Bioengineering Viral Subunits for Influenza Vaccine

Bioengineering Viral Subunits for Influenza Vaccine Development

Jarurin Waneesorn

Master of Science (Genetic Engineering)

A thesis submitted for the degree of Doctor of Philosophy at

The University of Queensland in 2017

Australian Institute for Bioengineering and Nanotechnology

i

Abstract

Avian influenza (AI) is a respiratory disease of birds resulting from infection with influenza A

viruses. Outbreaks of the highly pathogenic form of AI (HPAI) result in the loss of millions of birds

and remain a global threat to the poultry industry. The high mutation rate of the HPAI virus and a

proven ability to cross the species barrier raises concerns about the pandemic potential of this zoonotic

disease. The ongoing HPAI virus outbreaks in poultry, coupled with increasing reports of human

cases worldwide, demonstrate the need for practical means to control its rapid spread. Poultry

vaccination provides a means for prevention and control of AI by significantly decreasing the risk of

viral transmission. As part of control and prevention programs against HPAI, poultry vaccines are

commonly used in countries where HPAI viruses are endemic (i.e. parts of Asia and Africa).

However, the currently licensed poultry vaccines against AI, which are mainly produced in

embryonated chicken eggs, have potential drawbacks in terms of their production. The slow

production time and requirement for at least one egg per vaccine dose make it impractical to supply

sufficient doses for mass poultry vaccination during an outbreak. Vaccine manufacturing based on a

microbial platform offers a promising alternative due to the speed, cost and scale associated with

cultivation of the microorganism. Previously, process simulation of a bacterially-produced capsomere

platform, based on the VP1 capsid protein of murine polyomavirus, has indicated that 320 million

doses of capsomere vaccines could be produced within 2.3 days. Economic analysis has shown that

the capsomere vaccine can be produced at a cost of less than one cent per dose, presenting a financial

advantage over other vaccines in response to AI epidemics and pandemics. This thesis addresses some

of the challenges associated with AI vaccine production by demonstrating the design of a novel

modular capsomere platform and production process that allows a strain-matched vaccine to be

delivered at a speed, cost and scale suitable for effective disease control through poultry vaccination.

The major outcomes of this work were: (i) demonstration of the potential of structure-based modular

capsomeres as an AI vaccine candidate; (ii) demonstration of high immunogenicity of low purity

modular capsomeres prepared using a simpler method; and (iii) development of a simple and rapid

non-chromatographic purification process that is economical for use in poultry vaccine production.

This study is, to the best of my knowledge, the first to report on the development of a capsomere

vaccine platform (designated VP1dC) that allows modularisation (a strategy for displaying target

antigenic components of pathogens, known as modules, onto a platform base carrier) of a large

antigenic module: a structure-based designed influenza hemagglutinin (tHA1). This modular

ii

capsomere (CaptHA1) comprises five copies of modular VP1dC-tHA1 and is efficiently produced in

a stable form using Escherichia coli. CaptHA1 demonstrates high immunogenicity and confers

complete protection against viral challenge when administered to chickens. The simplified production

process allows CaptHA1 to be generated without solubility tags. CaptHA1 is purified using non-

chromatographic methods, without the requirement of a protein refolding process, and retains its

structural integrity and biological functions. The outcomes of this work contribute toward the

development of a microbial-based modular capsomere platform for large antigen presentation. The

platform has the potential for manufacture of vaccines for administration to poultry during AI

epidemics and pandemics.

iii

Declaration by author

This thesis is composed of my original work, and contains no material previously published or written

by another person except where due reference has been made in the text. I have clearly stated the

contribution by others to jointly-authored works that I have included in my thesis.

I have clearly stated the contribution of others to my thesis as a whole, including statistical assistance,

survey design, data analysis, significant technical procedures, professional editorial advice, financial

support and any other original research work used or reported in my thesis. The content of my thesis

is the result of work I have carried out since the commencement of my higher degree by research

candidature and does not include a substantial part of work that has been submitted to qualify for the

award of any other degree or diploma in any university or other tertiary institution. I have clearly

stated which parts of my thesis, if any, have been submitted to qualify for another award.

I acknowledge that an electronic copy of my thesis must be lodged with the University Library and,

subject to the policy and procedures of The University of Queensland, the thesis be made available

for research and study in accordance with the Copyright Act 1968 unless a period of embargo has

been approved by the Dean of the Graduate School.

I acknowledge that copyright of all material contained in my thesis resides with the copyright

holder(s) of that material. Where appropriate I have obtained copyright permission from the copyright

holder to reproduce material in this thesis and have sought permission from co-authors for any jointly

authored works included in the thesis.

iv

Publications during candidature

Peer-reviewed paper

1. Waneesorn J, Wibowo N, Bingham J, Middelberg APJ, Lua LHL*. Structural-based designed

modular capsomere comprising HA1 for low-cost poultry influenza vaccination. Vaccine,

2018. 36(22): p. 3064-3071. Available online 25 November 2016. ISSN: 0264-410X.

https://doi.org/10.1016/j.vaccine.2016.11.058.

*Corresponding author

Conference abstracts

1. Waneesorn J, Wibowo N, Middelberg APJ, Lua. LHL. Structural-based designed modular

capsomere comprising HA1 as low-cost poultry influenza vaccine. Australian Institute for

Bioengineering and Nanotechnology Student Conference. 2 October 2015. Brisbane,

Australia.

2. Waneesorn J, Middelberg APJ, Lua. LHL Structural-based designed modular capsomere

comprising HA1 as low-cost poultry influenza vaccine. Vaccine Technology VI. 12-17 June

2016. Albufeira, Portugal.

3. Waneesorn J, Middelberg APJ, Lua LHL. Bioprocess engineering of a bacterially-produced

low-cost modular capsomere vaccine for avian influenza. 10th Vaccine Congress. 4-7

September 2016. Amsterdam, The Netherlands.

4. Waneesorn J, Middelberg APJ, Lua LHL. Simple manufactured microbial platform as rapid

and low-cost modular capsomere vaccine for poultry vaccination. BioProcessing Asia. 5-8

December 2016. Phuket, Thailand.

v

Publications included in this thesis

1. Waneesorn J, Wibowo N, Bingham J, Middelberg APJ, Lua LHL. Structural-based designed

modular capsomere comprising HA1 for low-cost poultry influenza vaccination. Vaccine,

2018. 36(22): p. 3064-3071. Available online 25 November 2016. ISSN: 0264-410X.

https://doi.org/10.1016/j.vaccine.2016.11.058.

Incorporated as Chapter 3.

Contributor Statement of contribution

Waneesorn J (Candidate) Conception and design (60%)

Analysis and interpretation (65%)

Drafting and production (50%)

Wibowo N Conception and design (5%)

Analysis and interpretation (5 %)


Bingham J Conception and design (5%)



Middelberg APJ Conception and design (5%)



Lua LHL Conception and design (25%)



vi

Contributions by others to the thesis

This thesis was drafted and written by the candidate under the supervision of Professor Linda Lua

and Professor Anton Middelberg.

Challenge study with HPAI virus in Chapter 3 was performed by Jeff Butler and the Animal Studies

Team and Histology Team of Australian Animal Health Laboratory (AAHL).

In vivo immunogenicity study in Chapters 3 and 4 was performed by the candidate with assistance

from Nani Wibowo, Arjun Seth, Andrea Schaller and Nicolas Pichon.

Homology modelling in Chapter 4 was performed by Evelyne Deplazes.

Hemagglutination inhibition assay in Chapter 4 was performed by Julie Cook.

Protein expression in Chapter 5 was performed by the candidate in collaboration with Hayley

Charlton Hume.

Preparation of modular capsomere precipitation by using PEG was performed by Hayley Charlton

Hume.

All of the data and results presented in this thesis are solely the work of the candidate with exception

of the following figures in which the data were collected with collaborators:

Figure 3-6 data were obtained from John Bingham;

Figure 4-1 data were obtained from Evelyne Deplazes.

Statement of parts of the thesis submitted to qualify for the

award of another degree

None

vii

Research Involving Human or Animal Subjects

1. Animal ethics approval details (Appendix A):

Chief Investigator: Dr Nani Wibowo

Title: Development of low-cost, rapid-response influenza poultry vaccine candidates

AEC Approval Number: AIBN/235/14/QSF

Approval Duration: 02-Sep-2014 to 02-Sep-2015

Funding Body: QLD Smart Futures

Group: Production and Companion Animal

Other Staff/Student: Linda Lua, Jarurin Waneesorn, Arjun Seth, Thanh Doan, Andrea

Schaller, Vivek Gurusamy, Arun Kumar, Nicolas Pichon

Location: Gatton Bldg 8258-Poultry Research Unit

Approving committee: Production and Companion Animals (PCA) Ethics Committee

2. Animal Ethics approval details (Appendix B):

Memorandum to John Bingham

Project Title: Proof-of-concept testing of an avian influenza nanovaccine in chickens

Meeting Number: 2015/3

Project Protocol Number: 1767

Approval Duration: June-2015 to December 2015

Species and Number of Animals Approved: Chickens (24)

Approving committee: CSIRO Animal, Food and Health Sciences Australian Animal Health

Laboratory Animal Ethics committee

viii

Acknowledgements

Foremost, I would like to express my sincere gratitude to my principal supervisor and mentor,

Professor Linda Lua for her continuous support, valuable guidance and constant encouragement, and

particularly, her patience, motivation and immense knowledge throughout my Ph.D. study. Her

guidance helped me in all the time of research and writing of manuscripts and this thesis. I would also

like to thank my co-supervisor, Professor Anton Middelberg, for his unwavering support, motivation,

kindness and valuable insights into various aspects of my research. I am profoundly grateful to my

former co-supervisor, Dr. Nani Wibowo for her support, patience, and kindness.

Beside my supervisor, I would like to thank my examiners: Professor Lars Nielsen and Professor

Joanne Meers for their insightful comments, encouragement and their questions which helped me to

widen my research from various perspectives.

My sincere thanks go to all former and present members of the Centre for Biomolecular Engineering

for their support, friendship, companionship, particularly during those long hours and weekends in

the lab, and for all the fun we have had in the last four years. Special thanks to Nicolas Pichon, Arjun

Seth, and Andrea Schaller for their never-ending encouragement and companionship during the good

and bad times throughout my time in Australia. I am deeply grateful to Dr. David Wibowo for his

consistent encouragement and friendship. I am also grateful to Alemu Tekewe Mogus for stimulating

discussion and inspiring exchange of ideas. Thanks to everyone else who have supported me in many

ways: Dr. Natalie Connors, Dr. Frank Sainsbury, Dr. Chunxia Zhao, Dr. Balaji Somasundaram,

Lesley Forest, Noor Dashi, Liang Zhao, Yue Hui, Fiona Ritchie, Russell Wilson, Thejus Baby, Doan

Than Tam and Hossam Tayeb.

I would also like to thank all members of the Protein Expression Facility for their technical support

and helpful discussion. A very special gratitude goes to Dr. Hayley Charlton Hume for her consistent

and uncompromising help in the lab and for proof-reading manuscript drafts and this thesis. Special

thanks to Emilyn Tan, Yuanyuan Fan and Chris Munro for their encouragement and friendship

throughout my Ph.D. I am also grateful to all AIBN staff for their unfailing support and assistance.

I am thankful and sincere appreciation for a Royal Thai Government Scholarship and The University

of Queensland Top-Up Scholarship for financially supporting my study in Australia. I would also like

ix

to thank Ms Kamonwan Sattayayut, Ms Somchitr Wattanatassi, Ms Ploypaphat Ruanwong and Ms

Kewalee Somboon from Office of Education Affairs, Royal Thai Embassy, Australia for their kind

support throughout my study.

I would also like to thank my friends in Australia and Thailand and my colleagues at Regional

Medical Sciences Centre 1, Chiang Mai for their support and encouragement throughout my study.

My sincere thanks go to Alex Wilson and Penny James for their support, encouragement and warm

hospitality throughout my time in Australia.

Last but not the least, I would like to thank my family for their everlasting love and unconditional

support throughout my life. Without their support, my success would not have been possible.

Financial support

This research was support by:

Australian Research Council (ARC Discovery Project DP160102915)

Royal Thai Government Scholarship

The University of Queensland Top-Up Assistance

x

Keywords

Influenza, hemagglutinin, vaccine, poultry, murine polyomavirus, capsomere, structure, design,

modularisation, precipitation

Australian and New Zealand Standard Research

Classifications (ANZSRC)

ANZSRC code: 100703, Nanobiotechnology, 60%

ANZSRC code: 100302, Bioprocessing, Bioproduction and Bioproducts, 40%

Fields of Research (FoR) Classification

FoR code: 0903, Biomedical Engineering, 40%

FoR code: 1007, Nanotechnology, 40%

FoR code: 1107, Immunology, 20%

xi

Table of Contents

Abstract ................................................................................................................................................ i

Declaration by author .......................................................................................................................iii

Publications during candidature ..................................................................................................... iv

Publications included in this thesis ................................................................................................... v

Contributions by others to the thesis............................................................................................... vi

Statement of parts of the thesis submitted to qualify for the award of another degree ............. vi

Research Involving Human or Animal Subjects ........................................................................... vii

Acknowledgements..........................................................................................................................viii

Financial support .............................................................................................................................. ix

Keywords ............................................................................................................................................ x

Australian and New Zealand Standard Research Classifications (ANZSRC) ............................. x

Fields of Research (FoR) Classification ........................................................................................... x

Table of Contents .............................................................................................................................. xi

List of Figures ................................................................................................................................. xvii

List of Tables ................................................................................................................................... xix

List of Abbreviations ....................................................................................................................... xx

Chapter 1 ............................................................................................................................................ 1

Project overview .............................................................................................................................. 1

1.1 Background ............................................................................................................................ 1

xii

1.1.1 Viral capsomere as a vaccine platform ........................................................................... 2

1.1.2 Influenza hemagglutinin (HA1) as a vaccine antigen ..................................................... 4

1.2 Research objectives ................................................................................................................ 5

1.2.1 Structure-based design of modular capsomeres .............................................................. 6

1.2.2 Investigation of crudely purified modular capsomere components ................................ 6

1.2.3 Development of a simple bioprocess for modular capsomere production ...................... 7

1.3 Thesis organisation ................................................................................................................ 7

1.4 References .............................................................................................................................. 9

Chapter 2 .......................................................................................................................................... 13

Literature review ............................................................................................................................ 13

2.1 Influenza virus...................................................................................................................... 13

2.2 Influenza A virus .................................................................................................................. 13

2.2.1 Genome structure, envelope proteins and antigenic variation ...................................... 13

2.2.2 Influenza A virus life cycle ........................................................................................... 16

2.3 Avian influenza .................................................................................................................... 17

2.3.1 Ecology of avian influenza ........................................................................................... 18

2.3.2 Avian influenza outbreaks in poultry ............................................................................ 19

2.3.3 Impact of avian influenza outbreaks ............................................................................. 22

2.3.4 Avian influenza in humans ........................................................................................... 22

2.4 Strategies for control and prevention of avian influenza ..................................................... 25

2.5 AI vaccines for poultry application: manufacture and limitations ....................................... 26

xiii

2.5.1 Inactivated whole virus vaccines .................................................................................. 27

2.5.2 Live viral vector vaccines ............................................................................................. 28

2.6 Virus-like particle and capsomere as vaccine platform ....................................................... 29

2.6.1 Murine polyomavirus VP1 capsomere platform ........................................................... 29

2.7 Hemagglutinin as a vaccine target ....................................................................................... 31

2.8 Structure-based vaccine design strategy .............................................................................. 34

2.9 Linker design for modular capsomere presenting HA1 ....................................................... 35

2.10 References .......................................................................................................................... 37

Chapter 3 .......................................................................................................................................... 50

Structural-based designed modular capsomere comprising HA1 for low-cost poultry influenza

vaccination ..................................................................................................................................... 50

3.1 Abstract ................................................................................................................................ 51

3.2 Introduction .......................................................................................................................... 51

3.3 Materials and methods ......................................................................................................... 53

3.3.1 Plasmid construction ..................................................................................................... 53

3.3.2 Protein expression and purification............................................................................... 54

3.3.3 HA1 protein preparation ............................................................................................... 55

3.3.4 Capsomere characterisation .......................................................................................... 55

3.3.5 Hemagglutination assay ................................................................................................ 55

3.3.6 Immunogenicity study................................................................................................... 55

3.3.7 Enzyme-linked immunosorbent assay (ELISA) ........................................................... 56

xiv

3.3.8 Statistical analysis ......................................................................................................... 56

3.3.9 Challenge study with HPAI virus ................................................................................. 56

3.4 Results and discussion ......................................................................................................... 57

3.5 Acknowledgements .............................................................................................................. 66

3.6 Conflict of interest ............................................................................................................... 66

3.7 References ............................................................................................................................ 67

Chapter 4 .......................................................................................................................................... 71

Characterisation of modular capsomeres from purification modalities ......................................... 71

4.1 Abstract ................................................................................................................................ 72

4.2 Introduction .......................................................................................................................... 72


4.3.1 Plasmid construction and protein expression ................................................................ 74

4.3.2 Homology modelling .................................................................................................... 74

4.3.3 Preparation of highly purified CaptHA1 (cCaptHA1) .................................................. 74

4.3.4 Preparation of partially purified CaptHA1 (aCaptHA1)............................................... 74

4.3.5 Preparation of crudely purified CaptHA1 (pCaptHA1) ................................................ 75

4.3.6 Capsomere characterisation .......................................................................................... 75


4.3.8 Immunisation ................................................................................................................ 76

4.3.9 ELISA ........................................................................................................................... 76

4.3.10 Hemagglutination inhibition assay.............................................................................. 76

xv

4.4. Results and discussion ........................................................................................................ 76

4.5 Acknowledgements .............................................................................................................. 83

4.6 Conflict of interest ............................................................................................................... 83

4.7 References ............................................................................................................................ 84

Chapter 5 .......................................................................................................................................... 87

Simplified production and purification of modular capsomere-based vaccine for avian influenza

........................................................................................................................................................ 87

5.1 Abstract ................................................................................................................................ 88

5.2 Introduction .......................................................................................................................... 88


5.3.1 Plasmid construction ..................................................................................................... 90

5.3.2 Protein expression ......................................................................................................... 91

5.3.3 Chromatographically purified modular capsomere (GST-CaptHA1) ........................... 91

5.3.4 Crudely purified modular capsomere (CapVP1dC, CaptHA1 and mCap) using Na2SO4

................................................................................................................................................ 91

5.3.5 Crudely purified modular capsomere (CapVP1dC, CaptHA1 and mCap) using PEG

6000 ........................................................................................................................................ 92

5.3.6 Capsomere Characterisation ......................................................................................... 92


5.4 Results and discussion ......................................................................................................... 93

5.5 Acknowledgements ............................................................................................................ 103

5.6 Conflict of interest ............................................................................................................. 103

xvi

5.7 References .......................................................................................................................... 104

Chapter 6 ........................................................................................................................................ 107

Conclusions and future work ....................................................................................................... 107

6.1 Summary of research findings ........................................................................................... 107

6.1.1 Structure-based design of modular capsomeres .......................................................... 108

6.1.2 Characterisation of simpler purified modular capsomeres ......................................... 110

6.1.3 Simplified production and purification of modular capsomeres ................................. 112

6.2 Future work ........................................................................................................................ 113

6.3 Concluding thoughts .......................................................................................................... 115

6.4 References .......................................................................................................................... 117

Appendix A ..................................................................................................................................... 119

Animal Ethics Approval Certificate (1) ....................................................................................... 119

Appendix B ..................................................................................................................................... 121

Animal Ethics Approval Certificate (2) ....................................................................................... 121

xvii

List of Figures

Figure 1-1: Structure of murine polyomavirus VP1 protein.. .............................................................. 3

Figure 2-1: Structure of influenza A virus.. ....................................................................................... 14

Figure 2-2: Antigenic drift and antigenic shift of influenza A virus. ................................................ 16

Figure 2-3: Influenza virus replication cycle ..................................................................................... 17

Figure 2-4: The ecology of avian influenza viruses. .......................................................................... 19

Figure 2-5: Engineering VP1 capsomere. .......................................................................................... 31

Figure 2-6: Structural features and organisation of influenza virus hemagglutinin precursor, HA0. 33

Figure 2-7: Antigenic sites of HA1.. .................................................................................................. 34

Figure 3-1: Structure-based designed modular capsomere………………………………………… 58

Figure 3-2: Schematic representation of HA1 constructs and SDS-PAGE analysis of the expression

and solubility of modular capsomeres (Cap) monomer. .................................................................... 60

Figure 3-3: Characterisation of modular capsomeres by size exclusion chromatography, showing the

capsomeres separated from GST proteins and protein aggregates. ................................................... 62

Figure 3-4: Analysis of purified CaptHA1-3C .................................................................................. 63

Figure 3-5: HA1-specific antibody endpoint titre induced in chickens immunised with CaptHA1-

3C, VP1dC capsomere and HA1 protein. .......................................................................................... 64

Figure 3-6: Protection of chickens following immunisation and viral challenge. ............................. 65

Figure 4-1: Homology model of modular capsomere presenting influenza HA1(44-268) (tHA1)... 77

Figure 4-2: Major processing steps used to produce CaptHA1. ........................................................ 78

Figure 4-3: Characterisation of cCaptHA1, aCaptHA1 and pCaptHA1. ........................................... 79

xviii

Figure 4-4: Functional characterisation and homology modelling of modular capsomeres. ............. 80

Figure 4-5: Immunogenicity of modular capsomeres. ....................................................................... 82

Figure 5-1: Engineering modular capsomere presenting tHA1 (CaptHA1).......................................94

Figure 5-2: Modular capsomere protein expression and solubility.. .................................................. 95

Figure 5-3: Analysis of modular capsomere following precipitation using 1 M Na2SO4.................. 97

Figure 5-4: Optimisation of selective precipitation using Na2SO4 and PEG 6000 ............................ 98

Figure 5-5: Characterisation of precipitated CaptHA1. ..................................................................... 99

Figure 5-6: Functional characterisation of precipitated CaptHA1 ................................................... 101

Figure 5-7: Process for modular capsomere production. ................................................................. 102

xix

List of Tables

Table 2-1: Major outbreaks of HPAI H5 and H7 subtypes since 1959 ............................................. 21

Table 2-2: Laboratory-confirmed cases of human infection with avian influenza viruses. ............... 24

xx

List of Abbreviations

AI Avian Influenza

DTT Dithiothreitol

E. coli Escherichia coli

GAS Group A Streptococcus

GST Glutathione S-transferase

HA Hemagglutinin

HI Hemagglutination Inhibition

HBV Hepatitis B virus

HEV Hepatitis E virus

HPAI Highly pathogenic avian influenza

HPLC High performance liquid chromatography

HPV Human papillomavirus

IPTG Isopropyl -D-1-thiogalactopyranoside

kDa Kilodalton

LPAI Low pathogenic avian influenza

M1 Matrix protein 1

M2 Matrix protein 2

M2e Extracellular matrix protein 2

MuPyV Murine polyomavirus

xxi

MW Molecular weight

NA Neuraminidase

NEP Nuclear export protein

NP Nucleoprotein

NS1 Non-structural protein 1

NS2 Non-structural protein 2

PA Polymerase acid protein

PB1 Polymerase basic protein 1

PB2 Polymerase basic protein 2

PDB Protein data bank

PBS Phosphate buffered saline

PEG Poly (ethylene glycol)

RNA Ribonucleic acid

RNP Ribonucleoprotein

SD Standard deviation

SDS-PAGE Sodium dodecyl sulphate polyacrylamide gel electrophoresis

TEVp Tobacco etch virus protease

vRNP Viral ribonucleoprotein

VLP Virus-like particle

WHO World Health Organization

Wt Wild type

1

Chapter 1

Project overview

1.1 Background

Avian influenza (AI) has attracted international attention in recent years. Outbreaks of AI in poultry

are a major concern due to the serious consequences they hold for both livelihoods and international

trade. AI is caused by infection with avian influenza viruses which are members of the influenza A

genus. Two main forms of AI have been recognised: low pathogenic avian influenza (LPAI) and

highly pathogenic avian influenza (HPAI). Most AI viruses demonstrate low pathogenicity. However,

HPAI viruses can arise during viral circulation in poultry flocks as a result of antigenic drift and shift

[1, 2]. While LPAI in poultry is typically mild, HPAI exhibits high transmission and causes an almost

100% mortality rate [2, 3]. Whilst HPAI viruses are generally endemic in many developing countries,

in developed countries, where they are less common, outbreaks of HPAI are increasingly being

reported. Outbreaks of HPAI have been reported in the United States (US) in 2015 [4], Europe,

Middle East, Africa and Asia in 2016 [5] and Myanmar in 2017 [6]. Ongoing outbreaks of AI

worldwide, together with an increasing number of human cases with over 50% mortality for HPAI

infections [7], highlight the need for practical measures, such as disease surveillance and poultry

vaccination, to control its rapid spread.

Vaccination provides a measure for controlling influenza and preventing virus spread from animals

to humans. Poultry vaccination has been shown to increase resistance to field challenge and to reduce

viral shedding and transmission [8]. Adequate use of vaccines in conjunction with biosecurity,

education, culling, diagnostics and surveillance could result in effective eradication of AI during an

outbreak. However, the global practice of AI vaccination varies from infrequent use of vaccines in

the developed world (e.g. US and Europe), due to the uncommon occurrence of AI in these areas, to

routine use in some developing countries in Asia, Middle East, Central America and Africa where AI

viruses frequently circulate [9, 10].

Two types of vaccines have been commercialised for use in poultry: inactivated virus and

recombinant virus-vectored vaccines. Inactivated AI vaccines are the most widely used in commercial

2

poultry. These vaccines are mainly produced in embryonated chicken eggs and require at least one

egg per vaccine dose. A production time of approximately six months is a limitation of such vaccines

and is considerably slower than the potential global spread of AI, which could occur in a matter of

weeks. Furthermore, a potential AI pandemic could threaten global egg supplies, leading to a shortage

of the resources required for vaccine manufacture [11]. Inactivated vaccines also incur high

production costs due to viral inactivation, involving heat and chemicals, and the need for strong

adjuvants [12]. Recombinant viral vector vaccines, such as Newcastle disease virus-vectored vaccine

and Fowlpox virus-vectored vaccine, are produced in cell cultures [13, 14] which enable shorter

production times. It has been shown, however, that these virus-vectored vaccines may be inhibited

by maternal antibodies if poultry are vaccinated within two weeks of hatching [15]. For virus vectors

that can replicate in the host, the immune response to the AI portion of the vaccine can be supressed

by AI maternal antibody. In addition, pre-existing immunity to the virus vector can block or reduce

the subsequent immune response to the virus vector and the AI insert [16]. In addition to

commercialised vaccines, experimental second-generation AI vaccines, such as subunit and DNA

vaccines, have also been developed [17-20]. Lengthy production times make them unsuitable for use

as poultry vaccines in emergency outbreaks. Moreover, high production costs associated with these

alternative vaccines are, again, problematic for countries with low socioeconomic standing.

The limitations associated with current licensed AI vaccines underline the need for vaccines which

can be rapidly manufactured and involve low-cost processes in order to successfully respond to a

potential AI pandemic. Production of vaccines using a bacteria-based system offers a promising

alternative manufacturing process due to the speed, ease and low cost associated with recombinant

protein production in this microorganism. The overall aim of this thesis is to develop a novel vaccine

platform and simplified bioprocessing methods that may be useful in overcoming the limitations of

existing vaccine manufacturing processes.

1.1.1 Viral capsomere as a vaccine platform

A viral capsid that mimics the overall structure of a native virion, but lacks the genetic material

required for replication, is termed a virus-like particle (VLP). VLPs are safe and effective vaccine

candidates. Commercially licensed VLP vaccines comprised of unmodified capsid proteins are

available for human use against hepatitis B virus (HBV), human papillomavirus (HPV) and hepatitis

3

E virus (HEV) [21]. VLPs and their structural subunits (capsomeres) are also under development as

vaccine delivery platforms. Murine polyomavirus (MuPyV) VLPs and capsomeres based upon the

major structural protein VP1 have been successfully employed for the presentation of heterologous

antigens in vaccine production [22-24]. Heterologous expression of MuPyV VP1 in Escherichia coli

(E. coli) results in VP1 spontaneously assembling into a capsomere composed of five VP1 monomers.

This bacterially produced capsomere is a precursor of VLP formation which can be produced at gram-

per-litre levels [25] and purified from host proteins [26, 27]. Seventy-two copies of purified

capsomeres can be further assembled into a VLP in vitro under defined physiochemical conditions

[28] (Figure 1-1). However, the requirements for VLP in vitro-assembly adds additional cost to

vaccine production [29, 30] which may limit its use in low-cost markets such as the poultry industry.

Figure 1-1: Structure of murine polyomavirus VP1 protein. VP1 monomers self-assemble into a

pentameric subunit (capsomere) in E. coli, which further assemble into a virus-like particle (VLP) in

vitro. Structure of VP1 was taken from Protein Data Bank (http://www.rcsb.org) (PDB ID 1SD) and

visualised using UCSF Chimera.

Capsomeres have been shown to be less immunogenic than VLPs, presumably due to the low

complexity of their structures relative to the more complex VLPs. However, when administered with

adjuvants, capsomeres are capable of inducing antibody titres comparable to non-adjuvanted VLPs

[31]. Data obtained from process simulation showed that the unit cost of production of capsomeres is

up to 69% lower than that for VLPs. This is because of its simpler downstream process and a higher

product yield. Using capsomeres formulated with adjuvants may, therefore, achieve protective

immune responses and maintain low vaccine cost through elimination of the VLP assembly process.

The potential of capsomeres as a rapid-response and low-cost platform technology for use in vaccine

manufacturing was illustrated by a process simulation based on a conservative assumption of 50 µg

http://www.rcsb.org/

4

protein per vaccine dose [30]. Using a 10-kL fermentor, 320 million doses of modular capsomere

vaccine could be produced within three days at a cost of less than one cent per dose. Both capital

investment and operating cost were included in the simulation [30].

