 17DD BRAZILppVACCINE VIRUS WHEN GROWN IN SW13 AND C6-36 CELLS BY HAI ASSAYS

VIRUSB26 B37 B44 G6 G49 G65 H3 H6 Hll H20 H25 H31

17DDBRAZ 40* 80 40 40 - 20 40 80 640 80 80 320(SW13 p2)BRAZ 40 160 160 20 - 40 20 40 160 80 20 40(C636)

: HAI TITRE EXPRESSED AS RECIPROCAL OF THE DILUTIONINHIBITING HAEMAGGLUTINATION HAI TITRE LESS THAN 20

258

cell lines,

ii) PRNT’s

When both of the virus preparations were analysed in PRNT’s with McAbs previously shown to elicitneutralising activity, only one McAb would neutralise both viruses. This was McAb A6 which neutralised the preparation grown in SW13 cells (pi and p2) to an SNI of 0.8 and 0.9 respectively, while the C6-36 preparationwas neutralised to an SNI of 1.0. However, one McAb did differentiate the virus preparations, McAb H6 (17D specific) only neutralised the C6-36 preparation to n SNI of 0.8 and not the SW13 grown preparations. (TABLE8.4)

III. Discussion

Spontaneous temperature sensitive variants have been reported to arise in tissue culture for some flaviviruses. Paul in 1966, produced spontaneous variants of Kyasanur forest virus by repeated monkey epithelia cell passage, Halle and Zebovitz, (1977) producing a ts variant of Japanese Encephalitis virus (JE) in chick fibroblast cell cultures, and Eckels et al. (1980) used a diploid cell line of foetal rhesusmonkey lung cells (DBS-FRhL-2) to produce DEN-2 virus

259

TABLE 8.4 ANTIGENIC ANALYSIS OF THE 17DD BRAZILppVACCINE VIRUS WHEN GROWN IN SW13 ANDC6-36 CELLS BY PRNT’s

VIRUS McAbsA6 B26 B39 F14 G23 H3 H5 H6 H25 H33

17DDBRAZ (SW13 pi) 0.8'BRAZ (C636) 1.0BRAZ (Swl3 p2) 0.9

0.8 -

NT -

: ANTIBODY TITRE EXPRESSED ASLOG^q REDUCTION IN VIRUS TITREDENOTES NEUTRALISATION TITRE £ 0.6

NT: NOT TESTED

260

ts variants. Other workers have used chemical mutagenic agents to produce flavivirus ts variants. Tarr and Lubiniek (1976) used 5-azacytidine to produce ts variants of DEN-2 virus. However, in the work described above certain YF viruses were shown to be ts when grown in the mosquito C6-36 cell line but not when grown in the human adenocarcinoma SW13 cell line. Of the 17D derived vaccines manufactured in different Centres it was significant that only one early 17D-204 vaccine was ts while all the 17DD vaccines plus two plaque purified viruses derived from them were ts when grown in C6-36 cells. Also, two (Asibi and B4.1) out of the three wild- type viruses examined were ts when passaged in the C6-36 cell line, while neither the F W wild-type nor the FNV vaccine viruses derived from it were ts. Why the 17DD vaccines were ts when grown in the mosquito C6-36 cell line but not ts when grown in the vertebrate SW13 cell line is not known. It is known that mosquito cells lack sialic acid and have a different phospholipid content in their cell membranes to vertebrate cells (Stellar, 1976). Thus, this may suggest that ts is due to differences in post-translational modifications of YFvirus proteins in mosquito versus vertebrate cells.Since the E-protein of YF virus is glycosylated it ispossible that ts is due to these differences in glycosylation in the two cell types. Consequently,

261

differences in antigenicity of the envelope protein may be seen (Barrett et , 1990). Tentative evidencefavouring this hypothesis was found with McAb H31 in HAI assays (TABLE 8.3). However the ts of the SW13 (p2)passaged 17DD Brazilpp from C6-36 cells and the loss ofactivity of McAb H6 in PRNT’s (TABLE 8.4) suggests that ts is a stable change and not a conformational change. Thus there is no substantial evidence that ts is due to the E-protein. A similar action where a McAb only neutralised virus grown in C6-36 cells was reported by Barrett ^ (1989) with McAb B14 which onlyneutralised C6-36 grown vaccine viruses from the 17D-204 WHO secondary seed lot and the 17DD manufactured inBrazil. They postulated that this host-dependentneutralisation was due to differences in presentation of epitopes on the cell surface, due to mosquito cells lack sialic acid and the phosholipid content of invertebrate cells differs from that of vertebrate cells.

The area of ts of YF virus in cell culture requires further study, as it is conceivable that the YF vaccine will eventually be manufactured in a cell line where YF virus replicates to a high titre. Therefore, the study of ts properties of the YF virus vaccine in certain cell lines will be important for the thermal stability of the vaccine. Also, differential expression of epitopes in different cell lines requires more study

262

due to that invertebrate cells may not express epitopes involved in eliciting a protective immune response.

While Eckels et (1980) showed the Dengue-2 tsvariants to be attenuated, the 17DD Brazil plaque purified preparation grown in the C6-36 cell line though ts was still as pathogenic for mice as the original virus passaged in SW13 cells (A.D.T. Barrett, personal communication).

263

CHAPTER 9

REACTIVITIES OF SOME McAbs IN PRNT’s WITH VIRUSES GROWN IN VERTEBRATE AND INVERTEBRATE CELL LINES

264

I. IntroductionAs described in the last Chapter, growing vaccine

and wild-type strains of YF virus in different cell substrates (vertebrate or invertebrate) may affect the biological activity of the E-protein epitopes of YF virus. This is of importance as the manufacture of YF vaccines will ultimately be carried out in tissue culture. Before this can take place, a fullinvestigation will be required on the effects (immunological and biological) of growing them in different cell substrates. This chapter describes a preliminary analysis of the YF viruses grown in the vertebrate human adenocarcinoma cell line: SW13, and the invertebrate mosquito A. Albopictus cell line: C6-36using PRNT’s. The McAbs used were of two sources, the first a 17D-204 sub-strain specific McAb derived usingYF17D-204 UK as the immunogen, the second using McAbsthat were flavivirus subgroup and group specific and derived from other flaviviruses as immunogens. Except for McAb 117 which is YF wild-type specific and derived using YF virus strain Asibi as the immunogen. The properties of all the McAbs are described previously in Chapter 2.

265

II. Results

A) The YF 17D-204 sub-strain specific McAb

The 17D-204 sub-strain specific McAb 864 was used to examine in PRNT’s three 17D-204 viruses (Fr, SA, UK) and plaque purified preparations derived from them and three other non-plaque purified 17D-204 viruses grown in SW13 cells. As shown in TABLE 9.1 there were three types of reaction. First, where the virus was not neutralised(P-16065), although IIF tests indicated that this virushad the epitope recognised by McAb 864. The second where McAb 864 neutralised the virus to an SNI of about 1 (e.g. Fr, SA, UK and USA) and the last where the McAb neutralised the viruses to a high titre (an SNI of approximately 4 e.g. 17D-204 WHO). The viruses which were neutralised to a high SNI were grown in C6-36 cells and the resultant virus was examined in a PRNT with McAb 864. As shown in TABLE 9.1 the viruses which wereneutralised when grown in SW13 cells were alsoneutralised when grown in C636 cells to approximately the same level except for the 17D-204 UKpp- which had a significantly lower SNI (1.5 log^^ units or 30 fold difference).

266

TABLE 9.1 PRNT’s OF THE McAb 864 WITH YF VIRUSESGROWN IN SW13 AND C6-36 CELLS

Viruses Cell Substrate Difference in neutralisation

17D-204 SW13 C6-36 titre*

Fr 0.8* NT -

Frpp 0.9 - >0.5SA 1.3 NT -

SApp >4.4 >4.8 <-0.4WHO >4.0 4.0 >0.0UK 1.0 NT -

UKpp+ 3.9 >4.0 >-0.1UKpp- 3.6 2.1 1.5USA 0.8 NT -

P-16065 NT: ANTIBODY TITRES EXPRESSED AS LOG^q REDUCTION IN

VIRUS TITREDENOTES NEUTRALISATION TITRE _< 0.4

NT: NOT TESTEDDIFFERENCE EXPRESSED AS THE SNI (SW13) - SNI (C6-36)

267

B ) Flavivirus subgroup and group specific McAbs

Six different YF viruses (three wild-type and three different vaccine strains) were examined with six McAbs in PRNT’s. All McAbs neutralised one or more of the viruses when they were grown in SW13 cells, with the exception of McAb 117. McAbs 4G2 and 6A3C-4 partially neutralised the 17DD Brazil vaccine virus, while the 17D-204 USA vaccine virus was not neutralised by any of the McAbs used (TABLE 9.2). The FNV IP vaccine virus was neutralised only by McAbs 1B7 and 4G2. However, the wild-type viruses were neutralised by a number of McAbs; 1B7, 4G2, 6A3C-4 and 6B3B-3 neutralised both Asibi andB4.1. While 1B7 did not neutralise the F W virus, McAb 5B1D-2 neutralised the Asibi virus.

Each virus was grown in 06-36 cells and analysed with the same panel of McAbs. As seen in TABLE 9.3 only the FVV and B4.1 wild-type viruses were neutralised. F W was partially neutralised by McAbs 1B7 and 6B1D-2, while McAbs 6A3C-4 and 6B3B-3 had an increased SNI of 2 compared to that when the virus was grown in SW13 cells. However, B4.1 had reduced SNI’s using virus grown in 0636 versus that grown in SW13 cells for McAbs 6A30-4 and 6B3B-3.

268

TABLE 9.2 REACTIVITY OF FLAVIVIRUS SUBGROUP AND GROUP McAbs IN PRNT’S. WITH VIRUS GROWN IN SW13 CELLS

McAbsViruses 117 1B7 4G2 6A3C-4 6B3B-3 6B1D

17DD BRAZ - - 0.5* 0.5 - -

17D-204 USA (6145)

- - - - - -

FNV IP - 1.1 1.0 - - -

ASIBI - 1.0 4.0 4.0 >4.0 1.0FVV - - 1.3 0.7 1.1 -

B4.1 - 0.5 1.7 >3.0 >3.0 -

ANTIBODY TITRES EXPRESSED AS LOG^q REDUCTION IN VIRUS TITREDENOTES NEUTRALISATION TITRE <0.4

269

TABLE 9.3 REACTIVITY OF FLAVIVIRUS SUBGROUP AND GROUP McAbs IN PRNT’s WITH VIRUS GROWN IN C6-36 CELLS

Viruses17DD BRAZ

17D-204 USA (6145)

FNV IPASIBI

F WB4.1

McAbs117 1B7 4G2 6A3C-4 6B3B-3 6B1D-2

0.7 1.5 2.2 2.02.2 2.4 1.7

0.6

ANTIBODY TITRES EXPRESSED AS LOG^^ REDUCTION IN VIRUS TITREDENOTES NEUTRALISATION TITRE <0.4

270

Ill. Discussion

As shown in the last chapter, biological differences (ts) were shown between different strains of YF virus when grown in C6-36 cells, but no differences were seen when the various YF viruses were grown in SW13 cells.

In this Chapter different McAbs gave different results when examined in a biological assay (PRNT) with viruses grown in SW13 and C6-36 cells. One McAb (McAb 864) showed no difference in neutralising ability, while other McAbs were able to differentiate vaccine from wild-type strains and also differences between vaccine strains and between wild-type strains.

