ArticlePDF Available

In Silico Screening of Novel Inhibitors for HPV: A Rational Structure Based Approach (Docking Versus Pharmacophore Model Generation)


Abstract and Figures

The E2 protein from HPV 16 was selected as a molecular target and its known structures were exploited for broad scope of "hits" to be identified in the screening process. We compared both structure-based and ligand-based design approaches for virtual screening. Databases enriched in natural compounds were used for virtual screening based on molecular docking. In this study, we identified novel classes of HPV inhibitors by means of a structure-based drug-design protocol involving Pharmacophore based virtual screening with molecular docking simulation.
Content may be subject to copyright.
Letters in Drug Design & Discovery, 2009, 6, ???-??? 1
1570-1808/09 $55.00+.00 © 2009 Bentham Science Publishers Ltd.
In Silico Screening of Novel Inhibitors for HPV: A Rational Structure
Based Approach (Docking Versus Pharmacophore Model Generation)
G. Reshmi and M. Radhakrishna Pillai*
Translational Cancer Research Laboratory, Rajiv Gandhi Center for Biotechnology, Thycaud PO, Thiruvananthapu-
ram-695014, Kerala, India
Received March 05, 2009: Revised June 09, 2009: Accepted June 09, 2009
Abstract: The E2 protein from HPV 16 was selected as a molecular target and its known structures were exploited for
broad scope of “hits” to be identified in the screening process. We compared both structure-based and ligand-based design
approaches for virtual screening. Databases enriched in natural compounds were used for virtual screening based on mo-
lecular docking. In this study, we identified novel classes of HPV inhibitors by means of a structure-based drug-design
protocol involving Pharmacophore based virtual screening with molecular docking simulation.
Keywords: HPV, TAD, Molecular docking, Pharmacophore model, Virtual screening, GOLD.
Oncogenic Papillomaviruses (HPVs) are responsible for
approximately half a million cases of cervical cancer each
year. HPVs also cause genital warts, and are the most com-
mon sexually transmitted diseases in many countries. This
awareness has paved the way for new exciting approaches
for preventing cervical cancer, which is the second most
common malignancy of women worldwide [1]. The principal
agent in the etiology of cervical cancer, human papillomavi-
rus (HPV) type 16, encodes early proteins, E1 and E2, and is
involved with host cell systems to maintain and produce vi-
rus [2]. Loss of E2 function, which usually occurs in early
stage cancers, allows dysregulated expression of the viral
oncoproteins, E6 and E7 [3]. HPV E2 binds with high affin-
ity to specific sites in viral DNA and also binds to E1 and
targets this replication factor to the viral origin of replication.
HPV16 E2 and Cervical Tumor Progression
The major factor contributing to cervical cancer progres-
sion rests upon disruption or loss of the E2 gene product. E2
protein maintains a strict control over the E6 and E7 gene
expression while it is maintained in the extra chromosomal
form. During the progression of cervical cancer, the viral
DNA is integrated into the host genome such that E6 and E7
and the upstream transcriptional regulatory region are intact
while the expression of E2 is lost suggesting that these pro-
teins play a regulatory role in carcinogenic progression [4-5].
It has been postulated that in the absence of E2 polypeptides,
the transcription control over E6 and E7 is lifted leading to
their expression in a deregulated manner which may lead to
tumor formation.
Druggable Hot Spots of Molecular Target-HPVE2
E2 protein of human papillomavirus regulates the
primary transcription and replication of the viral genome.
*Address correspondence to this author at the Translational Cancer Re-
search Laboratory, Rajiv Gandhi Center for Biotechnology, Thycaud PO,
Thiruvananthapuram-695014, Kerala, India; Tel: +91- 471-2341716 |
2347975; Fax: 91- 471-2348096; E-mail:
Both activities are governed by a-200 amino acid N-terminal
module (E2NT) which is connected to a DNA binding C-
terminal module by a flexible linker [6]. The crystal structure
of the E2NT module from high-risk type 16 human
papillomavirus reveals an L-shaped molecule with two
closely packed domains, each with a novel fold. The
transactivation module is composed of two domains, N1 and
N2, arranged so as to give it an overall L-shaped appearance.
Arg37, Glu39 and Ile73 are three conserved amino acids,
their point mutational effects mainly control the two
principal functions of E2, transactivation and HPV DNA
replication [reviewed in 7-10]. The dimerisation surface of
E2 represents a good target for designing anti-viral drugs,
since it is essential for viral transcription, there is no
homologous human protein and the residues forming the
interface are highly conserved among different viral strains.
It has been suggested that E2 can regulate the switch
between early gene expression and viral genome replication
during HPV infection [11]. Inhibition of HPV E2 might lead
to increased expression of the E6 and E7; however, recent
publications question this assumption which is based primar-
ily on E2 over-expression. For these reasons targeting E2
might be effective in early lesions by inhibiting replication
and viral episome segregation.