The VP1 capsomere platform has been developed to present small peptide epitopes, such as influenza

M2e and J8 peptide, from group A streptococcus (GAS) M-protein [24]. Small peptides incorporated

into foreign proteins often adopt structures different to their native state [22]. Consequently,

antibodies induced against this non-native structure would be unable to bind the native pathogen. For

this reason, strategic design is required for peptide epitope presentation [22]. Fusion of large antigenic

modules has several benefits over small peptide epitopes, as larger antigens generally possess more

antigenic regions and are more likely to assume their native state [32]. Presentation of the whole

protein module, such as the globular domain of influenza hemagglutinin (HA1), has been shown to

increase immunogenicity due to the presence of secondary structural elements necessary for

conformational epitope formation which are more reflective of native pathogens [33]. However,

molecular fusion of large antigens on the capsomere platform may pose challenges. Structural

separation between each protein domain is crucial to maintaining structural integrity and preventing

protein aggregation [24]. Successful presentation of large antigens on the VP1 capsomere platform

has not been shown previously. In this study, a novel VP1 capsomere platform was engineered to

successfully present a large heterologous antigen, namely the globular domain of influenza A

hemagglutinin (HA1).

1.1.2 Influenza hemagglutinin (HA1) as a vaccine antigen

Hemagglutinin (HA) is a surface glycoprotein of influenza viruses. It is responsible for receptor

binding and facilitating virus entry into target cells [34]. Antibodies raised against HA can be

neutralising, binding to the virus and blocking attachment to host cell receptors, thus preventing

infection [35]. Crystallography studies have shown that HA forms homotrimers which are embedded

in the surface of the viral envelope but part of them remains surface-exposed [36]. Each HA monomer

consists of a globular head (HA1) and a “rod-like” stalk region (HA2), linked by a single disulfide

bond. HA2 is highly conserved among different HA subtypes and is a potential target of universal

vaccines. In contrast, HA1 contains the receptor binding site and the strain-specific antigenic sites of

the HA molecule. HA1 is known to induce strong antibody-mediated immune responses [37]. In

5

particular, the vast majority of antibodies are directed against several regions of HA1 [38], suggesting

that HA1 is a promising target for vaccines. HA1 is highly insoluble when expressed in bacteria [20].

Numerous studies have reported that E. coli-expressed HA1 requires purification from inclusion

bodies, denaturing and subsequent refolding [19, 33, 39]. HA1 consists of a compact head and a long

tail which interacts with the HA2 subunit. As the majority of antigenic epitopes are present on the

head of HA1, it was hypothesised that removing the HA1 tail would not impair immunogenicity of

HA1. Further to this, it was investigated if structure-based design could be employed to produce a

suitable HA1 vaccine antigen.

1.2 Research objectives

This Ph.D. project aims to address the limitations in the manufacture of current AI vaccines for use

in poultry. Using a microbial expression system, the aim is to generate a cost-effective poultry vaccine

that is both safe and efficacious. This vaccine is designed to be produced within weeks of a new strain

being identified.

The main objectives of this research are:

(i) to evaluate the potential of a structure-based designed modular capsomeres

presenting modified HA1 as a vaccine candidate. The capsomere is based on

the VP1 of MuPyV and the modular capsomere presenting modified HA1

vaccine is the ‘chimeric’ molecule;

(ii) to evaluate the immunogenicity of E. coli-produced modular capsomeres

prepared as a partially and crudely purified form compared with highly

purified capsomeres;

(iii) to develop a simple process for manufacture of modular capsomeres presenting

HA1 for application in poultry vaccination.

The rationale behind each of these research objectives is explained in the following sections.

6

1.2.1 Structure-based design of modular capsomeres

Previously, a VP1 modular capsomere platform lacking 28 amino acids at its N-terminus and 63

amino acids at its C-terminus was engineered to display the M2e epitope of influenza virus [40]. This

vaccine induced a strong immune response which further conferred complete protection against

challenge with the homologous virus in mice [40, 41]. Despite this, the platform could not be used to

present a large antigen such as influenza HA1. Fusion of HA1 on site N of the platform resulted in

inclusion bodies post expression in E. coli (unpublished data). It was speculated that modularising

full-length HA1, which has a molecular mass more than 10 times that of M2e, on the same platform

could have disrupted the structure of the modular capsomere or caused steric hindrance between the

capsomere platform and the HA1 insert, leading to protein misfolding [42]. Previous studies have

shown that truncated HA1 containing the necessary elements for conformational epitope formation,

is sufficient to stimulate protective immune responses [35, 43]. Taken together, it was hypothesised

that rational re-design of both the capsomere platform and the HA1 antigen, based on their structures,

may result in stable and functional modular capsomeres presenting structurally-relevant

immunogenic regions of truncated HA1. Two main questions derived from this hypothesis were: 1)

is a structure-based designed capsomere platform suitable for large antigen presentation; and 2) do

antibodies generated against modular capsomeres presenting structure-based designed HA1 confer

protection in chickens? Thus, this study aims to evaluate the potential of novel modular capsomere

designs as vaccine candidates in vivo.

1.2.2 Investigation of crudely purified modular capsomere components

The main goals of animal vaccination are to improve the productivity of livestock, to improve animal

welfare and to prevent transmission of zoonotic diseases [12]. For the poultry industry, the financial

benefit of vaccination is the key consideration for implementing a vaccination program [44]. Lower

production costs of poultry vaccines can be achieved if the purification steps using chromatography

are replaced with simpler, non-chromatographic methods. Studies have demonstrated that crude

microbial extracts increase potency and enhance protective antibody responses [45, 46]. In addition,

highly specific immunity resulting from the synergistic effect of an optimal amount of contaminants

in the vaccine formulation was obtained from chickens immunised with a single dose of adjuvanted

CapM2e crudely purified by selective precipitation [47]. Nonetheless, high levels of contaminants

7

may alter immunogenicity of the modular capsomere. This section evaluated the immunogenicity of

partially and crudely purified modular capsomeres compared with that of chromatographically

purified capsomeres. The research question posed was: can modular capsomeres of low purity deliver

immunogenicity equivalent to highly purified capsomeres?

1.2.3 Development of a simple bioprocess for modular capsomere production

As reported in section 1.2.2, the costs associated with purification processes contribute to high

vaccine costs. To achieve a cost-effective poultry vaccine, it is advantageous to merge molecular

strategies with bioprocessing approaches. Simple production of murine polyomavirus modular

capsomeres and VLPs presenting a rotavirus antigen as a vaccine candidate was achieved by

integrating synthetic biological and bioprocess engineering approaches [32, 48]. The complex process

involved in producing highly purified vaccines often introduces more costs while compromising

vaccine yield [30]. The existing capsomere format, as discussed in 1.2.1, was tagged with a

Glutathione S- transferase (GST) fusion protein for expression and purification purposes.

Nevertheless, the use of a GST tag brings disadvantages, such as poor binding between the GST tag

and affinity chromatography media (approximately 35% of GST-VP1 protein was captured by GST-

affinity chromatography) [49], formation of insoluble aggregates upon GST tag removal and the use

of costly chromatography media [49]. Moreover, subsequent enzyme-mediated tag removal

introduces complexity and extra cost to the bioprocess. With the aim of developing cost-effective and

safe influenza vaccine candidates for use in poultry, the bioprocess was simplified by using non-

tagged protein expression together with selective salting-out precipitation. This new process

eliminated costly affinity purification and subsequent GST tag release by enzymatic cleavage

(Chapter 5). The research questions posed were: 1) does the simplified process increase final

capsomere yield compared with the chromatography process; and 2) do the modular capsomeres

prepared using the simplified process retain their structural integrity and functional properties?

1.3 Thesis organisation

This Ph.D. thesis consists of six chapters including this introductory chapter.

8

Chapter 2 provides a literature review of key topics involved in this research. The chapter includes a

review of a vaccine platform based on the murine polyomavirus VP1 protein, limitations of current

vaccine technologies and influenza as a target disease.

Chapter 3 evaluates the efficacy of structure-based designed modular capsomeres presenting HA1 as

vaccine candidates for poultry.

Chapter 4 evaluates the immunogenicity of low purity modular capsomeres prepared using simple

methods.

Chapter 5 evaluates the potential of a simple bioprocess developed for low-cost modular capsomere

vaccine preparation.

Chapter 6 summarises the key findings from this research and suggests future research directions.

9

1.4 References

1. Kelly, T.R., et al., A review of highly pathogenic avian influenza in birds, with an emphasis

on Asian H5N1 and recommendations for prevention and control. Journal of Avian Medicine

and Surgery, 2008. 22(1): p. 1-16.

2. Capua, I. and S. Marangon, Control and prevention of avian influenza in an evolving scenario.

Vaccine, 2007. 25(30): p. 5645-5652.

3. Capua, I. and S. Marangon, Control of avian influenza in poultry. Emerging Infectious

Diseases, 2006. 12(9): p. 1319.

4. Jhung, M.A., et al., Outbreaks of avian influenza A (H5N2), (H5N8), and (H5N1) among

birds-United States, December 2014-January 2015. Morbidity and Morbidity Weekly Report

2015. 64(4): p. 111.

5. World Health Organization/Europe, WHO calls for heightened vigilance as avian influenza

continues to spread in Europe. [News] 2017 [cited 03.07.2018]; available from:

http://www.euro.who.int/en/health-topics/communicable-

diseases/influenza/news/news/2017/01/who-calls-for-heightened-vigilance-as-avian-

influenza-continues-to-spread-in-europe.

6. Reuters. Myanmar reports outbreak of H5N1 bird flu on poultry farm: World Organization

for Animal Health, 2017 [cited 04.08.2017]; available from:

http://www.businessinsider.com/r-myanmar-reports-outbreak-of-h5n1-bird-flu-on-poultry-

farm-oie-2017-7/?r=AU&IR=T.

7. World Health Organization. Cumulative number of confirmed human cases of avian influenza

A(H5N1) reported to WHO. 2017 [cited 24.04.2017]; available from:

http://www.who.int/influenza/human_animal_interface/H5N1_cumulative_table_archives/e

n/.

8. Marano, N., C. Rupprecht, and R. Regnery, Vaccines for emerging infections. Revue

Scientifique et Technique (International Office of Epizootics), 2007. 26(1): p. 203-215.

9. Swayne, D.E., Avian influenza vaccines and therapies for poultry. Comparative Immunology,

Microbiology and Infectious Diseases, 2009. 32(4): p. 351-363.

10. Kapczynski, D.R. and D.E. Swayne, Influenza vaccines for avian species, in Vaccines for

Pandemic Influenza. 2009, Springer. p. 133-152.

11. Broadbent, A.J. and K. Subbarao, Influenza virus vaccines: lessons from the 2009 H1N1

pandemic. Current Opinion in Virology, 2011. 1(4): p. 254-262.

12. Meeusen, E.N., et al., Current status of veterinary vaccines. Clinical Microbiology Reviews,

2007. 20(3): p. 489-510.

http://www.euro.who.int/en/health-topics/communicable-diseases/influenza/news/news/2017/01/who-calls-for-heightened-vigilance-as-avian-influenza-continues-to-spread-in-europe



10

13. Boyle, D.B. and B.E. Coupar, Construction of recombinant fowlpox viruses as vectors for

poultry vaccines. Virus Research, 1988. 10(4): p. 343-356.

14. Huang, Z., et al., Recombinant Newcastle disease virus as a vaccine vector. Poultry Science,

2003. 82(6): p. 899-906.

15. Maas, R., et al., Maternal immunity against avian influenza H5N1 in chickens: limited

protection and interference with vaccine efficacy. Avian Pathology, 2011. 40(1): p. 87-92.

16. Swayne, D.E., J.R. Beck, and N. Kinney, Failure of a recombinant fowl poxvirus vaccine

containing an avian influenza hemagglutinin gene to provide consistent protection against

influenza in chickens preimmunized with a fowl pox vaccine. Avian Diseases, 2000: p. 132-

137.

17. Murugan, S., et al., Recombinant haemagglutinin protein of highly pathogenic avian influenza

A (H5N1) virus expressed in Pichia pastoris elicits a neutralizing antibody response in mice.

Journal of Virological Methods, 2013. 187(1): p. 20-25.

18. Khurana, S., et al., Bacterial HA1 vaccine against pandemic H5N1 influenza virus: evidence

of oligomerization, hemagglutination, and cross-protective immunity in ferrets. Journal of

Virology, 2011. 85(3): p. 1246-1256.

19. Aguilar-Yáñez, J.M., et al., An influenza A/H1N1/2009 hemagglutinin vaccine produced in

Escherichia coli. PloS One, 2010. 5(7): p. e11694.

20. Jegerlehner, A., et al., Bacterially Produced Recombinant Influenza Vaccines Based on Virus-

Like Particles. PloS One, 2013. 8(11): p. e78947.

21. Zhao, Q., et al., Virus-like particle-based human vaccines: quality assessment based on

structural and functional properties. Trends in Biotechnology, 2013. 31(11): p. 654-663.

22. Anggraeni, M.R., et al., Sensitivity of immune response quality to influenza helix 190 antigen

structure displayed on a modular virus-like particle. Vaccine, 2013. 31(40): p. 4428-4435.

23. Rivera-Hernandez, T., et al., Self-adjuvanting modular virus-like particles for mucosal

vaccination against group A streptococcus (GAS). Vaccine, 2013. 31(15): p. 1950-1955.

24. Middelberg, A.P., et al., A microbial platform for rapid and low-cost virus-like particle and

capsomere vaccines. Vaccine, 2011. 29(41): p. 7154-7162.

25. Liew, M.W.O., Y.P. Chuan, and A.P.J. Middelberg, High-yield and scalable cell-free

assembly of virus-like particles by dilution. Biochemical Engineering Journal, 2012. 67: p.

88-96.

26. Lipin, D.I., et al., Affinity purification of viral protein having heterogeneous quaternary

structure: Modeling the impact of soluble aggregates on chromatographic performance.

Journal of Chromatography A, 2009. 1216(30): p. 5696-5708.

11

27. Chuan, Y.P., L.H. Lua, and A.P. Middelberg, High-level expression of soluble viral structural

protein in Escherichia coli. Journal of Biotechnology, 2008. 134(1): p. 64-71.

28. Chuan, Y.P., et al., Virus assembly occurs following a pH-or Ca2+-triggered switch in the

thermodynamic attraction between structural protein capsomeres. Journal of The Royal

Society Interface, 2010. 7(44): p. 409-421.

29. Pattenden, L.K., et al., Towards the preparative and large-scale precision manufacture of

virus-like particles. Trends in Biotechnology, 2005. 23(10): p. 523-529.

30. Chuan, Y.P., et al., The economics of virus-like particle and capsomere vaccines.

Biochemical. Engineering. Journal., 2014. 90: p. 255-263.

31. Thönes, N., et al., A direct comparison of human papillomavirus type 16 L1 particles reveals

a lower immunogenicity of capsomeres than viruslike particles with respect to the induced

antibody response. Journal of Virology, 2008. 82(11): p. 5472-5485.

32. Lua, L.H.L., et al., Synthetic biology design to display an 18 kDa rotavirus large antigen on

a modular virus-like particle. Vaccine, 2015. 33(44): p. 5937-5944.

33. Khurana, S., et al., Properly folded bacterially expressed H1N1 hemagglutinin globular head

and ectodomain vaccines protect ferrets against H1N1 pandemic influenza virus. PLoS One,

2010. 5(7): p. e11548.

34. Skehel, J.J. and D.C. Wiley, Receptor binding and membrane fusion in virus entry: the

influenza hemagglutinin. Annual Review of Biochemistry, 2000. 69(1): p. 531-569.

35. Jeon, S.H. and R. Arnon, Immunization with influenza virus hemagglutinin globular region

containing the receptor-binding pocket. Viral Immunology, 2002. 15(1): p. 165-176.

36. Caton, A.J., et al., The antigenic structure of the influenza virus A/PR/8/34 hemagglutinin (H1

subtype). Cell, 1982. 31(2): p. 417-427.

37. Thomas, S. and B.A. Luxon, Vaccines based on structure-based design provide protection

against infectious diseases. Expert Review of Vaccines, 2013. 12(11): p. 1301-1311.

38. Das, K., et al., Structures of influenza A proteins and insights into antiviral drug targets.

Nature Structural & Molecular Biology, 2010. 17(5): p. 530-538.

39. Hartinger, D., et al., Enhancement of solubility in Escherichia coli and purification of an

aminotransferase from Sphingopyxis sp. MTA144 for deamination of hydrolyzed fumonisin

B(1). Microbial Cell Factories, 2010. 9: p. 62.

40. Wibowo, N., et al., Modular engineering of a microbially-produced viral capsomere vaccine

for influenza. Chemical Engineering Science, 2013. 103: p. 12-20.

12

41. Wibowo, N., et al., Protective efficacy of a bacterially produced modular capsomere

presenting M2e from influenza: Extending the potential of broadly cross-protecting epitopes.

Vaccine, 2014. 32(29): p. 3651-3655.

42. Zhao, H.L., et al., Increasing the homogeneity, stability and activity of human serum albumin

and interferon-α2b fusion protein by linker engineering. Protein Expression and Purification,

2008. 61(1): p. 73-77.

43. Xuan, C., et al., Structural vaccinology: structure-based design of influenza A virus

hemagglutinin subtype-specific subunit vaccines. Protein & Cell, 2011. 2(12): p. 997-1005.

44. Heldens, J., et al., Veterinary vaccine development from an industrial perspective. The

Veterinary Journal, 2008. 178(1): p. 7-20.

45. Muniandy, N. and T. Mukkur. Comparative Efficacy of Whole Cell and Crude Extract

Vaccines of Pasteurella multocida Type 6: B in Mice. in The Seventh Veterinary Association

Malysia Scientific Congress. 1995. Seremban.

46. Rodríguez-Limas, W.A., et al., Immunogenicity and protective efficacy of yeast extracts

containing rotavirus-like particles: A potential veterinary vaccine. Vaccine, 2014. 32(24): p.

2794-2798

47. Wibowo, N., et al., Non-chromatographic preparation of a bacterially produced single-shot

modular virus-like particle capsomere vaccine for avian influenza. Vaccine, 2015. 33(44): p.

5960-5965.

48. Tekewe, A., et al., Integrated molecular and bioprocess engineering for bacterially produced

immunogenic modular virus‐like particle vaccine displaying 18 kDa rotavirus antigen.

Biotechnology and Bioengineering, 2017. 114(2): p. 397-406.

49. Lipin, D.I., L.H. Lua, and A.P. Middelberg, Quaternary size distribution of soluble

aggregates of glutathione-S-transferase-purified viral protein as determined by asymmetrical

flow field flow fractionation and dynamic light scattering. Journal of Chromatography A,

2008. 1190(1): p. 204-214.

13

Chapter 2

Literature review

2.1 Influenza virus

Influenza virus is a highly contagious RNA virus of the family Orthomyxoviridae. Based upon

antigenic differences in their nucleoprotein (NP) and matrix protein 1 (M1), these viruses are

classified into four types: Influenza A, B, C and D [1]. Influenza B and C viruses commonly infect

humans, although both viruses have been isolated from other mammalian species [2-4]. Influenza B

viruses exhibit low mutation rates and low genetic diversity, which seemingly restricts their

infectivity to epidemic proportions only. Influenza C viruses typically cause only mild disease in

children and are not considered to be of epidemic or pandemic concern [5]. Influenza D viruses

primarily infect cattle and are not known to cause illness in humans [6]. In contrast, most influenza

A viruses are extremely pathogenic and are capable of infecting a broad host range, for example birds

and numerous mammals, including humans [7]. These highly mutagenic viruses are considered to

present a significant global threat to human and animal health.

2.2 Influenza A virus

2.2.1 Genome structure, envelope proteins and antigenic variation

The roughly spherical influenza A virion is composed of an outer lipid envelope, an inner layer and

a viral core containing the genome (Figure 2-1). The lipid envelope contains three transmembrane

proteins, glycoproteins hemagglutinin (HA) and neuraminidase (NA), which protrude from the viral

surface, and the matrix protein 2 ion channel (M2). The inner layer resides underneath this lipid

envelope and contains the most abundant protein in the virion, the matrix protein 1 (M1). The viral

core is located beneath the inner layer and contains eight single-stranded negative-sense RNA

molecules. RNA segments 1-6 encode polymerase basic protein 2 (PB2), polymerase basic protein 1

(PB1), polymerase acidic protein (PA), HA, nucleoprotein (NP), and NA, respectively [8]. RNA

segment 7 encodes matrix protein 1 and 2 (M1 and M2). RNA segment 8 encodes non-structural

protein 1 (NS1) and non-structural protein 2 (NS2; also known as nuclear export protein, NEP) [8].

14

All RNA segments are encapsidated by multiple NP molecules and associated with PB1, PB2 and PA

to form the viral ribonucleoproteins (vRNP) complex. PB1, PB2 and PA form an RNA-dependent

RNA polymerase complex, which is responsible for transcription and replication of viral RNA [9].

Figure 2-1: Structure of influenza A virus. The viral particle is composed of an outer lipid envelope,

an inner layer and a viral core containing the viral genome. The genome consists of eight single-

stranded RNA molecules. RNA segments 1-6 encode polymerase basic protein 2 (PB2), polymerase

basic protein 1 (PB1), polymerase acidic protein (PA), hemagglutinin (HA), nucleoprotein (NP), and

neuraminidase (NA), respectively. RNA segment 7 encodes matrix protein 1 and 2 (M1 and M2). RNA

segment 8 encodes non-structural protein (NS). The outer envelope contains HA and NA, which are

the major surface glycoproteins of the virus, and M2, which is an integral membrane protein and

functions as an ion channel. The inner layer contains M1, which is the most abundant protein in the

virion. Adapted from [9].

The envelope glycoproteins, HA and NA, have important functional roles and are also used to

determine virus subtype. HA is the most abundant of these envelope proteins, constituting

approximately 80% (by mass) of the outer lipid envelope. NA accounts for a further 17% [10]. HA is

responsible for binding to sialic acid receptors on the surface of host cells and mediating membrane

fusion during the initial stage of infection. NA plays a major role during virus replication. It facilitates

the release of the new viruses from infected cells by cleaving sialic acid from the HA of viral progeny

that is attached to the host cell surface via HA-sialic acid interaction [2, 7]. Both HA and NA are

targets of protective immune responses. To date, 18 subtypes of HA and 11 subtypes of NA have

15

been identified [11, 12]. They are highly variable in sequence, with less than 50% and 30% conserved

amino acids of HAs and NAs, respectively, among all subtypes [11, 12].

Influenza A viruses are constantly evolving owing to two main types of variations: antigenic drift and

antigenic shift. Antigenic drift refers to changes in HA and/or NA surface proteins caused by the

accumulation of point mutations (substitutions, deletions and insertions) in the HA and NA genes due

to the inaccuracy of the RNA polymerases to insert the correct nucleotide and the lack of proofreading

function of the RNA polymerase (Figure 2-2A). These point mutations often encode amino acid

changes in the HA and/or NA, permitting the virus to escape neutralisation by antibodies raised

against the previous strain [13]. Antigenic drift may or may not affect the virulence of the virus with

the potential to cause influenza epidemics [14]. Antigenic shift is a major change in the virus, resulting

from reassortment of viral segmented RNA when cells are co-infected with two different parent

viruses (Figure 2-2B). Reassortment of RNA segments is an important mechanism for ensuring

antigenic variation in influenza A viruses. It occurs randomly, allowing the new generation of viruses

to contain novel combinations of genes. Reassortment of gene segments between different subtypes

of influenza A viruses (for example, between avian and human viruses, or avian and other animal

viruses) has previously led to new subtypes which have caused influenza outbreaks [15]. Indeed, the

viruses that caused the 1957 and 1968 pandemics resulted from reassortment between avian and

human influenza A viruses, generating new subtypes H2N2 and H3N2 [16, 17]. The recent influenza

A (H1N1) 2009 pandemic virus also resulted from the reassortment of several viruses, including

swine and avian-like swine viruses [15, 18].

16

Figure 2-2: Antigenic drift and antigenic shift of influenza A virus. (A) Antigenic drift results in

the production of a new HA or NA resulting from point mutations in existing HA or NA genes of the

influenza virus. (B) Antigenic shift is the reassortment of genes from genetically distinct viruses

during co-infection of the same host, resulting in a new subtype which is a mix of the parental viruses’

surface antigens. Adapted from https://courses.lumenlearning.com/microbiology/chapter/virulence-

factors-of-bacterial-and-viral-pathogens/.

2.2.2 Influenza A virus life cycle

Like all viruses, replication of the influenza A virus relies upon the use of host cellular machinery.

The life cycle of the virus involves several distinct stages, including virus attachment and entry into

host cells, replication of the viral genome and the subsequent budding of newly produced viral

particles [10, 19] (Figure 2-3). The initial attachment of virions to host cells is mediated through the

binding of HA to sialic residues on the host cell surface [20]. Upon binding, the virus enters the cell

by receptor-mediated endocytosis, a process by which the cell membrane invaginates and engulfs the

virus forming a vesicle known as the endosome [20]. The endosome harbours an acidic environment

(pH 5-6) which induces a conformational change within the HA proteins [21]. This results in fusion

of the viral and endosomal membranes. The low pH of the endosome mediates the opening of the M2

ion channels, enabling acidification of the virion itself [22]. This, in turn, releases the vRNP complex

from the M1 matrix protein, permitting their entry into the host cell cytoplasm and their migration to

the nucleus where transcription and replication occurs [23]. Newly synthesised viral proteins and

genetic material are produced in the nucleus and then transported to the host cell membrane where

the assembly of new viral particles takes place [23]. HA, NA and M2 localise at the site of budding

on the surface of the host cell membrane, while M1 assembles just beneath it. The vRNP complex is

packaged into the virions on the inner surface of the cell membrane. Packaged virions then undergo

17

a budding process by which HA and NA are incorporated into the host membrane. In the final stage

of the process, NA cleaves sialic acid residues from the HA glycoproteins in the new viral particles,

allowing their release from the host cell membrane [24].

Figure 2-3: Influenza virus replication cycle. The virus enters a host cell by endocytosis after

attachment of HA to the host receptor. The acidic environment within the endosome allows uncoating

of the virus particle. Viral RNA leaves the endosome and enters the cell nucleus where transcription

takes place. The viral proteins are translated in the cytoplasm. The viral proteins and viral RNA

translocate to the inner surface of the host cell membrane where the new virion’s components are

assembled. This is followed by budding of assembled parts from the host, taking with them part of the

host cell membrane containing the internal viral proteins. Adapted from

http://www.vetsoft.com.eg/article/ArticleData.aspx?ID=264.

2.3 Avian influenza

Avian influenza (AI), also known as avian flu, bird flu or fowl plague, is a respiratory disease of birds

caused by influenza A viruses [25]. AI viruses are classified as either highly pathogenic avian

influenza (HPAI) or low pathogenic avian influenza (LPAI) based on their molecular characteristics

and the ability to cause severe disease in inoculated young chickens [26]. HPAI viruses have an

intravenous pathogenicity index (IVPI) in 6-week-old chickens greater than 1.2 or cause at least 75%

mortality in 4- to 8-week old chickens infected intravenously [27, 28]. Poultry infected with LPAI

18

viruses typically show only mild symptoms which may often go unnoticed [29]. In contrast, HPAI

viruses cause severe infections that affect multiple internal organs and result in almost 100%

mortality, generally within 48 hours of infection [29, 30]. HPAI viruses are caused mainly by the H5

and H7 subtypes of influenza A virus [31]. Although the majority of AI viruses are of low

pathogenicity, HPAI viruses can arise from LPAI viruses during circulation in poultry flocks due to

their mutation-prone nature. A key determinant of AI virus pathogenicity is the HA glycoprotein [32].

HA is synthesised as a precursor molecule (HA0) which is cleaved into subunits, HA1 and HA2, by

host proteases [33]. This cleavage is required for viral infectivity [34]. The HA0 of LPAI viruses is

cleaved at a conserved arginine residue by trypsin-like proteases [35]. Replication of these viruses

can only occur where such enzymes are present, i.e. respiratory and intestinal tracts [8]. In contrast,

the HA0 of HPAI viruses have multiple basic amino acids within their cleavage sites, which permit

cleavage by ubiquitous intracellular proteases. Therefore, HPAI viruses are able to replicate in

extrapulmonary sites throughout the body, causing fatal systemic infections [8, 35].

The binding of HA glycoproteins present on the surface of influenza virus is primarily influenced by

the type of linkage that exists between sialic acid and galactose molecules present on host cells.

Human influenza viruses preferentially bind to sialic acids that are linked to galactose with α-2,6

linkages [36], i.e. those that are present in the human respiratory tract. Most avian influenza viruses

preferentially bind to sialic acids with α-2,3 linkages to galactose, i.e. those present in birds [36]. For

this reason, avian influenza viruses are largely confined to infecting birds. Swine influenza viruses,

however, recognise both α-2,6 and α-2,3 linkages [36]. The presence of α-2,6 and α-2,3 linkages in

the pig trachea explains why swine are believed to play a role in the adaptation of avian viruses to

human viruses, acting as mixing vessels where intra-subtype reassortment events could occur which

generate new pathogenic viruses [37].

2.3.1 Ecology of avian influenza

The ecological cycle of AI viruses can be divided into two cycles, a natural avian influenza cycle and

an animal and human influenza cycle (Figure 2-4). In the natural cycle, wild birds, such as shorebirds

and waterfowls, are considered the natural hosts for diverse populations of AI viruses [38]. AI viruses

circulating in these hosts serve as parent strains for HPAI and LPAI H5 and H7 subtypes [39].

Infections with influenza A viruses in these birds are mostly asymptomatic and are restricted to the

19

gastrointestinal and/or the respiratory tract of the infected birds [13]. This may be due to these sites

offering the optimal temperature and pH for replication of the HPAI viruses [40]. AI viruses are

spread through faecal contamination of water habitats [13, 41] and can be transmitted to domestic

birds via contamination of water on farms along the flyways of migratory birds [13] or as a result of

domestic birds intermingling with backyard poultry [25]. However, contact between wild birds and

farmed poultry is unlikely as wild birds have little opportunity to enter fully enclosed poultry houses

[41]. Transmission of AI viruses to humans possibly occurs through direct contact with infected birds

[25], the transfer of viruses via contaminated clothing or through the transportation of susceptible

birds [41]. Inter-species transmission of AI viruses to non-poultry hosts can also occur via direct

contact with swine which are susceptible to both avian and human influenza viruses, as previously

described [42].

Figure 2-4: The ecology of avian influenza viruses. Avian influenza viruses are transmitted from

their natural hosts to domestic birds. AI viruses can also be transmitted to humans directly via

domestic birds or swine, which are susceptible to both avian and human influenza viruses. Adapted

from [43].