McAb 864, a 17D-204 sub-strain specific McAb, has been used previously by Buckley and Could (1985) to examine neutralisation of YF viruses grown as 20% mouse brain suspensions. In this chapter, each virus was grown from a reconstituted freeze dried preparation in SW13 cells. As seen in TABLE 9.1 the SNI titres differ from what Buckley and Gould obtained, especially the UK, South Africa and French 17D-204 vaccine viruses. These differences can probably be explained by the way in which the virus was grown, as well as, this study using different batches of vaccine to that used by Buckley and

271

Gould, (1985). An interesting difference was that between the 17D-204 USA vaccine virus and the revertant virus P-16065, where McAb 864 neutralised the former but not the latter. Also, while the plaque purified preparations of the 17D-204 UK and 17D-204 SA were neutralised to high SNI, the parent vaccine viruses were neutralised to a low SNI. The plaque purified and the 17D-204 WHO secondary seed batch vaccine viruses were grown in C6-36 cells and their SNI titres compared to that of virus grown in SW13 cells. As shown in TABLE 8.1 there were no large differences in the level of neutralisation. This clearly is due to McAb 864 being a potent neutralising antibody that recognises the 17D-204 sub-strain epitope on viruses grown in both cell lines. This point was substantiated by indirect immunofluorescence tests (data not shown).

However, Barrett et al. (1990) examined YF virus strains using flavivirus subgroup and group specific McAbs prepared with immunogens other than YF virus in IIF tests. It was shown that the McAbs could distinguish between wild-type from 17D-204 viruses, when the viruses were grown in C6-36 cells but not when they were grown in SW13 cells. The other two vaccine strains FNV and 17DD were not differentiated by IIF tests to the wild-type viruses. A part of this study was to examine and compare the viruses grown in 06-36 and SW13 cells

272

biologically in PRNT’s and as shown in TABLES 9.2 and9.3 the viruses could be differentiated. While five out of the six McAbs used in the study neutralised wild-type Asibi virus when grown in SW13 cells, none of the McAbs examined would neutralise the virus grown in C6-36 cells. Also, none of the McAbs neutralised the 17D-204 USA vaccine virus grown in either cell line. Although McAbs 4G2 and 6A3C-4 partially neutralised the 17DD Brazil vaccine virus when grown in SW13 cells, none of the McAbs would neutralise the virus when grown in 06-36 cells. Of interest is McAb 1B7 which only neutralised the FNV IP vaccine and wild-type Asibi and B4.1 viruses when grown in SW13 cells, but partially neutralised the wild-type strain FVV grown in 06-36 cells. Kjellen and Zeipel (1984) working on Echovirus type 4 (strain Pesascek) discussed the influence of host cells on the neutralisation ability of McAbs. They reported a minor fraction of the Pesascek virus strain that resisted neutralisation when plagued on either the rhabdomyosarcoma strain RD or the green monkey kidney strain GMK-AHl cell lines. Although, another fraction of infectious virus was shown to be neutralised when plagued on the RD cell line but resisted neutralistion when plagued on the GMK-AHl cell line. However,it was Grady and Kinch (1985) who first showed different neutralisation activities of a bunyavirus grown in

273

invertebrate (C6-36) and vertebrate (BHK-21) cell lines. They showed that one McAb neutralised only the virus grown in vertebrate cells, while two other McAbs neutralised virus grown in both cell lines and a fourth McAb neutralised the virus grown only in invertebrate cells. It is of interest that three of these McAbs had been shown to be located on the same domain (Grady et al., 1983). As in this study McAbs with a flavivirusgroup specificity prepared against St. Louis Encephalitis virus recognising the same antigenic site previously described by Roehrig et (1983) (E-4b)detected similar differences. Thus, the McAb neutralisation differences seen by Grady and Kinch (1985) and in this study reported by Barrett et al. (1990) may represent a common phenomenon with arthropod borne viruses. Barrett et al. (1989, 1990) have reported a number of differences, both antigenically and biologically, when YF virus is passaged in the C6-36 cell line when compared to YF virus being passaged in vertebrate cell lines. These differences may be explained by the differences of each cell line in processing the glycosylated E-protein. Stollar (1976) has shown the absence of sialic acid and or different phospholipid content in invertebrate cells compared to vertebrate cells. As seen in the last chapter and reported by Barrett et (1989, 1990) virus grown in

274

C6-36 cells is biologically different to that grown in SW13 (vertebrate) cells. This will have important implications in the manufacture of YF virus vaccine or the production of other vaccines to important human flavivirus pathogens (e.g dengue) in mosquito cell lines.

275

CHAPTER 10

GENERAL DISCUSSION

276

This thesis has described the analysis of YF vaccine viruses produced in different countries using different conditions of manufacture and different seed viruses with McAbs directed against epitopes of the E- protein. Also, where appropriate, comparison has been made between vaccines and wild-type strains of YF virus.

The E-protein has been shown to be respnsible for a number of biological properties including those of eliciting neutralisation and haemagglutination. Therefore, a panel of McAbs were generated against the 17DD Brazil and 17D-204 WHO vaccine viruses and used to examine the the biological function of E-protein epitopes of YF virus in PRNT’s and HAI’s. The majority of the McAbs were found to elicit HAI, but not neutralisation activity. This suggests that the McAbs are recognising the same functional domain of YF virus. Other workers have produced many more McAbs against flaviviruses that elicit neutralising activity than those generated in the Author’s laboratory against YF virus (Schlesinger et ^ . , 1983; Gould et al., 1985).This observation and the fact that nearly every McAb generated in the Author’s laboratory recognises the E- protein (A.D.T. Barrett, personal communication) may be due to the immunogen used, which was tissue culture fluid containing virus rather than mouse brain

277

suspension, polyethylene glycol precipitates or purified virus to generate the McAbs (Kimura-Kuroda and Yasui, 1983; Roehrig et , 1983; Gould et , 1985).

The panels of McAbs were used to analyse the E-protein epitopes of 34 vaccine and wild-type YF viruses in HAI assays and PRNT's. The results from these biological assays and there combined analysisgenerates groupings of McAbs which define distinct antigenic determinants or epitopes revealing a serological complexity which goes far beyond that established with polyclonal anti-sera.

Using a computer programme to analyse the HAI data of 27 McAbs with the 34 YF viruses, it was found that the viruses were divided in a dendrogram into three groups. One of the groups contained the majority of the 17D-204 vaccine viruses manufactured in differentcentres around the world. Thus, although the 17D-204viruses were not identical they were very similar. The17DD vaccines were found to separate into a different second group. The last group contained two vaccines which were shown to be neurotropic in monkeys and mice. These were the discontinued FNV IP and the P-16065 revertant viruses. Monath et (1983) also obtained asimilar grouping when using T RNA oligionucleotide fingerprint mapping he showed a high degree of homology

278

between the 17D vaccines. However, the 17DD derived vaccines were differentiated from the 17D-204 vaccines by the loss of one oligonucleotide (No.37). They suggested that this was due to the additional 40 passages in chick embryos for the development of the 17DD vaccines.

Presumably the differences in biological function of the epitopes of the E-protein shown by McAbs used in this study are also due to different manufacturing conditions in different centres (see Chapter 1). Whether these differences have practical significance for the efficacy of YF vaccines from different manufacturers or reflects an academic question is not clear. However, the results obtained suggest that, at the very least, there should be some clinical studies undertaken with vaccinees to investigate and compare the efficacy of the vaccine produced in different centres.

Unfortunately, only a few of the McAbs recognised epitopes that elicit neutralising activity and the majority of these did only neutralised the viruses poorly. These results greatly hindered the generation of McAb neutralisation resistant (McAb^) variants to study the epitopes in detail. However, McAb* variants were generated using McAb 864 (17D-204 substrain specific), McAb B39 (YF type specific) and McAb

279

B26* (partial flavivirus group reactive). The B25* variants were unusual in that they had lost the ability to produce plaques though infectivity was demonstrated using the TCID^q assay. The B39* variants were shown to remain neurovirulent for mice, while the 864* variants had an altered neurovirulence for mice compared to their parent strains. The mouse virulence studies with the B26* variants yielded equivocal results on their alteration in mouse neurovirulence. Both the B39* and 864* variants had lost the ability to elicit HAI activity with the McAbs tested.

The new 17D vaccine manufactured in the USSR was examined antigenically and biologically. The results obtained with this vaccine were somewhat worrying. While the 17D USSR vaccine reacted similar to other YF 17D vaccines in PRNT's and HAI assays, it was antigenically distinct being differentiated from other 17D vaccines in IIF studies and biologically different as it contained virions that were more virulent in mice than "normal" 17D virions. In view of the virulence of the 17D USSR virus for mice it was surprising that the monkey which received the large inoculum of 3.5x10^ pfu of 17D USSR virus showed no clinical signs of infection following intracerebral challenge. However, it may be that the vaccine is only virulent in mice or that the virulent

280

virus will only cause problems in young vaccinees, as has been seen previously with the FNV and P-16065 viruses. Clearly, further investigations are required on the 17D vaccine manufactured in the USSR.

Studies with McAbs (series F) generated againstthe FNV IP virus revealed similar biological properties to the McAbs raised against the 17D vaccine viruses.However, the results of the F series of McAbs in HAIassays revealed a different dendrogram to that derived from the results with the 17D generated McAbs. Presumably this reflects differences in antigenicity of the 17D and FNV vaccine viruses. Examination of neurovirulence of the FNV vaccine viruses in monkeys and mice differentiated them from each other to some extent and separated them from all the 17D vaccine viruses. However, using both animal models, FNV IP was the most neurovirulent of the FNV vaccines. Comparitive sequence data for the E-protein gene of FNV and 17D viruses will be required to answer the question of differences in antigenicity and possibly for the increased neurovirulence of the FNV vaccine viruses.

The properties of the YF viruses grown ininvertebrate and vertebrate cell lines was compared. Virus grown in the mosquito A. albopictus C6- 36 cell line was antigenically distinct to that grown in

281

human SW13 cells in that epitopes eliciting HAI and neutralisation differed. Furthermore, viruses of the 17DD substrain were temperature sensitive (ts) at 39.5°C when grown in C6-36 cells. Presumably, this reflects the different passage histories of 17DD and 17D-204 substrains of the YF vaccine 17D. It was also found that this ts phenotype was maintained for 17DD Brazil vaccine virus that was grown in C6-36 cells and then passaged into SW13 cells.

Thus, it would seem ts was due to a genetic change in the virus induced by growing the virus at 28°C in the C6-36 cell line. Comparison of the nucleotide sequence of the virus passaged in both cell lines would be required to answer this question. Consequently, there was no evidence to associate the ts phenotype with the E-protein of YF virus. However, the use of a potential ts YF virus as a vaccine requires further investigation, particularly of the efficacy of 17DD versus 17D-204vaccine viruses when used in the climates of Africa andSouth America. Some wild-type strains of YF virus were also found to be ts. Unlike the 17DD vaccine viruses that had never been grown in C6-36 cells, the passage history of wild-type strains of YF virus had involvedpassage in C6-35 cells thus the ts phenotype of theseviruses can not be easily interpreted.

