PANDEMIC PREPAREDNESS
AGAINST HIGHLY PATHOGENIC AVIAN INFLUENZA VIRUSES
By
Ram Parkash Kamal

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Comparative Medicine and Integrative Biology – Doctor of Philosophy
2014

ABSTRACT
PANDEMIC PREPAREDNESS
AGAINST HIGHLY PATHOGENIC AVIAN INFLUENZA VIRUSES
By
Ram Parkash Kamal
Influenza is an infectious disease of birds and mammals (including humans) caused by
influenza viruses. In humans the disease has two outcomes; seasonal influenza
epidemics (3-5 million severe cases and 250,000-500,000 deaths worldwide each year)
and human influenza pandemics (even greater public health and economic loss).
Molecular characterization of pandemic viruses of 20th (1918 H1N1, 1957 H2N2, and
1968 H2N3) and 21st centuries (2009 H1N1pdm) have revealed that these pandemic
viruses evolved from avian influenza viruses. Among avian influenza viruses, highly
pathogenic H5N1 and H7 viruses pose the greatest pandemic potential. The risk posed
by these viruses is encountered by pandemic preparedness strategies; constant
surveillance of virus evolution, risk assessment and preparation of pre-pandemic
vaccine seeds. I identified two major needs in the pandemic preparedness; the first was
the evolution of novel PB1-F2 truncations in H5N1 viruses and the second was the poor
immunogenicity of H7 viruses and vaccines. To fulfill the needs, I worked on two
research projects which are compiled as two chapters of this dissertation.
Firstly, we studied evolution and prevalence of PB1-F2 truncations in mammalian
versus avian influenza viruses along with their risk assessment. We for the first time
found independent occurrence and evolution of PB1-F2 truncations in multiple clades of

H5N1 viruses. In our risk assessment, we found a virus expressing C-terminal (Nterminal truncated) PB1-F2 (A/chicken/Vietnam/NCVD-296/2009; VN/296) 1000 fold
more virulent than another very similar virus (A/chicken/Vietnam/NCVD-281/2009;
VN/281) expressing full length PB1-F2. In vitro studies with cloned PB1-F2 expressing
proteins of 0, 24 and 90 amino acids did not show any significant difference on cell
apoptosis and IFNβ expression. However protein expression from full length and
deleted (ATG-ACG) mutants were obvious in transfected A549 cells with fractional
mitochondrial localization. We propose the model that C-terminal fragment of PB1-F2 as
in virus VN/296 is responsible for the higher virulence and the phenotype is countered
by N-terminal fragment as present in VN/281.
Secondly, we evaluated and compared the immunogenicity of purified recombinant
hemagglutinin (HA) proteins of two H7 and two seasonal influenza viruses. Our studies
showed that H7 HAs are weaker immunogens than seasonal influenza virus HAs. This
suggests the use of higher doses of HA in H7 vaccines or possible use adjuvants.
To summarize, the PB1-F2 risk assessment showed that virus expressing C-terminal
fragment is 1000 times more virulent than another very similar virus having full length
PB1-F2. These findings highly recommend including PB1-F2 truncations in prepandemic virus surveillance and risk assessment. The H7 HA immunogenicity study
concludes that the poor immunogenicity of H7 viruses and vaccines is due to intrinsic
properties of their HA protein. Further recommendations include using either higher
doses of HA in H7 prepandemic vaccines or adjuvants to achieve accepted level of
protection. These recommendations will strengthen the current pandemic preparedness
studies.

I dedicate this work to my loving wife, Anita Kumari, and my parents, Bishan Singh and
Lakshmi Devi. Without your love, support, patience and encouragement, this
monumental accomplishment would not have been possible. You all have sacrificed a
lot in order for this to come to fruition. In addition, I also dedicate this work to late Dr.
Aleksander Lipatov (Sasha) who was a dear friend, great senior and colleague.

iv

ACKNOWLEDGMENTS

I have had the opportunity to work with some great people while pursuing my degree at
Michigan State University. First, I would like to thank Dr. Ian A. York for giving me the
opportunity to join his lab at MSU and later at CDC to pursue this degree. I feel very
lucky to have you as my PhD supervisor. Any combination of words won’t be able to
explain how thankful I am for your guidance and mentorship over the years. From you, I
have learned not only to be a good scientist, but also to be a good colleague,
collaborator and supervisor. I hope we maintain a relationship in the future. I would also
like to thank the members of my committee, Dr. Roger Maes, Dr. Henry Hunt, Dr.
Sungjin Kim and Dr. Andrea Amalfitano for being easily available for my committee
meetings, continuous review of my dissertation progress and your expert feedbacks.
I would like to specially thank CMIB Program Director, Dr. Vilma Yuzbasiyan-Gurkan for
helping me out every time whether it’s joining a lab, choosing a course, moving to CDC,
organizing committee meetings, preliminary exam and final PhD exam. You have
contributed significantly to my and other CMIB students’ achievements. The journey
would have been tougher if without your guidance and help.
A special thanks to Kristy, Pin and Amrita (CDC, Influenza Division) for your contribution
to the dissertation. Without your support it could not be completed in a timely manner. I
also thank my seniors and colleagues at CDC/ Influenza Division, Dr. Charles Todd, Dr.
James Steven, Dr. Terry Tumpey, Dr. S. Sambhara, Dr. Priya Ranjan, Dr. Li-Mei Chen,
Dr. Jaber Hossain, and Dr. Jessica Belser for guiding me through these years and
generously sharing your expertise and reagents. My thanks also go to my group
v

members, Ned, Jason, Adrian and Irina, for your friendship and good times as well as
scientific inputs and inspirations. In addition to our group members, I thank members of
other teams of CDC-Influenza Division, Missy, Kortney, Xiangjie and Laura for your help
and answering all my queries.
A big thanks to Dr. Jaqueline Katz and Dr. Ruben Donis for giving me the opportunity to
do my PhD dissertation at CDC, Influenza Division. I will always be grateful for the
opportunity, your support and performance tracking which led to the timely completion
of this dissertation.
I thank CMIB PhD Program, Dept. of MMG and Grad School at Michigan State
University (East Lansing, MI, USA) for assistantships and fellowships. I am grateful to
Oak Ridge Institute for Science and Education (Oak Ridge, TN) for its fellowships during
my dissertation research at CDC.
I would like to thank my MSU friends Yogesh, Sonika, Rahul, V, and Manish for your
friendship, inspiration and wonderful times we spent together in Lansing. In addition, I
thank all my friends for your wells wishes, support and fun-time.
Finally, I would like to thank my my parents, Bishan Singh and Lakshmi Devi, my
brother Nijesh Kumar for their support during this endeavor. It has meant a lot and I am
truly grateful. Above all, I must thank Anita, you are the best and I appreciate you and
the sacrifices that you have made more than words can say!

vi

TABLE OF CONTENTS

LIST OF TABLES ........................................................................................................... x
LIST OF FIGURES.......................................................................................................... xi
KEY TO ABBREVIATIONS ........................................................................................... xii
CHAPTER 1 .................................................................................................................... 1
GLOBAL INTRODUCTION............................................................................................. 1
Influenza Disease: .............................................................................................. 1
Influenza Viruses: .............................................................................................. 2
IAV Genes and Proteins: .................................................................................... 2
Public Health Burden of Influenza: ................................................................... 4
Influenza Pandemics: ......................................................................................... 5
Avian Influenza Viruses: .................................................................................... 7
Pandemic Preparedness: ................................................................................. 12
1. Determinants of Pathogenicity: ........................................................... 13
2. Pandemic Vaccines: ........................................................................... 14
Gaps in Pandemic Preparedness:................................................................... 14
1. H5N1 PB1-F2: .................................................................................... 15
2. Poor immunogenicity of H7 viruses: ................................................... 18
Summary: .......................................................................................................... 20
APPENDICES .................................................................................................... 22
APPENDIX 1A: Tables ........................................................................... 23
APPENDIX 1B: Figures .......................................................................... 26
REFERENCES ................................................................................................... 28
CHAPTER 2 ................................................................................................................. 38
PB1-F2 VARIANTS IN H5N1 AVIAN INFLUENZA VIRUSES ...................................... 38
Abstract: ........................................................................................................... 38
Introduction: .................................................................................................... 39
Materials and Methods: .................................................................................... 42
Database analysis of PB1-F2 truncations............................................ 42
Distribution of PB1-F2 truncations in H5N1 viruses ........................... 42
Identification of paired PB1-F2 full-length/truncation H5N1 viruses . 43
Preparation of viral stocks .................................................................... 43
EID50 ....................................................................................................... 44
Plaque assay .......................................................................................... 44
Pathogenicity in mice ............................................................................ 44
Cloning and mutagenesis of PB1-F2.................................................... 45
Western blot ........................................................................................... 45
Confocal microscopy ............................................................................ 46
Quantitative RT-PCR for IFNβ expression ........................................... 46
vii

Flow cytometry for cell apoptosis ........................................................ 46
Statistical analyses ................................................................................ 47
Results:
47
Evolution of PB1-F2 truncations in avian and mammalian viruses .. 47
Evolution of PB1-F2 truncations in H5N1 viruses .............................. 49
Paired viruses for PB1-F2 studies ........................................................ 50
Full length PB1-F2 (VN/281) is associated with reduced-mice virulence ............................................................................ 51
Full length H5N1 PB1-F2 localizes to mitochondria ........................... 52
PB1-F2 mutants had no effect on apoptosis ....................................... 53
PB1-F2 truncations and IFNb expression ............................................ 53
Discussion: ...................................................................................................... 54
APPENDICES .................................................................................................... 59
APPENDIX 2A: Tables ........................................................................... 60
APPENDIX 2B: Figures .......................................................................... 63
REFERENCES ................................................................................................... 73
JOINING CHAPTER ..................................................................................................... 78
PATHOGENICITY DETERMINACY TO PANDEMIC VACCINES ................................ 78
REFERENCES ................................................................................................... 81
CHAPTER 3 .................................................................................................................. 83
HEMAGGLUTININ OF THE H7 SUBTYPE IS INTRINSICALLY POORLY
IMMUNOGENIC ............................................................................................................ 83
Abstract: ........................................................................................................... 84
Introduction: ..................................................................................................... 84
Materials and Methods: .................................................................................... 86
Recombinant hemagglutinin................................................................. 86
Viruses .................................................................................................... 87
Immunization.......................................................................................... 87
Antibody assays .................................................................................... 87
HI assay .................................................................................................. 88
ELISA ...................................................................................................... 88
Avidity assay .......................................................................................... 89
Microneutralization assay ..................................................................... 90
Statistical analyses ................................................................................ 91
Results: ............................................................................................................. 91
ELISA responses to purified HA........................................................... 91
Hemagglutination inhibition (HI) antibodies ........................................ 92
Virus-neutralizing antibodies ................................................................ 92
Avidity ELISAs ....................................................................................... 93
Discussion: ...................................................................................................... 94
Acknowledgements: ......................................................................................... 99
APPENDIX ....................................................................................................... 100
REFERENCES ................................................................................................. 107

viii

CHAPTER 4 ............................................................................................................... 114
OVERALL DISCUSSION AND SUMMARY ................................................................ 114

ix

LIST OF TABLES

Table 1.1 Influenza A virus genes and proteins ....................................................... 23
Table 1.2 Reported cases of avian influenza virus infections
by viral isolation ........................................................................................ 24
Table 1.3 Molecular determinants of IAV pathogenicity ......................................... 25
Table 2.1 Oligonucleotides used for cloning and mutagenesis of PB1-F2 ............ 60
Table 2.2 Details of PB1-F2 mutants made in this study ......................................... 61
Table 2.3 Prevalence of PB1-F2 truncations among all subtypes
(human, swine and avian IAVs) versus H5N1………………………………. 62
Table 2.4 Amino acid differences between viruses VN/281 and VN/296................. 62
Table 2.5 Growth characteristics and mouse virulence........................................... 62

x

LIST OF FIGURES

Figure 1.1 Structure of influenza A virus………………………………………………... 26
Figure 1.2 Genetic components of 2009 H1N1pdm09 virus .................................... 27
Figure 1.3 Structure of influenza A virus ................................................................. 26
Figure 2.1 PB1-F2 truncation prevalence .................................................................. 63
Figure 2.2 Evolutionary relationship of PB1-F2 truncations in H5N1 viruses ....... 64
Figure 2.3 Mice virulence of paired viruses with and without PB1-F2.................... 65
Figure 2.4 Construction of PB1-F2 mutants’ recombinant plasmid clones ........... 67
Figure 2.5 Expression of PB1-F2 mutants in A549 cells .......................................... 68
Figure 2.6 PB1-F2 expression and mitochondrial localization ................................ 69
Figure 2.7 PB1-F2 effect on apoptosis ...................................................................... 71
Figure 2.8 PB1-F2 effect on IFNβ expression ........................................................... 72
Figure 3.1 ELISA responses to recombinant hemagglutinins ............................... 101
Figure 3.2 HI responses to recombinant hemagglutinins...................................... 103
Figure 3.3 MN responses to recombinant hemagglutinins………………..……… 104
Figure 3.4 HI and MN titers and avidity ELISA for mice with HI titers
between 80 and 160 ................................................................................. 106

xi

KEY TO ABBREVIATIONS

HA

Heamgglutinin

NA

Neuraminidase

IAV

Influenza A virus

AIV

Avian influenza virus

HPAIV

Highly pathogenic avian influenza virus

HP

Highly pathogenic

LV

Low pathogenic

LPAIV

Low pathogenic avian influenza virus

RNA

Ribonucleic acid

CDC

Centers for Disease Control and Preventions, Atlanta, GA USA

HA

Heamgglutination

HI

Heamgglutination Inhibition

MN

Microneutralization

ELISA

Enzyme linked immunosorbant assay

RDE

Receptor destroying enzyme

BSL

Biosafety level

OD

Optical density

OPD

ortho-phenylenediamine

BSA

Bovine serum albumin

RBC

Red blood cell

RBS

Receptor binding site

xii

LD50

Median lethal dose

qRT-PCR

Quantitative Reverse transcription- Polymerase Chain Reaction

GISAID

Global Initiative on Sharing Avian Influenza Data

EID50

Median egg infectious dose

TCID50

Median tissue culture infectious dose

GAPDH

Glyceraldehyde 3-phosphate dehydrogenase

PFU

Plaque forming unit

xiii

CHAPTER 1
GLOBAL INTRODUCTION
Influenza Disease:
Influenza is an infectious disease of birds and mammals which is caused by influenza
viruses. Wild aquatic birds are the reservoir of influenza viruses since they harbor all the
different types of influenza viruses and serve as sources for all influenza infection of
domestic poultry and mammals, including humans. Wild birds typically manifest very
mild or asymptomatic intestinal tract infection (Webster, Bean et al. 1992), while in
domestic poultry and humans the disease manifestation may range from mild or subclinical respiratory tract infection (Swayne 2007) to severe respiratory disease with high
lethality depending on the strain of influenza virus and the presence of underlying host
factors. Clinical symptoms of human influenza infection can range from asymptomatic to
serious sickness characterized by sudden onset of fever, myalgia, and respiratory
symptoms. In humans, seasonal influenza epidemics occur each year which result in 35 million severe cases and 250,000-500,000 deaths worldwide (Influenza (Seasonal)).
Occasionally influenza viruses of novel antigenic characteristics, which have attained
efficient human to human transmission capability, cause human influenza pandemics.
There were 3 influenza pandemics (and one pseudo-pandemic) in the 20
st

one in the 21 century.

1

th

century and

Influenza Viruses:
Influenza viruses were isolated from chickens at the beginning of the 20th century
(A/Brescia/1902 [H7N7]), from pigs in the late 1920s, from humans in the early 1930s,
from horses and domestic ducks in the 1950s, from terns in 1961, and from many
waterfowl and shorebirds since 1974 (reviewed in (Webster, Bean et al. 1992)). These
viruses are lipid enveloped and belong to the family Orthomyxoviridae. Influenza viruses
are divided into three types (A, B and C) based on their matrix and nucleoprotein; only
types A and B viruses are responsible for most of the human influenza infections. While
type B viruses have only one subtype, influenza A viruses (IAVs) are further divided into
multiple subtypes based on their surface glycoproteins, hemagglutinin (HA) and
neuraminidase (NA). IAVs can belong to any subtype based on the combination of 17
known HA and 10 NA genes (Li, Sun et al. 2012, Tong, Li et al. 2012, Zhu, Yang et al.
2012). Birds are the natural reservoir for most IAVs because they harbor viruses with
most of the HA and NA combinations. Three subtypes (H1, H2, and H3) have caused
pandemics in humans so far (Webster, Bean et al. 1992). However, a number of other
subtypes, including H5, H6, H7, H9, and H10 have also infected humans (Webster,
Bean et al. 1992, Li, Li et al. 2006, Zhu, Wang et al. 2010, Arzey, Kirkland et al. 2012).
IAV Genes and Proteins:
A mature virion of influenza A virus is composed of the nucleocapsid made from M1 and
the membrane envelope derived from host cell membrane. The virion envelope contains
three viral proteins; HA, NA and M2 (figure 1.1).

2

The IAV genome is comprised of 8 gene segments of negative-sense single-stranded
RNA (figure 1.1). Each segment contains a coding region that encodes one or more
proteins, flanked by short 5' and 3' non-coding sequences. IAV encode 10 core proteins,
and a variable number of “accessory” proteins that are variably present in different
strains of viruses (table 1.1). Two segments encode surface envelope glycoproteins that
function as viral antigens: haemagglutinin (HA; approximately 1700 nucleotides,
segment 4), which is responsible for binding to sialic acid receptors and entry into host
cells, and neuraminidase (NA; approximately 1400 nucleotides, segment 6), which is
involved in release of budding virions from infected cells. Besides HA and NA, there are
two additional structural proteins; one is nucleoprotein (NP; approximately 1500
nucleotides, segment 5), which binds to the viral RNA, and another is the matrix protein
M1 (approximately 750 nucleotides, segment 7) which comprises the main component
of the viral capsid. The seventh segment also encodes a second protein, M2 (291
nucleotides), which is an integral membrane protein that functions as an ion channel.
The IAV polymerase complex is composed of three polymerase proteins; basic
polymerase 2 (PB2; about 2,275 nucleotides in coding region, segment 1), which
controls the recognition of host-cell RNA; basic polymerase 1 (PB1; about 2,270
nucleotides in the coding region, segment 2), which is suggested to catalyze nucleotide
addition and the acidic protein (PA; about 2,150 nucleotides, segment 3), which
possesses an endonuclease and a cap-snatching activity. Segment 8 is the smallest
segment of the viral genome, and encodes two proteins. The first is a non-structural
protein, NS1 (690 nucleotides), with multiple functions that particularly target the host
innate immune response. The second protein, encoded by an overlapping reading
3

frame, is the NS2 protein (363 nucleotides), and has a role in viral RNA transport
(O'Neill, Talon et al. 1998). These proteins (HA, NA, PB1, PB2, PA, NP, NS1, NS2, M1,
and M2) are the ten core proteins that are found in all IAV strains.
Gene segment 2, in addition to PB1, also encodes another protein from its +1 open
reading frame (PB1-F2) which localizes to mitochondria and has been implicated in cell
apoptosis. In 2009, an additional protein (N40) was discovered which is an N-terminally
truncated version of PB1 protein and is translated from fifth AUG (codon 40) in PB1
gene (Wise, Foeglein et al. 2009). In 2012, another IAV protein (PA-X) was discovered
which is synthesized from Segment 3 (PA) (Jagger, Wise et al. 2012). A wide range of
other proteins, often comprised of N- or C-terminal truncations of longer core proteins,
have been observed in strains of influenza, but frequently are evolutionarily unstable;
many of these proteins have not been described in the peer-reviewed literature but have
been mentioned in conference proceedings and informal discussions.
Public Health Burden of Influenza:
Influenza infections in human population result in annual epidemics with significant
morbidity and mortality worldwide. Annual influenza epidemics account for 12% ($10
billion) of the total economic burden in direct medical costs, however the overall
economic burden is much higher ($87.1 billion) (Molinari, Ortega-Sanchez et al. 2007).
Influenza and pneumonia rank as the eighth leading cause of all deaths in US in 2011
(Centers for Disease and Prevention 2010, Hoyert and Xu 2012). Out of 2,512,873
deaths in 2011, 53,667 (2.14%) deaths were associated with influenza and pneumonia
(Hoyert and Xu 2012).

4

Besides seasonal epidemics, influenza sporadically causes more devastating
pandemics. Influenza A viruses have caused 3 pandemics in 20

th

century (1918, 1957,

1968) and one recent pandemic in 2009. The pandemics of 1918, 1957, and 1968 were
due to either purely avian viruses or to reassortants of avian and mammalian viruses,
and that of 2009 was due to a reassorted swine virus that included avian influenza
components (figure 1.2). Therefore, the evolution of avian viruses poses a constant
threat of human pandemics.
Influenza Pandemics:
The 1918 pandemic virus has been called “the mother of all influenza pandemic viruses”
since all subsequent human pandemic viruses have included components of the 1918
strain (reviewed in (Elderfield and Barclay 2011)). Viruses derived from 1918 pandemic
continued to circulate in humans as seasonal influenza viruses, usually causing annual
epidemics with moderate morbidity and mortality. However, the nature of the segmented
genome of the virus makes it capable of reassortment with circulating avian and animal
viruses and thus gives the potential for the emergence of new virus (reviewed in
(Elderfield and Barclay 2011)).
The 1957 (“Asian”) pandemic virus was a new subtype (H2N2) that derived its HA, NA
and PB1 genes by reassortment with avian H2N2 viruses, but its remaining five gene
segments were derived from the 1918 H1N1 pandemic virus. Since the human
population had not been exposed to the H2 HA, there was no population immunity to H2
and the virus spread rapidly in the susceptible population, resulting in a death toll of
about 2 million people (reviewed in (Elderfield and Barclay 2011)).
5

Another such event happened in 1968 (“Hong Kong Flu”), when the circulating human
H2N2 virus reassorted with an avian virus of H3 subtype and resulted in 1-2 million
deaths. This virus acquired two genes (HA and PB1) from an avian H3 virus, and
retained NA from the previously-circulating H2N2 virus and the rest of the 5 gene
segments from 1918 virus (reviewed in (Elderfield and Barclay 2011)).
The pandemic viruses of 1957 and 1968 replaced the previously circulating viruses,
which therefore became extinct in the wild. An unusual event occurred in 1977 (the
“Russian Flu” pseudo-pandemic) when H1N1 virus re-emerged into the human
population. This virus was later shown to be very similar to H1N1 strains circulating in
the 1950s (Nakajima, Desselberger et al. 1978) . Since viruses circulating in the wild
rapidly accumulate mutations, the most likely explanation for the re-emergence of this
strain is that it was stored in a laboratory for about 20 years until it was accidentally
released (Nakajima, Desselberger et al. 1978). Although people older than about 20
years were generally immune to this virus due to prior exposure, enough younger
people were non-immune for the H1N1 virus to sustain global infection. Since then, the
two subtypes H1N1 and H3N2 have continued circulating in the human population as
seasonal viruses, causing annual epidemics.
st

The world witnessed the first influenza pandemic of the 21

century in 2009-2010. The

pandemic started in April 2009, when a previously undescribed influenza virus of
subtype H1N1 was isolated from people in Mexico and US (“H1N1pdm09”). Within a
month of the initial identification, the virus had spread globally, infecting more than 8000
people and resulting in 74 deaths in 40 countries. According to CDC’s estimates,

6

between April 2009 and April 2010 H1N1pdm09 was responsible for about 61 million
infections, 274,000 hospitalizations, and 12,470 deaths in the US (Ross, Mahmood et
al. 2009), with about 201,200 deaths worldwide (Dawood, Iuliano et al. 2012).
Genetic analyses of the isolates suggested that the H1N1pdm09 virus derived its genes
from 3 genetic lineages of viruses: Eurasian swine lineage (an avian virus that entered
swine in 1979) provided NA and M; the classical swine H1N1 lineage (originally
descended from the 1918 human influenza, which had further evolved in swine since
shortly after 1918) provided HA, NS and NP; and the swine triple reassortant lineage (a
complex reassortant virus of swine that included avian and human segments, which
entered swine directly or through humans around 1998) provided PB1, PB2 and PA
(Garten, Davis et al. 2009) (figure 1.2).

