Understanding the adaptive immune responses against newly emerged viruses, SARS coronavirus and avian h5n1 influenza a virus 3

The aim of the project was to understand the adaptive immune response against two newly emerged viruses, SARS-CoV and the H5N1 influenza A virus.. It induced unfolded protein response in

1

CHAPTER 1: INTRODUCTION

1.1 Overview of emerging infectious diseases

Despite advances in medical research and treatments during the 20thcentury, infectious diseases remain the leading causes of death worldwide The elimination of smallpox in 1977 was a great achievement in the fight against infectious diseases However, infectious diseases continue to persist and caused great losses both in the social and economic aspects This is due to the emergence of new infectious diseases, re-emergence of old infectious diseases, and persistence of intractable infectious diseases Emerging infections can be defined as infections that have newly appeared in the population, or have existed but are rapidly increasing in incidence or geographic range (Morse & Schluederberg, 1990) Examples of recent emerging diseases include Acquired Immunodeficiency Syndrome (AIDS) cause by human immunodeficiency virus (HIV), hantavirus pulmonary

syndrome, Lyme disease and foodborne infection by O157:H7 Escherichia

coli

There are several factors that contribute to the emergence of new diseases These include ecologic changes (e.g deforestation), changes in human demographics and behavior (e.g urban migration and intravenous drug use), and increased international air travel Human populations were brought closer and more frequently to the source of the pathogens The outbreak of the new infection, severe acute respiratory syndrome (SARS) may be the result of increased use of exotic animals as food source Close contact with palm civet cats and raccoon dogs in China was identified as a potential route by which the

2

SARS coronavirus (SARS-CoV) was transmitted from animal to human (Guan

et al., 2003; Wang et al., 2005) In addition to the identification of new

human viruses, “old infectious pathogens” are also re-emerging Microbial adaptation and natural genetic recombination and mutation result in new strains of known pathogens, which are not recognized by the human immune system One example is the emergence of H5N1 influenza A virus strain

which has caused human infection since 1997 (Subbarao et al., 1998)

The aim of the project was to understand the adaptive immune response against two newly emerged viruses, SARS-CoV and the H5N1 influenza A virus For SARS-CoV, the focus was mainly to investigate the T cell response against the unique accessory 3a protein The study was also extended to T cell response against the structural nucleocapsid (N) protein in SARS recovered individuals The main objective for the H5N1 work was to demonstrate the use of the recombinant baculovirus-expressed hemagglutinin (HA) protein for use as vaccine and as a tool to generate neutralizing antibodies Further characterization of a potent monoclonal antibody (mAb) against the HA protein was also performed

1.2 Severe Acute Respiratory Syndrome (SARS)

1.2.1 Epidemiology of severe acute respiratory syndrome (SARS)

SARS first emerged in Guangdong, China in late 2002 (Zhong et al.,

2003) By March 2003, the disease spread to Hong Kong, and then to

Vietnam, Singapore and Canada (Lee et al., 2003; Poutanen et al., 2003; Tsang et al., 2003) Those infected by SARS were mainly healthcare workers

or household members who had cared for patients with severe respiratory

3

illness Contact tracing finally indicated the index case was a healthcare worker from Guangdong province who visited Hong Kong and transmitted the virus to several other guests who, further contributed to global dissemination

of the disease (Ksiazek et al., 2003) A novel coronavirus SARS coronavirus (SARS-CoV) was identified as the causative agent (Drosten et al., 2003; Ksiazek et al., 2003; Peiris et al., 2003b) The SARS-CoV went on to infect

more than 8000 people in 29 countries across 5 continents with 774 deaths reported by World Health Organization (WHO) (WHO, 2003b) The SARS epidemic was officially controlled by July 2003 with strict isolation of patients

The main route of transmission seems to be airborne droplets from

infected patients (Booth et al., 2005; Hui & Chan, 2010; Yu et al., 2004b) Blood and fecal-oral transmission has also been suggested (Poon et al., 2003)

There are no known vectors for coronaviruses but epidemiological evidence demonstrated that early cases of SARS were linked to exposure to wild game

animals in the live wet markets in Guangdong province (Guan et al., 2003) In the nasal and fecal swabs from masked palm civets (Paguma larvata) and raccoon dogs (Nyctereutes procyonoides) in the wet markets, SARS-like

viruses which are genetically and antigenically related to the human CoV were detected using reverse transcriptase polymerase chain reaction (RT-PCR) and electron microscopy of viral particles from infected cells The SARS-like viruses isolated from the animals were shown to have more than 99% homology with the human SARS-CoV Interestingly, high seroprevalence for the animal SARS-CoV antibodies were found in animal traders working with these live animals in the wet markets, although they did

SARS-4

not have a history of SARS-like disease (MMWR, 2003) These observations seem to suggest that the live animal market probably is the site for the animal SARS-CoV to amplify and allow interspecies transfer of the animal virus to the humans However, it is not clear whether these animals are the natural reservoirs of the SARS-CoV in the wild In 2005, reports from several groups identified a virus that was genetically closely related to human SARS-CoV in the Chinese horseshoe bats, suggesting that they may be the natural source of SARS-CoV although attempts to isolate the virus from bats have not been

successful (Lau et al., 2005; Li et al., 2005) Bats are known to be the

reservoir hosts of several zoonotic viruses, including the Hendra and Nipah

paramyxoviruses that have recently emerged in Australia and East Asia (Chua

et al., 2000; Murray et al., 1995) Finally, molecular epidemiology showed

that at least two strains of SARS-CoV have been found in patients in Hong

Kong (Guan et al., 2004a) This suggests that the virus had jumped from

animal source to human on two separate occasions This indicates the outbreak of SARS would have been inevitable and the potential of re-emergence is also high

1.2.2 Genome Organization

Coronaviruses are a diverse group of large, enveloped stranded RNA viruses They belong to the order Nilovirales, family Coronaviridae, genus Coronavirus and cause respiratory and enteric diseases

positive-in humans and other animals Three serologically distpositive-inct groups of coronaviruses have been identified Group I and II contain mammalian viruses, whereas group III contains only avian viruses After its discovery, the

5

SARS-CoV was initially placed in a new group (IV) of coronavirus as the sequence is distinct from those previously reported in animals and humans

(Marra et al., 2003; Rota et al., 2003) However, examining sequences within

regions of ORF 1a of SARS-CoV showed domains that are unique to the group II coronaviruses, suggesting that SARS-CoV may be more directly

related to group II viruses (Snijder et al., 2003) Group II coronaviruses

include the bovine coronavirus, human OC43 virus and murine hepatitis coronavirus (MHV) More evidence demonstrated that the 3’UTR of SARS-CoV could substitute functionally for that of MHV, but not with 3’ UTR from

group I coronavirus (Goebel et al., 2004) Thus some research groups

suggested that SARS-CoV belongs within group II coronaviruses, in a subgroup IIb

The genome of SARS-CoV is approximately 30 kb, with

polyadenylated positive-stranded RNA (Marra et al., 2003; Rota et al., 2003)

The genomic organization is typically of a coronavirus, with the characteristic gene order, with the first two open reading frames (1a and 1b) encoding the viral replicase and the downstream mRNAs encoding structural proteins spike (S), envelope (E), membrane (M) and nucleocaspid (N) However, the gene encoding hemagglutinin-esterase found in group II and some group III

coronaviruses was not found in SARS-CoV (Rota et al., 2003) The RNA is

packaged by the N protein into a helical nucleocapsid, with the S protein forming morphologically characteristic projections on the virion surface