Structure-Based Drug Discovery
Structure-based drug discovery has made significant pro-
gress in the past 30 years benefiting from recent advances in
high performance and distributed grid computing [12]. Com-
puter-aided or in silico design is being utilized to expedite
and facilitate hit identification, hit-to-lead selection, optimize
the absorption, distribution, metabolism, excretion and toxic-
ity profile and avoid safety issues [13]. Virtual screening
[14], which has shown a great promise in drug discovery,
will play an important role in digging out active leads from
large compound libraries. It is intended to reduce the size of
chemical space and thereby allow focus on more promising
candidates for lead discovery and optimization. Natural
products, containing inherently large scale structural diver-
sity than synthetic compounds, have been the major re-
sources of bioactive agents and will continually play as pro-
2 Letters in Drug Design & Discovery, 2009, Vol. 6, No. 7 Reshmi and Pillai
tagonists for discovering new drugs [15]. Here, we focus on
an integrated database screening strategy involving 3D data-
base searching approaches such as Pharmacophore mod-
els(ligand-based design) applicable as queries in 3D database
searches and molecular docking (structure-based design) of
potential ligands into a binding pocket. In the post-genomic
era, rational anti-cancer drug discovery aims to discover or
design small molecules that modulate the activity of key
therapeutic targets pivotal for carcinogenesis [16]. This work
mainly aims to discover novel small molecular inhibitors
against important molecular targets involved in cervical can-
This in silico study strongly highlights the importance of
computational approaches in drug discovery and an attempt
to identify putative small inhibitors that might inhibit the
activity of the target molecule chosen for the study (HPV
E2NT). GOLD [17] a commercial docking package was used
to dock ligands in the E2NT binding site by virtual screening
(VS) of downloaded ligand databases [18]. Since there are
no known inhibitors of HPV 16 E2 TAD, we developed the
ligand-based Pharmacophore model (generated by Ligand
Scout 2.02[19]) based on the ligand bound structure of HPV
11E2 TAD. The overall structure for both proteins has been
found to be highly similar. In our study we developed a bind-
ing site model for HPV 16 E2 TAD and this promising result
confirmed the applicability of our active site model for fur-
ther studies. To date, there have been no lead generation and
optimization studies reported for this transactivation domain
of HPV E2 receptor. In the present study, we attempt virtual
screening (VS) of a number of ligands on HPV E2 and
evaluate how different ligand poses change with respect to
the scoring functions.
Active Site Prediction
The X-ray crystallographic structure of the complete
transactivation domain (TAD) of human papillomavirus type
16 E2 protein was retrieved from the Protein Data Bank,
(PDB-ID: 1 DTO) [6]. We decided to exploit the protein-
ligand structure as a starting point for structure-based virtual
screening based on molecular docking. Since HPV 16 E2
TAD lacks any bound ligands, the binding site residues need
to be identified prior to docking. Experimental studies car-
ried out by Wang et al. [20] provide an insight into the ex-
treme similarity between HPV 11 and HPV 16 E2 TAD pro-
tein structures. The all C superposition of HPV 11 and
HPV 16 (187 residues) gave an RMSD of mere 1.54 Å. The
N terminal (residues 3-94) and C terminal (residues 95-195)
sub-domains of these proteins were superimposed individu-
ally and were shown to yield RMSD of 0.85 Å and 1.14 Å,
respectively. All the secondary structural elements were ob-
served to be conserved. Identification of homologous resi-
dues making up the active site of HPV16 E2 was carried out.
This was taken into account in conjunction with the results
obtained from Site Finder implemented in MOE (Chemical
Computing Group Inc., Montreal, Canada). HPV 16 E2 TAD
active site is a channel that is lined with hydrophobic resi-
dues and protrudes towards the centre .The binding pocket
includes both a deep and narrow cavity and Tyr 32 sterically
hinders the binding pocket. Active site residues of HPV 16
E2 TAD are Ile 15, Tyr 19, Asp 28, His 29, Ile 30, Tyr 32,
Trp 33, Met 36, Glu 39, Ala 63, Val 64, Ser 65, Asn 67, Lys
68, Ala 69, Gln 71, Ala 72, Leu 79, Thr 93, Leu 94, Gln 95,
Val 97, Ser 98, Leu 99, Glu 100, Val 101.
Preparation of the Protein and the Binding Site
After detection of active site of the TAD domain, a ‘clean
input file’ was generated by removing water molecules, ions,
ligands and subunits not involved in ligand binding from the
original structure file. Hydrogen atoms were then added to
the protein and the active site was inspected to make suitable
corrections for tautomeric states of histidines, hydroxyl
group orientations and protonation states of charged residues
[21-22]. Local minimization was then performed in the pres-
ence of restraints to relieve potential bad contacts, at the
same time maintaining the protein conformation very close
to that observed in the crystallographic model. The resulting
receptor model was saved to a PDB file (compatible with
GOLD input file formats).
Preparation of the Database
The chemical databases used in our virtual screening in-
cluded the ZINC (http:/, Chembank
(, and Ambinter (http://www. [23]. Collectively, these three databases,
offered a collection of ~one million small-molecule organic
compounds. From this databases physiochemical properties,
such as LogP, number of H-bond donors/acceptors, numbers
of rotatable bonds calculated and 2D filters were applied to
remove inorganic, and compounds that were not drug-like
[24]. The filtered datasets were converted 2-D to 3-D using
CORINA (Molecular Networks GmbH) program with the
standard settings.
