AN INVESTIGATION OF THE ROLE OF FC GAMMA RECEPTORS IN DENGUE VIRUS NEUTRALIZATION

78 3.1.3 FcRI requires less antibodies for complete neutralization and lowered virus yield in sub-neutralizing concentrations 80 3.1.4 DENV is preferentially phagocytosed by FcRI when

INVESTIGATING THE ROLE OF FC

2015

DECLARATION

I hereby declare that this thesis is my original work and

it has been written by me in its entirety I have duly acknowledged all the sources of information which have been used in the thesis

This thesis has also not been submitted for any degree

in any university previously

Tanu Chawla 20/01/2015

I am very much obliged to him I would like to extend my thanks to my thesis advisory committee members Associate Professor Paul MacAry, Professor Soman Abraham and Associate Professor Shee Mei Lok for their valuable suggestions during my research

I would like to express my heartfelt appreciation to my lab members Kuan Rong Chan, Hwee Cheng Tan, Summer Zhang, Eugenia Ong and Jason Tang for technical assistance, motivating discussions and friendly lab environment Special thanks to my other colleagues from DUKE-NUS for providing important reagents required for the project, to FACS core facility for cell sorting and especially to my friend Elena Okina for encouraging dialogues

Finally, I would like to thank my family for their invaluable support I am highly grateful to love of my life, my husband Sudhir Pasumarty, who has continuously supported and encouraged me and made my PhD a happy, smooth and successful journey I am thankful to my parents from bottom of my heart for trusting in me and allowing me to follow my dream They have always motivated me even though being far away in other country I am indebted to them for their unconditional love Lastly, I am thankful to my brother, Karan who made me laugh at times when I was stressed Not forgetting my friend, Nimrata, Swati and Soka association friends with whom I share my joys and sorrows, I am glad

to thank all

Table of Contents

Acknowledgements … ……… iii

Table of contents …… ……… iv

Summary ……… ……… viii

List of tables …… ……… x

List of figures …… ……… xi

List of abbreviations … ……… xiii

List of publications …… ……… xvii

S.No Topic Page Chapter 1: An Introduction to Dengue … ……… 1

1.1 Dengue …….……….……… 2

1.1.1 Dengue epidemiology ……… ……… 3

1.1.2 Dengue disease manifestations ……… …… 10

1.1.3 Dengue virus genome and structure ……… …… 13

1.2 Dengue prevention and control ……….………… 18

1.2.1 Vector control ……… ……… ……… 18

1.2.2 Antivirals ……… ……… ……… 23

1.2.3 Vaccine ……… ……… ……… ……… 25

1.2.4 Challenges in vaccine development ……… ……… 28

1.3 Immune correlates of DENV infection ……… … 32

1.3.1 T cell responses ……… ……… ……… 32

Pathogenesis by T cells ……… ……… 33

Protection by T cell responses … ………… ……… 34

1.3.2 Antibodies are protective or pathogenic …….……… 36

Protection by antibodies ……….… ……… 39

Antibody mediated DENV neutralization …… ……… 42

Pathogenesis by antibodies ……… ……… 44

1.4 FcRs Utilization in DENV enhancement and neutralization ….… 49 Study aims ……… ……….… 57

Chapter 2: Methods … ……… 59

2.1 Cells …….……….……… 60

2.2 Antibodies ……….……… 60

2.3 Virus culture and purification ……… …….……… 62

2.4 Plaque assay ……….…….……… 62

2.5 Titration of h3H5 antibody for complete neutralization in THP-1, K562 shControl and shFcRIIA cells ……… 63

2.6 Fluorescent Labeling of Viruses ……… ………… … 63

2.7 Infection for localization studies in THP-1 or K562 cells … … 64

2.8 Immunofluorescence Assay (IFA) ……… ………… 65

2.9 Fluorescence-activated cell sorting (FACS) ……… 66

2.10 siRNA transfection in THP1 or K562 ……… …… 66

2.11 Flow Cytometry to determine surface expression of FcRs … 68

2.12 Western Blot ……… ……… 68

2.13 Human Phospho-kinase array ……… ………… 69

2.14 Lentiviral particles transduction in THP-1 cells ……… 69

2.15 Rapamycin drug treatment ……… ………… 70

2.16 Statistical analysis ……… ………… 70

Chapter 3: Results … ……… 71

3.1 Stoichiometric antibody requirements for DENV neutralization,

when antibody-opsonized DENV is ligated with different

activating Fc gamma receptors (FcRI and FcRIIA) 73 3.1.1 FcRIIA mediated phagocytosis requires higher antibody