6

1.2.2.1 Replicase genes

The viral replicase genes (ORF 1a and 1b) translate into two

polyproteins, pp1a (486 kDa) and pp1ab (790 kDa) (Thiel et al., 2003)

Expression of the pp1ab is predicted to involve a ribosomal frameshifting into the -1 frame just upstream of the ORF1a translation termination codon The pp1a and pp1ab polyproteins undergo proteolytic processing by viral cysteine proteinases to yield the functional components of the membrane-bound

replicase complex and a group of 16 non-structural proteins (nsp) (Ziebuhr et

al., 2000) SARS-CoV uses only two proteinases, PL2pro, a papain-like cysteine proteinase (nsp3) and 3CLpro, a 3C-like proteinase (nsp5) , in contrast

to most coronaviruses that use three proteinases (Gao et al., 2003; Rota et al., 2003; Snijder et al., 2003) The replicase complex, which includes an RNA-

dependent RNA polymerase and RNA helicase mediates both genome replication and transcription of a “nested” set of subgenomic mRNAs The putative functions of the non-structural proteins are summarized in Figure 1.1 The functions have been shown by biochemical assays or predicted based on their functional domains or structural similarities to other proteins while others

remain to be further characterized [reviewed by (Cheng et al., 2007)]

7

Fig 1.1 Summary of the SARS-CoV genome organization and viral protein expression Replicase (ORF 1a and 1b), constituting the first 2/3 of the genome, which translates into two polyproteins, pp1a and pp1ab The putative functions of each of the nsps are shown in the text boxes Open reading frames (ORFs) in the remaining 1/3 of the genome are translated from eight subgenomic mRNAs Four of the ORFs encode the structural proteins, spike (S), membrane (M), and envelope (E) and nucleocapsid (N) Another eight unique ORFs encode accessory proteins (3a, 3b,

6, 7a, 7b, 8a, 8b and 9b), which have no significant sequence homology to viral proteins of other coronaviruses This figure was adapted and modified from Cheng,

V C C et al 2007 Clin Microbiol Rev 20(4): 660-694

P

Expression promoted degradation

of host endogenous mRNAs,

which may inhibit host protein

synthesis and prevented

endogenous IFNβ mRNA

accumulation

Deletion attenuates viral

growth and RNA synthesis

Papain-like protease 2;

ADP-ribose 1-phosphatase

Not known 3C-like protease

Three-dimensional crystal structure of a dimer which binds viral RNA and interacts with nsp8

Crystal structure suggests a nucleic acid binding function within a larger RNA binding protein complex for viral gene transcription and replication

RNA-dependent RNA polymerase

Helicase (dNTPase and RNA 5’-triphosphatase activities)

3’→5’-exoribonuclease; supplements the endoribonuclease activity in the replication of the giant RNA genome

Uridylate-specific endoribonuclease; Involved in the coronavirus replication cycle Putative 2’-O-ribose methyltransferase

8

1.2.2.2 Structural and accessory proteins

The subgenomic mRNAs encode the structural proteins, S, E, M and

N, and the set of accessory proteins (namely ORF 3a, 3b, 6, 7a, 7b, 8a, 8b and 9b) The surface S protein is involved in the attachment and entry of the host cell The N together with M and E are involved in the assembly of the virion The accessory proteins have no significant homology to viral proteins of other coronaviruses These proteins are dispensable for virus replication in cell

culture while some appear to contribute to viral pathogenesis (Narayanan et

al., 2008; Tan et al., 2006) The focus of our study is the largest SARS-CoV

accessory protein, 3a It is also known as U274 or X1 and this protein will be discussed in detail in section 1.2.2.3 The characteristic, functions, and/or putative roles of the four structural and seven other accessory proteins of the SARS-CoV are outlined in Table 1.1 and Table 1.2 respectively

A type I integral membrane glycoprotein which

is N-glycosylated, trimerized in endoplasmic

reticulum (ER) (Bosch et al., 2003; de Groot et

al., 1987; Rota et al., 2003)

It is divided into 2 subdomains of similar size, S1 and S2 with distinct functions S1 domain forms the globular portion of the spike, mediating binding to host cell receptor,

angiotensin-converting enzyme 2 (ACE2) (Li

et al., 2003) The receptor binding domain

(RBD) is localized to amino acids 318 to 510

The S2 ectodomain contains two regions with a

4, 3 hydrophobic (heptad) repeat, HR1 and HR2 and a putative, internal fusion peptide

(Bosch et al., 2004; Sainz et al., 2005)

Biochemical studies have shown that peptides corresponding to the HR1 and HR2 of the SARS- CoV S protein can associate into an anti-parallel six- helix bundle with structural features typical of class I

fusion proteins (Ingallinella et al., 2004; Liu et al., 2004; Tripet et al., 2004) This HR1-HR2 structure

brings the fusion peptide in close proximity to the

transmembrane domain (Bosch et al., 2004), leading

to the fusion of the viral and cellular membrane, and consequently the viral entry

The S protein was known to be responsible for inducing host immune responses and is the primary target for viral neutralizing

antibodies (Keng et al., 2005; Zhou et al.,

2004)

The functional region of S protein from amino acids 324-688 can induce the release

of IL-8 in lung cells (Stevens et al., 2006)

It induced unfolded protein response in cultured cells as SARS-CoV with accumulation of S protein in the ER, may

modulate viral replication (Yamada et al.,

2006)

10

Table 1.1 Continued Protein (No of

terminal region (Shen et al., 2003)

It was demonstrated that the E protein does form ion channels, which are are more selective for monovalent cations than monovalent anions

(Wilson et al., 2004)

It was shown to be important for viral assembly as

demonstrated by the formation of VLPs (Ho et al., 2004; Nal et al., 2005; Vennema et al., 1996)

It induced apoptosis in transfected Jurkat T cells in the absence of growth factors A novel BH3-like region located in the C-terminal cytosolic domain of SARS-CoV E protein can bind Bcl-xL, whose overexpression can

antagonize apoptosis (Durrer et al., 1996)

Membrane

protein (221)

The M protein contains a long cytoplasmic tail, 3 hydrophobic transmembrane domains, and a short glycosylated N- terminal ectodomain It has been shown to

be N-glycosylated at asparagines residue at

position 4 (Nal et al., 2005; Voss et al.,

2006)

Functional analysis showed that the N-terminal region of the M protein, comprising of the 3 transmembrane domains, is sufficient to mediate accumulation of M in the Golgi complex and recruit the S protein to the sites of viral assembly and budding in the ER Golgi-intermediate

compartment (Voss et al., 2009)

M protein induced apoptosis in HEK293T cells, which could be suppressed by caspase

inhibitor (Tsurudome et al., 1992)

11

Table 1.1 Continued Protein (No of

The N protein is highly charged basic protein

of 422 amino acids (~48 kDa) with seven successive hydrophobic residues near the middle of the protein At the N terminal, it contains a highly conserved motif

[FYYLGTGP] which is observed in all

coronavirus N proteins (Rota et al., 2003) It

is reported to be abundantly found in the cytoplasm and nucleus of SARS-CoV infected cells

