REVIEW ARTICLE
Progress for dengue virus diseases
Towards the NS2B–NS3pro inhibition for a therapeutic-based
approach
Sonia Melino and Maurizio Paci
Department of Chemical Science and Technology, University of Rome ‘Tor Vergata’, Italy
One hundred million cases of dengue fever (DF) are
estimated by the World Health Organization to
occur yearly, together with between 250 000 and
500 000 cases of dengue hemorrhagic fever (DHF).
Extensive plasma leakage in various serous cavities
of the body, including the pleura, the pericardium
and the peritoneal cavities, may result in profound
shock, the so-called dengue shock syndrome (DSS).
The case ⁄ fatality rate of DHF in most countries is
about 5%, although appropriate symptomatic treat-
ment has been successful in reducing the mortality
of DHF to less than 1%. Most fatalities occur
among children and young adults. DF and DHF are
primarily diseases of tropical and subtropical areas,
but represent a typical example of a global disease.
The transmission of dengue virus (DENv) has
Keywords
dengue hemorrhagic fever; dengue virus;
NS3; protease inhibitors; vaccines; viral
diseases; viral serine protease
Correspondence
S. Melino, Dipartimento di Scienze e
Tecnologie Chimiche, Universita
`
di Roma

of viral polyprotein processing.
Abbreviations
ADE, antibody-dependent enhancement; DENv, Dengue virus; DF, dengue fever; DHF, dengue hemorrhagic fever; DSS, dengue shock
syndrome; E protein, glycoprotein E; ER, endoplasmic reticulum; HCV, hepatitis C virus; NS, nonstructural; NS3pro, NS3 protease domain;
NTPase, nucleotide three phosphate hydrolase; protein C, nucleocapsid protein C; prM, protein M.
2986 FEBS Journal 274 (2007) 2986–3002 ª 2007 The Authors Journal compilation ª 2007 FEBS
increased considerably in recent years as a result of
the expansion of the Aedes aegypti mosquito to dif-
ferent geographic areas, and DHF has spread from
South East Asia to the Western Pacific and the
Americas. A substantial number of people travelling
to endemic regions are also infected each year. In the
last year, DHF has again been on the increase in
India and in several Asian countries because of sea-
sonal factors. Dengue is one of the most important
mosquito-borne viral diseases affecting humans; its
global distribution is comparable to that of malaria
and an estimated 2.5 billion people live in areas at
risk from epidemic transmission (Fig. 1). In 1906,
Bancroft published the first evidence implicating the
mosquito A. aegypti as the vector of DENv [1].
DENv was originally classified as an arthropod-borne
animal virus (arbovirus). The arboviruses comprise
infection agents that are biologically transmitted
between susceptible hosts by hematophagous arthro-
pods and are classified in different virus families
according to viral genes, virion structure and the viral
replication cycle. DENv belongs to the Flavivirus
genus of the family Flaviviridae that are members of
the positive-stranded virus supergroup 2 [2].

increase virus replication, and thus the probability of
developing DHF, by a process known as antibody-
dependent enhancement (ADE) [5–7]. However, there
are still cases of DHF and DSS that cannot be ade-
quately explained by ADE, an example being in the
confirmed cases of primary infection [8]. Accurate
knowledge of the viral life cycle is essential in order to
highlight potential targets for antiviral therapy and to
obtain key information for the rational design of anti-
viral drugs.
Virus structure and replicative cycle
The structure of the DENv is relatively simple. The
virions are spherical particles 40–50 nm in diameter,
containing three structural proteins: the nucleocapsid
protein C (C; 12–14 kDa); protein M (prM; an 8 kDa
nonglycosylated membrane protein); and the glyco-
protein E (E; 51–59 kDa), which is the major envel-
ope protein present as a homodimer. The DENv
genome is a single-stranded positive-sense RNA that
is encapsidated by protein C in an icosahedral struc-
ture. The genomic RNA presents a single long ORF
encoding the three structural proteins (C, prM and E)
and seven nonstructural (NS1–5) proteins (Fig. 2). It
is translated as a single polyprotein, which is cleaved
by proteases of viral and host origin. Outside the
ORF, there are the 5¢- and 3¢-UTRs, which have sec-
ondary structure and are crucial in the initiation and
regulation of translation, replication and virion
assembly [9–11].
The first step in the viral infection process is the

ciated with the NS3 protease to form an active serine
protease complex [26]. NS3 is implicated in the
C prM E NS1 NS2A NS2B NS3 NS4A NS5 NS4B
Structural proteins Non structural proteins
Furin cleavage
NS2B-NS3 protease cleavage
Signal peptidases cleavage
Unidentified
p
rotease in ER
5’ UTR 3’ UTR
ORF
CAP
Genome organization
Polyprotein and processing
Fig. 2. Organization of the dengue virus
(DENv) RNA genome and scheme of the
proteolytic processing of the DENv poly-
protein.
Progress for dengue virus diseases S. Melino and M. Paci
2988 FEBS Journal 274 (2007) 2986–3002 ª 2007 The Authors Journal compilation ª 2007 FEBS
polyprotein processing and RNA replication. The NS3
protein (69 kDa) is a multifunctional protein with an
N-terminal protease domain (NS3pro) (1–180), an
RNA triphosphatase, an RNA helicase and an RNA-
stimulated NTPase domain in the C-terminal region
[26,27]. The protease and NTPase enzymatic functions
share an overlapping region between residues 160 and
180 of the NS3 protein [28]. The RNA triphosphatase
may contribute to RNA capping [29], whereas the