Molecular Docking
The molecular docking program GOLD (Version 3.0.1)
[17], was used to perform the virtual screening [25]. For the
study, the binding pocket on the TAD was the region tar-
geted for docking. The receptor was defined from the crys-
tallographic coordinates of the ligand (residues within 3.5 Å
of the ligand). The binding site atoms were further defined
from a file listing the cavity atoms. Dockings were per-
formed under ‘Standard default settings’ mode. The docking
results were analyzed using SILVER software, which allows
visualization of the protein-ligand docking and calculation of
several descriptors such as feasible hydrogen bonding be-
tween the protein and the ligand [22]. The Smiles formula of
the molecules with the highest scores was taken from the
corresponding entries in the database and was checked for
fulfillment of the Lipinski rule of five using PubChem [24,
26]. GOLD is widely regarded as one of the best docking
programs [28-30], Goldscores and Chemscores fitness func-
tion was employed in scoring compounds. For each of the 10
independent Genetic Algorithm (GA) runs, with a selection
pressure 1.1, 100 000 and default operator weights were used
for crossover, mutation, and migration of 95, 95, and 10,
respectively. Default cutoff values of 3 Å for hydrogen
bonds and 4.0 Å for vdW were employed while docking, a
limited flexibility is allowed for the hydrogen atoms in the -
In Silico Screening of Novel Inhibitors for HPV Letters in Drug Design & Discovery, 2009, Vol. 6, No. 7 3
OH and -NH3 substituents of the side chains of Ser, Thr, and
Lys residues. The best docked solutions were subjected to
energy minimization, re-docking and re-scoring [31-33]. The
highest ranking molecules were re-docked independently
into the protein active site under the default settings and
were checked for satisfactory GOLD scores and exhibition
of correct binding to the active site of the protein molecule.
Pharmacophore Generation
In the pharmacophore generation procedure, is important
to know biological function of the studied target. In this
study, the Pharmacophore model was derived by means of a
new tool LigandScout [19], a program for automatic Phar-
macophore identification from ligand bound protein com-
plexes. Here we developed a model from ligand bound
HPV11 E2 transactivation domain because it shows extreme
similarity with HPV 16 E2 TAD. The E2-TAD of HPV11
shares 47% amino acid identity with HPV16 and at position
101, HPV16 has a valine, whereas “low-risk” HPV11 has a
methionine at this position. We referred common chemical
features derived from the bioactive ligand conformations of
protein complex of HPV11 E2 TAD and LigandScout gener-
ates models of each one. The chemical features of both the
ligands were effectively mapped and the pharmacophore
model developed for HPV 16 E2 TAD protein. This model
was used to perform a Pharmacophore search of 3-D com-
pound database to identify ‘hits’ that satisfy all conforma-
tional requirements.
In Silico Screening
After assessing the query Pharmacophore model, virtual
screening was carried out by using the tool MOE (Chemical
Computing Group Inc., Montreal, Canada).The flexible
search mode was adopted to screen the database which con-
tains the structural information of millions of chemicals. The
resulting hit molecules were ranked and the compounds with
highest fit values were extracted and subjected to docking
study to select hits which satisfy the hydrogen bond acceptor
(HBA) feature of the models.
Virtual Screening Using Molecular Docking
The application of docking offers the possibility of virtu-
ally testing the ability of a molecule to bind to a target.
Docking tools utilize the crystal structure of a target protein
and find different binding modes for the molecules of inter-
est. Before preceding to Virtual Screening it was necessary
to identify the binding pocket residues of HPV 16 E2 TAD
because its X-ray crystallographic structure retrieved from
the PDB lacks any bound inhibitors. Therefore in the present
case, the binding pocket was detected by Site Finder module
implemented in MOE (Chemical Computing Group Inc.,
Montreal, Canada) Fig. (1), with the help of shape-based
algorithm the tool identified concave regions on the protein
surface. The module also evaluates whether these cavities
have an appropriate composition of hydrophobic and hydro-
philic atoms. The genetic algorithm-based docking program
GOLD [27] was used to screen the 3D databases, flexibly
docking million compounds to HPV E2 TAD. A total of
1568 molecules from a pool of 86,723 molecules were iden-
tified as hits while 157 molecules were re-docked, re-scored
and indicated as highly active. The targeted region (HPV E2
TAD) defined in our virtual screening covers Tyr19, Tyr32,
Leu94 residues in active site. We hypothesize that a small
molecule that binds to this region will compete with the
above residues, consequently blocking the transactivation of
domain Fig. (2). Some of the hits retrieved in database
search also showed good Gold scores and have good interac-
tions with active site residues. Three hits out of 46 com-
pounds with high GOLD dock score >50 listed in Table (1)
were considered as final hits for further experimental evalua-
tion Fig. (3). These molecules showed extraordinary results
with respect to all properties like calculated drug-like proper-
ties [21, 24] listed in Table (2) and thus can be treated as
good putative hits in the design of potent inhibitors of HPV.
Fig. (1). Transactivation domain of HPV16 E2 protein showing
binding pocket.
Fig. (2). HPV16 E2 TAD domain with major active site residues.