Properties of the N protein include self-dimerization

(He et al., 2004), RNA-binding capabilities (Huang

et al., 2004), cleavage by caspase 3 (Ying et al.,

2004) and activation of signal transduction pathways

The N protein was shown to induce apoptosis in COS cells in the absence of

growth factors (Surjit et al., 2004) It

antagonized interferon (IFN) by inhibiting synthesis of IFNβ (Weber et al., 1994)

Nuclear factor kappa B (NF- κB) activation was observed in Vero E6 cells expressing the

N protein in dose dependent manner (Zhou

et al., 2004) It may cause inflammation of

the lungs by activating COX-2 gene resulting

in inflammation through multiple COX-2

signaling cascades (Keng et al., 2005)

Most of the sera obtained from convalescent SARS patients have antibodies against N

(Leung et al., 2004; Shi et al., 2003; Tan et

al., 2004a) In addition, it was reported that

the N protein can induce specific T cell

responses (Kim et al., 2004)

Protein characteristic [ref(s)] Protein’s function or putative function(s) [ref(s)] Effect on cellular response of host [ref(s)]

ORF 3b (154) 3b was expressed from an ORF which

overlaps the ORF3a and E by using an

internal ribosomal entry site (Rota et al.,

2003)

One study revealed 3b initially accumulates in the nucleus and subsequently translocates to the

mitochondria (Freundt et al., 2009) However, the

exact mechanism in which 3b contributes to SARS-CoV pathogenesis is still not known

Overexpression of 3b has been shown to induce cell cycle arrest at the G0/G1 phase and

apoptosis (Yuan et al., 2006a) It is also suggested to be a type I IFN antagonist (Freundt

et al., 2009; Kopecky-Bromberg et al., 2007)

ORF 6 (63) ORF 6 protein is a 63 amino acid,

membrane associated protein (Geng et al., 2005; Pewe et al., 2005) It is expressed in

virus-infected Vero E6 cells as well as in the lungs and intestine specimen of SARS patients It is mainly localized in the ER and Golgi compartments It is shown to

incorporated into SARS-CoV virus particles although the protein is also secreted from infected cells or cells transiently expressing

ORF6 (Huang et al., 2007a)

Several studies have indicated that ORF 6 might

be involved in viral replication and play a role in SARS pathogenesis It interacts with SARS-CoV nsp 8, which was proposed to be a low-fidelity primase producing short RNA primers utilized by the primer-dependent nsp 12 for initiation of viral

RNA replication (Kumar et al., 2007) It also

partially colocalized with nsp3, a marker for virus replication complexes and induced membranous structures similar to vesicles involved in virus

replication (Zhou et al., 2010)

In the mouse study done with attenuated MHV expressing SARS-CoV ORF6 showed the virus replicating to higher titers and exhibited higher

virulence in mice (Zhao et al., 2009a) This

could be explained by studies revealed that ORF

6 is a type I IFN antagonist, preventing the nuclear translocation of signal transducer and

activator of transcription (STAT) 1 (Frieman et

al., 2008; Kopecky-Bromberg et al., 2007) A

conformation-dependent motif involving both the C and N terminal of the ORF6 was shown to

be important for impeding nuclear import

(Hussain & Gallagher, 2010; Zhou et al., 2010)

13

Table 1.2 Continued Protein (No of

terminal tail (Nelson et al., 2005) It has an

ER retrieval motif (KRKTE), located at the C-terminus, important for transport of proteins back to the ER and mediates the recycling of 7a between the ER and Golgi

apparatus (Fielding et al., 2004)

The 7b protein is a 44 amino-acid integral membrane protein expressed in SARS-CoV infected cells and its transmembrane domain

is essential for its Golgi compartment

localization (Schaecher et al., 2008)

7a also interacts with 3a, which interacts with M,

E, S suggesting that they form complexes (Tan et

al., 2004b) Huang et al showed that in addition

to 3a, 7a protein was also identified as a CoV structural protein when co-expressed with

SAR-the oSAR-ther structural proteins (Huang et al., 2006)

However interactions between the proteins seem

to be non-essential for the incorporation of 7a protein into VLPs

Both 7a and 7b proteins seems to be dispensable for virus replication when SARS-CoV mutant lacking the 7a and 7b genes can be replicated

efficiently in cell culture and mice (Yount et al.,

2005)

Some biological effects of 7a include induction of apoptosis in various cell lines through caspase- dependent pathway and interaction with Bcl-X L

protein (Schaecher et al., 2007; Tan et al., 2007);

inhibition of cellular protein synthesis; activation

of p38 mitogen-activated protein kinase (MAPK)

(Kopecky-Bromberg et al., 2006); and suppression

of cell cycle progression at the G0/G1 phase (Yuan

et al., 2006b) It is also a potent suppressor of

host RNA silencing mechanism during infection

(Karjee et al., 2010)

Other studies have shown that expressed 7a protein interacts with small glutamine-rich tetratricopeptide repeat-containing protein

(Fielding et al., 2006) and human Ap4A-hyrolase (Vasilenko et al., 2010) with biological

significance needed to be elucidated

It has been suggested that 7b is a potential attenuating factor in the SARS-CoV genome, with

enhanced viral replication in hamsters (Pfefferle et

al., 2009)

14

Table 1.2 Continued Protein (No of

Accessory proteins 8a and 8b are distinct from

ORF 8 (122 amino-acids) in conformation (Keng

et al., 2006) Epidemiology studies showed that

most human isolates of SARS-CoV had a 29 nucleotide deletion in ORF 8, resulting in two discrete ORFs 8a and 8b The early human and

animal isolates contained only one ORF 8 (Guan

et al., 2003) This loss of 29-nt might be an

evolutionary adaptation of the virus to infect humans Insertion of this 29-nt sequence using reverse genetics did not have much impact on virus growth and RNA replication in cell culture

(Yount et al., 2005) Thus it probably does not

play any role in pathogenesis

8a protein can interact with S protein, while 8b interacts with M, E, 3a and 7a On the other hand, ORF 8 interacts with S, 3a and 7a

It was shown that 8a not only enhances viral replication but also induces apoptosis through a mitochondrion-

dependent pathway (Chen et al., 2007)

The expression of 8b significantly regulated the E protein level, but not its

down-mRNA level (Keng et al., 2006)

However, the biological significance of these proteins is not known

ORF 9b (98) The 9b protein is encoded by an internal ORF

within the N gene and is translated via a leaky ribosomal scanning mechanism into a 98 amino

acid protein (Xu et al., 2009) The expression of

this protein is demonstrated in SARS-CoV

infected cells and in clinical specimens (Chan et

al., 2005b)

Crystal structure of 9b revealed a novel dimeric like structure with amphipathic surface and a central hydrophobic cavity, which binds lipid molecules that probably allow it to associate with intracellular

tent-vesicles (Meier et al., 2006) Thus, 9b has been

suggested to contribute to virus assembly but the precise function still need to be further investigated

There has been evidence showing the presence of 9b

in VLPs and purified virions, indicating that it is also

a structural protein (Xu et al., 2009)

Antibodies against 9b have been detected

in serum of SARS patients (Qiu et al.,

2005)

15

1.2.2.3 Accessory protein, ORF 3a

3a, also known as ORF3, X1 or U274, is the largest of these accessory proteins with 274 amino acids It is an O-glycosylated protein It is expressed from subgenomic RNA3, which contains the 3a and 3b ORFs