potential risk of vaccination resulting in the ADE of
future heterotypic infection [36,39]. Different strategies
for the development of dengue vaccines include live
attenuated and inactivated viruses, recombinant sub-
units, protein expression in Escherichia coli, recombin-
ant baculoviruses, recombinant poxviruses, chimeric
viruses derived from infectious cDNA clones of DENv,
and naked DNA vaccines. In preclinical evaluation
using no-human primates, chimeric tetravalent vaccines
have been demonstrated to produce high levels of
neutralizing antibody and viremia protection against
all serotypes after a single dose, and clinical trials are
in progress [37,38,40]. Another type of dengue vaccine
is the DNA vaccine, which represents a promising
gene-based vaccine strategy considered suitable for
developing a dengue tetravalent vaccine [41,42]. Several
flavivirus DNA vaccines, including those against den-
gue, have already been developed [43–46]. Recently, a
new dengue tetravalent DNA vaccine against DENv-3
and DENv-4, based on a prM ⁄ E strategy and com-
bined with two previously constructed DNA vaccines
against DENv-1 and DENv-2, has been constructed
[47]. Molecular biology techniques have facilitated the
development of recombinant subunit vaccines. Several
structural (E and prM) and nonstructural proteins
stimulate immunity, and the nonstructural proteins
NS1 and NS3 are the dominant sources of cross-react-
ive CD4
+
and CD8

S. Melino and M. Paci Progress for dengue virus diseases
FEBS Journal 274 (2007) 2986–3002 ª 2007 The Authors Journal compilation ª 2007 FEBS 2989
simultaneous immunization with the DNA and protein
vaccine has been demonstrated [60–62].
Recently, a capsid protein of DEN-2 virus has also
been used in order to obtain statistically significant
protection against the infective homologous virus. This
suggests that effective protection against the four sero-
types might be attainable only by immunization with
the four corresponding capsids, or with one of them
including the immunodominant cytotoxic epitopes of
the others [63].
The major pharmaceutical companies are currently
developing a treatment against the disease. A tetra-
valent live attenuated vaccine was developed at the
Walter Reed Army Institute of Research, Silver
Spring, Maryland, licensed to GlaxoSmithKline [36];
this is the first two-dose vaccine to show a 100%
immune response against all four virus subtypes that
cause the disease. The vaccine is expected to enter
Phase III in 2007 and be commercially available there-
after, if its efficacy and safety is proven.
Therapeutic approaches – NS3 protease
inhibition as a response to DENv
Viral inhibitors have been widely studied in in vitro
systems as supportive medical care and for sympto-
matic treatment; they represent an important aid for
patients and for improving survival in severe forms of
disease. Antiviral therapeutic strategies involve virus-
binding blocking to prevent intracellular virus multipli-

The serine protease domain of NS3 protein plays a
central role in the replicative cycle of DENv [80]. Like
other viral proteases, the DENv NS3 protease repre-
sents an attractive therapeutic target for the develop-
ment of novel antiviral agents. Studies over the past
20 years have shown that many viruses encode one or
more proteases [81,82] that catalyze the processing of
viral polyprotein or maturational processing of precap-
sids and which are required for the production of
infectious virions. The discovery and development of
inhibitors of the viral protease activity assumed clinical
relevance, as has been demonstrated in cases involving
the treatment of patients with acquired immunodefi-
ciency syndrome (AIDS) or hepatitis C virus (HCV)
[83–87]. Studies on the viral protease significantly
increase our understanding of the life cycle of viruses,
the mechanism of proteolytic processing and the regu-
lation of cellular processes. A recurring theme from
structural and sequence analyses is the remarkable
compactness of these enzymes. In addition, most con-
tain no disulfide bridges, in contrast to many classical
cellular proteases, and, moreover, cofactors such as
metal ions or peptides are frequently required to stabil-
ize the viral protease [88–90]. Most viral proteases
have little sequence homology with cellular proteins,
even when they share the same backbone fold. These
characteristics lead to a very different substrate speci-
ficity of the viral proteases with very important impli-
cations for the design and development of their
efficient inhibitors, while undesirable cross reactivity

ses (serine, aspartic, metallo and cysteine) of proteases
[95]. The basic mechanism consists of a charge relay
system that transfers the negative charge on the buried
carboxyl via the histidine to the serine. The transfer of
the Ser Oc proton to the histidine converts the serine
into a strong nucleophile for the attack on the peptidyl
carbonyl of the substrate. The substrate is oriented by
the binding of the amino acid side chain of the P
1
resi-
due in the S
1
pocket [96], a hydrogen bond between
the backbone NH of the P
1
residue and two hydrogen
bonds between the carbonyl oxygen of the scissible
bond and two backbone NH groups of the enzyme
(oxyanion binding hole). The reaction is carried on
through a tetrahedral transition state with an acyl-
enzyme intermediate.
The DENv NS3 protease is also commonly desig-
nated as being a member of the flavivirin enzyme
family (EC 3.4.21.91 and S07.001 Peptidase MEROPS
peptidase database http://merops.sanger.ac.uk), which
comprise the NS2B–NS3 endoproteases of the Flavivi-
rus genus [97,98]. The presence of a small activating
cofactor protein is a prerequisite for the optimal cata-
lytic activity of the flaviviral proteases with natural
polyprotein substrates [99,100]. The DENv NS3 pro-