2.3.2 Avian influenza outbreaks in poultry

Avian influenza was first described as “fowl plague” in 1878 by Perroncito in Italy [44], but the virus

was not isolated until 1959 when the first outbreak of HPAI H5 subtype occurred in Scotland [45].

Prior to 1997, the H5N1 strain of AI viruses is considered to have begun circulating in poultry

20

populations throughout parts of Asia [13]. Similar to other AI viruses of the H5 and H7 subtypes, the

H5N1 virus initially caused only mild disease with symptoms such as ruffled feathers and reduced

egg production, allowing the virus to evade detection [46]. In 1997, following months of circulation

in chickens, a highly pathogenic form of the virus emerged, causing almost 100% mortality in flocks

and resulting in the first H5N1 outbreak in Hong Kong Special Administrative Region (SAR) [47].

At the end of 2003, the H5N1 HPAI virus became more widespread, affecting several Asian countries

including South Korea, Vietnam, Japan, Thailand, Cambodia, Laos, Indonesia and China [48]. This

was the world’s largest HPAI outbreak in over four decades, affecting more than 120 million birds

and causing large-scale consequences for both the commercial poultry industry as well as rural

subsistence farming [46, 48]. Large-scale control efforts, including prompt culling of infected and

exposed birds, proper disposal of carcasses and quarantine and rigorous disinfection of farms, were

implemented which resulted in rapid control of the virus outbreak [46]. However, these massive

control efforts did not eliminate the virus. More likely, H5N1 was merely quiescent or possibly still

active in rural areas where deaths in small backyard flocks were likely to avoid detection. In mid-

2004, further outbreaks of H5N1 were reported in Cambodia, China, Indonesia, Vietnam and

Malaysia, a country spared in the first outbreak [46]. These outbreaks were much smaller, affecting

less than one million poultry. Since then, evidence strongly indicates that H5N1 is endemic in parts

of Asia [49, 50] and AI outbreaks of H5 and H7 subtypes have been reported more frequently [51].

In December 2014, an HPAI outbreak occurred in the US. The US Department of Agriculture’s

Animal and Plant Health Inspection Service reported an outbreak of HPAI H5N1, H5N2 and H5N8

that resulted in substantial mortality in domestic poultry [52, 53]. The viruses were first detected in

commercial flocks in Iowa and Minnesota and quickly spread to 15 states by June 2015 [52]. This

2014-2015 outbreak is the worst US avian influenza outbreak to date, affecting more than 48 million

domestic flocks and wild birds [54]. In addition, outbreak of HPAI H5N8 were reported in poultry

farms in Asia and Europe in 2014 [55, 56]. Since then, outbreaks have continually been reported in

countries throughout Europe, the Middle East and Asia [57]. The latest outbreaks of H5N1 and H5N6

were reported in Myanmar and Philippines in 2017 [57]. Table 2-1 lists major outbreaks of HPAI H5

and H7 subtypes occurring in the period 1959 to 2017.

21

Table 2-1: Major outbreaks of HPAI H5 and H7 subtypes since 1959. Adapted from [23, 47].

Virus subtype Year Country or continent Number of poultry affected References

H5N1 1959 Scotland 1 small farm [45]

1991 England 8000 [59]

1997 Hong Kong 1.5 million [60]

2002 Hong Kong 800,000 [61]

2003-07 Asia, Europe and Africa 100s of millions [62]

2014-15 US > 48 million* [53]

2017 Myanmar > 20,000 [41]

H5N2 1983-84 US 17 million [63]

1994 Mexico millions [64]

1997 Italy 8,000 [65]

2014-15 US > 48 million* [47]

H5N6 2017 Philippines > 26 million [57]

H5N8 1983 Ireland 307,000 [66]

2014-15 US > 48 million* [53]

2016 Europe and Middle East > 100,000 [56, 57]

H7N3 1963 England 29,000 [68]

1992 Australia 18,000 [69]

1995 Pakistan > 3 million [70]

2002 Chile 700,000 [71]

H7N4 1997 Australia >160,000 [72]

H7N7 1976 Australia 58,000 [73]

1985 Australia 240,000 [74]

2003 The Netherlands > 33 million [44, 75]

2012 Australia 50,000 [76]

* Total number of poultry affected by HPAI H5N1, H5N2 and H5N8 outbreaks

22

2.3.3 Impact of avian influenza outbreaks

AI outbreaks can result in catastrophic economic decline, primarily due to the direct loss of birds

from AI or from measures designed to control the spread of the virus, i.e. the destruction and disposal

of domestic birds [77]. This was evident in the 2003-2004 HPAI outbreaks which resulted in the death

and destruction of more than 200 million birds worldwide, causing economic losses of over US$20

billion [78]. Direct losses from turkeys and egg-laying hens euthanised during the 2014-2015 HPAI

outbreak in the US are estimated to be nearly US$1.6 billion. This estimated cost does not include

activities such as cleaning-up and restocking [52]. Based on US animal health policies, when an AI

virus (HPAI or LPAI) is detected in the US, elimination of the virus is required to be performed

through culling and disposal of affected poultry and cleaning and disinfecting premises and

equipment. This is followed by testing to demonstrate that the virus has been eliminated, which is

required before farms can be repopulated. This process is labour-intensive and time-consuming.

Poultry producers stand to lose a significant income, not only through the loss of birds, but also the

loss of usable laying facilities which are out of production for several months [52].

As poultry farming is an intricate business with multiple upstream and downstream stakeholders, the

losses from outbreaks affect, not only the poultry farmers themselves, but also connected businesses,

such as feed suppliers, equipment sales, slaughterhouse services, processors, transport services and

retailers [54]. Moreover, there are consequential effects on their employees, such as the loss of

household income and potential job loss [54]. Consumer responses to AI outbreaks are often

immediate and dramatic, resulting in additional economic losses through reduced demand for poultry

products in affected countries [79]. The detection of H5N1 in either wild or domestic birds in Asia,

Europe and Africa has resulted in sharp declines in sales, prices and consumption [80]. The effects

have also been observed at an international level. As a consequence of the US outbreak of HPAI in

December 2014, 56 trade partners imposed full or partial bans on shipments of US poultry meat and

products. This resulted in a 13% decline in shipments to China, the European Union, Mexico, South

Korea and Taiwan [81].

2.3.4 Avian influenza in humans

AI is a zoonotic disease with demonstrated ability to mutate and infect humans. Thus, in addition to

the economic impact, the presence of HPAI outbreaks raises concerns about the emergence of new

23

viruses that could potentially cause influenza epidemics in humans. AI viruses rarely infect humans,

although sporadic human infections with AI viruses (subtypes H5, H7 and H9) have been documented

(Table 2-2) [26]. Symptoms of AI virus infections in humans vary from mild conjunctivitis and

influenza-like illness to severe respiratory illness and multi-organ failure leading to mortality [82].

To date, HPAI H5N1 and H7N9 viruses have proven to be the most virulent in humans, causing

severe illness and high mortality [82]. Transmission of avian influenza viruses to humans was

believed to be rare due to a limited host range and the possible requirement of reassortment between

human and avian influenza viruses [41]. However, direct transmission of avian influenza virus from

domestic poultry to humans was indicated during an outbreak of the HPAI H5N1 virus in poultry in

1997 in the SAR, Hong Kong [83]. This highlights the potential threat of HPAI to humans.

24

Table 2-2: Laboratory-confirmed cases of human infection with avian influenza viruses. Adapted

from [8, 84].

Virus

subtype Year Country

Number of

cases (deaths) References

H5N1 1997 Hong Kong 18 (6) [85]

2003-09

Azerbaijan, Bangladesh,

Cambodia, China, Djibouti,

Egypt, Indonesia, Iraq, Laos,

Myanmar, Nigeria, Pakistan,

Thailand, Turkey, Vietnam,

Canada

468 (282) [86]

2010-14 China, Egypt, Indonesia,

Vietnam, Bangladesh

233 (125) [86]

2015 Indonesia 145 (42) [86]

2016-17 Egypt 13 (4) [86]

H7N9 2013-17 China 1557 (623) [87]

H7N7 1996 UK 1 [88]

2003 The Netherlands 89 (1) [89]

H7N3 2002-03 Italy 7 [90]

2004 Canada 2 [91]

2006 UK 1 [92]

H7N2 2002-03 US 2 [93]

2007 UK 4 [84]

H9N2 1999 Hong Kong, China 7 [94]

2003 Hong Kong 1 [95]

H10N8 2013 China 1 [96]

H10N7 2004 Egypt 2 [8]

25

2.4 Strategies for control and prevention of avian influenza

Effective control of AI requires a combination of measures including biosecurity, education,

diagnostics, surveillance and depopulation [58, 97]. Biosecurity refers to measures that reduce the

risk of AI virus introduction and spread of infection. It represents the first and foremost means of

prevention. Biosecurity covers inclusion biosecurity (quarantine) to keep the virus within infected

premises and exclusion biosecurity to keep the virus out of virus-free premises and the surrounding

environment. Biosecurity measures include cleaning and disinfecting to remove the contaminating

virus and enforcing restrictions on the movement of poultry and poultry-associated items [98].

Education of all poultry- and allied-industry staff (i.e. owners, suppliers and wholesalers) is a critical

aspect of the control of AI. Education on how AI is introduced and spread and how such events can

be prevented, as well as raising awareness of individual behaviour and controlling fomite or aerosol

movement of the virus, can greatly reduce the spread of AI viruses [99]. Accurate and rapid diagnosis

of AI is a prerequisite to early and successful control. The speed at which AI is controlled is greatly

dependent upon how quickly the first case is detected, the existing biosecurity and how quickly

control strategies are implemented. Surveillance is crucial for ongoing evaluation of the success of

control strategies and for use in decision making to certify an AI-free area, or during an AI outbreak

to determine the area of infection for quarantine purposes [100]. During HPAI outbreaks,

depopulation or stamping-out policies of euthanising infected and at-risk flocks has been a common

and effective control measure in countries where the virus has been newly introduced. However, this

control strategy usually comes at a high social and economic cost and is not the desired method from

a disease control and animal welfare perspective [101, 102]. The implementation and maintenance of

high standard control measures at the working farm level (industrial, semi-industrial and backyard

farms) can prevent infection penetration and perpetuation in farm systems [101]. However,

implementation of control strategies differs between small backyard bird flocks and industrially

raised poultry and contraventions in the biosecurity system may occur [101]. This indicates the need

to establish additional control tools for AI.

Vaccination represents an additional level of defense against AI. It can be a powerful tool for

controlling AI when used in conjunction with other control methods. The scientific basis for using a

vaccination strategy is the induction of protective immunity in the target population. An effective

vaccination program would raise the level of protective immunity in poultry flocks. Poultry

26

vaccination has been shown to increase the resistance to influenza infection in the field, reduce viral

shedding levels (in terms of amount and duration) in vaccinated birds and reduce transmission [103].

The demonstrated effectiveness of poultry vaccination indicates its potential for successfully

controlling AI. To ensure optimal outcomes, vaccination programs must be part of a wider control

strategy that includes biosecurity and infection monitoring. Vaccination could prevent the

introduction of AI or reduce its spread, thus minimising the negative impact on poultry production

and reducing potential economic losses. Furthermore, poultry vaccination can also minimise the risk

of human exposure to AI viruses, reducing the number of humans affected by the virus and reducing

the likelihood of a human influenza pandemic.

Poultry vaccines are commonly used in control and prevention programs against H5N1 HPAI, in

countries where HPAI viruses are endemic (i.e. Bangladesh, China, India, Indonesia, Vietnam and

Egypt) [104]. All HPAI and LPAI viruses of H5 and H7 subtypes are reportable to the World

Organization for Animal Health (OIE) [102], often resulting in trade sanctions being imposed on the

reporting country. In the event of an H5 and H7 LPAI outbreak, controlled use of AI vaccines may

reduce virus shedding from vaccinated and then exposed birds and increase resistance to infection in

vaccinated birds [29]. However, introduction of vaccines against H5 and H7 avian influenza viruses

in commercialised poultry is likely to lead to trade restrictions and bans due to the presence of H5 or

H7 antibodies in both the vaccinated and the naturally infected birds. The DIVA (differentiate

infected from vaccinated animals) strategy has been designed to distinguish between vaccinated birds

and birds infected with the native virus [105]. This is achieved through the administration of vaccines

based on different strains (e.g. H5N3) than the current field strain (e.g. H5N1) and using a serological

test, such as ELISA, to detect markers that can differentiate between vaccine-induced antibodies and

antibodies against the field virus [105]. The DIVA principle has been proposed as a potential tool in

assuring the safe practice of moving poultry and poultry products even when vaccination is used for

the control of AI.

2.5 AI vaccines for poultry application: manufacture and limitations

Vaccine have been developed using various technologies in order to provide protection from LPAI

and HPAI viral infections, with experimental studies demonstrating their protective efficacy in

poultry [100]. Transgenic yeast vaccines [106] and DNA-based vaccines [107] are novel approaches

27

which have demonstrated protection, however, only two types of vaccines (inactivated vaccines and

viral vector vaccines) are currently available for commercial use [102].

2.5.1 Inactivated whole virus vaccines

The majority of currently licensed AI vaccines are prepared by conventional means. This involves

virus propagation in 9 to 11-day-old embryonated chicken eggs from specific pathogen-free (SPF)

flocks. The infected allantoic fluids containing culture-derived virions are then harvested, chemically

inactivated (using formalin, -propiolactone or binary ethylenimine) and subsequently formulated

with mineral oil emulsion [108].

Inactivated vaccines were previously prepared using strains of LPAI viruses isolated from outbreaks

in poultry or from surveillance in wild or domestic birds [109]. These conventionally inactivated

vaccines, aimed at H5, H7 and H9 subtypes, are commercially available and licensed for use in several

countries around the world [110]. Despite this, the manufacture of influenza vaccines remains a

challenge owing to the antigenic variation typical of the influenza virus which causes mismatching

between vaccine and field strains [111]. Reverse genetics technology represents a substantial

improvement in generating inactivated vaccines. This technology enables the modification of a HPAI

virus to genetically match a LPAI field virus by altering the HA0 cleavage site of the HPAI virus to

that of the LPAI virus. Modified HA and NA genes are inserted in a vaccine virus backbone. The six

internal genes that encode for virus internal proteins can be derived from an existing influenza A virus

vaccine strain that is known to produce high titre in eggs. This allows the production of a vaccine

seed strain that genetically matches a circulating virus but is safe and has a high growth rate in eggs.

For example, this technology is used to generate low pathogenicity H9N2/PR8 reassortant viruses

that derive their HA and NA genes from an epidemic strain (Korean-isolated H9N2 viruses) and six

internal genes from the A/Puerto Rico/8/34 egg-adapted, high-growth donor strain (PR8). These

reassortant viruses have a higher growth rate than the parental viruses [112]. However, the higher

growth rate may not be observed in all cases reported and not only the six internal genes but also the

HA genes are determinants of growth rate in eggs [113].

Despite their efficacy, egg-based vaccines have limitations relating to production time and egg

supply. Egg-based vaccine production requires one to two eggs per vaccine dose and it takes

approximately six months to manufacture and distribute the vaccine. This is a significant limitation

28

since the global spread of the AI virus can occur within one month [114, 115]. Moreover, AI outbreaks

causing high mortality in poultry flocks can lead to a drastically limited egg supply and impose

substantial limitations on vaccine manufacture during an emergency event, such as an influenza

pandemic [116]. In addition, maternal immunity, conferred via egg yolk to the embryo, can interfere

with the immune response generated by these vaccines [117].

2.5.2 Live viral vector vaccines

Live vector virus is an alternative strategy for creating AI vaccines. Specific genes from avian

influenza viruses, namely those which encode proteins that generate a protective immune response

such as the HA and NA genes, are inserted into a live virus vaccine vector. This results in expression

of the HA and NA proteins in the vaccinated chickens upon administration. Live viral vector vaccines

can stimulate both humoral and cellular immune responses to both the inserted and the vector virus

[118]. Live recombinant fowl poxvirus and Newcastle disease virus vaccines with an AI H5 gene

inserted (rFP-AIV-H5 and rNDV-AIV-H5, respectively) have been licensed and used in several

countries [110].

The rFP-AIV-H5 is the most common vectored technology used in AI vaccine manufacture and has

been licensed in the US since 1988 and is also licensed in Mexico, Guatemala, El Salvador and

Vietnam [119]. It can be administered to one-day-old chickens by subcutaneous injection or wing

web puncture. This strategy is compatible with the DIVA concept as this vaccine can be used to

distinguish between vaccinated birds and naturally infected birds because the vaccinated birds do not

develop antibodies against nucleoprotein or matrix protein [109]. However, rFP-AIV-H5 vaccine is

not effective in chickens that have previously received a poxvirus vaccine or have been infected by a

field strain of fowl poxvirus [119].

The rNDV-AIV-H5 vaccine has been shown to protect against both HPAI and virulent ND viruses.

This vaccine has been licensed and heavily used in China in recent years [120]. It has advantages

over the rFP-AIV-H5 vaccine in that it can be administered by aerosol spray or eye drop and, thus, is

less time-consuming and labour-intensive [109]. However, similar to inactivated vaccines, maternal

antibodies to the NDV vector or AI virus may limit vaccine replication, resulting in a suboptimal

protective immune response [109].

29

2.6 Virus-like particle and capsomere as vaccine platform

A virus-like particle (VLP) is a viral capsid that resembles a native virion but lacks genetic material,

making it non-infectious. The natural ability of VLPs to induce strong humoral and cellular immune

responses independent of adjuvant make them potential candidates for vaccines. Many studies have

proven the effectiveness of VLPs as vaccine candidates and demonstrated their safety, due to their

inability to cause disease. Commercial VLP vaccines for human use are licensed for protection against

hepatitis B virus (HBV), human papillomavirus (HPV) and Hepatitis E virus (HEV) [121]. VLPs vary

in their structural complexity from single protein to multi-protein capsids complete with lipid

envelopes [122]. Less complex VLPs which lack a lipid envelope can be produced using prokaryotic

systems, such as E. coli [123, 124]. In certain instances, viral capsid proteins have been engineered

to limit their assembly to precursor capsomeric subunits, which can be further assembled into VLPs

in vitro under controlled and optimal physicochemical conditions [125].

Capsomeres present the possibility of manufacturing low-cost and low complexity vaccines. Upon

expression in E. coli, viral capsid proteins spontaneously assemble into capsomere structures [124].

Studies have shown that capsomeres themselves are immunogenic and can elicit a protective immune

response, both as a vaccine against the parental pathogen [126, 127] and as a platform to present

heterologous antigens of unrelated pathogens [128, 129]. Capsomeres formulated with adjuvants

induce strong immune responses comparable to those of VLPs [130, 131]. In contrast to VLP

production which is complex and challenging, production of capsomere-based vaccines eliminates

complex in vitro assembly processes which contribute significantly to vaccine production costs.

2.6.1 Murine polyomavirus VP1 capsomere platform

Murine polyomavirus (MuPyV) is a member of the Polyomaviridae family [132]. It is non-enveloped

and contains a circular double-stranded DNA genome. The viral capsid is comprised of 360 identical

copies of the major capsid protein (VP1) which are arranged as 72 morphological subunits

(pentamers) to create an icosahedral capsid. The viral genome is approximately 50 nm in diameter

and encodes three structural proteins, the major structural protein, VP1, and two minor proteins, VP2

and VP3 [133]. VP1 is a 384 amino acid protein with a molecular weight of approximately 42 kDa

which forms the viral capsid [133]. VP2 and VP3 are not required for viral capsid formation, nor for

stability [134]. Expression of VP1 in E. coli results in a pentameric structure composed of five

30

identical copies of VP1. This pentameric structure is referred to as the VP1 capsomere, approximately

8 nm in diameter and has a barrel-shaped morphology [135, 136].

The potential of MuPyV VP1 capsomere as a platform to present foreign antigens has previously

been demonstrated [137]. Studies have shown that 63 amino acids at the C-terminal are crucial for

VLP assembly and, thus, removal of the C-terminal of VP1 inhibits VLP formation and restricts

assembly to capsomeres. This modified capsomere, designated VP163 (VP1dC), serves as a

platform for insertion of an antigenic sequence. VP1 was previously engineered to contain two

insertion sites at surface-exposed loops, S1 and S4, at amino acids 85 and 293 of wild-type VP1

(wtVP1), respectively (Figure 2-5A). These insertion sites are also present in VP1dC (Figure 2-5B).

Molecular fusion of a single module of extracellular matrix protein 2 (M2e) antigen from influenza

virus into the surface-exposed loop (S1) of VP1dC resulted in five copies of M2e per capsomere and

yielded stable modular capsomeres post expression in E. coli [137]. These modular capsomeres were

incapable of forming VLPs in vitro, even in the presence of Ca2+ which has been shown to drive VLP

formation [125, 138]. A study demonstrated that immunogenicity of the M2e antigen was enhanced

when multiple copies of the M2e antigen were modularised into a single capsomere [139]. VP1dC

was re-designed to contain four insertion sites for multiple antigenic modules insertions. The first 28

amino acids were removed from the N-terminus of VP1dC and then two insertion sites were

engineered at either N- or C-termini. This resulted in a capsomere with four insertion sites, designated

VP1NC (VP1dNdC) [139] (Figure 2-5C). Modularisation of the M2e antigen by insertion of one

each of the M2e sequence into three different insertion sites of VP1dNdC resulted in a modular

capsomere comprising a total of 15 M2e antigenic elements (CapM2e) [140]. Mice immunised with

CapM2e, formulated with adjuvant, developed high antigen-specific antibody levels, which further

conferred protection against homologous live virus infection [140]. This modified capsomere served

as carrier platform to present small peptide antigens, which are easier to modularise than large protein

domains and whole proteins.

31

Figure 2-5: Engineering VP1 capsomere. (A) Wild-type VP1 (wtVP1) contains surface-exposed

loops S1 (amino acid 85) and S4 (amino acid 293) and N- and C-termini. Two insertion sites were

engineered at loops S1 and S4 of wtVP1 for molecular fusion of unrelated antigenic modules. (B)

VP1dC was engineered by removal of 63 amino acids from the C-terminus of wtVP1 sequence. (C)

VP1dNdC was engineered by removal of 28 amino acids from the N-terminus of VP1dC. Two

insertion sites were engineered at N- and C-termini of VP1dNdC. This figure was generated based

on the information available in Middelberg et al. [137] and Wibowo et al. [139].

VP1 Capsomere platform enables capsomere vaccine to be produced using E. coli, which is a simple,

low-cost, and high yield expression system [141]. Furthermore, this capsomere platform has been

developed to enable large-scale production at gram-per-litre levels in a bioreactor [142]. A study

using process models coupled with simulations has demonstrated the potential of VP1 capsomere

platform to deliver rapidly manufactured, low-cost vaccines. Based on a simulation, 320 million doses

of capsomere vaccine can be produced using a 10-kL fermenter in 2.3 days at a cost of less than one

cent per dose [143].

2.7 Hemagglutinin as a vaccine target

32

HA and NA are major target antigens in the production of subunit vaccines against influenza as they

induce a specific humoral immune response that provides protection against, or recovery after,

infection [144]. Immunity induced by HA and NA varies considerably due to their functional

differences in the replication cycle of influenza viruses [145, 146]. Antibodies against HA, generated

either during infection or vaccination, are capable of neutralising influenza viruses and prevent virus

entry into host cells [14]. These neutralising antibodies represent the first line of defense against

influenza infection [147] and are the most significant in terms of protective efficacy [148]. In contrast,

anti-NA antibodies do not prevent viral infection but act at a later stage of infection by preventing the

release of the virus from infected cells [145]. Therefore, HA is considered as the main component of

subunit vaccines, while NA is an additional component to broaden the immune response and enhance

vaccine effectiveness [149].

HA is a homotrimeric glycoprotein present in the viral spikes that protrude from the lipid envelope

of the influenza virion. Its main function is to bind sialic acid receptors present on host cells and

thereby to mediate virus attachment and entry [2, 150]. HA is encoded by viral RNA segment 4 [151]

and is synthesised and assembled post-translationally as a precursor polypeptide (HA0) that is

composed of three identical monomers. Each monomer contains two distinct domains, a distal domain

of globular shape (a globular domain; HA1) and a proximal fibrous stem-like structure (a stalk region;

HA2), as shown in Figure 2-6. HA1 and HA2 are linked by a single disulfide bond [1]. Cleavage of

HA0 at a single arginine residue by host protease is essential for activation of membrane fusion and

virus infectivity [33]. HA1 contains the globular domain of HA which is distal from the viral surface.

It contains the conserved sialic acid-binding pocket surrounded by antigenically variable antibody-

binding sites [33, 152]. The sialic acid binding site is the cellular receptor for the influenza virus. It

binds to the receptor binding site at the membrane distal-tip of the HA molecule, which is composed

of residues conserved in all subtypes of influenza virus [152]. HA2 contains a fusion peptide and the

transmembrane domain (Figure 2-6). HA2 is more conserved than HA1. The most highly conserved

sequence of the HA molecule is the N-terminus of HA2, where only five substitutions in the first 23

amino acids have been observed [152]. It has been shown to induce a broad-spectrum, neutralising

immune response against different virus strains and subtypes and has been targeted for use in broad

cross-protection vaccines [153, 154]. However, the vast majority of neutralising antibodies are

directed against several regions within the globular head of HA1 [155], making HA1 a promising

candidate for subtype-specific vaccines.

33

Figure 2-6: Structural features and organisation of influenza virus hemagglutinin precursor,

HA0. HA0 monomer contains two subunits, HA1 and HA2. HA1 globular head (blue) contains sialic

acid binding site. HA2 (purple) contains a fusion peptide (red) and the transmembrane domain. HA0

trimer is composed of three identical monomers. Adapted from [156].

There are several antigenic sites within the globular head of the HA molecule. For example, the

globular head of influenza A PR8 (H1N1) virus, used in this study, contains four antigenic sites

(epitopes), Sa, Sb, Ca (Ca1 and Ca2) and Cb (Figure 2-7) [157]. Antigenic site Sa is located on the

upper part of the globular head of HA1 subunit, close to the receptor binding site of each monomer

of the HA trimer. Site Sb occupies the back region of the globular head. Site Ca consists of two

subsites, Ca1 and Ca2, located on opposite surfaces of the HA monomer, but they form one site in

the interface region of two HA monomers. Site Cb is located near the bottom of the globular head

(Figure 2-7). All of these antigenic sites are conformational and formed by polypeptide sequences

within the HA1 sequence that are located adjacent to each other [158]. Therefore, to generate the

immune response, it is essential that HA1 is presented in its native state. However, HA1 is commonly

34

expressed as inclusion bodies in E. coli. [159, 160]. E. coli-expressed HA1 protein has been refolded

from inclusion bodies to take on the native conformation, as reported in several studies [161-165].

Figure 2-7: Antigenic sites of HA1. HA1 showing antigenic sites, Sa (orange), Sb (red), Ca1(blue),

Ca2 (green), Cb (yellow) and sialic acid (receptor)-binding site. One monomer is shown in blue. The

other monomers are shown in dark green. Adapted from [166, 167].

2.8 Structure-based vaccine design strategy

Structural vaccinology uses protein structure information obtained from structural studies and/or

protein databases to rationally engineer antigens for vaccine production. It is increasingly applied in

vaccine design and has shown promise in generating novel vaccine antigens that are not possible

using conventional vaccine manufacturing methods [168]. It allows the rational design of vaccines

using information derived from the crystal structure of protein antigenic modules, containing antibody

binding sites in complex with a protective antibody, as the starting point [168]. Antigenic epitopes

are identified based on the protein amino acid sequences and the secondary and tertiary structures

[169].

35

The principle of structure-based design is built on the observation that an efficacious immune

response does not require recognition of the entire antigenic module but specific epitopes within the

module [170]. The contacts between antigen and neutralising antibody determine a structural epitope.

This also identifies the substructures that must be retained to preserve key epitopes, while other

unnecessary features can be eliminated to optimise biochemical and immunologic performance.

Using high-resolution structural information, optimised antigens can be engineered to be more stable,

homogeneous and efficiently produced, making immunisation more practical and affordable. For

example, structure-based design strategies have been utilised in the design of HA-based vaccines.

Xuan et al. [165] demonstrated that a protein fragment, HA1, containing the receptor-binding pocket

is capable of inducing both humoral and cellular immune responses against the intact virus which

further confers protection against a viral challenge. In addition, Bommakanti et al. [171] demonstrated

that an immunogen comprising HA2 from the H1N1 subtype (PR/8/34) provided complete protection

against homologous viral challenge in mice.

Moreover, understanding the structural basis of immunogenicity and immunodominance can improve

vaccine efficacy and broaden the range of vaccine-preventable diseases [172]. Structure-based

methods have been employed to develop vaccines for high diversity viruses, such as human

immunodeficiency virus (HIV-1), influenza virus and hepatitis C virus (HCV) [173]. Furthermore,

the structure-based vaccine developed for use against serogroup B meningococcus, the first genome-

derived vaccine, was recently approved for use in Europe [170]. Therefore, structure-based methods

may enable production of an immunogenic modular capsomere vaccine that subsequently elicits

protective immune responses.

2.9 Linker design for modular capsomere presenting HA1

Modularisation of large protein modules, such as HA1, may cause a steric effect between the two

protein domains (platform and insert) which may lead to protein misfolding and aggregation when

produced in E. coli [174]. To ensure adequate separation between the platform and the insert, flexible

linkers can be introduced between the two domains. Glycine (G)- and serine (S)- rich flexible linkers

(GS linkers) are commonly used. The most widely used GS linker has the sequence of (GGGGS)n

[175]. The length of this GS linker can be modified by adjusting the copy number (n). The GS linker

has been shown to increase correct folding and stability in several fusion proteins [175]. For instance,

36

Huston et al. [176] demonstrated that insertion of (GGGGS)3 between the C-terminus of heavy-chain

(VH) and the N-terminus of light-chain variable (VL) regions resulted in a single-chain antibody

fragment (scFv), which had similar specificity and affinity to the parent antibody [176]. For modular

proteins, such as modular VP1 capsomere presenting M2e peptides, GS linkers were introduced

between the surface-exposed loop and the M2e sequence to allow separation of the two domains

[137]. Another flexible linker, GSAGSAAGSGEF, was designed by Waldo et al. to detect correct

folding of green fluorescence protein (GFP)-fusion proteins [177]. This linker performs similarly to

(GGGGS)4 linker, but with reduced amount of homologous repeats in the DNA coding sequence.

Thus, it is less likely to be deleted during cloning using homologous recombination techniques [177].

