J. Biomedical Science and Engineering, 2009, 2, 123-127
Published Online April 2009 in SciRes. http://www.scirp.org/journal/jbise JBiSE
1
Bioinformatics analysis and characteristics of
envelop glycoprotein E epitopes of dengue virus
Hua Zhong1, Wei Zhao1, Liang Peng1, Shan-Feng Li1, Hong Cao1
1Department of Microbiology, School of Public Health and Tropical Medicine, Southern Medical University, Guangzhou, Guangdong 510515, China. Corre-
spondence should be addressed to Hong Cao ([e-mail]), Tel.:[phone].
Received Sep. 4th, 2008; revised Dec. 22nd, 2008; accepted Jan. 8th, 2009
ABSTRACT
The major envelope glycoprotein E of dengue
(DEN) virus plays a central role in the biology of
flaviviruses. It is capable of inducing a protective
immune response in vivo and responsible for the
viral binding to the cellular receptor. The crystal
structures of glycoprotein E ectodomains have
already been determined. However, it is still un-
clear where the well-defined B-cell epitopes for
glycoprotein E which induce the neutralizing an-
tibodies locates. Thus, in order to characterize the
role of glycoprotein E in the pathogenesis of
dengue virus infection, we first used network
servers (http://bio.dfci. harvard.edu/Tools/ &
http://www. imtech. res. in) to predict and analyze
the well defined B-cell and T-cell epitopes of the
glycoprotein of the DEN-1 HAWAII strain. Then
based on the highly conserved envelop glyco-
protein amino acids, the hydrophilicity, antigenic-
ity, accessibility and flexibility of envelop glyco-
protein E were further predicted by using Biotic
softwares (DNASTAR) and network servers
(http://bio. dfci.harvard.edu/Tools/), the secondary
structure was putatively obtained. In our study,
the sequence at 281-295 amino acid (aa) for den-
gue virus type 1 HAWAII strain and the sequence
at 345-359, 383-397 for dengue virus type 2 NGC
strain were predicted as the more prevalent epi-
topes by using multiple parameters and different
analysis softwares, respectively. Two epitopes of
DEN-2 and one of DEN-1 locate on the domain Ш
and domainⅡ of the protein E, respectively. Sub-
sequently, further studies will be carried out to
examine the antigenicity and protection of the
synthetic peptides with higher scores in the av-
erage antigen index (AI) and better hydrophilic
properties determined by our data.
Keywords: Dengue Virus, Glycoprotein E, Epi-
tope, Bioinformatics
1. INTRODUCTION
Dengue virus, a flavivirus belonging to the flaviviridae
family, is a mosquito-borne human pathogen that causes
dengue and dengue hemorrhagic fever which is currently
one of the serious public health threats throughout the
tropical and subtropical regions of the world [1]. Four
serotypes of DEN virus have been identified (DEN-1, 2,
3 and 4), and each of these serotypes can infect humans
and cause disease. This virus shares many characteristics
with other flaviviruses, having a single-stranded RNA
genome surrounded by an icosahedral scaffold and cov-
ered by a lipid envelope. The complete virion is 50 nm
in diameter and contains an 11-kb plus-sensed RNA
genome that is composed of seven nonstructural (NS)
protein genes and three structural protein genes, core (C,
100 amino acids), membrane (M, 75 amino acids), and
envelope (E, 495 amino acids) [2,3]. The order of pro-
teins encoded is 5’-CprM(M)-E-NS1-NS2A-NS2B-
NS3-NS4A-NS4B-NS5-3’[4].The 495-amino-acid (aa)
envelop (E) glycoprotein, one of the three structural
proteins, is the principal component of the external sur-
face of the virion [5], and it is responsible for a wide
range of biological activities, including binding to host
cell receptors, fusion to and entry into host cells, there-
fore, this protein directly affects host range, cellular tro-
pism, and, in part, the virulence of the virus [2,5]. Fur-
thermore, the E protein also stimulates host immunity by
inducing protective and neutralizing antibodies [6]. It is
a main target and important antigen for vaccine devel-
opment, and many attempts have been made to elucidate
the structure-function relationships of the dengue virus
glycoprotein E. The crystal structures of protein E ecto-
domains have already been determined. However, the
location of well-defined B-cell and T-cell epitopes for
glycoprotein E is largely unknown. Mapping of the
B-cell and T-cell epitopes should be important for im-
munoinfectomic studies of dengue virus infection. Ran-
dom peptide display has been applied in antigenic epi-
tope determination. However, a combination of compu-
tational methods (e.g., bioinformatics) and experimental
approaches of conventional biology should be a holistic
way to determine the rigorous B-cell and T-cell epitopes.