DEN-2 NS3pro the cofactor activity cannot be sup-
plied in trans with a small peptide derived from the
cofactor NS2B [105]. Other serine proteases (subtil-
isin, a-lytic protease) are also known to require a
pro-region, such as NS2B, for inducing a productive
folding leading to the active form. In these cases,
once the protein is folded, the necessary pro-region
does not remain bound to the active enzyme. The
results obtained regarding the NS2B–NS3pro complex
indicate that NS2B also functions as a molecular
chaperone in assisting the folding of NS3pro to the
active conformation [105,107]. A new construct of the
recombinant form of the NS3pro fused to a 40-resi-
due cofactor and corresponding to the hydrophilic
part of NS2B by a glycine linker was engineered and
expressed in E. coli, and demonstrated activity against
hexapeptide substrates modified as chromogenic para-
nitroanilide derivates [105]. Expression of the con-
struct CF40GlyNS3pro (the amino acid sequence is
shown in Fig. 3) resulted in substantially high yields
of the soluble and active recombinant protein, which
was significantly more active than the refolded
NS3pro and CF40NS3pro (lacking the Gly linker). In
fact, although the DENv NS3 protease exhibits
NS2B-independent activity with small substrates such
as N-a-benzoyl-l-arginine-p-nitroanilide, the activity
towards peptide substrates is stimulated significantly
in the presence of the NS2B protein [26,106].
Recently, it has been proposed that the Fx
3

be compartmentalized into specific membranous struc-
tures [111]. This finding suggests that the protease
activity may be affected by the membrane environ-
ment; in fact, the CF40GlyNS3pro activity in vitro was
increased by the presence of zwitterionic and nonionic
detergents at low concentrations [105].
Structural biological studies
The initial structural study was performed by Brink-
worth et al. [103], using a sequence homology
approach of NS3 protease with HCV NS3 protease,
which has been widely studied and whose structure
has been resolved by X-ray and NMR spectroscopy
[112–114]. By molecular modelling, a number of
insights concerning the cofactor interaction and sub-
strate specificity were obtained.
The model, by analogy with HCV NS4A, predicted
that the NS2B peptide encompassing residues Gly72–
Gly83 could be sufficient to function as a peptide
cofactor in vitro [103]. Moreover, the model suggested
a substrate specificity in the P1 position for the basic
Lys or Arg residues because of the presence of an aci-
dic Asp129 residue, present in the active cleft at six
residues before the catalytic Ser135 and conserved in
all the flavivirus sequences. Other interactions between
DEN2 NS3pro and the substrate have been predicted
by this model, such as a possible H-bond between the
Asp75 and the P2 residues and a hydrophobic interac-
tion between the P1¢ residue and the Val52 or Tyr41
residues. The resolution of the crystal structure of
NS3pro [115] at 2.1 A

and makes the determination of the solution structure
very unlikely [107].
Fig. 3. Sequence alignment between DENv-2
NS2B–NS3pro (2FOM pdb) and West Nile
virus NS2B–NS3pro (2FP7 pdb) obtained
using the
T-COFFEE program, version 1.41
(http://tcoffee.vital-it.ch/cgi-bin/Tcoffee/
tcoffee_cgi/index.cgi) [135]. The catalytic
residues are in bold and the numbers refer
to the DENv-2 NS2B–NS3pro sequence.
Progress for dengue virus diseases S. Melino and M. Paci
2992 FEBS Journal 274 (2007) 2986–3002 ª 2007 The Authors Journal compilation ª 2007 FEBS
Recently, the crystallographic structure (at 1.5 A
˚
resolution) of the active form of the NS2B–NS3pro
protein, including the 47-residue core region of NS2B
via a glycine linker (such as CF40GlyNS3pro), has
become available (2FOM pdb; Fig. 4B) [117]. Overall,
the structure is topologically close to that reported pre-
viously (six b-strands in two b-barrels with the cata-
lytic triad located at the cleft between the two barrels).
Nevertheless, it presents relevant differences in the sec-
ondary and tertiary structure that are important for
definition of the structural and functional roles of the
NS2B cofactor. However, the X-ray structure does not
appear to be structurally well defined in some regions.
This suggests that these regions may adopt multiple
conformations when passing from the solution to the
crystal state, as has also been observed in the NMR