37

2.10 References

1. Webster, R.G., et al., Evolution and ecology of influenza A viruses. Microbiological Reviews,

1992. 56(1): p. 152-179.

2. Zambon, M.C., The pathogenesis of influenza in humans. Reviews in Medical Virology, 2001.

11(4): p. 227-241.

3. Yuanji, G., et al., Isolation of influenza C virus from pigs and experimental infection of pigs

with influenza C virus. Journal of General Virology, 1983. 64(1): p. 177-182.

4. Osterhaus, A., et al., Influenza B virus in seals. Science, 2000. 288(5468): p. 1051-1053.

5. Centers for Disease Control and Prevention, Types of Influenza Viruses, 2017 [cited

29.08.2017]; available from: https://www.cdc.gov/flu/about/viruses/types.htm.

6. Hause, B.M., et al., Characterization of a novel influenza virus in cattle and swine: proposal

for a new genus in the Orthomyxoviridae family. mBio, 2014. 5(2): p. e00031-14.

7. Rumschlag-Booms, E. and L. Rong, Influenza A Virus Entry: Implications in Virulence and

Future Therapeutics. Advances in Virology, 2012. 2013.

8. Subbarao, K. and T. Joseph, Scientific barriers to developing vaccines against avian influenza

viruses. Nature Reviews Immunology, 2007. 7(4): p. 267-278.

9. Nelson, M.I. and E.C. Holmes, The evolution of epidemic influenza. Nature Reviews Genetics,

2007. 8(3): p. 196-205.

10. Samji, T., Influenza A: understanding the viral life cycle. Yale Journal of Biology and

Medicine, 2009. 82(4): p. 153.

11. Tong, S., et al., New world bats harbor diverse influenza A viruses. PLoS Pathogens, 2013.

9(10): p. e1003657.

12. Tong, S., et al., A distinct lineage of influenza A virus from bats. Proceedings of The National

Academy of Sciences, 2012. 109(11): p. 4269-4274.

13. Cox, N. and K. Subbarao, Global epidemiology of influenza: past and present. Annual Review

of Medicine, 2000. 51(1): p. 407-421.

14. Wilson, I.A. and N.J. Cox, Structural basis of immune recognition of influenza virus

hemagglutinin. Annual Review of Immunology, 1990. 8(1): p. 737-787.

15. Girard, M.P., et al., The 2009 A (H1N1) influenza virus pandemic: A review. Vaccine, 2010.

28(31): p. 4895-4902.

38

16. Kawaoka, Y., S. Krauss, and R.G. Webster, Avian-to-human transmission of the PB1 gene of

influenza A viruses in the 1957 and 1968 pandemics. Journal of Virology, 1989. 63(11): p.

4603-4608.

17. Nelson, M.I., et al., Multiple reassortment events in the evolutionary history of H1N1

influenza A virus since 1918. PLoS Pathogens, 2008. 4(2): p. e1000012.

18. Brown, I.H., The epidemiology and evolution of influenza viruses in pigs. Veterinary

Microbiology, 2000. 74(1): p. 29-46.

19. Haaheim, L., Basic influenza virology and immunology, in Introduction to pandemic

influenza, 2009. Cambridge University Press, Cambridge.

20. Matlin, K.S., et al., Infectious entry pathway of influenza virus in a canine kidney cell line.

Journal of Cell Biology, 1981. 91(3): p. 601-613.

21. Huang, Q., et al., Early steps of the conformational change of influenza virus hemagglutinin

to a fusion active state: stability and energetics of the hemagglutinin. Biochimica et

Biophysica Acta (BBA)-Biomembranes, 2003. 1614(1): p. 3-13.

22. Pinto, L.H., L.J. Holsinger, and R.A. Lamb, Influenza virus M2 protein has ion channel

activity. Cell, 1992. 69(3): p. 517-528.

23 Boulo, S., et al., Nuclear traffic of influenza virus proteins and ribonucleoprotein complexes.

Virus Research, 2007. 124(1): p. 12-21.

24. Garoff, H., R. Hewson, and D.-J.E. Opstelten, Virus maturation by budding. Microbiology

and Molecular Biology Reviews, 1998. 62(4): p. 1171-1190.

25. Kapoor, S. and K. Dhama, Epidemiology of Influenza Viruses, in Insight into Influenza

Viruses of Animals and Humans, 2014. Springer. p. 65-86.

26. Lee, C.-W. and Y.M. Saif, Avian influenza virus. Comparative Immunology, Microbiology

and Infectious Diseases, 2009. 32(4): p. 301-310.

27. Allan, W.H., et al., Use of virulence index tests for avian influenza viruses. Avian Diseases,

1977. 21(3): p. 359-363.

28. National Research Council (U.S.). Committee on Animal Health, Methods for the

examination of poultry biologics. 1963. The National Academies Press. .


Diseases, 2006. 12(9): p. 1319.

30. World Health Organization, Avian and other zoonotic influenza. 2017 [cited 30.08.2017];

available from: http://www.who.int/mediacentre/factsheets/avian_influenza/en/.

39

31. Alexander, D.J., An overview of the epidemiology of avian influenza. Vaccine, 2007. 25(30):

p. 5637-5644.

32. Rott, R., The pathogenic determinant of influenza virus. Veterinary Microbiology, 1992. 33(1-

4): p. 303-310.

33. Skehel, J.J. and D.C. Wiley, Receptor binding and membrane fusion in virus entry: the

influenza hemagglutinin. Annual Review of Biochemistry, 2000. 69(1): p. 531-569.

34. Lazarowitz, S.G. and P.W. Choppin, Enhancement of the infectivity of influenza A and B

viruses by proteolytic cleavage of the hemagglutinin polypeptide. Virology, 1975. 68(2): p.

440-454.

35. Senne, D., et al., Survey of the hemagglutinin (HA) cleavage site sequence of H5 and H7 avian

influenza viruses: amino acid sequence at the HA cleavage site as a marker of pathogenicity

potential. Avian Diseases, 1996: p. 425-437.

36. Connor, R.J., et al., Receptor specificity in human, avian, and equine H2 and H3 influenza

virus isolates. Virology, 1994. 205(1): p. 17-23.

37. Gambaryan, A., et al., Specification of receptor-binding phenotypes of influenza virus isolates

from different hosts using synthetic sialylglycopolymers: non-egg-adapted human H1 and H3

influenza A and influenza B viruses share a common high binding affinity for 6′-sialyl (N-

acetyllactosamine). Virology, 1997. 232(2): p. 345-350.

38. Vandegrift, K.J., et al., Ecology of avian influenza viruses in a changing world. Annals of the

New York Academy of Sciences, 2010. 1195(1): p. 113-128.

39. Alexander, D.J., A review of avian influenza in different bird species. Veterinary

Microbiology, 2000. 74(1): p. 3-13.

40. Webster, R.G., et al., Intestinal influenza: replication and characterization of influenza

viruses in ducks. Virology, 1978. 84(2): p. 268-278.

41. Kaplan, B.S. and R.J. Webby, The avian and mammalian host range of highly pathogenic

avian H5N1 influenza. Virus Research, 2013. 178(1): p. 3-11.

42. Kida, H., et al., Potential for transmission of avian influenza viruses to pigs. Journal of

General Virology, 1994. 75(9): p. 2183-2188.

43. Shi, Y., et al., Enabling the'host jump': structural determinants of receptor-binding specificity

in influenza A viruses. Nature Reviews. Microbiology, 2014. 12(12): p. 822.

44. Lupiani, B. and S.M. Reddy, The history of avian influenza. Comparative Immunology,


45. Pereira, H., B. Tůmová, and V. Law, Avian influenza A viruses. Bulletin of the World Health

Organization, 1965. 32(6): p. 855.

40

46. World Health Organization/Centers for Disease Control and Prevention, Avian influenza

assessing the pandemic threat. 2005 [cited 16.10.2017]; available from:

http://apps.who.int/iris/bitstream/10665/68985/1/WHO_CDS_2005.29.pdf.

47. Chan, P.K., Outbreak of avian influenza A (H5N1) virus infection in Hong Kong in 1997.

Clinical Infectious Diseases, 2002. 34(Supplement_2): p. S58-S64.

48. Rushton, J., et al., Impact of avian influenza outbreaks in the poultry sectors of five South East

Asian countries (Cambodia, Indonesia, Lao PDR, Thailand, Viet Nam) outbreak costs,

responses and potential long term control. World's Poultry Science Journal, 2005. 61(03): p.

491-514.

49. Gilbert, M., et al., Mapping H5N1 highly pathogenic avian influenza risk in Southeast Asia.

Proceedings of the National Academy of Sciences, 2008. 105(12): p. 4769-4774.

50. Li, K., et al., Genesis of a highly pathogenic and potentially pandemic H5N1 influenza virus

in eastern Asia. Nature, 2004. 430(6996): p. 209.

51. World Organization for Animal Health, Update on avian influenza in animals (types H5 and

H7). 2018 [cited 16.07.18]; available from: http://www.oie.int/animal-health-in-the-

world/update-on-avian-influenza/2011/.

52. Greene, J.L., Update on the highly-pathogenic avian influenza outbreak of 2014-2015. 2015

[cited 28.09.2017]; available from: https://fas.org/sgp/crs/misc/R44114.pdf.

53. Jhung, M.A. and D.I. Nelson, Outbreaks of avian influenza a (H5N2), (H5N8), and (H5N1)

among birds — United States, December 2014–January 2015. Morbidity and Mortality

Weekly Report, 2015. 64(4): p. 111.

54. Fry, E., What the worst bird flu outbreak in U.S. history means for farms. 2015 [cited

25.05.2015]; available from: http://fortune.com/2015/06/25/bird-flu-outbreak-farms/.

55. Lee, Y.-J., et al., Novel reassortant influenza A (H5N8) viruses, South Korea, 2014. Emerging

Infectious Diseases, 2014. 20(6): p. 1087.

56. European Centre for Disease Prevention and Control, Outbreaks of highly pathogenic avian

influenza A (H5N8) in Europe. 2016; available from:

http://ecdc.europa.eu/en/publications/Publications/risk-assessment-avian-influenza-H5N8-

europe.pdf.

57. World Organization for Animal Health, OIE Situation Report for Avian Influenza. 2017 [cited

28.09.2017]; available from:

http://www.oie.int/fileadmin/Home/eng/Animal_Health_in_the_World/docs/pdf/OIE_AI_sit

uation_report/OIE_SituationReport_AI_18September2017.pdf.

58. David E. Swayne, D.L.S., and Leslie D. Sims, Influenza, in Diseases of poultry, J.R.G.a.L.R.

McDougald, Editor. 2013, John Wiley & Sons, Inc.: John Wiley & Sons, Inc. p. 181-218.

http://www.oie.int/animal-health-in-the-world/update-on-avian-influenza/2011/

http://www.oie.int/animal-health-in-the-world/update-on-avian-influenza/2011/

41

59. Alexander, D., et al., An outbreak of highly pathogenic avian influenza in turkeys in Great

Britain in 1991. Veterinary Record, 1993. 132(21): p. 535-536.

60. Shortridge, K.F., Poultry and the influenza H5N1 outbreak in Hong Kong, 1997: abridged

chronology and virus isolation. Vaccine, 1999. 17: p. S26-S29.

61. Sims, L., et al., An update on avian influenza in Hong Kong 2002. Avian Diseases, 2003.

47(s3): p. 1083-1086.

62. Yee, K.S., T.E. Carpenter, and C.J. Cardona, Epidemiology of H5N1 avian influenza.

Comparative Immunology, Microbiology and Infectious Diseases, 2009. 32(4): p. 325-340.

63. Swayne, D. and D. Suarez, Highly pathogenic avian influenza. Revue Scientifique et

Technique (International Office of Epizootics), 2000. 19(2): p. 463-482.

64. Villareal, C. and A. Flores, The Mexican avian influenza (H5N2) outbreak. Avian Diseases,

2003. 47: p. 18-22.

65. Capua, I., et al., Outbreaks of highly pathogenic avian influenza (H5N2) in Italy during

October 1997 to January 1998. Avian Pathology, 1999. 28(5): p. 455-460.

66. McNulty, M., et al., Isolation of a highly pathogenic influenza virus from turkeys. Avian

Pathology, 1985. 14(1): p. 173-176.

67. Mantell, C. H5N8 Avian Flu Flies Across Europe. 2016. [cited 04.03.17]; available from:

http://www.healthmap.org/site/diseasedaily/article/h5n8-avian-flu-flies-across-europe-

112216.

68. Wells, R., An outbreak of fowl plague in turkeys. Veterinary. Record, 1963. 75: p. 783-786.

69. Selleck, P., et al., Identification and characterisation of an H7N3 influenza A virus from an

outbreak of virulent avian influenza in Victoria. Australian Veterinary Journal, 1997. 75(4):

p. 289-292.

70. Muhammad, K., M. Akram Muneer, and T. Yaqub, Isolation and characterization of avian

influenza virus from an outbreak in commercial poultry in Pakistan. Pakistan Veterinary

Journal, 1997. 17: p. 6-8.

71. Rojas, H., et al., Avian influenza in poultry in Chile. Veterinary Record, 2002. 151(6): p. 188-

188.

72. Selleck, P., et al., An outbreak of highly pathogenic avian influenza in Australia in 1997

caused by an H7N4 virus. Avian Diseases, 2003. 47(s3): p. 806-811.

73. Turner, A., The isolation of fowl plague virus in Victoria. Australian Veterinary Journal, 1976.

52(8): p. 384-384.

42

74. Barr, D., et al., Avian influenza on a multi‐age chicken farm. Australian Veterinary Journal,

1986. 63(6): p. 195-196.

75. Elbers, A.R.W., et al., The highly pathogenic avian influenza A (H7N7) virus epidemic in the

Netherlands in 2003—lessons learned from the first five outbreaks. Avian Diseases, 2004.

48(3): p. 691-705.

76. World Organization for Animal Health, Immediate notifications and follow-up reports of

highly pathogenic avian influenza (types H5 and H7). 2013 [cited 28.09.2017]; available

from:

http://www.oie.int/wahis_2/public%5C..%5Ctemp%5Creports/en_fup_0000013203_201303

27_111933.pdf.

77. Otte, J., et al., Impacts of avian influenza virus on animal production in developing countries.

CAB Reviews: Perspectives in Agriculture, Veterinary Science, Nutrition and Natural

Resources, 2008. 3(080): p. 18.

78. McLeod, A., et al. Economic and social impacts of avian influenza. in Proceedings of the joint

FAO/OMS/OIE/World Bank conference on avian influenza and human pandemic influenza,

November. 2005.

79. L. Stone, K., et al., Economic impacts of an outbreak of HPAI and potential market

disruptions. Journal of Agricultural and Resource Economics, 2009. 34(3): p. 545-545.

80. Beach, R.H., et al., The Effects of Avian Influenza News on Consumer Purchasing Behavior:

A Case Study of Italian Consumers' Retail Purchases. 2008, United States Department of

Agriculture, Economic Research Service.

81. Kuethe, J.N.a.T., Economic Implications of the 2014-2015 Bird Flu. 2015 [cited 04.08.2017];

available from: http://farmdocdaily.illinois.edu/2015/06/economic-implications-of-the-2014-

2015-bird-flu.html.

82. Centers for Disease Control and Prevention, Avian influenza A virus infectioms in humans.

2017 [cited 08.09.2017]; available from: https://www.cdc.gov/flu/avianflu/avian-in-

humans.htm.

83. Subbarao, K., et al., Characterization of an avian influenza A (H5N1) virus isolated from a

child with a fatal respiratory illness. Science, 1998. 279(5349): p. 393-396.

84. Belser, J.A., et al., Past, present, and possible future human infection with influenza virus A

subtype H7. Emerging Infectious Diseases, 2009. 15(6): p. 859.

85. Claas, E.C.J., et al., Human influenza A H5N1 virus related to a highly pathogenic avian

influenza virus. The Lancet, 1998. 351(9101): p. 472-477.

86. World Health Organization, Cumulative number of confirmed human cases for avain

influenza A (H5N1) reported to WHO, 2003-2007. 2017 [cited 28.08.2017]; available from:

http://www.who.int/influenza/human_animal_interface/2017_07_25_tableH5N1.pdf?ua=1.

43

87. World Health Organization, Asian Lineage Avian Influenza A (H7N9) Virus. 2017 [cited

28.08.2017]; available from: https://www.cdc.gov/flu/avianflu/h7n9-virus.htm.

88. Kurtz, J., R.J. Manvell, and J. Banks, Avian influenza virus isolated from a woman with

conjunctivitis. The Lancet, 1996. 348(9031): p. 901-902.

89. Koopmans, M., et al., Transmission of H7N7 avian influenza A virus to human beings during

a large outbreak in commercial poultry farms in the Netherlands. The Lancet, 2004.

363(9409): p. 587-593.

90. Puzelli, S., et al., Serological analysis of serum samples from humans exposed to avian H7

influenza viruses in Italy between 1999 and 2003. The Journal of Infectious Diseases, 2005.

192(8): p. 1318-1322.

91. Tweed, S.A., et al., Human illness from avian influenza H7N3, British Columbia. Emerging

Infectious Diseases, 2004. 10(12): p. 2196.

92. Nguyen-Van-Tam, J., et al., Outbreak of low pathogenicity H7N3 avian influenza in UK,

including associated case of human conjunctivitis. Euro surveillance: bulletin Europeen sur

les maladies transmissibles= European communicable disease bulletin, 2006. 11(5): p.

E060504. 2.

93. Ostrowsky, B., et al., Low pathogenic avian influenza A (H7N2) virus infection in

immunocompromised adult, New York, USA, 2003. Emerging Infectious Diseases, 2012.

18(7): p. 1128.

94. Matrosovich, M.N., S. Krauss, and R.G. Webster, H9N2 Influenza A Viruses from Poultry in

Asia Have Human Virus-like Receptor Specificity. Virology, 2001. 281(2): p. 156-162.

95. Butt, K., et al., Human infection with an avian H9N2 influenza A virus in Hong Kong in 2003.

Journal of Clinical Microbiology, 2005. 43(11): p. 5760-5767.

96. To, K.K.W., et al., Emergence in China of human disease due to avian influenza A(H10N8) –

Cause for concern? Journal of Infection, 2014. 68(3): p. 205-215.

97. Swayne, D.E., Principles for vaccine protection in chickens and domestic waterfowl against

avian influenza. Annals of the New York Academy of Sciences, 2006. 1081(1): p. 174-181.

98. Swayne, D.E., Epidemiology of avian influenza in agricultural and other man-made systems.

Avian Influenza, 2008: p. 59-85.

99. Waisbord, S., T. Michaelides, and M. Rasmuson, Communication and social capital in the

control of avian influenza: lessons from behaviour change experiences in the Mekong Region.

Global Public Health, 2008. 3(2): p. 197-213.

100. Swayne, D.E. and J.R. Glisson, Diseases of poultry editor-in-chief David E. Swayne ;

associate editors John R. Glisson ... [and four others], in Influenza, C. Ebooks, Editor. 2013,

Ames, Iowa : Wiley-Blackwell: Ames, Iowa. p. 181-218.

44


Vaccine, 2007. 25(30): p. 5645-5652.

102. Suarez, D.L., DIVA vaccination strategies for avian influenza virus. Avian Diseases, 2012.

56(4s1): p. 836-844.

103. Capua, I., G. Cattoli, and S. Marangon, DIVA--a vaccination strategy enabling the detection

of field exposure to avian influenza. Developments in Biologicals, 2003. 119: p. 229-233.

104. Centers for Disease Control and Prevention, Highly pathogenic avian influenza A (H5N1) in

birds and other animals. Birds and Poultry 2017 14.04.2017 [cited 28.10.2017]; available

from: https://www.cdc.gov/flu/avianflu/h5n1-animals.htm.

105. Suarez, D.L., Overview of avian influenza DIVA test strategies. Biologicals, 2005. 33(4): p.

221-226.

106. Wu, H., et al., Yeast-derived avian influenza virus hemagglutinin protein induced immune

response in SPF chicken. Journal of Animal and Veterinary Advances, 2011. 10(8): p. 999-

1002.

107. Suarez, D.L. and S. Schultz-Cherry, The effect of eukaryotic expression vectors and adjuvants

on DNA vaccines in chickens using an avian influenza model. Avian Diseases, 2000: p. 861-

868.

108. Zanella, A., G. Poli, and M. Bignami, Avian influenza: approaches in the control of disease

with inactivated vaccines in oil emulsion. Avian Diseases, 2003. 47: p. 180-183.

109. Swayne, D.E., Avian influenza vaccines and therapies for poultry. Comparative Immunology,


110. Swayne, D., et al., Assessment of national strategies for control of high-pathogenicity avian

influenza and low-pathogenicity notifiable avian influenza in poultry, with emphasis on

vaccines and vaccination. Revue Scientifique et Technique-OIE, 2011. 30(3): p. 839.

111. Yassine, H.M., et al., Genetic and antigenic relatedness of H3 subtype influenza A viruses

isolated from avian and mammalian species. Vaccine, 2008. 26(7): p. 966-977.

112. Song, J.M., et al., Generation and evaluation of reassortant influenza vaccines made by

reverse genetics for H9N2 avian influenza in Korea. Veterinary Microbiology, 2008. 130(3):

p. 268-276.

113 Hoffmann, E., et al., Eight-plasmid system for rapid generation of influenza virus vaccines.

Vaccine, 2002. 20(25-26): p. 3165-3170.

45

114. Broadbent, A.J. and K. Subbarao, Influenza virus vaccines: lessons from the 2009 H1N1

pandemic. Current Opinion in Virology, 2011. 1(4): p. 254-262.

115. Barrett, P.N., D. Portsmouth, and H.J. Ehrlich, Developing cell culture-derived pandemic

vaccines. Current Opinion in Molecular Therapeutics, 2010. 12(1): p. 21-30.

116. Doroshenko, A. and S.A. Halperin, Trivalent MDCK cell culture-derived influenza vaccine

Optaflu®(Novartis Vaccines). Expert Review of Vaccines, 2009. 8(6): p. 679-688.

117. Maas, R., et al., Maternal immunity against avian influenza H5N1 in chickens: limited

protection and interference with vaccine efficacy. Avian Pathology, 2011. 40(1): p. 87-92.

118. Meeusen, E.N., et al., Current status of veterinary vaccines. Clinical Microbiology Reviews,

2007. 20(3): p. 489-510.

119. Bublot, M., et al., Development and use of fowlpox vectored vaccines for avian influenza.

Annals of the New York Academy of Sciences, 2006. 1081(1): p. 193-201.

120. Ge, J., et al., Newcastle disease virus-based live attenuated vaccine completely protects

chickens and mice from lethal challenge of homologous and heterologous H5N1 avian

influenza viruses. Journal of Virology, 2007. 81(1): p. 150-158.

121. Zhao, Q., et al., Virus-like particle-based human vaccines: quality assessment based on

structural and functional properties. Trends in Biotechnology, 2013. 31(11): p. 654-663.

122. Lua, L.H., et al., Bioengineering virus‐like particles as vaccines. Biotechnology and

Bioengineering, 2013. 111(3): p. 425-440.

123. Chen, X.S., et al., Papillomavirus capsid protein expression in Escherichia coli: purification

and assembly of HPV11 and HPV16 L1. Journal of Molecular Biology, 2001. 307(1): p. 173-

182.

124. Salunke, D.M., D.L. Caspar, and R.L. Garcea, Self-assembly of purified polyomavirus capsid

protein VP1. Cell, 1986. 46(6): p. 895-904.


thermodynamic attraction between structural protein capsomeres. Journal of The Royal


126. Senger, T., et al., Virus-like particles and capsomeres are potent vaccines against cutaneous

alpha HPVs. Vaccine, 2010. 28(6): p. 1583-1593.

127. Rose, R.C., et al., Human papillomavirus type 11 recombinant L1 capsomeres induce virus-

neutralizing antibodies. Journal of Virology, 1998. 72(7): p. 6151-6154.

128. Chen, X.S., et al., Structure of small virus-like particles assembled from the L1 protein of

human papillomavirus 16. Molecular Cell, 2000. 5(3): p. 557-567.

46

129. Yuan, H., et al., Immunization with a pentameric L1 fusion protein protects against

papillomavirus infection. Journal of Virology, 2001. 75(17): p. 7848-7853.

130. Thönes, N., et al., A direct comparison of human papillomavirus type 16 L1 particles reveals

a lower immunogenicity of capsomeres than viruslike particles with respect to the induced

antibody response. Journal of Virology, 2008. 82(11): p. 5472-5485.

131. Fligge, C., et al., Induction of type-specific neutralizing antibodies by capsomeres of human

papillomavirus type 33. Virology, 2001. 283(2): p. 353-357.

132. Johne, R., et al., Taxonomical developments in the family Polyomaviridae. Archives of

Virology, 2011. 156(9): p. 1627-1634.

133. Belnap, D.M., et al., Conserved features in papillomavirus and polyomavirus capsids. Journal

of Molecular Biology, 1996. 259(2): p. 249-263.

134. Siray, H., et al., Capsid protein-encoding genes of hamster polyomavirus and properties of

the viral capsid. Virus Genes, 1999. 18(1): p. 39-47.

135. Stehle, T. and S.C. Harrison, Crystal structures of murine polyomavirus in complex with

straight-chain and branched-chain sialyloligosaccharide receptor fragments. Structure,

1996. 4(2): p. 183-194.

136. Baker, T., et al., Structures of bovine and human papillomaviruses. Analysis by cryoelectron

microscopy and three-dimensional image reconstruction. Biophysical Journal, 1991. 60(6):

p. 1445-1456.



138. Schmidt, U., R. Rudolph, and G. Böhm, Mechanism of assembly of recombinant murine

polyomavirus-like particles. Journal of Virology, 2000. 74(4): p. 1658-1662.





Vaccine, 2014. 32(29): p. 3651-3655.



142. Liew, M.W., A. Rajendran, and A.P. Middelberg, Microbial production of virus-like particle

vaccine protein at gram-per-litre levels. Journal of Biotechnology, 2010. 150(2): p. 224-231.

143. Chuan, Y.P., et al., The economics of virus-like particle and capsomere vaccines. Biochemical

Engineering Journal, 2014. 90: p. 255-263.

47

144. McMurry, J.A., B.E. Johansson, and A.S. De Groot, A call to cellular & humoral arms:

enlisting cognate T cell help to develop broad-spectrum vaccines against influenza A. Human

Vaccines, 2008. 4(2): p. 148-157.

145. Sylte, M.J. and D.L. Suarez, Influenza neuraminidase as a vaccine antigen, in Vaccines for

Pandemic Influenza. 2009, Springer. p. 227-241.

146. Johansson, B., D. Bucher, and E. Kilbourne, Purified influenza virus hemagglutinin and

neuraminidase are equivalent in stimulation of antibody response but induce contrasting

types of immunity to infection. Journal of Virology, 1989. 63(3): p. 1239-1246.

147. Jeon, S.H. and R. Arnon, Immunization with influenza virus hemagglutinin globular region

containing the receptor-binding pocket. Viral Immunology, 2002. 15(1): p. 165-176.

148. Potter, C., Inactivated influenza virus vaccine, in Basic and applied influenza research, A.S.

Beare, Editor. 1982, CRC Press, Boca Raton, FL. 1982. p. 119-158.

149. Johansson, B.E. and I.C. Brett, Changing perspective on immunization against influenza.

Vaccine, 2007. 25(16): p. 3062-3065.

150. Russell, R.J., et al., Structure of influenza hemagglutinin in complex with an inhibitor of

membrane fusion. Proceedings of the National Academy of Sciences, 2008. 105(46): p.

17736-17741.

151. Katz, J.M. and R.G. Webster, Amino acid sequence identity between the HA1 of influenza A

(H3N2) viruses grown in mammalian and primary chick kidney cells. Journal of General

Virology, 1992. 73(5): p. 1159-1165.

152. Wilson, I., J. Skehel, and D. Wiley, Structure of the haemagglutinin membrane glycoprotein

of influenza virus at 3 Å resolution. Nature, 1981. 289(5796): p. 366-373.

153. Sui, J., et al., Structural and functional bases for broad-spectrum neutralization of avian and

human influenza A viruses. Nature Structural & Molecular Biology, 2009. 16(3): p. 265-273.

154. Ekiert, D.C., et al., Antibody recognition of a highly conserved influenza virus epitope.

Science, 2009. 324(5924): p. 246.



156. Amorij, J.-P., et al., Development of stable influenza vaccine powder formulations: challenges

and possibilities. Pharmaceutical Research, 2008. 25(6): p. 1256-73.


subtype). Cell, 1982. 31(2): p. 417-427.

48

158. Jackson, D.C., et al., Antigenic determinants of influenza virus hemagglutinin I. Cyanogen

bromide peptides derived from A/MEMPHIS/72 hemagglutinin possess antigenic activity.

Virology, 1978. 89(1): p. 199-205.

159. Davis, A.R., et al., Immune response to human influenza virus hemagglutinin expressed in

Escherichia coli. Gene, 1983. 21(3): p. 273-284.

160. Shen, S., et al., Comparing the antibody responses against recombinant hemagglutinin

proteins of avian influenza A (H5N1) virus expressed in insect cells and bacteria. Journal of

Medical Virology, 2008. 80(11): p. 1972-1983.





2010. 5(7): p. e11548.



Virology, 2011. 85(3): p. 1246-1256.

164. Song, L., et al., Efficacious recombinant influenza vaccines produced by high yield bacterial

expression: a solution to global pandemic and seasonal needs. PLoS One, 2008. 3(5): p.

e2257.



166. Ellebedy, A.H. and R. Ahmed, Re-engaging cross-reactive memory B cells: The influenza

puzzle. Frontiers in Immunology, 2012. 3(53).

167. Castelán-Vega, J.A., et al., The hemagglutinin of the influenza A (H1N1) pdm09 is mutating

towards stability. Advances and Applications in Bioinformatics and Chemistry: AABC, 2014.

7: p. 37.

168. Kulp, D.W. and W.R. Schief, Advances in structure-based vaccine design. Current Opinion

in Virology, 2013. 3(3): p. 322-331.

169. Finco, O. and R. Rappuoli, Designing vaccines for the twenty-first century society. Frontiers

in Immunology, 2014. 5(12): p. 1-6.

170. Scarselli, M., et al., Rational design of a meningococcal antigen inducing broad protective

immunity. Science Translational Medicine, 2011. 3(91): p. 91ra62-91ra62.