Thus, in order to characterize the role of glycoprotein E
in the pathogenesis of dengue virus infection, we used
bioinformatics and molecular approaches to predict and
analyze its B-cell and T-cell antigen epitopes. Parameters
SciRes Copyright © 2009

124 H. Zhong et al. / J. Biomedical Science and Engineering 2 (2009) 123-127
SciRes Copyright © 2009 JBiSE
such as hydrophilicity, flexibility, accessibility, turns,
exposed surface, polarity and antegenic propensity of
polypeptide chains have been correlated with the loca-
tion of continuous epitopes in a few well-characterized
proteins. Net servers and the software DNA star are ap-
plied in our study.
2. MATERIALS AND METHODS
2.1. Antigenic Peptide Prediction
The online web server (http://bio.dfci. harvard. edu/
Tools/) give us a pathway to predict sequences of pep-
tides within a protein that are likely to be antigenic by
eliciting an antibody response. Antigenic peptides are
determined using the method of Kolaskar and Tongaon-
kar [7]. Predictions are based on a table that reflects the
occurrence of amino acid residues in experimentally
known segmental epitopes. We enter the amino acids of
dengue virus type 1 HAWAII strain as well as dengue
virus type 2 NGC (New Guinea C) strain, both of whom
are standard strains, then operate the applet.
2.2. Hydrophilicity Estimation
The website locating at (http://us.expasy.org/tools/
protscale.html) can give us a hydrophilicity prediction of
the envelop glycoprotein E, it is based on the method of
Hopp & woods.
2.3. Secondary Structure Presumption
Logging in the same web server mentioned in the third
step, the β-turn and and coil of the E protein can be ob-
tained, using the algorithm [8,9] of Levvit as well as
Deleage & Roux [10,11].
2.4. Surface Accessibility and Average
Flexibility Assumption
The Protean procedure of The DNASTAR software can
supply us with the E protein’s Surface accessibility and
flexibility using methods of Emini [12] and Karplus-
Schulz [13].
2.5. Tertiary Structure Prediction
After the process of secondary structure prediction, we
can use the method of JPRED to link in the PDB ID,
since the sequence of the protein E has 39 hits to the
PDB database with E values of less than 0.0001. Thus,
we make an entry in the PDB in the ID “IP58” to have a
look at the structure by using the viewer Software named
RASWIN32b2a.
3. RESULTS
3.1. Antigenic Peptide Prediction of the
Glycoprotein E
The prediction results for antigenic peptides of the
glycoprotein E for DEN-1 and DEN-2 are shown in
Figure 1.
(a) DEN-1 HAWAII strain
(b) DEN-2 NGC strain
Figure 1. Peptides predicted as antigenic epitope sites of the
protein E.
Table 1. Peptides predicted as B-cell epitope sites of the protein E.