282

This thesis has given an insight into the properties of YF virus vaccine viruses compared with those of wild-type YF viruses. Comparative studies on the efficacious nature of the YF vaccines manufactured in different centres is unlikely to take place as manufacturers will be reluctant to have their vaccine investigated in detail in case it is found unsuitable for vaccine purposes. Nonetheless there are still areas that can be investigated to continue research on the YF vaccine viruses. The first is comparative epitope mapping of the E-protein of vaccine and wild-type strains of YF virus with the McAbs used in this thesis and investigation of the relative location of epitopes and functional domains on different viruses. The second area is sequence data to support the biological results obtained in this thesis. Especially, the determination and comparison of the nucleotide (and deduced amino acid) sequence of the E-protein gene of the F W and FNV viruses and a number of the YF virus vaccines manufactured in different centres. Similarly, sequence data is required to examine the amino acid substitutions in the McAb* variants to identify the amino acids that constitute particular epitopes of the E-protein. Consequently, the changes in biological function of the E-protein epitopes could be physically mapped on to the

283

three dimensional structure of the E-proteinof YF or another flavivirus when it is determined. The third area is an investigation of induction of cytotoxic and helper T cell activity by different YF vaccine viruses, particularly the E-protein epitopes that elicit T cell activity and there conservation on different YF vaccine viruses, including those manufactured in different centres. Finally, as mosquito control is unlikely to succeed in eliminating YF infection in humans and YF vaccination will always be required, a more cost effective cell culture vaccine will need to be developed. Therefore, further analysis of the biological properties of the YF virus vaccines grown in different cell lines is needed. This new cell cultured produced vaccine would need to be biological stable, cheap and sufficiently attenuated to be gien to children just after birth like other vaccines. Such a YF vaccine would prevent infection and spread in humans and the risk of urban YF which has increased due to the re-introduction of Aedes Aegypti into Central America.

284

REFERENCESAmster, L.J. (1987). Carlos J. Finlay: The Mosquito Man.

Hospital Practice (May 15) 223 - 246.Anonymous. (1966). Fatal viral encephalitis following

17D yellow fever vaccine inoculation. The Journal of the American Medical Association 198, 671 - 672.

Barme, M. and Bronnert, C. (1984). Thermostabilisation du vaccin anati-amaril 17D lyophilise.- 1. Essai de substances protectrices. Journal of Biological Standardization 12, 435 - 442.

Barrett, A.D.T. & Gould, E.A. (1986 a). Comparison of neurovirulence of different strains of yellow fever virus in mice. Journal of General Virology 67, 631 - 637

Barrett, A.D.T. & Gould, E.A. (1986 b). Antibody-mediated early death in vivo after infection with yellow fever virus. Journal of General Virology

2539 - 2542Barrett, A.D.T. (1987). Yellow Fever Vaccines. Bulletin

Institute Pasteur 85, 103 - 124.Barrett, A.D.T., Pryde,A., Medlen, A.R., Ledger, T.N.,

Whitby, J.E., Gibson, C.A., Desilva, M., Groves, D.J., Langley, D.J. and Minor, P.D. (1989). Examination of the envelope glycoprotein of yellow fever vaccine viruses with monoclonal antibodies. Vaccine 7, 333 - 336.

Barrett, A.D.T., Mathews, J.H., Miller, B.R., Medlen, A.R., Ledger, T.N. and Roehrig, J.T. (1990). Identification of monoclonal antibodies that distinguish between 17D-204 and other strains of yellow fever virus. Journal of General Virology 71, 13 - 18.

Bauer, J.H. (1928). Transmission of yellow fever by mosquitoes other than Aedes aegypti. American Journal of Tropical Medicine 8, 261 - 282.

Bendiner, E. (1988). Max Theiler: Yellow Jack andthe Jackpot. Hospital Practice 23, 211 - 244.

285

Berge, T. (1975). (Ed.), and American Committee onArthropod-Borne viruses. International Catalogue of Arboviruses Including certain Other viruses of Vertebrates, USPHS, Washington.

Besselaar, T.G. and Blackburn, N.K. (1988). Antigenic analysis of West Nile virus strains using monoclonal antibodies. Archives of Virology 99, 75 - 88.

Biedrzycka, A., Cauchi, M.R., Bartholomeusz, A., Gorman, J.J. and Wright P.J. (1987). Characterization of protease cleavage sites involved in the formation of the Envelope glycoprotein and three Non- structural proteins of dengue virus type 2, New Guinea C strain. Journal of General Virology 68, 1317 - 1326.

Blake, J.B. (1968). Yellow fever in eighteenth century America. Bulletin of New York Academy of Science

673 - 686.Boquan, C.,Guofang, Z., Qinzhi, L., Xiuzhen, P.,

Guocui, L. and Ye, T. (1983). Production of monoclonal antibodies against Japanese B Encephalitis by mouse hybridoma technique. Acta Academiae Medicine sinicae 5, 210 - 215.

Boshell-Manrique. (1959). Outline history of yellow fever epidemic in middle America— 1948 - 1957.Proceedings 6th International Congress of Tropical Medicine and Malaria 5, 62 - 87.

Buckley, A. and Gould, E.A. (1985). Neutralization of yellow fever virus studied using monoclonal and polyclonal antibodies. Journal of General Virology 66, 2523 - 2531.

Bugher, J.C., Boshell-Manrique, J., Roca-Garcia, M., and Osorno-Mesa, E. (1944). Epidemiology of jungle yellow fever in eastern Colombia. American Journal of Hygiene 39, 16 - 51.

Bugher, J.C. (1951). The pathology of yellow fever.Yellow Fever. 137 - 163. Eds. Strode, G.K.Publishers, McGraw-hilIBook company. New York.

Burnet, F.M. (1957). A modification of Jernes Theory of antibody production using the concept of clonal selection. Australian Journal of Science 20, 6 7 -69.

286

Calisher, G.H., Karabatsos, N., Dalrymple, J.M., Shope, R.E., Porterfield, J.S., Westaway, E.G. and Brandt, W.E. (1989). Antigenic relationships between flaviviruses as determined by crossneutralisation tests with polyclonal antisera. Journal of General Virology 70, 37 - 43.

Cammack, N. and Gould, E.A. (1985). Conditions for haemolysis by flaviviruses and characterisation of the haemolysin. Journal of General Virology 66, 2291 - 2296.

Cammack, N. and Gould, E.A. (1986). Topographical analysis of epitope relationships on the envelope glycoprotein of yellow fever 17D vaccine and wild- type Asibi parent virus. Virology 150,333 - 341.

Cane, P.A. and Gould, E.A. (1988). Immunoblotting reveals differences in the accumulation of envelope protein by wild-type and vaccine strains of yellow fever virus. Journal of General Virology 70, 557 - 564.

Carey, D.E., Kemp, G.E., Troup, J.M., White, H.A.(1972). Epidemiological aspects of the 1969 yellow fever epidemic in Nigeria. Bulletin of the World Health Organization 46, 645 - 651.

Casals, J. St Brown, L.V. (1953). Hemagglutination with certain arthropod-borne viruses. Proceedings of the Soceity of Experimental Biology 83, 170 - 173.

Casals, J. (1957). The arthropod-borne group of animal viruses. Transactions N.Y. Academy of Science19, 219 - 235.

Casals, J. (1958). Viral encephalitis. In Viral Encephalitis: A Symposium. 5 - 2 1 . Eds. Fields,W.S. and Blattner, R.J. Charles C. Thomas, Springfield, IL.

Castle, E., Leidner, U., Nowak, T., Wengler, G. andWengler, G. (1986). Primary structure of the West Nile flavivirus genome region coding for all nonstructural proteins. Virology 149, 10 - 26.

Cecilia,D., Gadkari, D.A., Kedarnath, N., Ghosh, S.N.(1988). Epitope mapping of Japanese Encephalitis virus envelope protein using monoclonal antibodies against an Indian strain. Journal of General Virology 69, 2741 - 2747

287

Chamberlain, R.W. (1980). The Togaviruses. Biology, structure and replication. 175 - 227. Ed.Schlesinger, R.W. Publisher: Academic Press

Chambers T.J. and Rice, C.M. (1987).Molecular biology of the flaviviruses. Microbiological Sciences 4 , 219 - 223.

Chambers, T.J., McCourt, D.W.and Rice, C.M. (1989). Yellow fever virus proteins NS2A, NS2B, and NS4B: indentification and partial N-terminal Amino Acid sequence analysis. Virology 1 6 9 , 100 - 109.

Chambers, T.J., McCourt, D.W. and Rice C.M. (1990). Production of yellow fever virus proteins in infected cells: indentification of discretepolyprotein species and analysis of cleavage kinetics using region-specific polyclonal antisera. Virology 1 7 7 . 159 - 174.

Chanock, R.M. & Sabin, A.B. (1953). The hemagglutinin of St. Louis Encephalitis. 1. Recovery of stable hemagglutinin from the brains of infected mice. Journal of immunology 70, 271 - 285.

Clarke, D.H. and Casals, J. (1958). Techniques for hemagglutination and hemagglutination-inhibition with arthropod-borne viruses. American Journal of Tropical Medicine and Hygiene 7, 561 - 573.

Clarke, D.H. (1960). Antigenic analysis of certain Group B arthropod-borne viruses by antibody absorption. Journal of Experimental Medicine 1 1 1 , 1 - 20.

Clarke, D.H. (1964). Further studies on antigenic relationships among the viruses of the Group B tick-borne Complex. Bulletin W.H.O. 31, 45 - 56

Coia, G., Parker, M.D., Speight, G., Byrne, M.E. and Westaway, E.G. (1988). Nucleotide and complete amino acid sequences of Kunjin virus: Definitivegene order and characteristics of the virus- specified proteins. Journal of General Virology 69, 1 - 2 1 .

Cornet, M., Robin, Y., Hme, G., Adam, C., Renaudet, J., Valade, M. and Eyraud, M. (1979). Une pousse pizootique de fivre jaune selvatique au Sngal oriental. Isolement du virus de lots de moustiques adultes males et femelles. Medecine and Maladies Infectieuses 9, 63 - 66.

288

Davis, N.C., and Shannon, R.C. (1929). Studies on yellow fever in South America; transmission experiments with certain species of Culex and Aedes. Journal of Experimental Medicine 50, 803 - 808.

Davis, N.C. (1930 a). Susceptibility of capuchin (Cebus) monkeys to yellow fever virus. American Journal of Hygiene 11, 321 - 334.

Davis, N.C. (1930 b). Transmission of yellow fever;experiments with woolly monkey, spider monkey, and squirrel monkey. Journal of Experimental Medicine 51, 703 - 720.

Davis, N.C., and Shannon, R.C. (1931). Studies on yellow fever in South America: attempts to transmit virus with certain aedine and sabethine mosquitoes and with Triatomas (Hemiptera). American Journal of Tropical Medicine 11, 21 - 29.

DeCock, K.M., Monath, T.P., Nasidi, A., Tukei, P.M., Enriquez, J., Lichfield, P., Craven, R.B., Fabiyi, A., Okafor, B.C., Ravaonjanahary, C. (1988). Epidemic yellow fever in eastern Nigeria, 1986. The Lancet (March 19th) 630 - 633.

DeCock, K.M., and Monath, T.P. (1988). The Lancet (July 2nd) 4 0 - 4 0 .

Della-Porta, A.J. and Westaway, E.G. (1977 a). Immune response in rabbits to virion and non-virion antigens of the Flavivirus Kunjin. Infection and Immunity 15, 874 -882.

Della-Porta, A.J. (1977 b). Review Article. A multi-hit model for the neutralization of animal viruses. Journal of General Virology 38, 1 - 19.

