National Conference On Computational Biology (NCCB-09) 1
December 28th-29,2009
Sikhya O’ Anusandhan University (SOA University),Bhubaneswar
In-silico modeling and docking Studies of
Antiviral Drugs for Swine flu
Raghunath Satpathy*, Rashmiranjan Behera and Rajesh Kumar Guru
Department of Biotechnology, MIRC Laboratory,
Majhighariani Institute Of Technology & Science, Rayagada, Odisha, India-765017
E-mail: rnsatpathy@gmail.com
Abstract--Swine flu now becomes a deadest disease
throughout the world and the mortality rate is a great
international concern now. Influenza A viruses carry two
surface glycoprotein, the hemagglutinin (HA) and the
neuraminidase (NA) which recognize the same host cell
molecule, sialic acid. Here in this work an attempt has been
taken to improve the binding efficiency of the most common
drugs like Oseltamivir and Zanamivir.A homology based
modeling has been performed by MODELLER tool to get the
3D structure. From docking studies the functional groups of the
molecule responsible for binding to the modeled neuraminidase
enzyme was identified and for that several modification in the
functional groups has been done accordingly. Further docking
with analogs shows that modified functional group with –
CH2COOH in case of Oseltamivir and -CH 2OCH 2CH2OH in
Zanamivir having least energy value with high binding affinity.
I. INTRODUCTION
Human cases of new influenza A (H1N1) virus infection
have been identified recently in many countries [1, 2]. After
the detection of the first cases in Mexico and in the United
States and the spread of infection to further countries, the
World Health Organization (WHO) declared the outbreak of
a new influenza A (H1N1) swl (swine-like) virus infection to
be a public health emergency. The recent outbreak of the
novel strain of influenza A (H1N1) virus has raised a global
concern of the future risk of a pandemic. A recent report says
that currently isolates of H1N1 virus are resistant to presently
used neuraminidase inhibitor drugs like Zanamivir and
Oseltamivir [3]. To understand at the molecular level how
this new H1N1 virus can be inhibited by the current anti-
influenza drugs and which of these drugs it is likely to
already be resistant to homology modeling and MD
simulations have been applied on the H1N1
neuraminidase[4].Methods developed to facilitate and
speedup the drug designing process are Rational Drug Design
(RDD). These processes are used in biopharmaceutical
industry to discover and develop new drugs. RDD uses a
variety of computational methods to identify novel
compounds. One of those methods is docking of drug
molecules with receptors. The site of drug action, which is
ultimately responsible for the pharmaceutical affect, is a
receptor [5].

In the present study a computational approach is used to
predict the three dimensional structure of neuraminidase
enzyme sequence of a novel H1N1 virus in China by
homology modeling method [6]. The approach produces the
valid structural model with the available template which
having 88% amino acid identity. After validation of the
model the in-silico docking process with different analogs of
drug Oseltamivir and Zanamivir is performed this shows that
the analog having significant binding energy can as be used
as the probable lead molecule for the H1N1 virus. So by the
rational drug design approach, an effort has been made to
design novel modified drugs by using the conventional drugs
like Oseltamivir and Zanamivir and docking against the
predicted structure of H1N1 virus [6].
II. MATERIALS AND METHODS
.
A . Sequence retrieval and 3D model building
The sequence for the neuraminidase enzyme was retrieved
from NCBI having ID GQ232095 [7]. Then with this query
sequence a BLAST [8] search was performed against PDB
(Protein Databank) to retrieve the corresponding template for
the neuraminidase. Searching for the suitable template for the
neuraminidase in the PDB (protein data bank) was performed
basing on criteria viz, the native protein should have E value
<0.001, identity >85% with the target sequence. The model
was built by homology modeling and for this MODELLER
[9] program was used. The MODELLER program uses an
automated approach to comparative protein structure
modeling by satisfaction of spatial restraints. Briefly, the
core modeling procedure begins with an alignment of the
sequence to be modeled (target) with related known 3D
structures (templates). This alignment is usually the input to
the program. The output is a 3D model for the target
sequence containing all main chain and side chain non-
hydrogen atoms [10].
 2

receptor, inhibit its function, and thus act as drug. The

collection of Oseltamivir, Zanamivir and neuraminidase
complexes was identified via docking. The binding energy
was computed and the ligand binding pattern was observed.

III. RESULTS

A . Homology modeling of h1n1 neuraminidase

The sequence for the neuraminidase protein having 468
amino acid was obtained from NCBI having accession
number gq 232095.The sequence was directly submitted to
NCBI from Chinese national influenza centre on 4th June
2009. The template structure 3CYE [18] is considered as the
suitable template by considering the E-value (0.0), identity
(88%) for the homology modeling by Modeller program.