171. Bommakanti, G., et al., Design of an HA2-based Escherichia coli expressed influenza

immunogen that protects mice from pathogenic challenge. Proceedings of the National


49

172. Dormitzer, P.R., G. Grandi, and R. Rappuoli, Structural vaccinology starts to deliver. Nature

Reviews. Microbiology, 2012. 10(12): p. 807.

173. Burton, D.R., et al., Broadly neutralizing antibodies present new prospects to counter highly

antigenically diverse viruses. Science, 2012. 337(6091): p. 183.

174. Makrides, S.C., Strategies for achieving high-level expression of genes in Escherichia coli.

Microbiological Reviews, 1996. 60(3): p. 512-538.

175. Chen, X., J.L. Zaro, and W.-C. Shen, Fusion protein linkers: property, design and

functionality. Advanced Drug Delivery Reviews, 2013. 65(10): p. 1357-1369.

176. Huston, J.S., et al., Protein engineering of antibody binding sites: recovery of specific activity

in an anti-digoxin single-chain Fv analogue produced in Escherichia coli. Proceedings of the

National Academy of Sciences, 1988. 85(16): p. 5879-5883.

177. Waldo, G.S., et al., Rapid protein-folding assay using green fluorescent protein. Nature

Biotechnology, 1999. 17(7): p. 691-695.

50

Chapter 3

Structural-based designed modular capsomere comprising HA1 for low-

cost poultry influenza vaccination

The entire chapter consists of the peer-reviewed journal article published as:

Waneesorn J, Wibowo N, Bingham J, Middelberg AP, and Lua LH. Structural-based designed

modular capsomere comprising HA1 for low-cost poultry influenza vaccination. Vaccine, 2018.

36(22):p. 3064-3071. Available online 25 November 2016.

The following changes have been made to the text of this chapter:

- Page, figure, section and subsection numbering and reference style have been changed to be

consistent with the rest of the thesis.

51

3.1 Abstract

Highly pathogenic avian influenza (HPAI) viruses cause a severe and lethal infection in domestic

birds. The increasing number of HPAI outbreaks has demonstrated the lack of capabilities to control

the rapid spread of avian influenza. Poultry vaccination has been shown to not only reduce the viral

spread in animals but also reduce transmission of the virus to humans, preventing potential pandemic

development. However, existing vaccine technologies cannot respond to a new virus outbreak rapidly

and at a cost and scale that is commercially viable for poultry vaccination. Here, we developed a

modular capsomere, a subunit of virus-like particles, as a low-cost poultry influenza vaccine.

Modified murine polyomavirus (MuPyV) VP1 capsomere was used to present structure-based

influenza hemagglutinin (HA1) antigen. Six constructs of modular capsomeres presenting three

truncated versions of HA1 and two constructs of modular capsomeres presenting non-modified HA1

have been generated. These modular capsomeres were successfully produced in stable forms using

Escherichia coli, without the need for protein refolding. Based on ELISA results, this adjuvanted

modular capsomere (CaptHA1-3C) induced a strong antibody response (almost 105 endpoint titre)

when administered to chickens, similar to titres obtained in the group administered with insect cell-

derived HA1 proteins. Chickens that received adjuvanted CaptHA1-3C followed by challenge with

HPAI virus were fully protected. The results presented here indicate that this bacterially-produced

modular capsomere platform could potentially translate into a rapid-response and low-cost vaccine

manufacturing technology suitable for poultry vaccination.

3.2 Introduction

Influenza viruses cause severe respiratory tract infections, resulting in substantial morbidity and

mortality. The increasing number of human cases of avian influenza by H5N1 and H7N9 underlines

the threat of a possible pandemic. H5N1 avian influenza viruses have caused more than 600

confirmed cases with 60% fatality in humans [1]. The recent highly pathogenic avian influenza

(HPAI) outbreak in domestic poultry and wild birds in the US in 2015, affecting more than 48 million

birds and causing economic losses of billions of dollars due to the death and culling of poultry [2-4],

has demonstrated the lack of sophisticated capabilities to control the rapid spread of avian influenza.

Vaccination of poultry has been shown to not only reduce the spread of influenza in poultry but also

reduce virus transmission into humans [5], suggesting that poultry vaccination can prevent pandemic

52

development as part of a One Health approach. Although a number of alternative influenza vaccine

candidates have been developed [6-10], vaccine cost is still high and not suitable for poultry

vaccination, especially in developing countries [11].

The constraints of currently available vaccines underline the need for rapidly manufactured, safe and

low-cost vaccines for control of avian influenza. Previous studies have demonstrated the potential of

a platform technology based on modularised murine polyomavirus (MuPyV) capsid protein VP1

virus-like particles (VLPs) and their pentameric subunits, termed capsomeres, to present foreign

antigens [12]. VP1 can be produced and scaled-up to gram-per-liter levels in Escherichia coli (E.

coli) [13], followed by purification processes [14, 15]. Studies have demonstrated that the modular

VLPs and capsomeres induce strong immune responses against targeted antigens [12, 16].

Capsomeres are simpler, hence, faster and cheaper to manufacture than VLPs, and are therefore well

suited as a platform for poultry vaccine. A process simulation, based on a conservative assumption

of 50 µg protein per vaccine dose, demonstrated that 320 million doses of modular capsomere

vaccines can be produced in 2.3 days, at a cost of less than one cent per dose [17]. Both capital

investment and operating cost are included in the simulation. This technology can potentially translate

into a rapid-response and low-cost vaccine manufacturing technology suitable for poultry

vaccination.

VP1 capsomere has previously been used to present small peptide epitopes, such as an influenza virus

M2e peptide antigen [12, 18]. However, modularisation of the whole antigenic protein domain to VP1

platform has several advantages over small peptide epitopes. Presentation of the whole protein

domain, such as the globular head (HA1) of influenza virus hemagglutinin (HA), has been shown to

increase immunogenicity due to the presence of secondary structural elements which are necessary

to correctly display conformational epitopes [19] and increase the number of antigenic epitopes

presenting on the protein domain. Modularisation of large antigen to VP1 platform may pose

challenges. It is essential to allow structural separation between two protein domains to maintain their

intact structure, preventing protein aggregation [20]. The design of the linker is crucial to allowing

each protein domain to form a stable and independent domain. The flexible linker

(GSAGSAAGSGEF) designed by Waldo et al. has been shown to maintain correct folding of GFP-

fusion proteins for rapid protein-folding assay [21]. This linker reduced the amount of homologous

53

repeats in its sequence, which could have resulted in deletions during cloning using a homologous

recombination method [21].

Hemagglutinin (HA), an antigen glycoprotein found on the surface of the influenza virus, is known

to induce a strong antibody-mediated immune response [22]. The globular domain of HA (HA1)

contains the receptor binding sites and most of the antigenic sites of the HA molecule that are targets

for neutralising antibodies [23-25], suggesting that HA1 is a promising vaccine target.

In this study, we have developed a modular capsomere as a low-cost poultry influenza vaccine based

on VP1 platform. This platform was redesigned to contain two insertion sites at the N- and C-termini,

suitable for presenting large antigens. HA1 was redesigned based on its structure to retain necessary

elements for conformational epitope formation. Stable, immunogenic and protective modular

capsomeres comprising structure-based designed influenza hemagglutinin globular domain (HA1)

was produced and purified from E. coli without the need for protein refolding.

3.3 Materials and methods

3.3.1 Plasmid construction

Plasmid GST-wtVP1 (pGEXVP1) for expression of GST tagged wild-type VP1 [15] was used to

generate plasmid VP1dC (pVP1dC), the vector backbone in this study. Sixty-three amino acids were

removed from the C-terminus of wtVP1, resulting in the assembly-incompetent mutant VP1dC [26].

To allow insertion of heterologous protein module into VP1dC, two insertion sites (N and C) were

engineered. Site N was designated at the BamHI restriction site (amino acid position 5 of wtVP1).

Site C was generated by inserting a SnaBI restriction site at amino acid position 320 of wtVP1 by

site-directed mutagenesis with the QuikChange II Site Directed Mutagenesis kit (Agilent Stratagene,

CA, USA). Tobacco Etch Virus protease (TEVp) recognition site was inserted between GST and site

N using site-directed mutagenesis. Oligos 5’-

GGTAGCGCAGGTAGTGCAGCAGGTAGCGGTGAATTT-3’ encoding amino acids

GSAGSAAGSGEF (GSA linker) were introduced between GST and the TEVp recognition site, and

between the BamHI site and VP1dC sequence using QuikChange II Site Directed Mutagenesis kit.

To generate VP1-based modular capsomeres presenting HA1 constructs, amplified fragments of full-

length HA1 [HA11-326 (HA1)] and three versions of truncated HA1 [HA132-308 (tHA1-1), HA140-277

54

(tHA1-2) and HA144-268 (tHA1-3)] from influenza A strain A/Puerto Rico/8/34 (H1N1) (PR8 H1N1)

flanked by GSA linkers at its N- and C-termini were generated by PCR. Four constructs encoding

modular capsomeres presenting HA1 on site N, including CapHA1-N, CaptHA1-1N, CaptHA1-2N

and CaptHA1-3N, were generated by ligating these amplified fragments into the N-terminus of

BamHI-linearised plasmid VP1dC by using an in vivo homologous recombination method. The same

method was used to generate two constructs encoding modular capsomeres presenting HA1 on site

C, including CapHA1-C and CaptHA1-1C. In contrast, constructs encoding CaptHA1-2C, and

CaptHA1-3C were generated by ligating SnaBI- and XhoI-digested amplified fragments into SnaBI-

and XhoI-linearised pVP1dC. DNA sequences of all constructs were confirmed by DNA sequencing

(Australian Genome Research Facility (AGRF), Brisbane, Australia).

3.3.2 Protein expression and purification

Modular capsomere constructs were transformed separately into chemically competent E. coli Rosetta

(DE3) pLysS cells (Novagen, CA, USA). GST-tagged modular capsomeres were expressed as

previously described [15] with modification. The E. coli was induced with isopropyl--D-

thiogalactopyranoside (IPTG) at a concentration of 0.05 mM. The cultures were grown at 15 C and

180 rpm and were harvested after 8 h of induction. All culture media (Terrific Broth (TB) medium

and Luria-Bertani (LB) agar plates) were supplemented with 50 mg.L-1 ampicillin and 34 mg.L-1

chloramphenicol. For TB medium, tryptone was used instead of peptone. GST-tagged modular

capsomeres were purified by chromatographic methods as previously described [15] with

modification. For affinity chromatography, GST-tagged modular capsomeres were eluted with E

buffer (40 mM Tris, pH 8.0, 10 mM reduced glutathione, 300 mM NaCl, 1 mM EDTA, 5% (v/v)

glycerol, 5 mM DTT). For GST tag removal, TEVp, which was produced recombinantly as previously

described [27], was used at 50:1 (w/w) ratio (protein:TEVp), for 2 h at room temperature. TEVp was

removed by size exclusion chromatography (SEC) using Superdex 200 30/100 GL column, a flow

rate of 0.5 mL.min-1 (GE Healthcare, UK) and pre-equilibrated with Tris-NaCl buffer (40 mM Tris,

300 mM NaCl, pH 8.0). Endotoxin was removed from protein samples by anion exchange using a

Vivapure Q Mini H spin column (Sartorius Stedim, France) as previously described [12], except Tris-

NaCl buffer was used as the equilibration buffer.

55

3.3.3 HA1 protein preparation

HA1 protein, strain A PR8 H1N1, was produced in High FiveTM insect cells by the Protein Expression

Facility (The University of Queensland, Australia).

3.3.4 Capsomere characterisation

Qualitative analysis of protein samples was performed using standard SDS-PAGE using 10% gel [28]

and Western blot analysis. For Western blot analysis, the proteins from SDS-PAGE gel were

transferred onto a nitrocellulose membrane and probed with mouse anti-HA antibody followed by

horseradish peroxidase (HRP)-conjugated goat anti-mouse antibody (Bio-Rad, CA, USA). Binding

of antibody to the proteins was visualised via chemiluminescence with the NovexECL HRP

chemiluminescent substrate reagent kit (Novex, CA, USA). Protein samples were analysed by SEC-

HPLC using a Superdex 200 30/100 GL column (GE Healthcare, UK), a flow rate of 0.5 mL.min-1,

and pre-equilibrated with Tris-NaCl buffer (40 mM Tris, 300 mM NaCl, pH 8.0).

3.3.5 Hemagglutination assay

Hemagglutination assays were performed in a 96-well V-bottomed microtiter plate (Costar, Thermo

Fisher Scientific, MA, USA). Protein samples (VP1dC capsomeres, CaptHA1-3C and HA1 protein)

with starting concentration of 280 µg.mL-1, were serially 2-fold diluted in PBSA (137 mM NaCl, 2.7

mM KCl, 10.15 mM Na2HPO4, 1.76 mM KH2PO4, 0.1 mg.mL-1 BSA, pH 7.2). The plate was

incubated with 1.1% washed chicken red blood cells (RBCs) (Australian SPF Services Pty. Ltd.).

Agglutination was determined after incubation for 45 min at room temperature. The highest dilution

of antigen that showed complete agglutination of RBCs was defined as the endpoint titre of

hemagglutination. All samples were tested in duplicate.

3.3.6 Immunogenicity study

Three groups of eight Gallus gallus layer chickens (Bond Enterprises, Australia) at three weeks of

age were immunised intramuscularly with 70 µg of total protein on experimental Days 0 and 21.

Blood samples were collected from the wing vein before the first immunisation (Day 0), three weeks

after the first immunisation (Day 21) and two weeks after the second immunisation (Day 35). Group

1 was immunised with VP1dC capsomeres (negative control), Group 2 received CaptHA1-3C

56

(comprising 60% VP1dC and 40% tHA1-3C by mass), and Group 3 received insect cell-based HA1

protein (positive control). All samples were adjuvanted with aluminium hydroxide (Alhydrogel,

Brenntag, Germany) at a 1:1 (v/v) ratio of protein to Alhydrogel. Adjuvanted samples were prepared

as previously described [29]. All animal experimental work was reviewed and approved by The

University of Queensland Animal Ethics Committee (AIBN/235/14/QSF). All animals were cared for

humanely in accordance with The University of Queensland Animal Ethics Committee guidelines.

3.3.7 Enzyme-linked immunosorbent assay (ELISA)

ELISA was performed in 96-well Nunc-ImmunoTM MaxiSorpTM plates as described previously [29].

Insect cell-derived HA1 protein was used to coat the plates. Endpoint titres were determined as the

highest dilution of serum for which the OD was 3 standard deviations above the mean OD of blank

wells containing insect cell-derived HA1 protein and substrate.

3.3.8 Statistical analysis

Statistical analysis was performed on log-transformed data using GraphPad Prism Version 6.00

(GraphPad Software Inc., CA, USA). Comparison between multiple groups was performed with one-

way ANOVA and Tukey’s multiple comparison test. p < 0.05 was considered statistically significant.

3.3.9 Challenge study with HPAI virus

A construct encoding CaptHA1-3C containing HA1(45-262) from influenza virus

A/chicken/NSW/3122-1/2012 (H7N7) (H7N7CaptHA1-3C) was cloned and prepared as described in

section 3.3.1. Three-week-old White Leghorn chickens of either sex were randomly assigned to three

treatment groups (6 birds per group). The three groups received either H7N7CaptHA1-3C, VP1dC

capsomeres or no vaccine. The chickens were administered 0.5 mL of vaccine-Alhydrogel adjuvant

by the subcutaneous route at the base of the neck (Day 0), followed by a second immunisation at Day

21. On Day 35, the chickens were challenged with influenza virus A/chicken/NSW/3122-1/2012

(H7N7) by the mucosal (eyes, nostrils and mouth) administration of 0.2 mL inoculum containing 106.0

median egg infectious doses (EID50). Chickens were observed for disease and were euthanised once

they developed moderate clinical signs including hunched posture, head tucking, drooping wings,

lethargy and unwillingness to move on disturbance, facial swelling and redness, pallor of comb and

wattles, fluffed feathers and diarrhea. All surviving chickens were euthanised at Day 49. Clotted

57

blood samples were taken from wing vein prior to the first immunisation (Day 0) and at 7-day

intervals afterwards (Days 7, 21, 35 and 45). Sera were collected by centrifugation at 1000 x g for

10-15 min and were tested using a hemagglutination-inhibition assay using homologous virus [30].

Tissue samples were taken from all chickens after euthanasia and were fixed in 10% formalin [31].

Histological sections were stained with haematoxylin and eosin stain for histopathological

examination and by an immunohistochemical stain for detection of influenza A nucleoprotein antigen

as previously described [32].

3.4 Results and discussion

Major structural protein VP1 of murine polyomavirus has been previously modified for

modularisation of antigenic modules of small peptide antigens or large protein antigens, either in the

capsomere [12, 18] or VLP platform [12, 15]. Figure 3-1A illustrates the newly designed VP1

capsomere for modularisation of large antigens. Previously, a 23-amino acid M2e peptide was

inserted into truncated VP1 missing the first 28 amino acids and the last 63 amino acids [18]. This

CapM2e, formulated with adjuvant, has demonstrated protective efficacy in mice [16] and was highly

immunogenic in chickens [29]. However, this capsomere format (VP1dNdC) is not suitable for

modularisation of large protein antigens such as HA1 because the truncated N-terminus of VP1 is

hidden within the capsomere core. HA1 has a molecular weight 10-times higher than that of M2e.

Modularisation of HA1 onto the truncated N-terminus of VP1 is likely to cause a steric effect between

HA1 and VP1, resulting in modular capsomere instability [21, 33]. Retaining the native N-terminus

of VP1 may provide flexibility between HA1 and VP1, with the native 28 amino acid residues acting

as a linker between the carrier and module, possibly leading to the formation of a stable modular

capsomere. Therefore, a new VP1 capsomere format comprising VP1 with an intact N-terminus and

a truncation of the last 63 amino acids (VP1dC) is presented in this study (Figure 3-1A).

Figure 3-1B illustrates the rational design of antigenic modules derived from HA1 of influenza A

strain PR8 H1N1 for modularisation into the new VP1 capsomere format. HA monomer contains two

subunits, a globular head (HA1) and a stalk region (HA2), linked by a single disulfide bond. HA1 is

a target for neutralising antibodies as it contains the receptor binding site and most of the neutralising

epitopes of the HA molecule. However, HA1 is highly insoluble when expressed in E. coli [10, 19].

E. coli-expressed HA1 proteins have been purified from inclusion bodies, denatured and refolded [7,

58

8, 10, 35]. HA1 comprises a compact head and a long tail, which interacts with the HA2 subunit [10,

36, 37], as illustrated in Figure 3-1B.

Figure 3-1: Structure-based designed modular capsomere. (A) VP1dC capsomere platform. VP1dC

monomer contains two insertion sites, SN and SC, for presenting antigenic module from foreign

antigens. VP1dC expressed in E. coli spontaneously assemble into a pentameric structure, termed

capsomere as shown in Side-view and Top-view. (B) HA1 antigenic module. HA monomer is

composed of a globular head (HA1, beige) and a stalk region (HA2, red). Full-length HA1 contains

amino acid residues 1-326. Three versions of truncated HA1 include tHA1-1 comprised of amino

acids 32 to 308, tHA1-2 comprised of amino acids 40 to 277, and tHA1-3 comprised of amino acids

44 to 268.

All antigenic sites of HA1 are conformational and present on the head [19, 38]. Maintaining the

correct structure of the HA1 globular head is critical for its immunogenicity. In contrast, the HA1 tail

contains very few antigenic sites [24, 39], thus eliminating this part of the protein may not impair

59

HA1 immunogenicity. Therefore, three truncations of HA1 globular head were designed by

selectively removing the tail, to enhance the proteins’ solubilities in E. coli [40], while retaining the

necessary elements for formation of conformational epitopes. Four designs of HA1, including a full-

length HA1 [HA11-326 (HA1)] and three versions of truncated HA1 [HA132-308 (tHA1-1), HA140-277

(tHA1-2), and HA144-268 (tHA1-3)], were created. Each version of HA1 contains different secondary

structural elements. GSA linkers (GSAGSAAGSGEF) were introduced between GST, HA1 insert

either full length or truncations and VP1dC, to allow spatial separation between each protein domain

to maintain their intact structure. The theoretically calculated molecular weights of HA1, tHA1-1,

tHA1-2 and tHA1-3 are 36.6, 30.7, 27.0 and 25.5 kDa, respectively. Each version of HA1 was

modularised into either site N or C of VP1dC at the DNA level, yielding eight modular constructs

(Figure 3-2A).

These modular constructs were expressed as GST-tagged proteins in E. coli strain, Rosetta (DE3)

pLysS. The theoretically calculated molecular weights of CapHA1-N, CaptHA1-1N, CaptHA1-2N,

and CaptHA1-3N monomer are 101.5, 96.1, 91.9 and 90.4 kDa, respectively. The theoretically

calculated molecular weights of CapHA1-C, CaptHA1-1C, CaptHA1-2C and CaptHA1-3C monomer

are 101.7, 96.3, 92.0 and 90.5 kDa, respectively. Figure 3-2B and 3-2C show SDS-PAGE analysis of

the expression and solubility of the modular constructs. Expression was detected on SDS-PAGE gel

for all modular constructs. However, the level of expression and solubility varied between constructs.

Similar level of target protein expression was observed for CapHA1-N, CaptHA1-2N and CaptHA1-

3N. The lowest expression was detected for CaptHA1-1N (Figure 3-2B). For constructs with antigen

modularisation on the C-terminal, high expression levels were observed for CaptHA1-2C and

CaptHA1-3C, while a relatively lower expression level was detected for CapHA1-C. The lowest

expression level was observed for CaptHA1-1C (Figure 3-2C). The level of target protein solubility

appeared similar for all modular constructs and the soluble fraction was approximately 50% of total

expressed target protein. These results suggested that modularisation of antigenic module onto either

the N- or C-terminus of VP1dC has little effect on the target protein expression or solubility. The size

and structure of the antigenic module does affect the expression and solubility of modular protein.

The results suggest that modular protein expression and solubility is adversely affected either by

increase in the molecular weight or the presence of secondary structural elements of module HA1.

60

Figure 3-2: Schematic representation of HA1 constructs and SDS-PAGE analysis of the

expression and solubility of modular capsomeres (Cap) monomer. (A) Four designs of HA1

antigenic modules were cloned into the N-terminus of VP1dC to generate HA1-N, tHA1-1N, tHA1-

2N and tHA1-3N. Four designs of HA1 antigenic modules were cloned into the C-terminus of VP1dC

to generate HA1-C, tHA1-1C, tHA1-2C and tHA1-3C. (B) CapHA1-N, CaptHA1-1N, CaptHA1-2N

and CaptHA1-3N. (C) CapHA1-C, CaptHA1-1C, CaptHA1-2C and CaptHA1-3C. Lanes: (L)

molecular weight marker, (P) pre-induced lysate, (T) total lysate, and (S) soluble lysate. Predicted

molecular weights of the monomer of CapHA1-N, CaptHA1-1N, CaptHA1-2N, CaptHA1-3N,

CapHA1-1C, CaptHA1-1C, CaptHA1-2C and CaptHA1-3C are 101.5, 96.1, 91.9, 90.4, 101.7, 96.3,

92.0 and 90.5 kDa, respectively.

CaptHA1-1N and CaptHA1-1C were not taken through to purification due to poor expression and

solubility. CapHA1-N, CaptHA1-2N, CaptHA1-3N, CapHA1-C, CaptHA1-2C and CaptHA1-3C

were further purified by GST affinity chromatography, followed by removal of GST tag by TEVp

treatment [27]. Non-tagged modular capsomeres were separated from the GST tag by SEC. Endotoxin

was removed from the purified modular capsomeres by anion exchange chromatography. Figure 3-3

shows SEC chromatograms for the separation of the capsomeres, GST tag and protein aggregates.

61

Protein aggregates were eluted as an excluded peak at approximately 16 min, capsomeres were eluted

at approximately 22 to 24 min depending on molecular weight, and GST dimers were eluted at

approximately 30 min. Figure 3-3A shows VP1dC capsomeres being eluted at approximately 24 min,

as expected. Figure 3-3 (D-G) shows CaptHA1-3N, CapHA1-C, CaptHA1-2C, and CaptHA1-3C,

respectively, being eluted at approximately 22 min due to their higher molecular weights. Capsomere

peaks were not observed for CapHA1-N and CaptHA1-2N. It is likely that all modular capsomeres

were unstable post GST removal and formed aggregates under the selected buffer condition, as shown

in Figure 3-3B and 3-3C, respectively.

After the purification process, high capsomere yield was obtained from CaptHA1-3C (Figure 3-3G).

The SEC data suggested that modularisation of HA1 on the C-terminus of VP1dC leads to the

recovery of stable capsomeres as compared to poor capsomere recovery for the N-terminal modular

capsomere module. Based on the structure of VP1dC, it is likely that site C is more flexible than site

N, thus allowing HA1 to form a stable domain away from VP1dC. In contrast, modularising HA1 on

site N might disrupt structure of the capsomere or might cause a steric effect between each protein

domain leading to protein misfolding. Therefore, CaptHA1-3C was selected for subsequent study.

62

Figure 3-3: Characterisation of modular capsomeres by size exclusion chromatography, showing

the capsomeres separated from GST proteins and protein aggregates. (A) Size exclusion

chromatogram of VP1dC without inserts. (B)-(G) Size exclusion chromatograms of CapHA1-N,

CaptHA1-2N, CaptHA1-3N, CapHA1-C, CaptHA1-2C and CaptHA1-3C, respectively.

Purified CaptHA1-3C was characterised using analytical SEC (SEC-HPLC) (Figure 3-4A), SDS-

PAGE (Figure 3-4B) and Western blot analysis (Figure 3-4C). The SEC chromatogram shows a

capsomere peak eluted at the expected time (approximately 22 min) and the presence of a small

amount of higher molecular weight, which can be misfolded capsomeres or the oligomer form of

capsomeres, while protein aggregates were eluted as an excluded peak at approximately 16 min as

expected. Theoretically calculated molecular weights of VP1dC capsomere and CaptHA1-3C are

37.6 and 63.1 kDa, respectively, are shown in Figure 3-4B. The protein bands of correct molecular

weight were detected. Western blot analysis with mouse anti-HA sera shows specific detection of

63

CaptHA1-3C, while neither pre-induction host proteins nor VP1dC capsomere were detected (Figure

3-4C).

Receptor binding properties of purified CaptHA1-3C was analysed using a hemagglutination assay

(Figure 3-4D). Weak hemagglutination (endpoint titre of 4) was observed for VP1dC capsomere.

Strong hemagglutination was observed for CaptHA1-3C (endpoint titre of 1024), which was higher

than HA1 protein (endpoint titre of 128). The results suggested that, most likely, purified CaptHA1-

3C presented a structure containing sialic acid receptor-binding domains maintaining its ability to

bind to sialic acid on the surface of red blood cells.

Figure 3-4: Analysis of purified CaptHA1-3C. (A) Size exclusion chromatogram showing the

capsomeres separated from protein aggregates. (B) SDS-PAGE and (C) Western blot analysis with

HA specific antibody. Lanes: (L) molecular weight marker, (1) pre-induced lysate, (2) VP1dC and

(3) CaptHA1-3C (tHA1-3C). (D) Hemagglutination assay of VP1dC, CaptHA1-3C and HA1 proteins.

Two-fold serial dilutions of each protein are as indicated. The bars represent the endpoint titre of

each protein.

Purified and characterised CaptHA1-3C was tested for its immunogenicity in chickens. Figure 3-5

shows the HA1 specific antibody endpoint titres of sera from chickens following two intramuscular

64

immunisations with CaptHA1-3C, VP1dC capsomere as a negative control and HA1 proteins as a

positive control. All protein samples were adjuvanted with aluminum hydroxide (Alhydrogel). The

ELISA result shows that CaptHA1-3C induced a mean antibody endpoint titre of almost 105 after the

second immunisation (boost), comparable to HA1 proteins and significantly higher than VP1dC (p <

0.0001). This result indicated that microbial-based CaptHA1-3C produce similar anti-HA1 endpoint

titres as the insect cell-produced HA1.

Figure 3-5: HA1-specific antibody endpoint titre induced in chickens immunised with CaptHA1-

3C, VP1dC capsomere and HA1 protein. Lines represent mean titre. Statistical analysis of antibody

titres after 2nd immunisations (boost) is presented. **** = p < 0.0001; ns = not significant.

CaptHA1-3C containing HA1(45-262) from influenza virus A/chicken/NSW/3122-1/2012 (H7N7)

(H7N7CaptHA1-3C) was tested for its protective efficacy, as shown in Figure 3-6. Chickens that had

been immunised twice, 21 days apart, were totally protected against a mucosal challenge with the

homologous live virus compared to non-immunised or VP1dC-immunised chickens (Figure 3-6A),

which had higher levels of morbidity, mortality and histopathological lesions (encephalitis and/or

interstitial nephritis and/or myocarditis) (Figure 3-6B and 3-6C). Lesions were associated with viral

nucleoprotein antigen. Four of six of the immunised chickens seroconverted by Day 35, with HI titre

(using homologous antigen) ranging from 1:8 to 1:32. None of the unimmunised or VP1dC-

immunized chickens seroconverted against HA1.

65

Figure 3-6: Protection of chickens following immunisation and viral challenge. (A) Survival curves

(6 chickens per group). (B) Number of sick animals during 14-day period after challenge. (C)

Numbers of chickens in each group with (hatched bar) and without (clear bar) tissue lesions and

antigen.

This study demonstrates that stable modular capsomeres comprising influenza HA1 can be produced

in E. coli, which can be used as a fast, low-cost, and high yield expression system [15] without the

need for protein refolding. This modular capsomere induced strong antibody responses when

administered with aluminium hydroxide to chickens and induced full protection against challenge

with HPAI H7N7 virus. Aluminium hydroxide was chosen as an initial adjuvant allowing comparison

with previous capsomere benchmarks; subsequent work will explore animal-relevant adjuvants. The

results presented here indicate that this VP1 capsomere platform for bacterially produced-modular

capsomere could potentially translate into a rapid-response and low-cost vaccine manufacturing

technology suitable for poultry vaccination. Since the development of vaccines for veterinary use,

66

especially in agriculture sectors, has become more industrial and price sensitive [41], minimising the

cost of vaccine manufacture could contribute to low-cost vaccines. Further study is needed to explore

simpler methods of manufacturing poultry influenza vaccine while retaining immunogenicity and

protective efficacy. The modular capsomere combined with other adjuvants suitable for poultry

vaccination, such as oil emulsions [42] will be explore as well.