concentration for DENV neutralization …… 74 3.1.2 Inhibition of phagocytosis in K562 cells mediated

by FcRIIB ……… 78 3.1.3 FcRI requires less antibodies for complete neutralization

and lowered virus yield in sub-neutralizing concentrations 80 3.1.4 DENV is preferentially phagocytosed by FcRI when

opsonized with neutralizing antibody but not sub-neutralizing levels ……… 82 3.2 Analysis of signaling events regulated by FcRI and FcRIIA

involved in antibody-mediated neutralization in monocytes … 87 3.2.1 Akt-protein kinase pathway is activated

following infection with the h3H5 neutralizing antibody opsonized DENV in the presence of FcRI ……… 89 3.2.2 Deriving a stable FcRI and FcRIIA knockdown cell line 95 3.2.3 Akt and 4EBP-1 phosphorylation in shControl and

shFcRIIA Cells ……… ……… 100 3.2.4 ISG translation in shFcRIIA cells ……… 103 3.2.5 Low expression of eiF4E does not change antibody

requirement for DENV neutralization ……… ……… 104 3.2.6 Inhibiting of mechanistic target of rapamycin (mTORC1)

activity did not change the antibody requirement in DENV neutralization ……… 106

Chapter 4: Discussion … ……… 109

4.1 Lower antibody concentration required for DENV neutralization if immune complex is phagocytosed through FcRI compared to FcRIIA …….……….……… 112

4.2 Preferential interaction of FcRI and DENV opsonized with antibody at neutralizing but not sub-neutralizing levels … 116

4.3 Analysis of FcRI mediated signaling to unravel the path for DENV clearance 118

4.4 Future directions 121

4.5 Conclusions 123

Chapter 5: Bibliography 124

Appendices 155

Summary

Dengue is the most prevalent mosquito borne viral disease globally and is hence a cause of disease burden on all tropical nations Currently there is no licensed vaccine or therapeutic for dengue and vector control has not been sustainable in stemming the spread of disease In humans, DENV has been shown to infect monocytes, macrophages and dendritic cells, all of which express activating Fc gamma receptors (FcR) Consequently, neutralization of DENV by antibodies should be understood in the context of these cell types Using human acute monocytic leukemia cells expressing FcRI and FcRII (THP-1) and human myelogenous erythroleukemic cells (K562) expressing FcRII, this thesis examined how different activating FcRs influence the amount of antibody required to completely neutralize DENV Experimental data indicates that the amount of antibody needed for complete DENV neutralization is significantly lower when immune complexes are phagocytosed by FcRI than by FcRIIA Moreover, the data also demonstrates that FcRI is preferentially engaged by DENV when opsonized with neutralizing levels of antibody

by clustering this receptor Finally, we also show that lower antibody concentration required when antibody-opsonized DENV is phagocytosed by FcRI instead of FcRIIA cannot be explained by an antiviral response triggered by FcRI signaling that controls viral uncoating due to insufficient neutralization Collectively, data from this

thesis suggests that phagocytosis by FcRI is a more efficient pathway

to remove antibody-opsonized DENV from systemic circulation

List of Tables

1.1 List of DENV vaccines in progress …….…….………… 31

1.2 ADE in different RNA Viruses ……… ……… 44

2.1 List of antibodies used ……… ……… 61

2.2 List of siRNAs……… 67

3.1 List of Kinases in Human phospho kinase array from R&D systems ……….……… ……… 91

3.2 List of Kinases relevant to FcR pathway……… 93

List of Figures

1.1 South East Asian countries with dengue epidemic, 2014 ……… 5

1.2 Guidelines specified by WHO for progression of dengue to severe dengue ……….……… 12

1.3 DENV genome organization ……….…… 14

1.4 Summary of DENV life cycle ……… 16

1.5 E protein organization on surface of DENV ……… 17

1.6 Antibody response and viremia in primary and secondary DENV infection ……… 38

1.7 Flavivirus enhancement and neutralization ……… 41

1.8 Model for antibody-dependent enhancement (ADE) of DENV infection ……… 48

1.9 Family of Human FcRs ……… … 50

1.10 FcR mediated signaling pathway involved in phagocytosis…… 54

3.1 FcRIIA mediated phagocytosis requires higher h3H5 antibody concentration for DENV neutralization ……… 76

3.2 Higher antibody concentration inhibits DENV immune complex internalization in K562 cells ……… ……… 77

3.3 Inhibition of phagocytosis in K562 cells mediated by co-ligation of FcRIIB and no change in antibody requirement for DENV by neutralization by FcRIIB knockdown K562 cells ……… 79