Virtual Screening Using Pharmacophore Model
The pharmacophore model generated by the LigandScout
program includes five features: one hydrogen bond donor
4 Letters in Drug Design & Discovery, 2009, Vol. 6, No. 7
Reshmi and Pillai
Table 1. Potential Hit Compounds with High Gold Scores Obtained by Structure- Based Approach
Compound And Zinc id Gold Score Compound Structure
sulfanylacetyl] amino] phenyl] acetamide
Zinc id: 01083718
59.11 C22H20N2O4S2
Zinc id:03302168
59.01 C19H19N3O3S
-1,2,4-triazol-3-yl] sulfanyl]-N-
(3,4-difluorophenyl) acetamide
Zinc id:02393865
Mol.Wt - 419.449
57.35 C19H19F2N5O2S
Fig. (3). Docked conformation of potential hits and respective active site residues involved in binding (structure-based approach).
In Silico Screening of Novel Inhibitors for HPV Letters in Drug Design & Discovery, 2009, Vol. 6, No. 7 5
Table 2. Table Showing Values of Lipinski Parameters for Hits Obtained
No Zinc id Pubchem id Molecular weight H donors H Acceptors Log p
1 01083718 1269872 440.537 4 4 4.31
2 03302168 2417786 369.439 2 6 3.30
3 02393865 1992969 419.449 2 8 2.94
Fig. (4). Docked conformation of potential hits obtained from GOLD (Ligand-based approach).
Fig. (5). (A): Pharmacophore model of Ligand-1 used within LigandScout (red arrows, HBA; yellow spheres, hydrophobic sites; gray, ex-
cluded volumes).
(B): Pharmacophore model of Ligand-2 used within LigandScout (red arrows, HBA; yellow spheres, hydrophobic sites; gray, excluded vol-
6 Letters in Drug Design & Discovery, 2009, Vol. 6, No. 7
Reshmi and Pillai
(HBD), two hydrogen bond acceptors (HBA) and two hy-
drophobic groups. Besides, the program automatically gen-
erated several excluded volumes. Herein, we present struc-
ture-based PCM Fig. (5A) and Fig. (5B) which are used for
first -screening [34-35]. In order to select hits satisfying the
hydrogen bond acceptor (HBA) features of the models, mo-
lecular docking was carried out as second-screening [36].
The model query PCM Fig. (6) was exported to MOE for
subsequent VS with ZINC database (http:/blaster.docking.
org/zinc).Only the filtered subset of 5000 molecules was
used for the screening. VS with ZINC using ligand-based
Pharmacophore model yielded 82 hits that satisfied the
specified requirements and out of which three compounds
shows high GOLD dock score >50 listed in Table (3). All the
hits obtained in database searching underwent molecular
docking into HPV 16 E2 TAD binding pocket with GOLD
Fig. (4). These putative hits showed good drug-like proper-
ties listed in Table. (4). The results indicate that our model is
as a valuable tool for screening databases to identify mole-
cules with potential antiviral activity.
Fig. (6). LigandScout Pharmacophore model generated from
HPV11 E2 TAD/ligand complex (red arrows, HBA; yellow
spheres, hydrophobic sites; gray spheres, excluded volumes).
Table 3. Potential Hit Compounds with High Gold Scores Obtained by Ligand -Based Approach
Compound And Zinc id Gold Score Compound Structure
Bis (phenylmethoxyamino) phosphoryloxybenzene)
Zinc id: 03851138
56.2 C20H21 N2O4P
2-[[4-amino-5-(4-methyl phenyl)
-1,2,4-triazol-3-yl] sulfanyl]-N-
(5-chloro-2-methyl-phenyl) acetamide
Zinc id:02387705
55. 65 C18H18lCN5OS
Zinc id:02361742
Mol.Wt- 442.537
55.41 C24H22N6OS
In Silico Screening of Novel Inhibitors for HPV Letters in Drug Design & Discovery, 2009, Vol. 6, No. 7 7
Since HPV is the major causative factor of cervical can-
cer, an inhibitor of HPV E2 may block progression to inva-
sive cancer by inhibiting HPV replication. To find such puta-
tive inhibitors of HPV E2, the x-ray structure of the transac-
tivation domain of HPV E2 was chosen as a molecular tar-
get. Since experimental screening of therapeutic targets is
time-consuming and cost-intensive, in this study we used
computational screening approaches to identify putative
small molecular inhibitors of HPV16 E2 from natural com-
pound enriched databases. In the present work, we carried
out docking studies on the commercial dataset of million
molecules to HPV 16 E2 TAD protein, with the purpose of
developing a docking protocol fit for the target under study,
eventually applicable for more time-consuming virtual
screening of larger database of compounds. Usually, docking
to protein structures that do not have a ligand present, as in
the case of HPV 16 E2 TAD protein, dramatically reduces
the expected performance of structure-based methods. To
overcome such a limitation, we enhanced the simple docking
procedure by means of a combined structure-and ligand-
based drug design approach. The mapping information based
on the Pharmacophore model we developed is now being
taken advantage in the identification of novel lead molecules
through 3D databases searches.
In conclusion all of these goals have been realized and 6
putative inhibitors identified. Hence, these candidates can be
taken to the wet laboratory in order to carry out further test-
ing, optimization and investigation of their specific activities
and effectivity against HPV related diseases. In this respect,
this paper represents the first example of successful indi-
viduation of a potential hits through molecular docking
simulations on a HPV 16 E2 TAD target. At the same time,
compounds showing good docking scores and satisfy Lipin-
ski’s rule of five showing good drug-like properties derived
from this structure and its different binding modes, could
carry through further lead optimization to more potent HPV
16 E2 TAD inhibitors. Overall, the results from these Phar-
macophore versus docking studies underscore the impor-
tance and usefulness of employing computational biology
and chemoinformatics approaches in the areas of drug dis-
covery and molecular target identification.
