STUDY OF FLAVIVIRUS INFECTION - COnnecting REpositories fileSTUDY OF FLAVIVIRUS INFECTION ANIMAL MODELS, DETERMINANTS OF NEUROINVASION AND VECTOR SPECIFICITY Katholieke Universiteit

STUDY OF FLAVIVIRUS INFECTION ANIMAL MODELS, DETERMINANTS OF

NEUROINVASION AND VECTOR SPECIFICITY

Katholieke Universiteit Leuven

Faculteit Geneeskunde

Rega Instituut voor Medisch Onderzoek

Departement Microbiologie en Immunologie Laboratorium voor Virologie en Chemotherapie

Nathalie Charlier

TABLE OF CONTENTS ABBREVIATIONS

CHAPTER 1 General introduction

1. Flaviviridae .................................................................................................................... 2 2. Flavivirus ....................................................................................................................... 2 3. Pathology....................................................................................................................... 5 4. Genome structure ......................................................................................................... 6 5. Viral replication cycle .................................................................................................... 9 6. Structure of the virion ..................................................................................................10 7. Chimeric viruses ...........................................................................................................12 8. Neuroinvasion ..............................................................................................................20 9. Transmission cycle of flaviviruses................................................................................25 10. Aims and objectives ...................................................................................................28

CHAPTER 2 Infection of SCID mice with Montana Myotis leukoencephalitis virus as a model for Flavivirus encephalitis

SUMMARY ...........................................................................................................................32

INTRODUCTION ...............................................................................................................32

MATERIALS AND METHODS .........................................................................................34

RESULTS ..............................................................................................................................42 2.1. Antiviral susceptibility ...............................................................................................42 2.2. Ultrastructural analysis ..............................................................................................43 2.3. MMLV-induced encephalitis and mortality ..............................................................43 2.4. Viral RNA in blood samples and macrophages ........................................................44 2.5. Histopathology ..........................................................................................................45 2.6. Effect of treatment with an interferon-inducer on MMLV replication and associated mortality ..........................................................................................................46

DISCUSSION .......................................................................................................................49

CHAPTER 3 Complete genome sequence of Montana Myotis leukoencephalitis virus, phylogenetic analysis and comparative study of the 3' untranslated region of flaviviruses with not known vector

SUMMARY ...........................................................................................................................54

INTRODUCTION ...............................................................................................................54


RESULTS ..............................................................................................................................60 3.1. Genome and amino acid sequence ...........................................................................60 3.2. Predicted cleavage sites in the MMLV polyprotein ..................................................62 3.3. Characteristics of the 5’- and 3’-terminal nucleotide sequences ...............................63

3.4. Comparative study of the folding of the 3’-terminal sequences of four NKV flaviviruses ........................................................................................................................66 3.5. Phylogenetic analysis .................................................................................................69

DISCUSSION .......................................................................................................................71

CHAPTER 4 A rapid and convenient variant of fusion-PCR to construct chimeric flaviviruses

SUMMARY ...........................................................................................................................78

INTRODUCTION ...............................................................................................................78


RESULTS ..............................................................................................................................85 4.1. Construction of the pMODV-prM+E plasmid ........................................................85 4.2. Construction of chimeric viruses by fusion-PCR .....................................................85 4.3. Cloning of MODV prM and E genes .......................................................................86 4.4. Recovery of chimeric MODV/YFV viruses.............................................................87 4.5. Verification of the chimeric virus produced by BHK cells ......................................87

DISCUSSION .......................................................................................................................90

CHAPTER 5 Exchanging the yellow fever virus envelope proteins for Modoc virus prM and E results in a chimeric virus that is neuroinvasive in mice

SUMMARY ...........................................................................................................................92

INTRODUCTION ...............................................................................................................92


RESULTS ..............................................................................................................................98 5.1. Characterization of MODV/YFV chimera in cell culture........................................98 5.2. Confirmation of the chimeric geno- and phenotype ..............................................101 5.3. Infection of SCID mice with MODV, YFV 17D and MODV/YFV ....................102 5.4. MODV/YFV RNA in the brain .............................................................................104

DISCUSSION .....................................................................................................................105

CHAPTER 6 Construction of a Modoc virus full-length clone

SUMMARY .........................................................................................................................110

INTRODUCTION .............................................................................................................110

MATERIALS AND METHODS .......................................................................................112

RESULTS ............................................................................................................................116 6.1. RNA extraction .......................................................................................................116 6.2. Long PCR ................................................................................................................116

DISCUSSION .....................................................................................................................118

CHAPTER 7 The highly conserved pentanucleotide motif 5'-CACAG-3' in the 3' untranslated region of the flavivirus genome determines whether a flavivirus is able or not to replicate in mosquito cells

SUMMARY .........................................................................................................................122

INTRODUCTION .............................................................................................................122


RESULTS ............................................................................................................................126 7.1. Replication kinetics of MODV, YFV and MODV/YFV in Vero and mosquito cells ........................................................................................................................................126 7.2. Replication kinetics of MODV/YFV(CACAG) and MODV/YFV(CUCAG) in Vero and mosquito cells ................................................................................................127

DISCUSSION .....................................................................................................................129

CHAPTER 8 Construction of a chimeric dengue virus containing the structural proteins prM and E of Modoc virus

SUMMARY .........................................................................................................................134

INTRODUCTION .............................................................................................................134


RESULTS ............................................................................................................................140 8.1. Introduction of a unique restriction site in plasmid p4 ..........................................140 8.2. Construction of chimeric viruses by fusion-PCR ...................................................140 8.3. Cloning of MODV prM and E genes .....................................................................142

DISCUSSION .....................................................................................................................143

CHAPTER 9 General discussion and perspectives

1. The genome of MMLV, a NKV flavivirus ................................................................147 2. The MMLV mouse model .........................................................................................148 3. Construction and characterization of chimeric flaviviruses consisting of the backbone of YFV or DENV and the prM and E genes of a NKV flavivirus...............................149

REFERENCES ...................................................................................................................173

ABBREVIATIONS

APOIV Apoi virus

ATCC American Type Culture Collection

ATP adenosine triphosphate

BHK baby hamster kidney (cells)

BSA bovine serum albumine

BSL biosafety level

C capsid protein

CCID50 cell culture infective dose 50%

cDNA complementary DNA

CFAV cell fusing agent virus

CNS central nervous system

CPE cytopathic effect

CS conserved sequence

CTP cytidine triphosphate

DBV Dakar bat virus

DENV-1-4 dengue virus, serotype 1-4

DHF dengue hemorrhagic fever

DNA deoxyribonucleic acid

dNTP deoxynucleoside triphosphate

DSS dengue shock syndrome

E envelope protein

EC50 effective concentration 50%

FCS foetal calf serum

GTP guanosine triphosphate

HCV hepatitis C virus

HPRI human placenta RNase inhibitor

i.c. intracerebral(ly)

ICR Institute for Cancer Research (mice)

i.m. intramuscular(ly)

IMP inosine monophosphate

i.n. intranasal(ly)

i.p. intraperitoneal(ly)

JEV Japanese encephalitis virus

kDa kiloDalton

KUNV Kunjin virus

LGTV Langat virus

LIV Louping ill virus

LLC-MK2 Rhesus monkey kidney (cells)

3’ LSH 3’ long stable hairpin

MDP mean day of paralysis

MDD mean day of death

MEM minimum essential medium

MMLV Montana Myotis leukoencephalitis virus

MODV Modoc virus

M.O.I. multiplicity of infection

MVEV Murray Valley encephalitis virus

NGC New Guinea strain C (DENV)

NKV not known vector

NMRI Naval Medical Research Institute

NS non-structural protein

nt nucleotide

NTP nucleoside triphosphate

ORF open reading frame

PBS phosphate-buffered saline

PCR polymerase chain reaction

PCRU PCR unit

pfu plaque-forming units

poly(I).poly(C) polyinosinic-polycytidylic acid

POWV Powassan virus

(pr)M (precursor) membrane protein

RBV Rio Bravo virus

RCS repeated conserved sequence

RER rough endoplasmic reticulum

RNA ribonucleic acid

RT reverse transcription

s.c. subcutaneous(ly)

SCID severe combined immune deficiency (mice)

SLEV St. Louis encephalitis virus

TBEV tick-borne encephalitis virus

UTP uridine triphosphate

UTR untranslated region

Vero African green monkey kidney (cells)

WNV West Nile virus

wt wild type

YFV yellow fever virus

C h a p t e r 1

GENERAL INTRODUCTION

Chapter 1

Mosquito-borne

Not known vector

Tick-borne

MVEV

YFV

DENV-4

DENV-3 DENV-2 SLEV DENV-1

WNV KUNV

JEV

LGTV

POWV

TBEV

DBV

1. Flaviviridae

With nearly 80 pathogens in its ranks, the family of the Flaviviridae

contains several viruses that cause disease in humans. Among these is yellow

fever virus, the type virus of the Flaviviridae, from which the family derived

its name (flavus in Latin means "yellow").

Flaviviridae have been subdivided into three genera: Flavivirus,

Pestivirus and Hepacivirus. The Flavivirus genus contains several human

pathogens including yellow fever virus (YFV), dengue virus (DENV), and

Japanese encephalitis virus (JEV). The Pestivirus genus is home to bovine

viral diarrhea, classical swine fever virus and border disease virus of sheep,

but not known human pathogens. The genus Hepacivirus consists of hepatitis

C virus. Hepatitis G virus (also known as GB virus C) is classified within the

family of the Flaviviridae but has not (yet) been assigned to a genus.

2. Flavivirus

The genus Flavivirus contains (i) viruses that are transmitted by

mosquitoes or ticks (arthropod-borne) and (ii) viruses with not known

vector (NKV) (Chambers et al., 1990) (Fig. 1).

Fig. 1. Phylogenetic organization of the Flavivirus genus [abbreviations: yellow fever virus (YFV), dengue virus (DENV), St. Louis encephalitis virus (SLEV), West Nile virus (WNV), Kunjin virus (KUNV), Murray Valley encephalitis virus (MVEV), Japanese ence-phalitis virus (JEV), Powassan virus (POWV), Langat virus (LGTV), tick-borne encephalitis virus (TBEV), Dakar bat virus (DBV)]

Introduction

All flaviviruses of human importance belong to the arthropod-borne

group. The NKV-group holds a few viruses which have been isolated from

mice or bats and for which no arthropod-borne or natural route of

transmission has (yet) been demonstrated (Kuno et al., 1998).

One of the most important flaviviruses causing human disease is

dengue virus (DENV). It is estimated that there are worldwide annually as

many as 50 to 100 million cases of dengue fever and several hundred

thousand cases of dengue hemorrhagic fever (DHF), the latter with an

overall case fatality rate of ~5% (Gubler, 1997; Monath, 1994). The dengue

group of viruses includes four serotypes that exhibit a high level of genome

sequence homology and similar envelope protein antigenic properties (Blok

et al., 1992; Calisher et al., 1989; Kuno et al., 1998; Zanotto et al., 1996; World

Health Organization, 1997). Each virus can cause a primary infection in

humans, classically known as dengue fever and characterized by a nonfatal

febrile illness of variable severity, usually in older children or adults (Burke

& Monath, 2001). Infection with a single serotype confers protection against

reinfection with homologous strains but does not provide long-lasting cross-

protection against heterologous strains. Severe forms of dengue virus

infection [DHF and dengue shock syndrome (DSS)] are believed to result

from sequential infections with heterologous serotypes associated with

antibody-dependent enhancement of infection mediated by cross-reactive,

non-neutralizing antibody (Halstead & O'Rourke, 1977; Halstead, 1979;

Kliks et al., 1989). International travel and uncontrolled urbanization have

resulted in an increased spread of the mosquito vectors (Aedes aegypti and

Aedes albopictus). Infections in most tropical and subtropical regions now are

hyperendemic (prevalence of two or more serotypes), which enhances the

occurrence of DHF and DSS (Gubler, 1998).

The development of a vaccine for dengue viruses is a major public

health priority but remains problematic because of the need to induce

equally high levels of neutralizing antibodies against all four serotypes. A

Chapter 1

vaccine that is able to achieve this result is expected to provide uniform

protection against all serotypes and a low risk of antibody-dependent

enhancement. A variety of vaccine approaches have been undertaken,

including empirically derived and cDNA-derived live attenuated viruses,

recombinant subunit vaccines, inactivated virus vaccines, and DNA

vaccines. Although some candidates have progressed to clinical trials, there

have been problems with immunogenicity and reactivity of certain vaccines,

and it is not yet known which modality will be most suitable for use in

humans (Bhamarapravati et al., 1987; Bhamarapravati & Sutee, 2000; Bray et

al., 1996; Edelman et al., 1994; Guirakhoo et al., 2000, 2001; Hoke et al.,

1990; Huang et al., 2000; Kelly et al., 2000; Kochel et al., 2000; Konishi et al.,

2000; Porter et al., 1998; Putnak et al., 1996a, 1996b; Simmons et al., 1998,

2001). The strategy to develop chimeric flavivirus vaccines, which is

currently perhaps the approach that may be most successful, will be

discussed in section 7.

Despite the availability of a very efficacious vaccine, yellow fever

(YF) is still a major public health problem. The World Health Organization

estimates that there are annually 200,000 cases of YF, including 30,000

deaths, of which over 90% occur in Africa. Over the past 15 years, the

number of YF cases has increased tremendously, with most of the YFV

activity in West Africa. This increase in YFV activity is in part due to a

breakdown in YFV vaccination campaigns and mosquito control programs

(Mutebi & Barrett, 2002).

Japanese encephalitis (JE), a mosquito-borne arboviral infection, is

the leading cause of viral encephalitis in Asia (Hennessy et al., 1996).

Approximately 50,000 sporadic and epidemic cases of JE are reported

annually and result in high mortality (30%), while ~30% of the survivors

develop long-lasting neurological sequelae (Kalita & Misra, 1998; Misra et al.,

1998).

Introduction

Other important flaviviruses that cause encephalitis are also

responsible for high mortality rates or neurological sequelae. Tick-borne

encephalitis virus (TBEV) is believed to cause annually at least 11,000

human cases of encephalitis in Russia and about 3000 cases in the rest of

Europe. Related viruses within the same group, Louping ill virus (LIV),

Langat virus (LGTV) and Powassan virus (POWV), also cause human

encephalitis but rarely on an epidemic scale. Three other viruses within the

same group, Omsk hemorrhagic fever virus (OHFV), Kyasanur Forest

disease virus (KFDV) and Alkhurma virus (ALKV), are closely related to

the TBE complex viruses and tend to cause fatal hemorrhagic fevers rather

than encephalitis (Gritsun et al., 2003). Although the last large epidemic

caused by Murray Valley encephalitis virus (MVEV) occurred in 1974, new

cases of MVEV infection have been reported regularly, especially in Western

Australia (McCormack & Allworth, 2002). In 1996 an outbreak of West Nile

(WN) encephalitis with 373 cases and 17 deaths was reported in Romania

(Han et al., 1999; Tsai et al., 1998). In 1999, the disease appeared for the first

time in the northeastern United States and has continued to date to spread

across the United States and Canada. As of September 2003, there were

4156 total human cases reported in the USA, including 284 deaths (Nedry &

Mahon, 2003). Outbreaks of WN encephalitis occurred recently also in

Southern Russia and Israel (Lvov et al., 2000; Siegel-Itzkovich, 2000).

Although there are no recent reports of outbreaks or epidemics of St. Louis

encephalitis virus (SLEV), the virus causing this disease is endemic in the

western part of the United States and may account for a severe disease

(Kramer et al., 1997).

3. Pathology

The greatest body of information about flavivirus pathogenesis is

derived from experiments in mice and other laboratory rodents. These

animals provide a reasonably good model of flavivirus encephalitis but not

of other symptoms associated with human flavivirus infection (i.e. fever,

Chapter 1

arthralgia, rash and hemorrhagic fever). Viruses that produce these

syndromes in humans, including dengue fever and yellow fever, cause

encephalitic infections in laboratory rodents. The universal neurotropism of

flaviviruses in rodents and even in arthropod vectors (in which brain and

ganglia are major sites of replication) may reflect evolutionary conservation

of viral polypeptide structures involved in receptor interactions and of cell

membrane molecules that play a role in virus-receptor interactions.

Three patterns of pathogenesis have been described in flaviviral

encephalitis: (i) fatal encephalitis, usually preceded by early viremia and

extensive extraneural replication; (ii) subclinical encephalitis, usually

preceded by low viremia, late establishment of brain infection, and clearance

with minimal destructive pathology; and (iii) inapparent infection,

characterized by trace viremia, limited extraneural replication, and no

neuroinvasion (Johnson, 1982; Monath, 1986; Nathanson, 1980).

Pathological changes observed in humans and animals with flaviviral

encephalitis include (i) neuronal and glial damage directly caused by viral

replication and characterized by central chromatolysis, cytoplasmic

eosinophilia and cell shrinkage; (ii) inflammation, including perivascular

infiltration of small lymphocytes, plasma cells and macrophages; (iii) cellular

nodule formation composed of activated microglia and mononuclear cells;

and (iv) cerebral interstitial edema. At the ultrastructural level, infection of

neurons is characterized by marked proliferation and hypertrophy of rough

endoplasmic reticulum (RER), accumulation of vesicular structures derived

from the RER containing virus particles, and progressive degeneration of

the RER and Golgi apparatus (Hase, 1993a; Hase et al., 1990a).

4. Genome structure

The genome of the Flavivirus genus consists of a linear, positive-

sense, single-stranded RNA molecule of about 11 kilobases (kb) in length.

This RNA contains a methylated nucleotide cap (type I: m7G5’ppp5’A) at

Introduction

the 5' end and lacks a 3' polyadenylate tail. Surrounding the single open

reading frame (ORF) are 5’ and 3’ untranslated regions (UTRs) of about 100

nucleotides and 400 to 700 nucleotides, respectively (Fig. 2).

Fig. 2. Genomic organization of the Flaviviruses.

These regions contain conserved sequences and predicted RNA

structures that are likely to serve as cis-acting elements in the processes of

genome amplification, translation or packaging. Translation of the genome

results in synthesis of a single polyprotein precursor, that is co- and

posttranslationally processed by cellular proteases and a virally encoded

serine protease, into 10 polypeptides. The N-terminal one quarter of the

polyprotein encodes the structural proteins, and the remainder contains the

non-structural proteins, in the following order: C-prM-E-NS1-NS2A-NS2B-

NS3-NS4A-NS4B-NS5. The membrane topology on the endoplasmic

reticulum (ER) of the viral proteins is predicted based on the deduced

hydrophobicity (Fig. 3).

Fig. 3. Proposed topology of flavivirus polyprotein cleavage products with respect to the ER membrane. The proteins are drawn to scale (areas are proportional to the number of amino acids) and arranged in order (left to right) of their apparence in the polyprotein. Mature structural proteins are shaded and C-terminal membrane segments of M and E

are indicated. Cleavage sites for host signalase (♦), the viral serine protease (⇑), furin or

other Golgi-localized protease (♥), or unknown proteases (?) are indicated (Lindenbach & Rice, 2001).

C prM E NS1 NS2A 2B NS3 NS4A NS4B NS5 5’ UTR

Cap

Structural proteins Non-structural proteins

3’ UTR

Chapter 1

The prM protein which is precursor for the membrane (M) protein,

is, together with the envelope (E) and non-structural 1 (NS1) proteins,

translocated in the lumen of the ER, where their N-termini are generated by

signalase cleavages. The C-termini of the proteins remain anchored in the

membrane. The N-terminus of NS4B protein is also produced by the

signalase cleavage. The viral NS3 protease, which belongs to the family of

trypsin-like serine proteases, mediates, in combination with NS2B as

cofactor, the majority of other cleavages producing individual NS proteins,

as well as a short, mature form of C, which results from a cleavage upstream

from the hydrophobic signal for prM on the cytoplasmic side of the

membrane (Yamshchikov & Compans, 1993; Amberg & Rice, 1999) (Table 1).

Table 1

5' and 3' UTR, structural proteins, and non-structural proteins of members of the Flaviviruses and their function

Protein Flavivirus

5’ UTR 67-132 bases

Structural proteins

Cap type I m7Gpppm1NpN2

C nucleocapsid protein

prM precursor membrane glycoprotein

E envelope glycoprotein

Non-structural proteins

NS1 role in early RNA replication, secreted after glycosylation

NS2A function unknown; binds strongly to the 3' UTR, NS3, and NS5

NS2B cofactor of the NS3

NS3 serine protease; NTPase; helicase

NS4A function unknown;

binds strongly to most of the other non-structural proteins

NS4B function unknown

NS5 RNA-dependent RNA polymerase

3’ UTR 114-585 bases

Introduction

It is unclear which protease mediates cleavage at the NS1/NS2A

junction. The basic C protein interacts in the cytoplasm with progeny

genomic RNA to form the nucleocapsid, and immature virions are

produced by budding into the lumen of the ER, and are transported to the

cell surface through the exocytosis pathway. During this process, the prM, E

and NS1 are glycosylated. The prM protein is cleaved shortly prior to virus

release by cellular furin to produce the mature non-glycosylated M protein

present in extracellular virions (Stadler et al., 1997). It has been hypothesized

that the prM protein functions to protect the E protein from

conformational changes that could occur during passage of immature

virions through acidic cellular compartments (Lindenbach & Rice, 2001).

The prM protein folds very rapidly and is required for proper folding of E,

suggesting a chaperone-like role (Lorenz et al., 2002). Whereas most of the

flavivirus NS proteins are located intracellularly, the NS1 protein is secreted

from mammalian but not mosquito cells (Mason, 1989). The intracellular

NS1 protein is thought to play a role in viral RNA synthesis (Muylaert et al.,

1997; Lindenbach & Rice, 2001). The function of the extracellular, secreted

form of NS1 protein, is unknown.

5. Viral replication cycle

Binding and uptake of flaviviruses are believed to involve receptor-

mediated endocytosis via cellular receptors specific for the flavivirus

envelope proteins. Low pH-dependent fusion of the virion envelope with

the endosomal membrane delivers the nucleocapsid to the cytoplasm, where

the positive single-stranded RNA genome is uncoated. The 5’ UTR

containing the Cap structure, directs the RNA to the ribosomes, where

translation occurs. As mentioned in section 4, all viral proteins are produced

as part of a single polyprotein that is co- and posttranslationally cleaved in

combination of host and viral proteases. RNA replication occurs in

cytoplasmic replication complexes that are associated with perinuclear

membranes and occurs via synthesis of a genome-length minus-strand RNA

Chapter 1

intermediate. After replication, the viral genome is encapsidated in the

nucleocapsid proteins. Progeny virions are thought to assemble by budding

through intracellular membranes into cytoplasmic vesicles. The envelope

proteins become glycosylated in the host secretory pathway, i.e. the ER and

the Golgi apparatus. Finally, the vesicles fuse with the plasm membrane and

release mature virions into the extracellular compartment (Lindenbach &

Rice, 2001) (Fig. 4).

Fig. 4. Presumed replication cycle of flaviviruses. 1. binding; 2. receptor-mediated endocytosis; 3. membrane-fusion; 4. uncoating; 5. cap-mediated initiation of translation; 6. translation; 7. co- and posttranslational processing; 8. synthesis of template minus-strand and progeny plus-strand RNA; 9. encapsidation; 10. budding of virions in the ER; 11. transportation and maturation; 12. release of mature virions (Lindenbach & Rice, 2001).

6. Structure of the virion

The envelope of flaviviruses consists of a regular lattice formed by

two viral membrane proteins, namely and as outlined above, the major

envelope glycoprotein E (molecular mass, about 50 kDa) and the small

membrane protein M (molecular mass, about 7 kDa) (Lindenbach & Rice,

2001) (Fig. 5).

Introduction

The envelope glycoprotein E mediates virus entry into the cell via

receptor-mediated endocytosis, and carries the major antigenic epitopes

leading to a protective immune response (Heinz et al., 1994). The E

glycoprotein has two transmembrane segments at its carboxy terminus

linked by a short cytoplasmic loop. X-ray analysis of the ectodomain of E

revealed that the protein forms head-to-tail homodimers on the viral surface

(Rey et al., 1995). When exposed to low pH, the E proteins undergo an

irreversible rearrangement leading to dissociation of the dimers followed by

formation of trimers, and transitions which are apparently required for

fusion (Allison et al., 1995; Stiasny et al., 2001, 2002; Rey, 2003). Recent

experimental data provide evidence that the disulfide-linked loop at the tip

of the E protein functions as an internal fusion peptide (Allison et al., 2001).

Fig. 5. Envelope proteins of intracellular and extra-cellular flavivirus virions (Lindenbach & Rice, 2001).

The membrane protein M is synthesized as a precursor protein, prM

(molecular mass, about 26 kDa). It has been shown that prM has a

chaperone-like role in the folding and maturation of E (Konishi & Mason,

1993; Lorenz et al., 2002). Heterodimer formation between prM and E starts

soon after synthesis, a process that appears to be essential for E to reach its

final native conformation. The interaction between prM and E is also

important for later processing steps. It has been suggested that prM holds E

in an inactive conformation to prevent low-pH rearrangements during

transport through the acidic compartments of the trans-Golgi network

(Heinz & Allison, 2000a). Shortly before the virus is released from the cell,

Chapter 1

the pr portion is cleaved from prM by the cellular protease furin, leading to

mature virions consisting of E and M molecules (Stadler et al., 1997).

7. Chimeric viruses

Since there is no specific treatment for flavivirus infections, and

management of patients with the disease is often extremely problematic,

much emphasis has been placed on preventive vaccination. Vaccination is

currently possible against YFV, JEV and TBEV.

The wild type (wt) Asibi strain of YFV, isolated in 1927, was

attenuated by multiple passages in chicken embryo tissue, resulting in the

yellow fever virus 17D vaccine (YFV 17D). Only a single dose of the

vaccine is required to stimulate life-long protective immunity with an

efficacy of at least 99% (Monath, 1999a). Vaccination however is not

recommended for immunocompromised individuals on theoretical grounds,

although examples of successful vaccinations of HIV-infected persons have

been described (Receveur et al., 2000).

Vaccines available for the prevention of JE include both inactivated

and live-attenuated preparations. Although the formalin-inactivated vaccine

based on the wt Nakayama or Beijing-1 strains and produced in adult mouse

brains (JEV-Vax) is effective, the multiple-dose regimen, the relatively high

cost, and problems with immunogenicity complicate its use. In China, a

vaccine based on an attenuated JEV strain is used. This vaccine is prepared

from primary hamster kidney cell cultures and therefore not certified for use

in other parts of the world.

Inactivated cell culture-derived vaccines for TBE are currently used

in Russia and Europe (reviewed by Burke & Monath, 2001; Barrett et al.,

1999).

Recent introduction of the so called infectious clone technology has

opened new opportunities in flavivirus vaccine research. An infectious clone

Introduction

is a cDNA copy of an RNA virus genome that can be stably propagated in

bacterial plasmid vectors and from which RNA can be transcribed that is

infectious following transfection in the appropriate cells. Such infectious

clones offer the advantage that they allow easy manipulation of the genome

sequence. Infectious clones are now available for YFV 17D (Rice et al.,

1989), DENV-1, -2 and -4 (Rice et al., 1989; Lai et al., 1991; Kapoor et al.,

1995; Kinney et al., 1997; Polo et al., 1997; Gualano et al., 1998; Pur et al.,

2000), JEV (Sumiyoshi et al., 1992), TBEV (Mandl et al., 1997), LGTV

(Pletnev, 2001a), MVEV (Hurrelbrink et al., 1999), Kunjin virus (KUNV)

(Khromykh & Westaway, 1994), and WNV (Yamshchikov et al., 2001).

YFV 17D virus does not replicate in mosquitoes (although it does in

mosquito cells) and the post-vaccination viremia levels in humans are low,

which limits spread of the virus by mosquitoes in nature. These

characteristics make YFV 17D virus an attractive vector for engineering new

vaccines against other flaviviruses (e.g. the ChimeriVax vaccines, see below)

and possibly against HCV (Molenkamp et al., 2003). The ‘chimeric’ approach

to vaccine development is based on the recent observations that the

structural genes of one flavivirus can be replaced with homologous genes

from another flavivirus. This results in a viable chimeric virus that contains

an heterologous envelope.

The first chimeric flaviviruses were constructed by Bray and Lai,

who succeeded in the construction of the first intertypic DENV chimeras

(Bray & Lai, 1991). The entire structural region (C-prM-E genes) of the wt

DENV-4 (strain 814669) was successfully replaced with the corresponding

genes of DENV-1 (Western Pacific strain) and DENV-2 (New Guinea

strain C, NGC). With the information obtained from the first experiments,

new intertypic dengue virus chimeras were constructed by the same research

group: (i) DENV-2/DENV-4 chimeras that contained the prM-E genes or

only the NS1 gene of the neurovirulent DENV-2 NGC substituting for the

corresponding genes of the wt DENV-4 (strain 814669) (Bray et al., 1998),

Chapter 1

(ii) DENV-3/DENV-4 chimeras in which the C-prM-E genes of the wt

DENV-4 (strain 814669) were replaced with the corresponding genes of

DENV-3 (Chen et al., 1995). The chimeras displayed the expected serologic

specificities, and elicited protective levels of neutralizing antibodies in

monkeys (Bray et al., 1996). Intracerebral (i.c.) inoculation with DENV-2(C-

prM-E)/DENV-4 proved lethal for 3-day old outbred Swiss mice, whereas

DENV-2(NS1)/DENV-4 was not, suggesting that the major determinants

responsible for DENV-2 murine neurovirulence are located in the DENV-2

C-prM-E genes (Bray & Lai, 1991; Bray et al., 1998). This was also the first

demonstration that a tetravalent dengue vaccine composed of chimeric

viruses could be developed based on the genetic background of one

flavivirus. To achieve a desirable level of attenuation for vaccine candidates,

the wt DENV-4 backbone could be genetically modified by introduction of

non-lethal deletions in the 5’ and 3’ UTRs (Cahour et al., 1995; Men et al.,

1996; Durbin et al., 2001) or point mutations in the non-structural proteins

(Pethel et al., 1992). Huang and colleagues followed a similar approach to

develop intertypic chimeras as a tetravalent dengue vaccine. Based on an

earlier observation that attenuation markers of the DENV-2 vaccine

candidate PDK-53 virus are encoded by genetic loci located outside the

structural gene region of the PDK-53 virus, Butrapet and colleagues

expected that chimeric viruses containing the structural genes of other

DENV serotypes within the DENV-2 PDK-53 genetic background would

retain these phenotypic markers of attenuation (Butrapet et al., 2000; Huang

et al., 2000). The principal attenuation determinants of the PDK-53 strain

are a nucleotide change in the 5’ UTR and amino acid changes in NS1 and

NS3 that are responsible for small-plaque morphology and temperature-

sensitivity in simian (LLC-MK2) cells, reduced replication in mosquito cells,

and low neurovirulence in suckling mice. Several variants of DENV-1/

DENV-2 chimera containing the C-prM-E genes from the wt DENV-1

(strain 16007) or its attenuated derivative, PDK-13 virus, have been

reported, and their immunogenic potential has been demonstrated in mice.

Introduction

The remaining DENV-2/DENV-3 and DENV-2/DENV-4 chimeras are

being constructed (Huang et al., 2000).

The first chimera constructed by using genes from unrelated

flaviviruses was described by Pletnev and colleagues, who engineered a

chimeric TBEV/DENV-4 virus containing the prM and E genes of the Far

Eastern strain Sofjin of TBEV in the DENV-4 backbone described above

(Pletnev et al., 1992). The resulting chimeric virus grew more efficiently (i.e.

larger plaques and higher titers) in simian (LLC-MK2) cells, but less

efficiently in mosquito (C6/36) cells, as compared to the parental DENV-4.

TBEV/DENV-4 retained the neurovirulent properties of TBEV from

which it had acquired the prM and E genes, but was not neuroinvasive

when given to mice peripherally, thus suggesting that regions other than

prM and E are essential for the ability of this virus to spread from a

peripheral site to the brain (Pletnev et al., 1992; Pletnev et al., 1993). Since

TBEV/DENV-4 remained neurovirulent for mice, another chimera was

constructed that contained the prM and E genes from LGTV. LGTV strain

TP21 is the most attenuated of the tick-borne flaviviruses for humans, and

immunity to LGTV is cross-protective against TBEV (Pletnev & Men,

1998). In contrast to TBEV/DENV-4, the chimeric LGTV/DENV-4

replicated well in mosquito cells and was restricted in replication in simian

cell cultures. To develop the chimeras into vaccine candidates, they had to

be adapted to replicate efficiently in simian (Vero) cells, a satisfactory

substrate for human vaccine preparation (Pletnev et al., 2000). Neither the

LGTV/DENV-4 vaccine candidate nor the wt LGTV was able to infect

Toxorhynchites splendens, a mosquito species which is highly permissive for

dengue viruses (Pletnev et al., 2001b). Mouse neurovirulence of

LGTV/DENV-4 was significantly reduced as compared to the parental

LGTV and the TBEV/DENV-4 chimera, and lacked evidence of

neuroinvasiveness following intraperitoneal (i.p.) inoculation of Swiss or

SCID (severe combined immune deficiency) mice (Pletnev & Men, 1998;

Pletnev et al., 2000). Immunization of mice (Pletnev et al., 2000) and

Chapter 1

monkeys (Pletnev et al., 2001b) with LGTV/DENV-4 protected the animals

from subsequent challenge with highly virulent TBEV.

As discussed earlier, YFV 17D proved to be an attractive vector for

engineering new vaccines against other flaviviruses and plays a central role

in the ChimeriVax technology. The first YFV 17D chimeras were produced

by Chambers and colleagues, and contained the prM-E genes from a wt

JEV (Nakayama) or the SA14-14-2 strain (the Chinese live-attenuated JE

vaccine) (Chambers et al., 1999). In contrast to intertypic DENV chimeras,

attempts to generate viable YFV 17D chimeras incorporating the entire JEV

structural region (C-prM-E) could not be obtained, which was in accordance

with earlier observations during the construction of TBEV/DENV-4 and

LGTV/DENV-4 chimeras (Pletnev et al., 1992; Pletnev & Men, 1998). Like

the parental viruses JEV-N and YFV 17D, the JEV-N/YFV 17D chimera

was 100% neurovirulent in 3-to 4-week old ICR (Institute for Cancer

Research) and C57BL/6 mice. The JEV SA14-14-2/YFV 17D chimera was

unable to kill young adult ICR and C57BL/6 mice by both i.c. and i.p.

routes (Chambers et al., 1999). The latter chimera was selected as a primary

vaccine candidate designated ChimeriVax-JEV, since the virus (i) [inoculated

subcutaneously (s.c.) in mice] induced a solid protection against i.p.

challenge with a highly virulent strain of JEV and (ii) grew to high titers in

Vero (African green monkey kidney) cells that are certified for human

vaccine manufacturing (Guirakhoo et al., 1999).

Since RNA genomes of flaviviruses show high rates of mutation, it

is important to investigate which genetic determinants are responsible for

attenuation of neurovirulence. Arroyo and colleagues demonstrated that

multiple simultaneous reversions in distinct clusters of the E protein were

required for reversion of ChimeriVax-JEV to high virulence in mice (Arroyo

et al., 2001). ChimeriVax-JEV proved genetically stable and did not increase

its mouse neurovirulence following either 18 passages in cell culture or 6

mouse brain-to-brain passages. This reinforce the potential of ChimeriVax-

Introduction

JEV as a vaccine candidate for JEV (Guirakhoo et al., 1999). Furthermore,

similar to YFV 17D, and in contrast to the JEV SA14-14-2 parent,

ChimeriVax-JEV did not infect Aedes and Culex mosquitoes by oral feeding,

which indicated that spread of the chimera by mosquitoes is unlikely (Bhatt

et al., 2000). To investigate the safety and the ability to protect against a

subsequently challenge with a neurovirulent and neuroinvasive JEV,

ChimeriVax-JEV was extensively studied in rhesus monkeys. The chimeric

virus proved highly attenuated for this primate and less pathogenic than the

YFV 17D vaccine in all standard tests (Monath et al., 1999b, 2000).

ChimeriVax-JEV induced low, self-limited viremia following both i.c. and

s.c. inoculation, similar to viremias induced by YFV 17D. This is an

important feature that minimizes the possibility of neuroinvasion and

encephalitis in vaccines, and further reduces the chances of virus spread in

nature by feeding mosquitoes. A single dose of the chimera as low as 102

pfu (plaque-forming units) administered s.c. elicited high titers of JEV-

neutralizing antibodies. Animals receiving the vaccine were protected from a

severe i.c. challenge with a highly virulent wt JEV strain (Monath et al.,

1999b, 2000). In a randomized, double-blind, outpatient study the safety and

immunogenicity of ChimeriVax-JEV and YFV 17D was compared. S.c.

administration of a single dose of 104 or 105 pfu of ChimeriVax-JEV to

YFV-immune and naive volunteers caused no serious adverse events. The

rates of mild, transient injection site reactions and flu-like symptoms were

similar to control groups of subjects that received the YFV 17D vaccine.

Subjects inoculated with the chimera in both dose groups developed a

transient, low-titer viremia similar in magnitude and duration to that

following YFV 17D immunization. Neutralizing antibody seroconversion

rates to JEV were 100% in both high and low dose groups in both naive

and YFV-immune subjects. Mean JEV-neutralizing antibody responses were

higher in the high dose groups than in the low dose groups and also higher

in YFV-immune than in naive individuals (Monath et al., 2000).