(Marra et al., 2003; Rota et al., 2003) The topology of 3a was suggested to

be its N terminus facing the extracellular matrix, followed by three

transmembrane domains and its C terminus facing the cytoplasm (Tan et al., 2004b; Yu et al., 2004a; Zeng et al., 2004) 3a can be detected in alveolar

lining pneumocytes and some intra-alveolar cells of lung specimens of SARS patients as well as in SARS-infected cells It has been reported to localize to the Golgi apparatus, the plasma membrane and intracellular vesicles of

unknown origin (Yu et al., 2004a; Yuan et al., 2005) Transportation of 3a to

the cell surface depends on the juxtaposition of two sorting motifs, YxxΦ (where x is any amino acid and Φ is an amino acid with a hydrophobic side chain) and ExD (diacidic) motif, and it can subsequently undergoes

endocytosis (Tan et al., 2004b) Zeng et al reported disulfide-linked

complexes of S and 3a in the medium of SARS-CoV infected cells, suggesting that 3a was secreted together with S through the formation of virus particles

(Zeng et al., 2004) It was later confirmed that 3a indeed was a novel structural component of the SARS-CoV virion (Ito et al., 2005; Shen et al.,

2005) However, 3a protein is dispensible for VLP and SARS-CoV assembly

(Hsieh et al., 2005; Mortola & Roy, 2004) Studies also revealed that 3a can

interact specifically with structural proteins, M and E and accessory protein 7a

(Tan et al., 2004b; Yuan et al., 2005)

16

The cysteine-rich domain of 3a was known to be responsible for homo- and hetero-dimerization, which is crucial for its ion channel activity

(Lu et al., 2006b) Although the functional significance of the 3a’s ion

channel activity is still not well defined, it has been shown to be linked to its

pro-apoptotic function (Oostra et al., 2006) Two early studies have reported 3a protein to be pro-apoptotic (Law et al., 2005b; Wong et al., 2005) It was shown that 3a triggers apoptosis via capase-8 dependent pathway (Law et al.,

2005b) More recently, Padhan et al reported that 3a activates caspase-9 and other elements of the intrinsic pathway including cytochrome c release, Bax oligomerization, p53 upregulation and p38 MAP kinase (MAPK) activation

(Voss et al., 2009) This group later showed that the 3a protein elicits

apoptotic conditions by specific activation of ER stress and PERK signaling

pathway (Voss et al., 2006) They further proposed a potential role of 3a in

attenuating IFN responses and innate immunity through the induction of degradation of IFN alpha-receptor subunit 1 (IFNAR1) However, further studies are needed in animal model to confirm the extent of these effects in SARS-CoV pathogenesis

3a seems to play an important immunological role in SARS-CoV infection A strong and potentially protective humoral response is directed against the amino terminus of 3a protein in SARS patients Sera of SARS

convalescent patients showed immunoreactivity to 3a (Zhong et al., 2006)

48.8% of the recovered SARS patient had antibodies against 3a as compared

to only 7.4% of those who died from the disease Amino acids 15-28 in the ectodomain of 3a protein was able to induce neutralizing antibodies and

17

inhibit SARS-CoV propagation in Vero E6 cells (Akerstrom et al., 2006) 3a

was also reported to up-regulate the expression and secretion of fibrinogen

(Tan et al., 2005) and augment interleukin 8 (IL-8) and NF-κB promoter activities (Kanzawa et al., 2006), possibly through its RNA-binding activity of its C-terminal domain (Nal et al., 2005) These observations seem to suggest

that 3a might contribute to SARS pathogenesis by enhancing cytokine production

Some SARS patients were also reported to have bone problems during the convalesecent phase One study demonstrated that 3a does promote

osteoclastogenesis by direct and indirect mechanisms (Obitsu et al., 2009)

Other cellular host responses induced by 3a include perturbing Arf1-mediated vesicle trafficking, leading to vesicle formation and Golgi fragmentation

(Freundt et al., 2010); and inhibiting cell proliferation caused by a block in the G1 phase of cell cycle (Yuan et al., 2007)

18

1.3 H5N1 Influenza A

1.3.1 Epidemiology of H5N1 influenza A virus

Influenza A viruses can be divided into high and low pathogenicity (Swayne & Suarez, 2000) The low pathogenic strain of virus cause mild respiratory disease, a decrease in egg production, and/or depression in chickens In contrast, the highly pathogenic avian influenza (HPAI) viruses cause significant mortality in chickens and more recently in ducks, geese and wild waterfowl HPAI H5N1 influenza viruses were first isolated from sick

geese in Guangdong, China in 1996 (Xu et al., 1999) In 1997, these viruses

caused outbreaks in chickens in Hong Kong and were transmitted to humans

(Claas et al., 1998; Subbarao et al., 1998) It resulted in 18 infected cases with

6 deaths Infections were acquired by humans directly from the chickens Although the HPAI H5N1 virus was successfully eradicated by slaughtering all the poultry in Hong Kong, the donor of the hemagglutinin (HA) gene of the

1997 H5N1 strain, A/goose/Guangdong/1/96 continued to circulate in geese in southeastern China From 1997 through 2001, the HA on the various genotypes remained antigenically homogenous, but in 2002 it underwent marked antigenic drift, which rendered the virus highly pathogenic in the

ducks and other aquatic birds (Guan et al., 2004b; Sturm-Ramirez et al.,

2004) In 2003, the H5N1 virus re-emerged in a family in Hong Kong,

causing 2 deaths (Peiris et al., 2004) The strain was found to be antigenically

and molecularly similar to the strain that was pathogenic for ducks and chickens The HPAI H5N1 influenza virus then continued to cause a bird flu epidemic in Southeast Asia and then later spread to more than 60 countries

19

(Neumann et al., 2010) Although the outbreaks of the HPAI H5N1 virus in

these countries were mainly confined to poultry, the virus was transmitted to humans Since 2003, 504 individuals in 15 countries have been infected, with

299 fatal cases, a mortality rate of more than 60% (WHO, 12 August 2010)

Multiple genotypes were isolated from the infected birds, although one

particular genotype Z dominated (Li et al., 2004; Lipatov et al., 2004) It

drifted antigenically and the current circulating strain appears to be antigenically and genetically similar to A/Vietnam/1203/04 (VN04), which

infected some humans in Vietnam and Thailand (Lipatov et al., 2004) All the

HPAI H5N1 HAs possess a series of basic amino acids at the cleavage site RRKKR-), a characteristic of HPAI virus and most of their HA gene belong to

(-the A/goose/Guangdong/1/96 lineage (Hoffmann et al., 2000b; Li et al.,

2004) Based on the HA sequences, HPAI H5N1 viruses are now divided into

10 different clades as shown in Figure 1.2 (WHO, March 2009)

Human infection of the H5N1 virus was most likely acquired directly from the poultry and does not seem to spread efficiently among humans

(Wang et al., 2008a) However, cases of possible human-to-human transmission have been reported (Kandun et al., 2006; Wang et al., 2008a;

WHO, 2010) The continued evolution of the virus or a single reassortment with human influenza strains could render the virus successful in transmission between humans Thus there is a fear of possible H5N1 pandemic