De madrid, A,T, and Portefield, J.S. (1974). TheFlaviviruses (group B arboviruses):A crossneutralization study. Journal of General Virology

91 - 96.Despres,P., Cahour, A., Dupy, A., Duebel, V., Bouloy,

M., Digoutte, J.P. and Girard, M. (1987). High genetic stability of the region coding for thestructural proteins of yellow fever virus strain 17D. Journal of General Virology 68, 2245 - 2247.

289

Deubel, V., Camicas, J.L., Pandare, D., Robert, V., Digoutte, J.-P. and Germain, M. (1981). Développement de souches sauvages et vaccinales du virus de la fievre jaune dans les cellules de Aedes aegypti et transmission au souriceau. Annals Virology (Institute Pasteur) 132 E, 41 - 50.

Deubel, V., Ekue, E.K., Diop, M.M., Diop, A. and Digoutte, J.-P. (1986). Introduction A L ’analysede la virulence du virus de la fievre jaune (Flaviviridae): etudes genetique, immunochimiqueset biologiques comparative entre les souches attenuees vaccinales et leurs souches parentales. Annals Institut Pasteur 137E, 181- 192.

Deubel,V., Kinney, R.M. and Trent, D.W. (1988). Nucleotide sequence and deduced amino acid sequence of the nonstructural proteins of Dengue type 2, Jamaica Genotype: comparitive analysis of the full length genome. Virology 165, 234 - 244.

Dimmock, N.J. (1984). Review Article. Mechanisms of neutralisation of animal viruses. Journal of General Virology 65, 1015 - 1022.

Dittmar, D., Haines, H.G., and Castro, A. (1980). Monoclonal antibodies specific for dengue virus type 3. Journal of Clinical Microbiology 12, 74 - 78.

Downs, W.B. (1989). Arboviruses. In Viral Infections of Humans: Epidemiology and Control. Chapter 5.105 - 132. Eds. Evans, A.S. Publisher: PlenumMedical Book Company. New York.

Duffy, J. (1966). Sword of Pestilence: The New Orleans Yellow Fever Epidemic of 1853. Louisiana State University Press.

Duffy, J. (1968). Yellow fever in the continental United States during the Nineteenth Century. Bulletin of New York Academy of Science 44, 687 - 701.

Dupuy, A., Despres, P., Cahour, A., Girard, M. and Bouloy, M. (1989). Nucleotide sequence comparison of the genome of two 17D-204 yellow fever vaccines. Nucleic Acids Research 17, 3989

Durieux, C. (1956). Mass yellow fever vaccination in French Africa south of the Sahara. Yellow Fever Vaccination. WHO monograph series No. 30. Geneva

290

Eagle, H. (1959). Amino acid metabolism in mammalian cell cultures. Science 130, 432 - 437.

Earle, W.R. (1943). Production of malignancy vitro IV the mouse fibroblast cultures and changes seen in the living cells. Journal of the National cancer Institute 4, 165 - 212.

Eckels, K.H., Brandt, W.E., Harrison, V.R., McCowan, J.M., and Russell, P.K. (1976). Isolation of a temperature-sensitive Dengue-2 virus under conditions suitable for vaccine development. Infection and Immunity 14, 1221 - 1227.

Eckels, K.H., Harrison, V.R., Summers, P.L., Russell, P.K. (1980). Dengue-2 vaccine: preparation from a small-plaque virus clone. Infection and Immunity

175 - 180.Fenner, F. (1975/76). The classification and

nomenclature of viruses. Summary of results of meetings of the International Committee on Taxonomy of Viruses in Madrid, September 1975. Intervirology 6, 1 - 12.

Findlay, G.M. and Clarke, L.P. (1934). The susceptibility of the Hedgehog to yellow fever. Transactions of the Royal Soceity of Tropical medicine and Hygiene 28, 335 - 345.

Fitch, W.A. and Margoliash, E. (1967). Construction of Phylogenetic Trees: A method based on mutationdistances as estimated from cytochrome C sequences is of general applicability. Science 155, 279 - 284.

Fitzgeorge, R., and Bradish, C.J. (1980a). The in vivo differentiation of strains of yellow fever virus in mice. Journal of General Virology 46, 1 - 13.

Fitzgeorge, R., and Bradish, C.J. (1980b). Differentiation of strains of yellow fever in irradiated mice. Journal of General Virology 50, 345 - 356

Fox, J.P., Lennette, E.H., Manso, C. and Souza Aguiar, J.R. (1942). Encephalitis in man following vaccination with 17D yellow fever virus. American Journal of Hygiene 36, 117 - 141.

291

Fox, J.P. and Penna, H.A. (1943). Behaviour of 17Dyellow fever virus in rhesus monkeys; relation to substrain, dose and neural or extraneuralinoculation, American Journal of Hygiene 38, 152 - 172.

Fox, J.P. & Laemmert, H.W. (1947). The cultivation ofyellow fver virus. II. Observations on the infection of developing chick embryos. American Journal of Hygiene 46, 21 - 40

Franco, R., Martinez-Santamaria, J., and Toro-Villa, G. (1911). Fiebre amarilla o fiebre espiroquetal. Endemias of epidemias en Muzo de 1907 a 1910. Academia Nacional de Medicina, Sesiones Cientificas Centenario, Bogota 1, 169 - 228

Freestone, D.S. (1988). Yellow fever vaccine. Chapter 18, 387 - 419. Vaccines. Eds. Plotkin, S.A., andMortimer, E.A. Publisher; W.B. Saunders Company, Philadelphia.

Furuta, I., Takashima, I., Arikawa, J. and Hashimoto, N.(1984). Antigenic analysis of flaviviruses with monoclonal antibodies against Negishi virus. Microbiology and Immunology 28, 1023 - 1030.

Gaidamovich, S.Y. & Sokhey, J. (1973). Studies on antigenic peculiarities of West Nile virus strains isolated in the USSR by three serological tests. Acta Virology 17, 343 - 350.

Gentry, M.K., Henchal, E.A., Me. Cown, J.M., Brandt, W.E. and Dalrymple, J.M. (1982). Identification of distinct determinants on Dengue-2 virus using monoclonal antibodies. American Journal of Tropical Medicine and Hygiene 31, 548 - 555

Georges, A.-J., Tible, F., Meunier, D.-M.-Y., Gonzalez, J.P., Beraud, A.M., Sissoko-Dybdahl, N.-R., Abdul- Wahid, S., Fritzell, B., Girard, M. and Georges- Lourbot, M.-C. (1985). Thermostability andefficacy in the field of a new, stabilized yellow fever virus vaccine. Vaccine 3, 313 - 315.

Germain, M., Saluzzo, J.F., Cornet, J.P., Herv, J.P., Sureau, P., Camacas, J.L., Robin, Y., Salaun, J.J., and Heme, G. (1979). Isolement du virus de la fivre jaune partir de la ponte et de larves d ’une tique Amblyomma variegatum. C.R. Académie Science (Paris) (Ser. D) 289 635 - 637.

292

Geske, T., Nichtila, P., Seethaler, H., Koch, M.A., L ’age-Stehr, J. (1983). Establishment of hybridomas producing antibodies to viral surface epitopes related to pathogenic properties of yellow fever virus strains. Immunobiology 165, 263.

Gibson, C.A., Wills, M.R., Gould, E.A., Sanders, P.G. and Barrett, A.D.T. (1990). Effect of administration of sodium aurothiomalate on the virulence of yellow fever viruses in adult mice. Vaccine (In press).

Gollins, S.W. and Portefield, J.S. (1985). Flavivirus infection enhancement in macrophages : an electron microscope study of viral cellular entry. Journal of General Virology 66, 1969 - 1982.

Gollins, S.W. & Porterfield, J.S. (1986). A new mechanism for the neutralization of enveloped viruses by antiviral antibody. Nature 321,244 - 246.

Gould, E.A., Chanas, A.C., Buckley, A., Clegg, C.S.(1983). Monoclonal immunoglobulin N antibody to Japanese Encephalitis virus that can react with a nuclear antigen in mammalian cells. Infection and Immunity 41, 774 - 779.

Gould, E.A., Buckley, A., Cammack, N., Barrett, A.D.T., Clegg, J.C.S., Ishak, R. and Varma, M.G.R. (1985). Examination of the Immunological relationships between Flaviviruses using yellow fever virus monoclonal antibodies. Journal of General Virology 66, 1369 - 1382.

Gould,E.A., Buckley, A., Barrett, A.D.T., Cammack, N.(1986). Neutralizing (54K) and Non-neutralizing (54K and 48K) monoclonal antibodies against structural and non-structural yellow fever virus proteins confer immunity in mice. Journal of General Virology 67, 591 - 595.

Gould, E.A., Buckley, A., Groeger, B.K., Cane, P.A. and Doenhoff, M. (1987). Immune enhancement of yellow fever virus neurovirulence for mice; studies of mechanisms involved. Journal of General Virology 68, 3105 - 3112.

Gould, E.A. and Buckley, A. (1989). Antibody- dependent enhancement of yellow fever and Japanese encephalitis virus neurovirulence. Journal of General Virology 70, 1605 - 1608.

293

Gould, E.A., Cammack. specific identify vaccine1889 - 1894.

Buckley, A., Cane, P.A., Higgs, S. and (1989). Use of monoclonal antibody

for a wild-type yellow fever virus to a wild-type antigenic variant in 17D

pools. Journal of General Virology 70,

Grady, L.J., Sanders, M.L., and Campbell, W.P. (1983). Evidence for three separate antigenic sites on the G1 protein of La Crosse virus. Virology 126, 395 - 397.

Grady, L.J. and Kinch, W. (1985). Two monoclonal antibodies against La Crosse virus show host- dependent neutralising activity. Journal of General Virology 66, 2773 - 2776.

Grange, T., Bouloy, M., and Girard, M. (1985). Stable secondary structures at the 3'-end of the genome of yellow fever virus (17D vaccine strain). Federation of European Biochemical Societies Letters 188, 159 - 163.

Guirakhoo. F., Heinz, F.X. and Kunz, C. (1989). Epitope model of Tick-borne encephalitis virus envelope glycoprotein E: Analysis of structuralproperties,role of carbohydrate side chain, and conformational changes occurring at acidic pH. Virology 169, 90 - 99.

Haddow, A.J., Smithburn, K.C., Dick, G., Kitchen, S.F., and Lumsden, W. H. R. (1948). Implication of the mosquito Aedes (Stegomyia) africanus Theobald in forest cycle of yellow fever in Uganda. Ann. Tropical Medicine 42, 218 - 223.

Hahn, C.S., Darymple, J.M., Strauss, J.H. and Rice C.M.(1987). Comparison of the virulent Asibi strain of yellow fever virus with the 17D vaccine strain derived from it. Proceedings of the National Academy of Sciences 84, 2019 - 2023.

Hahn, Y.S., Galler, R., Hunkapiller, T., Dalrymple, J.M., Strauss, J.H., Strauss, E.G. (1988). Nucleotide sequence of Dengue-2 RNA and comparison of the encoded proteins with those of flaviviruses. Virology 162, 167 - 180.

Hall, R.A., Burgess, G.W., Kay, B.H. (1988). Type- specific monoclonal antibodies produced to proteins of Murray Valley Encephalitis virus. Immunology and Cell Biology 66, 51 - 56.

294

Hallauer, C. (1947). Proceedings of the 4th International Congress of Microbiology, Copenhagen. July 20th-26th, 257.