3.4 FcRI engagement requires lesser antibody for neutralization and lowered virus yield in enhancing concentrations … ……… 81

3.5 Preferential interaction of FcRI and DENV opsonized with neutralized h3H5 ……… ……… ……… 83

3.6 Clustering of FcRI when ligated with DENV in neutralizing condition ……… ……… ……… 85

3.7 Phospho kinase dot blot showing relative phosphorylation……… 92

3.8 Phosphorylation of Akt and PRAS40 in THP-1 cells in

phospho kinase array …….… ……… ……… 94 3.9 Characterization of shControl and shFcRIIA cell line …….…… 98 3.10 CRISPR-CAS9 transfection in THP-1 cells …….… ………… 99 3.11 Akt and 4EBP-1 phosphorylation in shFcRIIA cells …….… 102 3.12 ISG translation in shFcRIIA cells …….… ……… 103 3.13 Reduction in eif4e expression did not change antibody

requirement for DENV neutralization …….… ……… 105 3.14 mTORC1 inhibition by using rapamycin did not show difference in

antibody requirement for DENV neutralization …….… ………… 108

List of Abbreviations

CLEC5A C-type lectin domain family 5, member A

DALYs Disability-adjusted life years

DC-SIGN Dendritic cell-specific intercellular adhesion

Eif4E Eukaryotic translation initiation factor 4E

4EBP-1 Eukaryotic translation initiation factor 4E binding

protein FcR Fragment crystallisable receptors

HNE buffer HEPES, sodium chloride, EDTA buffer

IP-10 Interferon-inducible protein 10

IRF Interferon-regulatory factor

ITAM Immunoreceptor tyrosine activating motif ITIM Immunoreceptor tyrosine inhibitory motif

LILRB1 Leukocyte immunoglobulin-like receptor-B1

μm micrometers

MHC Major histocompatibility complex

mTOR Mechanistic or mammalian target of rapamycin

NK cells Natural killer cells

pfu/ml Plaque forming units per ml

PRNT Plaque-reduction neutralizing test

RIG-I Retinoic acid-inducible gene

SDS-PAGE Sodium dodecyl sulfate polyacrylamide gel

electrophoresis

Src Proto-oncogene tyrosine-protein kinase

STAT Signal transducer and activator of transcription

List of Publications

1 Research article

Chawla T, Chan KR, Zhang SL, Tan HC, Lim AP, Hanson BJ

and Ooi EE (2013) Dengue Virus Neutralization in Cells

Expressing Fc Gamma Receptors Plos one, Volume 8 /Issue 5 / e65231

2 Review article

Chawla T, Wilder-Smith A and Ooi EE Dengue, an expanding

neglected tropical disease Neglected tropical diseases in East Asia (in press)

CHAPTER 1

AN INTRODUCTION TO DENGUE

1.1 Dengue

Dengue is an acute viral infection caused by any of the four dengue virus serotypes (DENV1-4) that are immunologically different from each other

DENV is transmitted by Aedes mosquitoes from humans to humans in

urban centers (Simmons et al, 2012) and has emerged to be the most important mosquito borne viral disease in the world It is endemic throughout the tropical world and rapidly spreading to several sub-tropical regions The distribution of DENV and its mosquito vectors, principally

Aedes aegypti, places 50% of the world’s population at risk of dengue each year (Wilder-Smith et al, 2010) Latin America, South and South East (SE) Asia as well as parts of Africa experience high incidence of dengue annually, with Asia bearing the brunt with an estimated 75% of the global disease burden (WHO, 2012) However, its spread in non-endemic countries like Europe and United States of America (Alves et al, 2013; Anez & Rios, 2013; La Ruche et al, 2010; Murray et al, 2013a; Powell & Tabachnick, 2013) has meant that its geographical distribution is expanding globally creating a difficult situation for health organizations to control the disease