Like the 20

th

century pandemic viruses, therefore, the H1N1pdm09 virus’s genes were

originally derived from avian viruses. These genes entered the swine genetic lineage,
evolved, and reassorted multiple times to make the pandemic virus. Thus, all the
pandemic viruses since the 20

th

century were either direct avian-to-human transmission

(1918), were reassortants of circulating human subtypes with avian viruses (1957 and
1968), or incorporated avian viruses that had been filtered through swine. History
therefore teaches that avian influenza viruses pose a great threat for public health.
Avian Influenza Viruses:
In general, most emerging human infections are of zoonotic nature; wild animals are
responsible for more than 70% of all emerging infections (Taylor, Latham et al. 2001). In

7

the case of influenza, wild waterfowl are the original reservoir for all influenza A viruses
in poultry and mammals, including humans (Olsen, Munster et al. 2006). Novel strains
or subtypes of influenza viruses from this reservoir may cross host species barrier and
infect humans either directly by birds or via intermediate hosts such as swine.
Subsequently, adaptation of this cross species viruses may lead to human influenza
pandemics, as is already evident from history.
Avian influenza was first clearly described in 1878, in Northern Italy, as a severe
disease of domestic poultry and wild birds. It was originally known as “fowl plague” and
it was unclear if it was distinct from other avian diseases such as fowl cholera (reviewed
in (Kaleta and Rülke 2009)). In 1950, however, the disease was confirmed to be caused
by Influenza A virus (reviewed in (Krauss and Webster 2010)).
Avian influenza viruses (AIVs) are typically asymptomatic in their reservoir hosts, which
are usually wild waterfowl. Some subtypes of virus regularly infect domestic poultry
(chickens, ducks, and turkeys), and may be subclinical or cause clinical disease. Based
on their pathogenicity in domestic poultry, AIVs are categorized into two types: highly
pathogenic avian influenza viruses (HPAIV) and low pathogenicity avian influenza
viruses (LPAIV) (reviewed in (Senne 2010). To date, all known HPAIVs have belonged
to two HA types; H5 and H7 (reviewed in (Senne 2010)). A critical determinant of
pathogenicity is the cleavage site of the HA: In HPAIV, this cleavage site includes a
number of basic residues. This polybasic cleavage site can arise by mutation of the
normal site, so that HP viruses evolve from LP viruses (Banks, Speidel et al. 2001).

8

The first evidence of human infection with avian origin virus was identified during the
Asian influenza pandemic (H2N2, 1957) when an avian influenza virus isolated from a
turkey (A/Turkey/Massachusetts/3740/65(H6N2)) reacted strongly with antisera to the
1957 pandemic strain (Pereira, Tumova et al. 1967). Subsequently, human infections
with both LPAIV and HPAIV were detected. Human LPAIV infections have resulted from
cross-species transmission of LPAIV H9N2, H7N2, H7N3, and H7N7 and H7N9 (Table
2, reviewed in ref (Reperant, Kuiken et al. 2012)), as well as very rare instances of
H10N7 and H6N1 viruses (Li, Li et al. 2006, Zhu, Wang et al. 2010, Arzey, Kirkland et
al. 2012). Most of these LPAIV infections of humans have caused very mild or
subclinical infections, with conjunctivitis or mild respiratory symptoms most common. In
humans, HPAIV infections have resulted from cross-species transmission of HPAIV
H5N1, H7N3 and H7N7 (Table 1.2, reviewed in ref (Reperant, Kuiken et al. 2012)).
Although infection of humans with the HPAIV H5N1 strain is frequently a very severe
disease, virus pathogenicity in poultry is not a perfect predictor of human disease, since
the LPAIV H7N9, which causes little disease in poultry, has a high mortality rate in
humans, whereas HPAIV H7N3 viruses have not been associated with severe disease
to date (Tweed, Skowronski et al. 2004, Centers for Disease and Prevention 2012,
Lopez-Martinez, Balish et al. 2013).
As well as respiratory disease, atypical disease manifestations, such as gastrointestinal
or neurological symptoms, have been reported following infection with HPAIV H5N1,
while ocular infection was reported for HPAIV of the H7 subtype. In most of these
cases, HPAIV are transmitted from birds to humans via the respiratory or ocular route,
although some strains of HPAIV H5N1 may use both oral and respiratory routes of
9

infection in other mammals and possibly also in humans. So far, little human-to-human
transmission of LPAIV or HPAIV has been reported.
H5N1 viruses are now endemic in many countries. H5N1 became a public health
concern in 1997 when a poultry outbreak in Hong Kong resulted in 18 human infections
with six fatalities (Yuen, Chan et al. 1998). Since then, 608 human cases with case
fatality rate of 59% have been confirmed as of Aug. 10, 2012 (WHO 2012). H5N1
viruses are of particular pandemic concern because of their endemicity in many
countries, constant evolution and high case fatality rate.
H7 influenza viruses have also caused outbreaks in domestic birds in all parts of the
globe; USA (Suarez, Garcia et al. 1999, Spackman, Senne et al. 2003), Canada
(Pasick, Handel et al. 2005, Berhane, Hisanaga et al. 2009), China (Li, Li et al. 2006),
Pakistan (Abbas, Spackman et al. 2010), Chile (Spackman, McCracken et al. 2006),
Italy (Banks, Speidel et al. 2001, Campitelli, Mogavero et al. 2004, Di Trani, Bedini et al.
2004, Campitelli, Di Martino et al. 2008), Germany (Rohm, Suss et al. 1996),
Netherlands (Stegeman, Bouma et al. 2004) and Australia (Selleck, Arzey et al. 2003,
Bulach, Halpin et al. 2010). H7 human infections have been rare events (reviewed in
(Belser, Bridges et al. 2009)), with two exceptions in which large numbers of people
were infected. In 2003, a H7N7 outbreak occurred in The Netherlands, which resulted
in at least 89 human infections and one death (Koopmans, Wilbrink et al. 2004). The
number of infected people is based on virus isolate confirmed cases; however the
actual number is suggested to be much higher as evidenced by serological studies
(Meijer, Bosman et al. 2006). During this outbreak, both bird-to-human and limited

10

human-to-human transmission of H7 viruses were reported (Koopmans, Wilbrink et al.
2004).
Since 2003, sporadic H7 human infections have been reported in USA, UK and
Canada; until 2013, unlike H5N1 infections, H7 virus infections of humans mostly
caused ocular symptoms and no mortality (reviewed in (Belser, Bridges et al. 2009)). H7
viruses, although less virulent than the H5N1 subtype, are still of significant pandemic
concern because of their ability to infect humans and other mammals and the
compatibility of H7 HA with multiple NAs. Unlike most other HAs which only occur with 1
or 2 NAs, H7 has been found with 5 different NAs, 3 among which (H7N2, H7N3, H7N7)
have caused infection in humans (reviewed in (Belser, Bridges et al. 2009)). This
suggests that H7 viruses have more potential, among circulating strains, to make new
reassortant viruses and thus adapt to human population.
The risk of pandemics caused by avian influenza viruses is further emphasized by the
2013 Chinese influenza outbreak where a novel avian influenza subtype (H7N9)
entered the human population for the first time, leading to 135 laboratory-confirmed
human cases and 44 deaths (Campitelli, Di Martino et al. 2008). The outbreak started in
February 2013 and within 2 months virus spread to multiple geographical locations in
China and infected more than 100 people (Saikh, Martin et al. 1995). Following the
closing of live bird markets by Chinese Government, the incidence subsided through the
summer months. Like other avian influenza viruses, this virus also lacks efficient human
to human transmission; however it is unique among human infections with H7 viruses
because of its high case fatality rate (32.6%); unlike previous H7 viruses which are
almost non-lethal in humans with mild symptoms. It is important to note that this virus is
11

a LPAIV that rarely makes birds sick, complicating surveillance of the virus’s spread in
poultry. Further, this incidence underscores the need for constant surveillance of avian
influenza viruses and strengthening our pandemic preparedness strategies.
HPAIVs do not yet have the ability of efficient human-to-human transmission. Among
the reasons for this inefficient human transmission is their preferential attachment to α2,
3-linked sialic acids which are rarely present in the upper regions of the human
respiratory tract (Shinya, Ebina et al. 2006). Despite their poor human-to-human
transmission capability, both H5N1 and H7 viruses appear to pose the biggest
pandemic risk. For example, HPAIVs might reassort with a circulating human strain,
thereby potentially integrating lethality (of H5N1) and transmissibility (of human strains).
Thus, these viruses pose a constant threat of emergence of new viruses due to genetic
shift, and pose a continuous pandemic threat to humans.
Pandemic Preparedness:
New influenza viruses are constantly emerging in animal reservoirs due to reassortment
of gene segments and accumulation of point mutations. To counteract the constant
threat posed by newly emerging viruses, we need to constantly strengthen our
pandemic preparedness by animal and human surveillance, assessing the risk of new
viruses, and making vaccine seed stocks in advance. New viruses having significant
genotypic variations from the previously circulating strains are evaluated for their
phenotype (such as virulence and transmission) in suitable animal models (particularly
mice and ferrets). Simultaneously, the immunogenicity (the ability to mount protective
antibody response) of these viruses are evaluated by hemagglutination inhibition assays

12

(HI), ELISA and microneutralization assays (MN). Of these, HI is best correlated with
protection against seasonal influenza viruses, with a titer of 40 or more considered to be
protective in humans (J.C. de Jong 2000). However, clear correlates of protection have
not been determined for avian viruses (Keitel and Atmar 2009, Haaheim and Katz 2011,
Katz, Hancock et al. 2011). Other factors, including geographic distribution, population
immunity, and potential human exposure, are also considered as part of risk
assessment (Trock, Burke et al. 2012).
I focused my dissertation research on the two most significant components of pandemic
preparedness: Risk assessment and pre-pandemic vaccine planning.
1. Determinants of Pathogenicity:
Pathogenicity of influenza viruses is a multigene trait and is well reviewed in (Medina
and Garcia-Sastre 2011) (table 1.3). Broadly, pathogenicity includes the ability of the
virus to bind with appropriate host cell receptors (HA protein), inhibit the host immune
reponse (NS1), replicate efficiently (polymerase genes) and transmit efficiently between
indviduals (NA and multiple other gene products). Pathogenicity therefore depends on
multiple pathogenic determinants, the number of which keep increasing with the
evolution of new viruses and with new research.
Segment 2 of IAV genome captured our interest because of the fact that it was the only
gene besides surface glycoproteins (HA and NA) which was derived from avian viruses
in all the pandemic viruses. Furthermore, its role and mechanism in pandemic viruses is
still not fully understood. We and others have hypothesized that PB1 might be important
in viral virulence because of PB1-F2 (see below). All the 20
13

th

century pandemic viruses

expressed full length PB1-F2, however the 2009 H1N1pdm09 virus lacked fuctional
protein. Furthermore mammalian viruses generally have a truncated version of PB1-F2
while avian viruses are mostly full length, suggesting a role in host specific transmission
and pathogenic phenotypes. Analysis of this gene product became the first part of my
dissertation research.
2. Pandemic Vaccines:
After the identification of new viruses which differ significantly in their antigenicity,
virulence, and/or transmissibility in model animals or humans, there arises the need of
making new vaccine seed stocks. Selection of a virus for vaccine seed stock depends
on its ability to mount protective serum antibody titer and to show protection against a
majority of related viruses. Vaccine seed stocks are produced by reassortment of the
novel virus with a virus that is non-pathogenic to humans, A/Puerto Rico/8/34(H1N1), to
produce a reassortant virus that is antigenically like the novel virus but that can be
grown to high titers in eggs and manufactured safely.

In addition to classical

reassortment in eggs, vaccine stocks can be produced by reverse genetics (Hoffmann,
Neumann et al. 2000), and multiple other vaccine systems are being investigated
(Kreijtz, Osterhaus et al. 2009).
Gaps in Pandemic Preparedness:
In our review of literature we identified two significant gaps in the pandemic
preparedness of HPAIVs. (i) Risk assessment of truncated PB1-F2 protein in H5N1
viruses. (ii) Poor immunogenicity of H7 viruses.

14

1. H5N1 PB1-F2:
PB1-F2 is a short (87–90 a.a.) IAV protein encoded by the +1 reading frame of PB1
(Chen, Calvo et al. 2001). It was discovered serendipitously in 2001 during a search for
putative CD8 T cell epitopes encoded by alternative reading frames (Chen, Calvo et al.
2001). The potential importance of PB1 segment in mammalian pathogenicity and
adaptation is based on the fact that it was the only gene segment (other than the
glycoprotein genes, HA and NA) that was exchanged in the pandemic events of 1957
and 1968 (Kawaoka, Krauss et al. 1989). PB1’s role in pathogenicity and transmission
is still not fully understood but could be at least partially explained by the discovery of
PB1-F2 [30], which was present as a full-length protein in the first 3 pandemics but not
in the less-virulent 2009 pandemic.
PB1-F2 is relatively unstable and undergoes proteasomal degradation shortly after
expression (Chen, Calvo et al. 2001). It localizes in mitochondria and, at least in some
cell types, causes cell death (Chen, Calvo et al. 2001, Gibbs, Malide et al. 2003).
Although it is not required for viral replication in eggs or in cultured cells, it has been
associated with virulence in mice and ducks (Chen, Calvo et al. 2001, Zamarin,
Ortigoza et al. 2006, McAuley, Hornung et al. 2007, McAuley, Chipuk et al. 2010,
Schmolke, Manicassamy et al. 2011). Its potential importance was further demonstrated
by a study in mice that resulted in attenuation of 1918 H1N1 pandemic virus after its
polymerase segments (including PB1) were replaced with those of the modern H1N1
strain (Tumpey, Basler et al. 2005).Full length PB1-F2 inhibits IFN-β when compared
with deleted PB1-F2, and the N66S mutation also increases IFN-β inhibitory effect
(Conenello, Tisoncik et al. 2011, Dudek, Wixler et al. 2011, Varga, Ramos et al. 2011).
15

One recent study showed the opposite effect where PB1-F2 increased IFN-β expression
in epithelial cells (A549 and BEAS-2B) while it had no effect in immune cells (U-937 and
Jurkat) (Le Goffic, Bouguyon et al. 2010). Such conflicting studies suggest that the
effect of PB1-F2 may depend heavily on its genomic context
This virulence-enhancing effect of PB1-F2 is virus strain specific, since it is particularly
seen in the viruses which have a PB1 gene of recent avian origin (McAuley, Chipuk et
al. 2010). This effect was not appreciable in mice infected with a highly virulent virus
(A/WSN/33(H1N1)) but was obvious in low pathogenic viruses (McAuley, Hornung et al.
2007). This may be explained by masking of the PB1-F2 effect by the presence of
multiple

pathogenic

determinants

of

IAVs.

In

one

study

with

A/Viet

Nam/1203/2004(H5N1) virus (HPAIV), deletion of PB1-F2 caused delayed onset of
clinical symptoms and systemic spreading of virus (Schmolke, Manicassamy et al.
2011).
The mechanisms underlying the pathogenic effects of PB1-F2 are still not clear. It has
been suggested to be due to its mitochondrial targeting and apoptosis activity (Chen,
Calvo et al. 2001, Gibbs, Malide et al. 2003, Zamarin, Garcia-Sastre et al. 2005) but in
one study, PR8 virus that lacked PB1-F2 had reduced replication and virus spreading
independent of its pro-apoptotic features (Mazur, Anhlan et al. 2008). Whatever the
mechanism might be, deletion of PB1-F2 has resulted in altered viral virulence in a virus
strain- and host-specific manner.
Further, molecular analysis suggests a role of PB1-F2 in adaptation of AIVs to new
hosts following interspecies transmission. PB1-F2 was found to be under positive

16

selection based on the analysis of non-synonymous and synonymous mutations
(Obenauer, Denson et al. 2006, Smith, Naipospos et al. 2006). This inference was
questioned because of overlapping of PB1-F2 frame with PB1 (so that non-synonymous
PB1-F2 mutations might merely be due to synonymous mutations in PB1) (Holmes,
Lipman et al. 2006). However, a more recent study also found PB1-F2 under positive
selection pressure while considering PB1 frame (Suzuki 2006).
Several previous papers have noted that PB1-F2 of human influenza viruses has
become truncated due to premature stop codons. For example, Zell et al. analyzed
2226 PB1 sequences and found that the proportion of truncated PB1-F2 was higher in
human viruses (19%) than those of avian viruses (4%) (Table 1 and S1 in reference
(Zell, Krumbholz et al. 2007)). That study included data available until June 15, 2006,
and the number of available sequences has increased dramatically since then. We
confirmed this analysis using over 15,000 public and non-public influenza virus
sequences. However, in this analysis, we observed a new trend in avian influenza
viruses. Whereas in previous analyses B1-F2 truncations only appeared as sporadic,
transient events in avian influenza viruses, we found evidence for a truncation in PB1F2 of Vietnam H5N1 viruses that persisted for long periods, and expanded
geographically. Collectively the proposed role of truncated PB1-F2 in mammalian
adaptation, its interference with host innate immune response, constant evolution of
H5N1 viruses and novel persistence of PB1-F2 truncations in avian viruses, makes it
one of the most important targets for pandemic preparedness studies.
To summarize, the dramatically higher prevalence of truncated PB1-F2 in human
viruses when compared to avian viruses suggests that viral adaptation to mammals vs.
17

avian species may be enhanced by PB1-F2 truncation. To fill this gap in pandemic
preparedness of H5N1 viruses, we proposed to test the pathogenicity of H5N1 viruses
with full-length or truncated PB1-F2 in mice model and in vitro in mammalian and avian
cells (Chapter 2 of this dissertation).
2.

Poor Immunogenicity of H7 Viruses:

The second and the most important step in pandemic preparedness is the development
of pre-pandemic vaccine virus seed stocks. The pandemic vaccine candidates for H7,
H5 and H9 viruses have continually been made, updated, and as deemed appropriate,
tested in human clinical trials.
Avian influenza vaccines typically have not induced antibody titers comparable to the
seasonal flu vaccines. Plans for potential deployment of H5 and H9 vaccines include the
use of high HA doses, and administration of vaccine with adjuvants. These
modifications have generally induced antibody titers closer to those of seasonal
vaccines (reviewed in (Keitel and Atmar 2009)). However, even at high HA doses, in
limited trials H7 vaccines have performed poorly. If H7 vaccines are ineffective due to
poor immunogenicity, this would represent a significant gap in pandemic preparedness
(Joseph, McAuliffe et al. 2008, Cox, Madhun et al. 2009).
Several pieces of evidence suggest that H7 viruses and vaccines are poorly
immunogenic. First, human sero-surveillance studies rarely detect antibodies to H7
viruses (15, 16); even in the people who had close contact with H7-infected poultry
(Russell, Gamblin et al. 2004, Jia, de Vlas et al. 2009). Second, only 2 out of 89 RTPCR and virus isolation confirmed cases (in the 2003 Netherlands H7N7 outbreak)
showed HI titers of 40, which for human seasonal influenzas is considered protective
18

(Meijer, Bosman et al. 2006). Third, a clinical trial using a monovalent H7N7 inactivated
vaccine made by Sanofi Pasteur Inc yielded very poor antibody responses (Couch,
Patel et al. 2012). In this trial, each group of 24 or 25 people were immunized with two
doses of 7.5 μg, 15 μg (the usual dose of HA in seasonal vaccines), 45 μg and 90 μg (a
common dose for H5N1 non-adjuvanted vaccines). Only 1 of 99 vaccinees showed HI
titers of 40 or more (Couch, Patel et al. 2012). All these evidences suggest that H7
viruses and vaccines may be poorly immunogenic, but the reasons remain largely
unknown.
Poor immunogenicity of H7 viruses as suggested by HI and MN assays (13-16) could
be interpreted in different directions. First, it could be because of the absence of optimal
immunological assays and the incompletely understood correlates of protection for
these viruses. For example, H7 viruses may induce antibodies against the HA stalk
region than its globular head; such antibodies would be poorly detected by HI assays,
although MN assays would still detect anti-stalk antibodies that are neutralizing. Or,
protection may come from anti-NA or M2 antibodies, or cell-mediated immunity may
play an important role. By this interpretation, immunity induced by H7 vaccines may be
fully protective, but poorly detected. However, since the only way to test this in humans
may be to challenge people with virulent viruses, this interpretation is best kept
hypothetical.
Secondly, the poor immunogenicity may truly be an intrinsic characteristic of H7 viruses
and may reflect a true lack of protection following H7 immunization. This could also be
because of many reasons. For example, HA epitopes may be hidden because of
glycosylation sites; viruses may replicate poorly; or virus molecules may be poorly
19

detected by the by innate immune receptors etc. One potential reason for poor detection
by the innate immune response may be reduced overall glycosylation, since avian
influenza viruses tend to have fewer carbohydrates than do seasonal influenza viruses.
Based on our current knowledge of H7 viruses we cannot determine which of these is
most correct, because of lack of studies on H7 viruses. However, it is clear that there is
a significant need to study the immune response to H7 viruses and to standardize
protective correlates for these viruses. To fill this gap, we proposed the hypothesis that
“H7-HA is inherently less immunogenic than seasonal influenza viruses”. To test
our hypothesis, we evaluated and compared immune response of purified HAs of H7
and seasonal viruses using different serological tests (Chapter 3 of this dissertation).
Summary:
In summary, H5N1 and H7 influenza viruses may pose a great pandemic threat. H5N1
pandemic risk is high due to its endemicity in many bird species in many countries,
constant evolution in birds, and high case fatality rate in humans. PB1, particularly
originating from avian species, seems to play an important role in pathogenicity of
reassortants, and functional PB1-F2 may be implicated in viral fitness, pathogenicity,
and species adaptation. Since PB1-F2 is truncated in a large fraction of mammalian
viruses and present as a full length protein in avian species (Zell, Krumbholz et al.
2007), it may have some role in the adaptation of virus in mammalian species after
avian-mammalian transmission. The pandemic potential of H7 viruses is high because
of their abilities to infect humans, compatibility of H7 HA with multiple NAs, and their
potential for reassortment and human adaptation. Pandemic preparedness for H7

20

viruses is hampered by the lack of highly effective vaccine seed stocks because of their
poor immunogenicity. In this study, we propose to explore these gaps and strengthen
pandemic preparedness for HPAIVs.

21

APPENDICES

22

APPENDIX 1A: Tables

Table 1.1 Influenza A virus genes and proteins.
Gene
ID
1

Segment

Protein name

Protein function

Polymerase B2
(PB2)

Polymerase B2
(PB2)

Internal protein, Virus
Replication

2

Polymerase B1
(PB1)

Polymerase B1
(PB1)
PB1-F2

3

Polymerase A
(PA)

Polymerase A
(PA)

Internal protein, Virus
replication
Mitochondrial targeting,
Apoptosis
Internal protein, Virus
replication

4

Hemagglutinin
(HA)

Hemagglutinin
(HA)

5

Nucleoprotein
(NP)

Nucleoprotein
(NP)

6

Neuraminidase
(NA)

Neuraminidase
(NA)

7

Matrix (M)

Matrix 1 (M1)
Matrix 2 (M2)

8

Non-structural
(NS)

Non-structural
1 (NS1)
Non-structural
2 (NS2)

Surface glycoprotein, viral
attachment, Antigenic
determinant
Nucleocapsid protein, RNA
coating, nuclear targeting,
RNA transcription
Surface glycoprotein,
antigenic determinant, viral
release from host cells
Membrane protein stability
Membrane protein, viral
uncoating
Viral immune evasion
Regulation of virus life cycle,
specially mRNA transcription
and localization of viral
ribonucleic proteins

((adapted from (Lamb and Krug 2001) with minor modifications)).

23

Table 1.2 Reported cases of avian influenza virus infections by viral isolation.
Host of
origin
Wild birds,
Swans
Chicken,
domestic
ducks
Chicken
Chicken
Chicken

Year, Place

Chicken
Chicken,
turkeys
Turkeys
Domestic
ducks
Unknown
Unknown
Unknown
Unknown
Unknown

Pathotype*
HP

Subtype Associated activity

1997-present,
Asia and Africa

HP

H5N1

2007, UK
2006, UK
2004, Canada

LP
LP
HP

H7N2
H7N3
H7N3

2003,
Netherlands
2002-2003, Italy

HP

H7N7

LP

H7N3

Visiting bird market
Caregiving
Caregiving, managing
outbreak
Caregiving, managing
outbreak
Managing outbreak

2002, USA
1996, UK

LP
LP

H7N2
H7N7

Managing outbreak
Caregiving

2005, Azerbaijan

H5N1

De-feathering, meat
processing
Caregiving, Butchering,
De-feathering

2007, Hong Kong LP
H9N2
Unknown
2003, USA
LP
H7N2
Unknown
2003, Hong Kong LP
H9N2
Unknown
1999, Hong Kong LP
H9N2
Unknown
1998, China
LP
H9N2
Unknown
((adapted from (Reperant, Kuiken et al. 2012) with minor modifications)).