3.5 Acknowledgements

This project is funded by the Australian Research Council (ARC Discovery Project DP160102915).

The authors thank Arun Kumar, Thanh Tam Doan, Nicolas Pichon, Andrea Schaller, and Arjun Seth

for their technical support in the in vivo immunogenicity study, Jeff Butler for preparation and

characterisation of the virus challenge, the Animal Studies Team and the Histology Team of AAHL

for animal infection trial and histological preparations, respectively, and Julie Cooke for HI serology.

3.6 Conflict of interest

The University of Queensland (UQ) filed patents on the use of MuPyV as a vaccine platform.

L.H.L.L. and A.P.J.M. contributed to those patents and, through their employment with UQ, hold an

indirect interest in this intellectual property.

67

3.7 References

1. World Health Organization, Cumulative number of confirmed human cases of avian influenza

A (H5N1) reported to WHO. 2014 [cited 10.05.14]; available from:

http://www.who.int/influenza/human_animal_interface/en/.

2. Fry, E., What the worst bird flu outbreak in U.S. history means for farms. 2015 [cited

13.12.15]; available from: http://fortune.com/2015/06/25/bird-flu-outbreak-farms/.

3. McKenna, M., Bird flu cost the US $3.3 billion and worse could be coming. 2015 [cited

02.10.15]; available from: http://phenomena.nationalgeographic.com/2015/07/15/bird-flu-2/.

4. Kuethe, J.N.a.T. Economic implications of the 2014-2015 bird flu. 2015 [cited 12.05.16];

available from: http://farmdocdaily.illinois.edu/2015/06/economic-implications-of-the-2014-

2015-bird-flu.html.

5. Marano, N., C. Rupprecht, and R. Regnery, Vaccines for emerging infections. Revue

scientifique et technique (International Office of Epizootics), 2007. 26(1): p. 203-215.

6. Murugan, S., et al., Recombinant haemagglutinin protein of highly pathogenic avian influenza

A (H5N1) virus expressed in Pichia pastoris elicits a neutralizing antibody response in mice.

Journal of Virological Methods, 2013. 187(1): p. 20-25.



Virology, 2011. 85(3): p. 1246-1256.



9. Bommakanti, G., et al., Design of an HA2-based Escherichia coli expressed influenza

immunogen that protects mice from pathogenic challenge. Proceedings of the National


10. Jegerlehner, A., et al., Bacterially Produced Recombinant Influenza Vaccines Based on Virus-

Like Particles. PloS One, 2013. 8(11): p. e78947.

11. Peiris, J.M., M.D. De Jong, and Y. Guan, Avian influenza virus (H5N1): a threat to human

health. Clinical Microbiology Review, 2007. 20(2): p. 243-267.





68

14. Lipin, D.I., et al., Affinity purification of viral protein having heterogeneous quaternary

structure: Modeling the impact of soluble aggregates on chromatographic performance.

Journal of Chromatography A, 2009. 1216(30): p. 5696-5708.





Vaccine, 2014. 32(29): p. 3651-3655.







e2257.

20. Lua, L.H.L., et al., Synthetic biology design to display an 18kDa rotavirus large antigen on a

modular virus-like particle. Vaccine, 2015. 33(44): p. 5937-5944.


Biotechnology, 1999. 17(7): p. 691-695.



23. Stevens, J., et al., Structure of the uncleaved human H1 hemagglutinin from the extinct 1918

influenza virus. Science, 2004. 303(5665): p. 1866-1870.



25. Yu, X., et al., Neutralizing antibodies derived from the B cells of 1918 influenza pandemic

survivors. Nature, 2008. 455(7212): p. 532-536.


thermodynamic attraction between structural protein capsomeres. Journal of the Royal


27. Connors, N.K., et al., Improved fusion tag cleavage strategies in the downstream processing

of self-assembling virus-like particle vaccines. Food and Bioproducts Processing, 2014. 92(2):

p. 143-151.

69

28. Laemmli, U.K., Cleavage of structural proteins during the assembly of the head of

bacteriophage T4. Nature, 1970. 227(5259): p. 680.



5960-5965.

30. World Organization for animal Health, Avian influenza in Manual of Diagnostics Tests and

Vaccines for Terrestrial Animals 2016. 2016 [cited 2016 26.10.2016]; available from:

http://www.oie.int/international-standard-setting/terrestrial-manual/access-online/.

31. Wibawa, H., et al., The pathobiology of two Indonesian H5N1 avian influenza viruses

representing different clade 2.1 sublineages in chickens and ducks. Comparative

Immunology, Microbiology and Infectious Disease, 2013. 36(2): p. 175-191.

32. Bingham, J., et al., Infection studies with two highly pathogenic avian influenza strains

(Vietnamese and Indonesian) in Pekin ducks (Anas platyrhynchos), with particular reference

to clinical disease, tissue tropism and viral shedding. Avian Pathology, 2009. 38(4): p. 267-

278.

33 Zhao, H.L., et al., Increasing the homogeneity, stability and activity of human serum albumin

and interferon-α2b fusion protein by linker engineering. Protein Expression and Purification,

2008. 61(1): p. 73-77.

34. Salunke, D.M., D.L. Caspar, and R.L. Garcea, Self-assembly of purified polyomavirus capsid

protein VP1. Cell, 1986. 46(6): p. 895-904.

35. Hartinger, D., et al., Enhancement of solubility in Escherichia coli and purification of an

aminotransferase from Sphingopyxis sp. MTA 144 for deamination of hydrolyzed fumonisin

B 1. Microbial Cell Factories, 2010. 9: p. 62-62.

36. Graves, P.N., et al., Preparation of influenza virus subviral particles lacking the HA1 subunit

of hemagglutinin: Unmasking of cross-reactive HA2 determinants. Virology (New York,

N.Y.), 1983. 126(1): p. 106-116.

37. Brownlee, G.G. and E. Fodor, The Predicted Antigenicity of the Haemagglutinin of the 1918

Spanish Influenza Pandemic Suggests an Avian Origin. Philosophical Transactions of the

Royal Society of London. Series B Biological Sciences, 2001. 356(1416): p. 1871-1876.


subtype). Cell, 1982. 31(2): p. 417-427.

39. Igarashi, M., et al., Predicting the Antigenic Structure of the Pandemic (H1N1) 2009 Influenza

Virus Hemagglutinin. PloS One, 2010. 5(1): p. e8553.

40. Baneyx, F. and M. Mujacic, Recombinant protein folding and misfolding in Escherichia coli.

Nature Biotechnology, 2004. 22(11): p. 1399-1408.

70

41. Heldens, J., et al., Veterinary vaccine development from an industrial perspective. The

Veterinary Journal, 2008. 178(1): p. 7-20.

42. Hwang, S.D., et al., Single dose of oil-adjuvanted inactivated vaccine protects chickens from

lethal infections of highly pathogenic H5N1 influenza virus. Vaccine, 2011. 29(11): p. 2178-

2186.

71

Chapter 4

Characterisation of modular capsomeres from purification modalities

The entire chapter consists of the journal article submitted as:

Jarurin Waneesorn, Evelyne Deplazes, Nani Wibowo, John Bingham, Anton P.J. Middelberg, and

Linda H.L. Lua (2017). Characterisation of modular capsomeres from purification modalities.

Submitted to Molecular Biotechnology.


- Page, figure, section and subsection numbering and reference style were changed to be


72

4.1 Abstract

Economic losses resulting from highly pathogenic avian influenza outbreaks have soared into the

billions, directly affecting poultry farmers and having a significant ripple effect on downstream

businesses. Virus transmission is primarily animal to animal; however, increasing reports of animal

to human transmission are raising major concerns for human health. Development of a poultry

vaccine would not only prevent avian influenza outbreaks and associated socio-economic losses, but

also minimise the threat of virus spread to humans. We have previously demonstrated the

effectiveness of a modular capsomere presenting structure-based designed influenza hemagglutinin

(CaptHA1-3C; hereafter CaptHA1) produced in Escherichia coli (E. coli) for application in poultry

vaccination. The highly purified CaptHA1 vaccine candidate (cCaptHA1) induced complete

protection against virus challenge in administered chickens, but the costs associated with the process

make it impractical for low-cost markets. In this study, E. coli-produced CaptHA1 prepared as

partially purified CaptHA1 (aCaptHA1) and crudely purified CaptHA1 (pCaptHA1) were evaluated

for their immunogenicity, to investigate whether less pure vaccine candidates can deliver

immunogenicity equivalent to high-purity capsomeres. All three adjuvanted capsomere preparations

induced high antibody titers (~105 endpoint titre) with hemagglutinin inhibition (HI) activity (1024,

2048 and 32 endpoint titres for cCaptHA1, aCaptHA1 and pCaptHA1, respectively) after two doses

of immunisation. HI titers obtained for aCaptHA1 and pCaptHA1 suggested that immunisation with

these vaccine formulations could induce protection similar to cCaptHA1. The data presented here

demonstrate that a simplified purification process may be sufficient to generate an effective and low-

cost vaccine suitable for poultry vaccination.

4.2 Introduction

Avian influenza (AI) refers to a severe respiratory disease resulting from infection with avian

influenza A virus. The highly pathogenic form of AI (HPAI), caused mainly by H5 and H7 subtypes,

demonstrates high transmission and mortality rates in birds [1, 2]. The loss of millions of birds during

recent HPAI outbreaks greatly affected poultry production, processing and downstream businesses,

causing catastrophic economic loss on a global level [2-6].

73

The One Health Initiative recognises the connection between the health of humans, animals and

ecosystems and raises concerns regarding virus transmission between animals and humans [7].

Indeed, human infections with AI virus have been reported worldwide since 1997 [8], with the most

recent cases reported in January 2017 [9]. The global spread of the disease highlights the potential

pandemic spread of AI.

While most AI viruses are considered low pathogenic viruses (LPAI), viruses circulating in domestic

poultry can mutate from LPAI to HPAI by means of viral genome drift and shift [2]. Vaccination of

poultry against LPAI and HPAI presents a powerful tool to prevent AI outbreaks, reduce virus

transmission and mitigate economic loss. Nevertheless, current vaccines against avian influenza A

and the technologies used to produce them are more focused on human use. For veterinary vaccines,

major considerations are cost per dose and the effectiveness of a single-dose administration [10].

Previously, we have demonstrated the effectiveness of a modular capsomere-based vaccine produced

in E. coli for application in poultry. The modular capsomere presenting influenza hemagglutinin

engineering using structure-based design (CaptHA1) was produced as a stable modular capsomere

without the need for refolding [11]. A highly purified CaptHA1 (cCaptHA1) vaccine induced

complete protection when administered to chickens. This cCaptHA1 demonstrated promise as a

capsomere-based vaccine against AI. Notwithstanding, the use of cCaptHA1 as a cost-effective

poultry vaccine was hindered by the high costs associated with chromatographic purification

methods, which contribute more than 70% of the total operating cost of vaccine production [12].

Moreover, the recovery of cCaptHA1 prepared in this process was relatively low. A previous study

demonstrated that immunisation with crudely purified rotavirus-like particles containing certain

levels of contaminants resulted in immunogenicity and protective efficacy in mice [13]. In addition,

Wibowo et al. [14] demonstrated the high immunogenicity of precipitated capsomeres (CapM2e),

containing soluble aggregates and E. coli protein contaminants, as vaccine candidates against

influenza. This result suggests that soluble aggregates and other bacterial contaminants presented in

CapM2e do not affect vaccine immunogenicity [14]. With this result, simpler selective precipitation

methods may offer an alternative method of purification for CaptHA1.

To simplify purification steps without diminishing vaccine immunogenicity, it is essential to

investigate which vaccine components apart from the active component or vaccine antigen can remain

74

in the vaccine. In this study, we investigated vaccine candidates of varying purity for use in poultry.

Three CaptHA1 preparations, highly purified (cCaptHA1) containing mainly pure capsomeres,

partially purified CaptHA1 (aCaptHA1) containing soluble aggregates and capsomeres and crudely

purified CaptHA1 (pCaptHA1) containing soluble aggregates, capsomeres and other E. coli protein

contaminants, were prepared. Each preparation was evaluated for its safety and efficacious use as a

poultry vaccine candidate.


4.3.1 Plasmid construction and protein expression

Plasmid for the expression of GST-tagged modular capsomere presenting HA1(44-268) from influenza

A strain A/Puerto Rico/8/34 (H1N1) (PR8 H1N1), designated as GST-CaptHA1, was generated as

previously reported [11]. The GST-CaptHA1 was expressed in E. coli Rosetta (DE3) pLysS

(Novagen, CA, USA) as previously described [11].

4.3.2 Homology modelling

The homology model of CaptHA1 was created using Modeller v9.10 [15].The structure of the

pentameric VP1dC was used as a template. The structure of CaptHA1 was constructed using the

polyomavirus PDB ID 1SID [16] and influenza hemagglutinin PDB ID 1RUZ as described in Lua et

al. 2015 [17].

4.3.3 Preparation of highly purified CaptHA1 (cCaptHA1)

cCaptHA1 was prepared using GST-affinity chromatography followed by size exclusion

chromatography (SEC) as previously described [11]. GST was removed using TEVp as previously

described [11]. Endotoxin was removed from the protein sample by anion exchange using Vivapure

Q Mini H spin column (Sartorius Stedim, France) as previously described [11].

4.3.4 Preparation of partially purified CaptHA1 (aCaptHA1)

aCaptHA1 was prepared as follows. GST-CaptHA1 was initially captured by GST-affinity

chromatography as previously described [11]. Endotoxin was subsequently removed from the protein

sample by anion exchange using quaternary ammonium Sepharose resin (Q Sepharose resin; GE

75

Healthcare, Sweden). GST-CaptHA1 was mixed with Q Sepharose resin at a 10:1 (w/v) ratio of

protein to Q Sepharose resin. The mixture was slowly rotated at 4 C for 30 min, and the supernatant

was collected after centrifugation at 500 x g, 4 C for 5 min. This step was repeated five times. The

GST tag was cleaved from the fusion protein by enzymatic reaction using TEVp prior to the

separation of CaptHA1 from GST by SEC as previously described [11], except both fractions

corresponding to modular capsomere and aggregates peaks were collected. These fractions were

pooled to obtain aCaptHA1.

4.3.5 Preparation of crudely purified CaptHA1 (pCaptHA1)

Cell pellets containing GST-tagged CaptHA1 were resuspended in 40 mL L-buffer (40 mM Tris, 500

mM NaCl, 1 mM EDTA, 5% (v/v) glycerol, 5 mM DTT, pH 8.0). Cells were lysed by sonication

(Branson Sonifier 450 ultrasonicator, Branson Ultrasonics, CT, USA) on ice at an energy output of

60 W for four bursts of 40 seconds with at least 1 min pause between each burst. Soluble lysate was

separated from inclusion bodies and other cellular debris by centrifugation at 25,000 x g, 4 C for 20

min. The GST tag was removed enzymatically by TEVp as previously described [11], except L-buffer

was used. Soluble protein was precipitated with 1.0 M Na2SO4 for 2 h at room temperature. Following

centrifugation (22,000 x g, 4C for 15 min), the supernatant was removed and the precipitate

resuspended with L-buffer and further centrifuged (22,000 x g, 4C for 5 min). The supernatant was

buffer exchanged using Sephadex G-25 column (GE Healthcare, UK), a flow rate of 0.5 mL. min-1

and pre-equilibrated with Tris-NaCl buffer (40 mM Tris, 300 mM NaCl, pH 8.0). Endotoxin was

removed from protein sample by anion exchange using quaternary ammonium Sepharose resin (Q

Sepharose resin; GE Healthcare, Sweden) as described in Section 4.3.4.

4.3.6 Capsomere characterisation

Qualitative analysis of protein samples was performed using standard SDS-PAGE using 10% gel

[18], Western blot analysis with mouse anti-HA antibody and analytical SEC with Superdex 200

30/100 GL column (GE Healthcare, UK), as previously described [11].


Hemagglutination assay was performed as previously described [11].

76

4.3.8 Immunisation

Four groups of eight Gallus gallus layer chickens (Bond Enterprises, Australia) were cared for

humanely in accordance with The University of Queensland (UQ) Animal Ethics Committee

guidelines. All animal experimental work was reviewed and approved by The UQ Animal Ethics

Committee (AIBN/235/14/QSF).

All groups of chickens at three weeks of age were immunised with two doses of 70 µg of total protein

via the intramuscular route on experimental Day 0 and Day 21. Blood samples were collected from

the wing vein at Day 0, Day 21 and Day 35. The four groups received either VP1dC capsomeres,

cCaptHA1, aCaptHA1 or pCaptHA1. All samples were mixed with aluminium hydroxide

(Alhydrogel, Brenntag, Germany) at a 1:1 (v/v) ratio of protein to Alhydrogel for 1 h at room

temperature. Protein concentrations were determined using a Coomassie (Bradford) Protein Assay

Kit (Thermo Fisher Scientific).

4.3.9 ELISA

ELISA was performed to detect antibodies specific for HA1, as previously described [14].

4.3.10 Hemagglutination inhibition assay

Hemagglutination inhibition (HI) assay was performed according to OIE Terrestrial Manual 2016

(Chapter 2.3.4, p. 10-11) [19]. In brief, sera were serially diluted twofold in 1 x PBS pH 7.2-7.4 from

an initial dilution of 1:4. Sera ( 25 µL) were incubated with 4 HA units of influenza A strain A/Puerto

Rico/8/34 (H1N1) (PR8 H1N1) virus at room temperature for 1 h. An equal volume (25 µL) of 1.1

% chicken red blood cells was added and incubated at 4 C for 45-60 min. The HI titre was determined

as the highest dilution of serum that inhibited hemagglutination. HI values were presented as the

geometric mean with 95% confidence interval.

4.4. Results and discussion

The influenza virus hemagglutinin (HA) has a key function in virus binding and facilitation of virus

entry into the target cells. Bacterially-produced globular head (HA1) of the HA has been shown to

induce protective immunity in ferrets [20, 21]. The efficacy of a modular capsomere presenting

77

structure-based designed HA1 (CaptHA1) as a poultry influenza vaccine candidate was previously

evaluated [11]. Highly purified CaptHA1 (cCaptHA1) was highly immunogenic in chickens, and

when immunised with adjuvanted antigen, chickens were 100% protected against challenge with a

HPAI virus strain H7N7 [11].

CaptHA1 comprises five copies of HA1(44-268) (tHA1), each molecularly-fused to the C-terminus of

the capsid protein VP1dC (VP1(1-320)), as illustrated in Figure 4-1. Preventing structure perturbation

of the base module VP1dC by the inserted module and maintaining the conformational structure of

the antigenic module tHA1 are important for eliciting immunity against the targeted pathogen. A

GSA (GSAGSAAGSGEF) linker [22] was inserted between the base module VP1dC and antigenic

module tHA1 to provide flexibility and prevent a steric effect between the protein domains. Structure-

based designed CaptHA1 was previously described [11] to show the modularisation of antigenic

modules without compromising the structure integrity of the capsomere.

Figure 4-1: Homology model of modular capsomere presenting influenza HA1(44-268) (tHA1). Five

copies of modular VP1dC-tHA1 form a modular capsomere (CaptHA1). tHA1 is illustrated in red,

VP1dC is shown in grey and GSA linker (at the C-terminal of VP1dC) is shown in green. The

homology model was created using Modeller v9.10.

The established method [11] for the production of cCaptHA1 (Figure 4-2A) yielded modular

capsomeres displaying conformational HA1 that induced protective immunity in chickens. However,

the recovery of modular capsomeres was relatively low using this process. To investigate the

78

possibility of simplifying the processing steps required to minimally obtain immunogenic modular

CaptHA1 for poultry vaccination, two additional methods (Figure 4-2B and 4-2C) were used to obtain

less pure CaptHA1 preparations. One chromatographic method was used to obtain partially purified

CaptHA1 (aCaptHA1) containing capsomeres and soluble aggregates (Figure 4-2B). The third

method, selective salting-out precipitation, was used to obtain crudely purified CaptHA1 (pCaptHA1)

containing capsomeres, soluble aggregates and E. coli protein contaminants (Figure 4-2C).

Figure 4-2: Major processing steps used to produce CaptHA1. (A) Highly purified CaptHA1

(cCaptHA1) containing pure capsomeres. (B) Partially purified CaptHA1 (aCaptHA1) containing

capsomeres and protein aggregates. (C) Crudely purified CaptHA1 (pCaptHA1) containing

capsomeres, protein aggregates and host-cell proteins.

The cCaptHA1, aCaptHA1 and pCaptHA1 were characterised using analytical SEC (Figure 4-3A),

SDS-PAGE (Figure 4-3B) and Western blot analysis (Figure 4-3C). Figure 4-3A shows the analytical

SEC chromatograms of all three samples obtained. The cCaptHA1 contained mainly pure

capsomeres, aCaptHA1 contained capsomeres and soluble aggregates and pCaptHA1 contained

capsomeres, soluble aggregates and small molecular weight contaminants. Modular CaptHA1 has a

predicted molecular weight of 90.5 kDa and was eluted at approximately 22 min. A shoulder on the

left of the capsomere peak was observed for cCaptHA1 and aCaptHA1. This suggests that oligomeric

79

capsomeres or misfolded capsomeres or capsomeres in different conformational states were present

in the samples. The aCaptHA1 has significantly more soluble aggregates of capsomeres (eluted at

approximately 16 min) compared with cCaptHA1. For pCaptHA1, only a small peak was detected at

the elution time of modular capsomeres, but a large peak corresponding to soluble aggregates was

observed. Low molecular weight contaminants, which are likely to be host cell proteins, were detected

on the chromatogram (elution time at approximately 31 min) and SDS-PAGE gels (Figure 4-3B, Lane

3). Western blot analysis (Figure 4-3C) with anti-HA mouse sera confirmed that the protein bands at

63 kDa are monomeric VP1dC-tHA1. As observed on both SDS-PAGE gels and Western blot images,

the amount of CaptHA1 is lower in pCaptHA1 preparations compared with aCaptHA1 and cCaptHA1

preparations because the same protein mass of each sample was loaded onto the gel.

Figure 4-3: Characterisation of cCaptHA1, aCaptHA1 and pCaptHA1. (A) Analytical size

exclusion chromatograms of cCaptHA1, aCaptHA1 and pCaptHA1. Peaks corresponding to

aggregates, capsomeres and contaminants are indicated. All three samples were analysed on SDS-

PAGE gel (B) and Western blot (C) with mouse sera containing HA antibodies. Lanes: (L) molecular

weight marker, (1) cCaptHA1, (2) aCaptHA1 and (3) pCaptHA1.

Hemagglutination assay was conducted with all samples to test the binding of hemagglutinin to sialic

acid receptors. Hemagglutination titres of cCaptHA1, aCaptHA1and pCaptHA1 are indicated in

Figure 4-4A. Strong hemagglutination with an endpoint titre of 1024 was observed for high purity

cCaptHA1. A higher endpoint titre of 2048 was observed for aCaptHA1, while pCaptHA1 had an

endpoint titre of 32. Weak agglutination (endpoint titre of 4) was observed for VP1dC capsomeres,

as expected (data not shown). The results suggest that aCaptHA1 and pCaptHA1 presented

conformational HA1 with functional sialic acid-binding domains. The lower endpoint titre obtained

80

for pCaptHA1 is possibly a result of the lower amount of CaptHA1 in the protein sample. A

comparison of the homology models of CaptHA1 and trimeric HA suggests that a modular capsomere

with antigenic HA1 modules resembles a native trimeric HA (Figure 4-4B). Both the comparative top

and side views show the similarity of the globular head structure of HA1 between a modular

capsomere and trimeric HA. Previous findings have shown that stabilised trimeric HA, but not the

dimeric or monomeric forms, retain the ability to bind sialic acid on red blood cells, allowing

agglutination [21, 23]. This suggests that sialic acid-binding requires a quaternary structure composed

of a trimer of HA, and because the modular capsomere also causes hemagglutination, it may orient

the HA1 molecules in such a way that it mimics the relevant site in the HA trimer.

Figure 4-4: Functional characterisation and homology modelling of modular capsomeres. (A)

Hemagglutination assay of cCaptHA1, aCaptHA1, and pCaptHA1. Lines represent endpoint titres.

(B) Homology models of CaptHA1 and native trimeric HA protein (HA PDB file 1RUZ.pdb).

81

Well-characterised cCaptHA1, aCaptHA1and pCaptHA1 were tested for their immunogenicity in

chickens. Figure 4-5A illustrates HA1-specific antibody endpoint titres after the first immunisation

(prime) and the second immunisation (boost) with Alhydrogel-adjuvanted vaccine candidates. No

significant change was observed in the endpoint titre of antibody response (p < 0.05) between the

prime and boost for groups immunised with cCaptHA1, aCaptHA1 and VP1dC capsomere. For

pCaptHA1, the antibody endpoint titre increased from 103 (prime immunisation) to above 105 after

the boost immunisation. The lower titre after prime immunisation is likely caused by the lower ratio

of capsomeres to total protein in pCaptHA1 compared with the other two groups, cCaptHA1 and

aCaptHA1, as shown in Figure 4-5A. Notably, although there were fewer capsomeres in the

pCaptHA1 preparation, there was no difference between the titres of all groups after the boost. The

increase of antibody titre obtained from pCaptHA1 to the same level as cCaptHA1 and aCaptHA1

after the boost suggested fewer capsomeres in the boost still initiated a ‘maximal’ antibody titre.

However, the presence of aggregates and bacterial contaminants in the pCaptHA1 preparation may

also contribute to the increase in the antibody titre after the boost immunisation. Although aCaptHA1

contained fewer capsomeres but more soluble aggregates than cCaptHA1, similar immunogenicity

was obtained for both samples. This suggests that soluble aggregates form macromolecules that retain

antigenic properties to enhance vigorous immune responses, possibly via protein aggregate-mediated

B cell receptor cross-linking, as has been demonstrated in previous studies [24-26]. The bacterial

contaminants may also elicit the immune response through germline-encoded pattern recognition

receptors that recognise the bacterial components and lead to the activation of immune systems [27].

The data also suggest that a single dose at 70 g may be sufficient to induce protective immunity in

chickens, as the boost immunisation did not significantly elevate antibody titres for vaccine groups

cCaptHA1 and aCaptHA1. However, a single dose of 70 g pCaptHA1 may not be sufficient to

induce protective immunity in chickens.

82

Figure 4-5: Immunogenicity of modular capsomeres. (A) HA1-specific antibody titres induced in

chickens following two intramuscular immunisations with VP1dC capsomere and modular

capsomeres. Statistical analysis is presented between prime and boost titres and antibody titres after

the second immunisation (boost). **** = p < 0.0001; ns = not significant. (B) Hemagglutination

inhibition (HI) assay results showing HI titres of prime and boost sera following immunisation with

cCaptHA1, aCaptHA1 and pCaptHA1. Geometric mean with 95% confidence interval is presented.

Endotoxin levels of cCaptHA1, aCaptHA1 and pCaptHA1 were 225, 39023, and 59326 EU/mL,

respectively. Although endotoxin can evoke a pathophysiological effect through activation of the

immune system [28], the total amount of endotoxin per dose of injection (56.25, 9755.75 and 14831.5

EU for cCaptHA1, aCaptHA1 and pCaptHA1, respectively) was low compared with the endotoxin

content in bacterially-produced commercial vaccines. Typhus and cholera vaccines have reported

endotoxin levels of 1x105 and 1x106 EU/mL, respectively [29]. It is likely that chickens are less

susceptible to endotoxins compared with mammals [30]. The amounts of endotoxin were not toxic,

and no adverse effects were observed.

Hemagglutination inhibition (HI) test was conducted as a surrogate for vaccine efficacy in this study.

HI titres have been correlated to seroprotection by HI antibodies that inhibit the binding of influenza

hemagglutinin to sialic acid on target cells, preventing virus infectivity [31]. An HI titre of 16 against

4 HA units of antigen is considered positive for the immunoprotection of challenge chickens [19].

Nonetheless, HI titres ranging from 1:8 to 1:32 (geometric mean titre = 8) provided protection in

chickens challenged with the homologous virus [11]. Figure 4-5B illustrates HI titres following prime

and boost vaccinations with cCaptHA1, aCaptHA1 or pCaptHA1. After boost immunisation, HI titres

83

of cCaptHA1, aCaptHA1 and pCaptHA1 ranged from 1:4 to 1:256 (geometric mean = 20.7), 1:8 to

1:512 (geometric mean = 69.8) and <1:4 to 1:256 (geometric mean = 17.7), respectively. VP1dC

capsomere generated low HI titres (geometric mean = 1.3), as expected. The results suggest that HI

antibodies obtained from aCaptHA1 and pCaptHA1 are likely to confer protection, as was obtained

from the immunisation with cCaptHA1.

This study investigated the efficacy of partially purified aCaptHA1 and crudely purified pCaptHA1

as a vaccine antigen for AI compared with highly purified cCaptHA1, which was shown to confer

complete protection in immunised chickens in a previous study. The results show that presence of

soluble aggregates and low amounts of host cell contaminants do not affect the immunogenicity and

functionality of aCaptHA1 and pCaptHA1 after two immunisations. This warrants further

investigation to develop a simpler and cost-effective process to generate immunogenic modular

capsomere CaptHA1 that provides protection from AI.


This project is funded by the Australian Research Council (ARC Discovery Project DP160102915).

The authors thank Arun Kumar, Nicolas Pichon, Andrea Schaller, and Arjun Seth for their technical

support in the in vivo immunogenicity study, as well as Joanne Meers for valuable suggestions on

HA and HI assays.


The UQ filed patents on the use of MuPyV as a vaccine platform. L.H.L.L. and A.P.J.M. contributed

to those patents and, through their employment with UQ, hold an indirect interest in this intellectual

property. Other authors declare that there are no conflicts of interest.

84

4.7 References


Diseases, 2006. 12(9): p. 1319.

2. Spickler, A. R., Highly pathogenic avian influenza. 2016 [cited 03.03.17]; available from:

http://www.cfsph.iastate.edu/Factsheets/pdfs/highly_pathogenic_avian_influenza.pdf.

3. Foster, A., Bird flu outbreak MAPPED: avian influenza spreads across Europe—could the

UK be next? Sunday Express, 2016 [cited 04.03.17]; available from:

http://www.express.co.uk/news/world/740957/bird-flu-outbreak-map-avian-influenza-

Europe-spread-European-countries-UK-Britain.