Chapter 1

The first viable DENV-2/YFV 17D chimera (later designated

ChimeriVax-DENV-2) was constructed by insertion of the prM-E genes

from a wt strain of DENV-2 PUO-218 isolated from a patient in Thailand

(Guirakhoo et al., 2000). Two other laboratories constructed similar DENV-

2/YFV chimeras, based on the DENV-2 NGC strain (Caufour et al., 2001)

and strain PR-159/S1 (van der Most et al., 2000). Following their success

constructing a viable DENV-2/YFV 17D chimera, Guirakhoo and

colleagues developed three other chimeras, ChimeriVax-DENV-1, -3 and -4

(Guirakhoo et al., 2001, 2002) containing the prM-E genes from the wt

isolates DENV-1 PUO 359 (Thailand, 1980), DENV-3 PaH881/88

(Thailand, 1988), and DENV-4 1228 (Indonesia, 1978), respectively. Similar

to ChimeriVax-JEV, the ChimeriVax-DENV chimeras were not

neurovirulent in 3- to 4-week-old ICR mice, and significantly less

neurovirulent in suckling ICR mice than YFV 17D (Guirakhoo et al., 2000,

2001, 2002). All four ChimeriVax-DENV chimeras grew efficiently in a

vaccine-certified Vero cell line. All rhesus monkeys inoculated s.c. with a

single dose of these chimeric viruses (as mono- or tetravalent formulation),

developed viremia with magnitudes similar to that of the YFV 17D vaccine

strain but significantly lower than those of their parent DENV viruses.

Neutralizing antibody responses of the expected serotype specificities were

detectable in sera of immunized animals (Guirakhoo et al., 2000, 2001).

Animals immunized with the tetravalent DENV-1-2-3-4/YFV formulation

and 6 months later challenged with a second tetravalent dose, showed no

detectable viremia of the challenging viruses (Guirakhoo et al., 2001). In

monkeys immunized with the tetravalent formulation, the highest immune

response was directed toward ChimeriVax-DENV-2. This indicates that

appropriate formulation of the vaccine must be determined to achieve

uniform responses against all four serotypes. Following the observation that

(i) vaccination doses for the ChimeriVax-DENV-2 as low as 102 pfu were

effective, and (ii) immunization with ChimeriVax-DENV-2 in graded doses

ranging from 102 to 105 pfu/dose resulted in similar levels of DENV-2-

Introduction

neutralizing antibody titers on day 30, the dose of ChimeriVax-DENV-2

was lowered to 103 pfu in the tetravalent formulation (whereas doses of

ChimeriVax-DENV-1, -3 and -4 were 104.5, 103.6, 104.4, respectively). The

vaccine dose adjustment resulted in a more balanced immune response

against DENV-1, -2, and -3, but somewhat higher against DENV-4. Thus,

the immune response to the chimeric vaccine can be regulated by changing

the proportions of its components. Further studies will be necessary to

identify an optimal formulation of the four vaccine viruses. Administration

of a second tetravalent dose 2 months later to these animals increased

antibody titers to all four serotypes. Antibodies in sera from the immunized

animals efficiently neutralized various wt dengue isolates from different

geographic regions (Guirakhoo et al., 2002). The ChimeriVax-DENV-2

vaccine is currently being tested in a Phase I/II clinical trial (unpublished

data in Pugachev et al., 2003).

As outlined in section 2, an outbreak of WN encephalitis in New

York in the fall of 1999 and 2000 with 79 cases of laboratory-confirmed

WNV infection, 9 of which were fatal, received a lot of public attention

(Asnis et al., 2000; Briese et al., 1999). Chimeric flaviviruses expressing prM

and E of the WNV New York-1999 strain were constructed soon thereafter.

Monath and colleagues produced a chimeric WNV/YFV 17D based on the

ChimeriVax technology (Monath, 2001a; Monath et al., 2001b). This

chimeric virus is currently being tested in pre-clinical studies in monkeys.

Another chimeric WNV vaccine candidate, WNV/DENV-4, was recently

reported (Pletnev et al., 2002). The chimeric virus was highly attenuated in

neurovirulence and neuroinvasiveness in Swiss mice compared with its

WNV parent. Nonetheless, the WNV/DENV-4 chimera and a deletion

mutant derived from it were immunogenic and provided complete

protection against lethal WNV challenge (Pletnev et al., 2002).

Chapter 1

8. Neuroinvasion

The mechanism by which encephalitogenic flaviviruses cross the

blood-brain barrier and gain acces to the central nervous system (CNS) has

been debated for years and still remains largely unresolved (reviewed in

McMinn, 1997). Candidate routes for CNS invasion include, although not

exclusively: (i) infection of peripheral nerves or olfactory neurons

unprotected by the blood-brain barrier (McMinn et al., 1996a; Monath et al.,

1983); (ii) infection of vascular endothelial cells of capillaries in the brain,

transcytosis, and release of virus into the brain parenchyma (Liou & Hsu,

1998); and (iii) diffusion of virus between capillary endothelial cells into the

brain parenchyma (Kobiler et al., 1989; Lustig et al., 1992). A key factor for

neuroinvasion by these means is the capacity of neurotropic flaviviruses to

generate high titers and persistent viremia. Detailed analysis of the

pathology associated with SLEV neuroinvasion in hamsters revealed that,

following peripheral inoculation, the most likely mode of virus entry into

the CNS is the olfactory pathway, i.e. invasion of the olfactory epithelium

from the bloodstream followed by infection of olfactory neurons and entry

of the virus into the CNS (Monath et al., 1983; Burke & Monath, 2001).

Similarly, neuroinvasion by MVEV has been reported to occur via the

olfactory nerve (McMinn et al., 1996a). Although there are reports of

intranasal (i.n.) inoculation of mice and hamsters with WNV that result in

encephalitis (Nir et al., 1965; Goverdhan et al., 1992), Beasley and colleagues

noted that this is a rather inefficient process of which the characteristics

were influenced by mouse passage history of the virus strains studied. In

fact, neuroinvasion following i.p. inoculation was more efficient than

following i.n. inoculation (Beasley et al., 2002). These results suggest that

neuroinvasion via the olfactory nerve is not the prime mechanism of (WN)

virus entry into the brain in (the case of) WN encephalitis. Consistent with

this observation, Xiao and colleagues reported that WNV antigens could be

detected in the basal ganglia, brain stem, cerebellar cortex, and gray matter,

Introduction

but not within the olfactory bulb of hamsters inoculated i.p. with the WNV

USA99b strain (Xiao et al., 2001).

Viral determinants that are responsible for the neuroinvasiveness

properties of flaviviruses are even less well understood. The flavivirus major

envelope glycoprotein E is the predominant antigen to which neutralizing

antibodies and protective immune responses of the host are directed (van

der Most et al., 2002; Pletnev et al., 2001b; Guirakhoo et al., 2000, 2001;

Chambers et al., 2003; Zhao et al., 2003; Kroeger & McMinn, 2002; Monath

et al., 1999b; Velzing et al., 1999; Konishi et al., 1998; Simmons et al., 1998;

Seif et al., 1996; Lin & Wu, 2003; Butrapet et al., 1998). The E protein is also

the putative cell-receptor-binding protein and mediator of membrane fusion

and cell entry (Stiasny et al., 2002; Martinez-Barragan & del Angel, 2001;

Crill & Roehrig, 2001; Mandl et al., 2001; Bhardwaj et al., 2001; Kroschewski

et al., 2003; Despres et al., 1993; Guirakhoo et al., 1992). Several studies have

suggested that, in particular the E protein, plays a dominant role as a

determinant of flavivirus neurovirulence, since single amino acid

substitutions in the E protein were shown to cause dramatic effects on

neurovirulence (Monath et al., 2002; Holbrook et al., 2003; Lobigs et al.,

1990; Holzmann et al., 1990, 1997; Cecilia & Gould, 1991; Hasegawa et al.,

1992; Pletnev et al., 1993; Sumiyoshi et al., 1992, 1995; McMinn et al., 1995a,

1996a, 1996b). It is not well understood however how these various

mutations alter the functional properties of the flavivirus E protein so as to

increase the particular characteristics of neurovirulence.

Whether the determinants of neuroinvasiveness are also located in

the E protein is until this moment not ensured. Halevy and colleagues

showed that loss of neuroinvasiveness of WNV in ICR mice correlates with

the acquisition of an N-linked glycosylation site in the E protein (Halevy et

al., 1994). Later, however, it was shown that glycosylation does not directly

cause loss of neuroinvasiveness of WNV in mice; a second mutation (Leu68

→ Pro) within the envelope sequence would also be involved (Chambers et

Chapter 1

al., 1998). Also, Beasley and colleagues did not reveal a strong association

between the presence or absence of a potential glycosylation site and mouse

neuroinvasive phenotype. However, two neuroinvasive strains that lacked a

potential glycosylation site (EGY50 and SEN90) were about 10-fold less

lethal following i.p. inoculation than related neuroinvasive strains which

contained the glycosylation motif (Beasley et al., 2002). Several authors

demonstrated that loss of neuroinvasiveness can be achieved by introducing

specific amino acid substitutions within the E protein of other flaviviruses,

i.e. JEV, MVEV, LGTV, TBEV, and LIV (Cecilia & Gould, 1991;

Hasegawa et al., 1992; Sumiyoshi et al., 1995; McMinn et al., 1995a, 1995b,

1996a; Holzmann et al., 1990; Jiang et al., 1993; Pletnev et al., 1993; Pletnev,

2001a; Chambers et al., 1998; Mandl et al., 2000; Ryman et al., 1997a, 1997b,

1998; Bray et al., 1998). A revertant of WNV that regained the parental E

sequence, was only partially neuroinvasive. It was therefore concluded that

determinants outside the E region may also influence attenuation of

neuroinvasion. Whereas neurovirulent TBEV induces fatal encephalitis in a

large percentage of s.c. inoculated adult Swiss albino mice, attenuated

mutants (deletions of 4 to 21 amino acids in the capsid protein of TBEV)

did not cause disease but replicated and induced a specific antibody

response (Kofler et al., 2003). Two LGTV TP21 mutants exhibiting a

common change (Pro29 → Ser) in the non-structural NS2A protein showed

a reduced neuroinvasiveness in Swiss mice as compared to the parental

TP21 virus. Interestingly, both mutants did not differ from TP21 virus in

the amino acid sequence of their structural proteins (Campbell & Pletnev,

2000). A 320-nt deletion in the 3’ UTR of the attenuated LGTV E5 virus

resulted in a virus with reduced neuroinvasiveness in SCID mice as

compared to the progenitor E5 virus (Pletnev, 2001a). Also, a T10884 → C

substitution in the 3’ UTR of TBEV Vasilchenko (Vs) resulted in a

reduction of neuroinvasiveness in 3 to 4-week-old mice, as compared to the

original Vs virus (Gritsun et al., 2001). Ni and Barrett compared the genome

sequence of (i) a (for weanling mice) highly neurovirulent and neuroinvasive

Introduction

strain of JEV and (ii) two other neurovirulent, but not neuroinvasive, strains

of JEV and detected only mutations in NS2B, NS5 and 3’ UTR, suggesting

that these proteins and region are involved in neuroinvasiveness of JEV (Ni

& Barrett, 1996).

As discussed in section 7, chimeric flaviviruses, in which the prM

and E genes of one flavivirus have been replaced by the corresponding

genes of another flavivirus, have been constructed in attempts to develop

safe and effective vaccines against DENV, JEV and WNV (Chambers et al.,

1999; Guirakhoo et al., 1999, 2000, 2001; van der Most et al., 2000; Pugachev

et al., 2002; Sabchareon et al., 2002; Kanesa-thasan et al., 2001, 2002; Shlim &

Solomon, 2002; Monath, 2001a). Taken together, they provide usefull

information regarding the role of prM and E in the neuroinvasiveness in

mice. Pletnev and colleagues showed that a chimeric TBEV/DENV-4

(containing the envelope glycoproteins of TBEV), in contrast to DENV-4,

caused encephalitis in 6-week-old BALB/c mice following i.c. inoculation

and thus exhibited the neurovirulence phenotype of TBEV. However,

unlike the parental TBEV, this chimera was not pathogenic when young

adult BALB/c mice were inoculated i.p. or intramuscularly (i.m.), and,

hence, was not neuroinvasive (Pletnev et al., 1992). These findings suggest

that determinants other than prM and E may contribute to the

neuroinvasive properties of TBEV, or, alternatively, that the prM+E genes

of TBEV do not suffice to cause neuroinvasiveness in the context of the

DENV-4 genome. Similar observations were made with chimeric viruses

that consist of the DENV-4 “backbone” and the prM+E genes of LGTV.

LGTV (strain TP21) is neuroinvasive in Swiss mice and SCID mice. The

chimeric LGTV(TP21)/DENV-4 was neurovirulent, but not neuroinvasive

in SCID and Swiss mice (Pletnev & Men, 1998; Pletnev et al., 2000). Also,

chimeric WNV/DENV-4 viruses lost the neuroinvasive, but also the

neurovirulence properties of the neurovirulent WNV from which they had

obtained the prM+E genes, which is in contrast with the observation that

prM and E contain determinants of neurovirulence (Pletnev et al., 2002).

Chapter 1

Exchanging the prM and E genes of YFV 17D (a virus that is not

neuroinvasive in adult ICR mice) with the homologous genes of the

neuroinvasive JEV-Nakayama, did not result in a virus that was

neuroinvasive in adult ICR mice. Yet, in some (20%) adult C57BL/6 mice,

neuroinvasiveness was observed with this chimeric virus (Chambers et al.,

1999). The observed partial neuroinvasiveness is consistent with other

studies and indicate that the E protein of JEV contains determinants of

mouse neuroinvasiveness (Cecilia & Gould, 1991; Hasegawa et al., 1992).

Failure to observe neuroinvasiveness in ICR mice presumably indicates that

strain-specific differences exist in the susceptibility of young mice to the

chimeric JEV-N/YFV virus. Also, two mutations (His438 → Tyr in E

protein and Ala32 → Val in NS2B protein) in LGTV TP21 resulted in a

decrease in neuroinvasiveness for normal mice, but did not abrogate this

property completely when tested in highly permissive SCID mice (Campbell

& Pletnev, 2000).

Differences in neuroinvasiveness are not only mouse strain-

determined, but can also be influenced by (i) selective properties of the

blood-brain barrier, permitting some, and preventing other virus strains to

pass into the CNS; (ii) differences in the immunological response [e.g.

between suckling, young and adult animals or between immunocompetent

and immunodeficient mice; for example Pletnev and Men (Pletnev & Men,

1998) observed that SCID mice are 106- to 107-fold more permissive than

normal mice for detection of neuroinvasiveness]; (iii) and various levels of

virus multiplication in peripheral tissues resulting in viremias which are

either insufficient or too low to carry the virus into the brain (Albrecht,

1968).

Introduction

9. Transmission cycle of flaviviruses

As mentioned in section 2, flaviviruses are grouped into three

clusters: flaviviruses (i) transmitted by ticks, (ii) transmitted by mosquitoes

and (ii) flaviviruses that are not transmitted by a vector or that are

transmitted by a not yet known vector.

Fig. 6. Generalized transmission cycle of tick-borne flaviviruses, showing hosts for larval, nymphal, and adult ticks (Lindenbach & Rice, 2001).

Tick-borne flaviviruses are maintained in nature in a cycle involving

ticks and wild vertebrate hosts. Both male and female ticks are involved in

Chapter 1

transmission. Depending on the tick and environmental conditions, the life

cycle of a tick can range from a few months to two years. Each

developmental stage of a tick's life requires a blood meal in order to reach

the next stage. Virus is passed to succeeding tick stages during molting as

well as transovarially to progeny of adult ticks. TBEV (and possibly other

tick-borne flaviviruses) may be transmitted to uninfected ticks cofeeding on

a vertebrate host without the requirement for active viremic infection of the

host (Labuda et al., 1997). At least 10 species of rodents have been

implicated as TBEV-amplifying hosts (Gresikova & Beran, 1981; Gresikova

& Calisher, 1988). Insectivores (shrews, moles and hedgehogs, which have

relatively stable populations, in contrast to rodents), are believed to be

important reservoirs. Large mammals, such as goats, sheep and cattle, are

important blood-feeding hosts for adult Ixodid ticks but have low viremias

and are not considered to be important sources of tick infection. The virus

is excreted in the milk, however, and human infection may result from

consumption of unpasteurized goat or sheep milk or cheese (Fig. 6).

The general transmission cycle of mosquito-borne flaviviruses is

shown in Figure 7. Birds and mammals are effective viremic amplifying

hosts, serving as the source of infection of mosquito vectors. The main

epidemic vectors are Culex species. They breed in irrigated rice fields,

shallow ditches and pools. Humans are dead-end hosts and do not

participate in perpetuation of virus transmission. The maintenance of

mosquito-borne flaviviruses between epidemics has not been clearly

defined. Vertical transmission must be considered, i.e. the virus enters the

ovum in the female mosquito and survives there during the dry season. Less

than 1% of female mosquito progeny are infected. This is important for

survival of the virus during the dry season.

Introduction

Fig. 7. Generalized transmission cycle of mosquito-borne flaviviruses (Lindenbach & Rice, 2001).

YFV and DENV have, as compared to other mosquito-borne

flaviviruses, a different transmission pattern (Figure 8). The primary

transmission cycle (jungle or sylvatic cycle) involves wild nonhuman

primates and tree-hole breeding mosquitoes. It spreads from infected

mosquitoes to monkeys in the tropical rain forest. Infected monkeys can

then pass the virus onto other mosquitoes that feed on them. The infected

"wild" mosquitoes occasionally infect humans entering the forest (inter-

mediate cycle) and introducing the virus into areas with high human

Fig. 8. Transmission cycles of YFV and DENV (Lindenbach & Rice, 2001).

Or:

• Vertical transmission

• Hibernating vector

• Alternative vector (e.g. ticks)

Chapter 1

virus into areas with high human population density. Aedes aegypti

mosquitoes are domestic mosquitoes that breed in human-made containers.

If these mosquitoes have been infected by feeding on infected people, they

can cause an urban epidemic by infecting an unvaccinated human

population (urban cycle).

There exists thus far no insight into the viral determinants that are

responsible for vector-specificity, i.e. why do certain flaviviruses infect

mosquitoes and others ticks. Even more intriguing is why some flaviviruses,

i.e. those that belong to the NKV cluster, do not infect either ticks or

mosquitoes. This question is being addressed in Chapter 7.

10. Aims and objectives

The aim of this work is to establish a convenient model for the

study of flavivirus encephalitis. This model should resemble the course of

natural flavivirus infection and should be applicable for the study of

potential antiviral products against flavivirus infection, in particular those

that cause encephalitis. The virus used for this model should (i) be easy to

culture in vitro, (ii) cause 100% morbidity and mortality in small laboratory

animals, (iii) be safe to manipulate. Therefore we chose the bat flavivirus

Montana Myotis leukoencephalitis virus (MMLV).

The sequence of this virus [belonging to the biosafety level (BSL) 2

pathogens] has to be determined and analyzed. In particular, we should

search for the motifs that are present in those genes that are believed to be

interesting antiviral targets (such as the protease, helicase and polymerase).

Furthermore, the in vitro sensitivity to experimental potential anti-flavivirus

products should be compared with that of flaviviruses of human importance

(such as YFV). The pathogenesis of MMLV infection in mice will be

investigated, e.g. the time and ratio of paralysis and death, the image of brain

damage, localization of virus in the organs of the animal.

Introduction

If an antiviral product proves active against MMLV in vivo, it can

not be ensured that it will be active against other flaviviruses, such as YFV

or DENV. Therefore, a second approach to achieve our aim is through

construction of a chimeric virus. The new constructed virus should possess

the package characteristics of a flavivirus that is lethal to mice after i.p.

inoculation, and the replication machinery (and thus targets for antiviral

therapy) of a flavivirus that causes disease in man. By this approach we will

try to confer a flavivirus that is not infectious to mice, lethal to mice after

systemic inoculation.

The effect of exchanging the prM+E genes of a flavivirus of human

importance will be studied in vertebrate cells and mosquito cells.

Furthermore, the constructed chimeric viruses will be studied in mice.

Our third aim is to study the biology of NKV flaviviruses. For this

reason, an infectious full-length clone of a NKV flavivirus, which does not

yet exist, will be constructed.

C h a p t e r 2

INFECTION OF SCID MICE WITH MONTANA MYOTIS

LEUKOENCEPHALITIS VIRUS AS A MODEL FOR

FLAVIVIRUS ENCEPHALITIS

This chapter has been the subject of the following publication:

Charlier, N., Leyssen, P., Paeshuyse, J., Drosten, C., Schmitz, H., Van

Lommel, A., De Clercq, E. and Neyts, J. (2002). Infection of SCID mice

with Montana Myotis leukoencephalitis virus as a model for flavivirus

encephalitis. Journal of General Virology, 83, 1887-1896.

Chapter 2

SUMMARY

The bat flavivirus Montana Myotis leukoencephalitis virus (MMLV)

replicates well in Vero cells and appears to be equally sensitive as yellow

fever virus and dengue fever virus to a selection of (experimental) antiviral

agents. Cells infected with MMLV show dilatation of the RER, a

characteristic of flavivirus infection. MMLV particles have a size of

approximately 40 nm. Intraperitoneal, intranasal or direct intracerebral

inoculation of SCID mice with MMLV resulted in encephalitis ultimately

leading to death, whereas immunocompetent (NMRI) mice were refractory

to either i.n. or i.p. infection with MMLV. Viral RNA and/or antigens were

detected in the brain and serum of MMLV-infected SCID mice, but not in

any other organ examined: MMLV was detected in the olfactory lobes, the

cerebral cortex, the lymbic structures, the midbrain, cerebellum and medulla

oblongata. Infection was confined to neurons. Treatment with the interferon

α/β inducer poly(I).poly(C) protected SCID mice against MMLV-induced

morbidity and mortality, and this protection correlated with a reduction in

infectious virus titer and viral RNA load. This validates the MMLV model

for use in antiviral drug studies. The MMLV SCID model may, therefore, be

attractive for the study of chemoprophylactic or chemotherapeutic strategies

against flavivirus infections causing encephalitis.

INTRODUCTION

No specific antiviral therapy is available for flavivirus encephalitis.

One important reason is the lack of simple and convenient animal models.

Experimental infection of mice with JEV or TBEV by i.c. or peripheral

inoculation was reported to cause morbidity and mortality (Chiba et al.,

1999; Hase et al., 1990a). However, the study of these, for man, highly

pathogenic viruses [BSL3 for JEV, LIV, WNV and SLEV, and BSL4 for the

TBEV-complex, according to the American Committee on Arthropod-

MMLV encephalitis model

borne Viruses (ACAV); Subcommittee on Arbovirus Laboratory Safety

(SALS)] requires special laboratory facilities.

We here report a convenient model to study flavivirus encephalitis,

and the therapy thereof, using MMLV (BSL2 according to the ACAV;

SALS). MMLV was first isolated in 1958 from a mouse bitten in a

laboratory condition by a naturally infected little brown bat (Myotis lucifugus)

captured in Western Montana. The virus was subsequently isolated from

saliva, brain and various other tissues of other bats of the same species.

Serological studies suggested that the virus belongs to the flaviviruses (Bell

& Thomas, 1964). The virus was subsequently “forgotten” by the scientific

community. Based on both antigenic and molecular evidence [sequence of a

1 kb fragment of the genome (Kuno et al., 1998)], the virus was classified in

the Rio Bravo virus group within the genus Flavivirus (Heinz et al., 2000b).

The animal model that we present here may be particularly

convenient for the study of antiviral strategies against flavivirus infections,

in particular flavivirus encephalitis.

Chapter 2

MATERIALS AND METHODS

Virus propagation and RNA isolation

The original MMLV strain (Montana, 1958) was purchased from the

American Type Culture Collection (ATCC VR-537) and grown in Vero cells

at 37°C in minimum essential medium (MEM, Gibco, Paisley, Scotland)

supplemented with 10% inactivated foetal calf serum (FCS), 1% L-

glutamine and 0.3% bicarbonate. The viral RNA was extracted, using the

QIAamp Viral RNA kit (Qiagen, Hilten, Germany), according to the

Manufacturer’s instructions.

Animals

Eight- to 12-week-old SCID and NMRI (Naval Medical Research

Institute) mice (weighing 16-20 g) were used throughout the experiments.

All animals were bred at the Rega Institute under specific pathogen-free

conditions. Experiments on animals have been approved by, and are in

accordance with the guidelines, of the “Ethical Committee on Vertebrate

Animal Experiments” of the Katholieke Universiteit Leuven.

Compounds

Mycophenolic acid (MPA) was purchased from Sigma (St. Louis,

Missouri) and ribavirin [1-(β-D-ribofuranosyl)-1,2,4-triazole-3-carboxamide]

was from ICN (Costa Mesa, California). EICAR (5-ethynyl-1-β-D-

ribofuranosyl-imidazole-4-carboxamide) was synthesized by Dr. A. Matsuda

(Hokkaido University, Sapporo, Japan). Tiazofurin and selenazofurin were

provided by Dr. R. Cooney and Dr. D. G. Johns (National Institute of

Health, Bethesda, Maryland). Polyinosinic-polycytidylic acid [poly(I).poly(C)]

was purchased from Sigma-Aldrich (Sigma-Aldrich, Bornem, Belgium).


Plaque reduction assay

Serial dilutions of the test compounds were added to confluent Vero

cell cultures that were grown in 96-well microtiter trays. Cells were infected

with approximately 50 pfu of either YFV, DENV, or MMLV. Cultures were

further incubated at 37°C for 7 days, after which they were fixed with 70%

ethanol and stained with 2% Giemsa solution. Plaques were then counted

under an inverted microscope. The antiviral activity of compounds was

expressed as the EC50 (the effective concentration required to reduce plaque

formation by 50%).

Titration for infectious virus content

MMLV-infected mice were sacrified at various days after infection

by ether anesthesia. Brains were dissected aseptically, and 10% (w/v) tissue

homogenates were prepared in MEM supplemented with 2% FCS.

Confluent Vero cell cultures grown in 96-well trays were infected with 10-

fold serial dilutions of the tissue homogenates and incubated at 37°C for 1

hour, after which the inoculum was removed. Cultures were then washed

twice with warm medium and further incubated at 37°C for 7-9 days, after

which the cultures were evaluated for virus-induced cytopathic effect (CPE).

The titer was expressed in cell culture infective dose 50% (CCID50).

RT-PCR

Viral RNA was extracted from 140 µl of either the cell culture

supernatant or serum, lymphocytes or macrophages of infected animals

using the QIAamp Viral RNA kit (Qiagen). For the isolation of RNA from

cell cultures, cell debris was first spun down with two consecutive

centrifugations steps at 1500 x g. For the isolation of RNA from tissue

samples, the RNeasy Mini kit (Qiagen) was used. Prior to reverse

transcription (RT) 31.5 µl purified viral RNA was mixed with 10 µl 5 x RT-

buffer (Amersham Pharmacia Biotech, Roosendaal, the Netherlands), 0.5

Chapter 2

mM each of dATP, dTTP, dGTP, dCTP and 1.2 µM of antisense primer

(5’-GGGAGGCTGAGTAAGCATACACAAGC-3’, nt-position 3834-

3859) and denaturated at 92°C for 1 minute followed by chilling on ice. RT

reaction mixtures contained this denatured RNA plus 95 units of Human

Placenta RNase Inhibitor (HPRI, Amersham Pharmacia Biotech), 2.4 units

of RAV-2 reverse transcriptase (Amersham Pharmacia Biotech) and 1.35 µl

RNase-free water in a final volume of 50 µl. The reaction mixture was

incubated at 45°C for 1.5 hours. The polymerase chain reaction (PCR)

mixture was prepared as follows: 35.3 µl DNase-free water, 5 µl cDNA, 5 µl

10 x PCR-buffer, 2 µl dNTP-mix (100 µM of each dNTP) and 1.2 µM of

each primer (sense primer 5'-CGAGGTCAAAAACCCAAGGAGG-3' and

antisense primer 5'-CCACAGCATGGGCTCAGATGTT-3') were mixed

with 3.5 units of Super Taq (HT Biotechnology, Cambridge, United

Kingdom). The following PCR program was run: 10 min at 95°C, 30 cycles

of 30 s at 94°C, 30 s at 60°C, 1 min at 72°C and, for the final extension, 10

min at 72°C. The primers are specific MMLV primers that anneal in the

NS1-NS2A genes (sense primer at nt-position 3281-3302 and antisense

primer at nt-position 3639-3660).

Semi-quantitative RT-PCR

A calculated amount of 5 ng of total RNA isolated from tissue

fragments was used in a 50 µl semiquantitive RT-PCR assay using the

OneStep RT-PCR kit (Qiagen). GAPDH (glyceraldehyde phosphate

dehydrogenase) served as an internal control and a primer ratio

MMLV/GAPDH of 2/1 was used. Amplification primers were GAPDH

sense primer: 5'-GGTGAAGGTCGGTGTGAACG-3'; GAPDH antisense

primer: 5'-ATGTTCTGGACAACCCCACG-3'; MMLV sense primer: 5'-

GCACGAAGCTGCTGAGAGTGCC-3'; MMLV antisense primer: 5'-

TGAAAGCTGTGTAGGCGCTCCA-3'. After amplification (annealing at

60°C, 30 cycles) and separation of the amplicons (GAPDH 609 b, MMLV

978 b) by agarose gel electrophoresis, the gel was scanned and the two


bands were quantified using the ImageMaster software (Hoeffer Pharmacia

Biotech, Uppsala, Sweden).

Quantification of MMLV RNA by 5'-nuclease real-time RT-PCR

RNA preparation was performed using the Qiagen Viral RNA kit

(Qiagen) according to the Manufacturer's instructions. For elution of RNA,

the columns were incubated with 50 µl of RNase-free water at 80°C.

Primers MMLVS1 (5'-TCCGTAGGAAGCGTTGGTGT-3') and

MMLVAs1 (5'-ATTCCATTACTTCTGCCACTTCGA-3') were used to

amplify a 170-bp segment in the 5’ UTR/C region of MMLV. For detection

of PCR products, probe MMP (5'-AAGAGCGACGTGGTTTCCAGCC-3')

was used. It was labeled with FAM (6-carboxy-fluorescein) at the 5' end and

TAMRA (6-carboxy-N,N,N',N'-tetramethylrhodamine) at the 3' end. The

probe was phosphorylated at its 3' end to prevent elongation during PCR.

RT-PCR was done using the Superscript One-Step RT-PCR System with

Platinum Taq (Life Technology, Karlsruhe, Germany). A 20 µl reaction

volume contained 10 µl of 2 x reaction buffer, 2.5 mM additional

magnesium sulfate, 300 nM each of primers MMLVS1 and MMLVAs1, 200

nM of probe MMP, 0.8 µg BSA (Sigma-Aldrich), and 0.4 µl of Superscript

Reverse Transcriptase/Platinum Taq Enzyme mix. Prepared RNA (2 µl) was

added to each reaction. Thermal cycling in a Light-Cycler (Roche Molecular

Biochemicals, Mannheim, Germany) involved RT at 45°C for 20 min,

denaturation at 95°C for 5 min, followed by 45 cycles of 95°C for 5 s and

57°C for 35 s (Wittwer et al., 1997).

In the course of PCR, probe MMP was digested by 5'-exonuclease

activity of Taq DNA polymerase when specifically annealed to a generated

MMLV PCR product (Holland et al., 1991). Probe-digestion liberated the

FAM from the TAMRA dye, causing an increase in FAM-specific

fluorescence during PCR (Livak et al., 1995). FAM fluorescence was

measured on detection channel F1 (530 nm) and divided by fluorescence

Chapter 2

measured on channel F2 (640 nm) for normalization. As a quantification

standard for real-time PCR, a strong positive MMLV stock was diluted in

negative human plasma prior to RNA preparation. The amount of one PCR

unit (PCRU) of MMLV RNA was defined to be the lowest possible dilution

step that could be amplified in five of five replicative reactions. The

quantification procedure using the Light-Cycler was previously described

(De Silva et al., 1998).

Animal experiments

NMRI mice or SCID mice were inoculated with 104 pfu of MMLV

via either the intraperitoneal (i.p., 200 µl), intracerebral (i.c., 50 µl) or

intranasal route (i.n., 20 µl). The animals were examined daily for signs of

morbidity and mortality. Statistical significance of differences in the mean

day of death (MDD) was assessed by means of the Student's t test and

differences in the number of survivals by means of the χ2 test with Yates’

correction. To study the progression of MMLV infection in NMRI and

SCID mice, the animals were infected i.p. with MMLV and two to four mice

were sacrified every three days. Blood samples were taken and total RNA

was extracted from the serum and subsequently used for RT-PCR. Brain,

salivary glands, lung, liver, kidney, pancreas, seminal vesicles and spleen

were removed and subjected to immunostaining.

In some experiments (as indicated in the text) half of the infected

animals were implanted on day 0 and on day 14 with a mini-osmotic pump

(Alzet Model 2002, Alza Corporation, California, USA) filled with ribavirin

at a concentration of 83.3 µg/µl, resulting in a continuous s.c. calculated

release of 50 mg/kg/day of ribavirin. Mice were monitored for 4 weeks

after virus challenge.


Electron microscopy

Vero cells that had been infected with MMLV and that exhibited an

extensive CPE were fixed in a buffered glutaraldehyde solution for 30 min,

washed with isotonic phosphate buffer, postfixed with 1% OsO4 solution

for 1 h, collected, dehydrated, and embedded in Dow epoxy resin for

electron microscopy. Sections (1 µm) stained with toluidin blue were

examined with a light microscope. Sample sections (50 nm) of areas of

interest were collected on grids, stained with uranyl acetate and lead citrate,

and examined with a Philips CM10 electron microscope.

Histopathology

Moribund animals with severe signs of paralysis were euthanized

with ether anesthesia and were transcardially perfused with 20 ml of a

buffered 4% formaldehyde solution for histology. Fixed tissue samples were

embedded in paraffin and further processed by using standard methods.

Preparation of digoxigenin-labeled cRNA

MMLV cDNA encompassing 655 nt of the NS5 region of the

MMLV genome (sense primer 5'-TCATCCAGAGAAGAATTG-3' and

antisense primer 5'-TTCAAAAAGCCAAAGCTTTCAAATTC-3') was

cloned into the transcription vector pGEM-T (Promega, Madison, USA).

To generate run-off transcripts, the plasmid was linearized with NotI

(Promega) for 1.5 h at 37°C. The linearized plasmid DNA was precipitated

with 3 M sodium acetate (pH 4.6) and 2.5 vol of ethanol. After

centrifugation, the pellet was washed in 70% ethanol and dissolved in

RNase-free water at a concentration of ~ 100 ng/µl.

Transcription reactions (20 µl) consisted of 2 µl water, 10 µl

linearized cDNA (1 µg), 4 µl 5 x transcription buffer (Promega), 2 µl 10 x

Dig Labeling Mix (Roche Diagnostics GmbH, Mannheim, Germany) and 2

Chapter 2

µl T7 RNA polymerase (Promega). The reaction was incubated for 2 h at

37°C. The cRNA derived by T7 polymerase in vitro transcription was

isolated on a Mini Quick Spin RNA column (Boehringer Mannheim,

Germany) and the eluate was frozen at -80°C until use.

In situ hybridisation

Paraffin embedded tissue was sectioned, hydrated, fixed,

denaturated and acetylated according to standard procedures (Breitschopf et

al., 1992). The sections were permeabilized by digestion with proteinase K

(100 µg/ml) (Boehringer Mannheim) for 20 min. Prehybridisation was at

60°C for 30 min. Hybridisation solution was prepared by adding 200 ng of

digoxigenin (DIG)-cRNA per ml of hybridisation buffer (EasyHyb, Roche

Diagnostics). The hybridisation solution (150 µl/section) was added, the

section was heated for 4 min at 95°C to denature the probe and

hybridisation was allowed to occur at 60°C for 4 to 6 h in a humidified

chamber. The sections were incubated in 2 x SSC (0.15 M NaCl, 0.015 M

sodium citrate) for 12 h and washed with 1 x SSC. Upon completion of the

hybridisation step, the sections were incubated in buffer solution containing

0.5% (wt/vol) blocking reagent (Boehringer Mannheim) at room

temperature for 15 min, followed by incubation with anti-DIG polyclonal

antiserum alkaline phosphatase conjugate (1:500 in blocking buffer) for 1 h

and incubation overnight with the NBT/BICP reagent (nitro blue

tetrazolium chloride/5-bromo-4-chloro-3-indolyl phosphate; Roche

Molecular Biochemicals).

Immunohistochemistry

Primary antiserum was produced in New Zealand rabbits after

immunization with a suspension of UV-inactivated MMLV and complete

Freund's adjuvans. Tissue sections were rehydrated and blocked with 2%

skimmed milk. The sections were then incubated with 200 µl of primary

rabbit antiserum (dilution 1:100) for 1 h at 37°C. The sections were rinsed in


phosphate-buffered saline (PBS) and incubated with horse radish

peroxidase-conjugated donkey anti-rabbit antibody diluted 1:500 in PBS for

30 min at 37°C. After rinsing, endogenous peroxidase activity was quenched

by incubation overnight in 70 ml buffer (0.1 M acetate and 0.15 ml of a 3%

H2O2 solution) containing 20 mg AEC (3-amino-9-ethyl-carbazole; Sigma-

Aldrich) in 5 ml N,N-dimethylformamide.

To monitor MMLV infection in the brain, sections were stained

with hematoxylin and eosin (H&E) according to standard procedures.

Chapter 2

RESULTS

2.1. Antiviral susceptibility

MMLV produces an obvious CPE in Vero cells. The susceptibility

of MMLV to a selection of antiviral agents could therefore be easily

monitored. We compared the inhibitory effects of ribavirin, EICAR (the 5-

ethynyl derivative of ribavirin) (De Clercq et al., 1991), tiazofurin,

selenazofurin (an oncolytic C-nucleoside), and MPA (an inhibitor of IMP-

dehydrogenase) on the replication of YFV, DENV and MMLV.

Table 1

Susceptibilities of MMLV, YFV and DENV to a selection of (experimental) antiviral agents

EC50 (µg/ml)*

Antiviral agent YFV DENV MMLV

Ribavirin 22 ± 14 38 ± 11 25 ± 8

EICAR 0.8 ± 0.6 1.6 ± 0.6 0.4 ± 0.2

Mycophenolic acid 0.06 ± 0.04 0.10 ± 0.06 0.03 ± 0.02

Thiazofurin 17 ± 9 73 ± 25 18 ± 7

Selenazofurin 2.6 ± 1.5 12 ± 5 5 ± 3

* Concentration required to reduce virus-induced plaque formation in Vero cells by 50%.