*HP- Highly pathogenic; LP- low pathogenic.

24

Table 1.3 Molecular determinants of IAV pathogenicity.
Virulence
marker and
pathogenic
determinant

HA
Sialic acid
linkage
specificity
Receptor
binding
specificity
Multibasic
cleavage site
PB1–F2
S66
(associated
with increased
virulence), N66
or truncation
PB2
Adaptation to
mammalian
hosts
NS1
PDZ domainbinding motif

Pandemic virus

HPAI virus

Contemporary
seasonal
virus

1918
H1N1

1957
H2N2

1968
H3N2

2009 H5N1
H1N1

H7N7

H1N1

H3N2

α-2,6

α-2,6

α-2,6

α-2,3

α-2,3

α-2,6

α-2,6

D190
and
D225
No

Q226
and
N186
No

L226
and
S228
No

α-2,6
and
α-2,3
D190
and
D225
No

Q196,
Q226,
G228
Yes

Q226

D190
and
D225
No

L226
and
S228
No

S66

N66

N66

TR*

S66 or N66
N66

TR*

N66

K627

K627

K627

R591

K627
or
N701

K627
or
N701

K627

K627

KSEV

RSKV

RSKV TR*

ESEV
or
EPEV
Yes

ESEV

RSEV

RSKV

Yes

E91 associated No
No
No
No
No
No
No
with virulence
M2
Resistant to
No
No
No
Yes
Yes
No
No
Yes
adamantanes
and
(presence of
no
N31)
NA
Resistant to
No
No
No
No
No
No
Yes
No
oseltamavir
(presence of
Y275 & S294)
(adapted from (Medina and Garcia-Sastre 2011) with minor modifications).
TR* means truncated proteins which do not express the mentioned region.
25

APPENDIX 1B: Figures

Figure 1.1 Structure of influenza A virus (adapted from (Nelson and Holmes 2007)).
HA (Hemagglutinin) and NA (Neuraminidase) are the major glycoproteins on virus
envelop. M1 protein makes the virus capsid which contains 8 gene segments, each
bound to one molecule of each PA, PB1 and PB2 (polymerase complex) and multiple
molecules of NP (nucleoprotein); the complex together is called RNP
(ribonucleoprotein).

26

Figure 1.2 Genetic components of 2009 H1N1pdm09 virus (Adapted from (Garten,
Davis et al. 2009). The figure shows interspecies transmission and evolution of gene
segments of 2009 H1N1pdm09 virus.

27

REFERENCES

28

REFERENCES

Abbas, M. A., E. Spackman, D. E. Swayne, Z. Ahmed, L. Sarmento, N. Siddique, K.
Naeem, A. Hameed and S. Rehmani (2010). "Sequence and phylogenetic analysis of
H7N3 avian influenza viruses isolated from poultry in Pakistan 1995-2004." Virol J 7:
137.
Arzey, G. G., P. D. Kirkland, K. E. Arzey, M. Frost, P. Maywood, S. Conaty, A. C. Hurt,
Y. M. Deng, P. Iannello, I. Barr, D. E. Dwyer, M. Ratnamohan, K. McPhie and P. Selleck
(2012). "Influenza virus A (H10N7) in chickens and poultry abattoir workers, Australia."
Emerg Infect Dis 18(5): 814-816.
Banks, J., E. S. Speidel, E. Moore, L. Plowright, A. Piccirillo, I. Capua, P. Cordioli, A.
Fioretti and D. J. Alexander (2001). "Changes in the haemagglutinin and the
neuraminidase genes prior to the emergence of highly pathogenic H7N1 avian influenza
viruses in Italy." Arch Virol 146(5): 963-973.
Belser, J. A., C. B. Bridges, J. M. Katz and T. M. Tumpey (2009). "Past, present, and
possible future human infection with influenza virus A subtype H7." Emerg Infect Dis
15(6): 859-865.
Berhane, Y., T. Hisanaga, H. Kehler, J. Neufeld, L. Manning, C. Argue, K. Handel, K.
Hooper-McGrevy, M. Jonas, J. Robinson, R. G. Webster and J. Pasick (2009). "Highly
pathogenic avian influenza virus A (H7N3) in domestic poultry, Saskatchewan, Canada,
2007." Emerg Infect Dis 15(9): 1492-1495.
Bulach, D., R. Halpin, D. Spiro, L. Pomeroy, D. Janies and D. B. Boyle (2010).
"Molecular analysis of H7 avian influenza viruses from Australia and New Zealand:
genetic diversity and relationships from 1976 to 2007." J Virol 84(19): 9957-9966.
Campitelli, L., A. Di Martino, D. Spagnolo, G. J. Smith, L. Di Trani, M. Facchini, M. A. De
Marco, E. Foni, C. Chiapponi, A. M. Martin, H. Chen, Y. Guan, M. Delogu and I.
Donatelli (2008). "Molecular analysis of avian H7 influenza viruses circulating in Eurasia
in 1999-2005: detection of multiple reassortant virus genotypes." J Gen Virol 89(Pt 1):
48-59.
Campitelli, L., E. Mogavero, M. A. De Marco, M. Delogu, S. Puzelli, F. Frezza, M.
Facchini, C. Chiapponi, E. Foni, P. Cordioli, R. Webby, G. Barigazzi, R. G. Webster and
I. Donatelli (2004). "Interspecies transmission of an H7N3 influenza virus from wild birds
to intensively reared domestic poultry in Italy." Virology 323(1): 24-36.
Centers for Disease, C. and Prevention (2010). "Estimates of deaths associated with
seasonal influenza --- United States, 1976-2007." MMWR Morb Mortal Wkly Rep
59(33): 1057-1062.
29

Centers for Disease, C. and Prevention (2012). "Notes from the field: Highly pathogenic
avian influenza A (H7N3) virus infection in two poultry workers--Jalisco, Mexico, July
2012." MMWR Morb Mortal Wkly Rep 61(36): 726-727.
Chen, W., P. A. Calvo, D. Malide, J. Gibbs, U. Schubert, I. Bacik, S. Basta, R. O'Neill, J.
Schickli, P. Palese, P. Henklein, J. R. Bennink and J. W. Yewdell (2001). "A novel
influenza A virus mitochondrial protein that induces cell death." Nat Med 7(12): 13061312.
Conenello, G. M., J. R. Tisoncik, E. Rosenzweig, Z. T. Varga, P. Palese and M. G.
Katze (2011). "A single N66S mutation in the PB1-F2 protein of influenza A virus
increases virulence by inhibiting the early interferon response in vivo." J Virol 85(2):
652-662.
Couch, R. B., S. M. Patel, C. L. Wade-Bowers and D. Nino (2012). "A randomized
clinical trial of an inactivated avian influenza A (H7N7) vaccine." PLoS One 7(12):
e49704.
Cox, R. J., A. S. Madhun, S. Hauge, H. Sjursen, D. Major, M. Kuhne, K. Hoschler, M.
Saville, F. R. Vogel, W. Barclay, I. Donatelli, M. Zambon, J. Wood and L. R. Haaheim
(2009). "A phase I clinical trial of a PER.C6 cell grown influenza H7 virus vaccine."
Vaccine 27(13): 1889-1897.
Dawood, F. S., A. D. Iuliano, C. Reed, M. I. Meltzer, D. K. Shay, P. Y. Cheng, D.
Bandaranayake, R. F. Breiman, W. A. Brooks, P. Buchy, D. R. Feikin, K. B. Fowler, A.
Gordon, N. T. Hien, P. Horby, Q. S. Huang, M. A. Katz, A. Krishnan, R. Lal, J. M.
Montgomery, K. Molbak, R. Pebody, A. M. Presanis, H. Razuri, A. Steens, Y. O. Tinoco,
J. Wallinga, H. Yu, S. Vong, J. Bresee and M. A. Widdowson (2012). "Estimated global
mortality associated with the first 12 months of 2009 pandemic influenza A H1N1 virus
circulation: a modelling study." Lancet Infect Dis 12(9): 687-695.
Di Trani, L., B. Bedini, P. Cordioli, M. Muscillo, E. Vignolo, A. Moreno and M. Tollis
(2004). "Molecular Characterization of Low Pathogenicity H7N3 Avian Influenza Viruses
Isolated in Italy." Avian Dis 48(2): 376-383.
Dudek, S. E., L. Wixler, C. Nordhoff, A. Nordmann, D. Anhlan, V. Wixler and S. Ludwig
(2011). "The influenza virus PB1-F2 protein has interferon antagonistic activity." Biol
Chem 392(12): 1135-1144.
Elderfield, R. and W. Barclay (2011). "Influenza pandemics." Adv Exp Med Biol 719: 81103.
Garten, R. J., C. T. Davis, C. A. Russell, B. Shu, S. Lindstrom, A. Balish, W. M.
Sessions, X. Xu, E. Skepner, V. Deyde, M. Okomo-Adhiambo, L. Gubareva, J. Barnes,
C. B. Smith, S. L. Emery, M. J. Hillman, P. Rivailler, J. Smagala, M. de Graaf, D. F.
Burke, R. A. Fouchier, C. Pappas, C. M. Alpuche-Aranda, H. Lopez-Gatell, H. Olivera, I.
30

Lopez, C. A. Myers, D. Faix, P. J. Blair, C. Yu, K. M. Keene, P. D. Dotson, Jr., D.
Boxrud, A. R. Sambol, S. H. Abid, K. St George, T. Bannerman, A. L. Moore, D. J.
Stringer, P. Blevins, G. J. Demmler-Harrison, M. Ginsberg, P. Kriner, S. Waterman, S.
Smole, H. F. Guevara, E. A. Belongia, P. A. Clark, S. T. Beatrice, R. Donis, J. Katz, L.
Finelli, C. B. Bridges, M. Shaw, D. B. Jernigan, T. M. Uyeki, D. J. Smith, A. I. Klimov
and N. J. Cox (2009). "Antigenic and genetic characteristics of swine-origin 2009
A(H1N1) influenza viruses circulating in humans." Science 325(5937): 197-201.
Gibbs, J. S., D. Malide, F. Hornung, J. R. Bennink and J. W. Yewdell (2003). "The
influenza A virus PB1-F2 protein targets the inner mitochondrial membrane via a
predicted basic amphipathic helix that disrupts mitochondrial function." J Virol 77(13):
7214-7224.
Haaheim, L. R. and J. M. Katz (2011). "Immune correlates of protection against
influenza: challenges for licensure of seasonal and pandemic influenza vaccines, Miami,
FL, USA, March 1-3, 2010." Influenza Other Respi Viruses 5(4): 288-295.
Hoffmann, E., G. Neumann, Y. Kawaoka, G. Hobom and R. G. Webster (2000). "A DNA
transfection system for generation of influenza A virus from eight plasmids." Proc Natl
Acad Sci U S A 97(11): 6108-6113.
Holmes, E. C., D. J. Lipman, D. Zamarin and J. W. Yewdell (2006). "Comment on
"Large-scale sequence analysis of avian influenza isolates"." Science 313(5793): 1573;
author reply 1573.
Hoyert, D. L. and J. Xu (2012). "Deaths: Preliminary Data for 2011." National Vital
Statistics Reports 61(6): 1-56.
Influenza (Seasonal), W. H. O., April 2009. Retrieved August 10, 2013.
J.C. de Jong, W. E. B., A.M. Palache, G.F. Rimmelzwaan, A.D. Osterhaus (2000).
"Mismatch between the 1997/1998 influenza vaccine and the major epidemic A(H3N2)
virus strain as the cause of an inadequate vaccine-induced antibody response to this
strain in the elderly." J. Med. Virol 61(1): 94-99.
Jagger, B. W., H. M. Wise, J. C. Kash, K. A. Walters, N. M. Wills, Y. L. Xiao, R. L.
Dunfee, L. M. Schwartzman, A. Ozinsky, G. L. Bell, R. M. Dalton, A. Lo, S. Efstathiou, J.
F. Atkins, A. E. Firth, J. K. Taubenberger and P. Digard (2012). "An overlapping proteincoding region in influenza A virus segment 3 modulates the host response." Science
337(6091): 199-204.
Jia, N., S. J. de Vlas, Y. X. Liu, J. S. Zhang, L. Zhan, R. L. Dang, Y. H. Ma, X. J. Wang,
T. Liu, G. P. Yang, Q. L. Wen, J. H. Richardus, S. Lu and W. C. Cao (2009).
"Serological reports of human infections of H7 and H9 avian influenza viruses in
northern China." J Clin Virol 44(3): 225-229.

31

Joseph, T., J. McAuliffe, B. Lu, L. Vogel, D. Swayne, H. Jin, G. Kemble and K.
Subbarao (2008). "A live attenuated cold-adapted influenza A H7N3 virus vaccine
provides protection against homologous and heterologous H7 viruses in mice and
ferrets." Virology 378(1): 123-132.
Kaleta, E. F. and C. P. A. Rülke (2009). "The Beginning and Spread of Fowl Plague (H7
High Pathogenicity Avian Influenza) across Europe and Asia (1878–1955), in Avian
Influenza (ed D. E. Swayne), Blackwell Publishing Ltd., Oxford, UK. doi:
10.1002/9780813818634.ch7."
Katz, J. M., K. Hancock and X. Xu (2011). "Serologic assays for influenza surveillance,
diagnosis and vaccine evaluation." Expert Rev Anti Infect Ther 9(6): 669-683.
Kawaoka, Y., S. Krauss and R. G. Webster (1989). "Avian-to-human transmission of the
PB1 gene of influenza A viruses in the 1957 and 1968 pandemics." J Virol 63(11): 46034608.
Keitel, W. A. and R. L. Atmar (2009). "Vaccines for pandemic influenza: summary of
recent clinical trials." Curr Top Microbiol Immunol 333: 431-451.
Koopmans, M., B. Wilbrink, M. Conyn, G. Natrop, H. van der Nat, H. Vennema, A.
Meijer, J. van Steenbergen, R. Fouchier, A. Osterhaus and A. Bosman (2004).
"Transmission of H7N7 avian influenza A virus to human beings during a large outbreak
in commercial poultry farms in the Netherlands." Lancet 363(9409): 587-593.
Krauss, S. and R. G. Webster (2010). "Avian influenza virus surveillance and wild birds:
past and present." Avian Dis 54(1 Suppl): 394-398.
Kreijtz, J. H. C. M., A. D. M. E. Osterhaus and G. F. Rimmelzwaan (2009). "Vaccination
strategies and vaccine formulations for epidemic and pandemic influenza control."
Human Vaccines 5(3): 126-135.
Lamb, R. A. and R. M. Krug (2001). "Orthomyxoviridae: The viruses and their
replication. In: Fields BN, Knipe DM, Howley PM, eds. Fields Virology. 4th ed.
Philadelphia, Pennsylvania: Lippincott-Raven; 2001:1487-1532.".
Le Goffic, R., E. Bouguyon, C. Chevalier, J. Vidic, B. Da Costa, O. Leymarie, C.
Bourdieu, L. Decamps, S. Dhorne-Pollet and B. Delmas (2010). "Influenza A virus
protein PB1-F2 exacerbates IFN-beta expression of human respiratory epithelial cells."
J Immunol 185(8): 4812-4823.
Li, Q., X. Sun, Z. Li, Y. Liu, C. J. Vavricka, J. Qi and G. F. Gao (2012). "Structural and
functional characterization of neuraminidase-like molecule N10 derived from bat
influenza A virus." Proc Natl Acad Sci U S A 109(46): 18897-18902.

32

Li, Y., C. Li, L. Liu, H. Wang, C. Wang, G. Tian, R. G. Webster, K. Yu and H. Chen
(2006). "Characterization of an avian influenza virus of subtype H7N2 isolated from
chickens in northern China." Virus Genes 33(1): 117-122.
Lopez-Martinez, I., A. Balish, G. Barrera-Badillo, J. Jones, T. E. Nunez-Garcia, Y. Jang,
R. Aparicio-Antonio, E. Azziz-Baumgartner, J. A. Belser, J. E. Ramirez-Gonzalez, J. C.
Pedersen, J. Ortiz-Alcantara, E. Gonzalez-Duran, B. Shu, S. L. Emery, M. K. Poh, G.
Reyes-Teran, J. A. Vazquez-Perez, S. Avila-Rios, T. Uyeki, S. Lindstrom, J. Villanueva,
J. Tokars, C. Ruiz-Matus, J. F. Gonzalez-Roldan, B. Schmitt, A. Klimov, N. Cox, P. KuriMorales, C. T. Davis and J. A. Diaz-Quinonez (2013). "Highly Pathogenic Avian
Influenza A(H7N3) Virus in Poultry Workers, Mexico, 2012." Emerg Infect Dis 19(9).
Mazur, I., D. Anhlan, D. Mitzner, L. Wixler, U. Schubert and S. Ludwig (2008). "The
proapoptotic influenza A virus protein PB1-F2 regulates viral polymerase activity by
interaction with the PB1 protein." Cell Microbiol 10(5): 1140-1152.
McAuley, J. L., J. E. Chipuk, K. L. Boyd, N. Van De Velde, D. R. Green and J. A.
McCullers (2010). "PB1-F2 proteins from H5N1 and 20 century pandemic influenza
viruses cause immunopathology." PLoS Pathog 6(7): e1001014.
McAuley, J. L., F. Hornung, K. L. Boyd, A. M. Smith, R. McKeon, J. Bennink, J. W.
Yewdell and J. A. McCullers (2007). "Expression of the 1918 influenza A virus PB1-F2
enhances the pathogenesis of viral and secondary bacterial pneumonia." Cell Host
Microbe 2(4): 240-249.
Medina, R. A. and A. Garcia-Sastre (2011). "Influenza A viruses: new research
developments." Nat Rev Microbiol 9(8): 590-603.
Meijer, A., A. Bosman, E. E. van de Kamp, B. Wilbrink, M. Du Ry van Beest Holle and
M. Koopmans (2006). "Measurement of antibodies to avian influenza virus A(H7N7) in
humans by hemagglutination inhibition test." J Virol Methods 132(1-2): 113-120.
Molinari, N. A., I. R. Ortega-Sanchez, M. L. Messonnier, W. W. Thompson, P. M.
Wortley, E. Weintraub and C. B. Bridges (2007). "The annual impact of seasonal
influenza in the US: measuring disease burden and costs." Vaccine 25(27): 5086-5096.
Nakajima, K., U. Desselberger and P. Palese (1978). "Recent human influenza A
(H1N1) viruses are closely related genetically to strains isolated in 1950." Nature
274(5669): 334-339.
Nelson, M. I. and E. C. Holmes (2007). "The evolution of epidemic influenza." Nat Rev
Genet 8(3): 196-205.
O'Neill, R. E., J. Talon and P. Palese (1998). "The influenza virus NEP (NS2 protein)
mediates the nuclear export of viral ribonucleoproteins." EMBO J 17(1): 288-296.

33

Obenauer, J. C., J. Denson, P. K. Mehta, X. Su, S. Mukatira, D. B. Finkelstein, X. Xu, J.
Wang, J. Ma, Y. Fan, K. M. Rakestraw, R. G. Webster, E. Hoffmann, S. Krauss, J.
Zheng, Z. Zhang and C. W. Naeve (2006).
"Large-scale sequence analysis of avian influenza isolates." Science 311(5767): 15761580.
Olsen, B., V. J. Munster, A. Wallensten, J. Waldenstrom, A. D. Osterhaus and R. A.
Fouchier (2006). "Global patterns of influenza a virus in wild birds." Science 312(5772):
384-388.
Pasick, J., K. Handel, J. Robinson, J. Copps, D. Ridd, K. Hills, H. Kehler, C. CottamBirt, J. Neufeld, Y. Berhane and S. Czub (2005). "Intersegmental recombination
between the haemagglutinin and matrix genes was responsible for the emergence of a
highly pathogenic H7N3 avian influenza virus in British Columbia." J Gen Virol 86(Pt 3):
727-731.
Pereira, H. G., B. Tumova and R. G. Webster (1967). "Antigenic relationship between
influenza A viruses of human and avian origins." Nature 215(5104): 982-983.
Reperant, L. A., T. Kuiken and A. D. Osterhaus (2012). "Influenza viruses: from birds to
humans." Hum Vaccin Immunother 8(1): 7-16.
Rohm, C., J. Suss, V. Pohle and R. G. Webster (1996). "Different hemagglutinin
cleavage site variants of H7N7 in an influenza outbreak in chickens in Leipzig,
Germany." Virology 218(1): 253-257.
Ross, T. M., K. Mahmood, C. J. Crevar, K. Schneider-Ohrum, P. M. Heaton and R. A.
Bright (2009). "A trivalent virus-like particle vaccine elicits protective immune responses
against seasonal influenza strains in mice and ferrets." PLoS One 4(6): e6032.
Russell, R. J., S. J. Gamblin, L. F. Haire, D. J. Stevens, B. Xiao, Y. Ha and J. J. Skehel
(2004). "H1 and H7 influenza haemagglutinin structures extend a structural
classification of haemagglutinin subtypes." Virology 325(2): 287-296.
Saikh, K., J. Martin, A. Nishikawa and S. Dillon (1995). "Influenza A virus-specific H-2d
restricted cross-reactive cytotoxic T lymphocyte epitope(s) detected in the
hemagglutinin HA2 subunit of A/Udorn/72." Virology 214(2): 445-452.
Schmolke, M., B. Manicassamy, L. Pena, T. Sutton, R. Hai, Z. T. Varga, B. G. Hale, J.
Steel, D. R. Perez and A. Garcia-Sastre (2011). "Differential contribution of PB1-F2 to
the virulence of highly pathogenic H5N1 influenza A virus in mammalian and avian
species." PLoS Pathog 7(8): e1002186.