N-CFNS3d shows a good cross-peak dispersion,
indicating a stable folded state of the protein [107]. All
data confirm that the NS2B fragment D50–E80 has a
strong interaction with NS3pro and is also able to pro-
mote in trans the activity of the enzyme when correctly
folded. This finding indicates that this cofactor region
has an important role in the conformational stability
of the active site. In the crystal structure, the electron
density beyond the NS2B residue 76 is discontinuous,
revealing that this region may adopt several conforma-
tions probably as a consequence of its great flexibility
in solution. No evidence of direct interactions of NS2B
with the active site are found in this structure, giving
no structural explanation of its absolute requirement
by NS3pro for activity.
On the contrary, the crystallographic structure of
the homologous NS2B–NS3pro protein of West Nile
virus has shown direct interactions of the C-terminal
part of NS2B with the active site of the NS3pro. The
C-terminal part of NS2B wraps around NS3pro and,
in particular, the Arg78–Leu87 residues form a
b-strand in NS2B, which links the N-terminal tract of
NS3pro. The structure of the West Nile virus NS2B–
NS3 pro-inhibitor complex has also elucidated (2FP7
pdb) [117], the details of the S1 pocket, formed by
Gly151, Tyr161, Tyr150, Asp129 and the backbone
C-Term
Asp 75
A
N-Term

and may be useful in the development of drugs to treat
the flaviviral diseases.
Substrate specificity of NS2B–NS3pro
The first step towards designing an inhibitor of the
viral protease is to identify substrate specificity. The
selectivity of the proteases for particular substrates
results from the presence of specific binding sites
on the enzyme for amino acid side chains of the
substrate(s). The virus-encoded proteases display an
unusual degree of selectivity for their natural
polyprotein substrates and only very few cases are
known where the viral enzyme reacts with protein
substrates derived from the host cell [118,119]. In the
case of viral proteases, the identification of a high
turnover substrate is usually difficult [120] because
the kinetic parameters of synthetic peptides based on
the natural cleavage sites are generally unfavorable
[121].
The NS3 protease in the absence of the cofactor
reacts with small model substrates for serine proteases,
such as N-a-benzoyl-l-arginine-p-nitroanilide, and acti-
vity of the NS3 protease towards the substrate is
higher than that of the NS2–NS3 complex [106]. This
suggests that substrate recognition in the complex
requires additional interactions, extending beyond the
P1 site, for optimal activity. Other studies have indica-
ted that NS2B–NS3pro requires the presence of
Lys ⁄ Arg and Arg, respectively, at the P2 and P1 posi-
tions, for achieving substrate proteolysis, and that the
cleavage motifs have features in common with the phy-

the P3 and P4 positions also contribute significantly to
ground state binding, providing additional evidence for
enzyme–substrate interactions that extend beyond S2
to S2¢ [122]. The introduction of an arginine residue at
P3 results in an almost four-fold increase in k
cat
⁄ K
m
,
and the introduction of an arginine residue at P3 and
P4 in the capsid protein-derived tetrabasic sequence
RRRR results in a 30-fold increase in k
cat
⁄ K
m
.
A higher degree of selectivity for serine at the P3¢ posi-
tion is needed, whereas selection of residues at the P2¢,
and especially at the P4¢ positions seems to be relat-
ively unrestrained [123]. Recently, the specifics of sub-
strate recognition by NS3pro from DENv have been
mapped using a library of the 9-mer peptides to the
cleavable sequences with the general P4–P3–P2–P1–
P1¢–P2¢–P3¢–P4¢–Gly structure [124]. The N terminus
and the constant C-terminal Gly of the peptides were
tagged with a fluorescent tag and with a biotin tag,
respectively. The amino acid sequences of the peptides
corresponding to the junction regions efficiently clea-
ved by the DENv protease are shown in Table 2. In
addition, other potential sites of the NS2B–NS3pro

an Arg residue at the P2 position [125]. Thus, the
finding that DENv proteases exhibit a preference for
Arg at the P2 position could be explained by the
presence of Ser or Thr at NS2B-84 [125]. On the
basis of these recent studies, the DENv enzyme seems
to adopt a restricted specificity to process the natural
cleavage sites of the polyprotein precursor, but this
specificity is less stringent than the homologous viral
proteases.
Inhibition of NS3 protease, a therapeutic target
In a first step towards design of an inhibitor for the
DENv NS3 serine protease, the standard inhibitors of
serine proteases have been assayed. The serine protease
inhibitor, aprotinin, has been shown to inhibit the four
CF40GlyNS3pro proteases with high affinity (K
i
¼ 79,
25, 88, 6.4 pm for DEN 1–4 CF40GlyNS3pro pro-
teases, respectively), whereas other serine protease
inhibitors show a low ability in inhibiting the viral
protease [105,122]. Similarly to the HCV NS3 protease,
the existence of a high-affinity binding site in the non-
prime region of the enzyme offers the possibility of
developing effective inhibitors against the DENv pro-
tease by combinatorial optimization of the cleavage
sites. For this reason, small-molecule inhibitors based
upon the peptide substrates have been synthesized as
inhibitors of NS3pro (Table 3). N-terminal cleavage
site peptides, corresponding to the P6–P1 region of the
Table 3. Representative competitive inhibitors of the dengue NS2B–NS3pro serine protease.