As shown in Table 1, the antiviral efficacy of the different

compounds against all three viruses (YFV, DENV, and MMLV) ranked as

follows: MPA > EICAR > selenazofurin > tiazofurin > ribavirin. The

susceptibility of MMLV to these compounds proved more or less

comparable to the susceptibility of the human flaviviruses YFV and DENV

to these drugs.


N

RER

2.2. Ultrastructural analysis

MMLV-infected Vero cell cultures were studied by means of

electron microscopy. At 7 days post infection, the subcellular pathology was

characterized, as compared to noninfected cells, by extensive proliferation

and dilatation of the RER (Fig. 1), which is a characteristic for cells infected

with flaviviruses (Chambers et al., 1990a; Westaway, 1987). The size of the

viral particles inside the RER was determined to be ~ 40 nm, which is very

similar to the size of human flaviviruses (40 - 60 nm) (Monath & Heinz,

1996).

Fig. 1. Ultrastructural analysis of Vero cells 7 days post infection, showing extensive proliferation and dilatation of the RER (scale bar = 0,5 µm) (arrow: viral particles; N: nucleus; RER: rough endoplasmic reticulum) (inset: detail of particles).

2.3. MMLV-induced encephalitis and mortality

The effect of different routes of inoculation with MMLV on

morbidity and mortality in respectively SCID and NMRI mice, was studied.

Direct i.c. inoculation of the virus (104 pfu), in both SCID and NMRI mice,

led to paralysis within 6 days and mortality within 9 days [Fig. 2A and 2B;

mean day of paralysis (MDP): 6.1 ± 0.3 (SCID) and 7.1 ± 0.6 (NMRI);

mean day of death (MDD): 10.0 ± 1.2 (SCID) and 9.6 ± 0.7 (NMRI); 10

NMRI and 15 SCID mice]. Morbidity was characterized by ruffled fur,

paralysis of the hind legs and a wasting syndrome. I.p. or i.n. infection of

SCID mice with MMLV resulted in morbidity [Fig. 2A; MDP: 19.8 ± 2.6

Chapter 2

0

5

10

15

20

25

30

i.c. i.p. i.n.D

ay

s p

os

t in

fec

tio

n

0

5

10

15

20

25

30

i.c. i.p. i.n.

Da

ys

po

st

infe

cti

on

A B

No paralysis

No paralysis

No mortality

No mortality

(i.p. route, > 50 animals), 20.2 ± 1.1 (i.n. route, > 15 animals)] and mortality

[MDD: 25.1 ± 2.0 (i.p. route, > 50 animals) and 21.6 ± 0.5 (i.n. route, > 15

animals)]. Immunocompetent NMRI mice, infected with MMLV via i.p. or

i.n. route, remained healthy throughout the experiment (Fig. 2B).

Fig. 2. Paralysis (□) and mortality (■) following i.c., i.p., or i.n. inoculation of SCID mice (A) or NMRI mice (B) with MMLV. Each group consisted of 10 to > 50 mice. The columns represent the mean day of paralysis (MDP) and mean day of death (MDD) with standard deviations (SD).

2.4. Viral RNA in blood samples and macrophages

Total RNA was extracted from the serum and white blood cells of

i.p. MMLV-infected SCID mice that exhibited obvious signs of paralysis.

Following RT-PCR [detection limit 1.2 fg of viral RNA (data not shown)],

viral RNA was detected in the serum starting at day 6 post infection. This

was confirmed by real-time quantitative RT-PCR. The level of viral RNA in

the serum increased continously until the animals succumbed (Fig. 3). No

viral RNA was detected in the white blood cell population of infected

animals.


Fig. 3. Detection of viral RNA in the serum of SCID mice infected with MMLV via the intraperitoneal route. Sera from two mice were collected every third day after infection until day 21 post infection and were pooled. Quantification of the viral RNA load in the serum was done in triplicate by TaqMan RT-PCR (SD below 15%). [PCRU = PCR units :

One PCR unit is defined to be the lowest template copy number detectable in three of three replicative reactions as determined by limiting dilution series]

2.5. Histopathology

In SCID mice that had been inoculated with MMLV via the i.p.

route, the virus was first detected in the brain (by means of in situ

hybridisation) on day 12 post infection (data not shown). In one (out of

eight) animal a massive degeneration of the hippocampus was seen on day 18

(data not shown). The brain of the seven other animals showed a less

pronounced degeneration. At the time of paralysis, viral RNA was detected

in the gray matter of the olfactory lobes, the pyriform cerebral cortex, the

temporal cerebral cortex, the lymbic structures, the midbrain structures

(thalamus, hypothalamus), and in the medulla oblongata. Viral RNA also was

detected in the cerebellum with infection of virtual all of the Purkinje cells.

Overall, MMLV RNA was present in the cytoplasm of infected cells and

appeared to be confined to neurons. Neuron dysfunction may therefore be

expected to be the cause of encephalitis and ultimately death. The pattern of

MMLV infection, as assessed by in situ hybridisation, was confirmed by

immunohistochemistry (Fig. 4).

161298235

54

14

3

0

0

1

10

100

1000

3 6 9 12 15 18 21

Days post infection

Lo

g P

CR

U

Chapter 2

G I

D E F

G A B C

H

A B C

D

3

2 1 F

H

E

2.6. Effect of treatment with an interferon-inducer on MMLV

replication and associated mortality

To validate the MMLV model for future antiviral studies, we sought

to assess whether a correlation exists between the protective effect on

morbidity and mortality of a therapeutic or prophylactic agent, on the one

hand, and an inhibition of viral replication, on the other hand. Treatment

with ribavirin (given as a continuous infusion at 50 mg/kg/day for 14 days)

did not protect mice from MMLV-induced morbidity and mortality, nor did

it cause a reduction in viral RNA in the infected organs (data not shown).

Fig. 4. Localization of MMLV RNA or antigens in the brain of SCID mice at the time of paralysis (day 18 post infection by i.p. route) by in situ hybridisation and immunohistochemistry. Neurons in the cortex [stained by in situ hybridisation (A) (100x), and by immunohistochemistry (B (100x) and C (40x))], in the bulbus olfactorius [stained by in situ hybridisation (D) (2,5x)], in the hippocampus (E) [stained by in situ hybridisation (1 (2,5x) and 2 (25x)), or by immunohistochemistry (3) (25x)], in the temporal cortex (stained by in situ hybridisation) (F) (10x), in the cerebellum (stained by in situ hybridisation) (G) (10x). Purkinje cells in the cerebellum (stained by immunohistochemistry) (H) (25x). Sagital section through the mouse brain with indication of the position at which the microphotographs were taken (I).


We therefore studied the effect of poly(I).poly(C), an inducer of interferon

α/β. To this end, SCID mice were treated with a single dose of

poly(I).poly(C) at 15 mg/kg and were infected i.p. with MMLV 24 h later.

Treatment with poly(I).poly(C) resulted in a significant reduction in the

number of animals developing signs of paralysis (2/6 as compared to 6/6 in

the untreated control group) (data not shown) and also a 16-day delay in

virus-induced mortality (Fig. 5C). When poly(I).poly(C) was given 2 h after

the infection, all protective activity was gone. In a parallel group of mice, the

infection was monitored by real-time quantitative (Fig. 5B) and semi-

quantitative RT-PCR (Fig. 5A) and by titration (Fig. 5B) for infectious virus

content in the brain of MMLV-infected SCID mice that had received either

mock- or poly(I).poly(C) pretreatment. Tissue samples were taken at a time

at which untreated control animals had developed signs of paralysis (i.e. 18

days post infection). Levels of viral RNA detected in the brain of infected

SCID mice were markedly reduced (1.2 x 104 PCRU as compared to 1.18 x

108 PCRU in the untreated control group) in those animals that had been

pretreated with poly(I).poly(C) (Fig. 5B). Also titration for infectious virus

content (Fig. 5B) showed an important decrease in viral load in the brain of

the pretreatment group (1.5 x 107 CCID50/g tissue as compared to 5 x 1011

CCID50/g tissue). Thus, the protective effect of poly(I).poly(C) on virus-

induced morbidity and mortality was reflected by a marked reduction in the

infectious virus titer and viral RNA load.

Chapter 2

0

10

20

30

40

50

60

VC pre post

Mean

Day o

f D

eath

C

*

*

Fig. 5. MMLV RNA detected in the brain of MMLV-infected SCID mice that were treated with poly(I).poly(C) either 24 h before infection (pretreatment) or starting at 2 h post infection (posttreatment). (A) Gel electrophoresis of the amplicons obtained by One-Step RT-PCR (VC: untreated control group; pre: pretreatment group; post: posttreatment group). (B) Quantitation of viral load RNA using TaqMan technology (black bar) or infectious virus content (gray bar). (C) Mean day of death (* = p<0.05 as compared to the untreated control group). Dashed line: day at which animals in a parallel group were sacrificed for the determination of infectious virus titer and RNA load as depicted in panels A and B.

VC pre post

MMLV

GAPDH

A

B

CC

ID50/g

o

r P

CR

U

VC pre post

1012

1010

108

106

104

102

100


DISCUSSION

Several flaviviruses cause life-threathening neurological illness in

man. There is currently no specific antiviral drug available for the treatment

of infections caused by these viruses. The search for treatment strategies has

been hampered, in part, by the absence of a convenient animal model.

Several flaviviruses are pathogenic and lethal in mice, but only following

direct i.c. inoculation. Other flaviviruses can cause morbidity and mortality

following peripheral inoculation; however, these viruses require special

safety conditions for manipulation. Flaviviruses, such as DENV, YFV or

JEV can infect monkeys, but because of the costs involved and the

restricted availibility of monkeys, the number of studies using these animals

is limited.

We here present a model for the study of antiviral strategies against

flavivirus encephalitis, employing MMLV. This virus is (i) highly pathogenic

to mice, (ii) classified by the Subcommittee on Arbovirus Laboratory Safety

(SALS) as a BSL2 pathogen (http://www.cdc.gov/od/ohs/biosfty/bmbl4/

bmbl4s71.htm), has (iii), as discussed in Chapter 3, the same overall

organisation of the genome as flaviviruses of clinical importance, and (iv)

contains in those genes that are considered to be interesting antiviral targets

[i.e. the NS3 which encodes an NTPase/helicase and serine protease and,

the NS5, encoding a RNA-dependent RNA polymerase] the same

conserved motifs as flaviviruses that are infectious to man.

MMLV can be readily propagated in cell culture and the virus has a

susceptibility to a selection of antiviral drugs, including ribavirin and its 5-

ethynyl analogue EICAR (Leyssen et al., 2000), that is comparable to that of

YFV and DENV which further points to the relevance of MMLV for the

study of antiviral strategies against flaviviruses.

MMLV is neuroinvasive in SCID mice, but not in immunocom-

petent mice, in which it is only neurovirulent. In SCID mice infected via the

Chapter 2

peripheral route with MMLV, the virus is detected in the brain and serum;

no viral antigens or viral RNA are detectable in solid organs (by means of

immunohistochemistry or in situ hybridisation) nor in macrophages or other

blood-borne cells (by means of RT-PCR) (data not shown). Viral titers in

serum of MMLV-infected SCID mice start to rise well before the virus is

detected in the brain. It may be assumed that this viremia is the result of

low-level viral replication in peripheral organs, a level that is however too

low to be detected by in situ hybridisation or immunohistochemistry. Viral

replication in these organs was also not detectable by means of RT-PCR or

titration for infectious virus content, since, even following extensive

perfusion of the animals, traces of blood resulted in false positive signals for

all organs. Also Halevy and colleagues concluded that WNV replicates in yet

unidentified target tissues following peripheral inoculation of SCID mice

with the virus (Halevy et al., 1994). Modoc virus (MODV), a flavivirus

isolated from the white-footed deer mouse (Johnson, 1967) is, like MMLV,

neuroinvasive in SCID mice, but replicates, unlike MMLV, in peripheral

organs such as the salivary glands, and spleen (Leyssen et al., 2001). In a

rather recent study, a neuroadapted strain of YFV 17D, derived from a

multiple mouse brain passaged virus, proved neuroinvasive in SCID mice.

Unlike MMLV, this virus exhibited an efficient growth in peripheral tissues

of SCID mice (Chambers & Nickells, 2001). JEV was besides from the

brain, also isolated from liver, and spleen of newborn mice infected via

transplacental transmission of i.p. infected pregnant mice (Mathur et al.,

1981). It remains largely unclear which determinants influence the

organtropism of flaviviruses in peripheral organs. As is the case for the

murine MMLV and WNV models, in patients with WNV encephalitis, virus

was not detected in major organs such as lungs, liver, spleen and kidneys

(Nash et al., 2001; Sampson et al., 2000; Shieh et al., 2000).

MMLV-induced encephalitis in SCID mice is characterized by the

presence of viral RNA in virtual all regions of the brain and infection is

clearly confined to neurons. We did not detect other cells of the CNS that


were infected with MMLV. In the brain of mice infected with JEV, major

cytopathological changes were observed in neurons and developing neurons

(Ogata et al., 1991; Hase et al., 1993a, 1993b; Wang et al., 1998). Also in

infection models in rodents with viruses such as MVEV, YFV and WNV

viral replication in the brain was mainly confined to neurons (Matthews et

al., 2000; Schlesinger et al., 1996; Weiner et al., 1970; Eldadah & Nathanson,

1967). Presence of virus in all regions of the brain has also been observed in

patients with JE (Johnson et al., 1985). In patients with WN

(meningo)encephalitis, virus was detectable in neurons throughout the brain

(Nash et al., 2001; Nichter et al., 2000; Sampson et al., 2000; Shieh et al., 2000)

and, as seen in the MMLV model, also in the medulla oblongata. TBEV was

detected in the thalamus, substantia nigra, and cerebellum of the brain of patients

with TBE (Mazlo & Szanto, 1978).

In the CNS of MMLV-infected SCID mice, no, or little, cell

infiltration was observed. Since SCID mice do not carry (functional) B- and

T-cells, infiltration of the brain by lymphocytes is not possible. Yet, SCID

mice do harbour functional macrophages as well as natural killer (NK) cells;

also no, or little, infiltration of these cells in the brain of infected animals

was observed. In WNV-infected mice two populations of inflammatory cells

were detected in the CNS: NK cells and cytotoxic T cells (Liu et al., 1989).

In normal mice, JEV infection caused prominent inflammatory changes

with leukocytic infiltration and perivascular cuffing (Hase et al., 1990b). The

fact that no, or little, inflammation was observed in the brain of MMLV-

infected SCID mice points to the fact that direct viral damage is the main

(or sole) reason for brain dysfunction and virus-induced mortality. This may

make the MMLV model particularly attractive for the study of antiviral

strategies against flaviviruses. Indeed, a protective effect in the MMLV

model will likely be solely attribuable to a direct inhibitory effect of the

compound on viral replication in the brain. If such a molecule would prove

sufficiently protective in the MMLV model, it may be interesting to study its

effect in animal (including monkey) models with more pathogenic

Chapter 2

flaviviruses, and in which also an inflammatory response is involved in the

pathology of the disease.

To validate the MMLV model for future use in antiviral drug

studies, we sought to prove that protection against MMLV disease

progression correlates with an inhibitory effect on viral replication.

Treatment with ribavirin had a weak inhibitory effect on the replication of

MMLV in vitro (as well as YFV and DENV) and did also not protect mice

from MMLV-induced morbidity and mortality; nor did it cause a reduction

in viral RNA in the brain. This is in line with the fact that ribavirin did not

result in any protective effect in rhesus monkeys infected with DENV-1

(Malinoski et al., 1990). Ribavirin has also never been shown to be protective

against flaviviruses in man. Ribavirin did therefore not prove useful to

validate the MMLV model for antiviral drug studies. For this reason we

made use of the interferon α/β inducer poly(I).poly(C). This molecule has

previously been shown to elicit a protective effect against JEV and MODV

infections in mice and/or monkeys (Leyssen et al., 2001; Harrington et al.,

1977; Singh & Postic, 1970; Worthington et al., 1973). The interferon

inducer delayed MMLV-induced mortality and reduced viral titers, which

further demonstrates that the MMLV model may be useful in both

prophylactic and therapeutic studies.

In conclusion, the MMLV-mouse model has several clinical,

histopathological and virological features reminiscent of flavivirus

infections, in particular flavivirus encephalitis, in man. Therefore, the

MMLV model may be valuable for the study of antiviral strategies against

infections with flaviviruses, in particular those causing encephalitis.

C h a p t e r 3

COMPLETE GENOME SEQUENCE OF MONTANA MYOTIS

LEUKOENCEPHALITIS VIRUS, PHYLOGENETIC ANALYSIS

AND COMPARATIVE STUDY OF THE 3’ UNTRANSLATED

REGION OF FLAVIVIRUSES WITH NOT KNOWN VECTOR


Charlier, N., Leyssen, P., Pleij, C.W.A., Lemey, P., Billoir, F., Van Laethem,

K., Vandamme, A.M., De Clercq, E., de Lamballerie, X. and Neyts, J.

(2002). Complete genome sequence of the Montana Myotis leukoencephalitis

virus, phylogenetic analysis and comparative study of the 3’ untranslated

region of flaviviruses with not known vector. Journal of General Virology, 83,

1875-1885.

Chapter 3

SUMMARY

To obtain more insight in the biology of Montana Myotis

leukoencephalitis virus (MMLV) and its relation to other flaviviruses we

determined the complete genome sequence of the virus. The complete

MMLV genome is 10,690 nucleotides long and encodes a putative

polyprotein of 3374 amino acids. The virus contains the same conserved

motifs in genes that are believed to be interesting antiviral targets

[NTPase/helicase, serine protease and RNA-dependent RNA polymerase]

as flaviviruses of clinical importance. Phylogenetic analysis of the entire

coding region has confirmed the classification of MMLV in the clade of the

NKV flaviviruses and within this clade to the Rio Bravo branch (both

viruses have the bat as their vertebrate host). We have provided for the first

time a comparative analysis of the RNA folding of the 3’ UTR of the NKV

flaviviruses [MODV, Rio Bravo virus (RBV) and Apoi virus (APOIV); in

addition to MMLV]. Structural elements in the 3’ UTR that are preserved

among other flaviviruses have been revealed, as well as elements that

distinguish the NKV from the mosquito- and tick-borne flaviviruses. In

particular, the pentanucleotide sequence 5’-CACAG-3’, which is conserved

in all mosquito- and tick-borne flaviviruses, is replaced by the sequence 5’-

C(C/U)(C/U)AG-3’ in the loop of the 3’ long stable hairpin (3’ LSH)

structure of all four NKV flaviviruses. The availability of this latter sequence

motif allows us to designate a virus as either a NKV or a vector-borne

flavivirus.

INTRODUCTION

In recent years, the genomic sequences of an increasing number of

flaviviruses have been determined. Almost all of the sequence data available,

however, are from mosquito-borne and tick-borne flaviviruses due to their

impact on human health. Billoir and colleagues recently sequenced the

coding region of the NKV flaviviruses APOIV and RBV (Billoir et al.,

Complete genome sequence of MMLV

2000). Our research group reported for the first time the full-length

sequence of a NKV flavivirus, i.e. Modoc virus (MODV) (Leyssen et al.,

2002). In this chapter we present the complete genome sequence (coding

and non-coding regions) of MMLV, which gives us the opportunity to

investigate the particular characteristics (including the phylogeny) of the

NKV cluster. We describe for the first time a detailed and comparative

study of the organization and the secondary structure of the 3’ UTRs of

MMLV and three other NKV flaviviruses [RBV, MODV and APOIV].

Furthermore, we report that the pentanucleotide sequence CACAG, which

is conserved in the 3’ UTR of all arboflaviviruses, is replaced by the

sequence C(C/U)(C/U)AG in NKV flaviviruses. The availability of the full-

length sequences of NKV flaviviruses opens new possibilities to gather

more information about the determinants of vector specificity in the

genome of arthropod-borne flaviviruses.

Chapter 3


Virus

The original MMLV strain (Montana, 1958) was obtained from the

American Type Culture Collection (ATCC VR-537) and grown in Vero

cells.

Generation of PCR fragments

Viral RNA was extracted from 140 µl of supernatant medium of

virus-infected cells, using the QIAamp Viral RNA kit (Qiagen). An RT

reaction was designed employing the antisense primer 5’-GGGTCTCCTCT

AACCTCTAG-3’. Three sets of degenerated primers were designed based

on the alignment of full-length genome sequences of different flaviviruses

(Table 1). Other primers were designed based on: (i) the sequence of the

amplicons generated by the first primer sets; and (ii) the sequence of a

fragment of MMLV of approximately 1 kb [nt 8929-9939; Kuno et al., 1998]

(Fig. 1). All PCR amplifications were achieved under standard conditions

using Taq polymerase (HT Biotechnology) and 30 cycles including long

polymerization steps (1-2 min depending on the expected size of the

amplicons).

Fig. 1. Strategy employed for the sequencing of the MMLV genome. The locations of PCR fragments (dotted lines), specific primers (solid arrows) and random primers (open arrows) are indicated, along with the genomic organization of the virus.

C prM E NS1 NS2A NS2B NS3 NS4A NS4B NS5 5’ UTR Cap 3’ UTR

1 kb

3’ RACE 5’ RACE


Table 1

Flaviviruses employed in the phylogenetic analysis

Virus Vector Strain GenBank no.

Apoi virus

Montana Myotis leukoencephalitis virus

Rio Bravo virus

Modoc virus

Langat virus

Powassan virus

Tick-borne encephalitis virus



Dengue virus type 1

Dengue virus type 2

Dengue virus type 3

Dengue virus type 4

Japanese encephalitis virus

Kunjin virus

Murray Valley encephalitis virus

St. Louis encephalitis virus

West Nile virus

Yellow fever virus

Cell fusing agent virus

NKV

NKV

NKV

NKV

T

T

T

T

T

M

M

M

M

M

M

M

M

M, T

M

�

-

MML

RiMAR

M544

TP21

LB

Neudoerfl

Hypr

Vasilchenko

Singapore

New Guinea

H87

-

JaOArS982

MRM61C

Australia

MSI-7

EG-101

17D

-

AF160193

AJ299445

AF144692

AJ242984

AF253419

L06436

U27495

U39292

AF069066

M87512

AF038403

M93130

M14931

M18370

D00246

NC 000943

AF160194

M12294

X03700

M91671

M: mosquito-borne; T: tick-borne; NKV: not known vector

� Isolated only from mosquito cells (Aedes aegypti).

Amplification of genomic termini

The genomic 5’ and 3’ ends of MMLV RNA were determined by

RACE (Rapid Amplification of cDNA Ends). Amplification of the 3’ UTR

was achieved by the following method. An RNA oligonucleotide (5’-AAG

GAAAAAAGCGGCCGCAAAAGGAAAA-3’) was ligated to the 3’ end in

a reaction mixture containing 50 µl total RNA (10 µg), 15 µl 10 x T4 RNA

ligase buffer (Roche Molecular Biochemicals), 150 units of T4 RNA ligase

(Roche Molecular Biochemicals), 6 µl RNA oligonucleotide (2 µg), 142.5 U

HPRI (Amersham Pharmacia Biotech) and 62.5 µl RNase-free water. Fifty

µl of the reaction product was denaturated at 65°C for 10 minutes (to

Chapter 3

remove secondary structures) in the presence of 150 pmol of antisense

primer (5’-TTTTCCTTTTGCGGCCGCTTTTTTCCTT-3') and 14 µl

RNase-free water and chilled on ice. For the RT reaction, the following

were then added: 20 µl 5 x RT-buffer (Amersham Pharmacia Biotech), 1

mM each of dATP, dTTP, dGTP, dCTP, 95 units of HPRI and 40 units of

RAV-2 reverse transcriptase (Amersham Pharmacia Biotech). The mixture

was incubated at 45°C for 1.5 h and immediately chilled on ice. The

resulting cDNA was used in a 50 µl PCR reaction with 60 pmol of each

primer (antisense primer 5’-TTTTCCTTTTGCGGCCGCTTTTTTCCTT-

3' and sense primer 5’-CAGCAGTTCCAGCCAACTGGGTT-3’). The 5’

RACE was performed with the GeneRacer kit (Invitrogen, Groningen, The

Netherlands).

Cloning and sequence analysis

PCR products were cloned into the TOPO Cloning vector

(Invitrogen) or the pGEM-T Vector System I (Promega) and One Shot

competent E. coli cells (Invitrogen) were used for transformation. The

cloned inserts were sequenced in a cycle sequencing reaction with

fluorescent dye terminators and analyzed using an ABI 373 automatic

sequencer (Applied Biosystems Division, Foster City, CA).

Prediction of RNA secondary structure

The RNA secondary structure of the 3’ UTRs of MMLV, MODV,

RBV and APOIV was analysed using the STAR program (Gultyaev et al.,

1995). After folding each of the four sequences separately, the resulting

structures were searched for common structural elements as shown by the

occurrence of covariations in the stem regions in one or more of the other

three sequences. These sequences, containing proven structures, were

excised and replaced by five non-pairing nucleotides. The shortened

sequences were then submitted to a new cycle of folding and this was

repeated until no further common elements were detected.


Alignments and phylogenetic analysis

The complete amino acid sequences of the flaviviruses listed in

Table 1, were aligned using the ClustalW (1.74) software (Monath &

Lipman, 1988) and default alignment parameters, and manually edited in

McClade (Maddison & Maddison, 1989). Conserved motifs allowed an

unambiguous control of validity for alignment as previously reported (Billoir

et al., 2000).

Genetic distances were estimated using maximum-likelyhood

calculation in TreePuzzle-5.0 (Strimmer & Von Haeseler, 1996) based on

the Blosum62 (Henikoff & Henikoff, 1992) model of substitution and

taking into account a rate heterogeneity among sites with a discrete gamma

distribution of eight categories. An unrooted phylogenetic tree based on the

inferred distance matrix was constructed with Neighbor in Phylip 3.5.

Bootstrap analysis was performed according to the following algorithm:

1000 replicates were generated by Seqboot (Phylip) and redirected with

Puzzleboot to TreePuzzle-5.0 where distance matrices were estimated.

These distance matrices were subsequently used to infer phylogenetic trees

in Neighbor (Phylip) and a bootstrap consensus tree was generated with

Consense (Phylip).

Accession numbers

The MMLV sequence has been submitted to EMBL (accession no.

AJ299445). The MODV sequence is available at accession no. AJ242984.

The nucleotide sequence of the 3’ UTR of APOIV and RBV are available at

accession no. AF452049 and AF452050.

Chapter 3

RESULTS

3.1. Genome and amino acid sequence

The entire genome of MMLV (EMBL accession no. AJ299445) is

10,690 nucleotides long and encodes one long ORF extending from the

AUG start codon at nt-position 109 to the first in-frame stop codon at nt-

position 10,231, thereby encoding a 3374 amino acid long polyprotein

(10,122 nucleotides). The ORF is flanked by 5’ and 3’ UTRs that are,

respectively, 108 and 457 nucleotides long.

Comparison of the amino acid sequence of MMLV with that of

other flaviviruses revealed the presence of homologous protease cleavage

sites (Table 2), internal signal sequences and transmembrane sequences (the

C-terminal domains of the C, M, E and NS4A proteins are hydrophobic).

Table 2

Predicted protease cleavage sites in MMLV as compared to another NKV virus (RBV), a mosquito-borne virus (YFV) and a tick-borne flavivirus (TBEV)

Cleavage site (protease)

Cleavage after aa

MMLV RBV YFV TBEV

VirC↓CTHD

AnchC↓prM

Pr↓M

M↓E

E↓NS1

NS1↓NS2A

NS2A↓NS2B

NS2B↓NS3

NS3↓NS4A

NS4A↓2K

2K↓NS4B

NS4B↓NS5

90

107

194

269

753

1107

1330

1463

2079

2200

2223

2477

RKKQR↓SAKTV

ALMVA↓MEIEQ

ERAKR↓SLVIQ

APNLA↓TNCVS

TGVMG↓DQGCV

GLVSA↓QNEMS

MRGQP↓SKRAT

DGKRR↓SLYLL

AEKRR↓SSVLT

QGMQR↓TQIDT

LLVFA↓NEMRW

SPGRR↓GLSLS

KKQRR↓GGTES

TGLMA↓MQVSQ

HRLKR↓SLSIT

APSYS↓TQCVN

TGVMG↓DHGCA

GLVYA↓GSMTA

HRGQR↓ATDYT

DATQR↓SIIVF

AQMRR↓SGVLL

EGMQR↓TQIDS

VTVVA↓NEMRL

RSDRR↓GIVTS

SRKRR↓SHDVL

LMTGG↓VTLVR

RRSRR↓AIDLP

GPAYS↓AHCIG

LGVGA↓DQGCA

SWVTA↓GEIHA

IFGRR↓SIPVN

RGARR↓SGDVL

AEGRR↓GAAEV

PGQQR↓SIQDN

SAVAA↓NELGM

KTGRR↓GSANG

RGKRR↓SATDW

GMTLA↓ATVRK

SRTRR↓SVLIP

APVYA↓SRCTH

LGVGA↓DVGCA

SMVVA↓DNGEL

HRGRR↓SFSEP

RSSRR↓SDLVF

ASGRR↓SFGDV

AGKQR↓SSDDN

GLVAA↓NEMGF

SGGRR↓GGSEG

Abbreviations: VirC, mature virion C protein; CTHD, C-terminal hydrophobic domain; AnchC, anchored C protein (mature virion C protein + CTHD) [data for YFV and TBEV were from Chambers et al. (1990a) and for RBV from Billoir et al. (2000)].


Table 3

Putative processing of viral polyproteins of MMLV as compared to another NKV virus (RBV), a mosquito-borne virus (YFV) and a tick-borne flavivirus

(TBEV)

Protein MMLV RBV YFV TBEV

5’ UTR 108 NA 118 132

VirC

CTHD

Pr

M

E

NS1

NS2A

NS2B

NS3

NS4A

2K

NS4B

NS5

270 (90)

51 (17)

261 (87)

225 (75)

1452 (484)

1062 (354)

669 (223)

399 (133)

1848 (616)

363 (121)

69 (23)

762 (254)

2694 (897)

264 (88)

51 (17)

261 (87)

225 (75)

1452 (484)

1059 (353)

687 (229)

390 (130)

1857 (619)

357 (119)

69 (23)

774 (258)

2691 (897)

309 (103)

54 (18)

267 (89)

225 (75)

1479 (493)

1044 (348)

693 (231)

381 (127)

1869 (623)

378 (126)

69 (23)

750 (250)

2715 (905)

297 (99)

51 (17)

267 (89)

225 (75)

1488 (496)

1044 (348)

705 (235)

390 (130)

1863 (621)

378 (126)

69 (23)

756 (252)

2709 (903)

3’ UTR 457 NA 511 767

Total ORF 10122 (3374) 10137 (3379) 10233 (3411) 10242 (3414)

Total genome 10690 NA 10862 11141

The putative length of the 5’ UTR and 3’ UTR and of the respective genes are listed. The number of amino acids in each gene product is given between parentheses.

NA, not available.

As is the case for APOIV and RBV, the mature MMLV virC, E and

NS4A genes are markedly shorter than the corresponding genes of

arthropod-borne flaviviruses (Table 3).

We identified sequence motifs in the MMLV virus NS3 sequence

that are known to be associated with RNA helicase activity (1748-DEAH-

1751) and NTP-binding activity (1661-GSGKT-1665) (Pletnev et al., 1990).

The locations of these motifs proved to be perfectly conserved among the

flaviviruses and identical to the motifs found in APOIV and RBV. The NS3

proteins of flaviviruses also possess the active components of a serine

Chapter 3

protease (Chambers et al., 1990b). The location of such sequences is also

conserved in the NS3 protein of MMLV (1506-EGSFHTMWHVTRG-

1518, 1532-WANITEDLISYNGG-1545, 1590-PLDFPPGTSGSPIITSSG-

1609).

The flavivirus NS5 protein (the largest of the flavivirus-encoded

proteins) encodes an RNA-dependent RNA polymerase and also contains a

putative methyltransferase domain (Koonin, 1993). A heptapeptide

sequence, containing the characteristic GDD sequence motif (Kamer &

Argos, 1984; Poch et al., 1989), is also conserved in the MMLV NS5 gene

(3136-SGDDCVV-3142). The same heptapeptide motif is also present in

APOIV and RBV.

3.2. Predicted cleavage sites in the MMLV polyprotein

The N-termini of the proteins were defined, based on the position

of the cleavage sites of other flaviviruses. Cleavage by the viral protease

generally occurs following two dibasic residues and before an amino acid

with a short side chain, whereas processing with a host protease occurs at

sites obeying the (-3,-1) rule (Von Heijne, 1984). Table 2 summarizes the

cleavage sites for the processing of the MMLV polyprotein. At the N-

termini of prM, E and NS1 of MMLV predicted signalase cleavage sites are

detected which are also contributed by the C-terminal hydrophobic regions

of anchored C, prM and E, respectively. A signal sequence also precedes the

N-terminus of NS4B, suggesting that this hydrophobic protein is processed

in association with endoplasmic membranes. The N-terminus of NS2A

follows a cleavage site defined by the sequence Val-X-Ala (X = Ser, Thr,

Gln, Asn, Asp) (Cammisa-Parks et al., 1992; Von Heijne, 1984). In case of

MMLV the sequence consists of Val-Ser-Ala.

Five of the flavivirus polyprotein cleavages take place after two basic

amino acids (either Lys-Arg or Arg-Arg or Arg-Lys) (Chambers et al.,

1990a): i.e. anchored C-virion C, NS2A-NS2B, NS2B-NS3, NS3-NS4A and


NS4B-NS5. In case of MMLV, an Arg-Arg sequence is present at the C-

termini of NS2B, NS3 and NS4B. For the other two cleavages (anchored C-

virion C and NS2A-NS2B), suitable dibasic sequences could not be

identified. For anchored C-virion C, we suggest a cleavage site following a

Gln-Arg pair. For NS2A-NS2B cleavage may take place immediately after

Gln-Pro (Gln is also present in the DENV-2 and DENV-4 NS2A-NS2B

cleavage site) (Mandl et al., 1998). These sites were chosen based on the

sequence alignment with the polyproteins of other flaviviruses (Table 1) and

on the notion that the dibasic sequences are usually flanked by amino acids

with short side-chains, most commonly Gly, Ser or Ala. The prM protein is

a glycoprotein precursor, that undergoes delayed cleavage to form M and

the N-terminal "pr" segment. Akin to all flavivirus sequences the N-

terminus of the M-protein of MMLV immediately follows a pair of basic

amino acids believed to represent a cleavage site for either a viral or a host

protease. The two amino acids are flanked by an amino acid with a short

side-chain (Chambers et al., 1990a).

3.3. Characteristics of the 5’- and 3’-terminal nucleotide sequences

The ORF of the flavivirus genome is flanked by short non-coding

regions, which may contain elements involved in the regulation of essential

functions such as translation, replication or encapsidation of the genome

(Cammisa-Parks et al., 1992). The 5’ UTR of MMLV is 108 nucleotides long.

The MMLV 3’ UTR contains 460 nucleotides. As in other flaviviruses, the 3’

UTR is not extended by a poly(A) tract. At each end of the genome, two

terminal nucleotides, which are conserved among members of the whole

Flavivirus genus, were detected, i.e. 5’-AG and CU-3’. Besides these

conserved terminal nucleotides, there is only one nucleotide sequence motif

conserved among the mosquito- and tick-borne flaviviruses described. This

conserved motif is a pentanucleotide sequence (5’-CACAG-3’) located

approximately 45-61 nucleotides from the 3’ terminus (Wengler & Castle,

1986). It is predicted that it is located on a side-loop of a conserved 3’-

Chapter 3

terminal secondary structure, suggesting that this motif would induce the

formation of a circular RNA molecule, which could be important during

replication or encapsidation (Chambers et al., 1990a; Khromykh et al., 2001).

From all 21 vector-borne flaviviruses (Table 4) that were analysed, only

MVEV had a different pentanucleotide sequence, i.e. an A to C change at

position 4 (CACCG). We confirmed the presence of this deviating

pentanucleotide sequence in the MVEV genome by sequencing this

particular area of the genome of this virus. An A to a C change was also

noted at position 4 of this pentanucleotide sequence in the genome of cell

fusing agent virus (CFAV). The second position of this pentanucleotide (at

an analogous position, i.e. within the loop of a 3’-terminal stem and loop

structure) was either a U or a C instead of an A for all four NKV

flaviviruses. APOIV had, in addition, a C to a U change at position 3. This

pentanucleotide motif thus allows to discriminate between NKV and

vector-borne flaviviruses.


Mosquito-borne

NKV

Tick-borne

Table 4

The conserved pentanucleotide-sequence (5'-CACAG-3') in the last 61 nucleotides of the 3' UTR of mosquito- and tick-borne flaviviruses is a

C(C/U)(C/U)AG sequence for NKV flaviviruses

Strain Sequence alignment Position

TBEV 263

TBEVNEU

TBEVHYPR

TBEVVAS

LGTV

POWV

CCCAGACAC AGAUAGUCUGACAA -GGA---

CCCAGACAC AGGUAGUCUGACAA -GGA---


CCCAGACAC AGACAGUCUGACAA -GGA---

CCUAGACAC AGAUAGUCUGAAAA -GGA---

UCCAGGCAC AGAUAGCCUGACAA -GGA---

11,092

11,092

10,786

10,878

10,894

10,790

YFV 17D-213

YFV 17DD

YFV neurotropic

YFV viscerotropic

YFV Trinidad

YFV 85-82H

DENV-1

DENV-2

DENV-3

DENV-4

LIV

WNV

KUNV

JEV

MVEV

GUGAG-CAC AGUUUGCUCAAGAA -UAA---






UCCAGGCAC AGAACGCCAGAAAA UGGAAU-



UCCAGGCAC AGAGCGCCGCAAGA UGGAUU-


CACGG-CAC AGUGCGCC-GACAU AGGUG--

CACGG-CAC AGUGCGCC-GACAA UGGUG--

ACUAGGCAC AGAGCGCCGAAGUA UGUA---

-CAAGGCAC CGAGCGCCGAACAC UGUG---

10,811

10,811

10,811

10,811

10,709

10,811

10,689

10,677

10,650

10,603

10,822

10,916

11,345

10,928

10,966

CFAV -CCUCCCAC CGUUAG--GGAGUU UUGA--- 10,651

MMLV

RBV

MODV

APOIV

--AAGACUC AGAUUGUCUCAUGA -------

U---G-CCC AGAUUGC---AUGC UGG----

UCAAG-CUC AGAUUG-CUUACUA UGUA---

GUAAUCCCU AGGUUGGAUUAAAA UUAUCCU

10,644

NA

10,552

NA

Chapter 3

3.4. Comparative study of the folding of the 3’-terminal sequences of

four NKV flaviviruses

Strong support was found for the presence of four different RNA

regions (designated I, II, III and IV) in the 3’ UTR of MMLV, MODV and

RBV, but not in the 3’ UTR of APOIV (Fig. 2). The latter is assumed not to

have a region I equivalent.