1.1.1 Dengue epidemiology

DENV was first isolated in 1943, by Ren Kimura and Susumu Hotta while studying blood samples that were taken from patients during the 1943 dengue epidemic in Nagasaki, Japan A year later, Albert B Sabin and Walter Schlesinger also isolated DENV, and since then there has been a sharp rise in the spread of disease that is correlated with the increase in urbanization, international trade and travel (Gubler, 2002; Gubler, 2011; Tatem et al, 2012) Climatic change is also influencing the spread of dengue at higher altitude regions To preclude the spread of dengue, it is essential to understand the pattern of occurrence, including the factors that may have changed with time in endemic regions

Dengue cases, both endemic and imported, have been increasing worldwide Though the Second World War facilitated the geographical

expansion and population densities of Aedes aegypti, the unplanned

urbanization later provided the optimum conditions for DENV propagation People shifted to cities for work that resulted in unexpected growth of urban centers and housing with improper water supply, sanitation and

sewage systems that aided the breeding of Aedes mosquitoes The

presence of susceptible human hosts along with the mosquito vector led

to the emergence of cyclical dengue epidemics (Ooi & Gubler, 2009) The Philippines had two DHF outbreaks in a gap of two years, in 1954 and

1956 and Thailand had an epidemic in 1958 (Halstead, 1980) Prior to

1970, dengue epidemics had occurred only in nine countries Since then, however, dengue activity has been reported in over 128 countries and the scale of each epidemic have continued on an increasing trajectory (WHO, 2009; Brady et al, 2012) A recent study estimated that 390 million dengue infections occurred globally in 2010 and 96 million of these infections resulted in symptomatic disease Approximately 70% of these symptomatic cases resided in Asia (Bhatt et al, 2013) More recently, a study in India demonstrated that the true number of dengue cases between 2006 to 2012 was 282 times greater than what was officially reported (Shepard et al, 2014) Collectively, these fresh estimates of the global annual disease burden posed by dengue is several times more than the estimates reported previously by the World Health Organization (WHO, 2009)

Within SE and East Asia, dengue also has an expanded geographic distribution The Philippines, Vietnam, Cambodia, Thailand, Malaysia, Singapore and Indonesia have been hyper-endemic for dengue for many years Taiwan and China initially recorded imported cases due to travellers (Wu et al, 2010) but these countries now have seasonal epidemics of dengue Japan for first time after the Second World War had

an outbreak of dengue fever in August 2014 with nearly 150 confirmed cases of DF (CDC, 2014; WHO, 2014) These trends suggest that the burden of dengue will likely continue to increase in Asia in the coming years (Figure 1.1)

Figure 1.1: South East Asian countries with dengue epidemic, 2014

The increase in both vector and virus distribution geographically in areas with high human population density have also resulted in the co-circulation

of all DENV serotypes in many tropical urban centers The co-circulating DENV serotypes sustain themselves in low levels till the environmental factors are favorable causing epidemics every 3-5 years (Murray et al, 2013b) Such endemicity for all four DENV serotypes at any one time has been termed hyperendemicity (Gubler, 1998; Gubler, 2006; Gubler & Clark, 1995) Messina et al recently traced the global pattern of concurrent

or sequential emergence of all four DENV serotypes from 1943 to 2013 indicating the progression of hyperendemicity over the years (Messina et

al, 2014)

Changes in the prevalence of circulating DENV serotypes have triggered epidemics, likely through the low herd immunity in the population to the new serotype Furthermore, antibodies that develop to the serotypes that had circulated in the population could potentially also enhance the overall scale of the epidemic from the introduction of heterologous serotypes This is eloquently illustrated in the 1981 DHF epidemic in Cuba triggered

by a DENV2 introduction following circulation of DENV1 in that country Long term prospective cohort studies (1999-2005) in Iquitos, Peru, have underscored the role of serotype switches and epidemic emergence (Getis

et al, 2003; Morrison et al, 2004a; Morrison et al, 2004b) The authors observed that 80% of population included in their study had DENV antibodies against DENV1 and/or DENV2 Between the start of the study

in 1999 to last quarter of 2001, approximately 6% of the cohort were infected with either DENV1 or DENV2 each year In the last quarter of