Boronic acid Bz-Nle-Lys-Arg-Arg-B(OH)
2
0.043 [130]
Cyclohexenyl chalcone
derivative
Panduratin A 25 [133]
4-Hydroxypanduratin A 21 [133]
Table 2. The amino acid sequences of the cleavage sites of the NS2B–NS3pro protease in the precursor polyprotein.
fl
Denotes the scissile
bond, and the P1 residues are shown in bold.
S. Melino and M. Paci Progress for dengue virus diseases
FEBS Journal 274 (2007) 2986–3002 ª 2007 The Authors Journal compilation ª 2007 FEBS 2995
polyprotein, were found to act as competitive inhibi-
tors of the enzyme, with K
i
values ranging from 67 to
12 lm. The NS2A ⁄ NS2B cleavage site, RTSKKR, is
the peptide with the lowest K
i
value. However, in con-
trast to HCV NS3 protease, the cleavage products and
their analogs do not appreciably inhibit this protease.
In fact, the peptides corresponding to the P1¢–P5¢
region of the polyprotein cleavage sites do not show
any inhibitory effect on enzymatic activity, even at
1mm concentration [126].
Peptidic a-keto amide inhibitors have been well char-
acterized as reversible competitive inhibitors for other
serine proteases, including HCV NS3 protease [127–

are possible and stabilize inhibitor binding [131].
Potent inhibitors have been identified by incorporating
trifluoromethyl ketone and boronic acid onto the sub-
strate peptide [130]. Tetrapeptide boronic acid proved
to be the most potent inhibitor of DENv NS3 prote-
ase, having a K
i
of 43 nm, whereas the trifluoromethyl
ketone occupies an intermediate position between that
of peptide aldehyde and peptide boronic acid [131].
Recent studies have shown that some natural com-
pounds, such as chalcones isolated from Boesenbergia
rotunda (a common spice belonging to a member of
the ginger family), are able to inhibit the DENv NS3
protease. In particular, the cyclohexenyl chalcone deri-
vates, 4-hydroxypanduratin A and panduratin A, show
good competitive inhibitory activities against DEN-2
virus NS3 protease, having apparent K
i
values of 21
and 25 lm, respectively [133]. Although several inhibi-
tors of DENv proteases have been tested, selective
viral protease inhibition has not been obtained to date
and inhibitors for clinical trials are not yet available.
Conclusions
Albeit there are still no specific vaccines or chemother-
apy regimes for the prevention and treatment of DF
and DHF, the understanding and the biochemical
characterization of the life cycle of DENv have made
substantial progress over the past few years, and all

At present, no inhibitors of the cofactor–protease bind-
ing are available, although the resolution of the crystal-
lographic structure and the production of mutants can
help to develop specific inhibitors of the binding to the
cofactor. Some regions of the NS2B–NS3pro structure
are not yet well defined, and their resolution will be
important for the complete understanding of the struc-
ture–function correlations, such as the resolution of the
structure of its complex with an inhibitor.
On the other hand, the NTPase ⁄ helicase region of
the NS3 protein, and the surface of NS3 protein with
Progress for dengue virus diseases S. Melino and M. Paci
2996 FEBS Journal 274 (2007) 2986–3002 ª 2007 The Authors Journal compilation ª 2007 FEBS
the NS5 replicase, could represent alternative drug tar-
gets. Thus, the use of a pharmacological therapy using
combinations of different inhibitors, similarly to other
viral therapies, could minimize the development of
rapid resistance.
In conclusion, considerable effort has recently been
made towards inhibiting the viral replication of DENv.
Much remains to be performed to achieve results suit-
able for experimentation in clinical trials and to pro-
duce a drug for blocking the DENv spread. To date,
funding for a co-ordinated strategy against dengue has
been disappointing. This is probably attributable to
the spread of dengue diseases in the world’s poorest
countries. Our hope therefore is that the major inter-
national funding agencies will now seriously consider
increasing their commitment to combat these diseases.
The need is all the more urgent now that climate

dengue hemorrhagic fever. Am J Trop Med Hyg 40,
444–451.
7 Morens DM (1994) Antibody-dependent enhancement
of infection and the pathogenesis of viral disease. Clin
Infect Dis 19, 500–512.
8 Watts DM, Porter KR, Putvatana P, Vasquez B,
Calampa C, Hayes CG & Halstead SB (1999) Failure
of secondary infection with American genotype dengue
2 to cause dengue haemorrhagic fever. Lancet 354,
1431–1434.
9 Cahour A, Pletnev A, Vazielle-Falcoz M, Rosen L &
Lai CJ (1995) Growth-restricted dengue virus mutants
containing deletions in the 5¢ noncoding region of the
RNA genome. Virology 207, 68–76.
10 Zeng L, Falgout B & Markoff L (1998) Identification
of specific nucleotide sequences within the conserved
3¢-SL in the dengue type 2 virus genome required for
replication. J Virol 72, 7510–7522.
11 Chiu WW, Kinney RM & Dreher TW (2005) Control
of translation by the 5¢- and 3 ¢-terminal regions of the
dengue virus genome. J Virol 79, 8303–8315.
12 Littaua R, Kurane I & Ennis FA (1990) Human IgG Fc
receptor II mediates antibody-dependent enhancement
of dengue virus infection. J Immunol 144, 3183–3186.
13 Heinz FX, Auer G, Stiasny K, Holzmann H, Mandl C,
Guirakhoo F & Kunz C (1994) The interactions of the
flavivirus envelope proteins: implications for virus entry
and release. Arch Virol Suppl. 9, 339–348.
14 Heinz FX & Allison SL (2001) The machinery for fla-
vivirus fusion with host cell membranes. Curr Opin