Region I of MMLV, RBV and MODV consists of a long hairpin

with a branching stem-loop structure. At the 3’ end of region I of MMLV,

RBV and MODV, a conserved motif of 22 nucleotides [5’-UUGUAAAUA

(C/A)UU(U/G)(G/A)GCCAGUCA-3’] (labelled in bold in Fig. 2) was

observed. MMLV and RBV contain exactly the same sequence, whereas

MODV has an A to C, a G to U and an A to G change at positions 10, 13

and 14 of this sequence, respectively. This motif is not present in APOIV

which lacks region I.

The region between the stop codon and region I of the 3’ UTRs of

NKV flaviviruses is variable in length and is most probably single-stranded.

Region II of the 3’ UTR is predicted to form a Y-shaped structure.

The 3’ arm of the Y structure contains the conserved sequence 2 (CS2) (5’-

G(A/U)CUAGAGGUUAGAGGAGACCC-3’), which was present in all

four NKV viruses studied. The 5’ hairpin on the main stem of region II

varies considerably in length for the four NKV viruses, in contrast with

other flaviviruses. Upstream from region II, a repeated structure (IIbis) is

formed in the 3’ UTR of APOIV, which contains the repeated conserved

sequence 2 (RCS2) (5’-GACUAG(A/C)GGUUAGAGGAGACCC-3’).

Except for the CS2 sequence, the primary structure of region IIbis is not

identical to the sequence of region II; however, the secondary structure is

well conserved.


Three very similar hairpins (a, b and c) are predicted between

regions I and II for MMLV and RBV, the two NKV flaviviruses that are

also, according to the phylogenetic analysis (see below), most related. These

consist of a stem-loop (b), flanked by two shorter stem-loops (a and c); loop

b may form a pseudoknot at the 5’ side. The existence of this pseudoknot is

not only supported by its prediction with the STAR program but also by the

presence of one covariation in each of the two stems of the pseudoknot.

Region III folds into a Y shape for all four NKV viruses studied.

The two loops of the Y-structure of region III of the 3’ UTRs are formed

by a conserved stretch of nucleotides. However, the sequences of the stems

carrying these loops are not conserved, but rather show a large number of

compensatory base changes, strongly supporting the proposed secondary

structure.

In region IV, the 3’-terminal nucleotides of the 3’ UTR of the NKV

flaviviruses form a 3’ LSH, which preserves its shape despite significant

differences in sequence. This 3’ LSH was calculated to fold in the genome

of the four NKV flaviviruses with a similar position of the conserved

C(C/U)(C/U)AG motif (45-61 nucleotides from the 3’ terminus). At the 5’

side of the 3’ LSH, a small stem-loop (belonging to region IV and probably

coaxially stacking with the long 3’-terminal hairpin) is calculated for the four

NKV flaviviruses.

Inspection of the 3’ UTRs revealed the existence of a 69-79

nucleotide long sequence motif (e.g. 5’-GCUUUUGCUCCCGCGUUU

UUCAAAUUGCCUCAUCUUGAAUGGGGGGCGGCGUGGAUAUA

UACUCCAGCC-3’ for MMLV) located approximately 50 nucleotides away

from the 3’ terminus and representing an inverted repeat of another

conserved sequence element located approximately 54 nucleotides from the

5’ terminus (including the last 40 nt of the 5’ UTR and the first 29

nucleotides of the coding region). This may suggest a role in genome

Chapter 3

circularization. A similar but much smaller cyclization sequence has been

observed for the tick-borne flaviviruses (Khromykh et al., 2001). The

predicted folding of the four regions in the 3’ UTR as reported here is

supported by the fact that a large number of covariant and semi-covariant

sites occur in base-paired regions.


Fig. 2. Proposed secondary structure of the 3’ UTR of four NKV flaviviruses. The four regions (labelled I to IV) are delineated by boxes. Conserved motifs are shown in bold and boxed. For MMLV and RBV, the predicted pseudoknot is shown by connecting boxes. For MMLV and MODV, possible stem-loops are connected by dotted lines.

3.5. Phylogenetic analysis

A phylogenetic analysis was performed using complete coding

sequences (Fig. 3). Compared with the other flaviviruses, CFAV showed a

Fig. 3. Phylogenetic tree constructed based on the complete coding region of 20 flaviviruses by the neighbor-joining method. Bootstrap statistical analysis was applied with 1000 bootstrap samples (only bootstrap values below 100% are marked). The main vectors or hosts from which viruses were isolated are indicated.

0.1

YFV

DENV-4

DENV-3

DENV-2 SLEV

DENV-1

WNV

KUNV

MVEV

JEV

LGTV

POWV

RBV

APOIV

MMLV

9

Mosquito-borne

Not known vector

Tick-borne

New world

Rodent

Aedes

Culex

Aedes

TBEVVAS TBEVNEU

TBEVHYPR

MODV

Old world Rodent

Chapter 3

similarity below 30% and was therefore unreliable as an outgroup for

phylogenetic analysis of the complete genome of the flaviviruses. An

unrooted phylogenetic tree including the complete ORF sequences of 19

flaviviruses, and constructed with the neighbor-joining method, was

supported by high bootstrap values (ranging from 99,2 to 100 %). Three

major branches were observed: (i) the mosquito-borne virus branch; (ii) the

tick-borne virus branch; and (iii) the NKV virus branch. This confirms the

presence of MMLV in the group of the NKV viruses, as predicted by Kuno

and colleagues using a 1 kb fragment in NS5 (Kuno et al., 1998). The

bootstrap value of 100% allows us to conclude that MMLV belongs to the

RBV branch.


DISCUSSION

We have determined the complete sequence of MMLV, a flavivirus

with not known vector. The analysis of specific amino acid or nucleotide

patterns and the phylogenetic reconstructions based on the entire flavivirus

polyprotein confirm the taxonomic assignment of MMLV to the NKV

flaviviruses as previously suggested by the analysis of a ~ 1 kb fragment of

the NS5 gene (Kuno et al., 1998).

Moreover, the maximum bootstrap value allowed us to conclude

that MMLV belongs to the RBV branch, which is consistent with the fact

that both viruses have the bat as their vertebrate host. APOIV and MODV

(both isolated from rodents) are located in two distinct evolutionary

branches, MODV being more closely related to MMLV and RBV than to

APOIV.

The deduced amino acid sequence of MMLV revealed conservation

of the main features of flaviviruses, i.e. cleavage and glycosylation sites of

virus-specific proteins, the presence of highly conserved motifs important

for protease, helicase, methyltransferase and RNA-dependent RNA

polymerase activity (Monath & Heinz, 1996). The fact that the genome of

MMLV has the same organization as flaviviruses that are infectious to

humans as well as the same conserved regions in genes that can be

considered as antiviral targets, further points to the relevance of this model

in antiviral studies (Chapter 2).

We have studied the particular characteristics of the 3’ UTR of

NKV flaviviruses and therefore also included the 3’ UTR secondary

structures of MODV, RBV and APOIV in our analysis. The 3’ UTR

structures of flaviviruses have previously been suggested to be organized

into two distinct regions: (i) the 3’-terminal core element (approximately

330-400 nt in length for mosquito- and tick-borne flaviviruses), which is,

within the different serogroups, highly conserved in its primary sequence

Chapter 3

and RNA folding pattern (Wallner et al., 1995; Proutski et al., 1999); and (ii)

the variable region, which is inserted between the core element and the

coding region of the genome. This distinction corresponds with functional

differences between these two regions (Proutski et al., 1999). It was

suggested that the variable region could possibly act as a spacer separating

the folded 3’ UTR structure from the rest of the genome (Blackwell &

Brinton, 1995). Mandl et al. (1998) showed that deletion mutants of TBEV,

which lack the entire variable region, replicate as efficiently as the wt virus in

cell culture and mice. Akin to the situation in mosquito- and tick-borne

flaviviruses, the NKV flaviviruses also contain a variable region in their 3’

UTR, and this region also varies substantially in length. Most of this

variability is probably due to deletions or duplications in the region

immediately following the NS5 stop codon, as suggested for the mosquito-

borne flaviviruses (Shurtleff et al., 2001). The core element, with its stems

and loops, would constitute specific binding sites recognized by the virus-

encoded replicase, cellular proteins or viral capsid proteins, and would play

an important role in virus-specific transcription, translation and

encapsidation (Mandl et al., 1998; Gritsun et al., 1997; Proutski et al., 1997b;

Blackwell & Brinton, 1995, 1997; Chen et al., 1997). Using deletion mutants

of DENV-4, Proutski and colleagues proposed two parts within the core

element: (i) the most 3’-terminal structures/sequences that would act as a

viral promoter critical for the initiation of minus-strand RNA synthesis; and

(ii) more 5’ proximal structures/sequences that may function as enhancers

of viral RNA replication (Proutski et al., 1999). MODV has the shortest 3’

UTR (366 nucleotides) of the NKV flaviviruses discussed here and of all

flaviviruses sequenced so far. The folding pattern of this virus contains

possibly the (almost) basic 3’ UTR pattern that is necessary for replication

of a NKV flavivirus.

The phylogenetic tree based on the UTRs of NKV flaviviruses

showed similar topology to those constructed from the coding regions (data

not shown) indicating that the genetic information in these regions reflects


the evolutionary history of MMLV and the other NKV flaviviruses. Analysis

of the folding of the 3’ UTR points to the relatedness of MMLV, MODV

and RBV through a common folding pattern. Folding of the 3’ UTR of

these three viruses revealed four conformationally conserved structural

elements (region I-IV), that are supported by compensatory mutations,

which is suggestive for their functional importance. Six of the eight loops

expose conserved sequence motifs. Furthermore, at the 3’ end of region I, a

conserved motif of 22 nucleotides [5’-UUGUAAAUA(C/A)UU(U/G)

(G/A)GCCAGUCA-3’] was observed. This motif has not been described

for the mosquito- or the tick-borne flaviviruses and may be a particular

characteristic of New World NKV flaviviruses (MMLV, MODV, RBV)

(Fig. 3). APOIV, which can be considered a NKV flavivirus of the Old

World, lacks region I and thus this motif. Phylogenetically, APOIV is the

most distantly positioned flavivirus within the NKV flavivirus cluster. The

particular characteristics of the secondary structure of the 3’ UTR of

APOIV (absence of region I and presence of a duplicated region II)

corroborates this observation. Interestingly, MMLV and RBV, which both

have the bat as their natural host, share a common pseudoknot structure

(located between region I and II). Moreover, the sequence of region I in the

3’ UTR of MMLV and RBV is very similar, whereas the stems contain

compensatory mutations.

Several characteristics of the 3’ UTRs of the NKV flaviviruses are

comparable with those of either mosquito-borne or tick-borne flaviviruses

or both. A characteristic feature similar for mosquito-borne (with the

exception of YFV), but not tick-borne flaviviruses is the presence of

duplicated conserved RNA sequences (called CS2 and RCS2) (Chambers et

al., 1990a; Proutski et al., 1997b). It was assumed that they play an important

role in initiating viral transcription as cis-acting signals, either by virtue of

their exact nucleotide sequence (Hahn et al., 1987; Mangada & Igarashi,

1997) or through the interaction of secondary RNA structures with cellular

proteins (Blackwell & Brinton, 1995). The CS2 sequence is present in

Chapter 3

region II of the 3’ UTR folding pattern of NKV flaviviruses and is located

in a loop as in the mosquito-borne flaviviruses. According to Proutski and

colleagues, a single copy of either CS2 may be sufficient for normal viral

replication of DENV-4 (Proutski et al., 1999). However, deletion of both

stem-loop structures containing CS2 and RCS2 sequences led to an inability

of the mutants to replicate in mammalian cells. As is the case for the

mosquito-borne flaviviruses, APOIV contains both CS2 and RCS2, whereas

MMLV, MODV and RBV carry only one such sequence. This would be in

line with the observation of Proutski et al. (1999) that only one CS2

sequence is required for the efficient replication of mosquito-borne

flaviviruses. The mosquito-borne and NKV flaviviruses thus share a

common factor that may possibly be important for replication in

mammalian cells.

The 3’ UTR of NKV flaviviruses share also particular characteristics

with tick-borne flaviviruses. Like the tick-borne viruses, NKV flaviviruses

lack the small stem-loop located in region I of the 3’ UTR of mosquito-

borne flaviviruses (Proutski et al., 1997b, 1999). Deletion of this structure

led to a reduced efficiency of replication of DENV-4 in mosquito cells. It

was suggested that this structure may function as an enhancer of virus

replication in mosquito cells. The fact that NKV flaviviruses probably do

not (or inefficiently) replicate in mosquito cells, may reinforce this

hypothesis.

Region III of the NKV flaviviruses folds into a structure similar to

the one predicted in the 3’ UTR of tick-borne flaviviruses [where it is part of

a larger structure with three different branches of hairpins (Mandl et al.,

1998: Proutski et al., 1997b)], but that is not present in the 3’ UTR of

mosquito-borne flaviviruses (Hahn et al., 1987; Shurtleff et al., 2001). As for

the tick-borne flaviviruses, the two loops of the Y structure of region III of

the NKV 3’ UTRs are formed by a conserved stretch of nucleotides. These

can be detected at analogous positions [5’-AUUGGC-3’ and 5’-


(G/U)(G/U)UU-3’] (Gritsun et al., 1997; Mandl et al., 1993; Proutski et al.,

1997a, 1997b).

The very 3’ terminus of the 3’ UTR (region IV) folds in a manner

typical for all flaviviruses, forming the 3’ LSH structure and a small stem-

loop (belonging to region IV and probably coaxially stacking with the long

3’ terminal hairpin). The pseudoknot which is predicted between the small

stem-loop and the 3’ LSH of mosquito-borne, but not tick-borne

flaviviruses, is probably not formed in the 3’ UTR of the NKV flaviviruses.

The fact that the 3’ LSH is detected in mosquito-, tick-borne (Grange et al.,

1985; Brinton et al., 1986; Hahn et al., 1987; Mohan & Padmanabhan, 1991;

Mandl et al., 1993; Wallner et al., 1995; Shi et al., 1996; Proutski et al., 1997a,

1997b) and NKV flaviviruses strongly suggests that it plays a crucial role in

the replication of all flaviviruses. In particular, the presence of the highly

conserved pentanucleotide 5’-CACAG-3’ in the top loop has been

suggested to play an important role in virus replication (Khromykh et al.,

2001). Analysis of the 3’ UTRs of the NKV flaviviruses revealed a

pentanucleotide sequence motif [5’-C(C/U)(C/U)AG-3’] at an analogous

position. However, this pentanucleotide sequence is, at position 2 (and 3),

different from the 5’-CACAG-3’ motif of mosquito- and tick-borne

flaviviruses and appears to be unique for NKV flaviviruses. Indeed, whereas

the conserved CACAG motif was detected in all flaviviruses analysed so far

[with the exception of MVEV (and CFAV), which carries an A/C change at

position 4], all four NKV flaviviruses analysed carry a C(C/U)(C/U)AG

pentanucleotide motif. From a taxonomic point of view, knowledge of the

sequence of this pentanucleotide may thus be sufficient to allocate a

flavivirus either to the vector-borne or to the NKV flaviviruses.

C h a p t e r 4

A RAPID AND CONVENIENT VARIANT OF FUSION-PCR TO

CONSTRUCT CHIMERIC FLAVIVIRUSES


Charlier, N., Molenkamp, R., Leyssen, P., Vandamme, A.M., De Clercq, E.,

Bredenbeek, P.J., and Neyts, J. (2003). A rapid and convenient variant of

fusion-PCR to construct chimeric flaviviruses. Journal of Virological Methods,

108(1), 67-74.

Chapter 4

SUMMARY

So far, full-length cDNAs of chimeric flaviviruses have been

constructed by restriction-enzyme cleavage of the gene(s) to be exchanged

or by fusion-PCR of two amplified PCR fragments. The construction of a

chimeric flavivirus by a faster and more convenient variant of the standard

fusion-PCR is reported. A chimeric MODV/YFV virus was engineered in

which the structural prM and E genes of YFV 17D were replaced by the

homologous genes of MODV. In two PCR steps, a fusion was made

between the 3’ end of the C gene of YFV and the 5’ end of the prM gene of

MODV, and between the 3’ end of the E gene of MODV and the 5’ end of

the NS1 gene of YFV. For each of the two fusions between YFV and

MODV, a standard PCR was performed to amplify a short fragment with

one overlapping end that could be used as one of the primers in the

subsequent (fusion) PCR.

INTRODUCTION

In a fusion-PCR, complementary oligodeoxyribonucleotide (oligo)

primers are used to generate two DNA fragments with overlapping ends in

a PCR reaction. These fragments are combined in a subsequent 'fusion'

reaction in which the overlapping ends anneal, allowing the 3' overlap of

each strand to serve as a primer for the 3' extension of the complementary

strand. The resulting fusion product is further amplified by PCR. Compared

to the restriction method, which requires the incorporation of mutations to

create unique restriction sites, the fusion-PCR strategy has the advantage

that the gene fusion can be made at any position within the nucleotide

sequence.

Here we report a fast and straightforward approach for the

construction of chimeric (flavi)viruses. Instead of three amplification

reactions with four primers as used in a “classical fusion reaction”, a fusion

between two different genes is accomplished by only two amplification

Fusion-PCR for the construction of chimeric viruses

reactions (and the use of three primers), thereby reducing the risk of

accumulating mutations. This strategy was employed to construct a

MODV/YFV chimeric virus. YFV 17D was used as a backbone in which

the prM+E genes were replaced with those of MODV.

Chapter 4


Unless otherwise indicated, all the buffers used for restriction

enzymes, ligases, DNA and RNA polymerases were provided by the

suppliers and used according to their specifications.

Cells, viruses and plasmids

Vero and BHK-21 (Baby Hamster Kidney) cells, and MODV were

originally obtained from the ATCC (CCL-81, CCL-10 and VR-415

respectively). MODV was grown in Vero cells (Leyssen et al., 2001). The

clone pACNR-FLYF17Da contains a full-length cDNA of YFV 17D and is

identical to pACNR-FLYF17Dx (Bredenbeek et al., 2003) except for the

XhoI transcription run-off site which was changed to an AflII site in

pACNR-FLYF17Da. Plasmid pHYF-5’ is a derivative of pHYF-5’3’IV

(Bredenbeek et al., 2003) and contains a NotI-MluI fragment encompassing

the Sp6 promoter fused directly to the 5’ 2947 bases of YFV 17D.

Amplification and cloning of MODV prM+E cDNA

MODV RNA was extracted from 140 µl of cell culture supernatant

using the QIAamp Viral RNA kit (Qiagen) according to the Manufacturer’s

instructions. The cDNA was synthesized and amplified using the One Step

RT-PCR kit (Qiagen): 5 µl MODV RNA was added to 10 µl 5 x RT-PCR

buffer, 0.4 mM dNTP, 2 µl enzyme-mix, 95 units of HPRI (Amersham

Pharmacia Biotech), 0.6 µM of each primer [sense primer, 5’-AAGGTTTT

GGAAGATGACTCCGGC-3’ (nt-position 271-294); antisense primer, 5’-

GTTAATGACTGGTATGGGGGGTACA-3’ (nt-position 2444-2468)]

and 29 µl RNase-free water in a final volume of 50 µl. The following

amplification program was used: an RT at 50°C for 30 min, an initial PCR

activation step of 15 min at 95°C followed by 30 cycles of 30 s at 94°C, 30 s

at 60°C and 2 min at 72°C and a final extension phase of 10 min at 72°C.

The DNA fragment with the expected length of 2.2 kb was cloned into a


TOPO vector using the TOPO TA Cloning kit (version H) (Invitrogen) and

One Shot TOP10 E. coli cells (Invitrogen) to yield pMODV-prM+E.

Primers

Primers used for the construction of the chimeric region were

designed based on the nucleotide sequence of YFV 17D (GenBank

accession number X03700) (Rice et al., 1985) and MODV (GenBank

accession number NC_003635) (Leyssen et al., 2002). The nucleotide

sequences of the primers are listed in Table 1.

Table 1

Primers used for fusion-PCR

Name Polarity Sequence

A

B

C

D

E

F

S

AS

S

S

AS

S

5’-GGAATGCTGTTGATGACGGGTGGAACCATATTGTCAATTGAAGTTGTT-3’

5’-GCACTGAGGATCCCAATGGAA-3’

5’-GCAACGCGGCGGCCGCGCTAGCGATGACC-3’

5’-CTTTCACCACAGGAGTGATGGGAGATCAAGGATGCGCCATCAACTT-3’

5’-GCCTAAATTCAGTTGACTCC-3’

5’-TGCCTGTTGGAAAAGGATCGT-3’

Y/M

M

Y (NotI)

M/Y

Y

M

Abbreviations: S: sense; AS: antisense; Y: YFV; M: MODV

Fusion-PCR

Two short fragments (205 bp and 209 bp, Fig. 1), that were to serve

as primers in the subsequent fusion-PCR, were amplified in a reaction mix

of 50 µl consisting of 100 ng (pHYF-5’ or pMODV-prM+E) plasmid, 2

units of Pfu DNA polymerase (Promega), 5 µl of 10 x buffer supplied by the

Manufacturer, 400 µM dNTP, and 1.2 µM of each of the two primers (Table

1, A+B and D+E respectively). The conditions for the amplification

reactions were as follows: 30 s at 95°C, 30 s at 50°C and 1 min at 72°C

repeated for 25 cycles.

Chapter 4

For the fusion-PCR, the reaction mix of 50 µl consisted of 50 ng

(pHYF-5’ or pMODV-prM+E) plasmid, 2 units of Pfu DNA polymerase

(Promega), 5 µl of 10 x buffer supplied by the Manufacturer, 400 µM dNTP,

and 1.2 µM of each of the two primers (Table 1, C and F respectively). The

conditions for the amplification reactions were as follows: 1 min at 95°C, 1

min at 59°C and 2 min at 72°C repeated for 35 cycles.

Two fragments of 933 bp and 2051 bp respectively were thus

obtained (Fig. 1).

Construction of the recombinant plasmid

The fragment of 2051 bp and the pHYF-5’ vector were digested

with SacI and HpaI (Promega) (Fig. 1). Ligation of 100 ng vector with 65 ng

PCR fragment was carried out using the T4 DNA ligase in a 2 x rapid

ligation buffer (New England Biolabs GmbH, Frankfurt am Mainz,

Germany). The resulting plasmid pHYF-MO1 was digested with HpaI and

NotI (Promega) and served as a vector for the insertion of the HpaI-NotI

digested 933 bp fragment. The resulting plasmid pHYF-MO2 was digested

with MluI-NotI (Promega) and the DNA fragment [encompassing from 5’ to

3’ the Sp6 promotor fused to the YFV 5’ UTR and C gene, MODV prM

and E gene and part of the YFV NS1 gene] was ligated into NotI-MluI

digested pACNR-FLYF17Da to construct pACNR-MODV/YFV (Fig. 1).

The recombinant region was sequenced in a cycle sequencing

reaction with fluorescent dye terminators (Big Dye Terminator Cycle

Sequencing Ready Reaction kit, Applied Biosystems Division) and analyzed

using an ABI 373 automatic sequencer (Applied Biosystems Division).


pMODV-prM+E5889 bp

(1932 prM+E, 3957 TOPO)

prM

E

209 bp

2051 bpSacIHpaI

205 bp

YFV

MODV

pHYF-5’ pMODV-prM+E

prM

E

prM

E

C

NS

1

pHYF-5’6751 bp(2018 prM+E)

5’ UTR

HpaI

SacI

NotI

prM

E

C

NS

1

pHYF-5’5’ UTR

prM

E

C

NS

1

5’ UTR

vector

933 bpNotI HpaI

pHYF-MO2NotI

prM

E

C

NS

1

5’ UTR

prM

E

MluI

NotI

MluI

C

NS

1pr

M

E

pACNR-FLYF17Da(13,457 bp)

NS3

4A4

BN

S5

NS2A2B

5’ UTR3’ UTR

C

NS1

prM

E

pACNR-MODV/YFV

(13,430 bp)

NS3

4A

4B

NS

5

NS2A 2B

5’ UTR

3’ UTR

prM

E

A

D

C

F

E

B

AflII

Fusion-PCR 1 Fusion-PCR 2

ABDE

pMODV-prM+E5889 bp

(1932 prM+E, 3957 TOPO)

prM

E

209 bp

2051 bpSacIHpaI

205 bp

YFV

MODV

pHYF-5’ pMODV-prM+E

prM

E

prM

E

C

NS

1

pHYF-5’6751 bp(2018 prM+E)

5’ UTR

HpaI

SacI

NotI

prM

E

C

NS

1

pHYF-5’5’ UTR

prM

E

C

NS

1

5’ UTR

vector

933 bpNotI HpaI

pHYF-MO2NotI

prM

E

C

NS

1

5’ UTR

prM

E

MluI

NotI

MluI

C

NS

1pr

M

E

pACNR-FLYF17Da(13,457 bp)

NS3

4A4

BN

S5

NS2A2B

5’ UTR3’ UTR

C

NS1

prM

E

pACNR-MODV/YFV

(13,430 bp)

NS3

4A

4B

NS

5

NS2A 2B

5’ UTR

3’ UTR

prM

E

A

D

C

F

E

B

AflII


ABDE

Fig. 1. Construction of MODV/YFV chimera by using a variant of the fusion-PCR.

Chapter 4

Generation of recombinant viral RNA and transfection of BHK cells

Recombinant viral RNA was transcribed in vitro for 2 h at 37°C

using 5 µg AflII-linearized pACNR-MODV/YFV plasmid (Fig. 1) as a

template and Sp6 RNA-polymerase (1500 units/ml) and RNase inhibitor

(1000 units/ml) using the reaction conditions provided by the supplier. The

transcription reaction was spiked with 1 µCi [3H]-UTP (46 Ci/mmol) to

determine the yield of the transcription reaction.

BHK cells were transfected by electroporation essentially as

described (van Dinten et al., 1997). The medium was harvested from the

transfected cells when the CPE was nearly complete, clarified by

centrifugation and subsequently used to infect new Vero cells. At 48 h post

infection total cellular RNA was isolated from the infected Vero cells and

subsequently used in a RT-PCR to verify the chimeric nature of the virus.


RESULTS

4.1. Construction of the pMODV-prM+E plasmid

A PCR-fragment of 2.2 kb containing the prM and E genes of

MODV genome was cloned into the TOPO vector, creating the plasmid

pMODV-prM+E. Sequence analysis of the prM+E insert of this plasmid

demonstrated that the sequence of the MODV cDNA was identical to the

previous published sequence (Leyssen et al., 2002).

4.2. Construction of chimeric viruses by fusion-PCR

The nucleotide sequences at the 3’-termini of the YFV C and E

genes do not contain restriction enzyme sites that allow an exchange of the

YFV prM and E genes by the corresponding genes of MODV. Therefore

site-directed mutagenesis or fusion-PCR is required to create the desired

gene fusions. Two fusion-PCRs were designed to obtain the chimeric clone

pACNR-MODV/YFV, that contains the YFV genome with the prM and E

genes replaced by the prM and E genes of MODV (Fig. 1).

YFV C/MODV prM gene fusion

A short YFV/MODV fragment (205 bp) containing the 24 3’-

terminal nucleotides of the YFV C gene and the first 181 nucleotides of the

MODV prM gene was obtained by performing a standard hot start PCR

using the pMODV-prM+E plasmid as template. To this end, a sense

chimeric primer A (Table 1), consisting of the last 24 nucleotides of the

sequence of the YFV C gene and the first 24 nucleotides of the sequence of

the MODV prM gene, and an antisense primer B located in the prM gene of

MODV, were used. The amplified fragment served as antisense primer,

together with the sense primer C, which contained a NotI restriction site, to

generate a 933 bp fragment by fusion-PCR using the pHYF-5’ plasmid as

template.

Chapter 4

MODV E/YFV NS1 gene fusion

A short MODV/YFV fragment (209 bp) containing the 23 3’-

terminal nucleotides of the MODV E gene and the first 186 nucleotides of

the YFV NS1 gene was obtained by a standard PCR using pHYF-5’ as a

template and the chimeric primer D, consisting of the last 23 nucleotides in

the sequence of the MODV E gene and the first 23 nucleotides of the

sequence of the YFV NS1 gene and the antisense primer E located in the

NS1 gene of YFV. After purification, this amplified fragment was

subsequently used as an antisense primer, in combination with the sense

primer F to generate a 2051 bp fragment by fusion-PCR using the

pMODV-prM+E plasmid as template.

4.3. Cloning of MODV prM and E genes

First, the two fused fragments (933 en 2051 bp) were cloned into

the shuttle vector pHYF-5’. This vector contains the 5’ 2947 bases of YFV

17D fused to the Sp6 promoter and contains some restriction sites that are

essential for the cloning strategy and that are no longer unique in the much

larger pACNR-FLYF17Da.

The fragment of 2051 bp was digested with HpaI and SacI and

cloned into SacI and HpaI digested pHYF-5’. The resulting pHYF-MO1

contained the fusion between MODV E and YFV NS1. Plasmid pHYF-

MO1 was subsequently cut with HpaI and NotI to serve as a vector for the

HpaI and NotI digested 933 bp fragment which contained the YFV

C/MODV prM fusion. This resulted in the construction of pHYF-MO2.

To simplify Figure 1, the two-step ligation of the 933 and 2051 bp

fragments is presented as one cloning step.

The DNA fragment (encompassing Sp6 promotor, YFV 5’ UTR

and C gene, MODV prM and E genes and part of the YFV NS1 gene) was

isolated from pHYF-MO2 by digestion with MluI and NotI and cloned into


MluI-NotI digested pACNR-FLYF17Da to yield pACNR-MODV/YFV.

This final plasmid contains the chimeric MODV/YFV cDNA that consists

of full-length YFV cDNA in which the prM and E genes have been

replaced by the MODV prM and E genes. At the 5’ end, this cDNA is

directly fused to a Sp6 promotor to facilitate run-off transcription of full-

length, infectious RNA.

Sequencing of the chimeric region confirmed the expected sequence,

except for one mutation that was corrected by site-directed mutagenesis.

4.4. Recovery of chimeric MODV/YFV viruses

Full-length viral RNA was Sp6 transcribed in vitro using AflII

linearized pACNR-FLYF17Da and pACNR-MODV/YFV as a template

(Fig. 1). The transcripts were used to transfect BHK cells. Cells were seeded

in 10 ∅ cm dishes, incubated at 37°C. CPE was observed at 36 h post

electroporation (p.e.) in the cultures transfected with YFV 17D transcripts

and at approximately 48 h p.e. for the cultures transfected with the

MODV/YFV chimeric RNA.

4.5. Verification of the chimeric virus produced by BHK cells

To demonstrate that the chimeric MODV/YFV RNA is capable of

producing infectious virus, the medium of the transfected cells was

harvested and used to infect new Vero cells. At 48 h post infection total

cellular RNA was isolated and (chimeric) virus RT-PCRs were performed to

demonstrate the presence of viral RNA in the infected cells. Several sets of

primers were designed (Table 2) which allow unambiguous identification of

either YFV, MODV and chimeric MODV/YFV RNA (Fig. 2). RT-PCR

reactions were developed [using the Qiagen One Step RT-PCR kit (Qiagen)]

in which fragments of the MODV prM+E (primer set 1) and NS4B region

(primer set 2), the YFV E (primer set 3) and NS1 region (primer set 4) and

Chapter 4

Fig. 2. Control of the chimeric region. Localization of the designed primers in the genomes of YFV 17D, MODV and the chimeric virus MODV/YFV.

Table 2

Primers used to control the chimeric region

Primer set Sequence (sense/antisense) Nt-position

1

2

3

4

5

5’-GGACCCTGGAGAGGGATGGACAA-3’

5’-CAGATCAAAGATTCTCAGCCATTAATTTGATGGATCT-3’

5’-GTTTGTTGGTCTTGTGATTGC-3’

5’-GAGGAACAGCAAAACTAGCCT-3’

5’-TAGCCATTGATAGACCTGCTGAGGT-3’

5’-GCTTTGACACTCAAGGGGACATCC-3’

5’-GACTTGGGGTAAGAACCTTGTGTTC-3’

5’-CTGGTGCGCTCCTGGGTTACAGCT-3’

5’-GGACCCTGGAGAGGGATGGACAA-3’

5’-GCCTAAATTCAGTTGACTCC-3’

MODV: 1598-1620

MODV: 2166-2202

MODV: 6754-6774

MODV: 7710-7730

YFV: 1116-1140

YFV: 1838-1861

YFV: 2800-2824

YFV: 3485-3508

MODV: 1598-1620

YFV: 2619-2638

C prM E NS1 NS2A NS2B NS3 NS4A NS4B NS5

MODV

5’ UTR Cap 3’ UTR

2 1

YFV

C prM E NS1 NS2A NS2B NS3 NS4A NS4B NS5 5’ UTR Cap 3’ UTR

3 4


MODV/YFV

5

5’ UTR Cap 3’ UTR

4 1


the YFV C/MODV prM hinge region (primer set 5) could selectively be

amplified (Table 2). A fragment of the expected size of the YFV NS1 (lane

4, 709 bp), MODV prM+E (lane 1, 605 bp) and MODV/YFV gene fusion

(lane 5, 960 bp) was amplified using RNA that was isolated from cells

infected with the MODV/YFV virus. No amplicon was however generated

using the primers designed to amplify the YFV E (lane 3, 746 bp) and

MODV NS4B region (lane 2, 980 bp) (Fig. 3).

YFV MODV/YFV MODV

M 1 2 3 4 5 6 M 1 2 3 4 5 6 M 1 2 3 4 5 6

10 kb

2 kb

1 kb

0.5 kb

Fig. 3. Control of the chimeric region. Visualization of the fragments obtained after RT-PCR. M: 1 kb DNA ladder, lane 1: primer set 1 (605 bp, region MODV prM+E), lane 2: primer set 2 (980 bp, region MODV NS4b), lane 3: primer set 3 (746 bp, region YFV E), lane 4: primer set 4 (709 bp, region YFV NS1), lane 5: primer set 5 (960 bp, chimeric hinge region), lane 6: negative controle.

Chapter 4

DISCUSSION

Chimeric flaviviruses have conventionally been produced by

restriction of the genes to be exchanged (Huang et al., 2000; Guirakhoo et

al., 1999; Chambers et al., 1999; Monath et al., 1999b; Bray et al., 1996; Chen

et al., 1995; Pletnev et al., 1993; Bray and Lai, 1991) or by fusion-PCR of two

amplified PCR fragments (Caufour et al., 2001; Dekker et al., 2000; Van Der

Most et al., 2000). The restriction method requires the presence of a unique

restriction site at the point of fusion between the two desired genomic

fragments. Often, these unique restriction sites are not present and need to

be created, which is labor-intensive. A fusion-PCR provides the advantage

that a fusion between two sequences can be performed at any location.

“Classical fusion-PCR” has been described for the construction of chimeric

swine vesicular disease viruses (Dekker et al., 2000) as well as for studies

such as epitope mapping, introduction of mutations (Kitazono et al., 2002;

Karreman, 1998), generation of a gene disruption construct (Kuwayama et

al., 2002; Amberg et al., 1995), and production of heterodimeric Fab

antibody protein (Mullinax et al., 1992).

Here we report a very reliable and efficient fusion-PCR strategy

which allows a quick construction of gene fusion, as illustrated in this study

by the generation of a chimeric (flavi)virus. Instead of fusing two amplified

fragments in a third PCR, the presented method allows a fusion in only two

PCR steps. The first PCR amplifies a short fragment that will act as a primer

in a subsequent fusion-PCR. Thus, instead of three amplification reactions

with four primers, a fusion between two different genes is achieved by two

amplification reactions and only three primers. Care should be taken to

design the primers so that they have almost the same melting temperature

(Tm) value, which is crucial for annealing and amplification of the final PCR

product. This approach also offers the advantage that the risk of

accumulating mutations by PCR polymerases is being reduced. The method

is in principle readily amenable for the construction of any gene fusion.

C h a p t e r 5

EXCHANGING THE YELLOW FEVER VIRUS ENVELOPE

PROTEINS FOR MODOC VIRUS prM AND E RESULTS IN A

CHIMERIC VIRUS THAT IS NEUROINVASIVE IN MICE

This chapter will be the subject of the following publication:

Charlier, N., Molenkamp, R., Leyssen, P., Paeshuyse, J., Drosten, C., De

Clercq, E., Bredenbeek, P.J., and Neyts, J. Exchanging the yellow fever virus

envelope proteins for Modoc virus prM and E results in a chimeric virus

that is neuroinvasive in mice. Submitted for publication in the Journal of

Virology.