34

Selleck, P. W., G. Arzey, P. D. Kirkland, R. L. Reece, A. R. Gould, P. W. Daniels and H.
A. Westbury (2003). "An Outbreak of Highly Pathogenic Avian Influenza in Australia in
1997 Caused by an H7N4 Virus." Avian Dis 47(s3): 806-811.
Senne, D. A. (2010). "Avian influenza in North and South America, the Caribbean, and
Australia, 2006-2008." Avian Dis 54(1 Suppl): 179-186.
Shinya, K., M. Ebina, S. Yamada, M. Ono, N. Kasai and Y. Kawaoka (2006). "Avian flu:
influenza virus receptors in the human airway." Nature 440(7083): 435-436.
Smith, G. J., T. S. Naipospos, T. D. Nguyen, M. D. de Jong, D. Vijaykrishna, T. B.
Usman, S. S. Hassan, T. V. Nguyen, T. V. Dao, N. A. Bui, Y. H. Leung, C. L. Cheung, J.
M. Rayner, J. X. Zhang, L. J. Zhang, L. L. Poon, K. S. Li, V. C. Nguyen, T. T. Hien, J.
Farrar, R. G. Webster, H. Chen, J. S. Peiris and Y. Guan (2006). "Evolution and
adaptation of H5N1 influenza virus in avian and human hosts in Indonesia and
Vietnam." Virology 350(2): 258-268.
Spackman, E., K. G. McCracken, K. Winker and D. E. Swayne (2006). "H7N3 avian
influenza virus found in a South American wild duck is related to the Chilean 2002
poultry outbreak, contains genes from equine and North American wild bird lineages,
and is adapted to domestic turkeys." J Virol 80(15): 7760-7764.
Spackman, E., D. A. Senne, S. Davison and D. L. Suarez (2003). "Sequence analysis of
recent H7 avian influenza viruses associated with three different outbreaks in
commercial poultry in the United States." J Virol 77(24): 13399-13402.
Stegeman, A., A. Bouma, A. R. Elbers, M. C. de Jong, G. Nodelijk, F. de Klerk, G. Koch
and M. van Boven (2004). "Avian influenza A virus (H7N7) epidemic in The Netherlands
in 2003: course of the epidemic and effectiveness of control measures." J Infect Dis
190(12): 2088-2095.
Suarez, D. L., M. Garcia, J. Latimer, D. Senne and M. Perdue (1999). "Phylogenetic
analysis of H7 avian influenza viruses isolated from the live bird markets of the
Northeast United States." J Virol 73(5): 3567-3573.
Suzuki, Y. (2006). "Natural selection on the influenza virus genome." Mol Biol Evol
23(10): 1902-1911.
Swayne, D. E. (2007). "Understanding the complex pathobiology of high pathogenicity
avian influenza viruses in birds." Avian Dis 51(1 Suppl): 242-249.
Taylor, L. H., S. M. Latham and M. E. Woolhouse (2001). "Risk factors for human
disease emergence." Philos Trans R Soc Lond B Biol Sci 356(1411): 983-989.
Tong, S., Y. Li, P. Rivailler, C. Conrardy, D. A. Castillo, L. M. Chen, S. Recuenco, J. A.
Ellison, C. T. Davis, I. A. York, A. S. Turmelle, D. Moran, S. Rogers, M. Shi, Y. Tao, M.
35

R. Weil, K. Tang, L. A. Rowe, S. Sammons, X. Xu, M. Frace, K. A. Lindblade, N. J. Cox,
L. J. Anderson, C. E. Rupprecht and R. O. Donis (2012). "A distinct lineage of influenza
A virus from bats." Proc Natl Acad Sci U S A 109(11): 4269-4274.
Trock, S. C., S. A. Burke and N. J. Cox (2012). "Development of an influenza virologic
risk assessment tool." Avian Dis 56(4 Suppl): 1058-1061.
Tumpey, T. M., C. F. Basler, P. V. Aguilar, H. Zeng, A. Solorzano, D. E. Swayne, N. J.
Cox, J. M. Katz, J. K. Taubenberger, P. Palese and A. Garcia-Sastre (2005).
"Characterization of the reconstructed 1918 Spanish influenza pandemic virus." Science
310(5745): 77-80.
Tweed, S. A., D. M. Skowronski, S. T. David, A. Larder, M. Petric, W. Lees, Y. Li, J.
Katz, M. Krajden, R. Tellier, C. Halpert, M. Hirst, C. Astell, D. Lawrence and A. Mak
(2004). "Human illness from avian influenza H7N3, British Columbia." Emerg Infect Dis
10(12): 2196-2199.
Varga, Z. T., I. Ramos, R. Hai, M. Schmolke, A. Garcia-Sastre, A. Fernandez-Sesma
and P. Palese (2011). "The influenza virus protein PB1-F2 inhibits the induction of type i
interferon at the level of the MAVS adaptor protein." PLoS Pathog 7(6): e1002067.
Webster, R. G., W. J. Bean, O. T. Gorman, T. M. Chambers and Y. Kawaoka (1992).
"Evolution and ecology of influenza A viruses." Microbiol Rev 56(1): 152-179.
WHO (2012). "Cumulative number of confirmed human cases of avian influenza
A(H5N1) reported to WHO, 2003-2012. (updated Aug. 10, 2012).
http://www.who.int/influenza/human_animal_interface/EN_GIP_20120810CumulativeNu
mberH5N1cases.pdf."
Wise, H. M., A. Foeglein, J. Sun, R. M. Dalton, S. Patel, W. Howard, E. C. Anderson, W.
S. Barclay and P. Digard (2009). "A complicated message: Identification of a novel PB1related protein translated from influenza A virus segment 2 mRNA." J Virol 83(16):
8021-8031.
Yuen, K. Y., P. K. Chan, M. Peiris, D. N. Tsang, T. L. Que, K. F. Shortridge, P. T.
Cheung, W. K. To, E. T. Ho, R. Sung and A. F. Cheng (1998). "Clinical features and
rapid viral diagnosis of human disease associated with avian influenza A H5N1 virus."
Lancet 351(9101): 467-471.
Zamarin, D., A. Garcia-Sastre, X. Xiao, R. Wang and P. Palese (2005). "Influenza virus
PB1-F2 protein induces cell death through mitochondrial ANT3 and VDAC1." PLoS
Pathog 1(1): e4.
Zamarin, D., M. B. Ortigoza and P. Palese (2006). "Influenza A virus PB1-F2 protein
contributes to viral pathogenesis in mice." J Virol 80(16): 7976-7983.

36

Zell, R., A. Krumbholz, A. Eitner, R. Krieg, K. J. Halbhuber and P. Wutzler (2007).
"Prevalence of PB1-F2 of influenza A viruses." J Gen Virol 88(Pt 2): 536-546.
Zhu, H., J. Wang, P. Wang, W. Song, Z. Zheng, R. Chen, K. Guo, T. Zhang, J. Peiris, H.
Chen and Y. Guan (2010). "Substitution of lysine at 627 position in PB2 protein does not
change virulence of the 2009 pandemic H1N1 virus in mice." Virology 401(1): 1-5.
Zhu, X., H. Yang, Z. Guo, W. Yu, P. J. Carney, Y. Li, L. M. Chen, J. C. Paulson, R. O.
Donis, S. Tong, J. Stevens and I. A. Wilson (2012). "Crystal structures of two subtype
N10 neuraminidase-like proteins from bat influenza A viruses reveal a diverged putative
active site." Proc Natl Acad Sci U S A 109(46): 18903-18908.

37

CHAPTER 2
PB1-F2 VARIANTS IN H5N1 AVIAN INFLUENZA VIRUSES
Ram P. Kamal, C. Todd Davis, Amrita Kumar, Wen-Pin Tzeng, Tung Nguyen, Ruben
Donis, Jacqueline Katz, and Ian A. York
Abstract:
Influenza A viruses (IAV) express the protein PB1-F2 from an alternate reading frame
from the PB1 segment. Its roles are not well understood, but appear to involve
modulation of host cell responses. A previous study by Zell et al. (J Gen Virol 88: 536546 (2007)) showed that IAV of mammals frequently have premature stop codons that
are expected to cause truncations of PB1-F2, whereas avian IAV usually express fulllength PB1-F2 of 90 amino acids. Thousands of new IAV sequences have been added
to databases since the previous study. With this expanded dataset, we also found that
human and swine IAV frequently contain truncated PB1-F2, and that most avian IAV
have full-length PB1-F2. However, in contrast to other avian IAV, since 2010 highlypathogenic H5N1 viruses have had a high proportion of PB1-F2 truncations, due to
several independent mutations that have persisted and expanded. A pair of closely
related H5N1 viruses, differing mainly in the presence of PB1-F2, showed very different
mouse lethality, with the virus containing defective PB1-F2 virus being 1000-fold more
lethal in mice. Transient expression of full-length, truncated and deleted PB1-F2 in
mammalian and avian cells did not show any effect on cell apoptosis or interferon
expression. Full length PB1-F2 co-localized with mitochondria. We observed the

38 
 

expression of an in-frame protein from downstream ATGs in PB1-F2 plasmids in which
the PB1-F2 translation initiation codon was mutated as in some H5N1 viruses.
Introduction:
Highly pathogenic avian influenza viruses (HPAIV) of H5N1 subtype are now endemic in
many countries (OIE 2012). H5N1 became a public health concern in 1997 when a
poultry outbreak in Hong Kong resulted in 18 human infections with six fatalities (Yuen,
Chan et al. 1998). Since then, over 600 human cases with a case fatality rate of about
60% have been confirmed as of Oct. 8, 2013 (WHO 2013). Unlike seasonal influenza A
viruses (IAV) that circulate among humans, HPAIVs are inefficient in human-to-human
transmission. One of the key reason for inefficient human transmission of H5N1 viruses
is the preference of their hemagglutinin (HA) for α 2, 3-linked sialic acids that are
uncommon in the upper respiratory tract of humans (Shinya, Ebina et al. 2006).
Besides HA, other IAV genes have also been shown to contribute to viral virulence.
However, the PB1 gene segment is of particular interest, since this was the only gene
(other than the glycoprotein genes, HA and neuraminidase (NA)) reassorted in the
pandemic events of 1957 and 1968 (Kawaoka, Krauss et al. 1989). Replacement of
polymerase segments of 1918 H1N1 pandemic virus with those of the modern (pre
2009) H1N1 strain attenuated the former (Tumpey, Basler et al. 2005). One possible
factor in PB1’s role in pathogenicity and transmission of IAVs is the presence of a
protein, PB1-F2, encoded by its +1 reading frame (Chen, Calvo et al. 2001), which was
present in the first 3 pandemics but not in the less-virulent 2009 H1N1 pandemic.

39 
 

PB1-F2 is a short (87–90 amino acids (aa)) IAV protein which was discovered
serendipitously in 2001 during the search for CD8 T cell epitopes encoded by
alternative reading frames (Chen, Calvo et al. 2001). It was found to localize in
mitochondria and cause cell death (Chen, Calvo et al. 2001, Gibbs, Malide et al. 2003).
It is dispensable for viral replication but has been associated with viral pathogenesis in
mice and ducks (Chen, Calvo et al. 2001, Zamarin, Ortigoza et al. 2006, McAuley,
Hornung et al. 2007, McAuley, Chipuk et al. 2010, Schmolke, Manicassamy et al. 2011),
and truncation of PB1-F2 in the 2009 pandemic influenza virus (H1N1pdm09) is
hypothesized to be one cause of its less severe pathology and lower case fatality rate
(Ramakrishnan, Gramer et al. 2009). Its effect on viral phenotype is virus strain specific
(Pena, Vincent et al. 2012, Pena, Vincent et al. 2012, Solbak, Sharma et al. 2013),
particularly seen in the viruses which have a PB1 gene of recent avian origin (McAuley,
Chipuk et al. 2010), perhaps because its effects are masked by other pathogenic
determinants (McAuley, Hornung et al. 2007).
The mechanisms underlying its pathogenic effects are still not clear. It was suggested to
be due to its mitochondrial targeting and apoptosis activity (Chen, Calvo et al. 2001,
Gibbs, Malide et al. 2003, Zamarin, Garcia-Sastre et al. 2005) but in one study, PR8
virus that lacked PB1-F2 had reduced replication and virus spreading independent of its
proapoptotic ability (Mazur, Anhlan et al. 2008). Full-length PB1-F2 inhibits early host
cell responses including IFN-β (Conenello, Tisoncik et al. 2011, Dudek, Wixler et al.
2011, Varga, Ramos et al. 2011, Meunier and von Messling 2012, Varga, Grant et al.
2012, McAuley, Tate et al. 2013, Reis and McCauley 2013), although one recent study
showed the opposite effect where PB1-F2 increased IFN-β expression in some, but not
40 
 

all, cell types (Le Goffic, Bouguyon et al. 2010). PB1-F2 has also been linked to
bacterial infections following IAV infection (Alymova, Green et al. 2011, Huber 2012,
Weeks-Gorospe, Hurtig et al. 2012, Smith, Adler et al. 2013). However, most of these
phenotypes have been difficult to generalize to all strains of virus and all hosts.
A striking feature of PB1-F2 in the global IAV population is the frequent presence of
premature stop codons which render its protein product of varying lengths. For example,
although full-length PB1-F2 is 90 amino acids long, almost all 2009 H1N1pdm viruses
have PB1-F2 protein of 11 residues. Other common truncation produces protein that is
57 residues in length. In 2007, analysis of publicly-available PB1-F2 sequences
indicated that a large proportion (19%) of mammalian IAVs had truncated PB1-F2, but
such truncations were rare among avian IAVs (only 4%) (Zell, Krumbholz et al. 2007).
This bias has led to the hypothesis suggesting a role of PB1-F2 truncations in
adaptation of IAVs to mammalian hosts (Zell, Krumbholz et al. 2007).
Although PB1-F2 truncations have sporadically been observed in avian influenza
viruses, until recently it has been unusual for groups of related avian influenza viruses
to persistently carry this mutation. However, since 2009, an increasing number of H5N1
highly-pathogenic avian influenza virus (HPAIV) lineages have maintained PB1-F2
truncations over several years. Based on previous suggestions that these viruses may
be particularly well adapted to mammalian replication, we tested these viruses for their
ability to cause disease in mice and to alter cellular functions in mammalian and avian
cells.

41 
 

Materials and Methods:
Database analysis of PB1-F2 truncations. We retrieved full-length PB1 sequences
with collection dates prior to 2013 from GISAID and the NCBI Influenza databases
(accessed May 22, 2013), and merged the sets based on virus strain name to avoid
redundancy. We also included sequences for 169 H5N1 viruses sequenced at CDC
(sequences available in GISAID). Sequences were selected by host species (human,
swine, avian) and subtype (H5N1), and year of collection was included in the FASTA
header of the downloads. In the case of H5N1, sequences prior to 1996 (the year of
identification of the progenitor of current highly-pathogenic avian influenza (HPAI) H5N1
strains) and low pathogenicity H5N1 viruses were excluded. H5N1 sequences were
excluded from the avian subset since those differed significantly from other avian
subtypes in context of PB1-F2 truncations. PB1 sequences were aligned using MAFFT
(Katoh, Misawa et al. 2002), and PB1-F2 sequences were translated from each PB1.
The length of each PB1-F2 sequence was used to calculate the number of truncated
versions present in each subset, per year of collection. In cases where no start ATG
was present at the appropriate position the length of PB1-F2 was considered to be 0. In
addition, we merged the set of H5N1 PB1 genes in which PB1-F2 was truncated with
PB1 genes from the WHO/OIE/FAO H5N1 Evolution Working Group (annotated with the
HA-based clade of the virus), aligned the genes using MUSCLE (Edgar 2004) and
constructed a maximum likelihood tree with MEGA5 (Tamura, Peterson et al. 2011).
Distribution of PB1-F2 truncations in H5N1 viruses. For H5N1 viruses with PB1-F2
truncations, we obtained the corresponding hemagglutinin (HA) sequence from the
appropriate database, annotating HA sequences with the length of the truncated PB142 
 

F2. We also included HA sequences from the report of WHO/OIE/FAO H5N1 Evolution
Working Group in order to identify clades. We aligned HA sequences using MUSCLE
and constructed a maximum likelihood tree with MEGA5.
Identification of paired PB1-F2 full-length/truncation H5N1 viruses. We analyzed
the set of virus isolates available to us to identify a pair of similar viruses differing in
PB1-F2 length. Viral RNAs were isolated by using QIAamp Viral RNA Mini Kit (#52904,
QIAGEN Inc. – USA), reverse transcribed and amplified by AccessQuick™ RT-PCR
System (#A1702, Promega Corporation, USA). Sequencing reactions were performed
by using BigDye®

Terminator

v3.1

Cycle

Sequencing

Kit

(#4337456,

Life

Technologies, Grand Island, NY) and run in Applied Biosystems 3730xl DNA Analyzer
(Life Technologies, Grand Island, NY). RT-PCR and sequencing primers were selected
from the primers routinely used at Molecular Virology and Vaccines Branch, CDC. For
each H5N1 virus containing a PB1-F2 truncation, the amino acid sequence of each viral
protein was compared to the most closely related virus containing full-length PB1-F2.
Preparation of viral stocks. Viruses A/chicken/Vietnam/NCVD-281/2009 (VN/281)
(GISAID accession number EPI350009) and A/chicken/Vietnam/NCVD-296/2009
(VN/296) (GISAID accession number EPI350121) were grown in 10-11 days old
o

embryonated chicken eggs incubated at 37 C for 24 hours. Allantoic fluids from eggs,
infected with the highest dilution of virus inoculum that gave a hemagglutination (HA)
o

titer of 256 or more, were harvested, clarified by centrifugation and frozen at -80 C as
viral stocks.

43 
 

-5

EID50. Six dilutions (10

-10

to 10

) were made from each virus. Five 9-11 day old
o

embryonated hen eggs were inoculated per dilution and incubated at 37 C for 24 hours.
o

Eggs were chilled overnight at 4 C to kill the embryos. Egg infections were detected by
standard hemagglutination assay and EID50 was calculated by Reed and Meunch
method (REED and MUENCH 1938).
Plaque assay. Confluent MDCK cells were infected with six dilutions (10

-5

-10

to 10

) of

each virus, overlaid with 0.8% agarose medium and incubated at two different
o

o

temperatures (37 C and 40 C). 72 hours post infection, agarose medium were gently
removed from wells and cell monolayers were stained with crystal violet-formalin mix.
Plaques were counted in appropriate wells and EID50 values were calculated using the
Reed and Muench method (REED and MUENCH 1938).
Pathogenicity in mice. All animal experiments were approved by IACUC. 6-8 weeks
old female BALB/cJ mice (Jackson Laboratory, Bar Harbor, ME) were used in this
o

4

study. 5 mice each were intranasally infected with one of 5 virus doses (10 to 10 PFU
in 100 µl volume) of each virus; 5 mice were mock infected with PBS. Mice were
observed for weight loss and other clinical signs daily for a period of 16 days. Mice
weighing less than 75% of their initial weight or showing nervous symptoms at any point
during study were euthanized. Mouse LD50 values were calculated using the Reed and
Muench formula (REED and MUENCH 1938).

44 
 

Cloning and mutagenesis of PB1-F2. Wild type PB1-F2 sequences from
A/chicken/Vietnam/NCVD-281/2009 (VN/281) and A/chicken/Vietnam/NCVD-296/2009
(VN/296) were RT-PCR amplified, using primers Kpn1PB1F76-94 and Xba1PB1R38467 (summarized in Table 2.1) and directionally cloned into pTracer-CM2-GFP vector
(Life Technologies, Grand Island, NY). Mutants of cloned PB1-F2 (summarized in table
2.2) were made by site-directed mutagenesis using Quick Change Mutagenesis Kit
(Agilent Technologies, Santa Clara, CA) and mutagenesis primers (summarized in table
2.1 and 2.2). We also constructed PB-F2 constructs with a C-terminal HA tag so that the
expression could be probed by using anti HA-tag antibodies. Cloned PB1-F2 plasmids
were amplified in competent E. coli cells and were purified using Qiagen plasmid
purification kit (QIAGEN Inc, Valencia, CA). The sequence of all plasmid clones was
confirmed before use in experiments.
Western blot: PB1-F2 plasmid transfected cells were lysed using RIPA buffer (#R0278,
Sigma-Aldrich, St. Louis, MO). Samples were prepared in Novex Tricin SDS Sample
Buffer (#LC1676, Life Technologies, Grand Island, NY) and reduced with NuPAGE
Sample Reducing Agent (#NP0009, Life Technologies, Grand Island, NY). Samples
were run on Novex 10-20% Tricin Gels (#EC6625BOX, Life Technologies, Grand
Island, NY) using Novex Tricin SDS Running Buffer (#LC1675, Life Technologies,
Grand Island, NY). The gels were blotted to PVDF membrane and probed with mouse
anti-HA tag antibody [HA.C5] (#ab18181, Abcam, Cambridge, MA). β-actin was probed
as a loading control using mouse anti β-actin antibody (C4) (#sc-47778, SCBT Inc,
Santa Cruz, CA). Goat anti-mouse IgG-HRP (#sc-2055, SCBT Inc, Santa Cruz, CA)
was used as secondary conjugate. Blots were developed using SuperSignal West Pico
45 
 

Chemiluminescent Substrate (#34080, Thermo Fisher Scientific Inc, Rockford, IL) and
viewed in Kodak Image Station 4000R Pro (Carestream Health Inc, New Haven, CT).
Confocal microscopy: Transfected cells were washed with PBS, fixed in 4%
paraformaldehyde for 15 min, permeabilised with 0.5% Tween-20 in PBS for 10 min,
washed with PBST (0.2% Tween-20), blocked with 2% BSA in PBST for 2 h, washed
with PBST and incubated with primary antibody for HA and mitochondria MTCO2
(Abcam) 1:1000 dilution in 0.5% BSA in PBST overnight. Cells were then washed with
PBST, incubated with secondary antibody (Alexa 594(mito)/ 350 (HA) solution in 0.5%
BSA in PBST for 2 h and washed with PBST. After staining was complete, cover slips
were mounted with mounting media (Invitrogen). Images were captured using Zeiss
invert confocal microscope LSM 710 withX63 objective. Images were processed using
Zen 2010 (Zeiss) and Adobe Photoshop.
Quantitative RT-PCR for IFNβ expression: Human A549 (#CCL-185, ATCC) and
chicken DF1 cells (#CRL-12203, ATCC) , grown in 24-well cell culture plates were
transfected with 1.0 µg of plasmid DNA using TransIT-LT1 transfection reagent (Mirus
Bio, Madison, WI). Cells were lysed and lysate were directly used for qRT-PCR using
CellsDirect One-Step qRTPCR Kit (Life Technologies, Grand Island, NY) and TaqMan
gene expression assays for human and avian IFNβ (Life Technologies, Grand Island,
NY). The ∆ ∆ Cq method in Bio-Rad CFX Manager 2.1 software was used to analyze
the expression data. GAPDH expression was measured in each sample as a control.
Flow cytometry for cell apoptosis: A549 and DF1 cells were transfected as described
in the last paragraph. After harvesting, cells were stained for apoptosis markers using