20

Fig 1.2 Diagram showing the phylogenetic tree for the hemagglutinin gene of highly pathogenic avian influenza A (H5N1) viruses The geographic distributions refer to avian isolates, and the tree is based on publicly available sequences This figure was adapted from WHO, Continuing progress towards a unified nomenclature system for the highly pathogenic H5N1 avian influenza viruses (WHO, March 2009)

21

1.3.2 Genome organization

Influenza A is the prototype of the family, Orthomyxoviridae, which

comprises enveloped viruses with segmented negative-sense RNA genome

Based on the antigenicity of the HA and neuraminidase (NA) surface

glycoproteins, influenza A viruses currently form 16 HA (H1-H16) and 9 NA

(N1-N9) subtypes Over the past century, the only viruses that had infected

and circulated in humans are the H1N1, H1N2, H2N2, and H3N2 subtypes

However, recently humans infection are now associated with the H5, H7 and

H9 subtypes

The genome of influenza A virus is composed of eight single-stranded

negative-sense RNA segments which encodes for the 11 genes: HA, NA,

matrix 1 (M1), matrix 2 (M2), nucleoprotein (NP), nonstructural protein 1

(NS1), nonstructural protein 2 (NS2, also known as nuclear export protein,

NEP), polymerase acidic protein (PA), polymerase basic protein 1 (PB1),

polymerase basic protein 2 (PB2) and polymerase basic protein 1-F2

(PB1-F2) The most striking feature of the influenza A virion is the layer of

projections made of 2 distinct glycoproteins, HA and NA Embedded also in

the viral membrane is the M2 ion channel protein The M1 protein, sitting just

underneath the lipid membrane, forms a matrix holding the viral

ribonucleoproteins (vRNPs) The vRNPs are made up of viral RNA wrapped

up around NP and very small amounts of NEP At one end of the vRNPs are

the three polymerase proteins, PA, PB1, PB2, which make up the RNA

polymerase complex (Nayak et al., 2009; Nayak et al., 2004)

22

The properties and functions of the hemagglutinin of the H5N1 virus will be discussed in details in section 1.3.2.2 The characteristics, functions and/or putative roles of the other structural and non-structural proteins are outlined in Table 1.3

23

1.3.2.1 Structural and non-structural proteins

Table 1.3 Summary of structural and non-structural proteins of the H5N1 virus

of the protein is phosphorylated and whether phosphorylation is essential for its function

(Davey et al., 1985)

The primary function of NP is to encapsidate the virus genome for the purposes of RNA transcription, replication and packaging It is suggested to interact with cellular polypeptides, including actin, components of the nuclear import and export apparatus and a nuclear RNA helicase, indicating multiple functions of the protein

(Baudin et al., 1994; Portela & Digard, 2002) Recent

evidence showed that the NP can directly interact with the viral polymerase and mediates the switch from capped-primed viral mRNA synthesis to unprimed viral

RNA replication (Newcomb et al., 2009)

Lysine at position 184 of the H5N1 NP protein seems to induce earlier mortality

in chickens, with increased virus titers and nitric oxide levels and upregualated host immune genes, IFNα, IFNγ, Mx1 and inducible nitric oxide synthetase

II integral membrane protein, with its terminus in the cytoplasm, a transmembrane domain, the stalk, the head with the catalytic

N-active site (Varghese et al., 1983)

The role of NA in the influenza virus life cycle is still unclear NA can catalyze the cleavage of the α–ketosidic linkage between a terminal sialic acid and an adjacent D-

galactose or D-galactosamine (Palese et al., 1974) It is

suggested that NA functions to remove sialic acid from

HA, NA and the cell surface, thus the release of virus from the host cell It may also allow the transport of the virus through the mucin layer in the respiratory tract to the target epithelial cells

Immunization with H5N1 NA was shown to induce high titers of HPAI- neutralizing serum, although with lower response as compared to the HA protein

(Nayak et al., 2010) Anti-N1

cross-protecting antibodies against H5N1 can

be detected in some H1N1 infected

individuals (Frobert et al., 2010)

24

Table 1.3 Continued Protein (No of

been recognized (Lamb et al., 1981)

M1 is a 28 kDa protein and constitutes the most abundant protein in the virion The M2 protein is type III integral membrane protein, with N-terminal extracellular domain, transmembrane domain, and a cytoplasmic tail

Post-translation modifications such as phosphorylation on serine and palmitylation on cysteine can be found in the cytoplasmic

domain (Holsinger et al., 1994) The native

form of the M2 protein is homotetramer, either existing as a pair of disulfide-linked dimers or

disulfide-linked tetramers (Lamb et al., 1985)

The M1 protein underlies the viral lipid envelope and provides rigidity to the membrane of the virion (Martin & Helenius, 1991) It has been

demonstrated that the M1 protein interacts with RNA, although a zinc-binding motif within the M1 protein does not seems to influence binding to RNA The transport the M1 protein into the nucleus is required for the exit of the newly assembled RNPs from the nucleus An interaction between M1 and NS2 protein in purified virions

has also been proposed (Yasuda et al., 1993)

The M2 protein functions as an ion channel that permits ions to enter the virion during the uncoating as well as modulates the pH of intracellular compartments (Pinto & Lamb, 2006)

There has been evidence that the transmembrane domain constitutes the pore of the channel as specific changes in this domain alter the kinetics and ion selectivity of the channel

Amino acids Asp at position 30 and Ala at position 215 in the M1 protein are necessary

for the H5N1 lethality in mice (Fan et al.,

2009a)

Capase motif found in the N terminal of M2 protein has been suggested to play a significant role in virus pathogenicity shown

in chickens (Zhirnov & Syrtzev, 2009)

25

Table 1.3 Continued Protein (No of

be basic and PA is acidic based on their behavior on isoelectric focusing gels (Horisberger, 1980) The polymerase complex is a heterotrimer of subunits of

PA, PB1 and PB2 (~250kDa)

The polymerase complex is involved in both mRNA transcription and viral replication PA subunit has endonuclease and protease activities; involved in vRNA/cRNA promoter binding; interacts with the PB1 subunit The PB1 subunit contains the RNA-dependent RNA polymerase active site and interacts with PA and PB2 The PB2 subunit is responsible for cap binding and contains a bipartite nuclear localization signal at C terminus for nuclear import from the cytoplasm

[Reviewed by (Das et al., 2010)]

Lys at PB2-627 plays a role in the pathogenicity of the influenza virus

(Shinya et al., 2004) It supports efficient

virus replication in mammalian but not avian species It has been found in a substantial number of H5N1 viruses isolated from infected humans

PB1-F2 induces apoptosis, thus it has been proposed to kill host immune cells responding to influenza virus infection

(Chen et al., 2001). The presence of Ser in

place of Asn at position 66 of PB1-F2 conferred high pathogenicity to H5N1

virus in mice (Conenello et al., 2007).