Chapter 5

SUMMARY

A chimeric flavivirus infectious cDNA was constructed by

exchanging the prM and E genes of YFV 17D for the corresponding genes

of MODV, a flavivirus that belongs to the NKV cluster. Replication of in

vitro transcribed RNA from this chimeric virus was shown by 3H-uridine

labelling/RNA analysis. Expression of the MODV prM and E proteins was

monitored by radio-immunoprecipitation and revealed that the MODV

proteins were correctly and efficiently produced from the chimeric

precursor protein. The MODV E protein was shown to be N-linked

glycosylated, whereas prM, as predicted from the genome sequence, did not

contain N-linked carbohydrates. In Vero cells, the chimeric virus replicated

with a similar efficiency as the parental viruses, although it formed smaller

plaques than YFV 17D and MODV. In SCID mice that had been infected

intraperitoneally with the chimeric virus, viremia was detectable as of day 3

post infection, and the viral load increased steadily. The MODV/YFV virus,

like MODV from which it had acquired the prM and E structural proteins,

but unlike YFV, proved neuroinvasive in SCID mice. Animals developed

neurological symptoms (including paralysis) about 15 days after inoculation

and died shortly thereafter. The distribution of viral RNA in the brain of

MODV/YFV-infected mice was similar as in MODV-infected mice.

INTRODUCTION

All chimeric flaviviruses that have been constructed so far, are

between two vector-borne flaviviruses. These chimeric flaviviruses have

mainly been constructed in an attempt to develop vaccines against

flaviviruses that cause disease in man. Most of these chimeric viruses are

derived from the infectious clone of YFV 17D (Rice et al., 1989). YFV 17D

does not replicate in mosquitoes (although it replicates in mosquito cells)

and the post-vaccination viremia levels in humans are low, which limits

spread of the virus by mosquitoes in nature. YFV 17D has been used as a

Characteristics of a MODV/YFV chimeric virus

“vector” for engineering new vaccines against other flaviviruses [e.g. the

ChimeriVax vaccines ChimeriVax-JEV (Chambers et al., 1999; Monath et al.,

1999b), ChimeriVax-DENV-2 (Guirakhoo et al., 1999, 2000) and

ChimeriVax-DENV-1, -3 and -4 (Guirakhoo et al., 2001, 2002) and other

chimeras , i.e. WNV/YFV 17D (Monath, 2001a; Monath et al., 2001b) and

DENV-2/YFV17D that are based on other DENV-2 strains than the one

used for ChimeriVax-DENV-2 (Caufour et al., 2001; van der Most et al.,

2000)].

The mechanisms and determinants underlying the neuroinvasiveness

of flaviviruses are poorly understood. To further analyze the role of viral

envelope proteins in neuroinvasiveness of flaviviruses (in mice), we

constructed the first viable chimeric flavivirus between a vector-borne and a

NKV flavivirus. This chimeric flavivirus consists of the genetic backbone of

the non-neuroinvasive YFV 17D of which the prM+E genes were replaced

by the corresponding genes of the highly neuroinvasive MODV.

We report on the particular characteristics of this MODV/YFV

chimera that (i) proved instrumental in obtaining information on the

expression and processing of the envelope proteins of a NKV flavivirus and

that (ii) allowed to gain further insight into the determinants underlying

flavivirus neuroinvasiveness.

Chapter 5



Vero cells and MODV were obtained from the ATCC (CCL-81 and

VR-415, respectively). BHK-21J cells (Lindenbach & Rice, 1997) were

kindly provided by Dr. C.M. Rice. YFV 17D was derived from pACNR-

FLYF17Da (Bredenbeek et al., 2003; Molenkamp et al., 2003). The

construction of the chimeric vector pACNR-MODV/YFV containing the

full-length MODV/YFV cDNA (Fig. 1) by a convenient 2-step fusion-PCR

technology has been described in Chapter 4.

RNA transcription and transfection

Plasmid DNA containing full-length YFV or MODV/YFV cDNA

was linearized with AflII and purified by proteinase K digestion,

phenol/chloroform extraction and ethanol precipitation. Run-off RNA

transcripts were produced in vitro using Sp6 RNA-polymerase (Charlier et al.,

2003). Transcripts were used for electroporation of BHK-21J cells without

any additional purification (van Dinten et al., 1997). Cell culture medium was

harvested at the time the transfected cells displayed nearly complete

cytopathogenicity. Medium was cleared from cell debris by centrifugation

and subsequently used to prepare MODV/YFV stocks in Vero cells.

Monitoring viral kinetics

Monolayers of Vero cells were inoculated with 107 pfu of MODV,

YFV or MODV/YFV (Fig. 1) at 37°C in 25 cm2 culture flasks. Cell culture

medium was harvested every two days (between day 0 and 10), and titrated

for infectious virus content on Vero cells. Plaque assays were done in

confluent Vero cells as described previously (Bredenbeek et al., 2003). Real-

time quantitative RT-PCR (see below) to determine viral RNA load was

carried out on RNA extracted from the collected media.


Metabolic labeling and immunoprecipitation

Following electroporation with YFV 17D or MODV/YFV RNA,

BHK cells were seeded in 35 ∅ mm tissue culture dishes (2 ml/dish).

Alternatively, cells (1 x 106) were infected with MODV at M.O.I. 10. At 18 h

post transfection, the medium was replaced by 1 ml of RPMI 1640 medium

lacking FCS, methionine and cysteine (BioWhittaker). Cells were

metabolically labeled for 2 h with 100 µCi/ml of 35S-labeled methionine and

cysteine (Promix, Amersham Pharmacia Biotech). Cells were lysed with 1 ml

of 50 mM Tris-Cl (pH 7.5), 200 mM NaCl, 1 mM EDTA, 0.5% Triton X-

100, 0.1% SDS. For immunoprecipitation, 0.5 ml of lysate was mixed with 1

µl of hamster hyperimmune serum (Mod1). Following overnight incubation,

immunoprecipitates were collected with Pansorbin (Calbiochem) and

washed with 50 mM Tris-Cl (pH 7.5), 200 mM NaCl, 1 mM EDTA, 0.1%

SDS. The immunoprecipitated proteins were split in two samples. One

sample was treated overnight at 37°C with endo-β-N-acetylglucosamidase H

(EndoH, Boehringer Mannheim) while the second sample served as

untreated control. Precipitated proteins were analyzed by SDS-PAGE (15%

gel).

Quantitative RT-PCR of MODV, MODV/YFV and YFV RNA

RNA extraction was performed using the Qiagen Viral RNA kit


the columns were incubated with 50 µl of RNase-free water at 80°C.

Primers and probes were designed for MODV/YFV 17D: sense primer 5´-

TGGGTTTTGGTCTTCTAGCTTTCA-3’, antisense primer 5´-CTTGTTC

AGCCAGTCATCAGAGTCT-3’ and probe 5´-CAGGAGTGATGGGAA

ATCAAGGATGC-3’; YFV 17D: sense primer 5´-AATCGAGTTGCTAG

GCAATAAACAC-3’, antisense primer 5´-TCCCTGAGCTTTACGACCA

GA-3’ and probe 5´-ATCGTTCGTTGAGCGATTAGCAG-3’. Primers

and probe for quantitation of MODV RNA and the reaction conditions for

Chapter 5

MODV/YFV 17D, YFV 17D and MODV were as reported earlier (Leyssen

et al., 2001). Thermal cycling in a LightCycler (Roche Molecular

Biochemicals, Mannheim, Germany) involved reverse transcription (RT) at

45°C for 20 min, denaturation at 95°C for 5 min, followed by 45 cycles of 5

s at 95°C and 35 s at 57°C (for YFV) or 20 s at 60°C (for MODV and

MODV/YFV) (Wittwer et al., 1997; Drosten et al., 2002).

Immunofluorescence assay (IFA)

BHK cells that had been transfected with YFV 17D RNA or

MODV/YFV RNA, or that had been infected with MODV 24 h earlier,

were fixed in paraformaldehyde and prepared for immunofluorescence

(Bredenbeek et al., 2003). Mab 1A5 (kindly provided by Dr. J.J. Schlesinger)

which is specific for the YFV NS1 protein (Schlesinger et al., 1983), and

Mod1 (serum from hamsters that had recovered from MODV infection)

were used to detect the YFV NS1 and MODV antigens respectively.

Animals

SCID mice (10-day or 16-week-old and weighing 16-20 g) were used

throughout the experiments. All animals were bred at the Rega Institute

under specific pathogen-free conditions. All animal experiments were

conducted in accordance with the guidelines of the “Ethical Committee on

Vertebrate Animal Experiments” of the University of Leuven.

Infection of SCID mice with MODV, YFV 17D and MODV/YFV

Infection of SCID mice with the parental or chimeric viruses (all

passaged on Vero cells) was carried out in a BSL3 facility. Adult SCID mice

were inoculated with either 104 pfu of virus via the i.p. route or with 2.5x103

pfu via the i.c. route. The animals were examined daily for signs of morbidity

and for mortality. Statistical significance of differences in the MDD was

assessed by means of the Student's t test and differences in the number of


survivors was assessed by means of the Fisher exact test. Every three days,

three animals were euthanized, blood was collected by cardiac puncture and

total RNA was extracted from the serum and subsequently analyzed by real-

time quantitative RT-PCR. To obtain tissue samples for in situ hybridisation,

mice with severe signs of paralysis were euthanized and transcardially

perfused with 20 ml of a buffered 4% formaline solution. Fixed brain

samples were embedded in paraffin and further processed by using standard

histological procedures or were subjected to in situ hybridisation (see below).

Preparation of digoxigenin-labeled cRNA and in situ hybridisation

MODV cDNA encompassing 605 nucleotides of the E region of

the MODV genome (sense primer 5'-GGACCCTGGAGAGGGATGGAC

AA-3' and antisense primer 5'-CAGATCAAAGATTCTCAGCCATTAATT

TGATGGATCT-3') and YFV cDNA encompassing 709 nucleotides of the

NS1 region of the YFV genome (sense primer 5'-GACTTGGGGTAAGAA

CCTTGTGTTC-3' and antisense primer 5'-CTGGTGCGCTCCTGGGTT

ACAGCT-3') were each cloned into the transcription vector pGEM-T

(Promega). To generate run-off transcripts, the plasmids containing MODV

cDNA and YFV cDNA were linearized with NcoI or SacII (Promega),

respectively. Transcription reactions were carried out as described earlier

(Charlier et al., 2003). Paraffin-embedded tissue was sectioned, hydrated,

fixed, denaturated and acetylated according to standard procedures

(Breitschopf et al., 1992). In situ hybridisation was performed essentially as

described earlier (Charlier et al., 2002a)

Chapter 5

1 2 3

RESULTS

5.1. Characterization of MODV/YFV chimera in cell culture

Using a fusion-PCR strategy, a full-length MODV/YFV cDNA was

constructed in the low copy number vector pACNR (Chapter 4). In this

chimera the prM and E coding regions of YFV 17D were replaced with the

complete prM and E genes of MODV (Fig. 1). Translation of the chimeric

RNA should result in the synthesis of MODV/YFV precursor protein from

which MODV prM and E would be released by a signalase mediated

cleavage.

Fig. 1. Genomic organization of the MODV/YFV chimeric virus, that had been constructed by replacing the prM and E genes of YFV 17D by the homologous genes of MODV (red box). The amino acid sequence of the hinge regions C/prM and E/NS1 are indicated (sequence of YFV 17D in black, sequence of MODV in red, “/” = predicted cleavage site)

To analyze the replication of this chimeric RNA, BHK-21J cells

were electroporated with in vitro transcripts of the full-length MODV/YFV

cDNA and subsequently labeled with 3H-uridine in the presence of

actinomycin D. As shown in Figure 2, a labeled RNA molecule with a size

similar to YFV and MODV RNA was detected in the MODV/YFV RNA

transfected cells, which demonstrates that the chimeric RNA was replication

competent.

Fig. 2. Analysis of viral RNA replication (1: MODV; 2: YFV; 3: MODV/YFV).

5’ UTR Cap

Structural proteins Non-structural proteins

3’ UTR

LLMTGGTILS/IEV TTGVMG/DQGCAI



To study whether the translation of the chimeric MODV/YFV

RNA resulted in the proper expression of the MODV prM and E proteins,

transfected cells were labeled with 35S-methionine and 35S-cysteine, lysed and

used for radio-immunoprecipitation. Since the MODV proteins had never

been identified before, MODV-infected and YFV-RNA transfected cells

were used as controls, respectively. As shown in Figure 3, the hamster

polyclonal anti-MODV serum Mod1 reacted specifically with proteins of 18

kDa, 44 kDa and 53 kDa in the lysate of the MODV-infected BHK cells.

Treatment with EndoH resulted in a shift of the 53 kDa protein to 51 kDa

and of the 44 kDa protein to 39 kDa, whereas the 18 kDa protein was

unaffected by EndoH. On the basis of the predicted molecular weight of

the unglycosylated MODV proteins, the immunoprecipitated proteins were

identified as prM (18 kDa), NS1 (39 kDa) and E (51 kDa) respectively. The

results of the EndoH treatment were also consistent with the fact that

MODV E and NS1 contain putative N-linked glycosylation sites, whereas

prM contains none.

Figure 3. Radio-immunoprecipitation of MODV and YFV proteins with the polyclonal antiserum Mod1 before and after EndoH treatment.

14

20

30

45

66 97

220

MODV prM

MODV E

MODV NS1

- + - - + + EndoH

YFV MODV/YFV MODV

Chapter 5

MODV/YFV 17D YFV 17D MODV

The Mod1 antiserum precipitated an approximately 40 kDa protein

from the lysates of the cells that were transfected with YFV transcripts. The

molecular weight of this protein indicated that it was unlikely to be a YFV

protein and therefor probably of host origin. In addition to this presumed

host protein, MODV prM and E were precipitated from lysates of cells that

were transfected with MODV/YFV transcripts. The prM and E proteins

expressed by the chimera had a similar MW and EndoH sensitivity profile as

prM and E in the MODV-infected cells, indicating that the proteins were

properly released from the precursor proteins and post-translationally

modified.

Approximately 48 h post electroporation of the BHK cells with

MODV/YFV RNA, pronounced CPE was noted. The medium was

harvested and used for plaque assay on Vero cells. Following incubation for

4 days, the MODV/YFV infected cells showed clear, but small plaques as

compared to the YFV plaques obtained in a similar experiment. The average

size of the MODV/YFV plaques was 0.5 mm against 3 mm for the YFV

plaques and 2.5 mm for the MODV plaques (Fig. 4).

Fig. 4. Plaque morphology of YFV17D, MODV and the MODV/YFV chimeric virus on Vero cells four days after infection.


0

1

2

3

4

5

6

7

8

0 2 4 6 8 10

days post infection

log

PC

RU

Kinetics of the replication of the three viruses was assessed by

quantitative RT-PCR specific for each virus. All viruses reached a peak of

virus production at approximately 6 days post infection. MODV replicated

somewhat more efficiently than YFV 17D and MODV/YFV; the latter two

had very comparable kinetics of replication (Fig. 5).

Fig. 5. Growth kinetics of MODV (▲), YFV 17D (□)

and MODV/YFV (●) in Vero

cells. Quantification of viral RNA in the supernatant was performed (in triplicate) by real-time quantitative RT-PCR (standard deviation of triplicate determinations was below 15%, not shown).

5.2. Confirmation of the chimeric geno- and phenotype

Immunostainings were performed on Vero cells that had either been

infected with the MODV/YFV chimera, MODV or YFV. Monoclonal

antibodies directed against the YFV NS1 protein recognized epitopes in

BHK cells that had been transfected with YFV or MODV/YFV RNA, but

did not result in a positive signal in cells infected with MODV. Conversely,

Mod1 antiserum recognized epitopes of the MODV prM, E (and NS1)

protein in BHK cells that had been transfected with MODV/YFV RNA or

infected with MODV, but did not detect antigens in cells that had been

transfected with YFV RNA (Fig. 6).

Chapter 5

Fig. 6. Immunostaining of MODV, YFV 17D or MODV/YFV-infected BHK cells using either monoclonal antibody 1A5 (directed against the YFV NS1 protein) or polyclonal antiserum Mod1 (collected from hamsters 8 months after infection with MODV).

5.3. Infection of SCID mice with MODV, YFV 17D and MODV/YFV

Intracerebral inoculation of 10 day-old or adult SCID mice with

2.5x103 pfu of YFV 17D, MODV or the chimeric virus caused 100 %

mortality within two weeks post infection [MDD = 8.5 ± 0.85, 6.5 ± 0.58

and 9.3 ± 1.56 respectively] (Fig. 7). Morbidity was characterized by ruffled

fur, paralysis of the hind legs and wasting syndrome. The day of appearance

of the first symptoms was for all three viruses approximately two days

before death.

We next studied whether the chimeric virus would be neuroinvasive.

To this end, mice were infected i.p. with 104 pfu of YFV, MODV and

MODV/YFV. In contrast to the parental YFV 17D, but akin to MODV,

the MODV/YFV chimera proved highly neuropathic for SCID mice. All

(adult) mice developed paralysis and signs of encephalitis, and finally

YFV MODV

Mod1

1A5

MODV/YFV


0

5

10

15

20

25

MODV/YFV MODV YFV

Me

an

Da

y o

f D

ea

th

100%

100%

100%

100% 100%

No m

ortality

0

50

100

150

200

250

300

350

400

450

0 3 6 9 12 15 18 21

days post infection

PC

RU

succumbed [MDD = 16.4 ± 5.37 for MODV/YFV and 10.8 ± 0.6 for

MODV] (Fig. 7). MODV/YFV-infected animals died significantly later

than those infected with a similar titer of MODV. Interestingly, the mice

that had been infected with the chimeric virus showed neurological

abnormalities only about 24 h before they died, whereas MODV-infected

animals presented these symptoms for a period of at least 2 days before they

succumbed.

Fig. 7. Virus (MODV/YFV, MODV or YFV 17D)-induced mortality (Mean day of death

± standard deviations) in SCID mice following i.p. (black box) or i.c. (gray box) inoculation. Values on top of the bar indicate the percentage of animals that died in that particular group [each group consisted of 10 to 20 mice].

Total RNA was extracted every three days from the serum of SCID

mice infected i.p. with MODV/YFV. Viral RNA was detected in the serum

by means of a sensitive real-time RT-PCR assay as early as day 3 post

infection. The viral RNA load in the serum continued to increase till death

(Fig. 8).

Fig. 8. Detection of viral RNA (by means of real-time quantitative RT-PCR) in the serum of SCID mice infected with the MODV/YFV chimeric virus via the i.p. route. Serum from three mice was collected and pooled every third day after infection. Quantification of the viral RNA load was done in triplicate (standard deviation below 15%, not shown).

Chapter 5

C

A

D

B

5.4. MODV/YFV RNA in the brain

The brain of SCID mice, that had been infected i.p. with

MODV/YFV, was removed at the time the animals showed signs of

paralysis. The distribution of viral RNA was determined by in situ

hybridisation using probes that specifically target either the MODV E

region (Fig. 9) or the YFV NS1 region (data not shown). Viral RNA was

detected in the gray matter of the olfactory bulbs, pyriform cortex, temporal

lobes (Fig. 9, panel A and B), the midbrain structures (thalamus, hypothalamus)

(data not shown), the hippocampus (Fig. 9, panel C) and the medulla oblongata

(data not shown). Viral RNA was also detected in the cerebellum where

infection of virtually all Purkinje cells was noted (Fig. 9, panel D). Overall,

MODV/YFV RNA was present in the cytoplasm of infected cells and

infection appeared to be confined to the neurons. The brain of SCID mice

that had been infected i.p. with MODV or i.c. with YFV showed no signal,

when using the YFV and MODV probe, respectively (data not shown).

Fig. 9. Localization of MODV/YFV RNA in the brain of SCID mice at the time of paralysis (15 days post infection by the i.p. route) by in situ hybridisation with a MODV specific probe. Neurons in the temporal cortex [panel A (40x) and panel B (10x)] and in the hippocampus (panel C) (2.5x), and Purkinje cells in the cerebellum [panel D (10x)].


DISCUSSION

Introduction of the infectious clone technology has opened new

opportunities in flavivirus vaccine research. The use of YFV 17D as a

backbone for the construction of chimeric viruses has been applied to the

development of new vaccines for JEV and DENV (Chambers et al., 1999;

Monath et al., 1999b; Guirakhoo et al., 1999, 2000). The rationale for this

approach is based on the efficacy and safety of the YFV 17D vaccine. Here,

we report the characterization of a viable chimeric flavivirus based on the 5’

UTR, C gene, NS genes and 3’ UTR of the vector-borne YFV 17D and the

prM and E genes of MODV, a virus that belongs to the NKV cluster of

flaviviruses. The chimeric RNA replicated relatively efficiently in cell culture

and directed the production of chimeric virus particles that were infectious

for BHK and Vero cells. The chimeric virus was neuroinvasive and caused

paralysis and encephalitis in adult immunodeficient (SCID) mice.

As part of the characterization of this chimeric MODV/YFV virus

in cell culture, we obtained for the first time information about the

expression of proteins of NKV flaviviruses, in particular of MODV. The

MODV prM, E and NS1 proteins have a molecular weight of respectively

18 kDa, 53 kDa and 44 kDa. As expected on the basis of the translation of

the MODV nucleotide sequence (Leyssen et al., 2002), it was shown that

both the MODV E and NS1 protein were N-linked glycosylated. For the

MODV E protein three potential N-linked glycosylation sites at respectively

amino acid-position 426, 598 and 731 were predicted. Given the relative

shift in mobility after treatment of the immunoprecipitated MODV E, it

appears that only one of these sites is actually used for glycosylation, most

likely, the glycosylation site at position 427 (sequence N-V-S). The site at

position 599 has the sequence N-D-T which is often a poor acceptor for N-

linked glycans (Kornfeld & Kornfeld, 1985), whereas the potential

glycosylation site at position 732 (N-F-S) is unlikely to be used, because it is

located in the hydrofobic carboxy-terminal domain of the E protein and is

Chapter 5

likely to be buried in the RER membrane. The MODV NS1 protein

contains two potential N-linked glycosylation sites. Although mutational

analysis may be required to provide a definitive answer, detailed

measurements on the shifts in mobility between native and EndoH-treated

NS1 suggests that both glycosylation sites are used for the addition of N-

linked glycans. The MODV prM was shown not to be modified by N-linked

glycosylation, which is in agreement with the sequence prediction. This is

unique since prM of all other flaviviruses studied thus far (all of which

belong to the vector-borne cluster) (reviewed in Chambers et al., 1990a)

were shown to be modified by the addition of N-linked glycans. Also,

analysis of the amino acid sequence of the prM gene of MMLV, RBV and

APOIV (that together with MODV belong to the NKV cluster) revealed

the presence of N-linked glycosylation sites. The prM and E proteins

expressed by the MODV/YFV virus appear to have the same molecular

weight and EndoH profile as their counterparts expressed by MODV. This

indicates that the prM and E are released by signalase at the predicted fusion

sequences from the recombinant YFV-MODV-YFV precursor protein. This

cleavage appears to occur efficiently since a MODV prM or E containing

precursor was never detected in pulse-chase experiments in cells transfected

with MODV/YFV transcripts (data not shown). MODV/YFV grew in

Vero cells about as efficiently as YFV 17D.

We employed the chimeric MODV/YFV to gain further insight into

the mechanism underlying neuroinvasiveness of flaviviruses. The

MODV/YFV chimeric virus contains the prM and E genes of a for mice

highly neurotropic virus (i.e. MODV) in the genome of a non-neuroinvasive

virus (YFV 17D). Akin to the parental YFV 17D and MODV, the chimeric

MODV/YFV proved neurovirulent to adult (SCID) mice. According to the

aforementioned studies, in which the prM+E of neuroinvasive viruses such

as WNV, JEV and LGTV were not able to transfer the neuroinvasive

phenotype, one would predict the MODV/YFV not to be neuroinvasive.

However, unlike the parent YFV(17D) strain, and akin to MODV from


which it had acquired the envelope genes, the chimeric MODV/YFV was

neuroinvasive in 100% of the infected SCID mice. Interestingly, the

distribution of chimeric viral RNA in the brain was very similar to that

observed in MODV-infected mice (Leyssen et al., 2001).

Our findings, together with earlier published observations that

mutations within E can destroy the neuroinvasive properties of a flavivirus,

provide compelling evidence that the flavivirus envelope proteins are the

principal determinants of neuroinvasiveness and thus attest to the fact that

the characteristics of neuroinvasiveness of the donor of genes are conferred

to the receiving virus. This information may be instrumental to further

understand the molecular basis of flavivirus neuroinvasiveness and may

have implications for engineering live attenuated chimeric vaccines.

C h a p t e r 6

CONSTRUCTION OF A MODOC VIRUS FULL-LENGTH

CLONE


Charlier, N., Gritsun, T., Molenkamp, R., Leyssen, P., De Clercq, E.,

Bredenbeek, P.J., Gould, E., and Neyts, J. Construction of a Modoc virus

full-length clone. Journal of Virological methods. In preparation.

Chapter 6

SUMMARY

We prepared a full-length clone of the NKV flavivirus MODV.

Construction of infectious clones can be problematic because of instability,

toxicity, and recombination events occurring during the cloning of cDNA in

the bacterial vectors. Another problem is the synthesis of full-length cDNA.

To address this problem we have devised a rapid and simple method to

generate a full-length long RT-PCR product using viral RNA isolated from

either (i) supernatant of MODV-infected Vero cells or (ii) an unpurified

MODV-infected mouse brain suspension. Optimization of the primer

concentration was essential to achieve PCR synthesis of an 11 kb cDNA

copy of the RNA from infectious virus. These developments represent a

significant advance in recombinant technology and should be applicable to

positive-stranded RNA viruses which do not replicate to high titers in

mouse brain.

INTRODUCTION

Infectious clone technology offers a modern and direct approach to

analyse and modify virus genomes at the molecular level, thus resolving

problems relating to virus replication, virulence and pathogenesis.

Conventional methods of developing infectious clones for RNA viruses

utilise the production of short (up to 2 kb) cDNA molecules by RT-PCR,

cloning these fragments in bacterial vectors and fusion of overlapping

clones to produce full-length cDNA clones. The first infectious flavivirus

clone was produced by Rice and colleagues, involving the in vitro ligation of

two cDNA fragments of YFV (Rice et al., 1989). Since then, several

attempts have been made to construct infectious clones of YFV

(Bredenbeek et al., 2003) and other flaviviruses: JEV (Sumiyoshi et al., 1995;

Yun et al., 2003; Zhang et al., 2001), DENV (Sriburi et al., 2001; Gualano et

al., 1998; Kapoor et al., 1995; Kinney et al., 1997; Lai et al., 1991; Polo et al.,

1997; Pur et al., 2000), KUNV (Khromykh & Westaway, 1994), MVEV

Construction of a Modoc virus full-length clone

(Hurrelbrink et al., 1999), WNV (Shi et al., 2002a, 2002b), LGTV (Pletnev,

2001a) and TBEV (Mandl et al., 1997; Gritsun & Gould, 1998). However,

the generation of infectious clones of flaviviruses is known to be particular

difficult. Bacterial cloning of flavivirus cDNA products frequently leads to

unwanted recombination events and the products are often toxic to the

bacterial host (Boyer & Haenni, 1994; Rice et al., 1989). In some cases,

genome-length cDNA of a flavivirus (for example, JEV) was found to be

unstable in a bacterial host (Sumiyoshi et al., 1995), whereas in other cases,

stable clones were obtained in E. coli or yeast (Gritsun & Gould, 1998). To

date the mechanisms governing the stability of such a clone are not fully

understood. Success in these latter cases might be the result of the use of

different combinations of the cloning vector, bacterial host and incubation

temperature of the bacterial cultures (Gritsun & Gould, 1998).

Genetically engineered infectious viral RNA can also be obtained by

direct transcription using a full-length cDNA template from either a long

RT-PCR or fusion of overlapping RT-PCR fragments without cloning.

Using this approach, the problematic process of constructing a full-length

cDNA clone can be avoided and recombinant viruses can be made within

days (Gritsun & Gould, 1995).

As discussed in Chapter 1, the genus Flavivirus contains (i) viruses

that are transmitted by mosquitoes or ticks (arthropod-borne) and (ii) NKV

viruses (Chambers et al., 1990a). All flaviviruses of human importance

belong to the arthropod-borne group; manipulation of NKV flaviviruses is

not dangerous to man, which makes them particular attractive to use as a

tool for the investigation of flavivirus biology. No full-length cDNA clones

of NKV flaviviruses have so far been made available. We here describe the

construction of full-length MODV cDNA using viral RNA from (i) the

supernatant of MODV-infected Vero cells and (ii) a MODV-infected mouse

brain suspension.

Chapter 6



Vero (ATCC CCL-81) cells were grown in MEM supplemented with

10% FCS and were maintained in 5% CO2 at 37°C. MODV was obtained

from the ATCC (VR-415). The low copy number vector pACNR1180

which contains the polylinker cassette of pSL1180 (Ruggli et al., 1996) was

kindly provided by Dr. P.J. Bredenbeek (Bredenbeek et al., 2003).

Animals

Suckling (3 days old) and adult (>1 month old) SCID mice were

used throughout the experiments. All animals were bred at the Rega

Institute under specific pathogen-free conditions. All animal experiments

were conducted in accordance with the guidelines of the “Ethical

Committee on Vertebrate Animal Experiments” of the University of

Leuven.

Infection of Vero cells with MODV

Confluent Vero cells were infected with 5x104 pfu of MODV and

virus was harvested from cell culture media at 7 days post infection.

Infection of SCID mice with MODV

Infection of SCID mice with MODV (passaged in Vero cells) was

carried out in a BSL2 facility. Suckling and adult SCID mice were inoculated

with 2.5x103 pfu via the i.c. route. The animals were examined daily for signs

of morbidity and at the day the mice showed signs of paralysis, the animals

were euthanized and the brains were collected and frozen at –80°C until

further use.


Homogenization of brain samples

Homogenization was carried out using a sterile bottle with glass

beads. Each brain was vortexed, 1 ml of cold PBS was added and vortexed

for a second time. The obtained homogenized mixture was centrifuged for

5-10 min at 10,000 x g, and the supernatant was collected.

RNA extraction and reverse transcription

Several RNA extraction protocols were examined. According to the

manual of the Promega RNAgents Total RNA Isolation System, 500 µl of

denaturing solution and 50 µl of sodium acetate (2 M, pH 4.0) were added

to 250 µl of virus-infected mouse brain suspension or 200 µl supernatant

from MODV-infected Vero cells. The tube was inverted 4-5 times. After

addition of 500 µl of the lower organic phase of phenol:chloroform:isoamyl

alcohol, the tube was again inverted 3-4 times. After incubation on ice for

15 min, the sample was centrifuged at 10,000 x g for 5 min. The top

aqueous phase was transferred into a new tube and 500 µl of isopropanol

was added. After incubation at -20°C for 5-15 min, the sample was

centrifuged at 10,000 x g for 10 min and the supernatant was removed. The

pellet was stored in 70% ethanol at -20°C until use.

According to the manual of TRIzol (Gibco) 200 µl supernatant from

infected Vero cells or 500 µl of virus-infected mouse brain suspension was

mixed with 600 µl TRIzol (or 200 mg of brain tissue with 1.5 ml TRIzol)

and incubated for 5 min at room temperature. Subsequently, 160 µl of

chloroform was added and the tube was inverted 3-5 times. After incubation

at room temperature for 2-15 min, the sample was centrifuged at 10,000 x g

for 15 min. The aqueous phase was transferred into a new tube and an equal

volume of isopropanol was added to precipitate the RNA. After incubation

at room temperature for 15 min the RNA was pelleted by centrifuging at

10,000 x g for 10 min. The supernatant was discarded and the pellet was

stored in 70% ethanol at -20°C until use.

Chapter 6

Prior to use, the RNA sample was centrifuged at 10,000 x g for 5

min and the supernatant was discarded. The pellet was resuspended in 11 µl

of diethylpyrocarbonate (DEPC) treated water and two samples (22 µl) were

mixed with 10 µl of 10 µM antisense primer (5’-AGCGGAGGTCATAT-3’)

and incubated for 2 min at 95°C. Subsequently, 3 µl DTT (100 mM,

Invitrogen), 100 units of RNasin (Promega), 12 µl 5 x buffer, 6 µl dNTP (10

mM) and 2 µl Superscript II RNase H- Reverse Transcriptase (Invitrogen)

were added and the mixture was incubated for 4 h at 43°C followed by 20

min at 65°C.

cDNA synthesis and cloning

Following the RT reaction, 22.5 µl water, 15 µl buffer [buffer A, B

or C (Table 1)], 1 µl dNTP (10 mM), 3 µl magnesiumacetate (25 mM), 2.4 µl

Table 1.

Content of the buffer used for the full-length PCR optimization

Buffer A Buffer B Buffer C Final concentration in PCR mix

Tricine (pH 8.8)

Potassium acetate

Glycerol

Dimethyl sulfoxide

50 mM

250 mM

0%

0%

50 mM

250 mM

16.6%

16.6%

50 mM

250 mM

23.24%

23.24%

15 mM

75 mM

0%-5%-7%

0%-5%-7%

of each primer (1 µM, Table 2) and 0.8 µl of a 20:1 mixture of Hot Start

DNA polymerase (Westburg BV, Leusden, The Netherlands) and Deep

Vent DNA polymerase (New England Biolabs GmbH) were added to 3 µl

of the RT mixture. The full-length MODV genome was amplified in 30

cycles of 40 s at 94°C and 10 min at 72°C.


Table 2.

Specific primers used for full-length PCR

Polarity Sequence (Sp6 promotor underlined, restriction sites in bold)

Sense

Antisense

5’-AGCTGCGGCCGCATTTAGGTGACACTATAGAGTTGATCCTGCCAGCGGTGGGTCG-3’ (NotI)

5’-CCGCGACGCGTAGCGGAGGTCATATTCATGACCACACAGATTAC-3’ (MluI)

Construction of full-length cDNA clones of MODV

The full-length PCR product and the vector pACNR1180 were

restricted with NotI and MluI (New England Biolabs GmbH). Following

extraction from low melting agarose gel, the correct fragments were ligated

with T4 DNA ligase (Gibco) for 2 h at 37°C and transformed into E. coli

MC1061 competent cells.

Chapter 6

RESULTS

6.1. RNA extraction

Two RNA extraction kits were used and compared to isolate RNA

from three different starting materials. The first positive results were

obtained with RNA extracted from MODV-infected Vero cell supernatant

using the Promega RNAgents Total RNA Isolation System. After RNA

extraction, the RNA pellets were mixed and resuspended in 22 µl water, and

subjected to RT and long PCR. A correct full-length PCR product was also

obtained using the TRIzol kit, although with a sligthly lower yield (data not

shown). Using the RNAgents System we then compared three different

starting materials, i.e. (i) supernatant of Vero cell infected with MODV, (ii)

brain homogenate of MODV-infected suckling SCID mice and (iii) brain

homogenate of adult SCID mice infected with MODV. The yield of full-

length PCR product was as follows: brain homogenate of suckling mice >

Vero cell supernatant > brain homogenate of adult mice. Homogenizing the

brain with PBS (10% mouse brain suspension) followed by addition of the

correct amount of denaturation solution, resulted in a higher yield than

homogenizing the brain immediately with the denaturation solution as is

recommended by the Manufacturer.

6.2. Long PCR

When testing three different buffer conditions (Table 1), we noticed

that the addition of glycerol and DMSO had no influence on the

quantitative or qualitative outcome of the long PCR (data not shown).

The effect of primerset dilution was investigated by diluting each

primer (10 µM) from 1:1 to 1:10 to obtain a final concentration in the

amplification reaction of 480 nM to 48 nM. The dilution 1:8 resulted in the

highest product yield. Interestingly, varying the ratio of the two primers


influenced the quantity of the PCR product, i.e. ratio 1:2 and 2:1 of

sense:antisense primer resulted in a higher yield than ratio 1:1.

Taking all facts together, we selected the following reaction

conditions: 22.5 µl water, 15 µl buffer B, 1 µl dNTP (10 mM), 3 µl

magnesiumacetate (25 mM), 2.4 µl of each primer (1.25 µM) and 0.8 µl of a

20:1 mixture of Hot start DNA polymerase and Deep Vent DNA

polymerase were added to 3 µl of the RT mixture. The full-length MODV

genome was amplified in 30 cycles of 40 s at 94°C and 10 min at 72°C.

Chapter 6

DISCUSSION

Recombinant cDNA clones from which full-length infectious RNA

can be transcribed are a valuable tool for studying the molecular biology of

positive-stranded RNA viruses (Boyer & Haenni, 1994). The approach relies

on the infectious nature of the genome RNA of these viruses when

transfected into permissive host cells. However, the construction of such

functional cDNAs for flaviviruses has proven difficult (Ruggli & Rice,

1999). One obstacle is the synthesis of full-length cDNA. Of all available

full-length clones for flaviviruses only two have been constructed with one

full-genome PCR product (Zhang et al., 2001; Gritsun & Gould, 1998). To

this end, suckling mice were inoculated i.c. with the virus. Mice with severe

signs of paralysis were euthanized and the brain was collected. Viral RNA

was extracted from the mouse brain suspension. However, some positive-

stranded RNA viruses do not replicate to high titers in mouse brain. Here

we report the production of full-length PCR product using RNA from

supernatant of MODV-infected Vero cells.

With the availablity of RNase H- reverse transcriptases and low error

rate thermostable DNA polymerases, RT-PCR products as long as 20 kb

have been produced (Thiel et al., 2002). However, it is still necessary to

optimize the protocols empirically for a long RT-PCR experiment. Gritsun

and Gould succeeded in amplifying a full-length cDNA from TBEV using a

full-length long RT-PCR technique, essentially by optimization of the 3’

antisense primer concentration (Gritsun & Gould, 1998). Furthermore, full-

length PCR product was obtained starting from unpurified RT product.

Other research groups did not succeed obtaining full-length PCR product

unless the RT mixture was treated with RNase A and purified (Zhang et al.,

2001). In the present study, working with unpurified RT product did not

hamper the PCR conditions.