46 
 

Violet Chromatin Condensation/Dead Cell Apoptosis Kit with Vybrant® DyeCycle™
Violet and SYTOX® AADvanced™ for Flow Cytometry (Life Technologies, Grand
Island, NY) according to manufacturer’s instructions. Flow cytometry data were acquired
within one hour of staining. Data were acquired on FACSCanto II flow cytometer (BD,
Franklin Lakes, NJ), and analyzed using FlowJo software (Tree Star, Ashland, OR).
Statistical analyses. To compare weight loss, a linear mixed model with repeated
measures was used, using a cutoff of p<0.05 for statistical significance. Kaplan-Meier
survival curves were compared by the log-rank test. Effect on apoptosis and IFNβ
expression were analyzed by Kruskal-Wallis test in GraphPad Prism software.
Results:
Evolution of PB1-F2 truncations in avian and mammalian viruses. In a previous
study (Zell, Krumbholz et al. 2007), 2226 PB1 sequences having PB1-F2 ORF were
retrieved from NCBI database and analyzed for PB1-F2 truncations. The study found
that the proportion of truncated PB1-F2 was higher in human viruses (19%) than those
of avian viruses (4%) (Table 1 and S1 in reference (Zell, Krumbholz et al. 2007)). The
study included data available until June 15, 2006 and the number of available
sequences has increased significantly since then. We retrieved 19666 full-length PB1
sequences from the combined GISAID, NCBI and CDC Influenza databases: 9285
human, 2003 H5N1, 6705 avian (excluding H5N1), and 1572 swine sequences. We
identified 23 different truncations of PB1-F2, resulting in PB1-F2 length ranging from 087aa. A small number of viruses (53, 0.3% of total) had PB1-F2 that was longer than 90
amino acids at 101aa. 65.7% of viruses were full-length (87 or 90 aa). The prevalence
47 
 

of PB1-F2 truncations (less than 79 amino acids in length (Zell, Krumbholz et al. 2007))
was 58% for human viruses, 7.3% for H5N1, 3.1% for avian viruses excluding H5N1,
and 44.2% for swine viruses (table 2.3). Prevalence of PB1-F2 truncations by year of
collection is shown in Figure 2.1. The most common truncations were of 11 and 57
amino acids (23.2% and 7.6% of total, respectively), with no other truncation exceeding
1% of total. The most prevalent truncation in swine viruses is 11 aa (as seen in the
H1N1pdm virus) which has also replaced the previously most prevalent truncation
(57aa) in human viruses after the 2009 pandemic. Among avian viruses, 57 and 79residue versions were more prevalent before 2005, after which the 25aa variant became
more prevalent.
A 25 aa truncation was first detected in an avian virus (H2N1) collected from Delaware,
USA in 1988 (sequenced in 2006); but became persistent only after 2005. It has been
detected sporadically in swine and human viruses since 2001 and 2007, respectively.
Since its appearance its prevalence has been less than 3% in avian, human, and swine
viruses separately and combined.
It is important to note that these percentages are skewed due to sampling bias. For
example, human H1N1pdm viruses (almost all of which have truncated PB1-F2) were
much more intensively sequenced in 2009 than 2010; similarly, genome sequencing of
H5N1 differs greatly in different countries in which the virus is endemic, leading to
differential clade representation in the databases. Nevertheless, these data show that
the frequency of PB1-F2 truncations in H5N1 viruses has increased markedly in the
past five years, increasing from 10 of 346 samples (2.9%) in 2007, to 49 of 98 samples
(50%) in 2012, while other avian viruses consistently have full-length PB1-F2 and
48 
 

mammalian viruses have a high proportion of truncated PB1-F2. This provides further
support for the previous proposal (McAuley, Chipuk et al. 2010) that avian viruses may
receive a significant fitness benefit from full-length PB1-F2, while for viruses of humans
and swine either full-length PB1-F2 is dispensable, or a truncated version of PB1-F2 is
actively beneficial. In the latter case, the implication is that avian viruses containing
truncated PB1-F2 may be “pre-adapted” to mammalian infection.
Evolution of PB1-F2 truncations in H5N1 viruses. HPAIV H5N1 viruses have been
categorized into multiple clades based on their HA sequences. We constructed
phylogenetic trees, based on HA and full-length PB1 (Figure 2.2A and B), to analyze the
distribution of PB1-F2 truncations in H5N1 clades. Until 2009, PB1-F2 truncations were
rare (~2% of sequences, Figure 2.1) and sporadic, with few exceptions in which
truncations persisted in a particular lineage for several months (e.g. clade 1 in Thailand
from 2007-2008; see below). However, in 2009 and 2010, clade 2.3.4.1 viruses in
Vietnam and China consistently had PB1-F2 truncated at 25 or 24 amino acids. These
viruses included both poultry isolates and several human cases (A/Guizhou/1/2009, A
Hunan/2/2009).
In 2010 and 2011, the 2.3.2.1 clade became dominant in Vietnam, and the 2.3.4.1 clade
has not been isolated from Vietnam since 2010. However, in 2011/2012 a new PB1-F2
truncation of 25 amino acids arose in the clade 1.1a lineage in Vietnam. The earliest
instance of a truncated PB1-F2 in this clade is from 2007 (a pair of viruses from
Cambodia in the 1.1 lineage); however, since no further truncated PB1-F2 have been
identified from Cambodian isolates, and the 25-residue truncation in Vietnam appears to
be a separate lineage from the Cambodia isolates, these appear to have been sporadic
49 
 

mutations. In 2008, A/duck/Vietnam/NCVD-118/2008 had a PB1-F2 truncation; this may
represent the earliest version of the clade 1.1a mutation or may be a sporadic variant
that did not give rise to successful descendants. Since PB1 in these viruses is
phylogenetically distinct from that of the 2.3.4.1 lineage this mutation appears to have
arisen independently. To date, these truncations have only been isolated from poultry.
A third PB1-F2 truncation, of 57 amino acids, arose in the clade 2.3.2.1b lineage in
China and Vietnam in 2010 (first seen in A/duck/Vietnam/NCVD-885/2010) and this
variant has persisted through 2012 in both countries, with the majority of 2.3.2.1b
viruses containing truncated PB1-F2. Furthermore, in 2012 a reassortment event
occurred between clade 2.3.2.1b and 2.3.2.1c lineages in Vietnam, with the former
viruses donating 6 internal genes and the latter providing NA and the clade-determining
HA. Since the internal genes in this lineage includes the 57-aa truncation of PB1-F2 that
originally arose in 2.3.2.1a viruses, the vast majority of 2.3.2.1c viruses isolated in
Vietnam in 2012 also contain this PB1-F2 variant. An earlier example of a 57-aa
truncation of PB1-F2 arose in Thailand in 2007/2008 (Clade 1), although this variant has
not been identified since 2008.
Finally, a number of less common PB1-F2 truncations have been isolated in HPAIV
H5N1 since 2009, including 63, 34, 11, and 8 amino acids, and some variants that lack
a start codon altogether. However, since none of these variants have been isolated in a
subgroup of viruses for more than one year, we consider them to be sporadic mutations.
Paired viruses for PB1-F2 studies. We identified viruses A/chicken/Vietnam/NCVD281/2009

(VN/281)

(GISAID

accession

50 
 

number

EPI350009)

and

A/chicken/Vietnam/NCVD-296/2009 (VN/296) (GISAID accession number EPI350121)
as the most closely-matched pair of PB1-F2-disparate H5N1 viruses in our collection.
Amino acid differences between viral proteins of this pair are summarized in Table 2.4.
These viruses were isolated from chickens in 2009 in Vietnam and belong to clade
2.3.4.2. VN/281 contains a full length PB1-F2 of 90 amino acids, whereas virus VN/296
has a mutation in its translation initiation codon and hence is not expected to make full
length PB1-F2 protein. However, it potentially could express N-terminally truncated
versions of PB1-F2 from downstream in-frame ATGs at codon numbers 39, 46 and 51
(PB1-F2 ORF) as described previously (Zamarin, Ortigoza et al. 2006). Both the viruses
VN/281 and VN/296 replicated to similar titers in eggs and cells, as measured by EID50
and plaque assay (Table 2.5.).
Full length PB1-F2 (VN/281) is associated with reduced mice virulence. To study
the phenotype of selected pair of viruses, mice were infected with either virus VN/281 or
VN/296. These viruses are very similar elsewhere in the genome outside PB1-F2. The
virus pair differs only by 15 amino acids, none of which has been established as a
pathogenicity determinant REF. Mice infected with both viruses were severely affected,
showing weight loss and mortality (Figure 2.3 and Table 2.5). However, mice infected
with virus VN/296 (mutated start codon PB1-F2) were more severely affected than
those infected with VN/281 (full-length PB1-F2), showing marked weight loss and high
mortality at lower doses. The LD50 of virus VN/281 was approximately 1000-fold higher
than virus VN/296 (10

4.3

1.4

vs 10

PFU) making the later more virulent.

51 
 

Full length H5N1 PB1-F2 localizes to mitochondria. To find the mechanism
underlying 1000 fold pathogenicity difference of the studied viruses, we explored
phenotypes of cloned PB1-F2 plasmids in vitro in mammalian (A549) and chicken cells
(DF1). We constructed PB1-F2 variants differing in length as outlined in Figure 2.4. We
constructed 6 variants: 2 full-length versions (90 residues); 2 truncated at 24 residues;
and 2 in which the coding sequence was intact but the start codon (1st ATG) was
mutated to ACG. We confirmed the expression of the full length and truncated variants
in A549 cells by western blotting (Figure 2.5). No band was obvious for ATG-ACG
mutatnt. Both the full-length and the truncated versions of PB1-F2 were detected,
however the truncated version was at lower levels than for full-length.
After confirming the expression of PB1-F2 from plasmid constructs, we explored the
cellular localization of the protein in transfected cells using immunofluorescence.
Confocal microscopy revealed that the full-length PB1-F2 proteins were expressed and
were readily detected. The full length PB1-F2 from both the viruses localized with cell
mitochondria in both A549 (data not shown) and DF1 cells (Figure 2.6). The truncated
25-amino acid version of PB1-F2 was not detectable under standard conditions;
however, when the cells were treated with the proteasome inhibitor MG-132, some
expression was detectable (Figure 2.6). Interestingly, presence of an HA-tagged protein
was detectable by confocal microscopy in cells transfected with the ATG-mutated
version of PB1-F2; this was detected in the absence of MG-132, but was more obvious
in its presence. Since this presumably represents a protein in frame with the c-terminal
HA, this protein expression probably originates from an in-frame ATG (residue number
39, 46, or 51; figure 2.4b) downstream of the authentic start site, leading to expression
52 
 

of a truncated c-terminal fragment of the protein. The phenomenon has been described
previously by Zamarin et al in 2006 (Zamarin, Ortigoza et al. 2006). Neither the 25 aa
truncated version, nor the putative C-terminal fragment of PB1-F2, showed any
association with mitochondria (Figure 2.6)
PB1-F2 mutants had no effect on apoptosis. Mitochondrial localization may lead to
apoptosis and could be one explanation for the altered viral virulence. We transfected
A549 and DF1 with PB1-F2 variants, and harvested cells at 12, 24, 36 and 48 hours
post transfection, stained for apoptosis and cell death, and analyzed the cells by flow
cytometry. Statistical analysis of the apoptosis data was done using the Kruskal-Wallis
nonparametric test. No significant difference in percent of apoptotic or dead cells was
found between different PB1-F2 mutants and between the studied cell lines (Figure 2.7).
This is in accordance with previously published studied looking at the apoptosis
phenotype of H5N1 PB1-F2 (Chen, Chen et al. 2010, McAuley, Chipuk et al. 2010). One
of these studies evaluated this phenotype in all the 20th century pandemic viruses and
H5N1 viruses and found the phenotype only in PR8 and 1918 H1N1 viruses but not in
H5N1 and 1957 and 1968 pandemic viruses (McAuley, Chipuk et al. 2010).
PB1-F2 truncations and IFNb expression. A second possible mechanism for the
differences in virulence could be the interference with innate immune system. PB1-F2
from PR8 virus inhibits type I interferons (Dudek, Wixler et al. 2011, Varga, Ramos et al.
2011, Varga, Grant et al. 2012); however, that from the WSN/33 virus increases IFNβ
expression in epithelial cells (Le Goffic, Bouguyon et al. 2010). A549 cells were
transfected with PB1-F2 plasmids and IFNb expression was measured at different time
points using TaqMan assay qRT-PCR. We used empty plasmid vector as a negative
53 
 

control and tri-phosphate RNA as a positive control for IFNb induction. The expression
levels were normalized to GADPH and fold expression over empty vector transfection.
As expected, the positive control (TP-RNA) consistently induced IFNb expression (data
not shown). There was no significant difference in IFNb expression level with different
PB1-F2 plasmids in A549 cells (Figure 2.8) and DF1 cells (p>0.05, Kruskall-Wallis Test,
data not shown).
Discussion:
In 2007, Zell et al (Zell, Krumbholz et al. 2007) showed that the great majority of avian
influenza viruses contain a full-length PB1-F2 protein, whereas mammalian viruses
frequently have premature stop codons that truncate PB1-F2. Since the number of
influenza virus sequences has increased 7-fold since 2007, we repeated the analysis
and found a generally similar picture. In mammalian viruses (swine and human) PB1-F2
truncations are common (approximately 50% of sequences in the database). (A
possible exception to this rule is in the equine and canine influenza viruses, in which
PB1-F2 sequences are generally full length (not shown). However, a limited number of
PB1 sequences are available for these virus strains, so they were not included in the
present analysis. In contrast, in avian viruses over 95% of PB1-F2 sequences are full
length. The exception to this is in the highly pathogenic H5N1 virus group, in which the
frequency of PB1-F2 truncations in the databases has increased to 50% in 2012. It is
important to note that some of this increase may be due to surveillance bias, since the
published sequences of H5N1 are dominated by those from Vietnam and China where
PB1-F2 truncations are increasingly common in the endemic clades.

54 
 

We selected a pair of naturally-occurring viruses, VN/281 and VN/296, which differ only
by 0.36% (15 amino acids out of 4440 in the proteome) aside from PB1-F2, and tested
their replication in tissue culture and eggs, and virulence in mice. Although the viruses
replicated identically in eggs and in MDCK cells, their 50% mouse lethal dose (MLD50)
differed by 1000- fold. VN/281 (full length PB1-F2) was less pathogenic than VN/296
(deleted PB1-F2). This was unexpected since previous studies have linked PB1-F2 to
virulence, so the absence of PB1-F2 was expected to reduce the virulence of the virus.
On the other hand, the increased virulence of the PB1-F2-deleted virus is consistent
with the observation that in some contexts PB1-F2 increases IFN expression and
reduces viral replication (Le Goffic, Bouguyon et al. 2010, Le Goffic, Leymarie et al.
2011). The effect of PB1-F2 mutations on the H5N1 viruses has been rarely studied,
since most of those studies were done on A/Puerto Rico/08/1934(H1N1) (PR8) where
deletion of PB1-F2 decreases pathogenicity in mice. There are only two studies which
looked at mice pathogenicity in context of full length and deleted PB1-F2 of H5N1
viruses in mice (Schmolke, Manicassamy et al. 2011, Leymarie, Jouvion et al. 2013)
and neither of these found any significant effect. A slight reduction in lethality was
observed after PB1-F2 deletion in a previous H5N1 study (Leymarie, Jouvion et al.
2013). In another H5N1 study, PB1-F2 deletion resulted in reduced pathogenicity in
mice but not in ducks (Schmolke, Manicassamy et al. 2011). However, it has previously
been shown that the effects of PB1-F2 are highly dependent on genomic context and
the presence of other virulence factors (McAuley, Hornung et al. 2007, McAuley, Chipuk
et al. 2010, Pena, Vincent et al. 2012, Pena, Vincent et al. 2012, Solbak, Sharma et al.
2013), and the viruses in the present study have evolved for many years separately

55 
 

from the viruses (A/duck/Niger/2090/2006 and A/Vietnam/1203/2004) used in the former
studies.
In studies with PR8 viruses, the pathogenic effect of PB1-F2 is linked with its
mitochondrial localization, apoptosis in immune cells and regulation of innate immune
response in epithelial cells.

The only published study that has evaluated PB1-F2 of a

H5N1 virus (A/HK/156/1997) in context of mitochondrial targeting and apoptosis found
no effect on mitochondrial localization and apoptotic phenotype (Chen, Chen et al.
2010). However when PB1-F2 was mutated at Q69L and H75L, 50% of it localized to
mitochondria (Chen, Chen et al. 2010). The virus pair used in this study has 69Q and
75R but localized efficiently to mitochondria.
To help understand the reason for the difference in pathogenicity associated with fulllength vs. truncated PB1-F2, we assessed the effect of these variants on cellular
function in human and avian cells. Based on previous reports on PB1-F2 effects we
measured IFNβ response, and apoptosis induction. However, expression of PB1-F2
alone in human and chicken cells did not affect either apoptosis or IFN expression,
suggesting that any effect on virulence may be acting through as-yet unknown
functions.
It is intriguing that for the PB1-F2 plasmid lacking the initiating ATG, we observed
expression of an HA-tagged protein by confocal microscopy. Since this was detected
by staining for the HA tag at the C-terminus of the predicted PB1-F2 protein, the
simplest explanation is that this protein is produced by initiation at an in-frame ATG (at
position 39, 46, or 51) downstream of the authentic ATG, leading to an N-terminally

56 
 

truncated version (or versions) of PB1-F2 that also partially localized to mitochondria. A
similar observation has been described previously with PB1-F2 plasmid clones from
PR8 and WSN/33 viruses (Zamarin, Ortigoza et al. 2006). This is in accordance with
previous studies which have shown that the N-terminally truncated PB1-F2 and Cterminal peptides are sufficient and even more efficient in mitochondrial localization and
its apoptotic phenotypes (Gibbs, Malide et al. 2003, Yamada, Chounan et al. 2004);
however, those studies used a different virus, PR8. It is possible that the N-terminally
truncated PB1-F2 is constitutively expressed in the presence of full length protein, but is
expressed at a higher level when the standard ATG is not present. If so, this protein
may also play a role in influenza virus virulence in vivo.
Studies using deleted PB1-F2 viruses and plasmids have used different strategies.
These include mutation of multiple downstream ATGs along with the 1st start codon (Le
Goffic, Bouguyon et al. 2010), or introduction of a stop after the last in-frame ATG
(Zamarin, Ortigoza et al. 2006). These artificial strategies, which are not found in natural
isolates, would prevent expression of the putative C-terminal fragment of PB1-F2.
Other groups have both mutated the 1st start codon and inserted a stop codon at
positions 11 (McAuley, Hornung et al. 2007, Mazur, Anhlan et al. 2008, McAuley,
Chipuk et al. 2010, Meunier and von Messling 2012). The effect of these changes on
expression of the C-terminal fragment of PB1-F2 is unknown. Alternatively, the lack of
detection of a C-terminal fragment of PB1-F2 in these studies may be due to poor
binding of antibodies for the truncated protein, the use of antibodies recognizing the Nterminus of PB1-F2, or the general difficulty of detecting very small proteins.

57 
 

Our mouse observations are consistent with a recent study with A/USSR/90/77(H1N1),
a study in which PB1-F2 was eliminated by start codon mutation and insertion of a stop
codon at position 11 (Meunier and von Messling 2012), which may permit expression of
the C-terminal fragment of PB1-F2. In this study, infection of ferret lung explants with
the PB1-F2 deletion mutant led to increased pro-inflammatory cytokines, which in H5N1
infection may be associated with “cytokine storm” and increased virulence.
To summarize, historically rare PB1-F2 truncations are evolving among H5N1 HPAI
viruses. Since H5N1 virus pose a significant public health threat, assessment of the risk
of these truncations is very important for pandemic preparedness. To evaluate this, we
used a pair of similar viruses with full length and N-terminally truncated PB1-F2 which
differed by only 15 amino acids outside the PB1-F2 ORF. None among the 15 differing
amino acid is a known pathogenicity determinant. We found that a H5N1 virus with a
ATG-ACG mutation (VN/296) is 1000 fold more virulent than its paired virus (VN/281).
In vitro apoptosis and IFNb experiments did not show any significant difference among
different plasmids and between A549 and DF1 cells. Based on the previously published
and above mentioned studies, we propose that in HPAIV H5N1, the N-terminal of PB1F2 may reduce the pathogenic effects of the C-terminal fragment. Further immunopathogenesis studies will help clarify the mechanism of PB1-F2 pathogenicity.

58 
 

APPENDICES

59 
 

APPENDIX 2A: Tables

Table 2.1 Oligonucleotides used for cloning and mutagenesis of PB1-F2.
Primer name
Kpn1PB1F76-94
Xba1PB1R384-67

Primer Sequence
5’- ATTA GGTACC GGAGACCCTCCATACAGCC -3’
5’- GAGG TCTAGA GTCTGTCGACCTTGGGT -3’

PB1C167T_C170T_F

Used for
RT-PCR
amplification

5’TTCAGAAAAGGGGAAGTGGATAATAAACACAGAAACTGGAGCAC
-3’
PB1C96T_F
5’- AGACCCTCCATACAGCCATGGAACAGGGAC -3’
PB1T96C _F
5'- AGACCCTCCATACAGCCACGGAACAGGGAC-3'
Kpn1PB1F68-90
5’-TAG ACG ACT AAT CTA TAG TAA CGC ATC GAG GA G GTA
CC C TTA TAC TGG AGA CCC TCC ATA C -3’
Xba1HAtagPB1R364- 5’- GAGG TCTAGA TTA AGC GTA ATC TGG AAC ATC GTA TGG
GTA GTT TGT CCA CTC TTG TTT GTT G -3’
43

RT-PCR
amplification
with C-terminal
HA-tag

PB1F151-185
(HA_C167T_C170T)

Mutagenesis
with HA tag

5’- GAA AAG GGG AAG TGG A TAC CCA TAC GAT GTT CCA
GAT TAC GCT TAA TAA ACA CAG AAA CTG G-3’

Mutagenesis

Numberings in oligo names represent PB1 nt number. Restrictions sites are indicated as bold and underlined. Mutated
residues and stop codons are in red text. HA tag is purple and italicized.

60 
 

Table 2.2 Details of PB1-F2 mutants made in this study.
Sr.
No.

Source

PB1-F2

Mutagenesis Primer
Kpn1PB1F76-94 and Xba1PB1R384-67

1

VN/281/PB1-F2FU

Full

2

VN/281/PB1-F2TR

Truncated PB1C167T_C170T_F

3

VN/281/PB1-F2DL

Deleted*

PB1T96C _F

4

VN/296/ PB1-F2FU

Full

PB1C96T_F

5

VN/296/ PB1-F2TR

Truncated PB1C167T_C170T_F

6

VN/296/ PB1-F2DL

Deleted*

Kpn1PB1F76-94 and Xba1PB1R384-67

7

VN/281/PB1-F2FU-HA

Full

Kpn1PB1F68-90 and Xba1HAtagPB1R364-43

8

VN/281/PB1-F2TR-HA

Truncated PB1F151-185 (HA_C167T_C170T)

9

VN/281/PB1-F2DL-HA

Deleted*

PB1T96C _F

10

VN/296/ PB1-F2FU-HA

Full

PB1C96T_F

11

VN/296/ PB1-F2TR-HA

Truncated PB1F151-185 (HA_C167T_C170T)

12

VN/296/ PB1-F2DL-HA

Deleted*

Kpn1PB1F68-90 and Xba1HAtagPB1R364-43

*The first ATG is mutated to ACG, leaving the downstream in-frame ATGs intact.

61 
 

Table 2.3 Prevalence of PB1-F2 truncations among all subtypes (human, swine
and avian IAVs) versus H5N1.

Total
sequences
No. (%) >
79aa
No.
(%)
≤78aa

All

Human

Swine
1693

Avian (except
H5N1)
6703

19666

9268

13063
(66.4%)
6603
(33.6%)

3843
(41.5%)
5425
(58.5%)

H5N1
2003

880
(52.0%)
813
(48.0%)

6490
(96.8%)
213
(3.2%)

1850
(92.4%)
152
(7.6%)

Table 2.4 Amino acid differences between viruses VN/281 and VN/296.
Protein
Size (aa)
HA
567
M1
252
M2
97
NA
449
NP
498
NS1
2251
NS2
121
PA
716
PB1
756
PB1-F2
902
PB2
759
Total
4530
Total (excluding
4440
PB1-F2)
1
2

aa diffs
1
0
2
4
1
1
0
1
3
3
2
18
15

Both the viruses have a C-terminal truncation in NS1 that renders it 215aa long.
PB1-F2 is 90aa in virus VN/281 and deleted in virus VN/296.

Table 2.5 Growth characteristics and mouse virulence.
Virus

PB1-F2

EID50/ml

PFU/ml

VN/281
VN/296

Full length
Deleted

3.2 X10^9
6.8X10^8

7.0 X10^8
2.3 X10^8

62 
 

MLD50
(Log10
PFU)
4.3
1.4

APPENDIX 2B: Figures

100.0
80.0
60.0
40.0
20.0
0.0

Avian (No H5N1)
H5N1
Human
Swine

Figure 2.1 PB1-F2 truncation prevalence. PB1 sequences were retrieved from NCBI, GISAID and CDC databases. PB1
ORFs were aligned using MAFFT and looked for stop codons, hence length of the PB1-F2 protein. PB1-F2 proteins of
length <79aa were marked truncated and represented as percent of total viruses (PB1-F2 sequences) by host and year of
isolation.
63 
 

A

B

Clade 2.3.2.1

Clade 2.3.4.1

Figure 2.2 Evolutionary relationship of PB1-F2 truncations in H5N1 viruses.
Phylogenetic trees of HA gene (A) and PB1 gene (B) are plotted to show truncations in
PB1-F2 viruses. Truncations of 24/25, and 57 aa are colored red and green,
respectively. The blue font indicates virus strains having other truncations in PB1-F2
(63, 11, 8, and 0 amino acids). Clades were defined by taking representative sequences
from the WHO/OIE/FAO H5N1 Classification Working Group. Viruses used in this study
are indicated with green (VN/296, truncated PB1-F2) and red (VN/281, full-length PB1F2) arrows.
64 
 

A

100

G
Weight (% of day 0)

60
40

Mock
VN/281
VN/296

20
0
100

p=0.015

60
40

Mock
VN/281
VN/296

20
0
0

2

4

6

8

10

12

14

16

110
100
90
Mock

80

VN/281

70

VN/296

p=0.2284

60

F

80

Percent survival

1

10 PFU

B

p=0.317

Weight (% of day 0)

0

10 PFU

Percent survival

80

110
100
90
Mock

80

p=<0.0001

VN/281

70

VN/296

60
0

2

4

6

8

10

12

14

16

Days post-infection

Days post-infection

Figure 2.3 Mice virulence of paired viruses with and without PB1-F2. Five mice each were infected intranasally either
with PBS, VN/281 (45) or VN/296 (57) and observed daily for 16 days. Survival data (A-E) and body weight loss data (F-J)
of mice are shown here as percent of day “zero” values and are group averages of 5 mice. Statistical significances for
survival and body weight loss between two viruses were calculated using a linear mixed model with repeated measures
and Log Rank Test – Chi Square p-value, respectively.