26

Table 1.3 Continued Protein (No of

virus-stranded RNA (Cheng et al., 2009)

The NS2 protein (~11 kDa) is initially thought to be nonstructural but recently it is reported to be associated with M1 protein in

virions (Yasuda et al., 1993)

NS1 protein regulates the nuclear export of mRNA

and inhibits pre-mRNA splicing (Fortes et al.,

1994) It also contains two functional domains, RNA-binding and an effector domain, which

regulates the nuclear export of viral mRNA (Qian et

al., 1994) It helps to regulate the switch from early

to late protein synthesis by retaining the late viral mRNAs in the nucleus until the appropriate time for their expression

NS2/NEP has been shown previously to play a role

in the nucleocytoplasmic export of viral RNPs

(Paragas et al., 2001) Recently it was

demonstrated that the NS2 also changes RNA levels

by specific alteration of the viral transcription and

replication machinery (Robb et al., 2009)

The RNA binding domain of NS1 can inhibit IFN α/β-induced RNase L pathway by sequestering dsRNA away from 2’-5’-oligo synthetase (Min & Krug, 2006)

The effector domain of NS1 binds several cellular proteins: the p85β subunit of phosphatidylinositol-3 kinase (PI3K), leading

to the activation of PI3K signaling; (Hale et

al., 2006); protein kinase (PKR), inhibiting

protein kinase pathway and consequently

protein synthesis and viral replication (Min et

al., 2007); the 30-kDa subunit of cleavage and

polyadenylation specificity factor (CPSF30), inhibiting 3’-end processing of cellular pre- mRNAs, including IFNβ pre-mRNA

(Nemeroff et al., 1998)

27

1.3.2.2 Hemagglutinin protein

Hemagglutinin is the major surface glycoprotein of the influenza A virus and the main target of neutralizing antibodies It is responsible for viral binding to host receptor, enabling entry into host cell through endocytosis, and subsequent membrane fusion The HA spike glycoprotein, is a homotrimer and each monomer is synthesized as a single polypeptide HA0 (~ 76 kDa) that is cleaved by host proteases into two disulfide-linked chains HA1 (~ 47kDa) and

HA2 (~ 29 kDa) (Wiley et al., 1977) It is a type I integral membrane protein,

with ectodomain of 512 residues, a carboxyl-terminal proximal transmembrane domain of 27 residues, and a cytoplasmic tail of 10 residues

(Verhoeyen et al., 1980)

The crystal structure of the hemagglutinin protein of the A/Vietnam/1203/04 (VN04) H5N1 influenza A virus has been resolved

(Stevens et al., 2006) This VN04 virus was isolated from a 10-year old

Vietnamese boy who died from bird flu As expected, the VN04 HA trimer was very similar to other published HA proteins It has a globular head containing the receptor binding domain (RBD) and vestigial esterase domain, and a membrane proximal domain with central α-helical stalk and HA1/HA2 cleavage site The protein was predicted to be post-translationally modified with seven possible glycosylation sites per monomer One of the sites is found

in the cytoplasmic tail and thus unlikely to be glycosylated The HA protein is synthesized as a single-chain precursor in the endoplasmic reticulum, where it

is assembled as a trimer (Skehel & Wiley, 2000) It is then transported to the cell surface via the Golgi network where the HA0 is cleaved by specific host

28

proteases eg tryptase Clara into HA1 and HA2 The cleavage of the HA is required for the virus to be infectious and thus it has been linked to the virulence of H5N1 influenza virus (Garten & Klenk, 1999; Horimoto &

Kawaoka, 1994; Senne et al., 1996) Most HAs of low pathogenic strains

contains cleavage site (Q/E-X-R) [Q, Gln; E, Glu, R, Arg; X, any amino acid] which are cleaved only by certain proteolytic enzymes and thus restricts the infection only to the lung in mammals It is now known that the highly pathogenic H5 and H7 viruses contain a HA cleavage site composed of multiple basic amino acid residues (RRRKKR) that are recognized by ubiquitous proteases, such as furin and PC6, resulting in systemic infection Consequently, the tissue tropism for HPAI H5N1 virus in mammals is no

longer restricted to the lungs, but to other organs including the brain (Maines

et al., 2005) However, a recent study suggested that acquisition of a

polybasic cleavage site is only one necessary step for a low-pathogenic H5N1

strain to becoming a HPAI virus (Bogs et al., 2010) Other additional

virulence determinants found in the HA itself and in other viral proteins may already be present in these non-pathogenic strains

The entry of virus into the host cell mediated by HA involves 2 steps (Wiley & Skehel, 1987): First, it binds to the sialic acid-containing receptor

on the cell surface, allowing the attachment of a virus particle to the cell Then it is responsible for the penetration of the virus into the cell cytoplasm by mediating fusion of the viral membrane with the host endosomal membrane, resulting in the release of the viral RNA into the cytoplasm

29

The RBD is at the membrane distal end (HA1) of each HA monomer (Skehel & Wiley, 2000) The binding site comprises an α-helix (190-helix, HA1 188 to 190) and two loops (130-loop, HA1 134-138, and 220-loop, HA1

221 to 228) The conserved residues (Tyr98, Trp153, His183, Glu190, Leu194) form a pocket which is involved in receptor binding The RBD binds to

receptors containing glycans with terminal sialic acids (Suzuki et al., 2000)

The precise linkage determines the species preference Out of the 16 known avian and mammalian serotypes of HA, only 3 (H1, H2, and H3) have become adapted to the human population The host range which the virus infects is partly determined by the receptor-binding specificity of the HA protein Generally, the avian and equine influenza A viruses bind preferentially to sialic acid that is linked to galactose by an α2,3-linkage (SAα2,3Gal), whereas human influenza virus have higher binding affinity to α2,6-linkage (SAα2,6Gal) (Neumann & Kawaoka, 2006) Most of the HPAI H5N1 viruses have avian-type receptor specificity, therefore they usually infect the chickens, ducks and migratory birds The SAα2,3Gal receptor has been identified on

the ciliated cells of in vitro differentiated human epithelial cells from tracheal/bronchial tissues (Matrosovich et al., 2004), on the epithelial cells of the lower human respiratory tract (Shinya et al., 2006), and to some extent on the upper human respiratory tract (Nicholls et al., 2007) This may explain

reports of human infections with the avian H5N1 viruses Analysis of some recent human H5N1 isolates showed increased affinity for SAα2,6Gal receptor and two independent amino acid changes (Asn182-to-Lys and Gln192-to-Arg) were identified to be responsible for this change in receptor

specificity (Yamada et al., 2006) Thus this seems to suggest the possibility

endosomal and viral membranes (Bullough et al., 1994; Skehel & Wiley,

2000) A schematic diagram in Figure 1.3 shows the hypothetical mechanism for membrane fusion At neutral pH, the fusion peptide found at the N terminal of the HA2 subunit is buried in a pocket of ionizable residues adjacent to the site of precursor cleavage Low/fusion pH releases the peptide from its position and causes it to be inserted into the cellular membrane

(Durrer et al., 1996; Weber et al., 1994) A change in the pH of fusion can help influenza viruses adapt to different host species (Lin et al., 1997) and facilitate resistance to antiviral agents (Steinhauer et al., 1991) Residues

comprising the fusion peptide pocket were shown recently to be important in

regulating the pH of activation of the H5 HA protein (Reed et al., 2009)

Fig 1.3 Hypothetical mechanism for membrane fusion by hemagglutinin protein This figure was adapted from Hughson, F.M 1997 Current Biology 7:R565-69

31

1.4 Viral infection and the host immune system

A viral infection occurs when the body’s surface defense is breached The route of transmission can be through air droplets, blood, fecal-oral or through a vector Once inside the body, the next line of defense will be the host immune system which can be divided into two arms, the innate and the adaptive immunity The innate immune response is non-specific but it allows

a rapid response to invasion It is closely interlinked with adaptive response, which is highly specific, mediated by a specialized group of leukocytes This arm of defense also becomes more powerful following repeated encounters with the same antigen or pathogen