This study is supported by Department of Biotechnology
Government of India, (Grant No: - BT/PR5435/BID/07/095/
VS = Virtual screening
TAD = Transactivation domain
PCM = Pharmacophore model
HBA = Hydrogen bond acceptor
HBD = hydrogen bond donor
[1] Sherman, M.E.; Wang, S.S.; Carreon, J; Devesa, S.S. Mortality
Trends for cervical squamous and adenocarcinoma in the United
States. Relation to incidence and survival. Cancer, 2005, 103, 258-
[2] William Phelps, C; Peter Howley, M. Transcriptional transactiva-
tion by the human papillomavirus type 16 E2 gene product. J. Vi-
rol., 1987, 61(5), 1630-38.
[3] Brian Van Tine A.; Luan D. Dao.; Shwu-Yuan Wu, Timothy M. S.;
Biing Yuan Lin.; Nianxiang Zou.; Cheng-Ming Chiang.; Thomas
R. B. and Louise T. C. Human papillomavirus (HPV) origin bind-
ing protein associates with mitotic spindles to enable viral DNA
partitioning. Proc. Natl. Acad. Sci. USA, 2004, 101(12), 4030-
[4] McGlennen, R.C. Human papillomavirus oncogenesis. Clin. Lab.
Med., 2000, 20(2), 383-406.
[5] Romanczuk, H; Howley, P. M. Disruption of either the E1 or the
E2 regulatory gene of human papillomavirus type 16 increases viral
immortalization capacity. Proc. Natl. Acad. Sci. USA, 1992, 89(7),
[6] Alfred Antson, A.; Julie E. Burns.; Olga V. Moroz.; David J.
Scott.; Cyril M. Sanders.; Igor B. Bronstein.; G. Guy Dodson.;
Keith S. Wilson. and Norman J. Maitland. Structure of the intact
transactivation domain of the human papillomavirus E2 protein.
Nature, 2000, 403, 805-809.
[7] Cooper, C. S.; Upmeyer, S. N. and Winokur, P. L. Identification of
single amino acids in the human papillomavirus 11 E2 protein
critical for the transactivation or replication functions. Virology,
1998, 241, 312-322.
[8] Breiding, D. E.; Sverdrup, F.; Grossel, M. J.; Moscufo, N.;
Boonchai, W. and Androphy, E. J. Functional interaction of a novel
cellular protein with the papillomavirus E2 transactivation domain.
Mol. Cell Biol., 1997, 17, 7208-7219.
[9] Yao, J. M.; Breiding, D. E. and Androphy, E. J. Functional
interaction of the bovine papillomavirus E2 transactivation domain
with TFIIB. J. Virol., 1998, 72, 1013-1019.
[10] Hegde, R. S. and Androphy, E. J. Crystal structure of the E2 DNA-
binding domain from human papillomavirus type 16: Implications
for its DNA binding-site selection mechanism. J. Mol. Biol., 1998,
284, 1479-1489.
[11] Hegde, R. S. The papillomavirus E2 proteins: structure, function,
and biology. Annu. Rev. Biophys. Biomol. Struct., 2002, 31, 343-
[12] Kapetanovic, I.M. Computer-aided drug discovery and develop-
ment (caddd): in silico-chemico-biological approach. Chem.-Biol.
Interact., 2008, 71(2), 165–176.
[13] van de Waterbeemd, H; Gifford, E. ADMET in silico modelling:
towards prediction paradise? Nat. Rev. Drug Discov., 2003, 2(3),
[14] Kontoyianni, M.; Sokol, G.S.; McClellan, L.M. Evaluation of
library ranking efficacy in virtual screening. J. Comput. Chem.,
2005, 26(1), 11-22.
[15] Von Grotthuss, M.; Koczyk, G.; Pas, J.; Wyrwicz, L.S.;
Rychlewski, L. Ligand. Info small-molecule Meta-Database.
Comb. Chem. High Throughput Screen., 2004, 7, 757-761.
[16] Kitchen D.B.; Decornez H.; Furr J.R.; Bajorath J. Docking and
scoring in virtual screening for drug discovery: methods and appli-
cations. Nat. Rev. Drug Discov., 2004, 3(11), 935-49.
Table 4. Lipinski Parameters for Hits Obtained by Ligand -Based Approach
No Zinc id Pubchem id Molecular weight H donors H Acceptors Log p
1 03851138 2754935 384.366 2 6 5.03
2 02387705 1990445 337.887 2 5 2.96
3 02361742 2001378 442.537 2 6 3.94
8 Letters in Drug Design & Discovery, 2009, Vol. 6, No. 7 Reshmi and Pillai
[17] GOLD 3.0.1; Cambridge Crystallographic DataCentre: Cambridge
CB2 1EZ, UK.
[18] Irwin J.; Shoichet B.K. ZINC—a free database of commercially
available compounds for virtual screening. J. Chem. Inf. Model.,
2005, 182, 45-177.
[19] Wolber, G.; Langer, T. LigandScout: 3-D pharmacophores derived
from Protein-bound ligands and their use as virtual screening fil-
ters. J. Chem. Inf. Model., 2005, 45, 160-169.