65 
 

Figure 2.3 (cont’d)
H

100

Weight (% of day 0)

C

60

p=0.0625

40

Mock
VN/281
VN/296

20
0
100

D

I
Weight (% of day 0)

2

10 PFU

Percent survival

80

p=0.0585
60
40

Mock
VN/281
VN/296

20
0
100

E

J
p=0.00231

Percent survival

4

10 PFU

80
60
40

Mock
VN/281
VN/296

20
0

0

2

4

6

8

10

12

14

16

90
Mock

80

VN/281

70

p=0.0544

VN/296

60
110
100
90
Mock

80

VN/281

70

p=0.0005

VN/296

60

100
90
Mock

80

VN/281

70
60

p=0.0302

VN/296
0

2

4

6

8

10

12

Days post-infection

Days post-infection
66 
 

100

110

Weight (% of day 0)

3

10 PFU

Percent survival

80

110

14

16

PB1-F2

PB1

 

 1                                 25  26         39            46          51                                 90  Stop
F2FU

ATG

F2FU NO

ACG

CAA CAA

F2FU TR

ATG

TAA TAA

F2FU HA

ATG

F2FU NO HA

ACG

F2FUHA TR

ATG

CAA CAA

ATG

ATG
ATG

ATG

ATG

ATG

ATG

ATG

CAA CAA

ATG

ATG

ATG

CAA CAA

ATG

ATG

ATG

HA TAA TAA

ATG

TGA

 

ATG

ATG

TGA

 

TGA

 

HA TGA

 

HA TGA

 

ATG

TGA

 

Figure 2.4 Construction of PB1-F2 mutants’ recombinant plasmid clones. One full length (90aa), one truncated
(24aa) and one deleted (0 aa) variants of PB1-F2 were made from each of VN/281 and VN/296 viruses. Similar mutants
were also made with C-terminal HA tag. The mutated nucleotides and stop codons are shown in red.

67 
 

CMV2 vector
VN/281 FULL
VN/281  NO ATG
VN/281 TR
VN/296 FULL
VN/296  TR
MARKER

 

B‐actin loading 
control
Full 
(90aa)
Truncated 
(24aa)

15 kD
3.5 kD

Figure 2.5 Expression of PB1-F2 mutants in A549 cells. Cells were transfected with
either PB1-F2 plasmids having c-terminal HA tag or empty vector and examined for
protein expression, 24 hour post transfection. Cell lysates were run on tris tricin gradient
gel, blotted on PVDF membrane and probed with anti-HA antibody (abcam).

68 
 

Mitochondria

GFP

PB1F2/ HA tag

PB1F2+Mito
(merge)

a-d

e-h

i-l

Figure 2.6 PB1-F2 expression and mitochondrial localization. DF1 (chicken) cells
were grown on coverslips, transfected with either control plasmids or PB1-F2 plasmids
from virus VN/296. 30 hr post transfection, MG132 (proteasomal inhibitor) were added
to medium for 6 hr. 36 hr post transfection, cells were fixed with 4% paraformaldehyde
and permeablized. PB1-F2 was probed by staining HA-tag (green), mitochondria were
stained using MitoTracker (red) and examined under confocal microscope. a-d,
untransfected; e-h, CMV2/GFP (empty cloning vector); i-l, ERAP1-HA clone as positive
control for HA detection; m-p, q-t and u-x were transfected with full length, No-ATG and
Truncated PB1-F2 mutant clones of virus VN/296, respectively.

69 
 

Figure 2.6 (cont’d)

Mitochondria

GFP

PB1F2/ HA tag

m-p

q-t

u-x

70 
 

PB1F2+Mito
(merge)

b

c

CMV2 vector

VN/296 TR

VN/296 Del

VN/296 Full

VN/281 TR

VN/281 Del

d

VN/281 Full

Percent (Apoptotic + dead) cells as fold of mock (empty vector)

a

Figure 2.7 PB1-F2 effect on apoptosis. A549 cells were transfected with either empty
vector or PB1-F2 mutant plasmids. Cells were harvested at 12 hr (a), 24hr (b), 36hr (c)
and 48 hr (d) post transfection. Cells were stained apoptotic and necrotic dies. Data
represented dead + apoptotic cells as fold change over CMV2 vector (control).
Note: scale different for different time points.
71 
 

a

b

c

CMV2 vector

VN/296 TR

VN/296 Del

VN/296 Full

VN/281 TR

VN/281 Del

VN/281 Full

d

Figure 2.8 PB1-F2 effect on IFNb expression. A549 cells were transfected with either
empty vector or PB1-F2 mutant plasmids. Cells were lysed at 12 hr (a), 24hr (b), 36hr
(c) and 48 hr (d) post transfection. Expression was measured using TaqMan assay and
qRT-PCR. IFNb expression is normalized to GAPDH (internal control) and presented as
fold increase relative to CMV2-vector transfection.
72 
 

REFERENCES

73 
 

REFERENCES

Alymova, I. V., A. M. Green, N. van de Velde, J. L. McAuley, K. L. Boyd, H. E. Ghoneim
and J. A. McCullers (2011). "Immunopathogenic and antibacterial effects of H3N2
influenza A virus PB1-F2 map to amino acid residues 62, 75, 79, and 82." J Virol 85(23):
12324-12333.
Chen, C. J., G. W. Chen, C. H. Wang, C. H. Huang, Y. C. Wang and S. R. Shih (2010).
"Differential localization and function of PB1-F2 derived from different strains of
influenza A virus." J Virol 84(19): 10051-10062.
Chen, W., P. A. Calvo, D. Malide, J. Gibbs, U. Schubert, I. Bacik, S. Basta, R. O'Neill, J.
Schickli, P. Palese, P. Henklein, J. R. Bennink and J. W. Yewdell (2001). "A novel
influenza A virus mitochondrial protein that induces cell death." Nat Med 7(12): 13061312.
Conenello, G. M., J. R. Tisoncik, E. Rosenzweig, Z. T. Varga, P. Palese and M. G.
Katze (2011). "A single N66S mutation in the PB1-F2 protein of influenza A virus
increases virulence by inhibiting the early interferon response in vivo." J Virol 85(2):
652-662.
Dudek, S. E., L. Wixler, C. Nordhoff, A. Nordmann, D. Anhlan, V. Wixler and S. Ludwig
(2011). "The influenza virus PB1-F2 protein has interferon antagonistic activity." Biol
Chem 392(12): 1135-1144.
Edgar, R. C. (2004). "MUSCLE: multiple sequence alignment with high accuracy and
high throughput." Nucleic Acids Res 32(5): 1792-1797.
Gibbs, J. S., D. Malide, F. Hornung, J. R. Bennink and J. W. Yewdell (2003). "The
influenza A virus PB1-F2 protein targets the inner mitochondrial membrane via a
predicted basic amphipathic helix that disrupts mitochondrial function." J Virol 77(13):
7214-7224.
Huber, V. C. (2012). "Can surveillance of the influenza virus PB1-F2 gene be used to
predict the severity of secondary bacterial infections?" Virulence 3(6): 523-524.
Katoh, K., K. Misawa, K. Kuma and T. Miyata (2002). "MAFFT: a novel method for rapid
multiple sequence alignment based on fast Fourier transform." Nucleic Acids Res
30(14): 3059-3066.
Kawaoka, Y., S. Krauss and R. G. Webster (1989). "Avian-to-human transmission of the
PB1 gene of influenza A viruses in the 1957 and 1968 pandemics." J Virol 63(11): 46034608.
74 
 

Le Goffic, R., E. Bouguyon, C. Chevalier, J. Vidic, B. Da Costa, O. Leymarie, C.
Bourdieu, L. Decamps, S. Dhorne-Pollet and B. Delmas (2010). "Influenza A virus
protein PB1-F2 exacerbates IFN-beta expression of human respiratory epithelial cells."
J Immunol 185(8): 4812-4823.
Le Goffic, R., O. Leymarie, C. Chevalier, E. Rebours, B. Da Costa, J. Vidic, D.
Descamps, J. M. Sallenave, M. Rauch, M. Samson and B. Delmas (2011).
"Transcriptomic analysis of host immune and cell death responses associated with the
influenza A virus PB1-F2 protein." PLoS Pathog 7(8): e1002202.
Leymarie, O., G. Jouvion, P. L. Herve, C. Chevalier, V. Lorin, J. Lecardonnel, B. Da
Costa, B. Delmas, N. Escriou and R. Le Goffic (2013). "Kinetic Characterization of PB1F2-Mediated Immunopathology during Highly Pathogenic Avian H5N1 Influenza Virus
Infection." PLoS One 8(3): e57894.
Mazur, I., D. Anhlan, D. Mitzner, L. Wixler, U. Schubert and S. Ludwig (2008). "The
proapoptotic influenza A virus protein PB1-F2 regulates viral polymerase activity by
interaction with the PB1 protein." Cell Microbiol 10(5): 1140-1152.
McAuley, J. L., J. E. Chipuk, K. L. Boyd, N. Van De Velde, D. R. Green and J. A.
McCullers (2010). "PB1-F2 proteins from H5N1 and 20 century pandemic influenza
viruses cause immunopathology." PLoS Pathog 6(7): e1001014.
McAuley, J. L., F. Hornung, K. L. Boyd, A. M. Smith, R. McKeon, J. Bennink, J. W.
Yewdell and J. A. McCullers (2007). "Expression of the 1918 influenza A virus PB1-F2
enhances the pathogenesis of viral and secondary bacterial pneumonia." Cell Host
Microbe 2(4): 240-249.
McAuley, J. L., M. D. Tate, C. J. MacKenzie-Kludas, A. Pinar, W. Zeng, A. Stutz, E.
Latz, L. E. Brown and A. Mansell (2013). "Activation of the NLRP3 inflammasome by
IAV virulence protein PB1-F2 contributes to severe pathophysiology and disease." PLoS
Pathog 9(5): e1003392.
Meunier, I. and V. von Messling (2012). "PB1-F2 modulates early host responses but
does not affect the pathogenesis of H1N1 seasonal influenza virus." J Virol 86(8): 42714278.
OIE (2012). "Update on Highly Pathogenic Avian Influenza in Animals (type H5 and
H7).".
Pena, L., A. L. Vincent, C. L. Loving, J. N. Henningson, K. M. Lager, W. Li and D. R.
Perez (2012). "Strain-dependent effects of PB1-F2 of triple-reassortant H3N2 influenza
viruses in swine." J Gen Virol 93(Pt 10): 2204-2214.

75 
 

Pena, L., A. L. Vincent, C. L. Loving, J. N. Henningson, K. M. Lager, A. Lorusso and D.
R. Perez (2012). "Restored PB1-F2 in the 2009 pandemic H1N1 influenza virus has
minimal effects in swine." J Virol 86(10): 5523-5532.
Ramakrishnan, M. A., M. R. Gramer, S. M. Goyal and S. Sreevatsan (2009). "A
Serine12Stop mutation in PB1-F2 of the 2009 pandemic (H1N1) influenza A: a possible
reason for its enhanced transmission and pathogenicity to humans." J Vet Sci 10(4):
349-351.
REED, L. J. and H. MUENCH (1938). "A SIMPLE METHOD OF ESTIMATING FIFTY
PER CENT ENDPOINTS." Am. J. Epidemiol 27(3): 493-497.
Reis, A. L. and J. W. McCauley (2013). "The Influenza virus protein PB1-F2 interacts
with IKKbeta and modulates NF-kappaB signalling." PLoS One 8(5): e63852.
Schmolke, M., B. Manicassamy, L. Pena, T. Sutton, R. Hai, Z. T. Varga, B. G. Hale, J.
Steel, D. R. Perez and A. Garcia-Sastre (2011). "Differential contribution of PB1-F2 to
the virulence of highly pathogenic H5N1 influenza A virus in mammalian and avian
species." PLoS Pathog 7(8): e1002186.
Shinya, K., M. Ebina, S. Yamada, M. Ono, N. Kasai and Y. Kawaoka (2006). "Avian flu:
influenza virus receptors in the human airway." Nature 440(7083): 435-436.
Smith, A. M., F. R. Adler, R. M. Ribeiro, R. N. Gutenkunst, J. L. McAuley, J. A.
McCullers and A. S. Perelson (2013). "Kinetics of coinfection with influenza A virus and
Streptococcus pneumoniae." PLoS Pathog 9(3): e1003238.
Solbak, S. M., A. Sharma, K. Bruns, R. Roder, D. Mitzner, F. Hahn, R. Niebert, A.
Vedeler, P. Henklein, P. Henklein, U. Schubert, V. Wray and T. Fossen (2013).
"Influenza A virus protein PB1-F2 from different strains shows distinct structural
signatures." Biochim Biophys Acta 1834(2): 568-582.
Tamura, K., D. Peterson, N. Peterson, G. Stecher, M. Nei and S. Kumar (2011).
"MEGA5: molecular evolutionary genetics analysis using maximum likelihood,
evolutionary distance, and maximum parsimony methods." Mol Biol Evol 28(10): 27312739.
Tumpey, T. M., C. F. Basler, P. V. Aguilar, H. Zeng, A. Solorzano, D. E. Swayne, N. J.
Cox, J. M. Katz, J. K. Taubenberger, P. Palese and A. Garcia-Sastre (2005).
"Characterization of the reconstructed 1918 Spanish influenza pandemic virus." Science
310(5745): 77-80.
Varga, Z. T., A. Grant, B. Manicassamy and P. Palese (2012). "Influenza Virus Protein
PB1-F2 Inhibits the Induction of Type I Interferon by Binding to MAVS and Decreasing
Mitochondrial Membrane Potential." J Virol 86(16): 8359-8366.

76 
 

Varga, Z. T., I. Ramos, R. Hai, M. Schmolke, A. Garcia-Sastre, A. Fernandez-Sesma
and P. Palese (2011). "The influenza virus protein PB1-F2 inhibits the induction of type I
interferon at the level of the MAVS adaptor protein." PLoS Pathog 7(6): e1002067.
Weeks-Gorospe, J. N., H. R. Hurtig, A. R. Iverson, M. J. Schuneman, R. J. Webby, J. A.
McCullers and V. C. Huber (2012). "Naturally Occurring Swine Influenza A Virus PB1F2 Phenotypes That Contribute to Superinfection with Gram-Positive Respiratory
Pathogens." J Virol 86(17): 9035-9043.
WHO (2013). "Cumulative number of confirmed human cases of avian influenza
A(H5N1) reported to WHO, 2003- 2013. Update Oct. 8, 2013 (accessed Oct. 23,
2013).".
Yamada, H., R. Chounan, Y. Higashi, N. Kurihara and H. Kido (2004). "Mitochondrial
targeting sequence of the influenza A virus PB1-F2 protein and its function in
mitochondria." FEBS Lett 578(3): 331-336.
Yuen, K. Y., P. K. Chan, M. Peiris, D. N. Tsang, T. L. Que, K. F. Shortridge, P. T.
Cheung, W. K. To, E. T. Ho, R. Sung and A. F. Cheng (1998). "Clinical features and
rapid viral diagnosis of human disease associated with avian influenza A H5N1 virus."
Lancet 351(9101): 467-471.
Zamarin, D., A. Garcia-Sastre, X. Xiao, R. Wang and P. Palese (2005). "Influenza virus
PB1-F2 protein induces cell death through mitochondrial ANT3 and VDAC1." PLoS
Pathog 1(1): e4.
Zamarin, D., M. B. Ortigoza and P. Palese (2006). "Influenza A virus PB1-F2 protein
contributes to viral pathogenesis in mice." J Virol 80(16): 7976-7983.
Zell, R., A. Krumbholz, A. Eitner, R. Krieg, K. J. Halbhuber and P. Wutzler (2007).
"Prevalence of PB1-F2 of influenza A viruses." J Gen Virol 88(Pt 2): 536-546.

77 
 

JOINING CHAPTER
PATHOGENICITY DETERMINACY TO PANDEMIC VACCINES
Avian influenza viruses are undergoing constant evolution in domestic poultry and
occasionally infect other species including humans. All the 20th century and 21st century
human influenza pandemic viruses have derived multiple gene(s) from avian viruses
either directly or after further evolution in swine. While multiple avian viruses have
sporadically infected humans, including H6N1 (Yuan, Zhang et al. 2013), H9N2 (Pawar,
Tandale et al. 2012), and H10N7 (Arzey, Kirkland et al. 2012), viruses belonging to
subtype H5 and H7 have been predicted to be major public health risks. Both these HA
types include highly pathogenic viruses based on their phenotype in avian hosts.
Although the highly pathogenic phenotype in domestic poultry doesn’t necessarily
translate to human infections, both of these subtypes also include viruses that have
caused fatal infections in humans, and

due to constant evolution and possible

reassortment events, these subtypes pose the greatest human health threat. To
counteract the risk, US Dept. of Human Health and Services, along with its national and
international public health partners, constantly performs surveillance for new influenza
infections and studies the evolution and emergence of new influenza virus isolates.
Virus surveillance is followed by two major strategies of pandemic preparedness: Risk
assessment (especially determining pathogenicity and transmission), and risk mitigation
(especially preparing vaccines against potentially pandemic viruses).
In the first chapter of this dissertation, I focused on the risk assessment of HPAIV
containing PB1-F2 mutations. I studied the evolution of H5N1 influenza viruses and

78

found a new trend in the emergence and evolution of PB1-F2 proteins. Historically rare
in avian viruses, the truncated PB1-F2 has independently evolved in multiple different
clades. In a mouse model, the virus VN/296 with a defective initiating ATG of PB1-F2 is
1000-fold more virulent in mice than a paired virus VN/281 expressing full length PB1F2. I further explored the mechanism underlying the difference in virulence in vitro in
human (A549) and chicken (DF1) cells. I studied the previously described phenotypes
of PB1-F2 (apoptosis, IFNβ regulation and mitochondrial localization) with cloned
plasmids in transient expression system. Most of these phenotypes have been found in
studies with PR8 viruses while studies with H5N1 viruses are still lacking. As shown in
previous studies with H5N1 viruses, I did not find significant difference in apoptosis of
either of the studied cell lines. I also did not find any difference in the regulation of IFNβ
expression. I demonstrated that full length PB1-F2 localizes to mitochondria, the first
report showing the mitochondrial localization of H5N1 PB1-F2. I also demonstrated that
the absence of the initiating ATG still permits expression of a protein from the PB1-F2
region, which I suggest is due to translation from a down-stream ATG to produce a Cterminal fragment of PB1-F2 that does not appear to localize to mitochondria. Based on
my in vivo and in vitro data and previously published H5N1 studies, I propose a working
model for the virulence difference I observed in mice. I propose that the full length PB1F2 in VN/281 is playing a protective or neutralizing role against an exacerbated
inflammatory response elicited by the

C-terminal of PB1-F2 (present in VN/296).

Further studies in particular with isogenic viruses made by reverse genetics will
enhance the understanding and define the mechanism. Further research on the

79

mechanism was out of the reach of this PhD dissertation research due to the interim
moratorium on research involving H5N1 gain of function mutations.
My second project (discussed in the following chapter) focused on the second major
strategy of pandemic preparedness: Vaccines against potentially pandemic strains of
influenza. A pre-pandemic vaccine is a vaccine which is made based on risk
assessment of current and emerging influenza strains that could be used for protection
against future pandemic events. Pre-pandemic vaccines against avian influenza viruses
have been hindered by the poor immune response against these in humans. This is
particularly true for vaccines against H7 viruses.
Protection against influenza viruses is mainly produced by anti-HA antibodies.
Seasonal flu vaccines elicit protective antibody titers, for which the accepted serological
correlate is an HI titer of 40 or more. However, human serosurveillance studies and
results of H7 vaccine clinical trials have found very poor serum antibody titers as
measured by HI. These findings suggested that the HA of H7 viruses may be poorly
immunogenic. If so, pre-pandemic vaccines may not be effective when needed in the
future.
In spite of widespread anecdotal concern about H7 immunogenicity, no previously
published study had compared the immunogenicity of H7 and seasonal viruses. I
therefore performed this study as a second project in pandemic preparedness. My
specific aim was to evaluate and compare the immunogenicity of H7 HAs with that of
seasonal viruses. The project is described in the following chapter of this dissertation.

80

REFERENCES

81

REFERENCES

Arzey, G. G., P. D. Kirkland, K. E. Arzey, M. Frost, P. Maywood, S. Conaty, A. C. Hurt,
Y. M. Deng, P. Iannello, I. Barr, D. E. Dwyer, M. Ratnamohan, K. McPhie and P. Selleck
(2012). "Influenza virus A (H10N7) in chickens and poultry abattoir workers, Australia."
Emerg Infect Dis 18(5): 814-816.
Pawar, S. D., B. V. Tandale, C. G. Raut, S. S. Parkhi, T. D. Barde, Y. K. Gurav, S. S.
Kode and A. C. Mishra (2012). "Avian influenza H9N2 seroprevalence among poultry
workers in Pune, India, 2010." PLoS One 7(5): e36374.
Yuan, J., L. Zhang, X. Kan, L. Jiang, J. Yang, Z. Guo and Q. Ren (2013). "Origin and
Molecular Characteristics of a Novel 2013 Avian Influenza A(H6N1) Virus Causing
Human Infection in Taiwan." Clin Infect Dis.

82

CHAPTER 3
HEMAGGLUTININ OF THE H7 SUBTYPE IS INTRINSICALLY POORLY
IMMUNOGENIC
K. Blanchfield*, R. Kamal*, W.-P. Tzeng, N. Music, J. Wilson, A. Lipatov, R. O. Donis,
J.M. Katz, and I.A. York
1

Influenza Division, National Center for Immunization and Respiratory Diseases,

Centers for Disease Control and Prevention, Atlanta, Georgia, USA

* These authors contributed equally to this work

83

Abstract:
Of the influenza viruses within the seventeen known influenza hemagglutinin (HA)
subtypes, three have been responsible for pandemics in humans. Most subtypes
continuously circulate in wild birds and domestic poultry, and some of these avian
influenza viruses (AIVs) sporadically infect humans, with potential to cause a new
pandemic. Vaccines are the main defense against these potentially pandemic AIVs.
However, vaccines against AIVs often require high HA doses, or adjuvants, to achieve
acceptable serological titers. Viruses of the H7 subtype, in particular, frequently have
not induced strong antibody responses following infection or immunization. Here we
compared the immunogenicity, in naïve mice, of purified recombinant HA from two H7
viruses (A/Netherlands/219/2003(H7N7) and A/New York/107/2003(H7N2)) to that of
HA

from

seasonal

viruses

(A/California/07/2009(H1N1pdm09)

and

A/Perth16/2009(H3N2)). Mice produced antibodies to all HAs, but the response to
seasonal viruses was greater than to H7s. The difference was relatively minor when
measured by ELISA, yet greater when measured by hemagglutination inhibition assays,
and more marked by microneutralization assays, by which H7s induced a very weak or
no response. Antibodies induced by H7 were of significantly lower avidity than for H3 or
H1pdm09. We conclude that H7 HAs may be intrinsically less immunogenic than HA
from seasonal human influenza viruses.
Introduction:
Of the seventeen known subtypes of influenza virus hemagglutinin (HA), all but one
commonly circulate in birds, in which they generally cause mild to no disease.