Many organisms such as viruses and some bacteria developed ways to evade the immune system which can consequently result in the development

of disease in the host The clinical presentations of both SARS and H5N1 influenza virus are described below This will be followed by a detailed discussion of the humoral and cellular immune responses against these two viral infections in sections 1.4.2 and 1.4.3

1.4.1 Clinical Features of SARS and H5N1 influenza

Clinical presentations of both SARS and H5N1 influenza infection are similar Symptoms of fever, cough and shortness of breath were seen in H5N1 cases, which are clinically indistinguishable from severe human influenza

(Tran et al., 2004; Yuen et al., 1998) SARS-infected patients also presented initially with fever, myalgia, malaise, and chills or rigor (Lee et al., 2003; Peiris et al., 2003a; Zhao et al., 2003) Cough was common, but shortness of

breath, tachypnea, or pleursy was prominent only later in the course of a

32

SARS infection Abnormal initial chest radiographs were observed in both SARS and H5N1 influenza patients Abnormal chest radiographs observed in H5N1 patients include extensive, usually bilateral infiltration, lobar collapse,

focal consolidation, and air bronchograms (Apisarnthanarak et al., 2004)

Pneumocyte damage was also observed several months after the illness in the surviving H5N1 patients The most common abnormalities seen in chest

radiographs of SARS patients were ground-glass opacifications (Antonio et

al., 2003; Grinblat et al., 2003; Nicolaou et al., 2003)

Gastrointestinal symptoms were also commonly seen in both diseases

A large percentage of the H5N1 patients complained of gastrointestinal

symptoms like diarrhoea, vomiting and abdominal pain (Apisarnthanarak et

al., 2004) and watery diarrhoea were observed in some SARS patients later in

the course of illness Another common clinical feature seen in both SARS and H5N1 patients is lymphocytopenia An early onset of lymphopenia, thrombocytopenia and increased levels of serum transaminases were seen,

especially in the severe H5N1 cases (Tran et al., 2004; Yuen et al., 1998)

During SARS infection, there is a rapid decrease in CD4+ T cells being more severely reduced than CD8+ T cells (He et al., 2005) Lymphopenia was

prolonged, and can last till day 7 to 9 before the levels of T cells return to

normal (Wong et al., 2003) In some SARS patients, platelet count is also depressed (Peiris et al., 2003c; Tsang et al., 2003)

After the onset of SARS, cases can progress to a non-severe form of the disease characterized by mild respiratory symptoms More commonly, it progresses to moderate to severe with the development of dyspnea and

33

hypoxia Among these SARS patients, 20 to 30 percent of them need to be admitted in the intensive care unit and require mechanical ventilation The terminal event of SARS has been severe respiratory failure, multiple organ failure, sepsis or intercurrent medical illness such as acute myocardial

infarction (Booth et al., 2003; Chan et al., 2003; Tsang et al., 2003)

Similarly, severe H5N1 cases usually require ventilatory support within days after onset with rapid development of severe bilateral pneumonia Complications include acute respiratory distress syndrome (ARDS), renal failure, and multi-organ failure, with possible central nervous system

involvement (de Jong et al., 2005a; de Jong & Hien, 2006)

According to WHO, the fatality rate during SARS outbreak was estimated to be 9.6% (WHO, 2003b) and as of September 2009, the mortality rate for H5N1 influenza is close to 60% (WHO, 12 August 2010) Lung autopsy in SARS patients who died within 10 days after onset revealed diffuse alveolar damage (DAD), desquamation of pneumocytes, an inflammatory infiltrate, edema, and hyaline-membrane formation For those who died later

in the course of illness, organizing DAD was seen, with squamous metaplasia

and multinucleate giant cells of either macrophage or epithelial origin (Franks

et al., 2003; Ksiazek et al., 2003; Lee et al., 2003) Pathological examination

of patients who died from H5N1 disease showed reactive hemophagocytosis

as the most prominent feature Other findings include DAD with interstitial fibrosis, hepatic central lobular necrosis, acute renal tubular necrosis, and

lymphoid depletion (To et al., 2001)

34

1.4.2 Innate immune response against viral infection

The innate or non-specific immune defenses will be activated once the initial body surface defense is breached and the virus enters the host It is often a race between the virus and the host’s immune system at the early stage

of a viral infection There are several important players that constitute the innate immunity, the IFN response, the natural killer (NK) cells and macrophages (Nash, 2006) Viral infection usually leads to the production of type I IFN (IFN α/β), which can activate antiviral mechanisms in neighbouring cells which allow them to resist virus attack These mechanisms involve the dsRNA-dependent protein kinase R (PKR), 2’5’-oligoadenylate synthetase, and Mx proteins which block viral protein synthesis or transcription and degradation of viral mRNA The IFN is also important as it enhances the efficiency of the adaptive immunity and activates the NK cells and macrophages In the case of influenza virus infection, the macrophage-like plasmacytoid-derived DC2 dendritic cells are major producers of IFNα The

NK cells can be detected within 48 hours of a virus infection They mediate antiviral response by direct cytolysis of infected cells and through the production of IFNγ IFNγ can in turn protect cells from infection and activate macrophage antiviral mechanisms The macrophages present in the tissues of the body have three functions: i) phagocytosis of virus and virus-infected cells, ii) killing of virus-infected cells and iii) production of antiviral molecules such

as tumor necrosis factor-α (TNFα), nitric oxide and IFNα

Generally most of the cells in the body secrete high levels of type I IFN during a viral infection However, this was not the case during a SARS-CoV infection, as seen with increasing viral load during the first 10 days of the

35

infection Previously, it was reported that type I IFN, IFN-α and -β was not

detected in SARS patients, and it was not induced in vitro in infected cells, consistent in many clinical studies (Cheung et al., 2005; Jiang et

SARS-CoV-al., 2005) However, it is suggested that induction of type I IFN can be

detected early during SARS infection and are capable of activating neighbouring cells This was observed in SARS-CoV-infected macaques (de

Lang et al., 2009; de Lang et al., 2007) as well as in infected 2B4 bronchial epithelial cells (Yoshikawa et al., 2010) This recent microarray functional

study suggested that minute but physiological relevant amounts of IFNβ and IFNλs (type III IFNs) are produced by SARS-CoV infected 2B4 cells early in the infection to exert protective role even though their expression could not be detected until 48 hours post-infection

SARS-CoV like many viruses had learnt to evade the immune system

by modulating the expression of IFN-stimulated genes with antiviral activity

It was demonstrated that SARS-CoV-encoded (ORF) 3b, ORF6, ORF7, N,

nsp1, and most recently, M proteins function as IFN antagonists (Kamitani et

al., 2006; Kopecky-Bromberg et al., 2007; Siu et al., 2009; Spiegel et al.,

2005) These proteins were shown to either inhibit the IFN signaling through the JAK/STAT pathway, promote degradation of cellular RNAs or inhibit TBK1/IKKε-dependent activation of IRF3/IRF7 transcription factors

Besides IFN production, proteomic analysis of SARS patients showed activation of innate immune responses by SARS-CoV with increased acute-phase proteins such as serum amyloid A and mannose-binding lectin (MBL)