2001, entomological conditions became favourable; the increased Aedes

aegypti population density correlated with increase in infections initially by

DENV1 By March 2002, however, DENV1 was replaced by an introduction of DENV3 likely from other parts of Peru (Guagliardo et al, 2014; Kochel et al, 2008; Morrison et al, 2004b) and this DENV3 introduction resulted in a dengue outbreak (Morrison et al, 2010) A similar trend was seen in 2008 with the introduction of DENV4 (Forshey et al, 2009)

With hyperendemicity, the fluctuation in serotype prevalence is governed,

in part, by herd cross-reactive immunity levels A cohort study in Kamphaeng Phet, situated in northern Thailand, demonstrated that higher incidence of dengue infection in the previous year resulted in a milder disease in subsequent years (Endy et al, 2011) This observation suggests that although heterotypic immunity does not protect against infection beyond the first 3 months (Sabin, 1952; Snow et al, 2014), it does protect against more severe disease for a year or more after acute dengue This protection is short-lived Anderson et al, showed that shorter time intervals between first and second infection is associated with protection but as the antibodies wane from protective to enhancing levels,

it leads to severe form of dengue disease (Anderson et al, 2014) Hyperendemicity thus provides the necessary mix of virus serotypes that

is able to take advantage of increasingly susceptible population each time heterotypic immunity levels wane following a preceding epidemic More frequent cyclical epidemics is thus to be expected with hyperendemicity

Besides fluctuations in serotypes, Gubler and Trent have also postulated that hyperendemic transmission increases the likelihood of emergence of DENV strains with greater virulence or epidemic potential (Gubler & Trent, 1993) Indeed, studies on the viral genome have also yielded interesting insights into the selection and evolution of the dengue viruses, both temporally and geographically (Lewis et al, 1993; OhAinle et al, 2011; Raghwani et al, 2011; Rico-Hesse et al, 1998; Twiddy et al, 2002; Zhang

et al, 2006) Such genetic selection of DENV appears to influence epidemic dengue transmission The DHF outbreak in Puerto Rico in 1994 was not caused by the emergence of a new serotype, DENV2 Instead, surveillance data indicated that this serotype had been endemic in Puerto Rico since 1982 but that changes in the genomic sequence of DENV2 were associated with the DHF outbreak in 1994 (Bennett et al, 2003) (Bennett et al, 2006) In Sri Lanka, DENV3 caused an outbreak of DHF in

2000 despite the fact that this serotype was endemic in that country prior

to 2000 (Messer et al, 2002) As with the experience in Puerto Rico, the outbreak was associated with a change in the DENV3 genomic sequence (Messer et al, 2003) Likewise, the DENV2 outbreak in 2007 in Singapore was associated with a change in viral genotype (Lee et al, 2010) More

recent studies in Nicaragua indicated that a clade change in the circulating virus was associated with increased fitness in human and mosquito cells, along with greater rates of severe dengue (OhAinle et al, 2011) These studies suggest that with intensive transmission, the likelihood of selection for an epidemiologically fitter virus will increase However, genomic changes can also reduce the epidemic potential of virus Phylogenetic investigations into the DENV2 outbreak in the Pacific islands demonstrated that this virus evolved as it moved from island to island, inducing attenuation (Steel et al, 2010) that gave rise to a milder disease This culminated in Tonga where Gubler and colleagues observed high rates of mild or even asymptomatic DENV2 infections (Gubler et al, 1978) Collectively, the evolution of DENV can have critical impact on epidemiological outcomes

1.1.2 Dengue disease manifestations

DENV transmission is a loop from mosquito to human Aedes mosquito

injects virus into human skin, infecting immature dendritic cells (DCs) also called Langerhans cells (LCs) (Wu et al, 2000), these LCs then migrates

to the lymph nodes There, chemokines and cytokines released by LCs recruit monocytes and macrophages, which are target cells for infection (Kou et al, 2008; St John, 2013) The major organ system associated with DENV pathogenesis stated is the immune system, blood vessels (endothelial linings) and Liver (Martina et al, 2009)