Hallauer, C. (1959). Zuchtung von Gelbfiebervirus in menschlichen explantaten. ARCHIV FUR DIE GESAMTE. VIRUSFORSCHUNG 9, 428 - 441

Halle, S. and Zebovitz, E. (1977). A spontaneous temperature sensitive mutant of Japanese encephalitis virus: Preliminary Characterization.Archives of Virology 54, 165 - 176.

Hammam, H.M., Clarke, D.H. and Price, W.H. (1965).Antigenic variation of West Nile virus in relation to geography. American Journal of Epidemiology 82, 40 - 55

Hammam, H.M. & Price,W.H. (1966). Further oberservations on geographic variation in the antigenic character of West Nile and Japanese B viruses. American Journal of Epidemiology 83, 113 - 122.

Hashimoto, H., Nomoto, K., Watanabe, K., Mori, T.,Takezawa, T., Aizawa, C., Takegami, T. and Hiramatsu, K. (1988). Molecular cloning and complete nucleotide sequence of the genome of Japanese Encephalitis virus Beijing-1 strain.Virus Genes 1, 305 - 317.

Hanks, J.H. and Wallace, R.E. (1949). Relation of oxygen and temperature in the preservation of tissues by refrigeration. Proceedings of the Soceity of Experimental Medicine 71, 196 - 200.

Hardy, F.M. (1963). The growth of the Asibi strain of yellow fever virus in tissue cultures.-II. Modification of virus and cells. Journal of Infectious Diseases 113, 9 - 14.

Hase, T., Summers, P.L., Eckels, K.H. and Putnak, J.R.(1989). Morphogenesis of Flaviviruses. Subcellular Biochemistry 15, 275 - 305. Eds. Harris, J.R.Publisher: Plenum Press, New York.

Hawkes, R.A., Roehrig, J.T., Hunt, A.R. and Moore, G.A.(1988). Antigenic structure of MVE virus E glycoprotein. Journal of General Virology 69, 1105 - 1109.

Hearn, H.J. Soper, W.T. and Miller, W.S. (1968). Loss in virulence of yellow fever virus serially passed in Hela cells. Proceedings of the Soceity of Experimental Biology and medicine 119, 319 - 322.

295

Heinz, F.X., Tuma, W. and Kunz, C. (1981 a). Antigenic and immuno-genic properties of defined pjysical forms of tickborne encephalitis virus structural proteins. Infection and Immunity 33, 250 - 257.

Heinz, F.X. & Kunz, C.H. (1981 b). Homogeneity of the structural glycoprotein from European isolates of Tick-borne Encephalitis virus: comparison withother flaviviruses. Journal of General Virology 57, 263 - 274.

Heinz, F.X., Berger, R., Tuma, W., Kunz, C. (1983). A topological and functional model of epotopes on the structural glycoprotein of Tick-borne Encephalitis virus defined by monoclonal antibodies. Virology 126, 525 - 537.

Heinz, F.X. (1986). Epitope mapping of flavivirus glycoproteins. Advances in Virus Research 31, 103 - 168

Heinz, F.X., Holzmann, H., Mandl, C., Guirakhoo, F., Tuma, W. and Kunz, C. (1990). Molecular basis of antigenicity and attenuation of a Flavivirus: Tick-borne encephalitis virus. Vaccines 90 97 -100. Eds. Brown, F., Chanock, R.M., Ginsberg, H.S., Lerner, R.A. Publisher Cold Spring Harbour Laboratory Press.

Henchal, E.A., Gentry, M.K., McGowan, J.M. & Brandt, W.E. (1982). Dengue virus specific and flavivirus group determinants identified with Monoclonal antibodies by indirect immunofluorescence. American Journal of Tropical Medicine and Hygiene

830 - 836.Henchal, E.A., Brandt, W.E., Cowan, J.M., Gentry, M.K.,

Repik, P.B. and Padamanabhan, R.K. (1983). Proceedings of the International Conference on Dengue / Dengue hemorrhagic Fever, Kualar Lumpur.

Henchal, E.A., McGowan, J.M., Burke, D.S., Seguin, M.C. and Brandt, W.E. (1985). Epitopic analysis of antigenic determinants on the surface of Dengue-2 virions using monoclonal antibodies. American Journal of Tropical Medicine and Hygiene 34, 162 -169.

Hirst, G.K. (1941). The agglutination of red blood cells by allantoic fluid of chick embryos infected with influenza virus. Science 94, 22 - 23.

296

Irie, K.,Mohan, P.M., Sasaguri, Y., Putnak, R. andPadmanabhan, R. (1989). Sequence analysis ofcloned dengue virus type 2 genome (New guinea C strain). Gene 75 197 - 211.

lorio, R.M. (1988). Mini-review. Mechanisms ofneutralization of animal viruses: monoclonalantibodies provide a new perspective. Microbial Pathogenesis 5 , 1 - 7

Ishak, D. Tovey, D.G., and Howard, C.R. (1988). Morphogenesis of yellow fever virus 17D in infected cell cultures. Journal of General Virology 69, 325 - 335.

Karabatsos, N. (1985). (Ed.). International catalogue of Arboviruses 1985: including certain other virusesof vertebrates. American Society of Tropical Medicine and Hygiene, Subcommittee on information Exchange, San Antonio.

Kedarnath, N., Cecilia, D ., Sathe, P.S., Gadkari, D.A.A., Dandawate, C.N., Goverdhan, M.K. and Gosh, S.N. (1986). Monoclonal antibodies against Japanese encephalitis. Indian Journal of Medical Research 84, 125 - 133.

Kerr, J.A. (1951). The clinical aspects and diagnosis of yellow fever. Yellow fever 385 - 425. Eds.Strode, G. K. Publisher: McGraw-hill Book Company New york.

Kimura-Kuroda, J. & Yasui, K. (1983). Topological analysis of antigenic determinants on the envelope glycoprotein V3 (E) of Japanese Encephalitis virus using monoclonal antibodies. Journal of Virology 45, 124 - 132.

Kimura-Kuroda, J, & Yasui, K. (1986). Antigenic comparison of envelope protein E between Japanese Encephalitis virus and some other flaviviruses using monoclonal antibodies. Journal of General Virology 67, 2663 - 2672.

Kimura, T. & Ohyama, A. (1988). Association between pH- dependent conformational change of West Nile flavivirus envelope protein and virus mediated membrane fusion. Journal of General Virology 69, 1247 - 1254.

Kirk, R. (1941). An epidemic of yellow fever in the Nuba Mountains, Anglo-Egyptian Sudan. Annals of tropical Medicine and Parasitology 35, 67 - 112.

297

Kitano, T., Susuki, K. and Yamaguchi, T. (1974). Morphological chemical and biologicalcharacterization of Japanese Encephalitis virus virion and its haemagglutinin. Journal of Virology 14, 631 - 639.

Kiyotake, M., Takashima, I. and Hashimoto, N. (1983). Pathogenicity of Negishi virus in micecharacterized by age of susceptibility, routes of inoculation and growth of the virus in tissues. Japanese Journal of Veterinary Research 31, 7 - 13.

Kjellen, L. and von Zeipel, G. (1984). Influence of host cells on the infectivity and neutralizability of Echovirus type 4 (Pesascek). Intervirology 22, 32 - 40.

Kobayashi, Y., Hasegawa, H. Oyama, T., Tamai, T., and Kusaba, T. (1984). Antigenic analysis of Japanese Encephalitis virus by using monoclonal antibodies. Infection and Immunity 44, 117 - 123.

Kobayashi, Y. Hasegawa, H. and Yamaguchi, T. (1985).Studies on the antigenic structure of Japanese Encephalitis virus using monoclonal antibodies.Microbiology and Immunology 29, 1069.

Kohler, G. & Milstein, C. (1975). Continuous cultures of fused cells secreting antibody of predefined specificity. Nature 256, 495 - 497.

Lafferty, K.J. (1963). The Interaction between virus and antibody. I. Kinetic studies. Virology 21, 61 - 75.

Lafferty, K.J. (1963). The Interaction between virus and antibody. II. Mechanism of the reaction. Virology 21, 76 - 90.

Lafferty, K.J. & Oertelis, S. (1963). The Interaction between virus and antibody. III. Examination of virus-antibody complexes with the electron microscope. Virology 21, 91 - 99.

Laigret, J. (1937). Ann. Med. 42, 463Lewis, M., Hospedales, J., Hull, B., and Rawlins, S.

(1989). Yellow fever virus activity - Trinidad and Tobago. Morbidity and Mortality Weekly Report 38, 57 - 59.

298

Liprandi, F. (1981). Isolation of plaque variants differing in virulence from the 17D strain of yellow fever virus. Journal of General Virology

363 - 370.Liprandi, F. and Walter, R. (1983). Replication of

virulent and attenuated strains of yellow fever virus in human monocytes and macrophage-like cells (U.937). Archives of Virology 7 6 , 51 - 61.

Lloyd, W. (1931). The myocardium in yellow fever. II. The myocardial lesions in experimental yellow fever. American Heart Journal 6, 504 - 516.

Lloyd, W. and Penna, H.A. (1933). Yellow fever virus encephalitis in South American monkeys. American Journal of Tropical Medicine 1 3 , 243 - 264.

Lobigs, M., Dalgarno, L., Schlesinger, J.J. and Weir, R.C. (1987). Location of a neutralization determinant in the E protein of yellow fever virus (17D vaccine strain). Virology 1 6 1 , 474 - 478.

Lwoff, A. (1959). Factors influencing the evolution of viral diseases at the cellular level and in the organism. Bacteriological Reviews 23, 109 - 124.

Mackow, E., Makino, Y., Zhao, B., Zhang, Y. M., Markoff, L ., White, A.B., Guiler, M., Channock, R. and Lai, C. J. (1987). The nucleotide sequence of dengue type 4 virus: Analysis of genes coding fornonstructural proteins. Virology 1 5 9 , 217 - 228.

Macnamara, F.N. (1953). Reactions following neurotropic yellow fever vaccine given by scarification in Nigeria. Transactions of the Royal Society of Tropical Medicine and Hygiene 47, 199 - 208.

Mandel, B. (1976). Neutralisation of poliovirus: Ahypothesis to explain the mechanism and the one- hit character of the neutralization reaction. Virology 69, 500 - 510.

Mandl, C.W., Guirakhoo, F., Holzmann, H., Heinz, F.X. and Kunz, C. (1989 a). Antigenic structure of the flavivirus envelope protein E at the molecular level, using Tick-borne encephalitis virus as a model. Journal of Virology 63, 564 - 571.

299

Mandl, C.W., Heinz, F.X., Stockl, E. and Kunz, C.(1989 b). Genome sequence of Tick-borneencephalitis virus (Western subtype) andcomparative analysis of nonstructuralproteins with other flaviviruses. Virology 173,291 - 301.

Marchoux, E., and Simond, P.L. (1906a). Etudes sur la fievre jaune, deuxième mémoire de la mission française a Rio de Janeiro. Annals Institute Pasteur 20, 16 - 40.

Marchoux and Simond, (1906b). Etudessur la fievre jaune, troisième mémoire de la mission française a Rio de Janeiro. Annals Institute Pasteur 20, 104 - 148.

Markoff, L. (1989). In vitro processing of dengue virus structural proteins:cleavage of the pre-membrane protein. Journal of Virology 63, 3345 - 3352.