Figure1 Showing the working principle of MODELLER

program (Sali and Blundell 1993)

B. Model Validation
The MODELLER generated structure was further verified
by PROCHECK [11].The PROCHECK program provides the
information about the stereo chemical quality of a given
protein structure. The PROCHECK was used to generate
Ramachandran plot and the quality of the structure was Fig.2. Showing Rasmol [18] view of final 3D structure of the
computed in terms of % of residues in favorable regions, % h1n1 neuraminidase. Yellow color shows beta sheets and Red
of non Proline Glycine residues etc.The quality of structure one is the alpha helixes.
was also further accessed by using ERRAT [12]. ERRAT is a
protein structure verification algorithm that is especially Table 1 The structural information of the modeled
well-suited for evaluating the progress of crystallographic neuraminidase protein
model building and refinement. Along with ERRAT data the Structural features Information
DOPE (discrete optimized potential energy) score [13] per No. of residues 468
residues was observed. DOPE score is calculated by Modeler No. of atoms 3616
program which is the distance dependent statistical potential No. of Hydrogen bonds 252
based on probabilistic theory. This is extremely useful in No. of helices 3
making decisions about reliability. Then the backbone No. of sheets 42
alignment and RMSD (root mean square deviation) study was No. of turns 60
performed by using PYMOL tool [14]. PyMOL is an useful
molecular modeling software allowing the visualization of IV . VALIDATON OF THE MODEL
three dimensional molecular structures as well as for the
backbone alignment between the model and template. A. Structure validation by ERRAT and DOPE score.
To verify further the predicted structures, the coordinates
C. Preparation of drug analogs of the modeled protein was fed into the ERRAT Protein
From the PDB the drug molecular structure of Oseltamivir Verification Server. The overall quality factor was obtained
and Zanamivir were retrieved. Protein Data Bank (PDB) is a as 64. 978 which is satisfactory. The result obtained is shown
repository for the storing of X-ray and NMR retrieved in Figure 3. (B).The DOPE scores of both template and
structural data of molecules for proteins and nucleic acids. model obtained from Modeller output has been shown in
The structural analog for the above drugs were prepared by Figure 3(A). From the comparative chart the peak is showing
MARVIN SKETCH [15]. Marvin sketch is a Java based that there is no defect in the loop regions in the residues. So
chemical drawing tool which allows creating and editing of in the present case the loop refinement method is not
molecules in various file formats. required for the model.
along with the original molecule by the HEX tool. The
D. Docking study energy after each docking was computed given in table 3(A)
Docking was performed between the validated structure of and 3(B).
neuraminidase and the structural analogs by using HEX 5.0
tool [16]. HEX is an interactive molecular graphics program
for protein-ligand docking, assuming the ligand is rigid, and
it can superpose pairs of molecules using only knowledge of
their 3D shapes. It is still the only docking and superposition
program to use spherical polar Fourier (SPF) correlations to
accelerate the calculations [17]. The molecules binding to a
 3

% residues in favorable regions 81.7

% residues in additional residue regions 15.5
% residues in generously regions 1.3
% residues in disallowed regions 1.5
% of non Proline and non Glycine residues 100

B. Docking study
The docking study was performed between the modeled
structure and molecular analogs of Zanaminvir and
Oseltamivir
(A)
Table 3 (A) Docking results of modeled neuraminidase by
Zanmaivir

Drugs docked E -values

Zamanivir -207.88
Zamanivir analog 1 -205.81
Zamanivir analog 2 -202.92
Zamanivir analog 3 -215.49
Zamanivir analog 4 -204.48
Zamanivir analog 5 -203.74
Zamanivir analog 6 -206.54
Zamanivir analog 7 -217.52
Zamanivir analog 8 -212.19
(B) Zamanivir analog 9 -222.02
Fig.3 (A), (B).showing the DOPE score comparison between Zamanivir analog 10 -228.44
the model and template and the ERRAT validation server
output respectively. Table 3 (B) Docking results of modeled neuraminidase by
Oseltamivir
B. Structure validation by PROCHECK
The validation of the model was carried out using Drugs docked E -values
Ramachandran plot calculations computed with the Oseltamivir -193.53
PROCHECK program Figure 4. The Φ and Ψ distributions of Oseltamivir analog 1 -195.58
the Ramachandran plots of non-Glycine, non-Proline residues Oseltamivir analog 2 -201.36
are summarized in Figure 8 and table 2. Altogether 100% of Oseltamivir analog 3 -198.22
the residues were in favored and allowed regions. The overall Oseltamivir analog 4 -198.83
G-factor used was computed as -0.46. Oseltamivir analog 5 -207.61
Oseltamivir analog 6 -208.03
Oseltamivir analog 7 -229.53
Oseltamivir analog 8 -231.21
Oseltamivir analog 9 -209.74
Oseltamivir analog 10 -224.83

Fig.4.

Showing the Ramachandran plot view from PROCHECK

output.