Dengue has a wide range of clinical manifestations DENV infections can

be sub-clinical or symptomatic with a wide spectrum of disease, ranging from mild dengue fever (DF) to life threatening forms such as dengue haemorrhagic fever (DHF) and dengue shock syndrome (DSS) that is characterized by an extreme vascular permeability The number of mortality cases increase as the disease progresses from classical DF to more complex forms such as DHF/DSS DENV infection is followed by an incubation period of 3-7 days Following which, disease symptoms manifest abruptly starting with the febrile phase followed by the critical phase and then recovery phase (WHO, 2009) The initial febrile phase is characterized by high, sudden onset fever with severe headache, myalgia, retro-orbital pain, macular rash, leukopenia or nausea The febrile phase lasts for 3-7 days with most patients recovering from the symptoms

mentioned earlier Some DF cases may progress to severe disease at the time of fever defervescence At this critical phase, patients will experience haemorrhagic manifestations such as petechiae or mucosal bleeding, vascular leakage (hypovolaemia), thrombocytopenia and persistent vomiting If left untreated, fluid loss through vascular leakage would lead

to hypovolemic shock where mortality can be as high as 30% (Gubler, 1998; Heilman et al, 2014) Since there is no specific antiviral therapy available, the mainstay of case management is to provide appropriate fluid support to complications arising from hypovolemia With fluid therapy, most patients will survive the 24-48 hours of critical phase of illness and enter the recovery phase where they gradually revert to normal with reabsorption of fluids and an increase in WBC counts A small proportion

of patients experience prolonged lethargy and malaise, which could last for weeks after the acute illness (Simmons et al, 2012; WHO, 2012)

Predicting disease progression to the more severe forms of dengue is challenging The World Health Organization (WHO) has provided a list of warning signs for clinicians to guide patient monitoring during the acute phase of illness (Figure 1.2) These signs could help to identify patients that are likely to progress to DHF or DSS, allowing for early administration

of supportive treatments like intravenous fluid therapy (WHO, 2009) Its true performance in prognosticating impending DHF/DSS or other forms of severe dengue remains to be fully determined through prospective studies (WHO, 2009; Barniol et al, 2011)

Figure 1.2: Guidelines specified by WHO for progression of dengue

to severe dengue A criterion is set by WHO for dengue fever cases and

dengue with severe condition The warning signs in patients would indicate the requirement of strict observation and medical intervention as these patients may possibly develop severity of disease Those without warning signs may, also develop into severe dengue cases Severe dengue cases are characterized by severe plasma leakage, severe haemorrhage or severe organ impairment Adapted from WHO 2009

1.1.3 Dengue virus genome and structure

DENV belongs to genus Flavivirus in the family Flaviviridae, which

includes other viruses that are clinically relevant such as West Nile virus (WNV), Yellow fever virus (YFV) and Japanese encephalitis virus (JEV) The four DENV serotypes show 65-70% homology in their genome sequence The virion of dengue is ~50nm in diameter and is composed of

a single, positive-strand RNA genome of ~11Kb with a single open reading frame which translates mRNA to one long polypeptide that is then cleaved

by viral or host proteases into three structural proteins, capsid (C), envelope (E) and membrane (M) and seven nonstructural proteins NS1, NS2a, NS2b, NS3, NS4a, NS4b and NS5 (Figure 1.3) (Mukhopadhyay et

al, 2005) The C protein packages the positive sense RNA genome inside

it, which is then surrounded by a lipid bilayer with M and E proteins attached to it Nonstructural proteins are mainly involved in viral replication and modulation of the host response (Pastorino et al, 2010) The dengue genome has flanking 3’ and 5’ untranslated regions which are vital for virus replication and translation (Clyde & Harris, 2006)

Figure 1.3: DENV genome organization. DENV translated as a polyprotein is cleaved by viral and host proteases (represented by arrows) Structural proteins are cleaved by signalase cleavage in the ER (blue arrow) Non-structural (NS) proteins are mostly cleaved by the NS2B-NS3 viral protease in the cytoplasm (red arrow), except NS1 which

is released into the ER by an unidentified protease The NS proteins are essential for viral replication, assembly and modulation of host cell responses Adapted from Perera and Kuhn, 2008