4. Rushton, J., et al., Impact of avian influenza outbreaks in the poultry sectors of five South East

Asian countries (Cambodia, Indonesia, Lao PDR, Thailand, Viet Nam) outbreak costs,

responses and potential long term control. World’s Poultry Science Journal, 2005. 61(3): p.

491–514.

5. European Centre for Disease Prevention and Control, Outbreaks of highly pathogenic avian

influenza A(H5N8) in Europe. 2016; available from:

http://ecdc.europa.eu/en/publications/Publications/risk-assessment-avian-influenza-H5N8-

europe.pdf.

6. Akunzule, A., E. Koney, and M. Tiongco, Economic impact assessment of highly pathogenic

avian influenza on the poultry industry in Ghana. World’s Poultry Science Journal, 2009.

65(3): p. 517–528.

7. Fox, M., CDC concerned by H7N9 bird flu’s sudden spread in China. 2017 [cited 04.03.17];

available from: http://www.nbcnews.com/health/health-news/cdc-concerned-h7n9-bird-flu-

s-sudden-spread-china-n728946.

8. World Health Organization, Avian and other zoonotic influenza. 2016 [cited 02.03.17 ];

available from: http://www.who.int/mediacentre/factsheets/avian_influenza/en/.

9. World Health Organization, Human infection with avian influenza a(H7N9) virus—China.

2017 [cited 02.03.17 ]; available from: http://www.who.int/csr/don/27-february-2017-ah7n9-

china/en/.

10. Heldens, J., et al., Veterinary vaccine development from an industrial perspective. Veterinary

Journal, 2008. 178(1): p. 7–20.

11. Waneesorn, J., et al., Structural-based designed modular capsomere comprising HA1 for low-

cost poultry influenza vaccination. Vaccine, 2018. 36(22): p. 3064–3071.



85

13. Rodríguez-Limas, W.A., et al., Immunogenicity and protective efficacy of yeast extracts

containing rotavirus-like particles: a potential veterinary vaccine. Vaccine, 2014. 32(24): p.

2794–2798.



5960-5965.

15. Eswar, N., et al., Comparative protein structure modeling using MODELLER, in Current

Protocols in Protein Science. 2001, John Wiley & Sons.

16. Stehle, T. and S.C. Harrison, Crystal structures of murine polyomavirus in complex with

straight-chain and branched-chain sialyloligosaccharide receptor fragments. Structure,

1996. 4(2): p. 183-194.



18. Laemmli, U. K., Cleavage of structural proteins during the assembly of the head of


19. World Organisation for Animal Health, Avian influenza in manual of diagnostics tests and

vaccines for terrestrial animals 2016. 2016 [cited 26.10.2016]; available from:




2010. 5(7): p. e11548.



Virology, 2011. 85(3): p. 1246-1256.


Biotechnology, 1999. 17(7): p. 691-695.

23. Weldon, W. C., et al., Enhanced immunogenicity of stabilized trimeric soluble influenza

hemagglutinin. PLoS One, 2010. 5(9): p. e12466.

24. Braun, A., et al., Protein aggregates seem to play a key role among the parameters influencing

the antigenicity of interferon alpha (IFN-α) in normal and transgenic mice. Pharmaceutical

Research, 1997. 14(10): p. 1472–1478.

25. Rosenberg, A.S., Effects of protein aggregates: an immunologic perspective. The AAPS

Journal, 2006. 8(3): p. E501-E507.

26. Wang, W., et al., Immunogenicity of protein aggregates—concerns and realities. International

Journal of Pharmaceutics, 2012. 431(1): p. 1-11.

86

27. Akira, S., S. Uematsu, and O. Takeuchi, Pathogen recognition and innate immunity. Cell,

2006. 124(4): p. 783–801.

28. Magalhães, P. O., et al., Methods of endotoxin removal from biological preparations: a

review. Journal of Pharmacy and Pharmaceutical Sciences, 2007. 10(3): p. 388–404.

29. Brito, L.A. and M. Singh, Acceptable levels of endotoxin in vaccine formulations during

preclinical research. Journal of Pharmaceutical Sciences, 2011. 100(1): p. 34-37.

30. Roeder, D.J., M.-G. Lei, and D.C. Morrison, Endotoxic-lipopolysaccharide-specific binding

proteins on lymphoid cells of various animal species: association with endotoxin

susceptibility. Infection and Immunity, 1989. 57(4): p. 1054-1058.

31. Ohmit, S.E., et al., Influenza hemagglutination-inhibition antibody titer as a correlate of

vaccine-induced protection. Journal of Infectious Diseases, 2011. 24(12): p. 1879-1855.

87

Chapter 5

Simplified production and purification of modular capsomere-based

vaccine for avian influenza

The entire chapter consists of the journal article submitted as:

Jarurin Waneesorn, Hayley K. Charlton Hume, Anton P.J. Middelberg and Linda H.L. Lua (2017).

Simplified production and purification of modular capsomere-based vaccine for avian influenza.

Submitted to Biotechnology and Bioengineering. (Manuscript number: 17-793)


- Page, figure, section and subsection numbering and reference style were changed to be


88

5.1 Abstract

Highly pathogenic avian influenza (HPAI) is an extremely infectious disease caused by avian

influenza A viruses. As a result of exceedingly high transmission and mortality rates, HPAI outbreaks

have been responsible for huge economic losses in the poultry industry in both developed and

developing countries. Although it mainly affects birds, the increasing reports of HPAI crossing the

species barrier have fuelled major concerns that zoonotic disease may ultimately lead to a devastating

influenza pandemic among humans. Given the persistent threat to human health and the increasing

economic effect on the poultry industry, there is a pressing need to address the underlying problem-

that is, the lack of effective control measures. Poultry vaccination significantly decreases the risk of

viral transmission and transfer to humans when the vaccine is matched to the pathogenic virus. In

conjunction with appropriate biosecurity means, it can be instrumental in controlling the rapid spread

of avian influenza. A recently developed capsomere-based influenza vaccine proved to be 100%

protective against HPAI H7N7 virus challenge in chickens. The production and chromatographic

process by which this vaccine was produced is not economically attractive for use in the cost-sensitive

poultry industry. Here, the capsomere vaccine candidate was re-engineered, and a simple, non-

chromatographic purification process was developed. The bacterially-produced, non-fusion-tagged

modular capsomere presenting a truncated form of the globular head of hemagglutinin (CaptHA1),

can be selectively purified by precipitation. Direct comparison with an unoptimised chromatographic

process reveals at least a thirty-fold increase in the final yield of CaptHA1 using the simplified

column-free process, as well as the retention of CaptHA1 functional properties. While further

optimisation of the current process may result in further improved yield and purity, this study

describes a simplified process that is suitable for a rapidly produced, low-cost industrial poultry

vaccine.

5.2 Introduction

Highly pathogenic avian influenza (HPAI) remains a major concern for the world economy and

human health. Outbreaks of HPAI result in a significant loss of poultry through infection-induced

mortality or culling (to control viral transmission), thus leading to widespread devastation throughout

the poultry industry [1, 2]. The increased prevalence of HPAI, combined with the mutation-prone

nature of the virus, greatly increases the potential for viral spread to humans, raising genuine concerns

89

of a human influenza pandemic [3, 4]. The lack of effective measures to control HPAI within the

avian population highlights the risk of future influenza pandemics and further economic loss and is

driving the search for a suitable poultry vaccine. Studies suggest that when used in conjunction with

biosecurity means, appropriate vaccine use would be a powerful tool to control the spread of avian

influenza [5, 6].

Virus-like particle (VLP) vaccine technology offers the potential to manufacture rapid response

vaccines [7]. A modular capsomere platform based upon the major structural protein of murine

polyomavirus, VP1, has been developed as a means of rapid vaccine delivery [8]. This platform,

termed VP1dC, lacks the 63 C-terminal residues of VP1 that are responsible for mediating its

assembly into mature VLPs. Consequently, assembly is limited to pentameric capsomeres comprising

five identical copies of VP1dC. VP1dC was recently engineered to display a structurally designed

truncation of the globular head of influenza hemagglutinin (tHA1) at its C-terminus for the production

of a modular capsomere (CaptHA1) influenza vaccine in Escherichia coli (E. coli) [9]. Consistent

with previous work on VP1, a GST-tag was fused to the N-terminus of the protein [9]. The resulting

GST-tagged capsomere (GST-CaptHA1) comprised five copies of GST-tagged VP1dC-tHA1.

Chromatographically purified modular capsomere (CaptHA1) demonstrated high immunogenicity

and conferred 100% protection against challenge with HPAI H7N7 virus in chickens, demonstrating

the potential of this molecular construct as an avian influenza vaccine [9].

The final yield of CaptHA1 was surprisingly low. These low protein expression levels were further

exacerbated by poor affinity binding of the macromolecular construct, as noted previously for

capsomeres fused with GST [10]. Yield declined further as a result of capsomere aggregation

following GST-removal in the standard buffer. This resulted in just 2% of modular capsomeres being

recovered under standard process conditions with this construct (unpublished data).

Purification using chromatography can account for up to three-quarters of the total operating cost of

vaccine production [11]. High costs associated with affinity media and the complexities of GST

removal [10] make the multistep chromatography method used to prepare CaptHA1 unsuitable for

the production of a low-cost poultry vaccine. Employing cheaper, non-chromatographic purification

methods would be enabling for low-cost poultry vaccine markets. Previous studies from our

laboratory have shown that capsomere-based vaccines can be crudely purified using a selective

90

salting-out precipitation method [12, 13]. Importantly, capsomeres purified by precipitation retain

their immunogenicity, alleviating the need for a GST-tag and subsequent chromatographic

purification. Selective precipitation is widely used in the downstream processing of vaccines and

other biological products. Using salts or polymers as precipitants, target proteins are selectively

separated and concentrated from whole lysed cell preparations. In contrast to highly purified

preparations using chromatography approaches, precipitation increases the presence of host

contaminants. However, a recent study demonstrates that such contaminants do not diminish, and

may actually enhance, vaccine efficacy [12].

In this study, the production and purification process of a CaptHA1 vaccine candidate was simplified

by eliminating GST fusion protein production and adopting a simple non-chromatographic

purification method. Molecular engineering techniques were undertaken to generate soluble and

stable, non-tagged capsomeres displaying tHA1. Following expression in E. coli, capsomeres were

purified by selective precipitation. This method increased the final capsomere yield by more than 30

times when compared with the previous method using GST-affinity chromatography. More

importantly, we demonstrate that the capsomere structural integrity is retained. This simplified

process contributes to the rapid production of a modular capsomere-based strain-matched vaccine to

the avian influenza virus.


5.3.1 Plasmid construction

Expression plasmid for GST-tagged modular capsomere presenting HA1(44-268), from influenza A

strain A/Puerto Rico/8/34 (H1N1), designated as GST-CaptHA1, was generated as previously

described [9]. Murine polyomavirus VP1dC sequence (VP1 sequence lacking 63 amino acids at its

C-terminal) was cloned into the first multiple cloning site (MCS1) of pETDuet-1 (Novagen, WI)

between the NcoI and NotI restriction sites. This construct was used for the expression of non-tagged

capsomeres (CapVP1dC). Plasmid for the expression of non-tagged capsomeres presenting HA1(44-

268) (CaptHA1) was generated by inserting the VP1dC-GSA-HA1(44-268) sequence, amplified from the

GST-CaptHA1 plasmid above, into the second multiple cloning site (MCS2) of pETDuet-1 between

the NdeI and PacI restriction sites. A construct for the dual expression of both VP1dC and VP1dC-

tHA1 proteins was generated by inserting the VP1dC sequence into the MCS1 of the VP1dC-tHA1

91

pETDuet-1 plasmid (described above) between the NcoI and NotI restriction sites. This construct was

used for the expression of mixed modular capsomeres (mCap). All cloned constructs were generated

by the Protein Expression Facility, The University of Queensland, Australia. DNA sequences of all

constructs were verified by DNA sequencing (AGRF, Brisbane, Australia).

5.3.2 Protein expression

GST-tagged modular capsomere protein (GST-CaptHA1) was expressed in E. coli strain Rosetta

(DE3) pLysS, as described previously [9, 14]. Expression constructs for CapVP1dC, CaptHA1 and

mCap were transformed separately into chemically competent E. coli strains; Rosetta (DE3) pLysS

cells (Novagen, San Diego, CA), Rosetta 2 (DE3) cells (Novagen, San Diego, CA), and SHuffle T7

Express cells (New England Biolabs GmbH, Germany). Transformed Rosetta (DE3) pLysS and

Rosetta 2 (DE3) cells were grown separately in baffled flask at 37 C, 180 rpm to an OD600 of 0.5

using Terrific Broth (TB) medium containing 50 µg.mL-1 ampicillin (aMResco, Solon, OH) and 34

µg.mL-1 chloramphenicol (Astral Scientific Pty Ltd, Gymea, NSW, Australia). For protein

expression, cultures were induced with 0.2 mM isopropyl--D-thiogalactopyranoside (IPTG) (Astral

Scientific Pty Ltd, Gymea, NSW, Australia), grown at 25 C and harvested after 16 h. Transformed

SHuffle T7 Express cells were grown separately at 37 C to OD600 of 0.5 using TB medium containing

50 µg.mL-1 ampicillin (aMResco, Solon, OH). For protein expression, cultures were induced with 1

mM IPTG, grown at 25 C and harvested after 16 h.

5.3.3 Chromatographically purified modular capsomere (GST-CaptHA1)

Chromatographically purified GST-CaptHA1 was prepared as described previously [9, 14]. Sample

was stored at -80 C until further use.

5.3.4 Crudely purified modular capsomere (CapVP1dC, CaptHA1 and mCap) using Na2SO4

Cell pellets from 50 mL cultures of CapVP1dC, CaptHA1 and mCap were resuspended separately in

15 mL L-buffer (40 mM Tris, 500 mM NaCl, 1 mM EDTA, 5% (v/v) glycerol, 5 mM DTT, pH 8.5).

Cells were lysed by sonication (Branson Sonifier 450 ultrasonicator, Branson Ultrasonics, CT, USA)

at an energy output of 60 W for four bursts of 40 sec, and cell lysates were clarified by centrifugation

(25,200 x g, 4C, 20 min). Soluble protein was precipitated at different concentrations of Na2SO4

92

(0.6, 0.8, 0.9 and 1 M) for 2 h at room temperature. Following centrifugation (25,200 x g, 4C, 15

min), the supernatant was removed and the precipitate resuspended with L-buffer. The resuspension

was further centrifuged (25,200 x g, 4C, 5 min), and the supernatant containing the soluble

precipitate was taken for further analysis.

5.3.5 Crudely purified modular capsomere (CapVP1dC, CaptHA1 and mCap) using PEG 6000

Cell pellets from 50 mL cultures of CapVP1dC, CaptHA1 and mCap were resuspended separately in

15 mL of L-buffer (40 mM Tris, 500 mM NaCl, 1 mM EDTA, 5% (v/v) glycerol, 5 mM DTT, pH

8.0). Cells were lysed by sonication as above, and cell lysates were clarified by centrifugation (12,000

x g, 4C, 20 min). Soluble protein was precipitated at different concentrations of polyethylene glycol

(PEG) 6000 (2, 5, 8 and 10% w/v) (Chem-Supply Pty Ltd) for 2 h at room temperature. Following

centrifugation (12,000 g, 4C, 30 min), the supernatant was removed and the precipitates resuspended

with L-buffer. The resuspension was further centrifuged (12,000 x g, 4C, 20 min) and the supernatant

was taken for further analysis. Protein concentrations were determined using Coomassie (Bradford)

Protein Assay Kit (Thermo Fisher Scientific).

5.3.6 Capsomere Characterisation

Qualitative analysis of all protein samples was performed using standard SDS-PAGE [15], Western

blot and dot blot analysis. Western blot was performed as previously described [9], except chicken

anti-HA1 antisera and horseradish peroxidase (HRP)-conjugated rabbit anti-chicken IgG (A9046;

Sigma-Aldrich, USA) were used as the primary and secondary antibody, respectively. For dot blot

analysis, antigens (4 µg) were applied onto a PVDF membrane, blocked with 5% skim milk powder

in 0.1% PBS-T buffer and probed with chicken anti-HA1 antisera. HRP-conjugated rabbit anti-

chicken IgG (A9046; Sigma-Aldrich) was used for detection via chemiluminescence.


Hemagglutination assay was conducted with starting protein concentration 0.4 mg.mL-1 as described

previously (Section 2.3.4, p. 10-11) [16].

93

5.4 Results and discussion

VP1dC capsomere platform is based upon the VP1 structural protein of murine polyomavirus. To

generate a capsomere-based influenza vaccine, VP1dC was recently engineered to present a

structurally designed truncation of the globular head of influenza hemagglutinin (tHA1) at its C-

terminus (Figure 5-1). Modularisation at this flexible site allows antigenic module tHA1 to form a

stable domain adequately separated from the base module VP1dC [9]. VP1dC-tHA1 was previously

expressed as a GST fusion proteins to enhance protein solubility and aid affinity purification. When

expressed in E. coli, this modular protein self-assembles into a capsomere composed of five

monomers of GST-tagged VP1dC-tHA1 (GST-CaptHA1) (Figure 5-1A). Low capsomere yield and

high costs associated with GST-affinity purification highlighted the need to optimise the production

process. To remove complex chromatography steps, the VP1dC platform was redesigned to present

tHA1 at its C-terminus without a GST-tag (VP1dC-tHA1), ultimately producing a non-tagged

capsomere (CaptHA1) (Figure 5-1B). The globular head of hemagglutinin is highly insoluble when

expressed in E. coli, and is reported to form inclusion bodies [17-19]. The strategy for producing

fusion protein GST-VP1dC-tHA1 was postulated to enhance the solubility of tHA1 in E. coli. Thus,

removing the GST solubility tag may increase protein misfolding of VP1dC-tHA1 and lead to

capsomere instability. We hypothesised that titrating the tHA1 content within each capsomere would

maintain the structural integrity and protein stability of CaptHA1. A dual expression strategy,

employed to reduce steric hindrance of a large rotavirus antigen VP8* on the surface of VP1

capsomere [13, 20], was used here to reduce the number of tHA1 modules per capsomere. The dual

expression construct was designed to co-express VP1dC and modular VP1dC-tHA1. The resulting

population of capsomeres, denoted mCap, potentially comprised various ratios of VP1dC:VP1dC-

tHA1 (5:0, 4:1, 3:2, 2:3, 1:4, 0:5), as shown in Figure 5-1C.

94

Figure 5-1: Engineering modular capsomere presenting tHA1 (CaptHA1). (A) GST-CaptHA1, each

modular capsomere before and after removal of GST-tag, contains five copies of GST-tagged-VPdC-

tHA1 and VP1dC-tHA1, respectively. (B) CaptHA1, each modular capsomere contains five copies of

VP1dC-tHA1. (C) mCap, modular capsomeres composed of different ratios of VP1dC to VP1dC-

tHA1 (0:5, 1:4, 2:3, 3:2, 4:1 or 5:0).

The expression of non-tagged CaptHA1 and mCap was examined in three E. coli strains, namely

Rosetta (DE3) pLysS, Rosetta 2 (DE3) and SHuffle T7 Express cells (Figure 5-2). High expression

of CaptHA1 and mCap was observed in all host strains. In both Rosetta strains, the CaptHA1 soluble

protein was significantly reduced compared with the total protein, suggesting that the CaptHA1

protein misfolded during expression (Figure 5-2A). In contrast, approximately 70% of the total

CaptHA1 protein was soluble in SHuffle T7 Express cells. This host strain is specifically engineered

to enhance disulfide bonding in the cytoplasm [21]; thus, it may have contributed to the correct

folding and stability of tHA1, which contains two disulfide bonds. High solubility of mCap was

observed in all host strains, with Rosetta 2 (DE3) showing the highest levels (Figure 5-2B). This

observation supports the hypothesis that co-expression of VP1dC and VP1dC-tHA1, which titrates

95

tHA1 content per capsomere, alleviates the steric effects of tHA1 within the capsomere. This allows

capsomere structural integrity and stability to be maintained, in the same way we discovered for

rotavirus VP8* [13]. Consequently, CaptHA1 in SHuffle T7 Express and mCap in Rosetta 2 (DE3)

were chosen for further investigation.

Figure 5-2: Modular capsomere protein expression and solubility. (A) CaptHA. (B) mCap in E.coli

strain Rosetta (DE3) pLysS, Rosetta 2 (DE3) or SHuffle T7 Express. Lanes: (M) molecular weight

marker, (P) pre-induced lysate, (T) total lysate, (S) soluble lysate.

Selective salt precipitation can be used to enrich target proteins in complex preparations [12, 22].

Sodium sulfate (Na2SO4) at 1 M concentration was used to precipitate CaptHA1 and mCap. SDS-

PAGE analysis of precipitated proteins showed an enrichment of CaptHA1 and mCap proteins, as

demonstrated by dominant bands observed at the expected sizes for VP1dC-tHA1 (61.97 kDa) for

CaptHA1 (Figure 5-3A), and VP1dC-tHA1 and VP1dC (35.2 kDa) for mCap (Figure 5-3B). E. coli

host proteins were co-precipitated, but to a lesser degree than the CaptHA1 and mCap target proteins.

96

The structural integrity of modular capsomeres post-precipitation was first examined using analytical

size exclusion chromatography (SEC-HPLC). Figure 5-3C shows the SEC-HPLC analysis of the

CaptHA1 and mCap precipitates. For the CaptHA1 precipitate, two dominant peaks were observed

at approximately 16 min and 22 min, corresponding to the elution of soluble aggregates and

capsomeres, respectively. In contrast, only one dominant peak at 16 min, corresponding to soluble

aggregates, was observed for the mCap precipitate. Only a modest capsomere peak was observed for

the mCap precipitate. Fractions representing peaks were collected and analysed by SDS-PAGE

(Figure 5-3A and 5-3B). The CaptHA1 capsomere peak fraction contained the majority of the VP1dC-

tHA1 target protein (Figure 5-3A, Lane Ca), while the majority of the VP1dC-tHA1 and VP1dC

target proteins in the mCap fractions were contained in the aggregate peak fraction (Figure 5-3B,

Lane Ag). Taken together, these results suggest that precipitation using Na2SO4 affects the structural

integrity of mCap but not that of CaptHA1. Contaminating E. coli host proteins were observed in the

capsomere peak fractions for both CaptHA1 and mCap; however, distinct differences were observed.

An intense band at approximately 75 kDa was observed in the capsomere peak fraction of CaptHA1,

but not in the capsomere peak fraction of mCap. It was hypothesised that this is the heat shock

chaperone protein, hsp70, aiding in capsomere stability. As CaptHA1 presented mostly as stable

modular capsomeres, it was selected for continued investigation. Given the high levels of soluble

aggregates observed in the mCap precipitate and the complexities associated with mixed modular

capsomeres, no further studies were performed with mCap.

97

Figure 5-3: Analysis of modular capsomere following precipitation using 1 M Na2SO4. (A) SDS-

PAGE analysis of CaptHA1. (B) SDS-PAGE analysis of mCap. (C) Size exclusion chromatogram

(analytical SEC-HPLC) of CaptHA1 and mCap. Lanes: (M) molecular weight marker, (PC) protein

precipitate, (Ag) SEC fraction corresponding soluble aggregate peaks, (Ca) SEC fraction

corresponding capsomere peaks.

To further improve on the precipitation of CaptHA1, Na2SO4 and PEG 6000 were tested at different

concentrations. PEG 6000 is commonly used to concentrate viruses and precipitate a variety of

proteins [23-26]. Figure 5-4A shows SDS-PAGE analysis of CaptHA1 precipitation using 1, 0.9, 0.8

and 0.6 M Na2SO4. Precipitation of CaptHA1 was observed at concentrations equal to or greater than

0.8 M Na2SO4. Co-precipitation of contaminants increased with increasing concentrations of Na2SO4.

High CaptHA1 yield was recovered from precipitation using 1 M Na2SO4 while a moderate amount

of CaptHA1 remained in the supernatant fraction. Figure 5-4B shows SDS-PAGE analysis of

CaptHA1 precipitation using 10, 8, 5 and 2% PEG 6000. Precipitation of CaptHA1 was observed at

greater than or equal to 5% PEG 6000. Optimal yield of CaptHA1 was obtained with 8% PEG 6000.

As observed for Na2SO4 precipitation, contaminating E. coli host proteins were precipitated alongside

CaptHA1.

98

Figure 5-4: Optimisation of selective precipitation using Na2SO4 and PEG 6000. (A) SDS-PAGE

analysis of precipitation using 1, 0.9, 0.8 and 0.6 M Na2SO4. (B) SDS-PAGE analysis of precipitation

using 10, 8, 5 and 2% PEG 6000. Lanes: (M) molecular weight marker, (Su) supernatant, (PC)

protein precipitate.

CaptHA1 precipitated by either 1 M Na2SO4 or 8% PEG 6000, was characterised by SEC and SDS-

PAGE. Figure 5-5A and 5-5B show SEC chromatograms of 1 M Na2SO4- and 8% PEG 6000-

precipitated CaptHA1, respectively. As observed above, capsomeres eluted at approximately 22 min,

and soluble aggregates at approximately 16 min. As shown by SDS-PAGE analysis, an intense protein

band was observed at the expected molecular weight for VP1dC-tHA1 (61.97 kDa) in both the

Na2SO4 and PEG 6000 precipitates. The protein band at 75 kDa in the Na2SO4 precipitation sample,

posited to be hsp70, was not observed in the PEG-precipitated sample. This highlights the selective

nature of the different precipitants as a result of their specific physicochemical properties. An SEC

chromatogram shows that PEG-precipitated CaptHA1 contains several peaks in addition to the main

capsomere and aggregate peaks observed from 1 M Na2SO4-precipitated CaptHA1. This may be

explained by the anomalous chromatographic behaviour of PEG-proteins or PEG-sodium ion

complexes on SEC [27-29] or PEG variants and other impurities. Detailed characterisation of these

peaks using alternative techniques such as liquid chromatography–mass spectrophotometry (LC/MS)

or SEC with multi-angle light scattering (SEC-MALS) would help to determine such peaks and

support alternative explanations.

99

Figure 5-5: Characterisation of precipitated CaptHA1. (A) 1 M Na2SO4-precipitated CaptHA1 (B)

8% PEG 6000-precipitated CaptHA1. Size exclusion chromatogram (left) showing capsomeres

separated from aggregates. SDS-PAGE analysis (right) showing lanes: (PC) protein precipitate, (Ag)

SEC fraction corresponding soluble aggregate peaks, (Ca) SEC fraction corresponding capsomere

peaks.

To confirm protein identity, precipitated CaptHA1 was further characterised using Western blot and

dot blot analysis. Theoretical molecular weights of VP1dC, VP1dC-tHA1 and an insect cell-derived

HA1 protein (used as control) are 35.2, 61.97 and 37.6 kDa, respectively. SDS-PAGE analysis

showed that protein bands were detected at the expected molecular weights for VP1dC and VP1dC-

tHA1 (Figure 5-6A). In line with previous studies showing glycosylation of insect cell-derived HA1,

control HA1 was detected at a larger molecular weight than predicted [30-32]. Precipitated CaptHA1,

chromatographically purified CaptHA1 and HA1 control were detected by Western blot and dot blot

100

analyses (Figure 5-6B and 5-6C) using anti-HA1 chicken sera, demonstrating recognition in both the

denatured (Figure 5-6B) and native (Figure 5-6C) forms. VP1dC (base module used as negative

control) was not detected by either Western blot or dot blot analyses using anti-HA1 chicken sera.

Receptor binding properties of precipitated CaptHA1 were analysed using a hemagglutination assay

(Figure 5-6D). Strong hemagglutination with endpoint titres of 512 and 1024 were observed for PEG-

and Na2SO4- precipitated CaptHA1, respectively. These titres were higher than chromatographically

purified CaptHA1 (endpoint titre of 256). In contrast, no hemagglutination was observed for

precipitated CapVP1dC negative control. The results indicate that precipitated CaptHA1 at the same

starting protein concentration (0.4 mg.mL-1) exhibits strong receptor binding properties, suggesting

that tHA1 modules retain their native structure, enabling the binding of modular capsomeres to sialic

receptors on the surface of the red blood cells. Further, these results suggest that precipitated

CaptHA1 is just as effective as chromatographically purified CaptHA1. The focus of this study is to

develop both expression and purification methods for simplified processing of CaptHA1; thus, the

protective efficacy of precipitated CaptHA1 in chickens to challenge with HPAI virus is highly

relevant and will be determined in future studies.

101

Figure 5-6: Functional characterisation of precipitated CaptHA1. (A) SDS-PAGE analysis. (B)

Western blot analysis. (C) Dot blot analysis with HA1 specific antibody detected CaptHA1 and HA1

protein. (D) Hemagglutination assay of precipitated CaptHA1, chromatographically purified

CaptHA1 and CapVP1dC. Lines represent endpoint titres.

To simplify the production process of CaptHA1 for use as an avian influenza vaccine, an integrated

approach was adopted to express non-tagged modular capsomeres and eliminate the use of

chromatography for purification. Figure 5-7 compares the previously established methodologies for

the production of chromatographically purified CaptHA1 (Figure 5-7A) with the simplified process

developed in this study (Figure 5-7B). The approach using non-tagged CaptHA1 expression removes

the need for enzymatic tag removal and isolation of tag-free target proteins via SEC, directly

eliminating the high costs associated with these processing steps. From 1 L cell culture, the yield of

the tag-free CaptHA1 target protein, following TEVp digestion and SEC and using the unoptimised

GST expression and purification method, was approximately 0.22 mg with a purity of 91%. In

contrast, CaptHA1 recovered following precipitation with Na2SO4- and PEG- produced target protein

yields of 36 mg and 48 mg, respectively. However, the purity of precipitated CaptHA1 is much lower

than the chromatographically purified CaptHA1 and, as observed from the SDS-PAGE analysis

(Figure 5-6A), the purity of Na2SO4-precipitated CaptHA1 is lower than PEG-precipitated CaptHA1.

102

An in vivo study by Wibowo and colleagues [12] demonstrated that the presence of E. coli host

contaminants did not adversely affect the immunogenicity of a vaccine candidate, but in fact

suggested that the presence of an optimal amount of contaminating host-derived protein with adjuvant

may drive a preferential immune response. This may be beneficial, particularly for veterinary

vaccines, where purity considerations are less constraining.

Figure 5-7: Process for modular capsomere production. (A) Previously established process for

producing chromatographically purified CaptHA1. (B) Simplified process for producing crudely

purified CaptHA1.