84

Reassortment and mutation events periodically produce influenza viruses capable of
infecting mammals, and several influenza strains are now endemic in mammalian
species including swine, horses, dogs, and humans. In humans, this process has
resulted in several pandemics, during the 20

th

st

and 21

centuries, which have been

associated with high mortality (Morens and Taubenberger 2010, Taubenberger and
Kash 2010). Accordingly, public health services now closely monitor the sporadic
infections of humans with avian influenza viruses, in the hope of predicting and
preventing the next pandemic.
One avian influenza subtype with pandemic potential is the H7 lineage, in combination
with several neuraminidase (NA) subtypes (reviewed in (Belser, Bridges et al. 2009)).
These viruses have sporadically infected humans in multiple countries. In most cases,
these infections have been mild, often causing only conjunctivitis (Belser, Bridges et al.
2009). Furthermore, on several occasions, H7 viruses have caused outbreaks that have
infected larger contingents. In the Netherlands in 2003, at least 89 people were infected
with an H7N7 influenza virus, and there was one fatality due to severe pneumonia
(Fouchier, Schneeberger et al. 2004, Koopmans, Wilbrink et al. 2004). Recently, a
human outbreak by a novel H7N9 avian influenza virus has led to the infection of 135
people in China, with frequent severe disease and a mortality rate of over 30% (Chen,
Liang et al. 2013, Li, Zhou et al. 2013) as of August 12, 2013 (World Health
Organization 2013).
Serological testing of people for evidence of infection with H7 viruses has proven
difficult. Titers measured using hemagglutination inhibition (HI) and microneutralization

85

(MN) assays have generally been very low (Du Ry van Beest Holle, Meijer et al. 2005,
Puzelli, Di Trani et al. 2005, Skowronski, Li et al. 2007, Kayali, Setterquist et al. 2008,
Bos, Te Beest et al. 2010, Kayali, Ortiz et al. 2010, Chen, Liang et al. 2013, Li, Zhou et
al. 2013) even in people known to have been infected with H7 viruses (Meijer, Bosman
et al. 2006). This poor antibody response apparently extends to the vaccines against H7
that have been evaluated in pre-clinical and clinical studies (Joseph, McAuliffe et al.
2008, Cox, Madhun et al. 2009). For example, in a recent clinical trial of an H7N7
vaccine, only 2 of 99 vaccinees developed significant hemagglutination inhibition (HI)
titers against H7 (HI titer ≥32), even though high doses of the protein (up to 90 µg per
person) were used (Couch, Patel et al. 2012).
To our knowledge, the intrinsic immunogenicity of H7 HA has not been directly
compared to that of seasonal viruses (such as H3N2 and H1N1pdm09) that are known
to induce acceptable HI titers. We therefore used recombinant HA to compare
immunogenicity of several HA subtypes, including two H7 subtypes, in naïve mice.
Materials and Methods:

Recombinant hemagglutinin. The purified recombinant hemagglutinins (HA) and HA1
were expressed and purified from baculovirus vectors in insect cells, as described for
HA from A/Netherlands/219/2003(H7N7) (NL219) (Yang, Carney et al. 2012); A/New
York/107/2003(H7N3)

(NY107)

A/California/7/2009(H1N1pdm09)

(Yang,

(CA07)

(Chen,

Chen

et

Rivailler

et

al.
al.

2010);

2011);

and

A/Perth16/2009(H3N2) (Perth16) (Stevens, Chen et al. 2010). GenBank accession

86

numbers for the HAs are ACS71642.1 (Perth16), ACP41953.1 (CA07), AAR02640.1
(NL219), and ACC55270.1 (NY107).

Viruses. Wild-type Perth16 and 2:6 A/Puerto Rico/1/1934 (PR8) reassortants of NL219,
NY107, and A/Texas/5/2009(H1N1pdm09) (TX05, antigenically homologous to CA07)
were used in hemagglutination inhibition (HI) and microneutralization (MN) assays. All
the viruses used in this study were propagated in 10-11 days old embryonated chicken
eggs and were stored as allantoic fluid.

Immunization. Female 6-8 week old BALB/c mice (Jackson Laboratory, Bar Harbor,
ME) were used in this study. For immunization, groups of five mice were immunized
intramuscularly (IM) 2 times, with 3 weeks between injections, with either 3 or 10 µg of
purified recombinant HA in sterile PBS. Each immunization study included a control
group of 5 mice who were injected with sterile PBS. These doses were chosen based
on a pilot dose-response study (data not shown). Blood samples were collected on day
0 (before initial immunization), day 21 (before the boost), and on days 35, 49, and 63.

Animals were housed and handled under ABSL2 conditions. Live H7 viruses were
handled under BSL3+ conditions (Chosewood, Wilson et al. 2009). Animal research
was conducted under the guidance of the Centers for Disease Control and Prevention's
Institutional Animal Care and Use Committee in an Association for Assessment and
Accreditation of Laboratory Animal Care International-accredited animal facility.

Antibody assays. All sera were initially incubated with receptor-destroying enzyme
(RDE, from Vibrio cholerae, Denka Seiken, Tokyo, Japan), at the ratio of one volume of

87

o

o

serum and three volumes of RDE, overnight at 37 C. RDE was inactivated at 56 C for
30 minutes and then 6 volumes of PBS were added to each volume of serum, making
the final serum dilution of 1:10. Serum antibody titers to TX05, Perth16, NL219, and
NY107 were measured by HI, ELISA and MN. Each serum sample was tested at least
in duplicate. Known-positive ferret sera for homologous viruses were used as positive
controls.

HI assay. 96-well V-bottom plates were used for HA titration and HI assays. Freshly
prepared 0.5% turkey red blood cells RBC (TRBC) or 1.0% horse RBC (HRBC) were
used for HA titration and HI assay of seasonal and H7 sera samples, respectively, as
described (World Health Organization 2011). Briefly, two-fold serial dilutions of sera (25
µl) were incubated with 4 HA units (25 µl) of homologous virus for 30 min at room
temperature. Subsequently, 50 µl of RBCs were added for 30 min (TRBC, seasonal
samples) or 1 hour (HRBC, H7 samples) and wells were read for inhibition of
hemagglutination. HI titers were calculated as the reciprocal of the highest dilution of
serum that inhibited 100 % of the hemagglutination of 4 HA units of homologous virus.
As positive controls, we used serum from ferrets infected with wild-type CA07, Perth16,
NL219 or NY107. The minimum detectable limit of this assay was an HI titer of 10.

ELISA. Individual wells of 96-well high binding plates (Costar, Corning, NY) were
coated with purified recombinant full length ectodomain HA, or the HA1 domain, at a
o

concentration of 1 µg/ml at 4 C overnight. Recombinant full-length HAs used as antigen
for ELISA were obtained through the Influenza Reagent Resource (IRR: Influenza
Division, WHO Collaborating Center for Surveillance, Epidemiology and Control of
88

Influenza, Centers for Disease Control and Prevention, Atlanta, GA) with the following
catalog numbers: Perth16 HA, (FR-472), NL219 HA (FR-71), CA07 (FR-559), NY107
(FR-69). Wells were washed with PBS and then blocked with 0.1% bovine serum
o

albumin in PBS for one hour at 37 C. Blocking buffer was discarded and serially diluted
o

sera samples were added to the plates and incubated for 1 hour at 37 C. Plates were
washed with phosphate-buffered saline (pH 7.4) containing 0.05% Tween-20 (PBST)
and HRP-conjugated Goat Anti-Mouse IgG (Alpha Diagnostic, San Antonio, TX) was
o

diluted and applied to each well. Plates were incubated for 1 hr at 37 C and were then
washed with PBS. O-phenylenediamine (OPD) solution was added to each well as a
peroxidase substrate and the reaction was stopped after 10 minutes by adding 50 µl 3N
HCl. Absorbance was measured at 490 nm. Antibody titers are given as the reciprocal
of the highest dilution which gave an OD490 value greater than 2 times the average of
the background wells.
Avidity assay. For avidity ELISA assays (el Zayyat, Khalifa et al. 1998, Bjorkman,
Naslund et al. 1999, Rahbari, Keshavarz et al. 2012) we modified our ELISA protocol as
follows. Plates and sera were prepared as described above, and serum samples were
applied to the wells in duplicate. After one hour incubation, one replicate of samples
was incubated with PBST for 5 minutes and the other replicate of samples were
incubated in 4M urea (Sigma-Aldrich, St. Louis, MO) for 5 minutes. All samples were
then washed twice with PBST, after which the rest of the ELISA protocol was completed
as described above. The avidity index (AI) was calculated as previously described
(Rahbari, Keshavarz et al. 2012) for the lowest dilution (1:2500) as AI= (U+/U-) x100,
89

where “U+” is the OD490 for wells washed with urea and “U-” is the OD490 for wells
washed with PBST.
Microneutralization assay. MN assays were performed as described (World Health
Organization 2011). RDE-treated sera samples were added to 96-well flat bottom
microtiter plates (Costar) and twofold serially diluted with diluent (DMEM supplemented
with 1% BSA, 100 U/ml penicillin, 100 µg/ml streptomycin, 20 mM HEPES) for a final
well volume of 50 µl. 100 tissue culture infectious dose 50 (TCID50) of virus in 50 µl
diluent was then added to all sample wells. As well as test samples, each plate
contained 4 wells of cell control (cells without virus and serum), 4 wells of virus control
(cells with virus and no serum), and back titration (cells with 2 fold dilution of virus). As
positive controls, we used serum from ferrets infected with wild-type Perth16, CA07,
o

NL219 or NY107. Plates were then incubated at 37 C for one hour followed by addition
5

of 1.5 x 10 cells/ml in 100 µl to each well. Plates were incubated overnight (18-22 hrs)
o

at 37 C. Plates were washed once with PBS, and then treated with cold fixative (80%
acetone/20% PBS) for 8 minutes. An ELISA was then performed in order to determine
antibody titers of serum samples as follows. Fixed plates were washed twice with PBST
and 100 µl of anti-influenza A antibody (Millipore, Billerica, MA) diluted in blocking buffer
(1% BSA-PBST) was added to each well. After 1 hour incubation, plates were then
washed in PBST three times and 100 µl of horseradish peroxidase-conjugated goat
anti-mouse IgG diluted in blocking buffer was added to each well. The plates were
incubated at room temperature for 1 hour and then washed 4 times with PBST. 100 µl of
freshly prepared OPD in citrate buffer was then added to each well and the reaction was
90

stopped by adding 50 µl 3N HCL after the color in the virus control wells reached a deep
orange-yellow color. Absorbance was measured at 390 nm and the cutoff was
determined by the following equation: X = [(median OD of VC wells) + (median OD of
CC wells)]/2. All values equal to or less than X were considered positive for
neutralization. Antibody titers provided are the reciprocal of the serum dilution for the
corresponding well considered positive for neutralization.
Statistical analyses. Statistical analyses for serum HI, MN, and ELISA titers were
performed using a linear mixed model with repeated measures, implemented in SAS,
using a cutoff of p ≤ 0.05 for significance. For avidity ELISAs, 1-way ANOVA was
performed and significance was determined using a two-sided Student t test, with a
cutoff of p ≤ 0.05 for significance.
Results:
ELISA responses to purified HA. ELISA responses were lowest to the HA from
NY107 (Figure 3.1A, C, E). Perth16 HA induced the highest response, and this was
statistically significant (p<0.05) for the 10 µg group (Figure 3.1C, E). However, NL219
responses did not differ significantly from the responses to the seasonal viruses for the
3 µg dose (Figure 3.1A, E), and was similar to H1pdm09 for the 10 µg dose. The same
pattern was seen in considering only the maximum titer reached at any time point, with
Perth16 being the highest and NY107 being the lowest. CA07 and NL219 titers were
similar to each other, and in the 10 µg dose significantly lower than Perth16 (Figure
3.1E). Thus, although by ELISA NY107 was less immunogenic than the other three
viruses tested, NL219 was similar to the seasonal virus CA07.

91

Similar results were obtained using HA1 as the antigen in ELISA (Figure 3.1B, D, F),
although actual titers were significantly lower, as expected (since in this assay
antibodies raised against the conserved stalk component of HA would not be detected).
Perth16 HA induced the greatest HA1-specific response (p<0.05) while the response to
CA07 was similar response to that for NL219 or NY107.
Hemagglutination inhibition (HI) antibodies. While ELISA measures the overall
antibody response to HA, HI assays measure the antibodies that bind at or near the HA
receptor binding site. Both the H7 HAs gave significantly lower HI titers than both the
seasonal HAs at the 3 µg dose (Figure 3.2A). However, at the 10 µg dose, the two H7
viruses achieved similar titers to CA07, though all three of these HAs induced lower
titers than did Perth16 (Figure 3.2B). A similar pattern was seen when considering only
the GMT of the maximum titers induced by the HAs (Figure 3.2C).
Virus-neutralizing antibodies. Both the seasonal HAs gave MN titers that were
generally equal to, or higher than, their HI titers (Figure 3.3A-C). In contrast, only two of
40 mice immunized with H7 HAs gave detectable MN titer during the whole study time.
As positive controls we used sera from ferrets infected with wild-type NL219 or NY107.
These sera had HI titers of 1280 and 320 for NL219 and NY107, respectively (Figure
3.3D) and MN titers were readily measured for these sera (Figure 3.3D) indicating that
the MN assay is technically functional for H7 viruses.
Since the average HI GMT of H7-immunized mice were lower than that of seasonal HAimmunized mice, it was possible that MN titers were only proportionally reduced in H7immunized mice. Since no H7-immunized mice achieved HI titers greater than 160, we

92

selected sera from individual animals, immunized with each HA, with HI titers of 80 or
160, and compared MN titers for each sample. We identified samples from 6 mice each
for Perth16, CA07, NL219, and NY107. (Where the same animal had more than one
sample with titers of 80 or 160, only one sample was chosen.) HI titers were similar for
all HA, as expected, since these sera were specifically selected to be comparable
(Figure 3.4A). As noted, none of the H7-immunized mice had detectable MN titers.
However, all of the seasonal HAs induced easily-detectable MN responses, with MN
titers being similar to or higher than the HI titers (Figure 3.4A). Thus, even when HI
titers were similar, H7 HAs induced relatively low MN titers compared to H3 and
H1pdm09.
Avidity ELISAs. To compare the avidity of the antibodies induced by each recombinant
HA, we compared the effect of a 5 min PBST wash vs. a 5 min wash in 4 M urea on
ELISA titers, as previously described (Bjorkman, Naslund et al. 1999). Again, we used
the serum samples from mice with comparable HI titers of 80-160. Standard ELISA
titers for these samples were also comparable (Figure 3.4B, “PBST”), although the titer
for CA07 was slightly higher than for Perth16 (p=0.043). Adding the urea wash (Figure
3.4B, “Urea”) did not reduce the detected titer in serum samples from mice immunized
with Perth16 HA, but caused moderate (approximately 2-fold) reduction in titers for
CA07 and NL219 sera, and a 3.5-fold reduction in titer for the NY107 sera (Figure
3.4C). Following the urea wash, both NL219 and NY107 titers were significantly lower
than either Perth16 or CA07 titers (p < 0.05), while CA07 and Perth16 titers were still
comparable (p=0.363). The avidity index (AI) for the H7 HAs was significantly lower than
for either seasonal HA (Figure 3.4C) (p <0.05 vs. CA07; p <0.001 vs. Perth16).
93

Discussion:
Human serological responses to several strains of avian influenza viruses seem to be
weaker than those to seasonal influenza strains (Beare and Webster 1991, Treanor,
Wilkinson et al. 2001, Beigel, Voell et al. 2009, Brady, Treanor et al. 2009, Nicholson,
Thompson et al. 2009, Talaat, Karron et al. 2011). In particular, people infected with H7
subtype influenza viruses rarely develop high antibody titers, as measured by HI or MN
assays (Du Ry van Beest Holle, Meijer et al. 2005, Puzelli, Di Trani et al. 2005, Meijer,
Bosman et al. 2006, Skowronski, Li et al. 2007, Kayali, Setterquist et al. 2008, Bos, Te
Beest et al. 2010, Kayali, Ortiz et al. 2010, Chen, Liang et al. 2013, Li, Zhou et al.
2013). While this may reflect the nature of infection with H7 viruses, which commonly
cause superficial infections such as conjunctivitis rather than respiratory or systemic
disease, this was also seen with people immunized with H7 vaccines (Joseph, McAuliffe
et al. 2008, Cox, Madhun et al. 2009, Couch, Patel et al. 2012), suggesting that humans
may generally respond poorly to H7 subtypes of HA. However, in the absence of studies
directly comparing the response to H7 to the response to human seasonal influenza
viruses such as H1N1pdm09 and H3N2 strains, it has not been clear that H7 is
exceptional. Furthermore, it is not clear if any difference in response might be related to
intrinsic aspects of the HA, to pre-existing immunity in humans, or to some aspect of the
virus or the vaccine. To distinguish between these possibilities, we immunized naïve
mice with purified recombinant HA from two human seasonal viruses, H1N1pdm09 and
H3N2, and two different H7 viruses, and compared the antibody response to each. The
H7 HAs were from NL219 (H7N7), a highly pathogenic avian influenza virus (HPAIV)
representing the Eurasian lineage of H7 (Fouchier, Schneeberger et al. 2004), and
94

NY107, a low-pathogenicity avian influenza virus (LPAIV) representing the North
American lineage of H7 (Ostrowsky, Huang et al. 2012). Both viruses were isolated from
infected humans.
ELISA assays measure a broad range of antibodies, including those that are protective
as well as those that bind to HA but fail to protect. When measured by ELISA assays,
the HA from NY107 was somewhat less immunogenic than either seasonal HA.
However, the response to the HA from NL219 was similar to that to H1pdm09, though
lower than that to Perth16. When measured by HI assays, both H7 HAs were
considerably less immunogenic than either H3 or H1pdm09. HI assays measure the
ability of antibodies to prevent RBC hemagglutination by influenza viruses, mainly due
to blockade of the receptor-binding site (RBS) on the globular head of the HA (Katz,
Hancock et al. 2011). Due to their correlation with clinical protection (Katz, Hancock et
al. 2011), HI assays are widely used for determining antibody responses to influenza
viruses. Our results here are consistent with clinical findings that HI titers following
infection with H7 influenza viruses tend to be low.
The microneutralization assay (MN) is a third widely used assay for measuring antibody
responses to influenza viruses. MN assays functionally measure the ability of antibodies
to protect cells against infection, and as such are highly biologically relevant. By this
assay, the difference between the responses to H7 HA, and HA from seasonal viruses
was most dramatic. Even when limiting analysis to sera from H7-immunized mice where
an HI response to H7 was clearly detectable, MN responses were absent or very low
(Figure 3.4A). In contrast, all mice immunized with H3, and almost all those immunized
with H1pdm, developed strong MN titers.
95

The H7 HAs, especially NL219, thus showed discordant results depending on the assay
used to measure antibodies. Thus although by ELISA NL219 showed equivalent titers to
CA07, the corresponding HI titers were comparatively low, and MN titers were almost
undetectable in the same samples. If antibodies are being generated to H7 viruses (as
shown in ELISA), why are they not detected in HI and MN assays? Two possibilities are
that, although antibodies are being produced, they are either low affinity antibodies, or
they target non-neutralizing regions of the HA, such as the stalk. We used HA1 as the
antigen

in

ELISA

assays

to

test

whether

the

antibody

response

to

H7s

disproportionately targets the stalk region of HA. However, although titers against HA1
were significantly reduced for all HAs compared to those against the full-length HA,
NL219 and the seasonal HAs all showed a 40-50% reduction in titer (Figure 3.1D).
Titers for NY107 were very low for full-length HA and almost undetectable for HA1.
Since the HA1-specific responses for NL219 and NY107 were similar to that for CA07 (p
> 0.05), there is no evidence that the H7 and H1pdm09 HAs induced different levels of
HA1-specific antibodies.
The antibodies produced by B cells increase in affinity through the course of an
infection, as activated B cells undergo somatic hypermutation of their immunoglobulin
variable regions (reviewed in (Teng and Papavasiliou 2007). This affinity maturation is
enhanced by inflammation, including adjuvants (Khurana, Verma et al. 2011), as well as
repeated antigen exposure (Teng and Papavasiliou 2007). Testing the avidity of the
antisera suggested that H7, especially from NY107, did indeed induce low-affinity
antibodies, since a brief urea wash reduced the bound antibody titer dramatically
(Figure 3.4B, C). Antibodies induced by NL219 were also of relatively low affinity.
96

Interestingly, not only did Perth16 HA induce the highest titers of antibodies as
measured in all assays, the avidity of these antibodies is also apparently very high, with
titers being unaffected by the urea wash (avidity index = 100.3%).
The molecular mechanism(s) for the difference in immunogenicity is unknown. In
humans infected with avian influenza, weak immunogenicity may reflect the lack of prior
exposure to avian HA, compared to the repeated exposure to seasonal HA that may act
as priming immunization. Another possibility is that some feature of the vaccine or virus,
other than HA itself, may influence the response. However, in these experiments, the
use of naïve mice and purified recombinant HA suggests that intrinsic features of the
HA itself may affect immunogenicity.
One possible explanation is that the amount of glycosylation on the HA may influence
its interaction with the immune system (Crouch, Hartshorn et al. 2000, Reading, Tate et
al. 2007, Reading, Pickett et al. 2009, Wang, Chen et al. 2009, Tate, Brooks et al. 2011,
Tate, Job et al. 2011, Wanzeck, Boyd et al. 2011, Medina, Stertz et al. 2013). Although
H3N2 viruses are believed to have evolved increasing glycosylation over time in order
to prevent immune recognition of antigenic sites (Cherry, Lipman et al. 2009, Das,
Puigbo et al. 2010, Suzuki 2011, Kobayashi and Suzuki 2012), the protection is mainly
against memory responses rather than primary induction of antibodies. That is, added
glycosylation reduces recognition of the virus by pre-formed antibodies raised against
previous versions of the virus, whereas increasing glycosylation of HA increases the
immunogenicity, and reduces the pathogenicity, of H3N2 (Vigerust, Ulett et al. 2007,
Tate, Job et al. 2011, Wanzeck, Boyd et al. 2011) and other influenza viruses (Medina,
Stertz et al. 2013, Sun, Jayaraman et al. 2013). The findings here are consistent with
97

those findings, since the number of potential glycosylation sites on the HAs in these
experiments parallels their immunogenicity, with Perth16 (the most immunogenic HA)
having 11 potential carbohydrate addition sites, and CA07, NL219, and NY107 having
7, 6, and 5 sites, respectively.
It is important to note that these results are based on immunization with purified
recombinant protein. Purified proteins are intrinsically less immunogenic than the same
protein in a vaccine, which not only presents the proteins in multimeric form but also
provide adjuvant factors, such as viral nucleic acids (Jeisy-Scott, Kim et al. 2012).
Similarly, viral infection provides strong inflammatory signals that might be able to
overcome low intrinsic immunogenicity. For example, even though immunization of mice
with recombinant H7 from NL219 and NY107 induced weak or no MN titers, positivecontrol serum from ferrets infected with these viruses had easily detectable MN
antibodies (Figure 3.3D). It is interesting (and consistent with findings for humans
infected with H7 viruses) that for both cases the MN titer is lower than the HI titer, while
for seasonal viruses the reverse is typically true (compare Figures 3.2 and 3.3).
Nevertheless, our findings suggest that, with all other factors being equal, H7 HAs may
be less immunogenic than seasonal H3 or H1pdm09 HA, and H7 vaccines may require
higher doses or adjuvant, or may require introduced mutations in HA, in order to
achieve acceptable titers.

98

Acknowledgements:
We thank James Stevens, Paul Carney, and Hua Yang for purified recombinant
hemagglutinins. Amanda Balish and Jessica Belser for positive ferret sera, Li-mei Chen
and Jaber Hossain for PR8 ressortant viruses.

99

APPENDIX

100

APPENDIX: Figures

Full-length HA

HA1

A

B
*

ELISA GMT

65536

21
35
49
63

16384

4096

65536

*
*

16384

35
49
63

4096

1024

1024
Perth/16

CA/07

NL/219

NY/107

Perth/16 CA/07

NL/219 NY/107

Figure 3.1 ELISA responses to recombinant hemagglutinins. Mice were immunized with (A, B) 3 µg or (C, D)
10 µg of purified recombinant HA from Perth/16 (H3), CA/07 (H1pdm), NL/219 (H7, Eurasian lineage), or NY/107 (H7,
North American lineage) on two occasions 21 days apart. On days 0, 21, 35, 49, and 63, serum anti-HA titer was tested
by an ELISA assay, using purified recombinant homologous full-length HA (A,C, E) or for days 35, 49, and 63, HA1 (B, D,
F) as antigen.
(E,F) Same experiment as A-D, showing the highest titer achieved by each mouse at any time point.
Values are shown as geometric mean titers (GMT). Error bars represent 95% confidence intervals. Differences between
groups (A-D) or peak GMT (E,F) are indicated by * (0.0001<p<0.05) or ** (p<0.0001). One representative experiment of
at least 3 replicate experiments.