The E glycoprotein of DENV attaches itself to the putative cellular receptor and enters host cells via clathrin-mediated endocytosis Upon internalization, endocytic vesicle comprising of virus are trafficked to the Rab-5 positive early endosome within five mins of entry (van der Schaar et

al, 2008) Thereafter it matures to Rab-7 positive late endosomes, which provides a low pH environment for uncoating of viral RNA in to the cytoplasm (Harrison, 2008; van der Schaar et al, 2008) Viral RNA is then translated to a polyprotein and cleaved by viral and host proteases Viral

RNA and structural genes are then assembled in the endoplasmic reticulum (ER) to form immature viral particles that have E-prM heterodimers on the surface, making its appearance spiky (Zhang et al, 2003) These particles are then transported to trans-Golgi network (TGN) for maturation The mild acidic environment in TGN leads to the reorganization of E proteins into a flat dimeric conformation, providing a smooth appearance (Figure1.4) (Modis et al, 2004; Yu et al, 2008) Furin cleavage leads to dissociation of prM and extracellular secretion of an infectious mature virus (Zybert et al, 2008) The mature viruses are then released from the cells to infect other cells Some recent studies have shown that immature viruses can be infective in presence of antibodies (Dejnirattisai et al, 2010; Rodenhuis-Zybert et al, 2010)

Cryo-electron microscopy (cryoEM) has shown that there are 180 copies

of E proteins on the surface of the DENV forming an icosahedral symmetrical structure (Figure1.5) (Kostyuchenko et al, 2013) The E protein is the prime structural protein that induces antibody response Each E protein has three domains: Domain I (DI), Domain II (DII), and Domain III (DIII) (Modis et al, 2003; Zhang et al, 2004) DIII is believed to play a critical role in binding host cell receptors DII consist of the fusion loop that interacts with the endosomal membrane for fusion of the virus during entry into the cell (Rey et al, 1995) The hinge region that joins DI

to DII is quite flexible and a low pH environment in the endosome initiates

conformational change and bending of DII to uncover its fusion loop (Zhang et al, 2004)

Figure 1.4: Summary of DENV life cycle: The process of virus entry,

replication and assembly is presented in this figure: Adapted from Review Nature 2005 (Mukhopadhyay et al, 2005)

Figure 1.5: E protein organization on surface of DENV DI, DII, DIII

and fusion loop on DII are coloured in red, yellow, blue and green

respectively E protein dimers flatten on DENV surface and are arranged

in a herringbone pattern E proteins are organized in icosahedral

symmetry, and the black triangle indicates one asymmetric unit Adapted

from Lok, 2014

1.2 Dengue prevention and control

Despite more than three decades of research towards the development of

a dengue vaccine, there is no approved vaccine, drug or effective vector control However, efforts for prevention and control of disease are still in progress Different methods for improved vector control, development of antiviral drugs and vaccines for all four serotypes are currently underway and these are as follows:

1.2.1 Vector control

The prevention of dengue remains reliant on vector control which has been shown to be effective in Singapore (Ooi et al, 2006) and Cuba (Bonet et al, 2007) However, source reduction is costly (Carrasco et al, 2011) and few countries in dengue hyper-endemic regions can afford the intensive vector control programs used in Singapore and Cuba Moreover, such an approach is ultimately not sustainable The reasons for this lack

of sustainability have been reviewed (Ooi et al, 2006) Improved methods

to control dengue vectors are thus urgently needed

Use of insecticides has been a common method in controlling mosquito population for many years Insecticides such as pyrethroid deltamethrin or metofluthrin have been used and shown to kill mosquitoes (Ritchie & Devine, 2013) However there are problems related to the usage of insecticides such as dosage, method of delivery and the emergence of

insecticide resistance in mosquitoes (Maciel-de-Freitas et al, 2014) Most importantly, the use of insecticides involves high cost and the evidence of efficacy in reducing mosquito population density remains highly controversial (Eisen et al, 2009; Gubler, 1989) Use of chemical insecticides is thus not widely recommended

Given the lack of sustainability of source reduction and lack of effectiveness of chemical insecticides, alternative approaches to vector control have been on the agenda for some time The most promising

approach currently is using symbiotic bacteria, Wolbachia, to reduce the vectorial capacity of Aedes aegypti Wolbachia is found naturally in 60% of

insects including fruitflies, moths, butterflies and dragonflies but not in

Aedes aegypti (Sinkins et al, 1995) It is maternally inherited from one

generation to the next generation through insect eggs and can modify the

reproduction of insects it stays within If a male insect has Wolbachia but a female does not, then the eggs laid do not contain Wolbachia and hence they do not hatch However if the female insect contain Wolbachia but the male does not then the eggs would contain Wolbachia and they will hatch

producing new offspring (Bordenstein & Werren, 1998; Sinkins et al, 1995)