Mason, P.W., McAda, P.C., Dalrymple, J.M., Fournier,M.J. and Mason, T.L. (1987). Expression of Japanese Encephalitis virus antigens in E. coli. Virology 158, 361 - 372.

Massey, R.J. and Schochetman, G. (1981). Topographical analysis of viral epitopes using monoclonal antibodies: Mechanism of virus Neutralization.Virology 115, 20 - 32.

Mathis, C., Sellards, A.W., and Laigret, J. (1928). Sensibilit du Macca rhesus au virus de lafivre jaune. C.R. Académie Science (Paris). 188, 604 - 606.

Mathews, J.H. and Roehrig, J.T. (1984). Elucidation of the topography and determination of the protective epitopes on the E glycoprotein of St. Louis Encephalitis virus by passive transfer with monoclonal antibodies. Journal of Immunology 132, 1533 - 1537.

Meers, P.D. (1957). Yellow fever vaccination by scarification with the 17D strain: an appreciation of the present position. Transactions of the Royal Society of Tropical Medicine and Hygiene 51, 338 - 345.

Meers, P.D. (1959). Adaptation of the 17D yellow fever virus to mouse brain by serial passage. Transactions of the Royal Society of Tropical Medicine and Hygiene 53, 445 - 457.

300

Meers, P.D. (1986). Yellow fever in Swansea, 1865. Journal of Hygiene Cambridge 97, 185 - 191.

Mendez, M.R., Calisher, C.H., Kruger, H., Sipan, F., Snchez, S., and Lazuick, J.S. (1984). A continuing focus of yellow fever in the Apurimac river valley, Ayacucho, Peru, and the first isolation of yellow fever virus in that country. Bulletin Pan American Health Organization 18, 172 - 179.

Meunier, D.M.Y., Aron, N., and Mazzariol, M.J. (1988).The 1987 yellow fever epidemic in Mali: viral andimmunological diagnosis. Transactions of the Royal Society of Tropical Medicine and Hygiene 82, 767 - 767.

Miller, B.R., Barrett, A.D.T., Mitchell, C.J., and Monath, T.P. (1986). Biochemical and biological analysis of a yellow fever vaccine virus from a fatal human encephalitis case. Journal Cell Biochemistry, supplement lOD, 298

Miller, B.R. & Adkins, D. (1988). Biological characterization of plague-size variants of yellow fever virus in mosquitoes and mice. Acta Virology

227 - 234Monath, T.P., Wilson, D.C., Lee, V.H., Stroh, G. (1973).

The 1970 yellow fever epidemic in Okwoga District Benue Plateau State, Nigeria. Bulletin of the World Health Organization 49, 113 - 121.

Monath, T.P., Craven, R.B., Adjukiewicz, A., Germain, M.(1980). Yellow fever in the Gambia. 1978 - 1979: epidemiological aspects with observations on the occurrence of Orungo virus infections. American Journal of Tropical Medicine and Hygiene 29, 912 - 928.

Monath, T.P., Brinker, K.R., Chandler, F.W., Kemp, G.E.(1981). Pathophysiological correlations in a rhesus monkey model of yellow fever,with soecial observations on acute necrosis of B cell areas of lymphoid tissues. American Journal of Tropical Medicine and Hygiene 30, 431 - 443.

Monath, T.P., Kinney, R.M., Schlesinger, J.J., Brandriss, M.W. and Bres, P. (1983). Ontogeny of Yellow fever 17D Vaccine: RNA Oligonucleotidefingerprint and monoclonal antibody analyses of vaccines produced World-wide. Journal of General Virology 64, 627 - 637

301

Monath, T.P. (1984 a). Impact of Arthropod-borne virus diseases on Africa and the Middle East. Chapter 24. 377 - 399. Applied Virology. Eds. Kurstak, E., Al-nakib W., Kurstak, C. Publisher: AcademicPress.

Monath, T.P., Schlesinger, J.J., Brandriss, M.W., Cropp, C.B., Prange, W.C. (1984, b). Yellow fevermonoclonal antibodies: Type-specificand crossreactive determinants identified byimmunofluorecence. American Journal of Tropical Medicine and Hygiene 33, 695 - 698.

Monath, T.P. (1987). Yellow Fever: A medically neglected disease. Report on a seminar. Reviews of Infectious Disease 9, 165 - 175.

Monath, T.P. (1989). Yellow fever. Chapter 51.142 - 218. The Arboviruses: Epidemiology andEcology, volume V. Eds. Monath, T.P. Publisher: CRC Press, Florida.

Morens, D.M., Venkateshan, C.N. and Halstead, S.B.(1987). Dengue-4 virus monoclonal antibodies identify epitopes that mediate immune infection enhancement of dengue-2 virions. Journal of General Virology 68, 91 - 98

Motohide, Y., Takashima, I. and Hashimoto, N. (1988). Enhanced neutralization of Flavivirus by monoclonal antibodies against Negishi virus. Microbiology and Immunology 32, 351 - 361.

Murphy, F.A. (1980). Morphology and morphogenesis. In St. Louis Encephalitis. 65 - 193. Eds. Monath, T. Publisher: American Public Health Association.Washington, D.C.

Nasidi, A., Monath, T.P., DeCock. K., Tomori, O., Cordellier, R., Olaleye, O.D., Harry, T.O., Adeniyi, A., Sorungbe, A.O., Ajose-Cocker, A.O., Van Der Laan, G., and Oyediran, A.B.O. (1989). Transactions of the Royal Society of Tropical Medicine and Hygiene 83, 401 - 406.

Ng, L.M. and Lau. L.C.L. (1988). possible involvement of receptors in the entry of Kunjin virus into Vero cells. Archives of Virology 100, 199 - 211.

302

Ôsatomi, K. and Sumiyoshi, H. (1990). Complete■ nucleotide sequence of Dengue type 3 virus genome RNA. Virology 176, 643 - 647.

Nitayaphan, S., Grant, J.A., Chang, G-J. and Trent, D.W.(1990). Nucleotide sequence of the virulent SA-14 strain of Japanese Encephalitis virus and its attenuated vaccine derivative, SA-14-14-2. Virology 177, 541 - 552.

Noguchi, H. (1919). Eitology of yellow fever. Journal of Experimental Medicine 29, 565 - 584.

Noguchi, H. (1925). Yellow fever research, 1918 - 1919: summary L. icteroides. Journal of Tropical Medicine 28, 185 - 193.

Novak, M., Gresikova, M., Sekeyova, M., Russ, G., Zikan, J., Pospisil, M. and Ciampor, F. (1983). Production of monoclonal antibodies with haemagglutination-inhibition activity to the Skalica strain from the Tick-borne Encephalitis complex. Acta Virology 27, 34 - 42.

Novokhatsky, A.S., Gaidomovich, S. Ya., Shutkova, T.M., Melnikova, E.E., Mikheeva, T.G., Sveshnikova, N.A. (1987). Monoclonal antibodies to yellow fever virus. Antibot Med Biotekhol 32, 899 - 903.

Nowak, T. and Wengler, G. (1987). Analysis of disulfides present in the membrane proteins of the West Nile flavivirus. Virology 156, 127 - 137.

Nowak, T., Farber, P.M., Wengler, G. and Wengler G.(1989). Analyses of the terminal sequences of West Nile virus structural proteins and of the ^ vitro translation of these proteins allow the proposal of a complete scheme of the proteolytic cleavages involved in their synthesis. Virology 169,365 - 376.*

Pan American Health Organization / World HealthOrganization (1981). Meeting on Modernization of yellow fever vaccine production Techniques. Washington, D.C. 1 2 - 1 4 January

Pan American Health Organization / World HealthOrganization (1984). Meeting to develop guidelines and protocols for the production of yellow fever vaccine in cell cultures. Washington D.C.2 1 - 2 3 February

Patterson, R. (1989). Dr, William Gorgas and his war with the mosquito. Canadian Medical Association Journal 141, 596 - 599.

303

Paul, S.D. (1966). Some biological properties of two variants of Kayasanur Forest disease virus. Indian Journal of Medical Research 54, 419 - 424.

Peiris, J.S.M., Porterfield, J.S., Roehrig, J.T. (1982).Monoclonal antibodies against the flavivirus West Nile. Journal of General Virology 58, 283 - 289.

Peltier, M. (1947). Yellow fever vaccination, simple or associated with vaccination against smallpox, of the populations of French West Africa by the method of the Pastuer Institute of Dakar. American Journal of public Health 37, 1026 - 1032.

Phillpotts, R.J., Stephenson, J.R. and Porterfield, J.S.(1985). Antibody-dependent enhancement of Tick- borne Encephalitis virus. Journal of General Virology 1831 - 1837.

Phillpotts, R.J., Stephenson, J.R. and Porterfield, J.S. (1987). Passive immunization of mice with monoclonal antibodies raised against Tick-borne Encephalitis virus. Archives of Virology 93, 295 - 301.

Philip, C.B. (1929). Preliminary report of further tests with yellow fever transmission by mosquitoes other than Aedes aegypti. American Journal of Tropical Medicine 9, 267 - 269.

Philip, C.B. (1930). Studies in transmission of experimental yellow fever by mosquitoes other than Aedes aegypti. American Journal of Tropical Medicine 10 1 -16

Pletnev, A.G., Yamshchikov, V.F. and Blinov, V. (1990). Nucleotide sequence of the genome and complete amino acid sequence of the polyprotein of Tick-borne encephalitis virus. Virology 174,250 - 263.

Poland, J.D., Calisher, C.H., Monath, T.P., Downs, W.G. and Murphy, K. (1981). Persistence of neutralisation antibody 30 - 35 years afterimmunization with 17D yellow fever vaccine. Bulletin of the World Health organization 59,895 - 900.

Porterfield, J.S. (1957). Use of goose cells in haemagglutination tests with Arthropod-borne viruses. Nature 180, 1201 - 1202

304

Porterfield, J.S., Casals, J., Chumakov, M.P.,Gaidamovich, S. Ya., Hannoun, C., Holmes, I.H.,Horzinek, M.C., Mussgay, M., Oker-Blom, N. andRussell, P.K. (1975/76). Bunyaviruses and Bunyaviridae. Intervirology 6, 13 - 24.

Porterfield, J.S., Casals, J., Chumakov, M.P.,Gaidamovich, S.Ya., Hannoun, C., Holmes, I.H., Horzinek, M.C., Mussgay, M., Oker-Blom, N., Russell, P.K and Trent, D.W. (1978). Togaviridae. Intervirology 9, 129 - 148.

Porterfield, J.S. (1980). Antigenic characterisation and classification of Togaviridae. 13 - 46. In The Togaviruses. Publisher; Academic press.

Porterfield, J.S. (1989). Yellow fever in west Africa; a retrospective glance. British Medical Journal 2 9 9 , 1555 - 1557.

Quereshi, A.A. and Trent, D.W. (1973). Group B arbovirus structural and non-structural antigens: serological specificity of solubilized intracellular viral proteins. Infection and Immunity 8, 993 - 999.

Rawlins, S.C., Hull, B., Chadee, D.D., Martinez, R., LeMaitre, A., James, F., and Webb, L. (1990). Transactions of the Royal Society of Tropical Medicine and Hygiene 84, 142 - 143.

Rice, C.M., benches, E. M., Eddy, S.R., Shin, S.J.,Sheets, R.L. and Strauss, J.H. (1985). Nucleotide sequence of yellow fever virus: Implications forflavivirus gene expression and evolution. Science 229, 726 - 733.