101

Figure 3.1 (cont’d)

**

C

D

*

**

65536

21
35
49
63

ELISA GMT

*
16384

65536

*

35
49
63

16384

4096

4096

1024

1024
Perth/16

CA/07

NL/219

NY/107

Perth/16 CA/07

NL/219 NY/107

*

*

3 µg

10 µg

*
E

*
65536

ELISA GMT

F

*
Perth/16
CA/07
NL/219
NY/107

16384

65536

16384

4096

4096

1024

1024
33ugµg

10 µg
ug
10

102

Perth/16
CA/07
NL/219
NY/107

HI GMT

A

4096

21

*

1024

35
49

256

63

64
16
4
Perth/16

CA/07

B
HI GMT

4096

*

1024

NL/219

NY/107

**

21
35
49
63

256
64
16
4
Perth/16

CA/07

NL/219

C
4096

HI GMT

1024

NY/107

**
**

Perth/16
CA/07
NL/219
NY/107

256
64
16
4
3 ug

10 ug

Figure 3.2 HI responses to recombinant hemagglutinins. Mice were immunized
with (A) 3 µg or (B) 10 µg of purified recombinant HA from Perth/16, CA/07, NL/219, or
NY/107 on two occasions 21 days apart. On days 0, 21, 35, 49, and 63, serum anti-HA
titer was tested by an HI assay, using homologous live virus as antigen. (C) Same
experiment as A and B, showing the highest titer achieved by each mouse at any time
point. Values are shown as GMT. Error bars represent 95% confidence intervals.
Differences between groups (A, B) or peak GMT (C) are indicated by * (0.0001<p<0.05)
or ** (p<0.0001). One representative experiment of at least 3 replicates.
103

*

A

21
35
49
63

MN GMT

2048

*

512
128
32
8
Perth/16

B

NL/219

NY/107

*

2048

MN GMT

CA/07

21
35
49
63

512
128
32
8
Perth/16

CA/07

NL/219

NY/107

Figure 3.3 MN responses to recombinant hemagglutinins. Mice were immunized
with (A) 3 µg or (B) 10 µg of purified recombinant HA from Perth/16, CA/07, NL/219, or
NY/107 on two occasions 21 days apart. On days 0, 21, 35, 49, and 63, serum anti-HA
titer was tested by a MN assay, using homologous live virus as antigen. (C) Same
experiment as A and B, showing the highest titer achieved by each mouse at any time
point. (D) MN and HI values for positive control ferret sera, from ferrets infected with
live NL/219 or NY/107. Values are shown as GMT. Error bars represent 95%
confidence intervals. Differences between groups (A, B) or peak GMT (C) are indicated
by * (0.0001<p<0.05) or ** (p<0.0001). One representative experiment of at least 3
replicates.

104

Figure 3.3 (cont’d)

C
2048

**
*

MN GMT

512
128
32
8
3 ug

10 ug

D
HI
MN

MN GMT

2048
512
128
32
8
NL/219

NY/107

105

Perth/16
CA/07
NL/219
NY/107

A

*
512

Log2

128

Perth/16
CA/07
NL/219
NY/107

32
8
HI

B

MN

*

*

Log2 GMT

16384

Perth/16
CA/07
NL/219
NY/107

4096

1024
PBST

Avidity index

C

*
120
100
80
60
40
20
0

Urea

*

*

Figure 3.4 HI and MN titers and avidity ELISA for mice with HI titers between 80
and 160. Serum samples from mice with HI titers between 80 and 160 were selected
(n=6 mice per group), and avidity ELISAs were performed as described in the
manuscript. (A) HI and MN titers for each group. (B) To test avidity of antibodies,
ELISAs were performed according to the standard procedure with the addition of a 5
min PBST incubation (“PBST”, left grouping) or with a 5-minute incubation in 4M urea,
followed by two washes with PBST (“Urea”, right grouping). Values are shown as GMT.
(C) The avidity index (AI) was calculated for each sample. Values are shown as
average AI (percent). Error bars represent 95% confidence intervals. Differences
between groups are indicated by * (0.0001<p<0.05) or ** (p<0.0001).
106

REFERENCES

107

REFERENCES

Beare, A. S. and R. G. Webster (1991). "Replication of avian influenza viruses in
humans." Arch Virol 119(1-2): 37-42.
Beigel, J. H., J. Voell, C. Y. Huang, P. D. Burbelo and H. C. Lane (2009). "Safety and
immunogenicity of multiple and higher doses of an inactivated influenza A/H5N1
vaccine." J Infect Dis 200(4): 501-509.
Belser, J. A., C. B. Bridges, J. M. Katz and T. M. Tumpey (2009). "Past, present, and
possible future human infection with influenza virus A subtype H7." Emerg Infect Dis
15(6): 859-865.
Bjorkman, C., K. Naslund, S. Stenlund, S. W. Maley, D. Buxton and A. Uggla (1999).
"An IgG avidity ELISA to discriminate between recent and chronic Neospora caninum
infection." J Vet Diagn Invest 11(1): 41-44.
Bos, M. E., D. E. Te Beest, M. van Boven, M. R. van Beest Holle, A. Meijer, A. Bosman,
Y. M. Mulder, M. P. Koopmans and A. Stegeman (2010). "High probability of avian
influenza virus (H7N7) transmission from poultry to humans active in disease control on
infected farms." J Infect Dis 201(9): 1390-1396.
Brady, R. C., J. J. Treanor, R. L. Atmar, W. A. Keitel, R. Edelman, W. H. Chen, P.
Winokur, R. Belshe, I. L. Graham, D. L. Noah, K. Guo and H. Hill (2009). "Safety and
immunogenicity of a subvirion inactivated influenza A/H5N1 vaccine with or without
aluminum hydroxide among healthy elderly adults." Vaccine 27(37): 5091-5095.
Chen, L. M., P. Rivailler, J. Hossain, P. Carney, A. Balish, I. Perry, C. T. Davis, R.
Garten, B. Shu, X. Xu, A. Klimov, J. C. Paulson, N. J. Cox, S. Swenson, J. Stevens, A.
Vincent, M. Gramer and R. O. Donis (2011). "Receptor specificity of subtype H1
influenza A viruses isolated from swine and humans in the United States." Virology
412(2): 401-410.
Chen, Y., W. Liang, S. Yang, N. Wu, H. Gao, J. Sheng, H. Yao, J. Wo, Q. Fang, D. Cui,
Y. Li, X. Yao, Y. Zhang, H. Wu, S. Zheng, H. Diao, S. Xia, Y. Zhang, K. H. Chan, H. W.
Tsoi, J. L. Teng, W. Song, P. Wang, S. Y. Lau, M. Zheng, J. F. Chan, K. K. To, H. Chen,
L. Li and K. Y. Yuen (2013). "Human infections with the emerging avian influenza A
H7N9 virus from wet market poultry: clinical analysis and characterisation of viral
genome." Lancet.
Cherry, J. L., D. J. Lipman, A. Nikolskaya and Y. I. Wolf (2009). "Evolutionary dynamics
of N-glycosylation sites of influenza virus hemagglutinin." PLoS Curr 1: RRN1001.
108

Chosewood, L. C., D. E. Wilson, Centers for Disease Control and Prevention (U.S.) and
National Institutes of Health (U.S.) (2009). Biosafety in microbiological and biomedical
laboratories. Washington, D.C., U.S. Dept. of Health and Human Services, Public
Health Service, Centers for Disease Control and Prevention, National Institutes of
Health.
Couch, R. B., S. M. Patel, C. L. Wade-Bowers and D. Nino (2012). "A randomized
clinical trial of an inactivated avian influenza A (H7N7) vaccine." PLoS One 7(12):
e49704.
Cox, R. J., A. S. Madhun, S. Hauge, H. Sjursen, D. Major, M. Kuhne, K. Hoschler, M.
Saville, F. R. Vogel, W. Barclay, I. Donatelli, M. Zambon, J. Wood and L. R. Haaheim
(2009). "A phase I clinical trial of a PER.C6 cell grown influenza H7 virus vaccine."
Vaccine 27(13): 1889-1897.
Crouch, E., K. Hartshorn and I. Ofek (2000). "Collectins and pulmonary innate
immunity." Immunol Rev 173: 52-65.
Das, S. R., P. Puigbo, S. E. Hensley, D. E. Hurt, J. R. Bennink and J. W. Yewdell
(2010). "Glycosylation focuses sequence variation in the influenza A virus H1
hemagglutinin globular domain." PLoS Pathog 6(11): e1001211.
Du Ry van Beest Holle, M., A. Meijer, M. Koopmans and C. M. de Jager (2005).
"Human-to-human transmission of avian influenza A/H7N7, The Netherlands, 2003."
Euro Surveill 10(12): 264-268.
el Zayyat, E. A., K. E. Khalifa, I. M. el Gindy, A. Awad and M. E. Azab (1998).
"Evaluation of specific immunoglobulin G avidity enzyme linked immunosorbent assay
(IgG avidity ELISA) in diagnosis of early and chronic Schistosomiasis mansoni." J Egypt
Soc Parasitol 28(3): 739-752.
Fouchier, R. A., P. M. Schneeberger, F. W. Rozendaal, J. M. Broekman, S. A. Kemink,
V. Munster, T. Kuiken, G. F. Rimmelzwaan, M. Schutten, G. J. Van Doornum, G. Koch,
A. Bosman, M. Koopmans and A. D. Osterhaus (2004). "Avian influenza A virus (H7N7)
associated with human conjunctivitis and a fatal case of acute respiratory distress
syndrome." Proc Natl Acad Sci U S A 101(5): 1356-1361.
Jeisy-Scott, V., J. H. Kim, W. G. Davis, W. Cao, J. M. Katz and S. Sambhara (2012).
"TLR7 recognition is dispensable for influenza virus A infection but important for the
induction of hemagglutinin-specific antibodies in response to the 2009 pandemic split
vaccine in mice." J Virol 86(20): 10988-10998.
Joseph, T., J. McAuliffe, B. Lu, L. Vogel, D. Swayne, H. Jin, G. Kemble and K.
Subbarao (2008). "A live attenuated cold-adapted influenza A H7N3 virus vaccine
provides protection against homologous and heterologous H7 viruses in mice and
ferrets." Virology 378(1): 123-132.
109

Katz, J. M., K. Hancock and X. Xu (2011). "Serologic assays for influenza surveillance,
diagnosis and vaccine evaluation." Expert Rev Anti Infect Ther 9(6): 669-683.
Kayali, G., E. J. Ortiz, M. L. Chorazy and G. C. Gray (2010). "Evidence of previous
avian influenza infection among US turkey workers." Zoonoses Public Health 57(4):
265-272.
Kayali, G., S. F. Setterquist, A. W. Capuano, K. P. Myers, J. S. Gill and G. C. Gray
(2008). "Testing human sera for antibodies against avian influenza viruses: horse RBC
hemagglutination inhibition vs. microneutralization assays." J Clin Virol 43(1): 73-78.
Khurana, S., N. Verma, J. W. Yewdell, A. K. Hilbert, F. Castellino, M. Lattanzi, G. Del
Giudice, R. Rappuoli and H. Golding (2011). "MF59 adjuvant enhances diversity and
affinity of antibody-mediated immune response to pandemic influenza vaccines." Sci
Transl Med 3(85): 85ra48.
Kobayashi, Y. and Y. Suzuki (2012). "Evidence for N-glycan shielding of antigenic sites
during evolution of human influenza A virus hemagglutinin." J Virol 86(7): 3446-3451.
Koopmans, M., B. Wilbrink, M. Conyn, G. Natrop, H. van der Nat, H. Vennema, A.
Meijer, J. van Steenbergen, R. Fouchier, A. Osterhaus and A. Bosman (2004).
"Transmission of H7N7 avian influenza A virus to human beings during a large outbreak
in commercial poultry farms in the Netherlands." Lancet 363(9409): 587-593.
Li, Q., L. Zhou, M. Zhou, Z. Chen, F. Li, H. Wu, N. Xiang, E. Chen, F. Tang, D. Wang, L.
Meng, Z. Hong, W. Tu, Y. Cao, L. Li, F. Ding, B. Liu, M. Wang, R. Xie, R. Gao, X. Li, T.
Bai, S. Zou, J. He, J. Hu, Y. Xu, C. Chai, S. Wang, Y. Gao, L. Jin, Y. Zhang, H. Luo, H.
Yu, L. Gao, X. Pang, G. Liu, Y. Shu, W. Yang, T. M. Uyeki, Y. Wang, F. Wu and Z. Feng
(2013). "Preliminary Report: Epidemiology of the Avian Influenza A (H7N9) Outbreak in
China." N Engl J Med.
Medina, R. A., S. Stertz, B. Manicassamy, P. Zimmermann, X. Sun, R. A. Albrecht, H.
Uusi-Kerttula, O. Zagordi, R. B. Belshe, S. E. Frey, T. M. Tumpey and A. Garcia-Sastre
(2013). "Glycosylations in the globular head of the hemagglutinin protein modulate the
virulence and antigenic properties of the H1N1 influenza viruses." Sci Transl Med
5(187): 187ra170.
Meijer, A., A. Bosman, E. E. van de Kamp, B. Wilbrink, M. Du Ry van Beest Holle and
M. Koopmans (2006). "Measurement of antibodies to avian influenza virus A(H7N7) in
humans by hemagglutination inhibition test." J Virol Methods 132(1-2): 113-120.
Morens, D. M. and J. K. Taubenberger (2010). "Historical thoughts on influenza viral
ecosystems, or behold a pale horse, dead dogs, failing fowl, and sick swine." Influenza
Other Respi Viruses 4(6): 327-337.

110

Nicholson, K. G., C. I. Thompson, J. M. Klap, J. M. Wood, S. Batham, R. W. Newman,
R. Mischler, M. C. Zambon and I. Stephenson (2009). "Safety and immunogenicity of
whole-virus, alum-adjuvanted whole-virus, virosomal, and whole-virus intradermal
influenza A/H9N2 vaccine formulations." Vaccine 28(1): 171-178.
Ostrowsky, B., A. Huang, W. Terry, D. Anton, B. Brunagel, L. Traynor, S. Abid, G.
Johnson, M. Kacica, J. Katz, L. Edwards, S. Lindstrom, A. Klimov and T. M. Uyeki
(2012). "Low pathogenic avian influenza A (H7N2) virus infection in
immunocompromised adult, New York, USA, 2003." Emerg Infect Dis 18(7): 1128-1131.
Puzelli, S., L. Di Trani, C. Fabiani, L. Campitelli, M. A. De Marco, I. Capua, J. F.
Aguilera, M. Zambon and I. Donatelli (2005). "Serological analysis of serum samples
from humans exposed to avian H7 influenza viruses in Italy between 1999 and 2003." J
Infect Dis 192(8): 1318-1322.
Rahbari, A. H., H. Keshavarz, S. Shojaee, M. Mohebali and M. Rezaeian (2012). "IgG
avidity ELISA test for diagnosis of acute toxoplasmosis in humans." Korean J Parasitol
50(2): 99-102.
Reading, P. C., D. L. Pickett, M. D. Tate, P. G. Whitney, E. R. Job and A. G. Brooks
(2009). "Loss of a single N-linked glycan from the hemagglutinin of influenza virus is
associated with resistance to collectins and increased virulence in mice." Respir Res 10:
117.
Reading, P. C., M. D. Tate, D. L. Pickett and A. G. Brooks (2007). "Glycosylation as a
target for recognition of influenza viruses by the innate immune system." Adv Exp Med
Biol 598: 279-292.
Skowronski, D. M., Y. Li, S. A. Tweed, T. W. Tam, M. Petric, S. T. David, F. Marra, N.
Bastien, S. W. Lee, M. Krajden and R. C. Brunham (2007). "Protective measures and
human antibody response during an avian influenza H7N3 outbreak in poultry in British
Columbia, Canada." CMAJ 176(1): 47-53.
Stevens, J., L. M. Chen, P. J. Carney, R. Garten, A. Foust, J. Le, B. A. Pokorny, R.
Manojkumar, J. Silverman, R. Devis, K. Rhea, X. Xu, D. J. Bucher, J. C. Paulson, N. J.
Cox, A. Klimov and R. O. Donis (2010). "Receptor specificity of influenza A H3N2
viruses isolated in mammalian cells and embryonated chicken eggs." J Virol 84(16):
8287-8299.
Sun, X., A. Jayaraman, P. Maniprasad, R. Raman, K. V. Houser, C. Pappas, H. Zeng,
R. Sasisekharan, J. M. Katz and T. M. Tumpey (2013). "N-linked glycosylation of the
hemagglutinin protein influences virulence and antigenicity of the 1918 pandemic and
seasonal H1N1 influenza A viruses." J Virol 87(15): 8756-8766.
Suzuki, Y. (2011). "Positive selection for gains of N-linked glycosylation sites in
hemagglutinin during evolution of H3N2 human influenza A virus." Genes Genet Syst
86(5): 287-294.
111

Talaat, K. R., R. A. Karron, C. J. Luke, B. Thumar, B. A. McMahon, G. L. Chen, E. W.
Lamirande, H. Jin, K. L. Coelingh, G. Kemble and K. Subbarao (2011). "An open label
Phase I trial of a live attenuated H6N1 influenza virus vaccine in healthy adults."
Vaccine 29(17): 3144-3148.
Tate, M. D., A. G. Brooks and P. C. Reading (2011). "Inhibition of lectin-mediated innate
host defences in vivo modulates disease severity during influenza virus infection."
Immunol Cell Biol 89(3): 482-491.
Tate, M. D., E. R. Job, A. G. Brooks and P. C. Reading (2011). "Glycosylation of the
hemagglutinin modulates the sensitivity of H3N2 influenza viruses to innate proteins in
airway secretions and virulence in mice." Virology 413(1): 84-92.
Taubenberger, J. K. and J. C. Kash (2010). "Influenza virus evolution, host adaptation,
and pandemic formation." Cell Host Microbe 7(6): 440-451.
Teng, G. and F. N. Papavasiliou (2007). "Immunoglobulin somatic hypermutation." Annu
Rev Genet 41: 107-120.
Treanor, J. J., B. E. Wilkinson, F. Masseoud, J. Hu-Primmer, R. Battaglia, D. O'Brien,
M. Wolff, G. Rabinovich, W. Blackwelder and J. M. Katz (2001). "Safety and
immunogenicity of a recombinant hemagglutinin vaccine for H5 influenza in humans."
Vaccine 19(13-14): 1732-1737.
Vigerust, D. J., K. B. Ulett, K. L. Boyd, J. Madsen, S. Hawgood and J. A. McCullers
(2007). "N-linked glycosylation attenuates H3N2 influenza viruses." J Virol 81(16): 85938600.
Wang, C. C., J. R. Chen, Y. C. Tseng, C. H. Hsu, Y. F. Hung, S. W. Chen, C. M. Chen,
K. H. Khoo, T. J. Cheng, Y. S. Cheng, J. T. Jan, C. Y. Wu, C. Ma and C. H. Wong
(2009). "Glycans on influenza hemagglutinin affect receptor binding and immune
response." Proc Natl Acad Sci U S A 106(43): 18137-18142.
Wanzeck, K., K. L. Boyd and J. A. McCullers (2011). "Glycan shielding of the influenza
virus hemagglutinin contributes to immunopathology in mice." Am J Respir Crit Care
Med 183(6): 767-773.
World Health Organization (2011). WHO Global Influenza Surveillance Network: Manual
for the Laboratory diagnosis and virological surveillance of influenza, WHO Press,
World Health Organization, 20 Avenue Appia, 1211 Geneva 27, Switzerland.
World Health Organization (2013). "Number of confirmed human cases of avian
influenza A(H7N9) reported to WHO."
Retrieved Sept. 12, 2013, from
http://www.who.int/influenza/human_animal_interface/influenza_h7n9/09_ReportWebH7
N9Number.pdf.

112

Yang, H., P. J. Carney, R. O. Donis and J. Stevens (2012). "Structure and receptor
complexes of the hemagglutinin from a highly pathogenic H7N7 influenza virus." J Virol
86(16): 8645-8652.
Yang, H., L. M. Chen, P. J. Carney, R. O. Donis and J. Stevens (2010). "Structures of
receptor complexes of a North American H7N2 influenza hemagglutinin with a loop
deletion in the receptor binding site." PLoS Pathog 6(9): e1001081.

113

CHAPTER 4
OVERALL DISCUSSION AND SUMMARY
The Influenza Division at the CDC is tasked with influenza pandemic preparedness.
The history of the 20th and 21st centuries teach us that influenza pandemics appear at
unpredictable intervals, and arise from avian influenza viruses, either directly or filtered
through swine. However, a vast number of avian influenza viruses exist, and resources
are finite, so a main goal of the CDC is to identify avian influenza viruses that are
particular concerns for pandemics. A formal risk assessment process exists for this, but
this risk assessment depends heavily on research and evaluation of specific viral gene
products, to determine their potential contribution to adaptation to mammalian infection
and transmission. After risk assessment is performed, vaccines against pre-pandemic
influenza viruses are generated. However, the utility of these vaccines depends on their
immunogenicity. In my dissertation research, I described my contribution to this process
at the CDC. We identified two significant gaps in pandemic preparedness against
HPAIVs: One in risk assessment, the other in vaccine effectiveness.
For risk assessment, I analyzed the significance of historically rare but emerging PB1F2 truncations in H5N1 viruses. For vaccine effectiveness, I evaluated the
immunogenicity of H7 hemagglutinins (HA) in comparison with those of seasonal
viruses. We performed a comprehensive analysis of prevalence, phylogeny, and
evolutionary dynamics of PB1-F2 truncations in all the available influenza sequences of
human, swine, and avian isolates. Our database analyses results are largely in
accordance with the previously published studies which have found that the large

114

fraction of mammalian influenza viruses have truncated PB1-F2 while the prevalence of
such truncation are rare in avian influenza viruses. However, critically, we demonstrated
that PB1-F2 truncations are increasingly common in the H5N1 subtype, which is of
particular pandemic concern. We here report for the first time, the emergence and
evolution of PB1-F2 truncations independently in multiple different clades of H5N1
viruses. In the previously published study (2006) only 0.3% (1 out of a total of 328)
H5N1 viruses had truncated PB1-F2. Our database analyses found that 152 out of a
total of 2002 (7.6%) H5N1 viruses had truncated PB1-F2, a 25 fold increase from the
previously published study.
To achieve our first aim of risk assessment, we studied the pathogenicity of H5N1 PB1F2 truncations. We found that an H5N1 virus lacking expression of PB1-F2, due to
mutation of the initiating ATG, has much higher virulence in mice than a virus with fulllength PB1-F2. Since these different PB1-F2 did not show in vitro differences in
apoptosis or modulation of IFN (functions previously ascribed to PB1-F2) we propose
that PB1-F2 deletion may lead to altered virulence through as-yet unknown functions,
perhaps related to expression of a C-terminal fragment of PB1-F2 from downstream
ATG. We suggest that PB1-F2 truncations and ATG mutations be monitored as part of
ongoing avian influenza virus surveillance and risk assessment, and recommend that
the ATG deletion in particular be considered a potential risk factor for extreme virulence
in mammals.
In our second project under the overall aim of pandemic preparedness, we evaluated
the immunogenicity of H7 hemagglutinins in comparison with seasonal viruses. From
previously published serosurveillance studies and H7 vaccine clinical trials, it is known
115

that the H7 viruses and vaccines are poorly immunogenic. However, no one has
evaluated the immunogenicity of H7 viruses and compared it in parallel with seasonal
viruses. Furthermore, it is not clear if the differences in response is due to intrinsic
aspects of the HA, to pre-existing immunity in humans, or to some aspect of the virus or
the vaccine. We found that H7 viruses, of both the Eurasian and North American
lineages, are significantly less immunogenic than human seasonal virus HAs from the
H3N2 and H1N1pdm09 subtypes. This finding raises important considerations about
production of pre-pandemic vaccines against H7-subtype viruses.

In particular, it

seems likely that to achieve adequate protective immunity vaccination will require large
amounts of H7 HA, or the presence of an adjuvant. This means that US government
influenza vaccine stocks must either be much larger than previously anticipated, or that
vaccine strategies must be revised to account for the potentially lower amounts of
vaccine available.
These findings also point toward possible solutions for the concern. If the H7 HA is
intrinsically less immunogenic, then making different virus seed stocks is unlikely to
solve the problem; rather it is critical to identify the molecular causes of poor
immunogenicity and to identify ways of overcoming them. One possibility is that,
although antibodies are being produced, they are either low affinity antibodies, or they
target non-neutralizing regions of the HA, such as the stalk. Further experiments can
identify specific antigenic sites in H7 HAs, and in combination with new or different
serological assays may be used to identify protective titers more efficiently. As well, it
may be possible to genetically or chemically modify the H7 HA in order to overcome the
lack of intrinsic immunogenicity. For example, if the lack of glycosylation of avian
116

influenza HA results in poor immunogenicity, then new glycosylation sites may be
engineered into vaccine stocks to increase immunogenicity.
In summary, my dissertation research has helped advance the global pandemic
preparedness for avian influenza virus. Future work will extend these advances by
adding to the formal risk assessment process, and by enhancing pre-pandemic vaccine
stock generation.

117