Often the plasmids containing a full-length cDNA of flaviviruses are

unstable in E. coli. For some flaviviruses (e.g. YFV), this problem was

initially circumvented by using two plasmids and an in vitro ligation approach

(Rice et al., 1989). This strategy yielded the first functional flavivirus cDNA

for in vitro transcription of infectious viral RNA. In this report, we described

the construction of a stable, full-length MODV cDNA in an E. coli plasmid

vector. The MODV cDNA fragment that was used to construct this clone

was obtained from a designed full-length PCR. The low copy number

plasmid pACNR1180 was used as a vector for the construction of the full-

length MODV cDNA. This vector is derived from pACYC177 and has

previously been used to construct stable, infectious pestivirus and flavivirus

cDNAs (Mendez et al., 1998; Ruggli et al., 1996; Bredenbeek et al., 2003).

Once we have shown that this construct is able to generate

infectious MODV, a first aim will be to mutate the CUCAG motif in the 3’

UTR of the MODV genome to CACAG, the pentanucleotide motif typical

for vector-borne flaviviruses. Based on our observation made in Chapter 7

(where we demonstrated that mutating the CACAG motif of the chimeric

MODV/YFV 17D genome to CUCAG results in a virus that is no longer

able to replicate in mosquito cells), we expect that such mutated MODV, in

contrast to the wt MODV, will be able to replicate in mosquito cells.

C h a p t e r 7

THE HIGHLY CONSERVED PENTANUCLEOTIDE MOTIF

5’-CACAG-3’ IN THE 3’ UTR OF THE FLAVIVIRUS GENOME

DETERMINES WHETHER A FLAVIVIRUS IS ABLE OR NOT

TO REPLICATE IN MOSQUITO CELLS


Charlier, N., Molenkamp, R., Madder, M., Leyssen, P., Drosten, C., Panning,

M., De Clercq, E., Bredenbeek, P.J., and Neyts, J. The highly conserved

pentanucleotide motif 5’-CACAG-3’ in the 3’ UTR of the flavivirus genome

determines whether a flavivirus is able or not to replicate in mosquito cells.

Chapter 7

SUMMARY

So far no information is available on which factor(s) determines

whether a flavivirus is able to replicate in a vector (mosquito or tick) or not.

We show that a chimeric virus that consists of the YFV 17D genome, in

which the prM and E genes have been replaced by the corresponding prM

and E genes of the NKV flavivirus MODV, replicates as efficiently in

mosquito cells as the parental YFV 17D. This implies that the envelope

proteins of flaviviruses do not determine whether these viruses replicate in a

(mosquito) vector or not. The pentanucleotide motif CACAG in the 3’ LSH

of the 3’ UTR is highly conserved in all vector-borne flaviviruses. In viruses

that belong to the NKV cluster, this motif consists of C(C/U)(C/U)AG.

The pentanucleotide motif CACAG in the MODV/YFV chimeric virus was

mutated to the CUCAG motif which is characteristic for NKV flaviviruses.

The resulting virus replicated in Vero cells, but, unlike the parental virus, did

not longer replicate in mosquito cells. This provides strong evidence that

the pentanucleotide motif CACAG plays a crucial role in vector specificity.

INTRODUCTION

The genus Flavivirus contains (i) viruses that are transmitted by

mosquitoes or ticks (arthropod-borne) and (ii) NKV viruses (Chambers et

al., 1990a). No information is available about the factors that determine

whether a flavivirus is able or not to infect a vector (or vector cells). We

here report that the pentanucleotide motif CACAG in the 3’ LSH of the 3’

UTR of the flavivirus genome determines whether or not a flavivirus is able

to infect vector (mosquito) cells.

Determinants of vector specificity



enzymes, ligases, and polymerases were provided by the suppliers and used

according to their specifications.


Vero and BHK-21 cells were originally obtained from the ATCC

(CCL-81 and CCL-10 respectively). Mosquito (C6/36, Aedes albopictus) cells

were kindly provided by Dr. C.M. Rice. The clone pACNR-MODV/YFV

was constructed as described in Chapter 4. The clone pACNR-FLYF17Da

was obtained from Dr. P.J. Bredenbeek (Leiden, the Netherlands).

Amplification and cloning of a short YFV fragment containing the

CACAG pentanucleotide sequence

A fragment of 590 bp (nt-position 10,694-11,283 of the YFV 17D

genome), containing the CACAG pentanucleotide motif, was amplified in a

reaction mixture of 50 µl consisting of 30 ng of pACNR-FLYF17Da

plasmid, 2 units of Pfu DNA polymerase (Promega), 5 µl of 10 x buffer

supplied by the Manufacturer, 400 µM dNTP, and 1.2 µM of each of the

two primers (sense primer 5’-GTAGAAAGACGGGGTCTAGAGGT-3’

and antisense primer 5’-GGCACTGATGAGGGTGTCAGTG-3’). The

conditions for the amplification reaction were as follows: 30 s at 95°C, 30 s

at 60°C and 1 min at 72°C repeated for 25 cycles.

Following agarose gel electrophoresis the DNA fragment with the

expected length was cloned into a TOPO vector using the TOPO TA

Cloning kit (version H) (Invitrogen) and One Shot TOP10 E. coli cells

(Invitrogen) to yield pYFV-CACAG.

Chapter 7

Mutation of the pentanucleotide sequence

The adenosine at position two of the pentanucleotide motif

CACAG was mutated into a thymidine using the QuickChange Site-

Directed Mutagenesis Kit (Stratagene, Texas, USA): 50 ng of the pYFV-

CACAG plasmide was added to 5 µl 10 x buffer supplied by the

Manufacturer, 400 µM dNTP, 0.22 µM of each of the two primers (sense

primer and antisense primer 5’-GAGGTCTGTGAGCTCAGTTTGCTCA

AGAATAAGCAG-3’) and 2.5 units of Pfu Turbo DNA polymerase. The

conditions for the amplification reactions were 30 s at 95°C, 1 min at 55°C

and 9 min at 68°C repeated for 12 cycles. Following amplification, the

sample was treated with 10 units of the DpnI restriction enzyme and

transformed in One Shot TOP10 E. coli cells (Invitrogen) to yield pYFV-

CTCAG. To assure that the pentanucleotide sequence was altered correctly

and that no other mutations were inserted into the genome, the exchanged

fragment was sequenced.

Construction of the YFV full-length clone containing the CUCAG

motif

The plasmids pYFV-CTCAG and pACNR-MODV/YFV were

digested with the restriction enzymes XbaI and HaeII (Promega), and the

CTCAG containing fragment was inserted in the full-length clone pACNR-

MODV/YFV using T4 DNA ligase (Promega). Following ligation, which

yielded the plasmid pACNR-MODV/YFV(CTCAG), the latter was

transformed into MC1061 E. coli cells.

RNA transcription and transfection

Recombinant viral RNA was transcribed from 5 µg of AflII-

linearized pACNR-MODV/YFV(CTCAG) using Sp6 RNA-polymerase

(mMessage mMachine Kit; Ambion Ltd., Cambridgeshire, United

Kingdom). BHK cells were transfected by electroporation as described (van


Dinten et al., 1997). Cell culture medium was harvested at the time that the

transfected cells displayed nearly complete CPE. Medium was cleared from

cell debris by centrifugation and subsequently used to prepare layer stocks in

Vero cells.

Monitoring viral kinetics

Monolayers of Vero and C6/36 cells were inoculated with 107 pfu of

either MODV, YFV, MODV/YFV(CACAG) or the MODV/YFV

(CUCAG) chimeric virus at 37°C and 28°C in 25 cm2 culture flasks. Cell

culture medium was harvested every day or every two days (between day 0

and 10), and titrated for infectious virus content on Vero cells. Viral RNA

load was determined by real-time quantitative RT-PCR (see below) on RNA

extracted from the collected media.

Quantitative RT-PCR of MODV, YFV, MODV/YFV(CACAG) and

MODV/YFV(CUCAG) RNA

RNA extraction was performed using the Qiagen Viral RNA kit


the columns were incubated with 50 µl of RNase-free water at 80 °C. For

the quantitative determination of MODV, YFV and MODV/YFV RNA in

the supernatant of infected Vero and C6/36 cells, the reaction conditions

were as described in Chapter 5.

Chapter 7

0 1 2 3 4 5 6 7 8

0 2 4 6 8 10 Days post infection

log

PC

RU

Vero cells

0 1 2 3 4 5 6 7 8

0 2 4 6 8 10 Days post infection

log

PC

RU

C6/36 cells

RESULTS

7.1. Replication kinetics of MODV, YFV and MODV/YFV in Vero

and mosquito cells

To determine whether prM and E of flaviviruses play a role in the

vector specificity, the replication efficacy of MODV, YFV and

MODV/YFV was compared in Vero and mosquito cells. The kinetics of

replication of the three viruses were assessed by means of virus specific real-

time quantitative RT-PCR. All viruses reached a plateau of virus production

in Vero cells approximately 6 days post infection (Fig. 1). YFV 17D grew as

efficiently in mosquito C6/36 cells as in Vero cells. As expected, the NKV

flavivirus MODV did not replicate in C6/36 cells. If one assumes that the

prM and E proteins of MODV are responsible for the fact the virus does

not replicate in mosquito cells, one would expect the chimeric virus

MODV/YFV (containing the prM and E of MODV) not to replicate in

mosquito cells. However, in contrast to expectations, the MODV/YFV

replicated as efficiently as the parental YFV 17D in mosquito C6/36 cells

(Fig. 1). The factors that determine whether a flavivirus is able to replicate in

mosquito cells, are thus not located in the envelope proteins prM and E.

Fig. 1. Growth kinetics of YFV (�), MODV/YFV (�) and MODV (�) in Vero and

mosquito cells. Quantification of viral RNA in the supernatant was performed in triplicate by real-time quantitative RT-PCR using the probe methodology (standard deviation of triplicate determination below 15%) (PCRU = PCR units: One PCR unit is defined to be the lowest template copy number detectable in three of three replicative reactions as determined by limiting dilution series).


0

1

2

3

4

5

6

7

8

0 1 2 3 4 5 6 7 8

Days post infection

log

PC

RU

0

1

2

3

4

5

6

7

8

0 1 2 3 4 5 6 7 8

Days post infection

log

PC

RU

Vero cells C6/36 cells

7.2. Replication kinetics of MODV/YFV(CACAG) and MODV/YFV

(CUCAG) in Vero and mosquito cells

In Chapter 3, we reported that NKV flaviviruses carry, in contrast to

all 21 vector-borne flaviviruses (for which a sequence is today available), a

C(C/U)(C/U)AG pentanucleotide motif in the 3’ LSH of the 3’ UTR. We

mutated the pentanucleotide sequence CACAG in MODV/YFV into the

pentanucleotide motif that is characteristic for NKV flaviviruses. The

course of replication of the parental MODV/YFV(CACAG) and mutant

chimeric MODV/YFV(CUCAG) in Vero and C6/36 cells was compared in

two independent experiments. The kinetics of replication of the two viruses

was assessed by means of real-time quantitative RT-PCR. In Vero cells, the

chimeric MODV/YFV that contains the NKV pentanucleotide motif,

replicated somewhat slower than the chimeric virus that contains the vector-

borne CACAG motif. Eventually, both viruses efficiently produced CPE in

these cells. In contrast, the chimeric MODV/YFV virus that contained the

NKV specific pentanucleotide motif CUCAG, was not longer able to

replicate in mosquito (C6/36) cells (Fig. 2).

Fig. 2. Growth kinetics of MODV/YFV(CACAG) (●) and the mutant chimeric virus MODV/YFV(CUCAG) (○) in Vero and mosquito cells. Two indepent experiments were carried out that yielded similar findings. Data from one experiment are shown. Quantification of viral RNA in the supernatant was performed in duplicate by real-time quantitative RT-PCR using the probe methodology (standard deviation of triplicate determination below 15%).

Chapter 7

To study whether the inability of MODV/YFV(CUCAG) to

replicate in mosquito cells was solely due to the mutation in the

pentanucleotide motif (CACAG → CUCAG), the entire genome of the

virus was sequenced. No additional mutation was found.


DISCUSSION

All flaviviruses that cause disease in humans either belong to the

mosquito- or the tick-borne cluster. The third cluster, i.e. the NKV group,

consists of viruses for which no arthropod-borne route of transmission has

(yet) been demonstrated (Kuno et al., 1998). The factor(s) in the flavivirus

genome that determine whether a flavivirus is able or not to replicate in

vector (cells) have not been determined.

We first hypothesized that the viral envelope proteins (prM+E),

which are responsible for the initial interaction (i.e. binding and fusion) of

flaviviruses with the host cell, are responsible for whether a flavivirus is

either or not infectious to a vector. We therefore studied whether the

MODV/YFV chimeric virus (Chapters 4 and 5) is infectious to mosquito

cells. Since the MODV/YFV chimeric virus contains the envelope proteins

of the NKV flavivirus MODV, we assumed that the chimeric virus would

not be infectious to mosquito cells. However, the chimeric MODV/YFV

replicated as efficiently in mosquito cells as YFV 17D did. This thus proves

that the envelope proteins of the NKV flaviviruses are not responsible for

the fact that NKV viruses do not replicate in mosquito (vector) cells. Other

determinants must thus be responsible for the fact that NKV flaviviruses

(such as MODV) do not replicate in mosquito cells and, conversely, that

vector-borne flaviviruses do replicate in mosquito or tick (cells).

We earlier carried out a complete analysis of the 3’ UTR of NKV

flaviviruses (Chapter 3). The 3’ UTR of flaviviruses contains elements

involved in the regulation of essential functions such as translation,

replication or encapsidation of the genome (Khromykh et al., 2001; Proutski

et al., 1997a, 1997b). The 3’ UTR of NKV flaviviruses contains components

that are particularly characteristic for NKV flaviviruses, but also contains

components that are either typical for mosquito-borne or tick-borne

flaviviruses (or both) (Chapter 3). At each end of the genome, two terminal

Chapter 7

nucleotides, i.e. 5’-AG and CU-3’, are conserved among members of the

whole Flavivirus genus (vector-borne and NKV). Besides these conserved

terminal nucleotides, there is only one other nucleotide sequence motif

conserved in the 3’ UTR of the mosquito- and tick-borne flaviviruses, i.e.

the pentanucleotide sequence (5’-CACAG-3’) located approximately 45-61

nucleotides from the 3’ terminus (Wengler & Castle, 1986). This motif is

predicted to be located on a side-loop of a conserved 3’-terminal secondary

structure, which plays a role in the formation of a circular RNA molecule

(Chambers et al., 1990b; Khromykh et al., 2001). We sequenced the 3’ UTR

of four NKV flaviviruses (MMLV, MODV, RBV and APOIV) and

discovered that the second position of this pentanucleotide [which is

predicted to be located at an analogous position, i.e. within the loop of a 3’

LSH] was, in all 4 NKV flaviviruses studied, either U or C (instead of A in

the vector-borne flaviviruses). APOIV has, in addition, a C to U change at

position 3 of the pentanucleotide motif (Chapter 3). Since (i) all 21 vector-

borne flaviviruses (for which the sequence of the 3’ UTR is available)

contain A at position 2 of the pentanucleotide motif and (ii) all 4 NKV

flaviviruses carry either C or U at that position, it may be assumed that

evolutionary pressure is responsible for the observed sequence difference in

this motif between both groups. Since the common denominator in each

group is either “vector-borne” or “NKV”, it was thus tempting to speculate

that this pentanucleotide motif determines whether a flavivirus is able or not

to replicate in mosquito (vector) cells.

We therefore mutated the CACAG sequence in the 3’ UTR of the

MODV/YFV genome (which carries the 3’ UTR of YFV) into CUCAG, i.e.

the motif typical for NKV viruses that is also present in the MODV

genome. This novel mutant chimeric virus, assigned MODV/YFV

(CUCAG), failed, in contrast to the original chimeric virus

MODV/YFV(CACAG), to replicate in mosquito cells. We are currently

constructing a YFV mutant in which the pentanucleotide CACAG is

mutated to CUCAG and we will evaluate whether this YFV(CUCAG)


mutant has lost, as we expect, its capacity to replicate in mosquito cells.

Conversely, it will be important to study whether changing the NKV-

specific pentanucleotide motif CUCAG of a NKV flavivirus into the vector-

borne specific sequence CACAG, results in a NKV flavivirus that is able to

replicate in mosquito or tick cells. To this end, we need a full-length

infectious clone of a NKV virus. For this reason, we constructed a full-

length infectious clone of MODV (Chapter 6).

Insight into the determinants that are responsible for vector

specificity may also be important for the development of vaccines against

flaviviruses. As outlined in Chapter 5, various chimeric viruses have been

constructed in an attempt to produce safe vaccines against, amongst others,

DENV, WNV and JEV [e.g. (i) the ChimeriVax vaccines ChimeriVax-JEV

(Chambers et al., 1999; Monath et al., 1999b), ChimeriVax-DENV-2

(Guirakhoo et al., 1999; Guirakhoo et al., 2000) and ChimeriVax-DENV-1,

-3 and -4 (Guirakhoo et al., 2001; Guirakhoo et al., 2002), (ii) other chimeras

WNV/YFV 17D (Monath, 2001a; Monath et al., 2001b) and DENV-2/YFV

17D based on other DENV-2 strains than the one used for ChimeriVax-

DENV-2 (Caufour et al., 2001; van der Most et al., 2000), (iii)

WNV/DENV-4 (Pletnev et al., 2002)]. These chimeric vaccine viruses are

based on the YFV (or DENV) backbone and contain the 3’ UTR, and thus

the pentanucleotide motif CACAG of vector-borne flaviviruses. Such life

chimeric viruses can theoretically still be transmitted by mosquitoes from a

vaccinated person to a non-vaccinated person. It would be important to

disable these vaccine viruses to replicate in mosquito (cells). If our

observations with the MODV/YFV(CUCAG) and MODV/YFV(CACAG)

chimeric viruses can indeed be extended to other chimeric viruses, it may

allow the production of such chimeric vaccine viruses, as well as mutant

(CUCAG) flavivirus vaccines, that cannot longer be transmitted by a vector.

Because both mosquito and tick-borne flaviviruses contain the same

pentanucleotide motif it appears as though this motif is sufficient to allow

Chapter 7

replication of the viruses in mosquitoes and ticks. Still other determinants

must be responsible for a flavivirus to be either mosquito- or tick-specific.

This remains an intriguing issue worth of further investigation.

C h a p t e r 8

CONSTRUCTION OF A CHIMERIC DENGUE VIRUS

CONTAINING THE STRUCTURAL PROTEINS prM AND E OF

MODOC VIRUS


Charlier, N., Molenkamp, R., Leyssen, P., De Clercq, E., Bredenbeek, P.J.,

and Neyts, J. Construction of a chimeric dengue virus containing the

structural proteins prM and E of Modoc virus. Virus genes. In preparation.

Chapter 8

SUMMARY

We engineered a DENV-4 virus from which the genes coding for

the envelope proteins prM and E, were replaced by the homologous genes

of MODV. Our aim was to generate a DENV that is, unlike the parent

virus, able to replicate efficiently in mice. In two PCR steps, a fusion was

made between the 3’ end of the C gene of DENV-4 and the 5’ end of the

prM gene of MODV, and between the 3’ end of the E gene of MODV and

the 5’ end of the NS1 gene of DENV-4. For each of the two fusions

between DENV-4 and MODV, a standard PCR was performed to amplify a

short fragment with one overlapping end that could be used as one of the

primers in the subsequent (fusion) PCR.

INTRODUCTION

DENV is one of the most important flaviviruses that causes disease

in man. The dengue group of viruses includes four serotypes that exhibit a

high level of genome sequence homology and similar envelope protein

antigenic properties (Blok et al., 1992; Calisher et al., 1989; Kuno et al., 1998;

Zanotto et al., 1996; World Health Organization, 1997). Currently there is no

vaccine available against DENV, and there is, as for all other flaviviruses, no

specific therapy. Symptomatic treatment and intensive medical care are

essential to improve the chance of survival in patients with severe flavivirus

infections.

The mouse is not the natural host of DENV, although DENV can

replicate in the brains of suckling mice when infected i.c. (Raut et al., 1996).

SCID mice reconstituted with human peripheral blood lymphocytes and

infected subsequently i.p. with DENV-1, developed viremia, although virus

production was highly variable (Wu et al., 1995). Johnson and Roehrig

reported that mouse-adapted DENV-2 replicates in AG129 mice, which

lack IFN-α/β and IFN-γ receptor genes, after i.p. inoculation. Inoculation

of DENV-2 into SCID mice engrafted with human K562 cells (an

Construction of a MODV/DENV chimeric virus

erythroleukemia cell line) induced paralysis and death (Johnson & Roehrig,

1999; Lin et al., 1998). SCID mice transplanted with a human

hepatocarcinoma cell line (HepG2) and inoculated i.p. with DENV-2,

showed symptoms of encephalitis (An et al., 1999). When DENV-4 was

injected directly into i.p. grown tumors of the human liver cell line HuH-7

in SCID mice, virus was detected in the serum (Blaney et al., 2002).

Here we report the construction of a chimeric MODV/DENV-4,

that contains the envelope proteins prM and E of MODV, a virus that

readily replicates in mice, in the backbone of DENV-4. Such virus contains

the entire replication machinery of DENV-4. In case such chimeric

MODV/DENV-4 would be able to replicate in mice, this would allow to

assess antiviral strategies against DENV-4 in a small animal model.

Chapter 8



enzymes, ligases, DNA and RNA polymerases were provided by the

suppliers and used according to their specifications.


Vero and BHK-21 cells, and MODV were originally obtained from

the ATCC (CCL-81, CCL-10 and VR-415 respectively). BD1528 cells and

plasmid p4 containing a full-length cDNA of DENV-4 were kindly

provided by Dr. S. Whitehead (National Institute of Health, Bethesda,

Maryland). Plasmid pMODV-prM+E (Chapter 4) contains the prM and E

genes of MODV (nt-position 271-2468).

Primers

Primers used for the construction of the chimeric region were

designed based on the nucleotide sequence of DENV-4 (GenBank

accession number AF326573) (Durbin et al., 2001) and MODV (GenBank

accession number NC_003635) (Leyssen et al., 2002). The nucleotide

sequences of the primers are listed in Table 1.

Table 1

Primers used for PCR, site-directed mutagenesis and fusion-PCR

Name Polarity Sequence

A

B

C

D

E F

G

H

I

J

S

AS

S

AS

S AS

AS

S

AS

AS

5’-GCGCAACGTTGTTGCCATTGCTGC-3’

5’-CCACCAACAGCAGGGATTCTGAAGA-3’

5’-GGGAAAGGACCCTTACGTATGGTGCTAGCATTCATC-3’

5’-GGGAAAGGACCCTTACGTATGGTGCTAGCATTCATC-3’

5’-GGTTAGACCACCTTTCAATATGCTG-3’

5’-TGCTTGATTCCCACCGTAATGGCGACCATATTGTCAATTGAAGTTGTT-3’

5’-CATGATGAAGATGATCATTTCCCTGG-3’

5’-GGGAGTGTCATTGGGAAGGCTGTG-3’

5’-CTTTCACCACAGGAGTGATGGGAGACATGGGTTGTGTGGCGTCATGG-3’

5’-CACCAGGATACGTGCTGGGTGTG-3’

D

D

D (SnaBI) D (SnaBI) D D/M

M

M M/D

D

Abbreviations: S: sense; AS: antisense; D: DENV; M: MODV


Introduction of a unique restriction site in p4 plasmid

A unique restriction site at nt-position 230 of the DENV-4

sequence was created. Therefore, a fragment of 575 bp (nt-position 14,996

to 300 of the p4 plasmid) was amplified in a reaction mixture of 50 µl

consisting of 30 ng of p4 plasmid, 2 units of Pfu DNA polymerase


and 1.2 µM of each of the two primers (Table 1, A and B). The conditions

for the amplification reactions were as described in Chapter 7. Following

agarose gel electrophoresis, the DNA fragment with the expected length

was cloned into a TOPO vector using the TOPO TA Cloning kit

(Invitrogen) and One Shot TOP10 E. coli cells (Invitrogen) to yield pA.

Subsequently, a mutation to create a SnaBI restriction site, was

introduced by means of site-directed mutagenesis (QuickChange Site-

Directed Mutagenesis Kit, Stratagene): 50 ng of the pA plasmid was added

to 5 µl 10 x buffer supplied by the Manufacturer, 400 µM dNTP, 0.22 µM of

each of the two primers (Table 1, C and D) and 2.5 units of Pfu Turbo

DNA polymerase. The conditions for the amplification reactions were as

described in Chapter 4. Following amplification, the sample was treated with

10 units of the DpnI restriction enzyme and transformed in One Shot

TOP10 E. coli cells (Invitrogen) to yield pA-SnaBI. To assure that the

recognition site for SnaBI was formed correctly and that no other mutations

were inserted in the sequence, the exchanged fragment was sequenced.

The plasmids pA-SnaBI and p4 were digested with the restriction

enzyme NheI (Promega), and the fragment containing the SnaBI recognition

site, was inserted in the full-length clone p4 using T4 DNA ligase

(Promega). Following ligation, which yielded the plasmid p4-SnaBI, the

latter was transformed into BD1528 E. coli cells. Finally, the sequence

encompassing the exchanged fragment was sequenced to assure that the

fragment was inserted in the correct direction.

Chapter 8

Fusion-PCR

Two short fragments (343 and 312 bp, Fig. 1) were amplified in a

reaction mix of 50 µl consisting of 100 ng (p4-SnaBI or pMODV-prM+E)

plasmid, 2 units of Pfu DNA polymerase (Promega), 5 µl of 10 x buffer

supplied by the Manufacturer, 400 µM dNTP, and 1.2 µM of each of the

two primers (Table 1, E+F and H+I respectively). The conditions for the

amplification reactions were as follows: 30 s at 95°C, 30 s at 60°C and 1 min

at 72°C repeated for 25 cycles. The obtained fragments served as primers

EF and HI in the subsequent fusion-PCR.

For the fusion-PCR, the reaction mix of 50 µl consisted of 50 ng

(pMODV-prM+E or p4-SnaBI) plasmid, 2 units of Pfu DNA polymerase


and 1.2 µM of each of the two primers (primers EF+G and HI+J

respectively). The conditions for the amplification reactions were as follows:

1 min at 95°C, 1 min at 59°C and 2.5 min at 72°C repeated for 35 cycles.

Two fragments of 2162 and 1619 bp were thus obtained.

Construction of the recombinant plasmid

The fragment of 2162 bp and the p4-SnaBI vector were digested

with SnaBI and XhoI (Promega). Ligation of 100 ng vector with 30 ng PCR

fragment was carried out using the T4 DNA ligase in a “2x rapid ligation

buffer” (New England Biolabs GmbH). The resulting plasmid p4-MO1 was

digested with XhoI and StuI (Promega) and served as a vector for the

insertion of the XhoI-StuI digested 1619 bp fragment. This resulted in the

plasmid pMODV/DENV-4.

The recombinant region was sequenced in a cycle sequencing

reaction with fluorescent dye terminators (Big Dye Terminator Cycle

Sequencing Ready Reaction kit, Applied Biosystems Division) and analyzed


using an ABI 373 automatic sequencer (Perkin-Elmer, Applied Biosystems

Division). Mutations were corrected by means of site-directed mutagenesis.

Chapter 8

RESULTS

8.1. Introduction of a unique restriction site in plasmid p4

To exchange the prM and E genes of MODV, two unique

restriction sites surrounding the prM and E genes of DENV-4 in p4 had to

be available. At nt-position 3619 a unique recognition site for the restriction

enzyme StuI is available. However, plasmid p4 contains no unique

restriction site upstream of the C/prM hinge region. Therefore, we inserted

a restriction site by means of site-directed mutagenesis for a restriction

enzyme, i.e. SnaBI, that is otherwise not present in the p4 sequence.

8.2. Construction of chimeric viruses by fusion-PCR

The nucleotide sequences at the 3’-termini of the DENV-4 C and E

genes do not contain restriction enzyme sites that allow an exchange of the

DENV-4 prM and E genes by the corresponding genes of MODV.

Therefore site-directed mutagenesis or fusion-PCR appeared to be required

to create the desired gene fusions. Two fusion-PCRs were designed to

obtain the chimeric clone pMODV/DENV-4 (Fig. 1).

DENV-4 C/MODV prM gene fusion

A short DENV-4/MODV fragment (343 b) containing the 319 3’-

terminal nucleotides of the DENV-4 C gene and the first 24 nucleotides of

the MODV prM gene was obtained by performing a standard hot start PCR

using the p4-SnaBI plasmid as template. To this end, a sense primer E

located in the C gene of DENV-4 and an antisense chimeric primer F

(Table 1), consisting of the last 24 nucleotides of the sequence of the

DENV-4 C gene and the first 24 nucleotides of the sequence of the MODV

prM gene, were used. The amplified fragment served as sense primer,

together with the antisense primer G, to generate a 2162 bp fragment by

fusion-PCR using the pMODV-prM+E plasmid as template.


312 bp

1619 bpStuIXhoI

343 bp

DENV-4

MODV

vector

2162 bpSnaBI XhoI

P4-SnaBI

C

NS1prME

NS3

4A4BNS5

NS

2A

2B

StuISnaBI

JC

NS1prM

E

pMODV/DENV-4(15,231 bp)

NS

3

4A4BNS5

NS

2A2

B

5’ UTR

3’ UTR

prM

E

AflII


P4-SnaBI

(15,270 bp)

F

C

NS1prME

NS3

4A4BNS5

NS

2A

2B

5’ UTR

3’ UTR

E

prM

E

C

NS

1

pMODV-prM+E

prM

E

G

prM

E

C

NS

1

pMODV-prM+E

prM

E

H

I

P4-SnaBI

C

NS1prME

NS3

4A4BNS5

NS

2A

2BEF

HI

312 bp

1619 bpStuIXhoI

343 bp

DENV-4

MODV

vector

2162 bpSnaBI XhoI

P4-SnaBI

C

NS1prME

NS3

4A4BNS5

NS

2A

2B

StuISnaBI

P4-SnaBI

C

NS1prME

NS3

4A4BNS5

NS

2A

2B

StuISnaBI

JC

NS1prM

E


NS

3

4A4BNS5

NS

2A2

B

5’ UTR

3’ UTR

prM

E

AflII

C

NS1prM

E


NS

3

4A4BNS5

NS

2A2

B

5’ UTR

3’ UTR

prM

E

AflII


P4-SnaBI

(15,270 bp)

F

C

NS1prME

NS3

4A4BNS5

NS

2A

2B

5’ UTR

3’ UTR

E

P4-SnaBI

(15,270 bp)

F

C

NS1prME

NS3

4A4BNS5

NS

2A

2B

5’ UTR

3’ UTR

E

prM

E

C

NS

1

pMODV-prM+E

prM

E

G

prM

E

C

NS

1

pMODV-prM+E

prM

E

H

I

P4-SnaBI

C

NS1prME

NS3

4A4BNS5

NS

2A

2BEF

HI

Fig. 1. Construction of MODV/DENV-4 chimera by using a variant of the fusion-PCR.

Chapter 8

MODV E/DENV-4 NS1 gene fusion

A short MODV/YFV fragment (312 bp) containing the 288 3’-

terminal nucleotides of the MODV E gene and the first 24 nucleotides of

the DENV-4 NS1 gene was obtained by a standard PCR using pMODV-

prM+E as a template. To this end, a sense primer H located in the E gene

of MODV and an antisense chimeric primer I (Table 1), consisting of the

last 24 nucleotides of the sequence of the MODV E gene and the first 23

nucleotides of the sequence of the DENV-4 NS1 gene, were used. After

purification, this amplified fragment was subsequently used as a sense

primer, in combination with the antisense primer J to generate a 1619 bp

fragment by fusion-PCR using the p4-SnaBI plasmid as template.

8.3. Cloning of MODV prM and E genes

The fragment of 2162 bp was digested with SnaBI and XhoI and

cloned into SnaBI and XhoI digested p4-SnaBI. The resulting pMO1

contained the fusion between DENV-4 C and MODV prM. Plasmid pMO1

was subsequently cut with XhoI and StuI to serve as a vector for the XhoI

and StuI digested 1619 bp fragment which contained the MODV

E/DENV-4 NS1 fusion. To simplify Figure 1, the two-step ligation of the

2162 and 1619 bp fragments is presented as one cloning step. This resulted

in the construction of pMODV/DENV-4 which contains the chimeric

MODV/DENV-4 cDNA that consists of full-length DENV-4 cDNA in

which the DENV-4 prM and E genes in the DENV-4 ORF were replaced

by the MODV prM and E genes. At the 5’ end, this cDNA was directly

fused to a Sp6 promotor to facilitate run-off transcription of full-length,

infectious RNA.

Sequencing of the chimeric region confirmed the expected sequence,

except for two mutations that were corrected by site-directed mutagenesis.


DISCUSSION

The wt DENV-4 does not readily replicate in mice or other small

laboratory animals. Our aim was to engineer a DENV-4 that is able to

replicate readily in mice. MODV is lethal to mice following i.p. inoculation.

In Chapter 5, we reported that a chimeric virus, that consists of YFV 17D

from which the prM+E genes were replaced by the corresponding genes of

MODV, was unlike the parent YFV 17D, highly infectious to mice. Based

on these observations, we expect the MODV/DENV-4 chimeric virus

(containing the structural proteins of MODV) to replicate efficiently in

mice. It will be interesting to learn whether this chimeric virus is

neuroinvasive and whether it replicates (as shown for DENV-4 in man) in

other organs as well. Whatever the site of replication will be, such virus, that

contains the entire replication machinery of DENV, will be ideally suited to

assess the efficacy of molecules that have proven to be effective in vitro in

inhibiting DENV replication, by blocking the function of one of the

enzymes of the replication machinery of this virus.

C h a p t e r 9

GENERAL DISCUSSION AND PERSPECTIVES

Chapter 9

Flaviviruses are among the most important emerging viruses known

to man. They have a common ancestor of 10-20,000 years old and are

evolving rapidly to fill new ecological niches.

Flaviviruses cause a variety of diseases including encephalitis and

(hemorrhagic) fevers. Currently, there is no specific antiviral therapy

available for the treatment of infections with flaviviruses. Symptomatic

treatment and intensive medical care are essential to improve the chance of

survival in severely infected patients. Early diagnosis of the infection (i.e.

before the onset of serious disease) together with an early start of

symptomatic treatment, favor survival.

The availability of a convenient small animal model would be very

useful for the study of strategies for the treatment or prevention of

flavivirus infections. Monkeys infected with flaviviruses such as YFV,

DENV or JEV may develop infection, but because of the cost involved and

the restricted availability of monkeys, the number of studies that can be

carried out by using these animals is limited. Some flaviviruses (e.g. DENV,

YFV) cause morbidity and mortality in mice, but only when inoculated

intracerebrally and when virus strains are used that have been adapted by

serial passage in the brain of suckling mice. Other flaviviruses may cause

infection in mice following peripheral inoculation, although these viruses

require special research facilities, e.g. BSL3 conditions for manipulation of

JEV, and even BSL4 containment facilities for tick-borne encephalitis

viruses such as Russian spring-summer encephalitis virus (RSSEV) and

Omsk hemorrhagic fever virus (OHFV), which of course is an obstacle for

intensive research with animals.

Ideally, a virus used in a model for the evaluation of novel antiviral

strategies should (i) cause morbidity and mortality in small (adult) laboratory

animals following systemic inoculation, (ii) not be pathogenic to man, and

(iii) mimick human disease. Both Modoc virus and Montana Myotis

General discussion and perspectives

leukoencephalitis virus fulfill these criteria. MODV is a murine flavivirus

that was originally isolated from white-footed deer mice (Peromyscus

maniculatus) collected in Modoc county, California, in 1958. The bat

flavivirus MMLV was first isolated in 1958 from a mouse bitten in a

laboratory condition by a naturally infected little brown bat (Myotis lucifugus)

captured in Western Montana. Both flaviviruses have been assigned to the

cluster of the not known vector flaviviruses.

We wanted to develop novel models for the study of antiviral

therapy against flavivirus infections in rodents. Therefore, we sequenced

and characterized the complete genome of MMLV and studied the

characteristics of this virus in vitro and in vivo. Furthermore, we wanted to

make flaviviruses that cause disease in man (such as YFV and DENV), but

that are not infectious to adult mice upon systemic inoculation, infectious to

mice. We also hypothesized that detailed comparison of the viruses

belonging to the vector-borne and the NKV cluster should allow to define

the determinants that are responsible for the fact that a given flavivirus is

either able or not to replicate in cells of a vector (mosquitoes or ticks).

1. The genome of MMLV, a NKV flavivirus

The MMLV genome shares the same overall characteristics with

flaviviruses of human importance. First, MMLV has a similar genomic

organization, i.e. the complete genome is 10,690 nucleotides long and

contains one long open reading frame encoding a putative polyprotein of

3374 amino acids, that is flanked by a 5’ and 3’ UTR of, respectively, 108

and 457 nucleotides long. The coding region contains homologous protease

cleavage sites, internal signal sequences and transmembrane sequences.

Furthermore, the virus contains the same conserved motifs, in genes that

are believed to be interesting antiviral targets [NTPase/helicase, serine

protease and RNA-dependent RNA polymerase], as in flaviviruses that

cause disease in man. As a consequence, MMLV may serve in antiviral drug

studies as a good surrogate for flaviviruses that are of medical importance.

Chapter 9

So far, very little is known about the 3’ UTR of the NKV

flaviviruses. We therefore performed a comparative analysis of the RNA

folding of the 3’ UTR of the NKV flaviviruses (in addition to MMLV, we

used Modoc, Rio Bravo and Apoi viruses). Structural elements in the 3’

UTR that are preserved among other flaviviruses have been revealed, as well

as elements that distinguish the NKV from the mosquito- and tick-borne

flaviviruses. Most importantly, the pentanucleotide sequence 5’-CACAG-3’,

which is conserved in all mosquito- and tick-borne flaviviruses, is replaced

by the sequence 5’-C(C/U)(C/U)AG-3’ in the loop of the 3’ long stable

hairpin structure of all four NKV flaviviruses studied. Knowledge of this

pentanucleotide of flaviviruses thus allows to classify a flavivirus as either a

vector-borne or NKV flavivirus. The phenotypical implications of this

altered pentanucleotide motif were further investigated (see below).