Rice, C.M., Strauss, E.G. and Strauss J.H. (1986). Structure of the flavivirus genome. In The Togaviridae and Flaviridae 279 - 326.Eds. Schlesinger, S. and Schlesinger M.J.

Publisher: Plenum New York.Rice, C.M., Grakoui, A., Galler, R., Chambers, T.J.

(1989). Transcription of infectious yellow fever virus RNA from full-length cDNA templates produced by in vitro ligation. New Biologist 1, 285 - 296.

Rico-Hesse, Pallansch, M.A., Nottay, B.K. and Kew, O.M.(1987). Geographic distribution of wild poliovirus Type 1 genotypes. Virology 1 6 0 , 311 - 322.

305

Roehrig, J.T., Mathews, J.H., and Trent, D.W. (1983).Identification of epitopes on the E glycoprotein of St. Louis Encephalitis virus using monoclonal antibodies. Virology 128, 118 - 126.

Roehrig, J.T., Hunt, A.R., Johnson, A.J. and Hawkes, R.A. (1989). Synthetic peptides derived from deduced amino acid sequence of the E-glycoprotein of Murray Valley Encephalitis virus elicit antiviral antibody. Virology 171, 49 - 60.

Ruiz-Linares, A., Cahour, A., Despres P., Girard, M. and Bouloy, M. (1989). Processing of yellow fever virus polyprotein: role of cellular proteases inmaturation of the structural proteins. Journal of Virology 63, 4199 - 4209.

Sabin, A.B. (1951). Hemagglutination by viruses affecting the human nervous system. Federation Proceedings 10, 573 - 578.

Schindler, R. & Hallauer, C. (1963). Prufung von gelbfiebervirus-varianten aus menschlichen gewebekulturenim affenversuch. ARCHIV GESMATE VIRUSFORSCHUNG 33, 345 - 357.

Schlesinger, J.J. & Brandriss, M.W. (1981 a). Growth of 17D yellow fever virus in a macrophage-like cell line, U937: role of Fc and viral receptors inantibody-mediated infection. The Journal of Immunology 127, 659 - 665.

Schlesinger, J.J. & Brandriss,M.W. (1981 b). Antibody- mediated infection of Macrophage-like cell lines with 17D-yellow fever virus. Journal of Medical Virology 8, 103 - 117.

Schlesinger, J.J. & Brandriss, M.M. (1983). 17D yellow fever virus infection of P388D^ cells mediated by monoclonal antibodies: properties of themacrophage Fc receptor. Journal of General Virology 64, 1255 - 1262.

Schlesinger, J.J., Brandriss, M.W., and Monath, T.P.(1983). Monoclonal antibodies distinguish between wild type and vaccine strains of yellow fever virus by neutralisation, hemagglutination inhibition and immune precipitation of the virus envelope protein. Virology 125, 8 - 1 7 .

306

Schlesinger, J.J., Walsh, E.E. and Brandriss, M.W.(1984). Analysis of 17D yellow fever virus envelope protein epitopes using monoclonal antibodies. Journal of General Virology 65, 1637 - 1644.

Serie, C., Lindrec, A., Poirier, A., Andral, L., and Neri, P. (1968). Etudes sur la fievre jaune enEthiopie. Bulletin of the World HealthOrganization 38, 879 - 884.

Sever, J.L., Ley, A.C., Wolman, F., Caplan, B.M., Crockett, B.W., Turner, H.C. (1964). Utilization of disposable plastic plates, with a serologic microtechnic. The American Journal of Clinical Pathology 41, 167 - 170.

Shannon, R.C, Whitman, L., and Franca, M. (1938). Yellow fever virus in jungle mosquitoes. Science 88, 110.

Smith, C.E.G., and Gibson, M.E. (1986). Yellow fever in South Wales, 1865. Medical History 30,322 - 340.

Soper, F.L., Penna, H.A., Cardoso, E., Serafim, J., Frobisher, M., and Pinheiro, J. (1933). Yellow fever without Aedes aegypti: study of ruralepidemic in Valle do Chanaan, Espirito Santo, Brazil, 1932. American Journal of Hygiene 18,555 - 587.

Srivastava, A.K., Matsuo, S. and Igarashi, A. (1988). Monoclonal antibodies against Japanese Encephalitis virus envelope glycoprotein V3 (E). Tropical Medicine 30, 49 - 61.

Stephenson, J.R., Lee, J.M. and Wilton-smith, P.D.(1984). Antigenic variation among members of the Tick-borne Encephalitis virus complex. Journal of General Virology 65, 81 - 89.

Stokes, A., Bauer, J.H., and Hudson, N.P. (1928). Experimental transmission of yellow fever tolaboratory animals. American Journal ofTropical Medicine 8, 103 - 164.

307

Stollar, V., Stollar, B.D., Koo, R., Harrap, K.A. andSchlesinger, R.W. (1976). Sialic acid contents ofSindbis virus from vertebrate and mosquito cells. Equivalence of biological and immunological properties. Virology 69, 104

Strode, G.K. (1951). Yellow fever. Publisher: McGraw Hill book company New York.

Sumiyoshi, H., Mori, C., Fuke, I., Morita, K., Kuhara, S., Kondou, J., Kikuchi, Y., Nagamatu, H. andIgarashi, A. (1987). Complete nucleotide sequence of the Japanese Encephalitis virus genome RNA. Virology 161, 497 - 510.

Swinscow, T.D.V. (1976). Statistics at square one. Publisher: British Medical Association, London.

Tannock, G.A., Wark, M.C. & Hair, C.G. (1980). The development of an improved experimental yellowfever vaccine._____ Journal____of____BiologicalStandardization 8, 23 - 34.

Tarr, G.C. and Lubiniecki, A.S. (1976). Chemically induced Temperature-Sensitive Mutants of Dengue virus Type-2: Comparison of temperaturesensitivity in vitro with infectivity in suckling Mice, Hamsters and Rhesus monkeys. Infection and Immunity 13, 688 - 695.

Taylor, H.P. & Dimmock, N.J. (1985). Mechanisms of neutralization of influenza virus by IgM. Journal of General Virology 66, 903 - 907.

Theiler, M. (1930). Studies on action of yellow fever virus in mice. Annals Tropical Medicine 24, 249 - 272.

Theiler, M. (1933). The susceptibility of guinea pigs to the virus yellow fever. American Journal of Tropical Medicine 13, 399 - 414.

Theiler, M. and Smith, H.H. (1937 a). Use of yellow fever virus modified by in vitro cultivation for human immunization. Journal of Experimental Medicine 65, 787 - 800.

Theiler, M. and Smith, H.H. (1937 b). The effect of prolonged cultivation in vitro upon the pathogenicity of yellow fever virus. Journal of Experimental Medicine 65, 767 - 786.

308

Theiler, M, (1951). The Virus. 43 - 136. YellowFever. Eds. Strode G.K. Publisher: McGraw-HillBook company. New York.

Theiler, M. (1952). Yellow fever. Viral and Rickettsial infections of man. 531 - 554. 2nd edition.Philadelphia: Lippincott.

Theiler, M. and Downs, W.G. (1973). The arthropod-borne viruses of vertebrates. Publisher Yale University Press. New Haven.

Trapido, H., and Galindo, P. (1956). The epidemiology of yellow fever in Middle America. Experimental Parasitology 3, 285 - 323.

Trent, D.W. (1977). Antigenic characterization of flavivirus structural proteins separated by isoelectric focusing. Journal of Virology 22, 608 - 618.

Trent, D.W. and Naeve C.W. (1980). Biochemistry and replication. In St. Louis encephalitis 159 - 199. Ed. Monath, T. Publisher: American Public Health Association. Washington D.C.

Warren, A.J. (1951). Landmarks in the conquest of yellow fever. 1 - 3 9 . In Yellow Fever. Eds. Strode, G.K. Publisher: Mcgraw-hill book company

Waters, T.D., Anderson, P.S., Beebe, G.W., and Miller, R.W. (1972). Yellow fever vaccination,avian leukosis virus and cancer risk in man. Science 177, 76 - 77.

Weekly Epidemiological Record - World HealthOrganization (1989). 64, (4) 22 - 28.

Weekkly Epidemiology Record - World HealthOrganization (1989) 64, (6) 3 7 - 4 4 .

Wengler, G., Castle, E., Leidner, U., Nowak, Th.,Wengler, G. (1985). Sequence analysis of themembrane protein V3 of the flavivirus West Nile virus and of its gene. Virology 147, 264 - 274.

Wengler, G., Wengler, G., Nowak, T. and Wahn, K. (1987). Analysis of the influence of proteolytic cleavage on the structural organization of the surface of the West Nile flavivirus leads to the isolation of a protease-resistant E protein oligomer from theviral surface. Virology 160, 210 - 219.

309

Wengler, G and Wengler, G. (1989). Cell-associated West Nile flavivirus is covered with E+Pre-M protein heterodimers which are destroyed and reorganized by proteolytic cleavage during virus release. Journal of virology 63, 2521 - 2526.

Westaway, E.G. (1980). Replication of flaviviruses. In The Togaviruses. 531 - 581. Ed. Schlesinger, R.W. Publisher: Academic Press. New York.

Westaway, E.G., Brinton, M.A., Gaidamovich, S. Ya., Horzinek, M.C., Igarashi, A., Kaariainen, L., Lvov, D.K., Porterfield, J.S., Russell, P.K. and Trent, D.W. (1985). Flaviviridae. Intervirology

183 - 192.Westaway, E.G. (1987). Flavivirus replication strategy.

In Advances in virus research 33, 45 - 90.Publisher: Academic press. New York.

White, D.C. and Fenner, J.F. Togaviruses and Flaviviruses. Chapter 20. 479 - 508. In MedicalVirology. Publisher: Academic press. New York.

Whitman, L. (1935). The response to yellow fever virus in the nonsusceptible rabbit. Journal of Immunology

99 - 110.Whitman, L. (1939). Failure of Aedes Aegypti to transmit

yellow fever cultured virus (17D). American Journal of Tropical Medicine and Hygiene 19, 19 - 26.

WHO Technical Report Series, No. 179. (1959).WHO Technical Report Series, No. 479. (1971). Third

Report of the Expert Committee on Yellow Fever.WHO Technical Report Series, No. 658. (1981). WHO

committee on biological standardisation.WHO (1986). Prevention and control of yellow fever in

Africa. 94pp. Geneva.WHO Technical Report series. No. 771. (1988).

Requirements for yellow fever vaccine. 208 - 210.Wiktor, T.J. and Koprowski, H. (1980). Antigenic

variants of Rabies virus. Journal of Experimental Medicine 152, 99 - 112.

310

Winkler, G., Heinz, F.X. and Kunz, C. (1984). Exclusive use of high performance liquid chromatography techniques for the isolation and sequencing of a viral membrane protein. Journal of Chromatography 297, 63 - 7 3 .

Wreghitt, T.G. and Morgan-Capner, P. (1990). Elisa inthe Clinical Microbiology Laboratory. 2 - 5 . eds. Wreghitt, T.G. and Morgan-Capner, P. Published by the Public Health Laboratory Service.

Wright, P.J. (1982). Envelope protein of the flavivirus Kunjin is apparently not glycosylated. Journal of General Virology 59, 29 - 38.

Wright, P.J., Warr, H.M. and Westaway, E.G.(1983). Comparisons by peptide mapping of proteins specified by Kunjin, West Nile and Murray Valley Encephalitis viruses. Australian Journal of Experimental Biological and Medical Science 61, 641 - 653.