Table2 Ramachandran plot calculations on 3D model of

Neuraminidase protein computed with the PROCHECK
program

(A) Zanamivir (original molecule)
(B) Oseltamivir (original molecule)
(C) Zanamivir (Analog 10)
[5] Virupakshaiah DBM, Chandrakanth Kelmani, Rachanagouda Patil, and
Prasad Hegade,
[6] Huang IC, Li W, Sui J, Marasco W, Choe H, Farzan M , "Influenza A virus
neuraminidase limits viral super infection", J. Virol. vol 10, pp. 4834–4843,
2007.
[7] http://www.ncbi.nlm.nih.gov/nuccore/GQ232095.
[8] Altschul, S.F., Gish, W., Miller, W., Myers, E.W. & Lipman D.J” Basic
local alignment search tool.” J. Mol. Biol. vol 215, pp.403-410, 1990
[9] Eswar N, Eramian D, Webb B, Shen M, Sali A” Protein structure modeling
with
MODELLER Methods,” Mol. Biol, vol 426 pp.145-159, 2008.
[10] A. Sali & T.L. Blundell.” Comparative protein modeling by satisfaction
of spatial restraints” J. Mol. Biol., vol 234 pp.779-815, 1993.
4
(D) Oseltamivir (analog 8)
Fig.5. Showing the original drug molecule of Zanamivir and
Oseltamivir (A, B) and analogs of original drug molecule (C,
D).
V. CONCLUSION
The current drugs like Oseltamivir and Zamanivir are
potential Neuraminidase inhibitor hence used against swine
flu virus H1N1. In this work, we have constructed a
homology based 3D model of Neuraminidase of China H1N1
virus, using the MODELLER software. The final refined
model was further assessed by ERRAT and PROCHECK
program, and the results obtained that this model is reliable.
The stable model is further subjected for docking with the
different drug analogs of Oseltamivir and Zanamivir.
Docking results indicate that the analog 8 in case of
Oseltamivir and analog 10 in case of Zamanivir show more
potent binding activity than the original drugs, this
corresponds to inhibition activity against the viral
neuraminidase. More over this type of analysis show that the
some of the modified drugs having more potential activity
than the original one. So in our study the drug analogs
showing higher binding activities can be used as the probable
lead molecules than the rest of drugs for Swine flu.
ACKNOWLEDGMENT
We are thankful to CEO, Director and Dean of
Majhighariani Institute Of Technology & Science ,Ryagada
to provide us the MIRC lab for computing facility.
REFERENCES
[1] Salomon R, Webster RG, “The influenza virus enigma,” Cell, vol 136,
pp.402-410, 2009.
[2] Butler D,” Swine flu goes global,” Nature, vol 458, pp.1082-1083, 2009.
[3] Sheri mossad,”The resurgence of swine-origin influenza A (H1N1)
Cleveland clinical journal of medicine,”vol 76, pp.337-343, 2009.
[4] Rungrotmongkol T, etal,”Susceptibility of antiviral drugs against 2009
influenza A (H1N1) virus,” Biochem Biophys Res Commun. May 2009.
“Computer Aided Docking Studies on Antiviral Drugs for SARS “PWASET,
vol 24, pp. 297-299, 2008.
[11]Laskowski R A, MacArthur M W, Moss D S & Thornton J M:
“PROCHECK: a program to check the stereo chemical quality of protein
structures “ J. Appl. Cryst., vol 26 pp.283-291,1993
[12] V C. Colovos and T. O. Yeates “Verification of protein structures:
patterns of non bonded atomic interactions” Protein Sci vol 2, pp. 1511–1519,
1993
[13] Min-Yi Shen, Andrej Sali “Statistical potential for assessment and
prediction of
Protein structures” Protein Science vol 15 pp.2507-2524, 2006
[14] Delano, W.L “The PyMol Molecular Graphics System (2002) on World
Wide Web www.pymol.org”.

[15] M Péter Csizmadia,”Marvin sketch and Marvin view molecule applets
for the World Wide Web,” Third International Electronic Conference on
Synthetic Organic Chemistry (ECSOC-3), September 1-30.
[16]D.W. Ritchie,” Evaluation of Protein Docking Predictions Using Hex 3.1
in CAPRI Rounds 1 and 2: PROTEINS,” Struct. Funct. Genet. vol 52 pp 98-
106. 2003
National Conference On Computational Biology (NCCB-09)
December 28th-29,2009
Sikhya O’ Anusandhan University (SOA University),Bhubaneswar
5
[17] L. Mavridis and D.W. Ritchie, Toward High Throughput 3D Virtual
screening using Spherical Harmonic Molecular Surface Representations: J.
Chem. Inf. Model vol 47 pp1787-1796, 2008
[18] X .Zhu, X.Xu, I.A .Wilson “Structure Determination of the 1918 H1N1
neuraminidase from a crystal with lattice –translocation defects,
“unpublished.