particulate form. J Virol 69, 5816–5820.
21 Rey FA, Heinz FX, Mandl C, Kunz C & Harrison SC
(1995) The envelope glycoprotein from tick-borne
encephalitis virus at 2 A
˚
resolution. Nature 375, 291–
298.
22 Falgout B & Markoff L (1995) Evidence that flavivirus
NS1-NS2A cleavage is mediated by a membrane-bound
host protease in the endoplasmic reticulum. J Virol 69,
7232–7243.
23 Chambers TJ, Hahn CS, Galler R & Rice CM (1990)
Flavivirus genome organization, expression, and repli-
cation. Annu Rev Microbiol 44, 649–688.
24 Mackenzie JM, Jones MK & Young PR (1996) Immu-
nolocalization of the dengue virus nonstructural glyco-
protein NS1 suggests a role in viral RNA replication.
Virology 220, 232–240.
25 Jacobs MG, Robinson PJ, Bletchly C, Mackenzie JM
& Young PR (2000) Dengue virus nonstructural pro-
tein 1 is expressed in a glycosyl- phosphatidylinositol-
linked form that is capable of signal transduction.
Faseb J 14, 1603–1610.
26 Falgout B, Pethel M, Zhang YM & Lai CJ (1991) Both
nonstructural proteins NS2B and NS3 are required for
the proteolytic processing of dengue virus nonstruc-
tural proteins. J Virol 65, 2467–2475.
27 Gorbalenya AE, Donchenko AP, Koonin EV & Blinov
VM (1989) N-terminal domains of putative helicases of
flavi- and pestiviruses may be serine proteases. Nucleic

tein expressed in Escherichia coli exhibits RNA-depen-
dent RNA polymerase activity. Virology 216, 317–325.
36 Halstead SB & Deen J (2002) The future of dengue
vaccines. Lancet 360, 1243–1245.
37 Hombach J, Barrett AD, Cardosa MJ, Deubel V,
Guzman M, Kurane I, Roehrig JT, Sabchareon A &
Kieny MP (2005) Review on flavivirus vaccine develop-
ment. Proceedings of a meeting jointly organised by
the World Health Organization and the Thai Ministry
of Public Health, 26–27 April 2004, Bangkok, Thai-
land. Vaccine 23, 2689–2695.
38 Stephenson JR (2005) Understanding dengue patho-
genesis: implications for vaccine design. Bull World
Health Organ 83, 308–314.
39 Vaughn DW, Green S, Kalayanarooj S, Innis BL,
Nimmannitya S, Suntayakorn S, Endy TP,
Raengsakulrach B, Rothman AL, Ennis FA et al.
(2000) Dengue viremia titer, antibody response pattern,
and virus serotype correlate with disease severity.
J Infect Dis 181, 2–9.
40 Guirakhoo F, Pugachev K, Zhang Z, Myers G,
Levenbook I, Draper K, Lang J, Ocran S, Mitchell F,
Parsons M et al. (2004) Safety and efficacy of chimeric
yellow fever-dengue virus tetravalent vaccine formula-
tions in nonhuman primates. J Virol 78, 4761–4775.
41 Donnelly JJ, Ulmer JB, Shiver JW & Liu MA (1997)
DNA vaccines. Annu Rev Immunol 15, 617–648.
42 Schultz J, Dollenmaier G & Molling K (2000) Update
on antiviral DNA vaccine research (1998–2000). Inter-
virology 43, 197–217.

50 Kurane I, Brinton MA, Samson AL & Ennis FA
(1991) Dengue virus-specific, human CD4
+
CD8
–
cyto-
toxic T-cell clones: multiple patterns of virus cross-
reactivity recognized by NS3-specific T-cell clones.
J Virol 65, 1823–1828.
51 Gagnon SJ, Ennis FA & Rothman AL (1999) Bystan-
der target cell lysis and cytokine production by dengue
virus-specific human CD4(+) cytotoxic T-lymphocyte
clones. J Virol 73, 3623–3629.
52 Kaufman BM, Summers PL, Dubois DR & Eckels KH
(1987) Monoclonal antibodies against dengue 2 virus
E-glycoprotein protect mice against lethal dengue infec-
tion. Am J Trop Med Hyg 36, 427–434.
53 Kaufman BM, Summers PL, Dubois DR, Cohen WH,
Gentry MK, Timchak RL, Burke DS & Eckels KH
(1989) Monoclonal antibodies for dengue virus prM
glycoprotein protect mice against lethal dengue infec-
tion. Am J Trop Med Hyg 41, 576–580.
54 Brandriss MW, Schlesinger JJ, Walsh EE & Briselli M
(1986) Lethal 17D yellow fever encephalitis in mice.
I. Passive protection by monoclonal antibodies to the
envelope proteins of 17D yellow fever and dengue 2
viruses. J Gen Virol 67, 229–234.
55 Tan CH, Yap EH, Singh M, Deubel V & Chan YC
(1990) Passive protection studies in mice with
monoclonal antibodies directed against the non-struc-