311

APPENDIX

312

LOG^ HAI TITRES OF THE A, B, G AND H McAbs WITH VIRUSES USED IN THE GLUSTAN COMPUTER PROGRAMME

VIRUSES17D204 17DD

Wild Type vaccine vaccineNo. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1617D 17]A F B H F F W U U U R B D C C NS V 4 0 r r H K K K U R A 0 0 II V 1 L P 0 P P S Z K L L GB L P P P S PI + - PMcAbA3 1.0 2.2 1.3 1.9 1.9 2.2 0 1.6 1.9 1.9 2.2 1.9 1.9 2.2 1.0 l.(A4 1.5 0 1.0 1.3 1.9 1.0 0 1.0 1.6 1.6 1.6 1.0 1.3 0 1.3 1.'A6 1.3 0 1.0 1.6 2.2 1.9 0 1.6 1.9 1.9 2.2 1.3 1.6 1.6 1.6 2.!B19 0 0 0 0 0 1.3 0 0 1.0 1.0 0 1.0 1.3 1.0 1.0 0B26 1.6 1.9 0 1.0 1.3 2.2 1.6 1.0 1.0 1.3 1.0 1.6 1.6 2.5 2.2 2.;B37 1.0 1.6 0 1.9 1.9 1.6 0 1.6 1.9 1.9 2.2 1.3 1.3 2.2 1.6 2.:B39 1.6 1.9 1.6 1.9 2.2 1.3 1.6 2.2 1.9 1.9 2.5 0 1.0 1.9 1.6 1.'B44 1.6 2.2 1.9 2.2 2.2 1.9 1.3 2.5 2.8 2.5 2.2 1.9 1.9 2.2 1.6 2.:B45 1.9 1.3 1.3 1.6 1.6 2.2 0 1.3 2.2 1.9 1.6 1.3 1.0 2.5 1.9 1.'G6 1.0 1.9 1.9 1.0 1.0 1.3 2.5 1.6 1.6 1.0 1.3 1.0 1.3 0 1.6 0G23 2.2 0 0 1.9 1.9 2.2 1.9 1.6 2.5 2.2 1.9 0 0 2.2 2.5 2.:G46 1.0 1.6 0 1.0 1.3 0 1.6 1.6 1.3 1.3 1.0 1.0 1.3 1.1 1.3 0G49 1.3 0 0 1.3 1.0 0 0 1.3 1.0 1.0 1.0 0 1.0 1.0 1.0 1.;G65 1.9 1.6 0 2.2 2.5 2.2 1.6 2.2 2.2 2.2 2.5 0 1.0 1.9 2.2 2.1H5 0 0 0 1.6 1.3 1.3 1.9 1.6 1.9 1.6 1.6 1.9 2.2 1.6 1.9 1.;H6 1.0 1.0 1.0 1.0 1.0 1.0 1.9 1.0 1.0 1.0 1.0 1.6 1.9 1.0 1.6 1.H8 2.2 2.2 2.2 1.6 1.9 1.6 2.5 1.9 1.9 1.9 1.6 2.2 2.5 1.6 2.2 1.HID 1.0 1.0 1.0 1.3 1.3 1.0 1.3 1.3 1.3 1.6 1.3 1.6 1.6 1.6 1.6 1.Hll 2.2 1.9 2.2 1.6 1.3 1.6 1.9 1.9 2.8 1.9 1.6 1.6 1.9 1.0 1.6 2.H14 1.9 0 0 1.6 1.3 1.9 1.9 1.6 1.6 1.6 1.6 1.6 1.9 1.9 1.6 0H20 2.2 2.5 2.5 2.5 1.0 1.0 2.5 2.2 2.2 2.5 2.5 2.5 2.5 2.2 2.5 1.H23 0 0 0 1.3 1.0 0 1.9 1.0 1.0 1.3 1.0 1.9 1.9 0 1.9 0H25 0 0 0 1.0 1.3 2.2 0 1.0 1.9 2.2 1.0 0 0 2.2 0 2.H30 2.2 2.2 0 1.3 1.0 1.3 2.2 1.6 1.3 1.6 1.3 2.5 2.2 1.0 2.5 1.H31 1.9 1.9 2.2 1.6 1.3 1.3 1.9 1.9 1.9 1.6 1.6 1.9 1.9 1.0 1.9 1.H32 2.5 2.5 2.5 2.2 2.2 0 2.5 2.5 2.2 1.9 2.2 2.8 3.1 2.2 2.8 1.H33 2.2 1.9 2.2 1.6 1.3 1.6 1.9 1.6 1.5 1.6 1.3 2.2 1.9 1.6 2.2 1.

313

LOG Q HAI TITRES OF THE A, B, G AND H McAbs WITH VIRUSES USED IN THE CLUSTAN COMPUTER PROGRAMME

(CONTINUED)VIRUSES

YF17D-204Vaccines

FNVVaccines

YF17D20Vaccin

No. 17 18 19 20 21 22 23 24 25 26 27 28 29 30F S S S U A F F F IR A A A S U N N N NG P 8 1 A S V V V D

P 4 0 A B C D T N F I9 8 T C PA 0o X Y

McAbsz

A3 2.2 2.2 2.5 1.9 1.6 1.9 1.3 2.2 1.9 1.3 1.9 1.0 1.6 1.3A4 0 0 1.3 1.3 1.0 1.0 1.9 0 0 1.6 1.3 1.9 1.0 1.6A6 0 0 1.0 1.9 1.3 1.3 2.2 0 0 . 1.6 1.9 1.3 1.9 1.6B19 1.3 1.3 1.3 1.0 1.0 1.0 0 0 0 1.0 1.3 1.0 1.0 1.0B26 1.9 1.9 2.5 1.6 1.0 1.0 1.9 2.5 1.9 2.2 1.9 1.9 1.6 1.9B37 1.9 1.9 2.2 2.2 1.0 1.3 1.9 1.6 1.6 1.6 1.6 1.9 1.6 2.2B39 2.2 2.8 2.8 2.2 1.0 1.3 2.2 2.2 1.9 1.9 1.0 2.2 1.3 2.2B44 1.9 2.5 2.8 2.2 1.6 1.6 1.9 2.8 1.9 2.2 2.2 2.5 2.2 1.9B45 1.6 1.9 2.5 2.2 1.3 1.3 1.9 2.2 1.6 1.9 1.6 1.6 1.6 1.6G6 0 0 1.0 1.0 1.0 1.0 1.0 0 1.6 1.3 1.0 0 1.0 1.0G23 2.8 2.8 2.8 2.2 1.3 1.6 2.2 2.5 1.6 2.2 0 2.2 1.9 2.5G46 1.6 1.0 1.0 1.0 0 0 0 0 1.6 0 0 0 0 1.0G49 1.3 1.3 1.0 1.3 1.3 1.0 0 0 1.0 0 0 1.3 1.6 1.0G65 2.2 2.8 1.9 1.9 0 1.0 1.9 2.5 2.5 2.8 0 2.8 1.9 2.8H5 0 0 0 1.6 0 0 1.6 0 0 1.9 1.9 1.3 2.2 1.3H6 1.9 2.5 1.6 1.3 1.0 1.0 1.6 2.2 2.5 1.6 1.0 1.0 1.3 1.6H8 2.2 2.8 1.6 0 0 0 1.3 2.8 2.8 1.9 2.2 1.6 0 1.9HIO 1.0 2.5 1.6 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0Hll 0 2.2 2.2 1.3 2.5 1.0 2.2 0 0 1.3 1.9 2.2 1.6 1.9H14 0 0 0 1.9 2.2 1.3 1.9 0 0 0 1.6 0 2.8 1.6H20 0 0 0 1.0 2.2 0 1.6 0 0 2.2 2.5 2.8 2.2 1.6H23 2.2 2.8 2.2 1.0 1.0 1.0 1.0 2.8 1.6 1.0 0 1.0 2.2 0H25 1.9 1.6 1.6 1.6 2.5 1.0 2.2 1.6 1.3 2.2 0 2.8 2.5 2.2H30 1.3 0 0 2.2 1.0 1.6 1.0 0 2.2 1.9 2.2 1.0 2.2 1.9H31 2.2 1.3 1.0 1.6 1.0 1.3 0 0 1.6 1.6 1.6 1.6 1.9 1.6H32 2.2 1.3 1.3 1.6 1.0 1.0 0 0 2.5 1.6 2.8 1.0 1.3 1.6H33 1.9 1.3 1.3 1.0 1.0 1.0 1.9 0 2.2 1.9 2.2 1.9 1.0 1.9: USA VACCINE BATCHES; A: P-16065, B: 6145, C: 2848, D: 4E158 : AUST VACCINE BATCHES; X: 008, Y: 0331

314

LOG g HAI TITRES OF THE A, B, G AND H McAbs WITH VIRUSES USED IN THE CLUSTAN COMPUTER PROGRAMME

(CONTINUED)

VIRUSESYF17D-204 17DDVaccines Vaccine

No, 31 32 33 34

McAbsRF318 DAK COL BRZf

A3 1.6 2.5 1.0 1.0A4 1.6 1.9 1.3 1.0A6 1.6 2.2 1.0 1.6B19 0 0 0 0B26 1.3 1.6 1.3 1.6B37 1.3 2.2 2.2 2.2B39 0 0 1.9 2.2B44 1.9 2.5 2.2 2.2B45 1.6 2.5 1.6 1.9G6 0 0 1.0 1.3G23 1.9 2.8 1.0 1.3G46 0 0 1.0 1.0G49 0 0 0 0G65 0 2.2 1.0 1.6H5 1.9 2.5 1.3 2.2H6 1.9 2.2 1.0 1.6H8 1.6 1.3 1.3 1.3HIO 0 0 1.0 0Hll 1.6 2.5 1.6 2.2H14 0 0 0 0H20 0 2.5 1.9 1.9H23 0 2.5 0 1.0H25 0 0 0 1.3H30 2.8 2.5 0 0H31 2.5 2.2 1.3 1.6H32 0 0 1.9 2.5H33 2.2 1.6 1.0 1.0

315

OF HAI TITRES USING THE F McAbsMcAbs

FIO F17 F19 F20 F23 F25 F28 F33Viruseswild-type

F W 1.6 0 0 2.2 1.9 1.6 1.9 1.9Asibi 1.3 0 0 1.9 1.9 1.3 1.9 2.2

FNVFNV FC 1.0 2.2 1.0 1.3 2.2 1.3 1.3 1.6FNV IP 1.0 1.9 1.0 1.3 1.9 1.0 1.0 1.0FNV NT 1.3 2.2 0 1.3 1.9 1.3 1.0 1.3

17D-204USA 6145 1.9 0 0 1.6 2.2 1.3 2.2 1.9USA P-16065 1.3 1.6 1.0 1.0 1.3 1.0 1.0 1.0UK 0 1.9 1.6 1.0 1.9 1.0 1.0 1.6

17DDBrazil 1.3 0 0 1.0 1.3 1.0 1.0 1.3

17DUSSR 1.6 1.9 1.0 2.2 1.9 1.3 1.9 2.5

316

Documents

study of yellow fever vaccine viruses by Terence Neil ...epubs.surrey.ac.uk/855790/1/27606615.pdf · ANTIGENIC ANALYSIS OF EPITOPES ON YELLOW FEVER VIRUSES AND ITS VACCINES BY HAI