[20] Yong Wang.; René Coulombe.; Dale R. Cameron.; Louise
Thauvette.; Marie-Josée Massariol.; Lynn M. Amon.; Dominique
Fink.; Steve Titolo.; Ewald Welchner.; Christiane Y.; Jacques A.
and Peter W. W. Crystal Structure of the E2 Transactivation Do-
main of Human Papillomavirus Type 11 Bound to a Protein Inter-
action Inhibitor. J. Biol. Chem., 2004, 279, 6976-6985.
[21] Sarkhel, S.; Desiraju, G. R. N-H...O, O-H...O, and C-H...O hydro-
gen bonds in protein-ligand complexes: strong and weak interac-
tions in molecular recognition. Proteins, 2004, 54, 247-259.
[22] Desiraju, G.R.; Steiner, T. The Weak Hydrogen Bond in Structural
Chemistry and Biology. Oxford University Press, Oxford. 1999.
[23] Greene, J.; Kahn, S.; Savoj, H.; Sprague, P.; Teig, S. Chemical
Function queries for 3D database search. J. Chem. Inf. Comput.
Sci., 1994, 34, 1297-1308.
[24] Lipinski, C.A.; Lombardo, F.; Dominy, B.W.; Feeney, P.J. Ex-
perimental and computational approaches to estimate solubility and
permeability in drug discovery and development settings. Adv.
Drug Deliv. Rev., 2001, 46, 3-26.
[25] Verdonk, M. L.; Cole, J. C. Improved Protein-Ligand Docking
Using GOLD. Proteins, 2003, 52, 609-623.
[26] Veber, D.F.; Johnson, S.R.; Cheng, H.Y.; Smith, B.R.; Ward,
K.W.; Kopple, K.D. Molecular properties that influence the oral
bioavailability of drug candidates. J. Med. Chem., 2002, 45, 2615-
[27] Jones, G.; Willett, P.; Glen, R.C.; Leach, A.R.; Taylor, R. Devel-
opment and validation of a genetic algorithm for flexible docking.
J. Mol. Biol., 1997, 267, 727-748.
[28] Keseru, G. M. A virtual high throughput screen for high affinity
cytochrome P450cam substrates. Implications for in silico predic-
tion of drug metabolism. J. Comput.-Aided Mol. Des., 2001, 15,
[29] Kontoyianni, M.; McClellan, L. M.; Sokol, G. S. Evaluation of
docking performance: Comparative data on docking algorithms. J.
Med. Chem., 2004, 47, 558-565.
[30] Kroemer, R. T.; Vulpetti, A.; McDonald, J. J.; Rohrer, D. C.; Tros-
set, J. Y.; Giordanetto, F.; Cotesta, S.; McMartin, C.; Kihle´n, M.;
Stouten, P. F. W. Assesment of docking poses: interactions-based
accuracy classification (IBAC) versus crystal structure deviations.
J. Chem. Inf. Comput. Sci., 2004, 44, 871-881.
[31] Kitchen, D.B.; Decornez, H.; Furr, J.R.; Bajorath, J. Docking and
scoring in virtual screening for drug discovery: methods and appli-
cations. Nat. Rev. Drug Discov., 2004, 3, 935-949.
[32] Jacobsson, M.; Liden, P.; Stjernschantz, E.; Bostrom, H.; Norinder,
U. Improving structure-based virtual screening by multivariate
analysis of scoring data. J. Med. Chem., 2003, 46, 5781-5789.
[33] Jones, G.; Willett, P.; Glen, R.C.; Leach, A.R.; Taylor, R. Devel-
opment and validation of a genetic algorithm for flexible docking.
J. Mol. Biol., 1997, 267, 727-748.
[34] Joseph-McCarthy, D.; Alvarez, J.C. Automated generation of
MCSS- derived pharmacophoric DOCK site points for searching
multiconformation Databases. Proteins, 2003, 51, 189-202.
[35] Barreca, M.L.; De Luca, L.; Iraci, N.; Rao, A. ; Ferro, S.; Maga,
G.; Chimirri, A. Structure-based pharmacophore identification of
new chemical scaffolds as non-nucleoside reverse transcriptase in-
hibitors. J. Chem. Inf. Model, 2007, 47, 557-562.
[36] Schuster, D.; Maurer, E.M.; Laggner, C.; Nashev, L.G.; Wilckens,
T.; Langer, T.; Odermatt, A. The discovery of new 11b-
hydroxysteroid dehydrogenase type 1 inhibitors by common feature
pharmacophore modeling and virtual screening. J. Med. Chem.,
2006, 49(12), 3454-3466.
... The discovery of new lead compounds from large database of ligands is widely used in virtual screening method to identify the most promising compounds from the database for further study [12][13][14]. Successful virtual screenings have resulted in discovery of molecules either resembling the *Address correspondence to this author at the Faculty of Science, Lampan-gRajabhat University, Lampang 52100 Thailand; Tel (+66)5424-1052; Fax: (+66)5424-1052; E-mail: native ligands of the specific targets or in the discovery of the entire leads [15,16]. To develop new Cdk5 inhibitors, a Cdk5 template based on the protein target of Cdk5 inhibitors was employed [17]. ...