62 Konishi E, Terazawa A & Imoto J (2003) Simulta-
neous immunization with DNA and protein vaccines
against Japanese encephalitis or dengue synergistically
increases their own abilities to induce neutralizing anti-
body in mice. Vaccine 21, 1826–1832.
63 Lazo L, Hermida L, Zulueta A, Sanchez J, Lopez C,
Silva R, Guillen G & Guzman MG (2007) A recombi-
nant capsid protein from Dengue-2 induces protection
in mice against homologous virus. Vaccine 25, 1064–
1070.
64 Chen Y, Maguire T, Hileman RE, Fromm JR, Esko
JD, Linhardt RJ & Marks RM (1997) Dengue virus
infectivity depends on envelope protein binding to tar-
get cell heparan sulfate. Nat Med 3, 866–871.
65 Germi R, Crance JM, Garin D, Guimet J, Lortat-Ja-
cob H, Ruigrok RW, Zarski JP & Drouet E (2002)
Heparan sulfate-mediated binding of infectious dengue
virus type 2 and yellow fever virus. Virology 292, 162–
168.
66 Damonte EB, Matulewicz MC & Cerezo AS (2004)
Sulfated seaweed polysaccharides as antiviral agents.
Curr Med Chem 11, 2399–2419.
67 Talarico LB, Zibetti RG, Faria PC, Scolaro LA, Duar-
te ME, Noseda MD, Pujol CA & Damonte EB (2004)
Anti-herpes simplex virus activity of sulfated galactans
from the red seaweeds Gymnogongrus griffithsiae
and Cryptonemia crenulata. Int J Biol Macromol 34,
63–71.
68 Pujol CA, Estevez JM, Carlucci MJ, Ciancia M, Ce-
rezo AS & Damonte EB (2002) Novel dl-galactan

J Virol 74, 4957–4966.
75 Diamond MS & Harris E (2001) Interferon inhibits
dengue virus infection by preventing translation of
viral RNA through a PKR-independent mechanism.
Virology 289, 297–311.
76 Bartholomeusz A, Tomlinson E, Wright PJ, Birch C,
Locarnini S, Weigold H, Marcuccio S & Holan G
(1994) Use of a flavivirus RNA-dependent RNA poly-
merase assay to investigate the antiviral activity of
selected compounds. Antiviral Res 24, 341–350.
77 Neves-Souza PC, Azeredo EL, Zagne SM, Valls-de-
Souza R, Reis SR, Cerqueira DI, Nogueira RM &
Kubelka CF (2005) Inducible nitric oxide synthase
(iNOS) expression in monocytes during acute Dengue
Fever in patients and during in vitro infection. BMC
Infect Dis 5, 64.
78 Charnsilpa W, Takhampunya R, Endy TP, Mammen
MP Jr, Libraty DH & Ubol S (2005) Nitric oxide radi-
cal suppresses replication of wild-type dengue 2 viruses
in vitro. J Med Virol 77, 89–95.
79 Takhampunya R, Padmanabhan R & Ubol S (2006)
Antiviral action of nitric oxide on dengue virus type 2
replication. J Gen Virol 87, 3003–3011.
80 Valle RP & Falgout B (1998) Mutagenesis of the NS3
protease of dengue virus type 2. J Virol 72, 624–632.
81 Krausslich HG & Wimmer E (1988) Viral proteinases.
Annu Rev Biochem 57, 701–754.
82 Babe LM & Craik CS (1997) Viral proteases: evolution
of diverse structural motifs to optimize function. Cell
91, 427–430.

89 Love RA, Parge HE, Wickersham JA, Hostomsky Z,
Habuka N, Moomaw EW, Adachi T & Hostomska Z
(1996) The crystal structure of hepatitis C virus NS3
proteinase reveals a trypsin-like fold and a structural
zinc binding site. Cell 87, 331–342.
90 Shimizu Y, Yamaji K, Masuho Y, Yokota T, Inoue H,
Sudo K, Satoh S & Shimotohno K (1996) Identifica-
tion of the sequence on NS4A required for enhanced
cleavage of the NS5A ⁄ 5B site by hepatitis C virus NS3
protease. J Virol 70, 127–132.
91 Cahour A, Falgout B & Lai CJ (1992) Cleavage of the
dengue virus polyprotein at the NS3 ⁄ NS4A and
NS4B ⁄ NS5 junctions is mediated by viral protease
NS2B-NS3, whereas NS4A ⁄ NS4B may be processed by
a cellular protease. J Virol 66, 1535–1542.
92 Mukhopadhyay S, Kuhn RJ & Rossmann MG (2005)
A structural perspective of the flavivirus life cycle. Nat
Rev Microbiol 3, 13–22.
93 Beasley DW (2005) Recent advances in the mole-
cular biology of west nile virus. Curr Mol Med 5,
835–850.
94 Chambers TJ, Nestorowicz A, Amberg SM & Rice
CM (1993) Mutagenesis of the yellow fever virus NS2B
protein: effects on proteolytic processing, NS2B-NS3
complex formation, and viral replication. J Virol 67,
6797–6807.
95 Paetzel M & Dalbey RE (1997) Catalytic hydroxyl ⁄
amine dyads within serine proteases. Trends Biochem
Sci 22, 28–31.
96 Schechter I & Berger A (1967) On the size of the active