2. The MMLV mouse model

We next employed MMLV to elaborate a convenient flavivirus

animal model. So far, no infections in humans have been reported for this

virus, which points to the interesting safety features of this virus for

manipulation in a research laboratory.

MMLV replicates well in Vero cells, which makes the virus

convenient to use in cell culture systems and therefore for antiviral

evaluation. Cells infected with MMLV show dilatation of the endoplasmic

reticulum, a characteristic of flavivirus infection. The virus appears to be

equally sensitive as flaviviruses of human importance (i.e. YFV and DENV)

to a selection of (experimental) antiviral agents.

Experimental inoculation (intraperitoneal, intranasal or direct

intracerebral) of immunodeficient mice with MMLV resulted in encephalitis

ultimately leading to death. Viral RNA and/or antigens were detected in the

brain and serum of MMLV-infected SCID mice, but not in any other organ

examined. In the brain, MMLV was detected in the olfactory lobes, the


cerebral cortex, the lymbic structures, the midbrain, cerebellum and medulla

oblongata. Infection was confined to neurons. To validate the MMLV-model

for use in antiviral drug studies, we treated SCID mice with the interferon

α/β inducer poly(I).poly(C) before inoculation with MMLV. The mice were

protected against MMLV-induced morbidity and mortality, and this

protection correlated with a reduction in infectious virus titer and viral RNA

load. This validates the model for the study of chemoprophylactic or

chemotherapeutic strategies against flavivirus infections causing encephalitis.

3. Construction and characterization of chimeric flaviviruses

consisting of the backbone of YFV or DENV and the prM

and E genes of a NKV flavivirus

So far, chimeric flaviviruses consisting of part of a vector-borne and

NKV flavivirus have not been constructed. Our aim was to construct such a

chimeric virus that consists of the backbone of a vector-borne flavivirus, i.e.

either YFV 17D or DENV-4, from which the prM and E genes were

replaced by the corresponding genes of the NKV flavivirus MODV. The

study of the particular characteristics of such chimeric virus is important,

because (i) it allows to study whether flaviviruses of clinical importance that

are virtually not infectious to mice, can be rendered infectious to small

rodents when they contain the prM+E genes of a virus, such as MODV that

is highly infectious to mice, (ii) it allows to further define the role of the

prM+E genes in the neuroinvasive properties of flaviviruses (YFV and

DENV are not infectious to mice, whereas MODV is highly neuroinvasive)

and (iii) it allows to study whether the prM+E genes of flaviviruses

determine the tropism for a vector (mosquito or tick).

So far, full-length cDNAs of chimeric flaviviruses have been

constructed by restriction-enzyme cleavage of the gene(s) to be exchanged

or by fusion-PCR of two amplified PCR fragments. We developed a faster

and straightforward approach for the construction of chimeric (flavi)viruses.

Instead of three amplification reactions with four primers as used in a

Chapter 9

“classical fusion reaction”, a fusion between two different genes was

accomplished by only two amplification reactions (and the use of three

primers), thereby reducing the risk of accumulation of mutations.

Using this methodology, a chimeric MODV/YFV 17D was

engineered in which the structural prM and E genes of YFV 17D were

replaced by the homologous genes of MODV. In two PCR steps, a fusion

was made between the 3’ end of the C gene of YFV and the 5’ end of the

prM gene of MODV, and between the 3’ end of the E gene of MODV and

the 5’ end of the NS1 gene of YFV. For each of the two fusions between

YFV and MODV, a standard PCR was performed to amplify a short

fragment with one overlapping end that could be used as one of the primers

in the subsequent (fusion) PCR. Compared to the restriction method, which

requires the incorporation of mutations to create unique restriction sites, the

fusion-PCR strategy has the advantage that the gene fusion can be made at

any position within the nucleotide sequence.

The resulting MODV/YFV chimeric virus was characterized in vitro

and in vivo. MODV/YFV replicated (in Vero cells) about as efficiently as the

parental viruses. In SCID mice that had been infected intraperitoneally with

the chimeric virus, viremia developed at day 3 post infection, and the viral

load increased steadily. The MODV/YFV virus, like MODV from which it

had acquired the prM and E envelope proteins, but unlike YFV(17D),

proved neuroinvasive in suckling and adult SCID mice. All animals

developed neurological symptoms (including paralysis) and showed 100%

mortality. This provides evidence that by adapting YFV by chimerization, a

flavivirus that is responsible for disease in man, can be made infectious to

mice (following infection by the intraperitoneal route). This chimeric virus

contains the entire replication machinery of YFV and can thus be used to

study antiviral strategies against YFV in a mouse model. The distribution of

MODV/YFV RNA, as detected by in situ hybridisation, was similar to that

observed in MODV-infected mice, which provides compelling evidence that


the determinants of neuroinvasiveness of flaviviruses are entirely located in

the envelope proteins prM and E.

To study whether the same approach can be followed to generate

DENV that is readily infectious to mice, we engineered a chimeric

MODV/DENV-4 virus in which the structural prM and E genes of

DENV-4 were replaced by the homologous genes of MODV. In two PCR

steps, a fusion was made between the 3’ end of the C gene of DENV-4 and

the 5’ end of the prM gene of MODV, and between the 3’ end of the E

gene of MODV and the 5’ end of the NS1 gene of DENV-4.

Because MODV is not infectious to mosquito cells, which is in line

with the fact that the virus is not transmitted by mosquitoes, we speculated

that MODV/YFV would, in contrast to the parent virus YFV 17D, not be

infectious to mosquito cells. However, in contrast to expectations,

MODV/YFV appeared to replicate equally well in mosquito cells as in

vertebrate cells. This implicates that determinants other than prM+E in the

viral genome are responsible for the fact that a flavivirus is vector-borne or

a non vector-borne. In Chapter 3 we identified an altered pentanucleotide

motif in the genome of NKV flaviviruses when compared to vector-borne

flaviviruses.

The very 3’ terminus of the 3’ UTR (region IV) of the flavivirus

genome folds in a manner typical for all flaviviruses, forming a 3’ LSH

structure and a small stem-loop. It is suggested that the 3’ LSH plays a

crucial role in the replication of all flaviviruses. In particular, the presence of

the highly conserved pentanucleotide 5’-CACAG-3’ in the top loop has

been suggested to play an important role in virus replication (Khromykh et

al., 2001). Analysis of the 3’ UTR sequence of NKV flaviviruses revealed

that the pentanucleotide sequence (5’-CACAG-3’) which is conserved in all

mosquito- and tick-borne flaviviruses, consists of the sequence 5’-

C(C/U)(C/U)AG-3’ in the genome of all four NKV flaviviruses. Because

Chapter 9

prM+E has no effect on vector specificity, we investigated the possible

impact of this motif on the vector specificity of flaviviruses. We therefore

carried out site-directed mutagenesis on the pentanucleotide sequence of the

chimeric MODV/YFV virus, and evaluated whether the resulting mutant

[named MODV/YFV(CUCAG)] is able or not to replicate in mosquito

(C6/36) cells. The replication kinetics of MODV/YFV(CUCAG) were

compared with those of MODV, YFV 17D and the original chimeric

MODV/YFV. The MODV/YFV(CUCAG) replicated as efficiently in Vero

cells as MODV, YFV 17D and MODV/YFV(CACAG), but failed to

replicate in mosquito cells in contrast to MODV/YFV(CACAG). These

data provide compelling evidence that the pentanucleotide motif in the 3’

LSH determines whether a flavivirus is able to replicate in mosquito cells or

not. We are currently constructing a YFV 17D virus that contains the

CUCAG sequence to definitely corroborate this finding.

It will now be important to assess whether mutating the

pentanucleotide motif CUCAG of a NKV flavivirus to the CACAG

sequence of vector-borne flaviviruses results in NKV viruses that are

infectious to vector (mosquito or tick) cells. To this end, we need to start

from full-length infectious clones of NKV flaviviruses. Such clones do not

exist and we therefore decided to generate such clone.

A problem in the construction of infectious clones of flaviviruses is

the synthesis of full-length cDNA. To address this problem we have devised

a rapid and simple method to generate a full-length long RT-PCR product

using viral RNA isolated from either (i) the supernatant of MODV-infected

cells or (ii) an unpurified MODV-infected mouse brain suspension. These

developments represent a significant advance in recombinant technology

and the first method should be applicable to positive-stranded RNA viruses

which do not replicate to high titers in mouse brain.


Full-length clones will now be used to further evaluate the function

of the CACAG motif in the vector specificity and will also be a particular

importance to understand the biology of NKV flaviviruses.

REFERENCES

References

Albrecht, P. (1968). Pathogenesis of neurotropic arbovirus infections. Curr. Top. Microbiol.

Immunol. 43, 44-91. Allison, S.L., Schalich, J., Stiasny, K., Mandl, C.W., Kunz, C., and Heinz, F.X. (1995).

Oligomeric rearrangement of tick-borne encephalitis virus envelope proteins induced by an acidic pH. J. Virol. 69, 695-700.

Allison, S.L., Schalich, J., Stiasny, K., Mandl, C.W., and Heinz, F.X. (2001). Mutational evidence for an internal fusion peptide in flavivirus envelope protein E. J. Virol.

75, 4268-4275. Amberg, D.C., Botstein, D., and Beasley, E.M. (1995). Precise gene disruption in

Saccharomyces cerevisiae by double fusion polymerase chain reaction. Yeast 11, 1275-1280.

Amberg, S.M., and Rice, C.M. (1999). Mutagenesis of the NS2B-NS3-mediated cleavage site in the flavivirus capsid protein demonstrates a requirement for coordinated processing. J. Virol. 73, 8083-8094.

An, J., Kimura-Kuroda, J., Hirabayashi, Y., and Yasui, K. (1999). Development of a novel mouse model for dengue virus infection. Virology 263, 70-77.

Arroyo, J., Guirakhoo, F., Fenner, S., Zhang, Z.X., Monath, T.P., and Chambers, T.J. (2001). Molecular basis for attenuation of neurovirulence of a yellow fever virus/Japanese encephalitis virus chimera vaccine (ChimeriVax-JE). J. Virol. 75, 934-942.

Asnis, D.S., Conetta, R., Teixeira, A.A., Waldman, G., and Sampson, B.A. (2000). The West Nile Virus outbreak of 1999 in New York: the Flushing Hospital experience. Clin.

Infect. Dis. 30, 413-418. Barrett, A.D., and Gould, E.A. (1986). Comparison of neurovirulence of different strains of

yellow fever virus in mice. J. Gen. Virol. 67, 631-637. Barrett, P.N., Dorner, F., and Plotkin, S.A. (1999). Tick-borne encephalitis vaccine. Plotkin,

S., Orenstein, W.(Eds.), Vaccines, 3rd Edition, WB Saunders Company, Philadelphia, 767-780.

Beasley, D.W., Li, L., Suderman, M.T., and Barrett, A.D. (2002). Mouse neuroinvasive phenotype of West Nile virus strains varies depending upon virus genotype. Virology 296, 17-23.

Bell, J.F., and Thomas, L.A. (1964). A new virus, "MML", enzootic in bats (Myotis lucifugus) of Montana. Am. J. Trop. Med. Hyg. 607-612.

Bhamarapravati, N., Yoksan, S., Chayaniyayothin, T., Angsubphakorn, S., and Bunyaratvej, A. (1987). Immunization with a live attenuated dengue-2-virus candidate vaccine (16681-PDK 53): clinical, immunological and biological responses in adult volunteers. Bull. World Health Organ 65, 189-195.

Bhamarapravati, N., and Sutee, Y. (2000). Live attenuated tetravalent dengue vaccine. Vaccine 18, 44-47.

Bhardwaj, S., Holbrook, M., Shope, R.E., Barrett, A.D., and Watowich, S.J. (2001). Biophysical characterization and vector-specific antagonist activity of domain III of the tick-borne flavivirus envelope protein. J. Virol. 75, 4002-4007.

Bhatt, T.R., Crabtree, M.B., Guirakhoo, F., Monath, T.P., and Miller, B.R. (2000). Growth characteristics of the chimeric Japanese encephalitis virus vaccine candidate, ChimeriVax-JE (YF/JE SA14--14--2), in Culex tritaeniorhynchus, Aedes albopictus, and Aedes aegypti mosquitoes. Am. J. Trop. Med. Hyg. 62, 480-484.

Billoir, F., De Chesse, R., Tolou, H., De Micco, P., Gould, E.A., and de Lamballerie, X. (2000). Phylogeny of the genus Flavivirus using complete coding sequences of arthropod-borne viruses and viruses with no known vector. J. Gen. Virol. 81, 781-790.

Blackwell, J.L., and Brinton, M.A. (1995). BHK cell proteins that bind to the 3' stem-loop structure of the West Nile virus genome RNA. J. Virol. 69, 5650-5658.

References

Blackwell, J.L., and Brinton, M.A. (1997). Translation elongation factor-1 alpha interacts with the 3' stem-loop region of West Nile virus genomic RNA. J. Virol. 71, 6433-6444.

Blaney, J.E., Johnson, D.H., Manipon, G.G., Firestone, C.Y., Hanson, C.T., Murphy, B.R., and Whitehead, S.S. (2002). Genetic basis of attenuation of dengue virus type 4 small plaque mutants with restricted replication in suckling mice and in SCID mice transplanted with human liver cells. Virology 300, 125-139.

Blok, J., McWilliam, S.M., Butler, H.C., Gibbs, A.J., Weiller, G., Herring, B.L., Hemsley, A.C., Aaskov, J.G., Yoksan, S., and Bhamarapravati, N. (1992). Comparison of a dengue-2 virus and its candidate vaccine derivative: sequence relationships with the flaviviruses and other viruses. Virology 187, 573-590.

Boyer, J.C., and Haenni, A.L. (1994). Infectious transcripts and cDNA clones of RNA viruses. Virology 198, 415-426.

Bray, M., and Lai, C.J. (1991). Construction of intertypic chimeric dengue viruses by substitution of structural protein genes. Proc. Natl. Acad. Sci. U.S.A. 88, 10342-10346.

Bray, M., Men, R., and Lai, C.J. (1996). Monkeys immunized with intertypic chimeric dengue viruses are protected against wild-type virus challenge. J. Virol. 70, 4162-4166.

Bray, M., Men, R., Tokimatsu, I., and Lai, C.J. (1998). Genetic determinants responsible for acquisition of dengue type 2 virus mouse neurovirulence. J. Virol. 72, 1647-1651.

Bredenbeek, P.J, Kooi, E.A., Lindenbach, B., Huijkman, N., Rice, C.M., and Spaan, W.J. (2003). A stable full-length yellow fever virus cDNA clone and the role of conserved RNA elements in flavivirus replication. J. Gen. Virol. 84, 1261-1268.

Breitschopf, H., Suchanek, G., Gould, R.M., Coldman, D.R., and Lassmann, H. (1992). In situ hybridisation with digoxigenin-labeled probes: sensitive and reliable detection method applied to myelinating rat brain. Acta Neuropathol. Berl. 84, 581-587.

Briese, T., Jia, X.Y., Huang, C., Grady, L.J., and Lipkin, W.I. (1999). Identification of a Kunjin/West Nile-like flavivirus in brains of patients with New York encephalitis. Lancet 354, 1261-1262.

Brinton, M.A., Fernandez, A.V., and Dispoto, J.H. (1986). The 3'-nucleotides of flavivirus genomic RNA form a conserved secondary structure. Virology 153, 113-121.

Burke, D.S., and Monath, T.P. (2001). Flaviviruses. In Fields Virology, eds. Knipe, D. M. and Howley, P. M., pp. 1043-1125. Lippincott Williams and Wilkins, Philadelphia.

Butrapet, S., Kimura-Kuroda, J., Zhou, D.S., and Yasui, K. (1998). Neutralizing mechanism of a monoclonal antibody against Japanese encephalitis virus glycoprotein E. Am.

J. Trop. Med. Hyg. 58, 389-398. Butrapet, S., Huang, C.Y., Pierro, D.J., Bhamarapravati, N., Gubler, D.J., and Kinney, R.M.

(2000). Attenuation markers of a candidate dengue type 2 vaccine virus, strain 16681 (PDK-53), are defined by mutations in the 5' noncoding region and non-structural proteins 1 and 3. J. Virol. 74, 3011-3019.

Cahour, A., Pletnev, A., Vazielle-Falcoz, M., Rosen, L., and Lai, C.J. (1995). Growth-restricted dengue virus mutants containing deletions in the 5' noncoding region of the RNA genome. Virology 207, 68-76.

Calisher, C.H., Karabatsos, N., Dalrymple, J.M., Shope, R. E., Porterfield, J.S., Westaway, E.G., and Brandt, W.E. (1989). Antigenic relationships between flaviviruses as determined by cross-neutralization tests with polyclonal antisera. J. Gen. Virol. 70, 37-43.

Cammisa-Parks, H., Cisar, L.A., Kane, A., and Stollar, V. (1992). The complete nucleotide sequence of cell fusing agent (CFA): homology between the nonstructural proteins encoded by CFA and the nonstructural proteins encoded by arthropod-borne flaviviruses. Virology 189, 511-524.

References

Campbell, M.S., and Pletnev, A.G. (2000). Infectious cDNA clones of Langat tick-borne flavivirus that differ from their parent in peripheral neurovirulence. Virology 269, 225-237.

Caufour, P.S., Motta, M.C., Yamamura, A.M., Vazquez, S., Ferreira, I.I., Jabor, A.V., Bonaldo, M.C., Freire, M.S., and Galler, R. (2001). Construction, characterization and immunogenicity of recombinant yellow fever 17D-dengue type 2 viruses. Virus Res. 79, 1-14.

Cecilia, D., and Gould, E.A. (1991). Nucleotide changes responsible for loss of neuroinvasiveness in Japanese encephalitis virus neutralization-resistant mutants. Virology 181, 70-77.

Chambers, T.J., Hahn, C.S., Galler, R., and Rice, C.M. (1990a). Flavivirus genome organization, expression, and replication. Annu. Rev. Microbiol. 44, 649-688.

Chambers, T.J., Weir, R.C., Grakoui, A., McCourt, D.W., Bazan, J.F., Fletterick, R.J., and Rice, C.M. (1990b). Evidence that the N-terminal domain of nonstructural protein NS3 from yellow fever virus is a serine protease responsible for site-specific cleavages in the viral polyprotein. Proc. Natl. Acad. Sci. U.S.A. 87, 8898-8902.

Chambers, T.J., Halevy, M., Nestorowicz, A., Rice, C.M., and Lustig, S. (1998). West Nile virus envelope proteins: nucleotide sequence analysis of strains differing in mouse neuroinvasiveness. J. Gen. Virol. 79, 2375-2380.

Chambers, T.J., Nestorowicz, A., Mason, P.W., and Rice, C.M. (1999). Yellow fever Japanese encephalitis chimeric viruses: Construction and biological properties. J. Virol. 73, 3095-3101.

Chambers, T.J., and Nickells, M. (2001). Neuroadapted Yellow fever virus 17D: genetic and biological characterization of a highly mouse-neurovirulent virus and its infectious molecular clone. J. Virol. 75, 10912-10922.

Chambers, T.J., Liang, Y., Droll, D.A., Schlesinger, J.J., Davidson, A.D., Wright, P.J., and Jiang, X. (2003). Yellow fever virus/dengue-2 virus and yellow fever virus/dengue-4 virus chimeras: biological characterization, immunogenicity, and protection against dengue encephalitis in the mouse model. J. Virol. 77, 3655-3668.

Charlier, N., Leyssen, P., Paeshuyse, J., Drosten, C., Schmitz, H., Van, Lommel, A., De Clercq, E., and Neyts, J. (2002a). Infection of SCID mice with Montana Myotis leukoencephalitis virus as a model for flavivirus encephalitis. J. Gen. Virol. 83, 1887-1896.

Charlier, N., Leyssen, P., Pleij, C.W., Lemey, P., Billoir, F., Van Laethem, K., Vandamme, A.M., De Clercq, E., de Lamballerie, X., and Neyts, J. (2002b). Complete genome sequence of Montana Myotis leukoencephalitis virus, phylogenetic analysis and comparative study of the 3’ untranslated region of flaviviruses with no known vector. J. Gen. Virol. 83, 1875-1885.

Charlier, N., Molenkamp, R., Leyssen, P., Vandamme, A.M., De Clercq, E., Bredenbeek, P., and Neyts, J. (2003). A rapid and convenient variant of fusion-PCR to construct chimeric flaviviruses. J. Virol. Methods 108, 67-74.

Chen, C.J., Kuo, M.D., Chien, L.J., Hsu, S.L., Wang, Y.M., and Lin, J.H. (1997). RNA-protein interactions: involvement of NS3, NS5, and 3' noncoding regions of Japanese encephalitis virus genomic RNA. J. Virol. 71, 3466-3473.

Chen, W., Kawano, H., Men, R., Clark, D., and Lai, C.J. (1995). Construction of intertypic chimeric dengue viruses exhibiting type 3 antigenicity and neurovirulence for mice. J. Virol. 69, 5186-5190.

Chiba, N., Iwasaki, T., Mizutani, T., Kariwa, H., Kurata, T., and Takashima, L.. (1999). Pathogenicity of tick-borne encephalitis virus isolated in Hokkaido, Japan in mouse model. Vaccine 17, 779-787.

References

Crill, W.D., and Roehrig, J.T. (2001). Monoclonal antibodies that bind to domain III of dengue virus E glycoprotein are the most efficient blockers of virus adsorption to Vero cells. J. Virol. 75, 7769-7773.

De Clercq, E., Cools, M., Balzarini, J., Snoeck, R., Andrei, G., Hosoya, M., Shigeta, S., Ueda, T., Minakawa, N., and Matsuda, A. (1991). Antiviral activities of 5-ethynyl-1-beta-D-ribofuranosylimidazole-4-carboxamide and related compounds. Antimicrob. Agents Chemother. 35, 679-684.

Dekker, A., Leendertse, C.H., van Poelwijk, F., Rebel, J.M., and Moormann, R.J. (2000). Chimeric swine vesicular disease viruses produced by fusion-PCR: a new method for epitope mapping. J. Virol. Methods 86, 131-141.

De Silva, D., Reiser, A., Hermann, M., Tabiti, K., and Wittwer, C. (1998). Rapid genotyping and quantification on the LightCycler with hybridisation probes. Biochemica 2, 12-15.

Despres, P., Frenkiel, M.P., and Deubel, V. (1993). Differences between cell membrane fusion activities of two dengue type-1 isolates reflect modifications of viral structure. Virology 196, 209-219.

Drosten, C., Gottig, S., Schilling, S., Asper, M., Panning, M., Schmitz, H., and Gunther, S. (2002). Rapid detection and quantification of RNA of Ebola and Marburg viruses, Lassa virus, Crimean-Congo hemorrhagic fever virus, Rift Valley fever virus, dengue virus, and yellow fever virus by real-time reverse transcription-PCR. J. Clin. Microbiol. 40, 2323-2330.

Durbin, A.P., Karron, R.A., Sun, W., Vaughn, D.W., Reynolds, M.J., Perreault, J.R., Thumar, B., Men, R., Lai, C.J., Elkins, W.R., Chanock, R.M., Murphy, B.R., and Whitehead, S.S. (2001). Attenuation and immunogenicity in humans of a live dengue virus type-4 vaccine candidate with a 30 nucleotide deletion in its 3' untranslated region. Am. J. Trop. Med. Hyg. 65, 405-413.

Edelman, R., Tacket, C.O., Wasserman, S.S., Vaughn, D.W., Eckels, K.H., Dubois, D.R., Summers, P.L., and Hoke, C.H. (1994). A live attenuated dengue-1 vaccine candidate (45AZ5) passaged in primary dog kidney cell culture is attenuated and immunogenic for humans. J. Infect. Dis. 170, 1448-1455.

Eldadah, A.H., and Nathanson, N. (1967). Pathogenesis of West Nile Virus encepahlitis in mice and rats. II. Virus multiplication, evolution of immunofluorescence, and development of histological lesions in the brain. Am. J. Epidemiol. 86, 776-790.

Fitzgeorge, R., and Bradish, C.J. (1980). The in vivo differentiation of strains of yellow fever virus in mice. J. Gen. Virol. 46, 1-13.

Goverdhan, M.K., Kulkarni, A.B., Gupta, A.K., Tupe, C.D., and Rodrigues, J.J. (1992). Two-way cross-protection between West Nile and Japanese encephalitis viruses in bonnet macaques. Acta Virol. 36, 277-283.

Grange, T., Bouloy, M., and Girard, M. (1985). Stable secondary structures at the 3'-end of the genome of yellow fever virus (17 D vaccine strain). FEBS Lett. 188, 159-163.

Gresikova, M., and Beran, G.W. (1981). Tick-borne encephalitis. Beran G.W., ed.CRC handbook series of zoonoses, section B: Viral zoonoses.Vol I. Boca Raton, FL: CRC Press, 201-208.

Gresikova, M., and Calisher, C.H. (1988). Tick-borne encephalitis. Monath TP, ed.The arboviruses: Ecology and epidemiology, Vol IV. Boca Raton, FL: CRC Press , 177-203.

Gritsun, T.S., and Gould, E.A. (1995). Infectious transcripts of tick-borne encephalitis virus, generated in days by RT-PCR. Virology 214, 611-618.

Gritsun, T.S., Venugopal, K., Zanotto, P.M., Mikhailov, M.V., Sall, A.A., Holmes, E.C., Polkinghorne, I., Frolova, T.V., Pogodina, V.V., Lashkevich, V.A., and Gould, E.A. (1997). Complete sequence of two tick-borne flaviviruses isolated from Siberia and the UK: analysis and significance of the 5' and 3'-UTRs. Virus Res. 49, 27-39.

References

Gritsun, T.S., and Gould, E.A. (1998). Development and analysis of a tick-borne encephalitis virus infectious clone using a novel and rapid strategy. J. Virol. Methods 76, 109-120.

Gritsun, T.S., Desai, A., and Gould, E.A. (2001). The degree of attenuation of tick-borne encephalitis virus depends on the cumulative effects of point mutations. J. Gen.

Virol. 82, 1667-1675. Gritsun, T.S., Lashkevich, V.A., and Gould, E.A. (2003). Tick-borne encephalitis. Antiviral

Res. 57, 129-146. Gualano, R.C., Pryor, M.J., Cauchi, M.R., Wright, P.J., and Davidson, A.D. (1998).

Identification of a major determinant of mouse neurovirulence of dengue virus type 2 using stably cloned genomic-length cDNA. J. Gen. Virol. 79, 437-446.

Gubler, D.J. (1997). Dengue and dengue hemorrhagic fever: its history and resurgence as a global public health problem. D.J. Gubler and G. Kuno (ed.), Dengue and dengue hemorrhagic fever. CAB International, Wallingford, United Kingdom , 1-23.

Gubler, D.J. (1998). Dengue and dengue hemorrhagic fever. Clin. Microbiol. Rev. 11, 480-496. Guirakhoo, F., Bolin, R.A., and Roehrig, J.T. (1992). The Murray Valley encephalitis virus

prM protein confers acid resistance to virus particles and alters the expression of epitopes within the R2 domain of E glycoprotein. Virology 191, 921-931.

Guirakhoo, F., Zhang, Z.X., Chambers, T.J., Delagrave, S., Arroyo, J., Barrett, A.D., and Monath, T.P. (1999). Immunogenicity, genetic stability, and protective efficacy of a recombinant, chimeric yellow fever-Japanese encephalitis virus (ChimeriVax-JE) as a live, attenuated vaccine candidate against Japanese encephalitis. Virology 257, 363-372.

Guirakhoo, F., Weltzin, R., Chambers, T.J., Zhang, Z.X., Soike, K., Ratterree, M., Arroyo, J., Georgakopoulos, K., Catalan, J., and Monath, T.P. (2000). Recombinant chimeric yellow fever-dengue type 2 virus is immunogenic and protective in nonhuman primates. J. Virol. 74, 5477-5485.

Guirakhoo, F., Arroyo, J., Pugachev, K.V., Miller, C., Zhang, Z.X., Weltzin, R., Georgakopoulos, K., Catalan, J., Ocran, S., Soike, K., Ratterree, M., and Monath, T.P. (2001). Construction, safety, and immunogenicity in nonhuman primates of a chimeric yellow fever-dengue virus tetravalent vaccine. J. Virol. 75, 7290-7304.

Guirakhoo, F., Pugachev, K., Arroyo, J., Miller, C., Zhang, Z.X., Weltzin, R., Georgakopoulos, K., Catalan, J., Ocran, S., Draper, K., and Monath, T.P. (2002). Viremia and immunogenicity in nonhuman primates of a tetravalent yellow fever-dengue chimeric vaccine: genetic reconstructions, dose adjustment, and antibody responses against wild-type dengue virus isolates. Virology 298, 146-159.

Gultyaev, A.P., Van-Batenburg, F.H., and Pleij, C.W. (1995). The computer simulation of RNA folding pathways using a genetic algorithm. J. Mol. Biol. 250, 37-51.

Hahn, C.S., Hahn, Y.S., Rice, C.M., Lee, E., Dalgarno, L., Strauss, E.G., and Strauss, J.H. (1987). Conserved elements in the 3' untranslated region of flavivirus RNAs and potential cyclization sequences. J. Mol. Biol. 198, 33-41.

Halevy, M., Akov, Y., Ben, N.D., Kobiler, D., Lachmi, B., and Lustig, S. (1994). Loss of active neuroinvasiveness in attenuated strains of West Nile virus: pathogenicity in immunocompetent and SCID mice. Arch. Virol. 137, 355-370.

Halstead, S.B., and O'Rourke, E.J. (1977). Antibody-enhanced dengue virus infection in primate leukocytes. Nature 265, 739-741.

Halstead, S.B. (1979). In vivo enhancement of dengue virus infection in rhesus monkeys by passively transferred antibody. J. Infect. Dis. 140, 527-533.

Han, L.L., Popovici, F., Alexander, J.J., Laurentia, V., Tengelsen, L.A., Cernescu, C., Gary, J.H., Ion, N.N., Campbell, G.L., and Tsai, T.F. (1999). Risk factors for West Nile virus infection and meningoencephalitis, Romania, 1996. J. Infect. Dis. 179, 230-233.

References

Harrington, D.G., Hilmas, D.E., Elwell, M.R., Whitmire, R.E., and Stephen, E.I. (1977). Intranasal infection of monkeys with Japanese encephalitis virus: clinical response and treatment with a nuclease-resistant derivative of poly (I).poly (C). Am. J. Trop.

Med. Hyg. 26, 1191-1198. Hase, T., Dubois, D.R., and Summers, P.L. (1990a). Comparative study of mouse brains

infected with Japanese encephalitis virus by intracerebral or intraperitoneal inoculation. Int. J. Exp. Pathol. 71, 857-869.

Hase, T., Summers, P.L., and Dubois, D.R. (1990b). Ultrastructural changes of mouse brain neurons infected with Japanese encephalitis virus. Int. J. Exp. Pathol. 71, 493-505.

Hase, T. (1993a). Virus-neuron interactions in the mouse brain infected with Japanese encephalitis virus. Virchows Arch. B Cell Pathol. Incl. Mol. Pathol. 64, 161-170.

Hase, T., Dubois, D.R., Summers, P.L., Downs, M.B., and Ussery, M.A. (1993b). Comparison of replication rates and pathogenicities between the SA14 parent and SA14-14-2 vaccine strains of Japanese encephalitis virus in mouse brain neurons. Arch. Virol. 130, 131-143.

Hasegawa, H., Yoshida, M., Shiosaka, T., Fujita, S., and Kobayashi, Y. (1992). Mutations in the envelope protein of Japanese encephalitis virus affect entry into cultured cells and virulence in mice. Virology 191, 158-165.

Heinz, F.X., and Mandl, C.W. (1993). The molecular biology of tick-borne encephalitis virus. APMIS 101, 735-745.

Heinz, F.X., Auer, G., Stiasny, K., Holzmann, H., Mandl, C., Guirakhoo, F., and Kunz, C. (1994). The interactions of the flavivirus envelope proteins: implications for virus entry and release. Arch. Virol. 9, 339-348.

Heinz, F.X., and Allison, S.L. (2000a). Structures and mechanisms in flavivirus fusion. Adv.

Virus Res. 55, 231-269. Heinz, F.X., Collet, M. S., Purcell, R.H., Gould, E.A., Howard C.R., Houghton, M.,

Moorman, R.J. M., Rice, C.M., and Thiel, H.J. (2000b). Flaviviridae. In "Virus Taxonomy. Seventh Report of the International Committee for the Taxonomy of Viruses". Edited by M.H.V. Van Regenmortel, C.M. Fauquet, D.H.L. Bishop, C.H. Calisher, E.B. Carsten, M.K. Estes, S.M. Lemon, J. Maniloff, M.A. Mayo, D.J. McGeoch, C.R. Pringle, and R.B. Wickner. New York: Academic Press.

Henikoff, S., and Henikoff, J. (1992). Amino acid substitution matrices for protein blocks. Proc. Natl. Acad. Sci. U.S.A. 89, 10915-10919.

Hennessy, S., Liu, Z., Tsai, T.F., Strom, B.L., Wan, C.M., Liu, H.L., Wu, T.X., Yu, H.J., Liu, Q.M., Karabatsos, N., Bilker, W.B., and Halstead, S.B. (1996). Effectiveness of live-attenuated Japanese encephalitis vaccine (SA14-14-2): a case-control study. Lancet 347, 1583-1586.

Hoke, C.H., Jr., Malinoski, F.J., Eckels, K.H., Scott, R.M., Dubois, D.R., Summers, P.L., Simms, T., Burrous, J., Hasty, S.E., and Bancroft, W.H. (1990). Preparation of an attenuated dengue 4 (341750 Carib) virus vaccine. II. Safety and immunogenicity in humans. Am. J. Trop. Med. Hyg. 43, 219-226.

Holbrook, M.R., Ni, H., Shope, R.E., and Barrett, A.D. (2001). Amino acid substitution(s) in the stem-anchor region of langat virus envelope protein attenuates mouse neurovirulence. Virology 286, 54-61.

Holland, P.M., Abramson, R.D., Watson, R., and Gelfand, D.H. (1991). Detection of specific polymerase chain reaction product by utilizing the 5'----3' exonuclease activity of Thermus aquaticus DNA polymerase. Proc. Natl. Acad. Sci. U.S.A. 88, 7276-7280.

Holzmann, H., Heinz, F.X., Mandl, C.W., Guirakhoo, F., and Kunz, C. (1990). A single amino acid substitution in envelope protein E of tick-borne encephalitis virus leads to attenuation in the mouse model. J. Virol. 64, 5156-5159.

References

Holzmann, H., Stiasny, K., Ecker, M., Kunz, C., and Heinz, F.X. (1997). Characterization of monoclonal antibody-escape mutants of tick-borne encephalitis virus with reduced neuroinvasiveness in mice. J. Gen. Virol. 78, 31-37.

Huang, C.Y., Butrapet, S., Pierro, D.J., Chang, G.J., Hunt, A.R., Bhamarapravati, N., Gubler, D.J., and Kinney, R.M. (2000). Chimeric dengue type 2 (vaccine strain PDK-53)/dengue type 1 virus as a potential candidate dengue type 1 virus vaccine. J. Virol. 74, 3020-3028.

Hurrelbrink, R.J., Nestorowicz, A., and McMinn, P.C. (1999). Characterization of infectious Murray Valley encephalitis virus derived from a stably cloned genome-length cDNA. J. Gen. Virol. 80, 3115-3125.

Jiang, W.R., Lowe, A., Higgs, S., Reid, H., and Gould, E.A. (1993). Single amino acid codon changes detected in louping ill virus antibody-resistant mutants with reduced neurovirulence. J. Gen. Virol. 74, 931-935.

Johnson, H.N. (1967). Ecological implications of antigenically related mammalian viruses for which arthropod vectors are unknown and avian associated soft tick viruses. Jpn. J. Med. Sci. Biol. 20, 160-166.

Johnson, R.T. (1982).Viral infections of the nervous system. New York: Raven Press. Johnson, R.T., Burke, D.S., Elwell, M., Leake, C.J., Nisalak, A., Hoke, C.H., and

Lorsomrudee, W. (1985). Japanese encephalitis: immunocytochemical studies of viral antigen and inflammatory cells in fatal cases. Ann. Neurol. 18, 567-573.

Johnson, A.J., and Roehrig, J.T. (1999). New mouse model for dengue virus vaccine testing. J. Virol. 73, 783-786.

Kalita, J., and Misra, U.K. (1998). EEG in Japanese encephalitis: a clinico-radiological correlation. Electroencephalogr. Clin. Neurophysiol. 106, 238-243.

Kamer, G., and Argos, P. (1984). Primary structural comparison of RNA-dependent polymerases from plant, animal and bacterial viruses. Nucleic. Acids. Res. 12, 7269-7282.

Kanesa-thasan, N., Sun, W., Kim-Ahn, G., Van Albert, S., Putnak, J. R., King, A., Raengsakulsrach, B., Christ-Schmidt, H., Gilson, K., Zahradnik, J. M., Vaughn, D. W., Innis, B. L., Saluzzo, J. F., and Hoke, C.-H. J. (2001). Safety and immunogenicity of attenuated dengue virus vaccines (Aventis Pasteur) in human volunteers. Vaccine 19, 3179-3188.

Kanesa-thasan, N., Putnak, J.R., Mangiafico, J.A., Saluzzo, J.E., and Ludwig, G.V. (2002). Short report: absence of protective neutralizing antibodies to West Nile virus in subjects following vaccination with Japanese encephalitis or dengue vaccines. Am.

J. Trop. Med. Hyg. 66, 115-116. Kapoor, M., Zhang, L., Mohan, P.M., and Padmanabhan, R. (1995). Synthesis and

characterization of an infectious dengue virus type-2 RNA genome (New Guinea C strain). Gene 162, 175-180.