Cyclin-dependent kinase5 (Cdk5), involved in the processes of neuronal maturation and migration, is expressed in most tissues including proliferation cells. This research is aimed to use Cdk5 as a target for anticancer drug design. Virtual screening of Cdk5 by using Raccoon | AutoDock VS against the National Cancer Institute Diversity Set II has been performed. The selected compounds show binding energy between -10.43 to -8.53 kcal/mol and ligand efficiency values more than 0.40, which revealed 20 potential inhibitor candidates. Moreover, the binding energy of 10 compounds is related to their biological evaluation. These 10 inhibitors from NCI Diversity Set II serve as new Cdk5 inhibitors.
Full-text available
We identified a conditional transcriptional enhancer in the long control region (LCR) of human papillomavirus type 16 (HPV-16). This conditional enhancer requires activation in trans by a product of the viral early-region open reading frames (ORFs). Primer extension analysis of chloramphenicol acetyltransferase RNA isolated from transiently transfected CV-1 cells demonstrated that trans-activation of the HPV-16 LCR enhancer operated at the transcriptional level. Mutational analysis of the early ORFs demonstrated that the conditional enhancer of the LCR was trans-activated by the product of the E2 ORF. The E2 gene product of bovine papillomavirus type 1, which can trans-activate the conditional enhancer in the bovine papillomavirus type 1 LCR, was also capable of trans-activating the E2-responsive enhancer of HPV-16. The activity of the HPV-16 LCR enhancer was also assayed in two human cervical carcinoma cell lines, HeLa and SiHa, which harbor transcriptionally active, integrated HPV-18 and HPV-16 DNA sequences, respectively. No endogenous E2 or E2-like activity was detected in either cell line.
Full-text available
The transactivation domain (AD) of bovine papillomavirus type 1 E2 stimulates gene expression and DNA replication. To identify cellular proteins that interact with this 215-amino-acid domain, we used a transactivation-defective mutant as bait in the yeast two-hybrid screen. In vitro and in vivo results demonstrate that the cDNA of one plasmid isolated in this screen encodes a 37-kDa nuclear protein that specifically binds to an 82-amino-acid segment within the E2 AD. Mutants with point mutations within this E2 domain were isolated based on their inability to interact with AMF-1 and were found to be unable to stimulate transcription. These mutants also exhibited defects in viral DNA replication yet retained binding to the viral E1 replication initiator protein. Overexpression of AMF-1 stimulated transactivation by both wild-type E2 and a LexA fusion to the E2 AD, indicating that AMF-1 is a positive effector of the AD of E2. We conclude that interaction with AMF-1 is necessary for the transcriptional activation function of the E2 AD in mammalian cells.
Full-text available
Induction of gene expression by the papillomavirus E2 protein requires its approximately 220-amino-acid amino-terminal transactivation domain (TAD) to interact with cellular factors that lead to formation of an activated RNA polymerase complex. These interaction partners have yet to be identified and characterized. The E2 protein localizes the transcription complex to the target promoter through its carboxy-terminal sequence-specific DNA binding domain. This domain has been reported to bind the basal transcription factors TATA-binding protein and TFIIB. We present evidence establishing a direct interaction between amino acids 74 to 134 of the E2 TAD and TFIIB. Within this region, the E2 point mutant N127Y was partially defective and W99C was completely defective for TFIIB binding in vitro, and these mutants displayed reduced or no transcriptional activity, respectively, upon transfection into C33A cells. Overexpression of TFIIB specifically restored transactivation by N127Y to close to wild-type levels, while W99C remained inactive. To further demonstrate the functional interaction of TFIIB with the wild-type E2 TAD, this region was fused to a bacterial DNA binding domain (LexA:E2:1-216). Upon transfection with increasing amounts of LexA:E2:1-216, there was reduction of its transcriptional activity, a phenomenon thought to result from titration of limiting factors, or squelching. Squelching of LexA:E2:1-216, or the wild-type E2 activator, was partially relieved by overexpression of TFIIB. We conclude that a specific region of the E2 TAD functionally interacts with TFIIB.
The papillomavirus E2 protein is required for viral transcriptional regulation and replication. The E2 protein has a modular structure with two highly conserved domains, a sequence-specific DNA-binding and dimerization domain and a conserved N-terminus which is important for transcriptional transactivation, replication, and interaction with the E1 protein to determine which specific amino acids or regions in the N-terminus were important for the replication or transactivation functions. Single amino acid substitutions were created at highly conserved, highly charged amino acids in the HPV 11 E2 N-terminus. Each amino acid was mutated to a nonpolar alanine residue or a similarly charged amino acid. The mutated E2 proteins were analyzed for their abilities to support transcriptional transactivation and transient DNA replication and to enhance binding of E1 to the origin of replication. Single amino acid substitutions were identified which were defective for either the replication or transactivation functions, which demonstrated that the replication and transactivation functions within the N-terminus are separable. In several cases different amino acid substitutions at the same site had variable effects on transcription or replication, highlighting the importance of hydrophobic interactions or side chain structure at each site. The replication function appeared to correlate with the ability of E2 to enhance binding of E1 to the origin of replication though these studies also suggest that other functions performed by the E2 protein may be important for replication.