No Start
position Sequence End position
117 GATWVDVVLEHGSCVT 32
239 PTLDIELLKT 48
351 TNPAVLRKLCIE 62
487 DANFVCRR 94
5109 GKGSLITCAKFKCVTK 124
6126 EGKIVQYENLKYSVIVTVHT 145
7169 PTSEIQLTDYGALTLDCSP 187
8193 FNEMVLL 199
9204 KSWLVHKQWFLDLPLPW 220
10234 EDLLVTFKT 242
11246 KKQEVAVLG 254
12278 IFAGHLKCRL 287
13296 GMSYVMCTG 304
14316 QHGTVLVQVK 325
15329 TDAPCKIPF 337
16352 ITANPIVT 359
17373 FGESYIVVGA 382
18424 SIGGVFTSVGKLVHQIFGTAYGVLFSG 450
19458 GIGILLTW 465
20472 SASLSMTCIAVGMVTLYLGV 491

H. Zhong et al. / J. Biomedical Science and Engineering 2 (2009) 123-127 125
SciRes Copyright © 2009 JBiSE
The B-cell (Table 1) and T-cell epitopes (Table 2) of
the glycoprotein of DEN-1 HAWAII standard strain were
predicted by the means of the position of amino acids
with the online server respectively.
3.2. Hydrophilicity Prediction of the Glyco-
protein E
The prediction results for hydrophilicity of the protein E
of DEN-1 and DEN-2 are diagramed in Figure 2.
Table 2. Peptides predicted as T-cell epitope sites of the protein E.
HLA Sites Peptides Position
HLA-A2 206-215 117-126 483-492
HLA-A11 299-308 238-247 50-59
HLA-A24 298-307 439-446 325-334
HLA-B51 216-225 420-446 206-215
HLA-B60 48-57 313-322 256-265
HLA-B62 414-423 291-300 124-133
3.3. Secondary Structure Prediction
The prediction of protein E’s secondary structure is
shown in Figure 3.
3.4. Surface Probability of the Glycoprotein E
The surface probability assumption of the protein E of
DEN-1 and DEN-2 strains is diagramed in Figure 4.
3.5. Flexibility Presumption of the Glyco-
protein E
The Flexibility presumption of the glycoprotein E of
DEN-1 and DEN-2 strains is shown in Figure 5.
3.6. Putative Tertiary Structure of the Gly-
coprotein E
The putative tertiary structure of the glycoprotein E is
shown in Figure 6 (protein data bank). The ID is IP58.
4. DISCUSSION
It is well known that although urgently needed for den-
gue virus infection, specific drugs for treatment and ef-
fective vaccination for prevention are currently unavail-
able. This is the main reason why we have focused on
the so-called Antibody-dependent enhancement (ADE).
Studies suggest that during a secondary infection with a
different serotype, the presence of cross-reactive,
non-neutralizing antibodies enhances the efficiency with
which dengue virus infects susceptible cells. A molecu-
lar understanding of the events that lead to antibody
neutralization, enhancement, or escape will be critical to
the improvement of vaccines It is therefore important to
determine which surface features on the dengue virion
are responsible for inducing protective or enhancing
immune response in the different serotypes. Thus, the
structural and functional organizations of the dengue
virus proteins are of central interests for the understand-
ing of the biology of dengue virus and the mechanisms
of virus-cell interactions.
The dengue E ectodomain consists of structurally dis-
tinct domains: Ⅰ, Ⅱ and Ⅲ [14]. The domain Ⅲ ap-
pears to play an important part in host cell receptor
binding for viral entry and in inducing protective immu-
nity. The rigorous B-cell and T-cell epitopes were not
identified yet. In our study, we focused on the character-
izing the B-cell and T-cell epitopes of dengue virus en-
velop E glycoprotein by deploying the bioinformatics
approaches, the sequence at 281-295 amino acid (aa) for
dengue virus type 1 HAWAII strain and the sequence at
345-359, 383-397 for dengue virus type 2 NGC strain
were predicted as the more prevalent epitopes by using
multiple parameters and different analysis softwares,
respectively. The sequences selected not only have
higher scores in the average antigen index (AI), which
could predict the antigen epitope of envelop glycopro-
tein E, but also showed better hydrophilic properties.
Two epitopes of DEN-2 and one of DEN-1 locate on
(a) DEN-1 HAWAII strain (b) DEN-2 NGC strain
Figure 2. The hydrophilicity of the protein E.