This study details a simpler production process for obtaining purified CaptHA1, which is superior to

the previously established chromatographic approach in terms of ease of production and increased

protein yield. These improvements are achieved at the expense of purity, which may be desirable for

veterinary vaccines, noting the adjuvanting effect of contaminants. This process eliminates the tag

cleavage steps and complex chromatography steps, namely GST-affinity purification and subsequent

SEC. The process also demonstrates that higher target protein yield provide cost advantages for

vaccine provider. Through the redesign of the molecular construct and the adoption of selective

103

precipitation as the method for purification, the yield of CaptHA1 is improved by more than 30-fold.

Moreover, precipitated CaptHA1 is shown to retain its biological properties. Further studies will aim

to improve CaptHA1 yield using high-throughput experimentation systems [33, 34] and a fed-batch

high-cell-density process [35] as shown for VP1 gram-per-litre expression yield. Purity of CaptHA1

may improve through further optimisation of the simple and scalable selective precipitation method

[12, 13, 36] or through the implementation of membrane filter-based technologies [37, 38]. Given the

functional activity demonstrated in this study, future work will also examine the immunogenicity of

crudely purified CaptHA1 and its protective efficacy in chickens, in particular, with animal-relevant

adjuvants such as oil-based adjuvants [39].


The authors acknowledge research funding from the Australian Research Council (ARC Discovery

Project DP160102915) and scholarship support to J.W. from the Thai Government (Royal Thai

Government Scholarship) and The University of Queensland (UQ) Top-Up Assistance.


The UQ filed patents on the use of murine polyomavirus as a vaccine platform. L.H.L.L. and A.P.J.M.

contributed to those patents and, through their employment with UQ, hold an indirect interest in this

intellectual property. Other authors declare that there are no conflicts of interest.

104

5.7 References

1. Foster, A. Bird flu outbreak MAPPED: Avian influenza spreads across Europe-could the UK

be next? 2016 [cited 2017 04.03.17]; available from:

http://www.express.co.uk/news/world/740957/bird-flu-outbreak-map-avian-influenza-

Europe-spread-European-countries-UK-Britain.

2. McKenna, M. Bird flu cost the US $3.3 billion and worse could be coming. 2015 [cited 2017

25.05.2017]; available from: http://phenomena.nationalgeographic.com/2015/07/15/bird-flu-

2/.

3. Spickler, A. R., Highly pathogenic avian influenza. 2016 [cited 03.03.17]; available from:

http://www.cfsph.iastate.edu/Factsheets/pdfs/highly_pathogenic_avian_influenza.pdf.

4. Fox, M., CDC concerned by H7N9 bird flu’s sudden spread in China. 2017 [cited 04.03.17];

available from: http://www.nbcnews.com/health/health-news/cdc-concerned-h7n9-bird-flu-

s-sudden-spread-china-n728946.

5. Capua, I. and S. Marangon, The use of vaccination as an option for the control of avian

influenza. Avian Pathology, 2003. 32(4): p. 335-343.

6. Marangon, S. and L. Busani, The use of vaccination in poultry production. Revue Scientifique

et Technique-Office International des Epizooties, 2007. 26(1): p. 265.

7. Charlton Hume, H.K. and L.H. Lua, Platform technologies for modern vaccine

manufacturing. Vaccine, 2017. 35(35): p. 4480-4485.




cost poultry influenza vaccination. Vaccine, 2018. 36(22): p. 3064-3071.

10. Lipin, D.I., L.H. Lua, and A.P. Middelberg, Quaternary size distribution of soluble

aggregates of glutathione-S-transferase-purified viral protein as determined by asymmetrical

flow field flow fractionation and dynamic light scattering. Journal of Chromatography A,

2008. 1190(1): p. 204-214.





5960-5965.

105






15. Laemmli, U.K., Cleavage of structural proteins during the assembly of the head of


16. World Organisation for Animal Health, Avian influenza in manual of diagnostics tests and

vaccines for terrestrial animals 2016. 2016 [cited 2016 26.10.2016]; available from:






2010. 5(7): p. e11548.



e2257.



21. Lobstein, J., et al., SHuffle, a novel Escherichia coli protein expression strain capable of

correctly folding disulphide bonded proteins in its cytoplasm. Microbial Cell Factories, 2012.

11(1): p. 753.

22. Howe, P.E., The use of sodium sulfate as the globulin precipitant in the determination of

proteins in blood. Journal of Biological Chemistry, 1921. 49(1): p. 93-107.

23. Yamamoto, K.R., et al., Rapid bacteriophage sedimentation in the presence of polyethylene

glycol and its application to large-scale virus purification. Virology, 1970. 40(3): p. 734-744.

24. Lewis, G.D. and T.G. Metcalf, Polyethylene glycol precipitation for recovery of pathogenic

viruses, including hepatitis A virus and human rotavirus, from oyster, water, and sediment

samples. Applied and Environmental Microbiology, 1988. 54(8): p. 1983-1988.

25. Ingham, K.C., Precipitation of proteins with polyethylene glycol: Characterization of

albumin. Archives of Biochemistry and Biophysics, 1978. 186(1): p. 106-113.

26. Skoog, B., Determination of polyethylene glycols 4000 and 6000 in plasma protein

preparations. Vox Sanguinis, The International Journal of Transfusion Medicine, 1979. 37(6):

p. 345-349.

106

27. Kraus, S. and L.B. Rogers, Effects of salts on the size exclusion behavior of poly(ethylene

glycol). Journal of Chromatography A, 1983. 257: p. 237-245.

28. Güven, O., Size exclusion chromatography of poly (ethylene glycol). Polymer International,

1986. 18(6): p. 391-393.

29. Tayyab, S., S. Qamar, and M. Islam, Size exclusion chromatography and size exclusion HPLC

of proteins. Biochemical Education, 1991. 19(3): p. 149-152.

30. Chen, J.-R., C. Ma, and C.-H. Wong, Vaccine design of hemagglutinin glycoprotein against

influenza. Trends in Biotechnology, 2011. 29(9): p. 426-434.

31. Kuroda, K., et al., Expression of the influenza virus haemagglutinin in insect cells by a

baculovirus vector. The EMBO Journal, 1986. 5(6): p. 1359.

32. Wang, K., et al., Expression and purification of an influenza hemagglutinin—one step closer

to a recombinant protein-based influenza vaccine. Vaccine, 2006. 24(12): p. 2176-2185.

33. Rameez, S., et al., High‐throughput miniaturized bioreactors for cell culture process

development: Reproducibility, scalability, and control. Biotechnology Progress, 2014. 30(3):

p. 718-727.

34. Effio, C.L., et al., High-throughput process development of an alternative platform for the

production of virus-like particles in Escherichia coli. Journal of Biotechnology, 2016. 219: p.

7-19.



36. Matulis, D., Selective precipitation of proteins. Current Protocols in Protein Science, 1997: p.

4.5. 1-4.5. 37.

37. Saxena, A., et al., Membrane-based techniques for the separation and purification of proteins:

an overview. Advances in Colloid and Interface Science, 2009. 145(1): p. 1-22.

38. Besnard, L., et al., Clarification of vaccines: An overview of filter based technology trends

and best practices. Biotechnology Advances, 2016. 34(1): p. 1-13.

37. Lone, N.A., E. Spackman, and D. Kapczynski, Immunologic evaluation of 10 different

adjuvants for use in vaccines for chickens against highly pathogenic avian influenza virus.

Vaccine, 2017. 35(26): p. 3401-3408.

107

Chapter 6

Conclusions and future work

6.1 Summary of research findings

Avian influenza (AI) remains a worldwide threat to the poultry industry and human health. The

continuing outbreaks of highly pathogenic avian influenza (HPAI) across numerous continents have

necessitated the implementation of appropriate strategies to control the spread of the virus.

Vaccination is a powerful tool for managing viruses in poultry. The use of vaccination programmes,

in conjunction with biosecurity means, offers a potential measure for controlling the disease [1].

However, the limitations of current egg- and cell culture-based commercialised AI vaccines-in

particular, slow production time-underline the need to improve vaccine manufacturing processes to

successfully respond to HPAI outbreaks.

Microbial-based expression systems are successfully used for the production of licensed vaccines for

hepatitis E virus [2] and human papillomavirus [3, 4]. Such technologies allow vaccines to be

manufactured at a speed and scale that is unachievable using egg- and cell culture-based systems. A

bacterial-based expression system also allows an effective strain-matched vaccine to be produced

within weeks of the identification of the new strain. In this thesis, to address current AI vaccine

manufacturing limitations, a novel capsomere platform based upon the VP1 capsid protein of murine

polyomavirus, was designed and expressed in E. coli. This rapidly produced, scalable platform

(VP1dC) presents a low-cost alternative for vaccine manufacturing.

The experimental work described in this thesis was designed to investigate three key aspects in the

development of a modular capsomere as an influenza vaccine candidate for application in poultry.

These are outlined below:

(i) Structure-based design of modular capsomeres (Chapter 3): This study evaluated the

potential of a structure-based designed modular capsomere presenting modified HA1 as a

poultry vaccine candidate.

108

(ii) Characterisation of modular capsomeres purified using simpler methods (Chapter 4): This

study investigated the immunogenicity of E. coli-produced modular capsomere partially

purified by chromatographic methods or crudely purified using selective precipitation

techniques.

(iii) Simplified production and purification of modular capsomeres (Chapter 5): The

production process of modular capsomeres was simplified by adopting an integrated

approach to generate non-tagged modular capsomeres and replace chromatographic

methods with simpler selective precipitation methods suitable for cost-effective vaccine

manufacturing for price-sensitive markets.

The following sections summarise the key findings obtained from the experimental work in this study.

6.1.1 Structure-based design of modular capsomeres

This study evaluated the potential of a novel structure-based designed modular capsomere as an

influenza vaccine candidate. Rational design of both the capsomere platform and HA1 was

undertaken to overcome capsomere instability. A VP1 platform lacking the last 63 amino acids but

retaining the native 28 amino acids at its N-terminus with two insertion sites at its N- and C-termini,

respectively was designed.

Flexible linkers GSAGSAAGSGEF (GSA linkers) were also inserted at both termini to provide

structural separation between the platform and insert. This new capsomere platform was designated

as VP1dC (Figure 3-1A). HA1, derived from human influenza A strain PR8 H1N1, was redesigned

to improve protein solubility by selectively removing the HA1 tail. Secondary structural elements

within the globular head of HA1, responsible for the formation of conformational epitopes, were thus

retained. Three versions of this truncated HA1 (tHA1) were designed; HA132-308 (tHA1-1), HA140-277

(tHA1-2) and HA144-268 (tHA1-3). Each version of tHA1 contained different secondary structural

elements. Each version of tHA1 and unmodified HA1 (amino acid 1 to 36) were then independently

modularised into either the N- or C-terminus of VP1dC, resulting in eight modular VP1dC-HA1

constructs: four constructs presenting HA1 at the N-terminus of VP1dC (HA1-N, tHA1-1N, tHA1-

2N and tHA1-3N) and four constructs presenting HA1 at the C-terminus of VP1dC (HA1-C, tHA1-

1C, tHA1-2C and tHA1-3C). These modular constructs were self-assembled into capsomeres

109

consisting of pentamers of each VP1dC-HA1 and were expressed as GST-tagged proteins in E. coli

to improve protein solubility and aid protein purification.

The expression and solubility levels of each GST-tagged modular capsomere were examined by SDS-

PAGE analysis. The site of antigen modularisation, the N- or C-terminus, was shown to have little

effect on capsomere expression and solubility. In contrast, the expression and solubility of modular

capsomeres were adversely affected by increases in molecular weight and the presence of secondary

structural elements within the HA1 module.

To investigate the stability of modular capsomeres using SEC, the GST-tagged modular capsomeres

were purified by GST-affinity chromatography followed by enzymatic release of GST. Non-tagged

modular capsomeres were then separated from GST tags and protein aggregates using SEC. The

highest capsomere yield was obtained from CaptHA1-3C, which is a pentamer of VP1dC-tHA1-3C,

suggesting that modularisation of HA1 on the C-terminus of VP1dC facilitated the recovery of stable

capsomeres. These results suggest that modularisation at the C-terminus allows greater separation

between the platform and insert than modularisation at the N-terminus. It is likely that site C is more

flexible than site N, thus avoiding the steric barrier between each protein domain, and preventing

protein aggregation.

Western blot analysis of purified CaptHA1-3C showed specific detection of VP1dC-tHA1-3C

modules by mouse anti-HA sera. A hemagglutination assay was performed to assess the receptor

binding properties of purified CaptHA1-3C, demonstrating strong hemagglutination with endpoint

titre of 1024. Together, these results suggest that HA1 modules present in purified CaptHA1-3C

maintain their native structure, enabling binding to sialic acid on the surface of red blood cells.

An in vivo immunogenicity study conducted in chickens demonstrated that immunisation with two

doses of purified CaptHA1-3C contains HA144-268 from PR8 H1N1, adjuvanted with aluminium

hydroxide (Alhydrogel), inducing high levels of HA1-specific antibody endpoint titres (almost 105).

A protective efficacy study performed with CaptHA1-3C contains HA145-262 from HPAI virus strain

H7N7, showing that chickens immunised with two doses of adjuvanted CaptHA1-3C45-262

(H7N7CaptHA1-3C) were completely protected against challenge with the homologous live virus.

The immunogenicity and protective efficacy data illustrate the potential of E. coli-produced

CaptHA1-3C as a poultry vaccine candidate.

110

To the best of our knowledge, this study is the first modularisation of a re-engineered large antigenic

HA1 module on the capsomere platform. This modular capsomere was successfully expressed in E.

coli without the need for protein refolding. The immunogenicity and protective efficacy of this

modular capsomere indicated its potential as a vaccine candidate. In addition, this capsomere

platform, which offer a fast, low-cost and high-yield expression system, can be quickly adapted to

match circulating influenza strains.

6.1.2 Characterisation of simpler purified modular capsomeres

The results presented in Chapter 3 support CaptHA1-3C as a promising vaccine candidate. However,

high costs associated with the current chromatographic purification process make it unfeasible for the

development of a low-cost poultry vaccine. Studies have demonstrated the high immunogenicity of

precipitated capsomeres and VLPs (solubilised capsomeres and VLPs purified by precipitation) as

vaccine candidates against influenza and rotavirus [5, 6]. Thus, simpler selective precipitation

protocols may offer alternative methods of purification. However, selective precipitation increases

impurities such as protein aggregates and other E. coli protein contaminants in the vaccine

preparation. These host cell-derived impurities can enhance vaccine immunogenicity [7]. Therefore,

this study, aimed to investigate the effect of such impurities on the immunogenicity of the CaptHA1-

3C vaccine candidate.

CaptHA1-3C (hereafter CaptHA1) was purified using the chromatographic method (cCaptHA1).

Two additional purification processes were used to obtained less pure CaptHA1 preparations, and the

components of each preparation were then characterised. The immunogenicity and safety of each

preparation was evaluated and compared with cCaptHA1.

CaptHA1 contains five VP1 monomers, each displaying a C-terminal tHA1 (Figure 4-1). The

chromatographically purified CaptHA1 (cCaptHA1) contains mainly pure capsomeres

(approximately 90%). The chromatographically partially purified CaptHA1 (aCaptHA1) contains

around a 1:1 ratio of pure capsomeres to soluble aggregates, while pCaptHA1, which is crudely

purified by selective precipitation, contains mostly soluble aggregates, some E. coli host proteins and

a small amount of capsomeres.

111

Western blot analysis showed that both aCaptHA1 and pCaptHA1 were recognised by mouse anti-

HA sera. Strong hemagglutination of aCaptHA (endpoint titre of 2048) was observed in a

hemagglutination assay. An endpoint titre of 32 was shown for pCaptHA1, which was lower than

both aCaptHA1 and cCaptHA1 (endpoint titre of 1024). These results suggest that the presence of

soluble aggregates and E. coli contaminants do not inhibit the hemagglutination ability of CaptHA1.

Thus, aCaptHA1 and pCaptHA1 are likely to present intact structures similar to cCaptHA1,

resembling the native structure of trimeric HA containing both the globular head (HA1) and the stem

(HA2), as shown in Figure 4-4B. However, the endpoint titre obtained from pCaptHA1 was much

lower than those obtained from cCaptHA1 and aCaptHA1. It is hypothesised that the lower amounts

of CaptHA1 present in pCaptHA1. It is hypothesised that the lower amounts of CaptHA1 present in

pCaptHA1 resulted in there being fewer HA1 sites to bind to sialic acid on the red blood cells.

The presence of endotoxin in injectable vaccines and its toxicity is of concern in animals and humans

[8-10]. Therefore, prior to the in vivo immunogenicity study, bacterial endotoxins were removed from

cCaptHA1, aCaptHA1 and pCaptHA1 using anion exchange. The total amount of endotoxin per

injection dose for cCaptHA1, aCaptHA1 and pCaptHA1 was considered low compared with the

endotoxin levels present in other bacterially-produced commercial vaccines for human use [11]. No

pathophysiological effects were observed in the birds after immunisation, suggesting that the amount

of endotoxin present in each dose was not toxic.

The in vivo immunogenicity of aCaptHA1 and pCaptHA1 in chickens was evaluated and compared

with cCaptHA1. Each vaccine candidate was adjuvanted with Alhydrogel and administered in two

doses. The aCaptHA1 induced a strong antibody of more than 105 endpoint titres, which was superior

to that obtained for cCaptHA1. The pCaptHA1 induced strong antibody responses of almost 105

endpoint titres, similar to titres obtained for cCaptHA1. Interestingly, aCaptHA1, which contained

high levels of soluble aggregates, elicited a higher immune response than cCaptHA1, which contained

mostly pure capsomeres. It is hypothesized that the robust immune responses raised against

aCaptHA1 resulted from protein aggregate-mediated B cell receptor cross-linking [12, 13]. It is also

hypothesised that pCaptHA1, which contained mostly soluble aggregates, small molecular weight

contaminants of E. coli proteins and a small amount of capsomeres, provoked strong immune

responses cause by a synergistic effect resulting from an optimal amount of contaminants in the

vaccine formulation [5].

112

Given that HI titres have been shown to correlate with protection against virus infection [14], a

hemagglutination inhibition (HI) test was used as a surrogate assay to assess vaccine efficacy. A HI

titre with a geometric mean of 8 is reported to fully protect chickens against challenge with the

homologous live virus [15]. HI titres obtained from aCaptHA1 and pCaptHA1 (geometric mean titre

= 69.8 and 17.7, respectively) are therefore likely to confer protection as was obtained from

immunisation with the cCaptHA1 (geometric mean titre = 20.7).

The data presented in this study suggest that soluble aggregates, bacterial contaminants and

endotoxins present in the vaccine formulation do not impair vaccine efficacy. Rather, an optimal

amount of contaminants is likely to contribute to a more robust immune response, which may provide

a greater level of protection. Preservation of these vaccine components is therefore, considered

advantageous, particularly for veterinary vaccines, which favour highly immunogenic, single-dose

vaccines.

6.1.3 Simplified production and purification of modular capsomeres

Chapter 4 demonstrated that soluble aggregates, E. coli contaminants and endotoxins present in the

vaccine formulation do not alter the efficacy of the vaccine after two doses. Instead, these

contaminants contribute to robust immune responses. This result also suggests the potential of crudely

purified modular capsomeres as a poultry vaccine candidate. Motivated by this finding, this study

aimed to simplify the production process of CaptHA1 by employing molecular and bioprocess

engineering approaches. Non-tagged CaptHA1 constructs, denoted CaptHA1 and mCap, were

generated. Capsomeres were expressed in E. coli and purified by selective precipitation using sodium

sulfate (Na2SO4) and polyethylene glycol (PEG) 6000.

CaptHA1 contains five copies of VP1dC-tHA. The mCap was generated by adopting a dual

expression strategy to titrate the content of tHA1 presented in the capsomere to reduce a steric effect

in the capsomere. The high expression and solubility levels of CaptHA1 and mCap were observed

from the expression in E. coli SHuffle T7 Express cells and Rosetta 2 (DE3) cells, respectively.

However, CaptHA1 provided a higher capsomere yield, while mCap formed mostly soluble

aggregates following precipitation using 1 M Na2SO4.

113

Optimisation of precipitation conditions was conducted to improve capsomere yield. CaptHA1 was

precipitated using various concentrations of Na2SO4 and PEG 6000. Optimal yield of CaptHA1 was

obtained when precipitated with 1 M Na2SO4 and 8% PEG 6000. However, Na2SO4-precipitated

CaptHA1 contains more E. coli contaminants compared with PEG-precipitated CaptHA1.

The HA1 antigen in precipitated CaptHA1 was specifically detected using anti-HA1 chicken antisera

under both denatured and non-denatured conditions. Strong hemagglutination of precipitated

CaptHA1 was observed with endpoint titre of 512 and 1024 for PEG- and Na2SO4- precipitated

CaptHA1, respectively. This suggests that the presence of intact HA1 structures facilitates its binding

of sialic acid on the surface of red blood cells. It also suggests that precipitation using either PEG or

Na2SO4 does not destroy the structural integrity of CaptHA1. Additionally, these results reveal that

precipitated CaptHA1 is as effective as chromatographically purified CaptHA1.

The yield of Na2SO4- and PEG-precipitated CaptHA1 before the buffer exchange and endotoxin

removal processes was approximately 36 mg.L-1 and 48 mg.L-1, respectively. However, the

subsequent steps for buffer exchange and endotoxin removal may lead to decreases in final target

protein recovery.

Despite low purity levels, the yield of precipitated CaptHA1 was considerably higher than that of

chromatographically purified CaptHA1, which was only 2.4 mg.L-1 following GST-affinity

chromatography and 0.22 mg.L-1 with 91% purity following enzymatic release of GST and separation

of CaptHA1 from GST tags and protein aggregates using SEC.

This study demonstrated a simple chromatography-free process for producing CaptHA1, which is

superior to the previously established chromatographic approach in terms of production speed. In

addition, this simpler process eliminated complex chromatography steps, GST-affinity

chromatography and subsequent GST removal in particular; thus, production costs are likely to be

substantially reduced. A significant improvement was also achieved in the final yield, showing an

increase of more than 30-fold.

6.2 Future work

This thesis introduced a novel capsomere vaccine platform that is suitable for modularising large

heterologous antigenic modules and amendable to rapid industrial-scale vaccine manufacturing. This

114

thesis contributes important findings to the design and process development of platform-based

vaccine technology. These findings warrant further investigation in this field of study:

(i) The modular capsomere (CaptHA1) formulated with an aluminium hydroxide

(Alhydrogel) adjuvant induced strong immune responses when administered to

chickens. Aluminium hydroxide was chosen to enable comparisons with previous

studies; however, this adjuvant is not a standard adjuvant for animal vaccines. The

investigation of CaptHA1 formulated with animal-relevant adjuvants such as oil

emulsions [16] would identify the most suitable adjuvant for poultry vaccination.

Further, the duration of the protective immune response should be monitored to fully

evaluate effectiveness of vaccine candidates.

(ii) The simplified bioprocess developed in this thesis allows the production of functional

modular CaptHA1 at high yields without chromatography. To further increase

capsomere yield, optimisation of protein expression conditions [17, 18] should be

undertaken. Additionally, optimisation of selective precipitation conditions [5, 6, 19]

or integration of filter-based technology such as membrane filtration [20, 21] may

improve the final yield and purity of modular capsomeres. Future work should include

in vivo immunogenicity and protective efficacy studies in chickens using vaccine

candidates processed with newly developed methods.

(iii) Poultry vaccination is a powerful tool to support viral eradication programmes,

particularly when partnered with other control strategies. Vaccination programmes

must allow differentiation between infected and vaccinated animals (DIVA). The

DIVA strategy enables the detection of field exposure in vaccinated animals based on

heterologous vaccination. The CaptHA1 contains only hemagglutinin (HA1), making

it applicable to the DIVA concept. Further studies (field trials) are needed to evaluate

the use of CaptHA1 as a feasible DIVA vaccine. Differentiation of CaptHA1-

vaccinated chickens from vaccinated and then infected chickens can be achieved

through the detection of antibodies to other viral proteins such as nucleoprotein, matrix

protein, NS1 or neuraminidase.

115

(iv) The HA1 antigenic module in CaptHA1 generated in this thesis was derived from the

influenza A strain of either A/Puerto Rico/8/34 H1N1 (PR8 H1N1) or HPAI

A/chicken/NSW/3122-1/2012 H7N7. PR8 H1N1 was used in this study to benchmark

against previous studies, and HPAI H7N7 was used to allow viral challenge study in

chickens. The evaluation of modular capsomere comprising HA1 from persistently

circulating viruses among poultry populations, such as HPAI viruses subtype H5N1,

H5N8 and H7N9, is crucial for the development of a poultry vaccine to control HPAI

outbreaks. The adaptability of the capsomere vaccine platform to a variety of HA1

subtypes is a warranted study to support the implementation of this platform as a

strategy for AI preparedness. In addition, the VP1dC capsomere platform could be

applicable when designing improved influenza vaccines for other species, including

horses, pigs and humans. The need for an updated, rapidly produced and strain-

matched influenza vaccine is a significant hurdle in the control of influenza in all

species, including humans. This capsomere platform might also be interesting when

designing emergency vaccines for other diseases unrelated to avian influenza.

6.3 Concluding thoughts

Highly pathogenic avian influenza remains a global threat to animal and public health. Control

measures such as biosecurity and stamping out are not always achievable for social, economic and

ethical reasons. Vaccination provides a suitable means to control avian influenza outbreaks and

prevent potential transmission to humans. The constraints of commercialised poultry influenza

vaccines, which are produced mainly in eggs or cell cultures, underline the need for rapid vaccine

manufacturing for the effective control of avian influenza. This thesis sought to develop a rapidly

manufactured, low-cost poultry vaccine that is safe and efficacious. The newly-designed murine

polyomavirus capsomere platform (VP1dC) was modularised with the large structure-based designed

HA1 antigen (tHA1) from influenza A virus. The stable modular capsomere (CaptHA1) was

successfully produced in E. coli without the need for protein refolding. This CaptHA1 induced

protective immunity in chickens, verifying its potential as a poultry vaccine candidate. Further, the

simplified process for producing CaptHA1 demonstrated the possibility for producing modular

capsomeres at a scale, speed and cost that is unattainable using eggs or cell cultures. This microbial-

based modular capsomere platform for vaccine manufacturing would yield sufficient vaccines at an

116

affordable price for both developed and developing countries during avian influenza epidemics and

pandemics. Importantly, this modular vaccine platform and production strategy is applicable to other

vaccine designs for modularisation of large antigenic modules to generate vaccines that have not been

possible using conventional methods.

117

6.4 References


Vaccine, 2007. 25(30): p. 5645-5652.

2. Li, S.W., et al., A bacterially expressed particulate hepatitis E vaccine: antigenicity,

immunogenicity and protectivity on primates. Vaccine, 2005. 23(22): p. 2893-2901.

3. Zhao, Q., et al., Disassembly and reassembly improves morphology and thermal stability of

human papillomavirus type 16 virus-like particles. Nanomedicine: Nanotechnology, Biology

and Medicine, 2012. 8(7): p. 1182-1189.

4. Mach, H., et al., Disassembly and reassembly of yeast‐derived recombinant human

papillomavirus virus‐like particles (HPV VLPs). Journal of Pharmaceutical Sciences, 2006.

95(10): p. 2195-2206.



5960-5965.




7. Akira, S., S. Uematsu, and O. Takeuchi, Pathogen recognition and innate immunity. Cell,

2006. 124(4): p. 783-801.

8. Batista, P.d.O.M.D., et al., Methods of endotoxin removal from biological preparations: a

review. Journal of Pharmacy and Pharmaceutical Sciences, 2007. 10(3): p. 388-404.

9. Ogikubo, Y., et al., Evaluation of the bacterial endotoxin test for quantification of endotoxin

contamination of porcine vaccines. Biologicals, 2004. 32(2): p. 88-93.

10. Erridge, C., E. Bennett-Guerrero, and I.R. Poxton, Structure and function of

lipopolysaccharides. Microbes and Infection, 2002. 4(8): p. 837-851.

11. Brito, L.A. and M. Singh, Acceptable levels of endotoxin in vaccine formulations during

preclinical research. Journal of Pharmaceutical Sciences, 2011. 100(1): p. 34-37.

12. Braun, A., et al., Protein aggregates seem to play a key role among the parameters influencing

the antigenicity of interferon alpha (IFN-α) in normal and transgenic mice. Pharmaceutical

Research, 1997. 14(10): p. 1472-1478.

13. Wang, W., et al., Immunogenicity of protein aggregates—concerns and realities. International

Journal of Pharmaceutics, 2012. 431(1): p. 1-11.

14. Ohmit, S.E., et al., Influenza hemagglutination-inhibition antibody titer as a correlate of

vaccine-induced protection. Journal of Infectious Diseases, 2011. 24(12): p. 1879-1855.

118


cost poultry influenza vaccination. Vaccine, 2018. 36(22): p. 3064-3071.

16. Lone, N.A., E. Spackman, and D. Kapczynski, Immunologic evaluation of 10 different

adjuvants for use in vaccines for chickens against highly pathogenic avian influenza virus.

Vaccine, 2017. 35(26): p. 3401-3408.

17. Effio, C.L., et al., High-throughput process development of an alternative platform for the

production of virus-like particles in Escherichia coli. Journal of Biotechnology, 2016. 219: p.

7-19.

18. Rameez, S., et al., High‐throughput miniaturized bioreactors for cell culture process

development: Reproducibility, scalability, and control. Biotechnology Progress, 2014. 30(3):

p. 718-727.

19. Matulis, D., Selective precipitation of proteins. Current Protocols in Protein Science, 1997: p.

4.5. 1-4.5. 37.

20. Besnard, L., et al., Clarification of vaccines: An overview of filter based technology trends

and best practices. Biotechnology Advances, 2016. 34(1): p. 1-13.

21. Saxena, A., et al., Membrane-based techniques for the separation and purification of proteins:

an overview. Advances in Colloid and Interface Science, 2009. 145(1): p. 1-22.

22. Suarez, D.L., DIVA vaccination strategies for avian influenza virus. Avian Diseases, 2012.

56(4s1): p. 836-844.

119

Appendix A

Animal Ethics Approval Certificate (1)

120

121

Appendix B

Animal Ethics Approval Certificate (2)

122

Documents

Bioengineering Viral Subunits for Influenza Vaccine