A study done in La Reunion Island had shown that Aedes albopictus naturally infected with Wolbachia restricts virus density in salivary glands

and thus reduces transmission of DENV when compared to Aedes

albopictus without Wolbachia (Mousson et al, 2012)

Infection with Wolbachia can alter vectorial capacity as well as DENV

replication in Aedes, although the underlying mechanism by which

Wolbachia suppresses replication seems complex and remains to be fully elucidated (Rances et al, 2013) Scott O’ Neil’s group has developed several strains of Wolbachia that in laboratory and field studies have shown encouraging results (Walker et al, 2011) The main approach is to

infect field caught Aedes aegypti in the laboratory with strain of Wolbachia (wMel) before releasing them into the field The Wolbachia infected

mosquitoes will compete with wild-type Aedes aegypti to mate with female mosquitoes and thus, over several cycles, transmit the bacteria throughout the population (Ruang-Areerate & Kittayapong, 2006)

The first field trial was initiated by introducing wMel Aedes aegypti in

Yorkeys Knob and Gordonvale in the north of Queensland, Australia to assess the stability, persistence and fitness of bacterial infection

(Hoffmann et al, 2011) A year after the release of wMel Aedes aegypti,

field mosquitoes were caught and challenged with three DENV serotypes

Results indicated that a significant proportion of field caught Aedes

aegypti were infected with wMel and showed reduction in DENV

replication (Frentiu et al, 2014) A follow up study of almost 3 years

showed the persistence and stability of wMel infection among local Aedes

population (Hoffmann et al, 2014) Trials with such an approach to vector control are currently on-going in Vietnam, Indonesia and Brazil (2014a; 2014b)

The other approach to eliminate Aedes mosquitoes is to use genetically

modified (GM) mosquitoes (Alphey, 2014) The most advanced of such an approach is the one developed by Oxitec Termed ‘release of insects containing a dominant lethal’ (RIDL), this approach uses the insertion of a lethal gene, under the control of a chemically-inducible promoter, into the

Aedes aegypti genome using a transposon based system (Alphey et al,

2002) The GM male mosquito with lethal gene mates with wild-type female, the lethality trait is passed into the offspring which prevents them

from developing through the larval stages into imagoes The strain Aedes

aegypti OX513A has been successfully field-tested in Cayman Island

showing encouraging suppression of wild-type mosquitoes (Harris et al,

2012; Harris et al, 2011) OX513A Aedes aegypti have also been released

in Malaysia to assess its dispersal and survival in open field The survival was similar to that of wild type but dispersal was reduced in transgenic mosquitoes (Lacroix et al, 2012) Oxitec is currently collaborating with Moscamed, a company in Brazil for mass production of OX513A involving many stages such as egg production, larvae incubation and then sorting the male and female pupae or adults for field release (Carvalho et al, 2014; http://www.oxitec.com/health/our-products/aedes-agypti-

ox513a/ongoing-field-trials-of-ox513a-aedes-aegypti/) Field trials in Brazil and Panama are in progress

Several approaches have been taken to develop antiviral drugs for dengue One approach is to target viral proteins The structural proteins (C, M, and E) are involved in viral entry and packaging of viral genome and the non-structural proteins (NS1, NS2a, NS2b, NS3, NS4a, NS4b and NS5) are important for viral replication, assembly and evasion of host responses The Novartis Institute of Tropical Diseases (NITD) has been established in Singapore to develop an anti-dengue drug that inhibits any

of these viral proteins (Lim et al, 2013) Thus far, the Institute has made significant contributions to our understanding of viral replication (Dong et

al, 2012; Noble et al, 2010; Wang et al, 2011; Xie et al, 2011; Zou et al, 2011) although a lead compound (NITD008) that showed promising in vitro characteristics failed pre-clinical toxicity studies (Yin et al, 2009)

Another drug that inhibits the RNA dependent RNA polymerase activity of the NS5 protein was also tested in a phase II clinical trial in Vietnam This