The advent of three-dimensional (3D) molecular database searching motivates this investigation of how best to formulate queries for compounds likely to bind to an enzyme or receptor. 3D queries in the Literature generally refer to simple topological features (e.g., nitrogen or phenyl). To better capture the chemist's intent and to find functionally equivalent but structurally diverse compounds, generalized chemical function definitions are proposed for hydrogen bond accepters/donors, charge centers, and hydrophobes. Use of these function definitions in 5-HT3 antagonist and ACE inhibition queries is shown to identify dramatically more hits capable of forming the hypothesized interactions. Furthermore, false positives that have features inaccessible to the receptor are eliminated. Next, the literature on intermolecular interaction energy is reviewed to determine what geometric tolerances are chemically reasonable in queries. Finally, it is shown that the commonly used distance constraints poorly distinguish conformers that do and do not superimpose well with receptor features. An alternative, the location constraint, is proposed. Queries for angiotensin II antagonism and HLE inhibition are described and used to search 203 000 compounds.
The "high-risk" human papillomavirus types 16 (HPV-16) and 18 (HPV-18) have been etiologically implicated in the majority of human cervical carcinomas. In these cancers, the viral DNAs are often integrated into the host genome so that expression of the E1 and the E2 genes is lost, suggesting that disruption of these regulatory genes plays an important role in carcinogenic progression. Previous studies defining the viral genes affecting HPV-16 transformation functions have used the "prototype" viral genome, which was cloned from a human cervical carcinoma and later discovered to harbor a mutation in the E1 gene. In this study, we have corrected this mutation and have evaluated the effect of mutations of either the E1 or the E2 gene on the efficiency of HPV-16 immortalization of human keratinocytes. Mutation of either the E1 gene or the E2 gene in the background of a "wild-type" HPV-16 genome markedly increased immortalization capacity. Mutations were also generated in the E2-binding sites located upstream of the P97 promoter, which directs synthesis of the viral E6 and E7 transforming genes. E2 negatively regulates the P97 promoter through binding at adjacent sites. Surprisingly, the mutation of these sites only partially relieved the negative effect of E2 on viral immortalization, implicating additional mechanisms in the E2 repression of viral immortalization functions. Our results provide genetic evidence that the E1 and E2 gene products each can repress HPV-16 immortalization and support the hypothesis that a selective growth advantage is provided by integration of the viral genome in a manner that causes the loss of expression of either E1 or E2.
Prediction of small molecule binding modes to macromolecules of known three-dimensional structure is a problem of paramount importance in rational drug design (the "docking" problem). We report the development and validation of the program GOLD (Genetic Optimisation for Ligand Docking). GOLD is an automated ligand docking program that uses a genetic algorithm to explore the full range of ligand conformational flexibility with partial flexibility of the protein, and satisfies the fundamental requirement that the ligand must displace loosely bound water on binding. Numerous enhancements and modifications have been applied to the original technique resulting in a substantial increase in the reliability and the applicability of the algorithm. The advanced algorithm has been tested on a dataset of 100 complexes extracted from the Brookhaven Protein DataBank. When used to dock the ligand back into the binding site, GOLD achieved a 71% success rate in identifying the experimental binding mode.
The crystal structure of the E2 DNA-binding domain from the high-risk cervical cancer-associated strain human papillomavirus type 16 (HPV-16) is described here. The papillomavirus E2 proteins regulate transcription from all viral promoters and are required for the initiation of replication in vivo. They belong to a family of viral proteins that form dimeric beta-barrels and use surface alpha-helices for DNA interaction. Although all E2 proteins recognize the same consensus, palindromic DNA sequence, proteins from different viral strains differ in their abilities to discriminate among their specific DNA-binding sites. The structure reported here reveals that while the overall fold of the HPV-16 E2 DNA-binding domain resembles that of its counterpart from the related viral strain bovine papillomavirus type 1, the precise placement of the recognition helices is significantly different. Additionally, the charge distribution on the DNA-binding surfaces of the two proteins varies; HPV-16 E2 has a much less electropositive surface. HPV-16 E2 is thus less able to utilize charge neutralization of the phosphate groups on DNA to induce bending. These results correlate well with previous solution studies that showed decreased affinity between HPV-16 E2 and flexible DNA target sequences, and enhanced affinity towards A-tract-containing, pre-bent sequences. In summary, the crystal structure of the HPV-16 E2 DNA-binding domain shows that the protein presents a stereo-chemically and electrostatically unique surface to DNA, characteristics that can contribute to its mechanism of DNA target discrimination.
Papillomaviruses cause warts and proliferative lesions in skin and other epithelia. In a minority of papillomavirus types ('high risk, including human papillomaviruses 16, 18, 31, 33, 45 and 56), further transformation of the wart lesions can produce tumours. The papillomavirus E2 protein controls primary transcription and replication of the viral genome. Both activities are governed by a approximately 200 amino-acid amino-terminal module (E2NT) which is connected to a DNA-binding carboxy-terminal module by a flexible linker. Here we describe the crystal structure of the complete E2NT module from human papillomavirus 16. The E2NT module forms a dimer both in the crystal and in solution. Amino acids that are necessary for transactivation are located at the dimer interface, indicating that the dimer structure may be important in the interactions of E2NT with viral and cellular transcription factors. We propose that dimer formation may contribute to the stabilization of DNA loops which may serve to relocate distal DNA-binding transcription factors to the site of human papillomavirus transcription initiation.