126 H. Zhong et al. / J. Biomedical Science and Engineering 2 (2009) 123-127
SciRes Copyright © 2009 JBiSE
(a1) β-turn the coiled region of DEN-1 HAWAII strain (a2) β-turn the coiled region of DEN-1 HAWAII strain
(b1) β-turn the coiled region of DEN-2 NGC strain (b2) β-turn the coiled region of DEN-2 NGC strain
Figure 3. The prediction of the secondary structure of the protein E.
(a) DEN-1 HAWAII strain
(b) DEN-2 NGC strain
Figure 4. Surface probability of the glycoprotein E.
(a) DEN-1HAWAII strain
(b) DEN-2 NGC strain
Figure 5. The flexibility presumption of the glycoprotein E.
Figure 6. Putative tertiary structure of the glycoprotein E.
the domain Ш and domainⅡ of the protein E, respec-
tively. The domain Ш has been hypothesized to contain
multiple type- and subtype-specific epitopes eliciting
only virus-neutralizing monoclonal antibodies while the
domain Ⅱ is involved in virus-mediated membrane fu-
sion, and contains many cross-reactive epitopes eliciting
neutralizing and non-neutralizing monoclonal antibod-
ies. The predicted epitopes can be used for the devel-

H. Zhong et al. / J. Biomedical Science and Engineering 2 (2009) 123-127 127
SciRes Copyright © 2009 JBiSE
opment of vaccine and the dissection of the ADE effect.
The further experimental studies will be performed to
determine the immunogenicity and protection effect of
peptides with higher scores in the average antigen index
(AI) and better hydrophilic properties, and to identify
vaccine candidates.
ACKNOWLEDGEMENTS
This work was supported by Guangzhou Key Technology R&D Pro-
gram (No.2008Z1-E401 to H. Cao).
REFERENCES
[1] T. Monath.(1994) Dengue: the risk to developed and developing
countries. Proc Natl Acad Sci USA 91, 2395-2400.
[2] R. J. Kuhn, W. Zhang, M. G. Rossmann. (2002) Structure of
dengue virus: implications for flavivirus organization, matu-
-ration, and fusion. Cell 108, 717-725.
[3] E. A. Henchal, J. R. Putnak. (1990) The dengue viruses. Clin
Microbiol Rev 3, 376-396.
[4] C. M. Rice, B. N. Fields, D. M. Knipe, P. M. Howley.
(1996)Flaviviridae: the viruses and their replication. Virology 3,
931-959.
[5] J. T. Roehrig. Immunochemistry of the dengue viruses, 199-219
In: D. J. Gubler and G. Kuno (eds), Dengue and dengue
hemorrhagic fever. CAB International, New York, N. Y. 1997.
[6] G. J. Chang, Molecular biology of dengue viruses, 175-198 In :
D. J. Gubler, G. Kuno (eds.), Dengue and Dengue Hemorrhagic
Fever. CAB International, London, 1997.
[7] A. S. Kolaskar, P. C. Tongaonkar. (1990) A semi-empirical
method for prediction of antigenic determinants on protein anti-
gens, FEBS Lett 276, 172-174.
[8] M. Levitt, J. Greer. (1977) Automatic identification of secondary
structure in globular proteins. J Mol Biol 114, 181-239.
[9] J Cheng. (2008) A multi-template combination algorithm for
protein comparative modeling. BMC Structural Biology 8,
18-36.
[10] G. Deleage, B. Roux. (1987) An algorithm for protein secondary
structure prediction based on class prediction. Protein Eng 1,
289-294.
[11] J. Martin, J. F. Gibrat, F. Rodolphe. (2006) Analysis of an opti-
mal hidden Markov model for secondary structure prediction.
BMC Structural Biology 6, 25-45.
[12] E. A. Emini, J. V. Hughes, D. S. Peflow. (1985) Induction of
hepa-titis A vires-neutralizing antibody by a vires-specific syn-
thetic peptide. J Virol 55,836-839.
[13] P. A. Karplus, G. Schultz. (1985) Prediction of chain flexibility
in proteins. Naturwissenschaften 72, 212-213.
[14] Y. Modis, S. Ogata, D. Clements. (2003) A ligand-binding
pockets in the dengue virus envelop glycoprotein. Science 100,
6986-6991.