Karreman, C. (1998). Fusion-PCR, a one-step variant of the "megaprimer" method of mutagenesis. Biotechniques 24, 736,740,742.

Kelly, E.P., Greene, J.J., King, A.D., and Innis, B.L. (2000). Purified dengue 2 virus envelope glycoprotein aggregates produced by baculovirus are immunogenic in mice. Vaccine 18, 2549-2559.

Khromykh, A.A., and Westaway, E.G. (1994). Completion of Kunjin virus RNA sequence and recovery of an infectious RNA transcribed from stably cloned full-length cDNA. J. Virol. 68, 4580-4588.

Khromykh, A.A., Meka, H., Guyatt, K.J., and Westaway, E.G. (2001). Essential role of cyclization sequences in flavivirus RNA replication. J. Virol. 75, 6719-6728.

Kinney, R.M., Butrapet, S., Chang, G.J., Tsuchiya, K.R., Roehrig, J.T., Bhamarapravati, N., and Gubler, D.J. (1997). Construction of infectious cDNA clones for dengue 2 virus: strain 16681 and its attenuated vaccine derivative, strain PDK-53. Virology 230, 300-308.

References

Kitazono, A.A., Tobe, B.T., Kalton, H., Diamant, N., and Kron, S.J. (2002). Marker-fusion-PCR for one-step mutagenesis of essential genes in yeast. Yeast 19, 141-149.

Kliks, S.C., Nisalak, A., Brandt, W.E., Wahl, L., and Burke, D.S. (1989). Antibody-dependent enhancement of dengue virus growth in human monocytes as a risk factor for dengue hemorrhagic fever. Am. J. Trop. Med. Hyg. 40, 444-451.

Kobiler, D., Lustig, S., Gozes, Y., Ben, N.D., and Akov, Y. (1989). Sodium dodecylsulphate induces a breach in the blood-brain barrier and enables a West Nile virus variant to penetrate into mouse brain. Brain Res. 496, 314-316.

Kochel, T.J., Raviprakash, K., Hayes, C.G., Watts, D.M., Russell, K.L., Gozalo, A.S., Phillips, I.A., Ewing, D.F., Murphy, G.S., and Porter, K.R. (2000). A dengue virus serotype-1 DNA vaccine induces virus neutralizing antibodies and provides protection from viral challenge in Aotus monkeys. Vaccine 18, 3166-3173.

Kofler, R.M., Leitner, A., O'Riordain, G., Heinz, F.X., and Mandl, C.W. (2003). Spontaneous mutations restore the viability of tick-borne encephalitis virus mutants with large deletions in protein C. J. Virol. 77, 443-451.

Konishi, E., and Mason, P.W. (1993). Proper maturation of the Japanese encephalitis virus envelope glycoprotein requires cosynthesis with the premembrane protein. J. Virol. 67, 1672-1675.

Konishi, E., Yamaoka, M., Khin, S.W., Kurane, I., and Mason, P.W. (1998). Induction of protective immunity against Japanese encephalitis in mice by immunization with a plasmid encoding Japanese encephalitis virus premembrane and envelope genes. J. Virol. 72, 4925-4930.

Konishi, E., Yamaoka, M., Kurane, I., and Mason, P.W. (2000). A DNA vaccine expressing dengue type 2 virus premembrane and envelope genes induces neutralizing antibody and memory B cells in mice. Vaccine 18, 1133-1139.

Koonin, E.V. (1993). Computer-assisted identification of a putative methyltransferase domain in NS5 protein of flaviviruses and lambda 2 protein of reovirus. J. Gen.

Virol. 74, 733-740. Kornfeld, R., and Kornfeld, S. (1985). Assembly of asparagine-linked oligosaccharides.

Annu. Rev. Biochem. 54, 631-664. Kramer, L.D., Presser, S.B., Hardy, J.L., and Jackson, A.O. (1997). Genotypic and

phenotypic variation of selected Saint Louis encephalitis viral strains isolated in California. Am. J. Trop. Med. Hyg. 57, 222-229.

Kroeger, M.A., and McMinn, P.C. (2002). Murray Valley encephalitis virus recombinant subviral particles protect mice from lethal challenge with virulent wild-type virus. Arch. Virol. 147, 1155-1172.

Kroschewski, H., Allison, S.L., Heinz, F.X., and Mandl, C.W. (2003). Role of heparan sulfate for attachment and entry of tick-borne encephalitis virus. Virology 308, 92-100.

Kuno, G., Chang, G.J., Tsuchiya, K.R., Karabatsos, N., and Cropp, C.B. (1998). Phylogeny of the genus Flavivirus. J. Virol. 72, 73-83.

Kuwayama, H., Obara, S., Morio, T., Katoh, M., Urushihara, H., and Tanaka, Y. (2002). PCR-mediated generation of a gene disruption construct without the use of DNA ligase and plasmid vectors. Nucleic Acids Res. 30, E2.

Labuda, M., Kozuch, O., Zuffova, E., Eleckova, E., Hails, R.S., and Nuttall, P.A. (1997). Tick-borne encephalitis virus transmission between ticks cofeeding on specific immune natural rodent hosts. Virology 235, 138-143.

Lai, C.J., Zhao, B.T., Hori, H., and Bray, M. (1991). Infectious RNA transcribed from stably cloned full-length cDNA of dengue type 4 virus. Proc. Natl. Acad. Sci.

U.S.A. 88, 5139-5143. Leyssen, P., De Clercq, E., and Neyts, J. (2000). Perspectives for the treatment of infections

with Flaviviridae. Clin. Microbiol. Rev. 13, 67-82.

References

Leyssen, P., Van Lommel, A., Drosten, C., Schmitz, H., De Clercq, E., and Neyts, J. (2001). A novel model for the study of the therapy of flavivirus infections using the Modoc virus. Virology 279, 27-37.

Leyssen, P., Charlier, N., Lemey, P., Billoir, F., Van Damme, A.M., De Clercq, E., de Lamballerie, X., and Neyts, J. (2002). Complete genome sequence, taxonomic assignment, and comparatiev analysis of the untranslated regions of the Modoc virus, a flavivirus with no known vector. Virology 293, 125-140.

Lin, C.W., and Wu, S.C. (2003). A functional epitope determinant on domain III of the Japanese encephalitis virus envelope protein interacted with neutralizing-antibody combining sites. J. Virol. 77, 2600-2606.

Lin, Y.L., Liao, C.L., Chen, L.K., Yeh, C.T., Liu, C.I., Ma, S.H., Huang, Y.Y., Huang, Y.L., Kao, C.L., and King, C.C. (1998). Study of Dengue virus infection in SCID mice engrafted with human K562 cells. J. Virol. 72, 9729-9737.

Lindenbach, B.D., and Rice, C.M. (2001). Flaviviridae : The viruses and their replication. In: Fields Virology, 4th ed., Vol.1, Knipe, D.M.and Howley, P.M.(eds.) 1, 991-1041.

Liou, M.L., and Hsu, C.Y. (1998). Japanese encephalitis virus is transported across the cerebral blood vessels by endocytosis in mouse brain. Cell Tissue Res. 293, 389-394.

Liprandi, F. (1981). Isolation of plaque variants differing in virulence from the 17D strain of yellow fever virus. J. Gen. Virol. 56, 363-370.

Liu, Y., Blanden, R.V., and Mullbacher, A. (1989). Identification of cytolytic lymphocytes in West Nile virus-infected murine central nervous system. J. Gen. Virol. 70, 565-573.

Livak, K.J., Flood, S.J., Marmaro, J., Giusti, W., and Deetz, K. (1995). Oligonucleotides with fluorescent dyes at opposite ends provide a quenched probe system useful for detecting PCR product and nucleic acid hybridisation. PCR Methods Appl. 4, 357-362.

Lobigs, M., Usha, R., Nestorowicz, A., Marshall, I.D., Weir, R.C., and Dalgarno, L. (1990). Host cell selection of Murray Valley encephalitis virus variants altered at an RGD sequence in the envelope protein and in mouse virulence. Virology 176, 587-595.

Lorenz, I.C., Allison, S.L., Heinz, F.X., and Helenius, A. (2002). Folding and dimerization of tick-borne encephalitis virus envelope proteins prM and E in the endoplasmic reticulum. J. Virol. 76, 5480-5491.

Lustig, S., Danenberg, H.D., Kafri, Y., Kobiler, D., and Ben Nathan, D. (1992). Viral neuroinvasion and encephalitis induced by lipopolysaccharide and its mediators. J. Exp. Med. 176, 707-712.

Lvov, D.K., Butenko, A.M., Gromashevsky, V.L., Larichev, V.P., Gaidamovich, S.Y., Vyshemirsky, O.I., Zhukov, A.N., Lazorenko, V.V., Salko, V.N., Kovtunov, A.I., Galimzyanov, K.M., Platonov, A.E., Morozova, T.N., Khutoretskaya, N.V., Shishkina, E.O., and Skvortsova, T.M. (2000). Isolation of two strains of West Nile virus during an outbreak in southern Russia, 1999. Emerg. Infect. Dis. 6, 373-376.

Maddison, W.P., and Maddison, D.R. (1989). Interactive analysis of phylogeny and character evolution using the computer program MacClade. Folia Primatol. Basel. 53, 190-202.

Malinoski, F.J., Hasty, S.E., Ussery, M.A., and Dalrymple, J.M. (1990). Prophylactic ribavirin treatment of dengue type 1 infection in rhesus monkeys. Antiviral Res. 13, 139-149.

Mandl, C.W., Holzmann, H., Kunz, C., and Heinz, F.X. (1993). Complete genomic sequence of Powassan virus: evaluation of genetic elements in tick-borne versus mosquito-borne flaviviruses. Virology 194, 173-184.

Mandl, C.W., Ecker, M., Holzmann, H., Kunz, C., and Heinz, F.X. (1997). Infectious cDNA clones of tick-borne encephalitis virus European subtype prototypic strain Neudoerfl and high virulence strain Hypr. J. Gen. Virol. 78, 1049-1057.

References

Mandl, C.W., Holzmann, H., Meixner, T., Rauscher, S., Stadler, P.F., Allison, S.L., and Heinz, F.X. (1998). Spontaneous and engineered deletions in the 3' noncoding region of tick-borne encephalitis virus: construction of highly attenuated mutants of a flavivirus. J. Virol. 72, 2132-2140.

Mandl, C.W., Allison, S.L., Holzmann, H., Meixner, T., and Heinz, F.X. (2000). Attenuation of tick-borne encephalitis virus by structure-based site-specific mutagenesis of a putative flavivirus receptor binding site. J. Virol. 74, 9601-9609.

Mandl, C.W., Kroschewski, H., Allison, S.L., Kofler, R., Holzmann, H., Meixner, T., and Heinz, F.X. (2001). Adaptation of tick-borne encephalitis virus to BHK-21 cells results in the formation of multiple heparan sulfate binding sites in the envelope protein and attenuation in vivo. J. Virol. 75, 5627-5637.

Mangada, M.N., and Igarashi, A. (1997). Sequences of terminal non-coding regions from four dengue-2 viruses isolated from patients exhibiting different disease severities. Virus Genes 14, 5-12.

Martinez-Barragan, J.J., and del Angel, R.M. (2001). Identification of a putative coreceptor on Vero cells that participates in dengue 4 virus infection. J. Virol. 75, 7818-7827.

Mason, P.W. (1989). Maturation of Japanese encephalitis virus glycoproteins produced by infected mammalian and mosquito cells. Virology 169, 354-364.

Mathur, A., Arora, K.L., and Chaturvedi, U.C. (1981). Congenital infection of mice with Japanese encephalitis virus. Infect. Immun. 34, 26-29.

Matthews, V., Robertson, T., Kendrick, T., Abdo, M., Papadimitriou, J., and McMinn, P. (2000). Morphological features of Murray Valley encephalitis virus infection in the central nervous system of Swiss mice. Int. J. Exp. Pathol. 81, 31-40.

Mazlo, M., and Szanto, J. (1978). Morphological demonstration of the virus of tick-borne encephalitis in the human brain. Acta. Neuropathol. (Berl) 43, 251-253.

McCormack, J.G., and Allworth, A.M. (2002). Emerging viral infections in Australia. Med. J.

Aust. 177, 45-49. McMinn, P.C., Lee, E., Hartley, S., Roehrig, J.T., Dalgarno, L., and Weir, R.C. (1995a).

Murray valley encephalitis virus envelope protein antigenic variants with altered hemagglutination properties and reduced neuroinvasiveness in mice. Virology 211, 10-20.

McMinn, P.C., Marshall, I.D., and Dalgarno, L. (1995b). Neurovirulence and neuroinvasiveness of Murray Valley encephalitis virus mutants selected by passage in a monkey kidney cell line. J. Gen. Virol. 76, 865-872.

McMinn, P.C., Dalgarno, L., and Weir, R.C. (1996a). A comparison of the spread of Murray Valley encephalitis viruses of high or low neuroinvasiveness in the tissues of Swiss mice after peripheral inoculation. Virology 220, 414-423.

McMinn, P.C., Weir, R.C., and Dalgarno, L. (1996b). A mouse-attenuated envelope protein variant of Murray Valley encephalitis virus with altered fusion activity. J. Gen.

Virol. 77, 2085-2088. McMinn, P.C. (1997). The molecular basis of virulence of the encephalitogenic flaviviruses.

J. Gen. Virol. 78, 2711-2722. Men, R., Bray, M., Clark, D., Chanock, R.M., and Lai, C.J. (1996). Dengue type 4 virus

mutants containing deletions in the 3' noncoding region of the RNA genome: analysis of growth restriction in cell culture and altered viremia pattern and immunogenicity in rhesus monkeys. J. Virol. 70, 3930-3937.

Mendez, E., Ruggli, N., Collett, M.S., and Rice, C.M. (1998). Infectious bovine viral diarrhea virus (strain NADL) RNA from stable cDNA clones: a cellular insert determines NS3 production and viral cytopathogenicity. J.Virol. 72:4737-4745.

Misra, U.K., Kalita, J., and Srivastava, M. (1998). Prognosis of Japanese encephalitis: a multivariate analysis. J. Neurol. Sci. 161, 143-147.

Mohan, P.M., and Padmanabhan, R. (1991). Detection of stable secondary structure at the 3' terminus of dengue virus type 2 RNA. Gene 108, 185-191.

References

Molenkamp, R., Kooi, E.A., Lucassen, M.A., Greve, S., Thijssen, J.C., Spaan, W.J., and Bredenbeek, P.J. (2003). Yellow fever virus replicons as an expression system for hepatitis C virus structural proteins. J. Virol. 77, 1644-1648.

Monath, T.P., Cropp, C.B., and Harrison, A. K. (1983). Mode of entry of a neurotropic arbovirus into the central nervous system. Reinvestigation of an old controversy. Lab. Invest. 48, 399-410.

Monath, T.P. (1986). Pathobiology of the flaviviruses. Schlesinger S., Schlesinger M.J., eds. The Togaviridae and Flaviviridae. New York: Plenum Press, 375-440.

Monath, W.R., and Lipman, D.J. (1988). Improved tools for biological sequence comparison. Proc. Natl. Acad. Sci. U.S.A. 85, 2444-2448.

Monath, T.P. (1994). The risk to developed and developing countries. Proc. Natl. Acad. Sci.

U.S.A. 91, 2395-2400. Monath, T.P., and Heinz, F.X. (1996). Flaviviruses. In: Fields Virology, 3rd ed., Vol.1, Fields,

B.N., Knipe, D.M.and Howley, P.M.(eds.) 1, 961-1034. Monath, T.P. (1999a). Yellow fever. Plotkin, S., Orenstein, W.(Eds.), Vaccines, 3rd Edition,

WB Saunders Company, Philadelphia , 815-879. Monath, T.P., Soike, K., Levenbook, I., Zhang, Z.X., Arroyo, J., Delagrave, S., Myers, G.,

Barrett, A.T., Shope, R.E., Ratterree, M., Chambers, T.J., and Guirakhoo, F. (1999b). Recombinant, chimaeric live, attenuated vaccine (ChimeriVax (TM)) incorporating the envelope genes of Japanese encephalitis (SA14-14-2) virus and the capsid and non-structural genes of yellow fever (17D) virus is safe, immunogenic and protective in nonhuman primates. Vaccine 17, 1869-1882.

Monath, T.P., Levenbook, I., Soike, K., Zhang, Z.X., Ratterree, M., Draper, K., Barrett, A.D., Nichols, R., Weltzin, R., Arroyo, J., and Guirakhoo, F. (2000). Chimeric yellow fever virus 17D-Japanese encephalitis virus vaccine: dose-response effectiveness and extended safety testing in rhesus monkeys. J. Virol. 74, 1742-1751.

Monath, T.P. (2001a). Prospects for development of a vaccine against the West Nile virus. Ann. N.Y. Acad. Sci. 951, 1-12.

Monath, T.P., Arroyo, J., Miller, C., and Guirakhoo, F. (2001b). West Nile virus vaccine. Curr. Drug Targets Infect. Disord. 1, 37-50.

Monath, T.P., Arroyo, J., Levenbook, I., Zhang, Z.X., Catalan, J., Draper, K., and Guirakhoo, F. (2002). Single mutation in the flavivirus envelope protein hinge region increases neurovirulence for mice and monkeys but decreases viscerotropism for monkeys: relevance to development and safety testing of live, attenuated vaccines. J. Virol. 76, 1932-1943.

Mullinax, R.L., Gross, E.A., Hay, B.N., Amberg, J.R., Kubitz, M.M., and Sorge, J.A. (1992). Expression of a heterodimeric Fab antibody protein in one cloning step. Biotechniques 12, 864-869.

Mutebi, J.P., and Barrett, A.D. (2002). The epidemiology of yellow fever in Africa. Microbes.

Infect. 4, 1459-1468. Muylaert, I.R., Galler, R., and Rice, C.M. (1997). Genetic analysis of the yellow fever virus

NS1 protein: identification of a temperature-sensitive mutation which blocks RNA accumulation. J. Virol. 71, 291-298.

Nash, D., Mostashari, F., Fine, A., Miller, J., O'Leary, D., Murray, K., Huang, A., Rosenberg, A., Greenberg, A., Sherman, M., Wong, S., and Layton, M. (2001). The outbreak of West Nile virus infection in the New York City area in 1999. N.

Engl. J. Med. 344, 1807-1814. Nathanson, N. (1980). Pathogenesis. Monath TP, ed. St. Louis encephalitis. Washington,

DC: APHA, 201-236. Nedry, M., and Mahon, C.-R. (2003). West Nile virus: an emerging virus in North America.

Clin. Lab. Sci. 16, 43-49.

References

Ni, H., and Barrett, A.D. (1996). Molecular differences between wild-type Japanese encephalitis virus strains of high and low mouse neuroinvasiveness. J. Gen. Virol. 77, 1449-1455.

Nichter, C.A., Pavlakis, S.G., Shaikh, U., Cherian, K.A., Dobrosyzcki, J., Porricolo, M.E., and Chatturvedi, I. (2000). Rhombencephalitis caused by West Nile fever virus. Neurology 55, 153.

Nir, Y., Beemer, A., and Goldwasser, R.A. (1965). West Nile Virus infection in mice following exposure to a viral aerosol. Br. J. Exp. Pathol. 46, 443-449.

Ogata, A., Nagashima, K., Hall, W.W., Ichikawa, M., Kimura-Kuroda, J., and Yasui, K. (1991). Japanese encephalitis virus neurotropism is dependent on the degree of neuronal maturity. J. Virol. 65, 880-886.

Pethel, M., Falgout, B., and Lai, C.J. (1992). Mutational analysis of the octapeptide sequence motif at the NS1-NS2A cleavage junction of dengue type 4 virus. J. Virol. 66, 7225-7231.

Pletnev, A.G., Yamshchikov, V.F., and Blinov, V.M. (1990). Nucleotide sequence of the genome and complete amino acid sequence of the polyprotein of tick-borne encephalitis virus. Virology 174, 250-263.

Pletnev, A.G., Bray, M., Huggins, J., and Lai, C.J. (1992). Construction and characterization of chimeric tick-borne encephalitis/dengue type 4 viruses. Proc. Natl. Acad. Sci.

U.S.A. 89, 10532-10536. Pletnev, A.G., Bray, M., and Lai, C.J. (1993). Chimeric tick-borne encephalitis and dengue

type 4 viruses: effects of mutations on neurovirulence in mice. J. Virol. 67, 4956-4963.

Pletnev, A.G., and Men, R. (1998). Attenuation of the Langat tick-borne flavivirus by chimerization with mosquito-borne flavivirus dengue type 4. Proc. Natl. Acad. Sci.

U.S.A. 95, 1746-1751. Pletnev, A.G., Karganova, G.G., Dzhivanyan, T.I., Lashkevich, V.A., and Bray, M. (2000).

Chimeric Langat/Dengue viruses protect mice from heterologous challenge with the highly virulent strains of tick-borne encephalitis virus. Virology 274, 26-31.

Pletnev, A.G. (2001a). Infectious cDNA clone of attenuated Langat tick-borne flavivirus (strain E5) and a 3' deletion mutant constructed from it exhibit decreased neuroinvasiveness in immunodeficient mice. Virology 282, 288-300.

Pletnev, A.G., Bray, M., Hanley, K.A., Speicher, J., and Elkins, R. (2001b). Tick-borne Langat/mosquito-borne dengue flavivirus chimera, a candidate live attenuated vaccine for protection against disease caused by members of the tick-borne encephalitis virus complex: evaluation in rhesus monkeys and in mosquitoes. J. Virol. 75, 8259-8267.

Pletnev, A.G., Putnak, R., Speicher, J., Wagar, E.J., and Vaughn, D.W. (2002). West Nile virus/dengue type 4 virus chimeras that are reduced in neurovirulence and peripheral virulence without loss of immunogenicity or protective efficacy. Proc.

Natl. Acad. Sci. U.S.A. 99, 3036-3041. Poch, O., Sauvaget, I., Delarue, M., and Tordo, N. (1989). Identification of four conserved

motifs among the RNA-dependent polymerase encoding elements. EMBO J. 8, 3867-3874.

Polo, S., Ketner, G., Levis, R., and Falgout, B. (1997). Infectious RNA transcripts from full-length dengue virus type 2 cDNA clones made in yeast. J. Virol. 71, 5366-5374.

Porter, K.R., Kochel, T.J., Wu, S.J., Raviprakash, K., Phillips, I., and Hayes, C.G. (1998). Protective efficacy of a dengue 2 DNA vaccine in mice and the effect of CpG immuno-stimulatory motifs on antibody responses. Arch. Virol. 143, 997-1003.

Proutski, V., Gaunt, M.W., Gould, E.A., and Holmes, E.C. (1997a). Secondary structure of the 3'-untranslated region of yellow fever virus: implications for virulence, attenuation and vaccine development. J. Gen. Virol. 78, 1543-1549.

References

Proutski, V., Gould, E.A., and Holmes, E.C. (1997b). Secondary structure of the 3'-untranslated region of flaviviruses: similarities and differences. Nucleic. Acids. Res. 25, 1194-1202.

Proutski, V., Gristun, T.S., Gould, E.A., and Holmes, E.C. (1999). Biological consequences of deletions within the 3’ untranslated region of flaviviruses may be due to rearrangements of RNA secondary structure. Vir. Res. 64, 107-123.

Pugachev, K.V., Ocran, S.W., Guirakhoo, F., Furby, D., and Monath, T.P. (2002). Heterogeneous nature of the genome of the ARILVAX yellow fever 17D vaccine revealed by consensus sequencing. Vaccine 20, 996-999.

Pugachev, K.V., Guirakhoo, F., Trent, D.W., and Monath, T.P. (2003). Traditional and novel approaches to flavivirus vaccines. Int. J. Parasitol. 33, 567-582.

Pur, B., Polo, S., Hayes, C.G., and Falgout, B. (2000). Construction of a full length infectious clone for dengue-1 virus Western Pacific,74 strain. Virus Genes 20, 57-63.

Putnak, R., Barvir, D.A., Burrous, J.M., Dubois, D.R., D'Andrea, V.M., Hoke, C.H., Sadoff, J.C., and Eckels, K.H. (1996a). Development of a purified, inactivated, dengue-2 virus vaccine prototype in Vero cells: immunogenicity and protection in mice and rhesus monkeys. J. Infect. Dis. 174, 1176-1184.

Putnak, R., Cassidy, K., Conforti, N., Lee, R., Sollazzo, D., Truong, T., Ing, E., Dubois, D., Sparkuhl, J., Gastle, W., and Hoke, C. (1996b). Immunogenic and protective response in mice immunized with a purified, inactivated, Dengue-2 virus vaccine prototype made in fetal rhesus lung cells. Am. J. Trop. Med. Hyg. 55, 504-510.

Raut, C.G., Deolankar, R.P., Kolhapure, R.M., and Goverdhan, M.K. (1996). Susceptibility of laboratory-bred rodents to the experimental infection with dengue virus type 2. Acta Virol. 40, 143-146.

Receveur, M.C., Thiebaut, R., Vedy, S., Malvy, D., Mercie, P., and Bras, M.L. (2000). Yellow fever vaccination of human immunodeficiency virus-infected patients: report of 2 cases. Clin. Infect. Dis. 31, E7-E8.

Rey, F.A., Heinz, F.X., Mandl, C., Kunz, C., and Harrison, S.C. (1995). The envelope glycoprotein from tick-borne encephalitis virus at 2 A resolution. Nature 375, 291-298.

Rey, F.A. (2003). Dengue virus envelope glycoprotein structure: new insight into its interactions during viral entry. Proc. Natl. Acad. Sci. U.S.A. 100, 6899-6901.

Rice, C.M., Lenches, E.M., Eddy, S.R., Shin, S.J., Sheets, R.L., and Strauss, J.H. (1985). Nucleotide sequence of yellow fever virus: implications for flavivirus gene expression and evolution. Science 229, 726-733.

Rice, C.M., Grakoui, A., Galler, R., and Chambers, T.J. (1989). Transcription of infectious yellow fever RNA from full-length cDNA templates produced by in vitro ligation. New Biol. 1 , 285-296.

Ruggli, N., Tratschin, J.D., Mittelholzer, C., and Hofmann, M.A. (1996). Nucleotide sequence of classical swine fever virus strain Alfort/187 and transcription of infectious RNA from stably cloned full-length cDNA. J. Virol. 70, 3478-3487.

Ruggli, N., and Rice, C.M. (1999). Functional cDNA clones of the Flaviviridae: strategies and applications. Adv. Virus Res. 53, 183-207.

Ryman, K.D., Ledger, T.N., Weir, R.C., Schlesinger, J.J., and Barrett, A.D. (1997a). Yellow fever virus envelope protein has two discrete type-specific neutralizing epitopes. J. Gen. Virol. 78, 1353-1356.

Ryman, K.D., Xie, H., Ledger, T.N., Campbell, G.A., and Barrett, A.D. (1997b). Antigenic variants of yellow fever virus with an altered neurovirulence phenotype in mice. Virology 230, 376-380.

Ryman, K.D., Ledger, T.N., Campbell, G.A., Watowich, S.J., and Barrett, A.D. (1998). Mutation in a 17D-204 vaccine substrain-specific envelope protein epitope alters the pathogenesis of yellow fever virus in mice. Virology 244, 59-65.

References

Sabchareon, A., Lang, J., Chanthavanich, P., Yoksan, S., Forrat, R., Attanath, P., Sirivichayakul, C., Pengsaa, K., Pojjaroen-Anant, C., Chokejindachai, W., Jagsudee, A., Saluzzo, J.F., and Bhamarapravati, N. (2002). Safety and immunogenicity of tetravalent live-attenuated dengue vaccines in Thai adult volunteers: role of serotype concentration, ratio, and multiple doses. Am. J. Trop. Med. Hyg. 66, 264-272.

Sampson, B.A., Ambrosi, C., Charlot, A., Reiber, K., Verress, J.F., and Armbrustmacher, V. (2000). The pathology of human West Nile virus infection. Hum. Pathol. 31, 527-531.

Schlesinger, J.J., Brandriss, M.W., and Monath, T.P. (1983). Monoclonal antibodies distinguish between wild and vaccine strains of yellow fever virus by neutralization, hemagglutination inhibition, and immune precipitation of the virus envelope protein. Virology 125, 8-17.

Schlesinger, J.J., Chapman, S., Nestorowicz, A., Rice, C.M., Ginocchio, T.E., and Chambers, T.J. (1996). Replication of yellow fever virus in the mouse central nervous system: comparison of neuroadapted and non-neuroadapted virus and partial sequence analysis of the neuroadapted strain. J. Gen. Virol. 77, 1277-1285.

Seif, S.A., Morita, K., and Igarashi, A. (1996). A 27 amino acid coding region of JE virus E protein expressed in E. coli as fusion protein with glutathione-S-transferase elicit neutralizing antibody in mice. Virus Res. 43, 91-96.

Shi, P.Y., Brinton, M.A., Veal, J.M., Zhong, Y.Y., and Wilson, W.D. (1996). Evidence for the existence of a pseudoknot structure at the 3' terminus of the flavivirus genomic RNA. Biochemistry 35, 4222-4230.

Shi, P.Y., Tilgner, M., and Lo, M.K. (2002a). Construction and characterization of subgenomic replicons of New York strain of West Nile virus. Virology 296, 219-233.

Shi, P.Y., Tilgner, M., Lo, M.K., Kent, K.A., and Bernard, K.A. (2002b). Infectious cDNA clone of the epidemic west nile virus from New York City. J. Virol. 76, 5847-5856.

Shieh, W. J., Guarner, J., Layton, M., Fine, A., Miller, J., Nash, D., Campbell, G.I., Roehrig, J.T., Gubler, D.J., and Zaki, S.R. (2000). The role of pathology in an investigation of an outbreak of West Nile encephalitis in New York, 1999. Emerg. Infect. Dis. 6, 370-372.

Shlim, D.R., and Solomon, T. (2002). Japanese encephalitis vaccine for travelers: exploring the limits of risk. Clin. Infect. Dis. 35, 183-188.

Shurtleff, A.C., Beasley, D.W., Chen, J.J., Ni, H., Suderman, M.T., Wang, H., Xu, R., Wang, E., Weaver, S.C., Watts, D.M., Russell, K.L., and Barrett, A.D. (2001). Genetic variation in the 3' non-coding region of dengue viruses. Virology 281, 75-87.

Siegel-Itzkovich, J. (2000). Twelve die of West Nile virus in Israel. BMJ. 321, 724. Simmons, M., Nelson, W.M., Wu, S.J., and Hayes, C.G. (1998). Evaluation of the protective

efficacy of a recombinant dengue envelope B domain fusion protein against dengue 2 virus infection in mice. Am. J. Trop. Med. Hyg. 58, 655-662.

Simmons, M., Murphy, G.S., and Hayes, C.G. (2001). Short report: Antibody responses of mice immunized with a tetravalent dengue recombinant protein subunit vaccine. Am. J. Trop. Med. Hyg. 65, 159-161.

Singh, B., and Postic, B. (1970). Enhanced resistance of mice to virulent Japanese B encephalitis virus following inactivated vaccine and poly I:C. J. Infect. Dis. 122, 339-342.

Sriburi, R., Keelapang, P., Duangchinda, T., Pruksakorn, S., Maneekarn, N., Malasit, P., and Sittisombut, N. (2001). Construction of infectious dengue 2 virus cDNA clones using high copy number plasmid. J. Virol. Methods 92, 71-82.

Stadler, K., Allison, S.L., Schalich, J., and Heinz, F.X. (1997). Proteolytic activation of tick-borne encephalitis virus by furin. J. Virol. 71, 8475-8481.

References

Stiasny, K., Allison, S.L., Mandl, C.W., and Heinz, F.X. (2001). Role of metastability and acidic pH in membrane fusion by tick-borne encephalitis virus. J. Virol. 75, 7392-7398.

Stiasny, K., Allison, S.L., Schalich, J., and Heinz, F.X. (2002). Membrane interactions of the tick-borne encephalitis virus fusion protein E at low pH. J. Virol. 76, 3784-3790.

Strimmer, K., and Von Haeseler, A. (1996). Quartet puzzling: A quartet maximum likelihood method for reconstructing tree topologies. Mol. Biol. Evol. 13, 964-969.

Sumiyoshi, H., Hoke, C.H., and Trent, D.W. (1992). Infectious Japanese encephalitis virus RNA can be synthesized from in vitro-ligated cDNA templates. J. Virol. 66, 5425-5431.

Sumiyoshi, H., Tignor, G.H., and Shope, R.E. (1995). Characterization of a highly attenuated Japanese encephalitis virus generated from molecularly cloned cDNA. J. Infect. Dis. 171, 1144-1151.

Thiel, V., Herold, J., and Siddell, S.G. (2002). Long distance reverse-transcription PCR. Methods Mol. Biol. 192, 59-66.

Tsai, T.F., Popovici, F., Cernescu, C., Campbell, G.L., and Nedelcu, N.I. (1998). West Nile encephalitis epidemic in southeastern Romania. Lancet 352, 767-771.

van der Most, R.G., Murali-Krishna, K., Ahmed, R., and Strauss, J.H. (2000). Chimeric yellow fever/dengue virus as a candidate dengue vaccine: quantitation of the dengue virus-specific CD8 T-cell response. J. Virol. 74, 8094-8101.

van der Most, R.G., Harrington, L.E., Giuggio, V., Mahar, P.L., and Ahmed, R. (2002). Yellow fever virus 17D envelope and NS3 proteins are major targets of the antiviral T cell response in mice. Virology 296, 117-124.

van Dinten, L.C., den Boon, J.A., Wassenaar, A.L., Spaan W.J., and Snijder E.J. (1997). An infectious arterivirus cDNA clone: identification of a replicase point mutation that abolishes discontinuous mRNA transcription. Proc. Natl. Acad. Sci. U.S.A. 94, 991-996.

Velzing, J., Groen, J., Drouet, M.T., van Amerongen, G., Copra, C., Osterhaus, A.D., and Deubel, V. (1999). Induction of protective immunity against Dengue virus type 2: comparison of candidate live attenuated and recombinant vaccines. Vaccine 17, 1312-1320.

Von Heijne, G. (1984). How signal sequences maintain cleavage specificity. J. Mol. Biol. 173, 243-251.

Wallner, G., Mandl, C.W., Kunz, C., and Heinz, F.X. (1995). The flavivirus 3'-noncoding region: extensive size heterogeneity independent of evolutionary relationships among strains of tick-borne encephalitis virus. Virology 213, 169-178.

Wang, J.J., Liao, C.L., Yang, C.I., Lin, Y.L., Chiou, C.T., and Chen, L.K. (1998). Localizations of NS3 and E proteins in mouse brain infected with mutant strain of Japanese encephalitis virus. Arch. Virol. 143, 2353-2369.

Weiner, L.P., Cole, G.A., and Nathanson, N. (1970). Experimental encephalitis following peripheral inoculation of West Nile virus in mice of different ages. J. Hyg. (Lond) 68, 435-446.

Wengler, G., and Castle, E. (1986). Analysis of structural properties which possibly are characteristic for the 3'-terminal sequence of the genome RNA of flaviviruses. J. Gen. Virol. 67, 1183-1188.

Westaway, E.G. (1987). Flavivirus replication strategy. Adv. Virus Res. 33, 45-90. Wittwer, C.T., Ririe, K.M., Andrew, R.V., David, D.A., Gundry, R.A., and Balis, U.J.

(1997). The LightCycler: a microvolume multisample fluorimeter with rapid temperature control. Biotechniques 22, 176-181.

World Health Organization. Dengue bulletin. (1997). Bull. W.H.O. 21, 1-15. Worthington, M., Levy, H., and Rice, J. (1973). Late therapy of an arbovirus encephalitis in

mice with interferon and interferon stimulators. Proc. Soc. Exp. Biol. Med. 143, 638-643.

References

Wu, S.J., Hayes, C.G., Dubois, D.R., Windheuser, M.G., Kang, Y.H., Watts, D.M., and Sieckmann, D.G. (1995). Evaluation of the severe combined immunodeficient (SCID) mouse as an animal model for dengue viral infection. Am. J. Trop. Med.

Hyg. 52, 468-476. Xiao, S.Y., Guzman, H., Zhang, H., Travassos-da-Rosa, A.P., and Tesh, R.B. (2001). West

Nile virus infection in the golden hamster (Mesocricetus auratus): a model for West Nile encephalitis. Emerg. Infect. Dis. 7, 714-721.

Yamshchikov, V.F., and Compans, R.W. (1993). Regulation of the late events in flavivirus protein processing and maturation. Virology 192, 38-51.

Yamshchikov, V.F., Wengler, G., Perelygin, A.A., Brinton, M.A., and Compans, R.W. (2001). An infectious clone of the West Nile flavivirus. Virology 281, 294-304.

Yun, S.I., Kim, S.Y., Rice, C.M., and Lee, Y.M. (2003). Development and application of a reverse genetics system for Japanese encephalitis virus. J. Virol. 77, 6450-6465.

Zanotto, P.M., Gould, E.A., Gao, G.F., Harvey, P.H., and Holmes, E.C. (1996). Population dynamics of flaviviruses revealed by molecular phylogenies. Proc. Natl. Acad. Sci.

U.S.A. 93, 548-553. Zhang, F., Huang, Q., Ma, W., Jiang, S., Fan, Y., and Zhang, H. (2001). Amplification and

cloning of the full-length genome of Japanese encephalitis virus by a novel long RT-PCR protocol in a cosmid vector. J. Virol. Methods 96, 171-182.

Zhao, Z., Wakita, T., and Yasui, K. (2003). Inoculation of plasmids encoding Japanese encephalitis virus PrM-E proteins with colloidal gold elicits a protective immune response in BALB/c mice. J. Virol. 77, 4248-4260.

Documents

STUDY OF FLAVIVIRUS INFECTION - COnnecting REpositories fileSTUDY OF FLAVIVIRUS INFECTION ANIMAL MODELS, DETERMINANTS OF NEUROINVASION AND VECTOR SPECIFICITY Katholieke Universiteit