(Chen et al., 2004) A case study also revealed that low-MBL-producing

36

genotypes are shown to be associated with increased risk of SARS Thus, MBL is suggested to play a protective role in the host innate response It can bind the S protein through carbohydrate-recognition domains, resulting in protective biological effects in a calcium-dependent and mannan-inhibitable

fashion (Ip et al., 2005)

In the case of H5N1, data from some studies seem to indicate that the type of cells and the level of differentiation had a profound effect on the type I IFN response One group reported that the H5N1 virus strongly induces type I IFNs in alveolar epithelial cells when compared to the seasonal H1N1

influenza virus (Chan et al., 2005a) In contrast, Zeng et al showed that

H5N1 induces a weaker type I IFN response in differentiated bronchial

epithelium than H1N1 virus (Zeng et al., 2007), showing contradictory data

In a recent study, Chan et al went on to show that H5N1 virus replicated more efficiently and induced a stronger type I IFN response in the undifferentiated normal human bronchial epithelial (NHBE) cells as compared to the

differentiated NHBE cells (Chan et al., 2010) This was consistent with data

reported by Zeng et al It is of interest to note that similar to the SARS-CoV, the H5N1 virus was also shown to be resistant to the antiviral effects of the type I IFN and TNFα The NS1 gene seems to be associated with this

resistance (Seo et al., 2002)

ARDS is a severe form of acute lung failure which could be triggered

by SARS-CoV as well as H5N1 influenza virus Viral infection of macrophages and alveolar epithelium leads to the release of cytokines and chemokines, which can trigger pro-inflammatory cascades in both the infected

37

as well as the uninfected cells Cytokines such as the IFNα and IFNγ have antiviral effect, and chemokines lead to the infiltration of lymphocytes, macrophages and neutrophils and dendritic cells into the alveolar spaces Such cytokine cascades were observed in both the SARS and H5N1 influenza infections It has been suggested that the elevated pro-inflammatory cytokines and chemokines which lead to a ‘cytokine storm’ may have contributed to the

pathogenesis of lung damage in both of these viral infections (Cameron et al., 2008b; Peiris et al., 2009)

The hypothesis that the SARS is an immune-mediated disease remains controversial but had been supported by several reports Increased levels of CXCL10 in SARS patients throughout the disease had been demonstrated

(Cameron et al., 2008b) In the severe cases of SARS patients, the levels of

chemokines, CXCL10 and CCL2 [also known as MIP (macrophage inflammatory protein)-2α] levels were sustained as compared to non-severe patients Analysis of gene expression of PBMCs of SARS patients suggests that the response of these patients seems to be mainly an innate inflammatory response, rather than a specific immune response against a viral infection

(Reghunathan et al., 2005) In vitro experiments confirmed that SARS-CoV

infection of the macrophages and dendritic cells showed upregulation of expression of chemokines IP-10 (IFN-inducible protein-10) [also known as CXCL10], MCP (monocyte chemoattractant protein)-1, MIP-1α, and

RANTES (Law et al., 2005a; Yilla et al., 2005) Observation of elevated

levels of inflammatory cytokines and chemokines in the lungs of macaques that were infected with SARS-CoV was shown by microarray analysis and

38

quantitative RT-PCR (de Lang et al., 2007) A more recent study showed that

SARS-CoV infected aged macaques develop more severe pathology than their

younger counterparts, even though viral replication were similar (Smits et al.,

2010) Genomic analysis indicated that these aged macaques had an increase

in differential expression of genes associated with inflammation as compared

to the young macaques Thus it seems to indicate that the induced cytokines

or chemokines not only take part in the process of antiviral response, but are also involved in cell damage and even development of organ dysfunction

Similarly, several studies in humans with H5N1 diesease and in animal

models such as in ferrets (Cameron et al., 2008a) and macaques (Baskin et al.,

2009) infected with H5N1 and seasonal influenza viruses have demonstrated the increased pathology of the H5N1 virus is associated with enhanced innate host immune response It was observed that patients with H5N1 disease have higher serum levels of macrophage and neutrophil chemoattractant chemokines (CXCL10, CXCL2, IL-8) and pro- and anti-inflammatory cytokines (IL-6, IL-10, IFNγ) as compared with those with seasonal influenza

(de Jong et al., 2006) It is interesting to note that the not all the H5N1 virus

genotypes give a high-cytokine phenotype but those that are associated with

human disease seem to do so (Guan et al., 2004b; Mok et al., 2009) Thus it is

likely that the increased pathology of the H5N1 influenza virus is caused by a combination of both cytopathic effect (cell death) and dysregulated immune response (cytokine storm) (Korteweg & Gu, 2008)

39

1.4.3 Adaptive immune response against viral infection

The adaptive immune response unfolds as the viral infection proceeds This involves the antiviral antibodies, the helper T (TH) cells and the cytotoxic T lymphocytes (CTL)

1.4.3.1 Humoral immune response

Antibodies are crucial part of the adaptive immunity as they are usually the ones that block infectivity of the virus, thus preventing the infection or re-infection (Nash, 2006) Neutralizing antibodies are highly effective in inhibiting viral replication, while non-neutralizing antibodies also shown to be protective Antiviral effects of an antibody are summarized in the Table 1.4

Table 1.4 Antiviral effects of antibody Adapted and modified from D Male,

J Brostoff, D B Roth and I Roitt Immunology 2006 Seventh edition

Free virus

Antibody alone

• Blocks binding to cell

• Blocks entry into cell

• Blocks uncoating of virus Antibody +

complement

• Damage to virus envelope

• Blockade of virus receptor

Virus-infected cells

Antibody + complement

• Lysis of infected cell

• Opsonization of coated virus

or infected cells for phagocytosis Antibody bound to

infected cells

• Antibody-dependent cellular cytotoxicity (ADCC) by NK cells, macrophages, and neutrophils

40

Antibodies can be generated against any viral proteins and the neutralizing antibodies are usually generated against the surface proteins Antibodies against SARS-CoV can be found in patients and animals infected

with SARS-CoV (Hsueh et al., 2004) IgG can be detected as early as day 4

after the onset of illness During the time when patients seroconverting, at day

14 after onset of illness, levels of IgG, IgM and IgA against SARS-CoV can

be detected by immunofluorescent and ELISA assays against the nucleocaspid antigen In the large study of 623 SARS patients, neutralizing antibody levels peaked at 20-30 days and sustained for 5 months before dropping

progressively over time (Nie et al., 2004) These antibodies can neutralize the

pseudotyped viruses bearing the S protein of different strains of SARS-CoV, suggesting that these antibodies are cross-reactive It has been also shown that passive transfer of hyperimmune antisera containing high levels of neutralizing antibodies against SARS-CoV can protect senescent mice from

the infection (Vogel et al., 2007) Among the SARS structural proteins, the S protein was shown to elicit neutralizing antibodies (Buchholz et al., 2004; Lu

et al., 2004) Its major immunodominant epitope was found to lie at amino

acids 441-700

During primary influenza infection, antibodies against the HA, NA, NP and M proteins are produced (Potter & Oxford, 1979) Neutralizing antibodies against the surface proteins, HA and NA at the systemic or mucosal sites of infection are usually the ones that can provide immediate protection (Gerhard, 2001) HA-specific antibodies found in the nasal secretions are mainly IgA and IgM and infrequently the presence of IgG Serum antibody response can also be seen after a primary infection, with the detection of IgA, IgM and IgG