103 Brinkworth RI, Fairlie DP, Leung D & Young PR
(1999) Homology model of the dengue 2 virus NS3
protease: putative interactions with both substrate and
NS2B cofactor. J Gen Virol 80, 1167–1177.
104 Falgout B, Miller RH & Lai CJ (1993) Deletion analy-
sis of dengue virus type 4 nonstructural protein NS2B:
identification of a domain required for NS2B-NS3 pro-
tease activity. J Virol 67, 2034–2042.
105 Leung D, Schroder K, White H, Fang NX, Stoermer
MJ, Abbenante G, Martin JL, Young PR & Fairlie
DP (2001) Activity of recombinant dengue 2 virus NS3
protease in the presence of a truncated NS2B co-factor,
small peptide substrates, and inhibitors. J Biol Chem
276, 45762–45771.
106 Yusof R, Clum S, Wetzel M, Murthy HM & Padma-
nabhan R (2000) Purified NS2B ⁄ NS3 serine protease
of dengue virus type 2 exhibits cofactor NS2B depend-
ence for cleavage of substrates with dibasic amino acids
in vitro. J Biol Chem 275, 9963–9969.
107 Melino S, Fucito S, Campagna A, Wrubl F, Gamarnik
A, Cicero DO & Paci M (2006) The active essential
CFNS3d protein complex. FEBS J 273, 3650–3662.
108 Butkiewicz NJ, Yao N, Wright-Minogue J, Zhang R,
Ramanathan L, Lau JY, Hong Z & Dasmahapatra B
(2000) Hepatitis C NS3 protease: restoration of NS4A
cofactor activity by N-biotinylation of mutated NS4A
using synthetic peptides. Biochem Biophys Res Commun
267, 278–282.
109 Niyomrattanakit P, Winoyanuwattikun P, Chanprap-
aph S, Angsuthanasombat C, Panyim S & Katzenmeier

Dengue virus NS3 serine protease. Crystal structure
and insights into interaction of the active site with
substrates by molecular modeling and structural
analysis of mutational effects. J Biol Chem 274,
5573–5580.
116 Murthy HM, Judge K, DeLucas L & Padmanabhan R
(2000) Crystal structure of Dengue virus NS3 protease
in complex with a Bowman-Birk inhibitor: implications
for flaviviral polyprotein processing and drug design.
J Mol Biol 301, 759–767.
117 Erbel P, Schiering N, D’Arcy A, Renatus M, Kroemer
M, Lim SP, Yin Z, Keller TH, Vasudevan SG &
Hommel U (2006) Structural basis for the activation of
flaviviral NS3 proteases from dengue and West Nile
virus. Nat Struct Mol Biol 13, 372–373.
118 Kuyumcu-Martinez NM, Joachims M & Lloyd RE
(2002) Efficient cleavage of ribosome-associated poly
(A)-binding protein by enterovirus 3C protease. J Virol
76, 2062–2074.
119 Glaser W, Cencic R & Skern T (2001) Foot-and-mouth
disease virus leader proteinase: involvement of C-term-
inal residues in self-processing and cleavage of eIF4GI.
J Biol Chem 276, 35473–35481.
120 Kassel DB, Green MD, Wehbie RS, Swanstrom R &
Berman J (1995) HIV-1 protease specificity derived
from a complex mixture of synthetic substrates. Anal
Biochem 228, 259–266.
121 Pessi A (2001) A personal account of the role of pep-
tide research in drug discovery: the case of hepatitis C.
J Pept Sci 7, 2–14.

tors of elastase. Med Res Rev 14, 127–194.
128 Babine RE & Bender SL (1997) Molecular recognition
of proteinminus signligand complexes: applications to
drug design. Chem Rev 97, 1359–1472.
129 Han W, Hu Z, Jiang X & Decicco CP (2000) Alpha-
ketoamides, alpha-ketoesters and alpha-diketones as
HCV NS3 protease inhibitors. Bioorg Med Chem Lett
10, 711–713.
130 Yin Z, Patel SJ, Wang WL, Wang G, Chan WL, Rao
KR, Alam J, Jeyaraj DA, Ngew X, Patel V et al. (2006)
Peptide inhibitors of Dengue virus NS3 protease. Part
1: Warhead. Bioorg Med Chem Lett 16, 36–39.
131 Yin Z, Patel SJ, Wang WL, Chan WL, Ranga Rao
KR, Wang G, Ngew X, Patel V, Beer D, Knox Je
et al. (2006) Peptide inhibitors of dengue virus NS3
protease. Part 2: SAR study of tetrapeptide aldehyde
inhibitors. Bioorg Med Chem Lett 16, 40–43.
132 Ganesh VK, Muller N, Judge K, Luan CH, Padmana-
bhan R & Murthy KH (2005) Identification & charac-
terization of nonsubstrate based inhibitors of the
essential dengue and West Nile virus proteases. Bioorg
Med Chem 13, 257–264.
133 Kiat TS, Pippen R, Yusof R, Ibrahim H, Khalid N
& Rahman NA (2006) Inhibitory activity of cyclo-
hexenyl chalcone derivatives and flavonoids of finger-
root, Boesenbergia rotunda (L.), towards dengue-2
virus NS3 protease. Bioorg Med Chem Lett 16,
3337–3340.
134 Angelo PF, Lima AR, Alves FM, Blaber SI, Scaris-
brick IA, Blaber M, Juliano L & Juliano MA (2006)